UNIVERSIDAD COMPLUTENSE DE MADRID FACULTAD DE VETERINARIA TESIS DOCTORAL Interactions between high and low-virulence isolates of Neospora Caninum and the bovine placenta MEMORIA PARA OPTAR AL GRADO DE DOCTOR PRESENTADA POR Laura Jiménez Pelayo Supervisors Luis Miguel Ortega Mora Esther Collantes Fernández Javier Regidor Cerrillo Madrid © Laura Jiménez Pelayo, 2019 COMPLUTENSE UNIVERSITY OF MADRID VETERINARY FACULTY Animal Health Department INTERACTIONS BETWEEN HIGH- AND LOW- VIRULENCE ISOLATES OF NEOSPORA CANINUM AND THE BOVINE PLACENTA Supervisors: Luis Miguel Ortega Mora Esther Collantes Fernández Javier Regidor Cerrillo DOCTORAL THESIS Ms. Laura Jiménez Pelayo Madrid, July 2019 UNIVERSIDAD COMPLUTENSE DE MADRID FACULTAD DE VETERINARIA Departamento de Sanidad Animal INTERACCIONES ENTRE AISLADOS DE ALTA Y BAJA VIRULENCIA DE NEOSPORA CANINUM Y LA PLACENTA BOVINA Directores: Luis Miguel Ortega Mora Esther Collantes Fernández Javier Regidor Cerrillo TESIS DOCTORAL Dña. Laura Jiménez Pelayo Madrid, julio de 2019 Memoria presentada por Dña. Laura Jiménez Pelayo para optar al grado de Doctor por la Universidad Complutense de Madrid Madrid, 4 de julio de 2019 La realización de esta tesis doctoral ha sido posible gracias a la financiación de la Universidad Complutense de Madrid (UCM, mediante un contrato de Personal Investigador en Formación, del cual he sido beneficiaria. La financiación de las investigaciones ha sido posible gracias a los siguientes proyectos: • Ministerio de Economía y Competitividad (MINECO) del Gobierno de España (AGL2013- 44694-R y AGL201675935-C2-1-R). • Plataforma Tecnológica de Sanidad Animal de la Comunidad de Madrid (PLATESA) (S2013/ABI2906 y PLATESA2-CM P2018/BAA-4370). D. Luis Miguel Ortega Mora, Doctor en Veterinaria y Catedrático de Universidad adscrito al Departamento de Sanidad Animal de la Facultad de Veterinaria de la Universidad Complutense de Madrid, Dña. Esther Collantes Fernández, Doctora en Veterinaria y Profesora titular en el Departamento de Sanidad Animal de la Facultad de Veterinaria de la Universidad Complutense de Madrid, y D. Javier Regidor Cerrillo, Doctor en Farmacia e investigador contratado por la empresa Saluvet-innova, con base en el Departamento de Sanidad Animal de la Facultad de Veterinaria de la Universidad Complutense de Madrid CERTIFICAN: Que la tesis doctoral titulada “Interacciones entre aislados de alta y baja virulencia de Neospora caninum y la placenta bovina” que presenta la Licenciada en Veterinaria Dña. Laura Jiménez Pelayo ha sido realizada en las dependencias del Departamento de Sanidad Animal de la Facultad de Veterinaria, de la Universidad Complutense de Madrid bajo su supervisión y cumple todas las condiciones exigidas para optar al grado de Doctor por la Universidad Complutense de Madrid con Mención Internacional. De acuerdo con la normativa vigente, firmamos el presente certificado, autorizando su presentación como directores de la mencionada Tesis Doctoral. En Madrid, a 4 de julio de 2019 Fdo. Prof. Dr. Luis Miguel Ortega Mora Fdo. Prof. Dra. Esther Collantes Fernández Fdo. Dr. Javier Regidor Cerrillo DOCTORADO CON MENCIÓN INTERNACIONAL La presente tesis doctoral cumple con los requisitos exigidos por la Universidad Complutense de Madrid para obtener la mención de Doctor Internacional: 1) Realización de una estancia mínima de tres meses en una institución de enseñanza superior o centro de investigación fuera de España: • Centro receptor: Anatomisches Institut, Stiflung Tierärzliche Hochschule Hannover (TiHo), Germany. • Investigador principal: Prof. Christiane Pfarrer. • Duración de la estancia: 3 meses (15/09/2017-14/12/2017). 2) Los apartados de resumen, introducción, resultados, discusión y conclusiones de la tesis doctoral han sido redactados en una en una de las lenguas habituales para la comunicación científica en su campo de conocimiento, distinta a cualquiera de las lenguas oficiales en España (inglés). 3) La tesis doctoral ha sido evaluada por dos expertos pertenecientes a alguna institución de educación superior o instituto de investigación no español. 4) El Tribunal evaluador de la tesis está compuesto por, al menos, un experto perteneciente a alguna institución de educación superior o centro de investigación no español. “…sueña con metas altas y volarás como las águilas. Eso debes hacer; alcanzar las cumbres de la vida. Busca al que sea sabio y aprende con él. Usa bien la ambición sin por ello dañar a nadie. No hagas que tengan que recriminarte en tu trabajo, hazlo siempre bien. E intenta ganar cuando te hagan competir. No te dejes avasallar por nadie y aunque hayas nacido en un hogar humilde, no te consideres por ello indigno. Si luchas con esfuerzo, conseguirás todo lo que te propongas.” Gonzalo Giner, El sanador de caballos AGRADECIMIENTOS Mis primeras palabras de agradecimiento van dedicadas a mis directores. Luis, Esther, Javier, gracias por la paciencia y la dedicación demostrada durante estos años. Vuestros conocimientos junto con la capacidad para transmitirlos hacen posible la formación de muchos estudiantes. Me habéis transmitido el amor por la ciencia. Me siento orgullosa de que la universidad pública cuente con profesionales como vosotros, dedicación completa, sin fechas ni horarios. Espero que al leer este trabajo también vosotros os sintáis orgullosos de la formación profesional que he alcanzado, a la que habéis contribuido en gran medida. Extiendo mis agradecimientos a Gema (tu coherencia y tu inteligencia son un ejemplo para mí), Ignacio y Mercedes que, aunque no han estado implicados tan directamente como mis directores, siempre han tenido buenos consejos e ideas que han permitido mejorar esta tesis y cada uno de los trabajos llevados a cabo en estos años. A mis compañeros de la siempre new generation. Si tuviera que agradecer todo lo que habéis hecho por mí podría escribir una tesis sólo de agradecimientos. Ale-Alejandro desde que te conocí ya empecé a admirarte, sólo puedo darte las gracias por ser mi amigo y estar ahí siempre, incondicionalmente, para lo que sea. Sin duda, una de las cosas que más pena me da de cerrar esta etapa de mi vida es dejar de tenerte cerca todos los días, aunque sé que a nuestra amistad le quedan muchos capítulos por escribir (esperemos que mejores que el final de got). Martis te debo media tesis (¡¡y tú a mí otra media!!). Eres de esas casualidades de la vida que pasan porque tenían que pasar, porque nadie hubiera sido mejor compañera que tú, porque estoy segura de que con nadie me hubiera complementado mejor y nadie me hubiera aguantado tanto como tú (conservación en formol incluida para que te mantengas forever young, tú me entiendes…). Carlitos te debo el 90% de mi conocimiento del refranero español y todos esos momentos de risas hasta llorar, excepto cuando me metes en el montón de las menos feas, que eso no me hace ni p… gracia (conmigo no te metas que yo sí me cabreo). Gracias porque has sido una pieza fundamental para sobrellevar estos años, siempre con una broma preparada para enseñarnos a todos que la vida hay que tomársela con humor porque lo que hoy nos parece un drama mañana será una simple anécdota que nos hizo avanzar y mejorar. Roberto sé que sin mí serías ceniza pero, incluso debiéndome la vida, te quiero agradecer lo bien que me has tratado siempre y la paciencia que has tenido para soportar todas las bromas sin borrar la sonrisa de la cara. ¡Os voy a echar mucho de menos! A los post-docs más antiguos y a los recién llegados. Gracias a todos por ser un espejo en el que mirarse y una demostración de que con esfuerzo se consigue todo lo que uno se propone. Pili enhorabuena por todo lo que has conseguido y gracias por estar ahí cada vez que lo he necesitado durante todos estos años. Tu apoyo y tu ayuda han sido fundamentales para lograr esta tesis. Rafiki eres un luchador. Te mereces conseguir lo que quieres. Gracias por enseñarme que un poco de ambición es necesaria para alcanzar las metas y que hay que perseguir los sueños para alcanzarlos, aunque sigo pensando que deberías relajarte y disfrutar de vez en cuando jijiji. David gracias por hacerme ver el otro lado, por enseñarme que es posible conciliar el amor por la ciencia con la vida personal y que hay otros pilares en la vida que hay que cuidar y en los que puedes apoyarte cuando es necesario. A los Saluvet-innova group members. Javier te debo una pata de jamón y todas esas cosas bonitas que ya te he dicho en el primer párrafo y que no voy a repetir porque tú y yo no somos muy de decirnos cosas bonitas…jajaja. Vasca ¡te adoro! A pesar de que la coordinación no es tu fuerte y de que eres cuadriculada como un cubo de Rubik, ese hablar en listo y ese ser una vasca tan poco vasca, sin flequillo y veraneando en Torrevieja no voy a olvidarlos nunca. Espero tenerte cerca mucho tiempo y, si no nos va bien en la ciencia, siempre nos quedará Instagram. Ángela qué bien puesto tienes el nombre. Gracias por ser tan cariñosa, por hablar tanto y por entenderme siempre. Además de ser una persona genial, eres una trabajadora ejemplar y una mami molona. Otra gran persona que me ha regalado Saluvet y que me llevo para siempre. A las nuevas incorporaciones, Sheila y Sofía, desearos mucha suerte en el grupo, estoy segura de que os acogerán y os sentiréis como en casa. A la new new generation… Laurita gracias por ser tú. Todavía no conozco a nadie que pueda evitar quererte. Estoy segura de que triunfarás allá por donde vayas porque las personas buenas y trabajadoras, tarde o temprano, son valoradas. Conocerte ha sido una de las cosas más bonitas de estos años (¡¡y ha habido muchas y muy bonitas!!). Me llevo una amiga tan gafe como yo (¿será el nombre?). Aly mucho ánimo con tus cultivos. Tienes una tesis muy dura por delante, pero estoy segura de que podrás con ella. Por cierto, no te perdono el ceviche, la causa y el ají de gallina que nos debes jijiji. Merche mucha suerte con tus aislamientos, tus mataderos, tus ratoncillos… Gracias por tu buen humor, por tu alta capacidad para soportar las bromas de Carlos (lo siento, no he podido evitarlo) y hacernos pasar muy buenos momentos. María gracias por tu implicación, tu profesionalidad, tu exquisita educación y tu sonrisa permanente. Habéis traído aire fresco al laboratorio, muy necesario en esta etapa final de la tesis, aunque eso de que las nuevas generaciones sean más listas, más guapas y más simpáticas no lo llevo nada bien. Al resto de miembros de Saluvet. Gracias Lola por ser un poco la madre de todos, por preocuparte de nosotros y, sobre todo, por ocuparte de la maldita burocracia. Vane gracias por toda tu ayuda y por la alegría que desprendes por cada sitio por el que pasas. Gracias a Cinta, Javier Lobo, Isabel y Chema por ser tan amables y buenos compañeros, jamás habéis puesto ningún problema para compartir vuestro material e instalaciones cuando lo hemos necesitado. A los “becarios” de Cinta, Jaime, Juanjo, Ángeles y Laura por compartir con nosotros momentos de risas y estar siempre dispuestos a echar una mano. Gustavo, Javier, Abel y Alicia gracias por vuestra simpatía y el compañerismo, nunca nos ha faltado vuestra ayuda y vuestros consejos. A los que pasaron a mejor vida (laboralmente hablando). Raquelilla hace nada que te has ido y ya te echamos de menos. Gracias por toda tu ayuda procesando muestras y por tu profesionalidad, a pesar de ser insultantemente joven. Moreno, gracias por formar parte del dúo cómico más grande de la historia de Saluvet. Eres un genio. Nada más que añadir. A la old generation quiero agradecerle toda la ayuda que me brindaron cuando empecé en este camino y que me han seguido dando en estos años, algunos desde más cerca (Dani y Paula) y otros desde un poco más lejos (Iván y Deivid), pero siempre dispuestos a echar una mano e intentando que las nuevas generaciones no repitiéramos sus errores (que ya es bastante con los nuevos que se nos han ido ocurriendo…). Al dúo dinámico, Luis Corpa y Jorge, por todas las risas, el compañerismo y el amor a vuestro trabajo. Se necesitan más veterinarios como vosotros. A Dani princeso porque, aunque sus circunstancias no eran sencillas, siempre sacaba su lado más amable y generoso. A Anita, gracias por ocuparte de la organización del Apicowplexa, fue un congreso estupendo, y gran parte del éxito fue gracias a ti. No me olvido de aquellas personas que pasaron más rápidamente por mi aventura en Saluvet pero que, aun así, han contribuido a hacerla todavía mejor (Ofe, Joaqui, Eva, Yoli, Ali). A los grupos que me han acogido en las distintas estancias que he realizado durante los 5 años de tesis. Christiane thank you for being so kind and for sharing your experience with me. I hope that you enjoy reading this doctoral thesis, which have been possible thanks to you too. Thanks to all your fantastic team (Hanna, Doris, Jan, Christian, Missaka…). I am always grateful to have had this opportunity to be part of your research group. Danke :). A los españoles de Hannover (Bea, Cris, Inma, Ana, Andrés, Fran, Guille, Oriol) gracias por acogerme esos tres meses de frío cri-criminal, vosotros me distéis el calor que todo español necesita a 0 grados. Y gracias también a mis primos alemanes, Ali y Marco, por visitarme y enseñarme un trocito de Alemania. Thanks to Prof. Dirk Werling and his team for teaching me about bovine immunology, as difficult as important. A todo el Área de Investigación en Sistemas de Producción Animal del SERIDA, en especial a Koldo y Ali, por toda su ayuda durante los duros meses en Asturias. ¿Quién me iba a decir a mí que acabaría trabajando con vacas? Pues sí, también valgo para eso, aunque siempre maquillada que antes muerta que sencilla. No sé qué hubiéramos hecho las de Madrid sin vosotros con les vaques. Menos mal que estabais allí para echar siempre una mano. Gracias al Instituto de Ganadería de Montaña por acogerme durante un mes. Me gustaría agradecer especialmente a Julio por toda su ayuda, por su paciencia, por su carácter. No encuentro palabras de agradecimiento suficientes, aunque admitirás que unas cuantas risas a costa de MartaLaura te has echado… También gracias a ti Dani, mi postureta favorito, por integrarnos en tu nueva casa y por enseñarnos que fuera de Saluvet también hay vida y gente maravillosa (Carmen, Elora, Noive, Vero, Maikel, Marcos…). A los extranjeros que, sin duda, me han ayudado a crecer personal y profesionalmente y a tener una visión más amplia del mundo en el que vivimos. A los brasileños, Pomy, Müller, Wagner y Larissa gracias por vuestras sonrisas profident que llenaron el laboratorio de alegría. Espero poder visitaros pronto. Al italiano más simpático de la historia, Luca, que, aunque flipamos cuando llegó con sus 5 maletas para 2 meses, luego entendimos que Italia también is different y que el café y la pasta de la mama son sagrados. Ti voglio bene amico. A los argentinos, Marcelo, Yani, Nacho, Lucy, gracias por traernos una parte de Argentina a Madrid, pero no, el mate no es lo mío. Me he reído un montón con todos vosotros y siempre os recordaré con cariño. A las peruanas Lucy y Alessandra, gracias por ser tan buenas alumnas y aguantar todas mis chapas anotando cada punto y cada coma. A la Profesora Cristina, por sus consejos, su simpatía y todos los regalos que nos trae de Méjico siempre que vuelve. Gracias a todos, en Madrid tenéis una casa. Y, por último, pero no menos importante, gracias a mi familia. A la familia que no he elegido porque no hubiera podido imaginar nada mejor de lo que tengo. A mis padres, Feli y Jose, a los que les debo una carrera, un máster, una tesis, un hogar, un hermano, dos perras y la vida que cualquier persona podría desear. Gracias porque, además de darme todo eso, me aguantáis cada día. Soy consciente de que sufrís más que yo cuando tropiezo y disfrutáis más que yo cuando triunfo. Espero que vosotros también estéis orgullosos y que algún día pueda devolveros todo lo que hacéis por mí. Al resto de mi familia, a los que siguen aquí y a los que se fueron demasiado pronto, aunque nadie se va si alguien le sigue recordando cada día. Todos habéis contribuido a esto y una parte de esta tesis también es vuestra. Y a la familia que he elegido, mis amigos, los de toda la vida (Víc y Carol), los vets (Kris, Ceci, María, Carmen, Irene, Cid, Pale, Cris), a las chicas de New York (Paula y Lara), a todas las parejas de mis amigos, que ya son mis amigos también (en especial a Miguel Hernández, que ya tenía nombre artístico, y es el autor de algunas de las ilustraciones de esta tesis). Muchas gracias por quererme tal y como soy y por apoyarme en cada paso. Muchos me habéis visitado cuando estaba expatriada en diversas localizaciones de la geografía europea, otros me habéis aguantado horas hablando de mis movidas científicas, otros me habéis sacado de fiesta cuando lo necesitaba y todos tenéis la culpa de que hoy presente esta tesis. ¡Os debo una buena celebración! Porque dinero no tengo, pero historias que contar todas las que queráis. Y a te che sei, semplicemente sei <3. Don't wanna feel another touch, don't wanna start another fire, don't wanna know another kiss, no other name falling off my lips. Si mi famosa falta de suerte se debe a que la gasté el día en el que os encontré, entonces ha merecido la pena. TABLE OF CONTENTS ÍNDICE Table of contents III LIST OF TABLES AND BOXES/ÍNDICE DE TABLAS Y CUADROS………………………………………………………….VII LIST OF FIGURES/ÍNDICE DE FIGURAS…………………………………………………………………………………………….IX LIST OF ABBREVIATIONS/LISTADO DE ABREVIATURAS……………………………………………………………………XI CHAPTER I SUMMARY/RESUMEN…………………………………………………………………………………………………...3 CHAPTER II INTRODUCTION/INTRODUCCIÓN 1. Neospora caninum and bovine neosporosis………………………………………………………………………………13 1.1 Taxonomic classification and morphology of the parasite…………………………………………………….13 1.2 Life cycle and transmission……………………………………………………………………………………………….….16 1.3 Prevalence and economic impact……………………………………………………………………………………..…19 1.4 Pathogenesis, clinical signs and lesions………………………………………………………………………………..19 1.5 Diagnosis and control………………………………………………………………………………………………………..…21 2. Host-parasite interactions in bovine neosporosis: factors influencing abortion and transmission………………………………………………………………………………………………………………………………….24 2.1 Factors depending on the host………………………………………………………………………………………….…24 2.1.1 Gestation period……………………………………………………………………………………………………………24 2.1.2 Foetal immune responses……………………………………………………………………………………………..25 2.1.3 Maternal immune responses…………………………………………………………………………………………26 2.1.4 Bovine placenta and neosporosis….……………………………………………………………………………...28 2.1.4.1 Trophoblast cells in the maternal-foetal interface…………………………………………………..31 2.1.4.2 Extracelular matrix (ECM)……………………………………………………………………………………….32 2.1.4.3 Foetal membranes and allantoic and amniotic fluids………………………………………………33 2.1.4.4 Placental functions………………………………………………………………………………………………….34 2.1.4.4.1 Nutritive function……………………………………………………………………………………………..34 2.1.4.4.2 Gas exchange across the placenta……………………………………………………………………35 2.1.4.4.3 Endocrine function…………………………………………………………………………………………..35 2.1.4.4.4 Protective function…………………………………………………………………………………………..37 2.1.4.5 Cell cultures isolated from the bovine placenta………………………………………………………38 2.1.4.5.1 Bovine trophoblast cell cultures from cotyledon………………………………………………39 2.1.4.5.2 Epithelial cell cultures from bovine caruncle…………………………………………………….40 2.2 Factors depending on the parasite…………………………………………………………………………………….…42 2.2.1 Parasite stage, inoculation route and dose of infection…………………………………………………42 2.2.2 Isolates and intraspecific variability……………………………………………………………………………….43 2.2.2.1 Phenotypic variability in in vitro models………………………………………………………………….44 2.2.2.2 Variability in the murine model……………………………………………………………………………….47 2.2.2.3 Variability in the bovine model………………………………………………………………………………..50 2.2.2.3.1 Non-pregnant bovine model…………………………………………………………………………….50 2.2.2.3.2 Pregnant bovine model…………………………………………………………………………………….50 2.2.2.4 Molecular basis of the variability…………………………………………………………………………….52 2.2.2.4.1 Genomic studies……………………………………………………………………………………………….53 2.2.2.4.2 Transcriptomic studies……………………………………………………………………………………..54 2.2.2.4.3 Proteomic studies…………………………………………………………………………………………….56 CHAPTER III JUSTIFICATION AND OBJECTIVES/JUSTIFICACIÓN Y OBJETIVOS………………………………….61 CHAPTER IV RESULTS (PUBLICATIONS)/RESULTADOS (PUBLICACIONES)……………………………………….67 Table of contents IV Objective 1: Characterization of parasite interaction between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro………………………………………………………………………….69 Sub-objective 1.1: Characterization of the lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro. 1. Background…………………………………………………………………………………………………………………………….…72 2. Methods……………………………………………………………………………………………..…………………………………….73 2.1 Parasites and cell cultures…………………………………………………………………………………………………….73 2.2 Parasite invasion rate…………………………………………………………………………………………………………..74 2.3 Cell infection rate…………………………………………………………………………………………………………………74 2.4 Adhesion-invasion assay………………………………………………………………………………………………………75 2.5 Intracellular proliferation assays: Proliferation kinetics, doubling time and tachyzoite yield determinations………………………………………………………………………………………………………………………….75 2.6 Immunofluorescence staining……………………………………………………………………………………………..75 2.7 DNA extraction and real-time PCR……………………………………………………………………………………….75 2.8 Statistical analysis………………………………………………………………………………………………………………..76 3. Results ………………………………………………………………………………………………………………………………………76 3.1 Parasite invasion rate (pInvR) ……………………………………………………………………………………………..76 3.2 Cell infection rate (cInfR) …………………………………………………………………………………………………….76 3.3 Adhesion-invasion assay………………………………………………………………………………………………………77 3.4 Proliferation kinetics, doubling time comparisons and tachyzoite yield determination……….78 4. Discussion …………………………………………………………………………………………………………………………………80 5. Conclusions…………………………………………………………………………………………………………………………….…84 6. References………………………………………………………………………………………………………………………………..85 Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level. 1. Background……………………………………………………………………………………………………………………………….90 2. Methods……………………………………………………………………………………………………………………………………91 2.1 Parasites and cell cultures…………………………………………………………………………………………………….91 2.2 Infection of the cultures, collection and preservation of the samples…………………………………..91 2.3 RNA extraction, reverse transcription and quantitative real-time PCR………………………………….92 2.4 Measurement of cytokines in supernatants of BCEC-1 and F3 cell cultures by ELISA…………….92 2.5 Statistical analysis………………………………………………………………………………………………………………..92 3. Results ………………………………………………………………………………………………………………………………………93 3.1 Expression profile of TLR-2 …………………………………………………………………………………………………..93 3.2 Pro-inflammatory and regulatory cytokine modulation ……………………………………………………….94 3.3 Endothelial adhesion molecule (ICAM-1 and VCAM-1) expression ………………………………………94 4. Discussion ……………………………………………………………………………………………………………………………..….96 5. Conclusions……………………………………………………………………………………………………………………………….98 6. References………………………………………………………………………………………………………………………………..99 Objective 2: Investigation of the early infection by high- and low-virulence isolates of N. caninum in pregnant cattle at mid-gestation…………………………………………………………………………………………………107 Table of contents V Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid- gestation with high- and low-virulence isolates. 1. Introduction………………………………………………………………………………………………………….…………………110 2. Materials and methods……………………………………………………………………………………………………………111 2.1 Animals and experimental design………………………………………………………………………………………111 2.2 Parasites…………………………………………………………………………………………………………………………….111 2.3 Clinical monitoring and sampling…………………………………………………………………………..…………..111 2.4 Histopathology and lesion quantification……………………………………………………………………………112 2.5 Tissue DNA extraction and PCR determinations………………………………………………………………….112 2.6 IFN-γ responses in sera…………………………………………………………………………………………..………….112 2.7 N. caninum-specific IgG responses……………………………………………………………………………………..112 2.8 Statistical analysis………………………………………………………………………………………………………………112 3. Results…………………………………………………………………………………………………………………………………….113 3.1 Clinical observations………………………………………………………………………………………………………….113 3.2 Pathology and lesion quantification……………………………………………………………………………………113 3.3 Parasite distribution and burden in placental and foetal tissues………………………………………..115 3.4 IFN-γ kinetics in sera……………………………………………..…………………………………………………………..116 3.5 Specific anti-Neospora IgG responses in heifers and foetuses…………………………………………….116 4. Discussion……………………………………………………………………………………………………………………………….117 5. References………………………………………………………………………………………………………………………………121 Additional file 1: Materials and methods……………………………………………………………………………………..125 Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. 1. Background……………………………………………………………………………………………………………………………..130 2. Materials and methods……………………………………………………………………………………………………………131 2.1 Ethics statement………………………………………………………………………………………………………………..131 2.2 Animals and experimental design………………………………………………………………………………………131 2.3 Sample collection……………………………………………………………………………………………………………… 131 2.4 RNA extraction and reverse transcription……………………………………………………………………….….132 2.5 Quantitative real-time PCR (qPCR) …………………………………………………………………………………….132 2.6 Immunohistochemistry………………………………………………………………………………………………..……132 2.7 Statistical analysis……………………………………………………………………………………………………………...133 3. Results…………………………………………………………………………………………………………………………………….133 3.1 Messenger RNA expression levels………………………………………………………………………………………133 3.2 Detection and distribution of N. caninum antigens and characterization of inflammatory cell populations in the placenta……………………………………………………………………………………………………..137 3.3 Detection and distribution of ECM components in the placenta…………………………………….….143 3.4 Differences in the expression pattern between placentas from live and dead foetuses suggested the implication of certain molecules in foetal death……………………………………………….146 4. Discussion……………………………………………………………………………………………………………………………….146 5. Conclusions……………………………………………………………………………………………………..………………………150 6. References………………………………………………………………………………………………………………………………150 Table of contents VI CHAPTER V GENERAL DISCUSSION/DISCUSIÓN GENERAL……………………………………………………………161 CHAPTER VI CONCLUSIONS/CONCLUSIONES……………………………………………………………………………….175 CHAPTER VII REFERENCES/BIBLIOGRAFÍA……………………………………………………………………………………185 APPENDIX I ORIGINAL MANUSCRIPTS/MANUSCRITOS ORIGINALES…….……………………………….……..213 List of tables and boxes VII LIST OF TABLES AND BOXES CHAPTER II INTRODUCTION/INTRODUCCIÓN Box 1. Placentome structure……………………………………………………………………………………………………..30 Table 1. Summary of bovine trophoblast cultures isolated from blastocysts, cotyledons or by new methods and bovine caruncular cell cultures isolated from pregnant animals……………………………41 Table 2. Virulence classification of different isolates of N. caninum according to their behaviour in murine pregnant models and in vitro……………………………………………………………………………….……49 CHAPTER IV RESULTS (PUBLICATIONS)/RESULTADOS (PUBLICACIONES) Objective 1: Characterization of parasite interaction between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro Sub-objective 1.1: Characterization of the lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro. Table 1. Summary of virulence traits as a function of cell type and N. caninum isolate………………83 Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level. Table 1. Sequence of primers used for cytokine real-time PCR (qPCR) and standard curve data….93 Table S1. Statistical test results for mRNA expression levels……………………………………………………103 Table S2. Statistical results for protein secretion……………………………………………………………………..105 Objective 2: Investigation of the early infection by high- and low-virulence isolates of N. caninum in pregnant cattle at mid-gestation. Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid- gestation with high- and low-virulence isolates. Table 1. Summary of early infection dynamics in heifers and foetuses from G-Control, G-NcSpain7 and G-NcSpain1H……………………………………………………………………………………………………………….……119 Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. Table 1. Presence and distribution of matrix metalloproteinases (MMPs), inhibitory factors of metalloproteinases (TIMPs) and some components of the extracellular matrix (ECM) in placental samples obtained from an experimental infection model in pregnant cattle……………………………146 List of tables and boxes VIII Table S1. Sequences of primers used for cytokine real-time PCR (qPCR) and standard curve data …………………………………………………………………………………………………………………………………………………...155 Table S2. Antibodies, specificity and immunohistochemical procedure used……………………………156 List of figures IX LIST OF FIGURES CHAPTER II INTRODUCTION/INTRODUCCIÓN Figure 1. Taxonomic classification of N. caninum…………………………………………………………………….…14 Figure 2. Graphical representation and microscopic images of N. caninum parasite stages…….…15 Figure 3. Graphic representation of the ultra-structure of the tachyzoite of N. caninum, with special mention of the cytoskeleton (left) and cytoplasm (right) components found only in apicomplexan protozoa……………..………………………………………………………………………………………….….16 Figure 4. Life cycle and transmission of N. caninum……………………………………………………………………18 Figure 5. Representative images of the bovine placenta……………………………………………………….……29 Figure 6. Representative image from histological slide of the bovine placenta stained with HE…..30 Figure 7. Graphical representation of the main placental functions……………………………………………36 Figure 8. Representative images of bovine trophoblast cells (F3) and bovine epithelial caruncular cells (BCEC-1) in culture…………………………………………………………………………………………………………….39 Figure 9. Graphic representation of the lytic cycle of N. caninum and the image in cell culture…45 CHAPTER IV RESULTS (PUBLICATIONS)/RESULTADOS (PUBLICACIONES) Objective 1: Characterization of parasite interaction between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro Sub-objective 1.1: Characterization of the lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro. Figure 1. Parasite invasion rates in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H isolates………………………………………………………………………………………………………………………………………77 Figure 2. Infection and multi-infection rates in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H isolates…………………………………………………………………………………………………………………..78 Figure 3. Adhesion assay in F3 and BCEC-1 infected by Nc-Spain7 and Nc-Spain1H at 4 hpi………79 Figure 4. Proliferation kinetics over time and tachyzoite yield at 58 hpi…………………………………….81 Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level. Figure 1. TLR-2, IL-8, TNF-α, IL-6, IL-12p40, TGF-β1, ICAM-1 and VCAM-1 transcript expression…95 Figure2. IL-8, TNF-α and IL-6 secretion levels in culture supernatants……………………………………….96 List of figures X Objective 2: Investigation of the early infection by high- and low-virulence isolates of N. caninum in pregnant cattle at mid-gestation. Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid- gestation with high- and low-virulence isolates. Figure 1. Rectal temperatures………………………………………………………………………………………………….114 Figure 2. Proteinaceous exudate at the haemophagus area of the placentome……………………….115 Figure 3. N. caninum burdens in placental and foetal tissues….……………………………………………….116 Figure 4. IFN-γ kinetics in sera………………………………………………….……………………………………………..117 Figure 5. N. caninum-specific humoural immune responses…………………………………………………...118 Additional file 2. Histological findings in placental and foetal samples………………………………………126 Additional file 3. Quantification of necrosis foci (NF), size (ASF) and affected area (%LES) of these foci……………………..…………………………………………………………………………………………………………………..127 Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. Figure 1. PRR, cytokine, chemokine, endothelial adhesion molecules, MMPs and TIMPs transcript expression……………………………………………………………………………………………………………………………….135 Figure 2. Immunohistochemical labelling of parasite antigen in the placenta…………….……….……137 Figure 3. Quantification of immunohistochemically labelled T and B lymphocytes and phagocytic cells in the placenta………………………………………………………………………………………………..……………….139 Figure 4. Comparison of the immunohistochemical labelling of lymphocytes in the placenta…..140 Figure 5. Comparison of the immunohistochemical labelling of phagocytic cells in the placenta ………………………………………………………………………………………………………………………………………………..142 Figure 6. Comparison of the immunohistochemical labelling of MMP, TIMPs and ECM components ………………………………………………………………………………………………………………………………………………..144 Figure 7. Comparison of the immunohistochemical labelling of ECM components……………………145 Figure 8. Comparison of PRR, cytokine, chemokine, endothelial adhesion molecules, MMPs and TIMPs transcript expression between animals carrying NVF and VF from G-NcSpain7…………….147 Figure 9. Comparison of immunohistochemically labelled T and B lymphocytes and macrophages in the placenta of animals carrying NVF and VF from G-NcSpain7……………………………………….…..147 Supplementary 3. Th1/Th2 balance in bovine caruncles (a) and cotyledons (b) infected by Nc-Spain1H and Nc-Spain7……………………………………………………………………………………………………...157 List of abbreviations XI LIST OF ABBREVIATIONS ASF Average size focus Tamaño medio del foco BCEC-1 Bovine caruncular epithelial cell line 1 Células epiteliales de la carúncula bovina BKI Bumped kinase inhibitor Inhibidores de la proteína quinasa BNC/TGC Binucleated trophoblast cells/Giant trophoblast cells Células del trofoblasto binucleares/Células del trofoblasto gigantes bp Base pairs Pares de bases bPL Bovine placental lactogen Lactógeno placentario bovino BUVECs Bovine umbilical vein endothelial cells Células endoteliales umbilicales bovinas BVD Bovine viral diarrhea Diarrea vírica bovina CA Caruncle Carúncula Ca Calcium Calcio CCL Chemokine (C-C motif) ligand Ligando de quimioquina (motivo C-C) CD Cluster of differentiation Cúmulo de diferenciación CDPK Calcium dependent protein kinase Proteína quinasa dependiente de calcio cInfR Cell infection rate Tasa de infección celular cm Centimetre Centímetro CNS/SNC Central nervous system Sistema nervioso central CO Cotyledon Cotiledón CO2 Carbon dioxide Dióxido de carbono CSF Colony stimulating factor Factor estimulador de colonias Ct Cycle threshold value Valor umbral de ciclo DAB 3,3′Diaminobenzidine 3,3′Diaminobenzidina DC/CD Dendritic cells Células dendríticas DMEM Dulbecco’s modified eagle medium Dulbecco’s modified eagle medium DNA/ADN Deoxyribonucleic acid Ácido desoxirribonucleico dpi Days post-infection Días post-infección dsDNA Double-stranded DNA ADN de doble cadena e.g. Exempli gratia (for instance) Exempli gratia (por ejemplo) ECM/MEC Extracellular matrix Matriz extracelular ELISA Enzyme-linked immunosorbent assay Ensayo inmunoenzimático EnTT Endogenous transplacental transmission Transmisión transplacentaria endógena ESTs Expressed sequence tag Etiqueta de secuencia expresada EU/UE European Union Unión Europea ExTT Exogenous transplacental transmission Transmisión transplacentaria exógena F3 Bovine placental trophoblast cell line Células del trofoblasto bovino FB Foetal brain Cerebro fetal FCS Foetal calf serum Suero fetal bovino FGF Fibroblast growth factor Factor de crecimiento de fibroblastos FL Foetal liver Hígado fetal GRA Dense granule antigen Antígeno de gránulos densos h Hours Horas HE Haematoxylin-eosin Hematoxilina-eosina HFF Human foreskin fibroblasts Fibroblastos de piel de prepucio humana HPCVE Human placental chorionic villi explants Explantes de vellosidades coriónicas placentarias humanas hpi Hours post-infection Horas post-infección i.e. Id est (this is) Id est (esto es) ICAM Intercellular adhesion molecule Molécula de adhesión intercelular List of abbreviations XII IFAT Indirect fluorescent antibody test Test de inmunofluorescencia indirecta IFN Interferon Interferón IgG Immunoglobulin G Inmunoglobulina G IHC/IHQ Immunohistochemistry Inmunohistoquímica IL Interleukin Interleuquina IM Intramuscular Intramuscular iNOS Inducible nitric oxide synthase Óxido nítrico sintasa inducible IP Intraperitoneal Intraperitoneal ITS-1 Internal transcribed spacer-1 Espacio transcrito interno-1 IV Intravenous Intravenoso/a K Potassium Potasio LAK Lymphokine activated killer cell Célula linfoide derivada de las células NK Lymphocyte Th T helper lymphocyte Linfocito T colaborador M Molar Molar MAPK Mitogen-activated protein kinase Proteína quinasa activada por mitógenos M‐CSF Macrophage colony-stimulating factor Factor estimulante de colonias de macrófagos MEK/ERK Mitogen-activated protein kinase/Extracellular signal-regulated kinase Proteína quinasa activada por mitógenos / Quinasa regulada por señal extracelular MHC Major histocompatibility complex Complejo mayor de histocompatibilidad MIC Microneme protein Proteína de micronemas min Minute Minuto ml Millilitre Mililitro mM Milli molar Milimolar mm Millimetre Milímetro MMP Matrix metalloproteinase Metaloproteinasas de la matriz extracelular MOI Multiplicity of infection Multiplicidad de infección MPSS Massively parallel signature sequencing Secuenciación masiva de firmas paralelas mRNA/ARNm Messenger ribonucleic acid Ácido ribonucleico mensajero n Sample number Número de muestra NA Not applicable No aplicable Na Sodium Sodio Nc Neospora caninum (prefix) Neospora caninum (prefijo) NC Data not comparable Datos no comparables NF Necrotic foci Focos de necrosis NGS Normal goat serum Suero normal de cabra NK Natural killer (cells) (Células) asesinas naturales NO Nitric oxide Óxido nítrico No. Number Número NOD Nucleotide-binding oligomerization domain-like receptors Receptores similares al dominio de oligomerización de unión a nucleótidos NS No significant differences Diferencias no significativas NVF Non-viable foetuses Fetos no viables O2 Oxigen Oxígeno OD Optical density Densidad óptica OR Odds ratio Odds ratio PAF Platelet-activating factor Factor activador de plaquetas PAGs Pregnancy-associated glycoproteins Glicoproteínas asociadas a la gestación List of abbreviations XIII PBS Phosphate buffer saline Tampón fosfato salino PCR Polymerase chain reaction Reacción en cadena de la polimerasa pg Picogram (10-12 grams) Picogramo (10-12 gramos) pi Post-infection Post-infección pInvR Parasite invasion rate Tasa de invasión del parásito pInvRT Total parasite invasion rate Tasa de invasión total del parásito PRP Prolactin related protein Proteína relacionada con la prolactina PRRs Pattern recognition receptors Receptores de reconocimiento de patrones qPCR Quantitative PCR PCR cuantitativa RAPD Randomly amplified polymorphic DNA ADN polimórfico amplificado al azar RIPC Relative index per cent Índice relativo por cien RNA/ARN Ribonucleic acid Ácido ribonucleico ROP Rhoptry bulb protein Proteína de cuerpo de roptrias ROS Reactive oxygen species Especies de oxígeno reactivas RT Room temperature Temperatura ambiente SAG Surface antigen Antígeno de superficie SAGE Serial analysis gene expression Análisis serial de expresión génica SC Subcutaneous Subcutánea SD Standard deviation Desviación estándar SERP Serpine Serpina SRS SAG1-related sequence Secuencia SAG1 relacionada Td Doubling time Tiempo de duplicación Tg Toxoplasma gondii (prefix) Toxoplasma gondii (prefijo) TGF Transforming growth factor Factor transformante de crecimiento TIMP Matrix metalloproteinase inhibitor Inhibidores de las metaloproteinasas de matriz extracelular TLR Toll-like receptor Receptor tipo toll TNF Tumoral necrosis factor Factor de necrosis tumoral TY Tachyzoite yield Recolección de taquizoítos UK United Kingdom Reino Unido USA United States of America Estados Unidos de América UTC Uninucleated trophoblast cells Células del trofoblasto mononucleares VCAM Vascular cell adhesion molecule Molécula de adhesión vascular VEGF Vascular endothelial growth factor Factor de crecimiento endotelial VF Viable foetuses Fetos viables WB Western blotting Western blot wpi Weeks post-infection Semanas post-infección μg Micrograms (10-6 grams) Microgramo (10-6 gramos) µm Micrometer Micrómetros Χ2 Chi square Chi cuadrado %LES Total area affected by necrosis Área total afectada por la necrosis °C Grades centigrade Grados centígrados 2-DE Two-dimensional electrophoresis Electroforesis en dos dimensiones CHAPTER I SUMMARY RESUMEN Chapter I ~ Summary 3 Bovine neosporosis is one of the most important causes of abortion in cattle worldwide and Neospora caninum, an apicomplexan protozoan parasite, is the causal agent of the disease. Millionaire economic losses have been estimated in infected cattle, mainly associated with the reproductive failure. N. caninum is very effective at crossing the placental barrier and placental damage is crucial in the pathogenesis of abortion. In addition, infection during pregnancy poses a challenge for the immune system of pregnant cows. Changes in the T helper 1 and T helper 2 balance in the placenta during gestation have been associated with abortion. Bovine trophoblast and caruncular cell layers are key cellular components at the maternal-foetal interface in placentomes, playing a fundamental role in placental functionality. In addition to their role as a barrier, those cells are able to recognize and induce immune responses against N. caninum, participating in the initiation of innate immune responses at the placental level as well as in the development of an adaptative immune response. However, the causes that lead to abortion remain unknown and the mechanisms by which N. caninum infects the placenta and reaches the foetus are poorly studied. Despite bovine placenta plays a fundamental role in the pathogenesis of bovine neosporosis and placental invasion, proliferation and dissemination to the foetus seem to be crucial events, few studies focused on the early host-cell interactions at the placental level have been approached. In fact, only one descriptive in vitro study in primary giant trophoblast cells was carried out up to date. Moreover, a limited number of in vivo studies have been conducted to investigate the consequences of N. caninum infection at mid-gestation, which is when most abortions occur in naturally infected cattle. Finally, the parasite intra-specific variability influences the outcome of infection. However, studies conducted to compare isolates of different virulence are scarce, and none has assessed comparisons of isolates in bovine placental target cells at early stages of infection neither in vitro or in vivo. Here, the role of the placenta in the pathogenesis of the bovine neosporosis and the influence of the biological variability of the isolate in placental damage have been investigated. The first objective of the present Doctoral Thesis was the characterization of the interactions between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro with the aim of clarifying the events happened in the placenta and the role played by cells conforming the maternal-fetal interface after infection with isolates of different virulence of N. caninum, as well as the factors that enable some isolates to be more effectively transmitted and cause fetal death than others. In order to reach this objective, we initially studied tachyzoite adhesion, invasion, proliferation and egress of high- (Nc-Spain7) and low- (Nc-Spain1H) virulence N. caninum isolates in established cultures of bovine caruncular epithelial (BCEC-1) and trophoblast (F3) cells by immunostaining plaque assay and real-time PCR (qPCR). N. caninum invaded and proliferated in both cell lines. However, invasion, infection and proliferation of both isolates were lower in BCEC-1 than in F3, suggesting resistance of caruncular layer to N. caninum infection and a role as a barrier for this cell line. Tachyzoites of both isolates showed an early egression in BCEC-1, showing that BCEC-1 cannot avoid transmission although they can restrict adhesion and multiplication. On the other hand, Nc-Spain7 showed higher invasion and infection than Nc-Spain1H, mainly in trophoblast cells, as well as higher proliferation abilities, as was demonstrated in previous stablished cell cultures. Parasitophorous vacuoles raised a considerably big size in bovine trophoblast, contrary to observed in BCEC-1, that may indicate an important role of foetal trophoblasts as a niche for N. caninum replication. Chapter I ~ Summary 4 After studying the events of the lytic cycle, modulation of the trophoblast and caruncular immune responses by both parasites was investigated in vitro by RNA extraction and RT-qPCR determination in order to establish the role of placental cells in the regulation of innate immune responses at the placental level and their reprogramming by N. caninum infection with isolates of variable virulence. Expression profiles of toll-like receptor-2 (TLR-2), Th1 and Th2 cytokines (IL- 4, IL-10, IL-8, IL-6, IL-12p40, IL-17, IFN-γ, TGF-β1, TNF-α), and endothelial adhesion molecules (ICAM-1 and VCAM-1) were investigated. A similar modulation in both cell lines was observed consisting in upregulation of pro-inflammatory TNF-α and IL-8 and downregulation of IL-6 and TGF-β1 in infected cultures. Protein secretion of IL-6, IL-8 and TNF-α was confirmed by ELISA. Therefore, N. caninum infection favoured a pro-inflammatory response in placental target cells in vitro. Comparing isolates, higher expression levels of TLR-2 and TNF-α were found in trophoblasts infected with the low-virulence isolate Nc-Spain1H, suggesting immunomodulation differences between high- and low-virulence isolates, which would partially explain the existing differences in virulence. The study of the early infection in a pregnant bovine model inoculated with high- and low- virulence isolates of N. caninum at mid-gestation was the second objective of this Doctoral Thesis. Host-pathogen interaction, specially in the first stages of infection, as well as the intrinsic virulence factors associated with the implicated isolate are crucial in the outcome of infection. To achieve this objective, an experimental infection was carried out by intravenously inoculation of 107 tachyzoites of Nc-Spain7 or Nc-Spain1H isolates in pregnant heifers, and subsequent culling of the animals at two early moments post-infection (10 and 20 days post-infection-dpi). Comparative analysis of the early infection dynamics (clinical outcome, parasite distribution and burden and lesion development in placental and foetal tissues), the specific antibody response and the early modulation of placental immune responses and the extracellular matrix (ECM) regulation were investigated. This study brings light into the role played by the immune responses at the maternal-fetal inteface in the clinical outcome or in the control of parasite transmission and proliferation in placental and foetal tissues and provide new insight into the host-pathogen interactions, validating at the same time the results obtained in vitro. Differences in the outcome of the infection with high- and low-virulence isolates were found. Fever was the first clinical sign associated with the infection. Nc-Spain7-infected animals presented a biphasic increase of the body temperature, with two peaks at 1 and 3 dpi, and 100% of animals were febrile. Differences in the febrile responses induced by high- and low-virulence isolates may be explained by a more efficient multiplication of Nc-Spain7 as was previously indicated and confirmed in placental cultures in vitro. Early seroconversion and higher levels of IFN-γ in sera, detected by ELISA, were also related to the high virulent isolate. However, these levels did not correlate with protection against the parasite since earlier and wider dissemination of Nc-Spain7 (10 dpi) and higher parasite burden (detected by PCR and qPCR respectively), and more severe lesions in placental and foetal samples (observed by microscopic histological examination), were found in Nc-Spain7-infected animals than in Nc-Spain1H. In addition, 40% of foetal death was detected in the heifers infected with Nc-Spain7 when they were culled at 20 dpi. Remarkably, the modulation of the local immune responses and the ECM of the placenta were also influenced by the parasite isolate. Elements of the innate and adaptative immune responses as pattern recognition receptors (PRRs), cytokines, chemoattractant genes and ECM components were studied in placental samples by immunohistochemistry (IHC) and qPCR. Nc-Spain7 was not detected by the placental PRRs traditionally implicated in intracellular parasite recognition at 10 dpi. Subsequently, this isolate did not induce upregulation of any investigated gene except for Chapter I ~ Summary 5 IFN-γ, IL-4 and TNF-α. However, parasite replication and lesion development at 20 dpi probably led to PRRs activation and exacerbated expression of cytokines (mainly IFN-γ, iNOS, TNF-α and IL- 12p40) and chemokines, with a Th1/Th2 balance displaced towards a predominant Th1 response, that probably contributes to foetal death. Moreover, IL-8, TNF-α and iNOS were upregulated and TGF-β1 downregulated in placental tissues from non-viable foetuses, suggesting that they may be directly related to N. caninum abortion. Expression of cytokines TGF-β1, IL-6 and IL-17A in bovine placenta was not upregulated by Nc-Spain7 infection at any time, indicating the lack of activation of certain repair processes. In addition, a profound alteration of the ECM, with decreased expression of MMP-2 and TIMP-2 and destruction of collagen, fibronectin and vimentin could contribute to the abortion mechanism. On the other hand, the low virulent isolate Nc-Spain1H induced upregulation of PRRs, cytokines, chemoattractant genes and ECM modulators as early as 10 dpi. This modulation was maintained at 20 dpi at constant levels, maintaining Th1/Th2 responses balanced. Rapid immune responses together with a minor replication ability of Nc-Spain1H may be the causes of the low burden found. In addition, a profound modulation of the ECM by Nc-Spain1H has been evidenced in the placenta in vivo. Altogether, different mechanisms seem to be implicated in high- and low-virulence isolates transmission. We propose that Nc-Spain7 arrives to the foetus using its efficient replication ability of and an evasion strategy of the placental immune response in the early stage of the infection. However, high multiplication leads to placental damage and exacerbated immune responses that may be the causes of the abortion. On the other hand, the low-virulence isolate is quickly recognized and controlled by the placental immune responses, which along with the lower proliferation of this isolate, mainly in caruncular cells, suggest a different mechanism to cross the placental barrier. Modulation of the ECM and the hijacking of the immune cells may be the strategies used by this parasite in order to be transmitted to the offspring avoiding placental damage. Chapter I ~ Resumen 7 La neosporosis bovina es una de las causas más importantes de aborto en el ganado vacuno a nivel mundial. Dicha enfermedad está causada por el protozoo apicomplejo Neospora caninum, un parásito intracelular obligado formador de quistes tisulares, estrechamente relacionado con Toxoplasma gondii. La infección con este parásito causa cuantiosas pérdidas económicas en todo el mundo, principalmente asociadas al fallo reproductivo. N. caninum es capaz de cruzar la barrera placentaria de forma muy eficaz y el daño producido en la placenta al transmitirse al feto es crucial en la patogénesis del aborto. Además, la infección durante la gestación representa un desafío para el sistema inmunológico de los animales infectados. De hecho, desequilibrios en el balance de la respuesta de tipo Th1 y Th2 en la placenta durante la gestación se han asociado con el aborto. Las células del trofoblasto bovino junto con las células epiteliales de la carúncula componen la interfaz materno-fetal de los placentomas, desempeñando un papel fundamental en la funcionalidad de la placenta. Además de su papel como barrera física, impidiendo el paso de sustancias tóxicas y agentes patógenos de la madre al feto, estas células son capaces de reconocer e inducir respuestas inmunitarias frente a N. caninum, participando en la iniciación de respuestas inmunitarias innatas a nivel de la placenta, así como en el desarrollo de una respuesta inmunitaria adaptativa. Sin embargo, las causas que conducen al aborto y los mecanismos por los cuales N. caninum infecta la placenta y se transmite al feto siguen siendo desconocidos. A pesar de que la placenta bovina desempeña un papel fundamental en la patogénesis de la neosporosis y de que la invasión de la placenta, junto con la proliferación del parásito en ésta, y su diseminación al feto parecen ser eventos cruciales en el desarrollo de esta enfermedad, hasta el día de hoy se han abordado pocos estudios centrados en las interacciones tempranas entre las células hospedadoras y la placenta. De hecho, se dispone de un único estudio in vitro, descriptivo y muy limitado, en células multinucleadas de trofoblasto bovino. Además, el número de investigaciones in vivo explorando las consecuencias de la infección por N. caninum a mitad de la gestación, cuando se producen la mayoría de los abortos en vacas infectadas naturalmente, es muy limitado. Finalmente, la variabilidad intraespecífica del parásito influye en el resultado de la infección. Sin embargo, los estudios realizados comparando aislados de distinta virulencia son escasos, y ningún estudio abordado anteriormente había comparado aislados de distinta virulencia en células diana de la placenta bovina en las primeras etapas de la infección. Teniendo en cuenta lo expuesto previamente, en la presente Tesis Doctoral se ha investigado el papel de la placenta en la patogenia de la neosporosis bovina y la influencia de la variabilidad biológica del aislado en el daño placentario. El primer objetivo de la presente Tesis Doctoral consistió en la caracterización de la interacción parásito-hospedador entre aislados de alta y baja virulencia de N. caninum y células diana de la placenta bovina in vitro. El fin de dicho objetivo fue identificar los eventos ocurridos en la placenta y el papel desempeñado por las células que conforman la interfaz materno-fetal después de la infección con aislados de distinta virulencia, así como elucidar los mecanismos utilizados por el parásito para atravesar la placenta y los factores que permiten que algunos aislados se transmitan de manera más efectiva que otros, causando la muerte fetal. Para alcanzarlo, en primer lugar, se estudiaron los mecanismos de adhesión, invasión, proliferación y egresión de los taquizoítos de los aislados Nc-Spain7 (de alta virulencia) y Nc-Spain1H (de baja virulencia) en cultivos establecidos de células epiteliales carunculares bovinas (BCEC-1) y trofoblásticas (F3) mediante ensayos de inmunofluorescencia en placa y de PCR en tiempo real (qPCR). N. caninum invadió y proliferó en ambas líneas celulares. Sin embargo, la invasión, la infección y la proliferación de Chapter I ~ Resumen 8 ambos aislados fueron menores en BCEC-1 que en F3, lo que sugiere que las células carunculares maternas son resistentes a la infección por N. caninum y dejan entrever el papel de barrera que tendría esta línea celular en la interfaz materno-fetal. Además, los taquizoítos de ambos aislados mostraron mecanismo de egresión temprana en BCEC-1, lo que demuestra que a pesar de que estas células pueden restringir la adhesión y la multiplicación del parásito, no pueden evitar su transmisión. Por otro lado, el aislado más virulento, Nc-Spain7, presentó una mayor capacidad de invasión y de infección que el aislado menos virulento, Nc-Spain1H, principalmente en células trofoblásticas, así como una mayor capacidad de proliferación. Las vacuolas parasitóforas alcanzaron un gran tamaño en las células del trofoblasto bovino, contrariamente a lo observado en las células carunculares, lo que podría indicar una importante implicación de los trofoblastos en la infección por N. caninum, siendo probablemente el nicho de replicación del parásito. Después de estudiar los eventos del ciclo lítico, se investigó la modulación de las respuestas inmunitarias del trofoblasto y la carúncula in vitro, mediante la extracción de ARN y RT-qPCR, con el objetivo de establecer el papel de las células de la placenta en la regulación de la respuesta inmunitaria innata y su reprogramación por la infección con aislados de distinta virulencia de N. caninum. Se investigaron los perfiles de expresión del receptor toll-like 2 (TLR-2), de citoquinas de tipo Th1 y Th2 (IL-4, IL-10, IL-8, IL-6, IL-12p40, IL-17, IFN-γ, TGF-β1, TNF-α), así como de moléculas de adhesión endotelial (ICAM-1 y VCAM-1), encontrándose una regulación similar en ambas líneas celulares. Dicha modulación consistió en un aumento de expresión del TNF-α y la IL-8 y en un descenso de la IL-6 y del TGF-β1 en los cultivos infectados. La secreción proteica de IL-6, IL-8 y TNF-α fue confirmada por ELISA. Por lo tanto, se determinó que la infección por N. caninum favoreció una respuesta proinflamatoria en las células diana de la placenta bovina in vitro. Cabe destacar que las comparaciones entre los aislados mostraron una mayor expresión del TLR-2 y del TNF-α en trofoblastos infectados con el aislado de baja virulencia Nc-Spain1H, lo que sugirió mecanismos de modulación distintos con cierta evasión de la respuesta inmunitaria por parte del aislado más virulento. Esto podría suponer una explicación, al menos parcial, a las diferencias biológicas observadas entre los aislados. El estudio de la infección temprana en un modelo bovino gestante inoculado con aislados de alta y baja virulencia de N. caninum en el segundo tercio de la gestación supuso el segundo objetivo de la presente Tesis Doctoral. La interacción parásito-hospedador, especialmente en las primeras etapas de la infección, así como los factores de virulencia intrínsecos asociados con el aislado implicado son cruciales en el resultado de la infección. Para lograr este objetivo se realizó una infección experimental mediante la inoculación intravenosa de 107 taquizoítos de los aislados Nc- Spain7 o Nc-Spain1H en novillas gestantes, y el posterior sacrificio de los animales en dos momentos tempranos tras la infección (10 y 20 días después de la infección - dpi). El análisis comparativo de la dinámica temprana de la infección (resultado clínico, distribución y carga parasitaria y desarrollo de lesiones en tejidos placentarios y fetales), de la respuesta humoral específica y de la modulación precoz de las respuestas inmunitarias de la placenta y la regulación de la matriz extracelular (MEC) ponen de manifiesto el papel que desempeñan las respuestas inmunitarias de la interfaz materno-fetal en el resultado clínico o en el control de la transmisión y proliferación del parásito en los tejidos placentarios y fetales, y proporcionan una nueva visión de las interacciones parásito-hospedador, validando al mismo tiempo los resultados obtenidos in vitro. Chapter I ~ Resumen 9 Se observaron diferencias en el resultado de la infección con aislados de alta y baja virulencia de N. caninum. El aumento de la temperatura corporal fue el primer signo clínico asociado con la infección. El 100% de los animales infectados con el aislado Nc-Spain7 presentaron fiebre, con una dinámica bifásica consistente en dos picos de fiebre a los 1 y 3 dpi. Las diferencias en las respuestas febriles inducidas por los aislados de alta y baja virulencia pueden explicarse por una multiplicación más eficiente del aislado más virulento, como ya fue indicado previamente y confirmado por los resultados de la presente Tesis Doctoral en cultivos de placenta bovina in vitro. La seroconversión temprana y los niveles más altos de IFN-γ en suero, detectados por ELISA, también se relacionaron con el aislado más virulento. Sin embargo, estos niveles no se correlacionaron con la protección frente al parásito, ya que en animales infectados con el aislado Nc-Spain7 se observó una mayor y más temprana diseminación del parásito (10 dpi), así como mayores cargas parasitarias, detectadas por PCR y qPCR respectivamente, además de lesiones más graves en muestras de placenta y fetos, observadas mediante examen histológico, que en los animales infectados con el aislado Nc-Spain1H. También se detectó un 40% de mortalidad fetal en las novillas infectadas con el aislado Nc-Spain7 cuando fueron sacrificadas a los 20 dpi. La modulación de la respuesta inmunitaria y de la MEC de la placenta también fueron influenciadas por el aislado del parásito. Elementos de la respuesta inmunitaria innata y adaptativa, como receptores de reconocimiento de patrones (PRRs), citoquinas, quimioquinas y componentes de la MEC, fueron estudiados en muestras de placenta mediante inmunohistoquímica y qPCR. El aislado Nc-Spain7 no fue detectado a los 10 dpi por los receptores de la placenta tradicionalmente implicados en el reconocimiento de parásitos intracelulares. En consecuencia, este aislado no indujo un aumento de la regulación de ningún gen investigado, salvo el IFN-γ, la IL-4 y el TNF-α. Sin embargo, la replicación del parásito y el desarrollo de lesiones observado a los 20 dpi condujeron probablemente a la activación de los receptores de reconocimiento en la placenta y a un aumento exacerbado de la expresión de citoquinas (principalmente IFN-γ, iNOS, TNF-α e IL- 12p40) y quimioquinas. El equilibrio de las respuestas de tipo Th1/Th2 fue desplazado hacia una respuesta predominantemente Th1, que probablemente contribuyó a la muerte fetal. Además, la expresión de IL-8, TNF-α e iNOS se encontró aumentada y la expresión del TGF-β1 disminuida en los placentomas de fetos no viables a los 20 dpi, lo que sugiere una relación directa entre la modificación de estos genes y la presencia de aborto inducido por la infección de N. caninum. La infección con el aislado virulento no aumentó la expresión de las citoquinas TGF-β1, IL-6 e IL-17A en la placenta bovina en ningún momento tras la infección, lo que indica la falta de activación de ciertos procesos de reparación tissular. Además, una alteración profunda de la MEC, con disminución de la expresión de MMP-2 y TIMP-2 y destrucción de colágeno, fibronectina y vimentina podría contribuir al mecanismo del aborto. Por otro lado, el aislado de baja virulencia Nc-Spain1H indujo un aumento en la expresión de los receptores de reconocimiento, citoquinas, quimioquinas y moduladores de la MEC en un estadio muy temprano de la infección (10 dpi). Esta modulación se mantuvo a los 20 dpi a niveles más o menos constantes, manteniendo equilibradas las respuestas Th1/Th2 a lo largo de todo el periodo experimental. La activación temprana de la respuesta inmunitaria, junto con la menor capacidad de replicación, pueden ser las causas de la baja carga parasitaria encontrada en los animales infectados con el aislado Nc-Spain1H. Además, el aislado Nc-Spain1H indujo una profunda modulación de la MEC de la placenta in vivo. El conjunto de todos los resultados observados sugiere un mecanismo de transmisión diferente en aislados de alta y baja virulencia. El aislado más virulento, Nc-Spain7, podría llegar al feto utilizando su eficiente capacidad de replicación y un mecanismo de evasión de la respuesta Chapter I ~ Resumen 10 inmunitaria de la placenta en la fase inicial de la infección. Sin embargo, la eficiente multiplicación del parásito provocaría daño tisular en la placenta, lo que podría ser la causa de la respuesta inmunológica exagerada, que, a su vez, podría ser una causa desencadenante del aborto. Por otro lado, el aislado de baja virulencia es rápidamente reconocido y controlado por la respuesta inmunitaria de la placenta, lo que unido a la menor proliferación de este aislado, principalmente en las células carunculares, sugieren un mecanismo diferente para cruzar la barrera placentaria. La modulación de la MEC y el secuestro de células inmunitarias podrían ser las estrategias utilizadas por este parásito para ser transmitido evitando el daño placentario. CHAPTER II INTRODUCTION INTRODUCCIÓN Chapter II ~ Introduction 13 1. Neospora caninum and bovine neosporosis Bovine neosporosis is a parasite disease that causes reproductive failure in cattle and is considered one of the main transmissible causes of abortion in this species leading to significant economic losses both in the dairy and the meat industry worldwide (Reichel et al., 2013). Since Neospora caninum, etiologic agent of bovine neosporosis, was recognised for the first time in dogs with severe nervous disorders in Norway (Bjerkǻs et al., 1984), this parasite has been deeply studied and neosporosis has emerged as a serious disease of cattle worldwide. The interest by N. caninum has increased in the veterinary field along the years, having conducted numerous investigations on the biology of the parasite, as well as on the diagnosis, epidemiology and control of the disease (Dubey et al., 2007; Dubey & Schares, 2011; Ortega-Mora et al., in press). It was not until 1988 when Dubey et al. proposed the description of a new genus, Neospora, and species, N. caninum, after identifying the parasite in dogs that presented neuromuscular clinical signs identical to those observed in 1984 (Dubey et al., 1988a). In 1989, N. caninum was described as the etiologic agent of bovine neosporosis after the observation of N. caninum-like protozoan organisms in the brain and the kidney of necropsied foetuses coming from aborted Holstein dairy cattle in New Mexico (Thilsted & Dubey, 1989). In subsequent years, N. caninum infection was described and associated with reproductive failure in cattle in many countries (Anderson et al., 1991; Barr et al., 1991b; Dubey & Lindsay, 1996). Transplacental transmission in the bovine species was experimentally demonstrated in 1992 by inoculation of N. caninum tachyzoites in three cows and detection of the parasite in two aborted foetuses (Dubey et al., 1992). One year later, Conrad et al. (1993a) obtained the first isolate of N. caninum from an aborted bovine foetus, that was used to experimentally infect pregnant cows and reproduce typical clinical signs of bovine neosporosis (Barr et al., 1994). The definitive host for N. caninum was not discovered until 1998 (McAllister et al., 1998), when the presence of N. caninum oocysts in dog faeces was evidenced. This finding was later confirmed by Lindsay et al. (1999), completing the biological cycle of the parasite. In 1998, Neospora hughesi, a second species of Neospora, was isolated from the nervous tissue of an adult horse with nervous clinical signs (Marsh et al., 1996; Marsh et al., 1998), and several authors have indicated structural, antigenic, genetic and pathogenic differences between the two species of the genus Neospora (Marsh et al., 1998; Walsh et al., 2001). 1.1 Taxonomic classification and morphology of the parasite Closely related to the zoonotic apicomplexan parasite Toxoplasma gondii, N. caninum is an obligate intracellular protozoan belonging to the subphylum Apicomplexa (Adl et al., 2012), which includes more than 6.000 species that parasitize practically all groups of animals. The organisms classified within this subphylum are characterized by the presence of an apical complex. More specifically, N. caninum belongs to the Sarcocystidae family because of the formation of tissue cysts in the hosts, the presence of heteroxeneous biological cycles with phases of sexual and asexual replication and the oocysts sporulation in the environment (Tenter et al., 2002). The taxonomic classification of N. caninum and its main characteristics are summarized in Figure 1. In addition to N. caninum, which mainly affects ungulates (Dubey et al., 1988a), the species N. hughesi also belongs to the Neospora genus. The exact phylogenetical relationship of N. caninum Chapter II ~ Introduction 14 to other members of the Apicomplexa has been, and still is, under controversy (Dubey et al., 2002; Heydorn & Mehlhorn, 2002). Figure 1. Taxonomic classification of N. caninum Up to date, three invasive stages of the parasite have been described: tachyzoites, bradyzoites within tissue cysts and sporozoites within oocysts (Figure 2). Tachyzoites and tissue cysts containing bradyzoites are the invasive stages located in intermediate host´s tissues whereas unsporulated oocysts are eliminated with definitive host´s faeces, sporulating in the environment and becoming infective. The tachyzoite is the fast-replicating stage of the parasite being responsible for the acute phase of infection and the damage in the intermediate host (Dubey et al., 2006). Progressive cycles of intracellular replication, tachyzoite release by egression with host cell lysis and infection of surrounding cells, together with related immunopathologic repercussions lead to lesion formation and, in some animals, clinical disease (Dubey et al., 2007). The tachyzoites are able to invade a wide variety of cell types such as neurons, macrophages, dendritic cells (DC), fibroblasts, endothelial cells, myocytes and hepatocytes, allowing a rapid and effective intra-organic dissemination (Barr et al., 1993; Dubey et al., 2002). In addition, the tachyzoite is the most studied and widely used stage of the parasite since its maintenance and growth in vitro is relatively simple. The bradyzoite is the slow-replicating stage of the parasite, responsible for the chronic phase of the infection in the intermediate host. After rapid multiplication of the tachyzoites, they differentiate into bradyzoites, the quiescent life stage of the parasite that forms under host immune pressure and produces a tissue cyst in the intracytoplasmic space of a single cell (Goodswen et al., 2013). Tissue cysts are mainly located in immunopriviledged tissues as central nervous system (CNS) and skeletal muscle tissue (Barr et al., 1991b; Dubey & Lindsay, 1996; Peters et al., 2001; Dubey, 2003), and facilitate long term parasite persistence and chronic Chapter II ~ Introduction 15 asymptomatic infections (Hemphill et al., 2006; Dubey et al., 2007; Dubey & Schares, 2011). Bradyzoite stage may switch to tachyzoite stage under some conditions such as immunosuppression or immunomodulation, situation that is well documented to occur in pregnant animals and permits tachyzoite spread to other tissues, including dissemination across the placenta and infection of the foetus (Williams et al., 2009). Finally, sporozoites are contained within the oocyst, which is released to the environment in the faeces of a definitive host and supposes the environmentally resistant form of the parasite. The oocyst sporulates in the environment after approximately 24 hours under favorable conditions of temperature and humidity and constitutes the infective phase for the intermediate host, being responsible for the horizontal transmission between a definitive and an intermediate host. Therefore, oocysts are the key in the epidemiology of neosporosis, but there are only a few reports of N. caninum oocyst shedding by naturally infected dogs (reviewed by Dubey et al., 2017). The sporulated oocyst is composed by two sporocysts with four sporozoites each and may contain a residual sporocystic body (McAllister et al., 1998; Lindsay et al., 1999; Dubey et al., 2002). Although the presence of entero-epithelial stages of the parasite in naturally infected dogs has been demonstrated (Kul et al., 2015), the stages of merozoite and gametocyte, widely studied in T. gondii, have not been well characterized in N. caninum (Dubey et al., 1998; Hehl et al., 2015). Figure 2. Graphical representation and microscopic images of N. caninum parasite stages Ultrastructurally, all three zoites present, in addition to the typical organelles of any eukaryotic cell, all the characteristic structures of the apicomplexan parasites (Figure 3). The outermost layer of the zoites is the membrane, composed by an external layer that constitutes the cytoplasmatic membrane, and two internal ones which interact with the glideosome, structure involved in the invasion of the host cell (Boucher & Bosch, 2015), and the subpellicular microtubes (Ouologuem & Roos, 2014). The subpellicular microtubules are anchored to the polar rings, acting as a cytoskeleton and allowing the movement of the zoite together with the conoid (Speer & Dubey, 1989; Lindsay et al., 1993; Speer et al., 1999). In the cytoplasm of apicomplexan parasites rhoptries, micronemes, dense granules and amylopectin granules (polysaccharide reserve) are found and their number and distribution confer taxonomic value. The rhoptries have a main function in the formation and maturation of the parasitophorous vacuole within the host cell whereas micronemes are small vesicles that secrete proteins, participating in the recognition and adhesion to the host cell (Hemphill et al., 2013). The apical complex is set by conoid, polar rings, Chapter II ~ Introduction 16 micronemes and rhoptries (Dubey et al., 1988a). The dense granules participate, by means of secretion products, both in the formation and in the maintenance of the parasitophorous vacuole inside the host cell (Hemphill, 1999). Certain differences in the arrangement and number of the secretory organelles are found between tachyzoite and bradyzoite stages (Speer & Dubey, 1989; Dubey et al., 2004). It should be noted the presence of a vestigial apicoplast, present in most of apicomplexan parasites. This apicoplast have not photosynthetic capacity, but it has importance as a pharmacological target since it contains genetic material codifying for essential elements of the parasite's metabolism such as fatty acids synthesis (Sheiner et al., 2013). Figure 3. Graphic representation of the ultra-structure of the tachyzoite of N. caninum, with special mention of the cytoskeleton (left) and cytoplasm (right) components found only in apicomplexan protozoa. Source Iván Pastor Fernández, Doctoral Thesis, 2015 1.2 Life cycle and transmission N. caninum is characterized by a complex facultative heteroxenous life cycle (Figure 4) that involves a definitive canid host and an intermediate host in which sexual and asexual replications take place, respectively. Up to date, the confirmed definitive hosts of N. caninum are the dog, the coyote, the dingo and the wolf. By resemblance to T. gondii, it is assumed that oocysts are generated in the intestinal tract of the definitive host after schizogony and gametogony and expelled in the faeces in an unsporulated (noninfectious) form. Oocysts become orally infectious after sporulation outside the host (reviewed by Dubey et al., 2017). The main intermediate host is cattle, although the exposure of a wide variety of domestic and wild animals to N. caninum has been shown in several studies. Viable tachyzoites have been isolated from the cow, the sheep, the water buffalo, the dog, the wolf, the horse, the bison and the white-tailed deer, although direct (parasite identification) or indirect (antibody responses) evidences of infection have been described in a range of other species. However, there is no evidence that N. caninum can infect humans because only low levels of antibodies have been Chapter II ~ Introduction 17 reported, and neither N. caninum DNA nor parasite presence has been demonstrated in human tissues (Dubey et al., 2007; McCann et al., 2008; Calero-Bernal et al., 2019). After the ingestion of the oocysts, sporozoites are released in the gastrointestinal tract infecting the entero-epithelial cells. Inside, they switch to tachyzoites, which reach the bloodstream and produce the intra- organic spread of the parasite thanks to their ability to invade a wide range of organs and tissues. Finally, host immune responses lead to tachyzoite conversion, remaining the bradyzoites in immunopriviledged tissues (reviewed by Dubey et al., 2017). Predation of tissues infected with cysts containing bradyzoites is presented as the only way of transmission from the intermediate to the definitive host, thus closing the biological cycle. In the intermediate host, apart from the horizontal route, vertical or transplacental transmission can also occur when tachyzoites cross the placenta in a pregnant cow and invade the foetus, being the most efficient transmission route and playing a fundamental role in the propagation and maintenance of the disease in a herd (reviewed by Dubey et al., 2017). Depending on the origin of the infection, two types of transplacental transmission are recognized in cattle: exogenous transplacental transmission (ExTT) and endogenous transplacental transmission (EnTT) (Trees & Williams, 2005). Pathogenic and epidemiological consequences as well as the control measures for both are different. ExTT occurs after ingestion of sporulated oocysts by naïve cattle and is associated with epidemic abortion storms in a herd (Williams et al., 2009). EnTT follows recrudescence of infection in a persistently infected cow during pregnancy and is the main mechanism responsible for maintaining the parasite within cattle populations, resulting in foetal transmission rates as high as 95% (Dubey et al., 2006; Williams et al., 2009; Reichel et al., 2013). Although vertical transmission of the pathogen predominates in cattle, some works have showed a decrease in the vertical transmission rates with age and absence of transplacental transmission in some cases, suggesting a higher importance of horizontal transmission than traditionally assumed (French et al., 1999; Dubey et al., 2007; Eiras et al., 2011). In this sense, EnTT appears as the predominant mode of transmission in many herds, while controversy exists regarding the relevance of ExTT to give rise to a chronic infection (McCann et al., 2007; Dijkstra et al., 2008). To date, only ExTT has been achieved experimentally in pregnant cows after infection with tachyzoites or oocysts (Williams et al., 2000; Innes et al., 2001; McCann et al., 2007). However, EnTT in postnatally infected animals has been observed in natural infections (Dijkstra et al., 2008). In Spain, different epidemiological scenarios have been described in which ExTT and EnTT can occur independently or in combination (Rojo-Montejo et al., 2009c). Spite of the ingestion of sporulated N. caninum oocysts is the only demonstrated natural mode of infection in cattle after birth, other forms of transmission have been postulated. The presence of N. caninum-DNA in milk, including colostrum, has been demonstrated. However, there is no conclusive evidence that lactogenic transmission of N. caninum occurs in the nature. Venereal transmission may be possible, but unlikely, as evidenced recently. Finally, studies indicate that N. caninum life cycle is maintained in a domestic cycle between dogs and cattle (reviewed by Dubey et al., 2017). Chapter II ~ Introduction 18 Fi gu re 4 . L if e cy cl e an d t ra n sm is si o n o f N . c a n in u m Chapter II ~ Introduction 19 Compelling evidence suggests a sylvatic cycle involving domestic and wild canids, and ruminant and herbivore species in North America (McAllister et al., 1998; Barling et al., 2000; Gondim et al., 2004b; Gondim, 2006). In Europe, similar sylvatic cycles involving wild ruminants and canids seem plausible but currently little evidence exists to support this hypothesis. 1.3 Prevalence and economic impact Differences in prevalence of N. caninum infection between countries and regions, and between different production systems were found. Despite the problems found for comparing available studies, the lowest seroprevalences in Europe were found in Sweden and Germany, whereas Holland and Spain showed moderate to high seroprevalences (Bartels et al., 2006). South America, Argentina and Brazil presented the highest seroprevalences (80.9-97.2%) (reviewed by Dubey et al., 2017). The first studies carried out in Spain showed a high herd prevalence, 55% and 83% in beef and dairy cattle, respectively, and individual prevalence values of 18% in beef and up to 36% in dairy cattle (Mainar-Jaime et al., 1999; Quintanilla-Gozalo et al., 1999). These data were similar to those found in later studies carried out in Galicia (northwestern region of Spain), where herd seroprevalences reached 88% in dairy cattle and 77% in beef cattle, and individual seroprevalences were stablished between 16% and 22% in dairy herds, and in 25% in beef herds (González-Warleta et al., 2008; Eiras et al., 2011). In relation to abortion, between 12% and 40% of the aborted foetuses are infected by N. caninum worldwide (Dubey et al., 2007). Two studies carried out in Spain revealed infection in 58% (González et al., 1999) and in 39% of analyzed foetuses (Pereira-Bueno et al., 2003), bringing to light the importance of N. caninum as an abortive agent in our country, even ahead of the virus of bovine viral diarrhea (BVDV). Therefore, bovine neosporosis has been recognized as one of the main causes of reproductive failure worldwide, causing a negative economic impact on livestock production derived, directly, from the abortion and neonatal mortality and, indirectly, from the increase in the calving interval, the reduction of milk production and growth rate, the loss of genetic value of the animals and the increase of veterinary costs (reviewed by Dubey et al., 2017). Economic losses seem to be more important in dairy industry due to a higher risk of abortion. A recent study focused on the major livestock powers, Spain among them, estimated the impact of the disease at a billion dollars per year. National cost in Spain reached 19.8 and 9.8 million dollars in dairy and beef cattle, respectively (Reichel et al., 2013). 1.4 Pathogenesis, clinical signs and lesions Despite being one of the most important abortifacient agents in cattle, the final mechanisms triggered by N. caninum which lead to the foetal death are not completely understood. Regardless the origin of the infection, N. caninum tachyzoites replicate and disseminate throughout the host tissues, arriving to the placenta, where they can multiplicate, cross the maternal-foetal interface and reach the foetus (Dubey et al., 2006). Four mechanisms have been Chapter II ~ Introduction 20 postulated so far as possible cause of abortion: tachyzoites multiplication in vital organs of the foetus; placental damage induced by parasite replication, with the consequent alteration of the placental functions; imbalance of the immune responses at the placenta; and prostaglandin secretion (Buxton et al., 2002; Innes et al., 2005; Innes, 2007). While infection with N. caninum in non-pregnant cattle is generally asymptomatic, abortion is the most relevant clinical sign in pregnant animals. Abortion may occur from the third month of pregnancy, although it is usually observed between 5 and 6 months of gestation (Almería & López- Gatius, 2013). If infection occurs in the first third of gestation, the foetus is usually reabsorbed and heat repetition is observed. If the foetal death occurs between 3 and 8 months, the foetus is usually eliminated presenting moderate autolysis, although dead foetuses before the fifth month could be mummified and retained in the uterus for months. However, infection from the fifth month usually leads to the birth of healthy but congenitally infected calves that, generally, will present precolostral antibodies (Quintanilla-Gozalo et al. al., 2000; Williams et al., 2000). It is not very common the birth of very weak calves with neurological signs. When those appear, they may range from mild incoordination to complete paralysis. In severe cases, spine malformations, hydrocephalus or pneumonia can be observed. These signs usually appear in the first week after parturition, although they may appear until two weeks after birth and evolve to total paralysis and death of the animal during the first month of life (reviewed by Dubey et al., 2017). Affected tissues of the host present microscopic lesions which facilitate the diagnosis of abortion. The lesions, mainly focused on placental and foetal tissues, are always inflammatory and non- suppurative. In the placenta, foci of necrosis and areas of intense inflammation with infiltration of mononuclear cells are usually observed, which, in advanced processes, can progress towards conjunctive regeneration with hyperplasia, fibrosis and even calcification of the necrotic foci (Barr et al., 1994; Maley et al., 2003). These infiltrations begin first in the maternal caruncles and then spread to the foetal cotyledons, with the appearance of areas of hemorrhage and necrosis. Separation of caruncle and cotyledon is often observed with release of serum into the maternal- foetal septum. In general, lesions are more severe when infection occurs earlier (Collantes- Fernández et al., 2006a; Collantes-Fernández et al., 2006c; Gibney et al., 2008). In foetuses with congenital infection, the lesions are mainly located in the CNS, with multifocal areas of lymphocytic infiltration, perivascular cuffs, microgliosis, presence of astrocytes and areas of necrosis (Barr et al., 1991a; Dubey & Lindsay, 1996). Foci of non-suppurative inflammation have also been reported in the heart and in the liver, and, less frequently, in the kidney, the skeletal muscle and the lung. Interestingly, lesions were more severe when the abortion occurred after ExTT and during the first and second third of gestation (Dubey et al., 2006; Collantes- Fernández et al., 2006a; Collantes-Fernández et al., 2006c). Congenitally infected calves with nervous signs usually have inflammatory lesions at the spinal level and the brain (Peters et al., 2001). In contrast, the presence of lesions in congenitally infected calves without clinical signs and in adult animals is unusual and restricted to the CNS (reviewed by Dubey et al., 2017). Chapter II ~ Introduction 21 1.5 Diagnosis and control • Diagnosis of neosporosis Due to the lack of vaccines and effective treatments, the best control strategy for neosporosis is based on diagnosis. However, the diagnosis of bovine neosporosis is difficult and an adequate and systematic approach is needed. Serological testing of animals is sometimes necessary prior to entry into the herd or as a preliminary step to restocking. In abortion cases, epidemiological data and clinical history may suggest the involvement of N. caninum as a cause, but a correct laboratory diagnosis is always required. Currently, there is a large battery of diagnostic techniques, although not all them provide the same information or have the same reliability. Different strategies should be approached depending on the subject, and the choice of the most appropriate diagnostic tool for each case as well as its correct interpretation play a fundamental role (Ortega-Mora et al., 2006). Whether diagnosis is approached in foetuses, samples of the foetus (brain, heart, liver) and placenta are necessary to investigate histological lesions and presence of the parasite by polymerase chain reaction (PCR). The detection of specific or compatible lesions on target organs is of great value as an initial diagnostic approach and allows the association between the lesions and the presence of the parasite. However, the PCR provides greater sensitivity and specificity than immunohistochemistry to detect the parasite. Thoracic and abdominal fluids from foetuses older than 5 months are also useful for the determination of antibodies by indirect immunofluorescence (IFAT), enzyme-linked immunosorbent assay (ELISA) or Western blotting (WB), since antibodies from maternal origin cannot cross the placental barrier (Ortega-Mora et al., in press). When diagnosis is carried out in individuals, serological diagnosis in both serum and milk samples can be used, which can also be analysed by IFAT, ELISA and WB. In order to demonstrate congenital transmission, selected samples should be precolostral serum or serum from calves older than six months, avoiding interference with colostral antibodies. If animals did not present abortion and purchase purposes are pursued, WB is the preferential technique since is more sensitive and specific. In these cases, sampling can be repeated after 4 weeks to observe seroconversion. After the abortion, there is an increase in antibody levels in many animals (Ortega-Mora et al., in press). Finally, if the objective is the serological diagnosis of the herd, the detection of antibodies in samples collected from the milk tank by ELISA is an optimal technique due to its low cost and speed. However, due to the dilution effect of the antibodies in the tank, it is only useful in herds with an intra-herd prevalence higher than 10-15%. Detection of antibodies in the milk tank is useful to know the initial state of the infection on a farm and its progress (surveillance). The detection of antibodies by ELISA or IFAT in aborted and unaborted cows allows the association of N. caninum infection with abortion by calculating the odds ratio. If the association between seropositivity and abortion is significant, we can ensure the involvement of neosporosis in abortion. The next step is to perform a serology on the different age groups on the farm. This information will be useful for the choice of the most effective control measures since we can have an idea of intra-herd seroprevalence, and we can also study the relationship between the seropositivity of related animals (mother-daughter) and know the predominant mode of Chapter II ~ Introduction 22 transmission within the herd. Finally, the avidity ELISA makes possible the differentiation of herds with a primoinfection and high risk of ExTT (low avidity in most animals) from herds with chronic infection and EnTT (high avidity) (Ortega-Mora et al., in press). In addition, differential diagnosis with other causes of abortion, both transmissible and non- transmissible, must be performed. • Control of N. caninum transmission All the information obtained through the diagnosis protocol is important for the implementation of a control program in the herd, which is the only measure available for controlling bovine neosporosis since effective vaccines and treatments are not currently available, although they have been postulated as potential alternatives for the future. However, the cost-benefit ratio of the control measures, the economic losses caused by the infection and the epidemiological situation of each farm (seroprevalence and mode of transmission) must be considered when stablishing a control program (reviewed by Dubey et al., 2017). o Management measures Farms with N. caninum free status must reinforce the biosecurity measures in order to maintain their sanitary status. Some basic biosecurity measures would include quarantine and serological analyses of new animals, control of dogs' access to the farms as well as other potential definitive hosts (wolf, fox), specially to food storage areas and feeding and drinking troughs, correct elimination of abortion wastes or animal tissues and control of wild reservoirs as rodents. Similarly, these practices should be implemented in those farms that have suffered an epidemic outbreak of abortions due to a primary infection and the consequent ExTT. Farms where the infection is endemic must implement control measures focused on reducing EnTT. In farms with very low prevalence gradual selective culling of infected animals, starting with aborted cows, is recommended. When prevalence rates are intermediate or high, selective culling is economically unfeasible. In such cases, progressive elimination of the infection can be considered by culling of aborting cows and selective replacement with daughters of seronegative cows. The offspring of seropositive cows may be used to meat production (Hall et al., 2005). In the case of seropositive animals of high genetic value, embryo transfer using seronegative cows as receptors or artificial insemination with meat breeds are possible alternatives (Dubey et al., 2007; Reichel et al., 2013). In any case, serological monitoring at the herd level should be used as an indicator of the effectiveness of these control programs. o Vaccination Vaccination against bovine neosporosis has proven to be the most effective control measure from an economic point of view (Reichel et al., 2013). However, developing a vaccine designed to prevent both abortion and transplacental transmission supposes a big challenge. A commercial inactivated vaccine against bovine neosporosis (Bovilis NeoguardTM) was available for several Chapter II ~ Introduction 23 years in America and Oceania (Barling et al., 2003). However, it was withdrawn from the market later, when some studies showed large differences in efficacy at the farm level and a lack of protection against vertical transmission (Weston et al., 2012). Live vaccination has delivered promising results and the best immune protection to date in experimental infections in cattle. However, there are still no commercial live vaccines since their large-scale production is difficult (Reichel et al., 2015). Live vaccines include the use of in vitro attenuated isolates, genetically modified isolates or live isolates of low virulence such as Nc- Spain1H. On the other hand, inactivated vaccines, constituted by dead tachyzoites or antigenic extracts of the tachyzoites, have emerged as interesting alternatives to live vaccines, enhancing safety and stability of the final product and reducing costs. However, results in terms of protection against neosporosis have been negative or ambiguous (reviewed by Dubey et al., 2017). Finally, new generation vaccines have been proposed as another long-term alternative, with advantages as immune response directed against specific targets or incorporation of specific markers that would differentiate between vaccinated and infected animals (DIVA) (Marugán- Hernández, 2017). Unfortunately, recombinant vaccines have shown very low efficacy compared to live vaccines in vivo. In this scenario, new approaches as the combination of different antigens, use of improved adjuvants and enhancers of the immune response (e.g. fusion to Toll-like receptor (TLR) ligands) or combinations of live parasites with new drugs are being tested (reviewed by Dubey et al., 2017). o Chemotherapy and chemoprophylaxis Although there is not an effective and safe treatment against neosporosis, recent research discovered promising compounds in vitro and in vivo (Müller & Hemphill, 2011). Diamido derivatives (Debache et al., 2011; Schorer et al., 2012), artemisinin (Mazuz et al., 2012), miltefosine (Debache & Hemphill, 2012a), organometallic complexes of ruthenium (Barna et al., 2013), inhibitors of protein kinase (BKIs) (Ojo et al., 2014) or buparvaquone (Müller et al., 2015) are some of these drugs. However, the greatest advances have been achieved with the use of some sulfadiazines, such as toltrazuril, which had some efficacy in experimental models, decreasing transplacental transmission in mice (Gottstein et al., 2005; Strohbusch et al., 2009). Despite these good results, there are some doubts regarding their practical use in natural infections, the possible presence of residues in products for human consumption and their high costs. BKIs have been postulated as promising compounds against neosporosis and other apicomplexan parasites (reviewed by Dubey et al., 2017; Jiménez-Meléndez et al., 2017). In vitro and in vivo models of Neospora demonstrated an inhibition of the viability and proliferation of the parasite (Ojo et al., 2014), and partial protection against abortion and foetal dissemination in sheep (Sánchez-Sánchez et al., 2018). The development of alternative formulations and the use of different routes of administration, drug dosages and dosing regimens have been proposed for improving their efficacy (Sánchez-Sánchez et al., 2018). Chapter II ~ Introduction 24 2. Host-parasite interactions in bovine neosporosis: factors influencing abortion and transmission Bovine neosporosis is one of the main causes of abortion in cattle. However, abortion is not the only consequence of the infection, the birth of weak calves or healthy but congenitally infected calves may also occur. As specified above, the specific causes determining the occurrence of abortion have not been completely elucidated yet, and different mechanism have been proposed as the possible triggers of abortion. However, the existence of factors influencing the transmission of the parasite and abortion have been previously demonstrated. These factors depend on the host and the parasite, being the most relevant the gestation period, the foetal and maternal immune responses and the transmission route as a part of host-dependent factors, and the parasite stage, the inoculation route, the dose of infection and the intraspecific variability of the N. caninum isolates as a part of parasite-dependent factors. These factors have been investigated using both pregnant and non-pregnant mice and ruminant (bovine, sheep, goat and water buffalo) experimental models. For example, experimental infection in pregnant cattle at three different stages of gestation allowed to discern the influence of the gestation period in which the infection occurs by studying parasite distribution, parasite loads and lesions in aborted foetuses (Benavides et al., 2012; Bartley et al., 2012; Caspe et al., 2012; Bartley et al., 2013; Cantón et al., 2014a). Also, the influence of the breed, as another factor dependent on the host, has been studied by epidemiological surveys that suggested that certain breeds, mainly dairy (e.g. Holstein-Friesian), are more susceptible to infection than others (e.g. Limousin) (López-Gatius et al., 2005a; López-Gatius et al., 2005b; Armengol et al., 2007). However, latest studies suggested that differences observed between breeds might have been caused by differences in the risk factors associated with the different production systems and not by differences in breed-related susceptibility to infection (Eiras et al., 2011), although this must be experimentally studied. In addition to in vivo models, in vitro investigations have been useful to study the intraspecific variability of the isolates and new -omics techniques have increased the knowledge of the molecular basis of that virulence by means the comparative analyses of transcriptome, proteome and secretome, and have increased information on host-cell regulation events during infection between different virulence isolates. 2.1 Factors depending on the host 2.1.1 Gestation period The time during gestation when infection occurs seems to be critical to the pregnancy outcome (Williams et al., 2000; Innes, 2007). Experimental infections carried out within the first trimester of pregnancy (at 70 days of gestation) led to foetal loss generally associated with a multiplication or presence of the parasite in the placenta and the foetus, producing serious lesions that cause foetal death, with reabsorption, autolysis or mummification (Dubey et al., 1992; Barr et al., 1994; Williams et al., 2000; Macaldowie et al., 2004; Rojo-Montejo et al., 2009a; Regidor-Cerrillo et al., 2014). However, the birth of healthy calves without evidence of infection has also been described in some experimental infections using the subcutaneous (SC) route, thus a slower and controllable distribution of the parasite could be expected in this case, while other animals Chapter II ~ Introduction 25 infected with the same dose and isolate have aborted (Barr et al., 1994, Williams et al., 2000; Macaldowie et al . al., 2004; Gondim et al., 2004c; McCann et al., 2007; Caspe et al., 2012). Experimental infections carried out in the second third of gestation usually lead to the birth of healthy but congenitally infected calves. Sometimes neurological signs can be observed in the calf (Barr et al., 1994; Innes et al., 2001; Almería et al., 2003; Gondim et al., 2004c). Infection at mid- gestation can also cause abortion (Dubey et al., 1992; Almería et al., 2016; Vázquez et al., submitted), although lesions during this period are less severe than those observed after infection in early pregnancy (Maley et al., 2003). In fact, recent works using different parasite doses and administration routes at mid-gestation showed transplacental transmission and abortion (Vázquez et al., submitted). Finally, infection in the last third of gestation commonly leads to the birth of healthy calves although they are born congenitally infected (Barr et al., 1994; Innes et al., 2001; Williams et al., 2000; Maley et al., 2003; Cantón et al., 2013). Moreover, lesions and parasite loads are lower than those found in infections carried out in the first and second third of gestation (Gibney et al., 2008; Benavides et al., 2012). In naturally infected cattle, recrudescence of infection within the first half of gestation results in foetal death, whereas recrudescence later in gestation causes congenital infection but does not lead to the death of the foetus. However, most of naturally Neospora-induced abortions arise during mid-gestation, at five to six months (Dubey et al., 2007; Almería & López-Gatius, 2013; Almería & López-Gatius, 2015). The dead foetus may be reabsorbed, may suffer autolysis or mummification or may be expelled before or during the parturition depending on the development of the foetus at the moment of the death (Dubey, 2005). However, in naturally infected cattle, the birth of infected and clinically healthy calves is the most frequent consequence of the infection, which suggests that transmission occurs mainly in late gestation (Quintanilla-Gozalo et al., 2000; Pereira-Bueno et al., 2003). Therefore, the outcome of the infection depends on the moment when infection or its reactivation takes place during gestation and it is related to maternal and foetal immune responses. 2.1.2 Foetal immune responses The immunocompetence of the foetus at the time of the infection may determine its ability to control parasite infection and, therefore, its survival. The earlier the infection occurs during pregnancy, the more serious the consequences will be (Innes et al., 2002; Collantes-Fernández et al., 2006c). Bovine foetal immunocompetence develops gradually during the second half of pregnancy (Osburn et al., 1982; Buxton et al., 2002), recognizing and responding to pathogens during the second third of gestation. Therefore, foetus is very sensitive to infection during the first trimester of pregnancy (Williams et al., 2000; Innes et al., 2001), and the probability of foetal death is high. In the second third of gestation, the foetus can develop an immune response to infection although it might not be enough for its control (Williams et al., 2000; Andrianarivo et al., 2001; Almería et al., 2003; Bartley et al., 2004; Innes et al., 2005). In the third trimester, the Chapter II ~ Introduction 26 foetus is already immunocompetent (Osburn et al., 1982), controlling the multiplication of the parasite and limiting lesion development (Innes et al., 2002; Dubey et al., 2006). Experimental infections have supported this theory. Infection in early pregnancy was associated with parasite dissemination and Neospora-specific necrotic lesions, but no inflammatory reactions were detected in the foetuses (Macaldowie et al., 2004; Gibney et al., 2008), while infection in mid and late pregnancy resulted in fewer lesions and limited parasite dissemination (Maley et al., 2003; Gibney et al., 2008). Based on these experimental results, it has been suggested that the immune response of the foetus is responsible for the outcome of the infection, rather than the maternal inflammatory response (Gibney et al., 2008). 2.1.3 Maternal immune responses Briefly, two components constitute the immune system: humoral immune system, consisting of antibodies produced by plasma cells derived from activated B lymphocytes, and cellular immune system, consisting of T lymphocytes, macrophages, natural killer (NK) and proteins secreted by them, cytokines and chemokines. When CD4+ T lymphocytes recognize antigens, expressed on the membrane of the antigen-presenting cells, the immune response is stimulated. CD4+ T lymphocytes are divided into two groups, T helper 1 (Th1) and T helper 2 (Th2), depending on the secreted cytokines. Cytokines and chemokines are the soluble mediators of the immune system, acting as a communication network between cells of both lymphoid and non-lymphoid origin (Entrican, 2002). These molecules can exert pleiotropic effects on their targets such as induction of cell migration between anatomical compartments, cell activation, cell proliferation and regulation of antibody production. Haematopoietic colony stimulating factors (CSFs), which induce leucocyte colony formation in the bone marrow, chemokines, which recruit these cells to anatomical sites where they are required to carry out their effector function, and interferons, that are part of the innate response effective at controlling certain virus and intracellular parasite infections, are also components of the immune response. Th1 cells secrete pro-inflammatory cytokines IFN-γ, IL-12 and TNF-α stimulating the cell-mediated response, important for initiating immune responses that lead to the generation of T cells with a Th1-type cytokine profile. Th1-type responses are particularly effective at controlling intracellular infections, tumors and intracellular parasites as Neospora. On the other hand, Th2 cells secrete mainly IL-4, IL-10 and TGF-β, inhibiting the cell-mediated activity and increasing the humoral response (Entrican, 2002). In non-pregnant animal, infection with N. caninum usually does not cause clinical disease since a combination of innate and acquired immunity mechanisms control the spread of the parasite by the organism (Innes et al., 2001; Staska et al., 2003; Almería et al., 2003; Andrianarivo et al., 2005; Rosbottom et al., 2008; Bartley et al., 2013; Regidor-Cerrillo et al., 2014). IFN-γ produced by CD4+ T cells (Marks et al, 1998) limits multiplication of intracellular parasites (Innes et al., 1995) and is an important component of protective immunity (Khan et al., 1997; Baszler et al., 1999). On the other hand, the role of TLR could be key during the initial phases of infection since they are present in antigen presenting cells (TLR-2), in B lymphocytes (TLR-9) (Werling et al., 2006), and in Chapter II ~ Introduction 27 trophoblasts and other cell types within the placenta (Koga & Mor, 2008), recognizing pathogens and triggering innate immune responses against the parasite. Specifically, TLR-3, 7 and 8 have been recently implicated in N. caninum recognition in the bovine placenta (Marin et al., 2017a; Marin et al., 2017b). However, in the early stages of infection there is no clear trend towards cellular (Th1) or humoral (Th2) immune responses, since significant amounts of IFN-γ and IL-4 are secreted (Rosbottom et al., 2007; Regidor-Cerrillo et al., 2014). As the infection progresses, a polarization towards a Th1- type response occurs, predominating IFN-γ secretion over IL-4 secretion (Gibney et al., 2008; Rojo-Montejo et al., 2009a; Caspe et al., 2012; Regidor-Cerrillo et al., 2014). IFN-γ activates CD4+ T cells, which stimulate CD8+ T cells, also capable of secreting IFN-γ (Maley et al., 2006; Rosbottom et al., 2007). This immune response eliminates most of infected cells, inhibiting the parasite proliferation through the secretion of IFN-γ and TNF-α and blocking the invasion of new cells through the antibodies generated against tachyzoites (Innes, 2007). In this process, the secretion of IL-17 by CD4+ T cells previously activated by macrophages also seems to play a key role in the control of the infection (Flynn & Marshall, 2011). The immune response described is capable of drastically reducing the number of tachyzoites disseminated by the organism, remaining only bradyzoites, which are located in immuno-privileged tissues. Limited data from experimental infections suggested that when horizontal transmission occurs in non-pregnant animals, those animals might develop enough immunity to protect against vertical transmission during a subsequent pregnancy (Dubey & Lindsay, 1996; Ho et al., 1997; Williams et al., 2000; Innes et al., 2001). It has been demonstrated that the route of transmission influences the maternal immune responses and the outcome of the infection. ExTT leads to lower vertical transmission rates than EnTT, and some animals are able to clear the infection (Dijkstra et al., 2008). ExTT of N. caninum in cattle have been related with epidemic abortion, characterized by high parasite burden and high rates of abortion. However, immune responses generated during ExTT seem to confer protection against the parasite in successive gestations (Innes et al., 2001; Williams et al., 2003; Trees & Williams, 2005; McCann et al., 2007), and there is some epidemiological evidence that this also happens in the field (McAllister et al., 2000). On the other hand, there is no evidence, to our knowledge, that protective immunity to EnTT in N. caninum can occur. Moreover, recrudescence of a persistent infection during gestation have been observed (Williams et al., 2003), and immune responses do not seem enough to avoid vertical transmission in successive gestations. Thus, the incidence of foetal infection and foetopathy in the population might be positively related to infection prevalence where EnTT is dominant, and negatively related to infection prevalence where only ExTT occurs (Williams & Trees, 2005). Finally, infection with N. caninum during pregnancy or recrudescence leads to the parasite multiplication in the placenta, altering the immunological balance at the maternal-foetal interface with an increase of local pro-inflammatory IFN-γ, IL-12p40 and TNF-α cytokines which could compromise the gestation, together with an increase in IL-4 and IL-10 levels (Rosbottom et al., 2008; Regidor-Cerrillo et al., 2014), which avoids the immunological rejection of the foetus but favours the multiplication and vertical transmission of the parasite (Entrican, 2002; Innes et al., 2002). Chapter II ~ Introduction 28 When infection occurs early in gestation, the immune system of the dam can mount a strong cell proliferation response with production of high levels of Th1 and Th2 cytokines at the maternal- foetal interface together with intense placental infiltration of CD4+ and γδ T lymphocytes, NK cells and a high production of local IFN-γ (Maley et al., 2006; Rosbottom et al., 2008; Regidor- Cerrillo et al., 2014; Cantón et al., 2014b). Therefore, it is assumed that the abortion may be caused by the presence of Th1 and Th2-type cytokines at the maternal-foetal interface causing intense inflammation and placental tissue damage as the maternal immune system fights the infection. This local immune response is not observed when the foetus survives (Maley et al., 2006). During gestation, especially in the second third, a significant downregulation of cellular response to mitogen, a reduction in cell proliferation in response to specific N. caninum antigens and a lower IFN-γ production is observed in the dam (Innes et al., 2001). This fact is extremely important to avoid a possible immunological rejection of the foetus by the mother (Innes et al., 2005) but, on the other hand, might be a trigger for recrudescence of infection in persistently infected cows, and the Th2 cytokine environment at the maternal-foetal interface would favour parasite invasion of the placenta and infection of the foetus. Moreover, when cows were infected late in gestation, no abortion occurred and no lesions were observed in the foetuses, suggesting a control of the infection by the maternal -and foetal- immune responses (Benavides et al., 2012). Although a high rate of vertical transmission was observed, parasites were observed later in the placenta, from 28 dpi onwards (Maley et al., 2003; Macaldowie et al., 2004). Thus, the immunomodulation of the dam in late pregnancy plays a role in limiting the initial infection (Benavides et al., 2012), probably due to the specific production of IFN-γ by the mother at least partly (Rosbottom et al., 2011). Some studies postulated that difference in foetal death between early and late stages of pregnancy was more related to the intensity or magnitude of the immune response than to an imbalance of the Th1/Th2 responses (Rosbottom et al., 2008). However, in one study an extensive inflammatory response characterized by an inflammatory infiltrate of CD4 + and CD8 + T cells and high levels of expression of IFN-γ and IL-4 was observed after the recrudescence of a natural infection in the second third of gestation (Rosbottom et al., 2011), suggesting that this placental response, far from affecting foetal survival, could contribute to the parasitosis control. Therefore, it is not clear whether cytokines are beneficial or harmful to the host. It seems that it may depend on their concentration and location, as well as the gestation period. In this way, they could play an important role in the transplacental transmission of the parasite or mediate tissue damage, especially in the maternal-foetal interface. 2.1.4 Bovine placenta and neosporosis As described above, many studies have been carried out in pregnant cattle. Results from those works indicated that the foetal-maternal interface is a key place where N. caninum replicates and has to cross in order to be transmitted to the progeny. The placenta is one of the most severely affected tissues, together with the foetal CNS (Dubey et al., 2006). Microscopical lesions consisting of a non-suppurative inflammatory reaction in placenta have been widely described Chapter II ~ Introduction 29 (Barr et al., 1990; Macaldowie et al., 2004; Dubey et al., 2006; Cantón et al., 2014a). However, the pathogenesis of the infection at the maternal-foetal interface is not clear. Multiplication of the parasite within the placenta, with the subsequent tissue damage and alteration of the normal functions of the placenta have been proposed as mechanisms triggering the abortion. On the other hand, exacerbated immunological reactions in the placenta after infection with N. caninum is another hypothesis that could explain abortion (Innes, 2007). Be that as it may, primary infection with the parasite or reactivation lead to multiplication of the tachyzoites in host-tissues, arriving to the bloodstream and reaching the placenta, where the parasite preferentially replicates and cross the placental barrier, arriving to the foetus. The placenta is a temporal, dynamic and diverse organ with important immunological features that facilitate embryonic and foetal development and survival. In addition, the polarized maternal and foetal cell layers conforming the maternal-foetal interface within the placentome function as a biological barrier, avoiding the passage of pathogens and toxic substances to the foetus. However, N. caninum is very effective at crossing this barrier. Studying the interactions between the parasite and the bovine placental host cells is necessary to understand the factors that precipitate abortion in some animals or allow transmission in others. In order to understand the role that may play the placenta in the pathogenesis of the bovine neosporosis and the works carried out during the present Doctoral Thesis, some features of the bovine placenta are explained below. Among them, the placentome structure, summarized in Box 1, is one of the most important characteristics of the bovine placenta since it is the functional unit where the main functions take place. Figure 5. Representative images of the bovine placenta. (a) Image of a foetus within the placental membranes, (b) structure of bovine placentomes, and (c) main layers that conform the maternal-foetal interface. Chapter II ~ Introduction 30 Figure 6. Representative image from histological slides of the bovine placenta stained with HE. The arrows indicate the most representative parts of the bovine placentome. 20×. Bar 100 µm Chapter II ~ Introduction 31 2.1.4.1 Trophoblast cells in the maternal-foetal interface In general, the bovine trophoblast is composed of two populations of trophoblast cells, polarized uninucleated trophoblast cells (UTC) and trophoblast giant cells (TGC). A unique feature of the synepitheliochorial bovine placenta are TGC, also named binucleated cells (BNC), which are the result of an acytokinetic mitosis (Klisch, 1999b). These cells are crucial for the feto-maternal interaction since they can migrate and fuse with maternal uterine epithelial cells, releasing a variety of products into the maternal compartment. However, nowadays the regulation of the processes of differentiation, genome multiplication, endocrine activities, migration and fusion of bovine binucleate cells are poorly understood. Migration of BNC or TGC from the foetal trophectoderm into the maternal epithelium is crucial for the development of the placentome. After migration, fusion of trophoblast giant cells with single uterine epithelial cells leads to the formation of feto-maternal hybrid cells which are true syncitia in the maternal compartment of the placentome (Wimsatt, 1951; Wathes & Wooding, 1980). The migrating cells are mostly binucleated, non-polarized, arise from UTC by acytokinetic mitosis (Klisch et al.,1999a). They start to differentiate within the trophoblast at around 16 days of gestation (Wooding, 1992), just before implantation begins in cows (Björkman, 1969). BNC or TGC represent between 15 and 25% of all the chorionic epithelium during ruminant gestation (Wooding et al., 1997), with the remaining 80% consisting of UTC (Hradecky et al., 1988). These BNC are distributed in a random manner throughout the trophoblast, persisting in the trophectoderm at a constant proportion from implantation up to one or two days before parturition, when there is a rapid decrease in their number (Wooding et al., 1986). They are characterized by a dark-stained cytoplasm as a result of the presence of ribosomes and mature large Golgi bodies (Wango et al., 1990), which produce large membrane bound granules occupying almost half of the volume of BNC (Rodriguez et al., 2004). Although BNC are present in every ruminant placenta, the uterine caruncular syncytium formed in cattle is different since in the cow the syncytium is displaced by maternal unicellular cells (Wooding & Wathes, 1980), degenerating (Wooding et al., 1997) and being mostly phagocytized by UTC (Klisch et al., 1999b). Therefore, the syncytium formed by BNC is transient and the bovine placenta is classified as synepitheliochorial (Wooding, 1992). In addition, the migration of TGC does not continue beyond the maternal basement membrane and this migration has been described as “restricted trophoblast invasion/migration” (Pfarrer et al., 2003). In the placentome, there are small and scattered areas of local separation at these microvillar junctions, which contain maternal blood released intermittently from the maternal circulation. The tall columnar UTC of these separations actively participates in phagocytosis of erythrocytes, thereby providing iron for the foetus (Peter, 2013). In addition, UTC are involved in the production of interferon-τ (Roberts et al., 1992). BNC are not only unique because of their ability to migrate and their DNA content (most of them are tetraploid) (Klisch et al., 1999b), but also due to the production of a variety of TGC specific glycoproteins and growth factors. After the BNC migrate and fuse with the maternal epithelium, they release their granules into the maternal tissue; these granules can be detected both in the maternal trinucleate cells and the syncytial places (Wooding & Wathes, 1980). These processes Chapter II ~ Introduction 32 of migration, fusion and granule release continue throughout gestation and are critical to the development of the conceptus (Wooding & Wathes, 1980). The products of the BNCs include bovine placental lactogen (bPL), prolactin related protein-I (PRP-I) (Milosavljevic et al., 1989), as well as, different pregnancy-associated glycoproteins (PAGs), which have been identified as potential moderators of interactions at the maternal-foetal interface and are expressed from implantation to term (Green et al., 2000; Klisch et al., 2005; Wooding et al., 2005). Recent studies have demonstrated that several growth factors like vascular endothelial growth factor (VEGF) (Pfarrer et al., 2006b), platelet-activating factor (PAF) (Bücher et al., 2006) and fibroblast growth factor (FGF) (Pfarrer et al., 2006a) could be colocalized in TGC. These findings implicate that TGC survival, migration and fusion with uterine epithelial cells is a very complex process, and evidence that TGC are key regulators of various biological processes (angiogenesis, cell growth and differentiation and tissue remodelling) in the bovine placenta. Finally, TGC participate in the synthesis of steroids (Ullmann & Reimers, 1989; Matamoros et al., 1994) and prostaglandins (Reimers et al., 1985), essentials to support pregnancy. Furthermore, local steroid production by the placenta, particularly estrogens and progesterone, may also help to control caruncular growth, differentiation, and function (Hoffman & Schuler, 2002; Peter, 2013). 2.1.4.2 Extracellular matrix (ECM) Formation of the placenta is a dynamic process that involves constant tissue remodelling, from implantation to delivery. Any tissue remodelling involves major changes/alterations of the ECM composition and relies on a delicate balance between matrix synthesis and degradation (Haeger et al., 2016). The ECM, which provides an environment for placental cells and regulates cell functions including signaling, proliferation, migration, and invasion, consists of a mixture of several unmodified and conjugated proteins such as proteoglycans and glycosaminoglycans and structural components such as several types of collagen that can be degraded by metalloproteinases (MMPs) (Kalluri & Zeisberg, 2006; Haeger et al., 2016). Proteins representing the cytoskeleton (actin, desmin and vimentin), the ECM (collagen, fibronectin and laminin), and integrin receptors have been detected in bovine placenta using immunohistochemistry and can show a spatiotemporal expression pattern (Pfarrer et al., 2006; Zeiler et al., 2007). Adhesion to the ECM in the bovine placenta is maintained by diverse proteins, including the adhesive transmembrane integrins, which are important receptors at the feto-maternal contact zone since they are bi-directional signal transducers (Haeger et al., 2016). The role of cadherins in implantation and maternal-foetal communication was confirmed using cultured epithelial cells from bovine placentomes and caruncles (Bridger et al., 2008). The active interplay of ECM proteins during pregnancy ensures appropriate implantation, placental development, and timely placental detachment. Proteoglycans, such as decorin and biglycan, are small leucine-rich molecules responsible for fibrillogenesis, which expression patterns change over the course of pregnancy and differ between pathological and healthy placentas (Borbely et al., 2014). Decorin was isolated from the bovine placenta (Batbayar et al., 2000), and together with dermatopontin form a complex that may facilitate the binding of decorin to TGF-β1. Moreover, dermatopontin can bind to fibronectin. Chapter II ~ Introduction 33 Therefore, this complex may play a role in accelerating fibroblast adhesion and collagen fibrillogenesis, participating in tissue repair. It is logic then the downregulation of decorin observed during proliferation, remodelling and vascularization (Guillomot et al., 2014), dynamic processes that need ECM remodeling. ECM molecules can be degraded by MMPs, zinc-dependent endopeptidases which comprise collagenases, gelatinases, matrilysins, and convertase-activatable MMPs (Fanjul-Fernández et al., 2010). Their activity can be regulated by tissue inhibitors of MMP (TIMPs), chemokines, cytokines, and growth factors (Clark et al., 2008). Both MMPs and TIMPs have been identified in bovine placentomes during physiological pregnancy contributing to placentome growth by degradation and remodelling of the ECM in both foetal and maternal compartments (Walter & Boos, 2001; Kizaki et al., 2008; Dilly et al., 2011), and under pathologic conditions. In fact, histochemical studies in bovine placentomes highlighted not only the MMP expression but also their activity and possible activators and inhibitors as factors that may be involved in mechanisms of physiological or pathological membrane separation (Walter & Boos, 2001). The MMP activity may be associated with the rigidity of the intravillous ECM within the bovine placenta (Franczyk et al., 2017), and MMP-2, 3, and 7 play an important role in the degradation of decorin. Among TIMPs, only TIMP-2 protein has been localized in bovine placentomes so far, concretely in TGC (Walter & Boos, 2001; Dilly et al., 2011), whereas mRNAs from MMP-1, 3, 9, 13, and 16 and TIMP-1, 3, and 4 were increased in placentomes during calving (Dilly et al., 2010; Dilly et al., 2011), suggesting their participation in placental remodelling. Therefore, a multitude of functions of ECM and its components and modulators were suggested such as mediating trophoblast attachment to the uterine endometrium, inducing cell differentiation (Nakano et al., 2002), being a reservoir of growth factors (Taipale & Keski-Oja, 1997), or serving as tracks on which cells can migrate along (Pfarrer et al., 2003). However, implication of MMPs have also been observed in pathological conditions. Latest transcriptomic studies in bovine placentomes infected by N. caninum showed an upregulation of MMPs, specially MMP-13 (Horcajo et al., 2017), and MMPs might be involved in the crossing of T. gondii through the human placental barrier (Wang & Lai, 2013). In addition, MMPs from host infected with these apicomplexan parasites seem to be involved in infected macrophage dissemination in a mechanism resembling metastasis (Brasil et al., 2017), that may facilitate their transmission. 2.1.4.3 Foetal membranes and allantoic and amniotic fluids The placenta is primarily made up of foetal membranes, which include the amnion, allantois, and chorion (collectively termed extraembryonic membranes). The inner membrane is the amnion, which is inverted around the embryo such that the ectoderm lines a cavity filled with amniotic liquor, composed by bronchial and salivary secretions of the foetus and a small part of urine, chlorate carbonate and sulfate of Ca, K and Na, creatine, creatinine, carbohydrates, albumin, variable amounts of mucin and foetal hairs, epithelial cells, epidermal desquamations or foetal faeces (meconium). The allantois, the intermediate membrane, is a secondary outgrowth of the extraembryonic splanchnopleure from the ventral hindgut. The mesoderm is vascular. The allantoic fluid is mainly made up of the urine of the foetus. Its pH is 6.5 to 7.5 and contains traces of uric acid, salts, creatine, allantoin and glucose. At the beginning of pregnancy, the Chapter II ~ Introduction 34 accumulation of fluids in the allantois distends the foetal membranes, which adhere to the uterine wall favoring nidation. In addition, foetal fluids have a protective role for the foetus, especially at the beginning of pregnancy, when the embryo is more fragile. Finally, the outer membrane, responsible for the interactions with the uterine endometrium, is the chorion. The transitional stage between embryonic and foetal life is a phase of greater demand for the embryo. Therefore, greater activity of its membranes is required in order to establish relationships with the maternal organism. In this period, the vascularization of the allantois increases and the chorion fuses with the allantois, leading to the formation of allantochorion, that is the main component of the placenta. In fact, the foetal allantochorion adheres to the caruncles of the endometrium. Three layers are found in the allantochorial placenta: the endothelium, which line the allantoic blood vessels, the mesodermal connective tissue, and the chorionic epithelium, which derives from the trophoblast. In the uterine mucosa three layers can also be considered: the endothelium of the vessels, the adjacent connective tissue and the lining epithelium. Therefore, the allantoic epithelium is clearly involved in the transplacental exchange process (Leiser, 1975). 2.1.4.4 Placental functions Placenta, with all its components, connects the foetus with the maternal uterus, providing nutrition and gaseous exchange. In bovine species, this connexion between mother and foetus occurs without direct contact between the foetal and maternal bloods (uterine-placental circulation). This organ acts as the digestive, respiratory, absorption, excretion, metabolic and endocrine systems of the foetus, allowing the passage of oxygen and nutritive principles from the mother to the foetus and catabolites from the foetus to the mother. The two umbilical arteries carry oxygenated (arterial) blood from the placenta to the foetus. Moreover, the placenta protects the foetus acting as a barrier or filter that prevents the passage of pathogens and toxic substances, and it is considered a supplementary and transient extra-foetal organ since also works as a gland of internal secretion and participates, along with the corpus luteum, in the maintenance of pregnancy. Concretely, the trophectoderm is the main part of the placenta, and it is remarkably versatile. It is characterized by a great capacity for invasion, cell fusion, hormone production, specific nutrient absorption, selective transport, active metabolism, and finally by its ability to resist maternal immunological attack. Placental functions are detailed below and summarized in Figure 7. 2.1.4.4.1 Nutritive function An epithelium-specific polarized morphology of both layers displaying apical microvilli and junctional complexes are characteristic for the maternal-foetal interface (Björkman, 1973; Leiser, 1975). The passive exchange of foetal and maternal metabolites is limited to small molecules only. The nutrient can cross the bovine placenta by simple diffusion, facilitated diffusion or active transport. Low molecular weight compounds can pass through the placenta by simple diffusion according to the Donnan's effect, from the part with higher concentration to the part with lower concentration, until the equilibrium is reached in both parts. It is a very fast process and energy Chapter II ~ Introduction 35 is not consumed. Water, electrolytes, urea, uric acid, creatine, oxygen, CO2 and some drugs are diffused by this mechanism. Other substances need to join a carrier in order to cross the placenta. It is also a very fast method and consumption of energy is not needed. Substances such as glucose, amino acids and fatty acids necessary for the foetus pass across the placental barrier using this mechanism. Finally, active transport allows the transit of complex compounds, which are divided into simple constituents and resynthesized after having crossed the placental membrane. Proteins, phospholipids and neutral fats go across the placental barrier by active transport with energy consumption. In addition, the exchanges depend on the hemodynamic flows in the placentome and, consequently, are linked to the maternal hemodynamic situation, the foetal cardiac efficiency and the contractions of the uterine musculature. 2.1.4.4.2 Gas exchange across the placenta The gaseous exchanges (O2-CO2) across the materno-foetal barrier depends on the qualitative characteristics of the exchange membrane (surface and thickness), the speed of blood circulation in the placenta, the O2-CO2 tensions present in maternal and foetal blood (concentration gradient), but above all, the characteristics and the amount of foetal hemoglobin. It should be noted that foetal blood has less affinity for CO2 than maternal blood, favoring the passage of CO2 from the foetus to the mother (Donnelly & Campling, 2016). 2.1.4.4.3 Endocrine function The placenta works as an endocrine organ. It secretes both peptide and steroid hormones, which act to maintain pregnancy and to prepare for parturition and lactation (Donnelly & Campling, 2016). During the peri-implantation period, from day 9 to day 24 of gestation approximately, UTC of the elongating conceptus secrete IFN-τ (Roberts, 2007; Ealy & Yang, 2009; Bazer et al., 2015), which is considered the major pregnancy recognition signal in ruminants (Roberts et al., 1992). Its biological actions on the endometrium have been widely investigated (Roberts, 2007; Spencer et al., 2007; Spencer et al., 2008) and the major roles for IFN-τ include the inhibition of uterine prostaglandin PGF2α secretion, therefore preventing luteolysis, and the modulation of the innate immune responses (Hansen, 2011). The anti-luteolytic activity of IFN-τ remain until the implantation of the embryo is completed. In the successive phases of gestation, the luteotrophic action is ensured by the placental hormones, concretely by bPL. The main peptide hormone secreted by the bovine placenta is bPL (Duello et al., 1986), which is produced by BNC and is important for the maintenance of pregnancy. The secretion occurs in the cotyledons, in the area of contact with the maternal caruncles and it is discharged into the maternal bloodstream. Chapter II ~ Introduction 36 Fi gu re 7 . G ra p h ic al r ep re se n ta ti o n o f th e m ai n p la ce n ta l f u n ct io n s Chapter II ~ Introduction 37 The role of the bPL includes many functions such as (i) preparation of the mammary gland for lactation; (ii) double effect in relation to insulin, stimulation of its production and secretion after the incorporation of carbohydrates or increase in peripheral resistance, mobilization of fats and glycogenolysis during fasting period; and (iii) immunosuppressive effect. The bPL together with the progesterone also has luteotrophic and protective action of the corpus luteum. The uterus under stimulation of the progesterone and bPL loses its tone, the wall becomes thinner and the blood supply increases. To evaluate the steroid production of the placenta, it is necessary to consider the participation of the foetus, the corpus luteum, the placenta itself and the fundamental stages of steroid production for each hormone. Thus, placental activities are closely integrated with the foetal activities, coining the term "foetus-placental unit". The major steroids secreted by the placenta are progesterone and estrogens. The placenta is able, at the trophoblastic level (UTC and TGC), to produce progesterone in a remarkable amount and estrogens. Estrogens are believed to have a positive effect on the maturation process of the placenta (Grunert et al., 1989). Local steroid production by the placenta, particularly estrogens and progesterone, may also help to control caruncular growth, differentiation and function (Hoffman & Schuler, 2002). It is important to note that prolactin and relaxin are also produced by the placenta. Their concentrations in maternal blood increases at the end of gestation and during delivery preparing the mammary glands for lactation and relaxing the ligaments of the pelvic belt and intervertebral tail, as well as shortening and dilation of the cervix, respectively. 2.1.4.4.4 Protective function The placenta is a protective barrier for the foetus against germs and drugs that circulate in the maternal blood. The polarized morphology of both layers that conform the maternal-foetal interface only allows the exchange of simple molecules. In fact, maternal antibodies, for example, cannot enter the foetal circulation, resulting necessary for the calf to receive colostrum within the first few hours after birth. However, viruses have, in relation to their small dimensions, the possibility of overcoming the placental filter in each period of gestation and their damage is related to the named embryogenetic calendar. In addition, some organisms, as N. caninum, can replicate within the cells conforming the maternal-foetal interface, the caruncular epithelial cells and the epithelial trophoblast cells, and pass across the placental barrier arriving to the foetus. In fact, N. caninum is one of the most efficiently transmitted pathogens by the transplacental route. Trophoblast and caruncular cells are able to recognize pathogens and secrete cytokines and chemokines that recruit immune cells in the damaged area (Montes et al., 1995; Steinborn et al., 1998a; Steinborn et al., 1998b). Moreover, an important increment of macrophages in the cotyledonary villi is observed in foetuses older than 6 months. These cells can produce proinflammatory cytokines and present antigens, suggesting that they work as sentinel cells, and hence they have an important role in foetal defence. On the other hand, it has been postulated than macrophages in foetal villi may also contribute to transplacental transmission of microbial agents (Schlafer et al., 2000). From the immunological point of view, gestation is considered a transplantation of immunologically foreign material, which must be tolerated by the mother. Therefore, physiological pregnancy is considered a Th2 phenomenon essentially supported by the cytokines Chapter II ~ Introduction 38 produced by Th2 lymphocytes. However, how the maternal immune system is able to tolerate the conceptus allowing the development of the foetus is not fully understood. In the last years, the Danger Model theory has offered an explanation for successful pregnancy. This theory states that placental structure is likely to influence the immunological features at the maternal-foetal interface in different species. The epitheliochorial bovine placenta is relatively non-invasive and the foetus would not suppose a risk for the mother, so the immune system of the dam does not react against the foetus. A danger signal would be necessary to activate the maternal immune system. Danger signals are tissue damage and destruction, distress and non-apoptotic cell death. Infection would constitute such a signal (Matzinger, 1994; Entrican, 2002). Anyway, it seems clear that a balance of inflammatory and regulatory cytokines is important to ensure that the response itself is not damaging to the host by causing immunopathological effects. During the implantation phase, the predominantly lymphocytic population at the endometrium are NK cells, whose number increase in the luteal phase and throughout the first trimester of gestation. The NK cells are lymphocytes characterized by a cytotoxic ability and those presents in the decidua have been associated with the ability to interrupt gestation. The endometrium seems to have the capacity to inhibit the activity of NK, producing immunosuppressive factors, as progesterone, that favour implantation and gestation. The trophoblast also represents a barrier capable of resisting the attack of conventional cytotoxic T cells, thanks to the secretion of TGF-β, with local immunosuppressive effect. Under pathological conditions, there is a displacement of the cytokine equilibrium with prevalence of the Th1-type cytokines, especially IL-12, which activate NK, transforming them into powerful lymphokine- activated killer (LAK) cells capable of destroying the trophoblast by degranulation and releasing of proteolytic substances (perforins)(Elli, 2006). Cytokines beneficial for pregnancy are IL-4, IL-10 and TGF-β (Innes et al., 2007), as well as colony stimulating factors CSF-1 and GM-CSF, by inhibiting NK cell activity, by promoting trophoblast survival or by counteracting effects of inflammatory cytokines (Brach et al., 1992; Irving & Lala, 1995; Chaouat et al., 1995). Among those cytokines that appear to be detrimental are TNF-α, IFN- γ and IL-12, which are usually expressed at very low levels or are absent in normal placenta (Raghupathy, 1997; Roberston, 2000). TNF-α can cause thrombosis and smooth muscle contraction. It seems that TNF-α production by macrophages activates NK cells to produce IFN-γ, which further activates the macrophages to produce more TNF-α until the activated NK cells kill the embryo (Entrican, 2002). Finally, some cytokines may have both beneficial and detrimental effects, depending on concentration or stage of pregnancy (Innes et al., 2005). 2.1.4.5 Cell cultures isolated from the bovine placenta Decades of research into the molecular mechanisms by which the placenta forms and functions have sought to improve prevention, diagnosis, and management of disorders of this vital tissue. This research is quite advanced in human placenta. However, the placental tissues in different species reach their mature and anatomically distinct states via routes whose commonality of mechanism is often more assumed than confirmed (Huckle, 2017). Although differentiation of the early conceptus is very similar in human and cattle in size, the initial establishment with the uterine epithelium and subsequent growth and differentiation of the placenta are very different Chapter II ~ Introduction 39 between these two species (Schlafer et al., 2000). Therefore, human placental cell cultures or explants do not seem the most adequate tool for the study of placental disease in cattle. Bovine epithelial trophoblast and caruncular cell cultures are a convenient tool for a variety of future applications such as elucidation of the function of trophoblast-specific molecules that play a crucial role during pregnancy, clarification of the mechanisms that regulated the function of trophoblast-specific factors, learning about the cellular characteristics of caruncle and trophoblast, studying pathways of transport, communication and modification of metabolic substances, as well as studying pathways of infection of pregnancy associated diseases caused by parasites (e.g. N. caninum), bacteria (e. g. Bacillus spp.) and viruses (e.g. BVDV) or environmental toxins (e.g. plants, mycotoxins) (Bridger et al., 2007b; Awad et al., 2014). For these purposes, several cell lines have been established. To develop a suitable in vitro model for both layers that conform the interface between the dam and the foetus, caruncular epithelial and trophoblast cells, has been a challenge in the last few years. While the isolation, characterization and long-term culture of caruncular epithelial cells was relatively easily achieved (Bridger et al., 2007a; Bridger et al., 2007b), the establishment of a trophoblastic cell line resulted more complex and many attempts were made to culture bovine placental trophoblast cells (Figure 8). Figure 8. Representative images of bovine trophoblast cells (F3) (a) and bovine epithelial caruncular cells (BCEC-1) (b) in culture. (a) 20×. Bar 50 µm, (b) 10×. Bar 100 µm. 2.1.4.5.1 Bovine trophoblast cell cultures from cotyledon Traditionally, the isolation of bovine trophoblast cells had two possible origins, cotyledons (Munson et al., 1988; Vanselow et al., 2008) or blastocysts (Talbot et al., 2000; Shimada et al., 2001) (Table 1). However, the success of these attempts has been limited, firstly because the trophoblast cell cultures turned out to be uncryopreservable, and secondly because the characterization of the isolated putative trophoblast cell cultures left doubts concerning the cellular origin. The difficulties in the isolation of pure trophoblastic cell cultures from the cotyledon due to the heterogeneity of the tissue can be avoided when using non-implanted blastocysts as a source for trophoblast cells (Talbot et al., 2000; Shimada et al., 2001). However, blastocyst-derived cells are less differentiated and may not possess all properties characteristic for the cotyledonary trophoblast. Therefore, these cultures from pre-implanted blastocysts are useful for studying pre- Chapter II ~ Introduction 40 implantation and peri-implantation processes and early embryogenesis, but not for investigating processes that occur later during gestation. For example, abortions caused by N. caninum are more frequent during the second trimester of gestation in natural conditions, so the study of the pathways of N. caninum infection during transplacental transmission, as well as the evaluation of critical factors involved in placental pathogenesis should be done in trophoblast and caruncular cells isolated from caruncles and cotyledons, the real target for the parasite. It was in the year 2010 when Hambruch and collaborators achieved the isolation and characterization of trophoblast cells from cotyledons of 5 months pregnant cows, obtaining the first establishment of a long-term stable bovine cotyledonary trophoblast cell line (F3) of confirmed origin. This cell culture presented a good growth efficiency and the cells maintained their characteristics of origin beyond 45 passages. In addition, trophoblast cells were preserved by freezing, overcoming the problems to freeze and low efficiency when passing found in other trophoblast cultures previously isolated. The obtained cell line showed epithelial morphology and characteristic BNC in small numbers through all passages. The trophoblastic origin of F3 cells was verified by amplification of a Y-chromosome specific DNA-sequence and the presence of bPL mRNA. Immunofluorescence demonstrated that F3 cells were continuously positive for zonula occludens-2, cytokeratin and vimentin. In conclusion, the F3 cell line showed several in vivo characteristics of bovine cotyledonary trophoblast cells. Several investigations have been approached using this cell line (Dilly et al., 2010; Haeger et al., 2011; Sobotta et al., 2017; Horcajo et al., 2017; Loch et al., 2018). Following the methodology described by Hambruch et al. (2010), different research groups have obtained, characterized and used bovine trophoblast cell lines from cotyledonary origin (Martino et al., 2015). Other authors isolated only BNC by Percoll gradient and infected them with N. caninum NC-1, demonstrating the phagocyte capacity of these cells and suggesting that BNC could potentially participate in the transplacental infection of bovine neosporosis (Machado et al., 2007). In the last years, several research groups have continued the generation of trophoblast cells using new methodologies (Suzuki et al., 2011; Kawaguchi et al., 2016) (Table 1). 2.1.4.5.2 Epithelial cell cultures from bovine caruncle Most of in vitro studies focussed on the physiology and pathology of reproduction in cows used primary bovine endometrial cell from non-pregnant animals, and seldom differentiated between caruncular and intercaruncular regions (Fortier et al., 1988; Horn et al., 1998; Kimmins et al., 2003), which results fundamental since epithelia differ from each other functionally (Banu et al., 2005; Pfarrer et al., 2006a; Pfarrer et al., 2006b). Only one bovine endometrial (epithelial) cell line (BEND) originating from a non-pregnant animal had been characterized and deposited at the American Type Culture Collection (ATCC) (Johnson et al., 1999). Cell cultures derived from the caruncle of pregnant animals have rarely been established (Table 1). The group of the professor Pfarrer from the University of Veterinary Medicine of Hannover, in the attempt to create a cell culture model to study foetal-maternal interactions, established and validated a method to isolate and cultivate primary epithelial cells from the caruncle. The resultant cell line, named BCEC-1 (bovine caruncular epithelial cell line-1), was isolated from a pregnant cow (Bos taurus) with an estimated gestational age of 4 months (Bridger et al., 2007a) Chapter II ~ Introduction 41 (Table 1). For characterization of bovine epithelial cells, criteria such as phase contrast microscopy (Munson et al., 1991; Bridger et al., 2007b), cytokeratin expression (Yamauchi et al., 2003; Bridger et al., 2007a), and measurement of transepithelial electrical resistance (Bridger et al., 2007b) were established (Haeger et al., 2016). BCEC-1 cells were proven to be of maternal origin, and it was demonstrated that they preserved their epithelial specific properties. Cell morphology, protein expression and the ability to form an intact epithelial barrier was maintained for at least 32 passages. Furthermore, cryopreservation of the cells was achieved. This cell line, was created in order to standardise future experimental conditions, representing a functional, differentiated caruncular epithelium from pregnant cows, and since its isolation some investigations using this cell line have been carried out (Bridger et al., 2008; Waterkotte et al., 2011; Sobotta et al., 2017). The actual in vivo involvement of F3 and BCEC-1 during pregnancy offer a variety of future applications such as the study of pathways of infection of pregnancy associated diseases caused by parasites (e.g. N. caninum). In fact, these cells constitute a polarised cell culture model derived from the tissue participating in vertical transmission in pregnant cattle, so they are appropriate in order to study the interactions between different isolates of N. caninum and the bovine placenta during gestation. Table 1. Summary of bovine trophoblast cultures isolated from blastocysts, cotyledons or by new methods and bovine caruncular cell cultures isolated from pregnant animals Origin Cell line Reference Scope B la st o cy st BE-13 Stringfellow et al., 1987 Establish a method by which trophectodermal cells originating from individual preimplantation embryos could be perpetuated in culture CT-1 Talbot et al., 2000 Isolation and characterization of blastocyst-derived cell lines (CT-1 and CT-5) BT-1 Shimada et al., 2001 Establishment of a bovine trophoblast-1 cell line (BT-1), derived from in vitro matured and fertilized blastocyst; in the absence of feeder cells - Ramos-Ibeas et al., 2014 Description of an efficient method to obtain a trophoblastic cell line from a single bovine embryo or from an embryo biopsy - Pillai et al., 2019 Blastocyst-derived trophoblast C o ty le d o n - Munson et al., 1988 Isolation and culture of bovine cotyledonary trophoblastic cells from early second trimester placentas by collagenase digestion - Bridger et al., 2007a Isolation and validation of primary epitheloid cell cultures from bovine placental caruncles and cotyledons - Vanselow et al., 2008 Establishment and evaluation of a tissue culture model for bovine trophoblasts by collagenase digestion F3 Hambruch et al., 2010 Isolation and characterization of a bovine trophoblast cell line F3 Line A Line B Martino et al., 2015 Involvement of the Kiss-1R/Kps system in regulating trophoblast invasion and proliferation in bovine primary placenta cotyledon cell cultures N ew m et h o d s BT-(A-L) Suzuki et al., 2011 Establishment of 12 bovine trophoblastic cell lines using bone morphogenetic protein 4 (BMP4) biTBCs Kawaguchi et al., 2016 Generation of trophoblast cell lines (biTBCs) from bovine amnion-derived cells (bADCs) using an induced pluripotent stem cell technique C ar u n cl e Several caruncular cells Bridger et al., 2007a Isolation and validation of primary epitheloid cell cultures from bovine placental caruncles and cotyledons BCEC-1 Bridger et al., 2007b Establishment of an epithelial cell line from caruncles of pregnant cows and development of a model to study restricted trophoblast invasion, pathogenesis of pregnancy associated diseases and pathways of infection and transport Chapter II ~ Introduction 42 2.2 Factors depending on the parasite 2.2.1 Parasite stage, inoculation route and dose of infection Most of the experimental studies on ruminants have been carried out with tachyzoites maintained in vitro since the infectious parasite stage for ruminants in natural infection, the oocyst, is difficult to obtain. Some authors have reproduced the infection by oral administration of sporulated oocysts (Trees et al., 2002; Gondim et al., 2004c; McCann et al., 2007), although the rates of abortion described were significantly lower than those observed in studies performed by parenteral inoculation of tachyzoites. The establishment of the infection and the transplacental transmission depended on the dose administered and the stage of gestation when oocysts were used (Trees et al., 2002; Gondim et al., 2004c; McCann et al., 2007). However, comparisons between different studies using oocysts are difficult because of the variability in the viability of oocysts (reviewed by Dubey et al., 2017). Due to their susceptibility to digestion, N. caninum tachyzoites have usually been inoculated through a parenteral route (SC, intravenous -IV- or intramuscular -IM-). The IV and SC routes have been the most assayed routes of inoculation and IV route is associated with a more severe and reproducible clinical presentation in terms of abortion compared with the SC route (Macaldowie et al., 2004). However, some authors have suggested that SC inoculation may model the natural infection better than IV inoculation, since it allows the parasite to be processed through the lymph nodes before being distributed through blood circulation (Dubey & Schares, 2006). On the other hand, the effect of the infection dose using tachyzoites as inocula is related to the route of administration and the host species. In cattle, IV doses have ranged from 1-2×106 tachyzoites (Weber et al., 2013) to 5×108 (Macaldowie et al., 2004), causing infection in all inoculated dams at day 76 of gestation and 75% of abortions, and both infection and abortion in all cows infected on day 70 of gestation, respectively. Recently, the administration of higher doses of the parasite have also been associated with higher rates of abortion and the development of more severe tissue lesions in a pregnant bovine model infected at mid-gestation (Vázquez et al., submitted). SC inoculation of cattle with doses of 107 and 5×108 tachyzoites caused transplacental transmission and lesions in the foetus when inoculated in the second trimester of gestation, although the higher dose was associated with earlier and more severe histopathological changes (Maley et al., 2003). The IM route was used in the first studies of experimental infections in cattle (Dubey et al., 1992), and later in combination with IV or SC inoculations (Conrad et al., 1993b; Barr et al., 1994; Andrianarivo et al., 2000; Andrianarivo et al., 2001). Other pathways such as conjunctival have been sporadically used, resulting in infection and induction of a specific immune response in pregnant cattle without transplacental transmission (de Yaniz et al., 2007; Moore et al., 2014). Finally, the possibility of venereal transmission of neosporosis has been investigated. Experimental studies have shown that heifers are highly susceptible to intrauterine inoculation of semen spiked with tachyzoites (Serrano-Martínez et al., 2007a; Serrano-Martínez et al., 2007b), whereas adult cows were less susceptible to becoming infected using this route of inoculation (Canada et al., 2006; Serrano-Martínez et al., 2007b). Therefore, the vast majority of studies use IV or SC routes and the majority of experimental infections have been performed with very high doses, demonstrating that IV inoculation led to Chapter II ~ Introduction 43 more serious consequences for the foetus and that higher doses of parasite are related to higher abortion rate and more severe lesions. 2.2.2 Isolates and intraspecific variability As indicated above, one of the most important factors influencing abortion and transmission directly related to the parasite is the intraspecific variability of N. caninum isolates, which has been widely demonstrated both in vivo and in vitro. An extensive variation in virulence was demonstrated in N. caninum, which may be associated with variation in the clinical presentation of the disease, although little is known about the strain variation with respect to pathogenicity. To date, more than 100 N. caninum isolates have been obtained from domestic animals (cattle, water buffalo, sheep and dog) and wild animals (white-tailed deer, European bison and wolf) (Ortega-Mora et al., in press). At first, isolation of N. caninum focused on severe cases of the disease, for example, isolates NC-1 and Nc-Liverpool (Nc-Liv) were obtained from the encephalon of dogs with clinical signs, whereas the isolate Nc-SweB1 was obtained from the brain of an aborted bovine foetus (Dubey et al., 1988b; Barber et al., 1995; Stenlund et al., 1997). In Spain, SALUVET research group has obtained 10 isolates, mainly from congenitally infected and clinically healthy calves, with the aim of studying isolates with differences in the behaviour (Regidor- Cerrillo et al., 2008; Rojo-Montejo et al., 2009b). The availability of these isolates has made possible their comparison, facilitating the study of the intra-specific variability of the parasite. There are not specific animal models suitable for testing strain variation, and a meaningful comparison with pregnant cattle would be economically and ethically prohibitive. However, intra-specific variability of N. caninum isolates has been approached in the last years using both in vitro (cell culture techniques) and in vivo models (murine and ruminant models). Up to date, several authors have shown differences in the invasion and proliferation (Schock et al., 2001; Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014b; García-Sánchez et al., 2019), as well as in the dissemination and transmigration abilities in vitro (Collantes-Fernández et al., 2012; García-Sánchez et al., 2019). In addition, the characterization of a large population of bovine and canine isolates in mice demonstrated that some isolates were more virulent and more efficiently transmitted to the progeny than others (Lindsay et al., 1995; Atkinson et al., 1999; Collantes-Fernández et al., 2006b; Rojo-Montejo et al., 2009b; Regidor-Cerrillo et al., 2010; Dellarupe et al., 2014a), and recent results appeared to indicate that these findings in mice could reflect the effect of different isolates in cattle (Rojo- Montejo et al., 2009a; Caspe et al., 2012; Regidor-Cerrillo et al., 2014; Almería et al., 2016a). The knowledge obtained from those comparisons between isolates opens up new revenues for research and may allow the identification of new virulence factors, which is essential for the development of pharmacological and immunoprophylactic tools. Nowadays, the proliferating molecular approaches are presented as very useful tools for the study of intraspecific variability of the parasite. Chapter II ~ Introduction 44 2.2.2.1 Phenotypic variability in in vitro models According to the National Research Council of USA, experimental models are a critical component in biomedical research. They include animals, cells and all kind of cultures, as well as computational and mathematical simulations to improve human and animal health. On the other hand, the scientific world is increasingly aware of the welfare and ethics of animal experimentation. In this sense, since the publication of The Principles of Humane Experimental Technique (Russell & Burch, 1959), scientists and government agencies have endorsed replacement, reduction, and refinement as essential tools for promoting the humane treatment of research animals. Replacement alternatives refer to methods that avoid or substitute the use of animals. In vivo procedures should be replaced whenever possible by alternative methods that do not use animals, such as mathematical models or in vitro biological systems. Reduction is referred to any strategy that consists of using a smaller number of animals with scientifically valid results, or the maximization of the information obtained per animal. The refinement of the experimental design and the selection of the most appropriate model contribute to the fulfilment of this principle. Finally, refinement alternatives include all procedures to minimize or eliminate pain, as well as all enrichment methods to ensure animal welfare. In this context, the use of in vitro cultures has been widely accepted in order to provide an initial approach to certain mechanisms involved in complex processes since they allow the obtaining of reproducible results and the reduction of experimental animals as well as the optimization of the methods that will be used later during in vivo experiments. It has been demonstrated that in vitro models have numerous applications and, concretely, in vitro culture systems represent powerful tools for the study of apicomplexan parasites (Muller &Hemphill, 2012). In addition, tachyzoites can be maintained in cell culture by continuous passage. However, several contradictory studies indicated that tachyzoites cultured for a long number of passes showed a diminished pathogenicity in the murine model (Long et al., 1998; Bartley et al., 2006) or adaptation to cell culture (Pérez-Zaballos et al., 2005). Be that as it may, prolongated culture of tachyzoites seems to alter their original characteristics. One of the most important aplications of in vitro models in the research of apicomplexan parasites has been the understaining of the processes implicated in the lytic cycle of these parasites. The lytic cycle of N. caninum and other intracellular apicomplexan parasites comprises the processes of adhesion and invasion of the host cell, adaptation to intracellular conditions, maturation of the parasitophorous vacuole and multiplication of the parasite and, finally, egression of the tachyzoites from the vacuole causing the lysis of the host cell (Figure 9).This sequence of events is required for parasite survival and propagation in the course of animal infection (Smith, 1995; Black & Boothroyd, 2000). All these processes are strictly regulated by the sequential secretion of proteins by different organelles of the parasite such as rhoptries (ROP), micronemes (MIC) and dense granules (GRA), which interact directly with the host cell and modulate multiple functions, favouring the intracellular multiplication of the parasite. These processes have been more widely investigated in the zoonotic agent T. gondii at the molecular level (Dowse & Soldati, 2004; Carruthers & Boothroyd, 2007; Lim et al., 2012), but it is assumed that they may be similar in N. caninum due to the existence of conserved proteins with a high degree of homology (Hemphill et al., 2006). It has been demonstrated that N. caninum tachyzoites are able of invading and multiplying in a wide variety of cell types (Hemphill et al., Chapter II ~ Introduction 45 2004; Pérez-Zaballos et al., 2005; Hemphill et al., 2006; Müller & Hemphill, 2013), from bovine monocytes and endothelial cells (Lindsay & Dubey, 1989) to stable cell lines as kidney cells, human fibroblasts (HFF) or cells derived from the kidney of the African green monkey (Vero, MARC-145), being the latest the most used (Hemphill et al., 1999; Risco-Castillo et al., 2004; Regidor-Cerrillo et al., 2011). In addition to lytic cycle studies, in vitro models have been used with different aims. Next, we will focus on the studies of isolate characterization and the existing heterogeneity of the N. caninum isolates. In this sense, several studies have revealed that N. caninum isolates differ in certain in vitro phenotypes related to virulence such as tachyzoite invasion and growth rate, tachyzoite dissemination and transmigration through biological barriers and immune response modulation or evasion (Saeij et al., 2005; Taylor et al., 2006; Lambert & Barragan, 2010; Pollar et al., 2012). These phenotypic traits are key in the N. caninum virulence and may directly influence dynamics and outcome of infection in the natural host. In vitro works focussed on the comparison of different isolates carried out up to date are discussed below. Figure 9. Graphic representation of the lytic cycle of N. caninum and the image in cell culture. Source David Arranz Solís, Doctoral Thesis, 2016 Chapter II ~ Introduction 46 • Tachyzoite invasion and growth. High invasion, replication and tachyzoite yield in vitro was correlated with faster proliferation of the parasite in host tissues, reaching high parasite burdens in short time periods, which cause severe tissue damage and abortion. An extensive variability in the tachyzoite growth of N. caninum isolates have been demonstrated (Schock et al., 2001; Rojo-Montejo et al., 2009a; Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014b), and results are summarized in Table 2. In addition, a significant correlation between tachyzoite yield and pup mortality was reported in a BALB/c pregnant model for neosporosis (Regidor-Cerrillo et al., 2010; Dellarupe et al., 2014a), which supports the hypothesis that isolates with the highest inherent proliferation capacities reach higher parasite burdens in target host-tissues (foetal and placental), contributing to host tissue damage and clinical signs. However, a limitation of this kind of studies is the adaptation of tachyzoites to the in vitro culture, modifying its in vivo biological behaviour and virulence (Pérez-Zaballos et al., 2005; Bartley et al., 2008), which makes necessary the use of parasites with a low number of passages in cell culture (Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014b). • Migration. Higher migration capacities have been associated with wider parasite dissemination as well as with easier access to immunopriviledged sites, such as the CNS or the foetus. In T. gondii there is evidence of two putative complementary pathways utilized for crossing restrictive cellular barriers. T. gondii clonal genotype I tachyzoites exhibit a transmigratory ability in epithelial cell monolayers, linked to parasite gliding motility and virulence in mice (Barragan & Sibley, 2002; Taylor et al., 2006) whereas genotypes II and III induce a potent hypermigratory phenotype in DC, which have been identified as systemic carriers (Trojan horses) of T. gondii tachyzoites (Lambert et al., 2009; Lambert et al., 2011). These dissemination strategies seem to be conserved among coccidian apicomplexan parasites, since a similar isolate- specific migratory pattern was observed for N. canimum in human DC (Collantes-Fernández et al., 2012). Some N. caninum isolates exhibited a higher dependency on DC-mediated transmigration whereas other isolates exhibited a relatively higher dependency on extracellular tachyzoite transmigration for efficient translocation across polarized BeWo monolayers in vitro. Recently, Nc-Spain7-infected bovine macrophages showed a significantly higher hypermotility than those infected with Nc-Spain1H (García-Sánchez et al., 2019). • Parasite interaction with the host immune system. The ability of T. gondii to interact with the host cell through virulence factors and its correlation with subversion or induction of protective immune responses have been described in a high number of studies (Saeij et al., 2005; Blader & Saeij et al., 2009; Melo et al., 2013). This ability is directly related to strain virulence. Macrophages infected with type I and III strains are alternatively activated through phosphorylated STAT6 due to their polymorphic rhoptry kinase ROP16, while type II infected macrophages are classically activated through activating of NF-κB by the dense granule protein GRA15 (Jensen et al., 2011). However, the interaction of N. caninum with host cell signalling, especially with immune cells, have been barely studied. Studies in murine macrophages recognized TLR2-MAPK, TLR3-TRIF, TLR11-MEK/ERK pathways and NLRP3-inflammasome activation to be implicated in host-resistance against N. caninum by inducing production of proinflammatory cytokines IL-12p40, IL-1β, IL-6, TNF-α and IFN-γ (Jin et al., 2017; Li et al., 2018; Wang et al., 2018). Mota and collaborators (2016) demonstrated that p38 phosphorylation was quickly triggered in murine macrophages stimulated by live tachyzoites, while its chemical inhibition resulted in upregulation of IL-12p40 production and augmented B7/MHC costimulatory Chapter II ~ Introduction 47 molecules expression, showing that N. caninum manipulates p38 phosphorylation in its benefit, in order to downregulate the host's innate immune responses. A recent work studied the infection of N. caninum NC-1 isolate in human monocyte-derived macrophages (THP-1 cells), showing that N. caninum infection produced increased expression of pro-inflammatory cytokines (TNFα, IL-1β, IL-8, IFNγ) as well as induced expression of host-defence peptides (cathelicidins) mediated by MEK 1/2. Secretion of cytokines and cathelicidins from N. caninum-infected human macrophages reduced parasite internalization and promoted the secretion of pro-inflammatory cytokines in naïve macrophages (Boucher et al., 2018). Although these studies provide important advances, remarkable differences regarding the immune response between mice and human and bovine species exist, including the lack of TLR11 and TLR12 in the genome of cattle, among others (Jungi et al., 2011). In addition, the previous works did not compare different N. caninum isolates and, therefore, they did not demonstrate the possible heterogenicity of N. caninum isolates in immune response modulation. Thus, with the aim of determine the ability of two N. caninum isolates, which showed marked differences in vivo and in vitro, to modulate innate immune responses in the bovine host, SALUVET group has recently characterized for the first time N. caninum replication in bovine monocyte-derived macrophages and detailed isolate-dependent differences in host cell responses to the isolates Nc-Spain7 and Nc-Spain1H (García-Sánchez et al., 2019). Although both isolates survived and replicated in bovine macrophages, Nc-Spain1H presented a delayed replication and a lower growth rate without an exponential growth pattern respect to Nc-Spain7, which was accompanied by higher reactive oxygen species (ROS) production, higher IL12p40 expression by Nc-Spain1H-infected macrophages, and higher IFN-γ release by lymphocytes than Nc-Spain7-infected cells. Furthermore, infected macrophages expressed higher levels of IL-10 and lower expression of MHC Class II, CD86 and CD1b molecules than uninfected macrophages, with non-significant differences between isolates. Recent transcriptome studies of infected bovine macrophages have confirmed capacity of N. caninum infection to induce a pro-inflammatory gene expression profile that varied according to the virulence of Nc-Spain1H and Nc-Spain7 isolates (García-Sánchez et al., submitted). In spite of the advances in the last years, little is known about the mechanisms used by N. caninum to cross the placenta and the factors that enable some isolates to be more effectively transmitted and cause foetal death than others. 2.2.2.2 Variability in the murine model There is a great heterogeneity of murine models described in the literature. So far, the non- pregnant or brain infection model (Collantes-Fernández et al., 2006b; Pereira García-Melo et al., 2010) and the pregnant or congenital infection model (López-Pérez et al., 2006; López-Pérez et al., 2008) are the most commonly used. The main advantage of the brain infection model is the ease of handling, and a great utility in the chronic phase of infection. The parameters evaluated include morbidity and mortality, detection and parasite burden in the CNS, and the humoral and cellular immune responses. On the other hand, the pregnant mouse model is useful in the study of congenital neosporosis, and it has been presented as a more appropriate alternative than the non-pregnant model in the evaluation of the safety and efficacy of drugs and vaccines (Aguado- Martínez et al., 2009; Marugán-Hernández et al., 2011b; Rojo-Montejo et al., 2011; Monney et al., 2012; Debache & Hemphill, 2013). In addition to parameters evaluated in the non-pregnant Chapter II ~ Introduction 48 model, pregnant murine models allow the evaluation of cytokines at the placental level in mothers and the foetal death, morbidity, post-natal mortality and vertical transmission in the offspring. However, pregnant mice models present some disadvantages such as the synchronization of the estrus by the Whitten effect (Whitten, 1957), and a correct diagnosis of gestation. To date, most models of congenital neosporosis in mice emulate the ExTT, inducing the primary infection during pregnancy (Liddell et al., 1999; Quinn et al., 2002a; López-Pérez et al., 2006; López-Pérez et al., 2008), whereas attempts to reproduce EnTT have not been satisfactory (Jiménez-Ruiz et al., 2013). Numerous studies have been carried out with the aim of revealing aspects related to the pathogenesis of the disease (Table 2), the study of the immune response (Tanaka et al., 2000; Nishikawa et al., 2001; Ritter et al., 2002; Viera da Silva et al., 2017; He et al., 2017), as well as for the evaluation of vaccine and pharmacological candidates (Cannas et al., 2003a; Cannas et al., 2003b; Ramamoorthy et al., 2006; Aguado-Martínez et al., 2009; Marugán-Hernández et al., 2011b; Debache & Hemphill, 2012a; Debache & Hemphill, 2012b; Rojo-Montejo et al., 2012; Pastor-Fernández et al., 2015). Chapter II ~ Introduction 49 Ta b le 2 . V ir u le n ce c la ss if ic at io n o f d if fe re n t is o la te s o f N . c a n in u m a cc o rd in g to t h e ir b e h av io u r in m u ri n e p re gn an t m o d e ls a n d i n v it ro Is o la te O ri gi n , s p ec ie s ( R ef er en ce ) P re gn an t m ic e m o d el In v it ro m o de l V ir u le n ce co n se n su s M at er n al m o rb id it y P ar as it e p re se n ce in m at er n al t is su es N eo n at al m o rb id it y N eo n at al m o rt al it y V er ti ca l tr an sm is si o n p In vR 4h TY 56 h N C -1 B ra in , d o g w it h n eu ro lo gi ca l s ig n s ( D u b ey e t al ., 1 9 8 8 b ) Lo w H ig h H ig h H ig h 7 6 .8 % H ig h 9 2 .8 % - M ild H ig h N c- 6 A rg Fa ec es ( o o cy st s) , d o g ( B as so e t a l., 2 0 0 1 ) Lo w H ig h Lo w Lo w 0 % M ild 5 3 .8 % M ild M ild Lo w -M ild N c- B ah ia D o gs w it h n eu ro lo gi ca l s ig n s ( G o n d im e t a l., 2 0 0 1 ) H ig h H ig h H ig h H ig h 1 0 0 % H ig h 1 0 0 % H ig h H ig h H ig h N c- G er 2 Fa ec es ( o o cy st s) , d o g (S ch ar es e t a l., 2 0 0 5 ) Lo w Lo w Lo w Lo w 7 .7 % Lo w 3 0 .8 % Lo w Lo w Lo w N c- G er 3 Fa ec es ( o o cy st s) , d o g (S ch ar es e t a l., 2 0 0 5 ) Lo w M ild M ild M ild 1 9 % Lo w 9 .1 % M ild H ig h M ild N c- G er 6 Fa ec es ( o o cy st s) , d o g (S ch ar es e t a l., 2 0 0 5 ) Lo w Lo w Lo w Lo w 4 % Lo w 2 3 .1 % Lo w Lo w Lo w N c- Li ve rp o o l B ra in , d o g w it h n eu ro lo gi ca l s ig n s (B ar b er e t a l., 1 9 9 5 ) H ig h H ig h H ig h H ig h 1 0 0 % H ig h 1 0 0 % H ig h H ig h H ig h N c- N o w ra B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (M ill er e t a l., 2 0 0 2 ) Lo w Lo w M ild Lo w H ig h 8 7 % - - Lo w N c- Sp ai n 1H B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R o jo -M o n te jo e t a l., 2 0 0 9 a) Lo w Lo w Lo w Lo w 0 .5 % Lo w 5 % Lo w Lo w Lo w N c- Sp ai n 2H B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) Lo w Lo w M ild M ild 2 0 % M ild 6 1 .3 % M ild Lo w Lo w -M ild N c- Sp ai n 3H B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) Lo w M ild Lo w Lo w 5 % H ig h 8 9 % Lo w Lo w Lo w -M ild N c- Sp ai n 4H B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) H ig h H ig h H ig h H ig h 1 0 0 % H ig h 9 7 .3 % H ig h H ig h H ig h N c- Sp ai n 5H B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) H ig h H ig h H ig h H ig h 9 5 % H ig h 1 0 0 % M ild H ig h H ig h N c- Sp ai n 6 B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) Lo w Lo w M ild M ild 3 0 % M ild 5 7 .6 % M ild H ig h Lo w -M ild N c- Sp ai n 7 B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) H ig h M ild H ig h H ig h 9 5 % H ig h 7 9 .1 % M ild H ig h H ig h N c- Sp ai n 8 B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) Lo w Lo w Lo w Lo w 1 .1 % M ild 5 6 .4 % M ild Lo w Lo w -M ild N c- Sp ai n 9 B ra in , a sy m p to m at ic b u t co n ge n it al ly in fe ct ed c al f (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) Lo w Lo w M ild M ild 3 0 % M ild 5 2 .6 % M ild M ild Lo w -M ild N c- Sp ai n 10 B ra in , c o n ge n it al ly in fe ct ed c al f w it h c lin ic al s ig n s (R eg id o r- C er ri llo e t a l., 2 0 0 8 ) Lo w M ild M ild M ild 1 8 -2 0 % M ild 6 5 .5 % M ild H ig h M ild N c- Sw eB 1 B ra in , a b o rt ed c al f (A tk in so n e t a l., 1 9 9 9 ) M ild M ild - Lo w - - Lo w M ild A tk in so n e t a l., 1 9 9 9 ; M ill er e t a l., 2 0 02 ; Q u in n e t a l., 2 0 0 2 ; C o lla n te s- Fe rn án d ez e t a l., 2 0 0 6 b ; R o jo -M o n te jo e t a l., 2 0 0 9 b ; R eg id o r- C er ri llo e t a l., 2 0 1 0a ; P er ei ra G ar cí a- M el o e t a l., 2 0 1 0 ; R eg id o r- C er ri llo e t a l., 2 0 1 1 ; J im én ez -R u iz e t a l., 2 0 1 3 b ; D el la ru p e et a l., 2 0 1 4 a; D el la ru p e et a l., 2 0 1 4 b A tk in so n e t a l., 1 9 9 9 ; M ill er e t a l., 2 0 0 2 ; Q u in n e t a l., 2 0 0 2 a; C o lla n te s- Fe rn án d ez e t al ., 2 0 0 6 b ; R o jo -M o n te jo e t a l., 2 0 0 9 b ; R eg id o r- C er ri llo e t a l., 2 0 10 ; P er ei ra G ar cí a- M el o et a l., 2 0 1 0; R eg id o r- C er ri llo e t a l., 2 01 1 ; J im én ez -R u iz e t a l., 2 0 13 ; D el la ru p e et a l., 2 0 1 4a ; D el la ru p e et a l., 2 0 1 4b Chapter II ~ Introduction 50 2.2.2.3 Variability in the bovine model The bovine model, as target species of neosporosis, is the ideal model to study the consequences of parasite infection (Benavides et al., 2014). This model will be useful to study host-pathogen interactions and host immunity at the local and systemic level, as well as for evaluating diagnostics, vaccines and therapeutics. Both pregnant and non-pregnant models have been developed in cattle (Benavides et al., 2014). However, a standardized bovine model has not been established yet since there is a great diversity of experimental conditions employed, which influence the result of the infection, therefore resulting in a high level of variability and a trouble for comparison between experiments. These factors, previously exposed, include the gestation period, breed, parasite isolate and stage, and dose and route of inoculation of the parasite (Benavides et al., 2014). Thus, standardization is needed to advance research in a more collaborative, timely and efficient manner. 2.2.2.3.1 Non-pregnant bovine model Most of the studies performed in non-pregnant ruminants aimed to conduct research on the parasite cycle or the possibility of post-natal infection (Benavides et al., 2014). Cyclical oral transmission between dog and cattle was proven using this model (Gondim et al. 2002). The possibility of post-natal transmission of neosporosis has been addressed in new-born calves. It was demonstrated that tachyzoites present in colostrum could transmit the infection, although the infection has not been reproduced in calves (Davison et al., 2001). The possibility of venereal transmission of the disease have also been investigated using bovine experimental models. N. caninum DNA was detected in the semen of infected bulls (Ortega-Mora et al., 2003), although venereal transmission has not been demonstrated yet (Ferre et al., 2008; Osoro et al., 2009). Finally, numerous experimental studies have been carried out to characterize the immune responses occurring after N. caninum infection (Conrad et al., 1993b; Marks et al., 1998; Lunden et al., 1998; De Marez et al., 1999; Maley et al., 2001). 2.2.2.3.2 Pregnant bovine model As occurred in the murine models, primoinfection models, valid for evaluating ExTT, are the most common and their main objective is the evaluation of abortion and transplacental transmission. However, EnTT has not been experimentally reproduced yet (McCann et al., 2007; Benavides et al., 2014). • Foetal mortality model: Foetal mortality has been induced efficiently by IV inoculation of a high number of tachyzoites (107 - 5×108) at 70 days of gestation, principally from NC-1, Nc-Liv and Nc-Spain7 isolates. Parameters such as clinical observation (foetal viability and fever), detection of parasites (spread of the parasite in different organs of the dam and the foetus), presence of lesions in the foetus and in the placenta, and study of the immune responses (systemic humoral and cellular immune Chapter II ~ Introduction 51 responses as well as local expression of cytokines in the placenta or maternal tissues) have been used to monitor foetal death and transmission. However, the model of foetal mortality do not seem the most adequate to represent the natural infection since Neospora associated abortions have been described mostly at mid-gestation in the field (Dubey et al., 2007). In addition, IV infection on day 70 with the doses usually used could be excessively aggressive to carry out a correct evaluation of the efficacy of drugs or vaccines, underestimating some potentially effective products (Benavides et al., 2014), and to establish N. caninum isolate virulence (Regidor-Cerrillo et al., 2014). • Vertical transmission model: In this model, the infection is carried out from the second third of gestation (Williams et al., 2000; Innes et al., 2001; Almería et al., 2003; De Yaniz et al., 2007; Rosbottom et al., 2007; Wiengcharoen et al., 2011; Rojo-Montejo et al., 2013; Almería et al., 2016a; Almería et al., 2016b). The same parameters described for the foetal death model are valid to monitor the characteristics and consequences of the infection of this model, considering that here the most important thing is to determine whether transplacental transmission has occurred or not. Therefore, techniques are usually applied for detection of specific precolostral antibodies (ELISA, IFAT), determination of parasite presence (PCR, IHC) and observation of compatible lesions in several tissues (histology). In addition, other analyses can provide additional information. For example, in live vaccine assays, the genotyping of N. caninum by microsatellites analysis in foetal tissues allows to discriminate the infection by the isolate used in the vaccine or in the challenge (Rojo-Montejo et al., 2013). The bovine models are very useful for the study of the pathogenesis of the infection and the developed immune response, both peripheral and local in the placenta and the foetus. Numerous studies have investigated the distribution of the parasite, the lesions and the type of immune response (innate and memory) that occurs after infection, both in non-pregnant and pregnant cows (Benavides et al., 2014). The role of TLR (Bartley et al., 2013), as well as the humoral (Williams et al., 2000; Bartley et al., 2004; Macaldowie et al., 2004; Serrano et al., 2006; Rojo- Montejo et al., 2009a; Rojo-Montejo et al., 2013; Regidor-Cerrillo et al., 2014) and cellular responses, highlighting CD4+ and CD8+ T lymphocytes (Marks et al., 1998; Staska et al., 2003; Tuo et al., 2005; Rosbottom et al., 2007; Rocchi et al., 2011) and cytokines at the placental level (Rosbottom et al., 2008; Rosbottom et al., 2011; Almería et al., 2011; Regidor-Cerrillo et al., 2014; Cantón et al., 2014b; Almería et al., 2016a; Darwich et al., 2016), among others, have been studied. Recently, the lectin-binding pattern in the placentas of cows infected experimentally with N. caninum was studied by IHC techniques (Dorsch et al., 2019), proving the importance of the changes occurred in the ECM during N. caninum infection. Bovine models have also been used to evaluate the efficacy and safety of pharmacological treatments and vaccines in their final phase of development (Andrianarivo et al., 1999; Kritzner et al., 2002; Miller et al., 2002; Williams et al., 2007; Baszler et al., 2008; Vanleeuwen et al., 2011; Moore et al., 2011; Hecker et al., 2013; Rojo- Montejo et al., 2013; Weber et al., 2013). Finally, bovine models have been used to assess the biological diversity of N. caninum. Differences between isolates previously characterized in the murine model have been demonstrated in the Chapter II ~ Introduction 52 bovine model. In the numerous experimental infections carried out in pregnant ruminants, several N. caninum isolates have been employed. However, the comparison of those isolates is difficult as the experimental design varied considerably among the studies, as commented above. In addition, economic and infrastructure constraints involved in experimental infection of large animals are also impediments for developing this type of studies (Benavides et al., 2014). For these reasons, few studies comparing isolates in bovine models have been carried out. In these studies, the Nc-Spain1H isolate (low-virulent isolate that failed to induce mortality and led to a low transplacental transmission in mice), did not damage the foetus after infection at an early stage of gestation, whereas NC-1 caused death in 3 of 5 foetuses (Rojo-Montejo et al., 2009b). Furthermore, more severe lesions as well as higher frequency of parasite detection in placental and foetal tissues from NC-1 infected animals were found. In a similar comparative study, the Nc- Bahia isolate showed lower pathogenicity than NC-1 isolate in cows and buffaloes (Chryssafidis et al., 2014). Although in both inoculated groups there was a high vertical transmission, lesions observed in NC-1 infected group were more severe, and the frequency of abortion was higher in this group (100% vs 12.5%). In other studies, foetal death occurred in all cattle inoculated at day 70 of pregnancy with two high virulence isolates, Nc-Liv and Nc-Spain7, according to pathogenicity established in mice (Caspe et al., 2012). In addition, the high virulent Nc-Spain7 isolate was compared with Nc-Spain8, which showed low-moderate virulence in vitro and in mouse models but a transplacental transmission rate higher than 50% (Regidor-Cerrillo et al., 2014). In contrast to observations made in the pregnant mice model, both isolates showed 100% of foetal mortality, although foetal death occurred earlier in Nc-Spain7 inoculated animals, which presented higher parasite burdens in placental and foetal tissues. In a recent study, IV inoculation of Nc-Spain7 at mid-gestation induced 50% of foetal death until 42 days post-infection (Almería et al., 2016a) and 66.6% foetal death when gestation lasted until term (Vázquez et al., submitted). These studies highlighted the importance of the isolate in the outcome of infection in cattle and indicated that N. caninum isolates showing higher rate of invasion or tachyzoite yield in vitro as well as high virulence in mice are also more pathogenic in cattle than those moderate-low virulence isolates (Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014a). 2.2.2.4 Molecular basis of the variability N. caninum isolates have been isolated from diverse hosts and geographic regions, suggesting a high capacity of adaptation to different ecological niches, which could involve a high genetic diversity, as previously described in T. gondii (Grigg et al., 2001). However, in contrast to T. gondii, the genetic basis of the biological variability in N. caninum remains unknown. The first studies of genetic variability among N. caninum isolates were based on the analysis of 18S rRNA, as well as the ITS-1 (Internal Transcribed Spacer-1) and the RAPD (Randomly Amplified Polymorphic DNA). The high conserved sequences of other analyzed markers indicated a lower genetic variability than in T. gondii (Beck et al., 2009). Later, an extensive genetic diversity was demonstrated in N. caninum based on highly polymorphic microsatellite sequences (Regidor-Cerrillo et al., 2006; Regidor-Cerrillo et al., 2008; Pedraza-Díaz et al., 2009; Basso et al., 2009; Basso et al., 2010; Regidor-Cerrillo et al., 2013). The multilocus analyses have allowed to associate certain genetic profiles of each isolate to their country of origin (Regidor-Cerrillo et al., 2013), although a link Chapter II ~ Introduction 53 between certain microsatellite markers and the virulence of N. caninum isolates have not been found. In addition, the study of Single-Nucleotide Polymorphism (SNPs) allowed the definition of the population structure of T. gondii, with up to 15 haplogroups that include the classic type I, II and III strains. In fact, the type III strains present SNPs that affect the TgROP5, TgROP16 and TgROP18 genes, and determine their virulence (Taylor; et al., 2006; Saeij et al., 2006; Reese & Boothroyd, 2011; Behnke et al., 2011). In N. caninum, the identification of SNPs has been carried out recently by sequencing different isolates. However, those analyses showed that all the genomes are broadly identical, suggesting that the current population of N. caninum is due to the expansion of a single clonal lineage that has been favoured by the vertical transmission of the parasite by asexual reproduction. However, mitochondrial DNA and DNA from the apicoplast presented important differences respect to the nuclear DNA, evidencing a process of sexual recombination in the definitive host that would explain the intra-specific variability of the different isolates of the parasite (Khan et al., 2015). Furthermore, protein expression profiles between high- and low-virulence isolates of N. caninum have been carried out, finding differences in the abundance and protein species involved in gliding motility, lytic cycle processes and oxidative stress (Regidor-Cerrillo et al., 2012), which suggest the importance of variation on protein expression in the biological diversity of N. caninum. Recent studies combining the “omics” techniques provide a global view of the elements involved in biological processes at the molecular level. The rise of the “omics” techniques began after the sequencing and publication of the genome of several important eukaryotic pathogens in the EuPathDB database (Eukaryotic Pathogen Database, www.eupathdb.org) (Aurrecoechea et al., 2016). Specifically, the ToxoDB database (www.toxodb.org) groups freely and collaboratively the available “omics” information of Eimeria, Hammondia, Toxoplasma, Neospora, Sarcocystis, Cystoisospora and Cyclospora (Gajria et al., 2007). High-throughput genome, transcriptome and proteome procedures are revealing molecular mechanisms involved in biological diversity and virulence of apicomplexan. Findings implicate variability in genetic background and regulation of gene and protein expression at the level of epigenetics, transcription, translation, and posttranslational modification. Advances in genomic, transcriptomic and proteomic studies of N. caninum, which have allowed to evidence some mechanisms responsible for the phenotypic traits of the different isolates and to identify the processes determining their virulence, are summarized below. 2.2.2.4.1 Genomic studies Genomics comprises molecular characterization of complete genomes through the study of their content, organization, function of their genes and evolution of their genetic information. Up to date, the unique available genome sequence of N. caninum Nc-Liv isolate has 61 Mb, 14 chromosomes and more than 7,000 coding genes (Reid et al., 2012) and has been compared to T. gondii genomes of TgME49 (Type II) and TgVEG (Type III). Chapter II ~ Introduction 54 More information is available for T. gondii than for N. caninum, and includes different isolates (ME49, GT1, VEG and RH). Both T. gondii and N. caninum genomes showed a high degree of synteny with a one-to-one correspondence between most of protein-coding genes (Reid et al., 2012; Ramaprasad et al., 2015), so that T. gondii has traditionally been used as a model for studies of N. caninum. However, there are also marked discrepancies on gene repertoire between both T. gondii and N. caninum genomes (Reid et al., 2012). For example, several orthologous of virulence factors for T. gondii are pseudogenes such as ROP18 and GRA15 or were non-predicted from the Nc-Liv genome sequence such as GRA24. In addition, new putative genes associated to micronemes and rhoptries (SAG1-Related Sequence -SRS- gene family) such as NcSRS67 were identified in N. caninum (Bezerra et al., 2017). On the other hand, the comparison of the genome of different apicomplexan parasites has defined the existence of families of genes that could be associated with evasion strategies of the immune system and other virulence factors (Reid, 2015), characterized for their high mutation rate, gene expansion and high polymorphism. In T. gondii, these families include the surface proteins SRS (Tomavo, 2001; Reid et al., 2012) and the kinases present in the roptrias (ROPK) (Talevich & Kannan, 2013). In N. caninum, different protein-encoding genes have been sequenced and compared amongst isolates from different hosts and geographical areas (Beck et al., 2009), resulting in poor gene polymorphism with likely limited phenotypic impact. However, clonal gene expansion, recently determined in N. caninum, may drive phenotypic variants by gene dosage (Adomako-Ankomah et al., 2014). N. caninum contained predicted protein-coding genes involving SRS family, ROP and NTPase with attributed roles on virulence (Jung et al., 2004; Adomako- Ankomah et al., 2014; Pastor-Fernández et al., 2016). Whether there is variation in the copy number within N. caninum remains to be determined. Finally, as exposed before, the apicoplast, that has its own genetic material as well as its own gene expression machinery, seems to be essential for the parasite and, for this reason, its genome has been completely sequenced in T. gondii, N. caninum and some species of the genus Plasmodium. 2.2.2.4.2 Transcriptomic studies Transcriptomics studies gene expression at one specific stage of development of a cell by analysing transcribed RNA. This information supposes a link between genomic and proteomic findings, although it does not consider regulation occurred between transcription and translation. Transcriptome analyses were firstly performed in T. gondii tachyzoites using SAGE (Serial Analysis Gene Expression), ESTs (expressed sequence tag), microarrays, MPSS (Massively Parallel Signature Sequencing) and TSSs (Transcription Start Site) techniques (Radke et al., 2005; Wastling et al., 2009; Yamagishi et al., 2010; Hassan et al., 2012). Mechanisms of host cell manipulation after infection have been investigated by transcriptomics. For example, it was discovered that TgROP16, TgGRA7 and TgGRA15 proteins may alter certain transcription routes after infection by different mechanisms (Yamamoto et al., 2009; Ong et al., 2010; Rosowski et al., 2011; Jensen et al., 2013; Bougdour et al., 2014). Interestingly, many of these factors presented important polymorphisms between different T. gondii isolates, conditioning certain characteristics of the Chapter II ~ Introduction 55 infection (Taylor et al., 2006; Saeij et al., 2006; Reese & Boothroyd, 2011; Behnke et al., 2011). On the other hand, an increase in the transcription of micro-RNAs and the c-Myc gene in infected host cells was described and related to regulation of transcription, progression of cell cycle and apoptosis (Zeiner & Boothroyd, 2010; Franco et al., 2014). Transcriptomic studies also revealed that infection by T. gondii increased the expression of immune response genes and that expression varied according to the implicated isolate (Gail et al., 2001; Chaussabel et al., 2003; Okomo-Adhiambo et al., 2006; Skariah et al., 2010; Jia et al., 2013; Tanaka et al., 2013; Pittman et al., 2014). Nowadays, there are few studies of the transcriptome of N. caninum. The comparison between N. caninum and T. gondii tachyzoite transcriptome showed an expansion of the genes encoding the SRS surface proteins, as well as a lower or nonexistent transcription of virulence factors involved in mechanisms of pathogenicity of T. gondii (NcROP18, NcROP16, NcROP5, NcSUB2) (Reid et al., 2012). Species-specific gene repertoire may explain the biological differences of host restriction, transmission strategies, zoonotic potential and the limited pathogenicity of N. caninum in mice (Reid et al., 2012). In addition, it has been demonstrated that N. caninum cannot increase the transcription of micro-RNAs and the c-Myc gene, as observed in T. gondii (Zeiner & Boothroyd, 2010; Franco et al., 2014). On the other hand, controlled changes in N. caninum transcriptome are expected during the lytic cycle in host cells, during differentiation in specific stages, and during its exposition to different environment and immune responses in the host. The Apetala2 (ApiAP2) family is the major group of transcription factors in apicomplexan controlling parasite cell cycle progression, stage transformation and virulence, and 54 putative NcApiAP2s may be differentially expressed or regulated among N. caninum isolates determining phenotypic variants similarly to described in T. gondii (Croke et al., 2014; Melo et al., 2015). In addition, post-transcriptional regulation for apiAP2 factors has been suggested (Ramaprasad et al., 2015). Moreover, an extensive repertoire of histone modification machinery is present in apicomplexan, including N. caninum, although its “histone code” has not been explored (Bougdour et al., 2010; Dixon et al., 2010; Nardille et al., 2013). Recently, a putative set of natural long antisense transcripts (lancRNA) and long intergenic non-coding RNAs (lincRNAs) has been identified in N. caninum (Ramaprasad et al., 2015). NcRNAs can interfere in post-transcriptional regulation of gene expression through RNA splicing, transport or nuclear retention of the corresponding sense RNA transcripts, mRNA stability and modulation of translation in eukaryotes (Villegas & Zhapiropoulos, 2015). In the last years, SALUVET group has carried out transcriptomic studies that provide information on the mechanisms used for two different virulence isolates of N. caninum to interact with the host cell. In a first work, RNA-Seq was used to investigate differences in transcriptome between Nc-Spain7, a high-virulence isolate, and Nc-Spain1H, a low-virulence isolate, when infecting MARC-145 culture cells. Transcriptomes from both isolates showed marked variations throughout the lytic cycle but were inconsistent with proteome results carried out in the same conditions. However, a pre-bradyzoite status of the low virulence isolate Nc-Spain1H was identified (Horcajo et al., 2018). In a second work, differential expression of N. caninum genes involved in host cell attachment and invasion (SAG-related and microneme proteins), glideosome, rhoptries, metabolic processes, cell cycle and stress response between both isolates was found in F3 infected with both isolates, evidencing their intra-specific variability (Horcajo et al., 2017). Chapter II ~ Introduction 56 In the last work, it was demonstrated that Nc-Spain7 and Nc-Spain1H infection modulate bovine macrophage host signaling pathways to escape cellular defenses by repression of apoptosis and lysosome degradation. In addition, differences between isolates were compared, finding a greater modulation by Nc-Spain1H infection (García-Sánchez et al., submitted). Another recent work compared the tachyzoite transcriptomes of Nc-Liv (virulent) and Nc-Nowra (avirulent) and identified 3130 SNPs and 6123 indels between both isolates (Calarco et al., 2018). Nine markers were Sanger sequenced for both strains and for other additional eight strains, identifying a genetic population structure comprised of two major clades with no geographical segregation. The variants were predominantly located in loci associated with protein binding, protein-protein interactions, transcription, and translation. Furthermore, 468 nonsynonymous SNPs identified within protein-coding genes were associated with protein kinase activity, protein binding, protein phosphorylation, and proteolysis. This work may implicate these processes and the specific proteins involved as novel effectors of N. caninum tachyzoite virulence. 2.2.2.4.3 Proteomic studies Proteomics comprises the global analysis of the proteins expressed at a specific time-point, and have emerged as very useful tools for the study of apicomplexan parasites, establishing their protein repertoire and understanding the biological processes involved in stage transformation and host-cell interactions (Weiss et al., 2009; Wastling et al., 2009; Marugán-Hernández et al.,2010; Bautista et al., 2014). It is assumed that N. caninum proteome is reflect of transcriptome, although observed discrepancies between the transcriptome and proteome in apicomplexan suggest additional translational control on their proteome (Xia et al.,2008; Wastling et al., 2009). First proteomic approaches used two-dimensional electrophoresis (2-DE) to define the protein composition of N. caninum tachyzoites (Lee et al., 2003; Lee et al., 2004), and to compare tachyzoite proteome between different N. caninum isolates, showing similar proteomes with variation in a small number of isolate-specific proteins (Lee et al., 2005; Shin et al., 2005a; Shin et al., 2005b). The proteomes of N. caninum and T. gondii were also compared by 2-DE (Zhang et al., 2011). The development of Two Dimensional-Difference Gel Electrophoresis technique (2D- DIGE) allowed to quantify the differences in abundance of those proteins differentially expressed between two different extracts. Comparison of the protein expression profiles of tachyzoite of two high (Nc-Liv and Nc-Spain7) and one low virulence isolate (Nc-Spain 1H) demonstrated the increased abundance of proteins and variation of the protein species involved in the gliding motility (e.g., ACT1 and MLC1), lytic cycle processes (e.g., NcROP40 and NcMIC1), and oxidative stress responses (e.g., G6PDH) of virulent isolates (Regidor-Cerrillo et al., 2012). Remarkably, a high number of protein species with variation in abundance were identified as the same protein, which suggest additional and different regulation by post-translational modification, protein turn- over and degradation among isolates in N. caninum. A later study comparing Nc-Spain7 and Nc- Spain1H showed different proteomes throughout the lytic cycle and a pre-bradyzoite status of the low virulence isolate Nc-Spain1H was identified (Horcajo et al., 2018). Thus, these studies suggest that variations in the levels of protein expression and modulation could be the predominant force driving the biological diversity of N. caninum. Chapter II ~ Introduction 57 On the other hand, the study of the antigenic diversity among N. caninum isolates revealed a low intraspecific variability, since the analysis of different isolates by WB showed similar profiles (Schock et al., 2001; Miller et al., 2002; Shin et al., 2005a; Shin et al., 2005b; Lee et al., 2005). A recent work suggested that there may be a different host immune response against different isolates (Regidor-Cerrillo et al., 2015). In this study, immunomes from two virulent isolates (Nc- Liv, Nc-Spain7) and a low virulent isolate (Nc-Spain1H) were compared, confronting the protein extracts with sera from mice infected with each isolate. The differences found were dependent on the serum and not on the extract analyzed, suggesting that the differences in the immunome are due to the immune response induced by each isolate. Finally, proteomic approaches have also been used to compare the proteome of the tachyzoite and bradyzoite phases (Marugán-Hernández et al., 2010), to identify specific proteins of secretion organelles (Marugán-Hernández et al., 2011a; Sohn et al., 2011), as well as antigens that stimulate bovine T cells CD4+ (Rocchi et al., 2011) and secreted proteins by artificially inducing the release of intracellular calcium deposits (Pollo-Oliveira et al., 2013). CHAPTER III JUSTIFICATION AND OBJECTIVES JUSTIFICACIÓN Y OBJETIVOS Chapter III ~ Justification and objectives 61 Neospora caninum is an apicomplexan cyst-forming protozoan parasite that is considered as one of the main causes of abortion in cattle worldwide, including Spain. Important economic losses associated with the disease have been quantified, mainly as a consequence of the reproductive failure (Dubey & Schares, 2011). Transplacental transmission is the main route of transmission in cattle leading to abortion, birth of a weak calf or birth of a clinically healthy but persistently infected calf (Innes et al., 2002), playing the bovine placenta a key role in the pathogenesis of this disease. The direct damage produced by the multiplication of the parasite in placental and foetal tissues has been proposed as one of the possible causes of foetal death observed during N. caninum natural infections. Invasion of the placenta, proliferation and dissemination to the foetus seem to be crucial events in the pathogenesis of bovine neosporosis (Hemphill et al., 2006). However, the mechanisms by which N. caninum infects the placenta and reaches the foetus are poorly studied (Robbins et al., 2012). Importantly, in addition to be the main barrier for parasite transmission, the placenta is considered an immune regulatory organ, modulating foetal and maternal immune responses. It has been shown that the multiplication of the parasite in the placenta (and in other maternal tissues/organs) may alter the immunological balance at the maternal-foetal interface, suggesting immuno-mediated pathogenesis as a possible cause of abortion (Entrican, 2002; Quinn et al., 2002b; Innes et al., 2007; Bartley et al., 2012; Bartley et al., 2013). Trophoblast and caruncular cell layers constitute the maternal-foetal interface. These cells are able to recognize pathogens and secrete cytokines and chemokines, recruiting immune cells in the damaged area (Montes et al., 1995; Steinborn et al., 1998a; Steinborn et al., 1998b), and, thus, participacing in the initiation of innate immune responses at the placental level as well as in the development of an adaptative immune response. However, those cytokines and chemokines could contribute to abortion or facilitate parasite passage across the placenta and congenital transmission (Entrican, 2002; Innes, 2007). On the other hand, modified expression of adhesion receptors in the placenta could determine parasite passage. Selective expression of these cytokines and adhesion molecules, as a consequence of placental infection, may also reflect virulence mechanisms. Parasite-host interactions are also determined by the parasite intra-specific variability, which also supose an essential factor on the outcome of infection. Previous findings demonstrated that N. caninum isolates of bovine or canine origin show a large biological diversity, although they are genetically similar (Regidor-Cerrillo et al., 2011). Differences among several N. caninum isolates in vitro were correlated with differences observed in animal models and in the cytokine profiles induced during the infection (Regidor-Cerrillo et al., 2011; Dellarrupe et al., 2014a; Dellarupe et al., 2014b). Specifically, pregnant heifers inoculated at day 70 of gestation with the low-virulence isolate Nc-Spain1H spared the foetus (Rojo-Montejo et al., 2009b), whereas foetal death occurred in all inoculated cattle with the highly virulent isolate Nc-Spain7 (Caspe et al., 2012; Regidor- Cerrillo et al., 2014). In vitro models suppose an alternative to in vivo experiments, having numerous applications, and, concretely, in vitro culture systems represent powerful tools for the study of apicomplexan parasites (Muller &Hemphill, 2012). In addition to the reduction of experimental animals, they facilitate the initial approach to certain mechanisms involved in complex processes obtaining reproducible results. Despite the great utility of in vitro models, the use of animal models is Chapter III ~ Justification and objectives 62 essential for the study of several aspects of the disease. Although murine and ovine models have been utilized for the study of bovine neosporosis, those models did not reproduce faithfully the disease due to species-specific physiological differences in the structure of the placenta and in the immunological mechanisms developed against pathogen infection. Moreover, cattle is the most relevant and economically important target host, so that investigations related to neosporosis should be performed in the target animal species, despite to economic constraints and space-limitations. In this scenario, the general aim of the present Doctoral Thesis was to investigate the interactions between two N. caninum isolates with marked differences in virulence and the cells that conform the maternal-foetal interface in the bovine placenta both in vivo and in vitro. The final purpose is to bring light to the role of the bovine placenta in the pathogenesis of neosporosis, and the phenotypic factors that enable some isolates to be more effectively transmitted and cause fetal death than others. The Spanish isolates Nc-Spain7 and Nc-Spain1H were used in all the experiments. The high- virulence isolate Nc-Spain7 was obtained from an asymptomatic calf (Regidor-Cerrillo et al., 2008) and has been thoroughly characterized in both in vitro and in vivo (murine, ovine and bovine) models (Regidor-Cerrillo et al., 2010; Regidor-Cerrillo et al., 2011; Caspe et al., 2012; Collantes- Fernández et al., 2012; Regidor-Cerrillo et al., 2014; Dellarupe et al., 2014a; Dellarupe et al., 2014b; Arranz-Solís et al., 2016; Almería et al. 2016a; Almería et al., 2016b; García-Sánchez et al., 2019; Vázquez et al., submitted). The low-virulence isolate Nc-Spain1H was obtained from the brain of a clinically healthy calf from a Spanish herd with high intra-herd N. caninum seroprevalence (Rojo-Montejo et al., 2009a). Nc-Spain1H was also characterized in vitro (Rojo- Montejo et al., 2009a; Regidor-Cerrillo et al., 2010; García-Sánchez et al., 2019) and in vivo (Rojo- Montejo et al., 2009a; Rojo-Montejo et al., 2009b). The potential of this isolate as a live vaccine candidate was also investigated due to its attenuated nature (Rojo-Montejo et al., 2013). Previous findings showed the biological diversity between both isolates (Rojo-Montejo et al., 2009b; Regidor-Cerrillo et al., 2010; Regidor-Cerrillo et al., 2011; Collantes-Fernández et al., 2012; Regidor-Cerrillo et al., 2014; García-Sánchez et al., 2019), despite of apparently poor genetic variability (Regidor-Cerrillo et al., 2013). Concretely, differences in the lytic cycle of Nc-Spain7 and Nc-Spain1H were found in vitro (Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014b; García- Sánchez et al., 2019), mainly in the proliferation ability, which was clearly enhanced in Nc-Spain7. Moreover, differences found in vitro are correlated with differences observed in virulence and vertical transmission in animal models. Specifically, higher parasite burden and more severe lesions were observed in a pregnant mice model infected with Nc-Spain7 together with high rates of neonatal mortality and vertical transmission (Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014a), whereas 0.5% and 5% of neonatal mortality and foetal transmission were detected in mice infected with Nc-Spain1H (Rojo-Montejo et al., 2009a). In pregnant bovine models, 100% of foetal mortality and vertical transmission was observed after Nc-Spain7 infection at early gestation (Caspe et al., 2012; Regidor-Cerrillo et al., 2014), and at least 50% of foetal mortality at mid-gestation (Almería et al., 2016a; Vázquez et al., submitted). However, Nc-Spain1H infection spared the foetus at early gestation (Rojo-Montejo et al., 2009b). In vitro cultures of two bovine placental target cells that conform the maternal-foetal interface (trophoblast and caruncular cells) have been used in the present Doctoral Thesis in order to characterize the parasite-host interactions. Bovine caruncular epithelial cells, named BCEC-1, and Chapter III ~ Justification and objectives 63 bovine trophoblast cells, named F3, kindly provided by Prof. Pfarrer from the Anatomy Institute of the University of Veterinary Medicine of Hannover, were obtained by trypsin digestion of caruncular and cotyledonary tissues after manual separation of bovine placentomes from 4 and 5 months pregnant cows, respectively. Both established cultures were obtained through spontaneous immortalization, which resulted positive in order to avoid the loss of specific cell properties derived from the immortalization of cells by techniques such as gamma rays or different viral systems, as was observed in human trophoblast (Khoo et al., 1998). In addition, both the origin and the presence of specific properties from the original tissues were assessed by different studies (Bridger et al., 2007a; Bridger et al., 2007b; Hambruch et al., 2010; Waterkotte et al., 2011), concluding that these cell cultures would represent good tools for the study of regulatory mechanisms of placental growth and function. In the present Doctoral Thesis, the following specific objectives were pursued: ➢ Objective 1: Characterization of parasite interaction between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro. The aim of this objective was to bring light into the events that take place in the placenta after infection with different virulence isolates. The different phases of the lytic cycle in caruncular and trophoblast cells as well as the modulation of the immune responses elicited by the infection in these cells were investigated. The results will allow us to find differences between isolates in the maternal-foetal interface of the bovine placenta and to discern whether isolates of different virulence induce a different modulation of the placental cells during the early phases of infection. Moreover, comparisons between two isolates with marked differences in virulence are useful to unveil critical phases of the lytic cycle that may determine the differences observed among them, and virulence mechanisms using by N. caninum to avoid immune responses and to cross the placental barrier. o Sub-objective 1.1: Characterization of the lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro. In this sub-objective investigation of tachyzoite adhesion, invasion, proliferation and egress of high- (Nc-Spain7) and low- (Nc-Spain1H) virulence isolates in established cultures of bovine caruncular epithelial (BCEC-1) and trophoblast (F3) cells was carried out with the aim of evaluating critical factors involved in placental pathogenesis. o Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level. In this study, expression profiles of toll-like receptor-2 (TLR-2), Th1 and Th2 cytokines (IL-4, IL-10, IL-8, IL-6, IL-12p40, IL-17A, IFN-γ, TGF-β1 and TNF-α), and endothelial adhesion molecules (ICAM- 1 and VCAM-1) were evaluated in bovine caruncular epithelial (BCEC-1) and trophoblast (F3) cells after the infection with high- (Nc-Spain7) and low- (Nc-Spain1H) virulence isolates of N. caninum. Results from this sub-objective will help to establish the role of placental cells in the regulation of the innate immune responses at the placental level and their reprogramming by N. caninum infection with isolates of variable virulence. Chapter III ~ Justification and objectives 64 ➢ Objective 2: Investigation of the early infection by high- and low-virulence isolates of N. caninum in pregnant cattle at mid gestation . In vitro findings were also tested in experimental infections in cattle. Comparative analysis of the high- and low-virulence isolates will help to identify the role of immune responses at the maternal-foetal interface in the clinical outcome or in the control of parasite transmission and proliferation in placental and foetal tissues. Special emphasis was given to investigate the consequences of parasite infection in the placenta in vivo based upon immune cell population and expression profiles of immune elements, and other key markers of interest related to innate immune response and pathogenesis. o Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid-gestation with high- and low-virulence isolates. The aim of this sub-objective was to investigate how the differences between high- (Nc-Spain7) and low- (Nc-Spain1H) virulence isolates of N. caninum influence the clinical outcome, parasite distribution and burden, lesion development in placental and foetal tissues, and the specific antibody response during early infection in pregnant heifers inoculated at mid-gestation. The results from this experimental model will clarify some of the key events involved in the pathogenesis of bovine neosporosis, providing new insight into the host-pathogen interactions in the placenta and the foetus and validating results obtained in vitro. o Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. This sub-objective included the study of the variations in the mRNA and protein expression of elements of the innate and adaptative immune responses as pattern recognition receptors (PRRs) (TLR-2, -3, -8 and -9 and NOD-2), pro-inflammatory (IL-1β, IL-6, IL-8, IL-12p40, IL-17A, IFN-γ and TNF-α) and anti-inflammatory/regulatory cytokines (IL-4, IL-10 and TGF-β1), iNOS, chemokines (CCL2, 4 and 5) and endothelial adhesion molecule genes (ICAM-1 and VCAM-1) triggered locally in the bovine placenta at two early time-points (10 and 20 dpi) after inoculation with Nc-Spain7 and Nc-Spain1H isolate. The inflammatory infiltrate (T and B lymphocytes and phagocytic cells) was also studied. Finally, the regulation of the extracellular matrix was investigated by analysing the mRNA and protein expression levels of its regulators (MMP-2, -13, -14, TIMP-1 and -2) and the protein expression of its components (fibronectin, vimentin and collagen type IV). These studies will allow to identify association of the infection with subversion or induction of protective responses and investigate critical immune elements involved in placental pathogenesis as well as changes in the ECM of the placenta. These results will help to confirm and validate those obtained in the in vitro models. CHAPTER IV RESULTS (PUBLICATIONS) RESULTADOS (PUBLICACIONES) 67 LIST OF PUBLICATIONS Objective 1: Characterization of parasite interaction between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro. Sub-objective 1.1: Characterization of the lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro. Jiménez-Pelayo L, García-Sánchez M, Regidor-Cerrillo J, Horcajo P, Collantes-Fernández E, Gómez- Bautista M, Hambruch N, Pfarrer C, Ortega-Mora LM. Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum. Published in Parasites & Vectors. (17th October 2017) 10:463. DOI: 10.1186/s13071- 017-2409-9. Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level. Jiménez-Pelayo L, García-Sánchez M, Regidor-Cerrillo J, Horcajo P, Collantes-Fernández E, Gómez- Bautista M, Hambruch N, Pfarrer C, Ortega-Mora LM. Immune response profile of caruncular and trophoblast cell lines infected by high- (Nc-Spain7) and low-virulence (Nc-Spain1H) isolates of Neospora caninum. Published in Parasites & Vectors. (8th May 2019) 12:218. DOI: 10.1186/s13071-019-3466-z. Objective 2: Investigation of the early infection by high- and low-virulence isolates of N. caninum in pregnant cattle at mid-gestation. Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid-gestation with high- and low-virulence isolates. Jiménez-Pelayo L, García-Sánchez M, Vázquez P, Regidor-Cerrillo J, Horcajo P, Collantes- Fernández E, Blanco-Murcia J, Gutiérrez-Expósito D, Román-Trufero A, Benavides J, Ortega-Mora LM. Early Neospora caninum infection dynamics in cattle after inoculation at mid gestation with high (Nc-Spain7)- or low (Nc-Spain1H)-virulence isolates. Submitted to Veterinary Research (18th June 2019). Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. Jiménez-Pelayo L, García-Sánchez P, Collantes-Fernández E, Regidor-Cerrillo J, Horcajo P, Gutiérrez-Expósito D, Espinosa J, Benavides J, Osoro K, Pfarrer C, Ortega-Mora LM. Differential immune responses and extracellular matrix organization are observed in the bovine placenta during early infection at mid-gestation with high- and low-virulence Neospora caninum isolates. Manuscript in preparation. 69 Objetivo 1 Caracterización de la interacción parásito-hospedador entre aislados de alta y baja virulencia de N. caninum y las células diana de la placenta bovina in vitro Neospora caninum supone una de las principales causas de aborto en el ganado bovino en todo el mundo en parte gracias a su gran capacidad para cruzar la barrera placentaria. De esta forma, el daño causado por el parásito en su paso por la placenta se ha indicado como uno de los factores clave en la patogénesis de la neosporosis. Además, la infección con N. caninum supone un desafío para el sistema inmunitario de las vacas gestantes. De hecho, los desequilibrios en el balance Th1/Th2 en la placenta durante la gestación como consecuencia de la infección por N. caninum también se han propuesto como una posible causa del aborto. Los trofoblastos, células epiteliales fetales, componen la interfaz materno-fetal en el placentoma bovino junto con las células carunculares maternas. Ambos tipos celulares son componentes claves de dicha interfaz, jugando un papel fundamental en la funcionalidad de la placenta. De hecho, tanto las células del cotiledón como las de la carúncula son capaces de reconocer a patógenos, N. caninum entre ellos, e inducir una respuesta inmunitaria frente a ellos. En este objetivo se estudió el ciclo celular de dos aislados de N. caninum de alta y baja virulencia in vitro, en cultivos establecidos de células del trofoblasto y la carúncula bovina, así como se caracterizó la respuesta inmunitaria inducida por la infección en ambos tipos celulares en un momento temprano y otro más tardío de la infección (4 y 24 horas post-infección). En concreto, la adhesión, la invasión, la proliferación y la egresión de los taquizoítos de los aislados Nc-Spain7 y Nc-Spain1H fueron evaluadas mediante ensayos de inmunofluorescencia en placa y PCR cuantitativa, mientras que la respuesta inmunitaria se valoró mediante PCR cuantitativa del ARNm del receptor tipo toll 2 (TLR-2), citoquinas Th1 y Th2 (IL-4, IL-10, IL-8, IL-6, IL-12p40, IL-17, IFN-γ, TGF-β1, TNF-α) y moléculas de adhesión endotelial (ICAM- 1 y VCAM-1), y mediante la medición de los niveles de proteína de la IL-6, la IL-8 y el TNF- α por ELISA. N. caninum invadió y se multiplicó en ambas líneas celulares, aunque las mayores tasas de invasión (pInvR), infección (cInfR) y proliferación (TY58H) se encontraron en trofoblastos infectados con el aislado más virulento Nc-Spain7, que sugirió que los trofoblastos pueden ser el nicho preferido de replicación del parásito. La multiplicación en ambas células fue exponencial entre las 22 hpi y las 58 hpi, excepto para el aislado menos virulento, Nc-Spain1H, que mostró no ajustarse a un crecimiento exponencial ni lineal en las células de la carúncula. En dichas células, una egresión temprana de ambos aislados se observó a partir de las 22 hpi, hecho interesante que no había sido observado anteriormente y que demuestra el interés de este tipo celular y sugiere un papel como barrera en la placenta bovina. Observando las diferencias entre aislados, el aislado más virulento mostró mejor capacidad de penetración en las células del trofoblasto, así como mayor tasa de infección y multiplicación que el aislado menos virulento Nc-Spain1H. Sin embargo, las diferencias entre los aislados fueron mucho menos marcadas en el caso de las células de la carúncula sugiriendo un papel importante de los trofoblastos en la patogénesis de la infección. 70 Por otro lado, la valoración de la respuesta inmunitaria mostró un patrón similar de expresión de ARNm en ambas líneas celulares, encontrándose una rápida respuesta pro-inflamatoria, con el aumento de citoquinas como el TNF-α y la IL-8 en las células infectadas y una disminución de citoquinas anti-inflamatorias/regulatorias como la IL-6 y el TGF-β1. Los niveles proteicos de IL-6, IL-8 y TGF-β1 en el sobrenadante confirmaron los resultados de los niveles de expresión de ARNm. Se observaron además pequeñas diferencias entre los aislados, con una mayor expresión de TLR- 2 y TNF-α en ambas líneas celulares infectadas con el aislado Nc-Spain1H, que podrían ayudar al control del aislado menos virulento por parte del sistema inmunitario, lo que explicaría, al menos parcialmente, las diferencias encontradas entre los dos aislados. 71 1SALUVET, Animal Health Department, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain 2Department of Anatomy, University of Veterinary Medicine Hannover, Bischofsholer Damm 15, 30173 Hannover, Germany †Authors contributed equally Published in Parasites & Vectors (17th October 2017) 10:463. DOI: 10.1186/s13071-017-2409-9. Presented as oral communication in the “Parasites, Poverty and Social commitment” congress organized by the SOIPA & the European Veterinary Parasitology College (21st-24th June 2016, Bari, Italy). Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 72 Abstract Background: Neospora caninum, one of the main causes of abortion in cattle, is very effective at crossing the placental barrier and placental damage is crucial in the pathogenesis of abortion. Bovine trophoblast and caruncular cell layers are key cellular components in the maternal-foetal interface in placentomes, playing a fundamental role in placental functionality. Methods: We studied tachyzoite adhesion, invasion, proliferation and egress of high- (Nc-Spain7) and low- (Nc-Spain1H) virulence N. caninum isolates in established cultures of bovine caruncular epithelial (BCEC-1) and trophoblast (F3) cells. The parasite invasion rate (pInvR) and the cell infection rate (cInfR) were determined by immunostaining plaque assay at different time points and multiplicities of infection (MOIs), respectively. In addition, tachyzoite growth kinetics were investigated using real-time PCR (qPCR) analysis and immunostaining plaque assay at different times. Results: Neospora caninum invaded and proliferated in both cell lines. The pInvR was higher in F3 compared to BCEC-1 cells for the Nc-Spain7 isolate (P < 0.05), and higher for the Nc-Spain7 than the Nc-Spain1H in F3 cells (P < 0.01). The cInfR was also higher in F3 cells than in BCEC-1 cells for both isolates (P < 0.0001), and the cInfR for the Nc-Spain7 isolate was higher than for the Nc-Spain1H isolate in both cell lines (P < 0.05). Tachyzoite growth kinetics showed tachyzoite exponential growth until egress at 58 hours post-infection (hpi) for both isolates in F3, whereas Nc-Spain1H showed a non-exponential growth pattern in BCEC-1. Asynchronous egress of both isolates was observed from 22 hpi onwards in BCEC-1. In addition, the tachyzoite yield (TY58h) was higher in F3 than in BCEC-1 infected by both isolates (P < 0.0001), highlighting better replication abilities of both parasites in F3. Nc-Spain7 showed shorter doubling times and higher TY58h compared to Nc-Spain1H in F3 cells; adhesion, invasion and proliferation mechanisms were very similar for both isolates in BCEC-1. Conclusions: Our results indicate a highly similar behavior of high- and low-virulence isolates in their interactions with maternal caruncular cells and suggest an important role of foetal trophoblasts in the pathogenesis of N. caninum infection. Keywords: Neospora caninum, Cattle, Isolates, Virulence, Placenta, Trophoblast, Caruncle, Adhesion, Invasion, Proliferation 1. Background Neospora caninum is an apicomplexan protozoan parasite, phylogenetically related to Toxoplasma gondii. This parasite is considered a major cause of reproductive failure in cattle worldwide (Innes et al., 2005; Dubey et al 2006; Dubey et al., 2007), resulting in great economic losses (Reichel et al., 2013). Infection in cattle may occur through horizontal transmission, when cattle ingest sporulated oocysts shed by a canid definitive host, or by endogenous congenital transmission, from a persistently infected dam to a foetus (Williams et al., 2009). Oral infection or recrudescence in a pregnant cow can result in abortion, birth of a weak calf or birth of a clinically healthy but persistently infected calf (Dubey et al., 2007). Neospora caninum is one of the most efficiently transplacentally-transmitted organisms in cattle (Williams et al., 2009). During natural infections, invasion of the placenta, proliferation and dissemination to the foetus are crucial events in the pathogenesis of bovine neosporosis and are related to the interactions of tachyzoites with host cells and its relationship with the local immune response at the maternal-foetal interface (Hemphill et al., 2006). In vivo studies demonstrated that N. caninum is able to infect the maternal caruncular septum before crossing to the foetal placental villus (Gibney et al., 2008; Benavides et al., 2014). Despite Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 73 the fact that the placenta is directly involved in the pathogenesis of the disease (Entrincan, 2002; Innes, 2007), the mechanisms by which N. caninum infects the placenta and reaches the fetus are poorly understood (Robbins et al., 2012). One reason could be the placental diversity (Leiser & Kaufman, 1994), which makes an extrapolation of findings from one species to the other difficult. To date, only one limited in vitro study investigating the potential involvement of bovine trophoblast in N. caninum infection has been published (Machado et al., 2007). In addition, no information is available regarding in vitro infection in bovine caruncular epithelial cells and the role of placental cell layers in vertical transmission. In addition, a key question in bovine neosporosis is the influence of the parasite intra-specific variability on the outcome of infection. The lytic cycle of N. caninum and other apicomplexan parasites comprises the processes of invasion, adaptation to intracellular conditions, proliferation, and egress from host cells (Hemphill et al., 2006; Plattner & Soldati-Favre, 2008; Santos et al., 2009). This sequence of events is required for parasite survival and propagation in the course of animal infection. Our previous findings demonstrated that N. caninum isolates of bovine or canine origin show a large biological diversity, despite being genetically similar (Regidor-Cerrillo et al., 2011). Moreover, differences found in the events of the lytic cycle among several N. caninum isolates in vitro are correlated with differences observed in virulence and vertical transmission in animal models (Regidor-Cerrillo et al., 2011; Dellarupe et al., 2014b). Specifically, pregnant heifers inoculated at day 70 of gestation with the low-virulence isolate Nc-Spain1H spared the foetus (Rojo-Montejo et al., 2009b), whereas foetal death occurred in all inoculated cattle with the highly virulent isolate Nc- Spain7 (Caspe et al 2012; Regidor-Cerrillo et al., 2014). There is no information concerning the kinetics of events in the placenta that lead to an understanding of how the parasite actually reaches the foetal tissues. The influence of biological variability of the isolate on placental damage is also poorly understood. The cow possesses a cotyledonary (Strahl, 1906) and synepitheliochorial placenta (Wooding, 1992), where foetal cotyledons interdigitate with maternal caruncles to form placentomes (Bridger et al., 2007a; Bridger et al 2007b; Zeiler et al., 2007). The trophoblast (epithelial surface of the foetal cotyledons) consists of uninucleated and binucleated cells. Binucleated cells are responsible for a “restricted” trophoblast invasion (Pfarrer et al., 2003), playing an important role in embryo implantation and successful pregnancy outcomes. Caruncular epithelial cells form a polarized barrier that the parasite encounters before reaching and multiplying in foetal tissues. Hence, the aim of this study was to investigate the interaction of two isolates of N. caninum with maternal and foetal bovine target cells. Here, we studied tachyzoite adhesion, invasion, proliferation and egress of high- (Nc- Spain7) and low- (Nc-Spain1H) virulence isolates in established cultures of bovine caruncular epithelial (BCEC-1) and trophoblast (F3) cells. Since BCEC-1 and F3 cells conserve some of the properties from their tissues of origin (Bridger et al., 2007b; Waterkotte et al., 2011; Hambruch et al., 2010), they are a useful tool to evaluate critical factors involved in placental pathogenesis, such as the mechanisms used by N. caninum to cross the placental barriers. 2. Methods 2.1 Parasites and cell cultures Nc-Spain7 and Nc-Spain1H isolates were obtained from healthy, congenitally infected calves (Regidor- Cerrillo et al., 2008; Rojo-Montejo et al., 2009a) and extensively characterized using in vitro, murine and bovine models (Rojo-Montejo et al., 2009a; Rojo- Montejo et al., 2009b; Regidor-Cerrillo et al., 2010; Regidor-Cerrillo et al., 2011; Regidor-Cerrillo et al., 2013; Regidor-Cerrillo et al., 2014). Tachyzoites were routinely maintained in a monolayer culture of the MARC-145 monkey kidney cell line as described previously (Regidor-Cerrillo et al., 2011). The N. caninum isolates used in this study were subjected to a limited number of culture passages (from 8 to 15) to ensure the maintenance of their in vivo biological behaviour and avoid their adaptation to the host cells (Pérez-Zaballos et al., 2005). A bovine trophoblast cell line F3 (Hambruch et al., 2010) and a bovine caruncular cell line BCEC-1 (Bridger et al., 2007b) were isolated from two BVD- free, pregnant cattle (Bos taurus) with an estimated gestational age of 5 and 4 months, respectively. Cells were grown as indicated by Hambruch et al. (2010). Briefly, cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM)/Ham’s F12 containing 10% foetal calf serum (FCS) checked for the absence of specific IgG against N. caninum by IFAT, 100 IU/ml Penicillin, 100 mg/ml Streptomycin Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 74 and 2 mM Glutamine. All experiments were carried out with cells below passage 27, when both cell lines maintained their morphological and functional features (Bridger et al., 2007b; Hambruch et al., 2010; Waterkotte et al., 2011). Tachyzoites used for in vitro assays were recovered from 2.5–3 day growth cultures of MARC-145, when the majority of the parasites were still intracellular, and purified using Disposable PD-10 Desalting Columns (G.E. Healthcare, Buckinghamshire, UK) as previously described (Regidor-Cerrillo et al., 2011). Tachyzoite viability was checked by trypan blue exclusion. F3 and BCEC- 1 cell monolayers were inoculated within 1 hour of tachyzoite collection from flasks. All in vitro experiments in F3 and BCEC-1 cell lines were assayed in quadruplicate, and two independent experiments were carried out. 2.2 Parasite invasion rate Parasite invasion rate (pInvR) was defined as the number of tachyzoites invading the host cell at different time-points (hours) post-infection (hpi) and were determined following the methodology described in Dellarupe et al. (2014a) with minimal modifications. In order to obtain a confluent monolayer of F3 and BCEC-1, cells were seeded with 2 × 105 and 3 × 105 cells per well, respectively. Different densities of both cell types were used because F3 cells are bigger than BCEC-1 cells and formed a monolayer composed of polygonal cells while as BCEC-1 are smaller and they tended to form colonies and did not spread out the entire surface of the well. A total of 100 purified tachyzoites of each isolate were added to 24-well culture plates. Cultures were washed three times with phosphate buffered saline (PBS) at different time points (1, 2, 4, 6 and 8 hpi) for removing non- adhered/non-invading tachyzoites. Unwashed cultures were also included in the study. All plates were fixed at 48 hpi, and the pInvR was determined using single immunofluorescence staining as described below. To determine the pInvR, events (medium and large parasitophorous vacuoles) present in each well were counted using an inverted fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan) at a magnification of 200×. The pInvR at 1, 2, 4, 6 and 8 hpi (pInvR1h, pInvR2h, pInvR4h, pInvR6h, pInvR8h, respectively) was determined as the number of events per well in cell monolayers washed at different time points, and the total parasite invasion rate (pInvRT) was determined as the number of events per well in unwashed cultures. 2.3 Cell infection rate Multiplicity of infection (MOI) was defined as the ratio of the number of tachyzoites added to a known number of cells in a culture. Cell infection rate (cInfR) was defined as the percentage of cells infected using different MOIs (1, 2, 4, 6, 8 and 10). Cells were cultured in 24-well plates at concentration of 2 × 105and 3 × 105 cells per well for F3 and BCEC-1 cells, respectively. Infected cells were washed 3 times with PBS after 4 hpi to facilitate the synchronization of the cultures. Finally, cells were fixed at 48 hpi and stained using single immunofluorescence staining as described below.The overall number of cells, the number of infected cells and the number of cells containing more than one vacuole (multi-infected cells) were counted in 10 arbitrarily selected fields using an inverted fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan) at a magnification of 200×. Counting of events was carried out on images taken with three different filters (white light for discrimination of cell limits, blue-DAPI for visualization of the nuclei and red-Alexa 594 for examination of the tachyzoites) using a Nikon DSL1 camera (Chiyoda, TYO, Japan) and overlaid using Photoshop® software (Adobe Systems Incorporated, Mountain View, CA, USA). A mean value of 50 cells was counted in each field. 2.4 Adhesion-invasion assay An adhesion-invasion assay was performed in F3 and BCEC-1 cultures seeded at concentration of 2 × 105 and 3 × 105 cells per well, respectively. Cells were infected at a MOI of 2, and cultures were washed with PBS at 4 hpi to remove non-adherent extracellular tachyzoites. Cultures were immediately fixed, and double immunofluorescence staining was carried out following the protocol described below. Adhered extracellular tachyzoites (green- and red-stained) and intracellular tachyzoites (red-stained only) were counted using a fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan) at a magnification of 400×. A total of 1000 tachyzoites was counted in each coverslip. The percentage of intracellular tachyzoites (red-stained) respect to the total number of tachyzoites (intracellular and extracellular adhered tachyzoites) (green-stained) at 4 hpi was calculated. Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 75 2.5 Intracellular proliferation assays: Proliferation kinetics, doubling time and tachyzoite yield determinations Proliferation kinetics of Nc-Spain7 and Nc-Spain1H isolates in F3 and BCEC-1 cells were determined by quantifying the number of tachyzoites at specific times (4, 10, 22, 34, 46, 58, 70 and 82 hpi) by real- time PCR (qPCR). Cells were cultured and infected as indicated above using a MOI of 2. Cultures were washed at 4 hpi and subsequently maintained at 37 °C in 5% CO2. The samples were collected adding 200 μl of PBS, 180 μl of lysis buffer and 20 μl of proteinase K (Qiagen, Hilden, Germany) to each well at 4, 10, 22, 34, 46, 58, 70 and 82 hpi, transferred to a microcentrifuge tube and frozen at -80 °C prior to DNA extraction. In parallel, replicates of cell cultures in coverslips were infected as described above and were labelled using double-immunostaining to study microscopically the proliferation kinetics of both isolates in F3 and BCEC-1 cells. Three coverslips were photographed for each condition using an inverted fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan). The doubling time (Td) was defined as the period of time required for a tachyzoite to duplicate during the exponential multiplication period, excluding the lag phase (period without parasite multiplication) and the egress phase (Regidor-Cerrillo et al., 2011). The Td was determined by applying non-linear regression analysis and an exponential growth equation using GraphPad (San Diego, CA, USA). We represented Td for each isolate and each cell line as the average value obtained from all the determinations that revealed a linear regression, R2 ≥ 0.95. The tachyzoite yield (TY58h) was defined as the average value of the number of tachyzoites quantified by qPCR at 58 hpi for each isolate and cell line. 2.6 Immunofluorescence staining Single immunofluorescence staining was carried out as specified previously (Dellarupe et al., 2014a) with minimal variations. Parasites in fixed cultures were stained using hyperimmune rabbit antiserum directed against N. caninum tachyzoites (1:1000) as a primary antibody and a 1:1000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor® 594 (red, Thermo Fisher Scientific, Waltham, MA, USA) as a secondary antibody. The nuclei were stained by washing the cells with a solution of 1:5000 DAPI in PBS. Double immunofluorescence staining was carried out following the protocol described by Regidor- Cerrillo et al. (2011) with minimal modifications. Fixed plates were treated with 3% BSA in PBS for 30 min at 25 °C to block unspecific antibody binding and autofluorescence. Then, cultures were treated with a 1:1000 dilution of anti-tachyzoite hyperimmune rabbit antiserum and a 1:1000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor® 488 (green, Thermo Fisher Scientific, Waltham, MA, USA). After this step, only extracellular tachyzoites were labelled in green. After the first staining, cells were permeabilized using a solution of 0.25% Triton 100X in PBS 0.3% BSA (30 min, 37 °C). Later, cultures were treated again with a dilution of anti-tachyzoite hyperimmune rabbit antiserum as primary antibody (1:1000) and a 1:1000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor® 594 as secondary antibody (red, Thermo Fisher Scientific, Waltham, MA, USA). Therefore, intracellular tachyzoites were labelled only in red, while extracellular tachyzoytes resulted labelled in green and in red. The nuclei were stained by washing the cells with a solution of 1:5000 DAPI in PBS and the coverslips were embedded in Fluoroprep (BioMerieux, Marcy- l’Étoile, France). 2.7 DNA extraction and real-time PCR Genomic DNA was extracted from cellular samples using the DNeasy® Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Genomic DNA was eluted in a volume of 60 μl of molecular-grade water. Concentrations of DNA were determined for each sample using a nanophotometer (Nanophotomer®, Implen GmbH, Munich, Germany) and samples were diluted 1:4 in molecular-grade water. Quantification of N. caninum DNA was performed by real-time PCR using an Applied Biosystems 7300 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The Nc-5 region was targeted as described elsewhere (Collantes-Fernández et al., 2002). Five μl of diluted DNA from each sample were used for the qPCR assays. The number of N. caninum tachyzoites was determined by interpolating the Ct values (cycle threshold value, which represents the fractional cycle number reflecting a positive PCR result) on a standard curve. The standard curve was designed for the quantification of 10-1– 104 tachyzoites according to Regidor-Cerrillo et al. (2011). To normalize the quantification of the parasites in each sample, a bovine β-actin standard curve was designed (from 64 ng of DNA per μl to 0.2 ng per μl). The results were expressed as the Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 76 relation between amounts of parasite DNA and cell DNA (R2 ≥ 0.99; slope values varied from -3.67 to - 3.13). 2.8 Statistical analysis The parametric one-way ANOVA test, followed by a Tukey’s multiple comparisons test, was performed to investigate the influence of time on the pInvR and MOI in the cInfR, and the two-way ANOVA test, followed by a Tukey’s multiple comparisons test, was performed to study the influence of the parasite isolate and the cell type on the pInvR and cInfR. A Chi-square test was carried out to investigate the differences in the percentages of intracellular tachyzoites at 4 hpi in both target cell types. Bonferroni correction was used to eliminate error associated with making multiple comparisons. Statistical significance was established as P < 0.05. Differences that showed P-values ≥ 0.05 and < 0.1 were considered to be trending towards statistical significance. GraphPad Prism 5 v.5.01 (San Diego, CA, USA) software was used to perform all statistical analyses and graphical illustrations. 3. Results 3.1 Parasite invasion rate (pInvR) To investigate the impact of the placental cell type in parasite invasion, the pInvR was evaluated in trophoblasts and caruncular cells at different time points post-infection (1, 2, 4, 6 and 8 hpi). The pInvRs of the Nc-Spain7 and Nc-Spain1H isolates in F3 and BCEC-1 cells are shown in Fig. 1. The number of invaded tachyzoites for both isolates significantly increased until 4 hpi in both F3 (Nc-Spain7, ANOVA: F(5,35) = 11.87, P < 0.0001; Nc-Spain1H, ANOVA: F (5,35) = 7.211, P < 0.0001, followed by a Tukey’s multiple comparisons test) and BCEC-1 cells (Nc-Spain7, ANOVA: F (5,35) = 9.825, P < 0.0001; Nc- Spain1H, ANOVA: F (5,35) = 9.156, P < 0.0001, followed by a Tukey’s multiple comparisons test). From 4 hpi onwards, significant differences werenot observed. Regarding the influence of the target cell type, a higher pInvR was observed in F3 cells compared to BCEC-1 cells from 6 hpi onwards for the Nc-Spain7 isolate (two-way ANOVA test: F (3,168) = 27.25, P < 0.0001, followed by a Tukey’s multiple comparisons test). No statistically significant differences were found when the pInvRs of Nc- Spain1H in both cell lines were investigated. The influence of the parasite isolate on the invasion of bovine trophoblasts and caruncular cells was also investigated by comparison of the pInvRs between the Nc-Spain1H and Nc-Spain7 isolates, assayed at different times of infection. Nc-Spain7 showed a pInvR significantly higher than Nc-Spain1H from 6 hpi onwards in F3 cells (two-way ANOVA test: F (3,168) = 27.25, P < 0.0001 followed by a Tukey’s multiple comparisons test) (Fig. 1a). However, no statistically significant differences in pInvR were found between isolates in BCEC-1 cells (Fig. 1b). 3.2 Cell infection rate (cInfR) The percentage of infected cells (cInfR) and the percentage of multi-infected cells were evaluated at different MOIs. The number of infected cells significantly increased with increasing MOIs in both cell lines, F3 (Nc-Spain7, ANOVA: F (5,42) = 228.5, P < 0.0001; Nc-Spain1H, ANOVA: F (5,42) = 273.4, P < 0.0001, followed by a Tukey’s multiple comparisons test) and BCEC-1 (Nc-Spain7, ANOVA: F (5,42) = 30.04, P < 0.0001; Nc-Spain1H, ANOVA: F (5,42) = 42.60, P < 0.0001, followed by a Tukey’s multiple comparisons test). The cInfRs were higher in infected F3 than in BCEC-1 cells for both Nc- Spain7 and Nc-Spain1H isolates at the same MOI (Fig. 2a,b) (two-way ANOVA test: F (3,168) = 222.4, P < 0.0001, followed by a Tukey’s multiple comparisons test). The percentage of cells containing more than a single vacuole (Fig. 2c, d) was also higher in F3 than in BCEC-1 infected by both isolates (two-way ANOVA test: F (3,168) = 93.64, P < 0.0001, followed by a Tukey’s multiple comparisons test). In addition, Nc-Spain7 showed a higher cInfR than Nc-Spain1H in both cell lines at different MOIs (two- way ANOVA test: F (3,168) = 222.4, P < 0.0001, followed by a Tukey’s multiple comparisons test) (Fig. 2a, b). We also observed that Nc-Spain7 showed a higher percentage of multi-infected cells than Nc-Spain1H in F3 cells at 6, 8 and 10 MOIs (two-way ANOVA test: F (3,168) = 93.64, P < 0.0001 followed by a Tukey’s multiple comparisons test) (Fig. 2c). However, no significant differences in the number of multi-infected cells were found between isolates in BCEC-1 cells (Fig. 2d). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig2/ Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 77 Figure 1. Parasite invasion rates in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H isolates. Graphs represent parasite infection rates in F3 (a) and BCEC-1 cells (b) defined as the percentage of invaded tachyzoites (number of events per well) studied at different time points for Nc-Spain7 and Nc-Spain1H. Each column and error bar represent the mean and the SD of 4 replicates from 2 independent assays at the indicated sampling times. The total number of invaded tachyzoites was determined by single immunofluorescence staining of events (parasitophorous vacuoles and lysis plaques) followed by counting using an inverted fluorescence microscope. Significantly higher pInvRs were found in F3 cells compared to BCEC-1 cells infected with Nc-Spain7 (P < 0.01), whereas no differences were found in the pInvRs of F3 and BCEC-1 cells infected by Nc-Spain1H (P > 0.05). * represents significant differences between isolates 3.3 Adhesion-invasion assay In the light of the differences in pInvR and cInfR between both cells lines, an adhesion-invasion assay was performed to investigate whether these differences could be attributed to a different adhesion ability of the tachyzoites in these two cell lines, a different ability to penetrate in the cells or both (Fig. 3a). In this assay, non-adhered tachyzoites were eliminated in the washing step at 4 hpi before the fixation and extra- and intracellular adhered tachyzoites were counted. The percentage of intracellular tachyzoites respect to the total adhered intra- and extracellular tachyzoites was calculated. Surprisingly, both isolates showed that almost 100% of adhered tachyzoites were intracellular at 4 hpi in BCEC-1 cells, whereas a minor percentage of intracellular tachyzoites was observed in F3 cells, 88 and 69% for Nc-Spain7 and Nc-Spain1H, respectively. A significantly higher number of adhered tachyzoites from both isolates were internalized in BCEC-1 cells than in F3 cells at 4 hpi (Chi-square test: χ 2 = 287.6, df = 3, P < 0.0001) (Fig. 3b). Differences between isolates were not observed in BCEC-1, although the high-virulence isolate Nc- Spain7 showed a better ability to penetrate than the low-virulence isolate Nc-Spain1H in F3 cells (Fisher’s exact test: P < 0.0001) (Fig. 3b). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig3/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig3/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig3/ Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 78 Figure 2. Infection and multi-infection rates in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H isolates. Graphs represent the cell infection rates as the percentage of infected cells in F3 (a) and BCEC-1 cells (b) for both isolates and the percentage of cells with multi-infection (more than one parasitophorous vacuole) in F3 (c) and BCEC- 1 cells (d). Each column and error bar represents the mean and the SD of 4 replicates from 2 independent assays using different MOIs. The total number of cells, the number of infected cells and the number of cells with multi-infection were determined by double immunofluorescence staining followed by counting using an inverted fluorescence microscope. The cInfRs were higher in F3 than in BCEC-1 cells infected by both isolates (P < 0.0001). The percentage of cells containing more than a single vacuole was also higher in F3 than in BCEC-1 cells infected by both isolates (P < 0.05). * represents significant differences between isolates 3.4 Proliferation kinetics, doubling time comparisons and tachyzoite yield determination An in vitro intracellular proliferation assay was carried out to study proliferation and egress events of the lytic cycle of N. caninum in trophoblast and caruncular cell cultures. Proliferation kinetics over time assessed by qPCR are represented in Fig. 4a, b. The growth curves of both isolates in F3 and the growth curve of Nc-Spain7 in BCEC-1 adjusted to exponential growth from 10 hpi to 70 hpi, whereas the growth curve of Nc-Spain1H in BCEC-1 did not adjust either to the exponential or linear growth mathematical model. Analysing the Td, we observed a delay in the multiplication of N. caninum in BCEC- 1 cells, with the average Td value of Nc-Spain7 1.5- times higher in BCEC-1 cells (14.603 ± 1.428) than in F3 cells (9.425 ± 0.239) (one-way ANOVA: F (2,21) = 6.966, P = 0.0048 followed by a Tukey’s multiple comparisons test). The Nc-Spain1H isolate showed an average Td value of 12.246 ± 0.893 in F3 cells. The Td value for Nc-Spain1H in BCEC-1 could not be calculated due to the lack of exponential growth of Nc-Spain1H in BCEC-1. Nevertheless, no significant differences were found in the average Td values for Nc-Spain7 and Nc-Spain1H isolates in F3 cells. A microscopic examination of cultures fixed at different time points showed that the multiplication of Nc-Spain7 and Nc-Spain1H isolates began between 10 and 22 hpi. Differences in the parasitophorous vacuole size between both isolates were observed in F3 cells from 34 hpi onwards, with https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig4/ Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 79 Figure 3. Adhesion assay in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H at 4 hpi. Double immunofluorescence staining was performed and adhered extracellular tachyzoites were stained with Alexa Fluor® 488 (green) and Alexa Fluor® 594 (red), whereas intracellular tachyzoites were stained with Alexa Fluor® 594 (red). Nuclei were stained with DAPI (blue). Tachyzoites were counted in 10 arbitrarily selected fields, and the percentage of intracellular tachyzoites relative to the number of total adhered tachyzoites at 4 hpi was calculated. Representative images at a magnification of 1000× (a) show the adhesion assay performed in F3 and BCEC-1 cells infected with both isolates. The graph (b) represents the percentage of intracellular tachyzoites of Nc-Spain7 and Nc-Spain1H relative to the total number of intra- and extracellular tachyzoites adhered to F3 and BCEC-1 cells. Each column and error bar represent the mean and the SD of 4 replicates from 2 independent assays. BCEC-1 cells showed a significantly higher percentage of intracellular tachyzoites than F3 cells (P < 0.0001, Chi-square test). The percentage of intracellular tachyzoites for Nc-Spain7 (88%) was significantly higher than for Nc-Spain1H (69%) in F3 (P < 0.0001), whereas the percentage of intracellular tachyzoites of both isolates in BCEC-1 was the same (96%). * represents significant differences between isolates. Scale-bars: a, 40 μm Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 80 bigger vacuoles in Nc-Spain7 infected cells. However, no differences between the isolates were demonstrated by immunofluorescence in BCEC-1 cells. Between 58 and 82 hpi, asynchronous rupture of host cells and egress of the tachyzoites were observed in F3 cells. However, interestingly, an early egression of tachyzoites from 22 hpi onwards was observed in BCEC-1 (Fig. 4c). The TY58h was assessed to determine the number of tachyzoites produced during the same intracellular period after invasion, prior to complete tachyzoite egress from cell cultures (Fig. 4d). The TY58h was 15- times higher in F3 cells than in BCEC-1 cells infected with Nc-Spain7, and 10-times higher in F3 cells infected with Nc-Spain1H. Comparing the isolates, Nc-Spain7 showed a higher TY58h than Nc-Spain1H in F3 (one-way ANOVA: F (3,28) = 37.35, P < 0.0001 followed by a Tukey’s multiple comparisons test), whereas no differences in the TY58h were found in BCEC-1. The results obtained in the present work are summarized in Table 1. 4. Discussion In the present study, we established for the first time a species- and organ-specific in vitro model for each of the host cell layers in the maternal-foetal interface of the bovine placenta to study N. caninum infection. To date, several in vitro studies have been carried out using established cell lines, such as Marc-145, HeLa, BeWo or ovine trophoblast cells, to investigate the invasion and proliferation of different N. caninum isolates (Haldorson et al., 2005; Carvalho et al., 2010; Regidor-Cerrillo et al., 2011). There is only one (limited) descriptive study concerning the interaction of the parasite with bovine trophoblast cells (Machado et al., 2007), and no data are available about the parasite’s interaction with bovine caruncular cells. The interaction between N. caninum and these target cells has been studied in the present work, using two isolates of different virulence and the two cell lines that represent the maternal-foetal interface. The cell lines used in this work (bovine trophoblast cells, F3, and bovine caruncular epithelial cells, BCEC-1) were isolated from fifth- and fourth-month pregnant heifers, respectively, and they have maintained at least part of their morphological and functional characteristics (Bridger et al., 2007b; Hambruch et al., 2010; Waterkotte et al., 2011). Thus, they may be a useful tool to investigate the pathways of N. caninum infection during transplacental transmission during the second trimester of pregnancy when the majority of abortions caused by N. caninum occur (Dubey et al., 2007). Investigations were focused on the lytic cycle of the tachyzoites (host-cell invasion, proliferation and egress). The processes implicated in the lytic cycle of the parasite are essential for the invasion of host tissues, the distribution of the parasite through the organism and its distribution to the placenta. As a consequence, abortion or transplacental transmission may occur (Hemphill et al., 2006; Dubey et al., 2007). Here, two N. caninum isolates with marked differences in virulence were able to establish themselves and multiply both in the maternal epithelium of the caruncle and in the foetal trophoblast, although differences in the infection of both cell types were found. Our results showed a lower infection rate, as well as a lower percentage of cells with multi-infection, for both parasite isolates in BCEC-1 cells relative to F3 cells. Therefore, N. caninum tachyzoites more efficiently infect trophoblast cells compared to bovine caruncular epithelial cells, meaning bovine trophoblast cells are more susceptible to N. caninum infection. In experimental infections, higher parasite burdens and more severe lesions have been found in the foetal part of the placenta (Caspe et al., 2012; Regidor-Cerrillo et al., 2014). As previously demonstrated, BCEC-1 is an established cell line showing in vitro characteristics of a morphologically and functionally intact epithelial barrier with apical microvilli and junctional complexes (zonula occludens, zonula adherens and desmosomes) (Bridger et al., 2007a; Bridger et al., 2007b) as described in vivo (Björkman, 1973; Leiser, 1975). This polarized barrier, with apical junctional complexes obliterating the paracellular space, establishes an effective paracellular barrier to diffusion of fluid and solutes, limiting the passage of foetal and maternal metabolites (Bridger, 2008). These characteristics may be hindering the paracellular passage of N. caninum across the epithelium. Foetal cells (F3) share many properties with maternal cells (BCEC-1), including apical microvilli and expression of the tight junctional zonula occludens protein both in vivo (Björkman, 1973; Leiser, 1975) and in vitro (Hambruch et al., 2010); however, in contrast to the maternal BCEC- 1 cells, mononuclear trophoblast cells have phagocytic phenotypes in vivo (Schlafer et al., 2000). Also bovine F3 cells may form binucleated cells (Hambruch et al., 2010), which have https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig4/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/figure/Fig4/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5634964/table/Tab1/ Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 81 Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 82 Figure 4. Proliferation kinetics over time and tachyzoite yield at 58 hpi. Graphs (a and b) represent the average number of tachyzoites for each time-point for all individual experiments with an R2 > 0.95, except for BCEC-1 cells infected by Nc-Spain1H, which showed a nonexponential growth pattern. Error bars indicate the SD. Representative images (c) show the proliferation kinetics over time of Nc-Spain7 and Nc-Spain1H isolates in F3 and BCEC-1 cultures. The bar graph (d) represents the tachyzoite yield at 58 hpi for Nc-Spain7 and NcSpain1H in F3 and BCEC-1 cells. Each column and error bar represent the mean and the SD of 4 replicates from 2 independent assays. The TY58h was fifteen times higher in F3 cells than in BCEC-1 cells infected with Nc-Spain7, and ten times higher in F3 cells infected with Nc-Spain1H. Statistical differences were found in the TY58h between isolates in F3 cells, with the TY58h of Nc-Spain7 significantly higher than for Nc-Spain1H (P < 0.0001). * represents significant differences between isolates. Scale-bars: c, 10 μm phagocytic activity as has been previously described for trophoblast giant cells of various species (Schlafer et al., 2000; Amarante-Paffaro et al., 2004; Bevilacqua et al., 2010). The phagocytic activity of mono- and binucleate trophoblast cells may be mediating parasite passage to the foetus (Machado et al., 2007). In fact, in a BALB/c mouse model infected with T. gondii, a higher frequency of infected placentas was observed at later stages of pregnancy, which has been correlated with a higher phagocytic efficiency of the placental tissues in this period (Wujcicka et al., 2014). Therefore, although differences in junctional complexes between both cell types should be investigated to evaluate their influence in parasite invasion, it seems probable that the phagocytic ability of trophoblast cells may partially explain the higher susceptibility to parasite invasion observed in these cells. Noting the differences in the invasion and infection rates between both cell lines, an adhesion-invasion assay was performed in order to elucidate whether these differences could be associated with lower adhesion, penetration or both. Our results revealed that, contrary to expectations, both isolates, which had presented lower cell infection rate in BCEC-1 cells compared to F3 cells, showed a highly efficient invasion with almost 100% penetration of the adhered tachyzoites in BCEC-1 cells at 4 hpi. Thus, the lower invasion of N. caninum observed in BCEC- 1 cells may be due to a lower ability to adhere or to a fragile adhesion to host-cell receptors. Tachyzoite adhesion occurs in two phases, as previously described (Hemphill et al., 2006). The first step is the establishment of low-affinity contact between tachyzoites and the host-cell surface membrane, where surface antigens of N. caninum tachyzoites such as NcSAG1 and NcSRS2 are involved. Later, the actual adhesion process occurs via to microneme proteins (especially NcMIC3), which bind to host- cell surface chondroitin sulfates. Studying the differences in the type and abundance of superficial receptors responsible for the high-affinity interaction with tachyzoites between both cell lines could aid the understanding of the diminished adhesion of N. caninum in bovine caruncular cells. Concerning the growth kinetics of N. caninum in foetal and maternal cells, our results showed a dramatically lower proliferation of both isolates in caruncular cells. Moreover, Nc-Spain1H did not adjust to an exponential growth in maternal cells. Differences observed between both cell types may be partially attributed to a different degree of maturation. While the foetus is not completely immunocompetent in the second trimester of gestation, maternal cells have immunocompetent abilities that may restrict the infection and proliferation of the parasite. In vivo studies have demonstrated the influence of the gestational stage Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 83 Table 1. Summary of virulence traits as a function of cell type and N. caninum isolate apInvR (parasite invasion rate): number of tachyzoites invading the host cell at different time points post-infection bcInfR (cell infection rate): percentage of cells infected using different parasite doses c% Multi-infection: percentage of cells containing more than one vacuole dInvasion efficiency: results from adhesion-invasion assay, percentage of intracellular tachyzoites relative to the total number of tachyzoites at 4 hpi eTY58H (tachyzoite yield at 58 hpi): average number of tachyzoites quantified by qPCR at 58 hpi fTd (Doubling time): period of time required for a tachyzoite to duplicate during the exponential multiplication period, excluding lag and egress phases g NC data not comparable. The Td value for Nc-Spain1H isolate in BCEC-1 cells could not be calculated due to the lack of exponential growth of Nc-Spain1H in BCEC-1 h NS no significant differences +/++/++++ indicate higher rates of each parameter assayed with a significance of P < 0.05, P < 0.01, P < 0.0001, respectively -/- -/- - - - indicate lower rates of each parameter assayed with a significance of P < 0.05, P < 0.01, P < 0.0001, respectively on the outcome of N. caninum infection (Williams et al., 2000; Malet et al., 2003; Macaldowie et al., 2004; Collantes-Fernández et al., 2006). It is known that the survival of the foetus depends on the state of development of its immune system, as higher abortion rates, higher parasite burdens and more severe lesions were observed in foetal tissues when infection occurred in the first and second trimester of gestation (Williams et al., 2000; Macaldowie et al., 2004; Collantes-Fernández et al., 2006c). On the other hand, the lower multiplication of N. caninum in the maternal side of the placenta supports the hypothesis that caruncles act as a barrier, limiting not only parasite infection via reduced adhesion but also its multiplication. In experimental infections, comparisons between cytokine mRNA levels in separated maternal and foetal placental tissues showed that maternal tissue was the major source of most cytokines (Rosbottom et al., 2008) and had a major lymphocyte cell infiltration, particularly in the maternal caruncle (Gibney et al., 2008; Rosbottom et al., 2011), which may indicate that the maternal immune system was actively responding to the parasite. Moreover, early egress was observed in caruncular cells, which could be employed by the parasite as an escape mechanism to facilitate the dissemination of the parasite to the foetal part of the placenta, which has been demonstrated in this work to be the parasite’s preferential target cell, and, thus, allow vertical transmission of the parasite. As mentioned above, the role of the parasite in the outcome of infection is also a determining factor. Processes involved in the lytic cycle, including parasite invasion and intracellular proliferation, are essential for the maintenance and multiplication of the parasite in vitro and for parasite survival and propagation in host tissues during the course of animal infection (Hemphill et al., 2006; Dubey et al., 2007). In previous studies, several N. caninum isolates showed differences during the in vitro lytic cycle and more virulence than others in animal models, associated with higher abortion and transmission rates (Rojo-Montejo et al., 2009a;Rojo-Montejo et al., 2009b; Pereira García- Melo et al., 2010; Regidor-Cerrillo et al., 2010; Regidor-Cerrillo et al., 2011; Regidor-Cerrillo et al., 2014). In trophoblast cells, both isolates, described as “highly prolific” (Nc-Spain7) and “less prolific” (Nc-Spain1H) in previous studies (Regidor-Cerrillo et al., 2011), showed the same in vitro characteristics. In particular, the virulent isolate Nc- Spain7 demonstrated greater invasion, infection and proliferation rates than Nc-Spain1H in trophoblast cells. These differences may be explained by their biological diversity, as has been demonstrated in previous in vivo studies (Rojo- Montejo et al., 2009a; Rojo-Montejo et al., 2009b; Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 84 Regidor-Cerrillo et al., 2011; Regidor-Cerrillo et al., 2012; Regidor-Cerrillo et al., 2013; Regidor-Cerrillo et al., 2014; Regidor-Cerrillo et al., 2015). Nc-Spain7 showed a high neonatal mortality (95%) and vertical transmission rate (nearly 80%) in a pregnant BALB/c mouse model (Regidor-Cerrillo et al., 2010), as well as a percentage of abortion and vertical transmission as high as 100% in a bovine model (Regidor-Cerrillo et al., 2014; Almería et al., 2016a). However, Nc-Spain1H showed a 100% offspring survival rate and a low vertical transmission rate (5%) in a pregnant mouse model (Rojo-Montejo et al., 2009a), and no foetal death was observed in experimentally infected cattle (Rojo-Montejo et al., 2009b). The higher proliferation ability of Nc-Spain7 in trophoblast cells found in the present study may be responsible for the increase in the quantity of parasites reaching the foetal tissues and, consequently, for the enhancement of parasite burdens and pathology, ultimately resulting in foetal death and abortion. These results agree with those obtained in previous studies, where higher parasite burdens in the brain and placental tissues, a wider spread and greater severity of histopathological lesions and clinical signs were observed in animals experimentally infected with Nc-Spain7 (Rojo-Montejo et al., 2009b; Pereira García-Melo et al., 2010; Regidor-Cerrillo et al., 2014). However, in Nc-Spain1H-infected animals, less severe lesions were observed in placentas and maternal and foetal tissues (Rojo-Montejo et al., 2009b), which may explain the absence of abortion. In terms of dissemination in vivo, isolates with low virulence could have a lower efficiency at crossing biological barriers. More interestingly, contrary to our observation in trophoblast cells, the behaviour of both isolates was very similar in bovine caruncular cells. Differences between isolates were limited to a slightly higher infection rate by the virulent isolate Nc-Spain7, whereas adhesion, invasion and proliferation mechanisms were very similar for both isolates. This fact has also been observed in the phylogenetically- related protozoan T. gondii, where comparisons between three strains showed no significant differences in their capacity to infect human placental explants (Robbins et al., 2012). The comparable behaviour showed by different virulence isolates, together with the lower invasion, infection and proliferation rates found in caruncular cells, leads us to hypothesize that isolates may have been selected because of a low virulence in the maternal part of the placenta despite their differences in virulence traits in other host cells, including other placental cells such as trophoblasts. This reduced virulence in the caruncle may facilitate, on the one hand, evasion from maternal immunity and the placental damage caused by parasite multiplication, leading to the abortion. On the other hand, this behaviour may facilitate vertical transmission to the progeny, which is the main route of transmission for N. caninum. In fact, Nc-Spain7 and Nc-Spain1H isolates were obtained from healthy but congenitally infected calves, as described above. 5. Conclusions This is the first study where an in vitro model of N. caninum infection has been implemented in bovine placental cells. Our findings confirm a differential competency of two isolates of N. caninum with different virulence to proliferate in bovine trophoblast cells. However, bovine caruncular cells were the first cell line assayed where different virulence isolates showed similar invasion, adhesion and proliferation kinetics. The low replication of both isolates in the maternal side of the placenta may facilitate the evasion of the immune response by the parasite, allowing their transplacental transmission. This fact may have constituted an evolutionary advantage for these isolates. Remarkably, limited parasite invasion and growth in caruncular cells suggest a putative barrier function for this cell type in the placenta, although early parasite egress may facilitate transmission to offspring. Furthermore, our results confirm the role of foetal trophoblasts as target cells for N. caninum. Future research to determine the differences in surface receptors and cell junctions between both placental cell types are needed. In addition, studies focused on co-cultures of maternal and foetal cells may be helpful as a model to study parasite transport across the maternal epithelium as part of the bovine placental barrier. Finally, the existence of differences in local immunomodulation and cellular mechanisms that take place in the placenta infected by high- and low-virulence isolates of N. caninum should be investigated. Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 85 6. References Almería, S., Serrano-Pérez, B., Darwich, L., Domingo, M., Mur-Novales, R., Regidor-Cerrillo, J., Cabezón, O., Pérez- Maillo, M., López-Helguera, I., Fernández-Aguilar, X., 2016a. Foetal death in naive heifers inoculated with Neospora caninum isolate Nc-Spain7 at 110 days of pregnancy. Exp. Parasitol. 168, 62-69. Amarante-Paffaro, A., Queiroz, G.S., Correa, S.T., Spira, B., Bevilacqua, E., 2004. Phagocytosis as a potential mechanism for microbial defense of mouse placental trophoblast cells. J. Reprod. Fertil. 128 (2), 207-218, 10.1530/rep.1.00214. Benavides, J., Collantes-Fernández, E., Ferre, I., Pérez, V., Campero, C., Mota, R., Innes, E., Ortega-Mora, L.M., 2014. Experimental ruminant models for bovine neosporosis: what is known and what is needed. Parasitology 141 (11), 1471-1488, 10.1017/S0031182014000638. Bevilacqua, E., Hoshida, M.S., Amarante-Paffaro, A., Albieri-Borges, A., Zago Gomes, S., 2010. Trophoblast phagocytic program: roles in different placental systems. Int. J. Dev. Biol. 54 (2-3), 495-505, 10.1387/ijdb.082761eb. Björkman, N., 1973. Fine structure of the fetal-maternal area of exchange in the epitheliochorial and endotheliochorial types of placentation. Cells Tissues Organs 86 (61), 1-22. Bridger, P., Haupt, S., Klisch, K., Leiser, R., Tinneberg, H., Pfarrer, C., 2007a. Validation of primary epitheloid cell cultures isolated from bovine placental caruncles and cotyledons. Theriogenology 68 (4), 592-603. Bridger, P.S., 2008. Validation and establishment of cell culture models to study invasion and feto-maternal interaction in the bovine placentome. (Doctoral Thesis) http://geb.uni-giessen.de/geb/volltexte/2008/5839/ Bridger, P.S., Menge, C., Leiser, R., Tinneberg, H.R., Pfarrer, C.D., 2007b. Bovine caruncular epithelial cell line (BCEC-1) isolated from the placenta forms a functional epithelial barrier in a polarised cell culture model. Placenta 28 (11-12), 1110-1117, S0143-4004(07)00184- 1. Carvalho, J.V., Alves, C.M., Cardoso, M.R., Mota, C.M., Barbosa, B.F., Ferro, E.A., Silva, N.M., Mineo, T.W., Mineo, J.R., Silva, D.A., 2010. Differential susceptibility of human trophoblastic (BeWo) and uterine cervical (HeLa) cells to Neospora caninum infection. Int. J. Parasitol. 40 (14), 1629-1637, 10.1016/j.ijpara.2010.06.010. Caspe, S.G., Moore, D.P., Leunda, M.R., Cano, D.B., Lischinsky, L., Regidor-Cerrillo, J., Álvarez-García, G., Echaide, I.G., Bacigalupe, D., Ortega-Mora, L.M., Odeon, A.C., Campero, C.M., 2012. The Neospora caninum-Spain 7 isolate induces placental damage, fetal death and abortion in cattle when inoculated in early gestation. Vet. Parasitol. 189 (2-4), 171-181, 10.1016/j.vetpar.2012.04.034. Collantes-Fernández, E., Rodríguez-Bertos, A., Arnaiz- Seco, I., Moreno, B., Adúriz, G., Ortega-Mora, L.M., 2006a. Influence of the stage of pregnancy on Neospora caninum distribution, parasite loads and lesions in aborted bovine foetuses. Theriogenology 65 (3), 629-641. Collantes-Fernández, E., Zaballos, A., Álvarez-García, G., Ortega-Mora, L.M., 2002. Quantitative detection of Neospora caninum in bovine aborted fetuses and experimentally infected mice by real-time PCR. J. Clin. Microbiol. 40 (4), 1194-1198. Dellarupe, A., Regidor-Cerrillo, J., Jiménez-Ruiz, E., Schares, G., Unzaga, J.M., Venturini, M.C., Ortega-Mora, L.M., 2014b. Comparison of host cell invasion and proliferation among Neospora caninum isolates obtained from oocysts and from clinical cases of naturally infected dogs. Exp. Parasitol. 145, 22-28, 10.1016/j.exppara.2014.07.003. Dubey, J.P., Buxton, D., Wouda, W., 2006. Pathogenesis of bovine neosporosis. J. Comp. Pathol. 134 (4), 267-289. Dubey, J.P., Schares, G., Ortega-Mora, L.M., 2007. Epidemiology and control of neosporosis and Neospora caninum. Clin. Microbiol. Rev. 20 (2), 323-367. Entrican, G., 2002. Immune regulation during pregnancy and host-pathogen interactions in infectious abortion. J. Comp. Pathol. 126 (2-3), 79-94, 10.1053/jcpa.2001.0539. Gibney, E.H., Kipar, A., Rosbottom, A., Guy, C.S., Smith, R.F., Hetzel, U., Trees, A.J., Williams, D.J., 2008. The extent of parasite-associated necrosis in the placenta and foetal tissues of cattle following Neospora caninum infection in early and late gestation correlates with foetal death. Int. J. Parasitol. 38 (5), 579-588, 10.1016/j.ijpara.2007.09.015. Haldorson, G.J., Mathison, B.A., Wenberg, K., Conrad, P.A., Dubey, J.P., Trees, A.J., Yamane, I., Baszler, T.V., 2005. Immunization with native surface protein NcSRS2 induces a Th2 immune response and reduces congenital Neospora caninum transmission in mice. Int. J. Parasitol. 35 (13), 1407-1415. Hambruch, N., Haeger, J.D., Dilly, M., Pfarrer, C., 2010. EGF stimulates proliferation in the bovine placental trophoblast cell line F3 via Ras and MAPK. Placenta 31 (1), 67-74, 10.1016/j.placenta.2009.10.011. Hemphill, A., Vonlaufen, N., Naguleswaran, A., 2006. Cellular and immunological basis of the host-parasite relationship during infection with Neospora caninum Parasitology 133, 261-278. Innes, E.A., 2007. The host-parasite relationship in pregnant cattle infected with Neospora caninum. Parasitology 134, 1903-1910. Innes, E.A., Wright, S., Bartley, P., Maley, S., Macaldowie, C., Esteban-Redondo, I., Buxton, D., 2005. The host- parasite relationship in bovine neosporosis. Vet. Immunol. Immunopathol. 108 (1-2), 29-36. Leiser, R., 1975. Development of contact between trophoblast and uterine epithelium during the early stages on implantation in the cow. Zentralbl. Veterinarmed. C. 4 (1), 63-86. Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 86 Leiser, R., Kaufmann, P., 1994. Placental structure: in a comparative aspect. Exp. Clin. Endocrinol. 102 (03), 122- 134. Macaldowie, C., Maley, S.W., Wright, S., Bartley, P., Esteban-Redondo, I., Buxton, D., Innes, E.A., 2004. Placental pathology associated with fetal death in cattle inoculated with Neospora caninum by two different routes in early pregnancy. J. Comp. Pathol. 131 (2-3), 142- 156. Machado, R.Z., Mineo, T.W., Landim, L.P.,Jr., Carvalho, A.F., Gennari, S.M., Miglino, M.A., 2007. Possible role of bovine trophoblast giant cells in transplacental transmission of Neospora caninum in cattle. Rev. Bras. Parasitol. Vet. 16 (1), 21-25. Maley, S.W., Buxton, D., Rae, A.G., Wright, S.E., Schock, A., Bartley, P.M., Esteban-Redondo, I., Swales, C., Hamilton, C.M., Sales, J., Innes, E.A., 2003. The pathogenesis of neosporosis in pregnant cattle: inoculation at mid-gestation. J. Comp. Pathol. 129 (2-3), 186-195. Pereira García-Melo, D., Regidor-Cerrillo, J., Collantes- Fernández, E., Aguado-Martínez, A., Del Pozo, I., Minguijón, E., Gómez-Bautista, M., Adúriz, G., Ortega- Mora, L.M., 2010. Pathogenic characterization in mice of Neospora caninum isolates obtained from asymptomatic calves. Parasitology 137(7),1057-1068, 10.1017/S0031182009991855. Pérez-Zaballos, F.J., Ortega-Mora, L.M., Álvarez-García, G., Collantes-Fernández, E., Navarro-Lozano, V., García- Villada, L., Costas, E., 2005. Adaptation of Neospora caninum isolates to cell-culture changes: an argument in favor of its clonal population structure. J. Parasitol. 91 (3), 507-510. Pfarrer, C., Hirsch, P., Guillomot, M., Leiser, R., 2003. Interaction of integrin receptors with extracellular matrix is involved in trophoblast giant cell migration in bovine placentomes. Placenta 24 (6), 588-597. Plattner, F., Soldati-Favre, D., 2008. Hijacking of host cellular functions by the Apicomplexa. Annu. Rev. Microbiol. 62, 471-487. Regidor-Cerrillo, J., Álvarez-García, G., Pastor-Fernández, I., Marugán-Hernández, V., Gómez-Bautista, M., Ortega- Mora, L.M., 2012. Proteome expression changes among virulent and attenuated Neospora caninum isolates. J. Proteomics 75 (8), 2306-2318, 10.1016/j.jprot.2012.01.039. Regidor-Cerrillo, J., Arranz-Solís, D., Benavides, J., Gómez- Bautista, M., Castro-Hermida, J.A., Mezo, M., Pérez, V., Ortega-Mora, L.M., González-Warleta, M., 2014. Neospora caninum infection during early pregnancy in cattle: how the isolate influences infection dynamics, clinical outcome and peripheral and local immune responses. Vet. Res. 45, 10, 10.1186/1297-9716-45-10. Regidor-Cerrillo, J., Díez-Fuertes, F., García-Culebras, A., Moore, D.P., González-Warleta, M., Cuevas, C., Schares, G., Katzer, F., Pedraza-Díaz, S., Mezo, M., Ortega-Mora, L.M., 2013. Genetic diversity and geographic population structure of bovine Neospora caninum determined by microsatellite genotyping analysis. PLoS One 8 (8), e72678, 10.1371/journal.pone.0072678. Regidor-Cerrillo, J., García-Lunar, P., Pastor-Fernández, I., Álvarez-García, G., Collantes-Fernández, E., Gómez- Bautista, M., Ortega-Mora, L.M., 2015. Neospora caninum tachyzoite immunome study reveals differences among three biologically different isolates. Vet. Parasitol. 212 (3-4), 92-99, 10.1016/j.vetpar.2015.08.020. Regidor-Cerrillo, J., Gómez-Bautista, M., Del Pozo, I., Jiménez-Ruiz, E., Adúriz, G., Ortega-Mora, L.M., 2010. Influence of Neospora caninum intra-specific variability in the outcome of infection in a pregnant BALB/c mouse model. Vet. Res. 41 (4), 52, 10.1051/vetres/2010024. Regidor-Cerrillo, J., Gómez-Bautista, M., Pereira-Bueno, J., Adúriz, G., Navarro-Lozano, V., Risco-Castillo, V., Fernández-García, A., Pedraza-Díaz, S., Ortega-Mora, L.M., 2008. Isolation and genetic characterization of Neospora caninum from asymptomatic calves in Spain. Parasitology 135 (14), 1651-1659. Regidor-Cerrillo, J., Gómez-Bautista, M., Sodupe, I., Adúriz, G., Álvarez-García, G., Del Pozo, I., Ortega-Mora, L.M., 2011. In vitro invasion efficiency and intracellular proliferation rate comprise virulence-related phenotypic traits of Neospora caninum. Vet. Res. 42 (1), 41, 10.1186/1297-9716-42-41. Reichel, M.P., Ayanegui-Alcérreca, M.A., Gondim, L.F.P., Ellis, J.T., 2013. What is the global economic impact of Neospora caninum in cattle – The billion dollar question. Int. J. Parasitol. 43 (2), 133-142, 10.1016/j.ijpara.2012.10.022. Robbins, J.R., Zeldovich, V.B., Poukchanski, A., Boothroyd, J.C., Bakardjiev, A.I., 2012. Tissue barriers of the human placenta to infection with Toxoplasma gondii. Infect. Immun. 80 (1), 418-428, 10.1128/IAI.05899-11. Rojo-Montejo, S., Collantes-Fernández, E., Blanco- Murcia, J., Rodríguez-Bertos, A., Risco-Castillo, V., Ortega- Mora, L.M., 2009b. Experimental infection with a low virulence isolate of Neospora caninum at 70 days gestation in cattle did not result in foetopathy. Vet. Res. 40 (5), 49, 10.1051/vetres/2009032. Rojo-Montejo, S., Collantes-Fernández, E., Regidor- Cerrillo, J., Álvarez-García, G., Marugán-Hernández, V., Pedraza-Díaz, S., Blanco-Murcia, J., Prenafeta, A., Ortega- Mora, L.M., 2009a. Isolation and characterization of a bovine isolate of Neospora caninum with low virulence. Vet. Parasitol. 159 (1), 7-16. Rosbottom, A., Gibney, E.H., Guy, C.S., Kipar, A., Smith, R.F., Kaiser, P., Trees, A.J., Williams, D.J., 2008. Upregulation of cytokines is detected in the placentas of cattle infected with Neospora caninum and is more marked early in gestation when fetal death is observed. Infect. Immun. 76 (6), 2352-2361. Rosbottom, A., Gibney, H., Kaiser, P., Hartley, C., Smith, R.F., Robinson, R., Kipar, A., Williams, D.J., 2011. Up regulation of the maternal immune response in the placenta of cattle naturally infected with Neospora caninum. PLoS One 6 (1), e15799, 10.1371/journal.pone.0015799. Chapter ~ IV Results Sub-objective 1.1: Lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro 87 Santos, J.M., Lebrun, M., Daher, W., Soldati, D., Dubremetz, J., 2009. Apicomplexan cytoskeleton and motors: key regulators in morphogenesis, cell division, transport and motility. Int. J. Parasitol. 39 (2), 153-162. Schlafer, D., Fisher, P., Davies, C., 2000. The bovine placenta before and after birth: placental development and function in health and disease. Anim. Reprod. Sci. 60, 145-160. Strahl, H., 1906. Die Embryonalhüllen der Säuger und die Placenta. In Hertwig’s Handbuch Der Vergleichenden Und Experimentellen Entwicklungslehre Der Wirbeltiere. G. Fischer Jena, pp. 235-368. Waterkotte, B., Hambruch, N., Doring, B., Geyer, J., Tinneberg, H.R., Pfarrer, C., 2011. P-glycoprotein is functionally expressed in the placenta-derived bovine caruncular epithelial cell line 1 (BCEC-1). Placenta 32 (2), 146-152, 10.1016/j.placenta.2010.11.009. Williams, D.J., Guy, C.S., McGarry, J.W., Guy, F., Tasker, L., Smith, R.F., MacEachern, K., Cripps, P.J., Kelly, D.F., Trees, A.J., 2000. Neospora caninum-associated abortion in cattle: the time of experimentally-induced parasitaemia during gestation determines foetal survival. Parasitology 121 (4), 347-358. Williams, D.J., Hartley, C.S., Björkman, C., Trees, A.J., 2009. Endogenous and exogenous transplacental transmission of Neospora caninum - how the route of transmission impacts on epidemiology and control of disease. Parasitology 136 (14), 1895-1900, 10.1017/S0031182009990588. Wooding, F., 1992. The synepitheliochorial placenta of ruminants: binucleate cell fusions and hormone production. Placenta 13 (2), 101-113. Wujcicka, W., Wilczyński, J., Nowakowska, D., 2014. Do the placental barrier, parasite genotype and Toll-like receptor polymorphisms contribute to the course of primary infection with various Toxoplasma gondii genotypes in pregnant women? Eur. J. Clin. Microbiol. Infect. Dis. 33 (5), 703-709. Zeiler, M., Leiser, R., Johnson, G.A., Tinneberg, H.R., Pfarrer, C., 2007. Development of an in vitro model for bovine placentation: a comparison of the in vivo and in vitro expression of integrins and components of extracellular matrix in bovine placental cells. Cells Tissues Organs 186 (4), 229-242. Acknowledgement Not applicable. Funding This work was supported by the Spanish Ministry of Economy and Competitiveness (AGL2013–44694-R) and the Community of Madrid (PLATESA S2013/ABI2906). LJP was financially supported by a fellowship from the Complutense University of Madrid and MGS was financially supported through a grant from the Spanish Ministry of Economy and Competitiveness (BES-2014-070723). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Availability of data and materials Not applicable. Authors’ contributions JRC, PH, ECF, MGB and LMO conceived the study and participated in its design. LJP and MGS wrote the manuscript, with interpretation of results and discussion inputs from JRC, ECF, MGB, LMO, NH and CP. LJP and MGS performed in vitro plaque and immunofluorescence assays. JRC, PH, LJP and MGS designed and performed RT-qPCR analyses. NH and CP isolated bovine trophoblast and caruncular cell lines used in the assays. LJP and MGS carried out statistical analyses and interpreted the results. All authors read and approved the final manuscript. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. 89 1SALUVET, Animal Health Department, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain 2Department of Anatomy, University of Veterinary Medicine Hannover, Bischofsholer Damm 15, 30173 Hannover, Germany †Authors contributed equally Published in Parasites & Vectors (8th May 2019) 12:218. DOI: 10.1186/s13071-019-3466-z. Presented as poster communication in the 4th International Meeting on Apicomplexa in Farm Animals (Apicowplexa) (11th-14th October 2017, Madrid, Spain). Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 90 Abstract Background: Bovine neosporosis, one of the main causes of reproductive failure in cattle worldwide, poses a challenge for the immune system of pregnant cows. Changes in the Th-1/Th-2 balance in the placenta during gestation have been associated with abortion. Cotyledon and caruncle cell layers form the maternal- foetal interface in the placenta and are able to recognize and induce immune responses against Neospora caninum among other pathogens. The objective of the present work was to elucidate the immunomodulation produced by high- (Nc-Spain7) and low-virulence (Nc-Spain1H) isolates of N. caninum in bovine trophoblast (F3) and caruncular cells (BCEC-1) at early and late points after infection. Variations in the mRNA expression levels of toll-like receptor-2 (TLR-2), Th1 and Th2 cytokines (IL-4, IL-10, IL-8, IL-6, IL-12p40, IL-17, IFN-γ, TGF-β1, TNF-α), and endothelial adhesion molecules (ICAM-1 and VCAM- 1) were investigated by RT-qPCR, and protein variations in culture supernatants were investigated by ELISA. Results: A similar pattern of modulation was found in both cell lines. The most upregulated cytokines in infected cells were pro-inflammatory TNF-α (P < 0.05–0.0001) and IL-8 (P < 0.05–0.001) whereas regulatory IL-6 (P < 0.05–0.001) and TGF-β1 (P < 0.05–0.001) were downregulated in both cell lines. The measurement of secreted IL-6, IL-8 and TNF-α confirmed the mRNA expression level results. Differences between isolates were found in the mRNA expression levels of TLR-2 (P < 0.05) in both cell lines and in the mRNA expression levels (P < 0.05) and protein secretion of TNF-α (P < 0.05), which were higher in the trophoblast cell line (F3) infected with the low-virulence isolate Nc-Spain1H. Conclusions: Neospora caninum infection is shown to favour a pro-inflammatory response in placental target cells in vitro. In addition, significant immunomodulation differences were observed between high- and low-virulence isolates, which would partially explain the differences in virulence. Keywords: Neospora caninum, Cattle, Immune response, Placenta, Trophoblast, Caruncle, Isolates, Virulence, Cytokines 1. Background Bovine neosporosis is one of the main transmissible causes of abortion in cattle worldwide (Innes et al., 2005; Dubey et al., 2006; Dubey et al., 2007). The etiological agent of bovine neosporosis is Neospora caninum, an obligate intracellular parasite closely related to the zoonotic agent Toxoplasma gondii. Transplacental transmission is the main route of transmission in cattle (Williams et al., 2009) and the placenta can play a key role in the pathogenesis of bovine neosporosis (Entrican, 2002; Innes, 2007). The direct damage produced by the multiplication of the parasite in placental and foetal tissues has been proposed as one of the possible causes of abortion observed during N. caninum infections. Importantly, the placenta is considered to be an immune regulatory organ since it acts as a modulator of foetal and maternal immune responses. In fact, an immune-mediated pathogenesis has also been suggested as a possible cause of abortion (Quinn et al., 2002b). It has been shown that the multiplication of the parasite in the placenta alters the immunological balance at the maternal-foetal interface with an increase of local pro-inflammatory IFN-γ, IL-12p40 and TNF-α cytokines which could compromise the gestation, together with an increase in IL-4 and IL-10 levels (Rosbottom et al., 2008; Regidor-Cerrillo et al., 2014), which avoids the immunological rejection of the foetus but favours the multiplication and vertical transmission of the parasite (Entrican, 2002; Innes et al., 2002). Trophoblast and caruncular cells are able to recognize pathogens and secrete cytokines and chemokines that recruit immune cells in the damaged area (Montes et al., 1995; Steinborn et al., 1998a; Steinborn et al., 1998b). Thus, both cell types play a fundamental role in the initiation of innate immune responses at the placental level as well as in the development of an adaptative immune response for the pregnant dam and foetus. Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 91 Previous in vivo studies have shown the influence of the isolate on the dynamics and outcome of the infection in pregnant bovine models and in the cytokine profiles induced during the infection (Rojo- Montejo et al., 2009b; Caspe et al., 2012; Regidor- Cerrillo et al., 2014; Dellarupe et al., 2014b, Jiménez-Pelayo et al. unpublished data). To date, only one recent study has utilized an in vitro model consisting of immortalized bovine trophoblasts (F3) from the foetus and caruncular cells (BCEC-1) from the dam. The aim of the study was to elucidate the interactions between tachyzoites and the host cells that resemble the maternal-foetal interface of the bovine placentome while also taking into account the influence of the isolate. Maternal cells, where both isolates showed similar phenotypic traits, presented higher resistance to the infection than trophoblast cells, where the high- (Nc-Spain7) and the low-virulence (Nc-Spain1H) isolates showed marked differences in proliferation (Jiménez-Pelayo et al., 2017). However, the interactions between the parasite and the placental target cells from an immunological point of view have not been investigated in vitro until now. Thus, the objective of the present study was to compare the immune response profiles of the bovine placental cells in vitro after the infection with two N. caninum isolates of different virulence. Messenger RNA expression levels of TLR-2, pro-inflammatory cytokines IL-8, IL-12p40, IL-17, IFN-γ, TNF-α, anti- inflammatory/regulatory cytokines TGF-β1, IL-4, IL- 6 and IL-10 as well as ICAM-1 and VCAM-1 endothelial adhesion molecules were determined at 4 and 24 hours post-infection (hpi) in maternal caruncular (BCEC-1) and foetal trophoblast (F3) cell cultures and protein secretion was assessed in culture supernatants by ELISA. 2. Methods 2.1 Parasites and cell cultures A full description of the Nc-Spain1H and Nc-Spain7 parasites and cell cultures of bovine caruncular epithelial (BCEC-1) and bovine trophoblast cells (F3) is provided in a previous report (Jiménez-Pelayo et al., 2017). Briefly, Nc-Spain7 and Nc-Spain1H isolates were obtained from healthy, congenitally infected calves (Regidor-Cerrillo et al., 2008; Rojo- Montejo et al., 2009a) and tachyzoites were maintained in a MARC-145 culture as described previously (Regidor-Cerrillo et al., 2011). The number of culture passages of both N. caninum isolates was limited (passages from 9 to 11) to maintain their biological in vivo behavior (Pérez- Zaballos et al., 2005). The BCEC-1 and F3 cell lines were kindly donated by Dr C. Pfarrer from the University of Veterinary Medicine Hannover and maintained following the protocols described in the literature (Bridger et al., 2007b; Hambruch et al., 2010). 2.2 Infection of the cultures, collection and preservation of the samples BCEC-1 and F3 cells were seeded in 25 cm2 culture flasks adjusting the number of cells in order to obtain a confluent monolayer after 24 h of culture. F3 was seeded at 106 cells per flask, whereas BCEC- 1 was seeded at a concentration of 2 × 106 cells per flask. Tachyzoites were recovered from MARC-145 cultures when most of the parasites were still inside parasitophorous vacuoles; tachyzoites were purified using disposable PD-10 Desalting Columns (G.E. Healthcare, Amersham, UK) as previously described (Regidor-Cerrillo et al., 2011). The parasite viability was checked by trypan blue exclusion, and the tachyzoites were counted. Multiplicity of infection (MOI) of 8 (8 × 106 tachyzoites in F3 and 16 × 106 tachyzoites in BCEC-1) and 10 (107tachyzoites in F3 and 2 × 107 tachyzoites in BCEC-1) from the Nc-Spain7 and Nc-Spain1H isolates, respectively, were inoculated into confluent monolayers of F3 and BCEC-1 quickly after collection. Due to the differences observed in the infection rate between isolates (Jiménez-Pelayo et al., 2017), different MOIs of each isolate were selected with the aim of obtaining cultures infected with the same quantity of each parasite at 4 and 24 hpi. This way possible differences in the modulation of the mRNA expression levels between isolates could be attributed to differences in their biological behavior and not to the differences in the parasite burden. In addition, cultures were infected with high doses of both parasites to get a high infection of the cultures at 4 and 24 hpi so that the RNA from uninfected cells did not mask possible differences in RNA expression levels induced by the infection. The flasks were incubated at 37 °C until collection of the samples. The supernatants were collected at different time points (4, 8, 24 and 56 hpi) and stored at −80 °C for the detection of proteins by ELISA. The cultures were harvested at 4 or 24 hpi by scraping, centrifugation at 1350×g for 15 min at 4 °C and resuspending the pellet in 300 µl of RNAlater® (Qiagen, Hilden, Germany). The samples were stored at −80 °C prior to RNA extraction. Two independent experiments were carried out and four replicates were obtained in each experiment. Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 92 2.3 RNA extraction, reverse transcription and quantitative real-time PCR The mRNA expression levels of TLR-2, pro- inflammatory cytokines IL-6, IL-8, IL-12p40, IL-17, IFN-γ, TNF-α, anti-inflammatory/regulatory cytokines TGF-β1, IL-4 and IL-10 as well as ICAM-1 and VCAM-1 endothelial adhesion molecules were determined by real-time RT-PCR in the F3 and BCEC-1 cell layers infected with the high-virulence isolate (Nc-Spain7) and the low-virulence isolate (Nc-Spain1H) of N. caninum at an early (4 hpi) and a late (24 hpi) time point. RNA was extracted using a commercial Maxwell® 16 LEV simplyRNA Purification kit (Promega, Madison, WI, USA) following the manufacturer’s recommendations. RNA integrity was checked by 1% agarose gel and RNA concentrations were determined using a NanoPhotometer® spectrophotometer (Implen, Munich, Germany). cDNA was obtained by reverse transcription of 2.5 µg of RNA using the master mix SuperScript® VILO™ cDNA Synthesis kit (Invitrogen, Paisley, UK), which was diluted 1:20 in molecular grade water for the qPCR assays. The PCRs were performed using 12.5 µl of Power SYBR® Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA), 10 pmol of each primer (except for TLR-2 primers which were used at a concentration of 22.5 pmol) and 5 μl of diluted cDNA samples in an ABI 7300 Real Time PCR System (Applied Biosystems). The primers used for the qPCR reactions are shown in Table 1. β-Actin and GAPDH were used as housekeeping genes, obtaining comparable Ct values for all the samples. For each target gene, a seven-point standard curve was included in each batch of amplifications based on 10-fold serial dilutions starting at 10 ng/µl of plasmid DNA. The relative quantification of the mRNA expression levels (x-fold change in expression) was carried out by the comparative 2−ΔΔCt method (Schmittgen & Livak, 2008). 2.4 Measurement of cytokines in supernatants of BCEC-1 and F3 cell cultures by ELISA Protein concentrations of the cytokines that showed variations in the mRNA expression levels were determined in the culture supernatants at 4, 8, 24 and 56 hpi using commercial ELISA kits. The levels of IL-6, IL-8 and TNF-α cytokines were measured in the supernatants of the BCEC-1 and F3 cells by sandwich ELISAs using a Bovine IL-6 ELISA Reagent kit (ESS0029; Thermo Fisher Scientific, Waltham, MA, USA), Bovine IL-8 (CXCL8) ELISA Development kit (3114-1A-6; Mabtech AB, Stockholm, Sweden) and Bovine TNF-α ELISA kit (EBTNF; Thermo Fisher Scientific) following the manufacturers’ instructions. The sensitivity limits of these assays were 78 pg/ml for IL-6, 25 pg/ml for IL- 8 and 100 pg/ml for TNF-α. 2.5 Statistical analysis TLR, cytokine and endothelial adhesion molecule mRNA expression levels, as well as differences in the protein secretion between infected and control groups, were analysed using the non-parametric Kruskal–Wallis test, followed by Dunn’s multiple comparison test for all pairwise comparisons. In addition, to assess differences between both infected groups a Mann–Whitney test was performed for each molecule analysed. The statistical significance for all the analyses was established with P < 0.05. GraphPad Prism v.5.01 software (GraphPad Software, San Diego, CA, USA) was used to perform all statistical analyses and create all the graphical illustrations. Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 93 Table 1. Sequences of primers used for cytokine real-time PCR (qPCR) and standard curve data aNCBI accession numbers are for cDNA sequences used in primer design. Primer annealing was also checked with the Bos taurus genomic DNA sequences (http://www.ncbi.nlm.nih.gov/nuccore) bMinimum coefficient of regression (R2) of standard curves for each PCR target in all batches of amplification cStandard curve slopes. Minimum and maximum values for slopes for each PCR target in all batches of amplification dPrimer first described by Arraz-Solís et al. (Arranz-Solís et al., 2016) ePrimer first described by Menzies & Ingham (Menzies & Ingham, 2006) fPrimer first described by Puech et al. (Puech et al., 2015) gPrimer first described by Regidor-Cerrillo et al. (Regidor-Cerrillo et al., 2014) hPrimer described in the present work for the first time 3. Results 3.1 Expression profile of TLR-2 Our results showed that N. caninum infection for 4 h in BCEC-1 cells resulted in a significant upregulation of TLR-2 expression in Nc-Spain1H- infected cells compared with that of negative control cells (Kruskal–Wallis H-test followed by Dunnʼs multiple comparison test: χ2 = 16.2, df = 2, P = 0.0001) and with that of BCEC-1 cells infected with the high-virulence isolate Nc-Spain7 (Mann– Whitney U-test: U(8) = 8, Z = 2.591, P = 0.0007). In F3 cultures, statistical significance was not found at either 4 or 24 hpi between infected groups and the control group. However, comparing both isolates, lower expression of TLR-2 was found in the F3 Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 94 cultures infected with Nc-Spain7 than in the F3 cultures infected with the low-virulence isolate Nc- Spain1H at 4 hpi (Mann–Whitney U- test: U(8) = 16, Z = 2.287, P = 0.0315) (Fig. 1a). 3.2 Pro-inflammatory and regulatory cytokine modulation The pro-inflammatory cytokines IL-8 (Kruskal– Wallis H-test: χ2 = 19.52, df = 2, P < 0.0001 in BCEC-1 and χ2 = 17.56, df = 2, P = 0.0002 in F3) and TNF-α (Kruskal–Wallis H-test: χ2 = 19.73, df = 2, P < 0.0001 in BCEC-1 and χ2 = 19.4, df = 2, P < 0.0001 in F3) were upregulated in both cell types at 4 hpi compared to the respective control groups (Fig. 1b, c). At 24 hpi, IL-8 expression was still increased in BCEC-1 cells infected by both isolates (Kruskal–Wallis H-test followed by Dunnʼs multiple comparison test: χ2 = 16.63, df = 2, P = 0.0003 and χ2 = 10.19, df = 2, P = 0.0117 for Nc-Spain7 and Nc-Spain1H, respectively); however, the increment of IL-8 had disappeared at 24 hpi in F3-infected cells with respect to the control group. When both isolates were compared, Nc-Spain1H induced a higher expression of TNF-α than the high-virulence isolate Nc-Spain7 at 4 hpi in F3 cells (Mann– Whitney U-test: U(8) = 0, Z = 2.579, P < 0.0001). Protein levels of the pro-inflammatory cytokines IL- 8 and TNF-α were also investigated in the supernatant of control and infected cultures at different time-points post-infection. A higher secretion of IL-8 was found for both isolates in BCEC-1 cells at 24 hpi (Kruskal–Wallis H-test: χ2 = 15.87, df = 2, P = 0.0004) and in F3 cells at 56 hpi (Kruskal–Wallis H-test: χ2 = 13.74, df = 2, P = 0.001) with respect to the control group (Fig. 2a, b). Secretion of TNF-α was higher in BCEC-1 cells infected by both isolates (Kruskal–Wallis H-test: χ2 = 18.9, df = 2, P < 0.0001; Fig. 2c) and in F3 cells infected by Nc-Spain1H (Kruskal–Wallis H-test followed by Dunnʼs multiple comparison test: χ2 = 16, df = 2, P < 0.0001) at 8 hpi, although an earlier secretion of TNF-α was also found in F3 cells infected by Nc-Spain1H (χ2 = 14, df = 2, P = 0.0003) at 4 hpi (Fig. 2d, e). As observed with the TNF-α mRNA expression, Nc-Spain1H induced a higher secretion of TNF-α than did the high-virulence isolate Nc- Spain7 at 4 hpi (Mann–Whitney U-test: U(8) = 0, Z = 2.736, P = 0.0002) and at 8 hpi (U(8) = 0, Z = 2.305, P = 0.0002) in the F3 cultures (Fig. 2d, e). The expression levels of other important cytokines associated with N. caninum infection, such as IL- 12p40 and IL-6 (Fig. 1d, e), were modified in placental cells after parasite infection. Specifically, IL-6 levels were downregulated in BCEC-1 infected by Nc-Spain1H and Nc-Spain7 at 4 hpi (Kruskal– Wallis H-test: χ2 = 16.08, df = 2, P = 0.0003) and F3 cultures infected by Nc-Spain1H at 24 hpi (χ2 = 10.5, df = 2, P = 0.0052). IL-12p40 was also downregulated but only in infected F3 cultures at 4 hpi (χ2 = 12.99, df = 2, P = 0.0015). Differences between isolates in the modulation of IL-6 and IL- 12p40 were not found. In addition, we observed that the caruncular cell layer did not express IL- 12p40 mRNA at any time point. The decrease in the expression of IL-6 observed in infected BCEC-1 cells was confirmed by the decrease in the secretion levels of that protein found in the supernatants from BCEC-1 cultures infected with both isolates at 4 hpi, although statistically significant differences were not found (Kruskal–Wallis H-test: χ2 = 2.765, df = 2, P = 0.251), probably because of the high deviation between samples (Fig. 2f). Finally, pro- inflammatory IL-17 and IFN-γ responses were not detected in any cell lines at 4 nor at 24 hpi. We also studied the mRNA levels of the anti- inflammatory cytokines TGF-β1, IL-4 and IL-10. Remarkably, we observed a decrease in the expression levels of TGF-β1 in both cell lines infected with both isolates. Specifically, a decrease was observed at 4 hpi in F3 cultures (Kruskal–Wallis H-test: χ2 = 18.44, df = 2, P < 0.0001) and at 24 hpi in BCEC-1 cultures (χ2 = 12.02, df = 2, P = 0.0025) (Fig. 1f). No differences between isolates were observed in the mRNA expression levels of TGF-β1. There was not detection in bovine placental cells of the anti- inflammatory cytokine IL-4 or the regulatory cytokine IL-10 at 4 and 24 hpi. 3.3 Endothelial adhesion molecule (ICAM-1 and VCAM-1) expression The adhesion molecule ICAM-1 was expressed by both cell lines at 4 and 24 hpi. However, only a slight decrease in the mRNA expression levels of ICAM-1 was observed in the BCEC-1 cultures infected with Nc-Spain7 at 24 hpi compared to the control group although statistical significance was not found (Kruskal–Wallis H-test: χ2 = 5.894, df = 2, P = 0.0525) (Fig. 1g). VCAM-1 expression was detected only in the F3 cultures at 24 hpi, but differences between the infected and the control groups were not found in this culture at this time point (Fig. 1h). Results of mRNA expression levels and protein secretion from statistical tests are reported in Additional file 1: Tables S1 and S2, respectively. Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 95 Figure 1. TLR-2, IL-8, TNF-α, IL-6, IL-12p40, TGF-β1, ICAM-1 and VCAM-1 transcript expression. Scatter-plot graphs of relative mRNA expression levels (as x-fold change) of TLR-2 (a), IL-8 (b), TNF-α (c), IL-6 (d), IL-12p40 (e), TGF-β1 (f), ICAM-1 (g) and VCAM-1 (h) in F3 and BCEC-1 cell cultures at 4 and 24 hpi with Nc-Spain7 and Nc-Spain1H isolates. Data are represented as individual points. Horizontal lines represent median values for each group. ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. Unbracketed symbols represent differences with respect to the control group, while significant differences between isolates are denoted by horizontal square brackets Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 96 4. Discussion Transmission of N. caninum across the placenta makes this organ key in the pathogenesis of bovine neosporosis. Innate immune signalling is crucial at the maternal-foetal interface, where vertical transmission of pathogens to the foetus can have profound pathological outcomes. Trophoblasts and other cell types within the placenta may also be involved in the physiological protection of the placenta (Bevilacqua et al., 2010). Trophoblast cells have been shown to respond to some infections by producing pro-inflammatory cytokines and chemokines and endometrial or decidual cells can produce and secrete a variety of cytokines, participating in the attraction and activation of immune effector cells (Mitsunari et al., 2006; Gillaux et al., 2011). However, these innate immune mechanisms are unexplored at the maternal-foetal interface during N. caninum infection in pregnant cattle (Marin et al., 2007). The expression of TLRs has been described in trophoblasts and other cell types within the placenta (Koga & Mor, 2008). Specifically, TLR-2 was overexpressed in bovine trophoblast cell cultures at 8 hpi (Horcajo et al., 2017) and TLR-3, 7 and 8 have been implicated in N. caninum recognition in the bovine placenta (Marin et al., 2017a; Marin et al., 2017b). In our study, differential activation of TLR-2 in the F3 and BCEC- 1 cultures was observed. Figure 2. IL-8, TNF-α and IL-6 secretion levels in culture supernatants. Scatter-plot graphs representing the concentration of IL-8 (pg/ml) in BCEC-1 (a) and F3 (b) supernatants infected with Nc-Spain7 and Nc-Spain1H at 24 and 56 hpi, respectively, the concentration of TNF-α (pg/ml) in the BCEC-1 supernatants at 8 hpi (c) and in the F3 supernatants at 4 hpi (d) and 8 hpi (e), and the concentration of IL-6 (pg/ml) in the BCEC-1 supernatants at 4 hpi (f). Data are represented as individual points. Horizontal lines represent median values for each group. ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. Unbracketed symbols represent differences with respect to the control group, while significant differences between isolates are denoted by horizontal square brackets Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 97 An upregulation of TLR-2 was found in BCEC-1- infected cultures, especially in those infected with the low-virulence isolate Nc-Spain1H. The caruncular part of the placentome showed a higher expression of several TLRs, suggesting that the initial recognition of N. caninum at the placental level would occur in the maternal side of placenta (Marin et al., 2017a). Taking into account the data shown in Jiménez-Pelayo et al. (2017), which confirm the higher proliferation of N. caninum in trophoblast cells, an important role of placental TLR-2 in the immune response against N. caninum seems plausible. TLR activation is crucial for initiating the innate immune responses responsible for the elimination of intracellular parasites such as N. caninum, and the signalling pathway activated by TLR-2 leads to an increase in the transcription factors NF-κβ and AP-1, which trigger the synthesis of pro-inflammatory cytokines (TNF-α, IL-6, IL-12 and IL-1β) and chemokines (IL-8, RANTES) (Liu & Cao, 2016). Despite differential modulation of host TLR-2, both cell types presented a similar variation in the IL-6, TNF-α and IL-8 expression levels in infected cultures. The pro- inflammatory IL-8 and TNF-α cytokines were upregulated, and secretion of the proteins in the supernatants of both cell lines was also detected by ELISA. IL-8, a cytokine with neutrophil chemotactic/ activating and T-cell chemotactic activity both in vivo and in vitro, is important in the recruitment of leukocytes to the endometrium and may be a potential mediator of placental macrophage infiltration (Adams & Rlloyd, 1997), which might help to eliminate the parasite. IL-8 upregulation has already been observed in bovine umbilical vein endothelial cells (BUVECs) infected by T. gondii and N. caninum (Taubert et al., 2006) as well as in bovine trophoblastic cells and placentomes from cows infected with Brucella abortus (Carvalho et al., 2008). TNF-α is an inflammatory cytokine whose expression has also been described for epithelial cells (Haider & Knöfler, 2009). TNF-α is expressed in all cell types of the trophoblastic lineage and provokes a variety of biological effects on placental and endometrial cell types (Haider & Knöfler, 2009). In addition, TNF-α, through the NF-κβ signalling pathway, coordinates the inflammatory response via the induction of other cytokines (IL-1 and IL-6) and chemokines (IL-8) and via the upregulation of adhesion molecules (ICAM-1 and VCAM-1) (Haraldsen et al., 1996; Cavalcanti et al., 2012), playing a role favouring protective immunity in infectious diseases (Robbins et al., 2012). There are several lines of experimental evidence indicating that TNF-α plays a role not only in immunity to N. caninum but also in the immunopathology of neosporosis. TNF-α expression and secretion may reduce the parasite presence in the placenta by inhibiting the intracellular multiplication of the parasite (Yamane et al., 2000) and participating in parasite proliferation control mechanisms (Jesus et al., 2013); however, TNF-α expression is detrimental to pregnancy maintenance (Rosbottom et al., 2008). IL-6 expression levels were diminished in infected cultures of BCEC-1 and F3 at 4 and 24 hpi, respectively. The classification of IL-6 as Th1 or Th2 has been considered controversial since it can have characteristics of both depending on the dose, the cellular source and the gestational stage studied (Jauniaux et al., 1996). Currently, the presence of IL- 6 displaces the Th1/Th2 balance towards a Th2 response (Diehl & Rincón, 2002). However, in vivo models of N. caninum infection have shown IL-6 upregulation (Pinheiro et al., 2010; Almería et al., 2011; Jiménez-Pelayo et al. unpublished data). This response pattern may be related to a protective action that protects the foetus and allows gestation even if the animals are born infected (Innes et al., 2002). The decrease in IL-6 observed could be explained by the following: (i) IL-6 expression levels were affected by the high antigenic dose administered (MOI 10), resulting in downregulation (Liao et al., 2018); (ii) the time points were not adequate for detecting the peak of IL-6 expression, and the observed decrease may be the consequence of the rapid reduction in IL-6 expression after a peak of expression (Gayle et al., 2004; Ashdown et al., 2006; Beloosesky et al., 2006); or (iii) other cell types are implicated in the upregulation of IL-6 that was observed in vivo. The anti-inflammatory cytokine TGF-β1 was also found to be downregulated in F3 and BCEC-1 cultures at 4 and 24 hpi, respectively. Several members of the TGF-β superfamily have been suggested to regulate trophoblast cell functions, and their dysregulation has been implicated in pregnancy-associated diseases. TGF-β1 is crucial in neutralizing the inflammatory responses induced by Th1-type cytokines (Entrican, 2002). This effect has already been observed in previous works where the reduction of TGF-β1 was shown to be beneficial for controlling N. caninum growth but detrimental for the adequate maintenance of pregnancy (Almería et al., 2011; Arranz-Solís et al., 2016). The reduction of pro-inflammatory IL-12p40 observed in trophoblast cells, together with the lack of expression of IL-12p40 in BCEC-1 cultures, disagrees with the results of previous experimental infections (Rosbottom et al., 2008; Rosbottom et al., 2011; Almería et al., 2011; Regidor-Cerrillo et al., 2014; Almería et al., 2016). Similarly, the lack of expression of pro-inflammatory IFN-γ, essential for controlling parasite infection (Baszler et al., 1999; Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 98 Innes et al., 2005); and anti-inflammatory IL-4 and IL-10, related to placental protection during N. caninum infections (Rosbottom et al., 2008; Regidor-Cerrillo et al., 2014), lead us to hypothesize that the upregulation of IL-12p40, IFN-γ, IL-4 and IL- 10 observed in vivo could be attributed to immune cells present in the placenta, such as dendritic cells, NK cells or macrophages. Therefore, trophoblast and/or caruncular cells would not be responsible for the direct production of these cytokines, although the assayed time points may not have been appropriate for their detection. Finally, ICAM-1 and VCAM-1 expression were not modulated by the parasite infection. These adhesion molecules participate in the recruitment of inflammatory immune cells (Etienne-Manneville et al., 1999) and promote the adherence of monocytes to endothelial cells (Deisher et al., 1993). The upregulation of ICAM-1 has been observed in in vitro infections with apicomplexan parasites (Taubert et al., 2006; Silva et al., 2015; Maksimov et al., 2016). The absence of modulation observed in this work may be explained by differences in the timing of the expression of ICAM- 1 and VCAM-1 (Taubert et al., 2006; Silva et al., 2015; Maksimov et al., 2016) or by the lack of stimuli such as the acute-phase protein C-reactive protein (CRP), which is produced by the liver in response to IL-6 (Zhang et al., 2008). As mentioned above, the parasite isolate is a key factor in the outcome of the infection. In general, differences in the modulation between high- and low-virulence isolates were not remarkable in trophoblast or caruncular cells, with the exception of the mRNA expression levels of TLR-2 and TNF-α. TLR-2 levels were more upregulated by Nc-Spain1H infection than by Nc-Spain7 infection in both cell lines, which led us to hypothesize that the high- virulence isolate would activate less of the TLR recognition system, reducing the immune responses triggered by TLR-2. The inhibition of the TLRs implicated in the recognition of Trypanosoma cruzi and T. gondii in HPCVE increased the parasite burden and, importantly, TLR-2 inhibition prevented the secretion of IL-6 and IL-10, increasing parasite damage (Castillo et al., 2017a; Castillo et al., 2017b). The low-virulence isolate Nc-Spain1H activates the expression of TLR-2, starting an inflammatory response, which may be the cause of the lower proliferation of this isolate (Regidor- Cerrillo et al., 2011; Jiménez-Pelayo et al., 2017), in addition to being one of the causes that explains the higher levels of TNF-α in Nc-Spain1H-infected cells, especially in trophoblast cells. Our results suggest that differential activation of the TLRs by the isolates of differing virulences should be subject to future research since they may be responsible for the biological differences observed both in vitro and in vivo. The low-virulence isolate Nc-Spain1H also induced higher expression of TNF-α in F3. A higher TNF-α response may more efficiently control the proliferation of Nc-Spain1H in F3 cultures, which could explain the observations made by Jiménez- Pelayo et al. (2017) where lower proliferation of Nc- Spain1H was observed in these cells. The lower expression of TNF-α observed during the early stage of infection of trophoblasts with the high-virulence isolate Nc-Spain7 supports the hypothesis that this isolate may modify by yet unknown mechanisms the pro-inflammatory response by trophoblast cells. However, how Nc-Spain7 is able to evade the immune response and maintain lower levels of TNF- α expression in F3 remains unknown. On the other hand, these results suggest that pro-inflammatory cytokines such as TNF-α could have a minor impact in placental damage than postulated in previous works (Rosbottom et al., 2008; Almería et al., 2017), but other mechanisms should be implicated in placental damage in vivo and the occurrence of abortion, such as the high multiplication ability showed by the high-virulence isolate Nc-Spain7 (Jiménez-Pelayo et al., 2017). 5. Conclusions The results presented in this manuscript suggest that placental cells participate in the innate immune response at the maternal-foetal interface via a rapid pro-inflammatory response characterized by the overexpression of IL-8 and TNF-α and the downregulation of TGF-β1 and IL-6. Slight differences were detected when the immunomodulatory response induced by the high and low virulent N. caninum isolates was compared. The higher expression of TLR-2 in the F3 and BCEC- 1 cells and the TNF-α in F3 cells infected with the low-virulence isolate Nc-Spain1H may indicate a higher stimulation of the immune response by this isolate or a higher immunomodulation of Nc- Spain7, which could explain the biological differences observed in vitro and in vivo. F3 and BCEC-1 cultures seem to be a good tool for the study of the TLR activation mechanisms by N. caninum. Finally, we observed that cytokines such as IFN-γ, IL-4 or IL-10, which are commonly upregulated in the placenta after N. caninum infection, are not expressed in F3 and BCEC-1 cells; we conclude that the trophoblast and caruncular epithelial cells are not implicated in the production of these cytokines in the placenta or that other pathways/cells/molecules are needed for their production. Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 99 6. References Adams, D.H., Rlloyd, A., 1997. Chemokines: leucocyte recruitment and activation cytokines. Lancet 349 (9050), 490-495. Almería, S, Serrano-Pérez, B, López-Gatius, F, 2017. Immune response in bovine neosporosis: Protection or contribution to the pathogenesis of abortion. Microb. Pathog. 109, 177-182, 10.1016/j.micpath.2017.05.042. Almería, S., Araujo, R., Darwich, L., Dubey, J., Gasbarre, L., 2011. Cytokine gene expression at the materno‐foetal interface after experimental Neospora caninum infection of heifers at 110 days of gestation. Parasite Immunol. 33 (9), 517-523. Almería, S., Serrano-Pérez, B., Darwich, L., Domingo, M., Mur-Novales, R., Regidor-Cerrillo, J., Cabezón, O., Pérez- Maillo, M., López-Helguera, I., Fernández-Aguilar, X., 2016a. Foetal death in naive heifers inoculated with Neospora caninum isolate Nc-Spain7 at 110 days of pregnancy. Exp. Parasitol. 168, 62-69. Arranz-Solís, D., Benavides, J., Regidor-Cerrillo, J., Horcajo, P., Castaño, P., del Carmen Ferreras, M., Jiménez-Pelayo, L., Collantes-Fernández, E., Ferre, I., Hemphill, A., 2016. Systemic and local immune responses in sheep after Neospora caninum experimental infection at early, mid and late gestation. Vet. Res. 47 (1), 1-13. Ashdown, H., Dumont, Y., Ng, M., Poole, S., Boksa, P., Luheshi, G., 2006. The role of cytokines in mediating effects of prenatal infection on the fetus: implications for schizophrenia. Mol. Psychiatry 11 (1), 47. Baszler, T.V., Long, M.T., McElwain, T.F., Mathison, B.A., 1999. Interferon-gamma and interleukin-12 mediate protection to acute Neospora caninum infection in BALB/c mice. Int. J. Parasitol. 29 (10), 1635-1646. Beloosesky, R., Gayle, D.A., Amidi, F., Nunez, S.E., Babu, J., Desai, M., Ross, M.G., 2006. N-acetyl-cysteine suppresses amniotic fluid and placenta inflammatory cytokine responses to lipopolysaccharide in rats. Obstet. Gynecol. 194 (1), 268-273. Bevilacqua, E., Hoshida, M.S., Amarante-Paffaro, A., Albieri-Borges, A., Zago Gomes, S., 2010. Trophoblast phagocytic program: roles in different placental systems. Int. J. Dev. Biol. 54 (2-3), 495-505, 10.1387/ijdb.082761eb. Bridger, P.S., Menge, C., Leiser, R., Tinneberg, H.R., Pfarrer, C.D., 2007b. Bovine caruncular epithelial cell line (BCEC-1) isolated from the placenta forms a functional epithelial barrier in a polarised cell culture model. Placenta 28 (11-12), 1110-1117, S0143-4004(07)00184- 1. Carvalho Neta, A.V., Stynen, A.P., Paixao, T.A., Miranda, K.L., Silva, F.L., Roux, C.M., TSolís, R.M., Everts, R.E., Lewin, H.A., Adams, L.G., Carvalho, A.F., Lage, A.P., Santos, R.L., 2008. Modulation of the bovine trophoblastic innate immune response by Brucella abortus. Infect. Immun. 76 (5), 1897-1907, 10.1128/IAI.01554-07. Caspe, S.G., Moore, D.P., Leunda, M.R., Cano, D.B., Lischinsky, L., Regidor-Cerrillo, J., Álvarez-García, G., Echaide, I.G., Bacigalupe, D., Ortega-Mora, L.M., Odeon, A.C., Campero, C.M., 2012. The Neospora caninum-Spain 7 isolate induces placental damage, fetal death and abortion in cattle when inoculated in early gestation. Vet. Parasitol. 189 (2-4), 171-181, 10.1016/j.vetpar.2012.04.034. Castillo, C., Muñoz, L., Carrillo, I., Liempi, A., Gallardo, C., Galanti, N., Maya, J.D., Kemmerling, U., 2017a. Ex vivo infection of human placental chorionic villi explants with Trypanosoma cruzi and Toxoplasma gondii induces different Toll‐like receptor expression and cytokine/chemokine profiles. Am. J.Reprod. Immunol. 78 (1), 10.1111/aji.12660. Castillo, C., Muñoz, L., Carrillo, I., Liempi, A., Medina, L., Galanti, N., Maya, J.D., Kemmerling, U., 2017b. Toll-like receptor-2 mediates local innate immune response against Trypanosoma cruzi in ex vivo infected human placental chorionic villi explants. Placenta 60, 40-46. Cavalcanti, Y.V., Brelaz, M.C., Neves, J.K., Ferraz, J.C., Pereira, V.R., 2012. Role of TNF-Alpha, IFN-Gamma, and IL-10 in the Development of Pulmonary Tuberculosis. Pulm. Med. 2012, 10.1155/2012/745483. Deisher, T.A., Haddix, T.L., Montgomery, K.F., Pohlman, T.H., Kaushansky, K., Harlan, J.M., 1993. The role of protein kinase C in the induction of VCAM-1 expression on human umbilical vein endothelial cells. FEBS Lett. 331 (3), 285-290. Dellarupe, A., Regidor-Cerrillo, J., Jiménez-Ruiz, E., Schares, G., Unzaga, J.M., Venturini, M.C., Ortega-Mora, L.M., 2014b. Comparison of host cell invasion and proliferation among Neospora caninum isolates obtained from oocysts and from clinical cases of naturally infected dogs. Exp. Parasitol. 145, 22-28, 10.1016/j.exppara.2014.07.003. Diehl, S., Rincón, M., 2002. The two faces of IL-6 on Th1/Th2 differentiation. Mol. Immunol. 39 (9), 531-536. Dubey, J.P., Buxton, D., Wouda, W., 2006. Pathogenesis of bovine neosporosis. J. Comp. Pathol. 134 (4), 267-289. Dubey, J.P., Schares, G., Ortega-Mora, L.M., 2007. Epidemiology and control of neosporosis and Neospora caninum. Clin. Microbiol. Rev. 20 (2), 323-367. Entrican, G., 2002. Immune regulation during pregnancy and host-pathogen interactions in infectious abortion. J. Comp. Pathol. 126 (2-3), 79-94, 10.1053/jcpa.2001.0539. Etienne-Manneville, S., Chaverot, N., Strosberg, A.D., Couraud, P.O., 1999. ICAM-1-coupled signaling pathways in astrocytes converge to cyclic AMP response element- binding protein phosphorylation and TNF-alpha secretion. J. Immunol. 163 (2), 668-674. Gayle, D.A., Beloosesky, R., Desai, M., Amidi, F., Nuñez, S.E., Ross, M.G., 2004. Maternal LPS induces cytokines in the amniotic fluid and corticotropin releasing hormone in the fetal rat brain. Am. J. Physiol. Regul. Integr. Comp. Physiol. 286 (6), R1024-1029. Gillaux, C., Mehats, C., Vaiman, D., Cabrol, D., Breuiller- Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 100 Fouche, M., 2011. Functional screening of TLRs in human amniotic epithelial cells. J. Immunol. 187 (5), 2766-2774, 10.4049/jimmunol.1100217. Haider, S., Knöfler, M., 2009. Human tumour necrosis factor: physiological and pathological roles in placenta and endometrium. Placenta 30 (2), 111-123. Hambruch, N., Haeger, J.D., Dilly, M., Pfarrer, C., 2010. EGF stimulates proliferation in the bovine placental trophoblast cell line F3 via Ras and MAPK. Placenta 31 (1), 67-74, 10.1016/j.placenta.2009.10.011. Haraldsen, G., Kvale, D., Lien, B., Farstad, I.N., Brandtzaeg, P., 1996. Cytokine-regulated expression of E-selectin, intercellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion molecule-1 (VCAM-1) in human microvascular endothelial cells. J. Immunol. 156 (7), 2558-2565. Horcajo, P., Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Collantes-Fernández, E., Rozas, D., Hambruch, N., Pfarrer, C., Ortega-Mora, L.M., 2017. Transcriptome modulation of bovine trophoblast cells in vitro by Neospora caninum. Int. J. Parasitol. 47 (12), 791- 799, S0020-7519(17)30252-7. Innes, E.A., 2007. The host-parasite relationship in pregnant cattle infected with Neospora caninum. Parasitology 134, 1903-1910. Innes, E.A., Andrianarivo, A.G., Björkman, C., Williams, D.J., Conrad, P.A., 2002. Immune responses to Neospora caninum and prospects for vaccination. Trends Parasitol. 18 (11), 497-504. Innes, E.A., Wright, S., Bartley, P., Maley, S., Macaldowie, C., Esteban-Redondo, I., Buxton, D., 2005. The host- parasite relationship in bovine neosporosis. Vet. Immunol. Immunopathol. 108 (1-2), 29-36. Jauniaux, E., Gulbis, B., Schandene, L., Collette, J., Hustin, J., 1996. Molecular interactions during pregnancy: distribution of interleukin-6 in maternal and embrionyc tissues during the first trimester. Mol. Hum. Reprod. 2, 239-243. Jesus, E.E., Pinheiro, A.M., Santos, A.B., Freire, S.M., Tardy, M.B., El-Bacha, R.S., Costa, S.L., Costa, M.F., 2013. Effects of IFN-gamma, TNF-alpha, IL-10 and TGF-beta on Neospora caninum infection in rat glial cells. Exp. Parasitol. 133 (3), 269-274, 10.1016/j.exppara.2012.11.016. Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Horcajo, P., Collantes-Fernández, E., Gómez-Bautista, M., Hambruch, N., Pfarrer, C., Ortega-Mora, L.M., 2017. Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum. Parasit. Vectors 10 (1), 463, 10.1186/s13071-017-2409-9. Koga, K., Mor, G., 2008. Expression and function of Toll- like receptors at the maternal—fetal interface. Reprod. Sci. 15 (3), 231-242, 10.1177/1933719108316391. Liao, Y., Zhang, Y., Liu, X., Lu, Y., Zhang, L., Xi, T., Shu, S., Fang, F., 2018. Maternal murine cytomegalovirus infection during pregnancy up-regulates the gene expression of Toll-like Receptor 2 and 4 in placenta. Curr. Med. Sci. 38 (4), 632-639. Liu, J., Cao, X., 2016. Cellular and molecular regulation of innate inflammatory responses. Cell. Mol. Immunol. 13 (6), 711. Maksimov, P., Hermosilla, C., Kleinertz, S., Hirzmann, J., Taubert, A., 2016. Besnoitia besnoiti infections activate primary bovine endothelial cells and promote PMN adhesion and NET formation under physiological flow condition. Parasitol. Res. 115 (5), 1991-2001. Marin, M.S., Hecker, Y.P., Quintana, S., Pérez, S., Leunda, M.R., Cantón, G., Cobo, E.R., Moore, D.P., Odeón, A.C., 2017b. Toll-like receptors 3, 7 and 8 are upregulated in the placental caruncle and fetal spleen of Neospora caninum experimentally infected cattle. Vet. Parasitol. 236, 58-61. Marin, M.S., Hecker, Y.P., Quintana, S., Pérez, S., Leunda, M.R., Cantón, G., Cobo, E.R., Moore, D.P., Odeón, A.C., 2017a. Immunization with inactivated antigens of Neospora caninum induces toll-like receptors 3, 7, 8 and 9 in maternal-fetal interface of infected pregnant heifers. Vet. Parasitol. 243, 12-17. Menzies, M., Ingham, A., 2006. Identification and expression of Toll-like receptors 1–10 in selected bovine and ovine tissues. Vet. Immunol. Immunopathol. 109 (1), 23-30. Mitsunari, M., Yoshida, S., Shoji, T., Tsukihara, S., Iwabe, T., Harada, T., Terakawa, N., 2006. Macrophage- activating lipopeptide-2 induces cyclooxygenase-2 and prostaglandin E2 via toll-like receptor 2 in human placental trophoblast cells. J. Reprod. Immunol. 72 (1-2), 46-59. Montes, M.J., Tortosa, C.G., Borja, C., Abadia, A.C., González-Gómez, F., Ruiz, C., Olivares, E.G., 1995. Constitutive secretion of interleukin‐6 by human decidual stromal cells in culture. Regulatory effect of progesterone. Am. J. Reprod. Inmunol. 34 (3), 188-194. Pérez-Zaballos, F.J., Ortega-Mora, L.M., Álvarez-García, G., Collantes-Fernández, E., Navarro-Lozano, V., García- Villada, L., Costas, E., 2005. Adaptation of Neospora caninum isolates to cell-culture changes: an argument in favor of its clonal population structure. J. Parasitol. 91 (3), 507-510. Pinheiro, A.M., Costa, S.L., Freire, S.M., Ribeiro, C.S., Tardy, M., El-Bacha, R.S., Costa, M.F., 2010. Neospora caninum: Early immune response of rat mixed glial cultures after tachyzoites infection. Exp. Parasitol. 124(4):442-447, 10.1016/j.exppara.2009.12.018. Puech, C., Dedieu, L., Chantal, I., Rodrigues, V., 2015. Design and evaluation of a unique SYBR Green real-time RT-PCR assay for quantification of five major cytokines in cattle, sheep and goats. BMC Vet. Res. 11, 65, 10.1186/s12917-015-0382-0. Quinn, H.E., Ellis, J.T., Smith, N.C., 2002b. Neospora caninum: a cause of immune-mediated failure of pregnancy? Trends Parasitol. 18 (9), 391-394. Regidor-Cerrillo, J., Arranz-Solís, D., Benavides, J., Gómez- Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 101 Bautista, M., Castro-Hermida, J.A., Mezo, M., Pérez, V., Ortega-Mora, L.M., González-Warleta, M., 2014. Neospora caninum infection during early pregnancy in cattle: how the isolate influences infection dynamics, clinical outcome and peripheral and local immune responses. Vet. Res. 45, 10, 10.1186/1297-9716-45-10. Regidor-Cerrillo, J., Gómez-Bautista, M., Pereira-Bueno, J., Adúriz, G., Navarro-Lozano, V., Risco-Castillo, V., Fernández-García, A., Pedraza-Díaz, S., Ortega-Mora, L.M., 2008. Isolation and genetic characterization of Neospora caninum from asymptomatic calves in Spain. Parasitology 135 (14), 1651-1659. Regidor-Cerrillo, J., Gómez-Bautista, M., Sodupe, I., Adúriz, G., Álvarez-García, G., Del Pozo, I., Ortega-Mora, L.M., 2011. In vitro invasion efficiency and intracellular proliferation rate comprise virulence-related phenotypic traits of Neospora caninum. Vet. Res. 42 (1), 41, 10.1186/1297-9716-42-41. Robbins, J.R., Zeldovich, V.B., Poukchanski, A., Boothroyd, J.C., Bakardjiev, A.I., 2012. Tissue barriers of the human placenta to infection with Toxoplasma gondii. Infect. Immun. 80 (1), 418-428, 10.1128/IAI.05899-11. Rojo-Montejo, S., Collantes-Fernández, E., Blanco- Murcia, J., Rodríguez-Bertos, A., Risco-Castillo, V., Ortega- Mora, L.M., 2009b. Experimental infection with a low virulence isolate of Neospora caninum at 70 days gestation in cattle did not result in foetopathy. Vet. Res. 40 (5), 49, 10.1051/vetres/2009032. Rojo-Montejo, S., Collantes-Fernández, E., Regidor- Cerrillo, J., Álvarez-García, G., Marugán-Hernández, V., Pedraza-Díaz, S., Blanco-Murcia, J., Prenafeta, A., Ortega- Mora, L.M., 2009a. Isolation and characterization of a bovine isolate of Neospora caninum with low virulence. Vet. Parasitol. 159 (1), 7-16. Rosbottom, A., Gibney, E.H., Guy, C.S., Kipar, A., Smith, R.F., Kaiser, P., Trees, A.J., Williams, D.J., 2008. Upregulation of cytokines is detected in the placentas of cattle infected with Neospora caninum and is more marked early in gestation when fetal death is observed. Infect. Immun. 76 (6), 2352-2361. Rosbottom, A., Gibney, H., Kaiser, P., Hartley, C., Smith, R.F., Robinson, R., Kipar, A., Williams, D.J., 2011. Up regulation of the maternal immune response in the placenta of cattle naturally infected with Neospora caninum. PLoS One 6 (1), e15799, 10.1371/journal.pone.0015799. Schmittgen, T.D., Livak, K.J., 2008. Analyzing real-time PCR data by the comparative C(T) method. Nat. Protoc. 3 (6), 1101-1108. Silva, L.M., Vila-Viçosa, M.J., Cortes, H.C., Taubert, A., Hermosilla, C., 2015. Suitable in vitro Eimeria arloingi macromeront formation in host endothelial cells and modulation of adhesion molecule, cytokine and chemokine gene transcription. Parasitol. Res. 114 (1), 113-124. Steinborn, A., Geisse, M., Kaufmann, M., 1998b. Expression of cytokine receptors in the placenta in term and preterm labour. Placenta 19 (2), 165-170. Steinborn, A., Von Gall, C., Hildenbrand, R., Stutte, H., Kaufmann, M., 1998a. Identification of placental cytokine-producing cells in term and preterm labor. Obstet. Gynecol. 91 (3), 329-335. Taubert, A., Krull, M., Zahner, H., Hermosilla, C., 2006. Toxoplasma gondii and Neospora caninum infections of bovine endothelial cells induce endothelial adhesion molecule gene transcription and subsequent PMN adhesion. Vet. Immunol. Immunopathol. 112 (3-4), 272- 283. Williams, D.J., Hartley, C.S., Björkman, C., Trees, A.J., 2009. Endogenous and exogenous transplacental transmission of Neospora caninum - how the route of transmission impacts on epidemiology and control of disease. Parasitology 136 (14), 1895-1900, 10.1017/S0031182009990588. Yamane, I., Kitani, H., Kokuho, T., Shibahara, T., Haritani, M., Hamaoka, T., Shimizu, S., Koiwai, M., Shimura, K., Yokomizo, Y., 2000. The inhibitory effect of interferon gamma and tumor necrosis factor alpha on intracellular multiplication of Neospora caninum in primary bovine brain cells. J. Vet. Med. Sci. 62 (3), 347-351. Zhang, D., Chen, L., Li, S., Gu, Z., Yan, J., 2008. Lipopolysaccharide (LPS) of Porphyromonas gingivalis induces IL-1β, TNF-α and IL-6 production by THP-1 cells in a way different from that of Escherichia coli LPS. Innate Immun. 14 (2), 99-107, 10.1177/1753425907088244. Acknowledgements Not applicable. Funding This work was supported by the Spanish Ministry of Economy and Competitiveness (AGL2013-44694-R) and the Community of Madrid (PLATESA2-CM P2018/BAA-4370). Laura Jiménez-Pelayo was financially supported by a fellowship from the Complutense University of Madrid and Marta García-Sánchez was financially supported through a grant from the Spanish Ministry of Economy and Competitiveness (BES-2014-070723). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Availability of data and materials Not applicable. Authors’ contributions JRC, PH, ECF, MGB and LMO conceived the study and participated in its design. LJP and MGS wrote the manuscript, with interpretation of results and discussion input from JRC, ECF, MGB, LMO, NH and CP. LJP and MGS performed in vitro infection of the cultures, collection of the samples and ELISA assays. JRC, PH, LJP and MGS designed and performed RT- Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 102 qPCR analyses. NH and CP isolated bovine trophoblast and caruncular cell lines used in the assays. LJP and MGS carried out statistical analyses and interpreted the results. All authors read and approved the final manuscript. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 103 A d d it io n al f ile 1 : T ab le S 1 . S ta ti st ic al t es t re su lt s fo r m R N A e xp re ss io n le ve ls . Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 104 A d d it io n al f ile 1 : T ab le S 1 . C o n ti n u ed . Chapter ~ IV Results Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level 105 A d d it io n al f ile 1 : T ab le S 2 . S ta ti st ic al r es u lt s fo r p ro te in s ec re ti o n . 107 Objetivo 2 Investigación de la dinámica de infección temprana en un modelo bovino gestante inoculado con aislados de alta y baja virulencia de N. caninum en el segundo tercio de la gestación. La neosporosis bovina es una de las causas más importantes de aborto en ganado bovino en todo el mundo, apareciendo con mayor frecuencia durante la mitad de la gestación en condiciones naturales. Además, la variabilidad biológica de N. caninum ha sido ampliamente descrita anteriormente, sin embargo, las causas que la producen no han sido descubiertas todavía. Con el objetivo de confirmar los hallazgos hechos en los ensayos in vitro en células placentarias bovinas de carúncula y cotiledón se implementó un modelo bovino gestante infectado a la mitad de gestación, en el que se compararon la dinámica de la infección así como la respuesta inmunitaria local en la placenta y la modulación de la matriz extracelular (MEC) inducida por dos aislados de distinta virulencia de N. caninum, Nc-Spain7, de alta virulencia, y Nc-Spain1H, de baja virulencia, en dos momentos tempranos tras la infección (10 y 20 días post-infección, dpi). Veinticuatro novillas de la raza Asturiana fueron separadas aleatoriamente en tres grupos: el grupo G-Control, formado por 6 animales, fue inoculado con PBS, el grupo G-NcSpain7 formado por 9 animales fue inoculado por vía intravenosa con 107 taquizoítos del aislado más virulento Nc-Spain7, y el grupo G-NcSpain1H, formado por 9 animales, fue inoculado por vía intravenosa con 107 taquizoítos del aislado menos virulento Nc-Spain1H. Tres animales del G-Control y 4 animales de cada grupo infectado fueron sacrificados en el día 10 pi mientras que el resto de animales fueron sacrificados a los 20 dpi. Las consecuencias clínicas, la distribución del parásito en los tejidos maternos, fetales y placentarios así como el desarrollo de lesiones fue comparado entre los dos aislados a los 10 y 20 dpi. Tras la inoculación con ambos aislados se detectó un pico de fiebre en el día 1 pi, que se repitió en el día 3 pi sólo en el grupo infectado con el aislado más virulento. A los 10 dpi, se detectó ADN del parásito en muestras de placenta de un animal infectado con el aislado Nc- Spain7, asociado a la aparición de necrosis focal. Además, se observó transmisión de este aislado a los tejidos fetales, aunque en este caso no se asoció con presencia de lesiones. Sin embargo, el aislado menos virulento no fue detectado a los 10 dpi en tejidos maternos, fetales ni placentarios, y tampoco se observaron lesiones asociadas a la presencia del parásito. A los 20 dpi, prácticamente el 100% de las muestras de placenta fueron positivas a la presencia de parásito en los animales infectados con el aislado Nc-Spain7, asociado a necrosis focalizada severa. Además, la transmisión al feto se produjo en todos los animales de este grupo, hallándose mortalidad fetal en dos de ellos. Sin embargo, el aislado menos virulento sólo fue detectado en muestras placentarias de un animal a los 20 dpi, que no fueron asociadas a lesiones. Tampoco se detectó transmisión ni lesiones en órganos fetales. En cuanto a la respuesta inmunitaria periférica, todos los animales infectados con Nc-Spain7 habían presentado anticuerpos específicos frente a N. caninum a los 20 dpi mientras que sólo 3 de los 5 animales infectados con Nc-Spain1H habían seroconvertido. Además, dicha seroconversión fue observada antes en los animales infectados con el aislado más virulento. En resumen, los taquizoítos del aislado más virulento, Nc-Spain7, alcanzaron antes la placenta, multiplicándose y dando lugar al desarrollo de lesiones, transmisión 108 y muerte fetal mientras que el aislado menos virulento mostró una infección más tardía de la placenta sin desarrollo de lesiones ni transmisión durante los primeros 20 días tras la infección. Todos estos hallazgos parecen estar relacionados con una mejor capacidad del aislado más virulento para invadir las células de la placenta, así como una mayor capacidad de proliferación, como se demostró in vitro. Por otro lado, las muestras de placenta, tanto materna como fetal, obtenidas en este experimento fueron analizadas por técnicas de inmunohistoquímica y PCR cuantitativa con el objetivo de comparar la respuesta inmunitaria local en la placenta inducida por aislados de distinta virulencia. Además, distintos componentes de la matriz extracelular (MEC) fueron también investigados, ya que la organización de dicha matriz ha sido indicada como uno de los principales puntos de modulación por la infección del parásito y se ha postulado como un mecanismo que podría facilitar el paso de barreras biológicas como la placenta. En general, una respuesta caracterizada por un aumento tanto de las citoquinas Th1 como Th2 (principalmente IFN-γ e IL-4) junto con una respuesta inflamatoria caracterizada por un infiltrado mayoritariamente de linfocitos T CD4+ y CD8+ fueron observadas en todas las muestras infectadas, aunque el incremento tanto de citoquinas como de células inflamatorias fue más marcado a los 20 dpi. Además, se detectaron diferencias interesantes entre ambos aislados. A los 10 dpi se observó un aumento generalizado de la expresión de todas las citoquinas únicamente en los animales infectados con el aislado menos virulento, desencadenando una respuesta inmunitaria frente a este aislado que puede ayudar a controlar la infección. Sin embargo, el reconocimiento del aislado más virulento, Nc-Spain7, no se produjo hasta los 20 dpi, lo que podría contribuir a su multiplicación y diseminación. La detección tardía del parásito, cuando las lesiones ya se han desarrollado, podría dar lugar a una respuesta inmunitaria exacerbada en la placenta, en un intento del hospedador por controlar la infección, que, a su vez, podría contribuir a la muerte fetal. Específicamente, mayores niveles de IL-8, TNF-α e iNOS, mayor infiltración de células inmunitarias y menor expresión del TGF-β1 fueron detectados en las placentas de los animales que presentaron muerte fetal en comparación con aquellos animales infectados con el mismo aislado que no presentaron muerte fetal a los 20 dpi, sugiriendo que éstos pueden estar directamente involucrados en el aborto asociado a N. caninum. Además, una profunda alteración de la MEC, con disminución de la expresión de MMP-2 y TIMP-2 y destrucción de colágeno, fibronectina y vimentina, y la falta de activación de procesos de reparación tisular podrían contribuir al mecanismo de aborto. Finalmente, la expresión de los componentes de la MEC se encontró aumentada en los animales infectados con el aislado menos virulento, Nc-Spain1H, tanto a los 10 como a los 20 dpi, demostrando que dicho aislado es capaz de inducir la remodelación tisular que, por un lado, podría contribuir a la migración de células inmunitarias periféricas al foco de la infección pero, por otro lado, podría contribuir también a la transmisión del parásito al feto. En resumen, se observaron diferentes patrones de interacción parásito- hospedador a nivel de la placenta bovina durante la infección temprana a mitad de gestación con aislados de alta y baja virulencia de N. caninum, lo que sugiere la existencia de diferentes estrategias adaptativas evolutivas utilizadas por este parásito para la transmisión a la descendencia. 109 Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid gestation with high- and low-virulence isolates. Jiménez-Pelayo L1, García-Sánchez M1, Vázquez P1, Regidor-Cerrillo J2, Horcajo P1, Collantes- Fernández E1, Blanco-Murcia J3, Gutiérrez-Expósito D4, Román-Trufero A5, Osoro K5, Benavides J4, Ortega-Mora LM1*. 1Saluvet, Animal Health Department, Faculty of Veterinary Sciences, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain 2Saluvet-innova, Faculty of Veterinary Sciences, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain 3Department of Animal Medicine and Surgery, Faculty of Veterinary Sciences, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain 4Instituto de Ganadería de Montaña (CSIC-Universidad de León), 24346, León, Spain. 5Regional Service for Research and Agri-Food Development (SERIDA), 33300 Villaviciosa, Asturias, Spain *Corresponding author: Luis Miguel Ortega-Mora Running title: Early Neospora caninum infection dynamics in cattle after inoculation at mid- gestation with high (Nc-Spain7)- or low (Nc-Spain1H)-virulence isolates Submitted to Veterinary Research (VETR-D-19-00167) Presented as oral communication in the 4th International Meeting on Apicomplexa in Farm Animals (Apicowplexa) (11th-14th October 2017, Madrid, Spain). Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 110 Abstract Early Neospora caninum infection dynamics were investigated in pregnant heifers intravenously inoculated with PBS (G-Control) or 107 tachyzoites of high (G-NcSpain7)- or low- (G-NcSpain1H) virulence isolates at 110 days of gestation. Serial culling at 10 and 20 days post-infection (dpi) was performed. Fever was detected at 1 dpi in both infected groups (P < 0.0001), and a second peak was detected at 3 dpi only in G- NcSpain7 (P < 0.0001). At 10 dpi, Nc-Spain7 was detected in placental samples from one animal related to focal necrosis, and Nc-Spain7 transmission was observed, although no foetal lesions were associated with this finding. The presence of Nc-Spain1H in the placenta or foetuses, as well as lesions, were not detected at 10 dpi. At 20 dpi, G-NcSpain7 animals showed almost 100% positive placental tissues and severe focal necrosis as well as 100% transmission. Remarkably, foetal mortality was detected in two G-NcSpain7 heifers. Only one animal from G-NcSpain1H presented positive placental samples. No foetal mortality was detected, and lesions and parasite transmission to the foetus were not observed in this group. Finally, 100% of G-NcSpain7 heifers at 20 dpi presented specific antibodies, while only 60% of G-NcSpain1H animals presented specific antibodies at 20 dpi. In addition, earlier seroconversion in G-Nc-Spain7 was observed. In conclusion, tachyzoites from Nc-Spain7 reached the placenta earlier and multiplied, leading to lesion development, transmission to the foetus and foetal mortality, whereas Nc-Spain1H showed delayed infection of the placenta and no lesional development or transmission during early infection. Keywords Neospora caninum, pregnant cattle, mid-gestation, Nc-Spain7, Nc-Spain1H, early infection, virulence 1. Introduction Neospora caninum is an apicomplexan protozoan parasite that is considered to be one of the main causes of abortion in cattle. Horizontal transmission via oocyst ingestion is possible, although transplacental transmission in cattle seems to be the most efficient infection route (Williams et al., 2009). In pregnant cattle, infection with this parasite may lead to abortion, birth of still-born calves, birth of new-born calves with clinical signs or birth of clinically healthy but persistently infected calves (Dubey et al., 2007; Dubey & Schares, 2011). The disease outcome is influenced by several factors, including the maternal immune response in the placenta and the relative immunocompetence of the foetus at the time of infection, which are two key variables (Benavides et al., 2014; Horcajo et al., 2016). Experimental N. caninum infection in pregnant cattle during the first term generally produces foetal death and abortion, and foetuses show more severe lesions (Williams et al., 2000; Regidor-Cerrillo et al., 2014). Experimental infection from the second trimester onward, which is when the foetal immune system begins to develop, generally results in clinically healthy but congenitally infected calves (Williams et al., 2000; Gibney et al., 2008; Almería et al., 2010; Benavides et al., 2012), although infection with the highly virulent isolate Nc-Spain7 induced at least 50% foetal death at 110 days of gestation (dg) (Almería et al., 2016, Vázquez et al., submitted). Under natural conditions, abortion caused by N. caninum is more frequent during the second trimester of pregnancy (Collantes-Fernández et al., 2006a; Almería & López-Gatius, 2013). A limited number of studies have been conducted to investigate the consequences of N. caninum infection at mid-gestation (Maley et al., 2003; Bartley et al., 2004; Almería et al., 2010; Almería et al., 2016; Vázquez et al., submitted). Recently, intravenous (IV) inoculation of 107 tachyzoites of Nc-Spain7 at mid-gestation produced 50% foetal death until 42 days post-infection (dpi) (Almería et al., 2016) and 66.6% foetal death when gestation lasted until term. Moreover, foetal death was Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 111 observed using lower doses of Nc-Spain7 tachyzoites, although a lower percentage of abortions and a delayed presentation were detected as the dose decreased (Vázquez et al., submitted). The outcome of the infection in pregnant cattle also depends on the isolate. Specifically, the high- virulence isolate Nc-Spain7 showed a percentage of abortion and vertical transmission of 100% in a bovine model at early gestation (Caspe et al., 2012; Regidor-Cerrillo et al., 2014), whereas the infection in experimentally infected cattle with the low- virulence isolate Nc-Spain1H did not result in foetal death (Rojo-Montejo et al., 2009b). In the current study, the aim was to investigate how the differences between high (Nc-Spain7)- and low(Nc-Spain1H)- virulence isolates of N. caninum influence the clinical outcome, parasite distribution and burden, lesion development in placental and foetal tissues, and the specific antibody response during early infection in pregnant heifers inoculated at mid-gestation. The lack of bovine models studying early infection and the lack of experimental infections comparing isolates make the implementation of this model necessary to elucidate the pathogenesis of bovine neosporosis at mid-gestation, which is when most abortions occur in naturally infected cattle (Anderson et al., 1995; López-Gatius et al., 2004). 2. Materials and methods 2.1 Animals and experimental design Asturiana heifers, aged 20-30 months, were selected after assessing their seronegativity to N. caninum, Infectious Bovine Rhinotracheitis (IBR) virus, Bovine Viral Diarrhoea (BVD) virus, Leptospira and Mycobacterium avium subsp. paratuberculosis by ELISA. The health and reproductive management of the animals is detailed in Additional File 1. Pregnant heifers (n=24) were randomly distributed into three experimental groups, G-Control (n=6), G- NcSpain7 (n=9) and G-NcSpain1H (n=9) and inoculated intravenously at 110 days of gestation with phosphate buffered saline (PBS) and 107 culture-derived tachyzoites of Nc-Spain7 and Nc- Spain1H isolates, respectively. Three animals from G-Control, four animals from G-NcSpain7 and four animals from G-NcSpain1H were culled at 10 dpi, while three animals from G-Control, five animals from G-NcSpain7 and five animals from G- NcSpain1H were culled at 20 dpi. 2.2 Parasites Nc-Spain7 and Nc-Spain1H tachyzoites were routinely maintained in cultured MARC-145 cells, and inoculum was prepared as described previously (Regidor-Cerrillo et al., 2010). The same limited parasite passage numbers for both isolates were used for the experimental infection (11) to ensure the maintenance of their in vivo biological characteristics and avoid adaptation to the host cell (Pérez-Zaballos et al., 2005). Briefly, tachyzoites were recovered from culture flasks when they were still largely intracellular, and at least 80% of the parasitophorous vacuoles were undisrupted. Tachyzoite numbers were determined by Trypan blue exclusion followed by counting in a Neubauer chamber, and parasites were resuspended in PBS at the required dose of 107 tachyzoites in a final volume of 2 ml. Tachyzoites were administered to heifers within 1 h of harvesting from tissue culture. 2.3 Clinical monitoring and sampling Cattle were observed daily before and after inoculation throughout the entire experimental period. Rectal temperatures were recorded daily from 6 days prior to challenge to 14 dpi and weekly from 14 dpi onward. Animals with temperatures above 39.5 ºC were considered to be febrile. Foetal viability was checked once a week by ultrasound scanning of foetal heartbeat and movements. Blood samples were collected by coccygeal venipuncture at days -6 and -1 and twice a week until the end of the experiment for further analyses. Animals were sedated with xylazine hydrochloride (Rompun; Bayer, Mannheim, Germany) and euthanised by an IV overdose of embutramide and mebezonio iodide (T61; Intervet, Salamanca, Spain). Post-mortem examination of the heifers was performed immediately after euthanasia. Foetuses were separated from the placenta, and 18 placentomes (6 cranial, medial and caudal) were randomly recovered from each placenta. Half of the placentomes from each area were carefully detached by hand, and maternal caruncles (CA) and foetal cotyledons (CO) were separated. Full placentomes were transversally cut in slices measuring 2-3 mm in thickness, which were distributed for storage in 10% formalin (Sigma- Aldrich, Saint Louis, MO, USA) for histopathological examinations. Both full placentomes and CA and CO sections were also stored at -80ºC for parasite DNA detection by PCR. Foetal tissues included the brain (FB), heart, liver (FL), lung and a portion of semitendinosus skeletal muscle, which were maintained at -80 ºC for DNA extraction and fixated in 10% formalin. Blood and foetal thoracic and abdominal fluids were also collected when possible and maintained at -80ºC for serological analysis. Heifer tissues, including pre-scapular and ileofemoral lymph nodes, were also collected for PCR and histopathological analysis. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 112 2.4 Histopathology and lesion quantification After fixation for two weeks, maternal and foetal samples and placentomes were trimmed and conventionally processed for embedding in paraffin wax and haematoxylin and eosin (HE) staining. Histological slides were studied under an optical microscope. Lesion quantification at placentome samples was performed through a computer- assisted morphometric analysis in HE stained sections following a previously described procedure (Arranz-Solis et al., 2015). Among the parameters evaluated were the number and size of necrotic foci (NF and ASF), as well as the total area of necrotic lesions (%LES) affecting the interdigitate area of the placentome. In addition, the accumulation of proteinaceous material (eosinophilic) and cellular debris in the haemophagus subchondral area of the placentome was also measured, and the results are expressed as a ratio between the area occupied by the proteinaceous exudate in the haemophagus zone and the total area of the placentome. 2.5 Tissue DNA extraction and PCR determinations DNA extraction and PCR determinations were carried out as described elsewhere (Regidor- Cerrillo et al., 2014; Arranz-Solís et al., 2015). Briefly, genomic DNA was extracted from 20-100 mg of maternal and foetal tissue samples using the Maxwell® 16 Mouse Tail DNA Purification Kit (Promega, Wisconsin, USA). Parasite DNA was detected by nested PCR adapted to a single tube from the internal transcribed spacer (ITS1) region of N. caninum using TgNN1-TgNN2 and NP1-NP-2 as external and internal primers, respectively (Buxton et al., 1998; Hurtado et al., 2001; Regidor-Cerrillo et al., 2014). DNA quantification was performed by real-time PCR using the equipment ABI 7500 FAST (Applied Biosystems, Foster City, CA, USA) and targeting Nc-5 as described previously (Collantes- Fernández et al., 2002). Detailed information concerning DNA extraction and PCR is given in Additional File 1. 2.6 IFN-γ responses in sera IFN-γ levels in sera from dams were measured by the Bovine IFN-γ ELISA development kit (Mabtech AB, Sweden) following the manufacturer’s recommendations. The colour reaction was developed by the addition of 3,3’,5,5’- tetramethylbenzidine substrate (TMB, Sigma- Aldrich, Spain) and incubated for 5-10 min in the dark. Reactions were stopped by adding 2N H2SO4 when the first point of the standard curve reached a DO of 0.7 at 620 nm. Then, plates were read at 450 nm. The cytokine concentrations were calculated by interpolation from a standard curve generated with recombinant cytokines provided with the Bovine IFN-γ ELISA development kit (Mabtech AB, Sweden). 2.7 N. caninum-specific IgG responses Neospora-specific IgG antibody levels were measured in maternal serum by ELISA (Regidor- Cerrillo et al., 2014). IgG1 and IgG2 subclasses were also assessed by ELISA using peroxidase-conjugated sheep anti-bovine IgG1 and IgG2 antibodies (Serotec, Oxford, UK) at 1:1000 as secondary conjugates. For each plate, the OD values were converted into a relative index percent (RIPC) using the following formula: RIPC = (OD405 sample - OD405 negative control)/(OD405 positive control - OD405 negative control)×100. A RIPC value ≥ 12 indicates a positive result. Indirect fluorescent antibody test (IFAT) and Western blotting (WB) were carried out to detect specific IgG anti-Neospora antibodies in foetal blood and foetal thoracic and abdominal fluids. IFAT was carried out following the methodology previously described (Álvarez-García et al., 2003). Samples were diluted at two-fold serial dilutions in PBS starting at 1:8 up to the end point titre. Intact tachyzoite membrane fluorescence at a titre ≥ 8 was considered a positive reaction. WB was carried out as described previously (Álvarez-García et al., 2002). After blocking overnight, the membranes containing tachyzoite extracts were incubated with foetal sera and fluids diluted 1:20 and incubated for 1.5 h at room temperature. After washing, the membranes were incubated with 1:1200 peroxidase-conjugated monoclonal goat anti- bovine IgG (Thermo Fisher Scientific, Waltham, MA, USA) for 1 h, washed and developed using 4-chloro- 1-naphtol (Bio-Rad Laboratories, CA, USA) as a substrate. 2.8 Statistical analysis Differences in PCR detection of parasite DNA in maternal, foetal and placental tissues were evaluated using X2 or Fisher’s exact F-test. Parasite burdens were analysed using the non-parametric Mann-Whitney U test. Occurrence of foetal death was analysed by the Kaplan-Meier survival method to estimate the percentage of viable foetuses (VF) at each time point (Bland & Altman, 1998). The foetal survival curves of the infected groups were then compared with the Gehan-Wilcoxon test. Differences in histological scoring were analysed using the non-parametric Kruskal–Wallis test followed by Dunn's test for all pairwise comparisons. Finally, a two-way ANOVA test followed by a Tukey´s multiple comparisons test, was performed to compare rectal temperatures, antibody responses and IFN-γ kinetics in sera. https://www.sciencedirect.com/topics/immunology-and-microbiology/monoclonal-antibody Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 113 Statistical significance for all analyses was established with P < 0.05. All statistical analyses were carried out using GraphPad Prism 5 v.5.01 software (San Diego, CA, USA) 3. Results A summary of results (clinical outcome, lesions, parasite distribution and IgG responses) in heifers and foetuses inoculated with PBS, 107 tachyzoites of Nc-Spain7 isolate or Nc-Spain1H isolate at 110 dg and culled at 10 or at 20 dpi is shown in Table 1. 3.1 Clinical observations The mean rectal temperatures of animals from G- Control, G-NcSpain7 and G-NcSpain1H are represented in Figure 1. Five animals from G- NcSpain7 and 5 animals from G-NcSpain1H exhibited fever at 1 dpi. Six animals from G- NcSpain7 also exhibited fever at 3 dpi, and two of these animals maintained fever until 4 dpi The mean rectal temperatures of G-NcSpain7 increased significantly (> 39.5 ºC) at 1 and 3 dpi, and the mean rectal temperature of G-NcSpain1H only increased significantly at 1 dpi compared to the uninfected G-Control group (P < 0.0001; two-way ANOVA test). Significant differences between infected groups were found at 3 dpi when a second peak of fever was detected in G-NcSpain7 but not in G-NcSpain1H (P < 0.0001; two-way ANOVA test). Interestingly, only 5 out of 9 G-Nc-Spain1H heifers presented fever, whereas all G-Nc-Spain7 heifers were febrile at some time during the experimental period. Rectal temperatures of G-Control animals remained below 39ºC. Foetal mortality was not detected until 20 dpi. Foetal death was detected during culling in two heifers (3581 and 7934) from G-NcSpain7 as detailed below. However, foetuses from G-Control and G-NcSpain1H remained viable throughout the experiment. The comparative analysis of foetal survival curves between infected groups showed non-significant differences (P = 0.13; Gehan- Wilcoxon test). 3.2 Pathology and lesion quantification Gross lesions Placental detachment from the uterus, and autolyzed CO, were found in the two heifers (3581 and 7934) from G-NcSpain7 that were culled at 20 dpi and presented foetal mortality. In these cases, foetuses were swollen because of subcutaneous oedema and showed a degree of autolysis. Apart from these findings, no evident gross lesions were found in the placentas, foetuses or maternal lymph nodes studied from any of the other heifers. Microscopic lesions Maternal lymph nodes: Histological changes were not found in any lymph node. Placentomes: Two different histological changes were found in the placentomes. The first change consisted of focal necrosis with a variable degree of inflammatory infiltrate adjacent to the lesion, randomly distributed within the interdigitate zone of the placentome (Additional File 2a, b). This lesion was only found in G-NcSpain7 heifers, but there were differences within this group, as only one heifer culled at 10 dpi (9665) showed this lesion, affecting only one out of nine studied placentomes. The lesion was also found in all G-Nc-Spain7 animals culled at 20 dpi. All the parameters quantified in these lesions, NF, ASF and %LES, were higher in animals culled at 20 dpi than those found in the only animal with placental lesions culled at 10 dpi. Among those culled at 20 dpi, one of them (3581, non-viable foetuses -NVF-) had more NF and more %LES than the rest (P <0.01; Kruskal-Wallis test) (Additional file 3a, c). The ASF was higher in two animals (3581, NVF and 5082, VF) than in the other three (Additional file 3b). When studying the influence of the location of the placentomes (cranial, medial or caudal) on the evaluated parameters, no differences were found between the five heifers culled at 20 dpi. However, when analysing those two animals with higher ASF, medial and caudal CO showed higher %LES (P <0.05; Kruskal-Wallis test) than the cranial ones (Additional file 3c). The second histological change found in placentomes was the accumulation of proteinaceous material (eosinophilic) and cellular debris at the haemophagus area of the placentome, i.e., extravasated plasma (Figure 2a). This accumulation was found in all the animals from the study, including G-Control. However, comparing the amount of extravasated plasma, measured as the relative area occupied by the proteinaceous material in the haemophagus area, there were clear differences between groups (Figure 2b). Placentomes from G-Control showed less accumulation than G-NcSpain7 (P <0.0001; Kruskal- Wallis test) and G-NcSpain1H (P <0.001; Kruskal- Wallis test), but the accumulation in G- NcSpain7was higher than in G-NcSpain1H (P <0.05; Mann-Whitney U test). Comparing the differences within each group depending on the day of culling (10 vs 20 dpi), differences were found only at G- NcSpain7, where animals culled at 10 dpi showed less accumulation than those culled at 20 dpi. Foetal viscera. Only the five foetuses from G- NcSpain7 heifers culled at 20 dpi showed Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 114 histological lesions. The livers of all five foetuses showed perivascular aggregation of lymphocytes, macrophages and plasma cells. NVF (3581 and 7934) also showed mild multifocal necrotic foci in the liver with scant presence of inflammatory cells related to them. In addition, all foetuses from G- NcSpain7 culled at 20 dpi showed scant, randomly distributed aggregation of mononuclear cells in the lung parenchyma. NVF (3581 and 7934) also showed similar lesions in the lungs plus perivascular infiltration of mononuclear cells and, in one of them (7934), mild multifocal necrosis. Finally, we also found mild mononuclear myositis and myocarditis (Additional File 2c) in all five foetuses. Multifocal randomly distributed small aggregation of mononuclear cells in the neuropil of the brain (Additional file 2d) was observed only in VF as the brain samples from NVF were too autolytic to allow proper histological evaluation. Figure 1. Rectal temperatures. The mean rectal temperatures of animals from G-Control, G-NcSpain7 and G-NcSpain1H during the experiment are represented in the graphic (a). The exact temperatures of each infected animal recorded during the first 5 dpi are represented in table (b). **** indicates P < 0.0001 significant differences. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 115 Figure 2. Proteinaceous exudate at the haemophagus area of the placentome. Representative image of accumulation of proteinaceous material and cellular debris at the haemophagus subchondral area of a placentome from G- NcSpain1H at 20 dpi. HE. 2× (a), and dot-plot graph showing significant differences between groups (b). Each dot represents individual values of relative area occupied by the exudate in each placentome analysed, and medians are represented as horizontal lines. ****, *** and ** indicate P < 0.0001, P < 0.001 and P < 0.01 significant differences. Bar 1000 µm. 3.3 Parasite distribution and burden in placental and foetal tissues Parasite burdens are represented in Figure 3. Maternal tissues N. caninum DNA was only detected in 1 out of 15 pre-scapular lymph node samples in one heifer from G-NcSpain1H culled at 20 dpi (9638). Placental tissues In G-NcSpain7, Neospora DNA was detected sporadically in CA (4/36) and CO (1/36) samples belonging to one animal culled at 10 dpi (9665) and in 44 out of 45 CA and 44 out of 45 CO samples of animals culled at 20 dpi. The differences in the frequency of parasite detection between animals from G-NcSpain7 culled at 10 and at 20 dpi were statistically significant for CA and for CO (P < 0.0001; Fisher exact test). In G-NcSpain1H, all CA and CO samples from animals culled at 10 dpi were negative, and only 1 out of 45 CA and 4 out of 45 CO samples from one animal culled at 20 dpi (9638) were N. caninum DNA positive. The frequency of detection in CA and CO in G-NcSpain7 was significantly higher than in G-NcSpain1H culled at 20 dpi (P < 0.0001; Fisher exact test). Placental samples from G-Control animals were negative. The parasite burden in CA and CO, measured as the number of tachyzoites per mg of tissue, was analysed in N. caninum DNA-positive samples. Higher parasite burdens were found in CA and CO from G-NcSpain7 at 20 dpi than in samples from G- NcSpain7 at 10 dpi (P < 0.0001; Mann-Whitney U test). Slightly higher parasite burdens were found in CO than in CA samples from G-NcSpain7 culled at 20 dpi, although the differences were not statistically significant (P > 0.05; Mann-Whitney U test) (Figure 3c). The parasite burden was higher in CA (P ≤ 0.001; Mann-Whitney U test) and CO (P < 0.0001; Mann-Whitney U test) from animals with NVF (3581 and 7934) than those carrying VF (7992, 4405 and 5082) of G-NcSpain7 at 20 dpi (Figure 3d). In contrast, differences in the parasite burden in CA and CO between animals from G-NcSpain1H culled at 10 and 20 dpi were not found (P > 0.5; Mann- Whitney U test). Comparing animals from infected groups culled at 20 dpi, higher parasite burdens in CA and CO were detected in G-NcSpain7 than in G- NcSpain1H (P <0.0001; Mann-Whitney U test) (Figure 3a). Foetal tissues Regarding foetal tissues, 12 out of 15 FB samples from G-NcSpain7 foetuses at 20 dpi were positive by PCR, whereas all FB samples from G-NcSpain7 foetuses at 10 dpi were negative. All FBs from G- NcSpain1H foetuses culled at 10 or 20 dpi were negative. On the other hand, 2 G-NcSpain7 foetuses at 10 dpi (3/12) and one G-NcSpain7 foetus at 20 dpi (3/15) presented N. caninum positive FL samples, although differences between culling at 10 or 20 dpi were not found (P = 1; Fisher exact test). FL from all G-NcSpain1H foetuses were negative for N. caninum DNA detection. FB and FL samples from G-Control were negative. Higher parasite burdens were found in FB samples from G-NcSpain7 culled at 20 dpi than at 10 dpi (P <0.0001; Mann-Whitney U test), but differences were not found in FL (P > 0.5; Mann-Whitney U test) Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 116 (Figure 3b). In addition, a higher parasite burden was found in FB than in FL in G-NcSpain7 at 20 dpi (Figure 3c). Comparing animals carrying VF and NVF from G-NcSpain7 at 20 dpi, higher parasite burdens were found in the FL of NVF (P < 0.05; Mann- Whitney U test), whereas no differences were found in the FB (P > 0.5; Mann-Whitney U test) (Figure 3d). 3.4 IFN-γ kinetics in sera A peak of IFN-γ production was detected at 2 dpi in both infected groups with respect to the control group G-Control (P < 0.0001 and P = 0.0013 in G- NcSpain7 and G-NcSpain1H, respectively; two-way ANOVA test). All animals from G-NcSpain7 and G- NcSpain1H presented increased levels of IFN-γ at 2 dpi. Differences between infected groups were also found, with the increase of IFN-γ being higher in Nc- Spain7-infected animals than in Nc-Spain1H- infected animals (P < 0.0002; two-way ANOVA test) (Figure 4). 3.5 Specific anti-Neospora IgG responses in heifers and foetuses N. caninum-specific antibody responses (total IgG, IgG1 and IgG2) are shown in Figure 5. An earlier detection of N. caninum antibodies was observed in G-NcSpain7 (9 dpi) (7934) than in G-NcSpain1H (13 dpi). All animals from G-NcSpain7 seroconverted from 13 dpi, while only 3 out of 5 animals from G- NcSpain1H seroconverted between 13 (9677) and 16 dpi (7649 and 9638). Figure 3. N. caninum burdens in placental and foetal tissues. Comparative of parasite burdens quantified by qPCR in nested-PCR positive samples from CA and CO (a) and FB and FL (b) from G-NcSpain7 and G-NcSpain1H heifers culled at 10 and 20 dpi. (c) Comparative of parasite quantification by qPCR between CA and CO and between FB and FL from G- NcSpain7 heifers culled at 20 dpi. (d) Comparative of parasite quantification by qPCR in samples from CA, CO, FB and FL from NVF and VF foetuses from G-NcSpain7 culled 20 dpi. Each dot represents individual values of parasite burden, and medians are represented as horizontal lines. The N. caninum detection limit by real-time PCR was 0.1 parasites, and negative samples (0 parasites) were represented on the log scale as < 0.1 (i.e., 10−2). ****, ***, and * indicate P < 0.0001, P < 0.001 and P < 0.05 significant differences. σ indicates P < 0.1 tendency towards significant differences. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 117 Figure 4. IFN-γ kinetics in sera. Sera concentration of IFN-γ determined by ELISA in G-Control, G-NcSpain7 and G- NcSpain1H. **** and *** indicate P < 0.0001 and P < 0.001 significant differences. Total IgG levels were significantly higher from 13 dpi until the end of the experiment in G-NcSpain7 compared to G-Control (P < 0.05; two-way ANOVA test). No significant increase in the antibody levels was found in G-NcSpain1H during the experimental period compared to the control group (P > 0.05; two-way ANOVA test). Total IgG levels of G- NcSpain7 were higher than G-NcSpain1H at 16 dpi (P < 0.05; two-way ANOVA test) and at 20 dpi (P < 0.0001; two-way ANOVA test). No significant differences were found in G-NcSpain7 at 20 dpi between animals carrying NVF (3581 and 7934) and VF (7992, 4405 and 5082) (Figure 5a). IgG1 and IgG2 kinetics were similar to those observed in total IgG levels. Higher levels of IgG1 and IgG2 were found in G-NcSpain7 than in G- Control at 16 and 20 dpi (P < 0.01; two-way ANOVA test). There were no significant differences in IgG1 and IgG2 levels between G-NcSpain1H and G- Control. Comparing both infected groups, higher IgG1 and IgG2 levels were found in G-NcSpain7 than in G-NcSpain1H at 16 and 20 dpi (P < 0.01; two-way ANOVA test), and higher IgG1 levels were also found in G-NcSpain7 than in G-NcSpain1H at 13 dpi (P < 0.01; two-way ANOVA test). No significant differences in IgG1 and IgG2 were found between animals carrying NVF (3581 and 7934) and VF (7992, 4405 and 5082) in G-NcSpain7 at 20 dpi (Figures 5b and 5c). Neospora-specific IgG was not detected in foetal serum or foetal fluids by IFAT and WB. 4. Discussion Abortion due to N. caninum infection appears more frequently at mid-gestation in naturally infected cattle. However, few studies of pregnant bovine models of neosporosis at this gestation time have been reported, especially those investigating early infection dynamics. In the present work comparisons between high (Nc-Spain7)- and low (Nc-Spain1H)-virulence isolates of N. caninum inoculated at mid-gestation were approached at early time points post-infection (10 and 20 dpi). The results from this experimental model will increase the knowledge about biological differences found between high- and low-virulence isolates in vivo, clarifying some of the key events involved in the pathogenesis of bovine neosporosis. Half of the animals from both infected groups showed fever as the first clinical sign associated with N. caninum infection, which is in agreement with previous reports where a transient rise in body temperature was recorded during the first week post-infection, likely the consequence of the first cycle of parasite replication in host tissues (Buxton et al., 1998; Williams et al., 2000; Macaldowie et al., 2004; Caspe et al., 2012; Regidor-Cerrillo et al., 2014; Almería et al., 2016; Vázquez et al., submitted). A second peak of fever was detected at 3 dpi only in Nc-Spain7-infected animals, which suggests an earlier and higher replication of this isolate, leading to a second antigenic exposition of Nc-Spain7 tachyzoites. Similarly, high doses of the Nc-1 isolate were associated with a bi-phasic Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 118 Figure 5. N. caninum-specific humoural immune responses. Serum levels of total IgG (a), IgG1 (b) and IgG2 (c) antibodies against N. caninum in G-Control, G-NcSpain7 and G-NcSpain1H. Immunoglobulin levels are expressed as a relative index percent (RIPC) according to RIPC = (OD405 sample - OD405 negative control)/(OD405 positive control - OD405 negative control)×100. Each point represents the mean + SD at different sampling times. The red line indicates the ELISA cut-off point from which samples are considered positive. **** and ** indicate P < 0.0001 and P < 0.001 significant differences. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 119 Table 1. Summary of early infection dynamics in heifers and foetuses from G-Control, G-NcSpain7 and G-NcSpain1H increase in rectal temperature (Maley et al., 2003). Moreover, all G-NcSpain7-heifers were febrile, whereas only half of G-NcSpain1H animals showed fever. Differences between isolates may be explained by a more efficient replication of Nc- Spain7, as was previously demonstrated in vitro (Regidor-Cerrillo et al., 2011; Jiménez-Pelayo et al., 2017; García-Sánchez et al., 2019) and in vivo (Regidor-Cerrillo et al., 2010). Peripheral immune responses, both cellular and humoural, were assessed in dams’ sera during the experiment. IFN-γ was detected in both infected groups at 2 dpi, demonstrating that N. caninum tachyzoites activated the cellular immune response, which is crucial for host defence against intracellular pathogens (Innes et al., 2002; Innes et al., 2005; Williams et al., 2007). Higher IFN-γ levels in G-Nc-Spain7 did not correlate with protection against infection, since the Nc-Spain7 isolate was detected earlier in the placenta and showed a more effective invasion and proliferation than the Nc- Spain1H isolate. Specific antibodies against N. caninum were detected in all Nc-Spain7 heifers slightly earlier than in previous works (Bartley et al., 2004; Almería et al., 2010; Almería et al., 2016), whereas later seroconversion, and in fewer animals, was found in three Nc-Spain1H-infected heifers. Previous reports suggested that Nc-Spain7 may be able to induce a higher antibody response, whereas antigenic stimulation seems to be more reduced in Nc-Spain1H, probably due to its lower ability to multiply in host tissues, its clearance by the immune system or both (Buxton, et al., 1991; Rojo-Montejo et al., 2009b; Regidor-Cerrillo et al., 2014). In this study, culling as early as 10 and 20 dpi was effective in showing clear differences between isolates of variable virulence. At 10 dpi, few placental samples from one Nc-Spain7-infected animal were positive for N. caninum, and one of Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 120 them showed focal necrosis, demonstrating the colonization of the placenta by this isolate. Early detection of Nc-Spain7 at 10 dpi may be associated with its higher abilities for invasion and proliferation in placental cells. To the best of our knowledge, none of the previous experiments studied the dynamics of the infection as early as 10 dpi, however similar to our observations, focal necrosis was described in placentomes at 14 dpi with the Nc-1 isolate (Maley et al., 2003) and at 2 wpi with the Nc-Spain7 isolate (Almería et al., 2016). In addition to focal necrosis, differences in the plasma extravasation were found between infected and control animals. The extravasation of erythrocytes and plasma into the haemophagus zone of the placentome is a normal finding in healthy animals (Murai & Yamauchi, 1986). However, in this study, higher extravasation was found in infected animals at 10 and 20 dpi. Previous studies have described serum leakage in relation to necrotic and inflammatory foci in the interdigitate area of the placenta (Maley et al., 2003; Almería et al., 2010), but the increase in the proteinaceous material in the haemophagus area has not been previously reported. This increase may appear as one of the initial changes in the placenta associated with N. caninum infection, and it is tempting to hypothesize that it might be related to changes in vascular permeability. It has been recently shown that complement and coagulation cascades were modified after N. caninum infection in trophoblast cells in vitro (Horcajo et al., 2017); therefore, studying early vascular events in the placenta after N. caninum infection could be an interesting future research. N. caninum was also detected in the liver of two Nc-Spain7-infected foetuses, indicating that the high-virulence isolate is already transmitted to the foetus at 10 dpi. Previous studies have suggested the crossing and presence of tachyzoites in foetal tissues as early as 10 dpi (Barr et al., 1994; Maley et al., 2003). Our findings suggest that the liver is the first target organ in the foetus, which most likely represents the gateway for the parasite to invade the foetus through the umbilical vein, replicating in the parenchyma and spreading through the foetal body, as previously observed in sheep (Arranz-Solís et al., 2015). However, antibodies against N. caninum were not found in serum or corporal fluids from any foetus, probably because antibodies cannot be produced in infected foetuses until at least 6 weeks after infection (Caspe et al., 2012; Almería et al., 2016). In contrast, Nc-Spain1H was not found in any placental or foetal tissue, and specific humoural responses were not found either in Nc-Spain1H- infected foetuses. Lesions were not found either in placentas or in foetuses from G-NcSpain1H, apart from serum extravasation in placentomes. However, IV inoculation is supposed to disseminate the parasite quickly through the organism. We hypothesize that the presence of Nc-Spain1H in the placenta at 10 dpi is very low, since we were unable to detect the parasite, which is associated with lower invasion and replication abilities, as previously observed in placental tissues in vitro, especially in caruncular cells (Jiménez-Pelayo et al., 2017). In addition, a higher stimulation of the innate immune responses by the low-virulence isolate at the placental level was suggested at early time points post-infection in vitro (Jiménez-Pelayo et al., 2019), which could explain the more effective control of the parasite, thereby contributing to its lower proliferation. At 20 dpi, foetal death was detected in two Nc- Spain7-infected heifers, whereas it was not detected in any Nc-Spain1H-infected heifer. All G- NcSpain7 animals presented almost 100% positive placental samples, which is in keeping with previous studies where Nc-Spain7 or other isolates showed dissemination in placental tissues early after infection (2-4 wpi) when inoculation was carried out at mid-gestation (110-140 dg) (Maley et al., 2003; Almería et al., 2010; Rojo-Montejo et al., 2013; Almería et al., 2016). Our results again demonstrated the “tropism” of N. caninum for the bovine placental tissue, which seems to be one of the most appropriate niches for its multiplication. Placental necrosis was observed in G-NcSpain7 animals at 20 dpi associated with high parasite burdens. The extravasation of proteinaceous material and cellular debris in the haemophagus area in G-NcSpain7 was larger at 20 dpi than at 10 dpi and larger than in G-NcSpain1H animals, suggesting a correlation between the presence and severity of this histological change and infection by N. caninum in the present work, since larger areas of extravasation were observed in animals with higher parasite burden and infected with more virulent isolates. Parasite was found in most FB samples from G-NcSpain7, and inflammatory infiltrate and lesions compatible with N. caninum were found especially in foetal CNS, which is in keeping with previous observations where the brain was defined as a target tissue for N. caninum (Collantes-Fernández et al., 2006c; Almería et al., 2010Almería et al., 2016) but also in lung, skeletal muscle, heart and liver. Inflammatory infiltrate in foetal organs supports the hypothesis that at least partial foetal immunocompetence is already developed at this time, although no specific antibodies were found in foetal sera or in foetal fluids, as explained above. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 121 In contrast to those results observed after infection with the isolate of high-virulence Nc-Spain7, the infection with the low-virulence isolate Nc-Spain1H did not induce foetal death, and only one G- NcSpain1H animal presented positive placental samples at 20 dpi, similar to a previous experimental study at early gestation (Rojo- Montejo et al., 2009b). In addition, there were no evident lesions at the placenta. Taken together, these results suggest a limited colonization of maternal placenta by Nc-Spain1H, which is consistent with the low proliferation rate of this isolate under in vitro conditions (Regidor-Cerrillo et al., 2011; Jiménez-Pelayo et al., 2017). In addition, four CO samples were positive, suggesting that as demonstrated in vitro in F3 cells, the foetal compartment of the placenta may be the target cell and the preferential niche for parasite multiplication, whereas caruncular cells seem to play a barrier role for the placenta, limiting the invasion and multiplication of the parasite (Jiménez-Pelayo et al., 2017). Moreover, higher activation of the innate immune responses, specifically TLR-2, on the maternal side as observed in vitro (Jiménez-Pelayo et al., 2019), may contribute to the elimination of the tachyzoites, diminishing the burden in the caruncle and limiting tissue damage. Despite the absence of parasite DNA, lesions or foetal antibodies in G-NcSpain1H foetuses at 20 dpi, the identification of parasite DNA on CO indicated the transmission of this isolate to the foetal compartment. In fact, the origin of this isolate (from a dairy herd with high intra-herd N. caninum seroprevalence) (Rojo-Montejo et al., 2009a) and a previous experimental infection at early gestation (Rojo-Montejo et al., 2009b) also corroborate that transmission of Nc-Spain1H to the foetus does occur. It is therefore tempting to hypothesize that if the experimental design of the study had allowed a longer gestation, the parasite might have been transmitted to the foetus. Related to the pathogenesis of abortion, in the present work, higher parasite burdens and more severe lesions were detected in placentomes from one animal carrying NVF (3581) compared to VF, demonstrating that replication of the parasite at the maternal-foetal interface may be an important factor of foetal mortality (Buxton et al., 2002; Innes et al., 2007). On the other hand, resolution of placental lesions was demonstrated at 42 dpi (Maley et al., 2003), which indicated that progression of infection had been halted by the dam and the foetus and could be a reversible process in some cases. Moreover, our results showed that the %LES of the placenta was low, lesions showed a focal distribution and severity of the lesions did not seem sufficient to justify the foetal death by themselves because hypoxia signs were not found in NVF and placental functions did not seem to be compromised. There were no differences in the parasite burden in FB between VF and NVF, and only slightly higher parasite burden in FL was found in NVF. In addition, similar lesions were found in the foetal brain, liver, lung, heart and skeletal muscle of all Nc-Spain7 foetuses at 20 dpi. Brain lesions could be evaluated only in VF since NVF presented autolysis of the CNS. A key question that remains unsolved is the roles of the maternal and foetal immune responses in the outcome of the infection. In summary, wider parasite dissemination with earlier transmission to the foetus and foetal death were found after infection with the high-virulent isolate Nc-Spain7 as soon as 10 and 20 dpi, respectively. All these findings seem to be related to a better capacity of this isolate to invade the placenta earlier and proliferate more efficiently. The pathogenesis of the abortion could not be determined with our findings, since placental and foetal burdens and lesions in VF and NVF would not explain by themselves the foetal death. Therefore, the roles of the maternal and foetal immune responses in the outcome of the infection should be investigated. However, this experiment was not designed to elucidate the cause of the abortion, and closer monitoring of the foetus and sequential sampling and culling are warranted in further research. 5. References Almería, S., Araujo, R., Tuo, W., López-Gatius, F., Dubey, J.P., Gasbarre, L.C., 2010. Fetal death in cows experimentally infected with Neospora caninum at 110 days of gestation. Vet. Parasitol., 10.1016/j.vetpar.2009.12.044. Almería, S., López-Gatius, F., 2013. Bovine neosporosis: Clinical and practical aspects. Res. Vet. Sci. 95 (2), 303- 309, 10.1016/j.rvsc.2013.04.008. Almería, S., Serrano-Pérez, B., Darwich, L., Domingo, M., Mur-Novales, R., Regidor-Cerrillo, J., Cabezón, O., Pérez- Maillo, M., López-Helguera, I., Fernández-Aguilar, X., 2016a. Foetal death in naive heifers inoculated with Neospora caninum isolate Nc-Spain7 at 110 days of pregnancy. Exp. Parasitol. 168, 62-69. Álvarez-García, G., Collantes-Fernández, E., Costas, E., Rebordosa, X., Ortega-Mora, L.M., 2003. Influence of age and purpose for testing on the cut-off selection of serological methods in bovine neosporosis. Vet. Res. 34 (3), 341-352, 10.1051/vetres:2003009. Álvarez-García, G., Pereira-Bueno, J., Gómez-Bautista, M., Ortega-Mora, L.M., 2002. Pattern of recognition of Neospora caninum tachyzoite antigens by naturally infected pregnant cattle and aborted foetuses. Vet. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 122 Parasitol. 107 (1-2), 15-27. Anderson, M.L., Palmer, C.W., Thurmond, M.C., Picanso, J.P., Blanchard, P.C., Breitmeyer, R.E., Layton, A.W., McAllister, M., Daft, B., Kinde, H., 1995. Evaluation of abortions in cattle attributable to neosporosis in selected dairy herds in California. J. Am. Vet. Med. Assoc. 207 (9), 1206-1210. Arranz-Solís, D., Benavides, J., Regidor-Cerrillo, J., Fuertes, M., Ferre, I., Ferreras Mdel, C., Collantes- Fernández, E., Hemphill, A., Pérez, V., Ortega-Mora, L.M., 2015. Influence of the gestational stage on the clinical course, lesional development and parasite distribution in experimental ovine neosporosis. Vet. Res. 46, 19-014- 0139-y, 10.1186/s13567-014-0139-y. Barr, B.C., Rowe, J.D., Sverlow, K.W., BonDurant, R.H., Ardans, A.A., Oliver, M.N., Conrad, P.A., 1994. Experimental reproduction of bovine fetal Neospora infection and death with a bovine Neospora isolate. J. Vet. Diagn. Invest. 6 (2), 207-215. Bartley, P.M., Kirvar, E., Wright, S., Swales, C., Esteban- Redondo, I., Buxton, D., Maley, S.W., Schock, A., Rae, A.G., Hamilton, C., Innes, E.A., 2004. Maternal and fetal immune responses of cattle inoculated with Neospora caninum at mid-gestation. J. Comp. Pathol. 130 (2-3), 81- 91. Benavides, J., Collantes-Fernández, E., Ferre, I., Pérez, V., Campero, C., Mota, R., Innes, E., Ortega-Mora, L.M., 2014. Experimental ruminant models for bovine neosporosis: what is known and what is needed. Parasitology 141 (11), 1471-1488, 10.1017/S0031182014000638. Benavides, J., Katzer, F., Maley, S.W., Bartley, P.M., Cantón, G., Palarea-Albaladejo, J., Purslow, C.A., Pang, Y., Rocchi, M.S., Chianini, F., Buxton, D., Innes, E.A., 2012. High rate of transplacental infection and transmission of Neospora caninum following experimental challenge of cattle at day 210 of gestation. Vet. Res. 43 (1), 83, 10.1186/1297-9716-43-83. Bland, J.M., Altman, D.G., 1998. Survival probabilities (the Kaplan-Meier method). BMJ 317 (7172), 1572. Buxton, D., Maley, S.W., Wright, S., Thomson, K.M., Rae, A.G., Innes, E.A., 1998. The pathogenesis of experimental neosporosis in pregnant sheep. J. Comp. Pathol. 118 (4), 267-279. Buxton, D., McAllister, M.M., Dubey, J.P., 2002. The comparative pathogenesis of neosporosis. Trends Parasitol. 18 (12), 546-552. Buxton, D., Thomson, K., Maley, S., Wright, S., Bos, H.J., 1991. Vaccination of sheep with a live incomplete strain (s48) of Toxoplasma gondii and their immunity to challenge when pregnant. Vet. Rec. 129 (5), 89-93. Caspe, S.G., Moore, D.P., Leunda, M.R., Cano, D.B., Lischinsky, L., Regidor-Cerrillo, J., Álvarez-García, G., Echaide, I.G., Bacigalupe, D., Ortega-Mora, L.M., Odeon, A.C., Campero, C.M., 2012. The Neospora caninum-Spain 7 isolate induces placental damage, fetal death and abortion in cattle when inoculated in early gestation. Vet. Parasitol. 189 (2-4), 171-181, 10.1016/j.vetpar.2012.04.034. Collantes-Fernández, E., Arnaiz-Seco, I., Burgos, B.M., Rodríguez-Bertos, A., Adúriz, G., Fernández-García, A., Ortega-Mora, L.M., 2006c. Comparison of Neospora caninum distribution, parasite loads and lesions between epidemic and endemic bovine abortion cases. Vet. Parasitol. 142 (1-2), 187-191. Collantes-Fernández, E., Rodríguez-Bertos, A., Arnaiz- Seco, I., Moreno, B., Adúriz, G., Ortega-Mora, L.M., 2006a. Influence of the stage of pregnancy on Neospora caninum distribution, parasite loads and lesions in aborted bovine foetuses. Theriogenology 65 (3), 629-641. Collantes-Fernández, E., Zaballos, A., Álvarez-García, G., Ortega-Mora, L.M., 2002. Quantitative detection of Neospora caninum in bovine aborted fetuses and experimentally infected mice by real-time PCR. J. Clin. Microbiol. 40 (4), 1194-1198. Dubey, J., Schares, G., 2011. Neosporosis in animals—the last five years. Vet. Parasitol. 180 (1), 90-108. Dubey, J.P., Schares, G., Ortega-Mora, L.M., 2007. Epidemiology and control of neosporosis and Neospora caninum. Clin. Microbiol. Rev. 20 (2), 323-367. García-Sánchez, M., Jiménez-Pelayo, L., Horcajo, P., Regidor-Cerrillo, J., Ólafsson, E.B., Bhandage, A.K., Barragán, A., Werling, D., Ortega-Mora, L.M., Collantes- Fernández, E., 2019. Differential responses of bovine monocyte-derived macrophages to infection by Neospora caninum isolates of high and low virulence. Front. Inmunol. 10, 915. Gibney, E.H., Kipar, A., Rosbottom, A., Guy, C.S., Smith, R.F., Hetzel, U., Trees, A.J., Williams, D.J., 2008. The extent of parasite-associated necrosis in the placenta and foetal tissues of cattle following Neospora caninum infection in early and late gestation correlates with foetal death. Int. J. Parasitol. 38 (5), 579-588, 10.1016/j.ijpara.2007.09.015. Horcajo, P., Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Collantes-Fernández, E., Rozas, D., Hambruch, N., Pfarrer, C., Ortega-Mora, L.M., 2017. Transcriptome modulation of bovine trophoblast cells in vitro by Neospora caninum. Int. J. Parasitol. 47 (12), 791- 799, S0020-7519(17)30252-7. Horcajo, P., Regidor‐Cerrillo, J., Aguado‐Martínez, A., Hemphill, A., Ortega‐Mora, L.M., 2016. Vaccines for bovine neosporosis: current status and key aspects for development. Parasite Immunol. 38 (12), 709-723. Hurtado, A., Adúriz, G., Moreno, B., Barandika, J., Garcıá- Pérez, A.L., 2001. Single tube nested PCR for the detection of Toxoplasma gondii in fetal tissues from naturally aborted ewes. Vet. Parasitol. 102 (1), 17-27. Innes, E.A., Andrianarivo, A.G., Björkman, C., Williams, D.J., Conrad, P.A., 2002. Immune responses to Neospora caninum and prospects for vaccination. Trends Parasitol. 18 (11), 497-504. Innes, E.A., Wright, S., Bartley, P., Maley, S., Macaldowie, C., Esteban-Redondo, I., Buxton, D., 2005. The host- Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 123 parasite relationship in bovine neosporosis. Vet. Immunol. Immunopathol. 108 (1-2), 29-36. Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Horcajo, P., Collantes-Fernández, E., Gómez-Bautista, M., Hambruch, N., Pfarrer, C., Ortega-Mora, L.M., 2017. Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum. Parasit. Vectors 10 (1), 463, 10.1186/s13071-017-2409-9. Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Horcajo, P., Collantes-Fernández, E., Gómez-Bautista, M., Hambruch, N., Pfarrer, C., Ortega-Mora, L., 2019. Immune response profile of caruncular and trophoblast cell lines infected by high- (Nc-Spain7) and low-virulence (Nc-Spain1H) isolates of Neospora caninum. Parasit Vectors 12 (1), 218, 10.1186/s13071-019-3466-z. López-Gatius, F., López-Bejar, M., Murugavel, K., Pabón, M., Ferrer, D., Almería, S., 2004. Neospora-associated abortion episode over a 1-year period in a dairy herd in north-east Spain. J. Vet. Med. B Infect. Dis. Vet. Public Health 51 (7), 348-352. Macaldowie, C., Maley, S.W., Wright, S., Bartley, P., Esteban-Redondo, I., Buxton, D., Innes, E.A., 2004. Placental pathology associated with fetal death in cattle inoculated with Neospora caninum by two different routes in early pregnancy. J. Comp. Pathol. 131 (2-3), 142- 156. Maley, S.W., Buxton, D., Rae, A.G., Wright, S.E., Schock, A., Bartley, P.M., Esteban-Redondo, I., Swales, C., Hamilton, C.M., Sales, J., Innes, E.A., 2003. The pathogenesis of neosporosis in pregnant cattle: inoculation at mid-gestation. J. Comp. Pathol. 129 (2-3), 186-195. Murai, T., Yamauchi, S., 1986. Erythrophagocytosis by the trophoblast in a bovine placentome. Jpn J Vet Sci 48 (1), 75-88. Pérez-Zaballos, F.J., Ortega-Mora, L.M., Álvarez-García, G., Collantes-Fernández, E., Navarro-Lozano, V., García- Villada, L., Costas, E., 2005. Adaptation of Neospora caninum isolates to cell-culture changes: an argument in favor of its clonal population structure. J. Parasitol. 91 (3), 507-510. Regidor-Cerrillo, J., Arranz-Solís, D., Benavides, J., Gómez- Bautista, M., Castro-Hermida, J.A., Mezo, M., Pérez, V., Ortega-Mora, L.M., González-Warleta, M., 2014. Neospora caninum infection during early pregnancy in cattle: how the isolate influences infection dynamics, clinical outcome and peripheral and local immune responses. Vet. Res. 45, 10, 10.1186/1297-9716-45-10. Regidor-Cerrillo, J., Gómez-Bautista, M., Del Pozo, I., Jiménez-Ruiz, E., Adúriz, G., Ortega-Mora, L.M., 2010. Influence of Neospora caninum intra-specific variability in the outcome of infection in a pregnant BALB/c mouse model. Vet. Res. 41 (4), 52, 10.1051/vetres/2010024. Regidor-Cerrillo, J., Gómez-Bautista, M., Sodupe, I., Adúriz, G., Álvarez-García, G., Del Pozo, I., Ortega-Mora, L.M., 2011. In vitro invasion efficiency and intracellular proliferation rate comprise virulence-related phenotypic traits of Neospora caninum. Vet. Res. 42 (1), 41, 10.1186/1297-9716-42-41. Rojo-Montejo, S., Collantes-Fernández, E., Blanco- Murcia, J., Rodríguez-Bertos, A., Risco-Castillo, V., Ortega- Mora, L.M., 2009b. Experimental infection with a low virulence isolate of Neospora caninum at 70 days gestation in cattle did not result in foetopathy. Vet. Res. 40 (5), 49, 10.1051/vetres/2009032. Rojo-Montejo, S., Collantes-Fernández, E., Pérez- Zaballos, F., Rodríguez-Marcos, S., Blanco-Murcia, J., Rodríguez-Bertos, A., Prenafeta, A., Ortega-Mora, L.M., 2013. Effect of vaccination of cattle with the low virulence Nc-Spain 1H isolate of Neospora caninum against a heterologous challenge in early and mid-gestation. Vet. Res. 44, 106, 10.1186/1297-9716-44-106. Rojo-Montejo, S., Collantes-Fernández, E., Regidor- Cerrillo, J., Álvarez-García, G., Marugán-Hernández, V., Pedraza-Díaz, S., Blanco-Murcia, J., Prenafeta, A., Ortega- Mora, L.M., 2009a. Isolation and characterization of a bovine isolate of Neospora caninum with low virulence. Vet. Parasitol. 159 (1), 7-16. Williams, D.J., Guy, C.S., McGarry, J.W., Guy, F., Tasker, L., Smith, R.F., MacEachern, K., Cripps, P.J., Kelly, D.F., Trees, A.J., 2000. Neospora caninum-associated abortion in cattle: the time of experimentally-induced parasitaemia during gestation determines foetal survival. Parasitology 121 (4), 347-358. Williams, D.J., Guy, C.S., Smith, R.F., Ellis, J., Björkman, C., Reichel, M.P., Trees, A.J., 2007. Immunization of cattle with live tachyzoites of Neospora caninum confers protection against fetal death. Infect. Immun. 75 (3), 1343-1348. Williams, D.J., Hartley, C.S., Björkman, C., Trees, A.J., 2009. Endogenous and exogenous transplacental transmission of Neospora caninum - how the route of transmission impacts on epidemiology and control of disease. Parasitology 136 (14), 1895-1900, 10.1017/S0031182009990588. Declarations Ethics approval and consent to participate All protocols involving animals were approved by the Animal Research Ethics Committee of the Principado de Asturias, Spain (reference number PROAE 25/2016), following the proceedings described in Spanish and EU legislations (Law 32/2007, R.D. 53/2013, and Council Directive 2010/63/EU). All animals were handled in strict accordance with good clinical practices and all efforts were made to minimize suffering. Consent for publication Not applicable. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 124 Availability of data and materials The datasets supporting the conclusions of this article are included within the article and its additional files. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the Spanish Ministry of Economy and Competitiveness (AGL2013-44694-R) and the Community of Madrid (PLATESA2-CM P2018/BAA-4370). Laura Jiménez-Pelayo was financially supported by a fellowship from the Complutense University of Madrid and Marta García-Sánchez was financially supported through a grant from the Spanish Ministry of Economy and Competitiveness (BES-2014-070723). Patricia Vázquez had a Juan de la Cierva-Formación post- doctoral contract (FJCI-2014-20982) from the Spanish Ministry of Economy and Competitiveness (MINECO). Alicia Román-Trufero was supported by a FPI-INIA fellowship from the Spanish National Institute for Agriculture and Food Research and Technology (INIA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Authors’ contributions JRC, PH, ECF, JB, KO and LMO conceived the study and participated in its design; LMO coordinated the study with the help of KO and obtained its funding. LJP and JB wrote the manuscript, with results interpretation and discussion inputs from PH, JRC, ECF, KO and LMO; JBM and ART performed animal reproductive and health management prior to infection. PH prepared the inoculum. LJP, MGS, PV and ART monitored the animals during the experimental period; LJP and MGS processed the samples during the experiment and after culling but histological processing and examination that were carried out by DGE and JB. LJP, MGS, PV, JB, DGE, ART and KO participated in the collection of samples during necropsy. LJP, MGS, JB and DGE carried out statistical analysis and interpreted the results. All authors read and approved the final manuscript. Acknowledgements Authors gratefully acknowledge to: 1) SERIDA (Regional Service of Agri-food Research and Development of Asturias) Institution and Personal for their facilities and personal support. Special thanks to David Iglesias for their clinical assistance; 2) Mountain Livestock Institute (IGM), University of León CSIC-ULE for their histopathological specialist support, especially to Miguel Fernández for his help during the sampling; 3) Saluvet Group members, especially to Alejandro Jiménez-Meléndez and Roberto Sánchez-Sánchez; 4) Saluvet-innova members, especially to Paula García-Lunar. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 125 Additional File 1: Materials and Methods. Description of the health and reproductive handling of the cattle and the tissue DNA extraction and PCR determinations. Health and reproductive handling of the cattle All animals were vaccinated against IBR virus and BVD virus with two doses of BOVILIS® BOVIPAST RSP (MSD Animal Health, Milton Keynes, UK), separated for 4 weeks, and treated with antiparasitic products (Endoex® and Albendex®, s.p. veterinaria, s.a., Riudoms, Spain; Animec Plus®, Cenavisa laboratorios, Tarragona, Spain) following the manufacturer’s recommendations. Heifers were oestrus synchronised with the administration of: i) CIDR (Zoetis, New Jersey, USA) + 100 µg of synthetic gonadorelin analogue (Cystoreline®, CEVA, Barcelona, Spain) (day 0); ii) CIDR removing + 400 UI of equine serum gonadotropin (Folligon®, MSD Animal Health, Milton Keynes, UK) + 25 mg synthetic prostaglandin F2α analogue (Dinolytic®, Zoetis, New Jersey, USA) (day 5); iii) 25 mg synthetic prostaglandin F2α analogue (Dinolytic®, Zoetis, New Jersey, USA) (day 6); iv) 100 µg of synthetic gonadorelin analogue (Cystoreline®, CEVA, Barcelona, Spain) (day 7). Fifty-six hours after the first administration of PGF2α, two artificial insemination, 12 h apart, were carried out using semen from two Asturiana bulls seronegative to N. caninum. Pregnancy was confirmed by ultrasound scanning on day 35 after insemination, and twenty-four pregnant animals were selected for the experiment. Tissue DNA extraction and PCR determinations DNA extraction and PCR determinations were carried out as described elsewhere (Regidor-Cerrillo et al., 2014; Arran-Solís et al., 2015). Briefly, genomic DNA was extracted from 20-100 mg of maternal and foetal tissue samples using the Maxwell® 16 Mouse Tail DNA Purification Kit (Promega, Wisconsin, USA), following the manufacturer´s recommendations. The DNA concentration was determined for each sample using a SynergyTM H1 microplate reader (Biotek Instruments Inc, Winooski, VT, USA), and the samples were adjusted to 100 ng/µl in molecular grade water. Parasite DNA detection was carried out by nested PCR adapted to a single tube from the internal transcribed spacer (ITS1) region of N. caninum using TgNN1- TgNN2 as external primers and NP1-NP-2 as internal primers (Buxton et al., 1998; Hurtado et al., 2001; Regidor-Cerrillo et al., 2014). Each reaction was performed in a final volume of 25 µl with 5 µl of sample DNA. PCR was carried out in 9 samples of CA and 9 samples of CO as well as 3 samples of maternal pre- scapular and ileofemoral lymph node and 3 samples of FB and FL. DNA samples from G-Control were included in each round of DNA extraction and PCR as negative controls. Positive PCR controls with N. caninum genomic DNA equivalent to 10, 1 and 0.1 tachyzoites in 100 ng of bovine DNA were also included in each batch of amplifications. Fifteen µl aliquots of the PCR products were visualized under UV light in a 1.5% agarose/GelRed™ (Biotium INC) gel to detect the N. caninum-specific 247 bp amplification product. Nested-PCR positive samples were adjusted to a concentration of DNA of 20 ng/µl, and quantification of N. caninum DNA was performed by real-time PCR using the equipment ABI 7500 FAST (Applied Biosystems, Foster City, CA, USA). The Nc-5 region was targeted as described elsewhere (Collantes-Fernández et al., 2005). A volume of 5 μL of diluted DNA from each sample was used for the qPCR assays. The number of N. caninum tachyzoites was determined by interpolating the Ct values (cycle threshold value) on a standard curve. The standard curve was designed for the quantification of 10-1 to 105 tachyzoites according to Regidor-Cerrillo et al. (2011). To normalize the quantification of the parasite in each sample, a bovine β- actin standard curve was designed (from 64 ng of DNA per μl to 0.2 ng per μl). The results were expressed as the relation between parasite DNA and cell DNA amount (R2 ≥ 0.99; Slope values varied from -3.63 to - 3.18). Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 126 Additional File 2. Histological findings in placental and foetal samples. (a) Placenta. G-NcSpain7 20 dpi. Three foci of necrotic placentitis with mild infiltration of inflammatory cells at the interdigitate area of the placentome HE. 4×. Bar 500 µm. (b) Placenta. G-NcSpain7 20 dpi. Focal necrosis with mild infiltration of inflammatory cells at the interdigitate area of the placentome. HE. 10×. Bar 200 µm. (c) Foetal heart. G-NcSpain7 20 dpi. Focal non-suppurative myocarditis. HE. 10×. Bar 200 µm. (d) Foetal brain. G-NcSpain7 20 dpi. Glia focus with a small area of necrosis at the centre of the lesions. HE. 10×. Bar 200 µm. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 127 Additional File 3. Quantification of necrosis foci (NF), size (ASF) and affected area (%LES) of these foci. Graphs representing median number of cells, lower and upper quartiles (boxes) and minimum and maximum values (whiskers) of (a) number of NF in G-NcSpain7 culled at 20 dpi studied by individual animal or by placentome location, (b) ASF in G-NcSpain7 culled at 20 dpi studied by individual animal or by placentome location and (c) %LES in G-NcSpain7 culled at 20 dpi studied by individual animal, by placentome location or by placentome location in the placentomes of animals with higher ASF (3581 and 5082). ****, ***, ** and * symbols indicate P < 0.0001, P < 0.001, P < 0.01 and P < 0.05 significant differences. Chapter ~ IV Results Sub-objective 2.1: Early infection dynamics of high- and low-virulence isolates in pregnant heifers at mid-gestation 128 Additional File 3. Continued. 129 Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. Jiménez-Pelayo L1, García-Sánchez M1, Collantes-Fernández E1, Regidor-Cerrillo J2, Horcajo P1, Gutiérrez-Expósito D4, Espinosa J4, Benavides J4, Osoro K4, Pfarrer C5, Ortega-Mora LM1*. 1SALUVET, Animal Health Department, Faculty of Veterinary Sciences, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain 2SALUVET-innova, Faculty of Veterinary Sciences, Complutense University of Madrid, Ciudad Universitariia s/n, 28040 Madrid, Spain 3Instituto de Ganadería de Montaña (CSIC-Universidad de León), 24346, León, Spain 4Regional Service for Research and Agri-Food Development (SERIDA), 33300 Villaviciosa, Asturias, Spain 5Department of Anatomy, University of Veterinary Medicine Hannover, Bischofsholer Damm 15, 30173 Hannover, Germany *Corresponding author: Luis Miguel Ortega-Mora Running title: Differential immune responses and extracellular matrix organization are observed in the bovine placenta during early infection at mid-gestation with high- and low-virulence Neospora caninum isolates. Manuscript in preparation This work has been accepted as a poster in the 5th International Meeting on Apicomplexa in Farm Animals (Apicowplexa) (2nd-4th October 2019, Berlin, Germany). http://hdl.handle.net/10261/205975 CHAPTER V GENERAL DISCUSSION DISCUSIÓN GENERAL Chapter V ~ General Discussion 161 Neospora caninum is considered to be a major cause of abortion in cattle worldwide, and bovine neosporosis has become a major concern for the scientific community in the last years (Dubey et al., 2017, Ortega-Mora et al., in press). Important economic losses have been related to this infection, mainly associated with abortion, lengthened calving interval, reduced stock value and increased culling rate (Trees et al., 1999; Reichel et al., 2013) Transplacental transmission is the main route of transmission in cattle, playing the bovine placenta a fundamental role in the pathogenesis of bovine neosporosis. Apart from its barrier function, the placenta can act as an immune regulatory organ. Trophoblasts and other cell types within the placenta may be involved in the physiological protection of the placenta, recognizing pathogens and leading to cytokine induction and regulation of co-estimulatory molecules (Montes et al., 1995; Steinborn et al., 1998a; Steinborn et al., 1998b). However, little is known about the interaction of the parasite with trophoblastic cells and other cell types in the placenta. Moreover, differences in the virulence of N. caninum isolates have been associated with variation in the clinical presentation of the disease. Nevertheless, the factors that enable some isolates to be more effectively transmitted and cause foetal death than others are still unclear. In this context, the initial events occurring after parasite infection are crucial. Recent transcriptomic and proteomic studies in vitro suggested a higher modulation of the host cell by the low-virulence isolate Nc-Spain1H and a different expression of genes involved in important metabolic, stress response and innate immune response processes (Regidor-Cerrillo et al., 2012; Horcajo et al., 2017; García-Sánchez et al., submitted). However, the mechanisms used by the parasite to modulate bovine host cells during the early phase of infection are not completely understood and the determining factor leading to foetal death is not already known. Four mechanisms have been postulated so far as possible cause of abortion: tachyzoites multiplication and the consequent damage in foetal organs; placental damage induced by the replication of the parasite; alteration of the immune balance at the placental level inducing placental damage; and prostaglandin secretion (Buxton et al., 2002; Innes et al., 2005; Dubey et al., 2006; Dubey et al., 2007; Innes, 2007). Here, the principal objective was to bring light to the role of the bovine placenta in pathogenesis of neosporosis and to identify potential pathways or biological networks at the maternal-foetal interphase that could explain why some isolates are more effectively transmitted and cause higher foetal death than others. In order to achieve this main objective, in vitro and in vivo investigations in placental target cells and in pregnant cattle have been carried out. Comparisons between high- and low-virulence isolates were carried out using two Spanish N. caninum isolates that previously demonstrated marked differences in virulence both in vitro and in mice and bovine models. First studies of this Doctoral Thesis were carried out in established bovine caruncular and trophoblast cell lines. In vitro assays offer a reduction in the usage of animals for experimentation and represent powerful tools for the study of apicomplexan parasites (Muller & Hemphill, 2012). These models have enabled studies on parasite virulence, host factors involved in innate resistance, parasite stage conversion and drug effectiveness, among others (Muller & Hemphill, 2012). The use of established cultures has been widely accepted in order to provide an initial approach to certain mechanisms involved in complex processes since they allow the obtaining of reproducible results and the reduction of experimental animals as well as the optimization of the methods that will be used later during in vivo experiments. This is the first study where a species- and organ-specific in vitro model for each of the host cell layers in the maternal-foetal interface Chapter V ~ General Discussion 162 of the bovine placenta has been implemented to study N. caninum infection in bovine placental cells. To date, several in vitro studies have been carried out using established cell lines, such as Marc-145, HeLa, BeWo or ovine trophoblast cells, to investigate the lytic cycle of different N. caninum isolates (Haldorson et al., 2005; Carvalho et al., 2010; Regidor-Cerrillo et al., 2011). There was only one limited, descriptive study concerning the interaction of the parasite with bovine trophoblast cells (Machado et al., 2007), and no data were available about the parasite interaction with bovine caruncular cells. Bovine trophoblast cells (F3) and bovine caruncular epithelial cells (BCEC-1) had been used previously for the study of drug transport across the maternal epithelium as part of the epitheliochorial placental barrier of the cow (Waterkotte et al., 2011), for the investigation of biological processes such as differentiation or invasion during placentation (Hambruch et al., 2010; Loch et al., 2018) or for the study of other abortifacient pathogen for cattle as Coxiella burnetii (Sobotta et al., 2017), showing a good reproducibility of the results. Results in this study have demonstrated that this model is very interesting for the study of N. caninum and a useful tool to evaluate critical factors involved in placental pathogenesis. In addition, BCEC-1 and F3 were isolated from fifth- and fourth-month pregnant heifers, respectively, and they have maintained at least part of their morphological and functional characteristics (Bridger et al., 2007b; Hambruch et al., 2010; Waterkotte et al., 2011). Thus, they are a good model for exploring N. caninum infection since transplacental transmission predominates in the second third of pregnancy, when the majority of abortions caused by N. caninum occur (Dubey et al., 2007). To develop all the experiments, two N. caninum isolates were selected based on their virulence in vitro and in vivo and maintained in cell culture. The number of passages was controlled in order to minimize the influence of culture adaptation as was demonstrated that continuous culture passages of the tachyzoites may lead to an adaptation of the parasite, modifying its in vivo biological behaviour and virulence (Pérez-Zaballos et al., 2005; Bartley et al., 2006). First, the high virulent isolate Nc-Spain7 was obtained from an asymptomatic calf (Regidor- Cerrillo et al., 2008) that presented 100% of brain samples positives to N. caninum DNA and more than 50% of the brain samples with characteristic lesions. Later studies in vitro showed mild rates of invasion and high rates of proliferation in MARC-145 cells (Regidor-Cerrillo et al., 2010). Those results correlated with a high virulence observed in the pregnant murine model, where high maternal mortality and neonatal morbidity as well as high rates of neonatal mortality and vertical transmission were found (Dellarupe et al., 2014a). Recent studies in bovine macrophages showed high proliferation and survival of Nc-Spain7 (García-Sánchez et al., 2019). In the present Doctoral Thesis, in vitro assays in bovine placental cells showed higher invasion and infection rates and higher proliferation (higher tachyzoite yield and lower doubling time) of Nc-Spain7 in bovine trophoblast cells. GRA17 expression was higher in Nc-Spain7 than Nc-Spain1H, and in T. gondii this protein has been related to faster growth (Gold et al., 2015; Horcajo et al., 2018). In addition, gliding associated proteins GAP40, GAP50, GAPM3, as well as five inner membrane complex (IMC) proteins, had higher expression in Nc-Spain7, and have been associated with a role in parasite motility and host cell invasion (Frenal et al., 2010), and probably in proliferation (Horcajo et al., 2018). However, Nc-Spain7 invasion, concretely adhesion, and proliferation were dramatically lower in caruncular cells than in trophoblast cells. Trophoblast cells are the first layer of the foetus, and our results suggested a role as a niche for parasite multiplication. On the other hand, Chapter V ~ General Discussion 163 the influence of the isolate on the cytokine profiles induced during the infection has been previously demonstrated (Rojo-Montejo et al., 2009b; Caspe et al., 2012; Regidor-Cerrillo et al., 2014). In a second work in vitro in bovine trophoblast and caruncular cells, the expression levels of molecules related to the immune responses and endothelial adhesion molecule genes were assessed after infection with both isolates at early and late stages post-infection. Several immune elements such as TLR-2, IL-6, IL-8, TNF-α and TGF-β1, showed to be expressed for trophoblast and caruncular cell cultures without any extra stimulus but the parasite, proving that placental cells are early responders to the infection and are implicated in defence against the parasite and participate in local immune responses. Those results were confirmed later through in vivo investigations, and recent in vitro experiments with human syncytiotrophoblast and cytotrophoblast infected by T. gondii reported comparable results (Ander et al., 2017). In addition, slight differences between isolates could be also detected, and observations made in vitro suggested an evasion strategy of Nc-Spain7 to avoid immune responses, with diminished expression of TLR-2 and TNF-α compared to Nc-Spain1H. The low-virulence isolate Nc-Spain1H was also isolated from a clinically healthy calf with precolostral N. caninum antibodies from a dairy herd with high N. caninum intraherd seroprevalence (Rojo-Montejo et al., 2009a). In vitro experiments suggested a low invasion and proliferation rate of this isolate (Rojo-Montejo et al., 2009a; Regidor-Cerrillo et al., 2010), later confirmed in the pregnant murine model. Presence of Nc-Spain1H in maternal tissues and maternal morbidity were low. In addition, only 5% of vertical transmission and 0.5% of neonatal mortality were observed (Rojo-Montejo et al., 2019; Pereira García-Melo et al., 2010). In bovine macrophages Nc-Spain1H also presented lower proliferation and survival (García-Sánchez et al., 2019). Here, the low virulence isolate showed lower invasion and proliferation in trophoblast cells than the Nc-Spain7 isolate. In fact, previous proteome studies revealed a patron similar to a pre- bradyzoite stage for Nc-Spain1H, associated with low proliferative isolates in T. gondii (Horcajo et al., 2018). In caruncular cells, as observed with the high-virulence isolate, Nc-Spain1H showed a remarkable reduced adhesion and proliferation, and differences between both isolates were minor in this cell line. Those results suggested a particularly remarkable feature of the bovine placenta with the presence of both permissive and resistant cells in very close proximity, indicating that cell autonomous features determine the fate of N. caninum upon interaction with trophoblast and caruncular cells. In vitro infection of trophoblast and caruncular cells showed that trophoblasts are the niche for parasite proliferation whereas caruncular cells seem to restrict the infection. Moreover, our in vitro investigations suggested that caruncular cells limited N. caninum infection at two critical stages of the parasite lytic cycle—at the point of parasite adhesion to the host cell and during intracellular growth. Parallel results in human second- trimester syncytiotrophoblasts and cytotrophoblasts infected by T. gondii showed reduced adhesion and replication of T. gondii in syncytiotrophoblasts, as the primary placental barrier formed by cells in direct contact with maternal blood. Nevertheless, high proliferation of the parasite was observed in cytotrophoblasts, as the subjacent migratory layer, supporting our result (Ander et al., 2017). Finally, in the second work, Nc-Spain1H showed higher stimulation of TLR-2 and TNF-α than Nc-Spain7, suggesting a better control of its multiplication by the immune system which may be an advantage to be transmitted to the progeny without causing foetal death and abortion. Chapter V ~ General Discussion 164 Despite of the importance and great utility of in vitro models, they do not consider information on environment influence. Hence, in vivo models are essential for the study of several aspects that otherwise could not be assessed by means of in vitro culture, being helpful for studies related to host-parasite relationship, immune response or pathogenesis (Reichel & Ellis, 2009). Results obtained in vitro were consolidated and complemented by studying the early consequences of infection (10 and 20 dpi) in a pregnant bovine model inoculated with high- (Nc-Spain7) and low- virulence (Nc-Spain1H) isolates of N. caninum at mid-gestation. Although abortion caused by N. caninum is more frequent during the second trimester of pregnancy in naturally infected cattle (Dubey et al., 2007) only few pregnant bovine models of neosporosis at this gestation time have been reported (Maley et al., 2003; Bartley et al., 2004; Almería et al., 2010; Almería et al., 2016a). On the other hand, the consequences of either primoinfection or recrudescence in a pregnant cow can be abortion, birth of a weak calf or birth of a clinically healthy but persistently infected calf. Many factors are implicated in the outcome of the infection, related to the host (gestation period, maternal and foetal immune responses, origin of the infection and breed) and the parasite (isolates and intraspecific variability, parasite stage, inoculation route and dose of infection). Parasite biological variability is one of the main factors influencing the outcome of the infection. Specifically, the high-virulence isolate Nc-Spain7 showed a percentage of abortion and vertical transmission of 100% in a bovine model at early gestation (Caspe et al., 2012; Regidor-Cerrillo et al., 2014), whereas the infection in experimentally infected cattle with the low-virulence isolate Nc-Spain1H did not result in foetal death (Rojo-Montejo et al., 2009b). Later studies carried out at mid-gestation showed that intravenous inoculation of 107 tachyzoites of Nc-Spain7 produced 50% foetal death until 42 dpi (Almería et al., 2016a) and 66.6% foetal death when gestation lasted until term. A recent work focused on the standardization of a pregnant bovine model of neosporosis during the second third of gestation (Vázquez et al., submitted), concluding that intravenous inoculation of 107 tachyzoites from the high virulent Nc-Spain7 isolate could be an adequate route and dose combination with more than 50% of abortions and efficient parasite transplacental transmission (100%) (Vázquez et al., submitted). However, early events occurred after the arriving of the parasite to the placenta and comparison between different virulence isolates inoculated at mid-gestation had not been investigated until now. Serial culling at 10 and 20 dpi of pregnant cattle infected by high- and low-virulence isolates was carried out in order to investigate the early infection dynamics as well as early immune responses and ECM modulation induced locally at the placenta in vivo. Results from the present Doctoral Thesis confirmed differences in virulence between isolates during infection in heifers at mid-gestation and suggested differences in the proliferation and modulation of the placental immune responses and the ECM by the high virulent isolate Nc-Spain7 and the low virulent isolate Nc-Spain1H. The first clinical sign detected in Nc-Spain7 animals was an increment in body temperature at 1 dpi in half of the animals, in agreement to previous reports (Buxton et al., 1998; Williams et al., 2000; Macaldowie et al., 2004; Caspe et al., 2012; Regidor-Cerrillo et al., 2014; Almería et al 2016a; Vázquez et al., submitted). In addition, only Nc-Spain7 infection induced a second peak of fever at 3 dpi, indicating earlier and higher replication of this isolate, leading to a second antigenic exposition of Nc-Spain7 tachyzoites. Moreover, fever was detected in all animals infected with the high virulent isolate along the experimental time. Higher febrile responses induced by the high-virulence isolate may be explained by a more efficient multiplication of Nc-Spain7 in host Chapter V ~ General Discussion 165 tissues as indicated previously (Regidor-Cerrillo et al., 2010; Regidor-Cerrillo et al., 2011; García- Sánchez et al., 2019), and confirmed in placental cells in vitro as a part of the investigations of the present Doctoral Thesis. Peripheral immune responses, cellular and humoral, were assessed in dams´ sera along the experimental infection, detecting a peak of IFN-γ in both infected groups at 2 dpi. Therefore, infection with N. caninum tachyzoites activated the cellular immune response, crucial for host defence to intracellular pathogens (Innes et al., 2002; Innes et al., 2005; Williams et al., 2007). However, despite Nc-Spain7-infected animals showed the highest IFN-γ levels in maternal sera, they did not correlate with protection against infection since Nc-Spain7 was detected earlier in the placenta and showed a more effective invasion and proliferation than the Nc-Spain1H isolate. On the other hand, specific antibodies against N. caninum were detected in all Nc-Spain7 heifers between 9 and 13 dpi, slightly earlier than in previous works (Maley et al., 2004; Almería et al., 2010; Almería et al., 2016a), and earlier than in Nc-Spain1H, probably associated again with the higher proliferation of this isolate. Nc-Spain7 arrived at the placenta at early stages of infection (10 dpi), with 5 placental samples positives to Neospora DNA by PCR, and one of them showed focal necrosis, demonstrating the colonization of the placenta by this isolate. Similar to our observations, focal necrosis was described in placentomes at 2 wpi with the Nc-Spain7 isolate (Almería et al., 2016a). Early detection of Nc-Spain7 at 10 dpi may be associated with its higher invasion and proliferation ability in placental cells, as demonstrated in vitro. However, immunohistochemical labelling of Neospora antigen in placentomes revealed absence of detection of parasite antigen, probably due to the limited sensibility of IHC analyses and reduced burdens reached by the parasite at this time. N. caninum was detected in the liver of two Nc-Spain7-infected foetuses, indicating that the high- virulence isolate is already transmitted to the foetus at 10 dpi, although lesions in foetal tissues were not found at this time. Previous studies have suggested the crossing and presence of tachyzoites in foetal tissues as early as 10 dpi (Barr et al., 1994; Maley et al., 2003). In addition, our findings may suggest that liver is the first target organ in the foetus when arriving from the placenta, which most probably represents the gateway for the parasite to invade the foetus through the umbilical vein, replicating in the parenchyma and after, spreading through the foetal body as was previously observed in sheep (Arranz-Solís et al., 2016). However, antibodies against N. caninum were not found in serum or corporal fluids from any foetus, probably because at least 6 weeks after infection are needed to detect antibodies in infected foetuses (Caspe et al., 2012; Almería et al., 2016a). Immune response profiles in the bovine placenta were also investigated during this experimental infection at 10 dpi. The highly virulent isolate Nc-Spain7 showed apparently the capacity to evade the immune responses in the placenta. Only IFN-γ, TNF-α and IL-4 were moderately upregulated whereas expression levels of PRRs, chemokines and endothelial adhesion molecules, the rest of cytokines and ECM components were not modified. In addition, in vitro results from this Doctoral Thesis demonstrated a down-regulation of TLR-2 in bovine trophoblast and caruncular cells infected with Nc-Spain7 at early time-points post-infection respect to Nc-Spain1H-infected cells Chapter V ~ General Discussion 166 as was demonstrated in vivo. Host cell modulation leading to evasion of local immune responses probably contributed to the more effective proliferation of Nc-Spain7 observed in placental tissues both in vitro and in vivo. In a similar way, Nc-Spain7 showed a higher ability to evade the host macrophage responses which, in conjunction with its higher replication and transmigration capacity, was associated with the higher virulence in vivo (García-Sánchez et al., 2019). Despite the lack of activation of classical PRRs involved in N. caninum detection at the placental level, increment in IFN-γ, TNF-α and IL-4 as well as T CD4+, CD8+ and B lymphocytes at 10 dpi suggest that other pathways of the immune system should be activated by the high virulent isolate. Other effector molecules of the innate immune response, as NLR proteins (NLRP12, NAIP, NLRC4), have been recently involved in the activation of the inflammatory responses in bovine macrophages after Nc-Spain7 and Nc-Spain1H infection in vitro (García-Sánchez et al., submitted). A higher concentration of Nc-Spain7 antigens or host cell components released during cell damage may be necessary to activate classical PRRs in the placenta during Nc-Spain7 infection. Further studies are necessary to identify new activation routes of the innate immune responses. Contrary to in vivo observations, IL-4 and IFN-γ were not expressed by bovine placental cells in vitro, showing that probably only T lymphocytes and other immune cells are implicated in their secretion after infection. Interestingly, CD8+ cells prevailed over CD4+ cells at 10 dpi both in infected and control groups as was previously observed in sheep (Arranz-Solís et al., 2016). Previous studies in cattle and buffaloes found the opposite (Maley et al., 2006; Rosbottom et al., 2011; Cantón et al., 2013; Cantón et al., 2014c), and CD4+ cells have been shown to be more relevant in the protection against neosporosis (Tanaka et al., 2000a). However, our results may be related with an early response to N. caninum infection since it was suggested that CD8+ cells could predominate over CD4+ as early responders to N. caninum (Correia et al., 2013). The increases in the immune cell populations were mainly associated to the maternal compartment of the placentome, a common finding given that N. caninum reaches the maternal caruncular septa via the circulation before infecting the foetal trophoblast villi cells and spreading to the foetus (Dubey et al., 2006). Therefore, it seems reasonable to assume that parasite is firstly recognized in the maternal part of the placenta (Marin et al., 2017), stimulating quickly the maternal immune response, which attracts inflammatory cells to the infection focus trying to control parasite dissemination to the foetus. Culling at 20 dpi revealed foetal death in two Nc-Spain7-infected heifers in agreement with previous experimental infections (Almería et al., 2016a; Vázquez et al., submitted). Essentially 100% of caruncle and cotyledon samples were positivity to Neospora DNA, evidencing one more time the “tropism” of N. caninum for the bovine placental tissues, which seems to be one of the most appropriate niches for its multiplication. However, parasite was only observed in foetal villi of Nc-Spain7-infected animals at 20 dpi by IHC, as postulated in vitro that trophoblast cells are the real target of N. caninum. Placental necrosis was observed in all NcSpain7-infected animals at 20 dpi associated with high parasite burdens. In addition, the extravasation of proteinaceous material and cellular debris found in the haemophagus area of the placentomes was larger at 20 dpi than at 10 dpi in Nc- Spain7-infected heifers and larger than in NcSpain1H-infected animals, suggesting a correlation Chapter V ~ General Discussion 167 between the presence and severity of this histological change and infection by N. caninum, since larger areas of extravasation were observed in animals with higher parasite burden and infected with more virulent isolates. In fact, multiplication of the parasite in the endothelium might damage the placental vasculature, increasing vascular permeability and, thus, facilitating serum infiltration. Increased expression of SERP-1, as a component of the complement and coagulation cascades, was observed in infected placentomes from the in vivo model implemented in the present Doctoral Thesis, confirming complement and coagulation cascades modulation by the infection. In addition, those expression levels were higher at 20 dpi and higher in placentomes from non-viable foetuses than in viable foetuses, which may suggest an implication in foetal death. However, the role of such modulation in the pathogenesis of bovine neosporosis should be investigated more in depth. Comparing parasite burden and lesion development in placentomes from animals carrying viable and non-viable foetuses, higher parasite burden in caruncles and cotyledons from heifers carrying non-viable foetuses suggested that a higher replication of the parasite took place in animals with foetal death. However, similar lesions in caruncles and cotyledons were found in all Nc-Spain7- infected heifers at 20 dpi. Moreover, the total affected area of the placenta was limited, lesions showed a focal distribution, and hypoxia or other signs of placental dysfunction were not found in non-viable foetuses. Therefore, our theory is that in addition to the placental damage, other factors are probably implicated in the foetal death. In relation to the foetal tissues, almost 100% of brain positive samples were found at 20 dpi, as well as inflammatory infiltrate and lesions compatible with N. caninum in foetal CNS, which is in keeping with previous observations that described the establishment and persistence of the parasite in the CNS, where N. caninum transforms into the tachyzoite-bradyzoite stage inside tissue cysts (Buxton et al., 2002). Viable and non-viable foetuses presented similar parasite burdens in the CNS with increasing charges along the experiment. Parasite burden in foetal liver tended to a decrease as have been previously postulated that foetal liver would act as the first target organ in the foetus to spread to immunopriviledged tissues. However, foetuses at mid- gestation are able to control and clear parasite infection at the liver (Buxton et al., 1998; Arranz- Solís et al., 2016), being an evidence of specific foetal immunocompetence at this gestation period, although no specific antibodies were found in foetal sera or in foetal fluids. Certain foetal immunocompetence may be a possible cause of the fewer foetal death observed in our work compared to the infection earlier in gestation (Regidor-Cerrillo et al., 2014). Brain lesions could not be compared between viable and non-viable foetuses since non-viable foetuses presented autolysis of the CNS. Slightly more acute lesions were found in liver and lung from non-viable foetuses, that presented mild multifocal necrotic foci in addition to the inflammation also observed in viable foetuses. Mild mononuclear myositis and myocarditis were also found in all these five foetuses. As observed in the placenta, differences in burden and lesions between viable and non-viable foetuses did not seem enough by themselves to explain the foetal death observed in two foetuses. An immune-mediated pathogenesis has been suggested as a possible cause of abortion (Quinn et al., 2002b). It has been shown that the multiplication of the parasite in the placenta alters the immunological balance at the maternal-foetal interface and this alteration depends on the isolate Chapter V ~ General Discussion 168 implicated in the infection (Rojo-Montejo et al., 2009b; Caspe et al., 2012; Regidor-Cerrillo et al., 2014). In the view of the evasion of the immune responses at 10 dpi by Nc-Spain7, we hypothesize that the intrinsic ability of rapid multiplication of this isolate led to a profound tissue damage with liberation of host cell components, which could be responsible for PRRs activation at 20 dpi. In consequence, the expression of all the pro-inflammatory cytokines and chemokines, specially of IFN-γ, TNF-α and IL-12p40, which correlated with an extensive inflammatory infiltrate of T CD4 +, T CD8+cells and macrophages, was dramatically enhanced. Contrary to observed in control and Nc-Spain1H-infected placentas, CD8+ still prevailed over CD4+ in Nc-Spain7-infected samples at 20 dpi. As indicated above, CD4/CD8 ratio is usually > 1 in cattle infected by N. caninum, in contrast to observations made in sheep (Arranz-Solís et al., 2016). Although the causes of this predominance remain unknown, it has been postulated that a ratio < 1 in cattle is associated to an early stage of the infection. However, it cannot be ruled out that this difference may be related to the isolate employed in the view of the results presented in this Doctoral Thesis. Remarkably, placental samples infected with the high virulent isolate showed a predominant Th1 response at 20 dpi, with 10-fold less IL-4 than IFN-γ. IFN-γ is required to limit parasite proliferation (Innes et al., 1995; Baszler et al., 1999; Innes et al., 2002), although a critical threshold of the IFN-γ response is also required to limit adverse effects on pregnancy (Almería et al., 2017). These results may indicate Th1/Th2 deviation towards a Th1 predominance as one possible cause of the foetal death. In addition to Th1/Th2 balance, the increases of other local pro-inflammatory cytokines (Il-12p40, IL-8, TNF-α), iNOS and immune cells were associated with more-severe histopathological changes in vivo (Rosbottom et al., 2008; Regidor-Cerrillo et al., 2014; Arranz- Solís et al., 2016), and exacerbated immunological reactions, particularly in the placenta, have been previously postulated as one of the causes of abortion (Raghupathy, 1997; Quinn et al., 2002b; Almería et al., 2010). Moreover, the number of T and B lymphocytes and phagocytic cells was higher in placentomes from animals carrying non-viable foetuses and molecules such IL-8, iNOS and TNF-α were up-regulated in those animals, supporting the hypothesis that exacerbate immune responses in Nc-Spain7-infected placentas may contribute to the foetal death. On the other hand, Nc-Spain7 multiplication and the consequent cell lysis led to a profound alteration of the ECM organization that may facilitate its direct transmission to the foetus but may also be implicated in the mechanism of foetal death. Loss of fibronectin, vimentin and collagen in focal necrotic areas together with down-regulation of MMP-2, MMP-14 and TIMP-2 in Nc-Spain7- infected placentas from animals carrying non-viable foetuses may indicate that the high virulent isolate can inhibit tissue repair, given that IL-17A and TGF-β1 have synergistic properties increasing IL-6 production during fibrosis processes (Dufour, 2018), and stimulating the release of procollagen and fibronectin (Choy & Rose-John, 2017). In addition, absence of IL-6 and IL-17A modulation by Nc-Spain7 may be understood as a part of the evasion strategy of this isolate, given that both cytokines have been associated with potent pro-inflammatory properties against N. caninum infection (Pinheiro et al., 2010; Almería et al., 2011; Flynn & Marshall, 2011; Peckham et al., 2014). The lack of stimulation by Nc-Spain7 was also observed in placental cells in vitro and in bovine macrophages (García-Sánchez et al., submitted). Interestingly, TGF-β1 was diminished in placentomes from animals carrying non-viable foetuses compared to viable foetuses, as was observed in vitro both in trophoblast and caruncular cells. TGF-β1 dysregulation might also contribute to the foetal death observed in Nc-Spain7 heifers since it is crucial in neutralizing the Th1 inflammatory responses (Entrican, 2002). Chemokines and endothelial adhesion molecules genes were also upregulated in Nc-Spain7-infected placentomes, associated with an extensive Chapter V ~ General Discussion 169 inflammatory infiltrate of macrophages and T cells and severe histopathological changes. Although CCLs expression have been linked to protection against N. caninum infection (Mineo et al., 2010; Kameyama et al., 2012; Abe et al., 2015), exaggerated levels of these molecules were related to an increased incidence of abortion via increasing iNOS (Brandonisio et al., 2002; Menezes-Souza et al., 2012) or the inflammatory infiltrated in damaged areas (Pfaff et al., 2005; Wang et al., 2007; Ferro et al., 2008). As introduced above, low virulence traits were observed in the Nc-Spain1H isolate in vitro and in mice model, and infection of pregnant heifers early in gestation did not lead to foetal death, although transmission was observed (Rojo-Montejo et al., 2009a; Rojo-Montejo et al., 2009b; Regidor-Cerrillo et al., 2010). In the experimental infection in pregnant heifers carried out during this Doctoral Thesis, the low-virulence isolate showed dissimilar early infection dynamics and immune response and ECM modulation that those described for the high-virulence isolate. The low virulent isolate Nc-Spain1H induced fever quickly after inoculation (1 dpi). However, only half of the animals presented fever along the experimental period and a second peak of fever was not detected in these animals compared to Nc-Spain7-infected heifers. In addition, lower systemic IFN-γ levels and delayed seroconversion respect to the high virulent isolate were observed in vivo. It is unknown if the immune response developed in Nc-Spain 1H-infected animals was able to reduce parasite burden, limiting the tissue damage or if the low capacity of the isolate to multiply in host tissues may be associated with the reduction or absence of repeated antigenic stimulus (Buxton et al.,1991; Rojo-Montejo et al., 2009b; Regidor-Cerrillo et al., 2014). Presence of Nc-Spain1H was not found in any placental or foetal tissue, and specific humoral responses were not found either in Nc-Spain1H-infected foetuses at 10 dpi. Lesions were not found either in placentas or in foetuses from animals infected with the NcSpain1H isolate, apart from serum extravasation in placentomes. However, intravenous inoculation is supposed to disseminate the parasite quickly through the organism. We hypothesize that the low rate of multiplication of Nc-Spain 1H tachyzoites, observed in placental tissues in vitro, especially in caruncular cells, where the growth of this isolate is clearly slow, might imply a low level of parasitaemia and a low risk of placental infection. Interestingly, despite the absence of parasite detection, Nc-Spain1H induced a solid Th1 response at early stages of infection (10 dpi), that remained steady at later stages (20 dpi). Clues of this rapid activation of the pro-inflammatory responses by the low-virulence isolate were already found in vitro, with higher levels of TLR-2, TNF-α and IL-8 in trophoblast and caruncular cells infected with Nc-Spani1H than with Nc-Spain7 at early times post-infection. In addition, this observation agrees with previous proteome and transcriptome studies where the low-virulence isolate induced higher changes in host cells (Horcajo et al., 2017; Horcajo et al., 2018; García- Sánchez et al., submitted), and a higher stimulation of the immune responses from bovine macrophages in vitro compared to the Nc-Spain7 isolate (García-Sánchez et al., 2019). Th1 responses in Nc-Spain1H-infected placentas would be counterbalanced by a higher expression of anti-inflammatory and regulatory cytokines, minimizing pathology. These results could explain the more effective control of the parasite, thereby contributing to its lower proliferation, and suggest a better adaptation of the low virulence isolate to replicate in the placenta without compromising gestation, maintaining a delicate balance between suppression and induction of the host immune response to ensure foetal survival and vertical transmission. How placental Chapter V ~ General Discussion 170 tissues can recognize and mount an appropriate immune response against this isolate without parasite detection is a fascinating question that remains unknown. Existence of soluble factors of Nc-Spain1H might contribute to earlier Nc-Spain1H recognition, as has been previously suggested for the interaction between other pathogens and some PRRs (Martínez-Colón et al., 2019). For example, higher expression of TLR-9 is observed in caruncular samples infected with Nc-Spain1H than Nc-Spain7 at 20 dpi, when Nc-Spain7 or the products derived from tissue damage have already been recognized by the rest of the PRRs, supporting the hypothesis that there are differences in the antigen exposition between high and low virulent isolates as well as different activation routes of the innate responses. TLR-9 expression by antigen presenting cells is essential for initiating the innate responses and developing an effective Th1-type immune response after T. gondii infection (Minns et al., 2006), and a recent transcriptomic work showed its activation in bovine macrophages after N. caninum infection (García-Sánchez et al., submitted). Higher expression of highly immunogenic SRS membrane proteins was observed in Nc-Spain1H than in Nc-Spain7 (Horcajo et al., 2018), supporting our theory of differential antigen exposition. Contrary to Nc-Spain7, Nc-Spain1H induced a balanced or even predominant Th2 response in terms of IL-4 and IFN-γ. This observation may explain the absence of foetal death observed here and in previous experimental infections with this low virulent isolate (Rojo-Montejo et al., 2009b), together with the lower proliferation and the absence of lesions in the Nc-Spain1H-infected placentas and foetuses. Therefore, differences in the Th- responses between isolates may be a cause of the different outcome observed. Apart from IFN-γ, other pro-inflammatory cytokines, as IL-1β, TNF-α, IL-12p40, IL-8 and IL-6, showed an important implication in host defence during N. caninum infection in the placenta (Entrincan, 2002; Rosbottom et al., 2008; Almería et al., 2011). Interestingly, IL-6 and IL-17A expression levels were only increased in Nc-Spain1H-infected placentas at 10 dpi, decreasing at 20 dpi. IL-6 and IL-17A might be relatively important in controlling N. caninum in the first moments post-infection, and the relation between IL-6, IL-17A and TGF-β1 may contribute to lesions resolution, which contributes to maintain the homeostasis of the bovine placenta. Together with IL-4 and TGF-β1, IL-10 increase in caruncle counter- regulates pro-inflammatory cytokines at the maternal-foetal interface to avoid rejection (Innes, 2007), although it allows parasite proliferation and vertical transmission. Curiously, a lower expression of IL-10 was detected in cotyledonary samples infected by both parasites. Although specific research is needed, IL-10 downregulation in foetal cotyledons could implicate an intrinsic control mechanism of foetal trophoblast against external aggressions. Rapid upregulation of chemokines and cytokines may be associated with a slight increment of macrophages in Nc- Spain1H placentomes at 10 dpi. Macrophages, the first defensive mechanism against bovine neosporosis, may contribute to parasite migration (Lambert et al., 2006; Lachenmaier et al., 2011; Furtado et al., 2012; García-Sánchez et al., 2019), acting as “Trojan horse”. In fact, recent finding demonstrated an enhanced motility of bovine macrophages infected with N. caninum (García- Sánchez et al., 2019). ICAM-1, also upregulated by Nc-Spain1H, is involved in paracellular transmigration of T. gondii (Barragan et al., 2005), facilitating parasite dissemination and vertical transmission. Therefore, paracellular migration and “Trojan horse” strategy may be important for parasite dissemination in vivo, in particular into immune-privileged tissues as the placenta and in low-virulence isolates (Barragan et al., 2005). Interestingly, our results showed upregulation of ECM modulators by Nc-Spain1H, as observed in trophoblast cells in vitro (Horcajo et al., 2017), and lack of ECM components alteration. ECM remodelling may also help to the arriving of Chapter V ~ General Discussion 171 leukocytes in infected areas and host MMPs are involved in infected macrophage dissemination (Seipel et al., 2010). Moreover, MMPs and TIMPs modulation by the infection might be involved in the crossing through the placental barrier (Wang & Lai, 2013), and it has been postulated that ameboid migration of infected macrophages diminished collagen destruction (Olafsson et al., 2018; García-Sánchez et al., 2019). Therefore, we hypothesize that ECM modification, together with the “Trojan horse” mechanism, may be used by the low virulent isolate in order to facilitate its passage throughout the placenta and be transmitted to the foetus. At 20 dpi, only one Nc-Spain1H-infected animal presented positive placental samples (5/9). From these 5 positive samples, only one caruncular sample was N. caninum DNA positive whereas four samples from cotyledon were positive in the animal infected with Nc-Spain1H that presented positivity in the placenta, confirming transmission of this isolate and, on the other hand, supporting findings made in vitro which suggested the presence of both permissive and resistant cells in very close proximity in the placenta. Low presence of Nc-Spain1H in placental samples was similar to a previous experimental study at early gestation (Rojo-Montejo et al., 2009b), and contrary to observed during infection with the high-virulence isolate Nc-Spain7 (Regidor-Cerrillo et al., 2014). In addition, there were no evident lesions at the placenta in Nc-Spain1H-infected animals at 20 dpi. Taken together, these results suggest a limited colonization of maternal placenta by Nc-Spain1H, which is consistent with the low proliferation rate of this isolate under in vitro conditions. Moreover, earlier activation of the innate immune responses observed in vivo and higher activation of TLR-2 observed in vitro, may also contribute to the development of an earlier and more efficient immune response for elimination of the tachyzoites, diminishing the burden in the caruncle and limiting tissue damage as postulated above. On the other hand, transmission of Nc-Spain1H to the foetal compartment occurs, although no parasite DNA, lesions or foetal antibodies were found in Nc-Spain1H-infected foetuses. However, the origin of this isolate and a previous experimental infection at early gestation corroborated the existence of transmission (Rojo-Montejo et al., 2009a; Rojo-Montejo et al., 2009b). As described above, exacerbated immune responses were not observed in Nc-Spain1H placentas. In fact, Th1/Th2 responses were always balanced or even deviated towards Th2 predominance in Nc-Spain1H-infected placentomes, and anti-inflammatory cytokines such as TGF-β1, crucial to maintain gestation, was always upregulated. In addition, expression levels of ICAM-1, chemokines and ECM modulators were maintained at 20 dpi and might implicate that Nc-Spain1H crosses the placental barrier by hijacking immune cells in a mechanism similar to a “Trojan horse”. It is therefore tempting to hypothesize that if the experimental design of the study had allowed a longer gestation, probably Nc-Spain1H had not induced foetal death and abortion contrary to Nc- Spain7 since its strategy to be transmitted seems to be based on higher modulation of the extracellular matrix and probably a higher transmission via immune cells, limiting tissue damage and abortion. In summary, differences between isolates are related to higher invasion and proliferation abilities of the Nc-Spain7 isolate and to the existence of an evasion mechanism consisting of a delayed activation of the immune responses of the high-virulence isolate in placental tissues compared to the low-virulence isolate. We propose that Nc-Spain7 arrives to the foetus using its efficient replication ability and an evasion of the placental immune response strategy in the early stage of Chapter V ~ General Discussion 172 the infection. However, high multiplication leads to placental damage and exacerbated immune responses that may be the causes of the abortion. Higher proliferation and lower activation of the immune responses may be associated with higher expression of certain proteins as ROP17 or glideosome proteins and the lack or the minor expression of certain highly immunogenic SRS proteins in the high-virulence isolate, respectively. Virulence may be beneficial if the goal of the parasite is to kill the host and to increase the chances of ingestion and infection of the second host, as may be happening with Nc-Spain7, that would exploit its intrinsic invasion and replication capabilities as well as an early evasion of the placental immune responses in order to arrive to the foetus, leading to alteration of the local Th1/Th2 balance, placental damage and abortion. Both findings were confirmed both in vitro an in vivo. On the other hand, the low-virulence isolate is quickly recognized and controlled by the placental immune responses, which along with the lower proliferation of this isolate, mainly in caruncular cells, suggest a different mechanism to cross the placental barrier. Modulation of the ECM and the hijacking of the immune cells may be the strategies used by this parasite in order to be transmitted avoiding placental damage. As previously suggested, when the parasite is transmitted from the dam to the offspring, a low virulence could be selected in order to preserve the success of transmission. Therefore, a correlation between low virulence and vertical transmission could be established (Rojo-Montejo et al., 2009b).Trying to go deeper in the pathogenesis of the bovine neosporosis, comparison between viable and non-viable foetuses from Nc-Spain7-infected animals showed that hardly placental and foetal burden and lesions would explain the foetal death by themselves, and an implication of the maternal and foetal immune responses in the outcome of the infection was suggested. However, this experiment was not designed to elucidate the cause of the abortion and a closer monitoring of the foetus and sequential sampling and culling would be necessary. In vitro results were confirmed in vivo proving the validity of both models for investigating the processes implicated in the lytic cycle of N. caninum tachyzoites and the early infection dynamics in the bovine placenta as well as for the study of the immune responses and ECM modulation induced in both cell lines after infection with two isolates of different virulence. These studies have revealed interesting insights into likely mechanisms involved in specific phenotypic traits and virulence in N. caninum and lay the foundations for further investigations. CHAPTER VI CONCLUSIONS CONCLUSIONES Chapter VI ~ Conclusions 175 Objective 1: Characterization of parasite interaction between high- and low-virulence isolates of N. caninum and bovine placental target cells in vitro. First. Bovine trophoblast and bovine caruncular epithelial cell lines are good models for exploring the pathways of N. caninum infection during transplacental transmission in the second term of pregnancy. Concretely, proliferation and processes implicated in the lytic cycle of N. caninum in the bovine placenta and immune responses induced in both cell lines after infection can be investigated using this model. Sub-objective 1.1: Characterization of the lytic cycle of high- and low-virulence isolates of N. caninum in bovine placental target cells in vitro. First. Parasite invasion, growth and egression mechanisms show clear differences between bovine trophoblast and caruncular cell lines which suggest that bovine caruncular cells act as a barrier in the bovine placenta, limiting the proliferation of N. caninum whereas bovine trophoblast cells act as a niche for parasite multiplication. Caruncular cells restrict N. caninum infection at two critical stages of the parasite lytic cycle: at the point of parasite adhesion to the host cell and intracellular replication. Despite the role as a barrier, an early egression mechanism described in caruncular cells and high proliferation in trophoblasts indicate that N. caninum is able to cross the caruncular layer of the bovine placenta facilitating rapid transmission of the parasite to the progeny. Second. The highly virulent isolate Nc-Spain7 shows higher invasion and infection rates as well as higher proliferation than the low-virulence isolate Nc-Spain1H in bovine placental target cells, especially in trophoblast cells, that are one of the bases of their differences in virulence. Sub-objective 1.2: In vitro interaction between N. caninum and the placental target cells from an immunological level. First. Placental cells participate in the innate immune response against N. caninum at the maternal-foetal interface without any extra stimulus but the parasite, via a rapid pro- inflammatory response characterized by the overexpression of pro-inflammatory IL-8 and TNF-α and the downregulation of regulatory TGF-β1 and IL-6. Second. Bovine trophoblast and caruncular cells are activated early after infection with the low- virulence isolate Nc-Spain1H inducing higher levels of pro-inflammatory TNF-α, which might lead to a better control of this isolate by the immune response. However, Nc-Spain7 does not activate TLR-2 either in trophoblast nor caruncular cells, indicating an evasion mechanism of the immune response by this isolate. Chapter VI ~ Conclusions 176 Objective 2: Study of the early infection in a pregnant bovine model inoculated with high- and low-virulence isolates of N. caninum at mid-gestation. First. Different host-parasite interaction patterns are observed in the bovine placenta with high- and low-virulence isolates of N. caninum during early stages of infection at mid-gestation, suggesting the existence of different evolutionary adaptative strategies used by this parasite for transmission to the offspring. Sub-objective 2.1: Early N. caninum infection dynamics in pregnant heifers after inoculation at mid-gestation with high- and low-virulence isolates. First. The clinical consequences of N. caninum infection at mid-gestation are dependent on the isolate implicated in the infection. Nc-Spain7 tachyzoites are detected earlier than Nc-Spain1H tachyzoites in the placenta (at 10 days post-infection), replicating and leading to lesion development, transmission to the foetus and foetal death, whereas the low virulent isolate Nc- Spain1H showed a lower and delayed dissemination in the placental tissues without lesions development nor transmission during the first 20 days post-infection. These findings may be related to higher proliferation abilities of the high-virulence isolate Nc-Spain7. Sub-objective 2.2: Placental immune response and extracellular matrix organization during the early stages of N. caninum infection in pregnant heifers inoculated with high- and low-virulence isolates at mid-gestation. First. Modulation of the placental immune response at early stages of infection in pregnant cattle inoculated at mid-gestation is dependent on the isolate and the time post-infection. At 10 dpi, the low-virulence isolate Nc-Spain1H triggers activation of classical PRRs and a general increase of the local immune responses, even though its presence is not detected in the placenta at this time, which may contribute to the control of this isolate. Contrary, the high-virulence isolate Nc-Spain7 is not recognized by the classical PRRs in the placenta until 20 dpi, which may indicate the existence of an evasion mechanism that may favour its multiplication and dissemination to the foetus. Detection of Nc-Spain7 when lesions are already delevoped might lead to exacerbate Th1 immune responses in an attempt to control the infection, which may contribute to the foetal death. Second. Despite the lack of classical PRRs activation at 10 dpi, the high virulent isolate induces certain upregulation of inflammatory responses and infiltration of immune cells in the placenta at this time, suggesting the activation of other alternatives routes of the immune responses. Third. A different mechanism of transmission between high and low virulent isolates is suggested. Extracellular matrix seems to be an important modulation point for Nc-Spain1H, which may use this mechanism to cross the placental barrier, maintaining the homeostasis of the placenta and avoiding foetal death. On the other hand, Nc-Spain7 would exploit its intrinsic invasion and replication abilities as well as an early evasion of the placental immune responses in order to arrive to the foetus, leading to placental damage, alteration of the local Th1/Th2 balance, with exacerbated immune responses in the placenta and, finally, to foetal death. Chapter VI ~ Conclusions 177 Fourth. An exacerbated pro-inflammatory immune response at the placenta, unbalanced in favour of Th1 responses, and with higher expression of abortifacient iNOS, TNF-α and IL-8 and wider inflammatory infiltrate in placentomes from animals carrying non-viable foetuses, together with a decrease of TGF-β1 may contribute to the foetal death. In addition, a profound alteration of the extracellular matrix, with decreased expression of MMP-2 and TIMP-2 and destruction of collagen, fibronectin and vimentin, and the lack of activation of tissue repair processes could contribute to the abortion mechanism. Chapter VI ~ Conclusiones 179 Objetivo 1: Caracterización de la interacción parásito-hospedador entre aislados de alta y baja virulencia de N. caninum y células diana de la placenta bovina in vitro. Primera. Las líneas celulares de trofoblasto bovino y de células epiteliales carunculares bovinas son buenos modelos para explorar las rutas de infección por N. caninum durante la transmisión transplacentaria en el segundo trimestre de la gestación. Concretamente, la proliferación y los procesos implicados en el ciclo lítico de N. caninum en la placenta bovina, así como las respuestas inmunitarias inducidas en ambas líneas celulares después de la infección pueden investigarse utilizando este modelo. Subobjetivo 1.1: Caracterización del ciclo lítico de aislados de alta y baja virulencia de N. caninum en células diana de la placenta bovina in vitro. Primera. Los mecanismos de invasión, proliferación y egresión del parásito muestran claras diferencias entre las células del trofoblasto bovino y las células carunculares. Los resultados sugieren que las células carunculares actúan como una barrera en la placenta bovina, limitando la proliferación de N. caninum, mientras que las células del trofoblasto actúan como un nicho para la multiplicación del parásito. Las células de la carúncula limitan la infección por N. caninum en dos etapas críticas del ciclo lítico del parásito: en el punto de adhesión del parásito a la célula hospedadora y en la replicación intracelular. A pesar del papel de barrera, un mecanismo de egresión temprana descrito en las células carunculares y la alta proliferación del parásito en las células del trofoblasto indican que N. caninum es capaz de atravesar la capa caruncular de la placenta bovina, facilitando la rápida transmisión del parásito a la descendencia. Segunda. El aislado de alta virulencia Nc-Spain7 presenta mayores tasas invasión e infección, así como una mayor proliferación que el aislado de baja virulencia Nc-Spain1H en las células diana de la placenta bovina, especialmente en las células del trofoblasto, lo que supone una de las bases de sus diferencias de virulencia. Subobjetivo 1.2: Interacción in vitro entre N. caninum y las células diana de la placenta bovina a nivel inmunológico. Primera. Las células placentarias participan en la respuesta inmunitaria innata frente a N. caninum en la interfaz materno-fetal sin necesidad de ningún estímulo adicional excepto el parásito, a través de una rápida respuesta proinflamatoria caracterizada por un aumento en la expresión de las citoquinas proinflamatorias IL-8 y TNF-α y la disminución de la citoquina reguladora TGF-β1 y la IL-6. Segunda. Las células del trofoblasto bovino y las células carunculares se activan poco después de la infección con el aislado de baja virulencia Nc-Spain1H, induciendo niveles más altos de la citoquina proinflamatoria TNF-α, lo que podría conducir a un mejor control de este aislado por parte de la respuesta inmunitaria. Sin embargo, el aislado de alta virulencia Nc-Spain7 no activa el receptor TLR-2 ni en las células de trofoblasto ni en las carunculares, lo que indica un mecanismo de evasión de la respuesta inmunitaria de este aislado. Chapter VI ~ Conclusiones 180 Objetivo 2: Estudio de la infección temprana en un modelo bovino gestante inoculado con aislados de alta y baja virulencia de N. caninum en el segundo tercio de la gestación. Primera. Se observan diferentes patrones de interacción parásito-hospedador en la placenta bovina con aislados de alta y baja virulencia de N. caninum durante la infección temprana en el segundo tercio de la gestación, lo que sugiere la existencia de diferentes estrategias evolutivas de adaptación utilizadas por este parásito para la transmisión a la descendencia. Subobjetivo 2.1: Dinámica de la infección temprana en novillas gestantes tras la inoculación de aislados de alta y baja virulencia de N. caninum en el segundo tercio de la gestación. Primera. Las consecuencias clínicas de la infección por N. caninum en el segundo tercio de la gestación dependen del aislado implicado en la infección. El aislado Nc-Spain7 se detecta antes en la placenta que el aislado Nc-Spain1H, replicándose y provocando el desarrollo de lesiones, la transmisión al feto y la muerte fetal. El aislado de baja virulencia Nc-Spain1H mostró una diseminación más baja y tardía en los tejidos placentarios sin que se produjeran lesiones ni se transmitiera durante los primeros 20 días tras la infección. Estos hallazgos pueden estar relacionados con la mayor capacidad de proliferación del aislado de alta virulencia Nc-Spain7 en la placenta bovina. Subobjetivo 2.2: Respuesta inmunitaria en la placenta y organización de la matriz extracelular durante las primeras etapas de la infección con aislados de alta y baja virulencia de N. caninum en novillas gestantes inoculadas en el segundo tercio de la gestación. Primera. La modulación de la respuesta inmunitaria de la placenta en las primeras etapas de la infección en novillas gestantes inoculadas en el segundo tercio de la gestación depende del aislado y del momento estudiado tras la infección. A los 10 días tras la infección, el aislado de baja virulencia Nc-Spain1H desencadena la activación de los PRRs clásicos y un aumento general de la respuesta inmunitaria local que puede contribuir al control de la infección, a pesar de no ser detectado en la placenta en dicho momento. Por el contrario, el aislado de alta virulencia Nc-Spain7 no es reconocido por los PRRs clásicos en la placenta hasta los 20 dpi, lo que puede indicar la existencia de un mecanismo de evasión que favorece su multiplicación y diseminación al feto. La detección del aislado Nc-Spain7 por parte de la respuesta inmunitaria de la placenta cuando las lesiones ya se han producido podría llevar a una exacerbación de la respuesta de tipo Th1 en un intento de controlar la infección, contribuyendo a la muerte fetal. Segunda. A pesar de la falta de activación de los PRRs clásicos a los 10 días post-infección, el aislado de alta virulencia induce un cierto aumento de la expresión de la respuesta inflamatoria, así como la infiltración de las células inmunitarias en la placenta, lo que sugiere la activación de otras rutas alternativas de la respuesta inmunitaria. Tercera. Se sugiere un mecanismo de transmisión distinto entre aislados de alta y baja virulencia. La matriz extracelular parece ser un importante punto de modulación para el aislado de baja virulencia Nc-Spain1H, que puede utilizar este mecanismo para cruzar la barrera placentaria, manteniendo la homeostasis de la placenta y evitando la muerte fetal. Por otro lado, el aislado de alta virulencia Nc-Spain7 aprovecharía sus capacidades intrínsecas de invasión y proliferación, Chapter VI ~ Conclusiones 181 así como la evasión temprana de la respuesta inmunitaria de la placenta para llegar al feto, lo que provocaría lesiones en la placenta, la alteración del equilibrio local de las respuestas Th1/Th2, con una respuesta inmunitaria exacerbada en la placenta y, finalmente, la muerte fetal. Cuarta. Una respuesta inmunitaria proinflamatoria exacerbada en la placenta, con desequilibrio a favor de la respuesta de tipo Th1, mayor expresión de citoquinas y enzimas abortivas como la iNOS, el TNF-α y la IL-8 y un infiltrado inflamatorio más amplio en las placentas de las novillas con fetos no viables, junto con una disminución del TGF-β1 pueden contribuir a la muerte fetal. Además, la profunda alteración de la matriz extracelular, con disminución de la expresión de la MMP-2 y el TIMP-2 y la destrucción de colágeno, fibronectina y vimentina, así como la falta de activación de los procesos de reparación tisular podrían contribuir al mecanismo del aborto. CHAPTER VII REFERENCES BIBLIOGRAFÍA Chapter VII ~ References 185 Abe, C., Tanaka, S., Nishimura, M., Ihara, F., Xuan, X., Nishikawa, Y., 2015. Role of the chemokine receptor CCR5- dependent host defense system in Neospora caninum infections. Parasit. Vectors 8 (1), 5. Adams, D.H., Rlloyd, A., 1997. Chemokines: leucocyte recruitment and activation cytokines. Lancet 349 (9050), 490- 495. Adeyinka, F.D., 2012. The development of the bovine placentome and associated structures during gestation. (Doctoral Thesis). https://mro.massey.ac.nz/handle/10179/4271. Adl, S.M., Simpson, A.G., Lane, C.E., Lukeš, J., Bass, D., Bowser, S.S., Brown, M.W., Burki, F., Dunthorn, M., Hampl, V., 2012. The revised classification of eukaryotes. J. Eukaryot. Microbiol. 59 (5), 429-514. Adomako-Ankomah, Y., Wier, G.M., Borges, A.L., Wand, H.E., Boyle, J.P., 2014. Differential locus expansion distinguishes Toxoplasmatinae species and closely related strains of Toxoplasma gondii. MBio 5 (1), e01003-13, 10.1128/mBio.01003-13. Aguado-Martínez, A., Álvarez-García, G., Fernández-García, A., Risco-Castillo, V., Marugán-Hernández, V., Ortega-Mora, L.M., 2009. Failure of a vaccine using immunogenic recombinant proteins rNcSAG4 and rNcGRA7 against neosporosis in mice. Vaccine 27 (52), 7331-7338, 10.1016/j.vaccine.2009.09.050. Almería, S., De Marez, T., Dawson, H., Araujo, R., Dubey, J.P., Gasbarre, L.C., 2003. Cytokine gene expression in dams and foetuses after experimental Neospora caninum infection of heifers at 110 days of gestation. Parasite Immunol. 25 (7), 383-392. Almería, S., Vidal, D., Ferrer, D., Pabon, M., Fernández-de-Mera, M.I., Ruiz-Fons, F., Alzaga, V., Marco, I., Calvete, C., Lavin, S., Gortazar, C., López-Gatius, F., Dubey, J.P., 2007. Seroprevalence of Neospora caninum in non-carnivorous wildlife from Spain. Vet. Parasitol. 143 (1), 21-28. Almería, S., Araujo, R., Tuo, W., López-Gatius, F., Dubey, J.P., Gasbarre, L.C., 2010. Fetal death in cows experimentally infected with Neospora caninum at 110 days of gestation. Vet. Parasitol., 10.1016/j.vetpar.2009.12.044. Almería, S., Araujo, R., Darwich, L., Dubey, J., Gasbarre, L., 2011. Cytokine gene expression at the materno‐foetal interface after experimental Neospora caninum infection of heifers at 110 days of gestation. Parasite Immunol. 33 (9), 517-523. Almería, S., López-Gatius, F., 2013. Bovine neosporosis: Clinical and practical aspects. Res. Vet. Sci. 95 (2), 303-309, 10.1016/j.rvsc.2013.04.008. Almería, S., Serrano-Pérez, B., Darwich, L., Domingo, M., Mur-Novales, R., Regidor-Cerrillo, J., Cabezón, O., Pérez-Maillo, M., López-Helguera, I., Fernández-Aguilar, X., 2016a. Foetal death in naive heifers inoculated with Neospora caninum isolate Nc-Spain7 at 110 days of pregnancy. Exp. Parasitol. 168, 62-69. Almería, S., Serrano-Pérez, B., Darwich, L., Mur-Novales, R., García-Ispierto, I., Cabezón, O., López-Gatius, F. 2016b. Cytokine gene expression in aborting and non-aborting dams and in their foetuses after experimental infection with Neospora caninum at 110 days of gestation. Vet. Parasitol. 227, 138-142. Almería, S, Serrano-Pérez, B, López-Gatius, F, 2017. Immune response in bovine neosporosis: Protection or contribution to the pathogenesis of abortion. Microb. Pathog. 109, 177-182, 10.1016/j.micpath.2017.05.042. Álvarez-García, G., Pereira-Bueno, J., Gómez-Bautista, M., Ortega-Mora, L.M., 2002. Pattern of recognition of Neospora caninum tachyzoite antigens by naturally infected pregnant cattle and aborted foetuses. Vet. Parasitol. 107 (1-2), 15- 27. Álvarez-García, G., Collantes-Fernández, E., Costas, E., Rebordosa, X., Ortega-Mora, L.M., 2003. Influence of age and purpose for testing on the cut-off selection of serological methods in bovine neosporosis. Vet. Res. 34 (3), 341-352, 10.1051/vetres:2003009. Amarante-Paffaro, A., Queiroz, G.S., Correa, S.T., Spira, B., Bevilacqua, E., 2004. Phagocytosis as a potential mechanism for microbial defense of mouse placental trophoblast cells. J. Reprod. Fertil. 128 (2), 207-218, 10.1530/rep.1.00214. Ander, S., Rudzki, E., Arora, N., Sadovsky, Y., Coyne, C.B., Boyle, J.P., 2017. Human placental syncytiotrophoblasts restrict Toxoplasma gondii vertical transmission at two distinct stages and induce CCL22 in response to infection. bioRxiv, 170944. Anderson, M.L., Blanchard, P.C., Barr, B.C., Dubey, J.P., Hoffman, R.L., Conrad, P.A., 1991. Neospora-like protozoan infection as a major cause of abortion in California dairy cattle. J. Am. Vet. Med. Assoc. 198 (2), 241-244. Anderson, M.L., Palmer, C.W., Thurmond, M.C., Picanso, J.P., Blanchard, P.C., Breitmeyer, R.E., Layton, A.W., McAllister, M., Daft, B., Kinde, H., 1995. Evaluation of abortions in cattle attributable to neosporosis in selected dairy herds in California. J. Am. Vet. Med. Assoc. 207 (9), 1206-1210. Andrianarivo, A.G., Choromanski, L., McDonough, S.P., Packham, A.E., Conrad, P.A., 1999. Immunogenicity of a killed Chapter VII ~ References 186 whole Neospora caninum tachyzoite preparation formulated with different adjuvants. Int. J. Parasitol. 29 (10), 1613- 1625. Andrianarivo, A.G., Rowe, J.D., Barr, B.C., Anderson, M.L., Packham, A.E., Sverlow, K.W., Choromanski, L., Loui, C., Grace, A., Conrad, P.A., 2000. A POLYGEN-adjuvanted killed Neospora caninum tachyzoite preparation failed to prevent foetal infection in pregnant cattle following i.v./i.m. experimental tachyzoite challenge. Int. J. Parasitol. 30 (9), 985-990. Andrianarivo, A.G., Barr, B.C., Anderson, M.L., Rowe, J.D., Packham, A.E., Sverlow, K.W., Conrad, P.A., 2001. Immune responses in pregnant cattle and bovine fetuses following experimental infection with Neospora caninum. Parasitol. Res. 87 (10), 817-825. Andrianarivo, A.G., Anderson, M.L., Rowe, J.D., Gardner, I.A., Reynolds, J.P., Choromanski, L., Conrad, P.A., 2005. Immune responses during pregnancy in heifers naturally infected with Neospora caninum with and without immunization. Parasitol. Res. 96 (1), 24-31. Armengol, R., Pabón, M., Santolaria, P., Cabezón, O., Adelantado, C., Yaniz, J., López-Gatius, F., Almería, S., 2007. Low seroprevalence of Neospora caninum infection associated with the limousin breed in cow-calf herds in Andorra, Europe. J. Parasitol. 93 (5), 1029-1032. Arranz-Solís, D., Benavides, J., Regidor-Cerrillo, J., Fuertes, M., Ferre, I., Ferreras Mdel, C., Collantes-Fernández, E., Hemphill, A., Pérez, V., Ortega-Mora, L.M., 2015. Influence of the gestational stage on the clinical course, lesional development and parasite distribution in experimental ovine neosporosis. Vet. Res. 46, 19-014-0139-y, 10.1186/s13567-014-0139-y. Arranz-Solís, D., Benavides, J., Regidor-Cerrillo, J., Horcajo, P., Castaño, P., del Carmen Ferreras, M., Jiménez-Pelayo, L., Collantes-Fernández, E., Ferre, I., Hemphill, A., 2016. Systemic and local immune responses in sheep after Neospora caninum experimental infection at early, mid and late gestation. Vet. Res. 47 (1), 1-13. Arranz-Solís, D., 2016. Nuevos modelos animales para el estudio de la infección por Neospora caninum durante la gestación. (Doctoral Thesis). Complutense University of Madrid Ashdown, H., Dumont, Y., Ng, M., Poole, S., Boksa, P., Luheshi, G., 2006. The role of cytokines in mediating effects of prenatal infection on the fetus: implications for schizophrenia. Mol. Psychiatry 11 (1), 47. Atkinson, B., King, G., Amoroso, E., 1984. Development of the caruncular and intercaruncular regions in the bovine endometrium. Biol. Reprod. 30 (3), 763-774. Atkinson, R., Harper, P.A., Ryce, C., Morrison, D.A., Ellis, J.T., 1999. Comparison of the biological characteristics of two isolates of Neospora caninum. Parasitology 118 (Pt 4), 363-370. Aurrecoechea, C., Barreto, A., Basenko, E.Y., Brestelli, J., Brunk, B.P., Cade, S., Crouch, K., Doherty, R., Falke, D., Fischer, S., 2016. EuPathDB: the eukaryotic pathogen genomics database resource. Nucleic Acids Res. 45 (D1), D581-D591. Awad, M., Koshi, K., Kizaki, K., Takahashi, T., Hashizume, K., 2014. SOLD1 is expressed in bovine trophoblast cell lines and regulates cell invasiveness. Reprod. Biol.and Endocrinol. 12 (1), 55. Banu, S., Arosh, J., Chapdelaine, P., Fortier, M., 2005. Expression of prostaglandin transporter in the bovine uterus and fetal membranes during pregnancy. Biol. Reprod. 73 (2), 230-236. Barber, J.S., Holmdahl, O.J., Owen, M.R., Guy, F., Uggla, A., Trees, A.J., 1995. Characterization of the first European isolate of Neospora caninum (Dubey, Carpenter, Speer, Topper and Uggla). Parasitology 111 (Pt 5), 563-568. Barling, K.S., McNeill, J.W., Thompson, J.A., Paschal, J.C., McCollum, F.T., Craig, T.M., Adams, L.G., 2000. Association of serologic status for Neospora caninum with postweaning weight gain and carcass measurements in beef calves. J. Am. Vet. Med. Assoc. 217 (9), 1356-1360. Barling, K.S., Lunt, D.K., Graham, S.L., Choromanski, L.J., 2003. Evaluation of an inactivated Neospora caninum vaccine in beef feedlot steers. J. Am. Vet. Med. Assoc. 222 (5), 624-627. Barna, F., Debache, K., Vock, C.A., Kuster, T., Hemphill, A., 2013. In vitro effects of novel ruthenium complexes in Neospora caninum and Toxoplasma gondii tachyzoites. Antimicrob. Agents Chemother. 57 (11), 5747-5754, 10.1128/AAC.02446-12. Barr, B.C., Conrad, P.A., Dubey, J., Anderson, M.L., 1991a. Neospora-like encephalomyelitis in a calf: pathology, ultrastructure, and immunoreactivity. J. Vet. Diagn. Invest. 3 (1), 39-46. Barr, B.C., Anderson, M.L., Dubey, J.P., Conrad, P.A., 1991b. Neospora-like protozoal infections associated with bovine abortions. Vet. Pathol. 28 (2), 110-116. Barr, B.C., Conrad, P.A., Breitmeyer, R., Sverlow, K., Anderson, M.L., Reynolds, J., Chauvet, A.E., Dubey, J.P., Ardans, A.A., 1993. Congenital Neospora infection in calves born from cows that had previously aborted Neospora-infected fetuses: four cases (1990-1992). J. Am. Vet. Med. Assoc. 202 (1), 113-117. Chapter VII ~ References 187 Barr, B.C., Rowe, J.D., Sverlow, K.W., BonDurant, R.H., Ardans, A.A., Oliver, M.N., Conrad, P.A., 1994. Experimental reproduction of bovine fetal Neospora infection and death with a bovine Neospora isolate. J. Vet. Diagn. Invest. 6 (2), 207-215. Barragan, A., Sibley, L.D., 2002. Transepithelial migration of Toxoplasma gondii is linked to parasite motility and virulence. J. Exp. Med. 195 (12), 1625-1633. Barragan, A., Brossier, F., Sibley, L.D., 2005. Transepithelial migration of Toxoplasma gondii involves an interaction of intercellular adhesion molecule 1 (ICAM‐1) with the parasite adhesin MIC2. Cell. Microbiol. 7 (4), 561-568. Bartels, C.J., van, S.G., Veldhuisen, J.P., van, d.B., Wouda, W., Dijkstra, T., 2006. Effect of Neospora caninum-serostatus on culling, reproductive performance and milk production in Dutch dairy herds with and without a history of Neospora caninum-associated abortion epidemics. Prev. Vet. Med. 77 (3-4), 186-198. Bartley, P.M., Kirvar, E., Wright, S., Swales, C., Esteban-Redondo, I., Buxton, D., Maley, S.W., Schock, A., Rae, A.G., Hamilton, C., Innes, E.A., 2004. Maternal and fetal immune responses of cattle inoculated with Neospora caninum at mid-gestation. J. Comp. Pathol. 130 (2-3), 81-91. Bartley, P.M., Wright, S., Sales, J., Chianini, F., Buxton, D., Innes, E.A., 2006. Long-term passage of tachyzoites in tissue culture can attenuate virulence of Neospora caninum in vivo. Parasitology 133, 421-432. Bartley, P.M., Wright, S., Chianini, F., Buxton, D., Innes, E.A., 2008. Inoculation of Balb/c mice with live attenuated tachyzoites protects against a lethal challenge of Neospora caninum Parasitology 135 (Pt 1), 13-21, 10.1017/S0031182007003526. Bartley PM, M., Wright SE, M., Maley SW, D., Macaldowie CN, D., Nath, M.D., Hamilton CM, D., Katzer, F.D., Buxton, D.D., Innes EA, P., 2012. Maternal and foetal immune responses of cattle following an experimental challenge with Neospora caninum at day 70 of gestation. Vet. Res. 43 (1), 38, 10.1186/1297-9716-43-38. Bartley, P.M., Katzer, F., Rocchi, M.S., Maley, S.W., Benavides, J., Nath, M., Pang, Y., Cantón, G., Thomson, J., Chianini, F., Innes, E.A., 2013. Development of maternal and foetal immune responses in cattle following experimental challenge with Neospora caninum at day 210 of gestation. Vet. Res. 44, 91-9716-44-91, 10.1186/1297-9716-44-91. Basso, W., Venturini, L., Venturini, M.C., Hill, D.E., Kwok, O.C., Shen, S.K., Dubey, J.P., 2001. First isolation of Neospora caninum from the feces of a naturally infected dog. J. Parasitol. 87 (3), 612-618. Basso, W., Schares, S., Barwald, A., Herrmann, D.C., Conraths, F.J., Pantchev, N., Vrhovec, M.G., Schares, G., 2009. Molecular comparison of Neospora caninum oocyst isolates from naturally infected dogs with cell culture-derived tachyzoites of the same isolates using nested polymerase chain reaction to amplify microsatellite markers. Vet. Parasitol. 160 (1-2), 43-50, 10.1016/j.vetpar.2008.10.085. Basso, W., Schares, S., Minke, L., Barwald, A., Maksimov, A., Peters, M., Schulze, C., Müller, M., Conraths, F.J., Schares, G., 2010. Microsatellite typing and avidity analysis suggest a common source of infection in herds with epidemic Neospora caninum-associated bovine abortion. Vet. Parasitol. 173 (1-2), 24-31, 10.1016/j.vetpar.2010.06.009. Baszler, T.V., Long, M.T., McElwain, T.F., Mathison, B.A., 1999. Interferon-gamma and interleukin-12 mediate protection to acute Neospora caninum infection in BALB/c mice. Int. J. Parasitol. 29 (10), 1635-1646. Batbayar, T., Nomura, Y., Ishij, Y., Shirai, K., 2017. Affinity of placental decorin for collagen. Biosci. Biotechnol. Biochem. 64 (11), 2478-2481. Bautista, J.M., Marín-García, P., Diez, A., Azcárate, I.G., Puyet, A., 2014. Malaria proteomics: Insights into the parasite– host interactions in the pathogenic space. J. Proteomics 97, 107-125. Bazer, F.W., Ying, W., Wang, X., Dunlap, K.A., Zhou, B., Johnson, G.A., Wu, G., 2015. The many faces of interferon tau. Amino Acids 47 (3), 449-460. Beck, H, Blake, D, Dardé, M, Felger, I, Pedraza-Díaz, S, Regidor-Cerrillo, J, Gómez-Bautista, M, Ortega-Mora, LM, Putignani, L, Shiels, B, Tait, A, Weir, W, 2009. Molecular approaches to diversity of populations of apicomplexan parasites. Int. J. Parasitol. 39 (2), 175-189 10.1016/j.ijpara.2008.10.001. Behnke, M.S., Khan, A., Wootton, J.C., Dubey, J.P., Tang, K., Sibley, L.D., 2011. Virulence differences in Toxoplasma mediated by amplification of a family of polymorphic pseudokinases. Proc. Natl. Acad. Sci. U.S.A. 108 (23), 9631-9636, 10.1073/pnas.1015338108. Beloosesky, R., Gayle, D.A., Amidi, F., Nunez, S.E., Babu, J., Desai, M., Ross, M.G., 2006. N-acetyl-cysteine suppresses amniotic fluid and placenta inflammatory cytokine responses to lipopolysaccharide in rats. Obstet. Gynecol. 194 (1), 268-273. Benavides, J., Katzer, F., Maley, S.W., Bartley, P.M., Cantón, G., Palarea-Albaladejo, J., Purslow, C.A., Pang, Y., Rocchi, M.S., Chianini, F., Buxton, D., Innes, E.A., 2012. High rate of transplacental infection and transmission of Neospora caninum following experimental challenge of cattle at day 210 of gestation. Vet. Res. 43 (1), 83, 10.1186/1297-9716- Chapter VII ~ References 188 43-83. Benavides, J., Collantes-Fernández, E., Ferre, I., Pérez, V., Campero, C., Mota, R., Innes, E., Ortega-Mora, L.M., 2014. Experimental ruminant models for bovine neosporosis: what is known and what is needed. Parasitology 141 (11), 1471- 1488, 10.1017/S0031182014000638. Benirschke, K., Kaufmann, P., 2000. Anatomy and pathology of the umbilical cord and major fetal vessels. In Pathology of the Human Placenta. Springer, pp. 335-398. Bevilacqua, E., Hoshida, M.S., Amarante-Paffaro, A., Albieri-Borges, A., Zago Gomes, S., 2010. Trophoblast phagocytic program: roles in different placental systems. Int. J. Dev. Biol. 54 (2-3), 495-505, 10.1387/ijdb.082761eb. Bezerra, M.A., Pereira, L.M., Bononi, A., Biella, C.A., Baroni, L., Pollo-Oliveira, L., Yatsuda, A.P., 2017. Constitutive expression and characterization of a surface SRS (NcSRS67) protein of Neospora caninum with no orthologue in Toxoplasma gondii. Parasitol. Int. 66 (2), 173-180. Bjerkas, I., Mohn, S.F., Presthus, J., 1984. Unidentified cyst-forming sporozoon causing encephalomyelitis and myositis in dogs. Z. Parasitenkd. 70 (2), 271-274. Björkman, N., 1969. Light and electron microscopic studies on cellular alterations in the normal bovine placentome. Anat. Rec. 163 (1), 17-29. Björkman, N., 1973. Fine structure of the fetal-maternal area of exchange in the epitheliochorial and endotheliochorial types of placentation. Cells Tissues Organs 86 (61), 1-22. Black, M.W., Boothroyd, J.C., 2000. Lytic cycle of Toxoplasma gondii. Microbiol. Mol. Biol. Rev. 64 (3), 607-623. Blader, I.J., Saeij, J.P., 2009. Communication between Toxoplasma gondii and its host: impact on parasite growth, development, immune evasion, and virulence. APMIS 117 (5-6), 458-476, 10.1111/j.1600-0463.2009.02453.x. Bland, J.M., Altman, D.G., 1998. Survival probabilities (the Kaplan-Meier method). BMJ 317 (7172), 1572. Borbely, A.U., Sandri, S., Fernandes, I.R., Prado, K.M., Cardoso, E.C., Correa-Silva, S., Albuquerque, R., Knöfler, M., Beltrão-Braga, P., Campa, A., Bevilacqua, E., 2014. The term basal plate of the human placenta as a source of functional extravillous trophoblast cells. Reprod. Biol. Endocrinol. 12, 7, 10.1186/1477-7827-12-7. Boucher, L.E., Bosch, J., 2015. The apicomplexan glideosome and adhesins - Structures and function. J. Struct. Biol. 190 (2), 93-114, 10.1016/j.jsb.2015.02.008. Boucher, E., Marin, M., Holani, R., Young-Speirs, M., Moore, D., Cobo, E., 2018. Characteristic pro-inflammatory cytokines and host defence cathelicidin peptide produced by human monocyte-derived macrophages infected with Neospora caninum. Parasitology 145 (7), 871-884. Bougdour, A., Tardieux, I., Hakimi, M.A., 2014. Toxoplasma exports dense granule proteins beyond the vacuole to the host cell nucleus and rewires the host genome expression. Cell. Microbiol. 16 (3), 334-343, 10.1111/cmi.12255. Boysen, P., Klevar, S., Olsen, I., Storset, A.K., 2006. The protozoan Neospora caninum directly triggers bovine NK cells to produce gamma interferon and to kill infected fibroblasts. Infect. Immun. 74 (2), 953-960. Brach, M.A., deVos, S., Gruss, H.J., Herrmann, F., 1992. Prolongation of survival of human polymorphonuclear neutrophils by granulocyte-macrophage colony-stimulating factor is caused by inhibition of programmed cell death. Blood 80 (11), 2920-2924. Brandonisio, O., Panaro, M., Fumarola, I., Sisto, M., Leogrande, D., Acquafredda, A., Spinelli, R., Mitolo, V., 2002. Macrophage chemotactic protein-1 and macrophage inflammatory protein-1α induce nitric oxide release and enhance parasite killing in Leishmania infantum-infected human macrophages. Clin. Exp. Med. 2 (3), 125-129. Brasil, T.R., Freire-de-Lima, C.G., Morrot, A., Vetö Arnholdt, A.C., 2017. Host-Toxoplasma gondii coadaptation leads to fine tuning of the immune response. Front. Immunol. 8, 1080. Bridger, P., Haupt, S., Klisch, K., Leiser, R., Tinneberg, H., Pfarrer, C., 2007a. Validation of primary epitheloid cell cultures isolated from bovine placental caruncles and cotyledons. Theriogenology 68 (4), 592-603. Bridger, P.S., Menge, C., Leiser, R., Tinneberg, H.R., Pfarrer, C.D., 2007b. Bovine caruncular epithelial cell line (BCEC-1) isolated from the placenta forms a functional epithelial barrier in a polarised cell culture model. Placenta 28 (11-12), 1110-1117, S0143-4004(07)00184-1. Bridger, P.S., Haupt, S., Leiser, R., Johnson, G.A., Burghardt, R.C., Tinneberg, H., Pfarrer, C., 2008. Integrin activation in bovine placentomes and in caruncular epithelial cells isolated from pregnant cows. Biol. Reprod. 79 (2), 274-282. Bridger, P.S., 2008. Validation and establishment of cell culture models to study invasion and feto-maternal interaction in the bovine placentome. (Doctoral Thesis) http://geb.uni-giessen.de/geb/volltexte/2008/5839/ Bücher, K., Leiser, R., Tiemann, U., Pfarrer, C., 2006. Platelet-activating factor receptor (PAF-R) and acetylhydrolase Chapter VII ~ References 189 (PAF-AH) are co-expressed in immature bovine trophoblast giant cells throughout gestation, but not at parturition. Prostaglandins Other Lipid Mediat. 79 (1-2), 74-83. Buxton, D., Thomson, K., Maley, S., Wright, S., Bos, H.J., 1991. Vaccination of sheep with a live incomplete strain (s48) of Toxoplasma gondii and their immunity to challenge when pregnant. Vet. Rec. 129 (5), 89-93. Buxton, D., Maley, S.W., Wright, S., Thomson, K.M., Rae, A.G., Innes, E.A., 1998. The pathogenesis of experimental neosporosis in pregnant sheep. J. Comp. Pathol. 118 (4), 267-279. Buxton, D., McAllister, M.M., Dubey, J.P., 2002. The comparative pathogenesis of neosporosis. Trends Parasitol. 18 (12), 546-552. Calarco, L., Barratt, J., Ellis, J., 2018. Genome wide identification of mutational hotspots in the apicomplexan parasite Neospora caninum and the implications for virulence. Genome Biol. Evol. 10 (9), 2417-2431. Calero-Bernal, R., Horcajo, P., Hernández, M., Ortega-Mora, L.M., Fuentes, I., 2019. Absence of Neospora caninum DNA in Human Clinical Samples, Spain. Emerg. Infect. Dis. 25 (6), 1226-1227, 10.3201/eid2506.181431. Canada, N., Meirele, C.S., Ferreira, P., da Costa, J.M.C., Rocha, A., 2006. Artificial insemination of cows with semen in vitro contaminated with Neospora caninum tachyzoites failed to induce neosporosis. Vet. Parasitol. 139 (1-3), 109-114. Cannas, A., Naguleswaran, A., Müller, N., Eperon, S., Gottstein, B., Hemphill, A., 2003a. Vaccination of mice against experimental Neospora caninum infection using NcSAG1-and NcSRS2-based recombinant antigens and DNA vaccines. Parasitology 126 (4), 303-312. Cannas, A., Naguleswaran, A., Müller, N., Gottstein, B., Hemphill, A., 2003b. Reduced cerebral infection of Neospora caninum-infected mice after vaccination with recombinant microneme protein NcMIC3 and ribi adjuvant. J. Parasitol. 89 (1), 44-50. Cantón, G.J., Katzer, F., Benavides-Silván, J., Maley, S.W., Palarea-Albaladejo, J., Pang, Y., Smith, S., Bartley, P.M., Rocchi, M., Innes, E.A., 2013. Phenotypic characterisation of the cellular immune infiltrate in placentas of cattle following experimental inoculation with Neospora caninum in late gestation. Vet. Res. 44 (1), 60. Cantón, G.J., Katzer, F., Maley, S.W., Bartley, P.M., Benavides-Silvan, J., Palarea-Albaladejo, J., Pang, Y., Smith, S.H., Rocchi, M.S., Buxton, D., Innes, E.A., Chianini, F., 2014a. Inflammatory infiltration into placentas of Neospora caninum challenged cattle correlates with clinical outcome of pregnancy. Vet. Res. 45, 11-9716-45-11, 10.1186/1297-9716-45- 11. Cantón, G.J., Konrad, J.L., Moore, D.P., Caspe, S.G., Palarea-Albaladejo, J., Campero, C.M., Chianini, F., 2014b. Characterization of immune cell infiltration in the placentome of water buffaloes (Bubalus bubalis) infected with Neospora caninum during pregnancy. J. Comp. Pathol. 150 (4), 463-468, 10.1016/j.jcpa.2013.12.003. Cantón, G.J., Katzer, F., Maley, S.W., Bartley, P.M., Benavides-Silván, J., Palarea-Albaladejo, J., Pang, Y., Smith, S.H., Rocchi, M., Buxton, D., Innes, E.A., Chianini, F., 2014c. Cytokine expression in the placenta of pregnant cattle after inoculation with Neospora caninum. Vet. Immunol. Immunopathol. 161 (1-2), 77-89, 10.1016/j.vetimm.2014.07.004. Carruthers, V., Boothroyd, J.C., 2007. Pulling together: an integrated model of Toxoplasma cell invasion. Curr. Opin. Microbiol. 10 (1), 83-89, S1369-5274(06)00100-7. Carvalho Neta, A.V., Stynen, A.P., Paixao, T.A., Miranda, K.L., Silva, F.L., Roux, C.M., TSolís, R.M., Everts, R.E., Lewin, H.A., Adams, L.G., Carvalho, A.F., Lage, A.P., Santos, R.L., 2008. Modulation of the bovine trophoblastic innate immune response by Brucella abortus. Infect. Immun. 76 (5), 1897-1907, 10.1128/IAI.01554-07. Carvalho, J.V., Alves, C.M., Cardoso, M.R., Mota, C.M., Barbosa, B.F., Ferro, E.A., Silva, N.M., Mineo, T.W., Mineo, J.R., Silva, D.A., 2010. Differential susceptibility of human trophoblastic (BeWo) and uterine cervical (HeLa) cells to Neospora caninum infection. Int. J. Parasitol. 40 (14), 1629-1637, 10.1016/j.ijpara.2010.06.010. Caspe, S.G., Moore, D.P., Leunda, M.R., Cano, D.B., Lischinsky, L., Regidor-Cerrillo, J., Álvarez-García, G., Echaide, I.G., Bacigalupe, D., Ortega-Mora, L.M., Odeon, A.C., Campero, C.M., 2012. The Neospora caninum-Spain 7 isolate induces placental damage, fetal death and abortion in cattle when inoculated in early gestation. Vet. Parasitol. 189 (2-4), 171- 181, 10.1016/j.vetpar.2012.04.034. Castillo, C., Muñoz, L., Carrillo, I., Liempi, A., Gallardo, C., Galanti, N., Maya, J.D., Kemmerling, U., 2017a. Ex vivo infection of human placental chorionic villi explants with Trypanosoma cruzi and Toxoplasma gondii induces different Toll‐like receptor expression and cytokine/chemokine profiles. Am. J.Reprod. Immunol. 78 (1), 10.1111/aji.12660. Castillo, C., Muñoz, L., Carrillo, I., Liempi, A., Medina, L., Galanti, N., Maya, J.D., Kemmerling, U., 2017b. Toll-like receptor-2 mediates local innate immune response against Trypanosoma cruzi in ex vivo infected human placental chorionic villi explants. Placenta 60, 40-46. Cavalcanti, Y.V., Brelaz, M.C., Neves, J.K., Ferraz, J.C., Pereira, V.R., 2012. Role of TNF-Alpha, IFN-Gamma, and IL-10 in the Development of Pulmonary Tuberculosis. Pulm. Med. 2012, 10.1155/2012/745483. Chapter VII ~ References 190 Chaouat, G., Assal Meliani, A., Martal, J., Raghupathy, R., Elliott, J.F., Mosmann, T., Wegmann, T.G., 1995. IL-10 prevents naturally occurring fetal loss in the CBA x DBA/2 mating combination, and local defect in IL-10 production in this abortion-prone combination is corrected by in vivo injection of IFN-tau. J. Immunol. 154 (9), 4261-4268. Chaussabel, D., Semnani, R.T., McDowell, M.A., Sacks, D., Sher, A., Nutman, T.B., 2003. Unique gene expression profiles of human macrophages and dendritic cells to phylogenetically distinct parasites. Blood 102 (2), 672-681, 10.1182/blood-2002-10-3232. Choy, E., Rose-John, S., 2017. Interleukin-6 as a multifunctional regulator: inflammation, immune response, and fibrosis. J. Scleroderma Relat. Disord. 2 (2), S1-S5. Chryssafidis, A.L., Cantón, G., Chianini, F., Innes, E.A., Madureira, E.H., Gennari, S.M., 2014. Pathogenicity of Nc-Bahia and Nc-1 strains of Neospora caninum in experimentally infected cows and buffaloes in early pregnancy. Parasitol. Res. 113 (4), 1521-1528, 10.1007/s00436-014-3796-x. Clark, D.A., Chaouat, G., Arck, P.C., Mittruecker, H.W., Levy, G.A., 1998. Cytokine-dependent abortion in CBA x DBA/2 mice is mediated by the procoagulant fgl2 prothrombinase. J. Immunol. 160 (2), 545-549. Clark, I.M., Swingler, T.E., Sampieri, C.L., Edwards, D.R., 2008. The regulation of matrix metalloproteinases and their inhibitors. Int. J. Biochem. Cell Biol. 40 (6-7), 1362-1378. Collantes-Fernández, E., Zaballos, A., Álvarez-García, G., Ortega-Mora, L.M., 2002. Quantitative detection of Neospora caninum in bovine aborted fetuses and experimentally infected mice by real-time PCR. J. Clin. Microbiol. 40 (4), 1194- 1198. Collantes-Fernández, E., Rodríguez-Bertos, A., Arnaiz-Seco, I., Moreno, B., Adúriz, G., Ortega-Mora, L.M., 2006a. Influence of the stage of pregnancy on Neospora caninum distribution, parasite loads and lesions in aborted bovine foetuses. Theriogenology 65 (3), 629-641. Collantes-Fernández, E., López-Pérez, I., Álvarez-García, G., Ortega-Mora, L.M., 2006b. Temporal distribution and parasite load kinetics in blood and tissues during Neospora caninum infection in mice. Infect. Immun. 74 (4), 2491- 2494. Collantes-Fernández, E., Arnaiz-Seco, I., Burgos, B.M., Rodríguez-Bertos, A., Adúriz, G., Fernández-García, A., Ortega- Mora, L.M., 2006c. Comparison of Neospora caninum distribution, parasite loads and lesions between epidemic and endemic bovine abortion cases. Vet. Parasitol. 142 (1-2), 187-191. Collantes-Fernández, E., Arrighi, R.B., Álvarez-García, G., Weidner, J.M., Regidor-Cerrillo, J., Boothroyd, J.C., Ortega- Mora, L.M., Barragán, A., 2012. Infected dendritic cells facilitate systemic dissemination and transplacental passage of the obligate intracellular parasite Neospora caninum in mice. PLoS One 7 (3), e32123, 10.1371/journal.pone.0032123. Conrad, P.A., Barr, B.C., Sverlow, K.W., Anderson, M., Daft, B., Kinde, H., Dubey, J.P., Munson, L., Ardans, A., 1993a. In vitro isolation and characterization of a Neospora sp. from aborted bovine foetuses. Parasitology 106 (3), 239-249. Conrad, P.A., Sverlow, K., Anderson, M., Rowe, J., BonDurant, R., Tuter, G., Breitmeyer, R., Palmer, C., Thurmond, M., Ardans, A., 1993b. Detection of serum antibody responses in cattle with natural or experimental Neospora infections. J. Vet. Diagn. Invest. 5 (4), 572-578. Correia, A., Ferreirinha, P., Botelho, S., Belinha, A., Leitão, C., Caramalho, Í, Teixeira, L., González-Fernandéz, Á, Appelberg, R., Vilanova, M., 2015. Predominant role of interferon-γ in the host protective effect of CD8 T cells against Neospora caninum infection. Sci. Rep. 5, 14913. Croken M.M., Ma, Y., Markillie, L.M., Taylor, R.C., Orr, G., Weiss, L.M., Kim, K., 2014. Distinct strains of Toxoplasma gondii feature divergent transcriptomes regardless of developmental stage. PLoS One 9 (11), :e111297, 10.1371/journal.pone.0111297. Darwich, L., Li, Y., Serrano-Pérez, B., Mur-Novales, R., García-Ispierto, I., Cabezón, O., López-Gatius, F., Almería, S., 2016. Maternal and foetal cytokine production in dams naturally and experimentally infected with Neospora caninum on gestation day 110. Res. Vet. Sci. 107, 55-61. Da Silva, M.V., Ferreira França, F.B., Mota, C.M., de Macedo Júnior, A.G., Ramos, E.L., Santiago, F.M., Mineo, J.R., Mineo, T.W., 2017. Dectin-1 compromises innate responses and host resistance against Neospora caninum infection. Front. Inmunol. 8, 245, 10.3389/fimmu.2017.00245. Davison, H.C., Guy, C.S., McGarry, J.W., Guy, F., Williams, D.J., Kelly, D.F., Trees, A.J., 2001. Experimental studies on the transmission of Neospora caninum between cattle. Res. Vet. Sci. 70 (2), 163-168. De Marez, T., Liddell, S., Dubey, J.P., Jenkins, M.C., Gasbarre, L., 1999. Oral infection of calves with Neospora caninum oocysts from dogs: humoral and cellular immune responses. Int. J. Parasitol. 29 (10), 1647-1657. De Yaniz, M.G., Moore, D.P., Odeon, A.C., Cano, A., Cano, D.B., Leunda, M.R., Campero, C.M., 2007. Humoral immune response in pregnant heifers inoculated with Neospora caninum tachyzoites by conjunctival route. Vet. Parasitol. 148 Chapter VII ~ References 191 (3-4), 213-218. Debache, K., Guionaud, C., Kropf, C., Boykin, D., Stephens, C.E., Hemphill, A., 2011. Experimental treatment of Neospora caninum-infected mice with the arylimidamide DB750 and the thiazolide nitazoxanide. Exp. Parasitol. 129 (2), 95-100, 10.1016/j.exppara.2011.07.010. Debache, K., Hemphill, A., 2012a. Effects of miltefosine treatment in fibroblast cell cultures and in mice experimentally infected with Neospora caninum tachyzoites. Parasitology 139 (7), 934-944, 10.1017/S0031182012000066. Debache, K., Hemphill, A., 2012b. Intra-cisternal vaccination induces high-level protection against Neospora caninum infection in mice. Vaccine 30 (28), 4209-4215, 10.1016/j.vaccine.2012.04.050. Debache, K., Hemphill, A., 2013. Differential effects of intranasal vaccination with recombinant NcPDI in different mouse models of Neospora caninum infection. Parasite Immunol. 35 (1), 11-20, 10.1111/pim.12013. Deisher, T.A., Haddix, T.L., Montgomery, K.F., Pohlman, T.H., Kaushansky, K., Harlan, J.M., 1993. The role of protein kinase C in the induction of VCAM-1 expression on human umbilical vein endothelial cells. FEBS Lett. 331 (3), 285-290. Dellarupe, A., Regidor-Cerrillo, J., Jiménez-Ruiz, E., Schares, G., Unzaga, J.M., Venturini, M.C., Ortega-Mora, L.M., 2014a. Clinical outcome and vertical transmission variability among canine Neospora caninum isolates in a pregnant mouse model of infection. Parasitology 141 (3), 356-366, 10.1017/S0031182013001479. Dellarupe, A., Regidor-Cerrillo, J., Jiménez-Ruiz, E., Schares, G., Unzaga, J.M., Venturini, M.C., Ortega-Mora, L.M., 2014b. Comparison of host cell invasion and proliferation among Neospora caninum isolates obtained from oocysts and from clinical cases of naturally infected dogs. Exp. Parasitol. 145, 22-28, 10.1016/j.exppara.2014.07.003. Diehl, S., Rincón, M., 2002. The two faces of IL-6 on Th1/Th2 differentiation. Mol. Immunol. 39 (9), 531-536. Dijkstra, T., Lam, T.J., Bartels, C.J., Eysker, M., Wouda, W., 2008. Natural postnatal Neospora caninum infection in cattle can persist and lead to endogenous transplacental infection. Vet. Parasitol. 152 (3-4), 220-225. Dilly, M., Hambruch, N., Haeger, J., Pfarrer, C., 2010. Epidermal growth factor (EGF) induces motility and upregulates MMP‐9 and TIMP‐1 in bovine trophoblast cells. Mol. Reprod. Dev. 77 (7), 622-629. Dilly, M., Hambruch, N., Shenavai, S., Schuler, G., Froehlich, R., Haeger, J., Ozalp, G., Pfarrer, C., 2011. Expression of matrix metalloproteinase (MMP)-2, MMP-14 and tissue inhibitor of matrix metalloproteinase (TIMP)-2 during bovine placentation and at term with or without placental retention. Theriogenology 75 (6), 1104-1114. Dixon, S.E., Stilger, K.L., Elias, E.V., Naguleswaran, A., Sullivan, W.J.,Jr, 2010. A decade of epigenetic research in Toxoplasma gondii. Mol. Biochem. Parasitol. 173 (1), 1-9, 10.1016/j.molbiopara.2010.05.001. Donahoe, S.L., Lindsay, S.A., Krockenberger, M., Phalen, D., Šlapeta, J., 2015. A review of neosporosis and pathologic findings of Neospora caninum infection in wildlife. Int. J. Parasitol. Parasites Wildl. 4 (2), 216-238, 10.1016/j.ijppaw.2015.04.002. Donnelly, L., Campling, G., 2016. Functions of the placenta. Anaesthesia & intensive care medicine 17 (7), 349-353. Dorsch, M., De Yaniz, M., Fiorani, F., Hecker, Y., Odeón, A., Morrell, E., Campero, C., Barbeito, C., Moore, D., 2019. A descriptive study of lectin histochemistry of the placenta in cattle following inoculation of Neospora caninum. J. Comp. Pathol. 166, 45-53. Dowse, T., Soldati, D., 2004. Host cell invasion by the apicomplexans: the significance of microneme protein proteolysis. Curr. Opin. Microbiol. 7 (4), 388-396. Dubey, J.P., Carpenter, J.L., Speer, C.A., Topper, M.J., Uggla, A., 1988a. Newly recognized fatal protozoan disease of dogs. J. Am. Vet. Med. Assoc. 192 (9), 1269-1285. Dubey, J.P., Hattel, A.L., Lindsay, D.S., Topper, M.J., 1988b. Neonatal Neospora caninum infection in dogs: isolation of the causative agent and experimental transmission. J Am Vet Med Assoc, 193(10), 1259-1263. Dubey, J.P., Lindsay, D.S., Anderson, M.L., Davis, S.W., Shen, S.K., 1992. Induced transplacental transmission of Neospora caninum in cattle. J. Am. Vet. Med. Assoc. 201 (5), 709-713. Dubey, J.P., Lahunta, A.d., 1993. Neosporosis associated congenital limb deformities in a calf. Appl. Parasitol. 34, 229- 233. Dubey, J.P., Lindsay, D.S., 1996. A review of Neospora caninum and neosporosis. Vet. Parasitol. 67 (1-2), 1-59. Dubey, J.P., Lindsay, D.S., Speer, C.A., 1998. Structures of Toxoplasma gondii tachyzoites, bradyzoites, and sporozoites and biology and development of tissue cysts. Clin. Microbiol. Rev. 11 (2), 267-299. Dubey, J.P., Barr, B.C., Barta, J.R., Bjerkas, I., Björkman, C., Blagburn, B.L., Bowman, D.D., Buxton, D., Ellis, J.T., Gottstein, B., Hemphill, A., Hill, D.E., Howe, D.K., Jenkins, M.C., Kobayashi, Y., Koudela, B., Marsh, A.E., Mattsson, J.G., McAllister, M.M., Modry, D., Omata, Y., Sibley, L.D., Speer, C.A., Trees, A.J., Uggla, A., Upton, S.J., Williams, D.J., Lindsay, D.S., 2002. Chapter VII ~ References 192 Redescription of Neospora caninum and its differentiation from related coccidia. Int. J. Parasitol. 32 (8), 929-946. Dubey, J.P., 2003. Review of Neospora caninum and neosporosis in animals. Korean J. Parasitol. 41 (1), 1-16. Dubey, J.P., Sreekumar, C., Knickman, E., Miska, K.B., Vianna, M.C.B., Kwok, O.C.H., Hill, D.E., Jenkins, M.C., Lindsay, D.S., Greene, C.E., 2004. Biologic, morphologic, and molecular characterisation of Neospora caninum isolates from littermate dogs. Int. J. Parasitol. 34 (10), 1157-1167. Dubey, J.P., 2005. Neosporosis in cattle. Vet. Clin. North Am. Food Anim. Pract. 21 (2), 473-483. Dubey, J.P., Schares, G., 2006. Diagnosis of bovine neosporosis. Vet. Parasitol. 140 (1-2), 1-34. Dubey, J.P., Buxton, D., Wouda, W., 2006. Pathogenesis of bovine neosporosis. J. Comp. Pathol. 134 (4), 267-289. Dubey, J.P., Schares, G., Ortega-Mora, L.M., 2007. Epidemiology and control of neosporosis and Neospora caninum. Clin. Microbiol. Rev. 20 (2), 323-367. Dubey, J., 2009. Toxoplasmosis in sheep—the last 20 years. Vet. Parasitol. 163 (1), 1-14. Dubey, J., Schares, G., 2011. Neosporosis in animals—the last five years. Vet. Parasitol. 180 (1), 90-108. Dubey, J.P., Hemphill, A., Calero-Bernal, R., Schares, G., 2017. Neosporosis in animals. CRC Press. Duello, T.M., Byatt, J.C., Bremel, R.D., 1986. Immunohistochemical localization of placental lactogen in binucleate cells of bovine placentomes. Endocrinology 119 (3), 1351-1355. Dufour, A.M., Álvarez, M., Russo, B., Chizzolini, C., 2018. Interleukin-6 and type-I collagen production by systemic sclerosis fibroblasts are differentially regulated by interleukin-17A in the presence of transforming growth factor-beta 1. Front. Immunol. 9, 1865, 10.3389/fimmu.2018.01865. Duivenvoorden, J., Lusis, P., 1995. Neospora abortions in eastern Ontario dairy herds. Can. Vet. J. 36 (10), 623. Ealy, A.D., Yang, Q.E., 2009. Control of interferon‐tau expression during early pregnancy in ruminants. Am. J. Reprod. Inmunol. 61 (2), 95-106. Eiras, C., Arnaiz, I., Álvarez-García, G., Ortega-Mora, L.M., Sanjuan, M.L., Yus, E., Dieguez, F.J., 2011. Neospora caninum seroprevalence in dairy and beef cattle from the northwest region of Spain, Galicia. Prev. Vet. Med. 98 (2-3), 128-132, 10.1016/j.prevetmed.2010.10.014. Elli, M., 2005. Manual De Reproducción En Ganado Vacuno. Servet Edit. Entrican, G., 2002. Immune regulation during pregnancy and host-pathogen interactions in infectious abortion. J. Comp. Pathol. 126 (2-3), 79-94, 10.1053/jcpa.2001.0539. Etienne-Manneville, S., Chaverot, N., Strosberg, A.D., Couraud, P.O., 1999. ICAM-1-coupled signaling pathways in astrocytes converge to cyclic AMP response element-binding protein phosphorylation and TNF-alpha secretion. J. Immunol. 163 (2), 668-674. Fanjul-Fernández, M., Folgueras, A.R., Cabrera, S., López-Otín, C., 2010. Matrix metalloproteinases: evolution, gene regulation and functional analysis in mouse models. Biochim. Biophys. Acta 1803 (1), 3-19, 10.1016/j.bbamcr.2009.07.004. Feng, J.S., Yang, Z., Zhu, Y.Z., Liu, Z., Guo, C.C., Zheng, X.B., 2014. Serum IL-17 and IL-6 increased accompany with TGF- beta and IL-13 respectively in ulcerative colitis patients. Int. J. Clin. Exp. Med. 7 (12), 5498-5504. Ferre, I., Serrano-Martínez, E., Martínez, A., Osoro, K., Mateos-Sanz, A., Del-Pozo, I., Adúriz, G., Tamargo, C., Hidalgo, C.O., Ortega-Mora, L.M., 2008. Effects of re-infection with Neospora caninum in bulls on parasite detection in semen and blood and immunological responses. Theriogenology 69 (7), 905-911. Ferro, E.A.V., Mineo, J.R., Ietta, F., Bechi, N., Romagnoli, R., Silva, D.A.O., Sorda, G., Bevilacqua, E., Paulesu, L.R., 2008. Macrophage migration inhibitory factor is up-regulated in human first-trimester placenta stimulated by soluble antigen of Toxoplasma gondii, resulting in increased monocyte adhesion on villous explants. Am. J. Pathol. 172 (1), 50-58. Flynn, R.J., Marshall, E.S., 2011. Parasite limiting macrophages promote IL-17 secretion in naive bovine CD4(+) T-cells during Neospora caninum infection. Vet. Immunol. Immunopathol. 144 (3-4), 423-429, 10.1016/j.vetimm.2011.09.008. Fortier, M., Guilbault, L., Grasso, F., 1988. Specific properties of epithelial and stromal cells from the endometrium of cows. J. Reprod. Fertil. 83 (1), 239-248. Franco, M., Shastri, A.J., Boothroyd, J.C., 2014. Infection by Toxoplasma gondii specifically induces host c-Myc and the genes this pivotal transcription factor regulates. Eukaryot. Cell. 13 (4), 483-493, 10.1128/EC.00316-13. Franczyk, M., Lopucki, M., Stachowicz, N., Morawska, D., Kankofer, M., 2017. Extracellular matrix proteins in healthy and retained placentas, comparing hemochorial and synepitheliochorial placentas. Placenta 50, 19-24, doi: 10.1016/j.placenta.2016.12.014. Chapter VII ~ References 193 Frénal, K., Polonais, V., Marg, J.B., Stratmann, R., Limenitakis, J., Soldati-Favre, D., 2010. Functional dissection of the apicomplexan glideosome molecular architecture. Cell Host Microbe 8 (4), 343-357, 10.1016/j.chom.2010.09.002. French, N.P., Clancy, D., Davison, H.C., Trees, A.J., 1999. Mathematical models of Neospora caninum infection in dairy cattle: transmission and options for control. Int. J. Parasitol. 29 (10), 1691-1704, 10.1016/S0020-7519(99)00131-9. Furtado, J.M., Bharadwaj, A.S., Ashander, L.M., Olivas, A., Smith, J.R., 2012. Migration of Toxoplasma gondii-infected dendritic cells across human retinal vascular endothelium. Invest. Ophthalmol. Vis. Sci. 53 (11), 6856-6862, 10.1167/iovs.12-10384. Gail, M., Gross, U., Bohne, W., 2001. Transcriptional profile of Toxoplasma gondii-infected human fibroblasts as revealed by gene-array hybridization. Mol. Genet. Genomics 265 (5), 905-912. Gajria, B., Bahl, A., Brestelli, J., Dommer, J., Fischer, S., Gao, X., Heiges, M., Iodice, J., Kissinger, J.C., Mackey, A.J., Pinney, D.F., Roos, D.S., Stoeckert, C.J.,Jr, Wang, H., Brunk, B.P., 2008. ToxoDB: an integrated Toxoplasma gondii database resource. Nucleic Acids Res. 36 (Database issue), D553-6, gkm981. García-Sánchez, M., Jiménez-Pelayo, L., Horcajo, P., Regidor-Cerrillo, J., Ólafsson, E.B., Bhandage, A.K., Barragán, A., Werling, D., Ortega-Mora, L.M., Collantes-Fernández, E., 2019. Differential responses of bovine monocyte-derived macrophages to infection by Neospora caninum isolates of high and low virulence. Front. Inmunol. 10, 915. Gayle, D.A., Beloosesky, R., Desai, M., Amidi, F., Nuñez, S.E., Ross, M.G., 2004. Maternal LPS induces cytokines in the amniotic fluid and corticotropin releasing hormone in the fetal rat brain. Am. J. Physiol. Regul. Integr. Comp. Physiol. 286 (6), R1024-1029. Gibney, E.H., Kipar, A., Rosbottom, A., Guy, C.S., Smith, R.F., Hetzel, U., Trees, A.J., Williams, D.J., 2008. The extent of parasite-associated necrosis in the placenta and foetal tissues of cattle following Neospora caninum infection in early and late gestation correlates with foetal death. Int. J. Parasitol. 38 (5), 579-588, 10.1016/j.ijpara.2007.09.015. Gillaux, C., Mehats, C., Vaiman, D., Cabrol, D., Breuiller-Fouche, M., 2011. Functional screening of TLRs in human amniotic epithelial cells. J. Immunol. 187 (5), 2766-2774, 10.4049/jimmunol.1100217. Gold, D.A., Kaplan, A.D., Lis, A., Bett, G.C., Rosowski, E.E., Cirelli, K.M., Bougdour, A., Sidik, S.M., Beck, J.R., Lourido, S., Egea, P.F., Bradley, P.J., Hakimi, M.A., Rasmusson, R.L., Saeij, J.P., 2015. The Toxoplasma dense granule proteins GRA17 and GRA23 mediate the movement of small molecules between the host and the parasitophorous vacuole. Cell Host Microbe, 17 (5), 642-652, 10.1016/j.chom.2015.04.003. Gondim, L.F., Pinheiro, A.M., Santos, P.O., Jesus, E.E., Ribeiro, M.B., Fernandes, H.S., Almeida, M.A., Freire, S.M., Meyer, R., McAllister, M.M., 2001. Isolation of Neospora caninum from the brain of a naturally infected dog, and production of encysted bradyzoites in gerbils. Vet. Parasitol. 101 (1), 1-7. Gondim, L.F.P., Gao, L., McAllister, M.M., 2002. Improved production of Neospora caninum oocysts, cyclical oral transmission between dogs and cattle, and in vitro isolation from oocysts. J. Parasitol. 88 (6), 1159-1163. Gondim, L.F., McAllister, M.M., Pitt, W.C., Zemlicka, D.E., 2004a. Coyotes (Canis latrans) are definitive hosts of Neospora caninum Int. J. Parasitol. 34 (2), 159-161. Gondim, L.F., McAllister, M.M., Mateus-Pinilla, N.E., Pitt, W.C., Mech, L.D., Nelson, M.E., 2004b. Transmission of Neospora caninum between wild and domestic animals. J. Parasitol. 90 (6), 1361-1365. Gondim, L.F., McAllister, M.M., Anderson-Sprecher, R.C., Björkman, C., Lock, T.F., Firkins, L.D., Gao, L., Fischer, W.R., 2004c. Transplacental transmission and abortion in cows administered Neospora caninum oocysts. J. Parasitol. 90 (6), 1394-1400. Gondim, L.F.P., 2006. Neospora caninum in wildlife. Trends Parasitol. 22 (6), 247-252. González, L., Buxton, D., Atxaerandio, R., Adúriz, G., Maley, S., Marco, J.C., Cuervo, L.A., 1999. Bovine abortion associated with Neospora caninum in northern Spain. Vet. Rec. 144 (6), 145-150. González-Warleta, M., Castro-Hermida, J.A., Carro-Corral, C., Cortizo-Mella, J., Mezo, M., 2008. Epidemiology of neosporosis in dairy cattle in Galicia (NW Spain). Parasitol. Res. 102 (2), 243-249, 10.1007/s00436-007-0753-y. Goodswen, S., Kennedy, P., Ellis, J.T., 2013. A review of the infection, genetics, and evolution of Neospora caninum: from the past to the present. Infect. Genet. Evol. 13, 133-150, 10.1016/j.meegid.2012.08.012. Gottstein, B., Razmi, G.R., Ammann, P., Sager, H., Müller, N., 2005. Toltrazuril treatment to control diaplacental Neospora caninum transmission in experimentally infected pregnant mice. Parasitology 130, 41-48. Green, J.A., Xie, S., Quan, X., Bao, B., Gan, X., Mathialagan, N., Beckers, J., Roberts, R.M., 2000. Pregnancy-associated bovine and ovine glycoproteins exhibit spatially and temporally distinct expression patterns during pregnancy. Biol. Reprod. 62 (6), 1624-1631. Chapter VII ~ References 194 Grigg, M.E., Bonnefoy, S., Hehl, A.B., Suzuki, Y., Boothroyd, J.C., 2001. Success and virulence in Toxoplasma as the result of sexual recombination between two distinct ancestries. Science 294 (5540), 161-165, 10.1126/science.1061888. Grunert, E., Ahlers, D., Heuwieser, W., 1989. The role of endogenous estrogens in the maturation process of the bovine placenta. Theriogenology 31 (5), 1081-1091. Guillomot, M., Campion, E., Prézelin, A., Sandra, O., Hue, I., Le Bourhis, D., Richard, C., Biase, F.H., Rabel, C., Wallace, R., Lewin, H., Renard, J.P., Jammes, H., 2000. Spatial and temporal changes of decorin, tipe I collagen and fibronectin expression in normal and clone bovine placenta. Placenta 35 (9), 737-747, 10.1016/j.placenta.2014.06.366. Haeger, J.D., Hambruch, N., Dilly, M., Froehlich, R., Pfarrer, C., 2011. Formation of bovine placental trophoblast spheroids. Cells Tissues Organs 193 (4), 274-284, 10.1159/000320544. Haeger, J.D., Hambruch, N., Pfarrer, C., 2016. The bovine placenta in vivo and in vitro. Theriogenology 86 (1), 306-312, 10.1016/j.theriogenology.2016.04.043. Haider, S., Knöfler, M., 2009. Human tumour necrosis factor: physiological and pathological roles in placenta and endometrium. Placenta 30 (2), 111-123. Haldorson, G.J., Mathison, B.A., Wenberg, K., Conrad, P.A., Dubey, J.P., Trees, A.J., Yamane, I., Baszler, T.V., 2005. Immunization with native surface protein NcSRS2 induces a Th2 immune response and reduces congenital Neospora caninum transmission in mice. Int. J. Parasitol. 35 (13), 1407-1415. Haldorson, G.J., Stanton, J.B., Mathison, B.A., Suarez, C.E., Baszler, T.V., 2006. Neospora caninum: antibodies directed against tachyzoite surface protein NcSRS2 inhibit parasite attachment and invasion of placental trophoblasts in vitro. Exp. Parasitol. 112 (3), 172-178. Hall, C.A., Reichel, M.P., Ellis, J.T., 2005. Neospora abortions in dairy cattle: diagnosis, mode of transmission and control. Vet. Parasitol. 128 (3-4), 231-241. Hambruch, N., Haeger, J.D., Dilly, M., Pfarrer, C., 2010. EGF stimulates proliferation in the bovine placental trophoblast cell line F3 via Ras and MAPK. Placenta 31 (1), 67-74, 10.1016/j.placenta.2009.10.011. Hansen, P., 2011. The immunology of early pregnancy in farm animals. Reprod. Domest. Anim. 46, 18-30. Haraldsen, G., Kvale, D., Lien, B., Farstad, I.N., Brandtzaeg, P., 1996. Cytokine-regulated expression of E-selectin, intercellular adhesion molecule-1 (ICAM-1), and vascular cell adhesion molecule-1 (VCAM-1) in human microvascular endothelial cells. J. Immunol. 156 (7), 2558-2565. Hassan, M.A., Melo, M.B., Haas, B., Jensen, K.D., Saeij, J.P., 2012. De novo reconstruction of the Toxoplasma gondii transcriptome improves on the current genome annotation and reveals alternatively spliced transcripts and putative long non-coding RNAs. BMC Genomics 13, 696-2164-13-696, 10.1186. He, X., Gong, P., Wei, Z., Liu, W., Wang, W., Li, J., Yang, Z., Zhang, X., 2017. Peroxisome proliferator-activated receptor- γ-mediated polarization of macrophages in Neospora caninum infection. Exp. Parasitol. 178, 37-44. Hecker, Y.P., Moore, D.P., Quattrocchi, V., Regidor-Cerrillo, J., Verna, A., Leunda, M.R., Morrell, E., Ortega-Mora, L.M., Zamorano, P., Venturini, M.C., Campero, C.M., 2013. Immune response and protection provided by live tachyzoites and native antigens from the NC-6 Argentina strain of Neospora caninum in pregnant heifers. Vet. Parasitol. 197 (3-4), 436- 446, 10.1016/j.vetpar.2013.07.027. Hehl, A.B., Basso, W.U., Lippuner, C., Ramakrishnan, C., Okoniewski, M., Walker, R.A., Grigg, M.E., Smith, N.C., Deplazes, P., 2015. Asexual expansion of Toxoplasma gondii merozoites is distinct from tachyzoites and entails expression of non- overlapping gene families to attach, invade, and replicate within feline enterocytes. BMC Genomics 16, 66, 10.1186/s12864-015-1225-x. Hemphill, A., 1999. The host-parasite relationship in neosporosis. Adv. Parasitol. 43, 47-104. Hemphill, A., Fuchs, N., Sonda, S., Hehl, A., 1999. The antigenic composition of Neospora caninum. Int. J. Parasitol. 29 (8), 1175-1188. Hemphill, A., Vonlaufen, N., Naguleswaran, A., Keller, N., Riesen, M., Guetg, N., Srinivasan, S., Alaeddine, F., 2004. Tissue culture and explant approaches to studying and visualizing Neospora caninum and its interactions with the host cell. Microsc. Microanal. 10 (5), 602-620. Hemphill, A., Vonlaufen, N., Naguleswaran, A., 2006. Cellular and immunological basis of the host-parasite relationship during infection with Neospora caninum Parasitology 133, 261-278. Hemphill, A., Debache, K., Monney, T., Schorer, M., Guionaud, C., Alaeddine, F., Mueller, N., Mueller, J., 2013. Proteins mediating the Neospora caninum-host cell interaction as targets for vaccination. Front. Biosci. 5, 23-36. Heydorn, A.O., Mehlhorn, H., 2002. A re-evaluation of Neospora and Hammondia spp. Trends Parasitol. 18 (6), 246. Chapter VII ~ References 195 Ho, M.S., Barr, B.C., Rowe, J.D., Anderson, M.L., Sverlow, K.W., Packham, A., Marsh, A.E., Conrad, P.A., 1997. Detection of Neospora sp. from infected bovine tissues by PCR and probe hybridization. J. Parasitol. 83 (3), 508-514. Horcajo, P., Regidor‐Cerrillo, J., Aguado‐Martínez, A., Hemphill, A., Ortega‐Mora, L.M., 2016. Vaccines for bovine neosporosis: current status and key aspects for development. Parasite Immunol. 38 (12), 709-723. Horcajo, P., Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Collantes-Fernández, E., Rozas, D., Hambruch, N., Pfarrer, C., Ortega-Mora, L.M., 2017. Transcriptome modulation of bovine trophoblast cells in vitro by Neospora caninum. Int. J. Parasitol. 47 (12), 791-799, S0020-7519(17)30252-7. Horcajo, P., Xia, D., Randle, N., Collantes-Fernández, E., Wastling, J., Ortega-Mora, L., Regidor-Cerrillo, J., 2018. Integrative transcriptome and proteome analyses define marked differences between Neospora caninum isolates throughout the tachyzoite lytic cycle. J. Proteomics 180, 108-119. Horn, S., Bathgate, R., Lioutas, C., Bracken, K., Ivell, R., 1998. Bovine endometrial epithelial cells as a model system to study oxytocin receptor regulation. Hum. Reprod. Update 4 (5), 605-614. Hradecký, P., Stover, J., Stott, G., 1988. Histology of a heifer placentome after interspecies transfer of a gaur embryo. Theriogenology 30 (3), 593-604. Huckle, W., 2017. Cell-and tissue-based models for study of placental development. In Progress in Molecular Biology and Translational Science. Elsevier, pp. 29-37. Hurtado, A., Adúriz, G., Moreno, B., Barandika, J., Garcıá-Pérez, A.L., 2001. Single tube nested PCR for the detection of Toxoplasma gondii in fetal tissues from naturally aborted ewes. Vet. Parasitol. 102 (1), 17-27. Innes, E.A., Panton, W., Marks, J., Trees, A., Holmdahl, J., Buxton, D., 1995. Interferon gamma inhibits the intracellular multiplication of Neospora caninum, as shown by incorporation of 3H uracil. J. Comp. Pathol. 113 (1), 95-100. Innes, E.A., Wright, S.E., Maley, S., Rae, A., Schock, A., Kirvar, E., Bartley, P., Hamilton, C., Carey, I.M., Buxton, D., 2001. Protection against vertical transmission in bovine neosporosis. Int. J. Parasitol. 31 (13), 1523-1534. Innes, E.A., Andrianarivo, A.G., Björkman, C., Williams, D.J., Conrad, P.A., 2002. Immune responses to Neospora caninum and prospects for vaccination. Trends Parasitol. 18 (11), 497-504. Innes, E.A., Wright, S., Bartley, P., Maley, S., Macaldowie, C., Esteban-Redondo, I., Buxton, D., 2005. The host-parasite relationship in bovine neosporosis. Vet. Immunol. Immunopathol. 108 (1-2), 29-36. Innes, E.A., 2007. The host-parasite relationship in pregnant cattle infected with Neospora caninum. Parasitology 134, 1903-1910. Innes, E.A., Bartley, P.M., Maley, S.W., Wright, S.E., Buxton, D., 2007. Comparative host-parasite relationships in ovine toxoplasmosis and bovine neosporosis and strategies for vaccination. Vaccine 25(30), 5495-5503. Irving, J., Lala, P., 1995. Functional role of cell surface integrins on human trophoblast cell migration: regulation by TGF- β, IGF-II, and IGFBP-1. Exp. Cell Res. 217 (2), 419-427. Jauniaux, E., Gulbis, B., Schandene, L., Collette, J., Hustin, J., 1996. Distribution of interleukin-6 in maternal and embryonic tissues during the first trimester. Mol. Hum. Reprod. 2 (4), 239-243. Jensen, K.D., Wang, Y., Wojno, E.D.T., Shastri, A.J., Hu, K., Cornel, L., Boedec, E., Ong, Y., Chien, Y., Hunter, C.A., 2011. Toxoplasma polymorphic effectors determine macrophage polarization and intestinal inflammation. Cell Host Microbe 9 (6), 472-483. Jensen, K.D., Hu, K., Whitmarsh, R.J., Hassan, M.A., Julien, L., Lu, D., Chen, L., Hunter, C.A., Saeij, J.P., 2013. Toxoplasma gondii rhoptry 16 kinase promotes host resistance to oral infection and intestinal inflammation only in the context of the dense granule protein GRA15. Infect. Immun. 81 (6), 2156-2167, 10.1128/IAI.01185-12. Jesus, E.E., Pinheiro, A.M., Santos, A.B., Freire, S.M., Tardy, M.B., El-Bacha, R.S., Costa, S.L., Costa, M.F., 2013. Effects of IFN-gamma, TNF-alpha, IL-10 and TGF-beta on Neospora caninum infection in rat glial cells. Exp. Parasitol. 133 (3), 269-274, 10.1016/j.exppara.2012.11.016. Jia, B., Lu, H., Liu, Q., Yin, J., Jiang, N., Chen, Q., 2013. Genome-wide comparative analysis revealed significant transcriptome changes in mice after Toxoplasma gondii infection. Parasit. Vectors 6, 161, 10.1186/1756-3305-6-161. Jiménez-Melendez, A., Ojo, K.K., Wallace, A.M., Smith, T.R., Hemphill, A., Balmer, V., Regidor-Cerrillo, J., Ortega-Mora, L.M., Hehl, A.B., Fan, E., Maly, D.J., Van Voorhis, W.C., Álvarez-García, G., 2017. In vitro efficacy of bumped kinase inhibitors against Besnoitia besnoiti tachyzoites. Int. J. Parasitol. 47 (12), 811-821, 10.1016/j.ijpara.2017.08.005. Jiménez-Meléndez, A., Rico-San Román, L., Hemphill, A., Balmer, V., Ortega-Mora, L.M., Álvarez-García, G., 2018. Repurposing of commercially available anti-coccidials identifies diclazuril and decoquinate as potential therapeutic candidates against Besnoitia besnoiti infection. Vet. Parasitol. 261, 77-85. Chapter VII ~ References 196 Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Horcajo, P., Collantes-Fernández, E., Gómez-Bautista, M., Hambruch, N., Pfarrer, C., Ortega-Mora, L.M., 2017. Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum. Parasit. Vectors 10 (1), 463, 10.1186/s13071-017-2409-9. Jiménez-Pelayo, L., García-Sánchez, M., Regidor-Cerrillo, J., Horcajo, P., Collantes-Fernández, E., Gómez-Bautista, M., Hambruch, N., Pfarrer, C., Ortega-Mora, L., 2019. Immune response profile of caruncular and trophoblast cell lines infected by high- (Nc-Spain7) and low-virulence (Nc-Spain1H) isolates of Neospora caninum. Parasit Vectors 12 (1), 218, 10.1186/s13071-019-3466-z. Jiménez-Ruiz, E., Álvarez-García, G., Aguado-Martínez, A., Ortega-Mora, L.M., 2013. Mice congenitally infected with low-to-moderate virulence Neospora caninum isolates exhibited clinical reactivation during the mating period without transmission to the next generation. Exp. Parasitol. 134 (2), 244-248, 10.1016/j.exppara.2013.03.002. Jin, X., Gong, P., Zhang, X., Li, G., Zhu, T., Zhang, M., Li, J., 2017. Activation of ERK signaling via TLR11 induces IL-12p40 production in peritoneal macrophages challenged by Neospora caninum. Front. Microbiol. 8, 1393, 10.3389/fmicb.2017.01393. Johnson, G.A., Burghardt, R.C., Newton, G.R., Bazer, F.W., Spencer, T.E., 1999. Development and characterization of immortalized ovine endometrial cell lines. Biol. Reprod. 61 (5), 1324-1330. Jung, C., Lee, C.Y., Grigg, M.E., 2004. The SRS superfamily of Toxoplasma surface proteins. Int. J. Parasitol. 34 (3), 285- 296, 10.1016/j.ijpara.2003.12.004. Jungi, T.W., Farhat, K., Burgener, I.A., Werling, D., 2011. Toll-like receptors in domestic animals. Cell Tissue Res. 343 (1), 107-120, 10.1007/s00441-010-1047-8. Kalluri, R., Zeisberg, M., 2006. Fibroblasts in cancer. Nature Reviews Cancer 6 (5), 392. Kameyama, K., Nishimura, M., Punsantsogvoo, M., Ibrahim, H.M., Xuan, X., Furuoka, H., Nishikawa, Y., 2012. Immunological characterization of Neospora caninum cyclophilin. Parasitology 139 (3), 294-301, 10.1017/S0031182011002022. Kawaguchi, T., Cho, D., Hayashi, M., Tsukiyama, T., Kimura, K., Matsuyama, S., Minami, N., Yamada, M., Imai, H., 2016. Derivation of induced trophoblast cell lines in cattle by doxycycline-inducible piggyBac vectors. PloS one 11 (12), e0167550, 10.1371/journal.pone.0167550. Khan, I.A., Schwartzman, J.D., Fonseka, S., Kasper, L.H., 1997. Neospora caninum: role for immune cytokines in host immunity. Exp. Parasitol. 85 (1), 24-34. Khan, A., Fujita, A.W., Randle, M., Quinones, K., Shen, A., Oler, N., Sundar, U., Ryan, J., Šlapeta, G., Schares, L., Ortega- Mora, L.M., Dubey, J.P., Grigg, M.E., 2015. Selective sweep of an inbred population of the protozoan pathogen Neospora caninum. Apicomplexa in farm animals 30th June-3rd July, Edinburgh (UK) (oral communication). Khoo, N.K., Bechberger, J.F., Shepherd, T., Bond, S.L., McCrae, K.R., Hamilton, G.S., Lala, P.K., 1998. SV40 Tag transformation of the normal invasive trophoblast results in a premalignant phenotype. I. Mechanisms responsible for hyperinvasivess and resistance to anti‐invasive action of TGFβ. Int. J. Cancer 77 (3), 429-439. Kim, Y.M., Romero, R., Oh, S.Y., Kim, C.J., Kilburn, B.A., Armant, D.R., Nien, J.K., Gómez, R., Mazor, M., Saito, S., 2005. Toll-like receptor 4: a potential link between “danger signals”, the innate immune system, and preeclampsia? Obstet. Gynecol. 193 (3), 921-927. Kimmins, S., Lim, H.C., Parent, J., Fortier, M.A., MacLaren, L.A., 2003. The effects of estrogen and progesterone on prostaglandins and integrin beta 3 (β3) subunit expression in primary cultures of bovine endometrial cells. Domest. Anim. Endocrinol. 25 (2), 141-154. Kizaki, K., Ushizawa, K., Takahashi, T., Yamada, O., Todoroki, J., Sato, T., Ito, A., Hashizume, K., 2008. Gelatinase (MMP- 2 and-9) expression profiles during gestation in the bovine endometrium. Reprod. Biol. Endocrinol. 6 (1), 66. Klevar, S., Kulberg, S., Boysen, P., Storset, A.K., Moldal, T., Björkman, C., Olsen, I., 2007. Natural killer cells act as early responders in an experimental infection with Neospora caninum in calves. Int. J. Parasitol. 37 (3-4), 329-339. Klisch, K., Hecht, W., Pfarrer, C., Schuler, G., Hoffmann, B., Leiser, R., 1999a. DNA content and ploidy level of bovine placentomal trophoblast giant cells. Placenta 20 (5-6), 451-458. Klisch, K., Pfarrer, C., Schuler, G., Hoffmann, B., Leiser, R., 1999b. Tripolar acytokinetic mitosis and formation of feto- maternal syncytia in the bovine placentome: different modes of the generation of multinuclear cells. Anat. Embryol. 200 (2), 229-237. Klisch, K., De Sousa, N.M., Beckers, J., Leiser, R., Pich, A., 2005. Pregnancy associated glycoprotein‐1,‐6,‐7, and‐17 are major products of bovine binucleate trophoblast giant cells at midpregnancy. Mol. Reprod. Dev. 71 (4), 453-460. Chapter VII ~ References 197 Klisch, K., Wooding, F., Jones, C., 2010. The glycosylation pattern of secretory granules in binucleate trophoblast cells is highly conserved in ruminants. Placenta 31 (1), 11-17. Koga, K., Mor, G., 2008. Expression and function of Toll-like receptors at the maternal—fetal interface. Reprod. Sci. 15 (3), 231-242, 10.1177/1933719108316391. Kritzner, S., Sager, H., Blum, J., Krebber, R., Greif, G., Gottstein, B., 2002. An explorative study to assess the efficacy of Toltrazuril-sulfone (Ponazuril) in calves experimentally infected with Neospora caninum. Ann. Clin. Microbiol. Antimicrob. 1 (1), 4. Kul, O., Atmaca, H.T., Anteplioglu, T., Ocal, N., Canpolat, S., 2015. Neospora caninum: the first demonstration of the enteroepithelial stages in the intestines of a naturally infected dog. J. Comp. Pathol. 153 (1), 9-13, 10.1016/j.jcpa.2015.03.005. Lachenmaier, S.M., Deli, M.A., Meissner, M., Liesenfeld, O., 2011. Intracellular transport of Toxoplasma gondii through the blood–brain barrier. J. Neuroimmunol. 232 (1-2), 119-130. Lambert, H., Hitziger, N., Dellacasa, I., Svensson, M., Barragán, A., 2006. Induction of dendritic cell migration upon Toxoplasma gondii infection potentiates parasite dissemination. Cell. Microbiol., 8 (10), 1611-1623. Lambert, H., Barragán, A., 2010. Modelling parasite dissemination: host cell subversion and immune evasion by Toxoplasma gondii. Cell. Microbiol. 12 (3), 292-300. Lambert, H., Dellacasa-Lindberg, I., Barragán, A., 2011. Migratory responses of leukocytes infected with Toxoplasma gondii. Microb. Infect. 13 (1), 96-102, 10.1016/j.micinf.2010.10.002. Laven, R., Peters, A., 2001. Gross morphometry of the bovine placentome during gestation. Reprod. Domest. Anim. 36 (6), 289-296. Lee, R.S., Peterson, A.J., Donnison, M.J., Ravelich, S., Ledgard, A.M., Li, N., Oliver, J.E., Miller, A.L., Tucker, F.C., Breier, B., 2004. Cloned cattle fetuses with the same nuclear genetics are more variable than contemporary half-siblings resulting from artificial insemination and exhibit fetal and placental growth deregulation even in the first trimester. Biol. Reprod. 70 (1), 1-11. Lee, E.G., Kim, J.H., Shin, Y.S., Shin, G.W., Suh, M.D., Kim, D.Y., Kim, Y.H., Kim, G.S., Jung, T.S., 2003. Establishment of a two-dimensional electrophoresis map for Neospora caninum tachyzoites by proteomics. Proteomics 3 (12), 2339-2350. Lee, E.G., Kim, J.H., Shin, Y.S., Shin, G.W., Kim, Y.H., Kim, G.S., Kim, D.Y., Jung, T.S., Suh, M.D., 2004. Two-dimensional gel electrophoresis and immunoblot analysis of Neospora caninum tachyzoites. J. Vet. Sci. 5 ( 2), 139-145. Lee, E.G., Kim, J.H., Shin, Y.S., Shin, G.W., Kim, Y.R., Palaksha, K.J., Kim, D.Y., Yamane, I., Kim, Y.H., Kim, G.S., Suh, M.D., Jung, T.S., 2005. Application of proteomics for comparison of proteome of Neospora caninum and Toxoplasma gondii tachyzoites. J Chromatogr. B Analyt. Technol. Biomed. Life Sci. 815 (1-2), 305-314. Leiser, R., 1975. Development of contact between trophoblast and uterine epithelium during the early stages on implantation in the cow. Zentralbl. Veterinarmed. C. 4 (1), 63-86. Leiser, R., Kaufmann, P., 1994. Placental structure: in a comparative aspect. Exp. Clin. Endocrinol. 102 (03), 122-134. Leiser, R., Krebs, C., Klisch, K., Ebert, B., Dantzer, V., Schuler, G., Hoffmann, B., 1997. Fetal villosity and microvasculature of the bovine placentome in the second half of gestation. The J. Anat. 191 (4), 517-527. Li, S., Gong, P., Tai, L., Li, X., Wang, X., Zhao, C., Zhang, X., Yang, Z., Yang, J., Li, J., Zhang, X., 2018. Extracellular vesicles secreted by Neospora caninum are recognized by Toll-Like Receptor 2 and modulate host cell innate immunity through the MAPK signaling pathway. Front. Immunol. 9, 1633, 10.3389/fimmu.2018.01633. Liao, Y., Zhang, Y., Liu, X., Lu, Y., Zhang, L., Xi, T., Shu, S., Fang, F., 2018. Maternal murine cytomegalovirus infection during pregnancy up-regulates the gene expression of Toll-like Receptor 2 and 4 in placenta. Curr. Med. Sci. 38 (4), 632-639. Liddell, S., Jenkins, M.C., Dubey, J.P., 1999. Vertical transmission of Neospora caninum in BALB/c mice determined by polymerase chain reaction detection. J. Parasitol. 85 (3), 550-555. Lim, D.C., Cooke, B.M., Doerig, C., Saeij, J.P., 2012. Toxoplasma and Plasmodium protein kinases: roles in invasion and host cell remodelling. Int. J. Parasitol. 42 (1), 21-32, 10.1016/j.ijpara.2011.11.007. Lindsay, D.S., Speer, C.A., Toivio-Kinnucan, M.A., Dubey, J.P., Blagburn, B.L., 1993. Use of infected cultured cells to compare ultrastructural features of Neospora caninum from dogs and Toxoplasma gondii. Am. J. Vet. Res. 54 (1), 103- 106. Lindsay, D.S., Rippey, N.S., Toivio-Kinnucan, M.A., Blagburn, B.L., 1995. Ultrastructural effects of diclazuril against Toxoplasma gondii and investigation of a diclazuril-resistant mutant. J. Parasitol. 81 (3), 459-466. Chapter VII ~ References 198 Lindsay, D.S., Dubey, J.P., Duncan, R.B., 1999. Confirmation that the dog is a definitive host for Neospora caninum. Vet. Parasitol. 82 (4), 327-333. Liu, H., Liu, Z., Chao, H., Li, Z., Song, Z., Yang, Y., Peng, J., 2014. High-dose interferon-γ promotes abortion in mice by suppressing Treg and Th17 polarization. J. Interferon Cytokine Res. 34 (5), 394-403. Liu, J., Cao, X., 2016. Cellular and molecular regulation of innate inflammatory responses. Cell. Mol. Immunol. 13 (6), 711. Loch, C., Haeger, J., Pfarrer, C., 2018. IFNτ mediates chemotaxis, motility, metabolism and CK18 downregulation in bovine trophoblast cells in vitro via STAT1 and MAPK42/44 signaling. Placenta 64, 17-26. Long, M.T., Baszler, T.V., Mathison, B.A., 1998. Comparison of intracerebral parasite load, lesion development, and systemic cytokines in mouse strains infected with Neospora caninum. J. Parasitol. 84 (2), 316-320. López-Gatius, F., López-Bejar, M., Murugavel, K., Pabón, M., Ferrer, D., Almería, S., 2004. Neospora-associated abortion episode over a 1-year period in a dairy herd in north-east Spain. J. Vet. Med. B Infect. Dis. Vet. Public Health 51 (7), 348-352. López-Gatius, F., Santolaria, P., Yániz, J.L., Garbayo, J.M., Almería, S., 2005a. The Use of Beef Bull Semen Reduced the Risk of Abortion in Neospora-seropositive Dairy Cows. J. Vet. Med. B Infect. Dis. Vet. Public Health 52 (2), 88-92. López-Gatius, F., García-Ispierto, I., Santolaria, P., Yaniz, J.L., López-Bejar, M., Nogareda, C., Almería, S., 2005b. Relationship between rainfall and Neospora caninum-associated abortion in two dairy herds in a dry environment. J. Vet. Med. B Infect. Dis. Vet. Public Health 52 (3), 147-152. López-Pérez, I.C., Risco-Castillo, V., Collantes-Fernández, E., Ortega-Mora, L.M., 2006. Comparative effect of Neospora caninum infection in BALB/c mice at three different gestation periods. J. Parasitol. 92 (6), 1286-1291. López-Pérez, I.C., Collantes-Fernández, E., Aguado-Martínez, A., Rodríguez-Bertos, A., Ortega-Mora, L.M., 2008. Influence of Neospora caninum infection in BALB/c mice during pregnancy in post-natal development. Vet. Parasitol. 155 (3-4), 175-183. Lunden, A., Marks, J., Maley, S.W., Innes, E.A., 1998. Cellular immune responses in cattle experimentally infected with Neospora caninum. Parasite Immunol. 20 (11), 519-526. Macaldowie, C., Maley, S.W., Wright, S., Bartley, P., Esteban-Redondo, I., Buxton, D., Innes, E.A., 2004. Placental pathology associated with fetal death in cattle inoculated with Neospora caninum by two different routes in early pregnancy. J. Comp. Pathol. 131 (2-3), 142-156. Machado, R.Z., Mineo, T.W., Landim, L.P.,Jr., Carvalho, A.F., Gennari, S.M., Miglino, M.A., 2007. Possible role of bovine trophoblast giant cells in transplacental transmission of Neospora caninum in cattle. Rev. Bras. Parasitol. Vet. 16 (1), 21-25. Mainar-Jaime, R.C., Thurmond, M.C., Berzal-Herranz, B., Hietala, S.K., 1999. Seroprevalence of Neospora caninum and abortion in dairy cows in northern Spain. Vet. Rec. 145 (3), 72-75. Maksimov, P., Hermosilla, C., Kleinertz, S., Hirzmann, J., Taubert, A., 2016. Besnoitia besnoiti infections activate primary bovine endothelial cells and promote PMN adhesion and NET formation under physiological flow condition. Parasitol. Res. 115 (5), 1991-2001. Maley, S.W., Buxton, D., Thomson, K.M., Schriefer, C.E., Innes, E.A., 2001. Serological analysis of calves experimentally infected with Neospora caninum: a 1-year study. Vet. Parasitol. 96 (1), 1-9. Maley, S.W., Buxton, D., Rae, A.G., Wright, S.E., Schock, A., Bartley, P.M., Esteban-Redondo, I., Swales, C., Hamilton, C.M., Sales, J., Innes, E.A., 2003. The pathogenesis of neosporosis in pregnant cattle: inoculation at mid-gestation. J. Comp. Pathol. 129 (2-3), 186-195. Maley, S.W., Buxton, D., Macaldowie, C.N., Anderson, I.E., Wright, S.E., Bartley, P.M., Esteban-Redondo, I., Hamilton, C.M., Storset, A.K., Innes, E.A., 2006. Characterization of the immune response in the placenta of cattle experimentally Infected with Neospora caninum in early gestation. J. Comp. Pathol. 135 (2-3), 130-141. Manicone, A.M., McGuire, J.K., 2008. Matrix metalloproteinases as modulators of inflammation. In: Seminars in cell & developmental biology. Elsevier, pp. 34-41. Marin, M.S., Hecker, Y.P., Quintana, S., Pérez, S., Leunda, M.R., Cantón, G., Cobo, E.R., Moore, D.P., Odeón, A.C., 2017a. Immunization with inactivated antigens of Neospora caninum induces toll-like receptors 3, 7, 8 and 9 in maternal-fetal interface of infected pregnant heifers. Vet. Parasitol. 243, 12-17. Marin, M.S., Hecker, Y.P., Quintana, S., Pérez, S., Leunda, M.R., Cantón, G., Cobo, E.R., Moore, D.P., Odeón, A.C., 2017b. Toll-like receptors 3, 7 and 8 are upregulated in the placental caruncle and fetal spleen of Neospora caninum experimentally infected cattle. Vet. Parasitol. 236, 58-61. Chapter VII ~ References 199 Marks, J., Lunden, A., Harkins, D., Innes, E., 1998. Identification of Neospora antigens recognized by CD4+ T cells and immune sera from experimentally infected cattle. Parasite Immunol. 20 (7), 303-309. Marsh, A.E., Barr, B.C., Madigan, J., Lakritz, J., Nordhausen, R., Conrad, P.A., 1996. Neosporosis as a cause of equine protozoal myeloencephalitis. J. Am. Vet. Med. Assoc. 209 (11), 1907-1913. Marsh, A.E., Barr, B.C., Packham, A.E., Conrad, P.A., 1998. Description of a new Neospora species (Protozoa: Apicomplexa: Sarcocystidae). J. Parasitol. 84 (5), 983-991. Martínez-Colón, G.J., Warheit-Niemi, H., Gurczynski, S.J., Taylor, Q.M., Wilke, C.A., Podsiad, A.B., Crespo, J., Bhan, U., Moore, B.B., 2019. Influenza-induced immune suppression to methicillin-resistant Staphylococcus aureus is mediated by TLR9. PLoS Pathog. 15 (1), e1007560, 10.1371/journal.ppat.1007560. Martino, N., Rizzo, A., Pizzi, F., Dell'Aquila, M., Sciorsci, R., 2015. Effects of kisspeptin-10 on in vitro proliferation and kisspeptin receptor expression in primary epithelial cell cultures isolated from bovine placental cotyledons of fetuses at the first trimester of pregnancy. Theriogenology 83 (6), 978-987. Marugán-Hernández, V., Álvarez-García, G., Risco-Castillo, V., Regidor-Cerrillo, J., Ortega-Mora, L.M., 2010. Identification of Neospora caninum proteins regulated during the differentiation process from tachyzoite to bradyzoite stage by DIGE. Proteomics 10 (9), 1740-1750, 10.1002/pmic.200900664. Marugán-Hernández, V., Álvarez-García, G., Tomley, F., Hemphill, A., Regidor-Cerrillo, J., Ortega-Mora, L.M., 2011a. Identification of novel rhoptry proteins in Neospora caninum by LC/MS-MS analysis of subcellular fractions. J. Proteomics 74 (5), 629-642, 10.1016/j.jprot.2011.02.004. Marugán-Hernández, V., Ortega-Mora, L.M., Aguado-Martínez, A., Jiménez-Ruiz, E., Álvarez-García, G., 2011b. Transgenic Neospora caninum strains constitutively expressing the bradyzoite NcSAG4 protein proved to be safe and conferred significant levels of protection against vertical transmission when used as live vaccines in mice. Vaccine 29 (44), 7867-7874, 10.1016/j.vaccine.2011.07.091. Marugán-Hernández, V., 2017. Neospora caninum and bovine neosporosis: current vaccine research. J. Comp. Pathol. 157 (2-3), 193-200. Matamoros, R.A., Caamano, L., Lamb, S.V., Reimers, T.J., 1994. Estrogen production by bovine binucleate and mononucleate trophoblastic cells in vitro. Biol. Reprod. 51 (3), 486-492. Matzinger, P., 1994. Tolerance, danger, and the extended family. Annu. Rev. Immunol. 12 (1), 991-1045. Mazuz, M.L., Haynes, R., Shkap, V., Fish, L., Wollkomirsky, R., Leibovich, B., Molad, T., Savitsky, I., Golenser, J., 2012. Neospora caninum: In vivo and in vitro treatment with artemisone. Vet. Parasitol. 187 (1-2), 99-104, 10.1016/j.vetpar.2011.12.020. McAllister, M.M., Jolley, W.R., Wills, R.A., Lindsay, D.S., McGuire, A.M., Tranas, J.D., 1998. Oral inoculation of cats with tissue cysts of Neospora caninum. Am. J. Vet. Res. 59 (4), 441-444. McAllister, M.M., Björkman, C., Anderson-Sprecher, R., Rogers, D.G., 2000. Evidence of point-source exposure to Neospora caninum and protective immunity in a herd of beef cows. J. Am. Vet. Med. Assoc. 217 (6), 881-887. McCann, C.M., McAllister, M.M., Gondim, L.F., Smith, R.F., Cripps, P.J., Kipar, A., Williams, D.J., Trees, A.J., 2007. Neospora caninum in cattle: Experimental infection with oocysts can result in exogenous transplacental infection, but not endogenous transplacental infection in the subsequent pregnancy. Int. J. Parasitol. 37 (14), 1631-1639. McCann, C.M., Vyse, A.J., Salmon, R.L., Thomas, D., Williams, D.J., McGarry, J.W., Pebody, R., Trees, A.J., 2008. Lack of serologic evidence of Neospora caninum in humans, England. Emerg. Infect. Dis. 14 (6), 978-980. Melo, M.B., Nguyen, Q.P., Cordeiro, C., Hassan, M.A., Yang, N., McKell, R., Rosowski, E.E., Julien, L., Butty, V., Darde, M.L., Ajzenberg, D., Fitzgerald, K., Young, L.H., Saeij, J.P., 2013. Transcriptional analysis of murine macrophages infected with different Toxoplasma strains identifies novel regulation of host signaling pathways. PLoS Pathog. 9 (12), e1003779, 10.1371/journal.ppat.1003779. Menezes-Souza, D., Guerra-Sá, R., Carneiro, C.M., Vitoriano-Souza, J., Giunchetti, R.C., Teixeira-Carvalho, A., Silveira- Lemos, D., Oliveira, G.C., Corrêa-Oliveira, R., Reis, A.B., 2012. Higher expression of CCL2, CCL4, CCL5, CCL21, and CXCL8 chemokines in the skin associated with parasite density in canine visceral leishmaniasis. PLoS Negl. Trop. Dis. 6 (4), e1566, 10.1371/journal.pntd.0001566. Menzies, M., Ingham, A., 2006. Identification and expression of Toll-like receptors 1–10 in selected bovine and ovine tissues. Vet. Immunol. Immunopathol. 109 (1), 23-30. Miglino, M.A., Pereira, F., Visintin, J., García, J., Meirelles, F., Rumpf, R., Ambrósio, C., Papa, P., Santos, T., Carvalho, A., 2007. Placentation in cloned cattle: structure and microvascular architecture. Theriogenology 68 (4), 604-617. Miller, C.M., Quinn, H.E., Windsor, P.A., Ellis, J.T., 2002. Characterisation of the first Australian isolate of Neospora Chapter VII ~ References 200 caninum from cattle. Aust. Vet. J. 80 (10), 620-625. Milosavljevic, M., Duello, T.M., Schuler, L.A., 1989. In situ localization of two prolactin-related messenger ribonucleic acids to binucleate cells of bovine placentomes. Endocrinology 125 (2), 883-888. Mineo, T.W., Oliveira, C.J., Silva, D.A., Oliveira, L.L., Abatepaulo, A.R., Ribeiro, D.P., Ferreira, B.R., Mineo, J.R., Silva, J.S., 2010. Neospora caninum excreted/secreted antigens trigger CC-chemokine receptor 5-dependent cell migration. Int. J. Parasitol. 40 (7), 797-805, 10.1016/j.ijpara.2009.12.003. Minns, L.A., Menard, L.C., Foureau, D.M., Darche, S., Ronet, C., Mielcarz, D.W., Buzoni-Gatel, D., Kasper, L.H., 2006. TLR9 is required for the gut-associated lymphoid tissue response following oral infection of Toxoplasma gondii. J. Immunol. 176 (12), 7589-7597. Mitsunari, M., Yoshida, S., Shoji, T., Tsukihara, S., Iwabe, T., Harada, T., Terakawa, N., 2006. Macrophage-activating lipopeptide-2 induces cyclooxygenase-2 and prostaglandin E2 via toll-like receptor 2 in human placental trophoblast cells. J. Reprod. Immunol. 72 (1-2), 46-59. Monney, T., Debache, K., Grandgirard, D., Leib, S.L., Hemphill, A., 2012. Vaccination with the recombinant chimeric antigen recNcMIC3-1-R induces a non-protective Th2-type immune response in the pregnant mouse model for Neospora caninum infection. Vaccine 30 (46), 6588-6594, 10.1016/j.vaccine.2012.08.024. Montes, M.J., Tortosa, C.G., Borja, C., Abadia, A.C., González-Gómez, F., Ruiz, C., Olivares, E.G., 1995. Constitutive secretion of interleukin‐6 by human decidual stromal cells in culture. Regulatory effect of progesterone. Am. J. Reprod. Inmunol. 34 (3), 188-194. Moore, D.P., Echaide, I., Verna, A.E., Leunda, M.R., Cano, A., Pereyra, S., Zamorano, P.I., Odeon, A.C., Campero, C.M., 2011. Immune response to Neospora caninum native antigens formulated with immune stimulating complexes in calves. Vet. Parasitol. 175 (3-4), 245-251, 10.1016/j.vetpar.2010.10.020. Moore, D., Álvarez-García, G., Chiapparrone, M., Regidor-Cerrillo, J., Lischinsky, L., De Yaniz, M., Odeón, A., Ortega- Mora, L., Campero, C., 2014. Neospora caninum tachyzoites inoculated by the conjunctival route are not vertically transmitted in pregnant cattle: a descriptive study. Vet. Parasitol. 199 (1-2), 1-7. Mossman, H.W., 1987. Vertebrate Fetal Membranes. Rutgers University Press. Mota, C.M., Oliveira, A., Davoli-Ferreira, M., Silva, M.V., Santiago, F.M., Nadipuram, S.M., Vashisht, A.A., Wohlschlegel, J.A., Bradley, P.J., Silva, J.S., 2016. Neospora caninum activates p38 MAPK as an evasion mechanism against innate immunity. Front. Microbiol. 7, 1456. Müller, J., Hemphill, A., 2011. Identification of a host cell target for the thiazolide class of broad-spectrum anti-parasitic drugs. Exp. Parasitol. 128 (2), 145-150, 10.1016/j.exppara.2011.02.007. Müller, J., Hemphill, A., 2012. In vitro culture systems for the study of apicomplexan parasites in farm animals. Int. J. Parasitol. 43(2), 115-124 , 10.1016/j.ijpara.2012.08.004. Müller, J., Hemphill, A., 2013. New approaches for the identification of drug targets in protozoan parasites. Int. Rev. Cell Mol. Biol. 301, 359-401, 10.1016/B978-0-12-407704-1.00007-5. Müller, J., Aguado-Martinez, A., Manser, V., Balmer, V., Winzer, P., Ritler, D., Hostettler, I., Arranz-Solís, D., Ortega- Mora, L., Hemphill, A., 2015. Buparvaquone is active against Neospora caninum in vitro and in experimentally infected mice. Int. J. Parasitol. Drugs Drug Resist 5 (1), 16-25, 10.1016/j.ijpddr.2015.02.001. Munson, L., Chandler, S.K., Schlafer, D.H., 1988. Long-term culture of bovine trophoblastic cells. J. Tissue Cult. Methods 11 (3), 123-128. Munson, L., Ellington, J.E., Schlafer, D.H., 1991. Bovine trophoblastic cell vesicle attachment to polarized endometrial epithelial cells in vitro. In vitro Cell. Dev. Biol. 27 (1), 31-38. Murai, T., Yamauchi, S., 1986. Erythrophagocytosis by the trophoblast in a bovine placentome. Jpn J Vet Sci 48 (1), 75- 88. Naguleswaran, A., Müller, N., Hemphill, A., 2003. Neospora caninum and Toxoplasma gondii: a novel adhesion/invasion assay reveals distinct differences in tachyzoite-host cell interactions. Exp. Parasitol. 104 (3-4), 149-158. Nakano, H., Shimada, A., Imai, K., Takezawa, T., Takahashi, T., Hashizume, K., 2002. Bovine trophoblastic cell differentiation on collagen substrata: formation of binucleate cells expressing placental lactogen. Cell Tissue Res. 307 (2), 225-235. Nardelli, S.C., Che, F.Y., Silmon de Monerri, N.C., Xiao, H., Nieves, E., Madrid-Aliste, C., Ángel, S.O., Sullivan, W.J.,Jr, Angeletti, R.H., Kim, K., Weiss, L.M., 2013. The histone code of Toxoplasma gondii comprises conserved and unique posttranslational modifications. MBio 4 (6), e00922-13, 10.1128/mBio.00922-13. Nickel, R., Schummer, A., Seiferle, E., 2004. Eingeweide. Lehrbuch Der Anatomie Der Haustiere 2. Georg Thieme Verlag. Chapter VII ~ References 201 Nishikawa, Y., Tragoolpua, K., Inoue, N., Makala, L., Nagasawa, H., Otsuka, H., Mikami, T., 2001. In the absence of endogenous gamma interferon, mice acutely infected with Neospora caninum succumb to a lethal immune response characterized by inactivation of peritoneal macrophages. Clin. Diagn. Lab. Immunol. 8 (4), 811-816. Nishikawa, Y., Inoue, N., Makala, L., Nagasawa, H., 2003. A role for balance of interferon-gamma and interleukin-4 production in protective immunity against Neospora caninum infection. Vet. Parasitol. 116 (3), 175-184. Nishikawa, Y., Shimoda, N., Fereig, R.M., Moritaka, T., Umeda, K., Nishimura, M., Ihara, F., Kobayashi, K., Himori, Y., Suzuki, Y., Furuoka, H., 2018. Neospora caninum dense granule protein 7 regulates the pathogenesis of neosporosis by modulating host immune response. Appl. Environ. Microbiol. 84 (18), e01350-18, 10.1128/AEM.01350-18. Ojo, K.K., Reid, M.C., Kallur Siddaramaiah, L., Müller, J., Winzer, P., Zhang, Z., Keyloun, K.R., Vidadala, R.S., Merritt, E.A., Hol, W.G., Maly, D.J., Fan, E., Van Voorhis, W.C., Hemphill, A., 2014. Neospora caninum calcium-dependent protein kinase 1 is an effective drug target for neosporosis therapy. PLoS One 9 (3), e92929, 10.1371/journal.pone.0092929. Okomo-Adhiambo, M., Beattie, C., Rink, A., 2006. cDNA microarray analysis of host-pathogen interactions in a porcine in vitro model for Toxoplasma gondii infection. Infect. Immun. 74 (7), 4254-4265. Ólafsson, E.B., Varas‐Godoy, M., Barragán, A., 2018. Toxoplasma gondii infection shifts dendritic cells into an amoeboid rapid migration mode encompassing podosome dissolution, secretion of TIMP‐1, and reduced proteolysis of extracellular matrix. Cell. Microbiol. 20 (3), e12808, 10.1111/cmi.12808. Ong, Y., Reese, M.L., Boothroyd, J.C., 2010. Toxoplasma rhoptry protein 16 (ROP16) subverts host function by direct tyrosine phosphorylation of STAT6. J. Biol. Chem. 285 (37), 28731-28740, 10.1074/jbc.M110.112359. Ortega-Mora, L.M., Ferre, I., del Pozo, I., Caetano-da-Silva, A., Collantes-Fernández, E., Regidor-Cerrillo, J., Ugarte- Garagalza, C., Adúriz, G., 2003. Detection of Neospora caninum in semen of bulls. Vet. Parasitol. 117 (4), 301-308. Ortega-Mora, L.M., Fernández-García, A., Gómez-Bautista, M., 2006. Diagnosis of bovine neosporosis: Recent advances and perspectives. Acta Parasitol. 51 (1), 1-14. Ortega-Mora, L.M., Calero-Bernal, R., Regidor-Cerrillo, J., In press. Chapter 32 Neosporosis. In Manual práctico de enfermedades infectocontagiosas en rumiantes. García-Bocanegra, I., 2019. 289-300. ISBN: 978-84-9113-353-7. Osburn, B.I., MacLachlan, N.J., Terrell, T.G., 1982. Ontogeny of the immune system. J. Am. Vet. Med. Assoc. 181 (10), 1049-1052. Osoro, K., Ortega-Mora, L.M., Martínez, A., Serrano-Martínez, E., Ferre, I., 2008. Natural breeding with bulls experimentally infected with Neospora caninum failed to induce seroconversion in dams. Theriogenology 71(4), 639- 642, 10.1016/j.theriogenology.2008.09.035. Ouologuem, D.T., Roos, D.S., 2014. Dynamics of the Toxoplasma gondii inner membrane complex. J. Cell. Sci. 127 (15), 3320-3330, 10.1242/jcs.147736. Pan, Y., Jansen, G.B., Duffield, T.F., Hietala, S., Kelton, D., Lin, C.Y., Peregrine, A.S., 2004. Genetic susceptibility to Neospora caninum infection in Holstein cattle in Ontario. J. Dairy Sci. 87 (11), 3967-3975. Parks, W.C., Wilson, C.L., López-Boado, Y.S., 2004. Matrix metalloproteinases as modulators of inflammation and innate immunity. Nat. Rev. Immunol. 4 (8), 617-629, 10.1038/nri1418. Pastor-Fernández, I., Arranz-Solís, D., Regidor-Cerrillo, J., Álvarez-García, G., Hemphill, A., García-Culebras, A., Cuevas- Martín, C., Ortega-Mora, L.M., 2015. A vaccine formulation combining rhoptry proteins NcROP40 and NcROP2 improves pup survival in a pregnant mouse model of neosporosis. Vet. Parasitol. 207 (3-4), 203-215, 10.1016/j.vetpar.2014.12.009. Pastor-Fernández, I., Regidor-Cerrillo, J., Álvarez-García, G., Marugán-Hernández, V., García-Lunar, P., Hemphill, A., Ortega-Mora, L.M., 2016. The tandemly repeated NTPase (NTPDase) from Neospora caninum is a canonical dense granule protein whose RNA expression, protein secretion and phosphorylation coincides with the tachyzoite egress. Parasit. Vectors 9 (1), 352, 10.1186/s13071-016-1620-4. Peckham, R., Brill, R., Foster, D.S., Bowen, A., Leigh, J.A., Coffey, T.J., Flynn, R.J., 2014. Two distinct populations of bovine IL-17 T-cells can be induced and WC1 IL-17 γδ T-cells are effective killers of protozoan parasites. Sci. Rep. 4, 5431, 10.1038/srep05431. Pedraza-Díaz, S., Marugán-Hernández, V., Collantes-Fernández, E., Regidor-Cerrillo, J., Rojo-Montejo, S., Gómez- Bautista, M., Ortega-Mora, L.M., 2009. Microsatellite markers for the molecular characterization of Neospora caninum: application to clinical samples. Vet. Parasitol. 166 (1-2), 38-46, 10.1016/j.vetpar.2009.07.043. Pereira García-Melo, D., Regidor-Cerrillo, J., Collantes-Fernández, E., Aguado-Martínez, A., Del Pozo, I., Minguijón, E., Gómez-Bautista, M., Adúriz, G., Ortega-Mora, L.M., 2010. Pathogenic characterization in mice of Neospora caninum isolates obtained from asymptomatic calves. Parasitology 137(7),1057-1068, 10.1017/S0031182009991855. Chapter VII ~ References 202 Pereira-Bueno, J., Quintanilla-Gozalo, A., Pérez-Pérez, V., Espi-Felgueroso, A., Álvarez-García, G., Collantes-Fernández, E., Ortega-Mora, L.M., 2003. Evaluation by different diagnostic techniques of bovine abortion associated with Neospora caninum in Spain. Vet. Parasitol. 111 (2-3), 143-152. Pérez-Zaballos, F.J., Ortega-Mora, L.M., Álvarez-García, G., Collantes-Fernández, E., Navarro-Lozano, V., García-Villada, L., Costas, E., 2005. Adaptation of Neospora caninum isolates to cell-culture changes: an argument in favor of its clonal population structure. J. Parasitol. 91 (3), 507-510. Peter, A.T., 2013. Bovine placenta: a review on morphology, components, and defects from terminology and clinical perspectives. Theriogenology 80 (7), 693-705. Peters, M., Lutkefels, E., Heckeroth, A.R., Schares, G., 2001. Immunohistochemical and ultrastructural evidence for Neospora caninum tissue cysts in skeletal muscles of naturally infected dogs and cattle. Int. J. Parasitol. 31 (10), 1144- 1148. Pfaff, A.W., Georges, S., Abou-Bacar, A., Letscher-Bru, V., Klein, J.P., Mousli, M., Candolfi, E., 2005. Toxoplasma gondii regulates ICAM-1 mediated monocyte adhesion to trophoblasts. Immunol. Cell Biol. 83 (5), 483-489. Pfarrer, C., Hirsch, P., Guillomot, M., Leiser, R., 2003. Interaction of integrin receptors with extracellular matrix is involved in trophoblast giant cell migration in bovine placentomes. Placenta 24 (6), 588-597. Pfarrer, C., Weise, S., Berisha, B., Schams, D., Leiser, R., Hoffmann, B., Schuler, G., 2006a. Fibroblast growth factor (FGF)- 1, FGF2, FGF7 and FGF receptors are uniformly expressed in trophoblast giant cells during restricted trophoblast invasion in cows. Placenta 27 (6-7), 758-770. Pfarrer, C., Ruziwa, S., Winther, H., Callesen, H., Leiser, R., Schams, D., Dantzer, V., 2006b. Localization of vascular endothelial growth factor (VEGF) and its receptors VEGFR-1 and VEGFR-2 in bovine placentomes from implantation until term. Placenta 27 (8), 889-898. Pillai, V.V., Siqueira, L.G., Das, M., Kei, T.G., Tu, L.N., Herren, A.W., Phinney, B.S., Cheong, S.H., Hansen, P.J., Selvaraj, V., 2019. Physiological profile of undifferentiated bovine blastocyst-derived trophoblasts. Biol. Open 8 (5), 10.1242/bio.037937. Pinheiro, A.M., Costa, S.L., Freire, S.M., Ribeiro, C.S., Tardy, M., El-Bacha, R.S., Costa, M.F., 2010. Neospora caninum: Early immune response of rat mixed glial cultures after tachyzoites infection. Exp. Parasitol. 124(4):442-447, 10.1016/j.exppara.2009.12.018. Pittman, K.J., Aliota, M.T., Knoll, L.J., 2014. Dual transcriptional profiling of mice and Toxoplasma gondii during acute and chronic infection. BMC Genomics 15, 806, 10.1186/1471-2164-15-806 . Plattner, F., Soldati-Favre, D., 2008. Hijacking of host cellular functions by the Apicomplexa. Annu. Rev. Microbiol. 62, 471-487. Pollo-Oliveira, L., Post, H., Acencio, M.L., Lemke, N., van den Toorn, H., Tragante, V., Heck, A.J., Altelaar, A.F., Yatsuda, A.P., 2013. Unravelling the Neospora caninum secretome through the secreted fraction (ESA) and quantification of the discharged tachyzoite using high-resolution mass spectrometry-based proteomics. Parasit. Vectors 6 (1), 335, 10.1186/1756-3305-6-335. Puech, C., Dedieu, L., Chantal, I., Rodrigues, V., 2015. Design and evaluation of a unique SYBR Green real-time RT-PCR assay for quantification of five major cytokines in cattle, sheep and goats. BMC Vet. Res. 11, 65, 10.1186/s12917-015- 0382-0. Quinn, H.E., Miller, C.M., Ryce, C., Windsor, P.A., Ellis, J.T., 2002a. Characterization of an outbred pregnant mouse model of Neospora caninum infection. J. Parasitol. 88 (4), 691-696. Quinn, H.E., Ellis, J.T., Smith, N.C., 2002b. Neospora caninum: a cause of immune-mediated failure of pregnancy? Trends Parasitol. 18 (9), 391-394. Quintanilla-Gozalo, A., Pereira-Bueno, J., Tabares, E., Innes, E., González-Paniello, R., Ortega-Mora, L., 1999. Seroprevalence of Neospora caninum infection in dairy and beef cattle in Spain. Int. J. Parasitol. 29 (8), 1201-1208. Quintanilla-Gozalo, A., Pereira-Bueno, J., Seijas-Carballedo, A., Costas, E., Ortega Mora, L.M., 2000. Observational studies in Neospora caninum infected dairy cattle: relationship infection-abortion and gestational antibody fluctuations. In Hemphill A., Gottstein B. A European perspective on Neospora caninum. Int. J. Parasitol. 30, 877-924. Radke, J.R., Behnke, M.S., Mackey, A.J., Radke, J.B., Roos, D.S., White, M.W., 2005. The transcriptome of Toxoplasma gondii. BMC Biol. 3, 26. Raghupathy, R., 1997. Th1-type immunity is incompatible with successful pregnancy. Immunol. Today 18 (10), 478-482. Ramamoorthy, S., Lindsay, D.S., Schurig, G.G., Boyle, S.M., Duncan, R.B., Vemulapalli, R., Sriranganathan, N., 2006. Vaccination with gamma-irradiated Neospora caninum tachyzoites protects mice against acute challenge with N. Chapter VII ~ References 203 caninum. J. Eukaryot. Microbiol. 53 (2), 151-156. Ramaprasad, A., Mourier, T., Naeem, R., Malas, T.B., Moussa, E., Panigrahi, A., Vermont, S.J., Otto, T.D., Wastling, J., Pain, A., 2015. Comprehensive evaluation of Toxoplasma gondii VEG and Neospora caninum LIV genomes with tachyzoite stage transcriptome and proteome defines novel transcript features. PLoS One 10 (4), e0124473, 10.1371/journal.pone.0124473. Ramos-Ibeas, P., Calle, A., Pericuesta, E., Laguna-Barraza, R., Moros-Mora, R., Lopera-Vásquez, R., Maillo, V., Yáñez-Mó, M., Gutiérrez-Adán, A., Rizos, D., 2014. An efficient system to establish biopsy-derived trophoblastic cell lines from bovine embryos. Biol. Reprod. 91 (1), 15, 10.1095/biolreprod.114.118430. Reese, M.L., Boothroyd, J.C., 2011. A conserved non-canonical motif in the pseudoactive site of the ROP5 pseudokinase domain mediates its effect on Toxoplasma virulence. J. Biol. Chem. 286 (33), 29366-29375, 10.1074/jbc.M111.253435. Regidor-Cerrillo, J., Pedraza-Díaz, S., Gómez-Bautista, M., Ortega-Mora, L.M., 2006. Multilocus microsatellite analysis reveals extensive genetic diversity in Neospora caninum. J. Parasitol. 92 (3), 517-524. Regidor-Cerrillo, J., Gómez-Bautista, M., Pereira-Bueno, J., Adúriz, G., Navarro-Lozano, V., Risco-Castillo, V., Fernández- García, A., Pedraza-Díaz, S., Ortega-Mora, L.M., 2008. Isolation and genetic characterization of Neospora caninum from asymptomatic calves in Spain. Parasitology 135 (14), 1651-1659. Regidor-Cerrillo, J., Gómez-Bautista, M., Del Pozo, I., Jiménez-Ruiz, E., Adúriz, G., Ortega-Mora, L.M., 2010. Influence of Neospora caninum intra-specific variability in the outcome of infection in a pregnant BALB/c mouse model. Vet. Res. 41 (4), 52, 10.1051/vetres/2010024. Regidor-Cerrillo, J., Gómez-Bautista, M., Sodupe, I., Adúriz, G., Álvarez-García, G., Del Pozo, I., Ortega-Mora, L.M., 2011. In vitro invasion efficiency and intracellular proliferation rate comprise virulence-related phenotypic traits of Neospora caninum. Vet. Res. 42 (1), 41, 10.1186/1297-9716-42-41. Regidor-Cerrillo, J., Álvarez-García, G., Pastor-Fernández, I., Marugán-Hernández, V., Gómez-Bautista, M., Ortega- Mora, L.M., 2012. Proteome expression changes among virulent and attenuated Neospora caninum isolates. J. Proteomics 75 (8), 2306-2318, 10.1016/j.jprot.2012.01.039. Regidor-Cerrillo, J., Díez-Fuertes, F., García-Culebras, A., Moore, D.P., González-Warleta, M., Cuevas, C., Schares, G., Katzer, F., Pedraza-Díaz, S., Mezo, M., Ortega-Mora, L.M., 2013. Genetic diversity and geographic population structure of bovine Neospora caninum determined by microsatellite genotyping analysis. PLoS One 8 (8), e72678, 10.1371/journal.pone.0072678. Regidor-Cerrillo, J., Arranz-Solís, D., Benavides, J., Gómez-Bautista, M., Castro-Hermida, J.A., Mezo, M., Pérez, V., Ortega-Mora, L.M., González-Warleta, M., 2014. Neospora caninum infection during early pregnancy in cattle: how the isolate influences infection dynamics, clinical outcome and peripheral and local immune responses. Vet. Res. 45, 10, 10.1186/1297-9716-45-10. Regidor-Cerrillo, J., García-Lunar, P., Pastor-Fernández, I., Álvarez-García, G., Collantes-Fernández, E., Gómez-Bautista, M., Ortega-Mora, L.M., 2015. Neospora caninum tachyzoite immunome study reveals differences among three biologically different isolates. Vet. Parasitol. 212 (3-4), 92-99, 10.1016/j.vetpar.2015.08.020. Reichel, M.P., Ellis, J.T., 2009. Neospora caninum--how close are we to development of an efficacious vaccine that prevents abortion in cattle? Int. J. Parasitol. 39 (11), 1173-1187, 10.1016/j.ijpara.2009.05.007. Reichel, M.P., Ayanegui-Alcérreca, M.A., Gondim, L.F.P., Ellis, J.T., 2013. What is the global economic impact of Neospora caninum in cattle – The billion dollar question. Int. J. Parasitol. 43 (2), 133-142, 10.1016/j.ijpara.2012.10.022. Reichel, M.P., Moore, D.P., Hemphill, A., Ortega-Mora, L.M., Dubey, J.P., Ellis, J.T., 2015. A live vaccine against Neospora caninum abortions in cattle. Vaccine 33 (11), 1299-1301, 10.1016/j.vaccine.2015.01.064. Reid, A.J., Vermont, S.J., Cotton, J.A., Harris, D., Hill-Cawthorne, G.A., Konen-Waisman, S., Latham, S.M., Mourier, T., Norton, R., Quail, M.A., Sanders, M., Shanmugam, D., Sohal, A., Wasmuth, J.D., Brunk, B., Grigg, M.E., Howard, J.C., Parkinson, J., Roos, D.S., Trees, A.J., Berriman, M., Pain, A., Wastling, J.M., 2012. Comparative genomics of the apicomplexan parasites Toxoplasma gondii and Neospora caninum: Coccidia differing in host range and transmission strategy. PLoS Pathog. 8 (3), e1002567, 10.1371/journal.ppat.1002567. Reid, A.J., 2015. Large, rapidly evolving gene families are at the forefront of host-parasite interactions in Apicomplexa. Parasitology 142 (Suppl 1), S57-70, 10.1017/S0031182014001528. Reimers, T.J., Ullmann, M.B., Hansel, W., 1985. Progesterone and prostanoid production by bovine binucleate trophoblastic cells. Biol. Reprod. 33 (5), 1227-1236. Reynolds, L., Millaway, D., Kirsch, J., Infeld, J., Redmer, D., 1990. Growth and in vitro metabolism of placental tissues of cows from day 100 to day 250 of gestation. J. Reprod. Fertil. 89 (1), 213-222. Risco-Castillo, V., Fernández-García, A., Ortega-Mora, L.M., 2004. Comparative analysis of stress agents in a simplified Chapter VII ~ References 204 in vitro system of Neospora caninum bradyzoite production. J. Parasitol. 90 (3), 466-470. Ritter, D.M., Kerlin, R., Sibert, G., Brake, D., 2002. Immune factors influencing the course of infection with Neospora caninum in the murine host. J. Parasitol. 88 (2), 271-280. Robbins, J.R., Zeldovich, V.B., Poukchanski, A., Boothroyd, J.C., Bakardjiev, A.I., 2012. Tissue barriers of the human placenta to infection with Toxoplasma gondii. Infect. Immun. 80 (1), 418-428, 10.1128/IAI.05899-11. Roberts, C., Alexander, J., 1992. Studies on a murine model of congenital toxoplasmosis: vertical disease transmission only occurs in BALB/c mice infected for the first time during pregnancy. Parasitology 104 (1), 19-23. Roberts, M.S., 2007. Dermal Absorption and Toxicity Assessment. CRC Press. Roberts, R.M., Cross, J.C., Leaman, D.W., 1992. Interferons as hormones of pregnancy. Endocr. Rev. 13 (3), 432-452. Robertson, S.A., 2000. Control of the immunological environment of the uterus. Rev. Reprod. 5 (3), 164-174. Roberts-Thomson, I.C., Fon, J., Uylaki, W., Cummins, A.G., Barry, S., 2011. Cells, cytokines and inflammatory bowel disease: a clinical perspective. Expert Rev. Gastroenterol. Hepatol. 5 (6), 703-716, 10.1586/egh.11.74. Rocchi, M.S., Bartley, P.M., Inglis, N.F., Collantes-Fernández, E., Entrican, G., Katzer, F., Innes, E.A., 2011. Selection of Neospora caninum antigens stimulating bovine CD4+ve T cell responses through immuno-potency screening and proteomic approaches. Vet. Res. 42 (1), 91, 10.1186/1297-9716-42-91. Rodríguez, T.A., Sparrow, D.B., Scott, A.N., Withington, S.L., Preis, J.I., Michalicek, J., Clements, M., Tsang, T.E., Shioda, T., Beddington, R.S., Dunwoodie, S.L., 2004. Cited1 is required in trophoblasts for placental development and for embryo growth and survival. Mol. Cell. Biol. 24 (1), 228-244, 10.1128/mcb.24.1.228-244.2004. Rojo-Montejo, S., Collantes-Fernández, E., Regidor-Cerrillo, J., Álvarez-García, G., Marugán-Hernández, V., Pedraza- Díaz, S., Blanco-Murcia, J., Prenafeta, A., Ortega-Mora, L.M., 2009a. Isolation and characterization of a bovine isolate of Neospora caninum with low virulence. Vet. Parasitol. 159 (1), 7-16. Rojo-Montejo, S., Collantes-Fernández, E., Blanco-Murcia, J., Rodríguez-Bertos, A., Risco-Castillo, V., Ortega-Mora, L.M., 2009b. Experimental infection with a low virulence isolate of Neospora caninum at 70 days gestation in cattle did not result in foetopathy. Vet. Res. 40 (5), 49, 10.1051/vetres/2009032. Rojo-Montejo, S., Zabala-Martínez, J., Vázquez-Moreno, E., Collantes-Fernández, E., Loste-Montoya, J.M., Ortega- Mora, L.M., 2009c. Relevancia de la transmisión horizontal de Neospora caninum en rebaños bovinos españoles. XIV Congreso Internacional ANEMBE de Medicina Bovina May 6-8th, La Coruña (Spain) (oral communication). Rojo-Montejo, S., Collantes-Fernández, E., Rodríguez-Marcos, S., Pérez-Zaballos, F., López-Pérez, I., Prenafeta, A., Ortega-Mora, L.M., 2011. Comparative efficacy of immunization with inactivated whole tachyzoites versus a tachyzoite- bradyzoite mixture against neosporosis in mice. Parasitology 138 (11), 1372-1383, 10.1017/S0031182011001156. Rojo-Montejo, S., Collantes-Fernández, E., López-Pérez, I., Risco-Castillo, V., Prenafeta, A., Ortega-Mora, L.M., 2012. Evaluation of the protection conferred by a naturally attenuated Neospora caninum isolate against congenital and cerebral neosporosis in mice. Vet. Res. 43 (1), 62, 10.1186/1297-9716-43-62. Rojo-Montejo, S., Collantes-Fernández, E., Pérez-Zaballos, F., Rodríguez-Marcos, S., Blanco-Murcia, J., Rodríguez- Bertos, A., Prenafeta, A., Ortega-Mora, L.M., 2013. Effect of vaccination of cattle with the low virulence Nc-Spain 1H isolate of Neospora caninum against a heterologous challenge in early and mid-gestation. Vet. Res. 44, 106, 10.1186/1297-9716-44-106. Rosbottom, A., Guy, C.S., Gibney, E.H., Smith, R.F., Valarcher, J.F., Taylor, G., Williams, D.J., 2007. Peripheral immune responses in pregnant cattle following Neospora caninum infection. Parasite Immunol. 29 (4), 219-228. Rosbottom, A., Gibney, E.H., Guy, C.S., Kipar, A., Smith, R.F., Kaiser, P., Trees, A.J., Williams, D.J., 2008. Upregulation of cytokines is detected in the placentas of cattle infected with Neospora caninum and is more marked early in gestation when fetal death is observed. Infect. Immun. 76 (6), 2352-2361. Rosbottom, A., Gibney, H., Kaiser, P., Hartley, C., Smith, R.F., Robinson, R., Kipar, A., Williams, D.J., 2011. Up regulation of the maternal immune response in the placenta of cattle naturally infected with Neospora caninum. PLoS One 6 (1), e15799, 10.1371/journal.pone.0015799. Rosowski, E.E., Lu, D., Julien, L., Rodda, L., Gaiser, R.A., Jensen, K.D., Saeij, J.P., 2011. Strain-specific activation of the NF-kappaB pathway by GRA15, a novel Toxoplasma gondii dense granule protein. J. Exp. Med. 208 (1), 195-212, 10.1084/jem.20100717. Russell, W.M.S., Burch, R.L., Hume, C.W., 1959. The Principles of Humane Experimental Technique. Methuen London. Saeij, J.P., Boyle, J.P., Boothroyd, J.C., 2005. Differences among the three major strains of Toxoplasma gondii and their specific interactions with the infected host. Trends Parasitol. 21 (10), 476-481. Chapter VII ~ References 205 Saeij, J.P., Boyle, J.P., Coller, S., Taylor, S., Sibley, L.D., Brooke-Powell, E.T., Ajioka, J.W., Boothroyd, J.C., 2006. Polymorphic secreted kinases are key virulence factors in toxoplasmosis. Science 314 (5806), 1780-1783, 10.1126/science.1133690. Sánchez-Sánchez, R., Vázquez, P., Ferre, I., Ortega-Mora, L.M., 2018. Treatment of toxoplasmosis and neosporosis in farm ruminants: state of knowledge and future trends. Curr. Top. Med. Chem. 18(15), 1304-1323, 10.2174/1568026618666181002113617. Santos, J.M., Lebrun, M., Daher, W., Soldati, D., Dubremetz, J., 2009. Apicomplexan cytoskeleton and motors: key regulators in morphogenesis, cell division, transport and motility. Int. J. Parasitol. 39 (2), 153-162. Schares, G., Pantchev, N., Barutzki, D., Heydorn, A.O., Bauer, C., Conraths, F.J., 2005. Oocysts of Neospora caninum, Hammondia heydorni, Toxoplasma gondii and Hammondia hammondi in faeces collected from dogs in Germany. Int. J. Parasitol. 35 (14), 1525-1537. Schlafer, D., Fisher, P., Davies, C., 2000. The bovine placenta before and after birth: placental development and function in health and disease. Anim. Reprod. Sci. 60, 145-160. Schmidt, S., Gerber, D., Soley, J.T., Aire, T.A., Boos, A., 2005. Histo-morphology of the uterus and early placenta of the African buffalo (Syncerus caffer) and comparative placentome morphology of the African buffalo and cattle (Bos taurus). Placenta 27 (8), 899-911. Schmittgen, T.D., Livak, K.J., 2008. Analyzing real-time PCR data by the comparative C(T) method. Nat. Protoc. 3 (6), 1101-1108. Schock, A., Innes, E.A., Yamane, I., Latham, S.M., Wastling, J.M., 2001. Genetic and biological diversity among isolates of Neospora caninum. Parasitology 123, 13-23. Schorer, M., Debache, K., Barna, F., Monney, T., Müller, J., Boykin, D.W., Stephens, C.E., Hemphill, A., 2012. Di-cationic arylimidamides act against Neospora caninum tachyzoites by interference in membrane structure and nucleolar integrity and are active against challenge infection in mice. Int. J. Parasitol. Drugs Drug. Resist. 2, 109-120, 10.1016/j.ijpddr.2012.03.001. Seipel, D., Oliveira, B.C., Resende, T.L., Schuindt, S.H., Pimentel, P.M., Kanashiro, M.M., Arnholdt, A.C., 2010. Toxoplasma gondii infection positively modulates the macrophages migratory molecular complex by increasing matrix metalloproteinases, CD44 and alpha v beta 3 integrin. Vet. Parasitol. 169 (3-4), 312-319, 10.1016/j.vetpar.2009.12.042. Serrano, E., Ferre, I., Osoro, K., Adúriz, G., Mateos-Sanz, A., Martínez, A., Atxaerandio, R., Hidalgo, C.O., Ortega-Mora, L.M., 2006. Intrauterine Neospora caninum inoculation of heifers. Vet. Parasitol. 135 (3-4), 197-203. Serrano-Martínez, E., Ferre, I., Osoro, K., Adúriz, G., Mota, R.A., Martínez, A., Del-Pozo, I., Hidalgo, C.O., Ortega-Mora, L.M., 2007a. Intrauterine Neospora caninum inoculation of heifers and cows using contaminated semen with different numbers of tachyzoites. Theriogenology 67 (4), 729-737. Serrano-Martínez, E., Ferre, I., Martínez, A., Osoro, K., Mateos-Sanz, A., Del-Pozo, I., Adúriz, G., Tamargo, C., Hidalgo, C.O., Ortega-Mora, L.M., 2007b. Experimental neosporosis in bulls: Parasite detection in semen and blood and specific antibody and interferon-gamma responses. Theriogenology 67 (6), 1175-1184. Sheiner, L., Vaidya, A.B., McFadden, G.I., 2013. The metabolic roles of the endosymbiotic organelles of Toxoplasma and Plasmodium spp. Curr. Opin. Microbiol. 16 (4), 452-458, 10.1016/j.mib.2013.07.003. Sheldon, I.M., Dobson, H., 2004. Postpartum uterine health in cattle. Anim. Reprod. Sci. 82-83, 295-306, 10.1016/j.anireprosci.2004.04.006. Shimada, A., Nakano, H., Takahashi, T., Imai, K., Hashizume, K., 2001. Isolation and characterization of a bovine blastocyst-derived trophoblastic cell line, BT-1: development of a culture system in the absence of feeder cell. Placenta 22 (7), 652-662. Shin, Y.S., Lee, E.G., Jung, T.S., 2005a. Exploration of immunoblot profiles of Neospora caninum probed with different bovine immunoglobulin classes. J Vet Sci. 6 (2), 157-160. Shin, Y.S., Shin, G.W., Kim, Y.R., Lee, E.Y., Yang, H.H., Palaksha, K.J., Youn, H.J., Kim, J.H., Kim, D.Y., Marsh, A.E., Lakritz, J., Jung, T.S., 2005b. Comparison of proteome and antigenic proteome between two Neospora caninum isolates. Vet. Parasitol. 134 (1-2), 41-52. Silva, L.M., Vila-Viçosa, M.J., Cortes, H.C., Taubert, A., Hermosilla, C., 2015. Suitable in vitro Eimeria arloingi macromeront formation in host endothelial cells and modulation of adhesion molecule, cytokine and chemokine gene transcription. Parasitol. Res. 114 (1), 113-124. Skariah, S., McIntyre, M.K., Mordue, D.G., 2010. Toxoplasma gondii: determinants of tachyzoite to bradyzoite conversion. Parasitol. Res. 107 (2), 253-260, 10.1007/s00436-010-1899-6. Chapter VII ~ References 206 Smith, J.E., 1995. A ubiquitous intracellular parasite: the cellular biology of Toxoplasma gondii Int. J. Parasitol. 25 (11), 1301-1309. Sobotta, K., Bonkowski, K., Liebler-Tenorio, E., Germon, P., Rainard, P., Hambruch, N., Pfarrer, C., Jacobsen, I.D., Menge, C., 2017. Permissiveness of bovine epithelial cells from lung, intestine, placenta and udder for infection with Coxiella burnetii. Vet. Res. 48 (1), 23. Sohn, C.S., Cheng, T.T., Drummond, M.L., Peng, E.D., Vermont, S.J., Xia, D., Cheng, S.J., Wastling, J.M., Bradley, P.J., 2011. Identification of novel proteins in Neospora caninum using an organelle purification and monoclonal antibody approach. PLoS One 6 (4), e18383, 10.1371/journal.pone.0018383. Sorokin, L., 2010. The impact of the extracellular matrix on inflammation. Nat. Rev.Immunol. 10 (10), 712-723, 10.1038/nri2852. Speer, C.A., Dubey, J.P., 1989. Ultrastructure of tachyzoites, bradyzoites and tissue cysts of Neospora caninum. J. Protozool. 36 (5), 458-463. Speer, C.A., Dubey, J.P., McAllister, M.M., Blixt, J.A., 1999. Comparative ultrastructure of tachyzoites, bradyzoites, and tissue cysts of Neospora caninum and Toxoplasma gondii. Int. J. Parasitol. 29 (10), 1509-1519. Spencer, T.E., Johnson, G.A., Bazer, F.W., Burghardt, R.C., 2007. Fetal-maternal interactions during the establishment of pregnancy in ruminants. Soc. Reprod. Fertil. Suppl. 64, 379-396. Spencer, T.E., Sandra, O., Wolf, E., 2008. Genes involved in conceptus-endometrial interactions in ruminants: insights from reductionism and thoughts on holistic approaches. J. Reprod. Fertil. 135 (2), 165-179, 10.1530/REP-07-0327. Staska, L.M., McGuire, T.C., Davies, C.J., Lewin, H.A., Baszler, T.V., 2003. Neospora caninum-infected cattle develop parasite-specific CD4+ cytotoxic T lymphocytes. Infect. Immun. 71 (6), 3272-3279. Steinborn, A., Von Gall, C., Hildenbrand, R., Stutte, H., Kaufmann, M., 1998a. Identification of placental cytokine- producing cells in term and preterm labor. Obstet. Gynecol. 91 (3), 329-335. Steinborn, A., Geisse, M., Kaufmann, M., 1998b. Expression of cytokine receptors in the placenta in term and preterm labour. Placenta 19 (2), 165-170. Stenlund, S., Björkman, C., Holmdahl, O.J., Kindahl, H., Uggla, A., 1997. Characterization of a Swedish bovine isolate of Neospora caninum. Parasitol. Res. 83 (3), 214-219. Strahl, H., 1906. Die Embryonalhüllen der Säuger und die Placenta. In Hertwig’s Handbuch Der Vergleichenden Und Experimentellen Entwicklungslehre Der Wirbeltiere. G. Fischer Jena, pp. 235-368. Stringfellow, D., Gray, B., Lauerman, L., Thomson, M., Rhodes, P., Bird, R., 1987. Monolayer culture of cells originating from a preimplantation bovine embryo. In vitro Cell. Dev. Biol. 23 (11), 750-754. Strohbusch, M., Müller, N., Hemphill, A., Krebber, R., Greif, G., Gottstein, B., 2009. Toltrazuril treatment of congenitally acquired Neospora caninum infection in newborn mice. Parasitol. Res. 104(6), 1335-1343, 10.1007/s00436-009-1328- x. Supasorn, O., Sringkarin, N., Srimanote, P., Angkasekwinai, P., 2016. Matrix metalloproteinases contribute to the regulation of chemokine expression and pulmonary inflammation in Cryptococcus infection. Clin. Exp. Immunol. 183 (3), 431-440, 10.1111/cei.12725. Suzuki, Y., Koshi, K., Imai, K., Takahashi, T., Kizaki, K., Hashizume, K., 2011. Bone morphogenetic protein 4 accelerates the establishment of bovine trophoblastic cell lines. J. Reprod. Fertil. 142 (5), 733-743. Taipale, J., Keski-Oja, J., 1997. Growth factors in the extracellular matrix. FASEB J. 11 (1), 51-59, 10.1096/fasebj.11.1.9034166. Talbot, N.C., Caperna, T.J., Edwards, J.L., Garrett, W., Wells, K.D., Ealy, A.D., 2000. Bovine blastocyst-derived trophectoderm and endoderm cell cultures: interferon tau and transferrin expression as respective in vitro markers. Biol. Reprod. 62 (2), 235-247. Talevich, E., Kannan, N., 2013. Structural and evolutionary adaptation of rhoptry kinases and pseudokinases, a family of coccidian virulence factors. BMC Evol. Biol. 13, 117, 10.1186/1471-2148-13-117. Tanaka, T., Hamada, T., Inoue, N., Nagasawa, H., Fujisaki, K., Suzuki, N., Mikami, T., 2000. The role of CD4(+) or CD8(+) T cells in the protective immune response of BALB/c mice to Neospora caninum infection. Vet. Parasitol. 90 (3), 183- 191. Tanaka, S., Nishimura, M., Ihara, F., Yamagishi, J., Suzuki, Y., Nishikawa, Y., 2013. Transcriptome analysis of mouse brain infected with Toxoplasma gondii. Infect. Immun. 81 (10), 3609-3619, 10.1128/IAI.00439-13. Taubert, A., Krull, M., Zahner, H., Hermosilla, C., 2006. Toxoplasma gondii and Neospora caninum infections of bovine Chapter VII ~ References 207 endothelial cells induce endothelial adhesion molecule gene transcription and subsequent PMN adhesion. Vet. Immunol. Immunopathol. 112 (3-4), 272-283. Taylor, S., Barragán, A., Su, C., Fux, B., Fentress, S.J., Tang, K., Beatty, W.L., Hajj, H.E., Jerome, M., Behnke, M.S., White, M., Wootton, J.C., Sibley, L.D., 2006. A secreted serine-threonine kinase determines virulence in the eukaryotic pathogen Toxoplasma gondii. Science 314 (5806), 1776-1780, 10.1126/science.1133643. Tenter, A.M., Barta, J.R., Beveridge, I., Duszynski, D.W., Mehlhorn, H., Morrison, D.A., Thompson, R.C., Conrad, P.A., 2002. The conceptual basis for a new classification of the coccidia. Int. J. Parasitol. 32 (5), 595-616. Thilsted, J.P., Dubey, J.P., 1989. Neosporosis-like abortions in a herd of dairy cattle. J. Vet. Diagn. Invest. 1 (3), 205-209. Tomavo, S., 2001. The differential expression of multiple isoenzyme forms during stage conversion of Toxoplasma gondii: an adaptive developmental strategy. Int. J. Parasitol. 31 (10), 1023-1031. Trees, A.J., Davison, H.C., Innes, E.A., Wastling, J.M., 1999. Towards evaluating the economic impact of bovine neosporosis. Int. J. Parasitol. 29 (8), 1195-1200. Trees, A.J., McAllister, M.M., Guy, C.S., McGarry, J.W., Smith, R.F., Williams, D.J.L., 2002. Neospora caninum: oocyst challenge of pregnant cows. Vet. Parasitol. 109 (1–2), 147-154, 10.1016/S0304-4017(02)00234-0. Trees, A.J., Williams, D.J.L., 2005. Endogenous and exogenous transplacental infection in Neospora caninum and Toxoplasma gondii Trends Parasitol. 21 (12), 558-561. Tuo, W., Fetterer, R.H., Davis, W.C., Jenkins, M.C., Dubey, J.P., 2005. Neospora caninum antigens defined by antigen- dependent bovine CD4+ T cells. J. Parasitol. 91 (3), 564-568. Ueno, N., Harker, K.S., Clarke, E.V., McWhorter, F.Y., Liu, W.F., Tenner, A.J., Lodoen, M.B., 2014. Real‐time imaging of Toxoplasma‐infected human monocytes under fluidic shear stress reveals rapid translocation of intracellular parasites across endothelial barriers. Cell. Microbiol. 16 (4), 580-595. Ullmann, M.B., Reimers, T.J., 1989. Progesterone production by binucleate trophoblastic cells of cows. J. Reprod. Fertil. Suppl. 37, 173-179. Van Lint, P., Libert, C., 2007. Chemokine and cytokine processing by matrix metalloproteinases and its effect on leukocyte migration and inflammation. J. Leukoc. Biol. 82 (6), 1375-1381. Vanleeuwen, J.A., Greenwood, S., Clark, F., Acorn, A., Markham, F., McCarron, J., O'Handley, R., 2011. Monensin use against Neospora caninum challenge in dairy cattle. Vet. Parasitol. 175 (3-4), 372-376, 10.1016/j.vetpar.2010.10.016. Vanselow, J., Fürbass, R., Tiemann, U., 2008. Cultured bovine trophoblast cells differentially express genes encoding key steroid synthesis enzymes. Placenta 29 (6), 531-538. Villegas, V.E., Zaphiropoulos, P.G., 2015. Neighboring gene regulation by antisense long non-coding RNAs. Int. J. Mol. Sci. 16 (2), 3251-3266, 10.3390/ijms16023251. Walsh, C.P., Vemulapalli, R., Sriranganathan, N., Zajac, A.M., Jenkins, M.C., Lindsay, D.S., 2001. Molecular comparison of the dense granule proteins GRA6 and GRA7 of Neospora hughesi and Neospora caninum. Int. J. Parasitol. 31 (3), 253- 258. Walter, I., Boos, A., 2001. Matrix metalloproteinases (MMP-2 and MMP-9) and tissue inhibitor-2 of matrix metalloproteinases (TIMP-2) in the placenta and interplacental uterine wall in normal cows and in cattle with retention of fetal membranes. Placenta 22 (5), 473-483. Wang, J., Li, T., Elsheikha, H.M., Chen, K., Cong, W., Yang, W., Bai, M., Huang, S., Zhu, X., 2018. Live-attenuated Pru: Δcdpk2 strain of Toxoplasma gondii protects against acute, chronic and congenital toxoplasmosis. J. Infect. Dis. 218(5), 768-777, 10.1093/infdis/jiy211. Wang, M., Lai, S., 2013. Fibronectin degradation by MMP-2/MMP-9 in the serum of pregnant women and umbilical cord with Toxoplasma gondii infection. J. Obstet. Gynaecol. 33 (4), 370-374, 10.3109/01443615.2013.769501. Wang, X., Michie, S.A., Suzuki, Y., 2007. Importance of IFN-gamma-mediated expression of endothelial VCAM-1 on recruitment of CD8+ T cells into the brain during chronic infection with Toxoplasma gondii. J Interferon Cytokine Res. 27 (4), 329-338. Wango, E.O., Wooding, F.B., Heap, R.B., 1990. The role of trophoblastic binucleate cells in implantation in the goat: a morphological study. J. Anat. 171, 241-257. Wastling, J.M., Xia, D., Sohal, A., Chaussepied, M., Pain, A., Langsley, G., 2009. Proteomes and transcriptomes of the Apicomplexa--where's the message? Int. J. Parasitol. 39 (2), 135-143, 10.1016/j.ijpara.2008.10.003. Waterkotte, B., Hambruch, N., Doring, B., Geyer, J., Tinneberg, H.R., Pfarrer, C., 2011. P-glycoprotein is functionally expressed in the placenta-derived bovine caruncular epithelial cell line 1 (BCEC-1). Placenta 32 (2), 146-152, Chapter VII ~ References 208 10.1016/j.placenta.2010.11.009. Wathes, D.C., Wooding, F., 1980. An electron microscopic study of implantation in the cow. Am. J. Anat. 159 (3), 285- 306. Weber, F.H., Jackson, J.A., Sobecki, B., Choromanski, L., Olsen, M., Meinert, T., Frank, R., Reichel, M.P., Ellis, J.T., 2013. On the efficacy and safety of vaccination with live tachyzoites of Neospora caninum for prevention of Neospora- associated fetal loss in cattle. Clin. Vaccine Immunol. 20 (1), 99-105, 10.1128/CVI.00225-12. Weiss, L.M., Fiser, A., Angeletti, R.H., Kim, K., 2009. Toxoplasma gondii proteomics. Expert Rev. Proteomics 6 (3), 303- 313, 10.1586/epr.09.16. Werling, D., Piercy, J., Coffey, T.J., 2006. Expression of Toll-like receptors (TLR) by bovine antigen-presenting cells- potential role in pathogen discrimination? Vet. Immunol. Immunopathol. 112 (1-2), 2-11. Weston, J.F., Heuer, C., Williamson, N.B., 2012. Efficacy of a Neospora caninum killed tachyzoite vaccine in preventing abortion and vertical transmission in dairy cattle. Prev. Vet. Med. 103 (2–3), 136-144, 10.1016/j.prevetmed.2011.08.010. Whitten, M.K., 1957. Effect of exteroceptive factors on the oestrous cycle of mice. Nature 180 (4599), 1436. Wiengcharoen, J., Thompson, R.C., Nakthong, C., Rattanakorn, P., Sukthana, Y., 2011. Transplacental transmission in cattle: is Toxoplasma gondii less potent than Neospora caninum? Parasitol. Res. 108 (5), 1235-1241, 10.1007/s00436- 010-2172-8. Wilczyński, J.R., 2005. Th1/Th2 cytokines balance—yin and yang of reproductive immunology. Eur. J. Obstet. Gynecolog. Reprod. Biol. 122 (2), 136-143. Williams, D.J., Guy, C.S., McGarry, J.W., Guy, F., Tasker, L., Smith, R.F., MacEachern, K., Cripps, P.J., Kelly, D.F., Trees, A.J., 2000. Neospora caninum-associated abortion in cattle: the time of experimentally-induced parasitaemia during gestation determines foetal survival. Parasitology 121 (4), 347-358. Williams, D.J., Guy, C.S., Smith, R.F., Guy, F., McGarry, J.W., McKay, J.S., Trees, A.J., 2003. First demonstration of protective immunity against foetopathy in cattle with latent Neospora caninum infection. Int. J. Parasitol. 33 (10), 1059- 1065. Williams, D.J., Trees, A.J., 2006. Protecting babies: vaccine strategies to prevent foetopathy in Neospora caninum- infected cattle. Parasite Immunol. 28 (3), 61-67. Williams, D.J., Guy, C.S., Smith, R.F., Ellis, J., Björkman, C., Reichel, M.P., Trees, A.J., 2007. Immunization of cattle with live tachyzoites of Neospora caninum confers protection against fetal death. Infect. Immun. 75 (3), 1343-1348. Williams, D.J., Hartley, C.S., Björkman, C., Trees, A.J., 2009. Endogenous and exogenous transplacental transmission of Neospora caninum - how the route of transmission impacts on epidemiology and control of disease. Parasitology 136 (14), 1895-1900, 10.1017/S0031182009990588. Wimsatt, W.A., 1951. Observations on the morphogenesis, cytochemistry, and significance of the binucleate giant cells of the placenta of ruminants. Am. J. Anat. 89 (2), 233-281. Wooding, F., Wathes, D.C., 1980. Binucleate cell migration in the bovine placentome. J. Reprod. Fertil. 59 (2), 425-430. Wooding, F., Flint, A., Heap, R., Morgan, G., Buttle, H., Young, I., 1986. Control of binucleate cell migration in the placenta of sheep and goats. J. Reprod. Fertil. 76 (2), 499-512. Wooding, F., 1992. The synepitheliochorial placenta of ruminants: binucleate cell fusions and hormone production. Placenta 13 (2), 101-113. Wooding, F., Morgan, G., Adam, C.L., 1997. Structure and function in the ruminant synepitheliochorial placenta: central role of the trophoblast binucleate cell in deer. Microsc. Res. Tech. 38 (1‐2), 88-99. Wooding, F., Roberts, R., Green, J., 2005. Light and electron microscope immunocytochemical studies of the distribution of pregnancy associated glycoproteins (PAGs) throughout pregnancy in the cow: possible functional implications. Placenta 26 (10), 807-827. Wujcicka, W., Wilczyński, J., Nowakowska, D., 2014. Do the placental barrier, parasite genotype and Toll-like receptor polymorphisms contribute to the course of primary infection with various Toxoplasma gondii genotypes in pregnant women? Eur. J. Clin. Microbiol. Infect. Dis. 33 (5), 703-709. Xia, D., Sanderson, S.J., Jones, A.R., Prieto, J.H., Yates, J.R., Bromley, E., Tomley, F.M., Lal, K., Sinden, R.E., Brunk, B.P., Roos, D.S., Wastling, J.M., 2008. The proteome of Toxoplasma gondii: integration with the genome provides novel insights into gene expression and annotation. Genome Biol. 9 (7), R116, 10.1186/gb-2008-9-7-r116. Yamagishi, J., Wakaguri, H., Ueno, A., Goo, Y.K., Tolba, M., Igarashi, M., Nishikawa, Y., Sugimoto, C., Sugano, S., Suzuki, Chapter VII ~ References 209 Y., Watanabe, J., Xuan, X., 2010. High-resolution characterization of Toxoplasma gondii transcriptome with a massive parallel sequencing method. DNA Res. 17 (4), 233-243, 10.1093/dnares/dsq013. Yamamoto, M., Standley, D.M., Takashima, S., Saiga, H., Okuyama, M., Kayama, H., Kubo, E., Ito, H., Takaura, M., Matsuda, T., Soldati-Favre, D., Takeda, K., 2009. A single polymorphic amino acid on Toxoplasma gondii kinase ROP16 determines the direct and strain-specific activation of Stat3. J. Exp. Med. 206 (12), 2747-2760, 10.1084/jem.20091703. Yamane, I., Kitani, H., Kokuho, T., Shibahara, T., Haritani, M., Hamaoka, T., Shimizu, S., Koiwai, M., Shimura, K., Yokomizo, Y., 2000. The inhibitory effect of interferon gamma and tumor necrosis factor alpha on intracellular multiplication of Neospora caninum in primary bovine brain cells. J. Vet. Med. Sci. 62 (3), 347-351. Yamauchi, N., Yamada, O., Takahashi, T., Imai, K., Sato, T., Ito, A., Hashizume, K., 2003. A three-dimensional cell culture model for bovine endometrium: regeneration of a multicellular spheroid using ascorbate. Placenta 24 (2-3), 258-269. Zeiler, M., Leiser, R., Johnson, G.A., Tinneberg, H.R., Pfarrer, C., 2007. Development of an in vitro model for bovine placentation: a comparison of the in vivo and in vitro expression of integrins and components of extracellular matrix in bovine placental cells. Cells Tissues Organs 186 (4), 229-242. Zeiner, G.M., Boothroyd, J.C., 2010. Use of two novel approaches to discriminate between closely related host microRNAs that are manipulated by Toxoplasma gondii during infection. RNA 16 (6), 1268-1274, 10.1261/rna.2069310. Zhang, D., Chen, L., Li, S., Gu, Z., Yan, J., 2008. Lipopolysaccharide (LPS) of Porphyromonas gingivalis induces IL-1β, TNF- α and IL-6 production by THP-1 cells in a way different from that of Escherichia coli LPS. Innate Immun. 14 (2), 99-107, 10.1177/1753425907088244. Zhang, H., Lee, E.G., Yu, L., Kawano, S., Huang, P., Liao, M., Kawase, O., Zhang, G., Zhou, J., Fujisaki, K., Nishikawa, Y., Xuan, X., 2011. Identification of the cross-reactive and species-specific antigens between Neospora caninum and Toxoplasma gondii tachyzoites by a proteomics approach. Parasitol. Res. 109 (3), 899-911, 10.1007/s00436-011-2332- 5. Zhang, W., Deng, C., Liu, Q., Liu, J., Wang, M., Tian, K.G., Yu, X.L., Hu, D.M., 2007. First identification of Neospora caninum infection in aborted bovine foetuses in China. Vet. Parasitol. 149 (1-2), 72-76. APPENDIX I ORIGINAL MANUSCRIPTS MANUSCRITOS ORIGINALES RESEARCH Open Access Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum Laura Jiménez-Pelayo1†, Marta García-Sánchez1†, Javier Regidor-Cerrillo1, Pilar Horcajo1, Esther Collantes-Fernández1, Mercedes Gómez-Bautista1, Nina Hambruch2, Christiane Pfarrer2 and Luis Miguel Ortega-Mora1* Abstract Background: Neospora caninum, one of the main causes of abortion in cattle, is very effective at crossing the placental barrier and placental damage is crucial in the pathogenesis of abortion. Bovine trophoblast and caruncular cell layers are key cellular components in the maternal-foetal interface in placentomes, playing a fundamental role in placental functionality. Methods: We studied tachyzoite adhesion, invasion, proliferation and egress of high- (Nc-Spain7) and low- (Nc- Spain1H) virulence N. caninum isolates in established cultures of bovine caruncular epithelial (BCEC-1) and trophoblast (F3) cells. The parasite invasion rate (pInvR) and the cell infection rate (cInfR) were determined by immunostaining plaque assay at different time points and multiplicities of infection (MOIs), respectively. In addition, tachyzoite growth kinetics were investigated using real-time PCR (qPCR) analysis and immunostaining plaque assay at different times. Results: Neospora caninum invaded and proliferated in both cell lines. The pInvR was higher in F3 compared to BCEC-1 cells for the Nc-Spain7 isolate (P < 0.05), and higher for the Nc-Spain7 than the Nc-Spain1H in F3 cells (P < 0.01). The cInfR was also higher in F3 cells than in BCEC-1 cells for both isolates (P < 0.0001), and the cInfR for the Nc-Spain7 isolate was higher than for the Nc-Spain1H isolate in both cell lines (P < 0.05). Tachyzoite growth kinetics showed tachyzoite exponential growth until egress at 58 hpi for both isolates in F3, whereas Nc-Spain1H showed a non- exponential growth pattern in BCEC-1. Asynchronous egress of both isolates was observed from 22 h post-infection onwards in BCEC-1. In addition, the tachyzoite yield (TY58h) was higher in F3 than in BCEC-1 infected by both isolates (P < 0.0001), highlighting better replication abilities of both parasites in F3. Nc-Spain7 showed shorter doubling times and higher TY58h compared to Nc-Spain1H in F3 cells; adhesion, invasion and proliferation mechanisms were very similar for both isolates in BCEC-1. Conclusions: Our results indicate a highly similar behavior of high- and low-virulence isolates in their interactions with maternal caruncular cells and suggest an important role of foetal trophoblasts in the pathogenesis of N. caninum infection. Keywords: Neospora caninum, Cattle, Isolates, Virulence, Placenta, Trophoblast, Caruncle, Adhesion, Invasion, Proliferation * Correspondence: luis.ortega@ucm.es †Equal contributors 1SALUVET, Animal Health Department, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain Full list of author information is available at the end of the article © The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 DOI 10.1186/s13071-017-2409-9 http://crossmark.crossref.org/dialog/?doi=10.1186/s13071-017-2409-9&domain=pdf mailto:luis.ortega@ucm.es http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/publicdomain/zero/1.0/ Background Neospora caninum is an apicomplexan protozoan para- site, phylogenetically related to Toxoplasma gondii. This parasite is considered a major cause of reproductive fail- ure in cattle worldwide [1–3], resulting in great economic losses [4]. Infection in cattle may occur through horizontal transmission, when cattle ingest sporulated oocysts shed by a canid definitive host, or by endogenous congenital transmission, from a persistently infected dam to a foetus [5]. Oral infection or recrudes- cence in a pregnant cow can result in abortion, birth of a weak calf or birth of a clinically healthy but persist- ently infected calf [1]. Neospora caninum is one of the most efficiently transplacentally-transmitted organisms in cattle [5]. Dur- ing natural infections, invasion of the placenta, prolifera- tion and dissemination to the foetus are crucial events in the pathogenesis of bovine neosporosis and are related to the interactions of tachyzoites with host cells and its relationship with the local immune response at the maternal-foetal interface [6]. In vivo studies demonstrated that N. caninum is able to infect the maternal caruncular septum before crossing to the foetal placental villus [7, 8]. Despite the fact that the placenta is directly involved in the pathogenesis of the disease [9, 10], the mechanisms by which N. caninum infects the placenta and reaches the fetus are poorly understood [11]. One reason could be the placental diversity [12], which makes an extrapolation of findings from one species to the other difficult. To date, only one limited in vitro study investigating the potential involvement of bovine trophoblast in N. caninum infec- tion has been published [13]. In addition, no information is available regarding in vitro infection in bovine caruncu- lar epithelial cells and the role of placental cell layers in vertical transmission. In addition, a key question in bovine neosporosis is the influence of the parasite intra-specific variability on the outcome of infection. The lytic cycle of N. caninum and other apicomplexan parasites comprises the pro- cesses of invasion, adaptation to intracellular conditions, proliferation, and egress from host cells [6, 14, 15]. This sequence of events is required for parasite survival and propagation in the course of animal infection. Our previ- ous findings demonstrated that N. caninum isolates of bovine or canine origin show a large biological diversity, despite being genetically similar [16]. Moreover, differ- ences found in the events of the lytic cycle among sev- eral N. caninum isolates in vitro are correlated with differences observed in virulence and vertical transmis- sion in animal models [16, 17]. Specifically, pregnant heifers inoculated at day 70 of gestation with the low- virulence isolate Nc-Spain1H spared the foetus [18], whereas foetal death occurred in all inoculated cattle with the highly virulent isolate Nc-Spain7 [19, 20]. There is no information concerning the kinetics of events in the placenta that lead to an understanding of how the parasite actually reaches the foetal tissues. The influence of biological variability of the isolate on pla- cental damage is also poorly understood. The cow pos- sesses a cotyledonary [21] and synepitheliochorial placenta [22], where foetal cotyledons interdigitate with maternal caruncles to form placentomes [23–25]. The trophoblast (epithelial surface of the foetal cotyledons) consists of uninucleated and binucleated cells. Binucle- ated cells are responsible for a “restricted” trophoblast invasion [26], playing an important role in embryo im- plantation and successful pregnancy outcomes. Caruncu- lar epithelial cells form a polarized barrier that the parasite encounters before reaching and multiplying in foetal tissues. Hence, the aim of this study was to inves- tigate the interaction of two isolates of N. caninum with maternal and foetal bovine target cells. Here, we studied tachyzoite adhesion, invasion, proliferation and egress of high- (Nc-Spain7) and low- (Nc-Spain1H) virulence iso- lates in established cultures of bovine caruncular epithe- lial (BCEC-1) and trophoblast (F3) cells. Since BCEC-1 and F3 cells conserve some of the properties from their tissues of origin [24, 27, 28], they are a useful tool to evaluate critical factors involved in placental pathogen- esis, such as the mechanisms used by N. caninum to cross the placental barriers. Methods Parasites and cell cultures Nc-Spain7 and Nc-Spain1H isolates were obtained from healthy, congenitally infected calves [29, 30] and exten- sively characterized using in vitro, murine and bovine models [16, 18, 20, 29, 31, 32]. Tachyzoites were rou- tinely maintained in a monolayer culture of the MARC- 145 monkey kidney cell line as described previously [16]. The N. caninum isolates used in this study were sub- jected to a limited number of culture passages (from 8 to 15) to ensure the maintenance of their in vivo bio- logical behaviour and avoid their adaptation to the host cells [33]. A bovine trophoblast cell line F3 [28] and a bovine caruncular cell line BCEC-1 [23] were isolated from two BVD-free, pregnant cattle (Bos taurus) with an estimated gestational age of 5 and 4 months, respectively. Cells were grown as indicated by Hambruch et al. [28]. Briefly, cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM)/Ham’s F12 containing 10% foetal calf serum (FCS) checked for the absence of specific IgG against N. caninum by IFAT, 100 IU/ml Penicillin, 100 mg/ml Streptomycin and 2 mM Glutamine. All ex- periments were carried out with cells below passage 27, when both cell lines maintained their morphological and functional features [24, 27, 28]. Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 2 of 13 Tachyzoites used for in vitro assays were recovered from 2.5–3 day growth cultures of MARC-145, when the majority of the parasites were still intracellular, and purified using Disposable PD-10 Desalting Columns (G.E. Healthcare, Buckinghamshire, UK) as previously described [16]. Tachyzoite viability was checked by try- pan blue exclusion. F3 and BCEC-1 cell monolayers were inoculated within 1 hour of tachyzoite collection from flasks. All in vitro experiments in F3 and BCEC-1 cell lines were assayed in quadruplicate, and two inde- pendent experiments were carried out. Parasite invasion rate Parasite invasion rate (pInvR) was defined as the number of tachyzoites invading the host cell at different time-points (hours) post-infection (hpi) and were determined following the methodology described in Dellarupe et al. [17] with minimal modifications. In order to obtain a confluent monolayer of F3 and BCEC-1, cells were seeded with 2 × 105 and 3 × 105 cells per well, respectively. Different density of both cell types were used because F3 cells are bigger than BCEC-1 cells and formed a monolayer com- posed of polygonal cells while as BCEC-1 are smaller and they tended to form colonies and did not spread out the entire surface of the well. A total of 100 purified tachyzoites of each isolate were added to 24-well culture plates. Cultures were washed three times with phosphate buffered saline (PBS) at different time points (1, 2, 4, 6 and 8 hpi) for removing non-adhered/non-invading tachyzoites. Un- washed cultures were also included in the study. All plates were fixed at 48 hpi, and the pInvR was determined using single immunofluorescence staining as described below. To determine the pInvR, events (medium and large parasito- phorous vacuoles) present in each well were counted using an inverted fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan) at a magnification of 200×. The pInvR at 1, 2, 4, 6 and 8 hpi (pInvR1h, pInvR2h, pInvR4h, pInvR6h, pInvR8h, respectively) was determined as the num- ber of events per well in cell monolayers washed at different time points, and the total parasite invasion rate (pInvRT) was determined as the number of events per well in unwashed cultures. Cell infection rate Multiplicity of infection (MOI) was defined as the ratio of the number of tachyzoites added to a known number of cells in a culture. Cell infection rate (cInfR) was defined as the percentage of cells infected using different MOIs (1, 2, 4, 6, 8 and 10). Cells were cultured in 24-well plates at concentration of 2 × 105 and 3 × 105 cells per well for F3 and BCEC-1 cells, respectively. Infected cells were washed 3 times with PBS after 4 hpi to facilitate the synchronization of the cultures. Finally, cells were fixed at 48 hpi and stained using single immunofluorescence stain- ing as described below. The overall number of cells, the number of infected cells and the number of cells containing more than one vacuole (multi-infected cells) were counted in 10 arbi- trarily selected fields using an inverted fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan) at a magnification of 200×. Counting of events was carried out on images taken with three different fil- ters (white light for discrimination of cell limits, blue- DAPI for visualization of the nuclei and red-Alexa 594 for examination of the tachyzoites) using a Nikon DSL1 camera (Chiyoda, TYO, Japan) and overlaid using Photo- shop® software (Adobe Systems Incorporated, Mountain View, CA, USA). A mean value of 50 cells was counted in each field. Adhesion-invasion assay An adhesion-invasion assay was performed in F3 and BCEC-1 cultures seeded at concentration of 2 × 105 and 3 × 105 cells per well, respectively. Cells were infected at a MOI of 2, and cultures were washed with PBS at 4 hpi to remove non-adherent extracellular tachyzoites. Cultures were immediately fixed and double immunofluorescence staining was carried out following the protocol described below. Adhered extracellular tachyzoites (green- and red- stained) and intracellular tachyzoites (red-stained only) were counted using a fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan) at a magnification of 400×. A total of 1000 tachyzoites was counted in each coverslip. The percentage of intracellular tachyzoites (red- stained) respect to the total number of tachyzoites (intra- cellular and extracellular adhered tachyzoites) (green- stained) at 4 hpi was calculated. Intracellular proliferation assays: Proliferation kinetics, doubling time and tachyzoite yield determinations Proliferation kinetics of Nc-Spain7 and Nc-Spain1H iso- lates in F3 and BCEC-1 cells were determined by quanti- fying the number of tachyzoites at specific times (4, 10, 22, 34, 46, 58, 70 and 82 hpi) by real-time PCR (qPCR). Cells were cultured and infected as indicated above using a MOI of 2. Cultures were washed at 4 hpi and subsequently maintained at 37 °C in 5% CO2. The sam- ples were collected adding 200 μl of PBS, 180 μl of lysis buffer and 20 μl of proteinase K (Qiagen, Hilden, Germany) to each well at 4, 10, 22, 34, 46, 58, 70 and 82 hpi, transferred to a microcentrifuge tube and frozen at -80 °C prior to DNA extraction. In parallel, replicates of cell cultures in coverslips were infected as described above and were labelled using double-inmunostaining to study microscopically the pro- liferation kinetics of both isolates in F3 and BCEC-1 cells. Three coverslips were photographed for each Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 3 of 13 condition using an inverted fluorescence microscope (Nikon Eclipse TE 200, Chiyoda, TYO, Japan). The doubling time (Td) was defined as the period of time required for a tachyzoite to duplicate during the ex- ponential multiplication period, excluding the lag phase (period without parasite multiplication) and the egress phase [16]. The Td was determined by applying non- linear regression analysis and an exponential growth equation using GraphPad (San Diego, CA, USA). We represented Td for each isolate and each cell line as the average value obtained from all the determinations that revealed a linear regression, R2 ≥ 0.95. The tachyzoite yield (TY58h) was defined as the aver- age value of the number of tachyzoites quantified by qPCR at 58 hpi for each isolate and cell line. Immunofluorescence staining Single immunofluorescence staining was carried out as specified previously [17] with minimal variations. Para- sites in fixed cultures were stained using hyperimmune rabbit antiserum directed against N. caninum tachyzoites (1:1000) as a primary antibody and a 1:1000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor® 594 (red, Thermo Fisher Scientific, Waltham, MA, USA) as a sec- ondary antibody. The nuclei were stained by washing the cells with a solution of 1:5000 DAPI in PBS. Double immunofluorescence staining was carried out fol- lowing the protocol described by Regidor-Cerrillo et al. [16] with minimal modifications. Fixed plates were treated with 3% BSA in PBS for 30 min at 25 °C to block unspecific anti- body binding and autofluorescence. Then, cultures were treated with a 1:1000 dilution of anti-tachyzoite hyper- immune rabbit antiserum and a 1:1000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor® 488 (green, Thermo Fisher Scientific, Waltham, MA, USA). After this step, only extracellular tachyzoites were labelled in green. After the first staining, cells were permeabilized using a solution of 0.25% Triton 100X in PBS 0.3% BSA (30 min, 37 °C). Later, cultures were treated again with a dilution of anti-tachyzoite hyperimmune rabbit anti- serum as primary antibody (1:1000) and a 1:1000 dilution of goat anti-rabbit IgG conjugated to Alexa Fluor® 594 as secondary antibody (red, Thermo Fisher Scientific, Waltham, MA, USA). Therefore, intracellular tachyzoites were labelled only in red, while extracellular tachyzoytes resulted labelled in green and in red. The nuclei were stained by washing the cells with a solution of 1:5000 DAPI in PBS and the coverslips were embed- ded in Fluoroprep (BioMerieux, Marcy-l’Étoile, France). DNA extraction and real-time PCR Genomic DNA was extracted from cellular samples using the DNeasy® Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Genomic DNA was eluted in a volume of 60 μl of molecular-grade water. Concentrations of DNA were determined for each sample using a nanophotometer (Nanophotomer®, Implen GmbH, Munich, Germany) and samples were diluted 1:4 in molecular-grade water. Quantification of N. caninum DNA was performed by real-time PCR using an Applied Biosystems 7300 Real- Time PCR System (Applied Biosystems, Foster City, CA, USA). The Nc-5 region was targeted as described else- where [34]. Five μl of diluted DNA from each sample were used for the qPCR assays. The number of N. caninum tachyzoites was determined by interpolating the Ct values (cycle threshold value, which represents the fractional cycle number reflecting a positive PCR result) on a standard curve. The standard curve was designed for the quantification of 10-1–104 tachyzoites according to Regidor-Cerrillo et al. [16]. To normalize the quan- tification of the parasites in each sample, a bovine β- actin standard curve was designed (from 64 ng of DNA per μl to 0.2 ng per μl). The results were expressed as the relation between amounts of parasite DNA and cell DNA (R2 ≥ 0.99; slope values varied from -3.67 to -3.13). Statistical analysis The parametric one-way ANOVA test, followed by a Tukey’s multiple comparisons test, was performed to investigate the influence of time on the pInvR and MOI in the cInfR, and the two-way ANOVA test, followed by a Tukey’s multiple comparisons test, was performed to study the influence of the parasite isolate and the cell type on the pInvR and cInfR. A Chi-square test was carried out to investigate the differences in the percentages of intra- cellular tachyzoites at 4 hpi in both target cell types. Bon- ferroni correction was used to eliminate error associated with making multiple comparisons. Statistical significance was established as P < 0.05. Differences that showed P- values ≥ 0.05 and < 0.1 were considered to be trending to- wards statistical significance. GraphPad Prism 5 v.5.01 (San Diego, CA, USA) software was used to perform all statistical analyses and graphical illustrations. Results Parasite invasion rate (pInvR) To investigate the impact of the placental cell type in para- site invasion, the pInvR was evaluated in trophoblasts and caruncular cells at different time points post-infection (1, 2, 4, 6 and 8 hpi). The pInvRs of the Nc-Spain7 and Nc- Spain1H isolates in F3 and BCEC-1 cells are shown in Fig. 1. The number of invaded tachyzoites for both isolates signifi- cantly increased until 4 hpi in both F3 (Nc-Spain7, ANOVA: F(5,35) = 11.87, P < 0.0001; Nc-Spain1H, ANOVA: F(5,35) = 7.211, P < 0.0001, followed by a Tukey’s multiple comparisons test) and BCEC-1 cells (Nc-Spain7, ANOVA: Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 4 of 13 F(5,35) = 9.825, P < 0.0001; Nc-Spain1H, ANOVA: F(5,35) = 9.156, P < 0.0001, followed by a Tukey’s multiple compari- sons test). From 4 hpi onwards, significant differences were not observed. Regarding the influence of the target cell type, a higher pInvR was observed in F3 cells compared to BCEC-1 cells from 6 hpi onwards for the Nc-Spain7 isolate (two- way ANOVA test: F(3,168) = 27.25, P < 0.0001, followed by a Tukey’s multiple comparisons test). No statistically sig- nificant differences were found when the pInvRs of Nc- Spain1H in both cell lines were investigated. The influence of the parasite isolate on the invasion of bovine trophoblasts and caruncular cells was also investi- gated by comparison of the pInvRs between the Nc- Spain1H and Nc-Spain7 isolates, assayed at different times of infection. Nc-Spain7 showed a pInvR significantly higher than Nc-Spain1H from 6 hpi onwards in F3 cells (two-way ANOVA test: F(3,168) = 27.25, P < 0.0001 followed by a Tukey’s multiple comparisons test) (Fig. 1a). However, no statistically significant differences in pInvR were found between isolates in BCEC-1 cells (Fig. 1b). Cell infection rate (cInfR) The percentage of infected cells (cInfR) and the percentage of multi-infected cells were evaluated at different MOIs. The number of infected cells significantly increased with increasing MOIs in both cell lines F3 (Nc-Spain7, ANOVA: F(5,42) = 228.5, P < 0.0001; Nc-Spain1H, ANOVA: F(5,42) = 273.4, P < 0.0001, followed by a Tukey’s multiple comparisons test) and BCEC-1 (Nc-Spain7, ANOVA: F(5,42) = 30.04, P < 0.0001; Nc-Spain1H, ANOVA: F(5,42) = 42.60, P < 0.0001, followed by a Tukey’s multiple comparisons test). The cInfRs were higher in in- fected F3 than in BCEC-1 cells for both Nc-Spain7 and Nc-Spain1H isolates at the same MOI (Fig. 2a, b) (two- way ANOVA test: F(3,168) = 222.4, P < 0.0001, followed by a Tukey’s multiple comparisons test). The percentage of cells containing more than a single vacuole (Fig. 2c, d) was also higher in F3 than in BCEC-1 infected by both isolates (two-way ANOVA test: F(3,168) = 93.64, P < 0.0001, followed by a Tukey’s multiple comparisons test). In addition, Nc-Spain7 showed a higher cInfR than Nc-Spain1H in both cell lines at different MOIs (two- way ANOVA test: F(3,168) = 222.4, P < 0.0001, followed by a Tukey’s multiple comparisons test) (Fig. 2a, b). We also observed that Nc-Spain7 showed a higher percentage of multi-infected cells than Nc-Spain1H in F3 cells at 6, 8 and 10 MOIs (two-way ANOVA test: F(3,168) = 93.64, P < 0.0001, followed by a Tukey’s multiple comparisons test) (Fig. 2c). However, no significant differ- ences in the number of multi-infected cells were found be- tween isolates in BCEC-1 cells (Fig. 2d). Adhesion-invasion assay In the light of the differences in pInvR and cInfR be- tween both cells lines, an adhesion-invasion assay was performed to investigate whether these differences could be attributed to a different adhesion ability of the tachy- zoites in these two cell lines, a different ability to pene- trate in the cells or both (Fig. 3a). In this assay, non- adhered tachyzoites were eliminated in the washing step at 4 hpi before the fixation and extra- and intracellular adhered tachyzoites were counted. The percentage of intracellular tachyzoites respect to the total adhered intra- and extracellular tachyzoites was calculated. Surprisingly, both isolates showed that almost 100% of adhered tachyzoites were intracellular at 4 hpi in BCEC- 1 cells, whereas a minor percentage of intracellular Fig. 1 Parasite invasion rates in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H isolates. Graphs represent parasite infection rates in F3 (a) and BCEC-1 cells (b) defined as the percentage of invaded tachyzoites (number of events per well) studied at different time points for Nc-Spain7 and Nc-Spain1H. Each column and error bar represents the mean and the SD of 4 replicates from 2 independent assays at the indicated sampling times. The total number of invaded tachyzoites was determined by single immunofluorescence staining of events (parasitophorous vacuoles and lysis plaques) followed by counting using an inverted fluorescence microscope. Significantly higher pInvRs were found in F3 cells compared to BCEC-1 cells infected with Nc-Spain7 (P < 0.01), whereas no differences were found in the pInvRs of F3 and BCEC-1 cells infected by Nc-Spain1H (P > 0.05). * represents significant differences between isolates Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 5 of 13 tachyzoites was observed in F3 cells, 88 and 69% for Nc- Spain7 and Nc-Spain1H, respectively. A significantly higher number of adhered tachyzoites from both isolates were internalized in BCEC-1 cells than in F3 cells at 4 hpi (Chi-square test: χ2 = 287.6, df = 3, P < 0.0001) (Fig. 3b). Differences between isolates were not observed in BCEC-1, although the high-virulence isolate Nc-Spain7 showed a better ability to penetrate than the low- virulence isolate Nc-Spain1H in F3 cells (Fisher’s exact test: P < 0.0001) (Fig. 3b). Proliferation kinetics, doubling time comparisons and tachyzoite yield determination An in vitro intracellular proliferation assay was carried out to study proliferation and egress events of the lytic cycle of N. caninum in trophoblast and caruncular cell cultures. Proliferation kinetics over time assessed by qPCR are represented in Fig. 4a, b. The growth curves of both isolates in F3 and the growth curve of Nc- Spain7 in BCEC-1 adjusted to exponential growth from 10 hpi to 70 hpi, whereas the growth curve of Nc- Spain1H in BCEC-1 did not adjust either to the exponential or linear growth mathematical model. Ana- lysing the Td, we observed a delay in the multiplication of N. caninum in BCEC-1 cells, with the average Td value of Nc-Spain7 1.5-times higher in BCEC-1 cells (14.603 ± 1.428) than in F3 cells (9.425 ± 0.239) (one-way ANOVA: F(2,21) = 6.966, P = 0.0048 followed by a Tukey’s multiple comparisons test). The Nc-Spain1H isolate showed an average Td value of 12.246 ± 0.893 in F3 cells. The Td value for Nc-Spain1H in BCEC-1 could not be calculated due to the lack of exponential growth of Nc- Spain1H in BCEC-1. Nevertheless, no significant differ- ences were found in the average Td values for Nc-Spain7 and Nc-Spain1H isolates in F3 cells. A microscopic examination of cultures fixed at different time points showed that the multiplication of Nc-Spain7 and Nc-Spain1H isolates began between 10 and 22 hpi. Differences in the parasitophorous vacuole size between both isolates were observed in F3 cells from 34 hpi on- wards, with bigger vacuoles in Nc-Spain7 infected cells. However, no differences between the isolates were demon- strated by immunofluorescence in BCEC-1 cells. Between 58 and 82 hpi, asynchronous rupture of host cells and Fig. 2 Infection and multi-infection rates in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H isolates. Graphs represent the cell infection rates as the percentage of infected cells in F3 (a) and BCEC-1 cells (b) for both isolates and the percentage of cells with multi-infection (more than one parasitophorous vacuole) in F3 (c) and BCEC-1 cells (d). Each column and error bar represents the mean and the SD of 4 replicates from 2 independent assays using different MOIs. The total number of cells, the number of infected cells and the number of cells with multi-infection were determined by double immunofluorescence staining followed by counting using an inverted fluorescence microscope. The cInfRs were higher in F3 than in BCEC-1 cells infected by both isolates (P < 0.0001). The percentage of cells containing more than a single vacuole was also higher in F3 than in BCEC-1 cells infected by both isolates (P < 0.05). * represents significant differences between isolates Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 6 of 13 egress of the tachyzoites were observed in F3 cells. How- ever, interestingly, an early egression of tachyzoites from 22 hpi onwards was observed in BCEC-1 (Fig. 4c). The TY58h was assessed to determine the number of tachyzoites produced during the same intracellular period after invasion, prior to complete tachyzoite egress from cell cultures (Fig. 4d). The TY58h was 15-times higher in F3 cells than in BCEC-1 cells infected with Nc-Spain7, and 10-times higher in F3 cells infected with Nc-Spain1H. Comparing the isolates, Nc-Spain7 showed a higher TY58h than Nc-Spain1H in F3 (one-way ANOVA: F(3,28) = 37.35, P < 0.0001 followed by a Tukey’s multiple comparisons test), whereas no differ- ences in the TY58h were found in BCEC-1. The results obtained in the present work are summa- rized in Table 1. Discussion In the present study, we established for the first time a species- and organ-specific in vitro model for each of the host cell layers in the maternal-foetal interface of the bovine placenta to study N. caninum infection. To date, several in vitro studies have been carried out using established cell lines, such as Marc-145, HeLa, BeWo or Fig. 3 Adhesion assay in F3 and BCEC-1 cells infected by Nc-Spain7 and Nc-Spain1H at 4 hpi. Double immunofluorescence staining was performed, and adhered extracellular tachyzoites were stained with Alexa Fluor® 488 (green) and Alexa Fluor® 594 (red), whereas intracellular tachyzoites were stained with Alexa Fluor® 594 (red). Nuclei were stained with DAPI (blue). Tachyzoites were counted in 10 arbitrarily selected fields, and the percentage of intracellular tachyzoites relative to the number of total adhered tachyzoites at 4 hpi was calculated. Representative images at a magnification of 1000× (a) show the adhesion assay performed in F3 and BCEC-1 cells infected with both isolates. The graph (b) represents the percentage of intracellular tachyzoites of Nc-Spain7 and Nc-Spain1H relative to the total number of intra- and extracellular tachyzoites adhered to F3 and BCEC-1 cells. Each column and error bar represents the mean and the SD of 4 replicates from 2 independent assays. BCEC-1 cells showed a significantly higher percentage of intracellular tachyzoites than F3 cells (P < 0.0001, Chi-square test). The percentage of intracellular tachyzoites for Nc-Spain7 (88%) was significantly higher than for Nc-Spain1H (69%) in F3 (P < 0.0001), whereas the percentage of intracellular tachyzoites of both isolates in BCEC-1 was the same (96%). * represents significant differences between isolates. Scale-bars: a, 40 μm Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 7 of 13 Fig. 4 (See legend on next page.) Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 8 of 13 ovine trophoblast cells, to investigate the invasion and proliferation of different N. caninum isolates [16, 35, 36]. There is only one (limited) descriptive study con- cerning the interaction of the parasite with bovine trophoblast cells [13], and no data are available about the parasite’s interaction with bovine caruncular cells. The interaction between N. caninum and these target cells has been studied in the present work, using two isolates of different virulence and the two cell lines that represent the maternal-foetal interface. The cell lines used in this work (bovine trophoblast cells, F3, and bo- vine caruncular epithelial cells, BCEC-1) were isolated from fifth- and fourth-month pregnant heifers, respect- ively, and they have maintained at least part of their morphological and functional characteristics [24, 27, 28]. Thus, they may be a useful tool to investigate the path- ways of N. caninum infection during transplacental transmission during the second trimester of pregnancy when the majority of abortions caused by N. caninum occur [1]. Investigations were focused on the lytic cycle of the tachyzoites (host-cell invasion, proliferation and egress). The processes implicated in the lytic cycle of the parasite are essential for the invasion of host tissues, the distribution of the parasite through the organism and its distribution to the placenta. As a consequence, abortion or transplacental transmission may occur [1, 6]. Here, two N. caninum isolates with marked differences in virulence were able to establish themselves and multiply both in the maternal ephitelium of the caruncle and in the foetal trophoblast, although differences in the infec- tion of both cell types were found. Our results showed a lower infection rate, as well as a lower percentage of cells with multi-infection, for both parasite isolates in BCEC-1 cells relative to F3 cells. Therefore, N. caninum tachyzoites more efficiently infect trophoblast cells compared to bovine caruncular epithe- lial cells, meaning bovine trophoblast cells are more sus- ceptible to N. caninum infection. In experimental infections, higher parasite burdens and more severe le- sions have been found in the foetal part of the placenta [19, 20]. As previously demonstrated, BCEC-1 is an established cell line showing in vitro characteristics of a morphologically and functionally intact epithelial barrier with apical microvilli and junctional complexes (zonula occludens, zonula adherens and desmosomes) [23, 24] as described in vivo [37, 38]. This polarized barrier, with apical junctional complexes obliterating the paracellular space, establishes an effective paracellular barrier to dif- fusion of fluid and solutes, limiting the passage of foetal and maternal metabolites [39]. These characteristics may be hindering the paracellular passage of N. caninum across the epithelium. Foetal cells (F3) share many prop- erties with maternal cells (BCEC-1), including apical microvilli and expression of the tight junctional zonula (See figure on previous page.) Fig. 4 Proliferation kinetics over time and tachyzoite yield at 58 hpi. Graphs (a and b) represent the average number of tachyzoites for each time-point for all individual experiments with an R2 > 0.95, except for BCEC-1 cells infected by Nc-Spain1H, which showed a non- exponential growth pattern. Error bars indicate the SD. Representative images (c) show the proliferation kinetics over time of Nc-Spain7 and Nc-Spain1H isolates in F3 and BCEC-1 cultures. The bar graph (d) represents the tachyzoite yield at 58 hpi for Nc-Spain7 and Nc- Spain1H in F3 and BCEC-1 cells. Each column and error bar represents the mean and the SD of 4 replicates from 2 independent assays. The TY58h was fifteen times higher in F3 cells than in BCEC-1 cells infected with Nc-Spain7, and ten times higher in F3 cells infected with Nc-Spain1H. Statistical differences were found in the TY58h between isolates in F3 cells, with the TY58h of Nc-Spain7 significantly higher than for Nc-Spain1H (P < 0.0001). * represents significant differences between isolates. Scale-bars: c, 10 μm Table 1 Summary of virulence traits as a function of cell type and N. caninum isolate pInvRa cInfRb % Multi-infectionc Invasion efficiencyd TY58H e Td f Cell type comparisons (F3 vs BCEC-1) Nc-Spain7 + ++++ + - - - - + - - Nc-Spain1H NSh ++++ + - - - - + NCg Isolate comparisons (Nc-Spain7 vs Nc-Spain1H) F3 ++ + ++++ ++++ ++ NS BCEC-1 NS + NS NS NS NCg apInvR (parasite invasion rate): number of tachyzoites invading the host cell at different time points post-infection bcInfR (cell infection rate): percentage of cells infected using different parasite doses c% Multi-infection: percentage of cells containing more than one vacuole dInvasion efficiency: results from adhesion-invasion assay, percentage of intracellular tachyzoites relative to the total number of tachyzoites at 4 hpi eTY58H (tachyzoite yield at 58 hpi): average number of tachyzoites quantified by qPCR at 58 hpi fTd (Doubling time): period of time required for a tachyzoite to duplicate during the exponential multiplication period, excluding lag and egress phases gNC data not comparable. The Td value for Nc-Spain1H isolate in BCEC-1 cells could not be calculated due to the lack of exponential growth of Nc-Spain1H in BCEC-1 hNS no significant differences +/++/++++ indicate higher rates of each parameter assayed with a significance of P < 0.05, P < 0.01, P < 0.0001, respectively -/- -/- - - - indicate lower rates of each parameter assayed with a significance of P < 0.05, P < 0.01, P < 0.0001, respectively Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 9 of 13 occludens protein both in vivo [37, 38] and in vitro [28]; however, in contrast to the maternal BCEC-1 cells, mononuclear trophoblast cells have phagocytic pheno- types in vivo [40]. Also bovine F3 cells may form binu- cleated cells [28], which have phagocytic activity as has been previously described for trophoblast giant cells of various species [40–42]. The phagocytic activity of mono- and binucleate trophoblast cells may be mediat- ing parasite passage to the foetus [13]. In fact, in a BALB/c mouse model infected with T. gondii, a higher frequency of infected placentas was observed at later stages of pregnancy, which has been correlated with a higher phagocytic efficiency of the placental tissues in this period [43]. Therefore, although differences in junc- tional complexes between both cell types should be in- vestigated to evaluate their influence in parasite invasion, it seems probable that the phagocytic ability of trophoblast cells may partially explain the higher suscep- tibility to parasite invasion observed in these cells. Noting the differences in the invasion and infection rates between both cell lines, an adhesion-invasion assay was performed in order to elucidate whether these dif- ferences could be associated with lower adhesion, pene- tration or both. Our results revealed that, contrary to expectations, both isolates, which had presented lower cell infection rate in BCEC-1 cells compared to F3 cells, showed a highly efficient invasion with almost 100% penetration of the adhered tachyzoites in BCEC-1 cells at 4 hpi. Thus, the lower invasion of N. caninum ob- served in BCEC-1 cells may be due to a lower ability to adhere or to a fragile adhesion to host-cell receptors. Tachyzoite adhesion occurs in two phases, as previously described [6]. The first step is the establishment of low- affinity contact between tachyzoites and the host-cell surface membrane, where surface antigens of N. cani- num tachyzoites such as NcSAG1 and NcSRS2 are in- volved. Later, the actual adhesion process occurs via to microneme proteins (especially NcMIC3), which bind to host-cell surface chondroitin sulfates. Studying the dif- ferences in the type and abundance of superficial recep- tors responsible for the high-affinity interaction with tachyzoites between both cell lines could aid the under- standing of the diminished adhesion of N. caninum in bovine caruncular cells. Concerning the growth kinetics of N. caninum in foetal and maternal cells, our results showed a dramatic- ally lower proliferation of both isolates in caruncular cells. Moreover, Nc-Spain1H did not adjust to an expo- nential growth in maternal cells. Differences observed between both cell types may be partially attributed to a different degree of maturation. While the foetus is not completely immunocompetent in the second trimester of gestation, maternal cells have immunocompetent abil- ities that may restrict the infection and proliferation of the parasite. In vivo studies have demonstrated the influ- ence of the gestational stage on the outcome of N. cani- num infection [44–47]. It is known that the survival of the foetus depends on the state of development of its immune system, as higher abortion rates, higher parasite burdens and more severe lesions were observed in foetal tissues when infection occurred in the first and second trimester of gestation [44, 46, 47]. On the other hand, the lower multiplication of N. caninum in the maternal side of the placenta supports the hypothesis that carun- cles act as a barrier, limiting not only parasite infection via reduced adhesion but also its multiplication. In ex- perimental infections, comparisons between cytokine mRNA levels in separated maternal and foetal placental tissues showed that maternal tissue was the major source of most cytokines [48] and had a major lympho- cyte cell infiltration, particularly in the maternal caruncle [7, 49], which may indicate that the maternal immune system was actively responding to the parasite. More- over, early egress was observed in caruncular cells, which could be employed by the parasite as an escape mechan- ism to facilitate the dissemination of the parasite to the foetal part of the placenta, which has been demonstrated in this work to be the parasite’s preferential target cell, and, thus, allow vertical transmission of the parasite. As mentioned above, the role of the parasite in the outcome of infection is also a determining factor. Pro- cesses involved in the lytic cycle, including parasite inva- sion and intracellular proliferation, are essential for the maintenance and multiplication of the parasite in vitro and for parasite survival and propagation in host tissues during the course of animal infection [1, 6]. In previous studies, several N. caninum isolates showed differences during the in vitro lytic cycle and more virulence than others in animal models, associated with higher abortion and transmission rates [16, 18, 20, 29, 31, 50]. In tropho- blast cells, both isolates, described as “highly prolific” (Nc-Spain7) and “less prolific” (Nc-Spain1H) in previous studies [16], showed the same in vitro characteristics. In particular, the virulent isolate Nc-Spain7 demonstrated greater invasion, infection and proliferation rates than Nc-Spain1H in trophoblast cells. These differences may be explained by their biological diversity, as has been demonstrated in previous in vivo studies [16, 18, 20, 29, 32, 51, 52]. Nc-Spain7 showed a high neonatal mortality (95%) and vertical transmission rate (nearly 80%) in a pregnant BALB/c mouse model [31], as well as a per- centage of abortion and vertical transmission as high as 100% in a bovine model [20, 53]. However, Nc-Spain1H showed a 100% offspring survival rate and a low vertical transmission rate (5%) in a pregnant mouse model [29], and no foetal death was observed in experimentally in- fected cattle [18]. The higher proliferation ability of Nc- Spain7 in trophoblast cells found in the present study Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 10 of 13 may be responsible for the increase in the quantity of parasites reaching the foetal tissues and, consequently, for the enhancement of parasite burdens and pathology, ultimately resulting in foetal death and abortion. These results agree with those obtained in previous studies, where higher parasite burdens in the brain and placental tissues, a wider spread and greater severity of histo- pathological lesions and clinical signs were observed in animals experimentally infected with Nc-Spain7 [18, 20, 50]. However, in Nc-Spain1H-infected animals, less se- vere lesions were observed in placentas and maternal and foetal tissues [18], which may explain the absence of abortion. In terms of dissemination in vivo, isolates with low virulence could have a lower efficiency at crossing biological barriers. More interestingly, contrary to our observation in trophoblast cells, the behaviour of both isolates was very similar in bovine caruncular cells. Differences between isolates were limited to a slightly higher infection rate by the virulent isolate Nc-Spain7, whereas adhesion, inva- sion and proliferation mechanisms were very similar for both isolates. This fact has also been observed in the phylogenetically-related protozoan T. gondii, where comparisons between three strains showed no significant differences in their capacity to infect human placental explants [11]. The comparable behaviour showed by dif- ferent virulence isolates, together with the lower inva- sion, infection and proliferation rates found in caruncular cells, leads us to hypothesize that isolates may have been selected because of a low virulence in the maternal part of the placenta despite their differences in virulence traits in other host cells, including other pla- cental cells such as trophoblasts. This reduced virulence in the caruncle may facilitate, on the one hand, evasion from maternal immunity and the placental damage caused by parasite multiplication, leading to the abor- tion. On the other hand, this behaviour may facilitate vertical transmission to the progeny, which is the main route of transmission for N. caninum. In fact, Nc-Spain7 and Nc-Spain1H isolates were obtained from healthy but congenitally infected calves, as described above. Conclusions This is the first study where an in vitro model of N. caninum infection has been implemented in bovine pla- cental cells. Our findings confirm a differential compe- tency of two isolates of N. caninum with different virulence to proliferate in bovine trophoblast cells. How- ever, bovine caruncular cells were the first cell line assayed where different virulence isolates showed similar invasion, adhesion and proliferation kinetics. The low replication of both isolates in the maternal side of the placenta may facilitate the evasion of the inmune re- sponse by the parasite, allowing their transplacental transmission. This fact may have constituted an evolu- tionary advantage for these isolates. Remarkably, limited parasite invasion and growth in caruncular cells suggest a putative barrier function for this cell type in the pla- centa, although early parasite egress may facilitate trans- mission to offspring. Furthermore, our results confirm the role of foetal tro- phoblasts as target cells for N. caninum. Future research to determine the differences in surface receptors and cell junctions between both placental cell types are needed. In addition, studies focused on co-cultures of maternal and foetal cells may be helpful as a model to study para- site transport across the maternal epithelium as part of the bovine placental barrier. Finally, the existence of differences in local immunomodulation and cellular mechanisms that take place in the placenta infected by high- and low-virulence isolates of N. caninum should be investigated. Abbreviations BCEC-1: Bovine caruncular epithelial cell line 1; BVD: Bovine viral diarrhea; CInfR: Cell infection rate; Ct: Cycle threshold value; DMEM: Dulbecco’s Modified Eagle Medium; F3: Bovine placental trophoblast cell line; FCS: Foetal calf serum; hpi: Hours post-infection; MOI: Multiplicity of infection; PInvR: Parasite invasion rate; pInvRT: Total parasite invasion rate; qPCR: Real-time PCR; SD: Standard deviation; Td: Doubling time; TY58h: Tachyzoite yield at 58 h post-infection Acknowledgements Not applicable. Funding This work was supported by the Spanish Ministry of Economy and Competitiveness (AGL2013–44694-R) and the Community of Madrid (PLATESA S2013/ABI2906). LJP was financially supported by a fellowship from the University Complutense of Madrid and MGS was financially supported through a grant from the Spanish Ministry of Economy and Competitiveness (BES-2014-070723). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Availability of data and materials Not applicable. Authors’ contributions JRC, PH, ECF, MGB and LMO conceived the study and participated in its design. LJP and MGS wrote the manuscript, with interpretation of results and discussion inputs from JRC, ECF, MGB, LMO, NH and CP. LJP and MGS performed in vitro plaque and immunofluorescence assays. JRC, PH, LJP and MGS designed and performed RT-qPCR analyses. NH and CP isolated bovine trophoblast and caruncular cell lines used in the assays. LJP and MGS carried out statistical analyses and interpreted the results. All authors read and approved the final manuscript. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 11 of 13 Author details 1SALUVET, Animal Health Department, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain. 2Department of Anatomy, University of Veterinary Medicine Hannover, Bischofsholer Damm 15, 30173 Hannover, Germany. Received: 24 April 2017 Accepted: 27 September 2017 References 1. Dubey JP, Schares G, Ortega-Mora LM. Epidemiology and control of neosporosis and Neospora caninum. Clin Microbiol Rev. 2007;20(2):323–67. 2. Innes EA, Wright S, Bartley P, Maley S, Macaldowie C, Esteban-Redondo I, et al. The host-parasite relationship in bovine neosporosis. Vet Immunol Immunopathol. 2005;108(1–2):29–36. 3. Dubey JP, Buxton D, Wouda W. Pathogenesis of bovine neosporosis. J Comp Pathol. 2006;134(4):267–89. 4. Reichel MP, Alejandra Ayanegui-Alcérreca M, LFP G, Ellis JT. What is the global economic impact of Neospora caninum in cattle - the billion dollar question. Int J Parasitol. 2013;43(2):133–42. 5. Williams DJ, Hartley CS, Bjorkman C, Trees AJ. Endogenous and exogenous transplacental transmission of Neospora caninum - how the route of transmission impacts on epidemiology and control of disease. Parasitology. 2009;136(14):1895–900. 6. Hemphill A, Vonlaufen N, Naguleswaran A. Cellular and immunological basis of the host-parasite relationship during infection with Neospora caninum. Parasitology. 2006;133:261–78. 7. Gibney EH, Kipar A, Rosbottom A, Guy CS, Smith RF, Hetzel U, et al. The extent of parasite-associated necrosis in the placenta and foetal tissues of cattle following Neospora caninum infection in early and late gestation correlates with foetal death. Int J Parasitol. 2008;38(5):579–88. 8. Benavides J, Collantes-Fernández E, Ferre I, Pérez V, Campero C, Mota R, et al. Experimental ruminant models for bovine neosporosis: what is known and what is needed. Parasitology. 2014;141(11):1471–88. 9. Entrican G. Immune regulation during pregnancy and host-pathogen interactions in infectious abortion. J Comp Pathol. 2002;126(2–3):79–94. 10. Innes EA. The host-parasite relationship in pregnant cattle infected with Neospora caninum. Parasitology. 2007;134:1903–10. 11. Robbins JR, Zeldovich VB, Poukchanski A, Boothroyd JC, Bakardjiev AI. Tissue barriers of the human placenta to infection with Toxoplasma gondii. Infect Immun. 2012;80(1):418–28. 12. Leiser R, Kaufmann P. Placental structure: in a comparative aspect. Exp Clin Endocrinol Diabetes. 1994;102(03):122–34. 13. Machado RZ, Mineo TW, Landim LP Jr, Carvalho AF, Gennari SM, Miglino MA. Possible role of bovine trophoblast giant cells in transplacental transmission of Neospora caninum in cattle. Rev Bras Parasitol Vet. 2007;1:21–5. 14. Plattner F, Soldati-Favre D. Hijacking of host cellular functions by the Apicomplexa. Annu Rev Microbiol. 2008;62:471–87. 15. Santos JM, Lebrun M, Daher W, Soldati D, Dubremetz J. Apicomplexan cytoskeleton and motors: key regulators in morphogenesis, cell division, transport and motility. Int J Parasitol. 2009;39(2):153–62. 16. Regidor-Cerrillo J, Gómez-Bautista M, Sodupe I, Aduriz G, Álvarez-García G, Del Pozo I, et al. In vitro invasion efficiency and intracellular proliferation rate comprise virulence-related phenotypic traits of Neospora caninum. Vet Res. 2011;42(1):41. 17. Dellarupe A, Regidor-Cerrillo J, Jiménez-Ruiz E, Schares G, Unzaga JM, Venturini MC, et al. Comparison of host cell invasion and proliferation among Neospora caninum isolates obtained from oocysts and from clinical cases of naturally infected dogs. Exp Parasitol. 2014;145:22–8. 18. Rojo-Montejo S, Collantes-Fernández E, Blanco-Murcia J, Rodríguez-Bertos A, Risco-Castillo V, Ortega-Mora LM. Experimental infection with a low virulence isolate of Neospora caninum at 70 days gestation in cattle did not result in foetopathy. Vet Res. 2009;40(5):49. 19. Caspe SG, Moore DP, Leunda MR, Cano DB, Lischinsky L, Regidor-Cerrillo J, et al. The Neospora caninum-Spain 7 isolate induces placental damage, fetal death and abortion in cattle when inoculated in early gestation. Vet Parasitol. 2012;189(2–4):171–81. 20. Regidor-Cerrillo J, Arranz-Solís D, Benavides J, Gómez-Bautista M, Castro-Hermida JA, Mezo M, et al. Neospora caninum infection during early pregnancy in cattle: how the isolate influences infection dynamics, clinical outcome and peripheral and local immune responses. Vet Res. 2014;45(1):10. 21. Strahl H, Die Embryonalhüllen d. Säuger und die Placenta. In: Anonymous Hertwig’s Handbuch der vergleichenden und experimentellen Entwicklungslehre der Wirbeltiere: G. Fischer Jena; 1906. p. 235–368. 22. Wooding F. The synepitheliochorial placenta of ruminants: binucleate cell fusions and hormone production. Placenta. 1992;13(2):101–13. 23. Bridger P, Haupt S, Klisch K, Leiser R, Tinneberg H, Pfarrer C. Validation of primary epitheloid cell cultures isolated from bovine placental caruncles and cotyledons. Theriogenology. 2007;68(4):592–603. 24. Bridger PS, Menge C, Leiser R, Tinneberg HR, Pfarrer CD. Bovine caruncular epithelial cell line (BCEC-1) isolated from the placenta forms a functional epithelial barrier in a polarised cell culture model. Placenta. 2007;28(11–12): 1110–7. 25. Zeiler M, Leiser R, Johnson GA, Tinneberg HR, Pfarrer C. Development of an in vitro model for bovine placentation: a comparison of the in vivo and in vitro expression of integrins and components of extracellular matrix in bovine placental cells. Cells Tissues Organs. 2007;186(4):229–42. 26. Pfarrer C, Hirsch P, Guillomot M, Leiser R. Interaction of integrin receptors with extracellular matrix is involved in trophoblast giant cell migration in bovine placentomes. Placenta. 2003;24(6):588–97. 27. Waterkotte B, Hambruch N, Doring B, Geyer J, Tinneberg HR, Pfarrer C. P-glycoprotein is functionally expressed in the placenta-derived bovine caruncular epithelial cell line 1 (BCEC-1). Placenta. 2011;32(2):146–52. 28. Hambruch N, Haeger JD, Dilly M, Pfarrer C. EGF stimulates proliferation in the bovine placental trophoblast cell line F3 via Ras and MAPK. Placenta. 2010;31(1):67–74. 29. Rojo-Montejo S, Collantes-Fernández E, Regidor-Cerrillo J, Álvarez-García G, Marugán-Hernández V, Pedraza-Díaz S, et al. Isolation and characterization of a bovine isolate of Neospora caninum with low virulence. Vet Parasitol. 2009;159(1):7–16. 30. Regidor-Cerrillo J, Gómez-Bautista M, Pereira-Bueno J, Adúriz G, Navarro- Lozano V, Risco-Castillo V, et al. Isolation and genetic characterization of Neospora caninum from asymptomatic calves in Spain. Parasitology. 2008; 135(14):1651–9. 31. Regidor-Cerrillo J, Gómez-Bautista M, Del Pozo I, Jiménez-Ruiz E, Aduriz G, Ortega-Mora LM. Influence of Neospora caninum intra-specific variability in the outcome of infection in a pregnant BALB/c mouse model. Vet Res. 2010;41(4):52. 32. Regidor-Cerrillo J, Díez-Fuertes F, García-Culebras A, Moore DP, González- Warleta M, Cuevas C, et al. Genetic diversity and geographic population structure of bovine Neospora caninum determined by microsatellite genotyping analysis. PLoS One. 2013;8(8):e72678. 33. Pérez-Zaballos FJ, Ortega-Mora LM, Álvarez-García G, Collantes-Fernández E, Navarro-Lozano V, García-Villada L, et al. Adaptation of Neospora caninum isolates to cell-culture changes: an argument in favor of its clonal population structure. J Parasitol. 2005;91(3):507–10. 34. Collantes-Fernández E, Zaballos A, Álvarez-García G, Ortega-Mora LM. Quantitative detection of Neospora caninum in bovine aborted fetuses and experimentally infected mice by real-time PCR. J Clin Microbiol. 2002;40(4): 1194–8. 35. Haldorson GJ, Mathison BA, Wenberg K, Conrad PA, Dubey JP, Trees AJ, et al. Immunization with native surface protein NcSRS2 induces a Th2 immune response and reduces congenital Neospora caninum transmission in mice. Int J Parasitol. 2005;35(13):1407–15. 36. Carvalho JV, Alves CM, Cardoso MR, Mota CM, Barbosa BF, Ferro EA, et al. Differential susceptibility of human trophoblastic (BeWo) and uterine cervical (HeLa) cells to Neospora caninum infection. Int J Parasitol. 2010; 40(14):1629–37. 37. Björkman N. Fine structure of the fetal-maternal area of exchange in the epitheliochorial and endotheliochorial types of placentation. Cells Tissues Organs. 1973;86(Suppl. 61):1–22. 38. Leiser R. Development of contact between trophoblast and uterine epithelium during the early stages on implantation in the cow. Zentralbl Veterinarmed C. 1975;4(1):63–86. 39. Bridger PS. Validation and establishment of cell culture models to study invasion and feto-maternal interaction in the bovine placentome: VVB Laufersweiler Verlag; 2008. 40. Schlafer D, Fisher P, Davies C. The bovine placenta before and after birth: placental development and function in health and disease. Anim Reprod Sci. 2000;60:145–60. Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 12 of 13 41. Amarante-Paffaro A, Queiroz GS, Correa ST, Spira B, Bevilacqua E. Phagocytosis as a potential mechanism for microbial defense of mouse placental trophoblast cells. Reproduction. 2004;128(2):207–18. 42. Bevilacqua E, Hoshida MS, Amarante-Paffaro A, Albieri-Borges A, Zago Gomes S. Trophoblast phagocytic program: roles in different placental systems. Int J Dev Biol. 2010;54(2–3):495–505. 43. Wujcicka W, Wilczyński J, Nowakowska D. Do the placental barrier, parasite genotype and Toll-like receptor polymorphisms contribute to the course of primary infection with various Toxoplasma gondii genotypes in pregnant women? Eur J Clin Microbiol Infect Dis. 2014;33(5):703–9. 44. Williams DJ, Guy CS, McGarry JW, Guy F, Tasker L, Smith RF, et al. Neospora caninum-associated abortion in cattle: the time of experimentally-induced parasitaemia during gestation determines foetal survival. Parasitology. 2000; 121(4):347–58. 45. Maley SW, Buxton D, Rae AG, Wright SE, Schock A, Bartley PM, et al. The pathogenesis of neosporosis in pregnant cattle: inoculation at mid- gestation. J Comp Pathol. 2003;129(2–3):186–95. 46. Macaldowie C, Maley SW, Wright S, Bartley P, Esteban-Redondo I, Buxton D, et al. Placental pathology associated with fetal death in cattle inoculated with Neospora caninum by two different routes in early pregnancy. J Comp Pathol. 2004;131(2–3):142–56. 47. Collantes-Fernández E, Rodríguez-Bertos A, Arnaiz-Seco I, Moreno B, Aduriz G, Ortega-Mora LM. Influence of the stage of pregnancy on Neospora caninum distribution, parasite loads and lesions in aborted bovine foetuses. Theriogenology. 2006;65(3):629–41. 48. Rosbottom A, Gibney EH, Guy CS, Kipar A, Smith RF, Kaiser P, et al. Upregulation of cytokines is detected in the placentas of cattle infected with Neospora caninum and is more marked early in gestation when fetal death is observed. Infect Immun. 2008;76(6):2352–61. 49. Rosbottom A, Gibney H, Kaiser P, Hartley C, Smith RF, Robinson R, et al. Up regulation of the maternal immune response in the placenta of cattle naturally infected with Neospora caninum. PLoS One. 2011;6(1):e15799. 50. Pereira García-Melo D, Regidor-Cerrillo J, Collantes-Fernández E, Aguado- Martínez A, Del Pozo I, Minguijon E, et al. Pathogenic characterization in mice of Neospora caninum isolates obtained from asymptomatic calves. Parasitology. 2010;137(7):1057–68. 51. Regidor-Cerrillo J, Álvarez-García G, Pastor-Fernández I, Marugán-Hernández V, Gómez-Bautista M, Ortega-Mora LM. Proteome expression changes among virulent and attenuated Neospora caninum isolates. J Proteome. 2012;75(8): 2306–18. 52. Regidor-Cerrillo J, García-Lunar P, Pastor-Fernández I, Álvarez-García G, Collantes-Fernández E, Gómez-Bautista M, et al. Neospora caninum tachyzoite immunome study reveals differences among three biologically different isolates. Vet Parasitol. 2015;212(3–4):92–9. 53. Almería S, Serrano-Pérez B, Darwich L, Domingo M, Mur-Novales R, Regidor-Cerrillo J, et al. Foetal death in naive heifers inoculated with Neospora caninum isolate Nc-Spain7 at 110 days of pregnancy. Exp Parasitol. 2016;168:62–9. • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: Jiménez-Pelayo et al. Parasites & Vectors (2017) 10:463 Page 13 of 13 Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 https://doi.org/10.1186/s13071‑019‑3466‑z RESEARCH Immune response profile of caruncular and trophoblast cell lines infected by high‑ (Nc‑Spain7) and low‑virulence (Nc‑Spain1H) isolates of Neospora caninum Laura Jiménez‑Pelayo1†, Marta García‑Sánchez1†, Javier Regidor‑Cerrillo1, Pilar Horcajo1, Esther Collantes‑Fernández1, Mercedes Gómez‑Bautista1, Nina Hambruch2, Christiane Pfarrer2 and Luis Miguel Ortega‑Mora1* Abstract Background: Bovine neosporosis, one of the main causes of reproductive failure in cattle worldwide, poses a chal‑ lenge for the immune system of pregnant cows. Changes in the Th‑1/Th‑2 balance in the placenta during gestation have been associated with abortion. Cotyledon and caruncle cell layers form the maternal‑foetal interface in the pla‑ centa and are able to recognize and induce immune responses against Neospora caninum among other pathogens. The objective of the present work was to elucidate the immunomodulation produced by high‑ (Nc‑Spain7) and low‑ virulence (Nc‑Spain1H) isolates of N. caninum in bovine trophoblast (F3) and caruncular cells (BCEC‑1) at early and late points after infection. Variations in the mRNA expression levels of toll‑like receptor‑2 (TLR‑2), Th1 and Th2 cytokines (IL‑4, IL‑10, IL‑8, IL‑6, IL‑12p40, IL‑17, IFN‑γ, TGF‑β1, TNF‑α), and endothelial adhesion molecules (ICAM‑1 and VCAM‑1) were investigated by RT‑qPCR, and protein variations in culture supernatants were investigated by ELISA. Results: A similar pattern of modulation was found in both cell lines. The most upregulated cytokines in infected cells were pro‑inflammatory TNF‑α (P < 0.05–0.0001) and IL‑8 (P < 0.05–0.001) whereas regulatory IL‑6 (P < 0.05–0.001) and TGF‑β1 (P < 0.05–0.001) were downregulated in both cell lines. The measurement of secreted IL‑6, IL‑8 and TNF‑α confirmed the mRNA expression level results. Differences between isolates were found in the mRNA expression levels of TLR‑2 (P < 0.05) in both cell lines and in the mRNA expression levels (P < 0.05) and protein secretion of TNF‑α (P < 0.05), which were higher in the trophoblast cell line (F3) infected with the low‑virulence isolate Nc‑Spain1H. Conclusions: Neospora caninum infection is shown to favor a pro‑inflammatory response in placental target cells in vitro. In addition, significant immunomodulation differences were observed between high‑ and low‑virulence isolates, which would partially explain the differences in virulence. Keywords: Neospora caninum, Cattle, Immune response, Placenta, Trophoblast, Caruncle, Isolates, Virulence, Cytokines © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Open Access Parasites & Vectors *Correspondence: luis.ortega@ucm.es †Laura Jiménez‑Pelayo and Marta García‑Sánchez contributed equally to this work 1 SALUVET, Animal Health Department, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain Full list of author information is available at the end of the article http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/publicdomain/zero/1.0/ http://creativecommons.org/publicdomain/zero/1.0/ http://crossmark.crossref.org/dialog/?doi=10.1186/s13071-019-3466-z&domain=pdf Page 2 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 Background Bovine neosporosis is one of the main transmissible causes of abortion in cattle worldwide [1–3]. The etio- logical agent of bovine neosporosis is Neospora caninum, an obligate intracellular parasite closely related to the zoonotic agent Toxoplasma gondii. Transplacental trans- mission is the main route of transmission in cattle [4] and the placenta can play a key role in the pathogenesis of bovine neosporosis [5, 6]. The direct damage produced by the multiplication of the parasite in placental and foetal tissues has been proposed as one of the possible causes of abortion observed during N. caninum infections. Impor- tantly, the placenta is considered to be an immune reg- ulatory organ since it acts as a modulator of foetal and maternal immune responses. In fact, an immune-medi- ated pathogenesis has also been suggested as a possible cause of abortion [7]. It has been shown that the multi- plication of the parasite in the placenta alters the immu- nological balance at the maternal-foetal interface with an increase of local pro-inflammatory IFN-γ, IL-12p40 and TNF-α cytokines which could compromise the gestation, together with an increase in IL-4 and IL-10 levels [8, 9], which avoids the immunological rejection of the foetus but favours the multiplication and vertical transmission of the parasite [5, 10]. Trophoblast and caruncular cells are able to recognize pathogens and secrete cytokines and chemokines that recruit immune cells in the dam- aged area [11–13]. Thus, both cell types play a funda- mental role in the initiation of innate immune responses at the placental level as well as in the development of an adaptative immune response for the pregnant dam and foetus. Previous in  vivo studies have shown the influence of the isolate on the dynamics and outcome of the infection in pregnant bovine models and in the cytokine profiles induced during the infection ([9, 14–16], Jiménez-Pelayo et  al. unpublished data). To date, only one recent study has utilized an in vitro model consisting of immortalized bovine trophoblasts (F3) from the foetus and caruncular cells (BCEC-1) from the dam. The aim of the study was to elucidate the interactions between tachyzoites and the host cells that resemble the maternal-foetal interface of the bovine placentome while also taking into account the influence of the isolate. Maternal cells, where both isolates showed similar phenotypic traits, presented higher resistance to the infection than trophoblast cells, where the high- (Nc-Spain7) and the low-virulence (Nc- Spain1H) isolates showed marked differences in prolif- eration [17]. However, the interactions between the parasite and the placental target cells from an immunological point of view have not been investigated in  vitro until now. Thus, the objective of the present study was to compare the immune response profiles of the bovine placental cells in  vitro after the infection with two N. caninum isolates of different virulence. Messenger RNA expres- sion levels of TLR-2, pro-inflammatory cytokines IL-8, IL-12p40, IL-17, IFN-γ, TNF-α, anti-inflammatory/regu- latory cytokines TGF-β1, IL-4, IL-6 and IL-10 as well as ICAM-1 and VCAM-1 endothelial adhesion molecules were determined at 4 and 24 hours post-infection (hpi) in maternal caruncular (BCEC-1) and foetal trophoblast (F3) cell cultures and protein secretion was assessed in culture supernatants by ELISA. Results Expression profile of TLR‑2 Our results showed that N. caninum infection for 4 h in BCEC-1 cells resulted in a significant upregulation of TLR-2 expression in Nc-Spain1H-infected cells compared with that of negative control cells (Kruskal–Wallis H-test followed by Dunnʼs multiple comparison test: χ2 = 16.2, df = 2, P = 0.0001) and with that of BCEC-1 cells infected with the high-virulence isolate Nc-Spain7 (Mann–Whit- ney U-test: U(8) = 8, Z = 2.591, P = 0.0007). In F3 cultures, statistical significance was not found at either 4 or 24 hpi between infected groups and the control group. However, comparing both isolates, lower expression of TLR-2 was found in the F3 cultures infected with Nc-Spain7 than in the F3 cultures infected with the low-virulence isolate Nc-Spain1H at 4  hpi (Mann–Whitney U-test: U(8) = 16, Z = 2.287, P = 0.0315) (Fig. 1a). Pro‑inflammatory and regulatory cytokine modulation The pro-inflammatory cytokines IL-8 (Kruskal–Wal- lis H-test: χ2 = 19.52, df = 2, P < 0.0001 in BCEC-1 and χ2 = 17.56, df = 2, P = 0.0002 in F3) and TNF-α (Kruskal– Wallis H-test: χ2 = 19.73, df = 2, P < 0.0001 in BCEC-1 and χ2 = 19.4, df = 2, P < 0.0001 in F3) were upregulated in both cell types at 4 hpi compared to the respective control groups (Fig.  1b, c). At 24 hpi, IL-8 expression was still increased in BCEC-1 cells infected by both iso- lates (Kruskal–Wallis H-test followed by Dunnʼs mul- tiple comparison test: χ2 = 16.63, df = 2, P = 0.0003 and χ2 = 10.19, df = 2, P = 0.0117 for Nc-Spain7 and Nc- Spain1H, respectively); however, the increment of IL-8 had disappeared at 24 hpi in F3-infected cells with respect to the control group. When both isolates were compared, Nc-Spain1H induced a higher expression of TNF-α than the high-virulence isolate Nc-Spain7 at 4 hpi in F3 cells (Mann–Whitney U-test: U(8) = 0, Z = 2.579, P < 0.0001). Protein levels of the pro-inflammatory cytokines IL-8 and TNF-α were also investigated in the supernatant of control and infected cultures at different time-points post-infection. A higher secretion of IL-8 was found for both isolates in BCEC-1 cells at 24  hpi (Kruskal–Wallis Page 3 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 Fig. 1 TLR‑2, IL‑8, TNF‑α, IL‑6, IL‑12p40, TGF‑β1, ICAM‑1 and VCAM‑1 transcript expression. Scatter‑plot graphs of relative mRNA expression levels (as x‑fold change) of TLR‑2 (a), IL‑8 (b), TNF‑α (c), IL‑6 (d), IL‑12p40 (e), TGF‑β1 (f), ICAM‑1 (g) and VCAM‑1 (h) in F3 and BCEC‑1 cell cultures at 4 and 24 hpi with Nc‑Spain7 and Nc‑Spain1H isolates. Data are represented as individual points. Horizontal lines represent median values for each group. ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. Unbracketed symbols represent differences with respect to the control group, while significant differences between isolates are denoted by horizontal square brackets Page 4 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 H-test: χ2 = 15.87, df = 2, P = 0.0004) and in F3 cells at 56 hpi (Kruskal–Wallis H-test: χ2 = 13.74, df = 2, P = 0.001) with respect to the control group (Fig. 2a, b). Secretion of TNF-α was higher in BCEC-1 cells infected by both iso- lates (Kruskal–Wallis H-test: χ2 = 18.9, df = 2, P < 0.0001; Fig. 2c) and in F3 cells infected by Nc-Spain1H (Kruskal– Wallis H-test followed by Dunnʼs multiple comparison test: χ2 = 16, df = 2, P < 0.0001) at 8 hpi, although an ear- lier secretion of TNF-α was also found in F3 cells infected by Nc-Spain1H (χ2 = 14, df = 2, P = 0.0003) at 4 hpi (Fig. 2d, e). As observed with the TNF-α mRNA expres- sion, Nc-Spain1H induced a higher secretion of TNF-α than did the high-virulence isolate Nc-Spain7 at 4 hpi (Mann–Whitney U-test: U(8) = 0, Z = 2.736, P = 0.0002) and at 8 hpi (U(8) = 0, Z = 2.305, P = 0.0002) in the F3 cul- tures (Fig. 2d, e). The expression levels of other important cytokines associated with N. caninum infection, such as IL-12p40 and IL-6 (Fig. 1d, e), were modified in placental cells after parasite infection. Specifically, IL-6 levels were down- regulated in BCEC-1 infected by Nc-Spain1H and Nc- Spain7 at 4 hpi (Kruskal–Wallis H-test: χ2 = 16.08, df = 2, P = 0.0003) and F3 cultures infected by Nc-Spain1H at 24  hpi (χ2 = 10.5, df = 2, P = 0.0052). IL-12p40 was also Fig. 2 IL‑8, TNF‑α and IL‑6 secretion levels in culture supernatants. Scatter‑plot graphs representing the concentration of IL‑8 (pg/ml) in BCEC‑1 (a) and F3 (b) supernatants infected with Nc‑Spain7 and Nc‑Spain1H at 24 and 56 hpi, respectively, the concentration of TNF‑α (pg/ml) in the BCEC‑1 supernatants at 8 hpi (c) and in the F3 supernatants at 4 hpi (d) and 8 hpi (e), and the concentration of IL‑6 (pg/ml) in the BCEC‑1 supernatants at 4 hpi (f). Data are represented as individual points. Horizontal lines represent median values for each group. ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. Unbracketed symbols represent differences with respect to the control group, while significant differences between isolates are denoted by horizontal square brackets Page 5 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 downregulated but only in infected F3 cultures at 4  hpi (χ2 = 12.99, df = 2, P = 0.0015). Differences between iso- lates in the modulation of IL-6 and IL-12p40 were not found. In addition, we observed that the caruncular cell layer did not express IL-12p40 mRNA at any time point. The decrease in the expression of IL-6 observed in infected BCEC-1 cells was confirmed by the decrease in the secretion levels of that protein found in the supernatants from BCEC-1 cultures infected with both isolates at 4 hpi, although statistically significant differ- ences were not found (Kruskal–Wallis H-test: χ2 = 2.765, df = 2, P = 0.251), probably because of the high deviation between samples (Fig. 2f ). Finally, pro-inflammatory IL-17 and IFN-γ responses were not detected in any cell lines at 4 nor at 24 hpi. We also studied the mRNA levels of the anti-inflamma- tory cytokines TGF-β1, IL-4 and IL-10. Remarkably, we observed a decrease in the expression levels of TGF-β1 in both cell lines infected with both isolates. Specifically, a decrease was observed at 4 hpi in F3 cultures (Kruskal– Wallis H-test: χ2 = 18.44, df = 2, P < 0.0001) and at 24 hpi in BCEC-1 cultures (χ2 = 12.02, df = 2, P = 0.0025) (Fig. 1f ). No differences between isolates were observed in the mRNA expression levels of TGF-β1. There was not detection in bovine placental cells of the anti-inflamma- tory cytokine IL-4 or the regulatory cytokine IL-10 at 4 and 24 hpi. Endothelial adhesion molecule (ICAM‑1 and VCAM‑1) expression The adhesion molecule ICAM-1 was expressed by both cell lines at 4 and 24 hpi. However, only a slight decrease in the mRNA expression levels of ICAM-1 was observed in the BCEC-1 cultures infected with Nc-Spain7 at 24 hpi compared to the control group although statistical signif- icance was not found (Kruskal–Wallis H-test: χ2 = 5.894, df = 2, P = 0.0525) (Fig.  1g). VCAM-1 expression was detected only in the F3 cultures at 24 hpi, but differences between the infected and the control groups were not found in this culture at this time point (Fig. 1h). Results of mRNA expression levels and protein secre- tion from statistical tests are reported in Additional file 1: Tables S1 and S2, respectively. Discussion Transmission of N. caninum across the placenta makes this organ key in the pathogenesis of bovine neosporosis. Innate immune signalling is crucial at the maternal-foetal interface, where vertical transmission of pathogens to the foetus can have profound pathological outcomes. Troph- oblasts and other cell types within the placenta may also be involved in the physiological protection of the pla- centa [18]. Trophoblast cells have been shown to respond to some infections by producing pro-inflammatory cytokines and chemokines and endometrial or decidual cells can produce and secrete a variety of cytokines, par- ticipating in the attraction and activation of immune effector cells [19, 20]. However, these innate immune mechanisms are unexplored at the maternal-foetal inter- face during N. caninum infection in pregnant cattle [21]. The expression of TLRs has been described in tropho- blasts and other cell types within the placenta [22]. Spe- cifically, TLR-2 was overexpressed in bovine trophoblast cell cultures at 8 hpi [23] and TLR-3, 7 and 8 have been implicated in N. caninum recognition in the bovine pla- centa [21, 24]. In our study, differential activation of TLR-2 in the F3 and BCEC-1 cultures was observed. An upregulation of TLR-2 was found in BCEC-1-infected cultures, especially in those infected with the low-vir- ulence isolate Nc-Spain1H. The caruncular part of the placentome showed a higher expression of several TLRs, suggesting that the initial recognition of N. caninum at the placental level would occur in the maternal side of placenta [21]. Taking into account the data shown in Jiménez-Pelayo et al. [17], which confirm the higher pro- liferation of N. caninum in trophoblast cells, an impor- tant role of placental TLR-2 in the immune response against N. caninum seems plausible. TLR activation is crucial for initiating the innate immune responses responsible for the elimination of intracellular parasites such as N. caninum, and the signalling pathway activated by TLR-2 leads to an increase in the transcription fac- tors NF-κβ and AP-1, which trigger the synthesis of pro- inflammatory cytokines (TNF-α, IL-6, IL-12 and IL-1β) and chemokines (IL-8, RANTES) [25]. Despite differential modulation of host TLR-2, both cell types presented a similar variation in the IL-6, TNF-α and IL-8 expression levels in infected cultures. The pro-inflammatory IL-8 and TNF-α cytokines were upregulated, and secretion of the proteins in the super- natants of both cell lines was also detected by ELISA. IL-8, a cytokine with neutrophil chemotactic/activat- ing and T-cell chemotactic activity both in  vivo and in  vitro, is important in the recruitment of leukocytes to the endometrium and may be a potential mediator of placental macrophage infiltration [26], which might help to eliminate the parasite. IL-8 upregulation has already been observed in bovine umbilical vein endothe- lial cells (BUVECs) infected by T. gondii and N. caninum [27] as well as in bovine trophoblastic cells and placen- tomes from cows infected with Brucella abortus [28]. TNF-α is an inflammatory cytokine whose expression has also been described for epithelial cells [29]. TNF-α is expressed in all cell types of the trophoblastic lineage and provokes a variety of biological effects on placental and endometrial cell types [29]. In addition, TNF-α, through Page 6 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 the NF-κβ signalling pathway, coordinates the inflamma- tory response via the induction of other cytokines (IL-1 and IL-6) and chemokines (IL-8) and via the upregulation of adhesion molecules (ICAM-1 and VCAM-1) [30, 31], playing a role favouring protective immunity in infec- tious diseases [32]. There are several lines of experimen- tal evidence indicating that TNF-α plays a role not only in immunity to N. caninum but also in the immunopa- thology of neosporosis. TNF-α expression and secre- tion may reduce the parasite presence in the placenta by inhibiting the intracellular multiplication of the parasite [33] and participating in parasite proliferation control mechanisms [34]; however, TNF-α expression is detri- mental to pregnancy maintenance [8]. IL-6 expression levels were diminished in infected cul- tures of BCEC-1 and F3 at 4 and 24 hpi, respectively. The classification of IL-6 as Th1 or Th2 has been considered controversial since it can have characteristics of both depending on the dose, the cellular source and the gesta- tional stage studied [35]. Currently, the presence of IL-6 displaces the Th1/Th2 balance towards a Th2 response [36]. However, in  vivo models of N. caninum infection have shown IL-6 upregulation [37, 38, Jiménez-Pelayo et  al. unpublished data]. This response pattern may be related to a protective action that protects the foetus and allows gestation even if the animals are born infected [10]. The decrease in IL-6 observed could be explained by the following: (i) IL-6 expression levels were affected by the high antigenic dose administered (MOI 10), result- ing in downregulation [39]; (ii) the time points were not adequate for detecting the peak of IL-6 expression, and the observed decrease may be the consequence of the rapid reduction in IL-6 expression after a peak of expres- sion [40–42]; or (iii) other cell types are implicated in the upregulation of IL-6 that was observed in vivo. The anti-inflammatory cytokine TGF-β1 was also found to be downregulated in F3 and BCEC-1 cultures at 4 and 24 hpi, respectively. Several members of the TGF-β superfamily have been suggested to regulate trophoblast cell functions, and their dysregulation has been impli- cated in pregnancy-associated diseases. TGF-β1 is cru- cial in neutralizing the inflammatory responses induced by Th1-type cytokines [5]. This effect has already been observed in previous works where the reduction of TGF- β1 was shown to be beneficial for controlling N. caninum growth but detrimental for the adequate maintenance of pregnancy [38, 43]. The reduction of pro-inflammatory IL-12p40 observed in trophoblast cells, together with the lack of expres- sion of IL-12p40 in BCEC-1 cultures, disagrees with the results of previous experimental infections [8, 9, 38, 44, 45]. Similarly, the lack of expression of pro-inflammatory IFN-γ, essential for controlling parasite infection [1, 46], and anti-inflammatory IL-4 and IL-10, related to placen- tal protection during N. caninum infections [8, 9], lead us to hypothesize that the upregulation of IL-12p40, IFN-γ, IL-4 and IL-10 observed in  vivo could be attributed to immune cells present in the placenta, such as dendritic cells, NK cells or macrophages. Therefore, trophoblast and/or caruncular cells would not be responsible for the direct production of these cytokines, although the assayed time points may not have been appropriate for their detection. Finally, ICAM-1 and VCAM-1 expression were not modulated by the parasite infection. These adhesion molecules participate in the recruitment of inflamma- tory immune cells [47] and promote the adherence of monocytes to endothelial cells [48]. The upregulation of ICAM-1 has been observed in in  vitro infections with apicomplexan parasites [27, 49, 50]. The absence of mod- ulation observed in this work may be explained by dif- ferences in the timing of the expression of ICAM-1 and VCAM-1 [27, 49, 50] or by the lack of stimuli such as the acute-phase protein C-reactive protein (CRP), which is produced by the liver in response to IL-6 [51]. As mentioned above, the parasite isolate is a key fac- tor in the outcome of the infection. In general, differences in the modulation between high- and low-virulence iso- lates were not remarkable in trophoblast or caruncular cells, with the exception of the mRNA expression levels of TLR-2 and TNF-α. TLR-2 levels were more upregu- lated by Nc-Spain1H infection than by Nc-Spain7 infec- tion in both cell lines, which led us to hypothesize that the high-virulence isolate would activate less of the TLR recognition system, reducing the immune responses trig- gered by TLR-2. The inhibition of the TLRs implicated in the recognition of Trypanosoma cruzi and T. gondii in HPCVE increased the parasite burden and, impor- tantly, TLR-2 inhibition prevented the secretion of IL-6 and IL-10, increasing parasite damage [52, 53]. The low- virulence isolate Nc-Spain1H activates the expression of TLR-2, starting an inflammatory response, which may be the cause of the lower proliferation of this isolate [17, 54] in addition to being one of the causes that explains the higher levels of TNF-α in Nc-Spain1H-infected cells, especially in trophoblast cells. Our results suggest that differential activation of the TLRs by the isolates of dif- fering virulences should be subject to future research since they may be responsible for the biological differ- ences observed both in vitro and in vivo. The low-virulence isolate Nc-Spain1H also induced higher expression of TNF-α in F3. A higher TNF-α response may more efficiently control the proliferation of Nc-Spain1H in F3 cultures, which could explain the observations made by Jiménez-Pelayo et  al. [17] where lower proliferation of Nc-Spain1H was observed in these Page 7 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 cells. The lower expression of TNF-α observed during the early stage of infection of trophoblasts with the high- virulence isolate Nc-Spain7 supports the hypothesis that this isolate may modify by yet unknown mechanisms the pro-inflammatory response by trophoblast cells. However, how Nc-Spain7 is able to evade the immune response and maintain lower levels of TNF-α expression in F3 remains unknown. On the other hand, these results suggest that pro-inflammatory cytokines such as TNF-α could have a minor impact in placental damage than pos- tulated in previous works [8, 55], but other mechanisms should be implicated in placental damage in vivo and the occurrence of abortion, such as the high multiplication ability showed by the high-virulence isolate Nc-Spain7 [17]. Conclusions The results presented in this manuscript suggest that pla- cental cells participate in the innate immune response at the maternal-foetal interface via a rapid pro-inflam- matory response characterized by the overexpression of IL-8 and TNF-α and the downregulation of TGF-β1 and IL-6. Slight differences were detected when the immu- nomodulatory response induced by the high and low virulent N. caninum isolates was compared. The higher expression of TLR-2 in the F3 and BCEC-1 cells and the TNF-α in F3 cells infected with the low-virulence iso- late Nc-Spain1H may indicate a higher stimulation of the immune response by this isolate or a higher immu- nomodulation of Nc-Spain7, which could explain the bio- logical differences observed in vitro and in vivo. F3 and BCEC-1 cultures seem to be a good tool for the study of the TLR activation mechanisms by N. caninum. Finally, we observed that cytokines such as IFN-γ, IL-4 or IL-10, which are commonly upregulated in the placenta after N. caninum infection, are not expressed in F3 and BCEC-1 cells; we conclude that the trophoblast and caruncular epithelial cells are not implicated in the production of these cytokines in the placenta or that other pathways/ cells/molecules are needed for their production. Methods Parasites and cell cultures A full description of the Nc-Spain1H and Nc-Spain7 parasites and cell cultures of bovine caruncular epithe- lial (BCEC-1) and bovine trophoblast cells (F3) is pro- vided in a previous report [17]. Briefly, Nc-Spain7 and Nc-Spain1H isolates were obtained from healthy, con- genitally infected calves [56, 57] and tachyzoites were maintained in a MARC-145 culture as described previ- ously [54]. The number of culture passages of both N. caninum isolates was limited (passages from 9 to 11) to maintain their biological in vivo behavior [58]. The BCEC-1 and F3 cell lines were kindly donated by Dr C. Pfarrer from the University of Veterinary Medi- cine Hannover and maintained following the protocols described in the literature [59, 60]. Infection of the cultures, collection and preservation of the samples BCEC-1 and F3 cells were seeded in 25 cm2 culture flasks adjusting the number of cells in order to obtain a con- fluent monolayer after 24  h of culture. F3 was seeded at 106 cells per flask, whereas BCEC-1 was seeded at a concentration of 2 × 106 cells per flask. Tachyzoites were recovered from MARC-145 cultures when most of the parasites were still inside parasitophorous vacuoles; tachyzoites were purified using disposable PD-10 Desalt- ing Columns (G.E. Healthcare, Amersham, UK) as previ- ously described [54]. The parasite viability was checked by trypan blue exclusion, and the tachyzoites were counted. Multiplicity of infection (MOI) of 8 (8 × 106 tachyzoites in F3 and 16 × 106 tachyzoites in BCEC-1) and 10 (107 tachyzoites in F3 and 2 × 107 tachyzoites in BCEC-1) from the Nc-Spain7 and Nc-Spain1H isolates, respectively, were inoculated into confluent monolayers of F3 and BCEC-1 quickly after collection. Due to the dif- ferences observed in the infection rate between isolates [17], different MOIs of each isolate were selected with the aim of obtaining cultures infected with the same quantity of each parasite at 4 and 24 hpi. This way possible differ- ences in the modulation of the mRNA expression levels between isolates could be attributed to differences in their biological behavior and not to the differences in the parasite burden. In addition, cultures were infected with high doses of both parasites to get a high infection of the cultures at 4 and 24 hpi so that the RNA from uninfected cells did not mask possible differences in RNA expression levels induced by the infection. The flasks were incubated at 37  °C until collection of the samples. The superna- tants were collected at different time points (4, 8, 24 and 56 hpi) and stored at −80 °C for the detection of proteins by ELISA. The cultures were harvested at 4 or 24 hpi by scraping, centrifugation at 1350×g for 15 min at 4 °C and resuspending the pellet in 300 µl of RNAlater® (Qiagen, Hilden, Germany). The samples were stored at −80  °C prior to RNA extraction. Two independent experiments were carried out and four replicates were obtained in each experiment. RNA extraction, reverse transcription and quantitative real‑time PCR The mRNA expression levels of TLR-2, pro-inflamma- tory cytokines IL-6, IL-8, IL-12p40, IL-17, IFN-γ, TNF- α, anti-inflammatory/regulatory cytokines TGF-β1, IL-4 and IL-10 as well as ICAM-1 and VCAM-1 endothelial Page 8 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 adhesion molecules were determined by real-time RT- PCR in the F3 and BCEC-1 cell layers infected with the high-virulence isolate (Nc-Spain7) and the low-virulence isolate (Nc-Spain1H) of N. caninum at an early (4  hpi) and a late (24 hpi) time point. RNA was extracted using a commercial Maxwell® 16 LEV simplyRNA Purification kit (Promega, Madi- son, WI, USA) following the manufacturer’s recom- mendations. RNA integrity was checked by 1% agarose gel and RNA concentrations were determined using a NanoPhotometer® spectrophotometer (Implen, Munich, Germany). cDNA was obtained by reverse transcription of 2.5  µg of RNA using the master mix SuperScript® VILO™ cDNA Synthesis kit (Invitrogen, Paisley, UK), which was diluted 1:20 in molecular grade water for the qPCR assays. The PCRs were performed using 12.5  µl of Power SYBR® Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA), 10 pmol of each primer (except for TLR-2 primers which were used at a concentration of 22.5  pmol) and 5 μl of diluted cDNA samples in an ABI 7300 Real Time PCR System (Applied Biosystems). The primers used for the qPCR reactions are shown in Table 1. β-Actin and GAPDH were used as housekeeping Table 1 Sequences of primers used for cytokine real‑time PCR (qPCR) and standard curve data a NCBI accession numbers are for cDNA sequences used in primer design. Primer annealing was also checked with the Bos taurus genomic DNA sequences (http:// www.ncbi.nlm.nih.gov/nucco re) b Minimum coefficient of regression (R2) of standard curves for each PCR target in all batches of amplification c Standard curve slopes. Minimum and maximum values for slopes for each PCR target in all batches of amplification d Primer first described by Arraz‑Solís et al. [43] e Primer first described by Menzies & Ingham [61] f Primer first described by Puech et al. [62] g Primer first described by Regidor‑Cerrillo et al. [9] h Primer described in the present work for the first time Targeta Primer Primer sequence (5′–3′) Product size (bp) R2 b Slopec IFN‑γ (NM_174086.1) QIFN‑UPg GAT TCA AAT TCC GGT GGA TG 110 0.994 (−3.47)–(−3.30) QIFN‑RPg TTC TCT TCC GCT TTC TGA GG TNF‑α (EU276079.1) QTNF‑UPg CCA GAG GGA AGA GCA GTC C 126 0.998 (−3.39)–(−3.27) QTNF‑RPg GGA GAG TTG ATG TCG GCT AC IL‑4 (M77120.1) QIL4‑UPg CTG CCC CAA AGA ACA CAA CT 169 0.995 (−3.33)–(−3.54) QIL4‑RPg GTG CTC GTC TTG GCT TCA TT IL‑6 (X68723.1) QIL‑6‑UPd CTG GGT TCA ATC AGG CGA TT 150 0.999 (−3.22)–(−3.20) QIL‑6‑RPd GGA TCT GGA TCA GTG TTC TGA IL‑8 (BC103310.1) qIL8‑Fwh CCA CAC CTT TCC ACC CCA AA 177 0.995 (−3.36)–(−3.23) qIL8‑Rwh CTT GCT TCT CAG CTC TCT TC IL‑10 (NM_174088.1) QIL10‑UPg TGC TGG ATG ACT TTA AGG GTT ACC 60 0.999 (−3.27)–(−3.42) QIL10‑RPg AAA ACT GGA TCA TTT CCG ACAAG IL‑12p40 (NM_174356.1) QIL12‑UPg AGT ACA CAG TGG AGT GTC AG 157 0.992 (−3.39)–(−3.35) QIL12‑RPg TTC TTG GGT GGG TCT GGT TT IL‑17 (NM_001008412.1) qIL17bov‑uph GAA CTT CAT CTA TGT CAC TGC 83 0.997 (−3.30)–(−3.18) qIL17bov‑revh TGG ACT CTG TGG GAT GAT GA TGF‑β1 (NM_001009400.1) QTGF‑UPd GGT GGA ATA CGG CAA CAA AA 117 0.999 (−3.60)–(−3.53) QTGF‑RPd CGA GAG AGC AAC ACA GGT TC TLR‑2 (NM_001048231.1) QTLR2‑UPe ACG ACG CCT TTG TGT CCT AC 192 0.993 (−3.74)–(−3.38) QTLR2‑RPe CCG AAA GCA CAA AGA TGG TT ICAM‑1 (NM_174348.2) qICAM‑Fwh AGA CCT ATG TCC TGC CAT CG 219 0.994 (−3.34)–(−3.30) qICAM‑Rwh GGT GCC CTC CTC ATT TTC CT VCAM (XM_005204079.2) qVCAM‑Fwh GAA CTG GAA GTC TAC ATC TC 128 0.998 (−3.36)–(−3.32) qVCAM‑Rwh CAG AGA ATC CGT GGA GCT GG GAPDH (NM_001034034) GAPDH‑Ff ATC TCG CTC CTG GAA GAT G 227 0.996 (−3.67)–(−3.58) GAPDH‑Rf TCG GAG TGA ACG GAT TCG β‑Actin (NM_173979.3) BACTIN‑UPg ACA CCG CAA CCA GTT CGC CAT 216 0.994 (−3.45)–(−3.36) BACT216‑RPg GTC AGG ATG CCT CTC TTG CT http://www.ncbi.nlm.nih.gov/nuccore http://www.ncbi.nlm.nih.gov/nuccore Page 9 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 genes, obtaining comparable Ct values for all the sam- ples. For each target gene, a seven-point standard curve was included in each batch of amplifications based on 10-fold serial dilutions starting at 10  ng/µl of plasmid DNA. The relative quantification of the mRNA expres- sion levels (x-fold change in expression) was carried out by the comparative 2−ΔΔCt method [63]. Measurement of cytokines in supernatants of BCEC‑1 and F3 cell cultures by ELISA Protein concentrations of the cytokines that showed vari- ations in the mRNA expression levels were determined in the culture supernatants at 4, 8, 24 and 56 hpi using com- mercial ELISA kits. The levels of IL-6, IL-8 and TNF-α cytokines were measured in the supernatants of the BCEC-1 and F3 cells by sandwich ELISAs using a Bovine IL-6 ELISA Reagent kit (ESS0029; Thermo Fisher Scien- tific, Waltham, MA, USA), Bovine IL-8 (CXCL8) ELISA Development kit (3114-1A-6; Mabtech AB, Stockholm, Sweden) and Bovine TNF-α ELISA kit (EBTNF; Thermo Fisher Scientific) following the manufacturers’ instruc- tions. The sensitivity limits of these assays were 78 pg/ml for IL-6, 25 pg/ml for IL-8 and 100 pg/ml for TNF-α. Statistical analysis TLR, cytokine and endothelial adhesion molecule mRNA expression levels, as well as differences in the protein secretion between infected and control groups, were ana- lysed using the non-parametric Kruskal–Wallis test, fol- lowed by Dunn’s multiple comparison test for all pairwise comparisons. In addition, to assess differences between both infected groups a Mann–Whitney test was per- formed for each molecule analysed. The statistical signifi- cance for all the analyses was established with P < 0.05. GraphPad Prism v.5.01 software (GraphPad Software, San Diego, CA, USA) was used to perform all statistical analyses and create all the graphical illustrations. Additional file Additional file 1:Table S1. Statistical test results for mRNA expression levels. Table S2. Statistical test results for protein secretion. Abbreviations BCEC‑1: bovine caruncular epithelial cell line 1; F3: bovine placental tropho‑ blast cell line; qPCR: real‑time polymerase chain reaction; hpi: hours post‑infec‑ tion; BVD: bovine viral diarrhea; DMEM: Dulbecco’s modified Eagle medium; FCS: foetal calf serum; PBS: phosphate buffered saline; TLR‑2: Toll‑like Receptor 2; IL: interleukin; IFN: interferon; TGF: transforming growth factor; TNF: tumoral necrosis factor; ICAM: intercellular adhesion molecule; VCAM: vascular cell adhesion molecule; ELISA: enzyme‑linked immunosorbent assay; BUVECs: bovine umbilical vein endothelial cells; HPCVE: human placental chorionic villi explants; MOI: multiplicity of infection. Acknowledgements Not applicable. Funding This work was supported by the Spanish Ministry of Economy and Competi‑ tiveness (AGL2013‑44694‑R) and the Community of Madrid (PLATESA2‑CM P2018/BAA‑4370). Laura Jiménez‑Pelayo was financially supported by a fellow‑ ship from the Complutense University of Madrid and Marta García‑Sánchez was financially supported through a grant from the Spanish Ministry of Economy and Competitiveness (BES‑2014‑070723). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Availability of data and materials Not applicable. Authors’ contributions JRC, PH, ECF, MGB and LMO conceived the study and participated in its design. LJP and MGS wrote the manuscript, with interpretation of results and discussion input from JRC, ECF, MGB, LMO, NH and CP. LJP and MGS performed in vitro infection of the cultures, collection of the samples and ELISA assays. JRC, PH, LJP and MGS designed and performed RT‑qPCR analyses. NH and CP isolated bovine trophoblast and caruncular cell lines used in the assays. LJP and MGS carried out statistical analyses and interpreted the results. All authors read and approved the final manuscript. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Author details 1 SALUVET, Animal Health Department, Complutense University of Madrid, Ciudad Universitaria s/n, 28040 Madrid, Spain. 2 Department of Anatomy, Uni‑ versity of Veterinary Medicine Hannover, Bischofsholer Damm 15, 30173 Han‑ nover, Germany. Received: 4 January 2019 Accepted: 29 April 2019 References 1. Innes EA, Wright S, Bartley P, Maley S, Macaldowie C, Esteban‑Redondo I, et al. The host‑parasite relationship in bovine neosporosis. Vet Immunol Immunopathol. 2005;108:29–36. 2. Dubey JP, Buxton D, Wouda W. Pathogenesis of bovine neosporosis. J Comp Pathol. 2006;134:267–89. 3. Dubey JP, Schares G, Ortega‑Mora LM. Epidemiology and control of neosporosis and Neospora caninum. Clin Microbiol Rev. 2007;20:323–67. 4. Williams DJ, Hartley CS, Bjorkman C, Trees AJ. Endogenous and exog‑ enous transplacental transmission of Neospora caninum—how the route of transmission impacts on epidemiology and control of disease. Parasitology. 2009;136:1895–900. 5. Entrican G. Immune regulation during pregnancy and host‑pathogen interactions in infectious abortion. J Comp Pathol. 2002;126:79–94. 6. Innes EA. The host‑parasite relationship in pregnant cattle infected with Neospora caninum. Parasitology. 2007;134:1903–10. 7. Quinn HE, Ellis JT, Smith NC. Neospora caninum: a cause of immune‑ mediated failure of pregnancy? Trends Parasitol. 2002;18:391–4. 8. Rosbottom A, Gibney EH, Guy CS, Kipar A, Smith RF, Kaiser P, et al. Upregu‑ lation of cytokines is detected in the placentas of cattle infected with https://doi.org/10.1186/s13071-019-3466-z Page 10 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 Neospora caninum and is more marked early in gestation when fetal death is observed. Infect Immun. 2008;76:2352–61. 9. Regidor‑Cerrillo J, Arranz‑Solis D, Benavides J, Gomez‑Bautista M, Castro‑Hermida JA, Mezo M, et al. Neospora caninum infection during early pregnancy in cattle: how the isolate influences infection dynamics, clinical outcome and peripheral and local immune responses. Vet Res. 2014;45:10. 10. Innes EA, Andrianarivo AG, Bjorkman C, Williams DJ, Conrad PA. Immune responses to Neospora caninum and prospects for vaccination. Trends Parasitol. 2002;18:497–504. 11. Montes MJ, Tortosa CG, Borja C, Abadia AC, González‑Gómez F, Ruiz C, et al. Constitutive secretion of interleukin‑6 by human decidual stromal cells in culture. Regulatory effect of progesterone. Am J Reprod Immunol. 1995;34:188–94. 12. Steinborn A, Von Gall C, Hildenbrand R, Stutte H, Kaufmann M. Identifica‑ tion of placental cytokine‑producing cells in term and preterm labor. Obstet Gynecol. 1998;91:329–35. 13. Steinborn A, Geisse M, Kaufmann M. Expression of cytokine receptors in the placenta in term and preterm labour. Placenta. 1998;19:165–70. 14. Rojo‑Montejo S, Collantes‑Fernández E, Blanco‑Murcia J, Rodríguez‑ Bertos A, Risco‑Castillo V, Ortega‑Mora LM. Experimental infection with a low virulence isolate of Neospora caninum at 70 days gestation in cattle did not result in foetopathy. Vet Res. 2009;40:49. 15. Caspe SG, Moore DP, Leunda MR, Cano DB, Lischinsky L, Regidor‑Cerrillo J, et al. The Neospora caninum‑Spain 7 isolate induces placental damage, fetal death and abortion in cattle when inoculated in early gestation. Vet Parasitol. 2012;189:171–81. 16. Dellarupe A, Regidor‑Cerrillo J, Jimenez‑Ruiz E, Schares G, Unzaga JM, Venturini MC, et al. Comparison of host cell invasion and proliferation among Neospora caninum isolates obtained from oocysts and from clini‑ cal cases of naturally infected dogs. Exp Parasitol. 2014;145:22–8. 17. Jiménez‑Pelayo L, García‑Sánchez M, Regidor‑Cerrillo J, Horcajo P, Collantes‑Fernández E, Gómez‑Bautista M, et al. Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum. Parasit Vectors. 2017;10:463. 18. Bevilacqua E, Hoshida MS, Amarante‑Paffaro A, Albieri‑Borges A, Zago Gomes S. Trophoblast phagocytic program: roles in different placental systems. Int J Dev Biol. 2010;54:495–505. 19. Mitsunari M, Yoshida S, Shoji T, Tsukihara S, Iwabe T, Harada T, et al. Macrophage‑activating lipopeptide‑2 induces cyclooxygenase‑2 and prostaglandin E2 via toll‑like receptor 2 in human placental trophoblast cells. J Reprod Immunol. 2006;72:46–59. 20. Gillaux C, Mehats C, Vaiman D, Cabrol D, Breuiller‑Fouche M. Func‑ tional screening of TLRs in human amniotic epithelial cells. J Immunol. 2011;187:2766–74. 21. Marin MS, Hecker YP, Quintana S, Pérez S, Leunda MR, Cantón G, et al. Toll‑like receptors 3, 7 and 8 are upregulated in the placental caruncle and fetal spleen of Neospora caninum experimentally infected cattle. Vet Parasitol. 2017;236:58–61. 22. Koga K, Mor G. expression and function of Toll‑like receptors at the maternal‑fetal interface. Reprod Sci. 2008;15:231–42. 23. Horcajo P, Jimenez‑Pelayo L, Garcia‑Sanchez M, Regidor‑Cerrillo J, Collantes‑Fernandez E, Rozas D, et al. Transcriptome modulation of bovine trophoblast cells in vitro by Neospora caninum. Int J Parasitol. 2017;47:791–9. 24. Marin MS, Hecker YP, Quintana S, Pérez S, Leunda MR, Cantón G, et al. Immunization with inactivated antigens of Neospora caninum induces toll‑like receptors 3, 7, 8 and 9 in maternal‑fetal interface of infected pregnant heifers. Vet Parasitol. 2017;243:12–7. 25. Liu J, Cao X. Cellular and molecular regulation of innate inflammatory responses. Cell Mol Immunol. 2016;13:711. 26. Adams DH, Rlloyd A. Chemokines: leucocyte recruitment and activa‑ tion cytokines. Lancet. 1997;349:490–5. 27. Taubert A, Krull M, Zahner H, Hermosilla C. Toxoplasma gondii and Neospora caninum infections of bovine endothelial cells induce endothelial adhesion molecule gene transcription and subsequent PMN adhesion. Vet Immunol Immunopathol. 2006;112:272–83. 28. Carvalho Neta AV, Stynen AP, Paixao TA, Miranda KL, Silva FL, Roux CM, et al. Modulation of the bovine trophoblastic innate immune response by Brucella abortus. Infect Immun. 2008;76:1897–907. 29. Haider S, Knöfler M. Human tumour necrosis factor: physiologi‑ cal and pathological roles in placenta and endometrium. Placenta. 2009;30:111–23. 30. Haraldsen G, Kvale D, Lien B, Farstad IN, Brandtzaeg P. Cytokine‑ regulated expression of E‑selectin, intercellular adhesion molecule‑1 (ICAM‑1), and vascular cell adhesion molecule‑1 (VCAM‑1) in human microvascular endothelial cells. J Immunol. 1996;156:2558–65. 31. Cavalcanti YV, Brelaz MC, Neves JK, Ferraz JC, Pereira VR. Role of TNF‑alpha, IFN‑gamma, and IL‑10 in the development of pulmonary tuberculosis. Pulm Med. 2012;2012:745483. 32. Robbins JR, Zeldovich VB, Poukchanski A, Boothroyd JC, Bakardjiev AI. Tissue barriers of the human placenta to infection with Toxoplasma gondii. Infect Immun. 2012;80:418–28. 33. Yamane I, Kitani H, Kokuho T, Shibahara T, Haritani M, Hamaoka T, et al. The inhibitory effect of interferon gamma and tumor necrosis factor alpha on intracellular multiplication of Neospora caninum in primary bovine brain cells. J Vet Med Sci. 2000;62:347–51. 34. Jesus EE, Pinheiro AM, Santos AB, Freire SM, Tardy MB, El‑Bacha RS, et al. Effects of IFN‑gamma, TNF‑alpha, IL‑10 and TGF‑beta on Neospora caninum infection in rat glial cells. Exp Parasitol. 2013;133:269–74. 35. Jauniaux E, Gulbis B, Schandene L, Collette J, Hustin J. Molecular interactions during pregnancy: distribution of interleukin‑6 in maternal and embryonic tissues during the first trimester. Mol Hum Reprod. 1996;2:239–43. 36. Diehl S, Rincón M. The two faces of IL‑6 on Th1/Th2 differentiation. Mol Immunol. 2002;39:531–6. 37. Pinheiro AM, Costa SL, Freire SM, Ribeiro CS, Tardy M, El‑Bacha RS, et al. Neospora caninum: early immune response of rat mixed glial cultures after tachyzoites infection. Exp Parasitol. 2010;124:442–7. 38. Almería S, Araujo RN, Darwich L, Dubey JP, Gasbarre LC. Cytokine gene expression at the materno‑foetal interface after experimental Neospora caninum infection of heifers at 110 days of gestation. Parasite Immunol. 2011;33:517–23. 39. Liao Y, Zhang Y, Liu X, Lu Y, Zhang L, Xi T, et al. Maternal murine cytomegalovirus infection during pregnancy up‑regulates the gene expression of Toll‑like receptor 2 and 4 in placenta. Curr Med Sci. 2018;38:632–9. 40. Gayle DA, Beloosesky R, Desai M, Amidi F, Nuñez SE, Ross MG. Maternal LPS induces cytokines in the amniotic fluid and corticotropin releasing hormone in the fetal rat brain. Am J Physiol Regul Integr Comp Physiol. 2004;286:1024–9. 41. Ashdown H, Dumont Y, Ng M, Poole S, Boksa P, Luheshi G. The role of cytokines in mediating effects of prenatal infection on the fetus: implications for schizophrenia. Mol Psychiatry. 2006;11:47. 42. Beloosesky R, Gayle DA, Amidi F, Nunez SE, Babu J, Desai M, et al. N‑Acetyl‑cysteine suppresses amniotic fluid and placenta inflamma‑ tory cytokine responses to lipopolysaccharide in rats. Obstet Gynecol. 2006;194:268–73. 43. Arranz‑Solís D, Benavides J, Regidor‑Cerrillo J, Horcajo P, Castaño P, del Carmen Ferreras M, et al. Systemic and local immune responses in sheep after Neospora caninum experimental infection at early, mid and late gestation. Vet Res. 2016;47:2. 44. Rosbottom A, Gibney H, Kaiser P, Hartley C, Smith RF, Robinson R, et al. Up regulation of the maternal immune response in the placenta of cat‑ tle naturally infected with Neospora caninum. PLoS ONE. 2011;6:e15799. 45. Almería S, Serrano‑Perez B, Darwich L, Domingo M, Mur‑Novales R, Regidor‑Cerrillo J, et al. Foetal death in naive heifers inoculated with Neospora caninum isolate Nc‑Spain7 at 110 days of pregnancy. Exp Parasitol. 2016;168:62–9. 46. Baszler TV, Long MT, McElwain TF, Mathison BA. Interferon‑gamma and interleukin‑12 mediate protection to acute Neospora caninum infection in BALB/c mice. Int J Parasitol. 1999;29:1635–46. 47. Etienne‑Manneville S, Chaverot N, Strosberg AD, Couraud PO. ICAM‑ 1‑coupled signaling pathways in astrocytes converge to cyclic AMP response element‑binding protein phosphorylation and TNF‑alpha secre‑ tion. J Immunol. 1999;163:66874. Page 11 of 11Jiménez‑Pelayo et al. Parasites Vectors (2019) 12:218 • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research ? Choose BMC and benefit from: 48. Deisher TA, Haddix TL, Montgomery KF, Pohlman TH, Kaushansky K, Harlan JM. The role of protein kinase C in the induction of VCAM‑1 expression on human umbilical vein endothelial cells. FEBS Lett. 1993;331:285–90. 49. Silva LM, Vila‑Viçosa MJ, Cortes HC, Taubert A, Hermosilla C. Suitable in vitro Eimeria arloingi macromeront formation in host endothelial cells and modulation of adhesion molecule, cytokine and chemokine gene transcription. Parasitol Res. 2015;114:113–24. 50. Maksimov P, Hermosilla C, Kleinertz S, Hirzmann J, Taubert A. Besnoitia besnoiti infections activate primary bovine endothelial cells and promote PMN adhesion and NET formation under physiological flow condition. Parasitol Res. 2016;115:1991–2001. 51. Zhang D, Chen L, Li S, Gu Z, Yan J. Lipopolysaccharide (LPS) of Porphy- romonas gingivalis induces IL‑1β, TNF‑α and IL‑6 production by THP‑1 cells in a way different from that of Escherichia coli LPS. Innate Immun. 2008;14:99–107. 52. Castillo C, Muñoz L, Carrillo I, Liempi A, Medina L, Galanti N, et al. Ex vivo infection of human placental chorionic villi explants with Trypanosoma cruzi and Toxoplasma gondii induces different Toll‑like receptor expression and cytokine/chemokine profiles. Am J Reprod Immunol. 2017;78:12660. 53. Castillo C, Muñoz L, Carrillo I, Liempi A, Medina L, Galanti N, et al. Toll‑like receptor‑2 mediates local innate immune response against Trypanosoma cruzi in ex vivo infected human placental chorionic villi explants. Placenta. 2017;60:40–6. 54. Regidor‑Cerrillo J, Gomez‑Bautista M, Sodupe I, Aduriz G, Alvarez‑Garcia G, Del Pozo I, et al. In vitro invasion efficiency and intracellular prolifera‑ tion rate comprise virulence‑related phenotypic traits of Neospora caninum. Vet Res. 2011;42:41. 55. Almería S, Serrano‑Perez B, López‑Gatius F. Immune response in bovine neosporosis: protection or contribution to the pathogenesis of abortion. Microb Pathog. 2017;109:177–82. 56. Regidor‑Cerrillo J, Gómez‑Bautista M, Pereira‑Bueno J, Adúriz G, Navarro‑ Lozano V, Risco‑Castillo V, et al. Isolation and genetic characterization of Neospora caninum from asymptomatic calves in Spain. Parasitology. 2008;135:1651–9. 57. Rojo‑Montejo S, Collantes‑Fernández E, Regidor‑Cerrillo J, Álvarez‑García G, Marugan‑Hernández V, Pedraza‑Díaz S, et al. Isolation and characteri‑ zation of a bovine isolate of Neospora caninum with low virulence. Vet Parasitol. 2009;159:7–16. 58. Pérez‑Zaballos FJ, Ortega‑Mora LM, Álvarez‑García G, Collantes‑Fernández E, Navarro‑Lozano V, García‑Villada L, et al. Adaptation of Neospora cani- num isolates to cell‑culture changes: an argument in favor of its clonal population structure. J Parasitol. 2005;91:507–10. 59. Bridger PS, Menge C, Leiser R, Tinneberg HR, Pfarrer CD. Bovine carun‑ cular epithelial cell line (BCEC‑1) isolated from the placenta forms a functional epithelial barrier in a polarised cell culture model. Placenta. 2007;28:1110–7. 60. Hambruch N, Haeger JD, Dilly M, Pfarrer C. EGF stimulates proliferation in the bovine placental trophoblast cell line F3 via Ras and MAPK. Placenta. 2010;31:67–74. 61. Menzies M, Ingham A. Identification and expression of Toll‑like receptors 1–10 in selected bovine and ovine tissues. Vet Immunol Immunopathol. 2006;109:23–30. 62. Puech C, Dedieu L, Chantal I, Rodrigues V. Design and evaluation of a unique SYBR Green real‑time RT‑PCR assay for quantification of five major cytokines in cattle, sheep and goats. BMC Vet Res. 2015;11:65. 63. Schmittgen TD, Livak KJ. Analyzing real‑time PCR data by the compara‑ tive C(T) method. Nat Protoc. 2008;3:1101–8. Tesis Laura Jiménez Pelayo PORTADA TABLE OF CONTENTS LIST OF TABLES AND BOXES LIST OF FIGURES LIST OF ABBREVIATIONS CHAPTER I. SUMMARY / RESUMEN CHAPTER II. INTRODUCTION / INTRODUCCIÓN CHAPTER III. JUSTIFICATION AND OBJECTIVES /JUSTIFICACIÓN Y OBJETIVOS CHAPTER IV. RESULTS (PUBLICATIONS) / RESULTADOS (PUBLICACIONES) LIST OF PUBLICATIONS CHAPTER V. GENERAL DISCUSSION / DISCUSIÓN GENERAL CHAPTER VI. CONCLUSIONS / CONCLUSIONES CHAPTER VII. REFERENCES / BIBLIOGRAFÍA APPENDIX I. ORIGINAL MANUSCRIPTS / MANUSCRITOS ORIGINALES Differential susceptibility of bovine caruncular and trophoblast cell lines to infection with high and low virulence isolates of Neospora caninum Immune response profile of caruncularand trophoblast cell lines infected by high‑(Nc‑Spain7) and low‑virulence (Nc‑Spain1H)isolates of Neospora caninum