https://doi.org/10.1177/03009858241250336 Veterinary Pathology 1–11 © The Author(s) 2024 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/03009858241250336 journals.sagepub.com/home/vet Original Article 1250336 VETXXX10.1177/03009858241250336Veterinary PathologyRebollada-Merino et al research-article2024 1Department of Population Medicine and Diagnostic Sciences, Section of Anatomic Pathology, College of Veterinary Medicine, Cornell University, Ithaca, NY 2World Organisation for Animal Health Collaborating Centre for the Health of Marine Mammals, Italian National Reference Centre for Diagnostic Activities in Stranded Marine Mammals (C.Re.Di.Ma), Istituto Zooprofilattico Sperimentale del Piemonte, Liguria e Valle d’Aosta, Torino, Italy 3Departament de Medicina i Cirurgía Animals, Universitat Autònoma de Barcelona, Bellaterra, Spain 4Departament de Sanitat i Anatomia Animals, Universitat Autònoma de Barcelona, Bellaterra, Spain 5VISAVET Health Surveillance Centre, Complutense University of Madrid, Madrid, Spain 6Department of Animal Health, Faculty of Veterinary Medicine, Complutense University of Madrid, Madrid, Spain 7Veterinary Department, Mundomar, Benidorm, Spain 8Istituto Zooprofilattico Sperimentale del Mezzogiorno, Portici, Italy 9Istituto Zooprofilattico Sperimentale della Puglia e della Basilicata, Foggia, Italy 10Unitat Mixta d’investigació IRTA-UAB en Sanitat Animal, Centre de Recerca en Sanitat Animal (CReSA), Universitat Autònoma de Barcelona, Bellaterra, Spain 11Department of Internal Medicine and Animal Surgery, Faculty of Veterinary Medicine, Complutense University of Madrid, Madrid, Spain Supplemental material for this article is available online. Corresponding Author: Agustín Rebollada-Merino, Department of Population Medicine and Diagnostic Sciences, Section of Anatomic Pathology, College of Veterinary Medicine, Cornell University; Animal Health Diagnostic Center, 240 Farrier Road, Ithaca, NY 14850, USA. Email: amr449@cornell.edu Neurobrucellosis (Brucella ceti) in striped dolphins (Stenella coeruleoalba): Immunohistochemical studies on immune response and neuroinflammation Agustín Rebollada-Merino1 , Federica Giorda2, Martí Pumarola3, Laura Martino4, Alberto Gomez-Buendia5,6 , Umberto Romani-Cremaschi7, Cristina Casalone2, Virginia Mattioda2, Fabio Di Nocera8, Giuseppe Lucifora8, Antonio Petrella9, Lucas Domínguez5,6, Mariano Domingo4,10, Carla Grattarola2, and Antonio Rodríguez-Bertos5,11 Abstract Neurobrucellosis is a shared condition of cetaceans and humans. However, the pathogenesis and immune response in cetacean neurobrucellosis has not been extensively studied. In this multicentric investigation, 21 striped dolphin (Stenella coeruleoalba) neurobrucellosis (Brucella ceti) cases diagnosed over a 10-year period (2012-2022) were retrospectively evaluated. For each case, morphological changes were assessed by evaluating 21 histological parameters. Furthermore, the immunohistochemical expression of Brucella antigen, glial fibrillary acid protein (GFAP), and a selection of inflammatory cell (IBA-1, CD3, and CD20) and cytokine (tumor necrosis factor-alpha [TNF-α], interferon-gamma [IFN-γ], interleukin [IL]-1β, IL-2, and IL-6) markers were investigated. Inflammation of the leptomeninges, ependyma, and/or choroid plexus was lymphohistiocytic, containing macrophages/microglia (IBA-1+), T-cells (CD3+), and B-cells (CD20+) in equal proportion. B-cells occasionally formed tertiary follicles. GFAP expression showed astrocytosis in most cases. Expression of TNF-α, IFN-γ, and IL-2 indicated an intense proinflammatory response, stimulating both macrophages and T-cells. Our results showed that the inflammation and neuroinflammation in neurobrucellosis of striped dolphins mimic human neurobrucellosis and in vitro and in vivo studies in laboratory animals. Cetacean disease surveillance can be exploited to expand the knowledge of the pathogenesis and immunology of infectious diseases, particularly brucellosis, under a One Health approach. Keywords Brucella ceti, cetacean, dolphin, human, immune response, immunohistochemistry, inflammation, neurobrucellosis https://us.sagepub.com/en-us/journals-permissions https://journals.sagepub.com/home/vet http://crossmark.crossref.org/dialog/?doi=10.1177%2F03009858241250336&domain=pdf&date_stamp=2024-05-17 2 Veterinary Pathology 00(0) Brucellosis is an important zoonosis endemic in many parts of the world, caused by some Brucella spp., most commonly Brucella melitensis, Brucella abortus, and Brucella suis (biovars 1 and 3). Other Brucella spp. can cause sporadic cases of brucellosis in humans, such as B. suis biovar 2 and 4, Brucella canis, Brucella neotomae, and marine mammal strains.46,57,58 Human infection with terrestrial Brucella mainly occurs via ingestion of contaminated unpasteurized dairy prod- ucts, contact with infected animals, or inhalation of aerosols.4 The multisystemic nature of brucellosis in humans is due to the involvement of the osteoarticular, lymphoid, hepatobiliary, and vascular systems.4 Brucella can also affect the central ner- vous system (CNS),31,38 resulting in neurobrucellosis in up to 25% of brucellosis patients,24,47 especially immunocompro- mised individuals.31 Neurological manifestations in neurobru- cellosis may occur in both acute and chronic infections.38 Neurobrucellosis in humans is clinically obscure. Brucella can affect any region of the CNS including the leptomeninges, cerebrum, cerebellum, brainstem, cranial nerves, spinal cord, and/or peripheral nerves, and this explains the broad spectrum of clinical presentations.14,31,35,55 In the postantibiotic era, mortality due to neurobrucellosis is around 5%, but permanent neu- rological deficits as sequel of infections are not infrequent.14,36 The clinical diagnosis of neurobrucellosis in humans is fre- quently achieved by exclusion. Neuroimaging supports clinical suspects, but neurobrucellosis confirmation requires ancillary techniques. Cerebrospinal fluid can be screened looking for cellular, protein, and/or glucose alterations. Brucella infection must be confirmed by means of bacterial isolation from cere- brospinal fluid or evidencing a systemic immune response against Brucella.31 Novel techniques such as metabolomics may contribute to a rapid diagnosis.63 Brucella reaches the CNS by hematogenous spread during early bacteremia or after reactivation in immunosuppressive states that facilitate Brucella immune evasion.31,38,43 The lesions induced during neurobrucellosis in humans comprise a direct effect of persisting Brucella, as well as indirect damage due to the host inflammatory response triggered by the bacteria or its endotoxins.31,42 Despite recent advances on the knowl- edge of neurobrucellosis in human beings, many aspects of the pathogenesis and inflammation in human neurobrucellosis remain elusive. This is partly because the lack of animal mod- els mimicking the disease in humans.43 Brucella ceti16 is a phenotypically smooth strain causing cetacean brucellosis in mysticetes and odontocetes worldwide.25 Brucellosis is regarded as endemic in many cetacean popula- tions.9,40 Transmission may occur both horizontally (secretions, milk, and semen) and vertically.10,26,29,39 Some carnivorous spe- cies, such as killer whales (Orcinus orca), are thought to become infected by ingesting tissues of other cetaceans with brucello- sis.11 Pulmonary nematodes (Halocerus spp. and Pseudalius inflexus) harboring B. ceti are suspected to contribute as vectors to transmission in bottlenose dolphin (Tursiops truncatus) and harbor porpoise (Phocoena phocoena) based on bacteriology and immunohistochemical studies.11,29 Cetaceans from which B. ceti has been isolated are often found stranded, moribund, or dead.46 Besides neurological disease, B. ceti causes reproductive (abortion, placentitis, endometritis, mastitis, and epididymo-orchitis), osteoarticular (arthritis and discospondylitis), cutaneous (blubber abscesses), and cardiac (myocarditis and endocarditis) lesions.21,22 The liver, kidney, spleen, lymph nodes, and lungs can also be affected.18,28,46 B. ceti involvement of the CNS (neurobrucellosis) has been particularly reported in the striped dolphin (Stenella coeruleo- alba).9,12,18,23,26 Striped dolphins stranded alive affected from neurobrucellosis commonly show neurologic signs such as uncoordinated swimming, lethargy, circling, lateralization, inability to maintain buoyancy, opisthotonos, tremors, and/or tonic-clonic seizures.26,28 Gross findings include leptomenin- geal clouding, acquired hydrocephalus, and turbid cerebrospi- nal fluid (CSF) in some cases.11,21 The hallmark histological lesion in the CNS is a nonsuppu- rative meningoencephalitis, periventriculitis, choroiditis, and/ or radiculitis in cerebrum, cerebellum, brainstem, and/or spinal cord, along with occasional perivascular cuffing.1,9,18,20,21,23,26,28 Despite B. ceti seeming to be a major cause of morbidity and mortal- ity in cetacean populations,9 the pathogenesis and immunology of cetacean neurobrucellosis have not been extensively stud- ied.13 In this study, we aimed to describe the immunopathology of cetacean neurobrucellosis in the striped dolphin and to com- pare our results with previous studies in humans and in vitro and in vivo laboratory animal models of neurobrucellosis. Materials and Methods Retrospective Database Search The databases of the World Organisation for Animal Health Collaborating Centre for the Health of Marine Mammals (Italian National Reference Centre for Diagnostic Activities in Stranded Marine Mammals, [C.Re.Di.Ma], Torino Italy) and of the Autonomous University of Barcelona (Spain) were searched for deceased striped dolphins (Stenella coeruleoalba) found stranded along the Italian or Catalonian coastlines, respectively, over a 10-year period (2012–2022). No ethical approval was required for this study because tissues from deceased wildlife animals submitted for routine diagnosis were used retrospectively. Inclusion Criteria Cases included (n = 21) had been submitted for autopsy and had available formalin-fixed, paraffin-embedded CNS sam- ples. All cases were diagnosed with neurobrucellosis based on (1) Brucella isolation in the CNS and other organs (at least spleen, liver, and ovary/testis) using Brucella medium agar base, modified Thayer-Martin agar, CITA media, Farrell media, Columbia agar, and/or MacConkey agar;9,18,28 (2) B. ceti molecular confirmation (see below); and/or (3) positive sero- logical results using a rapid serum agglutination test (Rose Bengal plate test) (distributed by the National Reference Laboratories for Brucellosis; reagent available only for official testing).9,23 For molecular confirmation in most cases, Brucella isolates were sub-cultured and subjected to DNA extraction using the Maxwell1 16 Tissue DNA Purification Kit and the Rebollada-Merino et al 3 Maxwell1 16 Instrument (Promega, Madison, Wisconsin) or the High Pure DNA Template Preparation kit (Roche Diagnostics, Mannheim, Germany) and were then identified as B. ceti using at least the PCR-RFLP (restriction fragment length polymorphism) method.8 Case 12 and 20 were PCR positive (fresh tissue) in brain, lung, spleen, and/or blubber abscesses; seropositive; and showed Brucella-compatible gross and histological lesions, thus confirming a diagnosis of brucellosis. Data regarding postmortem condition code were recorded when available: code 1 (extremely fresh carcass, just dead), code 2 (fresh), code 3 (moderate decomposition), code 4 (advanced decomposition), or code 5 (mummified or skele- tal remains) (Table 1). Additional Testing Additional testing results for other cetacean pathogens target- ing the CNS had been performed in fresh tissues, including cetacean morbillivirus (PCR),32,60 gamma-herpesvirus (PCR),59 Toxoplasma gondii (PCR),61 and Listeria monocytogenes (bacterial culture) (Table 1). In cases found to be coinfected (cases 5, 6, 11, 12, and 13), sections included in this study were negative by immunohistochemistry (IHC) for cetacean morbil- livirus, gamma-herpesvirus, and Toxoplasma gondii.9,23,52 L. monocytogenes and Sarcocystis spp. infection was excluded given the absence of characteristic histological lesions.19 Histology Necropsies had been performed, and a complete set of organs (CNS [cerebrum, cerebellum, brainstem, and spinal cord], lymph nodes, spleen, liver, pancreas, adrenal glands, kidneys, ovaries/testes, stomach, intestine, trachea, thyroid and parathy- roid glands, thymus, lungs, heart, skin, blubber, and skeletal muscle) were collected and fixed by immersion in 10%, buff- ered, pH 7.2 formalin for 24 to 72 hours, before being routinely processed to produce 4-µm thick, hematoxylin and eosin stained sections. The Kluver-Barrera stain was performed when chromatolysis or demyelination was suspected. For each case, the 21 histological parameters detailed in Table 2 were Table 1.  Striped dolphin (Stenella coeruleoalba) cases with neurobrucellosis (Brucella ceti). Case No. Country Region Found Dead/ Moribund Postmortem Condition Code (0-5) Brucella Isolation B. ceti Molecular Confirmation Serology Other CNS Pathogens 1 Italy Apulia Dead 2 CNS, spleen + NP - 2 Italy Apulia Dead 3 CNS + NP - 3 Italy Apulia Dead NP CNS + NP - 4 Italy Apulia Dead 2 CNS + NP - 5 Italy Apulia Dead 2 CNS + + CeMV 6 Italy Campania Dead 2 CNS, spleen, lymph node (prescapular) + NP HV, Toxoplasma gondii 7 Italy Campania Dead 2 CNS + NP - 8 Italy Apulia Dead 3 CNS + + - 9 Italy Apulia Dead 2 CNS, heart, lymph node (prescapular), uterus + + - 10 Italy Apulia Dead 2 CNS, lung, heart, lymph node (tracheobronchial), liver, uterus + + - 11 Italy Calabria Dead 2 CNS + - CeMV 12 Italy Liguria Dead 2 NP + + T. gondii, Listeria monocytogenes 13 Italy Liguria Dead 2 CNS, lung, liver, spleen, lymph nodes (prescapular, tracheobronchial) + - CeMV, T. gondii, Sarcocystis spp. 14 Italy Campania Dead 3 CNS, lung + NP - 15 Spain Catalonia Moribund 1 CNS + + - 16 Spain Catalonia Dead 2 CNS, lymph node (diaphragmatic) + - - 17 Spain Catalonia Moribund 1 CNS, lymph node (mesenteric) + + - 18 Spain Catalonia Moribund 1 CNS, blood + + - 19 Spain Catalonia Moribund 1 CNS + + - 20 Spain Catalonia Moribund 1 - + + - 21 Spain Catalonia Moribund 1 CNS NP - - Abbreviations: CeMV, cetacean morbillivirus; CNS, central nervous system; HV, gamma-herpesvirus; NP, not performed. 4 Veterinary Pathology 00(0) semiquantitatively evaluated using a score from 0 to 3 (grade 0: absence of lesions; grade 1: focal lesions; grade 2: multifocal lesions; grade 3: diffuse lesions) in the hematoxylin and eosin section displaying the most severe inflammation (Table 1). Immunohistochemical expression of GFAP was used to assess astrocytosis (see below). The postmortem condition code did not affect the histological evaluation as for most animals the carcass was fresh. Immunohistochemistry For each case, consecutive paraffin sections scored histologi- cally were placed on charged slides. Deparaffinization, rehy- dration, and heat-induced epitope retrieval was performed using a PT Lab Vision module (Thermo Fisher). Then, the endogenous peroxidase was quenched through immersion in a hydrogen peroxide solution in methanol (Panreac Química). Table 2.  Histopathological parameters evaluated in the central nervous system of striped dolphins (Stenella coeruleoalba) with neurobrucellosis (Brucella ceti). Neuronal and glial changes Chromatolysis Acidophilic necrosis Microglial nodules Wallerian-like degeneration Demyelination Astrocytosis Inflammation Ependymitis Choroiditis Leptomeningitis Meningeal fibrosis Metaplastic ossification Vascular changes Perivascular cuffing Vasculitis Neovascularization Thrombosis Endothelial hypertrophy Hemodynamic changes Vasogenic edema Cytotoxic edema Interstitial edema Hemorrhage Congestion Table 3.  Primary antibodies used for the immunohistochemical study of the central nervous system of striped dolphins (Stenella coeruleoalba) with neurobrucellosis (Brucella ceti). Primary Antibody (Clone) Dilution Incubation Manufacturer Protein Identity Rabbit polyclonal anti-Brucella (IgG) 1:400 37°C, 1 hour Bioss - Mouse monoclonal anti-CDV-NP 1:500 37°C 1 hour VMRD - Rabbit polyclonal anti-HSV-1 1:75 37°C 1 hour Abcam - Rabbit polyclonal anti-Toxoplasma gondii 1:50 37°C 1 hour Thermo Fisher - Rabbit polyclonal anti-GFAP 1:400 4°C ON Dako 93.3% (human) Mouse monoclonal anti-mouse IBA-1 (NCNP24) 1:100 37°C 1 hour Wako Chemicals 89.8% (mouse) Rabbit polyclonal anti-human CD3 (UCHT1) 1:100 37°C 1 hour Dako 86.2% (human) Rabbit polyclonal anti-human CD20 (PA5-32313) 1:200 37°C 1 hour Thermo Fisher 83.0% (human) Rabbit polyclonal anti-bovine TNF-α (AHP852B) 1:200 4°C ON AbD Serotec 84.2% (bovine) Mouse monoclonal anti-bovine IFN-γ (7B6) 1:25 4°C ON AbD Serotec 85.5% (bovine) Mouse monoclonal anti-ovine IL-1β (1D4) 1:25 4°C ON AbD Serotec 77.8% (ovine) Mouse monoclonal anti-ovine IL-2 (IE10) 1:25 4°C ON AbD Serotec 68.0% (ovine) Mouse monoclonal anti-bovine IL-6 (CC310) 1:25 4°C ON LSBio 67.1% (bovine) Abbreviations: CD, cluster of differentiation; CDV-NP, canine distemper virus nucleoprotein; GFAP, glial fibrillary acidic protein; HSV, herpes simplex virus; IFN, interferon; IL, interleukin; ON, overnight; TNF, tumoral necrosis factor. Rebollada-Merino et al 5 The samples were incubated, first in horse serum and then in primary antibodies (Table 3). Due to the high identity of cytokines between cetaceans and ruminants,15 commercial antibodies for ruminant cytokines were used in this study. Antibody protein sequence identity between the bottlenose dolphin (Tursiops truncatus) and the target species of the antibody (human, mouse, bovine, or ovine) was assessed using pBLAST tool (National Center for Biotechnology Information, Bethesda, MA) (Table 3). The bottlenose dolphin was used as the cetacean reference because most protein sequences of striped dol- phins are not characterized yet. Inflammatory cell and cytokine antibodies cross-reactivity (IBA-1, CD3, CD20, tumor necrosis factor-alpha [TNF-α], interferon-gamma [IFN-γ], interleukin [IL]-1β, IL-2, and IL-6), not specificity, was assessed in lymphoid tissues (spleen and lymph node) from the target species (normal tissue and a tuberculous granuloma), and bottlenose and striped dolphin (normal tissue and lymphadenitis or splenitis). For all reactions except CD20, commercial reagents were used for the secondary antibody (ImmPRESS-VR horse anti- rabbit IgG polymer kit, Vector) and chromogen (ImmPACT NovaRED peroxidase substrate, Vector). Only for CD20, immunolabeling was performed on an autostainer (Autostainer Plus, Dako) using procedures, buffers, and solutions provided by the manufacturer; slides were incubated in DAB-chromogen- hydrogen peroxide solution (Dako, Denmark). In all cases, sam- ples were counterstained with hematoxylin and coverslipped. Formalin-fixed paraffin-embedded samples from the CNS and lymphoid organs from unrelated striped and bottlenose dolphins; from the target species of the antibody; or from ani- mals infected by Brucella, cetacean morbillivirus, gamma-her- pesvirus, or Toxoplasma gondii were used as positive controls when needed (depending on the primary antibody used). For negative controls, the primary antibody was replaced with a commercial universal negative control reagent. Brucella immunolabeling was recorded as absent/present. Immunolabeling for the rest of antibodies was quantitatively evaluated in leptomeninges and/or choroid plexuses as follows: 100 cells were counted in 5 adjacent, non-overlapping fields under a high-power field (HPF) (400×, size = 0.237 mm2), and immunolabeled cells were set as a percentage of the cell count (0%-100%). A score from 0 to 3 was assigned to each sample for statistical purposes—grade 0, absence: 0% immunopositive cells; grade 1, low numbers (less than 30%) of immunopositive cells; grade 2, moderate numbers (30%-60%) of immunopositive cells; and grade 3, abundant (more than 60%) of immunopositive cells. Statistics To evaluate the relationship between the histopathological lesions observed and the immunohistochemical results, a Kendall tau-b correlation coefficient was calculated in R.48 Results Histopathology Histologic lesions were observed in the cerebrum (19/21; 90%), cerebellum (20/21; 95%), brainstem (19/19; 100%), and spinal cord (16/16; 100%) (Table 4). Neuronal changes were character- ized by minimal focal areas of chromatolysis (central) (mean score [MS] 0.76, 0.70 standard deviation [SD]) (Fig. 1g), microg- lial nodules (MS 0.38 ± 0.67 SD) (Fig. 1e), demyelination of the white matter (MS 0.38 ± 0.74 SD), and astrocytosis (GFAP immunolabeling) (MS 0.43 ± 0.75 SD) (Fig. 1f). Acidophilic necrosis and Wallerian-like degeneration were not found in any case. Inflammation (non-suppurative and lymphohistiocytic) was diffusely observed especially in the leptomeninges (MS 2.14 ± 0.96 SD) (Fig. 1a) compared with ependymal and subep- endymal areas (MS 0.52 ± 0.93 SD) (Fig. 1b), and the choroid plexus (MS 0.38 ± 0.86 SD) (Fig. 1c). Mild focal leptomenin- geal fibrosis was observed in a few cases (MS 0.43 ± 0.75 SD). Metaplastic ossification of the leptomeninges was not observed. Regarding vascular changes, multifocal perivascular cuffing (nonsuppurative, lymphohistiocytic) was the most common finding (MS 1.29 ± 0.78 SD) (Fig. 1d). Endothelial cells of leptomeninges and parenchymal blood vessels were occasion- ally hypertrophied (MS 0.86 ± 0.73 SD). Rarely, thrombosis was observed (MS 0.29 ± 0.46 SD). Vasculitis and neovascu- larization were not observed. Hemodynamic changes consisted of multifocal vasogenic (MS 0.86 ± 0.65 SD) (Fig. 1h) and cytotoxic (MS 0.29 ± 0.46 SD) edema, hemorrhage (MS 0.95 ± 0.74 SD), and congestion (MS 0.86 ± 0.65 SD). Interstitial edema was almost absent (MS 0.05 ± 0.22 SD). Individual histological parameters are shown in Supplemental Table S1. Immunohistochemistry Brucella immunolabeling was observed intracellularly in scat- tered macrophages located in the areas of leptomeningitis, ependymitis, or choroiditis in all cases (Fig. 1k) and occasion- ally in inflammatory cells surrounding noninflamed blood ves- sels in the neuroparenchyma. Inflammation consisted of IBA-1+ macrophages and microglia, CD3+ T-cells, and CD20+ B-cells. IBA-1+ cells (macrophages/microglia) were multifocally observed in the areas of inflammation in the leptomeninges, ependyma, or cho- roid plexus (Fig. 2a). IBA-1+ cells were concentrated mainly in the outer areas of perivascular inflammation in the leptomenin- ges, ependyma, and choroid plexus. IBA-1+ cells were in low (grade 1, 16/21; 76%) or moderate (grade 2, 5/21; 24%) num- bers (MS 1.24 ± 0.44 SD). CD3+ cells (T-cells) were in the leptomeningeal, ependymal, or choroidal areas of inflamma- tion in a random distribution (Fig. 2b), but also in areas of neu- ronophagia (microglial nodules). T-cells were in low numbers (grade 1, 14/21; 67%), moderate numbers (grade 2, 5/21; 24%), or abundant (grade 3, 2/21; 9%) (MS 1.43 ± 0.68 SD). Membranous expression of CD20 was seen in B-cells in the areas of inflammation. B-cells were found randomly but occa- sionally formed lymphoid follicles (6/21; 29%) (Fig. 2c). B-cells were absent (grade 0, 1/21; 5%), in low numbers (grade 1, 8/21; 38%), in moderate numbers (grade 2, 9/21; 43%), or abundant (grade 3, 3/21; 14%) (MS 1.67 ± 0.80 SD). Cytokines evaluated included TNF-α, IFN-γ, IL-2, IL-1β, and IL-6. TNF-α was expressed in the cytoplasm of 6 Veterinary Pathology 00(0) inflammatory cells (macrophages and lymphocytes based on cell morphology) (Fig. 2d), especially around blood vessels of inflamed leptomeninges, ependyma, and choroid plexuses. TNF-α-expressing cells were absent (grade 0, 1/21; 5%), in low numbers (grade 1, 14/21; 67%), in moderate numbers (grade 2, 5/21; 23%), or abundant (grade 3, 1/21; 5%) (MS 1.29 ± 0.64 SD). The expression of IFN-γ was observed in the cyto- plasm of inflammatory cells (lymphocytes and macrophages based on cell morphology) and was distributed randomly through the areas of inflammation (Fig. 2e). IFN-γ-expressing cells were absent (grade 0, 2/21; 9%), in low numbers (grade 1, 11/21; 53%), in moderate numbers (grade 2, 7/21; 33%), or abundant (grade 3, 1/21; 5%) (MS 1.33 ± 0.73 SD). IL-2 expression occurred in the cytoplasm of inflammatory cells (lymphocytes and microglia based on cell morphology), ran- domly and eccentrically distributed with respect to blood ves- sels in inflamed areas (Fig. 2f). IL-2-expressing cells were absent (grade 0, 2/21; 9%), in low numbers (grade 1, 10/21; 48%), or in moderate numbers (grade 2, 9/21; 43%) (MS 1.33 ± 0.66 SD). Antibodies targeting IL-1β and IL-6 did not show immunoreactivity in dolphin tissues. IL-1β immunoreactivity was observed in the target species but not in dolphin tissues, and IL-6 immunoreactivity was not observed in either the tar- get species or dolphins. Figure 1.  Histological and immunohistochemical findings in neurobrucellosis (Brucella ceti) of striped dolphins (Stenella coeruleoalba). (a) Spinal cord. Leptomeningitis, lymphohistiocytic, diffuse, chronic, severe. Case 4. Hematoxylin and eosin (HE). (b) Lateral ventricle. Choroid plexitis, lymphohistiocytic, diffuse, chronic, severe. Case 14. HE. (c) Cerebrum. Ependymitis and subependymal perivascular encephalitis, lymphohistiocytic, multifocal, chronic, and moderate. Case 3. HE. (d) Spinal cord. Blood vessels are surrounded by a non-suppurative inflammatory infiltrate (perivascular cuffing). Case 4. HE. (e) Cerebrum. Microglial nodule adjacent to an area of leptomeningitis. Case 3. HE. (f) Cerebrum. Astrocytosis assessed as an increased expression of glial fibrillary acidic protein (GFAP). Inset: GFAP expression in a negative control dolphin. Case 2. GFAP immunohistochemistry (IHC). (g) Cerebrum. Central chromatolysis. Case 3. HE. Inset: centrifugal distribution and condensation of Nissl granules and peripheral nuclear displacement. Kluver-Barrera. (H) Spinal cord. Vasogenic edema. Case 19. HE. (i) Positive control, presence of Brucella antigen in the placenta of a sheep experimentally infected with B. melitensis. Brucella IHC. (j) Negative control, absence of Brucella antigen in an area of leptomeningitis of a dolphin not infected by B. ceti. Brucella IHC. (k) Striped dolphin with neurobrucellosis. Presence of Brucella antigen in the cytoplasm of macrophages. Brucellai IHC. Rebollada-Merino et al 7 Figure 2.  Immunohistochemical findings in neurobrucellosis (Brucella ceti) of striped dolphins (Stenella coeruleoalba). (a) Macrophages/microglia are located multifocally in areas of leptomeningitis. Case 5. IBA-1 immunohistochemistry (IHC). (b) T-cells are multifocally distributed in the inflamed leptomeninges. Case 17. CD3 IHC. (c) B-cells are multifocally distributed and occasionally forming lymphoid follicles in the leptomeninges. Case 15. CD20 IHC. (d) Tumor necrosis factor-alpha (TNF-α) is expressed in inflammatory cells in a multifocal distribution. Case 9. TNF-α IHC. (e) Interferon-gamma (IFN-γ)-expressing inflammatory cells are multifocally distributed in inflamed areas. Case 2. IFN-γ IHC. (f) Interleukin-2 (IL-2) is expressed in inflammatory cells distributed multifocally in the inflamed leptomeninges. Case 10. IL-2 IHC. Table 4.  Anatomic distribution of histologic lesions in the central nervous system of striped dolphins (Stenella coeruleoalba) with neurobrucellosis (Brucella ceti). Case No. Affected Areas Selected Areas for HE Scoring and IHCCerebrum Cerebellum Brainstem Spinal Cord 1 + NP NP NP Cortex 2 + + + + Cortex 3 + + + + Cortex 4 + + + + Spinal cord 5 + + + + Cerebellum 6 + + + + Choroid plexus 7 + + + + Cortex 8 + + + + Cerebellum 9 + + + + Cortex 10 + + + + Cerebellum 11 + + NP NP Choroid plexus 12 + + + NP Choroid plexus 13 + + + NP Cerebellum 14 + + + NP Choroid plexus 15 + + + + Choroid plexus 16 + + + + Cerebellum 17 + + + + Brainstem 18 + + + + Brainstem 19 + + + + Spinal cord 20 - + + + Spinal cord 21 - + + + Cerebellum Abbreviations: HE, hematoxylin and eosin; IHC, immunohistochemistry; NP, histology not performed. 8 Veterinary Pathology 00(0) Individual immunohistochemical parameters and are shown in Supplemental Table S1. Individual immunohistochemical expression of IBA-1, CD3, CD20, TNF-α, IFN-γ, and IL-2 and distribution among histological grades is shown in Fig. 3. Statistics Statistical correlations between immunohistochemical and his- topathological score values showed a significant moderate positive correlation between IBA-1 expression and the degree of leptomeningitis (τB = 0.56 [95% confidence interval [CI] = 0.26–0.85]; P = .0078). Endothelial hypertrophy showed a sig- nificant weak correlation with CD20 expression (τB = 0.47 [95% CI = 0.23–0.71]; P = .019). Also, there was a significant moderate correlation between perivascular cuffing and TNF-α expression (τB = 0.50 [95% CI = 0.21–0.79]; P = .015). Astrocytosis (GFAP expression) and IL-2 expression showed a significant moderate correlation (τB = 0.50 [95% CI = 0.23– 0.77]; P = .017). Finally, there was a significant weak correla- tion between thrombosis and IL-2 expression (τB = 0.49 [95% CI = 0.19–0.80]; P = .017). No other significant correlations were observed between immunohistochemical and histopatho- logical score values (Supplemental Figure S1). Discussion The pathogenesis of neurobrucellosis is not completely under- stood.43 Bacteremia during brucellosis facilitates Brucella invasion of the CNS in a phagocyte-facilitated manner.14 In vitro investigations demonstrated that B. abortus transverses human brain microvascular endothelial cell monolayers using monocytes as a Trojan horse,43 infecting endothelial cells, and trans-endothelially in inflammatory cells.34,42,51 Localization of Brucella in the CNS lesions of dolphins was observed intracel- lularly in macrophages surrounding vascular areas in most cases in our study, but the presence of Brucella antigens was lower than expected considering the severity of the lesions. This agrees with the reported fastidious histochemical and immunohistochemical detection of Brucella in neurobrucello- sis in both humans and dolphins,21,25,38,54,56 supporting the hypothesis that the pathological consequences of neurobrucel- losis are more related to the immune response elicited by the host than to the presence of intralesional bacteria.4,46 Wildlife often exhibit comorbidities; in this study, in cases where co-infections were observed, cetacean neuropathogens were not detected by IHC. No anatomical predisposition within CNS was observed in our study; notably, the cerebrum was not affected in 2 cases. The results obtained coincide with previous investigations that reported that dolphins suffering from neuro- brucellosis develop inflammation in the leptomeninges, choroid plexuses, ependyma, and in a perivascular distribution within the neuroparenchyma.1,9,18,21,23,25,26,28,56 Leptomeninges are the preferred location for B. ceti to elicit an inflammatory response in dolphins. Concordantly, Brucella has a great affinity for lep- tomeninges according to reported biopsies or autopsies in human cases of neurobrucellosis.14,31,38 Lymphocytes stand out as the main cellular population in cetacean1,9,21,25,26,28,56 and human2,38,54 neurobrucellosis, even in the cerebrospinal fluid in humans.14,31,33,36,38,42 T-cells account for the most abundant lymphocyte population in human neuro- brucellosis cases.24,54 T-cells are pivotal for controlling intra- cellular pathogens in the CNS.54 However, autoreactive T-cells also facilitate tissue damage mediated by excessive inflamma- tion.49 B-cells have been reported in the leptomeninges in human neurobrucellosis patients in a lower proportion com- pared with T-cells.49 Conversely, we observed a significant B-cell population in cetacean neurobrucellosis, occasionally forming tertiary follicles in the leptomeninges. Tertiary folli- cles contribute to inflammation in the CNS acting as antigen presenters, proinflammatory cytokine secretors, and antibody producers in persistent infections.45 Therefore, B-cell tertiary follicles in cetacean neurobrucellosis may be indicators of dis- ease chronicity or severity, mimicking their function in other organs, and in the CNS in some human diseases.45 T-cell production of TNF-α and IFN-γ stimulate natural killer cells and macrophages to phagocytize and produce nitric oxide to control Brucella infection.5,27,30 Chronic brucellosis cannot be overcome by increasing the production of IFN-γ because Brucella inhibit the IFN-γ-mediated activation of mac- rophages and/or accelerates the inactivation of IFN-γ-activated macrophages.5,30 Therefore, the expression of IFN-γ would be inefficient to control cetacean neurobrucellosis based on the intralesional presence of Brucella antigens. Indeed, rats chroni- cally infected (2 months) with B. melitensis do not show an induction of the IFN-γ mRNA levels in the brain.41 The expres- sion of IBA-1 was correlated with the rate of leptomeningitis, suggesting that macrophages/microglia contribute in a relevant manner to the severity of inflammation. Unlike cetaceans, some human neurobrucellosis cases report cerebral non-necrotizing Figure 3.  Immunohistochemical findings in striped dolphins (Stenella coeruleoalba) with neurobrucellosis. Semiquantitative results (0-3) of immunohistochemical expression of IBA-1, CD3, CD20, tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-γ), and interleukin-2 (IL-2). Rebollada-Merino et al 9 granulomas or abscesses2,14,33,54,57 with occasional multinucle- ated giant cells57 or nongranulomatous inflammation of the leptomeninges.38,54 Brucella infection of monocytes in vitro precedes the infec- tion of astrocytes and microglia.43 Microgliosis and astrocytosis occur during human neurobrucellosis,2,17,33,38,54,57 in macaques infected with B. melitensis,34 and in the cetaceans studied here. B. abortus has been reported to target microglia in a preferential manner,17 and most of the neuronal death observed in neurobru- cellosis seem to occur due to the microglial release of nitric oxide.50 Thus, microglial nodules and neuronal death observed in dolphins are likely consequences of Brucella infection of microglia rather than due to a direct neurotropic effect, as sug- gested in mice that showed absence of neuronal death after B. abortus infection if neurons were not cocultured with microg- lia.50 Both astrocyte proliferation and apoptosis have been reported in neurobrucellosis. On one hand, astrocytes, but not microglia, are susceptible to B. abortus-induced apoptosis.17 On the other hand, astrocytic proliferation in neurobrucellosis fol- lows the secretion of proinflammatory cytokines by microglia.17 The expression of IL-2 in striped dolphin neurobrucellosis cases suggests a role of this cytokine in differentiation, stimula- tion, and amplification of T-cells in response to brucellosis.6 Microglia in cetacean neurobrucellosis may express IL-2 upon stimulation of Brucella lipopolysaccharide, as reported in mice,53 to stimulate T-cell to produce TNF-α.27 However, mice microglia and astrocytes can secrete TNF-α in vitro if stimulated by Brucella outer membrane proteins (Omp31 and Omp25),35,62 B. abortus Omp19,4,44 or B. abortus.17,37,42 TNF-α mRNA levels have been found to increase in the brain of rats chronically infected (1 month) with B. melitensis48 and in a murine monocyte-macrophage cell line (Raw 264.7 cells) infected with B. suis S2.64 Similarly, TNF-α was found to be overexpressed in the CNS of the B. ceti-infected dolphins studied here. TNF-α not only activates effector cells that limit Brucella survival,6 as is suggested here based on the correla- tion between TNF-α and the severity of perivascular cuffing, but also activates brain endothelium increasing endothelial permeabil- ity of the blood-brain barrier in vitro.42 A limitation of this study is that the specificity of antibodies could not be assessed without specific purified antigens. Despite several antibodies in ruminants (Ruminantiamorpha) recognize conserved epitopes in cetaceans (Cetancodontamorpha),15 some ruminant-targeted antibodies used (IL-1β and IL-6) did not react in cetacean tissues despite their moderate to high protein sequence identity. The contribution of IL-1β and IL-6 in ceta- cean neurobrucellosis is warranted given that these cytokines are upregulated in vivo and in vitro in brucellosis in humans and mice according to genomic and serologic studies.33,37,42,55,63–65 Stroke, aneurisms, and submeningeal hemorrhage have been attributed to a Brucella-induced vasculitis in some human neurobrucellosis cases based on magnetic resonance imag- ing,3,7,31,55 but these have not been reported in cetacean neurobrucellosis. However, excessive inflammation and neuro- inflammation in the CNS lead to blood-brain barrier dysfunc- tion manifesting as edema and hemorrhage.49 Brucella is known to infect endothelial cells to traverse the blood-brain barrier.34,51 Furthermore, in vitro studies showed that B. abor- tus stimulates microglia to produce IL-1β, activating the brain microvasculature in vitro,42 but also platelets that activate human brain microvascular endothelial cells and promote trans-endothelial migration of inflammatory cells.42,51 Both mechanisms explain the blood-brain barrier activation and dys- function reported during neurobrucellosis,42 as well as the hemodynamic changes of the dolphins here studied. Perivascular inflammation in cetacean neurobrucellosis proba- bly occurs due to the trans-endothelial migration of inflamma- tory cells in neurobrucellosis.51 Rarely, human neurobrucellosis can induce CNS demyelin- ation;14,31 however, this finding is not considered an outstand- ing histological feature based on studies performing specific myelin stains (luxol fast blue).54 Demyelination was not wide- spread in the CNS of dolphins in our study. However, periph- eral nervous system demyelination can occur in human patients suffering from brucellosis because of physical compression of the spinal cord or a Guillain-Barré-like syndrome leading to neuropathies.4,14,31 Therefore, peripheral nervous system sam- pling and evaluation should be systematically performed in future cetacean neurobrucellosis cases to assess whether lesions also occur in the peripheral nervous system of this species. To conclude, neurobrucellosis is a shared condition of ceta- ceans and humans. The immune response in neurobrucellosis of striped dolphins mimics in vitro and in vivo findings of neu- robrucellosis in humans and animal models. Inflammation of the leptomeninges, ependyma, and/or choroid plexus contained macrophages/microglia, T-cells, and B-cells. B-cells occasion- ally formed tertiary follicles. GFAP expression demonstrated astrocytosis. Expression of TNF-α, IFN-γ, and IL-2 indicates an intense proinflammatory response. Cetacean disease sur- veillance is valuable not only for diagnosing threats to endan- gered species but can also be exploited to expand the knowledge of the pathogenesis and immunology of infectious diseases, particularly brucellosis, under a One Health approach. Author Contributions AR-M, MD, and AR-B designed and performed the experiments; FG, MP, LM, UR-C, CC, VM, FDN, GL, AP, LD, and CG contributed to the experimental design; AR-M, FG, LM, MD, and AR-B performed histo- logic evaluations; AG-B performed statistical analysis; the manuscript was written by AR-M with contribution from the other authors. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) received no financial support for the research, author- ship, and/or publication of this article. ORCID iDs Agustín Rebollada-Merino https://orcid.org/0000-0003-4304-2713 Alberto Gomez-Buendia https://orcid.org/0000-0002-8968-0836 10 Veterinary Pathology 00(0) References 1. Alba P, Terracciano G, Franco A, et al. The presence of Brucella ceti ST26 in a striped dolphin (Stenella coeruleoalba) with meningoencephalitis from the Mediterranean Sea. Vet Microbiol. 2013;164(1–2):158–163. 2. Alqwaifly M, Al-Ajlan FS, Al-Hindi H, et al. Central nervous system bru- cellosis granuloma and white matter disease in immunocompromised patient. Emerg Infect Dis. 2017;23(6):978–981. 3. Al-Sous MW, Bohlega S, Al-Kawi MZ, et al. Neurobrucellosis: clinical and neuroimaging correlation. AJNR Am J Neuroradiol. 2004;25(3):395–401. 4. Baldi PC, Giambartolomei GH. Pathogenesis and pathobiology of zoonotic brucellosis in humans. Rev Sci Tech. 2013;32(1):117–125. 5. Baldwin CL, Jiang X, Fernandes DM. Macrophage control of Brucella abor- tus: influence of cytokines and iron. Trends Microbiol. 1993;1(3):99–104. 6. Carvalho Neta AV, Mol JP, Xavier MN, et al. Pathogenesis of bovine brucel- losis. Vet J. 2010;184(2):146–155. 7. Chaudhary H, Seetharaman K, Kasinathan A, et al. Transverse sinus thrombo- sis in neurobrucellosis: a rare complication. Neurologist. 2019;24(3):109–110. 8. Cloeckaert A, Verger J-M, Grayon M, et al. Restriction site polymorphism of the genes encoding the major 25 KDa and 36 KDa outer-membrane proteins of Brucella. Microbiol. 1995;141:2111–2121. 9. Cuvertoret-Sanz M, López-Figueroa C, O’Byrne A, et al. Causes of cetacean stranding and death on the Catalonian coast (western Mediterranean Sea), 2012-2019. Dis Aquat Organ. 2020;142:239–253. 10. Dagleish MP, Barley J, Finlayson J, et al. Brucella ceti associated pathology in the testicle of a harbour porpoise (Phocoena phocoena). J Comp Pathol. 2008;139 (1):54–59. 11. Davison NJ, Dagleish MP, Dale EJ, et al. First confirmed reports of the isola- tion of Brucella ceti from a Risso’s dolphin Grampus griseus and a killer whale Orcinus orca. Dis Aquat Organ. 2021;145:191–195. 12. Di Francesco G, Petrini A, D’Angelo AR, et al. Immunohistochemical inves- tigations on Brucella ceti-infected, neurobrucellosis-affected striped dolphins (Stenella coeruleoalba). Vet Ital. 2019;55(4):363–367. 13. Di Guardo G, Mazzariol S. Commentary: advancement of knowledge of Brucella over the past 50 years. Front Vet Sci. 2015;2:27. 14. Drevets DA, Leenen PJ, Greenfield RA. Invasion of the central nervous sys- tem by intracellular bacteria. Clin Microbiol Rev. 2004;17(2):323–347. 15. Elnaggar MM, Abdellrazeq GS, Sacco RE, et al. Comparative analysis of the specificity of monoclonal antibodies developed against the bottlenose dol- phin, Tursiops truncatus, TNF-α, IL1-β, IL-6, IL-8, IL-10 with monoclonal antibodies made against ovine IFN-γ bovine IL-17A and IL-1β revealed they recognize epitopes conserved on dolphin and bovine orthologues. Vet Immunol Immunopathol. 2022;250:110456. 16. Foster G, Osterman BS, Godfroid J, et al. Brucella ceti sp. nov. and Brucella pinnipedialis sp. nov. for Brucella strains with cetaceans and seals as their preferred hosts. Int J Syst Evol Microbiol. 2007;57(pt 11):2688–2693. 17. García Samartino C, Delpino MV, Pott Godoy C, et al. Brucella abortus induces the secretion of proinflammatory mediators from glial cells leading to astrocyte apoptosis. Am J Pathol. 2010;176(3):1323–1338. 18. Garofolo G, Petrella A, Lucifora G, et al. Occurrence of Brucella ceti in striped dolphins from Italian Seas. PLoS ONE. 2020;15(10):e0240178. 19. Giorda F, Romani-Cremaschi U, Marsh AE, et al. Evidence for unknown sarcocystis-like infection in stranded striped dolphins (Stenella coeruleoalba) from the Ligurian sea, Italy. Animals (Basel). 2021;11(5):1201. 20. González L, Patterson IA, Reid RJ, et al. Chronic meningoencephalitis asso- ciated with Brucella sp. infection in live-stranded striped dolphins (Stenella coeruleoalba). J Comp Pathol. 2002;126(2–3):147–152. 21. González-Barrientos R, Morales JA, Hernández-Mora G, et al. Pathology of striped dolphins (Stenella coeruleoalba) infected with Brucella ceti. J Comp Pathol. 2010;142(4):347–352. 22. Granados-Zapata A, Robles-Malagamba MJ, González-Barrientos R, et al. Pathological studies and postmortem computed tomography of dolphins with meningoencephalomyelitis and osteoarthritis caused by Brucella ceti. Oceans. 2022;3:189–203. 23. Grattarola C, Petrella A, Lucifora G, et al. Brucella ceti infection in striped dol- phins from Italian seas: associated lesions and epidemiological data. Animals. 2023;12:1034. 24. Guven T, Ugurlu K, Ergonul O, et al. Neurobrucellosis: clinical and diagnostic features. Clin Infect Dis. 2013;56(10):1407–1412. 25. Guzmán-Verri C, González-Barrientos R, Hernández-Mora G, et al. Brucella ceti and brucellosis in cetaceans. Front Cell Infect Microbiol. 2012;2:3. 26. Hernández-Mora G, González-Barrientos R, Morales JA, et al. Neurobrucellosis in stranded dolphins, Costa Rica [published correction appears in Emerg Infect Dis. 2008;14(11):1825. Baquero-Calvo E [corrected to Barquero-Calvo E]]. Emerg Infect Dis. 2008;14(9):1430–1433. 27. Horiuchi T, Mitoma H, Harashima S, et al. Transmembrane TNF-alpha: struc- ture, function and interaction with anti-TNF agents. Rheumatology (Oxford). 2010;49(7):1215–1228. 28. Isidoro-Ayza M, Ruiz-Villalobos N, Pérez L, et al. Brucella ceti infection in dolphins from the Western Mediterranean sea. BMC Vet Res. 2014;10:206. 29. Jauniaux TP, Brenez C, Fretin D, et al. Brucella ceti infection in harbor por- poise (Phocoena phocoena). Emerg Infect Dis. 2010;16(12):1966–1968. 30. Kim S, Lee DS, Watanabe K, et al. Interferon-gamma promotes abortion due to Brucella infection in pregnant mice. BMC Microbiol. 2005;5:22. 31. Kizilkilic O, Calli C. Neurobrucellosis. Neuroimaging Clin N Am. 2011;21(4):927–937. 32. Krafft A, Lichy JH, Lipscomb TP, et al. Postmortem diagnosis of morbillivirus infection in bottlenose dolphins (Tursiops truncatus) in the Atlantic and Gulf of Mexico epizootics by polymerase chain reaction-based assay. J Wildl Dis. 1995;31(3):410–415. 33. Krishnan C, Kaplin AI, Graber JS, et al. Recurrent transverse myelitis follow- ing neurobrucellosis: immunologic features and beneficial response to immu- nosuppression. J Neurovirol. 2005;11(2):225–231. 34. Lee KM, Chiu KB, Sansing HA, et al. Aerosol-induced brucellosis increases TLR-2 expression and increased complexity in the microanatomy of astroglia in rhesus macaques. Front Cell Infect Microbiol. 2013;3:86. 35. Ma QL, Liu AC, Ma XJ, et al. Brucella outer membrane protein Omp25 induces microglial cells in vitro to secrete inflammatory cytokines and inhibit apoptosis. Int J Clin Exp Med. 2015;8(10):17530–17535. 36. Mahajan SK, Sharma A, Kaushik M, et al. Neurobrucellosis: an often forgotten cause of chronic meningitis. Trop Doct. 2016;46(1):54–56. 37. Marim FM, Franco MMC, Gomes MTR, et al. The role of NLRP3 and AIM2 in inflammasome activation during Brucella abortus infection. Semin Immunopathol. 2017;39(2):215–223. 38. Martínez-Chamorro E, Muñoz A, Esparza J, et al. Focal cerebral involve- ment by neurobrucellosis: pathological and MRI findings. Eur J Radiol. 2002;43(1):28–30. 39. Mauroo NF, Poon RWS, Beh PSL, et al. Detection of Brucella ceti in two Indo-Pacific Finless porpoises (Neophocaena phocaenoides) stranded in Hong Kong. J Wildl Dis. 2020;56(3):698–701. 40. McFee WE, Wu D, Colegrove K, et al. Occurrence of Brucella ceti in stranded bottlenose dolphins Tursiops truncatus coincides with calving season. Dis Aquat Organ. 2020;141:185–193. 41. Melek IM, Erdogan S, Celik S, et al. Evaluation of oxidative stress and inflammation in long term Brucella melitensis infection. Mol Cell Biochem. 2006;293(1–2):203–209. 42. Miraglia MC, Costa Franco MM, Rodriguez AM, et al. Glial cell-elicited activation of brain microvasculature in response to Brucella abortus infec- tion requires ASC inflammasome-dependent IL-1β production. J Immunol. 2016;196(9):3794–3805. 43. Miraglia MC, Rodriguez AM, Barrionuevo P, et al. Brucella abortus traverses brain microvascular endothelial cells using infected monocytes as a Trojan horse. Front Cell Infect Microbiol. 2018;8:200. 44. Miraglia MC, Scian R, Samartino CG, et al. Brucella abortus induces TNF- α-dependent astroglial MMP-9 secretion through mitogen-activated protein kinases. J Neuroinflammation. 2013;10:47. 45. Mitsdoerffer M, Peters A. Tertiary lymphoid organs in central nervous system autoimmunity. Front Immunol. 2016;7:451. Rebollada-Merino et al 11 46. Olsen SC, Palmer MV. Advancement of knowledge of Brucella over the past 50 years. Vet Pathol. 2014;51(6):1076–1089. 47. Povar J, Aguirre JM, Arazo P, et al. Brucelosis con afectación del sistema nervioso [Brucellosis with nervous system involvement]. An Med Intern. 1991;8(8):387–390. 48. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing; 2019. Accessed June 14, 2023. https:// www.R-project.org/. 49. Ransohoff RM, Brown MA. Innate immunity in the central nervous system. J Clin Invest. 2012;122(4):1164–1171. 50. Rodríguez AM, Delpino MV, Miraglia MC, et al. Brucella abortus-acti- vated microglia induce neuronal death through primary phagocytosis. Glia. 2017;65(7):1137–1151. 51. Rodríguez AM, Trotta A, Melnyczajko AP, et al. Brucella abortus-stimulated platelets activate brain microvascular endothelial cells increasing cell transmi- gration through the Erk1/2 pathway. Pathogens. 2020;9(9):708. 52. Sacchini S, Herráez P, Arbelo M, et al. Methodology and neuromarkers for cetaceans’ brains. Vet Sci. 2022;9(2):38. 53. Sawada M, Suzumura A, Marunouchi T. Induction of functional interleukin-2 receptor in mouse microglia. J Neurochem. 1995;64(5):1973–1979. 54. Seidel G, Pardo CA, Newman-Toker D, et al. Neurobrucellosis presenting as leukoencephalopathy: the role of cytotoxic T lymphocytes. Arch Pathol Lab Med. 2003;127(9):e374–e377. 55. Shehata GA, Abdel-Baky L, Rashed H, et al. Neuropsychiatric evaluation of patients with brucellosis. J Neurovirol. 2010;16(1):48–55. 56. Sierra E, Fernández A, Felipe-Jiménez I, et al. Histopathological dif- ferential diagnosis of meningoencephalitis in cetaceans: Morbillivirus, Herpesvirus,Toxoplasma gondii, Brucella sp., and Nasitrema sp. Front Vet Sci. 2020;7:650. 57. Sohn AH, Probert WS, Glaser CA, et al. Human neurobrucellosis with intrace- rebral granuloma caused by a marine mammal Brucella spp. Emerg Infect Dis. 2003;9(4):485–488. 58. Suárez-Esquivel M, Ruiz-Villalobos N, Jiménez-Rojas C, et al. Brucella neo- tomae infection in humans, Costa Rica [published correction appears in Emerg Infect Dis. 2017;23(8):1435]. Emerg Infect Dis. 2017;23(6):997–1000. 59. VanDevanter DR, Warrener P, Bennett L, et al. Detection and analysis of diverse herpesviral species by consensus primer PCR. J Clin Microbiol. 1996;34(7):1666–1671. 60. Verna F, Giorda F, Miceli I, et al. Detection of morbillivirus infection by RT-PCR RFLP analysis in cetaceans and carnivores [published correction appears in J Virol Methods. 2018;262:89]. J Virol Methods. 2017;247:22–27. 61. Vitale M, Galluzzo P, Currò V, et al. A high sensitive nested PCR for Toxoplasma gondii detection in animal and food samples. J Microb Biochem Technol. 2013;5:39–41. 62. Wang Z, Wang G, Wang Y, et al. Omp31 of Brucella inhibits NF-κB p65 signaling pathway by inducing autophagy in BV-2 microglia. Neurochem Res. 2021;46(12):3264–3272. 63. Yang H, Wang Z, Shi S, et al. Identification of cerebrospinal fluid metabolites as biomarkers for neurobrucellosis by liquid chromatography-mass spectrom- etry approach. Bioengineered. 2022;13(3):6996–7010. 64. Zhi F, Zhou D, Li J, et al. Omp16, a conserved peptidoglycan-associated lipopro- tein, is involved in Brucella virulence in vitro. J Microbiol. 2020;58(9):793–804. 65. Zhou D, Zhi F, Fang J, et al. RNA-Seq analysis reveals the role of Omp16 in Brucella-infected RAW264.7 cells. Front Vet Sci. 2021;8:646839.