UN CO RR EC TE D PR OO F Journal of Colloid and Interface Science xxx (2018) xxx-xxx Contents lists available at ScienceDirect Journal of Colloid and Interface Science journal homepage: www.elsevier.com Regular Article Effects of a mesoporous bioactive glass on osteoblasts, osteoclasts and macrophages N. Gómez-Cerezoa, b, L. Casarrubiosc, I. Moralesc, M.J. Feitoc, M. Vallet-Regía, b, ⁎, D. Arcosa, b, ⁎, M.T. Portolésc, ⁎ a Departamento de Química en Ciencias Farmacéuticas, Facultad de Farmacia, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria Hospital 12 de Octubre i+12, Plaza Ramón y Cajal s/n, 28040 Madrid, Spain b CIBER de Bioingeniería, Biomateriales y Nanomedicina, CIBER-BBN, Madrid, Spain c Departamento de Bioquímica y Biología Molecular, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), 28040 Madrid, Spain A R T I C L E I N F O Article history: Received 18 April 2018 Received in revised form 23 May 2018 Accepted 27 May 2018 Available online xxx Keywords: Mesoporous bioactive glasses Osteoblasts Osteoclasts Macrophages A B S T R A C T A mesoporous bioactive glass (MBG) of molar composition 75SiO2-20CaO-5P2O5 (MBG-75S) has been syn- thetized as a potential bioceramic for bone regeneration purposes. X-ray diffraction (XRD), Fourier trans- form infrared spectroscopy (FT-IR), nitrogen adsorption studies and transmission electron microscopy (TEM) demonstrated that MBG-75S possess a highly ordered mesoporous structure with high surface area and poros- ity, which would explain the high ionic exchange rate (mainly calcium and silicon soluble species) with the surrounded media. MBG-75S showed high biocompatibility in contact with Saos-2 osteoblast-like cells. Con- centrations up to 1mg/ml did not lead to significant alterations on either morphology or cell cycle. Regard- ing the effects on osteoclasts, MBG-75S allowed the differentiation of RAW-264.7 macrophages into osteo- clast-like cells but exhibiting a decreased resorptive activity. These results point out that MBG-75S does not inhibit osteoclastogenesis but reduces the osteoclast bone-resorbing capability. Finally, in vitro studies fo- cused on the innate immune response, evidenced that MBG-75S allows the proliferation of macrophages with- out inducing their polarization towards the M1 pro-inflammatory phenotype. This in vitro behavior is indica- tive that MBG-75S would just induce the required innate immune response without further inflammatory com- plications under in vivo conditions. The overall behavior respect to osteoblasts, osteoclasts and macrophages, makes this MBG a very interesting candidate for bone grafting applications in osteoporotic patients. © 2018. 1. Introduction Mesoporous bioactive glasses (MBGs) are bioceramics intended for bone tissue regeneration purposes. Discovered in 2004 by Zhao et al. [1], MBGs mean a significant upgrade respect to the conven- tional sol-gel bioactive glasses prepared by Li et al. in 1991 [2]. Sim- ilarly, to sol-gel bioactive glasses, MBGs are commonly prepared in the ternary system SiO2-CaO-P2O5 [3–5] and, in the last decade, dif- ferent research groups have incorporated different ions with poten- tial therapeutic properties [6–11]. In the case of MBGs, the incorpo- ration of a structure directing agent (SDA) to the synthesis results in the formation of an ordered mesophase by the self-organization of the SDA into micelles. Soluble silica, phosphate and calcium species con- densates around this organic template, which leads to a mesoporous ⁎ Corresponding authors at: Departamento de Química en Ciencias Farmacéuticas, Facultad de Farmacia, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria Hospital 12 de Octubre i+12, Plaza Ramón y Cajal s/n, 28040 Madrid, Spain (M. Vallet-Regí and D. Arcos). Departamento de Bioquímica y Biología molecular, Facultad de Ciencias Químicas, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), 28040 Madrid, Spain (M.T. Portolés) Email addresses: vallet@ucm.es (M. Vallet-Regí); arcosd@ucm.es (D. Arcos); portoles@quim.ucm.es (M.T. Portolés) structure after calcination, thus providing higher textural properties compared to conventional sol-gel bioactive glasses [12,13]. The primary consequence on the biological behavior is a faster and more intense ionic exchange (mainly Ca2+ and silica species) between the MBG and the surrounding fluids [14]. In fact, some MBGs have shown the fastest in vitro bioactive behavior when soaked in simu- lated body fluid, in terms of the nucleation and growth of a carbonate nanocrystalline apatite on their surface, very similar to the biological one found in bones [15]. However, the MBG surface reactivity is not their only action mechanism. The ions released from MBG also stim- ulate the expression of several genes of osteoblastic cells and induce angiogenesis both in vitro and in vivo [16,17]. Recent studies suggest that these ions could also regulate immune responses by altering the ionic microenvironment between the implants and hosts [18]. The im- portance of the immune response during biomaterial-mediated osteo- genesis makes necessary the evaluation of the osteoimmunomodula- tory properties of biomaterials for bone tissue [19]. Recently, the in vivo response to these materials has been stud- ied in different animal models, evidencing certain advantages respect to other bioceramics. Due to their potential bone regeneration capa- bilities, MBGs are being considered as bone grafts in the case of os- teoporotic patients. Osteoporosis is produced by the bone remodeling https://doi.org/10.1016/j.jcis.2018.05.099 0021-9797/ © 2018. UN CO RR EC TE D PR OO F 2 Journal of Colloid and Interface Science xxx (2018) xxx-xxx disruption that is due to either increased bone resorption by osteo- clasts or decreased new bone formation by osteoblasts or both [20]. The biomaterials most commonly employed for treatment of osteo- porotic bone and bone regeneration have been designed to stimulate the osteogenesis process and bone formation by osteoblasts. For this reason, osteoblasts are commonly used for the in vitro evaluation of bone materials [21] but few studies are focused on the effects of these biomaterials on bone resorbing osteoclasts [22]. Osteoclasts are multi- nucleated giant cells which differentiate from hematopoietic stem cells of the monocyte/macrophage lineage through sequential steps [23] regulated by several growth factors and cytokines expressed by dif- ferent bone cell types [24,25]. Osteoclasts can also differentiate in vitro from macrophages by stimulation with the macrophage/mono- cyte-colony-stimulating factor (M-CSF) and the receptor activator of nuclear factor kappa-B ligand (RANKL) [22]. These agents induce the fusion of pre-osteoclasts, which become multinucleated cells, and the formation of “ruffled membrane”, critical for bone resorption, that in- volves the tight attachment of osteoclasts to the bone surface to create the “sealing zone” rich in F-actin [26]. During bone resorption, osteo- clasts isolate the resorptive space from the surrounding bone and re- lease matrix-degrading enzymes, hydrogen ions and chloride ions in- side the sealing zone, producing the bone matrix degradation and the dissolution of the bone mineral component, respectively [27]. The effects of MBGs on osteoblasts have been widely evaluated by different research groups [7,12,13,38,41], whereas the effects on other cell types involved in bone remodeling are practically unknown. The present study is focused on the effects of a potential mesoporous bioactive glass for bone regeneration, with molar composition 75- SiO2-20CaO-5P2O5 (MBG-75S), on osteoclast differentiation, bone resorption activity and macrophage activation towards pro-inflamma- tory M1 phenotype. CaO plays a fundamental role in the biologi- cal properties of SiO2-CaO-P2O5 MBGs. However, previous works demonstrated that compositions with higher CaO content led to dis- ordered mesoporous structures [4]. The composition MBG-75S was chosen with the aim of ensuring enough CaO content while keeping the highly ordered mesoporous structure. Previously, the dose-depen- dent action of this powdered material on osteoblasts has been evalu- ated through the analysis of cell cycle, morphology, size, complexity and apoptosis after the treatment with different doses of MBG-75S. 2. Materials and methods 2.1. Synthesis and characterization of MBG-75S Mesoporous bioactive glass MBG-75S with molar composition 75- SiO2-20CaO-5P2O5 was prepared by EISA method and using Pluronic F127 as structure directing agent. For this purpose, 32g of Pluronic F127 was dissolved in an ethanol-HCl (0.5 M) solution. Thereafter, 61.3ml of tetraethylorthosilane (TEOS), 6.28ml of triethylphosphate (TEP) and 17.6mg of Ca(NO3)2·4H2O were gradually added in 3h in- tervals. The mixture was stirred for 24h, poured into Petri dishes (9 cm in diameter) and introduced in an incubator at 30°C for 7days, un- til solvent evaporation and gelling. The transparent membranes so ob- tained were calcined at 700 °C for 3h under air atmosphere. The resul- tant powder was gently milled in dry conditions and sieved, collecting the grain fraction below 40µm. Chemicals of highest purity available have been used in the present study. X-ray diffraction experiment was carried out in a Philips X’Pert diffractometer equipped with a Cu Kα radiation (wavelength 1.5406Å). The patterns were collected between 0.5 and 6.5 2θ° angle using a Bragg-Brentano geometry. Fourier-transform infrared spec- troscopy was done using a Nicolet Magma IR 550 spectrometer and using the attenuated total reflectance (ATR) sampling technique with a Golden Gate accessory. Nitrogen adsorption/desorption isotherm was obtained with an ASAP 2020 equipment. The MBG-75S was previously degassed un- der vacuum for 15h, at 150°C. The surface area was determined using the Brunauer-Emmett-Teller (BET) method. The pore size distribution between 0.5 and 40nm was determined from the adsorption branch of the isotherm by means of the Barret-Joyner-Halenda (BJH) method. The surface area was calculated by the BET method and the pore size distribution was determined by the BJH method using the adsorption branch of the isotherm. Scanning electron microscopy (SEM) was carried out using a JEOL-6335F microscope, operating at 15kV. Transmission electron microscopy (TEM) was carried out using a JEOL-1400 microscope, operating at 300kV (Cs 0.6mm, resolution 1.7Å). Images were recorded using a CCD camera (model Keen view, SIS analyses size 1024 X 1024, pixel size 23.5mm × 23.5mm) at 60,000× magnification using a low-dose condition. 2.2. Soluble species release from MBG-75S to the culture medium The levels of soluble calcium, phosphates and silica species in the culture medium were measured by inductively coupled plasma (ICP) spectroscopy, after soaking MBG-75S in Dulbeccós Modified Eaglés Medium (DMEM, Sigma Chemical Company, St. Louis, MO, USA) (1 mg/ml) for 3 and 7days. 2.3. Culture of human Saos-2 osteoblasts Human Saos-2 osteoblasts (105 cells/ml) were cultured in the pres- ence of 0.5, 1 and 2mg/ml of powdered MBG-75S in Dulbecco’s Modified Eaglés Medium (DMEM, Sigma Chemical Company, St. Louis, MO, USA) supplemented with 10% (vol/vol) fetal bovine serum (FBS, Gibco, BRL), 1mM L-glutamine (BioWhittaker Europe, Belgium), penicillin (200 μg/ml, BioWhittaker Europe, Belgium), and streptomycin (200μg/ml, BioWhittaker Europe, Belgium), under a 5% CO2 atmosphere and at 37°C. Controls in the absence of material were carried out in parallel. After 24h, the culture medium was aspirated, the cells were washed with phosphate-buffered saline (PBS) and har- vested using 0.25% trypsin-ethylene diamine tetraacetic acid (EDTA). Cell suspensions were centrifuged at 310g for 10min and resuspended in fresh medium for the analysis of cell cycle, apoptosis, cell size and complexity by flow cytometry as described below. 2.4. Cell-cycle and apoptosis analysis by flow cytometry Cells were resuspended in PBS (0.5ml) and incubated with 4.5ml of ethanol 70% during 4h at 4°C. Then, cells were centrifuged at 310g for 10min, washed with PBS and resuspended in 0.5ml of PBS with Tritón X-100 0.1%, propidium iodide (IP) 20μg/ml and ribonuclease (RNAse) 0.2mg/ml (Sigma-Aldrich, St. Louis, MO, USA). After in- cubation at 37°C for 30min, the fluorescence of PI was excited by a 15mW laser tunning to 488nm and the emitted fluorescence was mea- sured with a 585/42 band pass filter in a FACScan Becton Dickinson flow cytometer. The cell percentage in each cycle phase: G0/G1, S and G2/M was calculated with the CellQuest Program of Becton Dickinson and the SubG1 fraction (cells with fragmented DNA) was used as in- dicative of apoptosis. For statistical significance, at least 10,000 cells were analyzed in each sample. UN CO RR EC TE D PR OO F Journal of Colloid and Interface Science xxx (2018) xxx-xxx 3 2.5. Cell size and complexity detection by flow cytometry Forward angle (FSC) and side angle (SSC) scatters were evaluated as indicative of cell size and complexity, respectively, using a FAC- Scan Becton Dickinson flow cytometer. 2.6. Osteoclast differentiation from murine RAW 264.7 macrophages Murine RAW-264.7 macrophages (2× 104 cells/ml) were seeded on glass coverslips and cultured in the presence of 1mg/ml of pow- dered MBG-75S in Dulbecco's Modified Eagle Medium (DMEM) without phenol red, supplemented with 10% fetal bovine serum (FBS, Gibco, BRL), 1mM L-glutamine (BioWhittaker Europe, Belgium), penicillin (200 μg/ml, BioWhittaker Europe, Belgium), and strepto- mycin (200 μg/ml, BioWhittaker Europe, Belgium). To stimulate os- teoclast differentiation, 40ng/ml of mouse receptor activator of nu- clear factor kappa-B ligand (RANKL) recombinant protein (TRANCE/RANKL, carrier-free, BioLegend, San Diego) and 25ng/ ml recombinant human macrophage-colony stimulating factor (M-CSF, Milipore, Temecula) were added to the culture medium. Cells were cultured under a 5% CO2 atmosphere and at 37 °C for 7days. Controls in the absence of material were carried out in parallel. 2.7. Lactate dehydrogenase (LDH) measurement To evaluate the plasma membrane integrity during osteoclast dif- ferentiation, lactate dehydrogenase (LDH) activity was measured in the culture medium by an enzymatic method at 340nm (Bio-Analítica) using a Beckman DU 640 UV–Visible spectrophotometer. 2.8. Morphological studies by confocal microscopy Cells were seeded on glass coverslips and cultured in the presence of different doses of MBG-75S for 24h. Controls in the absence of material were carried out in parallel. After washing with PBS, cells were fixed with 3.7% paraformaldehyde in PBS for 10min, perme- abilizated with 0.1% Triton X-100 for 3min and preincubated with PBS containing 1% bovine serum albumin (BSA) for 30min. Then, cells were incubated with rhodamine phalloidin (1:40, v/v Molecu- lar Probes) for 20min to stain F-actin filaments. Samples were then washed with PBS and cell nuclei were stained with 3µM 4′-6-di- amidino-2′-phenylindole (DAPI, Molecular Probes) for 5min. After staining and washing with PBS, cells were examined using a Leica SP2 Confocal Laser Scanning Microscope. Rhodamine fluorescence was excited at 540nm and measured at 565nm. DAPI fluorescence was excited at 405nm and measured at 420–480nm. 2.9. Osteoclast resorption activity To evaluate the resorption activity of osteoclasts, RAW-264.7 macrophages were seeded on the surface of nanocrystalline hydrox- yapatite disks and differentiate into osteoclasts in the presence of 1mg/ml of powdered MBG-75S as it is described above. Nanocrys- talline hydroxyapatite disks were prepared by controlled precipitation of calcium and phosphate salts and subsequently heated at temper- atures below the sintering point, as previously described by our re- search group [22]. Controls in the absence of material were carried out in parallel. After 7days of differentiation, cells were detached us- ing cell scrapers and disks were dehydrated, coated with gold-palla dium and examined with a JEOL JSM-6400 scanning electron micro- scope in order to observe the geometry of resorption cavities produced by osteoclasts on the surface of nano-HA disks. 2.10. Detection of pro-inflammatory M1 macrophage phenotype To study the effect of MBG-75S on macrophage polarization to- wards pro-inflammatory M1 phenotype, RAW-264.7 macrophages were cultured with 1mg/ml of this material for 24h in the pres- ence or the absence of E. coli lipopolysaccharide/interferon-γ (250ng/ ml LPS and 100ng/ml IFN, Sigma-Aldrich Corporation, St. Louis, MO, USA) as inflammatory stimuli [28] in Dulbecco's Modified Ea- gle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Gibco, BRL), 1mM L-glutamine (BioWhittaker Europe, Bel- gium), penicillin (200 μg/ml, BioWhittaker Europe, Belgium), and streptomycin (200 μg/ml, BioWhittaker Europe, Belgium) at 37°C under a CO2 (5%) atmosphere. Controls in the absence of material were carried out in parallel. For the analysis of macrophage prolif- eration, the attached RAW-264.7 cells were washed with phosphate buffered saline (PBS), harvested using cell scrapers and counted with a Neubauer hemocytometer. The expression of CD80 as M1 marker [29] was used to detect pro-inflammatory M1 macrophages by flow cytometry and confocal microscopy. For flow cytometry studies, cells were detached, cen- trifuged and incubated in 45µl of staining buffer (PBS, 2.5% FBS Gibco, BRL) with 5µl of normal mouse serum inactivated for 15min at 4 °C in order to block the Fc receptors on the macrophage plasma membrane and to prevent non-specific binding. Then, cells were incu- bated with phycoerythrin (PE) conjugated anti-mouse CD80 antibody (2.5 µg/ml, BioLegend, San Diego, California) for 30min at 4 °C in the dark. Labelled cells were analyzed using a FACSCalibur flow cy- tometer. PE fluorescence was excited at 488nm and measured at 585/ 42nm. For confocal microscopy studies, macrophages cultured on glass coverslips were fixed with 3.7% paraformaldehyde (Sigma-Aldrich Corporation, St. Louis, MO, USA) in PBS for 10min, washed with PBS and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich Cor- poration, St. Louis, MO, USA) for 3min. The samples were then washed with PBS and preincubated with PBS containing 1% BSA (Sigma-Aldrich Corporation, St. Louis, MO, USA) for 30 min to pre- vent non-specific binding. Samples were incubated in 1ml of stain- ing buffer with phycoerythrin (PE) conjugated anti-mouse CD80 anti- body (2.5 µg/ml, BioLegend, San Diego, California) for 30min at 4 °C in the dark. Samples were then washed with PBS and the cell nuclei were stained with 3μM DAPI (4′-6-diamidino-2′-phenylindole, Mol- ecular Probes) for 5min. Samples were examined using a Leica SP2 Confocal Laser Scanning Microscope. PE fluorescence was excited at 488nm and measured at 575–675nm. DAPI fluorescence was excited at 405nm and measured at 420–480nm. 2.11. Statistics Data are expressed as means + standard deviations of a represen- tative of three repetitive experiments carried out in triplicate. Statis- tical analysis was performed by using the Statistical Package for the Social Sciences (SPSS) version 22 software. Statistical comparisons were made by analysis of variance (ANOVA). Scheffé test was used for post hoc evaluations of differences among groups. In all statistical evaluations, p< 0.05 was considered as statistically significant. UN CO RR EC TE D PR OO F 4 Journal of Colloid and Interface Science xxx (2018) xxx-xxx 3. Results and discussion 3.1. Synthesis and characterization of MBG-75S The structural, chemical and textural properties of MBG-75S were determined prior to any cell culture test. Low angle XRD pattern (Fig. 1a) shows two diffraction maxima at 1.05 and 1.75 2θ° that could be assigned to the (1 0) and (1 1) reflections of p6m hexagonal planar group, as previously observed for similar MBGs prepared with F127 as SDA [12]. FTIR spectrum (Fig. 1b) shows the characteristic bands at 500 and 1080cm−1 corresponding to Si-O-Si bending and stretching mode, respectively. The weak adsorption band observed at 590cm−1 corresponds to the bending vibrational modes of PO4 3- groups in an amorphous environment. The isotherm obtained by nitrogen adsorption analysis (Fig. 1c) can be described as a type IV curve, characteristic of mesoporous materials. The isotherm has a type H1 hysteresis loop in the meso- pore range, which is characteristic of cylindrical pores open at both ends, having necks along the pores. The dV/d log D plot (data not shown) showed a monomodal distribution centered around 5.7nm (see Table 1). The structural parameters obtained by XRD together with ni- trogen adsorption data provide valuable information about the MBG structure. Table 1 shows the structural and textural parameters and a scheme of the MBG structure is shown in Fig. 1d. For instance, from the diffraction angle 2θ° of the (1 0) maxima and using the Bragg’s Law where n is a positive integer (in our case 1) and λ is the wavelength of the incident beam (1.5406 Å), the interplanar distance for the (0 1) was calculated as 8.41nm. Considering the hexagonal structure of the p6m planar group, the lattice parameter a of the mesoporous structure can be easily obtained (see Table 1). In a hexagonal planar structure, the lattice parameter corresponds to the distance form center of pore to center of pore, as it is shown in Fig. 1d. Using the pore size provided by the nitrogen adsorption analysis, we can also calculate the thickness of the pore walls. SEM observations (Fig. 2a) show that MBG-75S consist in particles of irregular shape, ranging in size between 10 and 40μm. TEM study confirmed the highly ordered mesoporous structure of the MBG-75S (Fig. 2b), as well as the channel-like morphology of the pores characteristic of a p6m planar group. Taken as a whole, we can describe the structure of MBG-75S as a highly porous material, with a regular arrangement of single modal pore size distribution. Considering the distance between pores and the pore size, we can conclude that the walls of MBG-75S are thinner than the pore diameters. This fact, together with the high surface area, pore volume and channel-like open morphology of the pores (deduced from the H1 hysteresis loop and TEM images) would facilitate the fast ionic dissolution and exchange, in contact with the surrounding fluids under both in vitro and in vivo conditions. 3.2. Dose-dependent effects of MBG-75S on human Saos-2 osteoblasts The dose-dependent action of powdered MBG-75S on osteoblasts has been studied with human Saos-2 cells as in vitro experimental model. This osteosarcoma cell line is commonly used in this kind of in vitro studies due to its osteoblastic properties as production of min- eralized matrix, high alkaline phosphatase levels, PTH receptors and osteonectin presence [30]. All these studies have been performed in Fig. 1. Structural, chemical and textural characterization of MBG-75S; (a) XRD pattern. The Miller indexes for a planar hexagonal unit cell p6m are indicated; (b) FTIR spectra; (c) Nitrogen adsorption/desorption isotherm and (d) a scheme of the MBG-75S porous structure calculated from the mentioned characterization techniques. (1) UN CO RR EC TE D PR OO F Journal of Colloid and Interface Science xxx (2018) xxx-xxx 5 Table 1 Textural and structural parameters for MBG-S75. Surface area (m2 g−1) Pore volume (cm3 g−1) Pore size (nm) d(1 0) (nm) Lattice parameter aa (nm) Wall thicknessb (nm) 305.5 0.46 5.7 8.41 9.71 4.01 a Calculated as a= d(10)·2/√3. b Calculated as a – pore size. direct contact between cells and powdered MBG. By culturing the cells in direct contact with the MBG particles, we can study not only the effects of the ions released from MBG but also the effect of the very fast release associated to the high textural properties of MBG. Different cell parameters (cell cycle, morphology, size, com- plexity and apoptosis) were evaluated after 24h of culture of Saos-2 osteoblasts with increasing doses of MBG-75S. The analysis of the cell cycle by flow cytometry allowed us to know the effects of this MBG on the proliferation of human Saos-2 osteoblasts through progressive stages: G0/G1 phase (Quiescence/ Gap1), S phase (Synthesis) and finally G2/M phase (Gap2 and Mi- tosis). This analysis also indicates the percentage of apoptotic cells with fragmented DNA corresponding to the SubG1 fraction. Figs. 3 and 4 show the cell cycle profiles of osteoblasts and the percentages of cells within each cycle phase, respectively, after 24 h of culture in the absence or the presence of different doses of MBG-75S. As it can be observed in these figures, no alterations were detected in the G0/G1, S and the G2/M phases after treatment with 0.5 and 1mg/ ml of this MBG. However, 2mg/ml induced significant decreases in the G0/G1 phase (p < 0.05) and the G2/M phase (p< 0.005). This ef- fect can be explained by the significant increase (p < 0.005) produced by 2mg/ml in the SubG1 fraction (apoptotic cells). Very low lev- els of apoptosis were detected either in the absence of material or in the presence of 0.5mg/ml and 1mg/ml of MBG-75S (lower than 5% ± 0.5%), Fig. 4). The high apoptosis levels (20% ± 1%) observed with 2mg/ml are probably due to the presence of this material in Fig. 2. SEM micrograph (a) and TEM image (b) obtained for MBG75-S. Fig. 3. Effects of different doses of MBG-75S on cell cycle profile of human Saos-2 osteoblasts after 24h of treatment. UN CO RR EC TE D PR OO F 6 Journal of Colloid and Interface Science xxx (2018) xxx-xxx Fig. 4. Effects of different doses of powdered MBG-75S on cell cycle phases of human Saos-2 osteoblasts after 24h of treatment. Statistical significance: *p< 0.05; *** p< 0.005. powdered form which can produce the loss of cell anchorage, induc- ing a kind of apoptosis defined as anoikis [31,32]. The cell size and complexity of osteoblasts in the absence or the presence of different doses of MBG-75S were also evaluated by flow cytometry through FSC and 90° SSC light scatters, respectively. These properties depend on cell size, plasma membrane and intracellular organelles [33]. Significant effects of MBG-75S on these two pa- rameters (p < 0.005) were observed (Fig. 5), evidencing dose-depen- dent decreases of osteoblast size and complexity induced in by this powdered material. These effects could be due to the loss of cell Fig. 5. Effects of different doses of powdered MBG-75S on cell size and complexity of human Saos-2 osteoblasts after 24h of treatment. Statistical significance: *** p< 0.005. anchorage produced by the powdered material and to changes of the ambient ionic concentration as consequence of ion release from the material. The morphology of human Saos-2 osteoblasts in the presence of different doses of MBG-75S was observed by confocal microscopy af- ter staining with rhodamine-phalloidin (for F-actin filaments in red) and DAPI (for nuclei in blue). The typical characteristics of this cell type were observed in the presence of 0.5 and 1mg/ml. However, 2mg/ml induced cell morphology alterations in agreement with the pronounced apoptosis increase obtained by flow cytometry after treat- ment with this high MBG-75S dose (Fig. 6). Considering the MBG-75S dose-dependent effects observed with Saos-2 osteoblasts, the dose chosen for studying the response of osteo- clasts and macrophages to this material was 1mg/ml. 3.3. Effects of MBG-75S on osteoclast differentiation and resorption activity Osteoclasts can differentiate in vitro from macrophages by stim- ulation with the macrophage/monocyte colony-stimulating factor (M-CSF) and the receptor activator of nuclear factor kappa-B lig- and (RANKL). RAW-264.7 is a mouse macrophage cell line retain- ing many of the characteristics of macrophages in vivo [34]. In the present study, osteoclast-like cells were obtained from RAW-264.7 macrophages after 7days of differentiation with RANKL and M-CSF, in the absence or in the presence of 1mg/ml MBG-75S. The mor- phology of these osteoclast-like cells was evaluated by confocal mi- croscopy (Fig. 7) which allowed us to observe multinucleated cells in the presence and in the absence of MBG-75S, revealing osteoclast-like cell differentiation from RAW macrophages after 7days in both con- ditions. The presence of numerous actin rings, critical to define the sealing zone required for osteoclast resorption activity, was also ob- served (Fig. 7). Two images for each group are shown in order to highlight the presence of multinucleated cells and the actin rings, that are the main characteristics of osteoclasts. A statistical analysis of the multinucleated cells has been carried out in the absence and in the presence of MBG-75S obtaining values of 10% ± 1% of multinucle UN CO RR EC TE D PR OO F Journal of Colloid and Interface Science xxx (2018) xxx-xxx 7 Fig. 6. Effects of different doses of powdered MBG-75S on the morphology of human Saos-2 osteoblasts observed by confocal microscopy after 24h of treatment. Actin was stained with rhodamine-phalloidin (red) and cell nuclei with DAPI (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) ated cells in both cases in agreement with previous studies [22]. Our experiments evidence that the presence of MBG-75S does not inhibit the osteoclastogenesis, at least at the particles concentration used in this work. The resorption cavities left by osteoclast-like cells on nanocrys- talline hydroxyapatite disks after 7days of differentiation, were evalu- ated by SEM. As it can be observed in Fig. 8, the morphology and the number of these cavities evidenced that the resorptive activity is sig- nificantly modified by the presence of MBG-75S. Certainly, the cavi- ties observed in both cases are of 20 to 40µm in size (30 ± 10μm), in agreement with the sizes of the actin rings previously observed (see Fig. 7). However, in the absence of MBG-75S particles, osteoclasts leave marks significantly deeper with well-defined rims (Fig. 8, con- trol images) in comparison with the marks left in the presence of the MBG, where only weak dark contrasts can be observed on the surface of the nano-HA substrate (Fig. 8, right images). The magnification of each image was chosen during the observation of the different samples in order to highlight the characteristics of the cavities left by osteo- clasts in the presence and in the absence of MBG-75S. Higher magni- fications (Fig. 8, bottom images) evidence the lower resorptive activ- ity of osteoclasts differentiated in the presence of MBG-75S. Whereas the mark left without the MBG is a well-outlined cavity, exhibiting several micrometers in depth and a rim indicating the actin sealing area, the mark left in the presence of the MBG is just a superficial ero- sion poorly outlined rims. To know the effects of MBG-75S on plasma membrane integrity during the differentiation process, the lactate dehydrogenase (LDH) released into the culture medium of osteoclast-like cells was mea- sured after 3 and 7days of treatment (Fig. 9). Although the presence of MBG-75S produced a significant increase of LDH levels after 3days, no significant differences between control and treated cells were ob- served after 7days of differentiation, thus indicating the integrity of plasma membrane of these cells after 7days of culture with this mate- rial. To investigate specifically the possible effect of ionic dissolution products of MBG-75S on the resorption activity of osteoclast-like cells, Ca2+, phosphate and soluble silica species levels were measured into the culture medium during the differentiation process after 3 and 7days of treatment with 1mg/ml of this powdered material (Fig. 10). A significant increase of Ca2+, a significant decrease of phosphate and a very pronounced significant increase of silica levels (p < 0.005) were observed after 3 and 7days of treatment with 1mg/ml MBG-75S (Fig. 10). Different authors have demonstrated that the bone-resorbing activ- ity of osteoclasts is regulated by extracellular Ca2+ concentration and that high levels of this ion produce osteoclast retraction and dissipa- tion of sealing zone, decreasing the cell spread area with actin reor- ganization and podosomal disassembly, which resulted in a dramatic reduction of bone resorption [35,36]. The high soluble silica levels de- tected into the medium can also contribute to this effect due to the UN CO RR EC TE D PR OO F 8 Journal of Colloid and Interface Science xxx (2018) xxx-xxx Fig. 7. Effects of 1mg/ml of powdered MBG-75S on the morphology of osteoclast-like cells observed by confocal microscopy after 7days of treatment. Actin was stained with rhodamine-phalloidin (red) and cell nuclei with DAPI (blue). inhibitory action of this ion on the osteoclastic activity [27,37]. In this sense, the ionic exchange between MBG-75S and the surrounded flu- ids would be facilitated by the high surface area, porosity, pore size/ wall thickness ratio and, in general, by the highly ordered mesoporous structure of the MBG-75S. Since Ca2+ and soluble silicate ions also stimulate the prolifera- tion and differentiation of osteoblasts [11,27,38], MBG-75S presents a high potential for bone regeneration due to its capability for releasing these two ions. 3.4. Effects of MBG-75S on polarization of RAW-264.7 macrophages towards pro-inflammatory M1 phenotype The implantation of a biomaterial is accompanied by tissue in- jury through the surgical procedure that initiates an inflammatory re- sponse, starting with the formation of a provisional matrix. Thus, af- ter the first contact between the biomaterial and the tissue, proteins from blood and interstitial fluids adsorb to the biomaterial surface, determining the activation of coagulation cascade, complement sys- tem, platelets and immune cells. These facts result in the formation of a transient provisional matrix and the onset of the inflammatory re- sponse. On the other hand, the immune response is additionally af- fected by the ions and the products eluted from the biomaterial im- planted in the body. All these products could reach the bloodstream affecting lymphocytes and macrophages which release reactive oxy- gen species (ROS) and cytokines, events which play an important role in the inflammatory response towards the biomaterial. An im- mune response involves the action of all types of macrophages, clas- sical activated macrophages (M1) in the early phase and wound-heal- ing macrophages (M2) in the resolution stage. However, when inflam- matory stimuli persist at the implant site, macrophages attached to the biomaterial can foster invasion of additional inflammatory cells by se- creting chemokines like IL-8, MCP-1, MIP-1b leading to chronic in- flammation [39]. In order to know if MBG-75S promotes the pro-inflammatory M1 macrophage phenotype, RAW-264.7 macrophages were cultured in the absence or the presence of 1mg/ml of this material for 24h with- out or with E. coli lipopolysaccharide and interferon-γ, as inflamma- tory stimuli [28]. The expression of CD80 as M1 marker [29], was quantified by flow cytometry and observed by confocal microscopy with a phycoerythrin (PE) conjugated anti-mouse CD80 antibody. Pre- viously, the effects of MBG-75S on RAW-264.7 proliferation were evaluated. As it can be observed in Fig. 11, MBG-75 allowed RAW-264.7 macrophages to proliferate but more slowly than control cells (white), inducing a significant decrease of the cell number (p< 0.005), as it has been previously observed with others powdered materials for bone tissue [40]. In previous studies with other mesoporous bioactive glass MBG-85, the release of high Ca2+ concentrations reduced Saos- UN CO RR EC TE D PR OO F Journal of Colloid and Interface Science xxx (2018) xxx-xxx 9 Fig. 8. Scanning electron microscopy images of the resorption cavities left by osteoclast-like cells cultured on nanocrystalline hydroxyapatite disks after 7days of culture in the ab- sence (control) and the presence of 1mg/ml of powdered MBG-75S. Bottom images shows a higher magnification of a resorption cavity in the absence (control) and the presence of MBG-75S. Fig. 9. Effects of 1mg/ml of powdered MBG-75S on lactate dehydrogenase (LDH) re- leased into the media of osteoclast-like cells during the differentiation process after 3 and 7days of treatment. Controls without material were carried out in parallel (white). Statistical significance: *** p< 0.005. 2 osteoblast proliferation whereas Si concentration did not produce negative effect on the cell proliferation [41]. The effects of 1mg/ml of powdered MBG-75S on pro-inflamma- tory M1 macrophage phenotype were quantified by flow cytometry through the CD80 expression as described above in both basal and LPS/IFN-γ stimulated conditions. As it can be observed in Fig. 12, the addition of LPS/IFN-γ induced a significant increase of CD80+ macrophages (p < 0.005) in the absence (white) and in the presence (black) of MBG-75S. However, no significant differences were ob- served between control cells and MBG-75S treated macrophages in both basal and LPS/IFN-γ stimulated conditions. Fig. 13 shows CD80 expression of M1 RAW-264.7 macrophages observed by confocal mi- croscopy after 24h of treatment with E. coli lipopolysaccharide and interferon-γ as inflammatory stimuli. These results evidence that MBG-75S did not induce the macrophage polarization towards M1 pro-inflammatory phenotype, ensuring an appropriate immune response to this mesoporous bioac- tive glass. Fig. 10. Ca2+, phosphorous and silicon levels measured into the culture medium of osteoclast-like cells during the differentiation process after 3 and 7days of treatment with 1mg/ml of powdered MBG-75S. Controls without material were carried out in parallel (white). Statistical significance: *** p< 0.005. UN CO RR EC TE D PR OO F 10 Journal of Colloid and Interface Science xxx (2018) xxx-xxx Fig. 11. Effects of 1mg/ml of powdered MBG-75S on proliferation of RAW-264.7 macrophages after 24 h of treatment. Controls without material were carried out in par- allel (white). Statistical significance: *** p< 0.005. Fig. 12. Effects of 1mg/ml of powdered MBG-75S on pro-inflammatory M1 macrophage phenotype after 24 h of treatment without or with E. coli lipopolysaccha- ride and interferon-γ (LPS/IFN-γ) as inflammatory stimuli. Controls without material were carried out in parallel (white). Statistical significance: ффф p< 0.005 (comparison between basal and LPS/IFN-γ stimulated conditions). Bone remodeling is a very complex process that involves differ- ent cellular types to reach an equilibrium between bone formation, bone resorption and inflammatory response. In a scenario of degener- ative bone disease as osteoporosis, bone formation by osteoblasts is decreased respect to resorption driven by osteoclasts. For this reason, the osteoinductive effect of MBGs has been widely studied by differ- ent research groups, demonstrating significant increases in osteoblast proliferation [42–44] and a key role as differentiation stimuli from mesenchymal stem cells to osteoblast phenotype [45,46]. However, the effect of MBGs over other cell types as osteoclasts and macrophages have been poorly studied. Bone grafts intended for bone regeneration purposes in osteo- porotic patients not only should boost the osteoblastic pathway, but also to decrease the resorptive activity of osteoclast without inhibiting osteoclastogenic function. The inhibition of osteoclastogenesis could result in adynamic bone scenarios, similarly to those observed with the systemic administration of bisphosphonates, thus preventing bone re- generation. Besides, the bone graft should not produce a strong inflammatory response, without inhibiting the innate immune response mediated by macrophages. This response is mandatory to trigger the tissue healing process, but it must be controlled to avoid chronic inflammations that would impede the appropriated bone healing. The results observed in this study with MBG-75S with the three cell types (osteoblasts, osteoclasts and macrophages) point out very interesting responses to be considered as bone grafts, especially for os- teoporotic patients. For concentration of 1mg/ml, MBG-75S does not alter the osteoblasts cycle and their morphology. Besides, osteoclas- togenesis from macrophages was not hindered in the presence of this material and their plasma membrane were not altered, indicating that the osteoclast formation pathway is not blocked by MBG-75S. How- ever, the resorptive capability of these osteoclasts was significantly hampered, probably due to the high and fast silicon and calcium re- lease from the materials to the culture media. In this sense, the meso- porous structure of this material is likely to play a very important role. The activity of MBG-75S over osteoclasts opens the possibility for ob- taining bone grafts, which could reduce the osteoclast resorption with- out resulting in adynamic bone scenarios. Finally, MBG-75S evidences an excellent behavior respect to in- nate immune response, as this material allows the appropriated devel- opment of macrophages without polarization towards M1 pro-inflam- matory phenotype. 4. Conclusions The ions released from MBGs can stimulate the expression of several genes of osteoblastic cells [16] and could also regulate im- mune responses by altering the ionic microenvironment between the implants and hosts [18]. In the present study, a mesoporous bioac- tive glass with molar composition 75SiO2-20CaO-5P2O5 (MBG-75S) has been synthetized and its effects on osteoblasts, osteoclasts and macrophages have been evaluated jointly. This MBG exhibits a high mesoporous order and large surface area and porosity, thus allowing an efficient ionic exchange of Ca2+ and soluble silica with the sur Fig. 13. CD80 expression of M1 RAW-264.7 macrophages observed by confocal microscopy after 24h of treatment with E. coli lipopolysaccharide and interferon-γ as inflammatory stimuli. CD80 (red) was detected with PE conjugated anti-mouse CD80 antibody and nuclei were stained with DAPI (blue). Controls without antibody were carried out in parallel (left). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) UN CO RR EC TE D PR OO F Journal of Colloid and Interface Science xxx (2018) xxx-xxx 11 rounding media. MBG-75S shows in vitro biocompatibility respect to osteoblasts and concentrations up to 1mg/ml do not alter cell cy- cle of Saos-2 cells. MBG-75S does not inhibit osteoclastogenesis but decreases the resorptive activity of osteoclast cells. This fact in- dicates that MBG-75S would maintain bone remodeling but would slow down the bone resorption, thus favoring faster bone regenera- tion. MBG-75S allows macrophage proliferation without inducing po- larization towards M1 pro-inflammatory phenotype. This fact is in- dicative that MBG-75S would allow the innate immune response re- quired for healing process without further inflammatory complica- tions. Overall, the in vitro results obtained with osteoblasts, osteo- clasts and macrophages suggest that MBG-75S is an interesting candi- date as bone graft for bone regeneration purposes, especially in osteo- porotic patients. Further studies are currently being performed with in vivo models in order to determine the advantages of this MBG respect to other bioceramics; these results will be included in a future manu- script. Acknowledgements M.V.R. acknowledges funding from the European Research Coun- cil (Advanced Grant VERDI; ERC-2015-AdG Proposal No. 694160). N.G.C. is greatly indebted to Ministerio de Ciencia e Innovación for her predoctoral fellowship. L.C. acknowledges the financial sup- port from Comunidad de Madrid (Spain, CT4/17/CT5/17/PEJD-2016/ BMD-2749). The authors also thank to Spanish MINECO (MAT2015-64831-R, MAT2016-75611-R AEI/FEDER, UE). The au- thors wish to thank the ICTS Centro Nacional de Microscopia Electrónica (Spain), CAI X-ray Diffraction, CAI NMR, CAI Flow Cytometry and Fluorescence Microscopy of the Universidad Com- plutense de Madrid (Spain) for their technical assistance. References [1] X. Yan, C. Yu, X. Zhou, J. Tang, D.Y. Zhao, Highly ordered mesoporous bioac- tive glasses with superior in vitro bone-forming bioactivities, Angew. Chem. Int. Edit. 43 (2004) 5980–5984. [2] R. Li, A.E. Clark, L.L. Hench, An investigation of bioactive glass powders by sol-gel processing, J. Appl. Biomat. 2 (1991) 231–239. [3] X.X. Yan, X.H. Huang, C.Z. Yu, H.X. Deng, Y. Wang, Z.D. Zhang, S.Z. Qiao, C.Q. Lu, D.Y. Zhao, The in-vitro bioactivity of mesoporous bioactive glasses, Biomaterials 27 (2006) 3396–3403. [4] A. López-Noriega, D. Arcos, I. Izquierdo-Barba, Y. Sakamoto, O. Terasaki, M. Vallet-Regí, Ordered mesoporous bioactive glasses for bone tissue regeneration, M. Chem. Mater. 18 (2006) 3137–3144. [5] A. Garcia, M. Cicuendez, I. Izquierdo-Barba, D. Arcos, M. Vallet-Regí, Essen- tial role of calcium phosphate heterogeneities in 2d-hexagonal and 3d-cubic SiO2-CaO-P2O5 mesoporous bioactive glasses, Chem. Mater. 21 (2009) 5474–5484. [6] C.T. Wu, J. Chang, Multifunctional mesoporous bioactive glasses for effective delivery of therapeutic ions and drug/growth factors, J. Control. Rel. 193 (2014) 282–295. [7] C. Wu, R. Miron, A. Sculean, S. Kaskel, T. Doert, R. Schulze, Y. Zhang, Prolif- eration, differentiation and gene expression of osteoblasts in boron-containing associated with dexamethasone deliver from mesoporous bioactive glass scaf- folds, Biomaterials 32 (2011) 7068–7078. [8] C. Wu, Y. Zhou, C. Lin, J. Chang, Y. Xiao, Strontium-containing mesoporous bioactive glass scaffolds with improved osteogenic/cementogenic differentiation of periodontal ligament cells for periodontal tissue engineering, Acta Biomater. 8 (2012) 3805–3815. [9] S. Shruti, A.J. Salinas, E. Ferrari, G. Malavasi, G.A. Lusvardi, A.L. Doadrio, L. Menabue, M. Vallet-Regí, Curcumin release from cerium, gallium and zinc con- taining mesoporous bioactive glasses, Micropor. Mesopor. Mat. 180 (2013) 92–101. [10] A. Philippart, N. Gómez-Cerezo, D. Arcos, A.J. Salinas, E. Boccardi, M. Val- let-Regi, Aldo R. Boccaccini, Novel ion-doped mesoporous glasses for bone tis- sue engineering: Study of their structural characteristics influenced by the pres- ence of phosphorous oxide, J. Non-Cryst. Solids 455 (2017) 90–97. [11] L. Mao, L. Xia, J. Chang, J. Liu, L. Jiang, C. Wu, B. Fang, The synergistic ef- fects of Sr and Si bioactive ions on osteogenesis, osteoclastogenesis and angio- genesis for osteoporotic bone regeneration, Acta Biomater. 61 (2017) 217–232. [12] N. Gómez-Cerezo, I. Izquierdo-Barba, D. Arcos, M. Vallet-Regí, Tailoring the biological response of mesoporous bioactive materials, J. Mater. Chem. B 3 (2015) 3810–3819. [13] C. Wu, J. Chang, Mesoporous bioactive glasses: structure characteristics, drug/ growth factor delivery and bone regeneration application, Interface Focus 2 (2012) 292–306. [14] I. Izquierdo-Barba, M. Vallet-Regí, Mesoporous bioactive glasses: Relevance of their porous structure compared to that of classical bioglasses, Biomed. Glasses 1 (2015) 140–150. [15] I. Izquierdo-Barba, D. Arcos, Y. Sakamoto, O. Terasaki, A. López-Noriega, M. Vallet-Regí, High-performance mesoporous bioceramics mimicking bone miner- alization, Chem. Mater. 20 (2008) 3191–3198. [16] A. Hoppe, N.S. Güldal, A.R. Boccaccini, A review of the biological response to ionic dissolution products from bioactive glasses and glass-ceramics, Biomateri- als 32 (2011) 2757–2774. [17] A.A. Gorustovich, J.A. Roether, A.R. Boccaccini, Effect of bioactive glasses on angiogenesis: A review of in vitro and in vivo evidences, Tissue Eng. Part. B Rev 16 (2009) 199–207. [18] Y. Huang, C. Wu, X. Zhang, J. Chang, K. Dai, Regulation of immune response by bioactive ions released from silicate bioceramics for bone regeneration, Acta Biomater. 66 (2018) 81–92. [19] Z. Chen, T. Klein, R.Z. Murray, R. Crawford, Ji. Chang, C. Wu, Y. Xiao, Os- teoimmunomodulation for the development of advanced bone biomaterials, Mater. Today 19 (2016) 304–321. [20] T. Luhmann, O. Germershaus, J. Groll, L. Meinel, Bone targeting for the treat- ment of osteoporosis, J. Control Release 161 (2012) 198–213. [21] M.C. Matesanz, J. Linares, M. Oñaderra, M.J. Feito, F.J. Martínez-Vázquez, D. Arcos, S. Sánchez-Salcedo, M.T. Portolés, M. Vallet-Regí, Response of os- teoblasts and preosteoblast to calcium deficient and Si substituted hydroxyap- atites treated at different temperatures, Colloids Surf. B 133 (2015) 304–313. [22] M.C. Matesanz, J. Linares, I. Lilue, S. Sánchez-Salcedo, M.J. Feito, D. Arcos, M. Vallet-Regí, M.T. Portolés, Nanocrystalline silicon substituted hydroxyap- atite effects on osteoclast differentiation and resorptive activity, J. Mater. Chem. B 2 (2014) 2910–2919. [23] X. Feng, Regulatory roles and molecular signaling of TNF family members in osteoclasts, Gene 350 (2005) 1–13. [24] H.K. Väänänen, T. Laitala-Leinonen, Osteoclast linage and function, Arch Biochem. Biophys. 473 (2008) 132–138. [25] G.J. Atkins, D.R. Haynes, S.M. Geary, M. Loric, T.N. Crotti, D.M. Findlay, Co- ordinated cytokine expression by stromal and hematopoietic cells during human osteoclast formation, Bone 26 (2000) 653–661. [26] J.E. Aubin, Perspectives: Osteoclast adhesion and resorption: The role of po- dosomes, J. Bone Miner. Res. 7 (1992) 365–368. [27] G.R. Beck, S.-W. Ha, C.E. Camalier, M. Yamaguchi, Y. Li, J.-K. Lee, M.N. Weitzmann, Bioactive silica-based nanoparticles stimulate bone-forming os- teoblasts, suppress bone-resorbing osteoclasts, and enhance bone mineral density in vivo, Nanomedicine 8 (2012) 793–803. [28] D.M. Mosser, J.P. Edwards, Exploring the full spectrum of macrophage activa- tion, Nat. Rev. Immunol. 8 (2008) 958–969. [29] S.E. Burastero, Z. Magnani, C. Confetti, L. Abbruzzese, S. Oddera, P. Balbo, G.A. Rossi, E. Crimi, Increased expression of the CD80 accessory molecule by alveolar macrophages in asthmatic subjects and its functional involvement in al- lergen presentation to autologous TH2 lymphocytes, J. Allergy Clin. Immunol. 103 (1999) 1136–1142. [30] S.B. Rodan, Y. Imai, M.A. Thiede, G. Wesolowski, D. Thompson, Z. Bar-Shavit, S. Shull, K. Mann, G.A., Rodan, Characterization of a human os- teosarcoma cell line (Saos-2) with osteoblastic properties, Cancer Res. 47 (1987) 4961–4966. [31] M. Alcaide, M.C. Serrano, J. Román, M.V. Cabañas, J. Peña, E. Sánchez-Zapar- diel, M. Vallet-Regí, M.T. Portolés, Suppression of anoikis by collagen coating of interconnected macroporous nanometric carbonated hydroxyapatite/agarose scaffolds, J. Biomed. Mater. Res. A 95 (2010) 793–800. [32] J. Grossmann, K. Walther, M. Artinger, S. Kiessling, J. Scholmerich, Apoptotic signaling during initiation of detachment-induced apoptosis (anoikis) of primary human intestinal epithelial cells, Cell Growth Differ. 12 (2001) 147–155. [33] J.M. Udall, R.A. Moscicki, F.I. Preffer, P.D. Ariniello, E.A. Carter, A.K. Bhan, K.J. Bloch, Flow cytometry: a new approach to the isolation and characterization of Kupffer cells, Adv. Exp. Med. Biol. 216 (1987) 821–827. [34] J. Scheel, S. Weimans, A. Thiemann, E. Heisler, M. Hermann, Exposure of the murine RAW-264.7 macrophage cell line to hydroxyapatite dispersions of vari- ous composition and morphology: assessment of cytotoxicity, activation and stress response, Toxicol. In Vitro 23 (2009) 531–538. [35] Z. Li, K. Kong, W. Qi, Osteoclast and its roles in calcium metabolism and bone development and remodeling, abbreviation, Biochem. Biophys. Res. Commun. 343 (2006) 345–350. [36] S.Y. Hwang, J.W. Putney, Calcium signaling in osteoclasts, Biochim. Biophys. Acta 2011 (1813) 979–983. UN CO RR EC TE D PR OO F 12 Journal of Colloid and Interface Science xxx (2018) xxx-xxx [37] Z. Mladenovic, A. Johansson, B. Willman, K. Shahabi, E. Bjorn, M. Ransjo, Soluble silica inhibits osteoclast formation and bone resorption in vitro, Acta Biomater. 10 (2014) 406–418. [38] X. Zhang, D. Zeng, N. Li, J. Wen, X. Jiang, C. Liu, Y. Li, Functionalized meso- porous bioactive glass scaffolds for enhanced bone tissue regeneration, Sci. Rep. 6 (2016) 19361–19375. [39] S. Franz, S. Rammelt, D. Scharnweber, J.C. Simon, Immune responses to im- plants: A review of the implications for the design of immunomodulatory bioma- terials, Biomaterials 32 (2011) 6692–6709. [40] M.C. Matesanz, M.J. Feito, M. Oñaderra, C. Ramírez-Santillán, C. da Casa, D. Arcos, M. Vallet-Regí, J.M. Rojo, M.T. Portolés, Early in vitro response of macrophages and T lymphocytes to nanocrystalline hydroxyapatites, J. Colloid Interface Sci. 416 (2014) 59–66. [41] M. Alcaide, P. Portolés, A. López-Noriega, D. Arcos, M. Vallet-Regí, M.T. Por- tolés, Interaction of an ordered mesoporous bioactive glass with osteoblasts, fi- broblasts and lymphocytes, demonstrating its biocompatibility as a potential bone graft material, Acta Biomater. 6 (2010) 892–899. [42] Y. Zhu, M. Zhu, X. He, J. Zhang, C. Tao, Substitutions of strontium in meso- porous calcium silicate and their physicochemical and biological properties, Acta Biomater. 9 (2013) 6723–6731. [43] J. Lu, J. Wei, Q. Gan, X. Lu, J. Hou, W. Song, Y. Yan, J. Mac, H. Guo, T. Xiao, C. Liu, Preparation, bioactivity, degradability and primary cell responses to an ordered mesoporous magnesium–calcium silicate, Micropor. Mesopor Mater. 163 (2012) 221–228. [44] H.-S. Yun, J.-W. Park, S.-H. Kim, Y.-J. Kim, J.-H. Jang, Effect of the pore struc- ture of bioactive glass balls on biocompatibility in vitro and in vivo, Acta Bio- mater. 7 (2011) 2651–2660. [45] C. Wu, Y. Zhou, W. Fan, P. Han, J. Chang, J. Yuen, M. Zhang, Y. Xiao, Hy- poxia-mimicking mesoporous bioactive glass scaffolds with controllable cobalt ion release for bone tissue engineering, Biomaterials 33 (2012) 2076–2085. [46] C. Wu, Y. Zhou, M. Xu, P. Han, L. Chen, J. Chang, Y. Xiao, Copper-containing mesoporous bioactive glass scaffolds with multifunctional properties of angio- genesis capacity, osteostimulation and antibacterial activity, Biomaterials 34 (2013) 422–433.