Human bone marrow MSCs from healthy donors (n = 8; Innoprot) were cultured in Mesenpro medium (Thermo Fisher Scientific) supplemented with glutamine and penicillin-streptomycin (Lonza) in a humidified incubator at 37°C with 5% CO₂ until reaching a confluence of 80%. Half of the culture medium was renewed every 3–4 days. MSCs were used between 3rd 7th passages.

MV130 (Bactek^®, Inmunotek S.L. Spain), is a preparation of whole-cell heat-inactivated bacterial species including 90% Gram-positive bacteria (Streptococcus pneumoniae, Staphylococcus aureus, Staphylococcus epidermidis) and 10% Gram-negative bacteria (Klebsiella pneumoniae, Moraxella catarrhalis, Haemophilus influenzae).

MSCs were treated with MV130 (10⁷ bacteria/mL; MV130-MSCs) for different time periods as indicated in each section. Where indicated, other concentrations of MV130 were used. After treatment, culture medium was removed and cells were washed twice with warm PBS to completely remove the unbound bacterial preparation. MSCs under control conditions were treated with the same volume of the excipient of MV130 (CTRL-MSCs).

To analyze if MSCs primed with MV130 during 24 h modify their response to a second challenge, MSCs were gently washed with warm PBS and cultured for another 3 days. Then, cells were re-stimulated with IFNγ (2 ng/mL; Immunotools) for another 24 h. Supernatants were collected and cells were used to carry out migration assays or to perform co-cultures with T cells (see scheme below).

Bacteria from MV130 were stained with CFSE (Biolegend) at 2.5 µM following the manufacturer’s instructions, to monitor their presence by immunofluorescence or flow cytometry.

MSCs were cultured at a cell density of 5,000 cells/cm² in 24-well culture plates for 6 days. MSCs were treated with MV130 (10⁷ bacteria/mL) or excipient and this treatment were repeated once again 24 h later. 72 h after the last dose of MV130, cells were collected, and expression of NANOG and Oct-4 was analyzed by quantitative PCR. In parallel, other cultures were switched to a specific adipogenic or osteogenic conditioning medium, as described previously (44). For osteogenic differentiation, alkaline phosphatase enzyme (ALP) levels were determined as a measure of MSC differentiation into osteoblasts, after 5 days of culture, while for adipogenic differentiation, Oil Red staining (ORO) was performed 10 days after differentiation conditions.

Peripheral blood mononuclear cells (PBMCs) were obtained by density gradient centrifugation using lymphocyte isolation solution (Rafer) from buffy coats of volunteer healthy donors (Centro de Transfusión de la Comunidad de Madrid, Spain). Monocytes were obtained from PBMCs by immunomagnetic isolation using anti-CD14 microbeads and VarioMACS cell separator (Miltenyi Biotec), following the manufacturer’s protocol. Non-adherent T lymphocyte-enriched cell suspensions were obtained from PBMCs by nylon wool enrichment and labeled with CFSE (Biolegend) at 2.5 µM following the manufacturer’s instructions. The percentage of CD3⁺ cells were always above 90%.

MSCs were seeded in 6-well plates at a concentration of 5 × 10⁵ cells/well in 2 mL. Following overnight adherence, MV130 treatment was added during 24 h, and then MSC cultures were gently washed to remove the unbound bacteria. Monocytes were added at 1:10 MSC/monocyte ratio in the different conditions described below.

Monocytes were cultured in RPMI 1640 medium (Lonza) supplemented with 10% fetal bovine serum (Gibco, Thermo Fisher Scientific), 100 U/mL penicillin, 100 µg/mL streptomycin, 2 mM L-glutamine and 1 mM pyruvate (all from Lonza), referred to as complete-RPMI medium, in the presence of GM-CSF (5 ng/mL; Gibco, Thermo Fisher Scientific) to induce M1 macrophages or with GM-CSF (20 ng/mL) and IL- 4 (20 ng/mL; Gibco, Thermo Fisher Scientific) to induce DC differentiation. After 3 days, additional 5 ng/mL GM-CSF was added to macrophage cultures and half of the medium was renewed in DC cultures. At day 6 of co-culture, macrophage and DC phenotypes were analyzed by flow cytometry within CD90^- population.

Macrophages and DCs were stimulated overnight with LPS (Invitrogen, Life Technologies) at 10 ng/mL or 50 ng/mL, respectively. Supernatants were collected and their allostimulatory function was analyzed by culturing in mixed lymphocyte reaction (MLR) with CFSE-labeled T lymphocytes (1:10 Macrophage or DC/T cell ratio). After 5 days of co-culture, T lymphocyte proliferation was analyzed using the CFSE dilution method by flow cytometry in the CD3⁺ population. Supernatants from different co-cultures were harvested at different times and cytokine secretion was measured.

2.5 × 10³ MSCs were seeded in duplicate and allowed to adhere to 48-well plates for 12 h and after that, primed with MV130 or under control conditions for 24 h. After treatment, MSCs were gently washed and CFSE-T lymphocytes were seeded at 1:25 MSC/T cell ratio in warm complete-RPMI with Dynabeads Human T-Activator CD3/CD28 for T Cell Expansion and Activation (1:4 Bead/T cell ratio; Gibco Thermo Fisher Scientific). Stimulated and non-stimulated T lymphocytes cultured without MSCs was carried out as control. After 3, 4 or 5 days of co-culture, supernatants were collected and proliferation of CFSE-T lymphocytes (CD3⁺ cells) and CD69 expression were analyzed by flow cytometry.

Mice were housed in the animal facility (Registration No. ES280790000183) at CIEMAT (Madrid, Spain). Mice were routinely screened for pathogens in accordance with FELASA procedures and received water and food ad libitum. All experimental procedures were carried out according to Spanish and European regulations (Spanish RD 53/2013 and Law 6/2013, European Directive 2010/63/UE). Procedures were approved by the CIEMAT Animal Experimentation Ethical Committee according to approved biosafety and bioethics guidelines.

Male mice C57 (3 weeks old) were sublingually treated with MV130 (10⁹ bacteria/mL) or excipient (control group) in two consecutive doses of 10 µL. This was repeated for 5 consecutive days with a booster two days later and 24 h before sacrifice (see scheme below). Sublingual administration was performed under anesthesia (mixture of 5% isoflurane in oxygen), to ensure proper delivery and prevent swallowing.

Peripheral lymph nodes (P-LN), submaxillary lymph nodes (SM-LN) and oral mucosa (OM) were excised and processed for flow cytometry. Lymph node cell suspensions were obtained by gentle mechanical disruption with a potter homogenizer until completely disaggregated. Oral mucosa cell suspensions were obtained by enzymatic digestion for 1 h with DNase (100 µg/mL), dispase (800 µg/mL), and collagenase (20 µg/mL) (Roche). Cells were stained with CD45, CD3, F4/80, CD29, MHC-II, and CD19 mAbs. Cells were gated based on forward/side scatter characteristics and their ability to exclude propidium iodide. Leukocyte populations were analyzed according to CD45⁺ expression and, CD3⁺ for T lymphocytes, F4/80⁺MHC-II^loCD19^- for macrophages and F4/80^-MHC-II⁺CD19^- for dendritic cells. MSCs were identified as CD45^-CD29⁺Sca-1⁺ cells. For tissue histological analysis, oral mucosa was carefully removed, embedded in OCT compound (Thermo Fisher Scientific) and stored at -80°C until processing. Sections of 10 μm were blocked with 5% normal donkey serum, following the staining with anti-mouse CD45 and Hoechst 33342 (Invitrogen, Life Technologies) for staining the nuclei. For immunofluorescence analysis, preparations were mounted using FluorSave (Millipore) and imaged using a fluorescence microscope (Nikon Eclipse Ci) with a digital camera (Nikon DS-U3) and Nis-Elements D software. Images were assembled using ImageJ software.

FVB/NJ mice, sedated with isoflurane, received a single injection of 40 μg of LPS in 30 µL of PBS into the right pad. At the same time, 30 μL of PBS were injected into the left pad as control. The baseline measurement was determined by measuring the pad thickness of each mouse with a digital caliper before LPS administration. 24 h after LPS injection, 5 × 10⁵ MSCs, treated as described above, were intravenously infused through the tail vein. To assess the efficacy of the different experimental groups of MSCs, the right pad thickness was measured 24, 48, and 72 h after LPS administration, comparing it to the control pad. At the end of the experiments mice were sacrificed by CO₂ inhalation. Footpads were extracted and processed for histological or flow cytometry analysis. Cell suspensions were obtained by gentle mechanical disruption with a potter homogenizer until completely disaggregated and were stained with CD45, CD3, F4/80 and Gr-1 mAbs to identify T-cells, inflammatory and classic macrophages, and granulocytes (45). For histological analysis, pad samples were fixed in buffered formalin and embedded in paraffin. Sections were stained with Gallego’s Trichrome.

The transfer of MV130-bacteria from MSCs to DCs, generated as described above, was studied by flow cytometry and immunofluorescence. 2 × 10⁴ MSCs were seeded in a 12-well plate and primed with MV130, CFSE-stained MV130 or none (control conditions). After 24 h for allowing the uptake of bacteria by MSCs, cultures were extensively washed and 2 × 10⁵ DCs (1:10 MSC/DC ratio) were seeded. MSCs and DCs were co-cultured for 24 h and the presence of CFSE-MV130 in both populations was analyzed by flow cytometry using CD90 and CD1a, as markers for MSCs and DCs, respectively. For the immunofluorescence study, cells were placed on a chamber slide in a 1:10 ratio (7 × 10³ MSCs/7 × 10⁴ DCs). After fixing, cells were stained with phalloidin, HLA-DR and Hoechst as described below.

Migration assays were performed in transwell inserts with 8 µm pore membrane (6.5 mm diameter; Costar). After migration, the upper and/or lower fractions were collected and suspended in the same volume. Quantification of cell number was performed by flow cytometry, acquiring all events gated according to forward/side scatter, for 180 s at a constant low flow rate. The percentage of migrated cells were calculated as follow:

To determine if MV130 priming modifies the migratory capability of MSCs, 10⁵ MSCs primed with MV130 (10⁶, 10⁷ o 10⁸ bacteria/mL) or excipient, were seeded in a transwell insert. Cell numbers in each fraction was quantified by flow cytometry after 20 h of culture, as described above.

To assess the possible chemotactic effect of MV130 on MSCs, 10⁵ MSCs were seeded in a transwell insert and placed on a well with culture media with MV130 (10⁷ bacteria/mL), glycerol or IFNγ (10 ng/mL). After 20 h of culture, the number of cells in the upper and lower fractions was quantified by flow cytometry and the percentage of migrating cells were calculated as described above.

To find out if MV130-MSCs have a chemoattractive effect on different leukocyte populations, 2 × 10⁵ MV130-MSCs or CTRL-MSCs were seeded in 24-well flat-bottom culture plates. 10⁶ PBMCs were added into the insert and cultured for 8 h. Migrating cells (present at the lower chamber) were collected and stained for CD14, CD3, CD19, CD56, HLA-DR, and CD90 for flow cytometry analyses. Transwell cultures without MSCs in the lower chamber were used as controls.

For cell viability studies, 2 × 10⁵ MSCs were cultured in a 12-well plate. The percentage of apoptotic (Anex⁺/IP^-) and necrotic (IP⁺) cells was analyzed by flow cytometry after 24 h of treatment, using annexin V conjugated with DY634 (Immunostep) and propidium iodide (Biolegend).

MSCs were lysed to perform RNA purification using Absolutely RNA Microprep kit (Agilent Technologies) following the manufacturer’s protocol. High capacity cDNA Reverse transcription kit (Applied Biosystems. Thermo Fisher Scientific) was used for the synthesis of the cDNA following the manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was performed using specific predesigned TaqMan Gene expression assays for different genes (Applied Biosystems) (Supplemental Table S1). All PCR reactions were set in duplicates using the TaqMan Gene Expression Master Mix (Applied Biosystems). The amplification and detection were performed using a 7.900HT Fast Real-time PCR System (Centro de Genómica, Complutense University of Madrid). ΔCT method was employed using GNB2L1 as reference gene to normalize gene expression.

Production of different cytokines was measured in supernatants from MSC cultures and MSC-Monocyte or MSC-T lymphocyte co-cultures. Levels of TNFα, IL-10 (Biolegend), IFNγ and PGE2 (R&D) were determined by ELISA and levels of IL-6, CXCL8, CCL2, CXCL10 and VEGF-A was determined by Cytometric Bead Array (CBA, BD Bioscience). TGF-β1 production was determined by LegendPlex (Biolegend) according to manufacturer’s instructions.

7 × 10³ MSCs were grown and treated with MV130 on chamber slides. After fixation with 4% paraformaldehyde and permeabilized with 0.05% saponin, nonspecific epitopes were blocked with PBS containing 10% donkey serum. Then, cells were sequentially incubated with the primary antibody for 45 minutes at room temperature: Texas Red-conjugated-Phalloidin (Thermo Fisher Scientific); anti-CD63 (46); anti-LAMP2 (Developmental Studies Hybridoma Bank at the University of Iowa); anti-HLA-DR (BD Biosciences); anti-paxillin (Sigma Aldrich) as appropriate. Next, cells were incubated for another 45 minutes with the appropriate secondary antibody: Alexa Fluor 594 conjugated donkey anti-mouse IgG; Alexa Fluor 488 conjugated donkey anti-rabbit IgG (both from Invitrogen, Life Technologies); DyLight 405 conjugated goat anti-mouse IgG (Jackson ImmunoResearch). Finally, a counter-staining of the nuclei was performed with Hoechst 33342 (Invitrogen, Life Technologies) for 10 minutes. After staining, preparations were mounted using FluorSave (Millipore) and imaged using a fluorescence microscope (Nikon Eclipse Ci) with a digital camera (Nikon DS-U3) and Nis-Elements D software. Images were assembled using ImageJ software.

Before staining with specific antibodies, cells were incubated at 4°C for 5 min with FcR Blocking Reagent (Milteny Biotec) to block nonspecific binding. Then, cells were stained with specific monoclonal antibodies (Supplemental Table S2) conjugated with different fluorochromes (Alexa Fluor 488, FITC, PE, PerCP, PE-Cy5, Alexa Fluor 647 or APC).

For the intracellular detection of Bcl-2, Bcl-x_L and Bax proteins, cells were treated with a FACS permeabilizing solution according to the manufacturer’s instructions (BD Biosciences), and stained with anti-human Bcl-2 or anti-human Bcl-x_L Abs (Supplemental Table S2) for 30 min. Analysis was performed on a FACSCalibur flow cytometer (BD Biosciences) (Centro de Citometría y Microscopía de Fluorescencia. Complutense University of Madrid) and analyzed with FCS Express V3 software.

All data are expressed as mean ± SEM of the indicated parameter. Data analysis was performed using GraphPad Prism version 8.0.2. Statistical significance was determined by Wilcoxon test. Values of *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.005 were considered to be statistically significant.

We first analyzed the effect of MV130 on MSC survival. MV130 significantly increased the viability of MSCs (Figures 1A, B and Supplemental Figure S1A), increasing significantly Bcl-2 and Bcl-X_L antiapoptotic protein expression (Figure 1C). Supplemental Figure S1B shows that the expression of NANOG and Oct-4 transcription factors, known to be involved in pluripotency and self-renewal of undifferentiated stem cells, were not affected. In addition, no significant differences on the adipogenic and osteogenic differentiation potential were seen when control and MV130-primed MSC cultures were compared (Supplemental Figures S1C, D). Thus, while MV130 favors MSC survival, no changes in their stemness and multipotent developmental properties were observed.

MV130 priming of MSCs did not change their expression of TLR1, TLR2, TLR3, TLR4, TLR5, TLR6, and TLR9 (Supplemental Figure S1E), CD86, CD40, or HLA-DR (Supplemental Figure S1F) under any of the assayed conditions. CD73, an ectoenzyme with powerful anti-inflammatory properties, was maintained highly expressed in MV130-treated MSCs (Supplemental Figure S1F). Relevantly, upon MV130 priming, MSCs significantly upregulated the expression of others immunosuppressive molecules such as PD-L1 and PD-L2, as well as the adhesion protein ICAM-1, key player in MSC-mediated functions (Figure 1D). Interestingly, MV130-primed MSCs exhibited enhanced mRNA expression of IL6, TGF 1, VEGFA and COX2 (Figure 1E), confirmed at protein level for IL-6, TGF-β1, and for the major COX2 product, PGE2, as detected in MSC culture supernatants (Figure 1F).

To determine whether resident MSCs present in the oral mucosa could be responding to MV130 treatment, mice were administered sublingually with MV130-bacteria labeled with CFSE. Flow cytometry analysis of cell suspensions obtained from the excised oral mucosa indicated that MSCs could uptake MV130-bacteria efficiently in vivo. As shown in a representative experiment (Figure 2A), the percentage of CD45^- CD29⁺ Sca1⁺ MSCs uptaking MV130 (41%) was significantly higher to that seen (28%) in MHC-II⁺/F4/80^-CD19^- cells (mostly DCs) but lower than in MHC-II^lo/F4/80⁺ macrophages (72%). The relative numbers of MSCs and T cells within the oral mucosa remained unchanged upon MV130 treatment, in contrast to the significant recruitment of granulocytes from extramucosal tissues (Figure 2B and Supplemental Figure S2A).

As no active recruitment of mucosal MSCs was seen in the in vivo mouse model after MV130 treatment, we were prompted to explore the migration features of human MSCs primed in vitro with MV130. As shown in Figure 2C, the migration ability of MV130-primed MSCs was significantly decreased relative to control MSCs. Also, conventional migration assays showed that MV130 did not represent a chemotactic stimulus for MSCs, unlike IFNγ used as positive control (Figure 2D). In consonance with the migration assays, the immunofluorescence study revealed that MV130-primed MSCs showed characteristics of slow-moving cells with sparse and large streak-like focal adhesion complexes at the periphery and a low polymerization degree of F-actin, whereas control MSCs showed small dot-like nascent focal adhesion sites and polymerized actin at the ruffles (Supplemental Figure 2B). Thus, the results indicate that MSCs may uptake MV30-bacteria while decreasing their migratory activity.

Next, we studied the fate of MV130-bacteria uptaken by MSCs. To this end, cells were treated 24 h with CFSE-labeled MV130 and then extensively washed to remove extracellular bacteria. Bacteria-containing MSCs were monitored by flow cytometry and studied by immunofluorescence at different times. After treatment, about 60-85% of MSCs showed fluorescent staining. Although this percentage gradually decreased in the following days, around 50% of the MSCs still showed significant levels of fluorescence 5 days later (Figure 3A). As shown under microscopy (Figure 3B), MV130-bacteria accumulated in the vesicular compartment near the plasma membrane at the beginning, suggesting their internalization into early endosomes. Later (120 h) they appeared translocated to the late endosomal/lysosomal compartments as indicated by the co-localization with CD63 and to a lesser extent with LAMP2 markers (Figures 3B, C). These results indicate that MSCs are able to internalize, process and maintain the MV130-bacteria over time.

As it has been shown that MSCs may retain TLR2 ligands and transfer them to immune cells (47), we were prompted if this might be also the case with MV130-bacteria. Thus, MSCs were treated with CFSE-labeled MV130, extensively washed to remove unbound bacteria and co-cultured with monocyte-derived DCs. After 24 h, a notable proportion (around 25%) of DCs co-cultured with MSCs showed significant labelling with CFSE-conjugated MV130, compared to only 2% of background staining (Figures 3D, E). Immunofluorescence studies of MSC-DC co-cultures showed that CFSE-labeled bacteria could be transferred from MSCs to neighboring DCs through transient cytoplasmic extensions (Figures 3F). In addition, CFSE-MV130 conjugates from MSCs could also reach DCs through extracellular vesicles since in the co-cultures it was possible to observe CD63⁺ vesicles loaded with CFSE-conjugated MV130 (Figure 3G). Interestingly, in control experiments using only DCs, a direct uptake of CFSE-MV130 by these cells was around 35%, whereas it reached to around 70% when only MSCs were used (Figure 3E), which underlined the high ability of MSCs to uptake MV130-bacteria.

Both control and MV130-primed MSCs similarly dampened the generation of CD14^– CD1a⁺ DCs from CD14⁺ CD1a^– monocytes (Figure 4A). DCs induced in the presence of MV130-primed MSCs showed a slight but constant increase in the expression of HLA-DR, PD-L1 and PD-L2, while decreasing CD86 co-stimulatory molecule expression (Supplemental Figure S3A) and inducing a significant increase of IL-6 production (Figure 4B). These DCs showed a similarly reduced allostimulation ability in co-cultures with T cells (Figures 4C, D).

To assess the effects of the priming of MSCs with MV130 on the adaptive immune response, the ability of MSCs to modulate TCR-triggered activation of T cells was also investigated by proliferation assays. As shown in Figure 4E, in the presence of MSCs, T cell proliferation was slightly increased on day 3 while significantly suppressed on day 5. These enhancing and suppressing effects were more evident when MV130-primed MSCs were used, reaching signification on day 5 (Figure 4E and Supplemental Figure S3B). On the other hand, to test whether this response was transient or could be maintained for a prolonged period of time, MSCs were primed with MV130 for 24 h, extensively washed to remove extracellular bacteria and then stimulated with IFNγ on day 4. Figure 4F shows that CD3/CD28-induced T-cell proliferation was inhibited by IFNγ-activated MSCs in a greater extent than non-activated MSCs; yet, such an inhibition was significantly higher using MSCs primed with MV130.

We also addressed whether MV130 priming could influence the effects of MSCs on the repolarization of monocytes toward M2-like macrophages. As shown in Figures 4G–I, monocytes undergoing differentiation to macrophages in the presence of MSCs showed a M2-like pattern. This effect was more notable in cultures with MV130-primed MSCs, where the proportion of CD14^high CD163^high PD-L2⁺ macrophages generated was significantly increased respect to control MSC cultures (Figures 4G–I). After 6 days of differentiation, macrophages generated in the different cultures were activated with LPS and the cytokine production was analyzed. As shown in Figure 4J, macrophages differentiated in the presence of MSCs showed a reduced pro-inflammatory TNFα/anti-inflammatory IL-10 cytokine production ratio when compared with control M1-like macrophages, and this reduction was more pronounced in macrophages generated in the presence of MV130-primed MSCs with significantly increased IL-10 levels (p<0.05). In addition, macrophages generated in the presence of MV130-MSCs induced a lower proliferative response of T cells than their control counterparts (Figure 4K). As recent studies have demonstrated that MSC-derived CCL2 is required for polarizing IL-10⁺ tissue macrophages (48–50), the effect of priming MSCs with MV130 on its production was studied. As shown in Figure 4L, CCL2 production was upregulated in MV130-primed MSCs at both mRNA and protein levels (Supplemental Figure S3C). This was also the case for CXCL8 and CXCL12 (Figures 4L and Supplemental Figure S3C). Since CCL2, CXCL12 and CXCL8 are the major chemokines driving monocyte extravasation, chemotaxis assays were performed with PBMCs in a transwell system. The results showed that monocytes were the main cell type migrating at higher numbers toward the compartment with MV130-primed MSCs, as compared with untreated MSCs (Figure 4M and Supplemental Figure S3D). In addition, priming of MSCs with MV130 also modified their capacity to recruit monocytes upon IFNγ re-stimulation (Figure 4N). Again, this effect was accompanied by an increase in CCL2, and also CXCL10, chemokine production (Figure 4O).

As both MSCs and M2-like macrophages are associated with wound healing and tissue repair, we were prompted to test whether MV130-primed MSCs may have a differential effect in an in vivo model of acute inflammation. To this end, mice were injected in the footpad with LPS alone or with control or MV130-primed MSCs 24 h later. Both MSCs significantly reduced the LPS-inflammatory response, measured by footpad thickness, without differences between MV130-primed and control MSCs (Figure 5A). However, the histological analysis showed a clear reduction of leukocyte infiltration in those mice co-injected with MSCs primed with MV130 (Figure 5B). Flow cytometry analysis of the excised tissues indicated that leukocyte infiltration was markedly decreased by the treatment with MV130-primed MSCs, in comparison to that using control MSCs (Figure 5C). As shown, granulocytes and inflammatory macrophages were reduced by 50–60% and 40–50%, respectively (Figure 5C).