Involvement of Mesenchymal Stem Cells in Oral Mucosal Bacterial Immunotherapy

Abstract

Recent clinical observations indicate that bacterial vaccines induce cross-protection against infections produced by different microorganisms. MV130, a polyvalent bacterial sublingual preparation designed to prevent recurrent respiratory infectious diseases, reduces the infection rate in patients with recurrent respiratory tract infections. On the other hand, mesenchymal stem cells (MSCs) are key cell components that contribute to the maintenance of tissue homeostasis and exert both immunostimulatory and immunosuppressive functions. Herein, we study the effects of MV130 in human MSC functionality as a potential mechanism that contributes to its clinical benefits. We provide evidence that during MV130 sublingual immunization of mice, resident oral mucosa MSCs can take up MV130 components and their numbers remain unchanged after vaccination, in contrast to granulocytes that are recruited from extramucosal tissues. MSCs treated in vitro with MV130 show an increased viability without affecting their differentiation potential. In the short-term, MSC treatment with MV130 induces higher leukocyte recruitment and T cell expansion. In contrast, once T-cell activation is initiated, MV130 stimulation induces an up-regulated expression of immunosuppressor factors in MSCs. Accordingly, MV130-primed MSCs reduce T lymphocyte proliferation, induce the differentiation of dendritic cells with immunosuppressive features and favor M2-like macrophage polarization, thus counterbalancing the immune response. In addition, MSCs trained with MV130 undergo functional changes, enhancing their immunomodulatory response to a secondary stimulus. Finally, we show that MSCs are able to uptake, process and retain a reservoir of the TLR ligands derived from MV130 digestion which can be subsequently transferred to dendritic cells, an additional feature that also may be associated to trained immunity.

Keywords: immunomodulation; mesenchymal stem cells; pattern recognition receptors; polybacterial preparation; short-term memory; sublingual mucosal immunotherapy; vaccine.

Conflict of interest statement

JS is the CEO of Inmunotek SL, a pharmaceutical company that manufactures bacterial vaccines. The authors declare that this study received funding from Inmunotek SL. The funder had the following involvement in the study: final writing and editing of the manuscript.

Copyright © 2020 Vázquez, Fernández-Sevilla, Jiménez, Pérez-Cabrera, Yañez, Subiza, Varas, Valencia and Vicente.

Figures

Scheme 1

Figure 1

Treatment with MV130 modifies biological properties of MSCs. MSCs were treated with MV130 for 2 h. (A, B) After 24 h, MSCs were stained with Annexin V and IP and cell viability was analysed by flow cytometry. Mean ± SEM of 5 independent experiments (A) and a representative experiment (B) are shown. (C) MSCs were treated with MV130 for 2 h. Bcl2 and Bcl-xL expression in MSCs were analyzed by flow cytometry 24 h after treatment. The percentages of positive cells are indicated in each histogram. Gray filled histograms represent isotype control staining. Data are representative of 4 independent experiments. (D) 48 h after treatment, the expression of different surface markers was studied on MSCs by flow cytometry. Representative histograms and MFI values are shown (n = 3-4). Gray histograms represent isotype controls. (E) mRNA expression for different immunomodulatory factors was studied on MSCs by qRT-PCR. Data represent mean ± SEM of 8 to 12 independent experiments relative to individual controls. (F) Supernatants from MSC cultures were collected 48 h after treatment. Protein secretion relative to individual controls is expressed as mean ± SEM from 15 independent experiments (*p < 0.05; **p < 0.01***p < 0.005 by Wilcoxon test).

Figure 2

MV130 is uptaken by oral MSCs in vivo. (A) Flow cytometry analysis of T lymphocytes (CD45+CD3+), macrophages (F4/80+MHC-IIlo), DCs (F4/80-MHC-II+) and MSCs (CD29+Sca-1+) present in the oral mucosa from mice after sublingual immunization with MV130-CFSE. Histograms show the percentage of uptake of CFSE-MV130 by each of these populations. Data are representative of 3 independent experiments. (B) Percentage of T cells, granulocytes and MSCs present in peripheral lymph nodes (P-LN), submaxillary lymph nodes (SM-LN) and oral mucosa (OM) from mice after sublingual immunization with MV130 respect to control mice (n = 6–13). (C) MV130 priming reduces MSC migration capacity. Bar graph shows the percentage of migrating cells in MV130 primed cultures. Results represent the mean ± SEM of 3 independent experiments (D) MV130 does not specifically attract MSCs. Bar graph shows the percentage of migrating cells to MV130 or IFNγ. Results represent the mean ± SEM of 4 independent experiments. (*p < 0.05; ***p < 0.005 by Wilcoxon test).

Figure 3

MSCs act as reservoirs of MV130 and are able to transfer it to DCs. MSCs were treated for 24 h with MV130 labeled with CFSE (CFSE-MV130; green) for monitoring. After washing cells to remove the drug, its uptake, processing and transference to DCs were studied. (A) Uptake and maintenance of CFSE-MV130. Histograms show the percentage of positive cells and MFI in brackets (n = 6–8 independent experiments). (B, C) Spatiotemporal monitoring of MV130 (green) in MSCs. Immunostaining in red for CD63 (B) or LAMP2 (C) in MV130-MSCs at different times. Inset indicated by arrows in B, scale bar: 10μm (24 h) or 1μm (120 h). Right: Higher magnification of the white square in (C) (120 h); examples of LAMP2-positive compartments with CFSE-MV130. Hoechst was used for nucleus staining (blue). Images are representative of 5 independent experiments. (D–G) DCs differentiated from monocytes were co-cultured with MSCs, previously treated with MV130 or CFSE-MV130 as described in Material & Methods section, and transfer of MV130 from MSCs to DCs was studied. A representative dot plot (D) and the mean ± SEM of five independent experiments (E) are shown. DCs and MSCs were gated according to CD1a or CD90 expression, respectively. In (E) control MSCs and DCs directly treated with CFSE-MV130 are also shown. (F) MV130 transfer from MSCs to DCs studied by immunofluorescence. Hoechst was used for nucleus staining in all cases (blue). Co-cultures were labeled with phalloidin (red) and anti-HLA-DR (magenta). The absence of HLA-DR expression on MSCs allow to distinguish it from DCs (HLA-DR+). White arrows in (F) indicate area of insert image magnification. (G) MSCs were labeled with anti-CD63 (red) and Hoechst was used for nucleus staining (blue). Co-localization of CD63+ extracellular microvesicles with CFSE-MV130 was observed free in the medium. White arrows indicate area of insert image magnification. Representative images of 3 independent experiments (*p < 0.05 significance by Wilcoxon test).

Figure 4

Immunomodulatory abilities of MSCs after activation with MV130. (A–D) Phenotype and function of monocyte-derived DCs differentiated in the presence or absence of CTRL-MSCs or MV130-MSCs. At day 6, CD1a and CD14 expression were analyzed by flow cytometry in the CD90- population. The percentage of positive cells is shown in each plot (A) and IL-6 production was measured in the supernatants (B). Results represent the mean ± SEM (n = 6). (C, D) DCs stimulated with LPS were cultured in MLR assays with CFSE-labeled T lymphocytes. After 5 days, the percentage of proliferating T cells was calculated by CFSE dilution method (gated on CD3+ cell population) (C) and supernatants from MLR co-cultures were analyzed for TNFα and IL-10 protein secretion (D). Data represent the mean ± SEM (n = 3–5). (E) Control or MV130-MSCs were co-cultured with CFSE-labeled T lymphocytes stimulated with CD3/CD28 beads, for different times. Histograms show CFSE staining in proliferating T cells in CD3+ gated cells. Proliferation index referred to unstimulated T lymphocytes (gray line) is indicated. Data are representative from four independent experiments. (F) MV130-MSCs re-stimulated with IFNγ, following protocol described in Material and Methods, were co-cultured with CFSE labeled T lymphocytes. Proliferation index is shown. Bar graph shows mean ± SEM (n = 4). (G–K) Control and MV130 primed MSCs were co-cultured with monocytes in the presence of GM-CSF to induce M1 macrophage differentiation. Monocytes alone were cultured as M1 control. (G–I) After 6 days, CD14, CD163 and PD-L2 expression was determined by flow cytometry in non-MSC population (CD90- cells) (n = 5–6). A representative experiment (G) and mean ± SEM of percentage of CD14+CD163+CD90- cells from five to six independent experiments (H) are shown. (I) Representative PD-L2 expression on macrophages. MFI is shown in each histogram. (J) After 6 days of co-culture, LPS was added and supernatants were analysed for TNFα and IL-10 production. Data represent TNFα/IL-10 ratio production at the different experimental conditions (mean ± SEM; n = 4). (K) Macrophages stimulated with LPS were used to carry out MLR cultures with CFSE-labeled T lymphocytes. After 5 days, the percentage of T cell proliferation was measured in the CD3+ cell population. (L) CCL2 and CXCL8 protein secretion measured in control and MV130-MSC culture supernatants. Bars represent the mean ± SEM relative to individual controls from 15 independent experiments. (M) PBMCs were placed in a transwell insert while MSCs, treated with or without MV130 for the 24 h previous, and seeded in the bottom chamber. After 8 h, migrating PBMCs (present in the lower chamber) were collected and stained for CD14, CD56, CD3, HLA-DR, and CD19, and different leukocyte populations were analyzed by flow cytometry. MSCs were excluded from the analysis by CD90 expression. (mean ± SEM; n = 4) (N) PBMCs migrating toward control or MV130 primed MSCs re-stimulated with IFNγ. Monocyte recruitment was analyzed by flow cytometry (mean ± SEM, n = 4). (O) Supernatants from MSC cultures following the protocol described in Material & Methods section were analyzed for CCL2, CXCL8, and CXCL10 protein secretion after IFNγ re-stimulation. Results represent mean ± SEM of four to six independent experiments relative to individual controls. (*p < 0.05; **p < 0.01, ***p < 0.005 significances relative to M1-macrophages or DC; #p < 0.05; ###p < 0.005 significances relative to CTRL-MSCs by Wilcoxon test).

Figure 5

Effects of MV130-MSCs in an in vivo model of acute inflammation. FVB/NJ mice were challenged in the footpad with 40μg of LPS and administered with or without control or MV130-primed MSCs 24 h later. (A) Footpad thickness increment was determined after 72 h as a measure of the efficacy of the different experimental groups of MSCs. Data shown are mean ± SEM of two independent experiments (three mice per group) (B) Images show histological sections of footpad tissue stained with Gallego’s Trichrome. Images are representative of 3 mice per group. (C) Percentage of CD45+ leukocytes infiltrating footpads in the different mouse groups analyzed by flow cytometry. The distribution of the different leukocyte subpopulations in CD45+ cells is also shown in each experimental group. Results represent increments relative to control animals (2 independent experiments with 3 mice per group) (*p < 0.05,**p < 0.01 versus LPS alone; #p < 0.05 versus CTRL-MSCs; by Wilcoxon test).

Figure 6

Proposed Model of MSC involvement in oral mucosal bacterial immunotherapy with MV130.