TYPE Original Research PUBLISHED 14 September 2022 DOI 10.3389/fimmu.2022.957797 Mesenchymal stromal cells OPEN ACCESS induced regulatory B cells are EDITED BY Paolo Casali, The University of Texas Health Science enriched in extracellular Center at San Antonio, United States REVIEWED BY matrix genes and IL-10 Raghavan Chinnadurai, Mercer University, United States Xiaoyong Chen, independent modulators Sun Yat-sen University, China *CORRESPONDENCE Sergio G. Garcia 1,2†, Noelia Sandoval-Hellı́n 1†, Marcella Franquesa mfranquesa@igtp.cat Marta Clos-Sansalvador 1,2, Laura Carreras-Planella 1, Francesc E. Borràs Miriam Morón-Font 1, Dolores Guerrero 3, Francesc E. Borràs 1* feborras@igtp.cat † These authors have contributed and Marcella Franquesa 1* equally to this work and share 1 REMAR-IGTP Group, Germans Trias i Pujol Research Institute (IGTP) & Nephrology Department, first authorship University Hospital Germans Trias i Pujol (HUGTiP), Can Ruti Campus, Badalona (Barcelona), Catalonia, Spain, 2 Department of Cell Biology, Physiology and Immunology, Autonomous University SPECIALTY SECTION of Barcelona, Bellaterra, Spain, 3 Otorhinolaryngology Department, Hospital Universitari Germans This article was submitted to Trias i Pujol, Badalona, Spain B Cell Biology, a section of the journal Frontiers in Immunology RECEIVED 31 May 2022 ACCEPTED 24 August 2022 Regulatory B cells (Breg) are essential players in tolerance and immune PUBLISHED 14 September 2022 homeostasis. However, lack of specific Breg markers limit their potential in CITATION clinical settings. Mesenchymal stromal cells (MSC) modulate B cell responses Garcia SG, Sandoval-Hellı´n N, and are described to induce Breg in vitro. The aim of this work was to Clos-Sansalvador M, Carreras-Planella L, Morón-Font M, characterize MSC induced Breg (iBreg) and identify specific Breg biomarkers Guerrero D, Borràs FE and by RNAseq. After 7-day coculture with adipose tissue-derived MSC, B cells Franquesa M (2022) Mesenchymal stromal cells induced regulatory B were enriched in transitional B cell populations, with increased expression and cells are enriched in extracellular secretion of IL-10 and no TNFa. In addition, iBreg showed potential to matrix genes and IL-10 modulate T cell proliferation at 2 to 1 cell ratios and their phenotype independent modulators. Front. Immunol. 13:957797. remained stable for 72h. RNAseq analysis of sorted IL-10 positive and doi: 10.3389/fimmu.2022.957797 negative iBreg populations identified over 1500 differentially expressed genes COPYRIGHT (DEG) among both populations. Analysis of biological processes of DEG © 2022 Garcia, Sandoval-Hellı´n, highlighted an enrichment of immune regulation and extracellular matrix Clos-Sansalvador, Carreras-Planella, Morón-Font, Guerrero, Borràs and genes in IL-10- iBreg populations, while IL-10+ iBreg DEG were mostly Franquesa. This is an open-access associated with cell activation. This was supported by T cells modulation article distributed under the terms of assays performed in the presence of anti-IL-10 neutralizing antibodies the Creative Commons Attribution License (CC BY). The use, distribution showing the non-essential role of IL-10 in the immunomodulatory capacity or reproduction in other forums is of iBregs on T cells. However, based on RNAseq results we explored the role of permitted, provided the original author(s) and the copyright owner(s) TGF-b and found out that it plays a major role on iBreg induction and iBreg are credited and that the original immunomodulatory properties. Therefore, we report that MSC induce B cell publication in this journal is cited, in populations characterized by the generation of extracellular matrix and accordance with accepted academic practice. No use, distribution or immune modulation independently of IL-10. reproduction is permitted which does not comply with these terms. KEYWORDS B cell, immunomodulation, mesenchymal stromal cell, regulatory B cell, IL-10 Frontiers in Immunology 01 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 Introduction the role of IL-10 as a key molecule of Breg’s immunomodulatory capacity. The results of this study show that MSC induce B lymphocytes are the highlight of the adaptive humoral regulatory B cells in vitro that were able to suppress T cell response mainly by their production of antigen-specific proliferation, independently of IL-10 action. Characterization by immunoglobulins (Ig) or antibodies (1, 2). Yet, B lymphocytes transcriptomic analysis of Breg IL-10 positive and negative have essential roles besides antibody production, including antigen populations showed an enrichment of immunomodulation- presentation and co-stimulation, and the production of secretory and extracellular matrix organization-related genes in IL-10- mediators (i.e. cytokines), that modulate cellular responses (3, 4). cells. This set of genes could become potential Breg biomarkers, A specific subset of B lymphocytes, known as regulatory B cells and highlight the importance of IL-10- Breg populations for (Breg), can modulate the immune response by suppressing T cell future Breg and MSC studies. activation and proliferation, inducing differentiation of regulatory T cells, and producing immunomodulatory factors such as IL-10 (5). Breg are essential in the maintenance of immune homeostasis and Materials and methods immunological tolerance, but despite extensive discussion and research done in recent years (5–8), several drawbacks including Mesenchymal stem cell collection the lack of unique defining markers, and the heterogeneity across and culture experimental settings and animal models are undermining their therapeutic potential. Regarding the first, the general consensus on MSC from subcutaneous adipose tissue were obtained from human studies is that IL-10 identifies Bregs in vivo and is crucial to healthy donors at “Hospital Germans Trias i Pujol”, MSC exert their immunomodulatory role. On the second, the term Breg multilineage differentiation potential and surface markers were may include several B cell subsets such as transitional B cell (8–10), tested (Supplementary Figure 1) upon culture to meet the GMZB+ B cells (11–13), or CD1d+ B cells (14–16), most of them minimal criteria for MSC identification (22) as described in [ (19, with immunoregulatory effect. In deep contrast to their T regulatory 24)]. Informed consent was obtained from all subjects, and the counterparts (17), no transcription factors nor surface markers have study protocol conformed to the principles outlined in the been unequivocally associated to human Breg. Declaration of Helsinki. Cells were frozen at -190°C until further Characterization of Breg is a priority for their future clinical use. To perform the experiments cells were cultured in 175 cm2 application. Breg are limitedly represented in peripheral flasks at 37°C, 5% CO2, and 95% humidity, in 20 mL of minimum circulation. Thus, Breg characterization is mostly performed essential medium-a (Lonza, Verviers, Belgium) supplemented with by culturing B cells in vitro using a plethora of B cell 1% penicillin (100 IU/ml, Cepa S.L., Madrid, Spain) and activating molecules, such as anti-BCR antibodies, CD40L, IL- streptomycin (100 mg/ml, Normon Laboratories S.A., Madrid, 21, CpG, or LPS, among others. In recent years, mesenchymal Spain), 1% L-glutamine (Sigma Aldrich, St. Louis, MO, USA), stromal/stem cells (MSC) have been also described to induce and 10% heat-inactivated fetal bovine serum (Lonza). Medium Breg in vitro (18–21). was renewed every 2-4 days and flasks were examined by optical MSC are immunomodulatory non-hematopoietic cells that microscopy and when 90% confluence was reached, cells were can be isolated from several tissues, that hold stem cell properties detached with trypsin-EDTA (Lonza) at a concentration of 0.05% and can be phenotypically defined by a set of minimal criteria during 5 minutes at 37°C. Passages 2 to 8 were used for (22, 23). MSC are widely studied for their broad clinical potential experiments. To asses surface marker expression, MSC were in immune modulation, wound healing, and tissue repair. In labelled for flow cytometry analysis with the following antibodies: recent years, MSC have been described to interact with several CD90-PE-Cy7 (BioLegend, clone 5E10, San Diego, CA, USA), immune cell types, including B cells, reducing their terminal CD105-FITC (ImmunoTools, clone MEM-226SJ25C1, Friesoythe, differentiation, inhibiting antibody production, and promoting Germany), CD73-PE (BD Pharmingen, clone AD2), CD45-PerCP their differentiation to Breg (18–21). Taking into account their (ImmunoTools, clone HI30), HLD-DR-APC-H7 (BD Biosciences, role in B cell modulation, MSC constitute an optimal candidate clone L243). to induce and expand Breg in vitro for downstream application. In the present study, we use a multicomponent coculture system to expand and study in vitro MSC-induced Breg (iBreg). B cell isolation from human tonsils First, we characterized iBreg according to surface markers, IL- 10/TNFa expression, and immunomodulatory capabilities. To isolate B cells, tonsils were obtained from patients that Secondly, to search for new Breg cell markers and increase underwent tonsillectomy at the Hospital Universitari Germans knowledge on Breg physiology, we purified IL-10+ and IL-10- Trias i Pujol (HUGTiP) to be anonymously used for research iBreg to perform transcriptomic characterization by RNA-seq purposes. All participants and/or their legal tutor gave written and identify differentially expressed genes. Finally, we assessed informed consent to participate. The study protocol followed the Frontiers in Immunology 02 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 principles of the Declaration of Helsinki and was approved by To quantify IL-10 and TNFa intracellular expression in B the Clinical Research Ethics Committee of our institution cells, we performed intracellular staining of both cytokines in (Comitè Ètic d’Investigació Clı́nica, Refs CEIC: PI-16-056 independent tubes, labelling each B cell tube with IL-10-PE (BD from 5 May 2016). Tonsils were mechanically disrupted and Pharmingen, Clone JES3-19F1) or TNFa-PE (BD Pharmingen, filtered through a 70 mm cell strainer, to then isolate Clone MAb11) using IntraStain kit (Dako Denmark). Monensin mononuclear cells by density gradient separation with Ficoll- (5 mM, BD Biosciences) 1/6000 solution was added twelve hours Paque (GE Healthcare, Uppsala, Sweden). Mononuclear cells before performing the intracellular staining procedure to block were frozen in 90% heat-inactivated fetal bovine serum (Lonza) cell protein secretion. In some cases, for PMA and ionomycin and 10% Dimethyl sulfoxide (DMSO) and stored at -190°C until samples, 4h before intracellular staining samples were stimulated further use. Upon thawing, resting B cells were isolated by with 50 ng/mL of PMA (Sigma) and 500 ng/mL of Ionomycin negative selection by labelling anti-CD43 magnetic beads (Merk, Sigma-Aldrich) and Monensin 1/6000 dilution was (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). Purity added at the same time to the samples. Blank samples without was confirmed by Flow cytometry (FACS Canto II, BD fluorochrome-conjugated antibodies were prepared as a control Bioscience, San Jose CA), labelling cells with the following for cytometry. Intracellular staining was performed following antibodies; 7AAD (BD Biosciences), CD3-PE (BD Biosciences, manufacturer’s instructions. Clone UCHT1) and CD19-BV510 (BD Horizon, clone SJ25C1), To study iBreg phenotypical stability, iBreg were collected at ensuring CD19 purities above 95%. day 7 from MSC-B cell cocultures and cultured for an additional 72h period in 96w round bottom culture plates in the presence of different B cell stimulation molecules: TCL stimuli as defined in MSC and B cell coculture the “MSC and B cell coculture” section, 10 µg/mL LPS (E. coli, O111:B4, Sigma-Aldrich) or 10 µg/mL CpG (oligodeoxynucleotide- For the generation of iBreg, B cells and MSC were co-cultured 2006, Invivogen). B cells were analyzed by flow cytometry using the in 96-well flat bottom plates for 7 days. 50000 B cells and 5000 MSC previously defined panel at 24h, 48h and 72h. were seeded per well. B cells were cultured in 200 mL of IMDM culture medium (Lonza) supplemented with 1% penicillin (100 IU/ ml, Cepa S.L., Madrid, Spain) + streptomycin (100 mg/ml, Normon T cell proliferation assays Laboratories S.A. Madrid, Spain), 1% L-glutamine (Sigma- Aldrich), and 10% heat-inactivated fetal bovine serum (Lonza). To test iBreg immunomodulatory capacity, purified CD43- Cells were stimulated with a combination of monoclonal antibodies B cells were co-cultured with MSCs. After 7 days of co-culture that mimic T helper cell action (T-cell like: TCL); 10 mg/mL of F(ab) their regulatory potential was tested on T cell proliferation. T 3 2 anti-IgM (Jackson, ImmunoResearch laboratories) 10 IU of IL-2 cells were isolated from tonsillar PBMCs to perform autologous (Peprotech, London, UK) and 2.5 ug/mL of CD40 agonist (Bioxcell, assays by magnetic separation using HLA kit according to Clone G28.5). For the generation of activated B cells, 50000 cells manufacturer’s protocol for lymph nodes samples. After were cultured in 96-well round bottom plates with no MSC. A isolation, purity was confirmed by flow cytometry (FACS minimum of 2 MSC donors and 3 Tonsil donors were used for Canto II, BD Bioscience, San Jose CA), labelling cells with the each experiment. following antibodies; 7AAD (BD Pharmingen), CD3-PE (BD Pharmingen, Clone UCHT1) and CD19-BV510 (BD Horizon, clone SJ25C1). Purities were above 85% in all samples. iBreg characterization T cells were labelled with CFSE (5 mM) 1/6000 solution in PBS for 10 min at RT and stimulated with aCD2/CD3/CD28 To asses surface marker expression, iBreg cultures were labelled beads (Miltenyi Biotec, T Cell Activation/Expansion Kit, for flow cytometry analysis with the following antibodies: IgD- human). 50000 T cells were cultured in 96-well round bottom APC-Cy7 (BioLegend, clone IA6-2, San Diego, CA, USA), CD19- plates in 200 mL of IMDM culture medium (Lonza) BV510 (BD Horizon, clone SJ25C1), CD24-APC (Invitrogen, clone supplemented with 1% penicillin (100 IU/ml, Cepa S.L., eBioSN3), CD27-PE-Cy7 (Invitrogen, clone 0323), CD38-PE Madrid, Spain) + streptomycin (100 mg/ml, Normon (Invitrogen, clone HB7) and 7AAD (BD Pharmingen). Laboratories S.A.,Madrid, Spain), 1% L-glutamine (Sigma- To asses cytokine production, supernatants were collected and Aldrich), and 10% heat-inactivated fetal bovine serum (Lonza) kept at -20°C for further quantification of IL-10 and TNFa in two with the following ratios of T to iBreg cells: 1:2, 1:1, 2:1, 5:1, 10:1, independent ELISA tests, both according to manufacturer’s and 50:1. Proper controls were added for the experiments, no instructions (U-Cytech, Utrecht, The Netherlands). Standards CFSE labelling, no proliferation stimuli, and no iBreg. 4 days were provided by the kit and a standard curve was prepared after, proliferation was assessed by Flow cytometry. Samples from 200 to 3 pg/mL. Samples were measured at 450 nm using an were labelled with CD19-BV510 (BD Horizon, clone SJ25C1) to absorbance plate reader (Varioskan). exclude B cells from the analysis. For experiments assessing IL- Frontiers in Immunology 03 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 10 role in T cell modulation, we used anti-IL-10 neutralizing fluorometry (Quantus, Promega) and qPCR (KAPA library antibodies (BD Biosciences) at concentrations of 10 and 20 mg/ quantification kit, Light Cycler 480). Final quantification of each mL. T o inhibit TGFBR1, t he inhibitor Ly-364947 library was obtained from the quantifications derived from (MedChemExpress, Sweden) was added to wells at 2 mM and fluorometry and qPCR corrected by the fragment size observed in 5 mM as described in the literature (25, 26). IL-10 neutralization capillary electrophoresis, and used to prepare an equimolar mixture capacity was confirmed by ELISA (U-Cytech). Recombinant IL- of all libraries. Libraries were denatured with sodic hydroxide 10 (Peprotech) was measured at 50 and 100 pg/mL. For reaching a 20 pM concentration library pool. Samples were loaded neutralized samples, 10 or 20 mg/mL of anti-IL10 antibody into an Illumina HighSeq sequencer, obtaining 20-25M reads per was added to IL-10 samples and incubated for 30 min at RT sample. Raw RNA-seq data quality was check by FASTQC. Paired- before adding samples to the ELISA wells (Supplementary Figure 2). end reads for each sample were quantified via alignment to version 38 of the Ensembl annotation of the human genome using HISAT2 (27, 28). The DESeq2 package (29) was used to normalize and Il-10 Capture assay and cell sorting analyze differential gene expression between groups, IL-10+ and IL- 10- cells. Batch effect was corrected during differential gene To isolate IL-10 positive and negative B cells, we performed expression analysis, considering biological variability from MSC flow cytometry cell sorting (BD Aria II) by labelling iBreg samples and B cell donor. Genes with a Benjamini–Hochberg false- with a modified version of the IL-10 capture assay (Miltenyi Biotec discovery rate (FDR) inferior to 0.01 and a foldchange (FC) GmbH, Bergisch Gladbach, Germany). 4 hours prior to collecting B superior to 2 were considered significantly differentially expressed. cells, 0.01 mL of PMA (Sigma) and 0.2 mL of Ionomycin (Merk, Gene ontology (GO), chord diagrams and volcano plots were Sigma-Aldrich) were added to the stimulated samples to boost generated in R using clusterProfiler (30, 31), circlize (32) and cytokine production on B cells. In brief, after collecting iBreg, cells ggplot2 (33), respectively. Only GO categories with an FDR under were washed with cold commercial MACS buffer (Miltenyi Biotec 10% were represented. Heatmaps were generated in R using GmbH, Bergisch Gladbach, Germany) and resuspended in 80 mL of Heatmap3 package (34). String (35) software was used to analyze cold IMDM complete medium. Then, B cells were labelled with 20 gene interaction networks based on their protein interactions, with a mL of IL-10 catch reagent per 10 million cells, with an incubation of special interest for functional clusters of biological processes. 5 min on ice. Next, we added 10 mL of warm medium and incubated at 37°C for 45 min under constant agitation to promote IL-10 secretion. Cells were washed with cold MACS qPCR buffer and resuspended in 80 mL cold MACS buffer. Next, cells were labelled with 20 mL of IL-10-PE (Biologend, clone JES3-9D7) To analyze specific gene expression, whole RNA content was detection antibody per 10 million cells, incubating for 10 min on ice. isolated from cells using the RNAeasy Mini kit (Qiagen). cDNA was At this point, instead of following manufacturer’s instructions to obtained from RNA samples using random hexamers and the One- perform magnetic separation, we performed Flow cytometry step RT-PCR kit (BioRad laboratories) according to the associated cell sorting. We sorted cells in two groups according to manufacturer’s protocol using Applied biosystems 2729 Thermal high-expressing IL-10+ and IL-10- iBreg. Cycler. Analysis of gene expression was performed using the primer sequences indicated in Supplementary Table 3. qPCR reaction was performed and analyzed on the LightCycler 480 II-Real-Time PCR RNAseq and data analysis System (Roche) using “Master Mix PowerUp SYBR green” kit (Thermo Fisher Scientific). Each sample was analyzed in triplicate. RNA extraction was performed using Norgen Single Cell RNA 18S gene was used as endogenous reference to quantify relative purification kit (Norgen Biotech, Cat. 51800) according to expression of each gene by the difference in threshold cycle method, manufacturer’s instructions, with no b-mercaptoethanol and 2(−DDCt) and represented as the log2FC. performing DNase I treatment (Qiagen RNase-free DNase I) for 15 min with a mix of 10 mL of DNase I with 70 mL of Qiagen’s RDD buffer added directly to the column. All RNA samples were Results quantified and their quality assed using Agilent 2200 Bioanalyzer System by the IGTP Genomics platform, with a High Sensitivity MSC induce B cells enriched in RNA ScreenTape® analysis. All samples had high RNA integrity transitional B cell populations with values (RIN ≥ 9). Ribosomal RNA was depleted using NEB Next increased IL-10 secretion rRNA Depletion Kit (Human/Mouse/Rat) and RNA libraries generated by the NEB Next Ultra II Directional RNA Library Prep Human studies have shown that Breg are mostly found Kit. Libraries’ quality was assessed on a Tape Station using a “High among transitional naïve population expressing IL-10 in the Sensitivity D1000 Screen Tape” (Agilent) and quantified by absence of proinflammatory cytokines. Concordantly, to study Frontiers in Immunology 04 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 Breg induction in our system, we analyzed in vitro induced iBreg showed a significant reduction on the percentage of regulatory B cells (iBreg) after 7-day coculture of tonsillar B cells transitional B cells. with MSC in the presence of a T-cell like stimuli (TCL) (Figure 1A). We assessed by flow cytometry transitional B cell and naïve/memory phenotypes, and by ELISA, we quantified IL- iBreg have high and stable IL-10+/TNFa+ 10 and TNFa secretion. cell ratios that allow the sorting of IL-10+ After 7 days in coculture with MSC, B cells were enriched in populations by flow cytometry for transitional B cells (CD19 + CD24 Hi CD38 Hi , 14%), and downstream analysis significantly increased naïve B cell populations (CD19+ CD27-, 84%), compared to activated B cells (2.26% and 53.3% To date, Breg studies define IL-10 as the most representative respectively) (Figures 1B-E). Kinetic study of the induction of Breg cell marker and the hallmark of Breg functionality in Transitional B cell populations, showed that they were humans. Thus, we aimed to characterize IL-10 expression on significantly increased over time peaking at day 7, while naïve our iBregs produced cells. B cell populations remained stable in the presence of MSC. At day 7, intracellular staining revealed that 40% of iBreg These Transitional B cells were predominantly of a naïve expressed IL-10, while only 4% were positive for TNFa staining. phenotype, as over 90% were identified as naïve B cells Although slightly increased, IL-10+/TNFa+ ratio in iBregs did (CD19 + IgD + CD27 - ) (Supplementary Figure 3). IL-10 not significantly differ from activated B cells (Figures 3A, B). In secretion increased after day 5, with a peak average production contrast, when PMA and Ionomycin (PI) boost was added to at day 7 of 95 pg/mL, (Figure 1F). Moreover, no TNFa was cells before intracellular staining, a significant reduction of IL- detected in supernatants (Figure 1G), indicating no production 10+/TNFa+ ratio was observed in activated B cells, associated to of pro-inflammatory mediators by iBreg. By contrast, activated B an increase of TNFa+ cells. In contrast, iBreg maintained a cells did not show any production of IL-10 nor TNFa after day 5, similar profile observed in non-PI iBreg conditions (Figure 3B). suggesting a lack of long-term cellular stimulation in the absence Of note, iBreg significantly increased IL-10+/TNFa+ ratio of MSC. compared to PI-boosted activated B cells. Moreover, we Based on these observations, and according to surface analyzed intracellular IL-10 expression in MSC-B cell markers and cytokine expression, we stablished day 7 as the coculture over 7-day period performing 4h PI-boosting at each reference timepoint for downstream experiments involving timepoint. In these conditions, IL-10 was significantly increased iBreg analysis due to transitional B cell enrichment and at day 6 and 7 compared to previous time-points (Figures 3C, maximum IL-10 secretion in all samples. D). These results indicate that a 4h PI stimulation did not significantly alter the cytokine profile of iBreg, while it promoted proinflammatory cytokines in activated B cells. iBreg inhibit T cell proliferation and are To better characterize iBreg populations obtained in our stable after MSC coculture MSC-B cell coculture conditions at day 7, we sorted iBreg populations according to IL-10 expression (Figure 3E). A total In order to validate the immunomodulatory role of enriched of 6 iBreg samples from 3 different tonsils and 2 MSC donors Transitional B cell and Breg induction in vitro we performed T were sorted into IL-10+ and IL-10- fractions. Only iBreg PI- cell proliferation assays to explore the immunomodulatory boosted samples allowed recovering enough number of IL10+ B potential of iBreg after 7 days of MSC-B cell coculture. T cells cells for further analyses. RNA was extracted from sorted isolated from same donor Tonsils were cocultured for 4 days in samples and analyzed, showing ideal yields and excellent the presence of iBreg. iBreg significantly reduced polyclonal T integrity (Supplementary Table 1) for transcriptomic analysis. cell proliferation in a dose dependent manner. This effect was clearly observed at 1:2 T cell to B cell ratios compared to activated B cells, but was not maintained at lower iBreg ratios IL-10+ iBreg are not a distinct B cell (Figures 2A, B). functional subset, while IL-10- iBreg are To investigate whether iBreg showed a stable phenotype, we enriched in extracellular matrix and explored the effect of different B cell stimulatory molecules immune regulation genes added to MSC-B cell cocultures at day 7. To this end, iBreg were recovered and four different Breg polarizing stimuli were RNA-seq analysis of sorted IL-10+ and IL-10 - iBreg added to iBreg in the absence of MSC: 1) TCL stimuli (IL-2, successfully identified over 1700 genes were differentially CD40 agonist and anti-IgM), 2) CpG and 3) LPS (Figure 2C). expressed among both populations (Figure 4A), and showed While iBreg under unstimulated, LPS and CPG stimulated distinct expression between groups (Figure 4C), that conditions, remained phenotypically stable, TCL stimulated discriminated both populations. Frontiers in Immunology 05 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 A B C D E F G FIGURE 1 MSC induce regulatory B cells enriched in transitional B cell phenotypes and secrete IL-10. (A) Graphical protocol for the generation of in vitro induced regulatory B cells (iBreg) and activated B cells and downstream analysis including gating strategy of flow cytometry samples. (B) Representative FACS plots showing the percentage of Transitional B (CD19+CD24HiCD38Hi) cells from alive B cells (CD19+7AAD-). (C) Summarized data for the frequency of Transitional B cells, showing a significant increase at day 7 (D) Representative FACS plots showing the percentage of naïve B cells (CD19+CD27-) from alive B cells (CD19+7AAD-). (E) Summarized data for the frequency of Naïve B cells showing a significant increase in the presence of MSC. (F) Supernatant cytokine quantification by ELISA of IL-10. IL-10 production increased after day 5, with a peak production observed at day 7. (G) Supernatant cytokine quantification by ELISA of TNFa. TNFa production was absent in iBreg and activated B cells. Data shows at least three independent experiments in each group. Cells from 2 MSC donors and 3 Tonsil donors were used. Error bars represent SD. Two-way ANOVA was performed to determine statistical significance. *p < 0.05, **p < 0.01, ****p < 0.0001. Frontiers in Immunology 06 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 A B C FIGURE 2 iBreg suppress T cell proliferation. iBreg stability was analyzed after day 7 in the absence of MSC showing high stability. (A) T cell proliferation assay showing percentage of MFI (median fluorescence intensity) reduction for each condition compared to stimulated T cells. iBreg inhibited T cell proliferation at high B to T cell ratios (B) Histogram representing CFSE expression on T cells (C) Graph represents iBreg transitional B cell populations in stability assays. Figure represents transitional values grouped by stimuli condition. iBreg were for the most parts stable under different stimuli after coculture with MSC. Only significantly different condition was TCL stimuli, decreasing the percentage of transitional B cell. Data shows at least three independent experiments in each group. Cells from 2 MSC donors and 3 Tonsil donors were used. Error bars represent SD. Two-way ANOVA was performed to determine statistical significance. *p < 0.05, ***p < 0.001. First, we studied IL-10+ B cells differentially expressed genes MSC induce extracellular matrix genes (DEG). Gene ontology analysis of biological processes by gene that are upregulated in iBreg when ontology identified enrichment in processes only related to compared to activated B cells cellular division and activation (Figure 4B), and no genes related to immune regulation, nor related to Breg function, To identify potential Breg biomarkers, we took a closer look at except for IL-10 expression. DEG with top statistical significances (Supplementary Table 2). Unexpectedly, we observed higher abundances and log2FC Representation of the top 20 significant genes biological processes values of expressed genes in the IL-10- DEG, when stablishing a (Figure 5A) recapitulated similar overarching processes in both cut-off value of ±1 log2FC. In addition, IL-10- DEG were enriched IL10+ and IL10- groups compared to the whole list of DEG. IL-10+ in processes related to the generation and organization of the genes were involved in cell processes such as cell activation and/or extracellular matrix (Figure 4B). Genes identified in previous mitosis, with very few genes being involved in immune regulation. works as Breg markers were differentially expressed in the IL- However, none of them were particularly interesting as candidates 10- iBreg DEG, such as GMZB, FASL, THSB1, IL1RN, CD1c, for downstream analysis due to heavily skewed expression among TIGIT or CD73 (Figure 4D, and Supplementary Figure 4). Other samples. In contrast, IL-10- genes showed promising potential due Breg biomarkers described in the literature were identified by to high and stable expressions across cell donors. Consistent with RNA-seq, but their expression was stable in both groups this data, IL-10- DEG involved in extracellular matrix and (Supplementary Figure 4). immune regulation were associated to functional protein Frontiers in Immunology 07 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 A B C D E FIGURE 3 iBreg show high IL-10+/TNFa+ cell ratios after day 6 of culture, that remain constant after 4h PMA and Ionomycin stimuli. IL-10 capture assay was performed and used for downstream IL-10+ and IL-10- flow cytometry associated cell sorting showing optimal results in PMA and Ionomycin samples. (A) Representative FACS plots showing the percentage of IL-10 and TNFa positive B cells in Activated B cell and iBreg conditions. (B) Ratio of IL-10+ and TNFa+ B cells after 7-day culture by intracellular staining after 4h stimulation with PMA and Ionomycin (PI) or without further stimulation (NS). In PI samples only iBreg retain high ratio IL-10+/TNFa+, showing higher phenotypical stability and regulatory cytokine profile, which was comparable to NS iBreg allowing for the use of PI stimuli for downstream applications. (C) Representative FACS plots showing the percentage of IL-10 positive iBreg at days 0, 4, 5, 7 of culture (D) Analysis of the percentage of IL-10+ B cells over 7-day culture performing 4h PMA and Ionomycin stimulation and intracellular staining. iBreg show a significant increase of IL-10+ B cells after day 6, suggesting optimal IL-10 expression at day 6 and 7 (C) Graphical protocol for IL-10 capture assay and the associated FACS-based cell sorting of IL-10+ and IL-10- populations (E) Percentage of IL-10+ cells by flow cytometry in sorted samples. A total of 6 samples were sorted in each condition. PI samples showed high percentages of IL-10+ cells that allowed for transcriptomic analysis. For figures A and B Data shows at least three independent experiments in each group. Cells from 2 MSC donors and 3 Tonsil donors were used. Figure D represents per condition 6 sorted samples, performed in three batches of 2 samples. Error bars represent SD. Two-way ANOVA, One-way ANOVA or non-parametric T- test was performed to determine statistical significance. *p < 0.05, ***p < 0.001. networks, as we could find several mediators of specific pathways To validate our RNA-seq results, we further analysed gene simultaneously expressed, composing two clearly defined and expression by qPCR. First, we reproduced RNA-seq results independent functional networks (Figure 5B). These findings comparing gene expression of 25 DEG genes in sorted IL-10+ may suggest an enrichment in these particular pathways that and IL-10- iBreg (data not shown). To further verify the may indicate an iBreg mechanism of action. biomarker potential of these genes, we compared gene Frontiers in Immunology 08 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 B A C D FIGURE 4 RNA-seq analysis of differentially expressed genes (DEG) shows an increased abundance of IL-10- DEG that are involved in extracellular matrix organization and immune regulation. IL-10+ are mainly involved in cellular division and activation. (A) Table shows number of DEG in both groups. We observed a higher relative abundance of IL-10- DEG (B) Gene ontology analysis of biological processes in IL-10+ and IL-10- DEG. IL-10+ DEG are enriched in cellular division processes suggesting that IL-10+ iBreg populations do not constitute a particular Breg population. IL-10- DEG are associated to immune modulation and extracellular matrix organization, suggesting a potential association to Breg functionality and elucidating a possible link between extracellular matrixes and iBreg function. (C) Heatmap of top 20 DEG genes from both groups. (D) Volcano plot of genes identified by RNA-seq. Top 10 significant DEG are named on the plot. IL-10- DEG show higher fold changes and p- values and constitute potential Breg biomarkers, while IL-10 is the only gene identified in IL-10+ populations with higher significance and fold change. Colored dots represent significant genes (p-value < 0.05) with a logFC value of ± 1. expression between total non-sorted iBreg samples and in vitro in iBreg compared to activated B cells included mostly those activated B cells (Figure 5C). Nine out of the 24 genes tested had involved in the aforementioned extracellular matrix processes, a differential expression in iBreg samples, while the others were FN1, THSB1, SPARC, TGFBI, COL1A2, IL-10, ANXA1, LY9 and overrepresented in activated B cells, or had similar expressions CXCR4. Overall, our results suggest that MSC induce B cells to in both populations. Genes with a significant higher expression express genes involved in extracellular matrix organization, and Frontiers in Immunology 09 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 A B C FIGURE 5 IL-10- DEG involved in extracellular matrix organization are significantly upregulated in total iBreg when compared to activated B cells (A) Chord diagrams show biological processes of the top 20 DEG according to adjusted p-value, recapitulating similar biological processes to those identified in analysis involving all DEG. (B) Protein interactions of top DEG showing clear aggrupation in two functional networks of immune regulation and extracellular matrix genes. (C) qPCR validation of identified cell markers comparing expression between iBreg and activated B cells. Markers involved in extracellular matrix organization have clear distinct expressions on iBreg, while immune related markers have similar expressions or are downregulated in iBreg. Positive logFC values represent higher expression in iBreg compared to activated B cells. Cells from 2 MSC donors and 4 Tonsil donors were used for experiments. to generate rich extracellular matrixes, as seen by the increase of when compared to IL-10-, and TGF-b related genes were fibronectin (FN1) and collagens (COL1A2) in addition to others upregulated in IL-10- populations. First, we performed 7-day not tested by qPCR. Furthermore, upregulated genes such as MSC-B cell coculture in the presence of LY-364947, a TGFBR1 TGFBI, THSB1 and the aforementioned extracellular matrix inhibitor (Figures 6A, B). The percentage of Transitional B cells genes could be related to TGF-b mediated signalling. was significantly lower in the presence of the TGF-b inhibitor (15.2% vs 7.09% and 5.45% for iBreg, 2 µM and 5 µM, respectively), which could be attributed to TGF-b from MSC iBreg inhibition of T cell proliferation is mediating transitional B cell induction. However, the percentage independent of IL-10 direct effect of transitional B cells in TGFBR1 inhibited B cells was significantly higher than in activated B cells, thus, TGF-b or TGFBR1 Finally, we investigated whether IL-10 and TGF-b were direct independent MSC mediators may also contribute to this effect. mediators of iBreg regulatory capabilities and iBreg induction, as Second, we performed T cell proliferation assays to analyse iBreg IL-10+ iBreg did not show any trait related to immune regulation inhibitory capacity in the presence of an IL-10 neutralizing Frontiers in Immunology 10 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 A B C D FIGURE 6 iBreg inhibition of T cell proliferation is independent of IL-10 effect on T cells. (A) Representative FACS plots showing the percentage of Transitional B (CD19+CD24HiCD38Hi) cells from alive B cells (CD19+7AAD-). (B) Summarized data for the frequency of Transitional B cells in iBreg. TGFBR1 inhibitors significantly reduce transitional B cells induction by MSC, but are not comparable to activated B cells levels, highlighting a partial inhibition of MSC effect. (C) Histogram representing CFSE expression on T cells. Lighter grey represents stimulated T cell. Light grey and dark grey represent, stimulated T cells coculture with iBreg at 1:2 T to B cell ratios in the presence of an IL-10 neutralizing antibody or a TGFBR1 inhibitor, respectively. (D) Bar graph representing reduction of proliferating T cells for each condition compared to stimulated T cells controls. iBreg and iBreg samples with IL-10 neutralizing antibodies (IL-10inh 10 and 20 µg/mL) show equal inhibition of T cell proliferation, which is significantly increased when compared to activated B cells. iBreg samples with TGFBR1 inhibitor (TGFBR1inh 2 and 5 µM) show reduced suppression capabilities, while not being significantly different from activated B cells and B cells from MSC cocultures that were cocultured in the presence of inhibitor from day 0 (TGFBR1-inhibited B cells). Data shows at least three independent experiments in each group. Cells from 2 MSC donors and 3 Tonsil donors were used. Error bars represent SD. One-way ANOVA was performed to determine statistical significance. *p < 0.05. antibody and the same TGFBR1 inhibitor. iBreg inhibited T cell Discussion proliferation even in the presence of two different concentrations of IL-10 neutralizing antibodies (Figures 6C, D). iBreg inhibition Breg have been extensively studied in the literature due to their of T cell proliferation was statistically significant when compared association with tolerance in kidney transplanted patients (36, 37) to activated B cells either in the presence or absence of IL-10 and autoimmune diseases (9). However, there is a lack of unique blocking antibodies. However, TGFBR1 inhibition significantly defining human Breg markers that limits their clinical application. reduced iBreg supression of T cell proliferation (17.28 vs 7.1 and The canonical view on human Breg was originally focused on IL-10 6.3% of inhibition for iBreg, 2 µM and 5 µM, respectively). as the gold standard to discern regulatory B cells, however, Breg Moreover, B cells cultured in the presence of 2 µM TGFBR1 research has diversified in recent years, transitioning from the study inhibitor in MSC-B cell cocultures (TGFBR1inh 2 µM B cell of Breg as a single subset, to the association of different markers to condition, Figure 6B) showed no suppression of T cell Breg across all B cell subsets (5–8). proliferation, and were not significantly different from activated In this work, we aimed to induce and characterize Breg by B cells (TGFBR1-inhibited B cell condition, Figure 6D). culturing B cells with mesenchymal stromal cells (MSC). MSC Frontiers in Immunology 11 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 have been thoughtfully studied and used for their regulatory and expression of GZMB in the RNAseq analysis and found it regenerative capabilities. Nevertheless, MSC regulatory effect is differentially expressed in IL10- B cells, although it was no not enough to induce Breg, as co-stimulation by T helper action overexpressed in iBregs compared to activated B cells and/or BCR, TLR and CD40 activation are necessary for in vitro (Figure 5D). In this line, markers and potential key mediators studies. This need for cell activation to induce Breg is in line with for Breg identified in the literature [reviewed in (43)] such as the reported increase of the number of Bregs in response to PD-L1, IL-35, IDO among others have been interrogated in the inflammation (38). Furthermore, MSC regulatory effect is also RNAseq data. The expression of these genes has not been found boosted by inflammatory conditions (39). Thus, B cell-MSC differentially expressed between the two groups of analysis. interaction might constitute an in vitro model that better In this study, we identified immune related genes, in recapitulates Breg expansion in response to inflammatory addition to extracellular matrix (ECM) organization markers, conditions to promote tissue homeostasis, if the as being over-represented in sorted IL-10- iBreg. We compared microenvironment possesses a correct balance of regulatory expression of this markers by qPCR between iBreg and in vitro and proinflammatory components. activated B cells. We found ECM genes to be upregulated in In this study, we started by culturing B cells with MSC, iBreg, with substantial overexpression of Fibronectin, Collagen soluble IL-2, anti-IgM and CD40 agonist (T cell-like and Thrombospondin-1. In addition, upregulation of stimulation). We found MSC and T cell-like stimulating extracellular matrix and markers such as Thrombospondin-1 factors to allow for the expansion of moderate numbers of IL- and TGFBI point at TGF-b, which is highly produced by MSC, 10 producing iBreg, enriched in transitional B cell populations, as a crucial mediator of iBreg induction. In fact, inhibition of with no associated TNFa induction. In agreement with previous TGFBR1 in MSC-B cell cocultures partially abrogated descriptions of MSC effect on B cells, most B cells maintained a transitional B cell induction. This effect was correlated to their naïve B cell phenotype, and were able to suppress T T cell modulation capabilities, as inhibition of TGFBR1 in the cell proliferation. MSC-B cell coculture fully inhibited B cell-mediated suppression Given the importance of IL-10+ Breg populations, we first of T cell proliferation (Figure 6D) rendering TGFBR1-inhibited determined their intracellular expression without further iBregs comparable to activated B cells. MSC secretome is a major stimulation compared with a 4h PMA and Ionomycin (PI) contributor to MSC therapeutic effect, and according to our in boost. Overall, iBreg showed high IL-10+/TNFa+ B cell ratios, vitro results, TGF-b could be a key component for Breg that were not altered by PI boost. We sorted IL-10+ and IL-10- induction. Moreover, inhibition of TGFBR1 was also directly iBreg by performing a modified IL-10 capture assay with flow involved in iBreg immunomodulatory action. Functional iBreg cytometry associated cell sorting to study differential gene displayed a reduced suppression of T cell proliferation in the expression by RNA-seq. Our major unexpected finding was presence of TGFBR1 inhibitor (Figure 6D) showing that TGF-b that IL-10 + iBreg populations bear no enrichment in is partially involved in iBreg mediated suppresion. regulatory markers, while IL-10- iBreg populations were Classically, B cells studies have focused on antibody enriched in regulatory and extracellular matrix genes. As a production, cellular costimulation, antigen presentation and proof of concept, we performed T cell suppression assays and cytokine production. However, MSC-B cell coculture found that IL-10 neutralization had no effect over iBreg highlights the importance of ECM regulation, which may be immunomodulatory capabilities. crucial for immune modulation in tissues and is omitted in most So far, IL-10 was considered crucial for MSC-induced in vitro settings. ECM components orchestrate cell survival, regulatory B cell populations, however, IL-10 has been proven activation, differentiation and proliferation, playing an a double edge sword in human Breg in vitro studies (40). IL-10 is important role in the modulation of tissue homeostasis and a regulatory cytokine highly expressed in vivo by regulatory cellular crosstalk. ECM proteins found in IL-10- iBreg, such as subsets such as Treg, but it is also expressed in B cells in response Fibronectin1, Thrombospondin1 and Laminin 4 to cellular activation, which might be artefactually overused to (Supplementary Figure 4) have been described to modulate T allow in vitro culture, compared to the degree of stimulation cell activation and differentiation (44–48), and may be a major present in vivo. Moreover, IL-10 production in vitro is associated mediator of Breg immunomodulation. to TRL ligation (41) which could explain the absence of bona fide B cells interact in lymph nodes with follicular T helper cell, IL-10+ Breg cells in our culture setting, as we did not use TLR but also stromal cells such as follicular reticular cells, that will ligands for B cell activation. Therefore, we cannot conclude that stablish the ECM. Upon circulation and migration to IL-10 is associated to Breg in our experiment settings, thus, inflammatory environments, B cells will be challenged by a MSC-B cell coculture may be an interesting platform for the particular ECM composition that will influence immune study of IL-10 independent Breg mechanisms. These findings are responses. Whether Breg function is dictated by ECM in line with other Breg in vitro culture systems describing composition or Breg have the capability to influence “foreign” secreted factors such as GZMB as the mechanism of action of ECM environments remains to be answered in future studies. In Breg, independently of IL-10 (42). In fact, we have studied the addition, a better understanding of such markers would allow for Frontiers in Immunology 12 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 the better monitoring of inflammatory responses in MSC of a grant from PERIS-PIF-Salut (SLT017/20/000158), MC-S is cell therapies. supported by a grant from ISCIII (FI20/00021), FB is a senior MSC-Breg induction elucidates B cell populations with researcher from Germans Trias i Pujol Health Science Research characteristic features that open up the study of novel Breg Institute, supported by the Health Department of the Catalan subsets characterized by the generation of extracellular matrixes Government (Generalitat de Catalunya) and MF is supported by and immune modulation independently of IL-10 direct action. ISCIII (MS19/00018), co-funded by ERDF/ESF, “Investing in In addition, TGF-b mediates MSC-induced Breg generation and your future”. immunomodulatory capabilities. These findings build the basis for further mechanistic studies, required to better understand the biology and function of Breg. Acknowledgments We thank the Flow cytometry and the genomic departments Data availability statement of the IGTP for their valuable technical help and their contribution to the experiments. The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI under accession ID: PRJNA858585 Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could Ethics statement be construed as a potential conflict of interest. All participants and/or their legal tutor gave written informed consent for participation. The study protocol followed the principles of the Declaration of Helsinki and was Publisher’s note approved by the Clinical Research Ethics Committee of our All claims expressed in this article are solely those of the institution (Comitè Ètic d’Investigació Clı́nica, Refs CEIC: PI- authors and do not necessarily represent those of their affiliated 16-056). organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed Author Contributions or endorsed by the publisher. SG: Designed the study, performed experiments, formal analysis, figure creation, data curation, and writing original draft. NS-H: Performed experiments, formal analysis and data Supplementary material curation. MC-S: Performed experiments and formal analysis. The Supplementary Material for this article can be found LC-P: Performed experiments and formal analysis. MM-F: online at: https://www.frontiersin.org/articles/10.3389/ Performed experiments. DG: Human sample and data fimmu.2022.957797/full#supplementary-material curation. FB: Study design and formal analysis. MF: Designed the study, performed experiments, formal analysis, data SUPPLEMENTARY FIGURE 1 curation. All authors interpreted, discussed the data and MSC differentiate to adipocyte and osteocyte cell types and express MSC surface markers. (A) Pictures from MSC multilineage differentiation assays. participated in reviewing and editing the draft. All authors Left column shows adipocyte differentiation from donors 1 and 2. Right contributed to the article and approved the submitted version. column shows osteocyte differentiation from donors 1 and 2. (B) Flow cytometry analysis of the percentage of positive cells for each MSC marker. Cells are positive for CD90, CD105 and CD73, and negative for CD45 and HLA-DR. Histograms show isotype control in red, and 4 MSC Funding donors in different colors. Bottom right bar graph represents mean percentage of positive cells for each marker. This work was supported in part by SGR program of SUPPLEMENTARY FIGURE 2 Generalitat de Catalunya (2017-SGR-301 REMAR Group) and Anti-IL-10 antibody neutralized recombinant IL-10 detection by ELISA. Instituto Carlos III project PI17/00335, integrated in the 100 and 50 pg/mL of recombinant IL-10 was detected by ELISA without anti-IL-10 antibody (-) or in the presence of 10 or 20 µg/mL of antibody. National R + D + I and funded by the ISCIII and the Two-way ANOVA was performed to determine statistical significance. ns European Regional Development Fund. SG is supported by the (non-significant) p > 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p Catalan Health department (“Departament de Salut”) in receipt < 0.0001. Frontiers in Immunology 13 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 SUPPLEMENTARY FIGURE 3 performed to determine statistical significance. ns (non-significant) p > MSC induce regulatory B cells that are primarily enriched in transitional 0.05, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. naïve B cells, with excellent cell viability and maintain low memory cell. (A) Flow cytometry analysis of the percentage of alive B cells, gated as cells SUPPLEMENTARY FIGURE 4 CD19+ and 7AAD-. (B) Flow cytometry analysis of the percentage of naïve Analysis of published Breg markers distribution between IL-10+ and IL- B cells, gated as cells CD27- and IgD+, in transitional B cell populations, 10- groups comparing normalize gene expression. Graphs include gated as CD19+CD24HiCD38Hi. (C) Flow cytometry analysis of the gene Name and p-value from the statistical analysis. Error bars percentage of memory switched B cells from total B cells, gated as cells represent SD. Non-parametric T-test was performed to determine CD19+CD27+IgD-. Error bars represent SD. Two-way ANOVA was statistical significance. References 1. Janeway CA, Ron J, Katz ME. The b cell is the initiating antigen-presenting 19. Franquesa M, Mensah FK, Huizinga R, Strini T, Boon L, Lombardo E, et al. cell in peripheral lymph nodes. J Immunol (1987) 138(4):1051–5. Human adipose tissue-derived mesenchymal stem cells abrogate plasmablast 2. Ochsenbein AF, Fehr T, Lutz C, Suter M, Brombacher F, Hengartner H, et al. formation and induce regulatory b cells independently of T helper cells: Control of early viral and bacterial distribution and disease by natural antibodies. Modulation of b-cell function by human ASC. Stem Cells (2015) 33(3):880–91. Science (1999) 286(5447):2156–9. doi: 10.1126/science.286.5447.2156 doi: 10.1002/stem.1881 3. Rodrı́guez-Pinto D. B cells as antigen presenting cells. Cell Immunol (2005) 20. Asari S, Itakura S, Ferreri K, Liu CP, Kuroda Y, Kandeel F, et al. 238(2):67–75. doi: 10.1016/j.cellimm.2006.02.005 Mesenchymal stem cells suppress b-cell terminal differentiation. Exp Hematol (2009) 37(5):604–15. doi: 10.1016/j.exphem.2009.01.005 4. Harris DP, Haynes L, Sayles PC, Duso DK, Eaton SM, Lepak NM, et al. Reciprocal regulation of polarized cytokine production by effector b and T cells. 21. Guo Y, Chan KH, Lai WH, Siu CW, Kwan SC, Tse HF, et al. Human Nat Immunol (2000) 1(6):475–82. doi: 10.1038/82717 mesenchymal stem cells upregulate CD1dCD5(+) regulatory b cells in experimental autoimmune encephalomyelitis. Neuroimmunomodulation (2013) 5. Rosser EC, Mauri C. Regulatory b cells: Origin, phenotype, and function. 20(5):294–303. doi: 10.1159/000351450 Immunity (2015) 42(4):607–12. doi: 10.1016/j.immuni.2015.04.005 22. Dominici M, Le Blanc K, Mueller I, Slaper-Cortenbach I, Marini FC, Krause 6. Shang J, Zha H, Sun Y. Phenotypes, functions, and clinical relevance of DS, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The regulatory b cells in cancer. Front Immunol (2020) 11:582657. doi: 10.3389/ international society for cellular therapy position statement. Cytotherapy (2006) 8 fimmu.2020.582657 (4):315–7. doi: 10.1080/14653240600855905 7. Long W, Zhang H, Yuan W, Lan G, Lin Z, Peng L, et al. The role of regulatory 23. Pittenger MF, Discher DE, Pé ault BM, Phinney DG, Hare JM, Caplan AI. b cells in kidney diseases. Front Immunol (2021) 12:683926. doi: 10.3389/ Mesenchymal stem cell perspective: Cell biology to clinical progress. NPJ Regener fimmu.2021.683926 Med (2019) 4(1):22. doi: 10.1038/s41536-019-0083-6 8. Catalá n D, Mansilla MA, Ferrier A, Soto L, Oleinika K, Aguilló n JC, et al. 24. Carreras-Planella L, Monguió -Tortajada M, Borràs FE, Franquesa M. Immunosuppressive mechanisms of regulatory b cells. Front Immunol (2021) Immunomodulatory effect of MSC on b cells is independent of secreted 12:611795. doi: 10.3389/fimmu.2021.611795 extracellular vesicles. Front Immunol (2019) 10:1288. doi: 10.3389/ 9. Blair PA, Noreña LY, Flores-Borja F, Rawlings DJ, Isenberg DA, Ehrenstein fimmu.2019.01288 MR, et al. CD19+CD24hiCD38hi b cells exhibit regulatory capacity in healthy 25. Li Z, Kupcsik L, Yao SJ, Alini M, Stoddart MJ. Mechanical load modulates individuals but are functionally impaired in systemic lupus erythematosus patients. chondrogenesis of human mesenchymal stem cells through the TGF-beta pathway. Immunity (2010) 32(1):129–40. doi: 10.1016/j.immuni.2009.11.009 J Cell Mol Med (2010) 14(6A):1338–46. doi: 10.1111/j.1582-4934.2009.00780.x 10. Flores-Borja F, Bosma A, Ng D, Reddy V, Ehrenstein MR, Isenberg DA, 26. Hung SP, Yang MH, Tseng KF, Lee OK. Hypoxia-induced secretion of TGF- et al. CD19+CD24hiCD38hi b cells maintain regulatory T cells while limiting TH1 b1 in mesenchymal stem cell promotes breast cancer cell progression. Cell Transpl and TH17 differentiation. Sci Transl Med (2013) 5(173):173ra23. doi: 10.1126/ (2013) 22(10):1869–82. doi: 10.3727/096368912X657954 scitranslmed.3005407 27. Kim D, Langmead B, Salzberg SL. HISAT: A fast spliced aligner with low 11. Hagn M, Sontheimer K, Dahlke K, Brueggemann S, Kaltenmeier C, Beyer T, memory requirements. Nat Methods (2015) 12(4):357–60. doi: 10.1038/ et al. Human b cells differentiate into granzyme b-secreting cytotoxic b nmeth.3317 lymphocytes upon incomplete T-cell help. Immunol Cell Biol (2012) 90(4):457– 67. doi: 10.1038/icb.2011.64 28. Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol 12. Lindner S, Dahlke K, Sontheimer K, Hagn M, Kaltenmeier C, Barth TFE, (2019) 37(8):907–15. doi: 10.1038/s41587-019-0201-4 et al. Interleukin 21–induced granzyme b–expressing b cells infiltrate tumors and regulate T cells. Cancer Res (2013) 73(8):2468–79. doi: 10.1158/0008-5472.CAN- 29. Love MI, Huber W, Anders S. Moderated estimation of fold change and 12-3450 dispersion for RNA-seq data with DESeq2. Genome Biol (2014) 15(12):550. doi: 10.1186/s13059-014-0550-8 13. Chesneau M, Michel L, Dugast E, Chenouard A, Baron D, Pallier A, et al. Tolerant kidney transplant patients produce b cells with regulatory properties. J 30. Yu G, Wang LG, Han Y, He QY. clusterProfiler: An r package for comparing Am Soc Nephrol (2015) 26(10):2588–98. doi: 10.1681/ASN.2014040404 biological themes among gene clusters. OMICS: A J Integr Biol (2012) 16(5):284–7. doi: 10.1089/omi.2011.0118 14. Yanaba K, Bouaziz JD, Haas KM, Poe JC, Fujimoto M, Tedder TF. A regulatory b cell subset with a unique CD1dhiCD5+ phenotype controls T cell- 31. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A dependent inflammatory responses. Immunity (2008) 28(5):639–50. doi: 10.1016/ universal enrichment tool for interpreting omics data. Innovation (2021) 2 j.immuni.2008.03.017 (3):100141. doi: 10.1016/j.xinn.2021.100141 15. Bankoti R, Gupta K, Levchenko A, Stäger S. Marginal zone b cells regulate 32. Gu Z, Gu L, Eils R, Schlesner M, Brors B. Circlize implements and enhances antigen-specific T cell responses during infection. JI (2012) 188(8):3961–71. doi: circular visualization in r. Bioinformatics (2014) 30(19):2811–2. doi: 10.1093/ 10.4049/jimmunol.1102880 bioinformatics/btu393 16. Zhang M, Zheng X, Zhang J, Zhu Y, Zhu X, Liu H, et al. CD19+CD1d+CD5+ b 33. Wickham H. ggplot2: Elegant graphics for data analysis. 2nd ed. 2016. Cham: cell frequencies are increased in patients with tuberculosis and suppress Th17 responses. Springer International Publishing: Imprint: Springer (2016). 1 p. Cell Immunol (2012) 274(1–2):89–97. doi: 10.1016/j.cellimm.2012.01.007 34. Zhao S, Guo Y, Sheng Q, Shyr Y. Heatmap3: an improved heatmap package 17. Rudensky AY. Regulatory T cells and Foxp3: Regulatory T cells and Foxp3. with more powerful and convenient features. BMC Bioinf (2014) 15(S10):P16, Immunol Rev (2011) 241(1):260–8. doi: 10.1111/j.1600-065X.2011.01018.x 1471–2105-15-S10-P16. doi: 10.1186/1471-2105-15-S10-P16 18. Comoli P, Ginevri F, Maccario R, Avanzini MA, Marconi M, Groff A, et al. 35. Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. Human mesenchymal stem cells inhibit antibody production induced in vitro by STRING v11: protein–protein association networks with increased coverage, allostimulation. Nephrol Dialysis Transpl (2007) 23(4):1196–202. doi: 10.1093/ndt/ supporting functional discovery in genome-wide experimental datasets. Nucleic gfm740 Acids Res (2019) 47(D1):D607–13. doi: 10.1093/nar/gky1131 Frontiers in Immunology 14 frontiersin.org Garcia et al. 10.3389/fimmu.2022.957797 36. Newell KA, Asare A, Kirk AD, Gisler TD, Bourcier K, Suthanthiran M, et al. 43. Garcia SG, Sandoval-Hellı́n N, Franquesa M. Regulatory b cell therapy in Identification of a b cell signature associated with renal transplant tolerance in kidney transplantation. Front Pharmacol (2021) 12:791450. doi: 10.3389/ humans. J Clin Invest (2010) 120(6):1836–47. doi: 10.1172/JCI39933 fphar.2021.791450 37. Sagoo P, Perucha E, Sawitzki B, Tomiuk S, Stephens DA, Miqueu P, et al. 44. Li L, Shirkey MW, Zhang T, Xiong Y, Piao W, Saxena V, et al. The lymph Development of a cross-platform biomarker signature to detect renal transplant node stromal laminin a5 shapes alloimmunity. J Clin Invest (2020) 130(5):2602– tolerance in humans. J Clin Invest (2010) 120(6):1848–61. doi: 10.1172/JCI39922 19. doi: 10.1172/JCI135099 38. Mauri C, Menon M. Human regulatory b cells in health and disease: 45. Fernandes NRJ, Reilly NS, Schrock DC, Hocking DC, Oakes PW, Fowell DJ. therapeutic potential. J Clin Invest (2017) 127(3):772–9. doi: 10.1172/JCI85113 CD4+ T cell interstitial migration controlled by fibronectin in the inflamed skin. 39. Luk F, Carreras-Planella L, Korevaar SS, de Witte SFH, Borràs FE, Betjes MGH, Front Immunol (2020) 11:1501. doi: 10.3389/fimmu.2020.01501 et al. Inflammatory conditions dictate the effect of mesenchymal stem or stromal cells on b cell function. Front Immunol (2017) 8:1042/full. doi: 10.3389/fimmu.2017.01042/full 46. Overstreet MG, Gaylo A, Angermann BR, Hughson A, Hyun YM, Lambert K, et al. Inflammation-induced interstitial migration of effector CD4+ 40. Lighaam LC, Unger PPA, Vredevoogd DW, Verhoeven D, Vermeulen E, T cells is dependent on integrin aV. Nat Immunol (2013) 14(9):949–58. doi: Turksma AW, et al. In vitro-induced human IL-10+ b cells do not show a subset- 10.1038/ni.2682 defining marker signature and plastically Co-express IL-10 with pro-inflammatory cytokines. Front Immunol (2018) 9:1913/full. doi: 10.3389/fimmu.2018.01913/full 47. Li Z, He L, Wilson KE, Roberts DD. Thrombospondin-1 inhibits TCR- 41. Liu BS, Cao Y, Huizinga TW, Hafler DA, Toes REM. TLR-mediated STAT3 mediated T lymphocyte early activation. J Immunol (2001) 166(4):2427–36. doi: and ERK activation controls IL-10 secretion by human b cells. Eur J Immunol 10.4049/jimmunol.166.4.2427 (2014) 44(7):2121–9. doi: 10.1002/eji.201344341 48. Khoder A, Sarvaria A, Alsuliman A, Chew C, Sekine T, Cooper N, et al. 42. Chesneau M, Mai HL, Danger R, Le Bot S, Nguyen TVH, Bernard J, et al. Regulatory b cells are enriched within the IgM memory and transitional subsets in Efficient expansion of human granzyme b–expressing b cells with potent regulatory healthy donors but are deficient in chronic GVHD. Blood (2014) 124(13):2034–45. properties. JI (2020) 205(9):2391–401. doi: 10.4049/jimmunol.2000335 doi: 10.1182/blood-2014-04-571125 Frontiers in Immunology 15 frontiersin.org