Abstract
Multiple sclerosis (MS) is a chronic inflammatory autoimmune disease of the central nervous system, due to an immune reaction against myelin proteins. Multipotent mesenchymal stromal cells (MSCs) present immunosuppressive effects and have been used for the treatment of autoimmune diseases. In our study, gene expression profile and in vitro immunomodulatory function tests were used to compare bone marrow-derived MSCs obtained from MS patients, at pre- and postautologous hematopoietic stem cell transplantation (AHSCT) with those from healthy donors. Patient MSCs comparatively exhibited i) senescence in culture; ii) similar osteogenic and adipogenic differentiation potential; iii) decreased expression of CD105, CD73, CD44, and HLA-A/B/C molecules; iv) distinct transcription at pre-AHSCT compared with control MSCs, yielding 618 differentially expressed genes, including the downregulation of TGFB1 and HGF genes and modulation of the FGF and HGF signaling pathways; v) reduced antiproliferative effects when pre-AHSCT MSCs were cocultured with allogeneic T-lymphocytes; vi) decreased secretion of IL-10 and TGF-β in supernatants of both cocultures (pre- and post-AHSCT MSCs); and vii) similar percentages of regulatory cells recovered after MSC cocultures. The transcriptional profile of patient MSCs isolated 6 months posttransplantation was closer to pre-AHSCT samples than from healthy MSCs. Considering that patient MSCs exhibited phenotypic changes, distinct transcriptional profile and functional defects implicated in MSC immunomodulatory and immunosuppressive activity, we suggest that further MS clinical studies should be conducted using allogeneic bone marrow MSCs derived from healthy donors. We also demonstrated that treatment of MS patients with AHSCT does not reverse the transcriptional and functional alterations observed in patient MSCs.
Keywords
Introduction
Multiple sclerosis (MS) is a chronic inflammatory autoimmune disorder of the central nervous system, affecting more than 1 million people worldwide (21,37). The disease is primarily mediated by cluster of differentiation 4-positive T helper 1 (CD4+ Th1) cells, Th17 cells, T CD8+ cells, B-lymphocytes, and activated microglia (37). These immune cells cross the blood–brain barrier and react against myelin proteins, causing demyelination, axonal damage, and progressive neurologic disability in young adults (47,48).
Conventional treatment for MS patients includes corticosteroids and immunomodulatory/immunosuppressive drugs that decrease inflammation in the central nervous system and stabilize disease activity (23,54). High-dose immunosuppressive regimens, followed by autologous hematopoietic stem cell transplantation (AHSCT), have been used with satisfactory results, controlling or reducing disease progression in most patients with refractory disease (7–9,35,42).
Several reports have shown that bone marrow-derived mesenchymal stromal cells (MSCs) are able to modulate innate and adaptive immunity cell responses and induce tolerance (13,16,30,52), thus supporting the rationale for their application in the treatment of autoimmune diseases (11,25,53). Among the immunosuppressive and immunoregulatory functions, MSCs have the ability to inhibit i) T CD4+ cell proliferation; ii) proinflammatory Th1 and Th17 cytokine secretion, iii) cytotoxicity and proliferation of natural killer (NK) and T CD8+ cells, iv) differentiation of monocytes into dendritic cells, v) apoptosis and respiratory burst of neutrophils, and vi) B-cell proliferation and antibody production (30,52). In addition, MSCs induce generation and expansion of regulatory T-cells (6,17–19,46). Several mechanisms have been shown to be responsible for the immunomodulatory properties of MSCs, and currently it is a consensus that soluble factors released by these cells play a key role in their suppressive and immunomodulatory activity (3,13). Some of these soluble factors are constitutively produced by MSCs, and other factors are induced after licensing of MSCs in inflammatory microenvironments (3,13,14). Among these soluble factors are indoleamine 2,3-dioxygenase (IDO), prostaglandin E2 (PGE2), transforming growth factor-β1 (TGF-β1; TGFB1), interleukin 6 (IL-6), hepatocyte growth factor (HGF), leukemia inhibitory factor (LIF), human leukocyte antigen class I G5 (HLA)-G5, chemokines, and metalloproteinases (3,13,52).
The administration of MSCs in mice with experimental autoimmune encephalomyelitis (EAE) suppressed specific T-cell proliferation; decreased interferon-γ (IFN-γ), tumor necrosis factor (TNF), and IL-17 secretion; shifted the cytokine profile to a Th2 polarization; and increased the number of regulatory T-cells (2,20,41,58). In addition, MSCs migrated to the demyelinated areas, where they diminished cellular infiltrate, stimulated proliferation of oligodendrocytes and astrocytes through the release of soluble factors, mainly HGF (1,2), and decreased demyelinated areas and clinical EAE disability scores (27,34, 59,60). Based on EAE studies, some phase I/II clinical trials have been performed, treating MS patients with autologous MSCs (12,26,51,55).
However, whether or not MSCs from MS patients have functional defects that could affect therapeutic results efficacy, in an autologous setting, is a question that has not been answered. To accomplish this goal, we evaluated the in vitro characteristics of bone marrow-isolated MSCs from MS patients and healthy controls, through their differential gene expression profiles and immunomodulatory function. We showed that MSCs from patients and controls exhibit different transcription profiles. We also demonstrated that patient MSCs secreted lower amounts of anti-inflammatory cytokines and exhibited decreased capacity to inhibit T-lymphocyte proliferation. Furthermore, to elucidate if treatment of MS patients with AHSCT could reverse the transcriptional and functional alterations observed in their MSCs, we evaluated the differential gene expression profile of MSCs isolated from MS patients before and 6 months after treatment. We showed that the transcriptional profile of patient MSCs after transplantation was closer to that from pretransplant MSC samples than from their healthy counterparts, suggesting that treatment with AHSCT does not reverse the alterations observed in MSCs from MS patients.
Materials and Methods
Patients and Controls
Demographic Features of Multiple Sclerosis Patients and Clinical Parameters at Baseline (Pretransplantation) and 6 Months Post-AHSCT
F, female; M, male; MS, multiple sclerosis; SP, secondary progressive; PP, primary progressive; RR, relapsing remitting; EDSS, expanded disability status scale; AHSCT, autologous hematopoietic stem cell transplantation; ND, not determined; MRI, magnetic resonance imaging; Gd–, absence of gadolinium-enhancing lesions; Gd+, presence of gadolinium-enhancing lesions; STER, IV corticosteroids; IFN, β interferon; GA, glatiramer acetate; CY, cyclophosphamide.
Isolation and Characterization of Bone Marrow-Derived MSCs
Bone marrow aspirate samples (3–5 ml) were collected in the presence of ethylenediaminetetraacetic acid (EDTA), and mononuclear cells were separated using Ficoll-Hypaque (Amersham-Pharmacia, Uppsala, Sweden) gradient density (5). Mononuclear cells (1–4 × 107) were resuspended in α-MEM (a modification of minimum essential medium; Gibco BRL Life Technologies, Grand Island, NY, USA) medium containing 15% HyClone characterized fetal bovine serum (FBS; Thermo Scientific, Rockford, IL, USA), supplemented with 100 μg/ml penicillin (Gibco), 100 μg/ml streptomycin (Gibco), and 2 mM L-glutamine (Gibco). Cells were plated and incubated at 37°C in 5% CO2 atmosphere. After 7 days, nonadherent cells were removed, and the culture medium was changed every 3 days. MSCs were harvested at subconfluence using 1% trypsin–EDTA solution (Sigma-Aldrich, St. Louis, MO, USA) and cultured until the third passage for cell characterization, RNA extraction, and coculture assays. Cell morphology was analyzed using an inverted microscope (Carl Zeiss, Jena, Germany) with the aid of the AxioVision 3.0 software (Carl Zeiss). MSCs isolated from patients (pretransplantation, n = 32, mean age = 37.9 years; posttransplantation, n = 24, mean age = 39.3 years) and healthy controls (n = 22, mean age = 34.1 years) were analyzed by flow cytometry using the following conjugated monoclonal antibodies: CD90-phycoerythrin (PE), CD13-allophycocyanine (APC), CD29-APC, CD105-APC, CD73-PE, CD44-fluorescein isothiocyanate (FITC), CD166-PE, HLA-ABC-PE, CD146-FITC, CD54-PE, CD106-FITC, HLA-DR-FITC, CD14-PE, CD31-FITC, CD34-APC, and CD45-FITC (Becton-Dickinson, San Jose, CA, USA). All monoclonal antibodies were used at concentrations recommended by the manufacturer. All flow cytometry analyses were performed using a FACSort apparatus (Becton-Dickinson), and 20,000 events were acquired and analyzed by the Cellquest software (Becton-Dickinson).
The ability of MSCs to differentiate into osteogenic and adipogenic lineages was assayed by culturing in specific media. Adipogenic medium consisted of α-MEM containing 15% FBS, supplemented with 10 μM dexamethasone (Prodome, Campinas, SP, Brazil), 10 μg/ml insulin (Sigma-Aldrich), and 100 mM indomethacin (Sigma-Aldrich). Osteogenic medium included α-MEM, 7.5% FBS, supplemented with 0.10 μM dexamethasone, 100 μM ascorbic acid (Reagen, Colombo, PR, Brazil), and 10 mM β-glycerophosphate (Reagen). Culture medium was changed every 3 days for 27 days. Cultures were stained with Sudan II-Scarlet (Oil red, Sigma-Aldrich) for detection of lipids and von Kossa (Sigma-Aldrich) for detection of mineralized extracellular matrix.
RNA Extraction, Microarray Hybridizations, and Bioinformatics Analysis
Total RNA was isolated using Trizol (Invitrogen Life Technologies, Carlsbad, CA, USA), according to the manufacturer's instructions (10), purified with RNeasy mini Kit (Qiagen, Valencia, CA, USA) and spectrophotometrically analyzed at 260 and 280 nm (NanoDrop, ND-1000 UV-VIS; Thermo Fisher Scientific, Waltham, MA, USA). Samples from the same patients at pretransplantation and post-AHSCT (pre-AHSCT, n = 4; post-AHSCT, n = 4; mean age = 36.7 years) and healthy subjects (n = 4; mean age = 31 years) were used. To evaluate global gene expression profiles of patient and control MSCs, we used the Agilent one-color [cyanine 3 (Cy3) fluorochrome] microarray-based gene expression platform according to manufacturer's instructions (Agilent Technologies, Santa Clara, CA, USA). Briefly, RNA (1 μg) was amplified and labeled using the low RNA input linear amplification kit plus (Agilent). Complementary RNA (cRNA) samples were hybridized onto whole human genome microarray slides 4 × 44K G4112F (Agilent) overnight at 65°C, followed by washing steps. Slides were scanned using GenePix 4000B (Axon Instruments, Foster City, CA, USA), and the hybridization signals were extracted using the Agilent Feature Extraction software. Bioinformatics analyses include Gene Set Enrichment Analysis (49); Database for Annotation, Visualization, and Integrated Discovery (DAVID) (24); and Ingenuity Pathways Analysis (IPA) systems (Ingenuity Systems, Redwood, CA, USA). The data for the microarray analyses used in this study are available online at the Minimum Information About a Microarray Experiment (MIAME) public database ArrayExpress (http://www.ebi.ac.uk/arrayexpress/), accession E-MEXP-3905.
Coculture Assays
To test the inhibitory effect of MSCs isolated from MS patients (pretransplantation, n = 10, mean age = 36.1 years; posttransplantation, n = 9, mean age = 38.8 years) and healthy controls (n = 16, mean age = 32.7 years) on allogeneic lymphocyte proliferation, we used the carboxyfluorescein diacetate succinimidyl ester (CFSE; Invitrogen) dilution method and assessed proliferation percentages by flow cytometry. Peripheral blood mononuclear cells (PBMCs) isolated from healthy donors, which differed from the healthy MSC donors, were separated by density gradient, labeled with CFSE (10 μM, for 10 min, 37°C), and resuspended in Roswell Park Memorial Institute (RPMI) 1640 medium (Gibco) supplemented with 5% human serum albumin (Vialebex® 200 mg/ml; LFB, Rio de Janeiro, RJ, Brazil). CFSE-labeled PBMCs (concentration of 0.5 × 106 cells/ml) were added to the wells (1 ml/well) containing previously adhered patient or control MSCs, in six different dilutions (MSCs/PBMCs = 1:2, 1:5, 1:10, 1:20, 1:50, and 1:100) in the presence of 0.5 μg/ml of phytohemagglutinin (PHA, Sigma-Aldrich). All experiments were performed in triplicates, and cocultures were incubated for 5 days at 37°C with 5% CO2. On the fifth day of coculture, PBMCs were harvested, stained with anti-CD3-PE (Becton-Dickinson) at the manufacturer's recommended concentration, and analyzed by flow cytometry (FACSort apparatus). Control wells consisted of CFSE-labeled PBMCs stimulated with PHA in the absence of MSCs.
To evaluate the effect of MSCs on the induction of regulatory cells, we studied the frequency of some regulatory subsets of T-cells, recovered after coculture assays containing patient or control MSCs with PHA-stimulated PBMCs. The same samples from the inhibitory assays were used. After 5 days of coculture, PBMCs were harvested and stained for T-cell subsets with the following fluorophore-conjugated monoclonal antibodies: CD3+CD4+, CD3+CD8+, CD3+ natural killer T-cell positive (NKT+), CD4+CD25hi, CD4+CD25hi forkhead box P3 positive (Foxp3+), CD4+CD25hi cytotoxic T-lymphocyte-associated protein 4 positive (CTLA-4+), CD4+CD25hi glucocorticoid-induced TNF receptor-related protein positive (GITR+), CD8+CD8-Foxp3+ (Becton-Dickinson), CD3+CD4+HLA-G+, and CD3+CD8+HLA-G+ (Abcam, Cambridge, UK), and evaluated by flow cytometry (FACSort apparatus). All the antibodies were used at the concentrations recommended by the manufacturer.
Cytokine Quantification
To evaluate the secretion of IFN-γ, TNF, IL-1β, IL-6, IL-10, TGF-β, chemokine C-X-C motif ligand 10 [CXCL-10; IFN-γ-induced protein 10 (IP-10)], and HGF, supernatants obtained from the coculture of patient or control MSCs with PBMCs (1 MSC:2 PBMC dilution) were collected after 5 days and quantified through a cytometric bead array (Becton-Dickson) platform. HGF quantification was performed using the HGF Human ELISA kit (Invitrogen). For this experiment, supernatants were obtained from the same samples used in the inhibitory assays, and all experiments were done in triplicates.
Statistical Analyses
Kruskal–Wallis and Dunn's multiple comparison test were used to compare the surface expression of cluster of differentiation markers between patient (pre- and posttransplantation) and control MSCs. Differences in the percentage of lymphocyte proliferation, in frequency of regulatory subsets, and in cytokine levels in coculture assays were analyzed using the parametric one-way ANOVA, followed by Tukey–Kramer's post hoc test or the nonparametric Kruskal–Wallis, followed by the Dunn's multiple comparison test, according to data distribution. All values of p < 0.05 were considered to be significant. Data were analyzed using the GraphPad Prism, version 5.0 (San Diego, CA, USA). All the statistical analyses were adjusted for multiple comparisons. Gene expression data of MSCs isolated from MS patients and healthy donors were analyzed using the Linear Models for Microarray data package (http://www.tm4.org/mev_manual/limma.html) and were considered to be significant at p < 0.01 and fold change ≥ 2.0. Data were represented by mean ± SEM for both parametric and nonparametric data for clarity.
Results
Characterization of Bone Marrow Multipotent MSCs From Patients and Controls
To characterize multipotent MSCs isolated from bone marrow of MS patients and controls, we evaluated cell morphology in culture, expression of typical surface markers, and their differentiation potential into mesodermal lineage cells. MSCs exhibited the following features. i) Both patient and control cells presented fibroblast-like morphology and ability to adhere to plastic during in vitro expansion; however, patient MSCs before and after transplantation had a senescent appearance, that is, at the third passage, patient MSCs were morphologically similar to control MSCs at the eighth passage and, additionally, demanded longer time to reach healthy MSC confluence (Fig. 1A, B). ii) Patient and control MSCs expressed CD73, CD90, and CD105 markers and lacked CD14, CD31, CD34, and CD45 markers; however, quantitative differences were observed. Patient MSCs exhibited decreased surface expression of CD29 (pre-AHSCT = 79.25 ± 3.41%; post-AHSCT = 76.41 ± 4.47%), CD105 (pre-AHSCT = 58.70 ± 7.43%; post-AHSCT = 53.52 ± 7.64%), CD73 (pre-AHSCT = 67.84 ± 4.82%; post-AHSCT = 69.30 ± 3.55%), and CD44 (pre-AHSCT = 65.92 ± 5.29%; post-AHSCT = 61.58 ± 6.42%) when compared with control MSCs (CD29 = 90.42 ± 2.0%; CD105 = 88.68 ± 1.40%; CD73 = 84.73 ± 2.72%; CD44 = 84.90 ± 1.80%) (p<0.05, for all comparisons). Furthermore, posttransplantation patient MSCs showed decreased surface expression of CD166 (49.37 ± 6.68%), HLA-ABC (42.02 ± 5.99%), and CD146 (21.91 ± 5.48%) when compared with control MSCs (CD166 = 72.66 ± 3.46%; HLA-ABC = 70.03 ± 4.35%; CD146 = 42.61 ± 7.38%) (p<0.05, for all comparisons) (Fig. 1C). iii) Both patient and control MSCs exhibited adipogenic and osteogenic differentiation potentials (Fig. 1D-G).
Characterization of multipotent mesenchymal stromal cells (MSCs). (A) Control MSCs and (B) patient MSCs exhibiting fibroblast-like morphology (40×). (C) Expression of typical surface immunophenotypic markers. (D) Control MSC and (F) patient MSC adipogenic differentiation (630×). (E) Control MSC and (G) patient MSC osteogenic differentiation (630×).
Patient and Control MSCs Have Distinct Transcriptional Profiles
To evaluate hybridization signatures between patient and healthy donor MSCs and to compare hybridization signatures before and after transplantation, we performed a transcriptome analysis using microarrays. To identify functional gene sets, we performed a gene set enrichment analysis (GSEA), disclosing some MSC-relevant processes, such as “cytokine–cytokine receptor interactions” [p = 0.004; false discovery rate (FDR) q -value = 0.16], “IL-10 pathway” (p = 0.012; FDR q-value = 0.21), and “cell adhesion molecule” (p = 0.012; FDR q-value = 0.21), enriched in patient MSCs compared with control MSCs. Euclidean hierarchical clustering analyses revealed that MSCs isolated from MS patients exhibit distinct gene expression profile when compared with healthy donor MSCs (Fig. 2A). Statistical analyses showed 618 differentially expressed genes [p<0.01; fold change (FC) ≥2.0], comprising 248 repressed and 370 induced genes, as illustrated by the Volcano plot (Fig. 2B). DAVID gene ontology analyses revealed that most of the induced genes were included in the following categories: G-protein-coupled signaling receptors, intracellular signaling cascades, cellular adhesion regulation, immune response processes regulation, and tissue remodeling regulation. Repressed genes were included in the following categories: transcriptional regulation, guanosine-5′-triphosphatases (GTPases)-mediated signal transduction, rat sarcoma (Ras)-protein signal transduction, and T-lymphocyte activation. Among the significant and differentially expressed genes, we observed modulation of genes related to the immunomodulatory activity of MSCs, such as a downregulation of TGFB1 and HGF genes and an upregulation of IL10, IL6, IFN-γ receptor 1 (IFNGR1), and IFNGR2, as shown in the heatmap (Fig. 3A). The asterisks in front of the genes indicate that they are included in the list of differentially expressed genes (FC ≥ 2, p < 0.01). Furthermore, in the search for relevant MSC-related pathways, data were analyzed through the use of IPA, and we found significant modulation of the fibroblast growth factor (FGF) (p = 0.046) and HGF (p = 0.048) signaling pathways (Fig. 3B, C). Concerning FGF signaling, many genes were repressed [FGF3, FGF-binding protein 3 (FGFBP3), FGF receptor 1 (FGFR1), growth factor receptor bound protein 2-associated protein 1 (GAB1), v-akt murine thymoma viral oncogene homolog 2 (AKT2), protein tyrosine phosphatase, nonreceptor type 11 (PTPN11)], exception made for FGF2, which was induced. Similarly, in HGF signaling, HGF and other transducing signaling genes [phospholipase C γ (PLCγ), phosphoinositide 3-kinase (PI3K), AKT, v-crk avian sarcoma virus ct10 oncogene homolog-like (CRKL), cell division cycle 42 (CDC42), cyclin-dependent kinase 2 (CDK2)] were repressed, while genes RAS, protein kinase c (PKC), ras-related protein (RAP), and IL-6, which encode downstream HGF signaling molecules, were induced.
Patient and control MSCs have distinct transcriptional profiles. (A) Hierarchical supervised clustering analysis from gene expression data obtained from healthy MSCs and patient MSCs (pretransplantation). (B) Volcano plot showing differential gene expression in patient MSCs (pretransplantation) compared with healthy MSCs. Patient and control MSCs have distinct transcriptional profiles. (A) Heatmap containing enriched genes in “cytokine-cytokine receptor interactions” process, obtained from gene set enrichment analysis (GSEA), comparing patient MSCs (pretransplantation) with healthy MSCs. (B, C) Modulation of FGF (p = 0.046) and HGF (p = 0.048) signaling pathways, obtained from Ingenuity Pathways Analysis (IPA; Ingenuity Systems), containing repressed (flagged in green) and induced genes (flagged in red) in patient MSCs (pretransplantation).
Regarding the transcriptional profiles of patient MSCs before and after autologous hematopoietic stem cell transplantation, hierarchical clustering analyses revealed two clusters, one grouping healthy donor MSCs and the other MSCs from MS patients (pre- and posttransplantation samples) (Fig. 4). The transcription profiles of patient MSCs before and after transplantation differed, and both were distinct from that displayed by healthy donor MSCs. Statistical analyses showed 81 differentially expressed genes after transplantation, and most of them were upregulated in posttransplantation MSCs (p < 0.01; FC ≥ 2.0). Data mining using the above-mentioned software programs did not reveal pathways with clear biological relevance for MSCs; however, induced genes were included in the following categories: DNA metabolic processes, DNA replication, ribonucleoprotein complex, structural constituents of ribosomes, protein complex assembly, and biogenesis protein complex.
Patient (pre- and posttransplantation) and control MSCs have distinct transcriptional profiles. Euclidean hierarchical clustering analysis from gene expression data obtained from healthy MSCs and patient MSCs (pre- and posttransplantation).
Patient MSCs Have Decreased Capacity to Inhibit T-Cell Proliferation
Once patient MSCs showed significant modulation of genes involved in the immunomodulatory cell activity, we evaluated the antiproliferative activity of patient (pre- and post-AHSCT) and healthy donor MSCs in coculture assays. We observed that both patient (pre- and post-AHSCT) and control MSCs inhibited in vitro lymphocyte proliferation in a dose-dependent manner (Fig. 5A–C). However, when we compared their relative inhibitory effects, the antiproliferative activity of patient MSCs at pre-AHSCT (52.15 ± 5.65%) was significantly reduced (p = 0.024) when compared with controls (69.37 ± 7.25%) at a 1:2 ratio (Fig. 5D). No significant differences were detected when the inhibitory activity of post-AHSCT MSCs was compared with pretransplantation or with healthy donor MSCs. In addition, we observed no correlation between patients' Expanded Disability Status Score (EDSS) and the antiproliferative activity of patient MSCs at pre- and posttransplantation periods.
Patient MSCs have decreased ability to inhibit T-cell proliferation. (A) Percentages of lymphocyte proliferation in cocultures with healthy MSCs. (B, C) Percentages of lymphocyte proliferation in cocultures with patient MSCs (pre- and posttransplantation). (D) Inhibitory activity of patients (pre- and posttransplantation) and healthy MSCs in cocultures with allogeneic lymphocytes. PBMCs, peripheral blood mononuclear cells; PHA, phytohemagglutinin.
Regulatory Lymphocyte Subpopulations Recovered After Cocultures of PHA-Stimulated-PBMCs With MSCs From Healthy or MS Patients (Pre- and Posttransplantation)
PHA, phytohemagglutinin; PBMCs, peripheral blood mononuclear cells; MSCs, mesenchymal stromal cells; MS, multiple sclerosis; Pre-Tx, before hematopoietic stem cell transplantation; Post-Tx, 6 months after hematopoietic stem cell transplantation; CD3, cluster of differentiation 3; NKT, natural killer T; FOXP3, forkhead box P3; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; GITR, glucocorticoid-induced tumor necrosis factor receptor; HLA-G, major histocompatibility complex, class I, G.
Patient MSCs Have Decreased Anti-Inflammatory Cytokine Secretion
Since we observed a decreased antiproliferative activity of patient MSCs, we quantified some soluble factors in the supernatants collected from coculture assays of PBMCs with patient (pre- and post-AHSCT) and control MSCs (Fig. 6). We found no significant differences (p > 0.05) in proinflammatory IFN-γ, TNF, IL-1β, and IL-6 cytokine levels in supernatants derived from cocultures with patient (pre- and post-AHSCT) and control MSCs (Fig. 6A–D). In addition, HGF and CXCL-10 secretions were similar in all analyzed cocultures (Fig. 6G, H). However, we observed that the IL-10 (p = 0.016) and TGF-β (p = 0.0002) secretions were significantly decreased in cocultures with patient MSCs at pre-AHSCT (IL-10 = 53.8 ± 13.7 pg/ml; TGF-β = 59.4 ± 6.8 pg/ml) and post-AHSCT (IL-10 = 54.5 ± 13.8 pg/ml; TGF-β = 122.0 ± 14.8 pg/ml), when compared with cocultures with control MSCs (IL-10: 126.2 ± 32.1 pg/ml; TGF-β: 226.6 ± 28.9 pg/ml) (Fig. 6E, F).
Patient MSCs secrete less anti-inflammatory cytokines. (A–D) Concentration (pg/ml) of proinflammatory (IFN-γ, TNF, IL-1β, and IL-6) and (E–F) anti-inflammatory cytokines (IL-10 and TGF-β), (G) HGF, and (H) chemokine C-X-C motif ligand 10 (CXCL-10) in supernatants from cocultures of PHA-stimulated PBMCs with healthy or MS patient MSCs (pre- and posttransplantation).
Discussion
Bone marrow MSCs have been considered an important source for autologous cell therapy for autoimmune diseases, particularly due to their immunosuppressive properties (15,31,40,51). The ability of MSCs to modulate the immune response suggests a possible role of these cells in tolerance induction in patients with autoimmune diseases and supports the rationale for MSC application in the treatment of MS patients (3,11,16). However, it is unclear whether or not MSCs isolated from bone marrow of patients with MS are normal or defective, regarding phenotypic, transcriptional profiles and functional properties. A previous study has shown that MSCs from patients with systemic sclerosis have normal functions (29). Conversely, other studies have reported that MSCs from patients with rheumatoid arthritis and systemic lupus erythematosus exhibit altered function and several other abnormalities, such as defective proliferative capacity and decreased telomerase activity (28,38,50). Papadaki and colleagues (39) demonstrated that bone marrow stromal cells from MS patients have normal hematopoietic supporting capacity.
Regarding phenotypic features of MSCs, we showed in our study that patient and control MSCs show fibroblast-like morphology, as well as adipogenic and osteogenic differentiation potentials. However, patient MSCs exhibited a senescent appearance and significantly decreased expression of surface markers (CD29, CD105, CD73, and CD44), when compared with control MSCs. Mallam and colleagues (32) analyzed five secondary progressive MS patients (mean age 46.6 years) and five controls (mean age 71 years) and detected no differences in i) morphology; ii) expansion potential; iii) ability to differentiate into adipocytes, chondrocytes, and osteocytes; iv) expression of immunophenotypic markers CD90, CD166, CD105, CD44, and CD45 at third or fourth passage (n = 3 for patients and controls). Similarly, Mazzanti and colleagues (33) evaluated 10 MS patients (RR and SP types) (mean age 37.6 years) and six healthy donors (age and sex matched) and observed no differences in i) morphology; ii) mean proliferation curves; iii) osteogenic and adipogenic differentiation capabilities; and iv) expression of surface markers CD29, CD44, CD166, CD90, CD73, HLA-ABC, CD105, and HLA-DP-DQDR. Our work differs from these previous studies, and we suggest that these discrepancies are due to differences in sample size and in experimental conditions, such as the number of passages after which MSCs were submitted to immunophenotyping analyses.
Concerning transcriptional profiles, there are no published studies comparing MSCs isolated from MS patients and healthy donors, as well as studies evaluating MSCs isolated from MS patients before and after treatment with autologous hematopoietic stem cell transplantation. When patient MSCs at pretransplantation were compared with controls, 618 differentially expressed genes were observed, some of them related to the immunosuppressive activity, with special emphasis to downregulation of HGF and TGFB1 genes and modulation of HGF and FGF signaling pathways. Once HGF and TGF-β are important molecules secreted by MSCs and involved in the suppressive functions (3,6), we suggest that both modulation of HGF signaling and decreased expression of HGF and TGFB1 genes in patient MSCs could be involved with the results of inhibition of lymphocyte proliferation and the decreased secretion of IL-10 and TGF-β in supernatants of patients' cocultures. Additionally, transcriptional profiles of patient MSCs at post-AHSCT were closer to patient MSCs at pre-AHSCT than to healthy MSCs. It is noteworthy that a previous study found 1,539 differentially expressed genes in MSCs isolated from rheumatoid arthritis patients, including those implicated in cell cycle progression from phases G1 to S, proliferation, survival, and cell adhesion processes (28).
In our study, we observed that both patient (pre- and post-AHSCT) and healthy donor MSCs inhibited lymphocyte proliferation in a dose-dependent manner. However, healthy donor MSCs exhibited a better antiproliferative effect, that is, 70% inhibition for healthy donor cells versus 50% inhibition for patient cells. Zafranskaya and colleagues (57) evaluated 25 MS patients (SP, PP, and RR types) (mean age 31.5 years) and showed that in both autologous (n = 13) and allogeneic (n = 12) conditions, MS-patient MSCs, at first or second passages, significantly reduced PHA/myelin-stimulated CD4 T-cell proliferation after 6 days of cocultivation (MSC/PBMC ratio = 1:10). Contrary to our results, Mazzanti and colleagues (33) showed similar antiproliferative activity of MS patient and healthy donor (HD) MSCs when added to mixed lymphocyte reactions (MLR) (MSC/PBMC 1:10 and 1:1,000 ratios). After 5 days of coculture, they measured the proliferative response by tritiated thymidine [(3H)-Thy] incorporation test and concluded that the in vitro immunosuppressive activity of MSCs during a MLR is not dependent on the MSC origin (MS or HD), but on their concentration. In other autoimmune diseases (rheumatoid arthritis, scleroderma, and Sjögren's syndrome), previous studies showed that MSCs isolated from healthy individuals and patients were capable of inhibiting proliferation of autologous or allogeneic lymphocytes in a dose-dependent manner and that this inhibition was not associated with clinical activity of the disease (4). We believe that our conclusions differed from those of previous studies due to experimental differences, which can hamper proper comparisons. Some points may be considered, such as i) number of passages after which MSCs were submitted to coculture assays, ii) lymphocyte proliferation stimulants, iii) period of cocultivation, iv) different ratios of cocultivation, v) different proliferation analyses.
Studies have demonstrated the role of human bone marrow MSCs on induction, recruitment, and proliferation of regulatory T-cells (17–19). The generation of Treg cells occurs through cell–cell contact, inducing the expression of Foxp3 and CD25 by CD4+ T-cells. In addition, soluble factors released by MSCs, particularly TGF-β1, PGE2, and, HLA-G5 play a pivotal role in the induction of CD4+CD25hiFoxp3 Tregs (18,19,46). Furthermore, MSCs induce CTLA-4 and GITR expression in T-lymphocytes which, in turn, exert suppressive activity in coculture assays (44). In our study, we detected similar percentages of regulatory cell subpopulations in cocultures, using either healthy donor or MS patient MSCs (pre- and post-AHSCT). Although we did not observe increased percentages of Tregs in cocultures, we cannot exclude the effect of MSCs on the function of these cells.
Several mechanisms have been proposed to explain the immunomodulatory role of MSCs, and currently there is consensus that soluble factors released by these cells play a key role in their suppressive capacity (3,13). Secretion of TGF-β by MSCs has been implicated in the activation and expansion of Tregs (6). Indeed, lymphocytes obtained from healthy individuals stimulated with alloantigen in the presence of MSCs produce large amounts of IL-10. Furthermore, blockage of IL-10 or its receptor reverses MSC-induced inhibition of lymphocyte proliferation (3,45,56). In addition, greater production of IL-10 and TGF-β is observed when cells are cocultured in close contact, compared to Transwell cultures, suggesting that the cell–cell contact is important for the suppressive and immunomodulatory activity of MSCs (3,36). In our study, concentrations of IL-10 and TGF-β were significantly lower in supernatants from cocultures of PBMCs with patient MSCs (pre- and post-AHSCT) than from cocultures with healthy donor MSCs, which could explain the reduced suppressive effect of the MS patientderived cells.
Our results are not representative of typical MS patients, since we included only those that were already refractory to conventional treatments. Nevertheless, our study shows that MS patient-derived MSCs present higher senescence, decreased expression of important immunophenotypic markers, and different transcriptional profiles, when compared with their healthy counterparts. In addition, patient MSCs exhibit decreased antiproliferative capacity and secrete lower concentrations of anti-inflammatory cytokines. We also conclude that autologous hematopoietic stem cell transplantation does not reverse the alterations observed in patient MSCs. Since the transcriptional and functional alterations are involved in the immunomodulatory and immunosuppressive activity of MSCs, we therefore suggest that further MS clinical studies should be conducted using allogeneic bone marrow MSCs derived from healthy donors.
Footnotes
Acknowledgments
We would like to thank and honor the memory of the initiator of the Bone Marrow Transplantation group of the University Hospital of the Ribeirão Preto School of Medicine, Professor Júlio César Voltarelli, mastermind of our hematopoietic stem cell transplantation studies for autoimmune disorders. The study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP, Foundation for the Support of Research in the State of São Paulo, grants #2008/58387–0); Fundação Hemocentro de Ribeirão Preto (FUNDHERP, Ribeirão Preto Blood Bank Foundation). The authors declare no conflicts of interest.