Immune impairments in multiple myeloma bone marrow mesenchymal stromal cells

Abstract

In multiple myeloma (MM), bone marrow mesenchymal stromal cells (BM-MSCs) play an important role in pathogenesis and disease progression by supporting myeloma cell growth and immune escape. Previous studies have suggested that direct and indirect interactions between malignant cells and BM-MSCs result in constitutive abnormal immunomodulatory capacities in MM BM-MSCs. The aim of this study was to investigate the mechanisms that underlie these MM BM-MSCs abnormalities. We demonstrated that MM BM-MSCs exhibit abnormal expression of CD40/40L, VCAM1, ICAM-1, LFA-3, HO-1, HLA-DR and HLA-ABC. Furthermore, an overproduction of IL-6 (1,806 ± 152.5 vs 719.6 ± 18.22 ng/mL; p = 0.035) and a reduced secretion of IL-10 (136 ± 15.02 vs 346.4 ± 35.32 ng/mL; p = 0.015) were quantified in culture medium when MM BM-MSCs were co-cultured with T lymphocytes compared to co-cultures with healthy donor (HD) BM-MSCs. An increased Th17/Treg ratio was observed when T cells were co-cultured with MM BM-MSCs compared to co-cultures with HD BM-MSCs (0.955 vs 0.055). Together, these observations demonstrated that altered immunomodulation capacities of MM BM-MSCs were linked to variations in their immunogenicity and secretion profile. These alterations lead not only to a reduced inhibition of T cell proliferation but also to a shift in the Th17/Treg balance. We identified factors that are potentially responsible for these alterations, such as IL-6, VCAM-1 and CD40, which could also be associated with MM pathogenesis and progression.

Electronic supplementary material

The online version of this article (doi:10.1007/s00262-014-1623-y) contains supplementary material, which is available to authorized users.

Introduction

Multiple myeloma (MM) is a hematopoietic neoplasm characterized by a monoclonal expansion of secreting plasma cells (PCs) in the bone marrow (BM), resulting in skeletal destruction, renal failure, anemia, hypercalcemia and recurrent infections. MM represents approximately 1 % of all malignant tumors, 10 % of hematopoietic neoplasms and 2 % of cancer deaths [1].

One of the characteristic specific to MM is that BM constitutes a required microenvironment for disease development and progression. Within this BM microenvironment, the malignant clone is able to elude immune surveillance by inducing abnormalities in immune cells (natural killer, dendritic and T cells) and by enhancing the release of immunoregulatory cytokines by microenvironmental cells [2–4]. Patients with MM exhibit a variety of numerical and functional abnormalities of T cells: abnormal T cell subsets (i.e., CD4:CD8 and Th1:Th2 ratios), reduced T cell diversity and abnormal functional responses. Tumor cells also affect the frequency and the inhibitory capacities of T regulatory (Treg) cells in MM patients [5, 6]. Finally, the proportion of T helper (Th) 17 cells and the plasma concentrations of Th17-associated cytokines are increased in MM patients [7].

Among these bone marrow microenvironmental cells, the mesenchymal stromal cells (MSCs) play a crucial role in pathogenesis through their contacts with MM-PCs. The constitutive abnormalities that are observed in MM BM-MSCs, such as a reduced inhibition of T lymphocyte proliferation, are one result of these contacts [8, 9]. MSCs possess immunomodulatory capacities and can modulate most immune effectors, particularly T cells. MSCs can inhibit the activation and the proliferation of activated T cells and induce a state of anergy. These capacities are mediated by the secretion of many soluble factors (e.g., HGF, TGF-β, IL-6, IL-10, PGE2), which requires a dynamic crosstalk between MSCs and T cells, as well as by direct cellular contacts [10].

MSCs can also impact two CD4⁺ T cell subpopulations: Th17 cells and Treg cells. Th17 cells are pro-inflammatory and are defined by a preferential secretion of interleukin (IL)-17. Tumor growth factor-β (TGF-β) and IL-6, with or without IL-21, IL-23 and IL-1, are necessary for the induction and expansion of Th17 cells from naïve CD4⁺ precursors. It has been demonstrated that MSCs induce regulatory characteristics in Th17 cells in inflammatory environments by downregulating RAR-related orphan receptor γt (RORγt) [11]. However, some evidence also exists for a Th17 cell-promoting effect of MSCs [12]. Treg cells have suppressor functions that are essentials for the prevention of autoimmunity and the resolution of inflammatory processes. These cells are characterized by surface expression of CD25 and by intracellular expression of forkhead box P3 (FOXP3). The development and proliferation of Treg cells are induced by IL-10, IL-2 and TGF-β. Many studies demonstrated an enhancement of Treg numbers and activity by MSCs [13].

Previous studies from our group and other groups [8, 9, 14] reported that the immunomodulatory functions of BM-MSCs derived from MM patients are impaired. A reduced inhibition of T lymphocyte proliferation, a lower ability to silence mitogen-stimulated T cells in G0/G1 phase, a reduced inhibition of T cell activation and a reduced rate of T cell apoptosis have been observed during co-culture of T cells with MM BM-MSCs compared with co-cultures with healthy donor (HD) BM-MSCs.

The aim of this study was to investigate the mechanisms that underlie the MM BM-MSCs immunomodulatory impairment observed in MM BM-MSCs and the impacts of these impairments on MM. We analyzed MM BM-MSCs expression of a variety of adhesion molecules and immune effectors in constitutive and inflammatory conditions. We also measured the secretion of immunoregulatory cytokines by MM BM-MSCs that were co-cultured with activated T cells. Finally, we evaluated the fate of activated T cells that were co-cultured with MM BM-MSCs.

Materials and methods

Patients

Samples were obtained after receiving written informed consent from patients and HD volunteers and after approval from the Ethical Committee of the Jules Bordet Institute and Erasmus Hospital (Brussels, Belgium). Thirty-four patients with MM and twelve HDs were included in this study, and the characteristics of these participants are listed in Supplementary Table 1. All MM patients who underwent treatment were in remission at the time of BM harvesting and did not receive grafts.

Isolation, culture and characterization of MSCs

BM was harvested from the sternum or iliac crest of the patients. BM-MSCs were isolated by using the classical adhesion method and cultured as previously described [15]. Adherent cells in the culture were identified as MSCs when they fulfilled the International Society for Cellular Therapy (ISCT) criteria [16]. BM-MSCs were cultured for 24 h in the presence or absence of inflammatory medium (i.e., Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 2 mM l-glutamine, 50 U/mL of penicillin, 50 µg/mL of streptomycin, 1,000 U/mL of interferon-γ (IFN-γ), 3,000 U/mL of IFN-α, 50 ng/mL of tumor necrosis factor-α (TNF-α) and 25 ng/mL of IL-1β) (R&D Systems, Abingdon, UK). The harvested cells were analyzed using flow cytometry with the following markers to establish the immunological profile of the MSCs: members of the human leukocyte antigen (HLA) family, co-stimulatory molecules, cell adhesion molecules and immunoregulatory factors. Briefly, the cells were washed with phosphate-buffered saline (GmbH, Bergisch, Germany) and incubated for 30 min with the following monoclonal antibodies: CD105-FITC (Ancell Corporation, Bayport, MN, USA), CD229-PE (Imtec Diagnostics, Antwerp, Belgium), CD58-FITC (BD Biosciences Pharmingen, Erembodegen, Belgium), CD106-PC5 (BD Biosciences Pharmingen), CD54-PE (BD Biosciences Pharmingen), CD95-FITC (Miltenyi Biotec, Leiden, the Netherlands), CD274-PE (BD Biosciences Pharmingen), HLA-ABC-PC5 and HLA-DR-PC5 (BD Biosciences Pharmingen), HLA-G-PE (EXBIO Praha, Vestec, Czech Republic), CD4-PE (Miltenyi Biotec), CD8-FITC (Miltenyi Biotec), CD40-PE (Miltenyi Biotec), CD154-PC5 (Imtec Diagnostics, Antwerp, Belgium), CD200R-PE (Imtec Diagnostics), CD200-PC5 (Imtec Diagnostics), CD70-PE (Imtec Diagnostics), CD27-PerCP (Imtec Diagnostics), CD80-PE (Imtec Diagnostics), CD86-PC5 (Imtec Diagnostics), CD134-FITC (Imtec Diagnostics), CD166 (BD Biosciences Pharmingen), SDF1-APC (R&D Systems, Abingdon, UK), MMP9-FITC (R&D Systems), CD44-FITC (Miltenyi Biotec), CD56-PE (ANALIS, Suarlée, Belgium), CD252-PE (Imtec Diagnostics), CD49c-PE (BD Biosciences Pharmingen), CD49d-PE (BD Biosciences Pharmingen), CD49b-PE (BD Biosciences Pharmingen), CD183-PE (BD Biosciences Pharmingen), CD184-PE (BD Biosciences Pharmingen), HO1-PE (ENZO Life Sciences, Farmingdale, NY, USA), CD138-PE (BD Biosciences Pharmingen), RORγt-APC (R&D Systems), IL-6-PE (ImmunoTools, Friesoythe, Germany), IFN-γ-FITC (Imtec Diagnostics), IL-4-PerCP-Cy5.5 (Imtec Diagnostics) and IL-23R-PE (R&D Systems). After washing with MACSQuant Running Buffer (Miltenyi Biotec), the cells were fixed with 4 % formaldehyde solution. Data were acquired using a MACSQuant Analyzer (Miltenyi Biotec) and analyzed using FCS Express 4 Flow Cytometry software (De Novo Software, Los Angeles, CA, USA).

Preparation of MSC conditioned media

We prepared conditioned media (CM) from MSC cultured alone for 3 days in DMEM without serum. The supernatants were collected and frozen at −20 °C until further use.

Isolation of T cells and T cell/BM-MSCs co-cultures

Peripheral blood samples were collected from HDs after informed consent was obtained. T cells were isolated from human blood by magnetic cell separation using human CD3 MicroBeads (Miltenyi Biotec). The mean percentage of CD3-positive cells was observed using flow cytometry was 95 % (data not shown). (a) Mixed lymphocyte reactions (MLRs) were performed as previously described [14]. Briefly, 10⁵ CD3⁺ T cells were co-cultured with 2 × 10⁴ irradiated allogeneic peripheral blood mononuclear cells (PBMCs) in a final volume of 250 µL in 96-well plates. The MLRs were prepared in triplicates in the presence or absence of 1.2 × 10⁴ irradiated MSCs. After 4 days, 150 µL of culture medium were harvested for enzyme-linked immunosorbent assay (ELISA) analyses. Lymphocyte proliferation was assessed using 5-bromo-2-deoxy-uridine (BrdU) incorporation. On day 4, 50 µM BrdU (Roche Applied Science, Mannheim, Germany) was added to the co-cultures at day 4. T cell proliferation was evaluated using a colorimetric assay to measure BrdU incorporation according to the manufacturer’s instructions. Data were expressed as the % of T cell proliferation. (b) 2.5 × 10⁴ BM-MSCs were co-cultured with 10⁵ CD3⁺ T cells in 1 mL of DMEM supplemented with 2 mM l-glutamine, 50 U/mL of penicillin, 50 µg/mL of streptomycin, 10 % fetal bovine serum, 1 mg/mL of interleukin 2 (IL-2) (R&D Systems) and 1 mg/mL of phytohemagglutinin (PHA) (Remel Europe, Kent, UK). For blockade experiments, we added 2 µg/mL of anti-hIL-6 antibodies (R&D Systems), 2 µg/mL of anti-hCD40 antibodies (R&D Systems) or 5 µg/mL of anti-hVCAM-1 antibodies (R&D Systems). After 5 days, the culture medium was harvested for ELISA analyses and flow cytometry analysis. The Th17/Treg ratio was evaluated by flow cytometry using the Human Th17/Treg phenotyping Kit (BD Biosciences Pharmingen). The CD4/CD8 and Th1/Th2 ratios were evaluated by flow cytometry using the antibodies combinations CD8-FITC/CD4-PE and IFN-γ-FITC/CD4-PE/IL-4-PerCPCy5.5 after cell permeabilization using FIX & PERM^® kit (Invitrogen). Lymphocyte stimulation was assessed after carboxyfluorescein succinimidyl ester (CFSE) labeling using the CellTrace CFSE cell proliferation kit (Invitrogen Molecular Probes, Eugene, OR, USA). CFSE fluorescence was visualized by flow cytometry, and the results were expressed as the percentage of positive T cells.

Cytokine expression

Cytokine concentrations in the culture media from the MLRs were determined using Quantikine ELISAs for IL-6, IL-10, hepatocyte growth factor (HGF) and TGF-β according to the manufacturer’s instructions (R&D Systems). Cytokine concentrations in the culture media from the BM-MSC-/mitogenic-activated T cell co-cultures were determined using ProcartaPlex™ Human Essential Th1/Th2 Cytokine Panel immunoassays for IL-4/IL-5/IL-12, IFNγ and TNF-α according to the manufacturer’s instructions (eBiosciences, Vienna, Austria).

Statistical analysis

The Mann–Whitney test was used to analyze differences between groups. All tests performed in this study were two-sided. The significance level was set at p < 0.05, and the results were expressed as the mean ± SEM (standard error of the mean). All analyses were performed using GraphPad Prism 5 software (GraphPad Software, San Diego, CA, USA).

Results

MM BM-MSCs displayed reduced inhibition of T cell proliferation

We used mixed lymphocyte reaction (MLRs) to evaluate the growth of activated T cells in the presence or absence of BM-MSCs. We observed that the inhibition of T cell proliferation by MM BM-MSCs was absent compared with that of HD BM-MSCs (Fig. 1a). On average, HD BM-MSCs reduced the T cell proliferation to 68 % (n = 5) of the proliferation of the T cells alone (100 %). In contrast, untreated MM BM-MSCs increased this proliferation to 146 % (n = 6) and patient treatments tended to reduce this proliferation to 125 % (n = 12). Similar results were observed when the T cells were activated using mitogenic factors (i.e., PHA/IL2) (data not shown).

Fig. 1.

Reduced inhibition of T cell proliferation by MM BM-MSC. a Percentage of T cell proliferation activated by MLR in co-culture with HD BM-MSC (n = 5), BM-MSC from untreated patients (n = 6) and from treated patients (n = 12) compared to a 100 % T cell proliferation without BM-MSC (MLR). Columns represent the mean ± SEM. b Representation of the expansion of T cells unstimulated, activated (PHA/IL-2), activated and co-cultured with HD BM-MSC and MM BM-MSC

Th17/Treg ratios among activated T cells co-cultured with MM BM-MSCs

We used flow cytometry to analyze the Th17/Treg ratio after 5 days of co-culture between activated CD3⁺ T cells and BM-MSCs (Fig. 2a). We observed a shift in the Th17/Treg balance, with an increased percentage of T cells expressing IL-17A (i.e., Th17 cells) and a reduced percentage of T cells expressing FOXP3 (i.e., Treg cells) when activated T cells were co-cultured with MM BM-MSCs compared to co-cultures with HD BM-MSCs. The Th17 cell expansion was confirmed by an increased percentage of cells expressing the IL-23 receptor (IL-23R) when activated T cells were co-cultured with MM BM-MSCs compared to co-cultures with HD BM-MSCs. A similar increase was observed for the Th17-defining transcriptional factor RORγt, but this difference was not significant (Fig. 2b). We did not observe differences for CD4⁺/CD8⁺ ratio and Th1/Th2 (CD4⁺-IFN-γ⁺/CD4⁺-IL-4⁺) ratio between MM and HD BM-MSCs (data not shown). We also did not observe differences for T cell differentiation when activated T cells were cultured with MM BM-MSC conditioned medium (CM) compared to HD BM-MSC CM (data not shown).

Fig. 2.

Altered Th17/Treg ratio in co-cultures with T cells and MM BM-MSC. a Percentage of activated T cells positive for IL-17 (Th17) and Foxp3 (Treg) after 5 days of co-culture with BM-MSC from healthy donors (n = 5), untreated (n = 8) and treated MM patients (n = 5). b % of activated T cells positive for IL-23R and RORγt after 5 days of co-culture with BM-MSC from healthy donors, untreated and treated MM patients. Columns and bars represent the mean ± SEM

MM BM-MSCs immunogenicity in inflammatory conditions

We used flow cytometry to analyze MM BM-MSCs expression of many factors involved in T cell activation and in inflammatory responses. After 24 h of incubation in the presence or absence of pro-inflammatory medium, the expression of 30 factors was analyzed. The factors for which the expression observed was not different between MM and HD BM-MSCs in any conditions are presented in Supplementary Table 2. In basal medium without inflammation (Fig. 3), the % of positive cells for HLA-DR, CD40 and CD106 was increased in untreated MM BM-MSCs compared with their HD counterparts. Untreated MM BM-MSCs also exhibited more robust expression of HLA-ABC. In inflammatory conditions (Fig. 4), the percentage of cells positive for CD40 and CD154 (CD40L) was increased, while the percentage of cells positive for CD58 and hemeoxygenase-1 (HO-1) was decreased in untreated MM BM-MSCs compared with HD BM-MSCs. Untreated MM BM-MSCs also exhibited less-intense expression of CD54 compared with HD BM-MSCs. Patient treatments tended to reduce the abnormal expression level of CD40, CD106, CD58, CD54, HLA-ABC and HLA-DR, but this trend was not observed for CD154, HO-1 and CD40 in inflammatory conditions.

Fig. 3.

MM BM-MSC immunogenicity in basal medium. a–c Percentage of positive BM-MSC from healthy donors, untreated and treated MM patients for CD40, CD106 and HLA-DR. d Mean fluorescence intensity of HLA-ABC expressed by BM-MSC from healthy donors, untreated and treated MM patients. Bars represent the mean ± SEM. ***p < 0.005, **p < 0.02 and *p < 0.05 compared to healthy donors

Fig. 4.

MM BM-MSC immunogenicity in inflammation medium. a–c Percentage of positive BM-MSC from healthy donors, untreated and treated MM patients for CD40, CD154 and CD58. d Mean fluorescence intensity of CD54 expressed by BM-MSC from healthy donors, untreated and treated MM patients. Bars represent the mean ± SEM. **p < 0.02 and *p < 0.05 compared to healthy donors

MM BM-MSCs cytokine expression in inflammatory conditions

We used ELISAs to measure the concentrations of a variety of immunomodulatory cytokines (i.e., IL-6/10, HGF and TGF-β) in the CM from co-cultures of MM BM-MSCs and activated T lymphocytes (Fig. 5). The analyzed conditioned media were generated using two different methods. First, we analyzed the conditioned media from the MLR experiments. We observed an increased concentration of IL-6 for MLRs that included untreated MM BM-MSCs compared with MLRs that included HD BM-MSCs. In contrast, IL-10 concentrations were reduced in the presence of MM BM-MSCs, and no differences were observed for TGF-β between the two groups (similar results were observed for the second method—data not shown). A tendency to overexpress HGF was observed in MM BM-MSCs, but the observed difference was not significant. Second, we analyzed the CM from a 5-day co-culture of BM-MSCs and T cells activated using PHA/IL-2. We observed no differences in the concentration of TNF-α, IFN-γ, IL-5 and IL-12 between MM and HD BM-MSCs and no production of IL-4 (data not shown). Finally, to dissect the origin of IL-6 overconcentration in the CM from co-cultures between MM BM-MSCs and activated T lymphocytes, we performed intracellular staining using flow cytometry (CD45/IL-6) to distinguish which cells secreted this cytokine. We demonstrated that <5 % (±1.87) of T cells (CD45⁺ cells) are positive for IL-6, while a mean of 58 % (±4.33) of MSC (CD45-cells) are positive for IL-6 (n = 12; p = 0.0001).

Fig. 5.

Cytokine concentration in the medium of MLR co-cultured with BM-MSC. Concentration of IL-6, IL-10, HGF and TGF-β in the medium of MLR co-cultured with and without BM-MSC from healthy donors and untreated MM patients. Bars represent the mean ± SEM

Involvement of IL-6, CD40 and vascular cell adhesion molecule-1 (VCAM-1) in the Th17 population increase

We used blocking antibodies directed against IL-6, CD40 and VCAM-1 in our co-culture experiments to test their participation in the increased Th17 population induced by MM BM-MSCs. The addition of anti-IL-6, anti-CD40 or anti-VCAM-1 antibodies partially reduced the % of Th17 cells when T cells were cultured in the presence of MM BM-MSCs compared to HD BM-MSCs (Fig. 6).

Fig. 6.

Involvement of IL-6, CD40 and VCAM-1 in the Th17 population increase. Percentage of CD4⁺/IL-17⁺ cells from co-cultures between T cells and HD-MSCs (n = 11) or MM-MSCs with or without antibodies directed against IL-6, CD40 and VCAM-1 (n = 12). Bars represent the mean ± SEM

Discussion

MM is unique in its capacity to evade immune surveillance via increased secretion of immunologically active compounds and modulation of immune cells [3]. A variety of immunotherapies for MM, such as novel immunomodulatory drugs (e.g., pomalidomide), infusion of vaccine-primed T cells and administration of dentritic cell/PC fusions, are under investigation. However, although these treatments are specific, they must overcome the microenvironment-mediated drug resistance of MM-PCs. In MM, the relationship between MM-PCs and the BM microenvironment is critical for malignant cell survival and proliferation. Among the MM tumor microenvironment partners, BM-MSCs play a crucial role in pathogenesis through their contacts with MM-PCs. One consequence of these contacts is the constitutive abnormalities observed in MM BM-MSCs, such as a reduced inhibition of T lymphocyte proliferation [8, 9]. In this study, we investigated the mechanisms that underlie this impaired immunoregulatory capacity.

We analyzed the immunogenicity of MM BM-MSCs by evaluating the expression of many factors involved in T cell activation, such as CD80, CD86, CD70 and CD40. We observed an overexpression of CD40, CD154 (CD40L), CD106 (VCAM-1), HLA-DR and HLA-ABC and a reduced expression of CD58 (lymphocyte function-associated antigen-3) (LFA-3), CD54 (intercellular adhesion molecule-1) (ICAM-1) and HO-1 compared with HD BM-MSCs. The CD40/CD40L system plays a major role in immunity and inflammation. Ligation of CD40 triggers the production of numerous chemokines (e.g., MCP-1, MIP-1α and CCL5) and cytokines (e.g., IL-6, IL-1, TNF-α), the upregulation of cell adhesion molecules (e.g., ICAM-1/CD54, VCAM-1/CD106) and the secretion of matrix metalloproteinases (e.g., MMP-1/2/3/9) [17, 18]. Interactions between CD40 and its ligand also contribute to T and B cell activation and proliferation [19]. Contrary to Wallace et al. [20], we observed an increased expression of VCAM-1 (CD106) by MM BM-MSCs compared with HD BM-MSCs. Using fluorescence microscopy, Wallace et al. observed a reduced intensity of VCAM-1 expression in MM BM-MSCs. However, this intensity of expression was evaluated using an arbitrary score in a 98 % positive population, while previous studies have observed a mean of 16–50 % VCAM-1-positive cells in the BM-MSC population [21]. VCAM-1 is an immunoglobulin-like transmembrane adhesion molecule that is involved in the human immune system, particularly in the extravasation of leukocytes through the endothelium to sites of inflammation. VCAM-1 promotes CD3-dependent T cell proliferation via the very late antigen-4 [22], and this function could contribute to the reduced inhibition of T cell proliferation by MM BM-MSCs. We also measured a reduced expression of ICAM-1 (CD54) by MM BM-MSCs in inflammatory conditions. The pro-inflammatory medium increased ICAM-1 expression in all of the studied BM-MSCs, but MM BM-MSCs expressed this adhesion molecule less robustly than HD BM-MSCs. ICAM-1 is involved in the immunomodulation capacities of BM-MSCs [23] and in the inhibition of Th17 cell differentiation [11]. Combined with the reduced expression of the Treg co-inducer LFA-3 (CD58) [24], the reduced ICAM-1 expression observed in MM BM-MSCs could be involved in the observed changes in the Th17/Treg balance when T cells were co-cultured with MM BM-MSCs compared to co-cultures with HD BM-MSCs. HO-1-reduced expression by MM BM-MSCs could also be related to Th17/Treg unbalance because HO-1 is involved in the BM-MSC immunosuppressive capacities [25]. Indeed, HO-1 promotes Treg and inhibits Th17 formation [26, 27]. Finally, the increased expression of HLA-ABC and HLA-DR is another indicator of the shift in MM BM-MSC immunogenicity toward a T cell-stimulating phenotype. Several groups have demonstrated that IFN-γ-mediated inflammation increases HLA-DR expression in human MSCs [28]. At the opposite, IL-10 could decrease the induced HLA expression, revealing their immunosuppressive capacity in vitro.

We then measured the secretion of a variety of immunomodulatory cytokines in the CM from co-cultures of MM BM-MSCs and activated T lymphocytes. The observation of an increased concentration of IL-6 associated with a reduced concentration of IL-10 is consistent with the Th17/Treg ratio shift and the reduced inhibition of T cell proliferation. IL-6 increases T cell proliferation, induces the differentiation of Th17 cells and inhibits the differentiation of Treg cells [29]. In contrast, IL-10 reduces T cell proliferation and induces the differentiation of Treg cells [30, 31]. To confirm the implication of these altered factors in the abnormal immunomodulation capacities of MM BM-MSCs, we blocked the activity of IL-6, CD40 and VCAM-1 by directed antibodies in co-cultures between BM-MSCs and activated T cells. We demonstrated that these three factors are involved in the increased Th17 population when the activated T cells were co-cultured with MM BM-MSCs, in comparison with HD BM-MSCs. The observed absence of IL-4 secretion and the lack of differences for IL-5, IL-12 and IFN-γ between MM and HD BM-MSCs further indicate that the alterations of MM BM-MSC immunomodulation capacities impact only Th17/Treg cells and not Th1/Th2 populations. Indeed, IFN-γ and IL-12 are the principal cytokines responsible for the differentiation of Th1 cells, IFN-γ can negatively regulate Th17 cell differentiation and IL-4 is one the principal cytokines responsible for the differentiation of Th2 cells [32, 33]. In addition, IL-5 and IL-4 are extensively secreted by Th2 cells [32].

Collectively, our observations demonstrated that the altered immunomodulation capacities of MM BM-MSCs were related to variations in their phenotype and secretion profile. These alterations lead not only to a reduced inhibition of T cell proliferation but also to a shift in the Th17/Treg ratio. We identified several factors that are potentially responsible for these alterations, such as IL-6, VCAM-1, HO-1 and CD40/40L. Furthermore, these alterations seem rather to be linked to MM cell impact than to the entry in senescence or the excessive passage by MM BM-MSCs compared to HD BM-MSCs. Indeed, the reduced inhibition of T cell proliferation and shift in the Th17/Treg ratio are in conflict to what is observed in senescent or old age BM-MSCs [34, 35].

These variations could impact MM pathogenesis and progression via diverse mechanisms. First, the overexpression of CD40 and its ligand could contribute to the overexpression of many factors involved in disease development and progression, such as IL-6, VCAM-1 (CD106), MMPs, MCP-1 and MIP-1α. The increased expression of these factors by MM BM-MSCs was demonstrated in a previous study by our group [14]. In addition, as demonstrated in studies of inflammatory bowel disease [36], activated T cells, which are observed in MM [37], express CD40L. This expression would allow T cells to bind to local mesenchymal CD40⁺ cells and activate p38 and NF-κB, leading to further cytokine and chemokine production and to the upregulation of cell adhesion molecules. CD40 activation in MM-PCs also modulates the adhesion of these cells to BM-MSCs and induces diverse downstream pathways (e.g., JAK/STAT, p38, JNK/SAPK), favoring the survival and drug resistance of MM-PCs [38, 39]. Recent studies demonstrate a role of the CD40 pathway in angiogenesis [40]. Second, the overexpression of VCAM-1 by MM BM-MSCs could impact MM bone disease because stromal expression of VCAM-1 is involved in and necessary for osteoclastogenesis [41]. VCAM-1 is also involved in cancer immune evasion, potentially by promoting the migration of CD8⁺ T cells [42]. Finally, VCAM-1 mediates adhesion between MM-PCs and MM BM-MSCs, leading to cell adhesion-mediated drug resistance (CAM-DR) [43]. Third, the increased capacity of MM BM-MSCs to induce a shift of CD4⁺ activated T cells toward a Th17 phenotype could be connected to the observed increased Th17 population in MM patients [7, 44]. This increased Th17 population is correlated with disease progression and related to lytic bone disease [45, 46].

Patient treatments, which primarily employed immunomodulator drugs (17/22 treated patients), tended to reduce the differences between MM and HD BM-MSCs. Immunomodulator drugs can inhibit the expression of adhesion molecules such as VCAM-1 [47] and the production of immune mediators such as IL-10, TNF-α and vascular endothelial growth factor [47, 48] and reduce the Th17 population [49]. However, this reduction is incomplete and therefore requires further investigation to elucidate its mechanism of action. In conclusion, this study has identified novel constitutive alterations in MM BM-MSCs that could potentially contribute to MM disease by modulating the immune compartment. These alterations constitute new targets for treatments that aim to interfere with the relationship between MM-PCs and their microenvironment.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

BM

Bone marrow

BrdU

5-Bromo-2-deoxy-uridine

CCL5

Chemokine (C–C motif) ligand 5

CFSE

Carboxyfluorescein succinimidyl ester

CM

Conditioned medium

ELISA

Enzyme-linked immunosorbent assay

FOXP3

Forkhead box P3

HD

Healthy donors

HGF

Hepatocyte growth factor

HLA

Human leukocyte antigen

HO-1

Hemeoxygenase-1

ICAM-1

Intercellular adhesion molecule-1

IFN

Interferon

IL

Interleukin

IL-23R

Interleukin-23 receptor

LFA-3

Lymphocyte function-associated antigen-3

MCP-1

Monocyte chemotactic protein-1

MIP-1α

Macrophage inflammatory protein-1 alpha

MLR

Mixed lymphocyte reactions

MM

Multiple myeloma

MMP

Matrix metalloproteinase

MSC

Mesenchymal stromal cell

PBMC

Peripheral blood mononuclear cells

PCs

Plasma cells

PGE2

Prostaglandin E2

PHA

Phytohemagglutinin

RORγt

RAR-related orphan receptor gamma t

SEM

Standard error of the mean

TGF

Tumor growth factor

Th

T helper cells

TNF

Tumor necrosis factor

Treg

T regulatory cells

VCAM-1

Vascular cell adhesion molecule-1