Myeloid-derived suppressor cells therapy enhance immunoregulatory properties in acute graft versus host disease with combination of regulatory T cells

Background Myeloid-derived suppressor cells (MDSCs) play a critical role in modulating the immune response and promoting immune tolerance in models of autoimmunity and transplantation. Regulatory T cells (Tregs) exert therapeutic potential due to their immunomodulatory properties, which have been demonstrated both in vitro and in clinical trials. Cell-based therapy for acute graft-versus-host disease (aGVHD) may enable induction of donor-specific tolerance in the preclinical setting. Methods We investigated whether the immunoregulatory activity of the combination of MDSCs and Tregs on T cell and B cell subset and alloreactive T cell response. We evaluated the therapeutic effects of combined cell therapy for a murine aGVHD model following MHC-mismatched bone marrow transplantation. We compared histologic analysis from the target tissues of each groups were and immune cell population by flow cytometric analysis. Results We report a novel approach to inducing immune tolerance using a combination of donor-derived MDSCs and Tregs. The combined cell-therapy modulated in vitro the proliferation of alloreactive T cells and the Treg/Th17 balance in mice and human system. Systemic infusion of MDSCs and Tregs ameliorated serverity and inflammation of aGVHD mouse model by reducing the populations of proinflammatory Th1/Th17 cells and the expression of proinflammatory cytokines in target tissue. The combined therapy promoted the differentiation of allogeneic T cells toward Foxp3 + Tregs and IL-10-producing regulatory B cells. The combination treatment control also activated human T and B cell subset. Conclusions Therefore, the combination of MDSCs and Tregs has immunomodulatory activity and induces immune tolerance to prevent of aGVHD severity. This could lead to the development of new clinical approaches to the prevent aGVHD.


Background
Graft-versus-host disease (GVHD) is the major complications after allogeneic hematopoietic stem cell transplantation (Allo-HSCT) [1,2]. During GVHD, allogenic T cells are differentiated into effector lineages and secrete proin ammatory cytokines [3]. Allogeneic T-cell response are suppressed by immunosuppressive drugs, reducing the risk for acute GVHD [4,5]. However, immunosuppressive drug for GVHD result in toxic side-effects. Thus, there is an unmet need for novel treatment strategies for GVHD with less toxicity and fewer side effects.
MDSCs can be differentiated and expanded in vitro by various methods using a variety of combinations of progenitor cells and cytokines [13]. MDSCs generated in vitro suppress autoimmune lupus [14] and prevent pre-clinical GVHD model [15][16][17], indicating therapeutic potential for T-cell-mediated diseases.
CD4+CD25+Foxp3+ Treg-based cellular therapy is effective for recipients of bone marrow and solid organ transplantation [20][21][22] and in patients with autoimmune diseases [23]. Donor-derived Tregs are typically used, as they share an MHC type with CD4+ and CD8+ T cells, which are primarily responsible for GVHD [24]. However, the therapeutic potential of Tregs is limited by their short lifespan and their plasticity under pathological conditions [25][26][27].
MDSC and Treg cell interactions involve a positive feedback signals in which MDSCs expand Treg cells while Treg cells control MDSC function [28]. MDSC support the induction of regulatory B (Breg) cells [14], type of B cell that releases IL-10 and and has immunosuppressive effects [29,30]. Given this background, combined cell therapy using MDSCs and Treg cells may be bene cial for the treatment of aGVHD. We investigated the effects of the combination of MDSCs and Tregs on the induction of tolerance to MHCmismatched transplants. The combination of MDSCs and Tregs reciprocally modulated the populations of alloreactive Th1/Th17 cells and Foxp3+ Treg cells. Furthermore, the combination therapy alleviated aGVHD clinically and histopathologically by regulating the effector T/B cell and Treg/Breg balance.
These ndings indicate that the combination of MDSCs and Tregs shows promise for alleviating aGVHD.

Methods
Mice 8-weeks-old C57BL/6 (B6, H-2b) and BALB/c (H-2d) female mice were purchased from Orient Bio (Sungnam, South Korea). The mice were maintained under speci c-pathogen-free conditions in an animal facility with controlled humidity (55 ± 5%), 12/12 h light/dark cycle, and temperature (22 ± 1°C). Mice were fed mouse chow and tap water ad libitum. All animal experiments were performed in accordance with the animal care and use committee of The Catholic University of Korea approved the protocols
After three days, the induced Treg cells were stained with CD4, CD25 and then sorted using a MoFlo cell sorter to obtain a ~95% pure CD4+CD25+ population.

Bone marrow transplantation
To develop the aGVHD model, Balb/c were lethally irradiated with 700 cGy and infused with 5 × 10 6 donor BM cells plus 5 × 10 6 splenocytes from Balb/c(syngenic) or C57BL/6(donor, allogenic) on day 0. On day1 and day7 after bone marrow transplantation (BMT), recipient mice received MDSCs (1 × 10 6 ) and Tregs (1 × 10 6 ) individually or in combination. The mice were monitored for clinical signs and body weight. The clinical GVHD was scored twice weekly using the clinical GVHD scoring system (Table 1) [31]. Each of the ve clinical parameters summed up to get the nal score at indicated time points.

Histological and immunohistochemical analyses
Mice were sacri ced on day 28 after BMT and organs captured, cryoembedded, and sectioned. Tissue specimens were xed in 10% formalin buffer and embedded in para n. Sections (6 mm thick) were stained with H&E and the histologic score was determined using established scoring systems [31,32].

Enzyme-linked immunosorbent assay
The concentrations of IL-17 and IFN-g in culture supernatants were measured in duplicate using a sandwich enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's protocol (DuoSet; R&D Systems, Lille, France).

Statistical analyses
Data are expressed as the means ± standard error of the mean (SEM). One-way analysis of variance followed by Bonferroni's post hoc test was used to compare the differences between three or more groups. Statistical signi cance was considered as P-value < 0.05. All statistical analysis was performed with Prism (standard version 5.01; GraphPad Software, San Diego, CA).

Results
The combination of MDSCs and Tregs controls the T cell and B cell response To evaluate the effects of the combination of MDSCs and Tregs on T-and B-cell subsets, we cultured CD4+T cells from C57BL/6 spleen with MDSCs and Tregs individually or in combination for 3 days in vitro. The Th1 and Th17 cell populations were signi cantly reduced, and the Foxp3+Treg population was signi cantly increased, by the combination treatment compared to either cell type alone ( Figure 1A).
C57BL/6 splenocytes were incubated with LPS for 3 days in the presence of MDSCs and Tregs individually or in combination. The combination treatment markedly increased the population of IL-10producing Bregs ( Figure 1B) and signi cantly reduced the plasma cell population compared to the control and MDSCS or Tregs alone ( Figure 1B).
To determine whether combination treatement of MDSC and Treg has a inhibitory effect on proliferation of alloreactive T cells, we performed in vitro alloreactive proliferation assay. Alloreactive CD4 + T cells had proliferated vigorously to allogenic stimulation. MDSCs or Tregs alone suppressed the proliferation of alloreactive T cells. The combination more potently diminished the proliferation of alloreactive T cells ( Figure 2A). Under alloreactive T cell-activation conditions, elevated interferon (IFN)-g and IL-17 levels were markedly reduced by the combination treatment, compared to treatment with either cell type alone ( Figure 2B).
Next, we assessed the effects of combined therapy with MDSCs and Tregs on alloreactive T cell subsets.
Combined therapy increased the population of CD4 + Foxp + Treg cells approximately threefold ( Figure 2C) but reduced the population of effector Th1 and Th17 cells ( Figure 2C).

The combination of MDSCs and Tregs ameliorates acute GVHD severity and in ammation
To evaluate whether combined cell-therapy with MDSCs and Treg could therapeutic potential for aGVHD model. We used a fully MHC-mismatched (BALB/c (H-2k d )→ C57BL/6 (H-2k b ) murine models ( Figure 3A).
There were no differences in weight loss among the groups, while the animals treated with the combination had lower clinical scores than the control and MSC-or Treg-alone treated animals ( Figure  3B). The skin, liver, lung, and intestine are the primary targets of aGVHD. According to histopathologic analyses of the intestine, skin, liver and lung, the extent of tissue damage, in ammation, and lymphocyte in ltration was signi cantly reduced by the combined cell therapy compared to those of control mice and MDSCs or Tregs alone treated mice ( Figure 3C-E).
Proin ammatory cytokines, including IL-6, IL-17 and TNF-a, are key mediators of injury to target organs.
To determine whether combined therapy with MDSCs and Tregs modulate the production of proin ammatory cytokines in target tissues, we performed immunohistological staining of IL-17, IL-6, TNF-a, and Foxp3 in the skin, liver, lung, and small intestine. As shown in Figure 4, the expression level of IL-17, IL-6, and TNF-a was lower, while the expression of Foxp3+ cells was much higher, in sections of skin, liver, lung, and intestine tissue by the combined cell therapy compared to the control and to each cell type individually.

The combination of MDSCs and Tregs modulates T-and B-cell populations in vivo
To determine the in vivo mechanism of action of the combination cell therapy during the development of aGVHD, we analyzed the T helper subsets by ow cytometry. In peripheral blood, the Th1 and Th17 cells were signi cantily reduced by the combination of MDSCs and Tregs ( Figure S1). The population of Th2 cells slightly expanded and did not differ among the groups. Interestingly, the Foxp3+Treg population was increased by the combined cell therapy compared to the control and to treatment with MDSCs or Tregs alone ( Figure S1). Furthermore, the frequency of Th1 and Th17 cell in the spleen were signi cantly decreased by the combination therapy with MDSCs and the populations of Tregs. In contrast, the proportion of Foxp3+ Tregs ( Figure 5A) and IL-10+ Bregs were markedly increased ( Figure 5B).
To explore the regulatory effects of the combined cell treatment on B-cell subsets in vivo, we analyzed the populations of germinal-center B cells, plasma cells, and IL-10 producing Bregs by ow cytometry. The combination signi cantly decreased the populations of germinal-center B cells and plasma cells. Interestingly, the population of IL-10+ Bregs was increased by the combined cell therapy compared to the control and treatment with MDSCs or Tregs alone ( Figure 5B).

Effects of combined therapy with MDSCs and Tregs on human T-and B-cell subsets
To determine the effects of combined therapy on human T-cell subsets, we analyzed the populations of effector T cell subsets by ow cytometry. As shown in Figure6A, in vitro coculture with MDSCs and Tregs reduced the populations of Th1 and Th17 cells compared to coculture with either of those cell types individually. By contrast, the population of Foxp3+Tregs was increased by the combined cell therapy compared to the control and treatment with MDSCs or Tregs alone.
Furthermore, the proportion of IL-10-producing B cells was increased by the combination cell therapy compared to the control and treatment with MDSCs or Tregs alone ( Figure 6B). However, the population of CD138+ plasma cells did not differ among the groups.
Discussion GVHD may develop after allogeneic HSCT is used to treat hematologic malignancies. The standard therapy for aGVHD is steroids and calcineurin inhibitors, but the long-term use of such nonspeci c immunosuppressive drugs causes severe side effects [5]. Thus, more effective treatment strategies for acute GVHD urgently needed.
MDSCs are potent immunosuppressive cells in various pathologic settings. They can inhibit T-cell responses both in vitro and in vivo by producing factors such as arginase 1, iNOS, reactive oxygen species, and TGF-b, which inactivate a variety of immune cell types, particularly T cells [6]. MDSC induce expansion of Tregs [28,33]. In addition, MDSCs promote the expansion of Breg cells (IL-10-producing B cells) via an iNOS-dependent pathway and ameliorate autoimmune disorders [14].
In recent studies, MDSCs have potential for the treatment of GVHD [16,17,34]. However, our data show that MDSCs alone had low clinical e cacy. High ll et al. showed that an IL-13-producing subset of MDSCs had a greater suppressive effect than MDSCs [15]. MDSC show promise as potential cell therapy for the treatment GVHD, however the regulatory role of myeloid cells is poorly understood and still lack of a clinical research for the application.
Tregs can control GVHD in mice [35,36] and humans [37,38] by inhibiting the alloreactive T-cell response. They have shown positive results in preclinical trials and their ability to ameliorate GVHD is under investigation. However, this approach requires standardization of Treg expansion methods and doses [39,40]. Inducible Tregs can readily be generated, but controversial preclinical ndings and phenotype instability have hampered their translation to the clinic [18].
In previous study, for the rst time that murine MDSCs induced expansion Breg cells (IL-10-producing B cells) have therapeutic effect in an animal model of SLE [14]. Bregs have signi cant immunosuppressive abilities both in vitro and in vivo. Bregs prevent the onset of GVHD by inhibiting the differentiation of Th1 and Th17 cells and promoting the expansion of Tregs [29,41,42]. Therefore, MDSC-mediated expansion of Bregs is important for the treatment of aGVHD. Our data con rm that our combination therapy with MDSCs Tregs has potential for ameliorating GVHD.
We rst analyzed the regulatory effect of T and B cell subset by combination therapy of MDSC and Treg. The combination of MDSCs and Tregs induced Treg skewing and Th1/Th17 suppression, increased the frequency of IL-10-producing Bregs, and reduced the frequency of plasma cells. It also controlled the proliferation of alloreactive T cells and the Treg/Th17 balance in vitro. These data suggest that the combination of MDSCs and Tregs has potent immunoregulatory effects in mice and humans.
Systemic combined treatment of MDSCs and Treg cells improved clinical GVHD severity, effectively modulating the T helper cell response, reduced the percentage of alloreactive of Th1, and Th17 cells, and increased the frequencies of Th2 and Foxp3+ Treg cells. In vivo, the combination decreased the population of germinal center B cells and plasma cells and increased that of IL-10-producing Bregs. Thus, the combination cell therapy directly modulated the reciprocal regulation of CD4+T-cell and B-cell subsets. Moreover, the expression of proin ammatory cytokine (IL-6, IL-17 and TNF-α) decreased, and that of Foxp3 increased, in the GVHD target tissues (skin, liver, lung, and intestine) from mice treated with MDSC and Treg, suggesting that the in vivo suppressive activity of the combination therapy is attributable in part to soluble mediators. Future studies should investigate the immunoregulatory mechanisms and the interactions of the combination of MDSCs and Tregs for their therapeutic application.
Previous reports suggest that the combination of MSCs and Treg cell therapy have enhanced therapeutic e cacy in transplantation model [43][44][45]. However, the mechanism underlying cross talk between MSC and Treg remains unclear [46]. MSC therapy have also been proposed as a treatment for steroid-refractory GVHD but the results are still controversial [47,48]. Furthermore the origin, delivery route, dose, and timing of infused MSCs in uences therapeutic e cacy. MSC therapy is also insu cient to inhibit production of pro-in ammatory cytokines and reversed MSC-mediated therapetic e cacy under in ammatory condition [49,50]. It was reported that procoagulant tissue factor expression of MSCs by intravascular infusion trigger the instant blood-mediated in ammatory reaction (IBMIR), which in turn activate coagulation. These various problems in MSC therapy impede clinical applications [51]. We therefore suggest that MDSCs plus Treg combination therapy would be an alternative new treatment against GVHD that overcomes the limitations of MSC based therapy. Human MDSC therapy have been di cult to clinical use due to the numerical limitations of MDSC. However recent studies show development of large scale expanded MDSC provides the possibility of being used in clinical studies [17] Our study has several limitations. First, we used only female recipient and donor mice. Sex chromosomelinked minor histocompatibility affect the incidence and severity of GHVD model. Gender-related factors signi cantly affects HSCT outcome [52,53]. Further studies will be needed to con rm the therapeutic effect by sexual differentiation between recipient and donor in our model. Second, we could not prove the GVL effect by combination therapy of MDSC and Treg. Successful allogeneic HSCT are necessary for optimize the balance between GVHD and GVL in allogeneic stem cell transplantation. MDSCs ameliorated GVHD severity and mortality while preserving GVL activity [54]. Alloreactive iTregs can attenuate GVHD, yet severely compromise GVL function [55]. Further studies are needed to con rm the GVL effect by the combination of MDSCs and Tregs. Despite these limitations, this was the rst study to evaluate the e cacy of combination therapy of MDSC and Treg in aGVHD model. Although previous studies have reported inhibitory effects by each MDSC and Treg single-treatment on GVHD, proof of synergy effect of GVHD treatment by combination therapy of MDSC and Treg in this study would provide potential strategies for alleviating GVHD by overcoming limitations on single cell therapy in the future.

Conclusion
In conclusion, the combination of MDSCs and Tregs had a synergistic immunoregulatory effect by increasing the Treg/Breg populations and decreasing the populations of effector Th1 and Th17 cells, which ameliorated GVHD development. Taken together, this study demonstrates that combined therapy of MDSCs and Treg cells with fewer side effects than current immunosuppressant drugs has potential for treating aGVHD and should be evaluated in clinical trials.