IL-21 accelerates xenogeneic graft-versus-host disease correlated with increased B-cell proliferation

Abstract

Graft-versus-host disease (GVHD) is a prevalent and potential complication of hematopoietic stem cell transplantation. An animal model, xenogeneic GVHD (X-GVHD), that mimics accurately the clinical presentation of GVHD would provide a tool for investigating the mechanism involved in disease pathogenesis. Murine models indicated that inhibiting IL-21 signaling was a good therapy to reduce GVHD by impairing T cell functions. We sought to investigate the effect of exogenous human IL-21 on the process of X-GVHD. In this study, human IL-21 was expressed by hydrodynamic gene delivery in BALB/c-Rag2^−/− IL-2Rγc^−/− (BRG) immunodeficient mice which were intravenously transplanted human peripheral blood mononuclear cells (hPBMCs). We found that human IL-21 exacerbated X-GVHD and resulted in rapid fatality. As early as 6 days after hPBMCs transplanted to BRG mice, a marked expansion of human CD19⁺ B cells, but not T cells, was observed in spleen of IL-21-treated mice. Compared with control group, IL-21 induced robust immunoglobulin secretion, which was accompanied by increased accumulation of CD19⁺ CD38^high plasma cells in spleen. In addition, we demonstrated that B-cell depletion was able to ameliorate X-GVHD. These results are the first to find in vivo expansion and differentiation of human B cells in response to IL-21, and reveal a correlation between the expansion of B cells and the exacerbation of xenogeneic GVHD. Our findings show evidence of the involvement of B cells in X-GVHD and may have implications in the treatment of the disease.

Graphical abstract

Keywords

IL-21 / B cell / xenogeneic GVHD / immunodeficient mice / immunoglobulin

INTRODUCTION

Allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative option for a variety of malignant hematological diseases (Welniak et al., 2007). Although HSCT has been successfully used for treatment of these diseases, but it is not without risk. The major complication following HSCT is the development of graft-versus-host disease (GVHD), sometimes with a fatal outcome (Shlomchik, 2007). An animal model, xenogeneic GVHD (X-GVHD), that mimics accurately the clinical GVHD, would help to develop therapies and provide a tool for investigating the mechanism involved in disease pathogenesis. Neither CB17-scid nor NOD-scid immunodeficent mice engrafted with human PBMCs provided a reproducible model of xenogeneic GVHD (Shultz et al., 2007). BALB/c- Rag2^−/− IL-2Rγc^−/− (BRG) mice, which have no B, T, or NK cells and no leakiness, showed improved human PBMCs engraftment compared with NOD-scid mice (van Rijn et al., 2003; Shultz et al., 2007; Ito et al., 2009; Shultz et al., 2011). Recently, BRG mice which transferred with hPBMCs (hu-PBMC BRG mice) were used as a suitable platform to explore the effect of novel GVHD therapy agents, such as immunoglobulin or regulatory T cells (Tregs) (Mutis et al., 2006; Gregoire-Gauthier et al., 2012).

T cells have been identified as a key player in the graft-versus-host reaction and both donor CD4 and CD8 T cells have crucial roles in the pathogenesis of GVHD (Shlomchik, 2007; Blazar et al., 2012). Because of their ability to differentiate into various effector T cell subsets—Th1 cells, Th2 cells, and Th17 cells, CD4 T cells were considered as potential targets for treatment of GVHD in the clinic (Blazar et al., 2012). Recently many approaches, such as T cell depletion (anti-thymocyte globulin, anti-CD7 antibody, etc.), T cell activation inhibition (PKC inhibitors and antibodies specific for CD2, CD3 or CD147) and induction of Treg cells (Bortezomib, low-dose IL-2 or HDAC inhibitors) have been used in clinical trials for the prevention or treatment of GVHD (van Oosterhout et al., 2000; Lopez et al., 2006; Evenou et al., 2009; Blazar et al., 2012). However, these approaches usually leave patients at risk of complications such as infection or cancer relapse. Recent studies showed an association between high numbers of donor B cells and the development of both acute and chronic GVHD (Shimabukuro-Vornhagen et al., 2009). And treatment with rituximab—a CD20-specific monoclonal antibody that depletes B cells before HSCT, reduced GVHD incidence (Christopeit et al., 2009). In addition, IVIG (intravenous immunoglobulin) was reported to prevent xenogeneic GVHD by inhibiting the reconstitution of B cells, but not T cells (Gregoire-Gauthier et al., 2012). Therefore, specific approaches which targeted to B cells might be beneficial for the treatment of GVHD.

IL-21 is produced by CD4⁺ T cells and NKT cells, which could promote the activation, differentiation, maturation of NK cells, B cells, CD8⁺ and CD4⁺ T cells, dendritic cells, and macrophages (Spolski and Leonard, 2008). Previous reports showed that blocking IL-21 signaling, either by genetically depletion or via neutralizing mAbs, reduced disease severity in murine models of GVHD (Bucher et al., 2009; Meguro et al., 2010; Hanash et al., 2011; Meguro et al., 2011). Similar results were obtained in xenogeneic GVHD model by injecting a neutralizing anti-human IL-21 mAb (Hippen et al., 2012). Disease amelioration was related with a decrease in the numbers of IFN-γ-secreting CD4⁺ T cells and an increase in the proportion of CD4⁺ Tregs. Using Treg-depleted donor cells, Bucher et al demonstrated that the increased frequency of Tregs in anti-IL-21 mAb treating group was due to the conversion of CD4⁺CD25^- T cells into inducible regulatory T cells (Bucher et al., 2009). However, in xenogeneic GVHD model the effect of IL-21 on B cells has not been reported. Therefore, we wondered the effect of IL-21 on human B cells in xenogeneic GVHD.

In this work, we treated Rag2^−/−IL-2Rγc^−/− immunodeficieny mice with hPBMCs intravenously to establish the X-GVHD model, and human IL-21 was systemically expressed using a hydrodynamics-based gene delivery technique. We found that administration of IL-21 plasmid DNA resulted in high levels of circulating IL-21 in vivo, which exacerbated the process of X-GVHD. Surprisingly, B cells in IL-21 group augmented on day 7, while there was no increase in the quantity of T cells in spleen. Furthermore, compared with control group, IL-21 induced higher hIgM and hIgG, and increased the accumulation of CD19⁺ CD38^high plasma cells in spleen. Finally, we demonstrated that B-cell depletion was able to abolish X-GVHD. Collectively, we found that IL-21 treatment could promote B cell expansion, plasma cell differentiation, and immunoglobulin secretion, which were correlated with the accelerated X-GVHD progression. Our finding may provide an outlook on the therapeutic prospect for GVHD.

RESULTS

Administration of hIL-21 plasmid resulted in high levels of expression in vivo

A full-length hIL-21 gene was amplified by RT-PCR from human PBMCs and then ligated into PTT3-Flag plasmid. We determined the kinetic expression level of hIL-21 in BRG mouse serum after intravenous hydrodynamic injection of PTT3-hIL-21-Flag plasmid DNA into the tail vein. As shown in Fig. 1A, one day after a single dose of 20 μg of PTT3-hIL-21-Flag plasmid, a high level of IL-21 was detected in BRG mouse serum (about 40,000 pg/mL) by a sandwich double antibody ELISA. Serum levels of hIL-21 decreased quickly but were still higher than 500 pg/mL on day 5 and returned to baseline on day 7. No detectable hIL-21 was found in serum from naïve mice and mice that received injections of the same amount of control plasmid DNA. Importantly, we did not observe obvious toxicity such as weight loss and capillary leaking, caused by the in vivo expression of hIL-21 plasmid DNA (data not shown).

Human IL-21 exacerbated the process of xenogeneic GVHD

To determine whether IL-21 could enhance GVHD, we chose an X-GVHD model using BRG mice. We expressed hIL-21 protein in BRG mice by hydrodynamic injection. 1 day after injection of IL-21 or control plasmid, the mice were sub-lethally irradiated and then injected with hPBMCs on the same day. The kinetic expression of IL-21 in hu-PBMC BRG mice was similar to that in naïve BRG mice (data not shown). Consistent with earlier reports, morbid mice in two groups developed hunched posture, inactivity, and labored breathing. As shown in Fig. 1B and 1C, hu-PBMC BRG mice in IL-21 group exhibited accelerated weight loss and significantly faster disease development compared with control group (median survival time, 14 days versus 31.5 days, P < 0.01). And BRG mice that were injected hIL-21 plasmid without human PBMCs displayed 100% survival (Fig. 1B). These results indicated that hIL-21 exerted an effect on hPBMCs to exacerbate the process of X-GVHD.

Human IL-21 induced B cell engraftment and robust secretion of immunoglobulin

Next, we wonder the effect of IL-21 on human immune cell sub-populations in vivo. Flow cytometric analysis of hu-PBMC-BRG mouse splenocytes was performed after plasmid administration. Because serum hIL-21 can sustain as long as 7 days, hu-PBMC-BRG mice were sacrificed on day 7. In agreement with previous studies (van Rijn et al., 2003), the majority of engrafted cells in BRG mice were CD3⁺ T cells and CD19⁺ B cells (>90%, Fig. 2A). Surprisingly, the percentage of CD19⁺ B cells in spleen significantly increased in IL-21-treated group compared with control group (23.5 ± 2.5 versus 5.1 ± 0.5, P < 0.001), but the percentage of CD3⁺ T cells decreased in IL-21-treated group (67.8 ± 3.2 versus 87.3 ± 1.8, P < 0.001, Fig. 2A). Because the total number of human lymphocytes increased after IL-21 plasmid administration (Fig. 2B), the absolute number of B cells and T cells was even more profound than the percentage. As shown in Fig. 2B, the absolute number of human B cells was still remarkably higher in IL-21 group than that in control group. However, there was no difference in the absolute number of human T cells between mice receiving IL-21 plasmid and control plasmid. Besides cell number, we further analyzed the cytokine producing ability of human T cells involved in X-GVHD. Human Th1 and Th2 cytokines were measured in serum of mice on day 7 (Fig. 2C). Differences in human IFN-γ, TNF-α, IL-10, IL-5, and IL-4 levels were not observed between IL-21-treated group and control group. Collectively, these observations suggested that the expression of hIL-21 in vivo had no obvious biological effects on T cells during the early stage of X-GVHD.

Because autoantibodies play a major role in many kinds of autoimmune disorders, such as SLE (Sherer et al., 2004), rheumatoid arthritis (Nielen et al., 2004), and so on, we hypothesized that the expanded B cells in IL-21-treated mice might produce more antibodies which were specific to recipient tissues. The concentrations of human IgG and IgM in serum were determined by ELISA, and we found that the levels of both kinds of immunoglobulin in IL-21-treated mice were about 20–30 times higher than those in control group (IgG, 44.31 ± 9.3 versus 1.9 ± 0.6 μg/mL; IgM, 59.6 ± 13.6 versus 1.6 ± 0.6 μg/mL, respectively, Fig. 2D). Taken together, our results indicated that human IL-21 could enhance B cell engraftment and immunoglobulin secretion in X-GVHD.

Human IL-21 promoted B cell proliferation and immunoglobulin secretion during the early stage of X-GVHD

In order to determine whether IL-21 promote B cell proliferation or survival after transfer, we further detected the kinetics of human B cells in hu-PBMC-BRG mice treated with hIL-21 plasmid. Human B cells in spleen were evaluated every two days after injection of IL-21 or control plasmid. As shown in Fig. 3A and 3B, there was no difference in CD19⁺ B cells between IL-21 group and control group on day 2 and day 4, but the percentage and absolute number of B cells were significantly higher in IL-21 group than those in control group on day 6. Since the absolute number of B cells decreased quickly in both groups from day 2 to day 4, IL-21 may have no effect on B cell survival. And after day 4, B cells in both groups began to proliferate, and IL-21 showed obvious promotive effect on B cells proliferation. We also detected the human immunoglobulin levels in hu-PBMC-BRG mice, which reflected the state of activated B cells. Before day 4, the IgG and IgM in both control group and IL-21 group were at a very low level (Fig. 3C and 3D). It is notable that large amounts of IgG and IgM were detected in IL-21 group as early as day 6, and the levels of immunoglobulin only slightly raised in control mice. These results suggested that IL-21 could promote B cell proliferation but not survival after PBMC transfer, and IL-21 might contributed to the differentiation of B cells.

IL-21 enhanced the differentiation of human B cells into plasma cells in vivo

It is well accepted that the frequency of plasma cells generally is correlated with the amount of Ig production in vivo (Ettinger et al., 2005), so the capacity of IL-21 to induce plasma cells differentiation was explored. 7 days after plasmid administration, flow cytometric analysis of hu-PBMC-BRG mouse splenocytes was performed. CD19⁺ CD45⁺ cells were gated and assessed for the expression of CD38. As shown in Fig. 4A, only a small fraction (5.4%) of freshly isolated peripheral B cells were CD19⁺ CD38^high plasma cells, which agreed with previous reports (Ettinger et al., 2008). After 6 days in vivo, the percentage of CD19⁺ CD38^high plasma cells was significantly higher in IL-21 treated group than that in control group (79.8 ± 1.6 versus 61.3 ± 4.7, Fig. 4B). And the absolute number of plasma cells was also higher in IL-21 group, compared with control group (Fig. 4C). Taken together, B cells could differentiate into plasma cells in the model of X-GVHD, and IL-21 could accelerate the process of plasma cell differentiation.

B cells were important to the process of X-GVHD

Given the significant expansion and differentiation of human B cells after IL-21 treatment, we next wondered the role of B cells in X-GVHD. We intravenously injected pre-irradiated BRG mice with PBMCs or non-B cell PBMCs which were isolated from PBMCs (Fig. 5A). Following injection, we assessed the development of X-GVHD by recording body weight every 3 days. As shown in Fig. 5B and 5C, BRG mice transferred with non-B cell PBMCs did not exhibit weight loss until day 25 and survived long-term without evidence of GVHD, whereas mice transferred with PBMCs developed lethal X-GVHD. These results demonstrated that B-cell depletion reduced the incidence of X-GVHD.

DISCUSSION

Since the physiological effect of IL-21 on B cells in vivo has not been well described, in this study, we expressed human IL-21 by hydrodynamic injection in an X-GVHD model using hu-PBMC BRG mice. We found that human IL-21 expression exacerbated X-GVHD and resulted in rapid fatality. In the early stage of X-GVHD, a marked expansion of human CD19⁺ B cells, but not T cells, was observed in IL-21 treated mice. The number of CD19⁺ CD38^high plasma cells also increased, accompanied by robust IgM and IgG production after IL-21 treatment. Our results firstly revealed the in vivo effects of IL-21 on B cells in X-GVHD model.

Previous studies showed that T cells played a major role in the progression of GVHD (Shlomchik, 2007), and IL-21 was demonstrated to promote the proliferation and cytokine secretion of both CD4 and CD8 cells (Leonard and Spolski, 2005). In xenogeneic GVHD model, neutralizing anti-IL-21 mAb could attenuate GVHD by inhibiting CD4 T cell function (Hippen et al., 2012). In our model, when hIL-21 was highly expressed in the early stage of X-GVHD, the number of human T cells decreased during the first 2–3 days. Then the number of T cells began to rise, which was similar to the kinetics of B cells (Fig. 3B). But the T-cell dynamic changes were nearly the same between IL-21 and control groups during the early stage of GVHD (data not shown). Besides cell number, we also analyzed the levels of cytokines which could reflect T cell functions in mice serum. And we found no significant difference in human IFN-γ, TNF-α, IL-10, IL-5, and IL-4 levels on day 7. Taken together, in the early phase of our X-GVHD model, the effect of IL-21 on T cells was not obvious. IL-21 has pleiotropic effects on NK cells, such as enhanced proliferation and increased cytotoxicity(Spolski and Leonard, 2008). Previous report showed that NK cells were very low in all compartments during the process of xenogeneic GVHD (van Rijn et al., 2003). And in our xenogeneic GVHD model, CD56⁺ CD3^- NK cells were almost absent in spleen of IL-21-treated group and control group on day 7. Therefore, these results indicated that IL-21 showed little help to NK cells in the early stage of xenogeneic GVHD. In addition, IL-21 is also a critical regulator of Th17 development (Spolski and Leonard, 2008). Because of the low number of human T cells in spleen on day 7, it is difficult for us to identify Th17 cells by their ability to express IL-17 after ex vivo stimulation. If possible, we will detect Th17 cells at the late stage of X-GVHD in the future.

The impact of IL-21 on human B-cell responsiveness was addressed in vitro, and all of the experiments were carried out, in which human B cells were isolated and stimulated with IL-21 in combination with anti-CD40 and/or anti-IgM, which mimic the BCR-antigen and B cell-T cell interaction (Ettinger et al., 2008; Konforte et al., 2009). Notably, IL-21 was found to induce proliferation and differentiation of human B cells isolated from PBMCs and spleen (Ettinger et al., 2005; Good et al., 2006). In our xenogeneic GVHD model, B cells were activated and proliferated after engaging mouse tissues with the help of human T cells. The number of human B cells and plasma cells also increased in hIL-21 group compared with control group. These results suggested that IL-21 treatment could promote B cell expansion and plasma cell differentiation in vivo, which was consistent with the studies in vitro. In addition, as early as 5 days after human PBMCs injection, large amounts of hIgG and hIgM could be assayed in IL-21 group, consistent with the in vitro studies that IL-21 efficiently induced Ig production from purified B cells stimulated with anti-CD40 and/or anti-IgM. Therefore, the physiological function of IL-21 on B cells in vivo is similar to that on isolated B cells stimulated in vitro.

Yoshizaki A et al. found that the severity of EAE increased when transferring Il21r^−/− CD1d^hi CD5⁺ B cells. They demonstrated that ex vivo IL-21 drove the development and expansion of CD1d^hi CD5⁺ B-cell, which produced IL-10 to inhibit disease symptoms when transferred into mice with established autoimmune disease (Yoshizaki et al., 2012). In our xenogeneic GVHD model, we found robust B cell proliferation and differentiation in IL-21-treated group, but we did not detect such B10 cells because of the very low number of human B cells. If B10 cells had been induced by exogenous IL-21, the T cell immune response should be inhibited. However, our results showed that the process of xenogeneic GVHD was accelerated after IL-21 treatment (Fig. 1B and 1C). In addition, based on our cytokine secretion results, the levels of IL-10 and IFN-γ in serum were similar between IL-21-treated group and control group, which indicated that the IL-10 producing B10 cells and the functions of T cells were not affected by IL-21. Taken together, the effect of IL-21 on B cells might vary between autoimmune EAE model and our xenogeneic GVHD model.

B cells in PBMCs mainly consist of naïve B cells and memory B cells. And naïve and memory B cells in PBMCs react differently to the antigen. In our experiments, there were about 60% naive B cells and 30% memory B cells in PBMCs, which varied among different donors. In X-GVHD model, both naïve B cells and memory B cells were injected into BRG mice, but we did not examine the difference of proliferation and differentiation between naïve and memory B cells in response to IL-21. Previous study showed that IL-21R was expressed on naïve B cells, but not on memory B cells (Good et al., 2006). Although memory B cells lacked IL-21R, it was rapidly induced following activation in vitro (Good et al., 2006). This may allow memory B cells to proliferate to the same extent as naïve B cells in in vitro culture. So we hypothesized that both naïve B cells and memory B cells in response to IL-21 proliferated in X-GVHD model.

We were also eager to know whether IL-21 had direct effect on B cells to accelerate X-GVHD and tried hard to solve this problem. However, it is important to note that the mechanism of GVHD is very complex, and CD4⁺ T cells, CD8⁺ T cells, and Th17 cells have crucial roles in the pathogenesis of GVHD (Blazar et al., 2012), which may compensate for the loss of B cells. So we did not prove this hypothesis and we will pay attention to it in the future experiments at the late stage of GVHD. Clinical trials showed that treatment with a CD20-specific antibody that depletes B cells, reduced GVHD incidence (Shimabukuro-Vornhagen et al., 2009; Sarantopoulos et al., 2011). In an allogeneic GVHD model, researchers found that the recipients given B-cell-depleted donor spleen (B220^-splenocytes) cells did not develop chronic GVHD over more than 40 days (Zhang et al., 2006). These studies indicated that donor B cells played an important role in the process of GVHD. And in our xenogeneic GVHD model, we also observed that the BRG mice given non-B cell PBMCs did not exhibit GVHD symptoms for 25 days. This result is consistent with previous report in allogeneic GVHD, and suggests that B cells were important to xenogeneic GVHD. We suppose that B cells act as APCs to augment T cell immune responses, which might be proved in the future.

Because autoantibodies play a major role in many kinds of autoimmune disorders, such as SLE, RA, and so on (Nielen et al., 2004; Sherer et al., 2004), we hypothesize that the expanded B cells in IL-21 treating mice may produce more antibodies which are specific to recipient tissues. Antibody production is one of the most important ways to protect us against infection and malignancy. When the immune system is weak and ineffective, autoantibodies, which directed at the body’s own (self) cellular components, are produced. Most autoimmune diseases are caused when specific autoantibodies develop and begin to injure the body’s tissues and cells. In our X-GVHD model, after engaging mouse tissues, human B cells were activated and proliferated, and then differentiated into plasma cells. High levels of IgM and IgG were produced by plasma cells which should be specific to mice tissues. For mice, these antibodies were actually autoantibodies. Because the production of autoantibodies was correlated with the progress of GVHD, we hypothesize that autoantibody is an important factor in the occurrence of GVHD. We are trying to confirm this by detecting or staining the complexes formed by autoantibodies and autoantigens.

The ability of IL-21 on proliferation of B cells indicates that it may have clinical application in conditions of immunodeficiency. For example, severe deficiency of memory B cells was observed in immunocompromised individuals, such as patients with immunodeficient states (Warnatz et al., 2002). Therefore, careful injection of IL-21 has the potential to restore the B cell compartment of these patients.

Taken together, our results firstly defined the effect of IL-21 on B cells in in vivo model. In X-GVHD model which well mimic the clinical presentation of GVHD, we found that IL-21 treatment could promote B cell expansion, plasma cell differentiation, and immunoglobulins secretion, which were related with the exacerbation of xenogeneic GVHD. Specifically targeting B cells or immunoglobulin may provide new approaches for the treatment of GVHD.

MATERIALS AND METHODS

Immunodeficient mice

BALB/c-Rag2^−/−γc^−/− mice were kindly provided by Pro. Lieping Chen, Department of Immunobiology, Yale University School of Medicine, New Haven, USA. Male and female BALB/c-Rag2^−/−γc^−/− mice used were between 8–12 weeks of age. The mice were maintained under specific pathogen-free conditions in the animal facility at the Institute of Biophysics, Chinese Academy of Sciences.

Plasmid in vivo transfection

The IL-21 cDNA was cloned from stimulated human PBMCs and inserted into PTT3-Flag to construct the IL-21 expression plasmid, while the mock PTT3-Flag was used as control plasmid in the study. To exclude endotoxin contamination, large scale preparation of PTT3-hIL-21-Flag and the control PTT3-flag plasmid DNA was purified by the EndoFree Plasmid Mega kit (Qiagen, Germany). Plasmid DNA was transfected in vivo by using a hydrodynamic-based gene transfer technique. Briefly, 20 μg/mouse plasmid DNA was diluted in 2 mL of PBS (about 0.1 mL/g body weight) and injected into the tail vein using a 27-gauge needle and syringe within a time period of 5–8 s. The dynamic expression of IL-21 in vivo was confirmed by detecting IL-21 concentration in the serum with an ELISA kit from eBioscience (88-7216, San Diego, CA, USA) at indicated time points.

Cell isolation and xenogeneic graft-versus-host disease (X-GVHD) model

Human PBMCs were isolated from heparinized blood of healthy donors by Ficoll-Hypaque density centrifugation, and then resuspended in PBS. For B cell depletion, PBMC were stained with FITC conjugated mAb specific for human CD19, followed by anti-FITC microbeads (Miltenyi, Bergisch Gladbach, Germany). The magnetically labeled B cells are depleted by retaining them on a MACS Column in the magnetic field of a MACS Separator, while the unlabeled non-B cell PBMC pass through the column. BALB/c-Rag2^−/−γc^−/− mice were hydrodynamic injected with PTT3-hIL21-Flag plasmid or mock control plasmid on day 0. On day 1, mice firstly received total body irradiation with a single dose of 3.5 Gy and then 1.0 × 10⁷ freshly isolated human PBMCs or non-B cell PBMC were intravenously injected via the tail vein on the same day. Body weight loss and other GVHD symptoms including hunched posture, ruffled fur, reduced mobility, and tachypnoea were measured.

Measurement of human immunoglobulins and cytokines

The concentrations of human IgG and IgM in serum of mice were determined using double antibody sandwich ELISA. All of these detection antibodies were purchased from Sigma-Aldrich (St. Louis, MO, USA). Anti-hIgG (I1886) and peroxidase conjugated anti-human IgG (A0170) were used for hIgG detection. Anti-human IgM (I2386) and peroxidase conjugated anti-human IgM (A0420) were used for hIgM detection.

The concentrations of human interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), IL-4, IL-5, IL-10 were measured using the human Th1/Th2 Cytokine Bead Array (BD pharmingen, San Diego, CA, USA).

FACS analysis

Mice were sacrificed at indicated time point. Cells were harvested from the spleens and stained with antibodies specific for human CD45, CD3, CD19, and/or CD38. Samples were analyzed by flow cytometry to calculate the percentages and absolute accumulation of human lymphocytes.

References

[1] Blazar, B.R., Murphy, W.J., and Abedi, M. (2012). Advances in graft-versus-host disease biology and therapy. Nat Rev Immunol 12, 443-458.

[2] Bucher, C., Koch, L., Vogtenhuber, C., Goren, E., Munger, M., Panoskaltsis-Mortari, A., Sivakumar, P., and Blazar, B.R. (2009). IL-21 blockade reduces graft-versus-host disease mortality by supporting inducible T regulatory cell generation. Blood 114, 5375-5384.

[3] Christopeit, M., Schutte, V., Theurich, S., Weber, T., Grothe, W., and Behre, G. (2009). Rituximab reduces the incidence of acute graft-versus-host disease. Blood 113, 3130-3131.

[4] Ettinger, R., Kuchen, S., and Lipsky, P.E. (2008). The role of IL-21 in regulating B-cell function in health and disease. Immunol Rev 223, 60-86.

[5] Ettinger, R., Sims, G.P., Fairhurst, A.M., Robbins, R., da Silva, Y.S., Spolski, R., Leonard, W.J., and Lipsky, P.E. (2005). IL-21 induces differentiation of human naive and memory B cells into antibody-secreting plasma cells. J Immunol 175, 7867-7879.

[6] Evenou, J.P., Wagner, J., Zenke, G., Brinkmann, V., Wagner, K., Kovarik, J., Welzenbach, K.A., Weitz-Schmidt, G., Guntermann, C., Towbin, H., et al. (2009). The potent protein kinase C-selective inhibitor AEB071 (sotrastaurin) represents a new class of immunosuppressive agents affecting early T-cell activation. J Pharmacol Exp Ther 330, 792-801.

[7] Good, K.L., Bryant, V.L., and Tangye, S.G. (2006). Kinetics of human B cell behavior and amplification of proliferative responses following stimulation with IL-21. J Immunol 177, 5236-5247.

[8] Gregoire-Gauthier, J., Durrieu, L., Duval, A., Fontaine, F., Dieng, M.M., Bourgey, M., Patey-Mariaud de Serre, N., Louis, I., and Haddad, E. (2012). Use of immunoglobulins in the prevention of GvHD in a xenogeneic NOD/SCID/gammac- mouse model. Bone Marrow Transplant 47, 439-450.

[9] Hanash, A.M., Kappel, L.W., Yim, N.L., Nejat, R.A., Goldberg, G.L., Smith, O.M., Rao, U.K., Dykstra, L., Na, I.K., Holland, A.M., et al. (2011). Abrogation of donor T-cell IL-21 signaling leads to tissue-specific modulation of immunity and separation of GVHD from GVL. Blood 118, 446-455.

[10] Hippen, K.L., Bucher, C., Schirm, D.K., Bearl, A.M., Brender, T., Mink, K.A., Waggie, K.S., Peffault de Latour, R., Janin, A., Curtsinger, J.M., et al. (2012). Blocking IL-21 signaling ameliorates xenogeneic GVHD induced by human lymphocytes. Blood 119, 619-628.

[11] Ito, R., Katano, I., Kawai, K., Hirata, H., Ogura, T., Kamisako, T., Eto, T., and Ito, M. (2009). Highly sensitive model for xenogenic GVHD using severe immunodeficient NOG mice. Transplantation 87, 1654-1658.

[12] Konforte, D., Simard, N., and Paige, C.J. (2009). IL-21: an executor of B cell fate. J Immunol 182, 1781-1787.

[13] Leonard, W.J., and Spolski, R. (2005). Interleukin-21: a modulator of lymphoid proliferation, apoptosis and differentiation. Nat Rev Immunol 5, 688-698.

[14] Lopez, M., Clarkson, M.R., Albin, M., Sayegh, M.H., and Najafian, N. (2006). A novel mechanism of action for anti-thymocyte globulin: induction of CD4+CD25+Foxp3+ regulatory T cells. J Am Soc Nephrol 17, 2844-2853.

[15] Meguro, A., Ozaki, K., Hatanaka, K., Oh, I., Sudo, K., Ohmori, T., Matsu, H., Tatara, R., Sato, K., Sakata, Y., et al. (2011). Lack of IL-21 signal attenuates graft-versus-leukemia effect in the absence of CD8 T-cells. Bone Marrow Transplant 46, 1557-1565.

[16] Meguro, A., Ozaki, K., Oh, I., Hatanaka, K., Matsu, H., Tatara, R., Sato, K., Leonard, W.J., and Ozawa, K. (2010). IL-21 is critical for GVHD in a mouse model. Bone Marrow Transplant 45, 723-729.

[17] Mutis, T., van Rijn, R.S., Simonetti, E.R., Aarts-Riemens, T., Emmelot, M.E., van Bloois, L., Martens, A., Verdonck, L.F., and Ebeling, S.B. (2006). Human regulatory T cells control xenogeneic graft-versus-host disease induced by autologous T cells in RAG2-/-gammac-/-immunodeficient mice. Clin Cancer Res 12, 5520-5525.

[18] Nielen, M.M., van Schaardenburg, D., Reesink, H.W., van de Stadt, R.J., van der Horst-Bruinsma, I.E., de Koning, M.H., Habibuw, M.R., Vandenbroucke, J.P., and Dijkmans, B.A. (2004). Specific autoantibodies precede the symptoms of rheumatoid arthritis: a study of serial measurements in blood donors. Arthritis Rheum 50, 380-386.

[19] Sarantopoulos, S., Stevenson, K.E., Kim, H.T., Washel, W.S., Bhuiya, N.S., Cutler, C.S., Alyea, E.P., Ho, V.T., Soiffer, R.J., Antin, J.H., et al. (2011). Recovery of B-cell homeostasis after rituximab in chronic graft-versus-host disease. Blood 117, 2275-2283.

[20] Sherer, Y., Gorstein, A., Fritzler, M.J., and Shoenfeld, Y. (2004). Autoantibody explosion in systemic lupus erythematosus: more than 100 different antibodies found in SLE patients. Semin Arthritis Rheum 34, 501-537.

[21] Shimabukuro-Vornhagen, A., Hallek, M.J., Storb, R.F., and von Bergwelt-Baildon, M.S. (2009). The role of B cells in the pathogenesis of graft-versus-host disease. Blood 114, 4919-4927.

[22] Shlomchik, W.D. (2007). Graft-versus-host disease. Nat Rev Immunol 7, 340-352.

[23] Shultz, L.D., Brehm, M.A., Bavari, S., and Greiner, D.L. (2011). Humanized mice as a preclinical tool for infectious disease and biomedical research. Ann N Y Acad Sci 1245, 50-54.

[24] Shultz, L.D., Ishikawa, F., and Greiner, D.L. (2007). Humanized mice in translational biomedical research. Nat Rev Immunol 7, 118-130.

[25] Spolski, R., and Leonard, W.J. (2008). Interleukin-21: basic biology and implications for cancer and autoimmunity. Annu Rev Immunol 26, 57-79.

[26] van Oosterhout, Y.V., van Emst, L., Schattenberg, A.V., Tax, W.J., Ruiter, D.J., Spits, H., Nagengast, F.M., Masereeuw, R., Evers, S., de Witte, T., et al. (2000). A combination of anti-CD3 and anti-CD7 ricin A-immunotoxins for the in vivo treatment of acute graft versus host disease. Blood 95, 3693-3701.

[27] van Rijn, R.S., Simonetti, E.R., Hagenbeek, A., Hogenes, M.C., de Weger, R.A., Canninga-van Dijk, M.R., Weijer, K., Spits, H., Storm, G., van Bloois, L., et al. (2003). A new xenograft model for graft-versus-host disease by intravenous transfer of human peripheral blood mononuclear cells in RAG2-/- gammac-/- double-mutant mice. Blood 102, 2522-2531.

[28] Warnatz, K., Denz, A., Drager, R., Braun, M., Groth, C., Wolff-Vorbeck, G., Eibel, H., Schlesier, M., and Peter, H.H. (2002). Severe deficiency of switched memory B cells (CD27(+)IgM(-)IgD(-)) in subgroups of patients with common variable immunodeficiency: a new approach to classify a heterogeneous disease. Blood 99, 1544-1551.

[29] Welniak, L.A., Blazar, B.R., and Murphy, W.J. (2007). Immunobiology of allogeneic hematopoietic stem cell transplantation. Annu Rev Immunol 25, 139-170.

[30] Yoshizaki, A., Miyagaki, T., DiLillo, D.J., Matsushita, T., Horikawa, M., Kountikov, E.I., Spolski, R., Poe, J.C., Leonard, W.J., and Tedder, T.F. (2012). Regulatory B cells control T-cell autoimmunity through IL-21-dependent cognate interactions. Nature 491, 264-268.

[31] Zhang, C., Todorov, I., Zhang, Z., Liu, Y., Kandeel, F., Forman, S., Strober, S., and Zeng, D. (2006). Donor CD4+ T and B cells in transplants induce chronic graft-versus-host disease with autoimmune manifestations. Blood 107, 2993-3001.

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg 2013