Search

Chronic graft-versus-host disease (cGVHD) is characterized as autoimmune-like fibrosis and antibody production mediated by pathogenic T cells and B cells. MicroRNA-17-92 (miR-17-92) influences the survival, differentiation, and function of lymphocytes in cancer, infections, and autoimmunity. To determine whether miR-17-92 regulates T- and B-cell responses in cGVHD, we generated mice conditionally deficient for miR-17-92 in T cells, B cells, or both. Using murine models of allogeneic bone marrow transplantation, we demonstrate that expression of miR-17-92 in donor T and B cells is essential for the induction of both scleroderma and bronchiolitis obliterans in cGVHD. Mechanistically, miR-17-92 expressed in T cells not only enhances the differentiation of pathogenic T helper 1 (Th1) and Th17 cells, but also promotes the generation of follicular Th cells, germinal center (GC) B cells, and plasma cells. In B cells, miR-17-92 expression is required for autoantibody production and immunoglobulin G deposition in the skin. Furthermore, we evaluated a translational approach using antagomirs specific for either miR-17 or miR-19, key members in miR-17-92 cluster. In a lupus-like cGVHD model, systemic administration of anti–miR-17, but not anti–miR-19, alleviates clinical manifestations and proteinuria incidence in recipients through inhibiting donor lymphocyte expansion, B-cell activation, and GC responses. Blockade of miR-17 also ameliorates skin damage by reducing Th17 differentiation in a scleroderma-cGVHD model. Taken together, our work reveals that miR-17-92 is required for T-cell and B-cell differentiation and function, and thus for the development of cGVHD. Furthermore, pharmacological inhibition of miR-17 represents a potential therapeutic strategy for the prevention of cGVHD.

The endoplasmic-reticulum-resident protein STING (Stimulator of IFN genes) is a downstream signaling effector of cytosolic DNA sensor cGAS (cyclic GMP-AMP synthase). STING-mediated innate immune activation plays a key role in tumor- and self-DNA elicited anti-tumor immunity and autoimmunity, respectively, yet the mechanism remains largely unclear. We utilized murine models of allogeneic hematopoietic cell transplantation (allo-HCT) to study the biology of STING in antigen-presetting cells (APCs) and T cells. STING expression in donor T cells was dispensable for their ability to induce graft-versus-host disease (GVHD), a major complication of allo-HCT in the clinic. However, when STING-deficient mice were used as recipients, more severe disease was induced after allo-HCT. Using bone marrow (BM) chimeras where STING was absent in different compartments, we found that STING-deficiency on host hematopoietic cells (Fig. A), but not on non-hematopoietic cells, was primarily responsible for exacerbating the disease. Furthermore, STING expression on host CD11c+ cells played a dominant role in the regulation of allogeneic T-cell responses (Fig. B). Mechanistically, STING deficiency resulted in increased survival, activation and function of irradiated APCs, including macrophages and dendritic cells (DCs, fig. C-D). To further determine the role of STING in APCs, we generated a STING V154M knock-in mouse model, in which V154M mutation in TMEM173 causes constitutive activation of STING. Consistently, constitutive activation of STING attenuated the survival, activation and function of APCs isolated from STING V154M knock-in mice. In addition, STING-deficient APCs augmented donor T-cell expansion, chemokine receptor expression and migration into intestinal tissues (Fig. E), resulting in accelerated/exacerbated disease. Using pharmacologic approaches, we demonstrate that systemic administration of a STING agonist (c-di-GMP) to recipient mice before transplantation significantly reduced GVHD mortality (Fig. F). In conclusion, we report an inhibitory role of STING in regulating survival and T-cell priming function of hematopoietic APCs, especially CD11c+ cells, after allo-HCT. We validate that pharmacological activation of STING may serve as a potential therapeutic strategy to constrain APCs and induce immune tolerance. Figure Disclosures No relevant conflicts of interest to declare.

Graft-versus-host disease (GVHD) remains one of the major complications after allogeneic bone marrow transplantation (allo-BMT). Sirtuin-1 (Sirt-1) plays a crucial role in various biological processes including cellular senescence, metabolism, and inflammatory responses. Sirt-1 deacetylation regulates different transcription factors that are important for modulating immune responses. In the current study, we addressed the role of Sirt-1 in GVHD induction by employing Sirt-1 conditional knockout mice as well as a pharmacological Sirt-1 inhibitor. Using major histocompatibility complex (MHC)–mismatched and MHC-matched murine BMT models, we found that Sirt-1−/− T cells had a reduced ability to induce acute GVHD (aGVHD) via enhanced p53 acetylation. Sirt-1-deficient T cells also promoted induced regulatory T cell (iTreg) differentiation and inhibited interferon-γ production after allo-BMT. Sirt-1 deletion in iTregs increased Foxp3 stability and restrained iTreg conversion into pathogenic T cells. Furthermore, we found that administration with a Sirt-1 inhibitor, Ex-527, significantly improved recipient survival and clinical scores, with no signs of tumor relapse. These results indicate that Sirt-1 inhibition can attenuate GVHD while preserving the graft-versus-leukemia effect. Consistently, Sirt-1-deficient T cells also displayed a remarkably reduced ability to induce chronic GVHD (cGVHD). Mechanistic studies revealed that Sirt-1 deficiency in T cells enhanced splenic B-cell reconstitution and reduced follicular T helper cell development. Sirt-1 deficiency in T cells modulated donor B-cell responses reducing both B-cell activation and plasma cell differentiation. In addition, therapeutic Sirt-1 inhibition could both prevent cGVHD and reduce established cGVHD. In conclusion, Sirt-1 is a promising therapeutic target for the control of aGVHD and cGVHD pathogenesis and possesses high potential for clinical application.

Allogeneic hematopoietic stem cell transplantation (allo-HCT) is an effective means by which to treat a wide variety of diseases resulting from hematological dysfunction. However, the development of graft-versus-host disease (GVHD) remains a major cause of morbidity and mortality post transplantation. The IL-12 family of cytokines is comprised of IL-12, IL-23, IL-27, IL-35, and potentially IL-39. IL-12 family members are unique in that each cytokine and cognate receptor is comprised of heterodimers in which either the a or b subunit is shared among the others. IL-12 (p35+p40) and IL-23 (p19+p40) have well documented proinflammatory functions responsible for Th1 differentiation and Th17 stabilization, respectively, and play critical roles in GVHD development. IL-12R and IL-23R share a β-chain (IL-12Rβ1) yet use distinct α-chains to mediate their respective receptor signaling. While both IL-12R and IL-23R are widely implicated in inflammatory disorders, the role of IL-12Rβ1 in this context remains much less defined. We therefore studied the impact of eliminating the common IL-12Rβ1 chain or the unique IL-23Rα chain in T cells on GVHD using murine models of allogeneic bone marrow transplantation (BMT). In agreement with previous publications, we found a pathogenic role for IL-23Rα on donor T cells in aGVHD. Strikingly, a similar effect was not seen for IL-12Rβ1 (Figure 1A, B). These data suggest that that IL-23Rα contributes to GVHD pathogenesis via a pathway independent of IL-12Rβ1. To confirm that functional differences existed between T cells deficient for IL-23Rα or IL-12Rβ1 in GVHD, we assessed cytokine profiles of these T cells in target organs 14 days post-BMT. We found that, while production of IFNγ and IL-17 in the spleen was similarly decreased in both cohorts, GM-CSF production by CD4+ T cells was reduced exclusively in T cells deficient for IL-23Rα. Further, a significant reduction of IFNγ and GM-CSF in target organs, such as the liver and gut, was only observed in T cells deficient for IL-23Rα. The newest member of the IL-12 family, IL-39, was recently shown to contribute to SLE pathogenesis; this cytokine has been described to be composed of IL-23p19 and EBI3. Given the cognate receptor for IL-39 includes IL-23Rα and gp130, we hypothesized that IL-39 may play a role in aGVHD as this would explain why IL-12Rβ1 is dispensable. To validate that p19 and EBI3 can form a heterodimer, we transfected SV40 cells with vectors containing control, IL-23p19, EBI3 or both cDNAs. We detected IL-39 heterodimers only in the supernatant of cells transfected with both IL-23p19 and EBI3 via ELISA (Figure 1C). Furthermore, we observed significantly increased levels of IL-39 in allogeneic recipients at day 14 post BMT compared to naïve mice or recipients of BM alone in two models of aGVHD (Figure 1D). This may implicate IL-39 in the GVHD development. Taken together, our studies indicate that IL-23Rα plays an essential role, whereas IL-12Rβ1 is dispensable, for donor T cells to induce aGVHD. Our proposed model is that in the absence of IL-12Rβ1, IL-39 could transmit IL-23Rα signaling, hypothetically by forming a heterodimer with gp130. This new finding indicates that IL-23Rα and IL-39 are potential therapeutic targets for controlling aGVHD in the clinic. Figure 1. Effect of IL-12R 𝛃 1 in aGVHD and the potential role of IL-39 Lethally irradiated BALB/c mice were transplanted with 5x106 TCD-BM alone or plus 1x106 purified T cells from WT B6, IL-12R𝛃1KO, or IL-23RαKO mice. Survival (A) and body weight loss (B) are shown. Supernatant from SV40 cells transiently transfected with vectors containing control, IL-23p19, EBI3 or both cDNAs. p19 and EBI3 heterodimers were detected via ELISA (C). Serum was collected from naïve or lethally irradiated mice transplanted with 5x106 TCD-BM alone or plus 1x106 purified T cells. Formation of p19 and EBI3 heterodimers were tested in serum at 14 days post BMT via ELISA (D). Disclosures No relevant conflicts of interest to declare.

MicroRNAs (miRs) repress gene expression at the post-transcriptional level via binding target mRNAs subsequently promoting mRNA degradation or impeding translation. Certain miRs influence the survival, differentiation, and function of T cells in cancer, infections, and autoimmunity. Uncontrolled T-cell allogeneic responses contribute to chronic graft-versus-host disease (cGVHD), a major cause of non-relapse mortality following allogeneic hematopoietic stem-cell transplantation (allo-HCT). Dysregulation of miR-31 is implicated in cancer, but how miR-31 impacts allogeneic T-cell response is unclear. Using oligonucleotide array, we found miR-31 was dramatically up-regulated in T cells under alloantigen-driven vs. homeostatic proliferation in recipient mice following bone marrow transplantation (BMT) (Figure 1A-B). To understand how miR-31 regulates T-cell responses to alloantigens, we utilized murine models of allogeneic BMT and donor mice with miR-31 conditional knock-out on their T cells. In a sclerodermatous cGVHD model (B6→BDF1), we observed that the recipients transferred with donor T cells deficient for miR-31 exhibited significantly alleviated disease (Figure 1C), reflected by attenuated fibrosis and pathologic damage in skin. In a bronchiolitis obliterans cGVHD model (B6→B10.BR), we found that the recipients of miR-31-deficient T cells had reduced airway resistance, elastance (Figure 1D) and pathological damage in the lungs as compared with those of WT T cells. The role of miR-31 in promoting T-cell pathogenicity was further confirmed when miR-31 was inhibited by administration of specific antagomir (locked nuclear acid anti-miR-31). We used mass cytometry to analyze donor immune cell reconstitution in recipient spleens after BMT. Consistent with attenuated disease manifestation, the recipients of miR-31-deficient T cells displayed improved reconstitution of donor T- and B- lymphocytes. The follicular T helper cells (TFH) instruct germinal center (GC) B-cell expansion, affinity maturation, and plasma cell differentiation, whereas follicular regulatory T cells (TFR) inhibit TFH -mediated B-cell activation and antibody production. We found that miR-31 deficient T cells differentiated into more TFR, but fewer TFH during alloresponses. As a result, the differentiation and activation of GC B cells and the generation of plasma cells were attenuated in the recipients of miR-31-deficient T cells. Furthermore, miR-31-deficient T cells exhibited defects in proliferation and survival in allogeneic recipients, resulting in fewer donor T cells in recipient thymus, skin and lungs. In the absence of miR-31, T cells differentiated less into Th17 cells but more towards Tregs (Figure 1E) in vivo. Those miR-31-deficient Tregs expressed higher levels of Neuropilin 1 and PD-L1, the markers associated with superior stability and suppressive function of Tregs. While miR-31 played little role on Th1 cell differentiation, it facilitated alloantigen-reactive iTregs losing Foxp3 and producing IFNγ after being transferred into allogeneic recipients. Hypoxia, a hallmark of inflamed and damaged tissue, can drive fibrosis and disease development through immune cell dysregulation. Upon activation, T cells rapidly increase their metabolic rate and switch from oxidative phosphorylation to aerobic glycolysis. We found that miR-31-deficient T cells exhibited reduced hypoxia-inducible factor 1α (HIF1α) signaling (glycolysis promotor), surface GLUT1 expression, and glucose uptake, but increased lipid droplet accumulation in allogeneic recipients, suggesting that miR-31 promotes metabolic switch from fatty acid β-oxidation to aerobic glycolysis in allo-reactive T cells (Figure 1F). Furthermore, under hypoxia (3% oxygen) but not normoxia (21% oxygen) conditions, miR-31 increased Th17 but decreased iTreg differentiation from naïve CD4 T cells in the presence of IL-6 and TGFβ in vitro. Taken together, miR-31 regulates T-cell expansion, differentiation, and metabolism via promoting HIF1α expression and hypoxia adaptation in allo-reactive T cells, which enhances T-cell effector function and pathogenicity after allogeneic BMT. Disclosures No relevant conflicts of interest to declare.

Allogeneic hematopoietic stem cell transplantation (allo-HCT) is a procedure undertaken to cure hematological malignancies, especially leukemia. Graft-Versus-Host Disease (GVHD) is a serious condition that often appears following allo-HCT. Friend Leukemia Virus Integration 1 (Fli-1) is a transcription factor highly expressed in cancers including erythroleukemia and acute myeloid leukemia while also implicated in pathogenesis of systemic lupus. We have interrogated the role of Fli-1 in T-cell responses by generating a novel T-cell specific conditional disruption of fli-1 where essential exons 3 and 4 of the gene are floxed and excised in the presence of CD4-Cre recombinase. Models of acute GVHD (aGVHD) and chronic GVHD (cGVHD) were tested utilizing hematopoietic cells from mice with a heterozygous (fli-1-/+ or Het) or homozygous (fli-1-/-or Null) disruption of the fli-1 gene, or from wild-type (fli-1+/+ or WT) littermate controls. At baseline, T cells among each of these three mouse strains showed no significant difference in CD44/CD62L expression or CD4+FoxP3+ (nTreg) frequencies. In the cGVHD model, BALB/c (H2Kd) recipients were infused with allogeneic B6 (H2Kb) genotype-respective bone marrow and splenocytes in order to induce cGVHD. Recipients that received fli-1-/+CD4Cre+ marrow and splenocytes demonstrated improved survival and mild cGVHD, whereas those that received fli-1-/-CD4Cre+ or WT donor cells developed serve cGVHD (Fig. 1 a-f). Cellular studies from lymphoid organs of cGVHD allo-HCT recipients revealed that disruption of fli-1 was associated with decreased frequencies of donor CD4+ cells expressing IL-17A, IFN-γ, T follicular-like (TFH-like) cell markers, and CD8+ cells expressing PD-1. In aGVHD settings, donor fli-1-/+CD4Cre+ T cells had a decreased ability to induce aGVHD compared to WT donor T cells and fli-1-/- donor T cells (Fig. 2 a-b). When investigating cellular mechanisms in aGVHD settings, we found that fli-1-/+CD4Cre+ T cells produced significantly lower IFN-γ early after T-cell activation in vivo compared to WT and fli-1-/-CD4Cre+ T cells (Fig. 2 c). To then investigate the role of Fli-1 in T cells beyond GVHD, we utilized a colitis model by transferring naïve CD4+ T cells (CD44-CD25-) into Rag2-/- syngeneic recipients. While fli-1+/+ T cells induced severe colitis as expected, fli-1-/- and fli--/+ T cells caused moderate and mild colitis, respectively. These results were consistent with those observed in GVHD models. To elucidate underlying mechanisms, we tested the effects of Fli-1 on antigen-specific T-cells using a MHC-II restricted TCR transgenic (Tg) mouse strain specific for HY-peptide. CD4+CD25- T cells from fli-1+/+, fli-1-/+, or fli-1-/- CD4Cre+ TCR-Tg mice were polarized in vitro with endogenous antigen presenting cells from spleen in presence of HY-peptide under iTreg- or Th17-polarizing conditions. Both fli-1-/+ and fli-1-/- TCR-tg T cells exhibited significant increases in iTreg (CD4+FoxP3+) frequencies and surface expression of iTreg functional markers (CD25, CD39, CD73, NRP-1), while also having significantly decreased frequencies of IL-17A producing T cells compared to WT-TCR-tg T cells. To further explore the molecular mechanisms, we retrieved fli-1+/+, fli-1-/+, or fli-1-/- donor T cells from recipient spleens after allo-HCT and did RNAseq analysis on these cells. RNAseq data reveals significant differences in mRNA expression within acute inflammatory response and positive regulation of the immune response enrichment pathways between fli-1-/+ T cells and littermate fli-1+/+ T cells, and to a lesser extent in fli-1-/- T cells (Fig. 2 d-f). Thus, reducing Fli-1 transcriptional activity represents a potential therapeutic concept toward ameliorating GVHD after allo-HCT, while simultaneously targeting cancers such as leukemia which typically overexpress Fli-1. Disclosures No relevant conflicts of interest to declare.

The diversity and composition of T cell receptor (TCR) repertoire, which is the result of V, D and J gene recombination in TCR gene locus, has been found to impact immune responses in autoimmune and infectious diseases. The correlation of T-cell repertoire with the pathogenesis and outcome of graft-versus-host disease (GVHD) remain undefined. Here, by utilizing high-throughput sequencing of the gene encoding the TCRβ-chain, we comprehensively analyzed the profile of T-cell repertoire in host lymphoid and GVHD target organs after bone marrow transplantation (BMT). To understand whether T-cell repertoire is affected by different strength of alloantigen stimulation, we transferred same donor T cells derived from C57BL/6 (B6) mice into irradiated BALB/c (MHC-fully mismatched), B6D2F1 (MHC-haploidentical), BALB.b (MHC-matched ) and B6 recipients (syngeneic). Fourteen days later, T cells were isolated from recipient peripheral blood, spleen, peripheral lymphoid nodes (pLN), mesenteric lymphoid nodes (mLN), liver, lung, gut and skin for TCR sequencing. Clonality of donor T cells, which is inversely associated with TCR diversity, was significantly increased in either syngeneic or allogeneic recipients when compared with naïve donor T-cells, consistent with the concept that TCR diversity is reduced after T-cell activation and expansion. Increased TCR clonality was observed in lymphoid organs of allogeneic compared with syngeneic recipients, confirming that donor T cells were further activated in allogeneic recipients. However, decreased TCR clonality was observed in GVHD target organs of allogeneic compared with syngeneic recipients, suggesting that only limited donor T-cell clones were able to migrate in target organs in syngeneic compared to allogeneic recipients. The frequency of top clones in total productive rearrangements was increased in GVHD target organs especially liver of allogenic than syngeneic receipts. Interestingly, the frequency of top clones was positively associated with MHC disparity between donor and host, ranging from low to high in syngeneic, MHC-matched, haploidentical, and fully-mismatched recipients, respectively. To understand the extent to which TCR rearrangement is shared among different organs after BMT, we analyzed the overlap of TCR clones across different organs in the same recipients. T-cell clones were highly overlapping across organs, especially among GVHD target organs, in the same recipients after allogeneic BMT, although much lower overlapping in recipients after syngeneic BMT. The results suggest that alloantigen stimulation selectively activate certain T-cell clones and enrich antigen specific clones. On the other hand, much fewer shared clones were found among different recipients within the same group, regardless of MHC-disparity between donor and host. These results suggest that specific T-cell clones activated and expanded by alloantigens stimulation were highly different in individual recipients even with the same MHC-disparity between donor and host. Interestingly, the levels of clone overlapping were different in distinct organs among individual recipients. The level of T-cell clone overlapping was found high in liver of individual recipients regardless of the strength of alloantigen stimulation. The level of T cell clone overlapping was relatively high in pLNs and skin of the recipients after haploidentical BMT; whereas the level of T cell clone overlapping was relatively high in mLNs and gut of the recipients after MHC-matched BMT. These results suggest that skin may be a dominant target in haploidentical BMT and gut as a dominant target in MHC-matched BMT; whereas liver is a common target organ regardless. In conclusion, the current study establishes the association between MHC disparity, T-cell activation, and GVHD development in the level of donor T-cell repertoire. While TCR repertoire of donor T cells in peripheral blood or lymph nodes likely is representative in any individual recipient/patient, it is nearly impossible to identify T-cell clones that are pathogenic and shared among groups of recipients/patients even with the same MHC-disparity between donor and host. Disclosures No relevant conflicts of interest to declare.

Inhibition of the endoplasmic reticulum (ER) stress response via blockade of inositol-requiring enzyme-1α (IRE-1α) is currently a promising therapeutic strategy to treat B-cell leukemia, lymphoma, and myeloma. Because B cells play an important role in the development of chronic graft-versus-host disease (cGVHD), we hypothesize that the ER stress response contributes to B-cell function and pathogenicity in cGVHD. Here, we report that the ER stress response mediated by IRE-1α and its target X-box binding protein-1 (XBP-1) plays a critical role in cGVHD pathophysiology and represents a potential therapeutic target to prevent cGVHD. We tested the role of XBP-1 specifically in B cells by testing XBP-1 conditional knockout B cell grafts (XBP1fl/flCD19Cre+) in two mouse models of cGVHD. In the first model (B6 to BALB/c), recipients given XBP-1-deficient donor grafts showed significantly reduced cGVHD clinical scores, which were associated with reduced frequencies of donor-derived CD4 helper T cells within the lungs compared to the recipients of XBP-1fl/flCD19Cre- littermate donor grafts. XBP-1-deficient B cells produced significantly higher levels of IL-10 compared to WT control B cells after activation ex vivo. In the second model (B6 to B10.BR), the conversion of donor B cells to plasma cells (B220+CD38+CD138+) was reduced in both the spleens and lungs of recipients transplanted with XBP1fl/flCD19Cre+ grafts compared to those of the recipients given XBP1fl/flCD19Cre- grafts. Recipients given XBP1fl/flCD19Cre+ grafts also showed significantly higher total splenocytes and vastly increased splenic B-cell populations when compared with the recipients of XBP1fl/flCD19Cre- grafts. To expand on these findings, we tested if systemic XBP-1 blockade via a novel IRE-1α inhibitor, B-I09, would attenuate cGVHD. In a cutaneous model of cGVHD (B10.D2 to BALB/c), we found that prophylactic administration of B-I09 significantly reduced clinical features of cGVHD compared to vehicle controls (Fig. 1A). Validating these findings, hematoxylin and eosin stained skin sections of B-I09-treated mice had significantly lower pathology scores compared to vehicle controls (Fig. 1B). Isolated skin lymphocytes from recipients treated with B-I09 showed significant reductions in donor derived T cells and DCs compared to those treated with vehicle controls (Fig. 1C and D). Taken together, our findings reveal a novel role of the IRE-1α/XBP-1 pathway of the ER stress response in cGVHD pathophysiology and provide a readily translatable strategy to prevent the development of cGVHD in the clinic. Disclosures No relevant conflicts of interest to declare.

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is an effective therapy for many hematological malignancies due to its potent graft-versus-leukemia (GVL) effect; yet its clinical application is limited by the development of graft-versus-host-disease (GVHD). Currently, the separation of GVHD and GVL is a great challenge in the field. Complement can be activated via three different pathways: the classical (CP), lectin (LP) and alternative pathway (AP). The AP can spontaneously activate, but it also serves as an amplification loop for the other pathways, and hence plays a central role in the development of autoimmune diseases such inflammatory bowel disease, which shares certain pathological similarities with intestinal GVHD. Complement activation has been implicated in GVHD development, yet how complement regulates GVHD is not known. In current study, deposition of complement activation product C3d occurs within GVHD target organs (Fig. 1A), suggesting that local complement activation contributes to GVHD pathogenesis. To functionally dissect the roles of each complement pathway, we used mice deficient in fB (no AP) or C1q/MBL (no CP or LP) as recipients in MHC-mismatched murine models of GVHD. GVHD severity and mortality was significantly reduced in fB, but not C1q/MLB, deficient recipients compared to wild type (WT) counterparts, suggesting a crucial role for the AP in GVHD (Fig. 1B). Furthermore, using chimeric mice with a fB-deficiency specifically in the hematopoietic compartment, we demonstrated that the AP in hematopoietic cells is required for GVHD development, but not for the GVL effect (Fig. 1D). For translational purposes, we evaluated GVHD and GVL responses using CR2-fH, an AP-specific protein inhibitor (Fig. 1C). CR2-fH consists of the N-terminus of mouse complement factor H, which contains the AP-inhibitory domain, linked to complement receptor 2 (CR2) targeting fragment that binds the complement activation product C3d. CR2-fH has been shown to specifically co-localize to sites of injury and C3d deposition. In recipients harboring C1498 acute myeloid leukemia, systemic treatment of recipients with CR2-fH significantly ameliorated GVHD with no associated tumor relapse (Fig. 1D), suggesting CR2-fH does not impact GVL activity. Mechanistic studies revealed that host AP deficiency significantly impacted the phenotype of immune cells in GVHD target organs rather than in lymphoid organs. Complement deposition was decreased in GVHD target organs of AP-deficient recipients. Gut integrity was significantly preserved in fB-/- compared to WT recipients (Fig. 1E). Activation and maturation of host DCs in the gut, lung, and thymus were also significantly decreased in fB-/- recipients, while GVHD-protective CD8+DCs were increased. In contrast, AP deficiency did not affect the phenotype of recipient DCs in the lymphoid organs. The numbers of donor T cells and particularly CD103+CD8+ T cells, a cell subset that critically mediates intestinal GVHD, were decreased in the gut of fB-/- recipients, which was associated with reduced expression of the gut homing chemokine CCR9 on donor T cells. Donor iTregs were increased in the liver of fB-/- recipients (Fig. 1F), while Th1 and Tc1 cells were significantly decreased; lower cytolytic activity was also seen in these Tc1 cells. Donor Th2 and Tc1 were diminished in the lung while thymic integrity was significantly improved in fB-/- recipients. We observed no differences in differentiation or function of donor T cells in lymphoid organs (Fig. 1G). These studies further indicate AP activation within target organs is important for GVHD pathogenesis. Taken together, the current data identifies a central role for the AP in GVHD and validates the AP as a therapeutic target for the control of GVHD. We provide evidence that local complement generation in GVHD target organs is important for GVHD development. Since site-specific AP inhibition is effective at controlling GVHD while preserving GVL activity, this study identifies a novel strategy to reduce GVHD and spare GVL effects. Because site-specific complement inhibition has been shown to reduce potentially dangerous side effects and since a humanized version of CR2-fH (TT30) has been shown to be safe and nonimmunogenic in humans, the current findings have a high potential for clinical translation. Disclosures No relevant conflicts of interest to declare.

The successful treatment of hematologic malignances with allogeneic hematopoietic cell transplantation (allo-HCT) is limited by acute graft-versus-host disease (GvHD). The complement system has been shown to modulate adaptive immunity via interaction between complement activation products and their receptors expressed on both innate and adaptive immune cells. Complement receptors play an important role in pathogen and danger sensing by translating information gathered by complement fluid phase sensors into cellular responses. The anaphylatoxins, C3a and C5a, are key effector molecules of the complement system. C3aR/C5aR signaling plays an important role in the survival, maturation and differentiation of antigen presenting cells (APCs), as well as for effective antigen presentation to T cells and the subsequent modulation of T-cell proliferation, differentiation and function. Given that host APCs play a crucial role in priming alloreactive donor T cells to induce and intensify aGvHD, we evaluated the role of C3aR/C5aR in the induction of aGvHD via the regulation of host APC function. Using the two distinct well-defined, clinically relevant aGVHD models, B6 (H2b)->BALB/c (H2d) and FVB (H2q)->B6 (H2b), we observed that host C3aR/5aR deficiency led to a significantly reduced aGvHD as indicated by a higher survival rate and milder GvHD clinical scores as compared to WT recipients (Figure 1A). The ameliorated GVHD in C3aR/C5aR-/- recipients was associated with reduced donor T-cell activation, survival, bioenergetic capacity and Th1 differentiation, and with increased iTreg generation. Donor T cells in C3aR/C5aR-/- recipients express lower levels of chemokine receptors, CXCR3 and CCR6, which is likely responsible for the decreased migration of the effector T cells to GvHD target organs. Utilizing BM chimeras to distinguish the role of C3aR/C5aR on host hematopoietic cells vs. parenchymal tissues in the development of GVHD, we found that C3aR/C5aR expressed on recipient hematopoietic APCs was primarily responsible for donor T-cell response and pathogenicity in aGvHD. Among APCs, DCs are considered to be the most efficacious APCs due to their superior ability to take up antigen, express co-stimulatory molecules, and produce pro-inflammatory cytokines to polarize alloreactive T cells. C3aR/C5aR expression was drastically upregulated in DCs after total body irradiation. In comparison with WT DCs, irradiated C3aR/C5aR-/- DCs displayed higher apoptosis which is associated with DC upregulation of Fas expression, higher autophagy, and decreased activation and antigen presenting capacity reflected by lower IFN-γ positive DC cells and reduced MHCII expression. Notably, while C3aR/C5aR on hematopoietic cells was required for GVHD development, it was largely dispensable for the graft-versus-leukemia (GVL) effect. For translational application, we evaluated the effect of systemic administration of a combination of C3aR and C5aR antagonists (C3aRA/C5aRA) to block C3aR/C5aR signaling, on GVHD and GVL activity. We found that prophylactic treatment of recipients with C3aRA/C5aRA prior to transplant effectively prevented GvHD while preserving GVL effect (Figure 1B). Taken together, the current work provides a strong rationale and demonstrates the feasibility to target host C3aR/C5aR for the control of GVHD while preserving GVL activity after allo-HCT. Disclosures No relevant conflicts of interest to declare.

Janus kinase 2 (JAK2) signal transduction is a critical mediator of immune response. JAK2 activation promotes T-cell allosensitization, as well as Th1 and Th17 differentiation. JAK2-mediated STAT3 phosphorylation limits the generation of induced regulatory T cells (iTregs) by disrupting interactions between STAT5 and Foxp3. JAK2 is implicated in the onset of graft-versus-host disease (GVHD), which is a significant cause of transplant-related mortality after allogeneic hematopoietic cell transplantation (allo-HCT). Using murine models of allo-HCT we show here that transfer of donor JAK2-/- T cells is associated with significantly less GVHD, compared with wild-type or JAK2 replete donors (Figure 1, P =.003 and 0.01, respectively). Th1 differentiation among JAK2-/- T-cells is dramatically decreased, compared with controls. Conversely, iTreg polarization and stability are significantly increased among the JAK2 deficient T cells. To investigate the effect of pharmacologic JAK2 inhibition on T-cell alloresponses, pacritinib (supplied by CTI BioPharma), was chosen as it does not induce myelosuppression or increase risk for opportunistic infections in myelofibrosis patients - unlike JAK1/JAK2 inhibitors. Pacritinib potently inhibits JAK2, but also has suppressive activity toward JAK3, CSF1R, and IRAK1. Pacritinib was administered at 100mg/kg twice a day by oral gavage for 4 weeks beginning on the day of MHC-disparate allo-HCT, significantly reducing GVHD in recipient mice compared with methylcellulose vehicle control. In allogeneic mixed leukocyte reactions using human cells, pacritinib (2.5μM) significantly reduces T-cell proliferation after 5 days of culture (P *BCB and DB contributed equally to this work. Figure 1. Decreased murine GVHD with JAK2 deficient T-cells Figure 1. Decreased murine GVHD with JAK2 deficient T-cells Figure 2. Decreased human skin graft rejection with JAK2 inhibitor, pacritinib Figure 2. Decreased human skin graft rejection with JAK2 inhibitor, pacritinib Disclosures Singer: CTI BioPharma, Corp: Employment, Equity Ownership.

Allogeneic hematopoietic cell transplantation (HCT) can cure a variety of benign and malignant hematopoietic disorders, but graft-versus-host disease (GVHD) remains a significant source of transplant-related mortality and morbidity. Mature T cells in donor stem cell graft are primarily responsible for the development of acute GVHD. PIM kinases are a family of serine/threonine kinases, including PIM1, PIM2 and PIM3, which are expressed following T-cell activation. The PIM kinases have been shown to inhibit apoptosis as well as to stimulate the cell metabolism and protein synthesis. In the current study, we investigated how PIM kinases regulate T-cell responses to alloantigens and GVHD development. Using genetically modified mice on an FVB background that are deficient for either single, double, or triple kinases of the PIM family, we evaluated the role of PIM kinases on T-cell alloresponses. T cells deficient for PIM1/PIM3 kinases had reduced survival and proliferation upon stimulation with alloantigens in vitro, whereas T cells without the PIM2 kinase displayed increased survival and IFNγ production as compared to WT T cells. After being transferred to lethally irradiated allogeneic mice in vivo, PIM1/PIM3 double knockout (KO) T cells had reduced expression of IFNγ in spleen, whereas PIM2 KO T cells showed decreased IL-4/5 expression, although similar IFNγ production when compared with WT T cells. These data suggest PIM2 kinase plays a distinct role by negatively regulating T-cell alloresponses. To test the role of PIM kinases on T-cell ability to induce GVHD, we transferred WT, PIM2 single KO, PIM1/PIM2 double KO, PIM1/PIM3 double KO, PIM2/PIM3 double KO, or PIM1/PIM2/PIM3 triple KO T cells together with WT bone marrow cells into lethally irradiated B6 mice. While PIM1/PIM3 double KO T cells induced milder GVHD, PIM2 KO T cells induced much more severe GVHD when compared to WT T cells. In addition, any types of T cells deficient for PIM2 caused significantly more severe GVHD than their PIM2-replete counterparts. Furthermore, restoration of PIM2 expression by gene transfection reduced the ability of PIM2-deficient T cells in the induction of GVHD. Enhanced pathogenicity of PIM2-deficient T cells was also confirmed in another murine model of allogeneic HCT. These results indicate that PIM2 plays a dominant role among PIM kinases in negatively regulating T-cell alloresponses and GVHD induction. To validate the data obtained from KO mice, we silenced the PIM2 kinase in PIM1/PIM3 KO T cells by administration of a PIM kinase pan-inhibitor (AZD, AstraZeneca R&D, Waltham, MA), and we observed exacerbated GVHD induced by PIM1/PIM3 KO T cells after PIM2 kinase silencing. Our study demonstrates the important result that PIM2 negatively regulates T-cell mediated alloresponse and GVHD development. The finding uncovers a novel biological function of PIM kinases as well as urges caution in inhibiting PIM kinases for the treatment of hematologic malignances in patients after allogeneic HCT. YW and AD contributed equally to this work. Disclosures No relevant conflicts of interest to declare.

Adoptive regulatory T-cell (Treg) therapy has enhanced the outcome of patients suffering from graft-versus-host (GVH) disease following allogeneic hematopoietic stem cell transplantation (allo-HCT); however, fear of broad immune suppression and subsequent dampening of the beneficial graft-versus-leukemic (GVL) responses remains a challenge. In order to subvert broad immune suppression, we generated alloantigen-specific induced Tregs (iTregs) from resting CD4 or CD8 T cells and tested the ability of iTregs to suppress GVH and maintain GVL responses. We utilized a clinically relevant murine model of haploidentical-HCT with the addition of host-original leukemia cell line to evaluate the effects of CD4 and CD8 iTregs in GVH and GVL responses. While alloantigen-specific CD4 iTregs were effective in preventing GVHD (Fig. 1 A and C), they completely abrogated the GVL effect against aggressive leukemia resulting in 100% tumor mortality (Fig. 1 B and D). Mechanistically, these CD4 iTregs were found to potently suppress the expansion of effector T cells (Teffs) and their ability to secrete IFNγ and granzyme B in the recipient spleen and liver, which may contribute to the impaired GVL activity. Using similar approach, we generated alloantigen-specific CD8 iTregs and found they express higher levels of granzyme B and CTLA-4 compared to nTreg and CD4 iTregs. In vivo studies showed these CD8 iTregs moderately attenuated GVHD (Fig. 1 A and C)while completely sparing the GVL effect (Fig. 1 B and D). We thus further reasoned that the combination of CD4 and CD8 iTregs could achieve the optimal goal of allo-HCT: GVHD suppression with GVL preservation. Indeed, the combination therapy potently suppressed GVHD resulting in increased survival and decreased pathological injury to target organs than either CD4 or CD8 iTreg singular therapy (Fig. 1 A and C). More importantly, the combination therapy maintained potent GVL responses reflected by significantly decreased tumor mortality and load (Fig. 1 B and D).Mechanistically, we observed addition of CD8 iTregs maintained the suppression of Teff expansion but restored the ability of Teffs in producing inflammatory cytokines (e.g. IFNγ and TFNα) and cytolytic effector molecules (e.g. granzyme B and TRAIL). To our knowledge the current findings are the first to support the use of combinational iTreg therapy to achieve optimal suppression of GVHD while maintaining GVL responses. This work was supported by NIH grants: R01 CA118116 and R01 CA169116 Disclosures No relevant conflicts of interest to declare.

Allogeneic hematopoietic cell transplantation (allo-HCT) is an effective therapy for hematologic malignancies through T cell-mediated graft-versus-leukemia (GVL) effects, but allogeneic T cells often lead to severe graft-versus-host disease (GVHD). Cell metabolism plays pivotal roles in T-cell activation, differentiation, and function. However, understanding of T cell-metabolism is still superficial, and even less is known how metabolism regulates T-cell response to alloantigens and GVHD induction after allo-HCT. In this study, using a high-throughput liquid-and gas-chromatography-based metabolic approach, we compared the metabolic process of allogeneic versus syngeneic T cells at day 4 (early preclinical stage), day 7 (preclinical stage), and day 14 (clinical stage) post bone marrow transplantation (BMT), with naïve T cells as additional controls. Over 180 metabolites were identified and quantified. T cells after being transferred into pre-conditioned recipients were undergoing metabolic reprogramming reflected by attenuated levels of metabolites involving anabolic pathways of lipids, amino acids, nucleotides and carbohydrates in allogeneic and syngeneic T cells compared to those in naïve T cells. In comparison with syngeneic T cells, allogeneic T cells exhibited increased oxidative stress, reflected by higher levels of eicosanoid, cyclooxygenase, and lipoxygenase-oxidized eicosanoids, and decreased levels of antioxidant compounds such as glutathione (GSH) and glutathione disulfide (GSSG). To obtain biomass for robust proliferation followed by alloantigen stimulation, allogeneic T cells further increased pentose phosphate and polyamine synthesis by day 7 post-BMT. We also observed that allogeneic T cells and syngeneic T cells expressed comparable levels of metabolites in fatty acid and glutamine oxidized in tricarboxylic acid (TCA) cycle, which was much lower than those of naïve T cells. Importantly, allogeneic T cells exhibited higher levels of metabolites in glycolysis as compared to syngeneic T cells regardless of time points. Consistently, using Seahorse approach, we also found that allogeneic T cells significantly increased aerobic glycolysis as compared to syngeneic T cells post-BMT, whereas oxidative phosphorylation was similar. Moreover, blocking glycolysis with 2-deoxyglucose remarkably inhibited donor T-cell proliferation, expansion and Th1 differentiation after allo-BMT. Thus, aerobic glycolysis rather than mitochondrial oxidative phosphorylation is the preferential metabolic process required for the optimal expansion and activation of allogeneic T cells. Given mechanistic target of rapamycin (mTOR) plays an essential role in controlling T-cell metabolism particularly in glycolysis, we hypothesized that targeting mTOR would prevent GVHD by inhibiting glycolytic metabolism. Using pharmacological and genetic approaches, we unequivocally demonstrated that mTOR, especially mTORC1, was essential for T-cell glycolytic activity and for GVHD induction. Mechanistically, mTORC1 promoted T-cell activation, expansion, Th1 differentiation, and migration into GVHD target organs, but inhibited the generation of induced T regulatory cells. In conclusion, the current work provides compelling evidence that allogeneic T cells utilize glycolysis as a predominant metabolic process after BMT. Furthermore, we validate glycolysis or its key regulator, such as mTORC1, to be a valid therapeutic target for the control of GVHD. Disclosures: No relevant conflicts of interest to declare.