Supplementary Figure 6. Anti-CD40/ICB-induced rejection of V6K is not mediated by macrophages or myeloid cells.

Supplementary Figure 2. Knockout tumor cell lines lack expression of antigen presentation molecules.

Supplementary Figure 7. Stromal cell gating for flow cytometry.

Within days of exposure to chronic viral infections, activated CD8+ T cells differentiate into Tcf1-Slamf6loTim3hi exhaustion-prone effector T (TEX_EFF) cells or self-renewing Tcf1+Slamf6hiTim3lo precursor exhausted T (TPEX) cells. Here we showed that early CD8+ TEX cell fates were imprinted by forming subset-specific, self-associating chromatin hubs. Chromatin hub assembly coincided with effector or stemness gene induction and identified the transcription cofactors Id2 and Id3 as key regulators that promoted CD8+ TEX_EFF and CD8+ TPEX cell fates, respectively. Id2 drove CD8+ TEX_EFF cell specification by activating effector genes, while suppressing genes involved in exhaustion and stemness. In contrast, Id3-repressed effector genes but upregulated IL-7Rα and AhR, thereby maintaining the CD8+ TPEX cell pool. Mechanistically, Id2 and Id3 exhibited a distinct impact on the chromatin accessibility landscape in early CD8+ TEX cells by engaging Runx3 and Tcf1 transcription factors along with E proteins. These findings indicated that reshaping chromatin architecture represents a critical means for specifying CD8+ TEX cell fates and ensuring lineage stability.

Supplementary Figure 3. Memory immune cells reject secondary tumors.

Supplementary Figure 8. Immunofluorescent (IF) gating strategy and quantification for ECs and CAFs.

Supplementary Table 1. Myeloid cell depletion strategies.

Supplementary Figure 1. Anti-CD40/ICB immunotherapy slows tumor growth in a CD4+ T cell-dependent, but CD8+ T cell- and NK1.1 cell-independent manner.

In graft-versus-host disease (GVHD), Ca2+ signals in alloreactive T cells are carefully controlled to determine whether cells survive or thrive, although how this is accomplished during GVHD remains poorly defined. We demonstrate that EZH2, a chromatin-modifying enzyme, promotes alloreactive T-cell survival in GVHD by acting as a Ca2+ signaling brake to limit excessive intracellular Ca2+ responses. Ezh2 loss led to the upregulation of gene programs that promote effector differentiation in activated T cells, coincident with enhanced intracellular Ca2+ responses that ultimately caused massive cell death. Conditional deletion of Stim1 (required for cytosolic Ca2+ entry) led to “synthetic rescue” of Ezh2-null T cells by protecting them from cell death without interfering with effector differentiation, resulting in severe GVHD. Interestingly, Stim1 expression was unaffected by EZH2, whereas the expression of the endoplasmic reticulum Ca2+ release channel inositol 1,4,5-trisphosphate receptor 2 (Itpr2) was suppressed by EZH2. Notably, EZH2 and Ca2+ signals served mutually opposing roles in controlling the expression of genes in chimeric antigen receptor (CAR) T cells. Inhibiting Ca²⁺ signaling restored EZH2 function in CAR-T cells, significantly improving their antitumor activity. Our findings reveal the interdependent roles of EZH2 and Ca2+ signals in coordinating antigen-activated T-cell responses that mediate alloimmunity and tumor immunity.

Supplementary Table 1. Myeloid cell depletion strategies.

Supplementary Figure 4. Immunotherapy-induced rejection of V6K is not mediated by perforin or FasL.

Living cell therapy for immune regulation could potentially achieve long-lasting effects on inhibiting graft-versus-host disease (GVHD), a life-threatening complication after allogeneic hematopoietic cell transplantation (allo-HCT). However, novel immune regulatory cells with better persistence and sustained immune modulatory function are needed. Here, we demonstrate that engineering human T cells to co-express IFN-α2 and PD-L1 (named αp-T cells) rendered them a potent capacity to inhibit xenogeneic GVHD without impairing anti-leukemia in immunodeficient mice. These αp-T cells maintained stable expression of PD-L1 and IFN-α and potent immunosuppressive effects in peripheral tissues during GVHD control. CD4+ αp-T cells activated transcriptional programs that promoted effector differentiation, exhaustion, and proliferation inhibition in both themselves and their treated conventional T (conv-T) cells, ultimately reducing GVHD. These CD4+ αp-T cells created a synergistic regulatory loop that modulated conv-T cell responses, in which IFN-α suppressed expansion and survival of activated conv-T cells, promoted their differentiation into PD-1+TIM3+ exhaustion-like T cells, and sensitized them to PD-L1-mediated suppression. We further verified these regulatory effects of murine CD4+ αp-T cells on reducing GVHD in immunocompetent mice of allo-HCT. Thus, αp-T cells may represent a novel translational strategy to improve the safety and efficacy of allo-HCT and inhibit other inflammatory disorders in a broad context.

Acute myeloid leukemia (AML) faces major clinical challenges characterized by high relapse and low cure rates. Chimeric antigen receptor (CAR) T cell therapy has revolutionized the treatment for B cell malignancies, but it has limited efficacy in treating AML. Poor anti-tumor efficacy is often attributed to tumor resistance mechanisms and T cell dysfunction. The dysregulation of cell death in tumor cells plays a crucial role in promoting tumor resistance to immunotherapies. However, effective strategies that can disrupt this resistance mechanism remain an unmet need. Immunogenic cell death, such as pyroptosis, promotes effective innate and adaptive immunity. Pyroptosis is triggered by inflammasome activation and executed by caspase (CASP) and gasdermin (GSDM) proteins. Here we demonstrate that a previously unrecognized role of the NLRP3 inflammasome-pyroptosis signaling in sensitizing AML cells to CAR T cell killing. To test the hypothesis that epigenetic therapy agents that can activate the inflammasome-pyroptosis signaling in target tumor cells could potentially enhance therapeutic efficacy of CAR T cells against AML, we used a hypothesis-driven drug screening assay and identified that the DNA methylation inhibitor decitabine (DAC) induced high levels NLRP3 in murine AML C1498 cells. Bulk RNA-sequencing (RNA-seq) analysis revealed that DAC pre-treatment of murine C1498 AML cells induced expression of genes that were associated with inflammasomes and cell death molecules (Nlrp3, Il1b, Casp1, Casp4, Fas). Pretreatment of murine C1498 AML cells expressing human CD19 (C1498hCD19 cells) with DAC significantly enhanced their sensitivity to murine BBz-CAR T cells directed against human CD19 (hCD19-CAR T cells). Furthermore, DAC pretreatment of immunocompetent mice bearing C1498hCD19 cells, resulted in significantly enhanced in vivo antitumor activity of hCD19-CAR T cells, improving the overall disease-free survival. This beneficial effect was achieved in a second AML mouse model induced by MLL-AF9 fusion gene (MAF9). Mechanism studies showed that DAC priming of leukemia mice promoted the expansion of memory-like CAR T cells but restrained the generation of exhaustion-like CAR T cells. As a result, DAC-primed C1498 leukemia-bearing mice generated and maintained memory CAR T cells with robust self-renewal capability and long-term immunosurveillance capability. In the canonical NLRP3-CASP1-GSDMD pathway, damage-associated molecular patterns (DMAPs) induce NLRP3 oligomerization to activate CASP1. We observed that pharmacological inhibition of NLRP3 inflammasome-pyroptosis signaling, including NLRP3, GSDMD and pan-caspases, in DAC-primed tumor cells increases their resistance to CAR T cell cytotoxicity. GSDMD is the down-stream effector of the NLRP3 inflammasome pathway. Intriguingly, ablating GSDMD in C1498hCD19 cells enabled them to resist CAR T cell therapy in vivo, indicating that the NLRP3-GSDMD-mediated pyroptosis plays crucial roles in CAR T cell eradication of tumors in DAC-primed AML-bearing mice. Finally, we found that DAC pretreatment of human AML THP-1 cells activated their expression of inflammasome gene program and enhanced their sensitivity to human CD33-CAR T cell cytotoxicity. In vivo DAC-priming of human xenograft THP-1 AML-bearing immunodeficient NSG mice resulted in significantly augmented therapeutic efficacy of human CD33-CAR T cells, leading to significantly overall survival rates of these leukemia mice. Collectively, our findings identified that the inflammasome-pyroptosis pathway in tumor cells to make them more vulnerable to CAR T cell cytotoxicity, thereby reducing tumor resistance and its-induced CAR T cell dysfunction. They also suggest that novel and translational strategies that can selectively activate the inflammasome-pyroptosis signaling in tumor cells could potentially enhance therapeutic efficacy of CAR T cells.

Graft-versus-host disease (GVHD) remains a major concern following allogeneic hematopoietic stem cell transplantation (allo-HSCT). Two principal in vivo T-cell depletion (TCD) strategies-post-transplant cyclophosphamide (PTCY)-based and anti-thymocyte globulin (ATG)-based GVHD prophylaxis-have effectively mitigated this challenge. In recent years, PTCY has expanded beyond haploidentical HSCT to include matched related donor (MRD), matched unrelated donor (MUD) and mismatched unrelated donor (MMUD) HSCT, resulting in favorable outcomes. ATG-based regimens have been optimized in terms of dosing and timing across various transplant settings. Individualized ATG administration shows potential in overcoming the highly heterogeneous pharmacokinetics. Some prospective studies have explored the combined use of ATG and PTCY in haploidentical and MUD HSCT. Only a few randomized controlled trials (RCTs) have compared ATG and PTCY head-to-head, and high-level evidence remains scarce. Ongoing clinical trials are expected to clarify which in vivo T-cell depletion protocol is better, and whether adjustments in timing, dose, or combination can yield better outcomes for patients.

Despite significant advances in graft-versus-host disease (GVHD) prevention and treatment, calcineurin inhibitor (CNI)-based standard GVHD prophylaxis in allogeneic hematopoietic stem cell transplantation (allo-HSCT) has limited efficacy in controlling acute and chronic GVHD. Thus, inhibition of calcium (Ca2+) signaling is insufficient to suppress the generation and maintenance of alloreactive T cells that mediate host tissue injury. Recent studies suggested that CNI-dependent alloreactive T cells possess great ability to persist and mediate chronic-like GVHD in mice. The molecular events by which these T cells breakthrough CNI inhibition have not been previously defined. Ezh2, a chromatin-modifying epigenetic regulator, silences expression of gene programs critical for multiple cellular processes. Ablating Ezh2 in T cells inhibits GVHD and anti-tumor activity, largely due to massive antigen-activated T cell death. Increased Ca2+ signals in activated T cells are known to induce their cell death and dysfunction. However, the relationship between Ezh2 and intracellular Ca2+ response generation in GVHD has never been previously examined. We report here that Ezh2 and Ca2+-mediated signals operate coordinately to regulate the viability and effector function of GVHD T cells. Blockade of Ca2+ signal by conditional deletion of Stim1, an endoplasmic reticulum (ER) Ca2+ sensor required for Ca2+ entry in T cells, rescued non-viable Ezh2-null alloreactive T cells, as well as restored their capacity to mediate GVHD in mice after allo-HSCT. Moreover, while STIM1-null T cells typically exhibitdecreased effector differentiation and function of GVHD T cells, this was restored by deletion of Ezh2 in Stim1-null T cells. These data identify the interdependent roles of Ezh2 and Ca2+ signals in activation, effector differentiation and survival of alloreactive T cells. To understand how Ezh2 acts as 'brake“ for Ca2+ signals in T cells, we performed bulk-RNA-sequencing analysis on Ezh2/STIM1 dual knockout T cells. Ezh2 directly repressedexpression Itpr2, which encodes the ER Ca2+ release channel 1,4,5-trisphosphate receptor (IP3R2), thereby interfering with ER Ca2+ release and subsequent cytosolic Ca2+ entry. Combined deletion of Ezh2 and Itpr2 genes restored the inability of allogeneic Ezh2-null T cells to induce lethal GVHD. Furthermore, the co-dependence of Ezh2 and Iptr2 were similarly observed in CD19-directed CAR-T mediated elimination of CD19-expressing C1498 acute myeloid leukemia in mice. Itpr2 loss in Ezh2-null CAR-T cells led to their improved survival, expansion, and production of IFN-g-producing effector CAR-T cells. Collectively, our findings identify that Ezh2 suppresses the expression of Iptr2 to prevent excessive Ca2+ signal generation and antigen-driven T cell death and T cell dysfunction. These observations reveal a potential therapeutic window for the treatment of GVHD focused on increasing intracellular Ca2+ signals to eliminate alloreactive T cells, which is opposing to the current concept of CNI treatment. Furthermore, targeting this Ezh2-Itpr2 axis may have broad implications in the regulation of other types of antigen-driven T cell responses, such as anti-tumor immunity, autoimmunity and graft rejection of solid organ transplantation.

More than 50% of B cell malignancy patients who respond to chimeric antigen receptor (CAR) T cell therapies eventually relapse, and CAR T cell therapies for solid tumors have been largely ineffective. Clinical observations associate dysregulated expression of cell death molecules (e.g., FAS, FADD, BID and BCL2) in malignant B cells with poor outcomes in CAR-T cell-treated patients. Thus, innovative strategies that can enhance the sensitivity of tumor cells to CAR T cell cytotoxicity, potentially through modulation of cell death signaling, may improve efficacy of CAR T cell treatment. Cyclin Dependent Kinase 9 (CDK9) is a global regulator of gene transcription and repression. While it is often dysregulated in hematological malignancies, inhibition of CDK9 in lymphoma and leukemia cells abrogates pro-survival gene programs and induces apoptosis. Given the strong correlation between CDK9 inhibition and cell death induction, we hypothesized that CDK9 inhibition could sensitize B malignant cells to CAR T cell cytotoxicity, thereby improving the therapeutic efficacy. We report here that CDK9 inhibitor (CDK9-i) pre-treatment increased the sensitivity of B cell leukemia cells (Raji and Nalm6) to CD19-CAR T cell cytotoxicity. Furthermore, in vivo administration of CDK9-i to human B cell leukemia-bearing NOD/SCID/IL2Rg−/− (NSG) mice significantly enhanced efficacy of CAR T cell therapy, by limiting the generation of PD-1+TIM3+ terminally exhausted CAR T cells, enhancing elimination of leukemia, and significantly improving overall survival. Bulk RNA-sequencing analysis revealed that CDK9-i pretreatment induced expression of genes associated with IFN-gamma response, TNFA signaling, inflammatory response and IFN-alpha response. Among these genes, cell death receptor signaling genes (TNFR1B, TNFRSF1A and IFNGR2) and cell death mediator genes (Caspase 8 (CASP8) and Gasdermin E (GSDME)) were significantly upregulated. Western Blot analysis revealed that CDK9-i priming and CAR T cell treatment significantly increased expression and cleavage of CASP8 and GSDME. These data suggest activation of cell apoptosis and pyroptosis in leukemia cells treated by CDK9-i. Competition assays showed that ablation of CASP8/MLKL or GSDME in CDK9-i-treated leukemia cells resulted in markedly increased resistance to CAR T cell killing in cultures. In vivo, NSG mice receiving CASP8/MLKL DKO cells were resistant to CDK9-i treatment, succumbing to disease a month before endpoints were reached for mice receiving WT Raji cells and CDK9-i treatment. Mechanistic studies revealed that CDK9-i regulated expression of CASP8 and GSDME were associated with activation of STAT5 and STAT3 signaling pathways in Raji cells. Inhibition of either STAT3 or STAT5 shows consistent decrease in expression of GSDME, CASP8 and FAS. These data indicate that CDK9-i treatment sensitizes leukemia cells to CAR T cell cytotoxicity via rewiring the expression and activity of cell death molecular pathways mediating apoptosis and pyroptosis. Importantly, we found that CDK9-i treatment-reprogrammed cell death pathways were associated with enhanced bystander killing of CD19-negative B cell leukemia cells likely through a mechanism of upregulating the expression of cell death signaling receptors such as FAS, TNFR1, TRAILR. Our findings demonstrate that CDK9-i-priming may potentiate elimination of tumor cells lacking specific antigen through a mechanism of enhanced bystander effect. In summary, combined CDK9-i treatment and CAR T cell therapy may have significant and broad implications in improving the efficacy of cellular immunotherapy against cancer.

Current strategies to prevent and treat graft-versus-host disease (GVHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT) primarily utilize pan-immune suppressive agents. These pharmacological approaches cause nonspecific immune suppression, require long-term treatment and are restricted by drug toxicity. Cellular therapies such as regulatory T cells and mesenchymal stem cells target alloreactive T cells to prevent lethal GVHD in humans without systemic immunosuppression. However, they also have major challenges, including cell purification, ex vivo expansion to an adequate amount to meet the therapy need, and loss of immune suppressive effects. New cellular therapies are needed to address these challenges. Our previous studies and others have shown that adoptive transfer of murine plasmacytoid dendritic cells (pDCs) can suppress alloreactive T cell responses and GVHD. These pDCs produce high levels of IFN-α and PD-L1 that directly suppress alloreactive murine T cell responses; however, they are a rare population in peripheral blood and also require purification and ex vivo expansion to meet the therapy requirement. We therefore hypothesize that delivering immune suppressive molecules derived from human pDCs using alternative hematopoietic cells could lead to the development of a novel living cell therapy for inhibiting GVHD. We report here that IFN-α2a- and PD-L1-overexpressing human T cells (αp-T cells) acquire potent capacity to inhibit xenogeneic GVHD (x-GVHD) but preserve the ability to eliminate human xenograft leukemia in NOD/SCID/IL2Rg−/− (NSG) mice. Our initial studies showed that a lower level of pDCs or PDL1+ pDCs in donor G-CSF-mobilized allografts correlated with a significantly higher risk of severe GVHD in patients undergoing allo-HSCT. Like their mouse counterparts, human pDCs expressed high levels of IFN-α and PD-L1, and dose-dependently suppressed expansion and survival of TCR-activated autologous CD4+ T cells in cultures. To assess whether human T cells transduced with IFN-α and PD-L1 may acquire immune suppressive functionality, we produced lentivirus encoding IFN-α2a and PD-L1 and transduced them into human T cells (both CD4 and CD8 T cells) to create αp-T cells. While 80% of NSG mice receiving vector control T cells died from x-GVHD, adoptive transfer of these αp-T cells (purity: 60~75%) significantly inhibited x-GVHD with all surviving over 70 days. Compared to vector control, αp-T cells showed 25-fold fewer total donor T cells and a 15-fold decrease in IFN-γ+ effector CD4+ T cells in peripheral blood at day 46 after transplantation, and had about 5-fold fewer donor CD4+ T cells in the bone marrow and spleen at the study's endpoint. The reduction of donor CD4+ T cells in αp-T cell-treated x-GVHD NSG mice was accompanied with increased cell death and elevation of PD-1+TIM3+ terminal exhaustion-like T cells. Furthermore, PD-L1-expressing αp-T cells persisted in these NSG mice throughout the observation period of 85 days. These data indicate that infused αp-T cells can reduce the survival and persistence capacity of x-GVHD-mediated effector T cells in NSG mice. In an immune-competent setting of allo-HSCT using a C57BL/6 mouse anti-Balb/c GVHD model, we verified that adoptive transfer of murine αp-T cells also significantly inhibited the production of lethal GVHD, with 50% of these recipients survived at 60 days after transplantation. Importantly, treatment by co-transfer of αp-T cells and vector control T cells eliminated human xenograft leukemia in NSG mice challenged by human Raji cells without causing x-GVHD. Notably, CD8+ T cells derived from αp-T cell-treated leukemia-bearing mice produced approximately 2-fold higher frequency of IFN-γ than those T cells from control group. Our findings identified for the first time that treatment with human αp-T cells may represent a new and clinically relevant cell therapy strategy to reduce GVHD while preserving potent graft-versus-leukemia effects, resulting in significantly improved overall survival of leukemia-bearing mice. Ongoing studies are exploring the effect of these αp-T cells on allogeneic CAR-T cell induction of GVHD and anti-tumor activity. If successful, αp-T cells may have additional applications to the utilization of allogeneic CAR T cells to eliminate tumors without causing GVHD in the context of allo-HSCT.

During the early/intermediate phases of the immune response, calcium (Ca2+) signals are crucial for T cell activation, proliferation and effector differentiation. Yet, inhibiting Ca2+ signaling-activated master transcription factor NFAT has achieved limited success in the long-term restriction of graft-versus-host disease (GVHD), while high remission and infection rates remain major issues. This underscores the need to better understand how different calcium channels coordinate to regulate T cell alloimmunity. Mitochondria is one of the two intracellular calcium stores that buffers and modulates cytosolic Ca2+ response. Mitochondria calcium uniporter (MCU) complex is the sole channel through which Ca2+ enters the mitochondria. The role of MCU in T cell-mediated GVHD remains undetermined. In this study, we report that enforced Mcu expression in donor T cells abolished their capacity to induce lethal GVHD. Using retroviral gene delivery system to induce or constitutively overexpress Mcu in T cells, we discovered that Mcu ectopic expression impaired donor T cell survival. RNA-seq analysis identified the activation of gene programs mediating death of activated T cells. This was induced by persistent alloantigen exposure (restimulation-induced cell death, RICD), leading to massive cell death in the liver, a GVHD target organ. Pharmacological enhancement of MCU function with a natural compound phenocopied this finding while preserving anti-leukemia potency. These data indicate that boosting MCU function in donor T cells could be a novel and effective strategy to mitigate GVHD after allogeneic hematopoietic stem cell transplantation. Interestingly, we discovered T cells naturally downregulated MCU function during differentiation, indicating intrinsic repressive mechanisms that protect T cells from RICD. This led us to interrogate whether MCU is required for T cell alloimmunity. We generated T cell-specific Mcu conditional knockout C57/BL6 mice (Mcu-cKO). Balb/c recipients infused with Mcu-KO B6 mouse T cells developed severe liver GVHD, manifested by the dramatically reduced liver size, bile duct lesion, portal and lobular inflammation associated with massive lymphocyte infiltration. However, these Mcu-KO T cell recipients survived longer with 69% compared to 100% mortality in WT T cell recipients. Mechanistic studies showed Mcu-KO recipients had significantly fewer donor T cells in the spleen and liver, attributed to reduced proliferation capacity; and decreased IFN-g-producing cells. Meanwhile, Mcu-KO donor T cells retained GMCSF- and granzyme B-producing capacity. These data demonstrate MCU promotes T cell alloresponse, distinguishing its dispensable role in autoimmune disease and anti-viral infection models. To delineate the molecular mechanisms through which MCU regulates T cell alloresponse, we performed transcriptome profiling on sort-purified alloreactive CD8 T cells recognizing the alloantigen H60 in balb/b mice. Mcu-KO alloreactive CD8 T cells were characterized with enhanced effector programming, loss of memory potential, positive enrichment of exhaustion feature through gene set enrichment and DEG analysis. In depth mechanistic analysis unveiled STAT5 as the master upstream regulator through which MCU modulated transcriptome. Elevated STAT5 and the expression of its target genes (Bcl2, Tox, Lef1, Gzma, etc) were observed in Mcu-KO CD8 T cells. STAT5 is a predominant IL2 signaling molecule. Mcu deficiency increased the level of CD25, which is required for the formation of high affinity IL2 receptor in CD8 T cells. IL2a (CD25) is transcriptionally activated by NFkb, the elevated activity of which was identified in Mcu-KO CD8 T cells. These observations suggest that Mcu ablation led to activation of IL2-CD25-STAT5 axis, sustaining the persistence of activated T cells while reinforcing effector programs. Findings from this study elucidate the intricate regulatory network through which MCU regulates T cell alloimmunity, highlighting the fundamental translational potential of enhancing MCU functionality in alloreactive T cells through gene editing or pharmacological approach to reduce GVHD.

Natural killer (NK) cells are the promoters in graft-versus-host disease (GVHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT), while demethylation can regulate NK cell function. We explored the mechanism of demethylation regulating NK cell function to affect GVHD after allo-HSCT. BALB/c mice were transfused with C57BL/6 mouse-derived NK and bone marrow cells to establish GVHD models, followed by isolation and in-vitro expansion of NK cells. NK cell purity, cytokine levels, proliferation, and cytokine-producing NK cell levels were measured via flow cytometry. KIR2DL1/2/3 methylation was tested by Methylation-specific polymerase chain reaction (MSP), with determination of mouse survival and GVHD scores. KIR2DL1/2/3 and DNMT1 expression was detected through qRT-PCR and/or western blot. Methylation levels were upregulated and KIR2DL1/2/3 expression was downregulated in GVHD mouse model-derived NK cells following IL-2 stimulation. DNMT1 silencing promoted KIR2DL1/2/3 expression, proliferation, and the secretion of Granzyme, Perforin, and Interferon-γ (IFN-γ) in C57BL/6 mouse-derived NK cells. DNMT1 silencing also enhanced mouse survival, reduced GVHD scores, promoted KIR2DL1/2/3 expression on the NK cell surface, and increased the secretion of Granzyme, Perforin, IFN-γ, and the number of cytokine-producing NK cells in the spleen, liver, and lung tissues of the models. Collectively, DNMT1 silencing induced KIR2DL1/2/3 expression in NK cells through reducing methylation to alleviate GVHD after allo-HSCT.

Persisting alloreactive donor T cells in target tissues are a determinant of graft-versushost disease (GVHD), but the transcriptional regulators that control the persistence and function of tissue-infiltrating T cells remain elusive. We demonstrate here that Id3, a DNAbinding inhibitor, is critical for sustaining T-cell responses in GVHD target tissues in mice, including the liver and intestine. Id3 loss results in aberrantly expressed PD -1 in polyfunctional T helper 1 (Th1) cells, decreased tissue-infiltrating PD -1+ polyfunctional Th1 cell numbers, impaired maintenance of liver TCF-1+ progenitor-like T cells, and inhibition of GVHD. PD -1 blockade restores the capacity of Id3-ablated donor T cells to mediate GVHD. Single-cell RNA-sequencing analysis revealed that Id3 loss leads to significantly decreased CD28- and PI3K/AKT-signaling activity in tissue-infiltrating polyfunctional Th1 cells, an indicator of active PD-1/PD-L1 effects. Id3 is also required for protecting CD8+ T cells from the PD -1 pathway-mediated suppression during GVHD. Genome-wide RNAsequencing analysis reveals that Id3 represses transcription factors (e.g., Nfatc2, Fos, Jun, Ets1, and Prdm1) that are critical for PD -1 transcription, exuberant effector differentiation, and interferon responses and dysfunction of activated T cells. Id3 achieves these effects by restraining the chromatin accessibility for these transcription factors. Id3 ablation in donor T cells preserved their graft vs tumor effects in mice undergoing allogeneic hematopoietic stem cell transplantation. Furthermore, CRISPR/Cas9 knockout of ID3 in human CD19-directed chimeric antigen receptor T cells retained their antitumor activity in NOD/SCID/IL2Rg-/- mice early after administration. These findings identify that ID3 is an important target to reduce GVHD, and the gene-editing program of ID3 may have broad implications in T-cell-based immunotherapy.