ORP4L is a prerequisite for the induction of T-cell leukemogenesis associated with human T-cell leukemia virus 1

Human T-cell leukemia virus 1 (HTLV-1) causes adult T-cell leukemia (ATL), but the mechanism underlying its initiation remains elusive. Here we report that ORP4L is expressed in ATL cells but not normal T-cells. ORP4L ablation completely blocks T-cell leukemogenesis induced by the HTLV-1 oncoprotein Tax in mice while engineering ORP4L expression in T-cells results in T-cell leukemia in mice, suggesting the oncogenic properties and prerequisite of ORP4L for the initiation of T-cell leukemogenesis. For molecular insight, loss of miR-31 caused by HTLV-1 induces ORP4L expression in T-cells. ORP4L interacts with PI3Kδ to promote PI(3,4,5)P3 generation, contributing to AKT hyperactivation, NF-κB-dependent p53 inactivation induced pro-oncogenes expression and T-cell leukemogenesis. Consistently, ORP4L ablation eliminates human ATL cells in patient-derived xenograft ATL models. These results reveal a plausible mechanism of T-cells deterioration by HTLV-1 that can be therapeutically targeted.

The Histone Demethylase KDM6B Is a Genetic Dependency of NOTCH1-Driven T-ALL

T-cell acute lymphoblastic leukemia (T-ALL) arises from the accumulation of genomic abnormalities and the malignant proliferation of immature T-cells. Despite recent advancements in understanding the genetic alterations driving T cell leukemogenesis, patients still suffer from recurrent relapses and treatment-related toxicities. Genome sequencing has revealed significant heterogeneity and important insights into the genetic landscape of T-ALL. Mutations in epigenetic modifiers are frequently observed and serve as an attractive target for novel therapeutic approaches. Histone demethylase enzymes play a critical role in the regulation of gene expression programs in T-ALL. KDM6A (UTX) is known to behave as a tumor suppressor in most T-ALL subtypes. However, it's gene paralog, KDM6B (JMJD3), is never mutated and can be significantly overexpressed in patients. Here, we show that KDM6B is required for T-ALL initiation. Using genetic mouse models, Sca-1 enriched WBM from Vav-Cre: Kdm6b+/+, Vav-Cre: Kdm6bfl/+, and Vav-Cre: Kdm6bfl/fl adult mice was transduced with a retrovirus expressing Notch1 Intracellular Domain (NICD). NOTCH1 gain-of-function mutations are the most frequent driver events in adult T-ALL, and this model recapitulates many of the human pathologies. Transduced cells were transplanted into irradiated mice. While there was robust engraftment in all groups at four weeks post-transplant, T-ALL cells were not sustained in the genetic absence of Kdm6b. Mice receiving control NICD-GFP+ cells succumbed to T-ALL with median survival of 79 days, whereas the only mice receiving Kdm6b-null NICD-GFP+ cells that developed disease were found to retain one copy of the Kdm6b floxed allele (Fig 1). To investigate the translational potential, we targeted KDM6B for genetic inactivation by CRISPR/Cas9 in primary T-ALL patient cells, followed by xenograft into NSG mice. The effect of KDM6B targeting was quantified over time by monitoring the variant allele fraction (VAF) of the transplanted cells. In most patient samples, KDM6B-targeted cells were significantly outcompeted over time, thus further highlighting the requirement of KDM6B in sustaining T-ALL tumorigenesis. To examine the mechanism by which KDM6B sustain T-ALL cells, gene expression profiling was performed by RNA-seq on mouse T-ALL cells of genetic backgrounds Control, Kdm6b-Het, and Kdm6b-KO. Three gene sets were significantly downregulated in Kdm6b-deficient T-ALL cells compared to the Control group, all of which are involved in cell cycle processes. Additional validation of these findings with cell cycle functional studies is currently ongoing. Additionally, while Kdm6b has been described for its H3K27me3 histone demethylase function, recent studies have shown its involvement in regulating various gene expression programs through demethylase-independent mechanisms. To determine if the role of Kdm6b in T-ALL oncogenesis was related to it demethylase activity, we performed the same NICD retroviral transduction experiment with inclusion of a Kdm6b +/H1388Amouse genotype, a point mutation which renders Kdm6b catalytically dead. Our data shows that Kdm6b +/H1388Aphenocopies T-ALL with Kdm6b homozygous loss-of-function, showing no evidence of disease in the blood beyond 8-weeks post-transplant. We conclude from this that the leukemogenic role of Kdm6b requires it's H3K27me3 demethylase function. In summary, these data reveal Kdm6b as an oncogenic dependency in Notch1-driven T-ALL in both mouse and human systems, and present Kdm6b as a high value therapeutic target in adult T-ALL patients. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.

CongenitalSTAT3mutation Mediates Primary Persistent Polymorphic Inflammatory Activation and T Cell Leukemogenesis

Janus kinase/signal transducer and activator of transcription (JAK/STAT) pathways play a pivotal role in inflammation and immunity, among which, JAK/STAT3 pathway is the most potent and leads the crosstalk of immunity and oncogenesis. Somatic STAT3 activatingmutations have been found in about 40% of T cell large granular lymphocytic leukemia (T-LGLL) patients, most of which are located in exon 21 which encodes Src homology 2 (SH2) domain leading to the increased activity of aberrant STAT3 protein and the upregulation of its transcriptional targets. While germline STAT3activatingmutations represent a newly defined entity of immune dysregulations named infantile-onset multisystem autoimmune disease-1 (ADMIO1, #MIM 615952). Both the two diseases are rare and poorly understood. Here, we report a pedigree including a proband, a six-year-old girl, primarily manifesting as thrombocytopenia and lymphadenopathy and her father diagnosed as T-LGLL with pure red cell aplastic anemia without autoimmune disorders preceding or during his disease course. Morphology of the bone marrow smears of the proband indicated normal hyperplasia without evident dyspepsia or increased blast cells. However, the vacuoles in monocytes and the density and size of granules in neutrophils increased, and megaloblast transformation was observed in some neutrophils. (Fig. 1A, 1B) Biopsy of an enlarged lymph node showed the reactive follicular hyperplasia. (Fig. 1C) Whole exon sequencing and pedigree analysis of the family revealed the germline STAT3 c.833G>A/p.R278Hmutation harbored by the proband which originated de novo from her father who additionally carried a germline TAL1G62Rmutation and somatically accumulated an FLT3-ITD mutation. (Fig. 2) Through single-cell RNA sequencing, we also found the increase of circulating CD8+ T cells and the decrease of NK cells of the proband. (Fig. 3) The STAT3 target genes were generally overactivated, and the expression of cytokines decreased in transcription level. In the genes participating in JAK/STATs pathways, the expression of JAK3, STAT1, and STAT3was up-regulated significantly. (data not shown) Immunophenotype of the proband by flow cytometry confirmed change in immunocyte compartments, (Fig. 4) but the serum cytokine concentrations measured by flow cytometry yielded controversial results, that most of cytokines were moderately elevated, and IL-1β, IL-5, TNF-α, and IFN-γ were of the most evident. (data not shown) During the treatment and follow-up, Cyclosporin A (CsA) was efficient in maintaining her circulating platelets in the range of 166×109/L to 302×109/L, but the enlarged lymph nodes and hepatosplenomegaly had no response. Eleven months later, CsA was replaced by tacrolimusfor the severe gingival hyperplasia, which has efficiently stabilized her platelets count and normalized the enlarged lymph nodes, liver, and spleen. On the contrary, in the three and a half years' span of illness, the father was refractory to CsA and methotrexate (MTX), moreover, lethal bone marrow suppression was induced by one course of fludarabine. For the high level of HLA-I and HLA-II antibodies in the circulation, plantlets transfusions were only efficient after plasmapheresis. The father eventually died from pulmonary and gastrointestinal infection due to the failure of maternal HLA-haploidentical hematopoietic stem cell transplantation (HSCT). We comprehensively elaborated the immunophenotype of the proband and thoroughly elucidated the genetic alternations of the father which led to the T cell leukemogenesis, which brought new insight on these two rare diseases and highlighted a more scrupulous therapeutic strategy in T-LGLL with congenital mutations. Figure 1 Disclosures No relevant conflicts of interest to declare.

Direct Phosphorylation and Stabilization of MYC By Aurora B Kinase Promotes T Cell Leukemogenesis

The aberrant activation of oncogene MYC (also termed c-MYC) is one of the most common features in human cancer. The unleashed MYC oncogene frequently produces abundant MYC protein, which mediates a transcriptional response involved in a variety of biological processes, contributing to almost every aspect of tumorigenesis. The significance of MYC deregulation has been recognized in T-cell acute lymphoblastic leukemia (T-ALL), a life-threatening hematological malignancy with dismal outcome due to disease relapse and drug resistance. Therapeutic efforts aimed at inhibiting MYC expression or activity should have an important clinical relevance. However, attempts to directly disrupt MYC function have met with limited success, in part due to its "undruggable" protein structure. Aurora kinases, a multi-genic family of serine/threonine kinases, consist Aurora A (AURKA), Aurora B (AURKB) and Aurora C (AURKC). They are known to play an integral role in the regulation of cell division and chromosomal segregation. Amplification or overexpression of Aurora kinases is frequently found in human cancers with clear evidence of oncogenic potential, implicating Aurora kinases as rational anti-tumor targets. AURKB is the catalytic subunit of the chromosomal passenger protein complex (CPPC) which regulates multiple facets of cell division. Overexpression of AURKB has been reported in a variety of cancers and predicts poor overall survival. AZD1152 is a highly potent and selective inhibitor of Aurora B; preclinical evidence of anti-tumor efficacy with AZD1152 has extended to the clinical setting with tolerable toxicity. Despite considerable study, it remains largely unclear how expression of AURKB is elevated and, in particular, how elevated levels of AURKB reprogram cells to promote the cancer progression. We identified AURKB as a novel MYC binding partner using liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis. To assess the potential role of AURKB in regulating MYC, we inhibited AURKB in multiple T-ALL cell lines and found that MYC protein expression was diminished significantly. Notably, depletion of AURKB failed to downregulate MYC mRNA, suggesting a post-transcriptional regulation of the MYC protein. We then assessed MYC protein decay in the presence or absence of AURKB expression. Time course experiments revealed that knockdown of AURKB significantly shortened the half-life of endogenous MYC in T-ALL cells. Consistently, AURKB depletion resulted in significant downregulation of MYC-induced gene expression programs, corroborating an essential role of AURKB in sustaining MYC transcriptional activities. Regulation of MYC degradation by ubiquitin-proteasome system is dependent on MYC phosphorylation. Initiation of MYC turnover requires GSK3b-mediated phosphorylation of threonine 58 (p-T58). Time-course analysis showed that treatment with AZD1152 in T-ALL cells resulted in a notable increase in MYC p-T58. We performed in vitro kinase assays and found that MYC is phosphorylated directly by AURKB. This phosphorylation counteracted GSK3b-directed Thr58 phosphorylation and subsequent FBW7-mediated proteasomal degradation and enhanced the MYC protein stability. The AURKB mRNA and protein are generally more abundant in human T-ALL than normal thymocytes. We demonstrated that MYC, in concert with TAL1, directly binds to the AURKB promoter and activates its transcription, thus constituting a reciprocal regulation between MYC and AURKB. In order to clarify the biological significance of the AURKB-MYC circuit, we co-injected the one-cell-stage zebrafish embryos with the rag2:EGFP-Myc construct alone or together with rag2:AURKB followed by fluorescent microscopy analysis. Our results demonstrate that AURKB phosphorylation of MYC is functionally important for T cell leukemogenesis in vivo. Moreover, inhibition of AURKB elicits dramatic MYC degradation in association with robust apoptosis in wild type-FBW7 T-ALL cells, suggesting that FBW7 mutational status may serve as a genetic predictor of sensitivity to AURKB inhibitors. In this study, we unravel a non-mitotic role of AURKB in promoting MYC stabilization via direct MYC phosphorylation at an unconventional site. Our findings decipher a previously unsuspected mechanism involved in MYC deregulation and highlight disruption of the AURKB-MYC signaling circuit as a promising T-ALL therapeutic strategy. Disclosures No relevant conflicts of interest to declare.

KDM2B in polycomb repressive complex 1.1 functions as a tumor suppressor in the initiation of T-cell leukemogenesis

KDM2B together with RING1B, PCGF1, and BCOR or BCORL1 comprise polycomb repressive complex 1.1 (PRC1.1), a noncanonical PRC1 that catalyzes H2AK119ub1. It binds to nonmethylated CpG islands through its zinc finger-CxxC DNA binding domain and recruits the complex to target gene loci. Recent studies identified the loss of function mutations in the PRC1.1 gene, BCOR and BCORL1 in human T-cell acute lymphoblastic leukemia (T-ALL). We previously reported that Bcor insufficiency induces T-ALL in mice, supporting a tumor suppressor role for BCOR. However, the function of BCOR responsible for tumor suppression, either its corepressor function for BCL6 or that as a component of PRC1.1, remains unclear. We herein examined mice specifically lacking the zinc finger-CxxC domain of KDM2B in hematopoietic cells. Similar to Bcor-deficient mice, Kdm2b-deficient mice developed lethal T-ALL mostly in a NOTCH1-dependent manner. A chromatin immunoprecipitation sequence analysis of thymocytes revealed the binding of KDM2B at promoter regions, at which BCOR and EZH2 colocalized. KDM2B target genes markedly overlapped with those of NOTCH1 in human T-ALL cells, suggesting that noncanonical PRC1.1 antagonizes NOTCH1-mediated gene activation. KDM2B target genes were expressed at higher levels than the others and were marked with high levels of H2AK119ub1 and H3K4me3, but low levels of H3K27me3, suggesting that KDM2B target genes are transcriptionally active or primed for activation. These results indicate that PRC1.1 plays a key role in restricting excessive transcriptional activation by active NOTCH1, thereby acting as a tumor suppressor in the initiation of T-cell leukemogenesis.

The SUMO Pathway in Hematomalignancies and Their Response to Therapies

SUMO (Small Ubiquitin-related MOdifier) is a post-translational modifier of the ubiquitin family controlling the function and fate of thousands of proteins. SUMOylation is deregulated in various hematological malignancies, where it participates in both tumorigenesis and cancer cell response to therapies. This is the case for Acute Promyelocytic Leukemias (APL) where SUMOylation, and subsequent destruction, of the PML-RARα fusion oncoprotein are triggered by arsenic trioxide, which is used as front-line therapy in combination with retinoic acid to cure APL patients. A similar arsenic-induced SUMO-dependent degradation was also documented for Tax, a human T-cell lymphotropic virus type I (HTLV1) viral protein implicated in Adult T-cell Leukemogenesis. SUMOylation also participates in Acute Myeloid Leukemia (AML) response to both chemo- and differentiation therapies, in particular through its ability to regulate gene expression. In Multiple Myeloma, many enzymes of the SUMO pathway are overexpressed and their high expression correlates with lower response to melphalan-based chemotherapies. B-cell lymphomas overexpressing the c-Myc oncogene also overexpress most components of the SUMO pathway and are highly sensitive to SUMOylation inhibition. Targeting the SUMO pathway with recently discovered pharmacological inhibitors, alone or in combination with current therapies, might therefore constitute a powerful strategy to improve the treatment of these cancers.

Role of KDM2B and Non-Canonical PRC1.1 As a Tumor Suppressor in T-ALL

Introduction: Polycomb repressive complexes (PRCs) play an important role for the transcriptional repression of their target genes through histone modification. KDM2B is a component of non-canonical PRC1.1 and has a role for the recruitment of the complex to the target gene loci. It has a zinc finger-CxxC (ZF-CxxC) domain which specifically binds to unmethylated sequences in CpG islands (CGIs) and deletion of the CxxC domain induces complete loss of KDM2B occupancy and removal of other PRC1.1 components from CGIs. Recent studies revealed that loss of function mutations of several PRC component genes such as EZH2, EED, SUZ12 and BCOR were frequently detected in human T-cell acute lymphoblastic leukemia (T-ALL), which suggested PRCs have a tumor suppressive role in T cell development. Our group have reported conditional knock out of Bcor, encoding a component protein of non-canonical PRC1.1, induced T-ALL in mice. However, it is still unknown how KDM2B and non-canonical PRC1.1 regulate T cell leukemogenesis. Therefore, we performed detailed analysis of mice deficient for Kdm2b ZF-CxxC domain (Kdm2bΔCxxC/ΔCxxC) specifically in hematopoietic cells. Methods: We used the conditional Kdm2b allele (Kdm2bfl) mice, which contains LoxP sites flanking Kdm2b exon 13 encoding the ZF-CxxC domain. To generate hematopoietic cell specific Kdm2b KO mice, we transplanted Kdm2bfl/fl;Cre-ERT total BM cells into lethally irradiated CD45.1+ recipient mice and deleted Kdm2b 4 weeks after transplantation by intraperitoneally injecting of tamoxifen. Results: During the observation period of 300 days, almost all of the Kdm2b KO mice developed lethal T-ALL. They showed thymomegaly and splenomegaly and presented infiltration of donor-derived leukemic cells into the bone marrow, spleen, thymus and peripheral blood. Flow cytometric analysis revealed that T-ALL cells were mainly CD4 and CD8-double positive (DP). Notch1 active mutations in exons 26, 27, 28 or 34 were found in over a half of the T-ALL cases, indicating the Notch1 activation could be a driver for leukemic transformation in this mouse model. RNA sequence analysis of the DP cells revealed activation of Myc, which plays a key role in the development of T-ALL, and their downstream target genes in Kdm2b-deficient T-ALL. ChIP sequence analysis of DP thymocytes expressing 3xFLAG-KDM2B confirmed the binding peaks of KDM2B at the promotor of Myc. Peak calling analysis of the ChIP sequence data revealed that KDM2B was mainly located at transcript start sites (TSS), where KDM2B was co-localized with H2AK119ub1 and H3K27me3 histone marks. In addition, ChIP sequencing of H3K4me3 revealed that the KDM2B target genes include more bivalent genes than non-target genes. We next compared histone modification status around TSS in WT and Kdm2b KO DP cells and Kdm2b-deficient T-ALL cells. Global levels of H2AK119ub1 were significantly decreased in T-ALL cells and the reduction was mainly observed at the promoters of KDM2B target genes. Direct target genes of NOTCH1 including Myc also lost H2AK119ub1 at their promotors in T-ALL cells. Peak calling analysis of KDM2B, BCOR and NOTCH1 ChIP sequence data revealed that their target genes were closely overlapped at the promotor region. Moreover, EZH2 binding peaks were also overlapped with those of KDM2B and NOTCH1, suggesting that non-canonical PRC1.1 and PRC2 cooperatively antagonize NOTCH1-mediated gene activation. KEGG pathway analysis of the genes with overlapping binding peaks among KDM2B, EZH2 and NOTCH1 showed significant enrichment of T cell receptor signaling, Notch1 signaling and cell cycle pathway, all of which play an important role for the development of T-ALL. Of interest, H3K27me3 levels of the common target genes of EZH2, KDM2B, and NOTCH1 were much lower than the EZH2-specific target genes, indicating that H2AK119ub1 plays a key role in the repression of NOTCH1 targets. Conclusions: Our findings suggest that KDM2B recruits non-canonical PRC1.1 at the promotor regions of NOTCH1 targets to restraint thymocytes from transformation in concert with PRC2. Disclosures No relevant conflicts of interest to declare.