Search

Understanding the biological and clinical impact of copy number aberrations (CNAs) on the development of precision therapies in cancer remains an unmet challenge. Genetic amplification of chromosome 1q (chr1q-amp) is a major CNA conferring an adverse prognosis in several types of cancer, including in the blood cancer multiple myeloma (MM). Although several genes across chromosome 1 (chr1q) portend high-risk MM disease, the underpinning molecular etiology remains elusive. Here, with reference to the 3-dimensional (3D) chromatin structure, we integrate multi-omics data sets from patients with MM with genetic variables to obtain an associated clinical risk map across chr1q and to identify 103 adverse prognosis genes in chr1q-amp MM. Prominent among these genes, the transcription factor PBX1 is ectopically expressed by genetic amplification and epigenetic activation of its own preserved 3D regulatory domain. By binding to reprogrammed superenhancers, PBX1 directly regulates critical oncogenic pathways and a FOXM1-dependent transcriptional program. Together, PBX1 and FOXM1 activate a proliferative gene signature that predicts adverse prognosis across multiple types of cancer. Notably, pharmacological disruption of the PBX1-FOXM1 axis with existing agents (thiostrepton) and a novel PBX1 small molecule inhibitor (T417) is selectively toxic against chr1q-amp myeloma and solid tumor cells. Overall, our systems medicine approach successfully identifies CNA-driven oncogenic circuitries, links them to clinical phenotypes, and proposes novel CNA-targeted therapy strategies in MM and other types of cancer.

Ribosome dysfunction underlies the pathogenesis of many cancers and heritable ribosomopathies. Here, we investigate how mutations in either ribosomal protein large (RPL) or ribosomal protein small (RPS) subunit genes selectively affect erythroid progenitor development and clinical phenotypes in Diamond-Blackfan anemia (DBA), a rare ribosomopathy with limited therapeutic options. Using single-cell assays of patient-derived bone marrow, we delineated two distinct cellular trajectories segregating with ribosomal protein genotypes. Almost complete loss of erythroid specification was observed in RPS-DBA. In contrast, we observed relative preservation of qualitatively abnormal erythroid progenitors and precursors in RPL-DBA. Although both DBA genotypes exhibited a proinflammatory bone marrow milieu, RPS-DBA was characterized by erythroid differentiation arrest, whereas RPL-DBA was characterized by preserved GATA1 expression and activity. Compensatory stress erythropoiesis in RPL-DBA exhibited disordered differentiation underpinned by an altered glucocorticoid molecular signature, including reduced ZFP36L2 expression, leading to milder anemia and improved corticosteroid response. This integrative analysis approach identified distinct pathways of erythroid failure and defined genotype-phenotype correlations in DBA. These findings may help facilitate therapeutic target discovery.

Summary Human hematopoiesis is a dynamic process that starts in utero 18–21 days post-conception. Understanding the site- and stage-specific variation in hematopoiesis is important if we are to understand the origin of hematological disorders, many of which occur at specific points in the human lifespan. To unravel how the hematopoietic stem/progenitor cell (HSPC) compartment changes during human ontogeny and the underlying gene regulatory mechanisms, we compare 57,489 HSPCs from 5 different tissues spanning 4 developmental stages through the human lifetime. Single-cell transcriptomic analysis identifies significant site- and developmental stage-specific transitions in cellular architecture and gene regulatory networks. Hematopoietic stem cells show progression from cycling to quiescence and increased inflammatory signaling during ontogeny. We demonstrate the utility of this dataset for understanding aberrant hematopoiesis through comparison to two cancers that present at distinct time points in postnatal life—juvenile myelomonocytic leukemia, a childhood cancer, and myelofibrosis, which classically presents in older adults., Graphical abstract, Highlights • Sampling of >57,000 hematopoietic cells over 4 stages of human development • Gene regulatory networks underpin changes in lineage specification through ontogeny • Stem cells show progressive changes in cell cycling and inflammatory signaling • Resource enables understanding of hematopoietic disorders presenting at specific age, Roy et al. use single-cell transcriptomics to compare lineage specification and gene regulatory networks of hematopoietic stem/progenitor cells (HSPCs) from multiple tissues over human ontogeny. Comparison to HSPCs from two blood cancers demonstrates the utility of the dataset for understanding disease processes and the emergence of age-specific blood malignancies.

Summary Osteoclast (OC) development in response to nuclear factor kappa-Β ligand (RANKL) is critical for bone homeostasis in health and in disease. The early and direct chromatin regulatory changes imparted by the BET chromatin readers Brd2-4 and OC-affiliated transcription factors (TFs) during osteoclastogenesis are not known. Here, we demonstrate that in response to RANKL, early OC development entails regulation of two alternative cell fate transcriptional programmes, OC vs macrophage, with repression of the latter following activation of the former. Both programmes are regulated in a non-redundant manner by increased chromatin binding of Brd2 at promoters and of Brd4 at enhancers/super-enhancers. Myc, the top RANKL-induced TF, regulates OC development in co-operation with Brd2/4 and Max and by establishing negative and positive regulatory loops with other lineage-affiliated TFs. These insights into the transcriptional regulation of osteoclastogenesis suggest the clinical potential of selective targeting of Brd2/4 to abrogate pathological OC activation., Graphical Abstract, Highlights • Upregulation of osteoclast lineage commitment TFs occurs, Biological Sciences; Developmental Biology; Bioinformatics; Omics; Transcriptomics

Multiple myeloma is a malignancy of plasma cells initiated and driven by primary and secondary genetic events. However, myeloma plasma cell survival and proliferation might be sustained by non-genetic drivers. Z-DNA-binding protein 1 (ZBP1; also known as DAI) is an interferon-inducible, Z-nucleic acid sensor that triggers RIPK3-MLKL-mediated necroptosis in mice. ZBP1 also interacts with TBK1 and the transcription factor IRF3 but the function of this interaction is unclear, and the role of the ZBP1-IRF3 axis in cancer is not known. Here we show that ZBP1 is selectively expressed in late B-cell development in both human and murine cells and it is required for optimal T-cell-dependent humoral immune responses. In myeloma plasma cells, the interaction of constitutively expressed ZBP1 with TBK1 and IRF3 results in IRF3 phosphorylation. IRF3 directly binds and activates cell cycle genes, in part through co-operation with the plasma cell lineage-defining transcription factor IRF4, thereby promoting myeloma cell proliferation. This generates a novel, potentially therapeutically targetable and relatively selective myeloma cell addiction to the ZBP1-IRF3 axis. Our data also show a noncanonical function of constitutive ZBP1 in human cells and expand our knowledge of the role of cellular immune sensors in cancer biology.

ZBP1 is an inducible, non-constitutively expressed cellular nucleic acid sensor that triggers type I interferon (IFN) responses via phosphorylation and activation of the transcription factor (TF) IRF3 by TBK1. However, the role of the ZBP1-IRF3 axis in cancer is not known. Here we show that ZBP1 is selectively and constitutively expressed in late B cell development and it is required for optimal T cell-dependent humoral immune responses. In the plasma cell (PC) cancer multiple myeloma, interaction of constitutively expressed ZBP1 with TBK1 and IRF3 results in IRF3 phosphorylation. Notably, rather than IFN type I response genes, IRF3 directly activates, in part through co-operation with the PC lineage-defining TF IRF4, cell cycle genes thus promoting myeloma cell proliferation. This generates a novel, potentially therapeutically targetable and relatively selective myeloma cell addiction to the ZBP1-IRF3 axis. These data expand our knowledge of the role of cellular immune sensors in cancer biology.

Background: Diamond- Blackfan Anemia (DBA) is a rare, heritable ribosomopathy caused by mutations in ribosomal protein large (RPL) and small (RPS) subunit genes. The diagnostic criteria of DBA include presentation in infancy with virtually no mature erythroblasts (EB) on bone marrow (BM) examination, however atypical presentations in later life with milder haematological manifestations are increasingly reported. The cellular and molecular mechanisms underpinning variable clinical phenotypes, and how they relate to genotype, are yet to be elucidated. Furthermore, as many DBA patients do not respond to corticosteroids, understanding the pathological processes contributing to erythroid failure is a prerequisite for developing novel, precision-medicine therapies. Aim: To elucidate phenotypic and functional differences in erythropoiesis in RPS- and RPL-DBA, using primary BM samples from patients. Methods: We performed single-cell transcriptome profiling (scRNAseq), using the 10X Genomics chromium platform, of 45,488 CD34+ Lineage negative (Lin-) BM hematopoietic stem and progenitor cells (HSPC), purified by fluorescence-activated cell sorting. We included six patients with red cell transfusion-dependent DBA (aged 2-19y) with mutations in RPS19 (n=3), RPL11 (n=1) and RPL5 (n=2) and three healthy donors (aged 3-17y). We validated our findings using bulk RNAseq, functional assays, and deep immunophenotyping-based dissection of the haematopoietic architecture of DBA BM ex vivo. Results: High quality sequencing data was obtained for all nine donors; after quality control, 41,415 single cells were carried forward for analysis. Unsupervised clustering analysis and lineage identification revealed two divergent cellular patterns in DBA compared with age-matched control BM that segregated with genotype: a selective loss of erythromegakaryocyte (E/MK) progenitors in RPS-DBA, but relative preservation of the erythroid developmental trajectory in RPL-DBA, at the expense of megakaryopoiesis (Fig 1a). Gene Set Enrichment Analysis (GSEA) of differentially expressed genes between control and DBA HSPC clusters revealed p53-mediated apoptosis, TNFa-, IFNa- and IFNg-mediated inflammatory pathways in DBA EP. Although these pathways were enriched across all HSPC populations irrespective of genotype, they were detected at an earlier stage of the stem cell hierarchy, and more potently, in RPS- versus RPL-EP. Expression of transcriptional targets of the master E/MK transcription factor, GATA1, was significantly upregulated in RPL- versus RPS-DBA EP (Fig 1b), supporting the maintained erythroid program in RPL-DBA HSPC. However, expression of genes denoting erythroid differentiation, including adult haemoglobin (Hb) genes, was significantly elevated in RPL-DBA, suggesting aberrant accelerated differentiation to EB. These findings were confirmed by immunophenotypic examination of patient BM and bulk RNAseq of RPL-EB. Additionally, single-cell clonogenic assays of RPL-DBA EP showed severe qualitative defects. Thus, although erythroid commitment occurs in RPL-DBA, it is coupled with accelerated maturation beyond EP to functionally impaired mature EB, enriched in inflammatory and p53 pathways. Finally, we analysed the clinical characteristics of the 161 patients comprising the U.K DBA registry. In line with the milder erythroid specification defect, patients with RPL-DBA (n=44) presented with anaemia later (P=0.004), and with a higher average Hb concentration (P=0.04), than those with RPS-DBA (n=63). Furthermore, we identified higher corticosteroid responses in RPL-DBA, assessed at 6 months post initiation (P=0.006), consistent with our findings in RPL-DBA of preservation of the cellular EP populations that are targeted by corticosteroids. Impact: In conclusion, a preserved but distinct erythroid developmental trajectory, characterised by accelerated differentiation, underpins a milder haematological phenotype in RPL-DBA. Furthermore, we reveal the first single-cell transcriptomic dataset from haematopoietic cells in a ribosomopathy, uncovering novel cell intrinsic and extrinsic pathogenetic insights into failing erythropoiesis in DBA. Integration of these data with clinical genomics and phenomics provides a paradigm by which single cell approaches can be used to decipher genotype-phenotype relationships in Mendelian genetic disorders. Disclosures Mead: Novartis: Consultancy, Honoraria, Other: travel, accommodations, expenses, Research Funding, Speakers Bureau; Celgene/BMS: Consultancy, Honoraria, Other: travel, accommodations, expenses, Research Funding; Abbvie: Consultancy; CTI: Consultancy; Gilead: Consultancy.

Chimeric antigen receptor (CAR)-T cell immunotherapy is highly effective against B lineage cancers but at the expense of potentially serious toxicity such as cytokine release syndrome (CRS) and neurotoxicity. Recent work provided clinical evidence that allogeneic, cord blood CAR-NK cells induce high rates of non-durable clinical responses without CRS or neurotoxicity.

Human lymphopoiesis is a dynamic lifelong process that starts in utero 6 weeks post-conception. Although fetal B-lymphopoiesis remains poorly defined, it is key to understanding leukemia initiation in early life. Here, we provide a comprehensive analysis of the human fetal B-cell developmental hierarchy. We report the presence in fetal tissues of 2 distinct CD191 B-progenitors, an adult-type CD10+ve ProB-progenitor and a new CD10-ve PreProB-progenitor, and describe their molecular and functional characteristics. PreProB-progenitors and ProB-progenitors appear early in the first trimester in embryonic liver, followed by a sustained second wave of B-progenitor development in fetal bone marrow (BM), where together they form > 40% of the total hematopoietic stem cell/progenitor pool. Almost one-third of fetal B-progenitors are CD10-ve PreProB-progenitors, whereas, by contrast, PreProB-progenitors are almost undetectable (0.53% +/- 0.24%) in adult BM. Single-cell transcriptomics and functional assays place fetal PreProB-progenitors upstream of ProB-progenitors, identifying them as the first B-lymphoid-restricted progenitor in human fetal life. Although fetalBMPreProB-progenitors and ProB-progenitors both give rise solely to B-lineage cells, they are transcriptionally distinct. As with their fetal counterparts, adult BM PreProB-progenitors give rise only to B-lineage cells in vitro and express the expected B-lineage gene expression program. However, fetal PreProB-progenitors display a distinct, ontogeny-related gene expression pattern that is not seen in adult PreProB-progenitors, and they share transcriptomic signatures with CD10-ve B-progenitor infant acute lymphoblastic leukemia blast cells. These data identify PreProB-progenitors as the earliest B-lymphoid-restricted progenitor in human fetal life and suggest that this fetal-restricted committed B-progenitor might provide a permissive cellular context for prenatal B-progenitor leukemia initiation.

Summary Chimeric antigen receptor anti-CD19 (CAR19)-T cell immunotherapy-induced clinical remissions in CD19+ B cell lymphomas are often short lived. We tested whether CAR19-engineering of the CD1d-restricted invariant natural killer T (iNKT) cells would result in enhanced anti-lymphoma activity. CAR19-iNKT cells co-operatively activated by CD1d- and CAR19-CD19-dependent interactions are more effective than CAR19-T cells against CD1d-expressing lymphomas in vitro and in vivo. The swifter in vivo anti-lymphoma activity of CAR19-iNKT cells and their enhanced ability to eradicate brain lymphomas underpinned an improved tumor-free and overall survival. CD1D transcriptional de-repression by all-trans retinoic acid results in further enhanced cytotoxicity of CAR19-iNKT cells against CD19+ chronic lymphocytic leukemia cells. Thus, iNKT cells are a highly efficient platform for CAR-based immunotherapy of lymphomas and possibly other CD1d-expressing cancers.

International audience; VCP/p97 regulates numerous cellular functions by mediating protein degradation through its segregase activity. Its key role in governing protein homoeostasis has made VCP/p97 an appealing anticancer drug target. Here, we provide evidence that VCP/p97 acts as a regulator of cellular metabolism. We found that VCP/p97 was tied to multiple metabolic processes on the gene expression level in a diverse range of cancer cell lines and in patient-derived multiple myeloma cells. Cellular VCP/p97 dependency to maintain proteostasis was increased under conditions of glucose and glutamine limitation in a range of cancer cell lines from different tissues. Moreover, glutamine depletion led to increased VCP/p97 expression, whereas VCP/p97 inhibition perturbed metabolic processes and intracellular amino acid turnover. GCN2, an amino acid-sensing kinase, attenuated stress signalling and cell death triggered by VCP/p97 inhibition and nutrient shortages and modulated ERK activation, autophagy, and glycolytic metabolite turnover. Together, our data point to an interconnected role of VCP/p97 and GCN2 in maintaining cancer cell metabolic and protein homoeostasis.

CD1d is a nonpolymorphic, MHC class I–like molecule that presents phospholipid and glycosphingolipid Ags to a subset of CD1d-restricted T cells called invariant NKT (iNKT) cells. This CD1d–iNKT cell axis regulates nearly all aspects of both the innate and adaptive immune responses. Expression of CD1d on B cells is suggestive of the ability of these cells to present Ag to, and form cognate interactions with, iNKT cells. In this article, we summarize key evidence regarding the role and regulation of CD1d in normal B cells and in humoral immunity. We then extend the discussion to B cell disorders, with emphasis on autoimmune disease, viral infection, and neoplastic transformation of B lineage cells, in which CD1d expression can be altered as a mechanism of immune evasion and can have both diagnostic and prognostic importance. Finally, we highlight current and future therapeutic strategies that aim to target the CD1d–iNKT cell axis in B cells.

ZBP1 is an inducible nucleic acid (NA) sensor that is activated when pathogen NA bind to its Zα and Zβ domains. ZBP1 is required for TBK1-dependent phosphorylation of the transcription factor IRF3 (pIRF3) followed by its direct activation of type I interferon genes. However, the role, if any, of ZBP1 in tumour biology is not known. By searching for genes selectively expressed in multiple myeloma (MM) we identified ZBP1 mRNA expressed in 29 MM cell lines (MMCL) but not in >1000 other cancer cell lines (CCLE dataset); ZBP1 was expressed in all 766 patient myeloma PC (CoMMpass dataset) but not in normal blood cells (Blueprint) or 53 healthy tissues (GTex). We confirmed expression of ZBP1 mRNA and/or protein in MMCL, primary human and murine germinal centre B (GCB) and plasma cells (PC) as well as in myeloma PC. By inducing T cell-dependent humoral immune responses after ip alum-NP-KLH immunisation, we explored the role of selective and constitutive expression of Zbp1 in GCB to PC transition. We found no differences in the frequency of splenic GCB cells and PC between control WT and Zbp1-/- mice and in GCB cell frequency between immunised WT and Zbp1-/- mice. However, compared to WT, the increase in PC frequency in immunised Zbp1-/- mice was 50% lower (n=10/group, p To explore the function of ZBP1 in MM we depleted by 2 lentiviral shRNAs either isoform 1 (contains both Zα and Zβ domains; shRNA1) or both isoform 1 and isoform 2 (latter lacks Zα domain; shRNA2). Both shRNAs were toxic to all 5 MMCL tested suggesting that isoform 1 but not isoform 2 is essential for myeloma cell survival. This effect was specific because survival of K562 cells, which lack expression of ZBP1, was not affected by either shRNA and exogenous ZBP1 cDNA rescued cell death of ZBP1-depleted myeloma cells. Dox-induced ZBP1 depletion was toxic to MMCL in vitro and significantly inhibited myeloma cell growth in a subcutaneous NSG model of the MMCL H929 and MM.1S. Together, these findings reveal a novel myeloma cell-specific ZBP1 dependency. Transcriptome analysis of ZBP1-depleted H929 and MM.1S cells showed amongst the significantly downregulated genes enrichment for the cell cycle control and DNA repair pathways consistent with a critical role of ZBP1 in promoting myeloma cell proliferation. Flow-cytometric analysis of ZBP1-depleted MMCL as well as of patient-derived myeloma PC revealed cell cycle arrest at the G0/1 phase and increasing apoptosis. Exploring potential links with IRF3, we first observed that unlike in non-malignant cells, IRF3 was constitutively phosphorylated in MMCL. Using protein-co-immunoprecipitation we found that endogenous ZBP1 interacts with IRF3 and TBK1 while upon co-transfection with different ZBP1 deletion mutants, ZBP1-IRF3 interaction required primarily the ZBP1, RHIM domain-containing, C-terminus. Further, while in ZBP1-depleted myeloma cells total IRF3 and TBK1 levels were not altered, pIRF3 and pTBK1 levels decreased thus showing a post-translational dependency of constitutive pIRF3 and pTBK1 on ZBP1. Finally, pharmacological inhibition of TBK1 resulted in decrease of pIRF3 without affecting total IRF3. Importantly, shRNA-mediated IRF3 depletion resulted in cell cycle arrest and death of MMCL. By integrating histone mark and in-house IRF3 ChiP-seq with transcriptome of IRF3-depleted MM.1S cells we identified 770 down- and 330 up-regulated genes predicted to be directly regulated by IRF3. Pathway enrichment analysis confirmed cell cycle as the most highly regulated by IRF3. Notably, we observed no direct or indirect regulation of the interferon genes (e.g., IFNA1, IFNB1) by IRF3. As well as the IRF3 motif, IRF3 cistrome analysis revealed significant enrichment for the distinct IRF4 motif. Integration of the IRF3/IRF4 cistromes identified >80% IRF3 binding regions are co-occupied by IRF4 and co-regulation of cell cycle genes. Further we validated IRF3-IRF4 interaction at the IRF4 super-enhancer by ChIP-re-ChIP. These data show a novel dependency in MM comprising constitutive activation of the ZBP1-IRF3 pathway and regulation of cell cycle and proliferation by IRF3 thus providing opportunities for therapeutic targeting. Disclosures Caputo: GSK: Research Funding. Auner:Amgen: Other: Consultancy and Research Funding; Takeda: Consultancy; Karyopharm: Consultancy. Karadimitris:GSK: Research Funding.

BACKGROUND: The biological basis for the adverse prognosis of chr1q gain/amplification (1q+) present in ~30% of patients with multiple myeloma (MM) remains ill-defined. The transcription factor (TF) Pre-B-cell leukaemia homeobox 1 (PBX1), encoded on chr1q, acts as master regulator of early hematopoiesis and as an oncogene in leukemia and other malignancies. Herein, we hypothesized that PBX1 orchestrates proliferative regulatory networks that underpin the poor prognosis associated with 1q+ in MM. METHODS: We employed qPCR for mRNA quantification, western blotting and immunohistochemistry for protein analysis, lentiviral shRNA-mediated knock-down, Hoechst/Annexin V staining and flow-cytometry for apoptosis and cell cycle analysis, ChIPseq for cistrome and RNAseq after knock-down for transcriptome analysis. Additional data were obtained from MMRF/CoMMpass, Blueprint Consortium and Arkansas datasets. Computational analysis of clinical and "-omics" data was performed using standard bioinformatic work-flows; pathway enrichment analysis using EnrichR and GSEA. RESULTS: Combined genomic (WGS/WES) and transcriptomic (RNAseq) analysis of the CoMMpass dataset identified a subgroup of 1q+ MM patients (60%) characterized by aberrant PBX1 overexpression and amplification (PBX1amp); survival analysis revealed significantly worse outcome of this subgroup compared to 1q+ non-PBX1amp (p Depletion of PBX1 using 2 validated shRNAs was toxic to the 1q+ MM1.S t(14;16) and U266 t(11;14) myeloma cell lines (MMCL) in vitro, revealing a novel myeloma cell addiction to PBX1. Transcriptome analysis of PBX1-depleted cells showed significant de-regulation of genes (MM1.S: 753down/320up; U266: 598 down/288up) highly enriched in cell cycle and proteasome/ER stress response pathways. Flow-cytometric analysis confirmed that PBX1-depleted myeloma cells undergo cell cycle arrest (G0/1 phase) and apoptosis (n=3; p Genome-wide binding profiling of PBX1 identified 7,354 significantly enriched peaks shared between MM1.S and U266 MMCL. These peaks were distributed at active promoters (~50%) and active enhancers (~30%), as defined by ChromHMM maps; the PBX1 consensus binding motif was identified as top hit upon motif enrichment analysis. Transcriptome-cistrome integration identified 728 and 282 genes directly activated or repressed, respectively, by PBX1 in MM1.S cells. Again, pathway enrichment analysis of PBX1 direct targets demonstrated significant over-representation of cell cycle and proliferation pathways, as well as positive regulation of various metabolic processes (oxidative phosphorylation, glycolysis, unfolded protein response). Strikingly, enrichment analysis against public ChIPseq datasets revealed significant overlap with gene targets of FOXM1 (200) and E2F (150) TF, suggesting the presence of a united PBX1-FOXM1-E2F regulatory circuitry. In fact, PBX1 binding was detected at an active FOXM1 enhancer and at the E2F1/2 promoters, while FOXM1 and E2F1/2 mRNA levels decreased significantly upon PBX1 depletion in two MMCL. To further explore the PBX1-FOXM1 interaction, we overlaid the FOXM1-depleted transcriptome with the PBX1 network in MM.1S cells. This identified 160 genes to be co-regulated by PBX1 and FOXM1, including genes previously linked to myeloma cell drug resistance (NEK2, TOP2A, AURKA); however, PBX1 expression was not altered in FOXM1-depleted cells, suggesting that PBX1 is functionally upstream of FOXM1. Additional analysis across 814 primary MM patient CoMMpass transcriptomes confirmed significant co-expression of PBX1 with FOXM1, E2F1/2 and another 67 PBX1-FOXM1 highly expressed, co-regulated gene targets, conferring poor OS (p CONCLUSIONS: We found that genetic amplification and overexpression of PBX1 defines an ultra high-risk subgroup of 1q+ MM with proliferative features. Myeloma cells are addicted to aberrant PBX1 expression. As well as essential metabolic pathways, PBX1 regulates cell cycle progression by direct transcriptional regulation of FOXM1 and of its target genes. These data reveal PBX1 as a critical biological link between chr1q gain/amplification and poor prognosis in a subset of MM patients. Disclosures Auner: Karyopharm: Consultancy; Amgen: Other: Consultancy and Research Funding; Takeda: Consultancy. Caputo:GSK: Research Funding. Karadimitris:GSK: Research Funding.

BACKGROUND: Bone disease, a common source of morbidity in multiple myeloma (MM), is caused by RANKL-induced aberrant activation of osteoclasts (OC). RANKL-induced OC lineage commitment requires repression of an Irf-8 dependent macrophage inflammatory transcriptional programme commensurate with activation of an OC lineage-specific programme. Functional data have shown the requirement for the histone acetylation readers Brd2-4 BET proteins and of cMyc for OC lineage development. However, how Brd2-4 and Myc co-operate genome-wide to regulate transcriptome changes that underpin the very early stages of RANKL-induced OC lineage commitment has not been defined. METHODS: The OC progenitor-like murine RAW264.7 cell line was used for osteoclastogenesis. OC were assayed by TRAP staining. We performed RNA-seq for transcriptome analysis and ChIP-seq against Brd2-4, cMyc, and H3K27Ac mark for epigenomic profiling. The pan-Bet inhibitor IBET151 was used alone or in combination with RANKL. ChIP-seq/RNA-seq data were processed using standard bioinformatics pipelines; downstream analyses (pathway and motif enrichment, factor differential binding) were performed by various tools including EnrichR, R packages ChIPpeakAnno/DiffBind, Rose. RESULTS: Transcriptomic profiling of OC progenitors at 0, 4, 14 and 24h post-RANKL treatment identified 12 distinct clusters of expression trends. The 4h activated cluster includes OC master transcription factors (TFs; cMyc, Nfatc1, Fosl), and is enriched in OC-defining pathways. Notably, by 14h the majority of the genes required for mature OC formation and activation are already highly expressed (e.g. Ctsk, Mmp9). The downregulated clusters include monocyte defining TFs (e.g. Irf8, Mafb and Bcl6). These RANKL-dependent transcriptome changes are completely abrogated by iBET151, highlighting the critical role of Brd2-4 in osteoclastogenesis. Differential chromatin binding analysis upon RANKL induction revealed an overall enhanced Brd2-4 binding at already existing or de novo gained sites. This was more pronounced for Brd2&4 and much less for Brd3, with differentially binding sites (DBS) comprising 50% and 20% respectively of all binding sites in RANKL-treated cells. For Brd2&3, DBS were primarily distributed at promoters and for Brd4 at intergenic, candidate enhancers regions. Notably, nearly all gained DBS were sensitive to and abrogated by iBET151. Combinatorial profiling of Brd2 and Brd4 showed that almost half of Brd2 DBS peaks overlap with Brd4 (47%; 897/1896), while only 24% (766/3234) of Brd4 DBS peaks are co-occupied by Brd2. Transcriptome and Brd2&4 DBS integration in combination with motif enrichment analysis, identified genes that are predicted to be regulated by Brd2 and/or Brd4. EnrichR analysis suggests that enhanced binding of Brd2&4, singly or in combination, is required for activation of the critical OC lineage-specific and repression of the macrophage-defining transcriptional programs highlighting the non-redundant roles of Brd2&4 in OC development. Cell lineage commitment often requires 'commissioning' of cell-specific super-enhancers (SE). Combined analysis of genome-wide Brd4/H3K27ac profiles identified 678 RANKL-induced SE and their respective target genes. Further, 110 of these SE showed enhanced Brd4 binding in 2 peaks: 20/110 were linked to significantly up- and 90/100 to down-regulated genes. The repressed genes were significantly enriched to previously described Irf8, MafB and RunX1 targets, suggesting a critical role of SE in the repression of the monocyte/macrophage inflammatory programme during OC lineage commitment. Strikingly, among top hits, we detected a SE linked to the regulation of cMyc. To further investigate its role in OC development, we obtained the cistrome of cMyc after RANKL induction. We identified 560 binding sites which were highly enriched in cMyc, Max, Fli1, Fosl2 and Irf8 motifs. Cistrome-transcriptome integration suggested direct activation of 141 and repression of 52 genes by cMyc in response to RANKL; these are enriched in ribosome biogenesis pathways and Irf8-dependent targets respectively. CONCLUSIONS: Myc and Brd4 mark SE that repress an Irf8-dependent transcriptional programme, a requirement for OC lineage commitment. The non-redundant roles of Brd2&4 suggest that selective targeting of either could inhibit aberrant OC activation associated with MM. Disclosures Caputo: GSK: Research Funding. Auner:Amgen: Other: Consultancy and Research Funding; Takeda: Consultancy; Karyopharm: Consultancy. Karadimitris:GSK: Research Funding.

Overexpression of the transcription factor MAF, as a result of its juxtaposition to the IgH enhancer [MAF-translocated t(14;16)], is a myeloma-initiating event in 3-5% of patients with multiple myeloma (MM) and confers a poor prognosis. MAF is also overexpressed in another 40% of cases, often in co-operation with the oncogene MMSET. The mechanisms by which MAF overexpression impacts on the regulatory genome to generate the MAF-driven oncogenic transcriptome and its direct targets are not known. To address this, we employed a multi-layer -omics approach using primary myeloma plasma cells (PC) as well as myeloma cell lines (MMCL). First, we determined the chromatin accessibility and transcriptome profiles of MAF-translocated myeloma by performing ATAC-seq and RNA-seq, respectively, in purified bone marrow CD138+ PC from two patients with t(14;16) and three healthy donors. We identified 6,640 differentially accessible regions, 87% of which displayed enhanced chromatin accessibility in MAF samples compared to normal PC. Secondary analysis comparing this with ATAC-seq data from a set of 28 other MM samples, including hyperdiploid, MMSET and CCND1-translocated MM, revealed 33% of those regions to be MAF subgroup specific (1,949 regions), with the rest shared between MAF and other cytogenetic groups. Gene annotation and pathway enrichment analysis using GREAT confirmed overrepresentation of the MF myeloma patient signature, as previously identified in microarray datasets. RNA-seq detected significant upregulation of approximately 900 genes in MAF samples compared to normal counterparts, including MAF itself (top 4th hit) as well as its presumed targets (CCND2, ITGB7 and NUAK1). Next, we obtained the MAF cistrome using ChIP-seq in the MAF-translocated MMCL MM1.S and integrated it with the primary PC ATAC-seq data. This revealed that 31% (618/1,949) of the differentially accessible regions in MAF-translocated MM PC are also MAF-bound. Additional overlay with ENCODE ChromHMM epigenome map showed that 47% of MAF binding sites are on active enhancers and 42% on active promoters signifying potential direct regulation of the corresponding genes. Next, we superimposed the accessible and MAF-bound loci on the epigenomic landscapes of normal PC and other B-cell types using their corresponding ChromHMM maps (Blueprint consortium data). Interestingly, 56% (345/618) of the MAF-specific regions were not active in any stage of B cell development. This suggests that aberrant MAF overexpression and chromatin binding in PC is associated with de novo activation of these chromatin regions, over half of which (200/345; 58%) are enhancers; we termed these 'neo-enhancers'. Upon de novo motif analysis of MAF ChIP-seq in MAF-translocated JJN3 and MM1.S MMCL, we confirmed MAF as the first and, interestingly, IRF4 as the second top hit, suggesting a possible MAF-IRF4 functional interaction in myelomagenesis. Indeed, overlay of the accessible MAF-bound loci with IRF4 ChIP-seq data in MM1.S revealed 63% co-occupancy (including 62% of "neo-enhancers"), proposing a novel and extensive co-operative chromatin-based network between the two transcription factors. Final integration of the accessible MAF-bound regions with the paired transcriptomes of primary myeloma PC revealed a set 206 candidate enhancer-gene pairs. Strikingly, we identified two IRF4-cobound "neo-enhancers" linked to overexpression of TLR4 and CCR1, two genes known for their roles in myeloma cell proliferation and migration. We confirmed significant downregulation of both genes upon shRNA-mediated knockdown of MAF in the two MAF-translocated MMCL, MM1.S and JJN3, as well as the lethality of MAF depletion. Further, MAF overexpression in MAF-negative myeloma backgrounds led to transcriptional upregulation of these genes, further validating them as MAF targets. While CRISPR/Cas9i experiments targeting TLR4 are ongoing, preliminary results validated the functional role of the "neo-enhancer" in CCR1 gene expression. In conclusion, we demonstrate for the first time an extensive re-organisation of the PC chromatin conferred by oncogenic MAF in MM; we reveal its extensive co-operation with IRF4 in this process; we validate the directly MAF-regulated genes and functionally characterise neo-enhancers of key MAF-dependent genes that in addition to MAF itself are also critical for myeloma biology. Disclosures Hatjiharissi: Janssen: Honoraria. Caputo:GSK: Research Funding. Karadimitris:GSK: Research Funding.

In multiple myeloma (MM), a malignancy of the bone marrow plasma cells (BMPC), hyperdiploidy (HY) and oncogene over-expression via chromosomal translocation [including CCND1- t(11;14), MAF- t(14;16), MMSET-t(4;14)] are the primary myeloma initiating events (MIE) that drive distinct transcriptional programs. These are further shaped by secondary SNV and CNV events. This genetic heterogeneity converges, in most cases, to a functionally dichotomous state of CCND1 or CCND2 overexpression. The molecular mechanisms underlying each of the distinct myelomagenic transcriptomes and the CCND1 vs CCND2 dichotomy have not been defined. To address these questions, we obtained highly purified BMPC from 3 healthy donors and 30 MM patients (HY: 15; CCND1: 4; MMSET: 5; MAF: 2; other: 4), either at diagnosis or relapse, and mapped their chromatin accessibility and transcriptome profiles by ATAC-seq and RNA-seq, respectively. In total, we obtained ~300K regions with accessible chromatin in either MM or normal PC. Overall chromatin accessibility increased in myeloma compared to normal PC, particularly in MAF- and MMSET-translocated subtypes. Analysis of combined ATAC-seq/RNA-seq by Multi-Omics Factor Analysis (MOFA) resulted in a clearer samples distinction than either ATAC-seq or RNA-seq alone, with altered chromatin accessibility accounting for more of the variance than expression. Of the top five identified factors, the top two (one transcriptome driven, one accessibility driven) distinguished normal from MM samples, whilst two more separated MMSET, MAF and CCND1 subgroups. Ninety seven, 157, 256 and 348 overexpressed genes in the CCND1, HY, MMSET and MAF subgroups, respectively, were predicted to be regulated by differentially accessible enhancers. Twenty percent (165/858) of these genes were overexpressed in >1 subgroup suggesting a process of chromatin accessibility-based convergence evolution. Enrichment analysis suggested direct or indirect involvement of Polycomb and chromatin remodellers; significant enrichment was also found for genes involved in neurogenesis. ATAC-seq footprinting predicted binding sites for 250 expressed transcription factors (TFs), 116 of which displayed higher binding frequency in myeloma than in normal PC and included both known (e.g., XBP1, RELA, IRF4, PRDM1) and potentially novel regulators of myeloma biology (e.g., CXXC1 and NFE2L1). The remaining 134 TF were predicted to be present in at least one MM subgroup, but absent in normal PC. Amongst them, as expected, MAF was active in the MMSET- and more so in the MAF-translocated subgroups. DepMap database analysis suggested myeloma cell dependency on 181/250 TF (CRISPR/Cas9 CERES score < -0.1 in >3/14 MMCL analysed). In dissecting the regulatory basis of CCND2 vs CCND1 dichotomy, one MOFA factor completely separated MAF from CCND1 samples, placing extreme opposite weights on the expression of CCND2 and CCND1 respectively. Interestingly, the same factor identified open-chromatin clusters upstream of CCND2 and linked them to its over-expression. These clusters were also open in the MMSET group and in CCND2-expressing HY samples. Conversely, no accessibility was detected in the CCND1 group, the CCND1-expressing HY samples or in normal PCs. Further, super-enhancer calling using the H3K27ac histone mark in MAF-translocated JJN3 cells identified the region of interest as a bona fide super-enhancer. Chromatin long range interactions, as assessed by Capture-HiC, demonstrated high frequency interactions of the CCND2 promoter with the constituent elements of the putative super-enhancer. Experimental validation using a CRISPR/Cas9i system confirmed the functional role of all 4 super-enhancer constituents tested in the regulation of CCND2 expression, while TF footprinting predicted MAF binding to the super-enhancer in MAF-translocated PC. In conclusion, we show that distinct oncogenic transcriptomes in MM are underpinned by extensive chromatin changes, accompanied by TF activity 're-wiring' that does not necessarily require transcriptional deregulation of the TF themselves. We identify novel, non-oncogene TF dependencies that suggest therapeutic opportunities in MM and we discover and characterise the critical super-enhancer that drives overexpression of the CCND2 oncogene in MM. Disclosures Auner: Amgen: Other: Consultancy and Research Funding; Takeda: Consultancy; Karyopharm: Consultancy. Hatjiharissi:Janssen: Honoraria. Caputo:GSK: Research Funding. Karadimitris:GSK: Research Funding.

The mechanism of bone marrow failure (BMF) in paroxysmal nocturnal hemoglobinuria (PNH) is not yet known. Because in PNH the biosynthesis of the glycolipid molecule glycosylphosphatidylinositol (GPI) is disrupted in hematopoietic stem and progenitor cells by a somatic mutation in the PIG-A gene, BMF might result from an autoimmune attack, whereby T cells target GPI in normal cells, whereas PIG-A mutant GPI-negative cells are spared. In a deliberate test of this hypothesis, we have demonstrated in PNH patients the presence of CD8(+) T cells reactive against antigen-presenting cells (APCs) loaded with GPI. These T cells were significantly more abundant in PNH patients than in healthy controls; their reactivity depended on CD1d expression and they increased upon coculture with CD1d-expressing, GPI-positive APCs. In GPI-specific T cells captured by CD1d dimer technology, we identified, through global T-cell receptor α (TCRα) analysis, an invariant TCRVα21 sequence, which was then found at frequencies higher than background in the TCR repertoire of 6 of 11 PNH patients. Thus, a novel, autoreactive, CD1d-restricted, GPI-specific T-cell population, enriched in an invariant TCRα chain, is expanded in PNH patients and may be responsible for BMF in PNH.

Abstract 424 In Multiple myeloma (MM), a malignancy of plasma cells, autocrine and paracrine networks involving the malignant cells and their microenvironment, especially osteoclasts (OC), play a crucial role in disease pathogenesis. OC activation and bone destruction are common and devastating events in this disease. Tumour-derived glycosphingolipids (GSL) have been shown to modify the tumour microenvironment by promoting tumour growth, angiogenesis, immune evasion and metastasis. The aim of this work was two-fold: 1) to investigate the role of MM-derived GSL in OC development and activation and 2) to define the role of de novo GSL synthesis in OC development. Using MALDI-TOF MS and MS-MS, we first determined the GSL repertoire of primary CD138+ myeloma cells (n=3) and compared it with non-myeloma bone marrow cells (i.e., without CD138+ cells) (n=3), and various myeloma cell lines (n=5). We found that GM3 was the dominant GSL in primary myeloma cells and GM2/GM3 in myeloma cell lines; by contrast, in non-myeloma marrow the non-polar LacCer was the dominant GSL. As GM3 was the dominant GSL in myeloma cells, we tested its effect on osteoclast function. We found that exogenous GM3 synergistically enhanced the ability of M-CSF and RANKL to induce maturation of murine bone marrow OC in vitro. This, as shown by immunoblotting, was associated with increased ERK1/2, p38, JNK phosphorylation and NFATc dephosphorylation, signal transduction and transcriptional events respectively, required for OC differentiation and maturation in response to RANKL. Furthermore, GM3 further enhanced OC maturation in synergy with IGF-1, a growth factor known to promote myeloma growth and OC activation. Next we tested the effect of inhibition of de novo GSL biosynthesis on osteoclastogenesis. We found that the glucose ceramide synthase (GCS) inhibitors NB-DNJ inhibited RANKL- and M-CSF-dependent development of murine as well as human, monocyte-derived OC in a dose dependent manner when added either in the beginning or during OC differentiation cultures. This effect was associated with significantly reduced RANKL- and M-CSF-dependent phosphorylation of ERK, JNK and p38 as well as reduced localisation of NFATc in the nucleus. OC development in response to RANKL-RANK interaction requires movement of RANK into lipid rafts where it interacts with TRAF6, an adaptor crucial for downstream signalling and with cSrc which is required for actin ring formation and OC resorptive activity. Using sucrose gradient membrane fractionation and GM1 as a marker of rafts we found that GCS inhibitors partially disrupt the integrity of lipid rafts in developing osteoclasts and prevent RANKL-induced localisation of TRAF6 and Src in lipid rafts. Next we tested the ability of NB-DNJ to block OC activation caused by a single injection of alpha-galactosylceramide (αGC), an invariant NKT cell ligand known to rapidly activate the innate immune response. We found that in mice receiving αGC, serum C-telopeptide Type I collagen (CTX) levels increased by ∼50% (p0.05). Taken together, our data demonstrate a novel role of GSL in promoting OC differentiation and activation. Thus, GCS inhibitors may be of benefit in reducing OC activation and bone destruction in MM by preventing generation of tumour-derived, pre-osteoclastogenic GSL as well as by inhibiting de novo OC GSL synthesis and thus OC activation. Disclosures: No relevant conflicts of interest to declare.

Invariant NKT cells, a small subset of immunoregulatory T cells restricted by the glycolipid-presenting non-polymorphic CD1d molecule, are able to modulate a variety of innate and adaptive immune responses. Osteoclasts (OC) are bone resorbing polykaryons of hematopoietic lineage, that have the capacity to regulate myeloid cell egress from bone marrow (BM) thus making them an integral part of the innate immune response. We and others previously showed that NKT cells regulate hematopoiesis in mice as well as humans. In this work, we investigate the role of NKT cells in OC development and function in homeostasis and after their specific activation by the model glycolipid alpha-galactosylceramide (aGC). Using quantitative back scattered electron scanning microscopy, we found that TCR Ja18 −/− mice which selectively lack development of NKT cells, exhibit a moderate osteopetrotic phenotype affecting trabecular as well as cortical bone. Histologically, these mice had the same number of TRAP+ OC as WT mice suggesting a maturation rather developmental defect in the TCR Ja18 −/−-derived OC. In vitro differentiation in the presence of RANKL and M-CSF showed that while TCR Ja18 −/− BM cells are capable of forming multinucleated OC, these, as assessed by confocal microscopy, fail to form F-actin rings and sealing zone and thus are unable to resorb bone. Further underscoring the effect of NKT cells in this process, CD45.1+ BM cells highly purified from CD45.2+ WT/CD45.1+ TCRJa18 −/− mixed BM chimeras displayed restoration of their OC F-actin rings. Next we investigated whether in vivo activated NKT cells regulate OC function. We found that a single injection of aGC dramatically increased the number of CD3-B220-CD11b-c-fmshighc- kithigh BM OC progenitors and accelerated the in vitro development of OC in WT but not TCR Ja18 −/− mice. Furthermore, this resulted in high serum levels of IFN-g and IL-4 but not IL-1 or IL-17. An aGC-mediated increase of OC progenitors was observed in IFN-g −/− but not IL-4 −/−mice suggesting that NKT cell-derived IL-4 is the main cytokine promoting osteoclastogenesis in this context. Taken together, our data demonstrate a novel role of NKT cells in homeostatic bone mass regulation and in the orchestration of innate immune responses through regulation of OC development and function.

Invariant natural killer T cells (iNKT) are rare but powerful immunoregulatory T cells, evolutionarily conserved and characterized in humans by an invariant TCR Vα24-Jα18 chain, paired with a diverse TCR Vβ11 chain. Unlike conventional T cells which require peptide antigens presented by classical MHC molecules for their development and activation, iNKT cell selection and activation requires interaction with the non-polymorphic, MHC class I–like molecule CD1d presenting glyco(lipid) antigens. Upon invariant TCR – CD1d engagement, iNKT cells are activated and rapidly produce large amounts of cytokines. iNKT cells are functionally diverse and, either as immunoregulatory or effector cells, they can engage with nearly all aspects of the innate or adaptive immune response. As such, they can modulate emerging immune responses relevant to pathogens, autoimmunity, cancer, and alloreactivity. Although the biology of iNKT cells has been extensively studied in mice, their contribution to the pathogenesis of human disease remains elusive. Nevertheless, recent clinical observations in conjunction with insights from preclinical work highlight their importance in the alloresponse associated with hematopoietic stem cell transplantation and provide the impetus for exploring the therapeutic potential of iNKT cells.

Invariant natural killer T (iNKT) cells are powerful immunomodulatory cells that in mice regulate a variety of immune responses, including acute GVHD (aGVHD). However, their clinical relevance and in particular their role in clinical aGVHD are not known. We studied whether peripheral blood stem cell (PBSC) graft iNKT-cell dose affects on the occurrence of clinically significant grade II-IV aGVHD in patients (n = 57) undergoing sibling, HLA-identical allogeneic HSCT. In multivariate analysis, CD4− iNKT-cell dose was the only graft parameter to predict clinically significant aGVHD. The cumulative incidence of grade II-IV aGVHD in patients receiving CD4− iNKT-cell doses above and below the median were 24.2% and 71.4%, respectively (P = .0008); low CD4− iNKT-cell dose was associated with a relative risk of grade II-IV aGVHD of 4.27 (P = .0023; 95% CI, 1.68-10.85). Consistent with a role of iNKT cells in regulating aGVHD, in mixed lymphocyte reaction assays, CD4− iNKT cells effectively suppressed T-cell proliferation and IFN-γ secretion in a contact-dependent manner. In conclusion, higher doses of CD4− iNKT cells in PBSC grafts are associated with protection from aGVHD. This effect could be harnessed for prevention of aGVHD.

Invariant NKT (iNKT) cells modulate innate and adaptive immune responses through activation of myeloid dendritic cells and macrophages and via enhanced clonogenicity, differentiation, and egress of their shared myeloid progenitors. Because these same progenitors give rise to osteoclasts (OCs), which also mediate the egress of hematopoietic progenitors and orchestrate bone remodeling, we hypothesized that iNKT cells would extend their myeloid cell regulatory role to the development and function of OCs. In this study, we report that selective activation of iNKT cells by α-galactosylceramide causes myeloid cell egress, enhances OC progenitor and precursor development, modifies the intramedullary kinetics of mature OCs, and enhances their resorptive activity. OC progenitor activity is positively regulated by TNF-α and negatively regulated by IFN-γ, but is IL-4 and IL-17 independent. These data demonstrate a novel role of iNKT cells that couples osteoclastogenesis with myeloid cell egress in conditions of immune activation.

BACKGROUND: Patients with relapsed and/or refractory non-Hodgkin's lymphoma (NHL), especially those with aggressive lymphomas, have overall poor prognosis. Novel targets and therapies are under investigation. Molibresib (GSK525762) is a potent and specific inhibitor of the bromodomain and extraterminal domain (BET) family of proteins, the inhibition of which prevents transcriptional complex assembly and the subsequent expression of oncogenic drivers. Molibresib inhibits growth in NHL cell lines, both in vitro and in vivo. Study BET116183 was designed to evaluate the safety, tolerability, and preliminary efficacy of molibresib in relapsed and refractory hematologic malignancies. Here we report the results from the NHL dose escalation cohort. METHODS: Eligible subjects were adults with relapsed or refractory NHL. An accelerated dose titration was employed with one subject per dose level until the occurrence of a ≥Grade 2 drug-related toxicity; thereafter, subjects were enrolled in a standard 3+3 design. A Neuenschwander continual reassessment method (N-CRM) model was used to provide guidance for the next dose level. Dose escalation continued until the maximum tolerated dose (MTD) was identified. All data, including safety, tolerability, pharmacokinetics (PK), and efficacy, were used to identify the recommended part 2 dose (RP2D). RESULTS: From 14 May 2014 to the data cutoff date of 24 June 2018, 27 NHL subjects were enrolled and received at least one dose of study drug. Of these, 19 (70%) had B-cell lymphomas (diffuse large B-cell lymphoma [DLBCL], mantle cell lymphoma, marginal zone lymphoma, follicular lymphoma , and Burkitt's lymphoma); eight subjects (30%) had T-cell lymphomas (cutaneous T-cell lymphoma [CTCL], anaplastic T-cell lymphoma [ATCL], peripheral T-cell lymphoma, and adult T-cell leukemia/lymphoma). The median age was 64 years (range 24 to 76); 20 subjects (76%) were male and 7 subjects (24%) were female. The median number of prior treatments was 3 (range 1 to > 4). From the starting dose of 10 mg molibresib orally once daily (QD), the dose was escalated to 80 mg QD. The median time on study was 1.4 months (range 0.2 to 20 months). Two dose-limiting toxicities (DLTs) were identified in subjects treated at 60 mg QD, though one was subsequently determined not to be a DLT. One subject experienced Grade 4 thrombocytopenia related to study drug. A second subject experienced a Grade 2 mechanical fall; this event was later revised to unrelated to study drug. Across all dose levels, all subjects experienced an adverse event (AE); 25 subjects (93%) experienced at least one AE that was deemed to be related to molibresib treatment. The most common related AEs across all dose levels were thrombocytopenia (n = 21 [78%]), fatigue (n = 6 [22%]), nausea (n = 6 [22%]), diarrhea (n = 4 [15%]), and rash (n = 4 [15%]). Blood bilirubin was increased in 3 subjects (11%), and prothrombin time and activated partial thromboplastin time were prolonged in 2 subjects each (7%). Common Grade 3 and Grade 4 related events included thrombocytopenia (n = 19 [70%]), as well as anemia, asthenia, and increased blood bilirubin (n = 2 [7%] each). No Grade 5 related AEs were reported. Among all subjects, 11 (41%) required dose reduction for toxicity: 7 subjects at the 60 mg dose level (39% treated at that dose) and 4 at the 80 mg dose level (57% treated at that dose). PK analyses showed dose-proportionality after single and repeat dosing, with large variability between subjects. One subject with DLBCL achieved a complete remission that was durable through week 54 on study. Four additional subjects (one DLBCL and 3 CTCL) achieved partial remission, for an objective response rate (ORR) of 5/27 (18.5%). Five more subjects had stable disease as best response. Of six CTCL/ATCL subjects enrolled, three subjects had partial response for an ORR of 50% in this sub-population. CONCLUSIONS: This is the first study evaluating the safety and efficacy of the BET inhibitor molibresib in NHL subjects. Overall, thrombocytopenia and other AEs were monitorable, manageable and reversible. The RP2D was identified as 60 mg QD. Single-agent activity was observed across multiple NHL subtypes at both 60 mg and 80 mg doses; most notable was a 50% response rate in subjects with CTCL. Because of the promising data, Part 2 of the BET116183 study is currently open and enrolling subjects with CTCL to better define the clinical activity of BET bromodomain inhibition in this histology. Disclosures Dickinson: GSK: Consultancy. Kamdar:Genentech: Consultancy; Seattle Genetics: Speakers Bureau. Mateos:Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; GSK: Consultancy, Membership on an entity's Board of Directors or advisory committees; Amgen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Abbvie: Consultancy, Membership on an entity's Board of Directors or advisory committees. Alegre:Janssen: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees. Kim:Roche: Research Funding; Mundipharma: Research Funding; J&J: Research Funding; Novartis: Research Funding; Kyowa-Kirin: Research Funding; Celltrion: Research Funding; Takeda: Research Funding. Martín:Janssen: Honoraria, Other: Travel expenses; Celgene: Consultancy, Honoraria, Other: Travel expenses; Roche: Consultancy, Honoraria, Other: Travel expenses; Servier: Honoraria, Other: Travel expenses. Horner:GSK: Employment. Winnberg:GSK: Employment. Mathew:GSK: Employment. Botbyl:GSK: Employment. Karpinich:GSK: Employment. Kremer:GSK: Employment. Dhar:GSK: Employment. Karadimitris:GSK: Research Funding; Gilead: Honoraria; Celgene: Research Funding.

Background: Multiple myeloma (MM) is a remitting-relapsing malignancy with variable clinical outcome. Certain cytogenetic abnormalities, either present at diagnosis or emerged at later stages, predict for poor outcome and highlight the clinical importance of MM genetic heterogeneity. Whole genome and exome sequencing studies reveal a complex intraclonal genetic landscape, organised in linear and branching Darwinian patterns, which evolves in space and time. Clones with a more complex genetic architecture may be more fit to escape treatment and those patients are likely to have a worse clinical outcome. Clinical multicolour flow cytometry (MFC) is routinely used in MM diagnosis and detection of minimal residual disease. Previous studies have shown that MM cell subpopulations with discrete phenotypic features correspond to genetic subclones, therefore it is plausible that MFC data captures clonal heterogeneity. On that basis, we propose that clustering analysis of MM phenotypic subpopulations could be clinically relevant. Methods: We retrospectively analysed clinical MCF data at diagnosis from 44 patients eligible for autologous stem cell transplantation (AutoSCT) and 14 ineligible patients and data from 52 relapsed patients after first AutoSCT. All patients were treated between 2012 - 2018. The 8-colour MCF marker panel included CD138, CD38, CD56, CD45, CD20, CD19, cytoplasmic kappa and lambda light chains (cytLC). Data was analysed in FlowJo software and MM plasma cells were identified as CD38high, CD19-, cytLC+, within their FSC-A/SSC-A physical gate. The gated events were exported in a new fcs file. Clustering analysis was performed in Cytofkit, a R-based Bioconductor package, using the Rphenograph, Cluster-X and FlowSOM algorithms. All fcs files were subjected in the same clustering analysis, but CD56 positive and CD56 negative cases were analysed separately to offset bias from differential CD56 expression. Parameters inserted in the algorithms were FSC-A, CD138, CD38, CD45, CD20 and CD56. The number of clusters was produced by FlowSOM (k=4) and only clusters with size >1% of the total events were accepted. Results: At diagnosis, FlowSOM identified 1 (n=32, 56.1%) or 2 clusters (n=19, 33.3%) in most cases. Three clusters were found only in 5 patients (8.8%) and 4 clusters in 1 patient (1.8%). The number of clusters at diagnosis did not correlate with cytogenetic risk group or ISS. Also, the number of clusters did not predict for depth of response or relapse free survival post AutoSCT. On the contrary, phenotypic patterns at relapse post AutoSCT were more complex, with 1 cluster identified in 2 patients only (3.8%), 2 clusters in 23 (44.2%), 3 clusters in 24 (46.2%) and 4 clusters in 3 patients (5.8%). Patients with >2 clusters (n=27) had a shorter survival post relapse (median 17 months - 95% CI, 7-26.6) compared to those (n=25) with 1-2 phenotypic clusters (median not reached, Log rank p=0.06). A phenotypic cluster characterised by CD138low/- at relapse, was also associated with adverse outcome and higher risk cytogenetics. In 14 patients with available serial samples at diagnosis and relapse we observed 3 patterns of phenotypic evolution: a. sustained pattern, b. change of dominant cluster and c. emergence of completely new subpopulations. These evolving phenotypes resemble changes in clonal composition over time observed in genetic studies. Conclusion: Clinical MCF, in addition to routine diagnostics, can be informative of MM biological and clinical heterogeneity. Particularly in the relapsed setting, complex phenotypic patterns identified by clustering analysis may be of prognostic value. Validation of this preliminary study results in larger patient cohorts or clinical trials, could provide a useful and readily available tool for patient stratification and prognosis. Disclosures Apperley: Novartis: Honoraria, Research Funding, Speakers Bureau; Pfizer: Honoraria, Speakers Bureau; BMS: Honoraria, Speakers Bureau; Incyte: Honoraria, Speakers Bureau. Karadimitris:Celgene: Research Funding; GSK: Research Funding; Gilead: Honoraria.

Down-regulation of conventional human leukocyte antigen (HLA) class I and II molecules from the surface of tumor cells is an important mechanism for tumor immune evasion, survival, and progression. Whether CD1d, a nonconventional, glycolipid-presenting HLA class I–like molecule instructing the function of the immunoregulatory invariant NKT cells can affect tumor cell survival is not known. Here we show that CD1d is highly expressed in premalignant and early myeloma, but with disease progression its expression is reduced and eventually in advanced stages and myeloma cell lines is lost altogether, suggesting that CD1d impacts negatively on myeloma cell survival. Consistent with this, engagement of CD1d by anti-CD1d monoclonal antibodies (mAbs) induces cell death of myeloma cell lines with restored CD1d expression and primary myeloma cells. Cell death induced by monoclonal antibody engagement of CD1d is associated with overexpression of proapoptotic Bax and mitochondrial membrane potential loss but it is caspase-activation independent; in addition, it requires the cytoplasmic tail but not the Tyr residue critical for lysosomal sorting of CD1d. Finally, anti-CD1d cooperates with antimyeloma agents in the killing of myeloma cells. Thus, this work provides evidence linking a novel function of CD1d in the regulation of cell death with tumor survival and progression in humans.

The cellular mechanisms that control protein degradation may constitute a non-oncogenic cancer cell vulnerability and, therefore, a therapeutic target. Although this proposition is supported by the clinical success of proteasome inhibitors in some malignancies, most cancers are resistant to proteasome inhibition. The ATPase valosin-containing protein (VCP; p97) is an essential regulator of protein degradation in multiple pathways and has emerged as a target for cancer therapy. We found that pharmacological depletion of VCP enzymatic activity with mechanistically different inhibitors robustly induced proteotoxic stress in solid cancer and multiple myeloma cells, including cells that were insensitive, adapted, or clinically resistant to proteasome inhibition. VCP inhibition had an impact on two key regulators of protein synthesis, eukaryotic initiation factor 2α (eIF2α) and mechanistic target of rapamycin complex 1 (mTORC1), and attenuated global protein synthesis. However, a block on protein translation that was itself cytotoxic alleviated stress signaling and reduced cell death triggered by VCP inhibition. Some of the proteotoxic effects of VCP depletion depended on the eIF2α phosphatase, protein phosphatase 1 regulatory subunit 15A (PPP1R15A)/PP1c, but not on mTORC1, although there appeared to be cross-talk between them. Thus, cancer cell death following VCP inhibition was linked to inadequate fine-tuning of protein synthesis and activity of PPP1R15A/PP1c. VCP inhibitors also perturbed intracellular amino acid levels, activated eukaryotic translation initiation factor 2α kinase 4 (EIF2AK4), and enhanced cellular dependence on amino acid supplies, consistent with a failure of amino acid homeostasis. Many of the observed effects of VCP inhibition differed from the effects triggered by proteasome inhibition or by protein misfolding. Thus, depletion of VCP enzymatic activity triggers cancer cell death in part through inadequate regulation of protein synthesis and amino acid metabolism. The data provide novel insights into the maintenance of intracellular proteostasis by VCP and may have implications for the development of anti-cancer therapies.

Multiple myeloma (MM) is an incurable tumor of the plasma cells, the terminally differentiated immunoglobulin secreting B lineage cells. The genetic make-up of MM has been extensively characterized but its impact on the biology of the disease is incomplete without more precise knowledge of the identity and functional role of cells with multiple myeloma propagating activity (MMPA). We review here recent data that link MMPA with myeloma clonotypic populations organized in a cellular hierarchy that mirrors normal B cell development and also with drug resistance and disease relapse. We further propose a conceptual framework which, with optimal use of recent technological advances in genomics and phenomics, could allow dissection of the cellular and molecular properties of cells with MMPA, drug resistance and in vivo relapse in an integrated and patient-specific manner. There is real hope that these approaches will significantly contribute to further improvements in disease control, overall survival, and possibly cure of patients with MM. Stem Cells 2015;33:3205–3211

Diamond-Blackfan anemia (DBA) is a disorder characterized by a selective defect in erythropoiesis. Delineation of the precise defect is hampered by a lack of markers that define cells giving rise to erythroid burst- and erythroid colony-forming unit (BFU-E and CFU-E) colonies, the clonogenic assays that quantify early and late erythroid progenitor (EEP and LEP) potential, respectively. By combining flow cytometry, cell-sorting, and single-cell clonogenic assays, we identified Lin(-)CD34(+)CD38(+)CD45RA(-)CD123(-)CD71(+)CD41a(-)CD105(-)CD36(-) bone marrow cells as EEP giving rise to BFU-E, and Lin(-)CD34(+/-)CD38(+)CD45RA(-)CD123(-)CD71(+)CD41a(-)CD105(+)CD36(+) cells as LEP giving rise to CFU-E, in a hierarchical fashion. We then applied these definitions to DBA and identified that, compared with controls, frequency, and clonogenicity of DBA, EEP and LEP are significantly decreased in transfusion-dependent but restored in corticosteroid-responsive patients. Thus, both quantitative and qualitative defects in erythroid progenitor (EP) contribute to defective erythropoiesis in DBA. Prospective isolation of defined EPs will facilitate more incisive study of normal and aberrant erythropoiesis.

Invariant natural killer T cells (iNKT cells) are a small subset of immunoregulatory T cells highly conserved in humans and mice. On activation by glycolipids presented by the MHC-like molecule CD1d, iNKT cells promptly secrete T helper 1 and 2 (Th1/2) cytokines but also cytokines with hematopoietic potential such as GM-CSF. Here, we show that the myeloid clonogenic potential of human hematopoietic progenitors is increased in the presence of glycolipid-activated, GM-CSF–secreting NKT cells; conversely, short- and long-term progenitor activity is decreased in the absence of NKT cells, implying regulation of hematopoiesis in both the presence and the absence of immune activation. In accordance with these findings, iNKT-cell–deficient mice display impaired hematopoiesis characterized by peripheral-blood cytopenias, reduced marrow cellularity, lower frequency of hematopoietic stem cells (HSCs), and reduced early and late hematopoietic progenitors. We also show that CD1d is expressed on human HSCs. CD1d-expressing HSCs display short- and long-term clonogenic potential and can present the glycolipid α-galactosylceramide to iNKT cells. Thus, iNKT cells emerge as the first subset of regulatory T cells that are required for effective hematopoiesis in both steady-state conditions and under conditions of immune activation.

A rare point mutation in the core promoter -270GC-rich box of PIGM, a housekeeping gene, disrupts binding of the generic transcription factor (TF) Sp1 and causes inherited glycosylphosphatidylinositol (GPI) deficiency (IGD). We show that whereas PIGM messenger RNA levels and surface GPI expression in IGD B cells are low, GPI expression is near normal in IGD erythroid cells. This divergent phenotype results from differential promoter chromatin accessibility and binding of Sp1. Specifically, whereas PIGM transcription in B cells is dependent on Sp1 binding to the -270GC-rich box and is associated with lower promoter accessibility, in erythroid cells, Sp1 activates PIGM transcription by binding upstream of (but not to) the -270GC-rich box. These findings explain intact PIGM transcription in IGD erythroid cells and the lack of clinically significant intravascular hemolysis in patients with IGD. Furthermore, they provide novel insights into tissue-specific transcriptional control of a housekeeping gene by a generic TF.

HDAC inhibitors (HDACi) increase transcription of some genes through histone hyperacetylation. To test the hypothesis that HDACi-mediated enhanced transcription might be of therapeutic value for inherited enzyme deficiency disorders, we focused on the glycolytic and pentose phosphate pathways (GPPPs). We show that among the 16 genes of the GPPPs, HDACi selectively enhance transcription of glucose 6-phosphate dehydrogenase (G6PD). This requires enhanced recruitment of the generic transcription factor Sp1, with commensurate recruitment of histone acetyltransferases and deacetylases, increased histone acetylation, and polymerase II recruitment to G6PD. These G6PD-selective transcriptional and epigenetic events result in increased G6PD transcription and ultimately restored enzymatic activity in B cells and erythroid precursor cells from patients with G6PD deficiency, a disorder associated with acute or chronic hemolytic anemia. Therefore, restoration of enzymatic activity in G6PD-deficient nucleated cells is feasible through modulation of G6PD transcription. Our findings also suggest that clinical consequences of pathogenic missense mutations in proteins with enzymatic function can be overcome in some cases by enhancement of the transcriptional output of the affected gene.

Background: Bromodomains (BRDs) are domains found in a variety of proteins that recognize and bind to acetylated lysine residues in histone and other target proteins. The BRD and extra-terminal (BET) family of BRD-containing proteins bind to acetylated histone tails, alters chromatin structure and facilitates transcriptional complex localization to specific genes, thereby regulating gene transcription. The investigational agent GSK525762 is a potent small molecule inhibitor of the BET family of proteins that prevents assembly of macromolecular complexes and transcriptional response. GSK525762 inhibits growth in a broad spectrum of human hematological cancer cell lines, including cell lines derived from human patients with acute myeloid leukemia (AML), non-Hodgkin's lymphoma (NHL), and multiple myeloma (MM). GSK525762 is orally bio-available and can cause tumor reduction and improved animal survival in in vivo xenograft models of hematologic malignancies. Methods: This is a Phase I/II, open-label, 2-part study. Part 1 is a dose-escalation phase to determine the safety, tolerability, and recommended Phase 2 dose (RP2D) of GSK525762 in adult subjects with relapsed/refractory AML, NHL, and MM. Dose escalation is performed independently for each of these three cohorts. An accelerated dose titration was employed with one subject per dose level until the occurrence of a ≥Grade 2 drug-related toxicity or dose-limiting toxicity; thereafter, subjects have been enrolled in a standard 3+3 design. A Neuenschwander continual reassessment method (N-CRM) model is used at each dose escalation decision to provide guidance for the next dose escalation level. Starting dose is 5 mg GSK525762 orally once daily and dose escalation continues until the MTD is identified. All data, including safety, tolerability, pharmacokinetics (PK), and efficacy, are used to identify the RP2D. In Part 1, approximately 60-70 subjects will be enrolled (approximately 20 in each of three disease-specific cohorts); no hypothesis is being tested, and all analysis will be descriptive and exploratory. In Part 2, the clinical activity of GSK525762 (overall response rate) will be evaluated in expansion cohorts of subjects with AML, NHL, and MM. Up to 32 subjects may be enrolled into the AML and NHL cohorts, and up to 37 subjects may be enrolled in the MM cohort. Cohorts may be closed early if they do not exceed futility assessment. In addition, an exploratory cohort of subjects with double-hit lymphoma (DHL) and triple-hit lymphoma (THL) will be enrolled to evaluate clinical activity in this patient population. Additional study objectives include analysis of PK after single and repeat dosing, evaluation of pharmacodynamics (PD) and the relationship between GSK525762 exposure and safety/efficacy/PD parameters. Recruitment is ongoing across five centers (USA, UK, and Australia). Currently, 40 subjects have been enrolled (29 AML, 8 NHL, and 3 with MM). Study funded by GSK. Disclosures Stein: Seattle Genetics: Research Funding; Agios Pharmaceuticals: Other: Advisory Board, Research Funding; Celgene: Other: Advisory Board, Research Funding; Novartis: Consultancy. Huntly:Novartis: Speakers Bureau; BMS: Speakers Bureau; Ariad: Speakers Bureau; Pfizer: Speakers Bureau. Dickinson:GlaxoSmithKline: Consultancy, Research Funding. Horner:GlaxoSmithKline: Employment. Brennan:GlaxoSmithKline: Employment. Baron:GlaxoSmithKline: Employment. Kremer:GlaxoSmithKline: Employment, Equity Ownership. Dhar:GlaxoSmithKline: Employment.

Introduction: Development of mature B cells depends upon the expression of a functional B cell receptor (BCR) and of its immunoglobulin (Ig) heavy (H) and light (L) chains. The molecular hallmark of B cell development, somatic recombination of the genes that encode the IgH (V, D and J) and IgL (V and J) chains, takes place in B cells in primary B lymphopoiesis sites throughout development (i.e. fetal liver (FL), fetal bone marrow (FBM) and adult BM). However, the spatiotemporal relationship between the IgH repertoire in FL with that in FBM, and the impact of the fetal Ig repertoire on the long-term repertoire present in post-natal life, as well as the link between this and the development of disease, are largely unknown. Aim: To gain insights into the ontogeny of the human innate B cell repertoire by performing high resolution analysis of the IgH-Cmu repertoire of human FL, FBM and post-natal B cells. Methods: CD34negCD19pos B cells from human 2nd trimester FL and FBM, child peripheral blood (cPB) and adult peripheral blood (aPB) were isolated by flow sorting. For initial assessment of the IgM repertoire ontogeny in FL and FBM B cells, we performed CDR3 fragment length analysis of VH1-VH6 families in the B cells. To gain further insights into the ontogeny of IgH diversification, we employed 454 next generation sequencing (NGS)-based analysis of the IgVH-Cmu mRNA repertoire in FL, FBM and postnatal B cells. In total, 20 libraries generated from 17 individual, flow-sorted CD34negCD19pos B cell samples were sequenced: 5 FL (4 performed in independent duplicate libraries; GA 14-18 weeks), 3 FBM (GA 13-17 weeks), 3 cPB (age 4mo-4yrs) and 5 aPB (age 29-53 yrs) samples. Bioinformatic analysis was performed with the ARResT/Interrogate immunoprofiler, IMGT tools, and appropriate statistics. Results: Spectratyping and NGS data showed that the molecular mechanisms responsible for VDJ recombination-dependent repertoire diversification, such as convergent recombination, are active in early B cell development and as efficient as postnatal B cells in both FL and FBM. Comparably diversified B cell lymphopoiesis exists contemporaneously in FL and FBM. However, analysis of the top 100 most abundant clonotypes in each developmental stage showed that their mean abundance in FBM B cells was 10 times lower (0.12%) than in FL B cells (1.2%; p Conclusions: The lack of mature, antigen-responsive B cells in FBM and the pronounced clonotypic expansions in FL suggest that FL is the main source of IgM natural immunity during the 2nd trimester. Our results are consistent with persisting fetal IgM+ B cells being an important source of the natural IgM repertoire also in adult life. Further, the origins of specific stereotypic IgM+ BCR associated with CLL, can be traced back to fetal B cell lymphopoiesis, suggesting that persisting fetal B cells can be subject to malignant transformation late in life. Overall, these novel data provide unique insights into the ontogeny of physiological and malignant B lymphopoiesis that spans the human lifetime. Disclosures No relevant conflicts of interest to declare.

The process of allogeneic hematopoietic cell transplantation (HCT) is often associated with poor oral intake and as a result, nutritional status declines. Although it might seem obvious that optimal nutrition is likely to improve outcomes of transplantation, there are no clinical data that directly support this assumption. It is also unclear whether artificial nutrition support (ANS) should be provided as enteral tube feeding (ETF) or parenteral nutrition (PN). In this study we analysed day 100 non-relapse mortality (NRM) and incidence and severity of acute graft versus host disease (GvHD) according to both the route and adequacy of nutritional intake together with other known prognostic factors. A total of 484 patients who underwent HCT for hematological malignancy at a single institution between 2000 and 2014 were included. Myeloablative conditioning was used in 285 (59%) patients, 272 of whom received a TBI based regimen. Reduced intensity conditioning was given to 199 (41%) patients. For the 236 (49%) unrelated donor cell recipients in vivo T-cell depletion with alemtuzumab was used. Myeloablative HCT recipients were advised to have an enteral feeding tube inserted prophylactically at day 1 post HCT. ANS was initiated in 228 (47%) patients who met pre-defined feeding criteria that include actual or anticipated oral intake below 1/3 of estimated requirement for 5 or more days or significant weight loss or low body mass index. Total of 245 (51%) patients had adequate enteral nutrition (EN) either orally (N=198) or with use of ETF (N=47). Patients in whom ETF could not be established received PN (N=148, 31%) in order to provide adequate nutrition. The remaining 91 (19%) patients had inadequate nutrition due to either curtailed ANS (N=33) or a failure to start ANS because of a lack of feeding access via any route (N=58). The effects of patient, disease and transplant factors were studied in univariate analyses on NRM, acute GvHD and engraftment, following which multivariate analyses showed significant associations of NRM with age > 50 years (hazard ratio (HR) 2.3; 95% confidence interval (CI) 1.4 - 3.7; P=0.001); previous autograft (HR 2.4; 95% CI 1.3 - 4.5; P=0.007) and positive recipient CMV serology (HR 1.9; 95% CI 1.1 - 3.2; P=0.019). In addition, HRs were significantly increased in the PN and inadequate nutrition groups compared to those with adequate enteral intake: adequate PN: HR 3.2; 95% CI 1.7 - 6.0, inadequate nutrition: HR 4.4; 95% CI 2.4 - 8.0; all P grade 1 in patients who received PN (HR 2.0; 95% CI 1.2 - 3.3; P=0.006, and HR 1.8; 95% CI 1.1 - 3.0; P=0.018, respectively), but not in those with inadequate nutrition compared to adequate EN. Other significant covariates in the model for both increased overall and gut GvHD were the use of myeloablative versus reduced intensity conditioning P=0.001 and P In conclusion the data show that adequate nutrition during allogeneic HCT is associated with improved non-relapsed mortality at 100 days. Adequate EN is associated with significantly better results for this outcome than adequate PN. Furthermore adequate EN, predominantly via oral intake may be associated with lower incidence of overall and gut GvHD when compared to PN, perhaps because of its ability to maintain gut mucosal integrity and for support of the gastrointestinal tract environment, including gut microflora. Although the retrospective nature of this study can only indicate association, the significant hazard ratios reported warrant further research into optimising enteral nutrition in recipients of HCT. Figure Figure. Disclosures Milojkovic: Novartis: Honoraria; Bristol Myers Squibb: Honoraria; Pfizer: Honoraria; Ariad: Honoraria. Apperley:Incyte: Speakers Bureau; Novartis: Honoraria, Speakers Bureau; Pfizer: Honoraria, Speakers Bureau; Bristol Myers Squibb: Honoraria, Speakers Bureau; Ariad: Honoraria, Speakers Bureau.

Identification of prognostic indicators to predict outcomes of hematopoietic cell transplantation (HCT) is important for selection of patients most likely to benefit from the procedure. Widely validated systems such as the HCT-Comorbidity Index (HCT-CI) and the modified EBMT risk score provide consistent prediction of outcomes, in particular non-relapse mortality (NRM) and survival. In this retrospective study we identified neutropenia prior to initiation of conditioning regimen as an additional prognostic factor in 117 adults with high risk acute myeloid leukemia (AML) who received HCT at a single institution from January 2005 to December 2015. The main endpoints of the study were day 100 non-relapse mortality and survival at 5 years. The median age at HCT was 50 years (range 21-68); 65% had an EBMT score of ≥4. In total 107 (91%) patients had poor risk disease, with classification for poor risk including; patients undergoing HCT due to failure to achieve complete remission (CR) after induction chemotherapy, patients in 2nd CR, patients in 1st CR with adverse karyotype, patients with standard karyotype but with a FLT-3 mutation, patients with measurable residual disease (MRD) by immunophenotyping or molecular monitoring and patients with secondary AML. Conditioning was myeloablative (Cy-TBI: N=41, Bu-Cy: N=3) for patents younger than 51 years with good performance status; the intensity of conditioning was reduced for the others (N=73). Standard methotrexate and cyclosporine was used as graft versus host disease prophylaxis in recipients of sibling cells, and unrelated graft recipients also received alemtuzumab. With a median follow-up of 3.1 years (range 0.3 - 10.0), day 100 NRM for the whole group was 17.3% (95% confidence interval (CI) 11 - 25) and the probability of survival at 5 years was 40.3% (95% CI 31 - 50). The effects of patient, disease and transplant factors were analysed in univariate analyses on both NRM and survival. Significant factors were then entered into multivariate analyses which yielded significant associations with improved NRM for preconditioning absolute neutrophil count (ANC) >= 1000/microL and HCT-CI < 3 (Figure) (hazard ratio (HR) 0.36; 95% CI 0.1 - 0.9 and HR 0.39; 95% CI 0.2 - 0.9 respectively). The factors associated with improved survival in multivariate analysis included preconditioning neutrophil count >= 1000/microL (HR 2.9; 95% CI 1.6 -5.4, P= 0.01), absence of any measurable disease prior to conditioning (HR 2.1; 95% CI 1.1 -4.2, P= 0.03), and less than 2 disease risk defining criteria (HR 15.8; 95% CI 2.0 -124.4, P= 0.01). In conclusion, patients with ANC of 1000/microL and above prior to initiation of conditioning and those with HCT-CI below 3 had significantly better day 100 NRM. ANC of 1000/microL or above was also associated with better survival at 5 years together with disease specific factors, absence of measurable residual disease prior to initiation of condition and less than two adverse prognostic factors. A simple measurement such as the neutrophil count helps to predict outcome: this might reflect the presence of on-going disease but it is also possible that delaying transplant until neutrophil count recovery might impact favourably on the results. Our findings therefore advocate integration of preconditioning neutrophil count into existing risk assessment tools when considering HCT. Figure Figure. Disclosures Milojkovic: Bristol Myers Squibb: Honoraria; Pfizer: Honoraria; Ariad: Honoraria; Novartis: Honoraria.

The bromodomain and extraterminal (BET) protein BRD2-4 inhibitors hold therapeutic promise in preclinical models of hematologic malignancies. However, translation of these data to molecules suitable for clinical development has yet to be accomplished. Herein we expand the mechanistic understanding of BET inhibitors in multiple myeloma by using the chemical probe molecule I-BET151. I-BET151 induces apoptosis and exerts strong antiproliferative effect in vitro and in vivo. This is associated with contrasting effects on oncogenic MYC and HEXIM1, an inhibitor of the transcriptional activator P-TEFb. I-BET151 causes transcriptional repression of MYC and MYC-dependent programs by abrogating recruitment to the chromatin of the P-TEFb component CDK9 in a BRD2-4-dependent manner. In contrast, transcriptional upregulation of HEXIM1 is BRD2-4 independent. Finally, preclinical studies show that I-BET762 has a favorable pharmacologic profile as an oral agent and that it inhibits myeloma cell proliferation, resulting in survival advantage in a systemic myeloma xenograft model. These data provide a strong rationale for extending the clinical testing of the novel antimyeloma agent I-BET762 and reveal insights into biologic pathways required for myeloma cell proliferation.

Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired clonal disorder of the hematopoietic stem cell (HSC). Somatic mutations in thePIG-A gene result in the deficiency of several glycosylphosphatidylinositol-linked proteins from the surface of blood cells. This explains intravascular hemolysis but does not explain the mechanism of bone marrow failure that is almost invariably seen in PNH. In view of the close relationship between PNH and idiopathic aplastic anemia (IAA), it has been suggested that the 2 disorders might have a similar cellular pathogenesis, namely, that autoreactive T-cell clones are targeting HSCs. In this paper, we searched for abnormally expanded T-cell clones by size analysis of the complementarity-determining region 3 (CDR3) in the beta variable chain (BV) messenger RNA (mRNA) of the T-cell receptor (TCR) in 19 patients with PNH, in 7 multitransfused patients with hemoglobinopathy. and in 11 age-matched healthy individuals. We found a significantly higher degree of skewness in the TCR BV repertoire of patients with PNH, compared with controls (R2 values 0.82 vs 0.91,P

Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired clonal disorder of the hematopoietic stem cell (HSC). Somatic mutations in thePIG-A gene result in the deficiency of several glycosylphosphatidylinositol-linked proteins from the surface of blood cells. This explains intravascular hemolysis but does not explain the mechanism of bone marrow failure that is almost invariably seen in PNH. In view of the close relationship between PNH and idiopathic aplastic anemia (IAA), it has been suggested that the 2 disorders might have a similar cellular pathogenesis, namely, that autoreactive T-cell clones are targeting HSCs. In this paper, we searched for abnormally expanded T-cell clones by size analysis of the complementarity-determining region 3 (CDR3) in the beta variable chain (BV) messenger RNA (mRNA) of the T-cell receptor (TCR) in 19 patients with PNH, in 7 multitransfused patients with hemoglobinopathy. and in 11 age-matched healthy individuals. We found a significantly higher degree of skewness in the TCR BV repertoire of patients with PNH, compared with controls (R2 values 0.82 vs 0.91,P

A most distinctive feature of paroxysmal nocturnal hemoglobinuria (PNH) is that in each patient glycosylphosphatidylinositol-negative (GPI–) and GPI+ hematopoietic stem cells (HSCs) coexist, and both contribute to hematopoiesis. Telomere size correlates inversely with the cell division history of HSCs. In 10 patients with hemolytic PNH the telomeres in sorted GPI– granulocytes were shorter than in sorted GPI+ granulocytes in 4 cases, comparable in 2 cases, and longer in the remaining 4 cases. Furthermore, the telomeres of both GPI– and GPI+ hematopoietic cells were markedly shortened compared with age-matched controls. The short telomeres in the GPI– cells probably reflect the large number of cell divisions required for the progeny of a single cell to contribute a large proportion of hematopoiesis. The short telomeres of the GPI+ cells indicate that the residual hematopoiesis contributed by these cells is not normal. This epigenetic change is an additional feature shared by PNH and aplastic anemia.

Copyright @ 2013 Macmillan Publishers Limited. This is the author's accepted manuscript. The final published article is available from the link below. Regulation of cell survival is a key part of the pathogenesis of multiple myeloma (MM). Jun N-terminal kinase (JNK) signaling has been implicated in MM pathogenesis, but its function is unclear. To elucidate the role of JNK in MM, we evaluated the specific functions of the two major JNK proteins, JNK1 and JNK2. We show here that JNK2 is constitutively activated in a panel of MM cell lines and primary tumors. Using loss-of-function studies, we demonstrate that JNK2 is required for the survival of myeloma cells and constitutively suppresses JNK1-mediated apoptosis by affecting expression of poly(ADP-ribose) polymerase (PARP)14, a key regulator of B-cell survival. Strikingly, we found that PARP14 is highly expressed in myeloma plasma cells and associated with disease progression and poor survival. Overexpression of PARP14 completely rescued myeloma cells from apoptosis induced by JNK2 knockdown, indicating that PARP14 is critically involved in JNK2-dependent survival. Mechanistically, PARP14 was found to promote the survival of myeloma cells by binding and inhibiting JNK1. Moreover, inhibition of PARP14 enhances the sensitization of MM cells to anti-myeloma agents. Our findings reveal a novel regulatory pathway in myeloma cells through which JNK2 signals cell survival via PARP14, and identify PARP14 as a potential therapeutic target in myeloma. Kay Kendall Leukemia Fund, NIH, Cancer Research UK, Italian Association for Cancer Research and the Foundation for Liver Research.

Abstract 876 Children with Down syndrome (DS), under the age of 4 years, have a 150-fold excess risk of myeloid leukaemia (DS-ML), which arises in utero and is of erythro-megakaryocyte phenotype. Two known factors required for leukemic transformation- are: (i) presence of trisomy 21 (T21) in fetal liver (FL) hematopoietic cells and (ii) acquired N terminal-truncating mutation(s) of the transcription factor gene GATA1. Together, they lead to the preleukemic condition, Transient Abnormal Myelopoiesis (TAM). In humans, T21 provides the crucial initiating perturbation of hematopoiesis by inducing a marked expansion of FL megakaryocyte-erythroid progenitors (MEP) during the second trimester[1]. Mouse models have suggested that effects of T21 are confined to FL and are lost post-natally[2]. However, retrospective studies have reported blood counts abnormalities in some neonates with DS suggesting abnormal fetal hematopoiesis may persist beyond the second trimester into post-natal life. To investigate this, blood counts and smears in the first week of life from the first 100 babies with DS enrolled in a prospective, longitudinal study have been compared with blood counts and smears from normal term cords (n=100) and normal term and preterm neonates in the first week of life (n=200). 11/100 neonates with DS had a diagnosis of Transient Abnormal Myelopoiesis (TAM) by WHO criteria and confirmed by GATA1 mutation; 22 (2 with TAM) were preterm- gestational age (GA)< 37 weeks). In neonates with DS, with or without TAM, Hb concentrations, hematocrits and RBC counts were higher than GA matched neonates without DS; 51 (57.3%) of babies with DS had a Hb concentration of >20 g/dl (vs 2% of non DS neonates); only 2 (2.2%) of DS neonates had a Hb Although total white cell, neutrophil, lymphocyte and eosinophil counts in DS neonates without TAM were not significantly different from normal term cord or GA-matched peripheral blood samples, monocyte (median 0.97 vs 0.7 × 109/L) and basophil counts (median 0.2 vs 0 × 109/L) were increased. Dysplastic features, (abnormal granulation and nuclear morphology) were seen in neutrophils, monocytes and basophils in ∼80% of DS neonates without TAM and in 11/11 with TAM; these findings were rare in term and preterm neonates without DS. Circulating blasts were seen on blood smears from all neonates with DS both with TAM (median 34%; range 15–54%) and without TAM (median 3%; range 0.5– 11%) but rarely in normal controls (median 0, range 0–8%). Many, but not all, blasts in neonates with DS without TAM were morphologically megakaryoblasts and indistinguishable from those in TAM. Thus, abnormalities of hematopoiesis affecting all myeloid lineages are present in most, if not all, neonates with DS, regardless of whether or not they have TAM and GATA1 mutation(s). Consistent with this, we found that although cord blood from DS neonates without TAM had normal overall numbers of MEP, CMP and GMP compared to normal term cord, clonogenicity of megakaryocyte and erythroid progenitors cultured from flow sorted DS cord blood MEP and CMP was markedly increased compared to normal term cord blood (MEP 55 vs 18%; CMP 67 vs 10%; DS vs normal) whilst DS cord blood HSC and MPP populations gave rise to increased numbers of GM and myeloid blast colonies. These data show that in humans with DS, trilineage perturbation of fetal hematopoiesis persists into post-natal life, in contrast to mouse models. 1. Tunstall-Pedoe, O., et al., Abnormalities in the myeloid compartment in Down Syndrome fetal liver precede acquisition of GATA1 mutations. Blood, 2008. 112: 4507–11 2. Alford, K.A., et al., Perturbed hematopoiesis in the Tc1 mouse model of Down syndrome. Blood, 2010. 115: 2928-37 Disclosures: No relevant conflicts of interest to declare.

Introduction: Intracellular protein homeostasis requires a well-controlled protein degradation machinery to clear damaged and old proteins and to replenish intracellular amino acid pools. Cancer cells may be particularly susceptible to agents that disrupt protein degradation because of unbalanced protein levels related to gain or loss of genetic material, high protein turnover, or high-level production of specific proteins such as Ig in multiple myeloma (MM). The AAA ATPase VCP (p97) is a master regulator of protein degradation that has been implicated in oncogenesis. Small molecule VCP inhibition (VCP-i) rapidly activates caspases in cancer cells, induces endoplasmic reticulum (ER) stress, and has anti-tumor activity in murine xenograft models. A phase 1 trial of VCP-i in relapsed MM has recently opened. Methods: We investigated the cellular mechanisms that govern VCP-i mediated cancer cell death in MM cell lines including bortezomib-adapted AMO1 cells, MM cells from patients with bortezomib-resistant MM, as well as lung cancer (A549) and osteosarcoma (Saos2) cells. Results: The ATP-competitive inhibitor DBeQ and the allosteric inhibitor NMS873 induced cell line as well as primary MM cell death at similar low micromolar concentrations independently of bortezomib sensitivity. DBeQ and NMS873 caused phosphorylation of eIF2alpha in a time- and dose-dependent manner and resulted in strong transcriptional and translational up-regulation of ATF4, CHOP and GADD34. VCP-i also increased expression of ER chaperones BiP and p58IPK. Inhibition of eIF2alpha de-phosphorylation with guanabenz reduced S6 phosphorylation, a marker of protein translation, and increased A549 and OPM2 cell survival early (8h) after VCP-i. Direct inhibition of protein translation with cycloheximide also decreased early VCP-i mediated cell death. Using MEFs deficient in eIF2alpha kinases we show that eIF2alpha phosphorylation following VCP-i depends on both the unfolded protein response mediator PERK and the nutrient sensor GCN2. We found that DBeQ induces GCN2 phosphorylation in parallel with loss of mTORC1 signalling, induction of the key autophagy factor p62, and accumulation of LC3-II. DBeQ also induced a rapid decrease in free intracellular L-amino acids. Depletion of selected amino acids in the cell culture medium increased cell death and mRNA levels of CHOP and GADD34 following VCP-i with DBeQ or NMS873, but not following induction of ER stress with tunicamycin. Conclusion: Collectively, these data show that both ATP-competitive and allosteric VCP-i effectively kills cancer cells independently of bortezomib sensitivity. Disrupting VCP function induces early cell death via inappropriate eIF2alpha-regulated protein translation downstream of GADD34. VCP-i also depletes the intracellular free amino acid pool, resulting in cell death despite compensatory catabolic processes. Citation Format: Katarzyna Parzych, Sandra Loaiza, Tamara M. Chinn, Philippa May, Florentina Porsch, Anastasios Karadimitris, Christoph Driessen, Heather P. Harding, David Ron, Holger W. Auner. Comprehensive failure of intracellular protein homeostasis kills myeloma and solid cancer cells following VCP/p97 inhibition. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1261. doi:10.1158/1538-7445.AM2015-1261

In Down syndrome (DS), trisomy 21 (T21) causes perturbation of fetal liver (FL) hematopoiesis leading to expansion of megakaryocyte-erythroid (MK-E) progenitors. On this background, FL hematopoietic stem/progenitor cells (HSPC) may acquire GATA1 gene mutations which usually present as a neonatal preleukemic condition, Transient Myeloproliferative Disorder (TMD). TMD develops only in fetal life and most cases spontaneously regress after birth, suggesting unique features of the T21 fetal microenvironment may be important in driving both abnormal DS FL hematopoiesis and mutant GATA1 clones in TMD. Previous work showed activation of insulin-like growth factor (IGF) signalling in TMD and DS-AMKL and fetal-specific IGF dependence of murine MK development implicating T21 in altered fetal IGF signalling. To investigate the role of IGF signalling in fetal MK-E development in DS FL without GATA1 mutations, we performed gene expression profiling (GEP) of DS FL (n=5) and normal FL CD34+ cells (n=3). GSEA showed significant enrichment of IGF targets. We therefore measured expression of the principal IGF receptor, IGF1R, on DS FL HSPC (n=6). Surface IGF1R was uniformly expressed on all DS FL HSPC subpopulations. However, there was no difference in IGF1R surface or gene expression between DS and normal FL HSPC (n=8) suggesting that increased IGF1R expression on TMD and DS-AMKL cells is an early event and may reflect their fetal rather than T21 origin. To determine if the enrichment of IGF targets in DS FL HSPC might reflect increased IGF production in the FL microenvironment, we compared IGF expression in DS FL mesenchymal stromal cells (MSC)(n=4) and hepatocytes with normal FL MSC (n=4). Both DS and normal FL MSC expressed large amounts of IGF2, but little or no IGF1 with no difference between DS and normal MSC. Similarly, IGF2 was expressed by IHC at similar levels in both DS and normal fetal hepatocytes and MSC whereas IGF1 was barely detectable. Normal and DS fetal bone marrow (BM) MSC also expressed high levels of IGF2, but not IGF1 in contrast to adult BM MSC. Thus, IGF2 is the main IGF produced in the human fetal hematopoietic environment. Consistent with a specific role in DS FL hematopoiesis, IGF2 caused marked proliferation of DS FL clonogenic MK and MK-E cells (n=3) and stimulated MK proliferation in liquid culture while in normal FL, IGF2 stimulated EPO-independent BFU-E and Pre-BFU-E but had no effect on MK cells. Since differential effects of IGF2 on DS/normal FL HSPC might be due to altered expression of the Insulin Receptor (InsR), which binds IGF2 with low affinity, we compared InsR surface and gene expression in DS and normal FL HSPC but found no significant difference. IGF2 also binds to IGF2R, a negative regulator of IGF2, however there was no difference in IGF2R expression between DS and normal FL HSPC. Co-culture of normal FL CD34+ cells with DS or normal FL MSC (n=3) promoted MK and erythroid cell proliferation and differentiation. Since the effect of DS FL MSC on normal FL CD34+ cells was significantly greater (~4-fold) than that of normal FL MSC, the data above suggest this is unlikely to be due to IGF2 alone. To investigate whether T21 alters the MSC secretome, we performed GEP of DS FL MSC (n=3) and compared this with normal FL MSC (n=5). Both normal FL and DS FL expressed known hematopoietic cytokine genes, including SCF, CXCL12, G-CSF, M-CSF, Flt3 and IL-6, as well as IGF2 but not IGF1, EPO or TPO. Of the top 80 differentially expressed genes (Fold Change>2), ~40% encode secreted proteins, including several which affect IGF availability. Conclusion: T21 perturbs FL hematopoiesis through both HSPC-intrinsic mechanisms, including changes in IGF responsiveness, and through alterations to the FL microenvironment. Disclosures No relevant conflicts of interest to declare.

Invariant NKT cells (iNKT cells) are a small subset of immunoregulatory T cells highly conserved in humans and mice. Upon activation by glycolipids presented by the MHC-like molecule CD1d, iNKT cells promptly secrete Th1/2 cytokines but also cytokines with hematopoietic potential such as IL-3 and GM-CSF. In mice, NKT cells activated by alpha-galactosylceramide (alphaGC), a potent glycolipid ligand, cause an in increase in extramedullary hematopoietic committed progenitor activity through secretion of these cytokines. We tested the role of iNKT cells in regulating hematopoiesis under conditions of activation and in steady state hematopoiesis. We found that GM-CSF-secreting alphaGC -activated iNKT cells enhanced (by 64%, n=5, p In conclusion, iNKT cells when activated enhance the clonogenic capacity of myeloid progenitors and can thus modulate innate immune responses. Additionally, they are required for maintenance of normal hematopoiesis in the absence of immune activation. Our findings offer further evidence for a concerted regulation of the immune and hematopoietic systems and the potential for new therapeutic approaches for the manipulation of hematopoiesis.

Despite the therapeutic advances that followed the introduction of new immunomodulatory drugs and proteasome inhibitors, multiple myeloma (MM) remains an incurable malignancy and eventually all patients develop and succumb to chemo-refractory disease. The recently developed bromodomain and extra terminal (BET) protein inhibitors are novel agents targeting the acetyl-binding pockets of the BET family proteins BRD2-4 and BRDT. BET proteins activate transcription through their ability to bind to acetyl-modified lysine residues of histone tails, thereby serving as chromatin scaffolds that recruit the P-TEFb and PAFc1 complexes to Polymerase II (RNA Pol II), ensuring transcriptional initiation and elongation. In preclinical models, two classes of BET inhibitors, benzodiazepines (e.g JQ1) and quinolones (e.g. I-BET151), have been shown to have significant anti-proliferative activity against a variety of hematologic tumours. However translation of these data to molecules suitable for clinical development has yet to be disclosed. Herein, we tested the anti-myeloma activity and extended the mechanistic insights on two BET inhibitors: the chemical probe molecule I-BET151 and I-BET762, an orally active benzodiazepine suitable for clinical development. I-BET151 was tested in vitro in 6 myeloma cell lines (MMCL) with cytogenetic profiles representative of the most common translocations found in MM. I-BET151 induces apoptosis and cell cycle arrest in all MMCL in a time-dependent manner, with IC50 ranging from 133nM to 411nM at 72hrs. With the exception of KMS11 cells, IC50 was similar in stroma-free conditions and in co-culture with MS5 stromal cells. Similarly, I-BET151 induces apoptosis and cell cycle arrest in primary MM cells (n=4) cultured in the presence of IL-6 and stroma. In a subcutaneous MM mouse model, as compared to placebo, treatment with I-BET151 30mg/Kg/day i.p for 21 days resulted in 4-5 fold reduction in tumour size (p In conclusion, I-BET151 and I-BET762 show potent anti-myeloma activity in vitro and in vivo. I-BET151 inhibition of BRD2-4 binding mediates MYC transcriptional silencing and cell cycle arrest, but MYC-independent mechanism are also likely to mediate the I-BET biological effects in MM. Our data is the first example of an orally active BET inhibitor significantly delaying MM progression in vivo and provides strong rationale for clinical testing in phase I/II trials. Disclosures: Tough: GlaxoSmithKline: Employment. Smithers:GlaxoSmithKline: Employment. Bassil:GlaxoSmithKline: Employment. Chapman:GlaxoSmithKline: Employment. Harker:GlaxoSmithKline: Employment. Barbash:GlaxoSmithKline: Employment. Tummino:GlaxoSmithKline: Employment. Al-Mahdi:GlaxoSmithKline: Employment. Haynes:GlaxoSmithKline: Employment. Cutler:GlaxoSmithKline: Employment. Le:GlaxoSmithKline: Employment. Witherington:GlaxoSmithKline: Employment. Parr:GlaxoSmithKline: Employment. Prinjha:GlaxoSmithKline: Employment.

Inhibition of the proteasome is a widely used strategy for treating multiple myeloma that takes advantage of the heavy secretory load that multiple myeloma cells (MMCs) have to deal with. Resistance of MMCs to proteasome inhibition has been linked to incomplete disruption of proteasomal endoplasmic-reticulum (ER)-associated degradation (ERAD) and activation of non-proteasomal protein degradation pathways. The ATPase p97 (VCP/Cdc48) has key roles in mediating both ERAD and non-proteasomal protein degradation and can be targeted pharmacologically by small molecule inhibition. In this study, we compared the effects of p97 inhibition with Eeyarestatin 1 and DBeQ on the secretory apparatus of MMCs with the effects induced by the proteasome inhibitor bortezomib, and the effects caused by combined inhibition of p97 and the proteasome. We found that p97 inhibition elicits cellular responses that are different from those induced by proteasome inhibition, and that the responses differ considerably between MMC lines. Moreover, we found that dual inhibition of both p97 and the proteasome terminally disrupts ER configuration and intracellular protein metabolism in MMCs. Dual inhibition of p97 and the proteasome induced high levels of apoptosis in all of the MMC lines that we analysed, including bortezomib-adapted AMO-1 cells, and was also effective in killing primary MMCs. Only minor toxicity was observed in untransformed and non-secretory cells. Our observations highlight non-redundant roles of p97 and the proteasome in maintaining secretory homeostasis in MMCs and provide a preclinical conceptual framework for dual targeting of p97 and the proteasome as a potential new therapeutic strategy in multiple myeloma.

Abstract 977 Transcriptionally active genes, including housekeeping genes, are characterised by co-occupancy and antagonistic actions of histone deacetylases (HDAC) and acetyltransferases (HAT). Transcription is facilitated through the prevailing action of HAT, which maintain histone acetylation. HDAC inhibitors (HDACi) induce widespread histone hyperacetylation and as a consequence are expected to increase expression of transcriptionally active genes. We have previously demonstrated this effect in inherited glycosylphosphatidylinositol (GPI-anchor) deficiency, an autosomal recessive disorder characterised by histone hypoacetylation and transcriptional repression of PIGM due to an in cis mutation which disrupts binding of the transcription factor Sp1 to its core promoter cognate motif. We surmised HDACi-mediated hyperacetylation might lead to increased transcription of other housekeeping genes, such as those of the anaerobic glycolytic and pentose phosphate pathways (GPPP) disruption of which is ameliorated by relatively modest increases in enzymatic activity. HDACi could therefore be of therapeutic value in these disorders. To address these hypotheses, EBV B cell lines were treated with the HDACi sodium butyrate (NaBu; 3mM) and mRNA levels for GPPP genes assessed by RQ-PCR. Of 9 genes tested (glucose-6-phosphate dehydrogenase, G6PD; glucose-6-phosphate isomerase, GPI; triosephosphate isomerase, TPI; pyruvate kinase, PK; 6-phosphogluconolactonase, PGLS; phosphogluconate dehydrogenase, PGD; ribulose-5-phosphate-3-epimerase, RPE; ribose-5-phosphate isomerase A, RPIA; transketolase, TKT), only G6PD mRNA levels increased, in a time-dependent fashion, in response to NaBu (n=3; p The epigenetic correlates of G6PD mRNA induction were assessed by ChIP-Q-PCR. This revealed a dynamic, time-dependent, 3 to 4-fold increase in levels of histone 3 and 4 acetylation and a 4-fold increase in Sp1 and Polymerase II occupancy in the promoter of G6PD but not TPI or GPI. Preliminary pharmacological and shRNA experiments suggest that HDACi-mediated transcriptional upregulation of G6PD is Sp1-dependent. No differences were observed in baseline levels of histone acetylation or Sp1 occupancy of G6PD, TPI and GPI implying a yet to be defined cis acting determinant is required for the selective increase in Sp1 binding and histone hyperacetylation. Finally, in erythroblasts generated from peripheral blood mononuclear cells of two patients with Class I G6PD deficiency (G6PD Brighton and G6PD Harilaou), we confirmed the ability of NaBu to increase mutant G6PD mRNA and protein levels leading to increased G6PD enzymatic activity and its restoration to normal within 24hrs. In conclusion, we show that even within the same metabolic pathway, transcriptional upregulation of active genes in response to HDACi is selective rather than universal and is underpinned by enhanced Sp1 binding and histone hyperacetylation. Our findings raise the prospect of using HDACi to treat severe Class I G6PD deficiency. Disclosures: No relevant conflicts of interest to declare.