ALBERT

Description: Introduction: Acetylated histone and non-histone proteins are pharmacologic targets for both solid and hematological cancers including myeloproliferative neoplasms (MPNs), a group of clonal hematological malignancies driven by aberrant JAK2/STAT signaling. MPNs are characterized by epigenetic alterations, including aberrant acetylation, which makes this disease particularly interesting for targeting with HDAC inhibitors. Four classes of histone deacetylases (Class I-IV HDACs) regulate gene transcription and modulate cellular processes that drive the initiation and progression of cancer. Pan-HDAC and class I-selective HDAC inhibitors have gained traction in clinical settings, yet we reasoned that specific targeting of the 18 distinct HDAC proteins may establish roles for select HDACs as therapeutic vulnerabilities in MPNs. Methods: To explore the roles of individual HDACs in MPN, we first conducted an inhibitor screen of compounds having distinct HDAC selectivity based on electrophoretic mobility shift assays with full-length human HDAC proteins expressed in baculovirus and unique peptide substrates. Ultra-specific HDAC6 compounds were initially targeted for analysis based on its previously defined role in HSP90-mediated JAK2 stabilization and translation. Survival of MPN cell line models, MPN patient samples, leukemia cell lines, and MPN disease progression in mice transplanted with Hdac6-/-, and Hdac11-/- hematopoietic stem cells (HSCs) transduced with the MPLW515L oncogene, as well as Tg-Hdac11-eGfp mice were used to show the role of HDAC6 and HDAC11 in oncogene-driven and homeostatic hematopoiesis. As further proof of specificity, HDAC6 and HDAC11 were genetically ablated in MPN model cell lines using either RNA interference or inducible shRNA. For HDAC11 substrate identification, a combination of RNA-seq, acetylated proteome (SILAC), global metabolomics (LC-MS), Seahorse metabolic assays (Agilent Technologies), enzymatic assays, and acetylation-specific immunoblotting and mutation profiling were performed (Fig. 1). Results: Despite the established interplay between HDAC6, HSP90 and JAK2, neither a highly selective HDAC6 inhibitor, HDAC6 silencing, nor the Hdac6 deficiency suppressed MPN pathogenesis, although there were clear effects on the acetylation of α-tubulin, a well characterized HDAC6-selective substrate. Intriguingly, both inhibition of HDAC11 activity with highly-specific HDAC11 inhibitors and silencing HDAC11 using an inducible validated shRNA, identified HDAC11 as a therapeutic vulnerability for multiple human MPN cell lines. The Tg-Hdac11-eGFP reporter mice showed that HDAC11 is expressed in several hematopoietic cell types, including myeloid cells, erythroblasts, and megakaryocytes. Thus, Hdac11-/- and Hdac11+/+MPLWT bone marrow were examined for steady-state hematopoiesis and transplantation chimerism. These studies demonstrated that HDAC11 does not contribute to homeostatic or transplantated bone marrow reconstitution. However, in the oncogenic MPL model, recipient mice transplanted withoncogenic MPLW515L-expressing Hdac11-deficient HSCs displayed markedly impaired cytokine-independent colony-formation, had less fibrosis, and displayed improved survival in primary and secondary MPN hematopoietic stem cell transplantation; thus HDAC11 contributes to MPN pathogenesis (Fig. 1). Studies in additional leukemia cell lines, including THP-1, HL-60, and mantle lymphoma cell lines, but not in Ramos or K562 cells, established that HDAC11 contributes to oncogene-driven events in other cell types. Mechanistically, RNA-seq, SILAC proteomics, and metabolic profiling revealed that HDAC11 controls aerobic glycolysis by deacetylating Lys343 of the glycolytic enzyme enolase-1 (ENO1), functionally inactivating ENO1. Finally, the effects of targeting HDAC11 on metabolism were augmented by blocking compensatory pathways of oxidative phosphorylation that are induced via JAK2V617Fand MPLW515L oncogenic signaling. Conclusions: Our comprehensive screens of HDAC inhibitors, coupled with our biological, in vivo and molecular studies, indicate that HDAC11 is an attractive and potent target for disabling MPN metabolism and pathogenesis. These finding support the rationale for further development of clinical HDAC11 inhibitors for the treatment of metabolically-active cancers such as MPNs. Disclosures Pinilla Ibarz: Teva: Consultancy; TG Therapeutics: Consultancy; Sanofi: Speakers Bureau; Bayer: Speakers Bureau; Novartis: Consultancy; Bristol-Myers Squibb: Consultancy; Abbvie: Consultancy, Speakers Bureau; Takeda: Consultancy, Speakers Bureau; Janssen: Consultancy, Speakers Bureau. Reuther:Incyte Corporation: Research Funding. Levine:Loxo: Membership on an entity's Board of Directors or advisory committees; Roche: Consultancy, Research Funding; Lilly: Honoraria; C4 Therapeutics: Membership on an entity's Board of Directors or advisory committees; Isoplexis: Membership on an entity's Board of Directors or advisory committees; Imago Biosciences: Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy; Gilead: Consultancy; Celgene: Consultancy, Research Funding; Qiagen: Membership on an entity's Board of Directors or advisory committees; Prelude Therapeutics: Research Funding; Amgen: Honoraria. Verma:BMS: Research Funding; Janssen: Research Funding; Stelexis: Equity Ownership, Honoraria; Acceleron: Honoraria; Celgene: Honoraria. Epling-Burnette:Incyte Corporation: Research Funding; Celgene Corporation: Patents & Royalties, Research Funding; Forma Therapeutics: Research Funding.

Description: Myeloproliferative neoplasms (MPNs) are a group of hematopoietic stem cell disorders characterized by the abnormal production of various myeloid cells. Aberrant JAK2 signaling (e.g. induced by JAK2-V617F) plays an etiological role in MPN formation. While JAK2 inhibitors improve patient symptoms, they do not induce cell death as neoplastic cells appear to be rather insensitive to JAK2 inhibition and effectively rapidly become resistant to treatment. Therefore, the development of additional therapeutic approaches for MPNs is needed. Pim kinases are serine/threonine kinases that protect hematopoietic cells from apoptosis and also play a role in regulating hematopoietic stem cell growth. In mouse models, elevated Pim expression contributes to the development of lymphoma. Pims are constitutively active and thus regulated by protein expression, which is controlled by Pim gene expression and Pim protein stability. Pim1 gene expression is normally induced by JAK2/STAT5 signaling in response to extracellular growth factor stimulation, as Pim1 is a direct transcriptional target of STAT5. The deregulated JAK2 signaling in MPNs also induces Pim expression. STAT5 is required for MPN disease in mouse models, suggesting genes transcriptionally regulated by STAT5 are required for MPN disease formation. Together with the anti-apoptotic signaling and transforming properties of Pims, this suggests Pims may play a role in MPNs. We hypothesized that Pim kinases may offer a therapeutic target for MPNs and that Pim kinase inhibitors in combination with JAK inhibitors may cause neoplastic cytotoxicity, improving on current JAK2-inhibitor mono-therapy for MPNs. JAK2-V617F-dependentMPN model cells (HEL, SET2, Uke1, and BaF3/JAK2-V617F, including cells that are resistant to the JAK2 inhibitor ruxolitinib) as well as MPN patient cells, were treated with Pim kinase inhibitors, SGI-1776 and AZD1208, and the JAK2 inhibitor, ruxolitinib. The effects on cell growth, cell cycle, viability, and cell signaling were studied. High concentration SGI-1776 (10 μM) inhibited cell growth and viability of MPN model cells while lower doses (1 and 3 μM) had little effect on the growth and viability of these cells. Combination of 3 μM SGI-1776 with low dose ruxolitinib significantly enhanced growth inhibition and cell death of HEL and SET2 cells. Similar results were obtained with the much more effective and selective Pim inhibitor, AZD1208. We show that ruxolitinib inhibits Pim expression in MPN cells, and Pim expression is restored in ruxolitinib-resistant cells. Importantly, low dose SGI-1776 or AZD1208 (100 nM) re-sensitized ruxolitinib-resistant MPN cells to ruxolitinib treatment. Significantly, as single agents, both SGI-1776 and AZD1208 inhibited erythropoietin-independent erythroid colony formation of primary cells from MPN patients, but not erythroid colonies of normal controls. The combination of AZD1208 and ruxolitinib exhibited enhanced inhibition of colony formation of primary cells from MPN patients compared to treatment with either drug alone. These data indicate that Pim kinase inhibitors in combination with a JAK2 inhibitor may offer a more efficacious therapeutic approach over JAK2 inhibitor mono-therapy for MPNs. Disclosures No relevant conflicts of interest to declare.

Description: The optimism of anti-JAK2 therapy for the treatment of MPNs is largely based on the ability of JAK inhibitors to improve MPN patient constitutional symptoms. Unfortunately, such inhibitors are generally unable to induce remission or even allele burden, suggesting anti-JAK2 based therapies need refinement and optimization. Combination therapies continue to be widely investigated, however, such approaches could have significant complications when translating to patients. These include the difficulty determining proper dosing of each drug in combination in patients, the cost of treatment, and issues with combining drugs developed by different pharmaceutical companies. Recently, we and others have identified anti-BET bromodomain inhibitory properties of various kinase inhibitors, including the JAK2 inhibitors TG101209 and TG101308. Bromodomains are protein-protein interaction motifs that bind to acetylated lysine and, for example, can play roles in regulating gene expression by binding acetylated histones. These dual kinase-BET inhibitors bind to the acetylated lysine-binding pocket of BET bromodomains and thus inhibit the function of such domains. Importantly, cancer cells appear to be particularly sensitive to BET inhibitors compared to normal cells. Combining a BET inhibitor and a JAK2 inhibitor has been shown to be more effective against MPN cells than JAK2 inhibition alone. Our recent identification of dual kinase-BET inhibitors allows for the rational design of drugs with polypharmacology to inhibit more than one class of targets. To this end, we have optimized small molecules for dual anti-JAK2 and anti-BET activity. MA2-014 is a chemical derivative of TG101209 that exhibits similar anti-JAK2 activity, but about ten-fold improved anti-BET activity than TG101209. In fact, the activity of MA2-014 to target BET domains is similar to the prototypical BET inhibitor JQ1. For example, the IC50s of MA2-014 and JQ1 against the second bromodomain of the BET family member BRD4 are each about 20 nM. MA2-014 retains comparable biochemical activity against JAK2 as TG101209 and ruxolitinib (IC50s of low single digit nM, ~0.5 to 3 nM). In MPN cells, however, the ability of MA2-014 to inhibit JAK2-V617F signaling in MPN cells, as measured by P-STAT5, is about ten-fold improved over TG101209, and is comparable to ruxolitinib. Likewise, the ability of MA2-014 to inhibit the expression of c-Myc, which is widely used as a biomarker for BET inhibition, is about ten-fold better than TG101209 in MPN cells. The IC50 for MA2-014 for growth of Uke1 MPN cells is about 200 nM, compared to 500 nM for TG101209. Taken together, these data suggest that MA2-014 is a dual JAK2-BET inhibitor that exhibits superior BET inhibitory activity with similar if not better cellular JAK2 inhibitory activity than TG101209. MA2-014 efficiently inhibited the erythropoietin independent erythroid colony formation of myeloid progenitors from MPN patients, which is a hallmark of these cells and is widely used to test MPN therapeutics. The IC50 of MA2-014 in this assay is 50 nM, essentially identical to ruxolitinib, the only FDA approved JAK2 inhibitor for MPNs. The IC50 of TG101209 to inhibit this colony formation of primary MPN cells is 200 nM, four times greater than MA2-014. Interestingly, JAK2-V617F-driven MPN model cells that are resistant to ruxolitinib retained sensitivity to MA2-014. The IC50s of ruxolitinib and MA2-014 against the growth of BaF3-JAK2-V617F cells are about 100 nM. However, the same cells that are resistant to ruxolitinib (IC50 〉2000 nM) remain sensitive to MA2-014 (IC50 of 240 nM). Finally, in long-term culture assays, we have determined that JAK2-V617F driven MPN Uke1 cells are not able to become resistant to MA2-014 as readily as they do to TG101209 or ruxolitinib. These data suggest MA2-014 may be more resilient to drug resistance, the major hurdle in the clinical effectiveness of JAK2 inhibitors. Collectively, our work demonstrates that rationally-designed polypharmacology may be a novel approach to develop effective therapeutics for cancer, especially diseases that are driven by aberrant kinase signaling and are also sensitive to BET inhibition, as exemplified by MPNs. The use of such an optimized polypharmacologic therapeutic may provide the benefits of combination therapy with fewer complications associated with clinical development. Disclosures No relevant conflicts of interest to declare.

Description: Abstract 4191 JAK2 is a cytoplasmic tyrosine kinase that plays an important role in signaling following activation of various cytokine receptors. JAK2 activation promotes growth, survival, and differentiation of various cell types. Mutation of JAK2 is seen in numerous hematopoietic diseases, most notably in myeloproliferative neoplasms (MPNs). JAK2-V617F is a frequent mutation found in the classical MPNs: polycythemia vera, essential thrombocythemia, and primary myelofibrosis. The single amino acid change of valine to phenylalanine occurs in the pseudokinase domain of JAK2, relieving auto-inhibition of the kinase domain and allowing constitutive kinase activity. Numerous mouse models have demonstrated that JAK2-V617F can induce MPN-like disorders in mice. Thus, this point mutation, as well as other less common JAK2 mutations, is believed to play an important etiologic role in the development of MPNs in humans. The development and use of JAK2 inhibitors in clinical trials has shown promising results, again demonstrating the important role JAK2 plays in MPNs. While the JAK2-V617F mutation, as well as other JAK2 mutations, decreases auto-inhibition of JAK2 kinase activity, it is clear that mutated JAK2 still requires the expression of cytokine receptors to induce activation of transforming signals in hematopoietic cells. Normally, JAK2 binds to homodimeric and heterodimeric cytokine receptors through specific receptor motifs and is activated by various structural changes induced by cytokine stimulation. Following activation it utilizes receptor tyrosines as substrates for phosphorylation, leading to recruitment of downstream signaling molecules, such as STAT5, among others. JAK2 then activates STAT5 via phosphorylation and STAT5 then translocates to the nucleus to regulate transcription of target genes. JAK2-V617F does not require ligand for activation, but still requires the scaffolding function of cytokine receptors to facilitate its full activation and activation of downstream signaling via phosphorylation. Lipid rafts are microdomains of the plasma membrane that are enriched in cholesterol and sphingolipids. They have gained appreciation in signal transduction as sites of localization of signaling mediators, including membrane-bound receptors. Congregation of signaling proteins in lipid rafts within the plasma membrane promotes complex formation and signaling cascade activation. We have recently demonstrated that JAK2 is present in lipid rafts during erythropoietin signaling and that lipid raft integrity is required for erythropoietin-mediated signal transduction (Blood 2009, 114: 292). In our current study, we demonstrate that constitutive JAK-STAT signaling driven by JAK2-V617F is sensitive to lipid raft disruption. Human erythroleukemia (HEL) cells express constitutive activation of JAK-STAT signaling due to the presence of JAK2-V617F. Treatment of these cells with methyl-beta-cyclodextrin to disrupt lipid rafts abolished JAK2, STAT5, and STAT3 activation. Similar results are obtained in other cell lines harboring JAK2-V617F and that exhibit JAK-STAT activation that is dependent on this activated form of JAK2. We also demonstrate that JAK2-V617F co-localizes with lipid rafts, as shown by immunofluorescence, and that this co-localization is abolished by lipid raft disruption. This suggests the loss of JAK2-V617F-mediated JAK-STAT activation we observe following lipid raft disruption may be due to an inhibition of properly localized protein complex formation in the plasma membrane that is necessary for JAK2-V617F signaling. Lipid rafts may provide a site for an accumulation of JAK2-V617F-containing signaling complexes and may be necessary for the cellular signals initiated by JAK2-V617F. Our data show JAK2-V617F-driven JAK-STAT pathway activation is vulnerable to lipid raft disrupting agents and suggest lipid raft integrity as a potential therapeutic target for JAK2-V617F positive neoplasms. Targeting lipid rafts in combination with JAK2 kinase inhibitors may allow for more effective kinase inhibition at lower doses, potentially decreasing undesirable side effects associated with kinase inhibitor treatment. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 2955 Poster Board II-931 Mutational activation of Janus Kinase 2 (JAK2) is an important etiologic factor for the development of myeloproliferative neoplasms (MPNs). JAK2 is mutated in nearly all patients with polycythemia vera and about half of patients with essential thrombocythemia and primary myelofibrosis. The most prevalent mutation of JAK2 is valine 617 mutated to phenylalanine (V617F). The normal function of JAK2 is to interact with and become activated by cytokine receptors following their interaction with ligand. Interestingly, while JAK2-V617F is considered a constitutively activated kinase, there is evidence that it still requires interaction with a cytokine receptor to elicit its transforming signal. When expressed at less than or near endogenous JAK2 levels in hematopoietic cells, JAK2-V617F requires co-expression of a homodimeric receptor in order to become activated, transduce downstream signals, and induce transformation. When expressed at high levels in hematopoietic cells, co-expression of a homodimeric cytokine receptor is not needed. However, a functional cytokine receptor interacting domain is still required, suggesting that even when expressed at high levels JAK2-V617F requires interaction with a receptor. Also, a functional cytokine receptor interacting domain in JAK2-V617F is required for it to induce MPN-like disease in mouse models. The ability of cytokine receptor expression to activate JAK2-V617F has focused on homodimeric receptors. In this study we demonstrate that single components of heterodimeric receptors can also activate JAK2-V617F. Expression of interleukin-27 receptor alpha (IL27Ra), the ligand-binding component of the IL-27 receptor, enhances phosphorylation of JAK2-V617F on tyrosines 1007 and 1008. This activation of JAK2 also leads to tyrosine phosphorylation of signal transducers and activators of transcription-5 (STAT5), a main downstream effector of JAK2 activation. We obtain similar results when we utilize interleukin-12 receptor beta 1 (IL12RB1), a receptor belonging to the same family as IL27Ra. To extend these studies to other heterodimeric cytokine receptor components, we utilized the components of the interleukin-3 receptor, interleukin-3 receptor alpha (IL3Ra) and the beta common chain, which is also utilized in the receptors for interleukin-5 and granulocyte macrophage colony stimulating factor. Expression of each of these receptor subunits activates JAK2-V617F as well as STAT5. Importantly, we demonstrate that expression of IL27Ra can functionally replace the expression of a homodimeric cytokine receptor to support the activation of JAK2-V617F in BaF3 cells as well as promote the transforming activity of JAK2-V617F in these cytokine dependent hematopoietic cells. Interestingly, while IL12RB1 expression activates JAK2-V617F in 293T cells, it is not capable of enhancing the transforming signals of JAK2-V617F in BaF3 cells. Expression of IL3Ra or the beta common chain in BaF3 cells also enhances the ability of JAK2-V617F to transform these cells to cytokine independence. However, this enhancement is not immediate as it only becomes evident at later time points. Together our data demonstrate that in addition to homodimeric receptors, some heterodimeric receptor components may contribute to JAK2-V617F activation. It should also be considered that such receptors may play a role in JAK2-V617F-negative MPNs, perhaps through altered expression or activating receptor mutations, analogous to mutated thrombopoietin receptor proteins that play a role in the development of disease in a fraction of these MPN patients. Disclosures: No relevant conflicts of interest to declare.

Description: Philadelphia chromosome-negative myeloproliferative neoplasms (MPNs), including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis, are hematopoietic stem cell disorders that affect about 300,000 people in the U.S. PV and ET can progress to myelofibrosis where patients are susceptible to bone marrow failure, increased incidence of acute leukemia, and poor survival. MPNs are driven by aberrant JAK2 activity that is a consequence of mutations in JAK2, Mpl, or CalR. The JAK1/2 inhibitor ruxolitinib, which is FDA-approved for a subset of MPN patients, is effective in reducing patient symptomology but fails to induce remission, as it has little effect on the neoplastic cell burden. However, a modest increase in survival and decrease of allele burden has been suggested following long-term ruxolitinib treatment. The quality of life benefit afforded by ruxolitinib suggests it will likely remain a mainstay of future MPN therapies. Identifying the mechanism(s) by which cells persistently survive in the presence of ruxolitinib could prove pivotal to improving JAK2 inhibitor therapies in MPN patients. In order to identify mechanisms by which JAK2-driven MPN cells survive ruxolitinib treatment, we utilized global phospho-tyrosine proteomics to compare ruxolitinib-sensitive and persistent cells, the latter of which are drug persistent by nature due to their reacquisition of JAK2 inhibitor sensitivity upon removal of ruxolitinib. We utilized the JAK2-V617F-driven MPN model cell lines UKE1 and SET2 for this analysis. Pathway analysis from the datasets obtained indicated changes in phosphorylation of proteins involved in growth factor and cytokine signaling. Notably, we identified changes in phosphorylation of several tyrosine phosphatases, including SHP1 and SHP2. The SH2 domain-containing tyrosine phosphatase SHP2 (encoded by PTPN11) plays important roles in relaying signals from cell surface receptors, notably mediating activation of ERKs, and was the first phosphatase identified as leukemogenic. SHP2 regulates JAK2 signaling and is found mutated in ~8% of post-MPN leukemia, suggesting it may play a role in the biology of MPNs. In addition, SHP2 has been identified as a potential mediator of targeted therapy resistance in cancer. Therefore, we hypothesized that SHP2 activity may contribute to the persistent survival of MPN cells in ruxolitinib and provide a therapeutic target in MPNs. To test our hypothesis, we treated ruxolitinib-sensitive (drug naïve) and persistent cells with a specific allosteric inhibitor of SHP2, SHP099. Ruxolitinib-persistent cell growth was more sensitive to SHP2 inhibition, and SHP099 enhanced the growth inhibitory effects of ruxolitinib on ruxolitinib-sensitive cells. Treatment of UKE1 cells with a high concentration (~10X GI50) of ruxolitinib leads to the persistent survival of a small population of cells and this was prevented by SHP099. SHP2 inhibition was also deleterious to the maintenance of an existing drug persistent state in a high concentration of ruxolitinib. Interestingly, signaling pathways in cells persistently growing in ruxolitinib differentially respond to SHP2 inhibition. In both ruxolitinib-sensitive and persistent cells activation of ERK was inhibited by SHP099. However, in ruxolitinib-persistent UKE1 cells, this inhibition was more rapid, suggesting the kinetic regulation of ERK activation by SHP2 is altered in these cells. Also, SHP099 led to an increase in pSTAT3 in both ruxolitinib-sensitive and persistent cells. However, SHP099 led to an increase in pSTAT5 in ruxolitinib sensitive UKE1 cells, but not in cells persistently growing in ruxolitinib. Re-activation of signaling pathways following ruxolitinib treatment likely contributes to cell survival. We observed that ERK becomes re-activated shortly after ruxolitinib treatment, and this re-activation depends on SHP2 activity. Finally, we observed dose-dependent inhibition of the neoplastic growth of primary MPN patient cells by SHP099 and synergistic inhibition with ruxolitinib, suggesting co-targeting SHP2 may improve JAK2-targeted MPN therapies. In conclusion, our data suggest that re-wiring of signaling in cells challenged with ruxolitinib may create greater dependence on SHP2 activity that contributes to cell survival to JAK2 inhibition. Thus, targeting SHP2 may improve the efficacy of ruxolitinib as an anti-MPN therapy. Disclosures Haura: Incyte Corporation: Research Funding. Reuther:Incyte Corporation: Research Funding.

Description: Philadelphia chromosome negative myeloproliferative neoplasms (MPNs) are JAK2-driven disorders resulting from mutations in JAK2, MPL, or CALR. Ruxolitinib, the only FDA-approved JAK2 inhibitor for MPNs, alleviates patient symptomology and improves quality of life, but has little effect on reducing mutant allele burden. This persistent survival of MPN cells in the face of ruxolitinib, as well as other JAK2 inhibitors that have been clinically tested, is a major clinical bottleneck to the development of an effective targeted therapy for MPN patients. Identifying new therapeutic targets which play critical roles in MPN cells and/or in JAK2 inhibitor persistence may lead to improved MPN therapies. SHP2 is an oncogenic tyrosine phosphatase that is an effector of growth factor and cytokine receptor signaling. SHP2 plays a critical role in the activation of the RAS-ERK pathway and regulates JAK-STAT signaling via numerous phosphatase-dependent mechanisms. Activating mutations of SHP2(PTPN11) have been identified in leukemia, including 8% of MPN patients whose disease progressed to acute myeloid leukemia (AML). In addition, SHP2 has been shown to mediate adaptive resistance to targeted therapies in several cancers. Given the role of SHP2 in cytokine and JAK-STAT signaling, we envisaged a potential role of SHP2 in MPN cell growth and/or survival and ruxolitinib persistence. Treatment of JAK2-V617F-driven MPN model cell lines (UKE1, SET2, and BaF3-JAK2-V617F) with ruxolitinib blocked constitutive tyrosine phosphorylation of SHP2, including phosphorylation of Y542, a marker for activated SHP2. This phosphorylation, however, was restored in ruxolitinib persistent cells. Combination treatment of the allosteric SHP2 inhibitor RMC-4550 (Revolution Medicines) with ruxolitinib prevented the development of ruxolitinib persistent cells and pre-established persistent cells remained sensitive to SHP2 inhibition. RMC-4550 treatment led to significantly reduced levels of pERK consistent with the role of SHP2 in RAS signaling. Interestingly, pERK levels in persistent cells were more sensitive to SHP2 inhibition compared to drug naïve cells suggesting pERK was more dependent on SHP2 in ruxolitinib persistent cells. SHP2 inhibitor treatment increased pSTAT5(Y694) in drug naïve cells but this increase was not observed in similarly treated persistent cells. Furthermore, while ruxolitinib inhibited pERK levels in UKE1 and SET2 cells, pERK levels recovered within 24 hrs of treatment. SHP2 inhibition prevented the recovery of pERK in the presence of ruxolitinib. Collectively, these data suggest that signaling pathways in MPN cells treated with ruxolitinib can become rewired, gaining greater dependence on SHP2, concomitant with sustained pERK and cell survival/growth. Interestingly, we identified a known activating SHP2 mutation (F71L) in UKE1 cells obtained from two independent sources - consistent with the presence of PTPN11 mutations in post-MPN AML. The persistent survival of UKE1 cells in ruxolitinib was antagonized by CRISPR-mediated reduction of SHP2 expression, providing further evidence that SHP2 contributes to ruxolitinib persistence. To assess the effects of a SHP2 inhibitor on MPN progression in vivo, we employed the MPLW515Lbone marrow transplant mouse model of MPN. Initial assessment of therapeutic treatment of mice with an established MPN phenotype indicated that once daily treatment of RMC-4550 (10 or 30 mg/kg) antagonized the MPN phenotype. Complete blood counts indicated a significant reduction in white blood cells, monocytes, and neutrophils compared to vehicle treated mice, while flow cytometry analysis indicated RMC-4550 diminished CD11b+ cell numbers to near that observed in mice transplanted with MPLWT-transduced bone marrow. RMC-4550 improved the overall health of diseased mice, as indicated by increased weight, and significantly reduced organomegaly of the spleen and liver compared to vehicle treated MPN mice. Finally, erythropoietin independent erythroid colony formation of JAK2V617F-positive MPN patient cells was suppressed following SHP2 inhibition, which synergized or enhanced the inhibition induced by ruxolitinib in this assay. In summary, our results suggest that SHP2 inhibition may represent a potential MPN therapy in both ruxolitinib naïve and resistant patients and is an attractive therapeutic target for future clinical investigation. Disclosures Epling-Burnette: Incyte Corporation: Research Funding; Forma Therapeutics: Research Funding; Celgene Corporation: Patents & Royalties, Research Funding. Reuther:Incyte Corporation: Research Funding.

Description: Protein acetylation is an important contributor to cancer initiation. Histone deacetylase 6 (HDAC6) controls JAK2 translation and protein stability and has been implicated in JAK2-driven diseases best exemplified by myeloproliferative neoplasms (MPNs). By using novel classes of highly selective HDAC inhibitors and genetically deficient mouse models, we discovered that HDAC11 rather than HDAC6 is necessary for the proliferation and survival of oncogenic JAK2-driven MPN cells and patient samples. Notably, HDAC11 is variably expressed in primitive stem cells and is expressed largely upon lineage commitment. Although Hdac11is dispensable for normal homeostatic hematopoietic stem and progenitor cell differentiation based on chimeric bone marrow reconstitution, Hdac11 deficiency significantly reduced the abnormal megakaryocyte population, improved splenic architecture, reduced fibrosis, and increased survival in the MPLW515L-MPN mouse model during primary and secondary transplantation. Therefore, inhibitors of HDAC11 are an attractive therapy for treating patients with MPN. Although JAK2 inhibitor therapy provides substantial clinical benefit in MPN patients, the identification of alternative therapeutic targets is needed to reverse MPN pathogenesis and control malignant hematopoiesis. This study establishes HDAC11 as a unique type of target molecule that has therapeutic potential in MPN.

Description: The reciprocal translocation between chromosomes 9 and 22 that fuses coding sequences of the Bcr and Abl genes is responsible for a remarkably diverse group of hematologic malignancies. A newly described 230-kd form of Bcr-Abl has been associated with an indolent myeloproliferative syndrome referred to as chronic neutrophilic leukemia. We have cloned the corresponding gene and examined the biologic and biochemical properties of p230 Bcr-Abl after retroviral-mediated gene transfer into hematopoietic cell lines and primary bone marrow cells. p230 Bcr-Abl–expressing 32D myeloid cells were fully growth factor-independent and activated similar signal transduction pathways as the well-characterized p210 and p185 forms of Bcr-Abl. In contrast, primary mouse bone marrow cells expressing p230 required exogenous hematopoietic growth factors for optimal growth, whereas p185- and p210-expressing cells were independent of growth factors. The 3 Bcr-Abl proteins exerted different effects on differentiation of bone marrow cells. p185 induced outgrowth of lymphoid precursors capable of tumor formation in immunodeficient mice. In contrast, p210- and p230-expressing bone marrow cells caused limited outgrowth of lymphoid precursors that failed to form tumors in immunodeficient mice. Removal of cytokines and autologous stroma from Bcr-Abl–expressing bone marrow cultures produced the expansion of distinct lineages by the various Bcr-Abl proteins. p185 drove expansion of cytokine-independent lymphoid progenitors, while p210 and p230 generated cytokine-independent monocyte/myeloid cells. These findings suggest that the different Bcr-Abl fusion proteins drive the expansion of different hematopoietic populations, which may explain the association of the various Bcr-Abl oncoproteins with different spectra of human leukemias.