Search

Supplemental figure 2: A) Flow cytometry analysis showing that the TF1 erythroleukemia cells express high levels of CD34 marker on their cell surface. B) Twenty four hours post DOCK4 siRNA nucleofection, cells were used for immunoblot analysis for DOCK4 expression. Data are mean {plus minus}SEM from five independent experiments. C) Twenty four hours post DOCK4 knockdown, TF1 cells were briefly deprived of cytokines and then re-exposed to a cytokine cocktail (Stem cell factor (SCF), Interleukin-3 (IL-3) and Granulocyte-Macrophage colony stimulating factor (GM-CSF)) for 15 mins. Samples were resolved by immunoblotting for phospho-LYN (Y397), phospho-SHIP1 (Y1021) and phospho-SHP1 (Y536). Same membranes were probed with total LYN, SHIP1 and SHP1 antibodies as loading controls. D, E and F) Quantitation of the levels of phospho-LYN (Y397) (N=3), phospho-SHIP1 (Y1021) (N=5) and phospho-SHP1 (Y536) (N=4) respectively in TF1 cells from control and DOCK4 knocked down samples (*P < 0.05; Student's t test). Data are mean {plus minus}SEM from independent experiments as specified. G) Twenty four hours post DOCK4 knockdown HSCs were briefly deprived and exposed to 5 cytokine cocktails (TPO, SCF, FLT3 ligand, IL3 and IL6) for 15 mins. Following cytokine exposure, LYN kinase activity assay was performed to examine the endogeneous LYN kinase activity in HSCs expressing normal and reduced levels of DOCK4 (*P < 0.05; Student's t test). Data are represented as mean {plus minus}SEM from three technical replicates. H) Quantitation of the levels of phospho-SHIP1 (Y1021) in the in vitro kinase reaction products (***P < 0.0005; One way ANOVA). Data are mean {plus minus}SEM from three independent experiments. I) Quantitation of the levels of phospho-SHP1 (Y536) in the in vitro kinase reaction products (***P < 0.0005; One way ANOVA). Data are mean {plus minus}SEM from three independent experiments. J, K and L) Quantitation of the levels of phospho-LYN (Y397), phospho-SHIP1 (Y1021), and phospho-SHP1 (Y536) respectively in primary human HSCs treated with increasing doses of LYN/Src inhibitor, RK20449. Data are mean {plus minus}SEM from three biological replicates (****P < 0.00005, One way ANOVA; ns-not significant). M) LYN kinase activity assay was performed using recombinant LYN kinase and increasing amounts of recombinant DOCK4 C-terminus. JAK2 kinase and LYN/Src inhibitor were used as controls. Data are mean {plus minus}SEM from three independent experiments. (**P < 0.005; One way ANOVA). N) An in vitro binding assay was performed using flag-tagged DOCK4 C-terminus (D4CT) and recombinant LYN kinase to determine whether LYN kinase interacted with DOCK4-C-terminus. Samples were immunoprecipitated with anti-Flag antibody followed by immunoblot analysis using anti-Flag and anti-LYN antibodies.

Supplementary methods

Purpose:Myelodysplastic syndromes (MDS) with deletion of chromosome 7q/7 [-7/(del)7q MDS] is associated with worse outcomes and needs novel insights into pathogenesis. Reduced expression of signaling protein dedicator of cytokinesis 4 (DOCK4) in patients with -7/(del)7q MDS leads to a block in hematopoietic stem cell (HSC) differentiation. Identification of targetable signaling networks downstream of DOCK4 will provide means to restore hematopoietic differentiation in MDS.Experimental Design: We utilized phosphoproteomics approaches to identify signaling proteins perturbed as a result of reduced expression of DOCK4 in human HSCs and tested their functional significance in primary model systems.Results:We demonstrate that reduced levels of DOCK4 lead to increased global tyrosine phosphorylation of proteins in primary human HSCs. LYN kinase and phosphatases INPP5D (SHIP1) and PTPN6 (SHP1) displayed greatest levels of tyrosine phosphorylation when DOCK4 expression levels were reduced using DOCK4-specific siRNA. Our data also found that increased phosphorylation of SHIP1 and SHP1 phosphatases were due to LYN kinase targeting these phosphatases as substrates. Increased migration and impediment of HSC differentiation were consequences of these signaling alterations. Pharmacologic inhibition of SHP1 reversed these functional aberrations in HSCs expressing low DOCK4 levels. In addition, differentiation block seen in DOCK4 haplo-insufficient [-7/(del)7q] MDS was rescued by inhibition of SHP1 phosphatase.Conclusions:LYN kinase and phosphatases SHP1 and SHIP1 are perturbed when DOCK4 expression levels are low. Inhibition of SHP1 promotes erythroid differentiation in healthy HSCs and in -7/(del)7q MDS samples with low DOCK4 expression. Inhibitors of LYN, SHP1 and SHIP1 also abrogated increased migratory properties in HSCs expressing reduced levels of DOCK4.

Even though the Ten-eleven translocation (TET) enzymes catalyze the generation of 5-hydroxymethylcytosines required for lineage commitment and subsequent differentiation of stem cells into erythroid cells, the mechanisms that link extracellular signals to TET activation and DNA hydroxymethylation are unknown. We demonstrate that hematopoietic cytokines phosphorylate TET2, leading to its activation in erythroid progenitors. Specifically, cytokine receptor–associated JAK2 phosphorylates TET2 at tyrosines 1939 and 1964. Phosphorylated TET2 interacts with the erythroid transcription factor KLF1, and this interaction with TET2 is increased upon exposure to erythropoietin. The activating JAK2V617F mutation seen in myeloproliferative disease patient samples and in mouse models is associated with increased TET activity and cytosine hydroxymethylation as well as genome-wide loss of cytosine methylation. These epigenetic and functional changes are also associated with increased expression of several oncogenic transcripts. Thus, we demonstrate that JAK2-mediated TET2 phosphorylation provides a mechanistic link between extracellular signals and epigenetic changes during hematopoiesis. Significance: Identification of TET2 phosphorylation and activation by cytokine-stimulated JAK2 links extracellular signals to chromatin remodeling during hematopoietic differentiation. This provides potential avenues to regulate TET2 function in the context of myeloproliferative disorders and myelodysplastic syndromes associated with the JAK2V617F-activating mutation. This article is highlighted in the In This Issue feature, p. 681

Introduction: TP53 mutant (mTP53) MDS and AML represent a distinct molecular cohort with poor outcomes. Hypomethylating agents (HMA) have emerged as preferred treatment for these patients with CR rate of 20-30% and median OS of 6-12 months. APR-246 is a mutant p53 activator with single agent activity in mTP53 AML. We report initial phase 1b results of APR-246 + AZA in mTP53 MDS/AML. Methods: Eligible pts included HMA naïve mTP53 MDS and oligoblastic AML (≤ 30% blasts) ≥ 18 years of age. Pts received APR-246 in a 3+3 dose escalation design (50, 75, 100 mg/kg lean body weight) IV daily over 4 days in a lead-in phase (days -14 to -10) followed by the same dose of APR-246 (days 1-4) + AZA 75 mg/m2 SC/IV over 7 days (days 4-10 or 4-5 and 8-12) in 28 day cycles. Primary objective was safety with AEs graded by CTCAE v4.03 and DLT assessment over 6 weeks. Secondary endpoints included response by IWG 2006 criteria as well as serial next generation sequencing (NGS) and p53 IHC for evaluation of clonal suppression and remission depth. Results: As of Jan 1, 2017, 9 pts (33% male; median age 65 years (39-73)) have enrolled with 3 pts per cohort. Three pts had AML-MRC and 6 had MDS; all pts had poor risk cytogenetics (11% poor, 89% very poor) and higher risk disease by IPSS-R (22% high, 78% very high). Median BM blasts were 18% (4-30). Seven pts (78%) remain on study: 2 pts in the 50mg/kg cohort discontinued treatment (Tx), 1 pt due to infection during C2 who later died of sepsis unrelated to Tx, and 1 pt electively discontinued in durable marrow CR (mCR) after 5 cycles of therapy. Median time on study is 106 days (14-221). Tx related AEs during the lead-in phase (all G1) included ataxia (n=1), dizziness (n=1), and facial numbness (n=1). AEs occurring in > 1 pt included dizziness (n=3), nausea (n=3), neutropenia (n=3), thrombocytopenia (n=3), infection (n=3), headache (n=2), pain (n=2), weakness (n=2), falls (n=2), facial numbness (n=2), and ataxia (n=2); all G1/G2 except neutropenia/thrombocytopenia (G4). No Tx-related SAEs or DLTs have occurred to date. Five of six pts were response evaluable with 1 pt discontinuing tx prior to 1st disease assessment. ORR by IWG was 100% with 4 CR (80%, 3/3 in DL2) and 1 mCR. All CR patients achieved complete cytogenetic response. One CR patient achieved a mCR and partial cytogenetic response after APR-246 lead-in prior to combination therapy. All CR pts had high p53 positivity by IHC at baseline (55-70%) which normalized on serial assessment ( Conclusions: APR-246 + AZA combination is well tolerated in mTP53 MDS/AML. Responses have been achieved in all pts (80% CR) accompanied by deep molecular remissions. The maximum tolerated dose has not been reached and dose escalation is ongoing. Citation Format: David A. Sallman, Amy DeZern, Kendra Sweet, David P. Steensma, Thomas Cluzeau, Mikkael Sekeres, Guillermo Garcia-Manero, Gail Roboz, Amy McLemore, Kathy McGraw, John Puskas, Ling Zhang, Chirag Bhagat, Armin Graber, Najla H. Al Ali, Eric Padron, Roger Tell, Jeffrey E. Lancet, Pierre Fenaux, Alan F. List, Rami S. Komrokji. Phase Ib/II combination study of APR-246 and azacitidine (AZA) in patients with TP53 mutant myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr CT068.

Background: The myelodysplastic syndromes (MDS) are senescence-dependent stem cell malignancies. Although germ line mutations in RUNX1, CEBPA, ETV6, or GATA2 may underlie familial cases, inherited genetic polymorphisms influencing MDS susceptibility has not been evaluated. We performed the first genome-wide association study (GWAS) to identify single nucleotide polymorphisms (SNPs) linked to MDS predisposition. Methods: DNA from 1361 MDS patients from 9 international centers was genotyped on the Affymetrix Genome-Wide Human SNP Array 6.0; all patients signed informed consent. Data on 4597 healthy controls genotyped on the same platform were obtained from the Database of Genotypes and Phenotypes (dbGaP). Applying standard quality control metrics and limiting to individuals of European ancestry, we analysed 999 MDS patients and 4,309 controls at 545,924 markers. We used logistic regression to determine associations with MDS after appropriate adjustment assuming an additive genetic model. Results: We identified 15 SNPs at 9 loci associated with MDS, including rs6780298 (3q13; p = 5.0×10-6) and rs1206819 (20q13; p = 1.54×10-6) that mapped to the intragenic regions of MORC1 and EYA2, respectively; and rs2473784 (1p31; p = 3.32×10-7) that mapped downstream of DEPDC1. Other hits (p≤5×10-6) at 4q28, 5p14, 9p22, 6p22, 10q21 are in or near SLC7A11, GUSBP1, SH3GL2, MOG and TRIM27, respectively. Using publicly available gene expression profiling data, we confirmed all of these genes are expressed in the hematopoietic compartment and MORC1, EYA2, DEPDC1, and SH3GL2 are increased in leukemic samples suggesting that variation in these genes may be functionally relevant in myeloid malignancies. Notably, EYA2 (20q13) flanks the commonly deleted region in del(20q) MDS and has oncogenic properties in solid tumors. SH3GL2 is very highly upregulated in leukemic samples and interacts with Dynamin-1, a GTP-binding protein involved in cellular trafficking and is similarly associated with solid tumorigenesis. To validate our observed associations, we analysed the 15 SNP markers in an independent set of 12,385 individuals of whom 117 developed hematologic malignancy (including MDS) during follow up. The markers at 1p31 including rs2473784 and 4 others in strong linkage disequilibrium were associated with individuals who developed hematologic malignancy (p Conclusions: These data provide the first genome-wide identification of germline variants associated with MDS. Current work is underway to replicate these findings in an independent sample set of MDS patients and healthy controls. If confirmed, these SNPs may serve as biomarkers to identify individuals at risk for MDS and potentially support new prevention strategies. Citation Format: Kathy McGraw, Chia-Ho Cheng, Ann Chen, Giulio Genovese, Thomas Cluzeau, Hui-Yi Lin, Bartlomiej Przychodzen, Mar Mallo, Leonor Arenillas, Azim Mohamedali, Lionel Ades, Ashley Basiorka, Brittany Irvine, David Sallman, Eric Padron, Lubomir Sokol, Andrea Pellagatti, Chimene Brest, Sophie Raynaud, Bjorn Nilsson, Jacqueline Boultwood, Benjamin Ebert, Francesc Sole, Pierre Fenaux, Ghulam Mufti, Jaroslaw Maciejewski, Peter Kanetsky, Alan List. Identification of genetic polymorphisms associated with myelodysplastic syndromes by genome-wide association study. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2570.