Wild-type RUNX1 Research Articles

Number of RUNX1 mutations, wild-type allele loss and additional mutations impact on prognosis in adult RUNX1-mutated AML.

RUNX1-mutated acute myeloid leukemia (AML) show a distinct pattern of genetic abnormalities and an adverse prognosis. We analyzed the impact of multiple RUNX1 mutations and RUNX1 wild-type (WT) loss in 467 AML with RUNX1 mutations (mut): (1) RUNX1 WT loss (n=53), (2) >1 RUNX1mut (n=94) and (3) 1 RUNX1mut (n=323). In 1 RUNX1mut, +8 was most frequent, whereas in WT loss +13 was the most abundant trisomy (+8: 66% vs 31%, P=0.022; +13: 15% vs 62%, P<0.001). Analyses of 28 genes in 163 selected cases revealed SRSF2 (39%), ASXL1 (36%), DNMT3A (19%), IDH2 (17%) and SF3B1 (17%) as most frequently mutated genes. RUNX1 WT loss showed a higher frequency of ASXL1mut compared with the other cases (50% vs 29%, P=0.009). Median overall survival (OS) in the total cohort was 14 months. WT loss (OS: 5 months) and >1 RUNX1mut (14 months) showed an adverse impact on prognosis compared with 1 RUNX1mut (22 months; P=0.002 and 0.048, respectively). Mutations in ASXL1 and ⩾2 additional mutations correlated with shorter OS (10 vs 18 months, P=0.028; 12 vs 20 months, P=0.017). Thus, the number of RUNX1mut, RUNX1 WT loss and the number and type of additional mutations is biologically and clinically relevant.

Abstract LB-081: Novel and effective RUNX1-targeted therapy for AML expressing RUNX1 mutation

Abstract RUNX1 is a transcription factor involved in normal and malignant hematopoiesis. Somatic mutations in RUNX1 (mtRUNX1) have been documented in de novo and secondary AML (10%), MDS (~10%) and CMML (up to 37%). Germ-line, mono-allelic mutations and deletions in RUNX1 cause the highly-penetrant (~40%) autosomal dominant, Familial Platelet Disorder with a propensity to Myeloid Malignancy (FPD-MM). The majority of mtRUNX1 are missense mutations, large deletions or truncation-mutations in the DNA-binding ‘RUNT’ homology domain, or in the C-terminal transactivation domain. These mutations behave mostly as loss of function mutations, associated with relative resistance to standard chemotherapy and an unfavorable prognosis in AML. This highlights an unmet need to develop and test novel targeted therapies for AML due to germ-line or somatic mtRUNX1. In the present studies, we demonstrate that shRNA-mediated knockdown of mutant and wild-type RUNX1 repressed its target genes MYC, PU.1 and MPO (myeloperoxidase), as well as inhibited growth and induced apoptosis of AML cells expressing mtRUNX1 (OCI-AML5 and Mono-Mac-1). Ex vivo depletion of RUNX1 abrogated the leukemia initiating potential of OCI-AML5 cells. After engraftment, tetracycline-inducible shRNA-mediated in vivo knockdown of RUNX1, as compared to the non-induced controls, enhanced survival of the immune-depleted (NSG) mice engrafted with OCI-AML5 cells. RUNX1 transcription is driven by the BET protein BRD4-occupied super enhancer in the first intron of the RUNX1 gene. A heat map of Hi-C interaction scores within the RUNX1 TAD (topology-associated domain) showed triangle-shaped regions of high interaction, with CTCF binding sites defining the TAD boundaries, anchoring a loop that separates the RUNX1 TAD from other TADs. Consistent with this, our findings show that, shRNA-mediated depletion of BRD4 or treatment with BET protein (BETP) inhibitor (BETi) OTX015 reduced BRD4 occupancy at the enhancer and promoter of RUNX1. This was associated with depletion of RUNX1 and its target-gene expressions and apoptosis of cultured and patient-derived (PD), primary AML blast progenitor cells (BPCs). Additionally, treatment of NSG mice engrafted with luciferase-transduced OCI-AML5 cells with OTX015 (50 mg/kg/day X 5 days, for 3 weeks) reduced the AML burden and significantly improved their survival (p &amp;lt; 0.01). Co-treatment with the BETi and BCL2-antagonist venetoclax or CDK4/6 antagonist palbociclib or decitabine synergistically induced apoptosis of OCI-AML5 and PD AML BPCs (combination index values of &amp;lt; 1.0). Results describing the in vivo efficacy of BETi and BETP-PROTAC (proteolysis targeting chimera), alone and in combinations against AML expressing mtRUNX1 will be updated at the AACR meetings. Collectively, these findings highlight novel and effective, mechanistically-targeted, single agent or combination, BETP antagonist-based therapy of AML expressing mtRUNX1. Citation Format: Christopher P. Mill, Courtney DiNardo, Warren C. Fiskus, Dyana T. Saenz, Baohua Sun, Agnieszka J. Nowak, Misun Kim, Mark Routbort, Koichi Takahashi, Kapil N. Bhalla. Novel and effective RUNX1-targeted therapy for AML expressing RUNX1 mutation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-081. doi:10.1158/1538-7445.AM2017-LB-081

Clinical observations of AML expressing mutant RUNX1 and pre-clinical studies of RUNX1-targeted novel therapy of AML.

7030 Background: Runt-related transcription factor 1 ( RUNX1) is critically involved in normal and malignant hematopoiesis. Somatic mutations in RUNX1 occur in ~10% of AML, especially in older patients with history of radiation or antecedent hematologic disorder. Presence of mRUNX1 is reported to confer relative resistance to therapy and poorer prognosis in AML, and there are no mRUNX1-targeted or specific therapies available. Methods: We retrospectively analyzed outcomes of 94 mRUNX1 and 444 wild-type RUNX1 AML patients treated at our institution from 9/2013 to 12/2016. We also determined the pre-clinical efficacy of a targeted therapy against cultured and primary AML cells expressing mRUNX1. Results: 67% of mRUNX1 patients were &gt; 65 years of age. Co-occurring mutations with mRUNX1 were ASXL1 (33%), N/KRAS (20%), FLT3 (20%), IDH2 (18%), IDH1 (13%) and TET2 (10%). In patients &gt; 65 years treated with hypomethylating agent-based therapy, the presence or absence of mRUNX1 did not impact response rate (42% vs 46% CR/CRp, p = 0.67), median event-free survival (3.4 vs 4.7 mo, p = 0.82) or overall survival (11.5 vs 9.3 mo, p = 0.97). In mRUNX1 expressing AML OCI-AML5 and MonoMac1 cells, knockdown of RUNX1 by shRNA repressed its targets, e.g., MYC and PU.1, inhibited growth and induced apoptosis. Ex vivo knockdown of RUNX1 abrogated in vivo leukemia initiation by OCI-AML5 cells. After engraftment, inducible shRNA-mediated in vivo knockdown of RUNX1 restored survival of immune-depleted NSG mice engrafted with OCI-AML5 cells. RUNX1 transcription is driven by a super enhancer occupied by the bromodomain extraterminal protein (BETP), BRD4. Accordingly, BRD4 knockdown by shRNA or treatment with the BRD4-inhibitor OTX015 depleted RUNX1 and its targets, and induced apoptosis of AML cells. Treatment of OCI-AML5 cell-engrafted NSG mice with OTX015 (50 mg/Kg/day X 5, for 3 wks) reduced AML burden and improved survival (p &lt; 0.01). Co-treatment with the BETi and BCL2 inhibitor venetoclax or CDK4/6 antagonist palbociclib or decitabine synergistically induced apoptosis of OCI-AML5 and primary AML blasts. Conclusions: These findings highlight a novel, promising, BETP antagonist-based therapy of AML expressing mtRUNX1.

Functional analysis of novel RUNX2 mutations in cleidocranial dysplasia.

Cleidocranial dysplasia (CCD) is a rare autosomal dominant skeletal disorder caused by mutation of runt-related transcription factor 2 (RUNX2) gene. The purpose of this study was to explore novel RUNX2 mutations in seven individuals with CCD and investigate the function of the mutant RUNX2 proteins. DNA samples were prepared from the peripheral blood of the CCD individuals, and then subjected to DNA sequencing. Conservation and secondary structure analysis were performed based on RUNX2 sequencing results. pEGFP-C1 plasmids containing GFP-tagged wild-type RUNX2 and three novel RUNX2 mutations expression cassettes were constructed, and then transfected into HEK293T cells. Cell fluorescence, luciferase assay and western blotting were used to analyse the subcellular distribution and function of the mutant RUNX2 proteins. Three novel mutations (R193G, 258fs, Y400X) were found in the seven CCD patients. Conservation and structure analysis show one novel mutation (R193G) in Runt domain and two novel mutations (258fs and Y400X) in PST domain of RUNX2. Western blotting confirmed that the 258fs and Y400X mutations produced truncated proteins. Fluorescence detection showed that the three novel mutants localised exclusively in the nucleus. However, luciferase assay indicated all mutants severely impaired the transactivation activities of RUNX2 on osteocalcin promoter. Our results broaden the spectrum of RUNX2 mutations in CCD individuals and demonstrated that loss of function in RUNX2 is responsible for CCD.

Puerarin promotes the viability and differentiation of MC3T3‑E1 cells by miR‑204‑regulated Runx2 upregulation.

Puerarin has attracted increasing attention because of its beneficial effects on anti‑osteoporosis, but the molecular mechanisms underlying its actions on osteoblasts are not fully understood. The current study aimed to investigate the effect of puerarin on the cell viability and differentiation of mouse MC3T3‑E1 osteoblast‑like cells invitro and its underlying mechanisms. The results indicated that 0.01, 0.1 and 1mg/ml puerarin significantly promoted the viability of osteoblasts, enhanced alkaline phosphatase (ALP) activity and increased the expression of transforming growth factor‑β1, Smad2, Smad3 and Runt‑related transcription factor (Runx)2. Micro (mi)RNA target prediction programs predicted that miR‑204 may directly target Runx2. Following treatment with 0.1mg/ml puerarin for 48h, the expression level of miR‑204 was downregulated. Besides, miR‑204 dramatically repressed the luciferase activity of wildtype Runx2 3'‑UTR transfected cells, but not that of the mutant ones. Overexpression of miR‑204 in osteoblasts significantly decreased the protein expression of Runx2, while inhibition of miR‑204 enhanced Runx2's expression. In addition, overexpression of miR‑204 inhibited the cell viability and ALP activity of osteoblasts, while inhibition of miR‑204 had the opposite effect. The results suggested that puerarin may promote MC3T3‑E1 osteoblast‑like cell viability and differentiation, which may be related to the downregulation of miR‑204 and the following activation of Runx2.

MAP Kinase-Dependent RUNX2 Phosphorylation Is Necessary for Epigenetic Modification of Chromatin During Osteoblast Differentiation.

RUNX2, an essential transcription factor for osteoblast differentiation and bone formation is activated by ERK/MAP kinase-dependent phosphorylation. However, relationship between these early events and specific epigenetic modifications of chromatin during osteoblast differentiation have not been previously examined. Here, we explore these relationships using chromatin immunoprecipitation (ChIP) to detect chromatin modifications in RUNX2-binding regions of Bglap2 and Ibsp. Growth of MC3T3-E1c4 preosteoblast cells in differentiation conditions rapidly induced Bglap2 and lbsp mRNAs. For both genes, osteogenic stimulation increased chromatin-bound P-ERK, P-RUNX2, p300, and RNA polymerase II as well as histone H3K9 and H4K5 acetylation. The level of H3K4 di-methylation, another gene activation-associated histone mark, also increased. In contrast, levels of the gene repressive marks, H3K9 mono-, di-, and tri-methylation in the same regions were reduced. Inhibition of MAP kinase signaling blocked differentiation-dependent chromatin modifications and Bglap2 and Ibsp expression. To evaluate the role of RUNX2 phosphorylation in these responses, RUNX2-deficient C3H10T1/2 cells were transduced with adenovirus encoding wild type or phosphorylation site mutant RUNX2 (RUNX2 S301A/S319A). Wild type RUNX2, but not the non-phosphorylated mutant, increased H3K9 and H4K5 acetylation as well as chromatin-associated P-ERK, p300, and polymerase II. Thus, RUNX2 phosphorylation is necessary for subsequent epigenetic changes required for osteoblast gene expression. Taken together, this study reveals a molecular mechanism through which osteogenic genes are controlled by a MAPK and P-RUNX2-dependent process involving epigenetic modifications of specific promoter regions. J. Cell. Physiol. 232: 2427-2435, 2017. © 2016 Wiley Periodicals, Inc.

The Number of RUNX1 Mutations and the Presence of One Intact RUNX1 Allele Influence Cytogenetic Abnormalities, Additional Molecular Mutations and Prognosis in RUNX1 Mutated AML

Background: According to the revised WHO classification of 2016, AML with mutated RUNX1 constitutes a new provisional entity. We previously reported that the subgroup of RUNX1-mutated AML with RUNX1 wild-type (WT) loss is associated with a distinct pattern of cytogenetic and molecular genetic abnormalities and with an adverse prognosis. However, the impact of multiple RUNX1 mutations is unclear yet.Aims: Evaluation of associated genetic alterations and prognosis in AML with >1 RUNX1 mutation as compared to those with (1) RUNX1 WT loss and (2) one RUNX1 mutation.Patient cohorts and methods: The total cohort comprised 467 AML cases with RUNX1 mutations (mut) (296 male, 171 female). Median age was 72 years (range: 18-91 years). All patients were investigated using chromosome banding analysis (CBA) and amplicon sequencing of RUNX1. The cohort was split into the subgroups with RUNX1 WT loss (UPD or RUNX1 deletion detected by genomic arrays) (n=50), cases with >1 RUNX1mut (n = 94) and cases with 1 RUNX1mut and conservation of the RUNX1 WT allele (n = 323). Of these, 50, 55 and 58 cases, respectively, were selected for further mutation analyses of ASXL1, BCOR, CBL, CEBPA, DNMT3A, ETV6, EZH2, FLT3-ITD, FLT3-TKD, GATA2, IDH1, IDH2, KIT, KRAS, MLL-PTD, NPM, NRAS, SETBP1, SF3B1, SRSF2, TET2, TP53, U2AF1 and WT1. Variants of unknown significance were excluded from statistical analysis.Results: In the total cohort of 467 cases, CBA revealed a normal karyotype (NK) in 53% of patients, 18% harbored trisomies, 3% showed a complex karyotype (>3 abnormalities), 26% other aberrations. The proportion of cases with trisomies was largest in cases with RUNX1 WT loss (26%), followed by >1 RUNX1mut (19%) and 1 RUNX1mut (16%). In more detail, in the total cohort, 56% of cases with trisomies harbored +8, in 30% +13 was found. A similar pattern was observed for 1 RUNX1mut, whereas in cases with WT loss +13 was the most abundant trisomy (+8: 66% in 1 RUNX1mut vs. 31% in WT loss, p=0.022; +13: 15% vs. 62%, p<0.001). Cases with >1 RUNX1mut showed an intermediate distribution (+8: 44%, +13: 50%).Missense mutations were the most abundant mutation type (53%) inWT loss cases, followed by frameshift mutations (28%). By contrast, in cases with 1 RUNX1mut, frameshift mutations were found more frequent (45%, p=0.016), whereas missense mutations were detected at a frequency of 31% (p=0.006). In cases with >1 RUNX1mut, both were observed at similar frequencies (missense: 36%, frameshift: 38%).Mutation analyses of 163 selected cases revealed SRSF2 (39%), ASXL1 (36%), DNMT3A (19%), BCOR (18%), IDH2 (17%), SF3B1 (17%) and TET2 (17%) mutations as most frequently mutated genes in the total cohort. Cases with RUNX1 WT loss showed a higher frequency of ASXL1mut compared to the other cases (50% vs. 29%, p=0.013), while U2AF1mut were absent from this group (0% vs. 15%, p=0.019). Differences between cases with and without RUNX1 WT loss were also detected for DNMT3A, TET2, SF3B1 (more abundant in WT loss) and IDH2, WT1 (less abundant in WT loss), although these were not statistically significant. For many genes, the group of cases with >1 RUNX1mut showed an intermediate abundance pattern, or mutation frequencies similar to cases with 1 RUNX1mut were observed. Mutations in spliceosome genes (SF3B1, SRSF2, U2AF1, ZRSR2) were very abundant in all subgroups but less frequent (although not statistically significant) in cases with 1 RUNX1mut.Median overall survival (OS) in the total cohort was 14 months. WT loss (OS: 5 months) and >1 RUNX1mut (14 months) showed an adverse impact on prognosis compared to 1 RUNX1mut (22 months; p=0.002 and p=0.048, respectively). Mutations in ASXL1 and KRAS showed a negative impact on OS in the total cohort (10 vs. 18 months, p=0.028; 1 vs. 15 months, p<0.001), whereas DNMT3A mutations negatively affected OS in WT loss only (in WT loss: 1 vs. 7 months, p=0.005).Conclusion: 1) RUNX1 mutated AML cases with WT loss show a high frequency of +13, RUNX1 missense mutations and accompanying ASXL1 mutations. 2) They clearly separate from cases with 1 RUNX1mut, which depict a predominance of +8, RUNX1 frameshift mutations and mutations in IDH2 and WT1. 3) Patients with >1 RUNX1mut show a pattern intermediate between the former two subgroups. 4) Loss of RUNX1 WT and > 1 RUNX1mut showed an adverse impact on OS. Thus, not only the presence and number of RUNX1mut but also the conservation of an intact RUNX1 allele is biologically and clinically relevant. DisclosuresStengel:MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Perglerová:MLL2 s.r.o.: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

RUNX1 Mutation in Cytogenetically Normal Acute Myeloid Leukemia : Clinical Implications, Co-Mutation Analysis

Background and ObjectivesAcute Myeloid Leukemia (AML) is a cytogenetically and molecularly heterogeneous disease. In the recent decades, many genetic mutations and their clinical significances in AML have been identified with the development of new genomics technology. Based on these advances, new 2 entities were added to the WHO 2008 classification : AML with mutated NPM1 and AML with mutated CEBPA. Likewise, AML with RUNX1 mutation are now considered as a new provisional entity in the next update of WHO classification. In this work, we characterized patients with cytogenetically normal AML according to RUNX1 mutational status and analyzed several co-mutations by next generation sequencing.Patients and MethodsA total of 419 patients were included in the present study who met the following eligibility criteria: 1) age ≥ 15 years; 2) a diagnosis of AML with normal karyotype confirmed by conventional cytogenetic analysis. Analysis of genetic mutations were performed using targeted resequencing by Illumina Hiseq 2000 (Sureselect custom probe set targeting 94 myeloid gene panel including RUNX1 mutation). Samples for the confirmation of first complete response were also analyzed in 163 patients. The majority of patients (97%) received ’3+7’ standard induction chemotherapy. Median age was 53(range 15-84).ResultsOverall, most common mutations for this cohort were NPM1(33.9%), DNMT3A(30.3%), NRAS(20.2%), IDH2(15.0%), FLT3(12.2%), CEBPA(11.1%). RUNX1 mutations were found in 22 of 419 (5.4%) patients. 7 of 13 available samples in complete remission still had RUNX1 mutation. The patients with RUNX1 mutations were older than those with wild-type RUNX1. (p=0.006) and RUNX1 mutation had a trend of male preponderance. The WBC count and blast percentage of peripheral blood and bone marrow were not different according to RUNX1 mutational status. The complete response rate was significantly lower in RUNX1 mutated group compared with wild-type group. (57% vs. 84%, p=0.005) In univariable survival analysis, RUNX1 mutations were significantly associated with inferior event-free survival (EFS) (p<0.001), relapse-free survival (RFS) (p=0.009) and overall survival (OS) (p=0.002). However, in multivariable analysis, RUNX1 mutation was not an independent prognostic factor for inferior EFS (hazard ratio(HR) 1.48, p=0.286), RFS (HR 2.15, p=0.057) OS (HR 1.14, p=0.716). Co-mutation analysis revealed that ASXL1 (26%,p=0.001), KRAS (26%, p=0.009), BCOR (16%, p=0.032) were correlated with RUNX1 mutation. None of the patients with RUNX1 mutation had NPM1 mutation and only one patient had CEBPA mutation.ConclusionIn cytogenetically normal AML, RUNX1 mutation is observed in 5.4% and is mutually exclusive of the NPM1 and CEBPA mutation. Older age and lower complete response rate is correlated with RUNX1 mutation. In univariable survival analysis, RUNX1 mutation is associated with poor clinical outcomes. DisclosuresNo relevant conflicts of interest to declare.

Loss of Function RUNX1 Mutations Restrict Protein Biosynthesis during Pre-Leukemia and MDS Transition but Not after Leukemic Transformation

Runx1, a DNA binding subunit of core binding factors, is found frequently mutated in hematological malignancies. Runx1 mutation can be an early event in leukemogenesis endowing pre-leukemic stem cells with a selective advantage in the bone marrow, and is associated with an unfavorable outcome. In mouse models, loss of function (LOF) Runx1 mutations cause a broad decrease of ribosome biogenesis in hematopoietic stem and progenitor cells (HSPCs) by directly binding to ribosomal related genes essential for protein synthesis, and confers resistance to genotoxic stress (Cai et al. 2015 Cell Stem Cell 17(2):165-77). Paradoxially, leukemia cells generally require higher biosynthetic activity, and AML patients with LOF Runx1 mutations show upregulated ribosome signatures compared with those without Runx1 mutations (Silva et al., 2009 Blood 114:3001-3007). It remains unclear whether RUNX1 plays a role in regulating protein synthesis in leukemogenesis as in normal HSPCs, and if LOF Runx1 mutations are important for leukemia initiation, transformation and/or maintenance. To examine such mechanistic roles of RUNX1 in AML progression, we have used a previously reported MLL-PTD; Mx Cre; Runx1 Flox/Flox (Double mutant -DM) mouse model (Hayashi et al., 2015 Blood 126:303 ) that allows experimental tracking of the step-wise transition of HSPCs from pre-disease stage to a MDS-like stage, prior to full blown AML. Various subpopulations of the HSPCs, including genotypic HSCs, MPP, GMP, CMP, MEP, were isolated from the mice at pre-disease, MDS-like, and AML full-blown stages, and were assayed for protein synthesis rates by O-propargyl puromycin incorporation, DNA synthesis rates by BrdU labeling, and FACS analysis. At the pre-disease state, DM HSCs, as well as all the progenitor populations, had lower protein biosynthesis activity compared with similar populations of wild-type control or MLL-PTD mutant mice, consistent with LOF Runx1 mutations providing stress-resistance and survival advantage. As disease progressed, the DM mice developed MDS-like phenotypes including severe anemia and bone marrow fibrosis, with the HSCs (LSK CD34-Flt3- cells) showing increased protein synthesis rate compared with the pre-disease DM mice. Upon the onset of full-blown leukemia, the protein translation rates in all subpopulations of DM HSPCs were significantly faster than the control non-leukemic cells, regardless of the Runx1 mutant status. Importantly, preliminary analyses of two human AML samples found that CD34+ cells with LOF Runx1 mutations displayed a similarly enhanced protein synthesis rates than CD34+ leukemia bone marrow cells carrying wild type Runx1, as seen in the mouse model. Our results show that at early initiation, LOF Runx1 mutation supresses protein biosynthesis; during transition to MDS, the inhibitory regulation was bypassed in LT-HSCs (LSK CD34-Flt3-), suggesting that Runx1-controlled protein translation is involved in the early clonal selection of disease progression. In full-blown leukemia cells including the primitive subpopulations, however the protein synthesis rate appears to become uncoupled from Runx1 regulation possibly due to an activation of compensatory machineries. This study of the role of Runx1 mutation in pre-leukemia cell progression to full blown leukemia raises the question that while some tumor initiating mutations such as LOF Runx1 mutations contribute to the tumor initiation and transformation process, they may not be essential for maintaining certain crucial leukemia cell phenotypes such as protein biosynthesis. DisclosuresNo relevant conflicts of interest to declare.

Germline Variants of RUNX-1 in Myeloid Malignancy

Mutation of the master transcription factor RUNX1 (NM_001754) has a well characterized role in the pathogenesis of myeloid neoplasms. RUNX1 is located at 21q22 and is composed of a c-terminal transactivation domain (TAD) (269-480) and an n-terminal RUNT domain (amino acid 76-209) that binds DNA and mediates heterodimerization with CBF-β. If the RUNT domain is disrupted a RUNX1 mutant will behave in a dominant negative fashion. Mutations leaving the RUNT domain intact result in haploinsufficiency compounded by CBF-β binding site competition with the wild-type allele. Germline (GL) RUNX1 mutations cause the autosomal dominant predisposition syndrome familial platelet disorder with propensity to myeloid malignancy (FPDMM). In FPDMM, RUNX1 mutations, including RUNT domain missense mutations or nonsense mutations throughout the gene, confer mild chronic platelet disorders and a lifetime risk of MDS/AML. As with other GL predisposition syndromes a secondary event is considered necessary for progression, typically an additional mutation in the wild type RUNX1 allele. Onset of MDS/AML occurs at a median age of 33 years but the latency can be variable and ease of detection in younger cases likely distorts summary statistics. Inherited mutations in adults may be difficult to distinguish and have not been systematically explored. We hypothesize that a fraction of patients with otherwise typical RUNX1-positive MDS or related disorders are in fact carriers of a RUNX1 GL mutation.DNA obtained from a cohort of 1451 patients with myeloid neoplasia was analyzed using a multiamplicon deep next-generation sequencing (NGS) panel including all ORFs of RUNX1. A total of 124 patients (8.5%) were found to carry 117 unique RUNX1 mutations. Of these, 59 hits were missense and 58 nonsense; and in 94 (76%) of cases the RUNT domain was disrupted. Eighteen (15.3% RUNX1 positive patients) were previously described in FPDMM. We applied various bio analytic criteria to designate somatic status in 57 patients. The remaining 64 RUNX1 mutations were further investigated, when possible, by paired capillary sequencing of CD3- mononuclear DNA and in-vitro expanded CD3+ T-Cell DNA. We found 5/31 (16%) of these cases carried mutation in both myeloid and lymphoid lineages and were designated GL. This included 2 post-RUNT truncations (p.Y281*, p.S410*) and two RUNT domain missense mutations (p.R80C, p.S141L) all of which are expected to produce a dominant negative phenotype. A 5th GL mutation of uncertain significance in the TAD (p.M310I) was also discovered. Excluding this TAD missense mutation, each of these have been described as somatic factors in myeloid disease. The truncation at amino acid 410 is further downstream than all reported FPDMM variants to date. Chart review revealed an anamnestic presence of thrombocytopenia in 80% of proposed FPD patients. Suggestive family history was found in both truncated cases, with hematologic malignancy presenting in a 1st degree relative before 30 years of age. Median age at diagnosis of MDS/AML in GL cases was 46 years (range: 18-68 years) compared to 65 years in the cases designated somatic (range: 37-87 years). Two cases (p.S410* and p.M310I) had cytogenetic abnormalities on the 21st chromosome (trisomy 21 and t(8;21) respectively) and had no other molecular abnormalities detected by our NGS panel. A mutation in GPR98 of unknown significance was found with p.Y281*. Lastly the p.S141L mutant had cooperating mutations in chromatin modifiers ASXL1 and BCOR, and in RNA splicing gene LUC7L2. This study of an exemplary GL leukemia gene suggests that systematic searches for known and potential GL predisposition genes in otherwise typical adult cohorts may reveal a GL role in the evolution of myeloid neoplasms. Future work to determine genetic predisposition to leukemia in adults is essential. DisclosuresSekeres:Celgene: Membership on an entity's Board of Directors or advisory committees; Millenium/Takeda: Membership on an entity's Board of Directors or advisory committees. Carraway:Celgene: Research Funding, Speakers Bureau; Baxalta: Speakers Bureau; Incyte: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees.

Runx1 Is Strictly Required for Cbfb-MYH11 Induced Leukemia Development

Inversion of chromosome 16 is a consistent finding in patients with acute myeloid leukemia subtype M4 with eosinophilia (AML M4Eo), which generates a CBFB-MYH11 fusion gene. The prevailing hypothesis for the mechanism of leukemia development by CBFbeta-SMMHC, the fusion protein encoded by CBFB-MYH11, is that CBFbeta-SMMHC is a dominant negative repressor of RUNX1, a transcription factor that physically interacts with CBFbeta and CBFbeta-SMMHC. If this hypothesis is correct, reducing RUNX1 activity should facilitate leukemogenesis by CBFB-MYH11. In fact, loss-of-function mutations in RUNX1 are common in human AML, but not in inv(16) AML. However, we previously demonstrated that CBFB-MYH11 has RUNX1-repression independent functions (Hyde et al., Blood 115:1433, 2010). Moreover, we recently showed that a dominant negative allele of Runx1, Runx1-lz, delayed leukemogenesis by CBFB-MYH11 in a mouse model (Hyde et al., Leukemia 29:1771, 2015). These findings challenge the RUNX1-repression model for CBFbeta-SMMHC mediated leukemogenesis. However, our previous findings are not conclusive since the Runx1+/lz mice used in the previous study have one wild-type Runx1 allele, and still retain some Runx1 function.To definitively address this question, we crossed Cre-based conditional Runx1 knockout mice (Runx1f/f) with Cre-based conditional Cbfb-MYH11 knockin mice (Cbfb+/56M) and Mx1-Cre mice to generate Runx1f/f, Mx1-Cre, Cbfb+/56M mice, which express CBFbeta-SMMHC but not Runx1 after pIpC (poly I:C) treatment to induce Cre expression. Runx1f/f, Mx1-Cre, Cbfb+/56M mice had more severe platelet deficiencies and higher numbers of Lin-/Sca1-/C-kit+ progenitors and Lin-/Sca1+/C-kit+ hematopoietic stem cells in the bone marrow when comapred with Runx1f/f, Mx1-Cre mice. Unexpectedly Runx1f/f, Mx1-Cre, Cbfb+/56M mice also developed severe macrocytic anemia within two weeks after pIpC induction, which was lethal in about 1/3 of the mice. However, none of the Runx1f/f, Mx1-Cre, Cbfb+/56M mice developed leukemia up to one year after pIpC treatment. In contrast, all Mx1-Cre, Cbfb+/56M mice developed leukemia with an average survival of 4 months, as reported previously. These results suggest that Runx1 is strictly required for Cbfb-MYH11 induced leukemogenesis.To further study the mechanism of leukemogenesis, we performed RNA-Seq on C-kit+ bone marrow cells isolated from mice two weeks after pIpC treatment, to explore the global gene expression changes caused by Runx1 knockout on Cbfb-MYH11 expressing mice. Our preliminary data analysis showed that 1688 genes were differential expressed (Padj ≤0.05, FC ≥ 2) between Runx1f/f, Mx1-Cre, Cbfb+/56M and Mx1-Cre, Cbfb+/56M mice. Interestingly, many of these genes (48%) are Runx1 target genes. The above results suggest that mis-regulating the expression of Runx1 target genes contributes to leukemogenesis by CBFbeta-SMMHC. DisclosuresNo relevant conflicts of interest to declare.

The Hematopoietic Transcription Factors RUNX1 and ERG Prevent AML1-ETO Oncogene Overexpression and Onset of the Apoptosis Program in t(8;21) AMLs

The t(8;21) acute myeloid leukemia (AML)-associated oncoprotein AML1-ETO disrupts normal hematopoietic differentiation. Here, we have investigated its effects on the transcriptome and epigenome in t(8,21) patient cells. AML1-ETO binding was found at promoter regions of active genes with high levels of histone acetylation but also at distal elements characterized by low acetylation levels and binding of the hematopoietic transcription factors LYL1 and LMO2. In contrast, ERG, FLI1, TAL1, and RUNX1 bind at all AML1-ETO-occupied regulatory regions, including those of the AML1-ETO gene itself, suggesting their involvement in regulating AML1-ETO expression levels. While expression of AML1-ETO in myeloid differentiated induced pluripotent stem cells (iPSCs) induces leukemic characteristics, overexpression increases cell death. We find that expression of wild-type transcription factors RUNX1 and ERG in AML is required to prevent this oncogene overexpression. Together our results show that the interplay of the epigenome and transcription factors prevents apoptosis in t(8;21) AML cells.

Runx1 Phosphorylation by Src Increases Trans-activation via Augmented Stability, Reduced Histone Deacetylase (HDAC) Binding, and Increased DNA Affinity, and Activated Runx1 Favors Granulopoiesis

Src phosphorylates Runx1 on one central and four C-terminal tyrosines. We find that activated Src synergizes with Runx1 to activate a Runx1 luciferase reporter. Mutation of the four Runx1 C-terminal tyrosines to aspartate or glutamate to mimic phosphorylation increases trans-activation of the reporter in 293T cells and allows induction of Cebpa or Pu.1 mRNAs in 32Dcl3 myeloid cells, whereas mutation of these residues to phenylalanine to prevent phosphorylation obviates these effects. Three mechanisms contribute to increased Runx1 activity upon tyrosine modification as follows: increased stability, reduced histone deacetylase (HDAC) interaction, and increased DNA binding. Mutation of the five modified Runx1 tyrosines to aspartate markedly reduced co-immunoprecipitation with HDAC1 and HDAC3, markedly increased stability in cycloheximide or in the presence of co-expressed Cdh1, an E3 ubiquitin ligase coactivator, with reduced ubiquitination, and allowed DNA-binding in gel shift assay similar to wild-type Runx1. In contrast, mutation of these residues to phenylalanine modestly increased HDAC interaction, modestly reduced stability, and markedly reduced DNA binding in gel shift assays and as assessed by chromatin immunoprecipitation with the -14-kb Pu.1 or +37-kb Cebpa enhancers after stable expression in 32Dcl3 cells. Affinity for CBFβ, the Runx1 DNA-binding partner, was not affected by these tyrosine modifications, and in vitro translated CBFβ markedly increased DNA affinity of both the translated phenylalanine and aspartate Runx1 variants. Finally, further supporting a positive role for Runx1 tyrosine phosphorylation during granulopoiesis, mutation of the five Src-modified residues to aspartate but not phenylalanine allows Runx1 to increase Cebpa and granulocyte colony formation by Runx1-deleted murine marrow.

Prognostic Impact of Aberrant RUNX1 Expression in Patients with Acute Myeloid Leukemia Undergoing Allogeneic Hematopoietic Stem Cell Transplantation

Introduction:Despite intensive research most patients with acute myeloid leukemia (AML) still have a dismal prognosis. The transcription factor RUNX1 is a master regulator of myeloid differentiation and in AML its function is often disrupted by chromosomal translocations or mutations. These genetic alterations impact on leukemogenesis, disease progression and prognosis of AML patients. The expression of fusion proteins involving RUNX1 has been shown to exert a dominant-negative effect over wild-type RUNX1 leading to impaired myeloid differentiation and enhanced proliferation, while loss of RUNX1 may impact negatively on the frequency and self-renewal capacities of long-term hematopoietic stem cells. However, it remains unclear if differential expression of RUNX1 affects the AML phenotype. Here we analyzed the prognostic value of RUNX1 expression levels in AML patients undergoing allogeneic hematopoietic stem cell transplantation (HCT) after non-myeloablative conditioning (NMA).Patients & Methods:We analyzed 132 patients (median age 64 years [y], range 38-75y) with diagnostic bone marrow material available who received NMA-HCT (3x30mg/m2 Fludarabine on days -4 to -1 & 2Gy total body irradiation on day 0 followed by infusion of granulocyte-colony stimulating factor mobilized peripheral blood stem cells) at our institution between 2000 and 2012. Donors were human leukocyte (HLA)-matched related (15%) or HLA-matched (61%) or mismatched (24%) unrelated. 63 (48%) patients had a normal karyotype. Presence of FLT3 -ITD or FLT3 -TKD, mutational status of IDH1, IDH2, CEBPA, NPM1 and DNMT3A as well as expression levels of mir-9, mir-181a, BAALC, ERG and MN1 were determined at diagnosis. European LeukemiaNet cytogenetic classification was: 26% favorable, 27% intermediate-I, 20% intermediate-II and 27% adverse. RUNX1 expression at diagnosis was determined by qRT-PCR and normalized to 18S as internal standard. The median cut was used to define highand low RUNX1 expressers. Median follow-up was 3.9y for patients alive.Results:High RUNX1 expressers were more likely to have de novo AML (73% vs. 52%; P =.019) and higher % blasts in peripheral blood (median 38% vs. 19%; P =.004) and bone marrow (median 68% vs. 50%; P =.002) at diagnosis. Patients with high RUNX1 expression more often had IDH1 mutations by trend (P =.061) while there was no difference in IDH2 mutation frequency (P =.32). High RUNX1 expressers showed significantly higher ERG (P <.001), MN1 (P =.005) and mir-181a (P =.002) expression. High RUNX1 expression associated with longer overall (OS, P =.065) and event free survival (EFS, P =.109) by trend in the whole cohort (Figure 1 A, B). When we restricted our analysis to patients with a normal karyotype, high RUNX1 expression associated with a significantly longer OS (P =.035) and EFS (P =.041; Figure 1 C, D), while there was no prognostic impact of RUNX1 expression in patients with an abnormal karyotype (OS, P =.606 & EFS, P =.684). In multivariate analysis high RUNX1 expression independently associated with longer OS (Hazard Ratio 0.47 [95% Confidence Interval: 0.23 - 0.96]; P =.039) and longer EFS by trend (Hazard Ratio 0.57 [95% Confidence Interval: 0.29 - 1.14]; P =.113) in patients with normal karyotype.Conclusion:Our results revealed that high expression of the hematopoietic master regulator R UNX1 at diagnosis independently associated with survival in AML patients with normal karyotype receiving NMA-HCT. High RUNX1 expression associated with distinct clinical and molecular markers. Assessing pretreatment RUNX1 levels may help to refine risk stratification in AML patients undergoing NMA-HCT. [Display omitted] DisclosuresFranke:Novartis: Other: Travel Costs; MSD: Other: Travel Costs; BMS: Honoraria. Niederwieser:Novartis: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau.

Role of Runx2 phosphorylation in prostate cancer and association with metastatic disease.

The osteogenic transcription factor, Runx2, is abnormally expressed in prostate cancer (PCa) and associated with metastatic disease. During bone development, Runx2 is activated by signals known to be hyperactive in PCa including the RAS/MAP kinase pathway, which phosphorylates Runx2 on multiple serine residues including S301 and S319 (equivalent to S294 and S312 in human Runx2). This study examines the role of these phosphorylation sites in PCa. Runx2 was preferentially expressed in more invasive prostate cancer cell lines (PC3 > C4-2B > LNCaP). Furthermore, analysis using a P-S319-Runx2-specific antibody revealed that the ratio of P-S319-Runx2/total Runx2 as well as P-ERK/total ERK was highest in PC3 followed by C4-2B and LNCaP cells. These results were confirmed by immunofluorescence confocal microscopy, which showed a higher percentage of PC3 cells staining positive for P-S319-Runx2 relative to C4-2B and LNCaP cells. Phosphorylated Runx2 had an exclusively nuclear localization. When expressed in prostate cell lines, wild type Runx2 increased metastasis-associated gene expression, in vitro migratory and invasive activity as well as in vivo growth of tumor cell xenografts. In contrast, S301A/S319A phosphorylation site mutations greatly attenuated these Runx2 responses. Analysis of tissue microarrays from 129 patients revealed strong nuclear staining with the P-S319-Runx2 antibody in primary prostate cancers and metastases. P-S319-Runx2 staining was positively correlated with Gleason score and occurrence of lymph node metastases while little or no Runx2 phosphorylation was seen in normal prostate, benign prostate hyperplasia or prostatitis indicating that Runx2 S319 phosphorylation is closely associated with prostate cancer induction and progression towards an aggressive phenotype. These studies establish the importance of Runx2 phosphorylation in prostate tumor growth and highlight its value as a potential diagnostic marker and therapeutic target.

MiR-17 deregulates a core RUNX1-miRNA mechanism of CBF acute myeloid leukemia.

BackgroundCore Binding Factor acute myeloid leukemia (CBF-AML) with t(8;21) RUNX1-MTG8 or inv(16) CBFB-MYH11 fusion proteins often show upregulation of wild type or mutated KIT receptor. However, also non-CBF-AML frequently displays upregulated KIT expression. In the first part of this study we show that KIT expression can be also upregulated by miR-17, a regulator of RUNX1, the gene encoding a CBF subunit. Interestingly, both CBF leukemia fusion proteins and miR-17, which targets RUNX1-3′UTR, negatively affect a common core RUNX1-miRNA mechanism that forces myeloid cells into an undifferentiated, KIT-induced, proliferating state. In the second part of this study we took advantage of the conservation of the core RUNX1-miRNA mechanism in mouse and human, to mechanistically demonstrate in a mouse myeloid cell model that increased KIT-induced proliferation is per se a mechanism sufficient to delay myeloid differentiation.MethodsHuman (U937) or mouse (32D) myeloid clonal lines were used, respectively, to test: 1) the effect of RUNX1-MTG8 and CBFB-MYH11 fusion proteins, or upregulation of miR-17, on KIT-induced proliferation and myeloid differentiation, and 2) the effect of upregulation of KIT-induced proliferation per se on myeloid cell differentiation.ResultsIn the first part of this study we found that stable miR-17 upregulation affects, like the CBF-AML fusion proteins (RUNX1-MTG8 or CBFB-MYH11), a core RUNX1-miRNA mechanism leading to KIT-induced proliferation of differentiation-arrested U937 myeloid cells. In the second part of the study we harnessed the conservation of this core mechanism in human and mouse to demonstrate that the extent of KIT upregulation in 32D mouse myeloid cells with wild type RUNX1 can per se delay G-CSF-induced differentiation. The integrated information gathered from the two myeloid cell models shows that RUNX1 regulates myeloid differentiation not only by direct transcriptional regulation of coding and non-coding myeloid differentiation functions (e.g. miR-223), but also by modulating KIT-induced proliferation via non-coding miRNAs (e.g. miR-221).ConclusionsThe novelty of this study is dual. On the one hand, miRNAs (e.g. miR-17) can mimic the effects of CBF-AML fusion proteins by affecting a core RUNX1-miRNA mechanism of KIT-induced proliferation of undifferentiated myeloid cells. On the other hand, the extent of KIT-induced proliferation itself can modulate myeloid differentiation of cells with wild type RUNX1 function.Electronic supplementary materialThe online version of this article (doi:10.1186/s12943-014-0283-z) contains supplementary material, which is available to authorized users.

RUNX1/CBFβ Dosage Is Critical for MLL Leukemias Development

Transcription factors RUNX1/CBFβ play critical roles in hematopoiesis. Both of them are frequently involved in chromosomal translocations, point mutations, or deletions in acute leukemia. The mixed lineage leukemia (MLL) gene is also frequently involved in chromosomal translocations or partial tandem duplication in acute leukemia. We have previously shown that MLL, RUNX1, and CBFβ interact and form a regulatory complex to regulate downstream target genes. However, the functional consequence of MLL fusions on RUNX1/CBFβ activity remains unknown.To determine the impact of MLL fusion protein on RUNX1/CBFβ, we introduced either MLL, MLL-BP (longer N-terminal Flag-tagged MLL construct which contains CXXC domain; 1-1406), or MLL-fusions together with RUNX1, CBFβ, or both RUNX1 and CBFβ into 293T cells. MLL-BP and MLL fusions significantly decreased RUNX1 levels compared with controls (empty vector and MLL). CBFβ protein was mildly decreased by MLL-BP and MLL-fusions when expressed alone. However, when CBFβ was co-expressed with RUNX1, it was significantly decreased compared with controls. The expression levels of RUNX1 and CBFβ proteins in LSK cells from Mll-Af9 knock-in mice were significantly lower than those from wild-type (WT) mice. To confirm these findings in human acute myeloid leukemia (AML), we measured the expression of RUNX1 and CBFβ at both mRNA and protein levels in various leukemia cell lines. The expression levels of RUNX1 and CBFβ proteins were significantly decreased in AML cells with MLL fusion and MLL partial tandem duplication (MLL-PTD) compared with those in AML cells without MLL aberrations. MLL fusions still have CXXC domain. In MLL-PTD, the CXXC domain is duplicated. Our data showed that RUNX1 protein is not only down-regulated by MLL fusion proteins, but also by MLL-BP. Thus, to determine which region is involved in the down-regulation of RUNX1, we introduced a series of MLL deletion mutants into 293T cells and measured RUNX1 protein expression. MLL deletion mutants without CXXC domain had no effect on RUNX1 stability. The construct which contains point mutations in CXXC domain also lacked the ability to reduce RUNX1 expression. Furthermore, overexpression of only CXXC domain and flanking regions could down-regulate RUNX1 protein expression. These results suggest that MLL fusion proteins and the N-terminal MLL portion of MLL fusions down-regulate RUNX1 and CBFβ protein expression via the MLL CXXC domain and flanking regions.To understand the impact of RUNX1/CBFβ down-regulation on hematopoietic stem and progenitor cells (HSPCs), we generated RUNX1+/–/CBFβ+/– mice as a hypomorph model. The percentage of bone marrow (BM) LSK cells from RUNX1+/–/CBFβ+/– mice was significantly increased compared with that from WT mice. Using BM cells from these mice, we performed in vitro CFU assay and in vivo bone marrow transplantation (BMT) assay. BM cells from RUNX1+/–/CBFβ+/– mice provided more colonies in CFU assay compared with those from WT mice. To determine whether restoration of RUNX1 could repress the MLL mediated leukemogenesis, we retrovirally overexpressed WT RUNX1 in BM cells from Mll-Af9 knock-in mice. Using transduced BM cells, we performed in vitro CFU assay and in vivo BMT assay. RUNX1 overexpressed Mll-Af9 (Mll-Af9/RUNX1) cells underwent terminal differentiation after 2 times replating, while control vector transduced Mll-Af9 (Mll-Af9/Control) cells could still be replated more than 4 times. All the recipient mice transplanted with Mll-Af9/Control cells developed AML. In contrast, all the recipient mice transplanted with Mll-Af9/RUNX1 never develop AML. Furthermore, when we treated MLL leukemia cell lines with DOT1L inhibitor (EPZ-5676), RUNX1 protein levels in these MLL leukemia cell lines were significantly increased 48 hours after the treatment in comparing with controls treated with DMSO. However, there was no significant mRNA expression level change of RUNX1within 48 hours. Future studies are needed to fully understand the mechanism of whether this increasing RUNX1 protein level by DOT1L inhibitor is through blocking CXXC domain and flanking regions mediated degradation.In conclusion, MLL aberrations down-regulate RUNX1/CBFβ via their CXXC domain and flanking regions. Down-regulation of RUNX1/CBFβ plays critical role for MLL mediated leukemia development. Targeting RUNX1/CBFβ levels allows us to test novel therapies for MLL leukemias. DisclosuresMulloy:Celgene: Research Funding; Seattle Genetics: Research Funding; Amgen: Research Funding; NovImmune: Research Funding.

Mechanisms underlying platelet function defect in a pedigree with familial platelet disorder with a predisposition to acute myelogenous leukemia: potential role for candidate RUNX1 targets

Familial platelet disorder with a predisposition to acute myelogenous leukemia (FPD/AML) is an inherited platelet disorder caused by a germline RUNX1 mutation and characterized by thrombocytopenia, a platelet function defect, and leukemia predisposition. The mechanisms underlying FPD/AML platelet dysfunction remain incompletely clarified. We aimed to determine the contribution of platelet structural abnormalities and defective activation pathways to the platelet phenotype. In addition, by using a candidate gene approach, we sought to identify potential RUNX1-regulated genes involved in these defects. Lumiaggregometry, α-granule and dense granule content and release, platelet ultrastructure, αIIb β3 integrin activation and outside-in signaling were assessed in members of one FPD/AML pedigree. Expression levels of candidate genes were measured and luciferase reporter assays and chromatin immunoprecipitation were performed to study NF-E2 regulation by RUNX1. A severe decrease in platelet aggregation, defective αIIb β3 integrin activation and combined αδ storage pool deficiency were found. However, whereas the number of dense granules was markedly reduced, α-granule content was heterogeneous. A trend towards decreased platelet spreading was found, and β3 integrin phosphorylation was impaired, reflecting altered outside-in signaling. A decrease in the level of transcription factor p45 NF-E2 was shown in platelet RNA and lysates, and other deregulated genes included RAB27B and MYL9. RUNX1 was shown to bind to the NF-E2 promoter in primary megakaryocytes, and wild-type RUNX1, but not FPD/AML mutants, was able to activate NF-E2 expression. The FPD/AML platelet function defect represents a complex trait, and RUNX1 orchestrates platelet function by regulating diverse aspects of this process. This study highlights the RUNX1 target NF-E2 as part of the molecular network by which RUNX1 regulates platelet biogenesis and function.

Potential Role Of RUNX1 In The Pathogenesis Of Juvenile Myelomonocytic Leukemia (JMML)

Juvenile myelomonocytic leukemia (JMML) is an aggressive myeloproliferative neoplasm of young children. The only current curative treatment is bone marrow transplantation. Yet even with this aggressive therapy, ∼50% of children still die from their disease. Somatic mutations leading to constitutive activation of the tyrosine phosphatase Shp2 (also called PTPN11) or of RAS signaling occur in ∼90% cases of JMML. However, the transcription factors that act downstream of these aberrant signaling events have not been identified. We recently showed that RUNX1 is a direct interacting partner of Shp2 in megakaryocytic cells (Huang et al. 2012. Genes Dev 26: 1587-1601). Moreover, we showed that RUNX1 is normally negatively regulated by src-family kinase (SFK) mediated tyrosine phosphorylation in megakaryocytes and T-lymphocytes, and that Shp2 contributes to RUNX1 tyrosine dephosphorylation. We now show that overexpression of a mutant RUNX1 (RUNX1Y260F, Y375F, Y378F, Y379F, Y386F, “RUNX1-5F”), which is expected to mimic constitutive dephosphorylation by Shp2 in murine Lin- Sca-1+ c-kit+ (LSK) bone marrow cells is resistant to SFK-mediated tyrosine phosphorylation and leads to a dramatic expansion of CFU-M/CFU-GM and Gr1+Mac1+ cells in vitro and in vivo. In contrast, these effects are not seen when wild type RUNX1 or RUNX1Y260D, Y375D, Y378D, Y379D, Y386D (“RUNX1-5D”; mimicking constitutive RUNX1 tyrosine phosphorylation) are overexpressed. The RUNX1-5F expressing cells also have increased replating activity in serial colony forming assays, increased proliferation (BrdU incorporation), decreased apoptosis, and reduced cytokine dependence. This partially phenocopies conditional knock-in mice that express JMML associated activating Shp2 mutations. Flow sorted Gr1+Mac1+ cells from the RUNX1-5F transduced cultures expressed higher levels of the direct RUNX1 target gene PU.1, which plays a role in myelomonocytic growth, and Cyclin D1. To test whether RUNX1 is required for the myelomonocytic hyperproliferation in JMML, CD34+ peripheral blood cells from a patient with JMML and known activating Shp2 mutation (Shp2E76G) were lentivirally transduced with doxycycline-inducible RUNX1-5D or RUNX1-5F expression constructs and cultured under myeloid growth conditions. Upon doxycycline induction, the RUNX1-5D overexpressing cells (resistant to Shp2) exhibited at 32% reduction in BrdU incorporation. In contrast, the control RUNX1-5F expressing cells had no significant reduction in proliferation. These results are consistent with RUNX1 acting as an essential downstream target of activated Shp2 in JMML. As ERK mediated phosphorylation (downstream of RAS/MEK) is also known to increase RUNX1 activity, we propose that RUNX1 may be a common downstream transcriptional target of both activated Shp2 and RAS signaling in the pathogenesis of JMML. Disclosures: No relevant conflicts of interest to declare.

Src Kinase Can Activate RUNX1 Activity Via Phosphorylation Of C-Terminal Tyrosines and Activated RUNX1 Stimulates Granulopoiesis

We demonstrated that G-CSF signaling in lineage-negative marrow myeloid progenitors induces C-terminal SHP2 tyrosine phosphorylation more potently than does M-CSF signaling and that SHP2 knockdown in marrow or in the 32Dcl3 cell line reduces Cebpa gene transcription and impairs granulopoiesis.1,2 We also found that Runx1 directly activates Cebpa transcription via promoter elements and via a +37 kb enhancer and that Runx1 gene deletion reduces Cebpa mRNA and impairs granulopoiesis.3 Runx1 is phosphorylated by Src kinase on five tyrosines and dephosphorylated by SHP2, and RUNX1(5F) enhances megakaryopoiesis and rescues CD8 T cell development, whereas RUNX1(5D) is ineffective.4 Mutation to F prevents while change to D or E mimics phosphorylation. These findings suggested that in myeloid cells there exists a G-CSF to SHP2 to RUNX1 to Cebpa activation pathway directing granulopoiesis. We compared wild-type (WT) murine RUNX1b and a series of variants, Y260/375/378/379/386F (5F), Y260F (1F), Y375/378/379/386F (4F), 5D, 1D, and 4D for activation in 293T cells of a reporter with four RUNX1-binding sites linked to the TK promoter and the luciferase cDNA. Expression was compared by Western blotting. WT stimulated the reporter 5-fold on average, relative to empty CMV vector, 5D activated the reporter 15-fold, while 5F was nearly inactive. 4F was also nearly inactive, while 4D had WT activity, as did the 1F and 1D variants. Both 4F and 5F were expressed at or above WT levels; interestingly, 5D, but not 4D or 1D, was expressed at much higher levels than the other variants. A similar pattern was obtained with the MCSFR-Luc reporter. To further query the activity of the four C-terminal tyrosines, we evaluated the activity of Y375/378F (2F*), Y379/386F (2F), 2E, Y375F, Y378F, Y379F, and Y386F. The single residue mutations retained WT activity, whereas both the 2F* and 2F variants demonstrated significant reduction despite expression at or above WT levels. The 2E variant expressed at levels below WT but had increased activity. We then transfected the RUNX1 reporter with activated Src, Runx1, or both – in several experiments Src stimulated the reporter approximately 3-fold, on average, RUNX1 4-fold, and the combination 35-fold. In the same experiments, Src + 5F led to only 6.6-fold activation, and synergy was also lost if the reporter lacked the RUNX1-binding sites. These findings, together with the observation that SHP2 directly activates Src family kinases downstream of G-CSF receptor signaling,5 raises the possibility that in myeloid cells there actually is a G-CSF to SHP2 to Src to Runx1 to Cebpa activation pathway. Consistent with this idea, we find that the 5D variant has reduced interaction with co-expressed HDAC3 relative to WT, as assessed by co-IP, whereas the 5F variant has increased interaction. We previously found that RUNX1-ER stimulates endogenous Cebpa transcription when activated in 32Dcl3 cells. As the ER segment has trans-activating activity, to evaluate RUNX1 variants we utilizing the zinc-inducible MT promoter. Preliminary analysis indicates that both WT RUNX1b and its 5D variant can induce C/EBPa protein and RNA in this context. We previously found that exogenous RUNX1-ER can rescue granulopoiesis when transduced into marrow from Runx1(flox/flox);Mx1-Cre mice subjected 4 weeks earlier to pIpC injections. We now have constructed MIGC retroviral vectors expressing RUNX1, 5F, or 5D and GFP-Cre. These, or the empty MIG or MIGC vectors, were packaged and transduced into marrow isolated from RUNX1(flox/flox) mice. Three days later, lineage-depleted, GFP+ cells were cultured in methylcellulose with IL-3, IL-6, and SCF. Compared with MIG, MIGC reduced the percent CFU-G amongst CFU-G+CFU-M, from 55+/-1% to 10+/-2%; WT RUNX1 or 5D partially rescued granulopoiesis, to 31+/- 8% or 23+/-9% respectively, whereas 5F transduction led to only 4+/-1% CFU-G. In summary, in 293T cells and potentially in myeloid cells Src kinase activates Runx1 and this may favor granulopoiesis.1. Jack et al. Blood 114:2172, 2009. 2. Zhang and Friedman. Blood 118:2266, 2011. 3. Guo et al. Blood 119:4408, 2012. 4. Huang et al. Genes Dev. 26:1587, 2012. 5. Futami et al. Blood 118:1077, 2011. Disclosures:Cantor:AMGEN: Membership on an entity's Board of Directors or advisory committees.