Metabolomics,Unknown,Transcriptomics,Genomics,Proteomics

Dataset Information

Hematopoietic stem cells progressively deficient of glycogen synthase kinase initiate a pre-neoplastic state required for evolution to acute leukemia

ABSTRACT: The cellular origin and molecular progression towards aggressive cancers such as acute myeloid leukemia (AML) remain elusive. Clinically, Myelodysplastic syndromes (MDS) and related myeloproliferative neoplasias (MPN) disorders1-5 are believed to present as a precursor stage to lethal AML development. Despite the identification of cytogenetic abnormalities and increased activation of signaling in human MDS/MPN, specific pathways that either sustain or initiate disease progression and evolve into self-sustaining leukemic-initiating cells (L-ICs)13 have not been elucidated. Here we demonstrate that tissue specific loss of glycogen synthase kinase-3Î² (GSK3 betaÎ²) initiates the emergence of stable Pre-leukemic-ICs (PLIC) in vivo. In contrast to deletion or transgenic perturbation of pathways associated with AML eg. Î²-catenin/Wnt, serial transplantation of PL-IC produced abnormal hematological disease that phenotypically and molecularly resembles human MDS/MPN. PL-ICs were exclusively generated from GSK3 betaÎ² deficient hematopoietic stem cells (HSCs), indicating that disease initiation events collaborate with existing HSC self-renewal machinery. In the absence of GSK3 betaÎ², subsequent deletion of GSK3 betaÎ± caused rapid induction of L-ICs that give rise to lethal AML. As these processes were solely driven by dose-dependent deficiencies in GSK3 beta levels, our results suggest that perturbation of this pathway can sufficiently drive and recapitulate a step-wise progression of disease from HSCs to MDS/MPN and subsequent AML. Our study provides a molecular and cellular foundation to understand AML evolution from pre-leukemic precursors. We suggest that defining the molecular states of pre-neoplastic disease will allow patient stratification at early stages of MDS/MPN onset and aid in the development of therapeutic targeting of causal pathways responsible for the earliest stages of leukemic initiation events. 5 week post-transplanted recipient mice (CD45.1) with either Lin- bone marrow GSK3 alpha +/+ beta +/flx, alpha+/+ beta flx/flx, alpha -/- beta +/+ or alpha-/- beta fx/flx Rosa26-CreERTM cells (CD45.2) were induced 3 times with intraperitoneal injections of Tamoxifen (75mg/kg) at intervals of two to three days. Mice were sacrificed and analyzed 4 weeks after injection. Bone marrow lineage negative (Lin-), Sca1+, cKit+ cells were sorted from engrafted Lin- bone marrow GSK3 alpha +/+ beta +/flx, alpha+/+ beta flx/flx, alpha -/- beta +/+ or alpha-/- beta fx/flx cells for RNA extraction. Total RNA from purified populations was extracted and amplified as described previously (Shojaei et al., 2005). Amplified-labeled RNA was hybridized to Affymetrix Mouse Genome 430 2.0 Array.

ORGANISM(S): Mus musculus

SUBMITTER: Mickie Bhatia

PROVIDER: E-GEOD-53352 | biostudies-arrayexpress |

REPOSITORIES: biostudies-arrayexpress

ACCESS DATA

Similar Datasets

Project description:Full Title: Multilineage Dysplasia (MLD) in AML correlates with MDS-related cytogenetic abnormalities and a prior history of MDS or MDS/MPN but has no independent prognostic relevance: A comparison of 408 cases classified as “AML not otherwise specified” or “AML with myelodysplasia-related changes” The WHO classification of acute myeloid leukemia (AML) is hierarchically structured and integrates genetic information, data on patients’ history, and multilineage dysplasia (MLD). The category “AML with MDS-related changes” (AML-MRC) is separated from AML not otherwise specified (AML-NOS) by the presence of either MLD, MDS-related cytogenetics or MDS history. To clarify whether MLD alone is clinically relevant, we analyzed 408 adult patients categorized as AML-MRC or AML-NOS. EFS (Median 13.8 months vs. 16.0 months) and 3-year-OS (45.8% vs. 53.9%) did not differ significantly between patients with MLD and without. However, MLD correlated with pre-existing MDS (p<0.001) and MDS-related cytogenetics (p=0.035). Patients with MLD as sole AML-MRC criterion (AML-MLD-sole; n=90) had less frequently FLT3-ITD (p=0.032), and a lower median age than AML-NOS (n=232), but EFS, OS, and WBC did not differ significantly. Contrarily, patients with AML-NOS plus AML-MLD-sole (n=323) versus patients with MDS history or MDS-related cytogenetics (n=85) had better EFS (16.9 vs. 10.7 months; p=0.005) and 3-year-OS (55.8% vs. 32.5%; p=0.001). Gene expression profiles were measured for a subset of 96 patients. Analysis of the expression data showed distinct clusters for AML-MLD-sole and AML-NOS versus AML with MDS-related cytogenetics or MDS history. Thus, MLD demonstrated no independent clinical impact, while cytogenetics and MDS history were of prognostic relevance. This data suggest modifications in a revised WHO proposal. All 96 bone marrow samples were obtained from untreated patients at the time of diagnosis. Cells used for microarray analysis were collected from the purified fraction of mononuclear cells after Ficoll density centrifugation.

Project description:Transcriptional repressor Growth factor independence 1 (GFI1) is a key regulator of haematopoiesis. We previously established that the germline variant GFI1-36N promotes acute myeloid leukemia (AML) development, however the mechanism is not full elucidated. Here using multi-omics approach, we show GFI1-36N expression impedes DNA repair in leukemic cells. We demonstrate the presence of GFI1-36N is associated with increased frequency of chromosomal aberrations and mutational burden in murine and human AML cells. In particular, GFI1-36N modulates DNA repair pathways, O6-methylguanine-DNA-methyltransferase (MGMT) and homologous recombination repair (HR). Mechanistically, GFI1-36N exhibits impaired binding to Ndrg1 promoter element compared to GFI1-36S (wild type), causing decreased NDRG1 levels consequently leading to suppression of MGMT expression, imprinted at the transcriptome and proteome, thus leaving the AML cells vulnerable to DNA damaging agents. Targeting MGMT via temozolomide and HR via olaparib caused specifically extensive lethality in in vitro and ex vivo human and AML samples expressing GFI1-36N. Whereas the effects were insignificant on non-malignant GFI1-36S or GFI1-36N cells. Further, mice transplanted with GFI1-36N leukemic cells treated with combination of temozolomide and olaparib had a significantly longer AML-free survival than mice transplanted with GFI1-36S leukemic cells. In summary, we show that GFI1-36N disturbs DNA repair activity via NDRG1-MGMT axis and thus provides critical insights into novel therapeutic option for AML presented with GFI1-36N variant. Key Points Presence of GFI1-36N impedes Homologous DNA and MGMT DNA repair selectively in AML cells via the NDRG1-MGMT axis. Use of temozolomide and olaparib allows selectively targeting GFI1-36N leukemic cells. Introduction Gfi1 is a transcription factor which regulates the development of haematopoietic cells as well as neuronal and intestinal epithelial cells 1-5. We reported that a variant of GFI1, denominated GFI1-36N (characterized by an exchange of serine to asparagine at position 36), has a prevalence of 5-7% in a healthy control population but is found at an increased frequency of 10-15% among MDS and AML patients 6,7. The expression of germline variant GFI1-36N predisposes the carriers to develop de novo AML and MDS and correlates with a poor prognosis 6,7. Recently, we and other showed that malignant cells with GFI1-36N variant have increased H3K9-acetylation at target genes resulting in higher expression of genes required for cell survival and proliferation 8. GFI1 exerts its repressive role by recruiting histone-modifying enzymes (deacetylases HDAC1-3, demethylase LSD1, methyl transferase G9a) and regulates the accessibility of DNA to its target genes such as Hoxa9, Pbx1, Meis1, CSF1 and CSFR1 9-15. We also showed that GFI1 regulates apoptosis through its regulation of p53 in lymphoblastic leukemia 16 and we have demonstrated that GFI1 facilitates DNA repair 17. However, it is not known how these activities are affected in the GFI1-36N variant and whether the ability of GFI1 to regulate DNA repair pathways is maintained and how this might affect the development of myeloid malignancies. In this study, we leveraged multi-omics profiling to gain mechanistic insights into the molecular architecture that drives leukemia in the presence of GFI1-36N. We provide evidence that GFI1-36N interferes with DNA in leukemic myeloid cells, which leads to a higher frequency of genetic aberrations in MDS/AML cells. We also show that GFI1-36N myeloid leukemic cells are more sensitive to targeting MGMT and HR repair deficient cells, which opens a new selective therapeutic window to treat AML/MDS.

Dataset Information

Hematopoietic stem cells progressively deficient of glycogen synthase kinase initiate a pre-neoplastic state required for evolution to acute leukemia

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure