Spacing Constraints of Neighboring Zinc Finger Modules within GATA2

Abstract Genomic analyses in clinical and experimental contexts have accelerated discoveries of human genetic variants. While elucidating the consequences of conspicuously loss-of-function variants is highly tractable, decoding the impact of missense or non-coding variants is considerably more challenging. Previously, we described a germline variant in GATA2 in a patient with GATA2-deficiency syndrome, which inserts nine amino acids between the two zinc fingers (9aa-Ins), one of which mediates sequence-specific DNA binding (Cavalcante de Andrade Silva M. et al., Leukemia, 2021). Unlike other GATA2 coding region and enhancer variants identified (Bresnick E.H. et al., Blood Adv., 2020), it was unclear whether the altered zinc finger spacing would be inhibitory, stimulatory or of no consequence. The 9aa-Ins variant was defective in activating several target genes (Hdc, Ear2 and Tpsb2) in a genetic complementation assay with Gata2 -77 enhancer-mutant (-77 -/-) primary hematopoietic progenitor cells. As only several target genes were tested, we used RNA-seq to conduct a genome-wide comparison of the capacity of GATA2 and 9aa-Ins to activate and repress transcription. To elucidate mechanisms, we considered the following models: 1) 9aa-Ins fails to regulate all genes normally controlled by GATA2; 2) 9aa-Ins fails to repress all genes normally controlled by GATA2; 3) 9aa-Ins fails to activate genes normally controlled by GATA2; 4) 9aa-Ins ectopically regulates genes not controlled by GATA2. Using a genetic complementation approach with -77 -/- cells that were immortalized by the Hoxb8 transcription factor (hi-77 -/-) (Wang G.G. et al., Nat. Methods, 2006; Johnson K.D. et al., JEM, 2020), we compared GATA2 and 9aa-Ins activities when expressed at a comparable level. This analysis revealed 2,138 GATA2-regulated, 525 GATA2 and 9aa-Ins-regulated, and 414 ectopically-regulated genes (at least two-fold change, adjusted P-value <0.05). A similar number of genes were GATA2-activated (1,061) and repressed (1,077). Only 144 out of the 1,061 (14%) were 9aa-Ins-activated and 381 out of 1,077 (35%) were 9aa-Ins-repressed, illustrating the severe consequences of this mutation and a greater impact on activation versus repression. Statistical analysis with a range of P-values constraints (0.01 to 0.1) verified that activation by 9aa-Ins was more significantly impaired than repression (86% were no longer activated, and 65% were no longer repressed, P = 5.4 x 10 -6). Gene ontology analysis revealed that the 9aa insertion impaired GATA2-mediated activation of genes related to GPCR signaling and GATA2-mediated repression of genes related to innate immune machinery. The ectopically-regulated genes did not conform to a particular mechanism or pathway. Since it was unclear whether the transcriptional defects of the 9aa-Ins mutant reflect a unique inhibitory activity imparted by the 9aa sequence, we systematically varied the inter-zinc finger spacer length to establish whether any alterations can be tolerated. Using the genetic complementation assay, 2, 4, 6, 8, and 9aa spacer variants were compared with GATA2 for their capacity to regulate GATA2-activated genes (Hdc, Il1rl1, Gata1 and S1pr1) and repressed genes (Irf8, Il6st, Il6ra and Tifab). GATA2-mediated activation was compromised by insertions of two amino acids or more, whereas repression tolerated two and four amino acid insertions; 6, 8 and 9 were more inhibitory. Quantitative analyses revealed that a 6aa insertion reduced activation of the GATA2-activated genes by >50% of the wild type value, whereas the GATA2-repressed genes were still repressed by at least 50% (18% retention of activation and 83% retention of repression, P = 0.001). Thus, zinc finger spacing alterations differentially impacted activation versus repression. These results provide a rigorous foundation for interpreting variants that alter zinc finger spacing without disrupting vital finger residues. In vitro and in vivo functional analyses and molecular modeling are ongoing to further dissect the underlying mechanisms and ascertain the importance of genetic networks and circuits that are sensitive or resistant to human disease variants. Disclosures No relevant conflicts of interest to declare.

Download Full-text

Identification of ERG Specific Target Genes by Genome-Wide Screening in T-Lymphoblastic Leukemia

Blood ◽

10.1182/blood.v112.11.3788.3788 ◽

2008 ◽

Vol 112 (11) ◽

pp. 3788-3788

Author(s):

Liliana H Mochmann ◽

Konrad Neumann ◽

Juliane Bock ◽

Jutta Ortiz Tanchez ◽

Arend Bohne ◽

...

Keyword(s):

Target Genes ◽

Lymphoblastic Leukemia ◽

Specific Treatment ◽

P Value ◽

Dna Templates ◽

Genome Wide ◽

A Genome ◽

On Chip ◽

T Cell Leukemogenesis ◽

The Impact

Abstract The Ets related gene, ERG, encodes a transcription factor with a vital role in hematopoiesis. Recent findings have shown that ERG knockout mice require a minimum of one functional allele to ensure embryonic blood development and adult stem cell maintenance. Moreover, it was earlier reported that enforced expression of ERG induced oncogenic transformation in 3T3 cells. Overexpression of ERG, observed in a subset of acute T-lymphoblastic and acute myeloid leukemia patients, was associated with an inferior outcome. However, the impact of ERG contributing to this unfavourable phenotype has yet to be determined, as downstream targets of ERG in leukemia remain unknown. Herein, we conducted a genome-wide analysis of ERG target genes in T-lymphoblastic leukemia. Chromatin immunoprecipitation-on-chip array (ChIP-on-chip) analyses were performed using two ERG specific antibodies for the enrichment of ERG-bound DNA templates in T-lymphoblastic leukemia cells (Jurkat) with input DNA or IgG precipitated DNA as controls. Enriched DNA templates and control DNA were differentially labelled and co-hybridized to high resolution promoter chip arrays with 50–75mer probes (770,000) representing 29,000 annotated human transcripts (NimbleGen). Based on two independent ChIP-on-chip assays, bioinformatic analysis (ACME) yielded statistically significant enriched peaks (using a sliding window of 1000 bp, and a P-value < 0.0001) identifying promoter regions of 365 potential ERG target genes. From these genes, clustering by functional annotation was performed using the DAVID database and subsequently genes related to leukemia were further selected for quantitative PCR validation. The design of promoter primers included the highly conserved ETS GGAA DNA binding site. Genes with greater than two-fold enrichment (ERG ChIP versus control) included WNT2 (17-fold), OLIG2 (14-fold), WNT11 (7-fold), CCND1 (5-fold), WNT9A (4-fold), CD7 (3-fold), EPO (3-fold), ERBB4 (3-fold), RPBJL (3-fold), TRADD (3-fold), PIWIL1 (2-fold), TNFRSF25 (2-fold), TWIST1 (2-fold), and HDAC4 (2-fold). Interestingly, enriched target genes involved in developmental processes (WNT2, WNT9A, WNT11, TWIST1, PIWIL1, ERBB4, and OLIG2) have shown oncogenic potential when mutated or overexpressed. Thus, we hypothesize that overexpression of ERG may contribute to T-cell leukemogenesis by the deregulation of these oncogenic targets. Further disclosure of ERG directed downstream pathways may contribute to the design of specific treatment strategies (such as WNT inhibitors) with particular effectiveness in ERG deregulated leukemia.

Download Full-text

Mechanisms of Transcriptional Regulation and Transformation by HOXA9

Blood ◽

10.1182/blood.v120.21.sci-30.sci-30 ◽

2012 ◽

Vol 120 (21) ◽

pp. SCI-30-SCI-30

Author(s):

Jay L. Hess ◽

Cailin Collins ◽

Joel Bronstein ◽

Yuqing Sun ◽

Surya Nagaraja

Keyword(s):

Transcriptional Regulation ◽

Transcription Factors ◽

Transcriptional Activation ◽

Binding Sites ◽

Target Genes ◽

Conflicts Of Interest ◽

Hematopoietic Stem ◽

A Genome ◽

The Impact

Abstract Abstract SCI-30 HOXA9 plays important roles in both development and hematopoiesis and is overexpressed in more than 50 percent of acute myeloid leukemias (AML). Nearly all cases of AML with mixed lineage leukemia (MLL) translocations show increased HOXA9 expression, as well as cases with mutation of the nucleophosmin gene NPM1, overexpression of CDX2, and fusions of NUP98. In most cases, upregulation of HOXA9 is accompanied by upregulation of its homeodomain-containing cofactor MEIS1, which directly interacts with HOXA9. While HOXA9 alone is sufficient for transformation of hematopoietic stem cells in culture, the addition of MEIS1 increases the transformation efficiency and results in rapidly fatal leukemias in transplanted animals. Despite the crucial role that HOXA9 plays in development, hematopoiesis, and leukemia, its transcriptional targets and mechanisms of action are poorly understood. We have used ChIP-seq to identify Hoxa9 and Meis1 binding sites on a genome-wide level in myeloblastic cells, profiled their associated epigenetic modifications, identified the target genes regulated by HOXA9 and identified HOXA9 interacting proteins. HOXA9 and MEIS1 cobind at hundreds of promoter distal, highly evolutionarily conserved sites showing high levels of histone H3K4 monomethylation and CBP/P300 binding. These include many proleukemogenic gene loci, such as Erg, Flt3, Myb, Lmo2, and Sox4. In addition, HOXA9 binding sites overlap a subset of enhancers previously implicated in myeloid differentiation and inflammation. HOXA9 binding at enhancers stabilizes association of MEIS1 and lineage-restricted transcription factors, including C/EBPα, PU.1, and STAT5A/B thereby promoting CBP/p300 recruitment, histone acetylation, and transcriptional activation. Current efforts are focused on using both biochemical and genetic approaches to assess the role of HOXA9 “enhanceosome” components C/EBPα, PU.1, and STAT5A/B in transcriptional regulation and leukemogenesis. Studies to date suggest that C/EBPα and PU.1 binding can occur in the absence of HOXA9/MEIS1, supporting a model in which these proteins act as pioneer transcription factors for establishment of poised, but not activated, HOXA9-regulated enhancers. Work is under way to assess the impact of high-level HOXA9 and MEIS1 on enhanceosome assembly and the role of recruitment of transcriptional coactivators involved in target gene up- or downregulation, including histone acetyltransferases and chromatin remodeling complexes. Collectively, our findings suggest that HOXA9-regulated enhancers are a fundamental mechanism of HOX-mediated transcription in normal development that is deregulated in leukemia. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Mutation In Erythroid Specific Transcription Factor KLF1 Causes Hereditary Spherocytosis In the Nan (Neonatal Anemia) Hemolytic Anemia Mouse Model

Blood ◽

10.1182/blood.v116.21.3217.3217 ◽

2010 ◽

Vol 116 (21) ◽

pp. 3217-3217

Author(s):

Robert A White ◽

Daniel P. Heruth ◽

Troy Hawkins ◽

Derek Logsdon ◽

Margaret Gibson ◽

...

Keyword(s):

Amino Acid ◽

Hemolytic Anemia ◽

Zinc Finger ◽

Target Genes ◽

Zinc Finger Protein ◽

Hereditary Spherocytosis ◽

Competitive Inhibition ◽

Conflicts Of Interest ◽

Dominant Negative ◽

Zinc Finger Domain

Abstract Abstract 3217 The zinc finger protein Erythroid Krüuppel-like factor (EKLF, KLF1) regulates definitive erythropoiesis and terminal differentiation of red blood cells. KLF1 facilitates transcription through high affinity binding to CACCC elements within its erythroid-specific target genes which include genes encoding erythrocyte membrane skeleton (EMS) proteins. Deficiencies of EMS proteins lead to the hemolytic anemia Hereditary Spherocytosis (HS). We have identified a new HS gene by studying the hemolytic anemia mouse mutant Nan (Neonatal Anemia). Here we report that a mutation, E339D, in the second zinc finger domain of KLF1 is responsible for HS in Nan mice. The causative nature of the E339D mutation was verified with an allelic test cross between Nan/+ and heterozygous Klf1+/− knockout mice. Homology modeling predicted Nan KLF1 binds CACCC elements more tightly, suggesting that Nan KLF1 is a competitive inhibitor of wild type KLF1. Competitive inhibition may help explain the apparent disconnect between the finding that Nan/+ heterozygous mice are anemic, whereas Klf1+/− heterozygous mice are normal and haplo-sufficient. This is the first direct association of a KLF1mutation with a disease in adult mammals. After examining a small population of HS patients, we also discovered one HS patient with a KLF1 mutation, which resulted in a significant amino acid substitution (T251I) in the activator/repressor domain, 28 amino acid residues upstream of the first zinc finger domain. This HS subject had no known mutations in the exons or intron/exon boundaries of EMS genes (SPTA1, SPTB, ANK1, SLC4A1) which comprise 85% of HS mutations in humans. The lack of a known genetic mutation in EMS genes leaves this patient's KLF1 mutation as the leading candidate defect. The identification of the gene causing the Nan mutation is significant because the Nan mutant has allowed discovery of a new HS gene which may also cause this disease in humans. In addition, the putative dominant/negative competitive inhibition of the Nan mutation makes the Nan mouse an excellent model system to study the function of KLF1. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

The Impact of Distance to Treatment Center on the Outcome of AML

Blood ◽

10.1182/blood.v116.21.4742.4742 ◽

2010 ◽

Vol 116 (21) ◽

pp. 4742-4742 ◽

Cited By ~ 1

Author(s):

Bruno C. Medeiros ◽

Tamara J. Dunn ◽

Holbrook E Kohrt ◽

Steven Coutre ◽

Jason Gotlib ◽

...

Keyword(s):

Induction Chemotherapy ◽

Conflicts Of Interest ◽

Negative Impact ◽

Adverse Outcomes ◽

Treatment Center ◽

P Value ◽

Primary Therapy ◽

Newly Diagnosed ◽

Entire Cohort ◽

The Impact

Abstract Abstract 4742 Introduction While the distance patients travel to a treatment center (DTC) adversely impacts survival of patients with trauma, cardiac, or neurological disorders, as well as certain solid tumors, less is known of its influence in acute myeloid leukemia (AML). Care for patients with AML involves frequent emergent and urgent management, often complicating primary therapy provided in distant tertiary referral centers. We therefore hypothesized that increased DTC has a negative impact on outcome. We tested this hypothesis by assessing the effect of DTC on survival of patients with AML receiving care at a single institution. Patients and Methods Within the Stanford Leukemia Database, we identified 884 consecutive adult patients between 1993 and 2009 meeting the following criteria: age >=18, newly diagnosed AML (excluding APL), clinical management at Stanford University Medical Center (SUMC), and verified residence location available for DTC determination. Of these, 571 were deemed fit by the admitting physician to receive myelosuppressive induction chemotherapy. DTC was calculated by straight-line journey distance between home address at the time of diagnosis and treatment center. Results The median age for the entire cohort is 55 years and 322 patients (36%) are older than 60 years of age. Median survival for the entire cohort was 14.0 months. DTC was not univariately associated with outcome as a continuous variable. When testing for a critical DTC threshold impacting outcomes across the entire cohort, we found a significant correlation between longer DTC and adverse outcomes, shorter DTC was associated with lower OS. Patients living within 20 miles of SUMC had a worse median overall survival (10.4 months versus 15.0 months, HR 1.23, corrected p-value 0.02). However, when adjusted for administration of induction chemotherapy (p<0.0001), age at presentation (p<0.0001) and karyotype at diagnosis (CBF vs other; p-value- 0.92), the negative impact of DTC was lost (p=0.08). Conclusion After accounting for confounding factors, DTC has no significant impact on the outcome of newly diagnosed AML patients receiving care at our institution. Unlike non-hematologic malignancies, distance to treatment center likely does not adversely influence outcomes for patients with AML. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Regulation of Erythropoiesis by Histone Methyltransferase EZH2 Inhibitor 3-Deazaneplanocin A (DZNep)

Blood ◽

10.1182/blood.v120.21.982.982 ◽

2012 ◽

Vol 120 (21) ◽

pp. 982-982

Author(s):

Tohru Fujiwara ◽

Haruka Saitoh ◽

Yoko Okitsu ◽

Noriko Fukuhara ◽

Yasushi Onishi ◽

...

Keyword(s):

Molecular Mechanisms ◽

Target Genes ◽

Conflicts Of Interest ◽

K562 Cells ◽

Erythroid Differentiation ◽

Erythroid Cell ◽

Mrna Levels ◽

Wide Range ◽

The Impact ◽

Chip Analysis

Abstract Abstract 982 Background. EZH2, a core component of Polycomb repressive complex 2 (PRC2), plays a role in transcriptional repression through mediating trimethylation of histone H3 at lysine 27 (H3K27), and is involved in various biological processes, including hematopoiesis. Overexpression of EZH2 has been identified in a wide range of solid tumors as well as hematological malignancies. Recent studies indicated that 3-deazaneplanocin A (DZNep), an inhibitor of EZH2, preferentially induces apoptosis in cancer cells, including acute myeloid leukemia and myelodysplastic syndromes, implying that EZH2 may be a potential new target for epigenetic treatment. On the other hand, whereas PRC2 complex has been reported to participate in epigenetic silencing of a subset of GATA-1 target genes during erythroid differentiation (Yu et al. Mol Cell 2009; Ross et al. MCB 2012), the impact of DZNep on erythropoiesis has not been evaluated. Method. The K562 erythroid cell line was used for the analysis. The cells were treated with DZNep at doses of 0.2 and 1 microM for 72 h. Quantitative ChIP analysis was performed using antibodies to acetylated H3K9 and GATA-1 (Abcam). siRNA-mediated knockdown of EZH2 was conducted using Amaxa nucleofection technology™ (Amaxa Inc.). For transcription profiling, SurePrint G3 Human GE 8 × 60K (Agilent) and Human Oligo chip 25K (Toray) were used for DZNep-treated and EZH2 knockdown K562 cells, respectively. Gene Ontology was analyzed using the DAVID Bioinformatics Program (http://david.abcc.ncifcrf.gov/). Results. We first confirmed that DZNep treatment decreased EZH2 protein expression without significantly affecting EZH2 mRNA levels, suggesting that EZH2 was inhibited at the posttranscriptional level. We also confirmed that DZNep treatment significantly inhibited cell growth. Interestingly, the treatment significantly induced erythroid differentiation of K562 cells, as determined by benzidine staining. Transcriptional profiling with untreated and DZNep-treated K562 cells (1 microM) revealed that 789 and 698 genes were upregulated and downregulated (> 2-fold), respectively. The DZNep-induced gene ensemble included prototypical GATA-1 targets, such as SLC4A1, EPB42, ALAS2, HBA, HBG, and HBB. Concomitantly, DZNep treatment at both 0.2 and 1 microM upregulated GATA-1 protein level as determined by Western blotting, whereas the effect on its mRNA levels was weak (1.02- and 1.43-fold induction with 0.2 and 1 microM DZNep treatment, P = 0.73 and 0.026, respectively). Furthermore, analysis using cycloheximide treatment, which blocks protein synthesis, indicated that DZNep treatment could prolong the half-life of GATA-1 protein, suggesting that DZNep may stabilize GATA-1 protein, possibly by affecting proteolytic pathways. Quantitative ChIP analysis confirmed significantly increased GATA-1 occupancy as well as increased acetylated H3K9 levels at the regulatory regions of these target genes. Next, to examine whether the observed results of DZNep treatment were due to the direct inhibition of EZH2 or hitherto unrecognized effects of the compound, we conducted siRNA-mediated transient knockdown of EZH2 in K562 cells. Quantitative RT-PCR analysis demonstrated that siRNA-mediated EZH2 knockdown had no significant effect on the expression of GATA-1 as well as erythroid-lineage related genes. Furthermore, transcription profiles of the genes in the quantitative range of the array were quite similar between control and EZH2 siRNA-treated K562 cells, with a correlation efficient of 0.977. Based on our profiling results, we are currently exploring the molecular mechanisms by which DZNep promotes erythroid differentiation of K562 cells. Conclusion. DZNep promotes erythroid differentiation of K562 cells, presumably through a mechanism not directly related to EZH2 inhibition. Our microarray analysis of DZNep-treated K562 cells may provide a better understanding of the mechanism of action of DZNep. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Aberrant Expression of DOCK4 Leads to Disruption of the F-Actin Skeleton and Altered Membrane Stability in MDS Erythroblasts and Mature Erythrocytes

Blood ◽

10.1182/blood.v120.21.924.924 ◽

2012 ◽

Vol 120 (21) ◽

pp. 924-924 ◽

Cited By ~ 1

Author(s):

Sriram Sundaravel ◽

Tushar D. Bhagat ◽

Carolina Schinke ◽

David Ebenezer ◽

Hui Liu ◽

...

Keyword(s):

Actin Polymerization ◽

Conflicts Of Interest ◽

Osmotic Fragility ◽

Membrane Stability ◽

Refractory Anemia ◽

P Value ◽

Down Regulation ◽

Acute Leukemias ◽

Aberrant Expression ◽

The Impact

Abstract Abstract 924 Myelodysplastic syndromes (MDS) are a group of clonal hematopoietic disorders that are commonly characterized by anemia due to ineffective hematopoiesis. Even though a third of patients with MDS may transform to acute leukemias, cytopenias drive morbidity for most patients. Anemia remains a major cause of morbidity from fatigue. Most of the morbidity experienced by such patients is due to low red blood counts and therefore studies on the molecular pathogenesis of dysplastic erythropoiesis leading to anemia are critically needed. We have identified DOCK4, an important cofactor for various GTPases located on the chromosome 7q segment as a novel silenced gene in MDS and show that it's down regulation leads to disruption of normal erythropoiesis. In an attempt to uncover genes aberrantly expressed in MDS, we initially performed an integrative genomic analysis of primary hematopoietic cells from MDS patients. These studies revealed that DOCK4 is significantly under-expressed and hypermethylated in MDS stem and progenitor cells. Immunohistochemcial analysis revealed significantly reduced levels of DOCK4 in MDS erythroblasts. We then evaluated DOCK4 expression in a large published cohort of MDS gene expression datasets (N=183) and found that DOCK4 expression was strikingly reduced in the subset of 55 MDS patients with refractory anemia (RA; P value = 0.006). The RA subset of patients only has isolated anemia as the clinical presentation and has no apparent abnormalities in white cells or platelets. This association strongly alludes to a role of DOCK4 down-regulation in the erythroid dysplasia and anemia seen in this disease. Inorder to elucidate the functional implications of aberrant DOCK4 expression during erythropoiesis we used a dynamic model of human erythropoiesis to determine normal expression pattern during terminal differentiation and the impact of silencing DOCK4 expression on healthy primary erythroblasts. These studies revealed that DOCK4 is highly expressed during late stages of normal erythropoiesis and knockdown of DOCK4 in primary erythroblasts disrupted the F-actin skeleton. We then examined F-actin skeletal disruption in CD34+ derived erythroblasts from MDS patients. In order to quantify the extent of actin filament disruption directly in patient derived erythroblasts we first developed an assay based on multispectral flow cytometry (ImageStream™). This assay not only allowed us to visualize individually F-actin-stained cells but also allowed us to determine the percentages of cells in a given sample that contained shorter fragmented F-actin. These experiments revealed that approximately 85% of the cells in MDS patients with -7q deletion and/or hypermethylated promoter region in the DOCK4 gene contained shorter disrupted actin compared to the healthy controls that showed only 10% of the cells with disrupted F-actin. The level of F-actin disruption in healthy samples treated with cytochalasin D, an inhibitor of actin polymerization was 90%. We then examined the membrane stability of -7q MDS erythrocytes by performing osmotic fragility assays and found that these patients possessed erythrocytes that were more fragile compared to healthy erythrocytes. Based on these results we conclude that DOCK4 is an important signaling intermediate that is instrumental in maintaining erythroblast membrane homeostasis and silencing of DOCK4 in MDS contributes to anemia. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Large Conserved Domains Of Low DNA Methylation Maintained By 5-Hydroxymethycytosine and Dnmt3a

Blood ◽

10.1182/blood.v122.21.2406.2406 ◽

2013 ◽

Vol 122 (21) ◽

pp. 2406-2406

Author(s):

Mira Jeong ◽

Deqiang Sun ◽

Min Luo ◽

Yun Huang ◽

Myunggon Ko ◽

...

Keyword(s):

Gene Expression ◽

Dna Methylation ◽

Hox Genes ◽

Conflicts Of Interest ◽

P Value ◽

Genomic Feature ◽

Conserved Domains ◽

Hematopoietic Stem ◽

Genome Wide ◽

The Impact

Abstract Identification of recurrent leukemia-associated mutations in genes encoding regulators of DNA methylation such as DNMT3A and TET2 have underscored the critical importance of DNA methylation in maintenance of normal physiology. To gain insight into how DNA methylation exerts the central role, we sought to determine the genome-wide pattern of DNA methylation in the normal precursors of leukemia cells: the hematopoietic stem cell (HSC), and investigate the factors that affect alterations in DNA methylation and gene expression. We performed whole genome bisulfite sequencing (WGBS) on purified murine HSCs achieving a total of 1,121M reads, resulting in a combined average of 40X coverage. Using Hidden Markov Model we identified 32,325 under-methylated regions (UMRs) with average proportion of methylation ≤ 10% and by inspecting the UMR size distribution, we discovered exceptionally large “methylation Canyons” which span highly conserved domains frequently containing transcription factors and are quite distinct from CpG islands and shores. Methylation Canyons are a distinct genomic feature that is stable, albeit with subtle differences, across cell-types and species. Canyon-associated genes showed a striking pattern of enrichment for genes involved in transcriptional regulation (318 genes, P=6.2 x 10-123), as well as genes containing a homeobox domain (111 genes, P=3.9 x 10-85). We compared Canyons with TF binding sites as identified from more than 150 ChIP-seq data sets across a variety of blood lineages (>10)19 and found that TF binding peaks for 10 HSC pluripotency TFs are significantly enriched in entirety of Canyons compared with their surrounding regions. Low DNA methylation is usually associated with active gene expression. However, half of Canyon genes associated with H3K27me3 showed low or no expression regardless of their H3K4me3 association while H3K4me3-only Canyon genes were highly expressed. Because DNMT3A is mutated in a high frequency of human leukemias24, we examined the impact of loss of Dnmt3a on Canyon size. Upon knockout of Dnmt3a, the edges of the Canyons are hotspots of differential methylation while regions inside of Canyon are relatively resistant. The methylation loss in Dnmt3a KO HSCs led Canyon edge erosion, Canyon size expansion and addition of 861 new Canyons for a total of 1787 Canyons. Canyons marked with H3K4me3 only were most likely to expand after Dnmt3a KO and the canyons marked only with H3K27me3 or with both marks were more likely to contract. This suggests Dnmt3a specifically is acting to restrain Canyon size where active histone marks (and active transcription) are already present. WGBS cannot distinguish between 5mC and 5hmC, so we determined the genome-wide distribution of 5hmC in WT and Dnmt3a KO HSCs using the cytosine-5-methylenesulphonate (CMS)-Seq method in which sodium bisulfate treatment convert 5hmC to CMS; CMS-containing DNA fragments are then immunoprecipitated using a CMS specific antiserum. Strikingly, 5hmC peaks were enriched specifically at the borders of Canyons. In particular, expanding Canyons, typically associated with highest H3K4me3 marking, were highly enriched at the edges for the 5hmC signal suggesting a model in which Tet proteins and Dnmt3a act concomitantly on Canyon borders opposing each other in alternately effacing and restoring methylation at the edges, particularly at sites of active chromatin marks. Using Oncomine data, we tested whether Canyon-associated genes were likely to be associated with hematologic malignancy development and found Canyon genes were highly enriched in seven signatures of genes over-expressed in Leukemia patients compared to normal bone marrow; in contrast, four sets of control genes were not similarly enriched. Further using TCGA data, we found that expressed canyon genes are significantly enriched for differentially expressed genes between patients with and without DNMT3A mutation (p value<0.05) Overall, 76 expressed canyon genes, including multiple HOX genes, are significantly changed in patients with DNMT3A mutation (p=0.0031). Methylation Canyons, the novel epigenetic landscape we describe may provide a mechanism for the regulation of hematopoiesis and may contribute to leukemia development. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Genome-Wide CRISPR-Cas9 Screen Identifies Sensitizers to LSD1 Inhibition in MLL-Translocated Human AML Cells

Blood ◽

10.1182/blood-2018-178 ◽

2018 ◽

Vol 132 (Supplement 1) ◽

pp. 178-178

Author(s):

Gauri Deb ◽

Bettina Wingelhofer ◽

Emma Williams ◽

Hui-Sun Leong ◽

Tim CP Somervaille

Keyword(s):

Amino Acids ◽

Cell Proliferation ◽

Clinical Trials ◽

Conflicts Of Interest ◽

Myeloid Differentiation ◽

Functional Responses ◽

Pharmacological Inhibition ◽

Genome Wide ◽

A Genome ◽

Mtorc1 Pathway

Abstract Lysine-specific demethylase (LSD1, also known as KDM1A) is an epigenetic regulator that has recently emerged as a potential therapeutic target in acute myeloid leukemia (AML). It is a flavin dependent monoamine oxidase which can demethylate monomethyl or dimethyl lysine 4 of histone H3. Pharmacological inhibition of LSD1 induces differentiation of blast cells in MLL-translocated AML and has shown significant promise in pre-clinical studies. With LSD1 inhibitors advancing through early-phase clinical trials, there is a strong pre-clinical rationale for the identification of genes whose protein products collaborate with LSD1 to retard differentiation in cancer and which could potentially be targeted in combination therapies for enhanced therapeutic benefit. To identify potential drug-gene synthetic lethal interactions, we performed a genome wide loss-of-function CRISPR-Cas9 screen in human THP1 AML cells treated with a potent and selective tranylcypromine-derivative LSD1 inhibitor trans-N-((2-methoxypyridin-3-yl)methyl)-2-phenylcyclopropan-1-amine (OG86). THP1 cells exhibit a t(9;11) MLL gene rearrangement and display similar phenotypic and functional responses to those observed in primary MLL-translocated AML cells following LSD1 inhibition. The cells were transduced with lentiviral human CRISPR knockout (hGeCKOv2) library containing 122,411 sgRNAs targeting 19,050 protein coding genes and 1864 miRNA precursors in the human genome. The transduced cells were divided into two groups, treated with either DMSO or 250nM OG86 and maintained in culture for ~10 population doublings. To investigate sgRNA representation in the cell population harvested at different stages of the screen, sgRNA cassettes were PCR amplified from genomic DNA and deep sequenced; negatively selected genes were identified with the MAGeCK computational algorithm. After successful initial quality assessment of the screen, we next searched for genes selectively depleted in OG86-treated versus DMSO-treated THP1 cells in samples collected on Day 15 and Day 18. At a false discovery rate of 7.5% there were 10 expressed genes whose sgRNA representation was depleted at both time points. In particular, these included two genes coding for components of the MTOR signaling pathway: MTOR associated protein, LST8 homolog (MLST8) and Ras-related GTP-binding protein A (RRAGA). In the Day 18 comparison, an additional MTOR pathway gene LAMTOR2 scored among the ten most depleted. MLST8 is a core component of TORC1/TORC2 complex and RRAGA is involved in the activation of TORC1 by amino acids. Recruitment of both RAG proteins and TORC1 to lysosomal membranes in response to amino acids, and the consequent activation of TORC1 signalling, requires the trimeric Ragulator complex, of which LAMTOR2 is a member.Based on our screen, we hypothesized that THP1 AML cells exposed to pharmacologic inhibition of LSD1 exhibit increased sensitivity to concomitant inhibition of the amino acid sensing component of the TORC1 pathway. Using genetic knockdown and pharmacological inhibition strategies, we then validated our screen hits in combination with LSD1 inhibition in targeting human AML cells. RNAi based knockdown of RRAGA, LAMTOR2 and MLST8 in combination with LSD1 inhibition were found to promote myeloid differentiation and reduce cell proliferation in THP1 cells. Interestingly the mTORC1 pathway inhibitor everolimus (RAD001) showed at least additive effect in combination with OG86 to decrease THP1 cell proliferation and promote immunophenotypic differentiation. Comparison of transcription changes in combined versus single treatment conditions by RNA-seq analysis further confirmed a more extensive and wide-ranging upregulation of a myeloid differentiation program upon concomitant inhibition of LSD1 and mTORC1 pathway. In vitro studies performed in primary patient AML cells gave similar results. Finally, in vivo studies using AML patient-derived xenograft mouse model confirmed that combination treatment promotes a strong myeloid differentiation program. In conclusion, we report here that inhibition of mTORC1 sensitizes human MLL-translocated AML cells to LSD1 inhibitor-mediated differentiation therefore highlighting a novel combination approach for evaluation in clinical trials. Disclosures No relevant conflicts of interest to declare.

Download Full-text

High WT1 Expression Has An Independent Positive Influence On CR Rate and EFS in Non M3 AML Patients Treated with Chemotherapy.

Blood ◽

10.1182/blood.v114.22.4675.4675 ◽

2009 ◽

Vol 114 (22) ◽

pp. 4675-4675

Author(s):

Nicoletta Colombo ◽

Raffaella Grasso ◽

Maurizio Miglino ◽

Marino Clavio ◽

Gianmatteo Pica ◽

...

Keyword(s):

Regression Model ◽

Prognostic Value ◽

Cox Regression ◽

Conflicts Of Interest ◽

Categorical Variables ◽

P Value ◽

Prognostic Impact ◽

Chi Square Analysis ◽

The Impact

Abstract Abstract 4675 The prognostic value of WT1 expression at diagnosis is still controversial. It has been retrospectively evaluated in 99 consecutive non pretreated non M3 AML patients who had undergone a complete prognostic work up at diagnosis and had received intensive chemotherapy. Biological markers were evaluated on fresh marrow samples collected at diagnosis. WT1 expression was evaluated using TaqMan Gene Expression Assays as described. All patients received induction therapy with combination of fludarabine, Ara-C and anthracycline ± low dose gemtuzumab ozogamicin (n. 59) or with a conventional combination of Ara-C and anthracycline (n. 40) A conventional post-induction chemotherapy including intermediate dosage Ara-C was administered to all responding patients. Univariate comparisons between patients in CR vs non CR were performed using chi-square analysis or Fisher's exact test for categorical variables and t-test for continuous variables. P values < 0.05 were considered statistically significant. Analyses were performed using SPSS. The prognostic impact of WT1 expression was evaluated using quartiles as cut off point and selecting the one with the lowest p value. The event free survival and OS were calculated using the Kaplan Meier method. Non CR after the first induction course, relapse and death due to any cause were considered events. OS and EFS duration were calculated from start of treatment. The impact of multiple predictor variables was assessed by multivariate analyses according to the Cox regression model for OS and EFS while for the evaluation of RC was used the Logistic regression model. Median age of patients was 59 years (range 17-81). Cytogenetic alterations were prognostically favorable in 3 patients and belonged to the intermediate prognostic group in 77 patients (normal karyotype in 75 patients and +8 in two). Nineteen patients had a poor prognosis cytogenetics. For statistical analyses we considered two karyotipic groups: unfavorable (19 patients) and not unfavorable (80 patients). CRs were 60/99 (60%), of which 40 in 51 patients aged 60 or less (78%) and 20 in 48 older than 60 years (41%). Twenty-six patients relapsed, 54 are alive, 45 have died, with a median follow up of 360 days (range 20-2300). In Table 1 are reported clinical indicators of outcome being patients grouped according to the percentile of WT1 expression with the lowest p value (75th). Statystical analysis showed that all WT1 quartiles were balanced for other prognostic factors, such as cytogenetics, BAALC expression, FLT3 and NPMA and B mutations, age, blast count and therapy. The lack of consense on the role of WT1 level at diagnosis in the prognostic stratification indicate that further clinical studies are required. The clear correlation between the level of WT1 transcript and the tumor burden explains why WT1 is used in the follow up of leukemic patients as universal marker of residual disease, also in patients with specific chimeric products. On the contrary, the biological explanation of the prognostic impact of WT1 transcript level at diagnosis remains uncertain. Over the years WT1 gene has been considered as an oncogene or a tumor suppressor gene. In our experience the protective influence of high WT1 expression cannot be explained with an association with good prognosis biological features (such as mut NPM and / or low BAALC). The positive prognostic value of high WT1 expression might be implicated either with WT1 antioncogenic function, or with the stimulating effect of WT1 oncogene on leukemic cellular cycle, possibly associated with an enhanced response to chemotherapy.Table 1WT1 <= 2400 N./N.pts (%)WT1 > 2400 N./N.pts (%)p univ,p multiv.*RR (95% CI)CR (all karyotypes)41/ 75 (54)19/24 (82)0,0260.063.364 (0.927-12.202)CR (int/good karyot.)36/59 (61)19/210.010,0276.649 (1.240-35.645)CR (denovo AML int kar)31/45 (69)14/15 (98)0.020,03412.557 (1.218-129.446)CR (denovo, N.K.)26/40 (65)15/16 (94)0.0250.0413.430 (1.111-162.318)EFS at 24 months (all karyotypes)8%6%0.0020.050.486 (0.235-1.007)EFS at 24 months (int / good karyot.)9%64%0.0010.0230.360 (0.150-0.866)EFS at 24 months (de novo, N.K.)5%70%0.0010.0070.227 (0.077-0.671)OS (all karyot)15%55%0,110,660.837 (0.371-1.890)OS (int/good kar.)18%63%0,050,180.507 (0.186-1.381)Table 1 legend: * for multivariate analysis age, karyotype, FLT3, NPM mutation, BAALC expression, denovo/secondary disease were considered. Disclosures: No relevant conflicts of interest to declare.

Download Full-text

Polymorphisms in the CYP450 Genes Influences Regimen Related Toxicity and Outcome in Patients with Beta Thalassaemia Major Undergoing HSCT.

Blood ◽

10.1182/blood.v114.22.869.869 ◽

2009 ◽

Vol 114 (22) ◽

pp. 869-869

Author(s):

Poonkuzhali Balasubramanian ◽

Salamun Desire ◽

Vikram Mathews ◽

Kavitha M Lakshmi ◽

Shaji R Velayudhan ◽

...

Keyword(s):

Risk Factors ◽

Annual Meeting ◽

Conflicts Of Interest ◽

Univariate Analysis ◽

P Value ◽

Thalassaemia Major ◽

Hematopoietic Stem ◽

Beta Thalassaemia ◽

Variant Genotype ◽

The Impact

Abstract Abstract 869 Polymorphisms in drug metabolizing enzymes are known to contribute to inter-individual differences in the pharmacokinetics (PK) of the two most commonly used drugs for conditioning for hematopoietic stem cell transplantation (HSCT), busulfan (Bu) and cyclophosphamide (Cy) and their metabolites in plasma. We have previously reported the impact of CYP genes on the PK of Cy, [Blood (ASH Annual Meeting Abstracts), Nov 2004; 104: 99] and the influence of Cy PK on transplant outcome [Blood (ASH Annual Meeting Abstracts), Nov 2004; 104: 1820]. We have now extended this study to evaluate a total of 19 polymorphisms in 11 genes that are known to be involved in the metabolism of Bu and Cy. 180 of the 276 patients with thalassemia major who underwent HSCT between March 1991 and Dec 2008 and for who genomic DNA was available were included in the study. The following polymorphisms were screened using PCR followed by RFLP and/ or gel electrophoresis: GSTA1*B, GSTM1 and GSTT1 deletion, GSTP1*B, CYP2B6*2, *3, *4, *5 and *6, CYP2C9*2, *3 and *4, CYP2C19*2, *3, CYP3A4*1B, CYP3A5*3, *6 and ALDH1A1*2 and ALDH3A1*2. Polymorphism frequencies were associated with regimen related toxicities, other transplant related complications using Fischer's Exact test and Cox-proportional hazard's model.. Significant associations are shown in the Table. On univariate analysis, CYP2B6*4 variant genotype was associated with incidence of hemorrhagic cystitis (HC); CYP2C9*3 variant genotype was associated with the severity of HC; CYP2C19*3 and 2C9*2 genotypes were associated with overall and even-free survival (OS and EFS) and CYP2C9*2 and CYP2C9*3 genotype was associated with transplant related mortality (TRM). Multivariate analyses performed adjusting for known clinical risk factors still showed these genotypes to be significantly associated with outcome parameters. Variant genotypes of polymorphisms that result in decreased metabolism of Cy are protective against regimen related toxicities while these polymorphisms were risk factors for EFS and OS in the present study. This is the first report on the influence of common GST, CYP and ALDH polymorphisms on outcome of HSCT in patients with thalassaemia major. Screening for these polymorphisms in patients with beta thalassaemia undergoing HSCT can help identify patients at higher risk of complications.Table:EndpointGenotypeRelative risk (95% CI)P- value HCCYP2B6*4 variant0.3 (0.13-0.889)0.028 HC grade 1 vs. HC grade 2-4CYP2C9*3 variant0.2 (0.073-0.962)0.043 TRM2C9*2 variant2.7 (1.08-6.77)0.034 2C9*3 variant2.3 (1.0-5.7)0.049 OS and EFS2C19*3 variant3.3 (1.2-9.3)0.018 2C9*2 variant1.3 (0.93-2.03)0.070 Disclosures: No relevant conflicts of interest to declare.

Download Full-text