Abstract 2065

The fetal and adult hemoglobin genes located in the human β-like globin cluster on chromosome 11, undergo sequential activation and silencing during normal erythroid maturation. Pharmacologic agents which induce fetal hemoglobin are effective treatments for sickle cell disease; however mechanisms of γ-globin induction are not fully understood. To gain insights into globin gene regulation, we utilized a primary cell culture system to define transcriptome profiles associated with γ-globin silencing, β-globin activation and the γ to β globin switch. Erythroid progenitors were generated from peripheral blood mononuclear cells grown the entire period in the presence of stem cell factor (50ng/ml), Interleukin-3 (10ng/ml) and erythropoietin (4IU/ml) using a one-phase liquid culture system. We observed 90% γ-globin gene silencing from day 7 to 28 with simultaneous β-globin activation; moreover, the γ to β globin switch completed by day 21. As erythroid maturation progressed, an increased abundance of early (basophilic erythroblast) followed by late (orthochromatophilic erythroblast) erythroid progenitors was observed from day 16 to 31; we also observed a change in the Gγ/Aγ globin ratio from 3:1 to 1:1 by day 28. These data confirm that the one-phase system is a good cell model to study global gene profiling during in vitro erythropoiesis. To achieve this goal we performed microarray analysis on the Illumina Human v2.0 BeadChip using RNA isolated at days 7, 14, 21 and 28 of erythroid maturation. The data were normalized by model-based background correction using Bioconductor (R2.10.0 language processing) and supervised learning on the BRB-array platform. The erythroid transcriptome was analyzed in three datasets based on expression profiles corresponding to γ-globin silencing (profile-1), β-globin activation (profile-2) and the g to β globin switch (profile-3). Using the David Functional Annotation Bioinformatics platform, we identified 1,028 genes with functional annotation consistent with high expression at day 7 that silenced by day 28 (profile-1). Included in this analysis were genes with known biochemical and biological function and known information about regulation and expression patterns. Functional pathway analysis using IPA software demonstrated possible involvement of p38 MAPK (5.75), NF-kB (3.08), Interleukin 4 (2.18) and ERK/MAPK (2.06) signaling in γ-globin regulation; scores >1.3 supported involvement of that pathway in globin gene regulation. Profile-1 trans-factors were cross-referenced with putative binding sites in the β-like cluster locus control region, HBG1/HBG2 gene coding and 5’-upstream non-coding regions, by in silico Transcription Element Search Software (TESS) and TFSEARCH analysis. Twenty factors including CREB1, ATF2, GATA2, CEBPA, FOXQ1, and KLF4 were identified as possible γ-globin regulators. Similar studies were conducted for genes with increased expression from day 7 to 28 (profile-2), to identify proteins involved in β-globin activation. We identified 483 genes with functional annotation for the clathrin (2.6) and Insulin-like growth factor 1 pathway (1.56) related to cell processing and metabolism. In the same vein, TESS and TFSEARCH studies were extended to the HBB gene and 5’-upstream non-coding region. Among the 12 trans-factors related to profile-2, EKLF, GATA1, JUND, ATF1, and SP1 were identified. Finally, we completed the analysis of profile-3 genes with peak expression at day 21, when the γ to β switch occurs; 1,297 genes contain function annotation in cell cycle regulation (1.56), retinoid × receptor (1.57), and the ataxia telangiectasia mutated protein (3.16) among others. Collectively, these results demonstrated that our high-throughput transcriptome analysis established a system to define global pathways involved in globin gene expression. Specific proteins will be investigated using functional and applied genomics techniques to define mechanisms of γ-globin and β-globin gene regulation.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution