Fetal Hemoglobin Expression Is Differentially Affected by Inhibition of the Proposed Dred Complex Constituents, LSD1 and DNMT1

• Published in: Blood Vol. 120; no. 21; p. 3263
• Main Authors: Arvedson, Tara L, Tran, Lynn, Ross, Sandra L, Yoder, Sean, Hertz, Alexandra, Hale, Kari, Eschelbach, John, Dineen, Tom, Matyas, Stephanie, Hartley, Cynthia, Morgenstern, Kurt, Winters, Dwight, Cindy, Park, Molineux, Graham, Coxon, Angela
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 16-11-2012
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood.V120.21.3263.3263

Abstract 3263
Sickle cell disease and beta thalassemia are disorders caused by mutations in adult hemoglobin (HbA) or defects in HbA expression. A potential therapeutic solution is reactivation of fetal hemoglobin (HbF) expression. Although HbF, comprising two alpha and two gamma globin chains, is the primary form of hemoglobin expressed in utero, gamma globin expression is silenced in adults. One proposed mechanism of gamma globin silencing involves binding of the direct repeat erythroid definitive (DRED) repressor complex to sequences in the gamma globin promoter. The DRED complex is reported to include the orphan nuclear hormone receptors TR2 and TR4, lysine specific demethylase (LSD1) and DNA methyltransferase (DNMT1). As both LSD1 and DNMT1 are epigenetic modifiers, gamma globin repression is proposed to be mediated by LSD1- and DNMT1-induced epigenetic changes. To investigate the role of DNMT1 and LSD1 in HbF silencing, HbF expression was evaluated in an erythroid differentiation model where hematopoietic progenitor cells were treated with either DNMT1 or LSD1 small molecule inhibitors or siRNA.
Human hematopoietic progenitor cells from healthy donors were induced to become erythroid using a two step protocol including erythropoietin, SCF, IL-3 and hydrocortisone for days 1–7 and erythropoietin and SCF for days 8–14. Cultures were treated with a range of concentrations of either tranylcypromine or S2101 (LSD1 inhibitors) or 5-azacytidine (DNMT1 inhibitor) and compared to HbF-inducing, positive control small molecules pomalidomide and lenalidomide. Cultures were also treated with LSD1 siRNAs and compared to controls. The effect of treatment on gamma, beta and alpha globin transcription was determined by qRT-PCR. The effect of treatment on HbA and HbF levels was determined by ELISA, HPLC, flow cytometry and imaging. Differentiation was characterized by morphology and flow-based detection of CD34 and glycophorin. Effects on viability were characterized by ViCell and flow cytometry.
Treatment with a concentration range of 5-azacytidine increased the rate of red blood cell differentiation as measured by daily changes in CD34 and glycophorin and hemoglobinization. Quantitative ELISA demonstrated that HbF expression increased two-fold. In contrast, LSD1 inhibition reduced both the rate of proliferation and differentiation of erythroid progenitors. Consistent with impaired differentiation, both beta globin transcription and HbA expression were reduced by up to 84% (qRT-PCR) and 65% (quantitative ELISA), respectively. No increase in gamma globin transcription or HbF expression was observed in response to LSD1 inhibition. Control cultures differentiated as expected: after 14 days of treatment the majority of vehicle-, lenalidomide- or pomalidomide-treated cells were glycophorin-positive and enucleation was readily apparent. Both lenalidomide and pomalidomide treatment induced a two-fold increase in HbF expression, as previously reported.
Although both LSD1 and DNMT1 are reported to be components of the DRED complex and are proposed to be jointly responsible for epigenetically modifying the gamma globin promoter to silence HbF expression, inhibition of the two proteins had different outcomes on HbF expression. DNMT1 inhibition upregulated HbF expression to a similar extent as pomalidomide (currently in Phase 1 clinical trials for HbF induction), whereas LSD1 inhibition impaired erythroid differentiation and hemoglobinization. These results suggest that the mechanism of gamma globin silencing and the proposed role of the DRED complex require further evaluation. Furthermore, this work also suggests that LSD1 inhibition is not a therapeutic strategy for HbF induction in patients with sickle cell disease or beta thalassemia.
Arvedson:Amgen: Employment. Tran:Amgen: Employment. Ross:Amgen: Employment. Yoder:Amgen: Employment. Hertz:Amgen: Employment. Hale:Amgen: Employment. Eschelbach:Amgen: Employment. Dineen:Amgen: Employment. Matyas:Amgen: Employment. Hartley:Amgen: Employment. Morgenstern:Amgen: Employment. Winters:Amgen: Employment. Cindy:Amgen: Employment. Molineux:Amgen: Employment. Coxon:Amgen: Employment.