Elevated EGR2 Expression Provides a Potential Link Between Cell Cycle Arrest and Induced Differentiation in Myeloid Progenitor Cells

Similar to many immature cell types, myeloid progenitor cells need to exit cell cycle to undergo terminal differentiation, but the mechanism linking the two is still unclear. Elucidating this mechanism could lead to the development of new differentiation therapies to treat myeloid leukemia. Recent studies have suggested that the processes regulating myeloid differentiation and cell cycle progression together constitute a positive feedback loop where each process reciprocally affects the other. To study the relationship between these processes, we examined early cellular and molecular events associated with induced differentiation of the HL60 human promyelocytic leukemia cells. We treated HL60 cells with 3 classical inducers of differentiation (vitamin D3 analog EB1089 (EB), all-trans retinoic acid (ATRA), and dimethyl sulfoxide (DMSO)), along with PD0332991 (PD), a selective cyclin D-dependent kinase 4/6 inhibitor that caused G1-phase-specific cell-cycle arrest. We evaluated differentiation of the treated cells by flow cytometric analysis of CD11b (integrin α_M) and CD71 (transferrin receptor) expression. In untreated HL60 cells, a baseline subset of 3-5% of cells exhibits a differentiated, CD11b+CD71- phenotype. Exposure to the various inducers revealed a progressive increase in the percentage of CD11b+CD71- cells with time, such that by day 4 of treatment, it has increased to 50-90% in the treated samples, indicating that all 4 agents tested were effective in inducing myeloid differentiation. To understand how differentiation induced by each agent affects cell cycle progression, the cell cycle status of the induced cells were evaluated by a BrdU-incorporation assay after a 30-minute pulse of BrdU labeling. Uninduced cells exhibited a baseline cell cycle phase distribution of 64%-28%-8% (G1-S-G2/M phases). After 1 day of induction, EB-treated sample showed no changes in the distribution (58%-33%-9%), but ATRA, DMSO and PD-treated samples showed significant changes, with an increase of cell numbers in G1 phase and decrease in S phase (74%-18%-8%, 79%-13%-8%, and 93%-4%-3%, respectively). These results reveal that an early induction of G1 arrest was caused by treatment with ATRA, DMSO and PD, but not EB, and that the cell cycle arrest occurred before major changes in the myeloid phenotype were observed. To determine how the cell cycle perturbation relates to changes in the underlying genetic regulatory network, we examined by quantitative RT-PCR analysis the expression of several transcription factors associated with myeloid differentiation. PU.1 and CEBPA were found to be expressed at high levels but these levels did not change upon treatment with the inducing agents. Similarly, the expression levels of GFI1 and EGR1 did not change significantly with induction. In contrast, the expression level of EGR2 (Early Growth Response 2) was found to be low initially but became elevated upon treatment with 3 of the 4 inducers. EGR2 is a zinc finger transcription factor implicated in the control of a switch between pro- and anti-proliferation pathways. EGR2 has been shown to regulate the transition between differentiation states of Schwann cells, induction of anergic and regulatory T cells, growth and survival of osteoclasts, and proliferation and apoptosis of acute myeloid leukemia blasts. We found that EGR2 expression, after 1 day of treatment with ATRA, DMSO or PD, was increased by 5.2 ± 0.9, 7.6 ± 1.9, 5.8 ± 0.9 folds, respectively, whereas treatment with EB led to no significant change (1.5 ± 0.2 fold). We evaluated whether simultaneous treatment of the cells with 2 inducers would result in an additive effect. Treatment of HL60 cells with a combination of ATRA/DMSO, ATRA/PD, or DMSO/PD increased the percentage of CD11b+CD71- cells to 55%, 70% and 25% after just 1 day of treatment. In line with the enhanced phenotypic effect, the expression level of EGR2 was further elevated to 7.7 ± 1.4, 15.4 ± 3.5, and 11.3 ± 3.4 folds, respectively, when the cells were treated with the above inducer combinations, indicating a tight association between EGR2 expression and the phenotypic effect. In summary, our data suggest that elevated expression of EGR2 is an early event in the induction of myeloid differentiation in HL60 cells. Because of its known role in cell cycle regulation, EGR2 could function as a mechanistic link between cell cycle arrest and induced differentiation in myeloid progenitor cells.