Dexamethasone Targets E2F4 to Induce Erythroid Progenitor Renewal

Corticosteroids such as dexamethasone (Dex) are the first line of treatment to modulate erythroid output in patients with Diamond Blackfan Anemia (DBA), an inherited bone marrow failure syndrome characterized by ribosomal protein (RP) insufficiency. Dexamethasone is also used to enhance the yield of erythrocytes generated in vitro from adult CD34+ cells. While previous studies have addressed steroid resistance in DBA patients, the mechanism of steroid responsiveness (SR) in erythroid progenitors (EP) remains unresolved. Despite identifying targets such as p57^kip2, there is still limited information on other involved pathways. Corticosteroid treatment is associated with side effects, including growth defects, hypertension, osteoporosis, and increased risk of infection. Therefore, there is a need to better understand the mechanisms of corticosteroid action on EPs to identify new molecules with fewer side effects and to assist non-responsive patients.

In EP1 to EP4 progenitor subpopulations FACS-sorted from bone marrow (BM) of healthy individuals, we observed an overrepresentation of cell-cycle related proteins and notably the transcription factor E2F4. E2F4 is well known as a transcriptional repressor of the cell cycle at the G1/S transition. However, recent studies have demonstrated a pro-proliferative role for E2F4 in fast-cycling cells, including murine erythroid precursor cells. Thus, we hypothesized that E2F4 could be a target of dexamethasone in erythroid progenitor cells. To test this hypothesis, we confirmed that the increase in E2F4 expression levels observed in the primary bone marrow EPs was recapitulated in EPs derived from cultured CD34+ cells obtained from the peripheral blood (PB) of healthy donors. We then examined E2F4 expression in response to dexamethasone in EPs derived from different donors and followed by E2F4 knockout validations using CRISPR/Cas9 technology.

In EPs derived from healthy PB-derived CD34+ cultured for 4 days in the presence or absence of dexamethasone, we observed E2F4 expression was maximal in the EP4 population in the absence of Dex, but maximal E2F4 expression shifted to the EP3 population in the presence of Dex. These results confirm our previous studies that Dex acts at the immature CFU-E level (redefined as EP3) and identify E2F4 as a potential Dex target specifically acting on EP3, leading to the expansion of EP3 and EP4 progenitor populations.

Based on the increased expression of E2F4 in EP3 and its role as transcription factor, we investigated the differential cellular localization of E2F4 between cytoplasm and nucleus to address the hypothesis that Dex treatment would increase nuclear expression. Dex treatment resulted in increased expression of E2F4 in nuclear fraction compared to the cytoplasmic fraction. Consequently, we performed ChIP-seq analyses on sorted EPs obtained from PB-CD34+ cultured cells and FACS sorting. In EP3 cells, over 90% of E2F4 peaks were localized to gene promoters with the highest enriched motif matching the known E2F4 motif in the HOCOMOCO database. Even without treatment, E2F4 peaks were present at the CDKN3 and CCNE1 gene promoters. CCNE1's are cyclin kinases involved in cell cycle specifically in the EP3 population. Other gene promoters with E2F4 peaks were involved in DNA replication, chromosome segregation, and cell cycle regulation. These data confirm the hypothesis that E2F4 could regulate the cell cycle and be responsible for erythropoiesis recovery in steroid-responsive (SR) patients with DBA, specifically in EP3 cells.

To validate the hypothesis that dexamethasone exerts its EP recovery effect through E2F4, we used CRISPR/Cas9 editing technology to knockdown E2F4. We used three different sgRNAs, resulting in varying levels of knockdown. Using a serum-free culture system, we differentiated the edited CD34+ cells over 11 days with either sgRNA-AAVS1 or sgRNA-E2F4. In cultures with an 80-90% knockdown of E2F4, we observed a twofold decrease in cell numbers. Additionally, we noted accelerated terminal differentiation by day 14. Importantly, while Dex led to an increase in the cell numbers in control cells, this effect was inhibited in the E2F4 knockdown cells. Together, these results suggest that E2F4 is necessary in the response to steroids.

In summary, these findings offer new insights into the response of erythroid progenitors to dexamethasone and identify potential new targets for treatment.

No relevant conflicts of interest to declare.