Replication Factor C3 Is a Direct Target Of CREB, Promotes G1/S Transition Of Acute Myeloid Leukemia Cells, and Increases Hematopoietic Stem/Progenitor Cell Self-Renewal

CREB (cAMP Response Element Binding protein) promotes cellular transformation of hematopoietic cells and proliferation of myeloid leukemia cells. However, the underlying mechanisms of CREB function in leukemic transformation and hematopoiesis are not fully understood. To address this, we have investigated the downstream pathways of CREB activation in proliferation using a human acute myeloid leukemia (AML) cell line KG1 cells knocked-down for CREB with specific shRNAs. The CREB-knockdown KG1 cells were significantly defective in proliferative capability compared to control cells [cell number after 4d (X10⁵), seeding (1X10⁵), control vs. CREB-knockdown: 34.18 +/– 1.27 vs. 14.52 +/– 0.46, n=3, p< 0.01, mean +/– SEM]. In order to characterize the specific role of CREB in cell proliferation, we analyzed cell cycle progression patterns of CREB-knockdown and control KG1 cells after release from mitotic arrest. Our results indicated that G1 to S phase transition as assessed by % S phase was impeded by CREB-knockdown [S phase (%), control vs. CREB-knockdown cells, 8h after release: 53.29 +/– 0.54 vs. 23.57 +/– 1.69; 12h: 66.92 +/– 0.63 vs. 45.16 +/– 0.50, n=3, p< 0.01, mean +/– SEM]. To identify potential CREB target genes, we chose several cell cycle related genes such as CCNE1, CCNA1, CCNB1 and PCNA and compared their RNA expression levels in the CREB-knockdown with those in control KG1 cells after release from mitotic arrest. To our surprise, we failed to detect any noticeable differences in the mRNA expression levels of those genes between CREB-knockdown and control KG1 cells. In an effort to search for CREB responsive target genes, we analyzed additional CREB targets previously identified from microarray data (Pellegrini et al BMC Cancer 2008). We found that expression of replication factor C3 (RFC3), a 38kDa subunit of the RFC complex involved in DNA replication and repair processes, was significantly reduced in CREB-knockdown cells compared to control cells [38 +/– 1% of control, n=3, p<0.01]. CREB-knockdown also inhibited RFC3 mRNA expression in U937 and HL60 AML cell lines. Consistent with these results, mRNA expression levels of RFC3 appeared to be closely correlated with those of CREB when we examined bone marrow samples obtained from AML patients [n = 16, Pearson coefficient = 0.6366, p = 0.0008]. Moreover, we found that CREB directly interacted with the CRE site in the RFC3 promoter region in vivo, as assessed by chromatin immunoprecipitation assays. Exogenous overexpression of RFC3 in CREB-knockdown KG1 cells restored the defective G1/S progression [S phase (%), CREB-knockdown vs. CREB-knockdown with RFC3 overexpression, 9h after release: 38.97 +/– 0.45 vs. 62.24 +/– 1.06; 12h: 48.12 +/– 0.60 vs. 67.70 +/– 1.15, n=3, p< 0.01, mean +/– SEM]. Taken together, these results suggest that RFC3 may act as a novel downstream oncogenic target of activated CREB in AML cells.

We previously reported that CREB is a critical regulator of normal myelopoiesis (Cheng et al Blood 2008). To determine whether RFC3 could exert similar effects on normal hematopoiesis, we compared human umbilical cord blood derived CD34-positive cells with and without RFC3 overexpression for the capacity to form hematopoietic colonies. Overexpression of RFC3 in the CD34-positive cells resulted in significant increases of multi-potential CFU-GEMM colony numbers [without vs. with overexpression of RFC3 (per 1000 cells): 3.2 +/– 1.3 vs. 22.3 +/– 3.3, n=3, p< 0.01, mean +/– SEM]. The RFC3 effect on stimulating colony formation was magnified in secondary colony forming assays [without vs. with overexpression of RFC3 (per 100,000 cells): 10.7 +/– 3.5 vs. 180.2 +/– 44.4, n=3, p< 0.05, mean +/– SEM]. Since the formation of secondary colonies was derived mainly from residual stem/progenitor cell populations after long-term culture, RFC3 overexpression may enhance self-renewal of stem/progenitor cells. In conclusion, our results suggest that RFC3 is able to promote G1/S transition in a human AML cell line downstream of CREB activation. In addition, we provide evidence that RFC3 is involved in normal hematopoiesis and contributes to increased self-renewal potential of hematopoietic stem/progenitor cells. Our data demonstrate that RFC3 plays multiple roles in promoting AML cells proliferation as well as normal myelopoiesis through increasing the self-renewal potential of hematopoietic stem/progenitor cells in response to CREB activation.