Decreased Expression of HOXB4 Gene in Aplastic Anemia; Regulatory Interaction with GATA-2,

Aplastic anemia (AA) is characterized by a reduced number of hematopoietic stem cells (HSCs). It has been proposed that immunological injury in HSCs leads to reduced numbers of stem cells in the bone marrow. In addition, expression of the critical regulator of hematopoiesis GATA-2 is decreased in CD34-positive cells in AA (Fujimaki et al. Br J Haematol 2001). Despite the compelling results described above, only limited information has emerged regarding intrinsic abnormalities of hematopoietic stem cells in AA. It has been demonstrated that HOXB4 induces HSC expansion ex vivo (Antonchuk et al. Cell 2002), and restoring HOXB4 protein in HSCs from bone marrow failure patients promotes HSC expansion (Tang et al. Br J Haematol 2009). In conjunction with the evidence that recent genome-wide analysis of GATA factor chromatin occupancy identified GATA-2 peak at HOXB4 promoter (Fujiwara et al. Mol Cell 2009), we hypothesized that GATA-2 directly regulates HOXB4 expression in HSCs, which might contribute to the pathogenesis of AA. Here, we investigated possible link between GATA-2 and HOXB4, and also tested if HOXB4 is deregulated in CD34-positive cells from patients with AA.

For GATA-2 overexpression, human GATA-2 coding sequence was cloned into pcDNA3.1 expression vector as well as MSCV retroviral expression vector (Clontech). For GATA-2 knockdown, siRNA specific for human GATA-2 was transfected into CD34-positive cells or K562 cells by Amaxa Nucleofector kit (Amaxa Inc.). For promoter assay, DNA fragment of the HOXB4 gene promoter region (up to −262 from 1^st ATG) was cloned into pGL3-Basic (Promega), and the GATA deletion construct was subsequently created with QuickChange™Site-Directed Mutagenesis Kit (Stratagene). Quantitative chromatin immunoprecipitation (ChIP) analysis was performed using antibodies for GATA-2 (H-116, Santa Cruz). For analyzing clinical samples, informed consent was obtained in all cases and ethical considerations according to the declaration of Helsinki were followed.

To examine if GATA-2 and HOXB4 are functionally linked, we transfected a GATA-2 expression vector into K562 cells, and demonstrated that GATA-2 significantly upregulated endogeneous HOXB4 expression. Furthermore, siRNA-mediated GATA-2 knockdown in K562 cells significantly reduced HOXB4 expression, indicating that HOXB4 is a GATA-2 target gene. We overexpressed/reduced GATA-2 in cord blood-derived CD34+ cells, which also provided evidence for GATA-2 regulation of HOXB4 expression. Promoter analyses revealed that GATA sequence located at −160/-157 of the HOXB4 gene promoter region was required to confer luciferase activity in K562 cells. In vitro DNA binding studies and quantitative ChIP analysis revealed specific GATA-2 occupancy at a chromatin region containing this element. Finally, we demonstrated that HOXB4 gene expression was significantly decreased in CD34+ cells from patients with AA (n=10) compared to those with ITP (n=13). The expression levels of HOXB4 and GATA-2 also correlated in these populations (r=0.6573, p<0.01).

Based on these findings, we propose that decreased expression of GATA-2 in hematopoietic stem cells of AA leads to reduced HOXB4 transcription, which may have an important role in the development and/or progression of the disease.

No relevant conflicts of interest to declare.