Activating transcription factor-4 (ATF4) is a member of the ATF/CREB subfamily of basic region-leucine zipper proteins. ATF4 forms homodimers or heterodimers by interacting with itself or other proteins through its leucine zipper motif. Potential partners of ATF4 include members of the AP-1 and C/EBP families, c-maf, and Nfe2-related factors Nrf1 and Nrf2. Previous studies have shown that ATF4 is required for lens formation in mice. Masuoka and Townes (page 736) now report that ATF4 has a role in definitive hematopoietic development. ATF4-deficient embryos are severely anemic. The fetal livers from these embryos contain fewer hematopoietic progenitors, and the colonies contain fewer cells, than controls. Whereas the lens defect is due, at least in part, to p53-dependent apoptosis, no increase in apoptosis was noted in ATF4-deficient fetal livers. Thus, the fetal anemia appears to be due to a defect in erythroid proliferation. Furthermore, ATF4-deficient mice are approximately half the size of littermate controls and have delayed hair growth. Fibroblasts from ATF4-deficient embryos have a prolonged doubling time. These results suggest that ATF4 deficiency impairs cell growth in a variety of tissues.

ATF4 joins a growing list of genes, whose disruption causes transient fetal anemia. These genes, which include “flexed”, E2f4, and Stat5a/Stat5b,are not essential for erythropoiesis. Rather, they appear to be on ancillary pathways that support erythropoiesis at specific stages of development or under conditions of stress. Additional studies are needed to determine the exact role of ATF4 in hematopoietic regulation. It is interesting that some of the potential dimerization partners of ATF4 may also have a role in fetal hematopoiesis. Nrf1 deficiency is associated with fetal anemia, while c-maf deficiency is associated with fetal anemia and a defect in lens development. The similarity of these phenotypes to the ATF4 phenotype raises the possibility of a functionally relevant interaction between ATF4 and Nrf1 or c-maf.