The human epsilon-globin gene is transcribed in erythroid cells only during the embryonic stages of development. The developmental control of epsilon-globin gene is mediated by a transcriptional silencer characterized by cell transcription assays as the epsilon-globin gene silencer motif (epsilon GSM) located at -278 to -257 bp 5' to the cap site. Mutation in the GATA-1/YY1 sites within epsilon GSM resulted in expression of the human epsilon-globin gene in hAEC system as well as in transgenic mice. Recently, two other negative regulatory regions at - 3 kb (epsilon NRA) and at -1.7 kb (epsilon NRB) have been characterized in our laboratory; epsilon NRA contains two GATA motifs. To further understand the role of GATA transcription factors on epsilon-silencing, we have studied their effect on the epsilon-globin gene expression. We have transiently transfected embryonic erythroid K562 cells with increasing amounts of an expression vector for the human transcription factor GATA-1. We have observed that increasing amounts of GATA-1 transcripts have a negative effect on the level of expression of endogeneous epsilon-globin gene, while there is little or no effect in other globin genes, such as gamma or alpha globin or other genes containing functional GATA-binding sites in their promoter regions such as glycophorin C. We have also observed that the level of expression of transcription factor GATA-2 decreases as the expression of GATA-1 increases. By stable transfections, we have created K562 cell lines constitutively expressing high levels of transcription factors GATA-1 and confirmed the negative effect of increasing levels of GATA-1 on epsilon-globin gene expression. Interestingly, in K562 cells expressing high levels of GATA-2 transcription factor (through transient or stable transfections), increasing levels of endogenous epsilon-globin and gamma-globin genes are observed. RNAse protection assays are in progress to better characterize the effect of GATA-2 transcription factor on globin gene expression. These findings suggest that GATA-1 and -2 transcription factors might mediate transcriptional regulation of the epsilon-globin gene through a concentration/dependent mechanism. These should provide an interesting model to understand how quantitative changes in transcription factors may result in qualitative alteration in target gene expression.
