Regulation of the human protein S gene promoter by liver enriched transcription factors.

Protein S is expressed in a number of tissue types, one of the most physiologically relevant being the liver. However, transcriptional control of protein S gene expression is poorly understood. We have characterised a 638 bp area in the 5' flanking region of the human protein S gene, spanning all 10 previously reported transcription initiation sites, which demonstrates promoter activity in the human liver-derived cell line HepG2. More refined reporter gene analysis of this region enabled the identification of three transcription initiation sites whose absence is associated with significantly reduced promoter activity, together with a number of positively and negatively acting transcriptional regulatory elements. Consistent with these findings, DNaseI footprinting analysis identified eleven sites (I-XI) from within this 638 bp region that show evidence of binding nuclear proteins. We present evidence to show that the liver-specific factors hepatocyte nuclear factor 1 (HNF1) and HNF4 bind regions of the protein S promoter, which lie within the identified protein binding sites V and VIII, respectively, and that HNF4 activates the protein S promoter. Reporter gene analysis suggests that members of the CCAAT/enhancer binding protein (C/EBP) family of transcription factors are potent activators of protein S gene transcription in HepG2 cells.

Oestrogenic repression of human coagulation factor VII expression mediated through an oestrogen response element sequence motif in the promoter region.

Reporter gene analysis of two regions of the human factor VII (FVII) gene promoter (residues -658 to -1 and -348 to -1, where +1 is the start site of translation) in the mammalian liver-derived cell line HepG2 showed reduced transcriptional activity in the presence of oestrogenic factors. This effect was independent of promoter polymorphic haplotype. Similar analysis using a smaller region of the promoter spanning residues -187 to -1 failed to show any evidence of oestrogenic suppression. Electrophoretic mobility shift assays and supershift assays using recombinant oestrogen receptor alpha and anti-oestrogen receptor antibody localized the sequence motif to which oestrogen receptor was binding to residues -225 to -212 of the FVII promoter. The lack of oestrogenic suppression in a reporter gene construct spanning residues -658 to -1 modified to abolish oestrogen receptor binding at this site, confirmed the functional significance of this motif. Although superficially similar to the classical oestrogen response element (ORE), comprising two half sites separated by three spacer nucleotides, the FVII ORE represents an alternative type of ORE in which the two half sites are separated by just two spacer nucleotides. EMSAs indicated that increasing spacer nucleotide number from two to three in the FVII ORE, or decreasing it from three to two in a consensus ORE sequence motif, had a small effect on the binding affinity for oestrogen receptor. These data correlate with and provide a plausible mechanism for the inverse relationship between FVII and oestradiol levels observed during the menstrual cycle.