Regulation of the human relaxin genes H1 and H2 by steroid hormones.

Relaxin, a peptide hormone important to the outcome of human pregnancy is expressed in a tissue specific manner as two genes known as relaxins H1 and H2, in addition to a third human relaxin H3, expressed primarily in the brain. The H1 and H2 genes are highly homologous, differentially expressed in reproductive tissues and appear to activate the same receptor, but their regulation is poorly understood. Based upon the known physiology of these hormones and the response elements in their 5'- and 3'-flanking regions, the possibility that progesterone and/or the glucocorticoids might influence their differential expression was therefore investigated. The changes in the mRNA levels of the relaxin genes in response to either medroxyprogesterone acetate (MPA) or dexamethasone (Dex) were analyzed by RT-PCR using a choriocarcinoma cell line (JAR) as a model system, because the expression of these genes in any primary human cell type is too low for such a study. The addition of 0.5 microM MPA to JAR cells, significantly upregulated the mRNA of only the relaxin H2, while the addition of 0.5 microM Dex significantly upregulated the mRNAs for both the relaxins, after 6h of treatment. Promoter assays indicated an early activation of transcription (1 h), which by 6 h had decreased. Progesterone and/or glucocorticoids could exert their effects via the GRE motif found on the 5'-flanking region of the relaxin genes. The H1-GRE differs from the H2-GRE by a single nucleotide, which may affect H1-GRE binding to the progesterone receptor (PR) but not the glucocorticoid receptor (GR). The antiprogestin RU486 inhibited the binding of the GR to both H1-GRE and H2-GRE, while it enhanced the binding of the PR to these GREs. As determined by gel shift assays, this GRE motif could bind to both the PR and GR and was therefore considered to be functional. Thus, both progesterone and glucocorticoids are capable of differentially regulating the expression of the two human relaxin genes in a model system. Copyright 2004 Elsevier Ireland Ltd.