HUMAN FALLOPIAN-TUBE AS AN EXTRAOVARIAN SOURCE OF RELAXIN - MESSENGER-RIBONUCLEIC-ACID EXPRESSION AND CELLULAR-LOCALIZATION OF IMMUNOREACTIVE PROTEIN AND I-125 RELAXIN BINDING-SITES

By means of specific human relaxin primers that originated from relaxi n A and B chains, a monoclonal antibody, and I-125-relaxin, the expres sion of mRNA and immunoreactive protein and the presence of binding si tes for relaxin were investigated in human fallopian tubes. Reverse tr anscription polymerase chain reaction (RT-PCR) analysis of total RNA i solated from tubal tissues revealed the predicted 434-bp fragments ori ginating from both the H1 and H2 relaxin genes. Restriction enzyme dig estion of the RT-PCR products with Msp I (Hpa II), present only in the relaxin H1 sequence, resulted in the anticipated 175- and 259-bp frag ments, whereas digestion with Hpa I, which is present in the relaxin H 2 sequence and should have resulted in 188- and 246-bp fragments, indu ced a limited and partial digestion of the product. Codigestion of the RT-PCR product with Msp I + Hpa I also resulted in 175- and 259-bp fr agments. The immunoreactive relaxin protein was present primarily in t he tubal epithelial cells of the ampullary and isthmus regions, and wi th weaker intensity in tubal smooth muscle cells. Immunoreactive relax in either was barely present or was absent in other tubal cell types. The intensity of immunostaining for relaxin in the epithelial cells ap peared not to be cycle-dependent; however, these cells showed a lower immunostaining at the late secretory phase of the menstrual cycle than at other reproductive stages and an absence of immunostaining during the postmenopausal period. Quantitative autoradiography of I-125-relax in binding (net grain density/100 mu m(2)), calculated for different c ell types, indicated that tubal smooth muscle and stromal cells had a significantly higher grain density than epithelial cells (p < 0.05). T he results provide the first evidence that the human fallopian tube, l ike other nonovarian tissue, expresses relaxin H1 mRNA and also contai ns immunoreactive relaxin protein and specific relaxin binding sites. The data suggest that relaxin, through the presence of its specific re ceptor, may act as an autocrine/paracrine regulator of various tubal b iological activities, and that tubal epithelial-derived relaxin may be an additional source of immunoreactive relaxin previously reported to be present in peritoneal fluid.