Serum profiles of growth hormone (GW) were recorded for 24 h in women at different stages of normal pregnancy. Two monoclo~ial antibodies directcd against different epitopes and unaffected by human placental lactogen werc used in radioimniunoassays to distinguish the pituitary 22K-GH from the placental GH variant. The 'normal' episodic peak activity of GH in nun-pregnant and first trimester pregnant womcn was dramatically changed into a continuous very stable secretion during late pregnancy. This change was first observed at 17 weeks gestation. It is concluded that during the second half of pregnancy, serum rneasurcinents of GH reflect a major contribution frotn a non-cpisodically secreted placental GH variant and a concomitant suppression of pituitary GH. This specific signal, i.e. a continuous GH secretion, may be an important regulator of maternal liver metabolism during pregnancy.Growth hormone ((311) is a major mctabolic rcgulatory hormone which stimulates protein synthesis and has profound effects on carbohydrates and lipid metabolism (Merimee 1979). In recent studies on the physiology of growth hormone in pregnant women (Frankcnne rt al. 1988). we have shown that maternal serum levels of the pituitary 22K hGH, encoded by the GH-N

Several libraries of monoclonal antibodies have been produced against epitopes that reside on hCG, alpha hCG, and beta hCG. Having characterized them physically, we explored their use in the construction of highly specific and sensitive immunoradiometric assays. There were several important immunochemical considerations with respect to developing assays that accurately detect low levels of free subunits in serum in the presence of high concentrations of the native hormone. These include physical properties and specificities of the monoclonal antibodies, choice of capture antibody on the solid phase support, assay design, and purity of hormone standards. Using such assays, we found early pregnancy (in vitro fertilization) to be characterized by the sequential appearance of hCG, followed by beta hCG and then alpha hCG. Molar ratios of beta hCG to alpha hCG and beta hCG to hCG were highest in early gestation. However, there was a reversal of the beta hCG to alpha hCG ratio at 12-13 weeks gestation, and an excess of free alpha hCG was observed thereafter. Except for values obtained in very early pregnancy, the beta hCG to hCG ratio remained remarkably constant at approximately 0.5% throughout gestation. In contrast, choriocarcinoma was distinguished by absolute serum beta hCG concentrations 3-100 times greater than the maximum values observed during pregnancy and, more importantly, by exceedingly high beta hCG to hCG ratios. For comparison, we studied hCG, alpha hCG, and beta hCG levels in an additional 178 patients with nontrophoblastic tumors. Ectopic production of alpha hCG and beta hCG was rare (3%), and thus far, we have been unable to demonstrate the presence of hCG in such patients. Therefore, hCG and the free subunits appear not to be useful as serological markers for nontrophoblastic tumors.