The intercysteine "loop" sequence 38-57 in the beta subunit has been shown to be a determinant for expression of biological activity in human lutropin (hLH) and choriogonadotropin (hCG) [Keutmann, H. T., Charlesworth, M. C., Mason, K. A., Ostrea, T., Johnson, L., & Ryan, R. J. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 2038]. Together with other sequences, the 38-57 region may contribute to a multicomponent receptor binding domain in hLH/hCG. Because the structural features influencing activity in this important region are not easy to evaluate in the full-length subunit, we have used analogues of hLH beta-(38-57) prepared by solid-phase synthesis. The peptides were tested for inhibition of 125I-labeled hCG binding to rat ovarian membrane receptors. Secondary structure was analyzed by circular dichroism (CD) and by reactivity with antibodies to the native 38-57 peptide. An analogue lacking the 38-57 disulfide linkage retained 20% receptor binding and full immunoreactivity. "Far"-ultraviolet CD profiles were essentially identical with those of the disulfide-intact peptide; a transition from 10% to 30% alpha-helix in 90% trifluoroethanol was characteristic of both. The peptide thus appears not to require the disulfide bridge to retain a looped conformation with amphipathic secondary structure. An essential positive charge at position 43 was shown by complete loss of activity upon substitution of Asp or Ala for the Arg found in all known species of LH. Other analogues showed a requirement for a neutral residue at position 47, also highly conserved.(ABSTRACT TRUNCATED AT 250 WORDS)

Human chorionic gonadotropin (hCG) and its subunits were studied as substrates for cAMP-dependent protein kinase. Phosphorylated residues were identified in peptide fragments by sequence analysis following appropriate hydrolysis and purification. Only the free beta-subunit was phosphorylated. Intact beta subunit incorporated 1 mol of phosphate/mol of peptide. This was localized to Thr 97, within the sequence -Arg-Arg-Ser-Thr-, which resembles one type of acceptor site for cAMP-dependent phosphorylation. Since beta Thr 97 did not phosphorylate in native hCG, this residue may be masked by the alpha subunit. Reduced and carboxymethylated hCG beta phosphorylated to 2.8 mol of phosphate/mol of peptide. Besides Thr 97, phosphate was also found at Ser 66 and Ser 96. Ser 66 occurs within a segment recognized to be favored for phoshorylation with the general sequence -Arg-X-X-Arg-X-X-Ser-. The appearance of this site in the linearized peptide suggests that Ser 66 is "buried" in the native conformation of the beta subunit. Two synthetic peptides representing residues 93-102 of hCG beta were also examined. The disulfide form of the peptide phosphorylated at Thr 97 while the linear peptide phosphorylated at Ser 96. Conformation as well as primary structure can thus influence the site of phosphate incorporation.

The beta subunit of human choriogonadotropin (hCG) was previously shown to be phosphorylated by beef skeletal muscle cAMP-dependent protein kinase (Keutmann, H. T., Ratanabanangkoon, K., Pierce, M. W., Kitzmann, K., and Ryan, R. J. (1983) J. Biol. Chem. 258, 14522-14527). The phosphorylation site was primarily at Thr 97, located within the "determinant loop" region proposed by Ward and Moore (Ward, D. N., and Moore, W. T. (1979) Animal Models for Research in Fertility and Contraception (Alexander, N. J., ed) pp. 151-164, Harper and Row, Baltimore, MD) to be important for hormonal activity and specificity. Biological and immunological studies were carried out to determine the effect of this modification on the activity of hCG. The phosphorylated hCG beta recombined with hCG alpha to form phosphorylated hCG (*hCG). The recombined *hCG retained full immunological activity in a radioimmunoassay using anti-hCG serum and 125I-hCG. The biological activities of the *hCG on appropriate rat gonadal cells, as studied by hCG radioreceptor assay, follitropin radioreceptor assay, and adenylate cyclase assay, were 0.29 +/- 0.04, 0.29 +/- 0.07, and 0.69 +/- 0.13 times as potent as hCG, respectively. The reduced receptor binding activity of *hCG was not due to dissociation of the hormone into its subunits. The far ultraviolet CD spectrum of the *hCG showed no gross conformational change induced by phosphorylation. We conclude the following: 1) Thr 97 is not involved in subunit-subunit interaction and is not part of the hCG antigenic site recognized by this particular antiserum; 2) Thr 97 does appear to participate in the hCG-receptor interactions; 3) modification of Thr 97 does not result in an enhancement of follitropin-like activity.