Vol. 289, Issue 3, 1323-1333, June 1999

Measurement of Agonist and Antagonist Ligand-Binding Parameters at the Human Parathyroid Hormone Type 1 Receptor: Evaluation of Receptor States and Modulation by Guanine Nucleotide¹

Sam R. J. Hoare, Gerda de Vries² and Ted B. Usdin

Unit on Cell Biology, Laboratory of Genetics, National Institute of Mental Health, (S.R.J.H., T.B.U.) and Mathematical Research Branch, National Institute of Diabetes and Digestive and Kidney Diseases (G.dV.), National Institutes of Health, Bethesda, Maryland

Determination of ligand-binding constants for parathyroid hormone (PTH) receptors has been hampered by a lack of suitable experimental systems and mechanistic models for data analysis. In this study, ligand binding to the cloned human PTH-1 receptor was measured using membrane-based radioligand-binding assays. Guanosine 5'-O-(3-thiotriphosphate) (GTPS) (10 µM) reduced binding of agonist radioligands [¹²⁵I]rPTH(1-34) and [¹²⁵I]PTHrP(1-36) but only to a limited extent (by 29 ± 5 and 42 ± 3%, respectively). Radiolabeled agonist dissociation was described by three and two phases in the absence and presence of GTPS, respectively. GTPS treatment removed a pseudoirreversible binding phase. Inhibition of radiolabeled antagonist ([¹²⁵I]bPTH(3-34)) binding was measured using a 90-min incubation, which allowed binding of ligands to closely approach the asymptotic maximum. Agonist/[¹²⁵I]bPTH(3-34) displacement curves were fitted best by assuming two independent affinity states, both in the presence and absence of GTPS. After a 3-h incubation, binding of PTH agonists in the presence of GTPS was described by a single affinity state, indicating the presence of slow components in the binding reaction. Antagonist binding was described by a single affinity state and was not significantly affected by GTPS. The data were used to evaluate potential receptor-binding models. Although other models could not be excluded, all of the observations could be explained by assuming two binding sites on the receptor that recognize two corresponding sites on agonist ligands. Using the model, it was possible to estimate receptor-ligand-binding constants and to propose a direct method for identifying ligands that interact with a putative antagonist binding region of the receptor.