TSH and immunoglobulin G (IgG) preparations from patients with Graves' disease increase inositol phosphate as well as cAMP formation in Cos-7 cells transfected with rat TSH receptor (TSHR) cDNA. In a previous report, we mutated alanine 623 of the third cytoplasmic loop (residues 605-625) of the TSHR and showed it was critical for TSH and Graves' IgG initiation of phosphatidylinositol bisphosphate (PIP2) but not cAMP signaling. In this report, we substituted residues in the third loop of the TSHR with sequences from the N- and C-termini of the third loop of the alpha1- and beta2-adrenergic receptors (ARs), which computer analysis has identified as homologous to those in the TSHR. Alanine 623 is conserved in most ARs as well as in glycoprotein hormone receptors; there is, therefore, no change in alanine 623. After transfection of the mutant TSHR cDNAs into Cos-7 cells, we show that the mutant proteins are normally synthesized, processed, and incorporated into the membrane bilayer by Western blotting with a specific receptor antibody. We also show that the dissociation constant for TSH binding in all mutants is the same or lower than wild type TSHR. We then evaluated the ability of TSH or Graves' IgG to increase PIP2 and cAMP signals in each transfectant. Mutants A622 and B621 replace, respectively, residues 622-625 and 621-625 of the TSHR with alpha1- and beta2-AR residues from the C-terminus of the third cytoplasmic loop; mutants A607 and 8605 replace, respectively, TSHR residues 607-609 and 605-609 with N-terminus residues from alpha1- and beta2-AR. All four mutants, like the alanine 623 mutant, result in transfected cells which lose TSH and Graves' IgG initiation Of PIP2 but not cAMP signalling. Like the alanine 623 mutation to glutamic acid, the A607, B605, A622, and B621 mutants also result in decreased basal cAMP, but not inositol phosphate levels, relative to wild type receptor. In contrast to these results, mutants A610, B610, A617, and B617, which replace residues 610-613 or 617-620 of the TSHR with corresponding residues of the alpha1- and beta2-AR, retain TSH and Graves' IgG responsiveness in both inositol phosphate and cAMP assays. Mutation of residues 610-613, in fact, potentiates TSH-increased inositol phosphate production, despite having no effect on TSH-increased cAMP production. Mutation of residues 617-620 markedly elevates basal cAMP levels by comparison to wild type TSHR or pSG5 transfectants and causes a parallel, but smaller, increase in basal inositol phosphate production. This is the first description of a constitutively activated glycoprotein hormone receptor. In sum, the data indicate that the N- and C-terminal five residues of the third cytoplasmic loop of the TSHR, not only alanine 623, are important for TSH and Graves' IgG induction of the PIP2, but not their induction of the cAMP signal. In addition, they are important for the regulation of constitutive cAMP levels, but have only minimal effects on constitutive inositol phosphate levels. Substitution of native residues 617-620 appears to allow receptor activation of the cAMP signaling pathway in the absence of agonist; substitution of native residues 610-613 potentiates agonist-induced PIP2 but not cAMP signaling. Sequence homologies suggest these observations are applicable to gonadotropin receptors as well as all species of TSHR.