Intracellular retention of mutant gonadotropin receptors results in loss of hormone binding activity of the follitropin receptor but not of the lutropin/choriogonadotropin receptor.

It has recently been reported that Asp 397 of the rat lutropin/ choriogonadotropin receptor (rLHR) may be involved in transducing the signal from hormone binding to the stimulation of cAMP production. We examined the analogous region in the rat follitropin receptor (rFSHR) by substituting the Asp at position 404 (D404) of the rFSHR with either Glu (D404E), Ala (D404A), or Lys (D404K). Both in intact 293 cells and in detergent-solubilized extracts of 293 cells transiently transfected with the rFSHR constructs, only the wild type rFSHR exhibited detectable binding activity. Although the D404-substituted rFSHR mutants were visible on Western blots, in contrast to the wild type rFSHR which is present on Western blots as both mature and immature forms, only a single band comigrating with immature receptor was observed for the mutants. Furthermore, these mutants were sensitive to endoglycosidase H (Endo H), thus indicating that the mutant receptor proteins were retained intracellularly in the endoplasmic reticulum. To test whether the lack of binding of the D404-substituted rFSHR mutants was due to a perturbation of a binding site or to the intracellular retention of the mutants, a truncated rFSHR(t637) mutant, containing a cytoplasmic truncation that should not directly affect FSH binding, was examined. As with the D404-substitution mutants, rFSHR(t637) was stably expressed but sensitive to Endo H. Significantly, detergent-soluble extracts of cells expressing rFSHR(t637) were unable to bind FSH. From these results, we conclude that substitution of D404 of the rFSHR prevents hormone binding as a result of the intracellular retention of the mutants in the endoplasmic reticulum presumably in an incompletely folded state, as opposed to disruption of a hormone-binding site at D404. Comparable rLHR substitution (D397K) and truncation (t616) mutants were constructed and used to transfect 293 cells. For both rLHR(D397K) and rLHR(t616), human CG (hCG) binding to intact cells was not detectable, but high affinity hCG binding was observed in detergent-soluble extracts of the cells. Therefore, the rLHR differs from the rFSHR in that mutants of the rLHR that are retained in the endoplasmic reticulum have already been folded correctly and can bind hCG with high affinity as long as a hormone-binding site has not been perturbed by the mutation. In contrast, mutants of the rFSHR that are retained in the endoplasmic reticulum have not yet folded into a conformation that can bind hormone. This suggests a difference in the temporal pattern of folding between the two structurally related gonadotropin receptors. Our studies also demonstrate how mutagenesis studies of the rFSHR must be interpreted with caution, as FSHR mutants that are expressed but are retained intracellularly will most likely not be able to bind FSH even when a hormone-binding site has not been altered.