Glycosylation sites encoded by exon 2 of the human insulin receptor gene are not required for the oligomerization, ligand binding, or kinase activity of the insulin receptor.

Abstract

Asparagine-linked glycosylation of the insulin receptor is required for complete biosynthesis and acquisition of function. However, the relative role of each individual glycosylation site has not been elucidated. Previously, it has been shown that removal, by site-directed mutagenesis, of the four amino terminal glycosylation sites (N16,N25,N78, and N111) results in a mutant insulin receptor that remained in the endoplasmic reticulum as an unprocessed proreceptor (Collier E., Carpentier J.-L., Beitz L., Caro L. H. P., Taylor S. I., and Gorden P. [1993] Biochemistry 32, 7818-7823). In the present study, the contribution of these independent glycosylation sites to dimerization and insulin binding has been evaluated. Chinese hamster ovary cells were transfected with the wild-type human insulin receptor cDNA, or cDNA that had Q substituted for N at one, two, or all four of these glycosylation sites. Electrophoretic characterization of the proteins immunoprecipitated from 35S-labeled cells showed that both the wild-type and the quadruple mutant receptor had similar profiles, indicating that the mutant receptor is capable of undergoing dimerization. Analysis of the biochemical properties of this mutant showed that this receptor binds insulin, but ligand binding does not result in kinase stimulation. We demonstrated that the absence of kinase activation is not a property of the mutated receptor since the wild-type proreceptor behaves in a similar manner. Only partial glycosylation in this region of the receptor is required for its targeting to the cell membrane since single and double glycosylation mutants were found processed to their alpha and beta subunits on the cell surface.(ABSTRACT TRUNCATED AT 250 WORDS)