Insulin Receptor Substrate 1 Binds Two Novel Splice Variants of the Regulatory Subunit of Phosphatidylinositol 3-Kinase in Muscle and Brain

Abstract

We have identified two novel alternatively spliced forms of the p85a regulatory subunit of phosphatidylinositol (PI) 3-kinase by expression screening of a human skeletal muscle library with phosphorylated baculovirus-produced human insulin receptor substrate 1. One form is identical to p85a throughout the region which encodes both Src homology 2 (SH2) domains and the inter-SH2 domain/p110 binding region but diverges in sequence from p85a on the 5′ side of nucleotide 953, where the entire break point cluster gene and SH3 regions are replaced by a unique 34-amino-acid N terminus. This form has an estimated molecular mass of ~53 kDa and has been termed p85/AS53. The second form is identical to p85 and p85/AS53 except for a 24-nucleotide insert between the SH2 domains that results in a replacement of aspartic acid 605 with nine amino acids, adding two potential serine phosphorylation sites in the vicinity of the known serine autophosphorylation site (Ser-608). Northern (RNA) analyses reveal a wide tissue distribution of p85a, whereas p85/AS53 is dominant in skeletal muscle and brain, and the insert isoforms are restricted to cardiac muscle and skeletal muscle. Western blot (immunoblot) analyses using an anti-p85 polyclonal antibody and a specific anti-p85/AS53 antibody confirmed the tissue distribution of p85/AS53 protein and indicate a ~7-fold higher expression of p85/AS53 protein than of p85 in skeletal muscle. Both p85 and p85/AS53 bind to p110 in coprecipitation experiments, but p85a itself appears to have preferential binding to insulin receptor substrate 1 following insulin stimulation. These data indicate that the gene for the p85a regulatory subunit of PI 3-kinase can undergo tissue-specific alternative splicing. Two novel splice variants of the regulatory subunit of PI 3-kinase are present in skeletal muscle, cardiac muscle, and brain; these variants may have important functional differences in activity and may play a role in tissue-specific signals such as insulin-stimulated glucose transport or control of neurotransmitter secretion or action.