ISOFORM DIFFERENCES IN SUBSTRATE RECOGNITION BY GLYCOGEN-SYNTHASE KINASE-3-ALPHA AND KINASE-3-BETA IN THE PHOSPHORYLATION OF PHOSPHATASE INHIBITOR-2

Phosphorylation of inhibitor 2, the regulatory subunit of the ATP-Mg-d ependent protein phosphatase, by glycogen synthase kinase 3 (GSK-3) ca uses activation of the phosphatase. Prior phosphorylation by casein ki nase II has been shown to enhance both phosphorylation and activation of the phosphatase by GSK-3 (DePaoli-Roach, A. A. (1984) J. Biol. Chem . 259, 12144-12152). Reported here is a comparison of the phosphorylat ion of inhibitor 2 by two defined isoforms of GSK-3, GSK-3alpha and GS K-3beta. GSK-3beta was a significantly better inhibitor 2 kinase than was GSK-3alpha. The V(max)/K(m) value for GSK-3beta was approximately 10-fold higher than that for GSK-3alpha. GSK-3beta phosphorylated inhi bitor 2 to a stoichiometry of approximately 1.0 mol of phosphate/mol o f inhibitor 2. The phosphorylation by GSK-3beta was determined to be e xclusively at Thr-72 on the basis of the inability of the enzyme to mo dify a mutant inhibitor 2 in which Thr-72 was changed to alanine. Prio r phosphorylation by casein kinase II promoted the action of GSK-3alph a in keeping with earlier reports using undefined GSK-3 preparations. Phosphorylation by GSK-3beta, in contrast, was unaffected by the previ ous action of casein kinase II. These results suggest that there can b e important differences in substrate recognition by different isoforms of the same protein kinase and may help explain why some reported GSK -3 substrates require prior phosphorylation whereas others do not. It is proposed that substrate recognition by GSK-3 involves multiple dete rminants, the relative importance of each depending on both the substr ate and the GSK-3 isoform in question.