Control of yeast glycogen synthase-2 by COOH-terminal phosphorylation.

The budding yeast Saccharomyces cerevisiae expresses two isoforms of glycogen synthase, of which glycogen synthase-2 (GS-2) appears to be the most important determinant of glycogen accumulation (Farkas, I., Hardy, T. A., Goebl, M. G., and Roach, P. J. (1991) J. Biol. Chem. 266, 15602-15607). Partial proteolysis of purified yeast glycogen synthase activated the enzyme, mimicking the effects of dephosphorylation. The cleavage was localized to the COOH terminus of the molecule and trypsin treatment released 32P from enzyme labeled in vivo with 32P or in vitro by cyclic AMP-dependent protein kinase. Similarly, when cells were labeled with 32P, no radioactivity was incorporated into a mutant form of GS-2 truncated at residue 643 while the wild type enzyme was phosphorylated at both Ser and Thr residues. The 9 Ser and Thr residues COOH-terminal to position 643 were mutated individually to Ala, and the GS-2 mutants were expressed from a low copy plasmid in yeast that lacked functional chromosomal copies of the two glycogen synthase genes. Mutations at Ser-650, Ser-654, and Thr-667 resulted in significant activation of yeast glycogen synthase and elevation in the level of accumulated glycogen as compared with wild type. Likewise, expression of the truncated GS-2 resulted in hyperactive enzyme and the overaccumulation of glycogen. None of the other Ser or Thr mutations substantially affected glycogen synthase activity and glycogen storage. We conclude that Ser-650, Ser-654, and Thr-667 are regulatory phosphorylation sites in vivo. However, in vitro, cyclic AMP-dependent protein kinase modified Ser residue(s) COOH-terminal to position 659, and so the identity of the physiological GS-2 kinases is unclear. Yeast strains bearing glc7 and gac1 mutations are defective in genes encoding type 1 protein phosphatase components and are impaired in their ability to accumulate glycogen. Expression of the truncated GS-2 in these strains restored glycogen accumulation, as did the presence of GS-2 mutated at Ser-650, Ser-654, or Thr-667. These data are consistent with the hypothesis that type 1 phosphatase regulates GS-2 by controlling its phosphorylation state.