Calcium-dependent phosphorylation of glial fibrillary acidic protein (GFAP) in the rat hippocampus: a comparison of the kinase/phosphatase balance in immature and mature slices using tryptic phosphopeptide mapping

Abstract

In previous work we showed that phosphorylation of the astrocyte marker glial fibrillary acidic protein (GFAP) in hippocampal slices from adult rats is dependent on external Ca 2+, whereas in slices from immature rats aged 12–16 days postnatal 32P incorporation into GFAP is inhibited by external Ca 2+. The nature of this late developmental change in Ca 2+ sensitivity for GFAP phosphorylation was investigated in the present work by comparing in immature and adult animals phosphorylation of GFAP by endogenous protein kinase activity in cytoskeletal fractions and tryptic phosphopeptide maps prepared from cytoskeletal fractions labelled with [γ- 32P]ATP and from slices labelled with [ 32P]phosphate. Cytoskeletal fractions prepared from immature and adult hippocampus both contained endogenous protein kinase activity towards GFAP and other proteins stimulated by Ca 2+/calmodulin and by cyclic AMP. The maps of GFAP isolated from the cytoskeletal fractions labelled in the presence of Ca 2+/calmodulin were very similar and exhibited two major and several minor phosphopeptides. Comparison with maps derived from these fractions labelled in the presence of cyclic AMP showed that one of the major phosphopeptides was either directly or indirectly phosphorylated by Ca 2+/calmodulin-stimulated kinase activity. Maps derived from GFAP isolated from adult slices labelled in the presence of Ca 2+ and immature slices labelled in the absence of Ca 2+ were qualitatively identical, with minor differences from the cytoskeletal maps. At both ages the slice maps displayed the phosphopeptide phosphorylated through the activity of a Ca 2+/calmodulin kinase in the cytoskeletal fractions. By its migration properties this peptide appears to correspond to a sequence containing a site shown by other workers to be phosphorylated in vitro by CaM kinase II, suggesting that even in the absence of external Ca 2+, kinase activity directly or indirectly dependent on Ca 2+ was occurring in the immature slices. The near identity of the phosphorylation sites at the two ages suggest that the change in Ca 2+ sensitivity of GFAP phosphorylation during development is not due to a change in the balance of kinase and phosphatase activities, but rather to a change in the mechanism(s) whereby Ca 2+ controls the relative activity of these enzymes.