S. Ivar Walaas

Dopaminergic regulation of protein phosphorylation in the striatum: DARPP-32

Abstract DARPP-32 is a neuronal phosphoprotein that is specifically enriched in neurons possessing D 1 dopamine receptors, including the medium-sized spiny neurons of the striatum. DARPP-32 phosphorylation is regulated by dopamine acting through cyclic AMP. Biochemical studies have shown that phosphorylated DARPP-32 functions as a potent inhibitor of protein phosphatase-1 in vitro. In vivo , this inhibition may be an important component of the biochemical mechanisms by which dopamine, acting via D 1 receptors, exerts its neurophysiological effects. Regulation of DARPP-32 phosphorylation may also mediate specific interactions between dopamine, acting through cyclic AMP, and glutamate (or other first messengers), acting through Ca 2+ . Future studies of basal ganglion-specific phosphoproteins in general, and of DARPP-32 in particular, should lead to a clearer understanding of the molecular mechanisms underlying dopaminergic regulation of neuronal function.

Absence of synapsin I and II is accompanied by decreases in vesicular transport of specific neurotransmitters.

Studies of synapsin‐deficient mice have shown decreases in the number of synaptic vesicles but knowledge about the consequences of this decrease, and which classes of vesicles are being affected, has been lacking. In this study, glutamatergic, GABAergic and dopaminergic transport has been analysed in animals where the genes encoding synapsin I and II were inactivated. The levels of the vesicular glutamate transporter (VGLUT) 1, VGLUT2 and the vesicular GABA transporter (VGAT) were decreased by approximately 40% in adult forebrain from mice devoid of synapsin I and II, while vesicular monoamine transporter (VMAT) 2 and VGLUT3 were present in unchanged amounts compared with wild‐type mice. Functional studies on synaptic vesicles showed that the vesicular uptake of glutamate and GABA was decreased by 41 and 23%, respectively, while uptake of dopamine was unaffected by the lack of synapsin I and II. Double‐labelling studies showed that VGLUT1 and VGLUT2 colocalized fully with synapsin I and/or II in the hippocampus and neostriatum, respectively. VGAT showed partial colocalization, while VGLUT3 and VMAT2 did not colocalize with either synapsin I or II in the brain areas studied. In conclusion, distinct vesicular transporters show a variable degree of colocalization with synapsin proteins and, hence, distinct sensitivities to inactivation of the genes encoding synapsin I and II.

Multisite phosphorylation of microtubule-associated protein 2 (MAP-2) in rat brain: Peptide mapping distinguishes between cyclic AMP-, calcium/calmodulin-, and calcium/phospholipid-regulated phosphorylation mechanisms

Microtubule-associated protein 2 (MAP-2), a cytoskeletal protein of 280 kilodalton that is highly enriched in dendrites and neuronal perikarya, is subject to both cyclic AMP-, calcium/calmodulin-, and calcium/phospholipid-regulated phosphorylation when incubated with [γ-32P]ATP in vitro. We have analyzed the different sites in MAP-2 phosphorylated by these three kinases in fresh or boiled cytosol from different regions of the rat brain, in particular the olfactory bulb, where only one form (MAP-2B) is present, and the cerebral cortex, where both forms (MAP-2A and MAP-2B) are equally enriched. Cyclic AMP-dependent protein kinase and calcium/calmodulin-dependent protein kinase II phosphorylated four common phosphorylation sites, as well as a number of distinct sites that were unique to each enzyme. Calcium/phospholipid-dependent protein kinase phosphorylated a minimum of 15 sites, only one of which appeared to be shared with the other protein kinases. Only serine residues were phosphorylated by cyclic AMP-dependent and calcium/phospholipid-dependent protein kinases, while both serine and threonine residues were phosphorylated by calcium/calmodulin-dependent protein kinase II. No differences were observed in the peptide maps of phospho-MAP-2 prepared from different brain regions. These results emphasize the complexity of the phosphorylation systems that may regulate the function of MAP-2 in situ.

DARPP‐32 and Phosphatase Inhibitor‐1, Two Structurally Related Inhibitors of Protein Phosphatase‐1, Are Both Present in Striatonigral Neurons

Abstract: DARPP‐32 (dopamine‐ and cyclic AMP‐regulated phosphoprotein of Mr= 32,000) and phosphatase inhibitor‐1, two previously characterized inhibitors of protein phosphatase‐1, were identified in both the neostriatum and the substantia nigra. Phosphatase inhibitor‐1 was partially purified from bovine caudate nucleus and found to be distinct from DARPP‐32 in some of its biochemical properties. The neuronal localization of DARPP‐32 and phosphatase inhibitor‐1 within the rat neostriatum and substantia nigra was investigated by studying the effects of kainic acid. Injection into the neostriatum of kainic acid, which destroys striatonigral neurons and striatonigral fibers, decreased the amounts of DARPP‐32 and phosphatase inhibitor‐1 to the same extent, both in the lesioned neostriatum and in the ipsilateral substantia nigra. The specific activity of protein phosphatase‐1 in the neostriatum was unaffected by kainic acid. The results indicate that, in rat brain, DARPP‐32 and phosphatase inhibitor‐1 are both present in striatal neurons and in striatonigral fibers, and that they probably coexist in at least a subpopulation of striatonigral neurons. In contrast, protein phosphatase‐1 does not appear to be enriched in any specific neuronal subpopulation in the neostriatum.

Synapsin Ia, Synapsin Ib, Protein IIIa, and Protein IIIb, Four Related Synaptic Vesicle-Associated Phosphoproteins, Share Regional and Cellular Localization in Rat Brain

Abstract: The regional and cellular distribution of four synaptic vesicle‐associated proteins, synapsins Ia and Ib (Mr 86,000 and 80,000, collectively referred to as synapsin I) and proteins IIIa and IIIb (Mr 74,000 and 55,000, collectively referred to as protein III), has been compared in selected rat brain regions, using both radioimmunoassays and back‐phosphorylation assays. Lesions of several neuronal populations in the basal ganglia (corticostriatal fibers, intrinsic striatal neurons, striatonigral fibers, nigrostriatal fibers) induced decreases in the levels of these various proteins that were highly correlated (r= 0.96–0.97). Moreover, the synaptic vesicle‐associated phosphoproteins displayed a similar and widespread distribution throughout the CNS. This evidence for colocalization indicates that the majority of, and possibly all, CNS neurons and nerve terminals may contain both forms of synapsin I and both forms of protein III.

Cell-specific localization of the α-subunit of calcium/calmodulin-dependent protein kinase II in Purkinje cells in rodent cerebellum

Brain calcium/calmodulin-dependent protein kinase type II, a multimeric 600-650 kDa enzyme composed of alpha- (50 kDa) and beta/beta (60 and 58 kDa) subunits, may be formed by alpha- and beta-subunits combining in variable proportions in different types of neurons. This study presents evidence, using cerebella from mutant mice, that the alpha-subunit displays a restricted localization in the rodent cerebellum, being detectable only in Purkinje cells. Immunocytochemical analysis of normal rat cerebellum with an antibody selective for the alpha-subunit confirmed that this subunit was detectable only in Purkinje cells. In contrast, the beta/beta-subunits appeared to be present in all types of cerebellar mutants examined. These results indicate that different cells of the cerebellum express distinct isozymic forms of the multifunctional calcium/calmodulin-dependent protein kinase type II. It appears that Purkinje cells primarily contain an isoenzyme formed by both alpha- and beta/beta-subunits, and that non-Purkinje cells contain an isoenzyme formed primarily by beta/beta-subunits.

Phosphorylation of P1, a high mobility group‐like protein, catalyzed by casein kinase II, protein kinase C, cyclic AMP‐dependent protein kinase and calcium/calmodulin‐dependent protein kinase II

P1, a high mobility group‐like nuclear protein, phosphorylated by casein kinase II on multiple sites in situ, has been found to be phosphorylated in vitro by protein kinase C, cyclic AMP‐dependent protein kinase and calcium/calmodulin‐dependent protein kinase II on multiple and mostly distinct thennolytic peptides. All these enzymes phosphorylated predominantly serine residues, with casein kinase II and protein kinase C also labeling threonine residues. Both casein kinase II and second messenger‐regulated protein kinases, particularly protein kinase C, might therefore be involved in the physiological regulation of multisite phosphorylation of PI.

Phosphorylation of multiple sites in a 15 000 dalton proteolipid from rat skeletal muscle sarcolemma, catalyzed by adenosine 3′,5′-monophosphate-dependent and calcium / phospholipid-dependent protein kinases

This study reports a partial characterization of a 15,000 dalton (15 kDa) proteolipid present in rat skeletal muscle sarcolemma. The proteolipid is phosphorylated by both cyclic AMP-dependent and calcium/phospholipid-dependent protein kinases, displays an isoelectric point (pI) of 5.9, and can be extracted from sarcolemma by acidified chloroform/methanol (2:1) or non-ionic detergents. Phosphoamino acid analysis and tryptic fingerprinting of the phosphorylated proteolipid indicate that both cyclic AMP- and calcium/phospholipid-dependent protein kinases predominantly phosphorylate serine residue(s) on a single tryptic peptide. Additivity experiments and thermolytic fingerprinting demonstrate a minimum of two distinct phosphorylation sites on the proteolipid, the phosphorylation of which is independently catalyzed by cyclic AMP-dependent and calcium/phospholipid-dependent protein kinases in vitro. This sarcolemma proteolipid, which appears to be identified to a sarcolemma protein previously reported to be phosphorylated upon addition of insulin in a GTP-dependent manner (Walaas, O., Walaas, E., Rye-Alertsen, A. and Horn, R.S. (1979) Mol. Cell. Endocrinol. 16, 45-55), therefore represents a possible membrane target for those neuronal and hormonal stimuli which can regulate cyclic AMP-dependent or calcium/phospholipid-dependent protein kinase activities in skeletal muscle.

Widespread distribution of the c-src gene product in nerve cells and axon terminals in the adult rat brain

The regional and cellular distribution of the proto-oncogene product pp60c-src, a member of the family of membrane-associated tyrosine-specific protein kinases, was analysed in adult rat brain. High-resolution SDS-PAGE allowed analysis of both the fibroblast 60-kDa form and a variant, 61-kDa neuron-specific form of the c-src gene product which is encoded by an alternately processed c-src mRNA. Studies of microdissected brain regions showed that all CNS regions contained both forms of the enzyme, the 61-kDa form predominating in most regions with high content of gray matter and high density of synapses. Lesion-induced degenerations of specific neuronal elements in the basal ganglia decreased the level of both forms of the c-src gene product both in regions where cell bodies had been lesioned and in regions where nerve terminals had degenerated. The 61-kDa form of the enzyme appeared somewhat more sensitive to the effects caused by these lesions than the 60-kDa form. These results indicate that, within the mature mammalian brain, both cell body regions and nerve terminals of many, and possibly all, nerve cells contain both forms of the c-src gene product, the 61-kDa form being most highly enriched in the nerve cells. These results suggest that the enzyme may be involved in pleiotropic functions, including signal transduction in nerve terminals.

Skeletal muscle sarcolemma proteins as targets for adenosine 3′:5′-monophosphate-dependent and calcium-dependent protein kinases

The present study documents the existence in rat skeletal muscle plasma membrane (sarcolemma) of a distinct set of proteins, most of which represent unknown protein species, which can be phosphorylated in vitro by addition of cAMP-dependent or calcium-dependent protein kinases. Under the experimental conditions used, cAMP-regulated protein phosphorylation appeared to be the most important phosphorylation system in these membranes, followed by the calcium/phospholipid-regulated, and, with only a few substrates detected, the calcium/calmodulin-regulated systems. No specific substrate for cGMP-dependent protein kinase was found. In contrast, calcium/calmodulin-regulated protein phosphorylation was the most important in the sarcoplasmic reticulum fraction. Most of the cAMP-regulated and calcium/phospholipid-regulated sarcolemma phosphoproteins appeared to be intrinsic membrane proteins, at least three of which appeared to be phosphorylated by both these protein kinases. These phosphoproteins may represent membrane targets for multiple hormone or transmitter actions in skeletal muscle cells. Our results, therefore, suggest that protein phosphorylation systems, particularly those regulated by cAMP or calcium/phospholipid, may be more important in the regulation of sarcolemma function than hitherto believed.