Alterations in heavy and light neurofilament proteins in hippocampus following chronic ECS administration.

Chronic administration of electroconvulsive seizures (ECS), one of the most effective treatments for depression, induces sprouting of the mossy fibers in the hippocampus. This sprouting requires chronic ECS administration and appears to occur in the absence of hilar neuronal loss. Dynamic regulation of cytoarchitecture plays a vital role in such profound alterations of neuronal morphology. In particular, alterations in the neurofilament protein subunits have been implicated in neurite sprouting, neuronal regeneration, and growth. The present study was carried out to determine the influence of chronic ECS administration on the neurofilament subunits and other molecular markers of neuronal plasticity. Chronic ECS administration decreases the level of phosphorylated heavy neurofilament subunit (NF-H). In addition, the total level of the light neurofilament subunit (NF-L) but not the medium neurofilament subunit (NF-M) is decreased following chronic ECS treatment. Other cytoskeletal proteins, including actin, microtubule-associated protein (MAP-2), and tau, are not influenced by chronic ECS administration. Expression of the growth-associated protein (F1/GAP-43) also remains unchanged following chronic ECS treatment. The changes observed in neurofilaments may be part of the cytoskeletal remodeling that contributes to the mossy fiber sprouting induced by chronic ECS treatment.

Region-specific regulation of RGS4 (Regulator of G-protein-signaling protein type 4) in brain by stress and glucocorticoids: in vivo and in vitro studies.

The present study demonstrates that the regulator of G-protein-signaling protein type 4 (RGS4) is differentially regulated in the locus coeruleus (LC) and the paraventricular nucleus (PVN) of the hypothalamus by chronic stress and glucocorticoid treatments. Acute or chronic administration of corticosterone to adult rats decreased RGS4 mRNA levels in the PVN but increased these levels in the LC. Similarly, chronic unpredictable stress decreased RGS4 mRNA levels in the PVN but had a strong trend to increase these levels in the LC. Chronic stress also decreased RGS4 mRNA levels in the pituitary. The molecular mechanisms of RGS4 mRNA regulation were further investigated in vitro in the LC-like CATH.a cell line and the neuroendocrine AtT20 cell line using the synthetic corticosterone analog dexamethasone. Consistent with the findings in vivo, dexamethasone treatment caused a dose- and time-dependent decrease in RGS4 mRNA levels in AtT20 cells but a dose- and time-dependent increase in CATH.a cells. RGS4 mRNA regulation seen in these two cell lines seems to be attributable, at least in part, to opposite changes in mRNA stability. The differential regulation of RGS4 expression in the LC and in key relays of the hypothalamic-pituitary-adrenal axis could contribute to the brain's region-specific and long-term adaptations to stress.

Chronic ethanol administration regulates the expression of GABAA receptor alpha 1 and alpha 5 subunits in the ventral tegmental area and hippocampus.

Ethanol dependence and tolerance involve perturbation of GABAergic neurotransmission. Previous studies have demonstrated that ethanol treatment regulates the function and expression of GABAA receptors throughout the CNS. Conceivably, changes in receptor function may be associated with alterations of subunit composition. In the present study, a comprehensive (1-12 weeks) ethanol treatment paradigm was used to evaluate changes in GABAA receptor subunit expression in several brain regions including the cerebellum, cerebral cortex, ventral tegmental area (VTA) (a region implicated in drug reward/dependence), and the hippocampus (a region involved in memory/cognition). Expression of alpha 1 and alpha 5 subunits was regulated by ethanol in a region-specific and time-dependent manner. Following 2-4 weeks of administration, cortical and cerebellar alpha 1 and alpha 5 subunits immunoreactivity was reduced. In the VTA, levels of alpha 1 subunit immunoreactivity were significantly decreased after 12 weeks but not 1-4 weeks of treatment. Hippocampal alpha 1 subunit immunoreactivity and mRNA content were also significantly reduced after 12 but not after 4 weeks of treatment. In contrast, alpha 5 mRNA content was increased in this brain region. These data indicate that chronic ethanol administration alters GABAA receptor subunit expression in the VTA and hippocampus, effects that may play a role in the abuse potential and detrimental cognitive effects of alcohol.

Electroconvulsive seizure increases the expression of CREM (cyclic AMP response element modulator) and ICER (inducible cyclic AMP early repressor) in rat brain.

Rapid expression of ICER (inducible cyclic AMP early repressor), an inducible member of the CREM (cyclic AMP response element modulator) family of transcription factors, has been reported in neuroendocrine tissues and cell lines, but not in brain. In the present study, we demonstrate that acute electro-convulsive seizure (ECS) increases the expression of ICER in several rat brain regions. RNase protection analysis demonstrated that 1-2 h after administration of ECS, levels of mRNA for ICER and a splice variant, ICER gamma, were significantly increased in hippocampus, frontal cortex, and cerebellum. It is surprising that ECS also increased levels of mRNA for several CREM isoforms that previous studies have reported were not rapidly inducible. In situ hybridization analysis confirmed these findings and demonstrated that ECS induction of ICER was most obvious in the dentate gyrus granule cell layer of hippocampus and deep layers of cerebral cortex. Induction of ICER and CREM was accompanied by increased expression of two small CRE-binding complexes. Gel supershift analysis with CREM/ICER antisera confirmed that the inducible CRE-binding complexes contain CREM/ICER. Induction of CREM and ICER may contribute to negative feedback regulation of gene transcription that is increased by acute seizure and activation of CREB (cyclic AMP response element-binding protein.

Biochemical adaptations in the mesolimbic dopamine system in response to heroin self-administration.

Previous studies have shown that chronic, forced exposure to opiates produces specific biochemical adaptations in the ventral tegmental area (VTA) and nucleus accumbens (NAc). The functional consequences of these adaptations have been hypothesized to contribute to certain motivational aspects of drug addiction. In this study, the possibility that similar adaptations could occur in response to intermittent heroin self-administration was tested by comparing homogenates of VTA and NAc from rats self-administering heroin, rats receiving yoked injections of heroin, and rats receiving yoked injections of saline (controls). Tyrosine hydroxylase (TH) immunoreactivity was increased (31-38%) in the VTA and decreased (11%) in the NAc of heroin-exposed rats relative to controls. Heroin exposure also increased cAMP-dependent protein kinase (PKA) activity in both particulate (19-27%) and soluble (17-20%) fractions of the NAc, and decreased (16-17%) the level of Gi alpha immunoreactivity in this brain region. In contrast, no significant biochemical changes were found in the substantia nigra or caudate-putamen, indicating a selective effect on the mesolimbic dopamine system. Overall, adaptations in the VTA and NAc of heroin-exposed rats were similar to, but generally smaller in magnitude than, adaptations produced by chronic morphine administration. However, in contrast to morphine-treated animals, heroin-exposed animals failed to display overt signs of opiate physical dependence, suggesting that adaptations in motivational systems may occur more readily than adaptations in brain regions associated with physical dependence.

Influence of neurotrophic factors on morphine- and cocaine-induced biochemical changes in the mesolimbic dopamine system.

Previous research has shown an increase in tyrosine hydroxylase in the ventral tegmental area following chronic morphine and chronic cocaine treatments. Chronic morphine treatment also increases levels of glial fibrillary acidic protein in this brain region. In the present study, we investigated the effects of infusing neurotropic factors (nerve growth factor, brain-derived neurotrophic factor, neurotrophin-3, neurotrophin-4 or ciliary neurotrophic factor) via midline intra-ventral tegmental area cannulae on these biochemical changes. Our studies examined the effects of neurotrophic factor infusion alone, neurotrophic factor infusion followed by morphine treatment, morphine treatment followed by neurotrophic factor infusion, and concurrent neurotrophic factor infusion and cocaine treatment. Brain-derived neurotrophic factor, which by itself tended to decrease tyrosine hydroxylase levels in the ventral tegmental area, prevented the characteristic increase in tyrosine hydroxylase following morphine and cocaine exposure and reversed the increase in rats pretreated with morphine. Neurotrophin-4 and neurotrophin-3 exerted similar effects. In addition, neurotrophin-4 prevented the morphine-induced increase in glial fibrillary acidic protein. In contrast, ciliary neurotrophic factor infusions alone resulted in an increase in tyrosine hydroxylase levels, with no additional increase induced by morphine or cocaine coadministration. Nerve growth factor alone had no effect on tyrosine hydroxylase or glial fibrillary acidic protein levels and did not affect morphine's ability to induce these proteins. We also looked at the effects of intra-ventral tegmental area infusion of neurotrophic factor on cAMP-dependent protein kinase and adenylyl cyclase activity in the nucleus accumbens, both of which are increased by chronic morphine or cocaine exposure. In general, regulation of cAMP-dependent protein kinase and adenylyl cyclase morphine by neurotrophic factors paralleled effects seen in the ventral tegmental area. Intra-ventral tegmental area infusion of brain-derived neurotrophic factor (or neurotrophin-4) alone tended to decrease cAMP-dependent protein kinase and adenylyl cyclase activity in the nucleus accumbens and prevented the morphine-induced increases in these enzymes. These effects were not seen with ciliary neurotrophic factor or nerve growth factor. These studies demonstrate novel interactions within the ventral tegmental area, and its target the nucleus accumbens, between neurotrophic factors and drugs of abuse, which have potentially important implications for the pathophysiology and treatment of drug addiction.

Regulation of endogenous ADP-ribosylation by acute and chronic lithium in rat brain.

In the present study, we investigated the effects of lithium on endogenous ADP-ribosylation in rat brain. It was found that addition of lithium in vitro inhibits endogenous ADP-ribosylation activity in extracts of frontal cortex at therapeutically relevant concentrations. Inhibition is observed at concentrations as low as 0.3 mM and is maximal at 1 mM when 50% inhibition is obtained. A similar degree of inhibition of endogenous ADP-ribosylation was observed for all substrate proteins identified, including Gs alpha, suggesting that lithium's effect may be achieved at the level of ADP-ribosyltransferases and not specific substrate proteins. In contrast to lithium, chloride salts of sodium and potassium do not alter endogenous ADP-ribosylation activity in frontal cortex. To assess the possible in vivo relevance of this in vitro action of lithium, we studied the effect of chronic lithium administration on levels of endogenous ADP-ribosylation in frontal cortex. It was found that chronic lithium treatment, in contrast to the inhibitory effect of the drug in vitro, produced a > 35% increase in endogenous ADP-ribosylation activity. A similar degree of increase was observed for all of the substrate proteins identified. These novel findings raise the possibility that certain endogenous ADP-ribosyltransferases are among the acute targets of lithium in the brain and that adaptations in these enzymes may be part of the mechanisms underlying lithium's long-term effects on brain function.

Extracellular signal-regulated protein kinases (ERKs) and ERK kinase (MEK) in brain: regional distribution and regulation by chronic morphine.

Quantitative blot immunolabeling techniques were used to determine the concentrations of ERK1 (M(r) 44 kDa) and ERK2 (M(r) 42 kDa), the two major extracellular signal-regulated protein kinases, in different regions of rat brain. The aggregate ERK concentrations (ERK1 and ERK2) were relatively high in each of the brain regions studied, ranging from approximately 0.35 ng/microgram protein in cerebellum to approximately 1.2 ng/microgram protein in nucleus accumbens. However, differences in the regional distributions of ERK1 and ERK2 resulted in ratios of their relative abundance that differed by close to 10-fold among the regions studied. The ratios of ERK1 protein to ERK2 protein varied along a rostral-caudal gradient from a low of 0.16 in frontal cortex to a high of 1.5 in pons/medulla. In hypotonic homogenates from regions at either extreme of the gradient, ERK1 and ERK2 were both found to be predominantly (> 80%) soluble. In subcellular fractions prepared from sucrose homogenates of frontal cortex and pons/medulla, both ERK1 and ERK2 were enriched in the synaptosomal and cytosolic fractions, whereas ERK2 was also enriched in the microsomal fraction. By contrast, in subfractions containing purified nuclei, levels of ERK1 and ERK2 were about one-third of those seen in homogenates and, in subfractions enriched in mitochondria, both ERK1 and ERK2 were barely detectable. The catalytic activity of the ERKs paralleled their protein levels in all of the brain regions except the hippocampus, in which the activity and phosphotyrosine content were disproportionately high. As a possible explanation for this apparent disparity, the regional distribution of ERK kinase (MEK), which phosphorylates and activates the ERKs, was also investigated. The levels of immunoreactivity of the M(r) 45 kDa ERK kinase band differed by about threefold among the brain regions, with the highest levels being present in nucleus accumbens, hippocampus, substantia nigra, and caudate/putamen. Therefore, a higher concentration of ERK kinase immunoreactivity did not appear to account for the disproportionate levels of ERK activity and phosphotyrosine content in the hippocampus. Potential regulation of ERK and ERK kinase levels was also investigated in rats subjected to chronic morphine treatment. ERK1 and ERK2 levels were increased selectively in locus coeruleus and caudate/putamen after chronic morphine treatment, whereas ERK kinase immunoreactivity remained unchanged in all of the brain regions analyzed. In summary, the regional differences in ERK and ERK kinase expression and the region-specific regulation of ERK expression suggest that ERK-related signaling may play an important role in CNS function and its adaptive responses.

Chronic morphine administration increases beta-adrenergic receptor kinase (beta ARK) levels in the rat locus coeruleus.

Based on the established role of beta-adrenergic receptor kinase (beta ARK) and beta-arrestin in the desensitization of several G protein-coupled receptors, we investigated the effect of chronic morphine administration on beta ARK and beta-arrestin levels in selected brain areas. Levels of beta ARK were measured by blot immunolabeling analysis using antibodies specific for two known forms of beta ARK, i.e., beta ARK1 and beta ARK2. It was found that chronic morphine treatment produced an approximately 35% increase in levels of beta ARK1 immunoreactivity in the locus coeruleus, but not in several other brain regions studied. In contrast, chronic morphine treatment failed to alter levels of beta ARK2 immunoreactivity in any of the brain regions studied. Levels of beta-arrestin immunoreactivity, measured using an antiserum that recognizes two major forms of this protein in brain, were also found to increase (by approximately 20%) in the locus coeruleus. It is proposed that chronic morphine regulation of beta ARK1 and beta-arrestin levels may contribute to opioid-receptor tolerance that is known to occur in this brain region.

Inactivation of Gi and G(o) proteins in nucleus accumbens reduces both cocaine and heroin reinforcement.

The pertussis toxin (PTX)-sensitive G proteins Gi and G(o) may be implicated in drug reinforcement and addiction, since certain reward-related dopamine and opiate receptor subtypes are coupled to these G proteins, and since chronic exposure to cocaine or morphine alters levels of these G proteins in the nucleus accumbens (NAc). As a direct test of this hypothesis, Gi and G(o) proteins in the NAc were selectively inactivated by intra-accumbens injections of PTX in rats self-administering either cocaine or heroin. In control animals, bilateral injections of inactive PTX (0.1 microgram/1 microliter/side) in the NAc failed to alter baseline rates of cocaine and heroin self-administration. In contrast, the same dose of active PTX produced significant, long-lasting increases (up to 1 month) in the self-administration of both drugs, and shifted the dose-response curves to the right. These results suggest that PTX reduces or shortens the reinforcing efficacy of cocaine and heroin, leading to compensatory increases in drug self-administration. Similar NAc injections of PTX reduced the level of Gi alpha and G(o) alpha subunits as measured by both ADP-ribosylation and Western blot, without affecting levels of Gs alpha or G beta subunits. The effect of the toxin was mainly limited to the NAc, and no evidence of abnormal cell death or gliosis was observed. The onset of changes in self-administration rate coincided with the onset of changes in ADP-ribosylation, suggesting that, initially, the increased drug self-administration results directly from a reduction in functional Gi and G(o) proteins. After 28 d, self-administration baselines began to recover while levels of G protein ADP-ribosylation and immunoreactivity remained low, suggesting that adaptive mechanisms are involved at later time points. These results provide direct support for a common role of Gi and G(o) proteins in the NAc in the reinforcing and addictive properties of psychostimulant and opiate drugs.

Alterations in nitric oxide-stimulated endogenous ADP-ribosylation associated with long-term potentiation in rat hippocampus.

The present study examines the possible involvement of nitric oxide (NO)-stimulated endogenous ADP-ribosylation in long-term potentiation (LTP). LTP was induced in hippocampal slices by stimulation of Schaffer collateral inputs to the CA1 pyramidal neurons. Basal and sodium nitroprusside (SNP), which generates NO, stimulation of endogenous ADP-ribosylation was then studied in CA1 subfields isolated from the slices. Control slices received no treatment or were tetanized in the presence of aminophosphonovaleric acid, an NMDA receptor antagonist that blocks the development of LTP. SNP-stimulated ADP-ribosylation of endogenous proteins was reduced by 40-70% in LTP slices relative to control slices. LTP was also associated with a small but significant reduction in basal ADP-ribosylation activity. The results demonstrate that the induction of LTP is associated with regulation of endogenous ADP-ribosylation and suggest a role for this type of covalent modification in some aspect of LTP.

Lewis and Fischer rat strains display differences in biochemical, electrophysiological and behavioral parameters: studies in the nucleus accumbens and locus coeruleus of drug naive and morphine-treated animals.

In previous studies, we demonstrated that tyrosine hydroxylase and neurofilament proteins are regulated by chronic morphine and chronic cocaine treatments in the ventral tegmental area in Sprague-Dawley rats and that the inbred Lewis and Fischer 344 rat strains, under drug-naive conditions, show different levels of these proteins specifically in this brain region. In the current study, we compared Lewis and Fischer rats with respect to levels of adenylate cyclase, cyclic AMP-dependent protein kinase and G-proteins in the nucleus accumbens (NAc) and locus coeruleus (LC), brain regions in Sprague-Dawley rats where these proteins are regulated by chronic exposure to morphine or to cocaine. We found that levels of adenylate cyclase and cyclic AMP-dependent protein kinase activity are higher in the NAc and LC of Lewis rats compared to Fischer rats, whereas levels of Gi alpha and G beta were lower. These strain differences were not seen in several other brain regions analyzed and no strain differences were detected in levels of other G-protein subunits. Lewis and Fischer rats also differed in the ability of chronic morphine to regulate adenylate cyclase and cyclic AMP-dependent protein kinase in the NAc and LC. In the NAc, chronic morphine increased levels of the two enzymes in the Fischer strain only, whereas in the LC chronic morphine increased levels of the enzymes in both strains, with more robust effects seen in the Lewis rat. To understand possible physiological consequences of these strain differences in the cyclic AMP pathway, we studied LC neuronal activity under basal and chronic morphine-treated conditions. LC neurons of Lewis rats showed higher spontaneous firing rates in brain slices in vitro than those of Fischer rats and also showed greater morphine-induced increases in responsiveness to bath-applied 8-bromo-cyclic AMP. These electrophysiological findings are generally consistent with the biochemical observations. Moreover, Lewis and Fischer rats displayed very different opiate withdrawal syndromes, with different types of behaviors elicited upon precipitation of opiate withdrawal with the opiate receptor antagonist, naltrexone. The possible relationship between these behavioral findings and the biochemical and electrophysiological data is discussed. These studies provide further support for the possibility that Lewis and Fischer rat strains provide a useful model system in which some of the genetic factors that contribute to drug-related behaviors can be investigated.

Characterization and functional expression of a somatostatin receptor coupled to adenylyl cyclase.

Receptor ligand binding and functional studies indicate that somatostatin (SRIF) receptors belong to the G protein coupled receptor superfamily. Using the polymerase chain reaction (PCR) and degenerate primers to the third and sixth transmembrane domains of G protein coupled receptors, we have isolated a SRIF receptor cDNA from bovine locus coeruleus (LC). This SRIF receptor (referred to as LCR9) encodes a protein of 368 amino acids and is 94% identical to a human SRIF receptor recently isolated from pancreatic islet cells. Levels of LCR9 mRNA are highest in cerebral cortex, intermediate in thalamus and cerebellum, and lower in pons, LC, dorsal raphe, and substantia nigra. Expression in CHO cells demonstrates that LCR9 binds (125) I-Tyr(1)-SRIF with high affinity (K(d) approximately 0.5 nM). SRIF, SRIF28, and MK 678 (a SRIF(1) selective ligand) potently displaced SRIF ligand binding, while a CGP-23996 like compound (compound 1), which displays higher affinity for SRIF2, did not significantly influence ligand binding at concentrations up to 1 muM. Ligand binding to the expressed receptor was also significantly reduced by GTP. Unlike the receptor isolated from pancreas, expression of LCR9 resulted in SRIF receptor inhibition of adenylyl cyclase. SRIF, SRIF28, and MK 678 all potently inhibited forskolin-stimulated adenylyl cyclase in membranes from LCR9 transfected cells; in contrast the CGP-23996 like compound did not influence adenylyl cyclase activity. SRIF receptor inhibition of adenylyl cyclase was dependent on GTP and was sensitive to pretreatment of membranes with pertussis toxin. The results demonstrate that LCR9 encodes a high-affinity SRIF receptor that is negatively coupled to adenylyl cyclase.

Individual differences in locomotor activity are associated with levels of tyrosine hydroxylase and neurofilament proteins in the ventral tegmental area of sprague-dawley rats.

We have demonstrated previously that chronic morphine and cocaine treatments increase levels of tyrosine hydroxylase (TH), and decrease levels of neurofilament (NF) proteins, in the ventral tegmental area (VTA), a major dopaminergic brain reward region, of outbred Sprague-Dawley rats. We have also found inherent differences in levels of these proteins in the VTA of inbred rat strains that differ in their behavioral responses to opiates, cocaine, and other drugs of abuse, with the Lewis rat showing higher levels of TH and lower levels of NFs in the VTA compared to the Fischer 344 rat. Based on recent reports that individual differences in drug responses among outbred Sprague-Dawley rats are highly correlated with the animals' locomotor response to novelty, we determined in the present study whether such within-strain differences in locomotor behavior are also associated with differences in levels of TH and NFs in the VTA. Groups of 42 Sprague-Dawley rats were assessed for locomotor activity in a novel environment. The four animals from each group with the lowest locomotor responses (designated L rats), and the four with the highest locomotor responses (designated H rats), were analyzed for TH and NF immunoreactivity by immunoblotting procedures. It was found that the VTA of L rats exhibited higher levels of TH and lower levels of three major NF proteins, NF-200, NF-160, and NF-68, compared to the VTA of H rats. A tendency for similar L versus H differences in TH and NF levels were observed when groups of rats with the second lowest and second highest locomotor responses were compared; no differences were seen in groups whose locomotor responses were closer to the median. These biochemical differences between H and L rats showed regional specificity, with no significant differences seen in several other regions of brain or spinal cord studied. Differences were also observed between L and H rats in their locomotor responses to acute and repeated cocaine exposure. The possible relationship between the individual differences in TH and NFs and individual differences in locomotor activity and other drug-related behaviors is discussed.

Coordinate regulation of the cyclic AMP system with firing rate and expression of tyrosine hydroxylase in the rat locus coeruleus: effects of chronic stress and drug treatments.

Recent studies have demonstrated that chronic stress increases the firing rate and expression of tyrosine hydroxylase (TH) in neurons of the locus coeruleus (LC), the major noradrenergic nucleus in brain. The present study was undertaken to examine the influence of chronic stress and other treatments known to influence the activity of LC neurons on the cyclic AMP (cAMP) second messenger system in these neurons. Chronic (5 days) cold exposure significantly increased levels of TH immunoreactivity in the LC, as previously reported, but not in substantia nigra (SN) or ventral tegmentum (VT), two dopaminergic nuclei studied for comparison. Chronic cold exposure increased levels of cAMP-dependent protein kinase activity in soluble, but not particulate, fractions of the LC, and increased basal and GTP- and forskolin-stimulated adenylate cyclase activity in this brain region. In contrast, levels of the protein kinase and adenylate cyclase in VT, SN, and frontal cortex were not significantly influenced by cold exposure. To study further the relationship between regulation of LC firing rate, TH expression, and the cAMP system in the LC, other treatments known to influence TH were examined. Reserpine treatment, shown previously to increase levels of TH, was found to increase both LC firing rate and levels of soluble cAMP-dependent protein kinase activity in the LC. 6-Hydroxydopamine, shown previously to increase levels of TH and firing rate of LC neurons, also increased soluble levels of protein kinase activity. Other treatments known to either increase (adrenalectomy) or decrease (chronic imipramine) levels of TH in the LC were also found to increase or decrease, respectively, levels of cAMP-dependent protein kinase activity in this brain region.(ABSTRACT TRUNCATED AT 250 WORDS)

Endogenous ADP-ribosylation in brain: initial characterization of substrate proteins.

Cholera and pertussis toxin-mediated ADP-ribosylation has been used extensively to study regulation of guanine nucleotide binding proteins (G proteins) in the nervous system, but much less is known about possible endogenous ADP-ribosylation of G proteins in brain. The present study demonstrates endogenous ADP-ribosylation, in the absence of cholera and pertussis toxins, of four predominate proteins in homogenates of rat cerebral cortex. These proteins showed apparent molecular masses of 20, 42, 45, and 50 kDa by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The 42- and 45-kDa proteins comigrated precisely with the major cholera toxin-labeled bands. Furthermore, the endogenous ADP-ribosylated and cholera toxin-ADP-ribosylated bands yielded identical 32P-labeled peptide fragments by one-dimensional peptide mapping, indicating that they are probably the same proteins, presumably the alpha-subunits of Gs. In contrast, peptide maps of the 50-kDa protein, which migrated close to a 48-kDa cholera toxin-labeled band, demonstrated that this protein is distinct from the toxin-labeled band and from Gs alpha. Levels of endogenous ADP-ribosylation activity showed regional heterogeneity in brain, with a nearly threefold variation observed among the brain regions examined. Chronic administration (7 days) of corticosterone significantly increased overall levels of endogenous ADP-ribosylation, indicating that components of this system may be under hormonal control in vivo. Attempts to identify neurotransmitters or second messenger systems that regulate endogenous ADP-ribosylation activity in brain have so far been unsuccessful with one exception.(ABSTRACT TRUNCATED AT 250 WORDS)

A general role for adaptations in G-proteins and the cyclic AMP system in mediating the chronic actions of morphine and cocaine on neuronal function.

Previous studies have shown that chronic morphine increases levels of the G-protein subunits Gia and Goa, adenylate cyclase, cyclic AMP-dependent protein kinase, and certain phosphoproteins in the rat locus coeruleus, but not in several other brain regions studied, and that chronic morphine decreases levels of Gia and increases levels of adenylate cyclase in dorsal root ganglion/spinal cord (DRG-SC) co-cultures. These findings led us to survey the effects of chronic morphine on the G-protein/cyclic AMP system in a large number of brain regions to determine how widespread such regulation might be. We found that while most regions showed no regulation in response to chronic morphine, nucleus accumbens (NAc) and amygdala did show increases in adenylate cyclase and cyclic AMP-dependent protein kinase activity, and thalamus showed an increase in cyclic AMP-dependent protein kinase activity only. An increase in cyclic AMP-dependent protein kinase activity was also observed in DRG-SC co-cultures. Morphine regulation of G-proteins was variable, with decreased levels of Gia seen in the NAc, increased levels of Gia and Goa in amygdala, and no change in thalamus or the other brain regions studied. Interestingly, chronic treatment of rats with cocaine, but not with several non-abused drugs, produced similar changes compared to morphine in G-proteins, adenylate cyclase, and cyclic AMP-dependent protein kinase in the NAc, but not in the other brain regions studied. These results indicate that regulation of the G-protein/cyclic AMP system represents a mechanism by which a number of opiate-sensitive neurons adapt to chronic morphine and thereby develop aspects of opiate tolerance and/or dependence. The findings that chronic morphine and cocaine produce similar adaptations in the NAc, a brain region important for the reinforcing actions of many types of abused substances, suggest further that common mechanisms may underlie psychological aspects of drug addiction mediated by this brain region.

Chronic cocaine treatment decreases levels of the G protein subunits Gi alpha and Go alpha in discrete regions of rat brain.

A possible role for G proteins in contributing to the chronic actions of cocaine was investigated in three rat brain regions known to exhibit electrophysiological responses to chronic cocaine: the ventral tegmental area, nucleus accumbens, and locus coeruleus. It was found that chronic, but not acute, treatment of rats with cocaine produced a small (approximately 15%), but statistically significant, decrease in levels of pertussis toxin-mediated ADP-ribosylation of Gi alpha and Go alpha in each of these three brain regions. The decreased ADP-ribosylation levels of the G protein subunits were shown to be associated with 20-30% decreases in levels of their immunoreactivity. In contrast, chronic cocaine had no effect on levels of G protein ADP-ribosylation or immunoreactivity in other brain regions studied for comparison. Chronic cocaine also had no effect on levels of Gs alpha or G beta immunoreactivity in the ventral tegmental area and nucleus accumbens. Specific decreases in Gi alpha and Go alpha levels observed in response to chronic cocaine in the ventral tegmental area, nucleus accumbens, and locus coeruleus are consistent with the known electrophysiological actions of chronic cocaine on these neurons, raising the possibility that regulation of G proteins represents part of the biochemical changes that underlie chronic cocaine action in these brain regions.

Chronic antidepressant administration alters the subcellular distribution of cyclic AMP-dependent protein kinase in rat frontal cortex.

The influence of chronic administration of antidepressants on cyclic AMP-dependent protein kinase activity was examined in rat frontal cortex. Chronic administration of imipramine, tranylcypromine, or electroconvulsive seizures decreased cyclic AMP-dependent protein kinase activity in soluble fractions by approximately 25%, whereas enzyme activity was increased in the particulate fractions by approximately 20%. In contrast, enzyme activity in crude homogenates was not altered. This effect appears to be specific to antidepressant drugs, because representatives of several other classes of psychotropic drugs-namely, haloperidol, morphine, and diazepam--failed to alter either soluble or particulate levels of cyclic AMP-dependent protein kinase activity in this brain region following chronic administration. When the total particulate fraction was subfractionated, it was found that chronic imipramine treatment significantly increased the activity of cyclic AMP-dependent protein kinase in crude nuclear fractions but not in crude synaptosomal or microsomal fractions. Taken together, the data raise the possibility that chronic antidepressant treatments may stimulate the translocation of cyclic AMP-dependent protein kinase from the cytosol to the nucleus. This effect would represent a novel action of antidepressants that could contribute to the long-term adaptive changes in brain thought to be essential for the clinical actions of these treatments.

Sodium and potassium regulation of guanine nucleotide-stimulated adenylate cyclase in brain.

The present study examines the influence of potassium and sodium ions on guanine nucleotide regulation of adenylate cyclase in various brain regions, including the locus coeruleus (LC), dorsal raphe (DR), ventral tegmentum (VT), hippocampus (HP), frontal cortex (FC), substantia nigra (SN), neostriatum (NS) and cerebellum (CB). Guanine nucleotide regulation of adenylate cyclase was highest in the LC, DR and VT and lowest in NS and CB. Sodium and potassium ions were found to stimulate basal or GTP-activated adenylate cyclase in NS and SN, whereas the cations were found to specifically inhibit guanine nucleotide-stimulated enzyme activity in all other brain regions with the exception of CB, where there was no effect. With regard to stimulation of adenylate cyclase, lithium was more potent than sodium which was more potent than potassium in SN and NS. With regard to inhibition of the enzyme, potassium was equipotent to lithium which was greater than sodium in the other brain regions examined. Both stimulatory and inhibitory effects of cations in the different regions were significant (P less than 0.05) at 30 mM and were maximal at 90-120 mM. Sodium ion inhibition of GTP-stimulated adenylate cyclase in LC and DR was partially blocked by pertussis toxin treatment, whereas cation stimulation in NS was not affected by the toxin. The results demonstrate marked region-specific effects of sodium and potassium on adenylate cyclase, which could occur at either G-proteins or the catalytic unit of the enzyme. The possibility that ion fluxes alter G-protein function is discussed.