Cholera toxin differentially decreases membrane levels of alpha and beta subunits of G proteins in NG108-15 cells

Distribution of G-protein alpha subunits and neurotransmitter activation of g(alphai) and g(alphaq) in the brain of the lobster Homarus americanus

Immunocytochemistry using antisera specific for the G-protein alpha subunits G(alphai), G(alphaq), and G(alphas) revealed similar patterns of immunoreactivity in the lobster brain. Immunoreactivity was strongest in neuropil, especially the olfactory and accessory lobes, and was characterized by bundles of fine threads leading to dense concentrations of punctate staining in the glomeruli. This may reflect the concentration of G-protein alpha subunits at synapses. The major differences between the antisera were distinct patterns of staining intensity in subregions of glomeruli of the olfactory and accessory lobes. This result is potentially correlated with previous evidence that these subregions are neurochemically distinct. Neuronal cell bodies contained moderate levels of immunoreactivity at the plasma membrane and faint staining in the cytoplasm. The olfactory globular tract was moderately immunoreactive, but other fiber tracts were weakly immunoreactive. Immunoreactivity in the deutocerebral commissure consisted of small oval cell bodies and strands that formed a reticulated pattern, suggestive of glia. Photoaffinity labelling by using an analog of GTP demonstrated that histamine activated G(alphai) in brain homogenates. Further evidence of G-protein activation was obtained by showing that stimulation with a mixture of neuroactive substances increased the amount of phospholipase C-beta associated with membranes, G(alphaq), and G(beta). The lobster brain, especially in its neuropil regions, is richly endowed with neuromodulatory biochemical pathways involving G(alphai), G(alphaq), and G(alphas).

Inhibition of ADP-ribosyltransferase increases synthesis of Gs alpha in neuroblastoma x glioma hybrid cells and reverses iloprost-dependent heterologous loss of fluoride-sensitive adenylate cyclase

Exposure of NG108-15 cells to 50 mM nicotinamide [an inhibitor of mono(ADP-ribosyl)transferase] for 18 hr led to an increase in membrane associated Gs alpha measured either as cholera toxin substrate or by immunoblotting with a specific antiserum. Prolonged exposure of NG108-15 cells to iloprost is followed by homologous loss of iloprost sensitivity, and heterologous loss of fluoride-dependent activation of adenylate cyclase. Nicotinamide reversed the loss of fluoride sensitivity, but failed to restore iloprost-dependent activation of adenylate cyclase. These results with nicotinamide in NG108-15 cells contrasted with those from platelets, which also exhibit heterologous desensitization of fluoride sensitivity following prolonged exposure to iloprost. Treatment of platelets with 50 mM nicotinamide for 18 hr led to an increase of 75.0 +/- 19.4% in the amount of membrane associated cholera toxin substrate. However, there was no associated increase in the abundance of Gs alpha as determined by immunoblotting. Furthermore, in platelets there was no restoration by nicotinamide of the iloprost-dependent loss of fluoride-sensitive adenylate cyclase activity. It follows that heterologous desensitization in platelets is accompanied by inactivation of Gs alpha, which is retained within the plasma membrane in its inactive state. The nicotinamide-dependent increase in the abundance of membrane associated cholera toxin substrate and immunoreactive Gs alpha in NG108-15 cells is associated with an increase of 72.0 +/- 20.3% in the levels of mRNA encoding Gs alpha. The capacity of nicotinamide to increase the abundance of membrane associated Gs alpha was reversed when the cells were cultured in the presence of 20 micrograms/mL cycloheximide. These results suggest that the ability of nicotinamide to increase the abundance of Gs alpha in NG108-15 cells is mediated by de novo protein synthesis.

Cholera toxin-induced Gs alpha down-regulation in neural tissue: studies on the pineal gland

Cholera toxin (CT) treatment (50 micrograms/ml) was used to down regulate the alpha subunit of the stimulatory guanine nucleotide binding protein (Gs alpha) in pineal glands in organ culture, as has been seen in non-neural tissue. A 15 h treatment reduces Gs alpha by approximately 75% as measured using semi-quantitative Western blot technology. In contrast, this treatment does not alter the abundance of G beta, Gi alpha or Go alpha. This effect on Gs alpha was still apparent following a 36-h washout period. The 48-h CT treatment increased cyclic AMP accumulation 10- to 17-fold but blocked the norepinephrine (NE)-induced increase in cyclic AMP accumulation, presumably reflecting the loss of Gs alpha. This treatment did not, however, inhibit protein synthesis or stimulation of arylalkylamine N-acetyltransferase (NAT) activity produced by treatment with either DB-cyclic AMP (N6,2'-O-dibutyryl adenosine 3',5' monophosphate) or 8 Br-cyclic AMP, stable cyclic AMP derivatives. This indicates that a 48-h CT treatment was not generally toxic. In contrast, this treatment blocked subsequent CT stimulation of NAT. The effects of CT treatment on the adrenergic stimulation of NAT was examined using treatments which selectively produced alpha- or beta-adrenergic stimulation. alpha 1-Adrenergic activation of the pineal gland elevates [Ca2+]i, which potentiates effects of cyclic AMP; in these studies the response to alpha-adrenergic activation was markedly increased in 48 h CT-treated glands, reflecting Ca2+ potentiation of the effects of elevated levels of cyclic AMP.(ABSTRACT TRUNCATED AT 250 WORDS)

Correlation between prolactin secretion and Gs protein expression during sustained cholera-toxin stimulation

We have studied the chronic effect of cholera toxin (CTX) on prolactin synthesis and secretion in GH3 pituitary-tumour cells. Time-course analysis showed that prolactin secretion increased with time of CTX exposure, reached a peak at 3 h, and decreased thereafter. Prolactin synthesis was also shown to be stimulated by CTX. The basic and forskolin-stimulated cyclic AMP levels of the CTX-treated cells followed a biphasic time response similar to that of prolactin secretion. Exposure of cells to CTX for more than 3 h abolished the subsequent CTX-catalysed ADP-ribosylation in vitro. Moreover, a significant decrease in the pertussis-toxin-catalysed ADP-ribosylation was found after cells were exposed to CTX for longer than 6 h. Western-blot analysis indicated that the amount of Gs alpha (alpha-subunit of Gs) protein increased within 3 h, followed by a gradual decrease to 50% of the control level at 24 h. The accumulation of Gs alpha mRNA increased within 6 h of CTX exposure, and decreased thereafter to 40% of the basal level at 48 h. Our findings that prolonged treatment of CTX induced similar patterns of time responses in Gs alpha protein expression, cyclic AMP production and prolactin secretion indicate that CTX-induced changes in Gs alpha protein levels may be responsible for the cellular response leading to prolactin secretion.

Cholera toxin antagonizes morphine-induced catalepsy through a cyclic AMP-independent mechanism

We studied the effect of intracerebroventricular pretreatment with pertussis toxin and cholera toxin on morphine catalepsy in rats. Pertussis toxin (1 micrograms/rat, two, three and six days before) did not affect catalepsy evoked by central morphine. Cholera toxin (1 micrograms/rat) did not affect morphine catalepsy after 24 h and 48 h, but significantly reduced it (about 60%) after three and five days. Ten days later the morphine response had totally recovered. This effect was selective, since morphine analgesia was not modified. The reduction of catalepsy appeared unrelated to the ability of cholera toxin to raise cAMP levels, as demonstrated by the different time course of changes in striatal cholera toxin-stimulated adenylate cyclase activity. The effect required an intact cholera toxin molecule and did not occur with a similar dose of cholera toxin-B subunit. These findings demonstrate that catalepsy is an opioid effect not linked to pertussis toxin-sensitive G proteins and suggest that the Gs protein might be involved.

Gs alpha is a substrate for mono(ADP-ribosyl)transferase of NG108-15 cells. ADP-ribosylation regulates Gs alpha activity and abundance

NG108-15 neuroblastoma x glioma somatic hybrid cells were permeabilized in the presence of [32P]NAD+ and then cultured for 18 h. Resolution of the cell proteins on polyacrylamide gels revealed [32P]ADP-ribosylation of five major protein species with molecular mass values of 52 kDa, 44 kDa, 35 kDa, 30 kDa and 25 kDa. A similar pattern of labelling was also seen when NG108-15 cell membranes were incubated with [32P]NAD+ and hydrolysis of the product revealed mono(ADP-ribosyl)ation. Immunoprecipitation of these products with anti-Gs alpha antiserum revealed a single band identical to cholera toxin substrate. Culture of [32P]NAD(+)-loaded cells for 18 h in the presence of 50 mM-nicotinamide inhibited the eukaryotic mono(ADP-ribosyl)transferase activity. Inhibition of the eukaryotic enzyme was also accompanied by an increase in the abundance of Gs alpha, whether measured by Western blotting with anti-Gs alpha antibody (two separate antisera) or by cholera toxin-dependent [32P]ADP-ribosylation. There was no accompanying change in the abundance of G beta. The increase in Gs alpha abundance in nicotinamide-treated NG108-15 cells was accompanied by a 2-fold increase in basal adenylate cyclase activity (measured in the presence of GTP), and by a smaller but significant increase in iloprost-dependent activation of adenylate cyclase. Receptor number or affinity was not affected by nicotinamide, since this treatment did not alter the binding parameters of [3H]iloprost to NG108-15 cell membranes. Short-term exposure of cells to nicotinamide for 1 h revealed no significant difference in either basal or agonist-stimulated adenylate cyclase activity. These results reveal that mono(ADP-ribosyl)ation of Gs alpha by eukaryotic ADP-ribosyltransferase modifies the abundance and activity of Gs alpha in NG108-15 cells, and hence may play a role in the hormonal regulation of cell function.

The epithelial phenotype of human neuroblastoma cells express bradykinin, endothelin, and angiotensin II receptors that stimulate phosphoinositide hydrolysis

The neuroblastoma line SK-N-SH consists of distinct and interconverting cell types, which include a neuroblast phenotype (SH-SY5Y), an epithelial phenotype (SH-EP), and an intermediate cell type (SH-IN). In SH-SY5Y cells, only muscarinic receptor activation produced stimulation of phosphoinositide turnover, whereas in SH-EP cells, where muscarinic receptors are not present, the peptides bradykinin, endothelin, and angiotensin II stimulated phosphoinositide hydrolysis with EC50 values of 16, 6, and 0.7 nM, respectively, and a rank order of maximal effects of bradykinin greater than endothelin greater than angiotensin II. Fetal calf serum at concentrations between 1 and 10% was also a potent stimulator of phosphoinositide hydrolysis in SH-EP cells but not in SH-SY5Y cells. In the intermediate cell clone, SH-IN, phosphoinositide hydrolysis was stimulated not only by muscarinic receptors, but also by endothelin, bradykinin, and serum, an indication that this cell type harbors all the kinds of receptors that are differentially expressed in the other two cell types. The effects of the three peptides--bradykinin, endothelin, and angiotensin II--on phosphoinositide hydrolysis in SH-EP cells were additive, a result suggesting that the three kinds of receptors may activate distinct transducer proteins and/or phospholipase C subtypes. Pretreatment of intact SH-EP cells with pertussis toxin under conditions sufficient to ADP-ribosylate 90-95% of the endogenous guanine nucleotide regulatory protein substrates did not impair the ability of any of the receptors to stimulate phosphoinositide hydrolysis in any of the cell types. In contrast, short-term exposure to the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (1 microM) abolished the stimulation of phosphoinositide hydrolysis mediated by peptide receptors in SH-EP cells and partially inhibited that by muscarinic receptors in SH-SY5Y cells. Prolonged incubation of SH-EP cells with phorbol ester resulted in a recovery of receptor responsiveness, the extent and rate of which were different for each receptor type. In contrast, there was no recovery of responsiveness for muscarinic receptors in SH-SY5Y cells. The pattern of phorbol ester-mediated effects depended on the cell rather than on the receptor type. In fact, muscarinic receptor responsiveness in SH-IN, the intermediate cell type, was desensitized by and recovered from treatment with phorbol esters in a manner more similar to peptide receptors in SH-EP than to muscarinic receptors in SH-SY5Y. These data suggest that the transduction mechanisms by which distinct receptor types are coupled to phosphoinositide hydrolysis in the three cell phenotypes differ in sensitivity to feedback regulation by protein kinase C.

Analysis by mRNA levels of the expression of six G protein alpha-subunit genes in mammalian cells and tissues

The distribution and levels of expression of Gs alpha, Gi1 alpha, Gi2 alpha, Gi3 alpha, Go alpha, and Gx alpha mRNAs were compared by Northern blot analysis using several rat tissues and selected human and rat cell lines. Gi1 alpha, Go alpha, and Gx alpha, were detected in a limited number of tissue and cells whereas Gi2 alpha, Gi3 alpha, and Gs alpha, were expressed in all the tissues and cells tested albeit in varying amounts. The expression of these six genes appears to be differentially regulated during postnatal development of the rat brain. High expression levels particularly of Go alpha, in young rat brain may be related to the formation of neurites during differentiation of nerve cells.

Ras proteins activate calcium channels in neuronal cells

Ras (p21) proteins are involved in the control of cell growth and differentiation, but the mechanism by which they exert these effects is not yet known. Here we present evidence that c-Ha-ras (p21(Gly-12)) and its oncogenic mutant T24-ras (p21(Val-12)) selectively induce omega-conotoxin and dihydropyridine-sensitive Ca2+ currents within a few hours after introduction into the cytoplasm of neuroblastoma x glioma hybrid cells. Whereas control cells exhibited a mean Ca2+ current of 250 pA, it amounted to 730 pA in cells pretreated with ras protein. In cells loaded with p21(Gly-12), the effect occurred after 2 hours and was terminated after 8 hours. In contrast, introduction of p21(Val-12) resulted in a prolonged delay (6 hours) of the effect which lasted for more than 24 hours. When ras proteins were preactivated with the non-hydrolysable GTP analog GppNHp, the time courses of both p21(Gly-12) and p21(Val-12) effects were fast and sustained, suggesting that in intact cells (i) the GDP/GTP exchange is faster for p21(Gly-12) compared to p21(Val-12) and (ii) inactivation of p21(Gly-12) is mediated by GAP-induced GTPase activity. T-type Ca2+ currents and K+ currents were unaffected by ras proteins.