Differences in transcription and translation of long and short Gs alpha, the stimulatory G-protein, in human atrium

Quantification of G protein Gaalphas subunit splice variants in different human tissues and cells using pyrosequencing

The G protein Galphas is derived from four alternatively spliced transcripts, two long variants (Galphas(L)+CAG and Galphas(L)-CAG), which include an extra 45-bp segment, and two short variants (Galphas(S)+CAG and Galphas(S)-CAG). The long and short forms differ in each case by splicing in or out of a serine residue encoded at the 3' end of the variable exon 3. The relative expression of all four variants in human tissues is poorly investigated due to experimental limitations. We therefore established a method for reliable relative mRNA quantification of these splice variants based on the Pyrosequencing technology, and determined Galphas transcript ratios in various human tissues and cells. Galphas(S)/Galphas ratio was highest in blood mononuclear cells (0.84 +/- 0.02, n = 16) and lowest in the brain (0.51 +/- 0.14, n = 3). The different ranges resulted from differences in Galphas(S)+CAG ratios, which ranged from a total Galphas ratio of 0.32 +/- 0.07 (n = 12) in heart tissue to 0.57 +/- 0.03 (n = 16) in blood mononuclear cells (p < 0.0001), whereas the Galphas(S)-CAG ratio was rather constant and ranged from 0.22 +/- 0.04 (n = 7) in retinoblastoma cells to 0.27 +/- 0.04 in lymphocytes (p = 0.19). The Galphas(L)+CAG ratio ranged from 0.02 +/- 0.02 in heart tissue to 0.05 +/- 0.01 in retinoblastoma cells, with a varying proportion of Galphas(L)-CAG, which ranged from 0.14 +/- 0.02 in blood mononuclear cells to 0.41 +/- 0.08 in heart tissue. Stimulation of immortalized B lymphoblasts with isoproterenol resulted in significant changes of splice variant ratios. Our data indicate that changes of long and short ratios of Galphas in different tissues affected Galphas(L)-CAG and Gas(S)+CAG rather than Galphas(L)+CAG and Galphas(S-)CAG. Furthermore, stimulation of cells seemed to affect splice variant ratios. These results are, therefore, suggestive of different biological functions of these variants.

Impaired Gene Expression of .BETA.1-Adrenergic Receptor, but Not Stimulatory G-Protein Gs.ALPHA., in Rat Ventricular Myocardium Treated with Isoproterenol

We investigated the gene expression of beta(1)-adrenergic receptor (beta(1)AR) and stimulatory G-protein Gsalpha, important signal transduction elements for regulating heart rate and contractility, in ventricle after chronic treatment with isoproterenol (ISO) in rat. Rats were treated with ISO (4 mg/kg, intraperitoneal) twice a day for 4 d. Ventricle weight of the heart and ventricle weight/body weight ratio were increased by 23% and 25% compared with control, respectively. Positive inotropic responses to ISO in left atrial muscle preparations isolated from ISO-treated rats were markedly decreased. Northern blot hybridization showed that the mRNA transcript of beta(1)AR was significantly decreased in ventricle of ISO-treated rats, whereas Gsalpha mRNA level was unchanged. Present results demonstrate that the gene expression of myocardial beta(1)AR, but not Gsalpha, was decreased in rat myocardium of ISO-induced cardiac hypertrophy, and suggesting that decrease in the gene expression of beta(1)AR may be one of the mechanisms responsible for the diminished cardiac function.

Greater inotropic and cyclic AMP responses evoked by noradrenaline through Arg389 beta 1-adrenoceptors versus Gly389 beta 1-adrenoceptors in isolated human atrial myocardium

1. We studied the biochemical and contractile responses of isolated human myocardial tissue expressing native receptor variants of the 389G>R beta(1)-adrenoceptor polymorphism. 2. Right atrial appendage was obtained from homozygous RR patients (n=37) and homozygous GG patients (n=17) undergoing elective cardiac surgery. The positive inotropic effect of noradrenaline in these tissues, mediated through beta(1)-adrenoceptors, was studied using electrically stimulated (1 Hz) atrial strips, as well as the effects of noradrenaline on cyclic AMP levels and cyclic AMP-dependent protein kinase. 3. Tissue from RR homozygotes (n=14) showed significantly increased inotropic potency to noradrenaline (-log EC(50), M=6.92+/-0.12) compared to GG homozygotes (n=8, -log EC(50), M=6.36+/-0.11, P<0.005). This difference was not dependent on tissue basal force. 4. Tissue cyclic AMP levels (pmol mg(-1)) were also greater in RR homozygotes (basal 34.8+/-3.7 n=12, 300 nM noradrenaline 41.4+/-7.6 n=9, 30 micro M noradrenaline 45.2+/-3.2 n=22, 0.2 mM isoprenaline 48.3+/-4.2 n=16) compared to GG homozygotes (basal 30.7+/-4.4 n=5, 300 nM noradrenaline 32.6+/-6.92 n=5, 30 micro M noradrenaline 38.1+/-3.1 n=8, 0.2 mM isoprenaline 42.6+/-5.2 n=6, P=0.007). There were no differences between the variants in terms of cyclic AMP-dependent protein kinase activity. 5. These data provide the first evidence that enhanced G-protein coupling of the R389 beta(1)-adrenoceptor variant reported in rodent fibroblast expression systems is also present in native human receptors. The functional consequence of this is to significantly alter the inotropic potency of beta(1)-adrenoceptor activation depending on its genotype at the 389 position.

Effect of maturation and aging on beta-adrenergic signal transduction in rat kidney and liver

The characteristics of the beta-adrenergic signal transduction system were analyzed in kidney and liver membrane preparations from neonatal (2-3 days), mature (2 months), and old (2 years) rats. When comparing kidneys from adult to neonatal rats, we found a higher beta-receptor density and a higher percentage of beta(1)-receptor subtype, lower immunoreactive G(salpha)-protein, a lower ratio between the high and low molecular weight splice variant of G(salpha), lower immunoreactive G(ialpha)-protein, and lower basal adenylate cyclase activity. When comparing livers from adult to neonatal rats, we found lower beta-receptor density and basal adenylate cyclase activity. Very few differences could be detected when comparing mature to old kidneys or livers. Stimulated adenosine 3',5'-cyclic monophosphate (cAMP) synthesis was tissue- and age-dependent. In liver, G-protein- and beta-receptor-stimulated cAMP synthesis mirrored basal adenylate cyclase activity and was highest in liver from neonatal animals. In contrast, cAMP synthesis was significantly more stimulated in kidneys from mature animals than from neonatal and senescent rats. We conclude that: (i) the stoichiometry of the components within the beta-receptor/G-protein/adenylate cyclase complex is not fixed but is both tissue- and age-dependent; (ii) adenylate cyclase enzyme activity is possibly but not necessarily the rate-limiting step in the beta-receptor-mediated synthesis of cAMP; and (iii) there is in vivo evidence for a preferential co-expression of the large splice variant of the G(s)-protein and beta(2)-receptor subtype. It is speculated that this could have important physiological consequences for the development of the kidney.

Beta(2)-adrenergic and several other G protein-coupled receptors in human atrial membranes activate both G(s) and G(i)

Cardiac G protein-coupled receptors that couple to Galpha(s) and stimulate cAMP formation (eg, beta-adrenergic, histamine, serotonin, and glucagon receptors) play a key role in cardiac inotropy. Recent studies in rodent cardiac myocytes and transfected cells have revealed that one of these receptors, the beta(2)-adrenergic receptor (AR), also couples to the inhibitory G protein Galpha(i) (activation of which inhibits cAMP formation). If beta(2)ARs could be shown to couple to Galpha(i) in the human heart, it would have important ramifications, because levels of Galpha(i) increase with age and in failing human heart. Therefore, we investigated whether beta(2)ARs in the human heart activate Galpha(i). By photoaffinity labeling human atrial membranes with [(32)P]azidoanilido-GTP, followed by immunoprecipitation with antibodies specific for Galpha(i), we found that Galpha(i) is activated by stimulation of beta(2)ARs but not of beta(1)ARs. In addition, we found that other Galpha(s)-coupled receptors also couple to Galpha(i), including histamine, serotonin, and glucagon. When coupling of these receptors to Galpha(i) is disrupted by pertussis toxin, their ability to stimulate adenylyl cyclase is enhanced. These data provide the first evidence that beta(2)AR and many other Galpha(s)-coupled receptors in human atrium also couple to Galpha(i) and that abolishing the coupling of these receptors to Galpha(i) increases the receptor-mediated adenylyl cyclase activity.

Selective beta1-adrenoceptor blockade enhances the activity of the stimulatory G-protein in human atrial myocardium

1. Chronic selective beta1-adrenoceptor (beta1AR) blocker treatment enhances the sensitivity of beta2-adrenoceptor (beta2AR) in human heart (Hall et al., 1990; 1991). To clarify the mechanism of the cross-sensitization between beta1AR and beta2AR, we determined whether the stimulatory G-protein (G(s)alpha) function is increased in atria from beta1AR-blocker treated patients compared with non-beta-blocked patients, and investigated whether this change is caused by an alteration of post-translational modification of Gsalpha protein. 2. G(s)alpha function was determined by reconstitution of human atrial G(s)alpha into S49 cyc- cell membranes. In the reconstitution system, GTPgammaS stimulated cyclic AMP generation in a dose-dependent manner. Upon 10(-4) M GTPgammaS stimulation, G(s)alpha activity in the beta1AR-blocker, atenolol, treated group (78.2+/-10. 3 pmol cyclic AMP mg(-1) min(-1) 10(-3)) was 65% higher than that in non-beta-blocked patients (47.3+/-6.3 pmol cyclic AMP mg(-1) min(-1) 10(-3), n=15, P=0.02). 3. Isoelectric point (pI) valu G(s)alpha were measured by two dimensional gel electrophoresis (2D-E) and the amount of each isoform quantified by image analysis of a Western blot of the gel using specific antibody. Multiple isoforms of G(s)alpha were detected by 2D-E with different pI values. There were no significant differences between the groups of patients in either pI values or the proportions of the acidic isoforms of G(s)alpha to the main basic form (n=12, P>0.05). 4. The results suggest that chronic beta1AR-blockade enhances Gsalpha function in human atrium, and this may account in part for the hypersensitivity of beta2AR and other Gs-coupled receptors during beta1AR-blockade. The increased G(s)alpha function is unlikely to be caused directly by blockade of protein kinase A phosphorylation of G(s)alpha protein.

Agonist-induced release of splice variants of the alpha subunit of the stimulatory G-protein from rat cardiac membranes

It was the aim of the present study to evaluate whether, in physiological cardiac tissue, long and short splice variants of G(s)alpha (the alpha subunit of the stimulatory G-protein) differ in their susceptibility to guanine nucleotide-mediated activation. As a measure of G(s)alpha activation, we determined the proportion of G(s)alpha subunits which were released from the plasma membrane upon stimulation. Membrane preparations from heart ventricles of Wistar rats were incubated with increasing concentrations of the non-hydrolyzable GTP analogue guanylyl-imidodiphosphate (GppNHp, 0-100 micromol/L) in the absence or presence of the beta-adrenoceptor agonist isoprenaline (1 micromol/L). The 45 and 52 kDa forms of G(s)alpha (G(s)alpha-S and -L, respectively) were measured in the supernatant of the incubation mixture by immunoblotting and densitometry. The increase in cyclic AMP induced by GppNHp was measured in the same supernatant. In the absence of isoprenaline, GppNHp increased cyclic AMP formation and the concentration-dependent release of G(s)alpha-L (Friedman test, P < 0.05), whereas the amount of soluble G(s)alpha-S was not affected. After addition of isoprenaline, the redistribution of G(s)alpha-S into the soluble fraction could be stimulated by GppNHp in a concentration-dependent manner (P < 0.05). Kinetic experiments revealed that activation of G(s)alpha-S by GppNHp was rather slow, but could be markedly enhanced by isoprenaline. Thus, it is likely that the different susceptibilities of G(s)alpha-S and -L towards GppNHp reflects differences in the rate of spontaneous GDP release.