Dilated cardiomyopathy in two transgenic mouse lines expressing activated G protein alpha(q): lack of correlation between phospholipase C activation and the phenotype.

We previously described a transgenic mouse line (alpha(q)*52) in which cardiac-specific expression of activated G alpha(q)protein (HA alpha(q)*) leads to activation of phospholipase C beta (PLC beta), the immediate downstream target of HA alpha(q)*, with subsequent development of cardiac hypertrophy and dilation. We now describe a second, independent line in the same genetic background (alpha(q)*44h) with lower expression of HA alpha(q)* protein that ultimately results in the same phenotype: dilated cardiomyopathy (DCM) with severely impaired left ventricular systolic function (assessed by M-mode and 2D echocardiography), but with a much delayed disease onset. We asked if PLC activation correlates with the development of the phenotype. At 12-14 months, 65% of alpha(q)*44h mice still had normal cardiac function and ventricular weight/body weight ratios (VW/BW). However, their basal PLC activity, which began to increase in ventricles at 6 months, was threefold higher than in wild-type by 12 months. This increase was even more pronounced than in 2.5-month-old alpha(q)*52 mice, in which a twofold increase was accompanied by a 25% increase in VW/BW. Furthermore, at 12-14 months the increase in PLC activity in alpha(q)*44h mice with and without DCM was comparable. Thus, the delayed time course in alpha(q)*44h mice unmasked a lack of correlation between PLC activation and development of DCM in response to HA alpha(q)* expression, suggesting a role for additional pathways and/or mechanisms. It also revealed a differential temporal regulation of protein kinase C isoform expression. The markedly different ages of disease onset in these two mouse lines provide a model for studying both genetic modifying factors and potential environmental influences in DCM.

Reassembly of phospholipase C-beta2 from separated domains: analysis of basal and G protein-stimulated activities.

Phosphatidylinositol-specific phospholipase C-betas (PLC-betas) are the only PLC isoforms that are regulated by G protein subunits. To further understand the regulation of PLC-beta(2) by G proteins and the functional roles of PLC-beta(2) structural domains, we tested whether the separately expressed amino and carboxyl halves of PLC-beta(2) could associate to form catalytically active enzymes as two polypeptides, and we explored how the complexes thus formed would be regulated by G protein betagamma subunits (Gbetagamma). We expressed cDNA constructs encoding PLC-beta(2) fragments of different lengths in COS-7 cells and demonstrated by coimmunoprecipitation that the coexpressed fragments could assemble and functionally reconstitute an active PLC-beta(2). The pleckstrin homology domain of PLC-beta(2) was required for its targeting to the membrane and for substrate hydrolysis. Reconstituted enzymes that contained the linker region that joins the two catalytic domains were as active or more active than the wild-type PLC-beta(2). When the linker region was removed, basal PLC-beta(2) enzymatic activity was increased further, suggesting that the linker region exerts an inhibitory effect on basal PLC-beta(2) activity. The reconstituted enzymes, like wild-type PLC-beta(2), were activated by Gbetagamma; when the C-terminal region was present in these constructs, they were also activated by Galpha(q). Gbetagamma and Galpha(q) activated these PLC-beta(2) constructs equally in the presence or absence of the linker region. We conclude that the linker region is an inhibitory element in PLC-beta(2) and that Gbetagamma and Galpha(q) do not stimulate PLC-beta(2) through easing the inhibition of enzymatic activity by the linker region.

Altered regulation of potassium and calcium channels by GABA(B) and adenosine receptors in hippocampal neurons from mice lacking Galpha(o).

To examine the role of G(o) in modulation of ion channels by neurotransmitter receptors, we characterized modulation of ionic currents in hippocampal CA3 neurons from mice lacking both isoforms of Galpha(o). In CA3 neurons from Galpha(o)(-/-) mice, 2-chloro-adenosine and the GABA(B)-receptor agonist baclofen activated inwardly rectifying K(+) currents and inhibited voltage-dependent Ca(2+) currents just as effectively as in Galpha(o)(+/+) littermates. However, the kinetics of transmitter action were dramatically altered in Galpha(o)(-/-) mice in that recovery on washout of agonist was much slower. For example, recovery from 2-chloro-adenosine inhibition of calcium current was more than fourfold slower in neurons from Galpha(o)(-/-) mice [time constant of 12.0 +/- 0.8 (SE) s] than in neurons from Galpha(o)(+/+) mice (time constant of 2.6 +/- 0.2 s). Recovery from baclofen effects was affected similarly. In neurons from control mice, effects of both baclofen and 2-chloro-adenosine on Ca(2+) currents and K(+) currents were abolished by brief exposure to external N-ethyl-maleimide (NEM). In neurons lacking Galpha(o), some inhibition of Ca(2+) currents by baclofen remained after NEM treatment, whereas baclofen activation of K(+) currents and both effects of 2-chloro-adenosine were abolished. These results show that modulation of Ca(2+) and K(+) currents by G protein-coupled receptors in hippocampal neurons does not have an absolute requirement for Galpha(o). However, modulation is changed in the absence of Galpha(o) in having much slower recovery kinetics. A likely possibility is that the very abundant Galpha(o) is normally used but, when absent, can readily be replaced by G proteins with different properties.

Signal transduction in atria and ventricles of mice with transient cardiac expression of activated G protein alpha(q).

We recently showed that the transient expression of a hemagglutinin (HA) epitope-tagged, constitutively active mutant of the G protein alpha(q) subunit (HAalpha(q)*) in the hearts of transgenic mice is sufficient to induce cardiac hypertrophy and dilatation that continue to progress after HAalpha(q)* protein becomes undetectable. We demonstrated that the activity of phospholipase Cbeta, the immediate downstream target of activated Galpha(q), is increased at 2 weeks, when HAalpha(q)* is expressed, but also at 10 weeks, when HAalpha(q)* is no longer detectable. This observation suggested that the transient HAalpha(q)* expression causes multiple, persistent changes in cellular signaling pathways. We now demonstrate changes in the level, activity, or both of several signaling components, including changes in the amount and hormone responsiveness of phospholipase Cbeta enzymes, in the basal level of diacylglycerol (which predominantly reflects activation of phospholipase D), in the amount or distribution of protein kinase C (PKC) isoforms (PKCalpha, PKCdelta, and PKCepsilon), and in the amount of several endogenous G proteins. These changes vary depending on the isoform of the signaling molecule, the chamber in which it is expressed, and the presence or absence of HAalpha(q)*. Our results suggest that a network of linked signaling functions determines the development of hypertrophy. They also suggest that atria and ventricles represent different signaling domains. It is likely that such changes occur in other model systems in which the activity of a single signaling component is increased, either due to an activating mutation or due to overexpression of the wild type.

The WD repeat: a common architecture for diverse functions.

Our knowledge of the large family of proteins that contain the WD repeat continues to accumulate. The WD-repeat proteins are found in all eukaryotes and are implicated in a wide variety of crucial functions. The solution of the three-dimensional structure of one WD-repeat protein and the assumption that the structure will be common to all members of this family has allowed subfamilies of WD-repeat proteins to be defined on the basis of probable surface similarity. Proteins that have very similar surfaces are likely to have common binding partners and similar functions.

Cardiac myocytes express mRNA for ten RGS proteins: changes in RGS mRNA expression in ventricular myocytes and cultured atria.

Regulators of G-protein signalling (RGS) are recently identified proteins that shorten the lifetime of the activated G protein. We now show that rat cardiac myocytes express mRNA for at least 10 RGS. The mRNA for RGS-r is barely detectable in rat ventricles, but increases more than 20-fold during the 60- to 90-min process of isolating ventricular myocytes, and after 90 min of culture of atrial pieces in medium with Ca2+. Both in myocytes and in atria, the rise in RGS-r is transient. The mRNA for cardiac RGS5, but not RGS-r, is developmentally regulated. These studies suggest that rapid regulation of RGS levels may be a new mechanism that governs how signals are transmitted across the cardiac cell membrane.

Transient cardiac expression of constitutively active Galphaq leads to hypertrophy and dilated cardiomyopathy by calcineurin-dependent and independent pathways.

Cardiac hypertrophy and dilatation can result from stimulation of signal transduction pathways mediated by heterotrimeric G proteins, especially Gq, whose alpha subunit activates phospholipase Cbeta (PLCbeta). We now report that transient, modest expression of a hemagglutinin (HA) epitope-tagged, constitutively active mutant of the Gq alpha subunit (HAalpha*q) in hearts of transgenic mice is sufficient to induce cardiac hypertrophy and dilatation that continue to progress after the initiating stimulus becomes undetectable. At 2 weeks, HAalpha*q protein is expressed at less than 50% of endogenous alphaq/11, and the transgenic hearts are essentially normal morphologically. Although HAalpha*q protein declines at 4 weeks and is undetectable by 10 weeks, the animals develop cardiac hypertrophy and dilatation and die between 8 and 30 weeks in heart failure. As the pathology develops, endogenous alphaq/11 rises (2.9-fold in atria; 1.8-fold in ventricles). At 2 weeks, basal PLC activity is increased 9- to 10-fold in atria but not ventricles. By 10 weeks, it is elevated in both, presumably because of the rise in endogenous alphaq/11. We conclude that the pathological changes initiated by early, transient HAalpha*q expression are maintained in part by compensatory changes in signal transduction and other pathways. Cyclosporin A (CsA) prevents hypertrophy caused by activation of calcineurin [Molkentin, J. D., Lu, J.-R., Antos, C. L., Markham, B., Richardson, J., Robbins, J., Grant, S. R. & Olson, E. N. (1998) Cell 93, 215-228]. Because HAalpha*q acts upstream of calcineurin, we hypothesized that HAalpha*q might initiate additional pathways leading to hypertrophy and dilatation. Treating HAalpha*q mice with CsA diminished some, but not all, aspects of the hypertrophic phenotype, suggesting that multiple pathways are involved.

Effect of deletion of the major brain G-protein alpha subunit (alpha(o)) on coordination of G-protein subunits and on adenylyl cyclase activity.

Heterotrimeric G-proteins, composed of alpha and betagamma subunits, transmit signals from cell-surface receptors to cellular effectors and ion channels. Cellular responses to receptor agonists depend on not only the type and amount of G-protein subunits expressed but also the ratio of alpha and betagamma subunits. Thus far, little is known about how the amounts of alpha and betagamma subunits are coordinated. Targeted disruption of the alpha(o) gene leads to loss of both isoforms of alpha(o), the most abundant alpha subunit in the brain. We demonstrate that loss of alpha(o) protein in the brain is accompanied by a reduction of beta protein to 32+/-2% (n = 4) of wild type. Sucrose density gradient experiments show that all of the betagamma remaining in the brains of alpha(o)-/- mice sediments as a heterotrimer (s20,w = 4.4 S, n = 2), with no detectable free alpha or betagamma subunits. Thus, the level of the remaining betagamma subunits matches that of the remaining alpha subunits. Protein levels of alpha subunits other than alpha(o) are unchanged, suggesting that they are controlled independently. Coordination of betagamma to alpha occurs posttranscriptionally because the mRNA level of the predominant beta1 subtype in the brains of alpha(o)-/- mice was unchanged. Adenylyl cyclase can be positively or negatively regulated by betagamma. Because the level of other alpha subunits is unchanged and alpha(o) itself has little or no effect on adenylyl cyclase, we could examine how a large change in the level of betagamma affects this enzyme. Surprisingly, we could not detect any difference in the adenylyl cyclase activity between brain membranes from wild-type and alpha(o)-/- mice. We propose that alpha(o) and its associated betagamma are sequestered in a distinct pool of membranes that does not contribute to the regulation of adenylyl cyclase.

Sites important for PLCbeta2 activation by the G protein betagamma subunit map to the sides of the beta propeller structure.

The betagamma subunits of the heterotrimeric GTP-binding proteins (G proteins) that couple heptahelical, plasma membrane-bound receptors to intracellular effector enzymes or ion channels directly regulate several types of effectors, including phospholipase Cbeta and adenylyl cyclase. The beta subunit is made up of two structurally different regions: an N-terminal alpha helix followed by a toroidal structure made up of 7 blades, each of which is a twisted beta sheet composed of four anti-parallel beta strands (Wall, M. A., Coleman, D. E., Lee, E., Iñiguez-Lluhi, J. A., Posner, B. A., Gilman, A. G., and Sprang, S. R. (1995) Cell 83, 1047-1058; Lambright, D. G., Sondek, J., Bohm, A., Skiba, N. P., Hamm, H. E., and Sigler, P. B. (1996) Nature 379, 311-319). We have previously shown that sites for activation of PLCbeta2, PLCbeta3, and adenylyl cyclase II overlap on the "top" surface of the propeller, where Galpha also binds (Li, Y., Sternweis, P. M., Charnecki, S., Smith, T. F., Gilman, A. G., Neer, E. J., and Kozasa, T. (1998) J. Biol. Chem. 273, 16265-16272). The present study was undertaken to identify the regions on the side of the torus that might be important for effector interactions. We made mutations in each of the outer beta strands of the G protein beta1 propeller, as well as mutations in the loops that connect the outer strands to the adjacent beta strands. Our results suggest that activation of PLCbeta2 involves residues in the outer strands of blades 2, 6, and 7 of the propeller. We tested three of the mutations that most severely affected PLCbeta2 activity against two forms of adenylyl cyclase (ACI and ACII). Both inhibition of ACI and activation of ACII were unaffected by these mutations, suggesting that if ACI and ACII contact the outer strands, the sites of contact are different from those for PLCbeta2. We propose that distinct sets of contacts along the sides of the propeller will define the specificity of the interaction of betagamma with effectors.

Sites for Galpha binding on the G protein beta subunit overlap with sites for regulation of phospholipase Cbeta and adenylyl cyclase.

Heterotrimeric G proteins, composed of alpha and betagamma subunits, forward signals from transmembrane receptors to intracellular effector enzymes and ion channels. Free betagamma activates downstream targets, but its action is terminated by association with GDP-liganded alpha subunits. Because alpha can inhibit activation of many effectors by betagamma, it is likely that the alpha subunit binding surfaces on betagamma overlap the surfaces necessary for effector activation. To test this hypothesis, we mutated residues on beta shown to contact alpha in the recently published crystal structures of the alphabetagamma heterotrimer (Wall, M. A., Coleman, D. E., Lee, E., Iniguez-Lluhi, J. A., Posner, B. A., Gilman, A. G., and Sprang, S. R. (1995) Cell 83, 1047-1058; Lambright, D. G., Sondek, J., Bohm, A., Skiba, N. P., Hamm, H. E., and Sigler, P. B. (1996) Nature 379, 311-319.). The alpha subunit binds to the flat, top surface of the toroidal beta subunit and also extends a helix along the side of the beta subunit at blade 1. We mutated four residues on the top surface of beta (Hbeta1[L117A], Hbeta1[D228R], Hbeta1[D246S], and Hbeta1[W332A]) and two residues on the side of beta that contacts alpha (Hbeta1[N88A/K89A]). Each of the mutant proteins was able to form beta gamma dimers, but they differed in their ability to bind alpha and to activate phospholipase C beta2 (PLCbeta2), PLCbeta3, and adenylyl cyclase II. Mutation of residues along the side of the torus at blade 1 diminish affinity for alpha but do not prevent activation of any of the effectors. Mutations on the alpha binding surface differentially affected PLCbeta2, PLCbeta3, and adenylyl cyclase II. Residues that affect PLCbeta and adenylyl cyclase II activity are found on opposite sides of the central tunnel, suggesting that PLC and adenylyl cyclase, like the alpha subunit, make many contacts on the top surface. None of the mutations affected the ability of betagamma to inhibit adenylyl cyclase I. We conclude that alpha, PLCbeta2, PLCbeta3, and adenylyl cyclase II share an interaction on the top surface of beta. The importance of individual residues is different for alpha binding and for effector activation and differs even between closely related isoforms of the same effector.

Folding a WD repeat propeller. Role of highly conserved aspartic acid residues in the G protein beta subunit and Sec13.

The beta subunit of the heterotrimeric G proteins that transduce signals across the plasma membrane is made up of an amino-terminal alpha-helical segment followed by seven repeating units called WD (Trp-Asp) repeats that occur in about 140 different proteins. The seven WD repeats in Gbeta, the only WD repeat protein whose crystal structure is known, form seven antiparallel beta sheets making up the blades of a toroidal propeller structure (Wall, M. A., Coleman, D. E., Lee, E., Iniguez-Lluhi, J. A., Posner, B. A., Gilman, A. G., and Sprang, S. R. (1995) Cell 83, 1047-1058; Sondek, J., Bohm, A., Lambright, D. G., Hamm, H. E., and Sigler, P. B. (1996) Nature 379, 369-374). It is likely that all proteins with WD repeats form a propeller structure. Alignment of the sequence of 918 unique WD repeats reveals that 85% of the repeats have an aspartic acid (D) residue (not the D of WD) in the turn connecting beta strands b and c of each putative propeller blade. We mutated each of these conserved Asp residues to Gly individually and in pairs in Gbeta and in Sec13, a yeast WD repeat protein involved in vesicular traffic, and then analyzed the ability of the mutant proteins to fold in vitro and in COS-7 cells. In vitro, most single mutant Gbeta subunits fold into Gbetagamma dimers more slowly than wild type to a degree that varies with the blade. In contrast, all single mutants form normal amounts of Gbetagamma in COS-7 cells, although some dimers show subtle local distortions of structure. Most double mutants assemble poorly in both systems. We conclude that the conserved Asp residues are not equivalent and not all are essential for the folding of the propeller structure. Some may affect the folding pathway or the affinity for chaperonins. Mutations of the conserved Asp in Sec13 affect folding equally in vitro and in COS-7 cells. The repeats that most affected folding were not at the same position in Sec13 and Gbeta. Our finding, both in Gbeta and in Sec13, that no mutation of the conserved Asp entirely prevents folding suggests that there is no obligatory folding order for each repeat and that the folding order is probably not the same for different WD repeat proteins, or even necessarily constant for the same protein.

G alpha(o) is necessary for muscarinic regulation of Ca2+ channels in mouse heart.

Heterotrimeric G proteins, composed of G alpha and G betagamma subunits, transmit signals from cell surface receptors to cellular effector enzymes and ion channels. The G alpha(o) protein is the most abundant G alpha subtype in the nervous system, but it is also found in the heart. Its function is not completely known, although it is required for regulation of N-type Ca2+ channels in GH3 cells and also interacts with GAP43, a major protein in growth cones, suggesting a role in neuronal pathfinding. To analyze the function of G alpha(o), we have generated mice lacking both isoforms of G alpha(o) by homologous recombination. Surprisingly, the nervous system is grossly intact, despite the fact that G alpha(o) makes up 0.2-0.5% of brain particulate protein and 10% of the growth cone membrane. The G alpha(o)-/- mice do suffer tremors and occasional seizures, but there is no obvious histologic abnormality in the nervous system. In contrast, G alpha(o)-/- mice have a clear and specific defect in ion channel regulation in the heart. Normal muscarinic regulation of L-type calcium channels in ventricular myocytes is absent in the mutant mice. The L-type calcium channel responds normally to isoproterenol, but there is no evident muscarinic inhibition. Muscarinic regulation of atrial K+ channels is normal, as is the electrocardiogram. The levels of other G alpha subunits (G alpha(s), G alpha(q), and G alpha(i)) are unchanged in the hearts of G alpha(o)-/- mice, but the amount of G betagamma is decreased. Whichever subunit, G alpha(o) or G betagamma, carries the signal forward, these studies show that muscarinic inhibition of L-type Ca2+ channels requires coupling of the muscarinic receptor to G alpha(o). Other cardiac G alpha subunits cannot substitute.

G protein beta gamma subunits.

Guanine nucleotide binding (G) proteins relay extracellular signals encoded in light, small molecules, peptides, and proteins to activate or inhibit intracellular enzymes and ion channels. The larger G proteins, made up of G alpha beta gamma heterotrimers, dissociate into G alpha and G beta gamma subunits that separately activate intracellular effector molecules. Only recently has the G beta gamma subunit been recognized as a signal transduction molecule in its own right; G beta gamma is now known to directly regulate as many different protein targets as the G alpha subunit. Recent X-ray crystallography of G alpha, G beta gamma, and G alpha beta gamma subunits will guide the investigation of structure-function relationships.

Intracellular signalling: turning down G-protein signals.

The recently discovered family of proteins known as 'regulators of G-protein signalling' offers a solution to an important puzzle about the termination of signalling by G proteins and may also be important in more long-term modulation of signalling via G proteins.

Analysis of the physical properties and molecular modeling of Sec13: A WD repeat protein involved in vesicular traffic.

WD repeat proteins are a family of proteins that contain a series of highly conserved internal repeat motifs, usually ending with WD (Trp-Asp). The G beta subunit of heterotrimeric guanine nucleotide binding protein is a member of this family, and its crystal structure has been recently solved at high resolution (Wall et al. (1995) Cell 83, 1047-1058; Sondek et al. (1996) Nature 379, 369-374). Based on the coordinates of G beta, we have constructed a model for the structure of Sec13, a 33 kDa WD repeat protein from Saccharomyces cerevesiae essential for vesicular traffic. The model has been tested using a combination of biophysical and biochemical methods. Sec13 was expressed in Escherichia coli as a hexa-His-tagged protein (H6Sec13) and purified to homogeneity. In contrast to some other WD repeat proteins that are unable to fold into monomeric structures when expressed in E. coli, H6Sec13 was soluble and monomeric in the absence of detergent. The far-UV circular dichroism (CD) spectra of H6Sec13 indicated less than 10% alpha-helix consistent with the model which predicts primarily beta-sheets. H6Sec13 shows a cooperative and irreversible thermal denaturation curve consistent with a tightly packed structure. The CD spectrum shows an unusual positive ellipticity at 229 nm that was attributed to interactions of surface tryptophans since the 229 nm maximum could be abolished by modification of 6.3 +/- 0.3 (n = 3) tryptophans (out of 15 total in the molecule) with N-bromosuccinimide. Our model predicts that three sets of tryptophans are clustered near the surface. As predicted by the model, purified H6Sec13 was completely resistant to trypsin digestion. The concordance of the model of Sec13 presented in this paper with the biochemical and biophysical studies suggests that this model can be useful as a guide to further experiments designed to elucidate the function of Sec13 in vesicular traffic.

Folding of proteins with WD-repeats: comparison of six members of the WD-repeat superfamily to the G protein beta subunit.

The family of WD-repeat proteins comprises over 30 different proteins that share a highly conserved repeating motif [Neer, E. J., Schmidt, C. J., Nambudripad, R., & Smith, T. F. (1994) Nature 371, 297-300]. Members of this family include the signal-transducing G protein beta subunit, as well as other proteins that regulate signal transduction, transcription, pre-mRNA splicing, cytoskeletal organization, and vesicular fusion. The crystal structure of one WD-repeat protein (G beta) has now been solved (Wall et al., 1995; Sondek et al, 1996) and reveals that the seven repeating units form a circular, propeller-like structure with seven blades each made up of four beta strands. It is very likely that all WD-repeat proteins form a similar structure. If so, it will be possible to use information about important surface regions of one family member to predict properties of another. If WD proteins form structures similar to G beta, their hydrodynamic properties should be those of compact, globular proteins, and they should be resistant to cleavage by trypsin. However, the only studied example of a WD-repeat protein, G beta, synthesized in vitro in a rabbit reticulocyte lysate, is unable to fold into a native structure without its partner protein G gamma. The non-WD-repeat amino terminal alpha helix of G beta does not inhibit folding because G beta does not fold even when this region is removed. It is not known whether all WD-repeat proteins are unable to fold when synthesized in an in vitro system. We synthesized seven members of the family in a rabbit reticulocyte lysate, determined their Stokes radius, sedimentation coefficient, and frictional ratio, and assayed their stability to trypsin. Our working definition of folding was that the proteins from globular, trypsin-resistant structures because, except for G beta gamma, their functions are not known or cannot be assayed in reticulocyte lysates. We chose proteins that include amino and carboxyl extensions as well as proteins that are made up entirely of WD-repeats. We show that unlike G beta, several proteins with WD-repeats are able to fold into globular proteins in a rabbit reticulocyte lysate. One protein, beta Trcp, formed large aggregates like G beta, suggesting that it may also require a partner protein. Despite the presence of many potential tryptic cleavage sites, all of the proteins that did fold gave stable large products on tryptic proteolysis, as predicted on the basis of the structure of G beta. These studies suggest that other WD-repeat proteins are likely to form propeller structures similar to G beta.

Maintenance of cellular levels of G-proteins: different efficiencies of alpha s and alpha o synthesis in GH3 cells.

G-proteins couple membrane-bound receptors to intracellular effectors. Each cell has a characteristic complement of G-protein alpha, beta and gamma subunits that partly determines the cell's response to external signals. Very little is known about the mechanisms that set and maintain cellular levels of G-proteins or about potential points of regulation. We have assayed the steady-state levels of mRNA and protein for two types of G-protein subunits, alpha s and alpha o, in rat brain, heart and GH3 cells, and found that in all these cases, it takes 9- to 20-fold more mRNA to produce a given amount of alpha s protein than to produce the same amount of alpha o protein. Such a situation could arise from a relatively rapid rate of alpha s protein degradation, requiring rapid protein synthesis to compensate, or from relatively inefficient translation of alpha s mRNA compared with alpha o mRNA. The latter appears to be the case in GH3 cells. These cells contain 94 times more mRNA for alpha s than for alpha o, yet the rate of alpha s protein synthesis is only 9 times greater than alpha o protein synthesis. The degradation rates of the two proteins are similar (13 h for alpha s and 18 h for alpha o). To begin to define the mechanism that accounts for the fact that it takes more mRNA to synthesize a given amount of alpha s than alpha o, we asked whether there is a pool of alpha s mRNA that does not participate in protein synthesis. We found that virtually all alpha s and alpha o mRNA is associated with ribosomes. Therefore, all the mRNA is likely to be capable of directing protein synthesis. Since the rate-limiting step in protein synthesis is usually binding of the ribosome to mRNA at initiation, our results suggest that the relatively slow rate of alpha s protein synthesis is regulated by a mechanism that acts beyond initiation at peptide elongation and/or termination.

Intersubunit surfaces in G protein alpha beta gamma heterotrimers. Analysis by cross-linking and mutagenesis of beta gamma.

Heterotrimeric guanine nucleotide binding proteins (G proteins) are made up of alpha, beta, and gamma subunits, the last two forming a very tight complex. Stimulation of cell surface receptors promotes dissociation of alpha from the beta gamma dimer, which, in turn, allows both components to interact with intracellular enzymes or ion channels and modulate their activity. At present, little is known about the conformation of the beta gamma dimer or about the areas of beta gamma that interact with alpha. Direct information on the orientation of protein surfaces can be obtained from the analysis of chemically cross-linked products. Previous work in this laboratory showed that 1,6-bismaleimidohexane, which reacts with cysteine residues, specifically cross-links alpha to beta and beta to gamma (Yi, F., Denker, B. M., and Neer, E. J. (1991) J. Biol. Chem. 266, 3900-3906). To identify the residues in beta and gamma involved in cross-linking to each other or to alpha, we have mutated the cysteines in beta 1, gamma 2, and gamma 3 and analyzed the mutated proteins by in vitro translation in a rabbit reticulocyte lysate. All the mutants were able to form beta gamma dimers that could interact with the alpha subunit. We found that 1,6-bismaleimidohexane can cross-link beta 1 to gamma 3 but not to gamma 2. The cross-link goes from Cys25 in beta 1 to Cys30 in gamma 3. This cysteine is absent from any of the other known gamma isoforms and therefore confers a distinctive property to gamma 3. The beta subunit in the beta 1 gamma 2 dimer can be cross-linked to an unidentified protein in the rabbit reticulocyte lysate, generating a product slightly larger than cross-linked beta 1 gamma 3. The beta subunit can also be cross-linked to alpha, giving rise to two products on SDS-polyacrylamide gel electrophoresis, both of which were previously shown to be formed by cross-linking beta to Cys215 in alpha o (Thomas, T. C., Schmidt, C. J., and Neer, E. J. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 10295-10299). Mutation of Cys204 in beta 1 abolished one of these two products, whereas mutation of Cys271 abolished the other. Because both alpha-beta cross-linked products are formed in approximately equal amounts, Cys204 and Cys271 in beta are equally accessible from Cys215 in alpha o. Our findings begin to define intersubunit surfaces, and they pose structural constraints upon any model of the beta gamma dimer.

Localization of G alpha 0 to growth cones in PC12 cells: role of G alpha 0 association with receptors and G beta gamma.

The heterotrimeric G protein G0 is highly enriched in the growth cones of neuronal cells and makes up 10% of the membrane protein of growth cones from neonatal rat brain. We have used PC12 cells, a cell line that differentiates to a neuron-like phenotype, as a model with which to study the mechanism of G protein localization. First, the role of the beta gamma-subunit was investigated. The attachment of the beta gamma-subunit to the membrane depends on the isoprenylation of the gamma-subunit. The drug lovastatin blocks isoprenylation by inhibiting a key enzyme in the biosynthetic pathway. After treatment of PC12 cells with 10 microM lovastatin for 48 hours 50% of the beta gamma-subunits were cytosolic compared with 100% membrane bound beta gamma in control cells, as determined by cell fractionation, gel electrophoresis and western blot. Addition of 200 microM mevalonic acid reverses this effect. However, lovastatin affects neither the membrane attachment of alpha 0 nor its localization to the growth cones as determined by immunohistochemistry. This suggests that the localization and retention of alpha 0 are independent of the membrane attachment of the full complement of beta gamma-subunits. Second, pertussis toxin was used to block the interaction between alpha 0 and receptors. PC12 cells were treated with 0.1 microgram/ml pertussis toxin prior to and during nerve growth factor-induced differentiation. In vitro [32P]ADP-ribosylation confirmed that alpha 0 and alpha i were completely ADP-ribosylated by this treatment. The ADP-ribosylation by pertussis toxin did not interfere with neurite outgrowth. The localization of alpha 0 to the growth cones was indistinguishable from that in untreated cells. We conclude that G protein-receptor interaction is not necessary for the distribution of alpha 0 to growth cones.