Age increases cardiac Galpha(i2) expression, resulting in enhanced coupling to G protein-coupled receptors.

Cardiac G protein-coupled receptors that function through stimulatory G protein Galpha(s), such as beta(1)- and beta(2)-adrenergic receptors (beta(1)ARs and beta(2)ARs), play a key role in cardiac contractility. Recent data indicate that several Galpha(s)-coupled receptors in heart also activate Galpha(i), including beta(2)ARs (but not beta(1)ARs). Coupling of cardiac beta(2)ARs to Galpha(i) inhibits adenylyl cyclase and opposes beta(1)AR-mediated apoptosis. Dual coupling of beta(2)AR to both Galpha(s) and Galpha(i) is likely to alter beta(2)AR function in disease, such as congestive heart failure in which Galpha(i) levels are increased. Indeed, heart failure is characterized by reduced responsiveness of betaARs. Cardiac betaAR-responsiveness is also decreased with aging. However, whether age increases cardiac Galpha(i) has been controversial, with some studies reporting an increase and others reporting no change. The present study examines Galpha(i) in left ventricular membranes from young and old Fisher 344 rats by employing a comprehensive battery of biochemical assays. Immunoblotting reveals significant increases with age in left ventricular Galpha(i2), but no changes in Galpha(i3), Galpha(o), Galpha(s), Gbeta(1), or Gbeta(2). Aging also increases ADP-ribosylation of pertussis toxin-sensitive G proteins. Consistent with these results, basal as well as receptor-mediated incorporation of photoaffinity label [(32)P]azidoanilido-GTP indicates higher amounts of Galpha(i2) in older left ventricular membranes. Moreover, both basal and receptor-mediated adenylyl cyclase activities are lower in left ventricular membranes from older rats, and disabling of Galpha(i) with pertussis toxin increases both basal and receptor-stimulated adenylyl cyclase activity. Finally, age produces small but significant increases in muscarinic potency for the inhibition of both beta(1)AR- and beta(2)AR-stimulated adenylyl cyclase activity. The present study establishes that Galpha(i2) increases with age and provides data indicating that this increase dampens adenylyl cyclase activity.

tors (␤ 1 ARs and ␤ 2 ARs), that function through stimulatory GTP-binding regulatory proteins (G␣ s ) and activate adenylyl cyclase (AC). Activation of AC increases the formation of cAMP, which activates cAMP-dependent protein kinase A resulting in the phosphorylation of proteins controlling cardiac excitationcontraction (1). An important recent discovery is that ␤ 2 ARs (but not ␤ 1 ARs) in both rat (2,3) and human heart (4) also activate G␣ i , a G␣-subunit that inhibits AC (5). We also demonstrated that G␣ i couples to several other G␣ s -coupled receptors in human heart, including receptors for histamine, glucagon, and serotonin (4). Coupling of ␤ 2 AR and other G␣ s -coupled receptors to G␣ i is relevant to cardiac function because inactivation of G␣ i by pertussis toxin (PTX) increases myocyte contractility in rat heart (6) and increases both basal and receptormediated AC activity in human heart (4). In addition to inhibiting AC, the G␣ i pathway in heart is involved in antiapoptotic effects (7)(8)(9).
The dual coupling of ␤ 2 AR to both G␣ s and G␣ i is likely to alter ␤AR function in diseases in which cardiac G␣ i levels are increased, such as congestive heart failure and hypertensive cardiac hypertrophy (10 -13). Indeed, both congestive heart failure and cardiac hypertrophy are characterized by a reduced responsiveness of ␤ARs. Similarly, a decline in the responsiveness of cardiac ␤ARs due to aging has been demonstrated in both humans (14 -17) and rodents (18 -21). The age-induced decrease in ␤AR responsiveness is characterized at the molecular level by decreased stimulation of AC and at the whole organ level by a decrease in heart rate, ejection fraction, and cardiac output.
There have been conflicting reports about the effect of age on cardiac G␣ i levels in both humans and rodents. In one study of human heart, G␣ i levels were measured in atrial appendages received from surgical patients, and it was found that G␣ i expression increased with age (17). Another study examined G␣ i expression in human ventricles from hearts that were not suitable for transplant and found no change in G␣ i with age (16). Similar studies in rat heart have yielded inconsistent results even when the same strain of rat was used. In Fisher 344 rats, Roth et al. (21) reported an age-associated increase in cardiac G␣ i , which is reduced by chronic dynamic exercise. Johnson et al. (22) found an increase in G␣ i mRNA but no change in G␣ i protein. Two other reports found no increase in cardiac G␣ i protein (23,24). In Sprague-Dawley rats, Bohm et al. (19) found that age increases cardiac G␣ i , and the increase in G␣ i is reduced by exercise. In Wistar rats, Bazan et al. (25) reported an increase in cardiac G␣ i with age, but Xiao et al. (6) and Miyamoto et al. (26) found no change. Some of the reported differences on the effect of age may be attributable to experimental design. For example, the finding of Chin et al. (24) that age does not increase cardiac G␣ i in Fisher 344 rats may be explained by the fact that these investigators used 16-monthold rats for their old age group versus 24-month-old rats used by others. Moreover, most studies examined G␣ i expression by immunoblotting only, and different G␣ i antibodies were used. Nevertheless, the available evidence favors an increase in G␣ i with age, as indicated by the recent review of Roka et al. (27). However, two recent reviews by Lakatta (28,29) on global changes in cardiovascular aging state that G␣ i in heart does not increase with age. We believe that this conclusion is premature.
The importance of G␣ i in cardiac function underscores the need to establish whether or not cardiac G␣ i is affected with age. Therefore, we undertook a detailed study on the effect of age on G␣ i expression in rat ventricular membranes. We used Fisher 344 rats because these rats have been the most widely used rat strain for aging studies (30), and age-induced changes in cardiac structure have been characterized (31). Expression of the predominant cardiac subtypes of G␣ i , as well as of G␣ s , G␣ o , and the major subtypes of G␤, was assessed by immunoblotting. Levels of PTX-sensitive G␣ i /G␣ o proteins were also examined by radiolabeling through PTX-catalyzed ADP-ribosylation. In addition, G␣ i2 activity was assessed using photoaffinity labeling with [ 32 P]azidoanilido-GTP (AA-[ 32 P]GTP). We show age-dependent increases in both G␣ i2 expression levels and activation of G␣ i2 by ␤ 2 AR and other G protein-coupled receptors in heart. The age-induced increase in G␣ i2 has the functional effect of suppressing AC activity, which is restored by disrupting receptor-G␣ i coupling with PTX. I]CYP), RM/1 antibody specific for G␣ s raised against the peptide sequence RMHLRQYELL, and AS/7 antibody specific for G␣ i1 and G␣ i2 raised against the peptide sequence KNNLKDCGLF, were from PerkinElmer Life Sciences. Antibody specific for G␣ i1 and G␣ i3 , antibodies specific for G␤ 1 and G␤ 2 , and goat anti-rabbit IgG secondary antibody conjugated to horseradish peroxidase were from Santa Cruz Biotechnology (Santa Cruz, CA). GK-1 antibody for G␣ o was prepared by our laboratory and was characterized previously (32). Recombinant G␣ protein standards from Escherichia coli were from Calbiochem (La Jolla, CA). Protein A-Sepharose and [ 32 P]nicotinamide adenine dinucleotide ([ 32 P]NAD) were from Amersham Biosciences. PTX was from List Biological Laboratories (Campbell, CA). SuperSignal chemiluminescent reagent was from Pierce.

Materials-R(Ϫ)-isoproterenol-(ϩ)-
Animals-Sixteen Fisher 344 rats (eight 3-month-olds and eight 24-month-olds) were obtained from the National Institute on Aging under an Institutional Animal Care and Use Committee-approved protocol. Animals were sacrificed by decapitation, and left ventricles were extracted immediately, frozen in liquid nitrogen, and stored at Ϫ80°C.
Immunoblot Analysis-Immunoblotting was performed by separating 10 g of membrane protein by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by electrophoretic transfer onto polyvinylidene difluoride membranes. The membranes were stained for 5 min with 0.5% Ponceau S in 1% acetic acid to check for equal protein loading and transfer. The membranes were then blocked for 1 h with 10% nonfat milk in 20 mM Tris (pH 7.4), 500 mM NaCl, and 0.1% Tween 20, probed with either a 1:6000 dilution of RM/1, AS/7, or GK-1 antibody, a 1:200 dilution of anti-G␣ i3 antibody, or a 1:500 dilution of G␤ antibodies, then incubated with a 1:2000 dilution of goatanti-rabbit secondary antibody coupled to horseradish peroxidase. The immunoreactive proteins were visualized by incubation with Super-Signal chemiluminescent reagent and exposure to x-ray film. Protein bands were quantified by densitometry using Quantity One software (Bio-Rad, Hercules, CA). Electrophoresis and subsequent analysis of various amounts of protein yielded linear densitometric results.
Photoaffinity Labeling with AA-[ 32 P]GTP-AA-[ 32 P]GTP was synthesized according to published procedures and purified by thin layer chromatography on polyethylenimine cellulose (J.T. Baker, Phillipsburg, NJ) (32,33). Photoaffinity labeling of membranes (30 g of protein) was performed using 2 Ci of AA-[ 32 P]GTP in a total volume of 60 l in the presence of various drugs indicated in the figure legends. Photolabeled G␣ subunits were separated and subsequently analyzed by SDS-PAGE/autoradiography. The autoradiograms showed a prominent band below the 45-kDa molecular mass marker that represented G␣ i2 because it was immunoprecipitated with AS/7 antibody.
Pertussin Toxin Treatment of Ventricular Membranes for AC Assays-PTX (50 ng/l) was activated by incubation with 100 mM DTT and 0.25% SDS for 30 min at 30°C as described (34). The activation reaction was stopped by the addition of four volumes of 1 mg/ml ice-cold bovine serum albumin (BSA). Activated PTX (1 ng/l) was added to membranes (1.0 -1.5 mg/ml protein) in a buffer containing 37.5 mM Tris (pH 7.4), 6.25 mM MgCl 2 , 1 mM EDTA, 5 mM NAD, 2.5 mM ATP, 4 mM GTP, 10 mM thymidine, 10 mM DTT, 0.005% SDS, 0.08 mg/ml BSA, 0.5 mM benzamidine, 5 g/ml soybean trypsin inhibitor, 5 g/ml leupeptin, and 2.5 g/ml aprotinin. The mixture was incubated for 30 min at 30°C, then an equal volume was added of ice-cold 50 mM Tris (pH 7.4) with 1 mM benzamidine, 10 g/ml soybean trypsin inhibitor, 10 g/ml leupeptin, and 5 g/ml aprotinin. Membranes were pelleted by centrifugation at 100,000 ϫ g in an Optima TL ultracentrifuge and resuspended in the final resuspension buffer containing 75 mM Tris (pH 7.4), 12.5 mM MgCl 2 , 2 mM EDTA, 1 mM benzamidine, 10 g/ml soybean trypsin inhibitor, 10 g/ml leupeptin, and 5 g/ml aprotinin, at ϳ0.5-1.5 mg of protein/ml. A second tube containing the same volumes of all constituents, with the exception of H 2 O in place of PTX, was treated in the same fashion and used as a control.
ADP-ribosylation of PTX-sensitive G Proteins-PTX (50 ng/l) was activated by incubation with 100 mM DTT and 0.25% SDS for 30 min at 30°C as described (34). The activation reaction was stopped by the addition of four volumes of 1 mg/ml ice-cold BSA. Activated PTX (1 ng/l) was added to membranes (1.0 -1.5 mg/ml protein) in a buffer containing 37.5 mM Tris (pH 7.4), 6.25 mM MgCl 2 , 1 mM EDTA, 2.5 mM ATP, 4 mM GTP, 10 mM thymidine, 10 mM DTT, 0.005% SDS, 50 M NAD, 10 Ci/mmol [ 32 P]NAD, 0.08 mg/ml BSA, 0.5 mM benzamidine, 5 g/ml soybean trypsin inhibitor, 5 g/ml leupeptin, and 2.5 g/ml aprotinin. The mixture was incubated for 30 min at 30°C, and then 100 l of sample buffer (22.5 mM Tris (pH 6.8), 7.2% SDS, 9% glycerol, 0.01% bromphenol blue, and 10% 2-mercaptoethanol) was added to each tube, and samples were separated by SDS-PAGE and exposed to autoradiographic film. A second tube containing the same volumes of all constituents, with the exception of H 2 O in place of PTX, was treated in the same fashion and used as a control. Specific phosphorylation bands were identified on autoradiograms of the dried gels. 32  Statistical Analysis-Results are presented as mean Ϯ S.E. Each set of data was analyzed by Shapiro-Wilk tests for normality, then the statistical significance of comparisons between data from young and old samples was determined by performing unpaired Student's t tests and exact Wilcoxon Mann-Whitney tests on the mean values of each data set. p Ͻ 0.05 was considered significant for all comparisons. Fig. 1 shows immunoblots for several G␣ and G␤ protein subunits in left ventricular membranes from young (3 months) and old (24 months) rat hearts. We examined the expression of G␣ i2 , G␣ i3 , G␣ o , G␣ s , G␤ 1 , and G␤ 2 , all of which previously have been detected in rat heart (26,37). Quantitation of bands by densitometry reveals that, on average, there is a significant increase of 58 Ϯ 7% in G␣ i2 in old rat ventricles. In contrast, immunoblotting of each of the other G protein subtypes indicates that expression is unchanged between young and old rat ventricles.

PTX-sensitive G Proteins in Left Ventricular Membranes
Increase with Age-To confirm the increase in G␣ i2 observed by immunoblotting by another independent method, we examined ADP-ribosylation of PTX-sensitive G proteins (G␣ i /G␣ o ) in ventricles from young and old rats. This method has been used previously by Feldman et al. (10) to demonstrate a G␣ i increase in the failing human heart and by Bohm et al. (19) to demonstrate a G␣ i increase in the hearts of old Sprague-Dawley rats. As Fig. 2 shows, the amount of G␣ i /G␣ o labeled with [ 32 P]NAD in PTX-treated membranes is significantly increased by 39 Ϯ 8% in ventricles from older rats.
More G␣ i2 Is Activated in Left Ventricular Membranes from Older Hearts-We next determined whether the increased expression of G␣ i2 in older left ventricular membranes is accompanied by increased activation of G␣ i2 . To this end, we assessed the effect of age on the ability of ␤ 2 AR and other GPCRs to stimulate photoaffinity labeling of G␣ i2 with AA-[ 32 P]GTP. Activated GPCRs catalyze the exchange of GTP for GDP on ␣-subunits of G proteins associated with the GPCR, so the amount of AA-[ 32 P]GTP incorporated into the ␣-subunit gives a direct measure of the extent of G protein activation. Fig. 3A shows that stimulations through ␤ 2 ARs and glucagon receptors are significantly increased with age, from 102 Ϯ 25% and 101 Ϯ 31% above basal levels in young ventricles, to 226 Ϯ 33% and 244 Ϯ 40% above young basal levels, respectively. These results indicate that, as age increases, more G␣ i2 is activated upon stimulation of ␤ 2 ARs and glucagon receptors. Basal labeling of G␣ i2 also significantly increases (by 83 Ϯ 18%) in older membranes (Fig. 3A), consistent with the increased expression of G␣ i2 in older ventricles shown in Fig. 1. Photoaffinity labeling of stimulatory G␣ s is not altered with age (data not shown).
We next determined whether activation of G␣ i2 is also increased in aged heart through muscarinic acetylcholine receptor, a GPCR that interacts with G␣ i but not G␣ s . As shown in Fig. 3B, stimulation of muscarinic receptors in older left ventricular membranes produces greater photoaffinity labeling of G␣ i2 with AA-[ 32 P]GTP than in young membranes.
AC Activity Is Decreased in Older Rat Ventricles-Figs. 1, 2, and 3 show increased G␣ i2 protein expression and activation in old rat heart. Because stimulation of G␣ i -coupled receptors inhibits AC, we determined whether there are age-dependent increases in inhibition of cardiac AC production of cAMP. As shown in Table I, basal AC activity is significantly lower in old hearts, as are stimulations by ␤ 2 ARs (isoproterenol ϩ ␤ 1 AR antagonist CGP 20712A) and glucagon receptors. Age-dependent decreases in stimulations by ␤ 1 ARs (isoproterenol ϩ ␤ 2 AR antagonist ICI 118,551) and forskolin are not significant. To ascertain that the decrease in AC activity in older membranes was not because of a decrease in receptor number, we examined the expression level of ␤ARs in both young and old ventricular membranes. We determined ␤AR density using a saturating concentration of [ 125 I]CYP, and found no difference in the mean density of ␤ARs between young and old rat ventricles (Table I). These results are similar to other reports on ␤AR density in Fisher 344 rat hearts (21, 38 -42), though Mader et al. (43) did report a decrease in affinity. In other rat strains there may be moderate decreases in receptor number (6,19,25,44).
We also examined whether there are age-dependent changes in the ability of muscarinic receptors to inhibit ␤ 1 AR-and ␤ 2 AR-stimulated AC in rat ventricle. As shown in Table I, stimulation of muscarinic receptors strongly inhibits ␤ARstimulated AC in both young and old rat hearts. Because muscarinic receptors are able to inhibit AC stimulation almost completely in these membrane preparations, there is no significant difference between the maximal percent inhibition in young and old hearts, despite the increased coupling of muscarinic receptors to G␣ i2 shown in Fig. 3B. When the dose-response relationships were examined between muscarinic agonist concentration and inhibition of ␤AR-stimulated AC activity, small but significant increases were seen in the potency of muscarinic inhibition (Fig. 4). EC 50 s for muscarinic inhibition of ␤ 1 AR-and ␤ 2 AR-stimulated activity are 0.84 Ϯ 0.10 and 0.25 Ϯ 0.06 M in young ventricles, respectively, and 0.48 Ϯ 0.10 and 0.14 Ϯ 0.04 M in old ventricles. Previous   FIG. 1. Age increases G␣ i2 in rat left ventricle. Young (Y) and old (O) rat ventricular membranes were separated by SDS-PAGE, transferred to polyvinylidene difluoride membranes, probed with antibodies selective for G␣ and G␤ subunits, and visualized by chemiluminescence as described under "Experimental Procedures." n ϭ 6 for each age group. Each tissue was assayed at least twice, and representative immunoblots are shown.

FIG. 2. Age increases levels of PTX-sensitive G proteins.
Young (Y) and old (O) rat ventricular membranes were ADP-ribosylated by incubation with activated PTX and [ 32 P]NAD then subjected to SDS-PAGE followed by autoradiography. The radioactive bands were excised and counted as described under "Experimental Procedures." n ϭ 6 for each age group. Each tissue was assayed twice, and a representative autoradiogram is shown. studies (45) in Fisher 344 rats also indicate that maximal muscarinic inhibition of ␤AR-stimulated AC activity is similar in young and old rats. Detailed dose-response curves were not reported.
PTX Treatment Causes a Greater Increase in AC Activity in Old Hearts than in Young Hearts-We further examined the relationship between increased G␣ i2 levels and decreased basal and receptor-stimulated AC activity by eliminating G␣ i activity through PTX treatment. PTX ADP-ribosylates G␣ i /G␣ o and prevents their interaction with receptors, thereby removing the inhibition of AC by these G protein subtypes (46,47). Our previous studies have shown that PTX treatment of human atrial membranes results in an increase in AC stimulation through ␤ 2 AR and other G␣ s -coupled receptors (4), and others have shown that PTX treatment of myocytes increases ␤ 2 ARmediated contractility (2,6). As shown in Table II, the inactivation of G␣ i /G␣ o by PTX increases both basal and GPCRstimulated AC activity, and this effect is greater in the old than in the young left ventricular membranes for the G␣ i -coupled ␤ 2 ARs and glucagon receptors. G␣ i inactivation in old membranes restores receptor-stimulated cAMP production to the levels achieved in the younger tissue. These data are consistent with our finding that more G␣ i2 is activated upon stimulation of ␤ 2 ARs and other receptors in old heart (Fig. 3). ␤ 1 ARs do not couple to G␣ i . Accordingly, the increase in ␤ 1 AR-mediated stimulation of AC following PTX treatment was not signifi-cantly affected by age. The increase in stimulation by forskolin, a direct activator of AC, also is not affected by age. As expected, PTX treatment completely eliminated the inhibition of ␤ARstimulated AC activity by muscarinic receptors (data not shown). DISCUSSION The main finding of the present study is that G␣ i2 expression in rat left ventricle increases with age. The increase in G␣ i2 expression is demonstrated using several biochemical techniques including immunoblotting, ADP-ribosylation, and photoaffinity labeling with AA-[ 32 P]GTP. The increase in G␣ i2 results in enhanced coupling to G␣ s -coupled receptors such as FIG. 4. Age increases the potency of muscarinic inhibition of ␤AR-mediated stimulation of AC. Dose-response assays reveal small but significant decreases in the EC 50 for carbachol-mediated inhibition of cAMP production stimulated by either ␤ 1 ARs or ␤ 2 ARs in old (open squares or circles, respectively) versus young (filled squares or circles, respectively) rat ventricular membranes. ␤ 1 AR stimulation was achieved with 100 M isoproterenol ϩ 100 M ICI 118,551 (␤ 2 AR antagonist), and ␤ 2 AR stimulation was achieved with 100 M isoproterenol ϩ 100 M CGP 20712A (␤ 1 AR antagonist). n ϭ 4 for each age group. Assays were performed in duplicate, and representative curves are shown.

TABLE I
AC activity and ␤AR expression in young and old rat left ventricular membranes Receptors tested were ␤ 1 ARs (measured by inclusion of 100 M isoproterenol ϩ 100 M ICI 118,551 (␤ 2 AR antagonist)), ␤ 2 ARs (100 M isoproterenol ϩ 100 M CGP 20712A (␤ 1 AR antagonist)), glucagon receptors (50 M glucagon) and muscarinic receptors (100 M carbachol (Carb)). Direct activation of AC was tested with forskolin (50 M). n ϭ 6 for each age group, AC assays were performed in duplicate, and binding assays were performed in triplicate. *, p Ͻ 0.05 compared with AC stimulation by same condition in young membranes. ␤ 2 ARs and glucagon receptors, as well as to G␣ i -coupled muscarinic receptors. Thus, the net effect of the increase in G␣ i2 expression with age is an increase in G␣ i2 signaling. This results in reductions in both basal and receptor-mediated AC activities in aged heart, both of which are restored by disabling of G␣ i with PTX.
The present study examined the effect of age on cardiac G␣ i , an important issue about which conflicting data exist in the literature. Determination of the effects of age on G␣ i has become increasingly important in light of the recent demonstrations that many cardiac G␣ s -coupled receptors, including ␤ 2 AR, also couple to G␣ i (2)(3)(4). The data in the present study clearly indicate an increased level of G␣ i2 in old Fisher 344 rats, and importantly, also demonstrate an elevation in the receptorstimulated activation of G␣ i2 . Prior studies in rats had not provided a consensus as to the effects of age on cardiac G␣ i . There have been reports of increased G␣ i expression in Fisher 344 (21), Sprague-Dawley (19) and Wistar rats (25), as well as an increase in Fisher 344 G␣ i mRNA (22). However, there also have been reports of no change in G␣ i expression in Fisher 344 (22,23) and Wistar rats (6,26). These differences make it necessary to examine G␣ i by multiple biochemical approaches. Our data indicating G␣ i2 increases in immunoblotting, ADPribosylation and photoaffinity labeling lead us to conclude that there is a significant age-dependent elevation in Fisher 344 cardiac G␣ i2 . The increase in G␣ i2 in aged hearts is likely not due to hypertrophy because in Fisher 344 rats hypertrophy is seen at senescence, which occurs at 27-30 months of age (31).
In agreement with previous reports in both human (15)(16)(17) and rat heart (21, 23, 48 -51), we also demonstrate decreases in AC activity in old heart. To explain this phenomenon, some studies have implicated changes at the level of AC. There has been a report of a decrease in the number of forskolin binding sites in old rats (23), indicating a lower amount of AC in old tissue, though there have been no reports on AC mRNA levels in Fisher 344 rats. Although decreases in AC activity with age could be due to decreased amounts of AC or its targets, the fact that PTX treatment restores ␤ 2 AR-and glucagon receptorstimulated AC signals in older hearts to the levels in younger hearts indicates that PTX-sensitive proteins are responsible for the decreased receptor-stimulated enzyme activity. PTX treatment in both guinea pig heart and rat bladder results in similar reversals of age-induced decreases in AC (20,52). We conclude that elevated G␣ i is the main cause of reduced AC activity in aged rat left ventricles.
Apparently, an increase in cardiac G␣ i in aged rat heart does not affect ␤AR-mediated contractility. Although coupling of ␤ 1 ARs to stimulation of AC and contractility through G␣ s is widely accepted, questions remain as to whether effects on contractility through ␤ 2 AR involve cAMP. In humans, stimulation of contractility via ␤ 2 ARs has been reported to occur through a cAMP-dependent mechanism that results in protein kinase A-catalyzed phosphorylation of phospholamban, troponin I, and C-protein, as well as enhancement of both inotropy and lusitropy, in both non-failing (53) and failing human heart (54), as well as in non-failing myocardium from infants with Fallot tetralogy (55). Therefore, one would expect that in humans, an increase in the coupling of ␤ 2 AR to G␣ i would lower contractility. In contrast, cAMP-independent pathways control ␤ 2 AR-mediated contractility in rat (2, 56 -58), cat (59), sheep (60), and dog (61). Thus, in these species, an age-induced increase in G␣ i would not decrease ␤ 2 AR-mediated contractility. Consistent with this notion is the finding of Jain et al. (62) who found no difference in basal contractile and relaxation function in mice lacking either G␣ i2 or G␣ i3 . An important functional consequence of an age-induced increase in the coupling of cardiac ␤ 2 AR to G␣ i2 may be increased inhibition of apoptosis. It was shown recently that norepinephrine, acting through a G␣ s pathway, increases cardiac apoptosis (63,64). More recent data indicate that ␤ 1 ARs cause apoptosis, whereas ␤ 2 ARs acting through a G␣ i pathway oppose apoptosis (7)(8)(9). In adult rat ventricular myocytes, stimulation of a PTX-sensitive G␣ i -coupled pathway by ␤ 2 AR inhibits the number of apoptotic cells as measured by flow cytometry (7). Using a neonatal rat myocyte model, it was shown that ␤ 2 AR/G␣ i -mediated protection from apoptosis occurs through phosphatidylinositol 3-kinase (PI 3-kinase) and Akt/protein kinase B pathways (8). The ␤ 2 AR/G␣ i /PI 3-kinase signaling mechanism also has been shown to mediate the stimulation of NO production (65), a key mechanism in the cardioprotection conferred by ischemic preconditioning (66). Finally, ␤ 2 ARmediated protection from apoptosis recently has been reported to occur through a G␣ i -dependent stimulation of p38 kinase (67), though another study reports that ␤ 2 AR activates p38 kinase through a protein kinase A-dependent pathway that does not involve G␣ i (68). Thus, although the downstream signaling molecules involved have yet to be fully elucidated, an increase in cardiac G␣ i seen in aging or failing heart may be an adaptive mechanism to protect the heart from apoptosis, because apoptosis has been shown to occur in both aged (69,70) and failing heart (71,72).
In summary, the present study provides evidence that age increases G␣ i2 in older rat ventricles, and this results in more activated G␣ i2 upon stimulation of various GPCRs.

TABLE II
Effect of PTX on AC activity in young and old rat left ventricular membranes Receptors tested were ␤ 1 ARs (measured by inclusion of 100 M isoproterenol ϩ 100 M ICI 118,551 (␤ 2 AR antagonist)), ␤ 2 ARs (100 M isoproterenol ϩ 100 M CGP 20712A (␤ 1 AR antagonist)), and glucagon receptors (50 M glucagon). Direct activation of AC was tested with forskolin (50 M). Percent Change is the increase in cAMP production above the control level in the respective tissue in Table I. n ϭ 6 for each age group, and assays were performed in duplicate. *, p Ͻ 0.05 compared with percent change in response following PTX treatment in young membranes.