Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

In the elderly, there is an increased incidence of benign prostatic hyperplasia and of urinary incontinence, along with decreased bladder capacity, bladder compliance, urinary flow rate, and ability to postpone voiding (Brocklehurst and Dilane, 1966; Drach et al., 1979; Chun et al., 1988). -Adrenergic receptor activation plays an important role in the facilitation of urine storage (Edvardsen, 1968; de Groat and Saum, 1972). With aging, there is a decreased relaxation response of the rat bladder detrusor to -agonists in muscle strips contracted with KCl. In addition, the maximum relaxant response to isoproterenol on electric field stimulation-induced contractions was reduced significantly in the bladder from old rats compared with younger rats (Nishimoto et al., 1995).

Binding of an agonist to the -adrenergic receptors results in the binding of GTP to the stimulatory G-protein, Gs, which activates the catalytic component of adenylyl cyclase (Gilman, 1995). Activated -adrenergic receptors can activate both Gs and the inhibitory G-protein, Gi (Asano et al., 1984; Raymond, 1995; Lefkowitz, 1998). Because stimulation of Gi inhibits cAMP production, increases in Gi levels with aging may cause decreases in cAMP production. The age-dependent decrease in relaxation response of bladder detrusor to forskolin, which directly activates the catalytic component of adenylyl cyclase, and to isoproterenol, is related to decreased cAMP production (Wheeler et al., 1990). Relaxation by dibutyryl cAMP is age-independent, however, indicating that decreased -responsiveness does not appear to involve events distal to cAMP synthesis (Nishimoto et al., 1995). In addition, aging in the rat detrusor does not affect shortening velocity or the content of contractile and cytoskeletal proteins (Sjuva et al., 1997).

To assess the role of G-proteins in the age-dependent decrease in -adrenergic-induced detrusor relaxation, we assessed the quantity and function of G-proteins, including Gs, Gi, and Go. Gs and Gi were evaluated because they regulate adenylyl cyclase stimulation and inhibition, respectively. Go, which inhibits voltage-gated Ca²⁺ channels and is the other member of the pertussis toxin (PTX)-sensitive Gi/o family, also was quantified. The quantity of Gs, Gi, and Go was assayed by Western immunoblot analysis using primary antibodies directed at the unique regions of their -subunits. Gi/o protein function was quantified using PTX-catalyzed ADP-ribosylation with [³²P]NAD⁺. G-proteins were evaluated in bladder muscle from weanling (21 days old), young adult (90 days old), adult (6 months old), and aged (24 months old) rats to access changes in maturation and senescence. Because Gi increased with aging in the bladder, we evaluated 1) the effect of PTX treatment on isoproterenol-responsive adenylyl cyclase in young adult and old rat bladder particulates; and 2) the effect of cAMP-dependent protein kinase (PKA) inhibitors, myristoylated protein kinase inhibitor (myr-PKI) and KT-5720, on isoproterenol- and forskolin-stimulated cAMP production in bladder from old rats.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Polyclonal antisera directed against -subunits of Gs and Go were purchased from DuPont-New England Nuclear (Boston, MA). Polyclonal antisera directed against Gi (Gi3 > Gi1 or Gi2) was purchased from Upstate Biotechnology (Lake Placid, NY). Gradient gels and protein reagent were obtained from Bio-Rad (Hercules, CA). Rainbow high-molecular-weight markers, enhanced chemiluminescence (ECL) kits, and nitrocellulose membranes were purchased from Amersham International (Buckinghamshire, UK). [³²P]NAD⁺ was purchased from DuPont-New England Nuclear. PTX, ATP, dithiothreitol (DTT), protease inhibitors, thymidine, and NAD⁺ were obtained from Sigma Chemical Co. (St. Louis, MO). KT-5720 and myristoylated PKI amide (n-myristoyl-Gly-Arg-Thr-Gly-Arg-Arg-Asn-Ala-Ile-NH₂) were purchased from BIOMOL (Plymouth Meeting, PA).

Animals. The procedures described here were approved by the Animal Care Committee of the Yale University School of Medicine. Male Fischer 344 rats were obtained from the contract colonies of the National Institute on Aging (Harlan Sprague-Dawley Inc., Madison, WI). Bladders from male Fischer rats of the following ages and weights were removed immediately to ice-cold PBS: 21-day-old weanling (42.8 ± 2.1 g); 90-day-old young adult (247.0 ± 7.6 g); 6-month-old adult (395.2 ± 12.4 g), and 24-month-old aged (474.5 ± 15.3 g) rats. Heart ventricles were removed to be used as positive control tissue, as they have been well characterized with respect to G-protein content (Michel et al., 1993; Shu and Scarpace, 1994). For adenylyl cyclase and cAMP experiments, bladder domes from young adult (90-day-old, 378 ± 17 g) and old adult (632 ± 29 g) rats were purchased from Charles River (Boston, MA).

Membrane Sample Preparation. After removal of serosa, fat, and mucosa, bladders were placed in PBS at 4°C and minced. Heart ventricle samples from the different age groups were combined. Subsequently, minced bladder and heart ventricle samples were transferred to ice-cold homogenization buffer (1 ml/100 mg of tissue wet wt.) consisting of 20 mM HEPES, pH 7.2, 1 mM DTT; 0.05 mM phenylmethylsulfonyl fluoride; leupeptin (1 mg/500 ml), and 1 mg/100 ml concentration of each of the following: soybean trypsin inhibitor, pepstatin, antipain, and chymostatin (HEPES buffer plus inhibitors). The tissues were homogenized with a Polytron (Brinkmann Instruments, Westbury, NY) using three 10-s pulses at 80% power (4°C). The Polytron probe was cleaned with distilled water between homogenization of different samples. Bladder and heart ventricle tissue homogenates then were centrifuged (4°C, 20,000g, 20 min). The pellet was resuspended in homogenization buffer (1 ml/333 mg of tissue). Subsequently, aliquots were removed from the resultant samples for protein determination. The remainder of the samples was stored at 80°C for ADP-ribosylation or combined with SDS sample buffer and heated (95°C, 10 min) before storage at 20°C for Western blot analysis. Protein concentrations were measured (Bradford, 1976) using bovine -globulin as a standard.

Western Immunoblotting. Levels of immunoreactive -subunits of Gs, Gi, and Go in detrusor and heart ventricle plasma membranes were assessed by Western immunoblot analysis. Plasma membrane samples (40 µg of protein) were subjected to SDS-polyacrylamide gel electrophoresis (Laemmli, 1970) on 4 to 20% gradient gels and then transferred to nitrocellulose membranes. After transfer, the gels were stained with Coomassie blue to ascertain that the protein transfer was complete. After blocking with 5% milk-PBS, the nitrocellulose membrane blots were incubated for 1 h with the primary antibody (1:1000 dilution in 5% milk PBS-Tween 20 for the anti-Gs and anti-Go antibodies; 1:500 dilution for the anti-Gi antibody). Antibody-labeled G-proteins were detected by incubating the blots for 30 min with the rabbit anti-rat Ig conjugated to horseradish peroxidase (1:1000 dilution). The immunoreactive proteins were demonstrated using enhanced chemiluminescence.

ADP-Ribosylation with PTX. PTX-catalyzed ADP-ribosylation of Gi and Go was carried out as described by Kopf and Woolkalis (1991) and Carty (1994). PTX (5 µl, 100 µg/ml in 50% glycerol containing 50 mM Tris, 10 mM glycine, 0.5 M NaCl, pH 7.5) was activated by incubation with 5 µl of 200 mM DTT and 1 µl of 2.5% SDS (20 min, 32°C). BSA (1 mg/ml, 50 µl) was then added to the activated PTX mix. Membranes (10 µg of protein) were then incubated (32°C, 60 min) with a PTX reaction mix containing (total volume 50 µl) 5 µl activated PTX mix or vehicle, 1 mM EDTA, 5 mM DTT, 10 mM thymidine, 5 µM NAD⁺ containing 500,000 cpm [³²P]NAD⁺, 10 mM ATP, and 0.1 mM GTP in HEPES buffer plus protease inhibitors. The reaction was terminated by the addition of 1 ml of ice-cold 10% trichloroacetic acid. After 30 min on ice, samples were centrifuged (14,000g, 10 min). The supernatant was discarded, and 1 ml of ice-cold acetone was added to each sample. Samples were again centrifuged (14,000g, 10 min), and the resultant pellets were resuspended in 20-µl aliquots of HEPES buffer plus inhibitors plus 0.2% Lubrol (ICN Biomedicals Inc., Aurora, OH). SDS sample buffer was added to each resuspended pellet, the mixture was heated (95°C, 10 min), and then samples were subjected to SDS-polyacrylamide gel electrophoresis on 4 to 20% gradient gels. Specific phosphorylation bands were identified on autoradiograms of the dried gels. In addition, ³²P incorporation was quantified by cutting the appropriate bands from the gels and counting them in a scintillation counter. ³²P incorporation in controls (no toxin) was subtracted from the data obtained in the other lanes to obtain toxin-specific ³²P incorporation.

Adenylyl Cyclase Assay. Bladder dome was minced; suspended in 20 mM HEPES, pH 7.4, containing 1 mM EDTA, 0.2 mg/ml BSA, 7.5 µg/ml aprotinin, and a protease inhibitor cocktail (Complete; Boehringer-Mannheim, Mannheim, Germany); and homogenized as described earlier. Vehicle or PTX (5 µM, final), activated with DTT as described above except without SDS, was added to the bladder homogenate containing 0.2 mM NAD⁺, 1 mM ATP, 0.1 mM GTP, 10 mM thymidine, and 1 mM EDTA, and the mixture was incubated for 40 min (30°C). The ribosylation mixture was diluted with homogenization buffer and centrifuged (10,000g, 10 min), and the membranes were suspended in homogenization buffer plus phentolamine (0.5 µM). Adenylyl cyclase activity was then assayed in the membrane fraction (10 min, 30°C) in a reaction mix containing (final concentration) 35 mM Tris, pH 7.4, 1 mM EGTA, 1.5 mM DTT, 0.75 mM 3-isobutyl-1-methylxanthine, and 1 mM ATP with an ATP regenerating system [creatine phosphokinase (25 U/ml), phosphocreatinine (5 mM) and myokinase (150 U/ml)]. Incubations with isoproterenol were done in the presence of GTP (10 µM). The reaction was stopped by heating (70°C, 3 min), and cAMP concentration was measured in the supernatant with a radioimmunoassay (Biomedical Technologies, Inc., Stoughton, MA).

cAMP Measurement. Bladder dome muscle is minced and incubated for 30 min at 37°C in Krebs-Ringer-HEPES (K-H) solution containing 108 mM NaCl, 4.7 mM KCl, 1.3 mM CaCl₂, 1.0 mM MgCl₂, 20.0 mM NaHCO₃, 0.8 mM Na₂HPO₄, 0.4 mM NaH₂PO₄, 20.0 mM HEPES, and 10.0 mM dextrose, oxygenated with 95% O₂, 5% CO₂. The minced bladder muscle was then resuspended in fresh K-H containing dimethyl sulfoxide or PKA inhibitors and incubated for 30 min at 37°C. Aliquots of the muscles in K-H buffer or K-H buffer plus PKA inhibitors (200 µl) were then incubated with isoproterenol (10 µM) for 1 min or forskolin for 10 min. The reaction was stopped with ice-cold 1 N PCA (200 µl), 1 ml of sodium acetate buffer was added, and the samples were homogenized and centrifuged (9000g, 20 min, 4°C). The pellet was hydrolyzed in 1 N NaOH for protein assay. cAMP was quantified in the supernatant as described earlier after the supernatant was neutralized with 3 N KOH and centrifuged to remove the salt precipitate.

Data Analysis. Intensity (absorbance × area) of labeled bands (45 and 52 kDa for Gs, 40-41 kDa for Gi, and 39 kDa for Go) was quantified by densitometry using NIH Image 1.47 (National Institutes of Health) software. All results (except for the positive controls) were measured as percent optical intensity relative to 21-day-old levels, which were assigned the value of 100%. Results from multiple experiments were presented as mean value ± S.E. Statistical comparisons between age groups and concentrations were performed with ANOVA and the Scheffé F test (P < .05).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Western Blot Analysis. The bands detected in all bladder immunoblots for Gs, Gi, and Go were of identical molecular weight and of equal or greater band intensity than that of heart ventricle preparations. After Western immunoblot analysis for Gs (Figs. 1 and 2), the relative intensity of bands for 21-day-, 90-day-, 6-month-, and 24-month-old rat bladder was 100.0, 60.8 ± 2.8, 98.0 ± 7.4, and 108.8 ± 9.1%, respectively. Gs intensity for 90-day-old rats was significantly smaller than that for the 21-day-old rats. Relative intensity values for Gs in 6- and 24-month-old rats were not significantly different from the value for the 21-day-old. After Western immunoblot analysis for Go (Figs. 1 and 2), the relative intensity of bands for 21-day-, 90-day-, 6-month-, and 24-month-old rats was 100.0, 33.8 ± 3.7, 72.9 ± 4.8, and 109.3 ± 4.6%, respectively.

View larger version (82K): [in this window] [in a new window]   Fig. 1.   Representative immunoblots showing presence of Gi, Go, and Gs in aging rat bladder.

View larger version (28K): [in this window] [in a new window]   Fig. 2.   Analysis of immunoblots for Gi, Go, and Gs in aging rat bladder. Results are analyzed from three rats in each age group. *, significant differences from 21-day-old results (P < .05).

ADP-Ribosylation. With PTX-catalyzed ADP-ribosylation, a single band with a molecular weight of 41 to 42 kDa was ribosylated in the bladders of rats from all ages tested (Fig. 3A). The values for ADP-ribosylation for 21-day-, 90-day-, 6-month-, and 24-month-old rats were 9175 ± 1325, 4912 ± 1757, 11,525 ± 2825 and 22,512 ± 1157 cpm/mg of protein, respectively. ADP-ribosylation in 90-day-old rats was significantly smaller than that for the 21-day-old rat bladders, whereas ADP-ribosylation for the 24-month-old rats was significantly greater than values (cpm/mg of protein) of 21-day-, 90-day-, and 6-month-old rats (Fig. 3B).

View larger version (18K): [in this window] [in a new window]   Fig. 3.   A, representative autoradiogram showing the single band that was ADP-ribosylated by [32P]NAD+ in the presence of PTX. B, analysis of ADP-ribosylation with PTX in aging rat bladder membranes. Results are analyzed by counting the single gel band at approximately 41 kDa (n = 3 rat bladders in each group). *, significant differences from 21-day-old results (P < .05).

Effect of PTX Treatment on Adenylyl Cyclase Activity. Homogenates from young adult and old adult rat bladders were treated with either vehicle or PTX, and then adenylyl cyclase activity was assayed.

View this table: [in this window] [in a new window]   TABLE 1 Effect of PTX on GTP-isoproterenol-, and AlF4-stimulated adenylyl cyclase in bladder particulates from young adult (n = 3) and old (n = 4) rat All determinations contain 10 µM GTP and were performed in duplicate. Values are mean ± S.E. from duplicate measurements.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Effect of isoproterenol on adenylyl cyclase activity from control and PTX-treated bladder dome membranes from old adult rats. Results are analyzed from four rats, with each concentration in duplicate. Curves are significantly different by ANOVA. *, significant differences from basal (GTP alone) values (P < .05).

Effect of PKA Inhibitors on cAMP Accumulation. After incubation with either vehicle (dimethyl sulfoxide) or PKA inhibitors myr-PKI (5 µM) or KT-5720 (3 µM) in the presence of phentolamine (0.5 µM) for 30 min, cAMP accumulation was measured in bladder dome muscle obtained from old adult rats. Phosphodiesterase inhibitors were excluded to assay the effect of feedback mechanisms on cAMP accumulation. Agonist incubation times and concentrations had been previously determined to produce significant accumulations of cAMP (isoproterenol, 1 min; forskolin, 10 min). Isoproterenol (10 µM) induced cAMP accumulation was significantly increased by incubation with myr-PKI and KT-5720 (Table 2). Forskolin (3 µM)-induced cAMP accumulation was not altered significantly by either Myr-PKI or KT-5720 (Table 2).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Previously, we showed that the age-dependent decreases in -adrenergic relaxation response to both isoproterenol and norepinephrine in rat bladder were not due to postadenylyl cyclase modifications, because there were no age-dependent differences in relaxation to dibutyryl cAMP (Nishimoto et al., 1995). Furthermore, the age-dependent decreases in relaxation were probably not due to changes in -receptor number or function because although receptor density decreased with aging, there was no decrease in the number of -receptors per bladder or in the K_D values.

Isoproterenol does not increase adenylyl cyclase activity in the 22-month-old rat bladder but produces increases in adenylyl cyclase activity in the 21- and 90-day-old rat bladder (Wheeler et al., 1990). Catalytic adenylyl cyclase activity as measured by forskolin-induced adenylyl cyclase activity also decreases with aging, when 21-day-old rat bladder is compared with 90-day- and 22-month-old rat bladder. However, activation of adenylyl cyclase by Gpp(NH)p, a nonhydrolyzable GTP analog or AlF₄, G-protein activators, increases in the aged rat bladder compared with both the 21- and 90-day-old bladder (Wheeler et al., 1990). Because isoproterenol-, but not Gs-, induced adenylyl cyclase decreases with aging and because activated -adrenergic receptors can activate both Gs and Gi (Asano et al., 1984; Raymond, 1995; Lefkowitz, 1998), one locus of the age-dependent changes in contractile response may be Gi.

Previously, Gi1,2,3; Gq; and to a lesser extent Go and Gz, the neuronally localized PTX-insensitive G-protein (Fields and Casey, 1997), were described in rat bladder (Wang et al., 1995). We found Gs content is similar in the 21-day-, 6-month-, and 24-month-old rat bladder but lower in the 90-day-old rat bladder. Gi content, as well as PTX-catalyzed ADP-ribosylation, is higher in 6- and 24-month-old detrusor than in the 21- and 90-day-old detrusor. Go content is lower in the 90-day- and 6-month-old rat bladder compared with the 21-day- and 24-month-old bladders, which are equivalent.

The decreased ability of isoproterenol to increase adenylyl cyclase activity and to relax the detrusor with age appears be due to an increase in the functional activity of PTX-sensitive G-proteins. Therefore, the effects of PTX-catalyzed ADP-ribosylation on adenylyl cyclase activity were compared in bladders from 90-day- and 22-month-old rats. When rat bladders from 90-day- and 22-month-old rats are compared, norepinephrine-, isoproterenol-, and forskolin-induced relaxation responses, along with isoproterenol-stimulated adenylyl cyclase activity, are higher in the 90-day-old than in the 22-month-old Fischer rat bladder. On the other hand, Gpp(NH)p-stimulated adenylyl cyclase (Wheeler et al., 1990; Nishimoto et al., 1995); Gs, Go, and Gi immunoreactivity, and PTX-sensitive ADP-ribosylation are significantly lower when 90-day-old bladders are compared with 22-month-old bladders. Although PTX increases isoproterenol stimulation of adenylyl cyclase activity in membranes from young adult rats, it also increases GTP-stimulated adenylyl cyclase activity. Thus, in PTX-treated young adult membranes, isoproterenol does not increase adenylyl cyclase activity compared with that resulting from GTP stimulation. In bladder membranes from old adult rats, isoproterenol increases adenylyl cyclase significantly after PTX treatment, to an extent similar to the isoproterenol-induced increase in adenylyl cyclase activity previously noted in bladder from younger rats (Wheeler et al., 1990). The effect of PTX on contraction and relaxation response in young and old adult rats may clarify the age-dependent differences seen in GTP-stimulated adenylyl cyclase activity.

PTX rescue of -responsiveness has been noted in murine cardiac myocytes where ₂-receptors, but not ₁-receptors, couple to Gi (Xiao et al., 1999a). Although chronic heart failure in humans and animal models is associated with marked increases in Gi (Xiao et al., 1999b), the age-associated decrease in -responsiveness in the heart is not due to an increase in Gi and PTX does not reverse this diminution (Xiao et al., 1998). On the other hand, atopic sensitized airway smooth muscle has increased Gi expression and an attenuated relaxation response to isoproterenol. This attenuated response to isoproterenol is reversed by ADP-ribosylation with PTX (Hakonarson et al., 1995). Although there are multiple mechanisms that decrease -responsiveness with aging, the increases in Gi content and function in the bladder probably account for most of the decrease in -responsiveness with age.

The ₁-, ₂-, and ₃-adrenoceptors have been identified in rat bladder (Fujimura et al., 1999), and all three subtypes have been shown to mediate relaxation (Longhurst and Levendusky, 1999). Traditionally, agonist occupancy of the -receptor leads to activation of adenylyl cyclase and of PKA. PKA phosphorylation of the -receptor rapidly (seconds to minutes) may uncouple it from Gs and facilitates its coupling to Gi (Daaka et al., 1997; Lefkowitz, 1998). Increases in Gi seen in aged rat bladder may enhance adenylyl cyclase inhibition by increasing the rate or the amount of phosphorylation of the -receptor. Although isoproterenol-induced cAMP accumulation and relaxation response are decreased in bladder from aged rats compared with younger rats, it still is measurable. Therefore, isoproterenol-induced cAMP accumulation was measured after incubation with PKA inhibitors myr-PKI, a cell-permeable heat-stable PKI peptide sequence, and KT-5720, a hexyl derivative of K-252a, using agonist incubation times similar to those that produce bladder relaxation. Treatment with myr-PKI (5 µM) and KT-5720 (3 µM) increases the isoproterenol-induced increase in cAMP content in bladder from old adult rats, indicating that phosphorylation of the -receptor through Gi may be a possible mechanism for the age-dependent decline in isoproterenol-induced relaxation response and adenylyl cyclase activity (Wheeler et al., 1990; Nishimoto et al., 1995).

In the detrusor, the contractile response to cholinergic agonists is mediated mainly through M3 muscarinic receptors (Mutoh et al., 1997) via phosphatidyl inositol hydrolysis (Noronha-Blob et al., 1989; Wheeler et al., 1995) and presumably the G-protein Gq. Although contractile force is mediated mainly through M3 receptors, by far the most abundant of the muscarinic receptors in the bladder detrusor are the M2 receptors. The ratio of M2/M3 receptors in rat bladder is 9:1 (Wang et al., 1995). Additionally, activation of muscarinic receptors inhibits adenylyl cyclase in the guinea pig detrusor (Wheeler et al., 1995), a process that occurs via M2 receptors (Noronha-Blob et al., 1989). Recent data suggest that activation of M2 receptors also can cause contraction of the rat bladder in vitro and may mediate reflex- or volume-induced bladder contractions in vivo. The proposed mechanism for M2 receptor activation involves the reversal of -adrenergic-mediated relaxation (Hedge et al., 1997). In the case of the aging bladder where Gi is increased, M2 muscarinic activation of Gi may cause a decrease in -adrenergic-induced relaxation and may be a factor in incontinence, decreased bladder capacity, and the inability to postpone voiding that is observed with aging.

Alterations in G-proteins that lead to loss or gain of function are documented both in aging and in diseases, including hypertension (Feldman et al., 1995; Siffert et al., 1995), diabetes mellitus (Moxham and Malbon, 1996), pseudohypoparathyroidism, and some ovarian and adrenocortical tumors (Spiegel, 1997). In diabetes and ovarian and adrenocortical tumors, Gi appears to be increased. In lymphocytes isolated from older patients, increases in Gi but not Gs ADP-ribosylation are observed, and in immortalized lymphoblasts obtained from hypertensive patients, there is enhanced signal transduction and cell proliferation that are blocked by PTX. (Siffert et al., 1995). In the aging liver, the amount of Gs is increased along with adrenergic stimulation of adenylyl cyclase (Eakes et al., 1996).

Age-dependent decreases in -adrenergic-induced relaxation of rat detrusor are coupled to the large increases in Gi but not Go or Gs. Increased Gi activity may decrease the response to isoproterenol through phosphorylation of the -receptor, because treatment with PTX or PKA inhibitors reverses the age-dependent decline in isoproterenol-responsive adenylyl cyclase activity. Increases in Gi, are seen not only in aging but also in diabetes and hypertension and may have consequences in the voiding disorders prevalent in the older population.