cAMP signaling mechanisms with aging in rats

mtdmimsofagaing

andddopnmnt

Mechanisms of Ageing and Development 86 (1996) ll--26

CAMP signaling mechanisms with aging in rats Jane H. Chin, Anand N. Hiremath,

Brian B. Hoffman*

Depurtment 01’ Medicine, Stcmfbrd Lhiversity School of‘ Medicine und Geriatric Research. Education and Clinical Center. Veterans Aflhirs Medical Center. lK?B, 3801 Miranda Avenue. Palo Alto, CA 94304, USA

Received 10 July 1995; revised 6 October 1995; accepted 9 October 1995

Abstract Blunted CAMP responses to /?-adrenergic agonists play a major role in diminished smooth muscle relaxation in blood vessels from older animals, although the mechanisms remain uncertain. A diminished CAMP response could potentially arise from changes in the expression of adenylyl cylcase-coupled G proteins, such as a diminished expression of G, or an increased expression of G,. We tested the hypothesis that a loss in G, or increased expression of Gi could occur in tissues such as the aorta, heart and kidney with aging, which would provide a unifying explanation for blunted CAMP responses to many hormones with aging in a variety of cells. Using Western blotting with specific antibodies, we found no generalized changes in G protein expression with aging. Also, injection of pertussis toxin (which functionally inactivates Gi) into older animals did not restore vascular relaxation mediated by p-adrenergic receptors. We previously found an elevated ratio of regulatory to catalytic subunits of protein kinase A in the aorta of older rats, which would tend to impair activation of the catalytic unit; this alteration was not generalized to other organs such as the heart and kidney. Old rats fed a low salt diet did not show the restored /?-adrenergic agonist-induced vasodilation previously found in elderly humans, suggesting that there are species differences in the development of this deficit. Altogether, these results suggest that altered G protein expression does not provide a general explanation for blunted activation of adenylyl cyclase with aging. Keywords: Aging: G proteins; Vascular relaxation;

B-adrenergic receptors

* Corresponding author. Present address, VA Medical Center, 182B, 3801 Miranda Avenue. Alto.

CA 94304, USA. Tel.:

0047-6374/96/%15.00 SSDI

+ 1 415 8583933:

0 1996 Elsevier

0047-6374(95)01676-Q

Fax:

Science Ireland

+ 1 415 8559437.

Ltd. All rights

reserved

Palo

12

J.H.

Chin et al. / Mechanisms of’ Ageing and Development 86 (1996) II-26

1. Introduction

There is substantial evidence indicating that responsiveness to many drugs and hormones is diminished with increasing age. In particular, responses to many hormones that activate adenylyl cyclase have been shown to be attenuated with age in a variety of tissues. For example, responses to j’-adrenergic agonists, which activate adenylyl cyclase via G-proteins, are attenuated in the elderly, which leads to modification of responses to stress and exercise [I]. In addition, responsiveness to parathyroid hormone, which also activates adenylyl cyclase, is impaired in the kidneys of older rats [2,3]. It is not known whether there are common underlying mechanisms that are responsible for both the decline in the function of peptide hormones and catecholamine effects mediated by adenylyl cyclase. Previous studies have shown that p -adrenergic agonist-induced vascular relaxation is blunted with age in humans progressively across the age range from 20 to 80 years [4] and in rats [5-71. p- adrenergic agonist-stimulated CAMP accumulation is attenuated in blood vessels from older animals. These changes involve both decreased total CAMP accumulation [5,6] and CAMP bound to the regulatory subunit of protein kinase A [8]. The impaired vascular relaxation and CAMP accumulation by P-adrenergic agonists in vessels from older animals also occurs in the presence of phosphodiesterase inhibitors [5.9, lo], suggesting that the reduction in CAMP accumulation is not due to enhanced degradation, but rather reflects a decreased rate of synthesis of CAMP. In contrast, forskolin, which activates adenylyl cyclase independently of cell surface receptors, induces full relaxation in vessels from older animals [6,9]. In addition, CAMP accumulation (both maximal total CAMP accumulation [5,6] and bound CAMP accumulation [8]) are increased to similar extents in old and young aortas. Also, forskolin-induced protein kinase A activation is similar in the vessels of young and old animals [6]. Since the density of p-receptors was not different between mesenteric arteries from young and old animals [5], and the intact forskolin responses suggest the availability of sufficient isoforms of adenylyl cyclase sensitive to forskolin, the deficit in P-adrenergic stimulation of CAMP may involve alterations in receptor coupling to G proteins. Many cell surface receptors act through specific guanine nucleotide regulatory proteins (G proteins) which transduce signals from the receptors to various effecters. The G proteins are part of a large superfamily of proteins regulated by guanine nucleotides and function as on-off switches for cellular signaling. Ligands, such as isoproterenol, bind to p-adrenergic receptors which activate stimulatory heterotrimeric G proteins termed G,. Activation of G, leads to dissociation of CI, from the PI> subunits, with activation of adenylyl cyclase by CI,. Inhibition of adenylyl cyclase is mediated by G proteins termed Gi. G, typically inhibit adenylyl cyclase when activated by ‘inhibitory’ receptors, such as the a, adrenergic or the m2 muscarinic cholinergic receptors. The CIsubunit of Gi is a substrate for ADP-ribosylation by pertussis toxin, which impairs activation of this signaling pathway by agonists. On the other hand, the CIsubunits of G, are substrates for ADP-ribosylation by cholera toxin, which irreversibly activates the G, capacity to enhance adenylyl cyclase activity [ 11,121.

J.H.

Chin et al. :/Mechanisms of Aping and Del~elopment 86 (1996) II --26

13

In the present paper, we have carried out studies to test several possibilities for the blunted CAMP response to ,!?-adrenergic agonists. Experiments were done to investigate the possibility that a change in the expression of G, or Gi could contribute to the blunted response to P-adrenergic agonists in blood vessels. To test whether enhanced effects of Gi could also contribute to impaired activation of adenylyl cyclase in vascular smooth muscle, we injected pertussis toxin into older rats and determined B-receptor activation of smooth muscle relaxation. In addition, we have tested the hypothesis that there might be some general unifying mechanisms that could account for the alterations in response to a variety of hormones that occur with aging. Specifically, we have explored whether a loss in G,, or increased expression of G,, could occur in other tissues, such as heart or kidney, with aging would provide an underlying explanation for the blunted responses to many hormones with aging. Also, we have previously found that the expression of the regulatory subunit of CAMP-dependent protein kinase (protein kinase A) is increased relative to the expression of the catalytic activity of the enzyme in the aortas from older rats compared to young rats [8]. Such a change could result in impaired activation of the kinase in response to a rise in CAMP accumulation. In the present study, we have tested whether a similar age-related change in the expression of the subunits of protein kinase A occurs in other tissues, potentially contributing to the blunted physiological responses to hormones that activate CAMP accumulation. In addition, Feldman [13] found that elderly humans fed a low salt diet for 4 days have restored fi-adrenergic agonist-induced vasodilation. To determine whether this interesting effect occurs also in older rats, rats were fed a low salt diet for 4 days, or 4 weeks, and vascular relaxation to isoproterenol was determined in aortic rings from these animals.

2. Materials and methods 2.1. Materials

[3H]cAMP, [y-32P]ATP were purchased from Amersham (Arlington Heights, IL). Cellulose ester filters (HAWP, 0.45 mm) were obtained from Millipore (Bedford, MA). Protein kinase inhibitor from bovine heart (type II, P-8140), kemptide, histone type IIA, phenylephrine, ( -) isoproterenol, and histamine were obtained from Sigma Chemical Co. (St Louis, MO). Primary antibody RMjl (for G,) and AS/7 (for Gi, and G,,) were generous gifts from Dr. Alan Spiegel (Bethesda, MD). Biotinylated goat anti-rabbit IgG, alkaline phosphatase-streptavidin conjugate, 5-bromo-4-chloro-3-indolyl phosphate (BCIP), and nitroblue tetrazolium (NBT) were purchased from Zymed Laboratories Inc. (South San Francisco, CA). 2.2. Methods 2.2.1. Pertussis

toxin

In some experiments Sprague-Dawley male rats (2 months or 9-12 months) were

14

J.H.

Chit? et al. / Mechanisms of‘ Ageing und Development 86 (1996) II-26

injected intraperitoneally with 3 18 OPU of pertussis vaccine (Lederle-Praxis Biologicals, Pearl River, NY). Four days later the animals were decapitated and aortic rings were used to measure relaxation induced by isoproterenol after contraction with phenylephrine. 2.2.2. Low salt diet In some experiments Sprague-Dawley male rats (9912 months) were fed either a low-Salt diet (modified LM-485) containing 0.06% NaCl or a regular-Salt diet (LM-485) containing 0.31% NaCl (purchased from Teklad, Madison, WI). After 4 days, or 4 weeks, on the diets, isoproterenol-induced relaxation in isolated aortic rings was measured. 2.2.3. Measurement of aortic relaxation Thoracic aortas were removed from decapitated male Sprague-Dawley, or Fischer 344, rats ranging in age from 6 weeks to 16 months. After dissecting the surrounding connective tissue, aortic rings (3-4 mm) were mounted in tissue bath chambers (20 ml) containing modified Krebs solution (in mmol/l: NaCl 118, KC1 4.7, NaHCO, 24.9, glucose 10, CaCl, 2.5, MgSO, 1.2, and KH,PO, 1.2). The tissue was aerated with 95% O,-5% CO, at 37°C. Rings were equilibrated for 90 min before exposure to drugs. The isometric force of contraction was measured with force-displacement transducers and was recorded on a Grass Model 7D polygraph. Relaxation responses were obtained by adding cumulative doses of isoproterenol (10-9-10~5 M), or histamine (1O-9-lO-5 M), to vessels contracted with 3 x lop6 M of phenylephrine, which elicited 80% of the maximum contraction as determined from concentration response curves for phenylephrine (10 -.9- 10 - ’ M). 2.2.4. Membrane preparations Thoracic aorta, heart and kidney were removed, cleaned of adherent connective tissue and were frozen, or used, immediately. All tissues were homogenized in cold Tris-sucrose buffer containing a cocktail of protease inhibitors (20 mM Tris HCl, 2mM EDTA, 2.5 mM EGTA, 0.25 M sucrose, 50 mM mercaptoethanol, 4 pg/ml leupeptin, 10 pg/ml aprotinin, and 4 pg/ml antipain, pH 7.4). The total suspension was centrifuged at 30000 x g for 10 min at 4”C, and the pellet containing the crude membranes was solubilized in 1% Triton X-100 for 1 h at 4°C. Protein content was determined by the method of Peterson [14]. 2.2.5. Western immunoblotting of G proteins Triton extracted membrane proteins (lo-40 pg) were separated by polyacrylamide gel electrophoresis (12% precast acrylamide gel) using a mini-electrophoresis system. Protein bands were electrophoretically transferred to an Immobilon P membrane (Millipore Corp., Bedford, MA) in CAPS [3-(cyclohexylamino)-lpropanesulphonic acid] buffer, pH 11.O, containing 10% methanol for 2 h at 300 mA in a Transblot apparatus (Bio-Rad Laboratories, Richmond, CA). Non-specific

J.H. Chin et al. ! Mechanisms

of‘ Ageing and Deewlopment

86 (1996) I1 -26

15

binding sites on the Immobilon blot were blocked for 12-14 h by 3% BSA in buffer A (25 mM Tris-HCl and 500 mM NaCl) at 37°C. The blots were washed 3 times in 1% BSA in Buffer A, and then incubated for 12- 14 h at 22°C with specific rabbit antiserum (RM/I to recognize G,, and AS/7 to recognize Gil,, Gi21, or transducin a) diluted 1:500 to I:2000 in 1% BSA in buffer A. The filters were washed 3 times, and then incubated for 1 h at 22°C with the secondary antibody, biotinylated goat anti-rabbit IgG diluted 1:250 in Tris-buffered saline (25 mM Tris-HCl and 137 mM NaCl, pH 7.6) containing 0.05% Tween 20. After 4 washes, the immunoreactive regions were detected enzymatically using an alkaline phosphatase-strepavidin conjugate and the substrates BCIP and NBT. The immunodetected bands were quantified by laser densitometric scanning of the filters directly (Molecular Dynamics, Sunnyvale, CA or Pharmacia LKB, Piscataway, NJ). For each G protein band, the densitometer area was expressed relative to that of a brain standard used as an internal control in each experiment. The mean ratio associated with the data from 6 week old animals (young) was given the relative value of 100%.

2.2.6. Measurement

of CAMP

binding to regulatory

subunits

und CAMP-dependent

kinase activity

The number of CAMP binding regulatory subunits and the total amount of CAMP-dependent kinase activity were determined in homogenates from heart and kidney, as previously described for the aorta [S&The total CAMP bound to the regulatory subunit of the CAMP-dependent protein kinase was isolated by adsorption to cellulose filters, and then eluted from the filters by boiling phosphatebuffered saline containing 1 mM theophylline. Homogenates from the heart and kidney were added to binding buffer (50 mM phosphate buffer, pH 6.8, 10 mM theophylline, 10 mM magnesium acetate, 2 M sodium chloride, 1 mM isobutylmethyl xanthine, 10 mM mercaptoethanol, 0.5 mg/ml histone type IIA, and 1 mg/ml bovine serum albumin), in the presence of 200 nM CAMP, to determine the total number of CAMP binding sites. After incubation for 90 min at 4°C the suspension was added directly to presoaked HAWP cellulose filters, and the eluted total bound CAMP was adjusted to pH 3 and analyzed by radioimmunoassay. Protein kinase A catalytic activity was measured by the incorporation of phosphate into the specific peptide substrate kemptide in reaction mixtures containing 14 mM KH,PO, (pH 6.8), 5 mM magnesium acetate, 12.5 mM sodium fluoride, 50 ,uM kemptide, 0.275 mM [Y-~~P]ATP (60 dpmipmol) and either 1.5 PM CAMP, or CAMP-dependent protein kinase inhibitor (0.6 mg/ml). The reaction was initiated by the addition of 20 ,~l of the heart or kidney homogenate that had been previously forced through a 25-gauge needle. The mixture was incubated for 5 min at 30°C and terminated by transfer to phosphocellulose paper [15]. Protein kinase A activity was obtained by subtracting the activity in the presence of the inhibitor of CAMP-dependent protein kinase from that in the presence of added CAMP. The specific kinase A activity in the heart and kidney was linear with the amount of protein up to 200 pg.

16

J.H.

Chin et al. / Mechanisms

af Ageing and Development 86 (1996) 11-26

3. Results 3.1. Aortic

smooth muscle relaxation

To measure relaxation in aortic rings, the vessels were initially contracted with phenylephrine (3 x 10W6 M), which elicited approximately 80% maximal contraction. Concentration response curves for isoproterenol-induced relaxation were then constructed in aortic rings from rats varying in age from 6 weeks to 16 months old. As expected, isoproterenol-mediated relaxation was markedly impaired in the aortas from older Fischer rats (Fig. 1, left), and these results were similar in the aortas from Sprague-Dawley rats (data not shown). In contrast, there was no differ&e in maximal relaxation induced by histamine between adult and old rats (Fig, 1, right). In rat aorta, histamine induces vascular relaxation via activation of H, histamine receptors, which release end&elk4 cdl derived relaxing factor (most likely nitric oxide) from endothelial cells, which stimulate vascular relaxation independently of CAMP mediated mechanisms [ 16,171.

ISOPROTERENOL

0

1.5

4

8 16 AGE (months)

HISTAMINE

8

16

Fig. 1. Maximal relaxation induced by isaproterenol and histamine in aortas from Fischer 344 rats ranging in age from 1.5 to 16 months old. Vessels were initially contracted with phenylephrine, as described m Methods. Relaxation was induced by maximally effective concentrations of isoproterenol (IO- 5 M) or histamine ji0- ’ M). Bars represent the mean f SEM of relaxation in separak rings from 4, 4, 3, 9, 2, and 9 animals at 1.5,4, 8, 16 months for isoproterenol and 8 and 16 months for histamine, respectively: *P < 0.0s compared to relaxation m 1.5 month rats.

J.N. Chin er al. ! Mechanisms of Ageing and Development 86 (1996) 1I-26

a

YOUNG

[XXI

OLDER

17

Fig. 2. Effect of pertussis vaccine on isoproterenol-induced relaxation in aortas from young (2 months) and older (9- 12 months) Sprague-Dawley rats. Rats of both ages were injected intraperitoneally with either pertussis vaccine (318 OPU) (PT + ISO). or saline (SO). Four days later, relaxation of the aortic segments was measured to maximally effective concentrations of isoproterenol ( 10W5 M). Relaxation was markedly attenuated in aortas from older rats compared to that from young rats, but pertussis vaccine did not reverse the blunted response in the older animals; **P < 0.01 compared to relaxation in young rats.

3.2. Measurement

of G proteins by Western blotting

To test whether increased effects of Gi proteins were involved in impaired /?-adrenergic receptor relaxation of the aorta, rats were injected intraperitoneally with pertussis vaccine, which inactivates Gi, and p -adrenergic receptor-mediated smooth muscle relaxation was determined 4 days later. We have previously shown that this treatment of rats inactivates the Gi function in adipocytes [18]. The maximal isoproterenol-induced relaxation from pertussis-treated animals was similar to that of controls with 4-5 times greater relaxation in aortas from younger animals (3-4 months) than older animals (lo-- 12 months) (Fig. 2). These results do not support the hypothesis that enhanced activity of Gi is responsible for inhibiting p -adrenergic receptor-mediated vascular relaxation in aortas from older animals. To determine whether the mass of G, or Gi was altered with aging, we measured G proteins by Western blotting using specific rabbit antiserum. Laser densitometric measurements showed no age-related differences in the expression of either G, or Gi proteins in aorta membranes from Fischer or Sprague-Dawley rats, suggesting that changes in the amount of tl subunits of G, or Gi do not play a major role in the blunted responses to P-agonists (Fig. 3).

18

J.H.

Chin et al. / Mechanisms

of Ageing and Development

M W) Gs

12

86 (1996) II-26

Gi 34

66,

(a)

Gsh G.5l Gi (b)

Fischer

G

sh

Gal G.1

Sprague-Dawley

Fig. 3. Western blot analysis of aortic G proteins obtained from young (5-6 weeks old) and older (9- 12 months old) Fischer 344 and Sprague-Dawley rats. 20 pg of protein was loaded per lane for G, and 40 yg for Gi detection. (a) A representative Western blot is shown for Fischer rats. Lanes I and 3 are aortic samples from a young rat and lanes 2 and 4 from an older rat. (b) Data were quantitated by laser densitometry from Western blots as described in Methods, The mean data associated with young animals for each G- protein band was assigned the relative value of 100%. G,, and G,, refer to the higher and lower band for G,. Each bar is the mean + SEM for 12 Fischer rats and 6 Sprague-Dawley rats for each age group.

We then tested the hypothesis that changes in the CAMP system that occur in aorta with aging may be a more general mechanism that could occur in a variety of organs, and which may account for the blunted responses that is seen with aging

J.H.

Chin et al.

/ Mechanisms of‘ Ageing and Development 86 (1996) II-26

19

in response to catecholamines and some peptide hormones that activate adenylyl cyclase pathways. G proteins were quantified in membrane preparations from heart and kidney obtained from Fischer rats at the ages of 6 weeks, 4 months, 8 months and 16 months using Western blotting, as above. No age-related differences were observed in the heart or kidney for the OLsubunits of G, or G, (Fig. 4 and Fig. 5). 3.3. Proportion of regulatory and catalytic moieties of CAMP-dependent protein kinase in heart and kidney We tested whether the increase in the proportion of regulatory and catalytic components of protein kinase A observed in the aorta [8] also occurs in the heart and kidney. Such a change could result in impaired activation of kinase A and blunted physiological responses to hormones that stimulate CAMP accumulation. The expression of the regulatory subunit, as measured by the total number of CAMP binding sites, and the catalytic activity have been determined in three age groups. No significant age-related changes were observed in either the regulatory or catalytic subunits of the heart or kidney (Table 1). 3.4. Lou salt diet Salt restriction has been shown to restore impaired adrenergic responses in blood vessels and lymphocytes in normal elderly humans [13]. To determine if a low salt diet could restore /3-adrenergic relaxation in vessels from older rats, rats were fed a low salt diet containing 0.06% NaCl, or the regular salt diet containing 0.31% NaCl, for either 4 days, or 4 weeks. No restoration of relaxation responses to isoproterenol was observed in the aortas from older rats fed a low salt diet (Fig. 6).

4. Discussion

p -adrenergic mediated responsiveness decreases with age in the cardiovascular system in both humans and laboratory animals [1,7]. We have confirmed that relaxation to the /3-adrenergic agonist isoproterenol, which utilizes the adenylyl cyclase pathway, is impaired in aortas from old rats. In contrast, relaxation to histamine, which acts independently of CAMP and is dependent upon an intact endothelium for the release of endothelium-derived relaxing factor (EDRF), was not different between vessels from young and old animals. These data suggest an age-related deficit specifically in the P-adrenergic signal transduction pathway, rather than a non-specific general inability of older vessels to relax. Our previous results have shown that diminished a-adrenergic stimulation of CAMP accumulation in rat aorta is a rate-limiting explanation for the loss in relaxation with aging in this vessel [6]. Consequently, to determine the potential role of G proteins in explaining the blunted CAMP response in the vascular system, we measured the amount of stimulatory G, and Gi in aortas from young and older rats by Western blotting. We found no age-related change in the amount of CIsubunits

20

J.H.

Chin et al. / Mechanisms qf’ Ageing and Development 86 (1996) II-26

MW Gi

Gs

1

2

3

4

5

6

7

8

66,

HEART

150

100

50

0

(b)

G sh

Gi

Fig. 4. Western blot analysis of heart G proteins obtained from 6 week old, and 4. 8, and 16 month old Fischer 344 rats. 15 pg of protein were loaded per lane for G, detection and 30 pg of protein for Gi detection. (a) A representative Western blot is shown for Fischer rats. Lanes 1 and 5 are heart samples from a 6 week old rat. lanes 2 and 6 from a 4 month old rat, lanes 3 and 7 from an 8 month old rat, and lanes 4 and 8 from a 16 month old rat. G, is the upper band in Lanes 5-8. (b) Data were quantitated by laser densitometry from Western blots as described in Methods. The mean data associated with a young (6 week old) animal for each G protein band was assigned the relative value of 100%. Grh and G,, refer to the higher and lower band for G,. Each bar is the mean + SEM for 4 rats for each age group.

of either G, or Gi. We also found that pertussis toxin, which inactivates Gi, did not reverse the blunted relaxation to isoproterenol in aortas from older animals. To determine whether other potentially common mechanisms underlie age-re-

J.H. Chin et ul.

: Mechanisms of’ Ageing and Derelopment 86 (1996) I1 -26

21

M W) Gi

GS

1

2

3

4

5

6

7

8

66-

31KIDNEY (a)

m 150

m &Gj m

6wk

4 mon 8 mon mon

16

T LOO

2

50

0

a” 0

(b)

G

sh

Gi

Fig. 5. Western blot analysis of kidney G proteins obtained from 6 weeks old and 4, 8, and 16 month old Fischer 344 rats. 20 ,ug of protein were loaded per lane for G, detection and 40 fig for G, detection. (a) A representative Western blot is shown for Fischer rats. Lanes 1 and 5 are kidney samples from a 6 week old rat, lanes 2 and 6 from a 4 month old rat, lanes 3 and 7 from an 8 month old rat, and lanes 4 and 8 from a 16 month old rat. Gi is the upper band in lanes 5-8. (b) Data were quantitated by laser densitometry from Western blots as described in Methods. The mean data associated with a young (6 week old) animal for each G protein band was assigned the relative value of lOO”/o.G,, and G,, refer to the higher and lower band for G,. Each bar is the mean + SEM for 4 rats for each age group.

lated impairment of hormones that act to stimulate adenylyl cyclase in other organs, we measured the amount of G proteins and the proportion of regulatory subunits and catalytic activity of CAMP-dependent protein kinase in heart and

22

J.H. Chin et al. 1 Mechanisms oj’ Ageing and Development 86 (1996) 11-26

Table I Regulatory and catalytical subunits of CAMP-dependent protein kinase in heart and kidney from aging rats Sample

Regulatory units (pmol cAMP/mg protein)

Catalytic Activity (pmol phosphatelmin per mg protein)

Reg/Cat ( x 100)

4.461 & 0.225 3.887 k 0.656 4.509 * 0.200

97 * 37 103 & 58 67 + 24

5.24 i 1.74 5.04 & 2.22 7.75 * 3.01

4.367 f 0.213 4.003 + 1.043 3.193 + 0.289

154269 54+ I4 166&68

3.47 + 1.42 8.44 f 4.07 2.39 k 1.16

HUtY

6 week 9 months I6 months Kidney 6 weeks

9 months I6 months

The total number of CAMP binding sites (regulatory subunits) and CAMP-dependent protein kinase activity (catalytic activity) were assayed as described in Methods. The Reg/Cat is the ratio of the number of regulatory subunits to catalytic activity. The data are expressed as a mean k SEM from 2 separate experiments (assayed in triplicate) for each age group.

kidney. Both of these measurements did not differ with age in heart and kidney. Also, a low salt diet in rats did not reverse the impaired relaxation to isoproterenol as was observed in elderly humans [13]. The diminished capacity of P-adrenergic agonists to stimulate adenylyl cyclase in blood vessels could be due to defects at one site, or multiple sites, along the adenylyl cyclase pathway including the ,8-adrenergic receptor, the catalytic unit of adenylyl cyclase itself, or G proteins which link the receptor to the effector. We found no decrease in the number of /?-receptors in vessels from older animals [5], or in fat cells from old rats [19]. In the hearts from older rats, no loss [20-221 or decrease [23] in p-receptors has been reported. Since vascular relaxation to NECA, an adenosine receptor agonist which also acts via adenylyl cyclase, is impaired in older animals [6], the defect may involve impaired coupling of these receptors to the stimulatory G, protein, a diminished capacity of G, to activate adenylyl cyclase, or a decrease in the amount of specific isoforms of adenylyl cyclase activated by p- or adenosine agonists. The ability of a variety of vasodilators, such as histamine and acetylcholine, to relax vascular smooth muscle is critically dependent upon the release of a relaxing factor, most likely nitric oxide (NO), from endothelial cells [24,25]. NO induces vascular relaxation by activating cytosolic guanylate cyclase in smooth muscle cells, leading to a marked, rapid increase in the intracellular concentration of cGMP in vascular smooth muscle [26]. The effects of maturation and aging on agents causing vasodilation by releasing EDRF from endothelial cells are complex, with reports of decreased, unchanged or increased relaxation in rat and dog blood vessels [7]. Complementary DNAs encoding a family of at least 8 adenylyl cyclases have

J.H.

Chin et al.

;

Mechanisms

qf‘ AgeinK

and Development

86 (1996)

1 I-33

73

been isolated [12,27]. The adenylyl cyclase isoforms expressed in blood vessels are not known; however, types 5 and 6 are the predominant forms found in heart [27]. We found that forskolin-stimulated CAMP accumulation is unaltered in vessels from young and old animals, suggesting sufficient adenylyl cyclase was available in older animals to produce maximal CAMP accumulation by forskolin. In contrast, the total amount of adenylyl cyclase in heart appears to decrease with aging. In heart from 24 month old rats, forskolin-stimulated adenylyl cyclase activity was decreased concomitant with a decrease in the number of forskolin binding sites, a measure of the amount of adenylyl cyclase [21]. The possible change in expression at the protein level of various isoforms of adenylyl cyclase with aging has not been explored and would require the availability of subtype specific antibodies identifying each isoform. Mechanisms for the inhibition of adenylyl cyclase depend upon the type of isoform present. Some isoforms of adenylyl cyclases are especially susceptible to inhibition by the Gi, subunits, others are inhibited in the presence of excess &subunits, which sequesters the G,, subunits, and some, such as types 5 and 6, adenylyl cyclases are apparently inhibited by CAMP-dependent protein kinase [27]. When hearts from fetal, neonatal (445 days), adult (8 week), and mature adult (9 months) rats were compared, mRNA measurements showed an age-related decrease in the content of type 6 adenylyl cyclase, which correlated with a decrease in the

6

CONTROL LOW SALT

20

4

DAYS

4

WEEKS

Fig. 6. Effect of a low salt diet on isoproterenol-induced relaxation in aortas from older rats. Sprague-Dawley rats (9- 12 months old) were fed for either 4 days, or 4 weeks, a low salt diet containing 0.06% NaCl (low salt), or a regular salt diet containing 0.31% NaCl (control). Relaxation of aortic segments from older rats was then measured to maximally etfective concentrations of isoproterenol (10m5 M). The blunted vascular relaxation was not restored by the low salt diet.

24

J.H. Chin et al. / Mechanisms

of Ageing and Development

86 (1996) II -26

catalytic activity induced by isoproterenol or forskolin [23]. Although not tested in blood vessels, it is possible that selective reduction in specific isoforms of adenylyl cyclase that are preferentially sensitive to P-adrenergic agonists may occur with aging, and this could be an area for future investigation. We asked whether there could be a decrease in the expression of G,, or an increase in the expression of Gi, with aging. In the present study, aortas from animals at an age when relaxation and CAMP accumulation were routinely diminished relative to that from younger animals, we found no overall change in the expression of G, or Gi measured by immunoblotting. In contrast, Johnson et al, [28] found a decrease in one of the four bands (42 kD) in G, and a decrease in the Gi and G, proteins in aortas from 24 month old rats by Western blotting, but no age-related change in the level of mRNA for these G proteins. The difference between the studies may be due to the age of the animal, or the concentration of primary antibody, used. The animals we used were younger than those used by Johnson et al. [28], and we used 5 times less primary antibody and detected only 2 bands (42 and 45 kD) instead of 4 bands (38,42,45, and 52 kD) for the G,, component in the aorta. Incubation of rings of aorta from 18 month old rats with cholera toxin, which irreversibly activates G, by covalent modification, resulted in less CAMP accumulation than that found in 2 month old rats [29]. However, it is difficult to interpret this result in terms of potentially significant changes in expression of G,. A diminished rise in CAMP in response to a-adrenergic agonists in vessels from older animals could be due to increased inhibitory effects of the Gi proteins. We previously found an enhanced inhibition of adenylyl cyclase mediated by Aladenosine receptors in adipocytes from older, obese Sprague-Dawley rats [ 181. When Gi was inactivated by pertussis toxin, the lipolytic response in fat cells from the older animals was restored to values seen in younger animals. To test whether the results in adipocytes could be generalized to vascular smooth muscle, we injected pertussis toxin into rats to inactivate G,. The current results indicating that pertussis toxin does not restore vascular relaxation in vessels from old rats, suggest that potential changes in Gi are not the explanation for the blunted relaxation responses to activation of p-adrenergic receptors. The CIsubunit of G, usually exists as 2 forms with apparent molecular weights of 45 and 52 kD, with their relative quantities varying amongst tissues [30]. The antiserum we used recognized both the 45 and 52 kD species in the heart and kidney, but mainly the 45 kD form found in the aorta, as discussed above. We found no differences in the G,, or Gi, subunits in the heart or kidney from old rats when measured by Western blotting. These results confirm those of others who found no age-related change in G, in rat ventricle [21,28] or human heart [31], or Gi in rat ventricle [21,28,32]. However, an increase [22] and decrease [23,32] in G, and Gi in heart have also been reported using immunoblotting. In older explanted human heart, cholera toxin-mediated ADP-ribosylation of G, was decreased, but the amount of G, detected by immunoblotting was unchanged [31]. In kidney from aged rats, both cholera toxin and pertussis toxin labeling were diminished [3] and G,, mRNA was decreased, but Gi,, was increased [33].

J.H.

Chin et al.

; Mechanisms

of‘ Ageing and Development

25

86 (1996) 1 I-26

Feldman [13] found that elderly humans fed a low salt diet for 4 days have restored isoproterenol-mediated vasodilation. This dramatic change in humans was not observed in rats on a low salt diet. We found no correction of the deficit in isoproterenol-mediated relaxation. Consequently, there may be differences in the mechanisms involved in the development of age-related impairment of smooth muscle relaxation in the two species. In summary, we have several hypotheses to elucidate the mechanism for age-related blunted CAMP responses to P-adrenergic agonists and other hormones in the vasculature and related tissues. Vascular relaxation to the p-adrenergic agonist isoproterenol decreased with maturation and aging, but the abundance of G proteins, G, and Gi, determined by Western blotting was not altered in the aorta, heart, or kidney with aging in our experiments. These results suggest that altered G, function, rather than the amount of G,, remains a possible explanation for the impaired P-adrenergic receptor-mediated relaxation with aging. Further studies are required to test G protein function with aging in blood vessels and to explore the possibility that altered expression of specific isoforms of adenylyl cyclase with aging may occur. Acknowledgements

This work was supported in part by the Research Service of the Department Veteran Affairs.

of

References [I] E.G. Lakatta, Cardiovascular

regulatory mechanisms in advanced age. Physiol.

Rec.. 73 (1993)

413s461. [2] R. Marcus and D. Gonzales, Age-related change in parathyroid hormone-dependent cyclic AMP formation in rat kidney. Me& Ageing Devel.. 20 (1982) 353-360. [3] H. Hanai, C.T. Liang. L. Cheng and 9. Sacktor, Desensitization to parathyroid hormone in renal cells from aged rats is associated with alterations in G-protein activity. J. Clin. Invest., 83 (1989) 268-277. [4] H.Y.-M. Pan, B.B. Hoffman, R.A. Persche and T.F. Blaschke. Decline in b-adrenergic receptor-mediated vascular relaxation with aging in man. J. Pharmacol. Eup. Thu., 239 (1986) 802-807. [5] G. Tsujimoto. C.-H. Lee and B.B. Hoffman, Age-related decrease in p-adrenergic receptor-mediated vascular smooth muscle relaxation. J. Pharmacol. Exp. Tiler., 239 (1986) 41 l-415. [6] T.A. Deisher, S. Mankani, and B.B. Hoffman, Role of cyclic AMP-dependent protein kinase in the diminished P-adrenergic responsiveness of vascular smooth muscle with increasing age. J. Pharmnco/. E.up. Ther., 249 (1989) 812-819. [7] J.R. Docherty. Cardiovascular responses in ageing: A review. Pharm. Rev., 42 (1990) 103-125.

[8] J.H. Chin and B.B. Hoffman, Age-related deficit in p-receptor stimulation of CAMP binding in blood vessels. Mech. Ageing Des., 53 (1990) 11 1 - 125. [9] S.R. O’Donnell and J.C. Wanstall, Thyroxine treatment of aged or young rats demonstrates that vascular responses mediated by /?-adrenoceptor subtypes can be differentially regulated. Br. J. Pharmacol., 88 (1986) 41-49. [lo] M.G. Kazanietz and M.A. Enero, Desensitization of the D-2 adrenoceptor-mediated

vasodilation in rat aorta after prolonged treatment with the b-2 adrenoceptor agonist clenbuterol. J. Pharm. E.Y~. Ther., 252 (1990) 758-764.

26

J.H. Chin et al. ! Mechanisms of Ageing and Development 86 (I 996) I I-26

[I I] M.I. Simon, M.P. Strathmann, and N. Gautam, Diversity of G proteins in signal transduction, Science, 252 (1991) 802-808. [I21 R. Taussig and A.G. Gilman, Mammalian membrane-bound adenylyl cyclases. J. Biol. Chem., 270 (1995) l-4. [13] R.D. Feldman, A low-sodium diet corrects the defect in p-adrenergic response in older subjects. Circulation, 85 (1992) 612-618. [14] G.L. Peterson, A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal. Biochem., 83 (1977) 346-356. [15] J.J. Witt and R. Roskoski, Rapid protein kinase assay using phosphocellulose-paper absorption. Anal. Biochem., 66 (1975) 253-258. [16] J. Van de Voorde and 1. Leusen, Role of the endothelium in the vasodilator response of rat thoracic aorta to histamine. Eur. J. Pharmacol., 87 (1983) 1133120. [I71 R.D. Tanz, K.S.K. Chang and T.S. Weller, Histamine relaxation of aortic rings from diabetic rats. Agents and Actions, 28 (1989) l-8. [18] B.B. Hoffman, H. Chang, Z. Farahbakhsh and G. Reaven, Inhibition of hpolysis by adenosine is potentiated with age. J. Clin. hvesr., 74 (1984) 1750-1755. [19] 9.9. Hoffman, H. Chang, Z. Farahbakhsh and G.M. Reaven, Age-related decrement in hormonestimulated lipolysis. Am. J. Physiol.. 247 (1984) E772-E777. [20] K. Urasawa, T. Murakami and H. Yasuda, Age-related alterations in adenylyl cyclase system of rat hearts. Jap. Civc. j., 55 (1991) 676-684. [21] Y. Shu and P.J. Scarpace, Forskolin binding sites and G-protein immunoreactivity in rat hearts during aging. J. Cardioousc. Pharmacol., 23 (1994) 18% 193. [22] A. Bazan, E. Van de Velde and N. Fraeyman, Effect of age on p-receptors, G,,- and G,,-proteins in rat heart. Biochem. Pharmacol., 48 (1994) 479-486. [23] K.Tobise, Y. Ishikawa, S.R. Holmer, M.-J. Im, J.B. Newell, H. Yoshie. M. Fujita, E.E. Susannie and C.J. Homey, Changes in Type VI adenylyl cyclase isoform expression correlate with a decreased capacity for CAMP generation in the aging ventricle. Circ. Res.. 74 (1994) 596-603. [24] R.F. Furchgott and P.M. Vanhoutte, Endothelium-derived relaxing and contracting factors. Fed. Am. Sot. Exp. Biol. J., 3 (1989) 2007-2018. [25] L.J. Ignarro, Biosynthesis and metabolism of endothelium-derived nitric oxide. Amzu. Reo. Pharmaco/. To.Gcol., 30 (1990) 535 -560. [26] R.M. Rapoport and F. Murad, Role of cGMP in endothelium-dependent relaxation of vascular smooth muscle. In P.M. Vanhoutte (ed.), Relaving and Contracting Fariors. Humana Press, Clifton, NJ, 1988, pp. 2199239. [27] R. Iyengar. Multiple families of G,-regulated adenylyl cyclases. In B.L. Brown and P.R.M. Dobson (eds.), Advances in Second Messenger and Phophoprotein Research, Vol. 28, Ravens Press, New York, 1993, pp. 27-36. [28] M.D. Johnson, Y. Zhou, E. Friedman. and J. Roberts. Expression of G protein r subunits in the aging cardiovascular system. J. Gerontology, 50A (1995) 814-819. [29] M.G. Kazanietz and M.A. Enero. Decreased b-adrenoceptor-mediated vasodilation in aorta from aged rats: possible involvement of a stimulatory GTP-binding protein. Eur. J. Pharmacol.. 198 (1991) 177-181. [30] A.G. Gilman, G proteins: transducers of receptor-generated signals. Annu. Rev. Biochem., 56 (1987) 615-649. [31] M. White, R. Roden. W. Minobe, F. Khan. P. Larrabee, M. Wollmering. J.D. Port. F. Anderson, D. Campbell. A.M. Feldman and M.R. Bristow, Age-related changes in /?-adrenergic neuroeffector systems in the human heart. Circulation, 90 (1994) 1225-1238. [32] A. Miyamoto, S. Kawdna, H. Kimura and H. Ohshika, Impaired expression of G,, protein mRNA in rat ventricular myocardium with aging. Eur. J. Phurmacol., 266 (1994) 147- 154. [33] C.T. Liang, J. Barnes, H. Hanai and M.A. Levine, Decrease in G, protein expression may impair adenylate cyclase activation in old kidneys. Am. J. Physiol.. 264 (1993) F770bF773.