0013-7227/90/1261-0241$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society

Vol. 126, No. 1 Printed in U.S.A.

Evidence that Testosterone Modulates in Vivo the Adenylate Cyclase Activity in Fat Cells* RENE PECQUERY, MARIE-NOELLE DIEUDONNE, MARIE-CHRISTINE LENEVEU, AND YVES GIUDICELLIt Department of Biochemistry, Faculty of Medicine Paris-Ouest, Centre Hospitalier de Poissy, 78303 Poissy Cedex, France

ABSTRACT. In male hamster fat cell membranes, the a2adrenoreceptor-mediated inhibitory response of adenylate cyclase was almost completely suppressed by castration and was restored to control values after testosterone treatment, whereas the cyclase inhibitory response to nicotinic acid was insensitive to androgenic status. Basal and forskolin-, guanylylimidodiphosphate- and isoproterenol-stimulated cyclase activities were decreased by 30-40% after castration and restored to control values after testosterone treatment. In addition, Mn2+ + forskolin-stimulated activity in the presence or absence of GDP/3S was lower (—30%) after castration and normalized after testosterone treatment. Finally, the effects of testosterone described above

were completely abolished when the potent androgen receptor antagonist RU 23908 was administered together with testosterone. These results indicate that both the inhibitory and stimulatory responses of adenylate cyclase are promoted by testosterone through an androgen receptor-dependent mechanism; promotion of the inhibitory response concerns specifically the «2-receptormediated pathway, whereas promotion of the stimulatory response appears unspecific and mainly due to increased activity of the cyclase catalytic subunit. (Endocrinology 126: 241-245, 1990)

CCUMULATING evidence has underlined the importance of the balance between a2 (antilipolytic)-

the rat. In another recent study (8), we have also shown that androgenic status can modify the functional a2/fi-

and j8 (lipolytic)-adrenoreceptor-mediated control of li-

adrenergic balance in hamster adipocytes; testosterone

polysis in adipocytes from several species (1-3). Changes in the distribution of these receptors seem to be at least in part responsible for some of the site- and sex-related differences in the lipolytic responses to catecholamines in human adipocytes (4-6), which suggests that sex steroid hormones could play an important role in regulation of the lipolytic process in adipose tissue. However, lipolysis is a distal event which depends directly on the phosphorylation of triglyceride lipase by cAMP-dependent protein kinase-A. Since both «2- and /3-adrenoreceptors control the cAMP-generating enzyme adenylate cyclase, it appears crucial to determine whether this enzyme activity is influenced by sex hormones in adipose tissue. We have recently reported that estradiol treatment of ovariectomized rats results in a large increase in the fat cell adenylate cyclase catalytic activity (7), a mechanism that helps to explain how estrogens promote lipolysis in

in vivo, while increasing the 0-adrenergic lipolytic action of epinephrine, promoted to a greater extent its a2adrenoreceptor-mediated antilipolytic potency, resulting in a low lipolytic response to this catecholamine. Moreover, in adipocytes from testosterone-treated castrated hamster, cAMP production was enhanced in the presence of either isobutylmethylxanthine (IBMX), a phosphodiesterase inhibitor, or forskolin, a receptor-independent stimulatory agent, suggesting that testosterone may promote the activity of the adenylate cyclase catalytic subunit. In the present study we now provide direct evidence that testosterone indeed enhances the activity of the fat cell adenylate cyclase catalytic subunit and that most of the effects of testosterone mentioned above can be abolished by the antiandrogen nilutamide (RU 23908) (9).

A

Materials and Methods Six-week-old hamsters (100 ± 10 g) were kept under a 06002000 h light schedule and had free access to food and tap water. Castration was performed under sodium pentobarbital anesthesia after a single middle ventral incision. Control animals were sham-operated. After 1 week, half of the castrated animals received one daily sc injection of testosterone propionate (10

Received June 23,1989. * This work was supported by the Institut National de la Sante et de la Recherche Medicale (INSERM Grant CRE 892 011) and the Direction de la Recherche (Universite Paris V). t To whom all correspondence and requests for reprints should be addressed. 241

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 19:04 For personal use only. No other uses without permission. . All rights reserved.

242

TESTOSTERONE AND FAT CELL ADENYLATE CYCLASE

mg/kg BW in 0.1 ml propylene glycol) for 10 days, while the other half and the controls received the vehicle only. One day after the last injection, animals were killed, and epididymal and perirenal adipose tissues removed. In some experiments testosterone-treated animals received one daily sc injection of RU 23908 (20 mg/kg BW in 0.1 ml propylene glycol). Fat cells were isolated by collagenase digestion, and adipocyte membranes were prepared in the presence of 5 fig/ml soybean trypsin inhibitor, 0.124 mM benzamidine and 100 pM phenylmethylsulfonylfluoride, as previously described (10). Incubations were performed with fresh membranes (10-20 ^g) at either 30 C (for stimulation studies) or 25 C (for inhibition studies) in a final volume of 0.1 ml containing 25 mM Tris-HCl (pH 7.4), 2 mM MgCl2, 0.1 mM EDTA, 1 mM IBMX, 0.2 mM ATP, 5 mM phosphocreatine, 10 IU/ml creatine kinase, 50 mU/ ml adenosine deaminase, and 0.1% albumin. At the indicated times, reactions were stopped, and cAMP production was determined (11). Protein concentrations were measured according to the method of Bradford (12) with bovine albumin as standard. [3H]cAMP and cAMP-binding protein were obtained from the Radiochemical Centre (Amersham, United Kingdom); (-)isoproterenol bitartrate, (—)-epinephrine bitartrate, ATP (A2383), creatine phosphate, creatine kinase, bovine albumin (fatty acid poor), testosterone propionate, and IBMX from Sigma Chemical Co. (St. Louis, MO); adenosine deaminase, GTP, guanylylimidodiphosphate [Gpp(NH)p], GDP/3S from Boehringer Mannheim (Mannheim, West Germany); bacterial collagenase (CLS) from Worthington Biochemical Corp. (Freehold, NJ); and forskolin from Calbiochem (Lucerne, Switzerland). RU 23908 was a generous gift from Roussel-UCLAF (Romainville, France). All other chemicals were of analytical grade. Statistical significance of the data was calculated using paired and unpaired Student's t tests.

Results and Discussion As shown in Fig. 1A, the inhibitory response of adenylate cyclase to epinephrine («2-adrenergic response tested in the presence of 10 ixM propranolol) was almost completely abolished after castration and restored to control values after testosterone treatment. These alterations appear, at least in part, related to changes in the fat cell «2-adrenergic receptor density, which, as previously reported by us (8), was either dramatically reduced after castration or higher than control values after testosterone treatment. As shown in Fig. IB, however, the sensitivity and magnitude of the inhibitory adenylate cyclase response to nicotinic acid were unimpaired by castration and testosterone treatment. The fact that G;, the inhibitory regulatory protein, is involved in the coupling of nicotinic acid receptors to adenylate cyclase (13) strongly suggests that Gi is not implicated in the altered cyclase «2-adrenergic responses described above. In castrated hamsters we found an overall decrease

Endo • 1990 Voll26-Nol

(30-40%) in basal and stimulated adenylate cyclase activities regardless of the stimulatory agent tested (Fig. 2). However, after testosterone treatment these activities were either restored to control values [responses to forskolin or Gpp(NH)p alone or combined with isoproterenol] or increased over control values (basal and Mn2+ responses). These data are also consistent with our previous studies (8) showing that both lipolytic activity and cAMP production are decreased after castration and similar to or even higher than control values after testosterone treatment. That these adenylate cyclase changes due to androgenic status were observed for both receptor- and nonreceptor-mediated stimulations strongly suggests that these changes are not linked to altered adenylate cyclase stimulatory receptors. In adipocytes from estrogen-treated ovariectomized rats we have recently found an abnormally high activity of the adenylate cyclase catalytic subunit (7). This observation has led us to examine whether androgenic status could also influence the adenylate cyclase catalytic activity in hamster fat cells. According to Krall et al. (14), the activity of the purified cyclase catalytic subunit is increased by the addition of forskolin and Mn2+ ions. However, to measure this activity in membranes, the stimulatory influence of the regulatory coupling protein G s has to be reduced by the addition of GDP/3S (500 /LIM) (15). Data in Table 1 extend the validity of these mechanisms to the hamster fat cell system by showing that GDP/3S could reduce to basal values the Gpp(NH)pstimulated adenylate cyclase activity in both the castrated and testosterone-treated groups. Moreover, in the three experimental groups, addition of GDP/3S reduced to about the same extent (-20 to —25%) the Mn2+- plus forskolin-stimulated activity, which was either lower (-30%; P < 0.05) after castration or slightly (but not significantly) higher than control values after testosterone treatment (Fig. 3). However, experiments investigating the sensitivity (EC50 values) of the cyclase stimulatory response to guanine nucleotides [GTP and Gpp(NH)p] failed to reveal any influence of androgenic status on the activity of Gs (data not shown). Finally, when the potent androgen receptor antagonist RU 23908 was simultaneously administered with testosterone propionate, the promoting effects of testosterone alone on both a2-adrenoreceptor-mediated cyclase inhibition and nonreceptor-mediated cyclase stimulatory responses were abolished, leading to adenylate cyclase activities similar to those found in castrated animals (Table 2). Preliminary experiments (not shown) also revealed identical adenylate cyclase inhibitory and stimulatory responses in adipocyte membranes from castrated and RU-23908-treated castrated hamsters. From these results, it thus appears that much of the influence of androgenic status on the adenylate cyclase

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 19:04 For personal use only. No other uses without permission. . All rights reserved.

TESTOSTERONE AND FAT CELL ADENYLATE CYCLASE -.

243

B

100

V) 10 X3

U

HI 0)

CJ CJ HI I< >•

> H UJ Q

-4

-5

L0G10 [-EPINEPHRINE]

-7

(M)

L0G10 [NICOTINIC ACID]

(M)

FIG. 1. Dose-dependent inhibition of adenylate cyclase by epinephrine plus 10 /*M propranolol (A) and nicotinic acid (B). Adipocyte membranes prepared from sham-operated (D), castrated (A), and testosterone-treated castrated (A) hamsters were incubated at 25 C for 10 min. Results are expressed as percent inhibition of the basal values obtained in the presence of 1 nM GTP and 120 mM NaCl. Each point represents the mean ± SEM of three independent experiments performed in duplicate. The mean (±SEM) basal values for the three experiments were 797 ± 193 (D), 615 ± 127 (A), and 1098 ± 275 (A) pmol/mg protein-10 min. *, P < 0.01 vs. sham-operated.

Gpp(NH)p Mn

2+

( 1 mM)

130 FIG. 2. Effects of castration and testosterone replacement on basal and stimulated adenylate cyclase activities. Results are expressed as a percentage of the adenylate cyclase activity found in membranes from sham-operated animals. Each point is the mean ± SEM of four separate experiments. Symbols are explained in Fig. 1. For each effector tested, 100% response ( ) corresponds to the adenylate cyclase activity found in membranes from sham-operated animals. Enzyme activities were measured at 37 C for 10 min and, except for Mn 2+ , in the presence of 2 mM MgCl2. *, P < 0.02; **, nonsignificant vs. sham-operated.

CJ

Evidence that testosterone modulates in vivo the adenylate cyclase activity in fat cells.

In male hamster fat cell membranes, the alpha 2-adrenoreceptor-mediated inhibitory response of adenylate cyclase was almost completely suppressed by c...
561KB Sizes 0 Downloads 0 Views