COMMUNICATION
Dual Regulation of Adenylate Cyclase and Guanylate Cyclase: @2-Adrenergic Signal Transduction in Adrenocortical Carcinoma Cells1
NEELAM JAISWAL AND RAMESHWAR K. SHARMA2
Department of Biochemistry, University of Tennessee Center for the Health Sciences, Memphis, Tennessee 38163
Received January 27, 1986, and in revised form June 2, 1986
Isolated adrenocortical carcinoma cells of rat contain @2- and 3-adrenergic receptors. When these cells are incubated with @2-adrenergic agonists, there is a concentration- dependent increase of cyclic GMP that is blocked by the @2-adrenergic antagonist yo- himbine but not by the 8-antagonist propranolol. Concomitantly, both p-aminoclonidine (20 p.M) and clonidine (100 µM), the @2-adrenergic agonists, stimulate membrane guanylate cyclase activity. In calcium free medium there is no @2-agonist-dependent increase in cyclic GMP. Isoproterenol, a 3-agonist, and forskolin cause an increase in cyclic AMP but not cyclic GMP. The cyclic AMP increase induced by isoproterenol is blocked by propranolol but not by yohimbine. Isoproterenol- and forskolin-dependent increases in cyclic AMP are inhibited by p-aminoclonidine and the inhibition is relieved by yohimbine. These results indicate a dual regulation of guanylate cyclase and adenylate cyclase by the @2-receptor signal: guanylate cyclase is coupled to the receptor in a positive fashion, whereas adenylate cyclase is coupled in a negative fashion. Calcium is obligatory in the cyclic GMP-mediated response. @ 1986 Academic Press, Inc.
The general concept is that @2-adrenergic receptors function solely on the basis of negative coupling with adenylate cyclase (1-4). Adrenocortical carcinoma cells contain both ß- and @2-adrenergic receptors but not @1-receptors (5). The presence of a2-receptors in these cells also has been established by isolation and biochemical characterization of homogeneous recep- tors (6).
In the present studies, measurements of the levels of cyclic AMP and cyclic GMP in these cells, as a mea- sure of @2-adrenergic signal transduction, indicate the dual regulation of guanylate and adenylate cyclases by the receptor signal; guanylate cyclase is coupled to the receptor in a positive fashion, whereas ade- nylate cyclase is coupled in a negative fashion. Calcium is obligatory in the cyclic GMP-mediated response.
MATERIALS AND METHODS
The isolated adrenocortical carcinoma cells were prepared by trypsin digestion (7, 8). The method of
preparation of cells for evaluation of the effect of cal- cium and the method of incubation with @2- and @- agonists and antagonists were as described (9). In general for each adrenocortical carcinoma cell prep- aration, 1.3 g of tumor tissue was used; the cells, rep- resenting 30-35 mg of adrenal tumor (~2 × 106 cells), were resuspended in Krebs-Ringer bicarbonate buffer, pH 7.4, containing 4% albumin and 0.2% glucose. Every incubation experiment was conducted in triplicate and repeated at least three times. Assays for cyclic GMP and cyclic AMP were conducted at 10 min as described (10). To determine the guanylate cyclase activity changes in response to the @2-agonists, the adreno- cortical carcinoma cells were incubated with p-ami- noclonidine (20 M) or clonidine (100 uM) for 10 min at 37℃ and centrifuged, and the pellet was sonicated for 10 s in 250 ul of ice-cold 10 mM Tris-HCl buffer pH 7.5, containing 0.5 mM EDTA and 10.0 mM 8-mer- captoethanol. Guanylate cyclase activity was mea- sured as described previously (10).
RESULTS AND DISCUSSION
Cyclic GMP levels increased in response to the @2- agonist clonidine (11): 20 AM clonidine caused a half-
1 This investigation was supported by Grant PCM 80-0873 from the National Science Foundation. 2 To whom correspondence should be addressed.
0003-9861/86 $3.00 Copyright @ 1986 by Academic Press, Inc.
maximal response and higher concentrations caused a progressive decline in the peak level of cyclic GMP (Fig. 1). The decline in cyclic GMP levels by higher concentrations of clonidine could possibly be due to the activation of cyclic GMP-specific phosphodiester- ase by cyclic GMP (13). p-Aminoclonidine also induced an increase in cyclic GMP levels in these cells (Fig. 1, inset). The adrenocortical carcinoma cells used in these studies are devoid of cyclic GMP and cyclic AMP phosphodiesterase activities (12); therefore, the hor- monal-dependent rise in cyclic GMP levels is reflective of guanylate cyclase activity. This interpretation is further supported by the observations that both p- aminoclonidine and clonidine concomitantly stimu- lated membrane guanylate cyclase activities (Fig. 2, inset). Sodium nitroprusside, an agent known to stimulate soluble guanylate cyclase (14) but ineffective toward particulate guanylate cyclase (15, 16), did not stimulate cyclic GMP formation in the adrenocortical carcinoma cells (data not shown), indicating the se- lective positive coupling of hormonally dependent
500
cyclic GMP ( % above basal level )
300
cyclic GMP, % above basal level
400
200
300
100
2
20
100
200
PAC (UM)
100
0
0
1
10
100
1000
CLONIDINE ( UM)
200
800
600
100
400
200
cyclic GMP ( % above basal level )
Guanylate Cyclase activity (% above basal level )
PAC 20 MM
CL 100 ULM
EPI
EPI
100 UM
300
+ Yoh
200
100
PAC 2 JAN
PAC 2 MM
PAC 20 10M
PAC 20 ALM
PAC 100 UM
PAC 100 UM
+ Yoh
+ Yoh
+Yoh
400
300
200
100
CL
CL 20 UM + Yoh
CL 1001 H
CL 100 MLN *
CL CL 100MM 100 MLM
Yoh
Pro
particulate guanylate cyclase with @2-adrenergic re- ceptors. This is further supported by the studies with the @2-antagonist, yohimbine (17), where epineph- rine-, p-aminoclonidine-, and clonidine-dependent rises of cyclic GMP were all blocked by yohimbine, but propranolol, a -antagonist, did not block cloni- dine-dependent rise of cyclic GMP (Fig. 2). Although, these results indicate the @2-receptor-mediated acti-
vation of membrane guanylate cyclase, they do not reveal whether the enzyme activation is a direct or indirect membrane phenomenon.
It is noteworthy that similar to the ACTH-depen- dent particulate guanylate cyclase (15), calcium is obligatory in the clonidine-dependent rise of cyclic GMP (Fig. 1).
To determine the interaction of @2-signal transduc- tion with adenylate cyclase, the adrenocortical car- cinoma cells were incubated with isoproterenol, a ß- agonist. In accordance with the previous results (7), isoproterenol did not stimulate the cyclic GMP levels, but did elevate (~ threefold) the cyclic AMP level (Table I). The cyclic AMP rise was blocked by the B- antagonist propranolol, but not by the @2-antagonist, yohimbine, indicating the ß-adrenergic signal coupling with cyclic AMP. When the cells were, however, in- cubated with isoproterenol in the presence of p-ami- noclonidine, the isoproterenol-dependent elevation of cyclic AMP was blocked (Table I). The inhibition was prevented by yohimbine. It is noteworthy that iso- proterenol did not inhibit the p-aminoclonidine-de- pendent rise of cyclic GMP. These results not only indicate the inhibitory coupling of adenylate cyclase in @2-adrenergic-mediated signal transduction cor- roborating the results obtained with other systems (1-4), but they also show that a corresponding negative coupling of guanylate cyclase with 8-adrenergic signal does not occur.
TABLE I
| Addition | pmol/10 min | |
|---|---|---|
| CAMP | cGMP | |
| None | 2.70 ±0.10 | 0.18 ± 0.04 |
| Isoproterenol (10 µM) | 7.67 ±1.53 | 0.23 ±0.07 |
| Propranolol (20 (M) | 3.13 ±0.14 | 0.16 ±0.02 |
| Isoproterenol + propranolol | 3.48 ± 0.42 | 0.22 ± 0.06 |
| PAC (20 uM) | 3.16 ±0.94 | 0.63 ±0.18 |
| PAC (20 uM) + isoproterenol | 3.08 ±0.46 | 0.53 ± 0.15 |
| PAC (20 uM) + propranolol | 3.79 ± 0.58 | 0.59 ±0.06 |
| Yohimbine (500 µM) | 2.98 ±1.07 | 0.16 ± 0.03 |
| PAC + yohimbine | 3.50 ± 0.16 | 0.22±0.06 |
| PAC + isoproterenol + yohimbine | 6.61 ±0.64 | 0.24 ±0.07 |
| Isoproterenol + yohimbine | 8.43 ±0.77 | 0.24 ± 0.01 |
| Forskolin (200 KM) | 11.54 ± 0.69 | 0.24 ±0.06 |
| Forskolin + PAC | 6.46 ±0.92 | 0.72 ±0.12 |
| Forskolin + PAC + yohimbine | 13.64 ±1.72 | 0.22 ±0.04 |
Note. The conditions of the experiment were identical to those in Fig. 2.
To gain some understanding at what level the cou- pling of adenylate cyclase exists in the @2-signal transduction, experiments were conducted with for- skolin. This agent is known to stimulate the levels of cyclic AMP at a site beyond the level of 0-adrenergic receptor, either by directly interacting with the cat- alytic unit (18) of adenylate cyclase or by modifying the interaction between GTP binding regulatory pro- tein and the catalytic unit of adenylate cyclase (19, 20). Forskolin stimulated (~ fivefold) the rise of cyclic AMP which was inhibited by p-aminoclonidine and the inhibition was relieved by yohimbine (Table I). Forskolin had no effect on cyclic GMP levels, however. These results support the above conclusion for the negative coupling of adenylate cyclase with @2-adren- ergic receptor signal and they also provide a clue on the probable location of the inhibitory site, which is clearly beyond the receptor level and may be at the catalytic or at one of the GTP components of adenylate cyclase. Previous studies with other systems support the view that the inhibitory component of the ade- nylate cyclase, probably a GTP-binding protein, is coupled to the @2-adrenergic receptor signal trans- duction (1-4). Our results are consistent with this idea.
The intriguing aspect of our study is the demon- stration of the positive coupling of @2-receptor with particulate guanylate cyclase, implicating cyclic GMP and calcium as mediators of @2-adrenergic signal transduction. Although original studies (21) impli- cated cyclic GMP as an effector molecule mediating the effect of acetylcholine through cholinergic recep- tors, the mediatory role of cyclic GMP in signal transduction was seriously compromised since sub- sequent attempts to demonstrate a hormonally de- pendent particulate guanylate cyclase failed in every tested system (21, 22), and the soluble guanylate cy- clase activity was nonspecifically stimulated by poly- unsaturated fatty acids, peroxides, hydroperoxides, free radicals, ascorbic acid, sodium nitroprusside, and several other agents that presumably affect the oxi- dation-reduction potential of the biochemical reac- tions (22). Some of these reservations appear to be overcome since two distinct types of guanylate cyclase, particulate and soluble, have since been demonstrated in the adrenal gland (15, 16, 23), adrenocortical car- cinoma (15, 16), and liver (24) of rat. Indeed, docu- mentation of the hormonally dependent particulate guanylate cyclase in rat adrenal gland (15, 16, 23) and adrenocortical carcinoma cells (15, 16), not stimulated by nonspecific agents, has been provided. More re- cently, the above results have been corroborated in various rat tissues by demonstrating that atrial na- triuretic factor selectively activates particulate guanylate cyclase (25) including the rat adrenal gland (26). Significantly, recent in vivo studies (27) have demonstrated the atrial natriuretic factor-dependent corticosterone production in rat adrenal glands. Sim- ilar results have independently been obtained in iso- lated fasciculata cells of rat adrenal cortex and have
been further extended in revealing that cyclic GMP elevation precedes steroidogenesis (28). Furthermore, particulate guanylate cyclase has been purified and biochemically characterized from hormonally re- sponsive sea urchin spermatozoa cells (29) and from the rat adrenocortical carcinoma cells (30), the cells used in the present studies.
Our studies thus suggest that @2-adrenergic-me- diated signal transduction occurs by the dual regu- lation of guanylate and the adenylate cyclases. Such a “signaling device” will transduce the positive signal through the particulate guanylate cyclase simulta- neously screening the ß-adrenergic signal by stimu- lating the inhibitory component of the adenylate cy- clase. Such a sophisticated regulatory mechanism of the signal transduction may not be unique to the ad- renocortical carcinoma cells and therefore warrants search in other mammalian cell systems. According to the concept proposed by Rasmussen (31), cyclic GMP, calcium, and cyclic AMP would be considered as “synarchic messengers” in @2-adrenergic signal transduction.
REFERENCES
1. LAD, P. M., AND RODBELL, M. (1977) J. Biol. Chem. 252, 7964-7966.
2. SMITH, S. K., AND LIMBIRD, L. E. (1982) J. Biol. Chem. 257, 10,471-10,478.
3. MICHEL, T., AND LEFKOWITZ, R. J. (1982) J. Biol. Chem. 257, 13,557-13,563.
4. KATADA, T., NORTHUP, J. K., BOKOCH, G. M., UI, M., AND GILMAN, A. G. (1984) J. Biol. Chem. 259, 3578-3585.
5. NAMBI, P., AIYAR, N. V., AND SHARMA, R. K. (1983) J. Nutr. Growth Cancer 1, 77-84.
6. JAISWAL, R. K., AND SHARMA, R. K. (1985) Biochem. Biophys. Res. Commun. 130, 58-64.
7. PERCHELLET, J .- P., AND SHARMA, R. K. (1980) En- docrinology 106, 1589-1593.
8. SHANKER, G., AND R. K. SHARMA, (1980) Endocri- nology 106, 1594-1598.
9. PERCHELLET, J-P., AND SHARMA, R. K. (1979) Sci- ence 203, 1259-1261.
10. NAMBI, P., AIYAR, N. V., ROBERTS, A., AND SHARMA, R. K. (1982) Endocrinology 111, 196-200.
11. U’PRICHARD, D. C., GREENBERG, D. A., AND SNY- DER, S. H. (1977) Mol. Pharmacol. 13, 454-473.
12. SHARMA, R. K. (1976) in Control Mechanisms in
Cancer (Criss, W. E., Ono, T., and Sabine, J. R., eds.), pp. 109-124, Raven Press, New York.
13. PERCHELLET, J .- P., AND SHARMA, R. K. (1979) En- docrinology 105, 879-883.
14. MURAD, F., ARNOLD, W. P., MITTAL, C. K., AND BRAUGHLER, J. M. (1979) Adv. Cyclic Nucleotide Res. 11, 175-204.
15. NAMBI, P., AND SHARMA, R. K. (1981) Biochem. Biophys. Res. Commun. 100, 508-514.
16. NAMBI, P., AIYAR, N. V., AND SHARMA, R. K. (1982) Arch. Biochem. Biophys. 217, 638-646.
17. MOTUSKY, H. J., SHATTIL, S. J., AND INSEL, P. A. (1980) Biochem. Biophys. Res. Commun. 97, 1562-1570.
18. SEAMON, K. B., AND DALY, J. W. (1981) J. Biol. Chem. 256, 9799-9801.
19. BAROVSKY, K., PEDONE, C., AND BROOKER, G. (1984) Mol. Pharmacol. 25, 256-260.
20. CLARK, R. B., GORKA, T. J., GREEN, D. A., BARBER, R., AND BUTCHER, R. W. (1982) Mol. Pharmacol. 22, 609-613.
21. GOLDBERG, N. D., AND HADDOX, M. K. (1977) Annu. Rev. Biochem. 46, 823-896.
22. MURAD, F., ARNOLD, W. P., MITTAL, C. K., AND BRAUGHLER, J. M. (1979) Adv. Cyclic Nucleotide Res. 11, 175-204.
23. NAMBI, P., AND SHARMA, R. K. (1981) Endocrinology 108, 2025-2027.
24. HAGUENAUER-TSAPIS, R., SALAH, A. B., LA-COMBE, M .- L., AND HANOUNE, J. (1981) J. Biol. Chem. 256, 1651-1655.
25. WALDMAN, S. A., RAPORT, R. M., AND MURAD, F. J. (1984) J. Biol. Chem. 259, 14,332-14,334.
26. WALDMAN, S. A., RAPORT, R. M., FISCUS, R. R., AND MURAD, F. (1985) Biochim. Biophys. Acta 845, 298-303.
27. NAKAMURA, M., ODAGUCHI, K., SHIMIZU, T., NAK- AMURA, Y., AND OKAMOTO, M. (1985) Eur. J. Pharmacol. 117, 285-286.
28. JAISWAL, N., PAUL, A. K., AND SHARMA, R. K. (1986) FEBS Lett. 199, 121-124.
29. RAMARAO, C. S., AND GARBERS, D. L. (1985) J. Biol. Chem. 260, 8390-8396.
30. PAUL, A. K., AND SHARMA, R. K. (1985) 67th Annual Endocrinology Society meeting, June 19-21, Abstr. 744.
31. RASMUSSEN, H. (1981) in Calcium and cAMP as Synarchic Messengers (Rasmussen, H. ed.), Wiley-Interscience, New York.