Vol. 179, No. 2, 1991

September 16, 1991

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 979-984

Eleanor Davies, Christopher R.W. Edwards and Brent C. Williams

Department of Medicine, Western General Hospital, Crewe Road, Edinburgh EH4 ZXU, U.K.

Received July 21, 1991

To investigate the role of calcium as a second messenger in serotonm- stimulated aldosterone secretion, radiolabelled calcium influx studies were carried out in purified rat adrenal zona glomerulosa cells using 45CaCl2. The results show that serotonin caused calcium influx within 45 seconds of addition and this continued for up to 105 seconds. Angiotensin II also caused calcium influx; however, the effect was significantly smaller than that of serotonin. Serotonin-stimulated calcium influx could be inhibited by the calcium antagonist verapamil and by methysergide, a selective serotonin receptor type-l/2 antagonist. The data indicate that serotonin directly stimulates calcium uptake in zona glomerulosa cells via calcium channels which are coupled to specific serotonin receptors. B 1991 Academic Press, Inc.

The stimulatory action of 5-HT on aldosterone secretion in vitro is well

known (1, 2). However, the cellular mechanism by which the effect is mediated

remains unclear, although several studies have indicated the presence of specific

5-HT2 receptors which appear to activate adenylate cyclase and increase cyclic

AMP secretion (3, 4, 5, 6). Although the role of extracellular Ca2+ in

serotonin-stimulated aldosterone secretion has been studied indirectly using Ca2+

-chelating agents such as EGTA and Ca*+ antagonists such as verapamil (7),

direct studies using radiolabelled Ca2+ have not been carried out. Using purified

rat adrenal zona glomerulosa cells, this study investigates the role of Ca2+ influx in

Abbreviations:

ANOVA; analysis of variance. 5-HT; serotonin. EDTA; 1,2-Di (2-aminoethoxy) ethane tetra-acetic acid. SEM; standard error of mean. Ca2+; calcium. PI; phosphatidylinositol. IP3; inositol 1,4,5, tri-phosphate. ACTH; adrenocorticotropic hormone. Ci; Curie. Cyclic AMP; adenosine 3’, 5’-cyclic monophosphate.

0006-291X/91 $1.50

919 Copyright 0 1991 by Academic Press, Inc.

All righrs of reproduction in any form reserved.

Vol. 179, No. 2, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

5HT stimulated aldosterone secretion by using 45CaC12. It also compares the

effect of 5HT with that of angiotensin II, which is known to increase aldosterone

secretion through activation of phospholipase C and subsequent release of Ca*+

from intracellular storage sites (8).

MATERIALS AND METHODS

Rat adrenal zona glomerulosa cells were isolated from female Wistar rats by collagenase (Worthington Biochemical Corp, U.S.A.) digestion according to previously published methods (2). After isolation the zona glomerulosa cells were purified by percoll density gradient centrifugation according to the method of McNamara et a/ 1980 (9). Ca *+ influx was measured using a modified version of the method of Kojima and colleagues (10). (106 cells) was added to 15 pl of

Briefly, 285 ul of the cell suspension 45CaCl2 (5 @i) (Amersham International,

Aylesbury, U.K.), 100 ~1 of Krebs Ringer and 100 pl of either 5-HT (creatinine sulphate complex; Sigma Chemical Company Ltd, Poole, U.K.) or angiotensin II (Universal Biologicals, Cambridge, U.K.). Angiotensin II and 5-HT were dissolved in Krebs Ringer to the appropriate concentration. Control tubes contained 200 ul of Krebs Ringer in addition to the cells and 45CaCl2. Where utilised, 50 ul of

verapamil (Abbott Laboratories Ltd, Kent, U.K.) or methysergide (Sandoz Pharmaceuticals, Middlesex, U.K) was pre-incubated with the cells for 30 minutes at 370C before addition of the stimulus and the volume of Krebs Ringer was adjusted accordingly so that the final volume of each incubation was 500 pl. The specific activity of the 45CaCl2 was lo-40 mCi/mg Ca*+. A 100 ul aliquot of the

incubation was removed at 15, 45, 75 and 105 second intervals. This was immediately diluted in 4 ml of ice cold Tris washing solution, containing 144 mM NaCI, 5 mM CaC12, 5 mM Tris/HCI, pH 7.4, filtered through a Whatman GF/C glass-fibre filter (Whatman International Ltd, Maidstone, U.K.) and washed a further 3 times with the washing solution. The 45Ca*+ taken up by the cells, i.e. that associated with the filter, was counted in a p-counter using Cocktail-T liquid scintillation fluid (BDH Ltd, Poole, U.K.).

RESULTS All results are expressed as mean + SEM. Radiolabelled Ca*+ influx is

expressed as a % uptake of the total radioactivity added to each incubation i.e. 5

l&i. Statistical significance was calculated using 2-way ANOVA and Duncan’s

Range Test. A P value of less than 0.05 was considered significant.

Fiaure 1 shows the uptake of 45Ca*+ over a 105 second time course by

unstimulated cells (control) and those stimulated with 5-HT or angiotensin II.

Compared to the control group, 5-HT caused a significant uptake of 45Ca*+ within

45 seconds of addition (PcO.01). The uptake continued at 75 seconds (PeO.01)

and up to the final point analyzed which was 105 seconds (PcO.01). Angiotensin II

980

Vol. 179, No. 2, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

--IF Control

-O- MiT(10.6Y)

-A- Anglotensln II (IO-%)

0.0 0 20 40 60 60 100 120

Time (seconds)

Fiaure 1 shows the effect of no stimulus (control), 5-HT or angiotensin II on 45Ca2+ uptake in isolated purified rat adrenal zona glomerulosa cells. The stimulus was applied at time 0. Values are means + SEM (n=5 / group).

also caused a significant increase in 45Ca2+ uptake. However, the effect was

much slower and was not statistically significant until 105 seconds (PcO.05).

Fiaure 2 shows the uptake of 45Ca2+ over a 105 second time course by

unstimulated cells (control), cells stimulated with 5HT, cells stimulated with 5-HT in

the presence of the Ca2+ antagonist verapamil and cells stimulated with 5-HT in

2.0 -

15-

--a- Control

-c- 5HT(1C&)

-A- 5-MT (lOaM) + verapamll(lZ.5 PM)

+ 5-HT (lObY) + methysfqide (106M)

Time (seconds)

shows the effect of no stimulus (control, n=5 , 5-HT (n=5), 5-HT plus Figure 2 verapamil (n=3) and 5-HT plus methysergide (n=3) on b 5Ca2+ uptake in isolated purified rat adrenal zona glomerulosa cells. The stimulus was applied at time 0, verapamil and methysergide were pre-incubated with the cells for 30 minutes before addition of the stimulus. Values are means f SEM.

981

Vol. 179, No. 2, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

the presence of the 5-HTt/2 receptor antagonist methysergide. Compared to the

control group, 5HT caused a significant uptake of 45Ca2+ which reached

significance at 75 (P<O.O5) and 105 seconds (PcO.01). Verapamil significantly

inhibited serotonin-stimulated 45Ca2+ uptake at 75 (P&.05) and 105 seconds

(PcO.01). Methysergide also significantly inhibited 45Ca2+ uptake at 75 (PcO.05)

and 105 seconds (PcO.01).

DISCUSSION

The second messenger systems coupled to ACTH, angiotensin II and

potassium, the major physiological regulators of aldosterone secretion, have

already been studied in great detail by a number of groups. It is now generally

accepted that ACTH and angiotensin II act through the adenylate cyclase and PI

second messenger systems respectively and each stimulus causes complex ionic

changes within the cell (8, 11). Potassium, in contrast acts predominantly by

causing cellular depolarization which results in Ca2+ influx. An increase in cyclic

AMP is also observed with potassium, but it is unclear if this is the primary event or

it is secondary to Ca 2+ influx (12). 5-HT also stimulates aldosterone secretion in

vitro (1, 2). However, its mechanism of action in the zona glomerulosa remains

unclear. Some studies have shown 5-HT2-like receptors to be present, which

appear to be coupled to adenylate cyclase (3, 4). Pharmacological studies have

indicated that Ca2+ influx is also an important part of the action of 5-HT, however,

they are indirect studies which are open to the non-specific actions of the drugs

utilised (7).

This study has attempted to investigate the role of Ca2+ influx as part of the

second messenger system coupled to 5-HT receptors in the zona glomerulosa and

compare this with the well documented action of angiotensin Ii. To confirm the

existence of an inward transmembrane Ca2+ flux in the action of 5-HT, Ca2+ influx

studies were carried out using 45CaC12, which can be detected inside the cell after

lysis, by P-counting. The results show that the introduction of 5-HT to zona

glomerulosa cells caused uptake of 45Ca2+ within 45 seconds and this continued

for up to 105 seconds. Angiotensin II also caused 45Ca2+ uptake, however the

effect was much slower and significantly smaller, probably because the Ca2+

required for stimulation of steroidogenesis by angiotensin II is derived

982

Vol. 179, No. 2, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

predominantly ifom intracellular storage sites in response to If3 in the early

transient phase of the response (8). However, it is clear that angiotensin II has

some requirement for Ca2+ influx and this corresponds to the later sustained

increase in intracellular Ca2+ (8). 5HT-stimulated 45Ca2+ influx was inhibited by

the Ca2+ antagonist vefapamil, indicating the involvement of voltage operated

Ca2+ channels. In addition, the influx was also inhibited by the selective 5-HT1/2

receptor antagonist methysergide, suggesting that Ca2+ influx may be receptor

mediated. Similar concentrations of verapamil and methysergide also inhibit

aldosterone secretion in vitro (data not shown). Recent reports have shown that

5-HT does not increase intracellular Ca*+ ’ m rat adrenal zona glomerulosa cells,

suggesting that Ca2+ may not be important as a second messenger in

5-HT-induced aldosterone secretion (13). However, these studies employed the

fluorescent indicator Fura- in cell suspensions, and therefore only monitor

absolute changes in intracellular Ca2+ concentration. They do not exclude the

possibility that 5-HT may stimulate Ca 2+ influx which may then be immediately

extruded or sequestered by the cell and therefore not affect overall intracellular

Ca2+ concentration. It is equally possible that the Ca2+ influx may form part of a

membrane bound Ca2+ pool which is not detected by Fura-2. In either case, Ca2+

influx cannot be excluded as an important component of the second messenger

system in response to 5-HT in these cells. In the present studies 4%a2+ efflux

experiments were not carried out as they have already been reported in detail by

other groups, who showed that in direct contrast to angiotensin II, 5-HT, ACTH and

potassium did not affect steady-state Ca 2+ efflux from zona glomefulosa ceils (14).

In summary, the fesutts suggest that 5-HT in the adrenal zona glomerulosa

acts, like ACTH, by binding to specific membrane receptors and activating

adenylate cyclase, resulting in an increase in cyclic AMP secretion. Closely linked

to the enzymatic changes are ionic changes, whereby Ca2+ moves from the

extracellular fluid across the plasma membrane via specific ion channels which

appear to be coupled to 5-HT receptors. The integration of these two signals

appears to be necessary for the stimulation of steroidogenesis by 5-HT.

This work was funded by MRC grant number G 86138985 awarded to BCW. Thanks to Dr LM Burrell for statistical advice.

983

Vol. 179, No. 2, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

1. 2.

3.

4. 5.

6.

7. 8. 9.

10.

11. 12.

13.

14.

Miiller, J. and Ziegler, W. H. (1968) Acta Endocrinol. 59, 23-35. Haning, R., Tait, S. A. S. and Tait, J. F. (1970) Endocrinology. 87, 1147-l 167. Albano, J. D. M., Brown, B. L., Ekins, R.P., Tait, S. A. S. and Tait, J. F. (1974) Biochem. J. 142, 391-400. Fujita, K., Aguilera, G. and Catt, K. J. (1979) J. Biol. Chem. 254, 8657-8674. Matsuoka, H., Ishii, M., Goto, A. and Sujgimoto, T. (1985) Am. J. Physiol. 249, E234-E238. Rocco, S., Boscaro, M., D’agostini, D., Armanini, D., Opocher, G. and Mantero, F. (1986) J. Hypertens. 4, 551-54. Ganguly, A. and Hampton, T. (1985) Life Sci. 36, 1459-1464. Spat, A. (1988) J. Steroid Biochem 29, 443-453. McNamara, B. C., Cranna, C. E. G., Booth, R. and Stansfield, D.A. (1980) Biochem. J. 192,559-567. Kojima, I., Kojima, K. and Rasmussen, H. (1985) J. Biol. Chem. 260, 9171-9176. Kojima, I., Kojima, K. and Rasmussen, H. (1985) Biochem. J. 228, 69-76. Kojima, I., Kojima, K. and Rasmussen, H. (1985) J. Biol. Chem. 260, 4248-4256. Rocco, S., Ambroz, A. and Aguilera, G. (1990) Endocrinology. 127, 3103-3110. Williams, B. C., McDougall, J. G., Tait, J. F. and Tait, S. A. S. (1981) Clin. Sci. 61, 541-551.

984