Direct stimulation of cortisol secretion from the human NCI H295 adrenocortical cell line by vasoactive intestinal polypeptide Vanessa J. Cobb, Brent C. Williams*, J. Ian Mason and Simon W. Walker
Objective To investigate a possible direct action of vasoactive intestinal polypeptide (VIP) on adrenal cortisol secretion and to define its mechanism of action.
Design The human adrenocortical carcinoma cell line NCI H295, which is not contaminated by medullary chromaffin cells, was used to aid distinction between a direct action of VIP on adrenocortical cells and an indirect mechanism involving VIP-stimulated release of catecholamines.
Methods NCI H295 cells were challenged with 10-11-10-7 mol/I VIP for 4 h, with or without prior exposure for 72 h to 10 mmol/l forskolin. Cortisol and cyclic AMP contents of the overlying media were measured using in-house radioimmunoassays. Cells were treated with 10-8-10-6 mol/l adrenaline or 3.3 x 10-8 mol/l VIP with and without 10-8-10-6 mol/l propranolol to exclude the possibility that an indirect mechanism of action involving B-adrenoceptors was operating.
Results VIP treatment produced an increase in cortisol secretion without pre-incubation, but this was markedly enhanced by prior exposure of cells to forskolin. VIP was potent, with a threshold of 10-11 mol/l (n = 4), reaching a maximum 3.9 ± 0.9-fold increase in effect on cells
Introduction
Elevated plasma cortisol levels have been associated with high blood pressure in young people with a familial history of high blood pressure and may be an indication of a predisposition to hypertension [1]. Amongst subjects of this group there was also an increased prevalence of the AA genotype of the glucocorticoid receptor. It is possible that higher plasma cortisol levels result from the presence of this particular receptor genotype but they may equally well be an independent event. Thus, factors that increase cortisol secretion may also be involved in the pathophys- iology of essential hypertension.
There is accumulating evidence to indicate that adreno- corticotrophic hormone is not the sole regulator of cortisol secretion and that neural control mechanisms influence steroidogenesis in the human and other mammalian species (for review [2,3]). The adrenal cortex of several species contains ganglion cells and nerve fibres that
pre-exposed to forskolin (n = 4) by 3.3 x 10 8 mol/l. This increase matched the 4 h response to 10 umol/l forskolin. Cortisol secretion was accompanied by a parallel, dose-dependent increase in accumulation of CAMP.
Conclusions VIP potently and directly stimulates secretion of cortisol from these adrenocortical cells of human origin via an adenylate cyclase-coupled VIP receptor. These findings raise the possibility of a significant and direct effect of VIP in the control of steroid secretion from the adrenal cortex in humans.
Journal of Hypertension 1997, 15:1735-1738
Keywords: vasoactive intestinal polypeptide, cortisol, adrenal, NCI H295
From the Departments of Clinical Biochemistry and *Medicine, University of Edinburgh, Edinburgh, Scotland, UK.
Sponsorship: The support of the University of Edinburgh in the award of a Crighton Scholarship to Vanessa Cobb is gratefully acknowledged.
Requests for reprints to Vanessa J. Cobb, Department of Clinical Biochemistry, University of Edinburgh, Royal Infirmary of Edinburgh, Edinburgh EH3 9YW, Scotland, UK.
@ Rapid Science Publishers ISSN 0263-6352
synapse in close apposition to adrenocortical cells as well as blood vessels within the cortex [4,5].
Several putative neurotransmitters have been implicated as regulators of adrenal steroidogenesis, including acetyl- choline, catecholamines, corticotrophin-releasing factor, neuropeptide Y and vasoactive intestinal polypeptide (VIP, for review [3]). In the case of VIP, it has been dem- onstrated that rat adrenal cortex contains VIP-immuno- reactive nerve fibres [4] together with the VIP receptor messenger RNA being present both in the adrenal cortex and in the medulla [6]. Stimulation of the splanchnic nerve results in an increase in glucocorticoid secretion, in the presence of exogenous adrenocorticotrophic hormone, from the adrenals of conscious hypophysectomized calves [7], an effect mimicked by infusion of exogenous VIP [8]. Furthermore, stimulation of the splanchnic nerve is accompanied by the appearance of VIP in the venous effluent from conscious calves [9]. In other species, VIP
5 6
7 8
9
1
2
3 4
5
6 7 8
9
1
2
2
3
4
5
6
7 8
9
1
2
3 4 5
6
7 8
9
1 2
3
4
5 6
7 8
9
1
2
3
4
3
4
5
6 7
8
9
1 2
3 4
5
6
7
8
9
1
2
3 4
5
6
7
8
9
1
2
3 4
5 6 7
8
9
1 2
3 4
5
6
7 8
9
1 2
3 4
5
was found to stimulate androstenedione secretion from isolated, perfused porcine adrenals [10] and to increase steroid production from intact capsule-zona glomerulosa preparations from the rat [11].
Controversy has surrounded the question of whether VIP can stimulate steroidogenesis directly. Hinson et al. [12] failed to obtain a response of steroid secretion to VIP in collagenase-dispersed rat adrenocortical cells but observed an increase in aldosterone secretion to VIP from an intact capsule-zona glomerulosa preparation. Because adminis- tration of VIP can stimulate release of catecholamines from medullary chromaffin cells [13], it has been postu- lated that it might act indirectly by releasing catechol- amines from islets of chromaffin cells known to be present in the cortex [14]. The catecholamines would then stim- ulate secretion of steroids by acting on ß-adrenoceptors present on the adrenocortical cells. In contrast, Mazzocchi et al. [15] reported that an increase in steroid secretion occurred with administration of VIP in dispersed rat adrenocortical cells. Because it is likely that collagenase- dispersed rat adrenocortical cell preparations could contain some contaminating medullary chromaffin cells, it is difficult to determine whether the response observed is direct, indirect or both. Bornstein et al. [16] reported that only an attenuation of VIP-induced secretion of steroid in response to propranolol treatment occurred in a primary culture of human adrenal cells (a mixed culture of adrenocortical and medullary cells) [16], indicating that two mechanisms of VIP action may be involved.
In order to investigate a possible direct action of VIP on adrenocortical cells, we studied the steroidogenic effects of VIP on the human adrenocortical carcinoma cell line NCI H295. This cell line was originally established from a primary invasive adrenal tumour in 1980, from a patient showing signs of mineralocorticoid, glucocorticoid and adrenal androgen excess [17]. It has since been charac- terized and shown to secrete steroids from the three major pathways in the adrenal cortex [18]. Because it is not cont- aminated by medullary chromaffin cells, the NCI H295 adrenocortical cell line is the model of choice for exper- iments to distinguish direct from indirect effects of VIP on adrenocortical steroidogenesis.
Materials and methods
NCI H295 cells (ATCC, Rockville, Maryland, USA) were seeded into 12-well plates (5x 105 cells per well) and maintained in Dulbecco’s modified Eagle’s medium F12, 2% Ultroser HY (Gibco, Renfrew, UK), 5 µg/ml insulin, 5 µg/ml transferrin and 5 ng/ml sodium selenite (as 1% ITS; Sigma, Poole, UK) at 37°℃ with 5% CO2-95% air. Cells were pretreated with the above medium containing either 10 mmol/l forskolin or no agonist (‘control’) for 72-96 h, replacing medium every 24 h. After pretreat- ment, cells were washed with Earle’s balanced salt solu- tion, incubated for 1 h in serum-free medium and finally
Fig. 1
(a)
5
CORTISOL Secretion (n-fold increase over basal)
I
4
I
3
2
1
0
10M fskn
-11
-10
-9
-8
-7
log [VIP] (M)
(b)
30
[cAMP] in media (n-fold increase over basal)
25
20
15
10
5
0
10uM fskn
-11
-10
-9
00
-7
log [VIP] (M)
Dose-dependent increases in (a) cortisol and (b) cyclic AMP (cAMP) secretions after 4 h challenge with vasoactive intestinal peptide (VIP) Maximal stimulation occurred in response to 3.3 x 10-8 mol/l VIP (a 3.9 ± 0.9-fold increase) and was comparable to that in response to a 4 h challenge with 10 umol/l forskolin (fskn). Combined data (expressed as mean n-fold increases relative to basal values + SD) from four experiments both for cortisol and CAMP.
exposed to fresh serum-free medium with or without 10-11-10-7 mol/l VIP for 4 h at 37°C. For B-adrenoceptor activation studies, cells were treated in the same manner but with various doses of adrenaline, propranolol or VIP plus propranolol, as stated in the Results.
After 4 h, medium was removed and the cortisol and cyclic AMP (cAMP) content measured, using direct radioim- munoassays. The cell monolayer was washed (in Eagle’s
2
2
3
4
5
6
7
8
9
1
2
3
4
7
8
9
1
2
3
4
5
6
7 8
9
1
2
3
4
5
6 7
8
9
1
2
3 4
5 6
7 8
9
1
2
3
Fig. 2
(a)
175
150
CORTISOL Secretion (pmol / mg protein / 4 h)
T
125
T
100
75
50
25
0
(b)
175
[cAMP] in media (pmol / mg protein / 4 h)
150
125
100
75
50
T
25
0
fskn 10uM
Fskn + P -6
basal
P -6
VIP
VIP + P -6
basal
A -6
Administrations of 10-8, 10-7 and 10-6 mol/l propranolol (P) failed to inhibit 4 h basal, 4 h forskolin (Fskn and fskn)-stimulated (10 µmol/l) and 4 h vasoactive intestinal peptide (VIP, 3.3 x 10-8 mol/l)-stimulated cortisol (a) and cyclic AMP (cAMP, b) secretions only 10-6 mol/l P data are shown). Representative experiment from three experiments both for cortisol and for cAMP. Four-hour challenges with 10-8, 10-7 and 10-6 mol/l adrenaline (A) failed to stimulate cortisol (a) and cAMP (b) secretion above basal levels (12; only 10-6 mol/l A data are shown). Representative experiment from a set of three experiments both for cortisol and for CAMP.
balanced salt solution alone) and taken up into 0.3 mol/l NaOH with 0.1% sodium dodecyl sulphate for subsequent protein determination. The steroid and cAMP measure- ments were corrected for cell protein concentration.
Results are expressed as means of triplicate wells ± SD for n experiments. Statistical tests were performed using an unpaired, two-tailed, Student’s t test. P < 0.05 was considered statistically significant.
Results
Administration of 10-11-10-7 mol/l VIP produced a dose- dependent increase in secretion of cortisol by H295 cells. The stimulation was statistically significant for ‘control’ cells (no pre-incubation), in that a dose of 10-8 mol/l VIP resulted in a 2.0 ± 0.3-fold stimulation of cortisol secretion relative to basal values (n = 4). However, this response was greatly enhanced for cells pretreated for 72 or 96 h with 10 pmol/l forskolin, for which 10-8 mol/l VIP elicited a 3.6 ± 0.6-fold increase relative to basal values (n = 4), the threshold change occurring at 10-11 mol/l (n = 4).
The maximal response to VIP occurred at a dose of 3.3 × 10-8 mol/l both for cortisol and for cAMP production and pEC50 (negative logarithm to the base 10 of the concentration for half the maximal effect) values were 9.22 ± 0.43 (n =4) and 8.57 ±0.45 (n=4) for cortisol and cAMP production, respectively (Fig. 1).
Administration of the ß-adrenoceptor antagonist propra- nolol did not affect either basal or VIP-stimulated cortisol or cAMP secretion, which was consistent with the VIP response occurring through a specific VIP receptor and not, therefore, supportive of the hypothesis of there being an interaction of VIP with ß-receptors (Fig. 2). Indeed, administration of the non-selective ß-receptor agonist adrenaline failed to elicit an increase either in cortisol or in cAMP levels relative to basal values from H295 cells, suggesting that there had been a loss of functional B-receptors on cells of this cell line (Fig. 2).
Discussion
The mechanism by which forskolin pretreatment en- hances the subsequent response to VIP is, at present, not known. Possible explanations include upregulation of VIP receptors at the cell surface and upregulation of steroid- ogenic enzyme expression (since administration of for- skolin markedly upregulates expression and activity of P450c17 [19]). The latter explanation is supported by the fact that pretreatment with forskolin also enhanced the response of cortisol secretion to a subsequent 4 h further administration of 10 mol/l forskolin. The maximal steroidogenic effect of VIP was equivalent to that produced by 10 pmol/l forskolin, emphasizing that VIP is a very potent stimulus of cortisol production in these cells.
Two subtypes of the VIP receptor, VIP1 and VIP2, have so far been isolated and cloned [20,21]. Activation of VIP1 and VIP2 receptors increases intracellular cAMP levels via activation of adenylate cyclase in a variety of tissues. Administration of VIP also elicited a dose-dependent increase in cAMP production in forskolin-pretreated H295 cells which paralleled the increase in cortisol production (see Fig.1).
In conclusion, VIP acts directly on H295 cells to stimu- late steroidogenesis with an accompanying dose-depen- dent increase in cAMP. These findings do not rule out an indirect mechanism of action on adrenocortical cells in vivo but raise the possibility of an accompanying direct effect of VIP in the control of steroid secretion from the adrenal cortex of man.
References
1 Watt GCM, Harrap SB, Foy CJW, Holton DW, Edwards HV, Davidson HR, et al .: Abnormalities of glucocorticoid metabolism and the renin-angiotensin system: a four-corners approach to the identification of genetic determinants of blood pressure. J Hypertens 1992, 10:473-482.
2 Charlton BG: Adrenal cortical innervation and glucocorticoid secretion. J Endocrino/ 1990, 126:5-8.
3 Edwards AV, Jones CT: Autonomic control of adrenal function. J Anat 1993, 183:291-307.
4 Holzwarth MA, Cunningham LA, Kleitman N: The role of adrenal nerves in the regulation of adrenocortical functions. Ann NY Acad Sci 1987, 512:449-464.
5 McNicol AM, Richmond J, Charlton BG: A study of general innervation of the human adrenal cortex using PGP 9.5 immunohistochemistry. Acta Anat 1994, 151:120-123.
6 Usdin TB, Bonner TI, Mezey E: Two receptors for vasoactive intestinal polypeptide with similar specificity and complementary distributions. Endocrinology 1994, 135:2662-2680.
7 Edwards AV, Jones CT: The effect of splanchnic nerve stimulation on adrenocortical activity in conscious calves. J Physio/ 1987, 382:385-396.
8 Bloom SR, Edwards AV, Jones CT: Adrenal-cortical responses to vasoactive intestinal peptide in conscious hypophysectomized calves. J Physio/ 1987, 391:441-450.
9 Bloom SR, Edwards AV, Jones CT: The release of certain neuropeptides in response to splanchnic nerve-stimulation in conscious calves [abstract]. J Physio/ 1987, 384:P29
10 Bornstein SR, Ehrhart-Bornstein M, Stromeyer HG, Adler G, Scherbaum WA, Holst JJ: Vasoactive intestinal peptide (VIP) stimulates androstenedione release in isolated perfused pig adrenals. Life Sci 1993, 52:135-140.
11 Cunningham LA, Holzwarth MA: Vasoactive intestinal peptide stimulates adrenal aldosterone and corticosterone secretion. Endocrinology 1988, 122:2090-2097.
12 Hinson JP, Kapas S, Orford CD, Vinson GP: Vasoactive-intestinal- peptide stimulation of aldosterone secretion by the rat adrenal-cortex may be mediated by the local release of catecholamines. J Endocrinol 1992, 133:253-258.
13 Yamaguchi N: In vivo evidence for adrenal catecholamine release mediated by nonnicotinic mechanism: local medullary effect of VIP. Am J Physiol 1993, 265:R766-R771.
14 Bornstein SR, Ehrhart-Bornstein M, Scherbaum WA, Pfeiffer EF, Holst JJ: Effects of splanchnic nerve stimulation on the adrenal cortex may be mediated by chromaffin cells in a paracrine manner. Endocrinology 1990, 127:900-906.
15 Mazzocchi G, Malendowicz LK, Nussdorfer GG: Stimulatory effect of vasoactive intestinal peptide (VIP) on the secretory activity of dispersed rat adrenocortical cells. Evidence for the interaction of VIP with ACTH receptors. J Steroid Biochem Mol Biol 1994, 48:507-510.
16 Bornstein SR, Haidan A, Ehrhart-Bornstein M: Cellular communication in the neuroadrenocortical axis - role of vasoactive intestinal polypeptide (VIP). Endocrin Res 1996, 22:819-829.
17 Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, et al .: Establishment and characterisation of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 1990, 50:5488-5496.
18 Rainey WE, Bird IM, Mason JI: The NCI-H295 cell line: a pluripotent model for human adrenocortical studies. Mol Cell Endocrino/ 1994, 100:45-50.
19 Rainey WE, Bird IM, Sawetawan C, Hanley NA, McCarthy JL, McGee EA, et al .: Regulation of human adrenal carcinoma cell (NCI-H295) production of C19 steroids. J Clin Endocrinol Metab 1993, 77: 731-737.
20 Ishihara T, Shigemoto R, Mori K, Takahashi K, Nagata S: Functional expression and tissue distribution of a novel receptor for vasoactive intestinal polypeptide. Neuron 1992, 8:811-819.
21 Lutz EM, Sheward WJ, West KM, Morrow JA, Fink G, Harmar AJ: The VIP2 receptor: molecular characterisation of a cDNA encoding a novel receptor for vasoactive intestinal peptide. FEBS Lett 1993, 334:3-8.
5 6
7
8 9
3 4
5 6 7
8
9
4 5
4
2
2