l)01U-7227/91/1291-0489$03.00/0 Endocrinolony Copyright n" 1991 by The Endocrine Society
Vol. 129, No. 1
Printed in U.S.A.
Parathyroid Hormone Modulates Angiotensin II-Induced Aldosterone Secretion from the Adrenal Glomerulosa Cell* CARLOS M. ISALESf, PAULA Q. BARRETT, MICHAEL BRINES, WENDY BOLLAG, AND HOWARD RASMUSSEN Departments of Internal Medicine and Physiology, Yale University School of Medicine, New Haven, Connecticut 06510
PTH increased the sensitivity to angiotensin-II, shifting the Ko for activation from 1.0 to 0.3 nM. In contrast, between 30-45 min of angiotensin-II stimulation, PTH elevated the maximal secretory response to angiotensin-II from 109 ± 3.4 to 219 ± 13.3 pg/min • million cells. By itself PTH elicited only a small increase in the intracellular Ca2+ concentration, as measured by aequorin luminescence in glomerulosa cells. In cells pretreated with angiotensin-II or 15 mM potassium, the intracellular calcium response to PTH was markedly potentiated. PTH was also found to cause a small increase in the cellular cAMP content. Thus, PTH stimulates aldosterone secretion from adrenal glomerulosa cells, both alone and in combination with angiotensinII. (Endocrinology 129: 489-495, 1991)
ABSTRACT. The effect of PTH on aldosterone secretion from isolated bovine adrenal glomerulosa cells was examined. PTH binding was autoradiographically localized to the adrenal cortex, suggesting a specific effect. This binding of PTH was displaceable by cold PTH, but not by ACTH. No binding was observed in the adrenal medulla. In addition, PTH was shown to stimulate aldosterone secretion in a dose-dependent manner and to potentiate aldosterone secretion in response to angiotensin-II, such that PTH (109 M) elevated the secretory rate from 58.6 ± 6.8 to 110.9 ± 19 pg/min-million cells in the presence of 10 nM angiotensin-II. The magnitude of the synergism between the two hormones depended on the concentrations of PTH and angiotensin-II as well as the time during which aldosterone secretion was measured. Within the first 15 min of stimulation,
S
subsequent presentation. On the other hand, Hulter et al. (6) have shown that the infusion of PTH over a 12day period in humans results in a rise in blood pressure that parallels the length of the infusion. Interestingly, these patients also had a transient rise in aldosterone secretion. In view of the difficulty of demonstrating direct effects of PTH on the renin-angiotensin (Ang-II)-aldosterone axis from the indirect effects seen in vivo (induced by changes in serum calcium levels), we evaluated the possibility that PTH might directly influence aldosterone production from isolated bovine adrenal glomerulosa cells. We demonstrate for the first time that there are binding sites for PTH on adrenal glomerulosa cells and that these sites can modulate Ang-II-induced aldosterone secretion. This effect appears to be mediated by an elevation of the intracellular calcium concentration and a modest increase in the cellular cAMP concentration. Our results confirm and extend the data of other investigators, who have shown a PTH effect on adrenal steroidogenesis (7-9).
EVERAL lines of investigation suggest that the principal hormones involved in maintaining calcium homeostasis, PTH and 1,25-hydroxyvitamin D, may play a role in blood pressure regulation. Patients with primary hyperparathyroidism have a greater incidence (between 30-70%) of high blood pressure than that seen in the normal population (1, 2). However, the exact role of the PTH-vitamin D axis in the development of hypertension in these patients remains controversial, since their high blood pressure does not always improve after successful parathyroidectomy (1, 2). Likewise, studies in an animal model of essential hypertension, the spontaneously hypertensive rat (SHR), suggest that PTH may play a pathogenetic role. These rats exhibit abnormal calcium metabolism, with an increased circulating concentration of PTH (3-5). Parathyroidectomy of these animals before the development of hypertension ameliorates its
Received January 7, 1991. Address requests for reprints to: Carlos M. Isales, M.D., Division of Endocrinology, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut 06510. * This work was supported by grants from the National Dairy Promotion and Research Board through the National Dairy Council, the NIDDK (DK-19813), and the NHLBI (HL-36977). t Supported by a Physician Scientist Award from the NIDDK (DK01825).
Materials and Methods Materials Collagenase was obtained from Boehringer Mannheim Biochemicals (Indianapolis, IN); Percoll was from Pharmacia (Pis489
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 January 2016. at 01:02 For personal use only. No other uses without permission. . All rights reserved.
PTH MODULATION OF ALDOSTERONE SECRETION
490
cataway, NJ); BSA and agarose type VII (low temperature gelling) were obtained from Sigma (St. Louis, MO); PTH-(134) and -(1-84), Nle8'18,Try34,PTH-(l-34), and ACTH-U-24) and -(7-38) were from Bachem, Inc. (Torrance, CA). Aequorin was purchased from the Mayo Clinic (Rochester, MN). All other chemicals were of the best reagent grade available. Glomerulosa cell isolation Glomerulosa cells were isolated from bovine adrenal glands as described previously (10). Briefly, bovine adrenal glands were obtained at a local slaughterhouse and stored in ice-cold saline until use. Fat and pericapsular tissue were removed, and thin slices containing the glomerulosa layer were isolated. The tissue was minced, and cells were dispersed by collagenase digestion, followed by mechanical agitation. Cells were suspended in a modified Krebs-Ringer bicarbonate buffer containing NaCl (120 mM), KC1 (3.5 mil), NaHCO3 (25 mM), MgS04 (1.2 mM), NaH2PO4 (1.2 mM), dextrose (5 mM), BSA (0.2%), and CaCl2 (1.25 mM) and equilibrated with 95% air and 5% CO2. Static incubations Glomerulosa cells were incubated at 37 C at a density of 250,000 cells/ml for 30 min before agonist addition. Incubations were terminated at the appropriate times by placing the samples in ice and immediately centrifuging at 1000 X g for 15 min (4 C). The supernatant was removed and stored at -20 C until the determination of aldosterone content. Aldosterone was measured using a solid phase RIA (Diagnostic Products, Los Angeles, CA). Except where specified, incubations were performed in triplicate, and each experiment was repeated three times using different cell preparations. The absolute value of the aldosterone secretory response from the isolated bovine adrenal glomerulosa cells can vary with the season (winter us. summer), but qualitatively the responses remain unchanged. Radioiodination Radioiodination of NNT-hPTH-(l-34) was performed as previously described, using a modification of the lactoperoxidase method (11). Briefly, PTH (1 fig/fd; 10 fig total) was added to Na125I (1 mCi; Amersham, Arlington Heights, IL), followed by lactoperoxidase (2 fig; Sigma Chemical Co., St. Louis, MO). The reaction was initiated by the addition of hydrogen peroxide (20 y\ 0.03% H2O2) and was maintained by three further 20-/ul additions of 0.03% H2O2 at 2.5-min intervals for a total of 10 min. The iodinated peptide was then passed through a C18 Sep-Pak cartridge (Water Associates, Milford, MA) to remove unreacted Na125I. The cartridge was washed with 3 ml 0.1% trifluoroacetic acid (TFA) and eluted with 3 ml 75% acetonitrile-25% H2O (vol/vol) containing 0.1% TFA into glass test tubes containing 30 /A 2% BSA. The eluate was lyophilized and purified by reverse phase HPLC using a 30-cm /u-Bondpak C18 column (Waters Associates). The column was equilibrated with H2O containing 0.1% TFA and developed with acetonitrile in 0.1% TFA. The gradient employed was a 60-min linear gradient of 33-43% acetonitrile. Eluted fractions were collected in borosilicate glass tubes (12 x 75) containing 30 jul 1% BSA and monitored for radioactivity in a 7-spectrometer.
Endo • 1991 Vol 129 • No 1
Autoradiography of adrenal slices Fresh adrenal glands were sliced, placed in cryostat embedding medium (O.C.T. Compound, Miles Scientific, Naperville, IL), and frozen at -70 C. The frozen specimens were serially cut into 16-/on thick sections with a cryostat, thawmounted on gelatin-subbed slides (24), air dried at 37 C for 5 min, and stored at -80 C until use (within 1 month). To generate autoradiographs, the sections were brought to room temperature and preincubated for 15 min at 25 C in a modified pH 5.0 Krebs-Ringer buffer (KRB) to dissociate any bound ligand. Excess solution was carefully removed without allowing the sections to dry. Approximately 300,000 counts of [125I]PTH-(l-34) in 800 fi\ buffer containing protease inhibitors (60 mg/ml bacitracin; 250,000 IU/ml aprotinin) with (nonspecific binding) or without (total binding) added unlabeled PTH(1-34) were then carefully layered onto each slide within individual humidified petri dishes. The slides were incubated at room temperature for 45 min, a time at which on the basis of pilot experiments slices were maximally labeled. The labeling reaction was terminated by rinsing three times in ice-cold buffer, followed by a rinse in iced water to remove buffer salts. Sections were then dried in dust-free cold air within 2 min. The slides were then exposed to x-ray film (Bmax, Amersham) for 1 week at 4 C, developed in D-19 (Eastman Kodak, Rochester, NY) for 5 min, and fixed for 10 min. Each slide was fixed in formalin and stained with hematoxylin (Gill's no. 3, Sigma) for correlation with the autoradiographically derived data. Aequorin loading and intracellular calcium measurements Aequorin measurements were performed as described in detail previously (12). Briefly, the glomerulosa cells were washed twice in a calcium-free KRB and then successively incubated in four solutions: 1) 120 mM potassium glutamate, 20 mM iV-Tris-(hydroxymethyl)methyl-2-aminoethanesulfonic acid (TES), 10 nM EGTA, 5 mM sodium ATP, and 2 mM magnesium chloride, pH 7.1, for 20 min at 4 C to effect permeabilization; 2) 120 mM potassium glutamate, 20 mM TES, 0.1 mM EGTA, 5 mM sodium ATP, 2 mM magnesium chloride, and 200 /xg/ml aequorin at pH 7.1 for 30 min at 4 C; 3) 120 mM potassium glutamate, 20 mM TES, 0.1 mM EGTA, 5 mM sodium ATP, and 10 mM magnesium chloride at pH 7.1 for 60 min at 4 C; and 4) calcium-free KRB containing 10 mM magnesium chloride and equilibrated with 95% air-5% CO2 for 60 min at room temperature. The extracellular calcium concentration was then restored to 1.25 mM incrementally over a 60-min period, and the cells were allowed to recover for at least 1 h before use. Cells were divided into aliquots of 14-20 X 106 cells, centrifuged, and resuspended in low temperature gelling agarose in a perifusion chamber for calcium measurements. cAMP measurements Glomerulosa cells were purified from contaminating red blood cells using Percoll (56%). Samples containing 1 X 106 cells were preincubated at 37 C for 30 min. The appropriate agents were added, and the incubations were continued for an additional 5 min. The incubations were stopped by adding trichloroacetic acid (TCA) at a final concentration of 5%. The samples were then stored overnight (-80 C).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 January 2016. at 01:02 For personal use only. No other uses without permission. . All rights reserved.
PTH MODULATION OF ALDOSTERONE SECRETION
491
TCA extraction Extraction of TCA was necessary before assay of cAMP. Each sample was sonicated for 15 pulses at a 1.5 setting, 50% duty cycle. A solution containing 2.5 ml tri-N-octylamine and 7.5 ml Freon 11 was mixed 1:1 with the TCA extract (4 C) and vortexed for 1 min to mix the two phases. Samples were spun for 10 min at 2000 rpm in a refrigerated centrifuge, and the top phase was removed and saved for assay. cAMP was measured with a RIA (Biomedical Technologies, Inc., Stoughton, MA). Statistical analysis Secretion studies were analyzed using unpaired Student's t test.
Results Autoradiographic localization of PTH binding PTH receptors were localized in slices of bovine adrenal glands by autoradiography. A representative photograph of an autoradiograph of such a slice is shown in Fig. 1. A microtome-sectioned slice of adrenal tissue was incubated with [125I]PTH and exposed to x-ray film for 7 days, as described in Materials and Methods. The areas in white represent those regions with the highest receptor density, while those in black posses the lowest receptor density. As can be seen, binding of radiolabeled PTH was limited to the outer cortex (Fig. 1, top panel) and could be displaced by the addition of unlabeled PTH (~80% specific; Fig. 1, middle panel), but not by unlabeled ACTH (Fig. 1, bottom panel). No binding was observed in the inner cortex or medulla. To address the issue of receptor specificity in another way, the ability of the ACTH antagonist ACTH-(7-38) to block PTH- and ACTH-stimulated aldosterone secretion was examined. ACTH (0.1 nM) alone caused an increase in aldosterone secretion from 18.4 ± 4.5 to 88.1 ± 19.4 pg/min-million cells. This aldosterone secretory response was blunted in the presence of the ACTH fragment 7-38 (100 nM; from 88.3 ± 9.4 to 57.8 ± 1.6 pg/ min-million cells). In contrast, ACTH-(7-38) caused no significant inhibition of PTH-induced enhancement of Ang-II-mediated aldosterone secretion [0.1 nM Ang-II, 31.5 ± 5.4 pg/min • million cells; 0.1 nM Ang-II plus 0.1 nM PTH, 57.8 ± 2.0 pg/min-million cells; 0.1 nM AngII, 0.1 nM PTH, and 100 nM ACTH-(7-38), 69.3 ± 9.6 pg/min-million cells]. PTH and aldosterone secretion
FlG. 1. PTH preferentially binds to the adrenal cortical layer. Autoradiographic exposure of the total binding of [125I]NNT(Nle818,Tyr34)hPTH-(l-34) to bovine adrenal cortical slices. Binding was evaluated in the absence {top panel) or presence (middle panel) of 1 MM unlabeled PTH or in the presence of 1 MM ACTH-(l-24) (bottom panel). Note the absence of binding to the capsule, high levels to the glomerulosa and outer fasiculata layer, and lower levels throughout the remainder of the cortex. The adrenal medulla was characterized by a very low level of binding of PTH.
Once specific PTH-binding sites had been demonstrated in the adrenal glomerulosa layer, the biological role of these receptors was assessed. A series of secretion studies were performed to examine the ability of the peptide, used alone or in conjunction with Ang-II, to stimulate aldosterone secretion from isolated bovine ad-
renal glomerulosa cells. As illustrated in Fig. 2, in the absence of other secretagogues, PTH over the concentration range of 0.01-10.0 nM elicited a modest increase in aldosterone secretion from glomerulosa cells. In response to a concentration of 10 nM PTH, the control secretory rate of 9.5 ± 0.7 pg/min • million cells was increased to a
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 January 2016. at 01:02 For personal use only. No other uses without permission. . All rights reserved.
PTH MODULATION OF ALDOSTERONE SECRETION
492
Endo'1991 Vol 129 • No 1
LU
70 60
50 •
20 10
0
11
10
9
- Log [PTH] FIG. 2. PTH stimulates aldosterone secretion. Bovine adrenal glomerulosa cells were incubated with PTH (10"n-108 M) in the absence (M) and presence (•) of Ang-II (10 nM). Aldosterone secretion was measured after 60 min of stimulation, and the average rate of secretion was calculated. *, P < 0.01, control vs. PTH and Ang-II vs. Ang-II plus PTH.
value of 34.8 ± 6.5 pg/min• million cells (P < 0.01). By comparison, as also shown in Fig. 2, when increasing concentrations of PTH (0.01-10 nM) were added to cells simultaneously exposed to 10 nM Ang-II, a potentiation of the secretory response was observed. Although in the presence of Ang-II, PTH increased the secretory rate (from 58.6 ± 6.8 to 131.0 ± 14.4 pg/min • million cells; P < 0.01), this change represented an absolute increase in the aldosterone secretory rate of 72 pg/min • million cells, rather than the 25 pg/min • million cell increase observed in the presence of PTH alone. Thus, PTH acted synergistically with Ang-II to stimulate aldosterone secretion. In these studies PTH-(1-34) and PTH-(l-84) were used interchangeably. The magnitude of the synergism varied according to the dose of Ang-II used as well as the time over which secretion was measured. This interaction was evaluated early (after 15 min of stimulation) and later (between 30-45 min of incubation) using a dose of PTH (0.1 nM) that by itself only slightly increased secretion (+15.7 pg/ min-million cells). The concentration of Ang-II was varied between 0.1-10.0 nM. As can be seen in Fig. 3, during the first 15 min of incubation (when the magnitude of the secretory response is thought to depend on the amount of calcium mobilized from the endoplasmic reticulum) (13), PTH shifted the Ang-II-aldosterone doseresponse curve to the left (Ka from 1.0-0.3 nM) without significantly increasing the maximal secretory response (46.4 ± 8.7 vs. 57.8 8.5 pg/min • million cells; P = NS). In contrast, during a sustained response to Ang-II between 30-45 min of stimulation (when the secretory response is maintained by calcium influx through voltage-depend-
1
3
5
7
9
11
[Ang II] nM FIG. 3. PTH shifts the Ang-II-aldosterone dose-response curve. Bovine adrenal glomerulosa cells were incubated with Ang-II in the absence (O) and presence (•) of PTH (0.1 nM). Aldosterone secretion was measured after 15 min of stimulation, and the average rate of secretion was calculated. *, P < 0.01, Ang-II us. Ang-II plus PTH.
11
[Ang II] nM FIG. 4. PTH enhances the maximal rate of Ang-II-stimulated aldosterone secretion. Bovine adrenal glomerulosa cells were incubated with Ang-II in the absence (O) and presence (•) of PTH (0.1 nM). Aldosterone secretion was measured after 30 and 45 min of stimulation in separate incubations, and the average rate of secretion between 30-45 min was calculated. *, P < 0.005, Ang-II vs. Ang-II plus PTH
ent channels) (13), PTH greatly enhanced the maximal secretory response (Ang-II, 109.4 ± 3.4 pg/min-million cells; Ang-II plus PTH, 219.2 ± 13.3 pg/min • million cells; P < 0.005) without shifting the Ka for activation (Fig. 4). Effects of PTH on intracellular Ca2+ and cAMP concentrations In an effort to investigate the mechanism of the stimulatory action of PTH, the hormone was added to ae-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 January 2016. at 01:02 For personal use only. No other uses without permission. . All rights reserved.
PTH MODULATION OF ALDOSTERONE SECRETION quorin-loaded cells in the absence or presence of additional secretagogues, either 15 mM potassium or 10 nM Ang-II. As shown in Fig. 5A, PTH alone (10 nM) caused only a small elevation of intracellular calcium. However, when the cells were pretreated with either Ang-II (Fig. 5B) or 15 mM potassium (Fig. 5C), a much larger and more sustained elevation of intracellular calcium was observed upon the addition of PTH. Since PTH is known to increase the cAMP concentration in its classic target cells, the effect of PTH on the cellular content of cAMP was measured in glomerulosa cells either treated with PTH alone or in the presence of Ang-II. As shown in Fig. 6, bovine PTH-(l-84) at low concentrations (0.1 nM) had a significant effect in increasing the cellular cAMP content (control, 1.54 ± 0.04 pmol/million cells; PTH, 2.15 ± 0.18 pmol/million cells; P < 0.01); cellular cAMP content was not further markedly altered by the addition of higher concentrations of PTH (0.1 nM PTH, 2.15 ± 0.18 pmol/million cells; 10 nM PTH, 2.28 ± 0.06 pmol/million cells) or by Ang-II and PTH (0.1 nM PTH, 2.15 ± 0.18; 0.1 nM PTH plus 10 nM Ang-II, 2.06 ± 0.18). However, in the presence of 1 mM isobutylmethylxanthinee (a phosphodiesterase inhibitor) the percent increase in cellular cAMP in the presence of PTH was enhanced (0.1 nM PTH, 140% without IBMX, 148% with IBMX; 10 nM PTH, 148% without IBMX, 182% with IBMX). Discussion The present data demonstrate that there are specific binding sites for PTH on the bovine adrenal cortex which are functionally linked to signalling systems. Activation of these systems leads to changes in both intracellular calcium and cAMP content. Even though the changes in intracellular calcium and cAMP are modest, and concentrations of PTH alone in the range of 0.1-10 nM evoke only a small aldosterone secretory response, in glomerulosa cells treated simultaneously with Ang-II, the effects
493
of PTH on aldosterone secretion are significantly greater. The basis for the potentiation of the Ang-II-induced aldosterone secretory response is unclear, but may relate to interactions between the second messengers underlying the response to both peptides. Thus, if these PTHgenerated second messengers were to affect the signals mediating the action of Ang-II, their interaction might underlie the PTH-elicited potentiation of Ang-II-induced aldosterone secretion. In fact, there is evidence to suggest that the elevation of cAMP levels produced by PTH might influence one such signal, calcium influx. It has been proposed that an Ang-II-elicited increase in calcium influx together with elevated diacylglycerol levels support sustained aldosterone secretion. Indeed, agents that modulate the activity of voltage-dependent calcium channels, such as the calcium channel agonist BAY K 8644 and the antagonist nitrendipine, can either potentiate (BAY K 8644) or inhibit (nitrendipine) the sustained secretory response to Ang-II (14,15). Interestingly, the activity of the channel for which BAY K serves as a specific agonist, the L-type channel, has been shown to be modulated by a cAMP-dependent protein kinase, with a rise in cAMP enhancing calcium influx. Thus, the PTH-induced increase in cAMP production could, via an activation of cAMP-dependent protein kinase, potentiate the Ang-II-elicited increase in calcium influx through L-channels, thereby potentiating as well the aldosterone secretory response. Support for this suggestion is provided by our study using aequorin-loaded cells. In these experiments, exposure to PTH alone had little effect on intracellular calcium levels. However, if glomerulosa cells were first depolarized by pretreatment with Ang-II or 15 mM K+ (16), PTH elicited a large increase in the cytosolic calcium concentration. Because L-type channels are activated at more depolarized potentials, K+-stimulated cells are sensitized to respond to L-channel agonists such as BAY K 8644. Thus, the ability of depolarizing stimuli to
A.
c. Ang II (10 nM)
3 -