Kinetics of cytosolic calcium and aldosterone responses in rat adrenal glomerulosa cells.

0013-7227/91/1295-2431$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society

Vol. 129, No. 5 Printed in U.S.A.

Kinetics of Cytosolic Calcium and Aldosterone Responses in Rat Adrenal Glomerulosa Cells* STEPHEN J. QUINN, PETER ENYEDI, DOUGLAS L. TILLOTSON, AND GORDON H. WILLIAMS Department of Physiology, Boston University School of Medicine (D.L.T.), Boston, Massachusetts 02118; the Department of Physiology, Semmelweis University Medical School (P.E.), Budapest, Hungary; and the Endocrine-Hypertension Division, Department of Medicine, Brighamand Women's Hospital, and Harvard Medical School, Boston, Massachusetts 02115

ABSTRACT. To evaluate the relationship between cytosolic calcium (CaO and aldosterone production, rat adrenal zona glomerulosa (ZG) cells were studied during long-term stimulation by different secretagogues. Ca; was measured in single ZG cells using microspectrofluorimetry, and aldosterone was determined in cell populations using a superfusion system. For external potassium (K+), Ca; increases are sustained, with only a slight decrement over time, a feature shared by aldosterone production. The relationship between aldosterone output and Ca; is nonlinear, with a Cai value for half-maximal stimulation of approximately 500 nM. Furthermore, the sustained changes in Cas with external K+ indicate that ZG cells can use an amplitude-based Cai signal to stimulate aldosterone production. Ca; changes stimulated by angiotensin-II (Ang-II) show a complex doseresponse pattern, with high concentrations (>1 nM) of Ang-II eliciting a peak-plateau signal and lower doses (0.1 nM to 10 pM) producing repeated Caj oscillations. The peak amplitude of the Caj response in individual cells is not dose dependent, with the ZG cell experiencing peak levels repeatedly at the lowest Ang-II concentrations. However, the Caj transients are more frequent with increasing Ang-II concentrations between 0.1 nM and 10

A

DRENAL zona glomerulosa (ZG) cells integrate external signals from many sources for the regulation of aldosterone production. Angiotensin-II (AngII) and ACTH bind to plasma membrane receptors to initiate the transduction process. Ang-II acts through the phosphoinositide system by activation of phospholipase-C, while ACTH stimulates adenylate cyclase activity (1). Elevation of external potassium (K+) depolarizes the plasma membrane, due to a change in the equilibrium potential for K+ (1). Despite these divergent initial events, external calcium (Ca2+) is required by each of these agents, and there is strong evidence for the involvement of cytosolic calcium (CaO as an intracellular mesReceived April 9,1991. Address all correspondence and requests for reprints to: Dr. Stephen J. Quinn, Endocrine-Hypertension Division, Brigham and Women's Hospital, 221 Longwood Avenue, Boston, Massachusetts 02115. * This work was supported in part by grants from the NIH (DK40127, HL-42120, and HL-42354).

pM. When integrated over time, the mean Cai signal also shows only modest dose-dependency during the sustained phase of AngII stimulation. Unlike the integrated Ca; signal, aldosterone production increases steeply between 10 pM and 0.1 nM Ang-II, indicating that the Ca; signal is likely to be frequency-based. Conversely, the steroid response to high Ang-II closely mirrors the kinetics of the more sustained Cai signals, including the diminished Ca; and aldosterone levels during sustained stimulation with the highest Ang-II doses. Arginine vasopressin stimulated Caj and aldosterone responses, which closely resemble those elicited by 0.1 nM Ang-II, except that both Ca; and aldosterone return to basal values within 20 min of continuous presentation of arginine vasopressin. Each ZG secretagogue produces a distinct pattern of Caj and aldosterone response. In addition, Ca; response patterns can be divided into two general classes: a sustained Cai response, which appears to modulate cell activation by the amplitude of the Ca; signal, and an oscillating Cai response, which uses the frequency of the Cai transients to control the magnitude of stimulation. (Endocrinology 129: 2431-2441,1991)

senger (1, 2). Ca; studies using ZG cell populations have consistently reported increases in Cai in the presence of Ang-II and external K+ (3-12). Ca; responses to ACTH application appears to be species specific; bovine, but not rat, ZG cells display elevations in Ca; (7). Additional secretagogues can be grouped into two major categories depending on their activation of the phosphoinositide [i.e. arginine vasopressin (AVP)] or adenylate cyclase systems (i.e. serotonin); however stimulation of steroidogenesis in ZG cells is dependent on external Ca2+ in all cases (1,13,14). A number of studies have attempted to correlate Ca; responses with second messenger formation and steroid production. In general, establishing the link between Ca; and other cellular events is hindered by the complexity of the Ca; response pattern. For external K+, most investigators report a kinetically simple rise in Cai, which is dose dependent and sustained for the duration of stim-

2431

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 September 2016. at 15:01 For personal use only. No other uses without permission. . All rights reserved.

Ca; AND STEROID RESPONSES IN ZG CELLS

2432

ulation (4, 8, 9). However, studies using single ZG cells have demonstrated that secretagogue-stimulated Cai changes are often far more complex than previously indicated from results of cell population studies (15-18). One striking observation has been the number of cell types, including ZG cells, that display Cai oscillations during stimulation by Ang-II and AVP, which activate the phosphoinositide system (15, 16, 19) during short term stimulation. The objective of this study was to examine the relationship between Ca; responses from single rat ZG cells and the associated aldosterone production, as measured with superfusion of ZG cell populations. The study focuses on Ang-II, external K+, and AVP, which consistently elicit Cai responses in rat ZG cells. The duration of stimulation was sufficiently long so as to make temporal comparisons between Ca; and steroid changes induced by these secretagogues.

Materials and Methods Adrenal cell preparation and aldosterone determination Isolated rat ZG cells were prepared as previously described (20) Briefly, ZG cells were obtained from adrenal capsules of female Sprague-Dawley rats. The capsules were digested with collagenase and DNAse dissolved in the incubation medium described below, followed by mechanical dispersion of the tissue. Aldosterone production of the isolated ZG cells was examined in a system of superfusion columns (21). The incubation solution was HEPES-buffered medium 199 (Gibco, Grand Island, NY) with 3.6 mM KC1, pH 7.4, supplemented with 2 mg/ ml BSA. Cells from six rats were loaded into each column and supervised at a flow rate of 0.2 ml/min. After a 30-min control superfusion, cells were stimulated with AVP, Ang-II, or external K+ for 1 h. Effluent fractions were collected every 2-5 min and analyzed for aldosterone by a RIA kit (Diagnostic Products, Los Angeles, CA). The transit time of the superfusion system (6 min) was determined by radiolabeled Ang-II and was used as a correction factor when determining the kinetics of aldosterone production. Rat ZG cells loaded with fura-2/AM have similar aldosterone output as control conditions when monitored in static incubations (3, 5, 7). Measurement of Cat The protocol for Ca; measurement is similar to the methods described previously (15, 16, 22). Briefly, dispersed ZG cells were incubated for 20 min with fura-2/AM (Molecular Probes, Eugene, OR) at a concentration of 5 nM in Eagle's Minimum Essential Medium including HEPES (20 mM), CaCl2 (1.25 mM), MgSO4 (0.5 mM), and BSA (0.1%), pH 7.47. The cell suspensions were washed and used 1-4 h after loading with dye. A suspension of adrenal cortical cells was placed in a superfusion chamber mounted on a microscope stage, allowed to adhere to its treated glass coverslip bottom, then continuously superfused with warmed (35-38 C) solution (NaCl, 120 mM; KC1, 4 mM; Na2HPO4, 1.2 mM; CaCl2, 1.25 mM; MgSO4, 0.5 mM; HEPES, 20 mM; BSA, 0.1%; glucose, 0.2%; pH 7.4). The test solutions

Endo• 1991 Vol 129 • No 5

consisted of the standard bath solution and varying concentrations of AVP, Ang-II, or external K+. In the case of external K+, there was an equimolar adjustment of NaCl. The test solutions were applied by puffer superfusion, which allows rapid local solution changes around an identified ZG cell (15, 22). Secretagogue concentrations ranged from 10 pM to 10 nM for Ang-II, 4-8 mM for external K+, and 100 nM for AVP, with the duration of stimulation being 10-30 min. Cells selected for Ca; measurements were identified as ZG cells based on morphological characteristics (23). Over 95% of the chosen ZG cells by this criteria responded with a rise in Ca; to elevation of external K+ or application of Ang-II, while 60% of identified ZG cells responded to AVP stimulation (15,16, 22). The whole cell measurement system consisted of a dual wavelength light source (Photon Technology International, South Brunswick, NJ), an inverted microscope equipped for UV fluorimetry (Nikon Diaphot, Garden City, NY), and a photon-counting photomultiplier tube (Photon Technology International). Fura-2 was used to estimate Ca; by the ratiometric method. Excitation wavelengths were centered at 350 and 380 nm, with bandwidths of 5 nm. Calibration constants were determined by an in vitro method, and a Ka of 224 nM was used, as previously reported (24). Estimates of Caj concentrations fell within a range where fura-2 reliably monitors Ca*. Data analysis Ca; responses were analyzed by one-way analysis of variance and the Newman-Keuls multiple comparison test. All Ca; measurements were recorded at 5 data points per second and digitally filtered with a 21-point Savitsky-Golay smoothing routine. Oscillations were defined as a set of Cai transients, with each transient lasting at least 30 sec and having a minimum amplitude change of 50 nM.

Results Individual ZG cells from over 20 different cell preparations were examined. The mean resting Ca; concentration was 97 ± 8 nM (±SEM; n = 149) and was not significantly different for the subpopulations tested at each secretagogue concentration. Each ZG cell was stimulated with at least one concentration of Ang-II, external K+, or AVP. The aldosterone output under control conditions was 0.37 ± 0.02 ng/rat-h (±SEM; n = 26) and was not significantly different for the cell preparations tested at each secretagogue concentration. Cai response to probnaed stimulation by external potassium When external K+ was elevated for a long (30-min) period of stimulation, individual rat ZG cells displayed sustained increases in Ca;, which promptly rose and fell in response to the step changes in external K+ (Fig. 1). The rate of rise of the Caj response was only slightly dose dependent. The initial amplitude of the Cai response elicited by external K+ was maintained in most cases during the 30-min stimulation; however a more transient


i AND STEROID RESPONSES IN ZG CELLS external potassium

2433

mM

800

400

6" •8

1

60 20 40 time, minutes FIG. 2. Time course of the aldosterone responses to different external K+ concentrations. Shown are mean (±SEM) data from three or four separate experiments at each external K+ concentration: 5 mM (D), 6 mM (A), and 8 mM (O). The duration of column superfusion with the test solutions was 55-60 min, as indicated by the bar at the bottom of the graph. The time lag of the superfusion system was taken into account when determining the starting point. Sample intervals were 2 or 5 min, as indicated by the data points.

0

400

0L 400r

0L 30 10 20 time, minutes FIG. 1. Time course of the Cai responses to different external K+ concentrations. Shown are representative Cai recordings from three different ZG cells during puffer superfusion with the indicated concentrations of external K+. The duration of stimulation was 30 min, as indicated by the horizontal bar above the traces. Similar results were found for 9-22 cells at each concentration.

o

initial response was found in some cells (i.e. stimulation with 8 mM K+; Fig. 1). The amplitude of the Cai response was strongly dose dependent, rising sharply between 48 mM external K+ (Fig. 3A). In addition, there was a consistent, but modest, decrease (10-30%) in the response after the maximal rise at all concentrations. Aldosterone production during stimulation by external potassium The results of a set of superfusion experiments (Fig. 2) using the external K+ concentration as the secretagogue indicated that the kinetics of the aldosterone response to external K+ qualitatively resembled the abovementioned C^ responses. There was a steep initial rise in steroid output whose time course was independent of the external K+ concentration. The rate of steroid production gradually diminished after the output reached its maximal values, showing declines of 5-10% after 30 min and 20-35% after 60 min of external K+ stimulation, with the lower K+ concentrations showing the greatest decline in steroidogenesis. As with the Ca; response, a clear dose dependency of the maximal amplitude of the steroid response (Fig. 3A) to external K+ was observed. This relationship is plotted in Fig. 3A and shows a steep rise with increasing K+ concentration. The early kinetics of enhanced aldosterone production were independent of the external K+ concentration and the extent of steroi-

dogenesis stimulated by elevation of K+ (see Fig. 8). The incubation media for steroid and Ca; measurements had slightly different K+ concentrations; however, there was no significant difference in steroid production when the external K+ concentration was raised from 3.6 to 4.0 mM (n = 2). Cai response to prolonged stimulation by Ang-II

Stimulation by Ang-II produced complex Ca; changes in individual ZG cells (Fig. 4). At high Ang-II concentrations (10 and 1 nM), ZG cells displayed an initial Cai transient, which decayed to a plateau level, still above basal Ca; values.. At hormone concentrations of 0.1 nM and below, the Ca; signal made a transition from a relatively stable response to clearly oscillating patterns, particularly at the lower Ang-II doses. These responses continued unabated for the duration of hormone application (30 min). One prominent dose-dependent feature of these Caj responses elicited by Ang-II is the delay between the application of the Ang-II and the first Ca; transient (Fig. 4) (15, 16). At low Ang-II concentrations (0.1 nM and below), another dose-dependent feature is the oscillatory pattern of the Ca; response. The frequency of the Ca; oscillations declines while the percentage of ZG cells displaying oscillations increases as the dose of Ang-II is reduced (Table 1). For responsive ZG cells showing a peak-plateau Ca; response, the plateau Ca; level has a tendency to decrease as the Ang-II concentration increases. As noted in previous reports, the amplitude of the peak Cai transients show little dose dependency. This result was confirmed in the present study, with no significant differences found between 10 nM Ang-II and 10 pM Ang-II (Table 1). When the Cai responses to Ang-II are averaged, much



2434

700

Endo • 1991 Vol 129 • No 5 angiotensin II

500

10 nM

0 1000

lnM

external potassium, mM

60

B

40

'''

g 20 10 20 time, minutes

CJ O

2 CO

o

*

0 •

•

•

1

1

1

100 300 500 700 cytosolic calcium, nM

FIG. 3. Relationship among Cai, aldosterone output, and external K+ concentration. A, Peak aldosterone production (O) and peak Ca; levels are plotted against the external K+ concentration. Peak Ca; levels are mean (±SEM) values from 9-22 cells/external K+ concentration, while peak steroid values represent mean (±SEM) values from 2-4 separate experiments. The 2 data points for 8 mM K+ are superimposed. For some data points, the SEM are included in the symbol. B, Aldosterone production is plotted against the mean Ca; level found at each external K+ concentration, using mean data from A for each external K+ concentration.

of the detail found in the individual records is lost (Figs. 5, 7B, and 8A). However, the averaged Caj response does provide the temporal information for a representative ZG subpopulation and also allows for a simplified integration of the Cai change. The averaged Caj change closely resembles the individual cell records for stimulations of 1 nM and above, while this relationship is lost at lower hormone concentrations due to a lack of synchrony of the Cai signal for individual ZG cells. The sustained

30

FIG. 4. Time course of the Ca; responses to different Ang-II concentrations. Traces are representative Cai recordings from 5 separate ZG cells during puffer superfusion with the indicated concentrations of Ang-II. The duration of stimulation was 30 min, as indicated by the horizontal bar above the traces. Similar results were found for 10-15 cells at each hormone concentration.

phase of the averaged Ca4 response was similar for all Ang-II concentrations. Aldosterone production during stimulation by Ang-II Ang-II elicited a robust steroidogenic response during the 60 min of column superfusion (Fig. 6). Steroid output was stimulated more than 2-fold by 10 pM Ang-II (data not shown) and greater than 100-fold by 1 nM Ang-II. From a basal rate of aldosterone production of 0.37 ng/ rat-h, aldosterone output at the end of a 60-min superfusion with Ang-II was 4.6 ng/rat-h for 30 pM Ang-II, 22.8 ng/rat-h for 100 pM, 40.5 ng/rat-h for 1 nM, and 29.6 ng/rat-h for 10 nM (Fig. 6). Examination of the temporal features of the aldosterone responses stimulated by Ang-II reveals complexities not found with K+ stimulation (Fig. 6). At the highest Ang-II concentrations, there was a rapid increase in steroid production; however, the aldosterone response was somewhat transient. The steroid output declined by


i AND STEROID RESPONSES IN ZG CELLS

2435

TABLE 1. Cai response characteristics for Ang-II stimulation of rat ZG cells Ang-II cone. 10 1 100 30 10

nM nM pM pM pM

10 10 24 12 10

Responsive (%) 100 100 85 83 80

Basal Cai

Peak Cai

(nM)

(nM)

73 58 73 95 140

±19 ±25 ±20 ±32 ± 35

334 ± 265 ± 267 ± 378 ± 364 ±

40 46 41 70 86

Oscillatory (%) 0 0 59 80 100

Oscillations (frequency)

0.64 ± 0.05 0.45 ± 0.03 0.37 ± 0.04

Plateau Cai (nM)

95 ± 3 3 150 ± 42 197 ± 46 238 ± 88

Cai responses of individual rat ZG cells (n = number) during stimulation with four different Ang-II concentrations. Ang-II was applied for 30 min, except for 100 pM Ang-II, where stimulations ranged from 10-30 min. The percentage of responsive (responsive) cells is indicated as well as the percentage of responsive cells that gave oscillatory responses (oscillatory). The remaining responsive cells demonstrated more stable Cai recordings. The basal Ca; values are the mean (±SEM) basal levels for the subpopulation of cells tested with each Ang II concentration. The peak Cai values are the mean maximal (±SEM) changes above basal levels for individual cell records. The plateau Ca; values are the mean (±SEM) Caj increases above basal levels after 30 min of Ang-II stimulation for those responsive ZG cells that did not oscillate. The oscillation (oscillations) frequency is the mean (±SEM) number of Caj transients per min recorded from the initial Ca; transient to the end of Ang-II stimulation. 300 200

FIG. 5. Mean Caj responses to different Ang-II concentrations. Each graph represents the mean (±SEM) values from 10-15 individual ZG cells tested at the hormone concentrations and stimulation durations indicated in Fig. 4. Sample intervals were every 0.5-2 min. The AngII concentration is noted in the righthand corner of each graph. The last graph includes the mean data points from all 5 Ang-II concentrations: 10 nM (•), 1 nM (A), 100 pM (•), 30 pM (T), and 10 pM

(•)•

300 100 pM 200

100

10 20 time, minutes

25% at 30 min and 55% at 60 min with 10 nM Ang-II, and by 10% at 30 min and 40% at 60 min with 1 nM Ang-II. This decaying phase of steroid output at high Ang-II concentrations was greater than that found at all external K+ concentrations tested. At the lower Ang-II concentrations, the elevation of aldosterone production began at a later time and was much slower (Fig. 6). In addition, the maximal steroid output decreased with Ang-II concentration (Fig. 7A). Interestingly, a decline from peak steroid output was not observed at these lower

30

10 20 time, minutes

doses over the 60-min measurement period. At 10 pM Ang-II, aldosterone production more than doubled from basal levels at later time periods; however, the response was too small to perform kinetic analysis (data not shown). When the steroid responses were normalized to the peak output at each Ang-II concentration, a pronounced delay was demonstrated which was strongly dose dependent (Fig. SB). Analysis of the response delay for cytosolic calcium, the half-time (t1/2) to the peak steroid


Endo • 1991 Voll29«No5


2436

s PI

4

g 60 20 40 time, minutes FlG. 6. Time course of the aldosterone responses to different Ang-II concentrations. Shown are mean (±SEM) data from three or four separate experiments at each Ang-II concentration: 30 pM (A), 100 pM (D), 1 nM (A), and 10 nM (O). The duration of column superfusion with the test solutions was 55-60 min. as indicated by the bar at the bottom of the graph. The time lag of the superfusion system was taken into account when determining the starting point. Sample intervals were 2 or 5 min, as indicated by the data points. The final aldosterone production rates were 4.6 ± 0.8 ng/rat-h for 30 pM Ang-II, 22.8 ± 4.1 for 100 pM, 40.5 ± 9.7 for 1 nM, and 29.6 ± 4.8 for 10 nM.

0

production, and the rate of rise of aldosterone output all showed this dose-dependency, with slower kinetics at lower Ang-II concentrations. Conversely, the kinetics of the early steroid response in external K+ stimulation showed no dose-dependency (Fig. 8C). Due in part to the kinetics of the early phase of steroidogenesis and the aldosterone decline seen at later times with high Ang-II concentrations, the steroid response in Ang-II displayed a shift in sensitivity over time during prolonged stimulation by the hormone. The EC50 for Ang-II was close to 0.5 nM for the first 20 min, but was less than 0.1 nM for the last 20 min of stimulation. In contrast, the sensitivity remained constant during prolonged stimulation by external K+. Ccii response to prolonged stimulation by A VP

Long stimulation with a maximal steroidogenic dose of AVP (100 nM) produced a transient Ca; response in rat ZG cells (Figs. 9 and 11). In the first 10 min, the Ca* changes were similar to those elicited by 0.1 nM Ang-II and were characterized by a response delay of roughly 30 sec and an initial rapid Caj transient, followed by additional Cai transients in some cases. Unlike the responses to Ang-II, the Cai changes stimulated by AVP were shortlived, with a termination of Cai transients and a decline to resting Cai values usually within 20 min (n = 7). Aldosterone production during stimulation by AVP With 100 nM AVP, steroid production was stimulated to a modest extent and for a brief period of time (Figs. 10 and 11). Aldosterone output rose 5-fold from basal levels, reaching a peak value by the 7 min mark. This peak response was followed by a decline in steroid production, which returned to basal values within 30 min

o o u o -u CO

O J

O.lnM lOnM angiotensin II 300

0

O.lnM lOnM angiotensin II

FlG. 7. Relationship among Cas, aldosterone output, and Ang-II concentration. A, Steroid production is plotted against Ang-II concentration for peak (O) and final (A) aldosterone output. Data points represent mean (±SEM) values from three or four separate experiments. The peak aldosterone production rates were 4.87 ± 0.9 for 30 pM Ang-II, 27.7 ± 4.8 for 100 pM, 73.6 ± 16.1 for 1 nM, and 61.4 ± 9.0 for 10 nM. The final steroid output rates were 4.6 ± 0.8 ng/rat • h for 30 pM AngII, 22.8 ± 4.1 for 100 pM, 40.5 ± 9.7 for 1 nM, and 29.6 ± 4.8 for 10 nM. B, Mean Cai responses are plotted against Ang-II concentration for peak Cai changes (•) and plateau or final Ca; increases (A).

and remained at control values for the next 30 min. The response delay and the rate of rise of steroid output were similar for 100 nM AVP and 0.1 nM Ang-II; however, the ti/2 was much shorter for 100 nM AVP than for 0.1 nM Ang-II, since the aldosterone response to Ang-II continued to rise during prolonged stimulation, while the response to AVP began to decline within 10 min.

Discussion Adrenal glomerulosa cells respond to a wide array of secretagogues, which require external Ca2+ to stimulate



2437 vasopressin, 100 nM

800

400 0 400r § 200

L

400 200

200

10 20 time, minutes

30

FIG. 9. Time course of the Caj response to stimulation by vasopressin. Traces are representative Ca; recordings from 4 different ZG cells during puffer superfusion with 100 nM AVP. The duration of stimulation was 30 min, an indicated by the horizontal bar above the traces. Similar results were found for all 10 ZG cells that responded to this concentration of AVP. 20

40

60

time, minutes FIG. 8. Kinetics of the early responses of Ca; and aldosterone. A, Mean Cai responses to different Ang-II concentrations, as shown in Fig. 5, are presented with an expanded time scale to better illustrate the dosedependent delay in the Ca response with Ang-II stimulation. Data are included from the following Ang-II concentrations: 10 nM (•), 1 nM (A), 100 pM (•), 30 pM (T).; and 10 pM (•). B, Aldosterone data are normalized to the peak steroid output at each Ang-II concentration and plotted us. time. The half-time (ti/2) for the initial rise in steroid production is the point at which the normalized aldosterone output is 0.5. These values are for different Ang-II concentrations: 5 min, 10 nM; 6 min, 1 nM; 12 min, 100 pM; and 24 min, 30 pM. C, Aldosterone data are normalized to the peak steroid output at each external K+ concentration and plotted vs. time. The half-time (ti/2) for the initial rise in steroid production is the point at which the normalized aldosterone output is 0.5. These values are for different external K+ concentrations: 5.5 min, 8 mM; 6 min, 6 mM; and 7 min, 5 mM. The duration of secretagogue stimulation is indicated by the bar at the bottom of each graph. Note the difference in the time scales for A us. B and C.

aldosterone production. This dependency on external Ca2+ probably is linked to the use of Ca; as an intracellular messenger for external K+, Ang-II, AVP, and, in some species, ACTH. Understanding the relationship between Cai changes and steroid production during stimulation with secretagogues has been complicated by diverse kinetics reported in ZG cell populations, ranging

from relatively sustained increases in Ca; (3, 4, 8, 9, 18, 22) to brief transient responses (11). In addition, the duration of the Ca; experiments has tended to be substantially shorter (5-15 min) than the incubation time for steroid determinations (1-2 h). Despite these shortcomings, the correlation between Ca; and aldosterone output has been reported to be relatively straightforward, especially when sustained increases in Caj were found. Ca; measurements in single ZG cells are critically important for our understanding of Ca2+ signalling, given the complexity and diversity of the Ca; response patterns observed with secretagogue stimulation. Likewise, aldosterone measurements using a column superfusion system allows for better understanding of the kinetics of steroidogenesis and the relationship between intracellular messengers and aldosterone production. However, isolated ZG cells may respond differently from more dense cell preparations or the intact adrenal gland for a variety of reasons. For example, cells from the adrenal gland produce a number of local factors, including insulin-like growth factor and Ang-II, which may influence both Ca; and aldosterone responses. Despite these complications, Ca; measurements using microspectrofluorimetry and aldosterone determinations using column superfusion are the techniques currently favored for in vitro research.


2438

Ca; AND STEROID RESPONSES IN ZG CELLS 300

A


II.

100 nM A V P .

20 £" ° 3 S

100

0

cytosol

u

10

20

30 300 Q

30

100 pM Ang II

S 20 10 00000 6 0 0 0 0 0 —o-

100 nM AVP o—o—o—o—o

2

0

20

30

time, minutes

0

40 60 20 time, minutes FIG. 10. Time course of the aldosterone response to vasopressin stimulation. Shown in A and B are mean ( ± S E M ) data from four separate experiments using 100 nM AVP (O) and 100 pM Ang-II (A). The duration of column superfusion with the test solutions was 55-60 min, indicated by the bar at the bottom of the graph. The time lag of the superfusion system was taken into account when determining the starting point. Sample intervals were 2 or 5 min, as indicated by the data points. The scale of aldosterone output was changed in the two graphs to better demonstrate the steroid change to AVP stimulation as well as the relationship between AVP and Ang-II stimulation.

External potassium In the present study Ca; responses to external K+ were monitored in single rat ZG cells for 30 min of stimulation. Ca; increased monotonically with the external K+ concentration between 4-8 mM. The kinetics of these Ca; responses were simple, consisting of an immediate rapid rise in Ca; and a sustained elevation, which declined to a minor degree during the 30 min of elevated K+. The characteristics of these Caj responses agree with most previous reports on ZG cell populations, though not all. Because the kinetics of the Caj responses are not dose dependent, external K+ appears to deliver its steroidogenic signal as an amplitude-based Ca; change in the ZG cell. This interpretation is consistent with steroid results, which demonstrate a similar dependence of aldosterone output on external K+ concentration. While both Ca; and aldosterone levels are elevated, these two parameters are not directly proportional, as reported in previous studies (3, 4, 8). Ca; rises more steeply at lower K+ concentrations than does aldosterone production. Moreover, other studies indicate that Ca{ continues to increase sharply between 8-10 mM K+ (22), while aldosterone

FIG. 11. Relationship between Caj and aldosterone output during stimulation by AVP. A, The mean Caj response (±SEM) from 10 different ZG cells is plotted against time. B, The mean Caj change (•) and aldosterone output (O) during stimulation with 100 nM AVP are plotted together over time. The duration of AVP stimulation is indicated by the bar at the bottom of each graph.

production begins to plateau (1). Taken together, these data suggest that half-maximal steroid production is attained with a Cai level close to 500 nM. The kinetics of the Ca; response and steroid output are qualitatively well matched for their initial rise, sustained levels, and modest declines over time. Unlike stimulation with Ang-II, the ti/2 of the steroid response is not dependent on the external K+ concentration, indicating a direct pathway to the steroidogenic pathway, which can elicit a prompt steroid response. Ang-II The relationship between Ca; and aldosterone output during Ang-II stimulation is far more complex. When examining the effects of high doses (>1 nM) of Ang-II, previous studies using ZG cell populations have reported either a transient increase in Cai, which decayed to basal levels within 10 min (10-12), or an initial peak response followed by a plateau Cai somewhat above basal levels (4-6, 8, 9). In a few cases, C^ responses to a wide range of Ang-II concentrations were documented, showing that lower doses of Ang-II elicited a gradual rise in Ca; to a plateau level without the initial peak response (15). Typically, the peak Ca; responses were compared to cumulative aldosterone output at one or more concentrations of Ang-II. While strong correlations between peak Ca; values and steroidogenesis are found, the simplest interpretation of these data (i.e. a causal relationship between peak Ca; and steroid output) is certainly incorrect, as a



dose-dependency for a change in peak Ca; is lacking at the single cell level, while the lack of dose-dependency for sustained Ca; responses is ignored. Single ZG cell studies have revealed complex Ca; responses to Ang-II, including Ca; oscillations. However, the peak Ca; amplitudes in these individual ZG cells are not dose dependent; rather, it is the waveform of the Cai response that depends on the dose (15). The differences in the peak Ca; amplitudes demonstrated in ZG cell populations are a function of the asynchrony of the individual ZG responses, not the peak Cai level reached in each ZG cell. As with studies examining external K+, most studies of Ang-II have examined Ca; for shorter periods of time than steroid production. In the present study, maintained but complex Ca; patterns were observed when single rat ZG cells were examined during 30 min of Ang-II stimulation, similar to previous reports in these cells (15,16). Although there is a continuum of Ca; responses across a wide range of Ang-II concentrations (10 nM to 10 pM), the Cai patterns can be divided into responses to low (1 nm) doses of secretagogue. At low doses, the oscillation frequency increases with Ang-II concentration, while the percentage of ZG cells displaying sustained vs. oscillatory patterns also increases. A sustained or peakplateau response is typically found at high Ang-II doses, with the plateau Cai level inversely related to the Ang-II concentration. Interestingly, averaged Cai responses show little difference in their plateau levels despite these changes in oscillation behavior and the transition from oscillatory to peak-plateau waveforms. Likewise, the average peak calcium response also shows little dose-dependency. Another prominent feature of these Ca; responses is the response delay, which is brief (3 min) at the lowest doses. As previously reported, the peak Ca; amplitude is similar at all Ang-II doses. Steroid output during Ang-II stimulation shows more complex kinetics than with elevation of external K+. Initial increases in aldosterone production occur after a distinct delay, with slower kinetics than the response stimulated by external K+. These initial steroid responses parallel the delay in the Caj response; however, the steroid response lags behind the Caj changes, especially at low doses of Ang-II, possibly requiring the buildup of the steroidogenic capacity when the ZG cell experiences Ca; oscillations. The magnitude of the aldosterone change is clearly dose dependent, with 1 nM Ang-II stimulating the greatest steroid production and 10-100 pM Ang-II causing the smallest stimulation. Interestingly, the EC5o for Ang-II-stimulated steroid output shifts to the left when stimulation is maintained for 60 min due to the gradual increase in steroidogenesis at the low Ang-II doses and the gradual decline at high Ang-II

2439

doses. A similar shift was seen for Ca;. At low Ang-II concentrations, the common Ca; pattern is oscillatory. The ZG cell may interpret these Ca; patterns in two basic ways, as an integrator or a frequency sensor. If the ZG cell is a strict integrator, the averaged Ca; response during stimulation by Ang-II and the resultant steroid response should be comparable to the relationship between Ca; and aldosterone output during K+ stimulation. This does not appear to be the case. The mean plateau calcium values near 150 nM at all Ang-II concentrations except 10 pM, which is approximately 100 nM above baseline values. However, there are significantly different rates of steroidogenesis at the end of a 60-min incubation for all Ang-II doses. Indeed, the mean Cai amplitude change with 0.1 nM Ang-II is close to 150 nM, while steroid production is between 20-30 ng/h-rat; however, 5 mM external K+ produced a larger (>200 nM) sustained Ca; increase, but far less aldosterone (8 ng/h • rat). Since the ZG cell does not appear to be a strict integrator of the Ca; response to low doses of Ang-II, it is possible that the frequency of oscillations plays a role in the regulation of steroidogenesis, perhaps through the time constant of enzymes comprising the steroidogenic pathway. With high Ang-II doses, the Ca; responses consist of a peak-plateau or sustained waveforms. As with external K+, the Ca; and steroid changes display parallel kinetics, including a decline in steroidogenesis at later times. This decrease in aldosterone production with the highest AngII concentrations cannot result from exhaustion of the steroidogenic capacity of the cells, as steroid output can be sustained at a comparable level with high concentrations of external K+. The present data support the hypothesis that high doses of Ang-II blunt the sustained Ca; response through the blocking of voltage-dependent Ca2+ channels (25) or some other influx pathway, which leads to a diminished steroid output. Other potential mechanisms for this decline include inhibition of adenylate cyclase (26-28) or down-regulation of Ang-II receptors (29, 30). An alternative interpretation may involve the recruitment of ZG cells, in the steroidogenic response when the Cai change makes the transition from an oscillatory to a sustained waveform. The percentage of ZG cells eliciting sustained Caj responses is proportional to the Ang-II concentration for 1 nM and below. At 10 pM, most ZG cells are responsive, and all demonstrate pronounced Ca; oscillations; however, there is a barely measurable increase in aldosterone production. As the Ang-II dose increases, more ZG cells may be recruited to produce additional aldosterone when a sustained Cai response is elicited. At higher Ang-II concentrations, all ZG cells produce a sustained Ca; response, and the amplitude of this sustained Ca; level may be the principle determinant



2440

of steroid output. Clearly, measurement of aldosterone in single ZG cells is required to assess the role of recruitment in the steroidogenic response to physiological levels of Ang-II. The impact of protein kinase-C activation on steriodogenesis remains problematic for any correlative relationship between Ca; and aldosterone. Most studies find that pharmacological activation with phorbol esters and diacylglycerol analogs has little or no effect alone, but must be coupled to an increase in Cai to elicit their full steroidogenic effect (31). Other studies indicate that substantial down-regulation of protein kinase-C has little effect on aldosterone production, with subsequent stimulation by Ang-II (32). It is not clear whether the large changes in steroid output found over a wide range of Ang-II concentrations are due primarily to the distinct Ca; response patterns or to a combination of protein kinase-C activation and those Ca; responses. Likewise, other transduction mechanisms, such as the modulation of adenylate cyclase activity, changes in intracellular sodium and hydrogen ion concentrations, as well as possible physiological actions of inositol polyphosphate metabolites, may regulate steroidogenesis independently or through interactions with the Cai signal. Vasopressin AVP produces a distinctly different set of Ca; and steroid responses, which are transient. Most studies find that AVP is a weak aldosterone secretagogue in rat ZG cells (14, 16, 33, 34). This hormone activates the phosphoinositide system, but to a much lesser extent than Ang-II. Superfusion studies detect only a transient increase in steroid output with AVP, despite continued production of inositol phosphates (16, 34). The present findings confirm the weak steroidogenic nature of AVP and indicate that the transiency of the aldosterone response may be due to the termination of the Ca; response. A concentration of 100 nM AVP elicits an initial Cai response similar to that to 100 pM Ang-II; however, the Ca; oscillations die out within 15 min, and Caj returns to basal levels (16). The steroid pattern mirrors the kinetics of the averaged Cai change, except for a temporal lag in the steroid response. Interestingly, previous studies have shown that AVP produces a sustained activation of phospholipase-C and, presumably, protein kinase-C, yet both Cai and steroid responses are transient. Summary Adrenal ZG cells produce distinctive Ca; and steroid responses to external K+, Ang-II, and AVP. External K+ elicits Cai responses that display strong dose-dependency in the amplitude of the Cai change. The steroid responses closely parallel these increases in Ca;, although the relationship does not appear to be linear. Ang-II produces


Ca; signals consisting of Ca; oscillations at low doses and peak-plateau waveforms at high doses of secretagogue. The ZG cell does not appear to act as a simple integrator of the oscillatory Ca; responses, indicating that the oscillation frequency may be sensed by the ZG cell and contribute to modulation of steroid production. AVP produces transient changes in Ca; and aldosterone, despite sustained activation of the phosphoinositide system. These results stress the importance of Ca; as an intracellular messenger and the ability of ZG cells to produce secretagogue-specific Ca; responses. Moreover, ZG cells appear to have the capacity to interpret diverse Cai signals based on sustained and oscillatory patterns. Acknowledgments Expert technical assistance was provided by Alpheen Menarchery, Tham Yao, and Paul Lartouris.

References 1. Quinn SJ, Williams GH 1988 Regulation of aldosterone secretion. Annu Rev Physiol 50:409-426 2. Barrett PQ, Bollag WB, Isales CM, McCarthy RT, Rasmussen H 1989 Role of calcium in angiotensin II-mediated aldosterone secretion. Endocr Rev 10:496-518 3. Braley L, Menachery A, Brown E, Williams G 1984 The effect of extracellular K+ and angiotensin II on cytosolic Ca++ and steroidogenesis in adrenal glomerulosa cells. Biochem Biophy Res Commun 123:810-815 4. Capponi AM, Lew PD, Jornot L, Vallotton MB 1984 Correlation between cytosolic free Ca++ and aldosterone production in bovine adrenal glomerulosa cells. J Biol Chem 259:8863-8869 5. Kramer RE 1988 Angiotensin II causes sustained elevations in cytosolic calcium in glomerulosa cells. Am J Physiol 255:E338E346 6. Balla T, Hausdorff WP, Baukal AJ, Catt KJ 1989 Inositol polyphosphate production and regulation of cytosolic calcium during the biphasic activation of adrenal glomerulosa cells by angiotensin II. Arch Biochem Biophys 270:398-403 7. Kojima I, Ogata E 1986 Direct demonstration of adrenocorticotropin-induced changes in cytoplasmic free calcium with aequorin in adrenal glomerulosa cells. J Biol Chem 261:9832-9838 8. Capponi AM, Lew PD, Vallotton MB 1987 Quantitative analysis of the cytosolic-free-Ca2+-dependency of aldosterone production in bovine adrenal glomerulosa cells: different requirements for angiotensin II and K+. Biochem J 247:335-340 9. Pratt JH, Rothrock JK, Dominguez JH 1989 Evidence that angiotensin II and potassium collaborate to increase cytosolic calcium and stimulate the secretion of aldosterone. Endocrinology 125:2463-2469 10. Apfeldorf WJ, Rasmussen H 1988 Simultaneous determination of intracellular free calcium and aldosterone production in bovine adrenal zona glomerulosa. Cell Calcium 9:71-80 11. Hausdorff WP, Catt KJ 1988 Activation of dihydropyridine-sensitive calcium channels and biphasic cytosolic calcium responses by angiotensin II in rat adrenal glomerulosa cells. Endocrinology 123:2818-2826 12. Ganz MB, Rasmussen J, Bollag WB, Rasmussen H 1990 Effect of buffer systems and pHi on the measurement of [Ca2+]j with fura 2. FASEB J 4:1638-1644 13. Ganguly A, Hampton T 1985 Calcium-dependence of serotoninmediated aldosterone secretion and differential effects of calcium antagonists. Life Sci 36:1459-1464 14. Guillon G, Balestre M-N, Chouinard L, Gallo-Payet N 1990 Involvement of distinct G-proteins in the action of vasopressin on


Ca; AND STEROID RESPONSES IN ZG CELLS rat glomerulosa cells. Endocrinology 126:1699-1708 15. Quinn SJ, Williams GH, Tillotson DL 1988 Calcium oscillations in single adrenal glomerulosa cells stimulated by angiotensin II. Proc Natl Acad Sci USA 85:5754-5758 16. Quinn SJ, Enyedi P, Tillotson DL, Williams GH 1990 Cytosolic calcium and aldosterone response patterns of rat adrenal glomerulosa cells stimulated by vasopressin: comparison with angiotensin II. Endocrinology 127:541-548 17. Connor JA, Cornwall MC, Williams GH 1987 Spatially resolved cytosolic calcium response to angiotensin II and potassium in rat glomerulosa cells measured by digital imaging techniques. J Biol Chem 262:2919-2927 18. Johnson EIM, Capponi AM, Vallotton MB 1989 Cytosolic free calcium oscillates in single bovine adrenal glomerulosa cells in response to angiotensin II stimulation. J Endocrinol 122:391-402 19. Berridge MJ, Galione A 1988 Cytosolic calcium oscillators. FASEB J 2:3074-3086 20. Williams GH, McDonnell LM, Raux MC, Hollenberg NK 1974 Evidence for different angiotensin II receptors in rat adrenal glomerulosa and rabbit vascular smooth muscle cells. Circ Res 36:384-390 21. Enyedi P, Szabo B, Spat A 1985 Reduced responsiveness of glomerulosa cells after prolonged stimulation with angiotensin II. Am J Physiol 248:E209-E214 22. Quinn SJ, Williams GH, Tillotson DL 1988 The calcium response to external potassium in single adrenal glomerulosa cells. Am J Physiol 255:E488-E495 23. Tait JF, Tait SAS, Gould RP, Mee RSR 1974 The properties of adrenal zona glomerulosa cells after purification by gravitational sedimentation. Proc R Soc Lond [Biol] 185:375-407 24. Grynkiewicz G, Poenie M, Tsien RY 1985 A new generation of Ca2+ indicators with greatly improved fluorescence properties. J

2441

Biol Chem 260:3440-3450 25. Quinn SJ, Cronwall MC, Williams GH 1987 Electrophysiological responses to angiotensin II of isolated rat adrenal glomerulosa cells. Endocrinology 120:1581-1589 26. Woodcock EA, Johnston CI 1984 Inhibition of adenylate cyclase in rat adrenal glomerulosa cells by angiotensin II. Endocrinology 115:337-341 27. Hausdorff WP, Sekura RD, Aguilera G, Catt KJ 1987 Control of aldosterone production by angiotensin II is mediated by two guanine nucleotide regulatory proteins. Endocrinology 120:1668-1678 28. Marie J, Jard S 1983 Angiotensin II inhibits adenylate cyclase from adrenal cortex glomerulosa zone. FEBS Lett 159:97-101 29. Crozat A, Penhoat A, Saez JM 1986 Processing of angiotensin II (A-II) and (Sar\Ala8)A-II by cultured bovine adrenocortical cells. Endocrinology 118:2312-2318 30. Bianchi C, Gutkowska J, De Lean A, Ballak M, Anand-Srivstava MB, Genest J, Cantin M 1986 Fate of [125I]angiotensin II in adrenal zona glomerulosa cells. Endocrinology 118:2605-2607 31. Kojima I, Kojima K, Kreutter D, Rasmussen H 1984 The temporal integration of the aldosterone secretory response to angiotensin occurs via two intracellular pathways. J Biol Chem 259:1444814457 32. Nakano S, Carvallo P, Rocco S, Aguilera G 1990 Role of protein kinase C on steroidogenic effect of angiotensin II in the rat adrenal glomerulosa cell. Endocrinology 126:125-133 33. Woodcock EA, McLeod JK, Johnston CI 1986 Vasopressin stimulates phosphatidylinositol turnover and aldosterone synthesis in rat glomerulosa cells: comparision with angiotensin II. Endocrinology 118:2432-2436 34. Enyedi P, Balla T, Antoni FA, Spat A 1988 Effect of angiotensin II and arginine vasopressin on aldosterone production and phosphoinositide turnover in rat adrenal glomerulosa cells: a comparative study. J Mc'l Endocrinol 1:117-124


Cytosolic calcium and aldosterone response patterns of rat adrenal glomerulosa cells stimulated by vasopressin: comparison with angiotensin II.

Atrial natriuretic peptide inhibition of calcium ionophore A23187-stimulated aldosterone secretion in rat adrenal glomerulosa cells.

Calcium channels and control of cytosolic calcium in rat and bovine zona glomerulosa cells.

Background calcium permeable channels in glomerulosa cells from adrenal gland.

Role of voltage-gated calcium channels in the regulation of aldosterone production from zona glomerulosa cells of the adrenal cortex.

Serotonin stimulates calcium influx in isolated rat adrenal zona glomerulosa cells.

Dual effects of dopamine in rat adrenal glomerulosa cells.

Angiotensin II exerts its effect on aldosterone production and potassium permeability through receptor subtype AT1 in rat adrenal glomerulosa cells.

Pyridine nucleotide redox state parallels production of aldosterone in potassium-stimulated adrenal glomerulosa cells.

K+ pump in rat adrenal glomerulosa cells: possible contribution to stimulation of aldosterone production.

Phospholipase D activity underlies very-low-density lipoprotein (VLDL)-induced aldosterone production in adrenal glomerulosa cells.

Signal transduction mechanisms involved in carbachol-induced aldosterone secretion from bovine adrenal glomerulosa cells.

Actions of angiotensin II antagonists upon aldosterone production by isolated adrenal glomerulosa cells.

Na(+)-H+ exchanger kinetics in adrenal glomerulosa cells and its activation by angiotensin II.

Role of angiotensin II receptor subtypes on the regulation of aldosterone secretion in the adrenal glomerulosa zone in the rat.

Effect of the endothelins on aldosterone secretion by rat zona glomerulosa cells in vitro.

Effect of protein kinase inhibitors on ACTH-stimulated aldosterone production in rat zona glomerulosa cells.

Angiotensin II receptor-mediated calcium influx in bovine adrenal glomerulosa cells.

Transforming growth factor-beta 1 inhibits aldosterone and stimulates adrenal renin in cultured bovine zona glomerulosa cells.

A potential role for phospholipase-D in the angiotensin-II-induced stimulation of aldosterone secretion from bovine adrenal glomerulosa cells.

Effects of high potassium concentration and dihydropyridine Ca2(+)-channel agonists on cytoplasmic Ca2+ and aldosterone production in rat adrenal glomerulosa cells.

Metabolism of arachidonic acid by rat adrenal glomerulosa cells: synthesis of hydroxyeicosatetraenoic acids and epoxyeicosatrienoic acids.

Thapsigargin-induced increase in cytoplasmic Ca2+ concentration and aldosterone production in rat adrenal glomerulosa cells: interaction with potassium and angiotensin-II.

Regulation of renin gene expression in rat adrenal zona glomerulosa cells.