A model for studying regulation of aldosterone secretion: Freshly isolated cells or cultured cells?

Cellular Signalling VoL 5, No. 5, pp. 651-666, 1993. Printed in Great Britain,

0898-6568/93 $6.00 + .00! © 1993 Pergamon Press Ltd

A MODEL FOR STUDYING REGULATION OF ALDOSTERONE SECRETION: FRESHLY ISOLATED CELLS OR CULTURED CELLS? NICOLE GALLO-PAYET,* MARCEL-DANIEL PAYET,~ Lucre CHOUINARD,* MARIE-NOELLE BALESTRE~ and GILLES GUILLON~§ *Endocrine Service, Department of Medicine, and tDepartment of Physiology and Biophysics, Faculty of Medicine, University of Sherbrooke, Sherbrooke, Quebec, Canada JIH 5N4 and ,Centre CNRS-INSERM Pharmacologie-Endocrinologie, Rue de la Cardonille, 34094 Montpellier Cedex, France (Received 18 December 1992; and accepted 11 March 1993) Abstract--Practically all studies relating to zona glomeruiosa function have been performed either with freshly isolated cells or with cells used after 2 or 3 days in culture. This study compares the step-by-step response (binding, second messenger production and aldosterone response) of isolated glomerulosa cells vs cells maintained in primary culture to the main stimuli of aldosterone secretion. One day in culture induces a decrease of 77 and 65% in the basal level of corticosterone and aldosterone secretions, compared to that observed in freshly isolated cells. In these conditions, the cells become more sensitive to most of their stimuli, but not all: e.g. important differences are noted in the dose-response of aldosterone secretion to adrenocorticotropin (ACTH), which is often shifted to a lower concentration sensitivity in cultured cells. For example, 0.1 nM ACTH stimulates steroid secretion by three-fold in isolated cells while 1 pM ACTH already induces a 25 and nine-fold increase, respectively, in corticosterone and aldosterone output in cultured cells. Moreover, some stimuli such as isoproterenol do not have any effect in isolated cells but do stimulate steroid secretion in cultured cells. In contrast, other stimuli, such as serotonin or DA (via DA2 receptors) act preferentially in freshly isolated cells. The main observation derived from this study is that glomerulosa cells, under appropriate conditions, are able to respond to their main secretagogues even after 4 days in culture. At this time, glomerulosa cells maintain their uitrastructural characteristics and functional properties and, aside from a few exceptions, demonstrate higher sensitivity to their known stimuli. Culture conditions used in the past 5 years have helped demonstrate the regulatory role of a number of peptides which probably act via paracrine or autocrine pathways. However, the effects of certain stimuli must be studied in freshly isolated conditions since their respective receptor binding sites are lost in culture. In conclusion, comparative studies between isolated and cultured conditions must be undertaken before pursuing studies on either the mechanisms of action or interactions between newly identified regulators of aldosterone secretion. Key words: Glomerulosa cells, culture conditions, aldosterone secretion.

specific receptors located on glomerulosa cell membranes and generate a number of different second messengers, as well as Ca 2+ influx. For example, AII, AVP and muscarinic receptors stimulate phosphoinositide breakdown, inducing the production of inositol trisphosphate [Ins(l,4,5)P3] and diacylglycerol, which in turn are responsible for intracellular calcium mobilization and protein kinase C (PKC) activation, respectively (for review, see [!, 2]). These effects are mediated via different G-proteins, some insensitive to both cholera toxin and pertussis toxin (PT) (coupling between hormonal receptors and phospholipase

INTRODUCTION secretion by zona glomerulosa of the adrenal cortex is under multifactorial regulation. Aside from its most important stimuli, namely adrenocorticotropin (ACTH), angiotensin II (All) and K ÷ ions, other factors are also known to stimulate secretion, such as vasopressin (AVP), serotonin (5-HT), acetylcholine, epinephrine, dopamine (via D A I receptors), endothelin, vasoactive intestinal peptide (VIP) and insulin. These regulatory molecules act via ALDOSTERONE

§ A u t h o r to w h o m c o r r e s p o n d e n c e should be a d d r e s s e d .

651

652

N. GALLO-PAVETet al.

C, PLC), others sensitive to pertussis toxin (coupling between hormonal receptors and calcium channels) [3, 4]. By contrast, ACTH, VIP, epinephrine, DA~ receptors of DA largely use cAMP as their second messenger via the well-known stimulation of Gs-proteins (for review, see [1]). Some, such as ACTH could stimulate calcium influx by a cAMP-dependent mechanism [5, 6]. On the other hand, DA (via DA~ receptors), somatostatin and atrial natriuretic factor (ANF) decrease aldosterone secretion induced by stimuli (for review see [2]). For example, DA 2 receptor decreases adenylate cyclase activity via a Gi-protein, since PT treatment abolishes these effects [7]. The inhibitory effect of DA on AII-induced PLC activation is mediated partly by an inhibition of Ca 2+ influx through Gi-protein partly by a direct effect on PLC activity [7]. As for ANF, its mechanism of inhibition is not yet clearly understood, but involves inhibition of Ca 2÷ influx [8, 9]. Ca 2+ influx is associated with cell membrane depolarization and activation of Ca 2÷ channels. Patch clamp studies have revealed the existence of at least three types of Ca 2÷ channels [8, 10, 11] and two types of K ÷ channels [11, 12] in glomerulosa cells. However, use of different animal models and/or different experimental conditions generate conflicting results either on the effects of both ACTH and AII on the activity of these channels or on their role in the excitation-secretion coupling [5, 13-15]. All of the studies related to zona glomerulosa function have been performed either with freshly isolated cells or with cells used after 2 or 3 days in culture. The first methodology was initially established by Haning et al. in 1970 [16] and modified thereafter by Tait and Tait in 1974 [17]. The second approach was developed by Hornsby et al. [18]. For the most part, results from these two methodologies yield similar conclusions; however, when results obtained with cultured cells reveal phenomenology not seen in freshly isolated cells, upholders of isolated cell methodology often qualify these new findings as aberrant or artefactual [19]. However, during the past year, many of the more fierce supporters of isolated cell methodo-

logy have turned to cultured systems [8, 9, 20, 21]. Both of these systems have advantages and disadvantages, e.g. important differences are noted in the dose-response of aldosterone to stimuli, which is often shifted to a lower concentration sensitivity in cultured cells. This is particularly the case when comparing the effects of ACTH [22] or AVP [23-26] in cultured vs isolated cells. Moreoever, some stimuli such as epinephrine and norepinephrine did not have any effect in isolated cells [27, 28]. This study compares the step-by-step response (binding, second messenger production and aldosterone response) of isolated glomerulosa cells vs cells maintained in primary culture to the main stimuli of aldosterone secretion, namely ACTH and All.

MATERIALS AND METHODS Chemicals

The chemicals used in the present study were obtained from the followingsources: aldosterone and corticosterone antiserum from ICN Biochemicals (Costa-Mesa, CA); [3H]aldosterone (72 Ci/mmoi); [3H]corticosterone (56 Ci/mmol) from New England Nuclear (Boston, MA); myo-[3H]inositol (10--20 Ci/mmol) and [3H]adenine (25 Ci/mmol) from Amersham (Oakville, Ontario); ATP, cAMP and DNAse from Sigma (St. Louis, MO, U.S.A.); arginine-vasopressin and Val-5-angiotensin II from Bachem (Marina Delpen, CA); ACTH 1-24 peptide (Cortrosyn) from Organon (Toronto, Canada); collagenase and culture media from GIBCO (Burlington, Ontario); Dowex l-X8 (100-200 mesh), chloride form, from Fluka (Mauppauge, NY). All other chemicals were of A-grade purity. Solutions

Before experiments were performed, both isolated and/or cultured cells were washed with Hanks' buffered saline (HBS; NaCI, 130 raM; KCI, 3.5 raM; CaC12, 1.8 mM; MgCI~, 0.5 raM; NaHCO 3, 2.5 raM; HEPES, 5mM, glucose, I g/l, pH 7.4). All experiments were performed in the same buffer supplemented with 0.5% BSA. Preparation o f glomerulosa cells

The zonae glomeruiosa were obtained from adrenal glands of female Long Evans rats weighing

Regulation of aldosterone secretion 200-25Og, and were isolated according to the method described in detail elsewhere [29]. The successive steps of zona glomerulosa isolation and cell dissociation were performed in MEM Eagle’s medium (supplemented with 100 U/ml penicillin and 100 pg/ml streptomycin). After a 20-min incubation at 37°C in collagenase (2 mg/ml, 4 glomerulosa/ml) and DNAse (25 pg/ml), the cells were disrupted by gentle aspiration with a sterile S-ml pipette, filtered and centrifuged for 10 min at 1OOg and suspended in the appropriate medium. For studies using freshly isolated cells, red blood cells and broken cells were removed by centrifugation at 1OOg over a 60% Percoll barrier (Pharmacia Canada Inc.), The cells were then suspended in MEM Eagle’s culture medium and preincubated for 30 min at 37°C in a humidified atmosphere (95% air: 5% CO,). After this resting period, cells were centrifuged at 100 g for 10 min, washed and resuspended in HBS buffer. Cell number ranged from 50 to 100 x 10’ cells/tube. For studies using primary cultures, cells were suspended in OPTI-MEM medium supplemented with 2% foetal calf serum (FCS), 100 U/ml penicillin and lOOp/ml streptomycin and plated in tissue culture dishes (35mm in diameter for CAMP determinations) or multi-well plates (1.6 cm for steroid measurements; Flow Laboratories, Mississauga, Ontario) at a density of approximately 7-10 x lo4 cells/multi-well or dish, respectively. The culture medium was changed every day and, when appropriate, cells were used at a density of approximately 1.5-2.5 x 10’ cells/dish or multi-well plates.

Incubations for steroid secretion measurements

Before each experiment, the medium of the cultured cells was aspirated and the cells washed twice with cold HBS buffer. The cells were incubated in 1 ml consisting of 0.9ml HBS-glucose supplemented with 0.5% BSA+O.l mg/ml bacitracin and 0.1 ml of stimuli. After a 2-h incubation at 37°C in 95% air: 5% CO2 atmosphere, the incubation medium was removed by aspiration and kept at -20°C until steroid measurements were performed. Isolated cell suspensions were also washed in HBS-glucose, centrifuged at 1OOg for 10 min and resuspended at a concentration of 5 x 104 cells/ml of HBS buffer and incubated for 2 h as described above. After incubation, the tubes were centrifuged at 800g and the supernatants stored at -20°C. Aldosterone and corticosterone were determined by radioimmunoassay directly from the medium, using specific aldosterone and corticosterone antisera and [‘Hlaldosterone and [3H]corticosterone as tracers.

653

Binding studies

12sI-Labelled [TyrZ’]ACTH-( l-39) with a specific activity of 2000 Ci/mmol was obtained from Amersham International (U.K). The analogue Sari-Tyr4-ValS-angiotensin II was iodinated by the iodogen method and separated as previously 25-cm ACTH [30] on a described for C-l 8 p-Bondapak column (Waters, Milford, MA) with a linear gradient of 20-60% acetonitrile in a buffer of 7% isopropanol-O.25 M ammonium acetate, pH 5.0, at a flow rate of 1 ml/min during 30 min. Carrier-free monoiodinated product was obtained in a homogeneous peak after 20 min with a specific activity of about 2000 Ci/mmol, determined by radioreceptor assay using crude liver membranes [31]. Binding assays were performed as previously described [30, 321. In short, cultured cells (I .5-2.0 x 10’ cells/Petri dish) were washed with 2 ml of HBS buffer and incubated 15 min at 37°C in this same medium. The hormone binding reaction was initiated by quick aspiration of the HBS medium and addition to each Petri dish of 0.8ml of HBS containing the labelled peptide to be tested [ x 30 pmol (80,000 d.p.m.) and 24 pmol (100,000 d.p.m.)] for [‘251]ACTH and [1251]AII, respectively. Isolated cells (1.5-2.0 x 10’ cells/assay) were washed by centrifugation, suspended for 15 min at 37°C in HBS medium. In this case, reaction was initiated by adding 400~1 of the cell suspension to propylene tubes containing 50 ~1 iodinated hormone and 50 ~1 buffer or unlabelled hormone or in the presence of 1 PM of unlabelled peptide for determination of nonspecific binding. The cells were incubated for 15 min at 37°C with labelled AI1 and 30 min at 20°C with labelled ACTH. In the case of AII, both cell suspensions and cultured cells (the latter were rapidly detached from the substratum by scraping with a rubber policeman), were filtered through Metriciel Gelman filters and rinsed three times. Cell-bound [‘*‘I]ACTH was separated from nonbound [‘251]ACTH by centrifugation of the cell suspension in Beckman microfuge tubes containing 200~1 PBS buffer plus 1% BSA. After centrifugation at 15,OOOg for I min, the pellet was rinsed without resuspension with 200 ~1 PBS plus 1% BSA and 10% sucrose. The tips of microfuge tubes were then sectioned and counted, as filters, in a Beckman gamma counter. Cyclic AMP determination

Intracellular cyclic AMP production was determined by measuring the conversion of [‘H]ATP into [3H]cyclic AMP, as described previously [22]. In short, cultured or isolated cells were incubated at 37°C in the same MEM Eagle’s culture media containing 2&i/ml [‘Hladenine. After 1 h, the cultures were washed while isolated cells were diluted

654

N. GALLO-PAYETet al.

and centrifuged. They were further incubated in HBS-glucose containing 1 mM isobutyl methylxanthine (IBMX) for 15 min at 37°C. Hormones or drugs were then added to the incubation medium for a further 15-min incubation period at 37°C. The reaction was ended by aspiration of the media (for cultures) or by direct addition of 1 ml ice-cold 5% trichloroacetic acid (in suspended cells). The cultured cells were scraped with a rubber policeman and 100#1 of cold 5mM (ATP-cyclic AMP) was added to the mixture. Cell membranes were pelleted at 5000 g 15 min and the supernatants were sequentially chromatographed on Dowex and alumina columns allowing for the separation of [3H]adenine nucleotide (primarily [3H]ATP) and [3H]cyclic AMP. Cyclic AMP formation was expressed as: % conversion = ([3H]cyclic AMP/[3H]cyclic AMP + [3H]ATP) x 100 per 15 rain.

Determination of inositol phosphate production The experiments were performed as described previously [32]. Briefly, the cells were labelled in culture medium containing myo-[3H]inositol (2/zCi/ml, 1.5ml/dish/3 days for cultured cells; 10#Ci/ml/tube/3h for isolated cells). After this period, the radioactive medium was discarded by centrifugation (for isolated cells) or by aspiration (for cultured cells), and the cells incubated in isotope-free MEM for 1 h. The cells were then washed and incubated for 15 min in HBS-glucose buffer supplemented with 10mM LiCI. After this equilibration period, the medium was changed again and cells were further incubated for 15 min at 37°C with fresh HBS-glucose/LiC1 medium and various concentrations of the hormone to be tested. The incubation ended with aspiration of the medium (for cultured cells) or centrifugation of the cells (for isolated cells) and the addition of 1 ml of 5% (v/v) HCIO 4 and 200#1 of bovine serum albumin (20mg/ml). Inositol phosphates were separated as described previously [32] by chromatography on Dowex IX8 columns. Radioactivity found in the total inositol phosphate fractions (IPs) were determined by scintillation counting in gel phase in a Beckman beta counter, with a counting efficiency of 18%. All results were corrected for quenching and were expressed in disintegrations per minute (d.p.m.). Background counts obtained with IPs solvent averaged 15 + 2 d.p.m.

Measurement of intracellular calcium concentration Intracellular calcium measurements were performed as described previously [6]. Briefly, cells were loaded 30 min at 37°C with 3.3/tM of fluorescent calcium chelator Fura 2 AM, in serum-free culture medium supplemented with HEPES (20 mM,

pH 7.4). Following the incubation, cells were washed three times in HBS buffer. The coverslips containing the loaded cells were placed in the perfusion chamber and mounted on an inverted Nikon microscope. Ceils were then incubated at room temperature in HBS buffer with or without the agents to be tested. The cells were alternatively excited with 340 and 380nm light and the emission was measured at 510nm using a Fluoroplex II system as previously described [6].

Morphological investigation The everted capsular adrenal glands were fixed in Bouin's fluid, embedded in paraffin, sectioned at 5/tm, and stained with haematoxylin and eosin. Isolated and cultured cells were observed and photographed in their Petri dishes containing culture medium, using an inverted phase-contrast Zeiss microscope. For electron microscopy, cells in suspension or Petri dishes were fixed with 2.5% glutaraldehyde, post-fixed with 2% osmium, dehydrated and embedded in Epon 812. Gold to silver-gray sections were then stained with uranyl acetate and lead citrate and examined in a Phillips 300 electron microscope.

Data analysis The data are presented as means + S.E.M. and are normalized to 2 x 105 cells. The EOf,0 is the concentration which produces the half-maximal response. Estimates of the EDs0 for agonists were obtained directly from the dose-response curves. Statistical analysis of the data was performed using the oneway analysis of variance (ANOVA) test.

RESULTS

Morphological characterization o f isolated and cultured glomerulosa cells Capsular tissue obtained after dissection did not contain any trace o f fasciculata tissue or medulla and histological slices were identical to those illustrated by H o r n s b y et al. [18]. The occasional chromaftin cell observed in culture did not result from medulla c o n t a m i n a t i o n but was contained in the zona glomerulosa as described previously [33]. Figure 1A shows the morphological appearance o f freshly isolated cells, grouped in an h o m o g e n e o u s population. Using Limbro culture dishes for plating and O P T I - m e d i u m , 80% o f cells adhered to the substratum in less than 1 h. In these culture

FIG. 1. Morphological appearance of glomerulosa cells. (a, b) Phase-contrast microscopy of freshly isolated glomerulosa cells, x 780 (a), or after 3 days in culture, × 450 (b), in OPTI-MEM medium + 10% FCS. (c) Fixed cultured cells after Toluidine Blue staining. (d-f) Electron micrographs of glomerulosa cells after 3 days of culture. Some cells exhibit numerous lipid droplets (LD), × 5426 (d), although others reveal few lipid droplets, x 10,260 (e). In both cases, there is an abundance of smooth endoplasmic reticulum (ER), and the presence of several lysosomal structures (L). Moreover, cells are characterized by their typical mitochondria (M) with tubular cristae, × 15,960 (f). The symbols: LD, ER, L and M are not present on panels d, e, f.

655

Regulation of aldosterone secretion conditions, overgrowth of fibroblasts was not encountered. Figure l b and c illustrate cells after 3 days in culture as seen by phase-contrast microscopy (Fig. lb) or histologically, following fixation and staining of cells (Fig. lc). Proliferating cells were observed soon after plating and exhibited polygonal morphology after 3 and 4 days. After 24 h of culture, many of the steroidogenic cells formed contact points and became firmly attached to the substratum. The adhering cells became flattened and the number of cells increased until confluence was reached after about 4 days. Cultured cells varied in size and were irregularly shaped with numerous long polypodia that formed the contact points with the substratum and neighbouring cells. In spite of the purity, we did observe cell-to-cell variation with respect to their histological appearance: some were rounded, others appeared flattened, some exhibited abundant refringent lipid droplets (Fig. ld), while others did not contain any lipid droplets (Fig. le). Ultrastructural studies revealed that after 3 and 4 days, glomerulosa cells still retained their typical mitochondria with tubular cristae, which is considered a good criterion of cell purity (Fig. ld-f). Response of isolated and cultured glomerulosa cells to ACTH Basal conditions. Freshly isolated cells had a higher steroid secretion output and higher cAMP production than cultured cells (Fig. 2). Twenty-four hours in culture induced a marked decrease in these parameters, with values falling from 8.8 + 0.4 ng/2 x 105 cells to 2.0-t-0.5 ng/2 x 105 cells (n = 3, P < 0.001) for corticosterone (77% decrease), from 241+_ 22 pg/2 x105 cells to 83___12pg/2 x105 cells (n = 3, P < 0.01) for aldosterone (65% decrease) and from 0.37+0.05 to 0.13 +0.02% conversion of ATP to cAMP (n = 3, P < 0.02; 65% decrease). These values did not change significantly between 2 and 4 days in culture. Response to ACTH. Cell sensitivity to ACTH was higher in cultured cells than in freshly

657

isolated cells. Corticosterone secretion was stimulated 17-fold in freshly isolated cells and increased to 158-fold on the third day and to 267-fold over basal levels after 5 days of culture (Fig. 2A). This increased ratio was partly due to a significant decrease in basal secretion levels, but also to an increase in the capacity of cells to synthesize corticosterone, since stimulated values increased from 148 ng/2 x 105 cells on the first day of culture to 454 ng/2 x l0 s cells on the fifth day. Aldosterone secretion was stimulated 6-fold in freshly isolated cells and this ratio increased to 20-fold between 2 and 4 days of culture (Fig. 2B). In the latter case, this enhancement was due to a decrease in the basal rate of secretion, since the stimulated aldosterone response remained the same between the first and third day, but decreased thereafter. In the same manner, cAMP response increased gradually from 1.6-fold in isolated cells to 10fold after 5 days of culture (Fig. 2C). Comparative dose-dependent response curves in freshly isolated cells and 3-day cultured cells. Since most of the experiments reported in the literature, as well as in our own laboratory, were performed either with freshly isolated cells or with 3-day cultured cells, cell sensitivity to ACTH was compared in these two populations. In both conditions, ACTH induced a dosedependent increase in steroid output (Fig. 3). In isolated cells, the lower effective dose was 0.1 nM (with a stimulated ratio of 3-fold over basal values for both corticosterone and aldosterone secretions) while maximal stimulation reached a plateau at 100 nM (with a stimulated ratio of 5 and 7 respectively; Fig. 3A, C). The half-maximal effective concentration (EDso) is similar for both corticosterone (0.09 + 0.012 nM, n = 6) and aldosterone (0.12 + 0.05 nM, n = 6) secretions. In cultured cells, the response to ACTH shifted to a lower concentration sensitivity (Fig. 3B, D). The lower effective concentration is less than 1 pM, since at that concentration, corticosterone and aldosterone secretions were already stimulated by 25- and 9-fold, respectively (from 2.0+0.9 ng/2 x 105 cells to 5 0 + 4 ng/2 x 105 cells

658

N. GALLO-PAYETet al.

60O o

%

50O

x

400

A

-yr

~g 300 c o

200

o o 0 0

'° I 5

0

1

IC

1

2

3

4

5

Cultured cells (days)

1800

B

1500

=o fX ¢N

1200 900

qD

tO o "D

600 300 0 IC

1

2

3

4

5

Cultured cells (days)

0.90

C

0.75 n rJ

! I
~e

0.60 0.45 0.30 0.15 -

0.00IC

I

2

3

4

5

Cultured cells (days)

FIG. 2. Glomerulosa cell response to ACTH in isolated (IC) and cultured cells. Glomerulosa cells were prepared as previously described in Materials and Methods. On each experimental day, cells were either incubated in Hank's buffer without (control conditions, ['7), or with 10nM ACTH ( 1 ) . Corticosterone responses (A), aldosterone secretion (B) and cAMP production (C) were determined as described in Materials and Methods. Results are the mean + S.E.M. of three distinct experiments, each done in triplicate. Where no error bars are shown, they are contained within the symbols. When placed over blank bar columns, P is compared to the control value in IC, when placed on hatched bar column, P is compared to its own control, *P < 0.02, **P < 0.001.

Regulation of aldosterone secretion 4oo

A

IC

.

*

7.0

350

8.0

30o

5.0 4.0

250

200 A ca

X i'd "--.

%" 1~2/1( I C

-11

I

I

I

I

-10

-9

-8

-7

ACTH (Log M)

6o0

o o

500

* •

/

B

vX

1.0

~, ,~p-'"

0.0 f--~//'¢'

C

I

I

I

I

I

-11

-10

-9

-8

-7

*

*

' ¢.. ~

e-

e

IC

2.0

50

0

C

3.0

150 100

=,o-

659

cc

*

o~ 'o "ID

ACTH (Log M)

1.4

D

CC

.

1.2 1.0 0.8

200

0.6

~

0.4 0.2

0

--i'~f//I C

t

-12 -11

I

i

I

I

-10

-9

-8

-7

ACTH (Log M)

0.0 ~'///( C

-12

I

I

I

I

I

-11

-10

-9

"8

-7

ACTH (Log M)

FIG. 3. Dose-response curves of ACTH-stimulated aldosterone secretion in isolated and 3-day cultured cells. Cells were processed as described in Fig. 1, and incubated either in Hank's buffer without (control), or with increasing amounts of ACTH. Corticosterone and aldosterone responses were measured either in isolated cells (IC, O, A, C) or in cultured cells (CC, 0 , B, D) as described in Materials and Methods. Results are the mean+S.E.M, of three distinct experiments, each done in triplicate. Where no error bars are shown, they are contained within the symbols. Arrows indicate the concentration of ACTH leading to half-maximal stimulation. *P < 0.001, compared to control.

for corticosterone and from 2 9 + 6 p g / 2 x 105 cells to 259 + 2 4 pg/2 x 105 cells for aldosterone secretion). Stimulation increased to 268- and 45-fold over basal values after reaching a plateau at 10nM concentration. The EOs0 of ACTH-stimulated effect in cultured cells is lower than that observed in isolated cells (0.066+0.011nM, n = 12 for corticosterone and 0.061_0.012nM, n = 12 for aldosterone secretions).

Binding of ACTH in glomerulosa cells. Since the higher sensitivity of glomerulosa cells to A C T H may be due to an increase in affinity or

in the number of A C T H receptors, A C T H binding studies were performed after each day of culture. The nonspecific binding (obtained by adding 1 #M unlabelled peptide) represented 30-35% of total binding. Increasing concentrations of 1-24 A C T H added to 13 pM iodinated 1-39 peptide caused a dose-dependent displacement of A C T H binding, with a half-maximal binding inhibition at about 0.5 nM. In cultures, affinity constants did not change significantly with time, with a dissociation constant around 0.3 nM (Table 1). On the other hand, binding capacity decreases in culture, with the number of sites falling from the initial 4100 to 1500 per cell after 4 days in culture.

660

N. GALLO-PAYETet al. IC soo

A

IC

.

so

C

*

"X"

~ ~ 0

*

~-

-If

-'X-

40 300

30

200

20

u

100

=o

o

x r---//

O4

C

I

-10

3so- B

I

-7

.

*

.

150 100

I

*

~ ~ ~

~

][ -

r --/./,

C

-I I

'~

r-'//

~ ~

C

~

o

cc

200-

0

I

-9 -8 All (log M)

3oo-

so

10

o

"" X

'-o

~

=o.

20

'~-

16

-

~2

8

;

I

-10

D

,

,

,

-I0

-9

-8

-7

All (log M)

0

I

cc

I

-7 *

.

*

S

** y

l

4

,

I

-9 -8 All (log M)

i-~/=

C

-I 1

,

,

,

,

-I 0

-9

-8

-7

All (log M)

FIG. 4. Dose--response curves of angiotensin II (AII)-stimulated aldosterone secretion and inositol phosphates accumulation in isolated and 3-day cultured cells. Cells were processed as described in Fig. 1, and incubated either in Hank's buffer without (control), or with increasing amounts of All. Aldosterone secretion and inositol phosphate accumulation were measured either in isolated ceils (IC, ©, A, C) or in cultured cells (CC, O, B, D) as described in Materials and Methods. Results are the mean + S.E.M. of three distinct experiments, each done in triplicate. Where no error bars are shown, they are contained within the symbol. Arrows indicate the concentration of All leading to half-maximal stimulation. *P < 0.02, **P < 0.001, compared to control.

TABLE1. BINDINGPARAMETERS FOR A C T H GLOMERULOSACELLS Day

KD (nM)

0 1 2 3 4

0.28 0.20 0.67 0.34 0.41

IN RAT

Bm~ (fmol/1 x 106 cells) No sites/cell 6.8 4.0 4.8 3.9 2.4

4100 2400 2900 2400 1450

Glomerulosa ceils were prepared as described in Materials and Methods and processed for binding studies immediately or after 1, 2, 3 and 4 days in culture. Results are the mean value of one experiment, done in triplicate.

Response of isolated and cultured glomerulosa cells to AH A I I induced a dose-dependent stimulation o f both aldosterone secretion and IPs production (Fig. 4). The maximal aldosterone stimulation ratio was obtained in cultured conditions (5fold vs 2-fold in isolated cells, Fig. 4A, B). In contrast, the IPs production was better stimulated in isolated cells (14-fold vs 5-fold in cultured cells, Fig. 4C, D). Whatever the studied parameters (IPs or secretion) or experimental conditions used (isolated or cultured cells), the EDs0 were about 0.5 n M in all cases. As was observed for A C T H , A l l binding

661

Regulation of aldosterone secretion

TABLE 2. BINDING PARAMETERS FOR ANGIOTENSIN II IN RAT ADRENAL GLOMERULOSA CELLS

Day 0 1 2 3 4 5

D

nmax

(nM)

(fmol/1 x 106 cells)

No sites/cell

0.66 + 0.07 1.00 + 0.23 0.92+0.06 0.86 + 0.03 0.82+0.04 0.36

283 ___43 133 -t-21"* 57+ 7* 63 + 2* 64+ 8* 29

170,589 + 30,850 79,675 + 13,770 34,438+7955* 38,244_ 5237* 38,616+ 10,009" 17,810

Glomerulosa cells were prepared as described in Materials and Methods and processed for binding studies immediately or after 1, 2, 3, 4 and 5 days in culture. Results are the mean + S.E.M. of three experiments, each done in duplicate. *P < 0.01, **P < 0.02, compared to day 0. capacity decreased significantly in culture with the number of sites falling from the initial 171,000 to 39,000 per cell after 4 days in culture (Table 2). The dissociation constant K D (1.00-t-0.05 nM, n = 3) remained similar whatever the cultured conditions used.

Intracellular calcium (Cai) responses in isolated and cultured cells. Isolated cells exhibit high levels o f basal [Ca2+]i and do not respond significantly to stimulation with AII (Table 3). In contrast, cultured cells have a lower basal level o f [CaZ+]i and respond to AII-stimulation with a 3-fold increase in [Ca2+]~. TABLE 3. INTRACELLULAR CALCIUM RESPONSES TO ANG1OTENSIN I I ( A l l ) IN ISOLATED AND CULTURED CELLS

Experimental conditions

[Ca2+]i (nM) Isolated cells Cultured cells

Control AII (100 nM)

283 +40 (3) 330+59 (3)

96+ 10 (6) 314+64 (6)*

Glomerulosa cells were prepared as described in Materials and Methods. After loading with Fura 2 AM for 30 min at 37°C, isolated cells were placed on glass coverslips pretreated with polylysine and were used immediately after adherence (30 min). Cells used after 3 days were cultured directly on the glass coverslips. Intracellular calcium ([Ca2+]i) was recorded as described in Materials and Methods. The number of experiments is indicated in parentheses. *P < 0.001, compared to control.

Response of isolated and cultured glomerulosa cells to vasopressin, isoproterenol and serotonin In these experiments, the stimulated responses of freshly isolated and cultured cells to AVP, isoproterenol (ISO) and 5-HT, three other well-known stimuli of aldosterone secretion, were compared. As described previously, freshly isolated cells had a high aldosterone output masking the hormonal effect of AVP and ISO, since no stimulation is observed in these experimental conditions, although a 2fold increase over control is observed after 3 days in culture (Fig. 5). On the other hand, 5H T had a significant stimulatory effect only in isolated cells. Both ISO and 5-HT use c A M P as second messengers. Figure 6 compares typical dose-response curves for second messenger production and aldosterone secretion in response to these stimuli. Figure 6A and B show that ISO stimulated aldosterone and c A M P production by 3- and 7-fold in cultured cells, although 5-HT elicited a 3-fold and a 7fold increase for both parameters in isolated cells (Fig. 6C, D). On the other hand, in cultured conditions, AVP stimulated aldosterone secretion by 3-fold and the IPs production by 5-fold (Fig. 6E, F). DISCUSSION The main observation derived from this study is that glomerulosa cells, under appro-

662

N. GALLO-PAYETet al.

700 © ~3

600

% x 04

500

,oo-

¢g e-

300 9(-

2

200

"O <

100"J C A V P I$o 5HT

Isolated c e l l s

I ill=

C AVP Iso § H T

Cultured cells

Flo. 5. Aldosterone response to vasopressin (AVP), isoproterenol (ISO) and serotonin (5-HT) in isolated and cultured glomerulosa cells. Glomerulosa cells were prepared as described in Materials and Methods. On each experimental day, cells were incubated in Hank's buffer either without (control conditions, rq), with 100nM AVP (ll), 100nM ISO (ll) or 100#M 5-HT (1). Aldosterone secretion was determined using specific radioimmunoassay. Results are the mean__.S.E.M. of three distinct experiments, each done in triplicate. Where no error bars are shown, they are contained within the symbol. *P < 0.02, **P < 0.001, compared to control.

priate conditions, can be maintained for at least 4 days in culture. At this time, glomerulosa cells maintain their ultrastructural characteristics and functional properties, and, aside from a few exceptions, demonstrate higher sensitivity to their known stimuli. However, due to a desensitization process or to an absence of specific growth factors which may regulate the genetic expression of some receptors, some substances (such as serotonin or dopaminergic D 2 agonist) act preferentially in isolated cells. Comparative basal steroid output in f r e s h l y isolated cells vs cultured cells

As suggested by De Lean et al. [27] and Pratt et al. [28] the high level of steroid secretion measured in freshly isolated cells may be due to an over-stimulation by the various stimuli which act in combination on aldosterone secretion in vivo. Another possible explanation could

involve an uncontrolled steroid release caused by cell membrane damage following tissue dissociation by collagenase treatment. Indeed, measurements of intracellular Ca 2+ with Fura 2 confirm this hypothesis, since in freshly isolated cells, the basal concentration of intracellular Ca 2+ is very high and decreases gradually when cells are used after a recovery period. Comparative responses to A C T H

Both corticosterone and aldosterone responses to A C T H are increased in cultured cell conditions. For aldosterone, this increased sensitivity to stimulation is due to a decrease in the basal rate of steroid output, since stimulated values do not vary significantly during the culture period. Moreover, these stimulated values are probably very near the maximal capacity of the cell, since no other known stimulus to aldosterone secretion has the potency of ACTH. In contrast, corticosteronestimulated values increase in culture, which suggests an enhanced capacity of the cells to synthesize the steroid. This may be explained by a parallel increase in cAMP-stimulated values. One can also raise the possibility that the high level of basal intracellular calcium in isolated cells is responsible for the high level of steroid output. Consequently, a decrease in the basal level of Ca 2+ decreases the basal rate of secretion and enhances cell sensitivity to stimuli. Our results can be compared to those obtained by Williams et al. [34] and Bird et al. [35] on fasciculata cells. In both cases, an enhanced stimulating effect of A C T H was observed in culture. However, in fasciculata cells, the authors observed a progressive increase in the basal rate of secretion. This difference may be either due to cell type (fasciculata vs glomerulosa) or more probably to culture conditions (Ham FI2 medium enriched with 10% foetal calf serum vs Minimum Essential Medium with 2% FCS). During the 4-day culture period, the affinity of A C T H for its receptor remains the same and is similar to that published by Penhoat et al. [36], Kemcke and Pond [37] or Li et al. [38], although the number of binding sites

Regulation of aldosterone secretion

663

cc 300 250 200 150 100 “0

50

1

X

O.O

+-

N

-7 -6 lsoproterenol (log M)

C

-9

-8

-5

OTj-

E

C

-9

-8

-7

lsoproterenol

-6

-5

(log Mb

P 1.8

0.0-I//, -8

C

,

,

,

,

O’,,-

-7

-6

-5

-4

C

-8

5-HT (log MI

E

300 250

* *

cc * *

-

200 -

-7

*

m0 c

F

30 -

x

-5

-4

(log Mb

cc

25 -

::

:: *

:: *

-9

-8

-7

+* :

*

150 -

15

lOO-

10

50 - /

-6

5-HT

5 20!r

I

I

Or-+-

OTjC

-11

-10

-9

AVP (log MI

-8

-7

C

-10 AVP

(log Ml

FIG. 6. Dose-Responses curves of isoproterenol (ISO), serotonin (S-HT) and vasopressin (AVP) in isolated (IC) or 3 days cultured cells (CC) on aldosterone secretion and second messengers accumulation. Cells were processed as described in the Fig. I, and incubated either in Hanks’ buffer either without (control), or with increasing amounts of IS0 (A, B), S-HT (C, D) and AVP (E, F). Aldosterone secretion and second messenger accumulation were measured either in cultured cells (CC, 0, A, B, E, F) or in isolated cells (IC, 0, C, D) as described in Materials and Methods. Results are the mean & S.E.M. of three distinct experiments, each done in triplicate. Where no error bars are shown, they are contained within the symbol. Arrows indicate the concentration of stimuli leading to half-maximal stimulation. *P < 0.02, **P < 0.01, ***P < 0.001, compared to control.

664

N. GALLO-PAYETet al.

per cell decreases by 64% in 4-day cultured cells compared to freshly isolated cells. From these results and those obtained by Penhoat et al. [36], we can assume that a culture medium which does not contain A C T H induces a decrease in binding capacity for this hormone. Moreover, the present results further support the concept that one or more second messengers - - other than cAMP - - are involved in ACTH-stimulated aldosterone secretion, since the half-maximal dose for stimulating hormone release is about 0.1 nM for both cultured and isolated cells (0.14 + 0.05 nM in isolated cells and 0.06-1-0.01 nM in cultured cells) and the halfmaximal dose of cAMP production is 1.0 + 0.1 nM (n = 7). In this context, it is important to recall the electrophysiological properties of glomerulosa cells, which possess both potassium and calcium channels. Durroux et al. [5, 10] have characterized, in culture conditions, at least three types of Ca 2+ channels, and have shown that A C T H affects these channels differently [5]. Moreover, we [6] and other groups [39, 40] have shown that A C T H increases inositol phosphate breakdown and intracellular Ca 2+, this latter effect being a cAMP-dependent process. Using freshly isolated cells, Quinn et al. [11] did not observe the L-type calcium channel, and assumed that this channel only appeared under cultured conditions [2]. We postulate that this Ca 2÷ channel could play a role in the long-term regulation of glomerulosa cell function. Comparative responses to A H and vasopressin

The binding characteristics of All observed in isolated cells is similar to that previously described by Douglas et al. [41] and further confirmed by many other laboratories. As with ACTH, cultured cells are more sensitive to AII than isolated cells. In this case as well, the observed response is not due to an increase in hormone binding or second messenger production, since both decrease in culture. This is not surprising if we assume that in vivo, AII receptors may be up-regulated by growth factors

present in the cell environment [42]. The high levels of basal IPs production and steroid secretion observed in isolated cells could be explained by a high level of intracellular Ca 2+, since we and others have shown that Ca 2+ alone can stimulate IPs production and thus enhance hormone secretion [25, 43]. These high basal values mask or make the cell insensitive to any further stimulation (i.e. cell desensitization). This explanation could explain differences observed between our results on the effect of AVP on glomerulosa cells [25, 26] and those of Balla et aL [23] and Enyedi et al. [24] using isolated cells. They reported AVP binding and IPs production without any significant stimulation of aldosterone secretion. As in the case of ACTH, the mechanism of All and AVP action are certainly more complex than previously thought. In fact, the majority of the studies relating to the different intracellular calcium pools are performed with cultured cells [44, 45]. In particular, cell culture conditions provide a good model for understanding the role of Ca 2+ and protein kinase C on AVP or ACTH actions [25, 26, 46, 47] or the role of G protein on hormone action [48]. Indeed, Woodcock et al. [49], Baukal et al. [50] and Hunyady et al. [51] now use cultured cells in order to study the finely tuned regulation of and/or interaction between AII and ACTH action. Comparative responses to isoproterenol, seroton& and dopam&e

Calcium does not explain all the discrepancies observed between the stimulated responses of isolated and cultured cells. In the case of isoproterenol, our results confirm those obtained by De Lean et al. [27] and Pratt et al. [28], who did not observe any stimulation in freshly isolated cells, but did observe a 2-fold increase in aldosterone secretion in cultured conditions. This refractoriness of isolated cells to stimulation by adrenergic agonists can be attributed to the presence of endogenous catecholamines whose concentrations are already optimal for stimulation of aldosterone secretion. This involvement of catecholamines is

Regulation of aldosterone secretion probably true for many other stimuli produced not only in the medulla (5-HT, AVP, DA, ANF), but also present in the cortex [52]. All these peptides display a paracrine effect on aldosterone secretion either directly from the medullary tissue present in zona glomerulosa or via systemic or sympathetic pathways [33, 5254]. The influence of these endogenous stimuli decrease in cells maintained in culture thus enabling the study o f their direct effect and mechanism of action without any interaction with other endogenous stimuli. The cultured cell model has permitted us to show that AVP, CRF, epinephrine, VIP and many other peptides can regulate aldosterone secretion. Nevertheless, if cultured cells represent a good model for studying the majority of peptides, it is important to compare both conditions, since, in some cases, receptors present in isolated cells disappear in cultured conditions. Therefore caution must be taken before affirming that cultured conditions are the best model. This is particularly the case for dopamine receptors of the DA2 subtype [7] and serotonin receptors which, in cultured conditions, do not modify either aldosterone secretion or cAMP production. The results obtained for aldosterone- and cAMP-stimulated production by 5-HT in isolated cells are similar to those obtained by Albano [55] and Leboulanger [53]. In these cases, caution must be taken in preparing cells using gentle enzyme digestion. Moreover, cells must be used only after a 2- or 3 h resting period, since artefactual responses often observed in freshly isolated cells are due to high levels of basal Cai. In conclusion, it appears that glomerulosa cells can be maintained for 4 days at least in primary culture without change in their ultrastructural characteristics or their capacity to produce corticosterone and aldosterone under various hormonal stimulations. Using MEM culture medium + 10% FCS or OPTI-MEM medium + 2% FCS, cells maintain aldosterone secretion without any necessity to add antioxidants during this period [56]. In many cases, these conditions appear to be the most valuable for studying intracellular processes involved in

665

the complex mechanisms of action of regulatory factors. Culture conditions used in the last 5 years have allowed demonstration of regulatory role of a number o f peptides which probably act via paracrine or autorine pathways to regulate steroid secretion in adrenocortical cells. Acknowledgements---This work has been supported by MRC grants to N. GALLO-PAYETand M.-D. PAWT and CNRS-INSERM grants to G. GUmLON. N.G.-P. is a recipient of a development grant from the Medical Research Council of Canada (MRCC).

REFERENCES 1. Bird I. M., Walker S. W. and Williams B. C. (1990) J. molec. Endocr. 5, 191-209. 2. Quinn J. J. and Williams G. H. (1988) A. Rev. Physiol. 50, 409-426. 3. Hausdorff W. P., Sekura R. D., Aguilera G. and Catt K. J. (1987) Endocrinology 120, 1668-1678. 4. Kojima I., Shibata H. and Ogata E. (1986) FEBS Lett. 204, 347-351. 5. Durroux T., Gallo-Payet N. and Payet M.-D. (1991) Endocrinology 129, 2139-2147. 6. Tremblay E., Payet M.-D. and Gallo-Payet N. (1991) Cell Calcium 12, 655-673. 7. Gallo-Payet N., Chouinard L., Balestre M.-N. and Guiiion G. (1991) Molec. cell. Endocr. 81, 11-23.

8. Barrett P. Q., Isales C. M., Bolag W. B. and McCarthy R. T. (1991) Am. J. Physiol. 261, F706-F719. 9. Barrett P. Q., Isales C. M., Bollag W. B. and McCarthy R. T. (1991) Can. J. Physiol. Pharmac. 69, 1553-1560. 10. Durroux T., Gallo-Payet N. and Payet M.-D. (1988) J. Physiol, Lond. 404, 713-729. I 1. Quinn S. J., Cornwall M. C. and Williams G. H. (1987) Endocrinology 120, 903-914. 12. Payet M.-D., Benabderrazik M. and Gailo-Payet N. (1987) Endocrinology 121, 875-882. 13. Cohen C. J., McCarthy R. T., Barrett P. Q. and Rasmussen H. (1988) Proc. natn. Acad. Sci. U.S.A. 85, 2412-2416. 14. Hescheler J., Rosenthal W., Hinsch K. D., Wulfern M. and Trautwein W. (1988) EMBO J. 7, 619-624. 15. Quinn S. J., Cornwall M. C. and Williams G. H. (1987) Endocrinology 120, 1581-1589. 16. Haning R., Tait S. A. S. and Tait S. F. (1970) Endocrinology 87, 1147-1167. 17. Tait J. F., Tait S. A. S., Gould R. P. and Mee M. S. R. (1974) Proc. R. Soc. Lond. B. 185, 375-407.

666

N. GALDPAYET et al.

18. Hornsby P. J., O’Hare M. J. and Neville A. M. ( 1974) Endocrinology 95, 124Q- 125 1. 19. Barrett P. Q., Bollag W. B., Isales C. M., McCarthy R. T. and Rasmussen H. (1989)

39. Farese R. V., Rosic N., Babischkin J., Farese M. G., Foster R. and Davis J. S. (1986) Biochem.

Endoc. Res. 10, 496518. 20. Brauneis U., Vassilev

biophys. Res. Commun. 135, 742-748. 40. Yamaguchi T., Naito Z., Stoner

P. M., Quinn S. J., Williams G. H. and Tillotson D. L. (1991) Am. J. Physiol. 260, E772-E779. 21. Quinn S. J., Brauneis U., Tillotson D. L., Cornwall M. C. and Williams G. H. (1992) Am. J. Physiol. 262, C598-C606. 22. Gallo-Payet N. and Payet M. D. (1989) J. Endocr. 120,409421. 23. Balla T., Enyedi P., Spat A. and Antoni F. A. (1985) Endocrinology 117, 421-423. 24. Enyedi P., Balla T., Antoni F. A. and Spat A. (1988) J. molec. Endocr. 1, 117-124.

25. Gallo-Payet N., Chouinard L., Balestre M.-N. and Guillon G. (1991) Endocrinology 129, 623634. 26. Guillon G., Balestre M.-N., Chouinard L. and Gallo-Payet N. (1990) Endocrinology 126,

1699-1708. 27. De Lean A., Racz K., NcNicoll N. and Desrosiers M.-L. (1984) Endocrinology 115, 485-492. 28. Pratt J. H., Turner D. A., McAteer J. A., and Henry D. P. (1985) Endocrinology 117, 1189-l 194. 29. Payet N.and Lehoux J.-G. (1982) J. Physiof., Paris 78, 3 17-32 1. 30. Gallo-Payet N. and Escher Endocrinology 117, 38-46.

31. Freychet P., N. (1972) J. 32. Guillon G., Gallo-Payet 33. Gallo-Payet

E.

(1985)

Kahn R., Roth J. and Neville Jr D. biol. Chem. 247, 3953.

Balestre M.-N., Chouinard L. and N. (1986) Biochem. J. 235, 209-214. N., Pothier P. and Isler H. (1987)

Biochem. cell. Biol. 65, 588-592. 34. Williams

B. C., Lightly E. R. T., Ross A. R., Bird I. M. and Walker S. W. (1989) J. Endocr.

121, 317-324. 35. Bird I. M., Clyme C. D., Lightly

Williams

B. C. and Walker

E. R. T., S. W. (1992) J.

Endocr. 133, 21-28. 36. Penhoat A., Jaillard C. and Saez J. M. (1989) Proc. natn. Acad. Sci. U.S.A. 86, 4978-4981. 37. Klemcke H. G. and Pond W. G. (1991) Endocrinology 12s, 24762488.

38. Li Z., Queen G. and Labella Endocrinology 126, 1327-1333.

F. S. (1990)

G. D., Mulrow P. J. (1990) Hypertension 16, 63564 1. 41. Douglas J., Aguilera G., Kondo T. and Catt K. (1978) Endocrinology 102, 685-696. 42. Penhoat A., Jaillard C., Crozat A. and Saez J. M. (1988) Eur. J. Biochem. 172, 247-254. 43. Woodcock E. A., Smith A. I. and Schmauk White L. B. (1988) Endocrinology 122, France-Saenz

R. and

1053-1059. 44. Ely J. A., Ambroz C., Baukal A. J., Christensen S. B., Balla T. and Catt K. J. (1991) J. biol.

Chem. 266, 18,635-18,641. 45. Byron K. L., Babnigg G. and Villereal M. L. (1992) J. biol. Chem. 267, 108-118. 46. Cozza E. N., Vila M. D. C., Acevedo-Duncan M., Farese R. V. and Gomez-Sanchez C. E. (1990) Endocrinology 126,2169-2176. 47. Kimura E. and Armelin A. (1990) J. biol Chem. 265, 3518-3521. 48. Begeot M., Langiois D., Spiegel A. M. and Saez J. M. (1991) Endocrinology 128, 3162-3168. 49. Woodcock E. A. (1989) Molec. cell. Endocr. 63, 247-253. 50. Baukal A. J., Hunyady L., Balla T., Ely J. A. and Catt K. J. (1990) Molec. Endocr. 4, 1712-1719. 51. Hunyady L., Merelli F., Baukal A. J., Balla R. and Catt K. J. (1991) J. biol. Chem. 266, 2783-2788. 52. Hinson J. P. (1990) J. Endocr. 124, 7-9. 53. Leboulenger F., Benyamina M., Delarue C.,

Netchitailo P., Saint-Pierre S. (1992) Bruin Res. 453, 103-109. 54. Vinson G. P., Pudney J. A. and (1985) J. Endocr. 105, 285-294. 55. Albano J. D. M., Brown B. L., S. A. S. and Tait J. F. (1974)

and Vaudry H. Whitehouse B. J. Ekins R. P., Tait Biochem. J. 142,

391400. 56. Hornsby P. J. and Crivello J. F. (1983) Molec. cell. Endocr. 30, l-20.