0013-7227/92/1304-2067$03.00/0

Endocrinology Copyright 0 1992

by The Endocrine

SocieLy

Bradykinin Stimulates Aldosterone Release from Cultured Bovine Adrenocortical Cells through Bradykinin B2 Receptors* LORI J. ROSOLOWSKYt

AND

The University of Texas Southwestern Dallas, Texas 75235-9041

WILLIAM Medical

B. CAMPBELL Center, Department

ABSTRACT. The adrenal cortex contains a kallikrein-like enzyme that may lead to bradvkinin (BK) formation. This study was designed td determine whether BK acts on adrenocortical cells to stimulate steroid secretion. BK, Lys-BK, a specific BK 2 (B2) receptor agonist, and desArgg-BK, a specific BK 1 (Bl) receptor agonist, all stimulated aldosterone secretion from cultured bovine adrenal zona glomerulosa cells. BK and Lys-BK were equipotent (EC& = 2 x lo-’ M), whereas desAr$-BK was 1000-fold less potent. The maximal effects of BK and BK analogs were comparable to the maximal effects of adrenocorticotropin or angiotensin II. A B2, but not a Bl, receptor antagonist inhibited BK-stimulated aldosterone release. Verapamil and N,N-diethylamino-octyl-3,4,5-trimethoxybenzoate, which reduce intracellular calcium concentrations, reduced BK-stimu-

T

HE nonapeptide bradykinin (BK) is synthesized from kininogen, primarily by the proteolytic action of kallikreins, which are found in plasma and many tissues. Recently, a kallikrein-like kininogenase was identified in rat and canine adrenal glands (1). The enzyme is present in both the adrenal cortex and medulla. Thus, the adrenal gland has the capacity to synthesize BK. In addition, kininase II, which degrades BK, has been located on the membranes of vascular endothelial cells in both the adrenal capsule and medulla and on chromaffin cells of the medulla (2). The ability of bradykinin to stimulate catecholamine secretion from the adrenal medulla both in viuo and in vitro is well established (3-7). However, it is unknown whether BK acts similarly on adrenocortical cells to stimulate steroid secretion. Since the enzymes for the production and degradation of BK are present throughout the adrenal gland, it is possible that BK alters secretion from the cortex as Received September 25, 1991. Address all correspondence and requests for reprints to: Dr. William B. Campbell, Department of Pharmacology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75235-9041. *This work was supported by National Heart, Lung, and Blood Institutes Grant HL-21066. t Recipient of NIH Predoctoral Fellowship 5T32-GMO7062.

of

Pharmacology.

lated aldosteronr secretion. Although BK stimulated both prostacyclin and aldosterone production, indomethacin abolished prostacyclin production without affecting aldosterone secretion. In cultured adrenal fasciculata cells, high concentrations of BK stimulated cortisol release, but Bl or B2 receptor agonists were not effective. BK-stimulated cortisol secretion was reduced by NJ-diethylamino-octyl-3,4,5-trimethoxybenzoate but not by indomethacin. In summary, BK stimulates aldosterone release from cultured adrenal glomerulosa cells via high affinity B2 receptors. The effect is calcium-dependent and independent of prostaglandins. BK also increases cortisol release; however, this stimulation requires high concentrations of BK and may be mediated by an unknown receptor or by a receptor-independent mechanism. (&ndocrinolog.~ 130: 2067.-2075, 1992)

well as the medulla. Two subtypes of BK receptors have been characterized in vascular and other tissues. BK 2 (B2) receptors are the more prevalent and physiologically significant receptor subtype (8,9). They mediate endothelium-dependent relaxation and may also directly cause contraction of vascular smooth muscle. BK-stimulated catecholamine release from the adrenal medulla also occurs through B2 receptors (6). B2 receptors bind BK with high affinity (nanomolar range) and use inositol trisphosphate and calcium as intracellular second messengers. The B2 receptor is most responsive to BK and Lys-BK and is blocked by analogs in which the proline in position 7 of the peptide is replaced by a D-phenylalanine residue (10). The BK 1 (Bl) receptor is less abundant than the B2 receptor (8). It was first identified in rabbit aorta by Regoli and colleagues (11). This receptor subtype is selectively increased after some types of tissue injury (12), but its physiological significance in healthy tissue is unclear. Bl receptors are of low affinity and tend to use CAMP as a second messenger. The Bl receptor is activated by C-terminal desArg peptides, including desArg”‘,Lys-BK and desArg”-BK, and, in most tissues, inhibited by desArg”,Leu*-BK (9). The aims of the present study were to: 1) test the

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

2068

BRADYKININ

hypothesis that BK alters adrenal cultured bovine adrenocortical cells; effects of BK were mediated through and 3) examine the mechanism(s) of Materials

STIMULATES

steroid release in 2) determine if the Bl or B2 receptors; action of BK.

and Methods

Bovine adrenal zona glomerulosa cells were cultured as previously described by Rosolowsky and Campbell (13). Briefly, 8-12 glands (obtained from a local slaughter house) were sequentially dipped in 70% ethanol and Earle’s balanced salt solution (EBSS) containing 500 U/ml penicillin and 500 rg/ml streptomycin. Tissue preparation was performed at room temperature in a sterile cell culture hood. Adrenal glands were trimmed of fat and bisected. A Stadie-Riggs microtome (Thomas Scientific, Swedesboro, NJ) was used to cut sequential 500-r slices of tissue from the outer surface of the gland. The first slice, consisting of capsular tissue, was used to prepare zona glomerulosa cells. Adherent cells from the inner cortical zones (zona fasciculata-reticularis) were removed by scraping the inner surface. The second slice, consisting of cortical tissue, was used to prepare zona fasciculata-reticularis cells (see below). For preparation of zona glomerulosa cells, the tissue was minced finely and resuspended in digestion buffer of EBSS containing 25 mM HEPES, collagenase (1.8 mg/ml), hyaluronidase (0.75 mg/ml), dispase (1 mg/ml), fatty acid-free BSA (1 mg/ml), DNase (0.2 mg/ml), penicillin (500 U/ml), and streptomycin (500 pg/ml). Fragments were incubated for 30-45 min at 37 C and dispersed by repeated pipetting through a wide bore lo-ml plastic pipette. Fragments were allowed to settle, and the supernatant decanted through a 125-micron Nitex nylon filter (Tetko Co., Elmsford, NY). Isolated cells were collected by centrifugation at 150 x g for 3 min, washed once in EBSS, and resuspended in EBSS media containing horse serum (10% vol/vol) and antibiotics. Remaining fragments were resuspended in digestion buffer and incubated for an additional 45 min. Cells were dispersed and collected as described above. Cells from four or five digestion cycles were pooled and washed in EBSS containing 10% horse serum, 200 U/ml penicillin, and 200 I.cg/ml streptomycin. The final cell pellet was resuspended in growth medium consisting of modified Ham’s F-12 media supplemented with 14 mM NaCl and 14 mM NaHC03. Horse serum (10% vol/vol), butylated hydroxyanisole (50 PM), a-tocopherol (1.2 PM), sodium ascorbate (100 phi), sodium selenite (50 nM), glutathione (0.15 PM), insulin (20 nh4), transferrin (10 rg/ml), metyrapone (5 PM), penicillin (200 U/ml), streptomycin (200 rg/ml), nystatin (3 U/ml), and gentamicin (30 pg/ml) were also added to the growth medium. Cells were plated at a density of 4-5 x 10’ cells per well directly on Primaria 24-well culture dishes (Becton Dickenson, Lincoln Park, NJ). Cell viability at the time of plating was approximately 60% as measured by exclusion of trypan blue. Nonviable cells did not adhere and were removed when medium was replaced the following day. Cells were maintained at 37 C in a humidified atmosphere of 95% sir/5% COz. The medium was replaced every 24 h with growth medium in which the horse serum was omitted and fetal bovine serum (2%) was added. Cells were used upon reaching confluency, typically after 3-4 days in culture. Based on light microscopy, the purity of zona

ALDOSTERONE

RELEASE

Endo. 1992 Voll30. No 4

glomerulosa (ZG) cells was approximately 95%. Additionally, cortisol release from ZG cells under basal conditions was 0.3% of the amount of cortisol produced by zona fasciculata-reticularis cells. Bovine zona fasciculata-reticularis cells were cultured based on the method of Rainey et al. (14). Cortical slices (from two or three glands) were minced and incubated for 30-60 min at 37 C in digestion buffer consisting of trypsin (0.25%) and antibiotic-antimycotic solution (1%) (final concentrations of penicillin, 10 U/ml; streptomycin, 0.1 mg/ml; and Amphoteritin B, 0.25 pg/ml) in Dulbecco’s Modified Eagle’s Medium/ Ham’s F12 (DME/FlZ) buffer. The supernatant was discarded, and fresh digestion buffer was added for two more cycles. Subsequently, the slices were incubated in the trypsin solution for 20 min intervals until almost all of the tissue was dissolved (usually three to four cycles, with fresh trypsin solution being added for each cycle). After each cycle, tissue fragments were dispersed by repeated pipetting. Fragments were allowed to settle, and the supernatant was collected and filtered through cheesecloth. The cells were centrifuged and washed as described for preparation of glomerulosa cells. Cells were resuspended and plated in growth medium of DIME/F12 containing fetal bovine serum (5%), ITS+ premix (insulin-transferrin-selenium-BSA) (l%), and antibiotic-antimycotic solution (1%) in 24-well Corning (Corning, NY) or Falcon (Oxnard, CA) culture plates. Initial plating density was approximately 0.75-1.0 x lo6 cells per well. Fresh growth medium, with the concentration of fetal bovine serum reduced to 2%, was added to the cells every 24 h. Cells were used upon reaching confluency, typically 2-3 days in culture. Aldosterone production by ZFR cells under basal conditions or after stimulation with 3.5 x lo-’ M ACTH was undetectable. At confluence, each culture well was washed twice with 1 ml modified F-12 medium (glomerulosa) or DME/F-12 (fasciculata) containing 1 mg/ml BSA. The cells were incubated for 2 h in this medium. It was then replaced with 1 ml F-12 or DME/ F-12 containing 2 mg/ml BSA and 1.8 x low3 M calcium chloride, and test agents were added. The incubation was continued for 1 h at 37 C. As a positive control for the functional integrity of the cells, and to assessthe day-to-day variation in absolute rates of steroid production, ACTH was added routinely to one set of cells. All test agents were added in a vol of 10 ~1, and an equal vol of the vehicle was added to the control cells. BK and analogs of BK were dissolved in distilled water and stored at -20 C. ACTH (l-39) was dissolved in a buffer consisting of 5 X lo-’ M Tris-HCl (pH 5.5), 10-l M glycine, and 0.2% lysozyme and stored at -40 C. Test agents were diluted in incubation medium immediately before an experiment. In studies with inhibitors or receptor antagonists, the agent or its vehicle was added 10 min before addition of the stimulus, and the incubation was continued for an additional 60 min. After incubation, the medium was removed and stored frozen until assayed. All experiments were performed on two to five different cell preparations. The data represent pooled results from multiple incubations from different cell preparations or summarized results from a representative experiment that was performed on two or three cell preparations. Aldosterone was measured by direct RIA. In brief, the sample

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

BRADYKININ

STIMULATES

(typically 5-100 ~1)was incubated with PBS, pH 7.4, containing 0.1% sodium azide and 0.1% polyvinylpyrrolidone buffer (PBSP) to which 5000 cpm [“Hlaldosterone and sheep antialdoaterone antibody (1:450,000 final dilution) were added in a total vol of 0.3 ml. Unlabeled aldosterone standards for the standard curve (O-200 pg) were added in 10 ~1 ethanol and adjusted with appropriate volumes of incubation medium. All samples and standards were assayed in duplicate. After incubating overnight at 4 C, bound and free aldosterone were separated with dextran-coated charcoal [dextran T-70 (0.075%) and charcoal (0.75%] in PBSP buffer. The bound counts were measured by liquid scintillation spectrometry. The sensitivity of the assay was less than 5 pg/O.3 ml, and 50% displacement of [3H]aIdosterone occurred between 15-20 pg/O.3 ml aldosterone. The aldosterone antiserum cross-reacted 0.004% with cortisol, 0.03% with corticosterone, 0.1% with progesterone, and 0.15% with deoxycorticosterone. Cortisol was measured by a similar procedure with the following modifications. The cortisol antiserum was purchased from Endocrine Sciences (Tarzana, CA) and diluted according to their instructions. Unlabeled cortisol standards (o-4000 pg) were added in 0.1 ml PBSP buffer. Assay sensitivity was less than 15 pg/O.3 ml and 50% displacement of [3H]cortisol occurred at 100 pg/O.3 ml of cortisol. The antiserum cross-reacted 4.5% with desoxycortisol, 2.9% with corticosterone, less than 0.01% with aldoeterone, less than 0.5% with progesterone, and less than 0.02% with desoxycorticosterone. The production of 6-keto-PGF,,, the stable metabolite of prostacyclin, was measured by RIA as previously described (15). Cell protein was measured by the method of Lowry et al. (16), and the results were expressed as picograms of steroid per rg protein. Statistical analysis was performed by using a one way analysis of variance, followed by the Sidak multiple comparisons test when differences were found to be significant. P < 0.05 was considered statistically significant. The following reagents were used: a specially modified low sodium Ham’s F-12 medium and DME/F-12 medium (Sigma Chemical Co., St. Louis, MO); EBSS, horse serum, and antibiotic-antimycotic solutions (GIBCO, Grand Island, NY); fetal bovine serum (Hyclone, Logan, UT); collagenase-type I (Worthington Biochemical Corp., Freehold, NJ); dispase (Boehringer Mannheim Biochemicals, Indianapolis, IN); and ITS+ premix (Collaborative Research, Bedford, MA). BK and BK analogs were purchased from Peninsula Laboratories (Belmont, CA), Bachem, Inc. (Torrance, CA), or Sigma Chemical Co., and captopril was obtained from Squibb Institute (Princeton, NJ). Anticortisol serum was purchased from Endocrine Sciences (Tarzana, CA), and radiochemicals were obtained from New

England Nuclear (Boston, MA). N,N-diethylamino-octyl-3,4,5trimethoxybenzoate (TMB-8) was obtained from Aldrich Chemical Co. (Milwaukee, WI). All other reagents were purchased from Sigma Chemical Co. Results BK stimulated aldosterone release in a concentrationdependent manner. The average response to BK from five ZG cell preparations is shown in Fig. 1. A significant increase in aldosterone was observed with lo-” M BK

ALDOSTERONE

RELEASE

2069

(P < 0.006). The response to increasing concentrations of BK occurred in a stairstep manner. Aldosterone release increased over the range of 1O-“-1O-s M and then plateaued from 10-9-10-7 M. Aldosterone release increased further at higher concentrations (1O-“-1O-4 M). With 10e4 M BK, aldosterone production was approximately lo-fold over control. The sensitivity and EC& to BK was similar to the response to other steroidogenic agonists. For example, the sensitivity and ECso of ZG cells to angiotensin II (Ang II) was approximately lo-” M and 3 x 10-l’ M, respectively. Also, the sensitivity and ECso to ACTH was approximately 3 X lo-l2 M and lo-“’ M, respectively (13). In a given cell preparation, the sensitivity, maximal response, and absolute amount of aldosterone release to BK varied; however, this day-today variation also occurred with ACTH. In Figs. 1-3, the response to lo-* M BK varied but remained between 75100% of the response to 3.5 x 10e8 M ACTH. Although the cause of the variation in absolute aldosterone production was unknown, the variation in response to ACTH indicates that this phenomenon is not specific to BK. We tested whether the response to BK was mediated by Bl or B2 receptors using specific receptor agonists. BK and Lys-BK, a specific B2 receptor agonist, both stimulated aldosterone release with an EC!,, of 2 x lo-” M (Fig. 2). Lys-BK gave a greater maximal effect than BK. The Bl specific agonist desA$-BK also stimulated secretion, but it was 1000-fold less potent than BK or Lys-BK. Concentrations greater than lo-’ M were required for stimulation of aldosterone release. These results suggested that BK-induced aldosterone release is mainly mediated by high affinity B2 receptors. However, at high concentrations of BK and a Bl agonist, there was further stimulation of aldosterone release, which may be mediated by Bl receptors or an unknown mechanism. To determine if multiple BK receptors were involved, we tested the effects of Bl and B2 receptor antagonists on BK-stimulated aldosterone release. The B2 receptor antagonist D-Arg’,Hyp”,Thi”*‘,D-Phe7-BK (lo-’ M) inhibited aldosterone release that was stimulated by concentrations of BK below lo-’ M (P < 0.0001) but did not affect aldosterone release in the presence of higher concentrations of BK (Fig. 3, top). The Bl receptor antagonist, desArggLeuR-BK (lo-” M) enhanced basal aldosterone release by 25-fold and caused a parallel upward shift in the concentration response curve to BK (P < 0.02) (Fig. 3, bottom). The combination of BK and desArg’Leu*-BK appeared to be additive. At lower concentrations, desArg’Leus-BK

did not alter basal aldosterone

release but still enhanced BK-stimulated aldosterone release (data not shown). Taken together, the results with both antagonists and agonists suggest that BK-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

2070

BRADYKININ

STIMULATES

ALDOSTERONE

En&* :992

RELEASE Vol

130

l No

4

FIG. 1. Effect of BK on aldosterone release from cultured bovine adrenal glomerulosa cells. Cells were incubated with increasing concentrations of BK for 66 min. Incubation medium was removed and assayed for aldosterone by RIA as described in Mater&and Methods. Cells which were incubated with 3.5 x lo-’ M ACTH as a positive control released 4.5 + 0.5 pg aldosterone/pg protein +h. Results are the mean f SEM for lo-26 determinations from 5 cell preparations.

1 0

1

-12

-11

I

I

I

I

I

I

-10

-9

-8

-7

-6

-5

8

,

-7

-8

BRADYKININ

FIG. 2. Effect of BK, desA$-BK, a Bl receptor agonist and Lys-BK, a B2 receptor agonist, on aldosterone release. See legend to Fig. 1. Cells which were incubated with 3.5 X lOWaM ACTH released 0.91 f 0.14 pg aldosterone/rg protein. h. Values are the mean -orSEM for n = 5.

-4

(log Ml

0.25

0.00

L

1

I

0

-12

-11

stimulated aldosterone release is mediated by B2 receptors but not Bl receptors. Stimulation of aldosterone release by high concentrations of BK appears to occur via a previously undescribed low affinity BK receptor, another receptor previously unknown to bind BK, or a receptor-independent mechanism. We tested the possibility that BK’s actions were mediated by prostaglandins. BK (lo-’ M) stimulated the production of 6-keto PGF1, (the stable metabolite of prostacyclin) by 3.7-fold (P < 0.01) (Table 1). Inhibition of cyclooxygenase with indomethacin reduced basal 6keto PGF1, production and abolished BK-stimulated 6keto PGF1, production (P < 0.01). However, indometh-

1

-10 BRADYKININ

I

-9 AGONIST

I

-8 (log

-5

W)

acin did not alter basal or BK-stimulated aldosterone secretion. Similar results were obtained using 10m6M BK (data not shown). Thus, BK-stimulated prostacyclin production does not appear to mediate BK’s steroidogenic action in glomerulosa cells. In contrast to the effects of indomethacin, verapamil (10m4 M), which blocks L-type calcium channels in many cell types including adrenal glomerulosa cells (17-19) inhibited BK-stimulated aldosterone release by 80% (P < 0.001, compared to BK alone) (Fig. 4). The calcium antagonist TMB-8 (20) (4 x low4 M), completely abolished basal and BK-stimulated aldosterone release (P < 0.0001, compared to control). These findings suggest that BK’s actions are independ-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

BRADYKININ

STIMULATES

ALDOSTERONE

2071

RELEASE

4r

F I 5 % a

ha. 3. Effect of Bl and B2 receptor antagonieta on BK-stimulated aldosterone releaee. Cells were treated with the B2 receptor antagollist D-W, Hypa,ThiW,o-Phe’-BK (top; g, the Bl receptor antagonist deeArg%eu*-BK (bottom; l ), or vehicle @, 0) for 10 min. BK or vehicle wan added, and the incubation wee continued for 60 min. The medium wee removed and assayed for aldosterone. Celle which were incubated with 3.5 x lo+ Y ACTH produced 3.18 f 0.37 and 6.47 f 0.19 pg aldosksone/ pg protein.h for the upper and lower experiments, reepectively. Beeulta are themeanfSEMforn=6.

3

-B-

BK

+

BK + 82 ANTAQONIST

t

0

-12

-11

-10 -9 BRADYKININ

(log

-8

-7

BRADYKWIN

(log

-8 M)

-7

-6

-6

16

12

-

BK

-t

BK + Bl ANTAQONIST

t 9

6

0

ent of PGs and may be mediated by increasing intracellular calcium concentrations. To determine the specificity of BK’s action on adrenal steroidogenesis, we also measured the effect of BK on cortisol release in cultured adrenal fasciculata cells. BK stimulated cortisol release in a concentration-related manner (Fig. 5). However, unlike aldosterone release, stimulation only occurred at high concentrations of the peptide. BK stimulated cortisol release by g-fold and 33fold at concentrations of lo-’ M and lo4 M BK, respectively (P < 0.0001) (Fig. ‘5). For comparison, ACTH (3.5 x lo-la M) stimulated cortisol secretion by 172-fold.

-10

-9

-6

-6

-4

M)

Lower concentrations of BK were without effect. Treatment with captopril, which prevents inactivation of BK, did not increase the sensitivity to BK (data not shown). Thus, degradation of BK by kininase II does not appear to account for the low activity. The Bl receptor agonists de&$-BK and desArgl’-Lys-BK stimulated cortisol release by less than 2-fold, even at concentrations as high as 10” M (Fig. 5). Similar effects were seen with the B2 receptor agonists Lys-BK and Met-Lys-BK (maximum concentration tested was 10m6 M). The effects of high concentrations of BK do not appear to be a result of cell damage since uptake of ethidium bromide (an index of

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

BRADYKININ

2072

STIMULATES

ALDOSTERONE

TABLE 1. Effect of bradykinin and indomethacin on the release of aldosterone and 6-keto PGF1. by adrenal glomerulosa cells Treatment Indomethacin

Bradykinin

WM

Pm-

tain.h) + + +

+

0.13 0.22 0.71 0.64

f f f f

0.03 0.02 0.04 0.07”

6-Keto-PGF1. (pg/pg Protein.h) 0.13 0.08 0.48 0.08

f f f f

En& n 1997. Voll3O.Nc.4

/-

150

Response Aldosterone

RELEASE

-

BK

*

DeaArg’ BK

-

DesArg” Lys-BK

-

Lys-BK

-

Met-Lys-BK

125

0.02 0.02 0.06” 0.03

Cells were treated with indomethacin (lo+ M) or vehicle (0.1% ethanol) for 10 min. Bradykinin (lo+ M) or vehicle (HzO) was added, and the incubation was continued for an additional 60 min. The medium was removed and assayed for aldosterone and 6-keto-PGFI. by IUA as described in hfaterioLs Md M&o&. Results are the mean f SEM for n = 6. ‘P < 0.01 compared to vehicle alone. 2.sa

25 2.00

-

CONTROL

0

VERAPAMIL

m

TMB-8

c

s

0

0

g i .

-9 BRADYKININ

1.50

-8

-7

-6

AGONIST

(log

-6

-4

MI

5. Effect of BK, Bl-specific receptor agonists, and BP-specific receptor agonists on cortisol release from cultured bovine adrenal fasciculata-reticularis cells. Cells were incubated with BK (O), the Bl receptor agonists de&$-BK (A), or desArg”-Lys-BK (V), or the B2 receptor agonists Lys-BK (0) or Met-Lys-BK (0) for 60 min. Incubation medium was removed and assayed for co&sol by RIA. Cells treated with 3.5 X 10-l’ M ACTH produced 697 f 378 pg cortisol/pg protein. h. Values are the mean -CSEM for n = 6. FIG.

: p”

I P

-10

1 .oo

E !!i l 0.50

0.00 BASAL

BRADYKININ

FIG. 4. Effect of verapamil and TMB-8 on basal and BK-stimulated aldosterone release. Cells were treated with verapamil (lo-’ M), TMB8 (6 X lo-’ M), or vehicle for 10 min. BK (lo-’ M) or vehicle was added, and the incubation was continued for an additional 60 min. Values are the mean f SEM for six to eight determinations from two cell preparations.

cell death) and fluorescein diacetate (an index of cell viability) was similar in cells treated with BK and untreated cells (data not shown). These results suggest that BK does not act through the Bl or B2 receptor in zona fasciculata cells.

The Bl receptor antagonist desArg,Leu’-BK also stimulated cortisol secretion (Fig. 6). At 10m6 M desAr$,Leus-BK, cortisol release increased by 24- to 49fold over basal secretion, as compared to an 84-fold increase using lo-’ M BK. At concentrations below 10m6 M, desAr$,Leus-BK had no effect on basal secretion and did not block BK-stimulated secretion (data not shown). Indomethacin did not alter basal or BK-stimulated cortisol release (basal, 12.3 + 3 us. indomethacin, 6.2 f 0.8 pg; lo-’ M BK, 106.7 + 11.8, us. BK and indomethatin, 87.9 + 9.7 pg cortisol/pg protein. h, n = 7-9). Thus, prostaglandins do not appear to be involved in BKstimulated cortisol release. In contrast, TMB8 (5 X 10m4 M) reduced basal cortisol release by 25% (control, 1.9 f 0.1 us. TMB-8 1.4 + 0.1 pg cortisol/pg protein.h, P < 0.02). Additionally, TMB-8 reduced BK-stimulated cortisol release by 80% (10B4 M BK alone, 23 f 0.5 US. BK and TMB-8, 4.7 f 0.2 pg cortisol/rg protein. h, P c 0.0001). Thus, BK-stimulated cortisol release may be

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

BRADYKININ

STIMULATES

0 0

-9

-8 -7 BRADYKININ

-6 009 MI

-6

-4

FlG. 6. Effect of BK and a Bl receptor antagonist on cortisol release. Faaciculata celle were treated with the Bl receptor antagonist deeArgXe&BK (lo* M) (0) or vehicle (0) for 10 min. Increaeing concentratione of BK or vehicle were added, and the incubation was continued for 60 min. Values are the mean f SEM for n = 6. Ce& treated with 3.6 x 10-l* M ACTH produced 112 f 12 pg cortisol/pg protein. h.

dependent on calcium. The inability of specific BK receptor agonists to mimic the effect of BK suggests that the effect of BK on cortisol release may be mediated by a previously u&scribed, low affinity BK receptor, another receptor, or an unknown mechanism.

This study indicates that BK increases aldosterone and cortisol secretion from cultured bovine adrenocortical cells, but the mechanisms for the two responses appear to be distinct. Zona glomerulosa cells were much more sensitive than zona fasciculata cells to stimulation by BK. BK stimulated aldosterone release in a biphasic manner. The release of aldosterone by low concentrations of BK appears to be mediated .by B2 receptors, whereas at high concentrations of the peptide, aldosterone secretion appears to ‘occur via a nonreceptor-mediated mechanism. This conclusion is based on the following findings. First, when BK and Bl- and B2-specific receptor agonists were compared, the order of agonist

ALDOSTERONE

RELEASE

2073

potency was Lys-BK=BK>>desArg-BK. This order is characteristic of tissues which possess the B2 receptor (21). Second, the B2 receptor antagonist D-A$, Hyp3,Thi6$,D-Phe’-BK blocked aldosterone secretion which was stimulated by low, but not high, concentrations of BK. This result also supports the existence of a high affinity B2 receptor. Third, since the Bl antagonist desAr$,Leu’-BK enhanced basal and BK-stimulated aldosterone release, Bl receptors do not appear to account for stimulation of aldosterone secretion in the presence of high concentrations of BK or Bl agonists. Thus, stimulation of aldosterone release by high concentrations of BK appears to occur by an unknown mechanism, perhaps due to heterologous stimulation of another receptor by BK. Alternatively, this effect may be due to a previously undescribed low affinity BK receptor. In many cell types containing the B2 receptor, BK’s actions are associated with a rise in cytoplasmic calcium concentrations. For example, Dixon et al. (20) showed that BK increases the intracellular calcium concentration in cultured arterial smooth muscle cells. This increase is blocked by pretreatment with TMB-8. In the present study, TMB-8 also abolished BK-stimulated aldo&zone release. However, it should be mentioned that the specificity of TMBS has been disputed (22-24). Using adrenal glomerulosa cells, some reports indicate that TMB-8 inhibits calcium release (26, 26), whereas others show that it blocks calcium influx (22,27). It has further been suggested that the effects of the drug are unrelated to calcium metabolism (23). Therefore, we also tested the calcium channel blocker verapamil, which inhibits L-type channels in adrenal cells (17-19). Verapamil inhibited aldosterone release that was stimulated by BK. The results with both inhibitors suggest that BK may stimulate aldosterone release by increasing cytosolic calcium concentrations. Since indomethacin did not reduce BK-stimulated aldosterone release but abolished BK-stimulated production of 6-keto PGFI,, prostacyclin does not appear to mediate BK-stimulated aldosterone release. The response of fasciculata cells to BK was distinct from that of glomerulosa cells. Supraphysiological concentrations of BK were required to stimulate cortisol release. In contrast to their effects in glomerulosa cells, Bl and B2 receptor agonists did not stimulate cortisol release. The inability of captopril to increase the stimulation of fasciculata cells to BK supports the morphological data of Laliberte et al. (2) that kiiinase II (referred to as angiotensin converting enzyme in that report) is not found in any cell types within the zona fasciculata. At high concentrations, the effect of BK is similar in both glomerulosa and fasciculata cells. For example, TMB-8 reduced steroidogenesis in both cell types which was stimulated by high concentrations of BK, whereas

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

2074

BRADYKININ

STIMULATES

indomethacin was without effect. Also, the Bl antagonist desAr$,Leus-BK stimulated basal and BK-stimulated steroidogenesis in both cell types. Although the molecular mechanism(s) of these effects is not clear, similar observations have been made using high concentrations of BK in other cell types. For example, BK and other basic peptides, such as melittin, mastoparan, compound 48/80, substance P, and ACTH, stimulate exocytosis in mast cells (28-30). The EC& of these compounds for stimulation of histamine release in mast cells is in the range of 10-6-10-4 M. In contrast, nanomolar concentrations of BK, substance P, and ACTH are active in tissues which possess specific receptors for these ligands. In rat peritoneal mast cells, Bueb et al. (31) proposed that these compounds stimulate histamine release by direct activation of GTP-binding proteins. Additionally, Farmer et crl. (32) demonstrated that Arg’-D-Phe’-substituted BK analogs decrease BK- and vasopressin-induced contraction of uterine smooth muscle by a receptor-independent mechanism. These analogs do not compete for radiolabeled BK binding. This group speculated that these agents inhibit BK-stimulated contractions by blocking a membrane ion channel or possibly by altering membrane function. Using guinea pig trachea, Farmer et al. (33) provided evidence for a third type of BK receptor (B3). In that system, BK causes bronchoconstriction in uiuo and contractions in isolated, epithelium-denuded trachealis in vitro (5 X 1O-g-1O-5 M). These contractions are not inhibited by Bl or B2 antagonists. In the present study, the existence of a receptor which is distinct from Bl or B2 in fasciculata cells has not been ruled out, but it seems unlikely given the high concentrations of BK necessary to stimulate secretion. Thus, further studies will be required to determine the molecular target(s) of BK and desAr$-BK in fasciculata cells. It is unknown whether BK alters adrenal steroidogenesis under physiological conditions in viuo. The kinins have a plasma half-life of only 15 set (34, 35). These peptides are also degraded by kininase II during passage through the pulmonary vascular bed (36). Thus, if BK alters steroidogenesis in u1’uo, it probably acts as a paracrine hormone. This possibility is supported by the discovery that kallikrein, which synthesizes BK from kininogen, is present in the adrenal gland (1). The concentration of BK in the adrenal circulation has not been reported. However, using microdialysis probes, Hargreaves and Costello (37) demonstrated that immunoreactive BK rises from basal concentrations of 2 X lo-’ M in patients before surgery to 1.5 X lo-’ M after surgery. Using different model systems, others have reported values within this range (38). If concentrations of BK in the adrenal circulation are similar, then BK may alter aldosterone release in uiuo, although the peptide probably does not affect cortisol release.

ALDOSTERONE

RELEASE

EdO.1692 Vo113O.No4

The physiological significance of BK action in the adrenal cortex is also unclear. In general, the peripheral vascular action of systemically injected BK is depressor, whereas it exerts a pressor effect when introduced into the central nervous system. Although many investigators have supported the notion that Ang II and BK physiologically antagonize each other’s effects on blood pressure, there are many instances where the two peptides exert common actions. For example, both Arg II and BK stimulate uterine contraction in several species and also increase motility in the guinea pig ileum (10). Also, both Ang II and BK increase catecholamine secretion from the adrenal medulla (3-6) and, as shown in the present study, the two peptides increase aldosterone release in vitro. Although aldosterone inhibits the production of Ang II via suppression of renin release, it increases concentrations of renal kallikrein (39), which may increase concentrations of BK. These feedback loops suggest additional levels of regulation between these two peptides and aldosterone release. Since BK induces the release of multiple factors, including prostaglandins, endothelium-derived-relaxing-factor, catecholamines, and possibly steroid hormones, the net effect of BK on blood pressure in uiuo probably depends upon the balance of release of these factors. In light of the high concentrations of BK needed to see an effect on cortisol release in the present study, it appears unlikely that BK regulates cortisol release in vim. In summary, we have shown that BK stimulates aldosterone and cortisol release from cultured adrenal glomerulosa and fasciculata cells, respectively. BK stimulates aldosterone release via high affinity B2 receptors, and the effect is comparable to maximal concentrations of ACTH or Ang II. In contrast, BK stimulates cortisol release by an unknown mechanism. To our knowledge, this study is the first demonstration of an effect of BK on adrenal steroidogenesis. Thus, in addition to stimulating adrenal catecholamine release, BK stimulates adrenal steroidogenesis as well. Acknowledgments The authors thank Mrs. Martha Williams, Mrs. Bandi Fishbeck, and Dr. Bandy Bomig for their technical assistance. We are grateful to Dr. William Bainey of the Department of Obstetrics and Gynecology for helpful advice and assistance with the cell culture. We also thank Mrs. Juanita Coley for her secretarial assistance. The antialdosterone serum was generously provided by the Pituitary Hormone Distribution Program of the NIH.

References 1. Scicli G, Nolly H, Carretero OA, Scicli AG 1991 Glandular kallikrein-like enzyme in adrenal glands. Adv Exp Med Biol 247: 217-222 2. Lliberte F, Laliberte M-F, Alhenc-Gelas F, Chevillard C 1987 Cellular and subcellular immunohistochemical localization of an-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.

BRADYKININ

3. 4. 5. 6. 7.

8. 9.

10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

STIMULATES

aiotensin-convertina ensyme in the rat adrenal gland. Lab Invest S&364-371 Feldberg W, Lewis GP 1964 The action of peptides in the adrenal medulla. Release of adrenaline by bradykinin and angiotensin. J Physiol 171:98-108 Robinson RL 1967 Stimulation of the catecholamine output of the isolated. nerfused adrenal gland of the dog by anaiotensin and bradyki& J Pharmacol E& Ther 156:252~257 Bommer M, Hen A 1989 Neuropeptides and other secretagogues in bovine chromafFm cells: their effect on opioid peptide metabolism. Neuropeptides 18243-251 O’Sullivan AJ, Burgoyne RD 1989 A comparison of bradykinin, angiotensin II and muscarinic stimulation of cultured bovine chromaffii cells. Biosci Rep 9243-252 Gwen PJ, Marriott DB, Boarder MR 1989 Evidence for dihydropyridine-sensitive and conotoxin-insensitive release of noradrenaline and uptake of calcium in adrenal chromaftin cells. Br J Pharmacol97~33-138 Regoli D, Barabe J 1980 Pharmacology of bradykinin and related kinins. Pharmacol Rev 321-36 Gavras H, Gavras I 1991 Effects of bradykinin antagonists on the cardiovascular svstem. In: Burch RM (ed) Bradvkinin Antagonists: Basic and Clinical Research. Marcel ‘Dekker, New York, pp 171189 Vavrek RJ, Stewart JM 1985 Competitive antagonists of bradykinin. Peptides 6161-164 Regoli b, Bara& J, Park WK 1977 Receptors for bradykinin in rabbit aortae. Can J Phvsiol Pharmacol55855-867 deBlois D, Bouthillier J-, Marceau F 1989 Pharmacological modulation of the up-regulated responses to de&$-bradykinin in &JO and in vitro. Immunopharmacology 17~187-198 Rosolowsky LJ, Campbell WB 1990 Endothelin enhances adrenocorticotropin-stimulated aldosterone release from cultured bovine adrenal cells. Endocrinology 126~1860-1866 Ralney WE, Vii I, Mason JI, Cachet C, Chambax EM, Saez JM 1988 Effects of transforming growth factor beta on ovine adrenocortical cells Mol Cell Endocrinol60:189-198 Campbell WB, Gomez-Sanchez CE 1985 Absence of prostacyclin involvement in angiotensin-induced aldosterone secretion in rat adrenal cells. Endocrinology 114279-286 Lowry OH, Rosebrough NJ, Fsrr AL, Randall RJ 1951 Protein measurement with the Folin phenol reagent. J Biol Chem 193: 65-275 Fakunding JL, Catt KJ 1980 Dependence of aldosterone atimulation in adrenal glomerulosa cells on calcium uptake: effects of lanthanum and verapamil. Endocrinology 107~1345-1353 Schiffrin EL, Lis M, Gutkowska J, Genest J 1981 Role of Ca*+ in response of adrenal glomerulosa cells to angiotensin II, ACTH, K+, and ouabain. Am J Physiol241:E42-E46 Foster R, Rasmussen H 1983 Angiotensin-mediated calcium efflux from adrenal glomerulosa cells. Am J Physiol245zE281-E287 Dixon BS, Breckon R, Fortune J, Vavrek RJ, Stewart JM, ManecCalvert II, Linas SL 1990 Effects of kinins on cultured arterial smooth muscle. Am J Physiol258:C299-C308 Regoli D, Rhaleb N-E, Dion S, Drapeau G 1990 New selective bradykinin receptor antagonists and bradykinin Bz receptor character-h&ion. Trends Pharmacol Sci 11:156-161 Kojima I, Kojima K, Rasmussen H 1985 Mechanism of inhibitory

ALDOSTERONE

23. 24.

25.

26. 27.

28.

29. 30.

31. 32.

33. 34. 35. 36. 37. 38. 39.

RELEASE

2076

action of TMB-8 [8-NN-diethylamino)octyl-3,4,5-trimethoxybenxoate] on aldosterone secretion in adrenal glomerulosa cells. Biochem J 23287-92 Rossier MF, Krause K-H, Lew PD, Capponi AM, Vallotton MB 1987 Control of cytosolic free calcium by intracelhdar organelles in bovine adrenal glomerulosa cells. J Biol Chem 262:4053-4058 Lihrmann I, Delarue C, Homo-Delarche F, Feuilloley M, B&nger A, Vaudry H 1987 Effects of TMB8 and dantrolene on ACTHand angiotensin-induced steroidogenesis by frog in&renal gland evidence for a role of intracelhdar calcium in angiotensin action. Cell Calcium 8269-282 Garcia R, Laychock SG, Rubin RP 1982 Inhibition of dibutyryl cyclic AMP induced steroidogenesie in rat adrenocortical cells by the putative calcium antagonist TMB-8. J Steroid Biochem 16: 317322 Kramer RE 1988 Angiotensin II-stimulated changes in calcium metabolism in cultured glomerulosa cells. Mol Cell Endocrinol60: 199-210 Kojima I, Shibata H, Ogata E 1986 Action of TMB-8 (I-(NJVdiethyhunino)octyl-3,4,5-trimethoxybenxoate) on cytoplasmic fres calcium in adrenal glomerulosa cells. Biochim Biophys Acta 888: 25-29 Lawrence ID, Werner JA, Cohan VL, Licktenstein LM, KageySobotka A. Vavrek RJ. Stewart JM. Proud D 1989 Induction of histamine release fromhuman skin mast cells by bradykinin analogs. Biochem Pharmacol38:277-233 Devilher P, Renoux M, Drapeau G, Regoli D 1988 Hiitamine release from rat peritoneal mast cells by kinin antagonists. Eur J Pharmacol149:137-140 Mousli M, Bueb J-L, Bronner C, Rouot B, Landry Y 1990 G protein activation: a receptor-independent mode of action for cationic amphiphilic neuropeptides and venom peptides. Trends Pharmacol Sci 11:358-362 Bueb JL, Mousli M, Landry Y, Bronner C 1990 A pertussis totisensitive G protein is required to induce histamine release from rat peritoneal mast cells by bradykinii. Agents Actions 30~98-100 Farmer SC, Burch RM, Dehaas CJ, Togo J, Steranka LR 1989 [Argi-D-Phe’]-substituted bradykinin analogs inhibit bradykininand vasopressin-induced contractions of uterine smooth muscle. J Pharmacol Exp Ther 2483677-681 Farmer SG, Burch RM, Meeker SA, Wiis DE 1989 Evidence for a pulmonary Bs bradykinii receptor. Mol Pharmacol36:1-8 Erdos EG 1979 Kininases. In: Erdos EG (ed) Bradykinin, Kallidin and Kallikrein. Handbook of Experimental Pharmacology. Springer-Verlag, Heidelberg, ~0125427-487 Ryan JW 1982 Processing of endogenous polypeptides by the lungs. Annu Rev Physiol44:241-255 Ferreira SH, Vane JR 1967 Half-lives of peptides and amines in the circulation. Nature 216:1237-1240 Hargreaves KM, Costello A 1990 Glucocorticoids suppress levels of immunoreactive bradykinin in inflamed tissue as evaluated by microdialysis. Clin Pharmacol Ther 48:168-178 Hulthen UL, Lscerof H, Hiikfelt B 1977 Renal venous output of kinins in patients with hypertension and unilateral renal -artery stenosis. Acta Med Stand 202189-192 Margolius HS 1989 Tissue kallikreins and kinim regulation and roles in hypertensive and diabetic diseases. Annu Rev Pharmacol Toxic01 29343-364

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 14:23 For personal use only. No other uses without permission. . All rights reserved.