Benedict J. Canny$, Li-Guo Jia, and Denis A. LeongQ. From the Departments of Medicine, Neuroscience, and Physiology, University of Virginia Health Sciences ...
Vol. 267, No. 12, Issue of April 25, pp. 8325-8329. 1992 Printed in U.S. A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biologv, Inc.
Corticotropin-releasing Factor, butNot Arginine Vasopressin, Stimulates Concentration-dependent Increases in ACTH Secretion from a Single Corticotrope IMPLICATIONS FOR INTRACELLULAR SIGNALS INSTIMULUS-SECRETION
COUPLING*
(Received for publication, October 10, 1991)
Benedict J. Canny$, Li-Guo Jia, and Denis A. LeongQ From the Departments of Medicine, Neuroscience,and Physiology, University of Virginia Health Sciences Center, Charlottesville, Virginia 22908
The two fundamental parameters of corticotropin (ACTH) secretion are the number of secreting corticotropes and theamount of ACTH secreted by each cell. We have measured these parameters in rat corticotropes in response to increasing concentrationsof corticotropin-releasingfactor (CRF) orarginine vasopressin (AVP). Increasing concentrationsof AVP stimulated more corticotropes to secrete, while the amount of ACTH each cell secreted remained relatively fixed (nongraded secretory response). Conversely, increasing concentrations of CRF stimulated more ACTH secretion per cell (graded secretoryresponse), while the number of secretory cells remained relatively constant. When viewed from the perspectiveof a single corticotrope, it was clear that CRF and AVP induced completely distinct specific responses. We have previously shown, and provide further evidence here, that secretory responses to CRF or AVP occur in the same cell. It is therefore apparent that a single corticotrope is able to generate either a graded, or a nongraded secretory response. We have also considered the potential intracellular changes that must direct graded or nongraded secretion. It is generally accepted that CRF stimulates activation of adenylate cyclase, whereas AVP activates phosphoinositidase in pituitary corticotropes. Our findings, and others surveyed here, suggest that theactivation of adenylate cyclase results in graded secretion, while the activationof phosphoinositidase induces the nongraded secretion. Graded or nongraded secretion may therefore be linked to specific second messengers. It is hypothesized that the inositol 1,4,S-trisphosphate-mediatedrelease of an intracellular Ca2+ storeconstitutes amechanism whereby phosphoinositidase-coupled hormones set in motion the nongraded secretory response. These findings suggest novel functions for individualsecond messengers.
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Parts of this manuscript were presented in abstract form at the 73rd Annual Meeting of The Endocrine Society, Washington, DC, June 20-22, 1991. $ A C. J. Martin Fellowof the National Health and Medical Research Council of Australia. Present address: Baker Medical Research Institute, P. 0.Box 348, Prahran, Victoria 3181, Australia. S Supported in part by National Institutes of Health Grant DK35937 and National Science Foundation Center for Biological Timing Grant DIR-8920162. Corresponding author. Tel.: 804-982-0820; Fax: 804-982-0059.
The secretion of corticotropin (ACTH)’ from the anterior pituitary gland is regulated in a concentration-dependent fashion by the two hypothalamic hormones arginine vasopressin (AVP) and corticotropin-releasing factor (CRF) (1). It is generally accepted that theinitial post-receptor intracellular messengers involved in AVP- or CRF-induced ACTH release are distinct.AVP-induced ACTH secretion follows the activation of phosphoinositidase, which promotes the hydrolysis of phosphatidylinositol bisphosphate to inositol 1,4,5trisphosphate (InsP3) and diacylglycerol (2). InsP3 mobilizes an intracellular Ca2+store, which in turn appears to promote the influx of extracellular Ca2+(3), while diacylglycerol activates protein kinase C (4). Boththe mobilization of intracellular Ca2+ stores andthe activation of protein kinase C have been implicated in the regulation of AVP-induced ACTH secretion (5-8). CRF, on the other hand, activates adenylate cyclase to lead to theaccumulation of intracellular CAMPand the activation of CAMP-dependent protein kinase (9, 10). Total hormone secretion by a gland represents the arithmetical sum of the secretion from its individual cells, and therefore reflects the product of the number of cells secreting and how much each cell secretes. These two parameters of secretion are of fundamental regulatory significance. The recent development of the reverse hemolytic plaque assay (RHPA (11,12))makes single cell assays of secretion possible on a large population of cells. We have previously demonstrated that different corticotropes show different sensitivity to maximal concentrations of AVP or CRF (13). We have extended these findings and assessed the effects of increasing physiological concentrations AVP or CRF on single rat corticotropes, examining both the number of secretory cells, and the amount of ACTH secreted by each cell. These results were compared with those obtained from cells harvested from rats which had been adrenalectomized for 1 week (thus depriving the rats of glucocorticoids, which are the negative feedback regulators of corticotropes (14)). MATERIALSANDMETHODS
Animals-Female Sprague-Dawley rats (140-240 g, Charles River Corp., Wilmington, MA) that had been acclimatized for at least 7 days in a controlled environment room (lights on 0500-1900 h, 25 “C, 45% humidity) were used in all experiments. In the first experiment the effect of increasing concentrations of AVP or CRF on ACTH secretion in cells from control (sham-adrenalectomized) or adrenal-
’ The abbreviations used are: ACTH, corticotropin; AVP, arginine vasopressin; CRF, corticotropin-releasing factor; InsPe,inositol 1,4,5trisphosphate; RHPA,reverse hemolytic plaque assay; LHRH, luteinizing hormone-releasing factor; GHRF, growth hormone-releasing factor; PDBu, phorbol12,13-dibutyrate; 8-Br-cAMP,8-bromo-CAMP.
8325
8326
Intracellular Messengers Graded and
ectomized rats was compared in aRHPA. The sham-adrenalectomies were performed under methoxyflurane (Pitman-Moore, Mundelein, IL) anesthesia via the dorsal approach, and the adrenals were visualized but not removed. Adrenalectomy involved the same operative approach, but the adrenals wereremoved.Following operation all rats had free access to H 2 0 and 0.9% NaCl in H20. Seven days after the operation one sham-adrenalectomized and one adrenalectomized rat was sacrificed between 0900 and 1000 h, and anterior pituitary cells were harvested for the RHPA. Seven control (sham-adrenalectomized) and seven adrenalectomized rats were studied. Serum was collected for subsequent estimation of corticosterone levels to assess the efficacy of adrenalectomy. In thesecond experiment, the effect of pharmacological agents on ACTH secretion in the RHPA was examined; four normal (adrenal-intact) rats were used, again these animals were sacrificed between 0900 and 1000 h. RHPA-The RHPA was performed as previously described (13), but with a different anti-ACTH antiserum. Following sacrifice the anterior pituitary glands were removed, chopped finely, and treated with 0.1% trypsin (Difco, Detroit, MI)/O.l% bovine serum albumin (fraction V) in Ca2+-free Spinner's minimum essential medium (GIBCO) for 1 h a t 37 "C in a spinner flask, and final disruption of the tissue was achieved with a Pasteurpipette. A suspension of single cells in incubation medium (RPMI, GIBCO containing 0.1% bovine serum albumin and 2 pl/ml penicillin/streptomycin (10,000 units/ 10,000pg/ml, GIBCO)) was obtained and mixed with an equal volume of protein A-conjugated ovine red blood cells (20% in RPMI) togive a final concentration of 1 x lo6 pituitary cells/ml. The cells were plated on a poly-L-lysine-coated microscope slide on which a Cunningham chamber had been constructed. After allowing the cells to settle for 1 h, the cells were stimulated to secrete ACTH by the addition of various concentrations of ovine CRF and AVP in the presence of anti-ACTH antiserum (1:15) in incubation medium to which ascorbic acid (30 pg/ml) and CaC12 (40pg/ml) had been added. The Ca2+concentration of the medium was thus raised to 0.7 mM. Two h laterplaque formation was promoted by the addition of guinea pig complement (1:5, GIBCO) and the assay was completed by the addition of 2% glutaraldehyde. No plaques were seen if pituitary cells, conjugated red blood cells, or complement were omitted from the assay; preabsorbing antiserum with 1 pg/ml ACTH abolished plaque formation. There was no significant change in the number of ACTH plaques from basal levels if the cells wereexposed to luteinizing hormone-releasing hormone (LHRH, 100 nM), growth hormone-releasing factor (GHRF, 100 nM), thyrotropin-releasing hormone (10 nM), somatostatin (100 nM), or butyric acid (10 pM). Prior to measuring the number of plaque-forming cells, or their plaque areas, the identity of the slide was hidden to ensure no bias in the counting procedure. At least 1000 cells per slide were counted to gain an estimate of the percentage of plaque-forming cells as a proportion of the total anterior pituitary cell population, and the areas of all or100 plaque forming cells were measured using a BioQuant IV Image Analysis System (R & M Biometrics, Bethesda, MD). The plaque area is proportional to the amount of hormone secreted by the cell (12,15). Plaque areas vary from small to large in the population; the frequency distribution profile of plaque areas is unimodal, and does not differ from a normal distribution. The smallest plaque area measures 300 pm2,and thelargest plaque area ranged from 20-35,000 pm2,giving a dynamic range of the assay between 66 and 177 to 1. AVP and ovine CRF were obtained from Peninsula Laboratories, Inc. (Belmont, CA). Unless otherwise noted all other chemicals were obtained from Sigma. Corticosterone was estimated using a competitive protein-binding assay (16). All samples were measured in the same assay; the minimum detection limit of the assay was 0.5 pg/dl. Statistical Analysis-Concentration-response relationships in the RHPA were compared using repeated measures analysis of variance following logistic regression while controlling for individual rat effects. Other statistical comparison were made using the paired-sign rank test, or one-way analysis of variance with Dunnett's test to compare experimental and control groups. Differences with a p value of less than 0.05 were regarded as being significant. RESULTS
Control (Sham-adrenalectomized) Animals-Inthe presence of medium alone, 1.4 k 0.3% of the total anterior pituitary cell population formed ACTH plaques (Fig. lA). AVP (lo-" M) induced 3.1 f 0.5% of pituitary cells to form ACTH
or Nongraded Secretion CRF
I
P
8
1
. . zi 1 "
CRF
S
0
BASAL
12
10
8
-log[Agonist]
6
M
FIG. 1. The number of ACTH-secreting cells (A) and the amount of ACTH secreted per cell (mean plaque area, B ) in anterior pituitary cells from 1-weeksham-adrenalectomized rats (n = 7).The cells were stimulated with medium alone (basal, +) or a range of concentrations of either CRF (W) or AVP (0).
plaques, andthere was a significant positive relationship between increasing concentrations of AVP and thepercentage of ACTH plaque-forming cells ( p < 0.001,Fig. lA). Conversely, 5.2 f 0.8% of pituitary cells formed plaques in the presence of CRF (10"' M), but there was not a significant relationship between the concentration of CRF and percentage of ACTH plaque-forming cells (Fig. lA). The mean ACTH plaque area was 2360 & 210 pm2 in the presence of medium alone (Fig. 1B). There was a modest, however significant ( p < 0.01) positive relationship between the concentration of AVP and the mean plaque area (Fig. 1B). No concentration of AVP significantly changed the mean plaque area from that observed with medium alone. There was also a significant ( p < 0.001) relationship between CRF concentration and mean plaque area (Fig. 1B).The slope of the concentration-plaque area relationship was significantly ( p < 0.001) greater when the cells were stimulated with CRF rather than AVP (Fig. 1B). Taken together, AVP appears to regulate the number of cells secreting ACTH, having little effect on the amount of ACTH secreted by each cell, where CRF has predominant effects on the amount of ACTH secreted by each cell. The secretory response of a single corticotrope to AVP is nongraded, i.e. increasing concentrations had little effect on the amount of ACTH secreted, the secretory response to CRF is, in contrast, graded. Adrenalectomized Rats-Serum corticosterone levels were below the limit of sensitivity of the assay (
