Calcium release from pituitary secretory granules: modulation by thiols, disulfides, and dihydropyridine calcium channel blockers.

0013-7227/90/1265-2671$02.00/0 Endocrinology Copyright © 1990 by The Endocrine Society

Vol. 126, No. 5 Printed in U.S.A.

Calcium Release from Pituitary Secretory Granules: Modulation by Thiols, Disulfides, and Dihydropyridine Calcium Channel Blockers* MARY Y. LORENSON, MARY L. CUCCARO, AND LAURENCE S. JACOBS Endocrine-Metabolism Unit, Department of Medicine, and the Clinical Research Center, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642

ABSTRACT. The distribution of calcium in isolated bovine pituitary secretory granules was studied by atomic absorption. The total granule calcium (in 26 preparations) averaged 14.5 nmol/mg protein, or 21.2 ± 1.6% of the total pituitary homogenate calcium. Incubation of granules with KC1 resulted in calcium release (78% at 15 mM and 100% at 50 mM, for example). Calcium release was also pH dependent, with greater release at acidic pH values; it was not influenced by either 500 nM strontium or 500 nM lanthanum. Release was augmented bv reduced glutathione (GSH), with significant release observable at thiol levels as low as 10 fiM. In addition to GSH, cysteine also stimulated release; mercaptoethanol and dithiothreitol were less potent. Interestingly, the disulfides cystine and oxidized glutathione also stimulated calcium release. Since the latter compounds are known to inhibit hormone release from granules, calcium and protein release appear to be regulated independently. A number of dihydropyridines were tested as potential

blockers of calcium release from granules. Nimodipine inhibited basal calcium release at high concentrations and potently inhibited GSH-stimulated calcium release, with an apparent Kj in the 10-20 nM range; it also inhibited K+-stimulated release but to a lesser extent. Nimodipine, however, did not significantly influence protein or hormone release. GSH-stimulated calcium release was also inhibited by nifedipine, and this inhibition was qualitatively and quantitatively similar to that by nimodipine. Nisoldipine and nitrendipine, however, displayed no significant inhibition. In summary, it appears that the release of secretory granule calcium in vitro is independent of protein release. Thiols and some disulfides stimulate calcium release, and its inhibition by dihydropyridines suggests that granule membranes may contain specific ion channels. The role of granule calcium in the cell remains to be defined. (Endocrinology 126: 2671-2678, 1990)

C

HANGES in the cytosolic concentration of free calcium constitute a critical response to a variety of extracellular signals (reviewed in Refs. 1-4). The free calcium is maintained basally in the range of 100-200 nM by the combined action of plasma membrane calcium pumps and exchangers expelling the ion from cells and by energy-dependent calcium compartmentalization into intracellular organelles. Upon stimulation of contractile and secretory cells, the free calcium concentration rapidly rises, sometimes to 1000 nM or above. While sustained cell responses are dependent upon a source of extracellular calcium, early responses to signalling and stimulation may be independent of an extracellular source. In such cells, calcium is mobilized from intracellular locales, triggered by hydrolysis of phosphatidyl inositol-4,5-bisphosphate and production of inositol1,4,5-trisphosphate [Ins(l,4,5)P3] (5-8). Of the potential Received October 23, 1989. Address all correspondence and requests for reprints to: Mary Y. Lorenson, Ph.D., Box Med/CRC, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, New York 14642. * This work was supported by NIH Grants DK-31326 and RR00044.

intracellular sources of calcium, mitochondria have a high capacity and low affinity for calcium and only accumulate calcium at high free ion concentrations (>1 JUM) (1, 9). Although they are not viewed as important in the initial burst of calcium, they may be involved in reaccumulation from other stores (10). The endoplasmic reticulum, or a subfraction thereof, has been the favored intracellular store capable of rapid calcium release into the cytoplasm (11-13). Recently, however, Volpe and coworkers (14, 15) have provided experimental evidence from HL60 and PC12 cells and rat liver and pancreas that argues against this possibility. They propose that calcium mobilization is from calciosomes, small vesicles (50-250 nm in diameter) which demonstrate both membrane Ca2+-ATPase activity and high affinity calciumbinding capacity. The latter is attributed to a protein within the lumen that is similar or identical to sarcoplasmic reticulum calsequestrin (16). In addition, Rossier et al. (17) found in adrenal cortical cells that the distribution of Ins(l,4,5)P3-binding sites and Ins(l,4,5)P3-sensitive 45Ca2+ uptake were greater in fractions not containing endoplasmic reticulum markers.

2671

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 11 July 2015. at 07:17 For personal use only. No other uses without permission. . All rights reserved.

2672

CALCIUM CHANNELS IN SECRETORY GRANULES

Another potential source of intracellular calcium is the secretory granule. Our previous studies on isolated pituitary secretory granules have indicated the presence within their intragranular acidic milieu of calcium and other divalent cations, calmodulin and calmodulin-binding proteins, and reduced glutathione (GSH) (18-20). We have also characterized a secretory granule membrane ATPase, which appears, however, to be calcium independent (21). The interrelationships between calcium and other granule components are currently unclear. It is possible that calcium is present in granules serendipitously, being trapped from the extracellular space during the continuous membrane recycling from endocytic vesicles to other organelles. On the other hand, intragranular calcium may function directly or indirectly in hormone multimeric storage structures, play a role in the secretory mechanism, or have some other function, not presently known, in granule biology. The present studies quantitate more precisely the amount of calcium stored in granules, show that it can be released in vitro, and indicate that release is sensitive to the dihydropyridines nimodipine (ND) and nifedipine. These data raise the possibility that secretory granule calcium might be a potential source of intracellular calcium for stimulussecretion coupling. Materials and Methods Materials Divalent cations (chloride salts), GSH, EDTA, sodium dodecyl sulfate, sucrose, and buffers were purchased from Sigma Chemical Co. (St. Louis, MO), calcium reference standard was from VWR Scientific (Rochester, NY), and radiolabeled compounds were purchased from Amersham (Arlington Heights, IL). ND, [l,4-dihydro-2,6-dimethyl-4-(m-nitrophenyl)pyridine-3,5-dicarboxylic acid 3-isopropyl, 5-(2-methoxyethyl) ester], other dihydropyridines, and Bay K-8644 were gifts from Patricia Hinkle, Ph.D., and Robert Kass, Ph.D. (Rochester, NY); Bay K-8644 was also purchased from Behring Diagnostics (Sommerville, NJ). Purified bovine (b) GH (B-17) and PRL (B-5) as well as rabbit anti-bGH serum were gifts from the National Pituitary Agency, while rabbit antiovine PRL was raised in our laboratory and has been described previously (22). Precipitating second antibody (goat antirabbit 7-globulin) was purchased from Antibodies, Inc. (Davis, CA). Methods Large pituitary granules, rich in GH and PRL, were prepared by differential and sucrose density gradient centrifugation, as previously described (21). Preparations were routinely more than 95% pure, as estimated by measurement of marker enzymes and by morphological absence of mitochondria, plasma membranes, and other cellular membranes (21). For calcium determinations, all reagents used were of the highest available purity, all glassware and plasticware was

Endo • 1990 Vol 126 • No 5

washed in dilute acid and thoroughly rinsed in HPLC grade water before use, and all reagents and buffers were pretested for calcium contamination. In addition, no interference with calcium measurements was noted with any of the reagents or buffers used. All thiols and dihydropyridines were freshly prepared; the latter were solubilized in dimethylsulfoxide, and incubations were carried out in the dark. The concentration of dimethylsulfoxide in granule incubations was maintained below 2.5%, a level that by itself did not influence calcium or protein release. Routinely, granules at 5 mg protein/ml were incubated in microcentrifuge tubes at 30 C in 0.18 M sucrose (pH 6.8-7.0) for 0-30 min. Granules were then separated from incubation medium by 10 min of microcentrifugation in the cold, the supernatant was diluted in 0.5% lanthanum oxide-3 N HC1, and the released calcium was measured using a Perkin-Elmer 290B atomic absorption spectrophotometer (Norwalk, CT). An airacetylene mixture with a tank regulator pressure of 8 psi was used, with a slit width of 0.7 nm at a wavelength of 423 nm. The total calcium in granules was routinely assessed after removal of protein by precipitation with an equal volume of 5% sulfosalicylic acid or by solubilization in sodium dodecyl sulfate (1% final concentration). Both methods gave comparable results, but the detergent solution tended to clog the inlet port of the spectrophotometer and, therefore, was not used routinely. The granule membrane potential was determined by distribution of the lipophilic [14C]thiocyanate anion, as described for a number of secretory granule systems (23-25); granule water space was estimated with 3H2O and [14C]dextran (20). Protein concentrations were determined by the method of Lowry et al. (26), with BSA as the standard. Bovine granule GH and PRL levels were measured by RIA, as previously described (27). Samples from triplicate incubations were each immunoassayed in duplicate at three dilutions within the hormone assay range (0.08-5.0 ng). Calcium release data are presented as a percentage of the total granule calcium found in the supernatant (mean ± SEM), corrected for the calcium found in the supernatant immediately after granules were added to microfuge tubes that contained 0.18 M sucrose (pH 6.8-7.0) at 4 C. Statistical comparisons between means were determined with Student's two-tailed t test for unpaired data and were considered significant at the 5% level of probability or less.

Results Granule calcium content In 26 granule preparations, the total secretory granule calcium was 14.5 ± 0.8 nmol/mg protein. We previously found the total calcium in pituitary homogenates to be 226 ± 25 Mg/g wet wt (18) or approximately 32.6 nmol/ mg protein. Using GH and PRL as markers for granule recovery in seven separate granule preparations, the granule calcium accounted for 21.2 ± 1.6% of the total calcium measured in homogenates after removal of cell debris. Granule calcium, thus, represents a substantial portion of the total cellular calcium and would represent a significant calcium reservoir if it could be mobilized.


CALCIUM CHANNELS IN SECRETORY GRANULES Granule calcium release

We previously had shown that protein and hormone release from isolated pituitary secretory granules was pH sensitive, with little release below pH 7 and marked release above pH 8 (22). For the present studies on calcium release, we incubated granules close to neutrality in order to minimize the calcium release that might be associated with granule rupture. Table 1 summarizes results of experiments carried out in 0.18 M sucrose adjusted to pH 5, pH 6, or pH 7; at the end of 30 min at 30 C, the final pH values were pH 6.2, pH 6.5, and pH 6.8, respectively. For the zero time determinations, incubation mixtures were on ice before granule addition, and microfuging was begun within 30 sec after addition. At pH 7, only small amounts of protein and calcium were released at zero time; after 30 min, there was a 4-fold increase in calcium release to over 20% of the total, while protein release was roughly doubled to 7.4%. Incubation at lower pH resulted in minimal protein release, confirming prior results (22), but caused higher calcium release at both 0 and 30 min. Based on these results, we chose 0.18 M sucrose at pH 7.0 as our routine incubation condition, since the zero time calcium release was low, and the pH did not change significantly throughout the incubation. We tested whether added KC1 would influence calcium release from granules and found there was a marked effect. For example, addition of 15 mM KC1 to granule incubations resulted in 78% calcium release, and virtually all granule calcium was released with 50 mM KC1. In contrast, the addition of KC1 to granule incubations had no effect on protein release. Thus, the added KC1 presumably resulted in depolarization of the granule membrane; using the steady state distribution of [14C] thiocyanate, we found that the granule membrane potential was 22 mV at pH 6.9. Initial studies using sucrose buffered with phosphate, Tris, HEPES, or 3-(iV-Morpholino)propane sulfonic acid (MOPS) were all associated with acceleration of granule TABLE 1. Calcium and protein release from secretory granules incubated in 0.18 M sucrose Release (% of total)

Condition

pH5 pH6 pH7

0 min

30 min

Ca 2+

Protein

Ca 2+

Protein

11.5 ± 0.6 9.0 ± 0.2 5.5 ± 0.3

3.8 ± 0.1 4.0 ± 0.2 3.8 ± 0.1

32.4 ± 1.4 22.0 ± 1.0 20.6 ± 0.9

6.6 ± 0.1 6.3 ± 0.2 7.4 ± 0.1

Secretory granules (5 mg/ml) were incubated in 0.18 M sucrose previously adjusted to pH 5, pH 6, or pH 7 (no adjustment). The amount of calcium and protein released at 0 and 30 min at 30 C was determined as described in Materials and Methods. Values given are the mean ± SE for triplicate incubations.

2673

calcium release. For example, zero time release in 50 mM HEPES-0.18 M sucrose was 69.5%, 65%, and 47.5% at pH 5, 6, and 7, respectively. At 10-mM buffer concentrations, zero time release at pH 6.5 was 54% (phosphate), 42% (Tris), 47% (MOPS), and 43% (HEPES). Since such stimulation could obscure changes in calcium release in response to other agents, we carried out subsequent experiments in sucrose alone, adjusting the pH of all additives before addition. When incubations were initiated between pH 6.8 and pH 7.0, the medium pH did not change substantially; the final pH range was between 6.6-6.9. GSH stimulation of calcium release We have previously shown that GSH can influence hormone storage structures, stimulate hormone release, and disrupt granule integrity, especially under alkaline conditions (18, 20, 22, 27). We, therefore, were curious to determine its effect on calcium release; the results of incubations carried out at varying GSH concentrations are shown in Fig. 1. At pH 7.0, calcium release was more than tripled as the GSH concentration was raised to 1 mM and above. The GSH effect was also readily demonstrated at lower pH values, despite the higher baseline rate of release. Thus, at pH 6, for example, calcium release was 24.9% in the absence of GSH and 88.0% with 5 mM GSH. At this pH, protein release was 5.3% without GSH and 14.5% with the thiol. For convenience, the above results were obtained from 30-min incubations; stimulated calcium release from granules, however, was complete at much earlier time periods. A time study is shown in Fig. 2, in which granules were incubated at 30

"5 30 •

,2 Q UJ

20"

3 UJ IAJ 0C

10"

CO

O

GSH (mM) FIG. 1. Influence of GSH on calcium release. Triplicate incubations of granules at 5 mg protein/ml were exposed to the indicated concentrations of GSH at pH 7.0 for 30 min at 30 C, and the calcium released was determined. Values are expressed as a percentage of the total granule calcium, which was 15.1 nmol/mg; SEs were less than 4.2% of their means.



2674

100-

30

Time (min) FIG. 2. Time course of GSH-stimulated calcium release. Triplicate incubations of granules (5 mg/ml) were exposed to 5 mM GSH-0.18 M sucrose at pH 6 and 30 C. After 0, 0.5,1, 2, 5,15, or 30 min, incubations were microfuged at 4 C, and the calcium in the supernatant was determined as described in Materials and Methods; for the zero time condition, granules were added to buffer at 4 C and immediately microfuged. Calcium released at zero time at pH 7 without GSH (3.7% of the total) was subtracted from all values. TABLE 2. Lack of effect of strontium (SrCl2) and lanthanum (LaCl3) on calcium release Calcium released (% of total) Condition 0.18 M sucrose +500 MM SrCl2 +500 MM LaCl3

Without GSH

With GSH

6.5 9.1 7.8

24.6 24.6 24.6

Secretory granules (5 mg protein/ml) were incubated at 30 C in 0.18 M sucrose or 0.18 M sucrose-2.5 mM GSH in the absence or presence of 500 uM SrCl2 or LaCl3. After 30 min, the amount of calcium released into the supernatant was determined as described in Materials and Methods. Averages of triplicate incubations are given; the differences in calcium release in the absence of GSH were not significant.

C with 5 mM GSH at pH 6.0 for up to 30 min. More than 79% of the releasable calcium was found in the supernate when incubations were microfuged immediately after granule addition, and more than 88% was released within the first minute. Under these conditions, maximal release had occurred by 2 min. Both basal and GSH-stimulated calcium release were also measured in the presence of strontium chloride or lanthanum chloride; results are shown in Table 2. At 500 ixM added cation, no significant effect was observed on release from granules in either the absence or presence of GSH. Thiol and disulfide influence on calcium release Since GSH had a potent stimulatory effect on calcium release even at neutral pH, we initiated studies on calcium release in the presence of other thiols and disulfides;

Endo • 1990 Voll26«No5

these results are summarized in Table 3. Both thiols and disulfides stimulated release, although GSH and cysteine were more potent than their oxidized counterparts. Stimulation was not directly related to reducing potential, however, since mercaptoethanol (MCE) and dithiothreitol (DTT) were relatively impotent. These thiols were without effect at 1 mM MCE and 0.1 mM DTT, concentrations that would be approximately equal to the reducing potency of 5 mM GSH (data not shown). Studies were also carried out in which iV-ethylmaleimide (NEM) was incubated with granules. In the presence of 5 mM NEM alone, calcium release was slightly increased from 6.8% to 11.2% of the total. However, pretreatment of granules with 10 mM NEM partly blocked (by 17-28%) the stimulation elicited by subsequent exposure to 5 mM thiols. Dihydropyridine sensitivity of calcium release Pituitary cells possess well characterized voltage-dependent calcium channels (28); ND is a dihydropyridine that has been shown to block these channels and influence hormone secretion from pituitary cell lines (29, 30). Data shown in Fig. 3 indicate that ND inhibited the augmented calcium release seen with GSH; the calculated apparent K; was in the 10-20 nM range. Greater than 85% inhibition was noted with 50 nM ND. Under basal conditions without GSH, more variability in the extent of inhibition by ND was noted, but higher concentrations of inhibitor were needed to demonstrate suppression. In one experiment, for example, 50 nM ND had little if any effect on basal calcium release, but 100 nM ND reduced it from 7.7% to 1.2%. ND also inhibited KCl-stimulated calcium release; these results are shown in Table 4. In the experiment shown in Table 4, inhibition by ND was 18.8% without KC1, but in the 50% range in the presence of 0.05-50 mM KC1. In contrast to GSH-stimulated calcium release, we were unable to completely block KCl-stimulated calTABLE 3. Influence of thiols and disulfides on calcium release from granules Addition None GSH Oxidized GSH Cysteine Cystine

MCE DTT

Calcium released (% of total) 5.3 39.3 25.3 37.5 18.7 8.5 12.2

Secretory granules (5 mg protein/ml) were incubated for 30 min at 30 C in 0.18 M sucrose with the indicated thiols or disulfides. All were at a final concentration of 5 mM, except MCE and DTT which were at 2.5 mM. Values given are the means of triplicate incubations; SEs were not greater than 5.6% of their means.



2675

iuu 80"

o Ul

60"

tn

UJ «j

\ \

40"

UJ

oc (0

o

20" o10

100 Ca

NIMODIPINE (nM) FIG. 3. ND inhibition of GSH-stimulated calcium release. Secretory granules were incubated in the dark at 30 C for 30 min in 0.18 M sucrose-2.5 mM GSH and ND at 0, 1, 2, 10, 25, and 50 nM. ND was solubilized in dimethylsulfoxide, which was kept constant (2.5%) in all incubations. Data are presented as the mean percentage of maximal stimulated release (n = 3). SEs were less than 3.2% of their respective means. TABLE 4. Inhibition of KCl-stimulated release by ND KC1

Calcium released (% of total) Without ND

0 0.05

0.10 0.50 5.0

15.0 50.0

Inhibition (07 \

(mM) 6.9

13.9 16.7 19.4 43.1 77.9 100.0

With ND 5.6 6.2 6.9 8.3

22.2 42.4 55.1

18.8 55.4 58.7 57.2 48.5 45.6 44.9

Secretory granules (5 mg protein/ml) were incubated at 30 C in 0.18 M sucrose (pH 7.0) without or with 100 nM ND and the indicated KC1. After 30 min, the amount of calcium released was measured, as described in Materials and Methods.

cium release with ND. The effect of ND was also tested on the accelerated zero time release in the presence of HEPES or phosphate. No inhibition of release was seen (data not shown). These results suggest that high rates of calcium release in the presence of HEPES and phosphate occurred via a different mechanism than that responsible for basal or GSHstimulated calcium release. A comparison of ND effects on calcium and PRL release is shown in Fig. 4. Granules were incubated without or with 10 or 50 nM ND in the absence or presence of 2.5 mM GSH. In the absence of GSH, ND caused a slight dose-related but not statistically significant decrease in calcium release; there was no doserelated effect on PRL. When GSH was present, ND resulted in progressive inhibition of calcium release, with 36.7% inhibition at 10 nM and 90% inhibition at 50 nM.

PRL

-GSH

Ca

PRL

+GSH

FIG. 4. ND influence on calcium and PRL release in the absence or presence of 2.5 mM GSH. Calcium and PRL release in triplicate incubations were determined as described in Materials and Methods and Fig. 3, with or without 5.0 mM GSH and with 0 (M), 10 nM (•), or 50 nM (•) ND.

GSH-stimulated PRL release fell 26.5% with 10 nM ND (P < 0.05), but no statistically significant inhibition was noted at the higher ND dose. Nifedipine, nisoldipine, and nitrendipine were also tested for their influence on GSH-stimulated calcium release. Inhibition by nifedipine was qualitatively and quantitatively similar to inhibition by ND. In contrast, no significant inhibition by nisoldipine and nitrendipine was demonstrable. Lack of influence of Bay K-8644 on calcium release Calcium channels inhibited by dihydropyridines have been shown in other systems to be stimulated by Bay K8644 (30-32). For this reason, the drug was tested for its effects on calcium release from granules. No effects were demonstrable. Two separate batches of Bay K-8644 were tested under a variety of conditions. These included incubating (or preincubating for 10 min) granules with from 25 nM to 2 ^M Bay K-8644 alone, with Bay K-8644 in the presence of GSH up to 2.5 mM, or treating granules with 10 and 50 nM ND and Bay K-8644 in the absence or presence of GSH. Discussion These observations indicate that pituitary PRL and GH secretory granules comprise a quantitatively significant calcium reservoir within the cell. The magnitude of granule calcium stores is such that mobilization of even a small fraction could easily account for the entire range of cytosolic calcium transients known to occur


2676


under stimulated conditions (1-4). However, if granule calcium could not be mobilized, its presence would be irrelevant to dynamic cellular regulation. Morphological evidence has been presented for intragranular calcium localization around the granule periphery, just inside the limiting membrane, and for changes in cellular calcium distribution after secretory stimulation with theophylline and potassium (33). We are not aware of independent confirmation of the latter observation, but our studies demonstrating acute releasability of granule calcium are consistent with that report. Our studies of in vitro calcium release, reported here, demonstrate that release is rapid. There is also substantial pH dependency of both basal and stimulated calcium release, with greater release being seen at lower pH values. Acid lability may have implications for the nature of the intragranular calcium binding; maximum stability was seen with incubations carried out at pH 7.0, very close to normal intracellular pH. If this finding can be extrapolated to the intact cell, it would provide a stable baseline suitable for acute regulatory modulation. It should be noted that the intragranular environment is more acidic than the cytosol (20). The stimulation of granule calcium release by HEPES, phosphate, Tris, and MOPS buffers was quantitatively substantial; therefore, incubations were carried out in 0.18 M sucrose alone. The pH did not change more than 0.2 U when incubations were initiated between pH 6.8 and pH 7.0. Buffer-dependent calcium release was not inhibited by ND, which suggests a mechanism separate from the ND- and nifedipine-sensitive basal and GSHstimulated calcium release processes. We have no further information as to the nature of this phenomenon. The data shown in Fig. 1 indicate the ability of GSH to stimulate calcium release from granules under conditions where protein release (22) [and granule rupture (22, 34)] probably could not account for the released cation. Other data, as well as that shown in Fig. 1, indicate that GSH at concentrations of 1 mM or less exerts virtually no effect at neutral pH on granule protein release, while the vast majority of the stimulatory effect on calcium occurs between 0.01-1 mM GSH. The stimulatory effect of GSH could be only partially abolished (20-25%) by 10 mM NEM, suggesting that although some of the effect is exerted via free thiol groups accessible to NEM, more than one site and/or mechanism may be involved. These interpretations should be, viewed cautiously due to the variability of release and the quantitatively small influence of the thiol-blocking agent. The rapidity of calcium movements in the stimulated state is well documented by the data shown in Fig. 2; about 80% of the total releasable calcium was already in the medium in the most rapidly processed samples. Rapid kinetics would be predicted for any calcium reservoir

Endo • 1990 Vol 126 »No 5

involved in acute cellular responses, of course. In addition to GSH, other sulfhydryls and related compounds were also tested, as shown in Table 3. Potency was clearly not related only to reducing power, as shown by the limited calcium release seen with MCE and DTT. Disulfides were also capable of releasing calcium, but were somewhat less effective; these observations might be accounted for by conversion of added disulfides to thiols during incubation, thus producing stimulatory compounds. Thiohdisulfide interchange may occur nonenzymatically or might be facilitated by enzymes such as the pituitary GSH:disulfide oxidoreductase we have previously described (35); we have no data bearing on this question directly. The state of granule thiols and/or disulfides may also play a key role in the regulation of granule calcium transport. The concentration of intragranular GSH is about 0.9 ixg/mg protein or close to 3 mM, assuming a water space of about 1 ix\/mg (18, 20). It is not known whether granule GSH is bound or free and what its role is in the maintenance of hormone storage forms and granule integrity. It is also unclear whether the intragranular thiohdisulfide equilibria are influenced by thiols and disulfides added to the incubation medium. Effectiveness of added thiols (or disulfides) may be related to accessibility to the site(s) of action (i.e. hormone storage forms, alternative intragranular sites, the outer or inner granule membrane surface, or combinations of these). In sarcoplasmic reticulum, for example, calcium release appears to be regulated in part by the oxidation state of a key membrane sulfhydryl (36). At present, we cannot localize the effect within the granule; we hypothesize that if the data of Leuschen et al. (33) are correct, the interface area between granule contents and the granule limiting membrane is a topographically likely site of action, since that is where the calcium appears to be located. ND is a member of the class of dihydropyridines, which are potent calcium channel blockers (37). Its ability to interfere with GSH-stimulated calcium release, to partially inhibit KCl-stimulated calcium release, and to blunt basal calcium release suggests that the calcium movements into and from granules occur in large part via channels that are operationally similar to those in excitable tissues and those that can be blocked in intact pituitary cells (30). Pharmacological evidence for calcium channels also infers rapid kinetics, which is consistent with the data shown in Fig. 2. Compared to ND, nifedipine has a nitro group in the ortho rather than the meta position, and methyl esters in both the 3 and 5 positions of the heterocyclic ring. It had effects similar to ND. In contrast, nisoldipine and nitrendipine were ineffective; the former has an o-nitro group and the latter is mnitrophenyl; both have methyl esters in the 5 position, but the former is 3-isobutyl-, while the latter is 3-ethyl



ester. Thus, the structural requirements for inhibition are unclear, although differences in potency and selectivity in other systems are well established, and differences in relative order of inhibitory potency toward calcium channels in pituitary cells have been shown as well (30). More broadly, ND and nifedipine inhibition of GSHstimulated calcium release is consistent with the notion that granule calcium transport sites can be regulated. Further, the extent of the regulation is shown by the nearly complete inhibition of calcium release by ND. Granule breakage or rupture can, therefore, not be involved as a contributory mechanism. Coupled with the pH data shown in Table 1, the ability of ND to reduce calcium release without an effect on PRL release indicates the independence of movements of these two granule components. One of the unexpected results encountered was the lack of effect of Bay K-8644; it did not stimulate basal or GSH-stimulated calcium release, and it had no effect on inhibition by ND. This calcium channel agonist, which can stimulate secretion from pituitary cells in culture (30, 31), acts by increasing the open time of the channel gating function (32). Its failure to influence net calcium movement in pituitary secretory granules might be due to pharmacologically distinct channels in these granules or, possibly, lack of granule penetration by the compound. We believe that these data indicate the presence of calcium channels in secretory granules; they appear to be similar, but not identical, to conventional voltagesensitive calcium channels found in excitable tissues. Among the differences of note is the inactivity of Bay K-8644, strontium, and lanthanum in our system. Since lanthanum and strontium are without effect, it is unlikely that the calcium is on the surface of the granules and is merely being exchanged during incubation. Thus, pituitary secretory granules contain a large pool of potentially releasable calcium, although the precise mode of storage is not understood. A substantial portion of this calcium can be released in vitro in less than a minute, and release is independent of protein release, is pH sensitive, and can be stimulated in a complex fashion by thiols and disulfides. The release of calcium observed in these studies could in theory result from inhibition of uptake or stimulation of efflux; since no calcium was included in the buffers, and the granules are known to contain substantial calcium, our observations can only be explained by the net stimulation of efflux. Isotopic studies will, however, be required to evaluate the possibility that granule uptake of previously released calcium might also occur. Additional questions relate to the in vivo kinetics of granule calcium movements, the nature of the intragranular calcium binding, and the functional role(s) released granule calcium might play in the cell.

2677

The nature of potential in vivo regulating signals for granule calcium release is not known. Although we cannot currently answer these questions, and the ultimate significance of our findings hinges on these issues in important ways, we should, nonetheless, underscore that, to our knowledge, this is the first report of rapidly releasable granule calcium. Hopefully, additional studies will yield answers to some of these questions.

Acknowledgment We thank Ms. Sandra Webster for secretarial assistance in the preparation of this manuscript.

References 1. Carafoli E 1987 Intracellular calcium homeostasis. Annu Rev Biochem 56:395 2. Somlyo AP, Himpens B 1989 Cell calcium and its regulation in smooth muscle. FASEB J 3:2266 3. Campbell AK 1983 Intracellular Calcium. Its Universal Role as Regulator. Wiley and Sons, Chichester 4. Rasmussen H 1982 Calcium as intracellular messenger in hormone action. Adv Exp Med Biol 151:473 5. Berridge MJ, Irvine RF 1984 Inositol trisphosphate, a novel second messenger in cellular signal transduction. Nature 312:315 6. Berridge MJ 1986 Inositol phosphates as second messengers. In: Putney Jr JW (ed) Phosphoinositides and Receptor Mechanisms. Liss, New York, p 25 7. Gill DL, Mullaney JM, Ghosh TK 1988 Intracellular calcium translocation: mechanism of activation by guanine nucleotides and inositol phosphates. J Exp Biol 139:105 8. Gershengorn MC, Geras E, Purrello VS, Rebecchi MJ 1984 Inositol trisphosphate mediates thyrotropin-releasing hormone mobilization of nonmitochondrial calcium in rat mammotropic pituitary cells. J Biol Chem 259:10675 9. Crompton M, Spiegel E, Salzmann M, Carafoli E 1976 A kinetic study of the energy-linked influx of Ca2+ into heart mitochondria. Eur J Biochem 69:429 10. Biden TJ, Wolheim CB, Schlegel W 1986 Inositol 1,4,5-trisphosphate and intracellular Ca2+ homeostasis in clonal pituitary cells (GH3). Translocation of Ca2+ into mitochondria from a functionally discrete portion of the nonmitochondrial store. J Biol Chem 261:7223 11. Streb H, Burne RF, Berridge MJ, Shultz I 1983 Release of Ca2+ from a nonmitochondrial intracellular store in pancreatic acinar cells by inositol-l,4,5-trisphosphate. Nature 306:67 12. Prentki M, Biden TJ, Janijc D, Irvine RF, Berridge MJ, Wollheim CB 1984 Rapid mobilization of Ca2+ from rat insulinoma microsomes by inositol 1,4,5-trisphosphate. Nature 309:562 13. Bayerdorffer E, Streb H, Eckhardt L, Haase W, Schulz I 1984 Characterization of calcium uptake into rough endoplasmic reticulum of rat pancreas. J Membr Biol 81:69 14. Volpe P, Krause K-H, Hashimoto S, Zorzato F, Pozzan T, Meldolesi J, Lew DP 1988 "Calciosome," a cytoplasmic organelle: the inositol 1,4,5-trisphosphate-sensitive Ca2+ store of nonmuscle cells? Proc Natl Acad Sci USA 85:1091 15. Pozzan T, Volpe P, Zorzato F, Bravin M, Drause, KH, Lew DP, Hashimoto S, Bruno B, Meldolesi J 1988 The Ins(l,4,5)P3-sensitive Ca2+ store of non-muscle cells: endoplasmic reticulum or calciosomes? J Exp Biol 139:181 16. Campbell KP 1986 Protein components and their roles in sarcoplasmic reticulum function. In: Entman ML, Van Winkle WB (eds) Sarcoplasmic Reticulum in Muscle Physiology. CRC Press, Boca Raton, p 65 17. Rossier MF, Capponi AM, Vallotton MB 1989 The inositol 1,4,5trisphosphate binding site in adrenal corticol cells is distinct from the endoplasmic reticulum. J Biol Chem 264:14078


2678


18. Greenan J, Lorenson MY, Conconi MV, Walker AM 1990 Alterations in in situ prolactin secretory granule morphology and immunoactivity by thiols and divalent cations. Endocrinology 126:512 19. Nelson TY, Lorenson MY, Jacobs LS, Boyd III AE 1987 Distribution of calmodulin and calmodulin-binding proteins in bovine pituitary: association of myosin light chain kinase with pituitary secretory granule membranes. Mol Cell Biochem 74:83 20. Lorenson MY, Miska SP, Jacobs LS 1984 Molecular mechanism of prolactin release from pituitary secretory granules. In: Mena F, Valverde CM (eds) Prolactin Secretion: A Multidisciplinary Approach. Academic Press, New York, p 141 21. Lorenson MY, Lee Y-C, Jacobs LS 1981 Identification and characterization of an anion-sensitive Mg2-ATPase in pituitary secretory granule membranes. J Biol Chem 256:12802 22. Lorenson MY, Jacobs LS 1982 Thiol regulation of protein, growth hormone and prolactin release from isolated adenohypophyseal secretory granules. Endocrinology 110:1164 23. Carty SE, Johnson RG, Scarpa A 1982 Electrochemical proton gradient in dense granules isolated from anterior pituitary. J Biol Chem 257:7269 24. Johnson RG, Scarpa A 1979 Protonmotive force and catecholamine transport in isolated chromaffin granules. J Biol Chem 254:3750 25. Hutton JC 1982 The internal pH and membrane potential of the insulin secretory granule. Biochem J 204:171 26. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ 1951 Protein measurement with the Folin phenol reagent. J Biol Chem 193:265 27. Lorenson MY 1985 In vitro conditions modify immunoassayability of bovine pituitary prolactin and growth hormone: insights into their secretory granule storage forms. Endocrinology 116:1399 28. Tan K-N, Tashjian Jr AH 1984 Voltage-dependent calcium chan-

29. 30.

31.

32. 33. 34. 35. 36. 37.

Endo • 1990 Voll26«No5

nels in pituitary cells in culture. I. Characterization by 48Ca2+ fluxes. J Biol Chem 259:427 Enyeart JJ, Aizawa T, Hinkle PM 1985 Dihydropyridine Ca2+ antagonists: potent inhibition of secretion from normal and transformed pituitary cells. Am J Physiol 248:C510 Enyeart JJ, Sheu S-S, Hinkle PM 1987 Dihydropyridine modulators of voltage-sensitive Ca2+ channels specifically regulate prolactin production by GH4Ci pituitary tumor cells. J Biol Chem 262:3154 Enjalbert A, Musset F, Chenard C, Priam M, Kordon C, Heisler S 1988 Dopamine inhibits prolactin secretion stimulated by the calcium channel agonist Bay K-8644 through a pertussis toxin-sensitive G protein in anterior pituitary cells. Endocrinology 123:406 Nowycky MC, Fox AP, Tsien RW 1985 Novel mode of neuronal calcium channel gating and its promotion by the dihydropyridine calcium agonist Bay K 8644. Proc Natl Acad Sci USA 82:2178 Leuschen MP, Moriarty CM, Sampson HW, Piscopo I 1981 Cytochemical analysis of intracellular calcium distribution in the anterior pituitary of the rat. Cell Tissue Res 220:191 Lorenson MY, Jacobs LS 1987 Osmotic pressure regulation of prolactin and growth hormone release from bovine secretory granules. Endocrinology 120:365 Lorenson MY, Lee Y-C, Jacobs LS 1981 Regulation of hormonal and secretory granule membrane disulfides by adenohypophysial glutathione: disulfide oxidoreductase. Life Sci 28:2309 Trimm JL, Salama G, Abramson JJ 1986 Sulfhydryl oxidation induces rapid calcium release from sarcoplasmic reticulum vesicles. J Biol Chem 261:16092 Miller RJ, Freedman SH 1984 Minireview. Are dihydropyridine binding sites voltage sensitive calcium channels? Life Sci 34:1205


Oxidation of dihydropyridine calcium channel blockers and analogues by human liver cytochrome P-450 IIIA4.

Gingival enlargement induced by dihydropyridine calcium channel blockers in a young child.

Calcium-channel blockers and anaesthesia.

Pituitary calcium channel modulation and regulation of prolactin gene expression.

Identification of ascorbate as an endogenous substance that irreversibly inhibits binding of dihydropyridine calcium channel blockers.

Calcium channel blockers and excitatory amino acids.

Calcium channel blockers and gingival overgrowth.

Calcium channel blockers and congestive heart failure.

Discovery and Development of Calcium Channel Blockers.

Copper, zinc and calcium: imaging and quantification in anterior pituitary secretory granules.

Calcium channel blockers in cardiovascular pharmacotherapy.

Calcium channel blockers in myocardial infarction.

Antiatherosclerotic effects of calcium channel blockers.

Calcium channel blockers for primary Raynaud's phenomenon.

Modulation by monovalent anions of calcium and caffeine induced calcium release from heavy sarcoplasmic reticulum vesicles.

Dihydropyridine binding of the calcium channel complex from skeletal muscle is modulated by subunit interaction.

Toward an understanding of the dihydropyridine-sensitive calcium channel.

The cardiac sarcoplasmic reticulum calcium-release channel: modulation of ryanodine binding and single-channel activity.

The effects of dihydropyridine calcium channel modulators on pentylenetetrazole convulsions.

Calcium channel blockers for inhibiting preterm labour and birth.

Voltage-dependent modulation of rat sensory neurone calcium channel currents by G protein activation: effect of a dihydropyridine antagonist.

Interaction between ethanol and calcium channel blockers in humans.

Calcium signaling and secretory responses in agonist-stimulated pituitary gonadotrophs.

Comparative cytoprotective effect of dihydropyridine calcium channel blockers against the toxicity of oxidized low density lipoprotein for cultured lymphoid cells.