In fura-%loaded bovine adrenal chromaffin cells, 0.5 p~ angiotensin I1 (AII) stimulated a 185 f 19 nM in- crease of intracellular-free calcium ([Ca2+Ii) approxi-.
Vol ,264, No. 31, Issue of November 5,PP. 18349-18355,1989 Printed in U.S.A.
CHEMISTRY THEJOURNAL OF BIOLOGICAL 0 1989 by The American Society for Biochemistry and Molecular Biology, Inc.
Dissociation of Ca2+Entry andCa2+Mobilization Responsesto Angiotensin I1 in Bovine Adrenal Chromaffin Cells* (Received for publication, June 27, 1988, and in revised form, July 3, 1989)
Kenneth A. Stauderman andRebecca M. Pruss From the Merrell Dow Research Institute, Cincinnati, Ohio 45215 lar Ca2+store associated with the endoplasmic reticulum (3). The mechanism controlling Ca2+entry is still uncertain, although it is thought to be a channel-mediated eventand that inositol lipid hydrolysis may be involved (4). Several hypotheses have been proposed to explain the relationship between Ca2+mobilization and Ca2+entry, but the one that seems most consistent with the data is the capacitance model of Ca2+ entry, inwhich the depletion of intracellular Ca2+ stores accompanying Ca2+ mobilization by Ins(1,4,5)P3 stimulates a capacitative entry of Ca2+to refill those stores (5). In examining this and othermodels of Ca2+ entry, an important question that arises is whetherCa2+ mobilization is both a necesary and sufficient stimulus for agonist-induced Ca2+entry. For example, in some tissues it is possible that Ca2+ release from internal stores alone is an insufficient stimulus to promote Ca2+entry but that it is a prerequisite for Ca2+entry stimulated by some other mechanism, perhaps via a second messenger (6, 7). In thepresent studythe sufficiency of agonist-induced Ca2+ mobilization to stimulate Ca2+entry was investigated in bovine adrenal chromaffin cells. These cells possess many characteristics of neuronal tissue, including voltage-operated ion channels (8,9). Chromaffin cells also possess receptor-linked processes, as shown by their ability to respond to angiotensin I1 (AII)with an increase in intracellular free Ca2+ concentration ([Ca"]i). (10). The identity of the Ca2+pool(s) involved in the All response and thepossible involvement of the phosphoinositide-signaling system has not been thoroughly investigated in chromaffin cells, as it has in other AII-responsive cells (11, 12). We report that, in bovine adrenal chromaffin cells, A I 1 stimulates 1) Ca2+ mobilization from intracellular stores, 2) Ca2+entry through the plasma membrane, and 3) rapid formation of [3H]Ins(1,4,5)P3.Furthermore, evidence is presented that AII-induced Ca2+mobilization is not the stimAgonist-induced release of intracellularly sequestered Ca2+ ulus that causes Ca2+entry in response to All. (Ca2+mobilization) followed by the entryof Ca2+through the plasma membrane (Ca2+entry) appears to be an immutable EXPERIMENTALPROCEDURES sequence of eventsina variety of cells (1, 2). The Ca2+ Chromaffin Cell Preparation-Chromaffin cells were prepared from mobilization response is believed to result from the agonist- bovine adrenal glands obtained from a local slaughterhouse (Klustimulated generation of inositol 1,4,5-trisphosphate ener's, Cincinnati, OH) as described previously (13). The percentage (Ins(1,4,5)P3),lwhich causes Ca2+release from an intracellu- of chromaffin cells in each preparation, assessed by staining with
In fura-%loaded bovine adrenal chromaffin cells, 0.5 angiotensin I1 (AII) stimulated a 185 f 19 nM increase of intracellular-free calcium ([Ca2+Ii)approximately 3 s after addition. The time from the onset of the response until achieving 50%recovery (ts)was 67 f 10 s. Concomitantly, An stimulated both therelease of 46Caz+ from prelabeled cells, and a 4-5-fold increase of [3H]inositol 1,4,5-trisphosphate (['H]Ins(1,4,5)P3) levels. In thepresence of 60 p~ LaC13, or when extracellular-free Ca2+ ([Ca"'].) was less than 100 nM, AH still rapidly increased[Ca2+]iby 95-135 nM, but thet% for recovery was then only 23-27 s. In medium with 1 mM MnClz present, Art also stimulated a small amount of Mn2+influx, as judged by quenching of the fura-2 signal. When [Ca"'], was normal (1.1mM) or low (e60 nM), 1-2 pM ionomycin caused [Ca2+Iito increase 204 f 26 nM, while also releasing45-55% of bound 46Ca2+. With low [Ca"+],, ionomycin pretreatment abolished both the [Ca2+]iincrease and 46Ca2+release stimulated by AII. However, after ionomycin pretreatment in normal medium, AI^ produced a La3+-inhibitable increase of [Ca2+]i(103 f 13 nM) with a tLhof 89 f 8 s, but no 46Ca2+release. No pretreatment condition altered AI^induced formation of ['H]Ins(l,4,5)P3. We conclude that AII increased [Ca2+Iivia rapid and transientCa2+ mobilization from Ins( 1,4,5)P3- andionomycin-sensitive stores, accompanied (andlor followed) by Ca2+entry through a La3+-inhibitable divalent cation pathway. Furthermore, the ability of AII to activate Ca"' entry in the absence of Ca2+ mobilization (i.e. after ionomycin pretreatment) suggests areceptor-linked stimulus other than Ca2+ mobilization initiates Ca2+ entry. p~
neutral red, was at least 85%. For culturing, cells were plated overnight in 75-cm2 flasks (Costar) at 20 X lo6 cells/flask in 20 ml of the payment of page charges. This article must therefore be hereby Dulbecco's modified Eagle's medium (DMEM) containing 10% heatmarked "aduertisement" in accordance with 18 U.S.C. Section 1734 inactivated fetal bovine serum, 2 mM glutamine, 100 units/ml penicillin, 100 pg/ml streptomycin, and M cytosine arabinofuranosolely to indicate this fact. The abbreviations used are: Ins(1,4,5)P3, D-myo-inositol 1,4,5- side. Intracellular Free Ca2+Measurements-Freshly prepared or 24-h trisphosphate; AII,angiotensin 11; [Ca2+Ii,intracellular-free Ca2+concentration; [Ca2f]o, extracellular-free Caz+ concentration; [3H]InsP, cultured cells were washed and resuspended at a density of lo6 cells/ [3H]inositolphosphate (specific inositol phosphates are abbreviated); ml in HEPES-buffered Krebs saline (HBK; 118 mM NaCl, 4.6 mM tv,, time from the onset of the AILresponse to the time when [Ca2+Ij KC1,lO mM glucose, 25 mM sodium-HEPES (Sigma),1.2 mM MgSO,, had recovered to 50% of the maximum increase; DMEM, Dubecco's 1.1 mM CaC12, 0.1% bovine serum albumin, pH 7.4). After addition modified Eagle's medium; HEPES, N-2-hydroxyethylpiperazine-N'- of 2 +M fura-2 acetoxymethylester the mixture was incubated with 2-ethanesulfonic acid; HBK, HEPES-buffered Krebs saline solution; shaking a t 37"C for 30 min. The cells were then washed once in EGTA, [ethylenebis(oxyethylenenitrilo)]tetraaceticacid. HBK and incubated 10 min at 37 "C to allow deesterification of fura-
* The costs of publication of this article were defrayed in part by
18349
18350
AI[-induced Changes of [Ca"]i in Chromaffin Cells
2 acetoxymethylester. After a final wash, viable chromaffin cells were counted by staining with neutral red, resuspended to IO6 cellslml in HBK, and kept at 22 "C until use. A 2-mi aliquot of cells was transferred to a quartz cuvette placed in a thermostatted sample compartment of a dual-excitation wavelength spectrofluorometer (Photon Technology International, Inc.). The cells were prewarmed to 37 "C with stirring for 5 min prior to measurements. Fluorescence emission (510 nm) was monitored with excitation wavelengths cycling between 340 and 380 nm such that a 340:380 nm ratio could be measured 30 times/s. At the end of an experiment, fluorescence maximum and minimum values at each excitation wavelength were obtained by first lysing the cells with 0.1% Triton X-100 (maximum) and then adding 4 mM EGTA (minimum). With the maximum and minimum values, and after subtracting the values for cell autofluorescence, the 340:380 nm fluorescence ratios were converted into free calcium concentrations using the formula and fura-2 CaZ+binding constant (224 nM) described by Grynkiewiczet al. (14). Where applicable, extracellular-free Ca2+concentration ([Ca"].) was calculated using a computer program (15). In the experiments with MnC12, fluorescence emission was monitored with a single excitation wavelength of 358 nm, which was the isosbestic point for fura-2 in our system. The fluorescence output under these conditions was independent of changes in cytosolic calcium. PHlInsP(1,4,5)P3 Formation-Cells were labeled with ['H]inositol by a modification of the method described by Nabika et al. (16). Briefly, freshly prepared cells were cultured as described above except that themedium contained inositol-free DMEM plus dialyzed serum (24-h dialysis) and [3H]myo-inositol(20 pCi/ml; 15 or 35 Ci/mmol). Following a 20-h incubation at 37 "C the cells were harvested, washed, resuspended in HBK (see above) a t 5 X lo6 cellslml, and warmed at 37 "C for 20-30 min. Any pretreatments were performed during the last minutes of this period. Pretreatment with ionomycin was for 5 min, and LaC1, was always added 1 min prior to the addition of AI]. The reactions were initiated by adding 500 pl of the cell suspensions to polypropylene tubes containing 8 p1 of All or vehicle (HBK) and were terminated by adding 500 pl of ice-cold 20% (w/v) trichloroacetic acid. These mixtures were kept onice for 15 min and thencentrifuged in amicrocentrifuge for 5 min. The supernatants were then extracted with 4 X 4 volumes of water-saturated diethyl ether and neutralized with NHAOH. Separation of [3H]Ins(l,4,5)P~ was performed by the high pressure liquid chromatography method of Dean and Moyer (17)using on-line radioactivity detection (Radiomatic Flow-One Beta). The radioactivity detector measures total counts, which can be converted into counts per minute as follows: cpm = total counts/residence timein the detector (0.5 min). Q u a n t i ~ t i o nof [3H]Ins(1,4,5)P3was achieved by integration of the appropriate peak. Identification of individual ['HI InsPs was tentatively made by comparing the elution times with 3H standards. 45Ca2+Experiments-Cells cultured 20 h were harvested, washed, and resuspended at 10' cells/ml in HBK. After a 30-min preincubation a t 37 "C, 4sCa2+(5-7 pCi/ml) was added and thecells incubated another 45 min at 37 "C to allow equilibration with *'Ca2+.The cells were then split into separateexperimental groups (described in figure legends), pelleted, and resuspended in either normal HBK buffer containing 1.1 mM CaCl2,HBK containing 4.1mM CaC& and 3 mM EGTA (1.1mM free Cap+),or buffer with no added CaC12 plus 0.1 mM EGTA. The amount of 45Ca2+retained by BCCs was determined by filtering 1-ml aliquots of cells through prewetted Whatman GF/C filters followed by three washes with 5 ml of a quench buffer, which consisted of HBK buffer minus bovine serum albumin and CaC12, and containing 2 mM LaCl3. The amount of radioactivity on the filters was then determined by liquid scintillation spectrophotometry. Materials-The following compounds were purchased from the suppliers listed in parentheses: fura-2 acetoxymethylester (Molecular Probes, Inc., Eugene, OR), [3H]myo-inositol(American Radioligand Laboratories, St. Louis, MO), DMEM glutamine, penicillin, and streptomycin (GIBCO), fetal bovine serum (Advanced Biotech. Inc., Silver Spring, MD), bovine serum albumin (ICN, Lisle, IL), angiotensin I1 (Peninsula, Belmont, CA), and ionomycin free acid (Behring Diagnostics). All other chemicals were of the highest purity obtainable from Sigma or Fluka Biochemicals (Ronkonkoma, NY). t3H]InsP standards were either prepared (17) or obtained from Du Point-New England Nuclear. *'CaC12 (39.5 mCi/mg) was from Du Pont-New England Nuclear.
RESULTS
An-induced Changes of (Caz+],in Bouine Adrenal Chromaffin Ceh-In normal medium containing 1.1 mM CaClz the initial resting [CaZ+li(mean k S.E.) in the chromaffin cells was 140 rt 10 nM ( n = 14). During the course of individual experiments we observed that the [Ca2+Iisignal gradually increased at a rate of 5-8 nM/min, which was due to slow leakage of fura-2 out of the cells rather than anactual increase in the [CaZ+li. Addition of 0.5 p~ resulted in a rapid 185 rt 19 nM increase in [Ca2+jzthat peaked about 3 s after the onset of the response, followed by a gradual decline until a new steady-state [Ca2+Iiwas reached that was 10-50 nM higher than prestimulation levels(Fig. la; Table I). The normal tendency of the [Ca2+Iisignal to increase with time (see above) can account for the slightly raised [Ca"], at the end of the AII response. The time from the onset of the A11 response to the time when [Ca2+]i had recovered to 50% o f the maximum increase (tTA) was 67 2 10 s (Table I). Roles of ~ x t r ~ e ~and l uI ~nr ~ r ~ e Cu2+-The ~ u ~ r relative contribution of extracellular and intracellular Ca" pools to the response to AII was examined. After lowering the [Ca2+l0 to
