0013-7227/91/1294-2126I03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society
Vol. 129, No. 4 Printed in U.S.A.
Growth Hormone-Releasing Factor Stimulates Calcium Entry in the GH3 Pituitary Cell Line* L. BRESSON, M. FAHMI, P. SARTOR, B. DUFY, AND L. DUFY-BARBE CNRS URA 1200, Laboratoire de Neurophysiologie, Uniuersite de Bordeaux II, Bordeaux, France
spontaneously active cells or triggered [Ca2+]i oscillations in inactive cells. The response to GHRF was totally blocked by external Ca2+ free solutions and Ca2+ channel blockers. Combined electrophysiological and fluorescent experiments were carried out in 16 cells. Eleven responded to GHRF. In all cases, the Ca2+ transients triggered by GHRF were associated with action potentials. The Ca2+ responses observed in our experiments clearly show that GH3 cells possess membrane receptors to GHRF. Thus, it is likely that the lack of secretory response observed in GH3 cells does not result from the absence of binding sites to the peptide. It is more likely to be related to alterations of transduction mechanisms resulting in uncoupling between stimulation and secretion. (Endocrinology 129: 2126-2130, 1991)
ABSTRACT. The GH3 pituitary cell line has been extensively used to study various aspects of the stimulus secretion coupling process. It is known that GH3 cells release PRL and GH in the basal state and in response to various secretagogues. However, this cell line was considered unsuitable as a model for studying the effects of GHRF since the neuropeptide did not affect GH secretion or gene expression. This suggested that the GH3 cells may lack GHRF receptors. The present study investigates the effect of GHRF on free intracellular Ca2+ concentrations in GH3 cells. Cytosolic free calcium concentrations ([Ca2+]i) were monitored in individual cells by microspectrofluorimetry using the fluorescent dye indo 1. When the cells were challenged with a brief application of GHRF (100 nM; 15 sec), 36 out of 59 of these cells responded within a few seconds by a marked increase in [Ca2+]i. GHRF enhanced the frequency of [Ca2+]i oscillations in
S
basal state and respond to various secretagogues such as high K+, TRH, and derivatives of cAMP (7). Some years ago, this cell line was found unsuitable as a model for studying the effects of GHRF, since the neuropeptide did not affect GH secretion or gene expression (8). To our knowledge, the effect of GHRF on intracellular calcium concentrations has not been described in these cells. The recent availability of new techniques, especially those allowing the investigation of peptide effects at the single cell level on intracellular free calcium concentrations ([Ca2+]i), prompted the present study. Thus, we studied in single GH3 cells the effect of GHRF on [Ca2+]i as evaluated by microspectrofluorimetry using the fluorescent probe indo 1. To further analyze the nature of the observed modifications, in some experiments, we simultaneously recorded, in the same cell, changes in membrane ion conductances by electrophysiological techniques. Our results show that in approximately 60% of a GH3 cell population, GHRF stimulates Ca2+ entry by activating action potential firing.
ECRETION of GH by the pituitary is mainly regulated by two hypothalamic peptides: GHRF and somatostatin, a GH release inhibiting factor. GHRF acts on somatotrophs by binding to a cell surface receptor, thus activating intracellular messengers. Different transduction pathways have been associated with GHRF action. Early studies implicated both Ca2+ and cAMP in the first steps contributing to the GHRF-induced GH release. GHRF increases cAMP levels in cultured pituitary cells (1, 2). GHRF-induced GH secretion also depends on the presence of extracellular Ca2+ in the medium (1, 2) and is inhibited by blockers of voltagesensitive Ca2+ channels such as verapamil and cadmium or cobalt ions (2). In rat somatotrophs in primary culture, GHRF was recently shown to elevate cytosolic calcium levels secondary to an influx of extracellular Ca2+ through calcium channels (3-5). The pituitary cell line GH3 has been extensively used to study various aspects of the stimulus secretion coupling process (6). GH3 cells release PRL and GH in the Received April 22,1991. Address all correspondence and requests for reprints to: L. DufyBarbe, CNRS URA 1200, Laboratoire de Neurophysiologie, 146 rue Leo Saignat, 33076 Bordeaux Cedex (France). * This work was funded by the Centre National de la Recherche Scientifique, University of Bordeaux II, Etablissement Public Regional, and Fondation pour la Recherche Medicale.
Materials and Methods Maintenance of the GH3 cells GH3/B6 cells, a subclone of the GH3 cell line, were used in these experiments. They were initially given to our laboratory by A. Tixier-Vidal and D. Gourdji (College de France, Paris).
2126
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
GHRF ACTION ON GH3 CELLS In this study, they were routinely cultured in DMEM-F12 (50/ 50), (Seromed Strasbourg, France) supplemented by 8 mM/1 NaHCO3, 2 mM/1 1-glutamine, 1 mM/1 sodium pyruvate, and 10% heat inactivated fetal bovine serum (Seromed, Strasbourg, France). The cells were routinely grown as stocks in 25 cm2 plastic flasks (Nunc, Poly-Labo, Strasbourg, France) at 37 C in a humidified air/CO2 (93/7) atmosphere. The medium was changed twice a week and the cells were passaged every 10-12 days. No antibiotics were added to the cultures. The cells used in the experiments were subcultured on 30 mm diameter glass coverslips which had been pretreated with polyornithine (5 g/1). They were used 3-6 days after trypsinization. Microspectrofluorimetry [Ca2+]i was monitored in individual cells by microspectrofluorimetry using the fluorescent dye indo 1 as described by Mollard et al. (9). The cells were loaded by incubation for 30 min at room temperature in a Hank's balanced salt solution (HBSS, 142.6 mM NaCl, 5.6 mM KC1, 2 mM CaCl2, 0.8 mM MgCl2, 5 mM glucose, 10 mM HEPES-NaOH, and buffered to pH 7.3) containing 5 nM indo 1 penta-acetoxymethyl-ester (indo I/AM, Calbiochem, La Jolla, CA) and 0.02% Pluronic F127 (Molecular Probes, Eugene, OR). The cells were then rinsed with HBSS and the glass coverslip holding the cultured cells was sealed to a hollowed plastic Petri dish which was placed on the microscope stage. During the recordings, the cells were bathed with HBSS containing 5% fetal bovine serum since serum addition was found to improve the number of responses and the overall behavior of the cells. A thermostatic device maintained the temperature of the bathing medium at 36 ± 1 C for the duration of the experiments. GHRF and other test substances were applied to the recorded cell by low pressure ejection from micropipettes (tip diameter 3-5 nm) positioned at approximately 20 nm from the cell membrane. Measurements were performed using a dual emission microscope (9). Indo 1 was excited at 355 nm and the light reflected off a dichroic mirror (380 nm). Emitted fluorescence signal was passed through a diaphragm slightly larger than the observed cell and directed to another dichroic mirror (455 nm). Transmitted light was filtered at 480 nm, reflected light filtered at 405 nm, and the intensities recorded by separate photometers (Nikon, Garden City, NY). Single photon currents were converted to voltage signals which were divided on line by a monolithic laser trimmed two-quadrant divider (AD 535, Analog Devices, Norwood, MA). After substraction of the mean background (fluorescence recorded from unloaded cells), the ratio F 405/F 480 was recorded on line as a voltage signal traced with a pen recorder (Gould, Santa Clara, CA) and transformed to [Ca2+]i according to the method of Grynkiewicz et al. (10). Electrophysiological experiments Membrane potential and ionic currents were recorded by using the whole cell configuration of the tight-seal recording technique (whole cell recording) described in detail elsewhere (11). The amplifying system included a Dagan 8800 (Dagan Corp, Minneapolis, MN) equipped with a 100 megohm resistor in the headstage for whole cell recording. All other equipment
2127
and cell bathing solutions were as described above. Whole cell patch pipettes had tip resistances of 3-5 megohms when filled with a salt solution (referred to as K glu) containing: K gluconate, 140 mM; MgCl2, 2 mM; EGTA, 1.1 mM; and HEPES-KOH, 5 mM. All internal solutions were adjusted to pH 7.3 ± 0.01. The osmolality was maintained at 290-295 mOsm/kg, 5-10 mOsm/kg hypoosmolar to the external solution. In combined microspectrofluorimetric and electrophysiological experiments, the electrophysiological signal and the emission wavelength ratio F 405/F 480 were simultaneously displayed on line on the pen recorder and digitized for subsequent analysis on a IBM AT using P-Clamp software (Axon, Burlingame, CA). Chemicals All chemicals used in this study were from Sigma (L'Isle D'Abeau, Chesnes, France) unless otherwise specified. GHRF (human, pancreatic, l-44-NH2) was purchased from UCB Bioproducts (Nanterre, France). The peptide was stored at -80 C diluted in acetate buffer at 10"4 M. Further dilutions were made in HBSS. The calcium channel blocker nifedipine was initially stored dissolved in dimethylsulfoxide at 10 mM. An intermediate dilution at 10~4 M was prepared in dimethylsulfoxide and the final dilution was made in HBSS. Statistical analysis Results were expressed as mean ± SD. The unpaired t test was used for statistical analysis of the results. Results 2+
Measurement of [Ca Ji Basal [Ca2+]i. Basal [Ca2+]i was measured in a total of 125 cells. In the absence of applied neuropeptide, we found that 56% of the cells (70 out of 125 cells) exhibited spontaneous fluctuations of their [Ca2+]i (Fig. la). The frequency and amplitude of these oscillations varied widely between experiments and, within the same experiment, from cell to cell. Some cells showed slow (1-12/ min) fluctuations of wide amplitude (maximal amplitude recorded, 248 nM). Others oscillated more rapidly with a lower amplitude. Some examples of typical oscillatory patterns are shown in Fig. lal, Ia2, Ia3. In these oscillating cells, the mean nadir value measured during interspike intervals was chosen as mean basal [Ca2+]i. Conversely, 55 out of 125 cells did not oscillate and [Ca2+]i was fairly stable (Fig. Ia4). In these nonactive or "silent" cells, the basal [Ca2+]i varied according to the cells, from 62-167 nM. The mean basal [Ca2+]i of these two categories of cells (oscillating and nonoscillating cells), taken altogether was 128 ± 36 nM. In active cells, the ejection of calcium free HBSS solution immediately suppressed the spontaneous fluctuations in 6 out of 6 cells (Fig. lbl), as did the ejection
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
GHRF ACTION ON GH3 CELLS
2128 3
[Ca2-]i nM
Endo • 1991 Vol 129 • No 4
]i nM
GH-RF
190
250 i
110 GH-RF 250 200 1508025(h
GH-RF
20015010010sec 2+
FlG. 2. [Ca ]i responses to a brief application (100 nM; 15 sec) of
GHRF in single GH3 cells, a, Induction of Ca2+ spikes in a previously silent cell; b, acceleration and amplification of Ca2+ transients in a previously active cell; c, summation of Ca2+ transients. nM 10sec
FlG. 1. Basal [Ca2+]i in individual GH3 cells as monitored by microspectrofluorimetry. a, Resting levels in unstimulated cells: al,2,3: spontaneously active cells; a4: nonactive cell, b, Inhibition of Ca2+ spontaneous activity induced by the brief application of: bl: Ca2+-free HBSS containing 2 mM EGTA; b2: cadmium chloride (500 fiM) in HBSS; b3: nifedipine (200 nM) in HBSS.
of calcium channel blockers such as cadmium chloride (500 fiM) in 13 out of 13 cells (Fig. Ib2) or nifedipine (200 nM) in 4 out of 4 cells (Fig. Ib3). Moreover, the mean basal [Ca2+]i was also significantly decreased during the ejection of these substances. Effect of GHRF on [Ca2+]i. The effect of GHRF was tested on 59 randomly chosen cells of which 32 were spontaneously active. When the cells were challenged with a brief application of GHRF (100 nM; 15 sec), most of them (36 out of 59) responded within a few seconds by a marked increase in [Ca2+]i. Among the 36 responsive cells, 17 were previously active. The occurrence of a response was, thus, unrelated to the existence of a previous Ca2+ spontaneous activity. The time course and kinetics of the response were highly variable. In 4 out of 19 previously silent cells, GHRF elicited the apparition of a series of well individualized Ca2+ transients (Fig. 2a). In the other 15 previously silent cells, a step increase in [Ca2+]i was observed, followed by sustained oscillations and a progressive decline toward resting values (not shown). In spontaneously active cells, GHRF increased the amplitude and frequency of [Ca2+]i oscillations (Fig. 2b). In some cases, the summation of the oscillations resulted in a plateau on which oscillations of lower amplitude were superimposed (Fig. 2c). In 23 cells, no modification
OmM GH^RF
230
130 GH-RF
OmM Ca2*
GH-RF
CdClz
300 200 140
120 J 10sec 2+
FlG. 3. Effect of Ca -free HBSS and calcium channel blockers on the [Ca2+]i response to GHRF. a, Application of GHRF during the course of an ejection of Ca2+ free HBSS containing 2 mM EGTA; no response was observed in 11/11 cells; b, inhibition of GHRF-induced [Ca2+]i increase by the ejection of Ca2+-free HBSS containing 2 mM EGTA; c, by a solution of cadmium chloride (500 nu) in HBSS; d, by a solution of nifedipine (200 nM) in HBSS.
of [Ca2+]i was observed after GHRF. These cells were therefore classified as unresponsive. To further explore the origins of these transient fluctuations of [Ca2+]i, external Ca2+ free solutions and Ca2+ channel blockers were used. When GHRF was ejected during the course of an ejection of Ca2+-free HBSS, the response was totally blocked in 11 out of 11 cells tested (Fig. 3a). In 8 out of 8 cells, the ejection of Ca2+-free
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
GHRF ACTION ON GH3 CELLS
HBSS during the development of a response immediately lowered [Ca2+]i and suppressed the response which was partially restored as soon as Ca2+-free HBSS ejection stopped (Fig. 3b). Similarly, GHRF-induced [Ca2+]i increase was inhibited by the ejection of cadmium chloride or nifedipine (Fig. 3, c and d). The mean maximum [Ca2+]i reached after 100 nM GHRF was 256 ± 68 nM (n = 36). Due to technical limitations imposed by the use of indo 1 which results in some bleaching, it was often impossible to observe the return to baseline levels. Consequently, we could not properly evaluate the duration of the response in all the cells studied. Whenever possible, we measured the duration of the responses to this short application (100 nM; 15 sec) of GHRF which lasted between 30 sec and 3 min. Combined measurement of [Ca2+]i and electrophysiological activity
Electrophysiological recordings of GH3 cells have shown that these cells are excitable and display Ca2+ action potentials either spontaneously (active cells) or in response to depolarization (silent cells). The combination of electrophysiological recording techniques with the monitoring of [Ca2+]i at subsecond time resolution was used to investigate directly the link between [Ca2+]i and changes in membrane potential. In cells that were loaded with the fluorescent probe indo 1 (see Materials and Methods) action potential firing caused a transient elevation in !Ca2+]i with the expected kinetic features of a rapid onset with a transition time of less than 1 sec and a return to basal [Ca2+]i within a few seconds. When the simultaneous monitorings of [Ca2+]i and current clamp recording of membrane potential were performed in a total of 16 cells, 11 responded to GHRF. In silent cells, at a concentration of 100 nM, a 15-sec application of GHRF provoked a slight depolarization (5-10 mV) associated with a series of all-or-none action potentials (Fig. 4b). The effect was maximal 1-2 min after GHRF administration and was reversible (Fig. 4). Since Ca2+ was simultaneously monitored with the probe indo 1, it was observed, in all cases, that GHRFinduced action potentials caused transient elevations of [Ca2+]i (Fig. 4a). Under similar conditions, GHRF (100 nM) enhanced the frequency of firing of spontaneously active cells. The increase in firing rate was also associated with a rise in [Ca2+]i (not shown). Taken together, these data provide evidence that, in GH3 cells, GHRFinduced increases in [Ca2+]i are due to an enhancement of membrane excitability leading to Ca2+ entry.
Discussion We show here that GHRF significantly increases [Ca2+]i in most cells (61%) of a GH3 cell population.
[Ca2*]i 263-1
2129 GH-RF
193150-
0_
10sec
FIG. 4. Simultaneous recording of a microspectrofluorimetric and electrophysiological response to GHRF in a single GH3 cell, a, Microspectrofluorimetric measurement of [Ca2+]i; b, current clamp recording of the electrical activity. Note that each Ca2+ transient, as measured with the fluorescent probe indo 1, is associated with an action potential.
According to the cells, this increase results from either a step increase followed by a plateau and a progressive return toward basal levels or from the acceleration and/ or amplification of Ca2+ transients spontaneously occurring in more than half the cells. All the cells tested did not respond to the peptide. Pituitary cells show considerable heterogeneity of response type. Previously, most studies in which biochemical responses were investigated used populations of thousands if not millions of cells. Recent, more accurate, assays allow single cell response analysis. Among these are electrophysiological techniques and measurement of [Ca2+]i by the micromethod used in our study. With such methods, it is generally found that, although monoclonal in essence, the cells of the GH3 cell line of a given population are heterogenous regarding their response to neuropeptides. Winiger and Schlegel (12), looking at the Ca2+ response to TRH in individual GH3, found that 92% of the cells were responsive. Among them, 24% showed both the first and the second phases of the response, 25% showed only the first phase, and 43% only the second phase. The reasons for this heterogeneity are not clear. It is possible that some receptors or binding molecules are expressed during specific stages of the cell cycle. Since the cultures used in our studies were not synchronized, this might explain the coexistence of subpopulations of responsive and unresponsive cells to GHRF. In any case, this [Ca2+]i increase was immediately reversed when the calcium channel blockers nifedipine or cadmium chloride were ejected close to the surface of the cell membrane, which indicates that Ca2+ influx from the extracellular medium is responsible for this elevation. This finding is confirmed by the results of combined experiments which clearly show that increases in [Ca2+]i result from action potential firing triggered by the application of GHRF. Our results agree with those of previous studies in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
2130
GHRF ACTION ON GH3 CELLS
which GHRF was shown to increase [Ca2+]i in mixed populations of rat anterior pituitary cells (3), populations of purified rat somatotrophs (5), or individual rat somatotrophs identified by plaque assay (13). In all cases, this increase was shown to result from Ca2+ uptake from the external medium. The electrophysiological observations of our study corroborate those of Chen et al. (14) who found that GHRF induced a depolarizing response simultaneously with a decrease in membrane resistance and acceleration of action potential firing in somatotroph-enriched cultures from rat pituitaries. In the GH3 cell line, TRH, which stimulates the release of PRL and GH (7), induces a biphasic rise in [Ca2+]i involving an initial Ca2+ release from internal stores followed by a Ca2+ influx from the extracellular medium (15,16). The combination of microspectrofluorimetric monitoring of [Ca2+]i with electrophysiological recordings obtained using the patch clamp technique allows the establishment of correlations between changes in [Ca2+]i and alterations in ionic currents occurring at the plasma membrane level (17). During the time-course of the TRH response, Ca2+ mobilization from intracellular stores correlates with changes in conductance and current, which allows identification of the Ca2+-induced activation of a K+ current to the first phase of TRH action and Ca2+ mobilization (16). In the present study, we detected no contribution of Ca2+ release from internal stores to the effect of GHRF. The combination of electrophysiological recordings with the microfluorimetric determination of [Ca2+]i has already allowed us to identify spontaneous oscillations in [Ca2+]i of pituitary cells as being the consequence of action potential firing (9, 17). The simultaneous monitoring of [Ca2+]i and current clamp recording performed in this study showed that, in GH3 cells, the GHRF-induced increase in intracellular free Ca2+ results exclusively from Ca2+ entry. Thus, in its first stages, the mechanism of action of GHRF is similar in GH3 and normal rat somatotrophs in primary culture (2-5). In both cell types, an increased Ca2+ influx is exclusively responsible for the increase of [Ca2+]i provoked by the neuropeptide. The calcium responses observed in our experiments clearly show that GH3 cells possess membrane receptors to this important physiological stimulus, GHRF. Thus, it is likely that the lack of secretory response to GHRF previously reported in GH3 cells (8) does not result from the absence of binding sites to the peptide on the surface of these cells but could be related to alterations of transduction mechanisms resulting in uncoupling between stimulation and secretion. Additional experiments will be necessary to further clarify this point.
Endo • 1991 Vol 129 • No 4
Acknowledgments We are indebted to G. Gaurier, F. Bertrand, and D. Varoqueaux for their help in the preparation of the manuscript.
References 1. Bilezikjian LM, Vale WW 1983 Stimulation of adenosine 3',5' monophosphate production by growth hormone-releasing factor and its inhibition by somatostatin in anterior pituitary cells in vitro. Endocrinology 113:1726-1731 2. Brazeau P, Ling N, Esch F, Boehlen P, Mougin C, Guillemin R 1982 Somatocrinin (growth hormone-releasing factor) in vitro bioactivity: Ca2+ involvement, cAMP mediated action and additivity of effect with PGE2. Biochem Biophys Res Commun 190:588594 3. Schofl C, Sandow J, Knepel W, 1987 GRF elevates cytosolic calcium concentration in rat anterior pituitary cells. Am J Physiol 253:E591-E594 4. Holl RW, Thorner MO, Leong DA 1989 Cytosolic free calcium in normal somatotropes: effects of forskolin and phorbol ester. Proc Natl Acad Sci USA 256:E375-E379 5. Lussier BT, French MB, Moor BC, Kraicer J 1991 Free intracellular Ca2+ concentration ([Ca2+]i) and growth hormone release from purified rat somatotrophs. I-GH-releasing factor-induced Ca2+ influx raises [Ca2+]i. Endocrinology 128:570-582 6. Gershengorn MC 1986 Mechanism of thyrotropin-releasing hormone stimulation of pituitary hormone secretion. Annu Rev Physiol 48:515-526 7. Ostlund Jr RE, Leung JT, Vaerewyck SH, Winokur T, Melman M 1978 Acute stimulated hormone release from cultured GH3 pituitary cells. Endocrinology 103:1245-1252 8. Zeytin FN, Gick GG, Brazeau P, Ling N, McLaughlin M, Bancroft C 1984 Growth hormone (GH)-releasing factor does not regulate GH release or GH mRNA levels in GH3 cells. Endocrinology 114:2054-2059 9. Mollard P, Guerineau N, Audin J, Dufy B 1989 Measurement of Ca2+ transients using simultaneous dual-emission microspectrofluorimetry and electrophysiology in individual pituitary cells. Biochem Biophys Res Commun 164:1045-1052 10. Grynkiewicz G, Poenie M, Tsien RY 1985 A new generation of Ca2+ indicators with greatly improved fluorescent properties. J Biol Chem 260:3440-3450 11. Sartor P, Dufy-Barbe L, Corcuff JB, Taupignon A, Dufy B 1990 Electrophysiological response to thyrotropin-releasing hormone of rat lactotrophs in primary culture. Am J Physiol 258:E311-E319 12. Winiger BP, Schlegel W 1988 Rapid transient elevations of cytosolic calcium triggered by thyrotropin releasing hormone in individual cells of the pituitary line GH3/B6. Biochem J 255:161-167 13. Holl HW, Thorner O, Leong DA 1988 Intracellular calcium concentration and growth hormone secretion in individual somatotropes: effects of growth-hormone releasing factor and somatostatin. Endocrinology 122:2927-2932 14. Chen C, Israel JM, Vincent JD 1989 Electrophysiological responses of rat pituitary cells in somatotroph-enriched primary culture to human growth hormone releasing factor. Neuroendocrinology 50:679-687 15. Gershengorn MC, Thaw C 1985 Thyrotropin releasing-hormone (TRH) stimulates biphasic elevation of cytoplasmic free calcium in GH3 cells. Further evidence that TRH mobilizes cellular and extracellular Ca2+. Endocrinology 116:591-596 16. Mollard P, Dufy B, Barker JL, Schlegel W 1990 Thyrotropinreleasing-hormone activated a [Ca2+]i-dependent K+ current in GH3 pituitary cells via Ins (1,4,5) P3-sensitive and Ins (1,4,5) PSinsensitive mechanisms. Biochem J 268:345-352 17. Schlegel W, Winiger BP, Mollard P, Vacher P, Wuarin F, Zahnd G, Wolheim CB, Dufy B 1987 Oscillations of cytosolic Ca2+ in pituitary cells due to action potentials. Nature 329:719-721
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
Vol. 129, No. 4 Printed in U.S.A.
Growth Hormone-Releasing Factor Stimulates Calcium Entry in the GH3 Pituitary Cell Line* L. BRESSON, M. FAHMI, P. SARTOR, B. DUFY, AND L. DUFY-BARBE CNRS URA 1200, Laboratoire de Neurophysiologie, Uniuersite de Bordeaux II, Bordeaux, France
spontaneously active cells or triggered [Ca2+]i oscillations in inactive cells. The response to GHRF was totally blocked by external Ca2+ free solutions and Ca2+ channel blockers. Combined electrophysiological and fluorescent experiments were carried out in 16 cells. Eleven responded to GHRF. In all cases, the Ca2+ transients triggered by GHRF were associated with action potentials. The Ca2+ responses observed in our experiments clearly show that GH3 cells possess membrane receptors to GHRF. Thus, it is likely that the lack of secretory response observed in GH3 cells does not result from the absence of binding sites to the peptide. It is more likely to be related to alterations of transduction mechanisms resulting in uncoupling between stimulation and secretion. (Endocrinology 129: 2126-2130, 1991)
ABSTRACT. The GH3 pituitary cell line has been extensively used to study various aspects of the stimulus secretion coupling process. It is known that GH3 cells release PRL and GH in the basal state and in response to various secretagogues. However, this cell line was considered unsuitable as a model for studying the effects of GHRF since the neuropeptide did not affect GH secretion or gene expression. This suggested that the GH3 cells may lack GHRF receptors. The present study investigates the effect of GHRF on free intracellular Ca2+ concentrations in GH3 cells. Cytosolic free calcium concentrations ([Ca2+]i) were monitored in individual cells by microspectrofluorimetry using the fluorescent dye indo 1. When the cells were challenged with a brief application of GHRF (100 nM; 15 sec), 36 out of 59 of these cells responded within a few seconds by a marked increase in [Ca2+]i. GHRF enhanced the frequency of [Ca2+]i oscillations in
S
basal state and respond to various secretagogues such as high K+, TRH, and derivatives of cAMP (7). Some years ago, this cell line was found unsuitable as a model for studying the effects of GHRF, since the neuropeptide did not affect GH secretion or gene expression (8). To our knowledge, the effect of GHRF on intracellular calcium concentrations has not been described in these cells. The recent availability of new techniques, especially those allowing the investigation of peptide effects at the single cell level on intracellular free calcium concentrations ([Ca2+]i), prompted the present study. Thus, we studied in single GH3 cells the effect of GHRF on [Ca2+]i as evaluated by microspectrofluorimetry using the fluorescent probe indo 1. To further analyze the nature of the observed modifications, in some experiments, we simultaneously recorded, in the same cell, changes in membrane ion conductances by electrophysiological techniques. Our results show that in approximately 60% of a GH3 cell population, GHRF stimulates Ca2+ entry by activating action potential firing.
ECRETION of GH by the pituitary is mainly regulated by two hypothalamic peptides: GHRF and somatostatin, a GH release inhibiting factor. GHRF acts on somatotrophs by binding to a cell surface receptor, thus activating intracellular messengers. Different transduction pathways have been associated with GHRF action. Early studies implicated both Ca2+ and cAMP in the first steps contributing to the GHRF-induced GH release. GHRF increases cAMP levels in cultured pituitary cells (1, 2). GHRF-induced GH secretion also depends on the presence of extracellular Ca2+ in the medium (1, 2) and is inhibited by blockers of voltagesensitive Ca2+ channels such as verapamil and cadmium or cobalt ions (2). In rat somatotrophs in primary culture, GHRF was recently shown to elevate cytosolic calcium levels secondary to an influx of extracellular Ca2+ through calcium channels (3-5). The pituitary cell line GH3 has been extensively used to study various aspects of the stimulus secretion coupling process (6). GH3 cells release PRL and GH in the Received April 22,1991. Address all correspondence and requests for reprints to: L. DufyBarbe, CNRS URA 1200, Laboratoire de Neurophysiologie, 146 rue Leo Saignat, 33076 Bordeaux Cedex (France). * This work was funded by the Centre National de la Recherche Scientifique, University of Bordeaux II, Etablissement Public Regional, and Fondation pour la Recherche Medicale.
Materials and Methods Maintenance of the GH3 cells GH3/B6 cells, a subclone of the GH3 cell line, were used in these experiments. They were initially given to our laboratory by A. Tixier-Vidal and D. Gourdji (College de France, Paris).
2126
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
GHRF ACTION ON GH3 CELLS In this study, they were routinely cultured in DMEM-F12 (50/ 50), (Seromed Strasbourg, France) supplemented by 8 mM/1 NaHCO3, 2 mM/1 1-glutamine, 1 mM/1 sodium pyruvate, and 10% heat inactivated fetal bovine serum (Seromed, Strasbourg, France). The cells were routinely grown as stocks in 25 cm2 plastic flasks (Nunc, Poly-Labo, Strasbourg, France) at 37 C in a humidified air/CO2 (93/7) atmosphere. The medium was changed twice a week and the cells were passaged every 10-12 days. No antibiotics were added to the cultures. The cells used in the experiments were subcultured on 30 mm diameter glass coverslips which had been pretreated with polyornithine (5 g/1). They were used 3-6 days after trypsinization. Microspectrofluorimetry [Ca2+]i was monitored in individual cells by microspectrofluorimetry using the fluorescent dye indo 1 as described by Mollard et al. (9). The cells were loaded by incubation for 30 min at room temperature in a Hank's balanced salt solution (HBSS, 142.6 mM NaCl, 5.6 mM KC1, 2 mM CaCl2, 0.8 mM MgCl2, 5 mM glucose, 10 mM HEPES-NaOH, and buffered to pH 7.3) containing 5 nM indo 1 penta-acetoxymethyl-ester (indo I/AM, Calbiochem, La Jolla, CA) and 0.02% Pluronic F127 (Molecular Probes, Eugene, OR). The cells were then rinsed with HBSS and the glass coverslip holding the cultured cells was sealed to a hollowed plastic Petri dish which was placed on the microscope stage. During the recordings, the cells were bathed with HBSS containing 5% fetal bovine serum since serum addition was found to improve the number of responses and the overall behavior of the cells. A thermostatic device maintained the temperature of the bathing medium at 36 ± 1 C for the duration of the experiments. GHRF and other test substances were applied to the recorded cell by low pressure ejection from micropipettes (tip diameter 3-5 nm) positioned at approximately 20 nm from the cell membrane. Measurements were performed using a dual emission microscope (9). Indo 1 was excited at 355 nm and the light reflected off a dichroic mirror (380 nm). Emitted fluorescence signal was passed through a diaphragm slightly larger than the observed cell and directed to another dichroic mirror (455 nm). Transmitted light was filtered at 480 nm, reflected light filtered at 405 nm, and the intensities recorded by separate photometers (Nikon, Garden City, NY). Single photon currents were converted to voltage signals which were divided on line by a monolithic laser trimmed two-quadrant divider (AD 535, Analog Devices, Norwood, MA). After substraction of the mean background (fluorescence recorded from unloaded cells), the ratio F 405/F 480 was recorded on line as a voltage signal traced with a pen recorder (Gould, Santa Clara, CA) and transformed to [Ca2+]i according to the method of Grynkiewicz et al. (10). Electrophysiological experiments Membrane potential and ionic currents were recorded by using the whole cell configuration of the tight-seal recording technique (whole cell recording) described in detail elsewhere (11). The amplifying system included a Dagan 8800 (Dagan Corp, Minneapolis, MN) equipped with a 100 megohm resistor in the headstage for whole cell recording. All other equipment
2127
and cell bathing solutions were as described above. Whole cell patch pipettes had tip resistances of 3-5 megohms when filled with a salt solution (referred to as K glu) containing: K gluconate, 140 mM; MgCl2, 2 mM; EGTA, 1.1 mM; and HEPES-KOH, 5 mM. All internal solutions were adjusted to pH 7.3 ± 0.01. The osmolality was maintained at 290-295 mOsm/kg, 5-10 mOsm/kg hypoosmolar to the external solution. In combined microspectrofluorimetric and electrophysiological experiments, the electrophysiological signal and the emission wavelength ratio F 405/F 480 were simultaneously displayed on line on the pen recorder and digitized for subsequent analysis on a IBM AT using P-Clamp software (Axon, Burlingame, CA). Chemicals All chemicals used in this study were from Sigma (L'Isle D'Abeau, Chesnes, France) unless otherwise specified. GHRF (human, pancreatic, l-44-NH2) was purchased from UCB Bioproducts (Nanterre, France). The peptide was stored at -80 C diluted in acetate buffer at 10"4 M. Further dilutions were made in HBSS. The calcium channel blocker nifedipine was initially stored dissolved in dimethylsulfoxide at 10 mM. An intermediate dilution at 10~4 M was prepared in dimethylsulfoxide and the final dilution was made in HBSS. Statistical analysis Results were expressed as mean ± SD. The unpaired t test was used for statistical analysis of the results. Results 2+
Measurement of [Ca Ji Basal [Ca2+]i. Basal [Ca2+]i was measured in a total of 125 cells. In the absence of applied neuropeptide, we found that 56% of the cells (70 out of 125 cells) exhibited spontaneous fluctuations of their [Ca2+]i (Fig. la). The frequency and amplitude of these oscillations varied widely between experiments and, within the same experiment, from cell to cell. Some cells showed slow (1-12/ min) fluctuations of wide amplitude (maximal amplitude recorded, 248 nM). Others oscillated more rapidly with a lower amplitude. Some examples of typical oscillatory patterns are shown in Fig. lal, Ia2, Ia3. In these oscillating cells, the mean nadir value measured during interspike intervals was chosen as mean basal [Ca2+]i. Conversely, 55 out of 125 cells did not oscillate and [Ca2+]i was fairly stable (Fig. Ia4). In these nonactive or "silent" cells, the basal [Ca2+]i varied according to the cells, from 62-167 nM. The mean basal [Ca2+]i of these two categories of cells (oscillating and nonoscillating cells), taken altogether was 128 ± 36 nM. In active cells, the ejection of calcium free HBSS solution immediately suppressed the spontaneous fluctuations in 6 out of 6 cells (Fig. lbl), as did the ejection
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
GHRF ACTION ON GH3 CELLS
2128 3
[Ca2-]i nM
Endo • 1991 Vol 129 • No 4
]i nM
GH-RF
190
250 i
110 GH-RF 250 200 1508025(h
GH-RF
20015010010sec 2+
FlG. 2. [Ca ]i responses to a brief application (100 nM; 15 sec) of
GHRF in single GH3 cells, a, Induction of Ca2+ spikes in a previously silent cell; b, acceleration and amplification of Ca2+ transients in a previously active cell; c, summation of Ca2+ transients. nM 10sec
FlG. 1. Basal [Ca2+]i in individual GH3 cells as monitored by microspectrofluorimetry. a, Resting levels in unstimulated cells: al,2,3: spontaneously active cells; a4: nonactive cell, b, Inhibition of Ca2+ spontaneous activity induced by the brief application of: bl: Ca2+-free HBSS containing 2 mM EGTA; b2: cadmium chloride (500 fiM) in HBSS; b3: nifedipine (200 nM) in HBSS.
of calcium channel blockers such as cadmium chloride (500 fiM) in 13 out of 13 cells (Fig. Ib2) or nifedipine (200 nM) in 4 out of 4 cells (Fig. Ib3). Moreover, the mean basal [Ca2+]i was also significantly decreased during the ejection of these substances. Effect of GHRF on [Ca2+]i. The effect of GHRF was tested on 59 randomly chosen cells of which 32 were spontaneously active. When the cells were challenged with a brief application of GHRF (100 nM; 15 sec), most of them (36 out of 59) responded within a few seconds by a marked increase in [Ca2+]i. Among the 36 responsive cells, 17 were previously active. The occurrence of a response was, thus, unrelated to the existence of a previous Ca2+ spontaneous activity. The time course and kinetics of the response were highly variable. In 4 out of 19 previously silent cells, GHRF elicited the apparition of a series of well individualized Ca2+ transients (Fig. 2a). In the other 15 previously silent cells, a step increase in [Ca2+]i was observed, followed by sustained oscillations and a progressive decline toward resting values (not shown). In spontaneously active cells, GHRF increased the amplitude and frequency of [Ca2+]i oscillations (Fig. 2b). In some cases, the summation of the oscillations resulted in a plateau on which oscillations of lower amplitude were superimposed (Fig. 2c). In 23 cells, no modification
OmM GH^RF
230
130 GH-RF
OmM Ca2*
GH-RF
CdClz
300 200 140
120 J 10sec 2+
FlG. 3. Effect of Ca -free HBSS and calcium channel blockers on the [Ca2+]i response to GHRF. a, Application of GHRF during the course of an ejection of Ca2+ free HBSS containing 2 mM EGTA; no response was observed in 11/11 cells; b, inhibition of GHRF-induced [Ca2+]i increase by the ejection of Ca2+-free HBSS containing 2 mM EGTA; c, by a solution of cadmium chloride (500 nu) in HBSS; d, by a solution of nifedipine (200 nM) in HBSS.
of [Ca2+]i was observed after GHRF. These cells were therefore classified as unresponsive. To further explore the origins of these transient fluctuations of [Ca2+]i, external Ca2+ free solutions and Ca2+ channel blockers were used. When GHRF was ejected during the course of an ejection of Ca2+-free HBSS, the response was totally blocked in 11 out of 11 cells tested (Fig. 3a). In 8 out of 8 cells, the ejection of Ca2+-free
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
GHRF ACTION ON GH3 CELLS
HBSS during the development of a response immediately lowered [Ca2+]i and suppressed the response which was partially restored as soon as Ca2+-free HBSS ejection stopped (Fig. 3b). Similarly, GHRF-induced [Ca2+]i increase was inhibited by the ejection of cadmium chloride or nifedipine (Fig. 3, c and d). The mean maximum [Ca2+]i reached after 100 nM GHRF was 256 ± 68 nM (n = 36). Due to technical limitations imposed by the use of indo 1 which results in some bleaching, it was often impossible to observe the return to baseline levels. Consequently, we could not properly evaluate the duration of the response in all the cells studied. Whenever possible, we measured the duration of the responses to this short application (100 nM; 15 sec) of GHRF which lasted between 30 sec and 3 min. Combined measurement of [Ca2+]i and electrophysiological activity
Electrophysiological recordings of GH3 cells have shown that these cells are excitable and display Ca2+ action potentials either spontaneously (active cells) or in response to depolarization (silent cells). The combination of electrophysiological recording techniques with the monitoring of [Ca2+]i at subsecond time resolution was used to investigate directly the link between [Ca2+]i and changes in membrane potential. In cells that were loaded with the fluorescent probe indo 1 (see Materials and Methods) action potential firing caused a transient elevation in !Ca2+]i with the expected kinetic features of a rapid onset with a transition time of less than 1 sec and a return to basal [Ca2+]i within a few seconds. When the simultaneous monitorings of [Ca2+]i and current clamp recording of membrane potential were performed in a total of 16 cells, 11 responded to GHRF. In silent cells, at a concentration of 100 nM, a 15-sec application of GHRF provoked a slight depolarization (5-10 mV) associated with a series of all-or-none action potentials (Fig. 4b). The effect was maximal 1-2 min after GHRF administration and was reversible (Fig. 4). Since Ca2+ was simultaneously monitored with the probe indo 1, it was observed, in all cases, that GHRFinduced action potentials caused transient elevations of [Ca2+]i (Fig. 4a). Under similar conditions, GHRF (100 nM) enhanced the frequency of firing of spontaneously active cells. The increase in firing rate was also associated with a rise in [Ca2+]i (not shown). Taken together, these data provide evidence that, in GH3 cells, GHRFinduced increases in [Ca2+]i are due to an enhancement of membrane excitability leading to Ca2+ entry.
Discussion We show here that GHRF significantly increases [Ca2+]i in most cells (61%) of a GH3 cell population.
[Ca2*]i 263-1
2129 GH-RF
193150-
0_
10sec
FIG. 4. Simultaneous recording of a microspectrofluorimetric and electrophysiological response to GHRF in a single GH3 cell, a, Microspectrofluorimetric measurement of [Ca2+]i; b, current clamp recording of the electrical activity. Note that each Ca2+ transient, as measured with the fluorescent probe indo 1, is associated with an action potential.
According to the cells, this increase results from either a step increase followed by a plateau and a progressive return toward basal levels or from the acceleration and/ or amplification of Ca2+ transients spontaneously occurring in more than half the cells. All the cells tested did not respond to the peptide. Pituitary cells show considerable heterogeneity of response type. Previously, most studies in which biochemical responses were investigated used populations of thousands if not millions of cells. Recent, more accurate, assays allow single cell response analysis. Among these are electrophysiological techniques and measurement of [Ca2+]i by the micromethod used in our study. With such methods, it is generally found that, although monoclonal in essence, the cells of the GH3 cell line of a given population are heterogenous regarding their response to neuropeptides. Winiger and Schlegel (12), looking at the Ca2+ response to TRH in individual GH3, found that 92% of the cells were responsive. Among them, 24% showed both the first and the second phases of the response, 25% showed only the first phase, and 43% only the second phase. The reasons for this heterogeneity are not clear. It is possible that some receptors or binding molecules are expressed during specific stages of the cell cycle. Since the cultures used in our studies were not synchronized, this might explain the coexistence of subpopulations of responsive and unresponsive cells to GHRF. In any case, this [Ca2+]i increase was immediately reversed when the calcium channel blockers nifedipine or cadmium chloride were ejected close to the surface of the cell membrane, which indicates that Ca2+ influx from the extracellular medium is responsible for this elevation. This finding is confirmed by the results of combined experiments which clearly show that increases in [Ca2+]i result from action potential firing triggered by the application of GHRF. Our results agree with those of previous studies in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
2130
GHRF ACTION ON GH3 CELLS
which GHRF was shown to increase [Ca2+]i in mixed populations of rat anterior pituitary cells (3), populations of purified rat somatotrophs (5), or individual rat somatotrophs identified by plaque assay (13). In all cases, this increase was shown to result from Ca2+ uptake from the external medium. The electrophysiological observations of our study corroborate those of Chen et al. (14) who found that GHRF induced a depolarizing response simultaneously with a decrease in membrane resistance and acceleration of action potential firing in somatotroph-enriched cultures from rat pituitaries. In the GH3 cell line, TRH, which stimulates the release of PRL and GH (7), induces a biphasic rise in [Ca2+]i involving an initial Ca2+ release from internal stores followed by a Ca2+ influx from the extracellular medium (15,16). The combination of microspectrofluorimetric monitoring of [Ca2+]i with electrophysiological recordings obtained using the patch clamp technique allows the establishment of correlations between changes in [Ca2+]i and alterations in ionic currents occurring at the plasma membrane level (17). During the time-course of the TRH response, Ca2+ mobilization from intracellular stores correlates with changes in conductance and current, which allows identification of the Ca2+-induced activation of a K+ current to the first phase of TRH action and Ca2+ mobilization (16). In the present study, we detected no contribution of Ca2+ release from internal stores to the effect of GHRF. The combination of electrophysiological recordings with the microfluorimetric determination of [Ca2+]i has already allowed us to identify spontaneous oscillations in [Ca2+]i of pituitary cells as being the consequence of action potential firing (9, 17). The simultaneous monitoring of [Ca2+]i and current clamp recording performed in this study showed that, in GH3 cells, the GHRF-induced increase in intracellular free Ca2+ results exclusively from Ca2+ entry. Thus, in its first stages, the mechanism of action of GHRF is similar in GH3 and normal rat somatotrophs in primary culture (2-5). In both cell types, an increased Ca2+ influx is exclusively responsible for the increase of [Ca2+]i provoked by the neuropeptide. The calcium responses observed in our experiments clearly show that GH3 cells possess membrane receptors to this important physiological stimulus, GHRF. Thus, it is likely that the lack of secretory response to GHRF previously reported in GH3 cells (8) does not result from the absence of binding sites to the peptide on the surface of these cells but could be related to alterations of transduction mechanisms resulting in uncoupling between stimulation and secretion. Additional experiments will be necessary to further clarify this point.
Endo • 1991 Vol 129 • No 4
Acknowledgments We are indebted to G. Gaurier, F. Bertrand, and D. Varoqueaux for their help in the preparation of the manuscript.
References 1. Bilezikjian LM, Vale WW 1983 Stimulation of adenosine 3',5' monophosphate production by growth hormone-releasing factor and its inhibition by somatostatin in anterior pituitary cells in vitro. Endocrinology 113:1726-1731 2. Brazeau P, Ling N, Esch F, Boehlen P, Mougin C, Guillemin R 1982 Somatocrinin (growth hormone-releasing factor) in vitro bioactivity: Ca2+ involvement, cAMP mediated action and additivity of effect with PGE2. Biochem Biophys Res Commun 190:588594 3. Schofl C, Sandow J, Knepel W, 1987 GRF elevates cytosolic calcium concentration in rat anterior pituitary cells. Am J Physiol 253:E591-E594 4. Holl RW, Thorner MO, Leong DA 1989 Cytosolic free calcium in normal somatotropes: effects of forskolin and phorbol ester. Proc Natl Acad Sci USA 256:E375-E379 5. Lussier BT, French MB, Moor BC, Kraicer J 1991 Free intracellular Ca2+ concentration ([Ca2+]i) and growth hormone release from purified rat somatotrophs. I-GH-releasing factor-induced Ca2+ influx raises [Ca2+]i. Endocrinology 128:570-582 6. Gershengorn MC 1986 Mechanism of thyrotropin-releasing hormone stimulation of pituitary hormone secretion. Annu Rev Physiol 48:515-526 7. Ostlund Jr RE, Leung JT, Vaerewyck SH, Winokur T, Melman M 1978 Acute stimulated hormone release from cultured GH3 pituitary cells. Endocrinology 103:1245-1252 8. Zeytin FN, Gick GG, Brazeau P, Ling N, McLaughlin M, Bancroft C 1984 Growth hormone (GH)-releasing factor does not regulate GH release or GH mRNA levels in GH3 cells. Endocrinology 114:2054-2059 9. Mollard P, Guerineau N, Audin J, Dufy B 1989 Measurement of Ca2+ transients using simultaneous dual-emission microspectrofluorimetry and electrophysiology in individual pituitary cells. Biochem Biophys Res Commun 164:1045-1052 10. Grynkiewicz G, Poenie M, Tsien RY 1985 A new generation of Ca2+ indicators with greatly improved fluorescent properties. J Biol Chem 260:3440-3450 11. Sartor P, Dufy-Barbe L, Corcuff JB, Taupignon A, Dufy B 1990 Electrophysiological response to thyrotropin-releasing hormone of rat lactotrophs in primary culture. Am J Physiol 258:E311-E319 12. Winiger BP, Schlegel W 1988 Rapid transient elevations of cytosolic calcium triggered by thyrotropin releasing hormone in individual cells of the pituitary line GH3/B6. Biochem J 255:161-167 13. Holl HW, Thorner O, Leong DA 1988 Intracellular calcium concentration and growth hormone secretion in individual somatotropes: effects of growth-hormone releasing factor and somatostatin. Endocrinology 122:2927-2932 14. Chen C, Israel JM, Vincent JD 1989 Electrophysiological responses of rat pituitary cells in somatotroph-enriched primary culture to human growth hormone releasing factor. Neuroendocrinology 50:679-687 15. Gershengorn MC, Thaw C 1985 Thyrotropin releasing-hormone (TRH) stimulates biphasic elevation of cytoplasmic free calcium in GH3 cells. Further evidence that TRH mobilizes cellular and extracellular Ca2+. Endocrinology 116:591-596 16. Mollard P, Dufy B, Barker JL, Schlegel W 1990 Thyrotropinreleasing-hormone activated a [Ca2+]i-dependent K+ current in GH3 pituitary cells via Ins (1,4,5) P3-sensitive and Ins (1,4,5) PSinsensitive mechanisms. Biochem J 268:345-352 17. Schlegel W, Winiger BP, Mollard P, Vacher P, Wuarin F, Zahnd G, Wolheim CB, Dufy B 1987 Oscillations of cytosolic Ca2+ in pituitary cells due to action potentials. Nature 329:719-721
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 18:48 For personal use only. No other uses without permission. . All rights reserved.
