- Email: [email protected]
Three types of cytoplasmic Ca2+ oscillations in stimulated pancreatic β-cells
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 268, No. 1, January, pp. 404-407,1989
COMMUNICATION Three Types of Cytoplasmic
Ca2+ Oscillations in Stimulated Pancreatic ,&Cells’
EVA GRAPENGIESSER, ERIK GYLFE,2 Department
of Medical Cell Biology, Uppsala University,
AND
Biomedicum,
BO HELLMAN
Box 571, S-751 23 Uppsala, Sweden
Received August 181988, and in revised form September 30,1988
Oscillations of cytoplasmic Ca2+ (Cay) involved in cell regulation have recently attracted considerable attention. In the pancreatic P-cells an intermediate concentration of glucose (11 mM) induces large oscillations of Caf+ with periods of 2 to 6 min. Procedures stimulating insulin secretion further, such as raising glucose to 20-30 mM or adding carbachol, ATP, theophylline, glucagon, or forskolin, often changed these oscillations into a steady increase of Caf+. In addition, forskolin and glucagon triggered prominent 9- to 14-s Cai?+spikes during the intervals of increased Ca?+, whereas carbachol and ATP initiated a series of rapid spikes of decreasing magnitude and increasing duration (6-11 s). All types of oscillations depended on the presence of extracellular Caz+, but carbachol and ATP also induced single Cay transients in the absence of the cation. The results demonstrate hitherto unknown oscillations of Ca? in the pancreatic P-cell which are dependent in different ways on Ca2+entry. 0 1989 Academic Press, Inc.
Using time-sharing dual-wavelength fluorometry it was recently possible to verify the prediction (1) that glucose induces variations in Caf+ regulating insulin release with initial lowering dependent on intracellular sequestration and outward transport of Ca*+ followed by a rise due to influx of the ion (2). When combined with epifluorescence microscope fluorometry the time-sharing approach enabled the demonstration of glucose-induced oscillations of CaZ+ in individual p-cells (3). Such a phenomenon has been proposed on the basis of electrophysiological observations with submaximally stimulating glucose concentrations (1, 4-6). We have now characterized the glucose-induced oscillations further and studied the effects of other insulin secretagogs with known effects on the electrical activity of the fi-cells. The results indicate that there are at least three different types of Cay oscillations in stimulated pancreatic pcells.
i This work was supported by grants from the Swedish Medical Research Council (12X-562, 12X-62403, the Swedish Diabetes Association, the Nordic Insulin Foundation, Aage Louis-Hansens Memorial Foundation, the Swedish Hoechst and NOVO Companies, and the Family Ernfors Foundation. * To whom correspondence should be addressed. 0003-9861/89 $3.00 Copyright All rights
0 1989 by Academic Press, Inc. of reproduction in any form reserved
EXPERIMENTAL
PROCEDURES
The procedures and equipment employed have been described (3). Single cells were prepared from islets of Langerhans isolated from 3-month-old rats or lomonth-old ob/ob mice. The cells were allowed to attach to circular 25-mm cover glasses during 8-48 h in RPM1 1640 medium. Further experimental handling was performed with a basal medium physiologically balanced in cations with Cll as the sole anion (7). The cells were loaded with fura- during 40 min of incubation in 0.5 PM fura- acetoxymethylester (Molecular Probes). The cover glasses with the loaded cells were used as the bottom of an open chamber connected to a peristaltic pump allowing superfusion of a 2.5-mm medium layer at a rate of 1.0 ml/min. The chamber was placed on the stage of an inverted microscope (Leitz Diavert) within a climate box maintained at 37°C. The microscope was equipped for epifluorescence fluorometry with a 100X uv-fluorite objective (Nikon). A 75-W xenon arc lamp combined with 5- or lo-nm half-bandwidth (HBW)a interference filters in an air-turbine time-sharing multichannel spectro-
3 Abbreviations used: HBW, half-bandwidth; inositol1,4,5-trisphosphate. 404
IPa,
CYTOPLASMIC
Ca2+ OSCILLATIONS
IN PANCREATIC
405
B-CELLS RESULTS
TIME
(MIN)
FIG. 1. Glucose-induced oscillations of Cay in individual mouse p-cells. The 340/380 nm fluorescence excitation ratios of the indicator furaare shown to the right and the corresponding Cat’ values (nM) to the left. Each panel shows one typical experiment of four. (A) Effect of raising glucose from 3 to 11 mM (arrow) and further to 30 mM (bar) followed by return to 3 mM. (B) Effect of raising the Ca2+ concentration from 1.28 to 3.84 mM (bar) with 11 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 11 mM. (C)Effect of 1OpM methoxyverapamil (bar) with 11 mM glucose.
photofluorometer (8) provided I-ms excitation light flashes at 340 and 380 nm every 10 ms. Emission was measured with a photomultiplier at 510 nm using a 30-nm HBW filter. The background fluorescence from cells not containing fura- was negligible. An analog ratio meter allowed recording of the 3401380 nm fluorescence excitation ratio. Calibration was performed as previously described (3,9) using the equation,
Caf+= K,P(R - R,dWL,,
The appearance of Caf’ oscillations with 11 mM glucose and their transition into continuously increased Caf+ with 30 mM of the sugar are shown in Fig. 1A. These oscillations, which had periods of 2-6 min, were not exclusively confined to mouse p-cells, since they also occurred in the rat (not shown). The magnitude of the oscillations increased with rise of the extracellular Caa’ concentration (Fig. 1B) and disappeared in the presence of a blocker of the voltage-dependent Caa+ channels (Fig. IC). Figure 2A shows that addition of 2 mM theophylline to an oscillating P-cell causes a change toward steadily increased Caf’ with only short intervals of lowering. A similar reaction was obtained with 10 nM glucagon (not shown). However, the type of response varied in different cells and with concentration. With 5 mM theophylline the oscillations were sometimes extinguished and Caf’ lowered to a basal value (not shown). Also the Cay response to the adenylate cyclase activator forskolin was variable with a continuous increase only in some cases. A typical experiment is shown in Fig. 2B. It is evident that 5 PM forskolin extends the intervals at high Caf+ and causes superimposed pronounced spikes lasting for 9-14 s (Fig. 2C). Also a high concentration of glucagon (100 nM) triggered such irregular spikes but only during the initial minutes of exposure
1
A
- RI,
where p = F,,/F, and F. and Rminare the fura- fluorescence at 380 nm and the 3401380 nm fluorescence excitation ratio, respectively, in an “intracellular” K+-rich medium lacking Ca’+. F. and R,,,, are the corresponding data obtained at saturating Caa+ concentrations. With the 5-nm HBW excitation filters /3 was 8.0, R,,, 18.0, and R,i, 0.62, whereas the corresponding values were 8.1, 13.4, and 0.62 with the IO-nm HBW filters (only used in Fig. 2A). The K. employed was 231 nM (10). With the nonlinear Caa+ binding characteristics of the indicator (9) and substantial Ca2+ gradients in the cytoplasm, the fluorescence signal will tend to underestimate average Ca:+ (2). In determining the absolute values for average Caf+ in a stimulated cell this problem is likely to be considerably greater than imperfection of the calibration procedure and the possible presence of partially hydrolyzed fura- ester.
0
1
.-2
1
TIME 0.4IN,
FIG. 2. Modifications of glucose-induced oscillations of Caz+ by agents affecting CAMP. The 340/380 nm fluorescence excitation ratios of the indicator furaare shown to the right and the corresponding Ca:+ values (nM) to the left. Each panel shows one typical experiment of four. (A) Effect of 2 mM theophylline (bar) with 11 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 11 mM. (B) Effect of 5 PM forskolin (upper bar) with 11 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 11 mM. (C) The section in B indicated by the lower capped bar is shown on an expanded time scale.
406
GRAPENGIESSER,
GYLFE.
TlME 0.4INl
FIG. 3. Modifications of glucose-induced oscillations of Cap by agents stimulating IPaformation. The 340/ 380 nm fluorescence excitation ratios of the indicator fura- are shown to the right and the corresponding Ca’+ values (nM) to the left. (A) Effect of 1 @M ATP (upper bar) with 20 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 20 mM. This response is representative for 8 of 12 studied cells. (B) The section in A indicated by the lower capped bar is shown on an expanded time scale. (C) The initial effect of 10 pM carbachol with 20 mM glucose. This response is representative for 10 of 12 studied cells.
(not shown). The forskolin-induced spikes occurred only when Cat’ was increased and disappeared in the presence of 10 GM methoxyverapamil or after omission of extracellular Cazf (not shown). When the effects of earbachol and ATP were tested on oscillating fl-cells, both agents promptly induced five to seven Caz+ spikes of decreasing magnitude with durations increasing from 6 to 11 s (Figs. 3A3C). These spikes were obtained irrespectively of whether Cap was high or low before the introduction of the drug and were often followed by continuously increased Caf’ . When 10 PM methoxyverapamil was present, carbachol and ATP triggered only one to three Cav spikes. After omission of extracellular Cazf there were only single spikes (not shown). DISCUSSION Like the bursts of membrane depolarization (11, 12), the amplitude of the glucose-induced oscillations of Cav increased with rise of the extracellular Ca2+ concentration and disappeared in the presence of a blocker of the voltage-dependent Caz+channels. However, in contrast to the burst pattern, which is transformed into continuous depolarization with glucose concentrations stimulating secretion maximally (12), the corresponding change to a steady increase of Ca: was not an obligatory phenomenon (e.g., Fig. 3). More
AND
HELLMAN
important, each oscillation required 2-6 min, whereas the bursts of depolarization are in most cases considerably shorter than 2 min. Since variations in membrane potential were analyzed in electrically coupled P-cells, this discrepancy was first taken to indicate that the depolarization bursts are considerably longer in an individual cell (3). An alternative interpretation is that the Ca? oscillations reflect cyclic variations in the duration of bursts. With lo-11 mM glucose, some P-cells (20%) have been found to exhibit such variations with a period ranging from 3 to 6 min (13,14). However, it is questionable whether the variations in burst duration are sufficiently pronounced to explain the amplitude of the Gaff oscillations with minima often corresponding to basal Ca?+ A rise oflthe glucose concentration is not a prerequisite for the transition from a burst pattern into a more continuous depolarization. Agents increasing CAMP, like theophylline and forskolin, or the membrane-permeant dibutyryl CAMP have been found to prolong the bursts (15, 16), prevent cyclic variations in burst duration (17), or induce continuous depolarization at the burst plateau (16). It was apparent from the present data that addition of 2 mM theophylline or 10 nM of the adenylate cyclase activating agonist glucagon to an oscillating P-cell causes a corresponding change toward steadily increased Caf+ . Also the observation that 5 mM theophylline sometimes extinguished the oscillations and lowered Ca? to a basal value is consistent with electrophysiological observations. High concentrations of this phosphodiesterase inhibitor have thus been found to cause a paradoxical hyperpolarization (16). The different Caf’ response to the adenylate cyclase activator forskolin with pronounced 9- to 14-s spikes may result from an excess of CAMP since such spikes were also observed as an initial response to a high concentration of glucagon. However, forskolin also has an additional effect on the p-cells in mediating depolarization by blockage of delayed rectifying K+ channels (18). Muscarinic receptor activation has been found to increase the burst frequency in glucose-stimulated @cells (19, 20), to decrease the repolarization between the bursts (19-21), or to induce continuous depolarization at the plateau potential (19,21). Extracellular ATP mimics muscarinic receptor activation in stimulating the /?-cell formation of Ca’+-mobilizing inositol 1,4,5-trisphosphate (IP,) by binding to P,-purinoceptors (22-25). It is therefore not surprising that carbachol and ATP had almost identical actions on Caf+ in replacing the glucose-induced oscillations with continuously increased Gaff In the pancreatic p-cells carbachol and ATP have similar effects as IP3 in mobilizing calcium taken up into intracellular stores in response to glucose (22-25). Considering that carbachol and ATP induced single Caf’ spikes even in the absence of external Ca’+, it can be anticipated that intracellular Ca2+ is involved at least in the initiation of the response.
CYTOPLASMIC
Ca2+ OSCILLATIONS
Oscillations of Cay with different frequencies have been observed in a number of stimulated cells (2630). In some cases the phenomenon continues after removal of extracellular Ca’+. It has even been proposed that cellular processes may be frequency-modulated rather than amplitude-modulated by Ca? (27). The oscillations observed in pancreatic p-cells were of three distinct types. The slow 2- to 6-min cycles induced by glucose exhibited a strict dependence on extracellular Ca2+and resembled in many respects reported variations in the membrane potential (11,12). Also the 9- to 14-s Cay spikes invoked by forskolin probably reflect Ca2+ influx, since no effect was obtained in the presence of a blocker of the voltage-dependent Ca2+channels or when the medium was depleted of Ca’+. It is unclear whether influx of Ca’+ is directly involved also in the initial oscillations caused by carbachol and ATP, or if extracellular Ca2+is simply required to allow rapid refilling of the intracellular pool involved. The oscillations induced by carbachol, ATP, and glucagon were initial phenomena. Only forskolin generated a sustained Caf+ spike activity. Considering that the procedures stimulating secretion often led to transformation of oscillations into persistent increase of Caf+ , it seems unlikely that frequency-modulation represents a general mechanism for Cay control of insulin release. During the editorial handling of the present communication data about the actions of carbachol on individual cells of the clonal pancreatic P-cell line HIT were published (30). The effect of 200 PM carbachol on Ca’+ of these tumor cells was similar to that obtained in the present study with only 10 PM of the agonist, suggesting a lower receptor density on the HIT cells.
IN PANCREATIC
407
P-CELLS
G., POENIE, M., AND TSIEN, R. Y. (1985) J. Biol. Chem. 260,3440-3450.
9. GRYNKIEWICZ,
10. JOHANSSON, H., LARSSON, R., NYGREN, P., LINDH, E., RASTAD, J., AKERSTR~M, G., AND GYLFE, E.
11.
(1987) Biosci. Rep. 7,705-712. E. K., AND SAKAMOTO,
MATTHEWS,
Y. (1975)
J.
Physiol. 246,421-437.
12.
MEISSNER,
H. P., PREISSLER,
M., AND HENQUIN,
J. C. (1979) inDiabetes 1979 (Waldhlusl, W. K., Ed.), pp. 166-171, Excerpta Medica, Amsterdam. 13.
HENQUIN, J. C., MEISSNER, H. P., AND SCHMEER, W. (1982) Eur. J. Physiol. 393,322-327.
14. COOK,D. L. (1983) Metabolism 32,681-685. 15. HENQUIN, J. C., SCHMEER, W., AND MEISSNER, H. P. (1983) Endocrinology 112,2218-2220. 16. HENQUIN, J. C., AND MEISSNER, H. P. (1984) J. Physiol. 351,595-612. 17. HENQUIN, J. C., SCHMEER, W., NENQUIN, M., AND MEISSNER,
H. P. (1984)
B&hem.
Biophys. Res.
Cwnmun. 120,797-803. 18. ZUNKLER, B. J., TRUBE, G., AND OHNO-SHOSAKU, T. (1988) PJiigers Arch. 411,613-619. 19. COOK, D. L., CRILL, W. E., AND (1981) Diabetes 30,558-561. 20. HERMANS, M. P., SCHMEER, J. C. (1987) Endocrinology
21.
PORTE,
D., JR.
W., AND HENQUIN,
120,1765-1773.
GAGERMAN, E., IDAHL, L. A., MEISSNER, H. P., AND T;~LJEDAL, I. B. (1978) Amer. J. Physiol.
235, E493-E500. 22. HELLMAN,
B., AND GYLFE,
E. (1986)
Pharmacol-
ogy 32,257-267.
23.
GYLFE,
E., AND HELLMAN,
B. (1986) B&hem.
J
233,865-870. 24.
HELLMAN,
B., GYLFE,
E., AND WESSLEN,
N. (1986)
Biochem. Int. 13,383-389. 25. HELLMAN, B., GYLFE, E., WESSLEN, N., HALLBERG, A., GRAPENGIESSER, E., AND MARCSTRBM, A. (1989) E3cp. Clin. Endocrinol., in
REFERENCES 1. HELLMAN, B., AND GYLFE, E. (1986) in Calcium and Cell Function (Cheung, W. Y., Ed.), Vol. 6, pp. 253-326, Academic, Orlando. 2. GYLFE, E. (1988) J. Biol. Chem. 263,5044-5048. 3. GRAPENGIESSER, E., GYLFE, E., AND HELLMAN, B. (1988) Biochem.
Biophys.
Res. Commun.
151,
1299-1304. 4. CHAY, T. R., AND KEIZER, J. (1983) Biophys. J. 42, 181-190. 5. CHAY, T. R., AND KEIZER, J. (1985) Biophys. J. 48, 815-827. 6. PEREZ-ARMENDARIZ,E. M., AND ATWATER, I. (1986) in Biophysics of the Pancreatic P-Cell (Atwater, I., Rojas, E., and Soria, B., Eds.), pp. 31-51, Plenum, New York. ‘7. HELLMAN, B. (1975) Endocrinology 97,392-398. 8. CHANCE, B., LEGALLAIS, V., SORGE, J., AND GRAHAM, N. (1975) Anal. Biochem. 66,498-514.
press. 26. COBBOLD, P. H., WOODS, N. M.
CUTHBERTSON,
K.
S. R., AND
(1987) in Inositol Lipids in Cellular Signalling (Michell, R. H., and Putney, J. W., Jr., Eds.), pp. 119-124, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
27. BERRIDGE,
M. J. (1987)
Annu. Rev. Biochem. 56,
159-193. 28. SCHLEGEL, W., WINIGER, B. P., MOLLARD, P., VACHER, P., WUARIN, F., ZAHND, G. R., WOLLHEIM, C. B., AND DUFY, B. (1987) Nature (Lm-
don) 329,719-721. 29. AMBLER, S. K., POENIE, M., TSIEN, R., AND TAYLOR, P. (1988) J. Biol. Chem. 263,1952-1959. 30. PRENTKI, M., GLENNON, M. C., THOMAS, A. P., MORRIS, R. L., MATSCHINSKY, F. M., AND CORKEY, B. E. (1988) J. Biol. Chem. 263, 11044-
11047.
COMMUNICATION Three Types of Cytoplasmic
Ca2+ Oscillations in Stimulated Pancreatic ,&Cells’
EVA GRAPENGIESSER, ERIK GYLFE,2 Department
of Medical Cell Biology, Uppsala University,
AND
Biomedicum,
BO HELLMAN
Box 571, S-751 23 Uppsala, Sweden
Received August 181988, and in revised form September 30,1988
Oscillations of cytoplasmic Ca2+ (Cay) involved in cell regulation have recently attracted considerable attention. In the pancreatic P-cells an intermediate concentration of glucose (11 mM) induces large oscillations of Caf+ with periods of 2 to 6 min. Procedures stimulating insulin secretion further, such as raising glucose to 20-30 mM or adding carbachol, ATP, theophylline, glucagon, or forskolin, often changed these oscillations into a steady increase of Caf+. In addition, forskolin and glucagon triggered prominent 9- to 14-s Cai?+spikes during the intervals of increased Ca?+, whereas carbachol and ATP initiated a series of rapid spikes of decreasing magnitude and increasing duration (6-11 s). All types of oscillations depended on the presence of extracellular Caz+, but carbachol and ATP also induced single Cay transients in the absence of the cation. The results demonstrate hitherto unknown oscillations of Ca? in the pancreatic P-cell which are dependent in different ways on Ca2+entry. 0 1989 Academic Press, Inc.
Using time-sharing dual-wavelength fluorometry it was recently possible to verify the prediction (1) that glucose induces variations in Caf+ regulating insulin release with initial lowering dependent on intracellular sequestration and outward transport of Ca*+ followed by a rise due to influx of the ion (2). When combined with epifluorescence microscope fluorometry the time-sharing approach enabled the demonstration of glucose-induced oscillations of CaZ+ in individual p-cells (3). Such a phenomenon has been proposed on the basis of electrophysiological observations with submaximally stimulating glucose concentrations (1, 4-6). We have now characterized the glucose-induced oscillations further and studied the effects of other insulin secretagogs with known effects on the electrical activity of the fi-cells. The results indicate that there are at least three different types of Cay oscillations in stimulated pancreatic pcells.
i This work was supported by grants from the Swedish Medical Research Council (12X-562, 12X-62403, the Swedish Diabetes Association, the Nordic Insulin Foundation, Aage Louis-Hansens Memorial Foundation, the Swedish Hoechst and NOVO Companies, and the Family Ernfors Foundation. * To whom correspondence should be addressed. 0003-9861/89 $3.00 Copyright All rights
0 1989 by Academic Press, Inc. of reproduction in any form reserved
EXPERIMENTAL
PROCEDURES
The procedures and equipment employed have been described (3). Single cells were prepared from islets of Langerhans isolated from 3-month-old rats or lomonth-old ob/ob mice. The cells were allowed to attach to circular 25-mm cover glasses during 8-48 h in RPM1 1640 medium. Further experimental handling was performed with a basal medium physiologically balanced in cations with Cll as the sole anion (7). The cells were loaded with fura- during 40 min of incubation in 0.5 PM fura- acetoxymethylester (Molecular Probes). The cover glasses with the loaded cells were used as the bottom of an open chamber connected to a peristaltic pump allowing superfusion of a 2.5-mm medium layer at a rate of 1.0 ml/min. The chamber was placed on the stage of an inverted microscope (Leitz Diavert) within a climate box maintained at 37°C. The microscope was equipped for epifluorescence fluorometry with a 100X uv-fluorite objective (Nikon). A 75-W xenon arc lamp combined with 5- or lo-nm half-bandwidth (HBW)a interference filters in an air-turbine time-sharing multichannel spectro-
3 Abbreviations used: HBW, half-bandwidth; inositol1,4,5-trisphosphate. 404
IPa,
CYTOPLASMIC
Ca2+ OSCILLATIONS
IN PANCREATIC
405
B-CELLS RESULTS
TIME
(MIN)
FIG. 1. Glucose-induced oscillations of Cay in individual mouse p-cells. The 340/380 nm fluorescence excitation ratios of the indicator furaare shown to the right and the corresponding Cat’ values (nM) to the left. Each panel shows one typical experiment of four. (A) Effect of raising glucose from 3 to 11 mM (arrow) and further to 30 mM (bar) followed by return to 3 mM. (B) Effect of raising the Ca2+ concentration from 1.28 to 3.84 mM (bar) with 11 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 11 mM. (C)Effect of 1OpM methoxyverapamil (bar) with 11 mM glucose.
photofluorometer (8) provided I-ms excitation light flashes at 340 and 380 nm every 10 ms. Emission was measured with a photomultiplier at 510 nm using a 30-nm HBW filter. The background fluorescence from cells not containing fura- was negligible. An analog ratio meter allowed recording of the 3401380 nm fluorescence excitation ratio. Calibration was performed as previously described (3,9) using the equation,
Caf+= K,P(R - R,dWL,,
The appearance of Caf’ oscillations with 11 mM glucose and their transition into continuously increased Caf+ with 30 mM of the sugar are shown in Fig. 1A. These oscillations, which had periods of 2-6 min, were not exclusively confined to mouse p-cells, since they also occurred in the rat (not shown). The magnitude of the oscillations increased with rise of the extracellular Caa’ concentration (Fig. 1B) and disappeared in the presence of a blocker of the voltage-dependent Caa+ channels (Fig. IC). Figure 2A shows that addition of 2 mM theophylline to an oscillating P-cell causes a change toward steadily increased Caf’ with only short intervals of lowering. A similar reaction was obtained with 10 nM glucagon (not shown). However, the type of response varied in different cells and with concentration. With 5 mM theophylline the oscillations were sometimes extinguished and Caf’ lowered to a basal value (not shown). Also the Cay response to the adenylate cyclase activator forskolin was variable with a continuous increase only in some cases. A typical experiment is shown in Fig. 2B. It is evident that 5 PM forskolin extends the intervals at high Caf+ and causes superimposed pronounced spikes lasting for 9-14 s (Fig. 2C). Also a high concentration of glucagon (100 nM) triggered such irregular spikes but only during the initial minutes of exposure
1
A
- RI,
where p = F,,/F, and F. and Rminare the fura- fluorescence at 380 nm and the 3401380 nm fluorescence excitation ratio, respectively, in an “intracellular” K+-rich medium lacking Ca’+. F. and R,,,, are the corresponding data obtained at saturating Caa+ concentrations. With the 5-nm HBW excitation filters /3 was 8.0, R,,, 18.0, and R,i, 0.62, whereas the corresponding values were 8.1, 13.4, and 0.62 with the IO-nm HBW filters (only used in Fig. 2A). The K. employed was 231 nM (10). With the nonlinear Caa+ binding characteristics of the indicator (9) and substantial Ca2+ gradients in the cytoplasm, the fluorescence signal will tend to underestimate average Ca:+ (2). In determining the absolute values for average Caf+ in a stimulated cell this problem is likely to be considerably greater than imperfection of the calibration procedure and the possible presence of partially hydrolyzed fura- ester.
0
1
.-2
1
TIME 0.4IN,
FIG. 2. Modifications of glucose-induced oscillations of Caz+ by agents affecting CAMP. The 340/380 nm fluorescence excitation ratios of the indicator furaare shown to the right and the corresponding Ca:+ values (nM) to the left. Each panel shows one typical experiment of four. (A) Effect of 2 mM theophylline (bar) with 11 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 11 mM. (B) Effect of 5 PM forskolin (upper bar) with 11 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 11 mM. (C) The section in B indicated by the lower capped bar is shown on an expanded time scale.
406
GRAPENGIESSER,
GYLFE.
TlME 0.4INl
FIG. 3. Modifications of glucose-induced oscillations of Cap by agents stimulating IPaformation. The 340/ 380 nm fluorescence excitation ratios of the indicator fura- are shown to the right and the corresponding Ca’+ values (nM) to the left. (A) Effect of 1 @M ATP (upper bar) with 20 mM glucose. The arrow indicates the increase of the glucose concentration from 3 to 20 mM. This response is representative for 8 of 12 studied cells. (B) The section in A indicated by the lower capped bar is shown on an expanded time scale. (C) The initial effect of 10 pM carbachol with 20 mM glucose. This response is representative for 10 of 12 studied cells.
(not shown). The forskolin-induced spikes occurred only when Cat’ was increased and disappeared in the presence of 10 GM methoxyverapamil or after omission of extracellular Cazf (not shown). When the effects of earbachol and ATP were tested on oscillating fl-cells, both agents promptly induced five to seven Caz+ spikes of decreasing magnitude with durations increasing from 6 to 11 s (Figs. 3A3C). These spikes were obtained irrespectively of whether Cap was high or low before the introduction of the drug and were often followed by continuously increased Caf’ . When 10 PM methoxyverapamil was present, carbachol and ATP triggered only one to three Cav spikes. After omission of extracellular Cazf there were only single spikes (not shown). DISCUSSION Like the bursts of membrane depolarization (11, 12), the amplitude of the glucose-induced oscillations of Cav increased with rise of the extracellular Ca2+ concentration and disappeared in the presence of a blocker of the voltage-dependent Caz+channels. However, in contrast to the burst pattern, which is transformed into continuous depolarization with glucose concentrations stimulating secretion maximally (12), the corresponding change to a steady increase of Ca: was not an obligatory phenomenon (e.g., Fig. 3). More
AND
HELLMAN
important, each oscillation required 2-6 min, whereas the bursts of depolarization are in most cases considerably shorter than 2 min. Since variations in membrane potential were analyzed in electrically coupled P-cells, this discrepancy was first taken to indicate that the depolarization bursts are considerably longer in an individual cell (3). An alternative interpretation is that the Ca? oscillations reflect cyclic variations in the duration of bursts. With lo-11 mM glucose, some P-cells (20%) have been found to exhibit such variations with a period ranging from 3 to 6 min (13,14). However, it is questionable whether the variations in burst duration are sufficiently pronounced to explain the amplitude of the Gaff oscillations with minima often corresponding to basal Ca?+ A rise oflthe glucose concentration is not a prerequisite for the transition from a burst pattern into a more continuous depolarization. Agents increasing CAMP, like theophylline and forskolin, or the membrane-permeant dibutyryl CAMP have been found to prolong the bursts (15, 16), prevent cyclic variations in burst duration (17), or induce continuous depolarization at the burst plateau (16). It was apparent from the present data that addition of 2 mM theophylline or 10 nM of the adenylate cyclase activating agonist glucagon to an oscillating P-cell causes a corresponding change toward steadily increased Caf+ . Also the observation that 5 mM theophylline sometimes extinguished the oscillations and lowered Ca? to a basal value is consistent with electrophysiological observations. High concentrations of this phosphodiesterase inhibitor have thus been found to cause a paradoxical hyperpolarization (16). The different Caf’ response to the adenylate cyclase activator forskolin with pronounced 9- to 14-s spikes may result from an excess of CAMP since such spikes were also observed as an initial response to a high concentration of glucagon. However, forskolin also has an additional effect on the p-cells in mediating depolarization by blockage of delayed rectifying K+ channels (18). Muscarinic receptor activation has been found to increase the burst frequency in glucose-stimulated @cells (19, 20), to decrease the repolarization between the bursts (19-21), or to induce continuous depolarization at the plateau potential (19,21). Extracellular ATP mimics muscarinic receptor activation in stimulating the /?-cell formation of Ca’+-mobilizing inositol 1,4,5-trisphosphate (IP,) by binding to P,-purinoceptors (22-25). It is therefore not surprising that carbachol and ATP had almost identical actions on Caf+ in replacing the glucose-induced oscillations with continuously increased Gaff In the pancreatic p-cells carbachol and ATP have similar effects as IP3 in mobilizing calcium taken up into intracellular stores in response to glucose (22-25). Considering that carbachol and ATP induced single Caf’ spikes even in the absence of external Ca’+, it can be anticipated that intracellular Ca2+ is involved at least in the initiation of the response.
CYTOPLASMIC
Ca2+ OSCILLATIONS
Oscillations of Cay with different frequencies have been observed in a number of stimulated cells (2630). In some cases the phenomenon continues after removal of extracellular Ca’+. It has even been proposed that cellular processes may be frequency-modulated rather than amplitude-modulated by Ca? (27). The oscillations observed in pancreatic p-cells were of three distinct types. The slow 2- to 6-min cycles induced by glucose exhibited a strict dependence on extracellular Ca2+and resembled in many respects reported variations in the membrane potential (11,12). Also the 9- to 14-s Cay spikes invoked by forskolin probably reflect Ca2+ influx, since no effect was obtained in the presence of a blocker of the voltage-dependent Ca2+channels or when the medium was depleted of Ca’+. It is unclear whether influx of Ca’+ is directly involved also in the initial oscillations caused by carbachol and ATP, or if extracellular Ca2+is simply required to allow rapid refilling of the intracellular pool involved. The oscillations induced by carbachol, ATP, and glucagon were initial phenomena. Only forskolin generated a sustained Caf+ spike activity. Considering that the procedures stimulating secretion often led to transformation of oscillations into persistent increase of Caf+ , it seems unlikely that frequency-modulation represents a general mechanism for Cay control of insulin release. During the editorial handling of the present communication data about the actions of carbachol on individual cells of the clonal pancreatic P-cell line HIT were published (30). The effect of 200 PM carbachol on Ca’+ of these tumor cells was similar to that obtained in the present study with only 10 PM of the agonist, suggesting a lower receptor density on the HIT cells.
IN PANCREATIC
407
P-CELLS
G., POENIE, M., AND TSIEN, R. Y. (1985) J. Biol. Chem. 260,3440-3450.
9. GRYNKIEWICZ,
10. JOHANSSON, H., LARSSON, R., NYGREN, P., LINDH, E., RASTAD, J., AKERSTR~M, G., AND GYLFE, E.
11.
(1987) Biosci. Rep. 7,705-712. E. K., AND SAKAMOTO,
MATTHEWS,
Y. (1975)
J.
Physiol. 246,421-437.
12.
MEISSNER,
H. P., PREISSLER,
M., AND HENQUIN,
J. C. (1979) inDiabetes 1979 (Waldhlusl, W. K., Ed.), pp. 166-171, Excerpta Medica, Amsterdam. 13.
HENQUIN, J. C., MEISSNER, H. P., AND SCHMEER, W. (1982) Eur. J. Physiol. 393,322-327.
14. COOK,D. L. (1983) Metabolism 32,681-685. 15. HENQUIN, J. C., SCHMEER, W., AND MEISSNER, H. P. (1983) Endocrinology 112,2218-2220. 16. HENQUIN, J. C., AND MEISSNER, H. P. (1984) J. Physiol. 351,595-612. 17. HENQUIN, J. C., SCHMEER, W., NENQUIN, M., AND MEISSNER,
H. P. (1984)
B&hem.
Biophys. Res.
Cwnmun. 120,797-803. 18. ZUNKLER, B. J., TRUBE, G., AND OHNO-SHOSAKU, T. (1988) PJiigers Arch. 411,613-619. 19. COOK, D. L., CRILL, W. E., AND (1981) Diabetes 30,558-561. 20. HERMANS, M. P., SCHMEER, J. C. (1987) Endocrinology
21.
PORTE,
D., JR.
W., AND HENQUIN,
120,1765-1773.
GAGERMAN, E., IDAHL, L. A., MEISSNER, H. P., AND T;~LJEDAL, I. B. (1978) Amer. J. Physiol.
235, E493-E500. 22. HELLMAN,
B., AND GYLFE,
E. (1986)
Pharmacol-
ogy 32,257-267.
23.
GYLFE,
E., AND HELLMAN,
B. (1986) B&hem.
J
233,865-870. 24.
HELLMAN,
B., GYLFE,
E., AND WESSLEN,
N. (1986)
Biochem. Int. 13,383-389. 25. HELLMAN, B., GYLFE, E., WESSLEN, N., HALLBERG, A., GRAPENGIESSER, E., AND MARCSTRBM, A. (1989) E3cp. Clin. Endocrinol., in
REFERENCES 1. HELLMAN, B., AND GYLFE, E. (1986) in Calcium and Cell Function (Cheung, W. Y., Ed.), Vol. 6, pp. 253-326, Academic, Orlando. 2. GYLFE, E. (1988) J. Biol. Chem. 263,5044-5048. 3. GRAPENGIESSER, E., GYLFE, E., AND HELLMAN, B. (1988) Biochem.
Biophys.
Res. Commun.
151,
1299-1304. 4. CHAY, T. R., AND KEIZER, J. (1983) Biophys. J. 42, 181-190. 5. CHAY, T. R., AND KEIZER, J. (1985) Biophys. J. 48, 815-827. 6. PEREZ-ARMENDARIZ,E. M., AND ATWATER, I. (1986) in Biophysics of the Pancreatic P-Cell (Atwater, I., Rojas, E., and Soria, B., Eds.), pp. 31-51, Plenum, New York. ‘7. HELLMAN, B. (1975) Endocrinology 97,392-398. 8. CHANCE, B., LEGALLAIS, V., SORGE, J., AND GRAHAM, N. (1975) Anal. Biochem. 66,498-514.
press. 26. COBBOLD, P. H., WOODS, N. M.
CUTHBERTSON,
K.
S. R., AND
(1987) in Inositol Lipids in Cellular Signalling (Michell, R. H., and Putney, J. W., Jr., Eds.), pp. 119-124, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
27. BERRIDGE,
M. J. (1987)
Annu. Rev. Biochem. 56,
159-193. 28. SCHLEGEL, W., WINIGER, B. P., MOLLARD, P., VACHER, P., WUARIN, F., ZAHND, G. R., WOLLHEIM, C. B., AND DUFY, B. (1987) Nature (Lm-
don) 329,719-721. 29. AMBLER, S. K., POENIE, M., TSIEN, R., AND TAYLOR, P. (1988) J. Biol. Chem. 263,1952-1959. 30. PRENTKI, M., GLENNON, M. C., THOMAS, A. P., MORRIS, R. L., MATSCHINSKY, F. M., AND CORKEY, B. E. (1988) J. Biol. Chem. 263, 11044-
11047.