Glucose-induced mobilisation of intracellular Ca2+ in depolarised pancreatic islets

J Physiology (Paris)

(1998) 92,31-35

0 Elsevier,Paris

Glucose-induced mobilisation of intracellular in depolarised pancreatic islets

Ca2+

Hassan Jijakli, Willy J. Malaisse* Laboratory

of Experimental

Medicine,

Brussels

Free University,

808 Route de L.ennik, B-1070

Brussels,

Belgium

(Received 10 October 1997; accepted 6 February 1998)

- Perifused rat pancreatic islets, prelabelled with 45Ca, were exposed for 90 min to a medium containing 30 mM K+, 0.25 mM diazoxide and 0.5 mM EGTA, but deprived of CaC12. Either verapamil (0.05 mM) or Cd2+ (0.05 mM) were also present in the erifusate. Under these conditions a rise in D-glucose concentrations from either 2.8 to 16.7 mM or zero to 8.3 mM increased both 85Ca outflow and insulin release, after an initial and transient decrease in effluent radioactivity. These findings suggest that, in islets depolarised by exposure to a high extracellular concentration of K+, D-glucose provokes an intracellular redistribution of Ca2+ ions and subsequent stimulation of insulin release. The functional response to D-glucose is apparentll+?ot attributable to either the closing of ATP-sensitive K+ channels, which were actually activated by diazoxide, or stimulation of Ca Influx, which was prevented by the absence of extracellular Ca? The present experimental design thus reveals a novel component of the glucose-induced remodelling of Ca2+ fluxes in islet cells. Such an effect might also be operative under physiological conditions, when the hexose leads to depolarisation of the islet B-cells. (0 Elsevier, Paris). Abstract

pancreatic

islets / insulin

secretion

I Ca*+ fluxes

/ D-glucose

1. Introduction In the absenceof extracellular Ca*+, and even in the presence of EGTA, D-glucose is able to stimulate insulin release from rat pancreatic islets, provided that the incubation medium contains an agent, such as forskolin or theophylline, augmenting the cyclic AMP content of the islet cells [I, 8, 141. The present report reveals that, even in the absence of such agents, D-glucose mobilises intracellular Ca2+ and stimulatesinsulin releasefrom Ca*+-deprived islets when they are perifused in media containing both diazoxide, in order to activate ATP-sensitive K+ channels, and a high concentration of KCl, used to cause, nevertheless, depolarisation of the B-cell plasma membrane [5, 121.

2. Materials and methods All experiments were conducted in pancreatic islets isolated by the collagenase procedure [ 151 from fed female Wis-

tar rats(Proefdierencentrum, Heverlee,Belgium). The methods used to measure 45Ca efflux [4] and insulin output [2,6] from prelabelled perifused islets were previously described in the cited references, the islets being preincubated for 60minat 16.7mM D-glucosein salt-balanced mediacontaining 45Ca and then placed in a perifusion chamber. In all experiments, the concentration of K+ of the perifusion mediumwasraisedto 30mM by equimolarsubstitution of NaCl by KCI. Such a medium also contained diazoxide

*Correspondence

and reprints.

(0.25 mM) and EGTA (0.5 mM), but was deprived of CaC12. Diazoxide was first dissolved in dimethylsulfoxide, the final concentration of the solvent not exceeding 0.25 pL/mL. At this concentration dimethylsulfoxide fails to affect islet function [ 1I]. Verapamil(O.05 mM) orCdCl2 (0.04 mM) were also incorporated into the perifusion medium, as required. All results are presented as mean values (kS.E.M.) together with the number (n) of individual observations. The statistical significance of differences between mean values was assessed by use of Student’s t--test.

3.

Results

All experiments were conducted in islets that had been preincubated for 60 min at 16.7 mM D-glucose in salt-balanced media containing a tracer amount of Wa and were perifused for 90 min in the presenceof 30 mM K+, 0.25 mM diazoxide and 0.5 mM EGTA and absenceof CaC12. In the first series of experiments, the perifusate also contained 0.05 mM verapamil and the concentration of D-glucose was raised from 2.8 to 16.7 mM between the 46th and 70th min of the perifusion period. After 42-45 min of exposure to 2.8 mM D-glucose, the release of insulin averaged 1.13 + 0.08 @J/islet per min (n = S), as distinct (P < 0.001) from 0.31 + 0.04 pU/islet per min (n = 8) when the same experiments were conducted at normal extracellular Ca*+ concentration (1 .O mM) in the absence of EGTA (data not shown). The rise in D-glucose concentration from 2.8 to 16.7 mM provoked a transient

decrease

in insulin

output

(figure

1, lower

32

H. Jijakli, W.J. Malaisse

Glucose

(16.7

mM)

1.8 196

-F 12 .E -2 l,o c 8 Lt. 098 $’

0,6 OS4

02 w 30

Figure 1. Effect of a rise in D-glucose concentration from 2.8 to 16.7 mM (min 46 to 70) on 45Ca fractional outflow rate (upper panel) and insulin release (lower panel) from islets perifused at high extracellular K+ concentration (30 mM) in the presence of diazoxide (0.25 mM), verapamil(O.05 mM) and EGTA (0.5 mM) but absenceof CaC12.Mean values (+ S.E.M.) refer to eight individual experiments.

panel). Thus, after 2.3 + 0.5 min of exposure to Dglucose, the secretory rate reached a nadir value, which corresponded to a mean fall of 114 + 18 nU/islet per min2 below the paired control value recorded just before the switch of the media containing D-glucose in low or high concentration (min 44). Such a fall was much more pronounced (P < 0.001) than that recorded, over a comparable

50

70

90

Time (min)

period of time, up to the 44th min of perifusion (4 k 11 nU/islet per min2). Thereafter, however, the output of insulin progressively increased and eventually reached values in excess of the paired control readings. Over the entire of period of exposure to the high concentration of D-glucose (min 46 to 70), the integrated value for insulin releasewas 293 + 52 nU/islet per min higher (P < 0.001) than the

Intracellular Ca2+ in depolarised pancreatic islets

theoretical value calculated by exponential extrapolation of the data recorded both from min 31 to 45 and from min 84 to 90. When the concentration of D-glucose was returned from 16.7 to 2.8 mM, the output of insulin rapidly decreased, after a transient off-response. In this first series of experiments, the rise in Dglucose concentration only caused minor changes in 4sCa outflow (figure I, upper panel). Except in one out of eight individual experiments, a modest decrease in 45Cafractional outflow rate was first observed. Thus, it reached, after 1.8 f 0.3 min of exposure to 16.7 mM, a nadir value which corresponded to a mean relative fall of 7.6 f 3.4 10-z min-i (n = 7) as distinct (P < 0.07) from a reference value of 1.3 + 0.1 10-2 min-1 (n = S), calculated by exponential extrapolation of the data recorded from min 31 to 45 and min 86 to 90. During prolonged exposure to 16.7 mM D-glucose, however, the outflow of 4sCawas higher than the theoretical value based on an exponential decrease of effluent radioactivity. Indeed, the slope of the regression line defining the progressive fall in 4sCa fractional outflow rate (logarithmic value) averaged 2.39 & 0.28 and 0.81 f 0.60 10-z min-1 (n = 8 in both cases; P c 0.03) respectively before (min 31 to 45) and after (min 45 to 59) raising the concentration of D-glucose. As already noticed in the case of insulin release, a transient off-response in 4sCa outflow took place when the concentration of Dglucose was brought back to its initial value. The second series of experiments differed from the first one in two respects.First, the concentration of D-glucose was raised from zero to 8.3 mM, in which case the stimulation of Ca2+inflow normally evoked by the hexose under physiological conditions is thought to be attributable mainly to the gating of voltage-insensitive Caz+channels [9]. Second, Cdz+ (0.04 mM) was used, instead of verapamil, to selectively block L-type Caz+ channels [16]. After 42-45 min exposure of the islets to the glucose- and Ca2+-deprived medium containing 30 mM K+ and 0.25 mM diazoxide, the output of insulin averaged 0.85 * 0.13 pU/islet per min (n = 4), a value significantly higher (P < 0.02) than that recorded, under comparable experimental conditions, at a close-to-normal concentration of extracellular Caz+ (1.0 mM), i.e., 0.37 + 0.04 @/islet per min (n = 4). The rise in D-glucose concentration from zero to 8.3 mM provoked a rapid and progressive increase in insulin release (figure 2, lower panel). After 15 min of stimulation, the secretory rate was 1.51 f 0.39 l.tU/islet per min higher (P < 0.01) than the mean control value recorded just before the rise in hexose concentration. A transient off-response was again observed when the administration of D-

33

glucose was halted at the 70th min of perifusion. The output of insulin then rapidly returned to a low value. In this second series of experiments, the rise in hexose concentration first provoked a transient decrease in effluent radioactivity. The paired change in %a fractional outflow rate between the 45th and 47th min averaged 0.12 + 0.03 10-Z min-1 (n = 5; P < 0.05), as compared (P < 0.02) to 0.00 + 0.02 10-z min-1 over the preceding 2-min period (min 43 to 45). Over the ensuing 5 min, however, the 4sCa efflux progressively increased, peaking after 6.0 + 0.4 min exposure to D-glucose at a level 0.31 + 0.09 10-z min-i higher (n = 4; P -e 0.05) than the paired nadir value reached shortly after introduction of the hexose. Inversely, when the administration of D-glucose was interrupted, the efflux of 45Ca was first transiently increased and then reached, within 4 min, a level 0.28 f 0.03 1tP min-1 lower (rr = 4; P < 0.005) than the paired value recorded shortly after removal of the hexose from the perifusion medium.

4. Discussion The present results reveal that D-glucose is able to stimulate insulin release from islets deprived of extracellular Ca2+, provided that these are exposed to a high concentration of extracellular K+, in order to cause depolarisation of the B-cell plasma membrane [4]. Such a stimulation of insulin secretion does not occur when the islets are perifused, in the absenceof extracellular Ca2+,at normal K+ concentration [2]. In the present experiments, the stimulation of insulin release cannot be attributed to the closing of K+ATPchannels, since they were fully activated by d&oxide. An alternative explanation might consist in an intracellular redistribution of Ca2+. This process may be favoured by the presence of EGTA in the incubation medium, the intracellular stores of Caz+being probably more readily releasablein islets exposed to the Caz+-chelating agent [20]. However, no stimulation of 45Caefflux occurs in responseto a rise in D-glucose concentration when islets are perifused in the absence of Caz+ and presence of EGTA at normal extracellular K+ concentration [ 131. Hence, the postulated intracellular redistribution of Ca2+ions would appear as a voltage-dependent process. The increasein 4sCaefflux evoked by D-glucose under the present experimental conditions cannot be due to either an increase in Ca2f influx or the gating of L-type Ca2+ channels, the islets being deprived of extracellular Ca2+and either verapamil or CdCl2

34

H. Jijakh, W.J. MaIaisse

Glucose

(8.3

mM)

1.2

Figure 2. Effect of a rise in D-glucose concentrationfrom zero to 8.3 mM (min 46 to 70) on 45Cafractionaloutflow rate(upperpanel)andinsulinrelease (lower panel) from islets perifused at high extracellular K+ concentration (30 mM) in the presenceof diazoxide (0.25 mM), CdC12(0.04 mM) and EGTA (0.5 mM) but absenceof CaC12. Mean values(+ S.E.M.) refer to four individual experiments.

30

being used to block thesevoltage-sensitive channels [9, 161. It is unlikely to be due to an increase of the islet cyclic AMP content, since the absenceof extracellular Ca2+suppressesthe effect of D-glucose upon cyclic AMP formation, as currently attributed, under physiological conditions, to activation of adenylyl-cyclase by Ca2+-calmodulin[ 18, 191. It is conceivable, however, that D-glucose, which is reported to increase Na+ inflow into the islet cells [7], aug-

70

50 Time

90

(min)

ments the intracellular concentration of the monovalent cation in the depolarised islets exposed to EGTA, this leading in turn to the mobilisation of Ca2+ from intracellular storage sites. Although an interference of D-glucose with more distal events in the secretory sequence should not be ruled out, the present findings may reveal a component of the glucose-induced changes in Ca2+ fluxes, that is not detected under physiological con-

Intracellular Ca*+ in depolarised pancreatic islets

ditions. Such a proposal is compatible with the claim that nutrient secretagoguesindeed cause, independently of any change in Ca2+ inflow, an intracellular redistribution of Ca2+, as first documented in islets prelabelled with 4sCa during preincubation in the absenceof D-glucose [lo]. More recently, Roe et al. [17] again proposed that, in mouse islets, the glucose-induced rise in intracellular calcium concentration was not due solely to Ca2+ influx through voltage-dependent calcium channels but also to a voltage-dependent intracellular Ca*+ release. The voltage dependency of the latter process was postulated in the light of its suppression by d&oxide. However, these authors did not investigate whether, in the diazoxide-treated islets, a rise in extracellular K+ concentration could restore the cationic response to D-glucose. Moreover, its secretory consequencewas not examined. In conclusion, therefore, the present experiments draw attention to an often ignored mechanism by which D-glucose, and possibly other nutrients, may increase the cytosolic concentration of Ca2+, i.e., through an intracellular mobilisation of Ca2+. This process, which is not related to stimulation of Ca2+ entry into the islet cells, appears to require depolarisation of the B-cell plasma membraneto be fully expressed and may actually be favoured by an increaseof its permeability as a result of extracellular Ca*+ deprivation.

Acknowledgments This work was supported by a grant from the Belgian Foundation for Scientific Medical Research (3.45 13.94). We are grateful to M. Mahy for technical assistance and C. Demesmaeker for secretarial help.

References [I]

Brisson G.R., Malaisse-Lagae

F., Malaisse W.J., The

stimulus-secretion coupling of glucose-induced insulin release. VII. A proposed site of action for adenosine-3’.5’. cyclic monophosphate, J. Clin. Invest. 51 (1972) 232-241.

35

[5]

Jijakli H., Malaisse W.J., Cationic events stimulated by Dglucose in depolarized islet cells, Cell Signal. 9 (1997) 283290.

[6]

Jijakli H., Ulusoy S., Malaisse W.J., Dissociation between the potency and reversibility two meglitinide analogues, 108.

of the insulinotropic action of Pharmacol. Res. 34 (1996) 105-

[7]

Kawazu S., Boschero A.C., Malaisse W.J., The stimulussecretion coupling of glucose-induced insulin release. XXVIII. Effect of glucose on Na’ fluxes in isolated islets,

[8]

Komatsu M., Schermerhorn T., Aizawa T., Sharp G.W.G., Glucose stimulation of insulin release in the absence of extracellular Ca2+ and in the absence of any increase in intracellular Ca2+ in rat pancreatic islets, Proc. Natl. Acad. Sci. U.S.A. 92 (1995) 10728-10732.

Eur. J. Physiol. 375 (1978) 197-206.

[9]

Lebrun P, Malaisse W.J., Herchuelz A., Evidence for two distinct modalities of Ca*+ Influx into the pancreatic Am. J. Physiol. 242 (1982) E59-E66.

IO]

[ II]

B-cell,

Lebrun P., Malaisse W.J., Herchuelz A., Nutrient-induced intracellular calcium movement in the rat pancreatic B-cell, Am. J. Physiol. 243 (1982) E196-E205. Levy J., Herchuelz A., Sener A., Malaisse-Lagae F., Malaisse W.J., Cytochalasin B-induced impairment of glucose metabolism in islets of Langerhans, Endocrinology 98 (1976) 429-437.

121 Malaisse W.J., Stimulation of insulin release by D-glucose in depolarized pancreatic islets, Cell Signal. 9 (1997) 265268. [I31

Malaisse W.J., Brisson G.R., Baird L.E., The stimulus-secretion coupling of lucose-induced insulin release. X. Effect of glucose on ’ calcium efflux from perifused islets, Am. J. Physiol. 224 (1973) 389-394.

]I41

Malaisse W.J., Garcia-Martinez I?, Dufrane .%I?, Sener Valverde I., Forskolin-induced activation of adenylate clase, cyclic adenosine monophosphate production and sulin release in rat pancreatic islets, Endocrinology (1984) 2015-2020.

A., cyin115

[I51 Malaisse-Lagae

E, Malaisse W.J., Insulin release by pancreatic islets. in: Larner J, Pohl SL (Eds.), Methods in Diabetes Research, John Wiley & Sons, New York, Vol. I, part B, 1984, pp. 147-152. Plasman PO.$ermann M., Herchue$ A., LeFun I?, Sensitivity to Cd but resistance to NI of Ca mflow into rat pancreatic islets, Am. J. Physiol. 258 (1990) E529-E533. Roe M.W., Lancaster M.E., Mertz R.J., Worley III J.F., Dukes I.D., Voltage-dependent intracellular calcium release from mouse islets stimulated by glucose, J. Bio]. Chem. 268 (1993) 9953-9956. Valverde I., Garcia-Martinez P., Ghiglione M., Malaisse W.J., The stimulus-secretion coupling of glucose-induced insulin release. LIII. Calcium-dependency of the cyclic AMP response to nutrient secretagogues, Harm. Metab. Res. 15 (1983) 62-68.

[2]

Herchuelz A., Malaisse W.J., Regulation of calcium fluxes in pancreatic islets. Dissociation between calcium and insulin release, J. Physiol. (Lond.) 283 (1978) 409424.

[3]

Herchuelz A., Malaisse W.J., Regulation of calcium fluxes in pancreatic islets: two calcium movements dissociated response to glucose, Am. J. Physiol. 238 (1980) E87-E95.

[J91 Valverde

[4]

Herchuelz A., Thonnart N., Sener A., Malaisse W.J., Regulation of calcium fluxes in pancreatic islets: the role of membrane depolarization, Endocrinology 107 (1980) 491497.

[201 Worley III I.E. McIntyre

I., Vandermeers A., Ajaneyulu R., Malaisse W.J., Calmodulin activation of adenylate cyclase in pancreatic islets, Science 206 (1979) 225-227. M.S., Spencer B., Mertz R.J., Roe M.W., Dukes I.D., Endoplasmic reticulum calcium store regulates membrane potential in mouse islet B-cells, J. Biol. Chem. 269 (1994) 14359-14362.