Biphasic insulin release in rat islets of Langerhans and the role of Intracellular Ca++ stores

Article

Reference

Biphasic insulin release in rat islets of Langerhans and the role of Intracellular Ca++ stores

KIKUCHI, M, et al.

KIKUCHI, M, et al . Biphasic insulin release in rat islets of Langerhans and the role of Intracellular Ca++ stores. Endocrinology , 1979, vol. 105, no. 4, p. 1013-9

DOI : 10.1210/endo-105-4-1013 PMID : 383467

Available at:

http://archive-ouverte.unige.ch/unige:40548

Disclaimer: layout of this document may differ from the published version.

1 / 1

0013-7227/79/1054-1013$02.00/0 Endocrinology

Copyright© 1979 by The Endocrine Society

Vol. 105, No.4 Printed in U.S.A.

Biphasic Insulin Release in Rat Islets of Langer hans and the Role of Intracellular Ca++ Stores*

MASATOSHI KIKUCHI, CLAES B. WOLLHEIM,t EBERHARD G. SIEGEL, ALBERT E.

RENOLD, AND GEOFFREY W. G. SHARP

Institut de Biochimie Clinique, University of Geneva, Sentier de la Roseraie, 1211 Geneva 4, Switzerland

ABSTRACT. The role of extracellular Ca++ in the development of biphasic insulin release has been studied previously by using Verapamil to block the glucose-induced stimulation of Ca++

uptake. It was found that glucose-stimulated uptake of ca++

from the extracellular fluid was important for the second phase of insulin release but had no role in the first phase. In the current study, the role of intracellular ca++ stores in biphasic insulin release has been studied.

Isolated rat islets were maintained in tissue culture for 46 h with normal (1 mM) or high (5 mM) phosphate concentrations.

The intracellular calcium content of islets maintained in high phosphate was 50 times greater than the content of control islets.

Insulin release was studied by perifusion, and the Ca ++ concen- tration in the medium was decreased to 0.1 mM during stimula- tion with high glucose to minimize the effect of extracellular

S

TIMULATION of insulin release from the B cells of islets of Langerhans by a rapid increase in glucose concentration is characterized by a two-phase response (1-4). Characteristically, a rapid peak of release (first phase) is followed by a period of slowly increasing release (second phase). The mechanism responsible for this bi- phasicity has not been elucidated. It has been suggested that the first phase is due to 1) a small labile pool of insulin-containing granules distinct from the majority of granules in that they are more susceptible to release, perhaps by their sensitivity to fusion with the plasma membrane or by close apposition to structures critical to the release process (5-7); 2) individual B cells that have differential sensitivity to glucose (8); 3) an initial burst of insulin release which then causes a transient feedback inhibition of subsequent release (9, 10); 4) effects such as the concomitant release of somatostatin (11); or 5) a biphasic change in the concentration of an intracellular mediator of release (12). As Ca++ is known to be required for (13, 14) and can alone stimulate (15-20) insulin re- lease, a rise in intracellular Ca ++ concentration is now considered to be a major controlling influence in stimu-

Received September 11, 1978.

• This work was supported by the Swiss National Science Founda- tion (Grant 3.774-0.76SR).

t

To whom requests for reprints should be addressed.

1013

Ca++. The islets maintained in normal phosphate responded with a transient subnormal first phase insulin release. In con- trast, 16.7 mM glucose elicited a 3.5-fold greater first phase release from the ca++_loaded islets. This release was quantita- tively similar to the first phase response of islets perifused with 1 mM ca++ throughout. ca++ loading also increased the rate of second phase insulin release, but this was less marked than that observed for the first phase.

It is concluded that the handling of intracellular ca++ is primarily responsible for the first phase of insulin release, whereas both intracellular ca++ and increased uptake of ca++

from the extracellular compartment are involved in the genesis and development of the second phase. (Endocrinology 106:

1013, 1979)

Ius-secretion coupling in B cells, as suggested previously for other secretory cells (21, 22). Thus, it seems possible that the biphasic insulin release is a reflection of a biphasic change in the concentration of cytosol Ca++ (23).

In considering this possibility, it has been demonstrated that the effect of high glucose to increase the uptake of extracellular ca++ by islets (24-33) is necessary for sec- ond phase release but plays no part in the first phase response (23). If increased cytosol Ca++ is required for insulin release, it follows that intracellular Ca ++ must play the major role in the genesis of first phase release.

To investigate the role of intracellular Ca ++ stores in biphasic release of insulin, the responses to glucose by islets that have had their intracellular stores loaded with ca++ have been compared with those of islets containing normal Ca ++ stores.

Materials and Methods Isolation and ⁴⁵Ca ++ loading of the islets

Pancreatic islets were isolated by the collagenase digestion technique (34) from male Wistar rats, weighing 220-270 g. The islets were washed and maintained for 46 h in batches of 200- 250 in 2 ml Dulbeco's Modified Eagle's Medium (35) containing 10% heat-inactivated calf serum, 14 ^mM NaHC03, 8.3 ^mM glucose, 400 IU /ml sodium penicillin G, 200 !Lg/ml streptomycin sulfate, and 100 ILCilml ⁴⁵Ca++ at a final concentration of 1.05

1014 KIKUCHI ET AL. Endo • 1979 Vol105 • No4

mM. The phosphate concentrations were either 1.0 mM (normal

phosphate medium) or 5.0 mM (high phosphate medium) for the purpose of increasing intracellular Ca++ content (36). A 1 M solution of NaH2P04 was titrated with a 1 M solution of Na2HP04 to pH 7.40. To 50 ml culture medium, which did not contain phosphate, 50 or 250 j.Ll of this phosphate solution were added for normal or high phosphate medium, respectively. The islets were maintained at 37 C and pH 7.4 in 35-mm diameter plastic petri dishes and gassed with an air-C02 mixture.

Measurement

or

^{5Ca ++} content of islets

To measure 45Ca++ content, 10 islets were removed in 20 j.Ll culture medium after 46 h of maintenance. The islets were separated from the radioactive medium by centrifugation for 15 sec at 8000 X g in a Greiner microfuge (type ZF1, Greiner Scientific Corp., New York, NY) through 200 p.l of a mixture of dibutyl- and dinonyl-phthalate (10:3, vol/vol) into 20 j.Ll 6 M urea. The bottom of the tube was cut off above the urea layer and placed in 5 ml Instagel for liquid scintillation spectrometry.

The 45Ca++ content was then calculated by subtracting the 45Ca ++ contained in the extracellular space from the total 45Ca ++

content. Extracellular space was calculated from the distribu- tion space of [³H]sucrose (29). In parallel determinations, the extracellular space was 1.1 ± 0.11 nl/islet (n = 19) for islets maintained at 1 mM phosphate and 2.1 ± 0.25 (n = 18) for islets maintained at 5 mM phosphate. As the islets were in isotopic equilibrium with the medium 45Ca++, the calcium content was expressed as picomoles per islet by calculation from the specific activity of 45Ca++ in the medium (37).

Perifusion of the islets and measurement of insulin release The islets were perifused using 40 islets/chamber, as de- scribed in detail previously (4, 29, 37). The perifusate consisted of Krebs-Ringer bicarbonate buffer (KRB) containing 1.0 mM CaCh (except when stated), 0.5% dialyzed bovine serum albu- min, and 2.8 mM glucose. Phosphate concentrations were either 1.0 mM (normal KRB buffer) or 5 mM (high phosphate KRB buffer). The phosphate salt used was KH2P04, and the concen- tration of KCl in the high phosphate KRB was reduced to maintain a constant final concentration ofK+ and isoosmolarity.

The islets, after calcium loading, were placed directly in the perifusion chamber without washing. From zero time to 46 min, the islets were perifused with the appropriate normal or high phosphate KRB buffer containing 2.8 mM glucose. Unless oth- erwise stated, the islets were then exposed to 16.7 mM glucose, and the stimulation period was continued for another 44 min.

The chamber volume was 70 j.Ll, and the flow rate was 1.2 ml/

min. Changes in CaCh concentration are detailed in the text.

An aliquot of each sample was assayed for immunoreactive insulin by the method of Herbert et al. (38) using rat insulin as standard.

Measurement of islet insulin and glucagon content

Batches of 10 islets were placed in 1 ml acid ethanol (750:

235:15, vol/vol, ethanol-water-concentrated HCl) and left over- night at 4 C. The supernatants were then assayed at appropriate dilutions for immunoreactive insulin and immunoreactive glu-

cagon. Immunoreactive glucagon was measured by the method of Unger et al. (39) using pork glucagon as standard and rabbit antipork serum (serum 30K).

Materials used

All materials used were as described previously (23), except for pork glucagon used as a standard for immunoassay which was obtained from Novo Research Institute (Bagsvaerd, Den- mark) and glucagon antiserum 30K (obtained from Dr. R. H. Unger, University of Texas Southwestern Medical School, Dal- las, TX).

Statistics

Analyses were performed with Student's t test for unpaired data, except when stated otherwise. Values are expressed as the mean± SEM.

Results Islet Ca content

The time course of ⁴⁵Ca accumulation in islets main- tained in the presence of 1 or 5 mM phosphate was studied. Total exchangeable calcium was measured 1, 4, 24, 48, and 72 h after the addition of ⁴⁵Ca ++. In islets maintained in 1 mM phosphate (controls), the ⁴⁵Ca con- tent attained its maximum at 4 h, suggesting that the

45Ca in the islets was in equilibrium with the isotope in the medium. The ⁴⁵Ca content was taken to represent total Ca content and was 13.2 ± 2.2 pmol/islet (n

=

8) at 4 h. The ⁴⁵Ca content of islets maintained at 5 mM

phosphate was 100 ± 20% of paired control islets at 1 h, 105 ± 18% at 4 h, 380 ± 46% at 24 h, 1908 ± 38 at 48 h, and 9560 ± 1338% at 72 h (n = 6-8). Although definite documentation of equilibration has to await measure- ment of tissue specific activity, the results suggest that the high phosphate concentration does not change the time of equilibration (4 h) with the isotope. However, the high phosphate concentration markedly increased the Ca content beyond 4 h, and the net gain of Ca had not even reached its maximum at 48 h. The time course of Ca accumulation was similar when the phosphate concen- tration in the medium was increased from 1 to 5 mM 24 h after the addition of ⁴⁵Ca++, at which time the accu- mulation of ⁴⁵Ca in the islets had attained its plateau. In a large series of experiments, the Ca content of islets maintained for 46 h in 1 mM phosphate was 11.4 ± 1.0 pmol/islet (n = 31) and 547 ± 65 pmol/islet (n = 28) in islets maintained in 5 mM phosphate. Thus, the intracel- lular Ca content was increased approximately 50-fold.

Insulin release

Both sets of islets were examined for insulin release under basal conditions (2.8 mM glucose) and during stim-

Ca++ STORES IN BIPHASIC INSULIN RELEASE 1015

ulation with high glucose (16.7 mM). Two series of exper- iments were performed, and basal insulin release was assessed by estimating the mean insulin release between 41-46 min. The mean values for the first series (Fig. 1) were 7.2 ± 0.8 pg/islet· min (n =- 17) and 5.9 ± 0.6 (n = 16; P > 0.20) for islets maintained at 1 and 5 mM phos- phate, respectively; in the second series (Fig. 2), the values were 9.8 ± 0.9 (n

=

20) and 7. 7 ± 0.6 (n

=

21; P

> 0.05). For the study of glucose-stimulated release, in the first series of experiments, the conditions were chosen so that the effect of extracellular ca++ was minimized at the time that the high glucose stimulus was applied. To achieve this, the extracellular Ca ++ concentration was reduced to 0.1 mM as the glucose concentration was increased to 16.7 mM (these results are shown in Fig. 1).

Under these conditions, insulin release is usually much reduced. Islets maintained in 5 mM phosphate were per- ifused with 2.8 mM glucose, 1.0 mM ca++, and 5 mM

-

^c

- _{- -}

E ^~

-

Ill

_~

- w en

<{

w

...J

w

0::

z

...J

:::::»

en z

-

100

80

60

40

20

0

~~~H~9~~::1$:?:r;,:~~

~ I ca++ O.lmM

~ I zt-zzzn "?l"zz ' " ' z zzz ²²²

mean ~ SEM

- Pi 5mM (n=16l

o---o

Pi 1 mM ( n= 17l

40 50 60 70 80

90 minutes

FIG. 1. Glucose-stimulated insulin release by islets which were main- tained in 1 or 5 mM phosphate media for 46 h before zero time. The perifusate concentration of glucose changed from 2.8 to 16.7 mM, and the concentration of Ca changed from 1.0 to 0.1 mM at 46 min. Vertical lines represent ±SEM. P;, Inorganic phosphate.

-

^c

100

- ^{-; 80}

E

-

^Ill

-

^C)

_a.

- _w 60 en

<{

w

...J

w

0::

z 40

...J

20

:::::»

en z

-

0

~Ca++1mM~

I

40

I

I I I I I I I

I I I

I

I I

I

50

mean± SEM

-

^{Pi 5mM ln=21 l}

~ Pi 1 mM (n=20)

60 70 80

90 minutes

FIG. 2. Glucose-stimulated insulin release by islets which were main- tained in 1 or 5 mM phosphate media for 4 h before zero time. The perifusate concentration of glucose changed from 2.8 to 16.7 mM at 46 min. Ca remained unchanged at 1.0 mM throughout. Vertical lines represent ±SEM. P;, Inorganic phosphate.

phosphate to maintain high ca++ loading. At 46 min, the composition of the perifusate was changed to 16.7 mM

glucose and 0.1 mM ca++, leaving the 5 mM phosphate unchanged. Islets maintained in normal phosphate were switched similarly at 46 min to 16.7 mM glucose and 0.1

mM Ca ++, while leaving phosphate unchanged at 1.0 mM.

In a second series of experiments, the medium Ca ++

concentration was kept at 1.0 mM throughout the exper- iment (these results are shown in Fig. 2).

With low Ca++ (0.1 IDM) in the medium, the response of islets maintained under normal conditions and with normal stores of Ca ++ was characterized, as expected, by very small first and second phases of release. The time course of the first phase response was normal and was complete by the 53rd minute, a time which is at the nadir between the first and second phases in the normal con- trols (Fig. 2). The response of islets maintained in high phosphate and with high levels of intracellular Ca ++ is in

1016 KIKUCHI ET AL. Endo • 1979 Vol105 • No4

marked contrast to normally maintained islets. Despite the perifusate ca++ concentration of 0.1 mM, glucose induced a prompt and very large surge of insulin release.

The first phase response reached peak rates of release of 79 pg/islet. min. This may be compared with the peak rate of normally maintained islets, which was 25 pg/islet·

min. The decline in the insulin release rate from the loaded islets then appeared to have a normal time course.

The difference between the first phase insulin release by ca++_loaded islets and that by normal islets was deter- mined (Table 1) by integration of the insulin release above baseline for each individual perifusion chamber between min 47-51, a period of time which best quanti- tates the first phase with minimal interference (overlap) by the second phase. The ca++_loaded islets released 185 pg insulin/islet, in contrast to 53 pg by the normal islets.

Thus, the ca++_loaded islets released 3.5 times more during the first phase than the normal islets. Further- more, comparison with Fig. 2 shows that the Ca ++-loaded islets released as much insulin during the first phase in the face of 0.1 mM ca++ in the medium as did normal islets in the presence of 1.0 mM Ca ++.

Second phase insulin release was reduced in both nor- mal and loaded islets by the presence of 0.1 mM ca++ in the medium (Figs. 1 and 2), although the loaded islets were secreting insulin at a higher rate than the normal islets (Fig. 1 and Table 1). Interestingly, in both situations (i.e. with normal and ca++_loaded islets), the second phase release was constant from the end of the first phase to the end of the observation period. Thus, no slowly rising second phase was detected in the presence of 0.1

mM ca++ as is usually the case in the presence of 1.0 mM

ca++.

When the responses of the normal and ca++_loaded islets were compared in the presence of the usuall.O mM

Ca ++ in the medium, no significant differences were detected. First and second phases followed similar courses. This suggests that the Ca ++-loading procedure has not impaired the integrity of the islets, as does the similarity of islet contents of insulin and glucagon (see results in Table 2).

Two series of control experiments were performed. In the first, the possibility that 5 mM phosphate acutely enhances glucose-induced insulin release was considered.

Islets maintained under normal phosphate conditions were perifused in normal KRB buffer up to the min 46.

At this time, one group of islets was exposed to a medium containing 16.7 mM glucose, 0.1 mM ca++, and 5 mM

phosphate, and another group was exposed to 16.7 mM

glucose and 0.1 mM ca++ with normal phosphate main- tained. First phase insulin release was similar in the two groups. In normal phosphate, the release was 32 ± 3 pg insulin/islet, and in 5 mM phosphate, the release was 31

± 6 pg insulin/islet (P

=

0.8; n

=

4). Therefore, the presence of a high phosphate concentration per se under "

these conditions had no influence on glucose-induced release. The second control series investigated the pos- sibility that in islets maintained for 46 h in 5 ^mM phos- phate, the change in perifusate concentration from 1.0 to 0.1 mM ca++ was responsible for the large stimulation of insulin release rather than the increased glucose concen- tration. Therefore, islets maintained under high' and nor- mal phosphate conditions were perifused in high and normal phosphate KRB buffer, respectively, throughout the experiments. After 46 min, both sets of islets were changed from 1.0 to 0.1 mM ca++ but with no change in the 2.8 mM glucose concentration. The change in Ca ++

caused only a small and transient insulin release. In the normal islets, total incremental insulin release above baseline was 8 ± 4 pg/islet during min 47-51 (n = 18), TABLE 1. Incremental insulin release above baseline from islets maintained in normal or high phosphate media and exposed to 16.7 mM glucose in the presence of 1.0 and 0.1 mM ca++

Phosphate (1 mM) 1st Phase release [pg/islet (±SEM). 5 min]

ca++

l.OmM 148 ± 14 (20)

O.lmM 53± 10 (17)

%Change -64

F 37.41

p <0.001

2nd Phase release [pg/islet (±SEM) ·39 min]

ca++

l.OmM 1307 ± 84 (20)

O.lmM 237 ± 17 (17)

%Change -82

F _104.39

p <0.001

Phosphate (5 mM)

171 ± 15 (21) 185 ±42 (16)

8 0.29 NS

1214 ± 78 (21) 608 ±99 (16)

-50 21.16

<0.001

F

1.71 10.11

0.22 14.38

p

NS

<0.005

NS

<0.001

The first phase represents samples collected at min 47-51 (first 5 min after stimulation), and the second phase represents the samples collected at min 52-90. Statistical analysis was performed by one-way analysis of variance. Nonsignificant values refer toP> 0.20 or higher. The number of observations is in parentheses.

ca++ STORES IN BIPHASIC INSULIN RELEASE ¹⁰¹⁷ TABLE 2. Contents of insulin and glucagon in islets maintained for 46

h in normal or high phosphate media

Insulin Glucagon

1mM 38 ± 4 (15)

5 ± 1 (15)

Phosphate

5mM 43 ± 2 (20)"

5 ± 1 (13) Results are expressed as nanograms per islet(± SEM). The number of determinations is shown in parentheses.

"P> 0.20.

while the Ca ++-loaded islets released 12 ± 3 pg/islet during min 47-51 (n = 10); the release rates were not significantly different (P > 0.3). After 51 min, the release from both sets of islets returned to the basal rates seen before the switch in Ca ++ concentration. These results were not changed by the addition of 13.9 mM L-glucose at 46 min (data not shown).

Discussion

The in vitro maintenance of isolated pancreatic islets permits prolonged exposure of the islets to various con- ditions. The dramatic increase of total exchangeable calcium in the islets after maintenance for 2 days in the presence of 5 mM phosphate had no adverse effect on their content or release of insulin. Neither basal nor glucose-stimulated insulin release from islets maintained and perifused at high (5 mM) phosphate concentrations differed significantly from those of control islets when the perifusate Ca ++ was 1 mM. Since, in addition, insulin content of the two types was similar, it is unlikely that high phosphate concentrations induce changes in insulin biosynthesis or glucose recognition by the islets. Fur- thermore, the addition of 5 mM phosphate only during glucose stimulation did not alter the insulin release. This appears to be at variance with the results of Hellman and Andersson, who reported that 10 mM phosphate inhibit glucose-induced insulin release from freshly isolated ob/

ob mouse islets (40) when added together with the stim- ulus. These authors also reported that high phosphate concentrations stimulate the uptake of ⁴⁵Ca++ (40).

Two effects of glucose on the net movement of Ca across the plasma membrane have been documented in islets of Langerhans which are thought to be related to the stimulation of insulin release. These are: 1) inhibition of ca++ efflux (37, 41-43) and 2) increased ca++ uptake (23-30). A major role for the precise control of cytosol Ca ++ has been given to the membranes of intracellular organelles which store Ca because, despite a lack of knowledge of flux rate constants in intact cells, the intra- cellular storage capacity is large, calcium uptake mech- anisms have been demonstrated (44-46), and the surface area of intracellular membranes is large relative to the surface area of the plasma membrane. In seeking an

explanation for the biphasic nature of glucose-induced insulin release, the possibility has been considered that the interplay of the effects of glucose on intracellular and extracellular Ca handling is responsible for a biphasic change in the cytosol ca++ concentration which, in turn, determines the biphasic pattern of release. As a first step, the role of extracellular Ca and glucose-stimulated up- take of ⁴⁵Ca++ was studied. It was found that by the use of Verapamil, an agent known to block Ca ++ channels, glucose-stimulated ⁴⁵Ca++ uptake could be completely blocked without any significant effect on the first phase of insulin release (23). Therefore, glucose-stimulated Ca uptake plays no role in the first phase of release. As a consequence, if an increase of cytosol Ca ++ is required for insulin release, then utilization of intracellular Ca ++

must be involved in the causation of first phase release.

Evidence in favor of this conclusion can be seen in the paper by Lundquist et al. (47) in which an experimental protocol similar to that used in this study was applied, i.e. when the isolated perfused pancreas was subjected to a change from low glucose and normal Ca ++ to high glucose and zero Ca ++. In this situation, a first phase response was primarily observed. Similarly, experiments by Henquin (48) involving exposure of islets to high glucose and approximately 20 p.M Ca ++ resulted in almost complete suppression of the second phase but only 50%

inhibition of the first phase. As the first and second phases were not clearly distinct in these experiments, some of the 50% inhibition observed during the first phase could be ascribed to inhibition of the overlapping second phase component (23). Thus, the first phase shows considerable resistance to decreased extracellular Ca ++ concentrations, as would be expected if it were largely dependent upon intracellular Ca++. However, on the basis of experiments with theophylline in the pres- ence of 20 p.M Ca ++ in which an apparent second, but not first, phase response was resto~ed, Henquin concluded that the second phase response was more dependent upon intracellular Ca ++ than the first phase one. This conclusion is based on the assumption that theophylline mobilizes intracellular Ca ++, and we are unable to rec- oncile it with our own data.

In the present study, both phases of glucose-induced insulin release in control islets were reduced in the pres- ence of low extracellular Ca ++, albeit the first phase less than the second phase (Table 1). This difference between the two phases was best seen in islets which had their intracellular Ca ++ stores loaded. When perifused in the presence of low ca++, the rates of insulin release during the first phase were elevated to values similar to those seen at normal extracellular ca++. However, the release rates during the second phase, while being higher than those seen in control islets at low Ca ++, failed to reach the rates observed at normal ca++ (50% reduction). It

1018 KIKUCHI ET AL. Endo • 1979 Vol lOS • No4

appears, therefore, that intracellular ea++ stores are in- volved in the control of both phases of insulin release, but that their role is particularly important for the full development of the first phase. This agrees with our previous data that Verapamil blockade of glucose-stim- ulated calcium uptake did not inhibit first phase release but inhibited second phase insulin release by 55%. The combined data suggest strongly that the first phase re- sponse to glucose is generated by utilization of intracel- lular ea++ and the second phase requires both intracel- lular ea ++ and increased uptake of extracellular ea ++ for its full development. The reason glucose did not elicit a higher insulin release from the islets with increased ea stores at 1 mM ea++ is unclear. It may be that insulin release in response to 16.7 mM glucose in normal islets perifused at 1 mM ea ++ is already maximal and cannot be further augmented by increased ea stores. This is supported by our earlier observations that insulin release in response to 16.7 mM glucose was not further enhanced at 10 mM extracellular ea++ (37), a ea++ concentration which by itself stimulates insulin release (16).

When assessing the effects of glucose on cytosolic ea ++

concentrations, it has been proposed that •sea ++ efflux from preloaded islets may reflect the cytosolic ea ++

concentration. This view is supported by the fact that glucose-stimulated insulin release can be completely in- hibited without major changes in glucose-stimUlated

•sea++ efflux (29, 37, 40, 49). This makes it unlikely that

•sea++ is released together with the secretory granule content during exocytosis. It is therefore of particular interest that a biphasic ⁴sea ++ efflux pattern is associated with the biphasic insulin release when two distinct phases are observed, as is the case in this and an earlier study (23) in islets maintained for 2 days in the presence of

•sea++. Based on these considerations and the results from the previous (23) and present studies, a model for biphasic insulin release is proposed. The initial inhibitory effect of glucose on ea++ efflux across the plasma mem- brane and a possible inhibition of intracellular ea++

sequestration by glucose or its metabolites (46) would both contribute to raise cytosolic ea++. It appears that cytosolic ea ++ rises to high values and is then pumped down by an increased activity of the ea ++-extruding mechanism in the plasma membrane seen as a spike in

•sea++ efflux (23, 37, 49) and possibly by intracellular storage mechanisms as these try to adjust to a new equilibrium position. This overshoot phenomenon would determine the first phase insulin response. The new equilibrium position is thought to be reached at the nadir between the two phases of insulin release. This is indi- cated by the constant insulin release rate during the second phase when the influence of extracellular ea ++

was minimized by perifusing both normal and ea ++- loaded islets at 0.1 mM ea++ (Fig. 1). Similarly, when

glucose-stimulated ea ++ uptake was inhibited by Vera- pamil, second phase insulin release was reduced to a simple plateau (23). Indeed, under these conditions, glu- cose-stimulated ⁴⁵ea ++ efflux does not show a secondary rise during the second phase but, after the first phase spike, remains parallel with basal ⁴⁵ea++ efflux (23). The

45ea ++ efflux and insulin release pattern thus support the concept of a new equilibrium in ea++ fluxes reached at the nadir when the effects of extracellular ea ++ are reduced. The rising portion of the second phase insulin release seen at normal extracellular ea ++ could be due to a gradually increasing cytosolic ea++ concentration, as a gradual net gain of ea ++results from the glucose-induced imbalance of plasma membrane fluxes.

Acknowledgments

The authors are grateful to Mrs. Savina Kalfopoulos and Miss Danielle Cassard for their skilled technical assistance.

References

1. Cerasi, E., and R. Luft, The plasma insulin response to glucose infusion in healthy subjects and in diabetes mellitus, Acta Endo- crinol (Kbh) 55: 278, 1967.

2. Curry, D. L., L. L. Bennett, and G. M. Grodsky, Dynamics of insulin secretion by the perfused rat pancreas, Endocrinology 83: 572, 1968.

3. Porte, D., and A. A. Pupo, Insulin responses to glucose: evidence for a two pool system in man, J Clin Invest 48: 2309, 1969.

4. Rabinovitch, A., A. Gutzeit, M. Kikuchi, E. Cerasi, and A. E.

Renold, Defective early phase insulin release in perifused isolated pancreatic islets of spiny mice (Acomys cahirinus), Diabetologia 11: 457, 1975.

5. Lacy, P. E., S. L. Howell, D. Y. Young, and C. V. Fink, New hypothesis of insulin secretion, Nature (Lond) 219: 1177, 1968.

6. Grodsky, G. M., A threshold distribution hypothesis for packet storage of insulin and its mathematical modeling, J Clin Invest 51:

2047, 1972.

7. Malaisse, W. J., E. Van Obberghen, G. Devis, G. Somers, and M.

Ravazzola, Dynamics of insulin release and microtubular-microfi- lamentous system, Eur J Clin Invest 4: 313, 1974.

8. Mathews, E. K., and P.M. Dean, Electrical activity in islet cells, In Falkmer, S., B. Hellman, and I.-B. Tii.ljedal (eds.), The Structure and Metabolism of the Pancreatic Islets, Pergamon Press, Oxford, 1970, pp. 305-312.

9. Iversen, J., and D. W. Miles, Evidence for a feedback inhibition of insulin secretion in the isolated perfused canine pancreas, Diabetes 20: 1, 1971.

10. Pace, C. S., F. M. Matschinsky, P. E. Lacy, and S. Conant, Electro- physiological evidence for the autoregulation of ,8-cell secretion by insulin, Biochim Biophys Acta 497: 408, 1977.

11. Orci, L., and R. H. Unger, Functional subdivision of islets of Langerhans and possible role of D-cells, Lancet 2: 1243, 1975.

12. Randle, P. J., and C. N. Hales, Insulin release mechanisms, In Steiner, D. F., and N. Freinkel (eds.), Handbook of Physiology, vol.

1, sect. 7, American Physiological Society, Washington DC, 1972, pp. 219-235.

13. Grodsky, G. M., and L. L. Bennett, Cation requirements for insulin secretion in the isolated perfused pancreas, Diabetes 15: 910, 1966.

14. Milner, R. D. G., and C. N. Hales, The role of calcium and magnesium in insulin secretion from rabbit pancreas studied in vitro, Diabetologia 3: 47, 1967.

15. Devis, G., G. Somers, and W. J. Malaisse, Stimulation of insulin release by calcium, Biochem Biophys Res Commun 67: 525, 1975.

16. Hellman, B., Stimulation of insulin release after raising extracellu-

ca++ STORES IN BIPHASIC INSULIN RELEASE 1019

Jar calcium, FEBS Lett 63: 125, 1976.

17. Wollheim, C. B., B. Blonde!, P. A. Trueheart, A. E. Renold, and G.

W. G. Sharp, Calcium-induced insulin release in monolayer culture of the endocrine pancreas. Studies with ionophore A23187, J Bioi Chern 250: 1354, 1975.

18. Ashby, J.P., and R.N. Speake, Insulin and glucagon secretion from isolated islets of Langerhans. The effect of calcium ionophores, Biochem J 150: 89, 1975.

19. Karl, R. C., W. S. Zawalich, J. A. Ferrendelli, and F. M. Matschin- sky, The role ofCa++ and adenosine 3'-5'-monophosphate in insulin release induced in vitro by the divalent cation ionophore A23187, J Bioi Chern 250: 4575, 1975.

20. Charles, M.A., J. Lawecki, R. Pictet, and G. M. Grodsky, Insulin secretion. Interrelationships of glucose, adenosine 3' -5' -monophos- phate and calciun, J Bioi Chern 250: 6134, 1975.

21. Douglas, W. W., Stimulus-secretion coupling: the concept and clues from chromaffin and other cells. The First Caddum Memorial Lecture, Cambridge, September 1967, Br J Pharmacal 34: 451, 1968.

22. Rubin, R. P ., The role of calcium in the release of neurotransmitter substances and hormones, Pharmacal Rev 22: 389, 1970.

23. Wollheim, C. B., M. Kikuchi, A. E. Renold, and G. W. G. Sharp, The role of intracellular and extracellular ca++ in glucose-stimu- lated biphasic insulin release, J Clin Invest 62: 451, 1978.

24. Malaisse-Lagae, F., and W. J. Malaisse, Stimulus-secretion coupling of glucose-induced insulin release. Ill. Uptake of ⁴⁵calcium by isolated islets of Langerhans, Endocrinology 88: 72, 1971.

25. Hellman, B., J. Sehlin, and I.-B. Tiiljedal, Calcium uptake by pancreatic /J-cells as measured with the aid of ⁴⁵Ca and mannitol-

"H, Am J Physiol221: 1795, 1971.

26. Hellman, B., J. Sehlin, and I.-B. Tiiljedal, Effects of glucose on

•sea++ uptake by pancreatic islets as studied with the lanthanum method, J Physiol (Lond) 254: 639, 1976.

27. Henquin, J.-C., and A. E. Lambert, Cobalt inhibition of insulin secretion and calcium uptake by isolated rat islets, Am J Physiol 228: 1669, 1975.

28. Hellman, B., J. Sehlin, and I.-B. Tiiljedal, Calcium and secretion:

distinction between two pools of glucose-sensitive calcium in pan- creatic islets, Science 194: 1421, 1976.

29. Wollheim, C. B., M. Kikuchi, A. E. Renold, and G. W. G. Sharp, Somatostatin and epinephrine induced modifications of •sea++

fluxes and insulin release in rat pancreatic islets maintained in tissue culture, J Clin Invest 60: 1165, 1977.

30. Naber, S. P., M. L. McDaniel, and P. E. Lacy, The effect of glucose on the acute uptake and efflux of calcium-45 in isolated rat islets, Endocrinology 101: 686, 1977.

31. Malaisse, W. J., J. C. Hutton, A. Sener, J. Levy, A. Herchuelz, G.

Devis, and G. Somers, Calcium antagonists and islet function. VII.

Effect of calcium deprivation, J Membr Bioi 38: 193, 1978.

32. Herman, L., T. Salo, and C. N. Hales, The electron microscopic localization of cations to pancreatic islets of Langerhans and their possible role in insulin secretion, J Ultrastruct Res 42: 298, 1973.

33. Ravazzola, M., F. Malaisse-Lagae, M. Amherdt, A. Perrelet, W. J.

Malaisse, and L. Orci, Patterns of calcium localization in pancreatic endocrine cells, J Cell Sci 27: 107, 1976.

34. Lacy, P., and M. Kostianowsky, Method for the isolation of intact islets of Langerhans from the rat pancreas, Diabetes 16: 35, 1967.

35. Morton, H. J., A survey of commercially available tissue culture media, In Vitro 6: 89, 1970.

36. Borle, A. B., Kinetic analysis of calcium movements in cell culture.

V. Intracellular calcium distribution in kidney cells, J Membr Bioi 10: 45, 1972.

37. Kikuchi, M., C. B. Wollheim, G. Cuendet, A. E. Renold, and G. W.

G. Sharp, Studies on the dual effects of glucose on •sea++ efflux from isolated rat islets, Endocrinology 102: 1339, 1978.

38. Herbert, V., K.-S. Lau, C. W. Gottlieb, and S. J. Bleicher, Coated charcoal immunoassay of insulin, J Clin Endocrol Metab 25: 1375, 1965.

39. Unger, R. H., E. Aguilar-Parada, W. A. Muller, and A.M. Eisen- traut, Studies on pancreatic alpha cell function in normal and diabetic subjects, J Clin Invest 49: 837, 1970.

40. Hellman, B., and T. Andersson, Calcium and pancreatic /J-cell function. IV. Evidence that glucose and phosphate stimulate cal- cium-45 incorporation into different intracellular pools, Biochim Biophys Acta 541: 483, 1978.

41. Malaisse, W. J., G. R. Brisson, and L. E. Baird, Stimulus-secretion coupling of glucose-induced insulin release. X. Effect of glucose on

45Ca++ efflux from perifused islets, Am J Physiol224: 389, 1973.

42. Bukowiecki, L., and N. Freinkel, Relationship between efflux of ionic calcium and phosphorus during excitation of pancreatic islets with glucose, Biochim Biophys Acta 436: 190, 1976.

43. Gylfe, E., and B. Hellman, Calcium and pancreatic /J-cell function.

II. Mobilization of glucose-sensitive ⁴°Ca ++ from perifused islets rich in /1-cells, Biochim Biophys Acta 538: 249, 1978.

44. Howell, S. L., and W. Montague, Regulation by nucleotides of

45calcium uptake in homogenates of rat islets of Langerhans, FEBS Lett 52: 48, 1975.

45. Sehlin, J., Calcium uptake by subcellular fractions of pancreatic islets. Effects of nucleotides and theophylline, Biochem J 156: 63, 1976.

46. Sugden, M. C., and S. J. H. Ashcroft, Effects of phosphoenolpyru- vate, other glycolytic intermediates and methylxanthines on cal- cium uptake by a mitochondrial fraction from rat pancreatic islets, Diabetologia 15: 173, 1978.

47. Lundquist, 1., R. Fanska, and G. M. Grodsky, Interaction of calcium and glucose on glucagon secretion, Endocrinology 99: 1304, 1976.

48. Henquin, J.-C., Relative importance of extracellular and intracel- lular calcium for the two phases of glucose-stimulated insulin release. Studies with theophylline, Endocrinology 102: 723, 1978.

49. Malaisse, W. J., A. Herchuelz, G. Devis, G. Somers, A. C. Boschero, J. C. Hutton, S. Kawazu, A. Sener, I. J. Atwater, G. Duncan, B.

Ribalet, and E. Rojas, Regulation of calcium fluxes and their regulatory roles in pancreatic islets, Ann. NY Acad Sci 307: 562, 1978.