Vol.
178,
August
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages 1182-1187
15, IS91
MEASURING
THE BALANCE BETWEEN INSULIN INSULIN RELEASE
SYNTHESIS
AND
FransC. Schuit, Rita Kiekens* and Daniel G. Pipeleers* Dept. of Biochemistry and*Dept. of Metabolism & Endocrinology, Faculteit Geneeskunde,Vrije Universiteit Brussel, Laarbeeklaan 103, B-1090 Brussels,Belgium Received
June
13,
1991
The absolute rates of hormone synthesis and release were determined in purified pancreatic B cells. Newly synthesizedproteins were labeledwith L-[3,5-3Hltyrosine or L-[2,5-3HJhistidine. When medium glucose was 5 10 mM, the production of insulin exceededor equaledits release. Raisingthe glucoselevels above 10 mM did not further increasethe rate of insulin synthesis(67+10 fmol/lO3 cells/2 hour) but elevated that of insulin releaseup to 3-fold the production rates (181flO fmol/lO3 cells/2 hour). In the presenceof glucagon or of the phorbol ester 12-O-tetradecanoylphorbol 13-acetatethe cells also reIeased 3-fold more hormone that they synthesized; releasewas however reducedto 25 % of the rate of production in the presenceof epinephrine. It is concluded that glucoseaswell ashormonalregulatorsof islet B cellscan influence, bi-directionally, the balancebetweenthe ratesof insulin synthesisand release. 0 1991Academic Press,Inc.
A reduction in pancreatic insulin reserve can lead to impaired glucose homeostasis.Its causemay be attributableto a decreasein the numberof pancreaticB cellsas well asto an imbalance between the relative rates of hormone production, degradation and release. Each of the latter processesis subject to physiologic regulation (I-5).
Glucose is
considered as a major component in thesecontrol mechanisms. The sugar has been shownto influence, dose-dependently,a numberof stepsbetweenthe transcription of the insulin gene(s)and the exocytosis of the processedhormone(6-12). It is so far unclear whether glucose itself can induce an imbalance between the relative rates of hormone production and release, and whether such effect would be a direct result of its concentration or of the simultaneouspresenceof (neuro)hormonalsignals. The present study addressesthis question in vitro. The selectedexperimental conditions allow a comparisonof the absoluterates of insulin biosynthesisand releaseat different glucose concentrations. It is examined to which extent hormonal and neural agents alter a glucose-inducedequilibrium betweenhormoneproduction andrelease. Materials and Methods Preparation of purified B cells. The techniquesfor islet isolation and B cell purification have been previously described (13). Briefly, isolated islets were prepared by collagenasedigestion of adult rat pancreasand dissociatedin a calcium-free medium containing trypsin and DNase. islet B and endocrine non-B cells were purified by autofluorescence-activatedcell sorting using a FACS-IV (13). The purified B cells were 0006-291X/91 Copyright All rights
$1.50
0 1991 by Academic Press, Inc. of reproduction in any form reserved.
1182
Vol.
178, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
reaggregatedfor 3 to 5 h. in a rotatory shakingincubator (Braun, Melsungen, FRG) and cultured for 16 h. (37’C, 5% Co2/95% air) in bacteriological petri dishes. The tissue culture medium was HAM-F10 (Gibco, Strathclyde, UK), supplementedwith 2 mM glutamine, 10% heat-inactivatedfetal calf serum,penicillin (0.1 mg/ml) and streptomycin (0.1 mg/ml). Incubations ofpure B cells. After culture, batchesof 2.5 x 104 reaggregatedB cells were labeled for 2 h. at 37°C in a total volume of 0.2 ml Earles-HEPESbuffer (EH-for composition, seeref. 13) containing 50 l,tCi L-[3,5-3Hltyrosine or L-[2,5JH]histidine (Amersham, Bucks, UK). The incubation was stopped by dilution in 1 ml cold EH buffer containing 1.4 mM glucose and 1 mM unlabeled tyrosine or histidine. After centrifugation the supernatantfraction was sampledfor the insulin radioimmunoassay. The cell pelletswere washedthree timesand sonicatedin 1 ml 2 M acetic acid containing 0.25M BSA (Sigma, St. Louis, MO, USA); the cell extracts were assayed for their content in 3H-labeledprotein and (pro)insulin. Total protein and proinsulin biosynthesis.Total protein synthesiswasdeterminedasthe radioactivity of the cellular pelletsand incubationsupematantsin 10%trichloroacetic acid (TCA) (14). The TCA soluble fraction represented lessthan 10 percent of the total cellular radioactivity: 1.6 & 0.2 percent at 10 mM glucoseand 5.9 + 2.2 percent at 1.3 mM glucose (mean-+ SD; n = 9). The counting efficiency of the g-counter (Beckman, Fullerton, CA) was 62 + 1% (mean+ SD, n = 7). Proinsulin biosynthesiswasmeasured by irnmunoprecipitationof the cellular extracts with guineapig anti-porcineinsulin serum (kindly provided by Dr. C. van Schravendijk from our department) and protein Asepharose CL4B (Pharmacia Uppsala, Sweden) (15). The recovery of the immunoprecipitation procedurewasestimatedby adding 5x104 cpm W-insulin to some of the samples;it varied between85 and95%. Expressionofresults. All resultsrepresentmeansk SEM of at least3 experiments. The significanceof differencesbetweenmeanvalues wascalculatedby the Student’st test. Results Measurementof protein synthesisi/zpure B cells. The incorporation of 3H-tyrosine and 3H-histidine into B cell protein was measuredat low (1.4 mM) and high (20 mM) glucoseconcentration (Table 1). Using tracersof high specific activity (56-60 Ci/mmol), the high glucosemediumwas found to induce 15-fold more protein synthesisand 30-fold more proinsulin synthesisthan the low glucosemedium. The ratio of proinsulin over
Table 1. Incorporation of 3H-tyrosine and 3H-histidine into B cell proteins. Purified B cells were incubated for 2 h in the presence of 3H-tyrosine or sH-histidinc at low (5.6-6 Ci/mmol) or high (5660 Ci/mmol) specific activity. Protein synthesis was measured at low (1.4 mM) and high (‘20 mM) glucose concentration. Data represent mean values + S.E.M. of n experiments. Tracer Ammo acid pM Ci/mmol aH-tyrosine 4.5
56
45
5.6
3H-histidine 4.2
60
42
6
[Ghicosc] mM
Protein synthesis (cpm/2 h/B cell) Total Proinsulin Non-insulin 0.8fO. 1 21.W1.2
2.6kO.4 23.4*2.3
0.25f0.03 0.49~0.04
5 5
0.08ti.01 2.6kO.2
0.24+0.06 2.8k0.3
0.27kO.06 0.48+0.04
3 3
1.8+0.2 19.6k1.6 0.22kO.02 2.4+0.1
0.19f0.02 0.38+0.03 0.15kO.03 0.37AO.02
5 5 3 3
1.4 20 1.4 20
3.4Lko.4 45.W1.8 0.328.06 5.3kO.3
1.4 20 1.4 20
0.4&O.1 2.2Iko.2 31.4k1.5 11.8kO.8 0.26kO.03 0.04kO.01 1.4f0.2 3.8M.3
1183
Vol.
778,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
total protein synthesis was two-fold higher at 20 mM glucose that at 1.4 mM glucose (f’ < 0.001 by paired t-testing). The difference between total protein synthesis and proinsulin
synthesis was defined as non-insulin protein (NIP) synthesis.
glucose, NIP synthesis was also lo-fold
At 20 mM
higher than at 1.4 mM glucose (P < 0.001).
Using the same tracers at a lo-fold lower specific activity decreased the incorporation of radioactive amino acid in protein and proinsulin approximately lo-fold (Table 1). The calculated rates of total protein and proinsulin biosynthesis
were therefore the same at
both specific activities. At 20 mM glucose, 145 k 10 fmol histidine and 290 k 16 fmol tyrosine were incorporated into the immunoreactive insulin pool of 18 islet B cells. The 2-fold higher incorporation of tyrosine molecules reflects insulin’s 2-fold higher content in tyrosine (4 residues per molecule) than in histidine (2 residues per molecule) (16). It was thus found that at 1.4 and 20 mM glucose, islet B cells synthesize respectively 3 f 0.2 and 73 + 5 fmol insulin per IO3 cells. Under the selected experimental conditions, the release of newly synthesised protein was negliglible as compared to the cellular pool of labeled protein: after 2 hour iabeling with 3H-tyrosine at 20 mM gtucose, only 1500 + 300 cpm were released per 103 B cells which is less than 3 % of the cellular 3H-labeled protein (58.000 k 11.000 cpm per 103 B cells - mean k SEM; n = 3). Comparison of the rates of glucose-induced insulin synthesis and release. Glucose stimulated dose-dependently the process of insulin synthesis and that of insulin release (Fig. 1). At 2.5 mM glucose the rate of proinsulin biosynthesis (4 k 1 fmol/103 B cells) did not statistically differ from that of insulin release (10 rt 5 fmol/lO3 B cells - P > 0.3).
At 7.5 mM glucose, insulin production increased to 56 + 9 fmol/103 cells
200
-
s
I 10
Glucose (mM)
Fig,
1. Effect
of glucose
on the rate
of insulin
synthesis
and
release. at the indicated glucose concentrations. The rates of hormone synthesis and release are expressed as fmol insulin per 103 B cells. Biosynthetic rates were calculated from the radioactivity incorporated into cellular (pro)insulin, considering that four tyrosine residues are incorporated per (pro)insulin molecule. Data represent mean values k SEM
PurifiedB cellswereincubatedfor 2 h with [sH]tyrosine (55 Ci/mmol; 2.50 @i/ml)
of 5 experiments.The paired Student’st-test was usedto calculate the statistical significance of differences between the rates of synthesis and release at a particular glucose concentration. * P < 0.05.
1184
Vol.
178,
No.
3, 1991
BIOCHEMICAL
AND
Insulin
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Synthesis
NS
NS
NS
L TPA (lo-* M) Gluca (lO~*M) Epi (10.‘M)
-
+ -
+ +
+ +
- + + +
Fig. 2. Effect of TPA, glucagon and epinephrine upon the rates of glucose-induced insulin synthesis and release. Purified B cells were incubated at 10 mM glucose in the presence or absence of 12-0tetradecanoylphorbol 13-acetate (TPA), glucagon (Gluca), or epinephrine (Epi) at the indicated concentrations. The rates of insulin biosynthesis and release are expressed as fmol per 103 B cells. Data represent mean values z?zSEM of 5 experiments. The significance of differences between control (glucose alone) and experimental conditions was calculated by the unpaired Students t-test modified by Bonferroni for multiple comparisons. NS: not significantly different; * P < 0.01 versus control.
whereas insulin release occurred at a 2- to 3-fold lower rate (21 I!I 3 fmo1/103 cells; P < 0.01).
At 10 mM glucose, the rate of hormone release (66 IL 13 fmol/lO3 cells) was
equal to that of insulin production (61 + 9 fmol/103 cells; Fig. 1). At 20 mM, 2- to 3fold more hormone was released (18 1 + 36 fmol/lO3 cells) than synthesised (67 f 10 fmol/lO3 cells, P c 0.02). When glucagon (10-g M) was added to the islet B cells at 10 mM glucose, insulin release was increased 3-fold (P < 0.01 vs glucose alone) but the rate of insulin biosynthesis was not affected (Fig. 2). A similar effect was observed with the phorbol ester 12-O-tetradecanoylphorbol
13-acetate (TPA) which increased glucose-
induced insulin release 2.2-fold without influencing the rate of insulin biosynthesis. Epinephrine (10m7M) suppressed 9.5% of the secretory activity of islet B cells at 10 mM glucose plus 10-g M glucagon (P < 0.01) but did neither affect the rate of hormone synthesis (Fig. 2). Discussion The process of insulin biosynthesis and that of insulin release have been the subject of However, quantitative data on the balance between both numerous investigations. cellular functions are scarce (2,3). The present work compares the absolute rates of both processes under conditions of a basal, stimulated and suppressed secretory activity. This study became feasible with the availability of purified B cell preparations which can be examined without having to account for possible interactions with other cell types. The choice of tyrosine and histidine as tracers allowed a more precise quantification of the 1185
vol.
178,
No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
biosynthetic processthan with previously usedmarkers. These amino acids are only incorporatedin the insulin segmentof the preproinsulinmoleculeandnot in the pre- or Cregions (16), so that the radioactivity of the insulin immunoreactive material is independent of posttranslational preproinsulin processing and can be taken as a quantitative index for the amount of newly synthesizedhormone. It is also known that rat preproinsulinI andII do not differ in their respectivenumberof tyrosine and histidine residues(16) so that the presentestimation of insulin biosynthesisis not influenced by differencesin expressionof the insulin I and II genes(17). In a static incubation at I .4 mM glucose, purified B cells synthesized approximately 3 fmol insulin per 103cells over 2 hours. This hormoneproduction should be interpreted asthe activity of 5 percent of the cells (14) and thus representsa rate of 60 fmol per 103 active cells, which is comparableto the production rates under conditions where most cells have beenrecruited (67 fmol/lOj cells at 10 mM glucose). During the sameperiod, 11 fmol insulin was releasedin the medium, part of which may correspondto leakage from damagedcells,part to active secretionfrom an asyet undeterminedfraction of the B cells (18). Up to 5 mM glucose no statistical difference was noticed between the respective rates of hormone synthesisand release. In this concentration range, glucose stimulated hormone synthesisdose-dependentlybut failed to induce a detectable rise in the releasedhormone. It is not excluded that a subpopulationof cells wasrecruited into a higher secretory activity but that the amountof glucose-inducedinsulin releaseremained undetectable against the background dischargefrom the total population. At 7.5 mM glucose,the islet B cellsproducedtwo-fold moreinsulin than they releasedover the same time period. Hormone degradationand secretionof newly synthesised(pro)insulin was negligible: at the end of incubation both the TCA soluble radioactivity in the cells and TCA precipitable radioactivity in the mediumremained under 5 percent of the cellular pool of newly synthesisedprotein. The cells have thus enlargedtheir storedinsulin pool by approximately 40 fmol per 103cells during this 2 hour incubation; this correspondsto approximately 0.5 percent of their initial insulin content. Above 7.5 mM, glucosedid not further elevate the rate of insulin biosynthesisbut extendedits dose-dependentstimulation of insulin releaseup to the ratesof hormoneproduction at 10 mM, and higher above this concentration. Degranulation of islet B cells can thus be causedby glucoseif the sugar levels exceed 10 mM. At lower concentrations, glucose augmentsthe cellular insulin store or leaves it unaffected, so that a reduction of the insulin reserves under this condition should rather be attributed to (nenro) hormonal influences which stimulate releasewithout augmentingsynthesis. The resultsobtained with glucagonand with TPA suggestthat increasedactivities in celluIar protein kinase A or C may mediate such process. The c&adrenergic receptor seems,on the other hand, capableof counteracting the depletion of insulin stores. Since (neuro) hormonal signalsplay an important role in the secretory responsivenessof normal B cells (19, 20), it is also realistic to consider them aspotential causesfor the degranulation of islet B cells, in particular at glucose levels under 10mM. 1186
Vol.
178, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Acknowledgments We thank E. Quartier and R. De Proft for technical assistance,N. Fennersfor secretarial help and C.F.H. van Schravendijk for supplying the anti-insulin antibody. This study was supportedby grantsfrom the Belgian Ministery of Scientific Policy (86/91-102) and by the Fund for Medical Scientific Research(3.00.59.86and 3.001389). Rita Kiekens is researchfellow at the National Fund for Scientific Research.
References 1. Howell, S.L. and Taylor, K.W. (1966) Biochim. Biophys. Acta 130, 519-521. 2. Sando, H., Borg, .I. and Steiner, D.F. (1972) J. Clin. Invest. 51, 1476-1485. 3. Halban, P.A. and Wollheim, C.B. (1980) J. Biol. Chem. 255,6003-6006. 4. Malaisse, W.J. (1973) Diabetologia 9, 167-173. 5. Ashcroft, S.j.h. (1980) Diabetologia 18, 5-15. 6. Permutt, M.A. and Kipnis, D.M.(1972) J. Biol. Chem. 247, 1194-1199. 7. Giddings, S.J., Chirgwin, J. and Permutt, M.A. (1981) J. Clin. Invest. 67, 952-960. 8. Brunstedt, J. and Chan, S.J. (1982) Biochem. Biophys. Res. Commun. 106, 1383-1389. 9. Permutt, M.A. (1974) J. Biol. Chem. 248,2738-2742. 10. Welsh, M., Scherberg, N., Gilmore, R. and Steiner, D.F. (1986) Biochem. J. 235, 459467. 11. Pipeleers,D.G. Pipeleers-Marichal, M. and Kipnis, D.M. (1976) Science 191, 88-90. 12. C&i, L., Amherdt, M., Malaisse-Lagae,F., Rouiller, C. and Renold, A.E. (1973) Science 179, 82-84. 13. Pipeleers, D.G., In ‘t Veld, P.A., Van De Winkel, M., Maes, E., Schuit, F.C. and Gepts, W. (1985) Endocrinology 117, 806-816. 14. Schuit, F.C., In ‘t Veld, P.A. and Pipeleers, D.G. (1988) Proc. Natl. Acad. Sci. (USA) 85, 3865-3869. 15. Berne, C. (1975) Endocrinology 97, 1241-1247. 16. Lomedico, P., Rosenthal, N., Efstratiadis, A., Gilbert, W., Kolodner, R. and Tizard, R. (1979) Cell 18,545-558. 17. Rhodes,C.J., Lucas, C.A. and Halban, P.A. (1987) FEBS Lett. 215, 179-182. 18. Salomon, D. and Meda, P. (1986) Exp. Cell Res. 162, 507-520. 19. Schuit, F.C. and Pipeleers,D.G. (1986) Science 232, 875-877. 20. Pipeleers,D.G. (1987) Diabetologia 30, 277-291.
1187
178,
August
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages 1182-1187
15, IS91
MEASURING
THE BALANCE BETWEEN INSULIN INSULIN RELEASE
SYNTHESIS
AND
FransC. Schuit, Rita Kiekens* and Daniel G. Pipeleers* Dept. of Biochemistry and*Dept. of Metabolism & Endocrinology, Faculteit Geneeskunde,Vrije Universiteit Brussel, Laarbeeklaan 103, B-1090 Brussels,Belgium Received
June
13,
1991
The absolute rates of hormone synthesis and release were determined in purified pancreatic B cells. Newly synthesizedproteins were labeledwith L-[3,5-3Hltyrosine or L-[2,5-3HJhistidine. When medium glucose was 5 10 mM, the production of insulin exceededor equaledits release. Raisingthe glucoselevels above 10 mM did not further increasethe rate of insulin synthesis(67+10 fmol/lO3 cells/2 hour) but elevated that of insulin releaseup to 3-fold the production rates (181flO fmol/lO3 cells/2 hour). In the presenceof glucagon or of the phorbol ester 12-O-tetradecanoylphorbol 13-acetatethe cells also reIeased 3-fold more hormone that they synthesized; releasewas however reducedto 25 % of the rate of production in the presenceof epinephrine. It is concluded that glucoseaswell ashormonalregulatorsof islet B cellscan influence, bi-directionally, the balancebetweenthe ratesof insulin synthesisand release. 0 1991Academic Press,Inc.
A reduction in pancreatic insulin reserve can lead to impaired glucose homeostasis.Its causemay be attributableto a decreasein the numberof pancreaticB cellsas well asto an imbalance between the relative rates of hormone production, degradation and release. Each of the latter processesis subject to physiologic regulation (I-5).
Glucose is
considered as a major component in thesecontrol mechanisms. The sugar has been shownto influence, dose-dependently,a numberof stepsbetweenthe transcription of the insulin gene(s)and the exocytosis of the processedhormone(6-12). It is so far unclear whether glucose itself can induce an imbalance between the relative rates of hormone production and release, and whether such effect would be a direct result of its concentration or of the simultaneouspresenceof (neuro)hormonalsignals. The present study addressesthis question in vitro. The selectedexperimental conditions allow a comparisonof the absoluterates of insulin biosynthesisand releaseat different glucose concentrations. It is examined to which extent hormonal and neural agents alter a glucose-inducedequilibrium betweenhormoneproduction andrelease. Materials and Methods Preparation of purified B cells. The techniquesfor islet isolation and B cell purification have been previously described (13). Briefly, isolated islets were prepared by collagenasedigestion of adult rat pancreasand dissociatedin a calcium-free medium containing trypsin and DNase. islet B and endocrine non-B cells were purified by autofluorescence-activatedcell sorting using a FACS-IV (13). The purified B cells were 0006-291X/91 Copyright All rights
$1.50
0 1991 by Academic Press, Inc. of reproduction in any form reserved.
1182
Vol.
178, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
reaggregatedfor 3 to 5 h. in a rotatory shakingincubator (Braun, Melsungen, FRG) and cultured for 16 h. (37’C, 5% Co2/95% air) in bacteriological petri dishes. The tissue culture medium was HAM-F10 (Gibco, Strathclyde, UK), supplementedwith 2 mM glutamine, 10% heat-inactivatedfetal calf serum,penicillin (0.1 mg/ml) and streptomycin (0.1 mg/ml). Incubations ofpure B cells. After culture, batchesof 2.5 x 104 reaggregatedB cells were labeled for 2 h. at 37°C in a total volume of 0.2 ml Earles-HEPESbuffer (EH-for composition, seeref. 13) containing 50 l,tCi L-[3,5-3Hltyrosine or L-[2,5JH]histidine (Amersham, Bucks, UK). The incubation was stopped by dilution in 1 ml cold EH buffer containing 1.4 mM glucose and 1 mM unlabeled tyrosine or histidine. After centrifugation the supernatantfraction was sampledfor the insulin radioimmunoassay. The cell pelletswere washedthree timesand sonicatedin 1 ml 2 M acetic acid containing 0.25M BSA (Sigma, St. Louis, MO, USA); the cell extracts were assayed for their content in 3H-labeledprotein and (pro)insulin. Total protein and proinsulin biosynthesis.Total protein synthesiswasdeterminedasthe radioactivity of the cellular pelletsand incubationsupematantsin 10%trichloroacetic acid (TCA) (14). The TCA soluble fraction represented lessthan 10 percent of the total cellular radioactivity: 1.6 & 0.2 percent at 10 mM glucoseand 5.9 + 2.2 percent at 1.3 mM glucose (mean-+ SD; n = 9). The counting efficiency of the g-counter (Beckman, Fullerton, CA) was 62 + 1% (mean+ SD, n = 7). Proinsulin biosynthesiswasmeasured by irnmunoprecipitationof the cellular extracts with guineapig anti-porcineinsulin serum (kindly provided by Dr. C. van Schravendijk from our department) and protein Asepharose CL4B (Pharmacia Uppsala, Sweden) (15). The recovery of the immunoprecipitation procedurewasestimatedby adding 5x104 cpm W-insulin to some of the samples;it varied between85 and95%. Expressionofresults. All resultsrepresentmeansk SEM of at least3 experiments. The significanceof differencesbetweenmeanvalues wascalculatedby the Student’st test. Results Measurementof protein synthesisi/zpure B cells. The incorporation of 3H-tyrosine and 3H-histidine into B cell protein was measuredat low (1.4 mM) and high (20 mM) glucoseconcentration (Table 1). Using tracersof high specific activity (56-60 Ci/mmol), the high glucosemediumwas found to induce 15-fold more protein synthesisand 30-fold more proinsulin synthesisthan the low glucosemedium. The ratio of proinsulin over
Table 1. Incorporation of 3H-tyrosine and 3H-histidine into B cell proteins. Purified B cells were incubated for 2 h in the presence of 3H-tyrosine or sH-histidinc at low (5.6-6 Ci/mmol) or high (5660 Ci/mmol) specific activity. Protein synthesis was measured at low (1.4 mM) and high (‘20 mM) glucose concentration. Data represent mean values + S.E.M. of n experiments. Tracer Ammo acid pM Ci/mmol aH-tyrosine 4.5
56
45
5.6
3H-histidine 4.2
60
42
6
[Ghicosc] mM
Protein synthesis (cpm/2 h/B cell) Total Proinsulin Non-insulin 0.8fO. 1 21.W1.2
2.6kO.4 23.4*2.3
0.25f0.03 0.49~0.04
5 5
0.08ti.01 2.6kO.2
0.24+0.06 2.8k0.3
0.27kO.06 0.48+0.04
3 3
1.8+0.2 19.6k1.6 0.22kO.02 2.4+0.1
0.19f0.02 0.38+0.03 0.15kO.03 0.37AO.02
5 5 3 3
1.4 20 1.4 20
3.4Lko.4 45.W1.8 0.328.06 5.3kO.3
1.4 20 1.4 20
0.4&O.1 2.2Iko.2 31.4k1.5 11.8kO.8 0.26kO.03 0.04kO.01 1.4f0.2 3.8M.3
1183
Vol.
778,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
total protein synthesis was two-fold higher at 20 mM glucose that at 1.4 mM glucose (f’ < 0.001 by paired t-testing). The difference between total protein synthesis and proinsulin
synthesis was defined as non-insulin protein (NIP) synthesis.
glucose, NIP synthesis was also lo-fold
At 20 mM
higher than at 1.4 mM glucose (P < 0.001).
Using the same tracers at a lo-fold lower specific activity decreased the incorporation of radioactive amino acid in protein and proinsulin approximately lo-fold (Table 1). The calculated rates of total protein and proinsulin biosynthesis
were therefore the same at
both specific activities. At 20 mM glucose, 145 k 10 fmol histidine and 290 k 16 fmol tyrosine were incorporated into the immunoreactive insulin pool of 18 islet B cells. The 2-fold higher incorporation of tyrosine molecules reflects insulin’s 2-fold higher content in tyrosine (4 residues per molecule) than in histidine (2 residues per molecule) (16). It was thus found that at 1.4 and 20 mM glucose, islet B cells synthesize respectively 3 f 0.2 and 73 + 5 fmol insulin per IO3 cells. Under the selected experimental conditions, the release of newly synthesised protein was negliglible as compared to the cellular pool of labeled protein: after 2 hour iabeling with 3H-tyrosine at 20 mM gtucose, only 1500 + 300 cpm were released per 103 B cells which is less than 3 % of the cellular 3H-labeled protein (58.000 k 11.000 cpm per 103 B cells - mean k SEM; n = 3). Comparison of the rates of glucose-induced insulin synthesis and release. Glucose stimulated dose-dependently the process of insulin synthesis and that of insulin release (Fig. 1). At 2.5 mM glucose the rate of proinsulin biosynthesis (4 k 1 fmol/103 B cells) did not statistically differ from that of insulin release (10 rt 5 fmol/lO3 B cells - P > 0.3).
At 7.5 mM glucose, insulin production increased to 56 + 9 fmol/103 cells
200
-
s
I 10
Glucose (mM)
Fig,
1. Effect
of glucose
on the rate
of insulin
synthesis
and
release. at the indicated glucose concentrations. The rates of hormone synthesis and release are expressed as fmol insulin per 103 B cells. Biosynthetic rates were calculated from the radioactivity incorporated into cellular (pro)insulin, considering that four tyrosine residues are incorporated per (pro)insulin molecule. Data represent mean values k SEM
PurifiedB cellswereincubatedfor 2 h with [sH]tyrosine (55 Ci/mmol; 2.50 @i/ml)
of 5 experiments.The paired Student’st-test was usedto calculate the statistical significance of differences between the rates of synthesis and release at a particular glucose concentration. * P < 0.05.
1184
Vol.
178,
No.
3, 1991
BIOCHEMICAL
AND
Insulin
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Synthesis
NS
NS
NS
L TPA (lo-* M) Gluca (lO~*M) Epi (10.‘M)
-
+ -
+ +
+ +
- + + +
Fig. 2. Effect of TPA, glucagon and epinephrine upon the rates of glucose-induced insulin synthesis and release. Purified B cells were incubated at 10 mM glucose in the presence or absence of 12-0tetradecanoylphorbol 13-acetate (TPA), glucagon (Gluca), or epinephrine (Epi) at the indicated concentrations. The rates of insulin biosynthesis and release are expressed as fmol per 103 B cells. Data represent mean values z?zSEM of 5 experiments. The significance of differences between control (glucose alone) and experimental conditions was calculated by the unpaired Students t-test modified by Bonferroni for multiple comparisons. NS: not significantly different; * P < 0.01 versus control.
whereas insulin release occurred at a 2- to 3-fold lower rate (21 I!I 3 fmo1/103 cells; P < 0.01).
At 10 mM glucose, the rate of hormone release (66 IL 13 fmol/lO3 cells) was
equal to that of insulin production (61 + 9 fmol/103 cells; Fig. 1). At 20 mM, 2- to 3fold more hormone was released (18 1 + 36 fmol/lO3 cells) than synthesised (67 f 10 fmol/lO3 cells, P c 0.02). When glucagon (10-g M) was added to the islet B cells at 10 mM glucose, insulin release was increased 3-fold (P < 0.01 vs glucose alone) but the rate of insulin biosynthesis was not affected (Fig. 2). A similar effect was observed with the phorbol ester 12-O-tetradecanoylphorbol
13-acetate (TPA) which increased glucose-
induced insulin release 2.2-fold without influencing the rate of insulin biosynthesis. Epinephrine (10m7M) suppressed 9.5% of the secretory activity of islet B cells at 10 mM glucose plus 10-g M glucagon (P < 0.01) but did neither affect the rate of hormone synthesis (Fig. 2). Discussion The process of insulin biosynthesis and that of insulin release have been the subject of However, quantitative data on the balance between both numerous investigations. cellular functions are scarce (2,3). The present work compares the absolute rates of both processes under conditions of a basal, stimulated and suppressed secretory activity. This study became feasible with the availability of purified B cell preparations which can be examined without having to account for possible interactions with other cell types. The choice of tyrosine and histidine as tracers allowed a more precise quantification of the 1185
vol.
178,
No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
biosynthetic processthan with previously usedmarkers. These amino acids are only incorporatedin the insulin segmentof the preproinsulinmoleculeandnot in the pre- or Cregions (16), so that the radioactivity of the insulin immunoreactive material is independent of posttranslational preproinsulin processing and can be taken as a quantitative index for the amount of newly synthesizedhormone. It is also known that rat preproinsulinI andII do not differ in their respectivenumberof tyrosine and histidine residues(16) so that the presentestimation of insulin biosynthesisis not influenced by differencesin expressionof the insulin I and II genes(17). In a static incubation at I .4 mM glucose, purified B cells synthesized approximately 3 fmol insulin per 103cells over 2 hours. This hormoneproduction should be interpreted asthe activity of 5 percent of the cells (14) and thus representsa rate of 60 fmol per 103 active cells, which is comparableto the production rates under conditions where most cells have beenrecruited (67 fmol/lOj cells at 10 mM glucose). During the sameperiod, 11 fmol insulin was releasedin the medium, part of which may correspondto leakage from damagedcells,part to active secretionfrom an asyet undeterminedfraction of the B cells (18). Up to 5 mM glucose no statistical difference was noticed between the respective rates of hormone synthesisand release. In this concentration range, glucose stimulated hormone synthesisdose-dependentlybut failed to induce a detectable rise in the releasedhormone. It is not excluded that a subpopulationof cells wasrecruited into a higher secretory activity but that the amountof glucose-inducedinsulin releaseremained undetectable against the background dischargefrom the total population. At 7.5 mM glucose,the islet B cellsproducedtwo-fold moreinsulin than they releasedover the same time period. Hormone degradationand secretionof newly synthesised(pro)insulin was negligible: at the end of incubation both the TCA soluble radioactivity in the cells and TCA precipitable radioactivity in the mediumremained under 5 percent of the cellular pool of newly synthesisedprotein. The cells have thus enlargedtheir storedinsulin pool by approximately 40 fmol per 103cells during this 2 hour incubation; this correspondsto approximately 0.5 percent of their initial insulin content. Above 7.5 mM, glucosedid not further elevate the rate of insulin biosynthesisbut extendedits dose-dependentstimulation of insulin releaseup to the ratesof hormoneproduction at 10 mM, and higher above this concentration. Degranulation of islet B cells can thus be causedby glucoseif the sugar levels exceed 10 mM. At lower concentrations, glucose augmentsthe cellular insulin store or leaves it unaffected, so that a reduction of the insulin reserves under this condition should rather be attributed to (nenro) hormonal influences which stimulate releasewithout augmentingsynthesis. The resultsobtained with glucagonand with TPA suggestthat increasedactivities in celluIar protein kinase A or C may mediate such process. The c&adrenergic receptor seems,on the other hand, capableof counteracting the depletion of insulin stores. Since (neuro) hormonal signalsplay an important role in the secretory responsivenessof normal B cells (19, 20), it is also realistic to consider them aspotential causesfor the degranulation of islet B cells, in particular at glucose levels under 10mM. 1186
Vol.
178, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Acknowledgments We thank E. Quartier and R. De Proft for technical assistance,N. Fennersfor secretarial help and C.F.H. van Schravendijk for supplying the anti-insulin antibody. This study was supportedby grantsfrom the Belgian Ministery of Scientific Policy (86/91-102) and by the Fund for Medical Scientific Research(3.00.59.86and 3.001389). Rita Kiekens is researchfellow at the National Fund for Scientific Research.
References 1. Howell, S.L. and Taylor, K.W. (1966) Biochim. Biophys. Acta 130, 519-521. 2. Sando, H., Borg, .I. and Steiner, D.F. (1972) J. Clin. Invest. 51, 1476-1485. 3. Halban, P.A. and Wollheim, C.B. (1980) J. Biol. Chem. 255,6003-6006. 4. Malaisse, W.J. (1973) Diabetologia 9, 167-173. 5. Ashcroft, S.j.h. (1980) Diabetologia 18, 5-15. 6. Permutt, M.A. and Kipnis, D.M.(1972) J. Biol. Chem. 247, 1194-1199. 7. Giddings, S.J., Chirgwin, J. and Permutt, M.A. (1981) J. Clin. Invest. 67, 952-960. 8. Brunstedt, J. and Chan, S.J. (1982) Biochem. Biophys. Res. Commun. 106, 1383-1389. 9. Permutt, M.A. (1974) J. Biol. Chem. 248,2738-2742. 10. Welsh, M., Scherberg, N., Gilmore, R. and Steiner, D.F. (1986) Biochem. J. 235, 459467. 11. Pipeleers,D.G. Pipeleers-Marichal, M. and Kipnis, D.M. (1976) Science 191, 88-90. 12. C&i, L., Amherdt, M., Malaisse-Lagae,F., Rouiller, C. and Renold, A.E. (1973) Science 179, 82-84. 13. Pipeleers, D.G., In ‘t Veld, P.A., Van De Winkel, M., Maes, E., Schuit, F.C. and Gepts, W. (1985) Endocrinology 117, 806-816. 14. Schuit, F.C., In ‘t Veld, P.A. and Pipeleers, D.G. (1988) Proc. Natl. Acad. Sci. (USA) 85, 3865-3869. 15. Berne, C. (1975) Endocrinology 97, 1241-1247. 16. Lomedico, P., Rosenthal, N., Efstratiadis, A., Gilbert, W., Kolodner, R. and Tizard, R. (1979) Cell 18,545-558. 17. Rhodes,C.J., Lucas, C.A. and Halban, P.A. (1987) FEBS Lett. 215, 179-182. 18. Salomon, D. and Meda, P. (1986) Exp. Cell Res. 162, 507-520. 19. Schuit, F.C. and Pipeleers,D.G. (1986) Science 232, 875-877. 20. Pipeleers,D.G. (1987) Diabetologia 30, 277-291.
1187
