- Email: [email protected]
Alteration of anomeric preference of glucose-induced insulin secretion by glyceraldehyde
Life Sciences, Vol. 53, pp. 1373-1382 Printed in the USA
Pergamon Press
ALTERATION OF ANOMERIC PREFERENCE OF GLUCOSEINDUCED INSULIN SECRETION BY GLYCERALDEHYDE T. Murata, I. Miwa, and J. Okuda Department of Clinical Biochemistry, Faculty of Pharmacy, Meijo University, Tempaku-ku, Nagoya 468, Japan (Received in final form August 23, 1993)
Summary Glucokinase activity in pancreatic islets was dose-dependently inactivated by D-glyceraldehyde, whereas islet hexokinase activity was not altered. In untreated islets, cz-D-glucose stimulated insulin secretion more efficiently than ~-D-glucose at a glucose concentration of 10 mM. However, glyceraldehyde highly attenuated the insulinsecretory response of pancreatic islets to cz-D-glucose compared with that to 13-D-glucose. Thus, there was apparently no anomeric preference of glucose-induced insulin secretion in glyceraldehydetreated islets. Glyceraldehyde affected neither the ct-preference of glucose phosphorylation by glucokinase nor the 13-preference of glucose phosphorylation by hexokinase. Our study suggests that defective discrimination of glucose anomers by glyceraldehydetreated islets may be caused by inactivation of glucokinase. It is widely known that the tx anomer of glucose is a better stimulant for insulin secretion than the 13anomer at physiological glucose concentrations, i.e. below about 15 mM (1-5). From the kinetic analysis of glucokinase with glucose anomers, the o,-preferential insulinotropism of glucose has been thought to be due to the o~-anomeric preference of glucokinase (6, 7). We recently reported that glucokinase is responsible for the anomeric preference of both glucose utilization and glucose-induced insulin secretion in pancreatic islets both at physiological and at higher glucose concentrations up to 60 mM (8). We also found that glucose-induced insulin secretion is impaired by glyceraldehyde through inactivation of glucokinase (9). In the present study, we examined the effect of glyceraldehyde on the insulin-secretory response of pancreatic islets to Author for correspondence: Ichitomo Miwa, Ph.D., Department of Clinical Biochemistry, Faculty of Pharmacy, Meijo University, Tempaku-ku, Nagoya 468, Japan 0024-3205/93 $6.00 + .00 Copyright © 1993 Pergamon Press Ltd All rights reserved.
1374
Glucose Anomeric Preference in Insulin Secretion
Vol. 53, No. 17, 1993
glucose anomers. Since liver glucokinase is very similar to islet glucokinase in various properties (10) and is easy to obtain in large amounts, we used rat liver glucokinase instead of islet glucokinase in this study. Rat erythrocyte hexokinase was used instead of islet hexokinase, since both are the type I isoform
(11). Methods Treatment of oancreatic islets. Pancreatic islets were isolated from nonfasted female Wistar rats (weighing 300-350 g, Clea Japan, Tokyo, Japan) according to the method described previously (12). Three kinds of batch sizes of islets were used: batches of about 250 islets for the kinetic analysis of glucokinase and hexokinase; batches of 20 islets for the determination of glyceraldehyde-concentration dependency of inactivation of glucokinase and hexokinase; batches of 5 islets for the assay of insulin secretion induced by glucose anomers. These batches were incubated for 1 h at 37 °C in 1 ml of buffer A (118 mM NaC1, 4.7 mM KCI, 2.5 mM CaC12, 1.2 mM KH2PO4, 1.2 mM MgSO4, 24.2 mM NaHCO3, 0.2 % bovine serum albumin, and 2.8 mM glucose [pH 7.4] ) supplemented with 6-18 mM glyceraldehyde (Aldrich, Milwaukee, WI, USA), washed with buffer A at room temperature, and then further incubated for 30 min at 37°C in 1 ml of buffer A in an attempt to eliminate the possible direct influence of glyceraldehyde on subsequent assays for glucose phosphorylation and glucose-induced insulin secretion. All the incubations were performed in an atmosphere consisting of a mixture of 95% 02 and 5% CO2. After the treatment, the islets were immediately used for the assay of glucose phosphorylation or insulin secretion. Assavs for islet hexokinase and glucokinase. Pancreatic islets were sonicated according to the method of Trus et al. (13), and the supernatant was obtained by centrifuging the sonicate at 12,000 Xg for 15 min. E n z y m e activities were assayed fluorometrically by a modification of the method of Trus et al. (13). Glucose phosphorylation was conducted in 300 B1 of 50 mM Hepes-NaOH buffer (pH 7.6) containing 100 mM KC1, 8 mM MgC12, 0.5 mM NAD, 5 mM ATP, 1 mM dithiothreitol, 0.5 mg/ml bovine serum albumin, 1 unit/ml glucose6-phosphate dehydrogenase, islet extract, and glucose. Incubations were performed for 90 min at 30 °C and stopped by addition of 150 B1 of 500 mM sodium phosphate buffer (pH 8.0) containing 0.9 mM sodium dodecylsulfate. The Vmax and Km of hexokinase were determined by the Hanes-Woolf plot with five levels of glucose concentration over a range of 0.03-0.5 mM; and those of glucokinase, with five levels of glucose concentration over a range of 6-100 mM. The Vmax of hexokinase was subtracted from the velocities observed at glucose concentrations above 6 mM to give glucokinase activity. For examination of the concentration dependency of glyceraldehyde on glucose-phosphorylating activity, the hexokinase activity was measured at a glucose concentration of 0.5 mM, and the glucokinase activity was estimated as the difference between activities at 0.5 and 50 mM glucose, because the Km of each enzyme for glucose
Vol. 53, No. 17, 1993
Glucose Anomeric Preference in Insulin Secretion
1375
was not altered by treatment of pancreatic islets with glyceraldehyde, as described in RESULTS. Assav of insulin ~ecretion. Insulin secretion induced by glucose anomers was performed as described previously (8) except that two consecutive incubations (3 min each) of pancreatic islets were carried out in 1 ml, instead of 0.5 ml, of incubation medium. The two incubation media were pooled and stored at -70 °C until insulin measurement could be performed. Insulin was measured by an enzyme immunoassay kit from Mitsui (Tokyo, Japan), with human insulin as the standard. Purification of enzymes. Rat liver glucokinase was purified as described previously (14). Rat erythrocyte hexokinase was partially purified by chromatography on DE-52 (Whatman, Maidstone, England) as described elsewhere (15). Treatment of enzymes. Enzyme solutions (about 0.3 unit/ml) of rat liver glucokinase or rat erythrocyte hexokinase were prepared with buffer B (pH 7.0) consisting of 20 mM KH2PO4, 100 mM KCI, 1 mM MgCI2, 1 mM EDTA, 1 mM dithiothreitol, 5 % (v/v) glycerol, and 1 mg/ml bovine serum albumin. An aliquot (90 ~tl) of the enzyme solution was preincubated for 1 min at 37 °C, and incubation was started by addition of 10 ~tl of glyceraldehyde solution (120 mM). Incubations were performed for 1 h at 37 °C. The reaction mixture was taken out and immediately used for the assay of phosphorylation of glucose anomers, since glyceraldehyde contained in the reaction mixture did not affect glucokinase and hexokinase activities in subsequent assays. Control experiments were conducted in parallel with test experiments by replacing the glyceraldehyde solution with buffer B. Phosphorylation of ~lucose anomers. Phosphorylation of glucose anomers by liver glucokinase and erythrocyte hexokinase was determined fluorometrically by a modification of the method of Meglasson and Matschinsky (6). Briefly, 5 ~tl of the reaction mixture described above was added to 85 ~tl of 59 mM HepesNaOH buffer (pH 7.6) containing 118 mM KCI, 9.4 mM MgC12, 5.9 mM ATP, 1.2 m M d i t h i o t h r e i t o l , and 0.6 m g / m l b o v i n e s e r u m albumin. After preincubation for 1 min at 37 °C, glucose phosphorylation was started by addition of 10 ~tl of glucose anomer solution (100 mM), prepared with ice-cold water just before use. Incubations were performed for 3 min at 37 °C and stopped by addition of 10 I.tl of 0.75 M HC1. After the pH had been adjusted to a neutral range by addition of 10 ~tl of 800 mM Hepes-NaOH buffer (pH 7.6) containing 0.75 M NaOH, the formation of NADH was started by addition of 10 l-tl of 50 mM Hepes-NaOH buffer (pH 7.6) containing 6.5 mM NAD, 13 units/ml glucose-6-phosphate dehydrogenase (from Leucon0stoc mesenteroides, Oriental Yeast, Osaka, Japan), and 0.5 mg/ml bovine serum albumin. The formation was conducted for 10 min at 37 °C and terminated by addition of 410 I.tl of 500 mM sodium phosphate buffer (pH 8.0) containing 0.4 mM sodium dodecylsulfate.
1376
Glucose Anomeric Preference in Insulin Secretion
Vol. 53, No. 17, 1993
Assay of islet DNA. The content of DNA in pancreatic islets was assayed by fluorometry according to the method of Labarca and Paigen (16). The islet sonicate prepared by the method of Trus et al. (13) was used for the assay of DNA. Data analysis. Statistical analyses were performed using the unpaired, twotailed Student's t-test. Results Table I shows kinetic characteristics of glucokinase and hexokinase in untreated and glyceraldehyde-treated islets. The Vmax of glucose phosphorylation by glucokinase was reduced to 58% by treatment of pancreatic islets with 12 mM glyceraldehyde for 1 h at 37 °C, while the Vmax of glucose phosphorylation by hexokinase was unaffected by the treatment. The Km of each enzyme for glucose was not significantly different between untreated and glyceraldehydetreated islets.
TABLE I Kinetic Characteristics of Glucokinase and Hexokinase in Untreated and Glyceraldehyde-Treated Islets
Untreated islets
Glyceraldehydetreated islets
Glucokinase Vmax (mol glucose/ 90 min/kg DNA) Km (mM)
2.98 _ 0.10 11.5 _+0.61
1.72 _+0.13" 12.3 _+ 0.52
Hexokinase Vmax (mol glucose/ 90 min/kg DNA) Km ~ M )
1.89 _ 0.09 38.3 _+6.2
2.01 __.0.08 40.1 _+ 5.7
Pancreatic islets were incubated with 12 mM glyceraldehyde for 1 h at 37 °C, washed with buffer A for 30 min at 37 °C, and then assayed for glucokinase and hexokinase activities with various concentrations of glucose. Data are means _ SEM for 4 experiments. *P< 0.001 vs. untreated islets.
Vol. 53, No. 17, 1993
Glucose Anomeric Preference in Insulin Secretion
1377
Glyceraldehyde inactivated islet glucokinase in a concentration-dependent manner when pancreatic islets were incubated with glyceraldehyde (0-18 mM) for 1 h at 37 °C (Fig. 1). Islet hexokinase activity was not, however, altered by glyceraldehyde at any of the concentrations tested. In untreated islets, the o~ anomer of glucose was more effective than the 13 anomer in stimulating insulin secretion at a glucose concentration of 10 mM ( Fig. 2). When pancreatic islets were treated with 12 mM glyceraldehyde for 1 h at 37 °C, t~- and ~D-glucose-induced insulin secretion was inhibited by 53% and 30%, respectively (Fig. 2). Thus, there was apparently no anomeric preference of glucose-induced insulin secretion in glyceraldehyde-treated islets at a physiological glucose concentration (10 mM).
100
6o o
_
_
>
•.~
40
E
20
O
a E
LLI 0
I
0
I
6
1'2
18
Glyceraldehyde(mM) Fig. 1 Effect of glyceraldehyde concentration on glucokinase and hexokinase activities in pancreatic islets. Pancreatic islets were incubated with the indicated concentrations of glyceraldehyde for 1 h at 37 °C, washed with buffer A for 30 min at 37 °C, and then assayed for glucokinase ( 0 ) and hexokinase (O) activities. Data are mean percentages _ SEM of the enzyme activities (glucokinase, 56 _.+2.6 pmol/90 m i n / i s l e t ; hexokinase, 69 __.4.1 pmol/90 min/islet) in control experiments; n=4.
1378
Glucose Anomeric Preference in Insulin Secretion
I
Vol. 53, No. 17, 1993
P<0.001
200 [
P<0.001
I
03
.=_ E
150 P<0.05
I
¢,D O
E v
100 -
tO
im
~,'//.//'///'//,
.m
m
I
T
,/z//zzzzt/ t'fff//fJ///
f/#//"//J///
~444444~-";4": gZdd.'f.'f_dZZ/_.
50-
//////////.
03 t'-
~/////////~
~'///////.~
o~
03 c-
NS
¢/////1///t
_
r/f_1_i/mzffj .!f!!!F~!!~
a-glucose
I]-glucose
Untreated islets
c~-glucose
13-glucose
Glyceraldehydetreated islets
Fig. 2 Effect of glyceraldehyde on insulin-secretory response of pancreatic islets to either tx- or ~D-glucose. Pancreatic islets were treated with 12 mM glyceraldehyde for 1 h at 37 °C and then washed with buffer A for 30 rain at 37 °C. Incubations for insulin s e c r e t i o n were performed in the presence of 10 mM ct- or ~-D-glucose. Each value is expressed by subtraction of the basal insulin secretion (120 _+ 9.7 and 156 _ 8.7 fmol/6 min/islet in untreated and g l y c e r a l d e h y d e treated islets, respectively) from the observed insulin secretion. Data are means + SEM for 6-8 experiments. NS, not significant.
Liver glucokinase phosphorylated tx-D-glucose more rapidly than ~D-glucose at a glucose concentration of 10 mM (Table II). Glyceraldehyde treatment of glucokinase lowered the phosphorylation rates of both cx- and I~-D-glucose to a similar extent (45 and 41% inhibition, respectively), indicating that glyceraldehyde-treated glucokinase also displayed a preference for the tx anomer of glucose. Erythrocyte hexokinase phosphorylated I]-D-glucose more rapidly than o~-D-glucose (Table II). Glyceraldehyde treatment did not affect either the activity or the anomeric preference of hexokinase.
55 _.+6*
Glyceraldehyde
40 _ 4*
73 _ 2 1.38
1.37
99 _ 7
100 _+5
126 __.7
125 _ 5
1.27
1.25
Ratio (vl~/va)
Each enzyme was incubated with 12 mM glyceraldehyde for 1 h at 37 °C and then assayed for phosphorylation of glucose anomers (10 mM). Data are mean percentages _ SEM of the rate of phosphorylation of (x-Dglucose by glucokinase and hexokinase in control experiments; n=5. The vct and vl~ refer to the velocity of phosphorylation of ct- and 13-D-glucose, respectively. There was a significant difference (P<0.001) in phosphorylation rate between the (~ and 13anomers of glucose in all of the four pairs. *P < 0.001, vs. control.
100_+ 3
13-Anomer
(x-Anomer
(va/vl~)
o~-Anomer 13-Anomer Ratio
Control
Treatment
Phosphorylation by hexokinase
Phosphorylation by glucokinase
TABLE II Effect of Glyceraldehyde on Phosphorylation of Glucose Anomers by Liver Glucokinase and Erythrocyte Hexokinase
<
ca
ca
ca
O
ca
z
7~ 9
O
1380
Glucose Anomeric Preference in Insulin Secretion
aldose reductase Glucose
sorbitol dehydrogenase
• Sorbitol
Vol. 53, No. 17, 1993
ketohexokinase
> Fructose
aldolase B
D-Glyceraldehyde
Fructose 1-phosphate
pl~?dr~xyacetOne
Fig. 3 Possible pathway of D-glyceraldehyde formation in pancreatic islets.
Discussion The o~-anomeric preference in glucose-induced insulin secretion at physiological glucose concentrations (below 15 mM) has been accounted for by the txanomeric preference of glucose phosphorylation by glucokinase (6-8). Because glyceraldehyde had no effect on the ~-anomeric preference of glucokinase (Table II), the perturbation of the anomeric preference of insulinotropism of glucose in pancreatic islets by treatment with glyceraldehyde cannot be explained by alteration of the stereospecificity of glucokinase. Glucose phosphorylation in pancreatic islets is catalyzed not only by glucokinase but also by hexokinase. Hexokinase activity, however, is considered to be substantially inhibited by hexose phosphates such as glucose 6-phosphate and glucose 1, 6-bisphosphate in intact islets (17). Since the deterioration in islet glycolysis through inactivation of glucokinase by glyceraldehyde could lead to a decrease in the concentration of hexose phosphates, the inhibition of hexokinase activity by hexose phosphates would diminish in glyceraldehyde-treated islets. Consequently, the ratio of hexokinase activity to glucokinase activity might increase in glyceraldehyde-treated islets. This increase might cause the disappearance of the anomeric preference of glucose-induced insulin secretion. It seems possible that glyceraldehyde is produced from glucose in islet [3-cells by the pathway shown in Fig. 3, since all of the enzymes in the pathway have been reported to exist in pancreatic islets (9). In the diabetic state, fructose formation through the polyol pathway is increased in many tissues (18). We therefore suggested the possibility that the production of glyceraldehyde in islet ~cells may be accelerated by chronic hyperglycemia (9). The above-mentioned pathway also produces fructose 1-phosphate. Van Schaftingen (19) reported that glucokinase in hepatocytes is inhibited by its regulatory protein in the pres-
Vol. 53, No. 17, 1993
Glucose Anomeric Preference in Insulin Secretion
1381
ence of fructose 6-phosphate and that this inhibition is relieved by fructose 1phosphate. The glucokinase activity in pancreatic islets may be regulated in a manner similar to that operating in hepatocytes, but with lower efficiency (20). Thus, even if fructose 1-phosphate in islet 13-cells is increased by chronic hyperglycemia, it might not appreciably affect the activity of islet glucokinase. The loss of the o~-anomeric preference of glucose-induced insulin secretion has been observed in non-insulin-dependent diabetic patients (21) and experimental animal models of diabetes (22-25). An explanation for the cause has been proposed (26), but it has not yet been fully verified. Our view on the production of glyceraldehyde from glucose suggests that it would be worthwhile to elucidate whether glucokinase inactivation by glyceraldehyde may participate in such alterations induced by chronic hyperglycemia. Acknowledgment This study was supported in part by a grant-in-aid for scientific research (No. 04671377) from the Ministry of Education, Science, and Culture of Japan. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
A. NIKI, H. NIKI, I. MIWA, and J. OKUDA, Science 186150-151 (1974). G.M. GRODSKY, R. FANSKA, L. WEST, and M. MANNING, Science 186536-538 (1974). A.A. ROSSINI, J.S. SOELDNER, J.M. HIEBERT, G.C. WEIR, and R.E. GLEASON, Diabetologia 10795-799 (1974). G.M. GRODSKY, R. FANSKA, and I. LUNDQUIST, Endocrinology 9__7_7 573-580 (1975). F.M. MATSCHINSKY, A.S. PAGLIARA, B.A. HOVER, M.W. HAYMOND, and S.N. STILLINGS, Diabetes 24369-372 (1975). M.D. MEGLASSON and F.M. MATSCHINSKY, J. Biol. Chem. 258 6705- 6708 (1983). I. MIWA, K. INAGAKI, and J. OKUDA, Biochem. Int. 7449-454 (1983). I. MIWA, T. MURATA, and J. OKUDA, Biochem. Biophys. Res. Commun. 180709-715 (1991). T. MURATA, I. MIWA, Y. TOYODA, and J. OKUDA, Diabetes 4_.22 1003-1009 (1993). M.D. MEGLASSON, P.T. BURCH, D.K. BERNER, H. NAJAFI, A.P. VOGIN, and F.M. MATSCHINSKY, Proc. Natl. Acad. Sci. USA 8___008589 (1983). T. SHIMIZU, B.B. KNOWLES, and F.M. MATSCHINSKY, Diabetes 3__7_7 563-568 (1988). I. MIWA, T. MURATA, S. MITSUYAMA, and J. OKUDA, Diabetes 3___99 1170-1176 (1990).
1382
13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26.
Glucose Anomeric Preference in Insulin Secretion
Vol. 53, No. 17, 1993
M.D. TRUS, W.S. ZAWALICH, P.T. BURCH, D.K. BERNER, V.A. WEILL, and F.M. MATSCHINSKY, Diabetes 30911-922 (1981). I. MIWA, S. MITSUYAMA, Y. TOYODA, T. MURATA, and J. OKUDA, Prep. Biochem. 2_.0 163-178 (1990). Y.-J. DUAN, H. FUKATSU, I. MIWA, and J. OKUDA, Mol. Cell. Biochem. 112 23-28 (1992). C. L A B A R C A and K. PAIGEN, Anal. Biochem. 102 344-352 (1980). M.-H. GIROIX, A. SENER, D.G. PIPELEERS, and W.J. MALAISSE, Biochem. J. 223 447-453 (1984). K.H. GABBAY, N. Engl. J. Med. 288 831-836 (1973). E. VAN SCHAFTINGEN, Eur. J. Biochem. 179 179-184 (1989). W.J. MALAISSE, F. MALAISSE-LAGAE, D.R. DAVIES, A. VANDERCAMMEN, and E. VAN SCHAFTINGEN, Eur. J. Biochem. 190 539-545 (1990). A. ROVIRA, F.J. GARROTE, I. VALVERDE, and W.J. MALAISSE, Diabetes Res. 5 119-124 (1987). V. LECLERCQ-MEYER, J. MARCHAND, and W.J. MALAISSE, Med. Sci. Res. 15 1535-1536 (1987). A. NIKI, H. NIKI, I. NIKI, and Y. KUNOH, Diabetologia 3__!_165-67 (1988). A. NIKI, H. NIKI, T. HASHIOKA, I. NIKI, K. SUZUKI, and Y. GOTO, Diabetes Res. Clin. Pract. 7 $87-$92 (1989). F. FICHAUX, J. MARCHAND, B. YAYLALI, V. LECLERCQ-MEYER, J. CATALA, and W.J. MALAISSE, Int. J. Pancreatol. 8 151-167 (1991). W.J. MALAISSE, Horm. Metab. Res. 2__33307-311 (1991).
Pergamon Press
ALTERATION OF ANOMERIC PREFERENCE OF GLUCOSEINDUCED INSULIN SECRETION BY GLYCERALDEHYDE T. Murata, I. Miwa, and J. Okuda Department of Clinical Biochemistry, Faculty of Pharmacy, Meijo University, Tempaku-ku, Nagoya 468, Japan (Received in final form August 23, 1993)
Summary Glucokinase activity in pancreatic islets was dose-dependently inactivated by D-glyceraldehyde, whereas islet hexokinase activity was not altered. In untreated islets, cz-D-glucose stimulated insulin secretion more efficiently than ~-D-glucose at a glucose concentration of 10 mM. However, glyceraldehyde highly attenuated the insulinsecretory response of pancreatic islets to cz-D-glucose compared with that to 13-D-glucose. Thus, there was apparently no anomeric preference of glucose-induced insulin secretion in glyceraldehydetreated islets. Glyceraldehyde affected neither the ct-preference of glucose phosphorylation by glucokinase nor the 13-preference of glucose phosphorylation by hexokinase. Our study suggests that defective discrimination of glucose anomers by glyceraldehydetreated islets may be caused by inactivation of glucokinase. It is widely known that the tx anomer of glucose is a better stimulant for insulin secretion than the 13anomer at physiological glucose concentrations, i.e. below about 15 mM (1-5). From the kinetic analysis of glucokinase with glucose anomers, the o,-preferential insulinotropism of glucose has been thought to be due to the o~-anomeric preference of glucokinase (6, 7). We recently reported that glucokinase is responsible for the anomeric preference of both glucose utilization and glucose-induced insulin secretion in pancreatic islets both at physiological and at higher glucose concentrations up to 60 mM (8). We also found that glucose-induced insulin secretion is impaired by glyceraldehyde through inactivation of glucokinase (9). In the present study, we examined the effect of glyceraldehyde on the insulin-secretory response of pancreatic islets to Author for correspondence: Ichitomo Miwa, Ph.D., Department of Clinical Biochemistry, Faculty of Pharmacy, Meijo University, Tempaku-ku, Nagoya 468, Japan 0024-3205/93 $6.00 + .00 Copyright © 1993 Pergamon Press Ltd All rights reserved.
1374
Glucose Anomeric Preference in Insulin Secretion
Vol. 53, No. 17, 1993
glucose anomers. Since liver glucokinase is very similar to islet glucokinase in various properties (10) and is easy to obtain in large amounts, we used rat liver glucokinase instead of islet glucokinase in this study. Rat erythrocyte hexokinase was used instead of islet hexokinase, since both are the type I isoform
(11). Methods Treatment of oancreatic islets. Pancreatic islets were isolated from nonfasted female Wistar rats (weighing 300-350 g, Clea Japan, Tokyo, Japan) according to the method described previously (12). Three kinds of batch sizes of islets were used: batches of about 250 islets for the kinetic analysis of glucokinase and hexokinase; batches of 20 islets for the determination of glyceraldehyde-concentration dependency of inactivation of glucokinase and hexokinase; batches of 5 islets for the assay of insulin secretion induced by glucose anomers. These batches were incubated for 1 h at 37 °C in 1 ml of buffer A (118 mM NaC1, 4.7 mM KCI, 2.5 mM CaC12, 1.2 mM KH2PO4, 1.2 mM MgSO4, 24.2 mM NaHCO3, 0.2 % bovine serum albumin, and 2.8 mM glucose [pH 7.4] ) supplemented with 6-18 mM glyceraldehyde (Aldrich, Milwaukee, WI, USA), washed with buffer A at room temperature, and then further incubated for 30 min at 37°C in 1 ml of buffer A in an attempt to eliminate the possible direct influence of glyceraldehyde on subsequent assays for glucose phosphorylation and glucose-induced insulin secretion. All the incubations were performed in an atmosphere consisting of a mixture of 95% 02 and 5% CO2. After the treatment, the islets were immediately used for the assay of glucose phosphorylation or insulin secretion. Assavs for islet hexokinase and glucokinase. Pancreatic islets were sonicated according to the method of Trus et al. (13), and the supernatant was obtained by centrifuging the sonicate at 12,000 Xg for 15 min. E n z y m e activities were assayed fluorometrically by a modification of the method of Trus et al. (13). Glucose phosphorylation was conducted in 300 B1 of 50 mM Hepes-NaOH buffer (pH 7.6) containing 100 mM KC1, 8 mM MgC12, 0.5 mM NAD, 5 mM ATP, 1 mM dithiothreitol, 0.5 mg/ml bovine serum albumin, 1 unit/ml glucose6-phosphate dehydrogenase, islet extract, and glucose. Incubations were performed for 90 min at 30 °C and stopped by addition of 150 B1 of 500 mM sodium phosphate buffer (pH 8.0) containing 0.9 mM sodium dodecylsulfate. The Vmax and Km of hexokinase were determined by the Hanes-Woolf plot with five levels of glucose concentration over a range of 0.03-0.5 mM; and those of glucokinase, with five levels of glucose concentration over a range of 6-100 mM. The Vmax of hexokinase was subtracted from the velocities observed at glucose concentrations above 6 mM to give glucokinase activity. For examination of the concentration dependency of glyceraldehyde on glucose-phosphorylating activity, the hexokinase activity was measured at a glucose concentration of 0.5 mM, and the glucokinase activity was estimated as the difference between activities at 0.5 and 50 mM glucose, because the Km of each enzyme for glucose
Vol. 53, No. 17, 1993
Glucose Anomeric Preference in Insulin Secretion
1375
was not altered by treatment of pancreatic islets with glyceraldehyde, as described in RESULTS. Assav of insulin ~ecretion. Insulin secretion induced by glucose anomers was performed as described previously (8) except that two consecutive incubations (3 min each) of pancreatic islets were carried out in 1 ml, instead of 0.5 ml, of incubation medium. The two incubation media were pooled and stored at -70 °C until insulin measurement could be performed. Insulin was measured by an enzyme immunoassay kit from Mitsui (Tokyo, Japan), with human insulin as the standard. Purification of enzymes. Rat liver glucokinase was purified as described previously (14). Rat erythrocyte hexokinase was partially purified by chromatography on DE-52 (Whatman, Maidstone, England) as described elsewhere (15). Treatment of enzymes. Enzyme solutions (about 0.3 unit/ml) of rat liver glucokinase or rat erythrocyte hexokinase were prepared with buffer B (pH 7.0) consisting of 20 mM KH2PO4, 100 mM KCI, 1 mM MgCI2, 1 mM EDTA, 1 mM dithiothreitol, 5 % (v/v) glycerol, and 1 mg/ml bovine serum albumin. An aliquot (90 ~tl) of the enzyme solution was preincubated for 1 min at 37 °C, and incubation was started by addition of 10 ~tl of glyceraldehyde solution (120 mM). Incubations were performed for 1 h at 37 °C. The reaction mixture was taken out and immediately used for the assay of phosphorylation of glucose anomers, since glyceraldehyde contained in the reaction mixture did not affect glucokinase and hexokinase activities in subsequent assays. Control experiments were conducted in parallel with test experiments by replacing the glyceraldehyde solution with buffer B. Phosphorylation of ~lucose anomers. Phosphorylation of glucose anomers by liver glucokinase and erythrocyte hexokinase was determined fluorometrically by a modification of the method of Meglasson and Matschinsky (6). Briefly, 5 ~tl of the reaction mixture described above was added to 85 ~tl of 59 mM HepesNaOH buffer (pH 7.6) containing 118 mM KCI, 9.4 mM MgC12, 5.9 mM ATP, 1.2 m M d i t h i o t h r e i t o l , and 0.6 m g / m l b o v i n e s e r u m albumin. After preincubation for 1 min at 37 °C, glucose phosphorylation was started by addition of 10 ~tl of glucose anomer solution (100 mM), prepared with ice-cold water just before use. Incubations were performed for 3 min at 37 °C and stopped by addition of 10 I.tl of 0.75 M HC1. After the pH had been adjusted to a neutral range by addition of 10 ~tl of 800 mM Hepes-NaOH buffer (pH 7.6) containing 0.75 M NaOH, the formation of NADH was started by addition of 10 l-tl of 50 mM Hepes-NaOH buffer (pH 7.6) containing 6.5 mM NAD, 13 units/ml glucose-6-phosphate dehydrogenase (from Leucon0stoc mesenteroides, Oriental Yeast, Osaka, Japan), and 0.5 mg/ml bovine serum albumin. The formation was conducted for 10 min at 37 °C and terminated by addition of 410 I.tl of 500 mM sodium phosphate buffer (pH 8.0) containing 0.4 mM sodium dodecylsulfate.
1376
Glucose Anomeric Preference in Insulin Secretion
Vol. 53, No. 17, 1993
Assay of islet DNA. The content of DNA in pancreatic islets was assayed by fluorometry according to the method of Labarca and Paigen (16). The islet sonicate prepared by the method of Trus et al. (13) was used for the assay of DNA. Data analysis. Statistical analyses were performed using the unpaired, twotailed Student's t-test. Results Table I shows kinetic characteristics of glucokinase and hexokinase in untreated and glyceraldehyde-treated islets. The Vmax of glucose phosphorylation by glucokinase was reduced to 58% by treatment of pancreatic islets with 12 mM glyceraldehyde for 1 h at 37 °C, while the Vmax of glucose phosphorylation by hexokinase was unaffected by the treatment. The Km of each enzyme for glucose was not significantly different between untreated and glyceraldehydetreated islets.
TABLE I Kinetic Characteristics of Glucokinase and Hexokinase in Untreated and Glyceraldehyde-Treated Islets
Untreated islets
Glyceraldehydetreated islets
Glucokinase Vmax (mol glucose/ 90 min/kg DNA) Km (mM)
2.98 _ 0.10 11.5 _+0.61
1.72 _+0.13" 12.3 _+ 0.52
Hexokinase Vmax (mol glucose/ 90 min/kg DNA) Km ~ M )
1.89 _ 0.09 38.3 _+6.2
2.01 __.0.08 40.1 _+ 5.7
Pancreatic islets were incubated with 12 mM glyceraldehyde for 1 h at 37 °C, washed with buffer A for 30 min at 37 °C, and then assayed for glucokinase and hexokinase activities with various concentrations of glucose. Data are means _ SEM for 4 experiments. *P< 0.001 vs. untreated islets.
Vol. 53, No. 17, 1993
Glucose Anomeric Preference in Insulin Secretion
1377
Glyceraldehyde inactivated islet glucokinase in a concentration-dependent manner when pancreatic islets were incubated with glyceraldehyde (0-18 mM) for 1 h at 37 °C (Fig. 1). Islet hexokinase activity was not, however, altered by glyceraldehyde at any of the concentrations tested. In untreated islets, the o~ anomer of glucose was more effective than the 13 anomer in stimulating insulin secretion at a glucose concentration of 10 mM ( Fig. 2). When pancreatic islets were treated with 12 mM glyceraldehyde for 1 h at 37 °C, t~- and ~D-glucose-induced insulin secretion was inhibited by 53% and 30%, respectively (Fig. 2). Thus, there was apparently no anomeric preference of glucose-induced insulin secretion in glyceraldehyde-treated islets at a physiological glucose concentration (10 mM).
100
6o o
_
_
>
•.~
40
E
20
O
a E
LLI 0
I
0
I
6
1'2
18
Glyceraldehyde(mM) Fig. 1 Effect of glyceraldehyde concentration on glucokinase and hexokinase activities in pancreatic islets. Pancreatic islets were incubated with the indicated concentrations of glyceraldehyde for 1 h at 37 °C, washed with buffer A for 30 min at 37 °C, and then assayed for glucokinase ( 0 ) and hexokinase (O) activities. Data are mean percentages _ SEM of the enzyme activities (glucokinase, 56 _.+2.6 pmol/90 m i n / i s l e t ; hexokinase, 69 __.4.1 pmol/90 min/islet) in control experiments; n=4.
1378
Glucose Anomeric Preference in Insulin Secretion
I
Vol. 53, No. 17, 1993
P<0.001
200 [
P<0.001
I
03
.=_ E
150 P<0.05
I
¢,D O
E v
100 -
tO
im
~,'//.//'///'//,
.m
m
I
T
,/z//zzzzt/ t'fff//fJ///
f/#//"//J///
~444444~-";4": gZdd.'f.'f_dZZ/_.
50-
//////////.
03 t'-
~/////////~
~'///////.~
o~
03 c-
NS
¢/////1///t
_
r/f_1_i/mzffj .!f!!!F~!!~
a-glucose
I]-glucose
Untreated islets
c~-glucose
13-glucose
Glyceraldehydetreated islets
Fig. 2 Effect of glyceraldehyde on insulin-secretory response of pancreatic islets to either tx- or ~D-glucose. Pancreatic islets were treated with 12 mM glyceraldehyde for 1 h at 37 °C and then washed with buffer A for 30 rain at 37 °C. Incubations for insulin s e c r e t i o n were performed in the presence of 10 mM ct- or ~-D-glucose. Each value is expressed by subtraction of the basal insulin secretion (120 _+ 9.7 and 156 _ 8.7 fmol/6 min/islet in untreated and g l y c e r a l d e h y d e treated islets, respectively) from the observed insulin secretion. Data are means + SEM for 6-8 experiments. NS, not significant.
Liver glucokinase phosphorylated tx-D-glucose more rapidly than ~D-glucose at a glucose concentration of 10 mM (Table II). Glyceraldehyde treatment of glucokinase lowered the phosphorylation rates of both cx- and I~-D-glucose to a similar extent (45 and 41% inhibition, respectively), indicating that glyceraldehyde-treated glucokinase also displayed a preference for the tx anomer of glucose. Erythrocyte hexokinase phosphorylated I]-D-glucose more rapidly than o~-D-glucose (Table II). Glyceraldehyde treatment did not affect either the activity or the anomeric preference of hexokinase.
55 _.+6*
Glyceraldehyde
40 _ 4*
73 _ 2 1.38
1.37
99 _ 7
100 _+5
126 __.7
125 _ 5
1.27
1.25
Ratio (vl~/va)
Each enzyme was incubated with 12 mM glyceraldehyde for 1 h at 37 °C and then assayed for phosphorylation of glucose anomers (10 mM). Data are mean percentages _ SEM of the rate of phosphorylation of (x-Dglucose by glucokinase and hexokinase in control experiments; n=5. The vct and vl~ refer to the velocity of phosphorylation of ct- and 13-D-glucose, respectively. There was a significant difference (P<0.001) in phosphorylation rate between the (~ and 13anomers of glucose in all of the four pairs. *P < 0.001, vs. control.
100_+ 3
13-Anomer
(x-Anomer
(va/vl~)
o~-Anomer 13-Anomer Ratio
Control
Treatment
Phosphorylation by hexokinase
Phosphorylation by glucokinase
TABLE II Effect of Glyceraldehyde on Phosphorylation of Glucose Anomers by Liver Glucokinase and Erythrocyte Hexokinase
<
ca
ca
ca
O
ca
z
7~ 9
O
1380
Glucose Anomeric Preference in Insulin Secretion
aldose reductase Glucose
sorbitol dehydrogenase
• Sorbitol
Vol. 53, No. 17, 1993
ketohexokinase
> Fructose
aldolase B
D-Glyceraldehyde
Fructose 1-phosphate
pl~?dr~xyacetOne
Fig. 3 Possible pathway of D-glyceraldehyde formation in pancreatic islets.
Discussion The o~-anomeric preference in glucose-induced insulin secretion at physiological glucose concentrations (below 15 mM) has been accounted for by the txanomeric preference of glucose phosphorylation by glucokinase (6-8). Because glyceraldehyde had no effect on the ~-anomeric preference of glucokinase (Table II), the perturbation of the anomeric preference of insulinotropism of glucose in pancreatic islets by treatment with glyceraldehyde cannot be explained by alteration of the stereospecificity of glucokinase. Glucose phosphorylation in pancreatic islets is catalyzed not only by glucokinase but also by hexokinase. Hexokinase activity, however, is considered to be substantially inhibited by hexose phosphates such as glucose 6-phosphate and glucose 1, 6-bisphosphate in intact islets (17). Since the deterioration in islet glycolysis through inactivation of glucokinase by glyceraldehyde could lead to a decrease in the concentration of hexose phosphates, the inhibition of hexokinase activity by hexose phosphates would diminish in glyceraldehyde-treated islets. Consequently, the ratio of hexokinase activity to glucokinase activity might increase in glyceraldehyde-treated islets. This increase might cause the disappearance of the anomeric preference of glucose-induced insulin secretion. It seems possible that glyceraldehyde is produced from glucose in islet [3-cells by the pathway shown in Fig. 3, since all of the enzymes in the pathway have been reported to exist in pancreatic islets (9). In the diabetic state, fructose formation through the polyol pathway is increased in many tissues (18). We therefore suggested the possibility that the production of glyceraldehyde in islet ~cells may be accelerated by chronic hyperglycemia (9). The above-mentioned pathway also produces fructose 1-phosphate. Van Schaftingen (19) reported that glucokinase in hepatocytes is inhibited by its regulatory protein in the pres-
Vol. 53, No. 17, 1993
Glucose Anomeric Preference in Insulin Secretion
1381
ence of fructose 6-phosphate and that this inhibition is relieved by fructose 1phosphate. The glucokinase activity in pancreatic islets may be regulated in a manner similar to that operating in hepatocytes, but with lower efficiency (20). Thus, even if fructose 1-phosphate in islet 13-cells is increased by chronic hyperglycemia, it might not appreciably affect the activity of islet glucokinase. The loss of the o~-anomeric preference of glucose-induced insulin secretion has been observed in non-insulin-dependent diabetic patients (21) and experimental animal models of diabetes (22-25). An explanation for the cause has been proposed (26), but it has not yet been fully verified. Our view on the production of glyceraldehyde from glucose suggests that it would be worthwhile to elucidate whether glucokinase inactivation by glyceraldehyde may participate in such alterations induced by chronic hyperglycemia. Acknowledgment This study was supported in part by a grant-in-aid for scientific research (No. 04671377) from the Ministry of Education, Science, and Culture of Japan. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
A. NIKI, H. NIKI, I. MIWA, and J. OKUDA, Science 186150-151 (1974). G.M. GRODSKY, R. FANSKA, L. WEST, and M. MANNING, Science 186536-538 (1974). A.A. ROSSINI, J.S. SOELDNER, J.M. HIEBERT, G.C. WEIR, and R.E. GLEASON, Diabetologia 10795-799 (1974). G.M. GRODSKY, R. FANSKA, and I. LUNDQUIST, Endocrinology 9__7_7 573-580 (1975). F.M. MATSCHINSKY, A.S. PAGLIARA, B.A. HOVER, M.W. HAYMOND, and S.N. STILLINGS, Diabetes 24369-372 (1975). M.D. MEGLASSON and F.M. MATSCHINSKY, J. Biol. Chem. 258 6705- 6708 (1983). I. MIWA, K. INAGAKI, and J. OKUDA, Biochem. Int. 7449-454 (1983). I. MIWA, T. MURATA, and J. OKUDA, Biochem. Biophys. Res. Commun. 180709-715 (1991). T. MURATA, I. MIWA, Y. TOYODA, and J. OKUDA, Diabetes 4_.22 1003-1009 (1993). M.D. MEGLASSON, P.T. BURCH, D.K. BERNER, H. NAJAFI, A.P. VOGIN, and F.M. MATSCHINSKY, Proc. Natl. Acad. Sci. USA 8___008589 (1983). T. SHIMIZU, B.B. KNOWLES, and F.M. MATSCHINSKY, Diabetes 3__7_7 563-568 (1988). I. MIWA, T. MURATA, S. MITSUYAMA, and J. OKUDA, Diabetes 3___99 1170-1176 (1990).
1382
13. 14. 15. 16. 17. 18. 19. 20.
21. 22. 23. 24. 25. 26.
Glucose Anomeric Preference in Insulin Secretion
Vol. 53, No. 17, 1993
M.D. TRUS, W.S. ZAWALICH, P.T. BURCH, D.K. BERNER, V.A. WEILL, and F.M. MATSCHINSKY, Diabetes 30911-922 (1981). I. MIWA, S. MITSUYAMA, Y. TOYODA, T. MURATA, and J. OKUDA, Prep. Biochem. 2_.0 163-178 (1990). Y.-J. DUAN, H. FUKATSU, I. MIWA, and J. OKUDA, Mol. Cell. Biochem. 112 23-28 (1992). C. L A B A R C A and K. PAIGEN, Anal. Biochem. 102 344-352 (1980). M.-H. GIROIX, A. SENER, D.G. PIPELEERS, and W.J. MALAISSE, Biochem. J. 223 447-453 (1984). K.H. GABBAY, N. Engl. J. Med. 288 831-836 (1973). E. VAN SCHAFTINGEN, Eur. J. Biochem. 179 179-184 (1989). W.J. MALAISSE, F. MALAISSE-LAGAE, D.R. DAVIES, A. VANDERCAMMEN, and E. VAN SCHAFTINGEN, Eur. J. Biochem. 190 539-545 (1990). A. ROVIRA, F.J. GARROTE, I. VALVERDE, and W.J. MALAISSE, Diabetes Res. 5 119-124 (1987). V. LECLERCQ-MEYER, J. MARCHAND, and W.J. MALAISSE, Med. Sci. Res. 15 1535-1536 (1987). A. NIKI, H. NIKI, I. NIKI, and Y. KUNOH, Diabetologia 3__!_165-67 (1988). A. NIKI, H. NIKI, T. HASHIOKA, I. NIKI, K. SUZUKI, and Y. GOTO, Diabetes Res. Clin. Pract. 7 $87-$92 (1989). F. FICHAUX, J. MARCHAND, B. YAYLALI, V. LECLERCQ-MEYER, J. CATALA, and W.J. MALAISSE, Int. J. Pancreatol. 8 151-167 (1991). W.J. MALAISSE, Horm. Metab. Res. 2__33307-311 (1991).