CELL BIOCHEMISTRY AND FUNCTION
VOL.
9: 49-53 (1991)
Phosphorylation by Liver Glucokinase of D-Glucose Anomers at Anomeric Equilibrium DAGMAR ZAHNER AND WILLY J. MALAISSET Laboratory of Experimental Medicine, Brussels Free University, I15 Boulevard de Waterloo, B-1000 Brussels, Belgium
The relative contribution of each anomer of D-glucose to the overall phosphorylation rate of the hexose tested at anomeric equilibrium was examined in rat liver postmicrosomal supernatants under conditions aimed at characterizing the activity of glucokinase, with negligible interference of either hexokinase, N-acetyl-D-glucosamine kinase or glucose-6-phosphatase (acting as a phosphotransferase). Both at lo" and 30 "C, the relative contribution of each anomer was unaffected by the concentration of D-glucose. At both temperatures, the a/B ratio for the contribution of each anomer was slightly, but significantly, lower than the u/B ratio of anomer concentrations. These findings, which are consistent with the anomeric specificity of glucokinase in terms of affinity, cooperativity and maximal velocity, reveal that the preferred u-anomeric substrate for both glycogen synthesis and glycolysis is generated by glucokinase at a lower rate than is ~-~-glucose-6-phosphate. KEY WORDS--
Liver glucokinase; D-glucose anomers; D - ~ ~ U C Ophosphorylation. S~
INTRODUCTION Like many other enzymes involved in the early steps of D-glucose metabolism, liver glucokinase displays qualified anomeric specificity. For instance, the maximal velocity of the reaction catalysed by liver glucokinase was found by several investigators to be higher with /?-D-glucose than with a-~-glucose.'-~ On the other hand, the affinity of the enzyme for the a-anomer appears greater than for the f l - a n ~ m e r . It ' ~is not obvious, however, that the latter preference is sufficiently marked to be able to compensate for the difference in maximal velocity so that the phosphorylation rate of a-D-glucose could exceed that of /?-D-glucose at low concentrations of each a n ~ m e r . ' - ~Likewise, there seems to exist little anomeric specificity in terms of the cooperative behaviour of glucokinase, as estimated from the Hill number.'. The information so far available on the anomeric specificity of glucokinase is restricted in several further interrelated aspects. First, little attention has so far been paid to the influence of temperature upon the anomeric specificity of glucokinase. This issue should not be neglected, since it was recently tAddressee for correspondance. 0263-6484/9 1/010049-05%05.00 0 1991 by John Wiley & Sons, Ltd.
shown that, in the case of the low-K, hexokinase, a change in temperature may affect the anomeric specificity of the enzyme, in terms of both affinity and maximal velocity.' The latter finding made it possible to reconcile data collected at different temperatures by different investigators. Second, although the kinetics of D-glUCOSe phosphorylation by glucokinase are often analysed by means of a Hill plot, this classical equilibrium model of cooperativity is not fully adequate, since glucokinase is a monomeric protein with only one active site.6 Furthermore, it was recently emphasized that the Hill model cannot be applied to glucokinase when considering the anomeric heterogeneity of its ~ u b s t r a t e .Last, ~ virtually no information is available on the possible perturbation of the anomeric behaviour of glucokinase in insulinopenic situations, such as diabetes and starvation, cases in which other intrinsic kinetic properties of the enzyme were recently found to be pert~rbed.~". With these considerations in mind, the present study aimed at investigating, both at 10 "C and 30 "C,the relative contribution of each anomer to the overall phosphorylation rate by liver glucokinase of D-glucose tested at anomeric equilibrium.
50
D. ZAHNER AND W. J . M A L A I S E
For reasons already alluded to,' this information ''C]glucose (6.48 mM) or P-D-[U - ''C]glucose cannot be unambiguously derived from the experi- (1 1.52 mM), 10 pl of a solution of the homologous mental data so far available, which are all restricted unlabelled anomer used in increasing concentrato studies performed with each anomer tested tions (zero to either 317.5 mM a-D-glucose or separately. The present study takes advantage of 564.5 mM a-D-glucose) and 20 p1 of a solution of the preparation of pure labelled anomers," which the other anomer (5.76 to 288 mM 8-D-glucose or allow the fate of each anomer to be followed 3.24 to 162 mM a-D-glucose). All these solutions in a mixture of either labelled a-D-glucose and were freshly prepared in the reaction mixture (see above) and maintained at 4 "C. The final concenunlabelled /?-D-glucose or vice versa. tration of each anomer always reached an a/P ratio equal to 36/64. Taking into account the presence of MATERIALS AND METHODS D-glucose in the liver postmicrosomal fraction, the The a- and P-anomers of D-glucose were purchased final concentration of D-glucose ranged from 4 mM from Sigma (St Louis, MO). The a- and j?-anomzr to 102mM. The incubation was conducted over of D-[U - ''C]glucose were prepared as described 20 min at 10 "C or 5 min at 30 "C, being initiated by elsewhere.' the addition of the liver postmicrosomal fraction Samples of liver (1.0 g each) removed from fed (50 p1) and terminated by addition of 2.0 ml of iced female albino rats were homogenized, using a water. The samples were then frozen ( - 20 "C). On mechanical homogenizer (Braun, Melsungen, the next day, the separation of D-glucose and its FRG) with six passes at 1250 rpm, in 2.0 ml of a acidic metabolites was achieved by anion-exchange Hepes-NaOH buffer (20 mM, pH 7.4) containing chromatography.' KC1 (50 mM), sucrose (50 mM), mannitol(200 mM), All readings were corrected for the blank value EDTA (1 mM), dithioerythritol (1 mM), phenyl- recorded in the absence of liver extract. Such a methylsulfonylfluoride (1 mM) and antipain blank value was not affected by the concentration (10 pg ml- ';Sigma). After 10-min centrifugation at of unlabelled D-glucose and, relative to the total 1000 g and 4 "C, the postnuclear supernatant was radioactive content of each sample, averaged 3.7 & centrifuged for 60 min at 100000 g and 4 "C to 0.1 per cent and 3.8 & 0.2 per cent (n = 6 in both obtain a postmicrosomal supernatant, which con- cases) with labelled a- and a-D-glucose, respectained 1.93 0.09 (n = 6) mg of protein per 50 pl, tively. The total radioactive content of each sample corresponding to about 116 k 6 mg g-' liver wet averaged 460 27 x 1O3 cpm per sample (n = 12). weight. The postmicrosomal supernatant was then The protein content was measured by the dialysed for 180 min at 8 "C against a large volume method of Lowry et a1.,12using bovine albumin as (51) of a Hepes-NaOH buffer (20mM, pH 7-4) standard. containing KCl (50 mM) and dithioerythritol All results including those mentioned above are (1 mM). The D-glucose concentration of the liver presented as the mean ( k SEM) together with the extract, as measured by the glucose oxidase number of individual determinations (n). The stamethod, averaged 21.8 0.7 mM and 3.7 k 0.4 mM tistical significance of differences was assessed by before and after dialysis, respectively. The conta- use of Student's t-test. mination by the hexose present in the dialysed sample was taken into account in adjusting the RESULTS final D-glucose concentration of the assay medium. The labelled and unlabelled anomers of D-glu- At increasing concentrations of D-glucose (4 to cose were prepared in an iced Hepes-NaOH buffer 100 mM), the total phosphorylation rate of both (100 mM, pH 7.5) containing KCl(50 mM), MgCl, anomers ranged from 0.5 0.2 to 6.7 k (18 mM), dithioerythritol(5 mM), D-glucose 6-phos- 1.3 nmol min-' mg-' protein at 10 "C and from phate (6.8 mM) and ATP (56.2 mM). Since 40 pl of 2.1 & 0.3 to 25.7 & 2-8 nmol min-' mg-' protein this reaction mixture were eventually mixed with at 30 "C(n = 3 in all cases). As illustrated in Figure 50 p1 of the dialysed liver extract, the final concen- 1, such a total phosphorylation rate yielded, in a tration of ATP (25mM) was sufficient to avoid Hill plot, linear relationships at both temperatures. exhaustion of this substrate, even at high concen- As expected from prior observation^,^ the cooperativity of glucokinase and, even more so, its affinity, trations of D-glucose. The assay medium, prepared at 4 "C, contained as judged respectively from the Hill number and lop1 of a solution of labelled ~-D-[U- K,, appeared greater at high than low temperature,
'
*
*
51
GLUCOKINASE A N D D-GLUCOSE ANOMERS
10°C -
0
m >
>'
-0.5
--.
Y
L2.J
z
/
07 0
-
/
1
-1.0
/
-1.(
i Figure I . Hill plot for the phosphorylation Of D-glucose at 10 "C(left) (slope 1.033) and 30 "C(right) (slope 1.145). Mean values are derived from triplicate measurements collected with both anomers in three individual experiments. The concentration of D-glucose ([GI) is expressed as mM.
with mean 3O0C/10"C ratios of 1.11 (cooperativity) and 1.67 (affinity). The fractional contribution of a-D-glucose to the total phosphorylation rate was little affected by the concentration of the hexose, whether at 10°C or 30 "C (Figure 2). Pooling all available data, such a ratio averaged 33.7 & 0.5 per cent at 10 "C and 32.6 & 0.6 per cent at 30 "C ( n = 18). These two mean values were not significantly different from one another, but were both lower ( P < 0.001) than the relative abundance of the a-anomer, i.e. 36 per cent. DISCUSSION
The major purpose of the present study was to assess, at both 10 "C and 30 "C, the relative contribution of a- and P-D-glucose to the overall rate of hexose phosphorylation by liver glucokinase. Several aspects of the experimental design used for such a purpose deserve attention. First, preliminary experiments conducted with semi-purified rat liver glu~okinase'~ indicated that the amount
of enzyme required, in order to obtain a reaction velocity well above the blank value, would necessitate lyophilization of a large volume of material eluting from the DEAE-Trisacryl column. The inherent risk of a loss in enzymatic activity associated with such a procedure led us eventually to conduct the present experiments with freshly isolated liver postmicrosomal supernatants. Second, it was previously shown that, under the present experimental conditions, the contribution of hexokinase, of glucose-6-phosphatase (acting as a phosphotransferase) and of N-acetyl-D-glucosamine kinase to D-glucose phosphorylation is negligible.3s Third, the temperature and length of incubation were selected to minimize the extent of hexose an~merization.'~ Fourth, because of the contamination of the assay medium by D-glucose present in the liver extract, the respective concentrations of each anomer were fixed at values close to those found at anomeric equilibrium. Last, the amount of liver material and volume of medium used in each assay was selected to ensure sizeable reaction velocities (above the appropriate blank
52
D. ZXHNER AND
W. J. MALAISSE
anomer according to the following equation:'
4Or
I
o\"
30L
Y
0
I
I
10
30
I
100
Figure 2. Relative contribution of a-D-glucose to the overall rate of hexose phosphorylation at 10 "C(upper panel) and 30 "C (lower panel) in the presence of increasing concentrations of Dglucose (logarithmic scale) tested at anometric equilibrium. Mean values ( f SEM) are derived from triplicate measurements collected with each anomer in three individual experiments. The dotted line represents the relative abundance of the a-form (36 per cent).
value) whilst maintaining the cumulative amount of D-glucose 6-phosphate generated below 10 per cent of the initial amount of hexose present in each sample. The fractional contribution of a-D-glucose to the total phosphorylation rate was unaffected by either the temperature or concentration of the hexose. It was slightly, but significantly, lower than the relative abundance of the a-anomer. In the absence of cooperativity, the fractional contribution of each anomer to hexose phosphorylation is dictated by the ratio of affinity to maximal velocity for each
in which u, and up represent the reaction velocities for each anomer, V, and V, the corresponding maximal velocities, [a] and [PI the anomer concentrations, and K , and K , the Michaelis constants. Since the [ a ] / [ P ] ratio was kept constant in the present experiments, our results reinforce the view that the anomeric difference in maximal velocities is slightly more pronounced than the anomeric difference in affinity, resulting in a fractional contribution of the a-anomer slightly lower than its relative abundance at anomeric equilibrium. It could be objected that this reasoning ignores the phenomenon of cooperativity. However, the fact that the relative contribution of each anomer to the overall rate of hexose phosphorylation was indeed close to that expected from the reported K , and V,,, values for a- and /3-~-glucose'-~is precisely in good agreement with the knowledge that there is little if any anomeric difference in cooperativity.'. The fact that the relative contribution of each anomer to the total rate of hexose phosphorylation was not affected by the concentration of D-glucose provides further support to the latter concept. In conclusion, the present work reveals that, at anomeric equilibrium, the relative contribution of each D-glucose anomer to the total rate of hexose phosphorylation by glucokinase is close, albeit not identical, to their respective abundance. Since all the enzymes catalysing the conversion of D-glucose 6-phosphate to further metabolites display strict anomeric preferen~e'~ and in view of the recent concept of enzyme-to-enzyme substrate tunnelling,' it would appear that the preferential substrate for either glycogen synthesis or glycolysis, namely a-D-glucose 6-phosphate, when generated from D-glucose, is formed in the liver at a lower rate than P-D-glucose 6-phosphate, which is preferentially channelled into the less active-in terms of metabolic flow-pentose phosphate pathway. The possible advantage of such an arrangement in the hepatic regulation of D-glucose metabolism remains to be scrutinized.
REFERENCES 1.
Meglasson, M. D. and Matschinsky, F. M. (1983). Discrimination ofglucose anomers by glucokinase from liver and transplantable insulinoma. J . B i d . Chem., 258,6705-6708.
53
GLUCOKINASE AND D-GLUCOSE ANOMERS 2. Sener, A., Giroix, M.-H., Dufrane, S. P. and Malaise, W. J. (1985). Anomeric specificity of hexokinase and gluco-
10. Sener, A,, Leclercq-Meyer, V., Marchand, J., Giroix, M. H., Dufrane, S. P. and Malaise, W. J. (1985). Is glucokinase
kinase activities in liver and insulin-producing cells. Biochem. J., 230, 345-351. Zahner, D. and Malaise, W. J. (1990). Kinetic behaviour of liver glucokinase in diabetes. I. Alteration of glucokinase in streptozotocin-diabetic rats. Diab. Res., in press. Miwa, I., Inagaki, K. and Okuda, J. (1983). Preference of glucokinase for the a-anomer of hexose: relation to uanomer preference in hexose-induced insulin release by pancreatic islets. Biochem. Int., 7, 449-454. Malaise-Lagae, F., Giroix, M.-H., Sener, A. and Malaise, W. J. (1986). Temperature dependency of the anomeric specificity of yeast ad bovine hexokinases. Biol. Chem. Hoppe-Seyler, 367,411-416. Cardenas, M. L. Rabajille, E. and Niemeyer, H.(1984). Suppression of kinetic cooperativity of hexokinase D (glucokinase) by competitive inhibitors. A slow transition model. Eur. J . Biochem., 145, 163-171. Malaise, W. J., Zahner, D. and Marynissen, G. (1989). Anomeric specificity and kinetics of glucokinase: theoretical unsuitability of the Hill equation. Arch. internal. Physiol. Bioch., 97, 417-425. Zahner, D., Ramirez, R. and Malaise, W. J. (1990). Kinetic behaviour of liver glucokinase in diabetes. 11. Possible role of non-enzymatic protein glycation. Diab. Res., in press. Zahner. D.. Ramirez. R. and Malaise. W. J. (1990). Kinetic behaviour of liver glucokinase in diabetes. Ill. Possible role of insulinopenia. D i d . Res., in press
responsible for the anomeric specificity of glycolysis in pancreatic islets? J . Biol. Chem., 260, 12978-12981. Giroix, M.-H., Sener, A,, Pipeleers, D. G . and Malaise, W. J. (1984). Hexose metabolism in pancreatic islets. Biochem. J., 223,447-453. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, P. J. (1951). Protein measurement with the Folin reagent. J . Biol. Chem., 193,265-267. Zahner, D. and Malaise, W. J. (1991). Altered kinetic behaviour of liver glucokinase in starved rats. Biochimie., submitted for publication. Malaise, W. J., Giroix, M.-H., Dufrane, S. P., MalaisseLagae, F. and Sener, A. (1985). Anomeric specificity of hexokinase in rat, human, and murine tumor cells. Cancer Res.. 45, 6376-6378. Benkovic, S. J. and Schray, K. J. (1976). The anomeric specificity of glycolytic enzymes. Adv. Enzymol., 44,
3. 4.
5.
6.
7.
8. 9.
11. 12. 13. 14.
15.
139-164. 16. Batke, J. (1989). Channelling of glycolytic intermediates by temporary, stationary bi-enzyme complexes is probable in uitlo. Trends Biochem. Sci., 14,481-482.
Received 16 July 1990 Accepted 3 September 1990
VOL.
9: 49-53 (1991)
Phosphorylation by Liver Glucokinase of D-Glucose Anomers at Anomeric Equilibrium DAGMAR ZAHNER AND WILLY J. MALAISSET Laboratory of Experimental Medicine, Brussels Free University, I15 Boulevard de Waterloo, B-1000 Brussels, Belgium
The relative contribution of each anomer of D-glucose to the overall phosphorylation rate of the hexose tested at anomeric equilibrium was examined in rat liver postmicrosomal supernatants under conditions aimed at characterizing the activity of glucokinase, with negligible interference of either hexokinase, N-acetyl-D-glucosamine kinase or glucose-6-phosphatase (acting as a phosphotransferase). Both at lo" and 30 "C, the relative contribution of each anomer was unaffected by the concentration of D-glucose. At both temperatures, the a/B ratio for the contribution of each anomer was slightly, but significantly, lower than the u/B ratio of anomer concentrations. These findings, which are consistent with the anomeric specificity of glucokinase in terms of affinity, cooperativity and maximal velocity, reveal that the preferred u-anomeric substrate for both glycogen synthesis and glycolysis is generated by glucokinase at a lower rate than is ~-~-glucose-6-phosphate. KEY WORDS--
Liver glucokinase; D-glucose anomers; D - ~ ~ U C Ophosphorylation. S~
INTRODUCTION Like many other enzymes involved in the early steps of D-glucose metabolism, liver glucokinase displays qualified anomeric specificity. For instance, the maximal velocity of the reaction catalysed by liver glucokinase was found by several investigators to be higher with /?-D-glucose than with a-~-glucose.'-~ On the other hand, the affinity of the enzyme for the a-anomer appears greater than for the f l - a n ~ m e r . It ' ~is not obvious, however, that the latter preference is sufficiently marked to be able to compensate for the difference in maximal velocity so that the phosphorylation rate of a-D-glucose could exceed that of /?-D-glucose at low concentrations of each a n ~ m e r . ' - ~Likewise, there seems to exist little anomeric specificity in terms of the cooperative behaviour of glucokinase, as estimated from the Hill number.'. The information so far available on the anomeric specificity of glucokinase is restricted in several further interrelated aspects. First, little attention has so far been paid to the influence of temperature upon the anomeric specificity of glucokinase. This issue should not be neglected, since it was recently tAddressee for correspondance. 0263-6484/9 1/010049-05%05.00 0 1991 by John Wiley & Sons, Ltd.
shown that, in the case of the low-K, hexokinase, a change in temperature may affect the anomeric specificity of the enzyme, in terms of both affinity and maximal velocity.' The latter finding made it possible to reconcile data collected at different temperatures by different investigators. Second, although the kinetics of D-glUCOSe phosphorylation by glucokinase are often analysed by means of a Hill plot, this classical equilibrium model of cooperativity is not fully adequate, since glucokinase is a monomeric protein with only one active site.6 Furthermore, it was recently emphasized that the Hill model cannot be applied to glucokinase when considering the anomeric heterogeneity of its ~ u b s t r a t e .Last, ~ virtually no information is available on the possible perturbation of the anomeric behaviour of glucokinase in insulinopenic situations, such as diabetes and starvation, cases in which other intrinsic kinetic properties of the enzyme were recently found to be pert~rbed.~". With these considerations in mind, the present study aimed at investigating, both at 10 "C and 30 "C,the relative contribution of each anomer to the overall phosphorylation rate by liver glucokinase of D-glucose tested at anomeric equilibrium.
50
D. ZAHNER AND W. J . M A L A I S E
For reasons already alluded to,' this information ''C]glucose (6.48 mM) or P-D-[U - ''C]glucose cannot be unambiguously derived from the experi- (1 1.52 mM), 10 pl of a solution of the homologous mental data so far available, which are all restricted unlabelled anomer used in increasing concentrato studies performed with each anomer tested tions (zero to either 317.5 mM a-D-glucose or separately. The present study takes advantage of 564.5 mM a-D-glucose) and 20 p1 of a solution of the preparation of pure labelled anomers," which the other anomer (5.76 to 288 mM 8-D-glucose or allow the fate of each anomer to be followed 3.24 to 162 mM a-D-glucose). All these solutions in a mixture of either labelled a-D-glucose and were freshly prepared in the reaction mixture (see above) and maintained at 4 "C. The final concenunlabelled /?-D-glucose or vice versa. tration of each anomer always reached an a/P ratio equal to 36/64. Taking into account the presence of MATERIALS AND METHODS D-glucose in the liver postmicrosomal fraction, the The a- and P-anomers of D-glucose were purchased final concentration of D-glucose ranged from 4 mM from Sigma (St Louis, MO). The a- and j?-anomzr to 102mM. The incubation was conducted over of D-[U - ''C]glucose were prepared as described 20 min at 10 "C or 5 min at 30 "C, being initiated by elsewhere.' the addition of the liver postmicrosomal fraction Samples of liver (1.0 g each) removed from fed (50 p1) and terminated by addition of 2.0 ml of iced female albino rats were homogenized, using a water. The samples were then frozen ( - 20 "C). On mechanical homogenizer (Braun, Melsungen, the next day, the separation of D-glucose and its FRG) with six passes at 1250 rpm, in 2.0 ml of a acidic metabolites was achieved by anion-exchange Hepes-NaOH buffer (20 mM, pH 7.4) containing chromatography.' KC1 (50 mM), sucrose (50 mM), mannitol(200 mM), All readings were corrected for the blank value EDTA (1 mM), dithioerythritol (1 mM), phenyl- recorded in the absence of liver extract. Such a methylsulfonylfluoride (1 mM) and antipain blank value was not affected by the concentration (10 pg ml- ';Sigma). After 10-min centrifugation at of unlabelled D-glucose and, relative to the total 1000 g and 4 "C, the postnuclear supernatant was radioactive content of each sample, averaged 3.7 & centrifuged for 60 min at 100000 g and 4 "C to 0.1 per cent and 3.8 & 0.2 per cent (n = 6 in both obtain a postmicrosomal supernatant, which con- cases) with labelled a- and a-D-glucose, respectained 1.93 0.09 (n = 6) mg of protein per 50 pl, tively. The total radioactive content of each sample corresponding to about 116 k 6 mg g-' liver wet averaged 460 27 x 1O3 cpm per sample (n = 12). weight. The postmicrosomal supernatant was then The protein content was measured by the dialysed for 180 min at 8 "C against a large volume method of Lowry et a1.,12using bovine albumin as (51) of a Hepes-NaOH buffer (20mM, pH 7-4) standard. containing KCl (50 mM) and dithioerythritol All results including those mentioned above are (1 mM). The D-glucose concentration of the liver presented as the mean ( k SEM) together with the extract, as measured by the glucose oxidase number of individual determinations (n). The stamethod, averaged 21.8 0.7 mM and 3.7 k 0.4 mM tistical significance of differences was assessed by before and after dialysis, respectively. The conta- use of Student's t-test. mination by the hexose present in the dialysed sample was taken into account in adjusting the RESULTS final D-glucose concentration of the assay medium. The labelled and unlabelled anomers of D-glu- At increasing concentrations of D-glucose (4 to cose were prepared in an iced Hepes-NaOH buffer 100 mM), the total phosphorylation rate of both (100 mM, pH 7.5) containing KCl(50 mM), MgCl, anomers ranged from 0.5 0.2 to 6.7 k (18 mM), dithioerythritol(5 mM), D-glucose 6-phos- 1.3 nmol min-' mg-' protein at 10 "C and from phate (6.8 mM) and ATP (56.2 mM). Since 40 pl of 2.1 & 0.3 to 25.7 & 2-8 nmol min-' mg-' protein this reaction mixture were eventually mixed with at 30 "C(n = 3 in all cases). As illustrated in Figure 50 p1 of the dialysed liver extract, the final concen- 1, such a total phosphorylation rate yielded, in a tration of ATP (25mM) was sufficient to avoid Hill plot, linear relationships at both temperatures. exhaustion of this substrate, even at high concen- As expected from prior observation^,^ the cooperativity of glucokinase and, even more so, its affinity, trations of D-glucose. The assay medium, prepared at 4 "C, contained as judged respectively from the Hill number and lop1 of a solution of labelled ~-D-[U- K,, appeared greater at high than low temperature,
'
*
*
51
GLUCOKINASE A N D D-GLUCOSE ANOMERS
10°C -
0
m >
>'
-0.5
--.
Y
L2.J
z
/
07 0
-
/
1
-1.0
/
-1.(
i Figure I . Hill plot for the phosphorylation Of D-glucose at 10 "C(left) (slope 1.033) and 30 "C(right) (slope 1.145). Mean values are derived from triplicate measurements collected with both anomers in three individual experiments. The concentration of D-glucose ([GI) is expressed as mM.
with mean 3O0C/10"C ratios of 1.11 (cooperativity) and 1.67 (affinity). The fractional contribution of a-D-glucose to the total phosphorylation rate was little affected by the concentration of the hexose, whether at 10°C or 30 "C (Figure 2). Pooling all available data, such a ratio averaged 33.7 & 0.5 per cent at 10 "C and 32.6 & 0.6 per cent at 30 "C ( n = 18). These two mean values were not significantly different from one another, but were both lower ( P < 0.001) than the relative abundance of the a-anomer, i.e. 36 per cent. DISCUSSION
The major purpose of the present study was to assess, at both 10 "C and 30 "C, the relative contribution of a- and P-D-glucose to the overall rate of hexose phosphorylation by liver glucokinase. Several aspects of the experimental design used for such a purpose deserve attention. First, preliminary experiments conducted with semi-purified rat liver glu~okinase'~ indicated that the amount
of enzyme required, in order to obtain a reaction velocity well above the blank value, would necessitate lyophilization of a large volume of material eluting from the DEAE-Trisacryl column. The inherent risk of a loss in enzymatic activity associated with such a procedure led us eventually to conduct the present experiments with freshly isolated liver postmicrosomal supernatants. Second, it was previously shown that, under the present experimental conditions, the contribution of hexokinase, of glucose-6-phosphatase (acting as a phosphotransferase) and of N-acetyl-D-glucosamine kinase to D-glucose phosphorylation is negligible.3s Third, the temperature and length of incubation were selected to minimize the extent of hexose an~merization.'~ Fourth, because of the contamination of the assay medium by D-glucose present in the liver extract, the respective concentrations of each anomer were fixed at values close to those found at anomeric equilibrium. Last, the amount of liver material and volume of medium used in each assay was selected to ensure sizeable reaction velocities (above the appropriate blank
52
D. ZXHNER AND
W. J. MALAISSE
anomer according to the following equation:'
4Or
I
o\"
30L
Y
0
I
I
10
30
I
100
Figure 2. Relative contribution of a-D-glucose to the overall rate of hexose phosphorylation at 10 "C(upper panel) and 30 "C (lower panel) in the presence of increasing concentrations of Dglucose (logarithmic scale) tested at anometric equilibrium. Mean values ( f SEM) are derived from triplicate measurements collected with each anomer in three individual experiments. The dotted line represents the relative abundance of the a-form (36 per cent).
value) whilst maintaining the cumulative amount of D-glucose 6-phosphate generated below 10 per cent of the initial amount of hexose present in each sample. The fractional contribution of a-D-glucose to the total phosphorylation rate was unaffected by either the temperature or concentration of the hexose. It was slightly, but significantly, lower than the relative abundance of the a-anomer. In the absence of cooperativity, the fractional contribution of each anomer to hexose phosphorylation is dictated by the ratio of affinity to maximal velocity for each
in which u, and up represent the reaction velocities for each anomer, V, and V, the corresponding maximal velocities, [a] and [PI the anomer concentrations, and K , and K , the Michaelis constants. Since the [ a ] / [ P ] ratio was kept constant in the present experiments, our results reinforce the view that the anomeric difference in maximal velocities is slightly more pronounced than the anomeric difference in affinity, resulting in a fractional contribution of the a-anomer slightly lower than its relative abundance at anomeric equilibrium. It could be objected that this reasoning ignores the phenomenon of cooperativity. However, the fact that the relative contribution of each anomer to the overall rate of hexose phosphorylation was indeed close to that expected from the reported K , and V,,, values for a- and /3-~-glucose'-~is precisely in good agreement with the knowledge that there is little if any anomeric difference in cooperativity.'. The fact that the relative contribution of each anomer to the total rate of hexose phosphorylation was not affected by the concentration of D-glucose provides further support to the latter concept. In conclusion, the present work reveals that, at anomeric equilibrium, the relative contribution of each D-glucose anomer to the total rate of hexose phosphorylation by glucokinase is close, albeit not identical, to their respective abundance. Since all the enzymes catalysing the conversion of D-glucose 6-phosphate to further metabolites display strict anomeric preferen~e'~ and in view of the recent concept of enzyme-to-enzyme substrate tunnelling,' it would appear that the preferential substrate for either glycogen synthesis or glycolysis, namely a-D-glucose 6-phosphate, when generated from D-glucose, is formed in the liver at a lower rate than P-D-glucose 6-phosphate, which is preferentially channelled into the less active-in terms of metabolic flow-pentose phosphate pathway. The possible advantage of such an arrangement in the hepatic regulation of D-glucose metabolism remains to be scrutinized.
REFERENCES 1.
Meglasson, M. D. and Matschinsky, F. M. (1983). Discrimination ofglucose anomers by glucokinase from liver and transplantable insulinoma. J . B i d . Chem., 258,6705-6708.
53
GLUCOKINASE AND D-GLUCOSE ANOMERS 2. Sener, A., Giroix, M.-H., Dufrane, S. P. and Malaise, W. J. (1985). Anomeric specificity of hexokinase and gluco-
10. Sener, A,, Leclercq-Meyer, V., Marchand, J., Giroix, M. H., Dufrane, S. P. and Malaise, W. J. (1985). Is glucokinase
kinase activities in liver and insulin-producing cells. Biochem. J., 230, 345-351. Zahner, D. and Malaise, W. J. (1990). Kinetic behaviour of liver glucokinase in diabetes. I. Alteration of glucokinase in streptozotocin-diabetic rats. Diab. Res., in press. Miwa, I., Inagaki, K. and Okuda, J. (1983). Preference of glucokinase for the a-anomer of hexose: relation to uanomer preference in hexose-induced insulin release by pancreatic islets. Biochem. Int., 7, 449-454. Malaise-Lagae, F., Giroix, M.-H., Sener, A. and Malaise, W. J. (1986). Temperature dependency of the anomeric specificity of yeast ad bovine hexokinases. Biol. Chem. Hoppe-Seyler, 367,411-416. Cardenas, M. L. Rabajille, E. and Niemeyer, H.(1984). Suppression of kinetic cooperativity of hexokinase D (glucokinase) by competitive inhibitors. A slow transition model. Eur. J . Biochem., 145, 163-171. Malaise, W. J., Zahner, D. and Marynissen, G. (1989). Anomeric specificity and kinetics of glucokinase: theoretical unsuitability of the Hill equation. Arch. internal. Physiol. Bioch., 97, 417-425. Zahner, D., Ramirez, R. and Malaise, W. J. (1990). Kinetic behaviour of liver glucokinase in diabetes. 11. Possible role of non-enzymatic protein glycation. Diab. Res., in press. Zahner. D.. Ramirez. R. and Malaise. W. J. (1990). Kinetic behaviour of liver glucokinase in diabetes. Ill. Possible role of insulinopenia. D i d . Res., in press
responsible for the anomeric specificity of glycolysis in pancreatic islets? J . Biol. Chem., 260, 12978-12981. Giroix, M.-H., Sener, A,, Pipeleers, D. G . and Malaise, W. J. (1984). Hexose metabolism in pancreatic islets. Biochem. J., 223,447-453. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. and Randall, P. J. (1951). Protein measurement with the Folin reagent. J . Biol. Chem., 193,265-267. Zahner, D. and Malaise, W. J. (1991). Altered kinetic behaviour of liver glucokinase in starved rats. Biochimie., submitted for publication. Malaise, W. J., Giroix, M.-H., Dufrane, S. P., MalaisseLagae, F. and Sener, A. (1985). Anomeric specificity of hexokinase in rat, human, and murine tumor cells. Cancer Res.. 45, 6376-6378. Benkovic, S. J. and Schray, K. J. (1976). The anomeric specificity of glycolytic enzymes. Adv. Enzymol., 44,
3. 4.
5.
6.
7.
8. 9.
11. 12. 13. 14.
15.
139-164. 16. Batke, J. (1989). Channelling of glycolytic intermediates by temporary, stationary bi-enzyme complexes is probable in uitlo. Trends Biochem. Sci., 14,481-482.
Received 16 July 1990 Accepted 3 September 1990
