The insulin-resistant Zucker fa/fa rat has elevated hepatic glycolysis and activities of glucokinase and phosphofructokinase-2/fructose bisphosphatase-2 (PFK2). The latter catalyzes the formation and degradation of fructose-2,6-bisphosphate (fructose-2,6-P2) and is a glucokinase-binding protein. The contributions of glucokinase and PFK2 to the elevated glycolysis in fa/fa hepatocytes were determined by overexpressing these enzymes individually or in combination. Metabolic control analysis was used to determine enzyme coefficients on glycolysis and metabolite concentrations. Glucokinase had a high control coefficient on glycolysis in all hormonal conditions tested, whereas PFK2 had significant control only in the presence of glucagon, which phosphorylates PFK2 and suppresses glycolysis. Despite the high control strength of glucokinase, the elevated glycolysis in fa/fa hepatocytes could not be explained by the elevated glucokinase activity alone. In hepatocytes from fa/fa rats, glucokinase translocation between the nucleus and the cytoplasm was refractory to glucose but responsive to glucagon. Expression of a kinase-active PFK2 variant reversed the glucagon effect on glucokinase translocation and glucose phosphorylation, confirming the role for PFK2 in sequestering glucokinase in the cytoplasm. Glucokinase had a high control on glucose-6-phosphate content; however, like PFK2, it had a relative modest effect on the fructose-2,6-P2 content. However, combined overexpression of glucokinase and PFK2 had a synergistic effect on fructose-2,6-P2 levels, suggesting that interaction of these enzymes may be a prerequisite for formation of fructose-2,6-P2. Cumulatively, this study provides support for coordinate roles for glucokinase and PFK2 in the elevated hepatic glycolysis in fa/fa rats.
- glucose metabolism
the bifunctional enzyme 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase (PFK2) catalyzes both the formation of fructose-2,6-bisphosphate (fructose-2,6-P2) and its degradation (26, 27, 35). Fructose-2,6-P2 is a regulator of glycolysis because it is a potent activator of phosphofructokinase-1 and inhibitor of fructose-1,6-bisphosphatase-1. Various tissue-specific isoforms of PFK2 encoded by four genes are expressed in mammals. They differ in their relative kinase and bisphosphatase activities and also in their regulatory mechanisms (36). The liver isoform is regulated by phosphorylation of a serine residue at the NH2 terminus (Ser-32) by cAMP-dependent protein kinase, which leads to an increase in the bisphosphatase-to-kinase activity ratio. This mechanism accounts for the lowering of fructose-2,6-P2 and inhibition of glycolysis caused by glucagon (26).
Recent studies by Baltrusch et al. (8) showed that PFK2 binds to glucokinase through the bisphosphatase domain. Putative roles for this heterodimeric interaction have been proposed for both the hepatocyte and pancreatic β-cells, which express different isoforms of PFK2 (8, 31, 34). In pancreatic β-cells, binding of PFK2 activates glucokinase by posttranslational mechanisms (31), whereas in hepatocytes PFK2 has a dual role in regulating glucokinase expression and its subcellular compartmentation (34). Hepatic glucokinase shuttles between the cytoplasm and the nucleus depending on the substrate and hormonal conditions (1, 13). At low glucose, glucokinase is sequestered in the nucleus, through high-affinity binding to a 68-kDa protein (glucokinase regulatory protein), which is both a specific inhibitor for glucokinase and a nuclear receptor (13, 42). Binding of glucokinase to glucokinase regulatory protein is counteracted by fructose-1-phosphate, and accordingly precursors of fructose-1-phosphate cause translocation of glucokinase from the nucleus to the cytoplasm (2, 33). Translocation is also induced by elevated glucose concentrations by a mechanism that is synergistic with the effects of fructose-1-phosphate (4). Glucagon has the reverse effect and induces translocation of glucokinase from the cytoplasm to the nucleus (2, 13, 34). The role of PFK2 in mediating this effect of glucagon was shown by expression of a PFK2 variant lacking Ser-32 that functions as a constitutively active kinase (34). This PFK2 variant, when expressed at high levels, reverses the inhibitory effects of glucagon on glycolysis, fructose-2,6-P2, and glucokinase translocation, indicating a role for either PFK2 protein or its product fructose-2,6-P2 in mediating the effects of glucagon. It remains as yet undetermined whether retention of glucokinase in the cytoplasm at elevated glucose or low cAMP levels is mediated by binding of glucokinase to the dephosphorylated form of PFK2 or to independent cytoplasmic receptors through fructose-2,6-P2-dependent mechanisms (34).
Translocation of glucokinase between the nucleus and the cytoplasm has been studied in isolated hepatocytes during incubation with various substrates (2, 4, 33) and in vivo after fasting and refeeding (19) or after a glucose and insulin infusion (14). Defects in glucokinase translocation have been reported in animal models of Type 2 diabetes, including the Goto-Kakizaki rat and the Otsuka Long Evans Tokushima Fatty (OLETF) rat (41) and the Zucker diabetic fatty rat (20, 21). The Zucker diabetic rat had impaired glucokinase translocation during a glucose and insulin infusion (21). The mechanistic defect that accounts for the impaired translocation in these models of Type 2 diabetes is not known. A tentative hypothesis is that metabolic conditions associated with the diabetic or insulin-resistant state may alter the interaction of glucokinase with either nuclear or cytoplasmic receptors. In this study, we tested the hypothesis that PFK2 is a possible candidate for the impaired glucokinase translocation in a model of insulin resistance and Type 2 diabetes.
The hepatic activity of PFK2 is elevated in the insulin-resistant Zucker fa/fa rat (30), which is homozygous for a mutation in the leptin receptor gene. This model is characterized by hyperphagia, hyperinsulinemia, and mild hyperglycemia and is considered a good model for insulin resistance and Type 2 diabetes (40). The hepatic defect in this animal model is associated with increased partitioning of glucose-6-phosphate toward glycolysis and lipogenesis as opposed to glycogenesis and with various enzyme abnormalities, including elevated activities of glucokinase and glycolytic enzymes (5, 6, 39). In this study, we determined the contributions of glucokinase and PFK2 to the elevated glycolysis in this model by overexpressing these enzymes individually or in combination. We used metabolic control analysis to determine the sensitivity of glycolytic flux to changes in activity of these enzymes. We tested the role of PFK2 in the control of glucokinase translocation in this model of insulin resistance.
MATERIALS AND METHODS
Male Zucker Fa/? and fa/fa rats aged 11–13 wk (body wt of 311 ± 5 g for Fa/? and 461 ± 10 g for fa/fa) were obtained either from AstraZeneca (Alderley Park, Cheshire, UK) or from Harlan Olac (Bicester, UK). Rats were housed under standard conditions and fed ad libitum. Protocols adhered to the regulations of the United Kingdom Animal Scientific Procedures Act (1986), as outlined in the Project License approved by the local review committee and the Home Office. Hepatocytes were isolated by collagenase perfusion of the liver (34) and suspended in MEM containing 5% newborn calf serum and seeded in multiwell plates at a cell density of 8 × 104 cells/cm2.
Hepatocyte monolayer culture and enzyme overexpression.
After cell attachment (2 h), the medium was replaced by serum-free MEM containing adenoviral vectors for overexpression of rat liver glucokinase (Ad-LGK), the wild-type liver isoform of PFK2 (Ad-PFK2-W), or a kinase-active double mutant (S32A/H258A) variant of PFK2 (Ad-PFK2-M) described previously (7, 10). After a further 2 h, the medium was replaced by serum-free MEM containing 10 nM dexamethasone and 5 mM glucose, and the cells were cultured for 18 h to allow for enzyme expression (34).
For determination of rates of glucose phosphorylation or glycolysis, the hepatocyte monolayers were incubated for 3 h in MEM containing either [2-3H]glucose or [3-3H]glucose and the concentrations of glucose indicated without or with 100 nM glucagon or 10 nM insulin. Rates of glucose metabolism were linear during 3 h (16). At the end of the incubation, the medium was collected for determination of 3H2O and lactate (16). For determination of fructose-2,6-P2, the hepatocyte monolayer was extracted in 0.1 M NaOH and the plates were heated for 5 min at 80°C. Fructose-2,6-P2 was determined as described previously (43). For determination of glucose-6-phosphate, the plates were snap frozen in liquid nitrogen, and glucose-6-phosphate was determined fluorometrically in neutralized perchlorate extracts (25).
Glucokinase (free and bound) was determined spectrometrically after permeabilization of hepatocyte monolayers with digitonin (3), and free glucokinase activity is expressed as a percentage of total activity. Total activity is expressed as milliunits per milligram of protein (where 1 mU represents the amount converting 1 nmol of substrate/min). PFK2 activity was determined from the rate of conversion of fructose-6-phosphate to fructose-2,6-P2 during a 15-min incubation as in Ref. 9 (with activity expressed as pmol·min−1·mg−1) (34).
Metabolic control analysis.
Flux control coefficients of glucokinase or PFK2 on glucose metabolism, which are a measure of the fractional change in flux that results from a fractional change in enzyme activity (18, 28), were determined from experiments with titrated glucokinase or PFK2 overexpression using adenoviral vectors as described previously (16). The control coefficients were determined from the initial slopes of the double-log plots of rates of glucose metabolism (glucose phosphorylation or glycolysis) against the total activity for glucokinase or PFK2, respectively. Concentration control coefficients of glucokinase on glucose-6-phosphate or fructose-2,6-P2, representing the fractional change in cellular metabolite content that results from a fractional change in enzyme activity (28), were determined from experiments with titrated glucokinase overexpression from the slope of the double-log plot of metabolite content against enzyme activity (16, 25). The latter does not take into account possible subcellular compartmentation of these metabolites.
Immunoreactivity to glucokinase and PFK2 was determined by Western blotting using an antibody against human recombinant liver glucokinase (a kind gift from K. Brocklehurst, AstraZeneca, Macclesfield, Cheshire, UK) and an antibody to the bisphosphatase domain of rat liver PFK2 raised in chicken (31, 34).
After incubation of the hepatocyte monolayers on coverslips with the conditions indicated, they were rinsed in PBS and fixed in 4% paraformaldehyde-PBS (12). They were treated with NaBH4, preblocked, and stained as described previously (32). Antibody dilutions were as follows: for glucokinase, 1:100 dilution (rabbit IgG against human glucokinase residues 318-405, Santa Cruz); for PFK2, 1:50 dilution (chicken IgG against the bisphosphatase domain) (31). The secondary antibodies (donkey anti-rabbit or anti-chicken IgG, Jackson Immunoresearch) were FITC labeled (12). The cells were imaged with a Nikon Eclipse E400 epifluorescence microscope with narrow-band filters for FITC (B-2EC) and Nikon DXM1200 digital camera. Three representative fields were selected for each condition comprising between 50 and 90 nuclei, and the mean pixel intensity of the nuclei and cytoplasmic areas was analyzed from Gray images using Lucia G/F analysis software. For each incubation condition, the mean values for the nuclei and cytoplasm were determined and the results expressed as a nuclear-to-cytoplasmic ratio (N/C). Expression of results as a ratio corrects for drifts in fluorescence intensity (32).
Increased expression of glucokinase and PFK2 in fa/fa hepatocytes.
Hepatocytes from Zucker fatty fa/fa rats have altered activities of various enzymes involved in glucose metabolism (39). In this study, we confirmed that hepatocytes from fa/fa rats had a higher (P < 0.05) activity of glucokinase compared with Fa/? controls (10.6 ± 0.8 vs. 6.0 ± 0.6 mU/mg, means ± SE, n = 12). They also had a higher immunoreactivity to glucokinase (1.47 ± 0.04 vs. 1 relative densitometry units, n = 6) and to PFK2 (1.81 ± 0.04 vs. 1 relative densitometry units, n = 6).
Control of metabolic flux and metabolite concentrations by glucokinase.
To determine the contribution of glucokinase to the metabolic defect, hepatocytes from fa/fa and Fa/? rats were treated with 4 titers of recombinant adenovirus (Ad-LGK) to achieve titrated glucokinase overexpression by between two- and sevenfold above endogenous activity. The flux control coefficients of glucokinase on glucose phosphorylation and glycolysis were determined from the slopes of double-log plots of flux against total glucokinase activity (Fig. 1A) and are summarized in Table 1. Hepatocytes from fa/fa rats had higher (46–68%, P < 0.05) flux control coefficients of glucokinase on glucose phosphorylation (determined from metabolism of [2-3H]glucose) and on glycolysis, determined both from metabolism of [3-3H]glucose and formation of lactate than Fa/? controls (Table 1). However, the concentration control coefficients of glucokinase on glucose-6-phosphate and fructose-2,6-P2 (determined from double-log plots of metabolite content against glucokinase activity) were similar in fa/fa compared with Fa/? hepatocytes (Fig. 1B, Table 1). The concentration control coefficients of glucokinase on fructose-2,6-P2 were much lower than on glucose-6-phosphate.
Correlation between elevated glycolysis and fructose-2,6-P2.
The divergence in slope of metabolic flux against glucokinase activity (Fig. 1A) indicates that the higher rates of glycolysis in hepatocytes from fa/fa rats cannot be explained by the elevated glucokinase activity alone. In addition, the higher rates of glycolysis in fa/fa hepatocytes also cannot be explained by a higher glucose-6-phosphate content (Fig. 2A). However, they showed an apparent correlation (Fa/?: r = 0.97, P < 0.01; fa/fa: r = 0.97, P < 0.01) with the cell content of fructose-2,6-P2 (Fig. 2B), suggesting a possible contribution of either fructose-2,6-P2 or the dephosphorylated form of PFK2 protein.
Subcellular location of glucokinase and effects of glucose and glucagon.
The flux control coefficient of glucokinase on glucose metabolism is markedly dependent on its interaction with binding proteins (16, 25). The higher control coefficient of glucokinase on glycolysis in hepatocytes from fa/fa rats compared with controls (Table 1) could be due to differences in subcellular location and/or interaction with binding proteins. In hepatocytes from Fa/? rats, the N/C distribution of glucokinase was lower (P < 0.05) at 25 mM than at 10 mM glucose, and it was increased (P < 0.05) by glucagon at 25 mM glucose (Fig. 3), in agreement with previous findings on hepatocytes from Wistar rats (34). In fa/fa hepatocytes, there was no significant effect of glucose (25 mM vs. 10 mM) on the subcellular distribution of glucokinase. However, glucagon increased the N/C distribution at both 10 mM (P < 0.005) and 25 mM glucose (P < 0.005). This shows that hepatocytes from fa/fa rats are refractory to glucose but responsive to glucagon at both low and high glucose concentration. Hepatocytes from fa/fa rats incubated at 10 mM glucose also had a higher free glucokinase activity (determined by the digitonin release assay) compared with Fa/? controls (49 ± 2 vs. 43 ± 2 free glucokinase %total activity) as shown previously (5), indicating a greater distribution of glucokinase in the cytoplasm at low glucose.
Expression of PFK2-M counteracts the effect of glucagon on translocation.
Because PFK2 mediates the effects of glucagon on the N/C distribution of glucokinase (34), we determined the N/C distribution of glucokinase in hepatocytes after pretreatment without or with a recombinant virus for expression of a kinase-active PFK2 variant (PFK2-M: S32A, H258A). Expression of PFK2-M by threefold relative to endogenous PFK2 activity counteracted the effects of glucagon on translocation in hepatocytes from both fa/fa and Fa/? rats (Fig. 3), consistent with a role for PFK2 in mediating the effects of glucagon on glucokinase translocation in hepatocytes from fa/fa rats.
Regulation of glucose metabolism by glucagon: role of PFK2.
Because hepatocytes from fa/fa and Fa/? rats showed differences in glucose-induced but not glucagon-induced glucokinase translocation (Fig. 3), we evaluated the effects of glucagon and of expression of PFK2-M by threefold relative to endogenous activity on glucose metabolism. Rates of glucose phosphorylation and glycolysis were determined at 10 mM glucose in similar incubation conditions as in Fig. 3. In hepatocytes from both fa/fa and Fa/? rats, glucagon inhibited glycolysis by 42–45% and it caused a small but significant suppression of glucose phosphorylation (Fig. 4, A and B). The latter is consistent with the translocation of glucokinase to the nucleus with glucagon (Fig. 3). Expression of PFK2-M by threefold relative to endogenous activity reversed the inhibition of glucose phosphorylation caused by glucagon (Fig. 4A), and it caused partial reversal (42–45% to 29–30%) of the inhibition of glycolysis (Fig. 4B) but did not increase the fructose-2,6-P2 content (Fig. 4C). The latter is consistent with previous findings that expression of PFK2-M by ∼10-fold above endogenous activity is necessary to counteract the suppression of fructose-2,6-P2 caused by glucagon (34). The counteraction by PFK2-M of the inhibition of glucose phosphorylation by glucagon despite sustained suppression of fructose-2,6-P2 implicates a role for PFK2 protein as distinct from fructose-2,6-P2 in reversing the effects of glucagon on glucose phosphorylation and subcellular location.
Flux control coefficients of PFK2 on glucose metabolism.
In incubations with glucagon, the flux control coefficients of PFK2 on glucose phosphorylation (fa/fa: 0.18 ± 0.05 vs. Fa/?: 0.14 ± 0.05) and glycolysis (fa/fa: 0.36 ± 0.10 vs. Fa/?: 0.38 ± 0.05) were lower than the control coefficients for glucokinase (glucose phosphorylation: 0.52 ± 0.07 and 0.39 ± 0.03; glycolysis, 0.89 ± 0.14 and 0.68 ± 0.08). The latter coefficients of glucokinase were significantly higher (P < 0.05) than when flux was measured in the absence of glucagon (0.69 and 0.44; Table 1).
Effects of combined expression of glucokinase and PFK2-W on fructose-2,6-P2.
To determine the control of fructose-2,6-P2 concentration by glucokinase and PFK2, we overexpressed glucokinase and wild-type PFK2 separately and in combination by treatment with recombinant adenoviruses Ad-LGK and Ad-PFK2-W. Expression of either glucokinase or PFK2 alone by fivefold above endogenous levels had small effects (<50% increase) on the fructose-2,6-P2 content at both 10 mM glucose (Fig. 5A) and at 25 mM glucose (results not shown). In contrast, combined expression of glucokinase and PFK2 had a greater than additive effect on fructose-2,6-P2 levels, which were increased by threefold and fourfold in fa/fa and Fa/? hepatocytes, respectively (Fig. 5A). Similar synergism was observed in incubations with 25 mM glucose (results not shown).
During combined treatment with Ad-LGK and Ad-PFK2-W, glucokinase was overexpressed by 10- to 12-fold compared with 4- to 5-fold during treatment with Ad-LGK alone because PFK2 protein potentiates glucokinase expression (34). To normalize for differences in glucokinase activity between treatments with Ad-LGK alone or with additional Ad-PFK2, the effects of glucokinase on fructose-2,6-P2 were expressed as concentration control coefficients (as in Table 1). During combined PFK2 and glucokinase overexpression, the concentration control coefficients for glucokinase were greater (P < 0.01) than in cells overexpressing glucokinase alone in both Fa/? (10 mM glucose: 0.38 ± 0.04 vs. 0.28 ± 0.05; 25 mM glucose: 0.55 ± 0.06 vs. 0.16 ± 0.05) and fa/fa hepatocytes (10 mM glucose: 0.48 ± 0.05 vs. 0.27 ± 0.07; 25 mM glucose: 0.33 ± 0.04 vs. 0.18 ± 0.054), confirming synergistic effects by combined enzyme overexpression.
Glucokinase overexpression overrides the inhibition of glycolysis by glucagon.
Overexpression of PFK2-W by sixfold had a negligible effect (<15%) on glycolysis (Fig. 5A), similar to the lack of effect of expression of a kinase-active variant (PFK2-M) in the absence of glucagon (Fig. 4A). However, overexpression of glucokinase (4- to 5-fold) increased glycolysis by two- to threefold in both the absence and presence of glucagon (Fig. 5B), even though it did not override the suppression of fructose-2,6-P2 caused by glucagon (Fig. 5A). The stimulation of glycolysis caused by glucokinase overexpression correlated with the rate of glucose phosphorylation (Fig. 5C).
The Zucker fa/fa rat, which is homozygous for a mutation in the leptin receptor gene and is widely used as a model for insulin resistance and Type 2 diabetes, is characterized by high rates of glycolysis and lipogenesis and elevated activities of glucokinase and PFK2 (30, 39). In this study, we determined the contributions of these enzymes to the elevated glycolysis in hepatocytes from fa/fa rats by overexpressing glucokinase and PFK2 either individually or in combination. Three main findings emerged from this study. First, glucokinase had a much higher control strength on glycolysis than PFK2 in all experimental conditions tested, to the extent that glucokinase overexpression can override the inhibition of glycolysis caused by glucagon, despite sustained suppression of fructose-2,6-P2. Second, elevated activity of glucokinase alone cannot explain the higher rate of glycolysis in hepatocytes from fa/fa than in those from control rats, indicating the involvement of additional mechanisms that together impart a higher control strength of glucokinase on glycolysis in fa/fa hepatocytes. Third, the cell content of fructose-2,6-P2 in liver cells is coordinately controlled by glucokinase and PFK2.
Major but not exclusive role for glucokinase in the control of glycolysis.
Titrated glucokinase overexpression by up to sevenfold above endogenous levels resulted in a progressive increase in glycolysis when this was determined in both the absence of hormones and the presence of insulin or glucagon. These hormones affect the phosphorylation state of PFK2 and thereby the kinase-to-bisphosphatase ratio (26, 27). This is reflected by changes in fructose-2,6-P2, which was increased by insulin and decreased by glucagon. Overexpression of PFK2 as either a kinase-active variant (Fig. 4) or the wild-type enzyme (Fig. 5) did not significantly increase glycolysis when this was determined in the absence of glucagon. However, the kinase-active variant PFK2-M increased glycolysis in the presence of glucagon. This indicates that the endogenous level of PFK2 sustains near-maximal rates of glycolysis when it is in its dephosphorylated state (high kinase-to-bisphosphatase ratio). In these conditions, control of glycolysis resides predominantly at glucokinase. However, in the presence of glucagon when PFK2 is phosphorylated, control is shared between glucokinase and PFK2. It is noteworthy that glucokinase overexpression increased glycolysis in the presence of glucagon, despite sustained suppression of fructose-2,6-P2 content, indicating the overriding role of glucose phosphorylation in controlling glycolysis in conditions of suppressed fructose-2,6-P2 but elevated glucose phosphorylation.
Despite the high control strength of glucokinase on glycolysis, the elevated rate of glycolysis in hepatocytes from fa/fa rats cannot be explained by the elevated activity of glucokinase alone. This is supported by the upward shift in the correlation between glycolysis and glucokinase activity in hepatocytes from fa/fa rats compared with controls (Fig. 1A). The higher control strength of glucokinase on glycolysis in fa/fa hepatocytes (represented by the slope of the double log plot in Fig. 1A) also indicates involvement of additional factors that act synergistically with glucokinase in fa/fa hepatocytes.
Impaired glucose-induced glucokinase translocation in fa/fa hepatocytes.
Previous studies have shown that glucose-induced translocation of glucokinase from the nucleus to the cytoplasm is impaired in models of insulin resistance and diabetes such as the Goto-Kakizaki, OLETF (41), and Zucker diabetic fatty rats (20, 21). However, the underlying mechanisms have not been determined. We show in this study that glucokinase translocation in hepatocytes from fa/fa rats was refractory to glucose (25 mM vs. 10 mM) but responsive to glucagon at both 25 and 10 mM glucose. The refractoriness to glucose could be due to either impaired translocation at 25 mM glucose or to increased accumulation of glucokinase in the cytoplasm at 10 mM glucose. We cannot unequivocally exclude the former possibility. However, the increase in free glucokinase activity at 10 mM glucose, in fa/fa hepatocytes compared with controls, supports the latter hypothesis. A higher cell content of the dephosphorylated form of PFK2 in hepatocytes from fa/fa rats, as suggested by the higher cell content of fructose-2,6-P2, could explain the sequestration of glucokinase in the cytoplasm at low glucose.
Expression of a kinase-active PFK2 variant counteracts glucagon-induced glucokinase translocation independently of fructose-2,6-P2.
Previous work showed that, when a kinase-active variant of PFK2 (lacking Ser-32 and His-258) is expressed by 10-fold excess over endogenous PFK2 activity, it reverses the suppression of fructose-2,6-P2 caused by glucagon (34). The requirement for such a high level of expression of this variant was tentatively explained by the high activity of the endogenous bisphosphatase at saturating glucagon concentration (34). This study shows that expression of the kinase-active PFK2 variant by threefold relative to endogenous enzyme had negligible effects on the cell content of fructose-2,6-P2, particularly in hepatocytes from fa/fa rats (Fig. 4C), but nonetheless reversed both the translocation of glucokinase from the cytoplasm to the nucleus and the suppression of glucose phosphorylation caused by glucagon (Fig. 4A). This supports a role for PFK2 protein as distinct from its product fructose-2,6-P2 in controlling glucokinase location and glucose phosphorylation. At this level of PFK2 expression, there was only partial reversal of the inhibition of glycolysis by glucagon, consistent with the dual control of glycolysis by the rate of glucose phosphorylation (determined by glucokinase) and by allosteric activation of phosphofructokinase-1 by fructose-2,6-P2.
Synergistic control of fructose-2,6-P2 by glucokinase and PFK2.
Overexpression of glucokinase had a large effect on the cell content of glucose-6-phosphate in agreement with previous findings (25, 37), but it had a comparatively modest effect on fructose-2,6-P2 levels as shown by the lower concentration control coefficients (0.3 vs. 1.1). This indicates that neither glucose-6-phosphate nor fructose-6-phosphate, which is in equilibrium with glucose-6-phosphate, is a major determinant of the cell content of fructose-2,6-P2. Expression of PFK2-like expression of glucokinase also had a modest effect on fructose-2,6-P2 levels (Fig. 5A). However, combined overexpression of glucokinase and PFK2 had a synergistic effect on fructose-2,6-P2 levels. This is supported by the higher concentration control coefficient of glucokinase on fructose-2,6-P2 when PFK2 was overexpressed. This clearly establishes coordinate roles of glucokinase and PFK2 in regulating the fructose-2,6-P2 content. A tentative hypothesis is that, in the absence of phosphorylation of Ser-32, glucokinase can bind to PFK2 and facilitate formation of fructose-2,6-P2 through either inhibition of the bisphosphatase, because glucokinase binds to the bisphosphatase domain (8), or channelling involving phosphoglucoisomerase. A further possibility is that glucokinase may increase the kinase activity of PFK2 as suggested from a study on liver homogenates (38). The higher cell content of fructose-2,6-P2 with titrated glucokinase overexpression in hepatocytes from fa/fa compared with Fa/? controls (Fig. 1B) is consistent with the higher expression of PFK2 in fa/fa hepatocytes and with coordinate regulation by glucokinase and PFK2.
Glucokinase translocates from the nucleus to the cytoplasm in response to a rise in extracellular glucose concentration in the absorptive state (1, 14, 19). From the present and previous findings (34), it is proposed that glucokinase binding to the dephosphorylated form of PFK2 through the bisphosphatase domain (8) sequesters gluocokinase in the cytoplasm. As glucose absorption from the gut declines, the rise in glucagon results in phosphorylation of PFK2 (35), dissociation of glucokinase, and translocation to the nucleus. Pathophysiological states associated with increased expression of PFK2 or with its dephosphorylation would favor sequestration of glucokinase in the cytoplasm and increased glycolysis.
The work was supported by project grant support from the Wellcome Trust and equipment grants from the Medical Research Council (JREI, G0100348) and Diabetes UK (RD01/0002364) to L. Agius and by National Institutes of Health Grant RO1-38354 to A. J. Lange.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society