Glucose phosphorylation in tumor cells. Cloning, sequencing, and overexpression in active form of a full-length cDNA encoding a mitochondrial bindable form of hexokinase

In rapidly growing tumor cells exhibiting high glucose catabolic rates, the enzyme hexokinase is markedly elevated and bound in large amounts (50-80% of the total cell activity) to the outer mitochondrial membrane (Arora, K.K., and Pedersen, P.L. (1988) J. Biol. Chem. 263, 17422-17428; Parry, D.M., and Pedersen, P.L. (1983) J. Biol. Chem. 258, 10904-10912). In extending these studies, we have isolated a cDNA clone of hexokinase from a lambda gt11 library of the highly glycolytic, c37 mouse hepatoma cell line. This clone, comprising 4,198 base pairs, contains a single open reading frame of 2,754 nucleotides which encode a 918-amino acid hexokinase with a mass of 102,272 daltons. This enzyme exhibits, respectively, 68 and 32 amino acid differences, including several charge differences, from the recently sequenced human kidney and rat brain enzymes. The putative glucose and ATP binding domains present in the latter two enzymes and in rat liver glucokinase are conserved in the tumor enzyme. At its N-terminal region, tumor hexokinase has a 12-amino acid hydrophobic stretch which is present in the rat brain enzyme but absent in the rat liver glucokinase, a cytoplasmic enzyme. The mature tumor hexokinase protein has been overexpressed in active form in Escherichia coli and purified 9-fold. The overexpressed enzyme binds to rat liver mitochondria in the presence of MgCl2. This is the first report describing the cloning and sequencing of a tumor hexokinase, and the first report documenting the overexpression of any hexokinase type in E. coli. Questions pertinent to the enzyme's mechanism, regulation, binding to mitochondria, and its marked elevation in tumor cells can now be addressed.     

Inactivation and reactivation of hexokinase type II

Hexokinase isozyme II which loses activity rapidly in the absence of glucose ($\text{t}{}_{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}\text{～- 10}\text{min}$) is stabilized in the presence of glucose-6-P, P_i and ADP when glucose is also present but not by kinetically inert analogs. Enzyme inactivated by incubation in the absence of glucose is fully and rapidly recovered ($\text{t}{}_{\mathrm{<mtext>1</mtext><mtext>2</mtext><mtext>～-\; 10\; </mtext><mtext>min</mtext>}}$) by addition of both glucose and mercaptoethanol, each at 0.1 m. In the presence of 0.1 mm glucose, both glucose-6-P and P, facilitate the reactivation. Reactivation proceeds in two steps both with unfavorable equilibria: a fast reduction followed by a slow renaturation. Native enzyme is much more resistant to irreversible inactivation by trypsin than is enzyme that has lost its activity by incubation in the absence of glucose. The latter form shows no protection from trypsin action by glucose. Streptozotocin-diabetic rats that have lost hexokinase II preferentially in their insulin-sensitive tissues do not contain an activatable form of hexokinase in at least one of these, heart. The greater sensitivity of inactivated hexokinase to denaturation by trypsin suggests that such a “reservoir” form may be destroyed rapidly in vivo. Glucose may be important in determining the steady-state level of hexokinase II by “guiding” the folding of translation product. In this view insulin would act through its effect on glucose permeability.     

Factors controlling the metabolic response to glucose in isolated rat lung cells

Glucose decreases the oxygen utilization by isolated rat lung cells. Its effect displays saturation type kinetics with a “Ki” of 2.2. mM. The similarity of this value with the reported “Km” of 2.4 mM described for glucose uptake by these cells, suggests that both processes may be intimately related and both of them are under the control of the same rate limiting step. Several arguments point to glucose transport into these cells as the most important rate limiting step for its utilization: 1) Phloridzin prevented glucose inhibition of oxygen uptake while mannoheptulose did not; 2) The activity of hexokinase which is the least active glycolytic enzyme in these cells far exceeded the observed rates of glucose utilization and a decrease of 45 per cent in its activity in starved animals did not affect the rate of glucose uptake; 3) The “Km” of hexokinase for glucose is two orders of magnitude below the observed “Km” for glucose uptake and the “Ki” for glucose inhibition of respiration.     

Complete amino acid sequence of the type II isozyme of rat hexokinase, deduced from the cloned cDNA: comparison with a hexokinase from novikoff ascites tumor

The 917-residue amino acid sequence of the Type II isozyme of rat hexokinase has been deduced from the nucleotide sequence of cloned cDNA. The sequences of 197 nucleotides in the 5' untranslated region and 687 bases of the 3' untranslated region have also been determined. A region of overlap between two discrete cDNA clones was confirmed by isolation and sequencing of a genomic DNA clone that spanned the region. Within this region, the 634-nucleotide coding sequence was divided into three exons, each of 150-250 nucleotides; these results suggest that the gene encoding Type II hexokinase is likely to be quite complex. There is extensive similarity between the sequences of the N- and C-terminal halves of the Type II isozyme, as previously seen with the Type I and III isozymes; this is consistent with the view that these enzymes evolved by a process of gene duplication and fusion. A cDNA encoding the entire C-terminal half of a hexokinase from Novikoff ascites tumor cells was also isolated and found to be identical to a cDNA encoding the corresponding region of the Type II isozyme of skeletal muscle. Northern analysis indicated that a single mRNA, approx 5200 nucleotides in length, encoded both the skeletal muscle and the tumor enzymes. These results do not support previous speculation that the hexokinase isozymes of normal tissue are distinct from those of tumors, and suggest the possibility that post-translational modifications of a single protein species might account for apparent differences between the isozymes of normal and tumor tissues.     

Changing patterns of placental hexokinase isozymes during the course of gestation.

Hexokinase, an enzyme that is capable of regulating the entry of glucose into metabolic sequences, is known to exist in four isozyme forms in a variety of mammalian tissues. Each of these isozymes possesses different kinetic properties, suggesting that they may serve in different regulatory capacities. In addition, the proportions of the four isozymes are characteristically different for various differentiated tissues, presumably reflecting different metabolic capabilities of the tissues.Extracts of rabbit and human placentas from early gestational age and term pregnancies were chromatographed and assayed for hexokinase activity. Four peaks of activity were observed. Elution positions of the four placental hexokinase isozymes were comparable to those from liver; however, the relative proportions were considerably different. In both rabbit and human placentas, the proportion of Type I hexokinase increased during gestation and Type III decreased, while Types II and IV remained essentially unchanged. Implications of the gestational changes for placental carbohydrate metabolism are discussed.     

Pig red blood cell hexokinase: Regulatory characteristics and possible physiological role

The regulatory properties of pig erythrocyte hexokinase III have been studied. Among mammalian erythrocyte hexokinases, the pig enzyme shows the highest affinity for glucose and a positive cooperative effect with n_H = 1.5 at all the MgATP concentrations studied (for 0.5 to 5 mm). Glucose at high concentrations is also an inhibitor of hexokinase III. Similarly, the apparent affinity constant for MgATP is independent of glucose concentration. Uncomplexed ATP and Mg are both competitive inhibitors with respect to MgATP. Glucose 6-phosphate, known as a stronger inhibitor of all mammalian erythrocyte hexokinases, is a poor inhibitor for the pig enzyme (K_i = 120 μm). Furthermore, this inhibition is not relieved by orthophosphate as with other mammalian red blood cell hexokinases. A variety of red blood cell-phosphorylated compounds were tested and found to be inhibitors of pig hexokinase III. Of these, glucose 1,6-diphosphate and 2,3-diphosphoglycerate displayed inhibition constants in the range of their intracellular concentrations. In an attempt to investigate the role of hexokinase type III in pig erythrocytes some metabolic properties of this cell have been studied. The adult pig erythrocyte is able to utilize 0.27 μmol of glucose/h/ml red blood cells (RBC) compared with values of 0.56–2.85 μmol/h/ml RBC for the other mammalian species. This reduced capacity to metabolize glucose results from a relatively poor ability of the cell membrane to transport glucose. In fact, all the glycolytic enzymes were present and a low intracellular glucose concentration was measured (0.5 mm against a plasma level of 5 mm). Furthermore, transport and utilization were concentration-dependent processes. Inosine, proposed as the major energy substrate of the pig erythrocyte, at physiological concentrations is not as efficient as glucose in maintaining reduced glutathione levels under oxidative stress. Furthermore, newborn pig erythrocytes (fully permeable to glucose) possess hexokinase type II as the predominant glucose-phosphorylating activity. This fact and the information derived from the study of the regulatory characteristics of hexokinase III and from metabolic studies on intact pig erythrocytes permit the hypothesis that the presence of this peculiar hexokinase isozyme (type III) enables the adult pig erythrocyte to metabolize low but appreciable amounts of glucose.     

Hexokinase II-deficient mice. Prenatal death of homozygotes without disturbances in glucose tolerance in heterozygotes.

Type 2 diabetes is characterized by decreased rates of insulin-stimulated glucose uptake and utilization, reduced hexokinase II mRNA and enzyme production, and low basal levels of glucose 6-phosphate in insulin-sensitive skeletal muscle and adipose tissues. Hexokinase II is primarily expressed in muscle and adipose tissues where it catalyzes the phosphorylation of glucose to glucose 6-phosphate, a possible rate-limiting step for glucose disposal. To investigate the role of hexokinase II in insulin action and in glucose homeostasis as well as in mouse development, we generated a hexokinase II knock-out mouse. Mice homozygous for hexokinase II deficiency (HKII(-/-)) died at approximately 7.5 days post-fertilization, indicating that hexokinase II is vital for mouse embryogenesis after implantation and before organogenesis. HKII(+/-) mice were viable, fertile, and grew normally. Surprisingly, even though HKII(+/-) mice had significantly reduced (by 50%) hexokinase II mRNA and activity levels in skeletal muscle, heart, and adipose tissue, they did not exhibit impaired insulin action or glucose tolerance even when challenged with a high-fat diet.     

The effect of insulin on the catalytic efficacy of rat skeletal muscle hexokinase isoenzyme II]

The effect of insulin on the intracellular localization of rat skeletal muscle hexokinase isozyme II (hexokinase II) was studied in vivo. It was found that after injection of the hormone the glucose concentration in the muscle gradually increases in parallel with the hexokinase II redistribution between the cytosol and the mitochondrial fraction in the direction of the bound form of the enzyme. This effect of insulin is due to glucose, an indispensable participant of the complex formation between the enzyme and the mitochondrial membrane. It was shown that the effect of glucose as a hexokinase II adsorbing reagent is a highly specific one. The hexokinase II binding to mitochondria in the presence of glucose is accompanied by changes in some kinetic properties of the enzyme. A kinetic analysis of catalytic efficiency of the free and bound hexokinase II forms revealed that the catalytic efficiency of hexokinase II within the composition of the enzyme-membrane complex exceeds by two orders of magnitude that of the free enzyme. The data obtained are discussed in the framework of an adsorption mechanism of hexokinase activity regulation in the cell.     

重组荞麦胰蛋白酶抑制剂通过下调己糖激酶Ⅱ抑制HepG2细胞生长

糖酵解过度活跃是肿瘤细胞能量代谢的显著特征。抑制过度糖酵解已经成为一种新的癌症疗法。重组荞麦胰蛋白酶抑制剂 (recombinant buckwheat trypsin inhibitor, rBTI)可以通过上调磷酸酶及张力蛋白同源基因 (PTEN) 进而抑制HepG2细胞增殖。有关rBTI对肿瘤细胞能量代谢的影响仍未见报道。本研究中的MTT和ATP检测分析表明,rBTI以剂量依赖性方式抑制细胞活力及胞内ATP含量。qRT-PCR和Western印迹分析表明,rBTI处理HepG2细胞后,己糖激酶Ⅱ转录显著下调,但是糖酵解过程中的其他酶及葡萄糖转运蛋白基因在转录水平未发生显著变化,同时己糖激酶Ⅱ蛋白水平的表达也显著下调。酶活性分析也表明,rBTI能显著降低己糖激酶的活性。进一步分析表明, rBTI使细胞内PTEN转录及表达水平明显上调,己糖激酶Ⅱ转录和p-AKT,p-mTOR、己糖激酶Ⅱ的表达下调。当PTEN抑制剂phen存在时,可阻断rBTI诱导的己糖激酶 Ⅱ表达下降,表明rBTI能通过上调PTEN进而影响己糖激酶Ⅱ的表达。免疫荧光及Western印迹分析显示,rBTI作用后减弱了己糖激酶 Ⅱ在线粒体的定位,导致己糖激酶Ⅱ与线粒体电压依赖性阴离子通道蛋白 (voltage-dependent anion channel, VDAC) 分离,促使己糖激酶Ⅱ从线粒体转位到细胞质,降低糖酵解的效率。上述结果证明,rBTI对肿瘤细胞能量代谢的调控作用主要通过抑制PI3K/AKT信号通路,下调己糖激酶Ⅱ的表达并影响空间定位,进而抑制肿瘤细胞糖酵解过程,导致癌细胞生长受到抑制。     

Direct inhibition of hexokinase activity by metformin at least partially impairs glucose metabolism and tumor growth in experimental breast cancer

Emerging evidence suggests that metformin, a widely used anti-diabetic drug, may be useful in the prevention and treatment of different cancers. In the present study, we demonstrate that metformin directly inhibits the enzymatic function of hexokinase (HK) I and II in a cell line of triple-negative breast cancer (MDA-MB-231). The inhibition is selective for these isoforms, as documented by experiments with purified HK I and II as well as with cell lysates. Measurements of ¹⁸F-fluoro-deoxyglycose uptake document that it is dose- and time-dependent and powerful enough to virtually abolish glucose consumption despite unchanged availability of membrane glucose transporters. The profound energetic imbalance activates phosphorylation and is subsequently followed by cell death. More importantly, the “in vivo” relevance of this effect is confirmed by studies of orthotopic xenografts of MDA-MB-231 cells in athymic (nu/nu) mice. Administration of high drug doses after tumor development caused an evident tumor necrosis in a time as short as 48 h. On the other hand, 1 mo metformin treatment markedly reduced cancer glucose consumption and growth. Taken together, our results strongly suggest that HK inhibition contributes to metformin therapeutic and preventive potential in breast cancer.     

Oncogenes and inflammation rewire host energy metabolism in the tumor microenvironment

Here, we developed a model system to evaluate the metabolic effects of oncogene(s) on the host microenvironment. A matched set of “normal” and oncogenically transformed epithelial cell lines were co-cultured with human fibroblasts, to determine the “bystander” effects of oncogenes on stromal cells. ROS production and glucose uptake were measured by FACS analysis. In addition, expression of a panel of metabolic protein biomarkers (Caveolin-1, MCT1, and MCT4) was analyzed in parallel. Interestingly, oncogene activation in cancer cells was sufficient to induce the metabolic reprogramming of cancer-associated fibroblasts toward glycolysis, via oxidative stress. Evidence for “metabolic symbiosis” between oxidative cancer cells and glycolytic fibroblasts was provided by MCT1/4 immunostaining. As such, oncogenes drive the establishment of a stromal-epithelial “lactate-shuttle”, to fuel the anabolic growth of cancer cells. Similar results were obtained with two divergent oncogenes (RAS and NFκB), indicating that ROS production and inflammation metabolically converge on the tumor stroma, driving glycolysis and upregulation of MCT4. These findings make stromal MCT4 an attractive target for new drug discovery, as MCT4 is a shared endpoint for the metabolic effects of many oncogenic stimuli. Thus, diverse oncogenes stimulate a common metabolic response in the tumor stroma. Conversely, we also show that fibroblasts protect cancer cells against oncogenic stress and senescence by reducing ROS production in tumor cells. Ras-transformed cells were also able to metabolically reprogram normal adjacent epithelia, indicating that cancer cells can use either fibroblasts or epithelial cells as “partners” for metabolic symbiosis. The antioxidant N-acetyl-cysteine (NAC) selectively halted mitochondrial biogenesis in Ras-transformed cells, but not in normal epithelia. NAC also blocked stromal induction of MCT4, indicating that NAC effectively functions as an “MCT4 inhibitor”. Taken together, our data provide new strategies for achieving more effective anticancer therapy. We conclude that oncogenes enable cancer cells to behave as selfish “metabolic parasites”, like foreign organisms (bacteria, fungi, viruses). Thus, we should consider treating cancer like an infectious disease, with new classes of metabolically targeted “antibiotics” to selectively starve cancer cells. Our results provide new support for the “seed and soil” hypothesis, which was first proposed in 1889 by the English surgeon, Stephen Paget.     

Hexokinase isoenzymes of RIN-m5F insulinoma cells. Expression of glucokinase gene in insulin-producing cells.

We have analysed the pattern of expression of the hexokinase isoenzyme group in RIN-m5F insulinoma cells. Three hexokinase forms were resolved by DEAE-cellulose chromatography. The most abundant isoenzyme co-eluted with hexokinase type II from rat adipose tissue and displayed a Km for glucose of 0.15 mM, similar to the adipose-tissue enzyme. Hexokinase type II was in large part associated with a particulate subcellular fraction in RIN-m5F cells. The two other hexokinases separated by ion-exchange chromatography were an enzyme similar to hexokinase type I from brain and glucokinase (or hexokinase type IV). The latter isoenzyme was identified as the liver-type glucokinase by the following properties: co-elution with hepatic glucokinase from DEAE-cellulose and DEAE-Sephadex; sigmoid saturation kinetics with glucose with half-maximal velocity at 5.6 mM and Hill coefficient (h) of 1.54; suppression of enzyme activity by antibodies raised against rat liver glucokinase; apparent Mr of 56,500 and pI of 5.6, as shown by immunoblotting after one- and two-dimensional gel electrophoresis; peptide map identical with that of hepatic glucokinase after proteolysis with chymotrypsin and papain. These data indicate that the gene coding for hepatic glucokinase is expressed in RIN-m5F cells, a finding consistent with indirect evidence for the presence of glucokinase in the beta-cell of the islet of Langerhans. On the other hand, the overall pattern of hexokinases is distinctly different in RIN-m5F cells and islets of Langerhans, since hexokinase type II appears to be lacking in islets. Alteration in hexokinase expression after tumoral transformation has been reported in other systems.     

Distribution of hexokinase isoenzymes depending on a carbon source in Saccharomyces cerevisiae.

DEAE cellulose chromatography and agar gel electrophoresis of glucose-phosphorylating enzymes in Saccharomyces cerevisiae showed the existence of glucokinase and two hexokinase isoenzymes ( designated as hexokinase I and II ). The distribution of hexokinase isoenzymes was dependent on a carbon source in the medium, while that of glucokinase was not dependent. The cells grown on 3 % ethanol as carbon source showed the isoenzyme pattern with predominant hexokinase I and a little hexokinase II. The isoenzyme pattern of the cells grown on 6 % glucose, which was differnt from that of the cells grown on ethanol, showed that hexokinase I and II were minor and major parts respectively. When the cells grown on 3 % ethanol were incubated on the medium containing 6 % glucose, hexokinase I was repressed and hexokinase II inducted. These facts suggest that two hexokinase isoenzymes, but not glucokinase, are adaptive enzyme.     

Identification of regulatory properties of metabolic networks by graph theoretical modeling

An earlier graph theoretical model of metabolic and gene-expression networks has been modified and extended to include the effect of electrical potentials on binding constants, representation of uncatalyzed processes, and treatment of parallel reactions catalyzed by a single enzyme. Formal operations on the graph, which are facilitated by a set of standardized guidelines, identify the feedback signals in the network and rank them according to their influence. The technique was applied to a model of glycolysis in ascites tumor cells in the absence and presence of 12.5 mM exogenous glucose. Feedback regulation was widely distributed and mostly due to binding of adenine nucleotide cofactors to the enzymes of the network. The major changes in feedback regulation on adding glucose is the relief of inhibition of hexokinase and phosphofructokinase and the activation of pyruvate kinase. We conclude that regulation of tumor cell glycolysis is not restricted to hexokinase or to (Na+,K+)-ATPase as was previously suggested by others.     

Stabilization of hexokinases I and II of ELD cells by binding to mitochondria

Significance of the binding of hexokinase to mitochondria was examined with respect to stabilization of the enzyme by the binding. Stability during the incubation of the mitochondria-bound forms of hexokinases I and II, both prepared from Ehrlich-Lettre ascites hyperdiploid tumor cells (ELD cells), were compared with that of the corresponding free forms. During the incubation at pH 7.4 and 37 degrees C up to 60 min, hexokinase activities decreased gradually, and the decrease in the activity of the free form was much more marked than that of the bound form for both hexokinases. Hexokinase II was much less stable than I, and the activity of the free form of the former was almost lost by the incubation for 15 min. But, more than a half of the original activity of hexokinase II was retained even after 60 min of the incubation when the enzyme was bound to mitochondria. Addition of 50 mM glucose increased the stability of hexokinase II, but the stabilizing effect was less marked for hexokinase I. On the other hand, addition of 28 mg/ml of bovine serum albumin markedly stabilized hexokinase I to almost the same extent as was observed with mitochondria. On the contrary, the serum albumin had little stabilizing effect on hexokinase II. These findings indicate that the binding to mitochondria stabilizes the hexokinases of ELD cells, though the stability is different by nature between hexokinases I and II.     

Polyamines stimulate the binding of hexokinase type II to mitochondria

Spermine and spermidine enhanced the binding of hexokinase isoenzyme type II to mitochondria, both of which were prepared from Ehrlich-Lettre hyperdiploid ascites tumor cells, at much lower concentrations than Mg2+. Chymotrypsin-treated hexokinase II could not bind to the mitochondrial membrane in the presence of either spermine or Mg2+, indicating that the effect of spermine is not a nonspecific action, since the treatment of chymotrypsin cleaves only the region essential for the binding without any significant effect of the catalytic activity. Both spermine and Mg2+ antagonized the glucose 6-phosphate-induced release of mitochondria-bound hexokinase, and promoted the binding of the solubilized hexokinase II even in the presence of glucose 6-phosphate. However, inhibition of the activity of soluble hexokinase by glucose 6-phosphate was not reversed by spermine and Mg2+. Hexokinase II rebound to mitochondria with spermine and Mg2+ produced glucose 6-phosphate using ATP generated inside the mitochondria, and no difference was observed between the spermine- and Mg2+-rebound systems. Significance of the binding of hexokinase to mitochondria, especially with polyamines, is discussed with reference to high glycolytic rate in tumor cells.     

Altered metabolic states do not change the intracellular distribution of hexokinase in Zajdela hepatoma ascites cells.

The distribution of hexokinase between bound and soluble forms was studied by digitonin fractionation of Zajdela hepatoma ascites cells maintained under various metabolic conditions. Addition of glucose to Zajdela cells respiring on endogenous substrates induces an immediate inhibition of respiration by 50-60% ( Crabtree effect), and a production of acid due to glycolysis. Acid production decreases abruptly after 60s to 50% of the initial rate. The ATP/ADP ratio is not altered by the addition of glucose or by different rates of glycolysis. The uncoupling agent carbonyl cyanide m-chlorophenylhydrazone decreases the ATP/ADP ratio by 10-fold in cells respiring on endogenous substrate, but has little effect on cells oxidizing glucose. Rapid fractionation of the cells under these various metabolic conditions revealed no change in the distribution of hexokinase. Approx. 75% of hexokinase is bound in all cases, in contrast with lactate dehydrogenase, 95% of which was in the soluble form. Longer-term incubations (to 20 min) revealed only slight (10-15%) increases in soluble hexokinase in cells incubated with glucose. Various metabolic inhibitors had little additional affect on the subcellular distribution of hexokinase. Thus a rapid release of hexokinase from mitochondrial membrane is not a mechanism by which glycolysis is regulated in rapidly growing Zajdela hepatoma.