Rat brain hexokinase: location of the allosteric regulatory site in a structural domain at the N-terminus of the enzyme

After denaturation in 0.6 M guanidine hydrochloride, rat brain hexokinase becomes highly susceptible to proteolysis by trypsin. Glucose 6-phosphate (Glc-6-P) and its analog, 1,5-anhydroglucitol 6-phosphate, selectively protect the N-terminal half of the molecule from proteolysis. These compounds do not protect the C-terminal half of the molecule, nor do they protect enzyme activity; the Glc analog, N-acetylglucosamine, does protect the C-terminal domain and catalytic activity, but does not prevent proteolysis of the N-terminal half of the molecule. These results are consistent with previous work [M. Nemat-Gorgani and J. E. Wilson (1986) Arch. Biochem. Biophys. 251, 97-103; D. M. Schirch and J. E. Wilson (1987) Arch. Biochem. Biophys. 254, 385-396] demonstrating that binding sites for both hexose and nucleotide substrates, and thus catalytic function, are associated with a 40-kDa domain located at the C-terminus of the enzyme. They further demonstrate that the binding site for the allosteric effector, Glc-6-P, lies in the N-terminal half of the molecule and is distinct from the catalytic site. Using protection against proteolysis as a reflection of binding, it is shown that the Glc-6-P binding site in the N-terminal region has all the characteristics described for the allosteric effector site on this enzyme in terms of affinity for Glc-6-P, specificity, and synergistic interactions with the hexose binding site in the C-terminal region of the molecule. This disposition of catalytic and regulatory functions in discrete halves of the molecule is consistent with suggestions by several investigators that mammalian hexokinases evolved by a process of duplication and fusion of an ancestral gene coding for a hexokinase similar to the present-day yeast enzyme, with the regulatory site of mammalian hexokinases having evolved from what was originally a catalytic site.

Proteolysis of hexokinase PII is not the triggering signal of carbon catabolite derepression in Saccharomyces cerevisiae

The role of hexokinase PII in mediating carbon catabolite derepression in yeast has been examined. Hexokinase isoenzyme PII (EC 2.7.1.1) was partially degraded when protease inhibitors were omitted from the buffer used for preparation of cell-free extracts. The hexokinase PII inactivation induced by D-xylose was correlated with derepression of maltase (EC 3.2.1.20) in the wild-type strain Saccharomyces cerevisiae G-517 and in D.308.3, a strain that contains the cloned hexokinase PII gene on a multicopy plasmid. This inactivation was not correlated with the loss of hexokinase PII protein as assayed by immunoblotting. We conclude that during the derepression process there is no release of proteolytic peptides from hexokinase PII.

Inhibition of hexokinase activity by a fructose 2,6-bisphosphate-dependent cytosolic protein from liver

Mammalian and yeast hexokinases were found to be reversibly inhibited by fructose 2,6-bisphosphate, an effect requiring the presence of a cytosolic protein factor. Experimental evidence suggests that this factor (inhibitor) is a regulatory protein, the interactions of which with hexokinases are modulated by fructose 2,6-bisphosphate. The Vmax of hexokinase D was decreased, and no changes on other kinetic parameters were observed. The inhibitor was present in fresh liver cytosol filtered through Sephadex G-25 and was partially isolated by negative absorption on DEAE-cellulose followed by ammonium sulfate fractionation. The inhibitor was also present in brain and kidney, but not in muscle. A molecular mass of 200,000 was determined by gel filtration. The inhibition was dependent on the concentrations of both the inhibitory protein and fructose 2,6-bisphosphate. No delay in fructose 2,6-bisphosphate inhibition was observed. Several other hexose phosphates were tested and were not effective. In the presence of amounts of inhibitor sufficient to produce complete inhibition of hexokinase D, the concentration of fructose 2,6-bisphosphate required to produce 50% inhibition was about 0.5 microM. The inhibitor was unstable and was stabilized by the presence of fructose 2,6-bisphosphate.

Kinetic properties of hexokinase as assembled with a microcomputer data base

We have constructed a relational data base containing the kinetic properties of the isoenzymes of hexokinase using the Knowledgeman data-base program on an IBM PC microcomputer. The natural subunit of this data base is the refereed publication, 165 of which were included. Reported values for the Mr (approx. 97,000) are in good agreement, but this agreement becomes progressively worse as one examines the Km values for glucose and ATP and the Ki for glucose 6-phosphate, where the reported values are spread over three orders of magnitude. Some quantities are very thinly or unreliably determined. Experimental conditions, especially free Mg2+ concentration, are rarely close to physiological. Reasons for the spread or uncertainty of numbers, and the distinctions that can be made between isoenzymes despite this spread, are discussed.

Rat brain hexokinase: location of the substrate hexose binding site in a structural domain at the C-terminus of the enzyme

A glucose analog, N-(bromoacetyl)-D-glucosamine (GlcNBrAc), previously used to label the glucose binding sites of rat muscle Type II and bovine brain Type I hexokinases, also inactivates rat brain hexokinase (ATP:D-hexose 6-phosphotransferase, EC 2.7.1.1) with pseudo-first-order kinetics. Inactivation occurs predominantly via a "specific" pathway involving formation of a complex between hexokinase and GlcNBrAc, but significant nonspecific (i.e., without prior complex formation) inactivation also occurs, and equations to describe this behavior are derived. Inactivation is dependent on deprotonation of a residue with an alkaline pKa, consistent with the modified residue being a sulfhydryl group as reported to be the case with the hexokinase of bovine brain. The affinity label modifies three residues (per molecule of enzyme) at indistinguishable rates, but only one of these residues appears to be critical for activity. Amino acid analysis of the modified enzyme indicates derivatization of three cysteine residues; there was no indication of modification of other residues potentially reactive with haloacetyl derivatives. Kinetic analysis and effects of protective ligands were consistent with location of the critical sulfhydryl at the glucose binding site. Peptide mapping techniques permitted localization of the critical residue, and thus the glucose binding site, in a 40-kDa domain at the C-terminus of the enzyme. This is the same domain recently shown to include the ATP binding site. Thus, catalytic function is assigned to the C-terminal domain of rat brain hexokinase.

Effect of the presence of a reversible inhibitor on the time course of slow-binding inhibition

The half-time for the initial burst seen when a slow-binding inhibitor is present in an enzyme assay decreases from 0.693/k4 to 0.693/(k3 + k4) as the concentration of the slow-binding inhibitor is increased from zero to infinity (k3 and k4 are forward and reverse rate constants for the isomerization causing the slow-binding behavior). If the inhibitor solution contains a classical reversible inhibitor in addition to the slow-binding one, the half-time decreases from the same limit at zero inhibitor to a level which is higher at infinite inhibitor concentration (k3 is divided by (1 + xKi/Kj), where x is the ratio of classical and slow-binding inhibitor concentrations, and Ki and Kj are their initial inhibition constants before the slow-binding phase). Thus if one is using a racemic inhibitor, both enantiomers of which inhibit initially but only one of which shows slow-binding behavior, one will not obtain the correct parameters for the pure slow-binding inhibitor. A similar situation would apply if one were using a mixture of inhibitors such as antibiotics, several of which inhibit initially, but only one of which is a slow-binding inhibitor. This theory is illustrated by determining the half-times for the slow-binding inhibition of yeast hexokinase by various levels of TmATP in the presence and absence of HoATP, which shows little slow-binding behavior.

Enzymatic differences between hycanthone-resistant and sensitive strains of Schistosoma mansoni

1. Hycanthone-sensitive and resistant adult worms of Schistosoma mansoni were found to have generally similar specific activities in ten enzymes of carbohydrate metabolism. 2. Kinetic analyses revealed that pyruvate kinase, glucose-6-phosphate (G6P) dehydrogenase and malate dehydrogenase from both strains possessed similar Michaelis-Menten constants and were not inhibited by hycanthone. 3. Hexokinase and lactate dehydrogenase from the drug-resistant strain were not inhibited by hycanthone and showed three to five times greater Km values than those from the drug-sensitive worms which were also inhibitable by hycanthone. 4. Hycanthone more drastically affected the Vmax of phosphofructokinase from the hycanthone-sensitive parasite. 5. These data showed that the hycanthone inhibitable enzymes were generally from the drug-sensitive strain whereas the enzymes from drug-resistant worms are mostly hycanthone insensitive.

Mechanism of inactivation of hexokinase PII of Saccharomyces cerevisiae by D-xylose

The mechanism of inactivation of hexokinase PII of Saccharomyces cerevisiae by D-xylose was characterized. Inactivation was dependent on the presence of MgATP and was irreversible. Inactivation involved phosphorylation of the protein. Observation of the carbon catabolite repression of selected enzymes showed that invertase and maltase synthesis were not repressed when hexokinase PII was phosphorylated.

Allosteric character of the inhibition of rat-muscle hexokinase B by glucose 6-phosphate

We previously provided evidence from isotope-exchange measurements under non-equilibrium conditions that hexokinase B from rat muscle follows a compulsory-order mechanism with glucose binding before MgATP, and with both glucose 6-phosphate and MgATP capable of binding allosterically [Gregoriou, M., Trayer, I. P. & Cornish-Bowden, A. (1983) Eur. J. Biochem. 134, 283-288]. We have now re-examined this work in the light of recent criticisms [Ganson, N. J. & Fromm, H. J. (1985) J. Biol. Chem. 260, 12099-12105]. There is no difficulty in obtaining valid estimates of initial rates of isotope exchange when the equilibrium constant is unfavourable, if one uses highly radioactive reactants and low enzyme concentrations, as we did in the experiments we reported previously. However, our earlier suggestion that MgADP can be released within the inhibitory pathway, which was made for the sake of consistency with the catalytic pathway rather than because of any compelling experimental evidence, must be revised to avoid predicting that the rate must be zero in the absence of MgADP. Although our mechanism admits the possibility of substrate inhibition by MgATP, calculations show that there is no need for this to be observable under ordinary conditions. Indeed, with plausible values assumed for the kinetic constants one can calculate theoretical behaviour according to our model that closely resembles the experimental inhibition experiments that have been claimed as evidence against it.

Mechanistic origin of the sigmoidal rate behaviour of rat liver hexokinase D ('glucokinase')

Two recent proposals to account for the kinetic co-operativity of hexokinase D ('glucokinase') from rat liver are examined. A model in which the deviations from Michaelis-Menten kinetics result from a random order of binding of the substrates [Pettersson (1986) Biochem. J. 233, 347-350] accounts satisfactorily for the behaviour as a function of glucose concentrations, but it also predicts observable substrate inhibition by MgATP, which is in fact not observed. An alternative proposal in which the deviations arise from recycling of an enzyme-MgADP complex [Pettersson (1986) Eur. J. Biochem. 154, 167-170] also accounts satisfactorily for some of the data, but the required enzyme-MgADP complex could not be detected in isotope-exchange measurements. Thus the mnemonical mechanism proposed originally [Storer & Cornish-Bowden (1977) Biochem. J. 165, 61-69], which explains the deviations in terms of a relatively slow interconversion between two forms of free enzyme, remains the most parsimonious explanation of the behavior of hexokinase D.

Effect of xylose incubation on the glucose transport system in Saccharomyces cerevisiae

Incubation of Saccharomyces cerevisiae with xylose and ethanol for 16 hours leads to a decrease of hexokinase (and glucokinase) activity in the cells. It does not alter the levels of polyphosphate, orthophosphate and ATP. The transport of the glucose derivative 2-deoxy-D-glucose, a sugar that can be phosphorylated, is inhibited after this treatment, whereas transport of 6-deoxy-D-glucose, which has a blocked phosphorylation site, is not inhibited. Even though, both deoxyglucoses use the same transport system. The decrease in initial velocity of 2-deoxy-D-glucose transport is most pronounced under anaerobic conditions. Incubation of the cells with antimycin A, a treatment which has a similar effect as anaerobiosis, shows, that the inhibition of the transport of 2-deoxy-D-glucose is presumably the result of an increase in the Km of the carrier transport. Transport of glucose is probably regulated by kinase enzymes.

Red cell metabolism in hereditary pyrimidine 5'-nucleotidase deficiency: effect of magnesium

Since pyrimidine nucleotides avidly bind magnesium, we tested the hypothesis that the haemolytic anaemia in hereditary pyrimidine 5'-nucleotidase (P5N) deficiency is due to a state of functional magnesium depletion in the red cell (RBC). In haemolysates from normal subjects, cytidine triphosphate (CTP) inhibited the activity of pyruvate kinase in a competitive manner for magnesium. The CTP Ki was 0.4 mmol/l. CTP inhibited the activity of hexokinase in a competitive manner for ATP (Mg-ATP2-) with a Ki of 4 mmol/l. The inhibitory effect of CTP on both enzymes was overcome by increasing the magnesium content of the test system. Since CTP appeared to inhibit enzymes which required magnesium as a cofactor or Mg-ATP2- as a substrate, we tested the effect of exogenous magnesium on the metabolism of P5N deficient RBC. The autohaemolysis test, the incubated Heinz body assay and the rate of glucose oxidation by the pentose phosphate shunt were abnormal in the intact RBC from a patient with hereditary P5N deficiency. The addition of MgCl2 (6-10 mmol/l) did not improve these abnormal in vitro measures of metabolism in the P5N deficient RBC. This lack of effect of exogenous magnesium may be due to the slow uptake of magnesium by the human RBC.

[Interaction of hexokinase II isoenzyme from rat skeletal muscles with lecithin liposomes]

It was found that in the presence of Mg2+ (pH 7.5) rat skeletal muscle hexokinase isozyme II is firmly adsorbed on mitochondrial and artificial phospholipid membranes (lecithin liposomes). In both cases the adsorption isotherm has similar quantitative and qualitative characteristics, which points to the absence of specific binding sites on the membranes. Under these conditions, immobilization of hexokinase on various membranes is concomitant with similar changes in the enzyme stability upon storage as well as with the pH-dependence of the enzyme activity. It was demonstrated that the bound hexokinase form has a greater value of V, an increased affinity for glucose and a decreased sensitivity to the inhibitory action of glucose-6-phosphate as compared to the free form. Besides, this form is in a greater degree subjected to the inhibitory influence of ADP with respect to glucose. In this case, the enzyme affinity for ATP and the Ki value for ADP with respect to ATP is practically the same both for the free and membrane-bound forms. The data obtained suggest that the phospholipid component of mitochondrial membranes participates in the enzyme binding in the presence of Mg2+. It was assumed that the model system used in the present study, i.e., hexokinase-Mg2+-liposomes, may be successfully used for the analysis of an adsorption mechanism of regulation of hexokinase activity in the cell.

The effect of the sesquiterpene lactones from Geigeria on glycolytic enzymes

The effect of sesquiterpene lactones isolated from Geigeria was tested on three glycolytic enzymes. Phosphofructokinase was inhibited irreversibly by all of the sesquiterpene lactones, with ivalin(III) giving the highest extent of inhibition. Values for the kinetic constants Ki (1.3 mM) and kp (2.2 min-1) were established. Hexokinase and glyceraldehyde-3-phosphate dehydrogenase were also strongly inhibited at 1 mM and 3 mM concentrations of sesquiterpene lactones, respectively. Pre-incubation of ivalin with dithiothreitol decreased its inhibiting effect on phosphofructokinase, hexokinase and glyceraldehyde-3-phosphate dehydrogenase activities. Phosphofructokinase and hexokinase were protected against inhibition by ivalin by their respective substrates, adenosine-5'-triphosphate and glucose.

Anomeric specificity of hexokinase in rat, human, and murine tumor cells

In tumoral cells derived from the insulin-producing rat cell line RINm5F, both low- and high-Km glucose-phosphorylating enzymic activities were present. The hexokinase-like enzyme was inhibited by glucose 6-phosphate and displayed a greater affinity for but lower maximal velocity with alpha-D-glucose than beta-D-glucose. A comparable anomeric behavior of hexokinase was observed in breast cancer (MCF-7) and lymphocytic leukemia (P388) cells. Thus, the anomeric specificity of hexokinase in tumoral cells was not different from that recently characterized in normal mammalian cells.

Is glucokinase responsible for the anomeric specificity of glycolysis in pancreatic islets?

At a low concentration of D-glucose (3.3 mM), the phosphorylation rate of this hexose in rat pancreatic islet homogenates incubated at 8 degrees C is higher with the beta- than with the alpha-anomer, as expected from the anomeric specificity of hexokinase. In the presence of a high concentration of glucose 6-phosphate (3.0 mM), which inhibits hexokinase but not glucokinase, the phosphorylation rates of the two anomers are not significantly different from one another. Nevertheless, in intact islets exposed at 8 degrees C to the same low concentration of D-glucose, the alpha-anomer augments, more than the beta-anomer, the production of lactic acid and net uptake of 45Ca. At the same concentration (3.3 mM), the alpha-anomer is also more potent than the beta-anomer in enhancing insulin release from perfused pancreases stimulated at 37 degrees C by L-leucine or by the combination of Ba2+ and theophylline. It is concluded that the participation of glucokinase is not essential for the anomeric specificity of glycolysis and insulin release in rat pancreatic islets.

Inhibition and inactivation of glucose-phosphorylating enzymes from Saccharomyces cerevisiae by D-xylose

Three glucose-phosphorylating enzymes were separated from cell-free extracts of Saccharomyces cerevisiae by hydroxylapatite chromatography. Variations in the amounts of these enzymes in cells growing on glucose and on ethanol showed that hexokinase PI was a constitutive enzyme, whereas synthesis of hexokinase PII and glucokinase were regulated by the carbon source used. Glucokinase proved to be a glucomannokinase with Km values of 0.04 mM for both glucose and mannose. D-Xylose produced an irreversible inactivation of the three glucose-phosphorylating enzymes depending on the presence or absence of ATP. Hexokinase PI inactivation required ATP, while hexokinase PII was inactivated by D-xylose without ATP in the reaction mixture. Glucokinase was protected by ATP from this inactivation. D-Xylose acted as a competitive inhibitor of hexokinase PI and glucokinase and as a non-competitive inhibitor of hexokinase PII.

Hexokinase in human chorionic villi

The level, intracellular distribution, and isozymic pattern of hexokinase (EC 2.7.1.1) were determined on human chorionic villi obtained by trophoblast biopsy in the first trimester of pregnancy. About 50% of total hexokinase activity was found to be particle-bound and 96% of this in the overt form (i.e. assayable without the addition of detergents). Both soluble and particulate hexokinase show the same affinities for glucose but differ in the affinity for MgATP2- and in their sensitivity to glucose 1,6-diphosphate inhibition. By chromatographic and kinetic methods the soluble hexokinase was found to be represented by isozymes I, II and traces of hexokinase III while most of the bound enzyme was hexokinase type I. These results provide evidence for the expression of at least three hexokinase isozymes in early human development and can partially explain the high rate of glucose utilization of the placenta.

Allosteric inhibition of brain hexokinase by glucose 6-phosphate in the reverse reaction

A study of the reverse reaction of rat brain hexokinase (ATP:D-hexose 6-phosphotransferase, EC 2.7.1.1) has been performed using a photometric method based on a mutarotase-glucose oxidase-peroxidase-chromogen system to trap and visualize glucose, plus a glycerol kinase-glycerol system to trap ATP. Glucose 6-phosphate or 2-deoxyglucose 6-phosphate were used as phosphoryl donors at different concentrations of ADP. Variation of glucose 6-phosphate concentrations resulted in a biphasic curve from which apparent Km and Ki values of ca. 0.2 mM were calculated. In contrast, variation of 2-deoxyglucose 6-phosphate concentrations resulted in Michaelian kinetics with an apparent Km of 2 mM. The Km value for MgADP was 16 mM irrespective of the nature and concentration of the hexose 6-phosphate substrate. These results are fully consistent with an allosteric site for glucose 6-phosphate as an explanation for the inhibition of animal hexokinases by glucose 6-P and further indicate that the maximal rate is the parameter affected. From these observations and previous knowledge, the possible occurrence in animal hexokinases of a regulatory site for ATP to account for the competition between glucose 6-phosphate and ATP in the forward reaction is postulated.

High glucose concentrations partially release hexokinase from inhibition by glucose 6-phosphate

The phosphorylation of glucose by human erythrocyte hexokinase follows classical Michaelis-Menten kinetics; hexokinase manifests maximum activity at 5 mM glucose, and no further increase in activity can be measured at higher glucose concentrations. However, the erythrocytes of diabetics and normal erythrocytes incubated with high concentrations of glucose contain increased concentrations of glucose 6-phosphate. To elucidate the mechanism of accumulation of glucose 6-phosphate when erythrocytes are exposed to high glucose concentrations, hexokinase activity was examined in the presence of naturally occurring inhibitors, such as glucose 1,6-bisphosphate, 2,3-diphosphoglycerate, ADP, and glucose 6-phosphate at physiological concentrations. Without inhibitors or in the presence of glucose 1,6-bisphosphate,2,3-diphosphoglycerate, and ADP, maximum hexokinase activity was observed at 5 mM glucose concentration. On the contrary, in the presence of glucose 6-phosphate, hexokinase activity increased at glucose concentrations greater than 5 mM; inhibition by glucose 6-phosphate was partially competitive with glucose. The relief by glucose of glucose 6-phosphate inhibition of hexokinase is a possible explanation of the increased glucose 6-phosphate level in diabetic erythrocytes.

Glucose repression and hexokinase isoenzymes in yeast. Isolation and characterization of a modified hexokinase PII isoenzyme

Hexokinase PII, but not isoenzyme PI, has a unique role in glucose repression in yeasts [Entian, K.-D. (1980) Mol. Gen. Genet. 178, 633-637; Entian, K.-D. and Mecke, D. (1982) J. Biol. Chem. 257, 870-874; Entian, K.-D. and Fröhlich, K.-U. (1984) J. Bacteriol. 158, 29-35]. The number of hexokinase isoenzymes in crude extracts was re-examined by chromatofocusing. In addition to the known isoenzymes PI and PII, a third isoenzyme, PIIM, was detected. The activity of this enzyme was only about 5-10% of that of hexokinase PII and was independent of growth conditions. Experiments with hexokinase transformants and purified hexokinase isoenzymes clearly indicated that the PIIM form is also present in vivo. Fingerprint mapping of purified hexokinases showed that hexokinase PIIM is closely related to PII. Hybridization experiments between totally restricted yeast DNA and the previously isolated PII gene clearly indicated that PIIM is also coded by one of the two known hexokinase genes. No mRNA specific for hexokinase PIIM was detected after hybridization experiments with the previously cloned hexokinase PII gene [Fröhlich et al. (1984) Mol. Gen. Genet. 194, 144-148]. Hexokinase PIIM appears to be derived from hexokinase PII by a posttranslational event. The Km values of each of the purified isoenzymes, PII and PIIM, were identical for glucose, fructose and ATP. Both isoenzymes were strongly inhibited by high physiological concentrations for ATP; such inhibition has not been described previously. The possible role of hexokinase PIIM in glucose repression is discussed.

In vivo and in vitro inhibition of B16 melanoma growth by vitamin B6

The effect of vitamin B6 on the growth of B16 melanoma cells in vivo and in vitro was studied. B16 melanoma cells grown for three days in medium supplemented with 5.0 mM pyridoxine or 0.5 mM pyridoxal showed an 80% reduction in cell proliferation compared with control culture. Cells cultured for six hours in medium supplemented with 0.5 mM pyridoxal took up and incorporated 13 and 32% less [3H]thymidine, respectively, than did control cultures. A 17% reduction in [3H]glucose uptake was observed at this time point. When the incubation time was decreased to three hours, an inhibition of cellular uptake of [3H]thymidine (22%), [3H]uridine (14%), and [3H]glucose (15%) was observed; however, little or no inhibition in incorporation was detected. In in vivo studies, mice pretreated with pyridoxal for two weeks and then injected with B16 melanoma cells had a 62% reduction in tumor weight compared with controls at the end of a three-week period. If tumors were first established in mice and then treated with pyridoxal for six days, a 39% reduction in tumor growth was observed. There were no differences observed in body weights or liver weights in any of the animal groups. These results indicate that supraphysiological doses of vitamin B6 can inhibit the growth of B16 melanoma cells both in vitro and in vivo. The exact mechanism by which pyridoxal exerts its inhibitory effect was not ascertained, but experiments suggest that the vitamer may be acting on the plasma membrane to reduce precursor transport into the cell.

Differential effects of monovalent, divalent and trivalent metal ions on rat brain hexokinase

The effects of monovalent (Li+, Cs+) divalent (Cu2+, Ca2+, Sr2+, Ba2+, Zn2+, Cd2+, Hg2+, Pb2+, Mn2+, Fe2+, Co2+, Ni2+) and trivalent (Cr3+, Fe3+, Al3+) metals ions on hexokinase activity in rat brain cytosol were compared at 500 microM. The rank order of their potency as inhibitors of brain hexokinase was: Cr3+ (IC50 = 1.3 microM) greater than Hg2+ = Al3+ greater than Cu2+ greater than Pb2+ (IC50 = 80 microM) greater than Fe3+ (IC50 = 250 microM) greater than Cd2+ (IC50 = 540 microM) greater than Zn2+ (IC50 = 560 microM). However, at 500 microM Co2+ slightly stimulated brain hexokinase whereas the other metal ions were without effect. That inhibition of brain glucose metabolism may be an important mechanism in the neurotoxicity of metals is suggested.