Role of arginyl residues in yeast hexokinase PII.

Yeast hexokinase PII is rapidly inactivated (assayed at pH 8.0) by either butanedione in borate buffer or phenylglyoxal, reagents which are highly selective for the modification of arginyl residues. MgATP alone offers no protection against inactivation, consistent with low affinity of hexokinase for this nucleotide in the absence of sugar. Glucose provides slight protection against inactivation, while the combined presence of glucose and MgATP gives significant protection, suggesting that modified arginyl residues may lie at the active site, possibly serving to bind the anionic polyphosphate of the nucleotide in the ternary enzyme:sugar:nucleotide complex. Extrapolation to complete inactivation suggests that inactivation by butanedione correlates with the modification of 4.2 arginyl residues per subunit, and complete protection against inactivation by the combined presence of glucose and MgATP correlates with the protection of 2 to 3 arginyl residues per subunit. When the modified enzyme is assayed at pH 6.5, significant activity remains. However, modification by butanedione in borate buffer abolishes the burst-type slow transient process, observed when the enzyme is assayed at pH 6.5, to such an extent that after extensive modification the kinetic assays are characterized by a lag-type slow transient process. But even after extensive modification, hexokinase PII still demonstrates negative cooperativity with MgATP and is still strongly activated by citrate when assayed at pH 6.5.     

Specificity of ligand binding to yeast hexokinase PII studied by STD-NMR

Hexokinase catalyzes the phosphorylation of glucose and is the first enzyme in glycolysis. To investigate enzyme–ligand interactions in yeast hexokinase isoform PII under physiological conditions, we utilized the technique of Saturation Transfer Difference NMR (STD NMR) to monitor binding modes and binding affinities of different ligands at atomic resolution. These experiments clearly show that hexokinase tolerates several changes at C-2 of its main substrate glucose, whereas epimerization of C-4 significantly reduces ligand binding. Although both glucose anomers bind to yeast hexokinase, the α-form is the preferred form for the phosphorylation reaction. These findings allow mapping of tolerated and prohibited modification sites on the ligand. Furthermore, competitive titration experiments show that mannose has the highest binding affinity of all examined sugars. As several naturally occurring sugars in cells show binding affinities in a similar range, hexokinase may be considered as an ‘emergency enzyme’ in yeast cells. Taken together, our results represent a comprehensive analysis of ligand–enzyme interactions in hexokinase PII and provide a valuable basis for inhibitor design and metabolic engineering.     

Cloning and restriction analysis of the hexokinase PII gene of the yeast Saccharomyces cerevisiae

Summary Carbon catabolite repression in yeast depends on catalytic active hexokinase isoenzyme PII (Entian 1980a). A yeast strain lacking hexokinase isoenzymes PI and PII was transformed, using a recombinant pool with inserts of yeast nuclear DNA up to 10 kbp in length. One hundred transformants for hexokinase were obtained. All selected plasmids coded for hexokinase isoenzyme PII, none for hexokinase isoenzyme PI, and carbon catabolite repression was restored in the transformants. Thirty-five independently isolated stable plasmids were investigated further. Analysis with the restriction enzyme EcoRI showed that these plasmids fell into two classes with different restriction behaviour. One representative of each class was amplified in Escherichia coli and transferred back into the yeast hexokinase-deficient strain with concomitant complementation of the nuclear mutation. The two types of insert were analysed in detail with 16 restriction enzymes, having 0–3 cleavage sites on transformant vector YRp7. The plasmids differed from each other by the orientation of the yeast insert in the vector. After yeast transformation with fragments of one plasmid the hexokinase PII gene was localised within a region of 1.65 kbp.     

Genetic and biochemical evidence for hexokinase PII as a key enzyme involved in carbon catabolite repression in yeast

Summary Mutants with reduced hexokinase activity previously isolated as resistant to carbon catabolite repression of invertase and maltase (Zimmermann and Scheel, 1977) were allele tested with mutant strains of Lobo and Maitra (1977) which had defects in one or several of the genes coding for glucokinase and the two unspecific hexokinases. It could be demonstrated, that the mutation abolishing carbon catabolite repression had occurred in a gene allelic to the structural gene of hexokinase PII. Moreover, the defective mutant allele for hexokinase PII isolated by Lobo and Maitra (1977) was also defective in carbon catabolite repression. Neither glucokinase nor hexokinase PI showed any effect on this regulatory system. Biochemical analysis in crude extracts also showed altered kinetic properties of hexokinases in the hex1 mutants. The results directly support the hypothesis previously put forward, that one of the hexokinases is not only active as a catalytic, but also as a regulatory protein.     

The primary structure of the yeast hexokinase PII gene (HXK2) which is responsible for glucose repression

The nucleotide sequence of the Saccharomyces cerevisiae gene encoding the glycolytic isoenzyme hexokinase PII (HXK2), which is responsible for triggering glucose repression, has been determined. The reading frame was identified by comparison with the N-terminal undecameric amino acid (aa) sequence, determined previously [Schmidt and Colowick, Arch. Biochem. Biophys. 158 (1973) 458-470]. The codon sequence was not random, with 82.1% of the aa specified by only 25 codons. The structural gene sequence corresponded to 1455 bp, coding for 485 aa residues, corresponding to the Mr of 53 800 for the HXK2 monomer. Five initiation regions spanning 162 bp and three termination sites spanning 29 bp were detected. Sequences with similarities to a 5'-TATAAA-3' sequence were located 24-39 bp upstream of each initiation region. The most pronounced initiation region corresponded to the 5'-TATAAA-3' sequence at position -152. Two of the minor initiation sites were inside the coding sequence in front of two ATG codons.     

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.     

Genetics of yeast hexokinase

Two independent isolates of Saccharomyces cerevisiae lacking hexokinase activity (EC 2.7.1.1) are described. Both mutant strains grow on glucose but are unable to grow on fructose, and contain two mutant genes hxk1 and hxk2 each. The mutations are recessive and noncomplementing. Genetic analysis suggests that these two unlinked genes hxk1 and hxk2 determine, independently of each other, the synthesis of hexokinase isozymes P1 and P2, respectively. hxk1 is located on chromosome VIR distal to met10, and hxk2 is on chromosome IIIR distal to MAL2. Of four hexokinase-positive spontaneous reversions, one is very tightly linked to hxk1 and the other three to the hxk2 locus. The reverted enzymes are considerably more thermolabile than the respective wild-type enzymes, and in one case show altered immunological properties. Data are presented which suggest that the hxk1 and hxk2 mutations are missense mutations in the structural genes of hexokinase P1 and hexokinase P2, respectively. These are presumably the only enzymes that allow S. cerevisiae to grow on fructose.     

The hexokinase isoenzyme PII of Saccharomyces cerevisiae ia a protein kinase

The HXK2 gene product has an important role in controlling carbon catabolite repression in Saccharomyces cerevisiae. We have raised specific antibodies against the hexokinase PII protein and have demonstrated that it is a 58 kDa phosphoprotein with protein kinase activity. The predicted amino acid sequence of the HXK2 gene product has significant homology to the conserved catalytic domain of mammalian and yeast protein kinases. Protein kinase activity was located in a different domain of the protein from the hexose-phosphorylating activity. The hexokinase PII protein level remained unchanged in P2T22D mutant cells (hxk1 HXK2 glk1) growing in a complex medium with glucose. The protein kinase activity of hexokinase PII is regulated by the glucose concentration of the culture medium. Exit from the carbon catabolite repression phase and entry into derepression phase may be controlled, in part, by modulation of the 58 kDa protein kinase activity by changes in cyclic AMP concentration.     

The adenine nucleotide binding site on yeast hexokinase PII. Affinity labeling of Lys-111 by pyridoxal 5'-diphospho-5'-adenosine

The adenine nucleotide analog, [3H]pyridoxal 5'-diphospho-5'-adenosine (PLP-AMP), is shown to be a potent and specific inhibitor of yeast hexokinase PII. Evidence that the analog binds specifically at the ATP binding site includes the demonstration that glucose binding enhances PLP-AMP binding and that PLP-AMP and ATP bind competitively with an apparent Ki(PLP-AMP) = 23 microM. In addition, from the relationship between the degree of inhibition and extent of modification, it is estimated that the incorporation of 1 mol of PLP-AMP/mol of subunit is required for complete inhibition. Borohydride reduction of the Schiff's base complex formed between hexokinase and [3H]PLP-AMP gives a stable product. The reduced derivative was digested with trypsin and a single radioactive peptide was isolated by reversed-phase high-pressure liquid chromatography. Amino acid sequence analysis identified Lys-111 as the modified residue. Taking into account the known structures of the binary complexes (Shoham, M., and Steitz, T. A. (1980) J. Mol. Biol. 140, 1-14), the results suggest that Lys-111, located in the smaller of the two lobes of hexokinase, moves into the active site upon formation of the ternary complex.     

The high resolution crystal structure of yeast hexokinase PII with the correct primary sequence provides new insights into its mechanism of action

Hexokinase is the first enzyme in the glycolytic pathway, catalyzing the transfer of a phosphoryl group from ATP to glucose to form glucose 6-phosphate and ADP. Two yeast hexokinase isozymes are known, namely PI and PII. The crystal structure of yeast hexokinase PII from Saccharomyces cerevisiae without substrate or competitive inhibitor is determined and refined in a tetragonal crystal form at 2.2-A resolution. The folding of the peptide chain is very similar to that of Schistosoma mansoni and previous yeast hexokinase models despite only 30% sequence identity between them. Distinct differences in conformation are found that account for the absence of glucose in the binding site. Comparison of the current model with S. mansoni and yeast hexokinase PI structures both complexed with glucose shows in atomic detail the rigid body domain closure and specific loop movements as glucose binds. A hydrophobic channel formed by strictly conserved hydrophobic residues in the small domain of the hexokinase is identified. The channel's mouth is close to the active site and passes through the small domain to its surface. The possible role of the observed channel in proton transfer is discussed.     

Fluorescence quenching of dimeric and monomeric forms of yeast hexokinase (PII): effect of substrate binding steady-state and time-resolved fluorescence studies

Fluorescence quenching studies on the PII isoenzyme of yeast hexokinase have been performed using charged as well as polar uncharged quenchers. In both 'open' (i.e. in the absence of glucose) and 'closed' (i.e. in the presence of glucose) forms of the enzyme, bimolecular quenching rate constant (kq) for acrylamide is significantly larger than that of KI, indicating that all the tryptophans are not fully exposed to the solvent. Overall accessibility of tryptophans towards KI was greater in the presence of glucose than in the absence of glucose. At high ionic strength, the value of bimolecular quenching rate constant (kq) for KI did not change suggesting that the average environment of the accessible tryptophan residue(s) is almost neutral. Quenching by KI is dynamic in nature. Accessibility of tryptophans towards acrylamide at concentration > or = 0.2 M was more in the 'open' form of the enzyme than that observed in the 'closed' form whereas at concentration < or = 0.2 M no significant difference in the extent of quenching was observed. It is reasonable to conclude that glucose induced conformational change leads some tryptophan residue(s) to be more exposed and at the same time some tryptophan residue(s) in the hydrophobic region become more buried. Dimeric and monomeric forms of the enzyme behave similarly towards the quenching by acrylamide. In the unfolded state, the accessibility of tryptophans was considerably higher for both the quenchers. Temperature dependent study and the fluorescence lifetime data indicate that the mechanism of quenching by acrylamide is primarily dynamic in nature.     

Effect of alcohol on yeast hexokinase

Summary The effect of alcohol upon hexokinase activity has been examined and it has been found that in the concentration range of 0.015 to 1.5 M inhibition varies directly with concentration. Increasing alcohol concentration to 2.25 M resulted in less inhibition than at 1.5 M. No evidence was obtained that hexokinase activity was accelerated by use of very low concentrations (0.0015 M) alcohol.     

Isolation of hexokinase from baker's yeast

1. A method is described for the isolation of hexokinase from baker''s yeast. The method is based mainly on fractionation with alcohol and results See PDF for Structure in a 30-fold increase in specific activity. The final product could be crystallized from ammonium sulfate without change in specific activity. 2. The enzyme catalyzes a transfer of phosphate from adenosinetriphosphate to glucose, fructose, or mannose, the relative rates with these three sugars being 1:1.4:0.3. 3. With glucose as substrate, the turnover number for the crystalline enzyme is 13,000 moles of substrate per 10⁵ gm. of protein per minute at 30° and pH 7.5. The temperature coefficient (Q _10°) between 0 and 30° is 1.9. 4. Magnesium ions are necessary for the activity, the dissociation constant for the Mg⁺⁺ -protein complex being 2.6 x 10^–3. Fluoride in concentrations as high as 0.125 M has no inhibitory effect on the enzyme when the Mg⁺⁺ and orthophosphate concentrations are 6.5 x 10^–3 M and 1 x 10^–3 M, respectively. 5. The crystalline enzyme shows a loss in activity when highly diluted. This loss in activity can be prevented by diluting in the presence of small amounts of other proteins. Of the various protective proteins tested, insulin was the most effective, providing complete protection in a concentration of 6 micrograms per cc.; with serum albumin, a concentration of 60 micrograms per cc. was necessary. Thiol compounds (cysteine, glutathione) exerted no protective action. 6. The inactivation of the crystalline enzyme on incubation with trypsin can be prevented to a marked degree by the presence of glucose. The instability of crude preparations of yeast hexokinase may be attributed to the presence of proteolytic enzymes, since glucose or fructose has a remarkable protective effect on such preparations.