Mechanistic studies on DNA gyrase from Escherichia coli.

DNA gyrase is an essential bacterial type II topoisomerase which couples the free energy of ATP hydrolysis to the introduction of negative supercoils into DNA. This study concentrates on the interaction of the enzyme with ATP and on the way the free energy of hydrolysis of the nucleotide is coupled to the supercoiling reaction. Positional isotope exchange experiments, using ATP labelled with 18O in the ?-?-bridge position, have shown that no detectable scrambling of the label occurs during the gyrase ATPase. This is interpreted in terms of a slow ATP off-rate in gyrase. Gyrase-catalysed turnover of ATP?S was found to be 300 to 1000-fold slower than ATP from supercoiling and hydrolysis measurements. This precluded the use of isotopically chiral ATP?S to determine the stereochemistry of the gyrase ATPase, although the availability of more protein in the future may permit such a study. Various aspects of the interaction of gyrase with diastereoisomers of ATP?S and ATP?S have been explored which has yielded information on the type of MgATP complex handled by the enzyme and on the relationship between hydrolysis of the ATP analogues and their ability to support DNA supercoiling. Gyrase showed a strong preference for the Rp epimers of ATP?S and ATP?S in the presence of Mg2+ which is consistent with the Mg2+ ion being coordinated to the pro-S oxygens of the a and ?-phosphates in ATP bound to the enzyme; an ?,?,?-tridentate Mg-ATP complex with A-exo geometry is proposed. ATP?S(Rp) hydrolysis appears to be well coupled to DNA supercoiling whereas ATP?S(Rp) hydrolysis is only poorly coupled to the supercoiling reaction. Two hypotheses are proposed to account for this phenomenon. The divalent metal ion specificities of the DNA supercoiling, relaxation, and cleavage reactions of gyrase have been characterised. It was found that the cleavage reaction exhibited a greater tolerance towards different metal ions than the relaxation and supercoiling reactions; the mechanistic implications of this observation are discussed. An N-terminal fragment of the gyrase B protein was used as a simple model for gyrase-ATP interactions. This truncated protein showed the same stereospecificity towards the diastereoisomers of ATP?S and ATP?S as intact gyrase but the kinetics of hydrolysis of the two phosphorothioate ATP analogues showed significant differences from those with gyrase, which may be due to differences in the mechanisms of ATP hydrolysis for the two enzymes. Attempts to demonstrate reversal of stereospecifity on switching from a hard metal ion such as Mg2+ to a soft metal centre such as Cd2+ were thwarted by the inhibitory effects of Cd2+. The extent of supercoiling of pBR322 was investigated with ATP, ATP?S(Rp), and ATP?S(Rp). With ATP?S(Rp) the rate of DNA supercoiling was very slow (since its hydrolysis is inefficiently coupled to supercoiling; see above) and the reaction did not reach a limit even after long time-courses. With ATP and ATP?S(Rp), the supercoiling reaction was faster and did reach a limit: ATP gave a ?Lk of -46 and ATP?S(Rp) gave a ?Lk of -47. Measurements of the displacement of the equilibrium of the reaction catalysed by arginine kinase showed that ATP?S(RP) has a greater free energy of hydrolysis than ATP. This energy difference corresponds closely to the extra free energy required for the additional supercoiling seen with ATP?S(Rp). It is proposed that the extent of supercoiling in gyrase is limited by the free energy available from nucleotide hydrolysis.