Properties of a CH,S-blocked Creatine Kinase with Altered Catalytic Activity KINETIC CONSEQUENCES OF THE PRESENCE OF THE BLOCKING GROUP*

Steady state kinetic parameters for rabbit muscle creatine kinase (EC 2.7.3.2) and this enzyme stoichiometrically blocked at the iodoacetamide-sensitive cysteinyl residue with a CH,S-group have been measured at 30 + O.l”, pH 9.00, using Mg(II) as the required metal ion cofactor. The double reciprocal plots for the CH,S-blocked enzyme with MgATP as the variable substrate are biphasic, each curve showing a break at -1.9 mM MgATP, and suggest the possibility of negative cooperativity in metal-nucleotide binding. Furthermore, extrapolated lines at high MgATP concentrations intersect on the abscissa, indicating loss of synergism in binding of substrates. contrast, for creatine are, within experimental the same for both native and blocked

* The preceding paper in this series is the accompanying manuscript (1) Rabbit muscle creatine kinase has only one highly reactive cysteinyl residue per subunit (3). This sulfhydryl group has been blocked using numerous reagents, including iodoacetate (3,4), iodoacetamide (5), 2,4-dinitrofluorobenzene (3), disodium tetrathionate (6), 5,5'-dithiobis (2-nitrobenzoic acid) (7,8), and the spin labelN-(l-oxy-2,2,5,5-tetramethyl-3-pyrrolidinyl)iodoacetamide (9,10). In all of these cases, the authors reported that modification of this sulfhydryl group leads to essentially complete inactivation of the enzyme. In contrast, blocking this sulfhydryl group with the relatively small CHSS -group leads to enzyme with substantial (-20%) residual enzymatic activity (8,11). This residual activity provides a unique opportunity to examine the kinetic consequences of chemical modification of this important amino acid residue. In the present study a direct comparison is made between various steady state kinetic parameters of native and CH&blocked enzymes. In the preceding paper (1) proton relaxation rate enhancements and EPR spectra of native, CH,S-blocked, and iodoacetamide-inactivated creatine kinase are compared for a variety of ternary and quaternary enzyme. substrate complexes using the paramagnetic Mn(I1) ion as a spectroscopic probe.  (12). Methanethiolation of creatine kinase was carried out as previously described (8,11)  fully restores the properties of the native enzyme, also supports the contention that a sulfhydryl group is blocked (8,11). Furthermore, stoichiometric blocking of at least one of the other three sulfhydryl groups per subunit using organomercurials leads to no observable loss of catalytic activity, and leaves the enzyme completely sensitive to inactivation by iodoacetamide (11,18,19 (22), and the CH,S-blocking group may be Lines in Fig. 2b extrapolated from low MgATP concentrations (cl.9 mM MgATP) intersect above the abscissa, as is observed with the native enzyme. Lines in Fig. 2b extrapolated from higher MgATP concentrations (>1.9 mM) intersect close to or on the abscissa, suggesting a substantial decrease in the synergism in MgATP and creatine binding. Statistical analysis of the kinetic data (cf. Table I) confirms this loss of synergism. Comparison of maximal velocities (expressed as k,,, or turnover number) for the native and CH,S-blocked enzymes shows that the true maximal velocity of the modified enzyme is 28.1% of the value of the native enzyme, although the curvature of the double reciprocal plot would lead to an estimate of only 19.9% of the native enzyme value if only low MgATP (cl.9 mM) concentrations were examined.4 Fig. 3 shows intercept plots for MgATP for both native and CH,S-blocked creatine kinase. Also shown are extrapolated lines from data points obtained at high (>1.2 mM) and low (cl.2 mM) MgATP for the CH,S-blocked creatine kinase. The figure shows that, within experimental error, the observed value for K,,, (creatine) is unaltered by the presence of the CH,S-blocking group.

MATERIALS AND METHODS
Product Inhibition -Product inhibition patterns can distinguish among mechanistic schemes having rate equations of identical form. Fig. 4, a and b, shows selected product-inhibition patterns for both native and CH,S-blocked creatine kinase. Secondary plots (not shown) are all linear and were used to evaluate inhibition constants. MgADP is competitive with respect to MgATP and phosphocreatine is competitive with respect to creatine. These observations are consistent with a rapid equilibrium, random, sequential mechanistic scheme for both native and blocked enzyme. The results are summarized in Table II. For both enzymes the rate-determining step is the interconversion of the ternary E. MgATP . creatine and E. MgADP. phosphocreatine complexes. Thus, the kinetic constants reflect binding affinities of metal nucleotide and guanidino substrates and k,,, is the rate constant for interconversion of the ternary complexes.
Phosphocreatine shows only a slight increase in Ki from the value found for the native enzyme. This is similar to the lack of any significant difference of K,,, of creatine between the native enzyme and the CH,S-blocked creatine kinase at either low or high ATP concentration.
On the other hand, the nucleotide binding site is apparently somewhat altered since Ki (MgADP) for the CH,S-blocked enzyme is 3-fold greater than for the native enzyme.
Synergism in substrate binding is presumably associated with substrate-induced conformational changes within the enzyme's tertiary structure. In the presence of the CH,S-blocking group this synergism is lost at high MgATP concentrations; 4, (creatine) decreases a-fold to become equal to K,,, (creatine) and approaches K,,, (creatine) for the native enzyme. Thus it appears that in the presence of the CH,S-blocking group the conformation resulting from the binding of MgATP at one subunit may be transmitted to the adjacent protomer. This conformational change leads to weakening of the binding of MgATP since both K,,, (MgATP) and h?, (MgATP) are increased at high MgATP concentrations, relative to the native 4 In preliminary investigations of Dr. D. J. Smith (231, methyl methanethiolsulfonate treatment of a creatine kinase isoenzyme isolated from the cytoplasm of beef heart (24) led to a CH,S-blocked enzyme whose V,,, was 23 -c 2% that of the native enzyme. In addition, K,. values for creatine and MgATP were only slightly altered.
Intense clinical interest in the creatine kinase isoenzyme from heart tissue in the monitoring of myocardial infarctions (25) prompted our investigations of its comparative behavior to that of the rabbit muscle isoenzyme.  enzyme. Furthermore the affinity for MgADP is diminished slightly by modification, as was found for MnADP and free ADP by proton relaxation rate titrations (1). Thus the majority of the effects of the CH,S-label are reflected in the properties of the metal-nucleotide binding site.
Structural Predictions -Some insight into a possible structural role for the iodoacetamide-sensitive sulfhydryl group in native and blocked enzymes emerges using Chou-Fasman calculations (26,27). These calculations allow prediction of secondary structural features in globular proteins using empirically determined probabilities of occurrence of amino acid residues in regions of secondary structure; e.g. (Y helix, p sheet, and ,!3 turns. This method predicts the occurrence of 88% of the helical regions and 95% of the p sheet regions in the 19 proteins evaluated by Chou and Fasman (27  The symbols (P,), (P,), and (P,) represent the calculated Chou-Fasman parameters for a helical structure, p sheet, and p turns, respectively. The numerical subscripts indicate the length of the oligopeptide being considered in calculating the above mentioned parameters. For a complete description of the calculation and appli- The amino acid sequence surrounding the iodoacetate-sensitive cysteinyl residue of creatine kinase is: Ala-Gly-F'ro-His-Phe-Met-(Asp,His,Glu)-(Gly,Leu)-Tyr-Val-Leu-Thr-Cys-Pro-Ser-Asn-Leu-Gly-Thr-Gly-Leu-Arg (28), where the order of residues in parentheses was not determined. A Chou-Fasman analysis of this polypeptide yields the results shown in Table  III. The degeneracy in the sequence of residues 7-9 and lo-11 did not alter the predicted structure. The most striking feature of the calculations is the prediction that the "active" cysteinyl residue should occur at the beginning of a /3 turn which separates two portions of /3 sheet structure. In such a position this sulfhydryl group could well be involved in conformational changes in the protein. Kuntz (29) has observed that few p turns ever occur internally in the tertiary structure of proteins, a situation stemming from the hydrophilic nature of the amino acid side chains found in p turns. Conversion of the relatively hydrophilic thiol group in native creatine kinase to the more hydrophobic mixed disulfide in the CH,S-blocked enzyme could perturb the p turn structure, and any spatially associated regions of the protein.
Similarly, modification of the reactive sulfhydryl with a hydrophilic sulfhydryl reagent, e.g. iodoacetic acid, iodoacetamide, or 5,5'-dithiobis (2-nitrobenzoic acid), would be expected FIG. 4. Double reciprocal plots used to determine the nature and extent of product inhibition in the native and CH,S-blocked enzymes at pH 9.00, 30". Assay conditions were the same as those described for Figs. 1  If one postulates that the active sulfhydryl group necessarily becomes buried'dming catalysis (i.e. less solvated as a result of the conformational change associated with substrate binding), then chemical modifications resulting in a more hydrophilic nature for the modified cysteinyl side chain could prevent this burying and obliterate catalysis. Conversely, if chemical modifications lead to a more hydrophobic nature for this group, this hypothetical conformational change could still occur. Evidence linking conformational changes to substrate binding has come from many sources, including susceptibility to tryptic cleavage (301, reactivity of the active site sulfhydryl (311, magnetic resonance studies (1,31,32), intrinsic fluorescence (33), dye-binding studies (211, and immunological studies (34). While the role for the active cysteinyl residue in mediating conformational changes suggested above is purely speculative, it is nonetheless consistent with all previous observations of structural changes and kinetic changes.
Acknowledgments-We are grateful to Dr. W. J. O'Sullivan, University of New South Wales, for helpful discussions on assay conditions, to Dr. W. W. Cleland, University of Wisconsin, and Dr. J. F. Kirsch, University of California, Berkeley, for their advice on computer analysis, to Dr. E. C. Cordon, University of California, San Francisco, for adapting the SEQUEN computer program for use with the IBM 360 system, and to Dr. D. J. Smith for preliminary kinetic experiments.
Note Added in Proof-Very recently, Price and Hunter (35) have reported that in the presence of creatine, MgADP, and nitrate the active sulthydryls of the two subunits of rabbit muscle creatine kinase react with thiol blocking reagents at different rates from each other, behavior consistent with a conformational change in the second subunit brought about by modification of a thiol group of the first subunit.