Creatine kinase: The characteristics of the enzyme regenerated from the thio-methylated creatine kinase reflect a differentiation in function between the two reactive thiols

Biochimie (1996) 78, 219-226 © Soci6t6 franqaise de biochimie et biologie mol6culaire / Elsevier, Paris

Creatine kinase: The characteristics of the enzyme regenerated from the thio-methylated creatine kinase reflect a differentiation in function between the two reactive thio|s LX Hou, .IX Zhou National l_z~boratotw of Biomacromolecules, Institute of Biophysics, Academia Sinica. Beijing 100101. China

(P.eceived 19 February 1996; accepted 13 June 1996) Summary - - The activity of S-thiomethyl-modified creatine kinase is due to regeneration of the free thiol (Hou and Vollmer,Biochim Biophys Acta (1994) 1205, 83-88). Characteristics of enzyme regenerated from the S-thiomethyl-modified creatine kinase are reported in the present study. The intrinsic fluorescence of the regenerated enzyme is similar to that of the native enzyme in the presence or absence of the dead-end complex. Regenerated CK (rCK) with full activity has only one reactive thiol. The rate constant of the rCK-reactive thiol reacting with DTNB is close to that of the slow phase of the reactive thiols of the native enzyme. If the IAM-modified rCK is treated with the same method as that for obtaining the rCK, the thiol-methylated reactive thiol of the rCK is reduced to a free SH and a regenerated enzyme, RCK, is produced with about ! 0% of the rCK activity. Therefore, the different roles of the two reactive thiols of creatine kinase may stem from the characteristics of the rCK, which suggests that only one of the two reactive thiols is related to the activity of the enzyme and the slower phase thiol (the first SH) in the modification reaction with DTNB is directly related to the enzymatic activity while the faster phase thiol (the second SH) assists the first SH. This compensatory mechanism is proposed in the present study to inte~Tpretthe dispute on the reactive SH role in the enzymatic catalysis. creatine kinase I reactive thiols I chemical modification I regeneration Introduction

Creatine kinase (EC.2.7.3.2), an important enzyme in cellular energy metaoolism, reversibly transfers a phosphoryl group from ATP to creatine. It is a dimer with eight SH groups [1, 2], among which two have long been known to be far more reactive than the others. Although the role of the reactive thiols in the enzymatic catalysis has been thoroughly studied, no consensus has been reached on whether the reactive thiols are essential for the activity of the enzyme [3, 4] leaving the controversy so far unresolved [5-7]. Tsou and colleagues [8, 10] and Hou and Vollmer [9] reported results opposite to other authors [ 11-13], indicating that the reactive thiols of creatine kinase are important to its activity. Different roles for the reactive thiols in the function of the enzyme have also been noticed. Tsou [14] pointed out that only one of the two thiol groups of creatine kinase is

Abbreviations: nCK, native creatine kinase; rCK, CK regenerated from MMTS-modified creatine kinase; RCK, CK regenerated from IAMmodified rCK; IAM, iodoacetamide; DTNB, 5,5'-dithiobis(2-nitrobenzoic acid); MMTS, methyl methanethiosulfonate; 4-PDS, 4,4'-dithiodipyridine; mnCKuurs, nCK modified by MMTS; mnCKIAM,nCK modified by IAM; mrCKIAM,rCK modified by IAM.

required when the data of the enzyme inhibited by IAM from Watts et a! [15] were treated with Tsou's graphical raethod. Degani and Degani [16-18] suggested on the basis of the asymmetrical behavior of the two reactive thiols that each reactive thiol makes a different ccmribution to the activity of the enzyme. Price and Hunter [ 19] carefully studied the modification of creatine kinase by IAA and DTNB and found that in the absence of the substrates the reaction is monophasic, showing that the reactive thiols of both subunits react at the same rate; and that in the presence of the transition-state analogue the reaction is biphasic, suggesting that these two reactive thiols react at different rates while the enzyme is at the transition state dudng catalysis. Wang et al [20] found that when the enzyme was modified by either IAM or DTNB, the inactivation reaction in the pres~nce of all the substrates is monophasic, but the reaction of the enzyme modified with DTNB, in the presence of the transition-state analogue, becomes biphasic suggesting that the enzyme has to be in the dimeric state to be active. Grossman et ai [21] also reported significant differences in the flexibi!ities of the active site domains of homodimeric creatine kinase. Hou and Vollmer [9] showed that the activity of the Sthiomethyl-modified creatine kinase is due to the regeneration of a free thiol. The results reported in the present study further address some characteristics of creatine kinase regenerated (rCK) from the S-thiomethyl-modified enzyme, showing that there may be differentiation in function between the two reactive thiols of the native enzyme. Only the

220 slow-phase thiol (the first SH) of the enzyme is linked with the enzymatic activity. The fast-phase thiol (the second SH) may mainly serve as a defender of the first SH. These regeneration phenomena of the modified enzymes lead to the compensatory mechanism proposed in this paper, which may apparently be used to interpret the dispute on the role of Cys282 for the enzymatic activity.

Materials and methods The preparation and assay of rabbit muscle creatine kinase were as described before [22] except that bovine serum albumin was omitted from the assay mixture. The reaction of ATP with creatine, which releases a proton, was followed by measuring the absorbance change of the pH indicator thymol blue at 597 nm. The reaction mixture, 2.25 mL, contained 4 mM ATE 24 mM creatine, 5 mM Mg 2+ acetate, 0.008% thymol blue and 25 nM enzyme in 5 mM glycine buffer (pH 9.0). The reaction mixture was carefully adjusted to pH 9.0 before use, and a calibration curve was constructed to correlate the amount of protons generated and the absorbance change at 597 nm under the given conditions. Creatine kinase has a broad optimum pH range (8.0--9.0)[4], and a change of 0.5 pH unit in this range has no appreciable effect on its activity. The enzyme concentration was determined by absorbance at 280 nm

with Al~l cm= 8.8 [23]. The regenerated enzyme concentration was calibrated by the Lowry method [24]. MMTS, IAM, and DTNB were from Sigma, ATP from Fluka, and creatine from E Merck. All other reagents were local products of analytical grade used without further purification. The dead-end complex (DEC) [21] contained creatine kinase, 20 mM creatine, 4 mM MgCI2, and 3 mM ADP in 0.05 M (pH 9.0) glycine buffer. Fluorescence measurements were carried out at 25°C with a Hitachi F-4500 Model fluorescence spectrophotometer. Absorbance measurements were made with a Shimadzu UV-250 spectrophotometer. Modification of the enzyme SH groups by MMTS was carried out at 4°C with 78 lxM creatine kinase and 2.4 mM MMTS in pH 9.0, 0.1 M glycine buffer for 3 h, at the end of which the enzyme was completely devoid of reactive thiols. The modified enzyme was then separated from,the reagent in excess by centrifugation through a Penefsky [25] column with Sephadex G-25. Modification of the regenerated enzyme (rCK) by IAM was carried out under similar conditions as described in detail in the legend to figure 5. The regenerated enzyme, rCK, was prepared on the rationale that the substrate is not required for reactivation of the MMTS-modified enzyme [9]: the MMTS-modified enzyme was diluted to 2.0 l.tM with a pH 9.0, 5 mM glycine buffer, and then incubated at 30°C for about 3 h until the assay yielded a straight line. The regenerated enzyme obtained by dilution as above has the greatest fraction of the original enzyme activity possible. The determination of the number and type of reactive thiols for both the native and regenerated enzymes was carried out by Tsou's graphical method [ 14]. The calculation of the apparent rate constant of the activity regeneration for enzymes modified with different reagents was carried out by Tsou's method [26].

IlOi

,a.,

i 60

Results

4O

/4

i,° 0

300

Spectra comparison: the S-thiomethyl-modified and regenerated enzymes relative to the native enzyme

\

I

I

I

i

I

t

350

40O I

I

i

!

45O

WAVEI.I~NG'I'H (NM)

Fig 1. Comparison of spectra from the MMTS-modified and regenerated enzymes with the native enzyme. For the fl~Jorescence spectra, samples contained matched concentrations of protein (0.85 ~tM) in 0.05 M glycine buffer (pH 9.0). The dead-end complex was added from a stock solution to give a final concentration of 20 mM creatine, 3 mM ADP and 4 mM MgCI2 in a final volume of 2.0 mL. Excitation was at 295 nm. Curve 1, the native enzyme; curve 2, the regenerated enzyme; curve 3, the native enzyme with the dead-end complex; cu~v¢ 4, the regenerated enzyme with the dead-end complex. For the absorption spectra (inset), samples contained the same concentrations of protein (6.5 l.tM) in 0.04 M Tris-HCl buffer (pH 8.05). Curve 1, the native enzyme; curve 2, the native enzyme titrated by MMTS with a final concentration of 162 l.tM.

The fluorescence spectrum of nCK differs from that of DEC as reported by Grossman et al [21 ]. The rCK fluorescence spectrum is compared to that of the native enzyme in figure 1. The maximum rCK fluorescence intensity occurs at the same position (330 nm) as that of the native enzyme regardless of whether DEC is present or not. With no DEC, the rCK fluorescence intensity is largely consistent with that of nCK although there is less quenching (6%) (curves 1 and 2 in figure 1). In the presence of the DEC, both spectra curves nearly overlap (curves 3 and 4 in figure 1). The inset in figure 1 shows that the absorbance spectrum of the MMTSmodified enzyme looks very similar to that of the native enzyme. These results suggest that introducing thioi-methyl groups into the enzyme molecule does not significantly affect the conformation of the enzyme. Therefore, the stable conformation after introducing--S-CH3 groups into the enzyme molecule looks to constitute the basis of the activity regeneration which is accompanied with the thiol regeneration [9 ].

221

Identification of the number of reactive thiols in the native enzyme and in the regenerated enzyme

0 04

The decrease in enzyme activity is linearly related to the extent of masking of the SH groups during the modification with DTNB for both the native and regenerated enzymes. as shown in figure 2, indicating that there is one reactive SH relative to the enzymatic activity per active unit for both the native and regenerated enzymes [14]. Figure 2 also shows that complete inactivation occurs upon modification of one SH group per regenerated enzyme molecule and of two SH groups per native enzyme molecule. The above results suggest that only one of the two reactive thiols is related to the enzymatic activity for the native enzyme. The regenerated SH, which was confirmed to be the only reactive SH group in the regenerated enzyme rCK [9], is the one linked to the activity of the regenerated enzyme.

Comparison of modification rates for the regenerated and the native enzymes

0.03

'2_

0.02

0.01 0

20

40

60

80

100

t (s) 0

|

t

I

40

,~0

120

I

160

200

t (s) Modification of creatine kinase by DTNB was carefully studied by Price and Hunter [19]. The modification reaction in the absence of substrates is monophasic, while in the

i.0

0.8

0

0

0.6

04

02

0

Fig 3. The modification reaction of nCK with DTNB in the presence of the transition-state analogue. The reaction mixture contained 1.3 gtM of the native enzyme, which has a specific activity close to that of the rCK in figure 4; 48 gtM DTNB, 30 mM creatine, 2 mM ADP and 10 mM nitrate. The modification was followed spectrophotometrically at 412 nm [32]. The temperature was 25°C. The inset shows a semilogarithmic plot of the pseudo-first-order reaction: Full circles, direct plot; open circles, fast phase of the reaction.

0.4

08

1.2

16

2o

Fig 2, Relationship between the fractional activity remaining and the extent of modification of the reactive SH groups for the native and regenerated enzymes. Both enzymes, which have similar specific activity, were modified in 0. l M glycine buffer (pH 9.0) with different amounts of DTNB at 25°C for 60 min. The extent of modification was measured spectrophotometrically at 412 nm [32]. 'a' represents fractional activity remaining; 'm', number of DTNB bound per tool of enzyme. Curve l, rCK; curve 2, riCK.

presence of the transition-state analogue it becomes biphasic. Wang et al [20] compared the kinetics of the modification and inactivation reactions for this enzyme under similar conditions, finding that the DTNB inactivation re~action in the presence of all substrates is monophasic, but the DTNB modification reaction in the presence of the transition-state analogue is biphasic. In order to identify the differences between the regenerated and native enzymes, the DTNB modification reactions for both enzymes were followed under similar conditions. Following the methodology of Price and Hunter [19] and Wang [20], a secondorder plot of the modification of nCK with DTNB in the presence of the transition-state analogue shows that the reaction is biphasic (fig 3). The fast rate constant, as determined by a semilogarithmic plot of the reaction during the pseudo-first-order conditions [inset fig 3), and the slow phase rate constant differ sig;ai~4cantly (table I). Modification of rCK by DTNB under ~:he same conditions as described in figure 3 was determined to be a monophasic reaction (inset fig 4) with a rate constant close to that of the slow phase of the nCK modification (table I). These results strongly suggest that the slow phase thiol (the firs~ SH) of

222

Table I. Comparison of the reaction rate constants for both regenerated and native enzymes with DTNB in the presence of the transition-state analogue (S-i M- I).

riCK

Present results Reference [ 19]

Fast phase

Slow phase

rCK

1840 2100

650 420

500 -

0 020 L

0016

residual activity. As shown in scheme 1, the newly regenerated en::yme, RCK, with an activity of about 10% of the rCK, was indeed obtained when the IAM-modified rCK was added to the assay mixture of the forward reaction (curve 1 in figure 5). During the course of the reaction, the addition of excess NMTS to the reaction mixture containing the IAM-modified rCK after completr, activity regeneration brought the reaction to a complete standstill within a few minutes (point B on curve 2 in figure 5). This shows that the free reactive SH is indeed liberated from the surface - S - S - C H 3 group of the IAM-modified rCK molecule during the course of the reaction and that the liberated free SH is again blocked by the excess MMTS, leading to inactivation of the regenerated RCK enzyme. The addition of excess IAM to the RCK regeneration reaction mixture does not

0.012

100 o

0.008

B

0.004

0

20

40

60

80

_..A

I(}(}

t(s) I

0

40

,I

,

80

I

I

120

16(}

//

2(HI

t isl

/

_/

Fig 4. The modification reaction of rCK with DTNB in the presence of the transition-stateanalogue. The reaction mixture was the same as in figure 3 except that the rCK concentration was 1.2 ltM. 'The inset shows a scmilogtwithmicplot of the pseudo-first-order reaction. 0

2(J

I0

3,~

t (,ran)

the two reactive thiols in the native enzyme molecule is closely related to the activity of the enzyme, and that the fast phase thiol (the second SH) could defend the first SH from injury during modification.

Reactivation of IAM-modified rCK Previous results [9] and data presented in figures 2 and 4 have shown that rCK has only one reactive thiol. Therefore, of the two surface thiols in the rCK molecule, one is in the free form and another should be in the - S - S - C H 3 form. If this supposition is correct, the -S-CH3 surface thiol in the rCK molecule could be expected to be reduced when the free reactive thiol of the rCK molecule is alkylated by the modifier IAM, and the reduced enzyme could have some

Fig 5. Reactivation of IAM-modified rCK. 40 IlM of rCK was modified by 400 IlM IAM in 0. I M glycine buffer ( pH 9.0) at 25°C for 2 h. Tile excess IAM was removed by a Penefsky column [25]. The reaction mixture contained 260 nM of the IAM-modified rCK, 4 mM ATP, 24 mM creatine, 5 mM Mg "+, 0.008% thymoi blue in 5 mM glycine buffer (pH 9.0). Curve l, the reactivation course of the IAM-modified rCK in the above reaction mixture, Curve 2, effect of modifiers MMTS and IAM on the reactivation: at position A, 10 IlL of 21.5 mM IAM was added to the reactivated reaction mixture; at position B, 1 IlL of 108 mM MMTS was added to the above reactive mixture. Curve 3, the IAM-modified rCK diluted to 520 nM with 5 mM glycine buffer (pH 9.0) was incubated at 25°C for 20 rain. Then all the substrates were added to make the composition of the reactive mixture identical with curve 1. Curve 4, the remaining activity of the native enzyme modified by IAM under the same conditions as curve 1.

223

Aclivitv regeneration of the enzyme rood(fled by 4-PDS and DTNB

08

06it

I 0

I , 20

,1 'tO

Time

t 60

1 80

Although previous results [91 provide evidence that the activity of creatine kinase modified by MMTS can be regenerated, it is unclear whether only the MMTS-modified enzyme can be regenerated. To test for activity regeneration of CK modified with different modifiers, the native enzyme was modified with both 4-PDS and DNTB. Figure 6 compares the regeneration activity of both the 4-PDS- and DNTB-modificd enzymes with the acuvity of the MMTSmodified enzyme. The apparent rate constants of the reactivation course for the three modified enzymes were 2.8, 0.6 and 9.1 (s-J × 103), respectively. The additional experiments showed that the activity regeneration of the enzyme modified with various modifiers is related to the -SH regeneration, and that the activity regeneration accompanying the - S H is not relative to the buffer species (not shown). Therefore, the activity regeneration of creatine kinase may occur in the different buffer solutions with either MMTS-modifled enzymes or from the enzyme modified by other modifiers in the different buffer solutions.

(rain)

Discussion Fig 6. Comparison of the reactivation courses of both 4-PDS- and DTNB-modified enzyme to the MMTS-modifiedenzyme reactivation course. Two samples of native enzyrnc, 25 laM in 0. I M glycine buffer (pH 9.0), were treated with 0.75 mM 4-PDS and DTNB, rcspectivcly, for 3 h at room temperature. The excess was then removed by passing a column of Sephadex G-25. The activation regeneration course was monitored under conditions comparable to the previous work 191.The apparent rate constants were calculated by Tsou's method 1261. O represents the MMTS-modified enzyme; ~), the 4-PDS-modified enzyme: A, the DTNB-modified enzyme.

seem to affect the regenerated activity of the RCK (point A ir: curve 2 in figure 5), suggesting the protection of RCK against further inactivation by IAM. This protection seems to be in agreement with the report that the MMTS-modified enzyme has protection against further inactivation by IAM |27]. It is of interest that dilution of the lAM-modified rCK with 5 mM (pH 9.0) glycine buffer for 20 rain prior to addition of all substrates brought about the reactivation of the IAM-modified rCK and that the activity of the regenerated enzyme RCK was comparable to curve 1 (curve 3 in figure 5). The regenerated enzyme RCK, therefore, can also be obtained by dilution. Since the reactiva6on of the IAMmodified rCK in the assay mixture is obviously different from that of its control (the native enzyme~ with two reactive thiols (curve 4 in figure 5), then RCK should also have only one free thioi.

Creatine kinase is a homodimer [1] which is supposed to have an asymmetrical arrangement [18]. The hypothesis ot the asymmetric subunits is based on the heterogeneity of the two reactive thiols of this enzyme [16, 17]. Although the heterogeneity of the reactive thiols has been confirmed by an number of studies [ 19-21 ], its significance is unclear so far. The characteristics of rCK provide some information for understanding the different roles of the two reactive thiols of this enzyme. The results in figures 2 and 4 indicme that rCK has only one reactive thiol, which was coafirmed in a previous work [9]. Comparison of the modification with excess DTNB of the reactive thiols in rCK and nCK shows that, in the presence of the transition-state analogue, the former is monophasic (fig 4), while ihe latter is biphasic (fig 3), which is in agreement with tlae results of Price and Hunter [19] and Wang et ai [201. Since the rCK activity is ,comparable to the nCK activity [9] and the rate constant of 'he rCK thiol modification is close to that of the nCK slow phase thiol (the first SH) modification (table I), it can be deduced that the first thiol of creatine kinase is closely related to the enzymatic activity and that the active unit of the enzyme is a dimer with one free reactive thiol. On the other hand, the role of the other reactive thiol (~he second SH) conjugated with the first thiol in the nCK molecule is also important. As shown in scheme l, the free thiol in RCK is derived from the rCK-reactive thiol blocked by the CH3S-group, which is equivalent to the second thiol of the native enzyme. The results in fi~,ure 5 show that the RCK-reactive thiol could be regenerated by adding the IAM-modified rCK to the assay mixture or by diluting the

224 '~

SH •

......"~.... .

S-S-CI 13

SH

gly-Na()! I

-.511

~--

S - S - C t 13

5raM, p119 ()

mnCK~r,a

aCK

rCK IAM

')

, , S-CH2-CONH

S._CI_I2.CONt_I2

2

SI-[.

'-S-CH2-CONH 2

mnCKt,~ Scheme

RCK

~,

- - - S ' C H 2 - C O N q ~'2

gly-Na()l ! 5raM, pl{90

S-S-Ct-[ 3

rnrCKtAh~

1. Relationship between the native and regenerated enzymes.

IAM-modified rCK with 5 mM glycine buffer (pH 9.0). Therefore, the important condition for producing the reactive thioi in RKC is alkylation of the rCK-reactive thioi (namely the first thiol of nCK) by IAM, ie during modification of the enzymatic first thiol by IAM, the second thiol blocked by the -S-CH3 group could be converted to a free thiol. It can thus be inferred that there may be some coordination between the two reactive thiols. Since the rate constant of modification of the enzyme second SH with DTNB is faster than that of the enzyme's first SH (table I), the role of the enzyme's second SH may be to defend the first SH from modification. Altheugh RCK with the free second SH has about 10% of the rCK activity (fig 5), the other role of the second SH, namely accepting the modifier group blocking active thiol and then allowing its regeneration, looks to be more important than the former role since the latter appears to correspond to the principle of preventing waste in the metabolic control. These results show that the creatine kinase subunits are different from each other in their function and provide further evidence of the cooperation, defined by Degani and Degani [16-18], between the two subunits of this enzyme. Comparison ~f the number of thiols in unmodified, modified and reactivated dimerie enzymes unfolded in 6 M guanidine-HCl [9] suggests that the modifier group - S CH3 attached to the reactive SH in the MMTS-modified enzyme may shift to an internal thiol. Zhou and Tsou [ 10] showed that the dicyano derivative of creatine kinase migrated to internal thiols. Discovery of the oxidized creatine

kinase [28, 29] (see also below), which has intact reactive thiols and full activity but less internal thiols than the native enzyme, also suggests that the modifier group(s) attached to the reactive thiols may shift to internal thiol(s). These results imply that the internal thiols can accept the modifier group(s) attached to the reactive thiol(s) and are important for keeping the reactive thiol(s) free, although the internal thiols are unreactive in the absence of added denaturants and unrelated to the activity of the enzyme [30]. It is known that the activity of the MMTS-modified enzyme is due to the regeneration of the reactive thiol [9]. The reason of the reactive SH regeneration has been investigated by additional experiments. The reactivation rate constant of the MMTS-modified enzyme in the reverse reaction system (pH 7.0) is similar to that in the forward reaction system (pH 9.0) during monitoring the two reactions by the pH stat and colorimetric methods (not shown), respectively, indicating that the SH regeneration is due to neither the OH- catalyzed hydrolysis of the disulfide bond between the methanethio group and the reactive thiol nor due to other causes, such as the reaction direction or the assay method. The other experiments show that the categories of the buffer and modifier are not related to the SH regeneration. Influence of the possible reductant has also been negated by pre-oxidizing the substrate system prior to the activity regeneration reaction with iodine solution. The SH regeneration, therefore, may originate in the enzyme itself. The above inference can apparently be supported by the regeneration process characteristics described below. The

225

full activity of the MMTS-modified enzyme can be regenerated in both forward and reverse reactions with similar rate. Both the regenerated enzymes, rCK and RCK, can be obtained only by simple dilution. The activity regeneration of ~he modified enzyme always accompanies regeneration of the reactive SH. Of the two reactive thiols modified by MMTS, only one is regenerated. The thiol regenerated from the MMTS-modified enzyme is the slow phase thiol (the first SH) relating to the enzymatic activity. After blocking the free SH in rCK by the alkylating agent IAM, another thiol (the second SH) blocked by the -S-CH3 group could be transformed into the free SH and the new regeneration enzyme RCK has 10% of the rCK activity. The activity of DTNBand 4-PDS-modified enzymes could be similarly regenerated but to a different extent. These phenomena suggest that in the creatine kinase molecule there may be an ability compensatory to the enzymatic activity decrease brought by the environmental change. An actual example for the compensatory mechanism is the oxidized creatine kinase [28, 29], which has the full activity, the intact reactive thiols and the decreased internal thiols. The reasonable explanation for the oxidized creatine kinase should be that, under influence of the compensatory mechanism, the modifier group(s) attached to the reactive thiol(s) may si',ift to an internal thiol(s), leading to the freedom of the modified reactive thiol(s), the decrease of the internal SH number, and the maintenance of the enzymatic activity. Although details of the compensatory mechanism have not been cleared so far, the objectivity of this mechanism appears trustworthy. Under influence of the compensatory mechanism, the creatine kinase molecule could make as fully as possible its reactive thiols free in order to obtain ti~e best activity, suggesting that the reactive ~hiol(s) is important for the enzymatic activity. The results of site-specific mutagenesis [61 at Cys278 for the chicken cardiac mitochondrial creatine kinase also show that Cys278 is important for the enzymatic activity. For the mutants of C278G and C278S to continue to have some activity, the reaction must include the additional factor KCI. In the absence of KCI, all the mutants have a lower activity in the reverse reaction. For the forward reaction, even in the presence of KCI, the activity of all the mutants is still much lower than the wild-type enzyme. Therefore, the results from site-specific mutagenesis [6] also suggest that Cys278 is closely related to the enzymatic activit)'. The small residual activities for the mutant enzymes [6, 31] could apparently be regarded as the result of adjustment of the compensatory mechanism, implying that Cys282 (rabbit muscle) or Cys278 (chicken cardiac mitochondria) are not closely related to the enzymatic activity in the case of the site-specific mutagenesis. Furthen,',)re, t,~e creatine kinase molecule has this me,l.anism, so that a measurably ,-mall activity would be expected to remain when the enzyme was mutated at the alternative amino acid residue of the active site.

AcknowUedgment We would like to thank Professor CL T,~oq for his supervision of this projecl.

References 1 Putney S, Herlihy W, Royal N, Pang H, Aposhian HV, Pickering L. Belagaje R, Biemann K, Page D, Kuby S, Schimmel P ( ! 984) Rabbit muscle creatine phosphokinase, cDNA clonh,g, primary structure, and detection of human homologous. J Biol Chem 259, 1431714320 2 Ordahi CP, Evans GL, Kunz TA, Perriard JC (1984) Complete cDNAderived amino acid sequence of chick muscle creatine kinase. J Biol Chem 259, 15224-15227 3 Kenyon GL, Reed GH (1983) Creatine kinase: structure-activity relationships. Adv Enzymol 54, 367--426 4 Watts DC (1973) Creatine kinase, hi: The Enzyme (Boyer PD, ed) 8, 383--455 5 Buechter DD, Medzihradszky KF, Burlingame AL, Kenyon GL (1992) The active site of creatine kinase. Affinity labeling of cysteine 282 with N- (2,3-epoxypropyl)-N-amidinoglycine.J Biol Chem 267. 2173-2178 6 Furter R, Furter-Graves EM, Wailimann "i (1993) Creatine kinase: The reactive cysteine is required for synergism but is nonessential for catalysis. Biochemistry 32, 7022-7029 7 Lin L, Perryman MB, Friedman D, Roberts R, Ma TS (1994) Determination of the catalytic site of creatine kinase by site- directed mutagenesis. Biochim Biophys Acut 1206, 97-104 8 Wu H, Yao QZ, Tsou CL (1989) Creatine kinase is modified by 2-chloromercuri-4-nitrophenoi at the active site thiois with complete inactivation. Biochim Biophys Acta 997, 78-82 9 Hou LX, Vollmer S (1994) The activity of S-thiomethyl modified creatine kinase is due to the regeneration of free thiol at the active site. Biochim Biophys Acta 1205, 83-88 10 Zhou HM, Tsou CL (1987) The presence of reactive SH groups in the enzymatically active dicyano derivalive of creatine kinase. Biochim BiophysActa 91 I, 136--143 I I Quiocho FE. Thompson JW (1973) Substrate binding to ~n active creatine kinase with a thiol-bound mercurinitropi~enoi chromophoric probe. Proc Nat/Acad Sci USA 70, 2858-2,862 12 Smith DJ, Maggio ET, Kenyon GL (1975) Simple aikanethiol groups for temporary blocking of sulfhydryl groups of enzymes. Biochemistry !4, 766-77 I ! 3 Terrossian DE, Kassab R (1976) Preparation and properties of S-cyano derivatives of creatine kinase. Fur J Biochem 70, 623-628 14 Tsou CL (1962) Relation between modification of functional groups of proteins and their biological activity. Sci Sin I!, 1535-1558 15 Watts DC, Rabin BR, Crook EM (1961) The reaction of iodoacetate and iodoacetamide with proteins as determined with a silver/silver iodide electrode. Biochim Biophys Acta 48, 380-388 16 Degani Y, Degani C (1979) Subunit-selective chemical modification of creatine kinase. Evidence for asymmetrical association of the subunits. Biochemistl T 18, 5917-5923 17 Degani C, Degani Y ~1980) Further evidence for nonsymmetric subunit association and intersubunit cooperativity in creatine kinase. J Biol Chem 255, 8221-8228 18 Degani Y, Degani C (1980) Enzymes with asymmetrically arranged subunits. Trends Biochem Sci 5, 337- 341 19 Price NC. Hunter MG (1976) Non-identical behavior of the subunits of rabbit muscle creatino kinase. Biochim Biophys Acta 445, 364-376 20 Wang ZX, Preiss B, Tsou CL (1988) Kinetics of inactivation of creatine kinase during modificationof its thiol groups. Biochemistry 27, 5095-5100 21 Grossman Sh. France RM, Mattheis JR (1992) Heterogeneous flexL bilities of the active site domains of homodirneric creatine kinase. Biochim Biophys Acta 1159, 29-36

226 22 Yao QZ, Hou LX, Zhou HM, Tsou CL (1982) Conformational changes of creatine kinase during guanidine denaturation. Sci Sin 25B, I 186- i 193 23 Noda L, Kuby SA, Lardy H (1954) ATP-creatine transphosphory!ase. Methods Enzynzol 2, 605--610 24 Lowry OH, Roseborough NJ, Farr AL, Randall RJ ( 1951 ) Protein measurement with the folin phenol reagent. J Biol Chem 193, 265-275 25 Penefsky HS (1977) Reversible binding of Pi by beef heart mitochondral adnosine triphosphastase. J Biol Chem 252, 2891-2899 26 Tsou CL (1988) Kinetics of substrate reaction during irreversible modification of enzyme activity.Adv En~moi Relat Areas Mol Bioi 61,381-436 27 Smith DJ, Kenyon GL (1974) Non-essentiality of the active sulfhydryi group of rabbit muscle creatine kinase. J Biol Chenz 249, 3317-3318

28 Hot. LX, "['sou CL (1985) Comparison of the conf~)rmational and activity changes of creatine kinase during Li Dodecyisulfate denaturation. Acta Biochim Biophys Sinica 15, 393-398 29 Wang XC, Wang F, Zou XM, Zhou HM (1994) Oxidized form of creatine kinase. Scicm'e in Ct, imz (B) 37, 964-974 30 Yao QZ, Tian M. Tsou CL (1984) Comparison of the ratcs inactivation and comformational changes of creatine kinase during urea denaturation. Biochemistry" 23, 2740-2744 31 Chen LH, Babbit PC, Kenyon GL (1990) Expression and site-specific mutagenesis of rabbit muscle creatine kinase in E coli: Assessing the importance of cysteine 282. (1990) FASEB J 4, A 2119 32 EIIman TE (1959) Tissue sulfhydryl groups. Arch Biochent Biophys 82, 70-77