International Federation of Clinical Chemistry (IFCC): Scientific Division, Committee on Enzymes. IFCC methods for the measurement of catalytic concentration of enzymes. Part 7. IFCC method for creatine kinase (ATP: creatine (N-phosphotransferase, EC 2.7.3.2). IFCC Recommendation

Committee Members: R. Rej (US) (Chairman, M. tterder (DK) M. Mathieu (FR), L. M. Shaw (US) until December 1984,.]. H. Stremme (NOR) until December 1984, K. Lorentz (FRG) from 1985.06, and W. Gerhardt (SW) from June 1985. Optimization experiments were performed in collaboration with the Study Group on Creatine Kinase, Subcomnittee on Enzymes of the Committee on Standards, American Association for Clinical Chemistry. Members: R. C. Elser (Chairman), E.J. Sampson, A. R. Henderson, L. G. Morin and D. A. Nealon. Reprints are available from Dr Mogens Herder, Odense University Hospital, Department of Clinical Chemistry, DK-5000 Odense C, Denmark; or from Dr Robert Rej, NY State Department of Health, Wadsworth Center for Labs & Research, Empire State Plaza, Albany, New York 12201-0509, USA. dimeric forms of creatine kinase in serum. The method is also suited for the determination of creatine kinase variants which may be present in serum.


Introduction
The principles applied in the selection of the conditions of measurement are those stated in previous publications by this expert panel [1]. Human serum and tissue extracts have been used as the sources of enzymes. The final concentrations of substrates, auxiliary and indicator enzymes have been selected on the basis of experimental evidence and data in the literature dealing with the three In human serum, CK may be present in various forms: CK-MM may occur in several forms of which MM1, MM2 and MM3 are the best documented [18]. CK-MB. CK-BB.
Macro CK Type 2: probably a serum form of mitochondrial CK, although conclusive evidence is still lacking 16,21,22].
There are no reports of a dimeric mitochondrial CK in serum.
In order to ensure full catalytic activity, the creatine kinase molecule in serum must be reactivated by a reducing sulphhydryl compound.  pH (30 C) 6.60 +_ 0.05 Volume fraction of serum 0"0435 (1:23) Different kinetic properties have been found for each of the human creatine kinase isoenzymes, including those of human origin [23][24][25]. Kinetic data have also been compiled for mitochondrial CK [26]. Therefore, measurement of total creatine kinase in serum requires using an assay system whose reaction conditions are a compromise.
The proposed method for the measurement of the catalytic concentration of creatine kinase in serum is based on the principles proposed by Oliver [27] and later modified by Rosalki [28] and Szasz et al. in a series of publications [23,[29][30][31][32][33][34]. Recommended or standard methods published by professional societies in the United Kingdom [35], FR Germany [36], The Netherlands [37], France [38], Scandinavia [39,40] and Switzerland [41] are also based on these publications and have also been considered.
The primary reaction (1), catalysed by creatine kinase, and the coupled reactions (2) and (3) (1) Note: The concentrations apply to the complete reaction mixture. The catalytic concentrations of hexokinase and D-glucose-6-phosphate dehydrogenase must be determined as described in Appendix B.

IFCC conditions for measurement
The reaction conditions have been chosen on the basis of univariate and multivariate experiments with sera from healthy individuals, sera from patients with acute myocardial infarction, sera from patients with skeletal muscle diseases, and with isolated human isoenzymes [30, 38--40, 43, 44]. These IFCC conditions are optimized reaction conditions, which are defined 1] as those conditions that are most favourable for both the kinetic reactions and the technical aspects of the measurement, i.e. these conditions do not necessarily provide maximum activity.
The reaction is initiated by creatine phosphate. Following an initial lag phase, substrate conversion proceeds linearly with time and amount of enzyme until deceleration occurs due to build-up ofinhibiting NADPH and decreased NADP + concentration (see table 1 The equilibrium in (1) favours the formation of creatine and ATP at pH values around 6-7, due to the higher energy of creatine phosphate as compared to that of ATP and the lower Km values for ADP and creatine phosphate than for ATP and creatine [42]. This primary reaction is coupled through the auxiliary reaction (2), catalysed by hexokinase, to the NADPH forming indicator reaction (3) catalysed by glucose-6-phosphate dehydrogenase.

Instrumentation and equipment
A thermostatted spectrometer suitable for accurate measurement at 339 nm should be used.
The specifications for the equipment (for example, spectral band width, light path, accuracy of thermostats) should meet those of previous recommendations [1].
Instruments must be capable of monitoring the linear portion of the rate of conversion curve and should display both the initial absorbance of the reaction mixture and absorbance versus time during the measurement interval.
The temperature of the reaction mixture in the cuvette must be controlled at 30"0 _+ 0"05 C.
All volumetric glassware used for the preparation of reagents and for pipetting must meet US National Institute of Standards and Technology (NIST) Class A specifications, American Chemical Society Microchemical specifications (tolerance is 0"997 to 1"003) or other national equivalents, pH meters must be calibrated at 30 1"0 C by use of a standardized reference buffer (for example, NIST) with a pH value within unit of the reaction measurement pH.]" be prepared in calibrated flasks with water meeting the following standards [45]: Electrical resistivity: ->2"0 x 10 4 ohm/m at 25 C. pH: 6"0-7"0.
Dissolve the following components in approximately 950 ml of deionized or distilled water, which meets the above-mentioned requirements: imidazole 8"70 g; magnesium acetate tetrahydrate 2"74 g; and ethylenediaminetetraacetic acid, disodium dihydrate 968 mg. Adjust the pH to 7"3 at room temperature (20-26 C) with acetic acid, mol/1. Add water to a final volume of exactly 1.
The absorbance of the buffer solution (I) at 339 nm should be less than 0"050 [39].
Working solution IV should have an absorbance at 339 nm of less than 0"050 [39].
The catalytic concentration stability is apparently pH dependent, being maximal around pH 6"5 to 7"0 and minimal around pH 8 During the preincubation period, the solution in the cuvettes must attain a temperature of 30"0 0"05 C before initiating the reaction.   constant over a period of at least 60 s following the lag phase for sera with catalytic concentrations of creatine kinase up to 40 tkat/1 (2400 U/l), provided that the spectrometer is capable of making accurate absorbance readings up to 2"000 A. If the change of absorbance is greater than 0"01/s the serum sample .must be diluted with solution V. However, this will lead to inaccuracy due to nonlinearity of the dilution curve (see Appendix A for details).

Corrections for blank reactions
The rate of the overall reaction (A) is corrected for any sample blank reaction (B) as follows: a corrected aA aB.
The subscripts, A and B, indicate the composition of the reaction mixtures referred to in table 2. The corrected value of a is used in the following calculations. It equals the true rate of conversion catalysed by creatine kinase.
The reagent blank rate of conversion, C, does not enter into the calculations, but is used to ascertain the quality of the reagents. Its value should be less than a change in absorbance of 0"0007 per 60 s. If it is higher, the purity of the reagents must be reassessed (see Appendix B). For measurement at 339 nm, with sufficiently sensitive spectrometers, the limit of detectability using 180 s monitoring after the end of the lag phase is 24 +_ 8 nkat/1 (1.4 + 0.5 U/1 [38]. The analytical sensitivity has been established to be an increase in absorbance at 339 nm of 0"001 per 60 s, but will depend on the instrument used. recommendation, reference intervals of 0"30 to 1"50 tkat 1-1 118 to 90 U 1and 0"60 to 3"60 tkat 1-(36 to 216 U 1were found for women (aged 10-45 years) and men (aged 10-60 years), respectively [38].

Analytical variability
Data about inaccuracy are not available because no international reference material has yet been established.

Buffer type
The selection of an appropriate buffer for the reaction has been considered by Szasz et al. [8] and Morin [15].
The majority of CK investigative work since 1975 has been accomplished using imidazole as the buffer.
However, there has been some controversy surrounding the choice of this buffer. Imidazole was difficult to obtain in pure form, and decomposed during storage. This problem has now been solved by the manufacturers and much more stable and pure preparations are available having low absorbance at the wavelength of the measurement [2].
2,2-Bis (hydroxymethyl)-2,2',2"-nitrilotriethanol (Bis-Tris), due to its chelating ability and buffering capacity at the pH of the assay, has been proposed as an alternate buffer system to imidazole and EDTA [15]. However, studies by Szasz et al. 13], by Nealon et al. [20], by the SCE [24] and by the CK Study Group of the AACC [25] have shown that there is no distinct superiority of this buffer over imidazole with EDTA. Studies relating recovery of catalytic concentration to the ionic strength of the buffer show an inverse relationship between catalytic concentration and buffer concentration (see figure A2 and [13]). Buffer anions inhibit creatine kinase. A buffer of 100 mmol/1 imidazole containing 2 mmol/1 EDTA is therefore a compromise concentration between sufficient buffer capacity and minimal inhibition [8]. 3. Chelating substance, EDTA Inclusion of EDTA in the assay has several advantages: it prevents autoxidation of N-acetyl-L-cysteine and the formation of inhibitors from such oxidation [2,13], the stability of the CK reagent at 4 C is increased from less than 24 h to 5 days, and rates of conversion are increased by reversal of the apparent inhibition of CK by endogenous Ca 2+ [18] and Fe a+ [24].    Figure A3. EDTA influence on creatine kinase catalytic concentration determined in serum, based on data presented in [7].
individuals and from patients with acute myocardial infarction are from 1" to 1"2 times greater measured by an assay at 30 C containing EDTA, than in an assay, also at 30 C, without added EDTA [7,24] (see figure A3, [7] The volume fraction of the sample is critical. Changes of the volume fraction do not provide proportional changes in the rate of conversion [1]. Therefore, the catalytic concentrations of creatine kinase obtained with this IFCC method are defined specifically at a volume fraction of sample of 0"0435 (see figure A4, [2]).

Reactivation of catalytic activity by N-acetyl-L-cysteine
CK in serum is rapidly inactivated. Incubation with a thiol having a high redox potential reactivates CK. This thiol must fulfil several requirements, including rapid and complete reactivation of CK catalytic activity, no precipitation of proteins in the specimen, sufficient solubility in solution and lack of obnoxious odour. Employing freeze drying for incorporation into lyophilized reagent kits is not an absolute requirement for an IFCC method but it is convenient for routine methods []5, 5].
Glutathione was rejected as a reactivator because reactivation is incomplete and requires a longer time. in addition, glutathione reductase in serum causes decreased rates of conversion.
N-acetyl-L-cysteine has been found to be satisfactory. At a volume fraction of sample of 0"0435 and with 20 mmol/1 of N-acetyl-L-cysteine in the CK reagent, reactivation of CK catalytic activity in serum samples stored for one week at 4 C is 99% complete [1,7,8]. In addition, N-acetyl-L-cysteine does not cause microprecipitation of sample proteins [1] and is easily soluble at pHs between 6"5 and 6"7 at the required concentration.
Thiols in solution undergo irreversible oxidation. This has two effects: loss of available sulphhydryl groups [12,16], and formation of potent CK inhibitors [13,24]. Both processes are accelerated by certain polyvalent cations, and, therefore, are retarded by the presence of chelators. Inclusion of EDTA stabilizes N-acetyl-L-cysteine in the reagent for 24 h at room temperature and 5 days at 4 C [2, 13,24] (see figure A5 and [2]). N-acetyl-L-cysteine must be of the highest purity to avoid preformed inhibitors [7].  Figure A7. Dependence of apparent creatine kinase catalytic concentration on ADP concentration [8].

Catalytic mechanisms
The true substrates of the enzyme should be considered as the magnesium-nucleotide complex, which for this method is Mg-ADP, and free creatine phosphate. The enzyme appears to possess two binding sites on each subunit, one for the nucleotide and one for the guanidino substrate. The site for nucleotide binding presumably involves an arginine group and a lysyl group. It is speculated that a histidyl residue is involved in the guanidino site. Figure A6 has been adopted from Watts [27] and shows the rate equation leading to the formation of the transition complex, and the dissociation of the transition complex into products. It is thought that the process is a rapid equilibrium, random bimolecular mechanism in which either the nucleotide or the guanidino substrate may bind first [27].
Because the nucleotide binds to the active site as the magnesium ion complex, magnesium ion concentration is important in establishing the equilibrium concentration ofcomplexed nucleotide. Wevers et al. [28] estimated that the effective equilibrium concentrations of Mg-ADP and creatine phosphate were 1.'61 mmol/1 and 25"8 mmol/1, respectively, at the nominal reaction conditions. Small anions can occupy the site normally filled by the gamma-phosphoryl group of magnesium-ATP to form a dead-end complex, which resembles the transition state of the normal enzyme substrate complex. This explains the inhibitory effects of small anions such as nitrate, sulphate, and chloride [27,29].

Substrate concentrations
Although the true substrates are Mg-ADP and free creatine phosphate, for the sake of practicality total concentrations are given for ADP and creatine phosphate in the following paragraphs. This convenience is warranted because the influence ofimidazole, EDTA, AMP, and other reagent components on the concentration of the respective substrates is unknown.
Multivariate studies show that simultaneous increases of Mg 2+ to 15 mmol/1 and ADP to 3"5 mmol/1 increase CK reaction rates by 6-8% [25,30,31] (see Section 8). This change increases the ADP/AMP molar ratio from the current 0"4 to 0"7, and will increase residual adenylate kinase activity [9]. Magnesium is necessary for both creatine kinase and hexokinase activity. The interrelationship between the chelating agents and the substrates, particularly ADP and magnesium and calcium and other divalent cations endogenously present in the sample, is not known at the molecular level. However, the effects of each of these on the rate of conversion have been described [1,2,8,13,18]. Therefore Figure A8. Dependence of apparent creatine kinase catalytic concentration on creatine phosphate concentration [8]. and 40 mmol/1.30 mmol/1 has been selected as optimal in both univariate [1, 7 11] (see figure AS, [8]) and multivariate studies [25,30,31] (see also Section 8). 6 Adenylate kinase, also known as myokinase, is a remarkably stable enzyme and is present in the same tissues as creatine kinase. Conditions favouring the increased activity of CK also favour the increased activity of adenylate kinase. The pH optima for both enzymes are within two pH units of each other. The overall measured reaction rate includes catalytic activity attributable to adenylate kinase. Consequently, corrections must be made. This is accomplished by including inhibitors of adenylate kinase in the reaction mixture, and, in addition, compensation for a sample blank rate due to the residual adenylate kinase catalytic activity.
Adenylate kinase may be inhibited by various substances, including AMP [1,9,10], P1,ps-Di(adenosine-5'-pentaphosphate [9,10,33,36] and sodium fluoride [10,34,37]. Maximal inhibition of adenylate kinase by AMP was found when the AMP/ADP ratio was 10:1 [1,9,10]. However, AMP also inhibits creatine kinase [1,9]. CK inhibition is proportional to the concentration of AMP [31 although the degree of inhibition varies among sera. AMP inhibits adenylate kinase by a mechanism which is thought to involve the formation of a 'dead end' complex, either AMP-enzyme-AMP or ADP-enzyme-AMP [35]. Inhibition of adenylate kinase increases as AMP concentration increases and approximately 7% adenylate kinase catalytic activity remains at a concentration of 10 retool/1 of AMP. Because of the competitive nature of the inhibition, the adenylate kinase inhibition also depends on the ADP concentration. Inhibition ofCK by AMP also depends on the AMP/ADP ratio [9,10]. At a ratio of 5, CK is inhibited by approximately 10%. At a ratio of 2"5, CK is inhibited between 5 and 8%. Because both CK and adenylate kinase are sensitive to the AMP/ADP ratio, its value must be selected so that CK is minimally inhibited and adenylate kinase blanks are acceptably low. At an AMP/ADP ratio of 2"5, sample blank rates are decreased significantly, but the highest values of residual adenylate kinase amounted to 20% of the upper limit of the reference range for creatine kinase catalytic activity 10].
The dilemma of reducing residual adenylate kinase without concomitantly inhibiting CK is reduced by adding a second inhibitor.
Diadenosine polyphosphates have been examined as inhibitors of adenylate kinase. The compound with greatest inhibitory capability is P1,p5-Di(adenosine-5'-)pentaphosphate [36], a potent inhibitor of erythrocyte, muscle, and purified liver adenylate kinase. The pentaphosphate does not inhibit platelet adenylate kinase nearly as well. A combination of AMP and P1,ps-Di(adenosine-5'-)pentaphosphate has been recommended by Szasz et al. 10] at concentrations of 5 mmol/1 and 10 tmol/1, respectively. This results in inhibition of adenylate kinase from erythrocytes and muscle by 97%, from liver by 95% [33] and from platelets by 90% [10].
Fluoride is a non-competitive, adenylate kinase inhibitor with a Ki at 25 C of 2"5 mmol/1 [10]. Neither singly, nor in combination with AMP or with p1,p5_ Di(adenosine-5'-)pentaphosphate 10] does fluoride offer any advantage over the selected AMP and P1,ps-Di(adenosine-5'-)pentaphosphate inhibitor combination [9,10]. Fluoride has two disadvantages: it causes turbidity in the reagent due to precipitation of magnesium fluoride, and the lag-time before achieving maximum inhibition by fluoride is 6 rain at 30 C [10].
The inhibition potentials of the various components are summarized in table A3.
Adenylate kinase catalytic activity not completely in- 5 hibited by AMP and p1,p-Di(adenosine-5-)pentaphosphate is measured separately by the sample blank Table A3. Fractional residual adenylate kinase catalytic activity in the reagent mixture for measurement of creatine kinase with different inhibitors included. Modifiedfrom references (1,7,9,10,33 " Liver preparation obtained from Blue Sepharose column [33]. + + Liver extract clarified by centrifugation [33]. In samples from healthy persons this blank rate is zero or very low. In samples from patients with liver and heart diseases, residual adenylate kinase catalytic activity is significant [38]. In addition, in a reference method any possible side-reaction that may contribute to nonspecificity of the method should be compensated for. Therefore, measurement of the sample blank reaction is part of the IFCC method for creatine kinase. It has been established [39]  Hexokinase from baker's yeast is used as the auxiliary enzyme coupling the primary reaction to the indicator reaction. Both the form in which the enzyme is obtained and its final concentration (activity) in the reaction mixture influence the apparent CK activity that is measured. Because of the influence of small anions such as sulphate, nitrate, and chloride on the activity of creatine kinase in the formation of the 'dead end' transition complex (see below), ammonium sulphate suspensions of the auxiliary and indicator enzymes should be avoided. The catalytic concentration of hexokinase influences the duration of the lag phase of the reaction. At hexokinase catalytic concentrations lower than 50 btkat/1 (measured at 30 C in the CK reagent), the lag phase is greater than 120 s (see figure A9, [8]).
The indicator enzyme is glucose-6-phosphate dehydrogenase. Two types of this enzyme are commonly used for analytical purposes: one from yeast, the other from Leuconostoc mesenteroides. The coenzyme specificities of the enzymes from these sources are different. The yeast enzyme is specific for NADP + while the enzyme for L.
A-combination ofhexokinase at 50 btkat/1 (3000 U/I) and glucose-6-phosphate dehydrogenase at 33 btkat/1 (2000 U/l) has been chosen. This is not rate limiting in the measurement of the creatine kinase catalytic concentration below about 40 btkat/1 and it ensures a lag phase of less than 120 s (see figure A10, [8]). Under these conditions, the reagent blank rate of conversion will correspond to less than 0.04 btkat/1, provided the reagent specifications described in Appendix B are fulfilled.
120 to 180 s after the start will be linear with a catalytic concentration ofCK up to at least 40 tkat/1 (2400 U/l) [7,8]. 8  Rates of conversion are independent of the glucose concentration in the range of 10 mmol/l to 100 mmol/1.20 mmol/1 provides full catalytic activity with no sidereactions [1,7,8] (see figure A11, [8] The investment in time and manpower required to validate a new method not only for accuracy but also for transferability, precision, and expected reference values, is substantial. The decision not to replace well-accepted conditions as described in this IFCC method [1][2][3][4][5][6][7][8] by slightly changed conditions [30,31] was therefore made by the Expert Panel on Enzymes. This decision was based on the above criteria and was carefully considered. The 5% to 8% activity increase resulting from the change in total ADP and total magnesium concentrations was felt not to have significant importance to.justify a change. Additionally, no improvement in precision was demonstrated.  B1).
Let the increase in absorbance per second at 339 nm be a The total reaction volume, V, is 2"3 x 10 -a 1. The 'sample' volume, v, is 2 x 10 -a 1.

Conditions for measurement
The conditions for measurement of hexokinase catalytic concentration are similar to those of the IFCC creatine kinase method, except that creatine phosphate is omitted. Hexokinase is present at a rate limiting catalytic concentration. The reaction is initiated with adenosine-5'triphoshpate (ATP) (see table B2).
Let the increase in absorbance at 339 nm be a (s-l).
The total reaction volume, V, is 2"3 x 10 -3 1. The 'sample' volume, v, is 2 x 10 -3 1.  The conditions for measurement of 6-phospho-Dgluconate dehydrogenase catalytic activity concentration are similar to thoe of the IFCC creatine kinase method, except that creatine phosphate is omitted. The reaction is initiated with 6-phospho-D-gluconate.
Conditions for detection of 6-phospho-D-gluconate" NADP + 2oxidoreductase (decarboxylating) The conditions for detection of 6-phospho-D-gluconate: NADP + 2-oxidoreductase catalytic concentration are similar to those of the IFCC creatine kinase method, except that creatine phosphate is omitted. The reaction is initiated with 6-phospho-D-gluconate. The concentration of 6-phospho-D-gluconate corresponds to that which theoretically might arise from conversion of all NADP + in the assay mixture. This concentration does not necessarily support a linear rate of conversion, but will detect any significant interference from contaminating 6-phospho-D-gluconate dehydrogenase (see table B3).

Evaluation
If the increase of absorbance at 339 nm exceeds 0"001 per 60 s, this may be due to the presence of contaminating 6phospho-D-gluconate dehydrogenase in either the hexokinase or in the glucose-6-phosphate dehydrogenase stock solution. Consequently, the enzyme preparation in question must be replaced.