TY - JOUR
AU - Azzazy, Hassan, ME
AB - Abstract Background: The AACC assembled a committee to identify and validate a standard creatine kinase MB isoenzyme (CK-MB) material to improve the comparability of CK-MB mass assays. Methods: Three protocols were used. In protocol I, various CK-MB materials prepared in different matrices were screened as candidate standards. In protocol II, participating manufacturers calibrated their systems with concentrates of human heart CK-MB and then tested 20 patient samples to evaluate calibration bias. In protocol III, participating manufacturers calibrated their immunoassay systems using recombinant CK-MB2 (rCK-MB2) diluted into their respective sample diluents and measured 50 samples. Results: Candidate materials showed high recovery in stripped human serum, but bias improved only from 59% to 38%. These data led to the use of human heart CK-MB diluted in each manufacturer’s sample diluent. This strategy reduced bias from 31% to 15%. Because human heart CK-MB is difficult to provide, a lyophilized source of CK-MB2 was identified. rCK-MB2 was shown by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, reversed-phase HPLC, intrinsic protein fluorescence, circular dichroism, agarose gel electrophoresis, immunoreactivity studies, high and low temperature stability, and reconstituted stability to be equivalent to human heart CK-MB. Calibration of immunoassay systems with rCK-MB2 added into each respective manufacturer’s sample diluent showed a 13% between-manufacturer bias. Conclusion: Lyophilized rCK-MB2 was determined suitable for use as a reference material for CK-MB mass assays. Increased creatine kinase (CK;3 EC 2.7.3.2) isoenzyme MB (CK-MB) is considered the benchmark for diagnosis of acute myocardial infarction (AMI) (1)(2). A variety of methods are used to measure CK-MB, including electrophoresis, column chromatography, immunoinhibition/immunoprecipitation, and immunoassays (1)(3)(4). Additionally, assay methods such as immunoinhibition and electrophoresis rely on CK-MB activity (U/L). In recent years, many new CK-MB immunoassays have become available. These methods are termed “mass” assays because they measure CK-MB in terms of protein concentration, i.e., μg/L, using two-site immunoassay technology based on anti-M subunit (anti-CK-M), anti-B subunit (anti-CK-B), or anti-MB-specific antibodies. CK-MB mass assays can be automated, are highly sensitive (detection limit <1 μg/L) and specific, and have a rapid turnaround time (as low as 7 min), giving them characteristics that surpass any other technology or measurement of CK-MB (4). The characteristics of different CK-MB mass assays utilized for the correlation studies are displayed in Table 1. Table 1. Characteristics of various CK-MB mass immunoassays. Analyzer . Capture antibody . Detection antibody . Label . Detection scheme . Abbott AxSYM Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Abbott IMx Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Bayer Immuno 1 Anti-CK-MB Anti-CK-MB Alkaline phosphatase Absorbance Beckman Access Anti-CK-BB Anti-CK-MB Alkaline phosphatase Chemiluminescence Dade Behring Opus Plus Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Chiron ACS:1802 Anti-CK-BB Anti-CK-MB Acridinium ester Chemiluminescence Dade Behring Stratus II Anti-CK-MB Anti-CK-BB Alkaline phosphatase Fluorescence Dade Behring aca Plus Anti-CK-B Anti-CK-MB β-Galactosidase Fluorescence Hybritech Tandem-E3 Anti-CK-B Anti-CK-M Alkaline phosphatase Absorbance Ortho Vitros:ECI Anti-CK-BB Anti-CK-MB Horseradish peroxidase Luminescence Roche Elecsys Anti-CK-MB Anti-CK-MB Ruthenium complex Chemiluminescence TOSOH AIA 600 Anti-CK-BB Anti-CK-MB Alkaline phosphatase Fluorescence Analyzer . Capture antibody . Detection antibody . Label . Detection scheme . Abbott AxSYM Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Abbott IMx Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Bayer Immuno 1 Anti-CK-MB Anti-CK-MB Alkaline phosphatase Absorbance Beckman Access Anti-CK-BB Anti-CK-MB Alkaline phosphatase Chemiluminescence Dade Behring Opus Plus Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Chiron ACS:1802 Anti-CK-BB Anti-CK-MB Acridinium ester Chemiluminescence Dade Behring Stratus II Anti-CK-MB Anti-CK-BB Alkaline phosphatase Fluorescence Dade Behring aca Plus Anti-CK-B Anti-CK-MB β-Galactosidase Fluorescence Hybritech Tandem-E3 Anti-CK-B Anti-CK-M Alkaline phosphatase Absorbance Ortho Vitros:ECI Anti-CK-BB Anti-CK-MB Horseradish peroxidase Luminescence Roche Elecsys Anti-CK-MB Anti-CK-MB Ruthenium complex Chemiluminescence TOSOH AIA 600 Anti-CK-BB Anti-CK-MB Alkaline phosphatase Fluorescence 1 Goat polyclonal antibodies. All other antibodies are murine monoclonal antibodies. 2 Except for Chiron ACS:180, all anti-CK-MB antibodies are Conan antibodies. 3 Tandem-E is no longer manufactured. Open in new tab Table 1. Characteristics of various CK-MB mass immunoassays. Analyzer . Capture antibody . Detection antibody . Label . Detection scheme . Abbott AxSYM Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Abbott IMx Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Bayer Immuno 1 Anti-CK-MB Anti-CK-MB Alkaline phosphatase Absorbance Beckman Access Anti-CK-BB Anti-CK-MB Alkaline phosphatase Chemiluminescence Dade Behring Opus Plus Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Chiron ACS:1802 Anti-CK-BB Anti-CK-MB Acridinium ester Chemiluminescence Dade Behring Stratus II Anti-CK-MB Anti-CK-BB Alkaline phosphatase Fluorescence Dade Behring aca Plus Anti-CK-B Anti-CK-MB β-Galactosidase Fluorescence Hybritech Tandem-E3 Anti-CK-B Anti-CK-M Alkaline phosphatase Absorbance Ortho Vitros:ECI Anti-CK-BB Anti-CK-MB Horseradish peroxidase Luminescence Roche Elecsys Anti-CK-MB Anti-CK-MB Ruthenium complex Chemiluminescence TOSOH AIA 600 Anti-CK-BB Anti-CK-MB Alkaline phosphatase Fluorescence Analyzer . Capture antibody . Detection antibody . Label . Detection scheme . Abbott AxSYM Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Abbott IMx Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Bayer Immuno 1 Anti-CK-MB Anti-CK-MB Alkaline phosphatase Absorbance Beckman Access Anti-CK-BB Anti-CK-MB Alkaline phosphatase Chemiluminescence Dade Behring Opus Plus Anti-CK-MB Anti-CK-MM1 Alkaline phosphatase Fluorescence Chiron ACS:1802 Anti-CK-BB Anti-CK-MB Acridinium ester Chemiluminescence Dade Behring Stratus II Anti-CK-MB Anti-CK-BB Alkaline phosphatase Fluorescence Dade Behring aca Plus Anti-CK-B Anti-CK-MB β-Galactosidase Fluorescence Hybritech Tandem-E3 Anti-CK-B Anti-CK-M Alkaline phosphatase Absorbance Ortho Vitros:ECI Anti-CK-BB Anti-CK-MB Horseradish peroxidase Luminescence Roche Elecsys Anti-CK-MB Anti-CK-MB Ruthenium complex Chemiluminescence TOSOH AIA 600 Anti-CK-BB Anti-CK-MB Alkaline phosphatase Fluorescence 1 Goat polyclonal antibodies. All other antibodies are murine monoclonal antibodies. 2 Except for Chiron ACS:180, all anti-CK-MB antibodies are Conan antibodies. 3 Tandem-E is no longer manufactured. Open in new tab Split patient sample correlation studies have shown that commercially available CK-MB immunoassays correlate well (r >0.98) but are biased from one another, which is reflected by differences in the slopes of comparative analyses. When Henderson et al. (5) monitored the analytical performance of total CK (∼230 participants) and CK-MB (∼160 participants), they reported a wide dispersion of results between different analyzers as well as among identical analyzers. A major reason for such between-method bias is the lack of a CK-MB mass standard. To address this issue, the CK-MB Mass Assay Standardization Subcommittee was formed in January 1992 by the AACC to “identify or develop a standard material using human CK-MB that can be used to reduce laboratory-to-laboratory variation in the accuracy of the CK-MB mass assays”. The goals of this committee were to identify and validate a source of CK-MB standard, to establish purification and protein determination protocols, and to investigate the effect that various matrices and CK-MB materials (liquid or lyophilized) have on standardizing CK-MB mass assays. With cooperation among manufacturers, it is hoped that this reference material will be used to standardize assay methods such that between-assay bias will be reduced or eliminated. Materials and Methods expression and purification of rCK-MB The cDNA sequences encoding CK-M and CK-B were cloned from a human cardiac cDNA library (6)(7). The cDNA for each subunit was subcloned into different Escherichia coli expression vectors, and the plasmids were cotransformed into E. coli. A strain was shown to express active CK-MB isoenzyme in vivo. Recombinant human CK-MB2 (rCK-MB2) was purified according to a modification of the method of Chen et al. (8). Purified protein was concentrated by diafiltration, and its concentration was determined by its absorbance at 280 nm. For use as a standard material, rCK-MB2 is formulated with stabilizers, including a carrier protein, dispensed into vials, and lyophilized. sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis Native heart CK-MB and rCK-MB2 were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) to ascertain protein purity on 12% Tris-HCl ready gels (Bio-Rad). Gels were run on MiniProtein II systems (Bio-Rad) and stained with Coomassie blue. agarose gel electrophoresis Agarose gel electrophoresis was used to verify CK isoenzyme purity via separation by overall protein charge. Native heart CK-MB and rCK-MB2 were applied to Universal Gel/8 agarose gels and electrophoresed according to the manufacturer’s protocol (Chiron Diagnostics). Gels were stained for CK activity using the Creatine Phosphokinase Isoenzymes kit (Sigma Diagnostics). reversed-phase hplc The purity and hydrophobic nature of native heart CK-MB and rCK-MB2 were analyzed by reversed-phase HPLC using triplicate 10-μg sample injections on a 2.1 × 30 mm Brownlee RP-300 (C5) column on a Hewlett Packard 1090 HPLC system with photodiode array detection (Hewlett Packard). Proteins were eluted at a flow rate of 0.25 mL/min, using a linear gradient of 0.8 g/L trifluoroacetic acid in acetonitrile and detected at 214 nm. Hewlett Packard Chemstation software was used to integrate the data. intrinsic protein fluorescence The structural environments surrounding tryptophan residues in the proteins were assessed by measuring the intrinsic protein fluorescence of native heart CK-MB and rCK-MB2 at ambient temperature with a McPherson FL-750 spectrofluorescence detector (McPherson). Fluorescence excitation spectra were obtained by scanning the sample, 10 μmol/L CK-MB, from 200 to 300 nm at an emission maxima of 320 nm. The respective fluorescence emission spectra were measured by scanning from 300 to 400 nm at an excitation maxima of 290 nm. The relative fluorescence intensity data for both native heart CK-MB and rCK-MB2 were imported into a graphics program and displayed in overlapping plots. circular dichroism The protein-folding conformation of native heart CK-MB and rCK-MB2 was analyzed by measuring their circular dichroism at 25 °C on a JASCO J720 spectropolarimeter (Jasco). A far-ultraviolet circular dichroism spectrum was obtained by scanning 0.5 g/L samples from 178 to 260 nm with a 0.005-cm pathlength. immunoreactivity The dose–response of native heart CK-MB and rCK-MB2 added to serum was assessed using an IMx immunoanalyzer. A 5 g/L working stock solution of CK-MB was diluted into heat-inactivated newborn calf serum (Genzyme Diagnostics) to 0, 3, 10, 30, 100, and 300 μg/L CK-MB. These samples were assayed in duplicate in separate IMx analytical runs using the same stored calibration curve. Their fluorescence reaction rates were plotted against standard dilution factors. The native heart CK-MB and rCK-MB2 response curves in the immunoassay were overlayed for comparison. accelerated stability studies of rCK-MB2 To assess the storage shelf-life of lyophilized rCK-MB2 at 2–8 °C, accelerated stability studies were conducted by stressing vials of rCK-MB2 in isothermal incubators set at 45, 37, and 30 °C at staggered time intervals. For each temperature, stressed vials were reconstituted at the same time with Genzyme Diluent (Genzyme Diagnostics), diluted into assay range with heat-inactivated newborn calf serum (Genzyme Diagnostics), and assayed in duplicate in the same analytical run, using the IMx CK-MB mass assay. The percentages of recovery were determined relative to that of the value for non-stressed CK-MB stored at 2–8 °C. The rate constant for the loss of CK-MB mass at each incubation temperature was determined by linear regression analysis of the time plot data using a graphical program. An Arrhenius plot was constructed using the logarithm of the rate constants against the inverse of incubation temperature in degrees Kelvin. From the Arrhenius plot, the rate constant for the loss of rCK-MB mass at 8 °C can be extrapolated by linear regression analysis and used to calculate the time for rCK-MB mass to decrease 10% at 8 °C (9). reconstitution stability of rCK-MB2 The reconstituted or opened vial stability for rCK-MB2 was performed by reconstituting two vials with Genzyme CK-MB Diluent (Genzyme Diagnostics), storing them at 2–8 °C, and removing aliquots from these vials at different time intervals for assay in duplicate, using an IMx immunoanalyzer. The reconstituted product stability of rCK-MB2 at 2–8 °C was measured by the decrease in CK-MB mass to 90% of its day 0 value. shelf-life study Lyophilized rCK-MB2 was stored at 4 and −20 °C. At each selected time point over a period of ∼500 days, rCK-MB2 was reconstituted in aca Plus sample diluent, and its mass was measured using the aca Plus immunoanalyzer. A 10% loss of CK-MB mass in vials stored at 4 °C relative to that stored at −20 °C was used as the stability criterion. Results The standardization process followed in this study was divided into four phases. It is of note that all CK-MB immunoassays were compared to IMx because of its availability in all of the participating laboratories and not because it was considered the reference method. phase 1: screening of ck-mb candidates CK-MB candidates. Four different forms of CK-MB were considered for the development of a CK-MB standard: human heart CK-MB, hybrid CK-MB, rCK-MB2 (tissue isoform), and rCK-MB1 (serum isoform, missing the C-terminal lysine on the M subunit). The hybrid CK-MB, prepared by reassociation of the respective M and B subunits of denatured CK-MM and CK-BB, was obtained from Aalto Scientific. Different recombinant CK-MB forms were provided by Dr. M. Benjamin Perryman (University of Colorado Health Science Center, Denver, CO) (10). In addition, during the later stage of evaluation (phase 3), lyophilized rCK-MB2 was provided by Genzyme Diagnostics. Purification of CK-MB. Human heart CK-MB, rCK-MB2, and rCK-MB1 were purified using anti-CK-MB monoclonal antibody immunoaffinity chromatography according to established procedures (11). The final preparations of CK-MB were stored at −20 °C in 20 mmol/L Tris-HCl, pH 7.2 containing 500 mL/L glycerol. The purity of all CK-MB preparations was assessed by SDS-PAGE and agarose gel electrophoresis. Determination of CK-MB concentration. From the protein sequence of CK-MB, one can determine that the B subunit of CK contains 4 tryptophan, 9 tyrosine, and 2 cystine amino acid residues, whereas the M subunit of CK contains 4 tryptophan, 10 tyrosine, and 2 cystine amino acid residues (6)(10). Based on this amino acid content, the molar absorptivity of CK-MB was estimated as 0.82 L/cm-mol (absorbance of 1 g/L solution of CK-MB at 280 nm using 1-cm pathlength) (12). The concentration of the purified CK-MB was determined spectrophotometrically on a Model 8450A Spectrophotometer (Hewlett Packard). Lyophilized and liquid forms of human heart native CK-MB, rCK-MB1, rCK-MB2, and hybrid CK-MB were evaluated in different matrices, which yielded equivalent recovery results in immunoassays. Lyophilized rCK-MB2 was selected for use in all subsequent studies because of its long-term stability and ease of shipping and handling. phase 2: determination of optimum calibrator matrix Matrix selection. Three different potential sample matrices were evaluated: an unprocessed CK-MB-free human serum pool, stripped serum, and a synthetic matrix (50 mmol/L phosphate buffer containing 50 g/L bovine serum albumin and 0.15 mol/L NaCl, pH 7.4). The stripped human serum was prepared by mixing a human serum pool with a blend of anion- plus cation-exchange resin to extract residual CK. Some other serum components may have been removed unintentionally during this process. The stripped human serum was donated by Dade Behring (Glasgow, DE). Manufacturers’ sample diluents were subsequently evaluated as the study progressed. Formulation. To optimize the formulation of a standard for use in all CK-MB mass assays, a screening experiment was designed in which the four sources of CK-MB and the three matrices were evaluated. In this study, formulations containing 0, 10, or 100 μg/L of human heart CK-MB, hybrid CK-MB, rCK-MB1, or rCK-MB2 were prepared in three matrices. These samples were shipped to nine participating manufacturers of CK-MB mass assays. At each manufacturer’s site, each vial was hydrated with 2 mL of deionized water, and the CK-MB masses of the reconstituted candidate standards were determined on their respective immunoassay systems. Manufacturers’ correlation study: protocol I. Purified, lyophilized human heart CK-MB candidate standards (0, 10, and 100 μg/L) prepared in stripped human serum were assayed on each immunoassay analyzer and later used to correct original calibration curves via regression analysis. Additionally, 69 patient samples were analyzed for CK-MB mass. For these studies, the following seven immunoassay systems were used: ACS:180 (Chiron Diagnostics), Stratus II (Dade-Behring Diagnostics; analysis performed at the Glasgow, DE site), Tandem-E (Hybritech), OPUS Plus (Dade Behring; analysis performed at the Glasgow, DE site), aca Plus (Dade Behring; analysis performed at the Wilmington, DE site), AIA-600 (TOSOH), and IMx (Abbott Diagnostics). Assays were conducted according to respective manufacturers’ specifications. Manufacturers’ correlation study: protocol II. In an alternative approach for the standardization of CK-MB mass assays, the manufacturers of these assays were provided 20 patient sample pools (not used in the study described above) and purified human heart CK-MB from which each participating manufacturer prepared CK-MB candidate standards at 0, 10, 50, 100, and 200 μg/L in their sample diluent. The standards and the samples were analyzed in duplicate in their respective immunoassay systems. The slope bias of these methods against IMx was determined by linear regression analysis. For the different calibrator matrices, CK-MB recovery (data not shown) varied over a wide range as follows: stripped human serum (60–100%), human serum (23–60%), and synthetic medium (29–88%). Although stripped human serum produced the highest CK-MB recovery, the utilization of this matrix for reconstitution of CK-MB reference material only slightly reduced the slope range from ± 59% to ± 38% (Fig. 1). The slight reduction in between-manufacturer bias achieved with candidate standards made with stripped human serum is probably attributable to the variable effect of this matrix in different immunoassays. After dilution of human heart CK-MB reference material using the manufacturers, sample diluents, the range of slopes for the patient correlation plots was reduced from ± 31% to ± 15% (Fig. 2). Figure 1. Open in new tabDownload slide Patient correlation studies using protocol I. The correlation of patient CK-MB mass values (n = 69) measured on different immunoassay systems against patient CK-MB mass values measured on the IMx immunoanalyzer is shown before (left panels) and after (right panels) standardization with human heart CK-MB in stripped human serum. Slope values before and after standardization were as follows: ACS:180 (before, 1.92; after, 0.99), Stratus II (before, 0.96; after, 0.62), Tandem-E (before, 1.19; after, 1.23), OPUS Plus (before, 0.74; after, 0.48), aca Plus (before, 1.16; after, 1.06), and AIA-600 (before, 1.20; after, 0.88). Sy|x values before standardization were between 5.63 and 12.89 μg/L; after standardization, Sy|x values were between 6.27 and 16.83 μg/L. ♦, ACS:180; □, Stratus II; ▵, Tandem-E; ▴, OPUS Plus; ▪, aca Plus; ○, AIA-600. Figure 1. Open in new tabDownload slide Patient correlation studies using protocol I. The correlation of patient CK-MB mass values (n = 69) measured on different immunoassay systems against patient CK-MB mass values measured on the IMx immunoanalyzer is shown before (left panels) and after (right panels) standardization with human heart CK-MB in stripped human serum. Slope values before and after standardization were as follows: ACS:180 (before, 1.92; after, 0.99), Stratus II (before, 0.96; after, 0.62), Tandem-E (before, 1.19; after, 1.23), OPUS Plus (before, 0.74; after, 0.48), aca Plus (before, 1.16; after, 1.06), and AIA-600 (before, 1.20; after, 0.88). Sy|x values before standardization were between 5.63 and 12.89 μg/L; after standardization, Sy|x values were between 6.27 and 16.83 μg/L. ♦, ACS:180; □, Stratus II; ▵, Tandem-E; ▴, OPUS Plus; ▪, aca Plus; ○, AIA-600. Figure 2. Open in new tabDownload slide Patient correlation studies using protocol II. The correlation of patient CK-MB mass values (n = 20) measured on different immunoassay systems against patient CK-MB mass values measured on the IMx immunoanalyzer is shown before (top panel) and after (bottom panel) standardization with human heart CK-MB in each manufacturer’s sample diluent. Slope values before and after standardization were as follows: aca Plus (before, 0.79; after, 0.99), AIA-600 (before, 1.04; after, 1.02), Stratus II (before, 0.93; after, 1.13), and ACS:180 (before, 1.41; after, 0.82). Sy|x values before standardization were between 4.63 and 9.95 μg/L; after standardization, Sy|x values were between 1.43 and 5.24 μg/L. ♦, ACS:180; □, Stratus II; ▪, aca Plus; ○, AIA-600. Figure 2. Open in new tabDownload slide Patient correlation studies using protocol II. The correlation of patient CK-MB mass values (n = 20) measured on different immunoassay systems against patient CK-MB mass values measured on the IMx immunoanalyzer is shown before (top panel) and after (bottom panel) standardization with human heart CK-MB in each manufacturer’s sample diluent. Slope values before and after standardization were as follows: aca Plus (before, 0.79; after, 0.99), AIA-600 (before, 1.04; after, 1.02), Stratus II (before, 0.93; after, 1.13), and ACS:180 (before, 1.41; after, 0.82). Sy|x values before standardization were between 4.63 and 9.95 μg/L; after standardization, Sy|x values were between 1.43 and 5.24 μg/L. ♦, ACS:180; □, Stratus II; ▪, aca Plus; ○, AIA-600. Because all CK-MB isoforms were recovered similarly in different matrices and native human CK-MB preparations are difficult to obtain and prone to batch-to-batch variability, rCK-MB2 is proposed as a reference material. phase 3. preparation and properties of rCK-MB2 After phases 1 and 2 were completed, lyophilized rCK-MB2 was evaluated as a CK-MB reference standard. To validate its similarity to native human heart CK-MB, this material was analyzed in structural, immunological, and stability studies and compared to native CK-MB obtained from either BioProcessing, Inc., or provided by Dr. Jack Ladenson (Washington University, St. Louis, MO) (11). rCK-MB2 was expressed and purified from E. coli. The purified rCK-MB2 has an apparent molecular mass similar to native heart CK-MB, as shown by SDS-PAGE analysis (Fig. 3). The purity and structural similarity of rCK-MB2 to native heart CK-MB was further demonstrated by agarose gel electrophoresis (Fig. 4) and reversed-phase HPLC (Fig. 5) analyses. Figure 3. Open in new tabDownload slide SDS-PAGE analysis of rCK-MB2 and native heart CK-MB. Proteins were separated on 12% Tris-HCl ready gels (Bio-Rad) using a MiniProtein II system (Bio-Rad) and stained with Coomassie blue. The standard ladder (MW Stds) consists of 15 proteins ranging in molecular mass from 10 to 120 kDa. Figure 3. Open in new tabDownload slide SDS-PAGE analysis of rCK-MB2 and native heart CK-MB. Proteins were separated on 12% Tris-HCl ready gels (Bio-Rad) using a MiniProtein II system (Bio-Rad) and stained with Coomassie blue. The standard ladder (MW Stds) consists of 15 proteins ranging in molecular mass from 10 to 120 kDa. Figure 4. Open in new tabDownload slide Agarose gel electrophoresis of rCK-MB2 and native heart CK-MB. Analysis of the isoenzyme purity of rCK-MB2 and native heart CK-MB was performed using Universal Gel/8 agarose gels in a Universal Electrophoresis Cell (Chiron Diagnostics). The gel was stained to detect CK activity using a Creatine Phosphokinase Isoenzymes kit (Sigma Diagnostics). Figure 4. Open in new tabDownload slide Agarose gel electrophoresis of rCK-MB2 and native heart CK-MB. Analysis of the isoenzyme purity of rCK-MB2 and native heart CK-MB was performed using Universal Gel/8 agarose gels in a Universal Electrophoresis Cell (Chiron Diagnostics). The gel was stained to detect CK activity using a Creatine Phosphokinase Isoenzymes kit (Sigma Diagnostics). Figure 5. Open in new tabDownload slide Reversed-phase HPLC of rCK-MB2 and native heart CK-MB. CK-MB samples were injected onto a 2.1 × 30 mm Brownlee RP-300 column on a Hewlett Packard 1090 HPLC, eluted using a linear gradient of 0.8 g/L trifluoroacetic acid in acetonitrile, and detected at 214 nm. mAU, milliabsorbance units. Figure 5. Open in new tabDownload slide Reversed-phase HPLC of rCK-MB2 and native heart CK-MB. CK-MB samples were injected onto a 2.1 × 30 mm Brownlee RP-300 column on a Hewlett Packard 1090 HPLC, eluted using a linear gradient of 0.8 g/L trifluoroacetic acid in acetonitrile, and detected at 214 nm. mAU, milliabsorbance units. The intrinsic protein fluorescence spectra for native heart CK-MB and rCK-MB2 also are similar (Fig. 6). Circular dichroism studies of the native and recombinant CK-MB proteins revealed no distinguishable secondary structure (data not shown). Figure 6. Open in new tabDownload slide Intrinsic protein fluorescence of rCK-MB2 and native heart CK-MB. The intrinsic protein fluorescence was measured at ambient temperature using a McPherson FL-750 spectrofluorescence detector (McPherson). An excitation maximum of 290 nm and an emission maximum of 320 nm were used to generate the spectra. Fluor., fluorescence. Figure 6. Open in new tabDownload slide Intrinsic protein fluorescence of rCK-MB2 and native heart CK-MB. The intrinsic protein fluorescence was measured at ambient temperature using a McPherson FL-750 spectrofluorescence detector (McPherson). An excitation maximum of 290 nm and an emission maximum of 320 nm were used to generate the spectra. Fluor., fluorescence. The immunoreactivity responses of native heart CK-MB and rCK-MB2 were examined using the IMx CK-MB mass assay, which showed that both proteins have similar immunoreactivity (Fig. 7). Figure 7. Open in new tabDownload slide Immunoreactivity of rCK-MB2 and native heart CK-MB. Samples were assayed in duplicate in separate IMx analytical runs using the same stored calibration curve. The immunoreactivity curves for rCK-MB and native heart CK-MB, as measured by their fluorescence reaction rates, were overlayed for comparison purposes. The Sy|x for fluorescence rate vs inverse CK-MB dilution was 124.9 units (arbitrary) for rCK-MB and 97.9 units for native heart CK-MB. Figure 7. Open in new tabDownload slide Immunoreactivity of rCK-MB2 and native heart CK-MB. Samples were assayed in duplicate in separate IMx analytical runs using the same stored calibration curve. The immunoreactivity curves for rCK-MB and native heart CK-MB, as measured by their fluorescence reaction rates, were overlayed for comparison purposes. The Sy|x for fluorescence rate vs inverse CK-MB dilution was 124.9 units (arbitrary) for rCK-MB and 97.9 units for native heart CK-MB. Different studies were performed to evaluate the stability of rCK-MB2. In accelerated stability studies (Fig. 8), the percentages of recovery were determined relative to the value for CK-MB stored at 2–8 °C. The rate constant for the loss of rCK-MB mass at 8 °C was extrapolated by linear regression analysis from the Arrhenius plot and used to calculate the time for rCK-MB mass to decrease 10% at 8 °C. Studies of lyophilized rCK-MB2 showed that the refrigerated product shelf-life was 19 years (Fig. 9). The reconstituted product stability of rCK-MB2 at 2–8 °C was at least 21 days as measured by the decrease in CK-MB mass to 90% of its day 0 value (Fig. 10). The shelf-life of rCK-MB2 at 4 °C, as determined relative to its mass at −20 °C, was 555 days (data not shown). Figure 8. Open in new tabDownload slide Accelerated stability studies of rCK-MB2. Vials of lyophilized rCK-MB2 were stressed at 45, 37, and 30 °C, reconstituted with Genzyme Diluent, and assayed in duplicate in the same analytical run using the IMx CK-MB mass assay. The percentages of recovery were determined relative to that of the value for CK-MB stored at 2–8 °C. The Sy|x of the percentage of rCK-MB2 recovered vs incubation time was 2.0% at 45 °C, 6.6% at 37 °C, and 14.1% at 30 °C. Figure 8. Open in new tabDownload slide Accelerated stability studies of rCK-MB2. Vials of lyophilized rCK-MB2 were stressed at 45, 37, and 30 °C, reconstituted with Genzyme Diluent, and assayed in duplicate in the same analytical run using the IMx CK-MB mass assay. The percentages of recovery were determined relative to that of the value for CK-MB stored at 2–8 °C. The Sy|x of the percentage of rCK-MB2 recovered vs incubation time was 2.0% at 45 °C, 6.6% at 37 °C, and 14.1% at 30 °C. Figure 9. Open in new tabDownload slide Shelf-life stability of rCK-MB2. The rate constants for the loss of rCK-MB2 mass at different incubation temperatures were used to construct an Arrhenius plot of the logarithm of the rate constants (k) against the inverse of incubation temperature in degrees Kelvin [103/T (°K−1)]. The rate constant for the loss of rCK-MB2 mass at 8 °C was used to estimate the shelf-life stability of rCK-MB2 at 2–8 °C. The Sy|x of log k vs inverse temperature, in degrees Kelvin, was 0.13. Figure 9. Open in new tabDownload slide Shelf-life stability of rCK-MB2. The rate constants for the loss of rCK-MB2 mass at different incubation temperatures were used to construct an Arrhenius plot of the logarithm of the rate constants (k) against the inverse of incubation temperature in degrees Kelvin [103/T (°K−1)]. The rate constant for the loss of rCK-MB2 mass at 8 °C was used to estimate the shelf-life stability of rCK-MB2 at 2–8 °C. The Sy|x of log k vs inverse temperature, in degrees Kelvin, was 0.13. Figure 10. Open in new tabDownload slide Reconstitution stability for rCK-MB2. Aliquots of reconstituted rCK-MB2 and stored at 2–8 °C were removed at different time intervals for assay in duplicate using the IMx CK-MB mass assay. The criterion for determining reconstitution stability of rCK-MB2 at 2–8 °C was the decrease in CK-MB mass to 90% of its day 0 value. The Sy|x of the percentage of rCK-MB2 recovered against time at 5 °C was 3.5. Figure 10. Open in new tabDownload slide Reconstitution stability for rCK-MB2. Aliquots of reconstituted rCK-MB2 and stored at 2–8 °C were removed at different time intervals for assay in duplicate using the IMx CK-MB mass assay. The criterion for determining reconstitution stability of rCK-MB2 at 2–8 °C was the decrease in CK-MB mass to 90% of its day 0 value. The Sy|x of the percentage of rCK-MB2 recovered against time at 5 °C was 3.5. phase 4: final rCK-MB2 evaluation Each of the participating manufacturers was provided lyophilized rCK-MB2 and its diluent and 50 patient sample pools. Manufacturers reconstituted the rCK-MB2, prepared rCK-MB2 candidate standards at 0, 10, 50, 100, and 200 μg/L in their respective sample diluents, and determined the CK-MB mass values of the rCK-MB2 candidate standards and 50 samples. Data were analyzed for between-manufacturer bias by assessing patient sample correlation plots. A final correlation study was performed to evaluate the lyophilized rCK-MB2 on different CK-MB mass assays. As shown in Fig. 11 , before using rCK-MB2 to establish the calibration curves for different CK-MB immunoassays, the range of slopes was ± 40%. This slope range was reduced to ± 13% after rCK-MB2 was used to establish the calibration curves for the immunoassays. Figure 11. Open in new tabDownload slide Patient correlation studies using protocol III. The correlation of patient CK-MB mass values (n = 50) measured on different immunoassay systems against patient CK-MB mass values measured on the IMx immunoanalyzer is shown before (left panels) and after (right panels) standardization with rCK-MB2 in respective manufacturers’ sample diluents. Slope values before and after standardization were as follows: ACS:180 (before, 1.573; after, 1.018), aca Plus (before, 0.936; after, 1.192), OPUS Plus (before, 0.894; after, 0.974), AIA-600 (before, 1.15; after, 1.047), AxSYM (before, 1.016; after, 1.00), Access (before, 0.98; after, 1.168), and Stratus II (before, 0.755; after, 0.923). Sy|x values before standardization were between 3.39 and 7.90 μg/L; after standardization, Sy|x values were between 3.46 and 5.31 μg/L.♦, ACS:180; ▪, aca Plus; −, Access; •, AxSym; □, Stratus II; ▴, OPUS Plus; ○, AIA-600. Figure 11. Open in new tabDownload slide Patient correlation studies using protocol III. The correlation of patient CK-MB mass values (n = 50) measured on different immunoassay systems against patient CK-MB mass values measured on the IMx immunoanalyzer is shown before (left panels) and after (right panels) standardization with rCK-MB2 in respective manufacturers’ sample diluents. Slope values before and after standardization were as follows: ACS:180 (before, 1.573; after, 1.018), aca Plus (before, 0.936; after, 1.192), OPUS Plus (before, 0.894; after, 0.974), AIA-600 (before, 1.15; after, 1.047), AxSYM (before, 1.016; after, 1.00), Access (before, 0.98; after, 1.168), and Stratus II (before, 0.755; after, 0.923). Sy|x values before standardization were between 3.39 and 7.90 μg/L; after standardization, Sy|x values were between 3.46 and 5.31 μg/L.♦, ACS:180; ▪, aca Plus; −, Access; •, AxSym; □, Stratus II; ▴, OPUS Plus; ○, AIA-600. Discussion This study showed that a lyophilized rCK-MB material is suitable for use as a reference material for standardization of CK-MB mass assays. This is important for reducing method-related bias of CK-MB mass assays. Initially, the catalytic activity of CK in serum or plasma was measured to evaluate the diagnosis of AMI and to rule out cardiac necrosis. Recently, Gella et al. (13) described the preparation of lyophilized CK-MB purified from human heart for use as a reference material for CK-MB catalytic concentration measurements. Upon reconstitution, the catalytic concentration of the reference CK-MB material was reported to be ∼67 U/L when measured at 30 °C by the recommended method of the International Federation of Clinical Chemistry (13). CK-MB mass assays were developed to improve both the analytical and clinical sensitivity and specificity for measuring CK-MB compared with the enzymatic assay. These mass assays utilize monoclonal anti-CK-MB antibody in conjunction with anti-M or anti-B antibodies to accurately measure small changes in CK-MB concentrations during the early hours following AMI (Table 1). Despite the fact that CK-MB expression is not limited to cardiac tissue and that its diagnostic accuracy is low (∼50%) at 3 h after AMI, CK-MB is still regarded as the benchmark for diagnosis of AMI. In the Chest Pain Evaluation Center environment, Gibler et al. (14) examined the utility of assaying CK-MB mass in a strategy that included sampling at presentation and subsequently after 3, 6, and 9 h in >1000 low-risk nondiagnostic electrocardiogram patients. Although these patients were at low myocardial infarction risk, this study documented a sensitivity of 100% and specificity of 98.3% for myocardial infarction diagnosis in this population of chest pain patients (14). The field of biochemical markers of cardiac injury is in a dynamic state, in which assays for novel markers and current markers such as cardiac troponin I and cardiac troponin T continue to be developed to enhance the diagnosis of AMI (15). However, progress in the development of several different biochemical assays for new AMI markers has led to several analytical and interpretative issues, particularly those pertaining to comparisons with different CK-MB immunoassays. Therefore, it was deemed essential by the AACC to establish rigorous criteria for standardizing CK-MB mass assays. To accomplish this task, the AACC established a subcommittee to identify and develop a standard CK-MB material to reduce method-related variations in the analytic accuracy of CK-MB mass assays. To demonstrate the similarity in composition among CK-MB mass assays, the characteristics of 12 commercially available CK-MB immunoassays are displayed in Table 1. All of these assays use an anti-CK-MB monoclonal antibody as either the capture or detection antibody with the exception of the Hybritech Tandem-E assay, which utilizes anti-CK-B (capture) and anti-CK-M (detection) monoclonal antibodies. Ten of the remaining 11 assays use the “Conan” CK-MB antibody (11). The Chiron ACS:180 CK-MB Mark I mass assay uses antibodies developed by the manufacturer. In contrast, the methods listed in Table 1 use various labels and detection schemes. A detailed description of these assays has been reported previously (1)(4). The CK-MB standardization process was conducted in four phases: (a) identification of a CK-MB candidate reference material, (b) determination of an optimum assay matrix, (c) validation of the biochemical characteristics of the identified reference material, and (d) testing the selected material on different CK-MB mass assays. Although four different forms of CK-MB were examined, the results indicate that rCK-MB2 is the most suitable as a reference material because it can be easily prepared in large, homogeneous quantities compared with the other CK-MB forms. Lyophilized rCK-MB2 was favored over that in a liquid format because of its stability and ease of handling. Furthermore, validation experiments in this study have shown that rCK-MB2 has physicochemical characteristics similar to, if not equivalent to, CK-MB purified from human heart (11). Examination of three different assay matrices revealed that stripped human serum showed the best recovery. However, utilization of this matrix to reconstitute CK-MB material did not produce a significant reduction in method bias, which may be related to different manufacturers’ assay response to this matrix. Further experiments demonstrated that the use of each respective manufacturer’s sample diluent to reconstitute a CK-MB reference material substantially reduced method bias to ± 15% (Fig. 2). The findings of this study show that lyophilized rCK-MB2 can be used as a reference material for standardizing CK-MB mass assays. The systematic bias between the CK-MB methods was reduced from ± 40% to ± 13% with the use of rCK-MB2. Manufacturers should use this material to adjust their CK-MB mass assay calibration curves so that method-related bias is minimized or eliminated for this important AMI diagnostic tool. The next steps of this process include (a) production and purification of large quantities of rCK-MB2, and (b) performance of studies necessary to gain acceptance of lyophilized rCK-MB2 as a worldwide CK-MB standard. Finally, this study may serve as a model for other future standardization activities of new cardiac markers, such as troponin I and myoglobin. In particular, standardization of biochemical markers of cardiac injury will be essential to achieve cost-effectiveness of diagnostic testing of AMI. 1 Members of the AACC CK-MB Mass Assay Standardization Subcommittee. 2 Chairperson of the AACC CK-MB Mass Assay Standardization Subcommittee. 3 Nonstandard abbreviations: CK, creatine kinase; CK-MB, CK-MM, and CK-BB, isoenzymes MB, MM, and BB of CK, respectively; AMI, acute myocardial infarction; CK-M and CK-B, M and B subunits of CK, respectively; rCK-MB2, recombinant CK-MB2 (tissue isoform); SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; and rCK-MB1, recombinant CK-MB1 (serum isoform, missing the C-terminal lysine on the M subunit). We thank Marcella Holdridge (Genzyme Diagnostics) for technical assistance. We thank the participants from various diagnostic immunoassay manufacturers for collecting data for the patient sample correlation studies. Finally, we acknowledge the diligence of Jean Rhame in working with the AACC CK-MB Mass Assay Standardization Subcommittee and in assisting with the assembly of this article. References 1 Wu AHB, Schwartz JG. Update on creatine kinase isoenzyme assays. Diagn Clin Test 1989 ; 27 : 16 -20. 2 Lee TH, Rouan GW, Weisberg MC, Brand DA, Cook EF, Acampora D, Goldman L. Sensitivity of routine clinical criteria for diagnosing myocardial infarction within 24 h of hospitalization. Ann Intern Med 1987 ; 106 : 181 -186. Crossref Search ADS PubMed 3 Vaidya HC, Maynard Y, Dietzler DN, Ladenson JH. Direct measurement of creatine kinase-MB activity in serum after extraction with a monoclonal antibody specific to the MB isoenzyme. Clin Chem 1986 ; 32 : 657 -663. Crossref Search ADS PubMed 4 Delanghe JR, De Mol AM, De Buyzere ML, De Scheerder IK, Wieme RJ. Mass concentration and activity concentration of creatine kinase isoenzyme MB compared in serum after acute myocardial infarction. Clin Chem 1990 ; 36 : 149 -153. Crossref Search ADS PubMed 5 Henderson AR, Krishnan S, Webb S, Cheung CM, Nazir DJ, Richardson H. Proficiency testing of creatine kinase and creatine kinase-2: the experience of the Ontario laboratory proficiency testing program. Clin Chem 1998 ; 44 : 124 -133. Crossref Search ADS PubMed 6 Villarreal-Levy G, Ma TS, Kerner SA, Roberts R, Perryman MB. Human creatine kinase: isolation and sequence analysis of cDNA clones for the B subunit, development of subunit specific probes and determination of gene copy number. Biochem Biophys Res Commun 1987 ; 144 : 1116 -1127. Crossref Search ADS PubMed 7 Friedman DL, Kesterson R, Puleo P, Wu AHB, Perryman MB. Recombinant creatine kinase proteins and proposed standards for creatine kinase isoenzymes and subforms assays. Clin Chem 1993 ; 39 : 1598 -1601. Crossref Search ADS PubMed 8 Chen LH, Babbitt PC, Vasquez JR, West BL, Kenyon GL. Cloning and expression of functional rabbit muscle creatine kinase in Escherichia coli. Addressing the problem of microheterogeneity. J Biol Chem 1991 ; 266 : 12053 -12057. PubMed 9 Anderson G, Scott M. Determination of product shelf life and activation energy for five drugs of abuse. Clin Chem 1991 ; 37 : 398 -402. Crossref Search ADS PubMed 10 Perryman MB, Kerner SA, Bohlmeyer TJ, Roberts R. Isolation and sequence analysis of a full-length cDNA for human M creatine kinase. Biochem Biophys Res Commun 1986 ; 140 : 981 -989. Crossref Search ADS PubMed 11 Landt Y, Vaidya HC, Porter SE, Dietzler DN, Ladenson JH. Immunoaffinity purification of creatine kinase-MB from human, dog, and rabbit heart with the use of monoclonal antibody specific for CK-MB. Clin Chem 1989 ; 35 : 985 -989. Crossref Search ADS PubMed 12 Gill SC, von Hippel PH. Calculation of protein extinction coefficients from the amino acid sequence data. Anal Biochem 1989 ; 182 : 319 -326. Crossref Search ADS PubMed 13 Gella FJ, Frey E, Ceriotti F, Galan A, Hadjivassiliou AG, Horder M, et al. Production and certification of an enzyme reference material for creatine kinase isoenzyme 2 (CRM 608). Clin Chem Acta 1998 ; 276 : 35 -52. Crossref Search ADS 14 Gibler WB, Runyon JP, Levy RC, Sayre MR, Kacich R, Hattemer CR, et al. A rapid diagnostic and treatment center for patients with chest pain in the emergency department. Ann Emerg Med 1995 ; 25 : 1 -8. Crossref Search ADS PubMed 15 Christenson RH, Azzazy HME. Biochemical markers of the acute coronary syndromes. Clin Chem 1998 ; 44 : 1855 -1864. Crossref Search ADS PubMed © 1999 The American Association for Clinical Chemistry This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Standardization of Creatine Kinase-MB (CK-MB) Mass Assays: The Use of Recombinant CK-MB as a Reference Material
JF - Clinical Chemistry
DO - 10.1093/clinchem/45.9.1414
DA - 1999-09-01
UR - https://www.deepdyve.com/lp/oxford-university-press/standardization-of-creatine-kinase-mb-ck-mb-mass-assays-the-use-of-uYe55chA1g
SP - 1414
EP - 1423
VL - 45
IS - 9
DP - DeepDyve
ER -