Clinical Chemistry 49, No. 5, 2003
enzymes in the serum and modulate enzyme activity or interfere with the binding of polyclonal antibodies to the protein. The addition of polyclonal antibodies to cytosolic AST produced complete inhibition of the enzyme activity (5 ). The proposed immunologic method for measuring AST seems to have potential advantages over the conventional enzyme activity assay. In a single liver cell, two types of AST are present: a mitochondrial and a cytosolic form. The mitochondrial form is released into circulation in cases of more severe liver cell damage (2, 15 ). Thus, the degree of liver cell injury may be estimated by determining the ratio of serum mitochondrial and cytosolic AST. A higher concentration of mitochondrial AST may indicate more severe liver damage. Another issue to be considered is the degradation of enzymes in the circulation, with loss of activity by denaturation or degradation.
We thank Dr. K.W. Jeon (University of Tennessee, Knoxville, TN) for reading the manuscript and critical comments. This work was supported by a grant from the National Research Laboratory Program (M1-0104-000164) of the Korean Ministry of Science and Technology. References 1. Rej R. Aminotransferases in disease [Review]. Clin Lab 1989;9:667– 87. 2. Wada H, Kamiike W. Aspartate aminotransferase isozymes and their clinical significance. Prog Clin Biol Res 1990;344:853–75. 3. Dufour DR, Lott JA, Nolte FS, Gretch DR, Koff RS, Seeff LB. Diagnosis and monitoring of hepatic injury. I. Performance characteristics of laboratory tests. Clin Chem 2000;46:2027– 49. 4. Rej R. Quantitation of aspartate aminotransferase isoenzymes by immunologic methods: use of antibodies directed against the mitochondrial isoenzyme. Clin Biochem 1979;12:250 – 4. 5. Rej R. An immunological procedure for determination of mitochondrial aspartate aminotransferase in human serum. Clin Chem 1980;26:1694 – 700. 6. Rej R. Immunochemical quantitation of isoenzymes of aspartate aminotransferase and lactate dehydrogenase. Clin Biochem 1983;16:17–9. 7. Hirano K, Matsuda K, Adachi T, Watanabe Y, Sugiura M, Sawaki S. Enzyme immunoassay of human cytosolic aspartate aminotransferase. Clin Chim Acta 1984;144:49 –57. 8. Suzuki T, Kishi Y, Totani M, Kagamiyama H, Murachi T. Monoclonal and polyclonal antibodies against porcine mitochondrial aspartate aminotransferase: their inhibition modes and application to enzyme immunoassay. Biotechnol Appl Biochem 1987;9:170 – 80. 9. Doyle J, Schinina E, Bossa F, Doonan S. The amino acid sequence of cytosolic aspartate aminotransferase from human liver. Biochem J 1990; 270:651–7. 10. Niblock AE, Jablonsky G, Leung FY, Henderson AR. Changes in mass and catalytic activity concentrations of aspartate aminotransferase isoenzymes in serum after a myocardial infarction. Clin Chem 1986;32:496 –500. 11. Panteghini M. Aspartate aminotransferase isoenzymes. Clin Biochem 1990; 23:311–9. 12. Siest G, Schiele F, Galteau M-M, Panek E, Steinmetz J, Fagnani F, et al. Aspartate aminotransferase and alanine aminotransferase activities in plasma: statistical distributions, individual variations, and reference values. Clin Chem 1975;21:1077– 87. 13. Siest G, Henry J, Schiele F, Young DS. Interpretation of clinical laboratory tests: reference values and their biological variation. Foster City, CA: Biomedical Publications, 1985:459pp. 14. Varasteh A, Wellman M, Artur Y, Schiele F, Siest G. An avidin-biotin ELISA for the measurement of mitochondrial aspartate aminotransferase in human serum. J Immunol Methods 1990;128:203–9. 15. Nishimura T, Yoshida Y, Watanabe F, Koseki M, Nishida T, Tagawa K, et al. Blood level of mitochondrial aspartate aminotransferase as an indicator of the extent of ischemic necrosis of the rat liver. Hepatology 1986;6:701–7.
Method for Hemoglobin A1c Measurement Based on Peptide Analysis by Electrospray Ionization Mass Spectrometry with Deuterium-labeled Synthetic Peptides as Internal Standards, Toyofumi Nakanishi,1 Ken Iguchi,2 and Akira Shimizu1,2* (1 Department of Clinical Pathology and 2 Clinical Laboratory, Osaka Medical College, 2-7, Daigaku-machi, Takatsuki, Osaka 569-8686, Japan; * address correspondence to this author at: Department of Clinical Pathology, Osaka Medical College, 2-7, Daigakumachi, Takatsuki, Osaka 569-8686, Japan; fax 81-726-846548, e-mail [email protected]
) Hemoglobin A1c (HbA1c), which is defined as Hb that is irreversibly glycated at the N-terminal valine of the ␤-chain, is an important index in the monitoring of glucose control in patients with diabetes (1 ). Considerable discrepancies in the measured values of HbA1c have been observed among methods and among laboratories (2, 3 ). To standardize HbA1c methods, Kobold et al. (4 ) proposed a high-level reference method based on liquid chromatography combined with electrospray ionization mass spectrometry (ESI-MS) analysis of the glycated and nonglycated N-terminal hexapeptides of the Hb ␤-chains, which are released by enzymatic cleavage of the intact Hb molecule by endoproteinase Glu-C. Their method is now the officially approved IFCC reference method (5 ). In our laboratory, Hb variants have been measured by ESI-MS (6, 7 ). Specimens containing Hb variants showed unexpected HbA1c values, especially when analyzed by HPLC. To correctly estimate the degree of Hb glycation in such specimens, we have applied the peptide method using ESI-MS as proposed by Kobold et al. (4 ), but the ion signals of peptides occasionally fluctuate, depending on the MS conditions in our laboratory. We devised a method with a stable-isotope-labeled internal standard to obtain more reproducible values in laboratories where mass spectrometers are used for multiple purposes. We also used synthetic nonlabeled peptides to calibrate the measurement, as reported previously (6, 7 ). For HbA1c standardization, we used the calibrators provided by the IFCC working group. The calibrators were prepared by mixing isolated HbA0 and HbA1c (8 ). Four hexapeptides containing the six N-terminal amino acids of the Hb ␤-chain were chemically synthesized by Peptide Institute Inc. (Osaka, Japan): a nonglycated, unlabeled hexapeptide (Val-His-Leu-Thr-Pro-Glu; HD0, lot no. 749-901201), 1-deoxyfructosyl-hexapeptide (GD0; lot no. 480709), hexapeptide labeled with isopropyl-d7 leucine (HD7; lot no. 740-101295), and 1-deoxyfructosylhexapeptide labeled with the same deuterated amino acid (GD7; lot no. 740-102051). The purity of the peptides was ascertained by HPLC as 99.1%, 99.3%, 99.4%, and 99.5%, respectively. We weighed ⬃3 mg of each peptide and dissolved it in distilled water to adjust the concentration to 90 mmol/L. The peptides were mixed in the desired molar ratios. Endoproteinase Glu-C (lot no. PBIO 160016), an enzyme that specifically cleaves the carboxyl side of glutamic acid residues, was purchased from PE Biosystems. Other reagents (spectrophotometric grade) were
purchased from Nacalai Tesque and used without further purification. The MS system was a TSQ7000 triple-stage quadruple mass spectrometer with a conventional electrospray ion source (ThermoQuest). The HPLC system was an Ultrafast Microprotein Analyzer (Michrom BioResources) with a reversed-phase microcolumn [ZORBAX-SB-CN; 150 ⫻ 0.5 mm (i.d.); 5 m bead size]. The electrospray ion source was run with nitrogen at 0.4 MPa. Sheath gas and nitrogen auxiliary gas were used at an HPLC flow rate of 40 L/min. The spray voltage was 4.5 KV, and the transfer capillary temperature was 200 °C. The mass spectrometer was tuned and calibrated with Met-Arg-Phe-Ala and horse muscle apomyoglobin mixture. Univalent ions for the four peptides were selected for ion monitoring: m/z 695.4 for the HD0 hexapeptide, m/z 702.4 for the HD7 hexapeptide, m/z 857.4 for the GD0 hexapeptide, and m/z 864.4 for the GD7 hexapeptide. We chose univalent ions for monitoring because signal noise was lower for monitoring with univalent ions than with divalent ions. The signal intensities of the same deuterium-labeled and unlabeled peptides in the mixture were equivalent, and the ratios of signal intensities were identical to the ratio of molar concentrations of each peptide in the mixture. As internal standards, we added a mixture of labeled glycated and labeled nonglycated peptides in which the molar ratio of labeled glycated to labeled nonglycated peptides was 1:10 (10 L containing 20 moles of labeled glycated peptide and 200 moles of labeled nonglycated peptide) to test samples. To construct a calibration curve for the glycated hexapeptide, we added the internal standard to solutions containing various ratios of unlabeled glycated and nonglycated peptides. We analyzed 100 L of the solutions, which contained 200 pmoles of nonglycated hexapeptide and 0, 5.0, 9.5, 14.0, and 17.5 pmoles of glycated hexapeptide. To construct a calibration curve for HbA1c, we added the internal standard (mixture of labeled glycated and nonglycated peptides in a 1:10 molar ratio) to a solution of the calibrators provided by the IFCC Working Group for HbA1c Standardization after digestion with endoproteinase Glu-C. The digestion was performed according to the method of Kobold et al. (4 ). Two routine samples were measured and compared with both calibration curves to test the intra- and interassay variability of MS analyses. Analyses were repeated 10 times continuously in a day and 5 times on different days. The internal standard containing labeled glycated and nonglycated peptides (1:10 molar ratio) was added to the solution after digestion with endoproteinase Glu-C. Using selected-ion monitoring of the column eluate, we measured peak areas at m/z 695.4 for HD0, m/z 702.4 for HD7, m/z 857.4 for GD0, and m/z 864.4 for GD7 (all ions were univalent) and calculated the peak intensity ratio of glycated to nonglycated peptides by the following equation: 1/10 ⫻
Fig. 1. Calibration curves for the percentages of glycated peptide (A) and HbA1c (B). (A), calibrators were synthetic hexapeptides. The peak intensities (%) of glycated peptide were measured by the proposed ESI-MS method, using deuteriumlabeled internal standards. The percentage was calculated with use of the equation described in the text. All data points are the means (SE; error bars) of three experiments. Values on the x axis represent the percentage of glycated peptide/total peptides; values on the y axis represent the peak intensity (%) measured by the MS peptide method using labeled internal standards. The regression equation for the calibration line is: y ⫽ 0.97x ⫹ 0.11% (R2 ⫽ 0.9979). (B), calibrators were those provided by the IFCC Working Group for HbA1c Standardization. HbA1c was measured by the proposed ESI-MS method using deuterium-labeled internal standards. Values on the x axis represent the percentage of the target values certified by the IFCC working group; values on the y axis represent the peak height (%) measured by the MS peptide method using labeled internal standards. The results were calculated by the equation described in the text. All data points are the means (SE; error bars) of two experiments. The regression equation for the calibration line is: y ⫽ 1.05x (R2 ⫽ 0.9984).
Clinical Chemistry 49, No. 5, 2003
We then calculated the peak intensity (%) of the glycated peptides by the following equation: (Ratio) ⫻ 100 (1 ⫹ Ratio) As shown in Fig. 1A, we obtained a highly linear calibration curve for the glycated hexapeptides over a wide concentration range (0 –20%). We also obtained a highly linear calibration curve for HbA1c over a wide concentration range (Fig. 1B). As shown in Table 1, the peak intensity values for the two samples used to assess assay variability were highly reproducible (CVs ⱕ2%). The percentages of glycated peptide and HbA1c calculated on the basis of their respective calibration curves were similar. The causes of the small discrepancies between the values obtained with the two calibration curves should be identified by further experiments. Probable causes may include the different susceptibilities of glycated and nonglycated Hb to the enzyme, errors in mixing glycated and nonglycated materials, and impurities in the calibrators. We previously reported a method that involved monitoring of both univalent and divalent ions (6, 7 ), whereas Kobold et al. (4 ) measured only divalent ions, and in the present report, we measured only univalent ions. Use of the expression (0.5 ⫻ peak area of doubly protonated ions ⫹ 1 ⫻ peak area of univalent ions) for each peptide to calculate the ratio of the peak intensities for both peptides improved the reproducibility of the ratio over that calculated with use of divalent ions only. However, in repeated experiments, the values obtained without use of labeled internal standards varied, depending on instrument conditions. The values reported here were fairly constant among assays on different days. Daily calibration is not necessary because the peak intensity ratios for labeled glycated and nonglycated peptides were highly reproducible. The peak intensity ratio for glycated/nonglycated peptides measured with the standard material, in which the concentration ratio of glycated/nonglycated was 1:10, remained constant with a mean (SD) value of 0.189 (0.007) and a CV of 1.2%. We have used the proposed method to assess the HbA1c
Table 1. Results of studies on intra- and interassay variability. Peak intensity
Intraassay Sample I Sample II Interassay Sample I Sample II a
Number of analyses. Values were calculated with calibration curves constructed with synthetic hexapeptides. c Values were calculated with calibration curves constructed with HbA1c solution. b
values in specimens that contain abnormal Hb. The method is convenient and reliable for determining HbA1c in such specimens.
We thank the IFCC Working Group for HbA1c Standardization for providing the calibrators. References 1. Bunn HF. Evaluation of glycosylated hemoglobin diabetic patients. Diabetes 1981;30:613–7. 2. Weykamp CW, Penders TJ, Muskiet FAJ, van der Slik W. Influence of hemoglobin variants and derivatives on glycohemoglobin determinations, as investigated by 102 laboratories using 16 methods. Clin Chem 1993;39: 1717–23. 3. Goldstein DE, Little RR, Wiedmeyer HM, England JD, McKenzie EM. Glycated hemoglobin: methodologies and clinical applications. Clin Chem 1986;32: B64 –70. 4. Kobold U, Jeppsson JO, Du¨ lffer T, Finke A, Hoelzel W, Miedema K. Candidate reference methods for hemoglobin A1c based on peptide mapping. Clin Chem 1997;43:1944 –51. 5. Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem 2002;48:436 –72. 6. Nakanishi T, Miyazaki A, Iguchi K, Shimizu A. Effect of hemoglobin variants on routine glycohemoglobin measurements assessed by a mass spectrometric method. Clin Chem 2000;46:1689 –92. 7. Nakanishi T, Miyazaki A, Shimizu A, Yamaguchi A, Nishimura S. Assessment of the effect of hemoglobin variants on routine HbA1c measurements by electrospray ionization mass spectrometry. Clin Chim Acta 2002;323:89 – 101. 8. Finke A, Kobold U, Hoelzel W, Weykamp C, Miedema K, Jeppsson JO. Preparation of a candidate primary reference material for the international standardization of HbA1c determinations. Clin Chem Lab Med 1998;36:299 – 308.
Post-Race Kinetics of Cardiac Troponin T and I and N-Terminal Pro-Brain Natriuretic Peptide in Marathon Runners, Markus Herrmann,1,2 Ju¨ rgen Scharhag,2 Marina Miclea,2 Axel Urhausen,2 Wolfgang Herrmann,1* and Wilfried Kindermann2 (1 Department of Clinical Chemistry/Central Laboratory, University Hospital of Saarland, D-66421 Homburg/Saar, Germany; 2 Institute of Sports and Preventive Medicine, University of Saarland, Stadtwald, 66123 Saarbru¨ cken, Germany; * author for correspondence: fax 49-6841-1623109, e-mail [email protected]
) Annually there are cases of sudden cardiac death during and after marathon races (1–3 ), which has caused athletes and physicians to frequently ask whether marathon running damages the heart. Modern laboratory analyses, such as tests for cardiac troponin T and I (cTnT and cTnI) and N-terminal pro-brain natriuretic peptide (NTproBNP), provide additional information about cardiac cell damage and wall stress with high sensitivity and specificity (4 –7 ). Previous studies have investigated cTnT and cTnI in runners, cyclists, and triathletes (8 –16 ), but the results are controversial, mainly because the assays were first- (cTnI) or second-generation (cTnT) troponin assays and the cutoff points were inconsistent. In the present study we investigated cTnT and cTnI during a marathon race with third- (cTnT) and second-generation