Printer Friendly

Characterization of a new certified reference material for human cardiac troponin I.

The clinical measurement of serum cardiac troponin I (cTnI) (1) has become an important tool in the diagnosis of acute myocardial infarction (1) and myocardial damage (2-5). Unfortunately, variability in clinical cTnI assay results has been reported. A 10-fold difference in assay response between methods is common (6),but as much as a 100-fold difference in response has also been observed (7). With so much variation among assays, physicians and clinical laboratory staff are required to establish their own decision points for cTnI based on the assay used. Problems can arise when different assays are being used, such as when clinical laboratories change assays, when an attempt is made to correlate results between point-of-care testing platforms and a central laboratory, or when a clinician compares results from different laboratories. Standardization of clinical cTnI measurements is needed to provide more reliability in the use of cTnI assays for the diagnosis of myocardial infarction and damage.

With the assistance of the AACC/IFCC Cardiac Troponin I Standardization Committee (chaired by Stephen Kahn, Loyola University Medical Center), NIST has developed a Certified Reference Material for the standardization of clinical cTnI assays. NIST Standard Reference Material (SRM) 2921 is a solution of the human cardiac troponin complex, purified from human heart tissue. The choice of a human cardiac troponin complex, composed of the troponin I, troponin T (cTnT), and troponin C (cTnC) subunits, was made after 2 interlaboratory comparison studies by the AACC/IFCC Cardiac Troponin I Standardization Committee and NIST. The first comparison study (8) narrowed a field of 10 candidate reference materials, consisting of human and recombinant proteins with different degrees of troponin complexation. After evaluation of the abilities of the candidate reference materials to harmonize the 13 commercial cTnI assays evaluated, 2 candidate reference materials were chosen for further study in a second interlaboratory comparison study. Following the second study, which evaluated the 2 candidate reference materials (a recombinant troponin C-I complex and a human cardiac troponin C-I-T complex) using 15 commercial cTnI assays (unpublished work), the human troponin C-I-T complex was the troponin form chosen for SRM 2921.

We report here the results from the characterization and quantitative certification of SRM 2921. Extensive protein structural characterization was carried out on the troponin complex in SRM 2921. This information is provided so that clinical troponin assay manufacturers can use this reference material in the development of new or "next-generation[degrees] immunoassays with knowledge of the antigen s chemical structure as a guide in understanding assay performance. This troponin reference material is intended for use in calibrating clinical cTnI assays. Once commutability (9) with native clinical samples has been validated, the SRM will be useful to establish traceability to the International System of Units (SI) and for comparing the accuracy of clinical assays for the determination of cTnI in patient samples.

Materials and Methods


Approximately 30 mg (as determined by the manufacturer) of human cardiac troponin complex (lot no. 03/018T62), purified from healthy human hearts, was obtained from HyTest Ltd. The troponin complex was supplied in a buffer composed of 150 mmol/L sodium chloride, 5 mmol/L calcium chloride, 20 mmol/L Tris-HCI, pH 7.5. Purified human cTnI subunit (lot no. 02/10-8T53) was also obtained from HyTest. Both materials were shipped frozen, on dry ice, from HyTest and stored at NIST at -80 [degrees]C until used.

The HyTest human cardiac troponin complex solution was removed from the freezer and thawed for ~1 h at room temperature, then transferred to a 1-L Teflon bottle (Nalgene) and diluted to ~0.6 L with the addition of 0.57 L of buffer (150 mmol/L sodium chloride, 5 mmol/L calcium chloride, 20 mmol/L Tris-HCI, pH 7.5). The concentration of the resulting solution was ~50 mg/L, based on the manufacturer-reported concentration of the prediluted troponin solution.

After dilution, the troponin solution was aliquoted into 0.5-mL polypropylene screw-capped microcentrifuge tubes by use of a Digiflex CX pipettor (Titertek) with a 2-mL syringe and a 0.5-mm dispensing tip. Before aliquoting, the troponin solution was circulated through the pipettor and back into the Teflon bottle to equilibrate the wettable parts of the pipettor with the protein solution. During the aliquoting, the troponin solution was stirred slowly in the Teflon bottle in an ice bath. Approximately 4500 microcentrifuge tubes were each filled with 115 [micro]L of troponin solution. The tubes were placed in numbered storage racks of 100 tubes before being stored temporarily at -20 [degrees]C. After all of the tubes were filled and capped, 1 tube from each of the 45 storage trays was randomly removed, numbered according to the storage tray number, and stored at -80 [degrees]C until analysis. After remaining overnight at -20 [degrees]C, the remaining tubes were moved to a -80 [degrees]C freezer for permanent storage.


Reversed-phase liquid chromatography (RPLC) analysis of troponns was performed on a Model 1100 LC system (Agilent) consisting of a Mode11100 binary pump, Model 1100 thermostated well-plate autosampler, Model 1100 column oven, and Mode11100 variable wavelength absorbance detector. The column used for separation was a 150 x 2.1 mm (i.d) Zorbax 3005B Cg column (Agilent) heated to 35 [degrees]C in the column oven. The mobile phases used consisted of 1 mL/L trifluoroacetic acid (TFA) in water and 0.85 mL/L TFA in acetonitrile. Gradient elution was used for separation of the troponin subunits. The injection volume was 100 [micro]L. For detection, the ultraviolet (UV) absorbance was monitored at 220 mn. Samples of SR1VI 2921 for RPLC analysis were prepared as described above.

The procedure used for amino acid analysis was a modified version of the procedure developed by Agilent (10), which uses the amine-specific derivatization agent o-phthalaldehyde. The Agilent procedure uses the autosampler to mix the samples and o-phthalaldehyde reagent and then inject them for gradient LC analysis with UV absorbance detection. The column used for the separation was a Zorbax Eclipse AAA [150 X 4.6 mm (i.d.); 3.5-[micro]m particle size]. The Agilent 1100 liquid chromatograph described above was used for the analysis. NIST SRM 2389 (amino acids in 0.1 mol/L hydrochloric acid) was used to calibrate the amino acid analysis.

Relative molecular mass ([M.sub.r]) determinations of the troponin subunits were performed by capillary liquid chromatography-mass spectrometry (LC-MS) using a Waters CapLC liquid chromatograph coupled to a Waters ZMD single quadrupole mass spectrometer. The proteins were separated on a Vydac [C.sub.4] reversed-phase LC column [50 x 0.5 mm (i.d.)] by gradient elution with 0.5 mL/L TFA in water and 0.425 mL/L TFA in acetonitrile as mobile phases. The flow rate used was 25 [micro]L/min. Full-scan mass spectra were obtained of the column eluate by scanning from m/z 600 to m/z 1500 in 2 s in positive-ion mode using electrospray ionization.

Matrix-assisted laser desorption/ionization (MALDI) mass spectra were acquired by use of a delayed-extraction time-of-flight mass spectrometer (Voyager-DE STR; Applied Biosystems) operated in linear mode for protein analytes and reflector mode for the analysis of peptides. The acquisition rate was 50 shots/spectrum, and 5 spectra were averaged (accumulated) across random points on the sample spot. Sinapinic acid (3,5-dimethoxy-4-hydroxycinnamic acid) and [alpha]-cyano-4-hydroxycinnamic acid were the matrices used for proteins and peptides, respectively.


Protein amino acid sequences were obtained from the Internet database Swiss-Prot (11). Analysis of proteolytic digest peptide molecular mass data was assisted by the Internet-based PeptideMass (12) software from SwissProt as well as by use of Protein Prospector (13). A mass difference of 25 ppm or less was used as the acceptance criterion for a positive identification when comparing observed peptide masses with calculated masses in the analysis of MALDI-MS data. Deconvolution of the electrospray charge state distributions of proteins was accomplished with use of MaxEnt 1 software (Waters).



The concentration of the troponin I subunit in SRM 2921 was determined by a combination of methods. The first method used RPLC with UV absorbance detection (RPLCUV) and external calibration through purified cTnI standards. The second method used amino acid analysis to determine the concentration of cTnI in fractions collected from the RPLC separation of the 3 troponin subunits (cTnT, cTnI, and cTnC) that comprise the troponin complex. Additionally, amino acid analysis was used to determine the concentrations of the cTnT and cTnC subunits.

Preliminary RPLC-UV analysis of SRM 2921 was used to guide in the preparation of 2 calibration standards from purified human cTnI. The purity of the cTnI standards was assessed by LC-MS analysis, and no measurable protein contaminant was found. The RPLC-UV heights of the cTnI peak in the 2 calibration standards bracketed that of the cTnI in SRM 2921. The concentrations of these 2 cTnI standards were then determined by amino acid analysis from aliquots of each standard taken on each day of analysis. The troponin solutions in 4 SRM 2921 vials were analyzed each day, bracketed by injections of the 2 cTnI calibration standards. This analysis scheme was repeated on 4 separate days. A typical chromatogram from the RPLC-UV analysis of SRM 2921 is shown in Fig. 1. On each day of analysis, a calibration curve was prepared that related chromatographic peak heights to cTnI concentrations in the standards. From the peak height of the cTnI peak in the chromatogram of an injection of SRM 2921, the cTnI concentration in SRM 2921 could be calculated by use of the calibration curve. Because the cTnI calibration standards covered a very limited range of protein concentration (~25 to 40 mg/L), it was assumed that the UV absorbance of the protein would exhibit a linear relationship to protein concentration; therefore, only a 2-point, bracketing calibration curve was used. From this analysis procedure, the concentration of cTnI in SRM 2921 was determined to be 31.71 (1.60) mg/L. The uncertainty in the concentration (1.60 mg/L) is calculated as U = [ku.sub.c]. The quantity [u.sub.c] is the standard uncertainty calculated according to the 150 Guide (14,15), where [u.sub.c] is intended to represent the measurement error at the level of 1 SD. The coverage factor, k, is determined from the Student t-distribution corresponding to the appropriate associated degrees of freedom and a 95% level of confidence. A value of 3.544 was used for k in this instance.


The second method to determine the cTnI concentration in SRM 2921 used amino acid analysis of the cTnI subunit purified from the other troponin subunits by RPLC. During the RPLC-UV analysis described above, 2 fractions from the cTnI peak were collected on each of the 4 days of analysis and analyzed by amino acid analysis quantifying the amino acids alanine (Ala), valine (Val), and leucine (Leu). The decision to use these 3 amino acids was based on several factors, including their high abundance in cTnI, their chemical inertness (relative to their likelihood to be posttranslationally modified in the protein and their chemical stability during acid hydrolysis), and their chromatographic peak shape and resolution in the amino acid analysis. A comparison of the results obtained from each of these 3 amino acids for determination of the concentration of cTnI in SRM 2921 is shown in Table 1. Each amino acid provided a determination of the concentration of cTnI in the fraction. The mean of the results from all 3 amino acids was used to calculate the cTnI concentration in each RPLC fraction. The mean cTnI concentration based on all samples collected for amino acid analysis was determined to be 30.65 (2.37) mg/L, where again the uncertainty (2.37 mg/L) is given at the 95% level of confidence.

The measured concentrations of cTnI in SRM 2921, determined by both the RPLC-UV and amino acid methods, were combined to give the NIST certified concentration. The combination of the 2 values is the weighted mean, with the weighting based on a type B uncertainty analysis to estimate the systematic component of uncertainty in the combined value. The certified value of cTnI in SRM 2921 was determined to be 31.2 (1.4) mg/L at the 95% limit of confidence.


Additionally, amino acid analysis was used to determine the concentrations of the cTnT and cTnC subunits. Just as with the cTnI subunit, fractions were collected from the cTnT and cTnC peaks during RPLC-UV analysis. Two fractions were collected each day of 4 separate days of analysis and subjected to amino acid analysis. As with cTnI, the amino acids alanine, valine, and leucine were used for the amino acid quantification. This approach gave concentrations of the cTnT and cTnC subunits in SRM 2921 of 36.9 (3.8) mg/L and 24.2 (1.3) mg/L, respectively. The uncertainly in these values is expressed at the 95% confidence level. These values are provided on the NIST Certificate of Analysis for SRM 2921 as reference concentration values because they do not meet NIST criteria for certification and are provided with associated uncertainties that may reflect only measurement precision and may not include all sources of uncertainty.


Ideally, the 3 subunits that make up the troponin complex should be present in equimolar concentrations. The molar concentrations of the cTnT, cTnI, and cTnC subunits in SRM 2921 were compared to assess the purity and completeness of the troponin complex. Results from the amino acid analysis for all 3 subunits were obtained from 3 aliquots of SRM 2921. The observed mean molar concentration ratios for the subunit pairs cTnT/cTnI and cTnC/ cTnI were 0.84 (0.05) and 1.03 (0.04), respectively, and were obtained by the amino acid analysis of Val, Ala, and Leu. The uncertainties in the observed molar concentration ratios (in parentheses) are expressed at the 95% confidence level and were calculated as U = [ku.sub.c] . The coverage factor, k, was determined from the Student t-distribution corresponding to the appropriate associated degrees of freedom and a 95% level of confidence. A value of 3.182 was used for k in this instance.

The normalized ratios of the molar concentrations of the cTnI and cTnC subunits were essentially equal (1.03), but the normalized ratio of the cTnT concentration to that of cTnI was ~0.84, less than the equimolar value of 1.00. The cTnI and cTnC subunits are known to have a very strong noncovalent bond; therefore, the above results would suggest that the purity of the complete troponin complex (C-I-T complex) is ~85%, with the remaining 15% being the binary troponin C-I complex.


MS analysis of intact proteins and of the peptides resulting from proteolytic cleavage was used for structural characterization. A typical total ion chromatogram from the LC-MS analysis of SRM 2921 is shown in Fig. 1 in the Data Supplement that accompanies the online version of this article at Three abundant peaks are apparent, corresponding to the troponin T, troponin I, and troponin C subunits, respectively. From deconvolution of the charge-state distribution (16) of the averaged mass spectrum for each of the 3 main chromatographic peaks (shown in Fig. 2), the relative molecular masses ([M.sub.r]) of the most abundant species observed were determined and are listed in Table 2. Also listed in Table 2 are the relative molecular masses observed from MALDI-MS analysis of SRM 2921. The tentative identifications of the molecular species listed in Table 2 are based on the observed relative molecular masses of each species, the known posttranslational modifications of each troponin subunit, and when observed, the peptides in the enzymatic digest of each troponin subunit. The uncertainties in the [M.sub.r] values in Table 2 are expressed at the 95% level of confidence.

The first eluting subunit in the LC-MS analysis, cTnT, had 1 major component and several lower-abundance components (Fig. 2A). However, the observed molecular mass of the major component was ~44 mass units higher than the mean relative molecular mass of the principal known isoform of cTnT, calculated from its amino acid sequence (17). This mass discrepancy could indicate that the N[H.sub.2] terminus of cTnT is N-acetylated (increasing the relative nominal molecular mass by 42), a hypothesis supported by the fact that sequence analyses of troponin subunits have often demonstrated a blocked N[H.sub.2] terminus (18-22), commonly observed from N-terminal acetylation. However, the Swiss-Prot protein sequence database lists phosphoserine and not N-acetylserine as a possible posttranslational modification of the cTnT N[H.sub.2] terminus. The calculated relative molecular mass of monophosphorylated cTnT ([M.sub.r] = 34 539) does not correspond to any of the observed molecular species. If a relative molecular mass increase of 42 is considered, the peaks at mass 33 775 (8) and 34 373 (8) can be tentatively identified as C-terminal-truncated forms of cTnT, losing residues 282 through 287 (calculated [M.sub.r] = 33 773) or just residue 287 (calculated [M.sub.r] = 34 373), respectively. The protein database at the National Center for Biotechnology Information lists several known isoforms for human cTnT. Additionally, there are known variations in the amino acid sequence, polymorphisms that are characterized by single amino acid substitutions. Because the troponin complex used in SR1VI 2921 was obtained from several human hearts, it is possible that SR1VI 2921 might contain a mixture of variant and/or isoforms of cTnT. However, a comparison between the calculated relative molecular masses of known cTnT isoforms and variants indicated no close matches to any of the relative molecular masses observed for the cTnT peak in SR1VI 2921.


The observed molecular mass distribution of the cTnI in SR1VI 2921 (Fig. 2B) shows a substantial degree of heterogeneity. The pattern of components and, specifically, the mass differences between components indicate posttranslational modifications such as phosphorylation and truncation of the COOH terminus. Both types of posttranslational modifications have been reported previously for purified human cTnI standards (18). The types of cTnI species observed in SR1VI 2921 are nearly identical to those found in the 2 candidate reference materials that were obtained previously from HyTest. However, SR1VI 2921 differs in the relative distribution of these species: SR1VI 2921 appears to contain more C-terminally truncated materials than were observed in previous lots of the troponin complex. Additionally, a species with an [M.sub.r] of 23 918 (5)--tentatively assigned as nonphosphorylated, intact cTnI--was not observed in previous HyTest samples but gives the most abundant peak in the LC-MS analysis of SR1VI 2921.

A clean molecular mass distribution was observed for cTnC (Fig. 2C), with 1 major component and several minor components. The observed relative molecular mass of the major component [[M.sub.r] = 18 445 (4)] is in excellent agreement with the calculated relative molecular mass of cTnC (calculated [M.sub.r] =18 445), based on its known amino acid sequence and N-terminal acetylation. In our literature search on the amino acid sequence of cTnC, we found 1 variant and 2 sequence conflicts. The calculated mean relative molecular masses of the N-acetylated cTnC L29Q variant and the N-acetylated cTnC D115E sequence conflict form are 18 459.5 and 18 458.6, respectively; both masses are close to the mass [18 463 (8)] of one observed component for cTnC in SR1VI 2921. Structures for the other observed components could not be assigned based on known variants, isoforms, sequence conflicts, or posttranslational modifications of cTnC.

To confirm the molecular structural assignments made in Table 2 and for a more detailed investigation of potential posttranslational modifications, the 3 troponin subunits were subjected to enzymatic digestion and the resulting peptide mixtures were analyzed by MALDI-M5. To obtain peptides from each of the 3 troponin subunits separately, we performed the digestions on fractions collected from the RI'LC separation of SR1VI 2921 (see Fig. 1). Three separate enzymatic digestion protocols were used: digestion with trypsin, with endoproteinase Glu-C, and with endoproteinase Glu-C, followed by digestion with endoproteinase Lys-C. Because each proteolytic enzyme specifically cleaves after different amino acids, a different set of peptides was produced with each digestion protocol. The MALDI mass spectra of the peptide mixtures resulting from trypsin digestion of all 3 troponin subunits are shown in Fig. 2 of the online Data Supplement.

The resulting sequence coverage maps from all 3 digestion protocols are shown in Fig. 3 of the online Data Supplement. The amino acid sequence coverage, combining the results obtained from the 3 enzymatic digestion procedures, was 85% for cTnT, 96% for cTnI, and 52% for cTnC. For cTnT, digest peptides were not observed from the N-terminal region of the protein. The N[H.sub.2] terminus of cTnT is rich in glutamic acid residues, which produces very small peptides when digested with endoproteinase Glu-C, and low in lysine and arginine residues, which produces very large peptides when digested with trypsin or endoproteinase Lys-C. Neither very small nor very large peptides are readily seen in MALDI-MS when analyzing complex mixtures such as a proteolytic digest. Because of this, the N-terminal acetylation that was postulated based on the observed relative molecular mass of the intact protein could not be verified through peptide mapping. The 3 enzymatic digestion protocols produced very high sequence coverage for cTnI, including several peptides verifying acetylation of the N-terminal region. Unfortunately, phosphorylated peptides were not observed; therefore, the peptide mapping could not confirm the mono-and bisphosphorylation of cTnI that is suspected based on the relative molecular masses of the intact protein.

The low sequence coverage for cTnC is difficult to explain. Although many peptides were observed in the MALDI mass spectra of each enzymatic digest of the cTnC fraction, only a small number of these peptides could be attributed to expected digest peptides from cTnC: <50% in the tryptic digest, <40% for the endoproteinase Glu-C digest, and <50% in the combined endoproteinase GluC/Lys-C digest. Additionally, the identified digest peptides from each of the 3 digestion protocols tended to overlap the same regions of cTnC, leaving nearly 50% of the sequence unverified. The observed peptides that could not be attributed to expected peptides from cTnC were used to search the Swiss-Prot protein sequence database in the belief that these peptides might result from a contaminating protein that coelutes with cTnC. However, a database search did not produce any strong matches to known human proteins.


During the course of the RPLC fraction collection of the troponin subunit peaks for amino acid analysis, fractions were also collected for low-abundance peaks as well. These peaks are labeled I-1 through I-5 in the RPLC-UV chromatogram of SRM 2921 (Fig. 1). These 5 fractions were analyzed by MALDI-MS, both before and after digestion with trypsin. Fraction I-1, eluting just before cTnI, produced a distribution of ~12 low-intensity peaks between m/z 10 000 and m/z 13 000 when analyzed directly by MALDI-MS. After digestion with trypsin and analysis of the tryptic peptides by MALDI-MS, 12 peptides were observed with masses nearly identical to tryptic peptides observed for the cTnT subunit. Considering these results, it can be postulated that fraction I-1 contains a degradation product of cTnT.

Fraction I-2, was characterized similarly to fraction I-1. Direct MALDI-MS analysis produced a distribution of ~6 peaks between m/z 8900 and m/z 22 000. Again, ~12 peptides from the tryptic digest of fraction I-2 matched tryptic peptides of cTnT. Fraction I-2 also appears to contain a degradation product or products of cTnT.

Fraction I-3, eluting just before cTnI in RPLC, produced a MALDI mass spectrum with several strong signals, including a series of peaks closely matching the peak distribution for cTnI. In the tryptic digest of fraction I-3, eight peptides matched the masses of expected cTnI tryptic peptides. Additionally, there were strong signals at m/z 32 780 [+ or -] 50 and m/z 21 860 [+ or -] 30 in the MALDI mass spectrum of fraction I-3. Peptide mass fingerprinting (23) with protein sequence database searching of the tryptic peptides from fraction I-3 that were not attributable to cTnI produced 2 human proteins that can explain the MALDI-MS results for fraction I-3: human tropomyosin 1 a-chain (calculated [M.sub.r] = 21 801) and human myosin 1 light chain (calculated [M.sub.r] = 32 709). The database search of fraction I-3 tryptic peptides produced 5 peptides matching the expected masses of human myosin 1 tryptic peptides and 8 peptides matching tryptic peptides from human tropomyosin 1. Both proteins are found in human cardiac muscle tissue.

Analysis of fraction I-4 indicated that it possibly contained cTnI and/or degradation products of cTnI. Tryptic digestion of fraction I-5 produced 9 peptides with approximately the same observed relative molecular masses as unassigned peptides from the tryptic digest of cTnC. However, only 1 observed peptide mass from the tryptic digest of fraction I-5 matched an expected tryptic digest of cTnC. A database search using the observed relative molecular masses of the remaining peptides matched 11 peptides to expected tryptic peptides from human cardiac actin (24).


Through collaboration with the AACC and IFCC, manufacturers of clinical cTnI assays, and clinical practitioners, NIST has developed a certified reference material intended for the standardization and validation of cTnI assays. It has been a unique challenge to develop a reference material for an analyte whose chemical structure is not precisely known. Studies that indirectly probe the structures) of cTnI in the bloodstream, using antibodies of known specificity, were used to guide the selection of candidate reference materials. The expertise of clinical practitioners and assay manufacturers has been critical to the evaluation of candidate reference materials and finally in the selection of the cTnI preparation for SRM 2921.

It is very important to certify the concentration of cTnI in SRM 2921 in a way that provides traceability to the SI system of units. Because of their intrinsic structural heterogeneity, proteins such as cTnI were not considered "SI-traceable" analytes, unlike a smaller molecule with well-defined chemical structure, such as cholesterol. Recently, a new WHO reference standard for human chorionic gonadotropin and its metabolites was certified by amino acid analysis to provide an "amount-of-substance" or SI-traceable concentration value (25), demonstrating that SI traceability can be achieved for a clinical protein analyte. Amino acid analysis was also used for protein quantification of SRM 2921, in particular, amino acid analysis calibrated with a gravimetrically prepared amino acid standard (i.e., NIST SRM 2389). Therefore, once commutability has been validated, use of SRM 2921 for the calibration of clinical cTnI assays will provide SI traceability to patient cTnI values, an important step forward in global harmonization.

With strong evidence that structural changes in the cTnI molecule can significantly influence immunoassay results (26-28), characterization of the protein chemical structure was a necessary step in the evaluation of SRM 2921. Posttranslational modifications to the troponin complex can strongly affect the binding to antibodies if the antigen/antibody binding epitope is modified or adjacent to the modification. With this in mind, substantial effort was placed on the structural characterization of the troponin complex in SRM 2921. Through a combination of selective and sensitive analytical techniques, we were able to obtain detailed information of the chemical structure at the molecular level. This type of information should be useful when evaluating antibody binding behavior during the design of clinical immunoassays. All questions regarding the structure of the troponin complex and the individual subunits in SRM 2921 could not be answered. This is particularly true for cTnC because the masses of many of the peptides in enzymatic digests of cTnC could not be readily explained from knowledge about the chemical structure of cTnC. However, we believe this reference material is sufficiently well characterized to serve its intended purpose. As more becomes known about the detailed chemical structure of troponin released into the bloodstream after myocardial damage, additional characterization of SRM 2921 can be done.

In the preparation of SRM 2921, every attempt was made to adhere to 150 15194 (29), the international standard describing biological reference materials. However, 2 aspects of 150 15194 are not currently fulfilled: stability testing and commutability assessment. The stability of SRM 2921 stored at -80 [degrees]C will be evaluated on an ongoing basis. Accelerated stability studies (30, 31) are sometimes used to evaluate the storage stability of pharmaceuticals, particularly when ideal storage conditions cannot be guaranteed, for example, to predicate shelf-life of medication. We do not feel that an accelerated stability study for a protein reference material such as SR1VI 2921 would accurately predict stability when the material is stored under ideal conditions (e.g., kept frozen at -80 [degrees]C). HyTest Ltd. has evaluated the stability of their human cardiac troponin complex when stored at -80 [degrees]C for several years and has found no apparent degradation when evaluated by immunoassay. NIST will also monitor the stability of SRM 2921 during the course of its estimated lifetime.

A commutability study is another very important element in the successful development of a clinical reference standard and is a component of 150 15194. A commutability study of SRNI 2921 is underway. Through completion of this last study, SRNI 2921 should achieve compliance with 150 15194.

We thank Lorna Sniegoski, Susan Tai, Mary Satterfield, and Magali Theodore for assistance in the preparation of SRNI 2921, and Charles Hagwood, from the Statistical Engineering Division of NIST, for statistical evaluation of the quantitative results. We also thank the AACC/IFCC Cardiac Troponin I Standardization Committee, chaired by Stephen Kahn, for their guidance and technical support throughout the development process of this new reference material. Certain commercial equipment, instruments, or materials are identified in this report to specify adequately the experimental procedure. Such identification does not imply recommendation or endorsement by the NIST, nor does it imply that the materials or equipment identified are the best available for the purpose.

Received AprII 20, 2005; accepted November 3, 2005.

Previously published online at DOI: 10.1373/clinchem.2005.051359


(1.) Alpert JS, Antman E, Apple F, Armstrong PW, Bassand JP, de Luna AB, et al. Myocardial infarction redefined: a consensus document of the Joint European Society of Cardiology/American College of Cardiology committee for the redefinition of myocardial infarction. J Am Coll Cardiol 2000;36:959-69.

(2.) Sybrandy KC, Cramer MJM, Burgersdijk C. Diagnosing cardiac contusion: old wisdom and new insights. Heart 2003;89:485-9.

(3.) Plebani M. Biochemical markers of cardiac damage: from efficiency to effectiveness. Clin Chim Acta 2001;311:3-7.

(4.) Ross GE, Bever FN, Uddin Z, Devireddy L, Gardin JM. Common scenarios to clarify the interpretation of cardiac markers. J Am Osteopath Assoc 2004;104:165-76.

(5.) Davis GK. Role of cardiac troponin testing in percutaneous transluminal coronary angioplasty. Stand J Clin Lab Invest 2003;63: 167-74.

(6.) Christenson RH, Apple FS, Morgan DL, Alonsozana GL, Mascotti K, Olson M, et al. Cardiac troponin I measurement with the ACCESS[R] immunoassay system: analytical and clinical performance characteristics. Clin Chem 1998;44:52-60.

(7.) Apple FS, Maturen AJ, Mullins RE, Painter PC, Pessin-Minsley MS, Webster RA, et al. Multicenter clinical and analytical evaluation of the AxSYM troponin-I immunoassay to assist in the diagnosis of myocardial infarction. Clin Chem 1999;45:206-12.

(8.) Christenson RH, Duh SH, Apple FS, Bodor GS, Bunk DM, Dalluge J, et al. Standardization of cardiac troponin I assays: round robin of ten candidate reference materials. Clin Chem 2001;47:431-7.

(9.) Rej R, Drake P. The nature of calibrators in immunoassays: are they commutable with test samples-must they be? Stand J Clin Lab Invest 1991;51:47-54.

(10.) Henderson JW, Ricker RD, Bidlingmeyer BA, Woodward C. Rapid, accurate, sensitive, and reproducible analysis of amino acids. Agilent Technical Publication 5980-1193EN. [Last updated July 2005]. Palo Alto, CA: Agilent Technologies, 2002:10pp.

(11.) Boeckmann B, Bairoch A, Apweiler R, Blatter MC, Estreicher A, Gasteiger E, et al. The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003. Nucleic Acids Res 2003;31:365-70.

(12.) Wilkins MR, Lindskog I, Gasteiger E, Bairoch A, Sanchez JC, Hochstrasser DF, et al. Detailed peptide characterization using PEPTIDEMASS-a World-Wide-Web-accessible tool. Electrophoresis 1997;18:403-8.

(13.) Clauser KR, Baker P, Burlingame AL. Role of accurate mass measurement ([+ or -]10 ppm) in protein identification strategies employing MS or MS/MS and database searching. Anal Chem 1999;71:2871-82.

(14.) International Organization for Standardization. Guide to the expression of uncertainty in measurement. Geneva, Switzerland: International Organization for Standardization, 1993. Measurements.PublicationList?CLASSIFICATION=MEASUREMENTS (accessed September 2005).

(15.) Taylor BN, Kuyatt CE. Guide for evaluating and expressing uncertainty of NIST measurement results. NIST Technical Note 1297. Washington, DC: US Government Printing Office, 1994. (accessed September 2005).

(16.) Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM. Electrospray ionization for mass-spectrometry of large biomolecules. Science 1989;246:64-71.

(17.) Swiss-Prot Protein Sequence Database. TNNT2 Human. (accessed September 2005).

(18.) Bunk DM, Dalluge JJ, Welch MJ. Heterogeneity in human cardiac troponin I standards. Anal Biochem 2000;284:191-200.

(19.) McDonough JL, Arrell DK, Van Eyk JE. Troponin I degradation and covalent complex formation accompanies myocardial ischemia/ reperfusion injury. Circ Res 1999;84:9-20.

(20.) Gusev NB, Barskaya NV, Verin AD, Duzhenkova IV, Khuchua ZA, Zheltova A0. Some properties of cardiac troponin T structure. Biochem J 1983;213:123-9.

(21.) Wilkinson JM, Grand RJA. Amino acid sequence of chicken fast skeletal muscle troponin I. Eur J Biochem 1978;82:493-501.

(22.) Kobayashi T, Takagi T, Konishi K, Morimoto S, Ohtsuki I. Amino acid sequence of porcine cardiac muscle troponin C. J Biochem 1989;106:55-9.

(23.) Cottrell JS. Protein identification by peptide mass fingerprinting. Pept Res 1994;7:115-24.

(24.) Swiss-Prot Protein Sequence Database. ACTC_Human. (accessed September 2005).

(25.) Birken S, Berger P, Bidart JM, Weber M, Bristow A, Norman R, et al. Preparation and characterization of new WHO reference reagents for human chorionic gonadotropin and metabolites. Clin Chem 2003;49:144-54.

(26.) Katrukha AG, Bereznikova AV, Esakova TV, Pettersson K, Lovgren T, Severina ME. Troponin I is released in bloodstream of patients with acute myocardial infarction not in free form but as complex. Clin Chem 1997;43:1379-85.

(27.) Katrukha A, Bereznikova A, Filatov V, Esakova T. Biochemical factors influencing measurement of cardiac troponin I in serum. Clin Chem Lab Med 1999;37:1091-5.

(28.) Labugger R, Organ L, Collier C, Atar D, Van Eyk JE. Extensive troponin I and T modification detected in serum from patients with acute myocardial infarction. Circulation 2000;102:1221-6.

(29.) International Organization for Standardization. In vitro diagnostic medical devices: measurement of quantities in samples of biological origin-description of reference materials. ISO 15194: 2002. Geneva, Switzerland: ISO, 2005 Measurements.PublicationList? CLASSIFICATION=MEASUREMENTS (accessed September 2005).

(30.) Magari RT, Murphy KP, Fernandez T. Accelerated stability model for predicting shelf-life. J Clin Lab Anal 2002;16:221-6.

(31.) Magari RT. Estimating degradation in real time and accelerated stability tests with random lot-to-lot variation: a simulation study. J Pharm Sci 2002;91:893-9.

(1) Nonstandard abbreviations: cTnI, cTnT, and cTnC, cardiac troponin I, T, and C, respecfively; SRM, Standard Reference Material; SI, International System oi Units; RPLC, reversed-phase liquid chromatography; TFA, tritluoroacetic acid; UV, ultraviolet; LC-MS, liquid chromatography-mass spectrometry; and MALDI, matrix-assisted laser desorpfion/ionization.


Analytical Chemistry Division, Nafional Insfitute oi Standards and Technology, Gaithersburg, MD 20899.

* Author for correspondence. Fax 301-977-0685; e-mail
Table 1. Amino acid analysis results for the cTnI concentration
in SRM 2921 based on the amino acids Ala, Val, and Leu.

 Measured cTnI concentration (mg/L) from

Day Sample Ala Val Leu

1 1 29.61 32.44 30.93
 2 31.30 34.75 32.58
2 1 30.43 33.26 31.97
 2 29.97 33.00 33.06
3 1 29.55 30.89 29.86
 2 27.69 28.90 28.27
4 1 28.00 30.84 27.87
 2 28.84 29.83 29.99
Mean cTnI, 29.42 (1.67) 31.74 (3.01) 30.57 (2.93)
Combined mean 30.65 (2.37)
 cTnI, mg/L

(a) The uncertainties in the mean and combined mean values
are expressed at the 95% confidence level.

Table 2. Relative molecular masses observed in the LC-MS
and MALDI-MS analyses of SRM 2921.

 Observed [M.sub.r] (b)

Component [M.sun.r] (a) LC-MS MALDI-MS

 cTnT 34 459.0 33 775 (8) 33 771 (20)
 34 374 (8)
 34 503 (6) 34 501 (15)
 34 524 (8)
 cTnI 23 918.4 23 427 (8) 23 424 (20)
 23 556 (6) 23 552 (15)
 23 637 (8) 23 636 (20)
 23 704 (8) 23 700 (15)
 23 785 (8) 23 789 (20)
 23 918 (5) 23 920 (15)
 23 999 (6)
 24 079 (8) 24 079 (20)
 cTnC 18 444.6 18 426 (6)
 18 445 (4) 18 440 (10)
 18 463 (8)
 18 477 (8)

Component Tentative identification

 cTnT cTnT(Ac1-281) (c)
 N-Acetylated cTnT [cTnT(Ac1-287)]

 cTnI cTnI(Ac1-205)
 Monophosphorylated cTnI(Ac1-206)
 Monophosphorylated cTnI(Ac1-207)
 N-Acetylated cTnI [cTnI(Ac1-209)]
 Monophosphorylated cTnI(Ac1-209)
 Bisphosphorylated cTnI(Ac1-209)
 cTnC N-Acetylated cTnC [cTnC(Ac1-161)]

(a) Calculated from the amino acid sequence
obtained from the Swiss-Prot protein sequence
database (
using entry names TNNT2_HUMAN for cTnT, TNNI3_HUMAN
for cTnI, and TNNC1_HUMAN for cTnC.

(b) The component with the highest observed signal
intensity is indicated in bold. The values in
parentheses are the uncertainties calculated at
the 95% confidence level.

(c) Ac, acetylated.
COPYRIGHT 2006 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Proteomics and Protein Markers
Author:Bunk, David M.; Welch, Michael J.
Publication:Clinical Chemistry
Date:Feb 1, 2006
Previous Article:Detection of alternatively spliced transcripts in leukemia cell lines by minisequencing on microarrays.
Next Article:Western blotting of total lysate of Helicobacter pylori in cases of atrophic body gastritis.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters