Removal of inhibitory effects in a serum cardiac troponin I immunoassay.

The measurement of cardiac troponin I (cTnI) has become the gold standard for the clinical diagnosis of myocardial infarction (1 ). Varieties of commercial sandwich-type immunoassays are used for the measurement of cTnI concentration in human serum or plasma. Several of these assays use pairs of monoclonal antihuman cTnI antibodies with high specificities for the invariant part of the cTnI molecule (amino acid residues 30 –110), in accordance with the recommended guidelines (2 ). These assays are often based on magnetic beads or latex particles to take advantage of the ease of bead/particle washing, which can minimize assay interference by matrix proteins. Signal amplification by means of chemiluminescence or electrochemiluminescence allows the detection of serum cTnI at concentrations 0.05 g/L. As a part of the recent effort of the IFCC Working Group for Standardization of Troponin I to develop a cTnI reference measurement system (1 ), we have used various measurement techniques to investigate binding affinities between 6 monoclonal antibodies (mAbs) obtained from HyTest (each serving as either the capture or the detection antibody) and cTnI, in the form of either a reference material (NIST SRM 2921) or a cTnI-positive serum pool (PS). One of the techniques is the multiplexed bead array (3 ), which resembles the commercial beador particle-based immunoassays. The capture mAb molecules are covalently immobilized on different bead populations to capture cTnI. The bound cTnI is recognized by the biotinylated detection antibody, which is then bound to a fluorescent reporter, streptavidin– phycoerythrin (SA-PE). Although commercial cTnI assays show higher detection sensitivities, the ease of the assay multiplexing and the low cost of the instrumentation make the bead-array platform advantageous for optimization of antibody pairs for cTnI detection. It is well known that a human serum matrix can introduce inhibitory effects that cause immunoassays to report lower concentrations of cTnI. These inhibitory effects could be due to the presence of heterophilic antibodies and/or autoantibodies in the serum (4 ). Finding the optimal conditions to minimize the inhibitory effects and enhance cTnI detection is crucial to obtain the most robust and accurate immunoassay results. Therefore, bead arrays were used in the testing of 18 different conditions, including depletion of serum samples before analysis with commercial antibody-based multiprotein depletion columns, magnetic beads coupled with protein G or protein A, and carboxylated beads covalently coupled to antimouse IgG. In addition, various antihuman antibodies to serum protein (IgG, IgA, IgM, complement component C3), mouse IgG, detergent (Triton X-100), buffer solutions with different pH values (4.0 –7.4), EDTA, and a combination of both buffer solution and EDTA were added separately to serum samples before the cTnI-capture step. For both SRM 2921 spiked into normal (nonpathologic) human serum (NS) and a PS, the addition of both EDTA and a buffer solution at pH 5.2 to the sample substantially diminished the inhibitory effect of the matrix on bead assays performed with different antibody pairs. An optimized bead-array assay protocol is as follows: Initial addition of 12.5 L of a solution of 15 mmol/L EDTA in sodium acetate buffer (0.1 mol/L), pH 5.2, is followed by addition of 10 L NS, 7.5 L SRM 2921 spiked into PBS blocking buffer [BB in the letter refers to 0.01 mol/L PBS (0.138 mol/L NaCl, 0.0027 mol/L KCl), pH 7.4; 1% BSA; 0.05% NaN3.], and 10 L BB. After beads (14 L) are added and incubated for 30 min, 10 L biotinylated mAb (8.1 10 g/L) is added. After 30 min, 10 L SA-PE (1.0 10 g/L) is added. The left panel of Fig. 1 shows that for SRM 2921 spiked into NS, the addition of 12.5 L of acetate-EDTA solution to a total sample volume of 40 L before the addition of the capture mAb– coated beads produces a significant increase in assay response ( ), compared with the same dilution of the serum in BB (Œ). This increase in signal was also observed for PS (right panel). The addition of this buffer component to the serum samples increases the sensitivity of cTnI detection and enables more reliable measurement of cTnI in serum. Considering that the strong interaction between cTnI and troponin C (TnC) is Ca dependent (5 ) and that the high positive charge of cTnI (pI 9.87) increases interactions with negatively charged molecules such as TnC (pI 4.05), our result suggests that cTnI in both SRM 2921 and PS is largely associated with TnC, because both a low buffer pH and EDTA likely help to reduce the strong interaction between cTnI and TnC and thereby allow binding of the capture mAb to cTnI. Although this approach has not been evaluated on com1 Nonstandard abbreviations: cTnI, cardiac troponin I; mAb, monoclonal antibody; PS, cTnI-positive serum pool; SA-PE, streptavidin–phycoerythrin; NS, normal (nonpathologic) human serum; BB, PBS blocking buffer; TnC, troponin C. Clinical Chemistry 55:11 2055–2063 (2009) Letters to the Editor