Search

We have developed a fibrinogen-specific sandwich enzyme-linked immunosorbent assay (ELISA) microarray assay for use in qualitatively distinguishing between blood plasma and serum samples. Three capture antibodies (49D2, HPA001900, and F8512) were evaluated in conjunction with 1D6 as the detection antibody. The data show that 49D2 and (to a lesser extent) F8512 successfully identify previously unknown plasma and serum samples based on approximately a 28-fold difference in signal intensity between the sample types. This assay has utility in rapidly identifying previously archived clinical samples with incomplete annotation in a high-throughput manner prior to proteomic analyses., (Copyright © 2011 Elsevier Inc. All rights reserved.)

Sandwich ELISA microarrays have great potential for validating disease biomarkers. Each ELISA relies on robust-affinity reagents that retain activity when immobilized on a solid surface or when labeled for detection. Single-chain antibodies (scFv) are affinity reagents that have greater potential for high-throughput production than traditional IgG. Unfortunately, scFv are typically less active than IgG following immobilization on a solid surface and not always suitable for use in sandwich ELISAs. We therefore investigated different immobilization strategies and scFv constructs to determine a more robust strategy for using scFv as ELISA reagents. Two promising strategies emerged from these studies: (i) the precapture of epitope-tagged scFv using an antiepitope antibody and (ii) the direct printing of a thioredoxin (TRX)/scFv fusion protein on glass slides. Both strategies improved the stability of immobilized scFv and increased the sensitivity of the scFv ELISA microarray assays, although the antiepitope precapture method introduced a risk of reagent transfer. Using the direct printing method, we show that scFv against prostate-specific antigen (PSA) are highly specific when tested against 21 different IgG-based assays. In addition, the scFv microarray PSA assay gave comparable quantitative results (R(2) = 0.95) to a commercial 96-well ELISA when tested using human serum samples. In addition, we find that TRX-scFv fusions against epidermal growth factor and toxin X have good LOD. Overall, these results suggest that minor modifications of the scFv construct are sufficient to produce reagents that are suitable for use in multiplex assay systems.

Making sound proteomic inferences using ELISA microarray assay requires both an accurate prediction of protein concentration and a credible estimate of its error. We present a method using monotonic spline statistical models (MS), penalized constrained least squares fitting (PCLS) and Monte Carlo simulation (MC) to predict ELISA microarray protein concentrations and estimate their prediction errors. We contrast the MSMC (monotone spline Monte Carlo) method with a LNLS (logistic nonlinear least squares) method using simulated and real ELISA microarray data sets.MSMC rendered good fits in almost all tests, including those with left and/or right clipped standard curves. MS predictions were nominally more accurate; especially at the extremes of the prediction curve. MC provided credible asymmetric prediction intervals for both MS and LN fits that were superior to LNLS propagation-of-error intervals in achieving the target statistical confidence. MSMC was more reliable when automated prediction across simultaneous assays was applied routinely with minimal user guidance.