Adaptable Detection Strategies in Membrane-Based Immunoassays: Calibration-Free Quantitation with Surface-Enhanced Raman Scattering Readout.

This paper presents a method for immunometric biomarker quantitation that uses standard flow-through assay reagents and obviates the need for constructing a calibration curve. The approach relies on a nitrocellulose immunoassay substrate with multiple physical addresses for analyte capture, each modified with different amounts of an analyte-specific capture antibody. As such, each address generates a distinctly different readout signal that is proportional to the analyte concentration in the sample. To establish the feasibility of this concept, equations derived from antibody-antigen binding equilibrium were first applied in modeling experiments. Next, nitrocellulose membranes with multiple capture antibody addresses were fabricated for detection of a model analyte, human Immunoglobulin G (hIgG), by a heterogeneous sandwich immunoassay using antibody-modified gold nanoparticles (AuNPs) as the immunolabel. Counting the number of colored capture addresses visible to the unassisted eye enabled semiquantitative hIgG determination. We then demonstrated that, by leveraging the localized surface plasmon resonance of the AuNPs, surface-enhanced Raman spectroscopy (SERS) can be used for quantitative readout. By comparing the SERS signal intensities from each capture address with values predicted using immunoassay equilibrium theory, the concentration of hIgG can be determined (∼30% average absolute deviation) without reference to a calibration curve. This work also demonstrates the ability to manipulate the dynamic range of the assay over ∼4 orders of magnitude (from 2 ng mL to 10 μg mL). The potential prospects in applying this concept to point-of-need diagnostics are also discussed.