Cytokine detection and simultaneous assessment of rheumatoid factor interference in human serum and synovial fluid using high-sensitivity protein arrays on plasmonic gold chips

Abstract

Background: Fluorescence-enhancing microarray on plasmonic gold film is an attractive alternative to traditional enzyme-linked immunosorbent assay (ELISA) for cytokine detection because of the increased sensitivity. The assay chemistry is similar to an ELISA sandwich assay, but owing to the gold substrate, cytokine measurements are 10 to 100 times more sensitive and can be multiplexed. Plasmonic protein microarrays are, as other immunoassays, affected by the presence of heterophilic antibodies and rheumatoid factor may lead to analytical errors with serious implications for patient care. Here, we present a plasmonic gold substrate protein microarray for high-sensitivity detection of cytokines with simultaneous assessment of rheumatoid factor interference on a single chip. Results: Paired serum and synovial fluid samples from patients with rheumatoid arthritis (n = 18), osteoarthritis (n = 9) or healthy controls (n = 10) were arrayed on near-infrared fluorescence enhancing plasmonic gold chips spotted with cytokine-specific capture antibody and isotype control antibody. Possible rheumatoid factor interference was visualised by a non-specific signal from the isotype control antibody, and pre-treatment of samples with heat-aggregated animal IgG eliminated this background contamination. The platform was optimised using the cytokine IL-20. The protein microarray platform allowed for the detection of human IL-20 at levels <1 pg/ml with reliable IL-20 quantification over a 5-log dynamic range. Samples for which rheumatoid factor caused artefacts were identified and a method for eliminating rheumatoid factor interference was developed and validated. IL-20 protein levels were significantly higher in synovial fluid samples from patients with rheumatoid arthritis compared to osteoarthritis (p < 0.001), while serum levels of IL-20 did not differ between patients with rheumatoid arthritis, osteoarthritis or healthy controls. Conclusion: Using novel plasmonic gold chips, we developed a highly sensitive and accurate assay platform to detect lowly expressed cytokines in biological fluids, allowing for the elimination of rheumatoid factor interference in as little as 5 μl sample volume. The detection limit was below 1 pg/ml for IL-20 and linearity was achieved over a 5-log dynamic range. This technology is highly advantageous for cytokines where sensitivity or sample volume is critical or where assessment of rheumatoid factor interference needs addressed and eliminated.