Reproducibility of antigen-immobilized enzyme-linked immunosorbent assay (ELISA) and sandwich ELISA for quantitative detection of NNV particles.

Nervous necrosis virus (NNV) is a fish virus belonging to family Nodaviridae. In this study, we prepared partially aggregated and monometric NNV particles to determine reproducibility of two different enzyme-linked immunosorbent assays (ELISAs): antigen-immobilized ELISA and sandwich ELISA. Passing ratios of purified NNV particles through ultrafilters with molecular weight cut off (MWCO) of 10, 3 × 10 and 10 were 0%, 35.2% and 80.3%, respectively, suggesting that purified NNV particles were partially aggregated whereas those in filtrates with MWCO of 3 × 10 could be monometric. Both NNV particles were subjected to ELISAs. Reduction ratios of ELISA values by 2-fold dilution of antigens were 50% in sandwich ELISA regardless of aggregation state of NNV particles. In contrast, those in antigen-immobilized ELISA were 42% (partially aggregated NNV) to 43% (monometric NNV), which were lower than the theoretical value (50%). This could be due to changes in aggregation state of NNV particles during dry-immobilization. Sandwich ELISA has excellent reproducibility from five times of experiments, in comparison with antigen-immobilized ELISA. Furthermore, available range of regression lines (R > 0.99) in sandwich ELISA was wider than that in antigen-immobilized ELISA. These results revealed that sandwich ELISA had better quantitativeness, reproducibility and available range of ELISA values than antigen-immobilized ELISA.