An enzyme-linked immunosorbent assay (ELISA) was employed to investigate the expression of influenza A/Hong Kong/68 (H3N2) virus structural proteins on the surface of infected MDCK cells, and to detect viral antigens in culture media and cell extracts. Infected cells were fixed with 0.1 per cent glutaraldehyde before being examined for the presence of cell-surface antigens. Viral antigens were first observed on the surface of cells 4 hours after infection and reached a maximum 10–12 hours after infection, when measured by haemadsorption with chicken erythrocytes and by ELISA and immunofluorescence with hyperimmune antiserum to Hong Kong virus. A good correlation was found between the three assay systems. The presence of individual virion structural proteins on the cell surface was determined by ELISA using specific antibodies purified by differential affinity chromatography. Either or both of the internal matrix and nucleoprotein antigens were expressed from 2 to 6 hours after infection, with maximum expression after 2 hours, and the strain-specific and common antigenic determinants of haemagglutinin were observed on the cell surface from 4 hours after infection, and reached a maximum 8 to 10 hours after infection. Low levels of neuraminidase were detected between 4 and 8 hours after infection. Culture media and cell extracts were titrated by infectivity and haemagglutination assays, and by ELISA. Titres obtained from the culture media showed a close correlation between the three assay methods, with peak titres being attained 24 hours after infection. Viral antigens were first observed in cell extracts by ELISA 4 hours after infection, and infectious virions and haemagglutinin 2 hours later, but whereas maximum titres of infectious virus and haemagglutinin were found 10 hours after infection, the ELISA titre continued to rise until 24 hours after infection, which suggested that virus structural proteins were being accumulated in the cells after most of the progeny virions had been released. The results are discussed in terms of the potential use of ELISA in rapid virus diagnosis.
The results of comparative assays on cell extracts harvested at various times after infection indicated that virion structural proteins continue to be accumulated in the cell even after most progeny particles have matured. Moreover, the high correlation coefficients between infectivity, haemagglutination and ELISA of virus particles and antigens released into the culture media strongly suggest that only complete virus particles are released, and very little leakage of individual antigens occurred, even from dead cells.
ELISA has also been successfully employed to detect other cell-associated viral antigens, including rabies virus in brain tissue (3) and feline oncornavirus-associated cell membrane antigen in FL74 cells (22). In the latter study, the ELISA titres showed a close correlation in sensitivity with indirect fluorescence tests, but had the added advantage of being quantitative rather than qualitative. The results presented in this report, together with those using other viruses (3, 19, 22) suggest that ELISA is a potentially useful technique for the detection of specific virus infections in tissue culture, and as a tool for rapid virus diagnosis. Problems have been encountered using peroxidase rather than alkaline phosphatase as the enzyme in ELISA due to endogenous peroxidase in clinical material (13), but no evidence of non-specific staining was observed in this study, probably because the cells had been fixed initially with glutaraldehyde. We believe, therefore, that this study has demonstrated that ELISA can be employed as a rapid technique to detect and to distinguish subtype specificity of influenza A virus in tissue culture, and can potentially be applied to other viruses with cell surface-associated antigens. Moreover, the use of glutaraldehyde-fixation allows considerable flexibility in the timing of the assay and may prove suitable for viral diagnosis at a distance.