Indexed on: 19 Oct '04Published on: 19 Oct '04Published in: Journal of Virological Methods
Three assay methods were investigated for monitoring the time-course of neuraminidase (NA) activity of tissue culture derived equine influenza A virus from large-scale microcarrier cultivation and several steps of downstream processing required for the production of inactivated vaccines. Measurements of neuraminidase activity by a thiobarbituric acid (TBA) and a fluorometric method using Amplex Red as a fluorogen (FL-AR) did not correlate with the increase of hemagglutinin (HA) during virus replication. Samples analysed by the TBA method showed unspecific interference from low molecular weight compounds (< 3 kDa) of cell growth medium and virus maintenance medium. Further investigations showed that this was probably caused by interfering reactions between reducing sugars and amino acids that can be overcome by dialysis of samples. On the other hand, the sensitivity of the FL-AR method was not sufficient for the required measuring range. However, a reliable and sensitive fluorometric assay method (FL-MU-NANA) was obtained using 4-methylumbelliferyl-alpha-d-N-acetylneuraminic acid (4-MU-NANA) as a substrate, which allowed the detection of neuraminidase activities as low as 0.09 mU/mL. In this assay, time-course of neuraminidase activities correlated well with increasing hemagglutinin activities during virus replication in a bioreactor. Analysis of samples from various downstream processing steps comprising of clarification, inactivation, ultrafiltration (UF) and size-exclusion chromatography for the purification of influenza virus showed that neuraminidase activity was preserved at comparatively high levels. Based on the hemagglutinin and neuraminidase activity of the clarified and inactivated virus harvest, the overall recovery after gel filtration was about 34.4% and 119.5%, respectively.