Development of a flow cytometry-based assay for the quantification of active viral particles

In the present study two methods for the quantification of influenza viruses using flow cytometry have been developed. Common methods for virus titre quantification, the so-called infectivity assays, are based on the detection of cytopathic effects visible in cell populations which have been infected. These methods comprise the application of serial virus dilutions to cell monolayers, which requires an intensive manual labor and a microscopic read-out. Given that, the typical methods can be time and labor consuming – around 4 days depending on the virus type – and very susceptible to subjectivities. To overcome subjectivity, several flow cytometric methods have been designed which also determine the quantity of cells infected by a series of virus dilutions. With the procedure of the typical infectivity assay “tissue culture infective dose” (TCID50) and theoretical data from reported flow cytometry assays as a starting point, the first part of the work was to determine the time at which the infection should be stopped and measured. To do so, serial dilutions of a standard sample (influenza strain H1N1 A/PR/8/34 with known TCID50 titre) was used to infect adherent MDCK cells. The adequate infection time was considered as the moment at which a linear relationship between the proportion of infected cells and the concentration of active virus particles occurred. This linear correlation was found at 18 hours post infection. In order to inhibit or at least reduce multiple infection rounds which could lead to an overestimation of the titre, two methods were developed afterwards: one using zanamivir and the other using a low infection volume and the application of 15% serum to block the infection. For the first one, a linear correlation between the proportion of infected cells and the virus concentration was found at 24 hpi. For the case of reduced volume and 15% of serum, this linear correlation was first observed at 18 hpi, although it is likely to appear at shorter times. Generally, the non-controlled and the controlled methods offered an accurate calculation of titres for samples within a wide range of concentrations. A limit of detection of ~103 IU/mL was observed for the non-controlled procedure, whereas the controlled one showed a limit above 104 IU/mL. Moreover, both methods were shown to be adequate for the monitoring of virus titres of not only the influenza reference strain A/PR/8/34, but also of the cold adapted influenza strain H1N1 A/Singapore/1/57 ca produced in a bioreactor. Overall, the transfer from the TCID50 assay to a procedure applicable to flow cytometry was successful. The assay time could be reduced by 1 day with respect to the typical TCID50 assay which takes around 4 days. The biggest achievement, however, is the reduction of subjectivity for the read-out as this has successfully been transferred to a device and the lower standard deviations and thus higher reliability of titres obtained with the assay