Date Published: December 6, 2016
Publisher: Public Library of Science
Author(s): Efrat M. Cohen, Oren Kobiler, Edward Mocarski.
Viral gene expression varies significantly among genetically identical cells. The sources of these variations are not well understood and have been suggested to involve both deterministic host differences and stochastic viral host interactions. For herpesviruses, only a limited number of incoming viral genomes initiate expression and replication in each infected cell. To elucidate the effect of this limited number of productively infecting genomes on viral gene expression in single cells, we constructed a set of fluorescence-expressing genetically tagged herpes recombinants. The number of different barcodes originating from a single cell is a good representative of the number of incoming viral genomes replicating (NOIVGR) in that cell. We identified a positive correlation between the NOIVGR and viral gene expression, as measured by the fluorescent protein expressed from the viral genome. This correlation was identified in three distinct cell-types, although the average NOIVGR per cell differed among these cell-types. Among clonal single cells, high housekeeping gene expression levels are not supportive of high viral gene expression, suggesting specific host determinants effecting viral infection. We developed a model to predict NOIVGR from cellular parameters, which supports the notion that viral gene expression is tightly linked to the NOIVGR in single-cells. Our results support the hypothesis that the stochastic nature of viral infection and host cell determinants contribute together to the variability observed among infected cells.
Cell-to-cell variability is an important factor in cancer, development, evolution, host-pathogen interactions and other key biological processes [1–3]. The variability observed among single cells mainly arises from deterministic factors, i.e. preexisting molecular regulatory mechanisms [3, 4]. In the context of viral infections, it was suggested that stochastic interactions between a virus and individual host cells could contribute to variability in the outcome of infection in the entire infected organism [5–9]. Much of the variation in the outcome of infection can be attributed to the specific cell state prior to infection . Here, we provide evidence that the viral gene expression level also depends on the actual number of viral genomes initiating the infection process.
Cell-to-cell variations during viral infection contribute to significant differences in the outcomes of infection [36–39]. To study sources of single cell diversity during herpesvirus infection, we constructed fourteen barcoded, fluorescence-expressing viral recombinants. The expressed fluorescence from infected cells was found to be a surrogate for viral gene expression. We compared fluorescence levels in individual cells to the number of incoming viral genomes replicating (NOIVGR) in these cells. In the three cell types we examined, the NOIVGR correlated with fluorescence expressed from the viral genome, even though differences in the mean NOIVGR among the cell types were observed. Unexpectedly, our results suggest that cells with high basal gene expression levels are less permissive for viral infection. We also developed multiple linear regression models that predict the NOIVGR in a single cell according to cellular parameters, suggesting that the correlations we observed can be extrapolated to other experimental systems.