DYNAMICS OF VIRAL INFECTION IN THE OCEAN AT THE SINGLE-CELL LEVEL

Investigating host-pathogen interactions at the whole-population level tends to mask cell-to-cell variability, hampering the ability to discern between diverse phenotypes. To overcome this limitation, we applied single-cell dual RNA-seq to study the dynamic transcriptomes of a cosmopolitan bloom-forming alga, Emiliania huxleyi, and its specific large DNA virus EhV. Gene expression was profiled for multiple genomes (nuclear, plastid, mitochondrial and viral) using the massively parallel RNA single-cell sequencing method for cells sorted from infected cultures during a time course. Striking heterogeneity in the overall viral transcription was observed among individual cells from all samples, with cells exhibiting either high or very low viral transcript numbers, but very few at intermediate levels. The cells with high viral expression formed subpopulations characterized by clusters of strongly co-expressed viral genes. Though distinct in transcription profiles, almost every subpopulation was represented at all time points, with their relative proportions shifting over time. By ordering the cells based on the progression of infection states, we provide a pseudo-temporal scale on which viral and host expression could be compared. As the infection progressed, the viral genes were successively expressed in five partially overlapping kinetic classes that end with constitutive expression of the virion structural proteins. Furthermore, the infection states could be linked to host genes related to translation, organelle functions, cell fate regulation and lipid biosynthesis. Overall, this study shows the dynamic changes in within-population heterogeneity during viral infection at the single-cell resolution, sheds light on the strategies employed during marine host-virus arms races, and opens up the possibility of characterizing single-cell infection states and rare cell types in the field.