Viruses are completely dependent on their hosts for gene expression, especially protein synthesis. Through various strategies, viruses hijack the host cell translation machinery to recruit it for their own needs and co-currently block the cell’s anti-viral response. In the case of SARS-CoV-2, suppression of the interferon response has emerged as a major clinical determinant of pathogenicity. It is therefore critical to understand the strategies used by this virus to suppress cellular gene expression.

With the break of the COVID-19 pandemic, and while work in the lab was slowed down by lockdowns and stress, as virologists, we felt a strong drive to use our knowledge and experience working with other viruses to contribute to the global efforts of understanding this new and deadly virus. Specifically, we had worked in the past on analyzing host and viral gene expression in cells infected with herpesviruses or influenza, so we knew applying our methods on SARS-CoV-2 could highlight important biology.

Under the leadership of our PI, Noam Stern-Ginossar, and powered by a close collaboration with the Israeli Institute of Biological research (IIBR), we initiated a collective team effort and approached the challenge of studying SARS-CoV-2 biology. Our goal was to provide a better understanding of the mechanisms underlaying SARS-CoV-2 induced host shutoff. Taking part in this project has been an exceptional experience. The work was a fast-paced back-and-forth collaboration within the team. A large part of the time was spent alone, each in their own home office, while we stayed in contact through frequent virtual meetings. There was a lot of pressure from many directions, and it was not an easy time, but it felt gratifying to have something to contribute.

As a start, we aimed to track changes in host and viral gene expression along infection. To this end we used a combination of RNA-sequencing and ribosome profiling, which together give a snapshot of the expression level of each mRNA and the extent to which it is being translated. Combining these, we could calculate the relative translation efficiency of each gene, viral and cellular. Unexpectedly, we found that while viral mRNAs take over the mRNA pool in the cell, amounting to almost 80% of mRNA molecules, they are not translated more efficiently than cellular mRNAs. This suggests that rather than engaging uniquely with ribosomes and “stealing” them away from cellular mRNAs, the virus overwhelms the cell with viral transcripts and takes over by sheer numbers.

To better understand this observation, we quantified gene expression at an absolute level by taking global measurements of total RNA amounts and of total translation in the cells. These measurements revealed a drastic 70% reduction of translation in SARS-CoV-2 infected cells relative to uninfected cells. On the other hand, while total RNA amount in the cells remained constant throughout infection (mostly dominated by ribosomal RNAs), the levels of mRNA were increasing. Notably, this increase was driven by accumulation of viral mRNAs, while cellular mRNAs were, in fact, diminishing. This led us to suspect that infection induces massive degradation of host mRNA, a hypothesis we confirmed by directly measuring mRNA half-life; the stability of cellular mRNAs was greatly reduced in SARS-CoV-2 infected cells compared to uninfected cells. Looking more closely at which mRNAs are being degraded, we found a correlation between degradation and sub-cellular localization; transcripts that are mostly localized to the cytosol are more affected by the degradation induced by infection, and nuclear and mitochondrial mRNAs were protected from degradation.

Up to this point, we described changes happening at the RNA stability and global translation levels during infection. To get a more comprehensive picture, we asked whether specific translational control of cellular genes plays a role during infection. We used the calculated translation efficiency of host mRNAs at different times along infection to identify a group of genes whose translation became less and less efficient as infection progresses. The temporal profile of these genes revealed a clear signature; their mRNA levels were increasing in the absence of an equivalent increase at the translation level. As it turned out, these genes were strongly enriched for innate immune response pathways. Taken together, the results suggest that innate immune genes are transcriptionally induced during infection, and that the virus interferes with their ability to engage with ribosomes.

Our leading hypothesis on this point is that SARS-CoV-2 interrupts nuclear export of cellular transcripts, thus preventing newly transcribed mRNAs from meeting ribosomes at the cytosol and from being translated. To get more evidence on this, we performed RNA-seq on the cytosolic and the nuclear fractions separately. We could see that the increase in innate immune genes is specific to the nuclear fraction, supporting an export defect. However, interpretation of this result is complicated because most cellular mRNAs are depleted from the cytosol due to the degradation induced by the virus. Therefore, the point of whether SARS-CoV-2 inhibits translation through nuclear export inhibition remains to be settled, possibly by more direct measures.

Overall, our study allowed us to take an in-depth look into how SARS-CoV-2 expertly interferes with cellular gene expression, leading to the shutdown of host protein production through a multipronged strategy. Beyond the results described here, we found some interesting features of viral genes translation. You can read more in our published manuscript, here.