Forty years with coronaviruses

Weiss, Susan R.

Sequencing RNA genomes was difficult in the 1980s, and cDNA cloning was just emerging. The first complete genome sequence of a coronavirus was for infectious bronchitis virus in 1987 (Boursnell et al., 1987 ) and a few years later it was completed for MHV (Lee et al., 1991) . These genomes were assembled from many short cDNA clones and, when completed, indicated that the genome was ∼30 kb, significantly longer than had been estimated previously by sucrose gradient centrifugation, and that it was the longest of any known RNA virus. These sequences revealed two long open reading frames, ORF1a and 1b encoding 16 nonstructural proteins. It was also shown that both ORF1a and ORF1ab proteins were translated from genome RNA, and ORF1b via a translational frame shift at the end of ORF1a revealing new mechanisms of translational control (Bredenbeek et al., 1990) . Alexander Gorbalenya's insightful analyses of the proteins encoded in ORFs1a and 1b revealed several protease and other enzymatic domains (for example, ExoN, EndoU, and methylation enzymes) required for diverse aspects of replication and immune evasion. After the emergence of SARS-CoV, it became clear that there were some enzymes and proteins that were present in diverse coronaviruses and that there were other so-called accessory proteins that were species specific (Snijder et al., 2003) . Understanding the roles of these virally encoded proteins will undoubtedly reveal how these viruses evade host-directed responses. Moreover, antivirals targeting one of the essential enzymes conserved across the coronavirus family may ultimately be developed as a pan anti-coronavirus therapy. Indeed, current trials with the antiviral remdesivir, which targets the conserved RNA-dependent RNA polymerase, may indeed be active against a large number of coronaviruses (Agostini et al., 2018) . Indeed, it became clear that these the coronavirus accessory genes played a fundamental role in innate immune evasion and pathogenesis. We found that a poorly understood MHV protein, NS2, was a liverspecific antagonist of host innate immunity, and with Robert Silverman, now a longterm collaborator and friend, we found that NS2 was a phosphodiesterase that cleaved 2-5A, the activator of the antiviral OAS-RNase L pathway (Zhao et al., 2012) . This new research direction led to our two groups exploring several aspects of activation and antagonism of this pathway during infections with many viruses, as well as by endogenous double-stranded RNA that accumulates in certain autoimmune disease states. Importantly, we identified OAS-RNase L as a primary pathway that can be activated early in infection or during infections with viruses that shut down IFN production or signaling, and in bat cells as well as human cells (Li et al., 2019) . During this period, we found that MERS NS4b was a structural homologue of NS2 but was different from other viral phosphodiesterases in its primarily nuclear localization, which is a current direction of our research. This new and still ongoing research direction led to our discovery of new aspects of innate immune regulation by MERS-CoV (Comar et al., 2019) .