“It’s coming. We know it’s coming.”
That was Nevan Krogan’s blunt reminder that SARS-CoV-2 is not the first virus to wreak temporary havoc on humanity, nor will it be the last.
Indeed, SARS-CoV-2 is the third coronavirus this century to set off public health alarm bells, suggesting that coronaviruses as a whole, for reasons that are not yet known, are transforming from mostly harmless agents to major public health threats.
Krogan, who is the director of the Quantitative Biosciences Institute (QBI) at UC San Francisco and a senior investigator at Gladstone Institutes, is part of an international team of researchers that has been making the best of that transition by finding shared vulnerabilities of SARS-CoV-1, MERS-CoV and SARS-CoV-2, in the hope that the findings will prove helpful in fighting not just the current pandemic, but a possible future SARS-CoV-3 outbreak as well.
The team reported the latest results of that approach in the Oct. 15, 2020, online issue of Science.
In previous papers, the team had provided a comprehensive blueprint of the protein interactions between SARS-CoV-2 and infected host cells, and focused specifically on kinases that might provide druggable targets.
In their newest publication, the researchers combined proteomics and virology data to identify which viral-host protein interactions were common to SARS-CoV-2 and one or both of the other coronaviruses.
CRISPR-based screening, conducted in part by biotechnology company Synthego Corp., identified host proteins that affected the virus’ ability to grow. One of those proteins was the mitochondrial protein Tom70, a shuttle that moves other proteins into and out of mitochondria. Tom70 physically interacted with Orf9b in both SARS-CoV-1 and SARS-CoV-2.
The functional consequences of that binding for how infection progresses remain to be fully worked out. But one thing that became clear when the team conducted structural studies was that Orf9b, which is the only SARS-CoV-2 protein that homes to the host mitochondria, underwent a striking shape change when bound to Tom70. The difference between bound and unbound versions of Orf9b was so extreme that at a press conference highlighting key findings from the paper, Krogan argued that for targeting purposes, the bound and unbound versions might as well be two separate proteins.
In their paper, the team also used real-world evidence to search for existing drugs that may affect the clinical course of COVID-19 infections. Collaborators from Aetion Inc. used the company’s database of the clinical records of roughly 740,000 U.S. COVID-19 patients – close to 10% of total diagnosed U.S. cases – to look for medicines that were associated with better, or worse, outcomes.
Findings from that analysis suggested that the marketed nonsteroidal anti-inflammatory drug indomethacin, which inhibits prostaglandin E synthase 2 (PGES2), reduced the risk of hospitalization and the need for inpatient services, while typical antipsychotics such as haloperidol and chlorpromazine reduced the risk of needing mechanical ventilation in hospitalized patients.
Indomethacin and typical antipsychotics were not tested in randomized controlled trials – rather, the strategy was to compare them to matched patients that were taking either celecoxib rather than indomethacin, or atypical rather than typical antipsychotics. Patient groups were also small.
But both indomethacin’s target, PGES2, and the target of typical antipsychotics, the SigmaR1 receptor, did interact with coronavirus proteins in the preclinical studies that were also part of the paper.
Whether for new drug development or repurposing, the approach is set up to identify host-directed therapeutic possibilities.
Adolfo Garcia-Sastre, director of the Global Health and Emerging Pathogens Institute, explained that this focus on host proteins is a consequence of the team’s attempt to identify approaches that will work not only against SARS-CoV-2, but also against other coronaviruses. Viral targeting approaches, he said, are “very good for a specific virus, but not broad-spectrum.”
As co-author, Marc Vignuzzi, principal investigator at the Institut Pasteur, pointed out that this approach can also identify more targets overall: “The virus makes a dozen proteins, but targets hundreds of host proteins,” he said.