On April 30, 2020, the preprint study described in this story was published in Nature with additional data showing that several of the identified compounds limit the replication of novel coronavirus in cells. Read more here.
A team of scientists has found 69 drug candidates that might help treat COVID-19, the disease caused by infection with the novel coronavirus SARS-CoV-2. Forty-one of them are either already approved by the US Food and Drug Administration or currently in clinical trials for other uses, and might be more quickly available to patients if shown to be effective.
The team, led by molecular biologist Nevan Krogan of the University of California, San Francisco, identified 332 proteins that SARS-CoV-2 proteins bind to and use to replicate inside the human body (bioRxiv, 2020. DOI: 10.1101/2020.03.22.002386). This allowed them to pinpoint compounds that might interfere with the virus’s ability to interact with cells. “If the viral protein is hijacking the host proteins, and then our drug inhibits that protein, then we will be denying the virus the ability to hijack that host protein,” says UCSF molecular pharmacologist Kevan Shokat. There are currently no approved treatments for COVID-19. The team’s potential drug list includes diabetes, cancer, and hypertension drugs, in addition to chloroquine, which is currently in clinical trials. Ebola drug remdesivir, which targets a viral polymerase not targeted by the UCSF screen, is also in clinical trials.
Soon after SARS-CoV-2 emerged, scientists sequenced its genome; Krogan’s team downloaded it on Jan. 24. When he heard of community spread in California, Krogan immediately stopped work on other projects and focused on the virus. The scientists identified 29 viral genes corresponding to 29 proteins that human proteins interact with. “Essentially what we did was chop up the viral genome into its individual DNA pieces so that we could study the corresponding proteins in isolation,” Krogan says. The team then introduced each piece of DNA into human HEK293T kidney cells, and tagged each viral protein to allow the researchers to fish it out. “The virus and the viral proteins need our proteins in order to infect us,” Krogan says. “So what is that machinery? What are those proteins?” The team connected 26 of the viral proteins with 332 human proteins.
With these data in hand, Krogan called in the chemists to help find small molecules that would target the host proteins that they identified. Along with Shokat and Brian Shoichet, a computational biologist at UCSF, the team generated the list of drugs that target one or another of the highlighted host proteins.
The group is now testing the candidate drugs in infected cells, hoping to find one that blocks viral replication without harmful side effects. Scientists at Mount Sinai Hospital in New York and the Pasteur Institute in Paris are currently running viral assays with the candidates against coronavirus infected cells.
What normally takes a couple of years, the team got done in a matter of weeks. Krogan says this was possible because his team collaborated with hundreds of other scientists in 21 other laboratories.
The list of compounds is a good start, says Dennis Liotta, executive director for the Emory Institute for Drug Development, “but there’s a lot of work that needs to be done to validate the drug targets.”
The paper has been submitted for publication in a peer-reviewed journal.
This story was updated on March 26, 2020, to correct the disease that remdesivir treats. It treats Ebola, not arthritis. The number of drug candidates found in the study was corrected from over 60 to 69, and the number of those molecules that were already approved by the US Food and Drug Administration or in clinical trials was corrected from 38 to 41.
This story was updated on April 29, 2020, to correct how remdesivir acts on the virus. The drug targets a viral polymerase, not a viral protease.