QBI Coronavirus Research Group identifies drugs that block SARS-CoV-2 in the laboratory

In late April, a mere two months after setting out to identify new potential therapies for COVID-19, a consortium of scientists brought together by the UCSF Quantitative Biosciences Institute (QBI) announced that several existing medications for conditions as varied as cancer and psychiatric illness could suppress SARS-CoV-2 in the laboratory.

QBI director Nevan Krogan, PhD, founded the QBI Coronavirus Research Group (QCRG) at the outset of the COVID-19 pandemic and immediately got to work coordinating research into SARS-CoV-2, the virus that causes COVID-19, despite ongoing shelter-in-place orders and other pandemic-related disruptions.

QBI was founded to catalyze scientific collaborations across fields at UCSF and beyond, reporting through the School of Pharmacy. Krogan is a faculty member in the School of Medicine and an investigator at the Gladstone Institutes.

After identifying 69 existing drugs predicted to interfere with SARS-CoV-2 hijacking of human proteins, QCRG worked with collaborators in New York and Paris to test those compounds against the virus. The group’s newest paper revealed the results of those tests against live SARS-CoV-2 in the laboratory — several of the drugs were found to effectively fight the virus, and one actually made the viral infection worse.

The School of Pharmacy’s health and science writer, Levi Gadye, PhD, spoke with Krogan earlier this month about this groundbreaking collaboration and the implications of its recent findings.

Gadye: In March, your team identified 69 existing drugs with the potential to interfere with SARS-CoV-2 infection of human cells. How did you and your collaborators more recently winnow this down to the most promising candidates?

Krogan: We had reported in a bioRxiv paper, a blueprint or a map, if you will, of how SARS-2 hijacks and rewires the host by identifying over 300 different proteins that we found physically connected to at least one of the viral proteins. Working closely with chemical biologists like Kevan Shokat [in the UCSF School of Medicine] and Brian Shoichet [in the School of Pharmacy], we predicted 69 different drugs and compounds that would target at least one of these human proteins.

We did not have the virus growing in San Francisco. But through QBI, we had established tight collaborations with some of the best virological labs in the world, including one in Paris at the Pasteur Institute and one at the Department of Microbiology at Mount Sinai Hospital in New York. Over the last six weeks or so, they've been able to test about two-thirds, or close to 50, of these drugs and compounds. And this has allowed us to identify essentially two sets of drugs and compounds that we think are very exciting.

The first set are two drugs that are known to inhibit the process of translation, which is the process of our cells making proteins. The virus needs that translation machinery in order to make its own proteins, hijacking that machinery in a way we don't completely fully understand, but in a way that’s presumably beneficial for the virus.

Our map pointed out a couple of specific points in the translational process that the virus depends on, and we then identified two drugs targeting these parts of the translational machinery. One, which is being used in a clinical trial to see if it can fight off multiple myeloma, a particular type of cancer, is the drug Zotatifin, from a company that spun out at UCSF called eFFECTOR, co-founded by Kevan Shokat and Davide Ruggero [in the School of Medicine]. And then there's another compound that we predicted could have an effect, also inhibiting translation, called plitidepsin, and it’s actually approved for treating multiple myeloma in Europe.

Plitidepsin is currently in a clinical trial to see if it works against COVID-19, and eFFECTOR is looking to start a clinical trial with Zotatifin really soon on this. So those two drug compounds are looking good.

The other set of drugs and compounds hit a couple of receptors that we identified in our map called sigma R1 and sigma R2. These are important because they're very, very druggable. This is work spearheaded by Brian Shoichet. He made predictions about different drugs and compounds that bind to these receptors, including antihistamines, antipsychotics, cough suppressants, the female hormone progesterone, and anti-anxiety drugs, and some preclinical molecules, as well as hydroxychloroquine. They are all in the same category.

We found that all of these have relatively potent antiviral effects in the assays that we were doing both in New York and Paris. And what's interesting here is that when you just look at the chemical structures of all these compounds, there's no way you could link them all together. But the biology makes connections across all of these different particular compounds. This map allowed us to come up with this discovery of these receptors. We show that when we turn them down [with these drugs] there are antiviral effects. But the big question is, are any of the drugs that we identified going to actually work in humans?

That's a good question. Maybe not those ones. But the fact that we know what the receptors are, Brian Shoichet, and hopefully many others around the world, are going to come up with something that works, and hopefully it's an FDA approved drug that can be put into people.

Gadye: You were able to test the effects of some of the drugs that we've heard about in the news, like hydroxychloroquine. What were we able to find out about those drugs?

Krogan: In our laboratory assays, hydroxychloroquine does have antiviral effects. However, we did some more targeted in vitro analysis with hydroxychloroquine in comparison with some of these other drugs and compounds in the same category.

Hydroxychloroquine has been in the news with a lot of clinical trials, and one of the issues there is that it's toxic. Specifically there's cardiac toxicity, and that is to say that some individuals who take the drug will have heart problems and have heart attacks and die. There was a big trial that actually ended in Brazil a couple of weeks ago for that exact reason.

We think we now understand why. Our collaborator at the University of North Carolina, Bryan Roth, did in vitro binding experiments with hydroxychloroquine and a receptor on the heart, called HERG. Hydroxychloroquine binds more strongly to HERG than sigma R1 and R2, compared to some of these other compounds. So we think we understand why it is toxic. And we also think some of these other drugs and compounds will not have that same cardiotoxicity, though they could have other toxicity issues.

Gadye: Gotcha. You mentioned a few of the newer drugs that are showing new promise. Were any of them surprising? Either from the perspective of the diseases that they were originally approved to treat or in terms of their biology?

Krogan: Well, I think it's more about the biological mechanism. If you ask, “Oh, look, there's an antihistamine and there's an antipsychotic, what's the connection to potentially treating COVID-19?” The answer is the biology. If you know what the target is, there is a connection at the end of the day. So that's the power of doing the biology first and then going into pharmacology, [you find them and then] you go and test them. If they work, you go back to the biology and you can start to tweak things and make things much more potent.

All of those things that I've described are, for the most part, classified as antagonists in that they bind the receptors and turn them down.

Now Brian told me one morning that there was an agonist for the sigma R1 and R2 receptors. Agonists turn protein activity up. So if you add an antagonist, it binds to the receptors, turns down the function, you always see an antiviral effect, and if you turn it up, potentially you'd see something proviral, that's to say that the virus would actually grow better in the laboratory setting. And the one agonist he suggested that we looked at was dextromethorphan. And sure enough, in the laboratory setting, when you add it, the infection grows even more, nicely confirming we were looking at the right receptors.

But dextromethorphan is in the vast majority of cough suppressants that are out there. Obviously, one of the major symptoms of COVID-19 is cough, and people may be taking this drug. The big question is, would taking a cough suppressant when you're infected with COVID-19 have any effect on the infection?

We don't know. But at least in a laboratory setting, we see that dextromethorphan has a proviral effect. We're just reporting what we've seen in the lab. There’s a strong caveat, since we don't know if it's going to do the same thing in humans, but we knew we would have to responsibly report that.

Gadye: Does that mean that the antagonistic route, trying to block certain receptors or other proteins, will continue to be the one that receives a bit more focus?

Krogan: Absolutely.

Gadye: So zooming back a little bit, we are going to continue to see exciting results and get new details about how this infection works through research like yours. But people are desperate for solutions to COVID-19. How would you say people should try to interpret news like this?

Krogan: Well, there's still so much more work to do. We're not suggesting that any of the drugs that we find can be immediately put into humans with this disease, by no means. But the research provides us a kind of a lamppost to take a look under and points us in a direction that looks incredibly promising, because we're dealing with a couple of proteins that are very, very druggable, today.

In that regard, back to this idea of, if you know the biology, you're so much further ahead with respect to pharmacology. We're hopeful that we or somebody else will come up with the right molecule that will be very potent against these receptors that the virus uses to infect our cells, and then hopefully, a treatment comes out of that.

Personally, I think it will probably be a combinatorial approach where we use a couple drugs against a couple of human proteins, or one of our drugs with remdesivir, which is this Ebola drug that just got FDA approval. As we saw with HIV, a cocktail approach was the big breakthrough. My gut feeling is we're going to see a similar approach come to fruition and be successful with respect to this coronavirus.

Gadye: Sounds like there’s much to be hopeful for. One last thing, how does this work fit into QBI’s growth over the years?

Krogan: QBI was formed in March 2016 in the UCSF School of Pharmacy under the tutelage of Joe Guglielmo, who played an instrumental role in putting this together. The mantra of QBI really has been collaboration in breaking silos down, fostering collaborations between labs at UCSF and QBI, collaborations across different institutes around the world, and collaborations between pharmaceutical companies and academia. We've been working very hard over the last three, four years, setting up collaborations on a number of different levels.

That's why we were in such a perfect position, I would argue, to respond to this pandemic, by bringing together hundreds of different scientists here in San Francisco and around the world. That just doesn't happen by itself. It takes time to build up these connections. This is what we were being built to do. And we did it, and we're doing it. So, to me, that's very exciting.

Gadye: Thanks so much, Nevan.

More

Nature Publication Presents New ‘Blueprint’ Revealing How SARS-CoV-2 Virus Hijacks Human Cells (UCSF News)

UCSF team has discovered drugs that block coronavirus, paving way for 'a better drug sooner' (SF Chronicle)

A SARS-CoV-2 protein interaction map reveals targets for drug repurposing (Nature)