Protein Interactions Point the Way to Drugs Against COVID-19

Quantitative biologist Nevan Krogan is mapping how the coronavirus takes over human cells — drumming up clues that could show us how to fight it.

By Christopher King

Infected human colon cells produce filopodia protrusions (white) extending out from the cell surface containing viral particles (viral N protein in red). Credit: Dr. Robert Grosse, CIBSS, University of Freiburg.

In the face of unceasing investigation by the global research community, the SARS-CoV-2 virus is gradually yielding its secrets. Scientists are learning more about the molecular workings of COVID-19 disease and its seemingly (and disturbingly) random clinical course.

One researcher whose lab has contributed to major advances in recent weeks is Nevan Krogan, PhD, Director of the Quantitative Biosciences Institute (QBI) and Senior Investigator at the Gladstone Institutes, University of California, San Francisco, where he’s also part of the Data Science and Biotechnology Institute, led by Chan Zuckerberg Biohub investigator Katherine Pollard.

Nevan Krogan, PhD, is Director of Quantitative Biosciences Institute & Senior Investigator at the Gladstone Institutes, University of California, San Francisco

In his recent work, Krogan has spearheaded collaborations that have offered promising leads into how to precisely target existing drugs and compounds to combat COVID-19.

In a Cell report published earlier this month, Krogan’s team detail their findings on how SARS-CoV-2 modifies proteins within infected host cells to enhance the virus’s ability to invade neighboring cells. These viral modifications involve a route of infection not previously seen in the family of coronaviruses. The report also specifies at least half a dozen existing drugs that — with further testing and official approval — could be pressed into service against the virus.

Those results expand on work reported in a Nature paper published on April 30, in which Krogan and colleagues present a detailed mapping of more than 300 protein-protein interaction sites, revealing how the virus uses human cells. Each site also represents a potentially effective target for existing drugs.

Assessing the approach that led to these findings, Krogan says, “I don’t know anyone else who’s basically using the specific biology to inform drug discovery. A lot of researchers are just screening tens of thousands of drugs, seeing the hits, and then saying, ‘Okay, what now?’ What we’re doing is data-driven drug discovery — trying to understand in an unbiased way how the virus infects our cells, and then using that information to point us in different therapeutic directions.”

“What we’re doing is data-driven drug discovery — trying to understand in an unbiased way how SARS-CoV-2 infects our cells, and then using that information to point us in different therapeutic directions.”

In the coronavirus itself, Krogan sees “pluses and minuses.” On the latter score, he notes that, compared with his studies of the proteins in HIV and herpes, the number of SARS-CoV-2 proteins is similar to those other viruses. “But,” says Krogan, “I’ve never seen a virus get its fingers in so many different biological processes as this one. To me that’s remarkable, and it may explain why the virus manifests itself in so many different organs and symptoms.”

On the plus side, he says, the coronavirus does not appear to introduce as many mutations as influenza. This means that, unlike the annual updates required for flu inoculation, a successful treatment or vaccine for COVID-19 will likely be viable for the future.

A map of the interaction points between coronavirus proteins and human proteins — each site a potential drug target. (Credit: QBI Coronavirus Research Group)

From HIV to COVID-19

Krogan’s current work on the coronavirus also benefits from the foundation laid by his earlier research in tracking the interactions between viral and human-host proteins in HIV infection.

“We’re heavily involved in the development of technology,” he says, “and this technology is very disease-agnostic. It can be and is being applied to many different disease areas. We were funded to do HIV work, but the message in our coronavirus research is that it’s all connected: You can apply the same technology to HIV or Alzheimer’s or cancer or coronavirus. But what’s even more exciting is that the biology is the same. The genes being hijacked by HIV are the same ones being hijacked by coronavirus. And the genes being mutated in lung cancer are being hijacked by coronavirus. It’s all connected, and we’re in the business of making those connections on many levels.”

For their first SARS-CoV-2 paper, reported in Nature, Krogan and colleagues mapped out the protein-protein interactions between coronavirus and human cells — that is, identifying the precise intersections between coronavirus proteins and human-host proteins, and thereby locating potential sites for pharmaceutical intervention against the virus. “Most researchers are trying to find drugs to target the viral proteins. That makes sense. But we’re taking a different tack: We’re saying, ‘Let’s understand how this foe attacks us, identify what machinery it needs in our cells, and target that.’”

Based on their mapping and the potential sites for drug targeting, Krogan and his UCSF team identified 69 drugs and compounds approved by the U.S. Food and Drug Administration. Ultimately, their collaborators at the Institut Pasteur in Paris and Icahn School of Medicine at Mount Sinai in New York City tested 47 of the drugs against SARS-CoV-2 infection in the cells of African green monkeys, which approximate human cells in their response to viral infection.

In all, the results identified nearly a dozen existing drugs that specifically interfere with coronavirus infection, either by impeding the process of viral “translation,” in which the infected cell makes new viral particles out of viral RNA, or by decreasing the activity of host receptors that likely help the virus replicate itself.

Tentacle-like filopodia, studded with coronavirus particles, branch out from an infected cell. (Credit: Dr. Elizabeth Fischer
Tentacle-like filopodia, studded with coronavirus particles, branch out from an infected cell. Credit: Dr. Elizabeth Fischer, NIAID/NIH.

Coronavirus Spreads Out

In the subsequent collaboration reported in Cell, Krogan and colleagues examined kinases, which are proteins that function as control switches, regulating many aspects of cellular operation. In particular, the team investigated how the coronavirus, in the course of infecting cells, alters the actions and even the physical structure of host cells. As with the earlier study, the idea was to identify existing drugs that might interfere with the viral hijacking.

Specifically, the team examined protein phosphorylation, in which a kinase attaches a chemical marker to other proteins to initiate cellular action. Comparing phosphorylation in healthy cells against the process in infected cells, the researchers were able to pinpoint the kinases affected by the virus. They ultimately identified 49 such proteins.

One of these kinases, casein kinase 2 (CK2), drew particular attention from the team, as its functions include influence over cell structure. Creating a new collaboration with microscopy specialists at the University of Freiburg, Germany, the researchers learned that following viral infection, CK2 prods cellular machinery into creating tube-like projectiles called filopodia. Like tentacles, each laden with viral particles, these filopodia reach out to nearby cells, creating an avenue of infection separate from the angiotensin-converting enzyme 2 (ACE2) surface receptors by which the virus was already known to infect host cells. Until now, filopodia, although familiar in Ebola, Marburg, and other viral infections, had not yet been observed in SARS-CoV-2 or any other known coronavirus.

Until now, filopodia, although familiar in Ebola, Marburg, and other viral infections, had not yet been observed in SARS-CoV-2 or any other known coronavirus.

In checking the 49 kinases against 250 existing drugs that function by inhibiting kinase activity, the team found 87 compounds with potential effect against the cellular disruption caused by the virus. As in the earlier study, collaborators in New York and Paris tested 68 of these drugs and found several that kill the virus in cells. A handful that are already approved or in the process of approval for treating cancer and other conditions hold particular promise against the coronavirus. As Krogan says, “There are five or six that we’re pushing to get into clinical trials.”

Like much of today’s frenetically paced work on the coronavirus, these recent findings depended on rapid cooperation between far-flung scientists. In this, as Krogan notes, his team had an advantage.

“Most of the collaborations you’re seeing now in COVID research — you don’t just snap your fingers and they happen,” he says. “This represents years and years of building up trust. More than anything, I’m known for collaboration, and our mandate at the QBI — our self-imposed mandate, really — is about breaking down silos and bringing together on our campus scientists from academia and pharmaceutical companies around the world. So the relationships that are bearing fruit now — it’s been years putting those together.”

Stay up to date on the latest research on SARS-CoV-2 and COVID-19 — including the quest for therapeutic compounds — with Meta. Here are some feeds to get you started:

COVID-19

COVID-19 Therapeutic Drug Development

As a project of the Chan Zuckerberg Initiative, Meta is free — and accessible to researchers everywhere.

Meta is a free research discovery tool from the Chan Zuckerberg Initiative, providing a faster way to understand and explore science through personalized feeds.