Columbia University in the City of New York

Lab in the Time of Coronavirus: "Building a Better Antibody"

Meet a protein aficionado working on a new way to treat COVID-19.

KATHLEEN DURKIN I’m Kathleen Durkin of Columbia's Zuckerman Institute.

DEVIN POWELL And I’m Devin Powell of the Zuckerman Institute.

KATHLEEN Welcome to Lab in the Time of Coronavirus, where we explore how scientists are pivoting to tackle COVID-19.

DEVIN A few episodes ago, we met Larry Shapiro: a researcher hoping to create a new COVID treatment using antibodies collected from people who had recovered from the virus.

KATHLEEN Today we’re sitting down with his longtime collaborator Barry Honig, PhD. He’s a biophysical chemist who is working on pushing antibodies to the next level.

DEVIN Barry is a principal investigator at Columbia’s Zuckerman Institute, and he’s an expert in how individual cells stick together properly when growing. How cells in the heart, for instance, know to attach to other cells in the heart, and not to liver cells.

BARRY HONIG How two cells in our own bodies would recognize each other.

KATHLEEN Barry studies proteins that help cells recognize each other.

BARRY They have proteins on the surface of the cells, we call them cell surface proteins. When those proteins see other proteins they like to bind to, then the cells come together.

DEVIN He’s discovered that the cells in our brain have patterns of proteins that give them unique identities, like letters on a car’s license plate or barcodes on groceries. This helps growing neurons form proper connections, called synapses.

BARRY Neurons have to recognize who to form synapses with, who not to form synapses with. Every neuron has a unique identity, and we revealed the barcoding mechanism for vertebrate neurons.

KATHLEEN So what does this have to do with COVID?

BARRY So that sounds very remote from viruses, it is, except you could argue that the recognition in the brain is related to the recognition in COVID. The fundamental notion of recognition, of cells recognizing because proteins recognize and bind together, is the unifying principle. When viruses infect cells, they must penetrate the cells. And the way they get inside is the viral protein, for COVID it’s called the spike protein, binds to this protein on the surface of a human cell. Once they stick together they – it’s like a lock and key, the shape of one fits into the other – then that initiates a process that lets the entire virus into the cell.

KATHLEEN Barry just got a grant from the Gates Foundation to look for a treatment. He wants to use antibody proteins to interfere with the proteins on the surface of COVID that allow the viruses to stick to human cells.

BARRY That’s the basic goal of what we’re talking about: to use antibodies to attack viruses.

KATHLEEN Similar antibody treatments have been developed for AIDS.

BARRY So there’s a series of antibodies, proteins that basically can be used as drugs, effectively. They bind to a viral protein. In the case of AIDS they bind to GP120, which is responsible for entering the cell, and because they bind to it they don’t let it bind to its human target. And it stops the protein from entering the cell. Similarly, with COVID, Regeneron has proteins that bind to the spike protein of COVID, and once the Regeneron protein binds then the spike protein can’t enter the cell. So this is not for a vaccine, it’s for treatment basically.

KATHLEEN Like babyproofing an electrical socket by plugging it with a plastic guard, a tightly-fitting antibody protein renders the virus less dangerous.

BARRY The trouble is that it doesn’t bind tightly enough and isn’t effective enough when injected from the outside. So what you want is to design a protein that binds more effectively. What mutations can I make to make it bind better?

DEVIN Designing antibody proteins that interfere with COVID starts with using computers to predict what happens when you change the structure of these proteins when you mutate them.

BARRY Instead of randomly trying to make mutant proteins, does this one stick, does this one stick, we first use computational tools to say, this is a good idea, that’s a bad idea, this at least on the computer works well. And then we go to the experiment and see if we were right. The underlying rationale for this really comes from the chemistry department. What we’ll do is synthesize those mutant proteins in our lab, test experimentally whether we’ve actually succeeded in making them bind better, and then go back to the computation. So it’s a tight feedback loop between computation and experiment.

That’s the basic goal of what we’re talking about: to use antibodies to attack viruses.

DEVIN Another branch of Barry’s research compares the proteins made by viruses to the proteins that exist inside human cells. The goal here is to better understand how viruses take control of our cells once inside.

BARRY For viruses to be effective they somehow must disrupt human function in some way.

KATHLEEN Barry discovered that the most effective and most dangerous viral proteins mimic the shapes of human proteins.

BARRY By looking like them, by confusing the system, if you will. You see if a viral protein looks like a human protein, maybe it does what the human protein is doing. Maybe it does it a little better, so the human protein doesn’t get to do its thing anymore.

DEVIN Barry’s colleague at Columbia University Irving Medical Center, Sagi Shapira, PhD, took the research one step further. He noticed that many of the human proteins being mimicked were related to blood coagulation. So he looked at whether people who had macular degeneration, which is a disease associated with leaky blood vessels in the eye, had worse outcomes from COVID.

BARRY So they actually tested whether the outcome of people with macular degeneration was worse or better than most people, and there was a remarkable correlation. If you have macular degeneration you should definitely not get COVID because your probability of being intubated and dying goes way, way up. So that was an observation that came from this purely theoretical observation in our paper about the mimicry of proteins involved in this process.

DEVIN So over the years, studying proteins has led Barry to investigate everything from color vision to cancer.

BARRY I’m basically what you’d call a theoretician. I’m not an experimental scientist. I like to read other people’s experiments. But I always wanted to look for general principles in biology. Are there general laws, maybe not laws of physics, but are there general principles of how things work? That’s what really always interested me. And therefore it’s very easy for me to go from one problem to another because I see the connections, that’s sort of what I do.

KATHLEEN Thanks for listening to Lab in the Time of Coronavirus. Take a look at the show notes for links to all the things we discussed. You can find all of our episodes at or on iTunes. Take a moment to rate and review us on iTunes. That makes it easier for other people to find us.

DEVIN Special thanks to Professor Barry Honig for sitting down with us for this episode and to the entire Zuckerman team. The music, as always, was provided by Miguel Zenón, Jazz Artist-in-Residence at the Zuckerman Institute. If you have thoughts or questions, you can find us on Twitter and Instagram @ZuckermanBrain. As we wrap up this episode, it’s time as always for Kathleen’s last question.

KATHLEEN What is the first thing you plan to do post-pandemic?

BARRY Have a hell of a meal somewhere, probably.

Connect with us

Previous Episodes

View All News >