Skip to main content Skip to secondary navigation
Main content start

A drug hunter’s new tricks

Nathanael Gray on the rules that guide drug design for diseases like cancer and the new rules that could make more treatments possible

Cells are busy places, and proteins are almost always at the center of the action, orchestrating processes like making and recycling molecules, launching an immune response, and making your muscles contract. They also are at the center when those processes go awry in different diseases, making them alluring drug targets for researchers and clinicians.


Nathaneal Gray, with glasses, short brown hair, and a light blue shirt, stands in front of a simple grey backdrop.
Nathanael Gray

Nathanael Gray, the Krishan-Shah Family Professor of Chemical and Systems Biology, is a drug hunter. Gray, who recently joined Stanford ChEM-H as an Institute Scholar, is an expert in designing molecules that can bind to and interrupt the function of proteins to treat diseases like cancer. After building a career on designing drugs that target different proteins, he recently turned his focus to so-called “undruggable” proteins, those that cannot be blocked by a drug hunter’s usual tricks. His strategy, redirecting cellular functions to turn against dangerous proteins, could make more proteins—and more diseases—tractable.

He talked with ChEM-H about getting a toehold in proteins, how to drug the “undruggable,” and the importance of team science.

A lot of people may think of chemists as people who make molecules, but you talk about making “drug prototypes” instead of just molecules. What does that mean?

What’s cool about being a chemist is that you’re a jack of all trades. You can use your skills to build many kinds of molecules. But traditionally, people in chemistry are focused on the structure of the molecule, rather than the function or ultimate application. That means that they start by making a molecule and then they go out looking for the end-user of that molecule.

We take more of an engineering approach. We face the problem head-on, and we’re agnostic about the tool we build as long as it meets the challenge.

When we take that approach, what we build is a drug prototype, a molecule that we can use in a proof-of-concept experiment in cells to test if this kind of drug could work to treat a disease. From there, we can optimize the prototype by making it safer and more effective, for instance.

I think the most critical part of leading a research lab is training the next generation of drug hunters: scientists who are both expert practitioners of chemistry and also knowledgeable about biology and disease, people who know how to design, build, and optimize new kinds of drug prototypes.

How can you look at a problem and decide if it’s one that you can solve?

I’ve found that I need to make sure the project meets three criteria to be successful. I call it the “golden triangle” of translational research. First, you need to have some idea of how exactly your target protein fits into the disease biology. How confident are you that modulating the protein will impact the disease? Second, you need to have some way of assaying, or measuring, how well your drug prototype affects that protein, ideally in a physiological setting. And the third corner of the triangle is the drugging strategy. You need some way to get a ‘toehold’ into that protein by finding a molecule that directly or indirectly changes the protein, often by inhibiting its function

So you need to know that going after this protein will actually treat the disease, and you need a way to measure if the drug you use is affecting the protein. But how do you design a drug? How do you get one of those “toeholds?”

You first have to decide on your approach, like if you are going to use an antibody or engineer the cell’s genes. Recently there has been a resurgence of interest in a traditional approach: using a small molecule that fits into a pocket in the protein and forms a permanent chemical bond with the target.  Famous drugs like aspirin or penicillin are good examples of these kinds of compounds that permanently disable a protein’s function until the cell can make new proteins.

These inhibitors have conventionally been molecules found in nature, and people discovered their therapeutic properties through traditional use or through screening big libraries of molecules. But recently, we’ve gotten better at intentionally designing these molecules. We have more information about the structure of proteins, and we have a variety of chemistries we can use to react with specific amino acids in proteins. So we can start from the three-dimensional structure of the protein and design a molecule that will both fit nicely into the pocket and react with one of the atoms in that pocket to form a chemical bond.

Can you give me an example of how this strategy actually works?

Several years ago, a colleague of mine at the Dana Farber Cancer Institute, Pasi Jänne, was working on inhibitors of EGFR, a protein that is important in lung cancer. He had run the first clinical trial with an EGFR inhibitor for non-small cell lung cancer and discovered that patients who were relapsing had a version of EGFR with a specific mutation, a small alteration in the chemical structure of the protein. When I looked at the protein’s structure, I realized that we could design a drug that would be more selective for that mutant protein, fitting and binding better within that protein’s pocket. A proof-of-concept showed that this mutant selectivity strategy, which hadn’t yet been accepted universally, could treat those who relapsed. Our prototype drug would later evolve into a drug at AstraZeneca called Osimertinib.

So, designing these covalent inhibitor-type drugs works. But there are a whole lot of proteins out there that are “undruggable,” meaning they don’t have this “pocket” we can target with our molecules, so we have to come up with new strategies to deal with those.

What do you do then? How can you inhibit a protein if there’s no pocket for a molecule to fit into?

By harnessing the natural biochemistry of the cell. Cells are already equipped with proteins that do all kinds of things, like building, breaking apart, and moving around molecules. To drug “undruggable” proteins, we redirect those processes to affect our target protein.

You can imagine that an “undruggable” protein looks like a softball. And how do you get hold of a softball? With a mitt that goes around it. We use the same idea with proteins. We can make drug prototypes that do two things at once: wrap around the target protein and recruit some other protein within the cell that will change the fate of our target.

These are called proximity-based therapeutics. And we are just starting to figure out the rules that guide how we design these drugs.

I think of it like a matchmaking service. The magic is that we have to bring the protein close enough to another protein for just long enough that is has some effect.

And what’s an example of the kind of effect that other protein might have?

The most common example right now is induced degradation. So, say you have a protein that drives cancer growth. We can make a molecule that, on one end, wraps around that protein and, on the other end, recruits molecules that trigger the cell to chew up that protein and turn off cancer.

We’ve recently shown that we can use this strategy to get rid of a protein that dampens the immune system. By doing that, we’re able to boost the immune response in cancer cells.

But this is just the beginning. We’re working to figure out if we can trigger degradation in only certain parts of the body, so you can degrade proteins only in the kidney or only in the brain. We’re also trying to figure out what other cellular mechanisms we can co-opt. Instead of degrading a protein, can we change the way that protein behaves or what cellular neighborhood it lives in?

What brought you to Stanford?

There were four factors. One was the Stanford Cancer Institute whose director, Steve Artandi, was very interested in strengthening the cancer therapeutics program. The second was Stanford ChEM-H. I’ve known Carolyn Bertozzi since I was a first-year graduate student at Berkeley and she had just started her lab there in 1994.  Carolyn is a great scientist and leader and I’m excited about her vision. And the third was the Innovative Medicines Accelerator, which is led by Chaitan Khosla. The IMA is a partnership between ChEM-H and the School of Medicine to help more Stanford discoveries turn into real therapeutics; it’s fantastic to have support from the President’s office to help faculty move their scientific discovery towards societal impact. The combination of those three factors made Stanford a great place to do research.  Finally, I grew up in the Bay Area and it’s great to be home and near my family.

Tell me more about the IMA. What’s special about it?

The IMA is structured to enable researchers to develop drug prototypes

Biotech and pharmaceutical companies have what I call “widget-making” ability, which is expertise in building things like molecules, proteins, gene editing tools, or engineered cells. But they often lack biologists or clinicians who have spent their whole lives working on a disease target. Academia, in that sense, has incredible depth but that widget making isn’t available to everyone.

The IMA aims to bring that widget making to more Stanford researchers so that they can take their research in a more translational direction.

And, beyond that, the IMA encourages innovation. At a lot of companies, you are rewarded for not being associated with failure, and the only way to avoid failure is to not take chances. But in research, you have to take chances if you want to innovate.

What are you excited about?

What really excites me about being here at ChEM-H is the possibility of collaborating across labs and disciplines to develop new drugs. We’re in a golden period of drug development right now, and there’s a lot of excitement around what we can accomplish as teams. We’ve already witnessed the game-changing impacts of computer science on the life sciences in areas like predicting three-dimensional protein structure. To really understand the incredible complexity that underlies the orchestrated biochemistry of life, we need to embrace all fields. Team science is the future.

I’m also excited by the number of early-career scientists just coming on board at ChEM-H. It’s always a fun experiment to throw together so many brilliant young people and see what exciting discoveries emerge.


Gray is the Stanford Cancer Institute Co-Director of Cancer Drug Discovery and the Co-Lead of Medicinal Chemistry at the Innovative Medicines Accelerator.

More News Topics