This is From Our Neurons to Yours, a podcast from the Wu Tsai Neurosciences Institute at Stanford University, bringing you to the frontiers of brain science.
Today on the show, I'm excited to bring you my conversation with Carolyn Bertozzi. She's a Nobel laureate, a MacArthur genius grantee, and an all-around cool person. I failed to bring this up in our conversation, but she used to play with Tom Morello of Rage Against the Machine back in the day. So, if the Nobel Prize doesn't impress you, she's legit. Carolyn wouldn't call herself a neuroscientist. She's a chemist at heart, but her work is having a big impact on neuroscience.
That work focuses on the sugar molecules that stud the surface of our cells. It turns out these molecules are incredibly important for how cells communicate with each other. But understanding these molecules is so complex and challenging that Carolyn had to invent a new branch of chemistry in order to study them. So, we're going to be talking about how this research is opening doors not only within neuroscience, but across biology and drug development. We're going to get into some fun chemistry here, but we're in extremely good hands. So, let's jump right into my conversation with Carolyn Bertozzi.
Carolyn Bertozzi, welcome to From Our Neurons to Yours. I'm so glad to have you here.
Thank you. Great to be here.
I'm particularly excited to talk about this work because you've essentially developed a whole new branch of chemistry that's changing how we understand cell biology. And it's launched what, a dozen now, biotech companies working on therapies for a bunch of different diseases and really highlights the value and impact of basic, curiosity-driven science to change our understanding of ourselves, to change our ability to treat diseases. I want to get into all of that, particularly some of the implications for neuroscience since that's the focus of our show. But first I'd love to start with sort of a quick primer on the big picture of your work, which is focused on an aspect of our biology that's been ignored for a long time, if that's fair to say, or had been. The fact that nearly all of our cells throughout our bodies are covered with sugars. I think that might be hard to picture for a lot of people. Like, are our cells candy-coated like M&M's? I wonder if you could paint us a picture of what these sugars really look like and what we think they might be doing there.
You bet. You know, it's funny because when I give lectures to a non-scientific audience, I often show a slide with an M&M on the slide just to make exactly this point that cells have a sugar coating. But then I switch to another image of the planet Earth because I don't want people to think the sugars on our cells are as simple as they are on the M&M. They're much more complicated with features and patterns. So, I like to think of the cell as really like the landscape of a planet where you have forests, you have open fields, you've got tall trees and short shrubs and grass, and in many ways the complex carbohydrates form a kind of vegetation landscape on the surface of the cell. And they're moving.
And you know, I'm here in the Bay Area, so from time to time I'll go down to Monterey where there's a very famous aquarium, and in that aquarium there's a kelp forest where you stand there looking in this giant tank with the kelp sort of swaying back and forth, you know, in the waves of the water. And I like to think of the cell surface carbohydrates as being like that. They're these long chains protruding from the surface of the cell, swaying as fluid flows by our cells in our bodies. And they're very structurally complex and also a very rich code of information about the biological state of the cell. So different cell types have different kinds of forests on their cell surface, and that turns out to be important in the way that cells interact with other cells and the way that they form organs and tissues, and pretty much all human biology at some level involves these sugars.
I think, I mean, when I was taking biology classes and things, we learned about the cell surface proteins, things like, you know, there are ion channels, there are various mechanisms sitting in the lipid membrane of the cell. You're painting a picture, and I love this picture of the cell as the surface of a planet with different ecosystems and forests and things. What does that add to our picture of the biology of the cell to have a sense that there are these long chains of carbohydrates? What do we think they might be doing that's different from some of the proteins that we've been able to study for a long time?
Well, first of all, I should say that it's probably helpful to think about the sugars holistically along with those other molecules like the ion channels and the lipids. For example, all of those ion channels are decorated with polysaccharides. They're glycosylated. They're glycoproteins. And the sugars attached to the ion channels regulate their functions. So they affect their folding. They affect the way that subunits come together to form a multi-subunit complex. In the case of ion channels, they very explicitly can control basically the electric potential across the ion channel because some of these sugars are highly negatively charged. So they'll actually influence the rate of flow of cation ions through the channel.
And many of the lipids in your membranes are glycolipids. So they have sugars attached to them as well. So the sugars and the proteins and the lipids, they're all part of this molecular landscape. They're all part of this collection, and in many cases you can't really discuss the function of one without the other. Right? So a description of an ion channel is incomplete without also including its glycosylation pattern. And likewise, the description of what the glycans are doing would be incomplete without understanding what the underlying scaffolds are. And the kinds of biological processes that glycans participate in, there's a lot of overlap with proteins, right? They're involved in interactions with receptors. You know, they can be ligands for receptors. They can have structural roles. They can have biochemical interactions of their own. It's kind of all of the above.
Okay, so we shouldn't think of them at all in isolation, and we should think of them together and think about how they're complementing and working together in this broader ecosystem of what's going on on the cell surface. But for many decades, biologists did kind of ignore them, did look at the proteins and try to look at the lipids. Why is that? Why had these sugars been so hard to study? And maybe you can give us a sort of a primer on how you tackled that with the new chemistry you developed.
Yes. You know, it's funny. If you look back through kind of like the history of molecular biology, there was a time in the 1960s and 70s where carbohydrates had just as much attention, I would say, as other types of biopolymers. And people worked on them and tried to understand their biosynthetic origins, their structures, their roles in cell biology. But what happened was the molecular biology revolution, as it's often described. So this is the advent of recombinant DNA technology, you know, which allowed people for the first time to be able to clone a gene, express a protein heterologously, for example, take a human gene, put it in E. coli, use the E. coli as a factory to make the protein, and then study the protein. This was so powerful and so enabling, and people really gravitated to the areas of biology that could be accelerated immediately using these recombinant DNA technologies, and the glycans were not directly addressable with that technology because they're not part of the so-called central dogma. Right? So there's not like...
Gene to transcript to protein, right? That was what that was all based on.
Exactly. Proteins are primary gene products. So you clone a gene, you put it in an organism, it will make the protein. Glycans are not primary gene products, right? They are the product of hundreds of genes working together, right? Genes that encode all the different enzymes that are required to build the glycans. And so they weren't manipulatable using those molecular biology techniques. And as a consequence, you know, what happens in science is when there's a really powerful new tool, people want to use it, right? And science starts to move faster by virtue of having those tools, and so people gravitate to the areas of science that are moving fast. By contrast, areas of science that remain technically difficult, laborious, can't move at as quick a pace because the technologies aren't, you know, available. Those areas sometimes get left behind, and that's what happened with glycobiology.
So, you know, from the 60s and 70s where there was, you know, quite a significant bunch of people working in carbohydrates, fast forward to the 80s and 90s, and, you know, people just shifted away and towards other areas of biology. But not working on an area doesn't make it unimportant, right? And I think, you know, people who are thoughtful biologists recognize that obviously these molecules are very important. They're structurally complex, they're cell-type specific, they change in the context of disease. I mean, all the signs were there that...
I mean, isn't our blood types are based on these molecules, right?
That's correct, yes. If you're blood type A, right?
And that was discovered in the 1950s and 60s, right, at a time when people were putting a lot of thought into what cell surface sugars were doing. So, you know, during the kind of, I would say, the heady decades of the biotechnology industry and recombinant DNA technology, glycobiology kind of fell out of favor amongst biologists.
But interestingly, in parallel, it became interesting to chemists because chemists of the 1980s had developed some powerful tools to chemically synthesize large, complicated natural products. And this was a really hot area in organic synthesis was to pick the biggest, most difficult, complicated molecule you could find and then figure out how to make it from scratch, right? And that was like the Mount Everest type of science of the day was the bigger, the more complicated, like, you know, the more kudos you get for having been able to make it from scratch, right?
We're going to climb this mountain of making a really complex carbohydrate molecule.
Right? And so chemists started looking at these complex carbohydrates and thinking, well, that's pretty complicated. I wonder if we could make that from scratch. And it became sort of an interesting challenge in some of the very top organic chemistry labs in the world. So in a way, the center of gravity of glycobiology kind of shifted from the world of biology over to the world of chemistry in the 1980s and 90s. And those were the decades where I went to college and then graduate school and then did my postdoc. So, and I was a chemistry PhD student looking for interesting problems in synthesis, right? Interesting molecules to make. So that's how I was first introduced to the world of glycobiology was the challenge of the chemical synthesis, and that was my focus in my PhD work.
But all the while I was reading, you know, the papers from the biology labs around the interesting biological roles that glycans played, and I had in my head an idea that maybe I should go do a postdoctoral position in a biology lab, one of the small number at this point that actually focused on cell surface carbohydrates and their biological functions. And by the way, the 1980s was also the decade in which X-ray crystallography using synchrotron radiation sources became available to structural biologists.
And you can actually see the shape of some of these molecules.
You could see the shapes of proteins initially. So focus was on proteins. But what was interesting is there was a lot of interest in structure-guided drug development. People thought back then that if you could solve a structure of a protein with, you know, atomic resolution and, you know, you could see each bond in each atom, then you could just develop drugs, right? Just like that. And it turned out to be, of course...
Design a key to go in the lock, right?
It turns out that doesn't, it doesn't work quite as simply as that. Yeah.
Yeah. At the time, people thought all drugs would now be easy to make because of the availability of these structures. That turned out, of course, not to be true, but there were a few instances where it really was true. And one of those was in the development of the first generation of drugs for the flu.
And it turns out the flu is a virus that binds a sugar on the surface of host cells to kick off the infection. And this first wave of flu drugs were sugar analogs that were interacting with this sugar-binding receptor on the flu. So that was kind of a big story actually at the time, and it kind of pointed towards, you know, the fact that if you knew more about sugar biology and sugar chemistry, you might be able to make a lot of interesting drugs, you know. So that kind of also sort of sucked me in to the glycobiology part of it.
So tell us about, I mean, one of the challenges, if I understand correctly, was it's just hard to see what these sugars were doing. Like, where are they going? What are different kinds of these chains of carbohydrates?
And so it seems like the challenge was how do we just get a handle on these things and actually manipulate them without upsetting their actual function in the cell. Was that sort of at the heart of what you were trying to do?
That's part of it. Yeah. I mean, you know, what my experience was during my postdoctoral years. So this would have been like 1993, 4, 5, around that time. I had jumped ship out of a chemistry lab and into a biology lab. And that lab was studying the roles of cell surface sugars in the immune system. So it was an immunology lab essentially. And when I entered that lab, the biology was new to me. So I didn't have any experience as a cell biologist. So I was kind of learning from scratch these new techniques, and that's when I really saw the frustrations of glycobiology.
Because in the 90s, if you were studying proteins, you had really great automated commoditized tools at your disposal both to sequence the proteins and to image them in living systems. And those are two really important tools, like being able to just understand what proteins are there. Just sequencing the proteins from cells, we call that today proteomics, right? Being able to sequence a protein and know what proteins are there and how they change from this cell to that cell. That's an important thing to be able to do to study proteins. And then the other is to actually be able to visualize those proteins in a living cell or in a living animal and see how it moves around in its native habitat. And so the tool that people use for that is like the green fluorescent protein fusion, right? We have these genetically encoded fluorescent proteins that we can fuse at the DNA level to our protein of interest, and we express our protein as a GFP fusion, and then we can see where it goes, right? And that, by the way, that green fluorescent protein technology was recognized with a Nobel Prize in the 1990s, right? That's how impactful that was.
And those two capabilities, just sequencing the molecules so that you know what structures are there and being able to see where they are, those two things were not available for glycobiology.
So when I started my own lab as a new professor at Berkeley at the time, this is like 1996, the goal of my lab initially was to try and develop the tools for those two problems, right? How do you figure out what are the structures that are there, right? How do you sequence them, and then how do you see them in their native habitat and watch them, follow them around? Those were the two things that we kind of focused on.
And it seems like the immediate applications for that, I mean, you said that you had been in a, this was sort of developed with your experience in an immunology lab, but maybe the, we've been talking about these sugars as sort of a cellular identity is one of the things, one of the ways you can think about them. They're a way that the immune system can recognize different kinds of cells and whether this is a self cell versus another cell and what type of cell it is. And I'm sure there's more to it than that. But it seems like a lot of those early applications once you had these techniques to see these is that you could then start basically adding this to this growing field of immunotherapies where you could support the immune system in doing things like attacking cancer cells and the like. Is that a fair description of sort of how this went from now we can see these to what can we do with this understanding?
Yeah, I think that describes sort of like the first 20 years of my academic career. [laughter]
I hate to describe 20 years of important and difficult science in like two sentences. [laughter]
Well, you know, but I would say it took about 10 years for us to develop like a technology for the imaging part, for being able to see, you know, those sugars. And in fact, that was the motivation behind this type of chemistry that we developed called bioorthogonal chemistry, which now is kind of used across all of life sciences as an enabling tool for various applications. But its origin story was it was a tool that we developed to be able to visualize sugars on cells in cell culture and in animal models. And it took 10 years to kind of optimize that and get that all worked out. But once we had the ability to see the sugars and we could see how they were changing in different cell types, that's when we became interested in their role in cancer immune modulation, and that was another story that, you know, we started working in that area in Berkeley, continued that into my current lab at Stanford, and I would say that was kind of like, you know, the next decade of my research was focused on that. Yeah. And we did make a discovery, a fundamental discovery about, you know, cancer and immunology.
I hate to fast forward so much through these seminal discoveries, many of which have led directly to your Nobel Prize a couple of years ago, but I do want to get to some of the implications for neuroscience, and maybe there are some connections here between what you were discovering about the role of these cell surface sugars in cancer and what you and your team have been discovering in the brain.
So, do you mind if we jump into what is going on in glycobiology these days?
Absolutely. And it absolutely all does flow actually because the investment that we made in those first 15 years in developing tools to image the sugars and look at different classes of sugars, and we developed tools for different types that we could use to image them. And using those tools, you know, once you have those tools, right, you could look at the sugars in any context. You can look at them on cancers, you can look at them on immune cells. And it turns out you can look at them on the blood vessels that innervate the brain.
And so what happened was about six, seven years ago, a graduate student at Stanford who was in the chemistry department, she was her first year student, and in her first years they do rotations, lab rotations. So she rotated in my lab and she developed some reagents to image sugars, and the application at the time was around cancer. Okay. But then in her next rotation she went to a neuroscience lab, and that was the lab of Tony Wyss-Coray, who's a Stanford professor here, and he is kind of a leading figure in studies of aging in the brain and in how the blood-brain barrier changes as we age. And he had a project going in his lab where his students were looking for molecular changes in the brain endothelium. So these are the cells that line the blood vessels that are all innervating the brain, and they are the cells that constitute the blood-brain barrier.
Right? So these...
This is basically, just to make sure that all of our listeners are following along, this is basically the idea that the brain is supposed to be, and the nervous system are supposed to be sort of a specialized compartment. We don't want immune cells getting in there too easily. We don't want toxins and things from the blood able to access the brain. So there's this idea that there's a barrier that carefully controls the kinds of things that can get into the central nervous system. That's sort of like a big picture there.
That's exactly right. And the brain is full of blood vessels. In fact, someone told me that if you take all the blood vessels and put them in a big line, it's 400 miles of vasculature. So you could drive from Palo Alto to San Diego on the blood vasculature in your brain. It's a lot.
I think because your brain needs like continuous supply of energy and oxygen and everything, right? You know how it is. If you can't, if you don't breathe for a couple minutes, your brain, you know, you pass out, right? Your brain shuts down.
And that oxygen has to get in there, and your brain needs glucose, right? If you have low blood sugar, you can't think straight, right? So the blood-brain barrier keeps the bad stuff out because it's such an important organ to protect, but it also has to carefully regulate who comes in and things. A lot of things have to get in there, right? So it's not just a barrier like a physical barrier. It's a highly regulated barrier with transporters that bring in the good stuff and pumps that pump out the bad stuff. And it was known that as we age, our blood-brain barrier becomes more leaky.
