Chemistry to Build New Imaging Tools & Probe the Brain with Anand Muthusamy

Welcome to the Axial Podcast. Axial is an early-stage investment firm based in San Francisco. We partner with great founders and inventors investing in early-stage life science companies, often when they are no more than an idea. Axial is fanatical about helping the rare inventor who is compelled to build their own enduring business.

Hey, Don.

Hey, we're recording. I've been talking for like 5 minutes, you know, praising Luke Levis and you know, talk about, uh, some of the decisions you have to make. Um, but maybe we're really excited just to talk about your journey so far. You've actually, in a short period of time, like, have had a lot of diverse experiences, and then you have some really important decisions now. And so I want to touch base with you right now, so then we'll talk in 2 years, and then see what you're doing. And then over the next, you know, 10 years, I'm sure you're gonna have a massive success, and then hopefully we'll invest in those companies. But maybe you could just start off with a brief introduction yourself, and then we can go from there.

Yeah, absolutely. Yeah, uh, Josh, thanks for having me. It's a pleasure to be here, uh, pleasure to have gotten to know you over the past year, and, uh, really happy to do this. Um, yeah, about myself, the rundown is I was born and raised in Houston. I didn't always think I'd go into science. I think a lot of, like a lot of little kids, I loved basketball, football, wanted to go pro, quickly realized that wasn't gonna happen, but I always loved science. My dad, he just retired a year ago, but he worked at Royal Dutch Shell and their R&D headquarters is in Houston. So as a kid, he would take me to the labs. I just marveled at the facilities, the craftsmanship there, seeing the glassblowing. That was kind of a lost thing in chemistry and I was having a glassblower in your institute. But I just marveled at that, and I think that probably planted the seed for liking the lab and science. So fast forward, you know, I got into policy debate, really enjoyed philosophy, wanted to do science and philosophy, went to Penn for undergrad because they have a good liberal arts and science. So, you know, whenever I can't choose, I try and permute. So really enjoyed both, but Working in James Peterson's lab at Penn is what got me hooked on science. James happened to train with Dennis Jockrey at Caltech, physical organic chemist turned chemical biologist. So it was kind of that flavor of science. And yeah, we're making tools to study protein function. So we'd create fluorescent natural amino acids, incorporate them into proteins, watch them fold, kind of like a motion tracking suit for proteins. So that got me my first taste of science. I knew I wanted to go into research. And that's the quick origin.

Cool. And what's always fascinating is just like that starting point of your scientific work. It's steeped in like philosophy. It's steeped in like really wrestling with like just getting to know you and talking to you about your story. Just like you've been very methodical sometimes about decisions you've made, seemingly from the outside looking in. And so like maybe we can start off at Penn. So you started at Penn, like did you get to college and realize, okay, I'm gonna do science and philosophy? Did you know that going in? Or especially Penn, did you get roped into finance? Did you get kind of sucked into things? Like what was it kind of that first few years of how did you determine that this is the path you wanted to take?

Yeah, I did not get roped into finance. I took zero Wharton courses. One of my best friends, David, was a Wharton student. So I hear stories from him. But no, I was set on the science. I thought I'd go into medicine. I thought basic research would be a jumping off point to go into medicine eventually. So philosophy, matters of purpose of life, quality of life, medicine is the natural application of that. But yeah, once I got at the bench, being able to create especially creating tools, you feel in tune with nature when you can create tools to manipulate and hear from nature. So that got me hooked. So didn't think about finance. I wasn't distracted by that, but thought medicine, then got into research.

So then I always forget whose lab you were in college, but like—

James Peterson.

James Peterson's lab. And so any kind of initial first lab experiences, like, was that the first time? I guess because your father was an engineer, so you got some— you got exposure to, like, engineering and lab work. Any kind of funny lab experiences initially? Did you break anything? Did you get yelled at?

No, no, not yelled at. I hope so. The grad students I worked with As an undergrad, your experience is going to be shaped by the grad students, and they were very generous with their time. So John Warner, Lee Spate, Rebecca Wisner, these three really mentored me and another undergrad, Colin. So I remember our first summer just learning how to make LB was the first one. So you get a training wheels project to make a protein, which is probably one of the most satisfying things an undergrad can do when they first step foot is you make a protein. Um, so no bad experiences to start with, fortunately.

Yeah, I mean, I don't know, I think I spent a year trying to clone a thing called talons. It was okay. Yeah, precursor. And I was like, it was so hard to do. It was like one thing that was scarred on my face. And then I remember a Bunsen burner, I burnt like part of a bench, so— and then I almost broke an ultracentrifuge. You know, like, you know when you use a centrifuge And like, you think, oh, it's just a centrifuge, you know, I'm gonna put things in it and then spin it. And then, no, no, no, no, you need to make sure it's like perfectly balanced.

That's right.

Yeah, I almost broke this interest because I didn't think— I didn't know. And one thing is always amazing, I don't know how lazy your lab was. When I was in college, you join a lab and it just lets you do anything. And it's kind of insane, almost, in retrospect, just to let an 18-year-old do anything, uh, very little.

Yeah, well, I think that's good. It's good to have that, that playful atmosphere. It's good to break a few things. Yeah, exactly. And learn that way.

Um, okay, so you're at Penn and doing research. Any influences, any side projects, any kind of courses you took that helped you become a better thinker? That like maybe put you on the path of maybe being a little more entrepreneurial than most scientists?

Yeah, you know, back at Penn I wasn't thinking about entrepreneurship whatsoever. It wasn't until, you know, coming to Caltech a few years ago, and we can get into that, but at Penn it was just pure science, just the love of the craft there. And I would say the most influential First course was Ponzi Lu, who runs the biochemistry program there. He's a physicist turned biologist, so that's a recurring theme for me is I have mentors who are physicists who got into biology later. Ponzi has a course called Chem 22, which just forces you to think. So for example, the first project was back then there was a report about an organism that used arsenic instead of phosphorus. They said, you know, you can incorporate arsenic into the DNA backbone. And Ponzi's assignment was go figure out, based on the reagents they used, whether there was enough trace phosphorus to allow this organism to live off of phosphorus, basically invalidating this huge headline-grabbing study. And it turned out that that was the case. So that was just a good lesson, a good little scavenger hunt. And he actually sent us on a scavenger hunt all over Philadelphia to collect world science artifacts. Philadelphia has a lot of great museums and culture. American Philosophical Society is there. So yeah, Pomsey made you go outside the lab and explore culture and science.

Yeah, so can I get some— so one thing that makes you pretty unique is just the philosophy background you have. I've shopped philosophy courses and I've read a lot of philosophy books, and sometimes it just They're almost too hard to understand. But one, why did you start taking philosophy courses? I think you were a secondary or co-major of philosophy in college.

Not quite, uh, but, but took, took many philosophy courses. Um, yeah.

Why? And, uh, cause it's so hard. A lot of these classes are pretty hard. You know, it's like in terms of college courses, it's gonna be like linguistics. And then philosophy. So, uh, why'd you take those courses, and kind of what kind of influence have they had on you as a scientist and beyond?

Yeah, so, you know, uh, philosophy of science courses, logic courses, they teach you, you know, formal thinking. Uh, that, you know, I got into that, and I found that I didn't find the analytical philosophy as interesting. I wanted to read stuff about the human experience. So I took a bunch of philosophy courses, and then that led into taking medical anthropology courses. A lot of the philosophers I'm interested in are these post-World War II French philosophers, and you don't read those in the analytical classes. You read those in medical anthropology. So I just love thinking about human experience, like I said, quality of life questions. Once we've kind of ascended to the point where material needs are met, And that you think about kind of purpose, uh, and, and, uh, a lot of people in Europe after World War II, after they saw the heroes of war, thought about, you know, what, what is going on. So they have a lot of, you know, interesting lessons there. So down to medical anthropology. Um, yeah. And then, you know, that it kind of connects back to my interest in neuroscience because I think we're, you know, with tools to read and write physiology, we, you know, put the onus on really philosophy, ethics, culture to decide what we do with those tools. So that's kind of the direct line from philosophy to medanthro to, okay, we have great optical tools, we need the right tools and physiology, uh, what should we do with it?

Cool. More into then kind of going to Janelia, which is like, hey, you're in college, you're doing a lot of chemistry, a lot of science, Yeah, just in neuro building tools for neuro.

Mm-hmm.

Um, how did you determine to do Janelia, join Janelia, join Luke Levis's lab, um, versus, you know, PhD program?

Yeah.

Kinsey, get a job, you know, as he did. Near the end of college, it's always like, I don't know, it's very tough, very— I, I personally, I just interviewed for a lot of companies just for fun. Kind of knowing I didn't really want a job. I just went to grad school, probably shouldn't have done that, but like, like, did you have like— so like what I'm trying to say, I didn't really have an approach near the end of college. Did you have like a logical framework to say this is exactly what I'm going to do, or did you just kind of rope into Genelia?

No, I, I didn't have an approach. Honestly, I was a little bit sad towards the end of undergrad because I didn't know exactly what I should do, what I want to do. Um, and, you know, I did science, met anthro, was kind of getting away from medicine, wasn't sure how to start a career in science exactly, just knew that I really liked the intellectual life. I liked science, like working with these folks. Um, so, you know, I was just searching, soul-searching, exploring ideas. Um, and then, uh, it's a funny story how I even learned about Jamila and got to Luke's lab. Uh, it was probably spring my last year in college, and Luke happened to give a talk at a chemical biology symposium. Never met him, had never heard of Janelia. And back then it was called Janelia Farm Research Campus, so I just had a total misconception about what this place was because it had farm in the name. So I was like, you know, what goes on on the farm?

Luke says when he first interviewed at Janelia, his wife thought it was a cult or something.

Yeah, it's off in the woods a little bit. Never heard of Ashburn. You know, I, so I thought, okay, maybe backup plans, I just kind of stay around Penn and do some science and figure things out. But Luke gave a talk and it was just brilliant. He spoke about how he as a synthetic chemist makes tools for neuroscientists. And he joked about how, you know, he likes to ask about their problems, whether it's life or science. And that just, from there, like, that got me hooked. So after he gave the talk at the symposium, we're in the in the hall in Chem 73, and I just asked him a bunch of questions about what he does. And, you know, James said, hey, why don't you think about working with Luke? I said, oh, I didn't think about that. Yeah, maybe I should. So we sent him an email. Luke said, yeah, come down and interview. And so offered a job and started that way. So if I didn't show up to that symposium, I, you know, different life path.

Wow. It's always those like lucky moments where it's like Yeah, I went to grad school at Berkeley, and like, I almost— I, I missed that. I could— I was like, I got— I was like, got on that flight just a minute before it was going to take off. Like, I was so late. So I think those, in retrospect, those like lucky moments, or those small moments that like seem, uh, irrelevant at the time but then are really important. Um, you, Janelia?

Of Luke.

Luke is like the best in imaging probes. That's right. Like, did you kind of know what you're going to work on going in? Or I would assume you got roped into a lot of projects given Luke's kind of role at Janelia.

Yeah, I knew that we'd be making tools and I knew that it'd be a lot of synthetic chemistry. So I was excited by that. And in James's lab, I'd been making fluorescent unnatural amino acids. But to incorporate them into a larger biophysical assay. So, and then towards the tail end, I was learning a little more synthetic chemistry, learning a little bit of photophysics, some computational chemistry to like model excited states of dyes. So I'd been getting the tools to really run with them in Luke's lab. So I came in gung-ho to learn synthetic chemistry. Luke had a, you know, cadre of great synthetic chemists. And one cool thing about engineering is, you know, it's— I love grad students and postdocs,, but it's also great to be around folks who are veterans, 10 years in industry, right? So Luke had a few of those guys, and, uh, it was so great absorbing their synthetic knowledge. There's a lot of fingertip feel there. So, um, yeah, Luke had, you know, the whole rhodamine, uh, JF-DYE program going, and I just jumped into the fray. Yeah, so yeah, the background you set nicely, and I really enjoyed your blog post on this topic, uh, that, you know, Luke and John essentially said, let's revisit biochemistry It's been made, you know, 100, 150 years ago. We now have, you know, Buchwald-Hartwig coupling. We have palladium chemistry. We can make nitrogen-carbon bonds at will. So let's revisit these dyes. So John and Luke paved that way. They were the first to incorporate this palladium chemistry into making dyes. They figured out this azetidine substituent that increased the quantum yield of the dyes. And when I jumped in, The question was, can we make them brighter? Can we make more colors? And critically for live cell imaging, can you do no-wash experiments? Meaning, can you make the dye fluoresce only when it gets captured by its HaloTag? So it's, so really the question was, can you, you have this great scaffold, can you make a variety of those colors? Different biologists have their own setup. They may want this palette for this particular imaging experiment. So the The goal is to make a real toolkit, and then to share that toolkit widely with the academic world. So I was working with John learning synthesis and made several substituents on this rhodamine dye. And I'd say one of my unique contributions was incorporating computational modeling. So there's something called time-dependent density functional theory, just a a fancy way of modeling molecular orbitals. And you could actually use TD-DFT to predict the transition states, and you could predict, okay, if I add the substituent to this ring, I'll blue shift the molecule, I'll get this absorption, I'll get this emission, I'll have this Stoke shift. So that was a real— it just felt elegant to me that I could kind of use chemical intuition from the textbook plug it into a piece of software, get a prediction, go make the compound, and then just iterate. So yeah, that fine-tuning rhodamine paper made a whole palette of colors, you know, 10-nanometer steps across the visible spectrum.

Yeah, yeah, the paper you publish, I'll link in the description, just like you essentially can get any dye across the spectrum and you can synthesize it and it's pretty cheap. So, um, you know, Luke is still— his group is still, you know, shipping 100 of these probes to a bunch of labs and groups every week. So I think the work you did at the time is still like having an impact on all these people's research. Luke was telling me he'll go to a conference and people come up to him, be, oh, thank you, you helped me so much.

Yeah.

And he's like, who are you? I never met you before. So like It shows the power of new tools. You can create something that's generalizable and useful. It can just have an impact for years and years and years and just permeate every type of work. But at Janelia, I mean, very lucky. What was the kind of— one cool question is just being young at Janelia. It's in like Northern Virginia in the woods. Yeah.

Mm-hmm. Was there—

are there other young people there? Like, or was everyone— were you like the youngest person there?

I, I may have been the youngest or second youngest that I knew of. Uh, yeah. Like, like I said, a lot of senior scientists, a lot of folks that were, you know, grizzled veterans from academia or, or industry that came to this new place to explore. So I, yeah, in our, I think in our group I was the youngest. My entire time there. And at first that's hard socially coming from Penn, being in the college bubble, parties every day, to this quiet, just serene place off in the woods, total shift in culture. Frankly, a little bit harder for us again socially because these folks have kids. I'm a 21-year-old, little uncommon. But, you know, as time went on, I made really good friends with people across generations. And, you know, the housing world was kind of cool. You have these townhomes and apartments right on campus, but kind of behind the hill so that you, you don't see the actual research lab. So you're close but also perspective-wise far away. And, you know, a lot of folks that I worked with lived in the same place, so we'd go to the gym together, you know, on Sundays we'd have a workout routine, go get kebabs after. There's like a little, little home there. So, uh, yeah, hard at first, but then very rewarding, uh, uh, getting to know these folks. Um, yeah, so I, I— it was a very unique experience.

Yeah, it makes a lot— I mean, there's also— you went to Caltech, and Caltech is like a monastery of science. It's like Janelia is kind of cultish in terms of like, not the group, but just where it is. It's out in the woods, you're kind of alone. Caltech is definitely a monastery where it's like small. You can walk across the campus in 10 minutes. So while at Janelia, I think you were there for like maybe 2 years or longer. Yeah. So like, how did you end up making the decision to go to grad school? Did— versus staying at Janelia is a place you could stay at for a long time. You know, why did you end up saying, hey, I wanted to get a PhD?

Yeah, so I— at Janelia, I got introduced to neuroscience. When I first entered Janelia, I didn't understand circuit neuroscience. I thought behavioral science was silly because you couldn't quantify things hard, uh, like you could in chemistry. And after the 2 years, my view just flipped. You're around all these great biologists and they're asking, I just know how to form insightful questions. And I thought, yeah, I'd like to do that. Um, and then at Genelia they had a lot of great, you know, extracurriculars. I learned, you know, Python, and they have a robotics group, they had a VR group, uh, you know, just playing around. I thought, well, you know, grad school can be like this where by day I'm doing research and by evening before dinner we're playing around creating little VR worlds. And yeah, I'd like to do that for 5, 6 years. Um, and then I, you know, formed my own ideas about what I like to do in neuroscience, saw neuro as a wide open field for a toolmaker, and thought, yeah, let me, let me get serious about this. And, uh, I, I think you can nail the head that Caltech's a monster of science, and me liking that environment, Janelia drew me to Caltech. I knew that's where I needed to go regardless of who I worked with. That was the place.

Cool. Okay, so maybe also on this point, it seems around Janelia a lot of your current scientific ideas and a lot of the frameworks you use were kind of formed during that time. Like, what kind of big problems were you trying to solve before going to grad school? Like, did you formulate a list of problems you want to work on, or was it more around like, I want to do something in neuro, maybe build tools? Was it like, did you just like, how did you, what was the process for you to determine like, oh, this is stuff I'm interested in and I want to do a PhD in it?

Yeah, yeah. I thought, you know, I mentioned, you know, interest in science, philosophy, and then anthro, didn't know how to quite tie it together, and neuro just seemed like the perfect field to tie all that together. And so I knew if I do anything in neuro, it would be very fulfilling. I can imagine doing it the rest of my life. Life. So that was sufficient to go down that field. And then the specific problem that I was thinking about was, you know, how do you quantify physiology and make squishy concepts like emotion a little bit more tractable for quantitative sciences? Uh, I didn't have any grand solutions, but I knew that was an interesting problem. Uh, and, you know, folks like David Anderson at Caltech are, you know, interested in coming up with frameworks for thinking about emotion. So I like that Caltech tried to create a quantitative approach to something that's previously intractable. So yeah, just knowing that neuro would be the field was enough, and then knowing I wanted to make tools to quantify was the, you know, kind of final point to go to Caltech.

And then last question is, in terms of like, okay, you have a philosophical bedrock to some, a lot of decisions you make. But then, you know, Luke Levis, he's Mr. Useful, you know, he's all into utility. It's almost like a juxtaposition between, you know, thinking about neuroscience and chemistry from a philosophical lens versus Luke's group, which is very much like, let's do things that actually can make an impact right now. Um, what kind of When you look at your work now, what kind of like impact can you maybe trace back to Luke, trace back to even Penn, and that kind of lessons you've picked up along the way?

Yeah, absolutely. Yeah, I think the tension you mentioned there is real. I think it's nice to have that tension. They feed each other, right? You can have the greatest concepts, but you need to make tools to even get the data. So I like that tension. I like holding that. And you know what I really got from Luke was how just make tools that just simply work. You know, don't be content that it works in a toy system. Make a version, share it with a biologist, an expert in the field, get the feedback and iterate. Almost like lessons in product design. You don't really think about that in science, but Luke thinks about that. He has his own philosophy about product design about how, you know, to make a tool that just seamlessly integrates into someone's workflow. Um, even down to like the labels, you know, one thing he told me was, you know, uh, biologists are spending so much time with their experiment, they really want to trust the tool that they're using. So it's worth, you know, the minute to make a really nice label with very clear label, like, oh, wait, nanomoles and so on. So you just trust it from the get-go. I thought, you know what, that makes a lot of sense. It's not something that a lot of scientists would think about, but makes perfect sense.

Cool. I think that— okay, so we get to Caltech. Yeah, get to California. Was it first time in California, or first time living in California? First time living in California. Pasadena is beautiful. Um, okay, Caltech campus is pretty nice, honestly. It's like super nice. It's just so small, and I've never met a normal Caltech college students. I met normal Caltech grad students, professors, and whatnot, but kids in Google College— I have friends who go there— they're often a little out there. It's a little different. So like, one thing I love about Caltech— I used to love— I, I remember going to the Thomas Hunt Morgan Library. It's like, well, this is pretty cool, you know, very historical. But, um, you know, you're, you're in this environment where there's a lot of great researchers You have Francis Arnold, Stephen Mayo, you have Michael Levitz, like you have a bunch of cool people. Like, how did you— did you— yeah, how did you figure out who you want to work with? It seems like you had more for like a values reason and not so much like a master plan to join one lab. And so you land at Caltech How did you figure out what to do?

Yeah, so I was, you know, admitted to the chemistry program actually, uh, not biology. So I was admitted to chemistry, uh, I thought I'd go into chemical biology since that's what I've been doing. And I rotated through groups, uh, including Henry Lesser's group, which is in the biology division. And the nice thing about Caltech is that once you get here, you can join any group as long as you have a good rapport with the professor. So you can be and then into one division, then go join another, if that makes sense. So that's one real nice thing about Caltech. And really, I did the rotations, I looked around, I thought, do I have a really good transformative idea for chemistry? I thought, not really. I feel like I've kind of run my course in chemistry, but biology just seemed wide open. And for me, it felt like I could make the most impact there. So again, love neuro. Henry was a great match because again, he's a physicist by training. He studied physics and chemistry in undergrad, eventually got into neuro. He has a track record of turning good chemists into great neuroscientists. So it was just, you know, a perfect match in terms of values, topic, mentorship style. So I got very lucky there.

Cool. I didn't know that. I didn't know that part before. I didn't know. Okay, so you're in the Lester lab. Like, when you think about the research you did in grad school, you've done, like, did it, like, there's always some growing pains, I think, becoming like a professional scientist and having to do research without supervision. Did you find that came easy because you are generally already, and you kind of, okay, I've done this already, and and I could just run my own work. And then how did you also make the transition from, let's say, being told what to do to then formulating your own hypothesis?

Yeah, so I would say, you know, topically the transition was tough. You know, personal process-wise, it was easy. So, you know, Luke was great in letting you run free with your ideas. We meet like once a month, right? So it's, you meet as needed, kind of set the direction and go. So Harry learned a lot of independent process and John Grimm was just a, you know, a mensch of a mensch in terms of how efficiently he worked. So I learned a lot of that. Now topic-wise and experimentally, it was challenging at first in a good way because I had zero of the skills. I didn't do any imaging with my own hands. Cell culture, just physiology, bioimaging, animal work, all brand new to me as a chemist. So there's a learning curve, but fortunately, you know, we had some senior scientists and great staff that showed me the ropes.

So like, you bring a chemistry background to like, I don't know, a neuro lab. Did you have a nickname? Were you like Mr. Chemistry? Were you like the go-to person for like something specific? Because you bring a new— you brought a new expertise to the lab that maybe wasn't common. But, you know, any kind of anecdotes or like, did you— did people always ask you questions about this particular topic or something?

Yeah, I'd say, you know, there's a great senior scientist, Bruce, you know, who would ask me every single chemistry question he had. So he considered me the resident chemist. He's a physiologist by training. So yeah, he would just pepper me with chemistry questions. And yeah, but it was very nice to have that difference and then eventually learn their world.

Yeah, it's also— there's some sort of value to like being an outsider and coming into a lab versus just being like another biologist in a biology lab, you know. It's like you could be a physicist coming to a biology lab you could bring a whole new element to the research and you end up being the go-to person for a lot of different projects. There's an argument to be like being the outsider joining the lab. Some growing pains though, like you alluded to, it's like learn— steep learning curve. And so you're in this neuro lab, what kind of— what was your project? Did you have to do some really— because neuro That's some of the harder experiments to pull off. And you get to do a pinch-clap. There's some like hard neuro. I have friends who like image brains and stuff, and it's like brutal versus like just a PCR or even a synthesis sometimes. So what was your project on? And then any kind of like, how did you pick up these skills to then, you know, do something in neuro?

Yeah, absolutely.

Yeah.

So the project was to make biosensors for neural drugs. So Henry had done a sabbatical with Loren Luger at Janelia picking up protein engineering, biosensor design. And there's a big movement in neuroscience for optical physiology, meaning instead of sticking electrodes in the brain and recording from tens or hundreds of cells, what if you encoded a protein that lights up in the presence of some analyte? So it could be calcium for neuronal activity, it could be neurotransmitter to look at neurotransmission, it could be a drug to look at metabolism, flux, and organelles, and so on. So if you can do that, then you can record from many neurons, you can even record intracellular events that you couldn't with the electrode. So that's the field I was jumping into, and the project was to take what we call a periplasmic binding protein, looks like a Pac-Man or Venus flytrap, and evolved that to bind a variety of other ligands, uh, specific neural drugs, you know, more specifically opioids. So I ended up making these fluorescent sensors for methadone, fentanyl, morphine. Um, so it was protein engineering at first, uh, and, you know, having the Arnold group nearby meant I could go ask people, show me, you know, all the tips and tricks you have. Um, and then having, you know, the Guadnaro lab nearby allowed me to pick up, okay, show me how to make a virus, made an AAV, infect an animal, took her course in optogenetics to learn how do you get a probe into a brain and record in a freely behaving animal. So the way I picked it up was essentially just asking people or taking courses. And Caltech is so tight-knit that it's very easy to just walk across to another building and really see someone do it with their own hands. Yeah, so the project, yeah, making sensors. The two applications, one is at a subcellular level, so there's an increasing appreciation for what we call inside-out pharmacology, what others may call location bias in GPCR signaling. So it turns out that GPCRs can be activated within endosomes and Golgi. They signal from within these organelles, not just at the plasma membrane. So people are really rewriting the textbook on how GPCRs signal. So we're interested in the ligand half of that. Do ligands accumulate in acidic organelles? There's some early evidence that some drugs will accumulate in synaptic vesicles and be released for the synapse in an activity-dependent manner. So we're interested in that kind of subcellular pharmacokinetic question. So we looked at that, and I found that methadone accumulated in organelles, and there's some consequences to be figured out what that means for opioid receptor signaling. So that was the molecular level, and then I got really interested in the animal level. So I mentioned I thought animal behavior is too squishy, uh, and had the pleasure of working with a great grad student— now he's a fellow at Princeton— Matt Rosenberg, who taught me everything I know now about animal behavior. So I was interested in this question of, okay, how does an animal's behavior track with drug time course? You know, an assumption and a tenet of psychology is that when you administer a drug, it sets off a cascade of events, and that trickles up to affect behavior. But also there's some variance in that behavior, right? Not everyone who takes a drug develops substance use disorder. What explains that variance? They're really interested in that question, you know, to what extent behavior is stereotyped by a stimulus. So we essentially developed an assay to allowed animals to run through a maze and assess their deficiency when they take an opioid. So the reason to do that was the classic methods of looking at drug abuse was to look at self-administration. Does an animal take a lot of the drug? But that doesn't really match up with real human substance use disorder because it's not just that you like the drug. The problem is when you like the drug so much that you disregard other parts of your life. So you need to put that into a context where you take a drug, it's not just you like it, it's you disregard friendships, family, job, yourself, and so on. So we wanted to see how we could give an animal a task that was ecologically relevant, and that was to forage for water. So we look at how opioid administration would disrupt this natural task. Zero training of the animal, just let it do it, let the mouse be a mouse. You know, so those are kind of the two rewarding applications, subcellular pharmacokinetics and then animal behavior.

Yeah, I think that kind of you laying out the research you do or doing lays out how like it spans everything from protein engineering to behavioral analysis and then just how you learn all these different components. But maybe we can start from the beginning. Like, why are those bacterial proteins periplasmic? I always forget how to pronounce periplasmic.

Periplasmic binding proteins, or PBPs.

PBPs, exactly. Why did you focus on— because this kind of gets into the broader value of proteins in nature and then using them for applications. CRISPR is probably the most famous one. But how did you go after these periplasmic binding proteins? Why did you try using them? And why in the context of as an in vivo biosensor? Because it seems from the outside looking in, non-intuitive. But there's probably some lessons that could be extended to like other bacterial proteins and using them for other things as well.

Yeah, definitely. It's a great question. So an ideal biosensor will bind the analyte you care about and not do any crosstalk with other stuff in the cell. And ideally, you want a protein that does a lot of conformational change so that you can capture that work. And PVPs have been known, you know, for decades now to have this very stereotyped large conformational change, 10 to 30 angstrom movements, sort of like a Pac-Man closing. Venus flytrap closing. So the structural biology was there. So a lot of great bacterial stories, folks in the '70s and '80s solved a bunch of structures and just put it out there. And then later you figure out, oh, we can use this now that we have protein engineering tools. You can use it for a completely non-natural purpose. So that theme is present here as well. And then why PDP specifically? Well, I mentioned Henry did a sabbatical with Loren Luger, and he had pioneered using PVPs to detect neurotransmitters. So the seminal work was Johnny Marvin's 2013 paper for iGluSniffer, a glutamate sensing protein. They take a PVP that binds glutamate in nature and they tune it up, couple the GFP, it lights up in a dose-dependent manner. I'd say the non-intuitive leap that we made in the lab that we, you know, Henry drove was saying that, well, we want to detect drugs that nature's never seen, right? Bacteria never cared about methadone. It was synthesized in the '40s in Germany. How would they have evolved a starting point? Well, I mean, this is where a lot of the protein engineering intuition comes from, which is proteins aren't platonic ideals. They're kind of more like clay. You can shape them. So we said, let's throw a bunch of drugs at a bunch of PDPs that bind an amine. So an amine is common in neural drugs. You know, it's maybe a little bit outdated, but lipinsky 05 makes it such that they're, you know, kind of hydrophobic, they have amines. So we found an amine-binding protein that binds choline, evolved to bind nicotine, and then we said we want to bind a variety of neural drugs, so let's just throw libraries against each other. As long as you have a starting point, I think there's a lesson from Francis Arnold's work, as long as you have some starting point, we have 1% of the activity we want. Take it and start pushing on it, start evolving it, uh, so that we just, you know, implemented that playbook. Um, yeah, so, so PVP's nice conformational changes, non-intuitive part was making them bind classes of ligands that they would never have seen in nature.

Yeah, kind of need, yeah, some dynamic range of conformations, and then like it's never been exposed, then you want to measure it. And so then you can have an orthogonal system, a truly orthogonal system. So like, you know, you're wrapping your thesis right now and you've done a lot of work in terms of in vivo measurements. Could you talk how the imaging field is making a shift? Let's say imaging originally was on dead cells. Right now the cutting edge is live cells, and it's going to get to live animals. Could you talk about in the context of the work you do, maybe more broadly, of like what opportunities exist to image in vivo and, you know, kind of why it's important and things like that?

Yeah, absolutely. Yeah, you know, ideally any biologist would want maximum context. For their measurement. They only studied fixed cells because that was tractable. So there's a confluence of so many advances and now commodities from, say, Thorlabs that allow you to study in vivo. So now it's routine in neuroscience to— you can order a virus with a convectocore, you can order optical elements from Thorlabs, Jackson Labs will even do all the surgery for you. You can actually do all this even virtually. But yeah, preserving context is key. Neuroscience is just an exemplar because, you know, neuron in a dish is very different from a neuron in a circuit with glia around, right? So preserving that context is key. But then when you preserve the context, you need to make sure your tools are minimally invasive, biocompatible, and so on. So that becomes an engineering question, and protein engineering and the chemical synthesis reaction toolkit allows you to kind of weave around the constraints that biology imposes. So, you know, of course everyone admires, and I admire, Bertozzi's work in bioorthogonal chemistry. That's just such a beautiful concept for working in, say, aqueous environment. So we're getting more tools that allow us to go in vivo, and I think any biologist would have wanted to always start their leaves and have the tools Yeah.

You've also alluded to this concept of like reading and writing neural activity. And so, you know, the work you do now is kind of more on the reading side, maybe reading, like measuring drugs and metabolites in vivo. But, you know, the holy grail is to ultimately build out a set of tools that can read measured complete neural activity in a live animal and then do something about it, you know? Yeah. Um, you know, could you talk about, like, this is not— this is probably not the purview of your re— all your research you do in grad school, but maybe something else you do in the future. But could you talk about maybe the, the history and then the current state of the art? Because there's a lot of different companies trying to figure out ways to identify dysfunctional neurons and then fix them, or ways to develop optogenetic methods to measure neural activity and then try to, you know, kind of tilt them to certain function.

But like, what's kind of the history of the field and what's kind of the current state of the art Yeah, so history is, of course, you have maybe the first was deep brain stimulation electrode implants that you could justify in some very extreme cases for humans. And then for science, I think it really entered widespread use with, of course, optogenetics. Desirée Glab and others are able to read and write in a closed loop so they can read the code, discern it, and then write it back in. So in the lab we can do that already. Of course, in humans that's far away because we don't have a method to get this widespread expression in humans, and we probably don't have the ethical reason to justify that kind of experiment. So at the moment it seems really restricted to the lab, but there are ways, say, you know, Mikhail Shapiro's interested in ultrasound. Yeah, uh, so that is a less invasive way of interacting with neurons. Um, so, you know, does— is the Neuralink dream of reading and writing in humans possible? I think it's possible. Uh, what's the exact technical route? Uh, I think that's being, you know, that's shaking out right now. Um, but I think, you know, the nice thing about circuit neuroscience is it allows you to identify some targets. So when you can read and write in an animal, run hypotheses, get causality. I poke at this, I get this outcome, and I can inspire going after certain targets with more traditional therapeutics in humans. And then one thing I've been really interested in is, you know, you can— in the periphery, you can read and write a lot more easily. So the continuous glucose monitor and insulin pump being in a closed loop, which we call an artificial pancreas, is a real-world FDA-approved closed loop to read and write physiology. So I'm really interested in extending that concept to other metabolites, other therapeutics that'll be in that loop.

Interesting. Okay, so that's really helpful. So you're wrapping up your thesis. How long is it right now?

I think, um, as far as completed, it's probably like 50, 60 pages and probably, probably 200 or so once it's all done.

Oh yeah, a long way to go, man. Oh man, good luck. Be a tough few months. Um, yeah, while you're doing your wrapping up your thesis, you're exploring a bunch of options. That's right. Um, I mean, you know, startups, extending your research, bunch of— a lot of potential pathways. We can talk about like what excites you the most. You know what excites me? I'm really excited about imaging dyes for super resolution microscopy, but maybe from your perspective, what are some of the big problems you want to work on after grad school?

Yeah, so I think one in imaging specifically that I'm kind of obsessed with now is hyperspectral imaging. So that's where, you know, you are able to, in principle, image many colors at once.. And the way you do that is even if you have overlapping dyes and you only have so many filters, as long as each dye has a unique spectral signature, you can kind of see the absorption profile, emission profile, and get the signature for each. So as long as even if they're overlapping in their profiles, as long as they're offset, you can discern, you know, okay, that dye is— I labeled this protein. So Scott Fraser at USC has nice work doing, I think, 5 colors at once. And you can imagine you have color, you have lifetime, and Stokes shift. And those are 3 axes to play with. And I don't see why not doing tens of colors at once. In principle, it's possible with hyperspectral imaging. In practice, it's very messy. So I'd love to create tools that allow you to do that. And then you can ask any biologist, if they'd love 5, you know, 2 players at once, or 10 or 20 or so. I think most biologists would love more photons, more colors, more players that they're monitoring at once. So that's one in imaging. We're interested in CNS therapeutics. So, you know, we're getting the first inklings of antibodies that may or may not work in humans. And we're really increasing our toolkit in terms of doing imaging. So now I think there's an imaging marker for like TREM2 for Alzheimer's. So I'm really keen on human data for CNS. I'm really convinced by this book that John Allman and another physician, a friend of his, wrote. It's called The Human Illnesses, and it talks about why psychiatric disorders are uniquely human. It's because they have a language and emotional component, and even if that's present in other animals, we can't discern that. So whether it's autism, schizophrenia, or Alzheimer's, there are these language and emotional components that we can only see in humans. So I'm trying to figure— I'm grappling with this problem of how do you get really good quality human data in an ethical, effective fashion so that we can make real CNS therapeutics. A lot of psychiatric disorders and neurodegenerative disorders are among the highest burden. And probably if you look at progress in therapeutics, and burden, uh, it's, you know, one of the, the kind of most lopsided. So that's just a, you know, long-term thing I'm interested in. And the third thing is kind of like building new institutions. So I think it's, it's a very American thing to think about. We have— we used to have going westward, uh, and I'm interested in whatever that looks like intellectually. So, um, there are lots of, you know, great philanthropists and, um, new scientific organizations popping up, so I'm in the fray there helping how I can roadmap, consult, to think about, okay, how do you create an incentive structure that gets people focused on a problem and really solving it? So in the parlance, it's how do you make another Bell Labs? But maybe we can't create another Bell Labs. We need to think about some other institutional model. So yeah, hyperspectral imaging, human data for CNS, new institutions. That's, that's from my head.

A bunch of other ideas. I mean, I think that's gonna be exploring a lot of different things the next few months and beyond. But, uh, yeah, I think on the last part, you've been uniquely— you've been, have you been exposed to Genelia, which traces insulin use from Bell Labs because Eric Metzig, you know, was working at Bell Labs. So it's kind of, you know, I think— yeah, I totally agree. There's new— I think Luke even said this, like Science is moving away from competition to a generosity era. And so that requires new ways to collaborate, new incentives, like you said. And so I think the whole concept of how to do science is kind of being a little reorganized, not revolutionized. I don't know about that exactly, but kind of more options. I think there's more options for scientists. I think that's the key thing. If you're an inventor, you have a lot of options versus just like one, which is like become a professor. So, right, I think that's probably the best thing. It's like if you are— if you're a PhD student, postdoc, you have 20 different paths, right? And you get to choose the best one for you. And that's a lot of progress. Versus 20 years ago where you're a PhD, either professor or nothing. So that's awesome. Yeah, I think we'll talk in 2 years. We'll talk next week too, but like we'll do this again in 2 years. And by that time, I think you'll either have started a company or you're probably leading— you're part of some new research institution.

So, you know, that's, that's the—

maybe both. Let's do it. And then, um, but thanks for doing this, Nan. This is really awesome, and we'll catch up on this podcast 2 years from now, and then we'll catch up in general a week from now because it sounds great. But thanks for doing this.

I really appreciate it. Man, thanks for having me. It was a lot of fun. And, you know, if anyone listening, my, you know, message box is always open. I'm excited to always talk to great scientists.

We're gonna put your email— we're gonna put your email on it. You're gonna get 100 emails to you. So I get a lot of emails per day. You don't want— you don't want to even— I think it sounds cool in theory, but then the reality is the response— you have to respond to everyone. And it gets kind of brutal. So I'll ask you before, but I'll, I'll ask for your email and see.

I think I'm ready. I'm in a, uh, you know, talking mood. And, uh, you know, I think there's some serendipity will come from it.

So love it. Okay, well, thanks for doing this.

I really appreciate it.

Thanks, Josh.