Synthetic Biology and Rapid Evolution with Chang Liu

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's compelled to build their own enduring business. Hey, Cheng, welcome to this podcast.

Thanks.

Great to have you on. On this. You're just a legend. I've only heard rumors and, and, and, and like kind of gossip about your work and, and how you set up your lab. So I'm really looking forward to like just talking about your story and then the research you're doing now.

Yeah, well, thanks for your interest, and I'm happy to share. Cool.

Maybe just to start off, do you have a brief introduction on yourself and the work you do?

Sure. Yeah, so I'm Chang Liu. I run a lab at UC Irvine and have been here for 10 years. And our focus is broadly in the synthetic biology and protein engineering as well as protein evolution fields. Our kind of attack on the problem is that we've developed these genetic systems that can evolve very rapidly, namely the orthorep system or orthogonal replication system, where we have cells that in addition to their genomic replication has an independent replication system, in our case working in the cytoplasm. But this allows you to change the properties of replication without bothering the genome, therefore allowing the cell to survive and be this autonomously replicating agent that can evolve in a scalable and deep manner. And what we do with our orthogonal replication system is we increase its mutation rate. And so then we can kind of cross these speed limits of how fast you can mutate a genome and actually get to the point where we're driving the evolution of genes at rapid rates to produce biomolecules and to study the evolutionary process. So, you know, I live a typical, I suppose, science academia life, which involves teaching and running a group, mentoring students, mentoring members of the group, them mentoring me, and it's a very meaningful and exciting career.

Yeah, I think you do very outstanding research in synthetic biology and as you have your tentacles involved in a lot of the things from protein engineering, microbial engineering, and probably a lot more. Maybe we can go from the starting point. How did you get interested in science, or was it engineering that got you into this type of work initially?

Yeah, so I suppose we can go back pretty far. Maybe I'll start— well, when I was young, I was very serious about music. I was considering pursuing a career in, uh, well, my specialty was like piano, classical piano, uh, performance. And, you know, um, I was trained to be a classical pianist. Um, so I went to Harvard as an undergrad where there was a piano professor who took me on. Um, but also right before I went to Harvard, I took some chemistry classes and my high school, and I really enjoyed that. So I decided to see if I would even be a chem major. And so I started taking the chem curriculum at Harvard, and it was just, you know, awesome to me. And I decided to pursue a science career instead of a music career.

Okay, so two questions there. So what house were you in? Okay, I was in Leverett. What house were you in?

Okay, not too far. I was, uh, in Mather House.

Okay, that's probably— at least the worst would be Dunster. Dunster was like the worst. So, Dunster. So, okay, cool, you're in Mather. And then when was— like, you're, you're practicing piano.

Yeah.

Um, like, was, was this like a late decision? Like, you've been playing full-time piano and then like Senior in high school, say, okay, I'm gonna do science.

I think it was very gradual. I mean, my parents always believed that, you know, I shouldn't specialize too early. I mean, I was very serious about piano, but I was still keeping up with all my academic pursuits and doing well in high school and before that too. And so, yeah, it was just this kind of gradual movement towards more and more science. And ultimately, I think it's a good decision because, you know, the world needs like 50 good pianists. Yeah. But it needs a lot more good scientists. And so I think there's more value to it. And it's also less competitive in some way. By that. Yeah.

Okay, so you're in college. Did you go to Harvard for the music department, or did you go to Harvard for the chemistry or science? Because Harvard chemistry is pretty hardcore. It's like probably the most hardcore department in chemistry, or maybe hardcore department in general.

Yeah, it is a hardcore department. I mean, at, uh, at Harvard, as you know, um, they didn't have like minors. Right. You could do like dual concentrations. What I did was I took all of the music classes necessary for a major except for the history aspect of it. It was just like theory and, you know, the more like direct musical performance stuff.

Yeah.

And so I took that and then I also took the full and even more than the full like chem curriculum. So when I started, I think I was behind in chemistry in that even though in high school I really enjoyed it, you know, a lot of people entering Harvard had much more training, but I picked it up quickly and, you know, really liked it. And I was taking a load of 5 courses on average per semester instead of the usual 4. So I got a lot of work in.

That's pretty hardcore. I think the common theme here is, Chang, you're pretty hardcore. So you're, you're at Harvard, and any favorite experiences? I mean, I'm trying to think who is teaching chemistry. Like, you had Stuart Shriver in his prime. You had David Evans, a few other folks. So like any kind of formative experiences?

Yeah, so I took the You know, I wanted to go all in when I decided early on that chemistry could be a possible route that I would take. So, you know, I took like the Chem 20, 30 series, which is the organic chemistry for chemistry majors. There's also like the, I guess, 17, 27 series, which is for like the more bio-heavy and pre-med side. But in 2030, you know, one of my professors was Eric Jacobson, who was just this wonderful professor.

He taught me chemistry.

Yeah.

This guy would do reactions with both hands.

Yeah, exactly.

Insane.

Right. So, you know, it was funny to observe because he would write with his left hand and erase with his right. So he'd be erasing the board as he was writing the new thing. But like, I really enjoyed his intensity and The professors at Harvard really take the teaching part seriously. I mean, as we do here too, but they're very dedicated to teaching, at least most of them, and they're interested in communicating how they think about the field. And so that was a very formative experience. I remember in Chem 20, I was definitely behind compared to others. Eric Jacobson taught Chem 30. But I decided like midway through Chem 20 to, you know, spend a few, a couple of weeks just intensely like getting up to speed myself in my dorm room. And, you know, after that I was like up to speed with the rest of the class. Cool. And so, you know, ultimately I did well in Chem 20 and then Chem 30, I was really at the level where I could enjoy the nuances of things and became very passionate about it.

That's cool.

Yeah. After Chem 30, another formative class was George Whitesides taught a grad class on— I believe it was called Physical Chemistry, but it was very like organic chemistry. Heavy. And what was neat about George Weisheit's class is that he was very focused on, or, you know, he really revealed how he thought about the subject. And one of the things I think that governs his thinking about chemistry was like how to reduce things down to like the essence or the very simple, the simplest way of understanding. Something, right? Where are those like fundamental layers of understanding that other things are built on top of? And I really appreciated that because, you know, I think I naturally want to reduce things down to, you know, what is the crux of the problem here as well. So that class really helped me love the subject even more. Cool. Uh, after—

one quick question, were you like a Lamont person? Were you always in Lamont, or what was your favorite study spot? Was it a music room? I used to— honestly, I used to study in music rooms. I used to go to music rooms because they're so quiet and just like shut the door and just like sit there for 5 hours.

Yeah, I just stayed in my dorm room. Okay. Yeah, I— yeah, boring. And just close my door in my room.

At least you weren't a Lamontster. I had friends with Lamont. I was like, this is— I can't. It's Lamont's undergrad library. It feels very stressful to be there. But also during undergrad, did you join a lab? Was it like sophomore year? What time did you join a lab?

At the end of my sophomore year, I joined Stuart Schreiber's lab. Cool. I was talking to various people in the department, like postdocs and grad students. And, you know, there was this sense that a lot was happening on the interface between chemistry and biology. And I think also the sense that at least the idea in organic chemistry of being able to synthesize anything that's physically stable was already a solved problem, right? I mean, obviously there are major developments in organic chemistry still, and very meaningful ones where you can simplify synthesis and learn new principles and even start synthesizing exotic molecules that don't exist in nature. But the natural product side, which kind of drove traditional organic chemistry, I think by the time I was there was sort of a solved problem. But biology just seemed so open-ended and it seemed like there was a lot of, a lot of room for exploration there. So that's why I joined Stuart's lab. And he's also one of these like very philosophical, brilliant and engaging people. So that certainly attracts me to his lab as well.

Yeah, he— I mean, it's amazing what Stuart did. He brought chemical biology to Harvard Chemistry, which is like, you know, there's some faction, large faction of that department that is very organic, organic-oriented, and they kind of look down on chemical biology. Yeah. And for Stuart to like succeed, even just stay there, right?

Well, I didn't really appreciate at the time, but right, the, the name of Harvard's chemistry department is Chemistry and Chemical Biology.

And now I think, was it, was that the case then?

Or I think it was the case then too. And I think it was Stuart's doing.

Wow. I didn't know that.

So now I appreciate how hard that is, right? Like to change a departmental name at an academic institution, like impossible. And especially so many, like so much history and really egos involved as Harvard Chemistry. Yeah. So.

I had a friend who worked in a Jickinson lab. Yeah, he would just be there for till midnight every 7 days just running reactions. And I'm a biologist, I'm like, okay, I complain about biology, but at least I'm not these organic chemists.

Biology is cool, you know, in that you, if you get one thing, it self-propagates, right? Exactly. You get 0.01% yield. Well, you know, directed evolution is even less. And then it'll take off. Whereas, yeah, in synthesis you have to make everything and all the quantities that you get are what you've made. So you got to be there late.

I know. So you're near the end of college and you end up going to grad school, Peter Schultz lab. I think he was still at Berkeley or maybe he was at Scripps.

He was at Scripps.

He's at Scripps.

Okay.

He's at Scripps.

Yeah. You know, the other big influence in college in chemistry was David Evans. So his advanced organic chemistry class, which is just this masterfully crafted class, I mean, it was intense and a lot of work, but he was very nurturing. He had this gregarious, open personality where, you know, he wanted to make everybody feel that the heights of chemistry were quite accessible. It's just a matter of putting in the work and time. So I think I appreciated that a lot. And then he kind of took an interest in me where, you know, we would meet occasionally outside or somewhat frequently, I suppose, relatively speaking, outside of class. He asked me to TA his class the next year for— he was teaching Chem 30. And so that got me into really the academic experience where I was like doing research in Stuart's lab. I was teaching sections for Dave Evans, and yeah, just engaged in the kind of department outside of just the traditional taking the courses required kind of experience.

That's pretty special. I mean, David Evans is a legend in chemistry.

He—

I think he moved to Caltech and he recently passed away.

Yeah, he recently way. Yeah.

And then he invented ChemDraw with his wife.

Yeah.

In fact, on the side, he's like, yeah, I'm going to invent ChemDraw. And becomes this.

Yeah.

Incredible. Yeah. That's cool.

I mean, I had all these privileges of having these great influences early on that shaped my passion for chemistry. And, you know, when you work with those people, you kind of, they become both your guides and your heroes. So you tend to like think of a meaningful career as trying to do what they did, right? So that's, I think I've naturally stayed in academia just because, you know, those were the influences I had. You know, now it's clear that there are plenty of great career paths in science outside of academia. But I never really thought about them too much because of the influences I had. Yeah.

Also one thing too, I agree with that, but also like having early success helps.

Yeah.

And so did you, did you get a paper published in college or when was your first paper published? When's the first time you were author on a paper?

First paper we published, I mean, I was an undergrad in Stuart's lab and at the time he was really focused on what he called diversity-oriented synthesis, which is, I think, a kind of, like, supercharged combinatorial chemistry, where we were interested in diversifying molecules, not just by changing the substituents combinatorially that were on them, but trying to find transformations that would actually, like, diversify the structural connectivity of molecules. So we published a paper. I mean, I was second author. Yeah. The postdoc I worked with, obviously let that work, but our paper was on like how to control the stereoselectivity of this reaction called the Passerini reaction, which has like 3 components coming in, and you know, it, it's a nice sequence of reactions that all occur in one pot, so you can get a lot of complexity generated, and then we were trying to be able to diversify in controlling one stereochemistry versus the other.

That's cool. And so maybe you can talk about your thesis and also then just maybe talk about that transition. It seemed natural, I guess, going to a PhD program seemed natural in this context. Like, okay, I have all the success, I like the people I'm working with, I should get a PhD. How did you figure out like which program to go to? Yeah, to join, who, who to pick as a mentor?

Yeah, yeah, these are good questions and I have answers for them. So my thesis was basically on this reaction, right? And what we did in terms of using, you know, metals to control the stereochemistry. Organometallic catalysis. But at the time, so we were doing pretty like chem methodology stuff in the lab, but the other side of the lab was screening compounds for biological activity. I mean, the lab has a biology goal, and so I was exposed to biology, and I remember reading these papers by Peter Schultz, largely because Pete was the other big kind of the West Coast chemical biology, you know, big name that was a counterpart to Stuart's East Coast control, as it were. And so I read some of his papers and I was just very I thought it was amazing once I saw his work on expanded genetic codes that this was possible, right? And I remember reading a paper on having a biosynthetic pathway that would produce para-aminotyrosine that they had also tricked the synthetase into taking. And so you would have this E. coli strain that could just live off of or naturally use a new amino acid that you didn't have to feed it, right? Because it could make its own unnatural amino acid in this context. So that seemed to me to be incredible. And Pete was giving a talk at Harvard my senior year, so I made sure that I got to meet with him and have lunch with him. And I told him that I was very interested in his work. And he said, you know, if you come to Scripps, you can join my lab. And I said, okay, I will apply and come to Scripps. And so I applied to like 4 places, but I was eyeing Scripps as the main possibility.

Well, you know, the best part of grad school, at least I'm biased, I'm not as good a scientist as you. So the best part of grad school for me was the interviews. So I kind of feel bad that you didn't get to go to as many interviews as possible.

Yeah, that's true. I mean, I think if I were— it is a nice time where you can kind of get like the pulse of the field.

Exactly.

All these people. So I interviewed at— I applied to MIT, Caltech, Scripps, and Stanford, and then I interviewed at those places. And, you know, I decided to go to Scripps. Yeah, but I wish I had done a little more. I mean, it was also senior year. I was trying to, you know, senior year is nice because you can— you don't have to worry so much about grades. Yeah, I still want to get good grades, but it wasn't as important in terms of, you know, being a gatekeeper to some, you know, next stage in your career. And so I took a lot of like grad-level courses in physics, one in computer science, one in multiple in chemistry, just to explore my curiosity. Cool. They were hard classes, so I also didn't have that much time to go like, you know, do these trips, but jump on airplanes.

Yeah, it is very time intensive. Um, so you land at Scripps. Is this the first time in California? Like, because you escape Boston. Boston's a place you escape, and then you get out of Boston.

That's right.

You get to San Diego. Was this the first time you lived there, or?

The first time I lived there, yeah. I grew up in Tucson, Arizona, which is very close to California. So, I mean, especially Southern California. So we used to vacation in Southern California. I have some family, friends, and relatives there. Yeah, but it was the first time I lived there. And I didn't know San Diego, but it's just a lovely environment, right? I mean, we're kind of addicted to it now, San Diego.

And so maybe, so you're at Scripps, you join Peter's lab, right? What was in your head at the time in terms of like, were you saying, were you like, I'm gonna become a professor, I'm on this professor track, or was there something else, or were you more open-minded at the time, or were you just set on getting to the point where you can run your lab? What was the, what was kind of the, the motivating factor, or what was your vision at the time?

Yeah, a couple of things. I mean, I definitely was very focused on continuing in academia and eventually having my own lab. And looking back, I don't think I even really considered other career options. It just seemed like the natural course and the one that all my influences took. And so, you know, it was both comfort and my, like, inspiration from their careers that kept me on that path. And so, you know, so that kept me focused. And then the other thing is I really enjoyed doing experiments and the hustle of it. And, you know, I, I think I gamified doing experiments a little bit where obviously there's the longer-term vision and there's the research results you want, but just day to day, just being like, can I do this more efficiently than the previous day? Or can I get more experiments done and, you know, design them well so that I learn stuff about them even if they don't work? That was also like a short timescale motivator, just like seeing if I can be more efficient, get more done. So I think those two added up to, yeah, like a decent graduate school experience.

I have a friend, Ryan Flynn. He's a professor at Harvard now, and he's in Bertozzi's lab.

Oh, okay. I know who he is.

He works on microRNA. He is— you probably know, but he's really hardcore. And like, he would like, in terms of efficiency, he would run like, he'd have these blocks, he'd run like 3x the experiments somebody else could because he figured out a time to check in a lab.

And yeah, I mean, I, yeah, I tried to take the view that like experiments shouldn't wait for me. I should be waiting for the experiments. And so, you know, this is especially true with like cloning and things that have a natural time cycle where if you are willing to come in at some odd hours and be very organized about when you start things, you know, you can like save a day per experiment. And given how many experiments you run, that adds up.

This is a very separate topic, maybe just a little detour. How do you plan experiments? There seems to be a lot of variability. Everybody has their own little system to plan experiments, and there seems to be no standards, but any recommendation to scientists to plan their experiments a little better?

So I think this is somewhat antiquated now, but I, really enjoyed writing out the entire experiment and thinking through the steps and then annotating my lab notebook so that, you know, at night I would be like writing out experiments and then I would make sure by the daytime or, you know, before I left at night, I would just have like the reagents prepared so I could get going the next day. And then in the morning, oftentimes I would like my subconscious would work on some things for the experimental design, so I would make some modifications in the morning, but then get going.

Like a composer almost.

The writing process really helped a lot, and I relied on it to think through ideas in a very methodical way. Interesting, cool.

Okay, so you're in Peter's lab, and you join his lab, it seems like a little after the AMPR suppression paper. So I'm assuming that like a lot of—

Right, so there was an amber suppression paper on E. coli in 2001, right? So I started in 2005. Yeah. So between the time that the amber suppression paper was published and when I started, there was a really talented graduate student named Zhangming Um, let's see, what is his last name? I think it's Jiangming Shi. Uh, and he really like streamlined that process. So, you know, when the first paper was, uh, published, I mean, there were a lot of like technical things that still needed to be made more efficient and optimized, but he like designed all these libraries, synthetases, and had this real strong protocol going for like evolving new synthetases and how to make mutations in synthetases, how to choose those and how to do the selections right. So I benefited from starting at a time where all of that stuff was going really smoothly. And so I just leapt into some of these projects and things worked well early and I could kind of develop those early returns. Cool.

Talk about maybe the research you're doing in grad school. It seems related to genetic code expansion, like any particular projects.

Yeah, so I did a lot of projects early on, but the one that took off, um, I was interested in incorporating post-translational modifications via an expanded genetic code, and, uh, that was my first interest. And then my Second interest was to see if we could like evolve new protein functions now that we have unnatural amino acids available, and they kind of dovetailed with each other. So with the post-translational modification side of things, initially Pete suggested that I look into phosphotyrosine and phosphorylated residues, phosphoserine, phosphothreonine, and I synthesize a lot of these protected phosphates where, you know, you can't like deliver a phosphate to a cell or you can't deliver a phosphorylated amino acid to a cell because the transporters don't like to take in negatively charged amino acids. And so I was caging the negative charges by synthesizing these protecting groups on them. They had to be like really long phosphates. Interesting, because like if you release one phosphate, then if you want to cleave the other protecting group and your protecting group is very close to the other oxygen, your enzyme has to get really close to the phosphate. And if one is already uncaged, then that enzyme needs to be able to tolerate that negative charge, which is usually not the case, or it can't do. And so we were synthesizing these long protective groups that once they got into the cell, you get an esterase on the ends, act on the ends, and then there would be this like release of carbon dioxide that would shorten the linker. So it was, uh, you know, there were— the esterase activity would just start trigger a set of reactions that all the way at the end release the phosphate. So I worked on that. It didn't work because it turns out that the cells I was using— I didn't know this— they could just like eat the phosphate itself. So once you release the phosphate, there would be phosphatases in the cell that could remove the phosphate group. So at that time I was also looking at analogs of phosphate, and one analog is sulfotyrosine, which is a sulfate group instead of a phosphate. But when I looked into that analog, the, uh, there was already an associated biology with that post-translational modification. It's not just an analog for phosphate, but it has its own biological meaning. And that one actually incorporated very easily because it's just a single charge, a single negative charge, so it could get into the cell if you added a lot of it. And we didn't have to play this protecting group trick. And then when I tried selections to make a synthetase against it, they just worked very quickly. So early on we had phosphate running and I proved that it was incorporated. I found an interesting case where there's an anticoagulant that leeches make, which is sulfated, and its affinity for thrombin, which it inhibits, is based on the sulfation. And so I showed that we could produce the actual sulfated form rather than the desulfated form using the engineered synthase, and that was our first paper. Interesting.

That kind of highlights all of the theme of your overall work is just the power of direct evolution to find the answer for you versus maybe, you know, having to do it rationally altogether.

Yeah.

Could you maybe— so you're in Peter's lab doing a bunch of research, you're in a golden age of at least his work. Like any kind of— when you reflect back on that experience, Sometimes I have friends and being part of Lab reports or groups that are just at the peak of their powers.

Yeah.

Like, do you reflect on like any learning lessons? Like, like the intensity of it, how fast things are moving? I would assume in the short lap of the time there's like everyone's trying to collaborate with you, a lot of inbound, a lot of different projects to work on.

Yeah.

Like, how did you manage all that? And any kind of reflections on what it takes to build something of that, at that level of excellence?

So I think Pete's group, and it remains so, but I think it was especially true when I was there, was a really special case. There was just, it was a really special place. There were a lot of great scientists, budding scientists there, and everybody like challenged each other. By example. And so just having colleagues who are there all the time cranking on experiments, thinking deeply, being very creative was inspiring. I tried to do the same. And, you know, you also became like great friends with these colleagues that you were like spending day and night with in the lab. And so it just became like both a satisfactory, it satisfied both like your work and your life aspects of your existence. So yeah, so it was a great environment for research. Cool. Yeah, and I was doing directed evolution for these synthetases. Once we got these synthetases, I became very interested in this idea of like evolving proteins that contained unnatural amino acids in them. And so sulfate has a lot of biological privilege in that a lot of biology has been designed around this post-translational modification. And so we were asking whether we could exploit it in terms of leveraging some of these biological modes that it seems to prefer. Such as like tight binding of, you know, of target molecules. The sulfate group just seems to be very good at that. So, you know, we were trying to then do evolution experiments with sulfates against targets. And so that got me into also these deeper questions about like, why is it that we only have 20 amino acids? Is evolution Capable of using more? Would it benefit from having more? And it got me into thinking about like how to supercharge the evolutionary process itself rather than just use evolution to make the biomolecules we want. Interesting. And I think later on in my career, you know, the focus is not just on like applying evolution, but actually changing the nature of evolution and even potentiating it further in service of its demonstrated capability of innovating all the time.

Interesting.

That's a good point.

And so like you're forming all these ideas in your head in grad school. How did you determine to like do a postdoc in Anna Markin's lab? So you come to Berkeley. Berkeley's awesome. Adam's awesome. Why, why did you choose Adam's lab? And then why not? Did you have other options at the time? I think now, I don't know if at the time, I had a lot of grad school friends and they either pick a postdoc or they do a fellowship and some even start their own labs. But how did you determine to like do a postdoc with Adam?

Yeah, so I had a relatively short PhD. I graduated in 2009 and I started in 2005. And part of the motivation was my wife now, girlfriend at the time, was at Stanford Law during the later years of my PhD. So, you know, I had decided that I should go join her at some point. And therefore I focused on postdoc opportunities in the Bay Area. I remember there was this fellowship opportunity for the Miller Institute, which is at Berkeley. It's called the Miller Institute for Basic Research in Science. And they have this sweet fellowship program where you're funded through them. You're basically a member of that institute, and then you join a department. And I applied to it and I got it, and so that even like narrowed my— where I was looking further to, you know, labs in Berkeley.

So you were a Miller— one thing I always get confused— you were a Miller Fellow, but do Miller Fellows run their own groups as well? They have facilities and stuff?

They have like, yeah, they have like, you know, I mean, I would say that it is basically a typical postdoc, but with a little bit of more independence and yeah, kind of an understanding that like you should be afforded the independence by the institute. So that was how I was in Adam's lab. I mean, my choice of Adam's lab was because during the later years of my PhD, synthetic biology was all the rage. And it was obviously very relevant to what I was doing, but it seemed like genetic engineering and even protein engineering was being upgraded by synthetic biology ideas. And synthetic biology technology developments. So I was a little skeptical when reading about like how you could design genetic circuits easily and cover both dynamical properties and regulatory properties that normally you would just find in natural systems. But I wanted to test that skepticism by actually seeing if I could learn it and do something in that area. And so I joined Adam's lab to do so. Cool.

It seems to be related to the Schultz lab research, but a little, a little related but different. So Adam, he has a lot of, his projects are definitely on the, on the edge. And so kind of what were you working on during the postdoc? And then maybe we can use that to, how did that set you up for your own lab?

So in grad school, I think halfway or maybe even earlier on, I had this idea for orthogonal replication. And it was motivated by two things. One is the realization that, like, these cycles of directed evolution that I was doing all the time seemed like there had to be a better way. And the second was, you know, in Pete's lab, I mean, we were basically working on orthogonal translation systems. So it was not too far of a stretch to just imagine, you know, what would other orthogonal biosystems look like? And those two ideas merged in that having an orthogonal replication system, I realized you could really solve a lot of these problems with stepwise directed evolution. I mean, another influence in this regard was the thesis committee chair, or my thesis committee chair is Jerry Joyce, Gerald Joyce, who was working on continuous RNA evolution. And so the idea of like having mutation and selection all occur in an autonomous system was already being done in the RNA world. And I thought you could do it in the protein world as well. So I had this idea formed when I was a grad student, and I actually started in Adam's lab to do this idea where early on I spent like 3 months trying to refactor phage genomes in order to just steal their replication system and see if we could install it in bacteria. It didn't work because these phage systems had like severe toxicities associated with them. I think it's a working strategy, and I think there will be some work out from other groups that show it's a working strategy now, but at the time I didn't want to like screen through a lot of different systems. And Adam's lab had its own influence on me where they were working on really neat like predictable regulatory systems involving RNA, and I had some ideas on that, so I started some projects on that too, and those worked., and so I basically focused my efforts on those. So in Adam's lab, we worked on the design of RNA switches that could integrate multiple signals by doing so co-transcriptionally. So there are a lot of RNA switches, and I think the idea is that these are like very old regulatory forms where what they regulate is transcriptional elongation. And the neat thing about that is it's this processive process where you make a regulatory decision and then it's a regulatory decision that's imposed on what will come next. And what will come next could be another regulatory decision and another one. And this seemed like a really predictable way to bring together regulatory logics before you get to your gene of interest. That you're trying to control. So we started working on those systems and we made some nice progress on those by designing some suites of RNA switches that were controllable by various elements.

Cool. And then after working at Adam's lab, you become a professor at Irvine.

Yeah.

You start to raise a family. How did you go about establishing a research vision for your own group? And how does that relate to your past research and kind of any kind of lessons you've learned?

Yeah. So, you know, as I said, I had this idea that orthogonal replication was, was a good project and something I was very interested in and had ownership over. And I decided that I wanted to actually make it work in my own lab. So when I started, we decided that that is what we were going to do. Right before I was leaving Adam's lab, I spent half a year setting up various parts of it. And so that got me a head start on the, on the process. Yeah, but you know, it was an idea that I thought had mileage and I wanted to make it our own thing.

Cool. And then the rumor I always heard, I don't think it's true since you told me, was that like when you set up your lab at Irvine, it was just you. That's the rumor I was hearing from people in like the synthetic biology field, but I don't think that's true. Not true at all. You had a technician and a grad student. Could you talk about what was that first year of setting up your lab? What was it like? How did you recruit people and stuff?

I was very fortunate. When I was in my last year in Adam's lab, there was a rotation student in Adam's group named Arjun Ravi Kumar. Who sought me out because he heard that I was thinking about building an orthogonal replication system. And he said he was very interested in that idea. And so we start brainstorming and he decided to switch programs and move with me to Irvine. So when we came to Irvine, and I still remember driving a bunch of boxes with him, with some genetic constructs that we had made to bootstrap the system. When we got here, it was just like he and I were setting up the lab and just there 24/7 doing experiments to see if we can get this idea off the ground. So we hired a lab technician who was this very talented guy named Adrian Arieta, and it was the three of us doing experiments. You know, our situation was unique, right? Because I think the norm is that you develop something as a postdoc and then you kind of extend it and blossom from that, from the stuff you already have from your postdoc work rather than kind of bootstrap a new project altogether. But because we wanted to take the latter path, we realized that it was very important to demonstrate that this is possible, get an early publication before others in the field would believe that this had legs. So I decided to spend the first year just basically acting as a postdoc in my group.

Right. So that's a good way to put it.

I did. Yeah, I didn't write grants that first year. But then once we had our system working, or the architecture for the system working, it took us multiple additional years to get to higher mutation rates. But at least we showed this idea was possible. That's when I started applying for funding and then getting things off the ground on the practical lab building aspect.

So could you talk about, like, um, did you have, during that period of time when you were trying to essentially build a system from scratch, not an extension from maybe past work. Were there moments where you felt that, like, did you ever have a moment where this is not possible? It's gonna be too hard? And then can you then talk about the output of that work? Can you describe OrthoRep, OrthoRep, what it is and the significance of it?

Yeah, I mean, I think we convinced ourselves that this should be possible. Of course, we don't know how long it will take and We had a pretty good idea of how we would do it. Basically, you know, we found after doing a lot of literature search that there were these autonomously replicating plasmids in some weird strains of yeast that already encoded their own DNA polymerase. So the first strategy was that we would refactor viral systems. H systems and try to boot those up as a durable episome that self-replicates by its own DNA polymerase in the cell. But as I mentioned, I tried that for 3 months at the beginning of my postdoc. It didn't work because those were highly toxic systems. And so later on, I was searching to see if there were already like potentially leftover bits of viruses, I found that there was this weird replication system that had a lot of properties we were relying on from viral systems that was already in some strains of yeast. So we basically decided that that was the tack we would take. So we spent a lot of time like developing ways of genetically manipulating this, refactoring the DNA polymerase, making mutants of the DNA polymerase that based on literature on homologous structures could elevate their mutation rates. We didn't get too far on that for our first paper, but at least we got enough of a mutation rate difference that we could use mutation rates to prove orthogonality, and that was what we ultimately got published a year after we started the group. But yeah, there were like plenty of moments of doubt as to whether this would be viable in the long run or how long it would take. I think ultimately Arjun and I were just very convinced that this would work and that it would be useful. So we, you know, we were able to get past moments of doubt.

Yeah, I think it's just pretty rare that most people who start labs are pretty risk-averse initially, and then you try to get tenure and then do risky science. And often doesn't really work that way from outside looking in.

So yeah, I think it's right, you know, like ultimately I have this like 5, 6-year runway, uh, before I get fired, and you can do a lot with it and If I get fired, it's not like, you know, I'll be on the streets, right? I'll get a different job. So it didn't seem too much of a risk for one's own life. Similarly with Arjun. Yeah, it was. So it seemed like a risk worth taking. And like I had, I think I had realized, and I realized even more now that the thing you want to worry about in academia is not getting tenure, but being in a position where once you get tenure, you don't really have something that you're known for or that you like can carve out and build upon as your own, right? Because that's a bad existence. Like, when you're an assistant professor, everybody gives you a lot of leeway. They're very supportive, but it's when you get tenure that you're supposed to like fend for yourself. And if anything to base fending yourself off of, fending for yourself off of, then you're in a bad situation. So I think it's worth taking a big risk in the assistant professor phase so that you can establish something interesting that you can build upon for the rest of your career.

Totally agree. And so you definitely carved out your niche and you're the best at what you do in terms of hypermutation and developing new tools to design proteins. Could you talk about the stuff you're working on now and kind of the bigger opportunities you see in synthetic biology?

Yeah, so I'll start local with kind of the stuff we're doing in this hypermutation space. So one So once we had the orthogonal replication architecture established in 2014, Arjun spent these, you know, just heroic efforts trying to get the mutation rate to be high enough to really evolve stuff. And that took another like 2 to 3 years, but we got to 10 to the minus 5 substitutions per base. In like 2017, have a paper on in 2018. And the mutation rate of the yeast genome is 10 to the minus 10. So we were, you know, at that point we were cooking with gas and we had, you know, the ability to really drive a protein's evolution. But that's actually still an order, or an order of magnitude or more lower than what we think a gene can tolerate in terms of its mutation rate that maximizes the speed of adaptation. And so there's another student in my group, uh, he's going to graduate this year. His name is Gordon Ricks, and, uh, he's gone north of our mutation rate to be 10^-4 now. And that extra 10-fold, or I think we're at like 2 times 10 to the minus 4, we haven't published this yet, but it's a big deal because it, in a lot of cases where we're mutation limited, we are actually getting a 20-fold acceleration in things. So now I think we're actually like at the point where we're going to take a break from engineering OrthoRep to be faster and better and focus both on applications and on understanding basic aspects of evolutionary biology. Along the way, we've done a lot of applications, but at this point I think we'll put all our effort into applying and applying OrthoRep to basic science and protein engineering goals.

Do you see hypermutation in terms of the application side? Like having more value in maybe developing a better CRISPR protein or antibody, or do you see it as a maybe, oh, could it create fundamental discoveries about, you know, evolution and, you know, how—

I mean, I think the most long-term impact will come from the latter category. I mean, I don't know for sure, but certainly that's where my bet is. We are doing a lot on antibody evolution. We're doing a lot on enzyme evolution, and there's plenty of space to innovate there and get better enzymes and antibodies. So a lot of our lab is dedicated to this, but one of the things that I think nobody else can really do is mimic very long evolutionary pathways at laboratory timescale. Basically, the question that we're trying to answer now is how the history of evolution influences a gene's future. And in nature, the history of evolution is thousands to millions or billions of years. And a lot of what happened in a gene's past matters to its future because it's embedded in the sequence, right? A gene, when it adapts to something new, the thing that it adapted to before isn't just— doesn't just vanish. You know, the sequence patterns that allowed it to adapt to the previous condition are still in the gene. The sequence patterns that allowed it to adapt to the environment that was previous to that adaptation are still remembered in the gene to some extent. And those things accumulate over time. And I think those patterns really matter in shaping what the evolutionary potential of a gene is going forward. And so because we can compress evolutionary time, at least for a gene, by so many orders of magnitude, I think we can start sort of prospectively generating past and then testing in a systematic way what past did to the sequence of the gene and how that past influences the next phase of adaptation. Ultimately, we want to understand why evolution is so effective. I think it makes sense to me that there are people who believe in intelligent design, right? Amazing thing that all of this complexity, and even when you're looking at a gene, you know, the fine details that a gene has adapted to function at, it's amazing that that is possible, right? There's no like physical law that says if you're a sequence, you're going to be next to sequences that allow you to find some adaptation, right? So I think there's more than just a numbers game here. I think there is some force where over time what evolution has learned acts as a reservoir that it can draw on for additional innovation. And we'd like to both prove that to be the case in an actual systematic experiment or series of experiments and learn what those secrets are that allow evolution to work. And then those secrets are also useful in protein design and machine learning, right? Both are basically these learning algorithms that, you know, are trying to navigate some complex multidimensional space to find new objectives. Cool.

I think this is a great way to end it. So insightful at the end. We could touch base 2 years from now and get the update, but maybe If you have, you know, we already touched upon this throughout the conversation, and any final lessons you've kind of learned from your career so far, and any advice you'd give to maybe a scientist who is in grad school right now or trying to figure out what to do with their life?

Yeah, sure. I mean, I always, so I can, so the micro piece of advice is I think it's very important to design experiments in a way such that you learn something when they fail. This is something I try to teach my grad students. This means like building in the right controls, but even beyond that, just ask yourself like, when this experiment fails, what will I have learned? And if you can answer that question well, then you've designed the experiment in a way where it's going to save you a lot of time to diagnose after the fact. And so that's something I think should be kind of a mantra when you're in grad school. But the larger advice is you should be driven by your curiosity and your passion. And this is the time in one's life, if you're a PhD student, to like dedicate all of your effort and all of your energy to doing something interesting that just satisfies your curiosity, right? And you don't have too many other practical constraints, uh, that you have to cater to. And so make sure that you are taking advantage of exploring your curiosity in this time of life when you won't get too much of a chance again.

I totally agree. Yeah, this is awesome, Chang. Thanks for doing this. And we'll do this like 2 years from now, get the update. But thanks for taking time here and talking to me.

Yeah, thank you so much. All right.