Professor Alison Narayan
In this episode, Dan and Jackie talk to Alison Narayan, from the University of Michigan’s Life Sciences Institute and Department of Chemistry.
We talked broadly about the field of “biocatalysis,” about her DNA synthesis project with the JGI to explore flavin monooxygenases, and what we all need to do in the future to figure out how to make predictions about the functionality and chemical capabilities of enzymes. Also, the Prodcast formally declares its opposition to candy corn.
Links to work we discussed can be found below – you can jump to them here. Catch up on previous Natural Prodcast episodes here.
Transcript
DAN UDWARY: Hey everyone. Welcome back to Natural Prodcast. This continues the series of interviews we recorded at the Society for Industrial Microbiology and Biotechnology (SIMB) Natural Products meeting earlier this year. And today we have our conversation with Professor Alison Narayan, from the University of Michigan. I hope she’ll accept this description, but I’d say that Alison is not one of the more normal kinds of natural products people we usually talk to. She’s a chemist, and describes her work as exploring and exploiting biocatalysts – that is, using enzymes to do chemical reactions. Obviously, in my work, in genome mining, we have an awful lot of crossover, where us genome miners get really jazzed about finding enzymes that do unusual or unprecedented chemistry. So it’s super fun to hear about what folks are up to in a slightly different niche in chemical enzymology.
So, we talked about her JGI DNA synthesis project to explore the chemical capabilities of flavin monooxygenases, and more broadly about the field of biocatalysis and what needs to be done across science to to get a point where we can predict an enzymes synthetic capabilities. Also, along the way, I declared a position on candy corn, which is perfect timing for an October podcast release. I hope that doesn’t get me canceled.The monthly schedule seems to be working out, so expect lots more to come. Maybe a bonus episode soon, too. Thanks for listening! And enjoy this conversation with Alison Narayan.
JACKIE WINTER: Well, Welcome, everyone. It is my pleasure to introduce our guest today, professor Alison Narayan from the University of Michigan. Dr. Narayan is currently associated with the chemistry department as well as the life sciences department. So welcome, Alison to today’s podcast. And with your background, this is pretty fun. We’ve been talking with a lot of people who actually have been classically trained in synthetic chemistry as were you and then made their way into the field of natural products in some way, shape, or form.
And so do you want to tell us a little bit about what got you into natural products and got you excited about the world of natural products?
ALISON NARAYAN: Oh where did it all begin?
DAN UDWARY: Yeah, I love to hear people’s origin stories.
ALISON NARAYAN: For me, the initial attraction to natural products was dual, the beauty of the structures and also the challenge and thinking about making them. So I really came at it not necessarily based on the biology really just from as a simple organic chemist, can I make that big, beautiful molecule? And so I’d like to think I’ve evolved a bit since that time.
And I would say now, the primary reason I’m interested in the world of natural products is also twofold but for different reasons to learn about how nature builds molecules because nature does this with such elegance And also to take advantage of the biological properties that these molecules have and discover what’s possible from them. Because we have all of these molecules. We know very little about what the majority of them do. We have in-depth biological studies on a sliver of the molecules nature provides to us.
But in my lab, we like to ask questions about what don’t we know about this molecule. How do we discover more about what it can do, and as chemists take those structures as inspirations to really start making other molecules related to those scaffolds.
DAN UDWARY: So I think I associate your work with the term biocatalysis. I wonder would you say that’s fair?
ALISON NARAYAN: I think that’s fair. And I always use it as the first word in the title of my talk, 100% fair for you to make that association.
DAN UDWARY: So I guess what is the conceptual difference between a biocatalyst and an enzyme?
ALISON NARAYAN: I love that you asked this question because I have this question, we have this discussion internally in my group quite frequently. OK, so imagine this sentence comes across your desk. We’re developing this enzyme into a biocatalyst. Wasn’t it already a biocatalyst? And so–
DAN UDWARY: I mean, that’s what I was taught enzymes are, right?
ALISON NARAYAN: Yes, and so that’s why I don’t like that sentence. I think that sometimes, we think about enzyme function is we think about the natural function versus can we take this enzyme that nature has given us and develop it into a catalyst that we can use for many different things? I think they’re the same thing. So I would rephrase that sentence and say, we’re developing this enzyme, we’re exploring the function of this enzyme or its synthetic utility. But an enzyme is a biocatalyst. I would say those are the same in the Venn diagram.
DAN UDWARY: Do you need less specificity to be a broadly useful biocatalyst? Is that the difference?
ALISON NARAYAN: Oh, a broadly useful biocatalyst, so yeah, we have a lot of conversations, I think, a lot about this as well like is specificity good or bad when it comes to biocatalysis? You want your reactions to be very selective in terms of the products that they’re making. And you want to have good binding to your substrate and high turnovers. But if you have those things, can you also have something that works on every substrate that you give it? Usually not.
And so the question of do you need something that’s promiscuous for it to be useful? Not in a process chemistry route because there, you just need to build one molecule. But what I would say is where promiscuity is helpful is in an academic lab, for example, where you don’t want to necessarily engineer your enzyme for every different reaction that you want to do. If you’re not looking to develop a commercial process, you can have the ability to build different types of molecules.
Or if you’re a medicinal chemist, you want to have enzymes that you can apply in a route towards many different compounds in the same family. And so that you want it to have some promiscuity in the substrates it operates on.
DAN UDWARY: Got it, OK, OK.
JACKIE WINTER: So when you’re developing these biocatalysts coming from more of the chemistry background, you’re talking about promiscuity and substrate specificity, but when you kind of are applying them to making a compound, do you also take into account how are they going to behave in certain solvents or certain reactions or mixtures temperature, do you do take that into account? Or do you and your lab focus mainly on what can we feed this and what can it actually turn over?
ALISON NARAYAN: So that’s an excellent question. I think that we need enzymes with good physical properties to work with to make them good catalysts to then look at substrates go for example. So I think you can’t get to the second part without exploring the first part. And so I would say we don’t necessarily develop enzymes that have those properties. Sometimes we’ll do engineering to improve the physical properties of a protein and make it easier to work with.
But sometimes, what we’re looking to do is just find the natural enzymes with the best properties to then work with further. So for example, we’ve built a library of flavin independent monooxygenases with the help of JGI’s DNA synthesis grant. And that having a whole panel of these enzymes allows us to ask the questions which ones express the best in E coli? Which ones are the most thermally stable?
And those are important to us before we ask those next questions of will it work on many different substrates? Or does it do a reaction selectively?
DAN UDWARY: So it sort of does it work in the first place for instance? Is it useful?
ALISON NARAYAN: I like to think every enzyme is a little snowflake.
JACKIE WINTER: To treat it so it’s a delicate snowflake.
DAN UDWARY: It’s not wrong necessarily.
ALISON NARAYAN: Or maybe they’re like different types of snowballs, right? We’re in Michigan. It’s, well, we’re not physically in Michigan right now. But I am from Michigan, and it’s winter, so snow is on my mind. But if you have a delicate little snowflake that just melts in your hands, that’s not going to be great to work with. If you have an icy, hard packed snowball that you can’t destroy, we’ll pick that one.
DAN UDWARY: That makes sense, yeah.
ALISON NARAYAN: Because we want everyone in the lab to be able to work with it, not just the person that has the delicate protein touch, right? And we want other people around the world to be able to duplicate the results that we have with these enzymes, too, so the more robust, the better.
JACKIE WINTER: Nice.
DAN UDWARY: That makes sense. Yeah, it’s been a long time since I’ve actually isolated some proteins off the bench. I am well aware of the pitfalls of that not working out well.
JACKIE WINTER: The tears.
DAN UDWARY: Yeah, for sure.
JACKIE WINTER: I mean.
ALISON NARAYAN: Yeah, this is beyond my expertise and being able to predict which enzymes might have better proper physical properties to work with. But I mean, we found sometimes enzymes that are the most similar within our libraries can have drastically different properties in terms of their solubility, their stability.
JACKIE WINTER: It’s hard to predict how one enzyme will behave compared to another.
ALISON NARAYAN: And things that precipitate out or just have a very short half life.
JACKIE WINTER: So with your [project] with JGI, how many flavin monooxygenases are you sequencing, sort of along that vein, of if you have one or two and you compare them, and they behave very differently. Can you make any association that might help? Like, if you have an unknown sequence or new flavin monooxygenases, you could say, oh Alison sequenced all of these, she tried this, if you have this sort of motif, you might have a chance. And I guess with along that line, with JGI, like how many are you sequencing currently?
ALISON NARAYAN: Yeah. So– yeah, with our collaboration with JGI, we basically designed a library that includes hundreds of different flavin-dependent monooxygenases. And so there are different ways that we decided which sequences made it onto that list. So the majority of them are natural sequences. And so we built a library that was densely populated and sequenced space around a few enzymes that we have studied in depth.
So some of these that were initially identified by Russell Cox, like [INAUDIBLE]–
DAN UDWARY: Explain to my audience what a flavin monooxygenase does.
ALISON NARAYAN: Ah! OK. So some protein families are defined by their co-factor. And so, when I say flavin-dependent monooxygenases, I’m talking about the co-factor, which is flavin. One of the wonderful things about flavin proteins is that they’re bright yellow.
DAN UDWARY: Yeah.
ALISON NARAYAN: If the flavin co-factors is in the oxidized form, and so you see your proteins– very gratifying, when you do a protein purification out of this, like, neon yellow band coming off of your column and collecting that. So–
JACKIE WINTER: It’s a good day.
ALISON NARAYAN: It’s such a good day. And so the flavin is the co-factor and the kinds of flavin-dependent enzymes we’re working with the co-factor is bound throughout the catalytic cycle. So that co-factor stays in the active site. And then the other part of that, the monooxygenases, has to do with the mechanism. And so these enzymes use O2, and one of those oxygen atoms gets incorporated into our product, and one of those oxygen atoms from O2 ends up as a molecule of water.
DAN UDWARY: Mhm. So oxygenases, they’re oxidizing other molecules.
ALISON NARAYAN: Correct. Correct.
DAN UDWARY: Yep.
ALISON NARAYAN: And they’re beautiful flavins. They’re bright yellow, very fun.
So in our collaboration with JGI, we put together our wish list of sequences for flavin-dependent monooxygenases and there are different reasons each of those sequences made it onto the list. Some of them are natural sequences. Actually the vast majority are natural sequences, and we chose several of them to be highly related to enzymes that we have been studying since day one in my lab that do oxidative dearomatization chemistry on arenes.
The other reason that sequences made it onto our list is because we wanted to have diversity. And so we wanted a profile across that protein family to sample that sequence space and have as diverse of a protein library within that family as possible.
There’s one more category of sequences that we worked with JGI on, and those are ancestral protein sequences. And so starting from some of the enzymes that we had the most data on and knew the most about, working with my collaborator, Charlie Brooks, who’s also at the University of Michigan and a really talented graduate student, Chad, he predicted that ancestors, that those enzymes could have arisen from.
DAN UDWARY: OK, so sort of a phylogenetic reconstruction kind of a thing.
ALISON NARAYAN: Exactly, yeah. Working backwards in time.
DAN UDWARY: Yeah, OK, great. OK, what are we looking to monooxygenate?
ALISON NARAYAN: Oh, everything. [LAUGHING]
DAN UDWARY: Really?
ALISON NARAYAN: Well, we’ve asked different types of questions with our library, and so one of the questions we’ve asked this with this library is to try to understand the selectivity of the reaction and if there are certain sequence features that tell us about, or help us to predict where on the substrate that oxygen atom is going to be delivered in the product, and also the stereo selectivity of the reaction. So that is what phase of the starting material that oxygen atom is going to be delivered to–
DAN UDWARY: Right.
ALISON NARAYAN: –in the flavin-dependent reaction. So, we had some theories that we wanted to test out, about which residues were important. But one of the things that happened is when we would make substitutions to test that hypothesis through site-directed mutagenesis, we would end up with dead enzymes. So often, when you start monkeying around with your active site–
DAN UDWARY: Sure.
ALISON NARAYAN: –and making these blunt changes, you mess things up, which is maybe not surprising. And so taking this approach of using the– predicting the ancestors that could give rise to two different selectivities allowed us to resurrect those proteins–
DAN UDWARY: I see.
ALISON NARAYAN: –test their function, and they do the chemistry beautifully, which was very exciting, and also, then learn about the relationships of the sequence and the selectivity.
DAN UDWARY: Got it. OK, very cool.
ALISON NARAYAN: And one of also the amazing things that we learned through this study is that our predicted ancestral proteins are much more thermally stable than the natural enzymes that we’ve been working with.
DAN UDWARY: Really?
ALISON NARAYAN: And so we got some rocks of catalysts, like rocks in a good way in terms of their stability, that we can now work with in the lab. That’s something– ancestral protein prediction, reconstruction is a whole field. There’s a lot of work around this. And so that is something that people have shown in study after study that you can get to more thermally-stable proteins using this approach.
JACKIE WINTER: That’s fascinating.
DAN UDWARY: That’s weird. I hadn’t heard that before. Why is that? Why would that be the case?
ALISON NARAYAN: It’s a great question. If you dive into the literature, you’ll find different answers to that question. One that you will come across frequently is that we’re predicting the ancestors, right? We’re going back in time.
DAN UDWARY: Yeah?
ALISON NARAYAN: And the planet was hotter millions and billions of years ago.
DAN UDWARY: Like, really early on.
[LAUGHING]
JACKIE WINTER: Way back in time.
ALISON NARAYAN: It’s way back in time.
DAN UDWARY: OK. Like when these first evolved? Yeah.
ALISON NARAYAN: So that’s a theory that’s there. I don’t necessarily know if I buy into that. I’m curious about the algorithms that are used to do these predictions. And if there’s some feature of how the sequences that are predicted that gives rise to more stable proteins.
DAN UDWARY: Interesting.
ALISON NARAYAN: So I’m not sure why.
DAN UDWARY: Yeah.
ALISON NARAYAN: But it is–
DAN UDWARY: Really interesting.
ALISON NARAYAN: It is a result that has a functional benefit for us, being chemists working with enzymes in a lab.
DAN UDWARY: Totally, yeah.
ALISON NARAYAN: We’ve done this approach of the ancestral protein reconstruction in a few different families and we’ve seen that result consistently, getting to more thermally stable enzymes, which is always a happy day. In lab, when you’ve got a more thermally stable protein, less working in the cold room. You know?
DAN UDWARY: That’s right. Yeah.
JACKIE WINTER: So when you have these thermally-stable proteins, I mean, what– from your perspective, like, what temperature would you want to push them to be able to still be catalytically active or to still be–
ALISON NARAYAN: Yeah. So it’s an interesting question, and there is a lot of literature around enzymes from thermophilic organisms where their peak activity is at an elevated temperature. Right? We don’t necessarily see that the activity of our enzymes is better at higher temperatures, but it’s something that we’ll empirically determine with a given enzyme that we’re interested in studying to do a profile and look at the yields and kinetics of the reactions at different temperatures.
Yeah, and it’s not something that it’s like always the same. So I would say we run enzymatic reactions anywhere from 4 degrees to room temperature, to slightly elevated. We don’t have any reactions that we’re doing in boiling water.
DAN UDWARY: Yeah, OK. Do you have a– whether it’s in the JGI project or not, do you have a favorite biocatalysis story you could tell us?
ALISON NARAYAN: Oh a favorite biocatalysis story. You know, like every day.
[CHUCKLING]
DAN UDWARY: Every day there’s a new story?
ALISON NARAYAN: Next new favorite biocatalysis story.
DAN UDWARY: That’s great.
ALISON NARAYAN: Well, we talked a little bit about flavin-dependent monooxygenases, another class of enzymes that my lab loves to work with are non-heme enzymes, and we work with them in a couple of different flavors. So non-heme enzymes, it just means that they need iron.
DAN UDWARY: Right.
ALISON NARAYAN: They have iron in their active site, but it’s not in the form of a heme.
DAN UDWARY: A heme–
[INTERPOSING VOICES]
ALISON NARAYAN: So you don’t have a porphyrin around the circumference of that iron, we just have the iron itself litigated into the protein. So we work with Rieske oxygenases that are non-heme iron-dependent enzymes and also another class, it’s a mouthful to say, alpha ketoglutarate-dependent non-heme iron-dependent enzymes.
We need to come up with a catchy one because that doesn’t– a catchy name because it doesn’t really roll off–
JACKIE WINTER: No.
ALISON NARAYAN: Often, they’re abbreviated too from alpha ketoglutarate to just AKGs. In my lab, the students working with that class of enzymes call themselves the non-heme team.
JACKIE WINTER: Do they have shirts?
ALISON NARAYAN: I think we need shirts. [CHUCKLING]
Like screaming, put me on a t-shirt, non-heme team. They need like a whole superhero outfit, you know?
[CHUCKLING]
OK, so I will tell you a story that involves those two classes of enzymes, the non-hemes and also are lovely flavins.
DAN UDWARY: Great.
ALISON NARAYAN: OK, so we were interested in doing this two-enzyme sequence with one first reaction catalyzed by a flavin-dependent enzyme, second reaction catalyzed by a non-heme enzyme. They’re both enzymes that require oxygen for the chemistry to occur.
DAN UDWARY: Sure.
ALISON NARAYAN: And so we usually start on small scale in the lab. So we’ll run things in like little Eppendorf tubes, or in 96, or 384 well plates. But then the fun really starts when you’re scaling up, you know? How are you going to scale up your reaction?
I’ll never forget one day I walked in the lab and Tyler Doyon, who was one of the very first graduate students to work in my group, it looked like he had like a giant thing of like candy corn on his bench. And I was like, what is that? It’s beautiful . It’s like, imagine like a conical tube, like a 50 mil Falcon tube, that visually looks like a piece of candy corn. Right? It’s like colorless–
DAN UDWARY: Just layers of color?
ALISON NARAYAN: Colorless at the tip, bright yellow in the middle, and orange at the top.
DAN UDWARY: OK, uh-huh.
ALISON NARAYAN: I was like, “Tyler, what is that?” And he was like, “I’m making tropolone,” which is the product of that two enzyme sequence. Tropolones are bright orange in color.
DAN UDWARY: OK.
ALISON NARAYAN: So at the top where you have the interface between the atmosphere and the reaction, plenty of oxygen, everything’s going fine, the reaction– both reactions proceed through and you get the bright orange.
DAN UDWARY: Right.
ALISON NARAYAN: The product of the first reaction, the product of the flavin-dependent reaction, bright yellow, it turns out, and then in that tip, nothing was happening at all. No chemistry was going on. So–
DAN UDWARY: Because the oxygen is not getting there?
ALISON NARAYAN: Yeah.
DAN UDWARY: OK.
ALISON NARAYAN: Or we had consumed all of the oxygen that was in solution.
DAN UDWARY: Right.
ALISON NARAYAN: And so that became the rate limiting reactant. Yeah. Anyway, so that’s my favorite, and I’ve never seen a–
DAN UDWARY: That’s quite a visual, yeah.
ALISON NARAYAN: Recreated– to look quite that beautiful is like what Tyler was able to achieve that day.
And I don’t know if it’s related, but also like my lab loves candy corn. Candy corn–
DAN UDWARY: That’s just wrong.
ALISON NARAYAN: It’s kind of a–
DAN UDWARY: I don’t know what to tell you.
ALISON NARAYAN: It’s like a divisive candy. It’s like, people love it or they hate it.
DAN UDWARY: Hate it.
JACKIE WINTER: Yeah, I hate it.
ALISON NARAYAN: There’s no one that stands in the middle on that.
DAN UDWARY: Natural Prodcast stands against candy corn.
[LAUGHING]
JACKIE WINTER: We’re going to get that on a t-shirt.
DAN UDWARY: Yes, there you go.
ALISON NARAYAN: You know, I like, every year, I like to have a handful of candy corn and it starts out– and it’s good. The first couple of pieces, and then by the third, it’s like–
DAN UDWARY: You’re done.
ALISON NARAYAN: So sweet, so sweet. But then, imagine though– OK, my group– I brought in like a 10-pound thing of candy corn. I like put it in a candy dish in our group room. It was gone in one day.
[LAUGHING]
JACKIE WINTER: Oh my gosh.
DAN UDWARY: The non-heme team stands in favor of candy corn.
ALISON NARAYAN: It kind of was disturbing. I was hoping that people were, like hoarding it at their desks or something, but they confirmed that it had just been completely consumed.
DAN UDWARY: Grad students need calories from somewhere. If it’s not free pizza at the seminar, then it’s going to have to be the candy corn.
ALISON NARAYAN: I try to switch it up and bring in fruit and things like that in between the 10 pounds bags of candy corn.
JACKIE WINTER: Do you think we can get candy corn to sponsor the podcast?
DAN UDWARY: Sponsor Alison’s research?
JACKIE WINTER: Yeah, there we go.
DAN UDWARY: Totally.
ALISON NARAYAN: OK, so that’s maybe like my favorite memory of seeing a biocatalytic reaction, looking very cool but also then thinking about why does it look that way? It’s very interesting.
My favorite thing currently to do in lab is watch our robot. So we have a very simple, programmable robot, which is essentially a little arm with a multi-channel pipette and you can write a program that tells exactly what to do. You can get these things for like less than $10,000. And didn’t know if the group would use it, but just like, day one, set it up–
DAN UDWARY: Everybody loves a robot.
ALISON NARAYAN: Everyone loves a robot.
JACKIE WINTER: But the question is, does the robot like candy corn?
ALISON NARAYAN: I don’t know.
[LAUGHING]
But I can’t get enough of this thing. There’s something about watching this, like, set up thousands of reactions at a time, that’s just like so satisfying and so, I will often– I just– whenever– it’s an action, I go and I just stand there, and I’m like, oh this is so great. And I think back to when I was a graduate student doing total synthesis where our gold standard was, if you’re doing three reactions a day, like that’s a good level of productivity.
If you can do three reactions start to finish and understand what happened in each reaction, clean NMRs, like that was a good day.
DAN UDWARY: That was very productive.
ALISON NARAYAN: And now I look at this, and it’s like, it was amazing. You could do 3,000 reactions in a day, as long as you have a way to analyze all of them. And I think that’s one of the things I love about biocatalysis. You can really design your experiments in a way that’s data rich, and you can really just learn so much, which then enables you to ask bigger questions and accomplish bigger, lofty goals.
JACKIE WINTER: I think that– yeah that’s really neat. I think like– is your big picture idea, and one could envision a few years out, coming from a chemistry background, open up sigma catalog, you can pick out which reagents you need. Now you can almost go to a catalog, and you could say, oh I want to create this molecule. What enzyme could I go to that would install that moiety? Because maybe you can’t do it yet synthetically, because you have reactive groups around it. And so is that your goal, I guess in the next–
ALISON NARAYAN: Yes, you are describing my dream.
DAN UDWARY: OK.
ALISON NARAYAN: I think that there are several barriers to get to that place. So the first is, we don’t the substrate scope of many enzymes. And so we have limited information. There’s a lot of work to do, and that makes biocatalysis a great area to operate in, because there’s just so much information that we still need, and that’s the kind of information that chemists need to plan an enzyme into a synthesis and know they’re going to have that catalyst that will work to– or at least have a shot at it, right? To plan that in and make that a solid retrosynthetic design.
Also functionally, the supply of enzymes. Right?
DAN UDWARY: Yeah.
ALISON NARAYAN: And so, the three of us sitting here are not afraid of DNA–
[LAUGHING]
Or growing a little E. coli to make a protein of interest that you’d want to work in, but that’s not like a– it’s not a standard operation for most organic chemistry labs. And so, thinking about how do we, as a field, evolve to maybe even include that, or how do we make enzymes more available commercially? And it’s not new. Like you can buy panels of enzymes. Like you’ve always been able to get lipases and epoxide hydrolases, in a limited number. But what we don’t have is the same breadth as what we have and how you can buy ligands, right? We can’t buy enzymes the way that you can buy ligands at this point in time.
But that was going to change. It’s going to change.
DAN UDWARY: So how does that change? So like, so you know, I’m a data guy, right? And so I think about– I guess maybe what is your perspective then on being able to tackle that kind of specificity or function, or substrate promiscuity angle? Are we going to be able to ever predict that, do you think? From your experience with working with all these different kinds of enzymes and mutations and everything else, do you see a predictability to this that we will eventually be able to tackle? Or is that just going to end up too hard and too chemically weird?
ALISON NARAYAN: I think that it’s still a challenge at this point in time.
DAN UDWARY: For sure, it’s impossible right now. I think it’s impossible right now.
ALISON NARAYAN: But I think that is the direction we need to move in.
DAN UDWARY: Yeah.
ALISON NARAYAN: And we’ve seen work from Matt Sigman‘s lab, and Alán Aspuru-Guzik that shows the pyramidalization of phosphines and predictive models that allow you to choose the right phosphine ligand without having to screen.
DAN UDWARY: Yeah.
ALISON NARAYAN: Ad nauseum.
DAN UDWARY: Yeah.
ALISON NARAYAN: So a protein– an enzyme is just a big ligand. [LAUGHING] And so it’s more complicated, but I anticipate that we’ll be able to use the same types of strategies to develop models.
One of our limitations right now aside from having a much more complicated ligand to wrestle with computationally and pyramidalization is that we need good data sets. And also, with this question of substrate promiscuity, that’s also like another element that needs to be taken into consideration in building these models. And so I think that one of the reasons I’m so excited about my lab’s robot is that it gives us a tool to help us–
DAN UDWARY: You can generate that data.
ALISON NARAYAN: –generate–
JACKIE WINTER: Quickly.
ALISON NARAYAN: –those types of data sets and start moving in that direction, to at least have the data to do that. Even having that kind of data and sharing it with the community though can help inform how synthetic chemists might design an enzyme into a synthesis if you know what chemistry it does on a substrate, that’s in the neighborhood of a substrate that you might be working with.
DAN UDWARY: Yeah.
ALISON NARAYAN: And there are other tools that I think are really useful, that again are set up to take this data. One that I’ll point out specifically is a website created by Nick Turner’s lab. So Nick Turner is at the University of Manchester, and so he and a really talented student, Will Finnegan, and his group created RetroBioCat. And so they have many different enzyme classes represented their data from the literature, and that they’ve generated in their group where you can look at different substrates that they have data on for a given enzyme.
And so I think we just need that, but on–
DAN UDWARY: On a big scale.
ALISON NARAYAN: –massive scale, and then we’ll have that information for people, specifically chemists to really plan in a smart, strategic way, enzymes into an overall synthetic route.
Yes. So when is it going to happen? I don’t know. I think going to be in business for some time.
DAN UDWARY: Yeah.
ALISON NARAYAN: [LAUGHING]
DAN UDWARY: Getting physical things into tubes is always the– that’s the limiting reagent, I think.
ALISON NARAYAN: Yeah, maybe in the future chemists will have DNA synthesizers on their bench, and they can just make the DNA and do in vitro transcription/translation to make their protein and run a reaction that afternoon. Who knows? I mean, things– just the technology is evolving so quickly that enables this area of science. It’s another exciting reason, exciting thing about biocatalysis right now that I think is going to look totally different in five years than it does right now. So look out.
DAN UDWARY: All right.
ALISON NARAYAN: Enzymes are coming for you and your flasks.
DAN UDWARY: I’m not sure if that ends on a hopeful note or–
[LAUGHING]
ALISON NARAYAN: It was like hopeful, it wasn’t like a threat!
DAN UDWARY: It’s a good thing.
JACKIE WINTER: They’re coming for you.
DAN UDWARY: It’s a good thing.
JACKIE WINTER: Run.
ALISON NARAYAN: It’s a good thing, and they will play nicely with other types of reagents, like–
[LAUGHING]
DAN UDWARY: Yeah. OK. All right. Great. Well, Alison, thanks so much for talking to us today. This is really an excellent conversation. I’m so happy we did this.
ALISON NARAYAN: This was a lot of fun. Thanks for inviting me to be a part of your podcast.
DAN UDWARY: Absolutely. We’ll have you back. Thanks.
ALISON NARAYAN: Thank you.
Show Notes
- “Enzyme Library-enabled Chemoenzymatic Tropolone Synthesis”, (preprint) ChemRxiv. 2023. doi: 10.26434/chemrxiv-2023-lvngd
- “Enabling Broader Adoption of Biocatalysis in Organic Chemistry”, ACS AU. 2023. doi: 10.1021/jacsau.3c00263
- “Deciphering the evolution of flavin-dependent monooxygenase stereoselectivity using ancestral sequence reconstruction”, PNAS. 2023, 120, e2218248120. doi: 10.1073/pnas.2218248120