Natural Prodcast Episode 20: Brian Bachmann

DAN UDWARY: Hey everyone and welcome back to a new session of Natural Prodcast. I’ve got two things I wanted to tell you about in this introduction.

First and foremost, our interview this week is with Professor Brian Bachmann from Vanderbilt University’s Chemistry Department, where he is the PI of the Laboratory for Biosynthetic Studies. Brian and I are old friends and colleagues, having both graduated with our PhDs from Craig Townsend’s lab at Johns Hopkins. Brian went on to work at a groundbreaking company in the early 2000s called Ecopia, which was a very early genome mining company, back when DNA sequencing of natural product pathways was hard and expensive to do. As I said, Brian is now at Vanderbilt, and he’s doing some amazing research in and around drug discovery, which includes screening methods, genome mining in cave microbiomes, and using engineered biosynthetic pathways and enzyme systems to do synthetic analoging of some really complex bioactive natural product molecules. It’s a wide ranging conversation, and we could have gone on for a lot longer than we did, and I can say without reservation that Brian is one of the smartest people I’ve ever had the pleasure of getting to know.

Second, I want to take a minute to introduce a new voice to the Prodcast. Please take a second to join me in mentally welcoming Natural Prodcast’s new co-host, another old friend of mine, Professor Jaclyn Winter, or you’ll pretty much always hear me call her Jackie, from Pharmacology and Toxicology at the University of Utah. Jackie was a grad student while I was a postdoc at Brad Moore’s lab, went on to some cool post-doctoral work, and now her research in Utah touches on many areas of natural products biosynthesis and drug discovery.

If you enjoy the interview, I’d encourage you to come check out the links to some cool papers in the show notes, which you can always find at naturalprodcast.com or through the JGI website at jgi.doe.gov.

There’s a lot more to talk about there, and I’ll be getting together a proper introductory episode with her in the near future where we can dive into that. And I’m also dying to talk in more detail about the JGI Secondary Metabolism Collaboratory, which is the big huge monster thing that’s been keeping me busy for a long time now and almost ready for a real presentation to the world. … But, meanwhile, I’ve been sitting on all these great interviews recorded a few months ago at the SIMB Natural Products meeting and it’s time to get this podcast rolling again. I’ve committed to the JGI Comms team, and, therefore, to you, to put episodes out monthly, aiming for the first Thursday of the month moving forward. So, that’s the schedule, I’m committing to it, and you all can hold me accountable and yell at me if I slip again. My JGI coworkers are way too nice to yell at me, so that’s gonna be your job, audience. Send me those mean tweets and emails.

But anyway, today, please enjoy this great conversation with my good friend Brian Bachmann!

DAN UDWARY: Jackie, this is our first podcast together!

JACKIE WINTER: It is. Thank you for this opportunity.

DAN UDWARY: I don’t know if this is the first one we’ll put out. But this is the first one that we’re recording, so. And yeah, well–

BRIAN BACHMANN: Sometimes the first podcast don’t come out.

[LAUGHTER]

DAN UDWARY: That’s very true.

BRIAN BACHMANN: I have to say, sometimes the first “Prodcasts” recorded don’t come out.

DAN UDWARY: This is true. Yeah, yeah. So Brian and I– Brian was the very first recording I ever made practicing for the podcast. And– I did not know what I was doing.

BRIAN BACHMANN: Yeah, that makes two of us.

DAN UDWARY: So the audio quality was terrible. The hardware I was using was terrible. Now, we’ve got these fancy mics and everything.

BRIAN BACHMANN: All that talk about politics and religion too, it was not appropriate.

DAN UDWARY: Yeah. No, that was not going to fly. So yeah. So we’re giving him a second chance now that I know what I’m doing. And we have today Brian Bachmann from Vanderbilt.

BRIAN BACHMANN: Correct.

DAN UDWARY: Welcome, Brian.

BRIAN BACHMANN: Thank you. Good to be here.

DAN UDWARY: So Brian is a very old friend. We were graduate students together in Craig Townsend’s lab way back in the day. And then you did a postdoc for a little bit there and then went off to industry.

BRIAN BACHMANN: That’s right. Yeah, I went to a SIMB meeting, Society for Industrial Microbiology meeting and–

DAN UDWARY: Which is where we are today, recording today.

BRIAN BACHMANN: Yeah, in the year 2000. And I was trying to figure out what to do with myself after graduate school. I’d graduated in August. And the SIMB meeting was in August. And we started going to some talks. And at the time, it was really exciting when people would sequence [DNA].

The genome was– well, that hadn’t really happened yet. It was unpublished at that point. But people would push out gene clusters maybe once a month at most. Maybe twice a year, a full gene cluster would come out. And it was a very exciting event. And people in the research group would pore over it and argue about it and think about it and debate it.

But at the SIMB meeting, this guy at a company called Ecopia was talking about a gene cluster a week. They completed a gene cluster per week. And that really got me excited. And I stopped thinking about doing a postdoc and went to industry to see what they had, what data they had.

DAN UDWARY: And they had a lot of data for the time, right?

BRIAN BACHMANN: For the time. Nothing compared to now. But at the time, there was maybe, yeah, several thousand fully sequenced biosynthetic gene clusters from organisms that had not been previously evaluated. So we started to see this potential of looking for compounds based on these gene clusters. And that was my job at the company.

So I was the first chemist. So after having fun analyzing gene clusters for several months, they said, OK, now, you have to go get the molecules. So I built a program there to do that. Yeah.

JACKIE WINTER: Yeah, I think we actually use some of your information from those days. As a grad student, I think I use one of your clusters to help close my cluster at the time.

BRIAN BACHMANN: Oh. Oh, which one was that?

JACKIE WINTER: Napyradiomycin.

BRIAN BACHMANN: OK, yeah, yeah, we worked on that one. Yeah, we also worked in DIANs. We worked on that. We had separate projects with John Thorson and Ben Shen. [Listen here to Ben Shen on the Natural Prodcast.]. And they were supposed to be orthogonal. And then we much to our chagrin found that they had the same conserved polyketide synthase and then P3. We still don’t know why 20 years later.

JACKIE WINTER: It’s pretty amazing, though, how far we’ve come, though, too when you start talking about going after single gene clusters.

BRIAN BACHMANN: That’s right.

JACKIE WINTER: And it was a big deal. And you really had– I mean, the whole sequencing endeavor, had a cost, how much blood, sweat, and tears went into trying to identify that you had the right cluster before you sent it out for sequencing.

BRIAN BACHMANN: That’s right. And then just– and then in terms of there’s sequencing an organism by shotgun, which is what we did, we didn’t even sequence the whole genome. We just did a low-resolution shotgun pass where we sequenced 500 runs. And then we used those to find gene clusters in cosmid libraries and sequence those by shotgun.

We had a dozen people doing that to get that one gene cluster a week. It was a huge endeavor at the time.

DAN UDWARY: Yeah, yeah… Well, let’s back up a little bit. So maybe if we back up a little bit and talk about your research at Vanderbilt and maybe how that industry experience shaped what it is that you do today.

BRIAN BACHMANN: Yeah, so I left genome mining company doing natural product discovery and swore never to do it natural product discovery ever again. And I came– I had all these– I was really interested in synthetic biology. And I was really interested in mechanistic enzymology and biosynthesis. The field was doing it.

So I built a lab to do that. And then yes, we had several targets– mechanistic enzymology targets and biosynthetic gene cluster targets that we did.

DAN UDWARY: You were the first person I ever heard used the term “retrobiosynthesis.”

BRIAN BACHMANN: Yeah, that’s right. Yeah, and we had this– I had this idea while sitting in my bedroom in Montreal where my industry job was of evolving pathways one enzyme at a time from back to forward and using the terminal selection as a step. So it was just by retrosynthesis. And that’s what I really wanted to do. And that’s why I came back to academia. And I set that up. And we finally got that to work 10 years later.

But in the meantime there, we had people joining Vanderbilt who were doing drug discovery and neuroscience and other areas that were being very successful and taking preclinical candidates into the clinic and to INDs. And it seemed as though there had been a change from when I was in industry. Because we were coming up with new chemical compounds as well, getting the phase I trials, phase II trials. And at the time, nobody was interested in licensing that technology or doing anything with it.

So I was kind of discouraged. But then these– we had these– the Conn and Lindsley group at Vanderbilt came. And they started to do these with preclinical candidates and get lots of interest. And I thought, well, I know that we can do this. So in 2006 was when I started our cave-based natural product discovery program at that point. kept doing the other things as well. Yeah.

DAN UDWARY: Yes, cave-based. So yeah, down in caves. Tell us about that.

BRIAN BACHMANN: Yeah, so every natural product discovery campaign has to have a story. And it can be looking for drugs from marine organisms, in which case you get to go to exotic tropical locations and go snorkeling and scuba diving. And when my colleagues come back from these trips, they seem to be tanner and more rested looking than when they left.

And the problem with caves, of course, is they’re not quite as appealing as that. And we tend to– we actually maybe come out maybe a little bit paler than we actually went in and covered with bruises. So yeah. So it was– the story is, let’s look at an ecosystem that has never been exploited.

Because it’s been shown over and over that when we move into new chemical ecologies, new ecosystems, we find new chemistry. It worked in the marine ecosystems and marine actinomycetes and cyanobacteria. And we thought we could also– let’s see what caves have to offer because they’re unique environments.

These are structures that are millions of years old. This is — rooms that have existed for millions of years and relatively stable– no lights, constant temperature, constant humidity. And they were– so they were actually good environments for bacteria to live in.

So we thought, well, let’s see. But they’re also very, very low nutrients. So, they’re hypercompetitive. So our argument was that this is the kind of place where we want to look for hypercompetitive organisms that need to defend each other or subdue one another with small molecules.

DAN UDWARY: Antibiotics, yeah, to fight each other to over scarce resources. Yeah.

BRIAN BACHMANN: Or to communicate.

DAN UDWARY: Or communicate.

JACKIE WINTER: So I have a question. When you’re sampling the caves, do you go after particular sources? Do you go after sediment? Do you go after– if there’s any water– or I guess, when you’re down there, is it something that strikes your fancy when you start collecting samples for later processing?

BRIAN BACHMANN: Yeah, well, there’s a lot of– there’ s a couple of different schools of thought with that. So some people are interested in cave microbial– people are interested in the more extreme portions of caves where you have to go miles underneath the surface where no human has ever tread And that there’s very, very, very low nutrients down there.

And most of the energy is coming from chemoautotrophic and chemolithotrophic processes at that point. Those are very– those are deserts. They’re very, very sparse areas. And we have certainly– when we first started caving, we sampled off through these areas.

And you can do– with a teaspoon of soil from outside, you maybe have to do a 10,000-fold dilution to plate that out. The deeper you go in the cave, the less dilution. You can be down to 1 on 100 in a deep portion of a cave. So it just– it really, really, really falls rapidly.

But we eventually– after testing all these different zones, we eventually settled in on a intermediary zone where there was less nutrients but enough that you had a lot of biodiversity. So we choose environments that are close enough to energy inputs. So this is– for us, that’s dripping water, percolating through the soil through leaf litter and humic acids and those kind of things close enough.

And you can– we spot them by you can actually see consortia of microorganisms growing on the walls. They look like– they look like colonies, different colors, all sorts of different colors. I mean, where else in nature– and they’re not fungal. Where else nature can you go see colonies of bacteria just growing? They’re not monocultures there. They’re definitely consortia. There’s layer upon layer upon layer upon layer of them.

So we get them off of wet parts of caves– not dry, dusty, but wet. We like to get them where water is flowing and where formations are forming so that we have things to point out and say, we swabbed this stalactite. We used to swab that gypsum flower, that popcorn.

JACKIE WINTER: And when you bring back these samples to lab, you’re using culture-dependent method to look at that biodiversity. So how do you recapitulate the media that you’re trying to grow these organisms on?

BRIAN BACHMANN: That’s an interesting question. We’ve used a lot of different approaches. Some of them have been actinomycete-specific approaches. And you can enrich for rare actinomycetes by doing some tricks. Treating them with– boiling them in phenol, for instance, does a great job of–

DAN UDWARY: Really?

BRIAN BACHMANN: –of getting rid of all the non-spore formers. And yeah. And streptomyces don’t like that. So you can really, really enrich for non-streptomyces, actinomycetes, by mistreating your samples in various ways that a cave never experiences.

But we’ve also done things where we’ve grown organisms in ultra minimal media. So you could take International Streptomyces Protocol media number 2, ISP2, and dilute it by a factor of 100. And we got– we would find– we found organisms that will only grow on the minimal medium, which is probably more mirroring what happens in the cave environment. You get more unusual organisms that way too. But then they can subsequently be taught to grow on rich media by culturing them.

DAN UDWARY: OK, what’s the taxonomic distribution? Is it significantly different than what you might see in, say, soil or any other kind of environments?

BRIAN BACHMANN: No, I mean, they’re very rich environments. And they look a lot like most soil microbiome systems. And it makes sense because they are open. And they’re low. And things flowing– water flows into them and floods them on a regular basis.

So they are full of terrestrial microbes. So they’re open ecosystems. They’re not sealed off and closed. But there’s definitely types of organisms that are living there. And we’ve moved recently from actinobacteria now.

And we’re focusing much of our efforts on myxobacteria. And they’re also replete in caves, which makes sense because you have these biofilm walls. And it makes sense that there are micropredators. So our orientation in the cave project just now switching from intermicrobial interactions, studying intermicrobial interactions by imaging interactions between predatory and nonpredatory organisms–

DAN UDWARY: Oh, OK.

BRIAN BACHMANN: –from these environments.

DAN UDWARY: What would you say is the ultimate goal of this research then? Is it to understand that ecology? Or are you still doing drug discovery

BRIAN BACHMANN: Yeah, so we’re still doing drug discovery. So we have two separate projects. One is trying to understand the chemical ecology — the microbial chemical ecology, because we feel as though that stacks the deck for discovering molecules that do something.

And that’s a whole area of research. And it’s more basic science. And it’s less focused on translation. But we plug– we want to plug these organisms into a drug discovery platform.

So to me, the elephant in the room with regard to the genome mining-based approaches are that other than resistance genes, you have no way of knowing of whether or not a compound that you’re isolating is going to be active. And even if you do have resistance genes, it may not be better than something that already exists.

And the hit rate on if you have a target gene cluster and you have to find that compound, it’s low. It’s practically low. And you can get it if you want– if you want– if you really want to, you can do heterologous expression. You can do homologous expression techniques and stimulation techniques and that kind of thing. But it’s a lot of work.

So to me, to us, about 6 years ago, we thought the big gap in genome mining is not knowing if something is active before you isolate it, after isolating lots of things that were not active or new, that were not active. And so we built this as a platform that allows us for– it’s geared toward mammalian systems.

But it allows us to assay on a single cell level against cell lines and against primary cells isolated from patients up to 12 markers at a time. And these are phenotypic markers. So there are things like DNA damage. There’s a general marker for DNA damage, a gamma H2AX. Or you could look at apoptosis markers at caspase, or you could look at metabolism markers like phosphoprotein S6. So if there is a perturbagen present–

DAN UDWARY: “Perturbagen”–

BRIAN BACHMANN: A molecule that causes cells to be perturbed cells. We’ll find that. So it’s a generic– it’s a generic stress panel that we can now plug in. And it works on a metabolome. So we don’t have to– we can look at the metabolome level. We fractionated the metabolome into well plates. And we can use our fancy multiplexed activity metabolomics method to tell us what’s active, which peak is active before we even scale up.

The reason that we did that was to fit what we thought was a gap in the genome mining area, which is a potential wasted effort.

DAN UDWARY: Wasted effort– genome mining wasted effort. Come on, man. You’re killing me.

BRIAN BACHMANN: Well, it’s not. I mean, it– but it’s not. I mean, I’m a big believer and a big advocate of it, of the whole approach. But I mean, you live it. And you’re trying– you were at a company. And has it accelerated natural product discovery, microbial natural product discovery compared to activity-based screening?

DAN UDWARY: Oh, that’s a good question. Has it? I think it– I think it must have.

BRIAN BACHMANN: I mean, if you were to take an organism of interest and grow it under 20 conditions and look for cytotoxicity, do you think it would be faster than that?

DAN UDWARY: Yeah, yeah, yeah. Yeah, I think it’s an excellent point. So–

BRIAN BACHMANN: So that was why– that’s a major focus of the group. Now, the cave organisms are there as a raw material. Very familiar with working actinomycetes. We can also genetically manipulate them. We can do strain improvements. We can do all sorts of things.

So that’s just where we like to work. And it has the bonus of the chemical ecology projects, on the side. But we’re really– we are now doing genome mining projects where we’re targeting what we think are privileged classes of molecular diversity and then plugging that into these multiplex activity assays.

JACKIE WINTER: Oh, that’s great. I think– I mean, like you were saying, I think it’s great to be able to prioritize what to focus on because there’s so many gene clusters. But there’s [only] so many hours in the day. And you really have to be educated with what you spend your time on.

BRIAN BACHMANN: Yeah, and I think anybody in natural product discovery in any field and even 30 years ago would say it’s the [key to] success. And that is about choosing your battles.

DAN UDWARY: That’s right.

BRIAN BACHMANN: And then anything you can do to stack the deck. And genome mining is one way you can stack the deck. But it’s not sufficient by itself. You need more. You need other prioritization techniques, resistance genes or something.

DAN UDWARY: Right. Right. Right. Right. Right. Yeah, yeah, for sure. No, I think about this a lot in terms of what we’re doing with the JGI with the Secondary Metabolites Collaboratory and trying to have a good catalog of what’s out there so that we can do more comparative analysis and maybe– I don’t know– start to put some of these things into boxes more than they are right now.

But yeah, at the end of the day, getting access to the chemistry and seeing what a molecule actually does is not something that you can do with bioinformatics. And I think– I hope we don’t ever lose track of that. But definitely, in terms of me with what I’m doing in data wrangling, sometimes maybe I do.

BRIAN BACHMANN: No, but that stuff, that’s the raw material that you use to start to do the discovery. And more and more people are coming up with techniques. And my colleague Allison Walker at Vanderbilt is coming up with machine learning techniques just to look at clusters and start to predict what they do.

And that’s extremely exciting. That’s going to be sequence based. It’s all about giving you a good sequence, good curation, all that stuff, so. And I think that’s the other ways forward, then I’m going backward and looking at assays again in some ways. And maybe the forward way is these machine learning techniques and other things.

DAN UDWARY: Yeah, so let’s rewind a little bit more. What got you into natural product chemistry in the first place? What’s your origin story? I usually start with that. But I didn’t do that at this time.

BRIAN BACHMANN: OK, well, I started off in organic synthesis. And I worked as an undergraduate for a pretty well-known guy, infamous person, actually. His name is Tomas Hudlicky, who recently passed away. But he really got me interested in– and we were doing total synthesis. And I did that for several years. And it really got me interested in doing synthesis.

But I couldn’t stop thinking, well, these organisms are making these compounds. It looks a lot easier than what I’m doing. They’re breathing and pooping out molecules. And I’m using gallons of tetrahydrofuran and methylene chloride to make milligrams, in the end, of compounds.

And his other thing was that he had the biotransformation that he used in his work. And it was a chlorobenzene dioxygenase. So take chlorobenzene. And you bubbled it through a fermenter. And it came out with a cyclohexanediol, enantiomerically pure.

DAN UDWARY: Yeah, I remember this.

BRIAN BACHMANN: That’s what really, really impressed me. I thought, wow, if we could just harness what microbes can do and do synthesis. Can we do total synthesis in vivo? And most of my career could be described as trying to do that and failing for the last 20-plus years.

And that’s what led me to natural products. So after realizing that enzymes were specific for their substrates, that you couldn’t buy– there wasn’t a encyclopedia of enzymes at Aldridge that I could use to buy green uricase or “Wittig-ase” or whatever I was looking for. Then I backtracked. And I said, well, it’s possible to do this. And life has figured out how to do this.

So I started to look for people who were taking apart pathways to see how they work. So if you want to build watches, you should probably spend some time taking them apart. So that’s why I joined the Townsend Lab and moved away from synthesis and into biosynthesis and did that.

And in the meantime then, while we were in graduate school, then Frances Arnold’s papers came out, directed evolution, and Pim Stemmer. And at the time, I thought that problem’s solved. We have everything we need now to do this. And that’s what really triggered me to come back to academia. But then I realized at that point, after doing directed evolution for a few years, that there was also caveats with that.

And we had them– so synthetic biology now– and that would be the third research group area of my group now has become– so we call it “Fixing the Unfixable” projects, where you take a natural product that is too complicated to make by chemical synthesis and make analogs in combination of mutagenesis and direct evolution. Yeah.

DAN UDWARY: So it’s medicinal chemistry but with biological systems?

BRIAN BACHMANN: That’s right. And it’s actually not that uncommon. Or it could be called semisynthesis. So you can knock out an acyltransferase and then use a free hydroxyl group and do some chemistry and get around pharmacological problems.

But we wrote a review article that’s actually called “Fixing the Unfixable” in medicinal chemistry. And it has example after example either using synthetic biology or total synthesis or combination of those of overcoming pharmacological liabilities of natural products. So that’s about what about a third of my group is working on.

And they’re working on a molecule called everninomicin, which is an octasaccharide and contains orthoester linkages and a couple of aryl groups. And it’s a complicated molecule. But we’ve been able to generate about 20 analogs at this point.

And Casey and Nikolaou synthesized the everninomicin at Scripps, I guess, 22 years ago. And it’s over 130 steps to make everninomicin.

DAN UDWARY: Yeah, one of those.

BRIAN BACHMANN: So it’s not going to be analoging with that [synthetic pathway] anytime soon. So I’m pretty proud of that. We’ve been able to generate so many analogs of something that wouldn’t be doable using traditional chemical synthesis.

DAN UDWARY: Right. Right. Right. OK, and do they work? Do they do new things? Or it’s the same things, or?

BRIAN BACHMANN: Yeah, so we can make bioactive analogs that retain their activity and have different pharmacological properties. So they’re new compositions of matter. So one of the problems with the erythromycins and these everninomicins is that the intellectual property is dead at this point.

So no company would ever be interested in developing them, doing the substantial amount of effort to develop them because they’re not protectable, the intellectual property. But if you can come up with new compositions of matter that are even marginally better, then that changes the game. And you can reopen the book and come up with antibiotics for multidrug-resistant infections and that kind of thing.

So we’re still doing that, that thing that I started off doing as an undergraduate and not succeeding and still using all the tools from mechanistic enzymology to natural product biosynthesis to genomics to directed evolution– but it’s just the– it really is a straight line going right back to my sophomore year of college.

DAN UDWARY: So where do you see that in the future and, say, 10 years from now? I guess what I’m thinking is, it feels like if you’re working deeply in, specifically, in everninomicin and analogs of that, you’re really understanding one specific biosynthetic pathway. I guess how does this get bigger? Do you feel like you just have to know a pathway and really understand it in order to play with the chemistry?

BRIAN BACHMANN: So, it justifies mechanistic enzymology. It justifies taking apart pathways to understand how they work. You really do have to understand them. And I would say that if you looked at the grant applications for the last 40 years for studying biosynthetic pathways, all of them said that a rationale for doing this is to make analogs, improved analogs. But nobody does it! And not nobody, but very, very few people are actively engaged in doing it. And I think this is a huge potential for the field.

And you don’t have to discover new molecules. You can make new molecules. And you could– and you could study things that might be difficult to write a grant on right now like another P450 or glycosyltransferase. You probably would have a hard time justifying that basic science project.

But if you convert that to a synthetic biology project like [Jay] Keasling does, Taxol, artemisinins, or morphine analogs, then it’s a whole other area. So I hope that the field starts to merge with the synthetic biology community, with the metabolic engineering community because the skill sets are very complementary.

DAN UDWARY: So OK, yeah. So you don’t use a strictly synthetic biology approach to this? Or you’re just really like– molecular biology on plasmids and stuff to express things, or? What’s the–

BRIAN BACHMANN: Well–

DAN UDWARY: –I guess– I’m sorry.

BRIAN BACHMANN: –what you need is what JGI is providing, which is a giant warehouse full of microbial sequences. So let’s say you’re interested in improving a lead compound that you’ve discovered from microbes. Chances are you could find a dozen other producers of that compound. And–

DAN UDWARY: It’s getting there, yeah, for sure.

BRIAN BACHMANN: Right, and anybody who’s ever done any engineering work, you never want to stick with one organism. You want to have a dozen organisms work. Sometimes something won’t be soluble. Sometimes it won’t be active in a heterologous host.

So what you really need is a large tool kit of pieces in order to put them together to build pathways, to make novel compounds, or to find analogs. And that was Warp Drive. One of Warp Drive’s great ideas was to pick a lead and then use nature’s SAR by having large, large sequence collections.

DAN UDWARY: Warp Drive being the company that I used to work for, which is now part of Ginkgo. Yeah.

JACKIE WINTER: Now, I think that was a great example, too, with having a large resource to compare these gene clusters too if you find one of interest and there’s multiple organisms. I think Ben Shen has a beautiful example of this where an organism, you can beat it to death by culturing it. You just never find the compound. And then you take another strain. And it just produces it by just looking at it.

BRIAN BACHMANN: Exactly.

JACKIE WINTER: And so it’s nice to have alternatives to use and then figure out regulation and why certain clusters are more amenable to work with, or even strains. And I think then it lends more to microbiology. And the more we have, the more we can learn from those systems.

BRIAN BACHMANN: Exactly, and you can solve your problems by just finding a different producer. I mean, Bill Metcalf at UIUC used to go on about this. And they say they had a big genome mining program in phosphonates. And they could find one an interesting new gene cluster. And then they would find it in seven other organisms. And one of those would just produce tons, right?

DAN UDWARY: Right.

BRIAN BACHMANN: But I think it reaches into biocatalysis as well and biotransformations just for industrial manufacturing and getting away from petrochemical, petrochemistry, and going to the green chemistry. So I think it’s just a way for the field to really expand.

DAN UDWARY: But isn’t that really just genome mining?

BRIAN BACHMANN: Well, yeah.

[LAUGHTER]

JACKIE WINTER: It starts with genome mining.

BRIAN BACHMANN: Everything starts with genome mining. And a great database.

DAN UDWARY: All right, awesome. So you have a poster here at the conference. Tell us about that.

BRIAN BACHMANN: OK, so the poster is on something that’s very new for us, which is a drug development project. It’s one of these “Fixing the Unfixable” projects, where we have identified a lead using single cell chemical biology that has a really interesting activity against leukemia cells.

And we got motivated enough. And it’s remarkable activity and selectivity that we decided to identify the target of that organism. So we made chemical probes. And we did comparative proteomics. And then we did cryo-EM studies to find the binding site. And then we did CRISPR and directed evolution to evolve resistance in the active site.

So we really, really understand there’s a single target for this compound. It’s single-digit animal activity. And it’s actually an enzyme called ATP synthase. So it’s the last step in oxidative phosphorylation.

So now, what we’re trying to do is we’re trying to improve the pharmacology of it. It has some issues. It has a short half life in vivo. It’s toxic, has some toxicity issues.

So now, we’re trying to make analogs of that. And we’re doing this “Fixing the Unfixable” stuff, where we’re able to do semisynthesis with compounds that are the products of either the full biosynthetic gene cluster or gene knockout of the biosynthetic gene cluster to make a whole world of analogs, which we can then do classical pharmacological studies on from microsomal stability studies to pharmacokinetics and PKPD studies and that kind of thing.

So I’m trying– the whole poster goes from gene cluster down to an animal xenograft study. That’s where I am. And I’m super excited about that work right now. I’m putting a lot of effort into that.

DAN UDWARY: Unfixable in the sense?

BRIAN BACHMANN: Unfixable in the sense that if it’s going to require 130 steps to make an analog, it’s not fixable. But if we can then co-opt the life processes in making knockouts or knockins, doing some directed evolution, then it’s fixable.

DAN UDWARY: Got it. Yeah, OK, great.

JACKIE WINTER: Think green.

DAN UDWARY: Brian, thanks so much. This is really great conversation. So happy to see you. And looking forward to the next couple of days of the conference.

BRIAN BACHMANN: Should be great. Thanks.