4.5 Billion Molecules, One Mission: Cure the ‘Undruggable’ Diseases

Harsh Thakkar (00:05.594)
All right, we're live. I have a question for the audience. Just take a wild guess. How many diseases do you think are currently undruggable? So I went to Perplexity AI before this episode just to search for this and I couldn't find an answer. know, undruggable means, it's a term that refers to portion of some human diseases that
have been challenging to treat with conventional drug discovery methods. So we're talking diseases like Alzheimer's, Parkinson's, some genetic disorders, some types of cancer. So there's not a real number, but I have been looking at a lot of targeted drug discovery companies and also looking at companies that are using AI in a very innovative way. And that's how I landed on this
Philadelphia based biotech company called Third Law Molecular. And the guest I have today is Christian Schaffmeister, also known as Chris. He is the founder and president of Third Law Molecular, a biotech company based in Pennsylvania, Philadelphia area. And he's also a, has been a chemistry professor at Temple University since 2007.
And I have him here today because I want to understand the molecule that his team is working on and also understand the larger implications of what that means to the field of targeted drug discovery. So thanks for joining us, Chris. Welcome to the show.
Chris (01:49.829)
Hey, thank you. Thank you, Harsh, for inviting me.
Harsh Thakkar (01:52.984)
Yeah, I want to start off by asking you your work with the spiroligomer, spiroligomer. Maybe I'm not saying it right, but you can correct me there. Spiroligomer. Okay, thank you. Spiroligomer. So this molecule, and actually I should pull this up here on the screen while you talk about it. This is a new class. This molecule is pioneering.
Chris (02:06.472)
Spiroligomers is how we pronounce it. Yeah.
Harsh Thakkar (02:22.256)
sort of a new class of targeted therapeutics. To set the context for everything we're gonna be talking about today, can you explain how these molecules are different from traditional small molecules for biologic?
Chris (02:37.745)
Sure. So just when I was young, when I was in my twenties, I decided I wanted to make molecules that can get into the body and fix things. They could really like address proteins that we currently can't. And you can basically bin current therapeutics into two bins. We've got small molecules that bind into deep, greasy pockets on proteins. And those are typically about 500 Daltons or less.
following Lipinski's rules. But currently with those kind of molecules, we only target about 850 proteins in the human genome, which has maybe 30,000 proteins in it. More recently, we have gone for biologics like antibodies, and we can target proteins with those that are outside of cells, that are in the bloodstream or on cell surface receptors.
Where we're targeting is there's a big gap in size between those two where we don't make a lot of things. And the molecules that we're making are in that size range. They're typically from a thousand to 5,000 Daltons. So like five to 10 times or twice to 10 times bigger than a normal drug. And what's really special about them is that they have a whole bunch of fused rings that have embedded stereo centers.
that lets you control the shape of the backbone. So on the screen, you've got one up that this is an animation that we put on our website to sort of teach people how these molecules work. Yeah, and if you keep scrolling, you'll see these molecules are made up of all these rings that are fused together. And then within those rings, if you keep going, are all these stereocenters and we can control every one of them. So a stereocenter is a carbon that has four groups on it.
Harsh Thakkar (04:07.013)
Yeah.
Harsh Thakkar (04:20.645)
Hmm.
Chris (04:33.831)
And they can be arranged, those four groups can be arranged in two different ways. And so you have a choice of two different orientations on each of those lit up carbons. So there's, think there's nine of them in that structure. So there's two to that one, two, three, four, five, six, seven. Now that's a small one, seven. So there's like a 64, no, 128 different shapes that that molecule backbone can adopt.
Harsh Thakkar (04:44.56)
Hmm.
Harsh Thakkar (05:03.014)
Hmm.
Chris (05:04.315)
Then if you keep going, there are all these side chains that we have hanging off of the core scaffold. well, this is an illustration that if you change just one of those stereo centers, you will drastically change the shape of the molecule. You can twist one half of the molecule around 180 degrees relative to the bottom half.
Harsh Thakkar (05:28.006)
Okay, so it was one of those lit up pieces that if you change that you're changing this part and this part and how it's aligned. Okay.
Chris (05:39.205)
Yeah, the two halves of the molecule break apart, one half of it flips over 180 degrees, and then in the animation, it gets joined back together again. What you're doing is you're creating another molecule, that has a completely different shape and will have different properties. And it's sort of like a molecular Lego set, where we have these Lego bricks that you can snap together in different ways to make different shapes.
Harsh Thakkar (05:52.966)
Hmm.
Harsh Thakkar (05:57.318)
you
Harsh Thakkar (06:08.954)
That's interesting. love how you face that and how you put the example of the Lego bricks, that's exactly what I was thinking.
Chris (06:09.339)
This is.
Chris (06:18.107)
Yeah, I do want to say, give credit where credit's due, that term molecular Lego was popularized by Fraser Stoddard, who won the Nobel Prize a couple of years ago. yeah, so. Right.
Harsh Thakkar (06:21.135)
yeah.
Harsh Thakkar (06:26.118)
Okay. Okay.
Harsh Thakkar (06:31.952)
So yeah, go ahead.
Chris (06:34.681)
And then hanging off of this core scaffolding, we have all these side chains that are like side chains of peptides. And we can put hundreds of different groups into each position and choose from the menu. So it just fits in more with that molecular Lego idea that we have these bits and pieces that we can put together in different combinations to make different three-dimensional shaped molecules.
Harsh Thakkar (06:48.986)
Hmm.
Harsh Thakkar (07:03.44)
This is so interesting how I, I don't know who designed this animation for you, but this is very, very interesting.
Chris (07:13.969)
Thank you. Yeah, yeah, it's a, we have a wizard for a web designer.
Harsh Thakkar (07:20.006)
Okay. thanks for explaining this. Because sometimes when we're talking about these topics specifically for audience that doesn't understand these terms, it's easier to see something while you're talking about it so that they can understand what you're saying and pair it with an analogy of like Lego bricks, right?
Chris (07:44.135)
Sure.
Harsh Thakkar (07:48.286)
just to get the point across. What specific challenges do you see these molecules address in the field of targeted drug discovery that maybe the traditional ones cannot?
Chris (08:04.763)
Yeah. So we have evidence, we've demonstrated that molecules that are a bit bigger than these ones, they're about 1500 daltons. So like three times the size of a normal drug, they can get into cells by passive diffusion. So since it can get into cells, we could target proteins inside of cells with these molecules, even though they are quite large. And since they're large, they can bury a lot of surface area. And we think we're going to be able to
bind surfaces of proteins inside of cells, which would let us manipulate protein-protein sort of communication pathways. It would also let us bind to what are called intrinsically disordered proteins. So proteins that don't have a lot of shape to begin with, that are involved in a lot of diseases like diabetes, cancer, we can make molecules that have nice, well-defined shapes that could bind to them and stabilize
Harsh Thakkar (08:50.64)
Hmm.
Chris (09:04.283)
the disordered proteins in well-defined shakes.
Harsh Thakkar (09:09.719)
Interesting. so as a company, your team is working on these molecules, so where are you currently today? Like, are you testing these in different scenarios, or what are the applications for these molecules? Anything that you can share here?
Chris (09:31.451)
Yeah, the big thing that we're doing is because these are really modular, we are making very large libraries of these molecules so that we can screen them against any targets we want. So first we're building a search engine for these molecular LEGO. In July of last year, we completed a library of four and a half billion of the largest molecules, most complex molecules outside of nature.
Harsh Thakkar (09:35.386)
Mm-hmm.
Harsh Thakkar (09:42.957)
I see.
Harsh Thakkar (09:47.238)
Hmm.
Chris (09:58.459)
This is where we take four of these segments and we connect them together like fingers on your hand and attach them to a DNA barcode that codes for each one of them. So we swap different molecules out in each position, different segments, and created a molecule that's like four fingers on your hand, and they're big enough to wrap around proteins. And we can search this huge space of new space of molecules.
Harsh Thakkar (10:10.278)
Mmm.
Chris (10:28.195)
in a matter of a few days using this DNA encoded library technology.
Harsh Thakkar (10:33.734)
Hmm.
Chris (10:35.067)
The other thing that we're doing right now is making a library of a single segment where the groups that we're swapping out are the individual building blocks, the individual Lego. And that's going to be like a high resolution search tool for these molecules.
Harsh Thakkar (10:51.234)
I see, okay, so like a set of all the square pieces, a set of all rectangular pieces or like unique pieces, something along that line, okay, okay.
Chris (11:02.555)
Yeah, yeah, the different building blocks have, we have four different core building blocks. We put them together in different combinations, four at a time. And that gives you four to the four different backbones that you can make. And then all the side chains that we can decorate them with, it's just an astronomical number. So we're making a hundred million to begin with on that library. With those two libraries, we'll be able to search for both big molecules that bind proteins
Harsh Thakkar (11:07.654)
Mm-hmm.
Chris (11:32.385)
and these individual segments that we know get into cells.
Harsh Thakkar (11:37.132)
Okay, okay. yeah, go ahead.
Chris (11:38.619)
And then we're going to target all sorts of proteins, proteins involved in immunology, immunotherapy, cancer, metabolic diseases. We have a long list of targets that we're going to purchase and then screen against these libraries to sort of cherry pick what are the best molecules that we can discover to move forward.
Harsh Thakkar (12:06.18)
Hmm. Interesting. Yeah, because I think I don't know where I read that, but I was going to ask you about the library because I did read in some news article that the platform had 4.5 billion molecules in there. So that's an amazing... How long did it take your team from like starting from scratch to like get up to that point?
Chris (12:37.022)
I would say it took us about a year because we had to develop the chemistry. We did this with a wonderful company up in Boston named Xchem. These people are awesome chemists. And we had to develop some new chemistry to get this to work on DNA and then put it together. But they've told us, you know, it set new standards in terms of quality of DNA encoded library synthesis. And it's been a, that was a great collaboration.
Harsh Thakkar (12:40.303)
Okay.
Harsh Thakkar (12:47.259)
Mm-hmm.
Harsh Thakkar (13:06.02)
And I'm guessing there was AI involved in some way or shape here in designing these or.
Chris (13:11.783)
No, not yet. AI, so I do a lot of computational chemistry and I'm writing a lot of software to design these molecules. And the software is going to work something like Rosetta, which is a program developed by David Baker's group. He just won the Nobel Prize last year for this software. We're using a lot of those same algorithms to develop
Harsh Thakkar (13:15.855)
Not yet. Okay.
Chris (13:40.903)
software that can design spiruligamers that bind to proteins of known structure, where we can sort of fit this, find a spiruligamer that will fit into the surface of designed, of a known protein. Our goal then is to use that data to train Aon models to be able to build spiruligamers from scratch, to sort of hallucinate them up against protein surfaces.
Harsh Thakkar (13:47.621)
Hmm.
Harsh Thakkar (14:10.778)
Very interesting. Yeah, this is, and you, so your chemistry background, obviously, you know, working at Temple, how did you, like, what I wanna also ask you is, when did this idea of third law molecular form? Like, was it, were you doing some research at Temple?
where you were like, okay, I need to spin off or I need to build this company. How was third law molecular born? I guess that's my question.
Chris (14:47.207)
So I've been funded by the Department of Defense for about 18 years to develop this technology for a variety of applications. And in 2018, I told them I knew how to scale it up. I knew how to make kilograms of the building blocks and make millions and billions of molecules. And they said, I could use these to replace antibodies in diagnostic tests. And they have a lot of diagnostic tests. And they said, great.
Harsh Thakkar (14:54.758)
Okay.
Harsh Thakkar (15:11.248)
Okay.
Chris (15:15.015)
Chris, we love the idea, but we're not gonna fund your lab because we don't get much out of academic research. We want you to start a company. So they gave me a bunch of money and they told me start a company, hire some people, build a lab, and that's what we've been doing for the last five years. As we're developing and getting the technology scaled up for creating these molecules as diagnostic molecules, I've always had an eye on therapeutics. want to develop, I wanna develop.
I'm very interested in diseases of aging, primarily down the road. And so I want to create new tools that would allow us to tackle those. So things like Alzheimer's. so as we are developing this with the Department of Defense, I'm also sort of pivoting to develop these as therapeutics. We took some of the molecules and we fed them to mice. We showed that they have oral availability.
They're well tolerated. They get into cells. We did some of those experiments that were fairly cheap and now we're raising money to develop these as therapeutics.
Harsh Thakkar (16:23.994)
Very cool, yeah. That's an interesting story and also how fast, like you mentioned that you, there's also another thing that I was looking on your website, which is pretty interesting if anybody wants to check it out, is a table of how these molecules compare to other molecules and you have different check boxes and things of that nature.
That was a very interesting insight when I was reading about this. You've also talked about this term Goldilocks molecule or finding the perfect fit, right? So I'm guessing the idea comes from the story of Goldilocks where things need to be just right and you've introduced that concept.
to in finding a molecule that's just right for the job. So can you break down what, how did you come across that term or like what it means in.
Chris (17:29.967)
Well, I don't want to claim it because that's an old, you know, it's the old idea, right? Finding something that's just the right size. And I think it where it comes down to is a basic physical chemistry that when you want to tie two molecules together, you have to bury a certain amount of surface that you get interactions between those two molecules and you have to bury a certain amount of surface area. Small molecules do it by
Harsh Thakkar (17:33.542)
Yeah.
Harsh Thakkar (17:49.35)
Mm-hmm.
Chris (17:58.181)
burying themselves inside of caves. If you want to bind to surface, you need a bigger molecule. And that's where the Goldilocks idea comes in. You need something that is not too small. The not too big part, you're not going to get into cells if it's too big. So that's the not too small and the not too big part. But if you're just the right size,
about, I think about 1500 Daltons and you're structured, you can bind to a protein surface and still have good drug properties, still get into cells. That is our hope. That is what we're, what we believe we can do and what we're going to be shooting for. There is another large class of molecules that's become very popular in the last couple of years called cyclic peptides, where you take a peptide and you tie it into a ring and
Harsh Thakkar (18:31.013)
Hmm.
Harsh Thakkar (18:45.414)
Hmm.
Chris (18:49.047)
I invented one type of cyclic peptide called stapled peptides with Gregory Verdine. And it's been the basis of two companies, of Aleron and now more recently, Parabolous. So that's where you take a, a cross link across one or two turns of an alpha helical peptide. Those are also considered Goldilocks molecules. And I think Merck has really been using this term a lot with their PCSK9 inhibitor. It's a cyclic peptide that's going to be used as a cholesterol drug.
Harsh Thakkar (19:09.872)
and
Chris (19:18.407)
There's a lot of excitement in this idea of making molecules that are about, you one to two hundred, two thousand Daltons that are bigger, that can bind to the surfaces of proteins. There's a lot of excitement.
Harsh Thakkar (19:18.839)
I see.
Harsh Thakkar (19:31.878)
Yeah, and what are like other, so you gave us some examples of companies that are researching in this space and using this concept of Goldilocks molecules. What kind of diseases or conditions is that approach going to be really beneficial for? Do you have any ideas?
Chris (19:56.881)
Yeah, so, you know, the simplest one, the oldest one, cancer. There is an off switch for cancer that we've known about forever. It's called MIC. It's a protein that when it dimerizes with another protein called Max, it turns on cell proliferation, cell division. And if you could make a molecule that would bind to MIC and get into cells, you could switch off cancers.
Harsh Thakkar (20:01.775)
Hmm.
Chris (20:26.853)
That is that protein has no deep pocket. is no way you're going to get it. Well, I don't want to say no way. There's always could be something, but it doesn't have a deep pocket. It's it's it's considered like the one of the biggest undruggable targets. And that's one that we'd like to go after this, particularly because these spear ligamers that we make are long and extended and they could.
they could bind to the long extended helix of of Mick.
Harsh Thakkar (20:59.364)
Right, right. And you also mentioned at the top of the episode that you were working with this company in Boston that's really good in chemistry. Are you collaborating with other people within the industry just to learn about other developments or collaborating with them specifically to take the work you're doing to the next level?
Chris (21:24.187)
Yes, we're reaching out to pharmaceutical companies. mean, you need to work with Big Pharma. They're the only organizations that can do clinical trials. You you need a lot of resources to do clinical trials and to get things into the clinic. So, yes, we are reaching out to them. We do, our molecules are new. They're a new modality. There's some risk. There's the newness of it. It takes some time to get people on board with that idea.
Harsh Thakkar (21:28.879)
Yes.
Harsh Thakkar (21:48.752)
Mm-hmm.
Chris (21:53.893)
But I look at these and I see, like when I look at natural products and the kind of molecules that nature makes, the most active natural products are molecules that have a lot of fuserings from everything from steroids to morphine to things like brevetoxin and mitotoxin. Nature has figured out that if you want to make a molecule that has the most bang for buck.
that the least amount of molecule will bind tightly to a protein. You take your atoms and you tie them into a bunch of fused rings so that it buttresses itself, so it stabilizes itself. And that's the kind of idea that I took when I was young and ran with to try and come up with a modular way to do that.
Harsh Thakkar (22:42.402)
And you mentioned about, so like how these different therapeutic areas and also that because this is a new modality, there is betting on the unknown because yes, there is enough science behind it, but going from a lab to clinic is for a new molecule or for one that has been around for ages is a challenge because.
human body is so complex and when you run it into the trials, you ultimately have to show that data to the regulatory authorities to move to the next milestone. So how, if everything goes according to the plan, you find the right collaborative partners to take this into clinical trials and you get some unique or fast track designation because this
Chris (23:13.115)
So complex.
Harsh Thakkar (23:39.238)
these molecules are going to be solving what is undruggable or uncurable out there. So you might get some leeway to move faster. In your mind, how many years do you think until this is completely tested and commercialized, how many years do you think would it take?
Chris (23:59.911)
Oh my goodness. Um, um, I'm hoping like five, you know, five, not more than 10. It's, um, yeah, I'm 60 now, so I got to get cracking on it. Um, if the, the, the thing about our molecules is that we are become very good at synthesizing them, putting them together. put them together on robots now with peptide synthesizers.
Harsh Thakkar (24:05.594)
Okay.
Harsh Thakkar (24:10.266)
Okay.
Harsh Thakkar (24:24.507)
Hmm.
Chris (24:29.487)
And so even with my little team, we can make like 60 of these a week. So I think we can move pretty quickly in a drug discovery program.
Harsh Thakkar (24:37.614)
Interesting, yeah. And that's the thing because there are lot of companies that are, especially the new modalities or new areas of drug discovery where it's hard to anticipate the different challenges until you actually go through the different phases of the trial or
you know, run these experiments. So when you think about that path, like you said it's five years or let's say 10 years, whatever, have you ever reflected like, this could be a roadblock two years from now? Or something along those lines, can you share?
Chris (25:24.504)
Ahem.
Chris (25:28.615)
Yes. mean, five years ago, knew the roadblock was scaling this all up. Excuse me one second. I knew that the roadblock was scaling this up to be able to make these molecules at scale, large numbers of them at large quantities. We've spent the last five years really figuring that all out. These are just made out of carbon, hydrogen, oxygen and nitrogen. They're like peptides.
Harsh Thakkar (25:34.18)
Mmm. No worries.
Harsh Thakkar (25:44.838)
Hmm.
Chris (25:58.107)
They just have this ladder backbone instead of a flexible backbone. I don't see, and we decorate them. The functional groups we've chosen from lots of polar rings that are in drugs that are already being used. So we're using groups, we're decorating them groups that are already in drugs and so are known to be, you know, they're not toxic in and of themselves.
Harsh Thakkar (26:14.996)
yeah.
Chris (26:27.843)
So by putting that together, think these molecules, I don't see any inherent safety issues with these molecules. Not from any kind of off target or chemical reason. Yeah, scaling up was a big, big challenge. And that we did with the help of the support from the Department of Defense.
Harsh Thakkar (26:42.79)
Hmm.
Harsh Thakkar (26:53.838)
Interesting, yeah. And it seems like the way you've articulated the molecule and the different steps and how it works, sounds like you mentioned, you've kind of solved that challenge and you're now taking this into the next stage, which is finding real world applications or collaborative partners where you can use them.
Chris (27:18.215)
Exactly.
Harsh Thakkar (27:20.612)
When you look at this area of targeted drug discovery, do you see any other companies or other scientists or your peers, like what puzzle are they solving that you're like, wow, that's interesting?
Chris (27:36.931)
okay. So I'm really interested in what people are doing with cyclic peptides. like I've been very carefully following, read all the papers that Merck put out on the development of their cyclic peptide for cholesterol, their PCSK9 inhibitor.
Harsh Thakkar (27:47.504)
Mm-hmm.
Chris (28:05.573)
No, that's a fascinating story because they started with a cyclic peptide that came out of a library, just like we make it. And then how much work they had to do to turn it into a useful orally available drug. And I would sum it up as basically they needed to put additional cross-links across the inside of the ring to stabilize it more, to make it more structured and more buttressed, to make it be able to bind the target.
with very, very high affinity. We think with our molecules, they already come with that kind of structure. There's no additional cross-linking that's necessary, I think. We hope that we can move faster with that. But I am always thinking about how potent are we gonna be able to make these molecules? What are the difficulties gonna be there as we develop them in potency and
Harsh Thakkar (28:36.474)
Mm.
Chris (29:04.421)
as we put them into animals and into animal models to see what kind of safety profile they have and what kind of degradation pathways they go through, things like that. The ADME, the PKPD stuff is still open question.
Harsh Thakkar (29:16.166)
And
Yep.
Yeah, just so I, so this is another area that I don't know much about, but like, so this molecule, let's say when you collaborate with a company that is running this into the clinical trials or using it and, you know, maybe improvising it. what, like, how would this work? Like would they,
Because the molecule is trademarked to your company, right? What somebody else does using it to do some trials and figure out, you know, clinical applications, would it be like a collaborative partnership between you and that biotech or pharma company and your team would get, you know, I don't know, some proceeds because you own the molecule or how does it work?
Chris (30:23.013)
we haven't done it yet. I'm, I'm what the way I think it's going to work sort of scientifically is, we will.
Harsh Thakkar (30:25.136)
Okay.
Harsh Thakkar (30:30.576)
Mm-hmm.
Chris (31:02.902)
So that the way I see this working is a drug company would could come to us in a strategic partnership and ask us to screen, you know, a few targets that they've had difficulty with current modalities.
Harsh Thakkar (31:11.695)
Yep.
Chris (31:31.274)
We can screen it against our libraries, identify some lead compounds, and then work with the chemists, the scientists at the pharma company to down select the list that we would carry forward and then work closely with them as we redesign new molecules, synthesize molecules, test them with the help of the company and progress them through to an IND stage, progress them through those kind of studies.
Harsh Thakkar (31:31.366)
Mm-hmm.
Harsh Thakkar (31:59.555)
I see.
Chris (32:02.364)
If we get investment from VCs and they want us to focus on just a few targets, then we would be doing this kind of development internal.
Harsh Thakkar (32:13.638)
Okay, that makes a ton of sense. And it's been really an interesting conversation. When I was preparing for this episode, I have to admit it was difficult because I had, this is like way out of my league because I work mostly in software and FDA compliance, that side of the life sciences industry.
I don't really talk to a lot of scientists or have these kinds of discussions, but that's why I have this podcast so I can learn and understand. It's much easier for me to have this kind of conversation with you and let you, the expert, behind the science, talk about it rather than reading a paper, which I'll never do anyways, right? So thank you so much for coming here and explaining it in a very concise and approachable way.
Chris (33:10.878)
Harsh, thank you so much for inviting me. This has really been a pleasure.
Harsh Thakkar (33:14.574)
Yeah, and for whoever was listening to this episode or watching, learned about your company, maybe they want to follow the news and everything that's going on, what's the best way for them to follow you and the company?
Chris (33:27.99)
Yeah, so our website thirdlawtx.com, you know, has some information on it. For some of the more speculative things that we I've thought about for this Spirulium technology, I have a couple of presentations that are up on YouTube. And yeah, that's you know, reach out to me. I love to talk about this and find especially
Harsh Thakkar (33:48.847)
Okay.
Chris (33:57.562)
specifically in terms of drug discovery, I really want to push these forward as new therapeutics. I they could have tremendous impact.
Harsh Thakkar (34:06.958)
Yeah, yeah. No, thanks for, I, you know, just hearing you talk and explaining this, it seems like you have a really great understanding of the wide applications that this technology could have. And, you know, you've worked with the right people to bring it up to where it is and where you're planning to go.
So I'll definitely be following your company and you, and we'll link those videos that you mentioned about in the show notes of this episode. So if the audience enjoys this, they can also go watch those videos and learn more about the company. But yeah, thank you again for coming onto the show, and I wish you and the team all the very best.
Chris (34:51.53)
Thank you so much.