Many species have developed amazing mechanisms to cope with various drivers of aging. We have previously interviewed two prominent experts that study those evolutionary marvels: Steven Austad and Emma Teeling. However, this research is not purely academic anymore. A handful of bold entrepreneurs are trying to go commercial, and one of them is Ashley Zehnder, DVM, PhD, co-founder and CEO of Fauna Bio, a biotech startup that looks for protective genotypes in animals such as hibernating squirrels in order to weaponize them against human diseases.
I must ask: how did you muster enough courage to start such a company?
Myself and my two cofounders, who were in the same postdoc lab at Stanford, realized that we had a unique set of skills. My background is more from a clinical, translational side: I come from a veterinary background of comparative physiology, clinical disease management in different species. So, I’m always thinking about the clinical impact of the work that we do.
My cofounders were the two other legs of that stool. Linda (Goodman) came from a human genomics background and realized the power of using evolutionary genomics to try to highlight and prioritize targets that would be more impactful for human disease, simply speaking, using a hundred million years of evolution to shine the spotlight on what are the most important parts of the human genome. Katie’s (Grabek) background is in human genetics, but she also has particular expertise in this one species, the 13-lined ground squirrel, and in building an amazing databank of very precisely timed samples throughout a very dynamic hibernation time course. All this gave us a starting point.
It’s not all that dissimilar to how BioAge started with their biobank of centenarians, I think. We’re trying to look for aging protection in a particular and highly specialized biobank. It’s a similar approach in that we’re looking for those extreme phenotypes, but here, we’re looking at highly conserved genes. The platform that we built at Fauna is optimized to rapidly translate insights from genetic changes in these extreme model organisms, then map them to human diseases and rapidly identify compounds that we can move into the clinic.
The problem is that a lot of work that’s done in academia is not translationally focused. A lot of investigators will focus on one species, or even an organ system within a species, and they’ll just try to characterize it very well, but not always with an eye toward how this relates to an unmet need in human populations. That’s exactly the niche we fill, that link between the two, by trying to find the best labs and the most interesting new model organisms, and then linking that directly to disease signatures in people.
I suppose that’s easier said than done. Could you explain Fauna Bio to me? That is, the whole process, step by step?
Sure. We at Fauna start from animal biology. We start from tissues at time points and species that show natural protection. Animals that are protected from scarring or fibrosis, animals that can reverse early signs of neurodegeneration, tau phosphorylation in the brain, that can connect and disconnect neuronal dendrites in the brain.
A lot of those natural phenotypes evolved as protective mechanisms. Say, hibernating mammals. As they go down into the depths of torpor, they go to almost freezing, and they must have robust mechanisms to protect their own tissues from damage.
Every organ in the body – brain, heart, lung, kidney, liver – must be protected from periods of low oxygen, oxidative stress, inflammation, low temperatures. They must change their metabolism. One of the programs we’re working on right now is a gene that is essential for the brain to be able to switch from glucose to alternative fuel sources like beta-hydroxybutyrate and lactate. That’s how the hibernating brain stays alive even when it’s almost frozen.
Essentially, what we do at Fauna is define protective gene signatures in specific tissues at specific time points where those tissues are physiologically protected from damage. We then define that gene expression signature and map it to human diseases.
We have a knowledge graph called Centaur that brings in a large amount of genomic data, including data from UK Biobank and other repositories such as DisGeNet where people have defined what a gene expression signature looks like – for example, for a heart failure in humans. We can then say, “Okay, we have squirrel heart tissue where we know these animals are protected at these specific time points,” then we literally overlap those in space and ask, “What are the genes that go down in the protected species but go up in humans with heart failure?”
We then can statistically enrich for networks where we see this opposite regulation, where we have proof through nature that these genes are protective, if we can alter their regulation. Then, we can take those genes as genetic targets either through an AAV modulation, which we’re doing for one program, or we can map that gene expression signature directly to small molecules. We have a component of our platform called LEOTM, which maps compounds to gene expression signatures. This data has been optimized from an NIH dataset called L1000.
People will often use that dataset for repurposing by taking a disease signature from a human disease setting, computationally reversing it, and trying to find a compound that matches. We, on the other hand, just find this reverse signature in nature. We say: ok, where is nature optimized to reverse this damage? Then, we just map it directly to a small molecule.
That’s how we found our lead program, Faun1003: by looking how the 13-lined ground squirrel responds to low oxygen. We found a signature that was highly protective in squirrels and mapped it to a small molecule that we are now internally optimizing for use in pulmonary fibrosis.
I understand that, at least for now, you decided to bet on small molecules. Could you explain this decision?
At the core of Fauna Bio, we’re modality-agnostic. We just try to find the best genetic target for a disease that we’re working on. Internally, we’ve optimized the platform to be able to rapidly find small molecules that hit our genes of interest, because it’s a good starting point.
But, not all targets that we find are amenable to a small molecule approach. So, the second program that I mentioned, the metabolism program, is an AAV therapy program, where we are working on a specific gene that turns on in the brain when it needs to switch to beta-hydroxybutyrate or lactate.
As other investigators have found, that gene is very relevant for diseases like retinitis pigmentosa. So, we’re looking at that as a potential gene-agnostic therapy for retinitis pigmentosa, along with some other genes. Internally, we’ve built a few more modules for small molecules development, but we’re not exclusive to that.
Can you give me an example of one such animal superpower and how you worked with that?
Let’s take the 13-lined ground squirrel, which is one of the best-described models of deep hibernation. Every organ in their body is adapted to survive extreme changes in oxygen levels and temperature. They can increase their metabolic rate 235-fold in an hour. That’s, by the way, one of the reasons we are partnering with Novo Nordisk on programs for obesity: they’re looking for ways to increase energy expenditure in humans, particularly in those that have lost a lot of weight on drugs like semaglutide. After losing weight, people reset their metabolism to a lower level. It’s known as “the biggest loser effect”. You lost a lot of weight, but then your body slows down metabolism, and you’re stuck there. It’s hard to change this baseline, but these animals do it every couple of weeks.
That was the basis of that exploration. We have other partnerships around other species, we’ve expanded beyond the 13-lined ground squirrel. We now work with a lab at UNLV with Frank van Breukelen. He works on a species called tenrecs. They’re amazing, they look a bit like hedgehogs, but they are actually closely related to African elephants.
What’s interesting about tenrecs is that they’re able to maintain aspects of stem-like quality in their heart cells for much longer in adulthood, so their hearts retain some ability to regenerate even later in life. We are now sequencing tenrecs to see if we can replicate that ability to maintain stemness. That should help with cardiac repair and resistance to damage.
We’re also working with a group at the University of Florida, which happens to be where I went to vet school. There’s a consortium of six or seven labs that are working with a species called the spiny mouse. It’s an African mouse that’s highly adapted to be able to repair damage to many organs without scarring. People originally started looking into them because they could repair large skin defects without any scarring. Then scientists realized that those mice can actually repair parts of their brain, including spinal cord, parts of their kidneys, all that, without laying down any scar tissue, unlike us humans.
We have an intern from the University of Montana who studies highland-adapted deer mice – that is, mice adapted to high-altitude, low-oxygen environments. This is very similar to human populations in places like Nepal. There are many implications here for mitochondrial biology, how the body uses energy and ATP.
These are just examples. We have an internal team called The Dream Team, which stands for “Discovery, Research, and Emerging Animal Models”. It’s a cross-functional team of physiologists, genomicists, and wet lab folks who go out and talk to various investigators, trying to find people who are doing the best science with the most interesting models.
There are indeed so many marvels of nature around that it must be hard to choose which ones to work on.
It’s a bit like what happens when we talk to Big Pharma. They have a strategic roadmap of disease areas, indications, and data types they want to work with. Similarly, we’re looking for species that have well-documented protection phenotypes, where the related diseases map into our indication spaces of high unmet need and commercial tractability. We have a good partnering interest, and we know we can develop molecules in a disease area that has commercial potential. We have our own internal rubric, literally a flowchart of yes or no decisions about what partnerships make the most sense for us.
It’s impressive that an early-stage startup has forged partnerships with entities like NIH and Novo Nordisk. Have you also encountered a “you people are crazy” kind of attitude?
It’s really funny: people either love what we do and are so thrilled that we exist, because they’ve always thought a company like this should exist, but they never knew there was one, or they think we’re a bit crazy.
We have people working for us who found us through webinars and other activities. We didn’t even have any job openings, but they were, like, “I have to work for you, you’re the only company that I want to work for”. And we found homes for those people because they really believe in what we do. But yes, it’s usually either one or the other.
Fauna Bio is probably a very exciting place to be.
It’s a lot of fun. I just talked to a grad student last week. She’s looking at all the labs in the field, she had interviews with Emma Teeling and a few other folks, and she was, like, “oh, I didn’t know there was a company that was doing this”.
So, yes, we get a lot of that. People are so excited when they find out that there’s a way to actually translate the work that’s coming out of these labs. I think that’s where people really get frustrated as postdocs and early-stage PIs: how do you actually start to translate these amazing findings?
It takes a lot of setting up the infrastructure, figuring out what it takes to build drug programs, what are pharma partners looking for, what does a good tractable commercial indication look like? There’s a lot of learning to do in terms of how you translate the work, but the science is in a place where genomes are good enough, the sequencing is cheap enough. It’s not an impossible hill to climb anymore, with us being able to get good molecular characterization of some of these emerging model species and look directly at a link between them and human diseases. It’s this beautiful intersection of technology and some of the work that we do that allows us to get into these new species.
Would you say that you have created a blueprint for other companies to jump in?
I’d like to think that we’ve shown it’s possible, that there’s a path to directly translating these insights into therapies that can help humans live longer and healthier, which is what we all want to do, but there are many other kinds of interesting biology that are just not necessarily a good fit for us. For instance, there are people who work on different properties of venom, which, pharmacologically, has been a rich place for discovery. That’s where we got things like ACE inhibitors and GLP-1 agonists from. So, there are many different aspects of natural biology that are under-leveraged, and there’s a huge amount of opportunity here. People don’t think to look outside humans and a few model organisms. You have to know that the data is there and what to do with it.
Have you considered working with long-lived species, such as the naked mole rat or bats?
Yes, we get this question all the time. We do. We have been thinking about the right way to ask the right questions about these species. But I think we’re getting an advantage by working with species such as hibernating mammals. If you take them as a class, they tend to live about 30% to 50% longer than their non-hibernating cousins, and there are many links between aging and hibernation in terms of DNA repair and protection and cellular regeneration.
Another nice thing about hibernation, discovery-wise, is that it’s transient. Part of the year they’re protected, and part of the year they’re not. So, you can compare those timepoints directly, looking at RNAseq profiles and organs at different time points, and the signal just jumps out in terms of what genes are responsible.
If you have an animal like the naked mole rat, they’re protected all the time. Trying to narrow it down to what’s driving that protection is pretty hard, but we have an internal team that’s thinking about how we can use data from some of those long-lived species.
That’s a long-standing question for us: “What’s the right data type, what’s the right disease area?” There’s been a lot of work characterizing those models and not so much work in the translation space, I mean high-throughput, high-quality -omics data generated from some of those species as we would need to do our thing.
We’re actually talking to one pharma partner who has an interest in some of these long-lived species in certain disease areas, but part of this is that we would need to do some robust sequencing to get the kind of data we need to feed into our platform. The currently available datasets are just not good enough. So, that’s a great question, and we think about it internally all the time, but it requires the right datasets, the right time points, the right samples, and the right disease area.
So, it has something to do with you being a company, right? You can’t just study long-lived species, you have to work towards a certain indication.
Yes, obviously, you have to rapidly translate insights into action. We must think a lot about how we can rapidly find a chemical matter, or at least genetic targets that modify human diseases. Our first layer of validation in the wet lab is in human cells and disease-relevant tissues in either 2D or 3D. Things must pass that filter before we move into more traditional preclinical animal studies. So, yes, we have to be able to rapidly find things that we can test.
A somewhat related question: have you also looked at underlying causes of aging, such as oxidative stress or DNA damage that some of these species are so good at mitigating?
Yes, sure. That’s another reason why hibernating mammals make such a good starting point for this kind of investigation. A lot of oxidative stress and inflammation is generated when they re-warm over the course of an hour every couple of weeks. This rapid cycle of rewarming and cooling is particularly damaging. Some species have developed ways to elongate telomeres, others have enhanced DNA repair and protection.
In our own data, we see upregulation of a fetal genetic program. If you look at heart cells that are damaged at the peak of one of these rewarming cycles, you see genes coming up that are usually expressed in neonatal animals at the time points where they are still growing their heart tissue. They apparently reactivate their fetal regeneration programs as part of a reparative mechanism, and that very obviously links to aging mechanisms.
What’s next for Fauna.bio and your subfield in general, if there’s even a subfield?
I guess our subfield would be comparative physiology, writ large, and studying protective mechanisms in natural animal models. Like I said, it’s a really exciting time. I think this field of investigation is where the longevity field was about ten years ago: there are many interesting datasets, people are starting to look at this space, to translate insights out of it. There are also many academic labs working on it, and there are other companies that are looking at other species. Some of those companies are still in stealth, but they do exist. I think it will be interesting to see in the next few years how many companies will be taking a similar approach.
There are companies that are looking at other aspects of extreme biology, and with some of them, we share our VC backers. I’m talking about companies like Basecamp Research, which is looking at extreme proteins. Another company, called Enveda, is working with natural compounds and plants for drug discovery. There’s also one called Wild Biotech that’s looking at microbiomes in different species.
I think people are realizing that with tools currently available, many more areas have become open for discovery, and then the question is always how you translate it and commercialize it. But people are starting to figure out what those business models look like.
What’s in your pipeline? When will we see something tangible from Fauna Bio?
Hopefully, pretty soon. Like I said, we have a small molecule that’s in lead optimization right now. We’re hoping to have it in the clinic as early as 2025. It’s for some types of pulmonary fibrosis that are particularly related to the mechanism we’re working on.
We could also have a gene therapy coming along right behind that. Fauna Bio could become a clinical company not too long from now, which is amazing considering that we started from generating our own datasets not that many years ago. It’s a great time, and we’re finally seeing the fruits of all of our labors and getting some very good results.