[Paradigma Studios/Adobe Stock]
Could the genetic secrets of “extreme mammals” help unlock treatments for myriad human diseases? The Emeryville, California–based startup Fauna Bio is betting on that premise. One of the animals the firm is studying include the 13-lined ground squirrel, which can double its body weight in fat accumulation in summer before entering hibernation in an obese state. After entering hibernation for roughly half the year, the squirrels’ enter a state of extreme metabolic suppression where their metabolic rate dips to 1–3% of normal.
Fauna Bio has inked significant deals with pharmaceutical giants Eli Lilly and Novo Nordisk, tapping its Convergence AI platform to identify potential obesity drug targets across 452 mammal species, including 65 other hibernators in addition to the 13-lined ground squirrel. The insights from the research could pave the way for new treatments for a range of human diseases, including metabolic, fibrotic, and neurodegenerative disorders.
Animal inspiration for human treatments
This unique approach to drug discovery may sound unconventional, but the potential implications for treating human disease has precedent. For instance, the discovery of a current crop of blockbuster obesity drugs owes its origin to research on the gila monster. The venom of the Gila monster contains a hormone known as exendin-4, which is similar to the human hormone glucagon-like peptide-1 (GLP-1) that regulates blood sugar. Because exendin-4 degrades more slowly than human GLP-1, it was a sound model for drug development. This research helped pave the way to the development of a new class of megablockbuster obesity and diabetes drugs, including GLP-1 receptor agonists like semaglutide and tirzepatide (which is technically a dual GLP-1 and glucose-dependent insulinotropic polypeptide (GIP) receptor agonist).
Ashley Zehnder, D.V.M., Ph.D.
While animals have long been used in preclinical research for drug development, the direct exploration of animal genetics and physiology to uncover potential treatments for human health remains rare. Part of the reason is a sort of implicit assumption that “humans are unique in the world,” said Ashley Zehnder, CEO of Fauna Bio. “And they’re really not.” There’s a tendency to overlook “how much we can learn by observing how other species either get diseases or don’t,” Zehnder said. “A lot of that physiology is very shared, particularly among mammals.”
Exploring extreme physiology for clues
In terms of its research on mammals like the 13-lined ground squirrel, Fauna Bio is focused “on highly conserved genes that function slightly differently in other species,” said Zehnder, who is a veterinarian with a Ph.D. in oncology. “What we’re exploring is how these species optimize the same genetic program for disease reversal instead of progression.”
Genes and pathways involved in human cancer biology are highly conserved evolutionarily not only in mammals but also in birds, reptiles, and even more primitive organisms. This is also true for many potent oncogenes – the genes that often cause cancer in animals. “The Ras protein, for instance, is the same all the way back to yeast,” Zehnder said.
Converging on connections between animal and human disease
Fauna knowledge graph showing animal disease resistance data. Image courtesy of Fauna Bio.
To explore the connections between animals’ robust survival mechanisms and human health, Fauna Bio has created an AI platform known as Convergence that analyzes the genomics of species that endure extreme conditions, such as hibernation. One of the core modules under the Convergence umbrella is Orca, which enables direct comparison of gene expression signatures between disease-resistant animals and human disease states. Specifically, Orca identifies areas of overlap and similarities between gene expression changes in animals that show natural resistance to tissue damage.
“Hibernators are a great example of this because of their really extreme warming and cooling periods that they go through throughout hibernation,” Zehnder said. For instance, hibernators can go through rapid rewarming periods that are damaging to many major organs. “Then they have to reverse that damage every time that they drop back into torpor,” she added.
Fauna Bio’s AI platform analyzes these reversal time points, mapping them directly to “really different human diseases,” Zehnder said.
By comparing gene expression patterns between hibernators and humans with heart failure, for instance, Fauna Bio identified genetic networks that show contrasting regulatory behaviors. For instance, protective genes are upregulated in hibernators but downregulated in humans with heart failure. This contrast opens up promising avenues to explore potential protective mechanisms against organ damage that could be targeted for new heart failure therapies.
Hunting for protective genes across species
In its research, Fauna Bio specifically maps genes that are more active in hibernators and less active in humans with heart failure, pinpointing the specific genetic changes linked to protection from damage. “That allows us to then start to rank order these genetic networks,” Zehnder said. In this regard, the company is not only exploring the 13-lined ground squirrels and humans, but more than 50 other species curated from scientific literature that has been harmonized for use on the Convergence platform.
The comparison of mammalian genomes has also advanced as a result of the Zoonomia Project, a scientific initiative that compares mammalian genomes to understand the molecular basis of disease and the evolutionary history of mammals. Science profiled the output from the initiative in a special issue in April 2023.
Fauna Bio partnered with more than 150 researchers across the Zoonomia Consortium to align genomes and proteins from more than 400 mammals to identify natural genetic mechanisms in diverse species that resist inflammation, anemia, cancer, and other diseases. The research found multiple mammalian species with disease-resistant adaptations, including inflammation blockade in the Siberian musk deer, potential anemia resistance in the Java mouse-deer, anti-cancer attributes in the North American beaver, and a number of pathology reversals in hibernators including the 13-lined ground squirrel.
The research could inform new therapeutic strategies targeting these mechanisms in humans. “That allows us to work on genes and pathways that are highly conserved and more likely to be functional and driving diseases,” Zehnder said. “We then validate all of our genetic and compound predictions on the bench in our lab in Emeryville, which is largely dedicated to modeling human cells.”
Modeling disease in human cells
Fauna Bio uses a variety of cellular models, including iPSC-derived and primary human cells, along with both 2D and 3D disease models. “We test our predictions in those models,” Zehnder said. It funnels findings from those models into modules within its Convergence platform that collectively synthesize them with external data from sources such as the UK Biobank.
Fauna Bio amalgamates the data into its graph neural network (GNN) that enables it to use advanced graph neural networks to hunt for links between genetic targets and human diseases. As its sophistication grows, the GNN grows more accurate at predicting disease linkages. “Once we find a gene expression signature of interest, we can map directly to small molecules,” Zehnder said.
Although the company only currently has about 20 employees at the moment (it is now hiring), its ambitions are considerable. “We are going from original data generation all the way to small molecule chemistry predictions and everything in between,” Zehnder said. “That’s the breadth of the platform.”