Each year Nature Biotechnology profiles a company that has raised significant early-stage funding in the previous year. Empress Therapeutics is reverse-engineering microbial manufacturing.
Over the past two decades, there has been intensive research into the human microbiome and its impact on human health and disease, but relatively few successes have been achieved in terms of translating research findings into therapeutic approaches: As of mid-2024, only two microbiome-based therapeutics have been approved by the US Food and Drug Administration, both of which target recurrent bacterial infections of the gut.
Empress’s founding team (left to right): John Mendline, Executive Partner, Flagship Pioneering; Sabrina Yang, Chief Innovation Officer and Co-Founder, Empress Therapeutics; Principal, Flagship Pioneering; Doug Cole, General Partner, Flagship Pioneering; Jason Park, CEO and Co-Founder, Empress Therapeutics; Operating Partner, Flagship Pioneering. Courtesy of Empress Therapeutics
But there’s still good reason to believe there’s buried treasure in the microbial communities that live in the human body. “I’m trained as an immunologist, and microbial communities are fundamentally foreign – so why do we have all these cells living in our gut, on our skin, in our mouths?” asks Murray McKinnon, CSO at Empress Therapeutics. The answer, he says, is the powerful symbiotic relationship we have with these microbial communities. These microbial communities have been strongly shaped by a process of “co-evolution” with their human hosts. “Microbial communities produce chemical reactions that have beneficial effects on the immune system, which in turn have beneficial effects on metabolism.”
Empress uses this principle of coevolution to guide the discovery of medicines based on biologically active small molecules naturally produced by the human microbiome. Co-founded in 2020 by Jason Park, Sabrina Yang, and colleagues from Flagship Pioneering, the company emerged from stealth in June 2023 with an initial investment of $50 million from Flagship.
Park, who now serves as the company’s CEO, cited his experience in the world of small molecule drug development as one of his motivations for starting Empress. Park noted that this class of drugs remains a cornerstone of the pharmaceutical industry due to their long track record of versatility, flexible formulations and ability to easily enter cells. “The biggest problem we saw was uncertainty,” Park said, noting that the small molecule drug development process remains highly vulnerable to failure due to unexpected toxicity, poor bioavailability and other factors.
Turning to the microbiome could resolve some of these uncertainties. “We have a real opportunity to rethink how we make small molecule drugs,” Park says. The evolutionary forces that have shaped the symbiotic relationship between humans and resident microbes are expected to favor the generation of compounds that are non-toxic, stable, and biologically active in the human body. And judging by previous microbiome studies, these compounds have the innate potential to directly modulate physiological processes linked to a variety of pathologies, from autoimmunity to cancer to neurodegenerative diseases.
Empress’s Chemilogics discovery platform begins with a detailed metagenomic survey of the microbial populations of a large number of patients with a particular condition and healthy controls. These sequence data are fed into a machine learning-powered algorithm that seeks to distinguish between these two groups based on specific “biosynthetic gene clusters” (BGCs), an assembly line of tightly linked genes that work sequentially together to manufacture a specific biomolecular product. “It’s not just the chemicals that matter, it’s also the causal relationship between those chemicals and the clinical phenotype,” says McKinnon. This analysis is facilitated by the inclusion of patient transcriptomics data, which reveals disease-related changes in gene expression that correlate with the presence or absence of a candidate BGC of interest. Ideally, this also reveals the target genes or pathways that are modulated by the product of that BGC.
These newly identified BGCs will be evaluated by expressing them in genetically reprogrammed model bacterial species such as E. coli, giving the Empress team the opportunity to thoroughly characterize the chemical output of these clusters. Potential drug candidates identified at this stage will then be subjected to a range of biological assays to evaluate their therapeutic efficacy and characterize their mechanism of action.
Princeton University biochemist Mohammad Seyedsayandost, whose lab focuses on finding useful compounds from microbes, is pleased to see startups looking at such naturally derived products as a starting point for drug discovery. “I think this is in some ways an emerging field, because we’re just now realizing how huge the number of biosynthetic gene clusters are and how little we know about them,” he says. Seyedsayandost cautions that while identifying BGCs from metagenomic data should be a relatively easy task, characterizing such clusters experimentally can be much more difficult. “You can’t always just inject genes into E. coli and expect these compounds to be made… I think in many cases you need to go back to the original host,” he says. This can introduce significant complexity, as many microbial species remain difficult to culture and manipulate.
MacKinnon acknowledges that “biology is the rate-limiting factor” and that the analysis and development process will be quite challenging once the first wave of algorithmic analysis is complete. But the company benefits from a Scientific Advisory Board (SAB) that includes experts in the fields of microbiome research, metagenomics, synthetic biology, and metabolomics, who have been involved in developing and refining the Chemilogics platform since the company’s inception. “All of our SAB members — all six of us — have been with us for over five years,” Park says.
Empress is still in the preclinical stage, but the team is investigating a variety of indications, including immune and inflammatory diseases, oncology and metabolic diseases. Park says the first few years of work with the platform have been fruitful, generating about 15 promising drug candidates with in vivo activity against several different categories of targets. Park hopes to move the first programs into clinical trials in the next one to two years.
Importantly, this first set of leads also appears to meet the hope that microbiome-derived compounds might avoid many of the potential side effects of fully synthetic drugs. “We did some toxicity studies on the lead molecules and were surprised at how well these molecules were tolerated,” said MacKinnon, adding that they required much less chemical optimization and refinement than is typical in small-molecule drug discovery campaigns.
Empress’s approach also represents a significant departure from most drug discovery research in the microbiome field, which has focused primarily on treating patients with specific consortia of therapeutically beneficial microbes. MacKinnon notes that that therapeutic strategy faces numerous challenges, including manufacturing difficulties (many clinically useful microbes are difficult to culture) and regulatory issues that come with developing optimized live microbe-based therapeutics. In contrast, Empress is focused on exploring the inner workings of these microbial communities to find a better starting point for initiating traditional small molecule drug programs.
But that doesn’t mean the company’s focus ends at the borders of the microbiome. Park is optimistic that as the Empress team gains more understanding about microbial biosynthetic diversity, they will be able to break new ground in terms of engineering new chemical synthesis pathways. “It doesn’t have to be limited to symbiotic microbes,” Park says. “The same concepts can be applied anywhere there is biosynthetic DNA.”
