USDA Greenlights the First Engineered Microbe That Makes Antibodies in the Gut

Every drug we take is a losing race against our own clearance machinery. The body degrades the molecule, flushes it, and the level falls below the point where it does anything. So we dose again. Daily pills, weekly injections, monthly infusions. The schedule is not a feature of the disease. It is a feature of the fact that the drug cannot stay.

The alternative has a name and a long history of not working. Put a living cell in the body, have it take up residence, and let it make the drug on site, continuously, for as long as you need it. The microbiome is the obvious place to try it, because the gut already hosts trillions of bacteria that live there permanently and would be delighted to keep doing so.

On Sept 29, USDA's Center for Veterinary Biologics conditionally licensed a product that does exactly that. It's a strain of Lactobacillus reuteri engineered to secrete antibodies, and it was approved for chickens.

A $6 billion gut punch

Clostridium perfringens is not an invader. It lives in the gut of essentially every healthy chicken, causing no trouble at all. It is supposed to be there.

What changes is the host. Put a bird under stress, from a feed transition or a coccidiosis vaccination, and the colony shifts behavior and starts producing two proteins. Alpha toxin is an enzyme that dissolves the membranes of the cells lining the gut. NetB punches holes in them. One melts the wall, the other perforates it.

Both do the same job for the bacterium: they open up intestinal cells so the nutrients inside leak out. The colony eats what spills, and a colony that can feed itself keeps growing, which means more toxin, more damaged lining, more food. The bird is left with a damaged gut. It converts feed and absorbs nutrients poorly, gains less weight, and in the worst cases dies. That is necrotic enteritis, and across global poultry production it costs an estimated $6 billion a year.

The usual answer to a bacterial problem is an antibiotic, and here it is the wrong tool twice over. You cannot kill off C. perfringens, because under normal conditions it belongs in a healthy chicken's gut. And anything that tries will push the survivors toward resistance, which is exactly why antibiotics have been pulled out of broiler production in the first place.

So BiomEdit went after the proteins instead, the way you would go after any other disease-driving protein: with antibodies. The team immunized llamas with both toxoids, harvested the single-domain antibodies a llama's immune system produces naturally, and screened for the tightest and most stable binders to alpha toxin and NetB. Bind the toxins in the gut and they cannot open intestinal cells. The colony stops getting fed, and the bacterium stays where it was always meant to be.

“We're not shooting them. We're taking their weapons away.”

AARON SCHACHT, CEO, BIOMEDIT

The chassis is two strains of Limosilactobacillus reuteri isolated from chickens themselves, one from a young bird and one from an old one, both in the same clade. The expression cassette sits in the chromosome rather than on a plasmid, which is what makes the new strain genetically stable: a distinct organism carrying one extra gene, not a bacterium holding a plasmid it can drop or transfer.

Delivery takes advantage of the one day in a chicken's life when its gut is nearly empty. Chicks pass through a spray cabinet on the day they hatch. The product lands on their backs, they peck and preen each other, and it is the first thing they ever ingest. A newborn microbiome has very few incumbents to compete with or displace.

“I just park in an empty parking spot, establish a colony, and start making my stuff right there.”

AARON SCHACHT, CEO, BIOMEDIT

Colonization takes about a day. One application at hatch protects the gut for roughly three weeks.

Building the regulatory path four years before launch

In early 2023, months after BiomEdit's launch and BE-101's entry into early development, the company went to USDA and asked the agency to agree on a regulatory path. Schacht treated the agency's willingness to engage as the test of whether the product was ready for full development.

First they had to pick the right US regulatory agency. Animal health in the US has three possible doors, and they lead to different places.

FDA regulates drugs and feed additives. EPA regulates pesticides. USDA regulates biologics, which it defines as products that work on or like the immune system, and it is also where antitoxins are handled. A bacterium secreting toxin-neutralizing antibody fragments fits both halves of that definition, so USDA took jurisdiction and became the agency BiomEdit dealt with.

All three can issue a Finding of No Significant Impact (FONSI), the letter that says a new technology carries no environmental risk. Choosing the right one up front meant one conversation instead of three.

The agency had the criteria for licensing a biologic. It did not have a path for this one.

“There was a lot of back-and-forth during the process where we were creating the regulatory path together.”

AARON SCHACHT, CEO, BIOMEDIT

After reviewing BiomEdit's preclinical data, USDA determined the product qualified for the conditional licensure pathway, which gave the company a closer working relationship with the agency than a standard filing would. From there they negotiated the actual requirements: what the efficacy protocol would look like, what safety data was needed, what had to be true of manufacturing.

The safety package is where a live engineered organism takes more time and investment than a molecule. BiomEdit had to show what happens when the strain leaves the bird: shedding, spread to other animals, persistence in the environment. USDA rolled it into a risk analysis and in November 2025 issued the FONSI, concluding the product posed no environmental risk and no risk to people eating chicken raised on it.

Efficacy came from a challenge study run with Southern Poultry Research Group, affiliated with the University of Georgia, which maintains a gold-standard necrotic enteritis model. Birds were challenged with C. perfringens. Mortality fell from 13.1% in challenged controls to 6.6% in treated birds, a 48.7% reduction.

Healthier chickens means bigger chickens

Farmers do not only care whether their chickens are healthy. They care whether their chickens make money.

Necrotic enteritis is mostly a silent disease. The subclinical form is the common one, and a bird carrying it never looks sick.

Its gut lining is damaged, so it absorbs less of what it eats. It grows slower and needs more feed to reach market weight. Birds with the subclinical form run about 12% lighter and need about 11% more feed per pound of meat.

Feed is 65 to 75% of the cost of raising a broiler. So the disease shows up where it hurts most, and it shows up quietly. That is the $6 billion, and almost none of it is dead chickens.

So the number that sells this product is the feed conversion. In BiomEdit's published work, treated birds under subclinical challenge converted feed better and gained more weight. Same 42 days, same feed bill, more chicken.

That result came from a controlled challenge study, which is a few hundred birds per group in a research setting. Enough for BiomEdit to have confidence in its product candidate. Not enough for a producer deciding whether to put it in every house.

So the next trials are the commercial ones. USDA let BiomEdit start before the license came through, and the product is now in more than 35 million broilers across three major US producers. The three were chosen for variety: differing production systems, bird sizes, and bird genetics. The point is not to repeat the result, it is to find out where it holds and where it does not.

BiomEdit's vets were at the hatchery the day the first birds were sprayed. The data from all three producers comes together around January 2027, and that is the readout that determines whether this becomes a product or stays a milestone.

Looking beyond the hen house

Chickens are the beachhead.

CPerf LLV will be the only first-generation product BiomEdit ever sells. While it worked its way through USDA, the platform kept moving. The delivery strains now carry multiple integration sites. Operons can run several genes, or several gene products at once. Expression and secretion have been increased and optimized. The growth characteristics, or fitness, of the resulting engineered strains have been enhanced. Schacht calls the result Programmable Veterinary Biologics, and everything behind the lead product is built on it.

The payload does not have to be an antibody. Nanobodies came first and carried a unique challenge: delivering a correctly folded, actively binding antibody fragment. Enzymes and peptides were next, and BiomEdit could not assume how well those biomolecules would fare in their system. The company has since engineered a variety of strains across species and applications. One program puts enzymes in the gut that degrade inflammatory factors. Another engineers rumen strains to release anti-methanogenic compounds, replacing seaweed-derived bromoform or a synthetic like Bovaer with biology, and redirecting the hydrogen the methanogens would have used into meat or milk.

The step Schacht is most interested in is dogs.

Injectable biologics for inflammatory bowel disease are dose-limited. Push injectable antibodies targeting key inflammatory cytokines hard enough to control the gut and you suppress the immune system everywhere else. But the target biology is in the gut. A probiotic secreting those same nanobodies into the lumen keeps the drug where the disease is and out of systemic circulation, which means the dose can go up rather than down. The hope is that efficacy and response rates improve relative to the standards of care established in human medicine by injectable antibodies such as anti-TNF or anti-IL-23. Antibody drugs like these are not precedented in animal health.

“If I can do it in pets, I can do it in humans.”

AARON SCHACHT, CEO, BIOMEDIT

Many chronic conditions in pets are close analogues of human ones. Data in dogs would de-risk the human version of the same hypothesis, and that is the door to partnerships well outside animal health.

Schacht is careful about how large a claim to make. He is not careful about the question underneath it.

“I believe it's a new modality. We have an N of 1. If you think back to the first antibody that was ever made, this might be the same moment for this modality. I'm not saying it is, but it might be.”

AARON SCHACHT, CEO, BIOMEDIT

The first antibody drug led to Amgen and Genentech, and antibody drugs are now the standard of care in many diseases. Schacht's question is who plays that role for engineered microbes. Asked if that's BiomEdit, he didn't hesitate: “Yeah, why not?”