When the liver fails, toxins – such as ammonia – that should be filtered from the blood build up and reach the brain. The result is hepatic encephalopathy (HE), a devastating neurological complication of liver disease that can cause anxiety, confusion, memory loss and, in severe cases, coma. HE is a common endpoint of liver cirrhosis, driving frequent hospitalisations and placing a heavy burden on patients and healthcare systems worldwide.

Current treatments offer only partial relief. The two mainstay therapies — lactulose and the antibiotic rifaximin — work primarily by reducing ammonia production in the gut, but neither corrects the full spectrum of metabolic disruptions that drive the disease. Patients remain vulnerable to recurrence, and rifaximin carries the added risk of disrupting the gut’s natural microbiome. A fundamentally different approach is needed — one that can tackle several disease drivers at the same time.

A research team from the National University of Singapore (NUS), led by Professor Matthew Chang from the NUS Synthetic Biology for Clinical and Technological Innovation (SynCTI) has recently achieved a major breakthrough on this front.

The researchers, who are also from the NUS Yong Loo Lin School of Medicine, engineered strains of a naturally occurring beneficial gut bacterium to function as programmable therapeutics capable of restoring metabolic balance across the gut, liver and brain. The study was published in the scientific journal Cell on 24 April 2026.

Reprogramming bacteria to fight disease on multiple fronts

The researchers redesigned Lactobacillus plantarum WCFS1 — a well-characterised commensal bacterium — into two complementary therapeutic strains. The first strain absorbs excess ammonia from the gut and converts it into branched-chain amino acids (BCAAs), essential nutrients that are depleted in HE patients. The second strain breaks down L-glutamine in the gut before it can be converted into yet more ammonia, cutting off a key source of the toxin.

Laboratory studies using a cocktail of both strains for HE showed that the combination reduced circulating ammonia by up to 10-fold and lowered brain ammonia to levels comparable to those in healthy conditions. Key metabolic imbalances — including depleted BCAAs and elevated L-glutamine — were restored, alongside marked improvements in anxiety-like symptoms and cognitive function.

“We found that engineered gut bacteria can simultaneously remove toxic ammonia, restore essential nutrients, and improve brain-related outcomes,” explained Prof Chang. “This directly addresses a major limitation of current treatments, which typically target only a single root cause rather than the full spectrum of metabolic drivers.”

Distinct advantages over a front-line antibiotic

Compared with rifaximin, the engineered bacterial cocktail achieved stronger improvements in anxiety and short-term memory. In addition, neuronal signalling was normalised and neuroinflammation was reduced, suggesting gut metabolic correction can drive benefits in the central nervous system.

The engineered strains also preserved the natural diversity of the gut microbiome, a significant advantage over rifaximin, which markedly reduced microbial richness. In long-term safety studies, the bacteria were well tolerated, showed no signs of systemic toxicity, and were cleared within 72 hours of the final dose.

A platform for next-generation “living medicines”

The team’s findings point to a versatile platform for what Prof Chang calls a new class of precision therapeutics. As the bacterial strains are modular — each engineered to perform a specific metabolic task — they could be adapted to target other disorders involving the gut-liver-brain axis, including urea-cycle defects and other hyperammonaemic conditions.

“Our study demonstrates the development of a multi-functional, programmable microbial therapy that can coordinate several therapeutic actions simultaneously inside the body,” said Prof Chang. “Unlike standard treatments such as rifaximin, which broadly suppress gut bacteria, our approach uses live biotherapeutics to precisely reprogramme metabolism while preserving the natural gut ecosystem.”

A patent application has been filed to support translation of the technology towards clinical use. The team’s next steps include evaluating the long-term performance of the engineered strains and expanding the platform to target other diseases linked to metabolic imbalance.

“Our long-term goal is to translate this work into the clinic and develop a new class of programmable, microbe-based therapies,” added Prof Chang. “These findings establish a strong foundation toward realising that vision.”