The research, led by Irene Palenca of the Santa Lucia Foundation in Rome together with colleagues at Sapienza University of Rome and the University of Naples Federico II, builds on a concept the team calls the gut–liver axis. When chronic inflammation erodes the intestinal epithelium, the tight junctions that seal neighboring cells together fall apart. Microbial products such as lipopolysaccharide, the endotoxin shed by Gram-negative bacteria, then slip through the broken barrier into the portal vein, the vessel that delivers everything absorbed in the gut straight to the liver. There, the arriving inflammatory cargo activates Kupffer cells, the liver’s resident macrophages, and hepatic stellate cells, the fibrosis-driving support cells, igniting inflammation and scarring that can progress to lasting liver damage.

The therapeutic molecule at the heart of the study is palmitoylethanolamide, or PEA, an endogenous fatty acid amide belonging to a family of signaling lipids known as ALIAmides, short for autacoid local injury antagonist amides. PEA acts primarily through peroxisome proliferator-activated receptor-alpha, a nuclear receptor that governs inflammatory resolution and metabolic homeostasis in both gut and liver. In experimental colitis models, PEA has been shown to dampen tissue damage, restore tight junction proteins, and suppress TLR4-dependent inflammatory signaling. In liver injury models, it reduces fibrotic remodeling by inhibiting Kupffer cell and stellate cell activation. Yet PEA has a stubborn pharmaceutical flaw: it is highly lipophilic and virtually insoluble in water, so oral doses are poorly absorbed, and getting sustained amounts to the inflamed intestinal mucosa has proven difficult even with micronized formulations.

The Italian team’s solution was to stop shipping PEA to the gut and instead have the gut make it. The researchers engineered Lactobacillus paracasei subsp. paracasei F19, a well-characterized probiotic strain, to carry the human gene for N-acylphosphatidylethanolamine-specific phospholipase D, or NAPE-PLD, the enzyme that converts membrane precursors into PEA. The construct was delivered on a plasmid called pTRKH3-slp-NAPE-PLD and introduced into the bacteria by electroporation. When given together with an ultra-low dose of sodium palmitate, roughly 0.0003 micrograms per kilogram, the engineered strain, designated pNAPE-LP, uses the palmitate as a substrate to synthesize PEA in situ. A control strain carrying the empty plasmid, pLP, received the same palmitate but lacked the enzyme, allowing the team to isolate the effect of bacterial PEA production from the lipid itself.

The experiments used male C57BL/6J mice subjected to chronic dextran sulfate sodium colitis, a widely used model in which three seven-day cycles of DSS in drinking water, each followed by a recovery week, reproduce the relapsing damage of human inflammatory bowel disease. Twelve mice per group received daily oral gavage of the engineered bacteria, the control strain, or vehicle, and a fourth group received the engineered probiotic plus weekly injections of MK886, a pharmacological antagonist of PPAR-alpha, to test whether the receptor was required for the benefit. On day 70, the animals underwent colonoscopy before tissue and blood collection, with all analyses performed by blinded observers.

The results were unambiguous. Chronic DSS drove severe mucosal injury, with endoscopic injury scores rising from 1.58 in healthy controls to 6.58 in colitic mice, histological damage scores climbing from 1.77 to 4.89, body weight losses of nearly 16 percent, and colons shortened from 8.12 to 5.50 centimeters. Treatment with pNAPE-LP and palmitate significantly reversed all of these parameters, cutting endoscopic scores to 4.83, histological scores to 3.50, and restoring weight and colon length. The control strain did nothing, and MK886 abolished every protective effect, pinpointing PPAR-alpha as the indispensable molecular switch.

Quantification by HPLC–mass spectrometry confirmed the mechanism. Colonic PEA levels jumped from 1.95 to 5.02 nanograms per gram of tissue in treated animals, while hepatic PEA rose from 1.18 to 4.11 nanograms per gram and serum PEA climbed from 1.67 to 6.54 nanograms per milliliter. This systemic distribution indicates that PEA manufactured locally in the colon reaches the liver, most likely through the portal circulation. Functional testing with fluorescein-labeled dextran showed that the treatment cut intestinal permeability from 5.91 to 3.02 micrograms per milliliter of serum, restored circulating levels of the tight junction proteins ZO-1 and occludin, and reduced plasma endotoxin from 5.86 to 4.23 endotoxin units per milliliter. In short, the probiotic sealed the leaky barrier and throttled the flow of inflammatory cargo to the liver.

The liver itself told a parallel story. Colitic mice showed elevated hepatic injury scores, a collagen proportional area of 10.5 percent on Sirius red staining compared with 1.85 percent in controls, transaminase levels near 150 units per liter for both AST and ALT, and heightened neutrophil infiltration measured by myeloperoxidase activity. pNAPE-LP treatment reduced collagen deposition to 8.01 percent, lowered transaminases to around 75 units per liter, and cut hepatic myeloperoxidase activity from 39.59 to 26.07 milliunits per 100 milligrams of tissue. Hepatic TGF-beta1, the master profibrotic cytokine, fell from 411.7 to 127.8 nanograms per gram. Western blot analysis revealed the underlying signaling: the engineered probiotic suppressed the hepatic TLR4/NF-kappaB/NLRP3 inflammasome axis, reduced caspase-1 activation, lowered the cytokines IL-1beta, TNF-alpha, and IL-6, and decreased Bax expression alongside TUNEL-detected DNA fragmentation, indicating less apoptotic cell death. PPAR-alpha expression, which colitis had suppressed, was restored to near-control levels.

One finding puzzled the researchers. MK886 co-administration not only blocked the protective effects but also erased the rise in colonic, circulating, and hepatic PEA, even though the antagonist should act downstream of PEA biosynthesis. The authors note that bacterial viability and NAPE-PLD expression were not assessed in the presence of MK886, so an indirect effect of the antagonist on the engineered strain cannot be formally excluded, and further work will be needed to clarify the observation.

The study’s authors are careful about its limits. The DSS model does not fully recapitulate human inflammatory bowel disease, the analysis did not include stellate cell activation markers such as alpha-SMA or hydroxyproline content, and the treatment protocol assessed functional efficacy rather than stable gastrointestinal colonization. Translating a genetically engineered live biotherapeutic into the clinic will demand rigorous characterization of strain identity, plasmid retention, genetic stability, the risk of horizontal gene transfer, environmental shedding, and scalable manufacturing, alongside regulatory pathways far more demanding than those for conventional probiotics. Still, the proof of concept is compelling: a probiotic reprogrammed as a living drug factory can simultaneously mend the intestinal barrier and quiet the inflammatory and fibrogenic signaling that chronic colitis unleashes on the liver, pointing toward a new class of therapies that treat the gut and its downstream organs as the connected system they truly are.

Subject of Research: Engineered probiotic biosynthesis of palmitoylethanolamide to treat colitis-associated liver injury via the gut–liver axis

Article Title: Engineered probiotic-mediated palmitoylethanolamide biosynthesis protects against colitis-associated liver injury

Article References: Palenca, I., Franzin, S. B., Sarnelli, G., & Esposito, G. (2026). Engineered probiotic-mediated palmitoylethanolamide biosynthesis protects against colitis-associated liver injury. Frontiers in Pharmacology, 17, Article 1925341. https://doi.org/10.3389/fphar.2026.1925341

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1925341

Keywords: engineered probiotics, palmitoylethanolamide, gut–liver axis, colitis, liver injury, PPAR-alpha, NAPE-PLD, inflammatory bowel disease, Lactobacillus paracasei, NLRP3 inflammasome, intestinal barrier, live biotherapeutics