A low-calorie rare sugar that many shoppers already see on health-food shelves may do far more than sweeten products without adding calories. New research in diabetic mice suggests that D-psicose, better known commercially as D-allulose, can substantially reduce the fat buildup and inflammatory damage that accumulate in the livers of animals with type 2 diabetes. The study, published in the journal Food Science & Nutrition, goes a step further than most previous work by tracing the sugar’s effects through an unexpected route: the trillions of microbes in the gut, the chemical messages they send to the liver, and a molecular sensor called the aryl hydrocarbon receptor that sits at the crossroads of metabolism and inflammation.
The condition the researchers targeted is one of medicine’s quiet epidemics. Metabolic dysfunction-associated steatotic liver disease, or MASLD, affects the majority of people living with diabetes, and its incidence exceeds that of diabetic kidney disease. Because early-stage disease is usually symptom-free, it is frequently missed until fat overload, inflammation, and eventually fibrosis have taken hold. Left unchecked, it can progress to steatohepatitis and cirrhosis, sharply raising the risk of liver cancer, and it independently correlates with cardiovascular events and mortality. For a sugar that delivers roughly 70 percent of the sweetness of sucrose with minimal caloric contribution, the possibility of also easing this liver burden is a striking proposition.
To test the idea, a team at Jiangsu University worked with male C57BL/6J mice, feeding them a high-fat diet for four weeks and then injecting low doses of streptozotocin, a chemical that damages the insulin-producing beta cells of the pancreas. This combination reliably produces animals with high blood glucose, abnormal blood lipids, oxidative stress, and fat-laden livers, reproducing selected features of type 2 diabetes. One week after the final injection, mice with fasting blood glucose above 11.1 millimoles per liter were considered diabetic. The diabetic animals were then split into three groups of six: one left untreated, one given D-psicose at 2.5 percent of drinking water, and one given 5.0 percent. Treatment continued for twelve weeks alongside the high-fat diet.
The results were visible before any instrument was switched on. Livers from untreated diabetic mice were enlarged and pale, the classic gross appearance of severe lipid retention, while livers from the high-dose treatment group looked reddish and smooth, closely resembling those of healthy controls. Under the microscope the differences sharpened. Hematoxylin and eosin staining revealed that diabetic livers had lost their orderly cellular architecture, with hepatocytes ballooning and filling with vacuoles. Oil Red O staining, which highlights neutral triglycerides in bright red, confirmed massive fat deposits. D-psicose reversed these defects in a dose-dependent manner, preserving hepatic structure and clearing lipid droplets most effectively at the higher dose. Body weight, which had been falling steadily as the disease progressed, also stabilized in the treated animals.
Blood chemistry told the same story from a different angle. Untreated diabetic mice carried elevated total cholesterol, triglycerides, and LDL cholesterol, along with depressed HDL, the protective fraction. They also showed raised alanine and aspartate aminotransferases, the enzymes that leak into the bloodstream when liver cells are injured, and heightened circulating levels of the inflammatory cytokines TNF-alpha, IL-6, and IL-1beta. Twelve weeks of D-psicose shifted all of these indices back toward the healthy profile, with the high-dose regimen again producing the larger effect. The pattern suggests the sugar was not merely masking symptoms but correcting an underlying imbalance in how the diabetic liver handles fat and inflammation.
The molecular explanation begins with lipid metabolism genes. In diabetic livers, the transcription factors and enzymes that drive de novo fat synthesis, SREBP-1c and FASN, were significantly upregulated, while PPAR-alpha, the master regulator of fatty acid beta-oxidation, was suppressed. D-psicose flipped this program: SREBP-1c and FASN fell in a dose-dependent fashion and PPAR-alpha recovered toward normal levels. The authors describe this as a dual mechanism of increasing expenditure and reducing intake, simultaneously blocking new fat production and burning existing fat. The treatment also restored the transcription of antioxidant enzymes SOD1 and GPX1, which diabetes had suppressed, and pushed down Col1a1 and TGF-beta1, two markers of the pro-fibrotic signaling that threatens to scar the liver permanently.
The most novel part of the study concerns the gut-liver axis. Sequencing the V3-V4 region of the bacterial 16S rRNA gene from colonic contents showed that diabetes had skewed the microbial community, inflating the Firmicutes-to-Bacteroidota ratio and expanding Proteobacteria, a phylum often associated with inflammation. At the genus level, diabetic mice had depleted Akkermansia, Lactobacillus, and several other taxa, while Alistipes, Parabacteroides, and Anaerotruncus expanded. D-psicose treatment recalibrated these imbalances, pulling the community profile back toward that of healthy controls. In parallel, the sugar repaired the intestinal wall itself: villi that had become stunted and disorganized regained their height, the tight-junction proteins ZO-1 and Occludin, which seal the gaps between epithelial cells, recovered their continuous staining pattern, and serum lipopolysaccharide, a marker of bacterial endotoxin leaking through a leaky gut, dropped significantly.
Those microbial and barrier changes coincided with a measurable shift in tryptophan chemistry. Using targeted UPLC-MS/MS, the researchers found that diabetic mice had depleted serum tryptophan, elevated kynurenine, and a raised kynurenine-to-tryptophan ratio, a signature of overactive IDO/TDO catabolism, together with reduced levels of indole-3-propionic acid, a microbially derived metabolite with known anti-inflammatory properties. D-psicose reversed every one of these abnormalities. In the liver, the enzyme TDO2, the rate-limiting step of the tryptophan-kynurenine pathway, had been overexpressed in diabetes while the aryl hydrocarbon receptor, or AhR, a nuclear sensor for microbial indole metabolites, had been suppressed. Treatment lowered TDO2 and restored AhR at both the mRNA and protein levels. At the same time, the phosphorylation ratio of NF-kappaB p65, the activated form of the central inflammatory transcription factor, fell from its diabetic high back toward normal.
Untargeted serum metabolomics reinforced the picture. Principal component analysis separated the three groups in both ionization modes, and internally validated OPLS-DA models, checked with seven-fold cross-validation and 200-permutation testing, confirmed that D-psicose shifted the metabolic profile toward that of healthy animals. KEGG enrichment pointed to pathways including tryptophan metabolism, arachidonic acid metabolism, biotin metabolism, and steroid hormone biosynthesis as candidate biological contexts for the changes. The authors are careful, however, about what these data can and cannot prove. Because 16S sequencing is taxonomic rather than functional, the study cannot show that Akkermansia or Lactobacillus actually produced the altered metabolites, and because AhR activity was never directly perturbed, the gut microbiota-tryptophan-AhR chain remains a coherent hypothesis rather than a demonstrated causal mechanism.
The limitations are acknowledged candidly. The HFD/STZ model does not capture the full heterogeneity or natural history of human diabetic liver disease, the sample size was modest at six animals per group, and fasting glucose was used only to confirm diabetes induction, with no assessment of glucose tolerance, insulin sensitivity, or glycated hemoglobin. No fecal microbiota transplantation, germ-free experiments, or pharmacological AhR ligands were deployed to test causality. Still, the convergence of evidence is notable: a widely available rare sugar improved liver histology, blood lipids, inflammatory cytokines, antioxidant defenses, and fibrosis markers in diabetic mice, all while reshaping the gut ecosystem, restoring barrier integrity, and rebalancing a microbial-metabolic signaling axis increasingly implicated in metabolic disease. If future studies in complementary models and, ultimately, human trials confirm these coordinated effects, D-allulose could evolve from a niche sweetener into a genuine dietary tool against one of diabetes’ most overlooked complications.
Subject of Research: Effects of D-psicose (D-allulose) on hepatic steatosis, gut microbiota, and tryptophan-AhR signaling in a mouse model of type 2 diabetes
Article Title: D‐Psicose Attenuates Hepatic Steatosis and Inflammatory Injury in HFD/STZ‐Induced T2DM Mice and Is Associated With Alterations in Gut Microbiota, Tryptophan Metabolism, and Hepatic AhR‐Related Signaling
Article References: Wu, L., Dong, X., Sun, W., Xu, H., Xie, Y., Liu, X., Liu, X., Ali, A., & Shao, K. (2026). D‐Psicose Attenuates Hepatic Steatosis and Inflammatory Injury in HFD / STZ ‐Induced T2DM Mice and Is Associated With Alterations in Gut Microbiota, Tryptophan Metabolism, and Hepatic AhR ‐Related Signaling. Food Science & Nutrition, 14(10), Article e72455. https://doi.org/10.1002/fsn3.72455
Image Credits: AI Generated
DOI: 10.1002/fsn3.72455
Keywords: D-allulose, D-psicose, type 2 diabetes, fatty liver, MASLD, gut microbiota, tryptophan metabolism, aryl hydrocarbon receptor, NF-kappaB, intestinal barrier, Akkermansia, metabolomics
