IngredientsAllulose โ chemically D-allulose, also known as D-psicose or D-ribo-2-hexulose โ is a naturally occurring monosaccharide (simple sugar) classified as a "rare sugar." Allulose (D-ribo-2-hexulose or d-psicose) is a novel low-calorie functional rare sugar and the epimerization product of d-fructose at the C-3 position. It is a ketohexose that has the same empirical formula as common monosaccharides such as glucose and fructose, and it is epimeric with fructose at the 3-position. Its molecular formula is CโHโโOโ, identical to that of glucose and fructose, but its three-dimensional configuration differs, which accounts for its distinct metabolic behavior.
The name "psicose" has an unusual origin: the name "psicose" is derived from the antibiotic psicofuranine, from which it was isolated. The compound has a long history of chemical characterization prior to its modern commercial re-discovery: it was known as early as 1915 as D-pseudofructose, and German chemists H. Ohle and F. Just determined its structure in 1935 and renamed it D-psicose; in 1942, F. W. Zerban at the New York Sugar Trade Laboratory and Louis Sattler at Brooklyn College isolated it from commercial cane molasses.
Allulose occurs in nature only in trace amounts. Psicose has been found in wheat, Itea plants, and processed cane and beet molasses. It is also reported in small concentrations in certain fruits. D-Psicose is naturally found in fruits, wheat, and syrups, but only in minute concentrations. Because natural concentrations are so low, essentially all commercial allulose is produced biotechnologically rather than extracted directly from plants.
Modern production relies on enzymatic conversion of more abundant sugars. Prof. Ken Izumori of the Kagawa University Rare Sugar Research Centre developed a strategy for the preparation of all hexoses using cheap and widely available substrates, according to which d-allulose could be converted from d-fructose via ketose 3-epimerase. Commercial-scale manufacturing typically starts from corn-derived fructose: psicose is manufactured commercially using D-psicose-3-epimerase from A. tumefaciens; a fructose solution is treated with immobilized, nonviable C. glutamicum cells harboring the enzyme, and it takes 4โ8 hours at 50ยฐC to convert fructose to psicose. Newer enzymatic and microbial routes continue to be developed to improve yield and lower cost, including epimerases derived from various gut and environmental bacteria used in whole-cell biocatalysis. Food-grade microbial species such as Bacillus subtilis have been employed for the production of D-allulose using whole-cell catalysis to circumvent the laborious process of enzyme purification, yielding a 30% conversion rate.
Allulose has 70% of the sweetness of sucrose and ultra-low energy content. D-allulose has a sweetness equivalent to 70% of that of sucrose and calories equivalent to about 0.3% of that of sucrose, with similar taste and volume characteristics to sucrose, and can undergo Maillard reaction with amino acids or proteins in food. It has been described as having some cooling sensation and no bitterness. These properties โ bulk, browning capacity, and sucrose-like mouthfeel โ make it functionally distinct from high-intensity sweeteners (such as stevia or sucralose) and popular in low-calorie baked goods, beverages, and confectionery.
Allulose is sold and used as a crystalline granulated powder (a tabletop sugar substitute), as a syrup, and as an ingredient incorporated into packaged foods, beverages, sauces, baked goods, and confections. It is also studied experimentally as an aqueous solution or beverage additive in clinical trials, and has been trialed in oral nutritional supplement (ONS) formulations for people with diabetes.
Allulose does not have a history of traditional herbal, culinary, or medicinal use in the way that botanicals or long-recognized food ingredients do. It exists in nature only in trace quantities in items such as wheat, certain plant genera (Itea), and processed cane and beet molasses, but it was never isolated, concentrated, or deliberately used by any traditional culture as a sweetener, tonic, or remedy. Its identification is a matter of 20th-century analytical chemistry โ first characterized structurally in the 1930s and isolated from molasses in 1942 โ rather than of ethnobotanical or folk-medicine tradition. It was first identified in the 1940s, although the enzymes needed to produce it on an industrial scale were not discovered until the 1990s. Its practical availability as a commercial food ingredient and "supplement-adjacent" sweetener is therefore a recent development, driven mainly by Japanese rare-sugar research (the Kagawa University Rare Sugar Research Center, led by Ken Izumori) beginning in the 1990sโ2000s, and by subsequent U.S. and Korean commercialization from the 2010s onward. There is consequently no traditional-medicine record โ Ayurvedic, Traditional Chinese Medicine, European herbal, or otherwise โ associated with allulose, since it was neither recognized nor deliberately consumed as a distinct substance until modern times.
Allulose itself is a single, well-defined chemical entity (a monosaccharide), so there are no "active compounds" within it in the way there might be within a botanical extract. Its physiological effects instead derive from its distinctive absorption, transport, and (non-)metabolism pathways, which are increasingly well characterized.
In mammals, allulose utilizes in essence the same pathways as dietary fructose for uptake into the body, distribution, and excretion; intestinal absorption of allulose is mediated by the transporter GLUT5 in apical membranes of epithelial cells. The export across the basolateral membrane occurs via GLUT2. Animal studies confirm this transport route: D-allulose, D-sorbose, and D-tagatose are likely transported via GLUT5 but not SGLT1 in rat small intestine.
Critically, despite being absorbed like fructose, allulose is not metabolized like fructose once inside the body. In contrast to fructose, the human genome does not encode enzymes able to metabolize allulose, leading to an almost complete renal excretion of the absorbed dose and near-zero energetic yield. Human pharmacokinetic characterization supports this: allulose is rapidly absorbed in the small intestine but undergoes minimal metabolism, with approximately 70% being excreted unchanged in urine within 24 hours. The unabsorbed fraction has a secondary fate in the lower gut: due to the limited capacity of GLUT5 to transport allulose, ingested amounts not fully absorbed in the upper small intestine reach the ileum and the colon and may become metabolized by the commensal bacteria residing there.
A key proposed mechanism for allulose's metabolic effects is stimulation of the incretin hormone GLP-1 (glucagon-like peptide-1) from intestinal L-cells, distinct from the pathway used by digestible sugars. GLP-1 is normally secreted when stimulated by nutrients including metabolizable sugars such as glucose and fructose, but D-allulose, a C-3 isomer of D-fructose, is a rare sugar with anti-diabetic or anti-obese effects in animal models that also stimulates GLP-1. Notably, this occurs through GIP-independent, GLP-1-selective signaling: oral gavage of allulose potently and sustainably induced GLP-1 secretion in rats, stimulating GLP-1 but not affecting GIP secretion, in contrast to glucose.
Because allulose is not metabolized, researchers have sought an explanation for GLP-1 release that does not depend on intracellular sugar metabolism. Recent work points to a physical, rather than biochemical, trigger: as allulose is not metabolized, it is thought to promote GLP-1 secretion through a pathway distinct from that of glucose or fructose, and this is not mediated by the sweet taste receptor in a human study. Experimental results demonstrate that allulose promotes GLP-1 secretion, at least in part, via intestinal distension as a novel GLP-1 secretory mechanism. Downstream, GLP-1 released by allulose appears to act centrally: vagal afferent-specific Glp1r knockdown affects food intake and glucose metabolism, and D-allulose activates GLP-1 secretion, which in turn directly interacts with vagal afferent neurons to induce calcium signaling. One proposed explanation for allulose's particularly strong GLP-1 stimulation relative to glucose is prolonged luminal residence: because D-allulose is slowly absorbed in the intestine compared to glucose, it would stay in the intestinal lumen longer to stimulate L-cells.
Beyond incretin release, additional mechanisms have been proposed for allulose's acute glucose-lowering action. Matsuo and Izumori found that allulose inhibits ฮฑ-glucosidase activities, which leads to suppression of glycemic responses after carbohydrate ingestion. In the small intestine, allulose and other monosaccharides such as glucose and fructose share the same transporters, such as GLUT2 and GLUT5, implying a degree of competitive inhibition of glucose absorption. A hepatic mechanism has also been suggested: activation of glucokinase in the liver promotes glycogen synthesis, and meta-analytic authors summarize the overall model as one in which allulose modulates glucose through several pathways, likely inhibiting intestinal glucose absorption via glucose transporter interactions to reduce post-meal glucose spikes, and may also stimulate GLP-1 release, which enhances insulin sensitivity and decreases glucagon levels, promoting stable glucose without significant insulin secretion.
This is the most extensively studied application of allulose in humans, and the evidence is comparatively strong for acute effects, though modest in absolute magnitude.
In a landmark equivalence trial in people with type 2 diabetes: a double-blind, multiple-crossover, randomized, controlled, acute feeding equivalence trial in 24 participants with type 2 diabetes assigned six treatments โ fructose or allulose at 0 g (control), 5 g, or 10 g โ added to a 75-g glucose solution. Allulose significantly reduced plasma glucose incremental area under the curve (iAUC) by 8% at the 10 g dose compared with 0 g (717.4 ยฑ 38.3 vs. 777.5 ยฑ 39.9 mmol ร min/L, P = 0.015), with a linear dose-response gradient. This trial defined what the authors termed "catalytic doses": recent literature suggests that catalytic doses (โค10 g/meal or 36 g/day) of D-fructose and D-allulose may reduce postprandial blood glucose responses to carbohydrate loads in people with and without type 2 diabetes by inducing glycogen synthesis.
A more recent 2024 meta-analysis pooled randomized trials specifically in type 2 diabetes: six studies involving 126 participants were included, and allulose significantly reduced glucose AUC (SMD: โ0.6662, 95% CI [โ1.1360, โ0.1964], p = 0.0054) with moderate heterogeneity (Iยฒ = 58.3%). Secondary outcomes were more mixed: insulin AUC showed a non-significant reduction (SMD: โ0.3648, 95% CI [โ0.7783, 0.0488], p = 0.0839); fasting plasma glucose showed a non-significant reduction, while time-above-range (TAR) significantly decreased (MD: โ8.8204, 95% CI [โ14.4101, โ3.2307], p = 0.0020); no significant change was observed in time-in-range. This pattern โ a consistent, statistically significant reduction in postprandial glucose excursion, with weaker or non-significant effects on fasting glucose and insulin โ is characteristic of the literature as a whole and indicates a real but modest acute glycemic benefit, primarily relevant to the post-meal glucose spike rather than to overall glycemic control markers such as HbA1c.
Additional supporting trials include a 12-week crossover study of allulose 7 g twice daily versus aspartame in type 2 diabetes examining glucose homeostasis, metabolic parameters, incretin levels, and inflammatory markers... allulose 7 g twice daily or aspartame 0.03 g twice daily for 12 weeks, and a small crossover trial in people without diabetes: a prospective, randomized, double-blind, placebo-controlled crossover study in 30 subjects without diabetes gave a standard oral 50 g sucrose load with placebo or escalating doses of D-allulose (2.5, 5.0, 7.5, 10.0 g), finding a significant, dose-related reduction in plasma glucose iAUC.
Evidence here comes from a small number of human randomized trials, generally described by the study authors themselves as preliminary.
The principal human trial is a 12-week Korean RCT: a preliminary study with 121 Korean subjects (aged 20โ40 years, BMI โฅ 23 kg/mยฒ) used a randomized controlled trial design with placebo control (sucralose, 0.012 g ร 2 times/day), low d-allulose (4 g ร 2 times/day), and high d-allulose (7 g ร 2 times/day) groups, assessing body composition, nutrient intake, CT scan, and plasma lipid profiles. Body fat percentage and body fat mass were significantly decreased following d-allulose supplementation. More specifically, d-allulose supplementation for 12 weeks reduced body fat mass, body fat percentage, and subcutaneous fat (SAT) area, while there was no significant reduction in visceral fat (VAT) area. The authors' own conclusion was appropriately cautious: despite some limitations, the study demonstrated that d-allulose is able to reduce body fat mass in overweight or obese subjects, and the preliminary study also indicates that the effects of d-allulose supplementation are likely to be dose-dependent.
A separate secondary summary of this same trial characterizes the effect size as modest: in a 12-week trial of 121 generally healthy individuals randomized to consume either 7 g or 14 g of allulose or a sucralose placebo daily, a modest reduction in fat mass was reported but no significant differences were observed in cardiometabolic parameters. Overall, human weight/body-fat evidence is limited to essentially one primary RCT (with related secondary publications) and should be regarded as preliminary rather than conclusive; larger, longer, independently replicated trials are lacking.
A 48-week Japanese RCT investigated allulose in people with elevated LDL cholesterol and mild glucose intolerance: subjects were randomly divided into 3 groups: high-dose D-allulose (15 g/day), low-dose D-allulose (5 g/day), and placebo (0 g/day); each subject consumed a daily test beverage for 48 weeks, with clinical examinations every eight weeks. Results were reported as follows: no significant increases in total cholesterol and LDL-cholesterol between test groups were observed, and 48 weeks of D-allulose consumption did not change risk factors for atherosclerotic cardiovascular disease; additionally, significant improvements in hepatic enzyme activities, fatty liver score, and glucose metabolism after long-term D-allulose consumption were observed. A related summary specifies the preload protocol and subgroup finding: in this 48-week trial, 15 g of allulose consumed as a preload 30 minutes before breakfast significantly reduced fasting serum alanine aminotransferase and gamma-glutamyl transferase, and, in the subgroup of subjects with borderline diabetes, reduced 2-hour glucose AUC after a 75 g glucose challenge. This remains a single mid-sized trial (approximately 90 participants) rather than a body of replicated evidence, so findings on liver enzymes and fatty liver score should be considered promising but preliminary.
Allulose's relationship to dental caries has been evaluated by regulatory bodies and in laboratory (in vitro) oral-microbiology studies, though direct long-term human caries-incidence trials are not available.
The FDA's own safety review concluded: allulose, like other non-cariogenic carbohydrate sweeteners listed in 21 CFR 101.80(c)(2)(ii), does not result in a decrease in dental plaque pH below 5.7, which is associated with decalcification of the dental enamel. Based on this, given the low cariogenic potential of allulose, the FDA concluded that allulose does not promote dental caries. Laboratory (in vitro biofilm) research is broadly consistent: allulose supports lower bacterial virulence activity and minimal biofilm formation compared to common dietary sugars while preserving microbial diversity, highlighting its potential as a non-cariogenic sugar alternative.
However, at least one in vitro study introduces nuance, finding a transient pH drop and raising a caveat about root caries specifically: allulose, sucralose, and xylitol do not produce a sustained acid drop below the critical pH of 5.5 and are not considered cariogenic compared with sugars such as sucrose, glucose, and fructose, although allulose may contribute to root caries. The same study cautions against overstating benefit: these sugar substitutes do not inhibit an S mutans pH drop in the presence of glucose and should not be considered anticariogenic in the presence of other sugars. Overall, the evidence for non-cariogenicity is fairly strong (supported by regulatory review and multiple in vitro models), but evidence for an active "anticariogenic" (caries-reducing) effect, as opposed to simple non-cariogenicity, is weaker and largely mechanistic/in vitro.
A 12-week randomized, double-blind, placebo-controlled human trial specifically examined allulose's effect on gut microbial communities and pathogen levels: participants consumed 15 g/day of D-allulose or sucralose (placebo) for 12 weeks. The key findings were reassuring from a safety standpoint: this randomized, double-blind, placebo-controlled study confirms the safety of D-allulose consumption over 12 weeks; consumption of D-allulose did not significantly alter gut microbial diversity or pathogen levels, indicating its safety. This addresses a theoretical concern that rare sugars reaching the colon unabsorbed might promote pathogenic bacterial overgrowth; in this trial no such effect was detected.
Allulose has been studied as a powder/crystalline sweetener dissolved in beverages, as a syrup, and incorporated into functional beverages, chocolate, and oral nutritional supplements. Doses used across the clinical literature vary considerably by study objective:
Regulatory/tolerance-derived dosage benchmarks: in 2020, the U.S. FDA accepted the conclusion by Samyang that the maximum tolerable consumption for a 60 kg adult was 33 to 36 grams per day. A separate manufacturer-derived toxicological assessment (GRAS Notice 693) similarly concluded D-allulose intake up to 0.5โ0.6 g/kg body weight/day is safe.
Allulose has undergone FDA review via the GRAS (Generally Recognized as Safe) notification pathway. The FDA has evaluated allulose's safety through its GRAS process; applicants submit an expert scientific review to the FDA as a "GRAS notification," and the FDA responds with either a "No Objection letter" or notification that the ingredient doesn't provide a basis under GRAS regulations. Multiple such notifications have been filed and not objected to. The FDA has not objected to three GRAS notifications regarding the use of allulose as a sugar substitute in certain conventional foods and beverages (GRN 400, GRN 498, and GRN 693).
Because of its unusual metabolic fate, the FDA created a distinct labeling framework for allulose. According to FDA's Center for Food Safety and Applied Nutrition, the latest data suggested allulose is different from other sugars in that it is not metabolized by the human body in the same way as table sugar โ it has fewer calories, produces only negligible increases in blood glucose or insulin levels, and does not promote dental decay. Practically, this means: the FDA exercises enforcement discretion allowing Nutrition and Supplement Facts labels to (a) exclude allulose from Total Sugars and Added Sugars declarations, and (b) use as low as 0.4 kcal/g for allulose calorie count, but allulose must still be included as a Total Carbohydrate. This creates a practical caveat relevant to certain patients: using the grams of carbohydrate on the product's Nutrition Facts panel could present a risk of hypoglycemia in people with diabetes who take insulin and calculate their bolus insulin doses based on the carbohydrate they consume, since the listed carbohydrate grams overstate the glycemic impact of allulose-containing products.
Preclinical toxicology data submitted in the GRAS review found a wide margin of safety. Toxicity data reveal an LD50 of 15.8โ16.3 g/kg bw, indicating that even at the highest exposure, D-allulose is not a safety risk. A chronic feeding study likewise found no adverse effects at high dietary levels: a month-long chronic toxicity study showed that D-allulose at a dose of 3% of the diet (or 1,280 mg/kg bw/day), the highest level tested, did not show adverse effects. The overall regulatory conclusion was that D-allulose, like other monosaccharides, belongs to the group with the lowest toxicity rating and is classified as an ordinary carbohydrate substance, such that its use in foods and beverages is not expected to pose a safety concern.
The most well-documented and clinically relevant adverse effect of allulose is dose-dependent gastrointestinal intolerance, characterized specifically in a controlled human dose-escalation trial. A GI tolerance test for D-allulose was performed to establish its daily acceptable intake level; when the dose was gradually increased in steps of 0.1 g/kg body weight to identify the maximum single dose for occasional ingestion, no cases of severe diarrhea or GI symptoms were noted until a dose of 0.4 g/kg BW was reached, with severe symptoms of diarrhea noted at 0.5 g/kg BW. Compared directly against an equal dose of ordinary sugar, allulose produced significantly more symptoms: a correlation analysis revealed significantly higher frequencies of symptoms of diarrhea (p = 0.004), abdominal distention (p = 0.039), and abdominal pain (p = 0.031) after D-allulose intake compared to sugar at the same dose. At still higher intakes, more severe symptoms emerged: increasing the total daily D-allulose intake gradually to 1.0 g/kg BW for regular ingestion resulted in incidences of severe nausea, abdominal pain, headache, anorexia, and diarrheal symptoms; based on these results, the study authors suggested a maximum single dose and maximum total daily intake of D-allulose of 0.4 g/kg BW and 0.9 g/kg BW, respectively.
A separate, earlier dose-tolerance analysis (cited in a Center for Science in the Public Interest regulatory submission) produced slightly different but concordant thresholds: the authors of the clinical study concluded that the maximum tolerable levels in humans were 0.5 g/kg bw/day in men and 0.6 g/kg bw/day in women, corresponding to 33.3 g/day in men and 31.0 g/day in women, based on mean body weights, with symptoms ranging from diarrhea, borborygmi, lower abdominal pain, nausea, and distention reported at doses above 0.5 g/kg bw/day. At the single-dose level, the same data source detailed a marked difference from ordinary sugar: in the first experiment, when 0.5 g/kg bw allulose was administered, 13 of 29 subjects (44.83%) reported diarrhea and four (13.79%) had severe symptoms, whereas at the same dose of sugar only four subjects reported diarrhea, none severe.
More recent, shorter-duration exposure data suggest that continuous intake within moderate ranges is generally well tolerated, with symptoms concentrated early in the exposure period: the results of a 30-day randomized, double-blind study confirm that D-allulose has good gastrointestinal tolerance: GI symptoms were mild, transient (primarily during Days 1โ3 of intervention), and showed no dose-dependent differences, aligning with the safety profile of low-digestible carbohydrates.
Because allulose is disposed of largely unchanged via the kidney rather than being metabolized, its handling in renal tissue has drawn specific attention. The apical membrane of renal proximal tubular cells contains GLUT5, which means there will always be a certain level of allulose that is retained in the renal tissue following absorption and filtration โ a pharmacokinetic detail noted in review literature on allulose's fate in the body, though it has not been linked to demonstrated renal harm in the available human trial data (one interventional rat-model study specifically evaluated allulose's effect on diabetic nephropathy progression rather than allulose-induced kidney injury).
Clinical dosing and labeling guidance flags a specific practical interaction relevant to insulin-treated diabetes patients: because Nutrition Facts labels must still count allulose within "Total Carbohydrate" even though it does not raise blood glucose the way ordinary carbohydrate does, using the grams of carbohydrate on the product's Nutrition Facts panel could present a risk of hypoglycemia in people with diabetes who take insulin and calculate their bolus insulin doses based on the carbohydrate they consume. This is a labeling-derived risk of miscalculation rather than a direct pharmacological interaction with insulin itself.
Health conditions that Allulose may help support.
Animal studies show allulose supplementation boosts catalase activity and improves antioxidant capacity, reducing free radical damage. A rat model comparing carbohydrate diets found the allulose group had improved antioxidant markers versus glucose and control groups.
Animal research shows d-allulose activates appetite-suppressing (anorexigenic) neurons in the brain's arcuate nucleus and reduces food intake, partly via GLP-1 signaling. Human evidence on subjective appetite/hunger is preliminary and mixed.
Multiple randomized controlled trials and meta-analyses show allulose reduces postprandial blood glucose in healthy adults, prediabetics, and people with type 2 diabetes. A 2024 meta-analysis of six RCTs (126 participants) found allulose significantly reduced glucose AUC and time-above-range in T2DM patients.
Human and animal studies show allulose stimulates release of gut satiety hormones (GLP-1, PYY, CCK), a mechanism linked to appetite and glucose regulation. A rat model found allulose supplementation increased GLP-1 levels alongside reduced food intake and body weight gain.
Animal studies consistently show allulose reduces body weight gain, food intake, and fat accumulation on high-fat/high-sugar diets. Small human trials report modest reductions in body fat and BMI with allulose versus placebo, though sample sizes are small and durations short.
A meta-analysis of clinical trials notes allulose shows potential for improving insulin sensitivity, though evidence is preliminary. Animal studies using the gold-standard hyperinsulinemic-euglycemic clamp method show d-allulose improves insulin resistance in high-fat-diet-fed rats.
Human studies show d-allulose enhances postprandial fat oxidation and decreases carbohydrate oxidation, suggesting it may help regulate energy metabolism. Animal research shows allulose promotes mitochondrial uncoupling in adipose tissue and increases energy expenditure.
A randomized controlled trial in hypercholesterolemia patients on statin therapy found long-term D-allulose ingestion significantly decreased triglyceride levels (by roughly a quarter) in both healthy subjects and patients. Animal studies also show allulose lowers plasma and hepatic triglyceride levels.
Body systems that Allulose may help support.
Go back