Safety Assessment of 5-D-Keto-Fructose: Reconciling Discrepant Toxicological Data Through Standardized Testing and Mechanistic Analysis

Abstract

Background: 5-D-keto-fructose (5-KDF) is a rare ketose sugar under development as a low-calorie sweetener. A 2022 publication by Hövels et al. reported cytotoxicity findings and concluded that "use of 5-KF in the food sector should be avoided." However, this study employed non-standardized methods, uncharacterized test material, and lacked critical experimental controls, raising questions about the validity of these conclusions.

Objective: To provide definitive safety data on 5-KDF using specification-controlled test material and internationally harmonized test methods (OECD Test Guidelines), and to identify the specific methodological deficiencies that produced artifactual toxicity signals in the Hövels et al. study.

Methods: A comprehensive toxicology program was conducted including: (1) bacterial reverse mutation assay (OECD TG 471; five tester strains ±S9, up to 5,000 µg/plate), (2) in vitro mammalian chromosomal aberration test (OECD TG 473; human peripheral blood lymphocytes ±S9, up to 1,000 µg/mL), and (3) mechanistic cellular safety assessment in human intestinal (Caco-2) and hepatic (HepG2) cell models evaluating cytotoxicity, oxidative stress, mitochondrial function, and apoptosis/necrosis pathways. Test material was crystalline 5-KDF (≥98.6% purity, ≤0.10% ash) produced via ion-exchange chromatography to eliminate catalytic impurities that could drive in situ Maillard/glycation chemistry during biological testing. Critical design elements included osmolarity-matched controls (D-mannitol, D-glucose, D-fructose, D-allulose at equimolar concentrations) to distinguish chemical toxicity from hyperosmolar stress—a control entirely absent from the Hövels et al. study.

Results: 5-KDF was non-mutagenic in the bacterial reverse mutation assay (revertant colony counts 0.9–1.1-fold vehicle control across all strains and conditions). 5-KDF was non-clastogenic in the chromosomal aberration test (0.5–0.8% aberrant cells vs. 0.5% in vehicle controls; positive controls 18–22%). In mechanistic cellular studies, 5-KDF showed no cytotoxicity at concentrations ≤100 mM (viability 94–102% of control). At 200 mM, modest viability reductions (88.2–92.1% in Caco-2 and HepG2 cells) were statistically indistinguishable from equimolar osmotic controls (mannitol: 80.4–88.0%, glucose: 85.5–90.5%), conclusively demonstrating that observed effects were due to non-specific hyperosmolar stress rather than intrinsic toxicity. No oxidative stress (GSH/GSSG ratio 97.8–101.5% of control), mitochondrial dysfunction (membrane potential 95.4–97.1% of control; bioenergetic parameters 96.1–100.8% of control), or apoptosis/necrosis pathway activation (caspase-3/7 activity 1.00–1.08-fold; viable cells 94.8–97.2%) was detected at any concentration.

Conclusions: Under standardized regulatory test conditions with specification-controlled test material, 5-KDF demonstrates no genotoxic or cytotoxic hazard. The discrepancy with Hövels et al. (2022) is attributable to three critical methodological deficiencies in that study: (1) Test article quality: use of filtrate-derived material lacking ion-exchange purification, retaining catalytic ions and nitrogen-containing residues capable of generating Maillard reaction products in situ during incubation (evidenced by the authors' report of "intense brown coloring"), (2) Absence of osmotic controls: binary comparison of high-molar 5-KDF against water vehicle confounds osmotic stress with chemical toxicity, and (3) Non-validated endpoints: bacterial growth inhibition in chemically reactive media does not constitute evidence of human genetic or cellular hazard. The weight of evidence from OECD-conformant genotoxicity testing and mechanistic cellular studies with appropriate controls conclusively supports the safety of specification-controlled 5-KDF for use as a food ingredient.