Essential fructosuria (EF) is a benign, asymptomatic, autosomal recessive condition caused by loss-of-function variants in the ketohexokinase gene and characterized by intermittent appearance of fructose in the urine. Despite a basic understanding of the genetic and molecular basis of EF, relatively little is known about the long-term clinical consequences of ketohexokinase gene variants. We examined the frequency of ketohexokinase variants in the UK Biobank sample and compared the cardiometabolic profiles of groups of individuals with and without these variants alone or in combination. Study cohorts consisted of groups of participants defined based on the presence of one or more of the five ketohexokinase gene variants tested for in the Affymetrix assays used by the UK Biobank. The rs2304681:G>A (p.Val49Ile) variant was present on more than one-third (36.8%) of chromosomes; other variant alleles were rare (<1%). No participants with the compound heterozygous genotype present in subjects exhibiting the EF phenotype in the literature (Gly40Arg/Ala43Thr) were identified. The rs2304681:G>A (p.Val49Ile), rs41288797 (p.Val188Met), and rs114353144 (p.Val264Ile) variants were more common in white versus non-white participants. Otherwise, few statistically or clinically significant differences were observed after adjustment for multiple comparisons. These findings reinforce the current understanding of EF as a rare, benign, autosomal recessive condition.

Fructose consumption in humans and animals has been linked to enhanced de novo lipogenesis, dyslipidemia, and insulin resistance. Hereditary deficiency of ketohexokinase (KHK), the first enzymatic step in fructose metabolism, leads to essential fructosuria in humans, characterized by elevated levels of blood and urinary fructose following fructose ingestion but is otherwise clinically benign. To address whether KHK deficiency is associated with altered glucose and lipid metabolism, a Khk knockout (KO) mouse line was generated and characterized. NMR spectroscopic analysis of plasma following ingestion of [6-13C] fructose revealed striking differences in biomarkers of fructose metabolism. Significantly elevated urine and plasma 13C-fructose levels were observed in Khk KO vs. wild-type (WT) control mice, as was reduced conversion of 13C-fructose into plasma 13C-glucose and 13C-lactate. In addition, the observation of significant levels of fructose-6-phosphate in skeletal muscle tissue of Khk KO, but not WT, mice suggests a potential mechanism, whereby fructose is metabolized via muscle hexokinase in the absence of KHK. Khk KO mice on a standard chow diet displayed no metabolic abnormalities with respect to ambient glucose, glucose tolerance, body weight, food intake, and circulating trigylcerides, β-hydroxybutyrate, and lactate. When placed on a high-fat and high-fructose (HF/HFruc) diet, Khk KO mice had markedly reduced liver weight, triglyceride levels, and insulin levels. Together, these results suggest that Khk KO mice may serve as a good model for essential fructosuria in humans and that inhibition of KHK offers the potential to protect from diet-induced hepatic steatosis and insulin resistance.

Fructose is a major component of Western diets and is implicated in the pathogenesis of obesity and type 2 diabetes. In response to an oral challenge, the majority of fructose is cleared during "first-pass" liver metabolism, primarily via phosphorylation by ketohexokinase (KHK). A rare benign genetic deficiency in KHK, called essential fructosuria (EF), leads to altered fructose metabolism. The only reported symptom of EF is the appearance of fructose in the urine following either oral or intravenous fructose administration. Here we develop and use a mathematical model to investigate the adaptations to altered fructose metabolism in people with EF. First, the model is calibrated to fit available data in normal healthy subjects. Then, to mathematically represent EF subjects, we systematically implement metabolic adaptations such that model simulations match available data for this phenotype. We hypothesize that these modifications represent the major metabolic adaptations present in these subjects. This modeling approach suggests that several other aspects of fructose metabolism, beyond hepatic KHK deficiency, are altered and contribute to the etiology of this benign condition. Specifically, we predict that fructose absorption into the portal vein is altered, peripheral metabolism is slowed, renal reabsorption of fructose is mostly ablated, and alternate pathways for hepatic metabolism of fructose are upregulated. Moreover, these findings have implications for drug discovery and development, suggesting that the therapeutic targeting of fructose metabolism could lead to unexpected metabolic adaptations, potentially due to a physiological response to high-fructose conditions.