The increasing intake of fructose has been associated with an increase in obesity among US children and adolescents, but its "dose dependent" effects on insulin sensitivity and lipid metabolism has not been studied in detail. Methods: 36 healthy male adult human subjects will be included in this study. They will be randomly allocated into a low, moderate or high fructose, a moderate or high glucose and a sucrose diet for 3 weeks. In randomized order and with 4 week wash out intervals each subject will receive all six different diets. Three day food records will be used to measure total fructose and glucose intake. During the low fructose diet subjects will be instructed to avoid nutrients containing fructose aiming at consumption of less than 1g fructose/d. During the moderate fructose diet subjects will receive 3x13.3g of fructose or glucose, respectively. High fructose diet subjects will receive either 3x26.7g/d of fructose, 3x 26.7g/d of glucose or 3x 26.7g/d of sucrose in the form of three daily soft drinks taken together with the three main meals.

75 g oral glucose tolerance testing will be performed and composite insulin sensitivity index will be calculated The presence of phosphorylated fructose metabolites in plasma will be measured by targeted LC-MS/MS. In addition, metabolite biomarkers in plasma will be screened by untargeted metabolite profiling using both LC-MS and GC-MS. In a subgroup of 10 subjects an euglycemic hyperinsulinemic clamp will be performed using [6,6- 2H2]glucose.

Thus, the lipogenic potential of fructose in humans will be compared with isocaloric amounts of glucose. Particularly, the question whether lipogenic effects are continuously dose dependent or whether there is a lipogenic shunting and if yes, at what level of ingested fructose will be addressed.

Inclusion criteria:

• Healthy male volunteers aged 20-50y with a BMI between 19 and 25 kg/m2

Exclusion criteria:

• Acute infection, malignant disease, renal (abnormal creatinine), hepatic (more than two-fold increased transaminases), pulmonary, neurological (epilepsy) or major psychiatric diseases, manifested atherosclerosis
• Fasting plasma glucose > 5.6 mmol/l
• Known alcohol or drug abuse
• HIV-antibody positive
• Subjects likely to fail to comply with the study protocol
• Smoking (>1 cigarette/month)
• Subjects who do not give informed consent

Additional exclusion criteria after baseline measurements:

• Subjects with high baseline consumption of soft drinks (>60g of carbohydrates daily)