Systemic insulin has rapid access to the brain by a receptor-mediated transport system located in endothelial cells of brain microvessels. In several species, including man, parallel increases in the concentration of insulin in the cerebrospinal fluid (CSF) after acute elevation of plasma insulin levels are well documented. In animal studies central nervous insulin has been shown to play an important role in body weight regulation and memory function. Direct intracerebroventricular administration of insulin reduced food intake and caused weight loss. Moreover, insulin in the brain has been shown to improve memory function. However, in humans the investigation of central nervous effects of insulin has been hampered by the fact, that infusion of insulin and glucose is not practicable for a longer time to obtain changes in body weight. Recently, we provided evidence, that insulin enters the CSF compartment after intranasal administration without absorption into the blood stream. With this approach, we investigated the effect of long-term intranasal administration on body weight regulation in healthy subjects. After a baseline of 2 weeks 40 lean subjects (16 females, 24 males) were treated for 8 weeks with intranasal insulin (4×40 IU/day) or placebo in a double blind, between subject comparison. Body weight and body composition were measured weekly by bioimpedance analysis. Blood glucose and plasma insulin levels were identical during intranasal administration of insulin and placebo, indicating that no relevant amount of insulin entered the blood stream. There was no decrease in body weight and body fat mass during intranasal insulin administration in woman as compared to placebo. However, in contrast to women, insulin treated men continuously lost body weight compared to the placebo condition (baseline adjusted body weight at the end of the study, placebo: 79.84±0.57kg; insulin: 78.12±0.57kg, p<0.05). Most important, this weight loss was mainly a result of a reduction in body fat mass (placebo: 14.12±0.61kg, insulin: 12.25 0.61g, p<0.05). With the same approach, effects of intranasal administration of insulin (4×40 IU/day) or placebo for 8 weeks on memory function were investigated in 38 healthy subjects (14 females, 24 males). Every week subjects were presented with a word list containing 30 words. One week later they had to tell the words they remembered. Compared to the baseline condition intranasal administration of insulin significantly improved memory function (words recalled, placebo: 2.92±1.00, insulin: 6.20±1.03, p<0.05) at the end of the treatment period. There was no difference in the insulin effects on memory function in males and females. With the same approach, effects of intranasal insulin aspart even exceeded the beneficial effects of human insulin on body fat mass (placebo: +0.45±0.63kg, human insulin: -1.38±0.63kg, insulin aspart: -2.67±0.63kg, p<0.08) and memory function (words recalled, placebo: 1.83±0.75, human insulin: 5.17±1.56, insulin aspart: 8.35±1.09, p<0.05) in 36 healthy men. These results indicate an important role for insulin in body weight regulation and memory function in humans. Concerning recent findings of a reduced transport of insulin across the blood brain barrier in obese subjects and in Alzheimer patients, these findings could be of considerable clinical relevance.