Laboratory rodents are broadly used to investigate the blood glucose lowering effects of putative anti-diabetic treatments as well as the involved mechanisms of action. However, many such studies neglect the necessity to sort out, if and to what extent glucose lowering actually results from loss of appetite and body weight, as can arise from trivial “side effects” like nausea or other unspecific sickness. Using obese mice on a diabetogenic high fat diet, the present thesis analyzed three interventions, which had previously been associated with both, anti-hyperglycemic action and reduced food intake in experimental rodents. To assess effects on fuel metabolism independently of indirect actions via weight changes, control mice were weight-matched with the treated animals by restricted feeding.
(i) Mice were treated with metformin, which is the most frequently used oral drug for T2DM. While a single dose of metformin acutely lowered blood glucose, regular Treatment with metformin along protocols, as they are usually applied to mimic clinical use of metformin, even impaired glucose homeostasis under conditions of weight neutrality. This outcome suggests major differences of the metabolic actions of metformin in rodents versus humans. Species-dependent differences thus could have considerably contributed to the unsatisfactory progress, which has been made in the past in unraveling the antihyperglycemic mechanisms responsible for the anti-diabetic action of metformin in human patients.
(ii) Mice were treated with emodin, a plant compound from Traditional Chinese Medicine. Since glucose lowering action of emodin in rodents has been reported, emodin-containing plant remedies are advertised for diabetes treatment, particularly in the internet. The present study unmasks that under body weight neutrality regular treatment with emodin causes derangement rather than improvement of blood glucose and glucose tolerance. Glucose lowering effects were neither observed in response to a single dose of emodin. The results turn the acclaimed anti-diabetic candidate emodin into a possibly diabetogenic agent and unmask premature claims of therapeutic utility as misleading and potentially dangerous.
(iii) Mice were exposed to hypoxia, which has previously been associated with impaired food intake and, in some studies, with improved glucose homeostasis. When maintained for several months in a hypoxic environment comparable to 5.500 m above SL, obese mice showed distinct amelioration of hyperglycemia, glucose intolerance and hepatic steatosis. Although the benefits were to a large part explained by weight loss, impressive improvement persisted also under conditions excluding weight-dependent effects. The results suggested that hypoxia did neither act via improved insulin secretion nor via Insulin sensitization, hinting at a novel, not yet discovered mechanism of action. Deeper understanding of the involved mechanism might in the future help to develop new treatment regimens for T2DM.
In summary, the present thesis highlights the extraordinary importance of weight-matched and/or pair-fed control groups for the interpretation of animal studies, which deal with alleged anti-diabetic treatment regimens.