Prior Experimental Experience Abolishes Appetite Suppression from Intracerebroventricular Insulin in Mice

Insulin administered directly into the brain's ventricles (intracerebroventricularly, or ICV) is known to inhibit eating across various species, a process thought to be crucial for central nervous system regulation of energy balance. This effect is primarily mediated by the hypothalamic melanocortin system. However, published research shows significant variability, with ICV insulin sometimes failing to reduce food intake for unknown reasons. This inconsistency complicates our understanding of insulin's role in central appetite control and its potential therapeutic applications for metabolic disorders. This study investigates whether the experimental procedure itself, specifically the animal's recent experience with injections, could be a key factor contributing to this observed variability in the hypophagic (appetite-suppressing) action of insulin.

Researchers conducted several crossover trials in mice. Animals received intracerebroventricular infusions into the third ventricle (i3vt) of either insulin or a vehicle control. In the primary experiment, mice received two injections spaced two days apart, with one group getting vehicle then insulin, and the other getting insulin then vehicle. Doses tested were 0.4 μU, 4.0 μU, and 40.0 μU per mouse. A separate experiment spaced the injections 7 days apart. To confirm the integrity of the downstream pathway, they also administered the melanocortin receptor agonist melanotan-II (MT-II) to mice that were non-responsive to insulin. The primary endpoint was food intake.

The effect of insulin was highly dependent on trial order and timing. In the 2-day crossover design, 0.4 μU insulin significantly reduced food intake only in mice that received vehicle on the first trial and insulin on the second. The effect was completely absent in mice that received insulin first. Higher doses of insulin (4.0 μU and 40.0 μU) had no effect on food intake in this 2-day paradigm, regardless of order.

When the injections were spaced 7 days apart, the crossover effect vanished, and insulin successfully reduced food intake. Importantly, mice that failed to respond to high-dose insulin still showed a robust reduction in food intake when given i3vt melanotan-II. This indicates that the hypothalamic melanocortin system itself was functional and that the interference with insulin's signal occurred upstream of the melanocortin receptors.