Increased pineal melatonin content coupled to restricted water availability in a Pavlovian conditioning paradigm in rats.

The objective of this study was to assess whether rat pineal melatonin content could be modified in a classical conditioning paradigm. In rats kept under light (200 lux) from 06.00 to 18.00 h daily, the time of lights off was selected as the unconditioned stimulus (US). Restricted water availability (from 10 min before to 10 min after light-dark, LD, transition) was the conditioned stimulus (CS). The conditioned and unconditioned responses were measured as the changes in pineal melatonin levels 4 h after LD transition. In animals under regular lighting conditions, lights out at 18.00 h (the US) caused a 4.4-7.8-fold increase of pineal melatonin concentration 4 h after later, when compared to animals maintained under light for the 4 h-period. After a training period of 7 days of restricted water availability (the CS), significantly augmented pineal melatonin levels were found in rats that were exposed to water but were maintained under light for the 4 h period after expected LD transition. The control animals for this experiment, i.e., rats which had undergone the training period, were kept for 4 h under light after expected LD transition, and did not receive water at LD transition, exhibited very low pineal melatonin levels. The conditioned increase of pineal melatonin content attained lower values than those in rats exposed to normal lighting conditions. It also fulfilled the contingency criterion, that is, it caused at trial a significant elevation of pineal melatonin content only when water availability was applied from 10 min previously to LD transition during training, and not 20 min after LD transition. After a training period of 7 days, restricted water availability applied 4 h before lights off (at 14.00 h), caused an enhanced production of melatonin 4 h later, regardless of the animals being exposed either to a dark or to a light environment. The results indicate that pineal melatonin production can be manipulated in a classical conditioning paradigm, when an appropriate CS stimulus is used.