All these preliminary informations seem to indicate a close interrelationship between melatonin and the "biochemical markers" involved in the addiction - reward - craving states. This seems true for alcohol, as suggested above and by evidence that alcohol withdrawal syndrome produced a reduction of nocturnal pineal melatonin content with a concomitant elevation in pineal serotonin. In addition, a group of "abstinent" (polydrug addicted) showed remarkably higher melatonin levels than acute relapsive cases 3031. It could be possible, in view of these preliminary informations, that melatonin is directly involved in the development of a state of drug addiction.

Our preliminary hypothesis could be that in conditions of drug intake, synthesis and release of melatonin in the pineal is altered (probably abnormally increased), either primarily or following alteration of the other neural systems defined above (DA, 5HT, and/or opiate). In normal conditions melatonin might be acting as a cerebral "pacemaker", sensitive to endogenous as well as exogenous stimuli in the attempt to maintain an equilibrate circadian interaction between the cerebral activities of endogenous aminergic and opiates systems (probably via a circadian mechanism). However, abnormal states (i.e. drug intake, abuse, addiction) could result in chronically altered pineal functions, resulting in rhythmically higher synthesis and release of melatonin (increased turnover). This could be either an original effect of drug intake or subsequent to modified activity of aminergic (dopaminergic, serotonergic) and/or opioidergic systems following drug abuse 3233. This change in the pineal functions could result in a rhythmically abnormal change in the functions of cerebral opiates (which should be reduced) and/or monoamine neurotransmitters, i.e. altered functions of DA in the nAcc (with a rhythmically higher increase and decrease of DA activity) and/or opposite modification of serotonergic functions, which could end in the development of a "circadian" state of reward - urge for drug (craving) resulting in addiction (see the resulting Figure 1).

This hypothesis could be studied in models of addicted animals i.e. spontaneous alcohol drinking rats versus spontaneous non preferring water drinking rats selected as described earlier 34 then submitted to treatment with melatonin receptor antagonists and/or agonists.

The above mentioned compounds, as well as melatonin, will be injected (intracerebral and/or systemically) in control and addicted animals. If the "melatonin hypothesis" is correct, one would expect modification of ethanol intake in spontaneous alcohol drinking rats. Similarly, modifications should be detectable in the behaviour of addicted animals submitted to the conditioned place preference test as well as to self administration procedures (either i.v. 3536 or intracranial self administration of drugs 3738. Modifications of neuro-biochemical parameters such as amine neurotransmitters and/or endogenous enkephalin and endorphin functions and receptor activities should be correlated to these behavioural alterations and monitored by mean of in vitro methods i.e. autoradiography 39 and in vivo techniques such as intracranial microdialysis 40 and in vivo electrochemistry 41. In particular, previous in vitro and in vivo experiments using the electrochemical method of voltammetry 42 indicated that Melatonin is an electroactive indoleamine hormone and that it can be measurable in vivo in the pineal gland as well as in the suprachiasmatic nucleus of rat brain 43. This, together with the possibility to concomitant voltammetric monitoring of DA and 5-HT release and metabolism 4244 will be a very useful tool to perform such investigation and already preliminary data have proposed a significant influence of melatonin or its antagonism on alcohol consumption in ethanol drinking rats 8. Further data supporting the proposed hypothesis will be of help in designing new therapeutic approaches to tackle drug dependence.