A recent study by McClung's group (2005),1 expanding on an earlier report,2 provides mechanistic insight to the timekeeper gene, Clock, which may regulate dopaminergic transmission and cocaine reward. This work provides further evidence that cocaine-induced effects have circadian influences.

McClung and colleagues studied Clock/Clock mutant mice,3 with a single-nucleotide transversion that inactivates the CLOCK protein, and found that they have an increased level of locomotor activity with a circadian activity pattern. Consistent with the observed hyperactivity, Clock/Clock mutant mice displayed increased levels of tyrosine hydroxylase (TH; a rate-limiting enzyme of dopamine synthesis) in ventral tegmental area (VTA) cells, as well as increased bursting and firing activity. TH-positive cells in the VTA were also positive for CLOCK protein, indicating potential local regulation of TH by CLOCK. Microarray studies in these mutants revealed that several target genes of CLOCK were downregulated in VTA (notably Per1 and Per2). Intriguingly, other genes involved in excitatory and inhibitory neurotransmission (ie glutamatergic or GABAergic) were also regulated in these mutant mice. Several groups have shown that expression of timekeeper genes in rodents or flies increases after exposure to cocaine, amphetamines, alcohol and morphine.

McClung et al1 found that Clock/Clock mutants exhibited robust sensitization to the locomotor-stimulating effects of repeated cocaine, indicating that functional CLOCK protein is not necessary for this form of cocaine-induced plasticity. These mutant mice also displayed modestly increased cocaine-induced place preference, a model for the rewarding effects of this psychostimulant.

Recent studies clarified the core molecular mechanisms of the circadian clock in the suprachiasmatic nucleus of the hypothalamus, which consists of autoregulatory transcription–translation loops with a periodicity of about 24 h. The positive loop is constituted by transcription factors CLOCK and BMAL1 that activate transcription of Per1, Per2 and Cry genes. The PER and CRY proteins assist in the negative feedback by attenuation of the CLOCK/BMAL1 transcription, thus inhibiting their own activation.4 Timekeeper genes, as transcription factors, may have an impact on the expression of target genes with E-box sequences in their promoter regions, such as dopamine and glutamate transporters, D1 dopamine receptor.

Based on previous studies and this report, it seems that changes in function or expression of different members of the timekeeper gene family may lead to alterations in one or another aspect of drug-induced behaviors. The earliest studies, which were performed by Hirsch and colleagues in Drosophila,5 showed, in sharp contrast to the report herein, that deletion of four different timekeeper genes (Clock, Per, Cycle and Doubletime, but not Time-less) resulted in the complete elimination of sensitization to repeated cocaine administration. A study by Abarca's group6 showed differential roles of Per1 and Per2 genes in cocaine-induced behaviors in mice. Our microarray study showed that Per1 mRNA expression is increased in the caudate–putamen of rats by acute ‘binge’ cocaine, whereas Per2 mRNA is upregulated only after repeated binge cocaine.7 Per1 knockout mice did not exhibit behavioral sensitization to repeated cocaine administration, whereas Per2 knockout mice displayed more potent cocaine-induced place preference. Also, Per2 knockout mice showed a higher rate of alcohol consumption.8 In addition, mice with inactivated Per1 mRNA did not display morphine-induced place preference.9 Interestingly, chronic morphine-induced increases in the expression of Per2 gene in the rat frontal cortex persisted after naloxone-precipitated withdrawal.10 These data implicate timekeeper genes in common mechanisms of drug abuse-related behaviors (Figure 1).

Figure 1
figure 1

Central and basal forebrain molecular circadian clock. *Timekeeper genes identified to date in four brain regions involved in processes related to drug addiction, including dopaminergic neurons of the mesolimbic/mesocortical (a) and nigrostriatal (b) dopaminergic systems. Circadian genes: Bmal1, ARNT-like protein 1; Clock, Clock; Cry2, Cryptochrome 2; Per1, Period 1; Per2, Period 2; TIM, Timeless.

The various timekeeper genes, which may have different effects in different parts of the brain and periphery, have been studied to a limited extent, with respect to the genetic basis for specific human disorders. In contrast to numerous single-nucleotide polymorphisms (SNPs) found in other human timekeeper genes such as Per1, Per2, only two variants have been found in the Clock gene: one in the 5′-UTR (101 bp upstream of ATG codon) and 3111 T>C in the 3′-UTR regions.11 A number of studies demonstrated an association of the 3111 T>C SNP with major depression, as well as insomnia and mood disorders. Per2 gene polymorphisms have been associated with bipolar disorders, and the Per3 gene has been associated with delayed sleep phase syndrome, and extreme diurnal preference. This may be relevant for patients with addictive diseases, who frequently adopt abnormal sleep–wake patterns with drug self-administration, of especially alcohol, cocaine and other stimulants. Such self-administration occurs primarily in the early and late evening hours (and sometimes through the night). In contrast, heroin (or other short-acting opiate) addicts usually space their self-administration during regular intervals in daytime and evening, although they may shift their sleep period later than normal, and wake up in the morning in opiate withdrawal.

To date, only one of these genes has been studied for an association with addictive diseases. Spanagel and colleagues reported a study of Per2 gene in 215 alcohol-dependent subjects with low or high alcohol intake, and identified a haplotype of four gene variants associated with low alcohol intake.8 With rodent studies included in the same report, Spanagel et al8 found that Per2 mutant mice drank more alcohol than controls. Also, the brain of mutant mice contained excess levels of glutamate, a situation associated with both cocaine and other stimulant exposure, as well as alcoholism. This finding may be related to the reduction in astrocyte-expressed transporter EAAT1, coupled with a modest increase in a second transporter, EAAT2.8

Further studies of relationships of polymorphisms or haplotypes in timekeeper-related genes in specific addictive diseases would be of interest. Studies from our laboratory12 have identified a functional polymorphism of MOR (mu opioid receptor); we then hypothesized, and other laboratories subsequently have identified, that one copy of this SNP alters critical hypothalamic-pituitary-adrenal (HPA) responsivity to stress. Much earlier, our group and others have shown that the MOR plays a major role in the HPA axis, which is normally under circadian control. We have recently shown a very significant association of this A118G variant of the MOR with both heroin addiction and alcoholism (reviewed in Kreek et al, 2005).13 Therefore, it would be of great interest to determine if polymorphisms of one or more of the timekeeper genes are associated with specific addictive diseases, and possibly with alterations in the stress-responsive circadian HPA axis. This axis has been shown, in laboratory and human studies, to contribute to the acquisition, continuation and relapse to specific addictionsâ–ª