Adb_064.fm

••••••••Original ArticlesDRD4 and CNR1 not strongly related to alcohol cue-reactivityEsther van den Wildenberg et al.
GENETIC STUDY
Polymorphisms of the dopamine D4 receptor gene
(DRD4 VNTR) and cannabinoid CB1 receptor gene
(CNR1) are not strongly related to cue-reactivity after
alcohol exposure
Esther van den Wildenberg1, Rob G. J. H. Janssen2, Kent E. Hutchison3,
Gerard J. P. van Breukelen4 & Reinout W. Wiers1
Faculty of Psychology, Maastricht University, the Netherlands1, Department Population Genetics and Genome Centre Maastricht, the Netherlands 2, Department of Psychology, University of Colorado at Boulder, USA 3, Department of Methodology and Statistics, Maastricht University, the Netherlands4 ABSTRACT
Polymorphisms in the D4 dopamine receptor gene (DRD4) and the CB1 cannabinoid receptor gene (CNR1) have been associated with a differential response to alcohol after consumption. The goal of the present study was to investigate whether heavy drinkers with these polymorphisms would respond with enhanced cue-reactivity after alcohol expo- sure. Eighty-eight male heavy drinkers were genotyped for the DRD4 variable number of tandem repeats (VNTR) [either DRD4 long (L) or short (S)] and the CNR1 rs2023239 polymorphism (either CT/CC or TT). Participants were exposed to water and beer in 3-minute trials. Dependent variables of main interest were subjective craving for alcohol, subjective arousal and salivary reactivity. Overall, no strong evidence was found for stronger cue-reactivity (= outcomedifference between beer and water trial) in the DRD4 L and CNR1 C allele groups. The DRD4 VNTR polymorphism tended to moderate salivary reactivity such that DRD4 L participants showed a larger beverage effect than the DRD4 S participants. Unexpectedly, the DRD4 L participants reported, on average, less craving for alcohol and more subjective arousal during cue exposure, compared with the DRD4 S participants. As weekly alcohol consumption increased, the CNR1 C allele group tended to report more craving for alcohol during the alcohol exposure than the T allele group. The DRD4 and CNR1 polymorphisms do not appear to strongly moderate cue-reactivity after alcohol cue exposure, in male Keywords
Alcohol, CNR1, craving, cue-reactivity, DRD4, polymorphism.
Correspondence to: Esther van den Wildenberg or Reinout W. Wiers, Faculty of Psychology, Experimental Psychology, University of Maastricht, UNS 40, PO BOX 616, 6200 MD Maastricht, the Netherlands. E-mail: [email protected], [email protected] or R.Wiers@ Alcoholism continues to be one of the most prevalent psy- the development of alcoholism, one can also try to chiatric disorders. It is characterized by changes in the identify an endophenotype and relate this to a genetic brain reward centre which lead to compulsive drug seek- polymorphism of interest. An endophenotype is an inter- ing and taking and loss of control over drinking. More or mediate phenotype which can be measured in a con- less 50–60% in the variability for alcoholism vulnerabil- trolled fashion in the lab and which is supposed to reflect ity is determined by genes (Goldman, Oroszi & Ducci the biological underpinnings of the broader phenotype 2005). When investigating the association between a (Gottesman & Gould 2003). Examples of endophenotypes specific gene and the phenotype alcoholism, often no that have been suggested in relation to addictive behav- relationship is found (e.g. Arias, Feinn & Kranzler 2006).
iours are the P300 amplitude reduction (Yoon et al.
Given the fact that alcoholism is a heterogeneous disease 2006) and a low level of response (Schuckit & Smith which is influenced by multiple genes and environmental 1996). Another possible endophenotype might be cue- factors, this is not completely unexpected. Instead of elicited craving (e.g. Hutchison et al. 2002a,b). This study searching for a single gene which might be important to investigated the association of the D4 dopamine receptor 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
DRD4 and CNR1 not strongly related to alcohol cue-reactivity variable number of tandem repeats polymorphism (DRD4 Both have often been associated with alcohol and drug VNTR) and a CB1 cannabinoid receptor gene polymor- use and abuse. Previous research has shown that D2/D4 phism (CNR1) with alcohol cue-reactivity (subjective antagonists such as olanzapine and haloperidol can craving, subjective arousal and salivary reactivity). The reduce urge to smoke (Hutchison et al. 2004), urge to importance of the dopamine and the cannabinoid sys- drink (Hutchison et al. 2001) as well as alcohol consump- tems in addictive behaviours is explained below.
tion (Modell et al. 1993). Olanzapine was found to reduce According to Wise & Bozarth (1987) all drugs (stim- cue-elicited alcohol craving in heavy drinkers and alco- ulants as well as sedatives) result in an initial release of hol-dependent patients with the DRD4 L allele (Hutchi- dopamine in the mesolimbic system, mainly the nucleus son et al. 2003, 2006). These patients also reported accumbens. Robinson & Berridge (1993, 2003) report reduced alcohol consumption, while patients with the evidence that after repeated administration of the drug, short allele (DRD4 S) did not benefit from olanzapine this psychomotor stimulant reaction becomes sensitized, treatment (Hutchison et al. 2006).
or stronger, due to adaptive changes in the brain. Dopam- The dopamine system is not the only neurotransmit- ine release can be generated by drug administration (e.g.
ter system that is modulated by prolonged alcohol use.
a priming dose) and by exposure to alcohol or drug cues The glutamatergic, GABAergic, noradrenergic, opioider- and is often experienced as a craving for or ‘wanting’ of gic, serotonergic and cholinergic systems have all been the drug. Because of the important involvement of the found to play a role in the different aspects related to the dopamine system in reward, genetic variations, also effects of alcohol (Basavarajappa & Hungund 2002). A called polymorphisms, in the genes for these receptors more recent discovery of another system involved in alco- might influence gene expression and protein levels or hol and drug use is the endocannabinoid signalling sys- change the function of the protein. Thereby, these varia- tem. The cannabinoid receptor subtype we will focus on is tions could have an impact on craving and the vulnera- the CB1 receptor which is prevalent in many parts of the bility for developing alcohol and drug-related disorders.
brain, including the brain reward system, such as the The dopamine receptor subtype we will focus on is the D4 ventral tegmental area (VTA), nucleus accumbens and receptor. The D4 receptor is widely distributed through- amygdala (Basavarajappa & Hungund 2002). The can- out the brain and can be found in limbic structures that nabinoid receptor is activated by endogenous cannab- are important in incentive sensitization, such as the inoids that are released from the post-synaptic cell upon nucleus accumbens shell and the amygdala (Van Tol et al.
dopamine release in, e.g. the VTA (Maldonado, Valverde 1991; Schoots & Van Tol 2003). The D4 receptor gene & Berrendero 2006). Drugs such as alcohol, cannabis, (DRD4) has an exon 3 polymorphism, consisting of a nicotine and opioids increase the firing rate of dopamine highly variable number of tandem repeats (VNTR). The neurons, resulting in an increase of endocannabinoids in VNTR consists of a 48-base-pair (bp) motif which can be the VTA. Through a retrograde control mechanism these repeated two up to 11 times, in tandem (Van Tol et al.
neuromodulators activate the pre-synaptic CB1 receptors 1992). The most frequently occurring variants of the that can modulate dopamine activity in the VTA by inhib- DRD4 are the two, four and seven repeats (Van Tol et al.
iting the release of glutamate and gamma-aminobutyric 1992; Schoots & Van Tol 2003). The different repeat acid (GAB Aergic). Prolonged alcohol or drug use leads to sequences have been found to modulate gene expression elevated levels of endocannabinoids which overstimulate and could differentially affect RNA stability and transla- the CB1 receptors resulting in a down-regulation of CB1 tional efficiency (Schoots & Van Tol 2003). The seven- receptor function and its signal transduction (Basavara- repeat allele has been found to show a blunted jappa, Cooper & Hungund 1998; Hungund & Basavara- intracellular response to dopamine in comparison with the two- and four-repeat variants (Asghari et al. 1995).
Genetic variations in the CB1 receptor gene (CNR1) This affects subsequent signal transduction which could have been associated with the severity of alcohol depen- influence the experience of craving.
dence (Schmidt et al. 2002) and with polysubstance Presence of the long allele (≥ 7 repeats; DRD4 L) has abuse (Comings et al. 1997; Zhang et al. 2004). The been found to moderate urge to drink after alcohol con- CNR1 single nucleotide polymorphism (SNP) rs2023239 sumption (Hutchison et al. 2002b) and to moderate the is a T to C substitution that seems to result in increased urge to smoke after smoking cigarettes (Hutchison et al.
mRNA expression (thus increased CB1 receptor expres- 2002a). Cue-elicited urge to use heroin, was found to be sion) in several brain areas, due to a different transcrip- modulated by the presence of at least a copy of the five tion of the DNA. Hutchison (2006) showed that the repeat or longer (Shao et al. 2006). The seven-repeat rs2023239 polymorphism was associated with more allele has also been associated with behavioural disorders alcohol, cannabis and tobacco use, with stronger positive like attention-deficit hyperactivity disorder and with per- subjective reactions to alcohol and smoking and with sonality traits like novelty seeking (Faraone et al. 2001).
more activation in the nucleus accumbens and ventral 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
medial prefrontal cortex, as shown with neuroimaging, measure alcohol-related problems. The questionnaire is during alcohol cue exposure. Furthermore, it seems that divided into two subscales: items 1–3 measure alcohol this CNR1 SNP effectively moderates the effect of olanza- consumption and items 4–10 alcohol-related problems.
pine, both in alcohol-dependent and non-dependent C allele carriers. However, findings are mixed. A recent Drug use. The participants were asked to report the life- association study failed to replicate the findings of Zhang time use of 10 other drugs: cigarettes, cannabis, amphet- et al. (Herman et al. 2006). Also, a small study by amines/speed, XTC/MDMA, cocaine, hallucinogenic Pierucci-Lagha et al. (2006) failed to find stronger subjec- drugs (e.g. magic mushrooms, smartdrugs, LSD), heroin, tive effects after alcohol consumption in C allele heavy medicines/pills (e.g. sedatives, benzodiazepines), other opiates/analgesics (e.g. methadone) and inhalants (e.g.
The main goal of the present study was to investigate laughing gas, poppers). Based upon the question if the whether heavy drinking DRD4 L allele carriers of the participant had ever used any of the 10 different drugs, a DRD4 VNTR and C allele carriers of the CNR1 sum score was calculated ranging from 0 (never used any rs2023239 SNP would show more cue-reactivity (sub- of the drugs) to 10 (ever used all of the 10 drugs).
jective craving, subjective arousal and salivary reactivity) after alcohol cue exposure compared with participants BISBAS scales. In order to measure sensitivity to reward homozygous for the DRD4 S allele resp. CNR1 T allele.
and punishment, the Dutch translation (Franken 2002) of the behavioural activation/approach system (BAS) and behavioural inhibition system (BIS) scales was adminis- MATERIALS AND METHODS
tered (Carver & White 1994). The BIS/BAS questionnaire Participants
consists of 24 items divided into four subscales. The BIS scale (seven items) measures reactions to the anticipation Participants were 88 male heavy drinkers who had pre- of punishment (e.g. ‘I worry about making mistakes’).
viously participated in the cue-reactivity study, described Furthermore, three BAS subscales are included. The BAS below, in relation to the A118G polymorphism of the reward responsiveness scale measures the positive reac- mu-opioid receptor gene (OPRM1; results described tion to the occurrence or anticipation of reward (e.g.
elsewhere). Of the 109 participants who took part in the ‘When I get something I want, I feel excited and ener- original study, 88 provided a second written informed gized’). The BAS Drive items reflect the pursuit of desired consent for additional DNA analyses of the DRD4 and goals (e.g. ‘I go out of my way to get things I want’). The CNR1 polymorphisms. Participants were paid €15 for BAS fun seeking scale consists of items which focus on taking part in the cue-exposure experiment. They did not the willingness to impulsively approach a potentially receive an additional monetary reward for their informed rewarding event (e.g. ‘I crave excitement and new consent for the additional DNA analyses. The study was approved by the Medical Ethics Committee of the Aca- demic Hospital in Maastricht. The A118G polymorphism of the OPRM1 gene was not confounded with the poly- morphisms of the DRD4 VNTR (r = −0.06) or CNR1 SNP Craving for alcohol. Craving for alcohol was assessed by means of a 100-mm Visual Analogue Scale (VAS) rang- ing from ‘absolutely no urge’ to ‘an irresistible urge’ to Materials and measures
drink. A second craving measure was the Dutch transla- tion (Franken, Rosso & van Honk 2003) of the Desire for Alcohol Questionnaire (DAQ; Love, James & Willner Alcohol use. Alcohol use was measured with a self-report 1998). The 14 items assess the desire to drink alcohol at questionnaire (Wiers et al. 1997) based on the Time-Line the moment of testing itself. The DAQ is composed of four Follow-Back method (Sobell & Sobell 1990). Alcohol con- subscales: (1) strong desires and intentions to drink; (2) sumption for each day of the week prior to the experiment negative reinforcement; (3) control over drinking; and (4) was filled out and the participants indicated whether this mild desires to drink. As response format, 70-mm VAS was more, less or equal to what they would normally were used with the anchors ‘strongly disagree’ and drink on that day of the week. A standard drink contains ‘strongly agree’. The mean score across all 14 items was 10-g alcohol. Furthermore, the number of binges (six drinks or more on one occasion) of the past 2 weeks was Arousal. The Affect Grid (Russell, Weiss & Mendelsohn 1989) was used to assess subjective arousal. The mood Alcohol-related problems. The 10-item Alcohol Use Disor- grid is a square of 9 × 9 consisting of two dimensions: a der Identification Test (Saunders et al. 1993) was used to valence dimension ranging from ‘unpleasant feelings’ 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
DRD4 and CNR1 not strongly related to alcohol cue-reactivity (left side of the grid) to ‘pleasant feelings’ (right side) and Table 1 Allele and genotype frequencies for the DRD4 variable
an arousal dimension which ranged from ‘sleepiness’ number of tandem repeat polymorphism (n = 87).
(bottom) to ‘high arousal’ (top). The participant was asked to place an X in one of the 81 possible positions.
Affect descriptors were placed at the corners and sides of the grid to elucidate which mood state belonged to which part of the grid. Based on the position of the X, two scores could be computed (valence and arousal), both ranging from 1 to 9. Only the arousal score was used here.
Salivary reactivity. The salivation measure was based on the Strongin, Hinsie and Peck technique (in White 1977). Participants placed three cotton rolls in the mouth: one under the tongue and two between the inner cheek and lower gum, on each side of the mouth. Weight of the cotton rolls (in milligrams) was determined before and after the 3-minute cue-exposure trial. The amount of saliva that was produced during cue exposure was indi- cated by the weight difference. To prevent loss of saliva, the cotton rolls were put in sealed, small-sized plastic DRD4. The 48-bp VNTR in exon 3 of the DRD4 was assayed using a modification (Anchordoquy et al. 2003) of the method of Sander et al. (1997). The primer sequences (Lichter et al. 1993) were: forward, 5′-AG GACCCTCATGGCCTTG-3′ (fluorescently labelled), andreverse, 5′-GCGACTACGTGGTCTACTCG-3′. This methodresults in PCR product of (in bp): 379, 427, 475 (four repeats), 523, 571, 619 (seven repeats), 667, 715, 763 genotype was performed by Taqman® SNP genotyping and 811. The genotyping success rate was 99%. One assays (Applied Biosystems, Nieuwerkerk ald IJssel, the sample failed. The allele and genotype frequencies for the Netherlands). PCR primers and Minor Groove Binding remaining 87 participants are presented in Table 1.
allele specific probes were ordered from Applied Biosys- Allele frequencies demonstrated Hardy–Weinberg equi- tems (assay ID: C_11600616). Amplification and detec- librium and are consistent with other studies (e.g. Ler- tion were carried out in duplicate in 384w plates on an man et al. 1998). Participants were classified as DRD4 S ABI7900HT Real-time PCR system according to the (both alleles < 7 repeats; S/S) or DRD4 L (homozygous or manufacturer’s procedures, followed by automatic geno- heterozygous for an allele ≥ 7 repeats; L/L or S/L). This type analysis using the Sequence Detection Software v was based on other studies grouping short and long allele 2.3 (Applied Biosystems, Niewerkerk ald IJssel, the Neth- participants (e.g. Lerman et al. 1998) and based on erlands). Three participants were genotyped with the CC molecular studies suggesting a different D4 receptor combination (3.41%), 24 were heterozygous (CT; function for the seven-repeat allele (Asghari et al. 1995).
27.27%) and 61 were homozygous for the common T One participant was homozygous for the seven-repeat allele (TT; 69.32%). Allele frequencies demonstrated Hardy–Weinberg equilibrium. Allele frequency was 0.17 for the C allele and 0.83 for the T allele. In the analyses the CNR1. Genomic DNA was isolated from buccal cells (two Omniswabs per sample, Whatman, Hertogenbosch, the Netherlands) with the QIAamp DNA Mini Kit (QIAgen, Venlo, the Netherlands). For both procedures the manu- A maximum of five participants was tested simulta- facturer’s protocols were followed. DNA concentration neously, at individual tables, separated from each other and purity were measured with a Nanodrop spectropho- by screens. The beverages were hidden under two tometer. Determination of the CNR1 rs2023239 SNP inverted, opaque, tinfoil containers that were placed 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
in front of the participant. The container on the left Table 2 Crosstabs for the DRD4 and CNR1 polymorphisms
side always contained an empty glass and a bottle of commercially labelled spring water (‘Spa Blauw’). The container on the right side always contained an empty beer glass and a bottle of ‘Brand’ beer (all participants had previously indicated that beer was their favourite alcoholic beverage). Before the start of the cue-exposure trials, baseline measures of subjective craving, subjective arousal and saliva production were taken. The par- ticipants were instructed that it was important to per- form all the actions pointed out by the experimenter hypothesized) DRD4 × CNR1 interaction was too low The cue exposure started with a 3-minute relaxation (only seven participants were classified as having the period, based on Monti et al. (1987) and Rohsenow et al.
DRD4 L and CNR1 C allele). The correlation between (2000). Next, the water and alcohol exposure trials fol- both factors was 0.01; therefore, no indication of con- lowed. To prevent transfer effects the water exposure founding between the two genotypes was found. For each always came first and the alcohol exposure second (Monti of the four outcome variables (two for subjective craving, et al. 1987). Participants were asked to place the three one for arousal and one for saliva) the initial model cotton rolls in their mouth in the way previously included interactions of genotype with each covariate.
described to them: one sublingually and two buccally.
Each covariate was first centred by subtracting the total Subsequently, they were asked to lift the container on the sample mean from it, to prevent possible collinearity and left, take the empty glass and fill it up half with water from to allow interpretation of the within-subject effect as bev- the bottle. After this, the 3-minute exposure started. Par- erage effect (alcohol–water) for the average person in the ticipants were asked to raise the glass, to carefully look at sample, and interpretation of genotype × beverage inter- it and sniff the contents for 5 seconds. This procedure was action as interaction for persons scoring average on repeated 15 times in total. After 3 minutes they were asked to put down the glass and cover it with the con- As our main interest was in effects of beverage and tainer. Next, they removed the dental rolls, put them into genotype, our main focus was on the within-subject the plastic bag and sealed it well to prevent loss of saliva.
ANOVA part. This comes down to an ANCOVA with the To prevent a dry mouth from the cotton rolls they were beverage effect (outcome difference between alcohol and given a cup of water to take a sip. Then, the participants water) as dependent variable, which is our measure of filled out the Affect Grid to measure subjective arousal, cue-reactivity, and genotype and covariates as indepen- and the VAS and DAQ to measure craving for alcohol.
dent variables. Genotype × covariate interactions were After the questionnaires had been returned to the exper- deleted from this model if non-significant, using α = imenter, the same procedure was repeated for the 0.01 to correct for multiple testing. If the final model alcoholic beverage under the right container. After the for the within-subject ANOVA part did not contain any cue exposure, the participants received the BISBAS beverage × genotype interaction was not significant, The final part of the experiment consisted of the col- using α = 0.05, then the between-subject ANOVA part was lection of buccal cells with several omni swabs in order to considered. This comes down to an ANCOVA with the aver- determine the genotype. The drug use questionnaire was age of both conditions (water and alcohol) as dependent filled out on the next day, when they returned for another variable. If a beverage × genotype (×covariate) interac- tion was found, then a separate ANCOVA was run per con- dition (water, alcohol). In both cases, i.e. ANOVA of the average of both conditions, and ANOVA per condition, The study design was a 2 (Beverage: Water versus Alcohol) × 2 [Genotype: S versus L (DRD4) or TT versus removed from the model before testing main effects.1 CT/CC (CNR1)] factorial design with between-subject covariates alcohol use and baseline of the outcome at [1] As the participants were tested in groups with a maximum hand, and was analysed accordingly, using repeated mea- of five people, all analyses were repeated with the centred covari- sures ANOVA (SPSS version 12.0.1). Age was not entered ate Block (to test the linear effect of testing in groups) and the in the analyses as a covariate because no differences were square of the variable Block (quadratic effect). None of these found between the groups (Table 3). As can be seen in analyses revealed any significant effects. Therefore, these cova- Table 2, the power for testing a possible (though not riates were left out of the final analyses.
2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
DRD4 and CNR1 not strongly related to alcohol cue-reactivity Pre-test measures
< 1, = 1.44, < 1, < 1, = 1.82, < 1, < 1, < 1, < 1, = 1.62, = 1.09, < 1, = −1.33, As can be seen in Table 3, regarding weekly alcohol use, number of binges and alcohol-related problems, no sig- nificant differences were found between the DRD4 S and L participants or between the C and T allele participants on the CNR1. Also, lifetime drug use did not reveal any differences between the genotype groups.
The DRD4 L participants displayed a higher BAS reward sensitivity compared with the DRD4 S participants, t(85) = −2.44, P = 0.017 (see Table 3). However, thiseffect was no longer significant after multiple testing.
Including the BAS reward score as a covariate in the analyses did not lead to different results. No differences between DRD4 S and DRD4 L participants were found on DRD4 cue-reactivity effects
= Desire for Alcohol Questionnaire; ns Q VAS craving. Within-subjects, a main effect of beverage was found, F1,83 = 68.50, P < 0.001, indicating that par- < 1, ns = −1.54, < 1, ns < 1, ns < 1, ns = −2.44, < 1, ns < 1, ns = −1.15, = 1.24, = 1.71, < 1, ns = 1.10, ticipants were craving more for alcohol after the beer exposure. No significant effects of genotype were found in nor genotype by beverage by covariate. The between- subjects ANOVA showed a significant main effect of DRD4, such that on average over both trials, the DRD4 S group was craving more for alcohol than the DRD4 L group, F1,83 = 8.23, P = 0.005. Average craving of both exposure trials for the S group was 43.91 mm (SE = 1.37), versus 36.21 mm (SE = 2.26) for the L group.
DAQ. For craving on the DAQ, a main effect of beverage was found such that participants craved significantly more for alcohol after the beer exposure, F1,79 = 48.83, P < 0.001. No effects of genotype were found.
Within-subjects, a main effect of beverage was found, such that participants became more aroused after expo- sure to beer, F1,83 = 29.00, P < 0.001. No significant effects were found of genotype. The between-subjects ANOVA revealed a significant main effect of genotype such that the DRD4 L participants were on average more aroused during cue exposure compared with the DRD4 S 1,83 = 6.20, P = 0.015. The average arousal score for the S group was 4.73 (SE = 0.11) versus 5.26 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
Figure 2 Mean scores and SE for the beverage effect (alcohol–
water) on the Desire for Alcohol Questionnaire (DAQ). A border- line significant interaction was found between beverage × CNR1 Figure 1 Mean scores and SE for saliva production during the
× alcohol use (P = 0.019, α = 0.01), such that individuals with a copy water and alcohol exposure for both DRD4 short (S) and long (L) of the C allele tended to show a larger beverage effect (alcohol– allele groups. Participants with the DRD4 L allele reported a bor- water) on craving as alcohol use increased. Simple effects analysis derline significant larger beverage effect (alcohol–water) compared revealed that the effect of genotype was significant when alcohol use with the DRD4 S allele participants (P = 0.068). Separate between- was high (P = 0.041), but not when alcohol use was relatively mod- subjects effects per condition (water or alcohol) did not reveal any The within-subject analysis revealed a borderline signifi- cant beverage × genotype interaction, such that individ- uals with the DRD4 L allele tended to show a larger beverage effect than the DRD4 S individuals, F P = 0.068 (Fig. 1). Separate between-subjects effects per condition (water or alcohol) did not reveal any significant CNR1 cue-reactivity effects
Figure 3 Mean scores and SE for the beverage effect (alcohol–
water) of saliva. A borderline significant interaction was found VAS craving. No significant main or interaction effects between beverage × CNR1 × baseline saliva (P = 0.011, α = 0.01).
Simple effects analysis revealed that the effect of genotype on the beverage effect was significant when baseline saliva production was low (P = 0.009), and non-significant when baseline saliva production DAQ. The within-subjects analysis on the DAQ revealed a borderline significant three-way interaction between beverage × CNR1 × alcohol use, F increased, the C allele group tended to crave more for (α = 0.01, multiple testing), such that participants with a alcohol during the alcohol exposure than the TT group.
C allele showed a larger beverage effect on craving as No within- or between-subjects main effect was found for weekly alcohol consumption increased. For the T allele participants, this pattern was reversed. In order to easily plot the interaction (see Fig. 2), the covariate alcohol use was dichotomized as relatively low (‘moderate’ users) ver- No significant main or interaction effects were found of sus high, based on a median split (39 drinks/week). Sim- ple effects analysis revealed that the genotype effect on the beverage effect was significant only when alcohol use Within-subjects, a borderline significant three-way 1,40 = 4.78, P = 0.041), with the C allele participants showing a significantly larger beverage interaction between beverage × CNR1 × baseline saliva effect than the T allele participants. Separate between- revealed that C allele participants relatively low on base- subjects effects per condition (water and alcohol) line saliva showed a stronger beverage effect than T allele revealed that the interaction between CNR1 × alcohol participants low on baseline, F1,83 = 6.70, P = 0.011 use was borderline significant in the alcohol trial only, (α = 0.01, multiple testing). In order to easily plot the interaction (see Fig. 3), the covariate baseline saliva was 1,83 = 3.79, P = 0.055. Thus, as alcohol consumption 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
DRD4 and CNR1 not strongly related to alcohol cue-reactivity dichotomized as low versus high, based on a median split.
(Cohen 1992), assuming the present d to be the true pop- Simple effects analysis revealed that the genotype effect ulation d, of course. Our sample consisted of 63 DRD4 S on the beverage effect was significant only when baseline allele participants versus 24 participants in the DRD4 L salivary production was relatively low, F1,41 = 7.62, P = 0.009), with the C allele participants showing a sig-nificantly larger beverage effect than the T allele partici- DISCUSSION
pants. Separate between-subjects effects per condition (water and alcohol) revealed that the CNR1 × baseline The present study was designed to investigate whether saliva interaction was significant in the alcohol trial only, non-treatment seeking, heavy drinking male individuals F1,83 = 5.48, P = 0.022. Thus, C allele participants low on carrying a copy of the DRD4 L allele of the D4 receptor baseline saliva responded with significantly more sali- gene or a copy of the CNR1 C allele of the CB1 receptor vary reactivity during the alcohol exposure compared gene would respond with more cue-reactivity after alco- with T allele participants low on baseline saliva hol exposure, compared with participants homozygous for the major allele (DRD4 S or CNR1 T, respectively).
Overall, no strong evidence in support of the hypothe- Effect size
sized associations was found. First, no significant effect of In order to investigate whether the lack of strong geno- genotype was found on cue-reactivity, i.e. on the beverage type effects on the beverage effect (within-subjects) for effect (alcohol–water), for either the DRD4 VNTR or the the DRD4 VNTR as well as the CNR1 polymorphism were CNR1 SNP on any of the dependent variables (subjective due to a power problem, we calculated the 95% confi- craving, subjective arousal and salivary reactivity). One dence intervals and Cohen’s effect size d, for the genotype borderline significant within-subjects main effect was effect on the beverage effect for each outcome (Cohen found for the DRD4 genotype on saliva. The DRD4 L 1992). Also, confidence intervals and effect sizes for the group tended to show a larger beverage effect than the genotype main effect (between-subjects) are shown. As DRD4 S group. Second, between-subjects, two significant can be seen in Table 4 most within-subjects effect sizes main effects were found for the DRD4 genotype. Contrary were rather small. For the DRD4 genotype, a small to to what would be expected, the DRD4 S group reported a medium effect size was found for saliva (d = 0.44). Thus, higher average craving for alcohol on the VAS compared the effect of the DRD4 polymorphism on saliva reactivity with the DRD4 L group. With respect to subjective would have been significant if 41 participants per geno- arousal, the DRD4 L group reported feeling more type subgroup (DRD4 S versus DRD4 L) had been tested aroused on average than the DRD4 S group. Third, after Table 4 95% confidence intervals (CI) and effect sizes for both the DRD4 and CNR1 genotype effects on the within-subjects (WS) bev-
erage effect (alcohol–water) and on the between-subjects (BS) average effect [(water + alcohol)/2] of the four dependent variables.
Effect sizes [Cohen’s d; as computed from mean difference/SE * √(1/N1 + 1/N2) adjusted for covariates] are based on the same final model as 95% CI, i.e.
with covariates (alcohol use and baseline of the outcome at hand) and without three-way interactions. DAQ = Desire for Alcohol Questionnaire;VAS = Visual Analogue Scale.
2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
correcting for multiple testing, two potentially interesting genotype subgroups these would have to consist of at borderline significant interactions were found for the least 100 participants per group (with d = 0.28) and even CNR1 genotype. As weekly alcohol consumption more with the smaller effect sizes. In case of such small increased, the C allele participants showed a larger bev- effect sizes requiring many participants to find a differ- erage effect on the DAQ (due to more craving during the ence, one might question the clinical relevance of this alcohol exposure) than the T allele participants. Further- difference. It should be noted that also in the study by more, it seemed that C allele participants low on baseline Hutchison et al. (2002b) no difference with regard to saliva production showed a larger beverage effect (due to craving was found between the DRD4 S and L groups a higher salivary reactivity after alcohol exposure) com- (only between craving after alcohol consumption versus pared with the T allele participants low on baseline saliva.
a placebo in the DRD4 L group). A differential response in Fourth, contrary to what was expected, no differences subjective craving for the DRD4 S and L group has mainly were found between DRD4 S and DRD4 L participants or been found in individuals who were being treated with between C and T allele participants with respect to alco- olanzapine (Hutchison et al. 2003, 2006).
hol or drug use. Finally, the DRD4 L participants were sig- Previous research by Hutchison et al. (2002b) showed nificantly more sensitive to reward than the DRD4 S that heavy drinkers with the DRD4 L allele reported being participants as shown by a higher score on the BAS less aroused and less subjectively high than the DRD4 S reward subscale. Perhaps the DRD4 L allele is more participants across three alcoholic drinks. It has been related to reward sensitivity than to the related concept of suggested that the DRD4 L individuals might be more novelty seeking, but note that for this trait the initial pos- sensitive to ‘self-medicate’ in order to enhance their itive associations were also not replicated (Kluger, Sieg- dopamine levels due to their lower (baseline) levels of fried & Ebstein 2002). Hence, replication of this positive stimulation. The present study does not support this hypothesis. The DRD4 L participants showed, on average The lack of strong effects of genotype might be due to across both exposure trials, significantly higher levels of several factors. In previous studies the participants were subjective arousal and lower levels of subjective craving subjected to actual alcohol consumption or the smoking compared with the DRD4 S participants. Both findings of a cigarette (e.g. Hutchison et al. 2002a,b). Possibly, in are unexpected based on the idea of understimulation the present study, stronger differences between the geno- playing an important role in the tendency for DRD4 type groups would have been found if the participants L individuals to engage in drinking (Hutchison et al.
had received a ‘priming dose’ of the alcoholic beverage instead of being exposed to the cues only. Also, stronger The marginally significant finding that participants effects might be found in alcohol-dependent patients, with a C allele reported more craving for alcohol during because they usually respond with stronger cue- cue exposure to an alcoholic beverage as their weekly reactivity after exposure to an alcoholic beverage.
alcohol consumption increased could imply that espe- Another possible explanation could be a lack of power to cially the C allele carriers, become increasingly sensitive find differences between the genotype subgroups due to a to the effects of alcohol as they engage in heavier drinking relatively small sample size. In order to investigate this patterns. An increased experience of craving after alcohol possibility we calculated confidence intervals and effect exposure might render them more sensitive to engage in sizes (Cohen’s d) of all within- and between-subjects further drinking. The lack of a similar finding for craving effects of genotype. A small to medium effect size was on the VAS could be due to the single item measurement.
found for the DRD4 VNTR on the beverage effect of saliva.
The DAQ consists of 14 items, which tap different subtle If our DRD4 L group had been somewhat larger (41 subcomponents of craving, something which might not instead of 24 participants) a significant difference be assessed with a single item. The CNR1 interaction with between the two groups would have been found (assum- saliva suggests that the C allele carriers relatively low on ing our sample effect size to equal the population effect baseline saliva production respond with more physiolog- size), with the DRD4 L allele group showing a larger bev- ical cue-reactivity to alcohol cues than their TT counter- erage effect on salivary reactivity than the DRD4 S allele parts relatively low on baseline saliva.
group. Thus, it seems that the DRD4 VNTR moderates Although in the present study, overall, no strong dif- salivary responding to alcohol cues. The lack of a signifi- ferences were found between the genotype groups, with cant beverage × genotype effect on craving (either VAS or regard to alcohol use or cue-reactivity, the DRD4 VNTR DAQ) for the DRD4 as well as the CNR1 genotype cannot and the CNR1 SNP could still prove to be important poly- be explained by a lack of power as becomes clear from the morphisms in moderating the response to pharmacolog- very small effect sizes found for these two craving mea- ical treatment. As the research by Hutchison et al. (2003, sures (ranging from 0.08 to 0.28, see Table 4). In order 2006) showed, the response to the dopamine antagonist to find a significant difference in craving between the olanzapine was moderated by the DRD4 VNTR, with a 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
DRD4 and CNR1 not strongly related to alcohol cue-reactivity better response (less craving and less drinking) in the Comings DE, Muhleman D, Gade R, Johnson P, Verde R, Saucier DRD4 L individuals. As also the endocannabinoid system G, MacMurray J (1997) Cannabinoid receptor gene (CNR1):association with i.v. drug use. Mol Psychiatry 2:161–168.
is involved in addictive behaviours, future research is Faraone SV, Doyle AE, Mick E, Biederman J (2001) Meta- needed to develop and test medications that interact with analysis of the association between the 7-repeat allele of the the endocannabinoid receptor and reduce the urge to dopamine D(4) receptor gene and attention deficit hyper- use alcohol or drugs. Animal research and preliminary activity disorder. Am J Psychiatry 158:1052–1057.
human studies with the CB1 receptor antagonist rimon- Franken IHA (2002) Behavioral approach system (BAS) sensi- abant are promising (see Le Foll & Goldberg 2005 for a tivity predicts alcohol craving. Pers Individ Dif 32:349–355.
Franken IHA, Rosso M, van Honk J (2003) Selective memory for review). Possibly, individuals with specific CB1 receptor alcohol cues in alcoholics and its relation to craving. Cognit polymorphisms such as the rs2023239 will benefit more from treatment with CB1 antagonists such as rimona- Goldman D, Oroszi G, Ducci F (2005) The genetics of addictions: bant than individuals not carrying this polymorphism.
uncovering the genes. Nat Rev Genet 6:521–532.
Future research will elucidate these topics. Additionally, Gottesman II, Gould TD (2003) The endophenotype concept in psychiatry: etymology and strategic intentions. Am J Psychi- more research is necessary to investigate the role of cue- elicited craving as a potential new endophenotype. In the Herman AI, Kranzler HR, Cubells JF, Gelernter J, Covault J quest for specific genes that are involved in the sensitivity (2006) Association study of the CNR1 gene exon 3 alternative for developing alcohol- and drug-related problems it is promoter region polymorphisms and substance dependence.
recommended to measure subcomponents of the hetero- Am J Med Genet B Neuropsychiatr Genet 141:499–503.
Hungund BL, Basavarajappa BS (2004) Role of endocannab- geneous disease such as alcohol (or drug) dependence.
inoids and cannabinoid CB1 receptors in alcohol-related The chance of identifying relevant genes and polymor- behaviors. Ann N Y Acad Sci 1025:515–527.
phisms is optimized this way, which is beneficial for a Hutchison KE (2006) Cue-elicited craving and acute responses more targeted treatment and prevention.
to alcohol: influence of the dopamine, opioid, and cannab-inoid receptor genes. Alcohol Clin Exp Res 30(Suppl.):253A– Acknowledgements
Hutchison KE, LaChance H, Niaura R, Bryan A, Smolen A The authors would like to acknowledge Ellen Lambrichs, (2002a) The DRD4 VNTR polymorphism influences reactivity Brad Pemberton and Andrew Smolen for the genotyping to smoking cues. J Abnorm Psychol 111:134–143.
Hutchison KE, McGeary J, Smolen A, Bryan A, Swift RM and Lotte Voorham for help in the data collection. This (2002b) The DRD4 VNTR polymorphism moderates craving research was supported in part by a ‘VIDI’ grant from the after alcohol consumption. Health Psychol 21:139–146.
Dutch National Science Foundation (N.W.O.) to R.W.W.
Hutchison KE, Ray L, Sandman E, Rutter MC, Peters A, Davidson D, Swift R (2006) The effect of olanzapine on craving andalcohol consumption. Neuropsychopharmacology 31:1310– References
Hutchison KE, Rutter MC, Niaura R, Swift RM, Pickworth WB, Anchordoquy HC, McGeary C, Liu L, Krauter KS, Smolen A Sobik L (2004) Olanzapine attenuates cue-elicited craving for (2003) Genotyping of three candidate genes following whole tobacco. Psychopharmacology (Berl) 175:407–413.
genome preamplification of DNA collected from buccal cells.
Hutchison KE, Swift R, Rohsenow DJ, Monti PM, Davidson D, Almeida A (2001) Olanzapine reduces urge to drink after Arias A, Feinn R, Kranzler HR (2006) Association of an drinking cues and a priming dose of alcohol. Psychopharma- Asn40Asp (A118G) polymorphism in the mu-opioid receptor gene with substance dependence: a meta-analysis. Drug Alco- Hutchison KE, Wooden A, Swift RM, Smolen A, McGeary J, Adler L, Paris L (2003) Olanzapine reduces craving for alco- Asghari V, Sanyal S, Buchwaldt S, Paterson A, Jovanovic V, Van hol: a DRD4 VNTR polymorphism by pharmacotherapy Tol HH (1995) Modulation of intracellular cyclic AMP levels interaction. Neuropsychopharmacology 28:1882–1888.
by different human dopamine D4 receptor variants. J Neuro- Kluger AN, Siegfried Z, Ebstein RP (2002) A meta-analysis of the association between DRD4 polymorphism and novelty seek- Basavarajappa BS, Cooper TB, Hungund BL (1998) Chronic eth- anol administration down-regulates cannabinoid receptors in Le Foll B, Goldberg SR (2005) Cannabinoid CB1 receptor antag- mouse brain synaptic plasma membrane. Brain Res 793:212– onists as promising new medications for drug dependence. J Basavarajappa BS, Hungund BL (2002) Neuromodulatory role Lerman C, Caporaso N, Main D, Audrain J, Boyd NR, Bowman of the endocannabinoid signaling system in alcoholism: an ED, Shields PG (1998) Depression and self-medication with overview. Prostaglandins Leukot Essent Fatty Acids 66:287– nicotine: the modifying influence of the dopamine D4 receptor Carver CS, White TL (1994) Behavioral inhibition, behavioral Lichter JB, Barr CL, Kenedy JL, Van Tol HHM, Kidd KK, Livak KJ activation, and affective responses to impending reward and (1993) A hypervariable segment in the human dopamine punishment: the BIS/BAS Scales. J Pers Soc Psychol 67:319– receptor D4 (DRD4). Hum Mol Genet 2:767–773.
Love A, James D, Willner P (1998) A comparison of two alcohol Cohen J (1992) A power primer. Psychol Bull 112:155–159.
craving questionnaires. Addiction 93:1091–1102.
2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220
Maldonado R, Valverde O, Berrendero F (2006) Involvement of Schoots O, Van Tol HH (2003) The human dopamine D4 recep- the endocannabinoid system in drug addiction. Trends Neu- tor repeat sequences modulate expression. Pharmacogenom- Modell JG, Mountz JM, Glaser FB, Lee JY (1993) Effect of halo- Schuckit MA, Smith TL (1996) An 8-year follow-up of 450 sons peridol on measures of craving and impaired control in alco- of alcoholic and control subjects. Arch Gen Psychiatry holic subjects. Alcohol Clin Exp Res 17:234–240.
Monti PM, Binkoff JA, Abrams DB, Zwick WR, Nirenberg TD, Shao C, Li Y, Jiang K, Zhang D, Xu Y, Lin L, Wang Q, Zhao M, Jin Liepman MR (1987) Reactivity of alcoholics and nonalcohol- L (2006) Dopamine D4 receptor polymorphism modulates ics to drinking cues. J Abnorm Psychol 96:122–126.
cue-elicited heroin craving in Chinese. Psychopharmacology Pierucci-Lagha A, Covault J, Herman A, Kranzler HR (2006) The cannabinoid receptor gene (CNR1) moderates the subjec- Sobell LC, Sobell MB (1990) Self-report issues in alcohol abuse: tive response to alcohol in healthy individuals. Alcohol Clin state of the art and future directions. Behav Assess 12:77–90.
Van Tol HH, Bunzow JR, Guan HC, Sunahara RK, Seeman P, Robinson TE, Berridge KC (1993) The neural basis of drug crav- Niznik HB, Civelli O (1991) Cloning of the gene for a human ing: an incentive-sensitization theory of addiction. Brain Res dopamine D4 receptor with high affinity for the antipsychotic Robinson TE, Berridge KC (2003) Addiction. Annu Rev Psychol Van Tol HH, Wu CM, Guan HC, Ohara K, Bunzow JR, Civelli O, Kennedy J, Seeman P, Niznik HB, Jovanovic V (1992) Multiple Rohsenow DJ, Monti PM, Hutchison KE, Swift RM, Colby SM, dopamine D4 receptor variants in the human population.
Kaplan GB (2000) Naltrexone’s effects on reactivity to alcohol cues among alcoholic men. J Abnorm Psychol 109:738–742.
White KD (1977) Salivation: a review and experimental inves- Russell JA, Weiss A, Mendelsohn GA (1989) Affect Grid: a single- tigation of major techniques. Psychophysiology 14:831– item scale of pleasure and arousal. J Pers Soc Psychol 57:493– Wiers RW, Hoogeveen KJ, Sergeant JA, Gunning WB (1997) Sander T, Harms H, Dufeu P, Kuhn S, Rommelspacher H, High- and low-dose alcohol-related expectancies and the dif- Schmidt LG (1997a) Dopamine D4 receptor Exon III alleles ferential associations with drinking in male and female ado- and variation of novelty seeking in alcoholics. Am J Med Genet lescents and young adults. Addiction 92:871–888.
Wise RA, Bozarth MA (1987) A psychomotor stimulant theory Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M of addiction. Psychol Rev 94:469–492.
(1993) Development of the Alcohol Use Disorders Identifica- Yoon HH, Iacono WG, Malone SM, McGue M (2006) Using the tion Test (AUDIT): WHO collaborative project on early detec- brain P300 response to identify novel phenotypes reflecting tion of persons with harmful alcohol consumption. II.
genetic vulnerability for adolescent substance misuse. Addict Schmidt LG, Samochowiec J, Finckh U, Fiszer Piosik E, Horod- Zhang PW, Ishiguro H, Ohtsuki T, Hess J, Carillo F, Walther D, nicki J, Wendel B, Rommelspacher H, Hoehe MR (2002) Asso- Onaivi ES, Arinami T, Uhl GR (2004) Human cannabinoid ciation of a CB1 cannabinoid receptor gene (CNR1) receptor 1: 5′ exons, candidate regulatory regions, polymor- polymorphism with severe alcohol dependence. Drug Alcohol phisms, haplotypes and association with polysubstance 2007 The Authors. Journal compilation 2007 Society for the Study of Addiction Addiction Biology, 12, 210–220

Source: http://www.impliciet.eu/adapt/publications/2007%20VdWildenbergetalAddBiolDRD4CNR1.pdf