To our knowledge, this is the first study to examine genetic associations with differences in smoking reward and reinforcement due to acute mood manipulation. Past research has shown that negative mood acutely increases ad lib smoking behavior, indicating that smoking is more reinforcing during negative mood. The novel finding from this study is that this influence of mood is associated with genes in the dopamine reward pathway and endogenous opioid pathway.
Among the significant findings, two genes believed related to the dopamine D2 receptor, the DRD2/ANKK1 TaqIA and the DRD2 C957T, were associated with sensitivity to mood effects on smoking reward and reinforcement. Smokers carrying the minor “T” allele of DRD2/ANKK1 TaqIA reported increased cigarette liking during negative versus positive mood, while those heterozygous for the common variant liked the cigarette more during positive versus negative mood induction. The minor allele group also had an increased number of puffs and shorter latency to smoking during negative mood, if they were smoking a nicotine cigarette. These findings are consistent with evidence for an increased risk of cigarette consumption among carriers of the DRD2/ANKK1 TaqIA minor allele (Munafo et al. 2004
). However, this finding may not fit as well with prior evidence that this subgroup of smokers is less responsive to treatment with the antidepressant bupropion for smoking cessation (David et al. 2007b
). Smokers homozygous for the DRD2 C957T allele (CC) also took more cigarette puffs during negative versus positive mood, perhaps consistent with prior evidence for their increased risk for relapse while on placebo and their enhanced therapeutic response to bupropion (Lerman et al. 2006
Two other genes thought to be involved in the dopamine pathway were associated with increased smoking reinforcement due to negative mood. The increase in smoking amount due to negative mood among those given nicotine (but not denic) cigarettes was greater in smokers carrying the DRD4 7 repeat allele, which has been associated with increased risk of smoking relapse (Shields et al. 1998
; David et al. 2008
). We also found that carriers of the 9-repeat allele of the dopamine transporter (SLC6A3) polymorphism had a shorter latency to smoking and increased number of puffs during negative mood induction, consistent with associations of this allele with stress-induced cigarette craving (Erblich et al. 2004
Regarding the two non-dopamine related gene variants examined here, the serotonin transporter promoter variant was unrelated to any of the effects examined, while the functional OPRM1 A118G variant was associated with smoking reward under negative versus positive mood induction. Specifically, smokers carrying the common AA genotype liked the cigarette more in the negative mood condition, while ratings of liking did not differ significantly as a function of mood condition in those with the minor “G” allele. The minor G allele is thought to be a reduced activity allele for the mu opioid receptor, as it relates to reduced mRNA and protein expression (Zhang et al. 2005
). These in vitro data are consistent with evidence that smokers carrying the common AA OPRM1 genotype are more likely than those carrying the “reduced activity” G allele to relapse, and those with the AA genotype also report higher levels of abstinence-induced negative affect (Lerman et al. 2004
). Further, female smokers with the OPRM1 AA genotype are more likely to choose nicotine over denicotinized cigarettes than female smokers carrying the G allele (Ray et al. 2006
). The present findings, therefore, suggest that the AA genotype of OPRM1 may predispose to smoking relapse, in part, via effects of negative mood on smoking reward.
In addition to the novelty of the study focus, genetic associations with increased smoking reward and reinforcement due to negative mood, strengths of this study include some of the methods used. The mood induction procedure was successful in manipulating mood as intended, as we have previously demonstrated (Conklin and Perkins 2005
). Moreover, the magnitude of negative and positive affect changes due to the mood induction procedures did not vary by genotypes, ruling out the notion that these genetic associations with increased smoking reward and reinforcement due to negative mood resulted from some genotypes experiencing greater intensity of negative affect. Second, the within-subject manipulation of mood reduced error variance in the statistical comparisons, all of which involved mood as a factor, as well as genotype. Third, inclusion in analyses of only those smokers who believed their dose instructions resulted in participants whose expectancies for nicotine were successfully manipulated, allowing for a test of genetic association with the influence of expectancies on smoking behavior during negative mood, if present.
The study contained several limitations as well, primarily the small numbers of subjects in some of the genotype subgroups varying in actual nicotine or expectancy for nicotine (e.g., those with the DRD4–7 allele). Consequently, despite the within-subject manipulation of mood, we may not have had adequate power to detect other genetic influences on smoking reward and reinforcement due to negative mood. This problem reflects the practical conflict between studying a limited amount of data from a large sample versus extensive phenotyping of a smaller sample (e.g., assessing prospective responses to different mood induction conditions while varying actual and expected nicotine, as done in this study). A second limitation is that some of the significant effects may have resulted from chance, with 6 genes and 3 gene X mood effects of interest (gene X mood, gene X mood X nicotine, gene X mood X expectancy) for each of the 3 main dependent measures. Yet, the fact that interactions involving SLC6A3 X mood and DRD2/ANKK1 TaqIA X mood were significant for each dependent measure suggests a consistency in associations that is not likely due to chance. Given the complete lack of prior research on genetic associations with smoking reward and reinforcement due to mood, this study was exploratory in nature and the findings were intended to be heuristic and not conclusive. This study should be replicated with larger samples to verify the findings. Third, the ad lib smoking period during mood induction was only 14 mins, and more extended duration of mood may reveal greater or smaller changes in smoking behavior due to genotypes.
In summary, these findings suggest that some dopamine and opioid genes are associated with the degree to which negative mood increases acute smoking reward and reinforcement. Future research should examine associations of these genes with smoking reward and reinforcement under other mood conditions or other acute behavioral manipulations (e.g. stressful challenge, co-administration of alcohol). The associations of genes in other neurobiological pathways with smoking reward and reinforcement during negative mood should also be examined. Results may help identify possible mechanisms to explain increased risk of dependence onset and persistence (i.e. relapse) in those with certain genotypes.