While the heritability of cigarette smoking and nicotine dependence (ND) is well-documented, the contribution of specific genetic variants to specific phenotypes has not been closely examined. The objectives of this study were to test the associations between 321 tagging single-nucleotide polymorphisms (SNPs) that capture common genetic variation in 24 genes, and early smoking and ND phenotypes in novice adolescent smokers, and to assess if genetic predictors differ across these phenotypes.
In a prospective study of 1294 adolescents aged 12–13 years recruited from ten Montreal-area secondary schools, 544 participants who had smoked at least once during the 7–8 year follow-up provided DNA. 321 single-nucleotide polymorphisms (SNPs) in 24 candidate genes were tested for an association with number of cigarettes smoked in the past 3 months, and with five ND phenotypes (a modified version of the Fagerstrom Tolerance Questionnaire, the ICD-10 and three clusters of ND symptoms representing withdrawal symptoms, use of nicotine for self-medication, and a general ND/craving symptom indicator).
The pattern of SNP-gene associations differed across phenotypes. Sixteen SNPs in seven genes (ANKK1, CHRNA7, DDC, DRD2, COMT, OPRM1, SLC6A3 (also known as DAT1)) were associated with at least one phenotype with a p-value <0.01 using linear mixed models. After permutation and FDR adjustment, none of the associations remained statistically significant, although the p-values for the association between rs557748 in OPRM1 and the ND/craving and self-medication phenotypes were both 0.076.
Because the genetic predictors differ, specific cigarette smoking and ND phenotypes should be distinguished in genetic studies in adolescents. Fifteen of the 16 top-ranked SNPs identified in this study were from loci involved in dopaminergic pathways (ANKK1/DRD2, DDC, COMT, OPRM1, and SLC6A3).
Citation: O'Loughlin J, Sylvestre M-P, Labbe A, Low NC, Roy-Gagnon M-H, Dugas EN, et al. (2014) Genetic Variants and Early Cigarette Smoking and Nicotine Dependence Phenotypes in Adolescents. PLoS ONE 9(12): e115716. https://doi.org/10.1371/journal.pone.0115716
Editor: Huiping Zhang, Yale University, United States of America
Received: April 17, 2014; Accepted: November 30, 2014; Published: December 29, 2014
Copyright: © 2014 O'Loughlin et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that, for approved reasons, some access restrictions apply to the data underlying the findings. Ethical considerations prevent public sharing of data. Access to NDIT data is open to any university-appointed or affiliated investigator upon successful completion of the application process. Masters, doctoral and postdoctoral students may apply through their primary supervisor. To gain access, applicants must complete the ‘Request for NDIT Data/DNA Application Form’ (http://ndit.crchum.qc.ca/main.php?p=25) and return it to the principal investigator (firstname.lastname@example.org). Approval will be based on scientific merit, relevance, and overlap with other projects. The outcome of the approval process, including conditions (i.e. adding investigators), recommendations (i.e. reorienting objectives) and/or reason(s) for rejection (e.g. overlap with other analyses), will be communicated by e-mail to applicants within 4–6 weeks of receipt of the application. For more information, visit www.nditstudy.ca or contact the principal investigator.
Funding: J O'Loughlin holds a Canada Research Chair in the Early Determinants of Adult Chronic Disease. I Karp is a Fonds de la Recherche en Santé du Québec Junior 1 Scholar and Canadian Institutes of Health Research New Investigator. This work was supported by Canadian Cancer Society grant numbers 010271 and 017435 awarded to J O'Loughlin (http://www.cancer.ca/en/?region=qc). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Cigarette smoking remains a major preventable cause of disability and premature death. Nicotine dependence (ND), which involves a complex interplay between learned or conditioned factors, genetics, and social and environmental factors , underpins this burden because it causes (prolonged) exposure to carcinogens and the other deleterious constituents in cigarette smoke.
Pharmacologic reasons for nicotine use include mood enhancement, either directly or through relief of withdrawal symptoms, and augmentation of mental or physical function . The psychoactive effects of nicotine are mediated through its influence on the dopamine, serotonin, γ-aminobutyric acid, glutamate, and opioid peptide neurotransmitter systems , all of which are involved in the mesocorticolimbic dopamine system and its connections to the nucleus accumbens and amygdala  as parts of the dopaminergic reward system . The release of mesolimbic dopamine is triggered by nicotine, which activates striatal neuronal acetylcholine receptors and inhibits its reuptake, thereby increasing levels of synaptic dopamine .
Twin and family studies ,  concur that smoking has an important genetic component. The estimated heritability for cigarette smoking initiation is 37% in males and 55% in females. For smoking persistence, it is 59% in males and 46% in females . Sullivan et al.  reported a heritability estimate of 70% for tobacco dependence. Because smoking is a complex behavior, it is likely that many genes are involved, each making a small contribution .
A ‘reward deficiency syndrome’ has been postulated as one unifying theme to account for the role of diverse neurotransmitters in ND. Consequently many studies have evaluated genes in opioid, serotinergic, dopaminergic, nicotinic, muscarinic and cholinergic receptor genes  although candidate gene association studies frequently produce inconsistent results , , . Recently in a meta-analysis of 15 genome-wide linkage scans, Han et al.  identified several chromosomal regions showing evidence of linkage with smoking behavior. Several of these regions contained notable candidate genes including the nicotinic acetylcholine receptor α4 subunit gene (CHRNA4) for maximum number of cigarettes smoked in 24 hours, the PLEKHG1 gene and the μ-opioid receptor gene (OPRM1) for ND, and DRD4 and COMT for smoking behavior and maximum number of cigarettes smoked in 24 hours. However, linkage was not detected to regions that contain other candidate genes such as the ANKK1-DRD2 gene locus and DDC.
Genome-wide association studies (GWAS) have also confirmed the role of variants in several nicotinic receptor genes as well as in CYP2A6. Three meta-analyses of GWAS, , ,  all of which used cigarettes smoked per day as the phenotype, identified genetic variants (i.e., single-nucleotide polymorphisms (SNPs)) in the q25.1 region of chromosome 15, which contains a cluster of nicotinic receptor genes (CHRNA5-CHRNA3-CHRNB4). Two of the meta-analyses identified the q13.2 region of chromosome 19 that contains the EGLN2/CYP2A6 locus. One study identified the CHRNB3/CHRNA6 locus , and another observed an association between the DBH locus and smoking cessation .
The general lack of consistency across studies likely relates to several issues including the assorted phenotypes studied (i.e., cigarettes per day, maximum cigarettes per day, ND, smoking status, initiation, withdrawal, and regular smoking), differing candidate gene coverage, imperfect choice of controls in case control studies, and incomplete control of population heterogeneity , . The accurate parsing of broad constructs such as ND which likely meld numerous phenotypes, may help distinguish the unique importance of specific genes and pathways.
While several candidate gene studies have been undertaken in youth, , ,  most genetic studies on smoking have been conducted in adults. Accumulating evidence suggests that ND symptoms can occur within months of first puff in some young smokers, ,  challenging older assumptions that the development of ND takes at least two years after smoking onset . Therefore, better understanding of the genetic predictors of levels of cigarette smoking and the development of ND in the early stages of smoking onset is needed. The objectives of this study were to test the associations between 321 tagging single-nucleotide polymorphisms (SNPs) that capture common genetic variation in 24 genes, and early smoking and ND phenotypes in novice adolescent smokers, and to assess if genetic predictors differ across these phenotypes.
Materials and Methods
The Nicotine Dependence in Teens (NDIT) Study is a prospective cohort investigation of the natural course of ND in adolescents . 1294 students aged 12–13 years were recruited in 1999–2000 in a convenience sample of 10 Montreal-area secondary schools selected to include French and English schools, schools in urban, suburban, and rural areas, and schools in high, moderate and low-income neighbourhoods. Participants were recruited in all grade 7 classes. The baseline response proportion was 55%. Nonparticipation in the NDIT study at baseline primarily related to the need for blood sampling for genetic analysis and to a province-wide labour dispute that resulted in some teachers not collecting consent forms.
Data were collected in classroom-administered self-report questionnaires every three months during the 10-month school year for five years from grade 7 to 11 (1999–2005), for a total of 20 data collection cycles. Blood was collected from 523 participants in 2002. In 2007 participants (mean age 20 years) completed mailed self-report questionnaires in cycle 21, and 336 participants who had not provided blood provided saliva samples in Oragene DNA kits (DNA Genotek Inc. Kanata, Ontario, Canada). In total, 859 participants (66% of 1294) provided either blood or saliva. DNA was extracted using established protocols and genotyped at the McGill University and Genome Quebec Innovation Centre.
Parents/guardians provided informed written consent and participants, who had attained legal age, provided informed written consent in cycle 21. NDIT received institutional approval from the Centre de recherche du Centre Hospitalier de l'Université de Montréal (CRCHUM) (ND06.087).
Data on all six phenotypes were collected in each data collection cycle. Data on number of cigarettes smoked per month were collected in a 3-month recall  with two items for each month, one measuring number of days on which the participant had smoked, and one measuring average number of cigarettes smoked per day. Total number of cigarettes smoked in each of the past three months was computed and “number of cigarettes” represented the total number of cigarettes smoked in the past three months. The intraclass correlation coefficient quantifying test-retest reliability of number of cigarettes smoked per month based on the 3-month recall was 0.64 .
Table S1 in S1 File details the items that comprise each of the five ND indicators investigated including response options and psychometric properties. Briefly, well-known ND indicators included the modified Fagerstom Tolerance Questionnaire (mFTQ)  and an indicator based on the six criteria for tobacco dependence in the International Classification of Diseases 10th Revision (ICD-10) . The other three indicators were ND symptom clusters. In earlier principal components analyses , clusters of ND-related symptoms in young smokers were identified from among 46 items drawn from the Hooked on Nicotine Checklist , the Stanford Dependence Index , and from six focus groups with young smokers . This analysis resulted in the creation of three new scales: an indicator of withdrawal; an indicator of self-medication to alleviate negative affect; and a general ND indicator which included several items that measured cravings referred to herein as a general ND/craving symptom indicator.
We included genes from loci that met the genome-wide significance threshold in GWAS studies, including CHRNA5, CHRNA3, CHRNB4, CHRNB3, CHRNA6, EGLN2 and DBH (Table S2 in S1 File). In addition, we selected genes from recent reviews that were reported to be associated with a smoking or ND phenotype in at least two articles and which were involved in neurotransmitter pathways implicated in the development of substance dependence or abuse. These included BDNF, CHAT, CHRNA4, CHRNB2, COMT, DDC, ANKK1, DRD2, DRD4, MAO-A, OPRM1, SLC6A3, SLC6A4 and TH. Finally, we included three other genes (CHRNA7, GSTM1 and NR4A2) of specific interest to the authors, which made biological sense as stated above but for which there was only one article that reported an association. Of the 24 genes investigated, seven (ANKK1, DBH, DDC, DRD2, DRD4, SLC6A3 (DAT1), and TH) were in the dopaminergic pathway.
A ‘greedy’ pairwise tagging approach , as implemented in the software Haploview , was used to tag polymorphisms present in these genes using the International HapMap Project CEU data . We “forced in” 36 candidate SNPs of interest. We captured all SNPs with a minor allele frequency (MAF) >2.5% in the HapMap CEU population but selected tagging SNPs with MAF >5%. The linkage disequilibrium (LD) cut-off was r2 = 0.8. We tagged polymorphisms from 10 kb upstream to 10 kb downstream of each gene using genomic positions according to NCBI build 36. SequenomiPlex Gold technology (SequenomInc, San Diego, California) was used for genotyping. The sample and SNP call rate thresholds were 0.95 and 0.80, respectively. Genotype reproducibility was >0.998. A total of 394 SNPs from 24 genes were genotyped. Seventy-three SNPs were excluded (three were monomorphic, three had >20% missing values, 51 had MAF <0.05, and 19 were in Hardy-Weinberg disequilibrium in ≥1 ethnic group). A total of 321 SNPs (Table S2 in S1 File) were retained for analysis.
Of 859 participants with a DNA sample, 544 reported cigarette smoking in at least one data collection cycle. The association between each SNP and each phenotype was investigated with linear mixed models to account for the correlation between repeated measures within participants (i.e., a random intercept and a random linear slope for age were included to account for within-participant correlation). All models also included fixed effects for sex, ethnicity (European, French-Canadian, or Other/Mixed) and age, which was represented by both linear and quadratic terms.
The distribution of number of cigarettes was left-skewed and therefore transformed using the natural logarithm. Each SNP-phenotype association was investigated under an additive genetic model. A total of 21 SNPs (identified in Table S3 in S1 File) with a genotypic frequency ≤15 were recoded into a “dominant model”.
A permutation approach was used to take the dependence structure of the variables used for each of the 1,926 models considered into account, corresponding to the combinations of 321 SNPs from 24 candidate genes and six correlated phenotypes –. Permutation tests are more accurate when variables are correlated  as in this analysis, given the moderate to high correlations between phenotypes (Pearson correlation coefficients ranged from 0.47 to 0.83 (Table S4 in S1 File)) and the LD between some SNPs. We thus preserved the structure of the data by permuting individual vectors of 321 genotypes and individual sets of repeated measures of six phenotypes.
Associations with raw p-values <0.01 in the linear mixed models were regarded as statistically significant and multiple testing adjustments were performed using both the false discovery rate (FDR) method  and Westfall and Young maxT procedure . Analyses were conducted using R  and the lme4 package for linear mixed models . Permutation analyses were conducted on the supercomputer Briarée of the Université de Montréal managed by Calcul Québec and Compute Canada.
Forty-one percent of participants were male (Table 1). Most participants (76%) were of French-Canadian or other European ancestry. Mean (sd) age at cigarette smoking onset was 14.7 (2.5) years. Participants reported a median (IQR) of 13 (160) cigarettes smoked in the past 3 months.
The phenotypes investigated were moderately-to-strongly correlated (r = 0.47–0.83) (Table S4 in S1 File). Two SNPs in COMT (rs2020917 and rs8140265) were in high LD (r2 = 0.98) (Table S5 in S1 File). Similarly, the r2 for SNP rs2242592 in ANKK1 and rs6276 in DRD2 was 0.98 (hereafter referred to as DRD2/ANKK1). Finally two pairs of SNPs in OPRM1 were in high LD (r2 = 0.74 for rs510769 and rs557748, and r2 = 0.94 for rs590761 and rs613341).
Table S3 in S1 File shows the estimated regression coefficients for all SNP-phenotype associations investigated controlling for age, sex and ethnicity. While several associations were statistically significant in the linear mixed models, none remained significant after correction for multiple testing according to the significance level obtained from the maxT method (α = 0.000033). None of the FDR corrected permutation-based p-values were statistically significant, but the p-values for the association between rs557748 in OPRM1 and the ND/craving and self-medication phenotypes were both 0.07632.
Table 2 shows the estimated beta coefficients for all SNP-phenotype associations in which the association was nominally statistically significant (i.e., the p-value before permutation in the linear mixed models <0.01). Sixteen SNPs located in seven genes (ANKK1, DRD2, CHRNA7, COMT, DDC, OPRM1, and SLC6A3) were associated with at least one phenotype. MAFs for these 16 SNPs were ≥0.08 (Table 3).
Five SNPs from two loci (rs887200 in COMT, and rs510769, rs557748, rs590761 and rs613341 in OPMR1) were associated with more than one phenotype (Table 4). In OPRM1, rs557748 was associated with all six phenotypes and rs510769 was associated with all the phenotypes except number of cigarettes. Number of cigarettes was the only phenotype investigated that was associated with SNPs in ANKK1/DRD2 and with rs921451 in DDC.
From an analysis of 24 candidate genes in this longitudinal study of adolescents, we identified 16 tagging SNPs at seven genetic loci that were associated with number of cigarettes smoked in the past three months and/or ND phenotypes. Because none of the associations remained statistically significant after correction for multiple testing, the findings in this study must be viewed as preliminary. This discussion focuses on the nominally significant findings.
Despite the relatively high correlation between the phenotypes investigated, the specific pattern of SNP associations differed across phenotypes. In particular, the pattern of SNP-gene associations for number of cigarettes was distinct from the patterns of SNP-gene associations for the ND phenotypes. Number of cigarettes was the only phenotype investigated that was associated with SNPs in ANKK1/DRD2 and with one SNP in DDC. Overall these findings support the tenet that accurate parsing of broad constructs such as smoking and ND may help identify and distinguish the relative importance of specific genes and pathways.
The most notable finding in this study is that five of the seven genes with SNPs associated with the phenotypes in this study (i.e., ANKK1/DRD2, DDC, CHRNA7, and SLC6A3) are directly involved in the dopaminergic pathway. Further, the opioid pathway was also implicated (i.e., OPRM1). This suggests that the rewarding effects of nicotine may be salient during early smoking, and that novice smokers genetically predisposed to experiencing positive dopamine or opioid-related psychoactive effects from nicotine may be susceptible to continuing and/or escalating cigarette consumption and developing ND.
COMT is responsible for the breakdown of catecholamines (i.e., dopamine, epinephrine and norepinephrine) . COMT Val158Met (rs4680), which has a MAF of 0.50 in European populations , causes a valine (high-activity allele) to methionine (low-activity allele) substitution that results in a 3–4-fold difference in COMT enzyme activity , . In our study, rs4680 was not associated with any of the phenotypes investigated although four other COMT SNPs (two of which (rs2020917 and rs8140265) were in high LD) were associated with the ND indicators. In fact, rs887200 was associated with four of the five ND phenotypes, suggesting that COMT may underpin the early development of ND in novice smokers. COMT findings across studies are generally inconsistent , , .
The candidate gene DDC encodes DOPA decarboxylase which catalyzes the decarboxylation of DOPA to dopamine. Ma et al.  found that rs921451 was associated with both number of cigarettes smoked per day and the Heaviness of Smoking Index in a European sample of 200 nuclear families (p = 0.01–0.04). Further, Yu et al.  reported that rs921451 was associated with the Fagerstrom Test for Nicotine Dependence (FTND) score in a sample of 319 African-American and 302 European-American families selected for cocaine or opioid dependence (p = 0.002). In the current study, the DDC SNP rs4947644 was associated with one ND phenotype (i.e., ND/cravings), and rs921451 was associated with number of cigarettes.
Although some studies have not detected an association with variation at the ANKK1/DRD2 locus –, multiple studies including two meta-analyses ,  report that ANKK1/DRD2 is associated with smoking phenotypes , – including progression from never to current smoking , , . In addition, ANKK1/DRD2 haplotypes have been associated with daily smoking  and ND , . In our study, one ANKK1 SNP (rs2242592) and three DRD2 SNPs (rs4936270, rs4586205 and rs6276) were associated with number of cigarettes smoked in the past three months, but not with any of the ND phenotypes. The DRD2 SNP rs4586205 is part of the 4-SNP haplotype associated with the Heaviness of Smoking Index in 1266 African-Americans from 402 families . These findings underscore again that number of cigarettes smoked should not be viewed as a proxy measure for ND.
Our study found that rs11737901 in SLC6A3 (also known as DAT1 (i.e., dopamine active transporter 1)) was associated with withdrawal. This gene encodes a membrane-spanning protein that mediates the synaptic reuptake of dopamine. SLC6A3 is the primary regulator of dopamine neurotransmission and is expressed primarily in areas characterized by dopaminergic circuits (striatum and nucleus accumbens). Early studies found an association between SLC6A3 variants and smoking which several later studies failed to replicate , . However in a sample of 668 nicotine-dependent siblings age <18 years in China, the risk of ND (FTND ≥8) was 3-fold higher in those with the SLC6A3 rs27072-A allele . Bergen et al.  reported an association between three SNPs (rs2975226, rs2652510 and rs2652511) and baseline FTND scores in an adult sample of 828 white smokers in two trials of smoking cessation and a pedigree cohort. SLC6A3 contains a variable number of tandem repeats that affects protein quantity. This polymorphism has two common alleles, a 10-repeat allele and a 9-repeat allele. In a sample of 2,448 young adults from the National Study of Adolescent Health, never smokers and current non-smokers had a higher frequency of the 9-repeat allele, suggesting a protective effect . Stapleton et al. , in a meta-analysis of the dopamine transporter genes and smoking cessation, found that four of five cohorts reported a trend in favor of cessation in those carrying the 9-repeat allele. Finally in 2013, Hiemstra et al.  did not detect an association between SLC6A3 genotypes and smoking onset in a longitudinal sample of 365 adolescents.
Mu-opioid receptors are expressed in multiple brain regions and mediate feelings of reward, analgesia and withdrawal. Mu-receptors in the ventral tegmental area are found predominantly on GABAnergic neurons and decrease the level of GABA released, which in turn disinhibits dopamine neurons. The opioid receptor is the primary site of action for highly addictive opiates. Non-opiate drugs such as nicotine may also activate mu-opioid receptors by stimulating the release of endogenous endorphins . OPRM1 (opioid receptor, mu 1) variants have been associated with increased mu-receptor binding in smokers  as well as with smoking reward . In our study, the OPRM1 SNPs rs510769 and rs557748 were in high LD and the results for these two SNPs were very similar. While rs557748 was associated with all phenotypes, rs510769 was associated with five of the six phenotypes investigated. rs510769 was recently reported to be associated with the plasma concentration of cotinine . Our results support the role of this gene across several diverse early smoking phenotypes.
In addition to SNPs from genes involved in dopaminergic pathways, our study also identified an association between rs868437 in the CHRNA7 nicotinic receptor gene and the mFTQ. In earlier work using NDIT data, we reported that several SNPs located in CHRN genes (including rs7178176 in CHRNA7) are associated with dizziness at first inhalation, a smoking initiation phenotype that may relate to sustained smoking . There is considerable evidence from both candidate-gene and GWAS studies that nicotinic receptor genes are associated with smoking phenotypes including cigarettes per day , , , , , serum cotinine , withdrawal , the Fagerstrom test  and early smoking behaviors . However, variation in CHRNA7 has not been associated with smoking phenotypes as often as other members of the family. One notable exception was confined to African Americans . Several studies used longitudinal designs in population-based samples of youth, but the phenotypes investigated differed from those in our study , . Future investigations of nicotinic receptor genes need to differentiate between cigarette smoking and ND phenotypes, at least in novice smokers.
Limitations of this study include the use of self-reported phenotypes which could result in misclassification. However, self-reports of number of cigarettes smoked are generally considered to be valid and reliable in youth,  and there are no gold standard measures of ND. Even though we controlled for ethnicity, there may be residual differences in ND phenotypes as well as in LD structure between individuals of French-Canadian, European, and Other/Mixed ancestry. Finally, our study did not detect significant associations between SNPs and the six phenotypes after correction for multiple testing possibly due to a lack of statistical power.
Genetic predictors appear to differ across early smoking and ND phenotypes supporting the tenet that accurate parsing of broad constructs such as smoking and ND may help distinguish the relative importance of specific genes and pathways in early smoking and the development of ND. Five of the seven genetic loci identified in this study are directly involved in the dopaminergic pathway suggesting that the rewarding psychoactive effects of nicotine may be salient during early smoking in novice smokers.
This file contains Tables S1–S5. Table S1, Questionnaire Items, Response Options and Psychometric Properties of Five Nicotine Dependence Phenotypes. NDIT Study, 1999–2008. Table S2, List of 24 Genes and 321 Tag SNPs Tested for an Association with Number of Cigarettes Smoked and Nicotine Dependence Phenotypes, NDIT Study, 1999–2008. Table S3, Estimated Beta Coefficients and P-Values for the Associations Between 321 SNPs And Number of Cigarettes Smoked and Nicotine Dependence Phenotypes. (n = 544). NDIT Study, 1999–2008. Table S4, Correlation Coefficients Between Number of Cigarettes Smoked and Nicotine Dependence Phenotypes. NDIT Study (n = 544), 1999–2008. Table S5, Linkage Disequilibrium Correlation Coefficients for SNPs in Genes COMT, DDC, DRD2/ANKK1 and OPRM1 that Were Statistically Significantly Associated with the Number of Cigarettes Smoked and/or Nicotine Dependence Phenotypes. NDIT Study, 1999–2008.
The authors thank the NDIT participants, Katia Desbiens and Nicholas Rudzicz for technical assistance, and the Genome Quebec and Genome Canada funded McGill University and Genome Quebec Innovation Centre for expert genotyping. Operation of the supercomputer Briarée is funded by the Canada Foundation for Innovation, NanoQuébec, RMGA and the Fonds de recherche du Québec - Nature et technologies (FRQ-NT).
Conceived and designed the experiments: JOL. Performed the experiments: JOL. Analyzed the data: AL. Contributed reagents/materials/analysis tools: JOL MHRG NL ED. Wrote the paper: JOL AL NL ED MHRG IK MPS JE. Developed the survey instruments: JOL. Participated in the interpretation of the results: JOL AL NL ED MHRG JE IK MPS. Selected the candidate genes: NL ED JOL. Conducted the Tag SNP approach: MHRG. Reviewed the article critically and approved the final version: JOL AL NL ED MHRG JE IK MPS.
- 1. Carpenter CM, Wayne GF, Connolly GN (2007) The role of sensory perception in the development and targeting of tobacco products. Addiction 102:136–147.
- 2. Benowitz NL (1988) Drug therapy. Pharmacologic aspects of cigarette smoking and nicotine addiction. N Engl J Med 319:1318–1330.
- 3. Laucht M, Becker K, Frank J, Schmidt MH, Esser G, et al. (2008) Genetic Variation in Dopamine Pathways Differentially Associated With Smoking Progression in Adolescence. J Am Acad Child Adolesc Psychiatry 47:673–681.
- 4. Shippenberg TS, Herz A, Spanagel R, Bals-Kubik R, Stein C (1992) Conditioning of opioid reinforcement: neuroanatomical and neurochemical substrates. Ann N Y AcadSci 654:347–356.
- 5. Nestler EJ (2002) From neurobiology to treatment: progress against addiction. Nat Neurosci 5 Suppl: 1076–1079.
- 6. Leonard S, Bertrand D (2001) Neuronal nicotinic receptors: from structure to function. Nicotine Tob Res 3:203–223.
- 7. Li MD, Cheng R, Ma JZ, Swan GE (2003) A meta-analysis of estimated and environmental effects on smoking behavior in male and female adult twins. Addiction 98:23–31.
- 8. Sullivan PF, Kendler KS (1999) The genetic epidemiology of smoking. Nicotine Tob Res 1 Suppl 2 S51–57, S69–70.
- 9. National Cancer Institute (2009) Phenotypes and Endophenotypes: Foundations for Genetic Studies of Nicotine Use and Dependence. Tobacco Control Monograph No. 20. Bethesda: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute. NIH Publication No.09–6366.
- 10. Caporaso N, Gu F, Chatterjee N, Sheng-Chih J, Yu K, et al. (2009) Genome-wide and candidate gene association study of cigarette smoking behaviors. PloS One 4:e4653.
- 11. Munafò M, Johnstone EC (2008) Genes and cigarette smoking. Addiction 103:893–904.
- 12. Tobacco and Genetics Consortium (2010) Genome-wide meta-analyses identify multiple loci associated with smoking behavior. Nat Genet 42:441–447.
- 13. Han S, Gelernter J, Luo X, Yang BZ (2010) Meta-analysis of 15 genome-wide linkage scans of smoking behavior. Biol Psychiatry 67:12–19.
- 14. Thorgeirsson TE, Gudbjartsson DF, Surakka I, Vink JM, Amin N, et al. (2010) Sequencevariants at CHRNB3-CHRNA6 and CYP2A6 affect smoking behavior. Nat Genet 42:448–453.
- 15. Liu JZ, Tozzi F, Waterworth DM, Pillai SG, Muglia P, et al. (2010) Meta-analysis and imputation refines the association of 15q25 with smoking quantity. Nat Genet 42:436–440.
- 16. Munafò MR, Flint J (2004) Meta-analysis of genetic association studies. Trends Genet 20:439–444.
- 17. Audrain-McGovern J, Lerman C, Wileyto EP, Rodriguez D, Shields PG (2004) Interacting effects of genetic predisposition and depression on adolescent smoking progression. Am J Psychiatry 161:1224–1230.
- 18. Huang S, Cook DG, Hinks LJ, Chen XH, Ye S, et al. (2005) CYP2A6, MAOA, DBH, DRD4, and 5HT2A genotypes, smoking behaviour and cotinine levels in 1518 UK adolescents. Pharmacogenet Genomics 15:839–850.
- 19. DiFranza JR, Savageau JA, Fletcher K, O'Loughlin J, Pbert L, et al. (2007) Symptoms of tobacco dependence after brief intermittent use. The development and assessment of nicotine dependence in youth-2 study. Arch PediatrAdolesc Med 161:704–710.
- 20. O'Loughlin J, Gervais A, Dugas E, Meshefedjian G (2009) Milestones in the Process of Cessation Among Novice Adolescent Smokers. Am J Public Health 99:499–504.
- 21. U. S. Department of Health and Human Services (1994) Preventing Tobacco Use Among Young People: A Report of the Surgeon General Public Health Service. Atlanta: Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health.
- 22. O'Loughlin J, Dugas E, Brunet J, DiFranza J, Engert J, et al.. (2014) Cohort Profile: The Nicotine Dependence in Teens (NDIT) Study. Int J Epidemiol (in press).
- 23. Centers for Disease Control and Prevention (1998) Selected cigarette smoking initiation and quitting behaviors among high school students - United States, 1997. Morb Mortal Wkly Rep 47:386–389.
- 24. Eppel A, O'Loughlin J, Paradis G, Platt R (2006) Reliability of self-reports of cigarette use in novice smokers. Addict Behav 31:1700–1704.
- 25. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO (1991) The Fagerström Test for Nicotine Dependence: a revision of the Fagerström Tolerance Questionnaire. Br J Addict 86:1119–1127.
- 26. World Health Organization (1992) International statistical classification of diseases and related health problems. 10th revision. Geneva: World HealthOrganization.
- 27. O'Loughlin J, DiFranza J, Tarasuk J, Meshefedjian G, McMillan-Davey E, et al. (2002) Assessment of nicotine dependence symptoms in adolescents: a comparison of five indicators. Tob Control 11:354–360.
- 28. Wheeler KC, Fletcher KE, Wellman RJ, DiFranza JR (2004) Screening adolescents for nicotine dependence: the Hooked On Nicotine Checklist. J AdolescHealth 35:225–230.
- 29. Rojas NL, Killen JD, Haydel KF, Robinson TN (1998) Nicotine dependence among adolescent smokers. Arch PediatrAdolesc Med 152:15–16.
- 30. O'Loughlin J, Kishchuk N, DiFranza J, Tremblay M, Paradis G (2002) The hardest thing is the habit: a qualitative investigation of adolescent smokers' experience of nicotine dependence. NicTob Res 4:20–29.
- 31. de Bakker PI, Yelensky R, Pe'er I, Gabriel SB, Daly MJ, et al. (2005) Efficiency and power in genetic association studies. Nat Genet 37:1217–1223.
- 32. Barrett JC, Fry B, Maller J, Daly MJ (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21:263–265.
- 33. The International HapMap Consortium (2007) A second generation human haplotype map of over 3.1 million SNPs. Nature 449:851–861.
- 34. Westfall PH, Young SS (1993) Resampling-based multiple testing: Examples and methods for p-value adjustment. John Wiley & Sons.
- 35. Goeman JJ, Solari A (2014) Multiple hypothesis testing in genomics. Stat Med 33:1946–1978.
- 36. Benjamini B, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 57:289–300.
- 37. R Core Team (2014) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.
- 38. Bates D, Maechler M, Bolker B and Walker S (2014). lme4: Linear mixed-effects models using Eigen and S4. R package version 1.1-7, URL: http://CRAN.R-project.org/package=lme4.
- 39. Lerman CE, Schnoll RA, Munafo MR (2007) Genetics and smoking cessation improving outcomes in smokers at risk. Am J Prev Med 336:S398–405.
- 40. Breitling LP, Dahmen N, Illig T, Rujescu D, Nitz B, et al. (2009)Variants in COMT and spontaneous smoking cessation: retrospective cohort analysis of 925 cessation events. Pharmacogenet Genomics 19:657–659.
- 41. Tenhunen J, Salminen M, Lundström K, Kiviluoto T, Savolainen R, et al. (1994) Genomic organization of the human catechol O-methyltransferase gene and its expression from two distinct promoters. Eur J Biochem 223:1049–1059.
- 42. Shield AJ, Thomae BA, Eckloff BW, Wieben ED, Weinshilboum RM (2004) Human catechol O-methyltransferase genetic variation: gene resequencing and functional characterization of variant allozymes. Mol Psychiatry 9:151–160.
- 43. Omidvar M, Stolk L, Uitterlinden AG, Hofman A, Van Duijn CM, et al. (2009) The effect of catechol-O-methyltransferase Met/Val functional polymorphism on smoking cessation: retrospective and prospective analyses in a cohort study. Pharmacogenet Genomics 19:45–51.
- 44. Tammimäki AE, Männistö PT (2010) Are genetic variants of COMT associated with addiction? PharmacogenetGenomics 20:717–741.
- 45. Ma JZ, Beuten J, Payne TJ, Dupont RT, Elston RC, et al. (2005) Haplotype analysis indicates an association between the DOPA decarboxylase (DDC) gene and nicotine dependence. Hum Mol Genet 14:1691–1698.
- 46. Yu Y, Panhuysen C, Kranzler HR, Hesselbrock V, Rounsaville B, et al. (2006) Intronic variants in the dopa decarboxylase (DDC) gene are associated with smoking behavior in European-Americans and African-Americans. Hum Mol Genet 15:2192–2199.
- 47. Singleton AB, Thomson JH, Morris CM, Court JA, Lloyd S, et al. (1998) Lack of association between the dopamine D2 receptor gene allele DRD2*A1 and cigarette smoking in a United Kingdom population. Pharmacogenetics 8:125–128.
- 48. Batra A, Gelfort G, Bartels M, Smoltczyk H, Buchkremer G, et al. (2000) The dopamine D2 receptor (DRD2) gene-a genetic risk factor in heavy smoking? Addict Biol 5:429–436.
- 49. Hiemstra M, Kleinjan M, van Schayck OC, Engels RC, Otten R (2014) Environmental smoking and smoking onset in adolescence: the role of dopamine-related genes. Findings from two longitudinal studies. PLoS One 9:e86497.
- 50. Hiemstra M, Engels RC, Barker ED, van Schayck OC, Otten R (2013) Smoking-specific parenting and smoking onset in adolescence: the role of genes from the dopaminergic system (DRD2, DRD4, DAT1 genotypes). PLoS One 8:e61673.
- 51. Li MD, Ma JZ, Beuten J (2004) Progress in searching for susceptibility loci and genes for smoking-related behaviour. Clin Genet 66:382–392.
- 52. Munafò M, Clark T, Johnstone EC, Murphy M, Walton R (2004) The genetic basis for smoking behavior: a systematic review and meta-analysis. Nicotine Tob Res 6:583–597.
- 53. Comings DE, Ferry L, Bradshaw-Robinson S, Burchette R, Chiu C, et al. (1996) The dopamine D2 receptor (DRD2) gene: a genetic risk factor in smoking. Pharmacogenetics 6:73–79.
- 54. Erblich J, Lerman C, Self DW, Diaz GA, Bovbjerg DH (2004) Stress-induced cigarette craving: effects of the DRD2 TaqI RFLP and SLC6A3 VNTR polymorphisms. Pharmacogenomics J 4:102–109.
- 55. Li D, London SJ, Liu J, Lee W, Jiang X, et al. 2011) Association of the calcyon neuron-specific vesicular protein gene (CALY) with adolescent smoking initiation in China and California. Am J Epidemiol 173:1039–1048.
- 56. Lobo DS, Zawertailo L, Selby P, Kennedy JL (2012) The role of ANKK1 and TTC12 genes on drinking behaviour in tobacco dependent subjects. World J Biol Psychiatry 13:232–238.
- 57. Huang W, Payne TJ, Ma JZ, Beuten J, Dupont RT, et al. (2009) Significant association of ANKK1 and detection of a functional polymorphism with nicotine dependence in an African-American sample. Neuropsychopharmacology 34:319–330.
- 58. Radwan GN, El-Setouhy M, Mohamed MK, Hamid MA, Azem SA, et al. (2007) DRD2/ANKK1 TaqI polymorphism and smoking behavior of Egyptian male cigarette smokers. Nicotine Tob Res 9:1325–1329.
- 59. Ducci F, Kaakinen M, Pouta A, Hartikainen AL, Veijola J, et al. (2011) TTC12-ANKK1-DRD2 and CHRNA5-CHRNA3-CHRNB4 influence different pathways leading to smoking behavior from adolescence to mid-adulthood. Biol Psychiatry 69:650–660.
- 60. Doran N, Schweizer CA, Myers MG, Greenwood TA (2013) A prospective study of the effects of the DRD2/ANKK1 TaqIA polymorphism and impulsivity on smoking initiation. Subst Use Misuse 48:106–116.
- 61. Wernicke C, Reese J, Kraschewski A, Winterer G, Rommelspacher H, et al. (2009) Distinct haplogenotypes of the dopamine D2 receptor gene are associated with non-smoking behaviour and daily cigarette consumption. Pharmacopsychiatry 42:41–50.
- 62. Voisey J, Swagell CD, Hughes IP, van Daal A, Noble EP, et al. (2012) A DRD2 and ANKK1 haplotype is associated with nicotine dependence. Psychiatry Res 196:285–289.
- 63. Gelernter J, Yu Y, Weiss R, Brady K, Panhuysen C, et al. (2006) Haplotype spanning TTC12 and ANKK1, flanked by the DRD2 and NCAM1 loci, is strongly associated to nicotine dependence in two distinct American populations. Hum Mol Genet 15:3498–3507.
- 64. Vandenbergh DJ, Bennett CJ, Grant MD, Strasser AA, O'Connor R, et al. (2002) Smoking status and the human dopamine transporter variable number of tandem repeats (VNTR) polymorphism: failure to replicate and finding that never-smokers may be different. Nicotine Tob Res 4:333–340.
- 65. Jorm AF, Henderson AS, Jacomb PA, Christensen H, Korten AE, et al. (2000) Association of smoking and personality with a polymorphism of the dopamine transporter gene: results from a community survey. Am J Med Genet 96:331–334.
- 66. Ling D, Niu T, Feng Y, Xing H, Xu X (2004) Association between polymorphism of the dopamine transporter gene and early smoking onset: an interaction risk on nicotine dependence. J Hum Genet 49:35–39.
- 67. Timberlake DS, Haberstick BC, Lessem JM, Smolen A, Ehringer M, et al. (2006) An association between the DAT1 polymorphism and smoking behavior in young adults from the National Longitudinal Study of Adolescent Health. Health Psychol 25:190–197.
- 68. Bergen AW, Conti DV, Van Den Berg D, Lee W, Liu J, et al. (2009) Dopamine genes and nicotine dependence in treatment-seeking and community smokers. Neuropsychopharmacology 34:2252–2264.
- 69. Stapleton JA, Sutherland G, O'Gara C (2007) Association between dopamine transporter genotypes and smoking cessation: a meta-analysis. Addict Biol 12:221–226.
- 70. Hiemstra M, Engels RC, Barker ED, van Schayck OC, Otten R (2013) Smoking-specific parenting and smoking onset in adolescence: the role of genes from the dopaminergic system (DRD2, DRD4, DAT1 genotypes). PLoS One 8:e61673.
- 71. Zhang L, Kendler KS, Chen X (2006) The mu-opioid receptor gene and smoking initiation and nicotine dependence. Behav Brain Funct 2:28.
- 72. Ray R, Ruparel K, Newberg A, Wileyto EP, Loughead JW, et al. (2011) Human Mu Opioid Receptor (OPRM1 A118G) polymorphism is associated with brain mu-opioid receptor binding potential in smokers. ProcNatlAcadSci USA 108:9268–9273.
- 73. Perkins KA, Lerman C, Grottenthaler A, Ciccocioppo MM, Milanak M, et al. (2008) Dopamine and opioid gene variants are associated with increased smoking reward and reinforcement owing to negative mood. BehavPharmacol 19:641–649.
- 74. Chen YT, Tsou HH, Kuo HW, Fang CP, Wang SC, et al. (2012) OPRM1 genetic polymorphisms are associated with the plasma nicotine metabolite cotinine concentration in methadone maintenance patients: a cross sectional study. J Hum Genet 58:84–90.
- 75. Pedneault M, Labbe A, Roy-Gagnon MH, Low NC, Dugas E, et al. (2014) The association between CHRN genetic variants and dizziness at first inhalation of cigarette smoke. Addict Behav 39:316–320.
- 76. Sherva R, Wilhelmsen K, Pomerleau CS, Chasse SA, Rice JP, et al. (2008) Association of a single nucleotide polymorphism in neuronal acetylcholine receptor subunit alpha 5 (CHRNA5) with smoking status and with ‘pleasurable buzz’ during early experimentation with smoking. Addiction 103:1544–1552.
- 77. Keskitalo-Vuokko K, Pitkäniemi J, Broms U, Heliövaara M, Aromaa A, et al. (2011) Associations of nicotine intake measures with CHRN genes in Finnish smokers. Nicotine Tob Res 13:686–690.
- 78. Baker TB, Weiss RB, Bolt D, von Niederhausern A, Fiore MC, et al. (2009) Human neuronal acetylcholine receptor A5-A3-B4 haplotypes are associated with multiple nicotine dependence phenotypes. Nicotine Tob Res 11:785–796.
- 79. Saccone SF, Hinrichs AL, Saccone NL, Chase GA, Konvicka K, et al. (2007) Cholinergic nicotinic receptor genes implicated in a nicotine dependence association study targeting 348 candidate genes with 3713 SNPs. Hum Mol Genet 16:36–49.
- 80. Stephens SH, Hartz SM, Hoft NR, Saccone NL, Corley RC, et al. (2013)Distinct loci in the CHRNA5/CHRNA3/CHRNB4 gene cluster are associated with onset of regular smoking. Genet Epidemiol 37:846–859.
- 81. Saccone NL, Schwantes-An TH, Wang JC, Grucza RA, Breslau N, et al. (2010) Multiple cholinergic nicotinic receptor genes affect nicotine dependence risk in African and European Americans. Genes Brain Behav 9:741–750.
- 82. Wong SL, Shields M, Leatherdale S, Malaison E, Hammond D (2012) Assessment of validity of self-reported smoking status. HealthRep 23:47–53.