ORIGINAL CONTRIBUTION Genetic Analysis of a Population Heavy Drinking Phenotype Identifies Risk Variants in Whites Ajna Hamidovic, PharmD, MS,* Robert J. Goodloe, MS,Þ Taylor R. Young, MA,þ Mindi A. Styn, PhD,§ Kenneth J. Mukamal, MD,|| Helene Choquet, PhD,¶ Jay L. Kasberger, BS,¶ Sarah G. Buxbaum, PhD,# George J. Papanicolaou, PhD,** Wendy White, PhD,ÞÞ Kelly Volcik, PhD,þþ Bonnie Spring, PhD,* Brian Hitsman, PhD,* Daniel Levy, MD,§§ and Eric Jorgenson, PhDÞ Abstract: Genetic association studies thus far have used detailed diagnoses of alcoholism to identify loci associated with risk. This proof-of-concept analysis examined whether population data of lifetime heaviest alcohol consumption may be used to identify genetic loci that modulate risk. We conducted a genetic association study in European Americans between variants in approximately 2100 genes and alcohol consumption as part of the Candidate gene Association Resource project. We defined cases as individuals with a history of drinking 5 or more drinks per day almost every day of the week and controls as current light drinkers (1Y5 drinks per week). We cross-validated identified single nucleotide polymorphisms in a metaanalysis of 2 cohorts of unrelated individualsVAtherosclerosis Risk in Communities (ARIC) and Cardiovascular Health Study (CHS)Vand in a separate cohort of related individuals VFramingham Heart Study (FHS). The most significant variant in the meta-analysis of ARIC and CHS was rs6933598 in methylenetetrahydrofolate dehydrogenase (P = 7.46  10j05) with a P value in FHS of 0.042. The top variants in FHS were rs12249562 in cubulin (P = 3.03  10j05) and rs9839267 near cholecystokinin (P = 3.05  10j05) with a P value of 0.019 for rs9839267 in CHS. We have here shown feasibility in evaluating lifetime incidence of heavy alcohol drinking from population-based studies for the purpose of conducting genetic association analyses. Key Words: alcohol consumption, alcoholism, cholecystokinin, methylenetetrahydrofolate dehydrogenase (J Clin Psychopharmacol 2013;33: 206Y210) A lcohol dependence is a complex trait that underlies a range of physiologic and behavioral symptoms manifested as tolerance, loss of control, withdrawal, and desire or inability to From *Preventive Medicine, Northwestern University, Chicago, IL; †Center for Human Genetics, Vanderbilt University Medical Center, Nashville, TN; ‡Center for Health Sciences, SRI International, Menlo Park, CA; §Department of Health and Community Systems, School of Nursing, University of Pittsburgh, Pittsburgh, PA; ||Department of Medicine, Beth Israel Deaconess Medical Center, Boston, MA; ¶Ernest Gallo Clinic and Research Center, University of California, San Francisco, CA; #Jackson Heart Study, Jackson State University, Jackson, MS; **Division of Cardiovascular Sciences, Prevention and Population Sciences Program, National Heart, Lung, and Blood Institute, Bethesda, MD; ††Natural Science Division, Tougaloo College, Tougaloo, MS; ‡‡Human Genetics Center, School of Public Health, University of Texas Health Science Center, Houston, TX; and §§Framingham Heart Study, National Heart, Lung, and Blood Institute, Framingham, MA. Received June 10, 2011; accepted after revision July 20, 2012. Reprints: Ajna Hamidovic, PharmD, Preventive Medicine, Northwestern University, 680 N. Lakeshore Drive, Suite 1400, Chicago, IL 60611-4402 (e-mail: a-hamidovic@northwestern.edu; ajna.hamidovic@gmail.com). Copyright * 2013 by Lippincott Williams & Wilkins ISSN: 0271-0749 DOI: 10.1097/JCP.0b013e318287009a 206 www.psychopharmacology.com cut down. Considerable evidence from twin genetic studies1Y11 indicates that the trait is heritable with estimates ranging from 40% to 60% of the risk of alcohol dependence due to genetic factors. A measure of alcohol dependence is alcohol consumption.12Y18 Alcohol drinking, as alcohol dependence, has a complex, nonMendelian pattern of inheritance, indicating involvement of multiple genetic variants.19 Because alcohol use can influence numerous health outcomes, population studies with broad aims typically collect alcohol consumption data, and these data have the potential be investigated in genetic association studies. Grant et al20 (2009) and Kendler et al21 (2010) found that it is feasible to closely index the genetic risk for alcohol dependence by collecting relatively simple quantitative data on the alcohol consumption. Fitting a model that included 5 measures of alcohol consumption [lifetime events of heaviest alcohol use including maximum drinks consumed in 24 hours, lifetime maximal tolerance, typical number of drinks per occasion (lifetime), frequency consumed alcohol (heaviest period) and frequency of drinking to intoxication (heaviest period)], Grant et al20 found a high genetic correlation with alcohol dependence symptom scores (9+0.97). Evaluating the extent to which each individual measure reflected the genetic risk factor by sex, Kendler et al21 (2010) found that in men, maximum drinks consumed in a 24-hour period had the highest loading, followed closely by frequency of drinking to intoxication, whereas in women, frequency of drinking to intoxication loaded most strongly. These studies show that, in theory, a lifetime event of heavy alcohol use indexes the genetic risk for alcohol dependence; however, no study thus far reflects this concept in practice. Therefore, we used an alcohol consumption phenotype of lifetime history of intake of 5 or more drinks per day almost every day of the week that was collected in cardiovascular cohort studies from the Candidate gene Association Resource (CARe) project. We conducted a genetic association analysis with variants on a genotyping platform that densely covers approximately 2100 genes. This approach has the advantage of using dense genotyping coverage of a large number of genes without prespecifying biological hypotheses about the effect of individual genes and genetic variants. METHODS We analyzed alcohol consumption from the National Heart, Lung, and Blood Institute (NHLBI)Ysponsored CARe project.22 The CARe Project was launched in 2007 to create a resource for association studies of various phenotypes. The CARe project consists of 9 NHLBI cohorts. It is approved by the ethics committees of the participating studies and by the Massachusetts Institute of Technology. Subjects Our phenotype of interest was available in 3 white cohorts from the CARe project: Atherosclerosis Risk in Communities Journal of Clinical Psychopharmacology & Volume 33, Number 2, April 2013 Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Journal of Clinical Psychopharmacology & Volume 33, Number 2, April 2013 (ARIC, 1989), Framingham Heart Study (FHS),23Y25 and Cardiovascular Health Study (CHS).26 Our sample of subjects in ARIC included 2138 (632 cases and 1506 controls) unrelated individuals with a mean (SD) age of 59.8 (5.6) years, of which 40% were female. The CHS cohort also included unrelated individuals (n = 859; 358 cases and 501 controls) with a mean (SD) age of 72.2 (5.2) years, of which 36.3% were female. Subjects in FHS (offspring cohort) included 772 related individuals (265 cases and 507 controls) of which 49% were female with the mean age (SD) 65.1 (8.9) years for the total sample of analyzed individuals. Phenotype This analysis was a case-control comparison between light and heavy drinkers. We defined cases as individuals with a lifetime history of drinking 5 or more drinks per day almost every day of the week. We defined controls as current light drinkers of 1 to 5 drinks per week to ensure comparison with other drinkers. Among light drinkers, we excluded individuals who may be binge drinkers (Q4 drinks per occasion for women and Q5 drinks per occasion for men).27 Genotyping Assay The content of the genotyping array, ITMAT-Broad-CARe or ‘‘IBC chip,’’ is informed by genome-wide association studies (GWAS), expression quantitative trait loci, pathway-based approaches and comprehensive literature searching. It includes loci relevant to alcoholism, such as gamma-aminobutryic acid and alcohol metabolism genes. As an example, it contains densely spaced single nucleotide polymorphisms from 84 of the 130 genes from the ‘‘addiction array’’28 and additional genes that are not on the addiction array but were found to be associated with alcoholism in later genetic association studies. The loci on the IBC chip are divided into 3 groups: group 1 (n = 435 loci)Vgenes and regions with a high likelihood of functional significance [tag SNPs selected to capture known variation with minor allele frequency (MAF) 90.02 and an r2 of at least 0.8 in HapMap populations]; group 2 (n = 1349 loci)Vcandidate loci that are potentially involved in phenotypes of interest or established loci that required very large numbers of tagging SNPs (tag SNPs selected to capture known variation with MAF 90.05 with an r2 of at least 0.5 in HapMap populations); group 3 (n = 232 loci)Vcomposed mainly of the larger genes (100 kb), which were of lower interest a priori to the investigators (includes only nonsynonomous SNPs and known functional variants). The average number of SNPs across the group 1 and group 2 loci of IBC was compared with GWAS products. The average coverage for group 1 loci is approximately 36.5 SNPs per locus on the IBC chip. The Illumina Human1M (San Diego, CA) and Affymetrix 6.0 (Santa Clara, CA) platform, for comparison, have an average of approximately 28.0 and approximately 17.4 SNPs, respectively, across the equivalent IBC loci. The average number of SNPs observed for the group 2 loci is approximately 16.3 SNPs, which is comparable with the current GWAS products. Additional details regarding the design of the IBC chip have been described in Keating et al.29 In toto, 49,320 SNPs were chosen to map approximately 2100 candidate gene loci. For detailed genotyping and quality control information, see Musunuru et al.22 Statistical Analysis For cohorts including unrelated individuals (ARIC and CHS), we used logistic regression to test SNP-phenotype associations. Association analysis was performed in PLINK30 under an * 2013 Lippincott Williams & Wilkins Population Alcohol Consumption Analysis additive genetic model. Association results were combined across the 2 cohorts using an inverse variance meta-analysis approach as implemented in METAL.31 For the FHS cohort for which there were significant numbers of related individuals, we used genome-wide association analyses with family,32 which implements generalized estimating equations in the gee package (http://cran.r-project.org/web/packages/gee/) to test association between the light versus heavy phenotype and each SNP under the additive genetic model. To address population stratification, we conducted principal component analysis as implemented in EIGENSTRAT.33 The first 10 principal components were included as covariates in the genetic association analysis. In addition, age and sex were included as covariates in the association analysis. All results were adjusted for residual inflation using the genomic control method. Bonferroni adjustment for multiple comparisons was set at an > level of 2.3  10j06. Replication P significance value was set at less than 0.05. Imputation of ungenotyped variants was done using a combined CEU + YRI reference panel including SNPs segregating in both CEU and YRI and SNPs segregating in one panel and monomorphic and nonmissing in the other, resulting in approximately 270,000 total SNPs. The use of the CEU + YRI panel resulted in an allelic concordance rate of approximately 95.6%, calculated as 1 j 2  |imputed dosage Y chip dosage|. This rate is comparable with rates calculated for individuals of African descent imputed with the HapMap 2 YRI individuals.34 In the first step of imputation, individuals with pedigree relatedness or cryptic relatedness (pi_hat 9 0.05) were filtered out. Recombination and error rate estimates for the entire sample were calculated based on a subset of random individuals. Next, these rates were used to impute all sample individuals across the entire reference panel. SNPs with low imputation scores (r2 G 0.3) and MAF of less than 0.01 were filtered out. RESULTS The locus that was the most strongly associated with the lifetime heavy drinking phenotype in ARIC and CHS is presented in Figure 1. Each additional copy of the major rs6933598*C allele (frequency: HapMap CEU = 0.715) was associated with a decrease in risk of heavy drinking [results of combined metaanalysis: odds ratio (OR), 0.75; confidence interval (CI), 0.66Y0.86; P = 7.46  10j05; results for individual studies: ARIC OR, 0.76; CI, 0.64Y0.91; P = 0.00155; CHS OR, 0.75; CI, 0.59Y0.95; P = 0.017]. This locus replicated in FHS OR of 0.75 (CI, 0.57Y0.98), P = 0.042. In FHS, the strongest associations of similar significance were located on separate chromosomes. The first SNP, rs12249562, is located on chromosome 10 in cubulin, where each additional copy of minor rs12249562*A allele (HapMap CEU = 0.195) resulted in a decrease of risk for lifetime heavy drinking OR of 0.52 (CI, 0.38Y0.71), P = 3.03  10j05. This association was closely followed by an association on chromosome 3 near cholecystokinin (CCK) where each additional copy of the minor rs9839267*G allele (HapMap CEU = 0.085) resulted in an increase of risk OR of 2.39 (CI, 1.58Y3.60), P = 3.05  10j05. This SNP replicated in CHS at a P value of less than 0.05 [OR, 1.57 (CI, 1.08Y2.27); P = 0.019; see Fig. 2 for a graphic representation of this locus on our discovery cohort (FHS) and the replication cohort (CHS)]. The P value for rs9839267 in ARIC was not less than 0.05. DISCUSSION We examined a lifetime heavy alcohol drinking incidence phenotype for association with genetic variants from a large number of candidate genes in 3 cohorts from the CARe project. www.psychopharmacology.com Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 207 Journal of Clinical Psychopharmacology Hamidovic et al & Volume 33, Number 2, April 2013 FIGURE 1. Association between the MTHFD1L locus and lifetime incidence of heavy alcohol drinking in the CARe project cohorts. Left panel: Our discovery sample included unrelated individuals from ARIC and CHS. Right panel: Our top SNP rs6933598 (purple diamond) from meta-analysis of ARIC and CHS was also associated with the phenotype in the Framingham Heart Study (FHS; P = 0.042). We were able to identify variants in CCK and methylenetetrahydrofolate dehydrogenase (MTHFD1L) that modulate risk for heavy alcohol drinking. These results demonstrate the feasibility in evaluating lifetime incidence of heavy alcohol drinking from population-based studies for the purpose of conducting genetic association analyses. Cholecystokinin was originally found in the gut where it is involved in the secretion of pancreatic enzymes, gall bladder, and gut motility. However, it is distributed even more widely in the brain where it is one of the most abundant neuropeptides.35,36 Colocalization of CCK on cell bodies and terminals of classic neurotrasmitters implicated in alcohol abuse potential including gammaaminobutyric acid,37,38 serotinin,39 and opiates40 makes the neuropeptide a biologically plausible candidate. However, the role of CCK in regulation of dopamine turnover in the mesoaccumbal projectionVa region highly implicated in the primary effects of drugs of abuse and in the process of sensitizationVhas received the most attention in the relationship of CCK and addiction (see Rotzinger et al, 200341). Indeed, dopaminergic transmission was thought to be implicated in the behavioral finding that antagonism of CCK significantly reduced the intake of ethanol in naive adult male Wistar rats.42 Very early candidate gene association studies provided initial results regarding the role of CCK in alcoholism. Two previous studies showed that the promoter SNP rs1799923 modulates clinically diagnosed alcohol dependence in the Japanese population,43,44 but an attempt to show the same in the white population was not successful.45 This particular SNP was genotyped in CARe and was in fact associated with our phenotype of interest [P = 3.08  10j05 (FHS), P = 0.0196 (CHS); see Fig. 2 showing 8 SNPs (in red) in high linkage disequilibrium with our top SNP rs9839267, 7 of which were imputed based on the genotyped SNP rs1799923]. Our result in the CCK locus allows the following conclusions to be made. First, it demonstrates the significance of CCK beyond the previously examined single locus analysis when the genome is evaluated more extensively. In addition, it opposes the earlier negative finding in whites and shows the importance of the locus in a population ethnically different from the previously identified Japanese group. Folate metabolism is complex, and the mechanism(s) by which alcohol inhibits it have not been definitively established. Homocysteine is converted to methionine via tetrahydrofolate (THF). MTHFD1L is involved in THF synthesis by catalyzing the reversible synthesis of 10-formyl-THF to formate and THF. Elevated homocysteine levels were first reported by Hultberg et al46 (1993) in patients hospitalized for detoxification after severe alcohol abuse. This finding has since been independently replicated.47,48 Plasma homocysteine levels predict alcohol withdrawal.47 The association between alcohol intake and raised FIGURE 2. Association between the CCK locus and lifetime incidence of heavy alcohol consumption in the CARe project cohorts. Left panel: Our discovery sample included related individuals from FHS. Right panel: Our top SNP rs9839267 (purple diamond) in FHS was also associated with the phenotype in CHS (P = 0.019). 208 www.psychopharmacology.com * 2013 Lippincott Williams & Wilkins Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Journal of Clinical Psychopharmacology & Volume 33, Number 2, April 2013 plasma homocysteine levels, in fact, has also been observed in moderate alcohol consumers, with plasma homocysteine levels increasing over a 6-week drinking period.49 The MTHFDL1 gene resides on 6q25.1 and spans approximately 235 kb. As shown in Figure 1, the associated locus is in very low LD with the remaining variants in the region, permitting localization of the region of functional significance. There are no published reports of rs6933598 in genetic association analyses. However, a proxy of rs6933598 and rs6922269 (r2 = 0.924 between the 2 variants in CEU) was associated with coronary heart disease in The Wellcome Trust Case Control Consortium50 and German MI Family Study.51 The same narrow region of a cluster of SNPs in high LD was also recently associated with late-onset Alzheimer disease.52 Because alcohol consumption is implicated in a variety of chronic conditions, population studies most commonly collect data to derive the number of grams of alcohol a person consumes per week (or per day). However, this phenotype, usually collected as part of a dietary survey, is a cross-sectional snapshot of alcohol consumption, and the data tend to have an extreme violation of normality in distribution. The same problems are encountered with the ‘‘Max Drinks’’ phenotype sometimes collected in population studies (the maximum number of drinks in a 24-hour period in the last 30 days). Our results show that data that summarize lifetime heaviest use of alcohol in cases may be used successfully in future population studies attempting to identify variants implicated in alcoholism risk. This is a retrospective analysis based on self-report, which we could not objectively verify. Because comorbidities are generally not included in published genetic association studies of alcoholism, they were not incorporated here either. We are guided to believe that our approach is valid because we have shown that the same locus in CCK, previously associated with clinically diagnosed alcohol dependence in a single locus analysis of the Japanese population, also mediates heavy alcohol consumption in the ethnically different white population. Here, though, we have demonstrated the importance of this gene when the genome is evaluated more extensively, thereby aligning results of previous pharmacologic studies that stressed the importance of CCK in alcohol consumption. Our primary goal in this analysis was to show that, in genetic association studies, simple measurements of lifetime heaviest use of alcohol may serve in place of more laborious assessment of alcohol dependence, especially in ongoing cohort studies in which diagnostic batteries are, at best, impractical. Provided that cases and controls are defined carefully, this approach is convenient because it allows for collection of sample sizes that may be difficult to obtain with clinical assessment of alcohol dependence. Future studies should evaluate how the variants discussed here modify expression/structure to provide information regarding their influence on the final phenotype. AUTHOR DISCLOSURE INFORMATION All sources of support: MD Scientist Fellowship in Genetic Medicine (Northerstern Memorial Foundation; PI: A. Hamidovic), National Research Service Award F32DA024920 (NIH/NIDA; PI: A. Hamidovic), Dr. Bonnie Spring’s Professional Account at Northwestern Feinberg School of Medicine, NIAAA R21 AA021223-01, E. Jorgenson). The Candidate gene Association Resource (CARe) wishes to acknowledge the support of the National Heart, Lung and Blood Institute and the contributions of the research institutions, study investigators, field staff and study participants in creating this resource for biomedical research (NHLBI contract number HHSN268200960009C). The following eight parent studies have contributed parent study data, ancillary * 2013 Lippincott Williams & Wilkins Population Alcohol Consumption Analysis study data, and DNA samples through the Broad Institute (N01HC-65226) to create this genotype/phenotype database for wide dissemination to the biomedical research community: the Atherosclerosis Risk in Communities (ARIC) study, the CHS, the Cleveland Family Study (CFS), the Coronary Artery Risk Development in Young Adults (CARDIA) study, the Framingham Heart Study, the Jackson Heart Study (JHS), the Multi-Ethnic Study of Atherosclerosis (MESA), and the Sleep Heart Health Study (SHHS). Individual study funding attributions can be obtained at http:// public.nhlbi.nih.gov/GeneticsGenomics/home/care.aspx. The authors declare no conflicts of interest. REFERENCES 1. Heath AC, Bucholz KK, Madden PA, et al. Genetic and environmental contributions to alcohol dependence risk in a national twin sample: consistency of findings in women and men. Psychol Med. 1997;27(6):1381Y1396. 2. Cadoret RJ, O’Gorman TW, Troughton E, et al. Alcoholism and antisocial personality. Interrelationships, genetic and environmental factors. Arch Gen Psychiatry. 1985;42(2):161Y167. 3. Cadoret RJ, Troughton E, O’Gorman TW. Genetic and environmental factors in alcohol abuse and antisocial personality. J Stud Alcohol. 1987;48(1):1Y8. 4. Cloninger CR, Bohman M, Sigvardsson S. Inheritance of alcohol abuse. Cross-fostering analysis of adopted men. Arch Gen Psychiatry. 1981;38(8):861Y868. 5. Goodwin DW, Schulsinger F, Hermansen L, et al. Alcohol problems in adoptees raised apart from alcoholic biological parents. Arch Gen Psychiatry. 1973;28(2):238Y243. 6. Hrubec Z, Omenn GS. Evidence of genetic predisposition to alcoholic cirrhosis and psychosis: twin concordances for alcoholism and its biological end points by zygosity among male veterans. Alcohol Clin Exp Res. 1981;5(2):207Y215. 7. Kendler KS, Heath AC, Neale MC, et al. A population-based twin study of alcoholism in women. JAMA. 1992;268(14):1877Y1882. 8. Kendler KS, Prescott CA, Neale MC, et al. Temperance board registration for alcohol abuse in a national sample of Swedish male twins, born 1902 to 1949. Arch Gen Psychiatry. 1997;54(2):178Y184. 9. McGue M, Pickens RW, Svikis DS. Sex and age effects on the inheritance of alcohol problems: a twin study. J Abnorm Psychol. 1992;101(1):3Y17. 10. Pickens RW, Svikis DS, McGue M, et al. Heterogeneity in the inheritance of alcoholism. A study of male and female twins. Arch Gen Psychiatry. 1991;48(1):19Y28. 11. Prescott CA, Kendler KS. Genetic and environmental contributions to alcohol abuse and dependence in a population-based sample of male twins. Am J Psychiatry. 1999;156(1):34Y40. 12. Agrawal A, Grant JD, Littlefield A, et al. Developing a quantitative measure of alcohol consumption for genomic studies on prospective cohorts. J Stud Alcohol Drugs. 2009;70(2):157Y168. 13. Hansell NK, Agrawal A, Whitfield JB, et al. Long-term stability and heritability of telephone interview measures of alcohol consumption and dependence. Twin Res Hum Genet. 2008;11(3):287Y305. 14. Heath AC, Meyer J, Jardine R, et al. The inheritance of alcohol consumption patterns in a general population twin sample: II. Determinants of consumption frequency and quantity consumed. J Stud Alcohol. 1991;52(5):425Y433. 15. Hettema JM, Corey LA, Kendler KS. A multivariate genetic analysis of the use of tobacco, alcohol, and caffeine in a population based sample of male and female twins. Drug Alcohol Depend. 1999;57(1):69Y78. www.psychopharmacology.com Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 209 Journal of Clinical Psychopharmacology Hamidovic et al & Volume 33, Number 2, April 2013 16. Kaprio J, Rose RJ, Romanov K, et al. Genetic and environmental determinants of use and abuse of alcohol: the Finnish Twin Cohort studies. Alcohol Alcohol Suppl. 1991;1:131Y136. 35. Crawley JN. Comparative distribution of cholecystokinin and other neuropeptides. Why is this peptide different from all other peptides? Ann N Y Acad Sci. 1985;448:1Y8. 17. Prescott CA, Hewitt JK, Truett KR, et al. Genetic and environmental influences on lifetime alcohol-related problems in a volunteer sample of older twins. J Stud Alcohol. 1994;55(2):184Y202. 36. Moran TH, Schwartz GJ. Neurobiology of cholecystokinin. Crit Rev Neurobiol. 1994;9(1):1Y28. 18. Whitfield JB, Zhu G, Madden PA, et al. The genetics of alcohol intake and of alcohol dependence. Alcohol Clin Exp Res. 2004;28(8):1153Y1160. 37. Alho H, Ferrarese C, Vicini S, et al. Subsets of GABA-ergic neurons in dissociated cell cultures of neonatal rat cerebral cortex show co-localization with specific modulator peptides. Brain Res. 1988;467:193Y204. 19. Goldman D, Oroszi G, O’Malley S, et al. COMBINE genetics study: the pharmacogenetics of alcoholism treatment response: genes and mechanisms. J Stud Alcohol. 2005;66(suppl 15):56Y64; discussion 33. 38. Hendry SH, Jones EG, DeFelipe J, et al. Neuropeptide-containing neurons of the cerebral cortex are also GABAergic. Proc Natl Acad Sci U S A. 1984;81(20):6526Y6530. 20. Grant JD, Agrawal A, Bucholz KK, et al. Alcohol consumption indices of genetic risk for alcohol dependence. Biol Psychiatry. 2009;66(8):795Y800. 39. van der Kooy D, Hunt SP, Steinbusch HW, et al. Separate populations of cholecystokinin and 5-hydroxytryptamine-containing neuronal cells in the rat dorsal raphe, and their contribution to the ascending raphe projections. Neurosci Lett. 1981;26(1):25Y30. 21. Kendler KS, Myers J, Dick D, et al. The relationship between genetic influences on alcohol dependence and on patterns of alcohol consumption. Alcohol Clin Exp Res. 2010;34(6):1058Y1065. 22. Musunuru K, Lettre G, Young T, et al. Candidate gene association resource (CARe): design, methods, and proof of concept. Circ Cardiovasc Genet. 2010;3(3):267Y275. 23. Dawber TR, Meadors GF, Moore FE Jr. Epidemiological approaches to heart disease: the Framingham Study. Am J Public Health Nations Health. 1951;41(3):279Y281. 24. Feinleib M, Kannel WB, Garrison RJ, et al. The Framingham Offspring Study. Design and preliminary data. Prev Med. 1975;4(4):518Y525. 25. Splansky GL, Corey D, Yang Q, et al. The third generation cohort of the National Heart, Lung, and Blood Institute’s Framingham Heart Study: design, recruitment, and initial examination. Am J Epidemiol. 2007;165(11):1328Y1335. 26. Fried LP, Borhani NO, Enright P, et al. The Cardiovascular Health Study: design and rationale. Ann Epidemiol. 1991;1(3):263Y276. 40. Gall C, Lauterborn J, Burks D, et al. Co-localization of enkephalin and cholecystokinin in discrete areas of rat brain. Brain Res. 1987;403(2):403Y408. 41. Rotzinger S, Vaccarino FJ. Cholecystokinin receptor subtypes: role in the modulation of anxiety-related and reward-related behaviours in animal models. J Psychiatry Neurosci. 2003;28(3):171Y181. 42. Crespi F. The role of cholecystokinin (CCK), CCK-A or CCK-B receptor antagonists in the spontaneous preference for drugs of abuse (alcohol or cocaine) in naive rats. Methods Find Exp Clin Pharmacol. 1998;20(8):679Y697. 43. Harada S, Okubo T, Tsutsumi M, et al. A new genetic variant in the Sp1 binding cis-element of cholecystokinin gene promoter region and relationship to alcoholism. Alcohol Clin Exp Res. 1998;22(suppl 3):93SY96S. 44. Ishiguro H, Saito T, Akazawa S, et al. Association between drinking-related antisocial behavior and a polymorphism in the serotonin transporter gene in a Japanese population. Alcohol Clin Exp Res. 1999;23(7):1281Y1284. 27. Centers for Disease Control and Prevention. Alcohol and Public Health Fact Sheets: Binge Drinking. 2010; Available at: http://www.cdc.gov/ alcohol/fact-sheets/binge-drinking.htm. Accessed April 26, 2011. 45. Ishiguro H, Saito T, Shibuya H, et al. No association between C-45T polymorphism in the Sp1 binding site of the promoter region of the cholecystokinin gene and alcoholism. Psychiatry Res. 1999;85(2):209Y213. 28. Hodgkinson CA, Yuan Q, Xu K, et al. Addictions biology: haplotype-based analysis for 130 candidate genes on a single array. Alcohol Alcohol. 2008;43(5):505Y515. 46. Hultberg B, Berglund M, Andersson A, et al. Elevated plasma homocysteine in alcoholics. Alcohol Clin Exp Res. 1993;17(3):687Y689. 29. Keating BJ, Tischfield S, Murray SS, et al. Concept, design and implementation of a cardiovascular gene-centric 50 k SNP array for large-scale genomic association studies. PLoS One. 2008;3(10):e3583. 47. Bleich S, Degner D, Bandelow B, et al. Plasma homocysteine is a predictor of alcohol withdrawal seizures. Neuroreport. 2000;11(12):2749Y2752. 30. Purcell S, Neale B, Todd-Brown K, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559Y575. 48. Cravo ML, Gloria LM, Selhub J, et al. Hyperhomocysteinemia in chronic alcoholism: correlation with folate, vitamin B-12, and vitamin B-6 status. Am J Clin Nutr. 1996;63(2):220Y224. 31. Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010;26(17):2190Y2191. 49. Bleich S, Bleich K, Kropp S, et al. Moderate alcohol consumption in social drinkers raises plasma homocysteine levels: a contradiction to the ’French Paradox’? Alcohol Alcohol. 2001;36(3):189Y192. 32. Chen M-H, Larson MG, Hsu Y-H, et al. A three-stage approach for genome-wide association studies with family data for quantitative traits. BMC Genetics. 2010;11:40. 50. Wellcome Trust Case Control C. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature. 2007;447(7145):661Y678. 33. Price AL, Patterson NJ, Plenge RM, et al. Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006;38(8):904Y909. 51. Samani NJ, Erdmann J, Hall AS, et al. Genomewide association analysis of coronary artery disease. N Engl J Med. 2007;357(5):443Y453. 34. Huang L, Li Y, Singleton AB, et al. Genotype-imputation accuracy across worldwide human populations. Am J Hum Genet. 2009;84(2):235Y250. 210 www.psychopharmacology.com 52. Naj AC, Beecham GW, Martin ER, et al. Dementia revealed: novel chromosome 6 locus for late-onset Alzheimer disease provides genetic evidence for folate-pathway abnormalities. PLoS Genet. 2010;6(9):e1001130. * 2013 Lippincott Williams & Wilkins Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.