Alcoholism is a relatively common, chronic, disabling and often treatment-resistant disorder. Evidence from twin and adoption studies indicates a substantial genetic influence, with heritability estimates of 50–60%. We conducted a genome scan in the Irish Affected Sib Pair Study of Alcohol Dependence (IASPSAD). Most probands were ascertained through alcoholism treatment settings and were severely affected. Probands, affected siblings and parents were evaluated by structured interview. A 4 cM genome scan was conducted using 474 families of which most (96%) were comprised by affected sib pairs. Nonparametric and quantitative linkage analyses were conducted using DSM-IV alcohol dependence (AD) and number of DSM-IV AD symptoms (ADSX). Quantitative results indicate strong linkage for number of AD criteria to a broad region of chromosome 4, ranging from 4q22 to 4q32 (peak multipoint LOD=4.59, P=2.1 × 10−6, at D4S1611). Follow-up analyses suggest that the linkage may be due to variation in the symptoms of tolerance and out of control drinking. There was evidence of weak linkage (LODs of 1.0–2.0) to several other regions, including 1q44, 13q31, and 22q11 for AD along with 2q37, 9q21, 9q34 and 18p11 for ADSX. The location of the chromosome 4 peak is consistent with results from prior linkage studies and includes the alcohol dehydrogenase gene cluster. The results of this study suggest the importance of genetic variation in chromosome 4 in the etiology and severity of alcoholism in Caucasian populations.
Alcoholism is a chronic, disabling and often treatment-resistant disorder with an estimated lifetime prevalence in the US of 19% in men and 8% in women.1 Family, twin and adoption studies provide substantial evidence that genetic factors play a strong role in the etiology of alcoholism. Although early studies concluded that genetic factors were not as important an influence on vulnerability to alcoholism in women as in men, more recent findings suggest genetic factors account for 50–60% of the variance in both sexes (reviewed in Prescott 2).
Several linkage studies of alcoholism or alcohol-related phenotypes have been conducted or are in progress. Of the four published linkage studies of alcoholism, two were based on multiplex families recruited from two American Indian tribes,3, 4 and the others included multiplex families collected throughout the US5, 6 and in the Pittsburgh area.7 Linkage analyses of drinking-related phenotypes have also been conducted using the Framingham Heart Study sample8 and a sample of sib pairs ascertained in San Diego.9 Overall, the regions implicated by these studies do not show a great deal of consistency, and the results are modest (e.g., LOD scores <3.0). Possible contributing factors to this lack of consistency and the small effect sizes are low power as well as genetic and cultural heterogeneity among the US population samples and between the general population and Native American samples.
Detection of susceptibility genes is further complicated because alcoholism probably reflects a clinically and etiologically heterogeneous set of disorders.10 Diagnostic classifications distinguish between alcohol dependence (AD) and alcohol abuse. Under the current and widely used criteria from the American Psychiatric Association,11 symptoms of AD include direct indices of physiological dependence (withdrawal, tolerance) as well as behavioral indices of addiction (lack of control over amount consumed, binge drinking, inability to quit drinking, continued use despite serious medical or psychiatric consequences, and drinking to the exclusion of other activities). Individuals must manifest at least three of these seven criteria on multiple occasions to be classified as having AD. In contrast, alcohol abuse is defined as continued drinking despite experiencing one or more negative psychosocial consequences (difficulties at work, difficulties with relationships or legal problems).
Although broad definitions of alcoholism including alcohol abuse appear to have a genetic basis (e.g.,12), AD is closer to a core definition of physiological dependence and is less influenced by cultural or situational factors than are abuse symptoms. We therefore selected AD as the preferred phenotype for detection of susceptibility loci. We also use AD symptom count as an index of disorder severity. This was based on three considerations: the greater statistical power available from a semicontinuous measure versus a binary diagnosis; heritabilities from twin studies indicate that the heritability of symptom count is comparable to that of AD12; and other studies have reported linkage to symptom counts (e.g.,5).
We conducted a linkage study based on sib pairs affected with AD who were recruited in Ireland (we use Ireland and Irish to refer to both Ireland (sometimes called the Republic of Ireland) and Northern Ireland.). Although a sibling sample has less power (per person studied) than multigenerational pedigrees, siblings are better matched for cohort effects and other environmental factors that may influence liability to AD. Sib pair samples may also be more representative of the population than cases from multiplex pedigrees (due to the infrequency of high density pedigrees). We did not include unaffected siblings because of the difficulty of defining ‘unaffected’ in the case of AD (e.g., ambiguities in the classification of lifetime abstainers and heavy drinkers who acknowledge no symptoms). Furthermore, many siblings were still in their 20's and 30's, well within the risk period for developing AD.
We conducted this study in Ireland because it is more culturally and genetically homogeneous than the US. Cultural factors strongly influence ethanol consumption patterns13 and there is less variation in drinking patterns and attitudes towards alcohol in Ireland than in the US.14, 15 Alcohol consumption in Ireland, particularly of distilled beverages, has increased during the past 15 years and is now among the highest of Western countries. Between 1998 and 2002 (the period covered by the data collection) the average annual ethanol consumption in the Irish Republic was 14 l of pure ethanol per capita among the population aged over 15 years.16
In addition, although Ireland is not a genetic isolate, the degree of genetic variation in Ireland is substantially less than in a general US sample.17 The current Irish population derives largely from a series of prehistoric Celtic invasions, supplemented in the last 400 years by English and Scottish settlement.18, 19
Materials and methods
Clinical data collection
Data collection occurred between 1998 and 2002 and was a joint collaboration among Virginia Commonwealth University, the Health Research Board in Dublin and Shaftsbury Square Hospital in Belfast. Ascertainment of probands was conducted in community alcoholism treatment facilities and public and private hospitals in the Republic of Ireland and Northern Ireland. A minority of participants were recruited through community notices, media advertisements and word-of-mouth. Probands were eligible for study inclusion if they met the current American Psychiatric Association11 criteria for AD, and if all four grandparents had been born in Ireland, Northern Ireland, Scotland, Wales or England. Individuals with other substance use and psychiatric disorders were not excluded but we did assess the chronological relationship between the onsets of these disorders and AD. The sample includes 591 probands who met all eligibility criteria.
The sample includes 610 affected siblings, for whom the inclusion criteria were the same as for probands. We attempted to enroll all living, biological parents for whom the probands provided permission to contact. The final sample includes 82 fathers and 131 mothers, of whom 43 fathers (52%) and 21 mothers (16%) met criteria for AD. The sample includes 10 extended families, which included affected grandparents or uncles. Details of clinical characteristics and evidence of diagnostic validity are available in Prescott et al.20
Probands, siblings and parents were interviewed by clinically trained research interviewers (often with extensive clinical experience with alcoholism) usually in participants’ homes or a treatment facility. Siblings and parents were usually interviewed in their homes. A small proportion of siblings who lived outside Ireland were interviewed by telephone. Lifetime history of DSM-IV AD was assessed in probands and siblings using the SSAGA interview (version 11,21), modified to reduce assessment time (e.g., we deleted items assessing onset ages of most individual symptoms), and among parents using an adapted version of the SCID interview.22
We conducted linkage analyses for two phenotypes: AD and number of AD symptoms (ADSX) based on number of positive DSM-IV AD criteria. Alcohol dependence symptoms scores range from 3 to 7 (because all affected individuals have at least three positive criteria). Phenotypic information from parents was not included in symptom count analyses (unless the parent was part of an affected sib pair and received the SSAGA interview) because symptom counts based on the SCID versus SSAGA interview are not directly comparable. As expected for a clinically ascertained sample, the probands and siblings were severely affected. Endorsement rates for the seven AD criteria ranged from 87 to 97%. The distribution of ADSX among probands (P) and siblings (S) was: three symptoms, P=1.4%, S=5.1%; four, P=3.2%, S=6.2%; five, P=8.8%, S=10.3%; six, P=18.3%, S=18.5%; seven, P=68.4%, S=59.8%.
DNA extracted from blood samples provided by 66 volunteer control subjects was included in the genome scan to obtain allele frequency estimates. Controls were recruited in the Republic from the Garda Siochana (the national police force) and the Forsa Cosanta Aituil (the army reserve) and in Northern Ireland from volunteers donating at the Northern Ireland Blood Transfusion Service. Controls were screened and their samples excluded if they reported a history of heavy drinking or problem alcohol use.
All interviewed participants provided informed consent before assessment and sample collection. The study protocol and consent procedures were approved by the VCU Institutional Review Board, Western IRB, the Health Research Board of the Irish Republic, and the human subjects committees of the treatment facilities from which participants were recruited (where such committees existed).
All but seven of the 1414 family participants provided DNA samples. Individuals not willing to donate blood and those interviewed by telephone were asked to contribute buccal epithelial cells collected with four cytology brushes. Eighty-five percent provided blood or blood plus brush samples, 15% gave brush samples only. Blood samples were collected by trained phlebotomists and consisted of 17 ml in two tubes. High molecular weight DNA was extracted from blood samples using standard techniques and stored at −20 °C until shipment to the molecular genetics laboratory at Virginia Commonwealth University. Buccal cell samples were shipped directly from the study sites to the VCU laboratory where DNA extraction was performed.
An autosomal genomewide screen was conducted by deCODE Genetics (http://www.decode.com/) using a standard panel of polymorphic microsatellite markers. There were 1020 markers with an average spacing of 4 cM. The average marker heterozygosity was 72.5% (min 6.3%, max 91.8) as calculated from our linkage sample using PEDSTATS.23
The genome scan was conducted with samples from 1473 individuals, including 1407 sibs and family members and 66 control subjects. Genotyping reliability was assessed in 17 individuals for whom we submitted two blinded samples for genotyping. We estimated the percentage of called markers, which received the same genotype assignment for both samples. Between-sample agreement ranged from 98.44 to 100%, with an average of 99.68%.
Checking of reported familial relationships and marker-error detection was conducted using GRR.24 Sixty-five families were identified with genotypes inconsistent with reported family structure. Of these, 26 were resolvable due to problems such as nonpaternity or by removing a problem sample from a family with more than two affected siblings. The remaining 39 families were excluded either because the problem could not be resolved or excluding the problem sample made a family uninformative for linkage (no affected sib pair). After data cleaning, 1289 samples were included in the final scan. The genotyping success rate was 90.2%, reflecting 1 185 544 of a possible 1 314 780 genotypes (1289 subjects × 1020 markers). The final sample used for linkage analyses included 474 families ranging from 2 to 10 members with 81.1, 14.2 and 1.9% containing 2, 3 or 4 persons, respectively. The siblings formed 701 sibling pairs (counting all possible pairings).
Alcohol dependence diagnosis was analyzed using single- and multipoint nonparametric linkage (NPL) analyses performed using the program Merlin.25 Alcohol dependence symptoms symptom count was analyzed using the QTL linkage routine in Merlin and ordinal linkage analysis using the Mx program.26 Allele frequencies were estimated from the genotypes of the control subjects.
As a result of the clinical severity of our sample, the distribution of ADSX is negatively skewed. This violates the assumptions of the QTL method and could produce misleading results. We dealt with this problem in two ways. First, we calculated empirical P-values for the QTL method. We used the simulate option in Merlin to estimate the empirical confidence intervals for our linkage peak. Random data sets were generated that have similar distributions to the original data in terms of family structure, marker informativeness, spacing and missing data patterns. In addition, phenotypic measurements, quantitative traits and affection status are preserved. Merlin performs gene-dropping simulations under the null hypothesis of no linkage or association to the observed phenotype. These data sets replace the input data with simulated chromosomes. A total of 10 000 bootstrap samples were generated consisting of 474 informative families in each sample. An NPL analysis was conducted with each sample and the resulting 10 000 estimates of the Z statistic and LOD scores were rank-ordered. The probability of obtaining a more extreme value than our observed test statistic (LOD=4.59) under the null hypothesis is thus P<0.0001. The ‘suggestive’ genomewide linkage criterion of P<0.0007 proposed by Lander and Kruglyak27 corresponds to a LOD of approximately 2.4 in our sample.
Our second approach to addressing the skewness of ADSX was to conduct ordinal linkage analysis. This treated ADSX as an ordinal variable with an underlying normal distribution for liability. Mx uses the maximum likelihood method to estimate thresholds for each category. Based on the density function of the normal distribution, the area between two adjacent thresholds was calculated as the observed proportion of the sample found in that category. The variance of ADSX was explained by a QTL component, a shared familial component, and a residual error component. We then calculated the LOD score for each marker as the difference in the χ2 for a full model (including the QTL effect) and the χ2 of the submodel without the QTL effect. Owing to the computation time required for these analyses, we ran them only for markers for which the QTL analyses suggested evidence for linkage (P<.001).
Figure 1 displays the genome scan results for chromosomes 1–22 for AD. There were three regions where the NPLOD score exceeded 1.0, including chromosomes 1q44 (peak multipoint LOD=1.20, P=0.009 at D1S2670), 13q31.1 (LOD=1.08, P=0.013 at D13S271) and 22q11.21 (LOD=1.56, P=0.004 at D22S1638). Details of markers and multipoint P-values are displayed in Table 1.
Figure 2 displays the genome scan results for ADSX. A region of linkage on chromosome 4q24–27 obtained genomewide significance (peak multipoint NPL LOD=4.59, P=2.1 × 10−6 at D4S1611). The linkage region was broad; the area with LOD scores >2.0 encompassed 23 cM (from 103 to 126 cM) and weak evidence for linkage was seen as far away as 4q32.1 (NPL LOD=1.55, P=0.005 at D4S2918). The NPL LOD score for ADSX exceeded 1.0 for four additional regions: 2q37.3 (NPL LOD=1.32, P=0.007 at D2S140), 9q21.13 (NPL LOD=1.40, P=0.006 at D9S1876), 9q34.11 (NPL LOD=1.00, P=0.02 at D9S271) and 18p11.32 (NPL LOD=1.04, P=0.016 at D18S1105). The empirical P-values for these markers are shown in Table 1. They were very close to those produced by the Merlin QTL routine, suggesting these estimates are not biased by being based on a selected sample or ADSX measurement. The single point results showed four markers with LODs greater than 2.0 within the chromosome 4 multipoint peak, indicating the multipoint result is not due to a highly positive result at a single marker.
Figure 3 displays the results for AD and ADSX for chromosome 4. The LOD for AD is near 0 in the region with peak linkage to ADSX. This suggests that the ADSX peak is not due to a small linkage signal associated with the dichotomous trait (AD) being magnified by moving to an outcome with greater information. Instead, the peak arises from within-pair differences on ADSX being associated with within-pair genetic variation in this region. Single point results based on ordinal analyses are displayed for 24 markers under the chromosome 4 peak (cM 62–177) and listed in Table 1. The linkage peak from the ordinal analysis is located at the same position as that derived from the QTL method, although the LOD scores are lower, consistent with the reduced power available from ordinal data.
We conducted follow-up analyses to assess the validity of the linkage findings and clarify their interpretation. First, the results of sensitivity analyses indicate that the chromosome 4 peak was not due to the disproportionate influence of a few pedigrees. Excluding families with more than four siblings from the QTL linkage analysis altered the results minimally and actually produced a slightly higher peak on chromosome 4 (NPL LOD=5.1 at D4S1611). Positive LOD scores at this marker were obtained for 52% of the families and no family had a LOD greater than 0.2.
We then conducted two sets of linkage analyses to determine the basis for the difference between the ADSX and AD chromosome 4 results and to identify whether specific symptoms were responsible for the ADSX linkage peak. A series of seven analyses of each of the AD criteria produced small peaks (all NPL LOD ∼1.0) at the same location as the ADSX result, suggesting that each criterion was contributing a small amount of information.
Finally, we conducted a complementary set of seven analyses, sequentially dropping each of the seven criteria before forming the symptom count. As shown in Figure 4, in each analysis the peak NPL LOD was obtained in the same region of chromosome 4 and had the same shape as for ADSX, but the values ranged widely. For three criteria (withdrawal, lack of control and drinking to the exclusion of other activities), the LODs decreased only slightly. Excluding the continued drinking despite serious medical or psychiatric consequences criterion actually increased the LOD slightly (to 4.73). The largest effect was obtained when the tolerance criterion was excluded, resulting in a peak NPL LOD of 1.80. Substantial decreases in the LODs were also seen for binge drinking (LOD=2.67) and failed attempts to cut back or quit drinking (LOD=3.28). A direct interpretation of these results is complicated by the fact that several criteria were rated as positive for virtually all probands and siblings (e.g., >98% endorsed lack of control), and thus the apparent absence of contribution of some criteria to the linkage peak may result from lack of phenotypic variation rather than genetic involvement. This is also suggested by the finding that when included individually in linkage analyses, each criterion obtained a peak LOD of about 1. In summary, the results are consistent with a large proportion of the linkage peak being due to phenotypic variation in tolerance to ethanol and binge drinking.
We found strong evidence of linkage of AD symptom count to chromosome 4q22–q32, suggesting genetic variation within this region may be associated with variation in symptomatology of AD. Linkage of AD or alcohol-related phenotypes to this region (or nearby areas) of chromosome 4 has been previously reported in three other samples (Table 2). The strongest evidence previously reported for this region was from the Collaborative Study on the Genetics of Alcoholism (COGA;5) for two severity-related phenotypes: maximum drinks consumed in 24 h (peak LOD=3.5 at 121 cM,28) and a multiple-threshold definition of AD adjusted for age and sex (peak LOD=3.5 at 127 cM,29). Weaker evidence of linkage has been found in a subset of COGA families (the initial, but not replication sample) for number of AD criteria,5 AD when analyzed in combination with electrophysiological measures,30 and a psychometrically defined phenotype characterized by later onset of alcohol use and anxious personality.31 Excess IBD sharing in this region was also observed among pairs of siblings who had low severity (<8 positives out of 37 items assessing alcohol-related problems), suggesting a possible protective role.5
Linkage of this region to drinking-related phenotypes has also been reported in two other samples. Ehlers et al.4 found a linkage for AD severity based on a four-item symptom count in families from a Mission Indian tribe. Bergen et al.8 found a linkage to maximum drinks consumed in a 24-h period in the family sample from the Framingham Heart study. This region has also been implicated in association-based genome scans of illicit drug abuse conducted in samples of European and African ancestry (Uhl et al., 2002).32
Several studies have not reported any significant linkage to this region of chromosome 4, including: a study of AD in multiplex families,7 preliminary reports of AD samples collected in the UK33 and Sweden,34 a study of alcohol craving in the Mission Indian sample,35 and two studies of ethanol response based on self-report among the initial COGA sample36 and on physiological measures in a sib pair sample.9 The latter study reported some evidence (LODs 1.3–1.4) for linkage to regions ∼40 cM away from our peak (at 60 and 180 cM) which could overlap with the region we identified given the wide confidence intervals surrounding linkage peaks.37
Although the power of these studies to reject linkage is low, it is noteworthy that for only one of the six published samples has there been a linkage on chromosome 4 to AD diagnosis,3 but for three there has been linkage to a severity or drinking volume phenotype. This is consistent with our finding virtually no evidence of linkage of this region to diagnosis but strong evidence of linkage to symptom count. Our symptom-dropping analyses (Figure 4) found the evidence for linkage on chromosome 4 decreased substantially when the tolerance and binge drinking criteria were excluded, suggesting this region may contain genes affecting variation in drinking volume.
One obvious set of candidate genes in this region is the alcohol dehydrogenase (ADH) cluster, located on chromosome 4q21–23. In man, the ADH cluster includes seven genes coding for five classes of enzymes involved in ethanol metabolism and expressed primarily in liver and gut.38 ADH oxidizes ethanol to acetaldehyde, from which it is oxidized by aldehyde dehydrogenase (ALDH) to acetate, which can be excreted. Some combinations of ADH and ALDH variants result in build-up of acetaldehyde, which produces an unpleasant ‘flushing’ response characterized by increased blood flow, tachycardia and nausea. Individuals with such genotypes are at very low risk to develop AD. The effect is due in large part to an ALDH2 variant that is common in Asian populations but rare in other ethnic groups.39 However, there is evidence that variants in ADH1B and ADH1C (previously called ADH2 and ADH3, respectively) are also associated with variation in alcohol metabolism. Polymorphic forms of ADH1B and ADH1C are present in many populations and have been found to be associated with individual differences in alcohol consumption in Asian, Caucasian, Jewish, South Pacific and Native American samples (reviewed in40). Variation at ADH1B has been associated with risk for AD, particularly in Asian groups.41 Some studies suggest variation at ADH1C may also affect risk for AD in some populations (e.g.,41). Follow-up work in a Southwest US Indian tribe sample3 indicated the chromosome 4 linkage signal for AD was due in part to several variants in ADH1C.42 However, other evidence suggests the association between ADH1C and AD in some populations may be due to linkage disequilibrium between ADH1C and ADH1B.43
Alcohol dehydrogenase variants probably affect risk for AD indirectly, via limitations on alcohol metabolism and consumption. This could explain why the phenotypes in most studies reporting linkage to this region are related to the amount of ethanol consumed or to AD severity rather than to AD status. Alcohol consumption is not itself a diagnostic criterion for AD (although it is reflected in some criteria, particularly tolerance), and there are large individual differences in the intake level at which physiological dependence develops. Individuals with rapid metabolism can consume more ethanol before experiencing its sedating effects and thus may consume greater amounts than individuals with slower metabolic rates. This results in increased exposure of organ systems to the toxic effects of ethanol, particularly in liver and pancreas.44 Thus, it is plausible that these linkage results arise because individuals with rapid metabolism tend to develop a more severe clinical presentation, including high physiological tolerance and binge drinking.
Future work will include follow-up of the chromosome 4 linkage region, candidate gene studies and further psychometric analyses of clinical features. Our results suggest several other areas may merit follow-up genotyping, including regions on chromosomes 1, 13 and 22 for AD and regions on 2, 9 and 18 for AD symptom count. The evidence for these regions is weak in our sample, but they overlap with regions identified in other linkage studies of alcohol-related phenotypes on chromosome 1,8, 31 22,8 and possibly 2, 13 and 18.7
Study limitations and strengths
Our study is under-powered to detect susceptibility loci of small effect when using linkage analyses based on affected sib pairs (i.e. , AD as the outcome). Power analyses based on a variance component method using estimates from the twin literature (total genetic variance of 50%, no dominance variation, no residual familial resemblance) indicate that a sample size of 500 sib pairs has approximately 78% power to detect QTLs accounting for 25% of the variance, but only 29% power to detect QTLs of 10% or below when using a P level of 0.05 (http://statgen.iop.kcl.ac.uk/gpc). Our power could have been increased by including genotypic information from unaffected siblings, but for the reasons described previously, we opted to study only affected siblings.
The method used to analyze ADSX assumes a continuously distributed trait score and an unselected sample. Simulations based on our sample size and structure suggest that a linkage peak of this size would occur by chance fewer than one out of 10 ,000 times. Studies of population-based twin samples find heritability estimates based on symptom count are similar to those from AD diagnosis, suggesting symptom count is a valid index of genetic risk.12 The difference in evidence for linkage between the QTL and ordinal analyses of chromosome 4 are probably due to lower statistical power associated with the ordinal analyses, but could be related to failure to meet the QTL analysis assumption of a normal distribution of ADSX scores. However, the similarity of the peak locations under both methods, the similarity of this result to those obtained by other investigators, and the existence of a strong candidate gene under the peak argue against our result being a statistical artifact.
Strengths of the study include a well-characterized, severely affected sample, which is relatively genetically homogeneous and culturally homogeneous with respect to alcohol use. The use of an affected sib pair design means that these families are likely to be representative of the etiology of alcoholism in the general population.
Grant B . Prevalence and correlates of alcohol use and DSM-IV alcohol dependence in the United States: results of the National Longitudinal Alcohol Epidemiologic Survey. J Stud Alcohol 1997; 58: 464–473.
Prescott CA . The genetic epidemiology of alcoholism: sex differences and future directions. In: Agarwal DP, Seitz HK (eds). Alcohol in Health and Disease. Marcel Dekker: New York, NY, 2001, pp 125–149.
Long J, Knowler WC, Hanson RL, Robin RW, Urbanek M, Moore E et al. Evidence for genetic linkage to alcohol dependence on chromosomes 4 and 11 from an autosome-wide scan in an American Indian population. Am J Med Genet (Neuropsychiat Genet) 1998; 81B: 216–221.
Ehlers CL, Gilder DA, Wall TL, Phillips E, Feiler H, Wilhelmsen KC . Genomic screen for loci associated with alcohol dependence in Mission Indians. Am J Med Genet (Neuropsychiatr Genet) 2004; 129B: 110–115.
Reich T, Edenberg HJ, Goate A, William JT, Rice JP, Van Eerdewegh P et al. Genome-wide search for genes affecting risk for alcohol dependence. Am J Med Genet (Neuropsychiatr Genet) 1998; 81B: 206–215.
Foroud T, Edenberg HJ, Goate A, Rice J, Flury L, Koller DL et al. Alcoholism susceptibility loci: confirmation studies in a replicate sample and further mapping. Alcohol Clin Exp Res 2000; 24: 933–945.
Hill SY, Shen S, Zezza N, Hoffman EK, Perlin M, Allan W . A genome wide search for alcoholism susceptibility genes. Am J Med Genet (Neuropsychiatr Genet) 2004; 128B: 102–113.
Bergen AW, Yang XR, Bai Y, Beerman MB, Goldstein AM, Goldin LR, Framingham Heart Study. Genomic regions linked to alcohol consumption in the Framingham Heart Study. BMC Genet 2003; 4 (Suppl 1): S101.
Wilhelmsen KC, Schuckit M, Smith TL, Lee JV, Segall SK, Feller HS et al. The search for genes related to a low-level response to alcohol determined by alcohol challenges. Alcohol Clin Exp Res 2003; 27: 1041–1047.
Zucker RA, Gomberg ES . Etiology of alcoholism reconsidered: the case for a biopsychosocial process. Am Psychologist 1986; 41: 783–793.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edn. American Psychiatric Association: Washington, DC, 1994.
Prescott CA, Aggen SH, Kendler KS . Sex differences in the sources of genetic liability to alcohol abuse and dependence in a population-based sample of U. S twins Alcohol Clin Exp Res 1999; 23: 1136–1140.
Bennett LA, Janca A, Grant BF, Sartorius N . Boundaries between normal and pathological drinking: a cross-cultural comparison. Alcohol Health Res World 1993; 17: 190–195.
Conniffe D, McCoy D . Alcohol use in Ireland: Some economic and social implications. ERSI: Dublinm, Ireland, 1993.
Walsh D, Walsh B . Drowning the shamrock: Alcohol and drink in Ireland in the post-war period. In: Single E, Morgan P, deLint J (eds). Alcohol, Society and the State. Vol. 2. The Social History of Control Policy in Seven Countries, A Report of the International Study of International Control Experiences, in collaboration with the World Health Organization Regional Office for Europe Addiction Research Foundation: Toronto, Canada, 1981, pp 103–126.
Strategic Task Force on Alcohol. Second Report, September 2004. Department of Health and Children: Dublin, 2004.
Kendler KS, MacLean CJ, Ma Y, O’Neill FA, Walsh D, Straub RE . Marker-to-marker linkage disequilibrium on chromosomes 5q, 6p, and 8p in Irish high-density schizophrenia pedigrees. Am J Med Genet 1999; 88: 29–33.
Relethford JH . Genetic structure and population history of Ireland: a comparison of blood group and anthropometric analyses. Ann Hum Biol 1983; 10: 321–334.
Sunderland E, Tills D, Bouloux C, Doyl J . Genetic studies in Ireland. In: Roberts DF (ed). Genetic Variation in Britain XII. Taylor and Francis Ltd: London, 1973 pp 141–170.
Prescott CA, Sullivan PF, Myers JM, Patterson DG, Devitt M, Halberstadt LJ et al. The Irish affected sib pair study of alcohol dependence. Study methodology and validation of diagnosis by interview and family history. Alcohol Clin Exp Res 2005; 29: 417–429.
Bucholz KK, Cadoret RJ, Cloninger CR, Dinwiddie SH, Hesselbrock VM, Nurnberger Jr JI et al. A new, semi-structured psychiatric interview for use in genetic linkage studies: a report on the reliability of the SSAGA. J Stud Alcohol 1994; 55: 149–158.
Spitzer RL, Williams JBW . Structured Clinical Interview for DSM-III-R (SCID). Biometrics Research Department. New York State Psychiatric Institute: New York, 1985.
Wigginton JE, Abecasis GR . PEDSTATS: descriptive statistics, graphics and quality assessment for gene mapping data. Bioinformatics 2005; 21: 3445–3447.
Abecasis GR, Cherny SS, Cookson WO, Cardon LR . GRR: graphical representation of relationship errors. Bioinformatics 2001; 17: 742–743.
Abecasis GR, Cherny SS, Cookson WO, Cardon LR . Merlin-rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet 2002; 30: 97–101.
Neale MC, Boker S, Xie G, Maes H . Mx: Statistical Modeling. 6th ed. Virginia Commonwealth University, Department of Psychiatry: Richmond, VA, 2004.
Lander E, Kruglyak L . Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 1995; 11: 241–247.
Saccone NL, Kwon JM, Corbett J, Goate A, Rochberg N, Edenberg HJ et al. A genome screen of maximum number of drinks as an alcoholism phenotype. Am J Med Genet (Neuropsychiatr Genet) 2000; 96B: 632–637.
Corbett J, Saccone NL, Foroud T, Goate A, Edenberg H, Nurnberger J et al. A sex-adjusted and age-adjusted genome scan for nested alcohol dependence diagnoses. Psychiat Genet 2005; 15: 25–30.
Williams JT, Begleiter H, Porjesz B, Edenberg HJ, Foroud T, Reich T et al. Joint multipoint linkage analysis of multivariate qualitative and quantitative traits. II. Alcoholism and event-related potentials. Am J Hum Genet 1999; 65: 1148–1160.
Dick DM, Nurnberger Jr J, Edenberg HJ, Goate A, Crowe R, Rice J et al. Suggestive linkage on chromosome 1 for a quantitative alcohol-related phenotype. Alcohol Clin Exp Res 2002; 10: 1453–1460.
Uhl GR, Liu OR, Naiman D . Substance abuse vulnerability loci: converging genome scanning data. Trends Genet 2002; 18: 420–425.
Guerrini I, Cook CC, Kest W, Devitgh A, McQuillin A, Curtis D et al. Genetic linkage analysis supports the presence of two susceptibility loci for alcoholism and heavy drinking on chromosome 1p22.1–11.2 and 1q21.3–24.2. BMC Genet 2005; 6: 11–18.
Osby U, Bransome M, Schumacher J, Liljenberg J, Brandt L, Nothen M et al. A genetic linkage study of alcohol addiction in a Swedish affected sibpair sample. Am J Med Genet (Neuropsychiat Genet) 2002; 114B: 781–782.
Ehlers CL, Wilhelmsen KC . Genomic scan for alcohol craving in Mission Indians. Psychiatr Genet 2005; 15: 71–75.
Schuckit MA, Edenberg HJ, Kalmijn J, Flury L, Smith TL, Reich T et al. A genome-wide search for genes that relate to a low level of response to alcohol. Alcohol Clin Exp Res 2001; 25: 323–329.
Roberts SB, MacLean CJ, Neale MC, Eaves LJ, Kendler KS . Replication of linkage studies of complex traits: an examination of variation in location estimates. Am J Hum Genet 1999; 65: 876–884.
Lee S-L, Hoog J-O, Yin S-J . Functionality of allelic variations in human alcohol dehydrogenase gene family: assessment of a functional window for protection against alcoholism. Pharmacogenetics 2004; 14: 725–732.
Agarwal DP, Goedde HW . Pharmacogenetics of alcohol metabolism and alcoholism. Pharmacogenetics 1992; 2: 48–62.
Crabb DW, Matsumoto M, Chang D, You M . Overview of the role of alcohol dehydrogenase and aldehyde dehydrogenase and their variants in the genesis of alcohol-related pathology. Proc Nutri Soc 2004; 63: 49–63.
Shen Y-C, Fan J-H, Edenberg HJ, Li T-K, Cui Y-H, Wang Y-F et al. Polymorphism of ADH and ALDH genes among four ethnic groups in China and effects upon the risk for alcoholism. Alcohol Clin Exp Res 1997; 21: 1271–1277.
Mulligan CJ, Robin RW, Osier MV, Sambuughin N, Goldfarb LG, Kittles RA et al. Allelic variation at alcohol metabolism genes (ADH1B, ADH1C, ALDH2) and alcohol dependence in an American Indian population. Hum Genet 2003; 113: 325–336.
Osier MV, Pakstis AJ, Soodyall H, Comas D, Goldman D, Odunsi A et al. A global perspective on genetic variation at the ADH genes reveals unusual patterns of linkage disequilibrium and diversity. Am J Hum Genet 2002; 71: 84–99.
Yin S-J, Agarwal DP . Functional polymorphisms of alcohol and aldehyde dehydrogenases. In: Agarwal DP, Seitz HK (eds). Alcohol in Health and Disease. Marcel Dekker: New York, NY, 2001, pp 1–26.
We are grateful to the study participants and their families for their time and effort. We thank Aine Finnerty, Phil Gavigan, Craig Barton, John Cosgrove, Michael Crossan, Sara Dineen, Claire Killeen, Deirdre King, Siobhan McHugh, Amanda Mullan, Eileen Murphy, Brian O’Malley and Bernie Purcell for their roles in data collection and Ruth Barrington, Ros Moran and Carol Cronin for administrative support. F Anthony O’Neill of Queens University, Belfast NI, the Northern Ireland Blood Transfusion Service and the Irish Gardai provided assistance in obtaining control blood samples. John Myers assisted with data analysis, Indrani Ray and Cheryl Smith assisted with database management, Brandon Wormley assisted with genotyping, and Stacey Garnett, Jill Opalesky and Rebecca Ortiz provided administrative assistance at VCU. This work was funded by US National Institutes of Health grant R01-AA-11408 with administrative support from the Irish Health Research Board.
About this article
Cite this article
Prescott, C., Sullivan, P., Kuo, PH. et al. Genomewide linkage study in the Irish affected sib pair study of alcohol dependence: evidence for a susceptibility region for symptoms of alcohol dependence on chromosome 4. Mol Psychiatry 11, 603–611 (2006). https://doi.org/10.1038/sj.mp.4001811
- binge drinking
- family study
- molecular genetics
Behavior Genetics (2015)
Molecular Genetics and Genomics (2014)
What psychiatric genetics has taught us about the nature of psychiatric illness and what is left to learn
Molecular Psychiatry (2013)
Journal of Molecular Neuroscience (2013)
A genome-wide association study of alcohol-dependence symptom counts in extended pedigrees identifies C15orf53
Molecular Psychiatry (2013)