Familial aggregation occurs in many medical and psychiatric afflictions. Causal factors that place members of a family at high risk for developing the same disease can generally he derived from genes, environment, or both. An example of major effect from a single gene, or a monogenic trait, is found in familial hypercholesterolemia, a rare genetic disorder. However, the blended effects of many genes, or polygenic trails, may account for more than half the normal population's variation in serum cholesterol level.1 Even the absente of monogenic or polygenic traits, a person's life style, including dietary fai intake and level of physical activity, can still clearly have an effect on his cholesterol level.
Theoretically, the isolated effects of major genes, polygenes, or environmental variables could produce disease. However, it appears more likely that a combination of all three is often present. Furthermore, interactions between genetic and environmental variables can produce an even more dramatic increase in risk for disease. An example is cigarette smoking, which seems to confer a magnified risk for coronary heart disease in persons with a positive family history for this disease.2,3
The concept of familial aggregation appears to be valid for the disease of alcoholism, although our knowledge base is far from complete. Indeed, the idea that alcoholism runs in families dales back to antiquity. Aristotle declared that drunken women "'bring fiorili children like themselves." Plutarch said, "one drunk begets another." Still it was not until late in the 19th century that the fust systematic studies of families of alcoholics were initialed. Almost without exception, every family study of alcoholism, regardless of country of origin, has shown higher rates of alcoholism among the relatives of alcoholics than occur in the general population.4
Since alcoholism runs in families. an obvious question arises. What are the relative contributions of environmental and hereditary factors to this disease?' Some claim that alcoholism is a learned habit, passed down by example horn one generation to the next. Others maintain that a gene or a combination of genes place certain individuals at high risk for developing this disease. In an attempt to separate genetic from environmental factors, various investigators, both in the United States and abroad, have employed twin, sibling, and adoption studies5-9 The epidemiologic and clinical data converge to suggest a clear genetic component to the disease of alcoholism. Sons of alcoholics are about four times more likely to be alcoholic than arc sons of nonalcoholics. Moreover, the incidence of this disease in daughters adopted from alcoholic mothers is three times that of daughters adopted from nonalcoholic parents.
If alcoholism is an inherited disease, how is it represented in the central nervous system (CNS)? To answer this question, alcoholic and nonalcoholic individuals have been studied and compared. However, any differences found in the alcoholics may be obfuscated by the chronic cytotoxic effects of alcohol on brain functioning. Thus a better strategy might be to study children of alcoholics and nonalcoholics before they have begun to consume alcohol and other drugs.
Over the past 6 years, my colleagues and I have embarked on a multidisciplinary prospective study assessing CNS functioning in alcoholic and nonalcoholic men and their prepubescenl sons. For this study. 60 father-son pairs were screened rigorously to select three distinct groups. The high risk group (A+ ) had a strong family history of alcoholism. It consisted of 20 boys and their recovering alcoholic fathers; the fathers also had at least one first- or second-degree alcoholic relative. The low risk group (NA-) consisted of 20 boys and their nonalcoholic fathers who had no first- or second-degree alcoholic relatives. The third group (NA + ) consisted of 20 boys and their nonalcoholic lathers who had at least one first- or second-degree alcoholic relative.
These three groups of matched subjects were studied across a variety of CNS measures and their home environments were assessed. Moreover, the boys, who at entry averaged 10 years of age and had not used alcohol or other drugs, were followed with a behavioral questionnaire to determine their subsequent drug use behaviors.
Three aspects of CNS functioning were analyzed: neuropsychological and electrophysiological performance and personality traits. Neuropsychological tests of visuoperception and memory showed reduced performance in both A-I- sons and their lathers, compared to NA- sons and their fathers. No significant differences were found among any of the groups in the EEC, nor in a simple evoked-responsc paradigm involving a color discrimination task. However, the P300 (a component of the cortical evoked potential) generated by a more complex and demanding task (the Continuous Performance Task [CPT]) showed lower amplitude in both A + sons and fathers compared to NAsons and fathers.10
Begleiter et al11 and others have observed a similar reduction in P300 amplitude in sons of alcoholics. In addition, we found this reduction in amplitude (of the late positive complex of the eventrelated potential) to be significantly correlated with poorer visuoperceptual performance.10 Recently. Parsons et al12 validated this relationship in alcoholics with a positive family history of alcoholism when compared to nonalcoholic samples.
Similarly, using three different personality tests, several traits distinguished A+ sons and their fathers from control subjects. In particular, A 4- sons and their lathers were found to be more harm-avoidant compared to NA- sons and their lathers. Thev were also more tense, more easily fatigued, and less adventuresome. When a personality index or composite was computed which optimized group separation, it was found to be significantly correlated with the P300 amplitude, memory, and the Embedded Figures Test (a measure of visuoperception).
This evidence suggests that the underlying personal traits of A + sons and their fathers share a common heritable substrate with the deficits in mentor)', visuoperception, and CPT performance that appear to be unrelated to characteristics of the family environment.13 It is possible that this atypical CNS profile is a heritable anlage that may be related to the development of alcoholism. Detailed results of these findings have been published elsewhere.10,14,15
When alcohol and other drug use behaviors were assessed prospectively over the ensuing three years, important differences emerged among the three groups of boys. The greatest percentage of users of alcohol, tobacco, and marijuana were the A4- boys, followed in descending order by the NA+ and NA- boys. The difference in use of each of these three substances between the A+ and NA- boys was statistically significant. However, our subject boys are still young, and serious problem behaviors relative to these substances are yet to emerge. Therefore, follow-up studies are being conducted to determine whether, in fact, the atypical CNS profile observed as a trait marker can accurately predict (he development of alcoholism and other drug-related problems.
If epidemiologic and clinical studies suggest a heriiable component to alcoholism, and if children of alcoholics display trait phenotypes such as the (INS markers noted previously, is there a molecular genetic basis for this disease? The answer to this question would not have been possible 10 years ago; however, since then, the extraordinary progress in DNA technology has made feasible the identification of genes involved in several inherited diseases.
When Kenneth Blum and I decided to collaborale on molecular genetic studies in alcoholism, we asked ourselves which gene(s) should be considered. Because approximately 100 000 genes are present in humans, our research was necessarily limited. As research in alcoholism had previously shown that certain neurotransmitters are involved in alcohol-related behaviors, we chose, as candidate probes, those neurotransmitter genes and a few other genes ahead) suspected of involvement in alcoholism. Of particular interest to us was the dopaminergic system because alcohol, acutely administered, has been shown to stimulate brain reward svslenis,"1 in pari through release of dopamine.1' By a fortunate set of circumstances, Olivier Civelli and colleagues had just cloned and expressed the D,, dopamine receptor (D12DR) complementary DNA1* and had mapped the D9DR gene to the q22-q23 region of chromosome 1 1 in humans.'9 (avelli generously provided us a clone of D0DR gene (XhD2Ol) for our studies.
Out nexl consideration was which types of alcoholics to study. While there are many definitions of alcoholism, including those by die World Health Organization, the National Council on Alcohol and Drug Dependencv, and the American Psychiatric Association, a number of alcoholic types also have been described in ihe literature. Jellinek, in the 1940s, described five types of alcoholics: alpha, beta, gamma, delta, and epsilon. Alcoholics have also been characterized as essential versus reactive alcoholics, primary versus affective disorder alcoholics, and (on the basis of psychiatric syndromes) depressed versus neurotic versus psychotic alcoholics. Recently Cloninger-0 proposed two types of alcoholics, type 1 and type 2, distinguished by their characteristics of alcohol-related problems and personality traits. While this formulation has had a clear heuristic value, at leasl three groups of investigators have been unable to verify its construct validity in the clinical setting.-1'2*
There is still uncertainty about what constitutes various alcoholic types, and ongoing research in our laboratory and elsewhere is attempting to delineate alcoholic trait phenolypes. For our initial molecular genetic stitches we opted to investigate the association of certain genes with a tvpe of virulent alcoholism that has a fatal outcome. While this may constitute a relatively small proportion of alcoholics in ihe general population, we reasoned that the overwhelming and unremitting alcohol-seeking behavior of these patients could have a molecular genetic basis.
We were fortunate that brain samples for such an association study were being collected by the National Neurological Research Bank in Los Angeles, Calif., from deceased alcoholics and nonalcoholics. Moreover, extensive data were available on these subjects. Included were hospitalization and other clinical records, anatomical examination at autopsy for organ pathology (macroscopic and microscopic), and body fluid analysis for levels of alcohol and other drugs. Detailed interviews with next of kin and relatives provided demographic and personal information, with specific attention to drinking behavior including quantity, frequency, and kind of beverage consumed, and the type of alcohol and behavioral problems the deceased had experienced.
Two trained psychiatrists independendy examined these records, and using DSM-HI-R criteria, alcoholic (alcohol-dependent and alcohol abuse) and nonalcoholic diagnoses were made. Concordance between ihese two assessments was 100% in diagnosing alcoholism. From this sample, brains of 35 alcoholics and 85 nonalcoholics were chosen, matched for age, race, sex, and autolysis lime. Ii should be noted that a large majority of the alcoholics had experienced repealed treatment failures, and the cause of death was primarily attributed to severe alcohol-induced damage to their bodily systems. Autopsy findings included Laënnec's cirrhosis, anasarca, cardiopulmonary-renal failure, ruptured esophageal varices, among others. The causes of death in the nonalcoholics included cancer, myocardial infarction, stroke, accidents, homicide, and suicide.
The frontal brain cortex was removed at autopsy by a neuropathologist and frozen at -700C. The brain samples were then coded without reference to their group identity (nonalcoholic or alcoholic). Using techniques of restriction fragment length polymorphism (RFLP) these samples were processed for highmolecular-weight genomic DNA, and the DNAs were digested separately with four different restriction endonucleases (Taql, M.spl, IicoRl. and PsIl) as described.2'4 These digests were then hybridized with various putative alcoholism gene probes to determine polymorphism, i.e., whether more than one form of the gene prevails. Of the nine gene probes used, three showed polymorphism, however, of these three, two DNA probes (alcohol dehydrogenase and transferrin) revealed a polymorphism that was not significantly associated with alcoholism.
The only DNA probe that showed polymorphism to be significantly associated with alcoholism was the D2DR gene (MiD2G] probe). The brain samples showed the presence of both previously described forms (alleles) of this gene found in the general population:19 the less prevalent Al allele and die more prevalent A2 allele. In the nonalcoholic brain samples, 20% had the Al allele while 80%, did not. However, the brains of alcoholics showed the reverse pattern; 69% had the AI allele, while only 31% did not. When the proportion of the presence of the Al allele to the absence of this allele was determined, the presence of the AI allele was significantly different in alcoholics compared with nonalcoholics (/><0.001).
The data were also analyzed on die basis of race, revealing a similar pattern of Al distribution in while and black alcoholics and in nonalcoholics of die two races. When Af allelic distribution as considered between alcoholic and nonalcoholic men and women, no significant sex difference was observed.
DNA samples from the 70 brains were then grouped according to whether the Al allele was present. This grouping allowed a classification based on their unique allelic association with alcoholism. Of the 31 DNA samples diat possessed the Al allele, 24 (77%) were from alcoholics. In the remaining 39 DNA samples that lacked the Al allele, 28 (72%) were from nonalcoholics. Observed values were significantly different Dom assigned probability (0.50) for the presence of the AI allele in samples of alcoholics [p = .002) and for the absence of the Al allele in samples of nonalcoholics (p= .007).
Knowledge and understanding of human afflictions, and hence their diagnosis, treatment, and prevention, often occur in quantum leaps, frequently catalyzed when a new scientific technology is found and applied. This has been exemplified recent Iv by the discovery of restriction fragment length polymorphism (RFLP) and other molecular genetic techniques. The application of these techniques to heritable disorders, such as Huntington's chorea, Duchenne's muscular dystrophy, and cystic fibrosis, has led to the localization, identification, and cloning of genes responsible for a variety of somatic diseases. Soon this new knowledge will be applied to the identification of certain highrisk groups, and to the development of pharmacological agents to treat these diseases.
It should be expected that similar knowledge will soon be developed for psychiatric disorders. However, the complexity and diverse phenomena displayed in many of these disorders suggests environmental factors or die involvement of several predispositional genes interacting with environment in the expression of a specific behavioral aberration. In this respect, the multiplex nature of alcoholism, where gene(s) and environment act separately or together, may represent a prototypic model for studying other behavioral disorders. Unlike schizophrenia, manic-depressive psychosis, and anxiety disorders, in which environmental provocative factors have yet to be clearly identified, alcoholism specifically requires exposure to ethanol, a chemical possessing inherent addictive capability. Indeed, epidemiologic and clinical studies show that alcoholism does occur in individuals without a heritable background for this disease. Laboratory studies of animals and of nonalcoholic humans have shown that die full-blown symptoms of alcoholism can be induced when sufficient ethanol is administered over a prescribed time. Still, in the human condition, a significant majority of alcoholics in die clinical setting do have a heritable background for aleo holism. Therefore, the challenges for researchers attempting to understand alcoholism are at least fourfold:
* Identify biological and behavioral trait markers that distinguish alcoholic phenolypes.
* Delineate the genes that predispose individuals to alcoholism and how the products expressed by these genes manifest die various alcoholic" phenotypes.
* Identify environmental variables, including alcohol availability and psychosocial factors, that induce individuals to drink excessively.
* Determine ihe mechanisms through which predispositional genes and environmental variables interact in the development of alcoholism.
In this report, we have made a modest preliminary attempt to address the first two of these challenges. In the first, the demonstration of peculiar CNS functioning in sons of alcoholics with a family history of alcoholism must await prospective studies to determine whether this is indeed a trait marker for the development of alcoholism. It is of interest to note that the alcoholic fathers of these sons displayed an atypical CNS profile similar to their sons, suggesting a transgenerational communality in this measure. These fathers were recovering alcoholics who had, thus far, a good record of sobriety (average of 7 years' abstinence) and in whom the damaging effects of alcohol on bodily systems were minimal.
If this atypical CNS profile turns out to be an alcoholic phenotvpe. the possibility emerges thai other alcoholics who have a more virulent form of the disease, as well as their blood relatives, may display aberrant CNS functioning that differs from the pattern observed in our investigation. As alcoholism involves brain dysfunction, research on alcoholics and their pedigrees, defining profiles of CNS trait markers, may well lead to a more useful classification of alcoholic phenotypes than currently exists.
The second challenge, to delineate genes involved in alcoholism and its related problems (or genes that predispose individuals away from this disease), is almost certain to result in the identification of many such genes. These may include both those that control enzymes involved in alcohol metabolism and those that control the multiple effects that alcohol exerts on biological svstems. For example, ethyl alcohol, like certain other anesthetics, has a bi phasic effect: it is initially a stimulant, then a depressant. Taken over time and in sufficient quantities, it will produce both tolerance and dependence.
That some of these phenomena may have a genetic basis is suggested by recent animal studies in which inbred strains have been developed for high and low voluntan alcohol consumption, and also for differential sensitivity to the acute hypnotic effect of alcohol and to its withdrawal effects. Interestingly, a recent study in mice23 has shown that genes coding lot the CiABAx receptor may be a critical determinant of the differential sensitivity to acute doses of alcohol. In another type of alcohol sensitivitv known as the alcohol flush syndrome, prevalent among Asians, a structural mutation in the gene that codes for an enzvme involved in alcohol metabolism (aldehyde dehydrogenase) leads to the loss of the enzyme's ability to metabolize acetaldehvcle,21' a toxic product of alcohol metabolism. Such a gene may well act as a protective factor against heavy alcohol consumption in individuals who inherit il.
Alcoholism is also manifested in a variety of damages to bodily systems. As these van' from individual to individual, and are not absolutely dependent on the amount or duration of alcohol consumed, the role of predispositional genes is suggested for specific end-organ pathology induced by alcohol. For example, some alcoholic individuals have the severe organic brain syndromes of alcoholic dementia and Korsakoff's psychosis (WernickeKorsakoff syndrome). Since genes lor Alzheimer's disease (which is often difficult to differentiate from alcoholic dementia) are now being identified, it is possible that similar genes may be operating in alcoholic dementia. Furthermore, a genetic predisposition to Wernicke's encephalopathy has been posited through an enzymatic abnormality in thiamine metabolism.2' Similarly, genetic factors have been implicated in Laënnec's cirrhosis.28
It is of interest thai in a recent study, a specific pattern associated with a gene coding for type 1 collagen (the type most increased in dense fibrosis) was found to occur more frequently in alcoholic individuals with cirrhosis than in those without cirrhosis.2'1 Furthermore, certain alcoholics develop upper gastrointestinal cancers; it would not be surprising that in the presence of specifically mutated oncogenes, alcohol may stimulate the proliferation of uncontrolled growth. The list of predispositional genes may well extend in the future to other alcohol-related problems such as endocrine dysfunction (particularly gonadal atrophy), cardiovascular disease, or the fetal alcohol syndrome, to mention just a few.
Now that we have entered the era of molecular genetics, we may ven well begin to see many reports of genes said to be associated with alcoholism. However, the key issue for neuroscientists and psychiatrists - an issue specifically relevant to the second challenge - will be to identify the genes that induce some people to drink heavily. Identifying genes that render individuals susceptible to damaging effects of alcohol will certainly be useful in early identification and treatment of alcohol-associated disorders. However, our fundamental concern should be with genetic factors that enhance inherent susceptibility to increased chinking behavior. Finding an association of the D., DR gene with alcoholism21 is a step in that direction. Clearly, this study should be replicated by analyzing the brains of many other deceased alcoholics, and also by further studies of living alcoholics (and their pedigrees) in whom the alcoholism is unremitting and highly pernicious in nature.
In recent preliminary studies conducted with Blum and colleagues, we have found a significantiy higher prevalence of the Al allele of the D2DR gene in blood samples obtained from alcoholics and their relatives, compared to nonalcoholic controls. Moreover, in a recent report by Comings et al30 on 2392 individuals, the presence of severe problems with alcoholism and/or drug abuse was found in 14.4% of Tourette syndrome probands versus 4.4% of relatives of controls (/K. 0005). Furthermore, the Al allele of the D2DR was significantly associated with the alcoholic probands of Tourette syndrome compared to nonalcoholic controls (D.E. Comings, personal communication, 1990). Still, we must await additional studies to ascertain the role the D2DR gene plays in all forms of alcoholism.
If an association between the D9DR gene and alcoholism is confirmed, it should not be construed to mean that an "alcoholism gene" per se has been found. There is evidence that dopamine receptors are also involved in brain reward circuitry in cocaine dependence;"' and cocaine, like alcohol, is known to increase synaptic levels of dopamine in the brain.31 It is noteworthy that in a recent clinical study of 263 cocaine addicts, more than 50 had a family history of alcoholism, and more than 80% were alcoholics (The US Journal, September 1990; p. 9). Other studies (including one of our own) of cocaine dependent subjects confirm this observation.
Growing scientific evidence indicates that in addition to alcohol and cocaine, a variety of other substances with initial euphorigenic and reinforcing effects, such as opiates, amphetamines, phencyclidine (PCP), nicotine, caffeine, and even food, may be operating, in part, through the dopaminergic system.32 This raises the intriguing hypothesis of a genetic link, through the dopamine receptors, between certain heritable types of alcoholism and cocaine dependence, and perhaps even with the abuse of other psychoactive substances.
Even if such a hypothesis is tested and found to be valid, it is unlikely that the dopamine receptor gene(s) will be the only one(s) involved in the predisposition to alcoholism. After all, our own data show that a significant minority of alcoholics did not show the Al allele of the D9DR.21 This suggests that other genes or environmental factors may suffice to produce alcoholism. Such other genes may include those which encode elements related to the syntiiesis, disposition, or receptors of neurotransmitters or neuromodulators involved in the brain reward system. Among possible candidates are genes for the serotonergic, the adrenergic, and the opioid systems. Ultimately, research along these lines could lead to polygenic markers to detect the relative susceptibility of any individual with a family history of alcoholism.
Alcoholism is a disease of CNS dysfunction, with a large majority of alcoholics inheriting genes that predispose them to develop die disease. This complex human problem is just beginning to yield its secrets to puissant techniques recently developed for studying brain function and genes. The challenges are many and difficult; however, I have every expectation that, as we enter the 21st century, such studies will begin to identify individuals at the highest risk for developing the disease. This could lead to early targeted prevention approaches for those at risk. Moreover, through identification of predispositional genes and study of their biochemical expression, the development of specific pharmaceuticals for treating alcoholics becomes a realistic possibility. The hope and promise of this knowledge and its application will be great for the 18 million Americans afflicted with alcoholism and dieir 30 million children at risk. The benefits to our society could be enormous.
1. Williams RR. Hasstedt SJ, Wilson DE. et al. Evidence that men with familial hvpercbolesterolemia can avoid early coronary death: an analysis of 77 gene carriers in lenti Utah pedigrees. /AMA. 1986; 255:219-224.
2 Hopkins PN. Williams RR, Hunt SC. Magnified risk from cigarette smoking for coronary prone families in Utah. West /Med. 1984: 14 1:195-202.
3 Khavv K-T. Barre tt-Connor E. Family history of heart attack: a modifiable risk (actor? Circulation. 1986; 74:239-244.
4 Cotton NS. The familial incidence of alcoholism: a review. J Stud Alcohol. 1979; 40:89-116.
5 Goodwin DW, Schulsinger K, Hermaneen L, Giize SB, Winokur G, Alcohol problems in adoptees raised apart troni biological parents. Arch Cen Psychiatry, 1973:28:238-243.
5 Goodwin DW, Schulsinger F. Moller N1 Hermaneen L, Winokur G, Guze SB. Drinking problems in adopted and nonadopted sons of alcoholics. Arch Cen Psychiatry 1 974 ; 3 1 : 1 64- 1 69.
6 Schuckil MA. Twin studies on substance abuse: an overview. In: Gedda L, Palisi P, Nance W, eds. Twin Research 3. New York, NY: Alan R I.iss Ine; 1981:61-70.
7 Cloninger CR, Boh man M, Sigvardsson S. inheritance of alcohol abuse: fostering analysis of adopted men. Cen Psychiatry 1 98 1 ; 38:86 1 -868.
8 Schuckil MA. Studies of populations at high risk Ibi alcoholism. Psychiatr Den 1985:3:31-63.
9 Whipple SC, Parker ES, Noble EP. An atypical neurocognitive profile in alcoholic lathers and their sons. / Stud hol. 1988; 49:240-244.
10. Begleiter H. Porjesz B, Biliari B, Kissin B. Event-related potentials in boys at risk for alcoholism. Science. 1984; 225:14931496.
12. Parsons OA. Sinha R. Williams HL. tionships between neuropsychological lest performance and event-related potentials in alcoholic and nonalcoholic samples. Alcoholism, 1990: 14:746-755.
13. Moos RH, Moos BS. Family State Manual. 2nd ed. Palo Alto, Calif: Consulting Psychologist Press; 1986.
14. Noble EP. Alcoholic fathers and sons: neuropsychological, electrophysiol ogical, personali ly and family correlates. In: Cloninger RC, Begleiter H. Banbur Report 33: Genetics and Biology of Alcohol ism, ("old Spring Harbor, NY: Cold Spring Harbor Press: 1990:159-174.
15. Whipple SC, Noble EP. Personality acteristics of alcoholic fathers and sons. J Stud Alcohol In press.
16. Kornet-sky C, Bain GT, Unterwald Lewis MJ. Brain stimulation reward: fects of ethanol. Alcoholism. 1988: 12:609616.
17 Imperato A, Di Chiara G. Preferential stimulation of dopamine release in the nucleus accumbens of freely moving rats by ethanol. / Pharm Exp Ther. 1986: 239:219-228.
18. Bunzow JR. Van Toi HUM, Grandy DK, et al. (Honing and expression of a rat D, dopamine receptor cDNA. Nature. 1988; 330:783-787.
19. Grandy DK. Liu N. Allen I.. el al. The human dopamine D1, receptor gene is located on chromosome 1 I at q22-q23 and identifies a Taq1 RFLP. Am J Hum Genet. 1989; 45:778-785.
20. Cloninger RCl. Neurogenetic adaptive mechanisms in alcoholism. Science. 1987; 236:410-416.
21. Inviti M, Schuckit M, Smith TL. Clinical importance of age of onset in type 1 and type 2 primary alcoholics. Arch (ien Psychiatry. 1999; 47:320-324.
22. Penick EC. Powell BJ, Nickel EJ, Read MR. Gabrielli WF, Liskow Bl. Examination of Cloninger's type I and type II alcoholism with a sample of men alcoholics in treatment. Alcoholism. 1999: 14:623629.
28. Nixon SJ. Parsons OA. Application of the tridimensional pei sonali ? ? questionnaire lo a population of alcoholics and other substance abusers. Alcoholism. 1990; 14:51 3-5 17.
24. Blum K. Noble EP. Sheridan PJ, el al. Allelic association of human dopamine D., receptor gene in alcoholism. JAMA. 1 990: 263:2055-2060.
25. Waflbrd KA, Burnett DM, Dinwiddie TV, Harris RA. Genetic differences in the ethanol sensitivity of GABAA receptors expressed in Xenopus oocytes. Science. 1990;249:291-293.
26. Hatada S. Agarwal DP. Goedde HW Aldehyde dehydrogenase polymorphism and alcohol metabolism in alcoholics. Alcohol. 1985;2:391-392.
27. Reulei JB, Girant DE, Coon ey TG. Wernicke's encephalopathy. N Engl f Med. 1985: 312:1035-1039.
28. limbec /., Omenti GS. Evidence of genetic predisposition to alcoholic cirrhosis and psychosis: twin concordances for alcoholics and its biological end points by zygosity among male veterans. Alcoholism. 1981;5:207-215.
29. Werner FR. Eskreis DS, Compton KV. Oi rejo H, Zeni MA. Haplotype analysis of a type I collagen gene ami its association with alcoholic cirrhosis in man. MoI Aspects Med. 1988; 10:159-168.
30. Comings V)Y., Muhleman D. Diet/ O, Shahbabrami B, Tast D. Kovacs BW The dopamine D2 receptor gene is a modifier of the expression of lhe Tonrette syndrome (GTs) and ADHD gene. A in J Hum Genet. 1990: 47(suppl):A52.
31. Heikkila RE. Orlansky H. Cohen G. Studies on the distribution between uptake inhibition and release of ['1H !dopamine in ral brain tissue slices. Biochem Pharmacol. 1975;24:847-852.
32. Wise RA. The role of reward pathways in the development of drug dependence;. Pharmacol Vier. 1987; 35:227-263.