¹Department of Psychiatry, 'Ramón y Cajal' Hospital, Alcala University, Madrid, Spain

²Genetics Unit, Department of Biology, Autonoma University, Madrid, Spain

³Department of Psychiatry, Columbia University, New York State Psychiatric Institute, New York, USA

Correspondence to: A Ibañez, Servicio de Psiquiatría, Hospital Ramón y Cajal, 28034 Madrid, Spain. angela.ibanez@globalmed.es

This study was conducted to detect a possible association of MAOA and/or MAOB genes with pathological gambling (PG). DNA polymorphisms in MAOA and MAOB genes were screened by molecular analysis in 68 individuals (47 males and 21 females) meeting ICD-10 and DSM-IV criteria for pathological gambling and 68 healthy comparison controls matched for age and sex. There were no significant differences between pathological gamblers and healthy volunteers in overall allele distribution at the MAOA gene polymorphism. However there was a significant association between allele distribution and the subgroup of severe male gamblers (n = 31) compared to the males in the group of healthy volunteers (² = 5246; df = 1; P < 0.05 [bonferroni corrected]). No association was found between the MAOB polymorphic marker and PG. Allele variants at the MAOA, but not the MAOB gene may be a genetic liability factor in PG, at least in severe male gamblers. Molecular Psychiatry(2000) 5, 105-109.

pathological gambling; genetics; monoamine-oxidase genes; MAOA gene; MAOB gene; impulse control disorders

Pathological gambling is an impulse control disorder and it has also been proposed as a model of addiction without substance.¹ As defined by DSM-IV its essential feature is a persistent and recurrent maladaptive gambling behaviour. It is a progressive, chronic and highly disabling disorder that is poorly understood and often underdiagnosed despite affecting 1-3% of the adult population.² At present, little is known about the biological correlates of pathological gambling. A growing literature suggests the involvement of genetic factors in behavioral disorders related to pathological gambling such as alcoholism, substance abuse, attention-deficit-hyperactivity disorder, and smoking.³ An increased incidence of about 20% of pathological gambling in first-degree relatives within clinical samples of pathological gamblers has been reported^4,5 and it has led to consideration of the possible role of a genetic component in the development of this disorder. Moreover, there is evidence for genetic influence derived from a large twin study performed on 3359 twin pairs in United States. This study revealed that inherited factors explained 62% of the diagnosis of pathological gambling disorder.⁶ Recently, a positive association has been reported between pathological gambling and DNA polymorphism at the D2 and D4 receptor genes.^7,8

Monoamine oxidases A (MAOA) and B (MAOB) play a critical role in the degradation of several neurotransmitters which could be involved in the pathogenesis of pathological gambling. Decreased platelet MAOB activity has been reported in pathological gambling^9,10 and other impulse control disorders.^11,12 Although MAOA activity has never been studied in pathological gambling, previous research suggests that abnormal MAOA activity may play a role in the pathophysiology of disorders with impaired impulse control.¹³ Decreased MAOA activity was associated with impulsive behaviors in several affected males from a large Dutch family with a mutation in that locus.¹⁴ Differences in MAOA activity have been associated with specific alleles of the structural gene.¹⁵ Polymorphic variants at MAOA and/or MAOB genes have been reported to be associated with alcoholism, substance abuse and other impulse control disorders.^16,17,18 In light of these findings we sought to investigate the possibility of an association between genetic variants at MAOA and/or MAOB genes and pathological gambling.

The patient group consisted of 68 unrelated pathological gamblers (47 men and 21 women) with a mean age of 42.24 years (SD = 13.45). We found psychiatric comorbidity in 62% of patients, personality disorders being the most common psychiatric diagnosis associated (44.1%), followed by alcohol abuse/dependence (33.8%). The mean South Oaks Gambling Screen (SOGS) score was 12.65 (SD = 2.60; range 6-17) and the Global Clinical Impression (GCI) was in the severe range (4 = severe or 5 = very severe disorder) in 63% of patients, all of them with a SOGS score of 12. Family history of pathological gambling in first-degree relatives was found in 14 of 68 patients (20.5%) and in 25 of 68 (36.8%) when second-degree relatives were included. Control individuals were 68 unrelated and unaffected blood donors (47 men and 21 women) with a mean age of 42.9 years (SD = 12.8).

The allele distribution of MAOA polymorphism analyzed is shown in Table 1. Only three (A, B, and C) of the five different alleles described by Hinds et al¹⁹ were found in our sample. Genotype frequencies were distributed according to Hardy-Weinberg equilibrium both in pathological gamblers (² = 0.43; df = 1; P = 0.51) and controls (² = 2.84; df = 2; P = 0.41). Allele A was only found in one control female; therefore this subject was excluded from the global analysis in order to preserve statistical power as suggested by other authors.²⁰ We found differences in overall allele distribution, but they did not reach statistical significance (² = 2.354; df = 1; P = 0.085). When stratifying by gender, similar differences in overall allele distribution were found in males (² = 2.443; df = 1; P = 0.090) but not in females (² = 0.329; df = 1; P = 0.370). Interestingly, allele B was more represented in most severe cases ('severe' or 'very severe' in GCI) than in healthy volunteers (² = 4.254; df = 1; P = 0.031). However, differences were not significant when Bonferroni correction was applied. Nevertheless, when stratifying by gender, the most severe male pathological gamblers showed significant differences in allele distribution compared with male controls (² = 5.246; df = 1; P = 0.021 [P < 0.05, Bonferroni corrected]). No differences were found in females.

Of the genotypes, considering male hemizygotes and female homozygotes together, the most frequent genotype was 'C or C/C' (0.687 in controls and 0.573 in patients), followed by 'B or B/B' (0.179 in controls and 0.309 in patients) and 'B/C' only found in females (0.134 in controls and 0.118 in patients). No differences in overall genotype distribution were found between gamblers and controls (² = 3.083; df = 2; P = 0.214). Genotype 'B or B/B' was identified in 17 of the 43 most severe patients and in 12 of the 67 controls; differences were significant only in males (² = 5.246; df = 1; P < 0.05; Bonferroni corrected), but not in females (² = 1.340; df = 2; P = 0.512).

Table 2 shows the allele distribution of MAOB polymorphism analyzed.²¹ We found eight different alleles in our sample corresponding to one, two, three, four, five, six, seven and 11 repeats of the (GT)_n sequence. The seven-repeat allele was only found in one pathological gambler and the 11-repeat allele was only found in one control. Therefore they were grouped with the six-repeat allele into one category as it has been suggested in the analysis of other polymorphic genetic markers with more than five alleles.²² The genotype frequencies were distributed according to Hardy-Weinberg equilibrium both in controls (² = 14.66; df = 15; P = 0.47) and in pathological gamblers (² = 12.20; df = 15; P = 0.66). Data were grouped in 2 ´ 6 contingency tables to compare allele distribution and in 2 ´ 2 contingency tables to compare the frequencies of each allele against the others combined in a single allele class. As described above, male hemizygotes and female homozygotes were considered together for the purpose of genotypic analyses. No statistically significant differences were found either in total samples or in a subgroup of more severe pathological gamblers compared with controls either in allele distribution or in genotypic analyses.

Our findings of no association between specific alleles at MAOB intron 2 polymorphism and pathological gambling seem to indicate that decreased platelet MAOB activity reported in pathological gambling^9,10 may not be related to structural variants of the MAOB gene. This finding does not exclude a genetic influence in the low MAOB activity observed in pathological gamblers, since decreases in enzymatic activity may be due to alterations in regulatory DNA sequences.²³ Moreover, previous research has failed to identify an association between the number of GT(n) repeat units at polymorphic sequence in MAOB intron 2 and the level of platelet MAOB activity, suggesting that structural alterations at the gene for MAOB are not the primary determinant of activity levels in platelets.²⁴

Although MAOA activity appears to be more important than MAOB in metabolism of neurotransmitters in the brain, little research has been directed towards investigating the behavioral correlates of MAOA activity, probably due to the limited accessibility to MAOA in peripheral tissues (only in cultured skin fibroblasts) compared to MAOB.²⁵ A mutation in the MAOA gene that leads to lack of MAOA activity has been described in several affected males of a single large Dutch kindred, that exhibited borderline mental retardation and increased impulsive behaviours.¹⁴ In this study, we did not find an association between MAOA intron 1 polymorphic marker and the total group of pathological gamblers, although near significant differences in allele distribution were found when only males were considered. However, allele 'B' was found more frequently in the most severe pathological gamblers, and differences reached statistical significance in male severe gamblers compared to male comparison subjects. Nevertheless, it should be pointed out the risk of achieving a false-positive result with the traditional significance level of 0.05, given the low prior probability of selecting the 'right' candidate gene for association studies in behavioral traits.²⁶ However, interestingly we found the most positive findings in the most severe cases; these results could be in agreement with the polygenic model of genetic susceptibility to psychiatric disorders.

Recently, Sabol et al have reported a new MAOA polymorphism located in the gene promoter.²⁷ This polymorphism has been shown to affect the transcriptional activity of the MAOA gene promoter. The high activity promoter alleles have been associated with panic disorder in females.²⁸ It is worth noting that this functional polymorphism is in linkage disiquilibrium with other MAOA and MAOB markers.²⁷ Moreover, we have found the same linkage disiquilibrium with the MAOA polymorphism analyzed in this study, and this evidence could suggest a positive association between the low activity MAOA gene promoter alleles and pathological gambling (manuscript in preparation).

As mentioned above the association between allele B of MAOA polymorphism and more severe pathological gamblers was observed in males, but not in females. Other genetic gender differences in pathological gambling have been reported, such as the association between DRD4-7 repeat allele and female gamblers⁸ and between the short variant of 5-HTTLPR and male gamblers.²⁹ Interestingly in both cases the allele variants associated with pathological gambling lead to a poorer functioning of the expressed gene product.^30,31 Further research is needed to investigate the possible contribution of genetic factors in the pathogenesis of pathological gambling and to elucidate if this contribution could be gender related. Interestingly, there are significant clinical differences between male and female pathological gamblers. Male pathological gamblers have on average an earlier age of onset, higher prevalence of alcohol-related disorders, and more psychopathic traits than female gamblers. In contrast, female pathological gamblers tend to obtain higher scores on depression scales at admission, meet criteria for depressive disorders and have more dependency traits than male pathological gamblers.³² It is unclear to what extent those clinical differences may be related to genetic factors.

In summary, the results of this study suggest a possible association between the MAOA VNTR intron 1 polymorphism and liability to pathological gambling in males, at least in the most severe cases. However, the sample size is a limitation of the study and further research in larger and different samples is needed to confirm this positive association and to elucidate the functional meaning of this finding.

Methods

Subjects

Sixty-eight patients (47 males and 21 females) admitted to the Pathological Gambling Outpatient Program of 'Ramon y Cajal' Hospital (Madrid, Spain) were included in this study. All subjects met ICD-10³³ and DSM-IV² diagnostic criteria for pathological gambling. The clinical assessment of the patients was performed by a senior psychiatrist (AI) who was blind to their genotypes. Five items-Global Clinical Impression (GCI) and South Oaks Gambling Screen (SOGS)³⁴ were used to assess the severity of pathological gambling. Psychiatric comorbidity in axis I was assessed by clinical interview based on DSM-IV diagnostic criteria. Structured Clinical Interview for Personality Disorders³⁵ was used to assess axis II comorbidity. Family history was assessed according to FH-RDC.³⁶

An equal number of comparison subjects, matched for age and sex, were selected from the Blood Donation Unit of 'Ramon y Cajal' Hospital. People attending this Unit, as well as pathological gamblers, were non-paid voluntary subjects. A senior psychiatrist (AI) in the team interviewed all subjects in the comparison group to exclude subjects with personal or family history of psychiatric disorders, including problem gambling (score of zero in SOGS was required), alcohol and/or substance abuse, and other impulse-control disorders.

All patients and comparison subjects were Caucasian and living in Central Spain. All patients and healthy volunteers signed written informed consent before their inclusion in the study.

DNA amplification and genotyping

All patients and comparison subjects were screened by molecular analysis of specific DNA polymorphisms at the MAOA and MAOB genes. DNA was isolated from 10 ml of peripheral blood samples according to standard methods. DNA was amplified using polymerase chain reaction (PCR). For the MAOA gene we studied a VNTR marker located in the first intron of the gene; PCR amplification was made using the primers and conditions described by Hinds et al.¹⁹ For the MAOB gene we analyzed a dinucleotide sequence (GT)n in the second intron of the gene.²¹ Amplified products obtained by PCR were resolved by electrophoresis in denaturing polyacrylamide gels and detected by silver staining.

Statistical analyses

Before analyzing the data, chi-square goodness-of-fit test was used to verify Hardy-Weinberg equilibrium of genotype frequencies in the patients as well as in the comparison group. Data were analyzed in two different ways for each marker as suggested by Craddock et al.³⁷ First, we used a chi-square omnibus test to determine whether there was an overall association between pathological gambling and allele distribution at the marker locus. Data were tabulated as an a ´ 2 table where a is the number of alleles at the marker locus. The second analysis was aimed at uncovering possible associations between pathological gambling and each marker separately. For this analysis data were tabulated as a series of 2 ´ 2 tables in which all alleles except the one being studied were collapsed into a single allele class in the table. Chi-square tests were used to investigate the association between gambling status and individual alleles. A correction for multiple testing was applied, following the technique described by Walsh et al.³⁸ For a conventional significance value of 0.05, the critical P value was calculated from the formula: [1 - (1 - P) a = 0.05], and consequently P = [1 - (0.95)1/a], where a is the number of alleles at the marker locus. Of the genotypes, since the MAOA and MAOB genes are located in the X chromosome and males have only one allele for this gene, we considered male hemizygotes and female homozygotes together for the purpose of genotypic analyses. Then chi-square tests were used to analyze the association between gambling status and genotypes. Bonferroni correction for multiple testing was applied when analyzing subgroups of gamblers.

1 Dickerson M. Gambling: a dependence without a drug. Int Rev Psychiatry 1989; 1: 152-172,

2 American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th edn. APA: Washington DC, 1994,

3 Blum K, Sheridan PJ, Wood RC, Braverman ER, Chen TJ, Comings DE. Dopamine D2 receptor gene variants: association and linkage studies in impulsive-addictive-compulsive behaviour. Pharmacogenetics 1995; 5: 121-141, MEDLINE

4 Lesieur HR. The female pathological gambler. In: Eadington WR (ed). Gambling Research (vol 5). Bureau of Business & Economic Research, University of Nevada: Reno, 1988,

5 Saiz-Ruiz J, Moreno I, Lopez-lbor JJ. Ludopatia: estudio clínico y terapéutico-evolutivo de un grupo de jugadores patológicos. Actas Luso-Españolas de Neurologia, Psiquiatría y Ciencias Afines 1992; 20: 189-197, MEDLINE

6 Eisen SA, Lin N, Lyons MJ, Scherrer JF, Griffith K, True WR et al. Familial influences on gambling behavior: an analysis of 3359 twin pairs. Addiction 1998; 93: 1375-1384, MEDLINE

7 Comings DE, Rosenthal RJ, Lesieur HR, Rugle LJ, Muhleman D, Chiu C et al. A study of the dopamine D2 receptor gene in pathological gambling. Pharmacogenetics 1996; 6: 223-234, MEDLINE

8 Perez de Castro I, Ibañez A, Torres P, Saiz-Ruiz J, Fernandez-Piqueras J. Genetic association study between pathological gambling and a functional DNA polymorphism at the D4 receptor. Pharmacogenetics 1997; 7: 345-348, MEDLINE

9 Carrasco JL, Saiz-Ruiz J, Hollander E, Cesar J, Lopez-lbor JJ Jr. Low platelet monoamine oxidase activity in pathological gambling. Acta Psychiatr Scand 1994; 90: 427-431, MEDLINE

10 Blanco C, Oresanz-Muñoz L, Blanco-Jerez C, Saiz-Ruiz J. Pathological gambling and platelet MAO activity: a psychobiological study. Am J Psychiatry 1996; 153: 119-121, MEDLINE

11 Fowler CJ, Tipton KF, MacKay AVP, Youdim MBH. Human platelet monoamine oxidase¾a useful enzyme in the study of psychiatric disorders? Neuroscience 1982; 7: 1577-1594, MEDLINE

12 Gottfries CG, Oreland L, Wiberg A, Winblad B. Lowered monoamine oxidase activity in brains from alcoholic suicides. J Neurochem 1985; 25: 667-673,

13 Lenders JWM, Brunner HG, Murphy DL, Eisenhofer HG. Genetic deficiencies of monoamine oxidase enzymes: a key to understanding the function of the enzymes in humans. Adv Pharmacol 1998; 42: 297-301, MEDLINE

14 Brunner HG, Nelen M, Breakefield XO, Ropers HH, Van Oost BA. Abnormal behavior associated with a point mutation in the structural gene for monoamine oxidase A. Science 1993; 262: 578-580, MEDLINE

15 Hotamisligil GS, Breakefield XO. Human monoamine oxidase A gene determines levels of enzyme activity. Am J Hum Genet 1991; 49: 383-392, MEDLINE

16 Parsian A, Suarez BK, Tabakoff B, Hoffman P, Ovchinnikova L, Fisher L et al. Monoamine oxidases and alcoholism. I. Studies in unrelated alcoholics and normal controls. Am J Med Genet 1995; 60: 409-416, MEDLINE

17 Vanyukov MM, Moss HB, Yu LM, Tarter RE, Deka R. Preliminary evidence for an association of a dinucleotide repeat polymorphism at the MAO-A gene with early onset alcoholism/substance abuse. Am J Med Genet 1995; 60: 122-126, MEDLINE

18 Devor EJ, Abell CW, Hoffman PL, Tabakoff B, Cloninger CR. Platelet MAO activity in type I and type II alcoholism. Ann NY Acad Sci 1994; 708: 119-128, MEDLINE

19 Hinds HL, Hendriks RW, Craig IW, Chen ZY. Characterization of a highly polymorphic region near the first exon of the human MAOA gene containing a GT dinucleotide and a novel VNTR motif. Genomics 1992; 13: 896-897, MEDLINE

20 Lim LCC, Powell JF, Murray R, Gill M. Monoamine oxidase A gene and bipolar affective disorder. Am J Hum Genet 1994; 54: 1122-1124, MEDLINE

21 Konradi C, Ozelius L, Breakefield XO. Highly polymorphic (GT)_n repeat sequence in intron II of human MAOB gene. Genomics 1992; 12: 176-177, MEDLINE

22 Lim LCC, Powell J, Sham P, Castle D, Hunt N, Murray R et al. Evidence for a genetic association between alleles of monoamine oxidase A gene and bipolar affective disorder. Am J Med Genet 1995; 60: 325-331, MEDLINE

23 Hyman SE. Relevance of gene regulation to psychiatry. In: Review of Psychiatry, Volume 15, Section III¾Recent advances in Neuroscience Relevant to Psychiatry. American Psychiatric Press, Washington, 1996, pp 311-330.

24 Girmen AS, Baenziger J, Hotamisligil GS, Konradi C, Shalish C, Sullivan JL et al. Relationship between platelet monoamine oxidase B activity and alleles at the MAOB locus. J Neurochem 1992; 59: 2063-2066, MEDLINE

25 Lenders JW, Eisenhofer G, Abeling NG, Berger W, Murphy DL, Konings CH et al. Specific genetic deficiencies of the A and B isoenzymes of monoamine oxidase are characterized by distinct neurochemical and clinical phenotypes. J Clin Invest 1996; 97: 1010-1019, MEDLINE

26 Faraone SV, Tsuang MT. Methods in Psychiatric Genetics. In: Tsuang MT, Tohen M, Zahner GEP (eds). Textbook in Psychiatric Epidemiology. John Wiley & Sons (Wiley-Liss): New York, 1995, 81-134.

27 Sabol SZ, Hu S, Hamer D. A functional polymorphism in the monoamine oxidase A gene promoter. Hum Genet 1998; 103: 273-279, Article MEDLINE

28 Deckert J, Catalano M, Syagailo YV, Bosi M, Okladnova O, Di Bella D et al. Excess of high activity monoamine oxidase A gene promoter alleles in female patients with panic disorder. Hum Mol Genet 1999; 8: 621-624, Article MEDLINE

29 Perez de Castro I, Ibañez A, Saiz-Ruiz J, Fernandez-Piqueras J. Genetic contribution to pathological gambling: possible association between a functional DNA polymorphism at the serotonin transporter gene (5HTT) and affected men. Pharmacogenetics 1999; 9: 397-400, MEDLINE

30 Van Tol HH, Wu CM, Guan HC, Ohara K, Bunzow JR, Civelli O et al. Multiple dopamine D4 receptor variants in the human population. Nature 1992; 358: 149-152, MEDLINE

31 Heils A, Teufel A, Petri S, Stober G, Riederer P, Bengel D et al. Allelic variation of human serotonin transporter gene expression. J Neurochem 1996; 66: 2621-2624, MEDLINE

32 Ibañez A. Bases Genéticas de la Ludopatía. Doctoral Thesis, Alcala University, Madrid, Spain, 1997,

33 World Health Organization. The ICD-10 Classification of Mental and Behavioural Disorders: Diagnostic Criteria for Research. World Health Organization, 1993,

34 Lesieur HR, Blume SB. The South Oaks Gambling Screen (SOGS): a new instrument for the identification of pathological gamblers. Am J Psychiatry 1987; 144: 1184-1188, MEDLINE

35 Spitzer R, Williams J. Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II, 7/1/85). Biometrics Research Department, New York State Psychiatric Institute: New York, 1985,

36 Andreasen NC, Endicott J, Spitzer RL, Winokur G. The family history method using diagnostic criteria. Reliability and validity. Arch Gen Psychiatry 1977; 34: 1229-1235, MEDLINE

37 Craddock N, Daniels J, Roberts E, Rees M, McGuffin P, Owen MJ. No evidence for allelic association between bipolar disorder and monoamine oxidase A gene polymorphisms. Am J Med Genet 1995; 60: 322-324, MEDLINE

38 Walsh C, Hicks AA, Sham P, Castle D, Hunt N, Clements A et al. GABA-A receptor subunit genes as candidate genes for bipolar affective disorder. An association analysis. Psychiatry Genet 1992; 2: 239-247,

Table 1 Distribution of MAOA alleles in pathological gamblers and the comparison group

Table 2 Distribution of MAOB alleles in pathological gamblers and the comparison group