Molecular Psychiatry (1998) 3, 337–341  1998 Stockton Press All rights reserved 1359–4184/98 $12.00 ORIGINAL RESEARCH ARTICLE Dopamine D 3 receptor gene variants and substance abuse in schizophrenia M-O Krebs 1 , F Sautel 2 , M-C Bourdel 1 , P Sokoloff, 2 J-C Schwartz 2 , J-P Olie ´ 1 , H Lo ˆo 1 and M-F Poirier 1 1 Service Hospitalo-Universitaire de Sante ´ Mentale et The ´ rapeutique, Ho ˆ pital Sainte-Anne, Universite ´ Rene ´ Descartes (Paris V), Paris; 2 Unite ´ de Neurobiologie et Pharmacologie (U109, INSERM), Paris, France Keywords: association study; drug abuse; drug depen- dence; schizophrenia; comorbidity In an association study of the Bal I polymorphism in the dopamine D 3 receptor (DRD3) gene in a French Cauca- sian population, global comparison of patients with schizophrenia (n = 89, DSM-III-R criteria) and controls (n = 52) led to non-significant differences. However, the homozygosity was significantly more frequent in schizophrenic patients with lifetime substance abuse comorbidity (n = 36) as compared to patients with no history of substance abuse (P = 0.010) or to controls (P = 0.047) and in neuroleptic responder patients as com- pared to treatment-refractory patients (n = 19; P = 0.037). The combined characteristics treatment response and lifetime substance abuse were strongly associated with homozygosity. We propose that homozygosity for the Bal I polymorphism DRD3 gene is associated with pre- disposition to substance abuse and/or the pharmaco- sensitive characteristic of schizophrenia rather than with schizophrenia itself, an hypothesis in agreement with the positive association of this polymorphism with opiate dependence (see companion article by Duaux et al) and the involvement of DRD3 in both pharmacodep- endence mechanisms and antipsychotic effects of neur- oleptics. The dopamine D 3 receptor (DRD3) is selectively expressed in the mesocorticolimbic dopaminergic sys- tem, thought to be implicated in the pathophysiology of schizophrenia, and is a target for both typical and atypical neuroleptics. 1 The DRD3-binding site is increased in post-mortem brains of untreated patients with schizophrenia. 2 A positive association between homozygosity for the Bal I polymorphism at DRD3 gene and schizophrenia has been found in two Euro- pean independent populations 3 but not replicated in most subsequent studies. The reasons for these discrep- ancies may lie in the complexity and phenotypic het- erogeneity of schizophrenia. 4 A high prevalence (about 30–40%) of lifetime sub- stance abuse exists among patients with schizo- phrenia, 5 especially in the premorbid phase or during the first years of the disease. 6 Abused substances (alcohol, heroin and cocaine) share the property to unconditionally evoke dopamine release in the shell of nucleus accumbens, 7 where DRD3 is selectively expressed. 1 In addition, pharmacological evidence in rats 8 and monkeys 9 suggests that DRD3 mediates rein- forcing effects of psychostimulants. Moreover, DRD3 binding is elevated in the post-mortem human brain from cocaine overdose victims. 10 Therefore, the DRD3 gene appears as a candidate gene not only for schizo- phrenia, but also for alcoholism and/or drug addiction, in which the influence of genetic factors has also been established. 11 Association studies in alcoholism have led to contradictory results 12–14 but, in an accompanying paper, homozygozity at the DRD3 gene Bal I polymorphism was associated with opiate abuse. 15 The present study reports a positive associ- ation of homozygozity of DRD3 gene at Bal I polymor- phism in a French sample of schizophrenic patients with current or past substance abuse comorbidity and in patients who respond to neuroleptic treatment. Eighty-nine unrelated in- or out-patients with schizophrenia (n = 83) or schizo-affective disorder (n = 6, DSM-III-R criteria, APA 1987) and fifty-two unre- lated controls, mostly from medical staff, with no per- sonal and/or familial psychiatric or substance abuse history, as assessed with a semi-standardized interview (see Methods), were recruited from a French Causasian population. Patients who had at least once in their life fulfilled the DSM-III-R criteria for abuse and/or depen- dence on any kind of psychoactive substance except nicotine (but not occasional users) were defined as ‘substance-abusing’ patients (see Table 1, n = 36). Nine- teen patients had abuse of or dependence on more than one substance, alcohol and/or cannabis being the most frequent. Patients with no clinical remission despite several trials with different neuroleptics for a sufficient duration and requiring permanent day care were defined as ‘treatment refractory’ patients, while the remaining patients, who at least partly responded to neuroleptics were defined as responders. Genotype distributions in either controls or patients did not depart from the Hardy–Weinberg equilibrium. The total sample of schizophrenic patients (n = 89) did not significantly differ from controls (n = 52) with regard to allele frequencies ( 2 = 0.004, d.f. = 1, P = 0.95), genotype distribution ( 2 = 0.4, d.f. = 1, P = 0.53) Table 1 Substances used by patients with schizophrenia Substances Number of patients with abuse or dependence Alcohol 21 Cannabis 20 Opiate 5 Sedative 5 Hallucinogen 4 Cocaine or psychostimulant 3 Inhalant 2 Nineteen patients had abuse of and/or dependence on more than one substance.