Molecular Psychiatry (2000) 5, 270–274  2000 Macmillan Publishers Ltd All rights reserved 1359-4184/00 $15.00 www.nature.com/mp ORIGINAL RESEARCH ARTICLE D2 Ser/Cys 311 variant is associated with delusion and disorganization symptomatology in major psychoses A Serretti, E Lattuada, C Lorenzi, R Lilli and E Smeraldi

Istituto Scientifico Ospedale San Raffaele, Department of Neuropsychiatric Sciences, University of Milan School of Medicine, Italy

The D2 receptor (DRD2) is a binding site of many psychoactive drugs and it has been pro- posed as a genetic risk factor for psychiatric disorders. The aim of this investigation was to study the DRD2 S311C variant in major psychoses. We studied 1182 inpatients with diagnoses of bipolar disorder (n = 480), major depressive disorder (n = 269), (n = 366), delusional disorder (n = 44), psychotic disorder not otherwise specified (n = 23) and 267 heal- thy controls. Eight hundred and eighty-seven subjects were also scored for their lifetime symptomatology using the the Operational Criteria checklist for psychotic illness (OPCRIT). DRD2 variants were not associated with affected subjects even when possible confounders like gender and onset were considered. When we considered the 887 subjects with the symp- tomatologic analysis, we observed a significant association of the DRD2 S311C variant with both delusion and disorganization features. The association was present independently from diagnoses. Our results do not show that coding variants of the DRD2 S311C play a major role in conferring susceptibility to major psychoses, but they may be connected with disorganized and delusional symptomatology independently from diagnoses. Molecular Psychiatry (2000) 5, 270–274. Keywords: dopamine receptors; linkage disequilibrium; psychopathology; bipolar disorder; schizo- phrenia; depressive disorder; paranoid disorder

Disturbances of dopamine transmission have been a genetic risk factor for schizophrenia following the implicated in the pathogenesis of psychiatric dis- observation of an excess of S311C variants among orders. Inhibitors of dopamine reuptake, like schizophrenics.5 Nevertheless a number of replications amineptine, have an antidepressant action; dopamine failed to confirm the initial report,6–14 with only one , such as l-dopa, or bromocrip- replication in Caucasian schizophrenics.15 Mood dis- tine have psychotomimetic effects in schizophrenics. orders have also been investigated. An increased den- Conversely, neuroleptics have properties sity has been reported of D2 or D2-like receptors in the and can induce depression in affective patients caudate of bipolar patients with psychotic symptoms, through their blockade of D2 receptors.1 compared with both normal controls and nonpsychotic The D2 receptor is the major binding site of many bipolar patients.16 Moreover in the same chromosomal typical and atypical neuroleptic drugs and its has region a translocation has been observed cosegregating been cloned and mapped to 11q22–23.2 Itokawa et al3 with bipolar disorder.17 Knockout studies suggested a reported a polymorphism causing a structural change behavioral activity of DRD2. Balk et al18 used homolo- from Serine to Cysteine at codon 311 of DRD2 (S311C); gous recombination to generate D2-receptor-deficient this substitution occurs in the third intracellular loop mice. Absence of D2 receptors led to animals that were of the receptor. Preliminary results showed that DRD2 akinetic and bradykinetic in behavioral tests and rarer TaqI variants were associated with higher levels showed significantly reduced spontaneous movements. of the monoamine metabolites homovanillic acid and There is, therefore, the possibility that variants of the 3-methoxy-4-hydroxyphenylglycol in lumbar cerebro- DRD2 receptor may elicit symptomatologic effects. spinal fluid in controls.4 S311C has been proposed as Thus, it has been suggested that the influence of DRD2 variants should be analyzed more at the level of symp- toms rather than syndromes.19 Up to now both linkage Correspondence: A Serretti MD, Dept of Neuroscience, Istituto and association studies failed to detect any association Scientifico H San Raffaele, University of Milano School of Medi- of DRD2 with mood disorders.20–28 A small positive cine, Via Luigi Prinetti 29, 20127 Milano, Italy. E-mail: serretti. 29 alessandroȰhsr.it association was reported in a Spanish sample but it Received 28 May 1999; revised 22 November 1999; accepted 19 failed to be replicated and a positive association was December 1999 also found by Arinami et al,19 but only when consider- DRD2 and major psychoses A Serretti et al 271 ing mood disorder subjects with mood incongruent The possible association of DRD2 with symptoms psychotic features. was suggested by the observation of Arinami et al19 that Traditionally the definition of the phenotype has the frequency of individuals with DRD2 variant was been based on categorical diagnoses which do not significantly higher in mood disorders with mood guarantee biological homogeneity and have been con- incongruent psychotic symptoms, but similar in mood sidered reductive of the amount of information avail- disorders without psychotic features when compared able from the whole symptomatologic presentation of with the controls.19 Following this hypothesis, a symp- patients.30–32 Moreover, converging evidence indicates tomatologic analysis has been performed using SADS that susceptibility only elicit limited symp- items on a small sample of schizophrenics and no tomatologic effects on psychiatric syndromes.33–35 We association was detected.53 We did not replicate the have, therefore, developed a phenotype definition that finding in a preliminary report using the DRD2 VNTR is based on four symptomatological factors: Excite- marker in a sample of 47 bipolar subjects.54 However, ment, Depression, Delusion and Disorganization.36 Our we reported and confirmed an association of DRD2* results are coherent with previous studies performed S311C with disorganized symptomatology in a large both in schizophrenia and mixed psychotic samples.37–41 sample of 267 schizophrenics and delusional sub- Family studies evidenced a moderate but significant jects.51,52 This association may explain previous con- heritability for all psychopathological factors.42–48 The flicting results where disorganized symptomatology factors are stable over time49 and they are correlated was not considered and a variable rate of disorganized with regional anatomical abnormalities.50 subjects may have influenced DRD2 frequencies. We have previously detected and replicated an In the present article we used a phenotype definition association of the DRD2 S311C variant with disor- that is based on the lifetime excitement, depressive, ganized symptomatology of schizophrenia.51,52 In the delusional and disorganized symptomatology.36 This is present study we investigated the possible influence of an alternative to the traditional use of psychiatric diag- the DRD2 S311C polymorphisms in major psychoses. noses to define the affected phenotype for genetic stud- For DRD2 S311C polymorphism, bipolars, major ies.30,31 However, the use of symptoms for phenotype depressives, schizophrenics and delusionals did not definition may be considered a limitation as symp- show any overall genotypic association (Table 1). tomatologic presentation may vary across illness epi- When considering 2 × 2 tables a trend was observed sodes. We used a lifetime scoring perspective includ- toward an excess of DRD2 S311C among schizo- ing ever-presented symptoms to reduce this bias, and phrenics (␹2 = 4.29; df = 1; P = 0.038) as compared to follow-up studies evidenced that episodes are substan- controls. The association drops to non significance tially stable over time.55 We are currently collecting fol- when corrected for multiple testing. Early age of onset low-up data to further reduce this bias. (Ͻ25 years) and gender did not modify DRD2 variant Ethnic origin is also a cause of stratification bias but frequencies significantly. Control subjects and patients our sample was composed of subjects mainly collected in all diagnostic groups were in Hardy–Weinberg equi- in the North of Italy with Italian antecedents for at least librium. two generations. Although genetic heterogeneity has When we considered the 887 subjects with the symp- been evidenced for some isolated populations (such as tomatologic analysis, we observed a significant associ- Sardinia, not included in our sample), Italy is charac- ation of the DRD2 S311C variant with both delusion terized by a substantial genetic homogeneity.56 Con- and disorganization features (Table 2). The association trols were screened for the absence of mental illnesses. was present independently from diagnoses and gender. Although this may introduce a bias on allelic fre- Severity of illness is directly associated with symptom quencies, association studies comparing very common scores, and this could constitute a bias for the detection diseases like major depression should exclude these of liability factors. Then we repeated the calculation cases from the control sample, as they may obscure including clinical indicators of severity as covariants positive associations. in turn. We considered the age of onset, current age and The use of drugs could bias our ability to detect length of disease. None of those covariates significantly naive symptomatology, but this should not constitute influenced the results (data not shown). a major obstacle.36,57 Specifically, while most subjects

Table 1 Crosstabulation of DRD2 S311C genotypes and diagnoses. Frequencies were not different from controls

Genotypes Control Schizophrenia Bipolar Major Delusional Psychotic Totals sample disorder depressive disorder NOS disorder disorder

(n) Freq. (n) Freq. (n) Freq. (n) Freq. (n) Freq. (n) Freq. (n)

Ser311/Ser311 252 0.94 329 0.90 457 0.95 255 0.95 43 0.98 22 0.96 1358 Ser311/Cys311 15 0.06 37 0.10 23 0.05 14 0.05 1 0.02 1 0.04 91 Totals 267 366 480 269 44 23 1449

Molecular Psychiatry DRD2 and major psychoses A Serretti et al 272 Table 2 Dopamine D2 receptor genotypes and symptom scores. Means and standard deviations of factor scores are reported. Delusion and Disorganization symptomatology were significantly associated with the DRD2 S311C variant

Genotypes Excitement Delusion Depression Disorganization Negative Mean SD Mean SD Mean SD Mean SD Mean SD n

Ser311/Ser311 0.353 0.359 0.263 0.266 0.555 0.347 0.260 0.289 0.203 0.268 835 Ser311/Cys311 0.346 0.365 0.395 0.274 0.462 0.352 0.368 0.327 0.255 0.319 52 All Groups 0.352 0.359 0.270 0.268 0.550 0.348 0.266 0.292 0.206 0.271 887 F 0.017 12.087 3.565 6.742 1.772 P 0.897 0.0005 0.059 0.009 0.183

were admitted assuming lithium, antidepressant or and people attending the general lab of our hospital benzodiazepine treatments, we applied a lifetime per- (age = 46.62 ± 15.93; males 48%). Before drawing blood spective to rate the OPCRIT checklist, also using comp- samples, controls were closely investigated to deter- lementary sources of data like previous records and mine whether they or their first-degree relatives had family informants.58 This strategy allowed us to psychiatric disturbances or previous psychiatric treat- describe early drug-free illness phases. ment, through interviews with subjects and relatives Our sample is composed mainly of severe inpatients when possible. Only unaffected subjects with negative and this should indeed be considered a selection bias, family histories were included in the study. It should probably due to the fact that our center is a tertiary be noted that those data were not collected through the care setting. On the other hand the selection of severe detailed methodology used in family studies, and that subjects should guarantee against the possible a substantial rate of false negatives is commonly inclusion of phenocopies; in fact it has been proposed observed.61 Informed consent was obtained from all to exclude less severe forms of mood disorders like sin- subjects; these were unrelated and of Italian descent, gle episode major depression as they are considered to with antecedents from all parts of the country. have less genetic contribution.32 In conclusion, DRD2 S311C variant was not a major DNA analysis liability factor for the development of major psychoses, but it may be connected with the disorganized and Genomic DNA was extracted from anticoagulated delusional symptomatology independently from diag- thawed blood according to the method of Lahiri and noses. Nurnberger.62 PCR-based genotyping of Ser311/Cys311 variant was carried out according to the method described by Arinami et al.5 PCR products were Methods digested with Sau961 restriction enzyme (New Sample England Biolabs). The digested PCR products were One thousand one hundred and eighty-two inpatients resolved in 3.5% agarose gel by electrophoresis, and (age = 44.70 ± 14.35 years; onset = 30.20 ± 12.19 years; visualized with ethidium bromide staining by ultra- males 44%) consecutively admitted to the Department violet light. The Ser311 shows bands of 126 bp, of Psychiatry at the Institute H San Raffaele were 92 bp, 53 bp and 23 bp, whereas the Cys311 allele included in this study. Two hundred and sixty-seven shows bands of 149 bp, 92 bp and 53 bp. Genotyping subjects were included in previous papers reporting an has been checked by two readers who were blind to association with psychotic symptomatology.51,52 One clinical data. hundred and ninety-nine patients considered for this study were part of the sample collected for the Euro- Statistical analysis pean Collaborative Project on Affective Disorders,59 but they have not been typed for DRD2 S311C variants for The factoring process of the OPCRIT checklist has been the consortium. Lifetime diagnoses were assigned by previously presented.36,63 The factor composition is as two independent psychiatrists on the basis of clinical follows: Excitement (items: Excessive activity, interviews and medical records, according to DSM-IV Reduced need for sleep, Pressured speech, Elevated criteria following a best estimate method.58 A subsam- mood, Thoughts racing, Increased sociability, ple of 887 patients was evaluated using the OPCRIT Increased self esteem, Irritable mood, Distractibility, checklist with a lifetime perspective.60 Criteria for Agitated activity, Dysphoria, Grandiose delusions, inclusion in the study were diagnosis of major Reckless activity); Depression (items: Loss of pleasure, psychoses. The presence of concomitant diagnoses of Loss of energy/tiredness, Diminished libido, Excessive mental retardation or drug dependence, somatic or self reproach, Slowed activity, Poor appetite, Poor con- neurological illnesses that impaired psychiatric evalu- centration, Suicidal ideation, Weight loss, Diurnal vari- ation represented exclusion criteria. ation, Early morning waking, Delusions of guilt); Two-hundred and sixty-seven control subjects were Delusion (items: Persecutory delusions, Well organized recruited among both healthy department personnel delusions, Delusions of influence, Widespread

Molecular Psychiatry DRD2 and major psychoses A Serretti et al 273 delusions, Primary delusional perception, Delusions 12 Sobell J, Sigurdson D, Heston L, Sommer S. S311C D2DR variant: and hallucinations lasting for one week, no association with schizophrenia. Lancet 1994; 344: 621–622. 13 Tanaka T, Igarashi S, Onodera O, Tanaka H, Fukushima N, Taka- Persecutory/jealous delusions and hallucinations, hashi M et al. Lack of association between dopamine D2 receptor Thought insertion, Thought withdrawal, Thought gene Cys311 variant and schizophrenia. Am J Med Gen 1996; 67: broadcast, Third person auditory hallucinations); Dis- 208–211. organization (items: Speech difficult to understand, 14 Verga M, Macciardi F, Pedrini S, Cohen S, Smeraldi E. No associ- Incoherent, Positive formal thought disorder, Inappro- ation of the Ser/Cys311 DRD2 molecular variant with schizo- phrenia using a classical case control study and the haplotype rela- priate affect, Bizarre behavior, Bizarre delusions); and tive risk. Schizophr Res 1997; 25: 117–121. Negative (items: Blunted affect, Negative formal 15 Shaikh S, Collier D, Arranz M, Ball D, Gill M, Kerwin R. DRD2 thought disorder, Restricted affect, Catatonia). Factor Ser311/Cys311 polymorphism in schizophrenia. Lancet 1994; 343: scores were calculated as the sum of binarized items 1045–1046. divided by the number of items in each factor. They 16 Pearlson GD, Wong DF, Tune LE, Ross CA, Chase GA, Links JM et al. In vivo D2 density in psychotic and nonpsy- were considered to be dependent variables when chotic patients with bipolar disorder. Arch Gen Psychiatry 1995; investigating their distribution across variants. Differ- 52: 471–477. ences of factor scores among variants were assessed 17 Smith M, Wasmuth J, McPherson J, Wagner C, Grandy D, Civelli O using one-way Analysis of Variance (ANOVA). Analy- et al. Cosegregation of an 11q22.3–9q22 translocation with affective sis of Covariance (ANCOVA) was used when including disorder: proximity of the dopamine D2 receptor gene relative to the translocation breakpoint. Am J Hum Genet 1989; 45 (suppl): covariates. Alpha values were considered significant A220. when below 0.01 (Bonferroni corrected: 0.05 divided 18 Balk J, Picetti R, Salardi A, Thirlet G, Dierich A, Depaulls A et al. by five factors). The power of our sample to detect dif- Parkinsonian-like locomotor impairment in mice lacking dopamine ferences amongst DRD2 variants was calculated con- D2 receptors. Nature 1995; 377: 424–428. sidering an alpha value of 0.01, two-tailed. With these 19 Arinami T, Itokawa M, Aoki J, Shibuya H, Ookubo Y, Iwawaki A et al. Further association study on dopamine D2 receptor variant parameters in our sample we had a high power (0.80) S311C in schizophrenia and affective disorders. Am J Med Gen to detect a medium effect size (0.49), that corresponded 1996; 67: 133–138. to a difference of approximately 0.12 points between 20 Byerley W, Leppert M, O’Connell P, Mellon C, Holik J, Lubbers A et the two DRD2 variants on factor scores.64 Data were al. D2 Dopamine receptor not linked to manic-depression in three analyzed with a Chi-square test for the overall associ- families. Psych Genet 1990; 1: 55–62. × 21 Holmes D, Brynjolfsson J, Brett P, Curtis D, Petursson H, Sherring- ation with diagnoses; 2 2 tables composed by ton R et al. No evidence for a susceptibility predisposing to Ser311/Cys311 vs Ser311/Cys311 were used. manic depression in the region of the dopamine D2 receptor gene. Br J Psych 1991; 158: 635–641. References 22 Mitchell P, Selbie L, Waters B, Donald J, Vivero C, Tully M et al. Exclusion of close linkage of bipolar disorder to the dopamine D1 1 Willner P. mechanisms in depression and mania. In: and D2 receptor gene markers. J Aff Dis 1992; 25: 1–12. Bloom F, Kupfer D (eds). Psychopharmacology, the Fourth Gener- 23 No¨then M, Erdmann J, Ko¨rner J, Lanczik M, Fritze J, Fimmers R et ation of Progress. Raven Press: New York, 1995, pp 921–931. al. Lack of association between dopamine D1 and D2 receptor genes 2 Grandy D, Litt M, Allen L, Bunzow J, Marchionni M, Makam H et and bipolar affective disorder. Am J Psych 1992; 149: 199–201. al. The human D2 dopamine receptor gene is located on chromo- 24 Debruyn A, Mendelbaum K, Sandkuijl L, Delvenne V, Hirsch D, some 11 at q22–q23 and identifies a TaqI RFLP. Am J Hum Genet Staner L et al. Nonlinkage of bipolar illness to tyrosine 1989; 45: 778–785. hydroxylase, tyrosinase, and D2 and D4 dopamine receptor genes 3 Itokawa M, Arinami T, Futamura N, Hamaguchi H, Toru M. A on 11. Am J Psychiatry 1994; 151: 102–106. Structural polymorphism of human dopamine D2 receptor, D2 25 Nanko S, Fukuda R, Hattori M, Sasaki T, Dai XY, Kanba S et al. (Ser311→Cys). Biochem Biophys Res Commun 1993; 196: 1369– Linkage studies between affective disorder and dopamine D2, D3, 1375. and D4 receptor gene loci in four Japanese pedigrees. Psychiatry 4 Jonsson E, Sedvall G, Brene S, Gustavsson JP, Geijer T, Terenius L Res 1994; 52: 149–157. et al. Dopamine-related genes and their relationships to monoam- 26 Craddock N, Roberts Q, Williams N, McGuffin P, Owen MJ. Associ- ine metabolites in CSF. Biol Psychiatry 1996; 40: 1032–1043. ation study of bipolar disorder using a functional polymorphism 5 Arinami T, Itokawa M, Enguchi H, Tagaya H, Yano S, Shimizu H (Ser311→Cys) in the dopamine D2 receptor gene. Psychiatr Genet et al. Association of dopamine D2 receptor molecular variant with 1995; 5: 63–65. schizophrenia. Lancet 1994; 343: 703–704. 6 Nanko S, Hattori M, Dai XY, Fukuda R, Kazamatsuri H. DRD2 27 Oruc L, Furac I, Croux C, Jakovljevic M, Kracun I, Folnegovic V et Ser311/Cys311 polymorphism in schizophrenia. Lancet 1994; al. Association study between bipolar disorder and candidate genes 343: 1044. involved in dopamine- metabolism and GABAergic neur- 7 Nothen MM, Wildenauer D, Cichon S, Albus M, Maier W, Minges otransmission: a preliminary report. Psychiatr Genet 1996; 6: J et al. Dopamine D2 receptor molecular variant and schizophrenia. 213–217. Lancet 1994; 343: 1301–1302. 28 Souery D, Lipp O, Mahieu B, Mendelbaum K, De Martelaer V, Van 8 Chen CH, Chien SH, Hwu HG. No association of dopamine D2 Broeckhoven C et al. Association study of bipolar disorder with receptor molecular variant Cys311 and schizophrenia in Chinese candidate genes involved in catecholamine neurotransmission: patients. Am J Med Gen 1996; 67: 418–420. DRD2, DRD3, DAT1, and TH genes. Am J Med Gen 1996; 67: 9 Crawford F, Hoyne J, Cai X, Osborne A, Poston D, Zaglul J et al. 551–555. Dopamine DRD2/Cys311 is not associated with chronic schizo- 29 Perez de Castro I, Santos J, Torres P, Visedo G, Saiz-Ruiz J, Llinares phrenia. Am J Med Gen 1996; 67: 483–484. C et al. A weak association between TH and DRD2 genes and 10 Goldman D, Urbanek M, Guenther D, Robin R, Long JC. Linkage bipolar affective disorder in a Spanish sample. J Med Gen 1995; and association of a functional DRD2 variant [Ser311Cys] and 32: 131–134. DRD2 markers to alcoholism, substance abuse and schizophrenia in 30 Cloninger C. Turning point in the design of linkage studies of Southwestern American Indians. Am J Med Gen 1997; 74: 386–394. schizophrenia. Am J Med Gen 1994; 54: 83–92. 11 Laurent C, Bodeau-Pean S, Campion D, d’Amato T, Jay M, Dollfus 31 Ginsburg B, Werick T, Escobar J, Kugelmass S. Molecular genetics S et al. No major role for the dopamine D2 receptor Ser→Cys311 of psychopathologies: a search for simple answers to complex prob- mutation in schizophrenia. Psychiatr Genet 1994; 4: 229–230. lems. Behav Genet 1996; 26: 325–333.

Molecular Psychiatry DRD2 and major psychoses A Serretti et al 274 32 Tsuang M, Faraone S. The Genetics of Mood Disorders. The Johns in a systematic sample of patients with onset of schizophrenia in Hopkins University Press: Baltimore, 1990. childhood and early adolescence. II: Postnegative distinction and 33 Cloninger C, Adolfsson R, Svrakic N. Mapping genes for human childhood predictors of adult outcome. Br J Psychiatry 1996; 169: personality. Nat Genet 1996; 12: 3–4. 371–378. 34 Comings DE. Polygenic inheritance in psychiatric disorders. In: 50 Chua SE, Wright IC, Poline JB, Liddle PF, Murray RM, Frackowiak Blum K, Noble EP (eds). Handbook of Psychiatric Genetics. CRC RS et al. Grey matter correlates of syndromes in schizophrenia. A Press: Boca Raton, 1997, pp 235–260. semi-automated analysis of structural magnetic resonance images. 35 Risch N. Linkage strategies for genetically complex traits: I. Multi- Br J Psychiatry 1997; 170: 406–410. locus models. Am J Med Gen 1990; 46: 222–228. 51 Serretti A, Macciardi F, Smeraldi E. Dopamine receptor D2 Ser 36 Serretti A, Macciardi F, Smeraldi E. Identification of symptomato- Cys311 variant associated with disorganized symptomatology of logic patterns common to major psychoses: proposal for a pheno- schizophrenia. Schizophr Res 1998; 34: 207–210. type definition. Am J Med Gen 1996; 67: 393–400. 52 Serretti A, Lilli R, Lorenzi C, Smeraldi E. Further evidence support- 37 Liddle PF. The symptoms of chronic schizophrenia. A re-examin- ing the association between the dopamine receptor D2 Ser/Cys311 ation of the positive-negative dichotomy. Br J Psychiatry 1987; 151: variant and disorganized symptomatology of schizophrenia. Schi- 145–151. zophr Res (in press). 38 Andreasen NC, Arndt S, Alliger R, Miller D, Flaum M. Symptoms 53 Sasaki T, Macciardi FM, Badri F, Verga M, Meltzer HY, Lieberman of schizophrenia. Methods, meanings, and mechanisms. Arch Gen J et al. No evidence for association of dopamine D2 receptor variant Psychiatry 1995; 52: 341–351. (Ser311/Cys311) with major . Am J Med Gen 1996; 67: 39 Liddle PF, Barnes TRE. Syndromes of chronic schizophrenia. Br J 415–417. Psychiatry 1990; 157: 558–561. 54 Serretti A, Smeraldi E. Dopamine D2 receptor gene not associated 40 Peralta V, de Leon J, Cuesta MJ. Are there more than two syndromes with symptomatology of mood disorders. Am J Med Gen 1999; 88: in schizophrenia? A critique of the positive-negative dichotomy. Br 294–297. J Psychiatry 1992; 161: 335–343. 55 Coryell W, Winokur G, Shea T, Maser J, Endicott J, Akiskal H. The 41 Maziade M, Roy MA, Martinez M, Cliche D, Fournier JP, Garneau long-term stability of depressive subtypes. Am J Psychiatry 1994; Y et al. Negative, psychoticism, and disorganized dimensions in 151: 199–204. patients with familial schizophrenia or bipolar disorder: continuity 56 Gasparini P, Estivill X, Volpini V, Totaro A, Castellvi-Bel S, Govea and discontinuity between the major psychoses. Am J Psychiatry N et al. Linkage of DFNB1 to non-syndromic neurosensory autoso- 1995; 152: 1458–1463. mal-recessive deafness in Mediterranean families. Eur J Hum Genet 42 Van Os J, Marcelis M, Sham P, Jones P, Gilvarry K, Murray R. 1997; 5: 83–88. Psychopathological syndromes and familial morbid risk of psy- 57 Peralta V, Cuesta M, De Leon J. An empirical analysis of latent chosis. Br J Psychiatry 1997; 170: 241–246. structures underlying schizophrenic symptoms: a four-syndrome 43 Van Os J, Fahy TA, Jones P, Harvey I, Sham P, Lewis S et al. model. Biol Psychiatry 1994; 36: 726–736. Psychopathological syndromes in the functional psychoses: associ- 58 Leckman JF, Sholomskas D, Thompson WD, Belanger A, Weissman ations with course and outcome. Psychol Med 1996; 26: 161–176. MM. Best estimate of lifetime psychiatric diagnosis: a methodolog- 44 Burke JG, Murphy BM, Bray JC, Walsh D, Kendler KS. Clinical ical study. Arch Gen Psychiatry 1982; 39: 879–883. similarities in siblings with schizophrenia. Am J Med Gen 1996; 59 Souery D, Lipp O, Mahieu B, Serretti A, Cavallini C, Ackenheil M 67: 239–243. et al. European Collaborative Project on Affective Disorders: Inter- 45 Kendler KS, Karkowski-Shuman L, FA ON, Straub RE, MacLean actions between genetic and psychosocial vulnerability factors. CJ, Walsh D. Resemblance of psychotic symptoms and syndromes Psychiatr Genet 1998; 8: 197–206. in affected sibling pairs from the Irish Study of High-Density 60 McGuffin P, Farmer A, Harvey I. A polydiagnostic application of Schizophrenia Families: evidence for possible etiologic heterogen- operational criteria in studies of psychotic illness. Development eity. Am J Psychiatry 1997; 154: 191–198. and reliability of the OPCRIT system. Arch Gen Psychiatry 1991; 46 Loftus J, DeLisi LE, Crow TJ. Familial associations of subsyndromes 48: 764–770. of psychosis in affected sibling pairs with schizophrenia and schi- 61 Andreasen NC, Endicott J, Spitzer RL. The family history method zoaffective disorder. Psychiatry Res 1998; 80: 101–111. using diagnostic criteria: reliability and validity. Arch Gen Psy- 47 Cardno AG, Holmans PA, Harvey I, Williams MB, Owen MJ, chiatry 1977; 34: 1229–1233. McGuffin P. Factor-derived subsyndromes of schizophrenia and 62 Lahiri D, Nurnberger JJ. A rapid non-enzymatic method for the familial morbid risks. Schizophr Res 1997; 23: 231–238. preparation of HMW DNA from blood for RFLP studies. Nucleic 48 Cardno AG, Jones LA, Murphy KC, Sanders RD, Asherson P, Owen Acid Res 1991; 19: 5444. MJ et al. Sibling pairs with schizophrenia or schizoaffective dis- 63 Cardno AG, Jones LA, Murphy KC, Asherson P, Scott LC, Williams order—associations of subtypes, symptoms and demographic vari- J et al. Factor analysis of schizophrenic symptoms using the ables. Psychol Med 1998; 28: 815–823. OPCRIT checklist. Schizophr Res 1996; 22: 233–239. 49 Maziade M, Bouchard S, Gingras N, Charron L, Cardinal A, Roy MA 64 Cohen J. Statistical Power Analysis for the Behavioral Sciences. et al. Long-term stability of diagnosis and symptom dimensions Lawrence Erlbaum Associates: Hillsdale, NJ, 1988.

Molecular Psychiatry