- Email: [email protected]
No evidence for linkage or association between the dopamine transporter gene and schizophrenia in a French population
PSYCHIATRY RESEARCH ELSEVIER
Psychiatry Research 59 (1995) l-6
No evidence for linkage or association between the dopamine transporter gene and schizophrenia in a French population
a,
Dominique Campion b, Maurice Jay”, Sylvie Bodeau-PCan”, Claudine Laurent Florence Thibautb, Sonia Dollfusb, Michel Petitb, Daniele Samolyka, Thierry d’Amatod, Maria Martineze, Jacques Mallet*” aLaboratoire de GPnt?ique Mokulaire de la Neurotransmission et des Processus NeurodPgknPratifs. CNRS, Batiment CER VI, Hbpital de la Piti&SalpktrOre. 83 Bd de I’Hc5pital, 75013 Paris, France bCentre Hospitalier SpPcialisP du Rouvray. 4 rue Paul Elunrd, BP 45, 76301 Sotteville-les-Rouen, France ‘Centre Hospitalier SpPcialisk de Saint-Paul, La RPunion, France ‘SHU de Psychiatric Adulte, H6pital du Vinatier, 69677 Lyon-Bron. France =Unit&de GtWtique Epidkmiologique, Inserm UI55. Chateau de Longchamp, 75016 Paris, France
Received 14 February 1995; revision received 19 July 1995; accepted 8 August 1995
Abstract Pharmacological and clinical findings suggest that the dopamine transporter (DAT) gene may be involved in the genetic predisposition to schizophrenia. Linkage of a Taq I VNTR polymorphism in the DAT gene to schizophrenia was studied in multiplex schizophrenic families from Rouen, France (n = 10) and the Island of La Reunion (n = 21). Neither the lod score method nor nonparametric methods (the affected pedigree member method of Weeks and Lange [ 19881and the sibling method of Green and Woodrow [ 19771)provided any evidence for linkage. An association study, carried out within a group of 91 unrelated schizophrenic patients from Rouen and 91 matched control subjects, examined a 40 base-pair repeat polymorphism located in the 3 ’ nontranslated end of the DAT mRNA. There was no significant difference in allelic or genotypic frequencies between the two groups. These results exclude any substantial involvement of the DAT gene in the pathogenesis of schizophrenia in the population studied. Keywords:
Genetics; Linkage analysis; Association study
1. Introduction
extensive
investigation
by several
mine receptor genes have not There is pharmacological evidence suggesting that a dysfunction of the dopaminergic system could be involved in the etiology of schizophrenia (Goldstein and Deutch, 1992). However, despite l Corresponding author, Tel: +33 1 42177531; Fax: +33 I 42177533.
groups,
dopa-
been clearly implicated (for review, see Mallet et al., 1994). Possibly, genes coding for other molecules involved in the dopaminergic transmission may contribute to the predisposition for schizophrenia. One candidate is the dopamine transporter (DAT) gene, which encodes a protein responsible for the reuptake of dopamine into presynaptic terminals.
0165-1781/95/%09.50 0 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0165-1781(95)02789-Y
2
S. Bodeau-Plan et al. /Psychiatry Research 59 (1995) 1-6
This mechanism allows termination of synaptic transmission and recycling of neurotransmitters (Uhl, 1992). The transport is blocked by psychostimulant drugs including cocaine, which thereby induce more pronounced manifestations of preexisting psychotic symptoms in schizophrenic patients (Horn, 1990). The human DAT gene was recently cloned (Giros et al., 1992) and mapped to the ~15.3 region of chromosome 5. A linkage study including nine Caucasian pedigrees failed to demonstrate positive linkage between schizophrenia and a DAT gene VNTR (variable number of tandem repeats) marker (Byerley et al., 1993a). However, definite conclusions could not be drawn from the study because (1) only a small number of pedigrees were investigated and (2) misspecification of genetic parameters in the lod score analysis could have produced false-negative results (Clerget-Darpoux et al., 1986). We report two different analyses of the relationship between the DAT gene and schizophrenia: (1) a linkage study using a Taq I VNTR in 31 schizophrenic multiplex families, analyzed by parametric and nonparametric methods; and (2) an association study using a 40 base-pair repeat polymorphism in 91 schizophrenic patients and 91 control subjects. 2. Methods 2. I. Data collection
Thirty-one multiplex families (21 from the Island of La Reunion in the Indian Ocean, 10 from Rouen, France) from our previously described sample of families (d’Amato et al., 1992; Campion et al., 1994) were included in the linkage study. These families contained at least two living firstdegree relatives with diagnoses of schizophrenia. Trained psychiatrists used the French translations of the Lifetime Anxiety version of the Schedule for Affective Disorders and Schizophrenia (Fyer et al., 1985) and the Schedule for Schizotypal Personalities (Baron et al., 1981) for diagnosis according to DSM-III criteria (American Psychiatric Association, 1980). Ninety-one unrelated schizophrenic patients (mean age = 42 f 12; mean age of onset = 24 f
6.2; 29 females and 62 males) and 9 1 matched control subjects (mean age = 48 + 11; 41 females and 50 males) who were free of psychiatric disorder and had no familial history of psychosis were included in the association study. All participants were Caucasian and originated from the area of Rouen (Normandy, France). Diagnoses were obtained as described above for subjects of multiplex families. 2.2. Laboratory procedures Genomic DNA was extracted from lymphocytes in the venous blood collected from subjects after informed consent had been given (Maniatis et al., 1982). Two markers were studied: (1) The Taq I VNTR described by Byerley et al. (1993b) was typed in multiplex family members with a standard Southern blotting method (Maniatis et al., 1982) that used a 3.2 kb probe corresponding to the nine C-terminal putative transmembrane regions of the human DAT gene (snhl-4) (Giros et al., 1992). The 16 alleles described by Byerley et al. (1993b) were observed. The Taq I VNTR was not investigated in the association sample because of the excessively large number of alleles. (2) The 40 base-pair repeat polymorphism located in the 3 ’ untranslated region of the DAT mRNA (Vandenbergh et al., 1992; Sano et al., 1993) was investigated in each member of the association sample using polymerase chain reaction, as described by these authors. Five alleles were observed corresponding to 6, 9, 10, 11, and 12 copies of the 40 base-pair element. 2.3. Statistics Pedigree analysis with the Taq I VNTR. In the 3 1 families, 165 individuals, including 80 schizophrenic patients, were typed. Lod score analysis was carried out with the MLINK program from the computer package LINKAGE, version 5.1 (Lathrop et al., 1985). All nonaffected subjects were considered as “unknown phenotype” because of the difficulty of specifying the penetrance functions for the pathological genotypes. Two definitions of the affected phenotype were used: a narrow definition, which included only schizophrenia, and a broad one, which included all schizophrenia spectrum disorders (schizophrenia,
S. Bodeau-PPan et al. /Psychiatry Research 59 (1995) 1-6
3
a moderately frequent gene (q = 0.097) with a lower rate of phenocopy (faa = 0.0003). In addition, two nonparametric methods of linkage analysis were applied: (1) The computer package of Weeks and Lange (1988) was used for the affected pedigree member (APM) method. As specified by these authors, different weighting factors, functions of the marker allelic frequencies, were used. Consequently, more weight is given to the sharing of a rare than of a common allele between distantly related affected relatives. (2) Twenty-three sibships containing 41 affected siblings were examined for allele similarity with the sibling method of Green and Woodrow (1977). Marker allele frequencies used in the lod score and APM analysis were estimated independently from family data in both the French and La Reunion samples according to the method of Biiehnke (1991). Recombination
Association analysis with the 40 base-pair repeat polymorphism. The allelic and genotypic distribu-
Fraction
tions for the 40 base-pair marker in the 91 unrelated schizophrenic patients was compared with that in the 91 matched control subjects using a x2 test with Yate’s correction as appropriate.
Fig. I. Lod scores for schizophrenia (“narrow” diagnosis) or schizophrenia spectrum (“broad” diagnosis) and Taq I VNTR, under two different genetic models (see 2.3. Statistical method+ Lod score, log to the base IO of the probability that a given set of data about genetic recombination would by virtue of two loci being linked at a specified recombination fraction divided by the probability that the data would arise by nonlinkage. VNTR, variable number of tandem repeats.
3. Results 3.1. Linkage analysis Due to differences in the marker allelic frequencies (data not shown), lod scores for linkage between schizophrenia and the DAT gene Taq I VNTR were computed separately for the Rouen and La Reunion subsamples subsequently summed. For all clinical classifications and genetic models used, the lod score values were negative in both subsamples. Thus, linkage of schizophrenia to the DAT locus was excluded (Fig. 1).
schizophreniform disorder, atypical psychosis, schizoaffective disorder, and schizotypal personality). Schizophrenia was treated as an autosomal dominant disease. Two different genetic models were used according to the analysis of McGue and Gottesman (1991): the first assumes a rare susceptibility allele (q = 0.003) with a high rate of phenocopy (faa = 0.008); and the second assumes
Table 1 Green and Woodrow sibling analysis of Taq I VNTR (variable number of tandem repeats) polymorphism with narrow and broad diagnoses (schizophrenia and schizophrenia spectrum) Sibships
Narrow diagnosis Broad diagnosis
23 23
Sibship composition Pairs
Trios
Quartets
16 10
5 9
2 4
Total score
I statistic
36 48
1.06, NS -0.66, NS
4
S. Bodeau-Pkan et ul. / Psychiafry Research 59
I 1995) 1-6~
Table 2 Allelic and genotypic distributions of the 40 base-pair repeat polymorphism within the associaiton sample Genotypes
Control subjects (n = 91) Schizophrenic patients (n=91) Familial cases (n = 18)
Alleles
11
12
22
Other
I
2
Other
9
41
38
3
61
118
3
6
38
44
3
50
129
3
2
5
10
1
9
26
1
3
0 3 0
8 17 25
14 48 67
0 3 0
Clinical subtypes
Paranoid (n = II) Disorganized (n = 34) Undifferentiated (n = 46)
0
8
1
IS
15
5
15
26
Nore. The two major alleles (allele 1 = 9 repeats of the 40 base-pair elements; allele 2 = 10 repeats) have been detailed separately; the other alleles (6, 11, and 12 repeats have been grouped because of their very rare occurrence (Other)).
The Rouen and La Reunion subsamples were also considered separately for the APM analysis. For all subsamples considered, weighting factors applied and clinical classifications used, the t values obtained did not reject the null hypothesis of independent segregation between the marker locus and the disease locus (data not shown). We also calculated the t statistic according to the Green and Woodrow sibling method (Table 1). The t values were not statistically significant either for the narrow or the broad clinical classification (narrow definition: t = 1.06, NS; broad definition: t = -0.66, NS). 3.2. Association analysis The possible association between the 40 basepair repeat polymorphism and schizophrenia was studied (Table 2). The allelic distributions in control subjects and patients were not significantly different (x ’ = 1.58, df = 2, NS). No significant difference between the genotype distributions of control subjects and patients was found (x2 = 1.!5, df = 3, NS), and the Hardy-Weinberg equilibrium was conserved in both groups (schizophrenic patients: x2 = 0.24, df = 2, NS; control subjects: x2 = 0.11, df = 2, NS). The patients were subdivided into subgroups according to clinical diagnosis or familial history of schizophrenia and compared with control subjects. The results were again negative both for allelic dis-
tributions (familial cases: x2 = 1.13, df = 2, NS; clinical subgroups: paranoid subtype: xZ = 0.41, df = 2, NS; disorganized subtype: x2 = 2.97, df = 2, NS, undifferentiated subtype: x2 = 2.88, df = 2, NS) and for genotype distributions (familial cases: x2 = 1.94, df = 3, NS; clinical subgroups: paranoid subtype: x2 = 3.52, df = 3, NS; disorganized subtype: x2 = 3.11, df = 3, NS; undifferentiated subtype: x2 = 4.13, df = 3, NS). 4. Discussion Our analysis gave no evidence for linkage between a VNTR marker of the DAT gene and schizophrenia in a group of families containing individuals originating from metropolitan France and La Reunion Island. Lod score values at a recombination fraction of 8 = 0 were below -2, excluding linkage for simple Mendelian diseases. However, lod score results should be interpreted with caution in genetically complex diseases like schizophrenia because misspecification of genetic parameters can lead to false exclusion of linkage (Clerget-Darpoux et al., 1986). To minimize this problem, we chose a conservative procedure to analyze our data: an “affected only” analysis, which considered all unaffected subjects as “unknown phenotypes” while including their genotypes at the marker locus in the lod score computation. We also used nonparametric
S. Bodeau-Pt!an et al. /Psychiatry Research 59 (1995) l-6
methods of linkage analysis, which do not require specification of genetic parameters. The APM analysis also did not find significant results. Both the lod score (Ott, 1992) and the APM (Babron et al., 1993) methods are sensitive to misspecification of allelic frequencies at the marker locus. To circumvert this problem, allelic frequencies were calcuiated using data from relatives in both La R&union Island and metropolitan subsamples. The Green and Woodrow sibling method requires neither specification of genetic parameters nor allelic frequencies at the marker locus. This method also failed to provide evidence for linkage between the Taq I VNTR and schizophrenia in our sample. A case-control study was also carried out in a French metropolitan sample from Rouen. This approach offers a good alternative for detecting predisposition factors in genetically complex diseases (Strittmatter et al., 1993). In our population, there was no association between the 40 basepair repeat polymorphism of the DAT gene and schizophrenia. An identical result has been reported in a Chinese population (Li et al., 1994) and a very recent British study (Daniels et al., 1995). These linkage and association studies do not support any major involvement of the DAT gene in schizophrenia. A minor effect cannot be excluded, however, because of the limited power of the statistical methods, the sample size, and the informativity of the markers. Acknowledgments
References American Psychiatric Association. (1980) DSM-III: Diagnostic and Sfatisrical Manual of Mental Disorders. 3rd edn. APA, Washington, DC. Babron, M.C., Martinez, M., Bonai’ti-Pellit, C. and ClergetDarpoux, F. (1993) Linkage detection by the affectedpedigree-member method: what is really tested? Gene1 Epidemioi IO, 389-394.
Baron. M., Asnis, L. and Gruen, R. (1981) The Schedule for Schizotypal Personalities (SSP): a diagnostic interview for schizotypal features. Psychiafry Res 4, 213-228. Biiehnke, M. (1991) Allele frequency estimation from data on relatives Am J Hum Gene! 48, 22-25. Byerley, W., Coon, H., Hoff, M., Holik, J., Waldo, M., Freedman, R., Caron, M.G. and Giros, B. (1993a) Human dopamine transporter gene not linked to schizophrenia in multigenerational pedigrees. Hum Hered 43, 319-322. Byerley, W., Hoff, M., Holik, J., Caron, M. and Giros, B. (1993b) VNTR polymorphism for the human dopamine transporter gene. Hum Mat Genet 2, 335. Campion, D.. d’Amato, T., Bastard, C.. Laurent, C., Guedj, F., Jay, M., Gorwood, P.. Dolfus, S., Thibault, F., Petit, M.. Gorwood, P., Babron, M.C., Waksman, G., Martinez. M. and Mallet, J. (1994) Genetic study of dopamine D,, D,, and D4 receptors in schizophrenia. Psychiatry Res 51, 215-230.
Clerget-Darpoux, F., Bonai’ti-Pellie, C. and Hochez, J. (1986) Effects of misspecifying genetic parameters in lod score analysis. Biometrics 42. 393-399. d’Amato,T., Campion,D., Gorwood, P., Jay, M., SabatC, O., Petit, C., Abbar, M., Malafosse, A., Leboyer, M., Hillaire, D., Clerget-Darpoux, F., Feingold, J., Waksman, G. and Mallet, J. (1992) Evidence for a pseudoautosomal locus for schizophrenia: replication of a non-random segregation of alleles at the DXYS.14 locus. Br J Psychiatry 161, 59-62. Daniels,J., Williams, J., Asherson, P., McGuff~n, P. and Owen, M. (1995) No association between schizophrenia and polymorphisms within the genes for debrisoquine 4hydroxylase (CYP2D6) and the dopamine transporter (DAT). Am J Med Gener (Neuropsychiatr Genet) 60.85-87. Fyer, A.J., Endicott, J., Mannuzza, S. and Klein, D.F. (1985)
Schedule for Affective Disorders and Schizophrenia-Lifetime
We are grateful to B. Giros for providing the probe. This work received financial support from the Centre National de la Recherche Scientifique, the Association Francaise contre les Myopathies, the Groupement de Recherche et d’6tudes sur les GtnBmes, RhBne Poulenc Rorer, and the Assistance Publique des HBpitaux de Paris. Sylvie Bodeau-P6an was supported by a grant from the Ministkre de 1’Enseignement Supkrieur et de la Recherche, and Claudine Laurent by the Fondation pour la Recherche MCdicale.
Anxiety Version. New York State Psychiatric Institute, New York. Giros, B., El Mestikawy, S.. Godinot, N., Zheng, K., Han, H.. Yang-Feng, T. and Caron. M.G. (1992) Cloning, pharmacological characterization, and chromosome assignment of the human dopamine transporter. Mel Phormacol 42. 383-390.
Goldstein, M. and hutch, A.Y. (1992) Dopaminergic mechanisms in the pathogenesis of schizophrenia. FASEB J 6, 2413-2421.
Green. J.R. and Woodrow, J.C. (1977) Sibling method for detecting 31-35.
HLA-linked
genes in diseases.
Tissue Antigens 9,
6
S. Bodeau-Pian et al. /Psychiatry Research 59 (1995) 1-6
Horn, A.S. (1990) Dopamine uptake: a review of progress the last decade. Prog Neurobiol34, 387-400.
in
markers in human 283-290.
gene mapping
Am J Hum Cenet 51.
Lathrop, GM., Lalouel, J.M., Jullier, C. and Ott, J. (1985). Multilocus linkage analysis in humans: detection of linkage and estimation of recombination. Am J Hum Gener 37. 482-498.
Sano, A., Kondoh, K., Kakimoto, Y. and Kondo, I. (1993) A 40nucleotide repeat polymorphism in the human dopamine transporter gene. Hum Gene1 91, 405-406. Strittmatter, W.J., Saunders, A.M., Schmechel, D.. Pericak-
Li, T., Yang, L., Wiese, C., Xu, C.T., Zeng, Z., Giros, B., Caron, M., Moises, H.W. and Liu, X. (1994) No association between alleles or genotypes at the dopamine transporter gene and schizophrenia. Psychiarry Res 52, 17-23.
Vance, M.. Enghild, J., Salvesen, G.S. and Roses, A. (1993) Apolipoprotein E: high-avidity binding to amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Nail Acad Sci USA 90, l977- I98 I.
Mallet, J., Meloni, R. and Laurent, C. (1994) Catecholamine metabolism and psychiatric or behavioral disorders, Curr Opin Genet Dev 4, 419-426. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. McGue, M. and Gottesman, I. (1991) The genetic epidemiology of schizophrenia and the design of linkage studies. Eur Arch Psychiatry Clin Neurosci 240, I74- I8 I. Ott, J. (1992) Strategies for characterizing highly polymorphic
Uhl, G.R. (1992) Neurotransmitter transporters: a promising new gene family Trends Neurosci 15, 265-268. Vandenbergh, D.J., Persico, A.M., Hawkins, A.L., Griffin, C.A., Li, X., Jabs, E.W. and Uhl, G.R. (1992) Human dopamine transporter gene (DATI) maps to chromosome 5~15.3 and displays a VNTR. Genomics 14, 1104-l 106. Weeks, E.D. and Lange, K. (1988) The affected-pedigreemember method of linkage analysis. Am J Hum Genet 42, 315-326.
Psychiatry Research 59 (1995) l-6
No evidence for linkage or association between the dopamine transporter gene and schizophrenia in a French population
a,
Dominique Campion b, Maurice Jay”, Sylvie Bodeau-PCan”, Claudine Laurent Florence Thibautb, Sonia Dollfusb, Michel Petitb, Daniele Samolyka, Thierry d’Amatod, Maria Martineze, Jacques Mallet*” aLaboratoire de GPnt?ique Mokulaire de la Neurotransmission et des Processus NeurodPgknPratifs. CNRS, Batiment CER VI, Hbpital de la Piti&SalpktrOre. 83 Bd de I’Hc5pital, 75013 Paris, France bCentre Hospitalier SpPcialisP du Rouvray. 4 rue Paul Elunrd, BP 45, 76301 Sotteville-les-Rouen, France ‘Centre Hospitalier SpPcialisk de Saint-Paul, La RPunion, France ‘SHU de Psychiatric Adulte, H6pital du Vinatier, 69677 Lyon-Bron. France =Unit&de GtWtique Epidkmiologique, Inserm UI55. Chateau de Longchamp, 75016 Paris, France
Received 14 February 1995; revision received 19 July 1995; accepted 8 August 1995
Abstract Pharmacological and clinical findings suggest that the dopamine transporter (DAT) gene may be involved in the genetic predisposition to schizophrenia. Linkage of a Taq I VNTR polymorphism in the DAT gene to schizophrenia was studied in multiplex schizophrenic families from Rouen, France (n = 10) and the Island of La Reunion (n = 21). Neither the lod score method nor nonparametric methods (the affected pedigree member method of Weeks and Lange [ 19881and the sibling method of Green and Woodrow [ 19771)provided any evidence for linkage. An association study, carried out within a group of 91 unrelated schizophrenic patients from Rouen and 91 matched control subjects, examined a 40 base-pair repeat polymorphism located in the 3 ’ nontranslated end of the DAT mRNA. There was no significant difference in allelic or genotypic frequencies between the two groups. These results exclude any substantial involvement of the DAT gene in the pathogenesis of schizophrenia in the population studied. Keywords:
Genetics; Linkage analysis; Association study
1. Introduction
extensive
investigation
by several
mine receptor genes have not There is pharmacological evidence suggesting that a dysfunction of the dopaminergic system could be involved in the etiology of schizophrenia (Goldstein and Deutch, 1992). However, despite l Corresponding author, Tel: +33 1 42177531; Fax: +33 I 42177533.
groups,
dopa-
been clearly implicated (for review, see Mallet et al., 1994). Possibly, genes coding for other molecules involved in the dopaminergic transmission may contribute to the predisposition for schizophrenia. One candidate is the dopamine transporter (DAT) gene, which encodes a protein responsible for the reuptake of dopamine into presynaptic terminals.
0165-1781/95/%09.50 0 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0165-1781(95)02789-Y
2
S. Bodeau-Plan et al. /Psychiatry Research 59 (1995) 1-6
This mechanism allows termination of synaptic transmission and recycling of neurotransmitters (Uhl, 1992). The transport is blocked by psychostimulant drugs including cocaine, which thereby induce more pronounced manifestations of preexisting psychotic symptoms in schizophrenic patients (Horn, 1990). The human DAT gene was recently cloned (Giros et al., 1992) and mapped to the ~15.3 region of chromosome 5. A linkage study including nine Caucasian pedigrees failed to demonstrate positive linkage between schizophrenia and a DAT gene VNTR (variable number of tandem repeats) marker (Byerley et al., 1993a). However, definite conclusions could not be drawn from the study because (1) only a small number of pedigrees were investigated and (2) misspecification of genetic parameters in the lod score analysis could have produced false-negative results (Clerget-Darpoux et al., 1986). We report two different analyses of the relationship between the DAT gene and schizophrenia: (1) a linkage study using a Taq I VNTR in 31 schizophrenic multiplex families, analyzed by parametric and nonparametric methods; and (2) an association study using a 40 base-pair repeat polymorphism in 91 schizophrenic patients and 91 control subjects. 2. Methods 2. I. Data collection
Thirty-one multiplex families (21 from the Island of La Reunion in the Indian Ocean, 10 from Rouen, France) from our previously described sample of families (d’Amato et al., 1992; Campion et al., 1994) were included in the linkage study. These families contained at least two living firstdegree relatives with diagnoses of schizophrenia. Trained psychiatrists used the French translations of the Lifetime Anxiety version of the Schedule for Affective Disorders and Schizophrenia (Fyer et al., 1985) and the Schedule for Schizotypal Personalities (Baron et al., 1981) for diagnosis according to DSM-III criteria (American Psychiatric Association, 1980). Ninety-one unrelated schizophrenic patients (mean age = 42 f 12; mean age of onset = 24 f
6.2; 29 females and 62 males) and 9 1 matched control subjects (mean age = 48 + 11; 41 females and 50 males) who were free of psychiatric disorder and had no familial history of psychosis were included in the association study. All participants were Caucasian and originated from the area of Rouen (Normandy, France). Diagnoses were obtained as described above for subjects of multiplex families. 2.2. Laboratory procedures Genomic DNA was extracted from lymphocytes in the venous blood collected from subjects after informed consent had been given (Maniatis et al., 1982). Two markers were studied: (1) The Taq I VNTR described by Byerley et al. (1993b) was typed in multiplex family members with a standard Southern blotting method (Maniatis et al., 1982) that used a 3.2 kb probe corresponding to the nine C-terminal putative transmembrane regions of the human DAT gene (snhl-4) (Giros et al., 1992). The 16 alleles described by Byerley et al. (1993b) were observed. The Taq I VNTR was not investigated in the association sample because of the excessively large number of alleles. (2) The 40 base-pair repeat polymorphism located in the 3 ’ untranslated region of the DAT mRNA (Vandenbergh et al., 1992; Sano et al., 1993) was investigated in each member of the association sample using polymerase chain reaction, as described by these authors. Five alleles were observed corresponding to 6, 9, 10, 11, and 12 copies of the 40 base-pair element. 2.3. Statistics Pedigree analysis with the Taq I VNTR. In the 3 1 families, 165 individuals, including 80 schizophrenic patients, were typed. Lod score analysis was carried out with the MLINK program from the computer package LINKAGE, version 5.1 (Lathrop et al., 1985). All nonaffected subjects were considered as “unknown phenotype” because of the difficulty of specifying the penetrance functions for the pathological genotypes. Two definitions of the affected phenotype were used: a narrow definition, which included only schizophrenia, and a broad one, which included all schizophrenia spectrum disorders (schizophrenia,
S. Bodeau-PPan et al. /Psychiatry Research 59 (1995) 1-6
3
a moderately frequent gene (q = 0.097) with a lower rate of phenocopy (faa = 0.0003). In addition, two nonparametric methods of linkage analysis were applied: (1) The computer package of Weeks and Lange (1988) was used for the affected pedigree member (APM) method. As specified by these authors, different weighting factors, functions of the marker allelic frequencies, were used. Consequently, more weight is given to the sharing of a rare than of a common allele between distantly related affected relatives. (2) Twenty-three sibships containing 41 affected siblings were examined for allele similarity with the sibling method of Green and Woodrow (1977). Marker allele frequencies used in the lod score and APM analysis were estimated independently from family data in both the French and La Reunion samples according to the method of Biiehnke (1991). Recombination
Association analysis with the 40 base-pair repeat polymorphism. The allelic and genotypic distribu-
Fraction
tions for the 40 base-pair marker in the 91 unrelated schizophrenic patients was compared with that in the 91 matched control subjects using a x2 test with Yate’s correction as appropriate.
Fig. I. Lod scores for schizophrenia (“narrow” diagnosis) or schizophrenia spectrum (“broad” diagnosis) and Taq I VNTR, under two different genetic models (see 2.3. Statistical method+ Lod score, log to the base IO of the probability that a given set of data about genetic recombination would by virtue of two loci being linked at a specified recombination fraction divided by the probability that the data would arise by nonlinkage. VNTR, variable number of tandem repeats.
3. Results 3.1. Linkage analysis Due to differences in the marker allelic frequencies (data not shown), lod scores for linkage between schizophrenia and the DAT gene Taq I VNTR were computed separately for the Rouen and La Reunion subsamples subsequently summed. For all clinical classifications and genetic models used, the lod score values were negative in both subsamples. Thus, linkage of schizophrenia to the DAT locus was excluded (Fig. 1).
schizophreniform disorder, atypical psychosis, schizoaffective disorder, and schizotypal personality). Schizophrenia was treated as an autosomal dominant disease. Two different genetic models were used according to the analysis of McGue and Gottesman (1991): the first assumes a rare susceptibility allele (q = 0.003) with a high rate of phenocopy (faa = 0.008); and the second assumes
Table 1 Green and Woodrow sibling analysis of Taq I VNTR (variable number of tandem repeats) polymorphism with narrow and broad diagnoses (schizophrenia and schizophrenia spectrum) Sibships
Narrow diagnosis Broad diagnosis
23 23
Sibship composition Pairs
Trios
Quartets
16 10
5 9
2 4
Total score
I statistic
36 48
1.06, NS -0.66, NS
4
S. Bodeau-Pkan et ul. / Psychiafry Research 59
I 1995) 1-6~
Table 2 Allelic and genotypic distributions of the 40 base-pair repeat polymorphism within the associaiton sample Genotypes
Control subjects (n = 91) Schizophrenic patients (n=91) Familial cases (n = 18)
Alleles
11
12
22
Other
I
2
Other
9
41
38
3
61
118
3
6
38
44
3
50
129
3
2
5
10
1
9
26
1
3
0 3 0
8 17 25
14 48 67
0 3 0
Clinical subtypes
Paranoid (n = II) Disorganized (n = 34) Undifferentiated (n = 46)
0
8
1
IS
15
5
15
26
Nore. The two major alleles (allele 1 = 9 repeats of the 40 base-pair elements; allele 2 = 10 repeats) have been detailed separately; the other alleles (6, 11, and 12 repeats have been grouped because of their very rare occurrence (Other)).
The Rouen and La Reunion subsamples were also considered separately for the APM analysis. For all subsamples considered, weighting factors applied and clinical classifications used, the t values obtained did not reject the null hypothesis of independent segregation between the marker locus and the disease locus (data not shown). We also calculated the t statistic according to the Green and Woodrow sibling method (Table 1). The t values were not statistically significant either for the narrow or the broad clinical classification (narrow definition: t = 1.06, NS; broad definition: t = -0.66, NS). 3.2. Association analysis The possible association between the 40 basepair repeat polymorphism and schizophrenia was studied (Table 2). The allelic distributions in control subjects and patients were not significantly different (x ’ = 1.58, df = 2, NS). No significant difference between the genotype distributions of control subjects and patients was found (x2 = 1.!5, df = 3, NS), and the Hardy-Weinberg equilibrium was conserved in both groups (schizophrenic patients: x2 = 0.24, df = 2, NS; control subjects: x2 = 0.11, df = 2, NS). The patients were subdivided into subgroups according to clinical diagnosis or familial history of schizophrenia and compared with control subjects. The results were again negative both for allelic dis-
tributions (familial cases: x2 = 1.13, df = 2, NS; clinical subgroups: paranoid subtype: xZ = 0.41, df = 2, NS; disorganized subtype: x2 = 2.97, df = 2, NS, undifferentiated subtype: x2 = 2.88, df = 2, NS) and for genotype distributions (familial cases: x2 = 1.94, df = 3, NS; clinical subgroups: paranoid subtype: x2 = 3.52, df = 3, NS; disorganized subtype: x2 = 3.11, df = 3, NS; undifferentiated subtype: x2 = 4.13, df = 3, NS). 4. Discussion Our analysis gave no evidence for linkage between a VNTR marker of the DAT gene and schizophrenia in a group of families containing individuals originating from metropolitan France and La Reunion Island. Lod score values at a recombination fraction of 8 = 0 were below -2, excluding linkage for simple Mendelian diseases. However, lod score results should be interpreted with caution in genetically complex diseases like schizophrenia because misspecification of genetic parameters can lead to false exclusion of linkage (Clerget-Darpoux et al., 1986). To minimize this problem, we chose a conservative procedure to analyze our data: an “affected only” analysis, which considered all unaffected subjects as “unknown phenotypes” while including their genotypes at the marker locus in the lod score computation. We also used nonparametric
S. Bodeau-Pt!an et al. /Psychiatry Research 59 (1995) l-6
methods of linkage analysis, which do not require specification of genetic parameters. The APM analysis also did not find significant results. Both the lod score (Ott, 1992) and the APM (Babron et al., 1993) methods are sensitive to misspecification of allelic frequencies at the marker locus. To circumvert this problem, allelic frequencies were calcuiated using data from relatives in both La R&union Island and metropolitan subsamples. The Green and Woodrow sibling method requires neither specification of genetic parameters nor allelic frequencies at the marker locus. This method also failed to provide evidence for linkage between the Taq I VNTR and schizophrenia in our sample. A case-control study was also carried out in a French metropolitan sample from Rouen. This approach offers a good alternative for detecting predisposition factors in genetically complex diseases (Strittmatter et al., 1993). In our population, there was no association between the 40 basepair repeat polymorphism of the DAT gene and schizophrenia. An identical result has been reported in a Chinese population (Li et al., 1994) and a very recent British study (Daniels et al., 1995). These linkage and association studies do not support any major involvement of the DAT gene in schizophrenia. A minor effect cannot be excluded, however, because of the limited power of the statistical methods, the sample size, and the informativity of the markers. Acknowledgments
References American Psychiatric Association. (1980) DSM-III: Diagnostic and Sfatisrical Manual of Mental Disorders. 3rd edn. APA, Washington, DC. Babron, M.C., Martinez, M., Bonai’ti-Pellit, C. and ClergetDarpoux, F. (1993) Linkage detection by the affectedpedigree-member method: what is really tested? Gene1 Epidemioi IO, 389-394.
Baron. M., Asnis, L. and Gruen, R. (1981) The Schedule for Schizotypal Personalities (SSP): a diagnostic interview for schizotypal features. Psychiafry Res 4, 213-228. Biiehnke, M. (1991) Allele frequency estimation from data on relatives Am J Hum Gene! 48, 22-25. Byerley, W., Coon, H., Hoff, M., Holik, J., Waldo, M., Freedman, R., Caron, M.G. and Giros, B. (1993a) Human dopamine transporter gene not linked to schizophrenia in multigenerational pedigrees. Hum Hered 43, 319-322. Byerley, W., Hoff, M., Holik, J., Caron, M. and Giros, B. (1993b) VNTR polymorphism for the human dopamine transporter gene. Hum Mat Genet 2, 335. Campion, D.. d’Amato, T., Bastard, C.. Laurent, C., Guedj, F., Jay, M., Gorwood, P.. Dolfus, S., Thibault, F., Petit, M.. Gorwood, P., Babron, M.C., Waksman, G., Martinez. M. and Mallet, J. (1994) Genetic study of dopamine D,, D,, and D4 receptors in schizophrenia. Psychiatry Res 51, 215-230.
Clerget-Darpoux, F., Bonai’ti-Pellie, C. and Hochez, J. (1986) Effects of misspecifying genetic parameters in lod score analysis. Biometrics 42. 393-399. d’Amato,T., Campion,D., Gorwood, P., Jay, M., SabatC, O., Petit, C., Abbar, M., Malafosse, A., Leboyer, M., Hillaire, D., Clerget-Darpoux, F., Feingold, J., Waksman, G. and Mallet, J. (1992) Evidence for a pseudoautosomal locus for schizophrenia: replication of a non-random segregation of alleles at the DXYS.14 locus. Br J Psychiatry 161, 59-62. Daniels,J., Williams, J., Asherson, P., McGuff~n, P. and Owen, M. (1995) No association between schizophrenia and polymorphisms within the genes for debrisoquine 4hydroxylase (CYP2D6) and the dopamine transporter (DAT). Am J Med Gener (Neuropsychiatr Genet) 60.85-87. Fyer, A.J., Endicott, J., Mannuzza, S. and Klein, D.F. (1985)
Schedule for Affective Disorders and Schizophrenia-Lifetime
We are grateful to B. Giros for providing the probe. This work received financial support from the Centre National de la Recherche Scientifique, the Association Francaise contre les Myopathies, the Groupement de Recherche et d’6tudes sur les GtnBmes, RhBne Poulenc Rorer, and the Assistance Publique des HBpitaux de Paris. Sylvie Bodeau-P6an was supported by a grant from the Ministkre de 1’Enseignement Supkrieur et de la Recherche, and Claudine Laurent by the Fondation pour la Recherche MCdicale.
Anxiety Version. New York State Psychiatric Institute, New York. Giros, B., El Mestikawy, S.. Godinot, N., Zheng, K., Han, H.. Yang-Feng, T. and Caron. M.G. (1992) Cloning, pharmacological characterization, and chromosome assignment of the human dopamine transporter. Mel Phormacol 42. 383-390.
Goldstein, M. and hutch, A.Y. (1992) Dopaminergic mechanisms in the pathogenesis of schizophrenia. FASEB J 6, 2413-2421.
Green. J.R. and Woodrow, J.C. (1977) Sibling method for detecting 31-35.
HLA-linked
genes in diseases.
Tissue Antigens 9,
6
S. Bodeau-Pian et al. /Psychiatry Research 59 (1995) 1-6
Horn, A.S. (1990) Dopamine uptake: a review of progress the last decade. Prog Neurobiol34, 387-400.
in
markers in human 283-290.
gene mapping
Am J Hum Cenet 51.
Lathrop, GM., Lalouel, J.M., Jullier, C. and Ott, J. (1985). Multilocus linkage analysis in humans: detection of linkage and estimation of recombination. Am J Hum Gener 37. 482-498.
Sano, A., Kondoh, K., Kakimoto, Y. and Kondo, I. (1993) A 40nucleotide repeat polymorphism in the human dopamine transporter gene. Hum Gene1 91, 405-406. Strittmatter, W.J., Saunders, A.M., Schmechel, D.. Pericak-
Li, T., Yang, L., Wiese, C., Xu, C.T., Zeng, Z., Giros, B., Caron, M., Moises, H.W. and Liu, X. (1994) No association between alleles or genotypes at the dopamine transporter gene and schizophrenia. Psychiarry Res 52, 17-23.
Vance, M.. Enghild, J., Salvesen, G.S. and Roses, A. (1993) Apolipoprotein E: high-avidity binding to amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Nail Acad Sci USA 90, l977- I98 I.
Mallet, J., Meloni, R. and Laurent, C. (1994) Catecholamine metabolism and psychiatric or behavioral disorders, Curr Opin Genet Dev 4, 419-426. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. McGue, M. and Gottesman, I. (1991) The genetic epidemiology of schizophrenia and the design of linkage studies. Eur Arch Psychiatry Clin Neurosci 240, I74- I8 I. Ott, J. (1992) Strategies for characterizing highly polymorphic
Uhl, G.R. (1992) Neurotransmitter transporters: a promising new gene family Trends Neurosci 15, 265-268. Vandenbergh, D.J., Persico, A.M., Hawkins, A.L., Griffin, C.A., Li, X., Jabs, E.W. and Uhl, G.R. (1992) Human dopamine transporter gene (DATI) maps to chromosome 5~15.3 and displays a VNTR. Genomics 14, 1104-l 106. Weeks, E.D. and Lange, K. (1988) The affected-pedigreemember method of linkage analysis. Am J Hum Genet 42, 315-326.