Linkage analysis of candidate loci in families with recurrent major depression

Molecular Psychiatry (1998) 3, 162 168 1998 Stockton Press All rights reserved 1359 4184/98 $12.00 ORIGINAL RESEARCH ARTICLE Linkage analysis of candidate loci in families with recurrent major depression J Balciuniene 1, Q-P Yuan 2, C Engström 3, K Lindblad 2, PO Nylander 3, M Sundvall 1, M Schalling 2, U Pettersson 1, R Adolfsson 3 and EE Jazin 1 1 Unit of Medical Genetics, Dept Genetics and Pathology, Box 589, Uppsala University, S-751 23 Uppsala, Sweden; 2 Neurogenetics Unit, Dept Molecular Medicine, Karolinska Institute and Karolinska Hospital, S-171 76 Stockholm, Sweden; 3 Dept Psychiatry, Umeå University, S-901 85 Umeå, Sweden Recurrent major depression, RMD, is characterized by the occurrence of depressive episodes in the absence of mania and/or hypomania. In linkage studies, RMD (or, in general, unipolar depression) are frequently grouped together with bipolar illnesses into a broad definition of affective disorders. However, twin studies suggest that RMD and bipolar disorders might have different genetic determinants. The objective of this study was to test a set of families with RMD for linkage to chromosomes that have been recently proposed to contain susceptibility loci for bipolar disorders: chromosomes 16, 18, 21 and the short arm of chromosome 4. We analysed five large families from the northern part of Sweden ascertained through a proband with RMD and containing several patients with RMD. For the genetic analysis, we included only severely affected individuals (those who had at least three episodes that required medical treatment) to increase the chances of finding a larger degree of genetic determination. The genetic model led to a total disease prevalence of 5% in females and 3% in males. We did not find significant evidence for linkage to any of the candidate chromosomes in the combined family set. Only one of the families showed a slight indication for linkage with markers from the pericentromeric region of chromosome 18. A genome scan analysis on an extended collaborative family material with severely affected individuals with RMD should be performed to evaluate whether RMD and bipolar disorders have a different genetic etiology. Keywords: unipolar depression; depressive disorder; bipolar disorder; chromosomes, human, pair 4; chromosomes, human, pair 16; chromosomes, human, pair 18; chromosomes, human, pair 21; disease susceptibility; genetic markers; human; lod score Introduction According to DSM-IV criteria 1 the essential feature of major depression (or unipolar depression) is one or more major depressive episodes without a history of mania or hypomania. Major depression has been subdivided into major depression, single episode (only one episode) or recurrent major depression (two or more episodes separated by at least 2 months). Life time prevalence for major depression varies between 6% and 16% in women and between 3% and 9% in men. 2 Several studies have shown that unipolar disorders are familial. 3,4 However, different twin and adoption studies attempting to estimate a degree of heritability resulted in conflicting findings. 4 10 One of the reasons might be that different studies applied different diagnostic criteria to ascertain their sample set. Recently, McGuffin and colleagues 11 conducted a twin Correspondence: EE Jazin, Dept Medical Genetics, Box 589, Uppsala University, S-751 23 Uppsala, Sweden. E-mail: elena.jazin medgen.uu.se Received 28 August 1997; revised 17 December 1997; accepted 22 December 1997 study based on a hospital register of twins, ascertained through a major depression proband. They restricted their sample to cases of more severe pathology. The estimates of heritability for major depression were between 48% and 75%, depending on the assumed population risk. Shorter duration of episodes and an occurrence of more than two depressive episodes were associated with a higher degree of genetic determination. No evidence for an effect of shared family environment was found. Unipolar and bipolar disorders are clearly differentiated by the absence or presence of episodes of mania. However, in linkage studies, major depression has usually been included together with bipolar disorders into a broad definition of affective illnesses. There have been few recent attempts to perform association analyses exclusively on unipolar disorder cases, 12,13 while linkage studies on families with unipolar depression have been attempted only by one group during the 70s and 80s. 14,15 On the other hand, many linkage studies performed during the last 20 years on families with bipolar disorders have contributed to a long list of chromosomal regions with evidence for linkage to bipolar dis-

orders. 16 The original observations have not yet been clearly replicated by others. An initial report of linkage on chromosome 11p in an Old Order Amish family 17 was not replicated by others 18,19 and even weakened by a follow-up study of the same family. 20,21 Positive scores on chromosome Xq27 28 22,23 also were followed by several conflicting results 24,25 but supported in one. 26 More recent publications pointed to the possibility of linkage to chromosomes 4p, 27 18, 28 31 21q, 32 34 16p 35 and Xq24 26. 36 However, these regions are large. For example, the proposed region on chromosome 18 spans about 136 cm and includes both chromosomal arms. Several studies observed genetic anticipation in families with bipolar 37,38 and unipolar 39 cases. Investigations of trinucleotide repeat expansions (a possible explanation for genetic anticipation) in families with bipolar disorders resulted in contradictory findings. 40 43 These findings together with different linkage results suggest heterogeneity in affective illnesses. Therefore, a stringent distinction of different phenotypes of affective illnesses might help to unravel the genetic components of those disorders. The objective of our study was to select families with RMD cases and analyze them for linkage to chromosomes suggested to contain susceptibility loci for bipolar disorders. Since the regions that contain potential loci are large, our first approach was to screen for linkage to the complete autosomes 16, 18 and 21 and the short arm of chromosome 4. Chromosome X was excluded as a candidate because the structure of our families did not seem to be compatible with X-linked inheritance. Materials and methods Family ascertainment and diagnosis Fourteen multicase families with affective disorders were initially identified through a systematic screening of all records at the Department of Psychiatry, Umeå, Sweden. The families were ascertained through bipolar I, bipolar II and RMD probands. The complete experimental sample set consisted of 337 individuals. For further investigations, we have selected only families with RMD probands when at least two other members with RMD were available for the study. In total, five families with RMD (Figure 1; Table 1) consisting of 60 persons available for the study were chosen for the analysis. From 23 individuals having diagnosis of affective illnesses, 20 were diagnosed as having RMD, two had schizoaffective disorder and one had bipolar I. Diagnosis was performed in accordance with DSM- IV criteria. 1 All individuals included in this study gave informed consent for the use of their clinical data for research purposes. In order to avoid identification, the samples were coded with blood numbers. DNA analysis Genomic DNA was prepared from peripheral blood lymphocytes by phenol-chloroform extraction. 44 Microsatellite polymorphisms were amplified by polymerase chain reaction (PCR). Six microsatellite markers located on chromosome 18q were typed using a radioactive detection method as described elsewhere. 45 The others were analyzed using the multiplex fluorescent detection method specified previously. 44 The fluorescent markers listed in the Weber set 6 46 were ordered from Genset (Paris, France). PCR reactions had a total volume of 10 l containing 50 ng DNA, 2.5 pmol of each primer, 12.5 mm of each dntp and 0.5 U Ampli- Taq polymerase (Perkin Elmer, Norwalk, CT, USA). After an initial step of 95 C for 5 min, PCR conditions were as follows: 10 cycles of denaturation at 94 C for 30 s, annealing at 50 58 C for 45 s, and elongation at 72 C for 1 min, 20 cycles of 89 C for 30 s, 50 58 C for 1 min, 72 C for 1 min and a final extension step at 72 C for 10 min. Amplification reactions were performed in a PTC-225 thermocycler (MJ Research, Watertown, MA, USA). Fluorescent PCR products were resolved on 4% polyacrylamide gel and detected using an ABI 377 DNA Sequencer (Applied Biosystems, Foster City, CA, USA). Sizes of marker alleles were defined using Genescan analysis 2.0.1 software (Applied Biosystems). Analyzed data were imported to Genotyper version 1.1 software package (Applied Biosystems) for an allele calling procedure and to make a final table of genotypes. Overall, we have genotyped 27 microsatellite markers spanning the chromosomes 16, 18, 21 and the short arm of chromosome 4 with an average distance of 12 cm. There were two gaps (25 and 29 cm) on chromosome 16q telomeric end, two gaps of 21 cm on 21q telomeric end and 16p centromeric region. Distances for the markers were taken from the homepage of the Cooperative Human Linkage Center (http://www.chlc.org/homepage.html). Linkage analysis Pairwise linkage analysis was carried out with the MLINK option of the LINKAGE software package. 47 Affection status was defined as described in the Results section. As the structure of our pedigrees suggested that X-linked mode of inheritance is unlikely, we used models for dominant and recessive trait only. The frequency of the disease-causing allele in the dominant model was assumed to be 3%, while in the recessive model the disease allele frequency was 25%. The mutation rate for both models was set to be 10 6 and recombination rates for females and males were equal. To account for the age-dependent penetrance the individuals were classified into three groups: 20 40 years old, 40 60 years old and older than 60. To stress the differences in penetrance and phenocopy rates of the two sexes, females and males were divided into different liability classes. Maximum penetrance for females was 70% and for males 50% and phenocopy rates were 0.5% and 0.1%, respectively. The models led to a total disease prevalence of 5% in females and 3% in males. The calculations were performed assuming 100% homogeneity. Marker allele frequencies were calculated from the genotypes of typed persons. 163

164 Figure 1 Pedigrees of five families from northern Sweden ascertained through probands with recurrent major depression. Family numbers are indicated above each pedigree. Description of the symbols is presented in the lower right corner of the picture. Simulation analysis Simulation analysis was performed using SLINK of the LINKAGE package 47 to determine the probability of detecting linkage in our family set. The same genetic models and affection phenotype definition as described in the previous section and the results section, respectively, were used for this analysis. Genotypes were simulated for a marker with five alleles of equal frequencies. The marker was assumed to lie at a recombination distance of 0.01 (linked) or 0.5 (unlinked) to the disease gene. All the families were considered to be homogeneous. Simulations were carried out for 500 replicates. The results for dominant models and recessive models assuming 100% homogeneity under assumptions of tight linkage ( = 0.01) and no linkage ( = 0.5) are summarized in Table 2. For the combined family set under the assumption of tight linkage, the dominant model resulted in an average lod score of 1.7 and a maximum lod score of 4.5. The recessive model resulted in an average lod score of 1.17 and a maximum lod score of 3.5. Simulation results for each family separately indicated that only two families (No. 26 and No. 35, Figure 1) were large enough to contribute with lod scores of suggestive significance (higher than 1). Results Definition of the affected phenotype for linkage analysis The criterion for a person to be considered affected in the linkage analysis was determined following the suggestion by McGuffin et al that more than two depressive episodes and shorter duration of episodes are associated with higher degree of heritability. 11 Therefore, for the linkage analysis, we decided to include as affected only individuals who had psychiatric records revealing 3 depressive episodes (Table 1). Persons who had undergone a test of personality traits and did not indicate any signs of depression or other psychiatric disorders during an interview, were considered as unaffected. Individuals who did not meet the criteria mentioned above, did not have psychiatric records from a psychiatric institution, had other psychiatric diagnosis or persons younger than 20 years old were excluded from the linkage analysis (they are indicated with (?) in Figure 1). As seen in Table 1, there were 19 persons who met the criterion for affection status described above. Seventeen of them were patients with RMD and two had schizoaffective disorder. In spite of a tendency to con-

Table 1 Characteristics of affected individuals included in the analysis Family Pedigree Sex Age of Number of Number of No No onset a episodes b years c III:1 M 52 6 6 4 III:7 F 54 13 38 III:11 M 32 5 38 II:5 d M 65 3 4 12 II:7 M 38 20 29 III:3 e F 32 3 12 II:8 M 49 4 23 II:11 F 39 7 15 26 II:14 F 33 4 23 III:1 F 19 8 28 III:4 f M 25 4 15 III:2 F 61 8 15 III:5 F 49 22 24 35 III:6 F 63 4 10 III:7 M 33 9 32 IV:1 F 33 4 18 II:5 F 37 20 26 205 III:4 F 15 3 23 III:6 F 36 3 9 a Age of onset according to the first episode when treatment was received. b According to the medical records; therefore this is a minimum number of episodes; patients own recollections about previous episodes are not included. c Years between the first and last episode in which treatment was received. d Committed suicide. e,f Diagnosed as schizoaffective. sider schizoaffective disorders to be more related to bipolar disorders than to unipolar disorders, we decided to include the two schizoaffective persons into the analysis as affected because they had a substantial number of depressive episodes during the course of the illness and, thus, perfectly complied with the criterion. However, the exclusion of those individuals did not significantly change the results (not shown). Three recurrent depression patients who did not meet the criterion described above and the bipolar I patient were specified as phenotype unknown and did not influence the analysis significantly. Two-point linkage analysis for candidate chromosomes Results from two-point lod score analysis of the tested markers are summarized in Table 3. For the whole sample set, none of the loci tested disclosed a compelling evidence for linkage. Examining the results for each family separately, only family 26 gave lod score values higher than one for three markers in the pericentromeric region of chromosome 18 (D18S542, D18S34 and D18S69) under the dominant genetic model, and for two of the markers (D18S542 and D18S34) under the recessive model. In the table the lod scores are also indicated for other markers with values higher than Table 2 Simulation results for dominant and recessive genetic models a A. Linked at = 0.01 Family No Dominant model Recessive model Mean Max Mean Max 4 0.32 0.97 0.12 0.61 12 0.29 0.84 0.38 1.00 26 0.62 1.66 0.49 1.64 35 0.50 1.68 0.26 1.04 205 0.35 0.83 0.27 0.74 Study 1.73 4.51 1.17 3.50 B. Unlinked at = 0.5 Family No Dominant model Recessive model Mean Max Mean Max 4 0.11 0.88 0.06 0.52 12 0.10 0.84 0.10 1.00 26 0.12 1.38 0.07 0.98 35 0.09 1.39 0.09 0.93 205 0.10 0.82 0.10 0.70 Study 0.13 2.55 0.09 2.05 C Dominant model Recessive model Lod 1 Lod 2 Lod 3 Lod 1 Lod 2 Lod 3 % of 72.6 38.2 10.8 52.6 17.2 1.0 replicates D Dominant model Recessive model Lod Lod Lod Lod Lod Lod 0.5 1 1.5 0.5 1 1.5 % of 8.6 3.6 1.0 5.0 1.4 0.6 replicates a Simulation results are represented for both dominant and recessive genetic models. Parts (A) and (B) show the average and maximum lod scores obtained from 500 simulated replicates for each individual pedigree and for the combined family set under linked and unlinked order, respectively. Parts (C) and (D) show a percentage of replicates with lod scores higher than a given lod score threshold in the combined family set under linkage and no linkage conditions, respectively. one standard deviation above the expected average lod score, such as, marker D4S403 in family 4, marker D16S2624 in family 12, marker D18S59 in family 35, marker D16S422 and three markers, D18S69, D18S51, D18S43, on the long arm of chromosome 18 in family 205. 165

166 Table 3 Linkage analysis results for chromosomes 4p, 16, 18 and 21 a Recombination Family No 4 12 26 35 205 All families distance (cm) Model Markers dom rec dom rec dom rec dom rec dom rec dom rec 11.8 8.1 8.6 5 14.3 20.7 7.3 7.5 29.2 25.3 6.1 15.3 17.4 13.7 7 10.3 16.6 10.8 9.4 5.3 11.3 20.8 4 D4S2366 3.08 1.71 D4S403 0.70 0.54 0.83 0.23* D4S2639 2.83 0.99 D4S2397 0.98 0.91 D4S2408 2.17 0.93 16 D16S2622 0.39 0.43 D16S748 1.27 1.24 D16S403 2.45 1.72 D16S769 1.68 1.11 D16S753 2.02 1.96 D16S2624 0.72 1.50 0.79 D16S422 0.67 0.11 0.93 18p D18S59 0.74 0.59 0.64* D18S976 0.86 0.34 D18S843 0.46 0.44* D18S542 1.34 1.35 1.06 0.55 18q D18S34 1.25 1.31 0.69* 0.40* D18S69 1.23 0.63 0.60* 0.11 D18S64 0.63 0.94 D18S51 0.83 0.66* 0.38 D18S43 0.49 1.42 1.33 MBP 2.46 0.84 21 D21S1437 2.35 2.38 D21S1435 1.69 1.37 D21S1270 2.27 2.70 D21S1440 0.33 1.69 D21S1446 0.63 1.35 a The table shows only the lod scores higher than one standard deviation above the expected average lod score obtained by simulation for each family. Also, lod score values higher than one are in italic. For the whole family set, overall lod scores are shown in the right side of the table ( All families ) for = 0, except for the lod scores indicated with * where they reach max at = 0.1. The markers are listed according to their positions in each chromosome (second column), and the recombination distances (in cm) between them are indicated on the left side. Family numbers (families are shown in Figure 1) are indicated in the top row of the table. The genetic models under which the given lod scores were obtained were: dom (dominant model) and rec (recessive model). Discussion We have selected familes with RMD and performed a screening on the chromosomes recently suggested to contain susceptibility loci for bipolar disorders (chromosomes 4p, 16, 18 and 21). We found no significant or suggestive evidence for linkage to the candidate chromosomes in the whole family set while in one of the families there was an indication for linkage with markers from the pericentromeric region of chromosome 18. According to the results from the simulation analysis (Table 2), estimated average lod score under linked conditions for the whole sample set was 1.7 for the dominant model (1.2 for recessive) and the probability of obtaining a lod score over 3 was 10% (1% for recessive). Simulations under no linkage gave an average lod score of 0.13 (0.13 for recessive) and a 1% probability of obtaining a lod score over 1.5 (0.6% for recessive). These results suggested that our material might be able to provide results of pointwise significance testing candidate loci in the whole family set, though it would be too weak for a whole genome scan. 48 Moreover, the absence of compelling evidence for linkage for the whole family set should be taken with caution, since the size of the material is small. In previous studies, unipolar and bipolar disorders have usually been grouped into a broad definition of affective illnesses. 27,28,32,35,49 53 For example, in the most recent studies 49 53 three models of affection status have been used: model I considered as affected individuals with bipolar I and schizoaffective-bipolar diag-

noses, model II included those diagnosed under model I and those with bipolar II diagnosis, model III included persons diagnosed under model II and all individuals with unipolar recurrent depression. Therefore, unipolar disorders were not considered as an independent genetic disease. This was probably influenced by previous twin studies which showed a low degree of heritability (21 24%) of unipolar disorders 9,54 while heritability of bipolar disorders was much higher (up to 70%). 3 However, recent twin studies 10,11 using more stringent diagnostic criteria for major depression resulted in considerably higher heritability values (48 75%). These latest studies, together with our observation that families with bipolar probands usually have both unipolar and bipolar diagnoses within the family, while families with RMD probands had almost exclusively unipolar depression cases, suggest that severe unipolar depression might have a different or partially different genetic etiology than bipolar illnesses. In our study, the criterion for a person to be considered affected in the linkage analysis was that he or she had at least three depressive episodes that required medical treatment. The criterion was based on the observations by McGuffin et al 11 indicating that more than two episodes and shorter duration of episodes are associated with a higher degree of heritability. We thus selected the restrictive criterion for affection status described above to improve the chance to find genetic components and to reduce heterogeneity in the selected sample. Furthermore, all the families descended from the neighboring regions of the northern part of Sweden, an isolated region in the past. Therefore, we assumed that we had a genetically homogeneous sample and we conducted all our calculations assuming 100% homogeneity. The affection trait model was based on conservative genetic parameters for bipolar disorders, but we assumed different penetrance and phenocopy rates for females and males. As the mode of inheritance in our family set seemed to be inconsistent with X-linked inheritance, we tested only dominant and recessive models. The overall prevalence of the disease in our analysis was 5% in females and 3% in males, corresponding to the lower end of range for the assessed lifetime prevalence of DSM-III-R major depression. 2 Such reduced prevalence would be expected for the most severe cases of major depression. For the whole family set, pairwise linkage analysis of all candidate regions did not reveal any suggestive or significant linkage values. However, the analysis of each family separately showed three markers in the pericentromeric region of chromosome 18 (D18S542, D18S34 and D18S69) with lod scores higher than 1 in family 26 (Table 3). Interestingly, the marker D18S542 lies in the same region as markers with the highest scores in the analysis of families with mixed bipolar and unipolar cases. 28 However, according to the simultations there is a 2% probability of obtaining a lod score higher than 1 by chance in this family (data not shown). If there is a susceptibility locus on chromosome 18 in just one family, this might indicate heterogeneity in our family set or the presence of one of several loci contributing to the cause of the disease in families with unipolar disorders. The main question, whether there are common genetic determinants for unipolar and bipolar depression, remains to be answered. A possibility is that a certain part of the genetic components overlap in these illnesses while other genetic factors might be different. In summary, we have used stringent criteria to select families with several cases of severe RMD to test for linkage to candidate chromosomes for bipolar disorders (4p, 16, 18 and 21). All the families together did not result in a suggestive score for any of these chromosomes. However, a few markers in the pericentromeric region of chromosome 18 showed suggestive scores in one family. A complete genome screening on extended collaborative families with diagnosis exclusively of severe RMD might provide new susceptibility locus/loci for unipolar disorders. Such screening will help to distinguish whether unipolar and bipolar illnesses are caused by the same genetic factors or represent conditions of different genetic etiology. Acknowledgements We thank all the members of the families for participating in this study and Barbro Karlsson and Gunnel Johansson for the skilful work collecting the family material. This work was supported by grants K97-19X- 10211-05B, K97-13X-10909-04AK and B96-21X-10412-04A from the Swedish Medical Research Council, by a grant BMH4-97-2466 from Biomed 2, by Söderström- Königska Foundation and by a grant from the Beijer foundation. MS is the recipient of the Ireland Award. References 1 American Psychiatric Association. Diagnostic and Statistical Manual. Fourth edn. American Psychiatric Association: Washington, DC, 1994. 2 Smith AL, Wiessman MM. Epidemiology. In: Paykel ES (ed). Handbook of Affective Disorders. Churchill Livingstone: New York, NY, 1992, pp 111 130. 3 Tsuang MT, Faraone SV. The Genetics of Mood Disorders. John Hopkins University Press: Baltimore, MD, 1990. 4 McGuffin P, Katz R. Nature, nurture and affective disorders. In: Deakin JWK (ed). The Biology of Depression. Gaskell Press: London, 1986, pp 26 51. 5 von Knorring AL, Cloninger CR, Bohman M, Sigvardsson S. An adoption study of depressive disorders and substance abuse. Arch Gen Psychiatry 1983; 40: 943 950. 6 Cadoret RJ, O Gorman TW, Heywood E, Troughton E. Genetic and environmental factors in major depression. J Affect Disord 1985; 9: 155 164. 7 Wender PH, Kety SS, Rosenthal D, Schulsinger F, Ortmann J, Lunde I. Psychiatric disorders in the biological and adoptive families of adopted individuals with affective disorders. Arch Gen Psychiatry 1986; 43: 923 929. 8 Andrews G, Stewart G, Allen R, Henderson AS. The genetics of six neurotic disorders: a twin study. J Affect Disord 1990; 19: 23 29. 9 Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. A population-based twin study of major depression in women. The impact of varying definitions of illness. Arch Gen Psychiatry 1992; 49: 257 266. 167

168 10 Kendler KS, Neale MC, Kessler RC, Health AC, Eaves LJ. The lifetime history of major depression in women. Reliability of diagnosis and heritability. Arch Gen Psychiatry 1993; 50: 863 870. 11 McGuffin P, Katz R, Wilkins S, Rutherford J. A hospital-based twin register of the heritability of DSM-IV unipolar depression. Arch Gen Psychiatry 1996; 53: 129 136. 12 Manki H, Kanba S, Muramatsu T, Higuchi S, Suzuki E, Matsushita S et al. Dopamine D2, D3 and D4 receptor and transporter gene polymorphisms and mood disorders. J Affect Disord 1996; 40: 7 13. 13 Ogilvie AD, Battersby S, Bubb VJ, Fink G, Harmar AJ, Goodwim GM et al. Polymorphism in serotonin transporter gene associated with susceptibility to major depression. Lancet 1996; 347: 731 733. 14 Tanna VL, Winokur G, Elston RC, Go RC. A linkage study of pure depressive disease: the use of the sib-pair method. Biol Psychiatry 1976; 11: 767 771. 15 Tanna VL, Wilson AF, Winokur G, Elston RC. Linkage analysis of pure depressive disease. J Psychiatr Res 1989; 23: 99 107. 16 Risch N, Botstein D. A manic depressive history [news]. Nat Genet 1996; 12: 351 353. 17 Egeland JA, Gerhard DS, Pauls DL, Sussex JN, Kidd KK, Allen CR et al. Bipolar affective disorders linked to DNA markers on chromosome 11. Nature 1987; 325: 783 787. 18 Mendlewicz J, Leboyer M, De Bruyn A, Malafosse A, Sevy S, Hirsch D et al. Absence of linkage between chromosome 11p15 markers and manic-depressive illness in a Belgian pedigree. Am J Psychiatry 1991; 148: 1683 1687. 19 Mitchell P, Waters B, Morrison N, Shine J, Donald J, Eisman J. Close linkage of bipolar disorder to chromosome 11 markers is excluded in two large Australian pedigrees. J Affect Disord 1991; 21: 23 32. 20 Kelsoe JR, Ginns EI, Egeland JA, Gerhard DS, Goldstein AM, Bale SJ et al. Re-evaluation of the linkage relationship between chromosome 11p loci. Nature 1989; 342: 238 243. 21 Pauls DL, Gerhard DS, Lacy LG, Hostetter AM, Allen CR, Bland SD et al. Linkage of bipolar affective disorders to markers on chromosome 11p is excluded in a second lateral extension of Amish pedigree 110. Genomics 1991; 11: 730 736. 22 Mendlewicz J, Fleiss JL. Linkage studies with X-chromosome markers in bipolar (manic-depressive) and unipolar (depressive) illnesses. Biol Psychiatry 1974; 9: 261 294. 23 Baron M, Risch N, Hamburger R, Mandel B, Kushner S, Newman M et al. Genetic linkage between X-chromosome markers and bipolar affective illness. Nature 1987; 326: 289 292. 24 Gejman PV, Detera Wadleigh S, Martinez MM, Berrettini WH, Goldin LR, Gelernter J et al. Manic depressive illness not linked to factor IX region in an independent series of pedigrees. Genomics 1990; 8: 648 655. 25 Baron M, Freimer NF, Risch N, Lerer B, Alexander JR, Straub RE et al. Diminished support for linkage between manic depressive illness and X chromosome markers in three Israeli pedigrees. Nat Genet 1993; 3: 49 55. 26 Lucotte G, Landoulsi A, Berriche S, David F, Babron MC. Manic depressive illness is linked to factor IX in a French pedigree. Ann Genet 1992; 35: 93 95. 27 Blackwood DH, He L, Morris SW, McLean A, Whitton C, Thomson M et al. A locus for bipolar affective disorder on chromosome 4p. Nat Genet 1996; 12: 427 430. 28 Berrettini WH, Ferraro TN, Goldin LR, Weeks DE, Detera Wadleigh S, Nurnberger JI Jr et al. Chromosome 18 DNA markers and manicdepressive illness: evidence for a susceptibility gene. Proc Natl Acad Sci USA 1994; 91: 5918 5921. 29 Stine OC, Xu J, Koskela R, McMahon FJ, Gschwend M, Friddle C et al. Evidence for linkage of bipolar disorder to chromosome 18 with a parent-of-origin effect. Am J Hum Genet 1995; 57: 1384 1394. 30 Freimer NB, Reus VI, Escamilla MA, McInnes LA, Spesny M, Leon P et al. Genetic mapping using haplotype, association and linkage methods suggests a locus for severe bipolar disorder (BPI) at 18q22 q23. Nat Genet 1996; 12: 436 441. 31 Coon H, Hoff M, Holik J, Hadley D, Fang N, Reimherr F et al. Analysis of chromosome 18 DNA markers in multiplex pedigrees with manic depression. Biol Psychiatry 1996; 39: 689 696. 32 Straub RE, Lehner T, Luo Y, Loth JE, Shao W, Sharpe L et al. A possible vulnerability locus for bipolar affective disorder on chromosome 21q22.3. Nat Genet 1994; 8: 291 296. 33 Gurling H, Smyth C, Kalsi G, Moloney E, Rifkin L, O Neill J et al. Linkage findings in bipolar disorder [letter; comment]. Nat Genet 1995; 10: 8 9. 34 Detera Wadleigh SD, Badner JA, Goldin LR, Berrettini WH, Sanders AR, Rollins DY et al. Affected-sib-pair analyses reveal support of prior evidence for a susceptibility locus for bipolar disorder, on 21q. Am J Hum Genet 1996; 58: 1279 1285. 35 Ewald H, Mors O, Flint T, Koed K, Eiberg H, Kruse TA. A possible locus for manic depressive illness on chromosome 16p13. Psychiatr Genet 1995; 5: 71 81. 36 Pekkarinen P, Terwilliger J, Bredbacka PE, Lönnqvist J, Peltonen L. Evidence of a predisposing locus to bipolar disorder on Xq24 q27.1 in an extended Finnish pedigree. Genome Res 1995; 5: 105 115. 37 McInnis MG, McMahon FJ, Chase GA, Simpson SG, Ross CA, DePaulo JR, Jr. Anticipation in bipolar affective disorder. Am J Hum Genet 1993; 53: 385 390. 38 Nylander PO, Engstrom C, Chotai J, Wahlstrom J, Adolfsson R. Anticipation in Swedish families with bipolar affective disorder. J Med Genet 1994; 31: 686 689. 39 Engstrom C, Thornlund AS, Johansson EL, Langstrom M, Chotai J, Adolfsson R et al. Anticipation in unipolar affective disorder. J Affect Disord 1995; 35: 31 40. 40 Oruc L, Lindblad K, Verheyen GR, Ahlberg S, Jakovljevic M, Ivezic S et al. CAG repeat expansions in bipolar and unipolar disorders [letter]. Am J Hum Genet 1997; 60: 730 732. 41 O Donovan MC, Guy C, Craddock N, Bowen T, McKeon P, Macedo A et al. Confirmation of association between expanded CAG/CTG repeats and both schizophrenia and bipolar disorder. Psychol Med 1996; 26: 1145 1153. 42 Vincent JB, Klempan T, Parikh SS, Sasaki T, Meltzer HY, Sirugo G et al. Frequency analysis of large CAG/CTG trinucleotide repeats in schizophrenia and bipolar affective disorder. Mol Psychiatry 1996; 1: 141 148. 43 Guy C, Bowen T, Daniels JK, Speight G, McKeon P, Mynett-Johnson L et al. Exclusion of expansion of 50 CAG/CTG trinucleotide repeats in bipolar disorder. Am J Psychiatry 1997; 154: 1146 1147. 44 Lindqvist AK, Magnusson PK, Balciuniene J, Wadelius C, Lindholm E, Alarcon Riquelme ME et al. Chromosome-specific panels of tri- and tetranucleotide microsatellite markers for multiplex fluorescent detection and automated genotyping: evaluation of their utility in pathology and forensics. Genome Res 1996; 6: 1170 1176. 45 Forsman K, Graff C, Nordstrom S, Johansson K, Westermark E, Lundgren E et al. The gene for Best s macular dystrophy is located at 11q13 in a Swedish family. Clin Genet 1992; 42: 156 159. 46 Sheffield VC, Weber JL, Buetow KH, Murray JC, Even DA, Wiles K et al. A collection of tri- and tetranucleotide repeat markers used to generate high quality, high resolution human genome-wide linkage maps. Hum Mol Genet 1995; 4: 1837 1844. 47 Cottingham RW Jr, Idury RM, Schaffer AA. Faster sequential genetic linkage computations. Am J Hum Genet 1993; 53: 252 263. 48 Lander E, Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 1995; 11: 241 247. 49 Detera-Wadleigh SD, Badner JA, Yoshikawa T, Sanders AR, Goldin LR, Turner G et al. Initial genome scan of the NIMH genetics initiative bipolar pedigrees: chromosomes 4, 7, 9, 18, 19, 20 and 21q. Am J Med Genet 1997; 74: 254 262. 50 Stine OC, McMahon FJ, Chen L, Xu J, Meyers DA, MacKinnon DF et al. Initial genome screen for bipolar disorder in the NIMH genetics initiative pedigrees: chromosomes 2, 11, 13, 14, and X. Am J Med Genet 1997; 74: 263 269. 51 Edenberg HJ, Foroud T, Conneally PM, Sorbel JJ, Carr K, Crose C et al. Initial genomic scan of the NIMH genetics initiative bipolar pedigrees: chromosomes 3, 5, 15, 16, 17, and 22. Am J Med Genet 1997; 74: 238 246. 52 NIMH Genetics Initiative Group. Genomic survey of bipolar illness in the NIMH genetics initiative pedigrees: a preliminary report. Am J Med Genet 1997; 74: 227 237. 53 Rice JP, Goate A, Williams JT, Bierut L, Dorr D, Wu W et al. Initial genome scan of the NIMH genetics initiative bipolar pedigrees: chromosomes 1, 6, 8, 10, and 12. Am J Med Genet 1997; 74: 247 253. 54 McGuffin P, Katz R, Rutherford J. Nature, nurture and depression: a twin study. Psychol Med 1991; 21: 329 335.