Sex-specific association between bipolar affective disorder in women and GPR50, an X-linked orphan G protein-coupled receptor

Article metrics


GPR50 is an orphan G protein-coupled receptor (GPCR) located on Xq28, a region previously implicated in multiple genetic studies of bipolar affective disorder (BPAD). Allele frequencies of three polymorphisms in GPR50 were compared in case–control studies between subjects with BPAD (264), major depressive disorder (MDD) (226), or schizophrenia (SCZ) (263) and ethnically matched controls (562). Significant associations were found between an insertion/deletion polymorphism in exon 2 and both BPAD (P=0.0070), and MDD (P=0.011) with increased risk associated with the deletion variant (GPR50Δ502–505). When the analysis was restricted to female subjects, the associations with BPAD and MDD increased in significance (P=0.00023 and P=0.0064, respectively). Two other single-nucleotide polymorphisms (SNPs) tested within this gene showed associations between: the female MDD group and an SNP in exon 2 (P=0.0096); and female SCZ and an intronic SNP (P=0.0014). No association was detected in males with either MDD, BPAD or SCZ. These results suggest that GPR50Δ502–505, or a variant in tight linkage disequilibrium with this polymorphism, is a sex-specific risk factor for susceptibility to bipolar disorder, and that other variants in the gene may be sex-specific risk factors in the development of schizophrenia.


Bipolar affective disorder (BPAD) is a severe psychiatric disorder affecting approximately 1% of the world's population, and shows no difference in lifetime prevalence between male and female subjects. Twin and adoption studies have demonstrated a strong genetic component, with a concordance in BPAD between monozygotic twins of 60%.1 Major depressive disorder (MDD) has a lifetime prevalence of 17% with women twice as likely as men to develop the disorder.2 Estimates of the heritability of MDD vary, but a meta-analysis of studies gives a point estimate of heritability of liability to MDD of 0.37.3 Schizophrenia (SCZ), as with BPAD, has an estimated frequency of 1% in the population, but heritability estimates suggest that it has a larger genetic component than either BPAD or MDD, with monozygotic twins giving a point estimate of comorbidity of 0.81 in another recent meta-analysis.4 Despite the strong genetic component in these major psychiatric disorders, there are also strong environmental influences.

Linkage to Xq28 has been studied many times in BPAD. Two loci, colour-blindness (CB), and glucose-6-phosphate dehydrogenase (G6PD), have been detected through linkage and association in more than one population.5 Most significant are LOD scores of 8.1 and 7.35 between CB and BPAD in the American and Belgian populations, respectively, although reanalysis of much of the data from positive linkage results on the X chromosome has resulted in much reduced evidence for linkage and suggestions of ascertainment bias.6, 7, 8, 9 Several more recent studies have again renewed interest in the distal end of Xq, although the region implicated in these studies (Xq24–28) is larger than that depicted in the earlier reports.5, 10, 11, 12, 13

GPR50 is an X-linked orphan G protein-coupled receptor (also known as H9, melatonin-related receptor or ML1X) located on Xq28, and was initially cloned from a human pituitary cDNA library.14, 15 The gene encodes a protein of 617 amino acids that is 45% identical overall to human melatonin receptors MTNR1A and MTNR1B, with identity increasing to 55% when comparing just the transmembrane domains.15 Despite its close structural relationship to the melatonin receptors, GPR50 does not bind either [125I]melatonin or [3H]melatonin.15, 16

RNA in situ hybridisation experiments with human GPR50 detected expression in the mediobasal hypothalamus, in a region containing the ventromedial and arcuate nuclei; and the paraventricular nucleus, as well as in the infundibular stalk.15 Expression was also detected in the pituitary gland.15 The hypothalamus has a punctate pattern of expression. In contrast, the expression pattern in the pituitary is widespread and heterogeneous, occurring in both the pars tuberalis and the pars distalis. The hypothalamus has been implicated in a number of physiological and behavioural responses. These include endocrine regulation (dorsal medial-ventromedial nuclei), reproduction, circadian and stress responses (dorsomedial hypothalamus), fluid balance, ingestive behaviours (arcuate nucleus and dorsomedial hypothalamus) and thermoregulation (lateral hypothalamus). The pituitary controls the endocrine system through secretion of hormones. This suggests a role for GPR50 and its ligand in neuroendocrine function.

The combination of map position on Xq28 and expression pattern makes GPR50 a positional and functional candidate in affective disorder.

Materials and methods

The study was approved by the Scottish Multicentre Research Ethics Committee.

Clinical assessment of patients

Subjects were in-patients or outpatients of hospitals in South East or South Central Scotland, and gave their informed consent for their inclusion in the study. Subjects were interviewed by an experienced psychiatrist and a blood was sample given for DNA extraction. Diagnoses were made according to DSM-IV criteria17 based on case note review and interview using The Schedule for Affective Disorders and Schizophrenia — lifetime version (SADS-L).18 Final diagnoses were reached by consensus between two trained psychiatrists (DB and WM). Age at onset was defined as the age when the subject first received professional help for psychiatric symptoms and ‘early’ onset was defined as less than 25 years of age.

Subjects in the group bipolar affective disorder (BPAD) included those with diagnoses Bipolar 1 and Bipolar 2, and all had a history of at least two episodes of major depression. Subjects included in the MDD group had a history of at least two episodes of major depression. The interview of subjects with MDD included additional questions to elicit symptoms consistent with ‘bipolar spectrum disorder’ as defined by Ghaemi et al.19 The criteria for ‘bipolar spectrum disorder’ include: brief episodes of hypomanic symptoms of any duration less than 4 days; a first or second degree relative with bipolar disorder; symptoms of hypomania induced by antidepressant medication. In the first stages of analysis, subjects with MDD and ‘bipolar spectrum’ features were included in the group of major depression. The nosological status of ‘bipolar spectrum’ remains uncertain, but there are strong arguments for including these subjects in the bipolar group.20 Analyses were therefore carried out with MDD excluding ‘spectrum’, and bipolar disorder including ‘spectrum’.

Control subjects were drawn from the same population in South East and South Central Scotland, and recruited from donors for the Scottish National Blood Transfusion service or from hospital staff.

The composition of the samples available for the association study is given in Table 1.

Table 1 Composition of available sample split into controls (CTL), bipolar (BPAD), major depressive disorder (MDD), schizophrenia (SCZ); bipolar spectrum (BPspec, a subdivision of MDD); male and female subjects

Parents–offspring trios were ascertained by contacting parents through the probands. They gave informed consent for genetic studies and a venous blood sample was collected.

Genomic DNA was extracted from venous blood samples using standard protocols.

DNA sequencing

DNA sequencing was performed using ABI prism BigDye Terminator Cycle Sequencing Ready Reaction Kit. Purified PCR products were sequenced directly from 14 control individuals.


The insertion/deletion polymorphism (Δ502–505) was amplified using the following PCR: primers: IndexTerm5′ TTCATTTCAAGCCTGCTTCC 3′, IndexTerm5′ Hex.CTTAGGGTGGCTGGTAGTGG 3′. The fluorescently labelled PCR products were separated on an automated laser fluorescence DNA sequencer ABI 377 (Perkin-Elmer, Welleslay, MA, USA), and analysed using the GENESCAN (version 2.1) fragment-analysis software. The alleles were identified using the GENOTYPER program (version 2.0) (Perkin-Elmer).

The nonsynonymous single-nucleotide polymorphism (SNP), Val606Ile (rs13440581), was genotyped using SNaPshot™ reactions carried out according to the manufacturer's instructions (Applied Biosystems). The PCR products were separated on an ABI PRISM® 377 DNA Sequencer and analysed using GENESCAN (version 2.1) fragment-analysis software. Genotypes of a subset of samples were confirmed using an ABI PRISM® 7900HT Sequence Detection System and TaqMan technology. The SNP rs2072621 was available from ABI as an Assays-on-Demand™ TaqMan SNP genotyping assay. Genotypes were detected on an ABI PRISM® 7900HT Sequence Detection System.

For LD analysis, SNPs were either genotyped using the TaqMan ABI Assays-on-Demand™ system or by DNA sequencing performed using ABI prism BigDye Terminator Cycle Sequencing Ready Reaction Kit.

TaqMan genotyping was performed at the Wellcome Trust Clinical Research Facility, Genetics Core (

Statistical analysis

Observed genotype frequencies in female subjects for all case and control groups were tested for Hardy–Weinberg equilibrium using χ2 tests with 1 df. Allele frequencies from both male and female chromosomes of their own patient group were used to calculate expected genotype frequencies. The χ2 contingency table tests were used to test for allelic association with case–control status and for differences in allele frequencies between cases with maternal or paternal family history. Unconditional logistic regression21 as implemented in UNPHASED22 was used to test for association with two- and three-marker haplotypes (haplotype frequencies were estimated using the EM algorithm). UNPHASED was also used to calculate D and r2 measures of linkage disequilibrium from trio (parents and offspring) genotypes.


Sequencing and association study

Putative functional polymorphisms in GPR50 were identified by sequencing the exons in 14 control individuals. No polymorphisms were detected in the first exon, but three nonsynonymous polymorphisms were detected in exon 2 that encodes transmembrane domains 2–7 and the carboxyl-terminal cytoplasmic tail (see Figure 1a). An in-frame 12 bp insertion/deletion polymorphism was identified beginning at nucleotide 1573, with respect to the Ensembl transcript sequence ENST00000218316, which results in the loss of four amino acids (Thr.Thr.Gly.His) and is designated Δ502–505. The ‘insertion’ allele, which contains Thr.Thr.Gly.His at positions 502–505, was found to be the most frequent form in the control population analysed and therefore this allele was designated GPR50wt. GPR50wt encodes a full-length protein of 617 amino acids as annotated in the Ensembl database, ENSP00000218316. The Δ502–505 variant encodes a truncated protein of 613 amino acids as annotated in the Genbank and NCBI databases (gi:4758467, NM004224). In addition, two nonsynonymous polymorphisms were identified; a Thr532Ala (rs561077) and a Val606Ile (rs13440581). The nucleotide changes resulting in these amino-acid substitutions are detailed in Figure 1b. Analysis of genomic sequences indicated that variants Δ502–505 and Thr532Ala are in complete linkage disequilibrum with each other, but not with Val606Ile. Association analysis was therefore performed on the polymorphisms Δ502–505 and Val606Ile, as well as an intronic SNP described below.

Figure 1

GPR50 gene structure and position of polymorphisms. The GPR50 gene structure is shown and polymorphisms identified by sequencing genomic DNA from 14 control individuals are illustrated (a). The seven transmembrane domains are denoted by roman numerals I-VII. An arrow denotes the approximate position of rs2072621. The nucleotide changes resulting in alterations of the amino-acid sequence for each polymorphism are shown in (b). Numbering of nucleotides is relative to the Ensembl transcript sequence ENST00000218316 and is with reference to the GPR50wt allele that encodes a full-length protein of 617 amino acids (a, b). The numbering of protein residues is relative to the 617 amino-acid GPR50wt as described by Ensembl protein ENSP00000218316.

Association between the GPR50wt/Δ502–505 polymorphism and psychiatric illness was explored in a case–control association study in a population sampled from the South East of Scotland (Table 1). None of the female genotype frequencies in the control or case groups differed significantly from Hardy–Weinberg equilibrium. Significant association was seen between Δ502–505 polymorphism and both MDD and BPAD (see Table 2; P=0.011 and P=0.0070). No association was seen between Δ502–505 and SCZ (P=0.12). The female association was stronger than that for the combined group when the cases were subdivided by gender (MDD, P=0.0064 and BPAD, P=0.00023). No association was seen for either GPR50 variant and any male case group.

Table 2 Single marker association

The difference in allele frequency between male and female controls (43.4 vs 37.6% Table 2) is not significant P=0.11 and, if the allele frequency in BPAD female subjects is compared to the combined allele frequency of male and female controls, the association remains significant (P=0.00089). Genotype relative risks are 3.1 (95% confidence interval 1.7–5.6) for the GPR50Δ502–505 homozygote and 1.5 (95% confidence interval 0.9–2.5) for the heterozygote compared to the GPR50wt homozygote. This implies additive allelic action on the scale of risk, which is consistent with the lack of deviation from Hardy–Weinberg equilibrium in BPAD female subjects. The frequencies of the GPR50Δ502–505 homozygote, heterozygote and GPR50wt homozygote in BPAD female subjects vs controls are 27.3 vs 13.7%, 47.8 vs 47.6% and 25.2 vs 38.6%, implying a low penetrance value for the GPR50Δ502–505 homozygote of 2.0% (penetrance=0.273 × 0.01/0.137), assuming a population frequency for BPAD of 1%.

Linkage disequilibrium map of GPR50

In order to define the haplotype blocks involved in this association, and to exclude any neighbouring genes, further markers were genotyped on a panel of 31 BPAD parent–offspring trios with affected probands. The linkage disequilibrium map of GPR50 was constructed using 10 SNPs spanning 21.5 kb (Table 3). Despite GPR50 being only 5 kb in length, two independently segregating haplotype blocks were observed. The location of the break between these blocks was resolved to 286 bp between markers rs12558086 and rs12557475 in the single intron (D=0.04). The haplotype block containing the Δ502–505 polymorphism spans 4.5 kb covering part of the intron, all of exon 2 and approximately 2 kb of the 3′ region. This suggests that the haplotype block carrying the putative functional polymorphism in BPAD is limited to GPR50, although we cannot formally rule out the possibility that the association results from another gene in linkage disequilibrium with these markers.

Table 3 Pairwise measures of D and r2 calculated from 31 trios with BPAD probands using both transmitted and untransmitted chromosomes for 10 markers spanning the region 19.2 and 2.2 kb up- and downstream from GPR50Δ502–505

Additional marker association results

In order to investigate the association further, we genotyped two additional polymorphisms: Val606Ile (rs13440581), identified by sequencing 14 control samples (see Figure 1) and rs2072621 identified from the public databases23 (Table 2). For both of these polymorphisms, the genotype frequencies in the female groups did not differ significantly from those expected under Hardy–Weinberg equilibrium. Weak association was detected between the exon 2 Val606Ile polymorphism and the female MDD sample (P=0.0096), this SNP showed no significant associations with female BPAD or SCZ. The intronic SNP rs2072621 gave no association with either BPAD or MDD, but showed significant association in the female SCZ sample (P=0.0014). As for the Δ502–505 polymorphism, no significant associations were found between Val606Ile or rs2072621 and disease status in males. Investigation of two- and three-marker haplotypes gave significant association with disease status in female MDD, BPAD and SCZ, although none of these associations were significantly stronger than their initial single marker association (Table 4). No association was seen for any male case group between disease status and haplotype.

Table 4 Two- and three-marker haplotype associations (P-values)

Statistical considerations

If the P-value for declaration of significance (P=0.05) is adjusted for multiple testing, the Bonferroni corrected P-value for declaration of significance accounting for the nine tests in the initial study (three sex categories by three diagnosis categories on a single marker) is P=0.0056. Under this conservative cutoff, association is declared for female BPAD and the Δ502–505 polymorphism, and for female schizophrenics and the intronic SNP rs2072621. Power analysis of the male sample suggests that the lack of association within males is unlikely to be due to the smaller number of chromosomes tested. The sample of 115 BPAD and 288 control male chromosomes for which the insertion/deletion genotypes were obtained has 74 or 100% power to detect an association, assuming a haplotype risk of 1.5 or 3.1, respectively, allele frequency of 0.4 and type I error of 5%. This suggests that the lack of association found in males is unlikely to be due to the smaller number of X-chromosomes tested in males (115 chromosomes in male, 286 chromosomes in female subjects).

Parent of origin/age of onset analysis

Evidence for a female only association led us to investigate parent of origin effects (see Donald et al24). The female BPAD group was divided into those reporting major psychiatric disorders in one or more first- or second-degree relatives on their maternal (N=37) or paternal (N=32) families (Table 5). Allele frequencies between these groups did not differ (P=0.86). This suggests that there is no parent of origin effect in the association between BPAD and the Δ502–505 polymorphism. Age at onset was examined for the female BPAD identifying those with an early age of onset (24 years or less) of affective illness (N=75 BPAD). Association between early age of onset BPAD and Δ502–505 polymorphism was significant (Table 5, P=0.00016). Parent of origin and early age of onset groups were also defined for MDD. No significant associations were found with any of these groups (Table 5).

Table 5 Analysis of female GPR50Δ502–505 association

The number of observations in each class in the analyses of parent of origin and age of onset was quite small, these results should, therefore, be regarded as preliminary and interpreted with caution.

‘Bipolar spectrum’ analysis

Analysis revealed that the MDD association was largely driven by a group (N=15) fulfilling diagnostic criteria for bipolar spectrum disorder (BPspec, P=0.0015), and that the remaining MDD sample (N=232) gave only a weakly significant association (P=0.044; Table 2). When female bipolar spectrum disorder and BPAD cases are combined, the association between affected status and the Δ502–505 polymorphism is further increased (P=0.000026, Tables 2 and 4).


This is the first study investigating a possible link between putative functional polymorphisms in GPR50, an orphan G protein-coupled receptor, and major mental illness. Both the putative functional polymorphisms studied result in alterations of the amino-acid sequence in the carboxyl-terminus cytoplasmic tail. The deletion allele, Δ502–505, which is associated with both BPAD and MDD (global P-values 0.007 and 0.011), results in the deletion of four amino-acid residues, including two threonine residues (see Figure 1). The SNP at nucleotide 1885 is characterised by a single base-pair transition (G → A), resulting in a nonsynonymous Val606Ile amino-acid substitution (rs13440581). A third SNP at position 1663, which is in complete LD with Δ502–505, also results in a nonsynonymous Thr532Ala amino-acid substitution (rs561077) in the carboxyl tail. GPR50 has an unusually long carboxyl tail of >300 amino acids, longer than that for any other known GPCR. The carboxyl tail contains 41 serine and threonine residues and may be involved in receptor coupling to G proteins, tethering the receptor to the plasma membrane, receptor sequestration, desensitisation and downregulation.25, 26 The identification of a consensus site for a serine protease in the carboxyl tail also raises the possibility that it has enzymatic function, which may depend on receptor activation/phosphorylation state.15 Increasing numbers of proteins in addition to the G proteins are now known to interact with GPCRs.27 Many of these GPCR-associated proteins interact with the third intracellular loop or with the C-terminal cytoplasmic tail. A mutation in the distal portion of the β2-adrenergic receptor has been shown to disrupt a PDZ domain-mediated interaction, resulting in reduced receptor recycling and increased targeting to lysosomes.28 Cao et al28 also showed that amino-acid substitution of the serine residue Ser411, a potential site of phosphorylation by GRKs in the intact receptor, effects endocytic sorting to recycling or lysosomal pathways. Recently, Barrett et al29 have reported an association between a polymorphism in the G protein receptor kinase 3 gene (GRK3) and bipolar disorder. GRKs phosphorylate serine and threonine residues in the cytoplasmic tail or third intracellular loop of GPCRs in the presence of high agonist concentrations. Phosphorylation results in rapid termination of signalling through homologous desensitisation.30 Two of the polymorphisms identified, Δ502–505 and Thr532Ala, disrupt threonine residues and may therefore disrupt sites of GRK phosphorylation. The third coding polymorphism, Val606Ile, alters no known protein motifs, but may well alter downstream signalling or protein–protein interactions in the cytoplasmic tail through as yet uncharacterised interactions. There is no clear functional effect of the intronic SNP rs2072621; however, in other genes, intronic SNPs have been shown to alter splicing efficiency (eg Liu et al31).

The clear association with female subjects affected with major mental illness and not with male subjects is intriguing. There is no reported gender bias in the numbers of individuals affected by BPAD. MDD however is twice as common in female as compared to male subjects. A number of other groups have reported sex-specific results with X-linked candidate genes in both BPAD5, 13, 32, 33, 34, 35 and MDD.36 A study of affected sib pairs by Stine et al.37 reported excess allele sharing in affected sister–sister pairs in the Xq26–28 region that was not found in affected brother–brother or brother–sister pairs. This suggests that despite the lack of a sex bias in the numbers of individuals affected by BPAD, the genetic risk factors may differ between genders.

Analysis of the female genotype relative risk in our study suggests an additive effect of each allele on risk. Non-pseudoautosomal Xq28 genes are reported to undergo X-inactivation38 and therefore it is surprising to observe this effect. The additive effect of the GPR50 deletion in female subjects could in theory result from the differences between numbers of cells expressing the deletion form in heterozygotes, when the X-chromosomes are inactivated randomly, compared to expression of the deletion form in all cells in Δ502–505 homozygotes. Until the X-inactivation status of GPR50 is assessed directly, this remains speculative. This hypothesis would not explain the lack of association with BPAD in males carrying a single copy of the Δ502–505 GPR50 gene. However, another gene in this region, BGN, is reported to be expressed in a sex chromosome dosage-dependant manner despite being X-inactivated and having no copy on the Y-chromosome.39 Geerkens et al39 suggest that BGN protein expression may be controlled during transcription or translation by another sex chromosome-linked gene that escapes X-inactivation. In a similar manner, this mechanism could account for the additive effect of the deletion in female subjects observed in this study. Alternatively, studies of X–Y gene pairs suggest that some X-linked genes may be expressed at higher levels in the brains of female compared to male subjects, independently of X-inactivation status, at least in mice.40 It is important, therefore, that the expression levels of GPR50 in male and female subjects are studied at both RNA and protein levels before effects of protein levels on phenotype are discounted.

Other possibilities for the observed sex-specific pattern of association include imprinting and interactions with hormones. Imprinting on the X-chromosome has been shown to influence phenotype in the study of Turner's syndrome patients who inherit an XO karyotype.41 Skuse et al41 compared the levels of social cognition of women with Turner's syndrome who had inherited the maternal X-chromosome, compared to those who had inherited the paternal X-chromosome. Patients inheriting the maternal X-chromosome tended to show more social difficulties, including offensive and disruptive behaviour. Although genetic and epigenetic mechanisms are possible, these remain speculative. A perhaps more obvious possibility is a female-specific hormonal interaction, given the expression of GPR50 in both the hypothalamus and the pituitary.

The endogenous ligand and physiological role for GPR50 still remain to be identified. The localised expression of this receptor in hypothalamus and pituitary could, however, suggest a role for GPR50 in neuroendocrine function and HPA (hypothalamic-pituitary-adrenocortical) axis regulation. There are multiple lines of evidence which suggest that abnormally elevated HPA axis activity and raised cortisol levels are associated with the development of mood disorders, including MDD and BPAD.42, 43, 44 There is also considerable interplay between the HPA and the hypothalamo-pituitary-gonadal axes and certain reproductive hormones, including oestrogen and progesterone, have been implicated in the regulation of mood and the pathogenesis of psychiatric disease disorders (reviewed in Toren et al45, Halbreich et al46). The mechanisms behind these effects are not fully understood but one could speculate that interactions between gonadal hormones and the hypothalamus and pituitary could provide a possible mechanism to explain the female-only association for GPR50. Interestingly, another female-specific association has recently been reported in the literature and this identified a variant of the CREB gene which displayed significant evidence for linkage to familial recurrent early-onset major depression in female but not in male subjects.47, 48 Zubenko et al48 reported an LOD score of 8.19 in a region containing the CREB1 gene, which encodes a cAMP responsive element-binding protein (CREB). The CREB protein is thought to interact with oestrogen receptors to promote the transcription of genes containing oestrogen responsive elements, even in the absence of CREB-binding sites. This mechanism could implicate CREB with sex-specific patterns of gene expression.49

We have shown that GPR50 is a strong candidate gene for female-associated affective disorder, and that there is weaker evidence of an association with schizophrenia also in female subjects. In conclusion, our findings support the hypothesis of linkage between bipolar-related affective disorders and Xq28.


  1. 1

    Kendler KS, Pedersen N, Johnson L, Neale MC, Mathe AA . A pilot Swedish twin study of affective illness, including hospital- and population-ascertained subsamples. Arch Gen Psychiatry 1993; 50: 699–700.

  2. 2

    Jones I, Kent L, Craddock N . The genetics of affective disorders. In: McGuffin P, Owen MJ, Gottesman II (eds) Psychiatric Genetics and Genomics. Oxford university press: Oxford, England, 2002.

  3. 3

    Sullivan PF, Neale MC, Kendler KS . Genetic epidemiology of major depression: review and meta-analysis. Am J Psychiatry 2000; 157: 1552–1562.

  4. 4

    Sullivan PF, Kendler KS, Neale MC . Schizophrenia as a complex trait: evidence from a meta-analysis of twin studies. Arch Gen Psychiatry 2003; 60: 1187–1192.

  5. 5

    Massat I, Souery D, Del-Favero J, Oruc L, Noethen MM, Blackwood D et al. Excess of allele1 for alpha3 subunit GABA receptor gene (GABRA3) in bipolar patients: a multicentric association study. Mol Psychiatry 2002; 7: 201–207.

  6. 6

    Hebebrand J, Hennighausen K . A critical analysis of data presented in eight studies favouring X-linkage of bipolar illness with special emphasis on formal genetic aspects. Hum Genet 1992; 90: 289–293.

  7. 7

    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.

  8. 8

    Berrettini W . Progress and pitfalls: bipolar molecular linkage studies. J Affect Disord 1998; 50: 287–297.

  9. 9

    Baron M . Genetic linkage and bipolar disorder: a cautionary note. J Affect Disord 2001; 67: 267–273.

  10. 10

    Bocchetta A, Piccardi MP, Martinelli V, Quesada G, Del Zompo M . Maternal inheritance of manic depression in hemizygotes for the G6PD-Mediterranean mutation. Indirect evidence for Xq28 transmission in Sardinia. Psychiatr Genet 1999; 9: 63–68.

  11. 11

    Pekkarinen P, Terwilliger J, Bredbacka PE, Lonnqvist 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.

  12. 12

    Ekholm JM, Kieseppa T, Hiekkalinna T, Partonen T, Paunio T, Perola M et al. Evidence of susceptibility loci on 4q32 and 16p12 for bipolar disorder. Hum Mol Genet 2003; 12: 1907–1915.

  13. 13

    Muller DJ, Schulze TG, Jahnes E, Cichon S, Krauss H, Kesper K et al. Association between a polymorphism in the pseudoautosomal X-linked gene SYBL1 and bipolar affective disorder. Am J Med Genet 2002; 114: 74–78.

  14. 14

    Gubitz AK, Reppert SM . Assignment of the melatonin-related receptor to human chromosome X (GPR50) and mouse chromosome X (Gpr50). Genomics 1999; 55: 248–251.

  15. 15

    Reppert SM, Weaver DR, Ebisawa T, Mahle CD, Kolakowski Jr LF . Cloning of a melatonin-related receptor from human pituitary. FEBS Lett 1996; 386: 219–224.

  16. 16

    Conway S, Drew JE, Mowat ES, Barrett P, Delagrange P, Morgan PJ . Chimeric melatonin mt1 and melatonin-related receptors. Identification of domains and residues participating in ligand binding and receptor activation of the melatonin mt1 receptor. J Biol Chem 2000; 275: 20602–20609.

  17. 17

    American Psychiatric Association. 1994. Diagnostic and Statistical Manual of Mental Disorders, (4th Edn). APA: Washington, DC.

  18. 18

    Endicott J, Spitzer RL . A Diagnostic interview: the schedule for affective disorders and schizophrenia. Arch Gen Psych 1978; 35: 837–862.

  19. 19

    Ghaemi SN, Ko JY, Goodwin FK . ‘Cade's disease’ and beyond: misdiagnosis, antidepressant use, and a proposed definition for bipolar spectrum disorder. Can J Psychiatry 2002; 47: 125–134.

  20. 20

    Smith DJ, Harrison N, Muir W, Blackwood DHR . Bipolar spectrum disorders in young adults with recurrent depression: comparison of structured diagnostic criteria with a manic symptoms checklist. J Affect Disord 2004 (in press).

  21. 21

    Zhao JH, Curtis D, Sham PC . Model-free analysis and permutation tests for allelic associations. Hum Hered 2000; 50: 133–139.

  22. 22

    Dudbridge F . Pedigree disequilibrium tests for multilocus haplotypes. Genet Epidemiol 2003; 25: 115–221.

  23. 23

    UCSC Genome Browser.

  24. 24

    Donald JA, Salmon JA, Adams LJ, Littlejohn T, Maher A, Mitchell PB et al. Parental sex effects in bipolar affective disorder pedigrees. Genet Epidemiol 1997; 14: 611–616.

  25. 25

    Ferguson SS . Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev 2001; 53: 1–24, Review.

  26. 26

    Chuang TT, Iacovelli L, Sallese M, De Blasi A . G protein-coupled receptors: heterologous regulation of homologous desensitization and its implications. Trend Pharmacol Sci 1996; 17: 416–421.

  27. 27

    Brady AE, Limbird LE . G protein-coupled receptor interacting proteins: emerging roles in localization and signal transduction. Cell Signal 2002; 14: 297–309.

  28. 28

    Cao TT, Deacon HW, Reczek D, Bretscher A, von Zastrow M . A kinase-regulated PDZ-domain interaction controls endocytic sorting of the beta2-adrenergic receptor. Nature 1999; 401: 286–290.

  29. 29

    Barrett TB, Hauger RL, Kennedy JL, Sadovnick AD, Remick RA, Keck PE et al. Evidence that a single nucleotide polymorphism in the promoter of the G protein receptor kinase 3 gene is associated with bipolar disorder. Mol Psychiatry 2003; 8: 546–557.

  30. 30

    Bunemann M, Hosey MM . G-protein coupled receptor kinases as modulators of G-protein signalling. J Physiol 1999; 517 (Part 1): 5–23.

  31. 31

    Liu HX, Cartegni L, Zhang MQ, Krainer AR . A mechanism for exon skipping caused by nonsense or missense mutations in BRCA1 and other genes. Nat Genet 2001; 27: 55–58.

  32. 32

    Rubinsztein DC, Leggo J, Goodburn S, Walsh C, Jain S, Paykel ES . Genetic association between monoamine oxidase A microsatellite and RFLP alleles and bipolar affective disorder: analysis and meta-analysis. Hum Mol Genet 1996; 5: 779–782.

  33. 33

    Ho LW, Furlong RA, Rubinsztein JS, Walsh C, Paykel ES, Rubinsztein DC . Genetic associations with clinical characteristics in bipolar affective disorder and recurrent unipolar depressive disorder. Am J Med Genet 2000; 96: 36–42.

  34. 34

    Saito T, Parsia S, Papolos DF, Lachman HM . Analysis of the pseudoautosomal X-linked gene SYBL1in bipolar affective disorder: description of a new candidate allele for psychiatric disorders. Am J Med Genet 2000; 96: 317–323.

  35. 35

    Hawi Z, Mynett-Johnson L, Gill M, Murphy V, Straubl RE, Kendler KS et al. Pseudoautosomal gene: possible association with bipolar males but not with schizophrenia. Psychiatr Genet 1999; 9: 129–134, Erratum in: Psychiatr Genet 1999; 9: 209.

  36. 36

    Schulze TG, Muller DJ, Krauss H, Scherk H, Ohlraun S, Syagailo YV et al. Association between a functional polymorphism in the monoamine oxidase A gene promoter and major depressive disorder. Am J Med Genet 2000; 96: 801–803.

  37. 37

    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.

  38. 38

    Carrel L, Cottle AA, Goglin KC, Willard HF . A first-generation X-inactivation profile of the human X chromosome. Proc Natl Acad Sci USA 1999; 96: 14440–14444.

  39. 39

    Geerkens C, Vetter U, Just W, Fedarko NS, Fisher LW, Young MF et al. The X-chromosomal human biglycan gene BGN is subject to X inactivation but is transcribed like an X–Y homologous gene. Hum Genet 1995; 96: 44–52.

  40. 40

    Xu J, Burgoyne PS, Arnold AP . Sex differences in sex chromosome gene expression in mouse brain. Hum Mol Genet 2002; 11: 1409–1419.

  41. 41

    Skuse DH, James RS, Bishop DV, Coppin B, Dalton P, Aamodt-Leeper G et al. Evidence from Turner's syndrome of an imprinted X-linked locus affecting cognitive function. Nature 1997; 387: 705–708.

  42. 42

    Stokes PE . The potential role of excessive cortisol induced by HPA hyperfunction in the pathogenesis of depression. Eur Neuropsychopharmacol 1995; 5: 77–82.

  43. 43

    Rybakowski JK, Twardowska K . The dexamethasone/corticotropin-releasing hormone test in depression in bipolar and unipolar affective illness. J Psychiatr Res 1999; 33: 363–370.

  44. 44

    Pariante CM . Depression, stress and the adrenal axis. J Neuroendocrinol 2003; 15: 811–812.

  45. 45

    Toren P, Dor J, Rehavi M, Weizman A . Hypothalamic-pituitary-ovarian axis and mood. Biol Psychiatry 1996; 40: 1051–1055.

  46. 46

    Halbreich U, Lumley LA . The multiple interactional biological processes that might lead to depression and gender differences in its appearance. J Affect Disord 1993; 29: 159–173.

  47. 47

    Zubenko GS, Hughes III HB, Maher BS, Stiffler JS, Zubenko WN, Marazita ML . Genetic linkage of region containing the CREB1 gene to depressive disorders in women from families with recurrent, early-onset, major depression. Am J Med Genet 2002; 114: 980–987.

  48. 48

    Zubenko GS, Hughes III HB, Stiffler JS, Brechbiel A, Zubenko WN, Maher BS et al. Sequence variations in CREB1 cosegregate with depressive disorders in women. Mol Psychiatry 2003; 8: 611–618.

  49. 49

    Lazennec G, Thomas JA, Katzenellenbogen BS . Involvement of cyclic AMP response element binding protein (CREB) and estrogen receptor phosphorylation in the synergistic activation of the estrogen receptor by estradiol and protein kinase activators. J Steroid Biochem Mol Biol 2001; 77: 193–203.

Download references


We thank Peter Visscher for statistical advice and discussion, and Kirsty Millar for critical reading of this manuscript. This work was supported by Organon Laboratories Ltd, Newhouse, Lanarkshire, UK.

Author information

Correspondence to P A Thomson.

Rights and permissions

Reprints and Permissions

About this article


  • bipolar affective disorder
  • schizophrenia
  • G protein-coupled receptor
  • hypothalamus
  • pituitary
  • Xq28

Further reading