Rare susceptibility variants for bipolar disorder suggest a role for G protein-coupled receptors

Bipolar disorder (BD) is a prevalent mood disorder that tends to cluster in families. Despite high heritability estimates, few genetic susceptibility factors have been identified over decades of genetic research. One possible interpretation for the shortcomings of previous studies to detect causative genes is that BD is caused by highly penetrant rare variants in many genes. We explored this hypothesis by sequencing the exomes of affected individuals from 40 well-characterized multiplex families. We identified rare variants segregating with affected status in many interesting genes, and found an enrichment of deleterious variants in G protein-coupled receptor (GPCR) family genes, which are important drug targets. Furthermore, we showed targeted downstream GPCR dysregulation for some of the variants that may contribute to disease pathology. Particularly interesting was the finding of a rare and functionally relevant nonsense mutation in the corticotropin-releasing hormone receptor 2 (CRHR2) gene that tracked with affected status in one family. By focusing on rare variants in informative families, we identified key biochemical pathways likely implicated in this complex disorder.

Systematic resequencing of X-chromosome synaptic genes in autism spectrum disorder and schizophrenia

Autism spectrum disorder (ASD) and schizophrenia (SCZ) are two common neurodevelopmental syndromes that result from the combined effects of environmental and genetic factors. We set out to test the hypothesis that rare variants in many different genes, including de novo variants, could predispose to these conditions in a fraction of cases. In addition, for both disorders, males are either more significantly or more severely affected than females, which may be explained in part by X-linked genetic factors. Therefore, we directly sequenced 111 X-linked synaptic genes in individuals with ASD (n = 142; 122 males and 20 females) or SCZ (n = 143; 95 males and 48 females). We identified >200 non-synonymous variants, with an excess of rare damaging variants, which suggest the presence of disease-causing mutations. Truncating mutations in genes encoding the calcium-related protein IL1RAPL1 (already described in Piton et al. Hum Mol Genet 2008) and the monoamine degradation enzyme monoamine oxidase B were found in ASD and SCZ, respectively. Moreover, several promising non-synonymous rare variants were identified in genes encoding proteins involved in regulation of neurite outgrowth and other various synaptic functions (MECP2, TM4SF2/TSPAN7, PPP1R3F, PSMD10, MCF2, SLITRK2, GPRASP2, and OPHN1).

Mechanisms of uptake and resistance to troxacitabine, a novel deoxycytidine nucleoside analogue, in human leukemic and solid tumor cell lines

Troxacitabine (Troxatyl; BCH-4556; (-)-2'-deoxy-3'-oxacytidine), a deoxycytidine analogue with an unusual dioxolane structure and nonnatural L-configuration, has potent antitumor activity in animal models and is in clinical trials against human malignancies. The current work was undertaken to identify potential biochemical mechanisms of resistance to troxacitabine and to determine whether there are differences in resistance mechanisms between troxacitabine, gemcitabine, and cytarabine in human leukemic and solid tumor cell lines. The CCRF-CEM leukemia cell line was highly sensitive to the antiproliferative effects of troxacitabine, gemcitabine, and cytarabine with inhibition of proliferation by 50% observed at 160, 20, and 10 nM, respectively, whereas a deoxycytidine kinase (dCK)-deficient variant (CEM/dCK(-)) was resistant to all three drugs. In contrast, a nucleoside transport-deficient variant (CEM/ARAC8C) exhibited high levels of resistance to cytarabine (1150-fold) and gemcitabine (432-fold) but only minimal resistance to troxacitabine (7-fold). Analysis of troxacitabine transportability by the five molecularly characterized human nucleoside transporters [human equilibrative nucleoside transporters 1 and 2, human concentrative nucleoside transporter (hCNT) 1, hCNT2, and hCNT3] revealed that short- and long-term uptake of 10-30 microM [(3)H]troxacitabine was low and unaffected by the presence of either nucleoside transport inhibitors or high concentrations of nonradioactive troxacitabine. These results, which suggested that the major route of cellular uptake of troxacitabine was passive diffusion, demonstrated that deficiencies in nucleoside transport were unlikely to impart resistance to troxacitabine. A troxacitabine-resistant prostate cancer subline (DU145(R); 6300-fold) that exhibited reduced uptake of troxacitabine was cross-resistant to both gemcitabine (350-fold) and cytarabine (300-fold). dCK activity toward deoxycytidine in DU145(R) cell lysates was <20% of that in DU145 cell lysates, and no activity was detected toward troxacitabine. Sequence analysis of cDNAs encoding dCK revealed a mutation of a highly conserved amino acid (Trp(92)-->Leu) in DU145(R) dCK, providing a possible explanation for the reduced phosphorylation of troxacitabine in DU145(R) lysates. Reduced deamination of deoxycytidine was also observed in DU145(R) relative to DU145 cells, and this may have contributed to the overall resistance phenotype. These results, which demonstrated a different resistance profile for troxacitabine, gemcitabine, and cytarabine, suggest that troxacitabine may have an advantage over gemcitabine and cytarabine in human malignancies that lack or have low nucleoside transport activities.

Analysis of 14 CAG repeat-containing genes in schizophrenia

Recently, it has been suggested that trinucleotide repeat-containing genes may be involved in the etiology of schizophrenia. This study was aimed at investigating putative associations between allelic variants or expansions of CAG repeat-containing genes (CAGrCG) and schizophrenia or its variability with respect to responsiveness to conventional neuroleptics. CAG repeat allelic variants of 14 expressed sequences were compared among three groups of subjects: neuroleptic-responder (R; n = 43) and neuroleptic-nonresponder (NR; n = 63) schizophrenic patients, and a control group (C; n = 122). No CAG expansions, in the range of those observed in neurodegenerative diseases, were identified in these 14 expressed sequences. The sizes of CAG repeat for the hGT1 gene were marginally different among the three groups of subjects (Kruskal-Wallis H (2, 456) = 10.48, Bonferroni corrected P = 0.047). Comparisons among the different groups indicated that neuroleptic responders have shorter alleles compared to controls (Mann-Whitney adjusted Z = -3.23, P = 0.0012). NR patients were not different from controls. These preliminary results suggest that the hGT1 gene, or a gene in its vicinity, may be involved in the etiology of schizophrenia or in modifying the disease phenotype with regard to outcome and/or neuroleptic responsiveness. Am. J. Med. Genet. (Neuropsychiatr. Genet.) 88:694-699, 1999.

Instability of the EPM1 minisatellite

Inherited mutations in the cystatin B gene ( CSTB ) are responsible for progressive myoclonus epilepsy type 1 (EPM1; MIM 254800). This autosomal recessive disease is characterized by variable progression to mental retardation, dementia and ataxia. The majority of EPM1 alleles identified to date contain expansions of a dodecamer repeat located upstream of the transcription start site of the CSTB gene. Normal alleles contain two or three copies of the repeat, whereas pathogenic alleles contain >40 repeats. We examined the meiotic stability of pathogenic, expanded EPM1 alleles from 17 EPM1 families by employing a fluorescence-based PCR-based genotyping assay capable of detecting single dodecamer repeat unit differences on an automated DNA sequencer. We followed 74 expanded allele transmissions to 30 affected individuals and 22 carriers. Thirty-five of 74 expanded allele transmissions demonstrated either contraction or expansion of the minisatellite, typically by a single repeat unit. Thus expanded alleles of the EPM1 minisatellite demonstrate a mutation rate of 47%, the highest yet observed for pathogenetic alleles of a human minisatellite.

Lack of association between the hSKCa3 channel gene CAG polymorphism and schizophrenia

Genetic anticipation, a phenomenon characterized by increased severity of symptoms and earlier age at onset of a disease in successive generations, is believed to be present in schizophrenia. In several neurodegenerative diseases showing anticipation, the mutation causing the disease is an expanded trinucleotide repeat. Therefore, genes containing trinucleotide repeats prone to expansion have become a suitable family of candidate genes in schizophrenia. A human calcium-activated potassium channel gene (hSKCa3), possibly mapping to chromosome 22q11-13, a region previously linked to schizophrenia, was recently described. This gene contains two contiguous expressed CAG repeat stretches. Recently, long allelic variants of one of these CAG repeats were found to be overrepresented in schizophrenic patients compared to normal controls. In this study we attempted to replicate this result and to study the relationship between the length of this CAG repeat on the one hand and the severity and age at onset of the disease on the other hand. No association with the disease or correlation with the severity of schizophrenia was identified. In addition, hSKCa3 was mapped to chromosome 1. Our results do not support the involvement of this particular CAG repeat-containing gene in schizophrenia.

A radiation hybrid map of 48 loci including the clouston hidrotic ectodermal dysplasia locus in the pericentromeric region of chromosome 13q

To facilitate the identification of the gene responsible for Clouston hidrotic ectodermal dysplasia (HED), we used a chromosome 13-specific radiation hybrid panel to map 54 loci in the HED candidate region. The marker retention data were analyzed using RHMAP version 3. The 54 markers have an average retention frequency of 31.6% with decreasing retention as a function of distance from the centromere. Two-point analysis identified three linkage groups with a threshold lod score of 4.00; one linkage group consisted of 49 loci including the centromeric marker D13Z1 and the telomeric flanking marker for the HED candidate region D13S143. Assuming a centromeric retention model, multipoint maximum likelihood analysis of these 49 loci except D13Z1 provided a 1000:1 framework map ordering 29 loci with 21 unique map positions and approximately 2000 times more likely than the next order. Loci that could not be ordered with this level of support were positioned within a range of adjacent intervals. This map spans 347 cR9000, has an average resolution of 17.3 cR9000, and includes 3 genes (TUBA2, GJbeta2, and FGF-9), 18 ESTs, 19 polymorphic loci, and 8 single-copy DNA segments. Comparison of our RH map to a YAC contig showed an inconsistency in order involving a reversed interval of 6 loci. Fiber-FISH and FISH on interphase nuclei analyses with PACs isolated from this region supported our order. We also describe the isolation of 8 new chromosome 13q polymorphic (CA)n markers that have an average PIC value of 0.67. These data and mapping reagents will facilitate the isolation of disease genes from this region.

Short GCG expansions in the PABP2 gene cause oculopharyngeal muscular dystrophy

Autosomal dominant oculopharyngeal muscular dystrophy (OPMD) is an adult-onset disease with a world-wide distribution. It usually presents in the sixth decade with progressive swallowing difficulties (dysphagia), eyelid drooping (ptosis) and proximal limb weakness. Unique nuclear filament inclusions in skeletal muscle fibres are its pathological hallmark. We isolated the poly(A) binding protein 2 gene (PABP2) from a 217-kb candidate interval on chromosome 14q11 (B.B. et al., manuscript submitted). A (GCG)6 repeat encoding a polyalanine tract located at the N terminus of the protein was expanded to (GCG)8-13 in the 144 OPMD families screened. More severe phenotypes were observed in compound heterozygotes for the (GCG)9 mutation and a (GCG)7 allele that is found in 2% of the population, whereas homozygosity for the (GCG)7 allele leads to autosomal recessive OPMD. Thus the (GCG)7 allele is an example of a polymorphism which can act either as a modifier of a dominant phenotype or as a recessive mutation. Pathological expansions of the polyalanine tract may cause mutated PABP2 oligomers to accumulate as filament inclusions in nuclei.

Genomic structure of the human GT334 (EHOC-1) gene mapping to 21q22.3

Several inherited diseases have been mapped to the distal tip of human chromosome 21. In our recent efforts to clone candidate genes for some of these disorders, we have assembled a cosmid and BAC contig spanning 770 kb. We have identified expressed sequences from this contig by means of a cDNA hybrid selection scheme. We present here the isolation, cDNA sequence, genomic organization, and polymorphisms analysis of one such expressed sequence, GT334, which had been identified independently and designated EHOC-1. GT334 is split into 23 exons, and spans an estimated 95 kb of genomic DNA. A pseudogene of the histone H2AZ gene has been identified, and maps within the third intron. We have identified an ORF potentially encoding a protein 1259 amino acids in length, longer than that described in the EHOC-1 gene. The GT334 gene was screened for single base pair changes using single-strand conformation polymorphism (SSCP) analysis and we have identified seven sequence variations within this gene. These polymorphisms can be used as markers in the genetic mapping of other diseases localized to this region.

Unstable insertion in the 5' flanking region of the cystatin B gene is the most common mutation in progressive myoclonus epilepsy type 1, EPM1

Progressive myoclonus epilepsy type 1 (EPM1, also known as Unverricht-Lundborg disease) is an autosomal recessive disorder characterized by progressively worsening myoclonic jerks, frequent generalized tonic-clonic seizures, and a slowly progressive decline in cognition. Recently, two mutations in the cystatin B gene (also known as stefin B, STFB) mapping to 21q22.3 have been implicated in the EPM1 phenotype: a G-->C substitution in the last nucleotide of intron 1 that was predicted to cause a splicing defect in one family, and a C-->T substitution that would change an Arg codon (CGA) to a stop codon (TGA) at amino acid position 68, resulting in a truncated cystatin B protein in two other families. A fourth family showed undetectable amounts of STFB mRNA by northern blot analysis in an affected individual. We present haplotype and mutational analyses of our collection of 20 unrelated EPM1 patients and families from different ethnic groups. We identify four different mutations, the most common of which consists of an unstable approximately 600-900 bp insertion which is resistant to PCR amplification. This insertion maps to a 12-bp polymorphic tandem repeat located in the 5' flanking region of the STFB gene, in the region of the promoter. The size of the insertion varies between different EPM1 chromosomes sharing a common haplotype and a common origin, suggesting some level of meiotic instability over the course of many generations. This dynamic mutation, which appears distinct from conventional trinucleotide repeat expansions, may arise via a novel mechanism related to the instability of tandemly repeated sequences.

Isolation and characterization of GT335, a novel human gene conserved in Escherichia coli and mapping to 21q22.3

As part of efforts to identify candidate genes for disorders mapped to 21q22.3, we have constructed a 405-kb cosmid contig encompassing five tightly linked markers mapping to this region. A subset of these cosmids was used to identify cDNA fragments by the method of hybrid selection. We present here the cDNA sequence of one such gene (GT335) mapping to this region. The gene is expressed as a 1.7-kb transcript predominantly in heart and skeletal muscle, potentially displays alternate splicing, and is predicted to encode a protein 268 amino acids in length. GT335 spans an estimated 13 kb of genomic DNA and is split into seven exons. Five of the six introns conform to the GT . . . AG consensus for intronic splice junctions; the sixth contains nonconventional (AT . . . AC) intronic junctions. We screened this gene for single-basepair mutations using single-strand conformation polymorphism and sequence analysis of both cDNA and genomic DNA from a number of unrelated individuals and have identified several sequence variations, two of which cause conservative amino acid substitutions. This gene is well conserved evolutionarily, with homologs identified in zebrafish and Escherichia coli, suggesting that it plays an important role in basic cellular metabolism.

Isolation and genomic structure of a human homolog of the yeast periodic tryptophan protein 2 (PWP2) gene mapping to 21q22.3

As part of efforts to identify candidate genes for disease mapping to the 21q22.3 region, we have assembled a 770-kb cosmid and BAC contig containing eight tightly linked markers. These cosmids and BACs were restriction mapped using eight rare cutting enzymes, with the goal of identifying CpG-rich islands. One such island was identified by the clustering of NotI, EagI, SstII, and BssHII sites, and corresponded to the NotI linking clone LJ104 described previously. A 7.6-kb HindIII fragment containing this CpG-rich island was subcloned and partially sequenced. A homology search using the sequence obtained from either side of the NotI site identified an expressed sequence tag with homology to the yeast periodic tryptophan protein 2 (PWP2). Several cDNAs corresponding to the human PWP2 gene were identified and partially sequenced. Northern blot analysis revealed a 3.3-kb transcript that was well expressed in all tissues tested. A cDNA consensus of 3157 bp was obtained, and an open reading frame potentially encoding 919 amino acid residues was identified. The predicted protein shows 42% identity and 57% similarity at the amino acid level to the yeast PWP2 protein, which is a member of the WD-repeat containing superfamily, and potentially encodes a G-protein beta subunit. The PWP2 gene is split into 21 exons, ranging in size from 53 to 516 bp, and spans an estimated 25 kb. The gene is transcribed in a 21cen-->21qter direction, with its 5' end mapping approximately 195 kb proximal to the 5' end of the phosphofructokinase-liver isoform gene. Four single base-pair polymorphisms were identified using single-stranded conformation polymorphism analysis. Possible functions of the protein based on homology to other members of the WD-repeat-containing family are discussed.

A 405-kb cosmid contig and HindIII restriction map of the progressive myoclonus epilepsy type 1 (EPM1) candidate region in 21q22.3

As a step toward identifying the molecular defect in patients afflicted with progressive myoclonus epilepsy type 1 (EPM1), we have assembled a cosmid contig of the candidate EPM1 region in 21q22.3. The contig constitutes a collection of 87 different cosmids spanning 405 kb based on a derived HindIII restriction map. Potential CpG-rich islands have been identified based on the restriction map generated from eight different rare-cutting enzymes. This contig contains the genetic material required for the isolation of expressed sequences and the identification of the gene defective in EPM1 and possibly other disorders mapping to this region.

A novel transmembrane transporter encoded by the XPCT gene in Xq13.2

To study the regulation and chromosomal basis of X chromosome inactivation, we have physically characterized the region in Xq13.2 known to contain the X inactivation center (XIC), a locus required in cis for inactivation to occur. Here, we report a novel gene isolated by positional cloning in this region. The gene (previously identified as DXS128E) encodes a predicted 67 kDa protein containing twelve hydrophobic transmembrane domains, characteristic of a family of transporter proteins. Presence of an N-terminal PEST domain, consisting mainly of proline/glutamic acid repeats, suggests that the protein may be rapidly or conditionally degraded. We designate this gene XPCT for X-linked PEST-containing transporter. Expression studies suggest that XPCT is subject to X chromosome inactivation, being expressed only from the active X, despite mapping within 600 kb of the XIST gene which is expressed exclusively from the inactive X. Thus, a chromosomal switch in inactivation pattern occurs between these two genes on the X chromosome.

2.6 Mb YAC contig of the human X inactivation center region in Xq13: physical linkage of the RPS4X, PHKA1, XIST and DXS128E genes

X chromosome inactivation is a mechanism of dosage compensation that regulates the expression of mammalian X-linked genes between XY males and XX females. This phenomenon is cis-acting, clonally heritable, and requires the presence of an X inactivation center (XIC). In our attempts to characterize this phenomenon, we have focused on the physical organization of the human XIC localized to Xq13. From previous studies, we had determined that the candidate XIC interval contained two loci (DXS128 and XIST) and was bound by the breakpoints of two structurally abnormal inactivated X chromosomes, a t(X;14) and an idic(Xp). Here we present a refined mapping of the XIC-containing region using the breakpoint of a late replicating rearranged X (rea(X)), and the initial characterization of a set of 40 yeast artificial chromosomes (YACs) derived from the XIC-containing region. These YACs form a 2.6 Mb contig which completely covers the XIC, and physically links the RPS4X, PHKA1, XIST, and DXS128E genes, as well as a laminin receptor pseudogene (LAMRP4). Furthermore, we have determined the relative orientations of these four genes, and have derived a restriction map of the region using the rare cutter enzymes BssHII, EagI, MluI, NruI, SalI, SfiI, SstII (or SacII), and NotI. We have identified at least 9 CpG-rich islands within this region, and have discovered a large (approximately 125 kb) inverted duplication proximal to the XIC based on symmetrical restriction patterns and homologous probes. We estimate the maximum size of the XIC-containing interval to be between 680 kb and 1200 kb, based on the localization of the breakpoints of the rearranged X chromosomes mentioned above. This lays the groundwork for the further characterization of the XIC region and the isolation of other expressed sequences therefrom.

Pulsed-field map of Xq13 in the region of the human X inactivation center

We have used human/mouse hybrid cell lines to derive a pulsed-field map of the Xq13 region of the human X chromosome, in the vicinity of the X inactivation center (XIC). We have mapped nine loci within two separate clusters (I and II). Cluster I contains three loci (DXS227, XIST, and DXS128) linked within 1700 kb. This cluster also includes the breakpoint of a translocated X;14 chromosome used to define the proximal border of the XIC region. Cluster II covers an additional 1800 kb and physically links six loci (DXS56, DXS171, DXS325, DXS347, DXS356, and DXS441) located between the XIC and the genes for Menkes disease (MNK) and PGK1. Maps of cluster I loci derived from active (Xa) or inactive (Xi) X chromosomes differed, presumably due to methylation differences between the Xa and Xi. This map provides a basis for examining the organization of the Xq13.2-q13.3 region, in and around the XIC, and will assist in the further cloning of expressed sequences from this region.

The human XIST gene: analysis of a 17 kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus

X chromosome inactivation in mammalian females results in the cis-limited transcriptional inactivity of most of the genes on one X chromosome. The XIST gene is unique among X-linked genes in being expressed exclusively from the inactive X chromosome. Human XIST cDNAs containing at least eight exons and totaling 17 kb have been isolated and sequenced within the region on the X chromosome known to contain the X inactivation center. The XIST gene includes several tandem repeats, the most 5' of which are evolutionarily conserved. The gene does not contain any significant conserved ORFs and thus does not appear to encode a protein, suggesting that XIST may function as a structural RNA within the nucleus. Consistent with this, fluorescence in situ hybridization experiments demonstrate localization of XIST RNA within the nucleus to a position indistinguishable from the X inactivation-associated Barr body.

Refined localization of human connexin32 gene locus, GJB1, to Xq13.1

Connexins are the peptide subunits of gap junctions that interconnect cells to allow the direct, intercellular transfer of small molecules. Recently, the human connexin32 gene (locus designation GJB1) has been regionally mapped by three other laboratories to Xp11-q13, Xcen-q22, and Xp11-q22. The smallest region of overlap from these studies is Xcen-q13. By using a series of somatic cell hybrid mapping panels and a rat connexin32 cDNA probe, we have localized the human GJB1 locus to a much smaller region in proximal Xq13.1, in interval 8, as described by Lafrenière et al. (8).

Physical mapping of 60 DNA markers in the p21.1----q21.3 region of the human X chromosome

Using a panel of human/rodent somatic cell hybrids and human lymphoblast lines segregating 18 different human X-chromosome rearrangements and deletions, we have assigned 60 DNA markers to the physical map of the X chromosome from Xp21.1 to Xq21.3. Data from Southern blot hybridization and polymerase chain reaction (PCR) amplification assign these markers to 15 primary map intervals. This provides a basis for further long-range cloning and mapping of the pericentromeric region of the X chromosome.

An 18-locus linkage map of the pericentromeric region of the human X chromosome: genetic framework for mapping X-linked disorders

We report a high-resolution genetic linkage map of the region Xp11.4 to Xq13.3, spanning the centromere of the X chromosome and encompassing approximately 30 cM. This 18-locus map is composed of 11 intervals that are spaced on average about 3 cM apart. Markers incorporated into the map together detect 19 distinct polymorphisms and include five genes (TIMP, SYP, AR, CCG1, PGK1), the OATL1 cluster, the hypervariable locus DXS255, the centromeric locus DXZ1, and 10 other anonymous DNA segments. Given that this map spans roughly one-fifth of the length of the X chromosome and includes many loci currently used in both diagnosis and mapping of X-linked disorders, it should be useful for genetic counseling and for guiding efforts to clone disease genes in this region.