Cytogenetic and molecular characterization of A2BP1FOX1 as a candidate gene for autism.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 144B:869 –876 (2007) Cytogenetic and Molecular Characterization of A2BP1/FOX1 as a Candidate Gene for Autism Christa Lese Martin,1* Jacqueline A. Duvall,2 Yesim Ilkin,3 Jason S. Simon,4 M. Gladys Arreaza,4 Kristin Wilkes,2 Ana Alvarez-Retuerto,2 Amy Whichello,5 Cynthia M. Powell,6 Kathleen Rao,6 Edwin Cook,7 and Daniel H. Geschwind2 11 Department of Human Genetics, Emory University, Atlanta, Georgia Departments of Neurology and Human Genetics, and Center for Autism Research, Semel Insitute, UCLA, Los Angeles, California 3 Department of Human Genetics, University of Chicago, Chicago, Illinois 4 Schering-Plough Research Institute, Kenilworth, New Jersey 5 Duke University School of Nursing, Durham, North Carolina 6 Division of Pediatric Genetics and Metabolism, University of North Carolina, Chapel Hill, North Carolina 7 Institute for Juvenile Research, Department of Psychiatry, University of Illinois at Chicago, Chicago, Illinois 2 Cytogenetic imbalances are increasingly being realized as causes of autism. Here, we report a de novo translocation between the short arms of chromosomes 15 and 16 in a female with autism, epilepsy, and global developmental delay. FISH analysis identified a cryptic deletion of approximately 160 kb at the boundary of the first exon and first intron of the 1.7 Mb ataxin-2 binding protein-1 (A2BP1) gene, also called FOX1. Quantitative real time PCR (Q-PCR) analysis verified a deletion of exon 1 in the 50 promoter region of the A2BP1 gene. Reverse transcription PCR (qRT-PCR) showed reduced mRNA expression in the individual’s lymphocytes, demonstrating the functional consequence of the deletion. A2BP1 codes for a brainexpressed RNA binding or splicing factor. Because of emerging evidence in the role of RNA processing and gene regulation in pervasive developmental disorders, we performed further screening of A2BP1 in additional individuals with autism from the Autism Genetics Resource Exchange (AGRE) collection. Twenty-seven SNPs were genotyped across A2BP1 in 206 parent-child trios and two regions showed association at P 0.008 level. No additional deletions or clear mutations were identified in 88 probands by resequencing of all exons and surrounding intronic regions or quantitative PCR (Q-PCR) of exon 1. Although only nominal association was observed, and no obvious causal mutations were identified, these results suggest that A2BP1 may affect susceptibility or cause autism in a subset of patients. Further investigations in a larger sample may provide additional information regarding the involvement of this gene in the autistic phenotype. ß 2007 Wiley-Liss, Inc. This article contains supplementary material, which may be viewed at the American Journal of Medical Genetics website at http://www. interscience.wiley.com/jpages/1552-4841/suppmat/index.html. Grant sponsor: National Institute of Mental Health; Grant numbers: R01-MH64547. *Correspondence to: Christa Lese Martin, Ph.D., Department of Human Genetics, Emory University, Atlanta, Georgia 30322. E-mail: firstname.lastname@example.org Received 6 November 2006; Accepted 23 February 2007 DOI 10.1002/ajmg.b.30530 ß 2007 Wiley-Liss, Inc. KEY WORDS: cytogenetic imbalance; copy number variation; association analysis; splicing factor; RNA binding Please cite this article as follows: Martin CL, Duvall JA, Ilkin Y, Simon JS, Arreaza MG, Wilkes K, AlvarezRetuerto A, Whichello A, Powell CM, Rao K, Cook E, Geschwind DH. 2007. Cytogenetic and Molecular Characterization of A2BP1/FOX1 as a Candidate Gene for Autism. Am J Med Genet Part B 144B: 869–876. INTRODUCTION Autism is a neurodevelopmental disorder characterized by the clinical triad of social deficits, impaired communication and repetitive restrictive behavior patterns. The prevalence of Autism Spectrum Disorders (ASD) has been estimated to be as high as 1/166, with males being affected at least four times more frequently than females [Yeargin-Allsopp et al., 2003; Muhle et al., 2004]. Many different biological causes have been implicated in the etiology of autism, but genetic factors appear to be the most important: family studies have revealed a recurrence risk of 4% among siblings of affected probands and twin studies have shown a significant difference in the concordance rates of autism between monozygotic and dizygotic twins [Veenstra-VanderWeele and Cook, 2004]. The genetic mechanism underlying autism has been studied using various approaches, including linkage, candidate gene, and chromosome analyses, however, there is still a paucity of information regarding causative genetic mechanisms of the autistic phenotype. Multiple lines of evidence support the location of an autism gene in the chromosomal region 16p13. First, several genome-wide linkage scans and association studies have obtained peaks at 16p13 [IMGSAC, 1998; Philippe et al., 1999; IMGSAC, 2001; Lucarelli et al., 2003; Barnby et al., 2005]. One region, located between D16S407 (10 Mb from pter) and D16S3075 (12 Mb from pter) has been implicated by various genome-wide linkage studies with an MLS ranging from 0.74 [Philippe et al., 1999] to 2.19–2.52 [IMGSAC, 2001]. In addition, an association study focusing on the 16p region also identified a marginal association between D16S502 (7.8 Mb from pter) and autism [Lucarelli et al., 2003]. Interestingly, this region on 16p is close to the 1.7 Mb ataxin2 binding protein-1 (A2BP1) gene which binds to the ataxin-2 protein, implicated in Spinocerebellar Ataxia type 2 (SCA2). A2BP1 has been previously characterized and shown to code for an RNA-binding protein which is expressed in muscle and brain. The gene is highly conserved in C. elegans, 870 Martin et al. D. melanogaster, mouse and human. A previous cytogenetic study reported two unrelated individuals with abnormal phenotypes who carry de novo chromosome rearrangements that disrupt A2BP1 [Bhalla et al., 2004]. One individual was reported to have severe intellectual and developmental retardation and had a seizure when 5 days old. The second individual presented with mild mental retardation, a history of seizures and short absences, delayed milestones, behavioral problems, mild dysmorphism, clinodactyly of the fifth finger and mild cutaneous bilateral syndactyly of the second and third toes. Here, we report the detailed clinical, cytogenetic and molecular findings in a female with autism, epilepsy, mental retardation and other clinical features, who carries a translocation between the short arms of chromosomes 15 and 16 that disrupts the A2BP1 gene, affecting its expression level. Further, we present data from quantitative PCR, sequence and SNP association analyses that were utilized to characterize the A2BP1 gene in a large cohort of individuals from the Autism Genetics Resource Exchange (AGRE). CASE REPORT AND METHODS Clinical Report The proband was enrolled in AGRE (sample number AU077504) following approved Institutional Review Board protocols. A lymphoblastoid cell line from the proband and peripheral blood samples from the proband and her parents were utilized for the analyses in this study. Although most families in the AGRE collection are multiplex, the proband is the only affected individual in this AGRE family. The proband was a 3,500 g term infant of an uneventful pregnancy delivered by repeat C-section. At 6 weeks, liver enzymes (AST 119, ALT 173, GGT 650) were elevated and an abdominal ultrasound revealed a gallstone. She underwent a cholecystectomy at 9 months due to failure to thrive and irritability after feedings. Liver enzymes were within normal limits post-operatively. At 17 months, she had myringotomy and tube placement due to a history of chronic otitis media. Elevated liver enzymes (AST 132, ALT 165, GGT 370) were again detected by routine pre-operative laboratory testing. Over the next several months, close monitoring of liver enzymes showed fluctuations of AST, ALT, and GGT with no apparent clinical signs or symptoms. The highest levels were noted at 20 months (AST 1041, ALT 1564, GGT 668). Over the next 2 years she had three liver biopsies that were consistent with mild portal fibrosis and active cholangiolitis of uncertain etiology, with no evidence of chronic active hepatitis. Liver enzymes improved with the long-term administration of ursodeoxycholic acid. At 2 years 8 months she was referred for a genetic evaluation due to liver disease of unknown etiology, developmental delay, hypotonia, uneven gait and mild dysmorphic features, including slight frontal bossing, wide-set eyes, epicanthal folds and a flat nasal bridge. She weighed 25.5 pounds (just below 5th centile), her height was 34 inches (5th centile) and head circumference was 49 cm (50th centile). Routine and highresolution chromosome analyses at this time were reported as normal. Complex partial seizures were first observed at 3 years 3 months and an EEG showed frequent sharp waves and spikes originating in the right frontal and right anterior temporal region. These seizures persisted until age 6. After 4 years of no known clinical seizure activity, the patient was weaned off anti-epileptic medication, however, at 12 years, she began exhibiting frequent rapid eye blinking episodes and had early morning myoclonic seizures; an EEG showed multifocal sharp waves in the bilateral temporal, posterior and frontal head regions. A Magnetic Resonance Imaging (MRI) study was reported to be normal, but upon more detailed review was consistent with mild cerebellar atrophy. At a follow-up genetics evaluation at 10 years 3 months, additional dysmorphic features were noted, including slight upslanting palpebral fissures, slight confluence of eyebrows, full lips and a broad nasal tip. Long, thin fingers, a wide space between the first and second toes and a sacral dimple were also observed. Due to the constellation of clinical features suggestive of a chromosomal imbalance, repeat chromosome analysis was ordered; the findings from these studies are reported in the Results. Developmentally, the proband sat up at 6–7 months, crawled at 10 months, and began walking at 17 months. Her walking was broad based and slightly unstable due to lower extremity hypotonia. Evaluation by a developmental pediatrician demonstrated delays in both receptive and expressive language skills at 2 years 6 months. By 3 years, some hand biting and posturing of arms would occur with excitement or frustration. Distractibility, sensory defensiveness, perseveration, and occasional pinching of other children upon frustration during group therapy were also concerns. At 4 years 6 months, the proband began having drastic mood and behavior changes. Extreme irritability and aggression toward herself and others occurred frequently and she could sustain tantrums/rages for up to 2 or more hours. She also experienced loss of spontaneous language (e.g., no longer called family members by name) and increased sensory defensiveness. An initial diagnosis of autism was made at 5 years 6 months by pediatric psychiatry. At 6 years 2 months, a developmental evaluation reported a Vineland Adaptive Behavior Composite Standard Score of 46 and a MerrillPalmer ratio I.Q. of 30. She had a score of 39.5 on the Childhood Autism Rating Scale (CARS). Autism Diagnostic Interview— Revised (ADI—R) summary scores at age 12 years 2 months were consistent with autism (Social ¼ 29; Communication ¼ 29; Restricted and repetitive behaviors ¼ 10). Developmentally, the proband remains severely impaired and requires constant supervision due to symptoms of autistic disorder and she continues to have frequent treatment refractory myoclonic seizures which generalize up to10 times per month. Cytogenetic and FISH Analyses G-banded chromosome analysis was performed on peripheral blood samples from the proband and her parents following routine procedures. FISH analysis was carried out on metaphase cells from the proband with a probe specific to the 16p subtelomere region [Knight et al., 2000] (D16S3400, Abbott Molecular, Inc., Des Plaines, IL) and with probes for the chromosome 15 alpha satellite (D15Z4) and classical satellite regions (D15Z1). Further FISH characterization was pursued to identify the breakpoint of the rearrangement. BAC clones were identified from the telomeric region of the short arm of chromosome 16 based on publicly available genome resources (http:// genome.ucsc.edu/, www.ensembl.org). As shown in Figure 1, genomic clones were selected at approximately 1 Mb intervals covering the most distal 10 Mb of 16p. When the breakpoint was narrowed to a region of 2 Mb, additional clone coverage was identified for fine mapping. All clones are from the RPCI11 library, unless otherwise specified. All clones were PCRverified using STS primers and FISH was used to verify each clone’s cytogenetic position and unique localization (i.e., no cross-hybridization signals to other chromosomes). DNA was isolated using an automated DNA isolation system (AutoGen 740, Integrated Separation Systems, Natick, MA) and directly-labeled with either Spectrum Orange-dUTP (Abbott Molecular, Inc.), Spectrum Green-dUTP (Vysis, Inc.) A2BP1/FOX1 as a Candidate Gene for Autism 871 Fig. 1. Schematic diagram showing genomic clone coverage for the most distal 10 Mb of 16p with results from FISH mapping in the proband with a 15p;16p translocation that disrupts the A2BP1 gene. or Diethylaminocoumarin-5-dUTP (DEAC/aqua, PerkinElmer Life Sciences, Inc., Boston, MA) using a standard nicktranslation reaction. Slide preparation, probe preparation, hybridization and post-hybridization washing were completed using previously described methods [Chong et al., 1997; Martin et al., 2002]. At least 10 cells were analyzed using direct microscopic visualization and digital-imaging analysis (ViewPoint software, Abbott Molecular Inc.). Chromosome identification was achieved by inverted DAPI staining. Quantitative PCR Analysis Real-time quantitative PCR (Q-PCR) was performed on genomic DNA from the proband with the 15p;16p translocation, her parents, and 88 additional autistic probands from the AGRE collection to examine exon 1 of the A2BP1 gene for heterozygous deletions. Each sample was tested a total of eight times and normalized to the control gene MPZ. Primers utilized are listed in Supplementary Table I. The primer concentration for each gene was 100 nm. The samples were run on an ABI Prism 7700 using the Bio-Rad iTaq SYBR Green Supermix with ROX under the following conditions: an initial 958C denaturation (3 min) was followed by 45 cycles of 958C (15 sec) and 608C (45 sec). A heat dissociation curve was run at the end of the PCR to confirm the presence of a single product. Semi-Quantitative RT-PCR Analysis Total RNA was extracted from lymphoblasts from the proband with the 15p;16p translocation, the proband’s normal family members, and an unrelated normal control individual using the RNeasy1 Mini Kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions. First-strand cDNA was synthesized from purified RNA using the SuperscriptTM RT First-Strand cDNA Synthesis kit following the manufacturer’s protocol (Invitrogen, Carlsbad, CA). PCR primers for isoform 4 and isoforms 1, 2 and 3 were designed based on the human A2BP1 sequence (NCBI accession numbers: NM_145891; NM_145892; NM_145893; see Supplementary Table I for primer sequences). PCR components for each 50 ml reaction were: 2.2 ml First-Strand cDNA, 2.5 U Taq DNA polymerase (Fisher Scientific, Pittsburgh, PA), 5 ml 10 PCR buffer A (Fisher Scientific), 1.1 ml 10 mM A2BP1 forward and reverse primers, 2.2 ml 1 mM 18S rRNA forward and reverse primers, 1.5 ml 300 mM dNTPs. Initial PCR was performed and 8 ml was removed at 24, 26, 28, 30, 32, and 34 cycles to identify the plateau phase of PCR and final PCR was done prior to the plateau. Thermocycling conditions were: 2 min at 948C, followed by 26 thermal cycles of 30 sec at 948C, 30 sec at 598C, 30 sec at 728C and 5 min at 728C. Forty microliters of each PCR reaction were run on a 1.2% agarose gel and stained with 1 SYBR1 Green I nucleic acid gel stain (BMA, Rockland, ME) for 30 min. Stained gels were imaged using a TyphoonTM scanner and analyzed using ImageQuantTM software (Amersham Biosciences, Piscataway, NJ). Each band image volume was individually corrected for background. A2BP1 band volumes were normalized to the 18S rRNA load control band volumes. A two-sample t-test (assuming equal variances) was performed on the normalized A2BP1 volumes using three independent replicates per sample. SNP Genotyping and Association Analysis Twenty-seven SNPs from dbSNP and Celera databases were genotyped by Illumina, Inc. (San Diego, CA; www.illumina. com) in 206 case-parent trios from AGRE. SNPs spanned both exons and introns between exon 5 (isoform 4) to the 30 end of the gene with an average density of 63 kb/SNP (Table I). The Haploview program was used to assess the genotyping success (%geno), Mendelian Errors (ME), Hardy–Weinberg equilibrium (HWE) and minor allele frequencies (MAF) [Barrett et al., 2005]. Transmission Disequilibrium tests (TDT) were performed for individual SNPs as well as all possible groupings of 2, 3, and 4 consecutive SNPs using a sliding window TDT analysis with Genehunter 2.0 [Daly et al., 1998]. DNA Sequence Analysis All 16 exons of the A2BP1 gene, including approximately 200 bases of intronic flanking regions, were sequenced in 88 autistic probands. Supplementary Table II lists the primers utilized for sequence analysis. Following DNA amplification, PCR reactions were diluted to 50 ml in PCR buffer containing 0.5 ml of ExoSAP-IT (USB Corporation, Cleveland, OH) and incubated 15 min at 378C followed by inactivation of the 872 Martin et al. TABLE I. SNPs Covering the A2BP1 Gene NCBI ID rs1395579 rs1911492 rs1395585 rs1507030 rs1507031 rs7501006 rs2063087 rs1003614 rs1003615 rs1034977 rs8060733 rs1034980 rs1019190 rs1848173 rs1507008 rs4787048 rs7204945 rs3785228 rs1024650 rs740677 rs3785214 rs917544 rs763649 rs12149686 rs2191133 rs3785189 rs874584 Position (Build 36) Major allele Minor allele Minor allele Frequency 7527735 7534508 7534959 7551157 7551379 7560547 7564832 7570144 7570219 7572495 7572881 7573264 7574873 7576229 7588603 7604103 7608397 7619581 7646994 7651762 7657122 7663573 7669146 7672064 7684601 7696446 7700167 A A A C A A A A A A A A A C A A A A C A A A A A A A A G G G G G G G G G C T C G G G G G G G G C G C G G G G 0.34 0.09 0.23 0.38 0.48 0.41 0.03 0.41 0.41 0.12 0.15 0.11 0.04 0.17 0.16 0.30 0.29 0.16 0.33 0.42 0.45 0.32 0.50 0.18 0.33 0.24 0.34 enzymes at 808C for 15 min. Cycle sequencing in the forward and reverse directions was performed using ABI PRISM BigDye terminator v3.1 Cycle Sequencing DNA Sequencing Kit (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. Briefly, 1 ml of each PCR product was used as template and combined with 4 ml sequencing reaction mix containing 5 pmol M13 sequencing primer (21M13 or M13Rev), 0.5 Sequencing buffer and 0.25 ml BDTv3.1 mix. Sequencing reactions were denatured for 1 min at 968C followed by 25 cycles at 968C for 10 sec, 508C for 5 sec and 608C for 4 min. Sequencing reactions were purified by filtration using Montage SEQ384 plates (Millipore Corp. TABLE II. Locations and Frequencies of SNPs in A2BP1 in Autistic Individuals SNP namea SNP4020 (G662A) SNP3bpDel298040 SNP4027 (C298121G) SNP4028 (G635524A) SNP4029 (A635528T) SNP4030 (G635530C) SNP4031 (C635605A) SNP4032 (T635606A) 1bpDel31195 SNP4100 (T31272C) 4 bpDel311820 SNP4022 (A311931G) SNP4024 (G497405A) SNP4025 (C497570T) SNP4102 (C558948T) SNP4103 (C559111T) rs1507010 (A586531G) rs4616299 (A586590G) rs321705(2bp632944) SNP4088 (C632950G) SNP4089 (C633048T) SNP4090 (G688277A) SNP4091 (G688294A) SNP4092 (G688364T) SNP3570 (G690344T) a Chr. 16 locationb Gene location Bases from nearest exonc Minor allele frequencyd 6069906 6367284 6367365 6704768 6704772 6704774 6704849 6704850 7101836 7101913 7382461 7382572 7568046 7568211 7629589 7629752 7657172 7657231 7703585 7703591 7703689 7758918 7758935 7759005 7760985 Exon 1 (50 UTR) Intron 1 Intron 2 Intron 2 Intron 2 Intron 2 Intron 2 Intron 2 Intron 3 Intron 4 Intron 4 Exon 4b (50 UTR) Exon 5 Intron 5 Exon 6 Intron 6 Intron 10 Intron 10 Intron 11 Intron 11 Exon 12 Exon 15 Intron 15 Intron 15 Exon 16 (30 UTR) 482 (UTR) 4 þ12 130 126 124 49 48 20 þ15 85 22 (UTR) 100 þ21 12 þ31 þ33 þ92 31 25 74 62 þ3 þ73 562 (UTR) 0.60% (1/166) 0.56% (1/180) 0.56% (1/178) 31.88%(51/160) 17.07%(28/164) 31.71%(52/164) 23.26%(40/172) 25.00%(43/172) 38.10% (64/168) 0.59% (1/170) 19.32%(34/176) 0.81% (1/124) 0.57% (1/176) 9.20% (16/174) 0.57% (1/174) 1.69% (2/118) 48.85% (85/174) 33.91% (59/174) 43.60%(75/172) 0.58% (1/172) 0.57% (1/174) 1.14% (2/176) 5.68% (10/176) 25.00% (44/176) 1.28% (2/156) Bold ¼ reported in NCBI dbSNP Build 120; Italics ¼ indel. Chr. 16 location from UCSC Genome Browser, July 2003. Bold ¼ in coding sequence; ¼ upstream of 1st base of exon; þ ¼ downstream of last base of exon. d Number of chromosomes carrying the minor allele out of total number of chromosomes is shown in parentheses; bold ¼ >10%. b c A2BP1/FOX1 as a Candidate Gene for Autism 873 Bedford, MA), dissolved in 25 ml deionized water and resolved by capillary electrophoresis on an Applied Biosystems 3730XL DNA Analyzer. Chromatograms were transferred to a Unix workstation (DEC alpha, Compaq Corp), base called with Phred (version 0.990722.g), assembled with Phrap (version 3.01) and scanned with Polyphred (version 3.5) [Nickerson et al., 1997]. The final results were viewed with Consed (version 9.0) (Phred, Phrap, and Consed available at http:// www.genome.washington.edu, PolyPhred is available at http://droog.mbt.washington.edu). Analysis parameters were all maintained at the individual software’s default settings. RESULTS Cytogenetic and FISH Characterization of the t(15;16) G-banding analysis revealed a possible abnormality involving the short arms of chromosomes 15 and 16. FISH analysis using a probe for the 16p telomere region verified an apparently balanced translocation between the short arms of chromosomes 15 and 16, with one hybridization signal on the normal 16p and one signal on 15p. FISH analyses with a chromosome 15 alpha-satellite probe and a probe from the classical satellite region on 15p both showed signals only on the normal 15 and derivative 15; there was no reciprocal signal on 16p. Parental cytogenetic analysis was normal, indicating the rearrangement occurred de novo in the proband. Microsatellite analysis of the proband and her parents demonstrated a paternal origin of the deletion. The karyotype was therefore designated: 46,XX,?t(15;16)(p11.2;p13.3)dn.ish t(15;16) (D16S3400þ,D15Z1þ,D15Z4þ;D16S3400,D15Z1,D15Z4). The short arms of acrocentric chromosomes contain multiple copies of the ribosomal RNA genes, which can be lost without phenotypic consequences. Therefore, the breakpoint on chromosome 16 was targeted with additional FISH studies to determine if the rearrangement disrupted a gene that could be involved in causing the proband’s phenotype. As shown in Figure 1, BAC clones for FISH analysis were selected at approximately 1 Mb intervals for the most distal 10 Mb of the short arm of chromosome 16. The clone located approximately 5 Mb from the 16p telomere (RP11-35P16) was translocated to 15p, but the clone at 7 Mb (RP11-545E8) remained on 16p. Interestingly, the 6 Mb clone (RP11-509E10) showed an apparent partial deletion, with the hybridization signal on the derivative 16 chromosome showing a reduced intensity compared to the signal on the normal chromosome 16. Also shown in Figure 1, is additional clone coverage that was selected in this region for fine mapping, including clones distal to and proximal to the 6 Mb clone. Figure 2 shows representative images from FISH analysis using these clones. RP11-19H6 was shown to be translocated to 15p while RP11-167B4 remained on 16p. Only a single hybridization signal was observed on the normal chromosome 16 for RP11-578P21, demonstrating a deletion at the translocation breakpoint. Parental FISH analysis using clone RP11-578P21 was normal, ruling out an inherited deletion polymorphism of this clone and no copy number polymorphisms for this region have been reported. The gene content at the translocation breakpoint and deleted region was examined and the only known gene contained in this interval is the ataxin-2 binding protein (A2BP1) gene. From FISH analysis, the deletion was estimated to include exon 1 of the A2BP1 gene. Further confirmation of this result was pursued using molecular methods. Q-PCR Analysis for Deletion Detection of Exon 1 in A2BP1 To precisely determine if exon 1 of the A2BP1 gene is deleted in the proband, real-time Q-PCR using genomic DNA was Fig. 2. Representative FISH analysis showing results from breakpoint mapping of the 15p;16p translocation. The color coding used in the schematic above the FISH images is the same as that used in Figure 1. BAC 19H6 is distal to the translocation breakpoint and shows a hybridization signal (red) on the short arm of the derivative 15 and the normal 16p. BAC 578P21 is deleted and only shows a single hybridization signal (red) on the normal 16; the derivative 15 and normal 15 are marked by a centromere probe for chromosome 15 (aqua). performed in the proband, her parents, an unaffected brother, and an unrelated control individual. As shown in Figure 3, the genomic dose of A2BP1 in the proband is approximately 50% of that in the four other individuals tested, demonstrating a heterozygous deletion of exon 1 in the proband. Deletions of exon 1 in A2BP1 were also tested for by Q-PCR in an additional 88 individuals (that also had complete gene sequencing). No evidence for a deletion in exon 1 was observed in any of the individuals tested, consistent with the presumed rarity of this deletion. A2BP1 Expression Analysis of Isoforms 1–4 The A2BP1 gene has 4 known isoforms and the total gene is among the longest in the genome, 1.7 Mb (spanning from 6.0–7.7 Mb on 16p). Isoform 4 spans the entire 1.7 Mb, but isoforms 1, 2, and 3 cover only 380 kb at the 30 end of the gene. The first 3 exons and part of the 4th exon of isoform 4 consist of the untranslated region of the gene. Interestingly, these 4 exons cover 1 Mb total genomic sequence. Since the deletion in the proband occurs in an untranslated region that is more than a megabase away from the coding sequence, but was predicted to potentially contain regulatory sequences, we hypothesized that the deletion might alter A2BP1 levels. To assess whether the deletion had such functional consequences, expression of A2BP1 was examined by semi-quantitative RTPCR in lymphoblasts. As shown in Figure 4a, qRT-PCR for isoform 4 showed the expression was significantly decreased in the proband compared to controls (P < 0.0001). The same experiment was performed using primers specific for isoforms 1, 2, and 3. Expression of isoforms 1, 2, and 3 (collectively) was also significantly decreased, although to a lesser extent than isoform 4 (Fig. 4b). These results demonstrate that the deletion has a significant effect on A2BP1 expression. SNP Analysis and Sequencing Rare chromosomal abnormalities, such as the deletion described here, in people with disease symptoms may indicate that the interrupted gene contributes to the disease. Motivated by the identification of this rare disease-causing variant, we reasoned that common variants in this gene may contribute to autism. To test this in a preliminary manner we conducted a pilot association analysis using AGRE autism families. 874 Martin et al. Fig. 3. Results from Q-PCR analysis of the A2BP1 gene in the proband, the proband’s unaffected family members and an unrelated normal control showing reduced dosage for A2BP1 in the proband. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.] Twenty-seven SNPs, in both introns and exons of the A2BP1 gene (Fig. 5), from dbSNP and the Celera database (Table I) were genotyped in 206 parent child trios from AGRE. All SNPs had a %geno 99%, ME < 2, MAF 0.03, and HWE P 0.03. No single SNP showed significant association when tested for transmission disequilibrium (TD). However, nominally significant multipoint TD was observed between SNPs rs1395579, rs1911492, and rs1395585, which span a 7 kb region, and between SNPs rs917544, rs763649, rs12149686, and rs2191133, spanning 21 kb (TDT P ¼ 0.004 and P ¼ 0.008, respectively; Fig. 5). Linkage disequilibrium was assessed using the Haploview program. SNPs within each of the two regions did not appear to be in strong LD, with r2 < 0.54 among SNPs in each of the regions. Re-sequencing of A2BP1 was also carried out to identify potential mutations in this gene within the AGRE sample. A2BP1 was sequenced for all 16 exons and surrounding intronic regions in one proband chosen from 88 affected sibspairs who demonstrated the most allele sharing across the region between markers D16S2622 and D16S407, which flank the A2BP1 gene. No non-synonymous amino acid changes were identified in this cohort. Twenty-five single nucleotide variants were identified by sequencing the 88 autistic individuals (Table II). Seven of the 25 SNPs were located in exons. However, none caused an amino acid change and were not predicted to be deleterious. DISCUSSION Various genome-wide genetic approaches, including linkage scans, association studies, cytogenetic and FISH analyses have been pursued to identify chromosomal regions harboring Fig. 4. qRT-PCR results demonstrating reduced expression of (A) isoform 4 and (B) isoforms 1, 2, and 3 (collectively) of the A2BP1 gene in the proband compared to normal family members and a normal unrelated control. causative or susceptibility genes for autism. Using these strategies, several studies have implicated the chromosomal region 16p13 [IMGSAC, 1998; Philippe et al., 1999; IMGSAC, 2001; Lucarelli et al., 2003; Bhalla et al., 2004; Barnby et al., 2005; Vorstman et al., 2006]. We identified a translocation involving the short arms of chromosomes 15 and 16 in a female with autism, severe developmental disability, epilepsy, delayed walking with mild residual ataxia, behavioral regression, fluctuating liver function tests and mild cerebellar atrophy. Using a combination of FISH and Q-PCR analyses, we demonstrated that a deletion of approximately 160 kb is present in exon 1 of the A2BP1 gene at the breakpoint of this chromosomal rearrangement. Furthermore, rtPCR showed significantly reduced expression, consistent with the supposition that the A2BP1 deletion in this individual is functional. A2BP1 is one of the largest genes in the human genome, covering 1.7 Mb and encoding a 377 amino acid protein. It is highly conserved in C. elegans, D. melanogaster, mouse and human [Kiehl et al., 2001] and has been previously characterized in human [Shibata et al., 2000], C. elegans [Kiehl et al., 2000] and mouse [Kiehl et al., 2001]. In humans, A2BP1 codes for an RNA-binding protein that binds to the C-terminus of ataxin-2, which has been implicated in spinocerebellar ataxia type 2 (SCA2) [Shibata et al., 2000]. Recently, the A2BP1 protein was shown to regulate alternative splicing of tissuespecific exons by binding to the hexanucleotide UGCAUG through its RNA recognition motif and is considered a neuronspecific splicing factor [Nakahata and Kawamoto, 2005; Underwood et al., 2005; Auweter et al., 2006]. The A2BP1 protein is predominantly expressed in muscle and brain. Specifically in the brain, A2BP1 is found in the cytoplasm of Purkinje cells and dentate neurons in a punctate pattern [Shibata et al., 2000; Underwood et al., 2005]. Postmortem studies of brains from individuals with autism have shown neuroanatomic abnormalities of the cerebellum and limbic system, including the hippocampus [Bauman and Kemper, 2005]. Therefore, A2BP1 is a plausible candidate gene for ASD given its expression pattern. Fine mapping studies of the proband with the 15p;16p translocation revealed a deletion involving exon 1, an untranslated region of the A2BP1 gene. However, qRT-PCR analysis of lymphoblasts showed reduced expression of A2BP1 for all 4 isoforms compared to other normal family members and an unrelated normal control. Therefore, the chromosomal region 1.5 Mb up from the coding sequence is likely involved in regulating gene expression, as its deletion affects the expression of all known A2BP1 isoforms in this individual. Given the causative deletion of A2BP1 and because of emerging evidence in the role of RNA processing and gene regulation in pervasive developmental disorders, we performed further screening of this gene in additional individuals with autism from the AGRE collection to identify other rare A2BP1/FOX1 as a Candidate Gene for Autism 875 Fig. 5. A2BP1 gene structure and SNP association results. The A2BP1 gene is shown schematically at the top, spanning from 6,009 kb on chr16 to 7,702 kb based on NCBI Genome Build 36. The coding region, which was the focus of our association analysis is depicted by the brackets. Exons are represented by short vertical lines. The region of interest is magnified below with relative locations of genotyped SNPs. The LD plot generated in Haploview shows r2 values between genes with darker color indicating higher LD. The two haplotype blocks that showed significant association are boxed in red and SNPs in those blocks are shown under the magnified gene structure. variants. We did not detect any additional deletions of exon 1 or mutations in the 88 probands screened. However, we cannot rule out whole gene deletions or deletions of other parts of the gene since we only examined exon 1 of A2BP1 by Q-PCR. Association analysis of SNPs across A2BP1 identified two potential regions harboring risk alleles. Although only nominal association was observed, which is not significant if one applies a strict Bonferroni correction, these results remain suggestive, since the SNPs are not totally independent, and a Bonferroni correction may be overly strict. Similarly, while no obvious causal mutations were identified in this study, these results do not rule out the possibility that A2BP1 may affect susceptibility or cause autism in a subset of patients. Multiple lines of evidence, including linkage studies and cytogenetic deletions support A2BP1 as a good candidate gene for involvement in ASD. Therefore, further investigations of the potential risk variants identified here in a larger sample of multiplex families or in sporadic autism cases, both with and without phenotypic features similar to the case reported, may provide additional information regarding the involvement of this gene in the autistic phenotype. ACKNOWLEDGMENTS This work was supported by NIMH grant R01-MH64547, A Genome Wide Search for Autism Susceptibility Loci (to CLM and DHG). We gratefully acknowledge the resources provided by the Autism Genetic Resource Exchange (AGRE) Consortium1 and the participating AGRE families. AGRE is a 1 Dan Geschwind, M.D., Ph.D., UCLA, Los Angeles, CA; Maja Bucan, Ph.D., Univ. of Pennsylvania, Philadelphia, PA; W.Ted Brown, M.D., Ph.D., F.A.C.M.G., N.Y.S. Inst. for Basic Research in Developmental Disabilities, Staten Island, NY; Rita M. Cantor, Ph.D., UCLA, Los Angeles, CA; John N. Constantino, M.D., Washington Univ., St. Louis, MO; T.Conrad Gilliam, Ph.D., Univ. of Chicago, Chicago, IL; Martha Herbert, M.D., Ph.D., Harvard Medical School, Boston, MA; Clara Lajonchere, Ph.D, Cure Autism Now, Los Angeles, CA; David H. Ledbetter, Ph.D., Emory Univ., Atlanta, GA; Christa Lese Martin, Ph.D., Emory Univ., Atlanta, GA; Janet Miller, J.D., Ph.D., Cure Autism Now, Los Angeles, CA; Stanley F. Nelson, M.D., UCLA, Los Angeles, CA; Gerard D. Schellenberg, Ph.D., Univ. of Washington, Seattle, WA; Carol A. Samango-Sprouse, Ed.D., George Washington Univ., Washington, D.C.; Sarah Spence, M.D., Ph.D., UCLA, Los Angeles, CA; Matthew State, M.D., Ph.D., Yale Univ., New Haven, CT; Rudolph E. Tanzi, Ph.D., Massachusetts General Hospital, Boston, MA. program of Cure Autism Now and is supported, in part, by grant MH64547 from the National Institute of Mental Health to Daniel H. Geschwind (PI). The goal of the program is to facilitate more rapid progress in the identification of the genetic causes of autism and autism spectrum disorders by promoting sharing and collaboration. A full description of this resource has been published (Geschwind et al. (2001) Am J Hum Genet, 69:2) and is available on the web at www.agre.org. The authors would also like to thank Susan Christian, PhD, for microsatellite analysis to determine the parental origin of the deletion, and William Dobyns, M.D., for review of the MRI scans. REFERENCES Auweter SD, Fasan R, Reymond L, Underwood JG, Black DL, Pitsch S, Allain FH. 2006. Molecular basis of RNA recognition by the human alternative splicing factor Fox-1. EMBO J 25(1):163–173. Barnby G, Abbott A, Sykes N, Morris A, Weeks DE, Mott R, Lamb J, Bailey AJ, Monaco AP. 2005. Candidate-gene screening and association analysis at the autism-susceptibility locus on chromosome 16p: Evidence of association at GRIN2A and ABAT. Am J Hum Genet 76(6):950–966. Barrett JC, Fry B, Maller J, Daly MJ. 2005. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 21(2):263– 265. Bauman ML, Kemper TL. 2005. Neuroanatomic observations of the brain in autism: A review and future directions. Int J Dev Neurosci 23(2–3):183– 187. Bhalla K, Phillips HA, Crawford J, McKenzie OL, Mulley JC, Eyre H, Gardner AE, Kremmidiotis G, Callen DF. 2004. The de novo chromosome 16 translocations of two patients with abnormal phenotypes (mental retardation and epilepsy) disrupt the A2BP1 gene. J Hum Genet 49(6): 308–311. Chong SS, Pack SD, Roschke AV, Tanigami A, Carrozzo R, Smith AC, Dobyns WB, Ledbetter DH. 1997. A revision of the lissencephaly and Miller-Dieker syndrome critical regions in chromosome 17p13.3. Hum Mol Genet 6(2):147–155. Daly MJ, Pratt SC, Houstis N, Reeve-Daly MP, Kirby A, Lander ES. 1998. GENEHUNTER 2.0-a complete linkage analysis system. Am J Hum Genet Suppl 63:A286. IMGSAC. 1998. A full genome screen for autism with evidence for linkage to a region on chromosome 7q. International Molecular Genetic Study of Autism Consortium. Hum Mol Genet 7(3):571–578. IMGSAC. 2001. A genomewide screen for autism: Strong evidence for linkage to chromosomes 2q, 7q, and 16p. Am J Hum Genet 69(3):570– 581. Kiehl TR, Shibata H, Pulst SM. 2000. The ortholog of human ataxin-2 is essential for early embryonic patterning in C. elegans. J Mol Neurosci 15(3):231–241. 876 Martin et al. Kiehl TR, Shibata H, Vo T, Huynh DP, Pulst SM. 2001. Identification and expression of a mouse ortholog of A2 BP1. Mamm Genome 12(8):595– 601. Knight SJ, Lese CM, Precht KS, Kuc J, Ning Y, Lucas S, Regan R, Brenan M, Nicod A, Lawrie NM, et al. 2000. An optimized set of human telomere clones for studying telomere integrity and architecture. Am J Hum Genet 67(2):320–332. Lucarelli P, Palminiello S, Saccucci P, Bottini N, De Luca D, Elia M, Fiumara A, Curatolo P. 2003. Association study of autistic disorder and chromosome 16p. Am J Med Genet Part A 119A(2):242–246. Martin CL, Waggoner DJ, Wong A, Uhrig S, Roseberry JA, Hedrick JF, Pack SD, Russell K, Zackai E, Dobyns WB, et al. 2002. ‘‘Molecular rulers’’ for calibrating phenotypic effects of telomere imbalance. J Med Genet 39(10):734–740. Muhle R, Trentacoste SV, Rapin I. 2004. The genetics of autism. Pediatrics 113(5):e472–e486. Nakahata S, Kawamoto S. 2005. Tissue-dependent isoforms of mammalian Fox-1 homologs are associated with tissue-specific splicing activities. Nucleic Acids Res 33(7):2078–2089. Nickerson DA, Tobe VO, Taylor SL. 1997. PolyPhred: Automating the detection and genotyping of single nucleotide substitutions using fluorescence-based resequencing. Nucleic Acids Res 25(14):2745– 2751. Philippe A, Martinez M, Guilloud-Bataille M, Gillberg C, Rastam M, Sponheim E, Coleman M, Zappella M, Aschauer H, Van Maldergem L, et al. 1999. Genome-wide scan for autism susceptibility genes. Paris Autism Research International Sibpair Study. Hum Mol Genet 8(5):805–812. Shibata H, Huynh DP, Pulst SM. 2000. A novel protein with RNA-binding motifs interacts with ataxin-2. Hum Mol Genet 9(9):1303–1313. Underwood JG, Boutz PL, Dougherty JD, Stoilov P, Black DL. 2005. Homologues of the Caenorhabditis elegans Fox-1 protein are neuronal splicing regulators in mammals. Mol Cell Biol 25(22):10005–10016. Veenstra-VanderWeele J, Cook EH Jr. 2004. Molecular genetics of autism spectrum disorder. Mol Psychiatry 9(9):819–832. Vorstman JA, Staal WG, van Daalen E, van Engeland H, Hochstenbach PF, Franke L. 2006. Identification of novel autism candidate regions through analysis of reported cytogenetic abnormalities associated with autism. Mol Psychiatry 111:18–28. Yeargin-Allsopp M, Rice C, Karapurkar T, Doernberg N, Boyle C, Murphy C. 2003. Prevalence of autism in a US metropolitan area. JAMA 289(1): 49–55.