Flanking Exon Research Articles

Molecular mechanisms for aberrant expression of the human breast cancer specific gene 1 in breast cancer cells: control of transcription by DNA methylation and intronic sequences.

Breast cancer specific gene 1 (BCSG1), also referred as synuclein gamma, is the third member of a neuronal protein family synuclein. BCSG1 is not expressed in normal breast tissues but highly expressed in advanced infiltrating breast carcinomas. When over expressed, BCSG1 significantly stimulates breast cancer metastasis. To elucidate the molecular mechanisms underlying the abnormal transcription of BCSG1 in breast cancer cells, in this study, we isolated a 2195 base pair fragment of human BCSG1 gene. This fragment includes 1 kb 5'-flanking region, exon 1, and intron 1. By analysing the promoter activity and the methylation status of the exon 1 region, we show that (1) Intron 1 plays critical roles in the control of BCSG1 gene transcription through cis-regulatory sequences that affect BCSG1 transcription in cell type-specific and cell type-nonspecific manners. (2) The activator protein-1 (AP-1) is functionally involved in BCSG1 transcription in breast cancer cells through its binding to an AP-1 motif located in the intron 1. (3) The exon 1 region of BCSG1 gene contains a CpG island that is unmethylated in BCSG1-positive SKBR-3 and T47D cells but densely methylated in BCSG1-negative MCF-7 cells. (4) Treating MCF-7 cells with a demethylating agent 5-Aza-2'-deoxycytidine specifically activated BCSG1 transcription. Thus, our results suggest that while the cellular content of transcription activators and repressors that interact with the cis-regulatory sequences present in the intron 1 contribute prominently to the tissue-specific expression of BCSG1, demethylation of exon 1 is an important factor responsible for the aberrant expression of BCSG1 in breast carcinomas.

Lack of an association between cystatin C gene polymorphisms in Japanese patients with Alzheimer's disease.

Associations between polymorphisms of the cystatin C gene (CST3) at 5' flanking region and exon 1 in Caucasian patients with late onset AD and exon 1 in a US study of late onset AD have been reported. Clinically diagnosed Japanese patients with AD and Japanese normal control subjects were assessed for the presence of polymorphisms of CST3. The authors could not confirm the previously reported association between CST3 polymorphisms and AD in Japan. Age had no effect on the CST3 genotype.

TESTING THE REVERSE‐SPLICING MODEL OF RDNA INTRON ORIGIN

Bhattacharya, D., Simon, D., Chakravarty, S. & Huang, J.Department of Biological Sciences and Interdepartmental Genetics Program, University of Iowa, Iowa City, IA 52242 USAGroup I introns are widely distributed in the nuclear ribosomal RNA genes of algae and fungi. Two models are proposed for intron mobility. The first relies on an endonuclease open reading frame that allows intron “homing” at the DNA‐level into homologous allelic sites. The second model proposes mobility at the RNA‐level through reverse‐splicing. Whereas endonuclease‐mediated homing is depen‐dent on a significant sequence requirement (14‐35 nt) at the insertion site, reverse‐splicing requires as little as 4‐6 nt at the 5′ flanking sequence to initiate intron integration. For this reason, reverse‐splicing may play the more important role in intron lateral transfer in the nuclear genome. In this presentation, we test two major predictions of the reverse‐splicing model with introns in the small (SSU) and large subunit (LSU) nuclear rRNA: 1) Group I introns reverse‐splice into rRNA regions which contain a 4‐6 nt 5′‐exon flanking sequence that builds a helix with the intron internal guide sequence required for forward‐ and reverse‐splicing. 2) Introns are non‐randomly distributed, with most of them clustering in regions that are not “hidden” by rRNA tertiary structure. Phylogenetic analyses show that multiple group I introns in the SSU rDNA of lichen fungi have likely originated through reverse‐splicing of one intron into a heterologous site. The reverse‐splicing model is also supported by the observation that both group I and spliceosomal introns are primarily found in highly conserved rRNA regions that map on the surface of the SSU ribosome. These introns show a positive correlation with rRNA sites that form cross‐links with tRNA and mRNA.

Structural characterization of the mouse Foxf1a gene

Members in the superfamily of the forkhead/winged-helix transcription factors are known to play a critical role in the control of cell differentiation and tissue development. To understand the regulation and function of these genes, we have initially isolated and characterized the mouse Foxf1a gene, a novel forkhead gene predominantly expressed in the lung. The mouse gene consists of two exons with the forkhead domain contained in exon 1, and is located at band E1 on chromosome 8. Amino acid sequence of the mouse protein shares a high degree of homology to that of the corresponding human protein. The tissue specificity of expression of the mouse gene also resembles that found in the human gene. This gene is primarily expressed in the lung, and to a lesser extent in placenta and tissues in gastrointestinal tract. The transcription start site was mapped to 113 nucleotides upstream from the putative translation initiation site. The promoter of the mouse gene is highly GC rich and contains neither a CAAT nor a TATA box. A series of luciferase report constructs driven by the promoter and various deletions in the 5′ flanking region of the gene were constructed and employed in transient transfection studies using a line of SV40 transformed mouse lymph node endothelial cells (SVEC4–10), which express the endogenous Foxf1a gene, and a line of mouse hepatoma cells (Hepa 1–6), in which Foxf1a is not expressed. To our surprise, these reporter genes are equally active in both cell lines. Further studies have shown that the proximal 5′ flanking sequence and exon 1 of the endogenous gene are highly methylated in Hepa 1–6 cells but not in SVEC4–10 cells, suggesting that DNA methylation but not cell-specific transcription factor(s) regulates cell specificity of gene expression in these cultured cells.

The Dynamics of Myogenin Site-specific Demethylation Is Strongly Correlated with Its Expression and with Muscle Differentiation

The molecular mechanisms underlying the activation of tissue-specific genes have not yet been fully clarified. We analyzed the methylation status of specific CCGG sites in the 5'-flanking region and exon 1 of myogenin gene, a very important myogenic differentiation factor. We demonstrated a loss of methylation, at the onset of C2C12 muscle cell line differentiation, limited to the CCGG site of myogenin 5'-flanking region, which was strongly correlated with the transcriptional activation of this gene and with myogenic differentiation. The same CCGG site was also found to be hypomethylated, in vivo, in embryonic mouse muscle (a myogenin-expressing tissue), as opposed to nonmuscle (nonexpressing) tissues that had a fully methylated site. In a C2C12-derived clone with enhanced myogenic ability, demethylation occurred within 2 h of induction of differentiation, suggesting the involvement of some active demethylation mechanism(s) that occur in the absence of DNA replication. Exposure to drugs that inhibit DNA methylation by acting on the S-adenosylmethionine metabolism produced a further reduction, to a few minutes, in the duration of the demethylation dynamics. These effects suggest that the final site-specific DNA methylation pattern of tissue-specific genes is defined through a continuous, relatively fast interplay between active DNA demethylation and re-methylation mechanisms.

Correlations of partial and extensive methylation at the p14(ARF) locus with reduced mRNA expression in colorectal cancer cell lines and clinicopathological features in primary tumors.

p14(ARF) is a putative tumor suppressor gene thought to modify the levels of p53. CpG sites within the 5'-flanking region and exon 1beta of p14(ARF) are targets of aberrant methylation and transcriptional silencing in human colorectal cancer (CRC). Here we have developed methylation-specific polymerase chain reaction (MSPCR) methods to detect methylation of CpG sites in p14(ARF) in CRC cell lines and primary CRC tumors, and correlated p14(ARF) mRNA expression with methylation in CRC cell lines using competitive quantitative reverse transcription-polymerase chain reaction methods. Ten CRC cell lines were studied; three (DLD-1, HCT15 and SW48) showed extensive methylation and six (Colo320, SW480, HT29, Caco2, SW837 and WiDr) were unmethylated; the other cell line, LoVo, showed partial methylation that affected exon 1beta but not the immediate upstream CpG sites. p14(ARF) mRNA was expressed at extremely low levels in fully methylated cell lines and at 10(4)- to 10(5)-fold higher levels in unmethylated cell lines. p14(ARF) expression in the partially methylated LoVo cell line was intermediate. Treatment of LoVo cells with 2 microM 5-aza-2'-deoxycytidine for 72 h was associated with marked (100-fold) induction of mRNA levels. Of 119 primary CRCs, 18% contained p14(ARF) methylation, although partial methylation was the most common pattern observed (in 67% of methylated tumors). Methylation of p14(ARF) was often accompanied by p16(INK4a) methylation; however, 50% of p14(ARF) methylated tumors contained unmethylated p16(INK4a). Methylation at p14(ARF) was associated with female gender, greater age, proximal anatomic location and poor differentiation, but not stage at diagnosis. A two-step MSPCR method for assaying p14(ARF) methylation in human tumors is described.

Genetic association of a cystatin C gene polymorphism with late-onset Alzheimer disease.

To determine whether the cystatin C gene (CST3) is genetically associated with late-onset Alzheimer disease (AD). A case-control study with 2 independent study populations of patients with AD and age-matched, cognitively normal control subjects. The Alzheimer's Disease Research Unit at the University Hospital Hamburg-Eppendorf, Hamburg, Germany, for the initial study (n = 260). For the independent multicenter study (n = 647), an international consortium that included the Massachusetts Alzheimer's Disease Research Center at the Massachusetts General Hospital, Boston; the Scientific Institute for Research and Patient Care, Brescia, Italy; and Alzheimer's research units at the Universities of Basel and Zurich, Switzerland, and Bonn, Goettingen, and Hamburg, Germany. Five hundred seventeen patients with AD and 390 control subjects. Molecular testing of the KspI polymorphisms in the 5' flanking region and exon 1 of CST3 and the apolipoprotein E (APOE) genotype. Mini-Mental State Examination scores for both patients with AD and control subjects. Homozygosity for haplotype B of CST3 was significantly associated with late-onset AD in both study populations, with an odds ratio of 3.8 (95% confidence interval, 1.56-9.25) in the combined data set; heterozygosity was not associated with an increased risk. The odds ratios for CST3 B/B increased from 2.6 in those younger than 75 years to 8.8 for those aged 75 years and older. The association of CST3 B/B with AD was independent of APOE epsilon4; both genotypes independently reduced disease-free survival. CST3 is a susceptibility gene for late-onset AD, especially in patients aged 75 years and older. To our knowledge, CST3 B is the first autosomal recessive risk allele in late-onset AD.

Molecular cloning and analysis of the partial sequence ofRhinopithecus roxellanae growth hormone gene

Growth hormone gene (GH) ofRhinopithecus roxellanae was amplified by PCR based on the sequences of the reported mammalian growth hormone gene for the first time. The amplified fragment was about 1.8 kb. It was cloned and its upper stream was sequenced. This sequencing region consists of a 5′flanking regulatory region, exon I and part of exon II, intron I of growth hormone gene. Comparing the corresponding sequences of growth hormone gene betweenRhinopithecus roxellanae and the porcine, we concluded that the homology reached 81% in the region, and there was high conservation in the 5′flanking sequence. The kinds of amino acids of exon I and exon II for about 90% were the same to those in pig. Many mutations occurred in the degenerate site of the triplet code. In the nucleotides of intron I, there were only 72% homologies with those in pig. It means that introns and 3′flanking sequence maybe play an important part in growth hormone gene regulation of the different animals.

Molecular cloning, expression and physical mapping of the human methionine synthase reductase gene.

Methionine synthase reductase (EC 2.1.1.135) is a flavoprotein essential for maintenance of methionine synthase in an active state. We characterized the human gene for methionine synthase reductase (MTRR). The gene is approximately 34kb and comprises 15 exons, varying in size from 43 to 1213bp, and 14 introns whose sizes vary from 108bp to 5kb. The positions of several junctions are conserved between the MTRR gene and the C. elegans ortholog, as well as with the rat cytochrome P450 reductase gene. A 1.3kb CpG island encompasses the 5'-flanking region and exon 1 and extends into intron 1. A short region including the transcription start site is sufficient to confer promoter activity, with a better outcome when accompanied by intron 1. The promoter region contains putative binding sites for Sp1, AP-1, AP-2 as well as CAAT motifs, but no consensus TATA box. Primer extension analysis revealed a single major transcription start site, located 137bp upstream of the previously reported initiator ATG. An alternative splicing event involving a portion of exon 1 predicts that translation can potentially be initiated at two different ATG codons. The gene was physically assigned to a narrow area between markers WI1755 and D5S1957.

Factor IX Gene Analysis In 70 Unrelated Patients with Haemophilia B: Description of 13 New Mutations

Seventy unrelated patients suffering from haemophilia B have been screened for determining the molecular defect and for evaluating the spectrum of factor IX mutations in the Rhône Alpes region in France. Most patients were characterized with respect to factor IX antigen and factor IX coagulant activity. We have used denaturing gradient gel electrophoresis to obtain a full scanning of the whole coding, promoter, and exon flanking sequences of the factor IX gene. This technique enabled us to determine the molecular defect in 68 out of 70 families (97%), and the mutation was further identified in the two last patients with a direct sequencing of the gene. A total of 2 complete gene deletions in patients with antifactor IX inhibitor, 6 small insertions/deletions and 62 point mutations were found. Two of these nucleotide substitutions (Arg145His and Ala233Thr) were detected in 21 patients (30%) suggesting the existence of a local founder effect. Thirteen mutations were previously undescribed, including 7 missense mutations. The detection of mutations in patients affected with haemophilia B may shed some light in the structure-function relationship of factor IX molecule within the coagulation system.

The P-selectin gene is highly polymorphic: reduced frequency of the Pro715 allele carriers in patients with myocardial infarction.

P-selectin is an adhesion molecule, expressed at the surface of activated cells, that mediates the interaction of activated endothelial cells or platelets with leukocytes. P-selectin expression is increased in atherosclerotic plaques, and high plasma levels of this molecule have been observed in patients with unstable angina. We investigated the P-selectin gene as a possible candidate for myocardial infarction (MI). The P-selectin gene is situated on chromosome 1q21-q24, spans >50 kb and contains 17 exons. The sequences of the 5'-flanking region and exons of 40 alleles from patients with MI were screened for polymorphisms using polymerase chain reaction/single-strand conformation polymorphism (PCR-SSCP) and sequencing. Thirteen polymorphisms were identified: five in the 5'-flanking and eight in the exonic sequences. Four polymorphisms (Ser290Asn, Asn562Asp, Leu599Val and Thr715Pro) predicted a change in the amino acid sequence of the P-selectin protein. All P-selectin polymorphisms as well as a common E-selectin polymorphism, Ser128Arg which has been reported as being associated with an increased risk of premature coronary heart disease (CHD), and is in tight linkage disequilibrium with several P-selectin polymorphisms, were investigated in 647 patients with MI and 758 control subjects from four regions of France and Northern Ireland (the ECTIM study). The entire set of P-selectin polymorphisms provided a heterozygosity of 91%. The polymorphisms were tightly associated with one another and displayed patterns of linkage disequilibrium suggesting the existence of highly conserved ancestral haplotypes. The five polymorphisms in the 5'-flanking region of the gene were unrelated to MI or any relevant phenotype measured in the ECTIM study. We inferred that the four missense variants identified in the coding region predicted eight common forms of the P-selectin protein. The Pro715 allele which characterizes one of these forms was less frequent in France than in Northern Ireland ( P < 0.002) and in cases than in controls ( P < 0.002; P < 0.02 after correction for the number of tests). We conclude that the P-selectin gene is highly polymorphic and hypothesize that the Pro715 variant may be protective for MI. Whether this variant affects the properties of the P-selectin protein in a way which is compatible with this hypothesis needs to be checked experimentally.

Molecular Mechanism of Human CD38 Gene Expression by Retinoic Acid: IDENTIFICATION OF RETINOIC ACID RESPONSE ELEMENT IN THE FIRST INTRON

CD38 is a nonlineage-restricted type II transmembrane glycoprotein possessing ecto-NAD+ glycohydrolase activity. Because of its unique expression pattern in lymphocyte differentiation, it appears to function as an immunoregulatory molecule. We previously reported that CD38 was specifically induced by all-trans-retinoic acid (RA) in human promyelocytic leukemia HL-60 cells. Here we studied the molecular mechanism of the RA-dependent induction of human CD38. The expression of CD38 mRNA by RA appeared to be caused by the transcriptional stimulation of the gene, since it was blocked by an RNA synthesis inhibitor, but not by a protein synthesis inhibitor. In search of the RA response element (RARE) possibly present in human CD38 gene promoter, we isolated and sequenced the genomic DNA covering the 5'-flanking region, exon 1, and partial intron 1. Transient transfection experiments revealed that the responsiveness to RA was conferred through an RARE consisting of two direct repeat TGACCT-like hexamer motifs with a 5-nucleotide spacer, which was located in the first intron rather than the 5'-flanking region of the CD38 gene. This RARE interacted with heterodimer composed of RA receptor and retinoid X receptor in vitro. Thus, the RA-induced expression of the human CD38 gene was demonstrated to be mediated through the RARE located in the first intron.

U-richness is a defining feature of plant introns and may function as an intron recognition signal in maize.

Using a large set of plant gene sequences we compared individual introns to their flanking exons. Both Zea mays and Arabidopsis thaliana introns are U-rich but display no apparent bias for A. We identified fifteen 11-mer U-rich motifs as frequent elements of maize introns, and these are virtually absent from exons. By mutagenesis, we show that the single U-rich motif in the Bronze2 intron of maize plays a key role in intron processing in vivo.

Novel acceptor splice site mutation in the invariant AG of intron 6 of alpha-galactosidase A gene, causing Fabry disease. Mutations in brief no. 146. Online.

We found a novel acceptor splice site mutation in the invariant AG of intron 6 of alpha-galactosidase A (alpha-Gal A) gene (IVS6-1G->A) in a patient with Fabry disease by sequencing of genomic DNA. Sequencing of RT-PCR revealed the deletion of first base pair (c909del) of exon 7 in mRNA and a frameshift resulting in premature termination. This mutation gives rise to a rare aberrant splicing (Simultaneous 3' destruction and 3' creation).

A new informative Alw 26 I polymorphism in exon 10 of the human low density lipoprotein receptor gene. Application to prenatal diagnosis

Using DNA sequencing of the coding and exon flanking regions of the low density lipoprotein receptor (LDLR) gene we identified an Alw26 I site in exon 10 by a transition G1426A. The alleles are represented by one uncut fragment (A1 = 108 bp) or two fragments (A2 = 82 bp and 26 pb). Two other fragments (72 bp and 16 bp) were systematically found within the amplified product. The alleles were detected in 157 unrelated French Caucasians with A1 frequency = 0.58 and A2 = 0.42. The observed heterozygoty was 44.5%. Homozygous familial hypercholesterolemie (FH) has a severe clinical picture leading to death during childhood. Because it is very informative, the present polymorphism was very useful as genetic marker for clinical diagnosis and counseling as we described in linkage analysis at the LDLR locus for prenatal diagnosis in a fetus who could inherit two LDLR mutant alleles from FH heterozygote parents. Hum Mutat 11:483, 1998. © 1998 Wiley-Liss, Inc.

Intronic and exonic sequences modulate 5' splice site selection in plant nuclei.

Pre-mRNA transcripts in a variety of organisms, including plants, Drosophila and Caenorhabditis elegans, contain introns which are significantly richer in adenosine and uridine residues than their flanking exons. Previous analyses using exonic and intronic replacements between two nonequivalent 5'splice sites in the 469 nt long rbcS3A intron 1 provided the first evidence indicating that, in both tobacco and Drosophila nuclei, 5'splice site selection is strongly influenced by the position of that site relative to the AU transition point between exon and intron. To differentiate between two potential models for 5'splice site recognition, we have expressed a completely different set of intronic and exonic replacement constructs containing identical 5'splice sites upstream of beta-conglycinin intron 4 (115 nt). Mutagenesis and deletion of the upstream 5'splice site demonstrate that intronic AU-rich sequences function by promoting recognition of the most upstream 5'splice site rather than by masking the downstream 5'splice site. Sequence insertions define a role for AG-rich exonic sequences in plant pre-mRNA splicing by demonstrating that an AG-rich element is capable of promoting downstream 5'splice site recognition. We conclude that AU-rich intronic sequences, AG-rich exonic sequences and the 5'splice site itself collectively define 5'intron boundaries in dicot nuclei.

Molecular cloning, upstream sequence and promoter studies of the human gene for the regulatory subunit RII α of cAMP-dependent protein kinase

The gene for the regulatory subunit RII α of cAMP-dependent protein kinase is highly regulated during spermatogenesis and a strong signal from a distinct short mRNA form is observed postmeiotically during spermatid elongation. This report presents the isolation and characterization of the 5′-flanking region (1.2 kb) and exon 1 of the human RII α gene. S1 nuclease mapping and primer extension experiments revealed the presence of a major transcriptional start site located 208 nucleotides upstream of start for translation. The 5′-flanking region of the RII α gene did not contain a TATA box and was highly G/C-rich. A basal promoter directing high levels of chloramphenicol acetyl transferase (CAT) activity was identified in the 5′-flanking sequence. Several potential binding sites for transcription factors were identified in this region, which may be responsible for the germ cell-specific regulation of this gene. We have previously reported that the human testis RII α cDNA contains a region (amino acids 45-75) with little or no homology to the corresponding rat skeletal muscle cDNA (Øyen, O., Myklebust, F., Scott, J.D., Cadd, G.G., McKnight, G.S., Hansson, V. and Jahnsen, T. (1990) Biol. Reprod. 43, 46–54). We examined whether this difference could arise due to organ-specific splice mechanisms or represented a species difference. We show that the low homology region of the human RII α cDNA resides entirely within exon 1, and does not originate from a tissue-specific alternate splicing of this distinct region.

Characterization of the bovine prion protein gene: the expression requires interaction between the promoter and intron.

We cloned the part of the bovine PrP gene which contains the 5'-flanking region, exon 1, exon 2 and intron 1 to analyze its promoter region. The 5' non-coding region of the bovine PrP gene consisted of three exons and two introns, and its organization was similar to that of the mouse, rat and sheep PrP genes. The 5'-flanking region of the bovine PrP gene from the transcription start site to nucleotide position -88 was (G + C)-rich (78%) and contained three potential binding sites for the transcription factor Sp1, but no CCAAT-box or TATA-box. This region showed high homology (89%) with that of the sheep PrP gene, but relatively low homology (approximately 46-62%) with the same region of the mouse, rat, hamster and human PrP genes. The position from -88 to -30 within the 5'-flanking region of the bovine PrP gene showed major promoter activity. However, this region was able to function properly only in collaboration with the region at +123 to +891 of intron 1 of the bovine PrP gene.

Two new recurrent nucleotide mutations in theCOL1A1 gene in four patients with osteogenesis imperfecta: About one-fifth are recurrent

Previous observations on mutations causing osteogenesis imperfecta (OI) suggested that unrelated patients had private mutations. Here preliminary studies on two patients with type I OI indicated that some mutations in the COL1A1 gene for type I procollagen cannot be detected by analyses of cDNAs. Therefore, we developed a protocol whereby 43 exon and exon flanking sequences of the COL1A1 gene can be amplified by PCR and scanned for mutations by denaturing gradient gel electrophoresis. Two new recurrent nucleotide mutations in the gene were found in four apparently unrelated patients with OI. Analysis of previous publications indicated that up to one-fifth of the mutations causing OI are recurrent in the sense that they were identical in apparently unrelated probands. About 80% of these identical mutations were in CpG dinucleotide sequences.

The IL-9 Receptor Gene (IL9R): Genomic Structure, Chromosomal Localization in the Pseudoautosomal Region of the Long Arm of the Sex Chromosomes, and Identification of IL9R Pseudogenes at 9qter, 10pter, 16pter, and 18pter

Cosmids containing the human IL-9 receptor (R) gene (IL9R) have been isolated from a genomic library using the IL9R cDNA as a probe. We have shown that the human IL9R gene is composed of 11 exons and 10 introns, stretching over ∼17 kb, and is located within the pseudoautosomal region of the Xq and Yq chromosome, in the vicinity of the telomere. Analysis of the 5′ flanking region revealed multiple transcription initiation sites as well as potential binding motifs for AP1, AP2, AP3, Sp1, and NF-kB, although this region lacks a TATA box. Using the human IL9R cosmid as a probe to perform fluorescencein situhybridization, additional signals were identified in the subtelomeric regions of chromosomes 9q, 10p, 16p, and 18p. IL9R homologs located on chromosomes 9 and 18 were partially characterized, while those located on chromosomes 16 and 10 were completely sequenced. Although they are similar to the IL9R gene (∼90% identity), none of these copies encodes a functional receptor: none of them contains sequences homologous to the 5′ flanking region or exon 1 of the IL9R gene, and the remaining ORFs have been inactivated by various point mutations and deletions. Taken together, our results indicate that the IL9R gene is located at Xq28 and Yq12, in the long arm pseudoautosomal region, and that four IL9R pseudogenes are located on 9q34, 10p15, 16p13.3, and 18p11.3, probably dispersed as the result of translocations during evolution.