Long-range sequence analysis in Xq28: thirteen known and six candidate genes in 219.4 kb of high GC DNA between the RCP/GCP and G6PD loci

A 3-year-old Tanzanian Female with Glucose-6-Phosphate Dehydrogenase A- and a Novel Heterozygous PIEZO1 Mutation (2744A>G, N915S) Presenting with Severe Hemolytic Anemia

A 3-year-old Tanzanian female presented with severe hemolytic anemia of unknown etiology, necessitating multiple red blood cell transfusions. The patient was found to have glucose-6-phosphate dehydrogenase (G6PD) deficiency A- and a heterozygous Piezo-type mechanosensitive ion channel component 1 (PIEZO1) mutation (2744A>G, N915S). This case identifies a novel PIEZO1 mutation implicated in erythrocyte channelopathies occurring in conjunction with an X-linked enzymopathy in a female patient. This underscores the importance of keeping X-linked disorders in the differential diagnosis of hemolytic anemia in females, as well as presents the possibility for novel coexisting mutations to augment the phenotype.

Radiogenomic association of deep MR imaging features with genomic profiles and clinical characteristics in breast cancer

BACKGROUND

It has been believed that traditional handcrafted radiomic features extracted from magnetic resonance imaging (MRI) of tumors are normally shallow and low-ordered. Recent advancement in deep learning technology shows that the high-order deep radiomic features extracted automatically from tumor images can capture tumor heterogeneity in a more efficient way. We hypothesize that MRI-based deep radiomic phenotypes have significant associations with molecular profiles of breast cancer tumors. We aim to identify deep radiomic features (DRFs) from MRI, evaluate their significance in predicting breast cancer (BC) clinical characteristics and explore their associations with multi-level genomic factors.

METHODS

A denoising autoencoder was built to retrospectively extract 4,096 DRFs from 110 BC patients' MRI. Visualization and clustering were applied to these DRFs. Linear Mixed Effect models were used to test their associations with multi-level genomic features (GFs) (risk genes, gene signatures, and biological pathway activities) extracted from the same patients' mRNA expression profile. A Least Absolute Shrinkage and Selection Operator model was used to identify the most predictive DRFs for each clinical characteristic (tumor size (T), lymph node metastasis (N), estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status).

RESULTS

Thirty-six conventional radiomic features (CRFs) for 87 of the 110 BC patients provided by a previous study were used for comparison. More than 1,000 DRFs were associated with the risk genes, gene signatures, and biological pathways activities (adjusted P-value < 0.05). DRFs produced better performance in predicting T, N, ER, PR, and HER2 status (AUC > 0.9) using DRFs. These DRFs showed significant powers of stratifying patients, linking to relevant biological and clinical characteristics. As a contrast, only eight risk genes were associated with CRFs. The RFs performed worse in predicting clinical characteristics than DRFs.

CONCLUSIONS

The deep learning-based auto MRI features perform better in predicting BC clinical characteristics, which are more significantly associated with GFs than traditional semi-auto MRI features. Our radiogenomic approach for identifying MRI-based imaging signatures may pave potential pathways for the discovery of genetic mechanisms regulating specific tumor phenotypes and may enable a more rapid innovation of novel imaging modalities, hence accelerating their translation to personalized medicine.

Glucose-6-Phosphate Dehydrogenase Deficiency

G6PD is a housekeeping gene expressed in all cells. Glucose-6-phosphate dehydrogenase (G6PD) is part of the pentose phosphate pathway, and its main physiologic role is to provide NADPH. G6PD deficiency, one of the commonest inherited enzyme abnormalities in humans, arises through one of many possible mutations, most of which reduce the stability of the enzyme and its level as red cells age. G6PD-deficient persons are mostly asymptomatic, but they can develop severe jaundice during the neonatal period and acute hemolytic anemia when they ingest fava beans or when they are exposed to certain infections or drugs. G6PD deficiency is a global health issue.

Drosomycin-like defensin, a human homologue of Drosophila melanogaster drosomycin with antifungal activity

Innate antifungal defense in Drosophila melanogaster relies on the activation of the Toll molecule and the release of drosomycin, a defensin-like molecule with antifungal properties. Ten human homologues of Toll have been described, with central roles in activation of the innate host defense. In the present study, we report a putative human homologue of the Drosophila-derived drosomycin, designated drosomycin-like defensin (DLD). Synthetic DLD displays a broad spectrum of activity against Aspergillus spp. and other clinically relevant filamentous fungi. These effects are specific for filamentous fungi; no activity has been found against yeasts or gram-positive or gram-negative bacteria. Synthetic DLD also displays immunomodulatory effects on Aspergillus-stimulated cytokine production. In addition, we show the expression of DLD mRNA in several human tissues, particularly in the skin, consistent with its putative role as a defensin against invading microorganisms. This is the first indication of an endogenous human peptide with specific antifungal activity, which is probably central in the defense against infections with molds.

The systematic functional characterisation of Xq28 genes prioritises candidate disease genes

Background

Well known for its gene density and the large number of mapped diseases, the human sub-chromosomal region Xq28 has long been a focus of genome research. Over 40 of approximately 300 X-linked diseases map to this region, and systematic mapping, transcript identification, and mutation analysis has led to the identification of causative genes for 26 of these diseases, leaving another 17 diseases mapped to Xq28, where the causative gene is still unknown. To expedite disease gene identification, we have initiated the functional characterisation of all known Xq28 genes.

Results

By using a systematic approach, we describe the Xq28 genes by RNA in situ hybridisation and Northern blotting of the mouse orthologs, as well as subcellular localisation and data mining of the human genes. We have developed a relational web-accessible database with comprehensive query options integrating all experimental data. Using this database, we matched gene expression patterns with affected tissues for 16 of the 17 remaining Xq28 linked diseases, where the causative gene is unknown.

Conclusion

By using this systematic approach, we have prioritised genes in linkage regions of Xq28-mapped diseases to an amenable number for mutational screens. Our database can be queried by any researcher performing highly specified searches including diseases not listed in OMIM or diseases that might be linked to Xq28 in the future.

Mapping candidate hotspots of meiotic recombination in segments of human DNA cloned in the yeast Saccharomyces cerevisiae

The hotspots of meiotic recombination in the human genome can be localized by genetic techniques. The resolution of these techniques is in the range of kilobases and depends on the density of the physical markers identifying allelic variants of the chromosomal loci. We thought it would be interesting to localize these sites with higher resolution. Assuming that some human chromosomal sites conserve their propensity for recombination when cloned in yeast, we localized the hotspots of recombination in several yeast artificial chromosomes (YACs) carrying human DNA. A number of potential recombination hotspots could be identified in the clones studied. Among them there are two classes of sites that are particularly recombination prone also in human meiotic cells: sites associated with CpG islands and sites located in the vicinity of long minisatellite sequences.

Identification of a 650 kb duplication at the X chromosome breakpoint in a patient with 46,X,t(X;8)(q28;q12) and non-syndromic mental retardation

A female patient with non-syndromic mental retardation was shown by high resolution GTL banding to have inherited an apparently balanced translocation, 46,X,t(X;8)(q28;q12)mat. Replication studies in the mother and daughter showed a skewed X inactivation pattern in lymphocytes, with the normal X chromosome preferentially inactivated. The mother also had significant intellectual disability. To investigate the possibility that a novel candidate gene for XLMR was disrupted at the X chromosome translocation breakpoint, we mapped the breakpoint using fluorescence in situ hybridisation (FISH). This showed that the four known genes involved in non-syndromic mental retardation in Xq28, FMR2, SLC6A8, MECP2, and GDI1, were not involved in the translocation. Intriguingly, we found that the X chromosome breakpoint in the daughter could not be defined by a single breakpoint spanning genomic clone and further analysis showed a 650 kb submicroscopic duplication between DXS7067 and DXS7060 on either side of the X chromosome translocation breakpoint. This duplicated region contains 11 characterised genes, of which nine are expressed in brain. Duplication of one or several of the genes within the 650 kb interval is likely to be responsible for the mental retardation phenotype seen in our patient. Xq28 appears to be an unstable region of the human genome and genomic rearrangements are recognised as major causes of two single gene defects, haemophilia A and incontinentia pigmenti, which map within Xq28. This patient therefore provides further evidence for the instability of this genomic region.

Structural gene organization and evolutionary aspects of the V-ATPase accessory subunit Ac45

The vacuolar H+-ATPase (V-ATPase) is a multisubunit enzyme that couples ATP hydrolysis to proton pumping across membranes. The intracellular targeting and activity of the V-ATPase may be regulated via proteins that interact with the pump such as the accessory subunit Ac45. Here we report the isolation and characterization of the gene encoding Ac45. This single-copy gene is located in a gene-dense region of chromosome Xq and consists of 10 exons spanning approximately 8 kb in the mouse and human genomes. The gene structure is poorly conserved in that its invertebrate orthologs of Caenorhabditis elegans and Drosophila melanogaster encompass only six and four exons extending over 4.1 and 2.1 kb, respectively. Furthermore, the overall degree of amino acid sequence identity between the mammalian and invertebrate Ac45 proteins is very low (<18%), except for a surprisingly highly conserved putative targeting motif in the carboxy-terminal region. Primer extension analysis revealed that the mouse Ac45 gene contains two major transcription initiation sites. The start sites are not preceded by a clear CAAT-box and are located in a CpG island. The most downstream start site contains a TATA-box and transcriptional regulatory elements such as PEA-3, F2F, Maz and Sp1. The limited number of regulatory DNA elements common in the genes encoding Ac45 and V-ATPase subunits suggests a differential regulation of these genes. Together with the finding that Ac45 appears to occur only in multicellular organisms, these results indicate that this accessory subunit directs the V-ATPase to specialized and complex vacuolar systems.

Physical and genetic characterization reveals a pseudogene, an evolutionary junction, and unstable loci in distal Xq28

A large portion of human Xq28 has been completely characterized but the interval between G6PD and Xqter has remained poorly understood. Because of a lack of stable, high-density clone coverage in this region, we constructed a 1.6-Mb bacterial and P1 artificial chromosome (BAC and PAC, respectively) contig to expedite mapping, structural and evolutionary analysis, and sequencing. The contig helped to reposition previously mismapped genes and to characterize the XAP135 pseudogene near the int22h-2 repeat. BAC clones containing the distal int22h repeats also demonstrated spontaneous rearrangements and sparse coverage, which suggested that they were unstable. Because the int22h repeats are involved in genetic diseases, we examined them in great apes to see if they have always been unstable. Differences in copy number among the apes, due to duplications and deletions, indicated that they have been unstable throughout their evolution. Taking another approach toward understanding the genomic nature of distal Xq28, we examined the homologous mouse region and found an evolutionary junction near the distal int22h loci that separated the human distal Xq28 region into two segments on the mouse X chromosome. Finally, haplotype analysis showed that a segment within Xq28 has resisted excessive interchromosomal exchange through great ape evolution, potentially accounting for the linkage disequilibrium recently reported in this region. Collectively, these data highlight some interesting features of the genomic sequence in Xq28 and will be useful for positional cloning efforts, mouse mutagenesis studies, and further evolutionary analyses.

Plexin-A3 mediates semaphorin signaling and regulates the development of hippocampal axonal projections

Plexins are receptors implicated in mediating signaling by semaphorins, a family of axonal chemorepellents. The role of specific plexins in mediating semaphorin function in vivo has not, however, yet been examined in vertebrates. Here, we show that plexin-A3 is the most ubiquitously expressed plexin family member within regions of the developing mammalian nervous system known to contain semaphorin-responsive neurons. Using a chimeric receptor construct, we provide evidence that plexin-A3 can transduce a repulsive signal in growth cones in vitro. Analysis of plexin-A3 knockout mice shows that plexin-A3 contributes to Sema3F and Sema3A signaling and that plexin-A3 regulates the development of hippocampal axonal projections in vivo.

Compositional heterogeneity within and among isochores in mammalian genomes. II. Some general comments

The presence of long-range correlations and/or mosaicism in DNA sequences results in GC fluctuations, even within individual isochores that are much larger than expected correlation-free 'random' sequences. Neglecting the presence of such fluctuations can lead to incorrect conclusions regarding relative homogeneity or isochore borders. In this commentary, we address these and other methodological issues raised by the variations in GC level within human isochores. We also discuss some recent misconceptions.

Cytoprotection against mechanical forces delivered through beta 1 integrins requires induction of filamin A

Cells in mechanically active environments can activate cytoprotective mechanisms to maintain membrane integrity in the face of potentially lethal applied forces. Cytoprotection may be mediated by expression of membrane-associated cytoskeletal proteins including filamin A, an actin-binding protein that increases the rigidity of the subcortical actin cytoskeleton. In this study, we tested the hypotheses that applied forces induce the expression of filamin A specifically and that this putative protective response inhibits cell death. Magnetically generated forces were applied to protein-coated magnetite beads bound to human gingival fibroblasts, cells with constitutively low basal levels of filamin A mRNA and protein. Forces applied through collagen or fibronectin, but not bovine serum albumin or poly-l-lysine-coated beads, increased mRNA and protein content of filamin A by 3-7-fold. Forces had no effect on the expression of other filamin isotypes or other cytoskeletal proteins. This effect was dependent on the duration of force and was blocked by anti-beta(1) integrin antibodies. Force also stimulated a 60% increase in expression of luciferase under the control of a filamin A promoter in transiently transfected Rat2 fibroblasts and was dependent on Sp1 transcription factor binding sites located immediately upstream of the transcription start site. Experiments with actinomycin D-treated cells showed that the increased filamin A expression after force application was due in part to prolongation of mRNA half-life. Antisense filamin oligonucleotides blocked force-induced filamin A expression and increased cell death by >2-fold above controls. The force-induced regulation of filamin A was dependent on intact actin filaments. We conclude that cells from mechanically active environments can couple diverse signals from forces applied through beta-integrins to up-regulate the production of cytoprotective cytoskeletal proteins, typified by filamin A.

Human-mouse comparative sequence analysis of the NEMO gene reveals an alternative promoter within the neighboring G6PD gene

NEMO (NFkappaB essential modulator) is a non-catalytic subunit of the cytokine-dependent IkappaB kinase complex that is involved in activation of the transcription factor NFkappaB. The human NEMO gene maps to Xq28 and is arranged head to head with the proximal G6PD gene. Mutations in NEMO have recently been associated with Incontinentia Pigmenti (Smahi et al., Nature 405 (2000) 466), an X-linked dominant disorder. Three alternative transcripts with different non-coding 5' exons (1a, 1b and 1c) of NEMO have been described. In order to identify regulatory elements that control alternative transcription we have established the complete genomic sequence of the murine orthologs Nemo and G6pdx. Sequence comparison suggests the presence of two alternative promoters for NEMO/Nemo. First, a CpG island is shared by both genes driving expression of the NEMO/Nemo transcripts containing exons 1b and 1c in one direction and the housekeeping gene G6PD/G6pdx in the opposite direction. In contrast to human, an additional variant of exon 1c, named 1c+, was identified in several tissues of the mouse. This larger exon utilizes an alternative donor site located 1594 bp within intron 1c. The putative second promoter for NEMO/Nemo transcripts starting with exon 1a is unidirectional, and not associated with a CpG island. Surprisingly, this promoter is located in the second intron of G6PD/G6pdx. It shows very low basal activity and may be involved in stress/time- and/or tissue-dependent expression of NEMO. To our knowledge, an overlapping gene order similar to the G6PD/NEMO complex has not been described before.

Mapping of X chromosome inversion breakpoints [inv(X)(q11q28)] associated with FG syndrome: a second FG locus [FGS2]?

FG syndrome is an X-linked condition comprising mental retardation, congenital hypotonia, macrocephaly, distinctive facial changes, and constipation or anal malformations. In a linkage analysis, we mapped a major FG syndrome locus [FGS1] to Xq13, between loci DXS135 and DXS1066. The same data, however, clearly demonstrated genetic heterogeneity. Recently, we studied a French family in which an inversion [inv(X)(q12q28)] segregates with clinical symptoms of FG syndrome. This suggests that one of the breakpoints corresponds to a second FG syndrome locus [FGS2]. We report the results of fluorescence in situ hybridization analysis performed in this family using YACs and cosmids encompassing the Xq11q12 and Xq28 regions. Two YACs, one positive for the DXS1 locus at Xq11.2 and one positive for the color vision pigment genes and G6PD loci at Xq28, were found to cross the breakpoints, respectively. We postulate that a gene might be disrupted by one of the breakpoints.

Eleven densely clustered genes, six of them novel, in 176 kb of mouse t-complex DNA

Targeted sequencing of the mouse t-complex has started with a 176-kb, gene-rich BAC localized with six PCR-based markers in inversion 2/3 of the highly duplicated region. The sequence contains 11 genes recovered primarily as cDNAs from early embryonic collections, including Igfals (previously placed on chromosome 17), Nubp2 (a fully characterized gene), Jsap1 (a JNK-binding protein), Rsp29 (the mouse homologue of the rat gene), Ndk3 (a nucleoside diphosphate kinase), and six additional putative genes of unknown function. With 50% GC content, 75% of the DNA transcribed, and one gene/16.0 kb (on average), the region may qualify as one of the most gene-dense segments in the mouse genome and provides candidates for dosage-sensitive phenotypes and mouse embryonic lethals mapped to the vicinity.

Female mice heterozygous for IKK gamma/NEMO deficiencies develop a dermatopathy similar to the human X-linked disorder incontinentia pigmenti

IKK gamma/NEMO is the essential regulatory subunit of the I kappa B kinase (IKK), encoded by an X-linked gene in mice and humans. It is required for NF-kappa B activation and resistance to TNF-induced apoptosis. Female mice heterozygous for Ikk gamma/Nemo deficiency develop a unique dermatopathy characterized by keratinocyte hyperproliferation, skin inflammation, hyperkeratosis, and increased apoptosis. Although Ikk gamma+/- females eventually recover, Ikk gamma- males die in utero. These symptoms and inheritance pattern are very similar to those of incontinentia pigmenti (IP), a human genodermatosis, synthenic with the IKK gamma/NEMO locus. Indeed, biopsies and cells from IP patients exhibit defective IKK gamma/NEMO expression but normal expression of IKK catalytic subunits. This unique self-limiting disease, the first to be genetically linked to the IKK signaling pathway, is dependent on X-chromosome inactivation. We propose that the IKK gamma/NEMO-deficient cells trigger an inflammatory reaction that eventually leads to their death.

A high-density single-nucleotide polymorphism map of Xq25-q28

A high-density single-nucleotide polymorphism (SNP) map was developed for Xq25-q28 using a targeted approach to SNP discovery. This high-density map includes 217 new SNP markers, and 117 are informative in the CEPH parent population with >20% minor allele frequency. The average distance between SNP markers is 100 kb in the targeted regions. This is the densest genetic map of Xq25-q28 to date. The SNP markers are presented in order by their distance in megabases along the X chromosome, and the markers from the current genetic map are placed using the same scale to produce an integrated map of the region.

Highly specific localization of promoter regions in large genomic sequences by PromoterInspector: a novel context analysis approach

We present a new algorithm called PromoterInspector to locate eukaryotic polymase II promoter regions in large genomic sequences with a high degree of specificity. PromoterInspector focuses on the genetic context of promoters, rather than their exact location. Application of PromoterInspector can serve as a crucial pre-processing step for other methods to locate exactly, or to analyze promoters. PromoterInspector does not depend on heuristics, because it is purely based on libraries of IUPAC words extracted from training sequences by an unsupervised learning approach. We compared PromoterInspector to in silico promoter prediction tools using the sequences from the review by J.W. Fickett. PromoterInspector compared favourably on Fickett's evaluation scheme. A true positive to false positive ratio of 2.3 was obtained, surpassing the best ratio of 0.6, reported for TSSG. The application of our method to several large genomic sequences of over 1.3 million base-pairs in total resulted in even more specific predictions. The coverage of annotated promoters was comparable to other in silico promoter prediction methods, while the true positive predictions increased by up to 100% of total matches. PromoterInspector scans 100 kb in less than one minute on a workstation, and thus is especially applicable for large genome analysis. The method is available at http://genomatix.gsf. de/cgi-bin/promoterinspector/promoterinspector.pl.

Glucose-6-phosphate dehydrogenase deficiency

Glucose-6-phosphate dehydrogenase (G6PD) is expressed in all tissues, where it catalyses the first step in the pentose phosphate pathway. G6PD deficiency is prevalent throughout tropical and subtropical regions of the world because of the protection it affords during malaria infection. Although most affected individuals are asymptomatic, there is a risk of neonatal jaundice and acute haemolytic anaemia, triggered by infection and the ingestion of certain drugs and broad beans (favism). A rare but more severe form of G6PD deficiency is found throughout the world and is associated with chronic non-spherocytic haemolytic anaemia. Many deficient variants of G6PD have been described. DNA sequence analysis has shown that the vast majority of these are caused by single amino acid substitutions. The three-dimensional structure of G6PD shows a classical dinucleotide binding domain and a novel beta + alpha domain involved in dimerization.

Nuclease-hypersensitive chromatin formed by a CpG island in human DNA cloned as an artificial chromosome in yeast

CpG islands are mostly unmethylated GC-, and CpG-rich chromosomal segments overlapping promoter sequences in all housekeeping and many tissue-specific genes in vertebrates. Typically, these islands show an open chromatin structure, low in histone H1 and rich in acetylated histones. We have previously found that the island-like CGCG-rich sites in human DNA are hypersensitive to DNase I upon cloning in Saccharomyces cerevisiae. Here we studied, with a higher resolution, the chromatin formed in yeast by one such site, the CpG island accompanying the human glucose-6-phosphate dehydrogenase gene. We have found two strong hypersensitive sites and several positioned nucleosomes flanking the island despite the absence in yeast of such chromatin fiber-shaping factors as histone H1, methyltransferase, and the tissue-specific transcription factors. This finding, together with similar observations from our laboratories and others supports the idea that variations in GC and/or CpG content substantially contribute to the DNA sequence features modulating the structure of the chromatin. The composition-dependent fluctuations in the accessibility of DNA in the chromatin may constitute an evolutionary advantage and may explain the surprising compositional selection that acts in both the coding and non-coding segments of some genes during mammalian evolution.

Selection and mapping of replication origins from a 500-kb region of the human X chromosome and their relationship to gene expression

In higher eukaryotes the mechanism controlling initiation of DNA replication remains largely unknown. New technologies are needed to shed light on how DNA replication initiates along the genome in specific regions. To identify the human DNA sequence requirements for initiation of replication, we developed a new method that allows selection of replication origins starting from large genomic regions of human DNA. We repeatedly isolated 15 new putative replication origins (PROs) from a human DNA region of 500 kb in which 17 genes have previously been characterized. Fine-mapping of these PROs showed that DNA replication can initiate at many specific points along actively transcribed DNA in the cell lines used for our selection. In conclusion, in this paper we describe a new method to identify PROs that suggests that the availability of initiation sites is dependent on the transcriptional state of the DNA.

Human and mouse XAP-5 and XAP-5-like (X5L) genes: identification of an ancient functional retroposon differentially expressed in testis

Although most retroposons that arose by reverse transcription of cellular mRNAs and by reintegration into the genome are nonfunctional, several examples exist in which the retroposon acquired a novel function and became fixed in the genome as a functional gene. We identified another such case: the ubiquitously expressed X-linked XAP-5 gene with unknown function gave rise to its retroposed counterpart, XAP-5-like (X5L), which has an intronless open reading frame and is autosomal in human. Phylogenetic analysis of the human and mouse XAP-5 and X5L genes shows that the retroposition most likely took place before the radiation of eutherian mammals. The XAP-5 and X5L genes are expressed in a wide range of tissues but are differentially expressed in testis. The ancient origin and broad expression of the X5L retroposon indicate that the XAP-5 and X5L genes may have assumed different functions in somatic cells. In addition to this, because of its autosomal location and its high level and particular pattern of expression in spermatogenic cells, the X5L expression in testis may compensate for the X-linked XAP-5 gene, which may be silenced during spermatogenesis.

Genomic sequence, structural organization, molecular evolution, and aberrant rearrangement of promyelocytic leukemia zinc finger gene

The promyelocytic leukemia zinc finger gene (PLZF) is involved in chromosomal translocation t(11;17) associated with acute promyelocytic leukemia. In this work, a 201-kilobase genomic DNA region containing the entire PLZF gene was sequenced. Repeated elements account for 19.83%, and no obvious coding information other than PLZF is present over this region. PLZF contains six exons and five introns, and the exon organization corresponds well with protein domains. There are at least four alternative splicings (AS-I, -II, -III, and -IV) within exon 1. AS-I could be detected in most tissues tested whereas AS-II, -III, and -IV were present in the stomach, testis, and heart, respectively. Although splicing donor and acceptor signals at exon-intron boundaries for AS-I and exons 1-6 were classical (gt-ag), AS-II, -III, and -IV had atypical splicing sites. These alternative splicings, nevertheless, maintained the ORF and may encode isoforms with absence of important functional domains. In mRNA species without AS-I, there is a relatively long 5' UTR of 6.0 kilobases. A TATA box and several transcription factor binding sites were found in the putative promoter region upstream of the transcription start site. PLZF is a well conserved gene from Caenorhabditis elegans to human. PLZF paralogous sequences are found in human genome. The presence of two MLL/PLZF-like alignments on human chromosome 11q23 and 19 suggests a syntenic replication during evolution. The chromosomal breakpoints and joining sites in the index acute promyelocytic leukemia case with t(11;17) also were characterized, which suggests the involvement of DNA damage-repair mechanism.

Dyskeratosis Congenita Caused by a 3′ Deletion: Germline and Somatic Mosaicism in a Female Carrier

X-linked dyskeratosis congenita (DC) is a bone marrow failure syndrome caused by mutations in the DKC1 gene located at Xq28. By 20 years of age, most affected boys develop bone marrow failure, whereas female carriers show a skewed pattern of X-chromosome inactivation. The gene product, dyskerin, is homologous to a yeast protein involved in ribosomal RNA biogenesis, providing a unique insight into a cause of aplastic anemia. Whereas most causative mutations are single amino acid substitutions, and nonsense or frameshift mutations have not been observed, we present here a case of DC caused by a 2-kb deletion that removes the last exon of the gene. Normal levels of mRNA are produced from the deleted gene, with the transcripts using a cryptic polyadenylation site in the antisense strand of the adjacent MPP1 gene, normally located 1 kb downstream of DKC1 in a tail to tail orientation. The predicted truncated protein lacks a lysine-rich peptide that is less conserved than the rest of the dyskerin molecule and is dispensable in yeast, supporting the contention that it may retain some activity and that null mutations at this locus may be lethal. The affected boy had an unaffected brother with the same haplotype around the DKC1 gene and a sister who was heterozygous for the deletion. We conclude therefore that the mother must be a germline mosaic with respect to this deletion. Investigation of her blood cells and other somatic tissues showed that a small proportion of these cells also carried the deletion, making her a somatic mosaic and indicating that the deletion took place early in development.

Dyskeratosis Congenita Caused by a 3′ Deletion: Germline and Somatic Mosaicism in a Female Carrier

X-linked dyskeratosis congenita (DC) is a bone marrow failure syndrome caused by mutations in the DKC1 gene located at Xq28. By 20 years of age, most affected boys develop bone marrow failure, whereas female carriers show a skewed pattern of X-chromosome inactivation. The gene product, dyskerin, is homologous to a yeast protein involved in ribosomal RNA biogenesis, providing a unique insight into a cause of aplastic anemia. Whereas most causative mutations are single amino acid substitutions, and nonsense or frameshift mutations have not been observed, we present here a case of DC caused by a 2-kb deletion that removes the last exon of the gene. Normal levels of mRNA are produced from the deleted gene, with the transcripts using a cryptic polyadenylation site in the antisense strand of the adjacent MPP1 gene, normally located 1 kb downstream of DKC1 in a tail to tail orientation. The predicted truncated protein lacks a lysine-rich peptide that is less conserved than the rest of the dyskerin molecule and is dispensable in yeast, supporting the contention that it may retain some activity and that null mutations at this locus may be lethal. The affected boy had an unaffected brother with the same haplotype around the DKC1 gene and a sister who was heterozygous for the deletion. We conclude therefore that the mother must be a germline mosaic with respect to this deletion. Investigation of her blood cells and other somatic tissues showed that a small proportion of these cells also carried the deletion, making her a somatic mosaic and indicating that the deletion took place early in development.

XLMR syndrome characterized by multiple respiratory infections, hypertelorism, severe CNS deterioration and early death localizes to distal Xq28

We report on a family with severe X-linked mental retardation (XLMR) and progressive, severe central nervous system deterioration. Three of the five affected males died of secondary complications before the age of 10 years and none have survived past the age of 10. These complications included swallowing dysfunction and gastroesophageal reflux with secondary recurrent respiratory infections. In addition, hypotonia and a mild myopathy were also present. All had a characteristic facies, including downslanting palpebral fissures, hypertelorism, and a short nose with a low nasal bridge. The two older boys showed cerebral atrophy by CT. No metabolic abnormalities were identified. Three obligate carriers had an IQ less than 80. The causal gene has been localized distal to DXS8103 in Xq28, a region spanning 5cM. No other XLMR disorder with these manifestations have been localized to this region and this appears to be a new disorder.

Isolation of full-length cDNA and chromosomal localization of human NF-kappaB modulator NEMO to Xq28

NEMO is an essential component of the IkappaB kinase complex. Others have shown that expression of mouse NEMO can complement the lack of responsiveness to NF-kappaB stimuli in two NEMO-deficient cell lines. Here we report the isolation of a full-length human NEMO cDNA. Virtual translation of human NEMO cDNA predicts a 48-kD coiled-coil protein which shares 87.9% identity and 90.5% similarity with the mouse homolog. By sequence alignment, we mapped the human NEMO gene to chromosome Xq28. We note that the NEMO and the G6PD (glucose-6-phosphate dehydrogenase) loci are arranged in a head-to-head orientation separated by no more than 800 bp. This map location is further supported by the sequence of an alternatively spliced variant of human NEMO mRNA. Thus, human NEMO is an X-linked gene closely adjacent to the G6PD locus.

Inherited duplication Xq27-qter at Xp22.3 in severely affected males: molecular cytogenetic evaluation and clinical description in three unrelated families

We describe the clinical phenotype in four males from three families with duplication (X)(qter-->q27::p22.3-->qter). This is an unusual duplication of the distal long arm segment, Xq27-qter, onto the distal short arm of the X chromosome at Xp22.3, as shown by fluorescent in situ hybridization analysis with multiple X-specific probes. The patients are young male offspring of three unrelated, phenotypically normal carrier women. The affected males have similar clinical manifestations including severe growth retardation and developmental delay, severe axial hypotonia, and minor anomalies. Such clinical similarity in three unrelated families demonstrates that this chromosome abnormality results in a new and distinct clinical phenotype. Replication studies, performed on two of the mothers, provided evidence that inactivation of the abnormal X chromosome permitted the structural abnormality to persist in these families for a generation or more in females without phenotypic expression.

1.4 Mb candidate gene region for X linked dyskeratosis congenita defined by combined haplotype and X chromosome inactivation analysis

Dyskeratosis congenita (DC) is a rare inherited disorder characterised by the early onset of reticulate skin pigmentation, nail dystrophy, and mucosal leucoplakia. In over 80% of cases bone marrow failure develops and this is the main cause of early mortality. The DC1 gene responsible for the X linked form (MIM 305000) of dyskeratosis congenita has been mapped to Xq28. In order to narrow the candidate gene region, genetic linkage analysis was performed in eight X linked pedigrees using a set of markers spanning Xq28. A maximum lod score of 5.31 with no recombinations was achieved with marker DXS1073. Two recombination events were identified; one of these uses X chromosome inactivation pattern analysis to determine carrier status and haplotype analysis to fine map the recombination breakpoint. The fine mapping of these recombination events has enabled the candidate gene region for X linked dyskeratosis congenita to be defined as the 1.4 Mb interval between Xq3274 and DXS1108.

Genomic organization of a 225-kb region in Xq28 containing the gene for X-linked myotubular myopathy (MTM1) and a related gene (MTMR1)

MTM1 is responsible for X-linked recessive myotubular myopathy, which is a congenital muscle disorder linked to Xq28. MTM1 is highly conserved from yeast to humans. A number of related genes also exist. The MTM1 gene family contains a consensus sequence consisting of the active enzyme site of protein tyrosine phosphatases (PTPs), suggesting that they belong to a new family of PTPs. Database searches revealed homology of myotubularin and all related peptides to the cisplatin resistance-associated alpha protein, which implicates an as yet unknown function. In addition, homology to the Sbf1 protein (SET binding factor 1), involved in the oncogenic transformation of fibroblasts and differentiation of myoblasts, was also evident. We describe 225 kb of genomic sequence containing MTM1 and the related gene, MTMR1, which lies 20 kb distal to MTM1. Although there is only moderate conservation of the exons, the striking similarity in the gene structures indicates that these two genes arose by duplication. Calculations suggest that this event occurred early in evolution long before separation of the human and mouse lineages. So far, mutations have been identified in the coding sequence of only 65% of the patients analyzed, indicating that the remaining mutations may lie in noncoding regions of MTM1 or possibly in MTMR1. Knowledge of the genomic sequence will facilitate mutation analyses of the coding and noncoding sequences of MTM1 and MTMR1.

Pathological consequences of sequence duplications in the human genome

As large-scale sequencing accumulates momentum, an increasing number of instances are being revealed in which genes or other relatively rare sequences are duplicated, either in tandem or at nearby locations. Such duplications are a source of considerable polymorphism in populations, and also increase the evolutionary possibilities for the coregulation of juxtaposed sequences. As a further consequence, they promote inversions and deletions that are responsible for significant inherited pathology. Here we review known examples of genomic duplications present on the human X chromosome and autosomes.

Recombination trapping: an in-vivo approach to recover cDNAs encoded in YACs

We have developed an approach to identify and localize cDNAs encoded by YACs. In this scheme, a YAC truncation vector containing a cDNA library is used to interrupt the YAC by homologous recombination in yeast. This approach generates YACs truncated at the site of recombination between the cDNA and the cognate YAC sequence and thus localizes the gene in the YAC. This method results in the production of a large percentage of true recombinants identifying gene encoding regions of the genome. This approach is shown to identify an unique EST sequence from a YAC in Xp22, the recently described transketolase-like gene in a YAC from Xq28 and a putative kinesin-like gene in Xq13. This system should also be useful in the mapping of YACs by targeted integration. We have constructed a new telomere truncation vector, pGR8, which incorporates two selectable markers, HIS5 and LYS2. This vector overcomes problems of previous vectors including: incompatibility with most YAC libraries, vector homology with the YAC arms and high backgrounds resulting from the use of a single selectible marker. A third counterselection with 5-fluoroorotic acid (5FOA) against yeast clones retaining the URA3 gene was also employed to reduce background further. Therefore, this vector and approach should be useful to the transcriptional analysis of YAC maps of any genome.

Mutations in GDI1 are responsible for X-linked non-specific mental retardation

Rab GDP-dissociation inhibitors (GDI) are evolutionarily conserved proteins that play an essential role in the recycling of Rab GTPases required for vesicular transport through the secretory pathway. We have found mutations in the GDI1 gene (which encodes uGDI) in two families affected with X-linked non-specific mental retardation. One of the mutations caused a non-conservative substitution (L92P) which reduced binding and recycling of RAB3A, the second was a null mutation. Our results show that both functional and developmental alterations in the neuron may account for the severe impairment of learning abilities as a consequence of mutations in GDI1, emphasizing its critical role in development of human intellectual and learning abilities.

Sequencing multimegabase-template DNA with BigDye terminator chemistry

Using the recently introduced BigDye terminators, large-template DNA can be directly sequenced with custom primers on automated instruments. Cycle sequencing conditions are presented to sequence DNA samples isolated from a number of microbial genomes including 750-kb Ureaplasma urealyticum, 1.2-Mb Mycoplasma fermentans, 2.3-Mb Streptococcus pneumoniae, and 4.6-Mb Escherichia coli. Average read lengths of >700 bp from unique primer annealing sites are often sufficient to fill final gaps in microbial genome sequencing projects without additional manipulations of template DNA. The technique can also be applied to sequence-targeted regions, thereby bypassing tedious subcloning steps.

X-linked dyskeratosis congenita is caused by mutations in a highly conserved gene with putative nucleolar functions

X-linked recessive dyskeratosis congenita (DKC) is a rare bone-marrow failure disorder linked to Xq28. Hybridization screening with 28 candidate cDNAs resulted in the detection of a 3' deletion in one DKC patient with a cDNA probe (derived from XAP101). Five different missense mutations in five unrelated patients were subsequently identified in XAP101, indicating that it is the gene responsible for X-linked DKC (DKC1). DKC1 is highly conserved across species barriers and is the orthologue of rat NAP57 and Saccharomyces cerevisiae CBF5. The peptide dyskerin contains two TruB pseudouridine (psi) synthase motifs, multiple phosphorylation sites, and a carboxy-terminal lysine-rich repeat domain. By analogy to the function of the known dyskerin orthologues, involvement in the cell cycle and nucleolar function is predicted for the protein.

An oligo-screening strategy to fill gaps found during shotgun sequencing projects

During the course of projects to sequence human and nematode cosmids we encountered great difficulties generating contiguous sequence in regions with repetitive DNA (Alu repeats in humans and tandem or inverted repeats in the nematode). We have developed a simple and efficient strategy to fill gaps. By screening M13, plasmid or phagemid libraries with oligonucleotides flanking the gap, clones are identified that contiguate the cosmid sequence. Our method has been integrated into the GAP4 sequence assembly program. The strategy reduces both time and costs in large scale sequencing projects.

Emerin deletions occurring on both Xq28 inversion backgrounds

Emery-Dreifuss muscular dystrophy (EMD) is an X-linked disorder characterized by contractures, progressive weakness and cardiomyopathy. EMD is caused by mutations in the 2 kb emerin gene that is located within human Xq28. Emerin is immediately distal to the 26 kb filamin gene, and flanking the filamin-emerin region are two large inverted repeats. This entire region previously has been found to be inverted in approximately 20% of X chromosomes, presumably mediated by the inverted repeats. Only one complete emerin deletion has been reported previously. It was found to be due to a complex rearrangement involving the inverted repeats which partially duplicated filamin. We report here two additional EMD patients who have large deletions of 20 and 34 kb, respectively. Unlike the previously reported deletion, these deletions appear to be simple deletions, with each breakpoint junction showing only 2 bp of overlap, suggesting an end-joining mechanism. However, the two deletions were found on each of the two inverted backgrounds. The 20 kb deletion includes the entire emerin gene and extends well into most of the distal inverted repeat. In contrast, the 34 kb deletion occurs on the inverted X chromosome and extends centromeric, well beyond the proximal inverted repeat. In addition, at least three nearby putative genes detected by previous sequence analysis are deleted among these patients but without obvious deviation from a typical EMD phenotype. Similarly to the previously reported deletion, filamin remains intact in these two deletions. All three deletions involve distinct breakpoints within the 4.7 kb filamin-emerin intergenic region, suggesting that loss of filamin is a lethal event. Thus, the close proximity of filamin to emerin may place constraints upon potential emerin deletions and probably accounts for the rarity of complete emerin deletions in EMD patients.

Duplication and distribution of repetitive elements and non-unique regions in the human genome

Genome mapping efforts and the initial sequencing of large segments of human DNA permit ongoing assessment of the patterns and extent of sequence duplication and divergence in the human genome. Initial sequence data indicate that the most highly repetitive sequences show isochore-related enrichment and clustering produced by successive insertional recombination and local duplication of particular repetitive elements. Regional duplication is also observed for a number of otherwise unique genomic sequences and thereby makes these segments become repetitive. The consequences of these duplication events are: (1) clustering of related genes, along with a variety of coregulatory mechanisms; and (2) recombinations between the nearby homologous sequences, which can delete genes in individuals and account for a significant fraction of human genetic disease.

Sequence-based exon prediction around the synaptophysin locus reveals a gene-rich area containing novel genes in human proximal Xp

The human Xp11.23-p11.22 interval has been implicated in several inherited diseases including Wiskott-Aldrich syndrome; three forms of X-linked hypercalciuric nephrolithiaisis; and the eye disorders retinitis pigmentosa 2, congenital stationary night blindness, and Aland Island eye disease. In constructing YAC contigs spanning Xp11. 23-p11.22, we have previously shown that the region around the synaptophysin (SYP) gene is refractory to cloning in YACs, but highly stable in cosmids. Preliminary analysis of the latter suggested that this might reflect a high density of coding sequences and we therefore undertook the complete sequencing of a SYP-containing cosmid. Sequence data were extensively analyzed using computer programs such as CENSOR (to mask repeats), BLAST (for homology searches), and GRAIL and GENE-ID (to predict exons). This revealed the presence of 29 putative exons, organized into three genes, in addition to the 7 exons of the complete SYP coding region, all mapping within a 44-kb interval. Two genes are novel, one (CACNA1F) showing high homology to alpha1 subunits of calcium channels, the other (LMO6) encoding a product with significant similarity to LIM-domain proteins. RT-PCR and Northern blot studies confirmed that these loci are indeed transcribed. The third locus is the previously described, but not previously localized, A4 differentiation-dependent gene. Given that the intron-exon boundaries predicted by the analysis are consistent with previous information where available, we have been able to suggest the genomic organization of the novel genes with some confidence. The region has an elevated GC content (>53%), and we identified CpG islands associated with the 5' ends of SYP, A4, and LMO6. The order of loci was Xpter-A4-LMO6-SYP-CACNA1F-Xcen, with intergenic distances ranging from approximately 300 bp to approximately 5 kb. The density of transcribed sequences in this area (>80%) is comparable to that found in the highly gene-rich chromosomal band Xq28. Further studies may aid our understanding of the long-range organization surrounding such gene-enriched regions.

A gene in human chromosome band Xq28 (GABRE) defines a putative new subunit class of the GABAA neurotransmitter receptor

We have isolated and sequenced a novel human gene (GABRE) of the GABAA neurotransmitter receptor family. A cDNA sequence of the gene coding for a 506 amino acid protein was identified, representing a member of a putative new class (epsilon) of the GABAA receptor. The gene is transcribed at least at low level in several different tissues, with the highest levels being detected in adult heart and placenta. Alternative splicing of GABRE transcripts isolated from different tissues was observed at multiple positions of the gene, yielding an unusually complex variety of cDNA variants. The structure of the 5' region of most cDNAs is compatible with expression of protein sequence epsilon only in adult brain, whereas in other tissues, the majority of transcripts code for truncated protein sequences. The GABRE gene extends over 14 kb and is clustered together with the alpha 3 and the putative beta 4 GABAA receptor subunit genes in an approximately 0.8-Mb interval in chromosome band Xq28, located in the candidate regions of two different neurologic diseases. Based on features of conservation of protein sequences, gene structure, and genomic organization of GABAA receptor gene clusters, we propose that the epsilon and gamma subunit genes have a common ancestor and that GABAA receptor gene clusters in the human genome have diverged by multiple duplication events of an ancestral gene cluster containing one each alpha, beta, and gamma/epsilon precursor gene.

Analysis of exon/intron structure and 400 kb of genomic sequence surrounding the 5'-promoter and 3'-terminal ends of the human glypican 3 (GPC3) gene

GPC3, the gene modified in the Simpson-Golabi-Behmel gigantism/overgrowth syndrome (SGBS), is shown to span more than 500 kb of genomic sequence, with the transcript beginning 197 bp 5' of the translational start site. The Xq26.1 region containing GPC3 as the only known gene has been extended to > 900 kb by sequence analysis of flanking BAC clones. Two GC isochores (40.6 and 42.6% GC) are observed at the 5' and 3' ends of the locus, with a large repertoire of repetitive sequences that includes an unusual cluster of four L1 elements > 92% identical over 2.8 kb. Eight exons, accounting for the full 2.4-kb GPC3 cDNA, have been sequenced along with neighboring intronic regions. PCR assays have been developed to amplify each exon and exon/intron junction sequence, to help discriminate instances of SGBS among individuals with overgrowth syndromes and to facilitate mutational analysis of lesions in the gene.

Differential expression of XAP5, a candidate disease gene

We have isolated a full-length cDNA corresponding to the XAP5 gene in Xq28. An unusual feature of the cDNA is that it contains runs of CCG repeats in the 5' untranslated region, typical of genes that exhibit anticipation. It has a striking pattern of differential expression and is greatly enhanced in various fetal tissues. This predicted protein encodes a unique 339-amino-acid polypeptide that contains a large percentage of highly charged residues and a possible nuclear localization signal. A comparison to genomic sequence shows that XAP-5 comprises 13 exons spanning 6.5 kb. An examination of the human population indicates that the longest CCG run is polymorphic and varies in length from 8 to 12 repeats.

Identification of a gene within the tandem array of red and green color pigment genes

The tandem array of color pigment genes on chromosome Xq28 contains nested exons of a distinct gene. This gene (termed TEX28) is composed of five exons that span almost the entire distance between the protein-coding regions of the color pigment genes and a transketolase-related gene. Although most of the TEX28 gene is repeated within the color pigment gene array, the exclusion of exon 1 from the array is predicted to restrict transcription to a single copy of the gene. The TEX28 gene encodes a polypeptide of 410 amino acid residues. This polypeptide does not display significant homology with any known proteins in public databases. Transcripts of the gene (1.8 kb) were detected in testes, but not in any other tissue examined. Color vision disorders that result from the deletion of color pigment genes should be reappraised for associated phenotypes that may derive from disruption of the TEX28 gene.

Genomic structure of a novel LIM domain gene (ZNF185) in Xq28 and comparisons with the orthologous murine transcript

Construction of a transcript map in the DXS52 region in Xq28 had previously led to the isolation of a cDNA with a LIM zinc finger domain in the carboxyl terminus. In parallel, the orthologous murine transcript was isolated from the syntenic region. The human and mouse cDNAs have been designated ZNF185 and Zfp185, respectively. By integrating the cDNA sequence with the cosmid-derived genomic sequence the exon-intron structure of the 3' end of the ZNF185 gene was resolved. Comparative sequence analyses of the human genomic sequence with the full-length murine cDNA facilitated prediction of the 5' end of the gene. The selective expression of three transcripts corresponding to the ZNF185 gene and a related gene was shown by Northern and Southern blots. In situ hybridizations revealed a nonubiquitous and stage-specific expression of Zfp185, especially in differentiating connective tissue. Since LIM proteins regulate cellular proliferation and/or differentiation by diverse mechanisms, and some have directly been associated with disease, conceivably ZNF185 may represent a candidate for a disease-causing gene linked to Xq28. Knowledge of the genomic structure will permit detailed mutation analyses.

Ataxia-telangiectasia locus: sequence analysis of 184 kb of human genomic DNA containing the entire ATM gene

Ataxia-telangiectasia (A-T) is an autosomal recessive disorder involving cerebellar degeneration, immunodeficiency, chromosomal instability, radiosensitivity, and cancer predisposition. The genomic organization of the A-T gene, designated ATM, was established recently. To date, more than 100 A-T-associated mutations have been reported in the ATM gene that do not support the existence of one or several mutational hotspots. To allow genotype/phenotype correlations it will be important to find additional ATM mutations. The nature and location of the mutations will also provide insights into the molecular processes that underly the disease. To facilitate the search for ATM mutations and to establish the basis for the identification of transcriptional regulatory elements, we have sequenced and report here 184,490 bp of genomic sequence from the human 11q22-23 chromosomal region containing the entire ATM gene, spanning 146 kb, and 10 kb of the 5'-region of an adjacent gene named E14/NPAT. The latter shares a bidirectional promoter with ATM and is transcribed in the opposite direction. The entire region is transcribed to approximately 85% and translated to 5%. Genome-wide repeats were found to constitute 37.2%, with LINE (17.1%) and Alu (14.6%) being the main repetitive elements. The high representation of LINE repeats is attributable to the presence of three full-length LINE-1s, inserted in the same orientation in introns 18 and 63 as well as downstream of the ATM gene. Homology searches suggest that ATM exon 2 could have derived from a mammalian interspersed repeat (MIR). Promoter recognition algorithms identified divergent promoter elements within the CpG island, which lies between the ATM and E14/NPAT genes, and provide evidence for a putative second ATM promoter located within intron 3, immediately upstream of the first coding exon. The low G+C level (38.1%) of the ATM locus is reflected in a strongly biased codon and amino acid usage of the gene.

Identification by shotgun sequencing, genomic organization, and functional analysis of a fourth arylsulfatase gene (ARSF) from the Xp22.3 region

We recently reported the isolation of two new members of the sulfatase gene family, arylsulfatase D (ARSD) and E (ARSE), located approximately 50 kb from each other in the Xp22.3 region. Mutation analysis indicated ARSE as the gene responsible for X-linked recessive chondrodysplasia punctata. Expression of the ARSE gene in COS cells resulted in a heat-labile arylsulfatase activity that was inhibited by warfarin. At the same time, we detected the presence of a 1.2-kb fragment located at approximately 60 kb from ARSD and ARSE with significant homology to these two genes, suggesting the existence of another sulfatase gene, arylsulfatase F (ARSF), in Xp22.3. We have used a combined approach of long-range genomic sequencing and screening of cDNA libraries to isolate the ARSF gene. Expression of the ARSF cDNA in COS cells resulted in a heat-labile arylsulfatase activity that is not inhibited by warfarin, supporting our hypothesis that only ARSE is specifically inhibited by warfarin and is most likely involved in warfarin embryopathy. Genomic analysis revealed that ARSF has an intron/exon organization highly similar to those of ARSD and ARSE, which is also shared by another Xp22.3 sulfatase gene, ARSC (arylsulfatase C, also known as steroid sulfatase), with the splice sites occurring at the same position in all four genes. The data obtained from sequence analysis and presented in this paper indicate that the ARSC, ARSD, ARSE, and ARSF genes are more similar to each other than to other members of the sulfatase gene family, supporting our hypothesis that they represent a subfamily of related proteins created through duplication events that occurred in an ancestral pseudoautosomal region.