Duplication of a gene-rich cluster between 16p11.1 and Xq28: a novel pericentromeric-directed mechanism for paralogous genome evolution
Duplication of a gene-rich cluster between 16p11.1 and Xq28: a novel pericentromeric-directed mechanism for paralogous genome evolutionEvan E. Eichler[dagger], Fei Lu, Ying Shen, Rachele Antonacci, Vesna Jurecic, Norman A. Doggett1, Robert K. Moyzis1, Antonio Baldini, Richard A. Gibbs and David L. Nelson*
Department of Molecular and Human Genetics, Human Genome Center, Baylor College of Medicine, Houston, TX 77030, USA and 1Center for Human Genome Studies, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
Received February 22, 1996;Revised and Accepted April 26, 1996
We have identified a 26.5 kb gene-rich duplication shared by human Xq28 and 16p11.1. Complete comparative sequence analysis of cosmids from both loci has revealed identical Xq28 and 16p11.1 genomic structures for both the human creatine transporter gene (SLC6A8) and five exons of the CDM gene (DXS1357E). Overall nucleotide similarity within the duplication was found to be 94.6%, suggesting that this interchromosomal duplication occurred within recent evolutionary time (7-10 mya). Based on comparisons between genomic and cDNA sequence, both the Xq28 creatine transporter and DXS1357E genes are transcriptionally active. Predicted translation of exons and RT-PCR analysis reveal that chromosome 16 paralogs likely represent pseudogenes. Comparative fluorescent in situ hybridization (FISH) analyses of chromosomes from various primates indicate that this gene-rich segment has undergone several duplications. In gorilla and chimpanzee, multiple pericentromeric localizations on a variety of chromosomes were found using probes from the duplicated region. In other species, such as the orangutan and gibbon, FISH signals were only identified at the distal end of the X chromosome, suggesting that the Xq28 locus represents the ancestral copy. Sequencing of the 16p11.1/Xq28 duplication breakpoints has revealed the presence of repetitive immunoglobulin-like CAGGG pentamer sequences at or near the paralogy boundaries. The mobilization and dispersal of this gene-rich 27 kb element to the pericentromeric regions of primate chromosomes defines an unprecedented form of recent genome evolution and a novel mechanism for the generation of genetic diversity among closely related species.
Duplications of genes involving the conservation of structure and gene order have been observed both intrachromosomally and interchromosomally in a variety of organisms (1 -4 ). Intrachromosomal paralogies are generally restricted to discrete cytogenetic band locations. Recent investigations into the genomic organization of several human gene clusters such as the human CEA gene family (carcinoembryonic antigens) (5 ), human olfactory receptor cluster (6 ), the ZNF (zinc finger) gene family on human 19p12 (7 ), the pregnancy-specific glycoprotein gene family (5 ) and the MAGE (melanoma antigen gene) cluster on the short arm of the X chromosome (8 ) indicate that endoduplication of a single gene or an ancestral repertoire of genes has been the likely mechanism for the generation of the observed tandem array structures. Some cross-species comparisons reveal that such endoduplications have occurred quite recently in human evolution (7 ); while others, such as the duplications involved in the archetypal Hox cluster organization, are much more ancient, being closely conserved among all vertebrate species (4 ,9 ,10 ).
In contrast to intrachromosomal paralogies, duplications of genes between chromosomes are believed to be mechanistically and functionally distinct (11 ,12 ). Such duplications transfer cognate genes to new genomic environments, with the possibility to acquire novel combinations of long-range cis-acting regulatory elements and ultimately new functional roles in the organism (13 ). The large paralogous segments between S. cerevisae chromosomes II and IV involving the functional ARC and COR gene clusters (1 ), the two copies of the insulin-growth factor genes on human chromosomes 11 and 12 (14 ) and the four copies of the vertebrate HOX cluster are a few examples of interchromosomal duplications which are presumed to have acquired functional diversity.
One theory postulated to account for the occurrence of interchromosomal paralogous genes has involved the duplication of chromosome number associated with major chordate speciation events (15 -18 ). Tetraploidization followed by the re-establishment of the disomic state by macromutational processes such as translocations, inversions and Robertsonian fusions have been invoked to explain the large-scale paralogous gene organizations and the similarity in cytogenetic banding pattern observed between some human chromosomes, such as 11p and 12p (15 ,16 ,18 -20 ). Three tetraploidization events are believed to have occurred during the evolution of mammals from protochordate ancestor, with the latest occurring approximately 250 mya (18 ). Interestingly, interspecific comparisons of gene organization and the estimation of the timing of duplication events between the [alpha]- and [beta]-globins (21 ), the two insulin growth factor gene families (14 ,22 ) and the HOX gene clusters in vertebrates (4 ), all suggest that interchromosomal exchange are ancient events, occurring 300-500 million years ago. These results lend further support to the argument that interchromosomal paralogies in these families represent the remnants of early tetraploidization events (13 ,18 ).
We report the identification and complete sequence analysis of an interchromosomal paralogous region between human cytogenetic bands Xq28 and 16p11.1. At least two genes, the creatine transporter (SLC6A8) and the CDM (DXS1357E) gene, have been involved in this duplication event. Comparative sequence analysis of 26.5 kb of paralogous DNA and 12.5 kb of flanking sequence indicates an overall nucleotide similarity of 94.6%, suggesting that this duplication event has occurred relatively recently (7-10 mya) in the hominoid lineage of evolution. Comparative fluorescent in situ hybridization (FISH) analysis of orthologous 16p11.1 and Xq28 loci among the lesser and greater apes confirm that the duplications have occurred relatively recently during the pongid/hominid lineage of evolution and that the chromosome X paralogous gene-rich segment likely represents the ancestral template copy. Among gorillas and chimpanzees, duplications of the Xq28 sequence have been observed on multiple chromosomes which are non-syntenic to human chromosome 16. In each case, the targets of the duplication events have been within the pericentromeric regions of chromosomes, suggesting a region-specific directed mechanism for the generations of these interchromosomal duplications. Sequence determination of the breakpoints on human Xq28 and 16p11.1 has identified complicated interspersed CAGGG pentamer repeats located near or at the boundaries of interchromosomal paralogy. The similarity of these repeats to immunoglobulin switch-like recombination elements (CTGGG), the pericentromeric site-specificity of duplications, the burst of intrachromosomal paralogies involving these genes in a short period of evolution and the maintenance of the transcriptionally active X locus suggest that recombinogenic processes such as replicative transposition have been involved in the duplication, mobilization and integration of the Xq28 gene-rich segment to non-homologous regions of the primate genome. The documentation of such a large (26.5 kb) transposition which has occurred recently in the hominoid genome identifies a novel mechanism for interchromosomal paralogy which may have implications in generation of diversity of Na+Cl- dependent transporters and other multigene families of primates.
Representative cosmids from various Xq28 contigs were mapped by Southern hybridization to X chromosome somatic-cell hybrid deletion panels and by FISH in order to determine the physical location of each set of contiguous cosmid clones. Cosmid clones from one contig (designated X1) consistently demonstrated strong hybridization signals corresponding to cytogenetic bands Xq28 and 16p11.1 by FISH (Fig. 1 ). Screening of arrayed cDNA libraries with these cosmids identified two different placental cDNA clones, p3A1 and p96D7. Sequence analysis of the 1.9 kb and 1.25 kb NotI inserts of p96D7 and p3A1 clones revealed that the cDNAs corresponded to the ORF of the recently described human creatine transporter gene(s) (SLC6A8) (GenBank accession L31409; GenBank accession S74309) (23 ,24 ) and the CDM (DXS1357E) gene (GenBank accession Z31696) (25 ). The first exon of the CDM gene had been previously mapped to a location 450 bp proximal to the 5' UTR of the ALD (adrenoleukodystrophy) gene (25 ,26 ), confirming the Xq28 location of the X1 cosmid contig within the BGN (biglycan) and L1CAM (L1 cell-adhesion molecule) interval (Fig. 1 ) (27 ).
In order to estimate the size of duplication between 16p11.1 and Xq28, a cosmid contig of approximately 150 kb was developed proximal to the creatine transporter locus extending distally to the 3' UTR of the ALD gene of Xq28 (Fig. 1 ). In addition, various cDNA and genomic probes including pbgn900 (a 920 bp cDNA fragment of the biglycan locus), TA19 (a genomic clone including exons 6 and 7 of ALD), c6 (a 3.5 kb genomic subclone which includes the first exon of creatine transporter gene) and H8 (a 2.5 kb cDNA fragment corresponding to exons 1-9 of ALD) were employed to facilitate in the identification and orientation of cosmids within and flanking this region (26 ). The position of these clones and a minimum tiling path of cosmids in the contig is summarized in Figure 1 . Using representative cosmids of the region, FISH of human metaphase chromosome spreads was used to assay the extent of paralogy between X and 16 (Fig. 1 ). FISH analysis indicates that the duplicated region between X and 16 is relatively small, approximately 30 kb, and that it does not include the X chromosome BGN and ALD genes. Interestingly, a second domain of paralogy was identified distal to the X-16 duplicated region. Using intron/exon genomic clones (TA14) and ALD cDNA fragments as probes, FISH analysis indicates that a paralog of the adrenoleukodystrophy gene likely exists at cytogenetic band 2p11.1-2p11.2 (25 ,26 ) (Fig. 1 ).
A shotgun M13 library was prepared from an Xq28 cosmid, u56c4, which cross-hybridized strongly to both 16p11.1 and Xq28 and which contained substantial portions of the putative DXS1357E and SLC6A8 genes. M13 subclones were sequenced using dideoxy dye-primer cycle sequencing methods and assembled using XDAP software, reconstituting the 33 kb genomic insert of cosmid clone u56C4 (GenBank accession U36341). Comparative sequence analysis between u56C4 and previously published cDNA sequence of the SLC6A8 (GenBank accession L31409) and the DXS1357E gene (GenBank Z31696) revealed greater than 99.8% nucleotide similarity, resulting in identical predicted amino-acid composition. These data indicate that the Xq28 genes likely represent the transcriptionally active loci for published DXS1357E and SLC6A8 transcripts. The organization of the creatine transporter gene (4.3 kb transcript) (23 ) is compact. It consists of 13 exons spanning 8.5 kb of genomic DNA (Fig. 2 and Table 1 ). Alignment of the putative amino-acid sequence of the creatine transporter with the gene's physical structure reveals a modular exon organization, with one exon for every transmembrane domain over the first eight exons. In an area where compaction of intron/exon structure is greatest (exons 8-10; average intron size 50 bp), the modular organization breaks down with transmembrane protein domains spanning intron/exon boundaries (Fig. 2 and Table 1 ). The DXS1357E gene is located in close proximity (5 kb) to the SLC6A8 gene and is transcribed with opposite polarity. Five exons of the DXS1357E gene were sequenced corresponding to positions 474 to 1315 of the DXS1357E transcript (Fig. 2 and Table 1 ). Based on previous mapping and sequencing of the first exon of the DXS1357E gene, we estimate that the gene spans ~20 kb of sequence.
A second cosmid clone from the chromosome 16 library, c329B6, was sequenced in its entirety (32.5 kb) (GenBank accession #U41302) and compared to the paralogous sequence of u56C4 (GenBank accession #U36341). Bestfit alignment of the two sequences reveals that the duplication between 16p11.1 and Xq28 involves 26.5 kb of genomic DNA, including the entire creatine transporter (SLC6A8) and five exons of the DXS1357E gene (Fig. 2 ). A total of 76 insertions/deletions (indels) and 1122 transitions/transversions were observed within the 26.5 kb region of overlap, resulting in an overall similarity of 94.6% between the two paralogous sequences (Table 2 ). A significantly greater (p <0.0001; Z = 7.2) per cent similarity was observed for the putative coding portions of the 16-X paralogous genes when compared to the overall similarity of the sequences (94.6% vs. 97.1%, respectively) (Table 2 ). Since the chromosome 16 truncated DXS1357E gene was likely non-functional at the time of the duplication event between these two chromosomes, we compared the divergence rate between putative coding segments (CDS) of the DXS1357E and SLC6A8 genes between X and 16. The divergence rate for DXS1357E CDS exons is significantly greater (p <0.05; Z = 1.68) than the divergence rate for SLC6A8 coding exons. In addition, no indels were observed within the CDS of the chromosome 16 paralogous SLC6A8 gene (Fig. 2 ). Predicted amino-acid composition of the 16-creatine transporter paralog, however, does indicate the presence of a `premature' stop codon in exon 4 (TrpTGG>>AmberTGA). Sequence analysis of PCR amplified and subcloned products from three unrelated individuals confirmed the presence of this premature stop codon in the chromosome 16 cosmid.
Table 2 Per cent nucleotide similarity between various portions of the 16p11.1-Xq28 duplicationThe overall % similarity for 26.5 kb of paralogous sequence was calculated as 94.6% (see Materials and Methods). % nucleotide similarity as well as the number of indels and single bp changes among various genic and non-genic portions are compared. The percent similarity of the coding exons of the 19p11.1-Xq28 SLC6A8 gene is significantly greater than that of the compared CDS of the DXS1357E gene (p<0.05). The 5' UTR of the SLC6A8 gene show greater nucleotide divergence than non-genic DNA.
Figure 3 depicts the sequence immediately flanking the boundaries of the 16p11.1 and Xq28 paralogy domain. CAGGG or inverted CCCTG repeats were located near or within all four breakpoints (Fig. 3 a,b). The 3' flanking region (relative to the transcriptional orientation of the Xq28 SLC6A8 gene) of the 16p11.1 domain, in particular, shows a complex arrangement of CAGGG pentamer repeats. A total of 141 direct CAGGG interspersed repeats were observed in 3.1 kb flanking the 16p11.1 duplicated portion, resulting in a density of one repeat every 22.5 bp (Fig. 3 b). Several higher-order repeat structures were also observed within the repeat region, all involving the CAGGG pentamer. For example, 19 CAGGG doublets (CAGGG)2 as well as several hexamer doublets including (CAGGGC)2, (ACAGGG)2, (GCAGGG)2 and (CCAGGG)2 were found in this region. The most complicated higher-order repeat structure identified was a 38-mer which, once again, included two pentamer CAGGG motifs located at either end of the repeat (see Fig. 3 a,c for consensus). CAGGG repeats were also found distributed through the 5' flanking region of the Xq28 paralogy domain, albeit much less frequently (1 CAGGG/CCCTG repeat every 124 bp). In contrast to the 16p11.1 flank, many of the repeats are inverted (CCCTG). Hundreds of imperfect repeats (i.e. those which differ by one bp from the CAGGG/CCCTG consensus) were observed on both the 5'and 3' flanks of the Xq28 and 16p11.1, respectively. With the exception of a few imperfect (CAGGG) repeats located near or within the breakpoints (Fig. 3 a), the reciprocal flanks (the 5' region flanking the 16p11.1 breakpoint and the 3' region flanking the Xq28 boundary) are virtually devoid of these repeat structures. It may be noteworthy that the 3' breakpoint of Xq28 occurs within the cluster of direct Alu repeats (Fig. 2 ).
In order to assess the origin of the duplication event, comparative FISH analysis using both u56C4 (chromosome X cosmid) and c329B6 (chromosome 16 cosmid) as probes was performed on metaphase chromosome spreads from various primate species. With the exception of gibbon, signals from both probes co-localized within each species (Table 3 ; Fig. 4 ). While the Xq28 signals were generally punctate, those generated on the other chromosomes were much more intense and broader in form among all species. In the chimpanzee, c329B6 and u56C4 cohybridized to the syntenic region of Xq28. In addition, signals were observed in the pericentromeric region of chimpanzee chromosomes 19 and 12, which are syntenic to human chromosomes 7 and 2p (Table 3 ). In the gorilla, the duplication event appears to have occurred multiple times, resulting in signals syntenic to human chromosomes X, 7, 2p, 13, 15, 16 and 18. With the exception of the Xq28 locus, all signals were observed within the first cytogenetic band from the centromere (Fig. 4 ). Southern analysis using p96D7 (SLC6A8 cDNA fragment) and p3A1 (DXS1357E cDNA fragment) on genomic zooblots from various primates confirmed that the duplication in the gorilla and chimpanzee had included the creatine transporter and CDM genes, resulting in multiple cross-hybridizing bands and increased intensity of hybridization on some restriction fragments (data not shown). In contrast to human, gorilla and chimpanzee, FISH using c329B6 and u56C4 on orangutan metaphase spreads showed signals only on the distal end of X chromosome, suggesting that the Xq28 represents the ancestral template copy. Gibbon was the only species which demonstrated differential patterns of chromosome hybridization with the two probes. The c329B6 probe hybridized uniquely to gibbon pericentromeric regions chromosomes 8, 15 and 17 (syntenic to human chromosomes 22, 15 and 14). Both the u56C4 and the c329B6 probe, once again, hybridized to the distal end of the X chromosome.
Table 3 Summary of comparative FISH analysisThe FISH chromosomal assignments of the u56C4 and 329b6 probes for each species are summarized. Roman numberals indicate phylogenetic groups, which define syntenic chromosomal portions among these primate species. All non-Xq28 signals were located within the pericentromeric region of each chromosome.
We have surveyed by FISH distantly related groups of the human population (Mbuti Pygmy, Israeli Caucasian, Brazilian Karitiana, Melanesian Nasoi and Auca Indian) in order to assess potential polymorphic variability in the location of the duplication. All five individuals showed strong hybridization of the u56C4 probe to both 16p11.1 and Xq28. The c329B6 probe hybridized intensely to 16p11.1 and more weakly to Xq28 (data not shown). These results indicate that the 16p11.1 and Xq28 duplication has been fixed in the human population for at least 200 000 years. Interestingly, hybridization of c329B6 to 2p11.1 was observed in a few individuals (Caucasian, Karitiana and Pygmy).
The consistent cohybridization of 56c4 and c329B6 probes to the distal end of the X chromosome may be taken as strong evidence that the Xq28 region represents the ancestral locus of the CDM and creatine transporter genes. To determine if this region is syntenic in a more distantly related organism, a panel of murine YACs extending from the Bgn to G6pd locus (~1062 kb) (29 ) was hybridized with human cDNA probes p96D7 and p3A1. Both the creatine transporter and CDM probes hybridized to YACs D19H6 and C176B11, placing both genes in the same 185 kb interval between Bgn and L1cam, at a location at least 100 kb distal to the mouse Bgn locus. The position of the genes, thus, is highly conserved between man and mouse on the X chromosome. Only bands cognate to the Xq28 fragments were observed upon hybridization of the creatine transporter cDNA (p96D7) to mouse genomic Southerns, indicating that other duplications of this region in the mouse are unlikely.
We have identified and characterized an evolutionarily recent and proliferative duplication event involving a gene-rich region of Xq28 and the pericentromeric regions of primate chromosomes. The overall similarity between the human 16p11.1 and Xq28 regions over 26.5 kb of paralogous sequence is 94.6% (Table 1 ). Based on estimates of mutation rate for silent site substitutions (30 ,31 ) and for intronic sequences (32 ) (5 * 10-9-7 * 10-9 mutations per site per year), we calculate that the 16p11.1 and Xq28 may have diverged as recently as 7-10 mya. Most estimates of nucleotide divergence between ancestrally related sequences, however, have been determined by comparisons of orthologous sequences and the time of species separation predicted from paleontological lines of evidence (30 -33 ). Among orthologous gene sequences (including introns and some intergenic segments) (33 ), selection pressure may be operative, limiting the rate of mutation. The recent transfer of genetic material to new chromosomal positions and the different (or lack of) selection pressure which it may experience is clearly distinct from orthologous sequence comparisons. Perhaps more appropriate measures of the rate of divergence within the 16p11.1-Xq28 paralogy domain can be made from comparisons of the [beta]-globin gene and its paralogous pseudogenes. These estimates of divergence which range as high as 13 * 10-9 would predict that the 16p11.1 and Xq28 sequences may have separated as recently as 4 mya. Comparisons of paralogous gene segments located in tandem (globin gene cluster), once again, may not be entirely equivalent to interchromosomal parologies, due to the potential for sequence homogenization by unequal crossing over or gene conversion (11 ,12 ). The sequence characterization of a recent duplication of genes between Xq28 and 16p11.1, thus, should provide researchers with a unique opportunity to study the rates of divergence of DNA between non-homologous chromosomes within the same species.
Interestingly, the rate of divergence is not uniform throughout the entire 16p11.1-Xq28 duplicated region (Table 2 ). The coding segments of 16p11.1 SLC6A8 exons are more highly conserved than the intronic or intergenic sequence (97.1 % vs. 94.4%; Table 2 ). The predicted amino acid sequence of the 16p11.1 creatine transporter paralog, however, identifies a premature stop codon within exon 4 (compare position 8633 of GenBank accession #U41302 with position 1224 of paralogous Xq28 cDNA, GenBank accession L31409). This suggests that the 16p11.1 creatine transporter is non-functional or that it encodes a truncated four trans-membrane domain protein. RT-PCR assays specific for the chromosome 16 SLC6A8 transcript have failed to identify expression from various tissue sources (brain, hippocampus, retina, liver, intestine, lymphoblastoid, fibroblast and melanoma; data not shown). Interestingly the normal stop codon of the Xq28 SLC6A8 transcript is absent in the paralogous 16p11.1 sequence (TGA>>GGA) (compare position 12824 GenBank accession #U36341 with position 11753 GenBank accession #U41302).
In contrast to the SLC6A8 gene, the 16p11.1 DXS1357E `coding' sequences have diverged much more rapidly (95.1% similarity) and have incurred mutations which alter the putative translational reading frame of the protein (when compared to the Xq28 sequence). Since the paralogy domain breaks within an intron of the chromosome 16 DXS1357E sequence (Figs 2 ,3 ), effectively truncating the gene, it is likely that the 16p11.1 DXS1357E paralog was a non-functional pseudogene at the time of the duplication event. This may explain why the divergence of DXS1357E coding exons (4.9%) is greater than the coding exons of the SLC6A8 gene (2.9%). In toto, our investigations suggest that although the DXS1357E gene has been a non-processed (promoter-less) pseudogene since the time of its divergence, the high conservation of the coding exons of the 16p11.1 SLC6A8 suggest that it may have had a functional role at some time during hominoid evolution.
Two other interchromosomal duplications have previously been described involving the 16p11.1-16p11.2 cytogenetic interval; namely, a duplication of the VH segments of the immunoglobulin heavy chain locus from cytogenetic band 14q32.3 to 16p11.2 (37 ) and the identification of a >15 kb paralogy domain between 6p25-pter and 16p11.1 involving the minisatellite locus, [lambda]MS29 (38 ). There are striking similarities between these duplications and the Xq28-16p11.1 paralogy domain. All three examples involve the mobilization of a substantial portion of genomic sequence (15-30 kb) relatively recently (1-10 mya) to the 16p11.1-16p11.2 cytogenetic interval. In each case, the transposition has involved the exchange of genetic material between subtelomeric and pericentromeric loci. In addition, the subtelomeric paralogs (VH segments on 14q32.3, DNF21S1 on 6p25-pter; SLC6A8/DXS1357E region on Xq28) have been deemed antecedent (37 ,38 ). Interestingly, when such events have involved the transposition of genes, the 16p11.1-p11.2 paralogs have generally been found to be transcriptionally silent (37 ). These data, as well as the general intrachromosomal instability of this region (39 ,40 ), suggest that the 16p11.1-11.2 region may be particularly prone to undergo macromutational events resulting in local chromosomal duplications.
We have identified and characterized the sequence at the junctions of the 16p11.1 and Xq28 paralogy domain. The 3' junction (centromere to telomere orientation) of the Xq28 region is located within a cluster of three head-to-tail Alu repeats (Figs 2 and 3 a). A second cluster of Alu repeats was observed 2 kb proximal to the 5' junction of the Xq28 region arranged in opposite orientation to the Alu breakpoint cluster (Fig. 2 ). No other Alu repeats were observed in nearly 28 kb of genomic distance between these two clusters, suggesting a non-random distribution of Alu repeats in the creatine transporter-CDM region (41 ). Rearrangement of Alu clusters, resulting in the formation of local duplications and deletions, has been implicated in the etiology of various human genetic diseases (42 -45 ). Although the mechanism by which these events occurs is largely unknown (43 ), it has been proposed that the sequence homology between Alu repeats facilitates illegitimate recombination events, resulting in disruption of gene structure and concomitant gene dysfunction (46 ). The inverted clusters of Alu repeats at either end of the Xq28 breakpoints may, similarly, provide the sequence homology required to bring the Xq28 breakpoints in close proximity (Fig. 5 ) for intrachromosomal recombination to occur.
We have described a novel interchromosomal paralogy involving 26.5 kb of gene-rich sequence between Xq28 and 16p11.1. Our data indicate that this paralogy has arisen quite recently in human evolution (7-10 mya) and have likely involved duplications of the Xq28 `master' sequence to the pericentromeric regions of chromosomes, possibly through recombinogenesis of CAGGG sequence motifs. These data suggest that processes other than tetraploidization (followed by reestablishment of the disomic state) may be involved in the expansion of gene families in the primate genome. The rapid proliferation of other mammalian `cluster' multigene families which are dispersed near the telomeres and centromeres of chromosomes such as the ZNF (zinc finger) gene family (7 ) and the olfactory receptor cluster (62 ) may be mediated by similar bursts of transposition of short stretches of genomic DNA. Such genome plasticity would facilitate gene diversity, dramatically increasing the adaptive potential of an organism. The proliferation of the mammalian Na+ Cl- dependent transporter gene family, of which SLC6A8 is a member, may be regarded in this light. The conservation of gene structure and their diverse function of this 12 membrane spanning domain gene family in amino-acid, neurotoxin, neurotransmitter, osmolyte and ion transport, may have involved duplication and dispersal from an ancestral template copy. The pericentromeric- directed transposition of the creatine transporters may provide a clue to the mechanism mediating such evolutionary diversity in vertebrate genomes.
An arrayed X-chromosome library, (LLOXNCO1 `U'), was obtained from Lawrence Livermore National Laboratory consisting of 25 000 clones. Clones were grown in a 4 * 4 array format on 16 nylon membrane filters at a density of 1536 clones per filter. Filters were prehybridized for 1 h at 65oC with 0.25 M NaPO4, 0.25 M NaCl, 5% SDS, 10% PEG and 1 mM EDTA; and blocked with 20 [mu]g/ml herring sperm DNA. Random hexamer-generated and end-labelled probes were prepared using manufacturer's specifications (Pharmacia; NEB) and hybridizations were performed as described. End-labelled oligonucleotide and random hexamer-generated probes were purified through G-25 and G-50 Sephadex columns, respectively. All hybridizations were performed overnight at 65oC in a rotisserie oven. Filters were washed three times for 30 min each at 65oC with 0.05 M NaPO4, 0.5% SDS and 1 mM EDTA solution and exposed to autoradiographic film. A second flow-sorted chromosome 16-derived cosmid library (LA16NC02) was obtained from the Los Alamos National Laboratory (63 ). Three filters consisting of 4032 arrayed clones which had been pre-selected during the course of developing a physical map of chromosome 16 were screened with both cDNA and genomic clones. Total cosmid DNA was used as probe to screen an array of PCR-amplified clonal inserts from human placental and heart cDNA libraries in order to isolate corresponding cDNA clones (64 ).
Individual cosmid clones were labelled and hybridized to Southern blots of chromosome 16 and X chromosome somatic cell hybrid deletion panels as described previously (28 ,65 ,66 ). In order to further refine the map location of chromosome 16 cosmid, Southern blots of restriction digest panels of YAC clones which had been STS mapped to the 16p11.1-16p11.2 interval were probed. Due to the reported proximity of the Xq28 CDM gene to the ALD (adrenoleukodystrophy) gene (26 ), corresponding ALD cDNA (H8, 2.6 kb EcoR1 insert; exons 1-9) and genomic (TA19; 3.8 kb TaqI subclone of cosmid Qc11H12; exons 6 to 7) clones, as well as cDNA clones p96D7 and p3A1, were employed to confirm the ALD-BGN (biglycan) map interval location and to extend the cosmid physical map in this region (Fig. 1 ). A Southern blot of EcoR1 restriction digests of five murine YAC clones (D19H6, C176B11, D741C3, B7S6, and H864F2) (29 ) was probed with p96D7 and p3A1 to confirm the comparative map location of the Cdm and creatine transporter genes to the Bgn/L1cam interval of mouse.
Shotgun libraries from two cosmids u56C4 (a 33.023 kb X chromosome insert) and c329B6 (32.505 kb chromosome 16 insert) were prepared in the M13delta vector (67 ) using previously described protocols (68 ). Libraries were constructed using an adaptor-based approach that increases the number of random clones by driving the blunt-ended ligation step with the addition of excess adaptor (67 ). The average insert size of M13 clones was approximately 1.5 kb. M13 clones were picked to inoculate 5 ml cultures and allowed to grow for no more than 7 h with vigorous shaking and aeration. Single stranded DNA was prepared using a PEG precipitation followed by GFC filter capture (69 ). The quantity and quality of DNA isolated was confirmed by gel electrophoresis on 1% agarose with ethidium staining.
Dye primer sequencing was performed using a Biomek 1000 workstation (Beckman Instruments Inc.), PE 9600 thermocycler (Perkin Elmer Cetus) and reagents provided by Applied Biosystems (70 ). A modified asymmetric PCR protocol (71 ,72 ) was used in the directed reverse sequencing phase of the project. Sequencing reaction products were analysed on a ABI 373A DNA Sequencer (Applied Biosystems). The collected data was trimmed using Seqprep software (MBCR, Department of Cell Biology, BCM) and transferred to a UNIX platform for assembly and editing using XDAP software (73 ). Sequence map gaps (SMGs) and the known exon structure from 56C4 (Table 1 ) assisted in confirming assembly of the sequence reads from cosmid c329B6.
A GeneWorks software package (v.2.1, Intelligenetics) was used to analyse the GC composition and to generate a detailed restriction map of the XDAP consensus assembly of both cosmids. The locations of putative exons in the sequence was determined with the GRAIL 2 gene-recognition tool (74 ). A comparison of genomic (GenBank accession U36341 and U41302) and cDNA sequences (GenBank accession L31409 and GenBank Z31696) identified the position and size of DXS1357E and SLC6A8 exons (23 ,25 ). The location and identity of Alu repetitive elements was determined using the PYTHIA search algorithm (75 ). Internal repetitive sequence motifs were identified and characterized using the PRINTREPEATS program. BestFit alignment (GCG Software package) was used to compare the 26.5 kb of paralogous sequence between the two cosmids.
In order to assess potential human diversity of the 16-X duplication, bicolor FISH was performed on metaphase chromosomal preparations from five lymphoblastoid cell lines from five diverse human populations: GM10493 (African Mbuti pygmy), GM11523 (Israeli Jew from Galilee), GM10965 (Brazilian Karitiana), GM10540 (Nasoi-speaking Melanesian) and GM11776 (Auca Indian tribe Waorani). In addition, chromosome metaphase spreads were prepared from human peripheral blood lymphocytes of a female donor or from lymphoblastoid cell lines derived from various species [chimpanzee (Pan troglodytes), gorilla (Gorilla gorilla), orangutan (Pongo pygmaeus) and gibbon (Hylobates lar)]. Cosmid probes were labelled by nick-translation and hybridized to chromosomal preparations as previously described (76 ). High copy repeats were suppressed by reannealing the probe with 20-fold excess of human Cot 1 DNA (Life Technologies) prior to hybridization. Bicolor FISH was performed in which the X chromosome cosmid, u56C4, was biotin-labelled and detected using fluorescein isothiocyanate (FITC)-conjugated avidin (5 [mu]g/ml) (Vector laboratories); and the chromosome 16 cosmid, c329B6, was digoxigenin-labelled and detected using a rhodamine-conjugated anti-digoxigenin antibody (Boehringer Mannheim). 4'6-diamidino-2-phenyllindole (DAPI) staining was simultaneously performed to confirm chromosomal cytogenetic band location.
We would like to thank C.-O. Sarde, J.-L. Mandel, C. C. Lee, A. Chaterjee and G. Herman for providing many of the Xq28 reagents from the ALD-BGN interval. We are grateful to N. Archidiacono, E. Lindsay and D. Muzny for excellent technical assistance. The chromosome specific gene libraries LLOXNC01 and LA16NC02 used in this work were constructed at the Biomedical Sciences Division, Lawrence Livermore National Laboratory, Livermore CA 94550 and the Center for Human Genome Studies, Los Alamos National Laboratory, Los Alamos, NM 87545; respectively, under the auspices of the National Laboratory Gene Library Project sponsored by the US Department of Energy. This research was supported by US Dept. of Energy grant (94ER61830) to DLN, and by grants from NIH/NCHGR (HG00210 to DLN and HG00823 to RAG).
1 Melnick, L. and Sherman, F. (1993) The gene clusters ARC and COR on chromosomes 5 and 10, respectively, of Saccharomyces cerevisiae share a common ancestry. J. Mol. Biol.233, 327-388.
2 Hankeln, T. and Schmidt, E. (1993) Divergent evolution of an `orphan' histone gene cluster in Chironomus. J. Mol. Biol.234, 1301-7.MEDLINE Abstract
3 Arnold, N., Wienberg, J., Emert, K. and Zachau, H. (1995) Comparative mapping of DNA probes from the Vk immunoglobulin gene regions on human and great ape chromosomes by fluorescence in situ hybridization. Genomics26, 147-56.MEDLINE Abstract
4 Garcia-Fernandez, J. and Holland, P. (1994) Archetypal organization of the amphioxus Hox gene cluster. Nature370, 563-6.MEDLINE Abstract
5 Teglund, S., Olsen, A., Khan, W., Frangsmyr, L. and Hammarstrom, S. (1994) The pregnancy-specific glycoprotein (PSG) gene cluster on human chromosome 19: fine structure of the 11 PSG genes and identification of 6 new genes forming a third subgroup within the carcinoembryonic antigen (CEA) family. Genomics23, 669-84.MEDLINE Abstract
6 Ben-Arie, N., Lancet, D., Taylor, C., Khen, M., Walker, N., Ledbetter, D., Carrozzo, R., Patel, K., Sheer, D., Lehrach, H. and North, M. (1993) Olfactory receptor gene cluster on human chromosome 17: Possible duplication of an ancestral repertoire. Hum. Mol. Genet.3, 229-35.
7 Bellefroid, E., Marine, J.-C., Ried, T., Lecocq, P., Riviere, M., Amemiya, C., Poncelet, D., Coulie, P., deJong, P., Szpirer, C., Ward, D. and Martial, J. (1993) Clustered organization of homologous KRAB zinc-finger genes with enhanced expression in human T lymophoid cells. EMBO J.12, 1363-74.MEDLINE Abstract
8 Muscatelli, F., Walker, A., De Plaen, E., Stafford, A. and Monaco, A. (1995) Isolation and characterization of a MAGE gene family in the Xp21.3 region. Proc. Natl Acad. Sci. USA92, 4987-91.MEDLINE Abstract
9 McGinnis, W. and Krumlauf, R. (1992) Homeobox genes and axial patterning. Cell68, 283-302.MEDLINE Abstract
10 Kappen, C., Schughart, K. and Ruddle, F. (1989) Organization and expression of homeobox genes in mouse and man. Ann. NY Acad. Sci.567, 243-52.MEDLINE Abstract
11 Dover, G. (1982) Molecular drive: a cohesive mode of species evolution. Nature299, 111-17.MEDLINE Abstract
12 Dover, G. (1988) DNA turnover and the molecular clock. J. Mol. Evol.26, 47-58.
13 Bird, A. (1995) Gene number, noise reduction and biological complexity. Trends Genet.11, 14-21.
14 Chan, S., Cao, Q. and Steiner, D. (1990) Evolution of the insulin superfamily: cloning of a hybrid insulin/insulin-like growth factor cDNA from amphioxus. Proc. Natl Acad. Sci.87, 9319-823.MEDLINE Abstract
15 Ohno, S., Wolf, U. and Atkin, N. (1968) Evolution from fish to mammals by gene duplication. Hereditas59, 169-87.MEDLINE Abstract
16 Ohno, S. (1970) Evolution by Gene Duplication. Springer Verlag: Berlin,Heidelberg,New York.
17 Sidow, A. (1992) Diversification of the wnt gene family on the ancestral lineage of vertebrates. Proc. Natl Acad. Sci. USA89, 5098-102.MEDLINE Abstract
18 Lundin, L. (1993) Evolution of the vertebrate genome as reflected in paralogous chromosomal regions in man and the house mouse. Genomics16, 1-19.MEDLINE Abstract
19 Bailey, G., Poulter, R. and Stockwell, P. (1978) Gene duplication in tetraploid fish: Model for gene silencing at unlinked duplicated loci. Proc. Natl Acad. Sci. USA75, 5575-9.MEDLINE Abstract
20 Ferris, S. and Whitt, G. (1979) Evolution of the differential regulation of duplicate genes after polyploidization. J. Mol. Evol.12, 267-317.MEDLINE Abstract
21 Jeffreys, A., Wilson, V., Blanchetot, A., Weller, P., Geurts van Kessel, A., Spurr, N., Solomon, E. and Goodfellow, P. (1984) The human myoglobin gene: a third dispersed globin locus in the human genome. Nucleic Acids Res.12, 3235-43.MEDLINE Abstract
22 Grzeschik, K. and Kazazian, H. (1985) Report of the committee on the genetic constitution of chromosomes 10, 11 and 12. Cytogenet. Cell Genet.40, 179-205.MEDLINE Abstract
23 Nash, S., Giros, B., Kingsmore, S., Rochelle, J., Suter, S., Gregor, P., Seldin, M. and Caron, M. (1994) Cloning, pharmacological characterization and genomic localization of the human creatine transporter. Receptors and Channels2, 164-174.
24 Sora, I., Richman, J., Santoro, G., Wei, H., Wang, Y., Vanderah, T., Horvath, R., Nguyen, M., Waite, S., Roeske, W. and Yamamura, H. (1994) The cloning and expression of a human creatine transporter. Biochem. Biophys. Res. Comm.204, 419-27.MEDLINE Abstract
25 Mosser, J., Sarde, C.-O., Vicaire, S., Yates, J. and Mandel, J.-L. (1994) A new human gene (DXS1357E) with ubiquitous expression, located in Xq28 adjacent to the adrenoleukodystrophy gene. Genomics22, 469-71.MEDLINE Abstract
26 Sarde, C.-O., Mosser, J., Koschis, P., Kretz, C., Vicaire, S., Aubourg, P., Poustka, A. and Mandel, J.-L. (1994) Genomic organization of the adrenoleukodystrophy gene. Genomics22, 13-20.MEDLINE Abstract
27 Nelson, D., Ballabio, A., Cremers, F., Monaco, A. and Schlessinger, D. (1995) In Nelson, D., Ballabio, A., Cremers, F., Monaco, A. and Schlessinger, D., Banff, A. B. (eds), Report of the sixth international workshop on X chromosome mapping 1995.
28 Stallings, R., Doggett, N., Callen, D., Apostolou, S., Chen, L., Nancarrow, J., Whitmore, S., Harris, P., Michison, H., Brauning, M., Sarich, J., Fickett, J., Cintosky, M., Sorenson, D., Torney, D., Hildebrand, C. and Moyzis, R. (1992) Evaluation of a cosmid contig physical map of human chromosome 16. Genomics13, 1031-9.MEDLINE Abstract
29 Chatterjee, A., Faust, C., Molinari-Storey, L., Kiochis, P., Poustka, A. and Herman, G. (1994) A 2.3-Mb yeast artificial chromosome contig spanning from Gabra3 to G5pd on the mouse X chromosome. Genomics21, 49-57.MEDLINE Abstract
30 Hayashida, H. and Miyata, T. (1983) Unusual evolutionary conservation and frequent DNA segment exchange in class I genes of the major histocompatability complex. Proc. Natl Acad. Sci. USA80, 2671-2675.MEDLINE Abstract
31 Efstratiadis, A., Posakony, J., Maniatis, T., Lawn, R., O'Connell, C., Spritz, R., DeRiel, J., Forget, B., Weissman, S., Slightom, J., Blechl, A., Smithies, O., Bralle, F., Shoulders, C. and Proudfoot, N. (1980) The structure and evolution of the Human beta-globin gene family. Cell21, 653-68.MEDLINE Abstract
32 Perler, F., Efstratiadis, A., Lomedico, P., Gilbert, W., Kolodner, R. and Dodgson, J. (1980) The evolution of genes: the chicken preproinsulin gene. Cell20, 555-65.MEDLINE Abstract
33 Margot, J. B., Demers, G. W. and Hardison, R. C. (1989) Complete nucleotide sequence of the rabbit beta-like globin gene cluster. J. Mol. Biol.205, 15-40.MEDLINE Abstract
34 ISCN (1985) Report of the standing committee on human cytogenetic nomenclature. Birth Defects21, 1-117.
35 Horai, S., Hayasaka, K., Kondo, R., Tsugane, K. and Takahata, N. (1995) Recent African origin of modern humans revealed by complete sequences of hominoid mitochondrial DNAs. Proc. Natl Acad. Sci. USA92, 532-6.MEDLINE Abstract
36 Bowcock, A. M., Ruiz-Linares, A., Minch, E., Kidd, J. R. and Cavalli-Sforza, L. L. (1994) High resolution of human evolutionary trees with polymorphic microsatellites. Nature368, 455-7.MEDLINE Abstract
37 Tomlinson, I., Cook, G., Carter, N., Elaswarapu, R., Smith, S., Walter, G., Buluwela, L., Rabbitts, T. and Winter, G. (1994) Human immunglobulin VH and D segments on chromosomes 15q11.2 and 16p11.2. Hum. Mol. Genet.3, 853-60.MEDLINE Abstract
38 Wong, Z., Royle, N. and Jeffreys, A. (1990) A novel human DNA polymorphism resulting from transfer of DNA from chromosome 6 to chromosome 16. Genomics7, 222-234.MEDLINE Abstract
39 Dauwerse, J., Jumelet, E., Wessels, J., Saris, J., Hagemeijer, A., Beverstock, G., van Ommen, G. and Breuning, M. (1992) Extensive cross-homology between the long and the short arm of chromosome16 may explain leukemic inversions and translocations. Blood79, 1299-304.MEDLINE Abstract
40 Stallings, R., Doggett, N., Okumura, K. and Ward, D. (1992) Chromosome 16-specific repetitive DNA sequences that map to chromosomal regions known to undergo breakage/rearrangement in leukemia cells. Genomics7, 332-8.
41 Jurka, J. and Batzer, M. (1995) In Meyers, R. (ed.) Human repetitive elements (in press).
42 Marcus, S., Hellgren, D., Lambert, B., Fallstrom, S. and Wahlstrom, J. (1993) Duplication in the hypoxanthine phosphoribosyl-transferase gene caused by Alu-Alu recombination in a patient with Lesch Nyhan syndrome. Hum. Genet.90, 477-82.MEDLINE Abstract
43 Heikinnen, J., Hautala, T., Kivirikko, K. and Myllyla, R. (1994) Structure and expression of the human lysyl hydroxylase gene (PLOD): Introns 9 and 16 contain Alu sequences at the sites of recombination in Ehlers-Danlos syndrome Type VI patients. Genomics24, 464-71.
44 Markert, M., Hutton, J., Wiginton, D., States, J. and Kaufman, R. (1988) Adenosine deaminase (ADA) deficiency due to deletion of the ADA gene promoter and first exon by homologous recombination between two Alu elements. J. Clin. Invest. 81, 1323-7.MEDLINE Abstract
45 Nicholls, R., Fischel-Ghodsian, N. and Higgs, D. (1987) Recombination at the human alpha-globin gene cluster: sequence features and toplogical constraints. Cell49, 369-78.MEDLINE Abstract
46 Morris, T. and Thacker, J. (1993) Formation of large deletions by illegitimate recombination in the HPRT gene of primary human fibroblasts. Proc. Natl Acad. Sci. USA90, 1392-6.MEDLINE Abstract
47 Davis, M., Kim, S. and Hood, L. (1980) DNA sequences mediating class switching in alpha-immunoglobulins. Science209, 1360-5.MEDLINE Abstract
48 Hengstschläger, M., Maizels, M. and Leung, H. (1995) Targeting and regulation of immunoglobulin gene somatic hypermutation and isotype switch recombination. Prog. Nucleic Acids Res.50, 67-99.
49 Vogt, P. (1990) Potential genetic functions of tandem repeated DNA sequence blocks in the human genome are based on a highly conserved `chromatin folding code'. Hum. Genet.84, 301-36.MEDLINE Abstract
50 Arakawa, H., Iwasato, T., Hayashida, H., Shimizu, A., Honjo, T. and Yamagishi, H. (1993) The complete murine immunglobulin class switch region of the alpha heavy chain gene-hierarchic repetitive structure and recombination breakpoints. J. Biol. Chem.268, 4651-5.MEDLINE Abstract
51 Iwasato, T., Shimizu, A., Honjo, T. and Yamagishi, T. (1990) Circular DNA is excised by immunoglobulin class switch recombination. Cell62, 143-9.MEDLINE Abstract
52 Matsuoka, M., Yoshida, K., Maeda, T., Usuda, S. and Sakano, H. (1990) Switch circular DNA formed in cytokine-treated mouse splenocytes: Evidence for intramolecular DNA deletion in immunoglobulin class switching. Cell62, 135-42.MEDLINE Abstract
53 Wahls, W., Wallace, L. and Moore, P. (1989) Hypervariable minisatellite DNA is a hotspot for homologous recombination in human cells. Cell, 95-103.
54 Jeffreys, A., Neumann, R. and Wilson, V. (1990) Repeat unit sequence variation in minisatellites: a novel source of DNA polymorphism for studying variation and mutation. Cell60, 473-85.MEDLINE Abstract
55 Krowczynska, A., Rudders, R. and Krontiris, G. (1990) The human minisatellite consensu at breakpoints of oncogene translocations. Nucleic Acids Res.18, 1121-7.MEDLINE Abstract
56 Gibbs, M., Collick, A., Kelly, R. G. and Jeffreys, A. J. (1993) A tetranucleotide repeat mouse minisatellite dsplaying substantial somatic instability during early preimplantation development. Genomics17, 121-8.MEDLINE Abstract
57 Kelly, R., Bulfield, G., Collick, A., Gibbs, M. and Jeffreys, A. (1989) Characterization of a highly unstable mouse minisatellite locus: Evidence for somatic mutation during early development. Genomics5, 844-56.MEDLINE Abstract
58 Sundquist, W. and Klug, A. (1989) Telomeric DNA dimerizes by formation of guanine tetrads between hairpin loops. Nature342, 825-9.MEDLINE Abstract
59 Sen, D. and Gilbert, W. (1988) Formation of parallel four-stranded complexes by guanine-rich motifs in DNA and its implications in meiosis. Nature334, 364-6.MEDLINE Abstract
60 Sen, D. and Gilbert, W. (1990) A sodium-potassium switch in the formation of four-stranded G4-DNA. Nature344, 410-14.MEDLINE Abstract
61 Lowe, L., Heerema, N., Cheerva, A. and Palmer, C. (1992) A new nonrandom chromosomal abnormality, t(2:16)(p11.2;p11.2), possibly associated with poor outcome in childhood acute lymphoblastic leukemia. Cancer Genet. Cytogenet.54, 60-4.
62 Trask, B., Massa, H., Evans, J., Scherer, S., Friedman, C., Youngblom, J., Rouquier, S., Giorgi, D., Martin-Gallardo, A., Wong, D., Iadonato, S., Yokota, H., van den Engh, G., Hearst, J. and Sachs, R. (1995) In Bentley, D., Green, E. and Warterston, R. (eds), Applications of FISH in genome analysis. Cold Spring Harbor Press: Cold Spring Harbour, NY, p. 14.
63 Longmire, J., Brown, N., Meincke, L., Campbell, M., Albright, K., Fawcett, J., Campbell, E., Moyzis, R., Hildebrand, C., Evans, G. and Deaven, L. (1993) Construction and characterization of partial digest DNA libraries made from flow-sorted human chromosome 16. GATA10, 69-76.
64 Lee, C., Yazdani, A., Wehnert, M., Zhao, Z., Lindsay, E., Bailey, J., Coolbaugh, M., Couch, L., Xiong, M., Chinault, A. et al. (1995) Isolation of chromosome-specific genes by reciprocal probing of arrayed cDNA and cosmid libraries. Hum. Mol. Genet.4, 1373-80.MEDLINE Abstract
65 Parrish, J. E., Oostra, B. A., Verkerk, A. J. M. H., Richards, C. S., Reynolds, J., Spikes, A. S., Shaffer, L. G. and Nelson, D. L. (1994) Isolation of a GCC repeat showing expansion in FRAXF, a fragile site distal to FRAXA and FRAXE. Nature Genet.8, 229-35.MEDLINE Abstract
66 Stallings, R., Torney, D., Hildebrand, C., Longmire, J., Deaven, L., Jett, J., Doggett, N. and Moyzis, R. (1990) Physical mapping of human chromosomes by repetitive sequence fingerprinting. Proc. Natl Acad. Sci.87, 6218-22.MEDLINE Abstract
67 Povinelli, C. M. and Gibbs, R. A. (1993) Large-scale sequencing library production: an adaptor-based strategy. Analyt. Biochem.210, 16-26.MEDLINE Abstract
68 Bankier, A. T., T., W. K. and Barrel, B. G. (1987) Random cloning and sequencing by the M13 dideoxynucleotide termination melthod. Meth. Enzym.155, 51-93.MEDLINE Abstract
69 Kristensen, T., Voss, H. and Ansorge, W. (1987) A simple and rapid preparation of M13 templates for manual and automated dideoxy sequencing. Nucleic Acids Res.15, 5507-16.MEDLINE Abstract
70 Civitello, A. B., Richards, S. and Gibbs, R. A. (1992) A simple protocol for the automation of DNA cycle sequencing reactions and polymerase chain reactions. J. DNA Seq. Map.3, 17-23.
71 Munzy, D., Richards, S., Shen, Y. and Gibbs, R. (1993) In Venter, C. (ed.) PCR based strategies for gap closure in large scale sequencing projects. Harcourt, Brace Jovanovich: London.
72 Wilson, R., Chen, C. and Hood, L. (1990) Optimization of asymmetric polymerase chain reaction for rapid fluorescent DNA sequencing. BioTechniques8, 184-9.MEDLINE Abstract
73 Dear, S. and Staten, R. (1991) A sequence assembly and editing program for efficient management of large projects. Nucleic Acids Res.19, 3907-11.MEDLINE Abstract
74 Uberbacher, E. and Mural, R. (1991) Locating protein-coding regions in human DNA sequences by multiple sensor-neural network approach. Proc. Natl Acad. Sci. USA88, 11261-5.MEDLINE Abstract
75 Milosavljevic, A. (1993) In Hunter, L., Shavlik, J. and Searls, D. (eds) Discovering sequence similarity by the Algorithmic significance method. AAAI Press.
76 Baldini, A., Miller, D., Shridhar, V., Rocchi, M., Miller, O. and Ward, D. (1991) Comparative mapping of a gorilla-derived alpha satellite DNA on great ape and human chromosomes. Chromosoma101, 109-14.MEDLINE Abstract
*To whom correspondence should be addressed[dagger]Present address: Human Genome Center, Biology and Biotechnology Research Program, L-452, Lawrence Livermore National Laboratory, Livermore, CA 94551, USA
This page is maintained by OUP admin. Last updated Thu Oct 31 15:25:15 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
