| Human Molecular Genetics | Pages |
©1999 Oxford University Press |
Extensive gene order differences within regions of conserved synteny between the Fugu and human genomes: implications for chromosomal evolution and the cloning of disease genes
Introduction
Results
Identification of genes linked to the Surfeit genes in the Fugu genome
Mapping and ordering the chromosome 9 genes by PCR analysis of somatic cell and radiation hybrids
Isolation of human cosmids containing the 9q34 genes
Additional mapping by fluorescence in situhybridization (FISH)
Discussion
Materials And Methods
General procedures for DNA analysis
Generation of the enriched Fugu cDNA library
PCR analysis of panels of human-rodent somatic cell hybrids and radiation hybrids
Isolation of human chromosome 9-specific cosmid clones
Mapping and ordering of genes by FISH
Acknowledgements
References
Extensive gene order differences within regions of conserved synteny between the Fugu and human genomes: implications for chromosomal evolution and the cloning of disease genes
Received March 3, 1999; Revised and Accepted April 26, 1999
The suitability of the Fugu genome to facilitate the identification of candidate human disease genes using comparative positional cloning is dependent upon the extent to which synteny and gene order are conserved between the two species. We have cloned seven Fugu genes which are closely linked to Surfeit genes in two regions of the Fugu genome and have mapped and ordered their human homologues both by PCR analysis of the Genebridge 4 panel of radiation hybrids and by fluorescence in situ hybridization. All seven human genes map to a 3 Mb region of chromosome band 9q34.1, ~2-4 Mb proximal to the human Surfeit genes. Although both Fugu regions are syntenic with human chromosome band 9q34, the relative order of the genes differs greatly in the two species. Indeed, some of the genes that are adjacent in the Fugu genome are separated by at least 2-4 Mb in the human genome. This suggests that intra-chromosomal rearrangements, most probably inversions, have been common during the 900 million years of divergent evolution separating Fugu and human. The utility of Fugu to facilitate human disease gene identification by comparative positional cloning is questioned in light of these results.
INTRODUCTION
The genome of the Japanese puffer fish, Fugu rubripes (Fugu), is 7.5 times smaller than the human genome but is believed to possess a similar complement of genes (1). The relative sparsity of dispersed repetitive DNA elements, the generally much reduced size of introns and the high gene density predicted for this compact vertebrate genome greatly facilitate the isolation and analysis of genes. For these reasons, and because gene structure is largely conserved between Fugu and mammals, Fugu possesses a model genome for the study of vertebrate gene structure (2-9). Comparative analysis between Fugu and other vertebrates has also been useful in identifying important conserved gene regulatory elements (10-13).
Another potentially exciting role for the Fugu genome is its suggested use in facilitating the identification of human disease genes by comparative positional cloning (14-18). Only if significant regions of conserved synteny and conserved gene linkage (specifically gene order) are shown to exist between the Fugu and human genomes would such a role be effective. In this context, several examples of conserved synteny between the genomes of Fugu and human have been reported (12,16-25), but these often involve only limited numbers of genes. The extent to which short-range gene linkage relationships are also conserved is less clear. Whilst there are a number of confirmed examples of pairs of genes being adjacent in both human and Fugu (19-23), there are only three examples of conserved gene order spanning three or more genes. The first involves the Hox gene clusters (24) but could be considered exceptional since the ordered clustering of Hox genes is critically related to their patterns of expression. The two other examples involve FOSand two other genes (S3iii125 and S20i15)which encompass 600 kb at human chromosome band 14q24.3 but only 12.4 kb in Fugu (17), and the WT1, RCN1, PAX6 and two other putative genes which encompass 1.5 Mb in human but <100 kb in Fugu (18). In these cases, gene order may have been conserved in the absence of any selection pressure although the presence of genes in the comparatively large intergenic regions between some of the human genes in both of these studies (which would alter their comparative gene linkage relationships) cannot be ruled out at this time.
Previously, we compared the linkage relationships of the Fugu Surfeit genes with those of the mammalian Surfeit genes (whose chromosomal location and relative order were already known) and found extensive gene order differences (20). In this work, using the reverse approach, we firstly have identified seven Fugu genes as being linked to Surfeit genes in two regions of the Fugu genome and subsequently have mapped and ordered their human homologues in order to evaluate further the extent to which synteny and gene order are conserved. The comparative mapping data obtained indicate that, although the two Fugu regions represent regions of conserved synteny with human chromosome band 9q34, the relative order of genes within these regions differs extensively (several of the genes that are adjacent in Fugu are separated by at least 2-4 Mb in human). Our findings have implications both for the use of the Fugu genome to accelerate the isolation of disease genes using comparative positional cloning strategies and also for our understanding of the genetic events that have shuffled groups of genes during the evolution of vertebrate chromosomes.
RESULTS
Identification of genes linked to the Surfeit genes in the Fugu genome
Two genes, an ASS gene homologue and a gene homologous to human EST00098, have been identified previously as neighbours of the Fugu Surf-4 and Surf-2genes in the Fugu genome (Fig. 1) (20). Additional genes linked to some of the Fugu Surfeit genes have been identified using a cDNA selection technique (see Materials and Methods). Forty-two of 75 cDNAs selected by this technique were found to be either Fugu Surf-1, Surf-3/rpL7a, Surf-4, Surf-5 or ASS cDNAs. A homology search (BLAST) of the MRC Human Genome Mapping Project Resource Centre (HGMP-RC) Fugu sequence database (http://fugu.hgmp.mrc.ac.uk/ ) revealed that a further seven cDNAs contained repetitive sequences. Based on sequence overlap, the remaining 26 cDNAs were assembled into 12 groups. Southern blot analysis and sequencing of Fugu cosmid subclones revealed that all of these groups of cDNAs share sequence identity with one of the template cosmids which was found to be interrupted by the presence of probable introns in most cases. Homology searches (BLAST) of the public databases revealed that three of these 12 groups are homologous to known genes and/or human expressed sequence tags (ESTs) which theoretically represent putative human genes.
Figure 1. Organization of the three Fugu genomic regions containing Surfeit genes. The relative order, orientation and intron-exon structure of the genes located in the Fugu genomic regions containing the Surf-1, Surf-3/rpL7a and Surf-6 genes (A), the Surf-2 and Surf-4 genes (B) and the Surf-5 gene (C) are shown. In each panel, the continuous bold line represents genomic DNA, arrows indicate the direction of transcription of the genes and exons are shown as empty boxes. Dashed lines represent regions of genes where exon structure has not been determined. Selected restriction sites are shown and are labelled Xb, XbaI; Xh, XhoI; S, SpeI; and C, ClaI (only those unaffected by dam methylation are shown). Representative cosmid clones spanning each region are shown below each locus (7). A scale bar is shown. The positions and transcriptional orientations of genes were determined using restriction mapping, Southern blot analysis and DNA sequence contig formation. Putative genes that are only specified by cDNAs and are not homologous to human sequences have been found between the SIAT3C and LSFR2 genes and the Surf-3/rpL7a and LSFR3 genes but are not shown. All other putative Fugu genes isolated in this study map outside these regions. Southern blot analysis of Fugu genomic DNA has confirmed that the relative organization of these Fugu genes within their respective cosmids is identical to their true genomic organization (data not shown).
Members of the first group are derived from a gene on cosmid 186H17 (Fig. 1A) which is highly homologous to a Rattus norvegicus cDNA encoding [alpha]N-acetylgalactosamine [alpha]2,6-sialyltransferase (ST6GalNAc III) (26). It is also homologous to >50 overlapping human ESTs, presumably derived from the putative human homologue of this gene which provisionally has been named SIAT3C (representative ESTs are listed in Table 1). Analysis of cosmid 186H17 has enabled us to identify the last three coding exons of the Fugu SIAT3C gene which conceptually encode the last 234 amino acids of the protein (the rat protein is 305 amino acids). The Fugu SIAT3C gene is located ~12 kb 3[prime] to the Fugu Surf-6 gene (Fig. 1A).
Table 1. Representative ESTs and cosmids for the putative human homologues of Fugu genes identified in this study
| Gene | EST and THCa IDs | LL09NC01 cosmid clonesb | |||
| SIAT3C | aa31h05.s1 | THC173025 | P71G9 | P71H10 | P71H12 |
| THC138050 | P116F7 | P135F7 | 256H3 | ||
| 265B6 | |||||
| LSFR1 | zv66h02.r1 | THC175796 | P23E3 | P86B10 | P112E8 |
| zv66h02.s1 | THC203189 | P112H5 | P132C4 | P137E4 | |
| zx10g06.r1 | P203B9 | P266C7 | P282G12 | ||
| no40g02.s1 | P283B10 | ||||
| LSFR2 | EST06999 | P69F2 | P117H7 | P118B4 | |
| P123C2 | P140B9 | P156E8 | |||
| P273F11 | P275H10 | P292A11 | |||
| P296G7 | |||||
| LSFR3A | H23D05 | THC115824 | P4H6 | P59C6 | P112E6 |
| yo78g08.r1 | P112E7 | P112G9 | P104H6 | ||
| imbb-est107 | P166G8 | P172G9 | P176B10 | ||
| P177D2 | P231B11 | P240H8 | |||
| P246G8 | P247F4 | P281C8 | |||
| LSFR3B | zv22h05.r1 | THC86730 | N/A | ||
| zr99e08.r1 | THC206404 | ||||
| zr99e08.s1 | |||||
| CCBL1 | N/A | P29E2 | P49F11 | P66C1 | |
| P66E6 | P103E6 | P163F2 | |||
| P171G11 | P186F1 | ||||
| EST00098 | zo70d08.s1 | THC170443 | P65A5 | P65A7 | P76E12 |
| zt60h12.s1 | P86B10 | P96A10 | P132C4 | ||
| P203B9 | 266C7 | ||||
| ASS | N/A | P131D6 | P237B10 |
The two non-overlapping members of the second group are derived from a Fugu gene, LSFR1 (linked to Surfeit genes in Fugu rubripes [num]1), located on cosmid 139G11 (Fig. 1B). LSFR1 displays homology to >50 overlapping human ESTs (Table 1). By comparing the homologous cosmid 139G11 and human EST sequences, we have been able to identify the last eight coding exons of the Fugu LSFR1 gene (Fig. 1B). The Fugu LSFR1 and EST00098 genes are separated by <2 kb (Fig. 1B).
The sole member of the third group is derived from a Fugu gene, LSFR2, located at one end of cosmid 186H17 (Fig. 1A), and is homologous to only a single human EST (Table 1). Comparison between cosmid 186H17 sequences and this EST has enabled us to identify the last three coding exons of the Fugu gene which is separated from the Fugu SIAT3C gene by <10 kb (Fig. 1A).
Two additional Fugu genes homologous to known human genes or ESTs were identified by direct sequencing of cosmid subclones. A Fugu homologue of the human gene (CCBL1) encoding human kidney cysteine conjugate [beta]-lyase (27) was identified ~1 kb 3[prime] to the Fugu Surf-6 gene and ~8 kb from the SIAT3C gene on cosmid 186H17 (Fig. 1A). The Fugu CCBL1 gene spans 2.95 kb and possesses 12 coding exons. The intron-exon structure of the human gene has not been determined. The second gene, LSFR3, was identified 5-6 kb 5[prime] to the Fugu Surf-3/rpL7a gene (Fig. 1A) through its homology to two distinct groups of human ESTs. Although the two groups of ESTs are unrelated at the DNA level, they nevertheless are predicted to encode two highly homologous polypeptides, suggesting that they are derived from two paralogous human genes, LSFR3A and LSFR3B (Table 1). So far eleven 3[prime] coding exons of the Fugu LSFR3 gene have been identified by comparison with the human LSFR3A and LSFR3B ESTs.
Mapping and ordering the chromosome 9 genes by PCR analysis of somatic cell and radiation hybrids
The human Surfeit genes and ASS have previously been mapped to chromosome band 9q34, and CCBL1 and EST00098 have previously been mapped to human chromosome 9 (27-29). PCR analysis of a panel of human mono-chromosomal somatic cell hybrid DNAs (30) was now used to determine on which chromosome the putative human homologues of the other Fugu genes identified in this study are located (see Materials and Methods). PCR primer pairs for the SIAT3C, LSFR1, LSFR2 and LSFR3A genes amplified a human-specific product of the expected size from only the DNA of the hybrid cell line containing human chromosome 9 and those for the LSFR3B gene only from the DNA of the hybrid cell line containing human chromosome 1. The human LSFR3B gene was not analysed further.
A PCR analysis was next performed on the HGMP-RC subset of the Genebridge 4 radiation hybrid (RH) panel (31) to determine a more precise chromosomal localization and order of these genes (see Materials and Methods). CCBL1, EST00098, SIAT3C, LSFR1, LSFR2 and LSFR3A were fitted between markers D9S1721 and D9S1148 in a region spanning 0.3-10.3 centiRays (cR) proximal to the ASS gene (Fig. 2). The human Surf-6 gene (SURF6)is predicted by this method to map some 13.7 cR (~4 Mb) distal to ASS(Fig. 2). Two pairs of genes/putative genes, i.e. CCBL1 and LSFR2, and LSFR1 and EST00098, could not be separated by this analysis (Fig. 2).
Figure 2. The relative order and distances in centiRays (cR) separating human chromosome band 9q34 genes as determined by PCR analysis of the Genebridge 4 RH panel. All genes were fitted to the map with a lod score >3.0. The relationship of the genes to markers in the Whitehead framework map, their relative orders and the relative distances are shown on the left, and the Genebridge 4 typing data vector for each of the genes is shown on the right (0 = a negative PCR result, 1 = a positive PCR result and 2 = an ambiguous PCR result). For chromosome 9, it is predicted that 1 cR correlates to ~300 kb (31). The LSFR1 and EST00098 genes and the CCBL1 and LSFR2 genes produced typing vector data that could not be given different map positions (represented by a 0.0 cR separating distance).
Isolation of human cosmids containing the 9q34 genes
The human chromosome 9-specific cosmid library LL09NC01 `P' was screened with PCR products (see above) specific to the known or putative 9q34 genes identified in this study (see Materials and Methods). LL09NC01 cosmids positive for each probe/gene are listed in Table 1. Southern blot analysis, subcloning and sequencing of representative cosmids subsequently enabled us to determine the intron-exon structure for part of each gene. So far, the last three coding exons of the human CCBL1 gene, two coding exons of the human EST00098 gene, the last two coding exons of the human SIAT3C gene, the last eight coding exons of the human LSFR1 gene, the last two coding exons of the human LSFR2 gene and the last eight coding exons of the human LSFR3A gene have been partially or completely identified. The boundaries of these exons are conserved between Fugu and human, indicating that they are likely to be orthologues of the corresponding Fugu genes. Although we have identified only the first three and last two coding exons of the Fugu ASS gene, the intron-exon boundaries for these exons are in positions identical to the corresponding exons (exons 3, 4, 5, 15 and 16) of the mouse and human ASS genes (32,33).
Four of the LL09NC01 cosmid clones containing the human EST00098 gene were also identified as containing the human LSFR1 gene. PCR analysis revealed that the two human genes are orientated tail-to-tail and that the 3[prime] ends of these genes must be separated by <2 kb. This organization is identical to that found in Fugu (Fig. 1A).
Additional mapping by fluorescence in situhybridization (FISH)
FISH analysis of metaphase spreads of the SD-1 cell line containing the Philadelphia chromosome translocation t(9;22)(q34.1;q11) (34) confirmed that, like ASS(35), the human CCBL1, EST00098, SIAT3C, LSFR1, LSFR2 and LSFR3A genes all map centromeric (proximal) to ABL1 (data not shown), which was as predicted by the RH analysis. This contrasts with the human Surfeit genes that map distal to ABL1 (36).
In order to determine the order of the CCBL1 and LSFR2 genes and the LSFR1 and EST00098 genes relative to the other 9q34.1 genes and to confirm the ordering predicted by the Genebridge 4 RH analysis, a series of two-colour interphase FISH experiments (37) were devised. Their design was based largely on the order of the genes predicted by the Genebridge 4 RH analysis and were such that the relative order of groups of three genes could be determined in a minimum of two complementary experiments involving simultaneous hybridization of three cosmid probes (see Materials and Methods). In the first, two cosmid probes are labelled with biotin and the third with digoxigenin and, in the second, one of the previously biotinylated cosmids is now labelled with digoxigenin resulting in it being detected as a different colour. The most commonly observed orders of the signals seen in the two separate experiments were then combined to give the overall order of the three probes (representative examples of the results from the pair of experiments used to order the human CCBL1, LSFR2 and LSFR3Agenes are shown in Fig. 3).
Figure 3. Representative examples of two-colour interphase FISH experiments used to order the human CCBL1, LSFR2 and LSFR3A genes. (A) A cosmid probe containing the CCBL1 gene was detected as red, and cosmid probes containing the LSFR2 and LSFR3A genes were detected as green. Overall, of 38 clear orders observed, 32 (or 84%) were red-green-green as in the representative example shown. The order of the signals produced is, however, ambiguous with respect to the order of the probes representing the LSFR2 and LSFR3A genes. (B) The probes are identical to those in (A) except that the cosmid probe containing the LSFR3A gene is now detected as red rather than green. Of 37 clear orders observed, 29 (or 78%) were red-green-red as in the representative example shown (the three spots of green flourescence seen to the left are background signal). The order of the signals produced (red-green-red) is now unambiguous, indicating that the order of the genes must be CCBL1-LSFR2-LSFR3A. It is noted that (B) alone is sufficient to order these three probes; however, in other experiments to confirm the RH gene orders, it was necessary to combine the results of two experiments.
Overall, the results were in complete agreement with the Genebridge 4 RH analysis (Fig. 2) and additionally indicated that CCBL1 maps proximal to LSFR2(Fig. 3). However, it was still not possible to order the EST00098 and LSFR1 genes relative to the other genes with any degree of confidence since all the cosmids identified containing these genes overlap to some degree. Whilst 61% of the 51 unambiguous probe orders observed in an experiment performed to order the two genes relative to SIAT3C suggest that EST00098 maps proximal to LSFR1, this is a highly tentative assignation since, in an additional 33 cases, the order of the three probes could not be determined clearly. Although a specific analysis of the distances separating the signals has not been performed, it is estimated that, with the exception of the EST00098-LSFR1 gene pair, all the genes are separated by at least 150 kb and in most cases by significantly larger distances. The relative position and order of these genes and the Surfeit genes on human chromosome band 9q34 compared with the order and position of the homologous Fugu genes in the Fugu genome are shown in Figure 4.
Figure 4. Human and Fugu maps showing differences between the linkage relationships of genes at the three Fugu loci (A, B and C) containing Surfeit genes and their human homologues at human chromosome band 9q34. An ideogram of a G-banded human chromosome 9 is shown to the left, and regions of human chromosome band 9q34 are expanded to its right. Genes mapped and ordered in this study (bold typeface) are shown in relation to other mapped genes and markers (normal typeface) in these regions (29,35,36,38,39). Distances between genes are not to scale and have not been determined precisely. The three Fugu loci containing Surfeit gene homologues are represented as continuous bold lines on the right. The scales of the three Fugu loci are not comparable with the human region. Human and Fugu gene homologues are linked by a thin line to highlight the extent to which gene linkage has been disrupted during the 900 million years of divergent evolution separating the two species.
DISCUSSION
We have identified seven Fugu genes that are linked to Surfeit genes in the Fugu genome and which are also homologous to known human genes or human genes defined by ESTs. The Fugu CCBL1, SIAT3C, LSFR2 and LSFR3 genes are closely linked to the Fugu Surf-3/rpL7a, Surf-1 and Surf-6 genes in a region spanning 60 kb (Fig. 1A), and the Fugu ASS, EST00098 and LSFR1 genes are closely linked to the Fugu Surf-2 and Surf-4 gene homologues in a region spanning 35 kb (Fig. 1B). Mapping of the human homologues of these Fugu genes revealed that these two regions of the Fugu genome represent regions of conserved synteny with human chromosome band 9q34. However, whereas the Surfeit genes map some 2-4 Mb distal to ABL1(29,36,38,39), ASS and the other six human genes map centromeric (proximal) to it (Fig. 4). Therefore, several pairs of genes that are adjacent in the Fugu genome (e.g. Surf-4 and ASS, Surf-2 and EST00098, Surf-6 and CCBL1) are separated by at least 2-4 Mb in the human genome.
Relative ordering of the genes by PCR analysis of the Genebridge 4 panel of RH DNAs and by FISH (both independent methods being in general agreement) additionally has revealed that gene order is also often very different even when the human ABL1-proximal genes are not separated by Surfeit genes in the Fugu genome (e.g. CCBL1, SIAT3C and LSFR2). The only conserved short-range linkage relationship among these human genes is that between the EST00098 and LSFR1 genes which are adjacent and orientated tail-to-tail in both the Fugu and human genomes. Indeed, of the total of 13 genes identified by us within three regions of the Fugu genome, conservation of gene order spanning three genes has not yet been observed. Of course, because of the probable presence of additional genes between the Fugu SIAT3C and LSFR2 genes and the Surf-3/rpL7a and LSFR3 genes (Fig. 1), and the possible presence of additional genes between the human 9q34 genes ordered in this study, the existence of additional conserved short-range gene linkage relationships within these Fugu and human regions cannot be ruled out at this time.
Since our results clearly demonstrate that extensive differences in the linkage relationships of genes exist within syntenic regions of the Fugu genome and human chromosome band 9q34, ordering of human genes that map to this human chromosome band based solely on the order of their gene homologues in Fugu would be highly speculative. This has interesting implications for the proposed use of the Fugu genome in accelerating the isolation, by comparative positional cloning, of human disease genes that map to this band. Three such disease genes, TSC1 (tuberous sclerosis), DYT1 (torsion dystonia) and NPS1/LMX1B (nail patella syndrome), recently have been identified using mammalian-based techniques (40-42). TSC1 maps ~300 kb proximal to the Surfeit locus (40), DYT1 maps just proximal to ASS (43) and NPS1/LMX1B maps to a region which encompasses the SIAT3C gene (44). There currently is no evidence to suggest that the Fugu TSC1, DYT1or NPS/LMX1B genes are closely associated with any of the Fugu genes identified in this study (based on our data and cosmid sequence data present in the HGMP-RC Fugu database).
Whilst the results of this work have specific implications for human chromosome band 9q34, a search of the literature regarding Fugu mapping data reveals that gene order differences between other regions of the human genome and Fugu might be more common than has been suggested previously. A number of other, but more limited, examples of disrupted synteny and/or gene order differences within regions of conserved synteny between human and Fugu (or the related puffer Tetraodon fluviatilis) have been found (12,21,45-48). Furthermore, consideration of new human mapping data provided by GeneMap '98 (49) indicates that gene order differences may also exist in some of the other genomic regions where synteny has previously been found to be conserved between human and Fugu (e.g. refs 21-23). In light of these examples, it is possible that the conclusions drawn from the data presented in this work regarding the utility of the Fugu genome for facilitating human disease gene isolation might be more widely applicable since the above examples of gene linkage differences/disruption incorporate many different regions of the human genome. Attempts to isolate human disease genes using the Fugu genome might, therefore, be ill-advised without first considering the possibility that potentially profound differences in linkage relationships can exist between the Fugu and human genomes even when synteny is conserved.
The results presented in this work also suggest that intra-chromosomal rearrangement may have been common within the vertebrate chromosomal region homologous to human chromosome band 9q34 during the 900 million years of divergent evolution separating mammals and bony fish (450 million years in each separate lineage since their divergence). Comparisons of the orders of the genes (and orientations where known) in both Fugu and human suggest that inversions may have been particularly common. Because some of the genes are adjacent in Fugu but are separated by Mb-sized distances in human, these rearrangements are predicted to have been relatively extensive either in number or in size. It is noted additionally that at least two intra-chromosomal rearrangements (predicted to be inversions) have occurred between human chromosome band 9q34 and the syntenic portion of mouse chromosome 2 (50). The presence of several other intra-chromosomal rearrangements in the human-mouse comparative map (and other mammalian comparative maps) following high resolution mapping of specific chromosomal regions (51-56) suggests that human chromosome band 9q34 may not be unusual in this respect and that intra-chromosomal rearrangements may have been more common during the evolution of vertebrate chromosomes than anticipated previously.
Determining the extent of gene order conservation or non-conservation between the Fugu and human genomes, and not just the extent of conserved synteny, will not only assist researchers in further assessing the suitability of the Fugu genome for the comparative positional cloning of human genes, but should also be revealing as regards the type and frequency of chromosomal rearrangements that have shaped genomes during vertebrate evolution.
MATERIALS AND METHODS
General procedures for DNA analysis
Routine DNA manipulations were carried out according to standard protocols (57). Southern blotting and hybridizations were performed using standard protocols and Hybond-N membrane (Amersham International, Little Chalfont, UK). DNA probes were gel-purified using the Geneclean II kit (BIO 101) and labelled by random hexanucleotide priming (58). Plasmid DNAs were sequenced from vector primers or insert-specific primers using ABI PRISM Dye Terminator Cycle Sequencing (Perkin Elmer, Norwalk, CT) for use with the automated ABI PRISM 377 DNA sequencer (Perkin Elmer Applied Biosystems, Foster City, CA). Sequences were processed using the MacVector 6.0.1 sequence analysis software (Oxford Molecular Group, Oxford, UK).
Generation of the enriched Fugu cDNA library
A Fugu cDNA library enriched for sequences from Fugu cosmids 137L19, 139G11, 183B03, 028B13, 177E10, 070P11, 186H17, 194D10, 036L10 and 006C20 (7) was constructed essentially as described previously (59). Human Cot-1 DNA, Lawrist 4 vector and cloned Fugu repetitive DNA elements were used as competitors. Fugu cDNAs initially were PCR-amplified from a Fugu 5[prime]-STRETCH PLUS cDNA library (Clontech, Palo Alto, CA) in 100 µl reactions from 2 µl of library phage lysate using pTriplEx vector primers flanking the cDNA inserts (5[prime]-CTCGGGAAGCGCGCCATTGTGTTGG-3[prime] and 5[prime]-CGACTCACTATAGGGCGAATTGGCC-3[prime]). Twenty-five cycles of 30 s at 94°C, 20 s at 58°C and 3 min at 72°C were performed. Two rounds of enrichment were performed and cDNAs amplified as before except that after the second round the primers possessed the additional nucleotides 5[prime]-CUACUACUACUA-3[prime] at their 5[prime] ends. Amplified cDNAs were cloned using the CloneAmp pAMP10 System (Gibco BRL, Paisley, UK). A total of 576 clones were stamped in gridded arrays onto Hybond-N membranes and processed as described for Hybond-N. The enriched library was screened with restriction fragments derived from the template cosmids to identify positive clones derived from these cosmids. Cosmid sequences from which the cDNAs were derived were identified by Southern blot analysis and sequencing.
PCR analysis of panels of human-rodent somatic cell hybrids and radiation hybrids
The panel of mono-chromosomal somatic cell hybrids (30) and the HGMP-RC subset (85 of the total 93 hybrid cell line DNAs) of the Genebridge 4 RH panel (31) were supplied by the HGMP-RC (Hinxton, Cambridge, UK). PCR primer pairs used for the somatic cell hybrid analysis were: SIAT3C, 5[prime]-GAAGAGCCACCCCTCAGTGCC-3[prime] and 5[prime]-GGCAGGACGACGGAAGCTACTC-3[prime]; LSFR1, 5[prime]-GAGTTCCAGGACCACATGTCG-3[prime] and 5[prime]-CATCCTCTGTCTCCAGGACG-3[prime]; LSFR2, 5[prime]-GAAACTGCAGTGACCAGTGG-3[prime] and 5[prime]-GGCTCCACTTGGTGACTTGG-3[prime]; LSFR3A, 5[prime]-CAGAACGTCTGGGAGCACCTG-3[prime] and 5[prime]-GATGGAGCACACGTACAGGTGG-3[prime]; and LSFR3B, 5[prime]-GTAAAGTACAAAACTTTCCTGAG-3[prime] and 5[prime]-CATGTCTGAATCTTTTAAGGTGG-3[prime]. Those used for the RH analysis were: SIAT3C, LSFR2 andLSFR3A as above; EST00098, 5[prime]-CTCCTGTCTGGACCACCTGGAGG-3[prime] and 5[prime]-TGCTGCTGGAACTCCAGGCGT-3[prime]; CCBL1, 5[prime]-GAGCTGCTGGGTCAGGAGATAGACC-3[prime] and 5[prime]-CCTGCCATCATTGTCATGGGCTCG-3[prime]; LSFR1, 5[prime]-CATGCCTGCTCTAGCCTACAC-3[prime] and 5[prime]-CATCCTCTGTCTCCAGGACG-3[prime];ASS, 5[prime]-GGTTGGCCTAAGAAAACCAT-3[prime] and 5[prime]-TGGGGAGCTATAAAAATGAC-3[prime] (amplifying an intron 14 CA repeat); and human Surf-6 (SURF6), 5[prime]-GGTGAGAGTCGCACGCGCAG-3[prime] and 5[prime]-CGTTTCGCAGGCCCCTTAGT-3[prime]. Primers were tested for specificity on human, mouse and hamster genomic DNA and amplified products cloned and sequenced to confirm their identity. Annealing temperatures were 60°C for the CCBL1, EST00098, LSFR1 (both sets) and SURF6 primer pairs, 58°C for the LSFR3A and LSFR3B primer pairs, and 57, 56 and 54°C for the LSFR2, SIAT3C and ASS primer pairs, respectively. The 50 µl reactions for somatic cell hybrid PCR consisted of 1 µl of somatic cell hybrid DNA, 50 mM KCl, 10 mM Tris-HCl, pH 8.3, 1.5 mM MgCl2, 10% dimethyl sulfoxide, 200 µM dNTPs, 1 µM of each primer and 5 U of Taq polymerase. Thirty-five cycles of 30 s at 94°C, 20 s at the primer annealing temperature, and 3 min at 72°C were performed following an initial 3 min incubation at 94°C. Conditions in each 25 µl reaction for RH PCR were as above except that 1 µl of hybrid DNA and 2.5 U of Taq polymerase were used, and 45 cycles were performed. Analysis was performed three times for each set of primers. Amplified fragments were separated on 2.0-3.0% NuSieve GTG agarose gels (Flowgen, Lichfield, UK), typed as recommended and submitted to the Whitehead Institute/MIT for framework mapping via the HGMP-RC (http://www.hgmp.mrc.ac.uk/ ).
Isolation of human chromosome 9-specific cosmid clones
Human cosmid clones were identified from a chromosome 9-specific library LL09NC01 `P' (constructed at the Biomedical Sciences Division, Lawrence Livermore National Laboratory, Livermore, CA, under the auspices of the National Laboratory Gene Library Project, sponsored by the US Department of Energy and supplied by the HGMP-RC) by screening with the human somatic cell and RH gene-specific PCR-amplified products described above, except for the CCBL1 probe which was derived by PCR from primers 5[prime]-GTGGAAGGTGGAACTCTAGCC-3[prime] and 5[prime]-CTCACCTTTCAGACAGGAGGC-3[prime]. Clones were analysed for the presence of gene-specific sequences by PCR, Southern blot analysis and sequencing. The orientation and separation of the LSFR1andEST00098 genes on cosmid P132C4 were determined by standard PCR using primers 5[prime]-TCCTGGTGCCCGTGATGG-3[prime] and 5[prime]-CTCCTGTCTGGACCACCTGGAGG-3[prime].
Mapping and ordering of genes by FISH
Slides of metaphase chromosome spreads and interphase nuclei were prepared, according to standard procedures, from phytohaemagglutinin (PHA)-stimulated normal human lymphocytes and the chronic myelogenous leukaemic cell line SD-1 carrying the Philadelphia chromosome translocation t(9;22)(q34;q11) involving the ABL1 and BCR genes (34). Single and two-colour FISH were performed essentially as described previously (37) except that in some cases biotinylated and digoxigenin-labelled cosmid probes were detected by single layer detection with FITC-avidin DCS (Vector Laboratories, Burlingame, CA) and/or anti-digoxigenin-rhodamine (Boehringer Mannheim, Lewes, UK). Slides were mounted in Citifluor (Citifluor, Canterbury, UK) containing 0.15 µg/ml 4[prime],6-diamidino-2-phenylindole as a counterstain. Signals were observed with a Zeiss Axioplan epifluorescence microscope and images captured digitally using an attached Photometrics KAF 1400-500 cooled CCD camera and combined using IP Lab Spectrum software (Signal Analytics, Vienna, VA).
ACKNOWLEDGEMENTS
We would like thank Drs Anna-Marie Frischauf, Paul Nurse and Denise Sheer for their helpful comments during the preparation of this manuscript, and Tania Jones and Moira Read for advice and assistance with all aspects of the FISH analysis.
REFERENCES
*To whom correspondence should be addressed. Tel: +44 171 269 3297; Fax: +44 171 269 3093; Email: fried{at}icrf.icnet.uk
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  V. Katju and M. Lynch The Structure and Early Evolution of Recently Arisen Gene Duplicates in the Caenorhabditis elegans Genome Genetics, December 1, 2003; 165(4): 1793 - 1803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Barton, B. Gottgens, M. Gering, J. G. R. Gilbert, D. Grafham, J. Rogers, D. Bentley, R. Patient, and A. R. Green Regulation of the stem cell leukemia (SCL) gene: A tale of two fishes PNAS, May 24, 2001; (2001) 101532998. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Seoighe, N. Federspiel, T. Jones, N. Hansen, V. Bivolarovic, R. Surzycki, R. Tamse, C. Komp, L. Huizar, R. W. Davis, et al. Prevalence of small inversions in yeast gene order evolution PNAS, November 16, 2000; (2000) 240462997. [Abstract] [Full Text]
|
|
|
|
 |
|
 H. Davidson, M. S. Taylor, A. Doherty, A. C. Boyd, and D. J. Porteous Genomic Sequence Analysis of Fugu rubripes CFTR and Flanking Genes in a 60 kb Region Conserving Synteny with 800 kb of Human Chromosome 7 Genome Res., August 1, 2000; 10(8): 1194 - 1203. [Abstract] [Full Text]
|
|
|
|
|
|
G. Elgar, M. S. Clark, S. Meek, S. Smith, S. Warner, Y. J.K. Edwards, N. Bouchireb, A. Cottage, G. S.H. Yeo, Y. Umrania, et al. Generation and Analysis of 25 Mb of Genomic DNA from the Pufferfish Fugu rubripes by Sequence Scanning Genome Res., October 1, 1999; 9(10): 960 - 971. [Abstract] [Full Text]
|
|
|
|
|
|
L. M. Barton, B. Gottgens, M. Gering, J. G. R. Gilbert, D. Grafham, J. Rogers, D. Bentley, R. Patient, and A. R. Green From the Cover: Regulation of the stem cell leukemia (SCL) gene: A tale of two fishes PNAS, June 5, 2001; 98(12): 6747 - 6752. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Seoighe, N. Federspiel, T. Jones, N. Hansen, V. Bivolarovic, R. Surzycki, R. Tamse, C. Komp, L. Huizar, R. W. Davis, et al. Prevalence of small inversions in yeast gene order evolution PNAS, December 19, 2000; 97(26): 14433 - 14437. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||