Strategy to sequence the genome of Corynebacterium glutamicum ATCC 13032: use of a cosmid and a bacterial artificial chromosome library

The initial strategy of the Corynebacterium glutamicum genome project was to sequence overlapping inserts of an ordered cosmid library. High-density colony grids of approximately 28 genome equivalents were used for the identification of overlapping clones by Southern hybridization. Altogether 18 contiguous genomic segments comprising 95 overlapping cosmids were assembled. Systematic shotgun sequencing of the assembled cosmid set revealed that only 2.84 Mb (86.6%) of the C. glutamicum genome were represented by the cosmid library. To obtain a complete genome coverage, a bacterial artificial chromosome (BAC) library of the C. glutamicum chromosome was constructed in pBeloBAC11 and used for genome mapping. The BAC library consists of 3168 BACs and represents a theoretical 63-fold coverage of the C. glutamicum genome (3.28 Mb). Southern screening of 2304 BAC clones with PCR-amplified chromosomal markers and subsequent insert terminal sequencing allowed the identification of 119 BACs covering the entire chromosome of C. glutamicum. The minimal set representing a 100% genome coverage contains 44 unique BAC clones with an average overlap of 22 kb. A total of 21 BACs represented linking clones between previously sequenced cosmid contigs and provided a valuable tool for completing the genome sequence of C. glutamicum.     

Mapping analysis of the Xylella fastidiosa genome

A cosmid library was made of the 2.7 Mb genome of the Gram-negative plant pathogenic bacterium Xylella fastidiosa and analysed by hybridisation mapping. Clones taken from the library as well as genomic restriction fragments of rarely cutting enzymes were used as probes. The latter served as a backbone for ordering the initial map contigs and thus facilitated gap closure. Also, the co-linearity of the cosmid map, and thus the eventual sequence, could be confirmed by this process. A subset of the eventual clone coverage was distributed to the Brazilian X.fastidiosa sequencing network. Data from this effort confirmed more quantitatively initial results from the hybridisation mapping that the redundancy of clone coverage ranged between 0 and 45-fold across the genome, while the average was 15-fold by experimental design. Reasons for this not unexpected fluctuation and the actual gaps are being discussed, as is the use of this effect for functional studies.     

Construction and utility of 10-kb libraries for efficient clone-gap closure for rice genome sequencing

Rice is an important crop and a model system for monocot genomics, and is a target for whole genome sequencing by the International Rice Genome Sequencing Project (IRGSP). The IRGSP is using a clone by clone approach to sequence rice based on minimum tiles of BAC or PAC clones. For chromosomes 10 and 3 we are using an integrated physical map based on two fingerprinted and end-sequenced BAC libraries to identifying a minimum tiling path of clones. In this study we constructed and tested two rice genomic libraries with an average insert size of 10 kb (10-kb library) to support the gap closure and finishing phases of the rice genome sequencing project. The HaeIII library contains 166,752 clones covering approximately 4.6x rice genome equivalents with an average insert size of 10.5 kb. The Sau3AI library contains 138,960 clones covering 4.2x genome equivalents with an average insert size of 11.6 kb. Both libraries were gridded in duplicate onto 11 high-density filters in a 5 x 5 pattern to facilitate screening by hybridization. The libraries contain an unbiased coverage of the rice genome with less than 5% contamination by clones containing organelle DNA or no insert. An efficient method was developed, consisting of pooled overgo hybridization, the selection of 10-kb gap spanning clones using end sequences, transposon sequencing and utilization of in silico draft sequence, to close relatively small gaps between sequenced BAC clones. Using this method we were able to close a majority of the gaps (up to approximately 50 kb) identified during the finishing phase of chromosome-10 sequencing. This method represents a useful way to close clone gaps and thus to complete the entire rice genome.     

Delimitation of the fertility restorer locus Rfk1 to a 43-kb contig in Kosena radish (Raphanus sativus L.)

We are pursuing a positional cloning strategy to isolate the fertility restoration gene Rfk1 from radish. Random polymorphic DNA-sequence-tagged site (RAPD-STS) markers tightly linked to the gene in radish were isolated, and a RAPD map surrounding the Rfk1 locus was constructed. We surveyed 948 F2 plants with adjacent RAPD-STS markers to isolate recombinants for bulk segregant analysis. This analysis was effective in isolating tightly linked amplification fragment length polymorphism (AFLP) markers surrounding the gene of interest. Ten tightly linked AFLP markers were obtained and used to construct a high-resolution map of the region. The closest AFLP-STS markers flanking Rfk1 were 0.1 cM and 0.2 cM away. Using the four adjacent AFLP markers, we screened lambda and cosmid libraries. The lambda and cosmid clones were aligned by examination of end sequences and restriction fragment length polymorphism (RFLP) patterns for each clone, and by hybridization to the DNA isolated from recombinants. Finally, we constructed a 198-kb contig encompassing the Rfk1 gene and comprising 20 lambda and two cosmid clones. By analysis of the breakpoints in recombinants with the rfk1/rfk1 or Rfk1/- genotype, the Rfk1 locus could be assigned to a 43-kb region comprising four lambda clones and one cosmid clone. This pinpoint localization in the radish genome has made it possible for us to identify the gene by sequence analysis and genetic transformation of cytoplasmic male-sterile Brassica napus plants.     

A yeast artificial chromosome contig encompassing the cystic fibrosis locus.

The gene responsible for cystic fibrosis (CF) has recently been identified. Coding sequence for the cystic fibrosis transmembrane conductance regulator (CFTR) spans at least 230 kb of the human genome. Although all 27 exons of the gene are represented in cosmid or bacteriophage clones, there are still several gaps in the physical map of this region. It should be possible to complete the map and to clone the entire CFTR gene in a single fragment of DNA using a yeast artificial chromosome (YAC) vector. Herein we describe the construction and physical mapping of a 1.5-Mb YAC contig which encompasses D7S8 (J3.11) and D7S23 (KM19), two genetic loci flanking the CF locus. One of the clones in the contig, 37AB12, contains a 310-kb YAC which includes the entire CFTR gene and flanking sequence in both the 5' and 3' directions.     

Preselection of shotgun clones by oligonucleotide fingerprinting: an efficient and high throughput strategy to reduce redundancy in large-scale sequencing projects.

Large-scale genomic sequencing projects generally rely on random sequencing of shotgun clones, followed by different gap closing strategies. To reduce the overall effort and cost of those projects and to accelerate the sequencing throughput, we have developed an efficient, high throughput oligonucleotide fingerprinting protocol to select optimal shotgun clone sets prior to sequencing. Both computer simulations and experimental results, obtained from five PAC-derived shotgun libraries spanning 535 kb of the 17p11.2 region of the human genome, demonstrate that at least a 2-fold reduction in the number of sequence reads required to sequence an individual genomic clone (cosmid, PAC, etc.) can be achieved. Treatment of clone contigs with significant clone overlaps will allow an even greater reduction.     

Clone bank of the tobacco (Nicotiana tabacum) chloroplast genome as a set of overlapping restriction endonuclease fragments: Mapping of eleven ribosomal protein genes

Chloroplast DNA from tobacco (Nicotiana tabacum) was digested with BamHI, SalI, PstI, XhoI and PvuII. Fragments produced by partial digestion with BamHI were size-fractionated. The resulting fragments were cloned into plasmids pBR328 and pACYC177. Large fragments were cloned in a cosmid, pHC79. Using these clones a clone bank of the entire tobacco chloroplast genome as a set of overlapping fragments was obtained. BamHI and PstI cleavage maps were constructed by analysis of the cloned fragments of the chloroplast DNA. The 11 ribosomal protein genes and other major sequenced genes have been marked on the map.     

Comparison of two Xanthomonas campestris pathovar campestris genomes revealed differences in their gene composition

For the Xanthomonas campestris pathovar campestris wild-type strain B100 a plasmid-based clone library was constructed. The plasmids carried chromosomal fragments of 3-4 kb in size that were tagged in vitro with the artificial transposon KAN-2. More than 3000 of the transposon target sites were characterized by DNA sequencing. The sequences obtained were compared to the recently published genome of Xanthomonas campestris pathovar campestris strain ATCC 33913. Most of the sequenced clones derived from strain B100 matched the chromosomal sequence of strain ATCC 33913. An alignment to the circular map of this chromosome revealed that the similarities were statistically distributed over the entire genome of strain ATCC 33913. The similarity was obvious for protein coding sequences, as well as for mobile genetic elements. However, four regions in the genome of Xanthomonas campestris pathovar campestris strain ATCC 33913, ranging in size from 11 to 37 kb, were not represented in the sequenced clone library of Xanthomonas campestris pathovar campestris strain B100. On the other hand, 1.2% of the sequenced clones originating from Xanthomonas campestris pathovar campestris strain B100 showed no or insignificant similarities to the genome of strain ATCC 33913.     

Ordering of cosmid clones covering the herpes simplex virus type I (HSV-I) genome: a test case for fingerprinting by hybridisation.

To allow the efficient construction of ordered clone libraries, we have been investigating the use of 'oligonucleotide fingerprinting' as an approach to identify overlapping clones, and ultimately restore the linear order of the clone set. To test the effectiveness of the procedure, we have constructed a cosmid library from the genome of the human DNA virus HSV-I and used hybridisation to multiple oligonucleotides selected from the nucleotide sequence to reconstruct the order of clones and oligonucleotides on the genome.     

The Neurospora crassa genome: cosmid libraries sorted by chromosome

A Neurospora crassa cosmid library of 12,000 clones (at least nine genome equivalents) has been created using an improved cosmid vector pLorist6Xh, which contains a bacteriophage lambda origin of replication for low-copy-number replication in bacteria and the hygromycin phosphotransferase marker for direct selection in fungi. The electrophoretic karyotype of the seven chromosomes comprising the 42.9-Mb N. crassa genome was resolved using two translocation strains. Using gel-purified chromosomal DNAs as probes against the new cosmid library and the commonly used medium-copy-number pMOcosX N. crassa cosmid library in two independent screenings, the cosmids were assigned to chromosomes. Assignments of cosmids to linkage groups on the basis of the genetic map vs. the electrophoretic karyotype are 93 +/- 3% concordant. The size of each chromosome-specific subcollection of cosmids was found to be linearly proportional to the size of the particular chromosome. Sequencing of an entire cosmid containing the qa gene cluster indicated a gene density of 1 gene per 4 kbp; by extrapolation, 11,000 genes would be expected to be present in the N. crassa genome. By hybridizing 79 nonoverlapping cosmids with an average insert size of 34 kbp against cDNA arrays, the density of previously characterized expressed sequence tags (ESTs) was found to be slightly <1 per cosmid (i.e., 1 per 40 kbp), and most cosmids, on average, contained an identified N. crassa gene sequence as a starting point for gene identification.     

Progress towards construction of a total restriction fragment map of a human chromosome.

We present an approach to the construction of an overlapping restriction fragment map of a single human chromosome. A genomic cosmid library genome was constructed from a mouse-human hybrid cell line containing chromosome 17 as its only human genetic component. Cosmids containing human inserts were isolated by hybridisation to a human Alu sequence. DNAs from ninety-six randomly chosen cosmids were digested with either EcoRI or HindIII, end-labelled with 35S-dATP and analysed using agarose gel electrophoresis. Comparison of the restriction fragment patterns revealed two pairs of overlapping clones, that were confirmed by cross-hybridization of the overlapping fragments. The two pairs of cosmids both mapped to human chromosome 17, as shown by hybridization to a panel of somatic cell hybrids. These data demonstrate that the generation of an overlapping cosmid map along a human chromosome is feasible, representing an intermediate step towards the complete sequencing of a human chromosome.     

Physical map of the genome of Rhodobacter capsulatus SB 1003.

A map of the chromosome of Rhodobacter capsulatus was constructed by overlapping the large restriction fragments generated by endonucleases AseI and XbaI. The analyses were done by hybridization of single fragments with the restriction fragments blotted from pulsed-field gels and by grouping cosmids of a genomic library of R. capsulatus into contigs, corresponding to the restriction fragments, and further overlapping of the contigs. A technical difficulty due to a repeated sequence made it necessary to use hybridization with cloned genes and prior knowledge of the genetic map in order to close the physical circle in a unique way. In all, 41 restriction sites were mapped on the 3.6-Mb circular genome and 22 genes were positioned at 26 loci of the map. Cosmid clones were grouped in about 80 subcontigs, forming two groups, one corresponding to the chromosome of R. capsulatus and the other corresponding to a 134-kb plasmid. cos site end labeling and partial digestion of cosmids were used to construct a high-resolution EcoRV map of the 134-kb plasmid. The same method can be extended to the entire chromosome. The cosmid clones derived in this work can be used as a hybridization panel for the physical mapping of new genes as soon as they are cloned.     

A combined physical and genetic map of Pseudomonas aeruginosa PAO

A combined physical and genetic map of Pseudomonas aeruginosa PAO was constructed by pulsed-field gel electrophoresis and Southern hybridization using cosmid clones from a genomic library carrying known genes. A total of 37 SpeI restriction fragments have been mapped on the 5862 kb genome, and fragment contiguity demonstrated by hybridization with clones from a SpeI junction fragment library and fragments obtained by partial SpeI digestion, both derived from the P. aeruginosa PAO chromosome.     

A physical map of the highly heterozygous Populus genome: integration with the genome sequence and genetic map and analysis of haplotype variation

As part of a larger project to sequence the Populus genome and generate genomic resources for this emerging model tree, we constructed a physical map of the Populus genome, representing one of the few such maps of an undomesticated, highly heterozygous plant species. The physical map, consisting of 2802 contigs, was constructed from fingerprinted bacterial artificial chromosome (BAC) clones. The map represents approximately 9.4-fold coverage of the Populus genome, which has been estimated from the genome sequence assembly to be 485 ± 10 Mb in size. BAC ends were sequenced to assist long-range assembly of whole-genome shotgun sequence scaffolds and to anchor the physical map to the genome sequence. Simple sequence repeat-based markers were derived from the end sequences and used to initiate integration of the BAC and genetic maps. A total of 2411 physical map contigs, representing 97% of all clones assigned to contigs, were aligned to the sequence assembly (JGI Populus trichocarpa , version 1.0). These alignments represent a total coverage of 384 Mb (79%) of the entire poplar sequence assembly and 295 Mb (96%) of linkage group sequence assemblies. A striking result of the physical map contig alignments to the sequence assembly was the co-localization of multiple contigs across numerous regions of the 19 linkage groups. Targeted sequencing of BAC clones and genetic analysis in a small number of representative regions showed that these co-aligning contigs represent distinct haplotypes in the heterozygous individual sequenced, and revealed the nature of these haplotype sequence differences.     

An integrated map of Arabidopsis thaliana for functional analysis of its genome sequence.

The genome of the model plant species Arabidopsis thaliana has recently been sequenced. To accelerate its current genome research, we developed a whole-genome, BAC/BIBAC-based, integrated physical, genetic, and sequence map of the A. thaliana ecotype Columbia. This new map was constructed from the clones of a new plant-transformation-competent BIBAC library and is integrated with the existing sequence map. The clones were restriction fingerprinted by DNA sequencing gel-based electrophoresis, assembled into contigs, and anchored to an existing genetic map. The map consists of 194 BAC/BIBAC contigs, spanning 126 Mb of the 130-Mb Arabidopsis genome. A total of 120 contigs, spanning 114 Mb, were anchored to the chromosomes of Arabidopsis. Accuracy of the integrated map was verified using the existing physical and sequence maps and numerous DNA markers. Integration of the new map with the sequence map has enabled gap closure of the sequence map and will facilitate functional analysis of the genome sequence. The method used here has been demonstrated to be sufficient for whole-genome physical mapping from large-insert random bacterial clones and thus is applicable to rapid development of whole-genome physical maps for other species.     

cosmid克隆筛选的体内同源重组法——一个含有小鼠t复合体连锁DNA顺序的cosmid克隆的分离

一个从cosmid分子克隆库中筛选特别基因顺序的遗传学方法——体内同源重组(invlvo homologous recombination)法。即使探针DNA与分子克隆库中带有与探针同源顺序的克隆发生体内重组,然后以遗传学方法进行筛选。cosmid分子克隆库构建在rec宿主细胞内,经体内包装(in vivo Packaging)成λ噬菌体颗粒,把该噬菌体颗粒转入带有探针DNA的rec~+细胞内,探针是已被克隆在与cosmid载体没有同源顺序的质粒(如PUC8或PUC9)内的。经过一段时间(1—3小时),待重组发生后,把cosmid进行体内包装。此时探针DNA连同质粒已整合入cosmid基因组内,因此它带有原为两个载体所分别带有的双重抗性——Amp~r(氨苄青霉素,PUC8或PUC9)和Kan~r(卡那霉素,cosmid)。这种双重抗性菌落可在含有这2种抗菌素的培养平皿上选出,该重组cosmid借助于λ切除酶的作用将已被整合的探针质粒重新切除,再经体内包装后,该cosmid被还原并纯化,然后可用一含有Xgal的培皿识别和选出。本文用此法以有关DNA探针从cosmid分子克隆库中分离得到含有与小鼠t复合体连锁的基因组顺序的克隆,并对该克隆作了物理图谱分析。     

Cosmid cloning and sample sequencing of the genome of the uncultivable mollicute,Western X-disease phytoplasma,using DNA purified by pulsed-field gel electrophoresis

Western X-disease (WX) phytoplasma is an uncultivable, intracellular pathogen of plants and insects with an AT-rich genome of 670 kb. As part of the genome sequencing project of WX phytoplasma, we have cloned approximately 50% of its genome into the pcosRW2 cosmid vector using DNA purified from pulsed-field gels. One of the cosmid clones with an insert of 24.6 kb was sequenced, which along with the cosmid end sequences, represents 60 kb of unique sequence from the WX phytoplasma genome. The putative genes identified in this sequence represent a wide variety of functions and many had not been previously identified in a phytoplasma.     

High-Resolution Physical Map of the Immunoglobulin λ Variant Gene Cluster Assembled by Quantitative DNA Fiber Mapping

Quantitative DNA fiber mapping (QDFM) allows rapid construction of near-kilobase-resolution physical maps by hybridizing specific probes to individual stretched DNA molecules. We evaluated the utility of QDFM for the large-scale physical mapping of a rather unstable, repeat-rich 850-kb region encompassing the immunoglobulin λ variant (IGLV) gene segments. We mapped a minimal tiling path composed of 32 cosmid clones to three partially overlapping yeast artificial chromosome (YAC) clones and determined the physical size of each clone, the extent of overlap between clones, and contig orientation, as well as the sizes of gaps between adjacent contigs. Regions of germline DNA for which we had no YAC coverage were characterized by cosmid to cosmid hybridizations. Compared to other methods commonly used for physical map assembly, QDFM is a rapid, versatile technique delivering unambiguous data necessary for map closure and preparation of sequence-ready minimal tiling paths.     

A minimal tiling path cosmid library for functional analysis of the Pseudomonas aeruginosa PAO1 genome

Pseudomonas aeruginosa is an important pathogenic and environmental bacterium, with the most widely studied strain being PAO1. Using the PAO1 reference cosmid library and the recently completed PAO1 genome sequence, we have mapped a minimal tiling path across the genome using a two-step strategy. First, we sequenced both ends of a set of over 500 random and previously mapped clones to create a backbone. Second, we end-sequenced a second set of cosmid clones that were identified to lie within the larger gaps using hybridization of the reference library filters with probes designed against sequences at the center of each gap. The minimal tiling path was calculated using the program Domino (http://www.bit.uq.edu.au/download/), with the overlap between adjacent clones set to 5 kb (where possible) to minimize the chance of truncating genes. This yielded a minimal tiling cosmid library (334 clones) covering 93.7% of the genome in 57 contigs. This library has reduced to a workable set the number of clones required to represent the majority of the P. aeruginosa genome and gives the precise location of each cosmid, enabling most genes of interest to be located on clones without further screening. This library should prove a useful resource to accelerate functional analysis of the P. aeruginosa genome.