TECHNICAL ADVANCE
Advanced resources for plant genomics: a BAC library specific
for the short arm of wheat chromosome 1B
Jaroslav Janda1, Jan Sˇafa´rˇ1, Marie Kubala´kova´1,2, Jan Bartosˇ1, Pavlı´na Kova´rˇova´1, Pavla Sucha´nkova´1, Stephanie Pateyron3, Jarmila Cˇı´halı´kova´1,2, Pierre Sourdille4, Hana Sˇimkova´1, Patricia Faivre-Rampant3, Eva Hrˇibova´1, Michel Bernard4, Adam Lukaszewski5, Jaroslav Dolezˇel1,2,* and Boulos Chalhoub3
1
Laboratory of Molecular Cytogenetics and Cytometry, Institute of Experimental Botany, Sokolovska´ 6, CZ-77200 Olomouc, Czech Republic,
2
Department of Cell Biology and Genetics, Palacky´ University, Sˇlechtitelu˚ 11, CZ-78371 Olomouc, Czech Republic,
3Organization and Evolution of Plant Genomes, Unite´ de Recherches en Ge´nomique Ve´ge´tale (INRA-URGV), 2 rue Gaston
Cre´mieux, CP 5708, F-91057 E´vry Cedex, France,
4Ge´ne´tique Mole´culaire des Ce´re´ales, UMR INRA-UBP, Domaine de Crouelle, 234 Avenue du Bre´zet, F-63039 Clermont-Ferrand
Cedex 2, France, and
5Department of Botany and Plant Sciences, University of California, Riverside, CA 92521, USA
Received 1 November 2005; revised 2 May 2006; accepted 30 May 2006. *For correspondence (faxþ 420 585 205 853; e-mail dolezel@ueb.cas.cz).
Summary
Common wheat (Triticum aestivum L., 2n¼ 6x ¼ 42) is a polyploid species possessing one of the largest genomes among the cultivated crops (1C is approximately 17 000 Mb). The presence of three homoeologous genomes (A, B and D), and the prevalence of repetitive DNA make sequencing the wheat genome a daunting task. We have developed a novel ‘chromosome arm-based’ strategy for wheat genome sequencing to simplify this task; this relies on sub-genomic libraries of large DNA inserts. In this paper, we used a di-telosomic line of wheat to isolate six million copies of the short arm of chromosome 1B (1BS) by flow sorting. Chromosomal DNA was partially digested with HindIII and used to construct an arm-specific BAC library. The library consists of 65 280 clones with an average insert size of 82 kb. Almost half of the library (45%) has inserts larger than 100 kb, while 18% of the inserts range in size between 75 and 100 kb, and 37% are shorter than 75 kb. We estimated the chromosome arm coverage to be 14.5-fold, giving a 99.9% probability of identifying a clone corresponding to any sequence on the short arm of 1B. Each chromosome arm in wheat can be flow sorted from an appropriate cytogenetic stock, and we envisage that the availability of chromosome arm-specific BAC resources in wheat will greatly facilitate the development of ready-to-sequence physical maps and map-based gene cloning. Keywords: wheat, genomics, physical mapping, BAC resources, chromosome sorting, flow cytometry.
Introduction
Common wheat (Triticum aestivum L.) is grown on a larger acreage than any other crop plant. Because of its socio-economic importance, a sustained increase in pro-duction must be ensured to keep pace with the growing human population. It is expected that the breeding of improved varieties will be accelerated by a better know-ledge of the wheat genome. However, genomics in wheat has been hampered by the size and the complexity of its genome. Common wheat is an allohexaploid species consisting of three closely related genomes (AABBDD,
2n¼ 6x ¼ 42), resulting in sequence redundancy (Gill et al., 2004). The estimated size of the hexaploid wheat genome is 17 billion bases (Bennett and Smith, 1991), mainly consisting of repetitive DNA (approximately 80%, Smith and Flavell, 1975), and is >100 times larger than the genome of the model plant Arabidopsis thaliana. Each individual wheat chromosome is larger than the recently sequenced rice genome (Goff et al., 2002). These unique features pose great challenges for gene discovery and genome sequencing.
Because of its large size and complexity, sequencing of the wheat genome requires careful consideration and development of appropriate strategies and technologies. In addition to clone-by-clone sequencing and selective sequencing of random bacterial artificial chromosome (BAC) clones from gene-rich regions, shotgun sequencing of ‘gene-enriched’ DNA obtained after genome filtration was suggested (Gill et al., 2004). Regardless of which strategy is chosen, a robust physical map will be required. This holds true not only for the sequencing approaches based on large-insert DNA clones but also for short sequences obtained from gene-enriched DNA that will have to be ordered on the genomic scaffold (Meyers et al., 2004). Beside its utility for genome sequencing, the availability of a global physical map would greatly accelerate positional gene cloning.
Physical maps are produced by ordering DNA clones from large-insert (typically BAC) libraries on the basis of a clone fingerprint pattern (Luo et al., 2003). Several genomic BAC libraries have been constructed for bread wheat (Allouis et al., 2003; Liu et al., 2000; Ma et al., 2000; Nilmalgoda et al., 2003). To achieve the required genome coverage, one to two million BAC clones have to be fingerprinted. Although it is possible to fingerprint such large numbers of clones with the automated fingerprinting technique of Luo et al. (2003), it is doubtful that biological constraints due to polyploidy and current contig assembly technology would allow contigs and physical maps to be built from such a large number of fingerprints (Jan Dvorˇa´k, University of California, Davis, CA, USA, personal communication).
One approach to simplify this task is to employ diploid relatives of wheat whose genomes are smaller. Several BAC libraries representing genomes of individual wheat progen-itors have been constructed. Moullet et al. (1999) created a BAC library from the D genome donor of wheat Aegilops tauschii, and Lijavetzky et al. (1999) developed a BAC library from T. monococcum, which is related to T. urartu, the progenitor of the A genome. BAC libraries from T. urartu and Ae. speltoides, the presumed progenitor of the B ge-nome, were developed recently (Akhunov et al., 2005). However, the hexaploid wheat has diverged compared with its wild diploid progenitor species. Polyploidization events cause important DNA loss and rearrangements (Akhunov et al., 2003; Anderson et al., 2003; Chantret et al., 2005; Dvorak et al., 2004; Kong et al., 2004; Wicker et al., 2003) suggesting that the diploid species constitute poor surro-gates for genomic studies in bread wheat.
We believe that a fundamental solution to the problem of sequencing large genomes is to create BAC libraries specific for small and defined parts thereof. In line with this, we have developed procedures for purification of individual chromosomes and chromosome arms by flow cytometry (Kubala´kova´ et al., 2002, 2005; Vra´na et al., 2000) and a protocol for preparation of intact DNA from sorted chromo-somes that is suitable for cloning (Sˇimkova´ et al., 2003).
These advances, and a highly efficient protocol for BAC cloning (Chalhoub et al., 2004), have allowed the creation of two sub-genomic BAC libraries from hexaploid wheat: a composite library specific for chromosomes 1D, 4D and 6D (Janda et al., 2004) and a library specific for chromosome 3B (Sˇafa´rˇ et al., 2004). Our ability to sort single chromosome arms from polyploid wheat (Kubala´kova´ et al., 2002, 2005) offers the possibility of constructing BAC libraries from even smaller parts of the genome. The size of individual wheat chromosome arms ranges from 224 to 582 Mb, representing only 1.3–3.4% of the whole genome. This is within the size range of plant genomes that have already been sequenced and/or will be sequenced in the near future.
Even with a reliable chromosome arm purification proto-col, creation of a BAC library still represents a challenge because of the minute amounts of DNA obtained from flow-sorted chromosome arms. To the best of our knowledge, no chromosome arm-specific BAC library has been constructed for any organism. In this paper, we describe the construction and characterization of a BAC library specific for the short arm of chromosome 1B (1BS) of wheat. This chromosome arm represents only 1.9% of the hexaploid wheat genome and carries a large number of important genes, including disease and pest resistance genes and genes for storage proteins that directly affect grain quality (Erayman et al., 2004; Sandhu et al., 2001). The Wheat Gene Catalog (http:// wheat.pw.usda.gov/ggpages/wgc/98/) is a good reference to the gene content of chromosome arm 1BS. The ability to create a BAC library from a particular chromosome arm opens avenues for the development of a physical map of the complex hexaploid wheat genome and its analysis in a targeted and step-wise manner, working with one particular chromosome arm at a time. In addition to simplifying the physical mapping efforts by several orders of magnitude, this chromosome-based approach has the potential for division of labor during sequencing of the wheat genome (Gill et al., 2004).
Results
The short arm of chromosome 1B (1BS) was purified from a di-telosomic line of bread wheat cultivar ‘Pavon 76’. The work involved preparation of liquid suspensions of intact chromosomes from synchronized root tip meristems, staining with a DNA-specific fluorescent dye 4¢,6-diamidi-no-2-phenylindole (DAPI), and analysis of the fluorescence intensity of stained chromosomes using flow cytometry. We obtained a reproducible distribution of relative fluor-escence intensity (‘flow karyotype’), which was
character-ized by five peaks (Figure 1). We confirmed the
chromosome content of each peak by sorting particles from each peak onto microscope slides and performing fluorescence in situ hybridization (FISH) with probes for GAA microsatellites and the Afa repeat. The two probes
facilitated identification of any chromosome arm in hexaploid wheat (Pedersen and Langridge, 1997). We established that peaks I–III were composites of various chromosomes and that the small peak to the right of peak III represented chromosome 3B (Kubala´kova´ et al., 2002; Vra´na et al., 2000). The peak on the far left was found to represent the short arm of chromosome 1B.
Subsequently, we sorted six million copies of chromo-some arm 1BS representing approximately 4 lg DNA in batches of 105and embedded them in 60 agarose plugs. The sorting exercise took 27 working days, and approximately
7000 seeds were required to prepare 267 samples of chro-mosome suspensions. We repeatedly checked the purity of the flow-sorted fractions by microscopic observation of chromosomes sorted on a glass slide and subjected to FISH with probes for GAA microsatellites and Afa repeats. On average, the sorted fractions consisted of 89% chromosome 1BS. Contaminating particles were arms and chromatids of other chromosomes without an apparent prevalence of specific types. Pulsed-field gel electrophoresis showed that the DNA of sorted chromosome arms was intact, indicating its usefulness for BAC library construction (data not shown). The DNA of sorted chromosomes was partially digested with HindIII and used to prepare an 1BS-specific BAC library. The complete 1BS-specific BAC library (TA-1BS) consists of five sub-libraries created from five independent ligation reactions using different size classes of DNA: H01, 50–75 kb; H02, 75–100 kb; H1, 100–150 kb; H2, 150–200 kb; H3, 200– 250 kb. We estimated the average insert size of the library after the analysis of 96 BAC clones selected randomly from the five sub-libraries (Figure 2). The H01 sub-library has 9600 clones and an average insert size of 62 kb, and is similar to the H02 sub-library with 23 808 clones and average insert size 66 kb. Sub-library H1 is the largest, with 27 264 clones and an insert size of 96 kb, while sub-library H2 is the smallest sub-library, with only 384 clones and an average insert size of 110 kb. Finally, the H3 sub-library has 4224 clones and its average insert size is 118 kb. The complete 1BS-specific BAC library comprises 65 280 clones ordered in 170 384-well plates with an average insert size of 82 kb. Almost a half of the TA-1BS library (45%) has inserts larger than 100 kb, while 18% of inserts are 75–100 kb, and 37% of inserts are shorter than 75 kb (data not shown).
Taking into account the wheat genome size of
16 974 Mb 1C)1(Bennett and Smith, 1991) and the relative
I
II III
3B 1BS
Relative fluorescence intensity
Number o
f events
Figure 1. Histogram of relative fluorescence intensity (‘flow karyotype’) obtained after flow cytometric analysis of DAPI-stained suspensions of chromosomes prepared from a 1BS di-telosomic line of wheat cultivar ‘Pavon 76’. The flow karyotype consists of three composite peaks (I, II, and III) representing specific groups of chromosomes, and a peak representing chromosome 3B. In addition, a peak representing the short arm of chromo-some 1B (1BS) is clearly discriminated.
BAC clones M M –7.5 –48.5 –97 –145.5 kb
Figure 2. Insert size analysis of 22 randomly chosen clones from the TA-1BS BAC library (H1 sub-library).
BAC DNA was digested with NotI to release the insert, separated by pulsed-field gel electrophor-esis and stained with ethidium bromide. Lanes M contain a k PFGE marker. The 7.5 kb band represents the BAC vector.
chromosome lengths given by Gill et al. (1991), we estimate the molecular size of the short arm of chromosome 1BS to be 315 Mb. Thus, this chromosome arm represents only 1.9% of the hexaploid wheat genome. Given the number of clones, the average insert size and the frequency of con-taminating particles, this library represents 14.5 size equiv-alents of 1BS. Using the formula of Clarke and Carbon (1976), we estimated the probability of recovering any DNA sequence present on chromosome 1BS in the library to be 99.9%. On the other hand, the probability of finding any DNA sequence from other chromosomes is as low as 4.5%.
To confirm the chromosome arm specificity and verify the 14.5-fold genome coverage, we screened the TA-1BS library by PCR using a set of microsatellite (SSR) markers. To do this, we divided the library into 170 pools; each pool representing one 384-well plate. Then we screened the pools with nine SSR markers specific for the short arm of chromosome 1B. The number of pools that gave PCR products of the expected lengths with each SSR marker ranged from 10 to 21 (Table 1). Based on these results, we estimated the coverage of the TA-1BS BAC library to be at least 15.2-fold.
Cytogenetic mapping is needed to support the develop-ment of physical contig maps by anchoring BAC contigs onto chromosomes. In order to test the usefulness of our new BAC library for cytogenetic mapping, we hybridized 75 randomly selected ‘low-copy’ BAC clones to wheat chromo-somes by FISH. Only four of them (5.3%) localized exclu-sively on 1BS. Another BAC clone (4B9) gave a distinct signal on the satellite of 1BS and on chromosome arms 5AS, 5BS, 1DS and 5DS. Five BAC clones gave FISH patterns similar to that of repetitive DNA family pSc119 (Bedbrook et al., 1980), while FISH with four BAC clones resulted in a distinct signal on 1BS and dispersed signals on all wheat chromosomes. Fifty-two BAC clones hybridized to all wheat chromosomes along their length, indicating the presence of interspersed repeated sequences. The remaining nine BAC clones did not yield any hybridization signals.
With the aim of overcoming the difficulties observed with BAC-FISH, we explored a possibility of using ‘low-copy’ subclones for FISH. Out of the 75 BAC clones tested, eight BAC clones with different FISH patterns were chosen for subcloning (Table 2). This strategy provided shotgun sub-clones that hybridized exclusively to unique loci on 1BS. FISH with subclones selected from BAC clones 77E3, 79P7 and 81P18, which hybridized to a single locus on 1BS and multiple loci on all chromosomes, resulted in discrete signals on 1BS only. To some measure, the subcloning strategy was also successful with BAC clones that hybridized evenly over the whole chromosome complement. We suc-ceeded in selecting ‘low-copy’ subclones hybridizing exclu-sively to specific loci on 1BS from two such BAC clones (78M16 and 81L2). However, subclones obtained from two other BAC clones (80E19, 80M16) gave only dispersed hybridization signals. Finally, FISH with two subclones (4B9-A9 and 4B9-F6) resulted in the same hybridization pattern as FISH with the complete BAC 4B9. Representative examples of FISH with BAC clones and subclones are shown in Figure 3.
Selected ‘low-copy’ subclones that gave a range of FISH patterns were completely sequenced (Table 3). DNA se-quences were assembled and edited using BioEdit software (http://www.mbio.ncsu.edu/BioEdit/bioedit.html), and fur-ther analyzed using Tandem Repeats Finder (TRF; Benson, 1999) and JDotter (Brodie et al., 2003). All sequences were successfully assembled, except subclones 4B9-A9 and 78M16-G18, for which TRF and JDotter showed the presence of duplicated and tandem organized regions. In case of 4B9-A9, this result could provide an explanation for the FISH pattern obtained (Figure 3). A homology search was per-formed against the sequences deposited in GenBank, and against the Triticeae Repeat Sequence Database (TREP; http://wheat.pw.usda.gov/ITMI/Repeats/index.shtml). Most of the subclones were found to carry various parts of repetitive elements, the most frequent being various parts of retrotransposons (Table 3). No homology was found for subclones 81P18-B18 and 77E3-E15.
The cytogenetically mapped subclones were then used to test the feasibility of developing chromosome sequence tags. Based on the nucleotide sequences of BAC subclones, we designed primers for PCR and tested their specificity on flow-sorted chromosome arm 1BS, genomic DNA of ‘Pavon 76’, and di-telosomic line 1BL of ‘Pavon 76’. The primers designed for 81P18-B18 and 77E3-F15 subclones were 1BS-specific and gave one PCR product of about 1000 bp (Figure 4).
Discussion
In this paper, we report an important step forward in developing a new generation of genomic resources for bread wheat. We demonstrate that it is possible to create a Table 1 Results of TA-1BS library screening on pools of 384-well
plates with 1BS-specific SSR markers SSR marker
Number of positive pools Marker code Reference
gwm 18 a 21 gwm 264 a 14 gwm 413 a 17 gwm 550 a 12 gpw 2067 b 10 gpw 3122 b 13 gpw 4069 b 14 gpw 7059 b 21 gpw 7070 b 15 Average 15.2 a Ro¨der et al. (1998). b
BAC library from a specific chromosome arm. The TA-1BS library is specific for the short arm of chromosome 1B, and is the third sub-genomic BAC library produced from individual plant chromosomes, after the composite BAC library specific
for wheat chromosomes 1D, 4D and 6D (Janda et al., 2004) and a BAC library from wheat chromosome 3B (Sˇafa´rˇ et al., 2004). In addition to creating a unique resource for wheat genomics, the current work confirms that our procedure for Table 2 Patterns of hybridization to wheat chromosomes observed after FISH with probes for putative ‘low-copy’ BAC clones selected from the 1BS-specific BAC library, and for their ‘low-copy’ shotgun subclones
BAC clones BAC subclones
BAC address
Insert
size (kb) Hybridization pattern
Subclone number
Insert
size (bp) Hybridization pattern
4B9 75 Single locus on 1BS, 5AS, 5BS, 1DS and 5DS A9 2700 Single locus on 1BS, 5AS,
5BS, 1DS and 5DS
F6 2500 Single locus on 1BS, 5AS,
5BS, 1DS and 5DS
77E3 55 Single locus on 1BS and weakly
dispersed on all chromosomes
E15 2300 Single locus on 1BS
F15 2700 Single locus on 1BS
O23 2500 Dispersed on all chromosomes
78M16 70 Dispersed on all chromosomes A1 4900 No signal
C22 1500 No signal
G18 1900 Single locus on 1BS
79P7 45 Single locus on 1BS and weakly
dispersed on all chromosomes
B1 2500 Single locus on 1BS
80E19 50 Dispersed on all chromosomes B1 2800 Dispersed on all chromosomes
80M16 60 Dispersed on all chromosomes N3 3200 Dispersed on all chromosomes
O22 1200 No signal
81L2 35 Dispersed on all chromosomes A1 3000 Dispersed on all chromosomes
C3 2000 No signal
E17 2900 No signal
N3 2900 Single locus on 1BS
81P18 60 Single locus on 1BS and weakly dispersed on all chromosomes
B14 1000 No signal
B18 2500 Single locus on 1BS
C3 3500 Single locus on 1BS
Figure 3. Fluorescence in situ hybridization (FISH) on the short arm of wheat chromosome 1B (1BS) using three BAC clones that give three different types of hybridization pattern, and three subclones derived from them.
BAC 81L2 (left) and its subclone 81L2-A1 hybridize to the entire length of 1BS except the NOR region. BAC 4B9 (centre) and its subclone 4B9-A9 hybridize to a single locus on 1BS. In contrast, BAC 81P18 hybridizes strongly to a single locus and weakly along the entire length of 1BS, while its subclone 81P18-B18 maps to a single locus on 1BS. The probes were labeled by digoxigenin and the sites of probe hybridization were detected using anti-digoxigenin–FITC (yellow). The chromosomes were counterstained by DAPI (shown in red pseudocolor). Three representative examples of hybridizations on 1BS are given.
the construction of large-insert DNA libraries from DNA of flow-sorted chromosomes is reproducible, reliable and effi-cient. This advance, together with the availability of a com-plete series of telosomic lines in wheat (Sears, 1954) and the possibility of purifying wheat telosomes by flow cytometry (Kubala´kova´ et al., 2002), make it possible to create a BAC library from every chromosome arm of wheat.
The TA-1BS library consists of five sub-libraries differing in the number of clones and average insert sizes. These features offer considerable flexibility in the use of the library. Sub-libraries H1, H2 and H3, with an average insert size of about 100 kb and tenfold 1BS coverage, could be used preferentially for contig assembly and chromosome walk-ing. Sub-libraries H01 and H02, which contain clones with smaller inserts of around 65 kb, represent useful resources for FISH mapping. We have confirmed the representative coverage of the short arm of chromosome 1B by library screening with SSR markers that map to 1BS. The 15.2-fold 1BS coverage, as determined by screening with SSR mark-ers, suggests that our estimate of clone representation using average insert size (14.5-fold) is accurate.
Compared with other genomic and sub-genomic BAC resources for hexaploid wheat, the TA-1BS library has a comparable average insert size and the highest genome coverage. Liu et al. (2000) reported on the construction of a TAC (transformation-competent artificial chromosome) lib-rary with an average insert size of 54 kb and 3.07-fold genome coverage. A genomic BAC library prepared by Nilmalgoda et al. (2003) comprises 6.5· 105
clones, an average insert size of 79 kb, and 3.1-fold genome coverage. Table 3 DNA sequences obtained from ‘low-copy’ BAC subclones
Subclone address Sequence length (bp) GenBank accession number
Homology with known DNA sequences
GenBank TREP DNA sequence GenBank accession number Smallest sum
probability DNA sequence
TREP accession number Smallest sum probability 4B9-A9 1334* DX572372 Gypsy-like retrotransposon AY494981 8e)113 Retrotransposon,
gypsy
(Laura – 719C13)
TREP1238 9e)31
77E3-E15 4079 DX572368
77E3-F15 2787 DX572367 Putative non-LTR retroelement reverse transcriptase
NP922330 3e)28
77E3-O23 2549 DX572366 Transposon protein, CACTA ABA98219 3e)10 Transposon, CACTA TREP746 2e)22 78M16-G18m13r 590* DX572369 Germin-like protein 4 AAT67049 4e)14
78M16-G18r 718* DX572370 Germin-like protein 4 AAT67049 4e)14 78M16-G18m13f 642* DX572371 81L2-A1 2376 DX572365 Retrotransposon, LTR, gypsy TREP821 e)157 81P18-B18 2321 DX572364 81P18-C3 2513 DX572363 Unclassified TREP1218 6e)20
*Subclone could not be completely sequenced and/or assembled.
500– bp
1BS
M BAC Dt1BL Pavon
Figure 4. Agarose gel electrophoresis of PCR products obtained with primers (5¢-TCTCCTCCCATGTAAGCTCTG-3¢ and 5¢-CTAGAAGGATGCCGAGGATG-3¢) derived from BAC subclone 81P18-B18.
The reaction was performed on flow-sorted short arm of chromosome 1B (1BS), BAC clone 81P18 (BAC), genomic DNA prepared from a line of wheat cultivar ‘Pavon 76’ di-telosomic for 1BL (Dt1BL) and genomic DNA of ‘Pavon 76’. The absence of a product on Dt1BL and its presence on 1BS indicate specificity of the primers for the short arm of 1B. Lane M,100 bp DNA Ladder Plus marker (Fermentas, Vilnius, Lithuana).
Another published BAC library from hexaploid wheat was created by Allouis et al. (2003). The library consists of 1.2· 106clones, with an average insert size of 130 kb, and
represents 9.3 haploid genome equivalents.
A single but serious disadvantage of genomic libraries from hexaploid wheat is their large size, which makes their screening laborious and expensive. Maintenance and hand-ling of genomic libraries with approximately 106BAC clones remains a non-trivial task (Allouis et al., 2003). However, sub-genomic BAC libraries offer a solution to this problem. Our composite BAC library specific for chromosomes 1D, 4D and 6D, which consists of only 87 168 clones with an average insert size of 85 kb, represents at least 3.4-fold clone coverage (Janda et al., 2004), while the BAC library specific for chromosome 3B, which comprises only 67 968 clones with an insert size of 103 kb, represents at least 6.2 equiv-alents of the chromosome (Sˇafa´rˇ et al., 2004). The TA-1BS library described in this paper comprises 65 280 clones with an average insert size of 82 kb, and represents 14.5 size equivalents of the short arm of chromosome 1B.
In addition to chromosome coverage, specificity is an important quality parameter for chromosome- and chromo-some arm-specific BAC libraries. To ensure that our library is specific for chromosome 1BS, we have periodically checked the flow-sorted fractions for contamination. The average content of other chromosome arms and chromatids was 11% as determined by FISH using probes for the GAA microsatellite and the Afa repeat. This FISH approach offers unambiguous identification of all chromosome arms in hexaploid wheat (Kubala´kova´ et al., 2002). Our previous studies (Janda et al., 2004; Sˇafa´rˇ et al., 2004) demonstrated that FISH analysis of sorted chromosome fractions provided the most detailed and reliable data on the purity in sorted fractions and the extent of contamination by other chromo-somes. Hence, we did not consider it useful to check the TA-1BS library for contamination using PCR with markers specific for other wheat chromosomes.
DNA libraries cloned in a BAC vector are a favorable genomic resource for use in map-based cloning. The availability of a BAC library specific for the short arm of 1B, which represents only 1.9% of the bread wheat genome, should accelerate cloning of important genes in wheat. Preliminary results obtained with our BAC library specific for chromosome 3B (Sˇafa´rˇ et al., 2004) confirm these expecta-tions. The small size and the specificity of the library greatly facilitated map-based cloning of major QTL for Fusarium head blight resistance (Liu et al., 2005) and a durable rust-resistance gene (Kota et al., 2006).
Whole-genome physical maps provide a basis for genome sequencing and will be needed whatever the sequencing method employed (Meyers et al., 2004). However, the genome of hexaploid wheat is so complex that the devel-opment of a whole-genome physical map seems out of reach of the current technology (Jan Dvorˇa´k, University of
California, Davis, USA, personal communication). Chromo-some- and chromosome arm-specific BAC libraries provide a solution to this problem. As they represent only a small fraction of the whole genome, sufficient coverage of a particular chromosome can be achieved with about 50 000 BAC clones. Using a suitable high-throughput method, a library of this size can be fingerprinted in a few months (Luo et al., 2003). Furthermore, the reduced complexity should greatly simplify the assembly of BAC contigs. Recent results on the production of a physical contig map of wheat chro-mosome 3B (Paux et al., 2006) fully support these assump-tions. The ability to create a BAC library from a particular chromosome arm opens avenues for the development of a physical map of the complex hexaploid wheat genome and its analysis in a targeted and step-wise manner, working with a particular chromosome arm at a time. In addition to simplifying the physical mapping efforts by at least two orders of magnitude, the chromosome-based approach has the potential for a division of labor (Gill et al., 2004).
The development of physical contig maps requires support from cytogenetic mapping to confirm the precise physical positions of centromeres, to determine the distances from the actual chromosome termini of markers near the termini of linkage maps, to resolve the order of contigs and clones, and to estimate the numbers and sizes of gaps (Harper and Cande, 2000). Our results demonstrate that mapping complete BAC clones to specific loci in wheat may be difficult due to the presence of repetitive DNA sequences that hybridize throughout the genome. This conclusion is supported by the observations of Zhang et al. (2004). In this paper, we show that shotgun subcloning and the use of ‘low-copy’ subclones may overcome this problem. Our success in the localization of clones shorter than 2 kb was made possible by performing FISH on flow-sorted chromosomes that are free of cell wall and cytoplasmic remnants. Because thousands of somes can be sorted on one slide, FISH on sorted chromo-somes offers higher throughput, higher sensitivity and higher resolution when compared to mitotic metaphase spreads (Vala´rik et al., 2004). Hence, flow-based cytogenetic mapping could play an important role in the development of a sequence-ready physical map of bread wheat.
In conclusion, this paper demonstrates that high-quality sub-genomic BAC resources representing only few per cent of the whole genome can be prepared for large-genome species. While our two previous studies utilized flow-sorted chromosomes to construct sub-genomic BAC libraries (Jan-da et al., 2004; Sˇafa´rˇ et al., 2004), this study marks a fundamental advance and reports on the creation of a large-insert library specific for a particular chromosome arm. We envisage that the use of chromosome arm-specific BAC resources will accelerate the development of sequence-ready physical contig maps and gene cloning, as well as comparative analyses aiming at revealing the genome changes accompanying the evolution of polyploid wheat.
Experimental procedures Plant material
A di-telosomic line of hexaploid wheat Triticum aestivum L. cv. ‘Pa-von 76’ (2n¼ 40 þ 2t1BS) carrying the short arm of chromosome 1B (1BS) in the form of a telosome was used. Seeds were germinated in the dark at 25 0.5C on moistened filter paper for 3 days to obtain a root length around 2–3 cm. Approximately 7000 seeds were germi-nated in batches of 25–30 for the preparation of 267 chromosome suspensions which were further used for flow sorting.
Preparation of chromosome suspensions and flow-cytometric sorting
Cell-cycle synchronization and the preparation of suspensions of intact chromosomes were performed according to the method described by Vra´na et al. (2000). Briefly, each of the 267 samples of chromosome suspension was prepared by mechanical homoge-nization of 25 formaldehyde-fixed meristem root tips in 1 ml ice-cold isolation buffer (Sˇimkova´ et al., 2003). Chromosomes in sus-pension were stained with 2 lg ml)1DAPI (4¢,6-diamidino-2-phen-ylindole) and analyzed using a FACSVantage flow cytometer (Becton Dickinson, San Jose´, CA, USA) equipped with a UV laser set to multi-line UV and running at 300 mW output power. In order to sort the chromosome 1BS, the sort window was set on a dot plot of fluorescence pulse area versus fluorescence pulse width. The chromosome arm was sorted in aliquots of 1· 105
into 160 ll of 1.5x isolation buffer. The purity of sorted fractions was checked regularly by sorting 2000 particles into a 15 ll drop of PRINS buffer (Kubala´kova´ et al., 2001) supplemented with 5% sucrose on a microscope slide. After air-drying, sorted chromosomes were identified using fluorescence in situ hybridization (FISH) with probes for the GAA microsatellite and Afa repeat, which give chromosome-specific fluorescent labeling patterns (Kubala´kova´ et al., 2002).
BAC library construction
Preparation of high-molecular-weight DNA and BAC library con-struction were performed as described previously (Chalhoub et al., 2004; Janda et al., 2004; Sˇafa´rˇ et al., 2004). Briefly, each batch of 105 flow-sorted 1BS chromosomes was pelleted (200 g), and the chro-mosomes were embedded in 15 ll of low-melting-point agarose (0.8% w/v). In total, 60 agarose mini-plugs representing 6· 106 sorted chromosomes were prepared and used for DNA cloning. Chromosomal DNA was partially digested with HindIII and size-selected by pulsed-field gel electrophoresis (PFGE) in 0.25x TBE buffer at 6 V cm)1, with a 1–40 sec switch time ramp, angle 120, for 12 h, and with a 2.5–5.5 sec switch time ramp, angle 120, for 6 h at 14C. Five regions of the gel, corresponding to approximately 50– 75 kb (H01 fraction), 75–100 kb (H02 fraction), 100–150 kb (H1 frac-tion), 150–200 kb (H2 fraction) and 200–250 kb (H3 fraction) were excised, the high-molecular-weight DNA was electro-eluted and ligated into a dephosphorylated vector pIndigoBAC5 (Caltech, Pa-sadena, CA, USA), which was prepared according to the method described by Chalhoub et al. (2004). Ligation mixtures were incu-bated at 16C, and, after de-salting, transformed into Escherichia coli ElectroMAX DH10B competent cells (Gibco BRL, Gaithersburg, MD, USA) by electroporation. The library was ordered in 384-well microtitre plates filled with freezing medium (Woo et al., 1994). The plates were incubated at 37C overnight, and stored at)80C after replication.
Estimation of the insert size of BAC clones
Ninety-six BAC clones were randomly selected from all five liga-tions (H01–H3), and incubated overnight at 37C in 1.5 ml of 2YT medium (Sambrook and Russel, 2001) containing 12.5 lg ml)1 chloramphenicol. BAC DNA was extracted by a standard alkaline lysis method and digested with NotI restriction endonuclease (0.25 U/20 ll). PFGE was used to separate DNA fragments in 1% agarose gel with 0.5x TBE buffer at 6 V cm)1, with a 1–40 sec switch time ramp, angle 120, for 14 h at 14C. The insert sizes were esti-mated after comparison with a lambda size standard run in the same gel.
Screening the library with 1BS-specific probes
After growing overnight in 2YT medium supplemented with chlo-ramphenicol (12.5 lg ml)1) in 384-well plates, BAC clones from each of the 384-well plates were pooled, pelleted and resuspended in 4 ml TE buffer (10 mMTris, 1 mMEDTA). Bacterial suspensions were lysed at 95C for 30 min, pelleted at 3000 g for 60 min, and the supernatant was diluted 25-fold using deionized water for PCR reactions. The reaction mix (10 ll) consisted of 2 ll template DNA, 1x PCR buffer, 1.5 mMMgCl2, 0.2 mMdNTPs, 1.25 lMprimers and 0.5 U AmpliTaq DNA polymerase (Roche, Mannheim, Germany). The PCR reaction was performed as follows: nine cycles consisting of 5 min at 94C, 12 sec at 65C and 18 sec at 72C, and 35 cycles of 12 sec at 94C, 12 sec at 55C and 18 sec at 72C, and a final extension at 72C for 7 min. The presence of PCR products was checked by electrophoresis on 4% MetaPhor agarose gel (FMC Bioproducts, Philadelphia, PA, USA).
Selection of ‘low-copy’ BAC clones
A total of 6144 BAC clones from the H01 sub-library (sixteen 384-well plates) were spotted onto two 8· 12 cm Hybond Nþ filters (AP Biotech, Piscataway, NJ, USA). The filters were hybridized with wheat genomic DNA labeled with alkaline phosphatase using Alk-Phos Direct (Amersham Biosciences, Little Chalfont, Buckingham-shire, UK) to reveal clones potentially containing a low amount of repetitive DNA sequences. Then the filters were screened with a probe for BAC vector to avoid the selection of non-growing clones. Seventy-five ‘low-copy’ BAC clones showing no or weak hybrid-ization signals were tested as BAC-FISH, and eight of them were selected for subcloning. In addition, they were localized on flow-sorted wheat chromosomes using FISH as described previously (Sˇafa´rˇ et al., 2004). No blocking DNA was used in the hybridization mix for FISH.
Subcloning of ‘low-copy’ BAC clones
DNA of ‘low-copy’ BAC clones was isolated using the Large-Con-struct Kit (Qiagen, Valencia, CA, USA), and physically fragmented using a HydroShear DNA shearing device (GeneMachines, San Carlos, CA, USA). Fragments of 3–5 kb were ligated into pCR-XL-TOPO vector (Invitrogen Life Technologies, Carlsbad, CA, USA). Ligation mixtures were transformed into One-Shot TOP10 E. coli (Invitrogen). A total of 384 subclones from each of the eight BAC clones were ordered in 384-well plates filled with freezing medium (Woo et al., 1994), incubated at 37C overnight, and stored at)80C. All 3072 BAC subclones were spotted onto one 8· 12 cm Hybond Nþ filter and screened with labeled genomic DNA as des-cribed above. One to three subclones from each BAC showing a
weak signal after hybridization with labeled genomic DNA were randomly chosen for mapping using FISH.
Fluorescence in situ hybridization
Preparation of digoxigenin- or biotin-labeled probes and FISH on flow-sorted chromosomes were performed as described by Sˇafa´rˇ et al. (2004). The sites of probe hybridization were detected using anti-digoxigenin–fluorescein isothiocyanate and streptavidin-Cy3, and the chromosomes were counterstained with DAPI or propidium iodide. The preparations were evaluated using a Olympus BX60 microscope (Olympus, Tokyo, Japan) with a CCD camera interfaced to a PC running ISIS software (Metasystems, Altlussheim, Germany).
Sequence analysis of ‘low-copy’ BAC subclones
Six BAC subclones that were successfully localized by FISH to single loci on the short arm of chromosome 1B and two that gave hybridization patterns dispersed on all chromosomes were ran-domly chosen for sequencing. The sequencing was performed by Agowa (Berlin, Germany). The sequenced parts of the subclones were assembled in the Chromas software (Griffith University, Southport, Qld, Australia), edited using BioEdit (http://www. mbio.ncsu.edu/BioEdit/bioedit.html), and the assembled sequences were deposited in GenBank (GSS database). They were then sear-ched for similarities to known sequences in GenBank usingBLASTN
(Altschul et al., 1997), and tested for the presence of repetitive units using JDotter (Brodie et al., 2003) and Tandem Repeat Finder (Benson, 1999). Finally, the sequences of subclones were used to design primers for internal sequence amplification to verify their specificity for 1BS. PCR was performed as described by Vra´na et al. (2000) on flow-sorted chromosome arm 1BS, BAC DNA, genomic DNA of a di-telosomic line of cv. ‘Pavon 76’ carrying the long arm of 1B (Dt1BL), and genomic DNA of cv. ‘Pavon 76’.
Acknowledgements
We are grateful to our colleagues Arnaud Bellec, Michaela Lib-osva´rova´, Radka Tusˇkova´ and Jitka Weiserova´ for excellent techni-cal assistance. This work was supported by research grants from the Czech Science Foundation (grant awards 521/04/0607, 521/05/P257 and 521/05/H013), the Ministry of Education, Youth and Sports of the Czech Republic (grant award LC06004), research grants from the French Ministry of Research and Agriculture, and by a collaborative project ‘Barrande’ (2003-035-2) between the Ministry of Education, Youth and Sports of the Czech Republic and the French Ministry of Foreign Affairs and Education. J.J. acknowledges the receipt of a Marie Curie PhD student Fellowship at URGV-INRA.
References
Akhunov, E.D., Akhunova, A.R., Linkiewicz, A.M., et al. (2003) Synteny perturbations between wheat homoeologous chromo-somes caused by locus duplications and deletions correlate with recombination rates. Proc. Natl Acad. Sci. USA, 100, 10836–10841. Akhunov, E.D., Akhunova, A.R. and Dvorak, J. (2005) BAC libraries of Triticum urartu, Aegilops speltoides and Ae. tauschii, the diploid ancestors of polyploid wheat. Theor. Appl. Genet. 111, 1617–1622. Allouis, S., Moore, G., Bellec, A. et al. (2003) Construction and characterisation of a hexaploid wheat (Triticum aestivum L.) BAC library from the reference germplasm ‘Chinese Spring’. Cereals Res. Commun. 31, 331–338.
Altschul, S.F., Madden, T.L., Scha¨ffer, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.J. (1997) GappedBLASTandPSI-BLAST: a
new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402.
Anderson, O.D., Rausch, C., Moullet, O. and Lagudah, E.S. (2003) The wheat D-genome HMW-glutenin locus: BAC sequencing, gene distribution, and retrotransposon clusters. Funct. Integr. Genomics, 3, 56–68.
Bedbrook, J.R., Jones, J., Odell, M., Thompson, R. and Flavell, R.B. (1980) A molecular description of telomeric heterochromatin in Secale species. Cell, 19, 549–560.
Bennett, M.D. and Smith, J.B. (1991) Nuclear DNA amounts in an-giosperms. Proc. R. Soc. Lond. (Biol.) 334, 309–345.
Benson, G. (1999) Tandem repeat finger a program to analyze DNA sequences. Nucleic Acids Res. 27, 573–580.
Brodie, R., Roper, R.L., Upton, C. and Dotter, J. (2003) A Java interface to multiple Dotplots generated by dotter. Bioinformat-ics, 20, 279–281.
Chalhoub, B., Belcram, H. and Caboche, M. (2004) Efficient cloning of plant genomes into bacterial artificial chromosome (BAC) lib-raries with larger and more uniform insert size. Plant Biotechnol. J. 2, 181–188.
Chantret, N., Salse, J., Sabot, F. et al. (2005) Molecular basis of evolutionary events that shaped the hardness locus in diploid and polyploid wheat species (Triticum and Aegilops). Plant Cell, 17, 1033–1045.
Clarke, L. and Carbon, J. (1976) A colony bank containing synthetic Col El hybrid plasmids representative of the entire E. coli gen-ome. Cell, 9, 91–99.
Dvorak, J., Yang, Z.L., You, F.M. and Luo, M.C. (2004) Deletion polymorphism in wheat chromosome regions with contrasting recombination rates. Genetics, 168, 1665–1675.
Erayman, M., Sandhu, D., Sidhu, D., Dilbirligi, M., Baenziger, P.S. and Gill, K.S. (2004) Demarcating the gene-rich regions of the wheat genome. Nucleic Acids Res. 32, 3546–3565.
Gill, B.S., Friebe, B. and Endo, T.R. (1991) Standard karyotype and nomenclature system for description of chromosome bands and structural aberrations in wheat (Triticum aestivum). Genome, 34, 830–839.
Gill, B.S., Appels, R., Botha-Oberholster, A.M. et al. (2004) A workshop report on wheat genome sequencing: International Genome Research on Wheat Consortium. Genetics, 168, 1087– 1096.
Goff, S.A., Ricke, D., Lan, T.H. et al. (2002) A draft sequence of the rice genome (Oryza sativa L. ssp japonica). Science, 296, 92–100. Harper, L.C. and Cande, W.Z. (2000) Mapping a new frontier; development of integrated cytogenetic maps in plants. Funct. Integr. Genomics 1, 89–98.
Janda, J., Bartosˇ, J., Sˇafa´rˇ, J. et al. (2004) Construction of a sub-genomic BAC library specific for chromosomes 1D, 4D and 6D of hexaploid wheat. Theor. Appl. Genet. 109, 1337–1345.
Kong, X.Y., Gu, Y.Q., You, F.M., Dubcovsky, J. and Anderson, O.D. (2004) Dynamics of the evolution of orthologous and paralogous portions of a complex locus region of two genomes of allopol-yploid wheat. Plant. Mol. Biol. 54, 55–69.
Kota, R., Spielmeyer, W., Dolezˇel, J., Sˇafa´rˇ, J., Paux, E., McIntosh, R.A. and Lagudah, E.S. (2006) Towards isolating the durable stem rust resistance gene Sr2 from hexaploid wheat (Triticum aesti-vum L.). In Abstracts of the International Conference ‘Plant and Animal Genome XIV’. San Diego, CA: Sherago International, Inc., p. 181.
Kubala´kova´, M., Vra´na, J., Cˇı´halı´kova´, J., Lysa´k, M.A. and Dolezˇel, J. (2001) Localisation of DNA sequences on plant chromosomes using PRINS and C-PRINS. Methods Cell Sci. 23, 71–82.
Kubala´kova´, M., Vra´na, J., Cˇı´halı´kova´, J., Sˇimkova´, H. and Dolezˇel, J. (2002) Flow karyotyping and chromosome sorting in bread wheat (Triticum aestivum L.). Theor. Appl. Genet. 104, 1362–1372. Kubala´kova´, M., Kova´rˇova´, P., Sucha´nkova´, P., Cˇı´halı´kova´, J., Bar-tosˇ, J., Lucretti, S., Watanabe, N., Kianian, S.F. and Dolezˇel, J. (2005) Chromosome sorting in tetraploid wheat and its potential for genome analysis. Genetics, 170, 823–829.
Lijavetzky, D., Muzzi, G., Wicker, T., Keller, B., Wing, R. and Dub-covsky, J. (1999) Construction and characterization of a bacterial artificial chromosome (BAC) library for the A genome of wheat. Genome, 42, 1176–1182.
Liu, Y.G., Nagaki, K., Fujita, M., Kawaura, K., Uozumi, M. and Ogi-hara, Y. (2000) Development of an efficient maintenance and screening system for large-insert genomic DNA libraries of hexaploid wheat in a transformation-competent artificial chro-mosome (TAC) vector. Plant J. 23, 687–695.
Liu, S., Pumphrey, M.O., Zhang, X., Gill, B.S., Stack, R.W., Gill, J.S., Dolezˇel, J., Chalhoub, B. and Anderson, J. (2005) Toward posi-tional cloning of Qfhs.ndsu-3BS, a major QTL for Fusarium head blight resistance in wheat. In Abstracts of the International Con-ference ‘Plant and Animal Genome XIII’. San Diego, CA: Sherago International, Inc., p. 71.
Luo, M.C., Thomas, C., You, F.M., Hsiao, J., Ouyang, S., Buell, C.R., Malando, M., McGuire, P.E., Anderson, O.D. and Dvorak, J. (2003) High-throughput fingerprinting of bacterial chromosomes using the SNaPshot labeling kit and sizing of restriction fragments by capillary electrophoresis. Genomics, 82, 378–389.
Ma, Z., Weining, S., Sharp, P.J. and Liu, C. (2000) Non-gridded lib-rary: a new approach for BAC (bacterial artificial chromosome) exploitation in hexaploid wheat (Triticum aestivum). Nucleic Acids Res. 28, 106–111.
Meyers, B.C., Scalabrin, S. and Morgante, M. (2004) Mapping and sequencing complex genomes: let’s get physical! Nat. Rev. Genet. 5, 578–581.
Moullet, O., Zhang, H.B. and Lagudah, E.S. (1999) Construction and characterisation of a large DNA insert library from the D genome of wheat. Theor. Appl. Genet. 99, 305–313.
Nilmalgoda, S.D., Cloutier, S. and Walichnowski, A.Z. (2003) Con-struction and characterization of a bacterial arteficial chromo-some (BAC) library of hexaploid wheat (Triticum aestivum L.) and validation of genome coverage using locus-specific primers. Genome, 46, 870–878.
Paux, E., Sourdille, P., Salse, J. et al. (2006) Chromosome based strategies to decipher the hexaploid wheat genome:
chromo-some 3B, a case study. In Abstracts of the International Confer-ence ‘Plant and Animal Genome XIV’. San Diego, CA: Sherago International, Inc., p. 18.
Pedersen, C. and Langridge, P. (1997) Identification of the entire chromosome complement of bread wheat by two-colour FISH. Genome, 40, 589–593.
Ro¨der, M.S., Korsun, V., Wendehake, K., Plaschke, J., Tixier, M.H., Leroy, P. and Ganal, M.W. (1998) A microsatellite map of the wheat genome. Genetics, 149, 2007–2023.
Sˇafa´rˇ, J., Bartosˇ, J., Janda, J. et al. (2004) Dissecting large and complex genomes: flow sorting and BAC cloning of individual chromosomes from bread wheat. Plant J. 39, 960–968.
Sambrook, J. and Russel, D.W. (2001) Molecular Cloning: A Labor-atory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
Sandhu, D., Champoux, J.A., Bondareva, S.N. and Gill, K.S. (2001) Identification and physical localization of useful genes and markers to a major gene-rich region on wheat group 1S chro-mosomes. Genetics, 157, 1735–1747.
Sears, E.R. (1954) The aneuploids of common wheat. Univ. Missouri Res. Bull. 572, 1–58.
Sˇimkova´, H., Cˇı´halı´kova´, J., Vra´na, J., Lysa´k, M.A. and Dolezˇel, J. (2003) Preparation of HMW DNA from plant nuclei and chromo-somes isolated from root tips. Biol. Plant. 46, 369–373.
Smith, D.B. and Flavell, R.B. (1975) Characterisation of the wheat genome by renaturation kinetics. Chromosoma, 50, 223–242. Vala´rik, M., Bartosˇ, J., Kova´rˇova´, P., Kubala´kova´, M., de Jong, H.
and Dolezˇel, J. (2004) High-resolution FISH on super-stretched flow-sorted plant chromosomes. Plant J. 37, 940–950.
Vra´na, J., Kubala´kova´, M., Sˇimkova´, H., Cˇı´halı´kova´, J., Lysa´k, M.A. and Dolezˇel, J. (2000) Flow-sorting of mitotic chromosomes in common wheat (Triticum aestivum L.). Genetics, 156, 2033– 2041.
Wicker, T., Yahiaoui, N., Guyot, R., Schlagenhauf, E., Liu, Z.D., Dubcovsky, J. and Keller, B. (2003) Rapid genome divergence at orthologous low molecular weight glutenin loci of the A and Am genomes of wheat. Plant Cell, 15, 1186–1197.
Woo, S.S., Jiang, J., Gill, B.S., Paterson, A.H. and Wing, R.R. (1994) Construction and characterization of a bacterial artificial chro-mosome library of Sorghum bicolor. Nucleic Acids Res. 22, 4922– 4931.
Zhang, P., Li, W., Fellers, J., Friebe, B. and Gill, B.S. (2004) BAC-FISH identifies chromosome landmarks consisting of different types of transposable elements. Chromosoma, 112, 288–299.
