Skip Navigation


DNA Research Advance Access originally published online on January 11, 2006
DNA Research 2005 12(5):301-364; doi:10.1093/dnares/dsi018
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary data
Right arrowOA All Versions of this Article:
12/5/301    most recent
dsi018v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrow Request Permissions
Google Scholar
Right arrow Articles by Sato, S.
Right arrow Articles by Tabata, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sato, S.
Right arrow Articles by Tabata, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Kazusa DNA Research Institute
The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact journals.permissions@oxfordjournals.org

Comprehensive Structural Analysis of the Genome of Red Clover (Trifolium pratense L.)

Shusei Sato1 {dagger}, Sachiko Isobe2 {dagger}, Erika Asamizu1, Nobuko Ohmido3, Ryohei Kataoka3, Yasukazu Nakamura1, Takakazu Kaneko1, Nozomi Sakurai1, Kenji Okumura2, Irina Klimenko4, Shigemi Sasamoto1, Tsuyuko Wada1, Akiko Watanabe1, Mitsuyo Kohara1, Tsunakazu Fujishiro1 and Satoshi Tabata1,*

1Kazusa DNA Research Institute 2-6-7 Kazusa-kamatari, Kisarazu Chiba 292-0818, Japan
2National Agricultural Research Center for Hokkaido Region (NARCH) Hitsujigaoka 1, Toyohira, Sapporo, 062-8555, Hokkaido, Japan
3Faculty of Human Development, Kobe University Tsurukabuto 3-11, Nada, Kobe 657-8501, Japan
4All-Russian Williams Fodder Crop Research Institute 141055 Lugovaya, Moscow Region, Russia

Received 19 September 2005; revised 30 September 2005


   Abstract
Top
 Abstract
1. Introduction
 2. Materials and Methods
 3. Results
 4. Discussion
 Supplementary Material
 Appendix
 Acknowledgements
 References
 
With the aim of establishing the basic knowledge and resources needed for applied genetics, we investigated the genome structure of red clover Trifolium pratense L. by a combination of cytological, genomic and genetic approaches. The deduced genome size was ~440 Mb, as estimated by measuring the nuclear DNA content by flow cytometry. Seven chromosomes could be distinguished by microscopic observation of DAPI stained prometaphase chromosomes and fluorescence in situ hybridization using 28S and 5S rDNA probes and bacterial artificial chromosome probes containing microsatellite markers with known positions on a genetic linkage map. The average GC content of the genomes of chloroplast, mitochondrion and nucleus were shown to be 33.8, 42.9 and 34.2%, respectively, by the analysis of 1.4 Mb of random genomic sequences. A total of 26 356 expressed sequence tags (ESTs) that were grouped into 9339 non-redundant sequences were collected, and 78% of the ESTs showed sequence similarity to registered genes, mainly of Arabidopsis thaliana and rice. To facilitate basic and applied genetics in red clover, we generated a high-density genetic linkage map with gene-associated microsatellite markers. A total of 7159 primer pairs were designed to amplify simple sequence repeats (SSRs) identified in four different types of libraries. Based on sequence similarity, 82% of the SSRs were likely to be associated with genes. Polymorphism was examined using two parent plants, HR and R130, and 10 F1 progeny by agarose gel electrophoresis, followed by genotyping for the primer pairs showing polymorphisms using 188 F1 plants from the mapping population. The selected 1305 microsatellite markers as well as the previously developed 167 restriction fragment length polymorphism markers were subjected to linkage analysis. A total of 1434 loci detected by 1399 markers were successfully mapped onto seven linkage groups totaling 868.7 cM in length; 405 loci (28%) were bi-parental, 611 (43%) were specific to HR and 418 (29%) were specific to R130. Each genetic linkage group was linked to a corresponding chromosome by FISH analysis using seven microsatellite markers specific to each of the linkage groups as probes. Transferability of the developed microsatellite markers to other germplasms was confirmed by testing 268 selected markers on 88 red clover germplasms. Macrosynteny at the segmental level was observed between the genomes of red clover and two model legumes, Lotus japonicus and Medicago truncatula, strongly suggesting that the genome information for the model legumes is transferable to red clover for genetic investigations and experimental breeding.

Key words: red clover; EST; microsatellite marker; genetic linkage map


   1. Introduction
Top
 Abstract
 1. Introduction
2. Materials and Methods
 3. Results
 4. Discussion
 Supplementary Material
 Appendix
 Acknowledgements
 References
 
Plant genetics has experienced a drastic change in the last decade with the emergence of genomics. A large quantity of genomic and cDNA sequences have been accumulated and systematic analyses of gene function have been conducted at a genome-wide scale based on the available sequence information. Identification and isolation of causative genes for a variety of mutants have been accelerated by the use of genomic libraries and DNA markers that are generated in association with sequence analysis. Fusion of conventional genetics and genomics is typical in two well-known model plants, Arabidopsis thaliana and Oryza sativa L. (rice), and in several major crop plants to a lesser extent but has not been very widespread in other plant species including a variety of agronomically important plants. To understand the genetic systems in the plant kingdom and be of benefit to plant breeding processes, collection of genome-wide information and subsequent adoption of genomic approaches in a diversity of plants is urgently needed.

In the legume family, genomics and molecular genetics have rapidly advanced for the last several years with a central focus on two model legume species, Lotus japonicus and Medicago truncatula.1Go These species are closely related to two forage legume crops, birdsfoot trefoil (Lotus corniculatus L.) and alfalfa (Medicago sativa), respectively, and it is anticipated that their genomic and genetic information can be applied to crop legumes, including forage crops. Several comparative analyses among legumes have been reported 2Go–6Go and have demonstrated that genomic and genetic information of model plants may be beneficial for crop breeding. To facilitate such comparison and subsequent transfer of information among model legumes, the availability of genomic information and genomic resources for each species is a prerequisite. These resources would also be useful to investigate genetic systems specific to individual crop legumes.

Red clover (Trifolium pratense L.) is an important forage legume widely cultivated in most temperate regions because of its characteristics of high seedling vigor, rapid growth, and tolerance to acidic and humid conditions. It is also used as a green manure crop because of its high nutrient content resulting partly from symbiosis with nitrogen fixing bacteria of the genus Rhizobium. Red clover has a diploid genome (2n = 2x = 14), with a DNA content of 0.97 pg/2C,7Go which is slightly larger than that of rice.7Go Several studies have shown that the genome of red clover is extremely polymorphic due to its strongly self-incompatibile fertilization. In fact, intra-population genomic heterozygosity was higher than inter-population heterozygosity.8Go–10Go The genetic diversity of red clover has been intensively studied using amplified fragment length polymorphism (AFLP) markers.11Go,12Go The high level of heterozygosity has hampered intensive genetic and genomic analyses of red clover. Recently, the first genetic linkage map with 256 restriction fragment length polymorphism (RFLP) markers was constructed,13Go while in another Trifolium species, the white clover Trifolium repens, a genetic map with 566 microsatellite markers was reported.14Go

In this study, we investigated the genome structure of red clover using a variety of genomic technologies including fluorescence in situ hybridization (FISH) and genomic and cDNA library construction and sequencing. In parallel, we constructed a high-density genetic linkage map of the entire genome with a large number of DNA markers, which is invaluable for genome comparison, map-based gene identification and isolation, and marker-assisted breeding. We specifically focused on microsatellite markers associated with protein-coding genes, which are more informative and useful than those randomly distributed in the genome, in combination with a user-friendly and cost-effective detection system. Furthermore, transferability of these markers to other germplasms was examined to assess the feasibility of using these markers in selective breeding of red clover. In addition, macrosynteny with the genomes of two model legumes, L. japonicus and M. truncatula, was investigated using the developed microsatellite markers to explore the possibility of sharing knowledge among red clover, models and related crop plants.


   2. Materials and Methods
Top
 Abstract
 1. Introduction
 2. Materials and Methods
3. Results
 4. Discussion
 Supplementary Material
 Appendix
 Acknowledgements
 References
 
2.1. Plant materials
A cDNA library was constructed from the Japanese red clover variety ‘Hokuseki’.15Go A genetic linkage map was constructed using a mapping population of 188 F1 individuals derived from a double-pseudo test cross between HR as the female parent and R130 as the male parent. HR is an elite plant selected from early progenies between ‘Hokuseki’ and a Swiss variety ‘Renova’ with characteristics of early flowering, erect shape, red flower color and middle-sized leaves. R130 is a progeny of a mapping population used in the development of the previously reported RFLP map,13Go originating from a Russian variety and a wild variety from the Archangelsk region (approximately N 65°, E 40°), which exhibits late flowering, semi-prostrate shape, white flower color and small leaves.

2.2. Flow cytometry
Nuclear DNA content was measured by flow cytometry according to the method described by Galbraith et al.16Go and Ito et al.17Go, with minor modifications. The nuclei were isolated from fully expanded leaves of red clover and young leaves of A. thaliana, to provide a reference sample, by chopping the leaf tissues and filtering the minced tissue through double-layered nylon meshes (20 and 50 µm pore size). Nuclear samples were stained with 41.7 µg/ml of propidium iodide for 60 min, and then analyzed using a flow cytometer (FACScan, Becton Dickinson) according to Ito et al.17Go. More than 800 nuclei were measured with a minimum of triplicate analyses performed for each plant.

2.3. Generation of BAC libraries
Genomic DNA of HR was partially digested with MboI and cloned into CopyControl pCC1BAC (Epicentre, WI, USA.) according to the previously described method.18Go Two independent libraries containing genomic segments of different size ranges, 108 and 80 kb on average, were generated. Three-dimensional (3-D) DNA pools for PCR screening were generated as described previously.18Go

2.4. Chromosome analysis of condensation pattern and FISH
Chromosome samples for microscopic observation were prepared as previously reported.19Go Briefly, root tips of red clover were treated in 2 mM quinolin at room temperature for 4 h, followed by fixation in ethanol : acetic acid (3 : 1). On the next day, the tips were washed thoroughly and subjected to enzymatic maceration in a cocktail of 2.5% Pectolyase Y-23 (Seishin Pharmaceutical Co., Ltd, Japan) and 1% Cellulase Onozuka RS (Yakult Honsha Co., Ltd, Japan), and incubated in a thermal bath at 37°C for 40 min. The tips were then macerated in a few drops of methanol : acetic acid (1 : 1) fixative using the tips of a fine forceps and air dried on glass slides.

The 28S rDNA, the 5S rDNA and bacterial artificial chromosome (BAC) clones representing each linkage group were used as probes for FISH analysis. The 28S and the 5S rDNA probes were produced by PCR using primer pairs designed based on the rRNA and 5S RNA gene sequences in the red clover genome. The BAC probes were selected from the 3-D DNA pools of the BAC libraries by PCR using the primer pairs to amplify simple sequence repeats (SSRs).

The FISH analysis using rRNA genes and the BAC clones were performed on well-prepared chromosome spreads according to the method described previously by Ohmido et al.20Go, with modifications. The chromosome spread on glass slides, prepared as described above, were incubated in 100% ethanol at 90°C for 2 min and air dried. They were denatured in 70% formamide/2 x SSC at 70°C for 4 min, and then dehydrated in a series of 70 and 100% cooled ethanol for 5 min each and air dried. After hybridization with digoxigenin- or biotin-labeled probes at 37°C for 8–48 h, slides were incubated with Sheep-Anti-Dig FITC (Roche, Switzerland) and StreptAvidin-cy3 (Jackson Immuno. Res. Lab, USA) in 4 x SSC at 37°C for 30 min in a humid dark box, followed by incubation in 4 x SSC at 37°C for 30 min in the dark with Anti-sheep FITC (Vector Labs, USA) and StreptAvidin-cy3 to amplify the signal. After washing in 2 x SSC and air drying, the chromosome sample was stained with 1 µg/ml DAPI in Vectashield.

The preparations were observed under a fluorescence microscope (OLYMPUS BX50) equipped with a sensitive cooled CCD camera (PXL1400), and the prometaphase chromosome spreads with clear patterns were photographed by blue or green light excitation and emission filters. Captured images were digitally stored in a computer and analyzed using the CHIAS3 imaging software.21Go

2.5. Random sequencing of the genome
The total cellular DNA was extracted from leaves of HR using a DNeasy Plant kit (Qiagen, The Netherlands). The obtained DNA was segmented by sonication and size-fractionated by agarose gel electrophoresis, and DNA fragments of ~2 kb long were cloned into the HincII site of M13mp18 followed by introduction into Escherichia coli ElectroTen-Blue (Stratagene, USA) by electroporation. Single-stranded DNA was prepared from each colony according to the standard protocol, and was subjected to sequence analysis with DNA sequencers type 3700 and 3730 (Applied Biosystems, USA) using the Dye-terminator Cycle Sequencing kit according to the protocol recommended by the manufacturer. The sequence data can be retrieved from the DDBJ/EMBL/GenBank public DNA Database under the accession numbers DE244757–DE246660.

2.6. EST collection
Expressed sequence tags (ESTs) were collected from a variety ‘Hokuseki’ and a plant R130. cDNA libraries were constructed from 3-week-old plantlets of ‘Hokuseki’ and leaves of R130. Total RNA was extracted from 5 g of tissue by the guanidium thiocyanate/CsCl ultracentrifugation method, as described previously.22Go Purification of polyadenylated RNA and conversion to cDNA was performed as described previously.22Go Synthesized cDNA was resolved by 1% agarose gel electrophoresis, and a fraction ranging from 1 to 3 kb was recovered. The recovered fragments were cloned into the EcoRI–XhoI sites of a pBluescript II SK-plasmid vector (Stratagene, USA) and introduced into an E. coli ElectroTen-Blue strain (Stratagene, USA) by electroporation. Normalization of the libraries was performed by self-hybridization, as previously described.22Go For generation of ESTs, plasmid DNAs were prepared from the colonies and sequenced using the BigDye Terminator cycle sequencing ready reaction kit (Applied Biosystems, USA). The reaction mixtures were run on the automated DNA sequencer ABI PRISM 3730 (Applied Biosystems), and the collected data were processed as described below.

2.7. Sequence data analysis
Raw sequence data were quality-evaluated by the Phred program23Go,24Go prior to data analyses. BLAST programs were used to search public DNA databases, a proteome database of A. thaliana and datasets of the TIGR gene indices of three legumes, M. truncatula, Glycine max and L. japonicus. For DNA regions adjacent to SSRs, a cut-off value of E ≤ 10–5 or bit scores ≥30 against the predicted genes of A. thaliana and L. japonicus or bit scores ≥30 against sequences in the TIGR gene indices were considered significant.

EST data were analyzed as follows. Chromatograms were evaluated with Phred23Go,24Go and vector-derived sequences were trimmed with Crossmatch (P. Green, http://bozeman.mbt.washington.edu/phrap.docs/phrap.html).24Go The EST reads were quality-trimmed by the Phred quality score at a position where 5 ambiguous bases (Phred score under16) were found within 15 contiguous bases. Reads that comprised >50 bp of contiguous satisfying quality were submitted to the DDBJ/EMBL/GenBank databases with the accession numbers BB902456–BB928811. A similarity search against the UniProt Non-redundant Reference 100 (UniRef100) Databases (http://www.ebi.ac.uk/uniref/) was performed for each EST using the BLASTX program after translation into their respective amino acid sequences in six frames. EST sequences were clustered into non-overlapping groups by BLASTN using a criterion of 95% identity for >50 bases.

2.8. Identification of SSRs
Four types of genomic and cDNA libraries were constructed for identification of SSRs in the red clover genome.

An SSR-enriched genomic library was constructed from HR and R130. The total cellular DNA was extracted from leaves of each plant using the DNeasy Plant kit (Qiagen, The Netherlands). The isolated DNA was mixed and fragmented by sonication followed by size fractionation by agarose gel electrophoresis. DNA fragments of length ranging from 800 to 1000 bp were cloned into an EcoRV site of the pBluescriptII SK-plasmid vector (Stratagene, USA) and introduced into E. coli ElectroTen-Blue (Stratagene, USA) by electroporation. Enrichment of SSR sequences in the genomic library was performed by a method modified from a previous report.25Go Biotinylated oligos, 100 pmol each of (AAC)8, (AAG)8, (ATC)8, (GGA)8 and (GGT)8, were bound to Dynabeads (Dynal Biotech, Norway). Plasmid DNAs were prepared from the genomic library and the inserts were PCR-amplified using the vector-derived primers 5'-CGCTCTAGAACTAGTGGATCCC-3' (A) and 5'-TCGAGGTCGACGGTATCGATAAGC-3' (B). Predominantly single-stranded copies of the inserts were obtained by asymmetric PCR using a 1 : 10 ratio (0.5 and 5 pmol) of the A and B primers. Hybridization of biotinylated oligos and insert fragments was performed twice. After the second round of hybridization, the eluted supernatant was PCR-amplified using a equal amounts of vector-derived primers (5 pmol each). The PCR product was TA-cloned into pT7 Blue T-Vector (Novagen, Germany) according to the manufacturer's instructions.

A cDNA library was made from seedlings of ‘Hokuseki’ for generation of an SSR-enriched cDNA library. Extraction of total RNA and polyadenylated RNA and conversion to cDNA were carried out as described in Section 2.6. cDNA fragments ranging from 0.5 to 1 kb were cloned into pBluescript II SK-. Biotinylated oligos, 100 pmol each of (AAC)8, (AAG)8, (ATC)8, (GGA)8, (GGT)8, (CT)12, and (AAAG)6, were used as the driver. Hybridization of the biotinylated driver and PCR-amplified cDNA inserts and subsequent washing and cloning were performed as described above.

Construction of a methyl-filtration genomic library was performed by cloning the sonicated genomic DNA segments of ~1 kb length into the EcoRV site of pBluescript SK-, followed by introduction into three mcrBC+ E. coli hosts, JM109, DH5{alpha} and XLI-Blue.26Go

A normalized cDNA library was prepared from leaves and leafstalks of HR and leaves of R130 plants, as described above. The plasmid DNA was amplified directly from each colony for sequencing using the TempliPhi DNA amplification kit (Amersham, UK). Sequence analysis was performed from one end of each insert and the SSR motifs used for the enrichment process described above were searched. Only repeats equal to or longer than 15 bp were used for the subsequent steps.

2.9. Amplification of SSR-containing regions and detection of polymorphisms
Primer pairs for amplification of SSR-containing regions were designed based on the flanking sequences of each SSR with the assistance of the Primer 3 program27Go so that amplified fragment sizes were between 90 and 300 bp in length. PCR was performed in a total volume of 5 µl containing 0.5 ng of red clover genomic DNA, 1 x PCR buffer (TaKaRa Bio Inc., Japan), 0.2 U TaKaRa Taq (Takara Bio Inc., Japan), 0.2 mM dNTPs and 0.8 µM each of the primers. Reactions were run using a modified ‘touchdown PCR’ program:28Go 3 min at 94°C for the initial denaturation, 3 cycles of 30 s at 94°C and 30 s at 68°C, followed by 2 rounds of the same program in which the annealing and extension temperatures were decreased by 2° every 3 cycles, then 4 rounds of a 3-step program of 30 s at 94°C, 30 s at 62°C, 30 s at 72°C, followed by 3 rounds of the same program in which the annealing temperature was decreased by 2° every 3 cycles, with a final extension for 10 min at 72°C. PCR products were resolved either on 3% MetaPhor agarose gels (BMA, USA) or on 10% acrylamide gels. The primer pairs giving a polymorphism among the parents of the mapping population, HR and R130, and 10 F1 progenies were selected and used for scoring a mapping population of 188 F1 plants.

2.10. RFLP assays
Total DNA was isolated from young leaves of red clover using the CTAB extraction method described by Doyle and Doyle.29Go Approximately 3 µg of DNA were digested with each of the six restriction enzymes, BamHI, DraI, EcoRI, HindIII, KpnI and PstI. The restriction fragments were separated on 0.6% agarose gels and blotted onto nylon membranes according to standard procedures. Red clover cDNA probes which had been mapped on the previously generated RFLP map13Go were labeled using the ECL direct-labeling system (Amersham, UK) and used as probes for Southern hybridization. The signals were detected by chemi-luminescence using an X-ray film.

2.11. Linkage analysis
Segregation data obtained from a mapping population of 188 F1 plants using microsatellite and RFLP markers were analyzed by a combination of a color map method30Go and a JoinMap program version 3.0 (http://www.kyazma.nl). To increase the efficiency and accuracy of the map calculation, the scored markers were roughly classified into seven linkage groups using the color map method prior to the JoinMap calculation. Segregation data were scored by ‘cp’ population type codes for the JoinMap analysis, which was followed by conversion into color codes showing marker types according to the color map procedure. The color-coded genotypes were displayed in a matrix for each parent and classified into seven groups representing individual chromosomes. Two data subsets of individual linkage groups from each single parent were recalculated independently by the grouping module of JoinMap with LOD = 5. Two parental data subsets were merged to one data subset using a join module of JoinMap. The marker order was calculated by a mapping module of JoinMap with the following parameters: Kosambi's mapping function, LOD ≥ 2.0, REC ≤ 0.35.

2.12. Allele frequency of the markers of other red clover germplasms
A total of 11 red clover varieties bred in different countries were used for polymorphism analysis: ‘Natshyu’ (Japan), ‘Hokuseki’ (Japan), ‘Sapporo’ (Japan), ‘Hokuiku-20’ (Japan), ‘Rannij2’ (Russia), ‘Start’ (Czechoslovakia), ‘Krano’ (Denmark), ‘Renova’ (Switzerland), ‘Merviot’ (Belgium), ‘Kenland’ (USA) and ‘Altaswede’ (Canada). Genomic DNA was extracted from eight individuals of each variety and subjected to PCR examination with the 268 primer pairs for the selected microsatellite markers distributed throughout the entire genome. The presence or the absence of amplification and the number of different sized fragments, which was regarded as the number of alleles, were recorded. Loci where no amplification was observed were regarded as null. The heterozygous/homozygous ratio of single amplified fragments was estimated based on the ratio of individuals with no amplification to the total 88 individuals. The number of alleles and polymorphism information content (PIC) were estimated based on the SSR marker data obtained. PIC was calculated using the following equation:

Formula
where Pij is the frequency of the jth allele for the ith marker.


   3. Results
Top
 Abstract
 1. Introduction
 2. Materials and Methods
 3. Results
4. Discussion
 Supplementary Material
 Appendix
 Acknowledgements
 References
 
3.1. Cytological analysis of the red clover genome
The nuclear DNA contents of two red clover plants, HR and R130, were calculated to be 0.91 pg/2C (n = 7) and 0.89 pg/2C (n = 7), respectively, by comparison with an A. thaliana standard (0.32 pg/2C)31Go (data not shown). These figures differ somewhat from a previous report (0.97 pg/2C) by Arumuganathan and Earle 7Go, in which chicken red blood cells (2.33 pg/2C) were used as a standard. Given that 1 pg of DNA is equivalent to 980 Mb,32Go,33Go the genome sizes of HR and R130 are estimated to be 446 and 436 Mb, respectively.

A karyotype of the red clover genome was analyzed by microscopic observation of prometaphase chromosomes stained by DAPI (Fig. 1a). The lengths of the prometaphase chromosomes ranged from 5.1 to 7.4 µm, and uneven condensation patterns that have proven useful in chromosome identification were observed. The resolution of individual chromosomes was better than a previous report34Go, in which the length of condensed metaphase chromosomes ranged from 1.9 to 2.9 µm, but seven chromosomes could not be definitively distinguished.


Figure 1
View larger version (37K):
[in this window]
[in a new window]
  		Cytological analysis of the red clover genome. (a) Red clover chromosomes stained by DAPI. (b) FISH analysis using 28S rDNA (green signals) and 5S rDNA (red signals) in R130. Arrows and arrowheads indicate 28S rDNA loci on Chromosome 1 and 6, respectively. Open arrows indicate 5S rDNA loci on Chromosomes 1 and 2. (c) FISH analysis using RCS2546 (green signals indicated by open arrowheads) and 28S rDNA (red signals indicated by chromosome number) in HR. (d) Chromosome map of the red clover genome. Green circles, loci of seven BACs corresponding to LG-specific markers; red boxes, 28S rDNA loci; orange circles, 5S rDNA loci.

 
To further characterize individual chromosomes, FISH analysis was performed using 28S and 5S rDNA fragments as probes. As shown in Fig. 1b, the 28S rDNA (green signals) loci could be detected most intensely in the nucleolar organizer regions (NORs) on the short arm of Chromosome 1 (arrows) and as a less intense signal in the internal regions on the short arm of Chromosome 6 (arrowheads). An additional signal was observed on one Chromosome 5 homologue in HR, but not in R130 (Fig. 1b). The 5S rDNA loci (red signals) could be detected proximal to the NOR signals on Chromosome 1 (open arrows), and in an additional two loci on the short arm of Chromosome 2 (open arrows). The results are summarized in Fig. 1d.

3.2. Sequence features of the genome
To explore the general sequence features of the red clover genome, 960 plasmid clones from a random genomic library of HR were sequenced from both ends of the inserts, and 1920 sequence files with an average length of 732 bp and a Phred score of ≥20 totaling 1.4 Mb in length were generated. A total of 244 (12.7%) and 14 (0.7%) sequences showed a high degree of similarity (E ≤ 10–50) to the genomes of chloroplast and mitochondria of A. thaliana, respectively, indicating that these sequences are derived from these organelles. Ninety-three files (4.8%) matched higher plant rDNA sequences. The average GC content of the putative chloroplast and mitochondrial genome sequences were 33.8 and 42.9%, respectively, while that of the remaining sequence files, which are likely to originate from the nuclear genome, was 34.2%.

Di-, tri- and tetra-nucleotide motifs were extracted from the genomic sequences to assess the composition of SSRs in the red clover genome. A pattern search for SSRs of 15 nt or longer identified 126 such SSRs in the 1.2 Mb nuclear genomic sequences (one SSR in every 9.7 kb). The poly(AT)n (37 loci, 29% of the identified SSRs) was the most abundant motif, followed by poly(AAT)n (35 loci, 28%) and poly(GA)n (12 loci, 10%).

The proportion of gene spaces in the entire genome was roughly estimated by determining the protein-coding regions in the genomic sequences. The sequence files (1662) that were likely to have originated from the nuclear genome were used as a query to search the proteome of A. thaliana and the sequences of M. truncatula, G. max and L. japonicus available in the TIGR gene indices (http://www.tigr.org/tdb/tgi/plant.shtml), with a minimum cut-off value of E = 10–5. As a result, 1018 out of 1662 files showed significant similarity to sequences of protein-coding genes in the dataset examined, and of these 177 files (10.6%) showed sequence similarity to genes related to transposons. The total length of the sequence files containing putative protein-coding regions but not transposons was 518 kb and their proportion in the files of the entire nuclear sequences was 50.6%.

3.3. Gene features
A total of 33 024 clones from cDNA libraries were single-pass sequenced from their 5' ends and 26 356 ESTs which met the criteria described in Section 2 were obtained (Table 1). The quality of the libraries with respect to 5'-termini coverage was assessed by comparison of 5'-end sequences to known protein sequences. Among the randomly selected 100 clones, 50 contained a translation initiation codon. To determine the number of non-redundant ESTs, EST clustering was performed as described in Section 2, and the 26 356 sequences were classified into 9339 non-redundant groups comprising 3508 contig sequences and 5831 singletons.


View this table:
[in this window]
[in a new window]
  		List of cDNA libraries constructed for EST analysis.

 
The 9339 non-redundant sequences were searched against the UniRef100 database to assign putative functions. At the time of writing, 7264 (78%) showed significant similarity (E < 10–10) to registered sequences. Most of the clones (4279) showed the highest similarity to A. thaliana genes; 670 were most similar to rice genes. Genes conserved between red clover and A. thaliana were classified into functional categories according to the Arabidopsis Gene Ontology (GO), provided by The Arabidopsis Information Resource (http://www.arabidopsis.org/). The 7941 red clover genes showing similarity to A. thaliana genes (E < 10–5) were subjected to this analysis. The distribution of genes in the molecular function, biological process and cellular component categories are shown in Appendix Fig. 1. As for legume species, 251, 224, 100, 61 and 52 red clover genes were similar to genes in Pisum sativum, G. max, M. sativa, M. truncatula and L. japonicus, respectively. The EST sequence of each clone and similarity search results are provided at www.dnaresearch.oxfordjournals.org.


Figure 4
View larger version (36K):
[in this window]
[in a new window]
  		Classification of ESTs into functional categories according to Arabidopsis Gene Ontology. (a) Molecular function category; (b) biological process category; (c) cellular component category.

 
3.4. Development of microsatellite markers
To identify SSRs in the red clover genome, we generated four different types of libraries, SSR-enriched genomic and cDNA libraries, a methyl-filtration genomic library and a normalized cDNA library, as shown in Table 2. The preliminary analysis of the cDNA sequences of M. truncatula and G. max in the TIGR gene indices, genomic and cDNA sequences of L. japonicus and cDNA sequences of red clover revealed that specific SSR motifs frequently occur in the protein-coding sequences of the legume genomes (data not shown). These motifs were used for selective enrichment in the construction of the SSR-enriched genomic and cDNA libraries, as described in Section 2. It has been reported that low-copy number gene-containing regions of the genome can be enriched by methyl filtration.26Go We adopted this methodology to concentrate SSRs occurring in gene spaces in the red clover genome. A standard normalized cDNA library was also used as a source to isolate SSRs directly associated with expressed gene sequences.


View this table:
[in this window]
[in a new window]
  		List of libraries for identification of SSRs.

 
A total of 83 172 clones were isolated from the four libraries and subjected to sequence analysis. As summarized in Table 3, 15 427 (53%) and 6568 (57%) clones from the SSR-enriched genomic and cDNA libraries, respectively, contained SSRs of 15 nt or longer, while 533 (28%) and 6801 (25%) clones in the methyl-filtration genomic and the normalized cDNA libraries, respectively, contained such SSRs. Based on the nucleotide sequences obtained, we designed a total of 7244 primer pairs to amplify SSRs by PCR, with putative amplified products ranging from 90 to 300 bp in length. We performed similarity searches of unique sequences adjacent to each SSR against the sequences of the predicted genes of A. thaliana and L. japonicus as well as cDNA sequences in the TIGR gene indices. Cut-off values of E ≤ 10–5 or bit scores ≥30 were adopted according to the preliminary alignments examined by eye. As shown in Table 2, 5970 (82%) out of 7244 query sequences showed similarity to the protein-coding genes described above, indicating that a significant fraction of the SSRs identified in this study are associated with genespaces.


View this table:
[in this window]
[in a new window]
  		Description of the integrated linkage map.

 
Polymorphism was examined for the 7159 primer pairs using the mapping parent plants, HR and R130, and 10 F1 progeny as templates. Some of the SSRs found in the random genomic sequences and cDNA sequences used in the RFLP map construction were also tested. As a result, a total of 1488 primer pairs identified polymorphisms among the 12 plants on 3% agarose gels. The primer pairs producing multiple bands likely to have originated from multiple loci of the genome were excluded as much as possible to avoid confusion in map construction. The selected primer pairs were then used to score polymorphisms in 188 F1 plants of the mapping population. Ultimately, from a total of 1305 primer pairs, 1024 gave clear polymorphisms on 3% agarose gels and 281 on 10% acrylamide gels and these were used for construction of a linkage map.

3.5. Screening of RFLP markers
To integrate the previous genetic linkage map generated with RFLP markers13Go into the new map, 121 cDNA probes were examined for detection of RFLP among the HR and R130 mapping parent plants and 14 F1 progeny. Using 188 F1 plants from the mapping population, 95 probes that showed polymorphisms were used for further RFLP analysis. Of the 95 probes 37 produced 2–6 polymorphic bands. In these cases, each band was separately scored. In total, 167 RFLP markers were successfully scored for subsequent construction of the linkage map in combination with the microsatellite markers.

3.6. Construction of a genetic linkage map
Linkage analysis using 1472 informative markers (1305 microsatellite and 167 RFLP markers) resulted in the assignment of 1463 loci to seven linkage groups by the color map procedure. Next, 1434 loci derived from 1399 markers (1286 detected by microsatellite markers and 148 by RFLP markers) were mapped onto the seven linkage groups by JoinMap. The results are summarized in Table 3, Appendix Fig. 2, Appendix Tables 1 and 2, and Supplemental Data at www.dnaresearch.oxfordjournals.org. The total map length was 868.7 cM with the shortest linkage group being 108.2 cM (LG6) and the longest being 149.7 cM (LG2). The average locus distance was 0.61 cM ranging from 0.46 cM in LG3 to 0.81 cM in LG5. Marker locus density seemed to be slightly higher in the proximal regions and lower in the distal regions of each linkage group (Appendix Fig. 2). A total of 34 markers (33 microsatellite and 1 RFLP) detected duplicate loci, of which 24 were bi-parental, 6 were HR-specific and 4 were R130-specific, and 17 of the duplicate loci mapped between linkage groups, 17 were within linkage groups.


Figure 5
Figure 5
Figure 5
View larger version (143K):
[in this window]
[in a new window]
  		Genetic linkage map of the red clover genome. Each linkage group corresponds to that in the previously reported linkage map.13Go LG2, LG1, LG3, LG4 and LG5 are inversely shown to match the chromosome map in Fig. 1. Positions of bi-parental, HR-specific and R130-specific marker loci are indicated by purple, pink and blue boxes, respectively. Distorted loci are preferentially represented when multiple markers including the distorted loci are closely located.

 

View this table:
[in this window]
[in a new window]
  		List of microsatellite marker loci.

 

View this table:
[in this window]
[in a new window]
  		List of RFLP marker loci.

 
Bi-parental or parent-specific marker loci segregating from HR and R130 were distributed among the seven linkage groups. Out of 1434 loci 405 (28%) were bi-parental, while 611 (43%) and 418 (29%) were specific to HR and R130, respectively (Table 3). LG1 harbored HR-specific markers at the highest frequency (64% of all loci on LG1). Of 405 bi-parental loci, 167, 160 and 78 were <abxcd>, <efxeg>, and <hkxhk> segregation types, respectively (Table 4). Distorted segregation was observed for 27.1% of all the marker loci on the map (P < 0.05). The loci showing distortion were distributed among all the linkage groups (Table 3), but the proportion of distorted loci was different for each linkage group. Only 7.7% of the marker loci on LG5 exhibited segregation distortion, while 52.3% showed distortion in LG2. The distortion ratio was fairly similar among segregation types (22.7–31.9%), except for the <hkxhk> type (67.9%), as shown in Table 4.


View this table:
[in this window]
[in a new window]
  		Number of locus and distortion ratio at each segregation type.

 
3.7. Linkage between genetic and cytological maps
To confirm the authenticity of the genetic linkage map constructed above, each linkage group was assigned to a chromosome by FISH analysis. Seven microsatellite markers located close to the end of each linkage group were selected as representatives: RCS1777 (LG1), RCS1588 (LG2), RCS1627 (LG3), RCS1647 (LG4), RCS0036 (LG5), RCS0019 (LG6), RCS2546 (LG7) and the 3-D DNA pools of the BAC genomic libraries were screened by PCR using the corresponding primer pairs. The selected BAC clones were used as probes in FISH analysis for chromosome mapping. As shown in Fig. 1c and d, the BAC clones harboring the markers RCS1777, RCS1588, RCS1627, RCS0036, RCS0019 and RCS2546 exclusively hybridized to the distal regions of Chromosome 4, 2, 5, 1, 6 and 3, respectively. RCS1647 was detected in the distal portion of Chromosome 7 and the central portion of Chromosome 1 adjacent to the NOR.

3.8. Evaluation of allele frequency in germplasms by polymorphism analysis
To analyze transferability of the generated markers to other red clover germplasms, 268 randomly selected microsatellite markers were examined by PCR in 88 red clover individuals. The list of markers and the results are summarized in Appendix Table 3. All the tested markers produced amplification products (data not shown). Homozygous null alleles, where no bands were produced, were observed in at least 1 individual in 78 of the 268 markers tested. The number of alleles per locus ranged from 2 to 19 with a mean value of 6.5 (Fig. 2a). Markers detecting four alleles were most frequent. PIC values ranged from 0.05 to 0.89 with a mean value of 0.60 (Fig. 2b). Markers with PIC values between 0.7 and 0.8 were most common.


View this table:
[in this window]
[in a new window]
  		List of the marker loci for evaluation of allele frequency.

 

Figure 2
View larger version (23K):
[in this window]
[in a new window]
  		Allele frequency in germplasms. (a) The number of alleles per locus; (b) distribution of the PIC values.

 
3.9 Comparison with the genomes of L. japonicus and M. truncatula
We performed BLASTN searches of the red clover sequences adjacent to mapped microsatellite markers against the genome databases of two model legumes, L. japonicus and M. truncatula. Sequences with E ≤ 10–5 were considered similar. Of the 1286 sequences that corresponded to mapped microsatellite markers, 434 and 566 showed sequence similarities to the genomes of L. japonicus and M. truncatula, respectively, 257 of which were common to both genomes (Appendix Table 1). For 133 and 161 red clover sequences a single match was found to the genomes of L. japonicus and M. truncatula, respectively, but the remaining 301 and 405 sequences had multiple matches to the respective genomes (Appendix Table 1 and Supplemental Data at www.dnaresearch.oxfordjournals.org). Two or more neighboring marker sequences of 52 loci on the red clover genetic linkage map had hits to either a single clone or clones closely located in the genomes of L. japonicus and M. truncatula (data not shown), suggesting the presence of microsynteny between red clover and the two model legumes.

Because significant portions of the genomic sequences of L. japonicus and M. truncatula have been anchored on their respective genetic linkage maps, the syntenic relationship between the red clover and model legume genomes could be explored by simply comparing the map locations of the red clover DNA markers and the corresponding best-hit genomic sequences of the model legumes. As shown in Fig. 3, alignment of homologous sequence pairs along each linkage group revealed an obvious syntenic relationship. The alignment appears relatively simple in the case of the red clover (rc) LG1, L. japonicus (Lj) chr5 and M. truncatula (Mt) chr1. Lj chr1, on the other hand, appeared to correspond to two alignments, rc LG6–Mt chr7 and rc LG7–Mt chr3. In contrast, the relationship of other linkage groups of red clover with those of the two model legumes seemed to be more complex at the macro level. At the segmental level, however, syntenic relationships could be detected in all the linkage groups. Some of the linkages, such as rc LG 2–Lj chr2–Mt chr6 and rc LG 3–Lj chr4–Mt chr4, have been supported by the synteny analysis based on the genomic sequences of L. japonicus and M. truncatula.1Go


Figure 3
View larger version (32K):
[in this window]
[in a new window]
  		Macrosynteny between the genomes of red clover and two model legumes. Putative syntenic relations were estimated by ordering the red clover DNA markers along with the linkage map positions of the corresponding best-hit genomic clones of L. japonicus and M. truncatula. In cases where multiple red clover LGs were assigned in a region of the L. japonicus and M. truncatula linkage maps, a representative red clover LG was selected for every 10 cM interval. For each represented syntenic region one or two representative pairs were drawn with the color code indicating the number of assigned homologous sequence pairs represented by colored lines, blue: one; green: two to four; red: five or more.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Materials and Methods
 3. Results
 4. Discussion
Supplementary Material
 Appendix
 Acknowledgements
 References
 
The nuclear DNA contents of two red clover plants, HR and R130, were estimated to be 0.91 and 0.89 pg/2C, respectively, by flow cytometory using A. thaliana (0.32 pg/2C) as a standard. This result is not in close agreement with a previous report by Arumuganathan and Earle, which estimated the red clover DNA content to be 0.97 pg/2C and which used chicken red blood cells as a standard.7Go The discrepancy between these studies may be due to the use of different standards and/or the heterogeneous genome structure of red clover due to its cross-pollination breeding system. Using the conversion factor of 1 pg DNA/980 Mb,32Go,33Go the genome sizes of HR and R130 were estimated to be 446 and 436 Mb, respectively. In comparison, the nuclear DNA content and the corresponding genome size of rice were estimated to be 0.88–0.89 pg/2C and 435 Mb, respectively, under the same conditions (data not shown). However, this is an overestimation because the genome size of rice was shown to be 389 Mb by genome sequencing.35Go Therefore, it is likely that the genome size of red clover is smaller than our present estimation, possibly as small as that of rice. The genome size of HR was slightly but consistently larger than that of R130. Whether this reflects a difference in genome structure or variability within the error range of our measurements remains to be clarified.

We could not uniquely identify the seven chromosomes by DAPI staining because the banding patterns of the smaller chromosomes were not clear and some of the chromosomes were similar in size.34Go However, FISH analysis using 28S and 5S rDNA as landmarks discriminated four of the chromosomes. Moreover, the presence of minor loci for 28S rDNA in addition to the NOR is a novel finding. One of the Chromosome 5 homologues in HR hybridized with 28S rDNA, but no Chromosome 5 staining was detected in R130. This may reflect a difference in genome structure between the two haplotypes in the HR genome.

By random sequencing of the total cellular DNA, the GC content of three genomes in red clover was deduced as follows: chloroplast, 33.8%, mitochondrion, 42.9% and nucleus, 34.2%. The GC content of the nuclear genome was similar to those of two model legumes, L japonicus (37.0%) and M. truncatula (33.3%), as well as that of A. thaliana (34.8%). SSRs seemed to be distributed rather evenly throughout the genome. AT and AAT were the major motifs and comprised over 57% of all the di-, tri- and tetra-nucleotide motifs, whereas AT, AG and AAG were dominant in L. japonicus and M. truncatula (data not shown). With respect to each repeat motif, 35% of the AT and 37% of the AAT motifs were linked to protein-coding sequences (data not shown). In contrast, 100% of ATC and 71% of AC motifs were associated with coding sequences, although these represent only 3.2 and 5.6%, respectively, of all the SSR motifs examined (data not shown).

A similarity search against the registered sequences of plant genes indicated that approximately half of the red clover genomic sequences contained protein-coding genes. Considering that 78% of the red clover ESTs showed significant similarity to registered sequences, the genespace of red clover could be as much as 65% of the entire genome, while the remaining regions are likely to be occupied by known and unknown transposon-related and other repeat sequences. No highly repetitive sequences were detected in this analysis.

We generated 26223 red clover ESTs that were grouped into 9339 non-redundant species. At the time of writing, the EST database in GenBank (dbEST) contained only 53 ESTs from a clover, Trifolium purpureum. Undoubtedly, the EST information as well as the cDNA clones generated in this study will facilitate gene isolation and large-scale analysis of gene function in Trifolium species. A similarity search against the UniRef100 protein database found that ~80% of the ESTs showed significant similarity to the registered genes, indicating that the functions of the majority of genes obtained in this EST project can be deduced by similarity to known genes. The result of GO classification of the red clover ESTs indicated that the cDNA clones were rather evenly distributed among a variety of functional classes, suggesting that we obtained a representative selection of gene species, probably by normalization of the cDNA library. To encourage the use of these data by the research community, we have created a database to provide EST sequence information for each clone and primary annotations deduced by similarity to known protein sequences (www.dnaresearch.oxfordjournals.org).

In this study, we aimed to develop a genetic linkage map of the red clover genome with a sufficient number of gene-associated microsatellite markers to facilitate map-based gene cloning and precise mapping of quantitative trait loci genes. We intended to adopt a cost-effective system to make use of the generated markers for a wide variety of purposes including breeding. For identification of SSRs, we constructed four different libraries: SSR-enriched genomic and cDNA libraries, a methyl-filtrated genomic library and a normalized cDNA library. SSRs were found in 53 and 57% of the clones from the SSR-enriched genomic and cDNA libraries, respectively, indicating that enrichment of SSRs by hybridization was successful. It should be noted that even the methyl-filtrated genomic library and the normalized cDNA library that were generated without any SSR enrichment process contained substantial amounts of SSRs.

DNA markers associated with genes are more informative and useful than those randomly generated from genomes such as genomic microsatellite, RAPD and AFLP markers. A similarity search against the predicted gene sequences of A. thaliana and L. japonicus and the cDNA sequences in the TIGR gene indices was performed to evaluate the libraries derived from red clover genomic DNA. Methyl filtration proved to be effective in enrichment of gene sequences because >90% of the clones from the methyl-filtrated genomic library showed similarity to presumptive genes. Of the SSRs identified in the SSR-enriched genomic library 60% were likely to originate from genespaces partly because hybridization was performed for SSRs preferentially located in the protein-coding sequences. Together with the SSRs from the two types of cDNA libraries, >80% of the SSRs identified in this study were likely to be originated from the genespace regions, demonstrating that the SSRs isolated in this study are an excellent source for the generation of gene-associated microsatellite markers.

Out of 7159 primer pairs we selected 1488 (21%) to amplify SSRs based on detection of polymorphisms in a mapping population on 3% agarose gels. In fact, ~70% of the microsatellite markers corresponding to the selected 1488 primer pairs could be genotyped fairly clearly on agarose gels with the 188 F1 mapping population. This screening ratio (21%) is much lower than that in white clover (63%), where polymorphisms were detected by capillary electrophoresis arrays.14Go We adopted the agarose gel system instead of more cost-intensive but sensitive systems, such as a fluorescent capillary gel system, to meet the requirements of a wide variety of users including breeders. This decision affected the success rate of selection of polymorphic markers. However, the primer pairs were screened under stringent conditions, thus allowing stable amplification and polymorphism detection in other detection systems. Sequence analysis of the amplified products showed that SSRs were indicative of polymorphism for most of the bands ranging from 200 to 500 bp in length, while polymorphisms observed in bands longer than 500 bp were often allocated to non-SSR regions (data not shown).

We constructed a genetic linkage map of 868.7 cM composed of 1434 marker loci, most of which were detected by microsatellite markers. Recently, high-density genetic linkage maps have been reported in several plant species: a rice map with 2740 microsatellite loci (157 kb/locus) by electronic-PCR,36Go a sorghum map of 1713 cM with 2926 loci mostly detected by AFLP markers37Go and a integrated soybean map of 2524 cM with 1849 loci mostly detected by microsatellite markers.38Go The red clover linkage map in the present study is comparable to these maps in terms of locus density and marker quality (co-dominant and gene association). Furthermore, only 2.3% of the markers detected duplicate loci, a phenomenon which often leads to confusion in map construction. These data also demonstrate high level of transferability of our markers and the map to other red clover germplasms and other plant species.

The total length of the genetic linkage map generated in this study was 868.7 cM, which is substantially longer than that of the previous RFLP map (535.3 cM),13Go indicating that a larger number of marker loci extended coverage of the genome. An alternative explanation is that the lengths of linkage maps are variable depending on inherent differences in mapping populations.2Go The parents of the present map originated in Switzerland and Japan (HR) and Russia (R130), while those of the previous map were derived only from Russia. The proportion of HR-specific loci (43%) was significantly larger than R130-specific (29%) and bi-parental (28%) loci, indicating that genetic diversity between two the haplotypes in HR was wider than that in R130.

We observed large spaces between marker loci at many of the distal portions of the linkage groups. The causes of this uneven distribution of loci remain to be studied. Since the majority of markers generated in this study are likely to be gene-associated, it is possible that the lower density reflects a lower gene density due to the presence of repetitive sequences in the distal regions of the chromosomes. Another possibility would be that the markers were mapped on these regions simply by type I error in the linkage analysis. Nevertheless, the FISH analysis for the markers in the distal regions of each linkage group, especially RCS0019, mapped at the terminus of LG6, which demonstrates the authenticity of the genetic linkage map generated in this study.

The relative ratio of the physical length to the genetic distance can be roughly estimated to be 507 kb/cM by simply dividing the genome size (440.1 Mb) by the length of the linkage map. The relative physical/genetic distance is shorter in red clover than in other legumes: 970 kb/cM in M. truncatula,2Go 907 kb/cM in L. japonicus 39Go and 835 kb/cM in white clover14Go. The genome size of red clover is comparable to those of M. truncatula and L. japonicus. Generally, genome diversity is wider in allogamous plants than in autogamous plants, and this might influence genetic segregation, resulting in different relative ratios of the physical to genetic distance. Alternatively, the difference in physical/genetic distance between two allogamous plants, red clover and white clover, may simply reflect the difference in genome size, 440.1 Mb in red clover versus 956 Mb in white clover14Go.

The density of the marker loci varies among the seven linkage groups. LG5 is a relatively short linkage group with a low density of loci. LG5 corresponds to Chromosome 1 harboring a large satellite which hybridizes with 28S rDNA. The long stretch of 28S rDNA, where genetic recombination is restricted, might result in the short genetic distance. The small number of marker loci, on the other hand, may reflect a lower gene density in this chromosome. Sequence conservation of rDNA and of other regions of Chromosome 1 may also contribute to the low degree of polymorphism and the short length of this linkage group. LG1 is an intermediate sized linkage group with the lowest density of loci. The distinctive feature of this linkage group is the highest proportion of HR-specific marker loci (64%). As described above, HR originated from varieties that were bred in two countries, Japan and Switzerland. These data suggest that LG1 was much more conserved within the red clover germplasms before they spread worldwide. Interestingly, a high degree of macrosynteny is observed between red clover LG1 and the genomes of two model legumes, M. truncatula and L. japonicus. Further analysis of LG1 with respect to synteny and gene function may provide clues to the evolution of red clover, as well as of other legume species.

Approximately 27% of the marker loci exhibited segregation distortion in the present map, while 37% showed distortion in the previous RFLP map. A high degree of genomic diversity in the mapping population used in this study may have contributed to reduced levels of distortion. The distortion ratio was especially high (68%) in the <hkxhk> type, possibly because some mechanism to avoid inbreeding weakness affected segregation. Distortion ratios varied among the seven linkage groups, ranging from 8 to 52% (Table 3). LG2, the longest linkage group, showed the highest distortion ratio.

The linkage between the genetic map and the cytological map was investigated by FISH analysis. The marker RCS1647 for LG4 hybridized to two different loci, the distal portion of Chromosome 7 and the central portion of Chromosome 1. The signal on Chromosome 1 might be due to duplication of a gene or a chromosomal segment during genome evolution. All the signals except that of RCS1647 on Chromosome 1 were detected at the distal portions of each chromosome, which agrees well with the positions of the respective markers on the corresponding linkage groups. This, together with the clear one-to-one relationship between each linkage group and each chromosome, strongly demonstrates the authenticity of the genetic linkage map.

Breeding involves thousands of individuals with a wide variety of genetic variations, whereas genetic linkage maps generated using F1 mapping populations reflect only four haplotypes. Thus, in order to utilize DNA markers and linkage maps for breeding processes such as development of screening markers and diversity analysis, transferability of the markers is crucial. In this study, we confirmed transferability of 268 selected markers on the genetic linkage map to 88 red clover germplasms derived from 11 varieties. The number of alleles per locus ranged from 2 to 19 with a mean value of 6.5, which is greater than the number identified in the previous RFLP map (average 3.1 alleles per locus ranging from 1 to 10). This result strongly indicates that the microsatellite markers developed in this study are suitable tools for analyzing numerous red clover germplasms.

PIC estimates the number of polymorphic pairs among all possible pairs in a population. In this study, the average PIC of 268 microsatellite markers was 0.6, which means that 60% of the germplasms (or two haplotypes) have heterozygous loci. For construction of a map using F1 progeny at least one of four haplotypes needs to show polymorphisms. To estimate the polymorphism ratio in four alleles Formula, where Pij is the frequency of the jth allele for the ith marker, was calculated. The mean value of PICi/4 allele for the 268 microsatellite markers was 0.85 (data not shown), which means that the probability of detecting a polymorphism between a pair of mapping parents per marker is 85%. In this study, we observed distortions from the normal distribution of allele number per locus and PIC. The allele number per locus observed most frequently was 4, which is smaller than the mean value of 6.2. However, the PIC distribution peaked at 0.7–0.8, which is larger than the mean value of 0.6. Because PIC is determined by the number and deviation of alleles in each locus, these data suggest that the higher PIC value could be largely attributed to an even distribution of polymorphic alleles rather than the total number of alleles.

Utilizing the sequence and map information obtained in this study, we compared the genome structures of red clover and two model legumes, L. japonicus and M. truncatula. A similarity search indicated that a substantial proportion of the red clover marker sequences showed sequence similarities to the model legumes, even though the sequences of only half of the entire genomes of each model species are currently publicly available. This demonstrates that genomes of red clover and the model legumes can be linked by the use of the DNA markers and the corresponding sequences. It is implicative that approximately two-thirds of the red clover markers examined have two or more matching sequences in the genomes of L. japonicus and M. truncatula. This may be due to gene and/or segmental genome duplications, which have diversified the genome structures during evolution. Nevertheless, a moderate degree of macrosynteny was observed between the genomes of red clover and the two model legumes as shown in Fig. 3, and the presence of microsynteny was also suggested.

The genome information and resources that we generated for red clover in this study are not as abundant as those accumulated for A. thaliana and some major crop plants. However, even these resources are sufficient for multiple applications. For example, combination of the microsatellite markers with known gene sequences and the genomic BAC libraries, which provide a 7.9-fold coverage of the genome, would allow isolation of most of the genes homologous to those of agronomic importance. A combination of information and material resources of the genomes not only in red clover but also in other plant species would provide further possibilities, including identification and isolation of genes utilizing synteny information from the model plants, development of DNA markers for candidate genes based on sequences from other plants, generation of consensus genetic maps among intra- and inter-specific groups, evaluation of variation among genetic resources and an efficient association analysis.

One of our major intentions in this study was to connect genomics to breeding. There is a large gap between plant genomics and breeding, even though breeding has been considered one of the most important and expected outcomes of genomics. Although various reasons for this discord could be hypothesized, financial and technical difficulties are among the leading issues. As one of the attempts to fill this gap, we insisted on a cost-effective and simple marker system which is user-friendly to breeders. We believe that PCR-based microsatellite markers, together with the detection system adopting agarose gels rather than acrylamide or capillary gels, would facilitate interchange of information and technologies between the two contrastive research fields. Breeding is often compared to an art because it pursues the creation of ‘ideal genotypes’ which have never existed. It is said that a great artistic work is born from a mixture of essential knowledge and sensibility. In this study, basic knowledge on the genome structure of red clover as well as material resources has been provided. We hope that these, mixed with the sensibility of the breeders, will contribute to the birth to fabulous new varieties in the near future.


   Supplementary Material
Top
 Abstract
 1. Introduction
 2. Materials and Methods
 3. Results
 4. Discussion
 Supplementary Material
Appendix
 Acknowledgements
 References
 
Supplementary material with additional information is available online at http://www.kazusa.or.jp/en/plant/redclover/marker/ or http://dnaresearch.oxfordjournals.org


   Appendix
Top
 Abstract
 1. Introduction
 2. Materials and Methods
 3. Results
 4. Discussion
 Supplementary Material
 Appendix
Acknowledgements
 References
 


   Acknowledgements
Top
 Abstract
 1. Introduction
 2. Materials and Methods
 3. Results
 4. Discussion
 Supplementary Material
 Appendix
 Acknowledgements
References
 
This work was supported by the Kazusa DNA Research Institute Foundation and NARCH, supported by the Ministry of Agriculture, Forestry and Fisheries with the cooperation of the Rice Genome Research Program.


   Footnotes
 
*To whom correspondence should be addressed. Tel. +81-438-52-3933, Fax. +81-438-52-3934, E-mail: tabata{at}kazusa.or.jp

{dagger}These authors contributed equally to the work Back

Communicated by Masahiro Yano


   References
Top
 Abstract
 1. Introduction
 2. Materials and Methods
 3. Results
 4. Discussion
 Supplementary Material
 Appendix
 Acknowledgements
 References
 

  1. Young, N. D., Cannon, S. B., Sato, S., et al. 2005, Sequencing the genespaces of Medicago truncatula and Lotus japonicus, Plant Physiol., 137, 1174–1181.[Free Full Text]
  2. Choi, H. K., Kim, D., Uhm, T., et al. 2004, A sequence-based genetic map of Medicago truncatula and comparison of marker colinearity with M. sativa, Genetics, 166, 1463–1502.[Abstract/Free Full Text]
  3. Choi, H. K., Mun, J. H., Kim, D. J., et al. 2004, Estimating genome conservation between crop and model legume species, Proc. Natl Acad. Sci. USA, 101, 15289–15294.[Abstract/Free Full Text]
  4. Yan, L. Y., Xu, Z. Y., Goldbach, R., Kunrong, C., Prins, M. 2005, Nucleotide sequence analyses of genomic RNAs of Peanut stunt virus Mi, the type strain representative of a novel PSV subgroup from China, Arch Virol., 150, 1203–1211.[CrossRef][Web of Science][Medline]
  5. Gutierrez, M. V., Vaz Patto, M. C., Huguet, T., Cubero, J. I., Moreno, M. T., Torres, A. M. 2005, Cross-species amplification of Medicago truncatula microsatellites across three major pulse crops, Theor. Appl. Genet., 110, 1210–1217.[CrossRef][Web of Science][Medline]
  6. Kaló, P., Seres, A., Taylor, S. A., et al. 2004, Comparative mapping between Medicago sativa and Pisum sativum, Mol. Genet. Genomics, 272, 235–246.[CrossRef][Web of Science][Medline]
  7. Arumuganathan, K. and Earle, E. D. 1991, Nuclear DNA content of some important plant species, Plant Mol. Biol. Rep., 9, 208–218.
  8. Milligan, B. G. 1991, Chloroplast DNA diversity within and among population of Trifolium pratense, Curr. Genet., 19, 411–416.[CrossRef]
  9. Kongkiatngam, P., Waterway, M. J., Fortin, M. G., Coulman, B. E. 1995, Genetic variation within and between two cultivars of red clover (Trifolium pratense L.): Comparisons of morphological, isozyme, and RAPD markers, Euphytica, 84, 237–246.
  10. Campos-de-Quiroz, H. and Ortega-Klose, F. 2001, Genetic variability among elite red clover (Trifolium pratense L.) parents used in Chile as revealed by RAPD markers, Euphytica, 122, 61–67.[CrossRef]
  11. Kölliker, R., Herrmann, D., Boller, B., Widmer, F. 2003, Swiss Mattenklee landraces, a distinct and diverse genetic resource of red clover (Trifolium pratense L.), Theor. Appl. Genet., 107, 306–15.[CrossRef][Web of Science][Medline]
  12. Herrman, D., Boller, B., Widmer, F., Kölliker, R. 2005, Optimization of bulked AFLP analysis and its application for exploring diversity of natural and cultivated populations of red clover, Genome, 48, 474–486.[Medline]
  13. Isobe, S., Klimenko, I., Ivashuta, S., Gau, M., Kozlov, N. N. 2003, First RFLP linkage map of red clover (Trifolium pratense L.) base on cDNA probes and its transferability to other red clover germplasm, Theor. Appl. Genet, 108, 105–112.[CrossRef][Web of Science][Medline]
  14. Barrett, B., Griffiths, A., Schreiber, M., et al. 2004, A microsatellite map of white clover, Theor. Appl. Genet., 109, 596–608.[Web of Science][Medline]
  15. Yamaguchi, H., Sawai, A., Gau, M., et al. 2000, Breeding of red clover ‘Hokuseki’ and its characteristics, Res. Bull. Hokkaido Natl. Agric. Exp. Stn., 171, 17–32.
  16. Galbraith, D. W., Harkins, K. R., Maddox, J. M., Ayres, N. M., Sharma, D. P., Firoozabady, E. 1983, Rapid flow cytometric analysis of the cell cycle in intact plant tissues, Science, 220, 1049–1051.[Abstract/Free Full Text]
  17. Ito, M., Miyamoto, J., Mori, Y., et al. 2000, Genome and chromosome dimensions of Lotus japonicus, J. Plant Res., 113, 435–442.[CrossRef]
  18. Sato, S., Kaneko, T., Nakamura, Y., Asamizu, E., Kato, T., Tabata, S. 2001, Structural analysis of a Lotus japonicus genome. I. Sequence features and mapping of fifty-six TAC clones which cover the 5.4 Mb regions of the genome, DNA Res., 8, 311–318.[Abstract]
  19. Ohmido, N. and Fukui, K. 2002, FISH analysis, In Murooka, Y. (Ed.). Manual on Biotechnology Tokyo Baifukan Co., Ltd pp. 187–193.
  20. Ohmido, N., Kijima, K., Ashikawa, I., de Jong, J. H., Fukui, K. 2001, Visualization of the terminal structure of rice chromosomes 6 and 12 with multicolor FISH to chromosomes and extended DNA fibers, Plant Mol. Biol., 47, 413–421.[CrossRef][Web of Science][Medline]
  21. Kato, S., Ohmido, N., Fukui, K. 2003, Development of a quantitative pachytene chromosome map in Oryza sativa by imaging methods, Genes Genet. Syst., 78, 155–161.[CrossRef][Web of Science][Medline]
  22. Asamizu, E., Nakamura, Y., Sato, S., Fukuzawa, H., Tabata, S. 1999, A large scale structural analysis of cDNAs in a unicellular green alga, Chlamydomonas reinhardtii. I. Generation of 3433 non-redundant expressed sequence tags, DNA Res., 6, 369–373.[Abstract]
  23. Ewing, B., Hillier, L., Wendl, M. C., Green, P. 1998, Base-calling of automated sequencer traces using phred. I. Accuracy assessment, Genome Res., 8, 175–185.[Abstract/Free Full Text]
  24. Ewing, B. and Green, P. 1998, Base-calling of automated sequencer traces using phred. II. Error probabilities, Genome Res., 8, 186–194.[Abstract/Free Full Text]
  25. Kijas, J. M., Fowler, J. C., Garbett, C. A., Thomas, M. R. 1994, Enrichment of microsatellites from the citrus genome using biotinylated oligonucleotide sequences bound to streptavidin-coated magnetic particles, Biotechniques, 16, 656–660 662.[Web of Science][Medline]
  26. Rabinowicz, P. D., Schutz, K., Dedhia, N., et al. 1999, Differential methylation of genes and retrotransposons facilitates shotgun sequencing of the maize genome, Nat. Genet., 23, 305–308.[CrossRef][Web of Science][Medline]
  27. Steve, R. and Helen, J. S. 2000, Primer3 on the WWW for general users and for biologist programmers, In Krawetz, S. and Misener, S. (Eds.). Bioinformatics Methods and Protocols: Methods in Molecular Biology Totowa, NJ, USA Humana Press pp. 365–386.
  28. Don, R. H., Cox, P. T., Wainwright, B. J., Baker, K., Mattick, J. S. 1991, Touchdown PCR to circumvent spurious priming during gene amplification, Nucleic Acids Res., 19, 4008.[Free Full Text]
  29. Doyle, J. J. and Doyle, J. L. 1990, Isolation of plant DNA from fresh tissue, Focus, 12, 13–15.
  30. Kiss, G. B., Kereszt, A., Kiss, P., Endre, G. 1998, Colormapping; a non-mathematical procedure for genetic mapping, Acta Biol. Hung., 49, 125–142.
  31. Bennett, M. D., Leitch, I. J., Price, H. J., Johnston, J. S. 2003, Comparison with Caenorhabditis (~100 Mb) and Drosophila (~175 Mb) using flow cytometry show genome size in Arabidopsis to be ~157 Mb and thus ~25% larger than the Arabidopsis Genome Initiative estimate of ~125 Mb, Ann. Botany, 91, 547–557.[Abstract/Free Full Text]
  32. Bennett, M. D., Bhandol, P., Leitch, I. J. 2000, Nuclear DNA amounts in angiosperms and their modern uses—807 new estimates, Ann. Botany, 86, 859–909.[Abstract/Free Full Text]
  33. Cavalier-Smith, T. 1985, The Evolution of Genome Size, Chichester, UK John Wiley & Sons Ltd.
  34. Taylor, N. L. and Chen, K. 1988, Isolation of trisomics from crosses of diploid, triploid, and tetraploid red clover, Crop Sci., 28, 209–213.
  35. International Rice Genome Sequencing Project. 2005, The map-based sequence of the rice genome, Nature, 436, 793–800.[CrossRef][Medline]
  36. McCouch, S. R., Teytelman, L., Xu, Y., et al. 2002, Development and mapping of 2240 New SSR markers for rice (Oryza sativa L.), DNA Res., 9, 199–207.[Abstract]
  37. Menz, M. A., Klein, R. R., Mullet, J. E., Obert, J. A., Unruh, N. C., Klein, P. E. 2002, A high-density genetic map of Sorghum bicolor (L.) Moench based on 2926 AFLP, RFLP and SSR markers, Plant Mol. Biol., 48, 483–499.[CrossRef][Web of Science][Medline]
  38. Song, Q. J., Marek, L. F., Shoemaker, R. C., et al. 2004, A new integrated genetic linkage map of the soybean, Theor. Appl. Genet., 109, 122–128.[CrossRef][Web of Science][Medline]
  39. Hayashi, M., Miyahara, A., Sato, S., et al. 2001, Construction of a genetic linkage map of the model legume Lotus japonicus using an intraspecific F2 population, DNA Res., 8, 301–310.[Abstract]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Brief BioinformHome page
I. Armstead, L. Huang, A. Ravagnani, P. Robson, and H. Ougham
Bioinformatics in the orphan crops
Brief Bioinform, November 1, 2009; 10(6): 645 - 653.
[Abstract] [Full Text] [PDF]

Home page
 DNA ResHome page
T.-Y. Hwang, T. Sayama, M. Takahashi, Y. Takada, Y. Nakamoto, H. Funatsuki, H. Hisano, S. Sasamoto, S. Sato, S. Tabata, et al.
High-density Integrated Linkage Map Based on SSR Markers in Soybean
DNA Res, August 1, 2009; 16(4): 213 - 225.
[Abstract] [Full Text] [PDF]

Home page
 Plant Physiol.Home page
MichaelL. Sullivan
A Novel Red Clover Hydroxycinnamoyl Transferase Has Enzymatic Activities Consistent with a Role in Phaselic Acid Biosynthesis
Plant Physiology, August 1, 2009; 150(4): 1866 - 1879.
[Abstract] [Full Text] [PDF]

Home page
 DNA ResHome page
S. Sato, Y. Nakamura, T. Kaneko, E. Asamizu, T. Kato, M. Nakao, S. Sasamoto, A. Watanabe, A. Ono, K. Kawashima, et al.
Genome Structure of the Legume, Lotus japonicus
DNA Res, August 1, 2008; 15(4): 227 - 239.
[Abstract] [Full Text] [PDF]

Home page
 GeneticsHome page
B. K. Hougaard, L. H. Madsen, N. Sandal, M. de Carvalho Moretzsohn, J. Fredslund, L. Schauser, A. M. Nielsen, T. Rohde, S. Sato, S. Tabata, et al.
Legume Anchor Markers Link Syntenic Regions Between Phaseolus vulgaris, Lotus japonicus, Medicago truncatula and Arachis
Genetics, August 1, 2008; 179(4): 2299 - 2312.
[Abstract] [Full Text] [PDF]

Home page
 Crop Sci.Home page
D. Herrmann, B. Boller, B. Studer, F. Widmer, and R. Kolliker
Improving Persistence in Red Clover: Insights from QTL Analysis and Comparative Phenotypic Evaluation
Crop Sci., January 16, 2008; 48(1): 269 - 277.
[Abstract] [Full Text] [PDF]

Home page
 DNA ResHome page
H. Hisano, S. Sato, S. Isobe, S. Sasamoto, T. Wada, A. Matsuno, T. Fujishiro, M. Yamada, S. Nakayama, Y. Nakamura, et al.
Characterization of the Soybean Genome Using EST-derived Microsatellite Markers
DNA Res, January 11, 2008; (2008) dsm025v1.
[Abstract] [Full Text] [PDF]

Home page
 DNA ResHome page
S. Isobe, A. Nakaya, and S. Tabata
Genotype Matrix Mapping: Searching for Quantitative Trait Loci Interactions in Genetic Variation in Complex Traits
DNA Res, November 13, 2007; (2007) dsm020v1.
[Abstract] [Full Text] [PDF]

Home page
 Plant Physiol.Home page
S. Sato, Y. Nakamura, E. Asamizu, S. Isobe, and S. Tabata
Genome Sequencing and Genome Resources in Model Legumes
Plant Physiology, June 1, 2007; 144(2): 588 - 593.
[Full Text] [PDF]

Home page
 BioinformaticsHome page
B. E. Suzek, H. Huang, P. McGarvey, R. Mazumder, and C. H. Wu
UniRef: comprehensive and non-redundant UniProt reference clusters
Bioinformatics, May 15, 2007; 23(10): 1282 - 1288.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary data
Right arrowOA All Versions of this Article:
12/5/301    most recent
dsi018v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (12)
Right arrow Request Permissions
Google Scholar
Right arrow Articles by Sato, S.
Right arrow Articles by Tabata, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sato, S.
Right arrow Articles by Tabata, S.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?