Funct Integr Genomics (2004) 4: 231–240DOI 10.1007/s10142-004-0116-z

ORIGINAL PAPER

Zhongyi Li . Bingyan Huang . Lynette Rampling .Jun Wang . Jun Yu . Matthew Morell .Sadequr Rahman

Detailed comparison between the wheat chromosome group7 short arms and the rice chromosome arms 6S and 8L withspecial reference to genes involved in starch biosynthesisReceived: 12 November 2003 / Revised: 26 March 2004 / Accepted: 20 April 2004 / Published online: 6 August 2004# Springer-Verlag 2004

Abstract Rice bacterial artificial chromosome (BAC)clones have been identified that contain sequencesorthologous to each EST localized to wheat chromosome7AS deletion stocks by Southern blot hybridization. Thisinformation has been used to relate the DNA sequenceincluded in each wheat deletion stock to a complement ofrice BACs. A virtual contig was used that covered 90 cM(21 Mb) of DNA sequence (with a gap for the 6S/8Ljunction). Comparison of the positions of orthologousgenes on the rice virtual contig and on wheat chromosome7AS showed that there was an unexpectedly low level ofsynteny (31.4%) and a high level of chromosomerearrangements (68.6%). The non-syntenous loci were oftwo classes: wheat and rice genes found at differentlocations in the genome (32.6%), and ESTs in wheat notpresent in rice (36.0%). Four starch synthetic genes,GBSSI, SSI, SSIIa and DBEI, were located at similarpositions on wheat chromosome 7AS and the virtual ricecontig covering wheat chromosome 7AS. A preliminarycomparison between the short arms of chromosome 7Aand 7D in wheat showed that both chromosomes had asimilar level of sequence synteny with rice. Therefore,there appears to be considerable variation in gene orderbetween wheat chromosome 7S and rice chromosome 6Sand 8L.

Introduction

Wheat and rice are the most important food crops for theworld as a whole. However, they present a remarkablecontrast in their genetic structure. Wheat has a complexgenome, consisting of three sets of seven chromosomeswith an estimated genome size of 1010 bp. In contrast, ricehas 12 chromosomes with an estimated genome size of2.5×108 bp. Despite this discrepancy in size, syntenicrelationships have been detected in genome organization,and it has been proposed that the rice genome could beused as a framework for the genetic analysis of all cerealgenomes (Moore et al. 1995). Wheat chromosome group 7has been proven to be syntenic with rice chromosomes 6and 8 by restriction fragment length polymorphism(RFLP) mapping (Gale and Devos 1998). Through thecomparative genomics between wheat ESTs (wESTs) andrice genome DNA it has been determined that wheatchromosome group 7 short arms (7S) are orthologous tothe short arm of chromosome 6 (6S) and long arm ofchromosome 8 (8L) of rice (Sorrells et al. 2003).

A large number of rice expressed sequence tags (ESTs)have been mapped on the bacterial artificial chromosome(BAC; http://www.tigr.org/tdb/e2k1/osa1/). More than28,000 full-length cDNA clones were completely se-quenced and mapped to the rice genome in the publicdatabase (Kikuchi et al. 2003). Although the entiregenome sequence of wheat is not available, many genomicDNA sequences have been determined through thesequencing of specific genes (e.g. Li et al. 2003; Suzukiet al. 2003; Rostoks et al. 2002). More than 67,900 wheatunique ESTs have been sequenced in the TIGR database(http://www.tigr.org/tdb/tgi/plant.shtml). These ESTscover different developmental stages including youngleaves, flower, immature seeds and mature seeds. Theyalso cover many of the ESTs induced by biotic and abioticstress (http://wheat.pw.usda.gov/wEST/). By February2004, a subset of these ESTs (over 16,000) had beenmapped in the wheat genome by hybridization to a set ofwheat chromosome deletion lines, placing the ESTs inchromosome bins (U.S. National Science Foundation

Z. Li (*) . L. Rampling . M. Morell . S. RahmanCSIRO Plant Industry,G. P. O Box 1600 Canberra, ACT, 2601, Australiae-mail: zhongyi.li@csiro.au

B. HuangHenan Academy of Agricultural Sciences,Zhengzhou, Henan, People’s Republic of China

J. Wang . J. YuBeijing Genome Institute,Beijing, People’s Republic of China

wheat EST project, http://wheat.pw.usda.gov/index.shtml).This rapid accumulation of genomic DNA sequences andESTs in rice, wheat and other cereals generates anexcellent opportunity for comparative genomics in cereals.

In this paper, we report the comparison of rice BACclones from 90 cM (21 Mb) of rice chromosomes 6S and8L (http://www.tigr.org/tdb/e2k1/osa1/) and wESTs lo-cated on chromosome 7S of the A genome (7AS; http://wheat.pw.usda.gov/index.shtml) using tBlastX search. Therice chromosomes 6S and 8L were assembled as a virtualcontig that corresponds to wheat chromosome 7AS. WheatESTs in wheat chromosome 7AS deletion line bins(Sorrells et al. 2003; http://wheat.pw.usda.gov/index.shtml) were then located on the virtual rice contig. Theanalytical results provide further information at themolecular level regarding chromosome colinearity andsynteny between wheat and rice.

Materials and methods

Mapping of wheat cDNAs encoding GBSSI, SSI, SSIIa andDBEI onto rice chromosomes 6S and 8L

Wheat cDNAs encoding granule bound starch synthase I (GBSSI),starch synthase I (SSI), starch synthase IIa (SSIIa) and starchdebranching enzyme I (DBEI) were used to search for rice BACclones using the tBlastX and tBlastN program of the TIGR databasesequence search. wESTs corresponding to those genes were alsoused for sequence comparison using tBlastX and tBlastN with BACclones on rice chromosomes 6S and 8L.

Rice BAC clone arrays along rice chromosomes 6S and 8L

Rice BAC clones used in this research were from the TIGR website(http://www.tigr.org/tdb/e2k1/osa1/). The order of the rice BACarray was determined by the International Rice Genome SequencingProject (IRGSP: http://rgp.dna.affrc.go.jp/cgi-bin/statusdb/irgsp-sta-tus.cgi).

Assembly of wheat chromosome 7AS

The virtual rice contig composed of rice chromosomes 6S and 8Lcorresponding to wheat chromosome 7AS was assembled based onthe information of starch synthesis genes (Li et al. 2003) as well asthe comparison between the ESTs in wheat chromosome 7ASdeletion bins and BAC clones on rice chromosomes 6S and 8L(http://www.tigr.org/tdb/e2k1/osa1/; http://wheat.pw.usda.gov/index.shtml).

Comparison of the wheat ESTs of chromosome 7AS and riceBAC clones

Wheat ESTs (316 wESTs) from the US NSF wheat EST projectwebsite (http://wheat.pw.usda.gov/index.shtml) were separated infour bins of wheat chromosome 7AS or 7DS (Qi et al. 2003; Sorrellset al. 2003). Wheat ESTs were used for the sequence search of riceBAC clones in the TIGR database by tBlastX. Those wESTs thatshowed a high homology (<1×10−15) with rice BAC clones werethen assigned along rice chromosomes 6S and 8L.

Mapping of genetic markers on wheat chromosome 7AS ontorice chromosomes 6S and 8L

To determine the sections of rice chromosomes 6S and 8L for theassembly of wheat chromosome 7AS, 11 RFLP markers on wheatchromosome 7AS were selected for a search of the TIGR database tomap the markers on rice chromosomes 6S and 8L. Seven of theseRFLP markers (Xabg704a, Xbcd873a, Xcdo545b, Xbcd907b,Xbcd135a, Xcdo1395, Xabc158b) were from the GrainGenesdatabase (http://www.graingenes.org/cgi-bin/WebAce/webace?db=graingenes&class=Probe&object), three (Xcdo475a, accessionnumber AA231764; Xcdo673, accession number BE439224; andXcdo962b, accession number AA231726) were from GenBank, andone is from a cDNA sequence: Xpsr311a (CGAACACGAGTGCACGAACT GCACGATTTG AGTTGGGTCA TAGCCTAC-GA GTCTTCTGGG CCGGAGCCGT AATAACTTGT GTAAGC-CATG TCGAGTGTGT GTGGTAGATA CTTAAGCCAACGATGCCACT GTTAGGGTTC ATTTGAATAA GAAAGT-GAAT TGATCTCCAA CCTCTGTTGC TTCATCCCCCCAACTGCTCC TCTCGTCTCA CCCAAATCTC GGCTCCGATCTCTCTCCCTC TT).

Comparison between the rice and barley BAC clonescontaining GBSSI genes

Two BAC clones (GenBank accession number AF147343 for riceand AF474373 for barley) were first compared using the NCBIPairwise BLASTN (BLAST 2 Sequences) program. The rice BACAF147343 was fully annotated and was a part of BAC AP002542used for the assembly of rice chromosome 6S. The coding regionsencoded by barley BAC AF474373 were then used to search theTIGR rice database to find the BAC clones which had highhomology with those regions of barley DNA sequences.

Comparison of the genes encoding GBSSI, SSI, SSIIa andDBEI between rice and wheat

The gene sequences encoding starch synthetic enzymes in rice werethose from rice BAC clones AF473737 (homology to AP002542)for GBSSI, AB026295 for SSI, AP003509 for SSIIa and AP005509for DBEI. The genomic DNA sequences in wheat were AB019624for GBSSI, AF091802 for SSI, AY133248 for SSIIa and AF548379for DBEI. The intron and exon structures of these genes have beendetermined previously (Li et al. 2003; Suzuki et al. 2003; Fig. 1).

Results and discussion

Localization of wheat cDNAs encoding GBSSI, SSI,SSIIa and DBEI on rice chromosomes

Criteria for matching

Wheat chromosome 7AS was sectioned as four chromo-some bins (http://wheat.pw.usda.gov/cgi-bin/gbrowse?source=japonica/). Bin 1 corresponds to chromosomedeletion line 7AS1-0.89–1.00, bin 2 corresponds to7AS5-0.59–0.89, bin 3 corresponds to 7AS8-0.45–0.59and bin 4 corresponds to C-7AS8-0.45. To establish areliable method for sequence comparison between wheatcDNAs (or ESTs) and rice genomic DNA BAC clones,wheat cDNAs encoding GBSSI, SSI, SSIIa and DBEIwere selected as those genes that have been mapped onwheat chromosome 7S by our previous work using PCR

232

markers and Chinese Spring chromosome deletion lines(Li et al. 2003). These four genes are distributed in fourchromosome bins on 7AS. GBSSI is in bin 1, SSI is in bin2, SSIIa is in bin 3 and DBEI is in bin 4.

Wheat ESTs and cDNAs encoding GBSSI, SSI, SSIIaand DBEI were used to search rice BAC clones in TIGRrice genome databases (http://wheat.pw.usda.gov/wEST/)using tBlastN and tBlastX (Table 1). Multiple sequences

were detected from all searches using either ESTs orcDNAs for all four genes. However, the highest homologybetween wheat cDNAs (or wheat ESTs) and rice BACsalways corresponds to orthologous genes in both species(Table 1). Although there is some difference of E-valuebetween tBlastN and tBlastX search, it convincinglyshowed that both tBlastN and tBlastX search will producereliable data for the sequence comparison between wheatcDNA (or ESTs) and rice BACs. Therefore, use of wheatESTs to search rice BACs using tBlastX search in TIGRrice genome databases (http://www.tigr.org/tdb/e2k1/osa1/) was selected in this work. E-score values lowerthan 1×10−15 between wESTs and rice BAC clones wereconsidered to represent orthologous sequences. Sequenceswith E-scores higher than 1×10−15 were considered to benon-orthologous sequences. For example, wheat SSIcDNA has the lowest E-value with rice BAC clonescontaining SSI, although it also has a low E-value withother genes encoding starch synthase SSII-1, SSII-2, SSII-3 and GBSSI in rice BACs. When multiple BAC clonesmatched to one wheat EST (<1×10−15), it indicated that awheat EST has multiple isoforms of that gene in rice.

Localization of wheat cDNA encoding GBSSI, SSI,SSIIa and DBEI on rice chromosomes

Through tBlastX search wheat GBSSI cDNA has anorthologous gene in rice BAC clone AP002542 at 7.4 cMof chromosome 6S. Wheat SSI cDNA has homology withrice BAC clone AB026295 at 12.9 cM (6S). Wheat SSIIacDNA is homologous to rice BAC clone AP003509 at34.3 cM (6S). Wheat DBEI cDNA matched to rice BACclone AP005509 at 106.1 cM of chromosome 8L. Theresults confirmed the available information that wheatchromosome 7AS is homologous to rice chromosomes 6Sand 8L. Localization of these genes created a frame and

Fig. 1 Comparison of intron and exon structure of granule boundstarch synthase I (GBSSI), starch synthase I (SSI), starch synthase IIa(SSIIa) and starch debranching enzyme I (DBEI) between wheat andrice. Rice GBSSI Genomic DNA from rice BAC, GenBank numberAF488413 (homology with AP000559); Wheat GBSSI genomicDNA sequence from wheat (AB019624); Rice SSI genomic DNAfrom rice BAC (AB026295); Wheat SSI genomic DNA sequencefrom Aegilops tauschii (AF091802); Rice SSIIa genomic DNA fromrice BAC (AP003509); Wheat SSIIa genomic DNA sequence fromA. tauschii (AY133248); Rice DBEI genomic DNA sequence fromrice isoamylase gene (AB093426); Wheat DBEI genomic DNAsequence from A. tauschii isoamylase gene (AF548379)

Table 1 Comparison tBlastNand tBlastX search of theTIGR rice genomic DNA data-base with full length cDNAs andESTs of granule bound starchsynthase I (GBSSI), starch syn-thase I (SSI), starch synthase IIa(SSIIa) and starch debranchingenzyme I (DBEI)

Genes Wheat cDNA orEST

GenBankaccessionnumber

tBlastN tBlastX Rice BACs GenBank accessionnumber

GBSSI cDNA X57233 6.7×10−184

4.0×10−180

AP002542

EST AW44881 8.2×10−63

4.8×10−62

AP002542

SSI cDNA AF091803 2.1×10−226

3.2×10−217

AB026295

EST CD894386 1.1×10−87

6.4×10−87

AB026295

SSIIa cDNA AF155217 4.7×10−252

2.7×10−260

AP003509

EST BM137830 8.8×10−81

2.6×10−86

AP003509

DBEI cDNA AF548380 1.6×10−246

7.2×10−253

AP005509

EST BQ609582 6.8×10−64

5.0×10−67

AP005509

233

anchor for the sequence comparison between these twocereal species.

Synteny between wheat chromosome 7S and itscorresponding rice chromosomes 6S and 8L

Localization of wheat ESTs on rice chromosomes

More than 300 wESTs located on chromosome 7AS wereused to search the BAC clones on rice chromosomes 6S or8L. Of these, 118 wESTs were apparent orthologs of riceBAC clone sequences from chromosomes 6S and 8L(Fig. 2, Table 2). Most of the wESTs from bin 1 to bin 3 ofwheat matched the rice BAC clones corresponding to ricechromosome 6S, whereas wESTs in bin 4 of wheat mostlymatched to the BAC clones from rice chromosome 8L(Fig. 2).

Most of the wESTs in bin 1 that matched to ricechromosome 6S corresponded to BAC clones located atthe distal region of rice chromosome 6S covering riceBACs AP005621 to AP002542. However, four of themmatched to the BAC clones located at the central region(from BACs AP004234 to AP004727) and the proximalregion (from BACs AP005545 to AP003527) of ricechromosome 6S (Fig. 2).

In bin 2, the majority of wESTs that matched to ricechromosomes 6S corresponded to rice BAC clones in thecentral region of rice chromosome 6S. Two of themmatched to the distal region of rice chromosome 6S andanother six matched to rice BAC clones at the proximalregion of rice chromosome 6S (Fig. 2).

In bin 3, all wESTs which matched to rice chromosomes6S and 8L did so to rice BAC clones located at theproximal region of rice chromosome 6S (Fig. 2), exceptfor three which had homologous sequence to a rice BACclone on rice chromosome 8L.

In bin 4, one wEST matched to a BAC clone at theproximal region of rice chromosome 6S, and all otherwESTs had a high homology to the rice BAC clones onchromosome 8L (Fig. 2).

Creation of a virtual rice contig covering wheatchromosome 7AS

The comparison between the wESTs and rice BAC clonesclearly showed that wheat chromosome 7AS containssequences that have high identity to those found in ricechromosomes 6S and 8L (Gale and Devos 1998; Sorrellset al. 2003). wESTs have been mapped onto deletionstocks of wheat chromosomes and the results collated(http://wheat.pw.usda.gov/wEST/; Sorrells et al. 2003).Although the available analysis gives the location ofwESTs in four bins on wheat chromosome 7AS, it doesnot give the gene order for wESTs on wheat chromosome7AS as there is limited information on the genomesequence from that part of the wheat chromosome. Inorder to investigate the potential gene order on wheatchromosome 7AS, a virtual rice contig covering regions ofthe rice genome thought to be syntenous to wheatchromosome 7AS was created. The regions of ricechromosomes 6S [from 0.0 cM to 65.8 cM (17 Mb)covering rice BACs AP005621 to AP003527] and 8L[from 96.6 cM to 121.2 cM (4 Mb) covering rice BACsAP005411 to AP004557] were selected for the assemblyof wheat chromosome 7AS using the relevant rice BACclones (Fig. 2). Previous studies have suggested thatsequences present in rice chromosome 8 are adjacent tosequences present on rice chromosome 6 at the centromereregion in representations of wheat chromosome 7 (Galeand Devos 1998; Sorrells et al. 2003; Li et al. 2003).However, the orientation of rice chromosome 8L in wheat

Table 2 Synteny and its break-down between rice chromosomes 6S and 8L and wheat chromosome 7AS

Bins No. ofwESTs(a=b+d+e)

wESTs that map torice BACs onchromosome 6 and8 (identity <10−15)(b)

wESTs that map to thecorresponding sectionof rice BACs onchromosome 6 and 8(c)

wESTs that map to riceBACs on a non-corresponding section ofrice chromosome 6 and 8(identity <10−15) (d)

wESTs that donot map to riceBACs(identity >10−15) (e)

wESTsthat wereconservedin rice (c/a) %

wESTsthatrearrangedin rice (b−c+d/a) %

wESTs thatdo notmatch riceBACs (e/a) %

1 86 15 13 35 36 15.1 43.0 41.92 129 43 39 42 44 30.2 35.7 34.13 57 33 24 3 21 42.1 21.1 36.84 44 27 24 6 11 54.5 20.5 25.0Total 316 118 100 86 112 31.6 32.9 35.5

"Fig. 2 The distribution of the wESTs on wheat chromosome 7AS.A virtual contig covering wheat chromosome 7S was assembledfrom two contigs for rice chromosome 6S (thick open box) and 8L(thick filled box). The thin bars represented the four bins of wheatchromosome 7AS (refer to http://wheat.pw.usda.gov/wEST/): bin 1candidates in chromosome deletion lines of Chinese Spring 7AS1-0.89–1.00; bin 2 candidates in chromosome deletion lines ofChinese Spring 7AS5-0.59–0.89; bin 3 candidates in chromosomedeletion lines of Chinese Spring 7AS8-0.45–0.59; and bin 4candidates in chromosome deletion lines of Chinese Spring C-7AS8-0.45. Rice BAC All rice BAC clones used to form thecontiguous contig, wheat ESTs wESTs had a homology with riceBAC clones on rice chromosome 6S or 8L, Bin the location of thewESTs in the chromosome bins, M multiple rice sequences with anE-value <1×10−15, S a single rice sequence with an E-value<1×10−15, asterisk the rice BAC used has low homology (<1×10−15)but not the lowest one, IdentityE-value by tBlastX search

234

235

chromosome 7AS was not clear. In this study, wESTs,RFLP and microsatellite markers were used in an attemptto determine the orientation of rice chromosome 8L inrespect to wheat chromosome 7AS. However, no clearpicture emerged as a limited number of wESTs (Fig. 2),RFLP and microsatellite markers from bin 3 and 4 inwheat matched to rice BAC clones in the proximal regionof rice chromosome 6S and both ends of the contigproduced from rice chromosome 8L. Thus, there are noclear details of the apparent “junction” of these two partsof rice chromosomes in the recreation of wheat chromo-some 7AS.

Synteny between wheat chromosome 7S and ricechromosomes 6S and 8L

Recently, 4,485 wESTs were used for the localization ofhomologous genes on rice chromosomes (Sorrells et al.

2003). The work showed that there was a mosaicchromosome structure in wheat when the wESTs fromchromosome 7 were used for the comparison with riceBAC clones from chromosomes 6 and 8. In total, 28wESTs matched to rice BAC clones from chromosome 6Sand 11 wESTs matched to rice BAC clones fromchromosome 8L (supplementary material of Sorrells etal. 2003).

In this work, the available genome DNA sequencesfrom Japonica rice (IRGSP) were used to assemble avirtual contig of wheat chromosome 7AS, which coveredtwo contigs, 65.8 cM (17 Mb) of chromosome 6S and24.6 cM (4 Mb) of chromosome 8L. The virtual rice DNAcontig was then used for comparison with 316 wESTsplaced on wheat chromosome 7AS by Southern blothybridization using deletion stocks (http://wheat.pw.usda.gov/index.shtml). Rice chromosomes 6 and 8 have beenpreviously reported to have a high level of synteny withwheat chromosome 7 (Kurata et al. 1994; Moore et al.

Fig. 3 Wheat ESTs on wheatchromosome 7AS have ortholo-gous sequences within the ricegenome but not on rice chro-mosomes 6S or 8L. A virtualcontig covering wheat chromo-some 7AS was assembled fromtwo contigs for rice chromo-somes 6S (thick open box) and8L (thick filled box). The thinbars represented the four bins ofwheat chromosome 7AS (as inFig. 2). Rice BAC Rice BACclones had a homology withwESTs from rice chromosomes,wheat ESTs wESTs had a ho-mology with rice BAC cloneson the rice chromosomes ratherthan rice chromosomes 6S or8L, Identity the sequence iden-tity between the wESTs and therice BAC clones, Chr No thechromosome number of rice

236

1995; Gale and Devos 1998; Sorrells et al. 2003; Li et al.2003).

In total, 112 wESTs were homologs of the genomicDNA on the virtual rice contig representing wheatchromosome 7AS through the sequence comparisonbetween wESTs on chromosome 7AS (http://wheat.pw.usda.gov/index.shtml) and available rice genome se-quences (http://www.tigr.org/tdb/e2k1/osa1/; Fig. 2). Thedata showed that at present approximately 31.6% wESTsplaced on wheat chromosome 7AS bins analysed weresequence matched to sequences contained on the rice 6S-8L virtual contig. It is quite surprising that there is suchlow sequence homology between wESTs and rice BACclones in the chromosome regions compared. However, allfour genes involved in starch biosynthesis that had been

located to 7AS (Li et al. 2003) were located on the ricevirtual contig.

Breakdown of colinearity and synteny between wheatand rice chromosomes

Wheat ESTs not located on rice chromosomes 6S and8L

Although nearly 120 wESTs from the four bins matched tothe BAC clones aligned to chromosomes 6S and 8L of rice(Fig. 2, Table 2), 86 wESTs had high homologoussequences with other rice chromosomes (Fig. 3, Table 2)

Fig. 4 Wheat ESTs on wheatchromosome 7AS have noorthologous sequences withinthe rice genome. A virtual contigcovering wheat chromosome7AS was assembled from twocontigs for rice chromosome 6S(thick open box) and 8L (thickfilled box). The thin bars repre-sented the four bins of wheatchromosome 7AS (as in Fig. 2).wheat ESTs wESTs had no ho-mology with rice BAC cloneson the rice chromosomes, Iden-tity the sequence identity be-tween the wESTs and the riceBAC clones

237

including chromosomes 6L and 8S. These wESTs matchedto BAC clones covering all rice chromosomes.

Comparison with the data provided in the supplemen-tary material by Sorrells et al. (2003) showed that therewere many wESTs from wheat chromosome 7S that hadorthologous genes on rice chromosomes 1, 4, 5 and 12(Fig. 4). A detailed analysis of the location of such genesin rice indicated that they were not derived from anyparticular region of the chromosome except for two(orthologous to wESTs BE590763 and BF201979), whichwere within 288 kb of each other on rice chromosome 1.

Wheat ESTs not located in the rice genome

Out of 316 wESTs used for the sequence comparison, 112did not match to any rice BAC clones (>1×10−15; Table 2,Fig. 4).

Three types of synteny breakdown between wheat andrice chromosomes

The results of the sequence comparison between wESTsand rice BAC clones showed that 31.6% of wESTsmatched to the sequences in rice BAC clones (Table 2)that are clearly from chromosome regions 6S and 8L.However, there were three different types of breakdown ofsynteny.

The first was apparent chromosome rearrangementwithin wheat chromosome 7AS compared to rice chromo-somes 6S or 8L. Eighteen wESTs from four bins did notmatch to the rice BAC clones in the corresponding ricechromosome regions (Fig. 2). For example, BE638019and BE403826 from bin 2 have a high homology with riceBAC clones at the distal region of rice chromosome 6Swhere most bin 1 wESTs are located (Fig. 2). In othercases, one rice BAC had homology with two wESTs fromtwo different chromosome bins: rice BAC AP000559 fromthe distal region of rice chromosome 6S had homologywith one wESTs from bin 1 (BE490191) and another frombin 2 (BE403826: Fig. 2).

The second type of apparent chromosome rearrange-ment was the detection of wESTs hybridizing to chromo-some 7AS deletion stocks that show identity to sequenceslocated on rice chromosomes other than 6S and 8L(Fig. 3). As indicated in Fig. 3 and Table 2, 86 of 316wESTs in the wheat chromosome 7AS bins had homologywith the rice BAC clones from other rice chromosomes(the chromosomes implicated include chromosomes 1–12). The results suggest that 32.9% of the wESTs analysedare associated with the first and second types of apparentchromosome rearrangement in chromosome 7AS(Table 2). These apparent chromosome rearrangementswere unevenly distributed among four bins of wESTs onchromosome 7AS as bin 1 had the highest percentage andbin 4 the lowest percentage of the chromosome rearrange-ments (Table 2).

The third type of apparent chromosome rearrangementswere those wESTs in bins of chromosome 7S that do notmatch to any BAC clones in the rice genome (Table 2).The data showed that 35.5% of wESTs that match to thebins of wheat chromosome 7AS do not have homology(>1×10−15) with any genomic DNA sequences in the ricegenome database (http://www.tigr.org/tdb/e2k1/osa1/).These ESTs could represent genes that existed only inwheat as there are no corresponding sequences in the ricegenome (other possibilities are noted in the discussion).The data in Table 2 indicated a higher rate of this type ofapparent chromosome variation in the distal region (bin 1)compared to that in the proximal region (bin 4) ofchromosome 7AS.

Breakdown of synteny between wheat chromosome7AS and its corresponding rice chromosomes 6S and8L

Most early work in comparative genomics in cerealsindicates that the gene order in genomes among cereals areconserved at the macro level (Hulbert et al. 1990; Ahn etal. 1993; Kurata et al. 1994; VanDeynze et al. 1995;Moore et al. 1997; Gale and Devos 1998). Through thecomparison of 4,485 ESTs from wheat and BAC clonesfrom the rice genome, numerous discontinuities in geneorder between wheat and rice were revealed (Sorrells et al.2003). More detailed comparisons in this work betweenwESTs on wheat chromosome 7AS and rice BAC cloneson rice chromosomes 6S and 8L showed that 58.4%wESTs from wheat either have non-syntenous positions orhave no orthologous sequences in the rice genome(Table 2); this is unlikely to be due to the incompletenessof ESTs (see later). The discrepancy between the geneorder of the rice and wheat genome could represent, forexample, incorporation of the genes related to diseaseresistance and grain composition in wheat, the rearrange-ment of DNA fragments through transcribed repeats,inverted repeats or pseudo-gene sequences (Wicker et al.2003), duplication and deletion (Akhunov et al. 2003), themore variable portion of mRNAs (such as the untranslatedregion) or library construction artefacts. The wESTsequences that could not be matched to the rice genomecould also be present in gaps in the rice genome sequence.In all of the discussion above, the assumption has beenmade that Southern blot hybridization to wheat chromo-some 7AS (and 7DS) has been accurate; clearly errors insuch mapping will also affect the results of the compar-ison. As this work is based on the sequence homologybetween wESTs and rice genomic DNA it could not coverthose genes that have a low homology at non-codingregion. Full length cDNA clones from wheat will furtherclarify the colinearity and breakdown of synteny betweenwheat and rice.

238

Comparison of the BAC clones encoding GBSSI frombarley and rice

GBSS I is an important determinant of grain quality andhas been extensively studied from both a biochemical andevolutionary perspective (e.g. Yamamori and Quynh 2002;Matthews et al. 2002). The genes for GBSSI are located onwheat chromosomes 7A, 4A and 7D, the 4A locationbeing caused by a translocation of the distal end of thechromosome 7B. It was of interest therefore in this contextto make a detailed comparison of the synteny between riceand wheat genomes around the gene for GBSSI. However,such information is not currently available for wheat, so anavailable BAC clone from barley was used that includesthe gene for GBSSI (BAC AF474373). Although somevariations may exist between wheat and barley genomicDNA around the gene for GBSSI, the barley BAC clonewas still used because of the synteny among cerealgenomes seen from early work (Moore et al. 1995; Galeand Devos 1998).

Barley BAC AF474373 contains 9 putative genes andrice BAC AF488413 (which includes GBSSI gene)contains 13 (Table 2). Sequence analysis showed thatbeside from GBSSI there was no sequence homologybetween the two BAC clones. As there are no availablegenomic sequences from wheat BAC clones, the analysiswas extended to search the TIGR databases with theputative genes encoded by the barley BAC clone. Theresults show that it is possible that there was a complexchromosome integration and deletion event in this regionduring evolution (Table 3). For example, sequencescurrently detected in chromosomes 1, 2, 4 and 6 in riceare found in BAC AF474373 in barley. Barley BACAF474373 has a highly homologous (97% identity)inverted repeat region located from 57.5 to 62.9 kb and

from 101.1 to 108.2 kb. The first region (57.5–62.9 kb)locates before gene 4 in barley and the latter one (101–108.2 kb) after gene 8 (Rostoks et al. 2002). Thisrepetitive element is not found on the rice BACAF488413.

Comparison of the genes encoding the enzymesinvolved in starch biosynthesis located on wheatchromosome 7AS

The sequences from both wheat and rice encoding GBSSI,SSI, SSIIa and DBEI were analysed in detail. The resultsshowed that the genes analysed from both wheat and ricehad conserved gene structures. In both cereals, GBSSI has11 exons, SSI has 15, SSIIa has 8 and DBEI has 18 exons.Two of these four genes from rice (GBSSI and DBEI)contained longer introns, while in the other two genes (SSIand SSIIa) the introns were longer in wheat. Some of theseresults have been discussed before (Li et al. 2003; Suzukiet al. 2003 Fig. 1).

Comparison between wheat chromosomes 7A and 7D

Wheat chromosomes 7A, 7B and 7D have differentlineages. The A genome is surmised to have been obtainedfrom Triticum uratu, the B genome is possibly from T.speltioides and the D genome is descended from Aegilopstauschii. These species are believed to have fused to formthe modern bread wheat about 10,000 years ago. Withinthe A, B and D chromosomes differences are well known,perhaps the most notable being the translocation of thedistal end of chromosome 7B to chromosome 4A. Thedifferences between chromosome 7A and 7D are less well

Table 3 Comparison betweenthe rice and barley BACs con-taining the genes encodinggranule bound starch synthase I(GBSSI)

Rice BACAF488413 (chr 6)

Barley BACAF474373 similar torice BAC

Identity Location of genes onrice chromosome

Gene 1 Beta-1,3-glucanaseGene 2 Putative no apical

meristem proteinGene 3 Unknown AP004094 1×10−51 2Gene 4 Putative magnesium

protoporphyrin IXAP000616 7.0×10−41 6 at 1.4 cM

Gene 5 Unknown AP004365 1×10−13 1Gene 6 Unknown AL607006 1.5×10−08 4Gene 7 GBSSI AP002542,

AF488413(GBSSI)

3.1×10−162 6 at 7.4 cM

Gene 8 Unknown AL6006618 5×10−98 4Gene 9 Unknown AP004734 2.9×10−17 4Gene 10 Unknown AL732339 0.016 4Gene 11 Unknown AP005777 0.003 2Gene 12 Putative transketolaseGene 13 Putative 3

phosphoshikimate

239

known and it was therefore of interest to compare thesynteny of 7DS with rice chromosomes 6S and 8L, giventhe information above regarding the synteny with chro-mosome 7AS.

The genes encoding GBSSI, SSI, SSIIa and DBEI weremapped on wheat chromosome 7AS, 7BS and 7DS byPCR markers (Li et al. 2003; Appels et al. 2003). All fourgenes were located at chromosome positions similar to thaton wheat chromosome 7AS except for the GBSSI gene onwheat chromosome 4AL that is translocated from chro-mosome 7BS (Appels et al. 2003). The distribution ofSouthern blot hybridisation mapped wESTs to the threebins for 7DS was compared with the distribution of thebest hits on the rice genome. In bin 1 (furthest from thecentromere), 6% of the ESTs could be mapped tochromosome 8, 30% to chromosome 6, 48% to otherchromosomes and 22% could not be mapped. In bin 2,15% of the ESTs could be mapped to chromosome 8, 53%to chromosome 6, 28% to other chromosomes and 30%could not be mapped. In bin 3 for 7D (nearest to thecentromere), 44% of the ESTs could be mapped tochromosome 8, 6% to chromosome 6, 37% to otherchromosomes and 12% could not be mapped. Theseresults are in broad agreement with the results forchromosome 7AS.

Conclusion

Comparison of 316 wESTs from chromosome 7AS binswith a virtual rice contig showed that 31.6% wESTsmatched orthologous sequences in the rice virtual contigwhich were consistent with their bin location. Approxi-mately 32.9% of wESTs matched to rice sequences that areindicative of chromosome rearrangements between wheat7AS and the rice virtual contig. In addition 35.5% ofwESTs do not have orthologous sequences in the ricegenome. Therefore, 58.4% of wESTs used in this work didnot have the same gene order in wheat and rice. Theoverall genome comparison between rice and wheat wasanchored by selected genes as demonstrated by the fourgenes involved in starch biosynthesis. The available ricegenome sequences can be used for a guiding reference forthe location and cloning of the genes in wheat (Sutton etal. 2003; Yan et al. 2003). However, caution needs to beexercised due to the high frequency of chromosomerearrangements indicated by this work. The informationfrom this work provides the ability to predict syntenylevels between the wheat and rice genomes depending onchromosome position. Although there is a high level ofsynteny breakdown around the GBSSI gene, the genestructure for four of the genes involved in starchbiosynthesis is conserved between the two cereal species.

References

Ahn S, Anderson JA, Sorrells ME, Tanksley SD (1993) Homo-eologous relationships of rice, wheat and maize chromosomes.Mol Gen Genet 241:483–490

Akhunov ED, Akhunova AR, Linkiewicz AM, Dubcovsky J,Hummel D, Lazo GR, Chao S, Anderson OD, David J, Qi L,Echalier B, Gill BS, Miftahudin, Gustafson JP, La Rota M,Sorrells ME, Zhang D, Nguyen HT, Kalavacharla V, Hossain K,Kianian SF, Peng J, Lapitan NLV, Wennerlind EJ, Nduati V,Anderson JA, Sidhu D, Gill KS, McGuire PE, Qualset CO,Dvorak J (2003) Synteny perturbations between wheat homo-eologous chromosomes caused by locus duplications anddeletions correlate with recombination rates. Proc Natl AcadSci USA 100:10836–10841

Appels R, Francki M, Chibbar R (2003) Advances in cerealfunctional genomics. Funct Integr Genomics 3:1–24

Gale MD, Devos KM (1998) Comparative genetics in the grasses.Proc Natl Acad Sci USA 95:1971–1974

Hulbert SH, Richter TE, Axtell JD, Bennetzen JL (1990) Geneticmapping and characterization of sorghum and related crops bymeans of maize DNA probes. Proc Natl Acad Sci USA87:4251–4255

Kikuchi S, Satoh K, Nagata T, Kawagashira N, Doi K, Kishimoto N,Yazaki J, Ishikawa M, Yamada H, Ooka H, Hotta I, Kojima K,Namiki T et al (2003) Collection, mapping, and annotation ofover 28,000 cDNA clones from japonica rice. Science 301(5631):376–379

Kurata N, Moore G, Nagamura Y, Foote T, Yano M, Minobe Y, GaleM (1994) Conservation of genomic structure between rice andwheat. BioTechnology 12:276–278

Li Z, Sun F, Xu S, Chu X, Mukai Y, Yamamoto M, Ali S, RamplingL, Kosar-Hashemi B, Rahman S, Morell MK (2003) Thestructural organisation of the gene encoding class II starchsynthase of wheat and barley and the evolution of the genesencoding starch synthases in plants. Funct Integr Genomics3:76–85

Matthews S, Spangler RE, Mason-Gamer RJ, Kellogg EA (2002)Phylogeny of Andropogneae inferred from phytochrome B,GBSS I and ndhF. Int J Plant Sci 163:441–550

Moore G, Devos KM, Wang Z, Gale MD (1995) Grasses, line upand form a circle. Curr Biol 5:737

Moore G, Roberts M, Aragon-Alcaide L, Foote T (1997) Centro-meric sites and cereal chromosome evolution. Chromosoma35:17–23

Qi L, Echalier B, Friebe B, Gill BS (2003) Molecular characteriza-tion of a set of wheat deletion stocks for use in chromosome binmapping of ESTs. Funct Integr Genomics 3:39–55

Rostoks N, Park RW, Ma J, Druka A, Shiloff BA, SanMiguel PJ,Jiang Z, Brueggeman R, Sandhu D, Gill K, Bennetzen JL,Kleinhofs A (2002) Genomic sequencing reveals gene content,genomic organization, and recombination relationships inbarley. Funct Integr Genomics 2:51–59

Sorrells ME, La Rota M, Bermudez-Kandianis CE, Greene RA,Kantety R, Munkvold JD, Mahmound MA, Ma X, GustafsonPJ et al (2003) Comparative DNA sequence analysis of wheatand rice genome. Genome Res 13:1818–1827

Sutton T, Whitford R, Baumann U, Dong C, Able JA, Langridge P(2003) The Ph2 pairing homoelogous locus of wheat (Triticumaestivum): identification of candidate meiotic genes using acomparative genetics approach. Plant J 36:443–456

Suzuki G, Moriyama M, Fujioka K, Yamamoto M, SubrahmanyamNC, Li Z, Apples R, Morell M, Mukai Y, Rahman S (2003) Thestarch branching enzyme I locus from Aegilops tauschii, thedonor of the D genome to wheat. Funct Integr Genomics 3:69–75

VanDeynze AE, Nelson JC, Yglesias ES, Harrington SE, Braga DP,McCouch SR, Sorrells ME (1995) Comparative mapping ingrasses. Wheat relationships. Mol Gen Genet 248:744–754

Wicker T, Yahiaoui N, Guyot R, Schlagenhauf E, Liu Z-D,Dubcovsky J, Keller B (2003) Rapid genome divergence atorthologous low molecular weight glutenin loci of the A andAm genomes of wheat. Plant Cell 15:1186–1197

Yamamori M, Quynh NT (2002) Differential effects of Wx-A1, -B1and -D1 protein deficiencies on apparent amylose content andstarch pasting properties in common wheat. Theor Appl Genet100:32–38

Yan L, Loukoianov A, Tranquilli G, Helguera M, Fahima T,Dubcovsky J (2003) Positional cloning of the wheat vernaliza-tion gene VRN1. Proc Natl Acad Sci USA 100:6263–6268

240