Trends in genetic and genome analyses in wheat: a review

P u b l i s h i n g

Australian Journal of Agricultural ResearchCSIRO PublishingPO Box 1139 (150 Oxford St)Collingwood, Vic. 3066, Australia

Telephone: +61 3 9662 7628Fax: +61 3 9662 7611Email: [email protected]

Published by CSIRO Publishing for CSIRO and the Australian Academy of Science

w w w . p u b l i s h . c s i r o . a u / j o u r n a l s / a j a r

All enquiries and manuscripts should be directed to:

AustralianJournal ofAgriculturalResearch

Volume 52, 2001© CSIRO 2001

A journal for the publication of original contributionstowards the understanding of an agricultural system

© CSIRO 2001 10.1071/AR01082 0004-9409/01/111043

Aust. J. Agric. Res., 2001, 52, 1043–1077

Trends in genetic and genome analyses in wheat: a review

P. LangridgeA, E. S. LagudahB, T. A. HoltonC, R. AppelsB, P. J. SharpD, and K. J. ChalmersA

ACRC for Molecular Plant Breeding, Department of Plant Science, Adelaide University, Glen Osmond, SA 5064, Australia.

BCSIRO, Division of Plant Industry, PO Box 1600, Canberra, ACT 2601, Australia.CCRC for Molecular Plant Breeding, Centre for Plant Conservation Genetics, Southern Cross University, PO Box

157, Lismore, NSW 2480, Australia.DPlant Breeding Institute, University of Sydney, Cobbitty, PMB11, Camden, NSW 2570, Australia.

Abstract. The size and structure of the wheat genome makes it one of the most complex crop species for geneticanalysis. The development of molecular techniques for genetic analysis, in particular the use of molecular markersto monitor DNA sequence variation between varieties, landraces, and wild relatives of wheat and related grassspecies, has led to a dramatic expansion in our understanding of wheat genetics and the structure and behaviour ofthe wheat genome. This review provides an overview of these developments, examines some of the special issuesthat have arisen in applying molecular techniques to genetic studies in wheat, and looks at the applications of thesetechnologies to wheat breeding and to improving our understanding of the genetic basis of traits such as diseaseresistance and processing quality. The review also attempts to foreshadow some of the key molecular issues anddevelopments that may occur in wheat genetics and breeding over the next few years.

Additional keywords: Triticum aestivum, genetic maps, breeding, marker, RFLP, AFLP, SSR, disease, quality,selection.AR01082P. Langri dgeA, E. S. LagudahB, T. A. Holt onC, R. Appel sB, P. SharpD, and K. ChalmersP. Langri dgeet al .Trends in geneti c and genome anal yses i n wheat

Introduction

The success of wheat breeding has largely come from theapplication of new technologies to breeding and selection.Particularly important were the introgression ofchromosome regions from wild relatives and thedevelopment of new selection strategies. Biotechnologyoffers two new means for improving wheat, firstly throughgenetic engineering and secondly through the developmentand application of molecular markers. Field trials ofgenetically engineered wheat are now well advanced in manycountries including Australia. The prime targets so farconsidered are the engineering of resistance to herbicides,resistance to viral and some fungal pathogens, andmodification of the quality characteristics of wheat grain,particularly starch and protein composition. Thedevelopment of large public genomics programs for wheat inthe USA, Canada, UK, France, and Australia is expected todeliver many new targets for engineering. Further, the privatesector has made a large investment in wheat genetic andgenomic projects.

Molecular marker technology offers a wide range of novelapproaches to improving the efficiency of selectionstrategies. The techniques are based around the detection of

sequence variation between varieties or accessions of wheat.Where the sequence variant sits in a region of the genomeclosely linked to a trait of interest, such as a diseaseresistance locus, the variant can be used to predict thepresence or absence of the resistance allele. The strength ofthe prediction will depend upon the closeness of the geneticlinkage between the sequence variant and the target locus.Sequence variation is generally abundant so that an almostunlimited number of such markers is available. However, inaddition to providing a means for predicting and trackingimportant alleles, molecular markers are also important inthe analysis of the genetic control for specific characteristicsand in the selection and analysis of the whole genome ofplants.

In this review, the current status of molecular markerdevelopment will be outlined. The problems and issuesrelated to marker development in wheat will also bediscussed and an attempt will be made to look at futuredirections in marker development and, particularly,application.

The genomic problems of wheat

Three features of wheat add greatly to the complexity ofbreeding and selection.

1044 P. Langridge et al.

Wide range of end uses, each with differing but specific quality requirements

This has imposed many constraints on wheat breedingprograms and has, in many cases, restricted the diversity ofgermplasm that can be used in a specific program.Consequently, the level of diversity detected betweencommercial wheat varieties is generally lower than for manyother species. This complicates both the development andapplication of molecular markers.

The complexity of the polyploid wheat genome, i.e. hexaploid, as opposed to diploid in most other cereal crops

Two problems arise here. Firstly the size of the wheat(Triticum aestivum) genome at 16 × 109 bp compared withbarley (Hordeum vulgare) at 5 × 109, maize (Zea mays) at5 × 109, and rice (Oryza sativa), the smallest of the majorcereals, at 4 × 108 bp, makes the application of severalimportant techniques difficult. Southern hybridisation isdifficult for wheat and the rate of failure is significantlyhigher than for other cereals.

Secondly, the presence of 3 related genomes of wheat (A,B, and D) adds to the complexity of many marker assays,particularly RFLP analysis, since 3 sets of bands will usuallyappear, one for each genome. Although in some cases thismeans that 3 loci can be mapped simultaneously, it is morecommon that only one genome will reveal polymorphismswith a given probe/enzyme combination. The 3 genomesalso translate into 21 linkage groups in wheat, again addingto the complexity of mapping work.

Low level of polymorphism in bread wheat

There is generally a low level of polymorphism in wheatrelative to other cereal species and this means that a largernumber of markers usually needs to be screened than is thecase for rice, maize, or barley (Chao et al. 1989; Lui et al.1990). Further, the level of polymorphism is not consistentacross genomes or crosses. Commonly, the D genome ismore highly conserved between varieties and is,consequently, substantially more difficult to map. If randommarkers are being used in the mapping exercise, then themaps of the D genome tend to have the poorest coverage(Chalmers et al. 2001, this issue). The development ofmicrosatellite markers specifically targetted to individualgenomes (D genome; Pestova et al. 2000a), should helpaddress this problem.

Although presenting difficulties in mapping, thehexaploid nature of wheat does offer some advantages.

Use of nulli-tetrasomic, translocation, and deletion lines to assign linkage groups

The allohexaploid genome of bread wheat means that it ishighly tolerant of loss of individual pairs of chromosomes.Through the pioneering work of Sears (1954), a series of lines

was generated, based on the variety Chinese Spring, whereindividual pairs of chromosomes were deleted and replacedby an extra chromosome pair from the homoeologous group.For example, the line nullisomic 1A/tetrasomic 1B has lostboth copies of chromosome 1A but has 4 copies ofchromosome 1B and the normal 2 copies of 1D. The presenceof the tetrasomic chromosome compensates for thenullisomy, so that these lines are more vigorous and stablethan the corresponding nullisomics. These lines are availablefor most wheat chromosomes and have proved extremelyuseful in assigning markers to linkage groups. Chromosomes2A and 4B carry genes important for fertility so these lines areusually maintained as monosomics, and upon selfing, theyyield seed that is nullisomic for 2A or 4B, respectively. Anexample of the use of these lines to assign marker bands to aspecific linkage group is shown in Fig. 1. Scanning theSouthern blot shown in Fig. 1, the disappearance of bandsindicates that the missing band was located on the missingchromosome (3A, 3B, or 3D in the figure).

Further resolution of band location can be achieved usingditelosomic lines. In these lines one arm has been lost from achromosome; a ditelosomic 3AS line will have 2 copies of theshort arm of chromosome 3A but will have lost the long armand retained 2 copies each of 3B and 3D. Through the use ofthese lines, a marker can be assigned to a chromosome arm(Fig. 1). A still higher level of resolution can be obtained byusing a series of deletion lines generated by Endo and Gill(1996) and others. These lines have deletions of chromosomesegments, usually from the telomere, and can be used toassign markers to particular regions of a chromosome. Oncethe initial arm location is known the deletion lines can bescreened to identify the deletions where the marker ismissing. This locates the marker to a physical region of thechromosome (Fig. 1). An important aspect of the deletion lineseries is that it has permitted a link to be made betweengenetic and physical maps. This aspect was demonstrated byGill et al. (1996) in comparing physical and genetic maps forGroup 5 chromosomes. The deletion lines have also formedthe basis for a large mapping project launched in the USA inearly 2000 entitled ‘The structure and function of theexpressed portion of the wheat genomes’. This project aims tomap 10000 Expressed Sequence Tags (ESTs) to a set ofdeletion lines and will form the framework for a highlydetailed map of the wheat genome (http://wheat.pw.usda.gov/NSF/).

Information on the various aneuploid and chromosomalmutant lines of wheat that are available for assigning markersto linkage groups and for physical mapping can be found onthe GrainGenes website (http://wheat.pw.usda.gov/ggpages/germplasm.html).

Current status of wheat maps

Several linkage maps of wheat have been published and themajor ones are summarised in Table 1.

Trends in genetic and genome analyses in wheat 1045

A comprehensive single population wheat map is basedon the W7984 × Opata 85 cross (http://wheat.pw.usda.gov/).The 114 recombinant inbred lines analysed were derived bysingle-seed descent (F8) from the cross of the synthetichexaploid W7984 × Opata M 85 and have been widely usedas an international reference mapping population for wheat.It is also known as the ITMI Mapping population or theSynthetic × Opata cross. The map that was assessed for thisreview contained 1074 loci, mainly RFLP and SSR, and thecurrent status of this map is summarised in Appendix 1(Fig. A1.1–1.7).

Some of the key features of the map are also summarisedin Table 2. The map generally has good genome coverage butas with most maps of hexaploid wheat, the D genome is lesswell covered than the A or B genomes. This probably reflectsthe lower level of polymorphism seen in the D genome eventhough a synthetic hexaploid was used as one of the parents.

The maps have been reproduced using the information inthe GrainGenes website and transporting this data into MapManager (http://mapmgr.roswellpark.org/mmQTX.html)for detailed analysis. The ‘rearrange’ command was carriedout on each chromosome in order to derive a map that fittedbest in terms of consistency of marker order. Mostchromosomes provided marker orders consistent withpublished maps, except for chromosomes 1B and 4B and inthis case additional information from known physicallocations of markers was used to derive the maps shown.The alignment with the physical maps was based on

published locations of RFLP and microsatellite markers(Röder et al. 1998b; Boyko et al. 1999; Faris et al. 2000;Sarma et al. 2000; Weng et al. 2000) and was generallyconsistent with the genetic order. In some instances theRFLP probes clearly assay repetitive sequences (e.g.Xfbb237 on 1D also located on chromosomes 1, 3, 5) and inthis case the alignment was not possible. In particular, it hasbeen found that the polymorphism that is mapped cansometimes be identified in the nulli-tetra stocks of ChineseSpring and in this instance a specific relation between themapped RFLP and its physical location is possible(MacGuire and Qualset 1997; Chalmers et al. 2001, thisissue). For chromosome 5D there is an apparentinconsistency between the physical and genetic maps(regarding XksuD30 and Xcdo57; see Appendix, Fig. A1.5).

The alignment of the physical and genetic maps is a usefulmechanism for defining ‘bins’ of markers that willcontribute to a composite/synthesis map of wheat currentlybeing produced through an international collaboration. Thealignments also indicate that some chromosomes such as 2Band 2D have relatively few reference points between thephysical and genetic maps. The assignment of wheatmicrosatellites or simple sequence repeats (SSR) from the‘gwm’ series (Röder et al. 1998b) to the deletionchromosomes characterised by Endo and Gill (1996) (seealso Boyko et al. 1999), will provide a much improved cross-referencing of the two maps (P. Sourdille, pers. comm.;B. Gill, pers. comm.).

3AS

3DS

3BS

Ditelosomic lines3DS 3DL 3BL 3AL CS 2bCS 1A 1B 1D 2A 2B 2D 3A 3B 3D 4A 4B 4D 5A 5D5B 6A 6B 6D 7A 7B 7D

Null-tetrasomic lines

Fig. 1. Use of wheat nulli/tetrasomic, ditelosomic, and deletion lines to assign RFLP bands to specific chromosomes andregions. Absence of a band in a particular nulli/tetrasomic line indicates the chromosome carrying the correspondingsequence, whereas presence of a band in a ditelosomic line indicates the chromosome arm carrying the locus. For the nulli/tetrasomic lines, the chromosome missing in each line is given above each lane and for the ditelosomic lines, the chromosomearm present in each line is shown. The lane marked ‘2b’ refers to a line carrying a deletion on the long arm of 3DS. CSindicates the Chinese Spring control. The chromosome designation of each major band is shown on the right. The probe usedwas MSH6 and the DNA was digested with DraI.

1046 P. Langridge et al.

The genetic lengths of the chromosomes in the Synthetic× Opata map range in size from 200 to 400 cM when the dataare treated as a selfed recombinant inbred population in MapManager (see Appendix 1). The length of some of thechromosomes may be a reflection of the way the dataset wasconstructed. Several labs from around the world havecontributed data to the Synthetic × Opata map and it is

probable that the standard of scoring and quality control hasnot been uniform. This is reflected by the clustering ofmultiple crossovers (e.g. 3–5 adjacent, apparentrecombination events), suggesting that the re-scoring of theloci in the respective lines may be warranted (Chalmers et al.2001, this issue). It seems unlikely that these regionsrepresent regions of negative interference as found in durum

Table 1. Published maps of wheat, wheat chromosomes, and diploid progenitors of wheat

Cross Genome Population structure

Marker type(s)

Number of

markers

Mapsize (cM)

Reference

T. monococcum × A 66 F2s ISSR 49 862 Kojima et al. 1998T. boeoticum RAPD 36

T. monococcum A RFLP 335 Dubcovsky et al. 1996Langdon × Langdon–

T. dicoccoides 6A, 6BAB 66 RSL 6A RFLP 77 57 Du and Hart. 1998

85 RSL 6B RFLP 61 67Messapia (durum) × MG4343

(T. turgidum)AB 65 RIL AFLP 80 2063 Lotti et al. 2000

RFLP 8RFLP 213 1352 Blanco et al. 1998SSR 79 Korzun et al. 1999

Ae. tauschii D 56 F2s AFLPRFLP

546 Boyko et al. 1999

Opata 85 × W7984 D 70 RIL SSR 55 Pestsova et al. 2000aSchomburgk × Yarralinka ABD 150 RIL RFLP

SSRAFLP

701152

Parker et al. 1999

T. dicoccoides × Langdon AB 150 F2s AFLPRAPDSSR

3153914

3180 Peng et al. 2000

Forno (winter wheat) ×Oberkulmer (T. spelta)

ABD 204 RIL RFLP + SSR

230 2469 Messmer et al. 1999

Opata 85 × W7984 ABD SSR 279 Röder et al. 1998aRFLP 9095 http://wheat.pw.usda.gov/

1069 6254 Appendix 1, this reviewGroup 2 ABD 114 F7 RFLP 57 A

60 B56 D

∼200 each

Nelson et al. 1995b

Groups 4, 5, and 7

ABD 114 F7 RFLP 136 A 129 B77 D

57–235 Nelson et al. 1995a

Group 3 ABD 114 F7 RFLP 58 A62 B10 D

230260171

Nelson et al. 1995c

Group 1 ABD Consensus RFLP + trait

109 211 Vandeynze et al. 1995

Group 5 ABD RFLP Xie et al. 1993Group 5 ABD 65 deletion lines RFLP 80 Gill et al. 1996Group 7 ABD 54 deletion lines RFLP

RAPD97 Hohmann et al. 1994

Group 6 ABD 51 88 Marino et al. 1996Consensus ABD RFLP +

trait1057 2830 Gale et al. 1995

Chinese Spring × T. spelta ABD RFLP 194 1803 Liu and Tsunewaki 1991var. duhamelianum

Group 1 ABD Consensus RFLP + trait

109 211 Vandeynze et al. 1995

Trends in genetic and genome analyses in wheat 1047

(T. turgidum L.) by Peng et al. (2000). Inspection of thedatabase indicated relatively few non-recombinant (orparental) chromosomes in the lines used for mapping(discussed in Chalmers et al. 2001, this issue).

Marker systems: detecting sequence variation in wheat

A wide range of techniques is available to detect DNAsequence variation between individuals. Broadly speaking

these can be divided into Southern hybridisation-basedmethods (restriction fragment length polymorphisms,RFLP), PCR-based techniques (the most widely used arerandomly amplified polymorphic DNA, RAPD; amplifiedfragment length polymorphisms, AFLP; and simplesequence repeat, SSR), and techniques based around thedetection of single nucleotide polymorphisms (SNP). Forwheat, most work to date has used RFLP, RAPD, SSR, and

(a) (b) (c)SSR AFLP RAPD

Table 2. Number of markers and length of chromosomes in the W7984 × Opata 85 cross

Group A B DNumber Length

(cM)Number Length

(cM)Number Length

(cM)

1 46 223 73 343 35 1982 59 238 77 279 65 3433 61 305 67 354 39 2034 50 242 37 147 19 1375 47 266 57 227 36 2676 66 239 59 220 47 2477 53 296 47 276 25 206

Total 382 1809 415 1846 268 1601

Fig. 2. Examples of various marker systems screened across wheat lines. (a) SSR or microsatellite markers. The PCR products resulting fromthe SSR reaction were fractionated on a 6% polyacrylamide gel and stained with ethidium bromide. The multiple bands, or stuttering, arefrequently seen in SSR assays. (b) AFLP banding pattern. The AFLP products were generated from selective PCR reactions where one primerwas labelled with a fluorescent dye tag. The PCR products were fractionated on a polyacrylamide gel in an ABI373 DNA sequencer and thepattern displayed using GeneScan software. Although this figure is in black and white, the red dye was used to label a size marker. (c) RAPDpattern. This was generated using a single 10 base primer for the PCR. The products have been fractionated on a polyacrylamide gel and stainedwith ethidium bromide.

1048 P. Langridge et al.

AFLP. An example of an RFLP marker is shown in Fig. 1,whereas Fig. 2 shows examples of AFLP, SSR, and RAPDmarkers. The AFLP band pattern is typical of the patternsseen using fluorescently tagged primers. An example of asection of an AFLP pattern developed with radiolabelledprimers is shown (see Fig. 4). The relative advantages anddisadvantages of these techniques for wheat are summarisedin Table 3. However, although the emphasis will be on thesemarker systems, a range of other techniques, many basedaround these procedures, can be successfully used to detectDNA sequence variation in wheat (for example, inter-simplesequence repeats, ISSRs; Nagaoka and Ogihara 1997)

The data in Table 3 represent averages for these markersystems screened against commercial bread wheat varieties.Examples of each of the marker systems are shown in Fig. 2.The ability of the different marker systems to detectpolymorphisms in wheat is summarised in Table 4 where 2studies are compared. Shah et al. (2000) compared the wheatvarieties Cheyenne and Wichita, and Paull et al. (1998) andParker et al. (2001) looked at 124 and 101 different varieties,respectively (the 101 varieties were a subset of the 124varieties). These results suggest that RFLP and SSR markersare the most effective in detecting polymorphisms. However,given the difficulties of RFLP detection, the large amount ofDNA required, and the difficulties in automating RFLPdetection, SSRs are currently the most popular markersystem for wheat. Therefore the major emphasis in thissection of this review will be to provide an overview of thecurrent status of SSR markers in wheat. Other markersystems will be described only briefly.

Table 4 does not include AFLP information since theseare difficult to quantify in the same way as the other markersystems. However, if 2 varieties are compared, on average 5polymorphic bands are revealed per primer combination.This compares with close to 10 for barley (P. Langridge, K.Chalmers, A. Karakousis, J. Kretschmer unpubl. data).

RFLPs

RFLP markers have formed the basis of much of themapping work in wheat and other species. An example of an

RFLP is shown in Fig. 1. Their advantages anddisadvantages are shown in Table 3 and, as can be seen inTable 4, they are extremely effective in detectingpolymorphisms in wheat. The probes used to detect RFLPsare frequently derived from cDNA clones and thereforerepresent expressed genes and are able to hybridise togenomic DNA sequences that show as much as 80%similarity. This means that an RFLP probe can be used acrossa wide range of related species. In wheat mapping, probesfrom wheat, rye (Secale cereale), barley, and oats (Avenasativa) are widely used. Further, wheat sequencessuccessfully hybridise to genomic DNA from rice, maize,sorghum, and several other grass species. The cross-speciesuse of RFLP probes has demonstrated the relationshipsbetween chromosomes of different grass species andrevealed the conservation of gene order (synteny) across thegrasses (Moore et al. 1995).

However, the disadvantages of RFLPs have limited theirapplication to pragmatic breeding programs and their use isdecreasing. The problems associated with RFLPs can beaddressed most effectively through PCR-based methods.

SSRs or microsatellite markers

SSRs comprise short repeat units of 1–6 nucleotides. Theyappear to be ubiquitous in higher organisms, although thefrequency and type of SSRs vary between species. They areabundant, dispersed throughout the genome, and showhigher levels of polymorphism than other genetic markers.These features, coupled with their ease of detection andtransferability between researchers, have made them usefulmolecular markers.

Isolation of SSRs

Many approaches have been used to isolate SSRs andtheir flanking sequences (Powell et al. 1996a). Genomicclones containing SSRs can be isolated by screening withlabelled oligonucleotides containing the desired repeatsequences. To improve the efficiency of SSR isolation,enriched libraries were developed using a variety of methodswith selection either before or after library construction. The

Table 3. Comparison of the most widely used marker systems for wheat

Marker System

Loci detected per assay

DNA amount needed

Advantages Disadvantages

RFLP 3 5 µg Highly reliable, co-dominant, serve as reference to other cereal maps, large number available, target specific regions

Technically difficult, slow, requires large amount of DNA. Detect low level of polymorphism

SSR 1 0.2 µg Reliable, co-dominant, often genome-specific, target specific regions, amenable to automation

High development cost

AFLP 50 0.2 µg Reliable, detect large number of loci simultaneously, amenable to automation

Random, dominant

RAPD 10 0.2 µg Cheap, technically simple Unreliable, dominant

Trends in genetic and genome analyses in wheat 1049

most commonly used SSR markers are (AC)n and (AG)nrepeats. In species where many gene sequences are availableit is also possible to identify SSRs by examination of DNAsequence databases. Identification of (AT)n and (GC)nmicrosatellites from sequence databases is also possible. Theisolation of such SSRs by hybridisation screening withlabelled probes or enrichment is problematic due to the self-hybridising palindromic nature of these sequences.

Detection of polymorphisms

Variations in the length of tandem repeats can beidentified by amplification of the region containing therepeat via PCR using primers designed to the regionsflanking individual SSRs. The sequences flanking themicrosatellites are relatively conserved, allowing anchoringof the flanking primers. Polymorphisms are detected basedon size differences between alleles, which result fromdifferences in the number of repeats. SSR loci are believed toevolve in a step-wise manner by the addition or subtractionof a single repeat due to polymerase slippage at DNAreplication (Jarne and Lagoda 1996). Dinucleotide repeatssometimes produce complex patterns when amplified byPCR due to the in vitro production of ‘stutter’ bands inaddition to the target band size. For this reason, the rarerhigher-order repeats may be more desirable markers.

Advantages of SSRs over other markers

SSRs are easier to use than RFLPs due to the loweramount of DNA required, higher polymorphism (Powellet al. 1996a), and the ability to automate assays. SSRmarkers can easily be exchanged as information betweenresearchers because each locus is defined by the primersequences rather than the physical exchange required forclones used as RFLP probes. SSR assays are more robustthan RAPDs and more transferable between populations thanAFLPs. The co-dominant nature of most SSR markers is alsoan advantage for genetic mapping.

Use of SSRs in genetic mapping

Their ease of use and high information content hasensured that SSRs have largely replaced RFLP as a mappingtechnology in human research (Dib et al. 1996). The

development of SSRs in plant research is accelerating, andSSR loci are now being incorporated into established geneticmaps of all the major cereals (Liu et al. 1996; Korzun et al.1997; Smith et al. 1997; Stephenson et al. 1998).

The construction of high-resolution linkage maps basedexclusively on SSRs may be an unrealistic goal for most cropplants due to the great effort and cost involved in isolation ofmarkers from each species. However, the creation of‘skeletal’ genetic maps with SSRs is an achievable goal,where SSRs can provide anchor points for specific regions ofthe genome. Gaps between the SSR markers may then be‘filled in’ with other markers such as AFLPs.

Isolation of SSRs from wheat

Wheat has proven more difficult than many species foridentifying polymorphisms, but recently there has been arapid increase in identification of polymorphic SSR markers(Röder et al. 1998a; Stephenson et al. 1998; Pestsova et al.2000a) and the Agrogene wheat microsatellite consortium(http://www.microsatellites.agrogene.com/).

The frequency of SSRs in the wheat genome was estimat-ed by hybridisation of synthetic AC and AG oligomers to ge-nomic libraries by Röder et al. (1995) and Ma et al. (1996).The calculated frequencies were once every 292 kb or 704 kbfor AC, and once every 440 kb or 212 kb for AG. This meth-od is likely to underestimate the number of SSR loci becausesmall repeats are not efficiently identified. Trinucleotide re-peats were about 10 times less common than the two dincle-otide repeats tested and tetranucleotide repeats were rare.

Many studies have examined the level of polymorphismof SSR markers in wheat (Table 5).

Table 4. Comparison of the ability of marker systems to detect polymorphismsCheyenne v. Wichita, Shah et al. (2000); varietal set, 124 varieties (RFLP), Paull et al. (1998) and 101 varieties (a subset of the 124 varieties used by Paull et al. 1998) (SSR) Parker et al. (2001)

Marker type Number screened Number polymorphic % PolymorphicCheyennev. Wichita

Varietalset

Cheyennev. Wichita

Varietal set

Cheyennev. Wichita

Varietal set

STS 77 3 3.9RAPD 40 8 20RFLP 52 119 41 98 78.8 82SSR 10 31 6 19 60 61

Table 5. Screening of wheat varieties for SSR polymorphisms

No. of accessions

No. of SSRs screened

No. of alleles

Reference

18 15 69 Röder et al. 199540 23 142 Plaschke et al. 199510 49 169 Bryan et al. 199711 13 33 Bohn et al. 1999

101 31 160 Parker et al. 2001

1050 P. Langridge et al.

Stephenson et al. (1998) mapped 51 SSR loci in a ChineseSpring × SQ1 wheat population. Röder et al. (1998a)isolated a large number of useful wheat SSR markers byenriching for hypomethylated regions of the genome. A totalof 279 loci amplified by 230 primer sets were placed onto agenetic framework map composed of RFLPs previouslymapped in the Opata 85 × W7984 cross. Only 37 of 230primer sets produced more than one mappable locus. Themajority (80%) of 193 SSR markers developed weregenome-specific. Of the 279 SSRs mapped, 93 mapped tothe A genome, 115 to the B genome, and 71 to the D genome.

Another 65 SSR markers were developed by Pestsova etal. (2000a) from Aegilops tauschii, the D-genome donor ofhexaploid bread wheat. Forty-eight primer pairs detectedpolymorphisms between Opata 85 and synthetic wheatW7984. The majority of markers could be assigned to the Dgenome of wheat. In contrast to RFLP markers, most wheatSSR markers are genome-specific and amplify only onespecific locus containing an SSR in the A, B, or D genomeof bread wheat (Röder et al.1998a).

Transferability between species

The ability to use the same SSR primers in different plantspecies depends on the extent of sequence conservation inthe primer sites flanking the SSRs and the stability of theSSR during evolution. Transferability of SSR primers acrossspecies would increase the value of such markers andpotentially decrease the cost of development. However,wheat SSR primer sets do not generally amplify SSRs inbarley (Röder et al. 1995), suggesting the need to developseparate SSR primer sets for each species.

Gene sequences as a source of SSR markers

The increasing availability of Expressed Sequence Tags(ESTs), cDNA and genomic sequences from wheat andbarley is providing a potentially valuable source of new SSRmarkers. Using in silico discovery of SSRs it is also possibleto identify polymorphic loci directly when sequences of thesame gene are available from more than one variety. SSRmarkers derived from expressed gene sequences provide‘perfect’ markers for those genes. This may allow directidentification of alleles that cause important phenotypictraits. Repeat sequences such as AT and GC may also beisolated using this approach. Many trinucleotide and higherrepeats are also present within ESTs and these markers couldbe useful additions to the predominantly dinucleotide repeatsso far used in plant genotyping. These markers are likely tobe more highly conserved and therefore may be moretransferable between species than dinucleotide repeats.Assays for trinucleotide SSRs may also be more easilyautomated due to the absence of artefacts such as ‘stutter’,which often occur with dinucleotide repeats and the largersize differences between alleles.

The International Triticeae EST Cooperative (ITEC) wasset up in 1998 to establish a public database of ESTs fromspecies of the Triticeae. At the end of Phase One, there were23000 EST sequences available (http://wheat.pw.usda.gov/genome/). Initial screening of the ITEC EST sequences forSSRs has identified a large number of potential SSRmarkers, which are currently being screened forpolymorphisms. Preliminary analysis indicated that 1.9% ofthe EST sequences contain dinucleotide repeats and 1.5%contain trinucleotide repeats (n >6). There are also largewheat EST sequencing projects established in the USA andelsewhere. These have resulted in boosting the number ofESTs in the public databases to over 50000 at the end of2000. Existing plans for EST sequencing in wheat suggestthat over 200000 will be available by 2001 (Report of theDundee workshop in wheat and barley genomics, May2000).

The rice genome project is a consortium of researchersthat aims to completely sequence the rice genome by 2004.However, the recent release of a ‘working draft’ of the ricegenome sequence should greatly accelerate the completionof a fully annotated version of the rice genomesequence(http://rgp.dna.affrc.go.jp/). The availability of thecomplete sequence of a cereal genome should facilitateconstruction of genetic maps in cereals such as wheat bycomparative mapping and enable increased opportunities todevelop transferable markers.

Amplified fragment length polymorphisms (AFLP)

One of the principal applications of AFLP (Zabeau and Vos1993; Vos et al. 1995) is in genetic mapping, either in mapconstruction or identifying linkages using bulk segregantanalysis (see below). The high multiplex ratio of the AFLPtechnique is now being exploited in wheat and has thepotential to improve the efficiency of genetic mapconstruction and to generate high-density maps around locithat control commercially important traits. The reportedlevels of AFLP polymorphism observed in cereal mappingpopulations differ widely among cereal species from 26.6%in rice (Cho et al.1998), up to 48% in maize (Vuylsteke et al.1999), 11.3% in barley (Becker et al. 1995), 23.8% in durumwheat (Lotti et al. 2000), and 12.8% in bread wheat(Chalmers et al. 2001, this issue). AFLP loci appear to beevenly distributed across the A and B genomes but there is asignificant reduction in the level of polymorphism detectedin the D genome, mirroring the observed levels ofpolymorphism in microsatellite and RFLP mapping studies(Marino et al. 1996; Röder et al. 1998a).

AFLP markers differ in both frequency and distributionthroughout the wheat genome, and as with other cereals, thechoice of primer sets used in generating AFLP directlyinfluences the distribution of loci identified. Given that thebasis for the detection of AFLP polymorphism is pointmutation or insertion deletion, events that are generally

Trends in genetic and genome analyses in wheat 1051

randomly distributed throughout the genome, it would beexpected that AFLP loci would be likewise distributedthroughout the genome. Generally, AFLP markers, which areproduced by different combinations of restriction enzymes,are distributed throughout the genome, although clusteringof markers in centromeric regions has also been reported.The clustering of both AFLPs and RFLPs observed in thecentromeric regions indicates that there may be asuppression of recombination and is in agreement with otherpublished cereal maps (Hart 1994; Becker et al. 1995; Qiet al. 1998). However, there is also evidence that AFLPmarkers lie outside regions that are heavily populated withRFLPs and this frequently leads to increases in map lengththat have been attributed to gap filling and terminalextensions in maps with sparse coverage.

In cereals with relatively small genomes such as rice, 6-base-pair cutting enzymes such as EcoRI and PstI are usedwith MseI to generate the AFLP template. This thenundergoes selective-PCR-amplification with primerscontaining +2 selective bases to produce the requirednumber of bands, generally 100–150. For analysis of cerealswith larger genomes, including wheat, selective-PCR-amplification requires the presence of an additional selectivenucleotide to produce a primer with +3 selective bases. Theaddition of the extra selective nucleotide effectively reducesthe number of bands by a factor of 4. The use of AFLP inwheat have been reported using both PstI and EcoRI primersand the different restriction enzymes can also affect thenumber of polymorphisms detected. In wheat, for example,more polymorphisms are detected by the PstI/MseIcombination than with the EcoRI/MseI combination, andanalysis reveals that the distribution of markers generated bythe 2 types of enzyme combination is different. In addition,some primer combinations generate more polymorphismsthan others and this can be used to optimise the number ofbands visualised. Most plant genomes are AT-rich, and use ofAT-poor primers can be used to reduce the band complexityin plants with large genomes such as wheat.

Detection of AFLP products has been accomplished using[33P]-labelled primers, fluorescently labelled primers, orsilver staining. Research kits for radioactive and fluorescentlabelling of AFLP primers are now commercially availablefrom several companies. Whereas the use of [33P]-labelledprimers or fluorescently labelled primers gives generallycomparable results, the use of silver staining has severaldrawbacks. The first disadvantage associated with silverstaining is that there is a variation in band intensity accordingto the size of the product and this becomes particularlyevident in the size range below 80–100 bp. Anotherdisadvantage of silver staining is the presence of doublets dueto the staining of both strands of the same DNA fragment,which may migrate at different speeds when their basecomposition differs markedly. This is not a problem with thealternative approaches as only one of the primers is labelled.

AFLP is a robust and reliable technique that is findingincreased use in wheat genetics and is rapidly becoming thepreferred molecular technique for many types of studies. Thevalue of this technique has now been widely demonstratedacross most of the major cereals and the lessons learned canbe readily applied to wheat genetics.

Sequence tagged sites (STS)

RFLP clones isolated from related cereals often hybridise tohomologous sequences in other species. Conversion ofRFLP markers to a sequence tagged site (STS) marker offersseveral advantages over conventional RFLP analysis,including relative ease, greater throughput, sharing of primersequences, and the need for less genomic DNA (Erpeldinget al.1996). Sequence tagged site PCR is carried out usingprimers designed to flank a known DNA sequence. Thesemarkers are of particular value when they can be designed toidentify variation in genes of agronomic importance. Incontrast to SSRs, STSs derived from the conversion of RFLPmarkers are readily transferable between wheat and barley(Erpelding et al. 1996). There are many variations of STSmarkers, including SSRs, gene-specific markers, and singlenucleotide polymorphisms (SNPs).

Gene-specific markers

This class of markers is directed to amplify polymorphismswithin known gene sequences, allowing the gene to bemapped. Polymorphisms are generally detected by sizedifferences in the PCR amplification products or by post-PCR digestion with restriction endonucleases. In general, thegreater conservation of expressed gene sequences reducesthe likelihood of finding polymorphic markers within a genesequence; however, sequences in the 5′ and 3′ untranslatedregions are generally more variable than in the coding region,and intron sequences also provide higher levels of variabilitythan the coding regions. Currently, there are relatively fewexamples of this type of approach; however, this type ofmarker is becoming much more accessible with theavailability of extensive sequencing programs in manyspecies that target expressed sequences. Two gene-specificmarkers that have been adopted widely by the Australianwheat breeders are derived from GBSS (granule-boundstarch synthase) and Cre3 (controlling cereal cyst nematoderesistance).

Primers have been developed to flank a sizepolymorphism in the GBSS gene between the A, B, and Dgenomes of hexaploid wheat. The granule-bound starchsynthase gene is an important one controlling the content ofthe amylose fraction of starch and is an importantdeterminant of the end-use of wheat (Zhao et al. 1998). Thelocus situated on chromosome 4A is absent in certain wheatvarieties, which identifies the wheat cultivar as beingsuitable for the production of Udon noodles, a major end-use

1052 P. Langridge et al.

of wheat in Asia. Approximately 40% of Australian wheatsare null at this locus.

Single nucleotide polymorphisms

Single nucleotide polymorphism (SNP) markers arebecoming widely used in human genetics to investigatelinkages of genes to specific disease conditions (Weiss1998). The technique involves the identification of singlebase-pair changes at specific sites between target genomesand designing PCR-based amplification protocols to revealthe polymorphism. It is estimated that the human genomecontains one SNP every 1058 base pairs (Durrett and Limic2001). The frequency of SNPs in maize is even greater thanin humans (A. Rafalski, unpubl. results). This marker classhas not yet been widely exploited in plants, but represents arich source of polymorphisms to be utilised in the future.

The SNP approach has several advantages over SSRmarkers, for example, SNPs in specific genes can betargetted using existing EST databases; and many techniquesare available to type SNPs in an automated fashion and manyyield simple positive or negative outcomes that can beinterpreted easily.

Identification of SNPs

SNPs can be discovered in silico when gene sequences areavailable from more than one allele of a particular gene. It isalso possible to design primers from gene sequences, usePCR to amplify the region from a diverse range of cultivarsand sequence the amplification products directly. Wherepossible, 5′ and 3′ untranslated regions should be targetted toincrease the likelihood of finding SNPs in gene sequences.

The rice genome sequence may also provide a valuablesource of primer sequences for SNP discovery. Due to thehigh similarity of gene order and gene sequences in rice andwheat it is possible to use rice sequences to design PCRprimers and amplify the corresponding genes in wheat. Therice genome sequence can be used to help target gaps ingenetic maps or to isolate markers for multiple genes inparticular regions of the genome. In this case, primers shouldbe designed to coding regions flanking introns wherepossible. This will increase the chance that primers will workacross species, increase the chance of discriminatingbetween amplification products from the differenthomeologous loci in wheat, and increase the chance ofidentifying SNPs.

SNP assays

There are many technologies available to detect or screenfor known SNPs in populations, including: (1) Taqman(Morin et al. 1999), (2) molecular beacons (Marras et al.1999), (3) fluorescent ddNTPs, (4) restriction enzymepolymorphisms (Neff et al. 1998), (5) MALDI-TOF massspectroscopy (Griffin and Smith 2000), (6) microarray

hybridisations (Hirschhorn et al. 2000), (7) DHPLC (Grosset al. 1999), and (8) SNAP (Drenkard et al. 2000).

Which technique is used in a particular laboratorydepends upon the scale of SNP analysis being undertaken.The use of fluorescent dideoxynucleotides or MALDI-TOFis likely to be most useful for genetic map construction,microarrays for intensive screening of germplasm, andmolecular beacons and restriction enzyme polymorphismsfor targetting small numbers of specific markers inimplementation laboratories.

The ability of molecular beacons (oligonucleotides thatfluoresce only upon hybridisation) to analyse singlenucleotide differences has been exploited to develop rapidtests for SNP analysis. For instance, Piatek et al. (1998)employed this assay for detecting drug resistance inMycobacterium tuberculosis. Similarly, Giesendorf et al.(1998) used this assay to detect a point mutation in themethylenetetrahydrofolate reductase (MTHFR) gene inhumans. It is envisaged that the molecular beacon assays willhave the potential to rapidly analyse SNPs in plants, and thatthese probes will find their way into nucleic acid researchand diagnostics.

Two point linkages to economic traits

Many groups around the world have identified molecularmarkers linked to traits of economic importance. Acomprehensive list of traits and linked markers can be foundin the Gene Catalogue of McIntosh (1998). This catalogueand updated versions can be found in the 1999 AnnualWheat Newsletter, Wheat Information Service, and is listedon the GrainGenes Website (http://wheat.pw.usda.gov/). Alist of some of the most recent publications (1999 and 2000only) describing the identification of marker linkage toeconomically important traits is shown in Table 6.

An important feature to note in the examples listed inTable 6 is the recent shift to the development of markers thatwill be suitable for application to marker-assisted selection.Many of the early markers found associated with traits werebased on RFLP or RAPD markers. Neither marker system issuited to large-scale screening as discussed above. TheCAPS, SCAR, and STS markers listed in Table 6 weredeveloped from RFLP or RAPD markers with the specificaim of improving their applicability to selection programs.Table 6 also shows the effort being invested into identifyingmarkers linked to disease resistance loci. However, someimportant studies have identified markers linked tocomponents of quality.

QTL studies

Examples of marker–trait associations largely representtraits controlled by a single or a few genes (Table 6). Theseare also traits where the phenotypic assay generallyinvolves a yes or no answer; for example, the plant isresistant or susceptible to a particular disease. Traits that

Trends in genetic and genome analyses in wheat 1053

show a continuous distribution between two extremes areregarded as quantitative traits. This distribution pattern canarise for two major reasons: the trait is controlled by severalloci each of small effect and different combinations ofalleles at these loci can give different phenotypes; and/orthe trait is strongly influenced by the environment so that acombination of genetic and environmental factorsinfluences the phenotype.

The identification of markers linked to loci controllingquantitative traits (QTL) is substantially more complex thanfor simple traits. To date, most mapping of QTL hasdepended upon the construction of complete linkage maps ofpopulations segregating for the trait of interest. This is incontrast to simply inherited traits where bulked segregantanalysis (BSA) is a rapid and effective means for findinglinked markers (see below). In addition, where the

quantitative trait is strongly affected by the environment,testing over multiple environments is usually required sincethis will allow a measure of the heritability of the trait andmay permit correction of the data to allow for theenvironmental component. For the analysis of data fromfield trials, spatial analysis of the trial data can substantiallyincrease the heritability of the trait and the ease ofidentifying controlling QTL. A comprehensive descriptionof the statistical tools available for the analysis of QTL canbe found in Liu (1998) and Eckermann et al. (2001).

Although QTL analysis represents a substantially morecomplex target for marker identification than simplyinherited traits, the benefits to selection programs of havingmarkers associated with QTL are far greater. These areusually traits that are difficult to select for in a conventionalbreeding program but are frequently prime targets for wheat

Table 6. Recent publications describing the association of molecular markers with traits in bread wheatCAPS, cleavage amplified polymorphic sequence; SCAR, sequence characterised amplified region; STS, sequence tagged site

Locus Trait Source Marker Location Referencetype

Rusts

Lr47 T. speltoides CAPS 7A Helguera et al. 2000Lr35 Adult plant resist. STS 2B Seyfarth et al. 1999Lr21 Ae. tauschii RFLP 1DS Spielmeyer et al. 2000Yr10, YrVav RFLP 1BS Spielmeyer et al. 2000Yr28 Ae. tauschii RFLP 4DS Singh et al. 2000Yr18 Adult plant resist. RFLP 7DS Singh et al. 2000Yrns-B1 Lgst.79–74 RFLP 3BS Borner et al. 2000YrH52 T. dicocoides SSR 1B Peng et al. 2000Yr17 Ae. ventricosa RAPD/SCAR 2A Robert et al. 1999Yr17, Lr37, Sr38 Ae. ventricosa STS, RFLP 2AS Seah et al. 2000

Powdery mildew resistance

Pm1c and Pm4a AFLP Hart et al. 1999Pm21 H. villosa SCAR 6VS Liu et al. 1999Pm26 T. turgidum RFLP 2BS Rong et al. 2000Pm13 Ae. longissima STS 3S Cenci et al. 1999

Other disease resistances

Wheat streak mosaic by movirus

RFLP 2D Khan A.A. et al. 2000

Dn5 Diuraphis noxia resist. SCAR Venter and Botha 2000Rkn-mn1 Root-knot nematode Ae. variabilis RAPD Barloy et al. 2000

Stress

Boron tolerance RFLP 7B, 7D Jefferies et al. 2000Sprouting resistance 5AL, 6A, Zanetti et al. 2000

3B, 7BCold tolerance RFLP 5A Vagujfalvi et al. 2000Preharvest sprouting STS, SSR 6B, 7D Roy et al. 1999

Quality

Flour colour STS 7A Parker and Langridge 2000Milling yield RFLP 3A, 7D, 5A Parker et al. 1999Grain protein content SSR 2DL Prasad et al. 1999

Wx-A1 Waxy SSR 7D Shariflou and Sharp 1999High grain protein T. turgidum RFLP 6B Mesfin et al. 1999HMW glutenins STS Group 1 Ahmad 2000

1054 P. Langridge et al.

improvement. This applies for resistance to several majordiseases, tolerance to a variety of abiotic stresses, and manymajor components of processing quality. Consequently, it isanticipated that markers for QTL will have the biggestimpact on breeding and selection strategies.

In the past 2 years there have been many publishedreports of QTL analysis in wheat (Table 7). Many of thetraits listed in Table 7 are only poorly understood at thephysiological or biochemical level. One advantage of QTLanalysis is that it frequently allows the geneticcomponents of a trait to be separated. For example, in thestudy of QTL controlling tolerance to high soil boron(Jefferies et al. 2000), 4 different assays were used toassess boron tolerance. Through QTL analysis of these 4phenotypic measures, it was possible not only to identifythe major loci controlling tolerance, but also to assignpossible functions to these loci, such as reduction inboron uptake, increased cellular tolerance, and alteredboron distribution within the wheat plants.

Mapping and tagging of disease resistance

From the information presented in Tables 6 and 7 it is clearthat a key objective in many marker development programswas the identification of markers linked to disease resistanceloci. The main reason is, of course, that disease is a factorlimiting yield and quality in many environments. In additionto offering improved methods for selection for resistantcultivars, markers offer several new tools for the breeding ofresistant varieties; namely the ability to combine severalresistances to a single pathogen in one variety (resistancegene pyramiding) and the possibility of selecting forresistant genotypes in the absence of the pathogen. Further,mapping of resistance gene loci is central in the cloning ofdisease resistance genes. For these reasons, special emphasiswill be placed in this review on the current status of themapping and cloning of disease resistance loci.

Plant breeders over the years have successfullyincorporated resistance genes to protect plants againstdiseases. Genes expressing major effects as well as those

Table 7. Recent QTL studies in wheatThese studies used full linkage maps constructed with RFLP and SSR markers to identify marker–QTL

associations with the exception of the studies labelled ‘BSA’ where Bulked Segregant Analysis was used. One map, Bai et al. (1999), used AFLP makers. Population structures used were: RIL, recombinant inbred lines;

DH, doubled haploids

Target trait Population Marker loci Reference

Quality

Pre-harvest sprouting 204 RIL 183 Zanetti et al. 2000Grain weight 100 RIL BSA Varshney et al. 2000Grain protein 100 RIL BSA Prasad et al. 1999Kernel morphology and texture:

kernel length, width, area, perimeter,test weight, 1000-kernel weight

78 RIL Campbell et al. 1999

Milling yield 150 RIL 178 Parker et al. 1999Flour colour 150 RIL 147 Parker et al. 1998Bread-making quality—hardness, 187 DH 380 Perretant et al. 2000

protein, dough strength

Physiological traits

Ear compactness—spike length 187 DH 380 Sourdille et al. 2000bnumber of spikelets, compactness Courtout ×

Chinese SpringFree-threshing habit in tetraploid wheats 65 RIL 213 Simonetti et al. 1999Lodging resistance 226 RIL 182 Keller et al. 1999bCrossability 110 DH Tixier et al. 1998Heading time photoperiod response 187 DH 380 Sourdille et al. 2000a

Courtout ×Chinese Spring

Disease resistance and abiotic stress tolerance

Fusarium head blight resistance 112 RIL 360 Waldron et al. 1999Fusarium head blight resistance 133 RIL AFLP Bai et al. 1999Powdery mildew resistance 226 RIL 182 Keller et al. 1999aResistance to chlorosis induced by

Pyrenophora tritici-repentis112 RIL 542 Effertz et al. 1998

Boron tolerance 161 DH 545 Jefferies et al. 2000

Trends in genetic and genome analyses in wheat 1055

with a classical quantitative phenotype in the host responseto the pathogen are used in breeding programs. However, itwas not until the last decade that major genes for diseaseresistance were isolated through advances in strategies inplant gene isolation. These strategies include transposontagging and map-based cloning. Among the earlier genesisolated via insertional mutagenesis through transposonswere the resistance genes Hm1 against maize leaf spot, Nagainst tobacco mosaic virus, Cf9 against tomato wilt and L6against flax rust. Map-based or positional cloning strategieswere equally successful among some of the earlier genesisolated, which includes Pto resistance to bacterial spot intomato, and Rps2 and Rpm1 resistance against bacterial spotin Arabidopsis.

Structure of cloned resistance genes

Four distinct structural categories can be used to describe themajor plant disease resistance genes cloned so far. In a classof its own is the Hm1 gene in maize. This gene encodes anactive reductase enzyme that can detoxify the Hc toxinproduced by the leaf spot fungus (Cochliobolus carbonum)and thereby confers resistance to the fungal pathogen.Another unique resistance gene class is Pto in tomato, whichencodes a serine-threonine kinase.

The third and broad class of resistance genes have as acommon feature the presence of leucine-rich repeats (LRR)of variable length (Jones and Jones 1997). The LRR class ofresistance genes can be further divided into 3 subclasses.One of the subclasses belongs to a superfamily of genes witha predicted protein product harboring a nucleotide bindingsite and leucine rich repeats (NBS-LRR). Extensive reviewson plant NBS-LRR genes have been conducted (Baker et al.1997; Jones and Jones 1997; Ellis et al. 1999) as theyrepresent the largest type of cloned disease resistance genes.Resistance to viral, bacterial, and fungal pathogens as well asto nematode and sucking insect pests has been identifiedamong NBS-LRR genes, which makes this class ofsequences a curious and interesting gene family to study interms of the generation of resistance specificities. Thesecond subclass among the LRR are those encodingpredicted extracytoplasmic LRR by the Cf2, 4, 5, and 9 genesthat confer resistance to different races of Cladosporiumfulvum in tomato (Jones and Jones 1997). In wheat, a groupof LRR genes that resemble the Cf. class of resistance genesis associated with meiotic processes (Ji and Langridge 1994)but has not been shown to be involved in disease resistance.A third subclass of the LRR group is found in the rice Xa21gene product that combines a receptor kinase domain similarto Pto in the C terminal and an N terminal half with a LRRdomain. Xa21 confers resistance to several races ofXanthomonas oryzae pv oryzae that cause bacterial blight inrice.

The mlo resistance gene in barley represents the 4thdistinct resistance gene class. The gene product encodes a

trans-membrane spanning protein and the wild type gene isconsidered to function as a negative regulator of cell death.mlo provides broad spectrum resistance to a range of barleypowdery mildew isolates.

Resistance gene analogs

The knowledge gained from the sequence structure of clonedresistance genes has been exploited in the cloning ofnumerous related family members and is collectivelyreferred to as resistance gene analogs (RGA). Most of thestrategies for cloning RGAs rely on the use of degenerateprimers for PCR amplification based on conserved motifs inthe NBS and the neighbouring region (Fig. 3). The conservedmotifs found among cloned plant disease resistance genesdistinguish them from other plant NBS containing seguencessuch as the superfamily of GTP-binding proteins. (Meyers etal. 1999). A large collection of RGAs, in particular thosebased on NBS-LRR sequences, have been isolated frompotato, soybean, rice, maize, wheat, and barley and havebeen placed on genetic maps of the respective species.Several RGAs were shown to map to locations of knowndisease resistance phenotypes. Some of these co-segregatingRGAs with resistance phenotypes were later demonstrated tobe the actual resistance genes. For example, a combination ofa transposon-tagged site in a co-segregating RGA familymember at the rp1 locus in maize led to the unequivocalidentification of the Rp1D rust resistance gene (Collins et al.1999). In the absence of demonstrating gene function, RGAsco-segregating or closely linked with known resistancephenotypes have proven to be sources of perfect or tightlylinked markers for resistance breeding in wheat (Feuilletet al. 1997; Seah et al. 2000, 2001; Spielmeyer et al. 2000).NBS-LRR sequences isolated from the Cre3 locus (Lagudahet al. 1997) originally derived from Ae. tauschii have provento be versatile in developing perfect markers for other cerealcyst nematode resistance genes located elsewhere in thewheat genome. Homologues of the gene sequences from theCre3 locus were shown to be located at a non-syntenic locusthat was diagnostic for Cre1 (Lagudah et al. 1998) and werealso found on introgressed segments in wheat containingCre5 and Cre6 nematode resistance genes (Seah et al. 2000;Ogbonnaya et al. 2001).

Genome organisation of R genes

Among the major group of disease resistance genes, namelythe NBS-LRR sequences isolated in plant genomes, 2distinct classes can be found. This distinction is based on thepresence of a toll-interleukin receptor (TIR) domain in the Nterminal half of the predicted protein in one subclass and theabsence of the TIR domain in the second subclass.Interestingly the TIR class has not been isolated or foundamong monocotyledonous databases. Thus whereas bothTIR and non-TIR classes are present among dicotyledonousplants, only the latter class occurs in cereals. From the

1056 P. Langridge et al.

N B S

conserved sequence

primer

L R R s

MHDLGLPLKinase2P-loop

recently completed Arabidopsis genome sequencing project,150 NBS-LRR gene members (~0.6% of the genome) wereidentified. In rice, it is estimated that 750–1550 NBS-LRRgenes (5–10 times that of Arabidopsis) may be found in thegenome (Meyers et al. 1999; Michelmore 2000). Theanalysis of this superfamily of NBS-LRR genes in wheat hasrevealed that they are present in all homoeologous groups(Spielmeyer et al. 1998).

Disease resistance phenotypes often occur as clusters inplants and similarly cloned disease resistance genes and theiranalogs are also clustered in the plant genome. Some of theclusters may constitute a family of closely related genesequences arranged in tandem, whereas in other regions ofthe wheat genome, mixed clusters of non-cross hybridisinggene members have been identified (Leister et al.1998; Weiet al. 1999). However, within a family of gene membersclustered at a locus, both functional and pseudogenes mayoccur of which the latter may be characterised by deletions,base changes leading to frameshifts, and in some cases thepresence of retroelement insertions (Ronald 1998). Allelicvariation associated with clusters of disease genes hasrevealed divergent haplotypes within a species; a featurecharacteristic of rapidly evolving gene families (Leister et al.1998). In the D genome progenitor of wheat, RGAhaplotypes exhibiting a null variant with the complete loss ofgene members, were found to be as frequent as haplotypesranging from 2 to 5 gene members at a locus on chromosome2DS (Lagudah et al. 2001).

A significant proportion of resistance gene clustersanalysed so far appears to confer single specificity against apathogen; however, there is some evidence for multiplespecificities that arise from related gene sequences that aretightly linked (Botella et al. 1998). There are also exampleswhere homologues of a single resistance gene cluster inpotato confer resistance to a virus and a nematode (Van derVossen et al. 2000). Interestingly, there is an example wherea single gene member within a cluster confers resistance toboth nematode and aphid pests in tomato (Milligan et al.

1998; Rossi et al. 1998). The leucine-rich region of NBS-LRR gene members is usually the most variable part withina gene family and has been suggested to have a major role inthe evolution of resistance specificity. Comprehensivestudies based on mutants and domain swaps from the flaxrust resistance genes have shown that specificity is alsoinfluenced by parts of the predicted resistance gene productsother than the LRR domain (Ellis et al. 1999). The wheatgenome is full of different rust resistance genes with varyinginfection types against different races. The molecular basisof resistance specificity in wheat remains unknown untilsuch genes have been cloned, although there are sufficientgrounds to expect similar patterns in model plantexperimental systems to operate in wheat.

Adult plant resistance in wheat

Numerous combinations of seedling rust resistance genes areemployed in wheat breeding programs. These genes exhibitcharacteristics of the gene-for-gene system and in view ofthe development or appearance of virulent strains, there is aregular need to identify new sources of effective resistance.The number of well characterised adult plant resistancegenes to rust is few in comparison with the seedlingresistance genes. According to McIntosh et al. (1995), theadult plant resistance gene, Sr2, is argued to be the mostimportant stem rust resistance and among one of the keydisease resistance genes being used in modern wheatbreeding. Efforts at identifying molecular markers to trackSr2 in breeding are further described by Sharp et al. (2001,this issue). Another adult plant resistance gene or genecomplex that has been widely used is the Lr34 gene (Dyck1987; Singh 1992; McIntosh et al. 1995). In addition toproviding resistance to leaf rust, it is invariably associatedwith a stripe rust resistance gene, Yr18 as well as exhibitingtolerance to barley yellow dwarf virus. Another phenotypictrait associated with Lr34 is leaf tip necrosis under suitableclimatic conditions. Lr34 was reported to be on chromosome7DS and in some wheat lines there is evidence for a second

Fig. 3. Structure of a nucleotide binding site-leucine-rich repeat (NBS-LRR) disease resistance gene. The diagnostic markers forCCN resistance in wheat are derived from this class of sequences. Degenerate primers can be made for the conserved motifs (arrowedregions) and used in PCR amplification to isolate resistance gene analogs.

Trends in genetic and genome analyses in wheat 1057

location associated with a translocation event (Dyck et al.1994).

Because of the importance of Lr34 in rust resistancebreeding, and in some cases difficulties associated with itsselection in breeding programs, several attempts have beenmade to identify molecular markers tightly linked to the gene(Nelson et al. 1997; William et al. 1997; Faris et al. 1999).In one study, BSA was used to identify markers thatdistinguished pooled samples with and without the Lr34gene. However, genetic linkage analysis showed the markerto be present on the long arm of chromosome 7BL (Williamet al. 1997) instead of the expected short arm 7DS location.A study of a recombinant inbred population in which a majorQTL was identified as the predicted position for Lr34,revealed that when natural field infestations of leaf rustpathogens were the sole source of inoculum, a major regioncontributing to resistance was located on 7BL (Faris et al.1999). The cloning of the maize Rp1D rust resistance genefrom the rp1 locus and subsequent confirmation that Rp1homologues were present on barley chromosome 7HS(Collins et al. 1999; Ayliffe et al. 2000), opened up newpossibilities for further exploring syntenic appproaches toidentify markers linked to Lr34. Furthermore, some allelesof the rp1 locus with disease lesion mimic phenotypes, andbroad spectrum resistance was considered to be analogous tothe leaf tip necrosis trait associated with Lr34 (Hulbert1997). Heterologous probing using the maize Rp1 sequenceson wheat ‘isolines’ and parents differing for Lr34 wereunsuccessful in identifying polymorphisms for 7DS(Spielmeyer et al. 1998). The maize sequences were mappedto a polymorphic region on 7AS. However, the 7AS locationwas considered to be distally placed away from the predictedhomoeoloci for Lr34. Thus the hunt for a tightly linkedmolecular marker for Lr34 has so far proved to be elusive.

Mapping and tagging quality traits

Grain and flour quality traits are also key targets for thedevelopment and application of molecular markers in wheat(Tables 6 and 7) for 3 reasons. The first is the nature of thequality tests that are undertaken in breeding programs. Manyof these tests require considerable amounts of grain, and sothe tests can therefore only be performed in late generationsof the program. The second is that the quality of grain forvarious end products can be, and often is, affected byseasonal factors such as drought, frosting during graindevelopment, or pre-harvest rainfall. This can mean inparticular years, that a considerable proportion of thebreeding program will not have particular quality testsperformed, or the data obtained are unreliable. The thirdreason is the strict approval processes in many countries(especially developed) for the release of new cultivars intoquality grades, with the quality grade having a strong link tothe price producers can expect for their wheat. Incombination with the reasons for developing markers for

quality mentioned above that also affect mapping, thedifficulty of collecting quality trait information onpopulations and the low heritability of many qualitycomponents in wheat have meant slower progress in mappingand using markers for wheat quality traits than for the moredefined malting quality traits in barley.

Despite these difficulties, Tables 6 and 7 indicate progresswith wheat quality trait mapping and tagging. For breaddough properties, the effects of various alleles of the storageprotein loci are well known. For example, the high (HMW)and low (LMW) molecular weight glutenins, and gliadins areclassified by protein SDS-PAGE, and variation in many ofthese is known to be related to dough properties of bread anddurum wheats (Payne et al. 1979; Hamer et al. 1992; Zhenand Mares 1992). As a consequence, many breedingprograms screen for these traits by SDS-PAGE of the grainproteins. Many of the genes encoding these proteins are nowcloned and sequenced, and PCR-based DNA tests for anumber of these loci, or for particular alleles, are availablefor use in bread and durum wheats (e.g. D’Ovidio andAnderson 1994; Ahmad 2000). Such tests could replace theSDS-PAGE protein gel tests as breeding programs begin touse DNA-based markers, since the DNA will be available fortesting.

The grain protein content (GPC) is also a target trait ofconsiderable interest as GPC is a key determinant of gradeand crop price. Prasad et al. (1999) suggested that a QTL on2DL controlled some of the high GPC of an Indian cultivar,and that this can be identified by a linked SSR marker. Afurther locus that has been well studied was transferred tobread wheat (via durum) from T. turgidum var. dicoccoides(Joppa et al. 1997; Mesfin et al. 1999). This locus on 6BSincreases GPC without a decrease in yield (Mesfin et al.2000), and is tagged by a number of PCR-based (SSR andSTS) markers (Khan I. A. et al. 2000). This locus and theassociated markers are ideal for implementation intoprograms aimed at high GPC because the markers are PCR-based, tightly linked, and diagnostic. In addition, mostprograms will not have the beneficial allele present, and thebeneficial allele for GPC does not appear to have otherdetrimental effects.

Grain hardness is also a key quality determinant. Themutations in various genes controlling grain puroindolineproteins around the ‘hardness’ locus on 5DS are nowbecoming well studied and PCR-based tests for thesemutations are available (Lillemo and Morris 2000). Therelationships between these markers, grain hardness, and endproduct quality are under intense research (Martin et al.2000; Turnbull et al. 2000). Further minor genes controllinggrain hardness and markers linked to these are now knownfrom QTL studies (Campbell et al. 1999; Perretant et al.2000), and this information may be useful in breeding toselect for minor variation in hardness. The QTL study ofCampbell et al. (1999) revealed markers linked to loci

1058 P. Langridge et al.

controlling grain morphology traits. One of these, grainweight, was also studied by Varshney et al. (2000). Thesestudies are important as they relate to grain classification incommerce, and they may also be related to milling yield, akey quality trait as determined by the amount of flour amiller gets from wheat grain. QTL mapping of this key traitwas undertaken by Parker et al. (1999), and 3 QTLs wereidentified (Table 6).

Flour colour is also a key trait in wheat quality. Yellowflour is a major quality criterion for yellow alkaline noodleproduction, whereas white flour is desirable for breadmaking. Parker and Langridge (2000) have converted anAFLP marker (Fig. 3) for a locus on chromosome 7A thatgives a high level of carotenoids to the cultivar Schomburgkto a STS marker. Therefore, the locus can be more easilyfollowed in breeding programs and is being actively used insome programs.

As its economic impact on producers and regions can besevere, there is considerable interest in mapping and tagginggenes important for pre-harvest sprouting (PHS) tolerance,an environmentally induced grain quality defect. QTLstudies by Anderson et al. (1993) and Zanetti et al. (2000)indicated that a large number of loci, many different betweenthe two studies, are involved in PHS tolerance. A BSA study(Varshney et al. 2000) has also suggested another locus inIndian wheats. These studies mean that implementation ofthese markers in a breeding program will first requireconsiderable research to identify those regions segregatingwithin a particular program. Perhaps this trait requiresdissection into its components (Mares 1995) beforesignificant progress can be made.

The final quality trait where progress has occurred andwhere markers are used in breeding programs is in whitesalted noodle quality. Work by Zhao et al. (1998) indicatedthat the common null allele of the waxy locus onchromosome 4A (Wx-B1b) is strongly associated with, andprobably a diagnostic marker for, changed starch granuleproperties that result in higher textural quality in white saltednoodles. This allele can be followed by SDS-PAGE, and alsoby PCR-based DNA markers (Briney et al.1998;McLauchlan et al. 2001, this issue). There are also nullalleles at the other 2 Wx loci on chromosomes 7A and 7D,and PCR-based markers are available for these null alleles(Shariflou and Sharp 1999) to aid the marker-assisteddevelopment of waxy (zero amylose) wheat cultivars.

Marker development methods

Essentially, two techniques have been used to date to findmarkers linked to traits of interest. These have involved theconstruction of complete linkage maps or the use of BSA.The key features of the application of these two techniquesto wheat are outlined below.

There are 8 or 9 main steps in developing molecularmarkers (Table 8). In Table 8, markers developed from fully

mapped populations (assuming 10 loci amenable to MarkerAssisted Selection [MAS] per population) are comparedwith those developed from BSA (assuming one locus perpopulation). The information is based on costs associatedwith the National Wheat Molecular Marker Program inAustralia and is essentially historical. Methods for markerscreening have improved in recent years and costs wouldnow be substantially lower (less than half). However, markerdevelopment is still an expensive science with mapconstruction, phenotyping, and validation requiring thegreatest resources. The cost of markers derived from BSA isless than half that of fully mapped populations in wheat.However, the power of BSA is limited, in most instances, totraits controlled by one or two major genes.

Full linkage maps

Complete linkage maps generated from screening the progenyof a cross have provided the basis for most early markerdevelopment work. However, this is difficult and labour-intensive, particularly in wheat, where linkage maps must beconstructed for 21 chromosomes. Usually 10–20 markers arerequired for each chromosome to give reasonable genomecoverage. This means 200–400 mapped loci. The workinvolved in screening this number of markers is considerableif markers such as RFLPs are used. In most crosses with com-mercial wheat cultivars only about 20% of cDNA or genomicprobes will detect an RFLP. This means that over 1000 probesare usually required. Combining RFLP with various othermarker systems offers an alternative approach. AFLP markersare particularly useful due to the high multiplexing ratio. Eventhough the level of polymorphism of individual AFLP bands isusually lower than for RFLPs, the large number of bands thatcan be scored in a single assay makes this a valuable markersystem for filling in gaps in linkage maps. However, there arefrequently difficulties in combining co-dominant anddominant markers on the same map.

The marker choice is complicated in wheat by the 3genomes. Most RFLP probes will reveal 3 loci, one on eachgenome (for example in Fig. 1). In constructing completelinkage maps, probes are selected that will give good genomecoverage. The high level of synteny between wheat, barley,and rye allows the location of RFLP loci to be extrapolatedacross to bread wheat. However, this only allows localisationof RFLP loci to a chromosome group and not to a genome.Further, good genome coverage may be hard to achieve asthe level of polymorphism between 2 parents is not evenacross all genomes. The D genome in particular is lessdiverse in bread wheats than the A and B genomes (Pestsovaet al. 2000a). The availability of many SSR markers helps toovercome this difficulty since these markers are generallysimple to assay, co-dominant, their location is known and,most importantly, they are usually genome-specific, thussimplifying the assignment of linkage groups (Plaschke et al.1995; Röder et al. 1995).

Trends in genetic and genome analyses in wheat 1059

It is also often important to consider how many traits ofsignificance may be segregating in the population used formarker development. If several important traits aresegregating, the benefits derived from constructing a fulllinkage map may outweigh the time and costs associatedwith full map construction.

The major linkage maps of wheat and its relatives arelisted in Table 1, and several of the maps are in wide use.

Bulked segregant analysis

Full map construction is not always the most appropriateapproach to finding markers linked to a trait of interest. BSAis, in many cases, a more rapid and cost-effective approach(see Table 8).

The technique of BSA was first applied to identifyingmolecular markers associated with economic traits byMichelmore et al. (1991). It involves pooling individualsfrom the 2 phenotypic extremes of a segregating F2, doubledhaploid, or similar population. DNA isolated from the 2pools is then screened with DNA markers, usually RAPD orAFLP but RFLP can be used, and polymorphic bandsidentified. Clear polymorphisms seen between the 2 poolswill be derived from regions of the genome that are commonbetween the individuals that made up each pool but differentbetween pools. The remainder of the genome will be

randomly contributed by the parents and should show nopolymorphisms between the pools. An example of how thistechnique is applied is shown in Fig. 4 where markers wereidentified linked to genes controlling yellow flour colour(Parker et al. 1998).

Bulked segregant analysis was successfully applied todevelop markers linked to agronomically important traitsin wheat (for example, Cre3; Eastwood et al. 1994). Thistechnique has clear advantages for marker identificationrelative to construction of complete linkage maps; only afew weeks are required to screen the pools with the AFLPor RAPD primers. However, there are also disadvantages:the method is difficult (although not impossible) forcomplex traits controlled by several unlinked loci; theactual location of the gene of interest is not revealed, oneonly finds linked markers; and the method can only beefficiently used with PCR-based marker systems such asRAPD, SSR, and AFLP markers. Both RAPD and AFLPmarkers will usually need to be converted to a more stablemarker system, such as STS, before they can be used in aselection program.

These problems may be overcome through the use of SSRmarkers. The effort involved in the initial screening of thepools may be greater for SSR markers and a large numberwill be required, but the benefits in having markers that are

Table 8. Steps and costs in developing molecular markers via two alternative strategies in the Australian National Wheat Molecular Marker Program (NWMMP)

The various stages in identifying markers were separately costed using the NWMMP as an example

Strategy 1. Full map construction via mapping population Strategy 2. Bulked segregant analysis (BSA)Steps Cost (AU$)A Steps Cost (AU$)B

Characterise genetics of trait; complex trait (or many single gene traits)

Characterise genetics of trait; simply inherited traits

Choose parents; design and coordinate strategy

20000 (1.9%) Choose parents 2000 (3.6%)

Construct large population(c. 200 individuals)

40000 (4%) Construct population, c. 50 individuals

10000 (18%)

Multiply seed 5400 (0.54%) Multiply seed 1500 (2.7%)Map construction 450000 (45%)Phenotype mapping population 15000 (agronomic trials)

110000 (disease)200000 (quality) (31%)

Phenotype individuals 20000 (36%)

Construct bulks (including statistical analysis)

1000 (1.8%)

QTL analysis, including statistical analysis

10000 (1%) Genotype bulks 5000 (9%)

Identify putative markers 1000 (1.8%) Validation of putative markers 150000 (15%) Validate in related pedigree 15000 (27%)Total cost 1 000 400 Total cost 55500Cost per marker 100040 (assumes 10 loci

identified per population)Cost per marker 55500 (assumes 1 locus

identified per population)

APer ‘full map’ (percentage of total in parentheses).BPer BSA (percentage of total in parentheses).

1060 P. Langridge et al.

immediately applicable in selection programs may overcomethis drawback. It is probable that we will see reports of theuse of SSR markers in BSA over the next few years as moreSSR markers become available.

Application of molecular markers

Analysis of diversity

Measures of genetic distance within a crop provide anindication of the average genetic divergence between cul-tivars. A wide range of phenotypic traits have been used toassess relationships between cultivars, including agronomictraits that have been useful for the analysis of variability

(Smith 1984), the selection of parents for hybrid construction(Frei et al. 1986), and for the prediction of progeny per-formance (Grafius 1956). Quantitative agronomic charactershave also been used to select superior parental combinations(Whitehouse 1969) and to predict the specific combiningability in F1 progeny of spring wheat (Shamsuddin 1985).The level of genetic similarity between genotypes can bepredicted from the coefficient of co-ancestry (Malécot 1948)based on pedigree information, and an extensive pedigreedatabase for wheat has been produced by CIMMYT, Mexico:

http://www.cimmyt.mx/Research/Wheat/IWISFOL/IWISFOL.htm

White YellowParentParent

Fre

qu

en

cy

Yellow

Yellow Bulk

White

White Bulk

Flour colour score

Yellow BulkWhite Bulk

White ParentYellow Parent

Fig. 4. Bulked segregant analysis identifying AFLP markers linked to flour colour. The upper panel showsthe result of scoring an F7 population of the cross Schomburgk × Yarralinka for flour colour. DNA from the10 whitest lines was pooled to produce the ‘White Bulk’ and DNA from the 10 most yellow lines the ‘YellowBulk’. The lower panel shows a section from an AFLP gel where the products from the 2 parents and the 2bulks are compared. The arrows indicate bands that are present only in the AFLP products of the yellowparent and the Yellow Bulk. The products were generated using PCR primers labeled with 32P.

Table 9. Recent studies where molecular markers have been used to study aspects of germplasm diversity in wheat and its wild relatives

Germplasm Number of lines

Marker type Number screened

Reference

German and Austrian winter wheats

55 RFLP probesAFLP primersSSR

1171621

Bohn et al. 1999

Triticum accessions RAPD Cao et al. 1999Bread wheat 55 SSR 20 Prasad et al. 2000Aegilops tauschii 113 SSR 18 Pestsova et al. 2000bTriticum dicoccoides 21 SSR 23 Fahima et al. 1998Australian wheats 124 RFLP Paull et al. 1998Bread wheat 65 SSR 14 Donini et al. 1998

Trends in genetic and genome analyses in wheat 1061

Alternatively, genetic similarity can be obtained directly bymeasuring resemblance for biochemical or molecular mark-ers (Smith et al. 1991). A major advantage of molecularmarkers is that they offer an easily quantifiable measure ofgenetic variation within crop species. With these advantages,molecular markers have become a critical tool in studies ofdiversity in bread wheat, durum wheat, and the diploid pro-genitors. Table 9 summaries some of the more recent studies.

The study of molecular diversity in wheat has severalimportant implications. A reliable estimate of diversity andrelatedness allows a re-evaluation of lines kept in germplasmcollections and may help minimise duplication, therebyreducing and improving the quality of germplasmcollections. For example, CIMMYT has begun a program toDNA-fingerprint a substantial proportion of its wheatgermplasm collection. When combined with sound pedigreeand phenotypic databases, the germplasm collections canserve as a means for identifying loci and genome regionsimportant for adaptation to specific environments, toleranceto environmental stress, and even aspects of quality.

Measures of diversity can also prove useful in identifyingthe optimal parents for crosses. This applies to both pedigreeand backcross breeding strategies.

Taking a marker through to application in a breeding program

Although markers are valuable tools in the study of diversity,as outlined above, and in developing an understanding of thegenetic control of various traits, their key application is insupporting breeding and selection programs. However, theidentification of a marker linked to a trait of importance isonly the first step in its application. There are 3 importantrequirements for the effective implementation of molecularmarkers.

Firstly, the availability of reliable and efficient strategiesis needed for marker use in the breeding programs. Thisinvolves the analysis of polymorphism and marker behaviourin different genetic backgrounds. Importantly, the markermust be able to detect polymorphisms between and withinthe lines or varieties used in the breeding and selectionprogram. Breeders need to carefully evaluate their breedingmethodology to ensure that MAS is used effectively, both interms of selection and cost.

Secondly, markers need to be easy and cheap to use. PCR-based techniques are the preferred marker types at presentdue to the simplicity of the assays relative to RFLP. Theavailability of extensive collections of SSR markers and thegenome specificity of many of these markers make themideal. New techniques based on the detection of singlenucleotide polymorphisms may provide an improved markersystem since they may not require gel-based detection.

Thirdly, a skill and equipment base must be established toensure ready access by breeders to the markers and markertechnology.

Several steps are needed before a marker can be usedin a practical breeding program with marker validationbeing a key element. Irrespective of the technique usedto identify a marker linked to a target trait, themarker–trait association has been found in a particularset of circumstances; usually in the progeny of a specificcross. The reliability of the marker can be estimatedfrom the closeness of linkage. As the next step, theability of the marker to predict the target phenotype mustbe tested in further populations. Usually one would aimto keep one parent in common with the parents used inthe initial mapping population. Therefore, the predictivevalue of the marker in identifying the phenotype of theplant can be estimated and compared with the originalmapping result.

The usefulness of a marker in screening and assayinggermplasm in use in the breeding program is assessed asthe next step. The parents likely to be used in a markerscreening program are screened for polymorphismsbetween those that have and do not have the target trait.Given the narrow germplasm base in bread wheat, a highproportion of markers will fail this test of applicability(about 80%). It is then necessary to see if other markers canbe found in the vicinity of the target locus that could beused in screening. Ideally one would aim to have about 10marker loci within 10 cM of the target locus to have areasonable chance of being able to track the locus in thegermplasm base used in a breeding program. At this point,markers such as SSRs are particularly valuable since theytend to detect a greater level of diversity than RFLP-basedmarker systems and the genome specificity of SSR markershelps to overcome the problem of tracking a particularlocus. This is particularly important for marker applicationin wheat breeding.

Finally, a clear protocol for use of the markers must beprovided. This will include the following information.

� Which molecular marker to use for a given germplasmcombination.

� The protocol for the marker assay and the expected size ofthe alleles.

� The reliability of the marker in terms of the actual successrate of the assay; SSR markers can vary in their ease ofassay, some working almost 100% of the time and othersshowing less than 80% success rate. This is often due tothe sequence of the primers.

� The closeness of linkage; this provides a measure of thesuccess of the marker in predicting the target locus orallele.

Pedigree or progeny breeding

The possible roles of molecular markers in pedigree/progenybreeding can be summarised by reviewing the stages in apedigree breeding strategy. In selecting the parents for the

1062 P. Langridge et al.

cross, a sound understanding of the diversity and relationshipbetween possible parents has the potential to increase thelikelihood of desirable outcomes. Molecular fingerprintingof possible parents can be used to either broaden or narrowthe diversity in breeding material. In hybrid wheat programs,the maximisation of diversity may help optimise heteroticeffects. At this stage, markers can also be useful inconfirming the genotype of parents used in crosses.Fingerprinting of lines in the parent blocks can offersignificant savings. We estimate that as many as 10% of thelines currently used as parents in Australian wheat breedingprograms are mixtures, are not pure lines (particularly someold varieties that can be heterozygous at over 10% of loci),or were incorrectly labelled.

At the next stage, marker screening of F1s can offer majoradvantages through eliminating lines that do not carry thetarget loci. Importantly, complex intercrosses, such as F1intercrosses and topcrosses, can be assayed for presence oftarget loci. Although not currently being widely used inwheat breeding programs, this approach could offer majorimprovements in the rate of gain through breeding byaccelerating and expanding the number of traits beingselected for or against in a minimum time (number ofgenerations).

The use of markers in screening segregating populationsoffers clear advantages over the use of bioassays for a widerange of traits, including many quality, abiotic stresstolerance, and some disease resistance loci. The applicationof markers at this stage needs to be carefully compared withalternative screening procedures since in some cases thebioassay may be cheaper, faster or more reliable. To be trulyeffective in screening segregating populations, highthroughput marker screening procedures will be needed. Insome selection strategies, markers offer the only feasiblescreening method. This will frequently be the case wheremany desirable traits are being combined in parent building.

Currently, there appears to be little role for markers in lategeneration screening, although this may change as locicontrolling complex characteristics, such as yield stabilityand broad adaptation, are located.Finally, in both pedigree and backcross-based breedingstrategies, markers can be applied to testing line purity andin supporting the commercialisation and protection of newvarieties.

Backcrossing

In backcross breeding, markers can be applied at almostevery major stage. DNA fingerprint patterns of germplasmcollections can be used to identify the best parents for across; for example, to identify the most closely relatedparents so that the extent of backcrossing needed to re-establish the elite parental background is minimised. Inaddition, as described above, markers can be used to checkthe integrity of parental lines.

At each backcross cycle, markers can be used to identifylines carrying the target trait, to select individuals with thesmallest amount of donor genome, and to find lines with thesmallest introgressed segment around the target locus. Forrecessive traits, marker screening can save the extrageneration needed to produce homozygous individuals forreliable phenotyping.

The same principles apply in later generation screeningfor pedigree breeding. Markers provide a powerful techniqueto ensure purity of lines and varietal identification forcommercialisation.

Future directions

There are several developments that we can expect in markertechnologies over the next few years. The techniques forassaying DNA variation between lines will become cheaperand allow higher throughput than is presently possible.Although the cost per assay may come down, the cost ofrunning an advanced molecular marker laboratory will go upthrough the need for sophisticated DNA analysis anddetection systems and robotics. The cost of DNA isolationwill continue to be a major limitation, although this shouldalso come down to some extent. We can also confidentlyexpect the number of useful traits tagged with molecularmarkers to increase dramatically. The application of geneticengineering to wheat improvement will further increase thenumber of genes (loci) that will be monitored through abreeding and selection program. These trends will encouragebreeders to conduct detailed molecular analyses of lines andwill greatly expand the amount of information that isavailable on each line. With these points in mind, we believethat the following issues will dominate the development andapplication of MAS.

Polymorphic markers

In the implementation of molecular marker techniques towheat improvement the lack of polymorphism amongstparents is an important factor limiting MAS applications.Numerous SSR markers are currently being added to RFLP/AFLP-based maps. SSR markers show considerable promisebut are yet to be widely implemented in practical breedingprograms. Many wheat SSRs are now available but thenumber that will actually be needed for effective applicationwill be very high. For full genome coverage of wheat thereshould be around one SSR marker every 10 cM. Since thewheat genome is ~9000 cM, this means at least 900 SSRmarkers would be required. Further, approximately 60% ofSSRs are able to detect polymorphisms between varieties,meaning that ~1500 would really be needed. If one allows forthe unevenness in marker distribution this number could beeasily double. If MAS at targetted loci is desired, the numberof SSRs may in practice be around 10 to account for mostparental combinations. Although a large number of new

Trends in genetic and genome analyses in wheat 1063

SSRs are expected to emerge from the EST programs aroundthe world, EST-based SSRs are reported to detectsubstantially less variation than genomic-derived SSRs.Alternative marker systems such as those based aroundSNPs are therefore becoming increasingly attractive.

Validation of markers

Most groups undertaking MAS underestimate the cost of the‘validation’ process required after the initial identification ofa potential region of interest in the mapping population. Inmany cases, finding a linked marker in a mapping populationhas proved relatively straightforward, whereas implementingit in a breeding program has not. The validation process isdefined as the testing of an allele for its effect in geneticbackgrounds other than the original mapping population, thecharacterisation of the polymorphisms for a range ofcandidate markers in all combinations of parents, and thebench testing of the selected enzyme-marker/marker-protocol. Approximately 15–25% of the total MASresources of the Australian National Wheat MolecularMarker Program go into this process, although some staffbelieve that this may be an underestimate (Table 8). Theattractiveness of MAS would therefore be greatly enhancedby reducing the overheads associated with validation. It maybe that diagnostic markers are one possible way to reducethese costs.

Marker throughput

The following are the key limitations.

� The cost and time for DNA extraction. Rapid DNAisolation systems are available that may reduce the costbut this will continue to be a major cost in MAS.

� The availability of closely linked, polymorphic markers.The public and private efforts in wheat genomics mayprovide sufficient SSR markers to supplement thosecurrently available and thereby overcome this limitation.Some breeders still hold concerns about the use of MASleading to a narrowing of the germplasm pool to onlythose parents and traits for which validated, polymorphicmarkers are available.

� Staff resources for handling the plants, DNA, and markerallele data. Many breeding programs have funded theirinitial work in MAS from special research funds but thetime is fast approaching for public sector breeders, as didprivate breeders, to re-allocate resources within the‘breeding team’ in order to further implement MAS.

� New, robotic systems may be capable of analysing 104–105

loci per day but will challenge the information processingcapability of all but the largest breeding programs.

� The cost of robotics, enzymes, non-gel-baseddiscrimination systems, etc. for true, high-throughputsystems. Although such systems show great potential forhandling the targetted numbers, they appear at presentbeyond the budget of most wheat breeders and geneticists.

Risks and limitations of using MAS

The risk most frequently raised is the temptation to use onlyparents for which either markers and/or polymorphicmarkers exist, thus further narrowing genetic diversitywithin breeding programs. In particular, this mayconcentrate use of a few, well characterised diseaseresistance genes to the exclusion of less well documentedsources. This risk can be minimised by breeder discretionallocating a proportion of the program to ‘new’/‘uncharacterised’ sources as has already happended in somebreeding programs. It might be more useful to think of thisproblem as a challenge of how marker technology be used toexpand our useful genepool. Tanksley and Nelson (1996)propose one solution through their ‘advanced backcrossQTL analysis’ where the introgression of wild germplasmcan be accelerated.

Marker technology can also help breeders know theposition of the genes they are using; in the case of a diseaseresistance program, a breeder may be using many sources ofthe same, allelic or closely linked resistance in the belief thatgene pyramids are possible and risk is spread.

There is also concern that a strong emphasis on molecularmarkers in a breeding program may lead to a focus onbreeding strategies based on the technology rather than themost appropriate strategy and germplasm for a particularenvironment. For example, backcrossing is highly amenableto MAS but is a conservative breeding strategy and shouldnot become the prime focus of a breeding program. Againthis issue can be addressed if the breeders are aware of thepotential problems and include a diversity of strategies intheir program.

Another limitation is the power of QTL discrimination(Melchinger et al. 1998). Many QTL indicated in mappingpopulation studies ‘disappear’ in validation populations.These QTL may have been cross-specific, subject togenotype environment interaction effects, or illusory.Illusory QTL may also be artefacts of small mappingpopulations, error in the phenotyping experiments, or reflectfundamental limitations to QTL analysis methods.

Diagnostic markers

The attraction of ‘diagnostic markers’ (where the marker isderived from the actual functional gene sequence or showsno recombination with the target gene) is almostoverwhelming, particularly when coupled with a simpleanalysis system, such as SNP. It may be feasible to delete (orat least greatly reduce) the time-consuming validation andparental polymorphism stages and increase the precision ofmarker assays. However, the costs of developing suchmarkers, although not a limitation for the human genomeprojects, may be prohibitive for wheat and most crop species.The importance of diagnostic markers is also closely tied tothe breeding strategies being used. Marker systems are nowallowing the analysis of the whole genome and it is likely that

1064 P. Langridge et al.

selection strategies based around monitoring linkage blocksand identifying specific recombination events will gain insignificance (see below). In such circumstances, diagnosticmarkers will decrease in significance.

Alternative map construction techniques

Marker identification through linkage disequilibrium map-ping is a possible alternative to the current strategies basedon the development of mapping populations and BSA (Elsn-er et al. 1995). This technique is widely used in animal, par-ticularly human, mapping. Will such an approach change thecost of marker development? Since breeding populationsprovide the lines for analysis, it will reduce costs associatedwith population construction, seed preparation, and pheno-typing. However, many more individuals will need to be gen-otyped (c. 1000) and the overall cost of analysing apopulation may be of the same order of magnitude as for amapping population, described in Table 8. The question isthen ‘will the discrimination be better than the mapping pop-ulation or will the number of traits considered be larger?’The technique has been used to identify marker–trait associ-ations in several species including wheat (Paull et al. 1998)but is otherwise largely untested in plants. Experience in hu-man mapping indicates that the technique can resolve com-plex traits into many factors of small effect. A furtherattraction is that mapping results are achieved simultaneous-ly as the breeding population progresses through the selec-tion generations. In contrast, results from mappingpopulations are often achieved after the first cohort of rele-vant material has progressed through the selection system.

Data management and breeding software

From the information summarised in this review, it is clearthat a huge amount of genetic data is being generated forwheat. There are now several well mapped populations forwhich detailed marker information and extensive phenotypicdata are available. There have been several studies of geneticdiversity in wheat and its progenitors and there are severalprograms now underway around the world to fingerprintwheat varieties, breeding lines, land races, and wildgermplasm of wheat relatives and progenitors. In addition,there have been many publications describing marker traitassociations for both major genes and QTL. At present,much of this information is dispersed and not readilyaccessible to wheat breeders or geneticists. The GrainGenesdatabase has been addressing this problem and provides themost comprehensive collection of wheat genetic data (http://wheat.pw.usda.gov/). CIMMYT has also been developingsoftware and databases for the display of germplasm andpedigree information and data from extensive, worldwidefield evaluations (http://www.cimmyt.mx/). The next stagewill be to provide the information in a form that can bereadily used by wheat breeders to identify crosses most likelyto produce significant advances and determine the most

appropriate crossing and selection strategy. Several researchprograms around the world are addressing some of theseissues and looking at improved methods for theinterpretation and utilisation of genetic data. These includeprograms such as QU-Gene (http://pig.ag.uq.edu.au/qu-gene/) and ICIS (http://www.cimmyt.mx/), which aim toprovide tools for plant breeders to design and managebreeding programs that integrate molecular and otherdatasets. There is also commercially available software, suchas Geneflow and Synergene (http://www.geneflowinc.com/),that offer some tools for testing different selection strategiesand improving the ease with which molecular data can bedisplayed and used.

Managing linkage blocks

The wild relatives of wheat have proved valuable sources ofa wide range of useful traits, particularly disease resistances.However, the behaviour of these alien segments in the wheatgenome has presented some problems that are unique towheat. In many cases the alien chromosome or chromosomesegment has shown very low recombination with wheat evenwhen it has been derived from a close relative (for example,Sr22 on 7A from Triticum boeoticum; Paull et al.1994). Incases where homoeologous chromosomes from relativessuch as rye have been used, special techniques have beenrequired to induce recombination (for example, the short armof chromosome 1R from rye; Rogowsky et al. 1991).Molecular markers have proved valuable in identifying therare lines where recombination has occurred and incharacterising the recombination events. They also provide apowerful tool for monitoring the transmission of theselinkage blocks.

Although the low rate of recombination between alienchromosome segments and the bread wheat chromosomeswas predicted, fingerprinting of commercial varieties hasindicated that low recombination rates and transfer of largelinkage blocks may also characterise crosses made betweencommercial varieties. Varietal screens, such as described byPaull et al. (1998), suggest that the transfer of largechromosome segments during breeding and selection, andselection for recombinant events between chromosomes,have been far lower than in other crops such as barley (Barret al. 2000). New work involving association mapping inwheat appears to confirm these conclusions. These resultssuggests that the strategies used for applying molecularmarkers to wheat will be different from those that will beapplied to other major crops. In particular, the managementof linkage blocks and the identification and localisation ofrecombination events will become a key issue in developingnew wheat breeding strategies.

Access to EST databases

It is estimated that ~250000 wheat ESTs will becomepublicly available over the next 2 years (Dundee Workshop

Trends in genetic and genome analyses in wheat 1065

on Public Wheat and Barley Genomics Programs, May2000). This is expected to provide ~2000 new SSR markersand will help drive the development of SNP markers forwheat. The impact of this work will be to increase the densityof markers available on maps and help drive down the costsof marker-based screening. It is also expected that thetechniques of map-based cloning will become moreaccessible to wheat researchers so we can expect to see theisolation of a wide range of genes that may be applied togenetic engineering of wheat or to the development ofdiagnostic markers.

Integrating transgenes into breeding programs

MAS has been extremely important in the selection andmonitoring of transgenes following the development oftransgenic plants. Therefore, it is important that geneticengineering be included in the planning of breeding method-ologies. The expectation is that many genes will be amenableto manipulation via genetic engineering. The transgenes willneed to be closely monitored through the breeding programsto satisfy regulatory requirements and to ensure their rapiddeployment. This will have the effect of increasing thenumber of loci that need to be tracked and will requirebreeding strategies that effectively integrate the transgenescreening with other aspects of MAS and breeding targets.

Acknowledgments

This work was supported by the Grains Research andDevelopment Corporation through the National WheatMolecular Marker Program, the Australian ResearchCouncil, and the Cooperative Research Centre for MolecularPlant Breeding. We also thank Dr Bryan Whan for hisvaluable comments on the review and Dr Chongmei Dongand Margaret Pallotta for providing some of the figures. Thehelp and advice of the ITMI coordinators is also gratefullyacknowledged.

References

Ahmad M (2000) Molecular marker-assisted selection of HMWglutenin alleles related to wheat bread quality by PCR-generatedDNA markers. Theoretical and Applied Genetics 101, 892–896.

Anderson JA, Sorrells ME, Tanskley SD (1993) RFLP analysis ofgenomic regions associated with resistance to pre-harvest sproutingin wheat. Crop Science 33, 453–459.

Ayliffe MA, Collins NC, Ellis JG, Pryor A (2000) The maize rp1 rustresistance gene identifies homologues in barley that have beensubjected to diversifying selection. Theoretical and AppliedGenetics 100, 1144–1154.

Bai GH, Kolb FL, Shaner G, Domier LL (1999) Amplified fragmentlength polymorphism markers linked to a major quantitative traitlocus controlling scab resistance in wheat. Phytopathology 89,343–348.

Baker B, Zambryski P, Staskawicz B, Dinesh-Kumar SP (1997)Signalling in plant–microbe interactions. Science 276, 726–733.

Barloy D, Lemoine J, Dredryver F, Jahier J (2000) Molecular markerslinked to the Aegilops variabilis-derived root-knot nematoderesistance gene Rkn-mn1 in wheat. Plant Breeding 119, 169–172.

Barr AR, Jefferies SP, Warner P, Moody DP, Chalmers KJ, Langridge P(2000) Marker assisted selection in theory and practise. In‘Proceedings of the 8th International Barley Genetics Symposium’.(Ed. S Logue) pp. 167–178.

Becker J, Vos P, Kuiper M, Salamini F, Heun M (1995) Combinedmapping of AFLP and RFLP markers in barley. Molecular andGeneral Genetics 249, 65–73.

Blanco A, Bellomo MP, Cenci A, Degiovanni C, Dovidio R, Iacono E,Laddomada B, Pagnotta MA, Porceddu E, Sciancalepore A,Simeone R, Tanzarella OA (1998) A genetic linkage map of durumwheat. Theoretical and Applied Genetics 97, 721–728.

Bohn M, Utz HF, Melchinger AE (1999) Genetic similarities amongwinter wheat cultivars determined on the basis of RFLPs, AFLPs,and SSRs and their use for predicting progeny variance. CropScience 39, 228–237.

Borner A, Roder MS, Unger O, Meinel A (2000) The detection andmolecular mapping of a major gene for non-specific adult-plantdisease resistance against stripe rust (Puccinia striiformis) in wheat.Theoretical and Applied Genetics 100, 1095–1099.

Botella MA, Parker JE, Frost LN, Bittne-Eddy PD, Beynon JL, DanielsMJ, Holub ER, Jones JDG (1998) Three genes of the ArabidopsisRPP1 complex resistance locus recognize distinct Peronosporaparasitica avirulence determinants. Plant Cell 10, 1847–1860.

Boyko EV, Gill KS, Mickelson-Young L, Nasuda S, Raupp WJ, ZiegleJN, Singh S, Hassawi DS, Fritz AK, Namuth D, Lapitan NLV, GillBS (1999) A high-density genetic linkage map of Aegilops tauschii,the D-genome progenitor of bread wheat. Theoretical and AppliedGenetics 99, 16–26.

Briney A, Wilson R, Potter RG, Barclay I, Crosbie G, Appels R, JonesMGK (1998) A PCR-based marker for selection of starch andpotential noodle quality in wheat. Molecular Breeding 4, 427–433.

Bryan GJ, Collins AJ, Stephenson P, Orry A, Smith JB, Gale MD (1997)Isolation and characterisation of microsatellites from hexaploidbread wheat. Theoretical and Applied Genetics 94, 557–563.

Campbell KG, Bergman CJ, Gualberto DG, Anderson JA, Giroux MJ,Hareland G, Fulcher RG, Sorrells ME, Finney PL (1999)Quantitative trait loci associated with kernel traits in a soft × hardwheat cross. Crop Science 39, 1184–1195.

Cao W, Scoles G, Hucl P, Chibbar RN (1999) The use of RAPD analysisto classify Triticum accessions. Theoretical and Applied Genetics98, 602–607.

Cenci A, D’Ovidio R, Tanzarella OA, Ceoloni C, Porceddu E (1999)Identification of molecular markers linked to Pm13, an Aegilopslongissima gene conferring resistance to powdery mildew in wheat.Theoretical and Applied Genetics 98, 448–454.

Chalmers KJ, Campbell AW, Kretschmer J, Karakousis A, Henschke P,Pierens S, Harker N, Palotta M, Cornish GB, Shariflou MR,Rampling LR, McLauchlan A, Daggard G, Sharp PJ, Holton TA,Sutherland MW, Appels R, Langridge P (2001) Construction ofthree linkage maps in bread wheat, Triticum aestivum. AustralianJournal of Agricultural Research 52, 1089–1119.

Chao S, Sharp PJ, Worland AJ, Warham EJ, Koebner RMD, Gale MD(1989) RFLP-based genetic linkage maps of wheat homoeologousgroup 7 chromosomes. Theoretical and Applied Genetics 78,495–504.

Cho YG, McCouch SR, Kuiper M, Kang MR, Pot J, Groenen JTM, EunMY (1998) Integrated map of AFLP, SSLP and RFLP markers usinga recombinant inbred population of rice (Oryza sativa L.).Theoretical and Applied Genetics 97, 370–380.

Collins N, Drake J, Ayliffe M, Qing S, Ellis J, Hulbert S, Pryor T (1999)Molecular characterisation of the maize Rp1D rust resistancehaplotype and its mutants. Plant Cell 11, 1–13.

Dib C, Faure S, Fizames C, Samson D, Drouot N, Vignal A, MillasseauP, Marc S, Hazan J, Seboun E, Lathrop M, Gyapay G, Morissette J,

1066 P. Langridge et al.

Weissenbach J (1996) A comprehensive genetic map of the humangenome based on 5,264 microsatellites. Nature 380, 152–154.

Donini P, Stephenson P, Bryan GJ, Koebner RMD (1998) The potentialof microsatellites for high throughput genetic diversity assessmentin wheat and barley. Genetic Resources and Crop Evolution 45,415–421.

D’Ovidio R, Anderson OD (1994) PCR analysis to distinguish betweenalleles of a member of a multigene family correlated with wheatbread-making quality. Theoretical and Applied Genetics 88,759–763.

Drenkard E, Richter BG, Rozen S, Stutius LM, Angell NA, MindrinosM, Cho RJ, Oefner PJ, Davis RW, Ausubel FM (2000) A simpleprocedure for the analysis of single nucleotide polymorphismsfacilitates map-based cloning in Arabidopsis. Plant Physiology 124,1483–1492.

Du C, Hart GE (1998) Triticum turgidum L. 6A and 6B recombinantsubstitution lines: extended maps and characterization of residualbackground alien genetic variation. Theoretical and AppliedGenetics 96, 645–653.

Dubcovsky J, Luo MC, Dvorak J (1995) Differentiation betweenhomoeologous chromosomes 1A of wheat and 1A(M) of Triticummonococcum and its recognition by the wheat ph1 locus.Proceedings of the National Acadamy of Science USA 92,6645–6649.

Dubcovsky J, Luo MC, Zhong GY, Bransteitter R, Desai A, Kilian A,Kleinhofs A, Dvorak J (1996) Genetic map of diploid wheat,Triticum monococcum L., and its comparison with maps ofHordeum Vulgare L. Genetics 143, 983–999.

Durrett R, Limic V (2001) On the quantity and quality of singlenucleotide polymorphisms in the human genome. StochasticProcesses and Their Applications 93, 1–24.

Dyck PL (1987) The association of a gene for leaf rust resistance withthe chromosome 7D suppressor of stem rust resistance in commonwheat. Genome 29, 467–469.

Dyck PL, Kerber ER, Aung T (1994) An interchromosomal reciprocaltranslocation in wheat involving leaf rust resistance gene Lr34.Genome 37, 556–559.

Eastwood RF, Lagudah ES, Appels R (1994) A direct search for DNAsequences tightly linked to cereal cyst nematode resistance genes inTriticum tauschii. Genome 37, 311–319.

Eckermann PJ, Verbyla AP, Cullis BR, Thompson R (2001) Theanalysis of quantitative traits in wheat mapping populations.Australian Journal of Agricultural Research 52, 1195–1206.

Effertz RJ, Anderson JA, Francl LJ (1998) QTLs associated withresistance to chlorosis induction by Pyrenophora tritici-repentis inadult wheat. Canadian Journal of Plant Pathology-RevueCanadienne de Phytopathologie 20, 438–439.

Ellis JG, Lawrence GJ, Luck JE, Dodds PN (1999) Identification ofregions in alleles of the flax rust resistance gene L that determinedifferences in gene-for-gene specificity. Plant Cell 11, 495–506.

Elsner TI, Albertson H, Gerken SC, Cartwright P, White R (1995)Breakpoint analysis—precise localization of genetic markers bymeans of nonstatistical computation using relatively few genotypes.American Journal of Human Genetics 56, 500–507.

Endo TR, Gill BS (1996) The deletion stocks of common wheat.Journal of Heredity 87, 295–307.

Erpelding JE, Blake NK, Blake TK, Talbert LE (1996) Transfer ofsequence tagged site PCR markers between wheat and barley.Genome 39, 802–810.

Fahima T, Roder MS, Grama A, Nevo E (1998) Microsatellite DNApolymorphism divergence in Triticum dicoccoides accessionshighly resistant to yellow rust. Theoretical and Applied Genetics 96,187–195.

Faris JD, Haen KM, Gill BS (2000) Saturation mapping of a gene-richrecombination hotspot region in wheat. Genetics 154, 823–835.

Faris JD, Li WL, Chen PD, Gill BS (1999) Candidate gene analysis ofquantitative resistance in wheat. Theoretical and Applied Genetics98, 219–225.

Feuillet C, Schachermayr G, Keller B (1997) Molecular cloning of anew receptor-like kinase encoded at the Lr10 leaf rust resistancelocus in wheat. Plant Journal 11, 45–52.

Frei OM, Stuber CW, Goodman MM (1986) Use of allozymes asgenetic markers for predicting genetic performance in maize singlecross hybrids. Crop Science 26, 37–42.

Gale MD, Atkinson MD, Chinoy CN, Harcourt RL, Jia J, Li QY, DevosKM (1995) Genetic maps of hexaploid wheat. In ‘Proceedings ofthe 8th International Wheat Genetics Symposium, China’. (Eds ZSLi, ZY Xin) pp. 29–40. (Agric Scientech Press: Beijing) Genome37, 871–875.

Giesendorf BAJ, Vet JAM, Tyagi S, Mensink E, Trijbels FJM, Blom HJ(1998) Molecular beacons: a new approach for semiautomatedmutation analysis. Clinical Chemistry 44, 482–486.

Gill KS, Gill BS, Endo TR, Boyko EV (1996) Identification and high-density mapping of gene-rich regions in chromosome group 5 ofwheat. Genetics 143, 1001–1012.

Grafius JE (1956) Components of yield in oat: a geometricalinterpretation. Agronomy Journal 48, 419–423.

Griffin TJ, Smith LM (2000) Single-nucleotide polymorphism analysisby MALDI-TOF mass spectrometry [Review]. Trends inBiotechnology 18, 77–84.

Gross E, Arnold N, Goette J, Schwarz-Boeger U, Kiechle M (1999) Acomparison of BRCA1 mutation analysis by direct sequencing,SSCP and DHPLC. Human Genetics 105, 72–78.

Hamer RJ, Weegels WP, Marseille JP (1992) Prediction of bread-making quality of wheat: the use of HMW glutenin A subunit basedquality scoring systems. Journal of Cereal Science 15, 91–102.

Hart GE (1994) RFLP maps of bread wheat. In ‘DNA-based markers inplants’. (Eds RL Phillips, JK Vasil) pp. 327–358. (Kluwer:Dordrecht)

Hart L, Mohler V, Zeller FJ, Hsam SLK, Schweizer G (1999)Identification of AFLP markers closely linked to the powderymildew resistance genes Pm1c and Pm4a in common wheat(Triticum aestivum L.). Genome 42, 322–329.

Helguera M, Khan IA, Dubcovsky J (2000) Development of PCRmarkers for wheat leaf rust resistance gene Lr47. Theoretical andApplied Genetics 101, 625–631.

Hirschhorn JN, Sklar P, Lindblad-Toh K, Lim YM, Ruiz-Gutierrez M,Bolk S, Langhorst B, Schaffner S, Winchester E, Lander ES (2000)SBE-TAGS: an array-based method for efficient single-nucleotidepolymorphism genotyping. Proceeding of the National Academy ofSciences, USA 97, 12164–12169.

Hohmann U, Endo TR, Gill KS, Gill BS (1994) Comparison of geneticand physical maps of group 7 chromosomes from Triticum aestivumL. Molecular and General Genetics 245, 644–653.

Hulbert SH (1997) Structure and evolution of the Rpl complexconferring rust resistance in maize. Annual Review ofPhytopathology 35, 293–310.

Jarne P, Lagoda PJL (1996) Microsatellites, from molecules topopulations and back. Trends in Ecology & Evolution 11, 424–429.

Jefferies SP, Pallotta MA, Paull JG, Karakousis A, Kretschmer JM,Manning S, Islam AKMR, Langridge P, Chalmers KJ (2000)Mapping and validation of chromosome regions conferring borontoxicity tolerance in wheat (Triticum aestivum). Theoretical andApplied Genetics 101, 767–777.

Ji L-H, Langridge P (1994) An early meiosis cDNA clone from wheat.Molecular and General Genetics 243, 17–23.

Trends in genetic and genome analyses in wheat 1067

Jones DA, Jones JDG (1997) The role of leucine-rich repeat proteins inplant defenses. Advances in Botanical Research 24, 89–167.

Joppa LP, Du C, Hart GE, Hareland GA (1997) Mapping gene(s) forgrain protein in tetraploid wheat (Triticum turgidum L.) using apopulation of recombinant inbred chromosome lines. Crop Science37, 1486–1589.

Keller M, Keller B, Schachermayr G, Winzeler M, Schmid JE, Stamp P,Messmer MM (1999a) Quantitative trait loci for resistance againstpowdery mildew in a segregating wheat × spelt population.Theoretical and Applied Genetics 98, 903–912.

Keller M, Karutz C, Schmid JE, Stamp P, Winzeler M, Keller B,Messmer MM (1999b) Quantitative trait loci for lodging resistancein a segregating wheat × spelt population. Theoretical and AppliedGenetics 98, 1171–1182.

Khan AA, Bergstrom GC, Nelson JC, Sorrells ME (2000)Identification of RFLP markers for resistance to wheat spindlestreak mosaic bymovirus (WSSMV) disease. Genome 43, 477–482.

Khan IA, Procunier JD, Humphreys DG, Tranquilli G, Schlatter AR,Marcucci-Poltri S, Frohberg RC, Dubcovsky J (2000) Developmentof PCR-based markers for a high grain protein content gene fromTriticum turgidum ssp. dicoccoides transferred to bread wheat. CropScience 40, 528–524.

Kojima T, Nagaoka T, Noda K, Ogihara Y (1998) Genetic linkage mapof ISSR and RAPD markers in einkorn wheat in relation to that ofRLFP markers. Theoretical and Applied Genetics 96, 37–45.

Korzun V, Borner A, Worland AJ, Law CN, Röder MS (1997)Application of microsatellite markers to distinguish inter-varietalchromosome substitution lines of wheat (Triticum aestivum L.).Euphytica 95, 149–155.

Korzun V, Roder MS, Wendehake K, Pasqualone A, Lotti C, Ganal MW,Blanco A (1999) Integration of dinucleotide microsatellites fromhexaploid bread wheat into a genetic linkage map of durum wheat.Theoretical & Applied Genetics 98, 1202–1207.

Lagudah ES, Dubcovsky J, Powell W (2001) Wheat genomics. PlantPhysiology and Biochemistry 39, 335–344.

Lagudah ES, Moullet O, Appels R (1997) Map based cloning of a genesequence encoding a nucleotide binding domain and leucine richregion at the Cre3 nematode resistance locus of wheat. Genome 40,659–665.

Lagudah ES, Moullet O, Ogbonnaya F, Seah S, Eastwood R, Appels R,Jahier J, Lopez-Brana, Delibes A (1998) Cloning of diseaseresistance gene sequences at loci conferring cyst nematoderesistance genes in wheat. In ‘Proceedings of the 9th InternationalWheat Genetics Symposium, Saskatoon, Saskachetwan, Canada’.(Ed. AE Slinkard) pp. 184–186.

Leister D, Kurth J, Laurie DA, Yano M, Sasaki T, Devos K, Graner A,Schulze-Lefert P (1998) Rapid reorganisation of resistance genehomologues in cereal genomes. Proceedings of the NationalAcademy of Science USA 95, 370–375.

Lillemo M, Morris CF (2000) A leucine to proline mutation inpuroindoline b is frequently present in hard wheats from NorthernEurope. Theoretical and Applied Genetics 100, 1100–1107.

Liu BH (1998) ‘Statistical genomics.’ (CRC Press: Boca Raton, FL)Liu YG, Tsunewaki K (1991) Restriction fragment length

polymorphism (RFLP) analysis in wheat. II. Linkage maps of theRFLP sites in common wheat. Japanese Journal of Genetics 66,617–633.

Liu ZW, Biyashev RM, Saghai Maroof MA (1996) Development ofsimple sequence repeat DNA markers and their integration into abarley linkage map. Theoretical and Applied Genetics 93, 869–876.

Liu Z, Sun Q, Ni Z, Yang T (1999) Development of SCAR markerslinked to the Pm21 gene conferring resistance to powdery mildewin common wheat. Plant Breeding 118, 215–219.

Lotti C, Salvi S, Pasqualone A, Tuberosa R, Blanco A (2000)Integration of AFLP markers into an RFLP-based map of durumwheat. Plant Breeding 119, 393–401.

Lui YG, Mori N, Tsunewaki K (1990) Restriction fragment lengthpolymorphism (RFLP) analysis in wheat. I. Genomic DNA libraryconstruction and RFLP analysis in common wheat. JapaneseJournal of Genetics 65, 367–380.

Ma ZQ, Röder M, Sorrells ME (1996) Frequencies and sequencecharacteristics of di-, tri, and tetra- nucleotide microsatellites inwheat. Genome 39, 123–130.

MacGuire PE, Qualset CO (1997) Progress in the genome mapping ofwheat and related species. Report No. 18. (University of CaliforniaGenetic Resources Conservation Program: Davis, CA)

Malécot G (1948) ‘Les mathematiques de l’heredite.’ (Masson and Cie:Paris)

Manly KF, Meer JM, Cudmore RH Jr (2000) ‘User manual for MapManager QTX.’ (Roswell Park Cancer Institute) 161 pp.

Mares DJ (1995) Dormancy in white wheats: mechanisms and locationof genes. In ‘7th International Symposium for Pre-harvestSprouting in Cereals 1995, Osaka, Japan’. (Eds K Noda, DJ Mares)pp. 179–184.

Marino CL, Nelson JC, Lu YH, Sorrells ME, Leroy P, Tuleen NA,Lopes CR, Hart GE (1996) Molecular genetic maps of the group 6chromosomes of hexaploid wheat (Triticum aestivum L. em. Thell).Genome 39, 359–366.

Marras SAE, Kramer FR, Tyagi S (1999) Multiplex detection of single-nucleotide variations using molecular beacons. Genetic Analysis:Biomolecular Engineering 14, 151–156.

Martin JM, Frohberg RC, Morris CF, Talbert LE, Giroux MJ (2000)Milling and bread baking traits associated with puroindolinesequence type in hard red spring wheat. Crop Science 41, 228–234.

McIntosh RA, Hart GE, Devos KM, Gale MD, Rogers WJ (1998)Wheat gene catalogue. In ‘Proceedings of the 9th InternationalWheat Genetics Symposium’. Vol. 5. (Ed. AE Slinkard) (UniversityExtension Press, University of Saskatchewan: Saskatoon, Canada)

McIntosh RA, Wellings CR, Park RF (1995) ‘Wheat rusts: an atlas ofresistance genes.’ (CSIRO: Melbourne)

McLauchlan A, Ogbonnaya FC, Hollingsworth B, Carter M, Gale KR,Henry RJ, Holton TA, Morell MK, Rampling LR, Sharp PJ,Shariflou MR, Jones MGK, Appels R (2001) Development ofrobust PCR-based DNA markers for each homoeo-allele of granule-bound starch synthase and their application in wheat breedingprograms. Australian Journal of Agricultural Research 52,1409–1416.

Melchinger AE, Utz HF, Schon CC (1998) Quantitative trait locus(QTL) mapping using different testers and independent populationsamples in maize reveals low power of QTL detection and large biasin estimates of QTL effects. Genetics 149, 383–403.

Mesfin A, Frohberg RC, Anderson JA (1999) RFLP markers associatedwith high grain protein from Triticum turgidum L. var. dicoccoidesintrogressed into hard red spring wheat. Crop Science 39, 508–513.

Mesfin A, Frohberg RC, Khan K, Olson TC (2000) Increased grainprotein content and its association with agronomic and end-usequality in two hard red spring wheat populations derived fromTriticum turgidum L. var. dicoccoides. Euphytica 116, 237–242.

Messmer MM, Keller M, Zanetti S, Keller B (1999) Genetic linkagemap of a wheat × spelt cross. Theoretical and Applied Genetics 98,1163–1170.

Meyers BC, Dickerman AW, Michelmore RW, Sivaramakrishnan S,Sobral BW, Young ND (1999) Plant disease resistance genes encodemembers of an ancient and diverse protein family within thenucleotide-binding superfamily. Plant Journal 20, 317–332.

Michelmore R (2000) Genomic approaches to plant disease resistance.Current Opinion in Plant Biology 3, 125–131.

1068 P. Langridge et al.

Michelmore RW, Paran I, Kesseli RV (1991) Identification of markerslinked to disease-resistance genes by bulked segregant analysis: arapid method to detect markers in specific genomic regions by usingsegregating populations. Proceedings of the National Academy ofScience USA 88, 9828–9832.

Milligan SB, Bodeau J, Yaghoobi J, Kaloshian I, Zabel P, WilliamsonVM (1998) The root knot nematode resistance gene Mi from tomatois a member of the leucine zipper, nucleotide binding, leucine-richrepeat family of plant genes. Plant Cell 10, 1307–1319.

Moore G, Foote T, Helentjaris T, Devos K, Kurata N, Gale M (1995)Was there a single ancestral cereal chromosome [Letter]. Trends inGenetics 11, 81–82.

Morin PA, Saiz R, Monjazeb A (1999) High-throughput singlenucleotide polymorphism genotyping by fluorescent 5′ exonucleaseassay. Biotechniques 27, 538–552.

Nagaoka T, Ogihara Y (1997) Applicability of inter-simple sequencerepeat polymorphisms in wheat for use as DNA markers incomparison to RLFP and RAPD markers. Theoretical and AppliedGenetics 94, 597–602.

Neff MM, Neff JD, Chory J, Pepper AE (1998) dCAPS, a simpletechnique for the genetic analysis of single nucleotidepolymorphisms: experimental applications in Arabidopsis thalianagenetics. Plant Journal 14, 387–392.

Nelson JC, Singh RP, Autrique JE, Sorrells ME (1997) Mapping genesconferring and suppressing leaf rust resistance in wheat. CropScience 37, 1928–1935.

Nelson JC, Sorrells ME, Vandeynze AE, Lu YH, Atkinson M, BernardM, Leroy P, Faris JD, Anderson JA (1995a) Molecular mapping ofwheat—major genes and rearrangements in homoeologous groups4, 5, and 7. Genetics 141, 721–731.

Nelson JC, Vandeynze AE, Autrique E, Sorrells ME, Lu YH, MerlinoM, Atkinson M, Leroy P (1995b) Molecular mapping ofwheat—homoeologous group 2. Genome 38, 516–524.

Nelson JC, Vandeynze AE, Autrique E, Sorrells ME, Lu YH, Negre S,Bernard M, Leroy P (1995c) Molecular mapping ofwheat—homoeologous group 3. Genome 38, 525–533.

Ogbonnaya FC, Seah S, Delibes A, Jahier J, Lopez-Brana I, EastwoodRF, Lagudah ES (2001) Molecular-genetic characterisation of a newnematode resistance gene in wheat. Theoretical and AppliedGenetics 102, 623–629.

Parker GD, Chalmers KJ, Rathjen AJ, Langridge P (1998) Mapping lociassociated with flour colour in wheat (Triticum aestivum L.)Theoretical and Applied Genetics 97, 238–245.

Parker GD, Chalmers KJ, Rathjen AJ, Langridge P (1999) Mapping lociassociated with milling yield in wheat (Triticum aestivum L.).Molecular Breeding 5, 561–568.

Parker GD, Fox PN, Langridge P, Whan B, Ganter P (2001) Analysingthe genetic diversity within Australian wheat breeding programsbased on molecular and pedigree data using 101 cultivars.Euphytica (in press).

Parker GD, Langridge P (2000) Development of a STS marker linked toa major locus controlling flour colour in wheat (Triticum aestivumL.). Molecular Breeding 6, 169–174.

Paull JG, Chalmers KJ, Karakousis A, Kretschmer JM, Manning S,Langridge P (1998) Genetic diversity in Australian wheat varietiesand breeding material based on RFLP data. Theoretical and AppliedGenetics 96, 435–446.

Paull JG, Pallotta MA, Langridge P, The TT (1994) RFLP markersassociated with Sr22 and recombination between chromosome 7Aof bread wheat and the diploid species Triticum boeoticum.Theoretical and Applied Genetics 89, 1039–1045.

Payne PI, Corfield KG, Blackman JA (1979) Identification of a highmolecular weight subunit of glutenin whose presence correlatedwith breadmaking quality in wheats of related pedigree. Theoreticaland Applied Genetics 55, 153–159.

Peng JH, Fahima T, Röder MS, Li YC, Grama A, Nevo E (2000)Microsatellite high-density mapping of the stripe rust resistancegene YrH52 region on chromosome 1B and evaluation of its marker-assisted selection in the F-2 generation in wild emmer wheat. NewPhytologist 146, 141–154.

Peng J, Korol AB, Fahima T, Roder MS, Ronin YI, Li YC, Nevo E(2000) Molecular genetic maps in wild emmer wheat, Triticumdicoccoides: genome-wide coverage, massive negative interference,and putative quasi-linkage. Genome Research 10, 1509–1531.

Perretant MR, Cadalen T, Charmet G, Sourdille P, Nicolas P, Boeuf C,Tixier MH, Branlard G, Bernard S, Bernard M (2000) QTL analysisof bread-making quality in wheat using a doubled haploidpopulation. Theoretical and Applied Genetics 100, 1167–1175.

Pestsova E, Ganal MW, Röder MS (2000a) Isolation and mapping ofmicrosatellite markers specific for the D genome of bread wheat.Genome 43, 689–697.

Pestsova E, Korzun V, Goncharov NP, Hammer K, Ganal MW, RöderMS (2000b) Microsatellite analysis of Aegilops tauschiigermplasm. Theoretical and Applied Genetics 101, 100–106.

Piatek AS, Tyagi S, Pol AC, Telenti A, Miller LP, Kramer FR, Alland D(1998). Molecular beacon sequence analysis for detecting drugresistance in Mycobacterium tuberculosis. Nature Biotechnology16, 359–363.

Plaschke J, Ganal MW, Röder MS (1995) Detection of genetic diversityin closely related bread wheat using microsatellite markers.Theoretical and Applied Genetics 91, 1001–1007.

Powell W, Machray GC, Provan J (1996a) Polymorphism revealed bysimple sequence repeats. Trends in Plant Science 1, 215–222.

Powell W, Morgante M, Andre C, Hanafey M, Vogel J, Tingey S,Rafalski A (1996b) The comparison of RFLP, RAPD, AFLP andSSR (microsatellite) markers for germplasm analysis. MolecularBreeding 2, 225–238.

Prasad M, Varshney RK, Kumar A, Balyan HS, Sharma PC, EdwardsKJ, Singh H, Dhaliwal HS, Roy JK, Gupta PK (1999) Amicrosatellite marker associated with a QTL for grain proteincontent on chromosome arm 2DL of bread wheat. Theoretical andApplied Genetics 99, 341–345.

Prasad M, Varshney RK, Roy JK, Balyan HS, Gupta PK (2000) The useof microsatellites for detecting DNA polymorphism, genotypeidentification and genetic diversity in wheat. Theoretical andApplied Genetics 100, 584–592.

Qi X, Stam P, Lindhout P (1998) Use of locus-specific markers toconstruct a high density map in barley. Theoretical and AppliedGenetics 96, 376–384.

Robert O, Abelard C, Dedryver F (1999) Identification of molecularmarkers for the detection of the yellow rust resistance gene Yr17 inwheat. Molecular Breeding 5, 167–175.

Röder MS, Korzun V, Gill BS, Ganal MW (1998b) The physicalmapping of microsatellite markers in wheat. Genome 41, 278–283.

Röder MS, Korzun V, Wendehake K, Plaschke J, Tixier M-H, Leroy P,Ganal MW (1998a) A microsatellite map of wheat. Genetics 149,2007–2023.

Röder MS, Plaschke J, Konig SU, Borner A, Sorrells ME, Tanksley SD,Ganal MW (1995) Abundance, variability and chromosomallocation of microsatellites in wheat. Molecular & General Genetics246, 327–333.

Rogowsky P, Guidet FLY, Langridge P, Shepherd KW, Koebner RMD(1991) Isolation and characterisation of wheat–rye recombinantsinvolving chromosome arm 1DS of wheat. Theoretical and AppliedGenetics 82, 537–544.

Ronald P (1998) Resistance gene evolution. Current Opinion in PlantBiology 1, 294–298.

Rong JK, Millet E, Manisterski J, Feldman M (2000) A new powderymildew resistance gene: introgression from wild emmer into

Trends in genetic and genome analyses in wheat 1069

common wheat and RFLP-based mapping. Euphytica 115,121–126.

Rossi M, Goggin FL, Milligan SB, Kaloshian I, Ulman DE, WilliamsonVM (1998) The nematode resistance gene Mi of tomato confersresistance against the potato aphid. Proceedings of the NationalAcademy of Science USA 95, 9750–9754.

Roy JK, Prasad M, Varshney RK, Balyan HS, Blake TK, Dhaliwal HS,Singh H, Edwards KJ, Gupta PK (1999) Identification of amicrosatellite on chromosomes 6B and a STS on 7D of bread wheatshowing an association with preharvest sprouting tolerance.Theoretical and Applied Genetics 99, 336–340.

Sarma RN, Fish L, Gill BS, Snape JW (2000) Physical characterizationof the homoeologous group 5 chromosomes of wheat in terms ofrice linkage blocks, and physical mapping of some important genes.Genome 43, 191–198.

Seah S, Bariana H, Jahier J, Sivasithamparam K, Lagudah ES (2001)The introgressed segment carrying rust resistance genes Yr17, Lr37and Sr38 in wheat can be assayed by a cloned disease resistancegene-like sequence. Theoretical and Applied Genetics 102,600–605.

Seah S, Spielmeyer W, Jahier J, Sivasithamparam K, Lagudah ES(2000) Resistance gene analogues within an introgressedchromosomal segment derived from Triticum ventricosum thatconfers resistance to nematode and rust pathogens in wheat.Molecular Plant-Microbe Interaction 13, 334–341.

Sears ER (1954) The aneuploids of common wheat. Research Bulletin572, University of Missouri College of Agriculture AgricultureExperimental Station. pp. 3–58.

Seyfarth R, Feuillet C, Schachermayr G, Winzeler M, Keller B (1999)Development of a molecular marker for the adult plant leaf rustresistance gene Lr35 in wheat. Theoretical and Applied Genetics 99,554–560.

Shah MM, Yen Y, Gill KS, Baenziger PS (2000) Comparisons of RFLPand PCR-based markers to detect polymorphism between wheatcultivars. Euphytica 114, 135–142.

Shamsuddin AKM (1985) Genetic diversity in reaction to heterosis andcombining ability in spring wheat. Theoretical and AppliedGenetics 70, 306–308.

Shariflou MR, Sharp PJ (1999) A polymorphic microsatellite in the 3′end of ‘waxy’ genes of wheat, Triticum aestivum. Plant Breeding118, 275–277.

Simonetti MC, Bellomo MP, Laghetti G, Perrino P, Simeone R, BlancoA (1999) Quantitative trait loci influencing free-threshing habit intetraploid wheats. Genetic Resources & Crop Evolution 46,267–271.

Singh RP (1992) Genetic association between gene Lr34 for leaf rustresistance and leaf tip necrosis in bread wheats. Crop Science 32,874–878.

Singh RP, Nelson JC, Sorrells ME (2000) Mapping Yr28 and othergenes for resistance to stripe rust in wheat. Crop Science 40,1148–1155.

Smith JSC (1984) Genetic variability within US hybrid maize:multivariate analysis of isozyme data. Crop Science 24, 1041–1046.

Smith JSC, Chin ECL, Shu H, Smith OS, Wall SJ, Senior ML, MitchellSE, Kresovich S, Ziegle J (1997) An evaluation of the utility of SSRloci as molecular markers in maize (Zea mays L.): comparisons withdata from RFLPs and pedigree. Theoretical and Applied Genetics95, 163–173.

Smith JSC, Smith OS, Bowen SL, Tenborg RA, Wall SJ (1991) Thedescription and assessment of distances between inbred lines ofmaize: III A revised scheme for the testing of distinctivenessbetween inbred lines utilizing DNA RFLPs. Maydica 36, 213–226.

Sourdille P, Snape JW, Cadalen T, Charmet G, Nakata N, Bernard S,Bernard M (2000a) Detection of QTLs for heading time and

photoperiod response in wheat using a doubled-haploid population.Genome 43, 487–494.

Sourdille P, Tixier MH, Charmet G, Gay G, Cadalen T, Bernard S,Bernard M (2000b) Location of genes involved in ear compactnessin wheat (Triticum aestivum) by means of molecular markers.Molecular Breeding 6, 247–255.

Spielmeyer W, Huang L, Bariana H, Laroche A, Gill BS, Lagudah ES(2000) NBS-LRR sequence is associated with leaf and stripe rustresistance on the end of homoeologous chromosome group 1S ofwheat. Theoretical and Applied Genetics 101, 1139–1144.

Spielmeyer W, Robertson M, Collins N, Leister D, Schulze-Lefert P,Seah S, Moullet O, Lagudah ES (1998) A superfamily of diseaseresistance gene analogues is located on all homoeologouschromosome groups of wheat (Triticum aestivum). Genome 41,782–788.

Stephenson P, Bryan G, Kirby J, Collins A, Devos K, Busso C, Gale M(1998) Fifty new microsatellite loci for the wheat genetic map.Theoretical and Applied Genetics 97, 946–949.

Tanksley SD, Nelson JC (1996) Advanced backcross QTL analysis—amethod for the simultaneous discovery and transfer of valuableQTLs from unadapted germplasm into elite breeding linesTheoretical and Applied Genetics 92, 191–203.

Tixier MH, Sourdille P, Charmet G, Gay G, Jaby C, Cadalen T, BernardS, Nicolas P, Bernard M (1998) Detection of QTLs for crossabilityin wheat using a doubled haploid population. Theoretical andApplied Genetics 97, 1076–1082.

Turnbull K, Gaborit T, Marion D, Rahman S (2000) Variation inpuroindoline polypeptides in Australian wheat cultivars inrelationship to grain hardness. Australian Journal of PlantPhysiology 27, 143–158.

Vagujfalvi A, Crosatti C, Galiba G, Dubcovsky J, Cattivelli L (2000)Two loci on wheat chromosome 5A regulate the differential cold-dependent expression of the cor14b gene in frost-tolerant and frost-sensitive genotypes. Molecular and General Genetics 263,194–200.

Van der Vossen EAG, Rouppe van der Voort J, Kanyuka K,Bendahmane A, Sandbrink H, Baulcombe DC, Bakker J, StiekemaWJ, Klein-Lankhorst RM (2000) Homologues of a singleresistance-gene cluster in potato confer resistance to distinctpathogens: a virus and a nematode. Plant Journal 23, 567–576.

Vandeynze AE, Dubcovsky J, Gill KS, Nelson JC, Sorrells ME, DvorakJ, Gill BS, Lagudah ES, McCouch SR, Appels R (1995) Molecular-genetic maps for group 1 chromosomes of triticeae species and theirrelation to chromosomes in rice and oat. Genome 38, 45–59.

Varshney RK, Prasad M, Roy JK, Harjit-Singh NK, Dhaliwal HS,Balyan HS, Gupta PK (2000) Identification of eight chromosomesand a microsatellite marker on 1AS associated with QTL for grainweight in bread wheat. Theoretical and Applied Genetics 100,1290–1294.

Venter E, Botha AM (2000) Development of markers linked toDiuraphis noxia resistance in wheat using a novel PCR-RFLPapproach. Theoretical and Applied Genetics 100, 965–970.

Vos P, Hogers R, Bleeker M, Reijans M, van de Lee T, Hornes M,Freiters A, Pot J, Peleman J, Kuiper M, Zabeau M (1995) AFLP: anew concept for DNA fingerprinting. Nucleic Acids Research 23,4407–4414.

Vuylsteke M, Mank R, Antonise R, Bastiaans E, Senior ML, StuberCW, Melchinger AE, Lübberstedt T, Xia XC, Stam P, Zabeau M,Kuiper (1999) Two high density AFLP® linkage maps of Zea maysL.: analysis of distribution of AFLP markers. Theoretical andApplied Genetics 99, 921–935.

Waldron BL, Moreno-Sevilla B, Anderson JA, Stack RW, Frohberg RC(1999) RFLP mapping of QTL for fusarium head blight resistancein wheat. Crop Science 39, 805–811.

1070 P. Langridge et al.

Wei F, Gobelman-Werner K, Morroll SM, Kurth J, Mao L, Wing R,Leister D, Schulze-Lefert P, Wise RP (1999) The Mla (powderymildery) resistance cluster is associated with three NBS-LRR genefamilies and suppressed recombination within a 240kb DNAinterval on chromosome 5S (1HS) of barley. Genetics 153,1929–1948.

Weiss KM (1998) In search of human variation. Genome Research 8,691–697.

Weng Y, Tuleen NA, Hart GE (2000) Extended physical maps and theconsensus physical map of the homoeologous group-6chromosomes of wheat (Triticum aestivum L, em Thell.).Theoretical and Applied Genetics 100, 519–527.

Whitehouse RNH (1969) An application of canonical analysis to plantbreeding. Genetica Agraria 23, 61–69.

William HM, Hoisington D, Singh RP, Gonzalez de Leon D (1997)Detection of quantitative trait loci associated with leaf rustresistance in bread wheat. Genome 40, 253–260.

Xie DX, Devos KM, Moore G, Gale MD (1993) RFLP-based geneticmaps of the homoeologous group 5-chromosomes of bread wheat(Triticum-Aestivum L.). Theoretical and Applied Genetics 87,70–74.

Zabeau M, Vos P (1993) Selective restriction fragment amplification: ageneral method for DNA fingerprinting. European PatentApplication Number 92402629.7, Publication Number 0534,858A1.

Zanetti S, Winzeler M, Keller M, Keller B, Messmer M (2000) Geneticanalysis of pre-harvest sprouting resistance in a wheat × spelt cross.Crop Science 40, 1406–1417.

Zhang XQ, Ross K, Gustafson JP (2000) Physical location ofhomoeologous group 5 and 6 molecular markers mapped inTriticum aestivum L. Journal of Heredity 91, 441–445.

Zhao XC, Batey IL, Sharp PJ, Crosbie G, Barclay I, Wilson R, MorellMK, Appels R (1998) A single genetic locus associated with starchgranule properties and noodle quality in wheat. Journal of CerealScience 27, 7–13.

Zhen Z, Mares D (1992) A simple extraction and one-step SDS-PAGEsystem for separating HMW and LMW glutenin subunits of wheatand high molecular weight proteins of rye. Journal of CerealScience 15, 63–78.

Manuscript received 5 April 2001, accepted 17 July 2001

text). The genetic maps were then aligned with thephysical maps (shown as standard C-banded karyotypeson the left of the figure) using published maps wheremarkers have been assigned to regions of chromosomesusing the deletion stocks established by Endo and Gill(1996), as discussed in the text. The approximate locationsare indicated by vertical black lines next to the geneticmap and C-banded karyotype (joined by a gray line).Details of the section defined by the respective deletionsin which the probes are found are in Boyko et al. (1999),Faris et al. (2000), Sarma et al. (2000), Weng et al. (2000),and Zhang et al. (2000). The positions of the centromerein the genetic maps were defined as the region betweenmarkers located on the short and long arms of therespective chromosomes, using the ITMI database ofmarkers analysed using the genetic stocks of wheat(MacGuire and Qualset 1997).

Appendix 1.

Genetic linkage map of the wheat cross W7984 synthetichexaploid wheat × Opata 85. The map was generated usingdata at http://wheat.pw.usda.gov/. The distance betweenmarkers is shown on the left of each map in centiMorgans andthe marker locus is given on the right. Mapping data for theSynthetic × Opata cross were down-loaded from theGraingenes web page (http://wheat.pw.usda.gov/) andimported into Map Manager (version QTb27ppc1). The‘gwm’ microsatellites included in the map were mainly thosethat were mapped with a LOD >2.5 by Röder et al. (1998b).The validity of published maps was then checked by runningthe ‘rearrange’ command in Map Manager and checking theorders of markers in the 3 possible maps generated by thisoption. Generally, the shortest map fitted the published mapsbest (except for chromosomes 1B and 4B, as discussed in the

Trends in genetic and genome analyses in wheat 1071

0 cM

100

200

300

400

500

1A 1B 1D

XGli1-1a

XksuD14.2

Xcdo426Xabc156

XGli3XksuE18Xwmc104XksuE18D

Xwmc24Xbcd98-.1

XksuG9

Xmwg67Xcdo580

Xrz244

Xabg452, Xwg605Xgwm164, Xcdo98

Xbcd1072Xabg373Xwmc120Xcdo473Xbcd1407

Xcdo312Xwmc93Glu-A1XksuE3Xbcd808.2Xbcd1930

Xbcd808.1, Xbcd1889

Xcmwg733Xbcd265

Xcmwg706XksuH9

XksuH14

Xcdo1160Xmwg632Xmwg912XksuE11.2Xcdo393Xwg241XksuE11.1

gammaGli

Xwhs179XksuD14XGli1Xfba311

Xwmc85

Xmwg938Xcdo388

Xmwg68

Xabc156, XGli3XksuF43.2

XksuF43.1XksuE18

Xwmc230

Xwmc51

Xfbb67cXksuE19

XksuE18HcXwmc52Xcdo92, Xbcd338, Xcdo98Xcdo618, Xbcd200,Xwg811, Xbcd762Xbcd1796, Xcdo1173, Nor-B1Xbcd340Xbcd1124Xrz166Xcdo1340, Xcdo127

Xabg373,Xmwg837.2, Xwg605Xcmwg645

Xbcd12, Xcdo278Xbcd1449Xbcd386Xfbb35

Xcmwg758Xglk136Xcdo637Xwmc156Xbcd1150.1

Glu-B1Xbcd1150.2

Xcmwg733Xbcd442Xcdo1196Xbcd441, Xbcd304XksuG34, Xmwg69XksuE11.1Xbcd508Xbcd1562Xbcd1514, Xbcd1261, Xcdo1189Xwmc44Xcdo346Xwmc113cXksuE11.2Xmwg912XksuI27

XGli1

Xfbb250b

Xbcd1434XksuD14.2

Xwhs179.1

XksuD14.1Xmwg938Xwhs179.2

Xcnl7Xmwg837Xwmc222aXabc156Xfbb237d

Xfba250Xmwg967

Xcdo312

Xfbb194a

Xwmc216Glu-D1XksuD49, XksuG55XksuG2

Xcmwg733Xbcd1930

Xcmwg701XAdh

XATPase

Xcmwg693Xbcd1261

Xbcd508XksuE11XksuI27.1Xwg241

Xgwm136

Xgwm33.1

Xgwm357Xgwm498

Xgwm131

Xgwm403

Xgwm33.3

Xgwm337

Fig. A1.1

1072 P. Langridge et al.

2D

Xfba83XW2IXwmc111Xwmc112b, Xwmc25b

Xfba400, Xfba88

Xbcd102Xbcd718Xbcd1970Xbcd18Xcdo456

Xbcd611Xcdo1379Xcdo1479Xbcd262Xfba272Xfbb279

Xfba341.1, Xcdo405Xwsu1

Xbcd260Xgwm30a

Xwmc18, Xgwm30b, Xfba38Xwmc245

Xbcd120Xbcd111

Xfba74, Xfbb99Xtam8-2DXfba111, Xfbb122, Xfba64Xfbb32, Xfba62, Xfba61,Xfba341.2, Xfbb284Xfbb68

Xcdo1008Xfba116

Xfbb9Xfbb251Xfbb72XksuH9-2D

XksuD23

XksuH16

Xbcd410Xfba311

Xwmc167, Xrz444

Xfbb377Xcdo36

Xfba314Xfba209.1

Xcmwg682Xfdba349Xfbb189b

Xgwm296Xgwm261

Xgwm102

Xgwm515

Xglk558

Xfbb274Xfba280Xwmc154, Xwmc83bXwmc25aXfbb121, Xfbb40, Xfba4Xbcd18Xrz69Xrz444

Xfba272

Xfbb62, Xfba29Xbcd1184

Xfba106

Xmwg950

Xtam72Xwsu1

Xfbb47, Xbcd152Xfba374, Xfbb21Xfbb226, Xfbb4, Xfbb75Xcdo370, XksuF11, Xbcd260

Xwmc265

Xbcd1709Xwmc245

Xwmc272

Xbcd1119

Xwg996

Xcdo388

Xfba199, Xfbb335Xbcd445Xcdo684

Xbcd1779

Xcnl6-2BXwmc175Xmwg2025Xbcd307

Xfba62.2,Xfbb284, Xfba61, Xfba345Xfba276, Xbcd1095Xwsu2Xbcd135Xfba116Xmwg546Xfbb113Xcmwg660

Xfba359Xcdo678

Xgwm526, Xfba385Xglk558XksuD23Xcdo36Xbcd1231Xfba310

2A

Xbcd348.1Xcdo456.1

Xcmwg682, Xcdo447Xfba82, Xfba70.1Xcdo456.2Xbcd348.2

Xfba70.2Xfba280Xfbb289,Xfbb72,

Xbcd718, Xcdo57XksuD18

Xwmc177,Xfba178, Xfba198

Xfba300

Xfba272, Xfbb329Xbcd1184

Xfbb226

Xfba106Xcdo1281

Xbcd855, Xbcd161Xbcd152

Xrz395-2AXwsu1-2A

Xbcd1688, Xfbb359, Xcdo405

Xfbb212-2A

XksuG5Xfbb353Xwmc261Xbcd543

Xfba276Xbcd1095

Xfbb284

XksuE16XksuD22

Xpsr934

XksuF41Xcdo1410

Xcdo678

Xbcd292Xbcd410XksuH16XksuF11Xfba8

Xfba385Xfba314

Xcdo1376

0 cM

100

200

300

400

500

Xgwm636, Xgwm614Xgwm497

Xbcd1970, Xfbb61

Xgwm359Xgwm71.1

Xfba374, Xgwm71.2

Xgwm558

Xcdo405

Xtam18

Xgwm55.1

Xgwm16

2BFig. A1.2

Trends in genetic and genome analyses in wheat 1073

3B

Xbcd907Xfbb147.1Xfbb166Xfba190XksuG53Xgwm533.1Xgwm493Xfbb185Xcdo460Xtam47Xtam61Xgwm533.2Xfba91, Xwmc43b

Xabg471,Xcdo1164Xgwm264Xpsr689-3BXpsr903-3B

Xfbb117-3BXcnl8-3BXcdo583-3BXbcd1418-3BXbcd809-3BXATPase-3B.2Xbcd1380-3B.1,2XATPase-3B.1Xcdo328-3BXksuH7-3BXfbb156-3B

Xcdo718Xmwg818Xgwm112Xbcd1555, Xbcd141,Xbcd147, Xfbb147.2Xmwg69, Xfbb168Xfbb378Xabc174-3BXbcd131Xfbb274Xfba171Xfba133, Xfba360Xcdo105Xbg131

Xfbb316, Xfba217Xfbb293, Xfba310Xtam63Xmwg11

Xfba235

3A 3D

Xwmc11Xglk683, Xfbb370Xtam47Xtam61Xcdo460Xfba366Xfba91Xbcd1428

Xcdo638-3A

Xcdo395Xbcd1823,Xcdo1345

Xabg460Xcdo1164Xcdo1435Xmwg14Xmwg22, Xbcd1532Xbcd706-3A

XksuA6-3A

Xfbb332Xfba127Xbcd366XATPaseXcdo118Xmwg802,Xbcd828Xcdo1174,Xcdo54Xbcd1145Xfba175Xgwm5Xfbb271,Xtam33Xmwg961Xbcd2044, Xbcd115Xfbb353Xcdo281, Xbcd452XksuH2, Xwg177Xfbb277Xpsr903Xwmc264Xbcd372Xmwg30

Xgwm162Xfbb293.1Xbcd1773Xtam63Xabc172.1

Xabc172.2

Xfbb293.2Xmwg12Xfbb237, Xmwg11

Xcdo482Xbcd451

0 cM

100

200

300

400

500

Xabg471

Xfbb315, Xfbb142, Xfbb24

Xfba214,Xfba220, Xfba213,Xfbb348, Xfba242, Xfbb177

Xfba167, Xfba8

Xcdo549Xfbb370Xbcd907Xbcd1802, Xbcd1802Xgwm183Xabc171Xfba91Xfba241Xbcd1532Xcdo1435Xmwg22Xcdo407XksuA6Xwmc43c

Xbcd134Xmwg688Xfbb324, Xgwm52Xabc176Xcdo1406Xbcd288Xfbb237, Xfbb23Xfba330Xbcd372,Xbcd22Xgwm645Xbcd1555Xwg110Xbcd515Xgwm383XksuD19XksuH15Xbcd361XksuE14aXfba27Xfba310Xfbb117

Fig. A1.3

1074 P. Langridge et al.

4A 4B 4D

Xfbb1-4AXfbb227, Xfba320Xfbb176, Xfba359Xfba211, Xfba43Xfbb332, Xfba40Xfba78-4A

XksuF8aXwmc96aXbcd1652Xcdo1387, Xcdo959Xbcd1738Xrz574Xfba147Xbcd402Xbcd8XksuG12bXwg622Xfba65Xwmc258a,b, Xgwm637Xmwg549Xabg390Xpsr115Xbcd1670a, Xfba4Xmwg710Xcdo780Xcdo475XksuD9Xwmc262, Xbcd130Xfba231Xfbb154Xwmc232Xcdo665Xfbb114Xfba282Xtam72, Xcdo545Xbcd1975Xfba243Xfba253Xbcd129Xbcd588

Xwmc173aXwmc48bXbcd2026, Xfba147Xcsu25Xrz672Xcdo938, Xfba8Xcdo1128

Xfbb22Xbcd1250Xbcd749, Xfbb121Xwg110Xrz251Xcdo795Xbcd1051Xabg484Xbcd1265

Xbcd1262Xcdo1401Xmwg2025, Xfbb178, Xfba41XksuG10Xcnl7, XksuE2, Xcdo1312Xfbb58Xcdo20, Xfbb182Xwg114, Xcdo1528Xfbb67Xfbb255Xbcd402Xfba177

Xfba78

Xmwg634

Xcdo669Xbcd265Xfbb13, Xwmc48aXfbb336, Xfbb251Xbcd327

XksuF8Xgwm165Xbcd1117Xbcd15Xfbb178Xcdo1312Xbcd1431.1, Xcdo1081Xfbb226Xfba177Xbcd1431.2

0 cM

100

200

300

400

500

Fig. A1.4

Trends in genetic and genome analyses in wheat 1075

5D

0 cM

100

200

300

400

500

Xwmc233Xmta10Xfba393Xfba114Xfbb238Xfba137

XksuD30Xcmwg770Xmwg561Xcdo412Xcdo57Xbcd1874Xfbb156Xgwm583Xgwm174Xgwm182Xcdo1508

Xfbb26Xwmc215Xbcd450Xrz395Xfba209Xwmc97Xmwg900Xmwg922Xbcd1103Xcdo346

Xfbb213, Xfbb100

Xbcd1670

Xfba364Xbcd1421Xwmc161aXbcd197, Xcdo506Xbcd87

5A 5B

Xpsr170cXgbxR665Xgwm154wXgbxG625Xfba352Xgwm304Xfba131Xwmc96bXbcd1871Xcdo749

Xcdo785Xbcd1355Xbcd157, Xpsr128Xcdo412Xbcd926, Xbcd1949Xcdo57Xcdo1090aXbcd1088Xbcd981Xmwg522Xmwg624Xfbb2Xbcd1235.1Xfbb255Xfba166Xbcd183Xfba190Xrz395Xcdo1326Xfba68Xfbb199, Xfbb209Xfba351Xabg391Xcdo457, Xbcd1235.2Xcdo1312Xabg366Xfbb330Xfbb249Xcdo20, Xwg114Xcdo1528Xgwm126Xmwg2112

Xfbb277Xbcd873Xcdo959Xfbb121.1, Xfba367Xpsr170Dc, Xwmc149Xfba393Xfba342Xgwm66, Xwmc247Xbcd1871Xabg705Xpsr929, Xcdo749Xgwm540Xgwm544Xfba232

Xmwg561Xgwm213, Xgwm335Xpsr574Xpsr128Xbcd157Xwg909Xwg889Xabc164XksuA1Xfba127Xcdo412Xtam72Xfbb121.2Xgwm499Xbcd508Xmwg52, Xcdo1192Xabg473Xmwg914Xbcd307Xbcd9Xwg583, Xfba348Xfbb237Xfba166Xbcd1030, Xfba332Xbcd450, Xcdo504Xgwm408Xgwm604Xcdo1326, Xfbb328, Xcdo348Xfba351Xcdo584Xwmc235Xfbb323

Fig. A1.5

1076 P. Langridge et al.

6A 6B 6D

Xpsr167Xbcd342, Xpsr889Xbcd1821Xbcd21,Xcdo476Xfba65, Xfbb194Xfbb147, Xfbb209Xfba307Xpsr8(Cxp3)Xmwg67, XksuG48Xfbb166Xcmwg652Xpsr10(Gli-2)Xfba85, Xfba152XksuH4

Xpsr312Xfbb145Xfba148Xcdo1315Xcdo270

Xgwm494Xcdo1428, Xfba234, Xcdo29Xfba397, Xfbb95Xwmc256, Xtam36Xwmc163, Xpsr915Xpsr463(Prk)Xcdo772Xbcd758, Xbcd1860Xcdo204, Xfba367, Xfbb283Xfbb215, Xfbb192Xbcd506, Xfbb170Xcdo388Xwmc201Xmwg934Xcsb112(Dhn5)Xfbb164a, Xgwm570Xfba111Xgwm169Xfbb191, Xbcd1510,Xcdo836Xfba8,Xfbb40,Xfba20Xfbb221,Xfbb82,Xfbb70

Xcdo476Xpsr167Xbcd342Xrz995XksuH4Xfbb16Xfba152Xtam60Xwmc95Xfba357Nor-B2XksuH14Xfba67,Xfba399,Xfba328,Xfba345, Xfba344Xfba359, Xbcd1495Xglk479, Xpsr141,Xbcd1383Xgwm644Xgwm88, Xwmc105bXcnl3, Xbcd1299,Xcdo524, Xbcd102Xbcd1716, Xbcd357

Xbcd1407Xwmc182Xfbb364Xcdo507, Xbcd2014Xgwm626Xfbb164Xmwg934Xcdo341Xwg341Xfbb57, Xfbb169Xfba42XksuG30Xabg473Xfba111, Xfbb130Xgwm219Xfbb377Xfbb359, fbb307b,Xfbb327Xmwg74Xfbb82Xfbb59, Xfba251, Xfbb70Xmwg2053

Xpsr889Xbcd1821Xabg466Xbcd342Xfba307Xpsr964Xpsr8(Cxp3)Xgwm469XksuG48Xfba1Xfba187Xfbb354,Xfbb319,Xfba85Xfbb222Xpsr627Xpsr106Xmwg916Xfbb231.1,Xwmc113a,Xfbb283Xcdo270Xabc173Xfba336,Xbcd1398,Xcdo534Xtam31Xabg20Xbcd357, Xbcd1716Xabc175Xfba381Xfbb231.2Xfba81Xcmwg669Xtam28, Xcsb112(Dhn5)Xfbb169XksuE14Xbcd1319XksuD1Xcdo836Xbcd1510XksuD27Xfbb70, Xfbb59Xmwg2053c

XksuD27Xmwg2053Xmwg573

0 cM

100

200

300

400

500

Fig. A1.6

Trends in genetic and genome analyses in wheat 1077

http://www.publish.csiro.au/journals/ajar

Xfba42Xwmc76Xbcd385Xgwm68Xbcd1338Xfbb195Xfbb150Xcnl8Xbcd310Xwg180Xfbb226Xfba32Xgwm43, Xgwm297Xfbb193, Xfba371Xfba311, Xgwm16Xcdo551, Xbcd98Xbcd349, Xabc455Xbcd178Xrz476Xcnl7Xwg514Xfba354,Xfbb258Xfba301Xfbb179, Xfbb175

Xtam50Xabc310Xcdo686

Xfba259

Xfbb67XksuD2

Xcnl2

Xfbb189,Xfba21Xrz508Xwmc276Xcdo414

XksuE18

7A 7B 7D

Xfbb156Xfbb278, Xfba311, Xfba321Xfbb9, Xfba17, Xfba72Xfba93, Xfbb193, Xcdo545,Xbcd907-7AXfbb121-7A

Xfba127Xfba109Xfbb67, Xfbb186Xcdo475bXfba42Xcdo1395

Xabc158Xfba248

Xwmc83, Xwmc83c

Xfba340

Xfba354XksuG12

Xgwm276Xfba69Xfba97Xfbb18, Xfbb145Xrz682, Xmwg938Xfba382-7AXwmc116XksuD2Xfba204Xwg380Xfba350Xcdo962Xfba337, Xfbb218, Xfba234Xwmc17Xcdo17, Xfbb343Xcdo595, Xbcd1066Xfbb264

Xcdo347Xfba134Xfbb238XksuH9

Xwg686, Xfba305, Xfbb352

Xfba8

Xwg834

Xcnl6

Xbcd1438Xbcd1872

Xmwg710

Xrz2

Xgwm4, Xfba377

Xwmc121Xbcd707Xgmw111Xgwm437, Xwg719Xfbb112Xcdo775

Xwmc94

Xfba69Xfbb79Xfba264Xfbb325Xmwg975Xwg420Xfba204Xgwm428

0 cM

100

200

300

400

500

Fig. A1.7