Acid soils limit plant production worldwide because their high concentrations of soluble aluminium cations (Al3+) inhibit root growth. Major food crops such as wheat (Triticum aestivum L.) have evolved mechanisms to resist Al3+ toxicity, thus enabling wider distribution. The origins of Al3+ resistance in wheat are perplexing because all progenitors of this hexaploid species are reportedly sensitive to Al3+ stress. The large genotypic variation for Al3+ resistance in wheat is largely controlled by expression of an anion channel, TaALMT1, which releases malate anions from the root apices. A current hypothesis proposes that the malate anions protect this sensitive growth zone by binding to Al3+ in the apoplasm. We investigated the evolution of this trait in wheat, and demonstrated that it has multiple independent origins that enhance Al3+ resistance by increasing TaALMT1 expression. One origin is likely to be Aegilops tauschii while other origins occurred more recently from a series of cis mutations that have generated tandemly repeated elements in the TaALMT1 promoter. We generated transgenic plants to directly compare these promoter alleles and demonstrate that the tandemly repeated elements act to enhance gene expression. This study provides an example from higher eukaryotes in which perfect tandem repeats are linked with transcriptional regulation and phenotypic change in the context of evolutionary adaptation to a major abiotic stress.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Acid soils inhibit root growth and limit plant production worldwide. Forty per cent of total arable land is affected by acidity, and this is especially common in tropical and sub-tropical soils due to the higher rainfall (von Uexküll and Mutert, 1995). Consequently, acid soils threaten sustainable food production in many developing countries in Asia, Africa and Latin America. Plants growing on acid soils are exposed to many stresses, but inhibition of root elongation is primarily caused by the prevalence of soluble aluminium ions (Al3+), which damage the sensitive root apices (Taylor, 1988; Ryan et al., 1993; Horst, 1995; Sivaguru and Horst, 1998; Matsumoto, 2000; Kochian et al., 2004). Shorter roots with impaired function take up less water and nutrients, retarding plant growth.
Many species have evolved mechanisms to resist Al3+ stress, either by excluding it from the root and shoot tissues or by safely accommodating the cations taken up by the plant (Taylor, 1988, Taylor, 1991; Kochian et al., 2004; Hiradate et al., 2007). Major food crops such as wheat (Triticum aestivum), rice (Oryza sativa) and maize (Zea mays) display some resistance to Al3+ stress, suggesting that this trait is important for their wider cultivation (Aniol and Gustafson, 1984; Nguyen et al., 2003; Ninamango-Carenas et al., 2003). The significant variation between genotypes within species has been exploited by plant breeders to develop more Al3+-resistant crops (Garvin and Carver, 2003).
Common wheat (T. aestivum spp. aestivum L.; genome BBAADD) is a hexaploid species that emerged approximately 10 000 years ago from a spontaneous hybridization between tetraploid emmer wheat (T. turgidum ssp. dicoccum; genome BBAA) and the diploid goatgrass (Aegilops tauschii; genome DD). Domestication and distribution of this new species was assisted by mutations that resulted in free-threshing grains and a non-fragile rachis (McFadden and Sears, 1946; Salamini et al., 2002). Although the genotypic variation in Al3+ resistance among genotypes of T. aestivum is substantial, no resistance has been reported in any of its tetraploid or diploid progenitor species (Slootmaker, 1974; Berzonsky and Kimber, 1986; Cosic et al., 1994). Indeed, T. turgidum (emmer and durum wheat) is noted for its sensitivity to Al3+ stress (Cosic et al., 1994), a characteristic that is likely to have restricted its wider distribution in agriculture. As a consequence, Al3+ resistance is thought to have evolved in hexaploid wheat after hybridization of the tetraploid and diploid progenitors (Garvin and Carver, 2003).
We investigated the evolution of the major Al3+ resistance mechanism in T. aestivum by examining its progenitor species and by directly testing the effect of the various promoter alleles on TaALMT1 expression. We conclude that the resistance mechanism involving malate efflux has multiple independent origins, many of which involve cis mutations that increase TaALMT1 expression.
We performed an extensive screen of Al3+ resistance in Triticum species to date, focusing on the likely progenitors of hexaploid wheat (Table 1). The germplasm included accessions from the Watkin’s collection, which includes a diverse range of accessions of T. turgidum, T monococcum, T. urartu, T. dicoccoides, T. timopheevii and T. zhukovskyi (Miller et al., 2000), as well as accessions of Ae. tauschii collected throughout its distribution range (Lagudah et al., 1991). Of the 760 accessions screened, none possessing only the A and/or B genomes displayed significant Al3+ resistance (Table 1). However, five of the Ae. tauschii (DD) accessions screened did display moderate levels of resistance compared to the resistant hexaploid genotype ET8 (Figure 1a and Table S1). In this paper, we refer to the orthologue of TaALMT1 in Ae. tauschii as AetALMT1. Al3+ resistance in the five accessions of Ae. tauschii was positively correlated with malate efflux from roots and with the level of AetALMT1 expression (Figure 1), reflecting the major resistance mechanism in hexaploid wheat encoded by TaALMT1. To test whether this trait is expressed in a hexaploid plant, we examined the progeny from a hybridization between one of the Al3+-resistant Ae. tauschii accessions, AUS18913, and the tetraploid species T. turgidum spp. durum (cv. Langdon). The resulting synthetic hexaploid plants were found to be more resistant to Al3+ and showed a greater Al3+-activated efflux of malate than the tetraploid parent (Figure 2). These results demonstrate that a moderate level of Al3+ resistance occurs in Ae. tauschii, and this probably pre-dates the appearance of hexaploid wheat.
Table 1. The number of diploid and polyploid accessions of Triticum species screened for Al3+ resistance
Number of accessions screened
Number of accessions showing resistance
Triticum dicoccoides (wild emmer)
Triticum turgidum (durum)
We investigated the allelic diversity of AetALMT1 in 55 accessions of Ae. tauschii, including the five Al3+-resistant ones, by sequencing three polymorphic regions of the gene (regions 1, 2 and 3), and scoring them with a cleaved amplified polymorphic sequence (CAPS) marker that differentiates the two alleles of the TaALMT1 coding region (Figure 3a) (Sasaki et al., 2004). In phylogenetic trees generated by analysing the sequence from regions 2 and 3, the five Al3+-resistant accessions clustered together, indicating a common origin (Figure S1a,b). However the resistant accessions were not differentiated from the others in a phylogenetic tree generated from the promoter sequences of region 1 (Figure S1c). All 55 Ae. tauschii accessions, including the Al3+-resistant ones, lacked the tandemly repeated elements commonly detected in the TaALMT1 promoters of Al3+-resistant hexaploid wheat. We then tested an additional 305 accessions of Ae. tauschii (Table S2) by PCR, and none possessed tandem repeats upstream of AetALMT1. The absence of these repeats in a total of 360 accessions collected throughout the distribution range of this species (Table S2) suggests that they do not occur in Ae. tauschii. Instead, most of the AetALMT1 promoter sequences (region 1) were identical to the type I and Ia alleles of hexaploid wheat which, with few exceptions, are associated with Al3+ sensitivity (Sasaki et al., 2006). Additional single nucleotide polymorphisms (SNPs) and indels were also identified, and these define additional promoter alleles now designated as Ib, Ic, Id, Ie, If and Ig (Tables S1 and S3).
We tested whether the tandem repeats upstream of TaALMT1 contribute to Al3+ resistance in hexaploid wheat by enhancing TaALMT1 expression. Two promoter alleles containing mutliple repeats (types V and VI) and two without repeats (types I and Ia) were used to drive GFP expression in transgenic rice (Figure 3b). The type I promoter allele is common in Al3+-sensitive hexaploid wheat, including ES8 (Sasaki et al., 2006). Type Ia differs from type I by eleven SNPs or indels, and is present in Al3+-resistant and Al3+-sensitive accessions of Ae. tauschii (Table S1). The type V and VI promoter alleles each contain three perfect tandem repeats in different patterns, and have only been detected in Al3+-resistant hexaploid wheat (Sasaki et al., 2006; Raman et al., 2008). The relative strengths of these promoters were compared by quantifying fluorescence intensity and transcript level in transgenic calli and in the root apices of regenerated plants. Rice transformation is considerably more efficient than wheat transformation, and promoter efficiency was therefore analysed in this heterologous system. In a time-course experiment, the type V promoter generated greater GFP fluorescence in transgenic calli than the type I promoter after 6 days on regeneration medium (Figure 4). When all four promoters were compared directly in the calli under similar conditions, the two promoters with tandem repeats generated significantly more fluorescence and accumulated more transcript than the promoters without repeats (Figure 5a,b). Measurements on plants regenerated from the transgenic callus yielded similar results, although the differences were not as large. Seedlings transformed with the type V and VI promoters accumulated significantly more GFP transcript in the root apices than those transformed with the type I promoter that lacks repeats (Figure 5c). Data from experiments with the type Ia promoter are not included here because insufficient plants were regenerated from the callus. Further experiments expressing these constructs in Al3+-sensitive and -resistant wheat would help to establish whether additional factors affect the level of TaALMT1 expression between genotypes.
This study extends our understanding of the origins and evolution of Al3+ resistance in hexaploid wheat in two ways. First, we establish that Al3+ resistance occurs in Ae. tauschii, the D-genome donor of hexaploid wheat. This provides experimental evidence that a moderate level of Al3+ resistance pre-dates the appearance of T. aestivum. The mechanism involves malate efflux and enhanced AetALMT1 expression, but tandemly repeated elements are not present in the promoter of AetALMT1. Second, we demonstrate that the tandemly repeated elements upstream of TaALMT1 contribute directly to the Al3+ resistance of hexaploid wheat. We speculate that the repeats contain regulatory elements that enhance TaALMT1 expression when present in multiple copies.
Allelic variation among accessions of Ae. tauschii is generally greater than in the corresponding regions on the D-genome of hexaploid wheat, reflecting the genetic bottleneck that results from polyploidization (Caldwell et al., 2004). However, the tandem repeats upstream of TaALMT1 represent allelic variation in T. aestivum that is absent from Ae. tauschii. The simplest explanation for the absence of these promoter alleles among the 360 Ae. tauschii accessions tested here is that they appeared in hexaploid wheat during the last approximately 10 000 years. How these tandem repeats arose is unclear, but the rolling-circle DNA replication machinery used by transposable elements such as Helitrons might be involved, as suggested previously for the appearance of repeated regions within the barley Mlo gene (Piffanelli et al., 2004), especially as mobile elements can be activated by allopolyploidy in cereals (Feldman and Levy, 2005). This burst of allelic variation in hexaploid wheat is consistent with the genome plasticity that arises from polyploidy, and exemplifies how allopolyploidy can accelerate evolution (Feldman and Levy, 2005; Dubcovsky and Dvorak, 2007). The alternative explanation is that the seven promoter alleles detected in hexaploid wheat are present in Ae. tauschii but not in the accessions that we examined. This explanation is less likely as we tested several hundred accessions collected from throughout its distribution range. Furthermore, hexaploid wheat would need to have been generated by at least five independent hybridization events between the tetraploid and diploid progenitors to explain the allelic diversity present in modern genotypes (Sasaki et al., 2006; Raman et al., 2008). Current evidence supports a model in which modern hexaploid wheat is largely derived from two independent hybridization events (Caldwell et al., 2004; Dubcovsky and Dvorak, 2007).
Seven promoter alleles have been identified in TaALMT1 to date, based on the various patterns of tandemly repeated elements (Sasaki et al., 2006; Raman et al., 2008). Some of these alleles have the same pattern of repeats and differ only in the number of repeats (e.g. types IV and V, and types VI and VII; Sasaki et al., 2006). It is plausible that these are derived from one another as a result of unequal cross-overs during recombination (Delhaize et al., 2007; Raman et al., 2008). However, alleles with distinct patterns of repeats, such as types V and VI examined here, are likely to have arisen independently. As none of the Ae. tauschii accessions had AetALMT1 promoters that contained repeats, the enhanced AetALMT1 expression measured in the five Al3+-resistant accessions must be controlled by sequences beyond the promoter region analysed here. Most transcriptional enhancers lie within the first kilobase upstream of the transcription start site, but regulatory elements can be more distant or occur elsewhere in the genome (Blackwood and Kadonaga, 1998). ‘Cranbrook’ is one of very few Al3+-resistant cultivars of hexaploid wheat that lacks tandem repeats upstream of TaALMT1. Interestingly, Cranbrook is a moderately resistant cultivar (Figure 1a) that shares identical coding and promoter sequences with the resistant accessions of Ae. tauschii and aligns near them in cladograms generated from regions 2 and 3 (Figure S1) (Raman et al., 2008). One explanation for this finding is that Cranbrook inherited its Al3+ resistance trait unchanged from the original D-genome donor.
Large-scale comparisons of genomes from related species indicate that coding and non-coding sequences are subject to similar selection pressure (Shabalina et al., 2001). Nevertheless, until recently, the contribution of non-coding DNA to species diversification and evolution has received less attention than changes in protein function. Indeed, the hypothesis that mutations in cis-regulatory elements represent a major force for evolutionary change via transcriptional activity was initially controversial (Wray et al., 2003; Pennisi, 2008). An increasing number of studies now support a key role for cis mutations in controlling phenotype and speciation (Wray et al., 2003; Wittkopp, 2006; Hanikenne et al., 2008; Chan et al., 2010). For instance, a recent study in baboon populations demonstrated that allelic variation in a cis-regulatory region of the FY gene influenced gene expression and the susceptibility of individuals to the Hepatocystis pathogen (Tung et al., 2009). Fewer examples of cis mutations involving tandemly repeated elements have been reported. One study in yeast established a link between transcriptional activity and tandem repeats involving small A/T-rich repeats. Because these repeats were unstable, they allowed rapid changes in gene expression to occur, which could then be subject to selection (Vinces et al., 2009). Larger repeats have also been implicated in gene regulation. In certain strains of Penicllium digitatum, five tandemly arranged copies of a 126 bp transcriptional enhancer increase expression of cytochrome P450 sterol 14α-demethylase and provide resistance to fungicides containing demethylation inhibitors (Hamamoto et al., 2000). The promoter of the human prodynorphin gene contains an allelic variation involving tandem repeats of a 68 bp region. This region contains a putative binding site for an AP-1 transcription factor, and the multiple repeats enhance gene transcription (Zimprich et al., 2000; Rockman et al., 2005). A recent example from plants involves a series of 23 bp repeats that increase the expression of a gene (MYB10) encoding an anthocyanin-regulating transcription factor in apples (Malus x domestica) (Espley et al., 2009). The present study is an example from higher plants that links large tandemly repeated elements (200–300 bp) with phenotypic changes that facilitate adaptation to a major abiotic stress.
Aegilops tauschii is divided into the subspecies (ssp.) tauschii and strangulata. The original D-genome progenitors of hexaploid wheat most likely come from the strangulata gene pool (Dvorak et al., 1998; Ogbonnaya et al., 2005). Ae. tauschii has several varieties based on morphology and early classification systems included three of these in ssp. tauschii (anthera, meyeri and typica) and one in ssp. strangulata (strangulata). More recent genetic analyses indicate that, in some geographical regions, ssp. strangulata also encompasses the meyeri and typica varieties (Dvorak et al., 1998). Two of the five resistant accessions identified in the present study show strangulata traits (AUS110802 and AUS21711), two show meyeri traits (AUS18913 and AUS110812) and one is intermediate (AUS110668) (Table S1). All of the resistant accessions were collected from Rasht and Gorgan in northern Iran and from regions in the southern Caucasus, regions that overlap with the likely origin of hexaploid wheat (Dvorak et al., 1998). Interestingly, the frequency of the CAPS marker scores in the Ae. tauschii accessions varied with geographical locations. This marker was initially developed to distinguish the major coding alleles in hexaploid wheat (Sasaki et al., 2004; Raman et al., 2005, 2008). Both CAPS alleles are common in accessions collected from the southern and western shores of the Caspian Sea, the likely origin of hexaploid wheat, but the TaALMT1-2 (AetALMT1-2) allele occurs less frequently at locations further from this area (e.g. Armenia), and is absent in accessions collected from Pakistan, Afghanistan and Turkey. These findings are consistent with previous analyses of allelic distributions in Ae. tauschii that show greater diversity along the southern reaches of the Caspian sea (Dvorak et al., 1998; Lelley et al., 2000).
As Al3+-sensitive genotypes of Ae. taushchii and T. aestivum express the same functional protein as resistant genotypes, but at lower levels, it is interesting to consider how the gene has persisted in Al3+-sensitive genotypes. One explanation is that even low expression of this gene conferred some benefit under the selection pressure of mildly acidic soils. The distribution of acid soils in the regions in which Ae. tauschii evolved over the past 2–4 million years is not known, but some soils adjacent to the Black Sea are known to be acidic today (Adiloglu and Adiloglu, 2003). Therefore, it is plausible that soils in North-Eastern Turkey, perhaps extending up to Georgia, provided sufficient selection pressure to fix the Al3+ resistance trait in specific accessions of Ae. tauschii. Further selection pressure will have been exerted as the distribution of hexaploid wheat was extended by humans into regions in which acid soils are more prevalent. An alternative explanation is that AetALMT1 provided other benefits that helped it persist in the absence of direct selection pressure from acid soils (Ryan and Delhaize, 2010).
We provide evidence that a major mechanism for Al3+ resistance in hexaploid wheat has multiple independent origins that function by enhancing the expression of TaALMT1. One origin is Ae. tauschii while others probably arose since the appearance of hexaploid wheat and rely on cis mutations in the TaALMT1 promoter. These mutations were able to increase the Al3+ resistance above the levels present in Ae. tauschii. These promoter alleles, together with other minor resistance loci (Cai et al., 2008; Ryan et al., 2009), help to explain the variation of Al3+ resistance evident among modern wheat cultivars.
Seed was obtained from CSIRO Plant Industry (Canberra, Australia), the Industry and Investment NSW (Wagga Wagga, Australia) and the Australian Winter Cereal Collection of the Industry and Investment NSW (Tamworth, Australia). The 360 Ae. tauschii accessions studied here were collected throughout their distribution range in eastern Europe, the Middle East and Western Asia. Geographic origins for each are provided in Tables S1 and S2. Although the TaALMT1 genes of hexaploid wheat are derived from Ae. tauschii, we use the terminology AetALMT1 to distinguish the Ae. tauschii genes from those of hexaploid wheat and from the AtALMT1 gene of Arabidopsis thaliana (Hoekenga et al., 2006).
Growth conditions and measurement of Al3+ resistance
Seeds were planted in 20 L of aerated nutrient solution (Ryan et al., 1995b). Al3+ resistance was estimated on the basis of relative root length (RRL) and haematoxylin staining. RRL was estimated by measuring the length of the longest root on each seedling before and after 4 days of growth in control solution or 10 μm AlCl3. RRL was calculated as (net root growth under Al treatment/net root growth in control solution) × 100. Al3+ resistance for the Ae. tauschii accessions is presented relative to ET8, a standard Al3+-resistant hexaploid that was included in each trial to account for variation between experiments.
Malate efflux was measured from excised root apices in 50 μm AlCl3 as described previously (Ryan et al., 1995b). Malate efflux from each accession of Ae. tauschii is presented relative to that for a standard hexaploid genotype (ET8) that was included in each trial to account for variation between experiments.
Relative expression of the ALMT1 genes was analysed by real-time quantitative PCR on a Rotor-Gene (Corbett Research, http://www.corbettlifescience.com) using comparative quantification with Rotor-Gene software version 6.1. cDNA prepared from 5–10 root apices (4 mm) using a Qiagen RNeasy kit (http://www.qiagen.com/) was amplified using the forward and reverse primers 5′-CGTGAAAGCAGCGGAAAGCC-3′ and 5′-CCCT CGACTCACGGTA-CTAACAACG-3′, respectively. The two internal reference genes used were the phosphate transporter gene PT1 (Genbank accession number AF110180), which was amplified using forward and reverse primers 5′-GAAGGACATCTTCACGGCGATC-3′ and 5′-CACGGCCA-TGAAGAAGAAGC-3′, respectively, and the glyceraldehyde-3-phosphate dehydrogenase gene (Genbank accession number EF592180) using the forward and reverse primers 5′-TGTTGAG-GGTTTGATGACCAC-3′ and 5′-TCAGACTCCTCCTTGATAGC-3′, respectively. The reference genes yielded similar results, and data are presented for PT1 only.
Allelic diversity in Ae. tauschii
Allelic diversity of AetALMT1 was first examined in 55 accessions of Ae. tauschii (Tables S1 and S3) by sequencing three regions of the gene (regions 1, 2 and 3) and scoring the accession for a cleaved amplified polymorphic sequence (CAPS) marker (Figure 3a). Region 1 spans approximately 610 bp of the promoter, encompassing the tandem repeats present in Al3+-resistant hexaploid genotypes. Amplification was achieved using the forward and reverse primers 5′-GCTCCTACCACTATGGTTGCG-3′ and 5′-CAGGCCGACTTTGA-GCGAG-3′. Region 2 spans 519 bp in intron 3 and was amplified using the forward and reverse primers 5′-GACCAGGACCAGTTGCGCAC-3′ and 5′-GCCAGCGGACCTAAGGTTGC-3′. Region 3, a 634 bp fragment straddling intron 4 to exon 6, was amplified using the forward and reverse primers 5′-GGATACAGAGGGTGCGGGTTAC-3′ and 5′-AATGCAAGCTCATTTCGCCAC-3′. The CAPS marker was used as described previously (Sasaki et al., 2004). This marker targets a SNP in exon 4 of TaALMT1 that distinguishes the two major alleles in coding the coding region called TaALMT1-1 and TaALMT1-2. The presence or absence of tandem repeats in the promoter of 305 additional accessions (Table S2) was assessed on the basis of the lengths of PCR products amplified using primers 5′-GCTCCTACCACTATGGTTGCG-3′ (forward) and 5′-CTTGAGCTTGCATTGCATCTG-3′ (reverse). The promoter allele previously designated type I′ by Sasaki et al. (2006) is named type Ia in this study to allow for the additional allele types Ib–Ig detected in Ae. tauschii (Table S1).
Four promoter alleles of TaALMT1 and AetALMT1 representing types I, Ia, V and VI (Genbank accession numbers AB243162.1, AB243170.1, AB243166.1 and AB243167.1) were amplified using primers 5′-GATGAGGCGCGCCGGCAGATGCAATGCAAGCTC-3′ and 5′-CACTAGAGCGGCCGCTTAATTAATGAGCTTCCATGTCTATTG-3′ (restriction sites for PacI and AscI are underlined). Digested PCR products were ligated into the HvPht1 binary expression cassette to drive GFP expression (Schunmann et al., 2004). Rice was transformed with Agrobacterium using the procedure described by Toki et al. (2006). Expression was determined by quantifying GFP fluorescence and by real-time quantitative PCR. Each replicate represents an independently transformed callus or primary transgenic (T0) plant regenerated from the callus. GFP fluorescence in calli was quantified using a fluorescence microscope (Leica MZFLIII, http://www.leica.com/) and AnalySIS Five software (Olympus Soft Imaging Systems, http://www.olympus-global.com/). The green wavelengths in each image were first separated from the others. They were then converted to a grey scale so their intensity could be measured and compared with the intensities in other figures. Regions to be analysed were selected using a freehand polygon tool, and the mean grey scale value of the selection was then determined. A ‘blank’ value obtained from tissue transformed with a vector lacking the GFP gene was subtracted from the raw data. The transcript level in the calli and root apices (5 mm) was analysed by real-time quantitative PCR as described by Ryan et al. (2009). For GFP, the forward and reverse primers were 5′-GGTCACGAACTCCAGCAGGA-3′ and 5′-AGAACGGCATCAAGGTGAAC-3′, respectively. The reference for these experiments was Hpt, the antibiotic selection gene included in the binary vector that confers resistance to hygromycin. Use of Hpt as the reference gene enabled us to account for the likelihood that not all cells in the callus were transformed and for variations in gene expression as a consequence of where the transgenes were inserted into the genome. Forward and reverse primers for the reference gene were 5′-TCGGTTTCCACTATCGGCGAGTACTTC-3′ and 5′-ATCTTCTTCTGGAGGCCGTGGTTG-3′, respectively. Statistical analysis of GFP expression in transgenic calli and regenerated plants was performed using one-way analysis of variance.
Phylogenetic analyses were performed using Phylip-3.68 (Felsenstein, 1989). Base changes along the tree were analysed using the assumptions outlined previously (Fitch, 1971) and the sites were unweighted. Additional details are presented in the online legend to Figure S1.
The authors are grateful to Evans Lagudah (CSIRO Plant Industry, Canberra, Australia) for helpful discussions and comments on the manuscript and for providing access to the collection of DNA for Ae. tauschii accessions, and to Ms Belinda Taylor (I&I NSW) for technical assistance. S.G. was funded by the Department of Biotechnology, New Delhi, India, and by Deputation Leave granted by the Indian Council of Agricultural Research, New Delhi, India. This study was funded by CSIRO Plant Industry.