Introduction
Molecular markers are used for a wide range of purposes in crop genetics and breeding, including genetic linkage and comparative mapping, positional cloning, genetic diversity assessment, phylogenetic analysis, genotypic profiling, quantitative trait loci and marker-assisted selection. In recent years, genomic microsatellites (or simple sequence repeats, gSSR) have attracted more attention because of abundance in plant genome, reproducibility, high level of polymorphism and codominant inheritance (Nicot et al., Reference Nicot, Chiquet, Gandon, Amilhat, Legeai, Leroy, Bernard and Sourdille2004). The recent wide availability of data for expressed sequence tags (ESTs) increased EST-derived SSRs, which belong to the transcribed regions of DNA and are expected to be more conserved and have a higher rate of transferability across species than genomic SSR markers (Rudd, Reference Rudd2003). The objectives of the present work were to test the transferability and polymorphisms of gSSR and EST-SSR markers in Triticum and Aegilops species closely related to cultivated wheats, and to test their applicability for wheat phylogenetic analyses. The use of polymorphic SSR markers either for the characterization and evaluation of germplasm or for phylogenetic analysis of wheat was also discussed.
Materials and methods
Twelve species or sub-species were analyzed: Triticum aestivum (AABBDD), T. turgidum ssp. durum (AABB), T. timopheevii (AAGG); T. turgidum ssp. dicoccoides (AABB), T. monococcum, T. urartu, Aegilops squarrosa, A. speltoides (S), A. bicornis, A. longissima, A. sharonensis and A. searsii. A total of 20 genotypes, including ancestral tetraploid and diploid Triticum and Aegilops species, were finally used to assess SSR applicability for phylogenetic analysis of wheat.
SSR and EST-SSR primer sequences developed by La Rota et al. (Reference La Rota, Kantety, Yu and Sorrells2005), annealing temperature and expected PCR product size are reported in the web site http://wheat.pw.usda.gov.
DNA amplifications were carried out as described by Gadaleta et al. (Reference Gadaleta, Giancaspro, Giove, Zacheo, Mangini, Simeone, Signorile and Blanco2009).
Results and discussion
Primer pairs of 79 gSSRs and 61 EST-SSRs were tested in 20 genotypes belonging to the 12 species or subspecies of the Triticum–Aegilops complex under the same PCR conditions as originally applied for amplification in wheat.
The 79 gSSR and 61 EST-SSR markers were polymorphic across the 12 species or subspecies of the Triticum–Aegilops complex (Fig. 1(a)). Among the 2800 data points (140 SSR × 20 genotypes), c. 10% were missing data, true null alleles or failed PCR amplifications. The results of DNA amplification from the genotypes of the five Aegilops species indicated that the polymorphism was genotype dependent. The mean number of alleles among the species was 1.71 (ranging from 1.90 for A. speltoides to 3.43 for T. aestivum) for gSSR markers and 1.88 (ranging from 1.26 for A. searsii to 1.84 for T. aestivum) for EST-SSR markers. Polymorphism was relatively higher in the source species than in related species. As expected, polymorphism of EST-SSR markers was lower than that of gSSR markers, likely because the higher level of conservation of DNA sequences belonging to the transcribed region of the genome, as previously reported (Nicot et al., Reference Nicot, Chiquet, Gandon, Amilhat, Legeai, Leroy, Bernard and Sourdille2004).
Fig. 1 Average number of alleles/primer pair of wheat gSSR and EST-SSR markers (a) and their transferability in 20 accessions of Triticum and Aegilops species (b).
Results of DNA amplification from the genotypes of the Aegilops species indicated that the rate of transferability was genotype dependent. Cross-species transferability of EST-SSRs was observed in 670 out of 1115 combinations (60.1%), whereas gSSRs gave amplified products in 845 out of 1407 combinations (60.0%). As expected, transferability was higher in the Triticum species with respect to Aegilops species (Fig. 1(b)), as also reported by Sourdille et al. (Reference Sourdille, Tavaud, Charmet and Bernard2001).
DICE genetic similarity coefficients were used to prepare dendrograms using the UPGMA method.
The dendrograms (Fig. 2) based on gSSR and EST-SSR bands were not significantly different. Each species was separated and the genotypes tested for the Aegilops species were always grouped together. Phylogenetic trees were consistent with cytotaxonomical and molecular data on species relationships in the Triticum–Aegilops complex. Cluster analysis showed that:
(1) Triticum species were separated from Aegilops species;
(2) T. monococcum was clustered with T. urartu, both species having a common A genome;
(3) tetraploid and hexaploid species of Triticum formed a close group;
(4) close clustering of A. bicornis, A. longissima and A. sharonensis is consistent with cytotaxonomical data; A. searsii was less clustered with the above species;
(5) within the section Sitopsis, separation of A. speltoides from the remaining four species (A. bicornis, A. longissima, A. sharonensis and A. searsii) was consistent with the Eig's classification on morphological traits;
(6) A. squarrosa seemed to be closer to the B genome donor species in the EST-SSR dendrogram and to the A genome donor species in the genomic dendrogram.
Fig. 2 Dendograms based on (a) genomic SSR and (b) EST-SSR markers.
Thus, we can conclude that wheat EST-SSR markers show a high transferability across a range of species. This transferability makes them a powerful tool to work on orphan wild species, where less effort has been devoted to develop genomic resources such as molecular markers. Wild species are an important source of both abiotic and biotic resistances, and molecular markers are precious tools to use and reduce the linkage drag derived from the introgressions of genes (location and size of the introgression) from these species.
Acknowledgements
This research was supported by a grant from Ministero dell'Università e della Ricerca, Italy, project ‘AGROGEN’.
