Introduction
In order to meet the predicted global cereal grain demand for the next decades, efficient conventional and molecular cereal breeding programmes and proper wheat genetic resources are necessary for selecting the suitable breeding lines (Baenziger et al., Reference Baenziger, Graybosch, Dweikat, Wegulo, Hein, Eskridge, Appels, Eastwood, Lagudah, Mackay, McIntyre and Sharp2008).
Many dominant genes for adaptation and trait enhancement have been lost during cereal crop domestication, but they have been retained in the genome of the wild components of the Triticeae gene pools (Dwivedi et al., Reference Dwivedi, Upadhyaya, Stalker, Blair, Bertioli, Nielen, Ortiz and Janick2008). In natural habitat, wild Triticeae species such as Dasypyrum villosum (Dv), whose genome was exposed to million of years of climatic and environmental changes, are now expressing increased heading earliness, density stands and plant biomass. Deploying whole and dissected Dv nuclear genome in the homoeologous wheat genetic background through inter-specific hybridization and introgression could be a lower cost and effective option to help wheat breeders to merge and select the proper adapted gene pools to sustain the needed yearly grain yield increase. In this study, we show that combining the Dv genome with the T. turgidum var durum genomic background and deploying dissected Dv genome in Triticum aestivum provided wheat genetic resources with new trait enhancements.
Materials and methods
Combining the whole nuclear genomes of Dv and T. turgidum var. durum
A hulled and brittle rachis Dv ecotype (2n = 2x = 14; VV) collected near Bari (Puglia, Italy) was used as pollen parent in hybridization with the free-threshing and tough rachis T. turgidum var. durum cv ‘Modoc’ (2n = 4x = 28; AABB) (Jan et al., Reference Jan, De Pace, McGuire and Qualset1986). The resulting hexaploid amphiploid (2n = 42; AABBVV), labelled M × V-b1, was fertile and showed brittle rachis (Supplementary Fig. S1, available online only at http://journals.cambridge.org). A non-brittle rachis mutated amphiploid plant (M × V-nb1; AABBVV; 2n = 6x = 42) was discovered in the M × V-b1 plot grown in 1987 (De Pace et al., Reference De Pace, Jan, Caputi and Scarascia Mugnozza2003). In the following years, other genetically stable variants (‘Mut 7-04’, ‘Mut 12-04’ and ‘Mut 16-04’) were found among the M × V-nb1 plants.
Dissection of the Dv genome by [T. aestivum cv ‘Chinese Spring’ (CS)×Dv]×CS hybridization and backcrossing
We selected six Dv-introgressed wheat breeding lines (IBLs), sharing a common CS (2n = 6x = 42; A′A′B′B′D′D′) genetic background, from a population of 150 aneuploid lines developed at the University of Tuscia, Viterbo, Italy, from the backcross (CS × Dv) × CS (Supplementary Fig. S2, available online only at http://journals.cambridge.org). The IBLs CS × V63 and CS × V32 contained a disomic addition of 6V and a disomic substitution 6V(6B), respectively, and CS_1B-1V line contained a pair of 1BL-1VS chromosomes from a spontaneous exchange between chromosome 1B of wheat and 1V of Dv. The IBLs CS × V58, CS × V59 and CS × V60 were morphologically different from CS, although they did not contain apparent GISH detectable V chromatin (Minelli et al., Reference Minelli, Ceccarelli, Mariani, De Pace and Cionini2005; Caceres et al., Reference Caceres, Vaccino, Corbellini, Cionini, Sarri, Polizzi, Vittori, De Pace, Prohens and Badenes2008). The IBLs were evaluated for resistance to Blumeria graminis f. sp. tritici (Bgt), Puccinia triticina (Pt) and P. graminis f. sp. tritici (Pgt) isolates in Italy and Hungary, and for end-use grain quality traits.
Dissection of the Dv genome by hybridization of T. aestivum cv CS and M×V-nb1 hexaploid amphiploid
An F2-like breeding population with broad genetic diversity was obtained from selfing the F1 plants obtained after crossing the hexaploid amphiploid M × V-nb1 to CS (Supplementary Fig. S3, available online only at http://journals.cambridge.org). After two generations of selfing, three F4 lines, named ‘8-1’, ‘41-3’ and ‘Mut 3-04’, were tested for two consecutive growing seasons (2007/2008 and 2008/2009), in the field at two sites: S. Angelo Lodigiano, SAL, near Lodi in northern Italy, and Tolentino, TOL, near Macerata in central Italy. The plots were managed using low-input criteria. The hexaploid wheat cultivars ‘Bologna’ and ‘PR22R58’ were used as checks. Heading time (days from 1st April), yield components and rheological properties of flour dough were evaluated at both sites.
Results and discussion
Combining the whole nuclear genomes of Dv and T. turgidum var durum
The hexaploid amphiploid M × V-nb1 and the derived homoploid variants displayed ‘farro’ traits (tenacious glumes but tough rachis) and the typical adaptive traits of Dv, such as high resistance to diseases (caused by Tilletia tritici, Bgt, Pt and Pgt), fortified caryopses (+17 to +21% protein contents and >+25% of Fe and Zn content, compared with CS) and heading earliness. When compared with the conventional ‘farro’ (T. monococcum, T. dicoccoides, and T. spelta), new ‘farro’ types expressed earlier heading and shorter culms (Table 1), and also caryopses (Mut 12-04 and Mut 16-04) with larger size.
Mean values for heading time, plant height, number of spikelets/spike, 1000 kernel weight and protein content in M×V-nb1 hexaploid amphiploid and derived variants by spontaneous mutation in comparison with ‘farro’ species used as controlsa
a Data were taken from plots at S. Angelo Lodigiano (Lodi, Italy) in 2007 and 2008.
Dissection and deployment of the Dv genome by ‘(CS×Dv)×CS’ hybridization and backcrossing
This type of hybridization favoured the haploidization of the V genome in the heterohaploid A′B′D′V F1, the random assortment (dissection) of V chromosomes in 7A′+7B′+7D′+1V gametes and the formation of wheat IBLs with the disomic addition of one of the V chromosome or V chromosome arm after backcrossing to CS. The IBLs CS × V32 and CS × V63 contained the chromosome 6V#4 (De Pace et al., Reference De Pace, Vaccino, Cionini, Pasquini, Bizzarri, Qualset and Kole2011), conferring multiple disease resistance to virulent strains of Bgt, Pt and Pgt. The IBL CS_1BL-1VS expressed superior Sodium Dodecyl Sulfate (SDS) sedimentation and rheological properties of flour dough than CS. Other IBLs (CS × V58, CS × V59 and CS × V60) had cryptic Dv chromatin introgression exhibiting a − 7 to − 9 d earliness in heading time compared with CS in the field. Using specific primers for several DNA targets in the genome of these lines revealed Dv alleles, but not CS alleles, at two loci (Vrn-A1 and Vrn-B3) involved in the vernalization response pathway (Caceres et al., Reference Caceres, Vaccino, Corbellini, Cionini, Sarri, Polizzi, Vittori, De Pace, Prohens and Badenes2008).
F3 progenies derived from the hybridization of two disomic addition lines for chromosome 6V, CS × V63 (+6V#4) and CS+6V#1 (produced by Sears, Reference Sears1982, susceptible to Bgt, Pt and Pgt; Supplementary Fig. S4, available online only at http://journals.cambridge.org) expressed simple inheritance, due to Dv-derived Mendelian dominant genes governing resistance response to each of the Bgt, Pt and Pgt pathogen. The high SDS sedimentation value of CS_1BL-1VS was traced to the effect of one high-molecular weight glutenin subunit (1v in Supplementary Fig. S5, available online only at http://journals.cambridge.org) encoded at the Glu-V1 locus in 1VS of CS_1BL-1VS (Vaccino et al., Reference Vaccino, Banfi, Corbellini and De Pace2010).
Dissection and deployment of the Dv genome by hybridization of T. aestivum cv CS and M×V-nb1 hexaploid amphiploid
‘New’ chromosome assortments were achieved using the AABBVV hexaploid amphiploid as a bridge to combine the A and B genomes from durum wheat; the A′, B′ and D′ genomes of bread wheat, and the V genome of Dv. Tetraploid AA′BB′ lines, hexaploid AA′BB′D′D′ lines, and aneuploid AA′BB′ lines with the addition of one or more V and D′ chromosomes were obtained (Supplementary Figs. S3 and S6, available online only at http://journals.cambridge.org). Promising outcomes of this breeding scheme were the high frequency of euploid segregants (Supplementary Fig. S3, available online only at http://journals.cambridge.org), the phenotypic uniformity observed in the progenies after only two generations of selfing that followed the CS × (M × V-nb1) triparental hybridization and the good agronomic values of those lines. The three inbred breeding lines, ‘8-1’, ‘41-3’ and ‘Mut 3-04’, had chromosome counting of 2n = 42 and were as good as the best check for yield components, grain quality and rheological flour dough properties (Table 2 and Supplementary Table S1, available online only at http://journals.cambridge.org).
Average values for test weight (TW), protein content (PC), SDS sedimentation volume (SSV), specific SDS sedimentation volume (SSSV=SSV/PC), gluten index, rheological flour dough properties (Brabender farinograph and Chopin alveograph) and bread quality, determined for three inbred breeding lines obtained from hybridization of T. aestivum cv CS and M×V-nb1 hexaploid amphiploida
Stab., stability; BU, Brabender Units; abs, absorption; P, pression, measures the tenacity of the dough; L, length, measures the extensibility of the dough; W, work, measures the strength of the dough.
a The values are mean of plots raised at two locations (S. Angelo Lodigiano and Viterbo) in Italy for 2 years (2008 and 2009).
In conclusion, it is suggested that a sustainable breeding response to mitigate the severe threat to the world's wheat supply is necessary (Hovmøller et al., Reference Hovmøller, Walter and Justesen2010). We evidenced that deploying whole and dissected Dv nuclear genome in the homoeologous wheat genetic background, it is possible to prepare wheat germplasm with unexploited genes for rust disease resistance and enhanced grain yield and quality traits.
Acknowledgements
This research was funded by the ‘Ministero delle Politiche Agricole Alimentari e Forestali’, Italy, in the framework of the research project FRUMIGEN (D.M. 292/7303/05 issued oct. 12th, 2005).
