Fig 1.
Subpopulation structure of Triticum monococcum ssp. aegilopoides.
(a) Morphological differences of spikes between T. monococcum ssp. aegilopoides KU-3620 and T. urartu KU-199-16. White bars indicate 5 cm. (b) Estimation of the optimal number of sub-populations (k) from STRUCTURE with the ΔK method. (c) Proportion of membership of the 43 accessions for K = 2, as calculated by STRUCTURE and CLUMPP software based on the polymorphisms detected by SSR markers. (d) Geographic distribution of the lineage 1 (L1) and lineage 2 (L2) accessions of wild einkorn and one accession of T. urartu.
Table 1.
The Am-chromosome-specific CAPS markers developed in this study.
Fig 2.
Spike morphology and cytological analysis of AABBAmAm synthetic hexaploids.
(a) Spikes of AABBDD, AABBAmAm, and AABBAA synthetic hexaploids. (b, c, d) GISH analysis of Ldn/Triticum monococcum ssp. aegilopoides KU-3620. Chromosomes were counterstained with DAPI (blue) (b), stained with an Am (T. monococcum ssp. aegilopoides KU-3620 DNA, Alexa Fluor 555) genomic DNA probe for the Am genome DNA (red) (c), and the images of DAPI-stained chromosomes and GISH signals were merged (d). (e) Wild-type plants. (f) Hybrid dwarf plants.
Fig 3.
Confirmation of Am-genome chromosomes in the AABBAmAm synthetic hexaploids.
The presence of Am-genome chromosomes in the 40 AABBAmAm synthetic hexaploid lines was confirmed based on the amplification of Am-chromosome-specific CAPS markers. Their parents (Ldn and wild einkorn accessions) were used as controls. Restriction enzyme and marker names are shown in parentheses following the chromosome names on the left of each gel image. Details of the CAPS markers are described in Table 1. Size differences in amplicons between the AB and Am genomes were observed. Both amplicons from the AB and Am genomes were detected in the synthetic hexaploid lines. The full-length gel images are shown in S2 Fig.
Fig 4.
Phenotypic comparison between L1 and L2 of Triticum monococcum ssp. aegilopoides and the AABBAmAm synthetic hexaploids.
Posterior distributions of mean values for the 18 traits estimated under Bayesian GLMM. Violet and blue posterior distributions indicate mean values of L1 and L2, respectively. AABBAmAm indicates the synthetic hexaploids and AmAm indicates the einkorn accessions. The center point, thick line, and thin line below the posterior distribution designate mean, 80%, and 95% credible intervals, respectively. Dashed lines connect center points of each lineage between the synthetic hexaploids and the einkorn accessions. The difference between L1 and L2 (L2 − L1) is shown above the posterior distribution. Asterisks with the differences indicate that both the upper 95% credible interval and the lower 95% interval of the difference is above or below zero. Although the number of spikelets is discrete, a normal distribution was used as a probability distribution since a normal distribution was more supportive than a lognormal distribution and Poisson distribution in the model comparisons.
Fig 5.
Phenotypic comparison between the 2017–2018 season and 2018–2019 season of Triticum monococcum ssp. aegilopoides and the AABBAmAm synthetic hexaploids.
Posterior distributions of mean values for the 18 traits estimated under Bayesian GLMM. Violet and orange posterior distributions indicate mean values of 2017–2018 and 2018–2019 seasons, respectively. AABBAmAm indicates the synthetic hexaploids and AmAm indicates their parental einkorn accessions. The center point, thick line, and thin line below the posterior distribution designate mean, 80%, and 95% credible intervals, respectively. Dashed lines connect center points of each season between the synthetic hexaploids and the einkorn accessions. Differences between the seasons are shown above the posterior distributions. Asterisks with the differences indicate that both the upper 95% credible interval and the lower 95% interval of the difference is above or below zero.
Fig 6.
Comparisons of grain morphology between L1 and L2 of Triticum monococcum ssp. aegilopoides and the AABBAmAm synthetic hexaploids.
(a) Seed shapes among the AABBAmAm synthetic hexaploids, their parental wild einkorn accessions, the AABBDD hexaploid lines, and Ldn are shown. (b) Posterior distributions of mean values for the five grain traits estimated under Bayesian GLM without thermal days as an explanatory variable. (c) Posterior distributions of mean values for the five grain traits estimated under Bayesian GLMM with thermal days. Violet and blue posterior distributions indicate mean values of L1 and L2, respectively. AABBAmAm indicates the synthetic hexaploids and AmAm indicates their parental einkorn accessions. The center point, thick line, and thin line below the posterior distribution designate mean, 80%, and 95% credible intervals, respectively. Dashed lines connect center points of each lineage between the synthetic hexaploids and the einkorn accessions. Differences between L1 and L2 (L2 − L1) are shown above the posterior distributions. Asterisks with the differences indicate both the upper 95% credible interval and the lower 95% interval of the difference is above or below zero. Saturation and levels of the grain image were adjusted.
Fig 7.
Grain characteristics of the AABBAmAm synthetic hexaploids.
(a) Frequency distribution of the SKCS hardness values in 40 AABBAmAm hexaploid lines (b). Scanning electron microscopy of the transverse sections in Ldn/Triticum monococcum ssp. aegilopoides KU-101-3, Ldn/T. monococcum ssp. aegilopoides KU-8001, and Ldn/T. monococcum ssp. aegilopoides KU-8201 grains (b). The characteristics of all the AABBAmAm hexaploid lines were consistent with those of soft grains of common wheat.
Fig 8.
Phenotypic comparisons of the AABB (Ldn), AABBAA, AABBAmAm, and AABBDD hexaploids.
Posterior distributions of mean values for the 18 traits of Ldn and the synthetic hexaploids estimated under Bayesian GLMM. The thick line and thin line below the posterior distribution designate 80% and 95% credible intervals, respectively.