Unused natural variation can lift yield barriers in plant breeding

Abstract

Natural biodiversity is an underexploited sustainable resource that can enrich the genetic basis of cultivated plants with novel alleles that improve productivity and adaptation. We evaluated the progress in breeding for increased tomato (Solanum lycopersicum) yield using genotypes carrying a pyramid of three independent yield-promoting genomic regions introduced from the drought-tolerant green-fruited wild species Solanum pennellii. Yield of hybrids parented by the pyramided genotypes was more than 50% higher than that of a control market leader variety under both wet and dry field conditions that received 10% of the irrigation water. This demonstration of the breaking of agricultural yield barriers provides the rationale for implementing similar strategies for other agricultural organisms that are important for global food security.

Figure 1. Genetic Sources of BY Variation

(Exotic variation) Since M82, the multiple-introgression line (IL789), and their hybrid IL789 × M82 (ILH789) differ only in three S. pennellii segments (green chromosomes), any BY difference between them is associated with the exotic allelic variation. (Cultivated variation) Yield differences between M82 (red chromosomes), the four tomato tester inbreds (pink chromosomes), and their hybrids (M82 × Testers) result from allelic variation present in the cultivated tomato gene pool. (Cultivated + Exotic) The yield of the hybrids of IL789 with the four testers (IL789 × Testers) results from both cultivated and exotic variation.

Figure 2. Pyramiding of S. pennellii Introgressions That Increase Agricultural Yield Components

Introgression lines IL7-5-5, IL8-3, IL9-2-5, and IL789, which combines all three segments, were compared to M82 (percent difference from M82) in a homozygous (IL) and heterozygous (ILH) condition in wet and dry fields (1 plant/m²). The bars represent total yield (Y), brix (B), and brix × yield (BY) means (± standard error) from three growing seasons; these data were pooled, since no season × genotype interactions were found. The base line represents M82, where the mean BY values of M82 from the three seasons were 353 g/m² in the irrigated treatment (455 g/m² in 2001, 285 g/m² in 2002, and 320 g/m² in 2003) and 184 g/m² in the dry treatment (244 g/m² in 2001, 186 g/m² in 2002, and 122 g/m² in 2003). The additive effect (a) is half of the difference between each IL and M82. The dominance deviation (d) is the difference between ILH and the mid-value of its parents. Values marked by an asterisk are significant (p < 0.05). The bars in the gray background and their corresponding a and d values represent the expected values of IL789 and ILH789 assuming complete additivity of the introgression effects. Asterisks above the expected-value bars represent significant deviations from the observed means for IL789 or ILH789 as determined by a t test at a confidence level of 95%. All experiments were transplanted in a randomized block design with the following number of replications for each genotype under each irrigation regime: 2001, 10 replications; 2002, 15 replications; 2003, 15 replications.

Figure 3. Contribution of Cultivated and Exotic Variation to BY

(A) The contribution of cultivated and exotic variation to BY in five genetic backgrounds. BY phenotypes in the Akko wide-spacing wet and dry experiments that involved four independent tester inbreds and their hybrids with M82 and IL789 are shown. Included are mean values for a control background of M82 and BOS3155. The four inbreds are represented as horizontal lines with circles in the gray bars. Experiments were transplanted in a randomized block design with 20 replications for each genotype under each irrigation regime. In all cases BY of the hybrids containing the exotic introgressions (IL789 × Testers) was significantly higher than that of their nearly isogenic cultivated tomato hybrids (M82 × Testers; t test, p < 0.01). The exotic effect, which represents the BY differences between the IL789 × Testers hybrids and the M82 × Testers hybrids, was consistent for all genetic backgrounds in each of the irrigation regimes. This was determined by genetic background × exotic effect two-way ANOVA (p for the interaction is 0.75 for the wet treatment and 0.88 for the dry field). (B) The mean contribution of cultivated and exotic variation to BY in wet and dry fields. BOS3155 is a leading commercial tomato hybrid that was used as a reference. Values of tester inbreds, M82 × Testers, and IL789 × Testers (shown as Δ% from M82) represent the means of the four genotypes included in each group (see Figure 3A). Base line and the letters attached to it represent M82. Means for each irrigation treatment with different letters are significantly different using a multiple-range means comparison (Tukey-Kramer; p < 0.01). The deduced exotic effect on BY is marked as black bars, and the contribution of the cultivated variation to BY is marked in gray. The absolute BY values of M82 were 303 g/m² in the wet treatment and 122 g/m² in the dry treatment.

Figure 4. BY Phenotypes of M82, BOS3155, and the Best Hybrid Combination (IL789 × 76) in Six Independent Trials

Plants were grown in two locations, under two planting densities and two irrigation regimes. The locations were (i) the Western Galilee Experimental Station in Akko and (ii) Kibbutz Mevo-Hama (MH) in the Golan Heights. The planting densities were (i) single plants (SP; 1 plant per m²) and (ii) plots (14 plants per 4 m², or 3.5 plants/m²). The irrigation regimes were (i) wet (320 m³ of water per 1,000 m² of field throughout the growing season) and (ii) dry (30 m³ of water per 1,000 m² of field). All experiments were transplanted in a randomized block design with the following number of replications: Akko-SP-wet, 15 replications; Akko-SP-dry, 20 replications; Akko-plots-wet, 8 replications; Akko-plots-dry, 8 replications; MH-SP-wet, 15 replications; MH-SP-dry, 15 replications. Means not connected by the same letter are significantly different using a multiple-range means comparison (Tukey-Kramer; p < 0.01).