Breeding strategies for the adaptation of sorghum (Sorghum bicolor L. Moench) as a novel crop for temperate Europe

Abstract

In temperate Europe, sorghum (Sorghum bicolor L. Moench) is considered a promising novel bioenergy and fodder crop. Principal advantages compared to maize besides drought tolerance include nutrient efficiency and Diabrotica virgifera tolerance. However, the history of temperate-adaptation is much younger for sorghum than for maize. Presently, its early-stage chilling sensitivity delays sowing and notably limits the available vegetation period and yield potential. Several studies have identified quantitative trait loci (QTL) for chilling tolerance related traits, but up to now very little is known about the inheritance of these traits in F1 hybrids. Since due to the strong heterosis for grain and biomass yield almost all commercial varieties are hybrids, an efficient breeding strategy for the enhancement of chilling tolerance in hybrid cultivars requires a profound understanding of heterosis, combining ability and relation between parental line and hybrid performance for this trait. In this regard, a comprehensive approach comprising field trials and controlled environment experiments was conducted to analyze emergence and juvenile shoot- and root development of four sterile female lines, 16 male lines and their factorial hybrids. The results show that most traits are heterotic and that the mid-parent values are rather poor predictors of hybrid performance. Thus, hybrid breeding programs should focus on efficient general combining ability (GCA) tests rather than on a too strict selection among lines based on their per se performance. Nevertheless, to achieve substantial enhancements of sorghum chilling tolerance on the long-term, the development of better adapted inbred lines is essential. The medium to high heritabilities estimated for seedling emergence and juvenile biomass suggest that a robust breeding progress for these complex traits is feasible. For biogas use, another principal shortcoming compared to maize is that current sorghum varieties fail to combine a high biomass dry matter (DM) yield with an adequate DM content for silage and satisfying methane yield. In the present study, early-maturing silage type S. bicolor experimental hybrids which represent a novel, alternative variety type (ideotype) for biogas use were analyzed regarding their agronomic performance and energy density. The results show that under adequate conditions, they reach higher DM contents. Due to a higher ratio of starch containing grains, the methane yield per DM unit of silage types is significantly higher than that of existing biomass type standard varieties, which offsets their presently slightly lower DM yield and allows for a methane yield per area unit similar or superior to existing biomass type varieties. However, due to the high importance of grains for yield and quality parameters, seed set and sufficient maturity (dough stage) under cold nights which can induce pollen sterility are critical factors. Experiences in maize breeding show that both systematic exploitation of heterosis by using genetically distinct heterotic pools of hybrid parents and continuous improvements of line per se performance are crucial for an optimal breeding progress in yield and adaptation. Plant height and maturity of sorghum hybrids are strongly related to their parental lines, so that selection for these traits can be efficiently conducted on inbred lines. In contrast, for hybrid biomass yield the general combining ability (GCA) of inbred lines is a far better predictor than their per se performance, even though with ongoing enhancement of inbred lines their relation to hybrid yield is expected to increase. The observed predominance of GCA over specific combining ability (SCA) effects for sorghum hybrid biomass yield facilitates the identification of superior combiners. Heterotic groups in sorghum are not clearly defined yet, since heterotic pattern were compromised by early line breeding methods and the subsequent use of cytoplasmic-male sterility (CMS) for hybrid seed production. In this study, the phylogenetic relatedness in a broad sorghum diversity set (n=470) has been shown to be predominantly based on geographic origin and races, underlining the possibility to establish heterotic groups based on these characteristics. Generally, the observed correlation between genetic distance of hybrid parents and mid-parent heterosis and hybrid yield has been low or inexistent, possibly due to a higher importance of adaptation traits. However, the maximum level of mid-parent heterosis depended on the phylogenetic group of the male parent, and was comparatively low when both hybrid parents were from the same group, supporting the idea of genetically diverse pools to be developed and maintained separately. The results of this thesis contribute to the design of efficient breeding strategies for the adaptation of sorghum as a novel crop in temperate Europe, which is expected to be accomplished in the medium term.