Genome-wide association study for in vitro digestibility and related traits in triticale forage.

BACKGROUND: Triticale is making its way on dairy farms as an alternative forage crop. This requires the availability of high-yielding triticale varieties with good digestibility. Triticale forage breeding mainly focussed on biomass yield, but efforts to improve digestibility are increasing. We previously investigated the interrelationships among different quality traits in soft dough triticale: starch, acid detergent fibre and in vitro digestibility of organic matter (IVOMD) and of neutral detergent fibre (IVNDFD) of the total plant, IVNDFD and Klason lignin of the stems, and ear proportion and stem length. Here we determine the genetic control of these traits, using a genome-wide association (GWAS) approach. A total of 33,231 DArTseq SNP markers assessed in a collection of 118 winter triticale genotypes, including 101 varieties and 17 breeding lines, were used. RESULTS: The GWAS identified a total of 53 significant marker-trait associations (MTAs). The highest number of significantly associated SNP markers (n = 10) was identified for total plant IVNDFD. A SNP marker on chromosome 1A (4211801_19_C/T; 474,437,796 bp) was found to be significantly associated with ear proportion, and plant and stem IVNDFD, with the largest phenotypic variation for ear proportion (R²p = 0.23). Based on MTAs, candidate genes were identified which were of particular relevance for variation in in vitro digestibility (IVD) because they are putatively involved in plasma membrane transport, cytoskeleton organisation, carbohydrate metabolic processes, protein phosphorylation, and sterol and cell wall biogenesis. Interestingly, a xyloglucan-related candidate gene on chromosome 2R, SECCE2Rv1G0126340, was located in close proximity of a SNP significantly associated with stem IVNDFD. Furthermore, quantitative trait loci previously reported in wheat co-localized with significantly associated SNP markers in triticale. CONCLUSIONS: A collection of 118 winter triticale genotypes combined with DArTseq SNP markers served as a source for identifying 53 MTAs and several candidate genes for forage IVD and related traits through a GWAS approach. Taken together, the results of this study demonstrate that the genetic diversity available in this collection can be further exploited for research and breeding purposes to improve the IVD of triticale forage.

Variation in potential feeding value of triticale forage among plant fraction, maturity stage, growing season and genotype.

Cereal forages, such as triticale forage, progressively gain interest as alternative crop for maize. The main study objective was to investigate the variation in potential feeding value of triticale forage among maturity stage, growing season and genotype, using total plant and stem fractions. Therefore, near infrared spectroscopy (NIRS) was evaluated as fast screening tool. The prediction ability was good (ratio of prediction to deviation, RPD ≥3.0) for total plant residual moisture, starch, sugars and for stem crude ash (CAsh) and neutral detergent fibre (aNDFom); suitable for screening (2.0 ≤ RPD <3.0) for total plant CAsh, acid detergent fibre (ADFom), in vitro digestibility of organic matter (IVOMD), in vitro digestibility of neutral detergent fibre (IVNDFD) and for stem total lignin (TL) and IVNDFD; poor (1.5 ≤ RPD <2.0) for total plant crude protein, crude fat, aNDFom, lignin (sa) and for stem Klason lignin (KL); unreliable (RPD <1.5) for stem residual moisture and acid soluble lignin (ASL). The evolution in potential feeding value of 36 genotypes harvested at the medium and late milk to the early, soft and hard dough stage was followed. The most important changes occurred between the late milk and early dough stage, with little variation in quality after the soft dough stage. During 2 growing seasons, variation in feeding value of 120 genotypes harvested at the soft dough stage was demonstrated. Interestingly, variation in stem IVNDFD is almost twice as high as for the total plant (CV 12.4% versus 6.6%). Furthermore, Spearman correlations show no link between dry matter yield and digestibility of genotypes harvested at the soft dough stage. Based on linear regression models ADFom appears as main predictor of both plant IVOMD and plant IVNDFD. Stem IVNDFD is particularly determined by KL.