Louis Blois, Marina de Miguel, Pierre-François Bert, Nathalie Ollat, Bernadette Rubio, Kai P. Voss-Fels, Joachim Schmid & Elisa Marguerit

Theoretical and Applied Genetics (Nov. 2023) 136:223

https://doi.org/10.1007/s00122-023-04472-1

In woody perennial plants, quantitative genetics and association studies remain scarce for root-related traits, due to the time required to obtain mature plants and the complexity of phenotyping. In grapevine, a grafted cultivated plant, most of the rootstocks used are hybrids between American Vitis species (V. rupestris, V. riparia, and V. berlandieri). In this study, we used a wild population of an American Vitis species (V. berlandieri) to analyze the genetic architecture of the root-related traits of rootstocks in a grafted context. We studied a population consisting of 211 genotypes, with one to five replicates each (n = 846 individuals), plus four commercial rootstocks as control genotypes (110R, 5BB, Börner, and SO4). After two independent years of experimentation, the best linear unbiased estimates method revealed root-related traits with a moderate-to-high heritability (0.36–0.82) and coefficient of genetic variation (0.15–0.45). A genome-wide association study was performed with the BLINK model, leading to the detection of 11 QTL associated with four root-related traits (one QTL was associated with the total number of roots, four were associated with the number of small roots (< 1 mm in diameter), two were associated with the number of medium-sized roots (1 mm < diameter < 2 mm), and four were associated with mean diameter) accounting for up to 25.1% of the variance. Three genotypes were found to have better root-related trait performances than the commercial rootstocks and therefore constitute possible new candidates for use in grapevine rootstock breeding programs.

See https://link.springer.com/article/10.1007/s00122-023-04472-1

Fig. 5: Manhattan plot for SNP associations with mean root diameter (Av_Diam). The thresholds were calculated with the Bonferroni method for α = 0.05 (dashed line) and 0.01 (solid line). Significant signals are indicated by a small red dot for α = 0.05 and a larger red dot for α = 0.01. The corresponding QQ plot is presented in Figure S3A (color figure online)