High-throughput phenotyping of a maize introgression library under water deficit conditions
Résumé
A maize Introgression Library (IL) of 75 lines was derived from the cross between Gaspé Flint (an extremely early Canadian landrace, used as donor parent) and B73, used as recurrent parent (Salvi et al., 2011). The population was formerly shown to segregate for several flowering time QTLs as well as for Seminal Root Architecture (SRA) traits. In this experiment, the IL was characterized using the high-throughput phenotyping platform PhenoArch (INRA, Montpellier), a greenhouse platform equipped for large-scale automated plant imaging and transpiration measurements. Objective of this investigation was to a) to screen the IL for the presence of growth and water use efficiency (WUE) QTLs under different water regimes and b) to test the effect of previously identified QTLs for flowering time and SRA on the growth and WUE.
The maize IL lines were tested under well-watered and water-deficit conditions (soil water potential > -0.05 and ~ -0.4 MPa, respectively). Each pot was weighted at least once per day allowing for the accurate estimation of daily transpiration. By means of image analysis and suitable models (multiple regression and sigmoidal fitting), it was possible to estimate biomass accumulation and thus WUE. Preliminary results showed statistically significant genetic effect for several traits, including biomass accumulation, leaf emission rate and WUE (P < 0.05, Dunnet test).
Based on the collected phenotypic data and highly dense SNP genotyping, several QTLs for biomass accumulation and WUE were mapped on Chr. 1, 2, 3, 5, 8, 9, 10, both in well-watered and water-deficit conditions. QTLs for phenology and RSA traits seems to affect differently plant development, biomass accumulation and WUE. A particularly interesting QTL conferring increased WUE in both WW and WD conditions was mapped on chr. 9.
The data collected using the platform PhenoArch will contribute to the comprehensive multi-trait phenotypic characterization of the IL. The most interesting IL lines and corresponding QTLs will be considered for fine mapping and QTL cloning.