Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Titre : Leaf traits associated with drought adaptation in faba bean (Vicia faba L.)

Auteur : Khazaei, Hamid

Université de soutenance : University of Helsinki

Grade : Doctoral 2014

Résumé
The potential for use of faba bean (Vicia faba L.) is increasing worldwide due to its positive environmental impact and nutritional interest, but there are many challenges for faba bean breeding and cultivation. These include its mixed breeding system, its unknown origin and wild progenitor, its large genome being the biggest diploid genome among field crops, and its relative sensitivity to biotic and abiotic stresses (e.g., drought). Little is known about the ecological adaptation of faba bean germplasm, or about the locations and effects of genes that influence traits related to drought adaptation, especially stomatal morphology and function as key characters for gas exchange between plant and atmosphere. Thus, the current study had two main goals, a) to examine whether faba bean germplasm from drought-prone and drought-free environments differed in leaf traits related to drought adaptation while testing a novel genetic resources utilization tool, and b) to screen the genome for regions and candidate genes controlling morpho-physiological traits related to drought adaptation.
Two sets of faba bean germplasm each containing 201 accessions from dry and wet regions of the world were chosen according to the principles of FIGS (the Focused Identification of Germplasm Strategy). Leaf morpho-physiological traits related to drought response (e.g., stomatal characteristics and water status) were measured under well watered conditions in a climate-controlled greenhouse. Thereafter, two subsets of 10 accessions each were exposed to water deficit. The significance of the difference between the two sets indicated the potential of FIGS to search for traits related to abiotic stress adaptation. Machine-learning algorithms and multivariate statistics showed that the discrimination between the two sets could be based on pertinent physiological traits, particularly leaf temperature and relative water content. When the plants exposed to water deficit, leaf temperature was the most responsive trait.
Four bi-parental populations were developed, of which Mélodie/2 × ILB938/2 showed the highest number of polymorphic single nucleotide polymorphisms (SNPs) and was advanced to the F5 generation, in which 211 individuals were tested for 222 SNPs derived from Medicago truncatula sequence information. The population was phenotyped for several morpho-physiological traits during 2013 and 2014. In total, 188 polymorphic SNPs were assigned to nine linkage groups that covered 928 cM with an average inter-marker distance of 5.8 cM. The map showed a high degree of synteny with the genome of M. truncatula. Most of the detected QTLs for stomatal morphology and function were in a single region of faba bean chromosome II syntenic with a segment of M. truncatula chromosome IV that harbours receptor-like protein kinase. Furthermore, a novel locus, ssp1, for stipule pigmentation was mapped in a well conserved region of M. truncatula chromosome V containing some candidate Myb and bHLH transcription factor genes.
The difference between the leaf temperatures of the two FIGS sets (the wet and dry set) allowed us to find a reliable and cost-efficient phenotyping tool for screening drought adaptation related traits in this species. Furthermore, using an appropriate mapping population and using novel M. truncatula-derived SNPs all brought success to detect the genetic regions and to indicate candidate genes. Furthermore, our results confirm the genomic data from model plant species can easily be translated to faba bean. Finally, breeding faba bean for drought adaptation can be made more straightforward by combining the use of germplasm tools such as FIGS, rapidly assessed phenotypes such as canopy temperature, and genomic tools for detecting candidate genes.

Présentation (HELDA)

Version intégrale (0,83 Mb)