Tanzi, Alberto Stefano
(2019)
Genetic investigation of plant architecture, yield, and diversity in winged bean (psophocarpus tetragonolobus (L.) DC.) for breeding programmes.
PhD thesis, University of Nottingham.
Abstract
Winged bean (Psophocarpus tetragonolobus) is an underutilised leguminous crop, cultivated mainly in hot and humid countries in Asia and the pacific area. It is traditionally grown as vegetable, for its green pods and tuberous roots, and also as a pulse crop, for its grain, with all edible parts reported to be rich in protein, vitamins, minerals, and fibre. Its plant architecture is composed of long, intertwining stems and lateral branches, which grow indeterminately and impact on final yield as well as farming practices. Yet, this aspect has remained poorly understood. Efforts in winged bean improvement programme also remain constrained by the lack of molecular tools, for example to carry out genetic dissection of the traits of interest, and genetic analysis of germplasm.
Controlled crosses were therefore performed between different genotypes of winged bean in order to investigate morphological traits contributing to its plant architecture, and investigate their effect and interaction with yield components, for pod productivity. After initial assessment of parental and F1 individuals, a parental combination (M3 × FP15) was chosen in order to establish a segregating population and to carry out phenotypic analyses. These revealed how stem length could be predominantly under control of internode length (rs= .80; p< .01), while the average length of lateral branches could impact more on the final number of pods per plant (rs= .44; p< .001) than the branch number per plant alone (rs= .38; p< .001). Together, the results shed light on key morphological traits and their potential interaction with yield components.
Along with crosses and field assessment, molecular markers were developed. First, a set of 18 genic-Simple Sequence Repeat (SSR) markers was validated, and used to initially evaluate the diversity among crossed genotypes, their heterozygosity, and to validate the obtained hybrids. A second set of markers consisted of Single Nucleotide Polymorphisms (SNPs), developed through a genotyping-by-sequencing method on 91 accessions, including parental genotypes, and on the analysed F2 population.
The SNPs across the genotyped biparental F2 population have allowed the construction of the first genetic linkage map in winged bean. Nine linkage groups (LG), seemingly corresponding to the 9 chromosomes in winged bean (2n= 2x= 18), were obtained with an average of 199 SNPs per LG. High segregation distortion was revealed by analysing the segregation patterns of the markers, leading first towards the development of a framework map without distorted marker. A complete map was then constructed by re-introducing distorted markers, resulting in a map with a total length of 1171.6cM, with 395 spaced SNPs (3.1cM/SNP) and an overall 34% of segregation distortion.
QTL analysis was carried out to dissect measured traits, by combining recorded F2 phenotypic data and the genetic information from the genetic linkage map. Through Interval Mapping (IM), 5 segregating QTLs and 8 putative QTLs were found across 9 traits, all explaining more than 10% of observed phenotypic variance. These included markers linked to genomic regions underlying traits such as stem length, branch number per plant, length of branch, and dry pod length.
Using a total of 5891 SNPs discovered across 91 accessions allowed us to perform a preliminary genetic diversity analysis of winged bean germplasm, collected from public and private sources. Genetic distance analyses showed potential agreement between genetic and geographic structure, in contrast with previous studies. The average 7.3% of observed heterozygosity could support the idea of winged bean as a predominantly inbreeding species, although within populations were still found relatively high levels of diversity. Overall fixation index (Fst) and pairwise population comparisons revealed moderate significant differentiation between analysed population groups, although South American and African accessions could have been derived from South East Asian and Papua New Guinea stocks.
Overall, the phenotypic analysis provides new insight for the design of winged bean ideotypes with reduced vegetative growth and improved pod productivity, while the QTL analysis gives the first potential markers linked to important traits that could be considered breeding targets. Together, this information provides the initial basis for altering winged bean plant architecture and supporting breeding programmes for the improvement of this crop.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)