Abstract

Quinoa (Chenopodium quinoa Willd.) is an important food crop for subsistence farmers in the Altiplano (high plains) of Peru, Bolivia, and Argentina. Saponins are part of a diverse family of secondary metabolites that are found in high concentrations in the pericarp of many varieties of quinoa. Due to their bitter taste and anti-nutritive properties, saponins must be removed before the quinoa grain is consumed. There are ‘sweet’ varieties of quinoa that have significantly reduced levels of saponin. Previous research suggests saponin production is controlled by a single locus. The major objective of this research was to elucidate the genetic components in the saponin biosynthesis pathway. Thus, we report the development and annotation of the first large scale expressed sequence tag (EST) collection for quinoa based on Sanger and 454 pyrosequencing of maturing seed tissue expressing saponins. Sanger sequencing produced 18,325 reads with an average read length of 693 nucleotides, while 454 GS-FLX pyrosequencing generated 295,048 reads with an average read length of 202 nucleotides. A hybrid assembly of all sequences generated 39,366 unigenes, consisting of 16,728 contigs and 22,638 singletons. Repeat sequence analysis of the unigene set identified 291 new microsatellite markers. From the unigene set, a custom microarray was developed and used to assay transcriptional changes in developing seeds of saponin-containing and saponin-free quinoa lines. The microarray consisted of 102,834 oligonucleotide probes representing 37,716 sequences of the unigenes set. Three different statistical comparisons, based on comparisons of ‘sweet’ vs. ‘bitter’ seed tissue at two developmental stages, were assayed on the custom array. Using a p-value cutoff threshold of 0.01, we identified a list of 198 significantly differentially expressed candidate genes common to all three comparisons. We also identified a list of candidate genes (p-value ≤ 0.05) that are known to be associated with identified triterpenoid (saponin) biosynthetic pathways that were differentially expressed in all three comparisons. Included in this list are candidate genes that share homology to cytochrome P450s (20), cytochrome P450 monooxygenases (10), and glycosyltransferases (49) suggesting that transcriptional differences in the saponin biosynthesis pathway possibly responsible for the absence or presence of saponin in quinoa are determined after the formation of the β-amyrin skeleton. These candidate genes are suggested for use in future studies in the production of saponin in quinoa.

Degree

MS

College and Department

Life Sciences; Plant and Wildlife Sciences

Rights

http://lib.byu.edu/about/copyright/

Date Submitted

2009-12-02

Document Type

Thesis

Handle

http://hdl.lib.byu.edu/1877/etd3309

Keywords

Chenopodium quinoa, saponins, EST assembly, microarray, 454 sequencing, SSRs

Share

COinS