|Title||Combining LC-MS/MS and genetic analysis to unravel plant hormone metabolism in Arabidopsis thaliana|
|Source||Wageningen University. Promotor(en): H.J.M. Bouwmeester, co-promotor(en): D. Vreugdenhil; L.I. Sergeeva. - Wageningen : Wageningen University - ISBN 9789463952408 - 142|
Laboratory of Plant Physiology
|Publication type||Dissertation, internally prepared|
Plants synthesize a myriad of metabolites, far more than those produced by most other organisms. Plant hormones are a unique set of compounds, with distinctive metabolism and properties that affect physiological processes during development and growth. Their underlying genetic variation will have attracted much research attention by virtue of junctions of NGS technologies and metabolomics.
The aim of this thesis was to study natural variation of plant hormones in Arabidopsis. To obtain reliable quantitative data of levels of hormone using LC-ESI-MS, I first developed a set of methodological processes related to extraction and purification. A set of Arabidopsis ecotypes was chosen, and studied to earn various aspects of natural variation with traits of mature roots. Based on results of these works, my study has moved to classical linkage mapping analysis to find genetic loci responsible for levels of a few sets of hormone in roots and leaves of Arabidopsis.
Chapter 2 describes the problem of matrix effects caused by impurities in semi-purified extracts, on the accuracy of data derived from LC-ESI-mass spectrometry. Matrix effects may result in both ion suppression and enhancement, and severely affects quantitative data of hormone levels. Without proper ways to minimize matrix effects, hormone data would be unreliable, and would not allow accurate QTL mapping. I validated a few key-points that are critical for determining the levels of a wide range of plant hormones in Arabidopsis extracts based on a one-step solid phase extraction (SPE) method. For the simplified purification of Arabidopsis extracts, a commercially available HLB sorbent was chosen to recover a wide range of chemically diverse series of hormone analytes. Dilution using a much smaller starting sample (e.g., 2.5 mg) reduced the matrix effects considerably but additional measures were required for most of the analytes. Flushing the HLB-SPE column with acidic methanol was more effective to reduce matrix effect than acetonitrile based eluent. At the end, I proposed a series of steps and procedures to optimize the protocol for hormone analysis in LC-ESI-MS.
Chapter 3 describes natural variation of hormone levels found in 13 Arabidopsis ecotypes. Variations of hormone level among the accessions were remarkably small, viz., less than three-fold difference between extremes. For CKs, relatively larger variations were found for ribosides and glucosides, as compared to the free bases. Root phenotypic traits of these accessions were also measured, using a new parameter (mature root unit) for complex root systems, and correlation analyses were done between hormone data and mature root traits. For root phenotyping, length-related traits—lateral root length and total root length—showed larger variations than lateral root number-related ones. Antagonistic interactions between hormones (IAA and trans-zeatin) were detected for root weight. These findings provide enough basis to warrant a quantitative genetic analysis in plant hormone metabolism and crucial information for the choice of a proper segregating population.
Chapter 4 reports diverse QTLs that are responsible for hormone levels of CKs, SA and JA in roots of Arabidopsis Ler×Cvi RIL population. QTL analysis of two sub-populations, viz., vegetative and flowering plants revealed that many of the QTLs were development-specific, suggesting that the transition to flowering has a profound effect on hormone metabolism. Using near-isogenic lines, several significant QTLs were confirmed; three co-localized QTL regions were responsible for determining several CK metabolites. Using a knock-out plant, a functional role of zeatin N-glucosyltransferase gene (UGT76C2) underlying a large-effect QTL for levels of tZ N-glucosides and tZRMP was evaluated in CK metabolism. Pleotropic effects of this gene were found for levels of CK in both roots and leaves, but significant changes of morphological traits were observed only in roots. This suggests that CK N- glucosides play an important role in root development. I also advocated the possibility of genetic regulation of concentration ratio between tZ7G and tZ9G based on a newly observed QTL of the trait.
As a further step of QTL analysis after Chapter 4, the study was extended to leaves. Chapter 5 describes the analysis of 5 groups of hormones in rosette leaves of the same RIL population. QTL analysis showed a multitude of significant loci for levels of IAA, ABAs and CKs. Also for leaves, development-specific QTLs were detected in two sub-populations, vegetative and flowering lines. QTLs for ratios between hormone metabolites belonging to the same group but also to different groups were found and some of them partly co-localized with those of single compounds, implying that QTLs for single hormones may also affect the balance between hormones. The detection of QTLs for ratios between structurally unrelated hormones (e.g. auxin : ABA) and further fine mapping may help unravelling genetic elements underlying hormone interactions in the regulation of plant development and stress responses.
Finally, in Chapter 6, several issues arising from the separate experiments are taken into consideration. The main significance of this PhD thesis is the experimental confirmation that finding QTLs for hormone metabolism is feasible and worth being extended to other populations of Arabidopsis and to crop plants. I anticipate that in the near future metabolomics study towards natural variation of plant hormones will be part of interesting theme in quantitative genetics. It will provide us to gain a better understanding of the complexity of molecular mechanism underlying hormone metabolism in plants.