Understanding the interaction between RNA-directed DNA methylation and DNA demethylases and its role in Fusarium oxysporum disease response in Arabidopsis thaliana

Date

2016

Authors

Lee, Joanne

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

RNA-directed DNA methylation (RdDM), directed by 24 nt small interfering RNA (siRNA), is a plant-specific de novo cytosine methylation pathway responsible for natural suppression of repeat sequences such as transposons and centromeric repeats. RdDM can be counteracted by DNA demethylation, which in Arabidopsis is catalyzed by four DNA demethylases; Demeter (DME), Repressor of silencing (ROS1) and Demeter-like 2 and 3 (DML2, DML3). Early studies in our laboratory have suggested DNA demethylases and RdDM work together to regulate plant defense-related genes and mediate Arabidopsis resistance against the fungal pathogen Fusarium oxysporum. How these two pathways interact to regulate defense-related genes remained unclear. The aim of this study is to investigate how RdDM and DNA demethylases play a role in Fusarium resistance by analysing changes in small RNA (sRNA) accumulation, DNA methylation, and gene expression in wild-type Arabidopsis (Col-0) and the DNA demethylase mutant rdd (ros1 dml2 dml3) with and without Fusarium infection. Whole-genome sRNA deep sequencing plus northern blot hybridisation detected a large number of differentially accumulated (DA) siRNAs Col-0 and rdd plants. A large majority of these DA 24 nt siRNAs occurred in gene bodies, the 3kb flanking regions or transposable elements (TEs). These results indicate that DNA demethylases are involved in the accumulation of RdDM associated siRNAs, and that RdDM and DNA demethylases function co-ordinately to regulate gene expression through targeting TE sequences. Comparative analysis of gene expression and DNA methylation between Col-0 and rdd, using whole genome mRNA and bisulfite sequencing, supported by qRT-PCR analysis and existing DNA methylation data, provided evidence that DA siRNAs participate in DNA demethylase mediated regulation of nearby defense related gene targets. These analyses revealed; i) genes differentially expressed (DE) between Col-0 and rdd are enriched for stress response functions; ii) siRNAs and DNA methylation co-exist near transcription start sites (TSS) of the DE genes in the rdd background but are absent in Col-0; and iii) the new methylation in rdd spreads towards the TSS, and is immediately adjacent to the end of TEs that are methylated in Col-0 as well as in rdd. These features suggest that TEs in the promoters are the target of RdDM and have the potential to induce spread of RdDM to adjacent sequences near TSS resulting in gene repression. DNA demethylases are required to counter this effect ensuring active gene expression. In contrast to the Col-0 vs rdd comparison, of which a large proportion of DE genes have differential siRNA and DNA methylation, a very small proportion of the Fusarium responsive genes are associated with differential siRNA and DNA methylation between mock treated and Fusarium infected plants. This suggests the Fusarium responsive expression pattern is not directly controlled by RdDM changes. I propose the Fusarium response is determined by the intrinsic property of cis-elements present in stress response gene promoters, but their accessibility by transcription factors can be inhibited by DNA methylation. DNA demethylases therefore are required to maintain a low-level methylation state to regulate gene transcription activity. Previous studies state the DME demethylase is expressed and functional only in developing seed, and therefore Ros1, DML2 and DML3 should account for the bulk of demethylase activity in vegetative tissues. However, my mRNA sequencing data and our previous microarray expression data both indicated a high level of DME expression in vegetative tissues. This suggested the rdd mutant may retain some DME DNA demethylase activity in vegetative tissues, which would affect the interpretation of our data. To address this issue, I generated DME knockdown transgenic lines in the rdd background using hairpin RNA technology, and assayed the transgenic plants for Fusarium resistance. These ros1-dml2-dml3-hpDME plants exhibited enhanced Fusarium susceptibility than rdd, suggesting DME contributes to DNA demethylase activity in vegetative tissues. These transgenic materials will be valuable materials for future functional studies of DNA demethylases in Arabidopsis.

Description

Keywords

sRNA, Arabidopsis, Fusarium oxysporum, RNA-directed DNA methylation

Citation

Source

Type

Thesis (PhD)

Book Title

Entity type

Access Statement

License Rights

Restricted until

Downloads