Epigenetic control of transposon activity and genomic imprinting in hybrid seeds – EpiHYBRIDS
In flowering plants, hybridization between closely related species or parents with different ploidy often results in disrupted genomic imprinting in the endosperm and seed collapse. Previous studies have shown that paternal small RNAs mediate endosperm balance by targeting transposable elements (TEs) flanking imprinted loci in the seed, but the mechanisms remain largely unclear.
In this project we aimed at providing further understanding on the origins, dynamics and function of small RNA subsets responsible for hybrid seed collapse, as well as their target TEs.
The hypothesis that has laid foundation for the EpiHybrids project is that natural variation in transposon diversity and dosage between parental lines contributes to reproductive isolation, via the activity of pollen small RNAs that disrupt genomic imprinting in the endosperm of hybrid seeds. Therefore, we aimed at i) providing further understanding on the origins, dynamics and function of small RNA subsets responsible for hybrid seed collapse in interploidy crosses in A. thaliana, and ii) develop new strategies to bypass interploidy and interspecific hybridization barriers in plants.
1. Profiling small RNA and DNA methylation in pollen using RNA sequencing and bisulfite sequencing technologies.
2. Ectopic expression of the Arabidopsis RNaseIII-like protein RTL1 in the two different pollen cell types, in order to identify the source of TE-derived small RNAs involved in the triploid block.
3. Chemically-induced epimutagenesis in Arabidopsis plants exposed to the DNA methylation inhibitor 5-Azacytidine, in order to generate epimutants segregating hypomethylated loci.
4. Targeted DNA demethylation of imprinted loci in the male germline using DNA glycosylases.
We have developed a new method to identify TE-derived small-interfering RNAs (siRNAs) produced in developing pollen, by using a fluorescent tagged line (FTL) that is genetically linked to the wild-type allele of NRPD1a (gene encoding the largest subunit of RNA Polymerase IV). Conveniently, this FTL line contains a pollen-expressed fluorescent marker (pLat52::DsRed) that was crossed with nrpd1 T-DNA mutants, resulting in and F1 plant that is heterozygous for the nrpd1 and FTL alleles. This allowed the purification of wild-type (DsRed +) and nrpd1 (DsRed -) pollen by fluorescence-activated cell sorting (FACS) from the sample plants, and identification of Pol IV-dependent siRNAs that are actively produced during pollen development. We then proposed to perform ectopic expression of the RNaseIII-like enzyme RTL1 in different pollen cell types, as RTL1 degrades endogenous siRNAs in A.thaliana. The main goal of this experiment was to identify the source of siRNAs involved in the triploid block, and independent transgenic lines for the different constructs were analyzed to test the effect of RTL1 in siRNA biogenesis (small RNA sequencing and analysis) and in the triploid block (interploidy hybridizations). This analysis showed that Pol IV transcribes different TE subsets in the two pollen cell types, but siRNAs involved in the triploid block are produced only in the vegetative cell.
Pachamuthu K, Simon S, Borges F. Targeted suppression of siRNA biogenesis in Arabidopsis pollen promotes triploid seed viability. Nature Communications (2024)
We have also discovered that chemically-induced epimutagenesis in Arabidopsis plants exposed to the DNA methylation inhibitor 5-Azacytidine allows suppressing the triploid block in a dose-dependent manner. Strikingly, strong suppressor lines showed stable transgenerational inheritance of hypomethylation at TEs flanking imprinted loci, which are well-known targets of DNA glycosylases. We then performed targeted DNA demethylation of imprinted loci in the male germline using a promoter specifically expressed during early pollen development, which confirmed that hypermethylation at DNA glycosylase targets is functionally associated with a strong triploid block response.
Huc J, Dziasek K, Pachamuthu K, Woh T, Köhler C, Borges F. Bypassing reproductive barriers in hybrid seeds using chemically induced epimutagenesis. Plant Cell (2022)
The results obtained in this project have provided a strong basis for future research, to better understand the epigenetic basis of the triploid block response in A. thaliana. In addition, this work allowed developing precise molecular tools to overcome seed-based hybridization barriers that are of great interest for plant breeding and crop improvement. The identification of pollen-derived siRNAs involved in the triploid block, as well as particular TE families that must be hypermethylated to build a strong triploid block response, will now allow to investigate how this interaction is able to promote triploid seed collapse in Arabidopsis and important crop species.
Hybridization between plants with different ploidy often encounter post-zygotic reproductive barriers in the seed, which are mediated by transposon-derived small RNAs that disrupt genomic imprinting. Strikingly, this is highly reminiscent of hybrid dysgenesis in Drosophila, suggesting that plants and animals utilize similar small RNA guides to control transposon activity and dosage in hybrid genomes. EpiHYBRIDS aims to provide mechanistic insight into the origins and function of small RNAs responsible for hybrid seed collapse in Arabidopsis, by performing 1) large-scale small RNA depletion and 2) targeted epimutagenesis at imprinted loci with single-cell resolution. The overall goal of this project is to elucidate the epigenetic basis of genome dosage responses in interploidy crosses, but the results may provide valuable insight on additional small RNA-related phenotypes that are highly relevant for modern agriculture and food production, such as heterosis and apomixis.
Project coordination
Filipe Borges (Institut Jean-Pierre BOURGIN)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partnership
IJPB Institut Jean-Pierre BOURGIN
Help of the ANR 329,229 euros
Beginning and duration of the scientific project:
January 2020
- 48 Months
