In my point of vue, this paper is interesting because it’s in my domain of interest but very difficult to understand because they put lot of technical word without definition and they say very often see references, as it’s described in this paper making this paper very difficult to understand. Also in this paper, the aim is not very clear and also there is no conclusion. I have the feeling that they don’t know what they can conclure. But I will try to explain in few words the paper…
About the introduction.
It is well know that natural epigenetic variation provides a source for the generation of phenotypic diversity but it remains unclear how this epigenetic variation contributes to this diversity and the relationship between genetic variation and epigenetic mechanisms. Epigenetic is defined by heritable modification of genes expression. This modification can be heritate during the meiosis and/or mitosis but does not affect a changing in DNA sequences. Epigenetic modifications are mainly defined by cytosine methylation in a DNA level or histone methylation, histone acetylation.
In plant including Arabidopsis thaliana model, basically there are 3 different patterns of methylation: methylation in a cytosine in a CG context, on a CHG context and in a CHH context (where H can be a A,C or T).
In order to understand the types and extend of natural DNA methylation variants in A.thaliana, epigenomes were determined using methyl-C-sequencing – genomic DNA sequencing – and RNA sequencing. Integration of all these information (genomic and epigenomic data) allowed investigation into variable methylation states of both CG gene body methylation and loci targeted by RdDM (RNA Directed DNA methylation).
Results
Firstly they assessed SMP (single methylation polymorphisms) diversity to understand their frequency among different population (accession). In order to do this, they compare the COL methylome (wild type) and methylome of different Arabidopsis population. They found 23% of CH pattern, 13% of CHG pattern and 64% of CHH pattern. After, they construct an epigenome-based phylogeny tree and compare these SMP with SNP. They can make a correlation between CG-SMP and SNP. So they used this feature but investigate the pattern of SMP diversity in a chromosome-wide or a genome-wide analysis (figure 1).
The major conclusion for this figure 1 is that the methylation state of SMPs in CG and CHG contexts is towards the methylated form at the centromere regions and towards transposons. The CHH is mainly in an unmethylated context. And using RNA- sequencing, it also shown that the transcription is higher in single gene copy where there is a CG methylated context. These features make sense because in the centromere there are lot of repetitive sequence and transposable element and these sequences are not transcribed.
Then, in figure 2, they tried to show the population-wide variation of DMRs. In fact, spontaneous formation of SMPs represents one form of natural epigenetic variation and this variation exists in the form of differentially methylated regions (DMR). They found that CG-DMRs are enriched in gene bodies and C-DMRs (composed of CHH-DMR and CHG-DMR) are enriched in transposable element and mainly unmethylated. The pattern of GC-DMR is the same as CG-SMP meaning that transposable elements are mainly methylated by CHH and CHG pattern leading to a silencing mechanism by RdDM pathway.
Then they look at different tissues: leaves and inflorescence using 2 different methods: when they performed a cluster using methylation levels of CG-DMRs or C-DMRs, this population cluster by their genotype and when the same analysis is performed using RNA-sequencing, this population cluster by their tissue.
This mean that DNA methylation is less dynamic than gene expression patterns and plays only a role during specific developmental stages or cell types.
Now, it’s important to understand the genetic linkage and methylation variants (figure 3). To do this, they use linkage disequilibrium decay. Basically, they wanted to know if DMR – SNP and SMP are linked with local genetic variant or not. So, the more these features are closed the more they are linked and segregate together.
If the value is around 0 these features are independent and if the value is around 1 it’s mainly due to local genetic variant.
We can see than SMP and CG-DMR reached faster 0 meaning that there are independent but SNP and C-DMR reached 50% of their value after 2kb meaning that there are due to local genetic variants.
So, now we know that C-DMRs are due to local genetic variant but which kind of genetic variant? They do an association-mapping methylation variant. They revealed VIM3 and AGO2 as possible causal loci (it’s know that AGO2 is acting in the RdDM pathway).
To sum up this part, the mQTL revealed an association between some genetic variants and DNA methylation variants expecially for C-DMR.
Finally, in order to understand the role of regions of the epigenome that are less prone to natural epigenetic variation, they searched for loci that contained methylated alleles. And they found that most of them are in seeds and in pollen. This conclusion also makes senses because by definition the epigenetic modification can be heritated. So in Arabidopsis, the pollen and seeds are the germ line, which is transmitted during generation. But it’s also found that there are lot of transposable elements is this germ line so the plant and the cell has to silencing these transposable elements. So the plant can degraded the transposable element using this RdDM silencing mechanism pathway. This silencing pathway in pollen and seed allow and ensure proper gametophytic and embryonic development.
To conclude this paper provides evidence that RdDM targeted genes may have co-opted this transposable element silencing mechanism to maintain their silencing state in vegetative tissues and transgenerationally in order to ensure proper expression in pollen, seed and germ line development.
Schmitz, R., Schultz, M., Urich, M., Nery, J., Pelizzola, M., Libiger, O., Alix, A., McCosh, R., Chen, H., Schork, N., & Ecker, J. (2013). Patterns of population epigenomic diversity Nature, 495 (7440), 193-198 DOI: 10.1038/nature11968
References
38. Li, L. et al. Linking photoreceptor excitation to changes in plant architecture. Genes Dev. 26, 785–790 (2012).
39. Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memoryefficient alignment of short DNAsequences to the humangenome. Genome Biol.
10, R25 (2009).
40. Lister, R. et al. Human DNA methylomes at base resolution show widespread epigenomic differences. Nature 462, 315–322 (2009).
41. Storey, J. D. A direct approach to false discovery rates. J. R. Stat. Soc. A 64,479–498 (2002).
42. Breitling, R. et al. Genetical