Limits...
Whole genome comparative analysis of transposable elements provides new insight into mechanisms of their inactivation in fungal genomes.

Amselem J, Lebrun MH, Quesneville H - BMC Genomics (2015)

Bottom Line: Both mechanisms require specific cytosine DNA Methyltransferases (RID1/Masc1) of the Dnmt1 superfamily.We identified fungal genomes containing large numbers of TEs with many C to T mutations associated with species-specific dinucleotide signatures.In particular, an RID1-dependent RIP mechanism was found only in Ascomycota.

View Article: PubMed Central - PubMed

Affiliation: INRA, UR1164 URGI Research Unit in Genomics-Info, F-78026, Versailles, France. joelle.amselem@versailles.inra.fr.

ABSTRACT

Background: Transposable Elements (TEs) are key components that shape the organization and evolution of genomes. Fungi have developed defense mechanisms against TE invasion such as RIP (Repeat-Induced Point mutation), MIP (Methylation Induced Premeiotically) and Quelling (RNA interference). RIP inactivates repeated sequences by promoting Cytosine to Thymine mutations, whereas MIP only methylates TEs at C residues. Both mechanisms require specific cytosine DNA Methyltransferases (RID1/Masc1) of the Dnmt1 superfamily.

Results: We annotated TE sequences from 10 fungal genomes with different TE content (1-70%). We then used these TE sequences to carry out a genome-wide analysis of C to T mutations biases. Genomes from either Ascomycota or Basidiomycota that were massively invaded by TEs (Blumeria, Melampsora, Puccinia) were characterized by a low frequency of C to T mutation bias (10-20%), whereas other genomes displayed intermediate to high frequencies (25-75%). We identified several dinucleotide signatures at these C to T mutation sites (CpA, CpT, and CpG). Phylogenomic analysis of fungal Dnmt1 MTases revealed a previously unreported association between these dinucleotide signatures and the presence/absence of sub-classes of Dnmt1.

Conclusions: We identified fungal genomes containing large numbers of TEs with many C to T mutations associated with species-specific dinucleotide signatures. This bias suggests that a basic defense mechanism against TE invasion similar to RIP is widespread in fungi, although the efficiency and specificity of this mechanism differs between species. Our analysis revealed that dinucleotide signatures are associated with the presence/absence of specific Dnmt1 subfamilies. In particular, an RID1-dependent RIP mechanism was found only in Ascomycota.

Show MeSH

Related in: MedlinePlus

Comparison of GC content (%) in TE copies and genome sliding windows. GC content in TE copies: GC content in Low quality (TEcpLQ, gray) and High quality (TEcpHQ, green) TEs and in genome sliding windows (red) was calculated with sliding windows (windows = 2 Kb, increment = 2 Kb). Y axis: TE copies% or Sliding windows%. X axis: GC%.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4352252&req=5

Fig2: Comparison of GC content (%) in TE copies and genome sliding windows. GC content in TE copies: GC content in Low quality (TEcpLQ, gray) and High quality (TEcpHQ, green) TEs and in genome sliding windows (red) was calculated with sliding windows (windows = 2 Kb, increment = 2 Kb). Y axis: TE copies% or Sliding windows%. X axis: GC%.

Mentions: One major consequence of RIP involves genome G:C content, because RIP increases the A:T content of mutated TE copies. As a result, when TEs are clustered in large blocks, the RIP-mediated mutation of C:G to A:T generates A:T rich isochores, as observed in Lmac [19]. We sought to identify A:T rich regions associated with TEs; therefore, we compared the G:C content of high quality TEs (see Methods, hereafter referred to as TEcpHQ) or low quality TEs (hereafter referred to as TEcpLQ) with the G:C content of the whole genome across 2 Kb sliding windows denoted as GSW. Lmac was the only species of our sample that displayed a bimodal GSW. The peak at 36% G:C corresponds to A:T-rich isochores composed mainly of RIPed TE copies (Figure 2), which have been described previously [19]. Other genomes displayed a unimodal GSW with a peak around 45-50% G:C, with the exception of Mvio (60%) and the negative control Atha (42%). The distribution of G:C content in TEs frequently differed from that in the GSW. In the negative control Atha, the G:C content of TEcpHQ was bimodal. The low TEcpHQ values (20%) probably correspond to heavily mutated ancestral TEs that tend to be depleted in C:G sites because they are highly methylated [43,44]. In genomes invaded by TEs (Mvio, Tmel, Mlar, Pgra, Bgra) the distribution of G:C content was similar between TEs (TEcpHQ) and the whole genome (GSW, Figure 2). However, the high TE content of these genomes clearly introduces a bias because TE content makes up a larger proportion of genomic space than non-TE content. In other genomes, the distribution of G:C content between TEs and the whole genome was very different, in particular for TEcpHQ. For example, TEcpHQ elements in BcinT4 had four peaks of GC content; one of these peaks (44% G:C) was very similar to the whole genome (GSW) peak, whereas the three other peaks comprised two groups of TEs with low G:C content (20 and 40%) and one group with high G:C content (55%). Mory showed a similar profile consisting of two peaks (40 and 60%) surrounding another peak at 50% corresponding to the whole genome GC content. TEs peaks with low G:C content in BcinT4 may correspond to TEs copies that have undergone RIP-associated C to T mutations at various rates. Indeed, the TEcpHQ peak at 20% G:C content in BcinT4 comprises AT-rich TE copies that are even more abundant among TEcpLQ elements (Figure 2). These highly degenerated TEs may result from multiple rounds of RIP leading to the mutation of all their target C:G sites.Figure 2


Whole genome comparative analysis of transposable elements provides new insight into mechanisms of their inactivation in fungal genomes.

Amselem J, Lebrun MH, Quesneville H - BMC Genomics (2015)

Comparison of GC content (%) in TE copies and genome sliding windows. GC content in TE copies: GC content in Low quality (TEcpLQ, gray) and High quality (TEcpHQ, green) TEs and in genome sliding windows (red) was calculated with sliding windows (windows = 2 Kb, increment = 2 Kb). Y axis: TE copies% or Sliding windows%. X axis: GC%.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4352252&req=5

Fig2: Comparison of GC content (%) in TE copies and genome sliding windows. GC content in TE copies: GC content in Low quality (TEcpLQ, gray) and High quality (TEcpHQ, green) TEs and in genome sliding windows (red) was calculated with sliding windows (windows = 2 Kb, increment = 2 Kb). Y axis: TE copies% or Sliding windows%. X axis: GC%.
Mentions: One major consequence of RIP involves genome G:C content, because RIP increases the A:T content of mutated TE copies. As a result, when TEs are clustered in large blocks, the RIP-mediated mutation of C:G to A:T generates A:T rich isochores, as observed in Lmac [19]. We sought to identify A:T rich regions associated with TEs; therefore, we compared the G:C content of high quality TEs (see Methods, hereafter referred to as TEcpHQ) or low quality TEs (hereafter referred to as TEcpLQ) with the G:C content of the whole genome across 2 Kb sliding windows denoted as GSW. Lmac was the only species of our sample that displayed a bimodal GSW. The peak at 36% G:C corresponds to A:T-rich isochores composed mainly of RIPed TE copies (Figure 2), which have been described previously [19]. Other genomes displayed a unimodal GSW with a peak around 45-50% G:C, with the exception of Mvio (60%) and the negative control Atha (42%). The distribution of G:C content in TEs frequently differed from that in the GSW. In the negative control Atha, the G:C content of TEcpHQ was bimodal. The low TEcpHQ values (20%) probably correspond to heavily mutated ancestral TEs that tend to be depleted in C:G sites because they are highly methylated [43,44]. In genomes invaded by TEs (Mvio, Tmel, Mlar, Pgra, Bgra) the distribution of G:C content was similar between TEs (TEcpHQ) and the whole genome (GSW, Figure 2). However, the high TE content of these genomes clearly introduces a bias because TE content makes up a larger proportion of genomic space than non-TE content. In other genomes, the distribution of G:C content between TEs and the whole genome was very different, in particular for TEcpHQ. For example, TEcpHQ elements in BcinT4 had four peaks of GC content; one of these peaks (44% G:C) was very similar to the whole genome (GSW) peak, whereas the three other peaks comprised two groups of TEs with low G:C content (20 and 40%) and one group with high G:C content (55%). Mory showed a similar profile consisting of two peaks (40 and 60%) surrounding another peak at 50% corresponding to the whole genome GC content. TEs peaks with low G:C content in BcinT4 may correspond to TEs copies that have undergone RIP-associated C to T mutations at various rates. Indeed, the TEcpHQ peak at 20% G:C content in BcinT4 comprises AT-rich TE copies that are even more abundant among TEcpLQ elements (Figure 2). These highly degenerated TEs may result from multiple rounds of RIP leading to the mutation of all their target C:G sites.Figure 2

Bottom Line: Both mechanisms require specific cytosine DNA Methyltransferases (RID1/Masc1) of the Dnmt1 superfamily.We identified fungal genomes containing large numbers of TEs with many C to T mutations associated with species-specific dinucleotide signatures.In particular, an RID1-dependent RIP mechanism was found only in Ascomycota.

View Article: PubMed Central - PubMed

Affiliation: INRA, UR1164 URGI Research Unit in Genomics-Info, F-78026, Versailles, France. joelle.amselem@versailles.inra.fr.

ABSTRACT

Background: Transposable Elements (TEs) are key components that shape the organization and evolution of genomes. Fungi have developed defense mechanisms against TE invasion such as RIP (Repeat-Induced Point mutation), MIP (Methylation Induced Premeiotically) and Quelling (RNA interference). RIP inactivates repeated sequences by promoting Cytosine to Thymine mutations, whereas MIP only methylates TEs at C residues. Both mechanisms require specific cytosine DNA Methyltransferases (RID1/Masc1) of the Dnmt1 superfamily.

Results: We annotated TE sequences from 10 fungal genomes with different TE content (1-70%). We then used these TE sequences to carry out a genome-wide analysis of C to T mutations biases. Genomes from either Ascomycota or Basidiomycota that were massively invaded by TEs (Blumeria, Melampsora, Puccinia) were characterized by a low frequency of C to T mutation bias (10-20%), whereas other genomes displayed intermediate to high frequencies (25-75%). We identified several dinucleotide signatures at these C to T mutation sites (CpA, CpT, and CpG). Phylogenomic analysis of fungal Dnmt1 MTases revealed a previously unreported association between these dinucleotide signatures and the presence/absence of sub-classes of Dnmt1.

Conclusions: We identified fungal genomes containing large numbers of TEs with many C to T mutations associated with species-specific dinucleotide signatures. This bias suggests that a basic defense mechanism against TE invasion similar to RIP is widespread in fungi, although the efficiency and specificity of this mechanism differs between species. Our analysis revealed that dinucleotide signatures are associated with the presence/absence of specific Dnmt1 subfamilies. In particular, an RID1-dependent RIP mechanism was found only in Ascomycota.

Show MeSH
Related in: MedlinePlus