Global DNA cytosine methylation as an evolving trait: phylogenetic signal and correlated evolution with genome size in angiosperms

Abstract

DNA cytosine methylation is a widespread epigenetic mechanism in eukaryotes, and plant genomes commonly are densely methylated. Genomic methylation can be associated with functional consequences such as mutational events, genomic instability or altered gene expression, but little is known on interspecific variation in global cytosine methylation in plants. In this paper, we compare global cytosine methylation estimates obtained by HPLC and use a phylogenetically-informed analytical approach to test for significance of evolutionary signatures of this trait across 54 angiosperm species in 25 families. We evaluate whether interspecific variation in global cytosine methylation is statistically related to phylogenetic distance and also whether it is evolutionarily correlated with genome size (C-value). Global cytosine methylation varied widely between species, ranging between 5.3% (Arabidopsis) and 39.2% (Narcissus). Differences between species were related to their evolutionary trajectories, as denoted by the strong phylogenetic signal underlying interspecific variation. Global cytosine methylation and genome size were evolutionarily correlated, as revealed by the significant relationship between the corresponding phylogenetically independent contrasts. On average, a ten-fold increase in genome size entailed an increase of about 10% in global cytosine methylation. Results show that global cytosine methylation is an evolving trait in angiosperms whose evolutionary trajectory is significantly linked to changes in genome size, and suggest that the evolutionary implications of epigenetic mechanisms are likely to vary between plant lineages.

Keywords: C-value; DNA cytosine methylation; HPLC; angiosperms; correlated evolution; epigenetics; genome size; phylogenetic signal.

Figure 1

Dot plot of global DNA cytosine methylation in the sample of 54 angiosperm species studied. Each dot represents one species, and dot width corresponds to the maximum bin width as determined from the data using a dot-density binning algorithm. The position of data points on the axis are denoted by tick marks. “At” indicates the position of Arabidopsis thaliana, discussed in the text.

Figure 2

Phylogenetic tree depicting the inferred evolutionary relationships between the 54 angiosperm species considered in this study. Information on percent cytosine methylation and C-value for each species is coded as dots beside species names. The two variables were centered and scaled for the plot. See Tables S1, S2 for raw data, and Figure S1 for the distribution over the phylogenetic tree of the 12 species with missing C-value data.

Figure 3

Bivariate scatterplot of the relationship between phylogenetically independent contrasts in global DNA cytosine methylation (PIC %mC) and log₁₀-transformed genome size [PIC log(C-value)], and least squares-fitted regression line through the origin.

Figure 4

Distribution of species on the plane defined by global DNA cytosine methylation (%mC) and log₁₀-transformed genome size [log(C-value)]. Filled dots indicate, from left to right, the position of Arabidopsis thaliana, Oryza sativa and Zea mays, discussed in the text.