Epigenetics and Evolution: Transposons and the Stochastic Epigenetic Modification Model

Andrei S Rodin,Sergio Branciamore,Grigoriy Gogoshin,Arthur D Riggs

doi:10.3934/genet.2015.2.148

Abstract

AbstractIn addition to genetic variation, epigenetic variation and transposons can greatly affect the evolutionary fitnesses landscape and gene expression. Previously we proposed a mathematical treatment of a general epigenetic variation model that we called Stochastic Epigenetic Modification (SEM) model. In this study we follow up with a special case, the Transposon Silencing Model (TSM), with, once again, emphasis on quantitative treatment. We have investigated the evolutionary effects of epigenetic changes due to transposon (T) insertions; in particular, we have considered a typical gene locus A and postulated that (i) the expression level of gene A depends on the epigenetic state (active or inactive) of a cis-located transposon element T, (ii) stochastic variability in the epigenetic silencing of T occurs only in a short window of opportunity during development, (iii) the epigenetic state is then stable during further development, and (iv) the epigenetic memory is fully reset at each generation. We develop the model using two complementary approaches: a standard analytical population genetics framework (diffusion equations) and Monte-Carlo simulations. Both approaches led to similar estimates for the probability of fixation and time of fixation of locus TA with initial frequency P in a randomly mating diploid population of effective size Ne . We have ascertained the effect that ρ, the probability of transposon modification during the developmental window, has on the population (species). One of our principal conclusions is that as ρ increases, the pattern of fixation of the combined TA locus goes from “neutral” to “dominant” to “over-dominant”. We observe that, under realistic values of ρ, epigenetic modifications can provide an efficient mechanism for more rapid fixation of transposons and cis-located gene alleles. The results obtained suggest that epigenetic silencing, even if strictly transient (being reset at each generation), can still have significant macro-evolutionary effects. Importantly, this conclusion also holds for the static fitness landscape. To the best of our knowledge, no previous analytical modeling has treated stochastic epigenetic changes during a window of opportunity.

Highlights

Transposons, which are elements that can self-replicate and move from one chromosomal location to another, are a major source of genetic variation in both animal and plant genomes [1], [2], [3], and, as such, are significant contributors to evolutionary change [4] [5] [6]
Strong evidence is accumulating that epigenetic variation caused by genetic changes, such as the insertion of transposons, can affect expression of nearby genes, and importantly, epigenetic variation and transposons have been linked with human disease [11]
We find that stochastic epigenetic modification (SEM) of a newly inserted transposon during a developmental window can greatly increase both the rate and probability of fixation of the new element, with the advantage provided by Stochastic Epigenetic Modification (SEM) increasing with Ne

Summary

Introduction

Transposons, which are elements that can self-replicate and move from one chromosomal location to another, are a major source of genetic variation in both animal and plant genomes [1], [2], [3], and, as such, are significant contributors to evolutionary change [4] [5] [6]. The demethylation of transposons in the early embryo is incomplete; for mouse and humans 30-40 % methylation remains, increasing to 80 % or more after differentiation [28], [29], [30] While it is not yet certain whether demethylation (and remethylation) of transposon LTRs takes place stochastically and independently on the maternal or paternal genome, as we assume here, it is known that the phenotype of Agouti yellow mice, which is due to the Avy transponson, is variable and consistent with stochastically incomplete and variable demethylation and/or remethylation events at the LTR in the early embryo, before specification of the three primary germ layers [24]

Modeling

Findings

Discussion

Conclusion