Parental effects occur when the phenotype of one or both parent(s) affects the phenotype of offspring beyond direct effects of genetic inheritance. Effects can be maladaptive, for instance when offspring lifespan and/or reproductive success are reduced because of transgenerational senescence, or adaptive, for example when parasite exposure of parents primes the immune responsiveness of offspring. From a mechanistic perspective, the question which processes underlie transgenerational inheritance remains open. Moreover, from an evolutionary point of view, the fitness consequences of transgenerational inheritance and its impact on adaptation to changing environments are also mostly unknown. In this meeting, we aim to discuss examples of transgenerational inheritance in natural populations of non-model organisms (plenary 1, session 2), potential mechanisms underlying such patterns (sessions 3 and 4) as well as its evolutionary underpinnings and consequences (sessions 5 and 6).
Confirmed speakers are Pat Monaghan, Olivia Roth, Virpi Lummaa, Bram Kuijper, Alexei Maklakov, Eric Miska, Frank Johannes, and Tobias Uller.
Organizers: Britta Meyer (MPI, Plön), Miriam Liedvogel (MPI, Plön), Melanie Heckwolf (GEOMAR, Kiel), Sandra Bouwhuis (Institute for Avian Research, Wilhelmshaven)
TransGenIn@evolbio.mpg.de
Birds, in comparison to mammals, are relatively long lived for their size and were once thought to largely escape senescence. As longitudinal studies have accumulated, it has, however, become clear that many avian traits show signs of late-life deterioration. In addition, there is increasing evidence for transgenerational effects of parental age. Using individual-based data from a large colony of a long-lived seabird, the common tern Sterna hirundo, it was found that recruited daughters from older mothers suffer from reduced annual reproductive success. Additionally, recruited sons from older fathers suffer from reduced life span and both effects translate to reductions in offspring lifetime reproductive success. In this talk, I will describe these sex-specific parental age effects, and introduce current work on telomere length, de novo mutation and DNA methylation to study the potential mechanisms underlying them.
Using >2000 captive zebra finches that we have detailed information on early developmental stress experienced by these birds as well as their parents and grandparents, we quantify direct as well as trans-generational effects of early stress on multiple morphological and fitness-related traits. We meta-summarize the effects across multiple stressors and dependent traits in order to generate unbiased conclusions regarding average effect sizes that one can expect to find. Out of six parameters that quantify early stress, we find that nestling body mass measured at 8-day of age (md8) to be the most powerful measurement of early nutrition. Despite ad-libitum food available to parents, md8 varied up to six-fold (ranging 2-12 gram), illustrating that some nestlings were seriously undernourished. Despite such drastic variation, an individual’s md8 had only modest effects on its morphology as an adult (mean r = 0.20) and even less pronounced effects on lifespan and reproductive performance (mean r = 0.07). Overall, there were no trans-generational effects of early stress experienced by parental and grandparental generations, on an individual’s traits (upper boundary of 95%CI of mean effect was below r=0.028). However, the md8 of the mother still had a significant, yet weak effect on daughter fecundity (mean r =0.13). Our results suggest that morphology (body size) may be more sensitive to early developmental stress, whereas reproductive performance (fitness) is relatively resilient to episodes of undernourishment. Our findings also suggest that detrimental effects of early food stress are unlikely to be carried on over multiple generations in captive zebra finches.
Epigenetic changes function as flexible mechanisms increasing adaptability within a generation and transmitting the environmental information to subsequent generations in vertebrates. In times of global warming and changing vegetation, we aimed to study the epigenetic response of a wild mammal species, the wild guinea pig, to an increase in temperature (of 10°C more than ambient) as well as to a diet alteration (of 40% less protein). We measured the direct epigenetic response in male wild guinea pigs and also studied whether, and if so to what degree that response was also transmitted to their sons. We detected transmission of epigenetic information from fathers to their sons. Furthermore, we detected DNA methylation changes of two types, a general epigenetic response (reflecting that the environment had changed) and a specific epigenetic response which occurred only for one environmental factor (reflecting, what factor of the environment had changed) highly targeted at specific genes.
The social environment can affect the phenotype of breeding females and can have transgenerational effects on the phenotypes of their future offspring. Using Japanese quail (Coturnix japonica), we studied maternal (P0) effects of the social environment on female offspring (F1) housed under matched and mismatched conditions with respect to group size (pairs of two vs. groups of four birds). The P0 and F1 social environments had an interactive effect on F1 female body mass, but not on other traits (circulating corticosterone and androgen levels or reproduction). Daughters of pair-housed P0 females became heaviest when housed in groups (mismatching conditions) themselves. Furthermore, there was a main effect of the F1 females own social environment on growth: F1 females housed in groups grew faster than F1 females housed in pairs, irrespective of the social environment of their P0 mothers. In addition to growing faster and ending up heavier than F1 pair-housed females, F1 group-housed females laid heavier eggs with a higher hatching success and produced heavier F2 offspring. Although the fitness consequences of these observed maternal effects still require further study in order to understand their adaptive significance and underlying physiological mechanisms, our results show that maternal effects can act in a context-dependent way. We emphasise the importance of taking such context-dependent maternal effects into account when investigating the potential adaptive benefits of transgenerational effects.
see attachments, 8 posters, 3 min resp.
The impacts of climate change on marine organisms will depend on their ability to tolerate, acclimate and eventually adapt to the environmental change. We performed unique transgenerational experiments to determine the molecular response of a coral reef fish to ocean acidification. Our complex design allows us to: 1) investigate the importance of parental exposure to ocean acidification conditions and tease apart within- and across-generation specific reactions; 2) evaluate the influence of the parental behavioural phenotype on the offspring response and 3) unveil tissue-specific transcriptional programs that might control the transgenerational response to ocean acidification. Results from brain transcriptomes reveal large differences between within and across generations with altered gene expression for the majority of within-generation responses returning to baseline levels following parental exposure to elevated CO2 conditions. Furthermore, the transcriptomic responses of offspring depended largely on the behavioural phenotype of the parents, revealing a complex interaction between parental phenotype and parental exposure for the transgenerational signal in the brain of the offspring. Transgenerational reprogramming, however, shows tissue-specificity. We were able to evaluate the influence of parental phenotype and parental exposure as well as draw a whole-organism picture of the transgenerational signal in a coral reef fish as a response to near-future predicted ocean acidification conditions.
For a number of years we have been using a coral reef damselfish, Acanthochomis polyacanthus, to investigate the influence of thermal conditions experienced over multiple generations on performance. As the project’s focus was to understand transgenerational effects as a means for marine fish to acclimate and adapt to rapid environmental change, two future projected increases of +1.5°C and +3°C were used. Fish were maintained in these elevated conditions for three generations and during this time phenotypic traits were measured as well as shifts in gene expression patterns. A number of interesting patterns were found with the temperature experienced by the current, parent or grandparent generation affecting the phenotype of fish, with more gradual warming over generations resulting in greater plasticity allowing developmental plasticity occurring on top of transgenerational plasticity. In addition, we also found that the thermal conditions in which reproduction occurred interacted with thermal history to effect the phenotype of offspring produced. Finally, certain traits exhibited rapid plasticity consistently occurring within a generation, while others required the previous generation(s) to have experienced warmer conditions for phenotypic change to arise.
Extreme weather events like heatwaves, commonly defined as periods of five successive days where thermal maximums are 5oC greater than the long-term average, are predicted to be one of the most probable and hazardous challenges we face in a warming world. While natural populations have been showing climate related stress responses and extinctions for decades, the proximate mechanisms behind such responses are poorly understood. Male-specific infertility due to heat-stress is recently being recognised in ectotherms, causing declines in testes size, sperm counts, sperm viability, female sperm storage, and subsequent reproductive fitness. However, current research often does not consider the long-term consequences. We show, using a beetle (Tribolium castaneum), that heatwave simulations cause negative transgenerational effects, primarily via sperm, for offspring longevity and sons’ reproductive fitness.1. However, such damage may be overestimated, if there is potential for populations to adapt to elevated mean temperatures associated with climate change via evolution and/or acclimation. Here, we tested the potential benefit of adaptation to increased mean temperatures for heatwave tolerance by experimentally evolving T. castaneum populations in divergent thermal regimes. Relative to 30oC-adapted males, those from populations maintained at 38oC for 25 generations displayed a relative improvement in mean post-heatwave survival by 75% and post-heatwave reproductive fitness by 55%. Moreover, when in competition with 30oC-adapted males, 38oC-adapted males gained a 31% higher paternity share when competing in their local 38oC thermal regime. Unexpectedly, the 38oC-adapted male post-heatwave performance was best when they were developmentally acclimated to 30oC for a generation, suggesting that although transgenerational adaptation to warmer conditions is possible, it could exert substantial metabolic costs. Ultimately, we show that temperature extremes carry detrimental transgenerational effects while moderate increases can provide positive transgenerational effects relatively fast. The results begin to address the knowledge gap for the potential of biological traits to adapt to climate change. Future work aims to expand thermal scenarios and elucidate the mechanisms of inheritance and thermal adaptation. 1. Sales, K., Vasudeva, R., Dickinson, D.E., Godwin, J.L., Lumley, A.J., Michalczyk, L., Hebberecht, L., Thomas, P., Franco, A., Gage, M.J.G. 2018. Experimental heatwaves compromise sperm function and cause transgenerational damage in a model insect. Nature Comms, 9:4771.
Since August Weismann (1834-1914) formulated the distinction between innate and acquired characteristics at the end of the 19th century, the debate relating to the inheritance of acquired traits has raised many controversies in the scientific community. Following convincing arguments against (e.g. William Bateson) this debate was then set aside by the majority of the scientific community. However, a number of epigenetic phenomena involving RNA, histone modification or DNA methylation in many organisms have renewed interest in this area. Transgenerational effects likely have wide-ranging implications for human health, biological adaptation and evolution, however their mechanism and biology remain poorly understood. We recently demonstrated that a germline nuclear small RNA/chromatin pathway can maintain epi-allelic inheritance for many generations in C. elegans. We will discuss evidence for such inheritance in an emerging vertebrate model, the cichlid fish of Malawi.
Non-genetic inheritance (NGI) deals with inheritance mechanisms apart from DNA. Candidates for germ-cell mediated information transfer between generations are modified histones, DNA methylation, and small RNAs. These mechanisms are evolutionarily conserved. In addition, the heritability of environmental information in plants and animals has been demonstrated. Together, this has inspired a lively debate on the relevance of NGI for human health and evolutionary processes. However, answers universally satisfactory for medicine, ecology, and evolutionary biology have not yet been derived. Therefore, evolutionary biologists, ecologists, and molecular epigeneticists have joined forces to compile fundamental mechanistic knowledge on NGI, to derive conclusions on the role of NGI, and to develop recommendations for experiments and models. We find that knowledge of fundamental properties of the molecular machinery raises critical questions on the relevance of epialleles and suggests the introduction of an “information cloud” view. It also questions the relevance of long-term inheritance of a particular epigenetic mark and suggests the adoption of a “relay race” view of multiple pathways. A mechanistic approach reveals an inseparable entanglement of genetic and non-genetic aspects and thus resolves a point of debate in evolutionary biology: NGI is inextricably genetic and non-genetic at the same time. Finally, the study of molecular mechanisms reveals comprehensive species-specificity. This makes a unifying answer on relevance of non-genetic inheritance phenomena unlikely, and suggests a thorough reconsideration of the choice of animal models. Our findings have clear implications for experimental and modeling approaches in ecology and evolution. We assume that the inclusion of a mechanistic perspective will make both approaches more powerful and meaningful, and provide precise recommendations on how to proceed.
In insects, increasing evidence shows that epigenetic changes can be passed on to future generations. This is also true for diapause, an environmentally induced state of ‘hibernation’ that is characterised by arrested development and suppressed metabolism. In Lepidoptera there is a strong paternal influence on the diapause decision of the offspring. However, not much is known about the epigenetic control of diapause. In this study we characterized differences in whole-genome methylation (whole-genome bisulfite sequencing) between diapause-destined and directly developing siblings of the butterfly Pieris napi. We reveal marked changes in a number of genes, thereby shedding a first light on the epigenetic control of diapause in this species. Transgenerational inheritance can have a substantial effect on the evolution of diapause, possibly expediting selection on a recurring phenotype.
Our climate is changing rapidly and organisms must either move, adapt or acclimate in order to persist. Epigenetic modifications such as DNA methylation may be an important mechanism underlying fast, adaptive responses, as they can contribute to heritable changes within populations and drive rapid evolution. Crucial during gametogenesis and development (embryogenesis), epigenetic modifications can be inherited both mitotically within a generation, and/or meiotically across generations (transgenerational epigenetic inheritance). However, heritable epigenetic modifications must overcome two reprogramming phases, once in the germline and once in the early embryo. Despite being well documented in plant and mammalian systems, epigenetic reprogramming studies during fish gametogenesis and embryogenesis have focused only on a few model species. Here, we investigated DNA methylation patterns, together with the molecular characterization and mRNA expression profiles of DNA-methyltransferase enzymes (DNMT) during gametogenesis and embryogenesis of marine threespine stickleback. Additionally, the impact of multiple simulated ocean warming scenarios (ambient °C, +1.5°C and +4°C) was evaluated at the epigenetic and phenotypic level to establish a link between environment, epigenetic reprogramming and transgenerational plasticity. So far, we found that DNA methylation seems to be stable during gametogenesis, with male gonads being hypermethylated compared to female gonads. Nevertheless, environmental temperature significantly influenced mature male gonad DNA methylation, whereas for females, only egg size/number was affected, in line with previous work showing that cytoplasmic factors (e.g. mitochondria) are likely involved in transgenerational inheritance down the maternal line. Overall, our preliminary results suggest that gametogenesis in stickleback is differentially regulated compared to mammals and is sensitive to environmental perturbations, with possible implications for adaptive windows under climate change.
It is now well recognised that the early environment is not simply permissive of development, but can also shape the phenotype in ways that have long term consequences for performance and fitness. There are many routes, both direct and indirect, whereby such environmental effects on phenotypic development can come about and span generations. Of considerable importance in this context is parental state, which can influence offspring via effects operating at many stages, from parents’ germ cells through to the developmental and growth environment parents provide for their offspring. Offspring growth, development and life history can then be altered in adaptive and non-adaptive ways. In this talk, I will focus particularly on effects of parental age and of early exposure to stress, discuss the consequences over varying time scales and whether these appear to be adaptive or not. I will present illustrative data from unmanipulated natural populations, and from a range of taxa in which conditions have been experimentally manipulated in both the lab and the field. I will examine some physiological and molecular mechanisms that can mediate effects that can occur over relatively long time scales, including changes in stress sensitivity and telomere dynamics.
It is well-established that the yolk of eggs of birds contains androgens like testosterone. These maternally derived hormones have been found to influence the behaviour and growth of the chicks hatching from these eggs. Hence, maternal yolk hormones have been hypothesized to act as adaptive maternal effects shaping the offspring’s phenotype, particularly in relation to the rearing environment. However, the evidence for this idea is mixed. We collected a large data set in wild blue tits (Cyanistes caeruleus; >150 broods) to test three premises underlying the hypothesis of adaptive transgenerational shaping of the offspring phenotype. We investigated (1) whether there was between-clutch variation in yolk testosterone levels; (2) whether such variation related to characteristics of the female or her environment; and, finally, (3) whether yolk testosterone levels related to offspring fitness. Importantly, in order to disentangle maternal hormone-related and environment-related variation in offspring fitness we carried out a large-scale cross-fostering experiment. We found substantial between-clutch variation in yolk testosterone, which was partly explained by the female’s age and lay date, but not related to offspring growth or survival, providing only limited support for the adaptive significance of maternal yolk androgens.
The phenotype of an animal cannot be explained entirely by its genes. Among the factors, other than genome, which contribute to the development and homeostasis of multicellular animals, the microbial associated community can be rapidly modified by environmental cues and may represent a mechanism for rapid acclimation and adaptation of the host to a changing environment. The Starlet Sea Anemone, Nematostella vectensis, occurs in brackish habitats along a geographic range which spans a pronounced thermocline, appearing to have resulted in different thermal tolerance, optima for growth and unique microbial communities, suggesting potential co-evolution of the holobiont in response to environmental variation. N vectensis is a powerful invertebrate model for molecular research due to: (1) the availability of a well-annotated genome, (2) an extensive molecular toolkit including well-developed protocols for gene suppression and overexpression as well as transgenesis and (3) the ability to procure all developmental stages on a weekly basis. The aim of the project is to determine if microbes specific to anemones contribute to thermal physiology through microbial transplantation experiments. We could proof, that bacteria change according to long-term temperature acclimation and that acclimated bacteria get maternally transmitted. Currently, we are transplanting microbial communities from animals acclimated to different to germ-free animals of the same strain. Animals, recolonized with the transplanted microbiota will be tested for the thermal tolerance and resistance. In addition, we are analyzing host gene expression and epigenetic changes in the genome. The outcomes from our research will allow better predictions about the ability of marine organisms to acclimate and potentially adapt to a rapidly changing environment.
Phenotypic changes in response to environmental influences can be transferred from one generation to the next with the potential to boosting offspring performance. This also applies to parental parasitic experience that can influence offspring immune responses, a phenomenon known as transgenerational immune priming (TGIP), which is often coupled with a vertical transfer of microbes. While in many systems components of the maternal immune system are directly deposited in the egg, in systems with viviparity, transfer of immunological experience may also occur during pregnancy. In conventional sex role systems, both TGIP and transfer of microbes are thus maternal traits, where influences of transfer via egg and during parental investment are hard to disentangle. However, sex roles are reversed as in the pipefish Syngnathus typhle with its unique male pregnancy, the maternal transfer of immunological and microbial experience via eggs and the paternal immunological contribution during pregnancy can be disentangled. This permits to study and manipulate maternal and paternal transfer separately, addressing how maternal and paternal or biparental parasitic experiences influence the maturation of the offspring immune system.
In a scenario of environmental change, however, TGIP should not be studied in isolation, as several performance-related environmental factors are changing simultaneously. This lowers the predictability of offspring environmental conditions, potentially hampering the benefits of trans-generational plasticity. This evokes the question of how the combination of an abiotic (e.g. parental immunological experience) and a biotic environmental factor plays out as trans-generational effect in the offspring generation. Our data of a dominating abiotic effect over the biotic TGIP effect suggest that the available resources that can be allocated to transgenerational plasticity are limited. This could ask for a reassessment of TGP as short-term option to buffer environmental variation in the light of climate change.
Environments vary over time and if this variation is predictable, environments that are similar across generations should favour evolution of anticipatory parental effects to benefit offspring. In contrast, the absence of correlation between parental and offspring environments should select against parental effects. However, experimental evidence is scarce. We investigated the evolution of maternal effects using experimental evolution. Populations of the nematode Caenorhabditis remanei, adapted to 20°C, were exposed to a novel temperature (25°C) for 30 generations with either positive or zero correlation between parent and offspring temperature. We found that populations evolving in environments with positive correlations had a positive maternal effect, since they required maternal exposure to 25°C to achieve maximum reproduction and fitness in 25°C. In contrast, populations evolving under zero correlation had lost this positive maternal effect. This shows that parental effects can aid population viability in warming environments. Correspondingly, ill-fitting parental effects can be rapidly lost.
The capacity of different species to respond to a given selective environment often varies. While nucleotide diversity has long been recognised as the sole predictor of species and populations potential to adapt, current evidence suggest that populations or species may compensate for low genetic diversity through plastic responses. A possible heritable non-genetic mechanism underlying plasticity is DNA-methylation. Using two coexisting and phylogenetically related fish species, the three-spined and the nine-spined stickleback, we examined the contribution of DNA methylation-based epigenetic variation to local adaptation. To do so, we investigated natural populations of the two species from adjacent brackish water and freshwater habitats and compared the structure and variation of DNA methylation among populations and species. Furthermore, we assessed the relationship between epigenetic variation and habitat and screened for functional genes affected by DNA methylation changes. Here we will present our first findings.
Co-Authors: Gillian Durieux, Miriam Liedvogel
Bird Migratory behaviour depend on adaptations optimized for flight, timing, navigation and fuel allocation. The heritability of these adaptations has been shown with classical approaches of crossbreeding experiments suggesting that the genetics of migration rely on few genes with large effects. However, studies with candidate genes and population genomics are not conclusive about the identity of genes or molecular mechanisms related with bird migration. Demography and other evolutionary processes difficult the assessment of findings based on purely DNA sequences and therefore, identification of genes related to migration. Gene expression and gene regulation (mediated mainly by chromatin dynamics like chromatin accessibility), improve the limitation features of other approaches to identify functional elements related migration. We have designed a common garden experiment with populations of European blackcaps (Sylvia atricapilla) to sample gene expression and regulation patterns in brain areas relevant to seasonal migration. We focus on i) the suprachiasmatic nucleus (SCN), a brain region controlling circadian and circannual regulatory processes, ii) hippocampus a brain region involved in spatial navigation, and iii) cluster N, a brain region involved in magnetic compass orientation. We used ATAC-seq to compare chromatin accessibility in these three brain areas of birds on migratory season and birds off migratory season. Our preliminary results show differences in chromatin accessibility mainly between migratory and non-migratory birds. Few differences are present between brain regions. The chromatin accessible regions are located in non-coding regions like promoter sequences and other potential cis-regulatory sequences. Such differences are close to genes related with cell migration, neurotransmitter activity among other functions. These results open the possibility for a more complex picture than the one suggested earlier for the genetics of migration.
There is a bewildering diversity of perspectives on parental effects. My aim in this talk is to make sense of this diversity by drawing attention to the organising role of problem agendas in evolutionary biology. I discuss and exemplify how different evolutionary problems motivate different ways to represent the relationship between parents and offspring. These representations in turn influence what aspects of biology that are considered to be relevant to evolution. While the research programs that develop around problem agendas generally happily co-exist, I use the contemporary debates over the explanatory role of ‘non-genetic inheritance’ in evolution to illustrate how different perspectives sometimes cause (unnecessary) scientific controversy.
When environments are unstable, genetic parental information is often not sufficient for offspring to become optimally adapted to the environment and parental effects may be necessary to prepare offspring for the environment they are likely to experience. Seasonally distinct life history patterns over sequential cohorts in short-lived animals are one example in which cues of the early environment either indirectly through parents or by direct experience shape phenotypes across generations. In response to predictably different ecological conditions, individuals born in different seasons often differ consistently in behaviour and physiology throughout life, often with consequences for fitness. Wild cavies (Cavia aperea) born into spring or autumn differ in behavioural traits such as risk-taking or stress-coping and in physiological underpinnings such as hormone levels or metabolic rates throughout life. In a series of experiments I tested by which mechanisms (maternal hormones and epigenetic effects, maternal behaviour or direct environmental cues) these seasonal differences are shaped and if they impose fitness consequences in matching versus non-matching environmental conditions later in life. While maternal behaviour has only limited influence in this precocial species, direct environmental cues and differences in hormone concentrations in early life shape permanent differences between seasonal cohorts that translate to the onset of reproduction as well as investment into offspring in adult life. Reproduction under mismatching conditions resulted in elevated maternal metabolic rates and reduced fecundity. Taken together, these results suggest that seasonal programming of offspring phenotypes is 1) a predictable form of transgenerational plasticity and 2) it provides an adaptive mechanism to adjust offspring to the environment they encounter after birth.
In light of recent climate change phenotypic plasticity has received ample attention. Not only can plasticity evolve, it can also affect evolution by tinkering with the raw material for selection (phenotypes). Bet-hedging, however, is rarely considered in this context, despite being tightly linked to trans-generational plasticity. We argue that the neglect of bet-hedging is due to synonymizing only plasticity with phenotypic variance and GxE interactions, so we strive for clearer terminology which separates reaction norms from evolutionary outcomes. We describe logistic reaction norms by their mean, the degree of phenotypic variance and on how variance is allocated within and among environments. We then define six emerging strategies as extremes of these three continuous axes: conservative bet-hedging vs. arithmetic mean optimization; fixed vs. variable development; and plasticity vs. diversified bet-hedging. Transgenerational plasticity and diversified bet-hedging are thus mutually exclusive and arise from opposing changes in reaction norms, but both strategies increase phenotypic variance and may alter the course of evolution. We predicted that the position along the three strategy axes depends on environmental conditions. We thus conducted a meta-analysis on 57 studies of insect diapause and extracted 447 logistic reaction norms (including trans-generational plasticity as well as plasticity across life history stages), and correlated means and variance components with climate parameters. Three patterns emerged: First, mean diapause correlated with mean winter onset. Secondly, the degree of phenotypic variance did not depend on the amplitude of environmental change, likely caused by strong study bias towards non-canalized genotypes. Thirdly, bet-hedging was rare and correlated only weakly with environmental predictability. Genotypes are thus most vulnerable to decreases in climate predictability.
Natural selection acts upon heritable variation in fitness-related traits. In principle, epigenetic, non-DNA sequence based inheritance can potentially contribute to adaptation. Whether this is the case is largely unknown. To address this, we placed a URA3 reporter gene at different positions within subtelomeric, chromatin-silenced regions in the otherwise isogenic Sacharomyces cerevisiae strains, causing variable degrees of URA3 repression. We placed populations of cells under negative URA3 selection to determine mechanisms of adaptation and the role of heritable gene silencing in this process. We show that populations in which heritable gene silencing occurs have a higher rate of survival and an increased probability of producing genetic mutations. Adaptive mutations appear and fix in populations faster in strains with intermediate level of epigenetic silencing, suggesting that there is an optimal frequency of epigenetic switching that enables genetic assimilation. Genome sequencing of evolved strains indicates that epigenetic silencing of genes under selection allow cells to adapt, not only by inactivation of the URA3 pathway, but also by changing gene expression or by changing the rate of epigenetic switching itself. This work, for the first time, experimentally demonstrates the impact and mechanisms of how epigenetic forms of inheritance shape the evolutionary outcome of a population.
Epigenetic modifications, including DNA methylation, have a major role in gene regulation and in the preservation of genome integrity. In plants, aberrant gains or losses of DNA methylation (i.e. ‘epimutations’) are sometimes stably inherited to subsequent generations, independently of DNA sequence changes. Here I highlight our ongoing efforts to quantify the stability and phenotypic impact of such heritable epigenetic changes. Specifically, I show that experimentally-induced epimutations can generate heritable variation in a range of complex traits and contribute to heterotic phenotypes in F1 hybrids. I also present evidence that heritable epimutations in natural settings often arise from stochastic failures to maintain DNA methylation during somatic development. The high rate and regularity of these stochastic events define a ‘fast-ticking’ molecular clock, which can be used as a tool to study somatic as well as evolutionary phylogenies.