Systems Biology Project

Title Transgenerational effects and evolution

Principal Investigator Professor Stuart Townley

Collaborators Professor RA Johnstone (University of Cambridge), Dr R Hoyle (University of Surrey), Dr J Wells (University College London)

Summary The capacity for parents to induce phenotypic effects in their offspring, with no genetic basis, has been documented in most kingdoms of living organisms. The stimulus that evokes such parental effects may appear only transiently during the parent's lifetime, and may have only a transient impact on the parent, and yet may nevertheless exert a long-lasting influence on the offspring if it is imposed during a critical window of sensitivity in the developmental process. These offspring may then transmit such effects to their own offspring, generating longer-term trans-generational effects. Despite increasing evidence for these effects, why they have evolved, and why they differ so widely across species, remains poorly understood. A key reason for this poor understanding is the lack of a robust theoretical framework for investigating the phenomenon.

Transgenerational effects are particularly relevant to humans, as it is increasingly recognized that some aspects of phenotype are flexible in relation to prevailing conditions, whereas others are strongly influenced by ancestral experience. Processes such as climate change, economic development and migration all challenge organisms, including humans. How different organisms can respond to such challenges is determined in part by the variable contribution of trans-generational effects to phenotypic development. Thus, understanding transgenerational effects has many practical implications, for understanding how organisms will respond to different types of environmental change.

Our research hypothesis is that trans-generational effects are an adaptive consequence of evolved life-history strategies, and that we can therefore predict under what ecological conditions they will arise. We propose to use novel life-history models to explain the diversity of trans-generational effects across species. These models consider how organisms should divide their resources between competing processes such as growth versus reproduction. A 'decision' in the parent generation has implications for the phenotype of the offspring, and hence what 'decision' should be made in this generation. We will use 'evolutionary game theory' in these models, which considers how the best strategy for one individual must take into account the strategies of other individuals, such as their offspring.

In order to build these models, we will need to develop novel mathematical methods for modeling life-history evolution that can better accommodate trans-generational influences. Transgenerational effects pose some particularly intriguing challenges for mathematical modeling, because there are time lags between when information 'enters' phenotype (eg being a first-born), and when natural selection operates (eg on the offspring of that first-born individual). These time-lags are mathematically very complex, and they may be further complicated by asymmetries between the two parents. For example, the ovum develops during the mother's own fetal life, whereas the sperm develops during the father's adolescence. The two parents therefore provide information to the offspring from contrasting time periods, which might in turn influence how the offspring should respond.

We are also interested in how organisms can respond to 'extreme events', such as forest fires or famines. These events occur less often than every generation, hence many generations never actually experience them. We will investigate how such extreme events might contribute to the evolution of trans-generational effects.

Once developed, we will use our model to evaluate our theoretical analyses across different species. We will pay particular attention to humans, benefiting from data on cohort studies already available to the investigators. These cohort studies provide data on 3 or more generations, and will allow us to undertake analyses using insights from the mathematical modeling.

Funding body EPSRC

Timescale 2010 to 2013