02 Abstracts from February
Epigenomics of foetal programming
Date: Tue Aug 29 14:18:16 BST 2006Project Summary: Our study aims to determine (i) whether exposure to protein-restriction in utero causes changes in the mechanisms that control gene expression (ii) if these changes persist into adulthood (iii) if these changes are passed on to the next generation.
Rationale: It is well known that some of the predisposition to complex diseases, such as diabetes, is inherited. However, there is a growing body of evidence to suggest that inherited predisposition does not always reside in the inheritance of disease-predisposing genes. Rather, an adverse in utero environment can alter the susceptibility to developing the disease in future generations. For example, there is strong epidemiological evidence that maternal protein-restriction during pregnancy can contribute to diabetes risk in the offspring. Such environmental influences can alter hormonal levels in utero, resulting in changes in the function of various genes in the foetus that persist into adulthood. Even more worryingly, it seems that these changes in gene function can be transmitted to future generations, that is, in utero grandparental nutrition can influence the health of the grandchildren. Clearly, understanding the molecular mechanisms that underlie these phenomena, and why they are inherited, would provide new avenues for the treatment and prevention of various complex diseases.
Scientific unknowns: We would like to address the fundamental question of how the changes in gene function persist into adulthood, and potentially future generations. The cells in the affected individual must retain a ‘memory’ of the adverse in utero environment, but this ‘memory’ has so far proved to be enigmatic. Several lines of evidence now suggest that the ‘memory’ resides in epigenetic modifications of the DNA. Epigenetic modifications, such as DNA methylation, control gene expression, and are therefore indispensable for normal gene function. Two salient features of epigenetic modifications are that they do not change the underlying DNA sequence, and they are potentially reversible. It is thought that upon exposure to an adverse in utero environment, altered hormonal levels induce aberrant epigenetic changes (epimutations) in the foetus. These epimutations are propagated throughout the life of the individual, hence becoming the ‘memory’ of the adverse in utero environment, and eventually predisposing the individual to diseases. These epimutations are also likely to be the memory that is passed on to future generations. However, despite the potentially important role epimutations could play in the inherited predisposition to diabetes, we are largely unaware of the frequency, genomic location, or the potential for inheritance of in utero environment- induced epimutations. In large part, this has been due to technical limitations. Because of the functional genomics tools that have been developed in the last few years, we are in a unique position to lead the first systematic genome wide study of in utero environment-induced can induce epimutations.
Why animals have to be used and human or tissue studies or computer simulations could not be used: The origins of inherited predisposition in mammals can only be studied in live, breeding animals (and breeding is not a procedure that can be refined or replaced as yet). Performing controlled experiments on the transgenerational effects of diet on epimutations would not be possible in humans, and other systems such as cell lines would be irrelevant as they do not reproduce sexually. Therefore an appropriate mammalian model organism is required. Mice have been extensively used for studying the effects of in utero dietary influences (and hence there are many well established protocols). We would like to study epigenetic changes induced by a protein-restricted in utero diet to mimic the effects of poor nutrition (which is generally characterized by deficient protein levels) in humans.
General sequence of the work and how different options, including non-animal studies, contribute (e.g. cell culture studies before/alongside animal ones, pilot experiments precede definitive): Each experiment starts by breeding two adult female mice with a single adult male (termed the FO generation). Immediately after coitus, pregnant females will be separated from the male and kept in separate cages. One female (control) will be given a normal protein diet during pregnancy, and the other female will be given a protein-restricted diet (8% protein but isocaloric to the normal 23% protein diet). The protein-restricted diet (PRD) will continue till weaning of the offspring (termed the Fl generation). In fact, from this point onwards in the experiment, all mice in all subsequent generations will be given a normal protein diet. We will then breed one Fl male and one Fl female (chosen randomly) with normal mice to generate F2 mice. Before setting up breeding pairs from the F2 mice and carrying on with the experiment, we will analyse DNA methylation in the Fl mice (post-mortem). Breeding pairs from the F2 mice will be set up only if we find epigenetic changes. If epigenetic alterations are found in the Fl generation, one F2 male and one F2 female (chosen randomly from each litter) will be mated with normal mice to generate F3 mice. The experiment will terminate at this generation. We will then analyse DNA methylation at regions initially identified in the Fl generation. All mice will be sacrificed at 16-18 weeks of age (but not older than 24 weeks) according to Home Office recommendations.
How animal suffering will be kept to a minimum, strategies used to minimise adverse effects and numbers, and why the protocols, and the way they will be carried out, will involve the least suffering: Numbers used will be kept to an absolute minimum by strict breeding control and monitoring of the animals. They will have constant access to food, water and environmental enrichment, and the only procedure carried out on live animals will be monitoring their weight. All tissue samples will be taken post-mortem. Each mouse will be allowed to breed once (to continue the genetic line), allowed to maintain a good quality of life with monitoring for weight gain and general health, access to veterinary care if necessary (although this is not foreseen), environmental enrichment and easy access to food and water. The mice will be culled at 16-18 weeks (not older than 24 weeks) using home office approved humane methods.
Why the species or types of animals were chosen: C57/BL6J and 129Rr/J are two of the most commonly used inbred strains. They are ideal for our proposed study as there are several dietary protocols developed for that can be readily adapted to our study. Furthermore, the genomes have been sequenced, which is extremely important for our genome-wide scan for epimutations.
Estimated numbers of mice to be used, and what will happen to them in general terms (such as typical experience and likely range of adverse effects): The maximum number of mice we expect to use during the course of our study is —600. The only animals given low protein diet are those during the first breeding cycle (so-called FO); all subsequent generations are maintained on a normal diet. The only other procedure carried out on live animals will be monitoring their weight post weaning and at death.
Statement of benefits/advances from achieving the project’s objectives and science will advance: Specifically, we expect to identify regions of the mouse genome at which the epigenetic state is altered due to low protein levels in utero, and determine whether these changes in DNA are transmitted to future generations. The potential benefits of our study are:
1. There is increasing evidence that adverse in utero environments can predispose to a host of diseases (such as diabetes). Our study will elucidate some of the key molecular mechanisms and pathways that underlie the effects of an adverse in utero environment.
2. The genomic regions identified in this study can form the basis of future animal and human studies that focus on the effects of in utero protein-restriction – a common phenomenon in developing countries. Therefore, we expect our study to be a paradigm for future studies in this important field.
3. Because epimutations do not change the DNA sequence, and are potentially reversible, the study we propose is even more appealing for potential future therapies that would target the epimutations.