No description available.

Dietary restriction (DR) can increase lifespan across evolutionarily distinct species, from yeast to rodents. The NAD+-dependent (class III) histone deacetylase SIRT1 in mammals, and its ortholog in other species, may play a major role in this response, but may affect ‘healthspan’ (number of years of good health), rather than lifespan per se. Ageing is accompanied by changes in genome methylation, which may be causal in the ageing process. Since histones are one of the many substrates that are deacetylated by SIRT1, we hypothesised that epigenetic effects of SIRT1 activity – and specifically effects on DNA methylation, which is associated closely with histone acetylation – mediate some of the beneficial effects of DR that contribute to increased healthspan. We also propose that dietary polyphenols may act at the cellular level in a similar way. To test this hypothesis, we first investigated effects of altering SIRT1 expression, by overexpression of a transgene or siRNA-mediated knockdown, on global DNA methylation (methylation of the LINE-1 element) in the human intestinal cell line, Caco-2. We also measured effects on global DNA methylation of dietary isoflavones and resveratrol, under control conditions as well as conditions of SIRT1 overexpression. Measured effects on global DNA methylation revealed possible complex interactions between SIRT1 and these dietary polyphenols but were dependent on the assay used to measure methylation at the LINE-1 element (COBRA or pyrosequencing), so were not considered to be robust observations. We investigated factors that may affect SIRT1 expression – specifically we examined SIRT1 promoter activity in response to polyphenols, effects of promoter methylation and effects of age. Treatment of Caco-2 cells with dietary polyphenols had no effect on the SIRT1 promoter in a promoter-reporter construct. In contrast, methylation of the SIRT1 promoter reduced reporter gene expression in this model. Age did not appear to change the levels of SIRT1 protein expressed in mouse intestinal tissue when comparing young and older mice. An in silico analysis was carried out to investigate if overlaps between groups of genes compiled from published and publically-available data found to i) associate with SIRT1, ii) show altered expression in response to DR, and iii) show altered methylation with ageing were greater than expected by chance, which would support the hypothesis, and provided targets to investigate possible site-specific effects of SIRT1 on DNA methylation. Ten genes were found to fit into the ‘three way’ overlap, which was statistically greater than expected by chance. Pyrosequencing assays could be optimised for only 8 of these 10 genes, so we focused further investigations on this sub-set. Significant effects of SIRT1 overexpression and/or knockdown on methylation were observed on at least one CpG site in the promoters of six of these genes (CDC7, EIF5, IRX3, KLF3, PTPRG, TBX3) and expression at the mRNA level of all of the eight genes (also PCYT1A and SLC39A4) was affected significantly. To gain a more comprehensive view of the extent to which DNA methylation at specific loci may be affected by SIRT1 expression levels microarray-based analysis of the methylation pattern across the genome in Caco-2 cells was carried out under conditions where SIRT1 was overexpressed or where expression was reduced by siRNA. In parallel, we measured the response to expressing SIRT1 at different levels at the level of the transcriptome. Overlaps that were statistically greater than expected by chance between these data and the lists of genes compiled from published and publically-available data used for the in silico analysis were found to exist between the complied list of genes reported to respond to dietary restriction and the set of genes we found to show altered expression in response to changing the level of SIRT1 expression and the set of genes we found to be differentially-methylated in response to reducing the level of SIRT1 expression. This observation is in broad support of our overarching hypothesis. The findings of this study indicate that effects of SIRT1 on methylation of specific genes may correspond with altered expression under conditions of dietary restriction. The data reveal a large number of gene targets for which causal links between these modifications roles in modulating the ageing process could be investigated.

There is little doubt today that ageing is a partially inherited characteristic with the environment playing an equally important role. In this project our aim was to elucidate the gene-gene and gene-environment interactions relevant to ageing through the use of theoretical models and the evolutionary theories of ageing. With our first model we establish for the first time the plausibility of an immunogenetic trade-off between reproduction and survival under infection pressure from the environment while making detailed predictions for the expected point of balance in two different countries, one developing and the other developed, together with predictions about the surprising speed with which an evolutionary transition between the two states can occur. In our second model we develop a detailed simulation program based on epidemiological studies to account for the action of the apolipoprotein gene in western populations, its association with lifestyle parameters, and its evolution over the last 2 million years. We suggest a two-stage history for evolution of Apo E where the establishment of the E3 allele took place during the shift of humanoids to a meat-based diet and the £2 allele started to appear slowly as a rare mutation. Later, with the spread of agriculture and the increasing longevity of humans, the alleles began to be selected more and more towards their current frequencies. Finally, we show how a combination of socioeconomic factors and the stochasticity of mortality can be the driving forces behind the heterogeneity seen in human populations today and reveal the key factors generating this heterogeneity.

Les théories évolutives du vieillissement et la théorie du «disposable soma» en particulier ont été la base théorique d'une avance récente de recherche sur le vieillissement animal. Pourtant, leur hypothèse centrale sur la physiologie de l'entretien et de la réparation cellulaires n'a pas été testée empiriquement. Dans cette thèse, j'ai analysé la physiologie du vieillissement de Escherichia coli sous restriction de carbone, en tant que système modèle pour valider empiriquement les théories évolutives du vieillissement. Les outils microfluidiques sont utilisés pour isoler de larges populations de cellules isolées de E. coli et pour obtenir une restriction carbonée homogène. Malgré le partage de la même génétique et des conditions environnementales, les cellules individuelles de la population présentent des variations significatives de la durée de vie et de cause de décès. Les distributions de durée de vie présentent des caractéristiques typiques du processus de vieillissement, souvent observées en études démographiques animales et humaines. Le taux de vieillissement peut être modifié par des mutations de la réponse générale au stress. Comme la longévité induite par la restriction calorique, la réponse générale au stress prolonge la durée de vie d'E.coli en atténuant l'effet du vieillissement au détriment des besoins immédiats des cellules. Un modèle quantitatif de ce compromis physiologique est construit et correctement prédit des observations expérimentales. En conclusion, je confirme la théorie du «disposable soma» du vieillissement avec les détails physiologiques du vieillissement de E.coli en famine. / The evolutionary theories of aging and the disposable soma theory in particular, have been the theoretical basis for a recent surge of animal aging research. Yet their central assumption about the physiology of cellular maintenance and repair has not been empirically tested. In this thesis, I analysed the physiology of E.coli aging under carbon starvation, as a model system to empirically validate evolutionary theories of aging. Microfluidic tools are used to isolate large populations of isogenic single E.coli cells, and to achieve homogenous carbon starvation. Despite sharing the same genetical background and environmental conditions, individual cells in the population exhibit significant variations in lifespans and causes of death. Distributions of lifespans exhibit typical features of the aging process, often seen in animal and human demographic studies. The rate of aging can be altered by mutations of the general stress response pathway. Resembling caloric restriction induced longevity, the general stress response pathway extends starvation lifespans of E.coli by attenuating the effect of aging at the expense of immediate needs of the cells. A quantitative model of this physiological trade-off is constructed and correctly predicted experimental observations. As a conclusion, I substantiate the disposable soma theory of aging with the physiological details of E.coli aging in starvation.