Linking growing up and getting old

Long-term effects of developmental diet on adult life history traits and health have been identified across a range of organisms, including humans. However, there has been considerable debate about the adaptive significance of such effects and the mechanisms responsible. In particular, we still lack insight into whether and how much such effects depend on the adult environment (as has been hypothesised to be the case in humans), and if and how such effects manifest as long term effects on gene expression, a proposed mechanism for the long-term effects. The aim of this thesis was to use the fruit fly, Drosophila melanogaster, as a model to gain insight into the effects of developmental diet on adult phenotypes and gene expression, and if and how this effect is also contingent on the actual adult environment experienced. I addressed this question from two angles by assessing the effects of developmental diet both within a single generation (plasticity) and when experienced across many generations (adaptation). <strong>Chapters 2, 3 and 4</strong> address the plastic effects of developmental diet on phenotypes and gene expression across adult diets and across the lifespan, while <strong>Chapter 5</strong> uses experimental evolution to address how flies adapt to differing larval diets over many generations, and how this interacts with classical selection for increased age-at-reproduction. Throughout the thesis I used the same three diets (poor, control, and rich) which differ 10 fold in the amount of sugar and yeast they contain and thus represent drastically different nutritional environments. <strong>Chapter 2</strong> begins by determining the general effects of the three diets used throughout this thesis on both developmental and adult traits. It addresses the fundamental question of whether variation in developmental diet affects developmental and adult life history traits, with a particular emphasis on whether the effect of developmental diet depends on the reproductive potential of the adult environment. I found that both the rich and especially the poor developmental diet lead to what would classically be considered negative effects on developmental traits relative to the control - both slowing development and decreasing adult size in&nbsp; young adulthood, however, their effects on adult traits were distinct: the poor larval diet lead to increased virgin lifespan and increased female fecundity at certain ages and in certain adult reproductive environments relative to the control, while the rich larval diet had the opposite effect. This suggests that while over- and under-feeding share certain similarities with respect to their effects on traits in early life, their long-term effects differ. It also indicates differing nutritional optima between the developmental and adult stage in flies, as the poor diet is known to drastically decrease both lifespan and fecundity when experienced during adulthood, but is shown here to be largely beneficial for these traits when experienced in development. In addition, we found that the adult reproductive environment was considerably more important for determining traits than the developmental environment, thus, while long-term effects of developmental diet do exist, they are marginal when related to the plastic response effected in adulthood - a recurring theme in this thesis. In<strong> Chapter 3</strong> the focus shifts from the effect of developmental diet on life history traits to its effect on the adult transcriptome. It has frequently been proposed that long-term effects of developmental diet on adult phenotypes are mediated by changes in gene expression. Thus, in this chapter I addressed the scope for such effects in fruit flies, by determining the relative effect of developmental versus adult diet on gene expression in very young adult flies (one-day old) in both sexes. I used a full factorial design combining three larval and three adult diets (9 treatments total). I found that the largest contributor to transcriptional variation in one day old flies is larval, not adult diet, especially in females. Furthermore, the global effect of increasing caloric content of the larval diet on gene expression was not linear, but rather followed the same pattern as that observed for developmental phenotypes in <strong>Chapter 2 </strong>(i.e. rich-raised flies were intermediate between poor and control) suggesting that calories per se do not drive global patterns of gene expression variation. Next, using Weighted Gene Correlation Networks Analysis (WGCNA) I identified modules of co-expressed genes whose expression was affected by larval or adult dietary conditions. In females, larval diet modulated the relative expression levels of reproduction versus non-reproduction related genes, while in males a large portion of the transcriptome was unaffected by dietary conditions, suggesting a lesser role for both larval and adult diet in affecting gene expression. The modules affected by diet in males related primarily to nutrient sensing and metabolism and showed no evidence of the reproduction and cell-cycle related processes identified in females, however, their expression in external tissue specific data sets suggested a role for the gut and fat body in mediating the effects of diet in males, potentially through the carry over of the larval versions of these tissues into adulthood. Overall, these results suggested that there is scope for long-term effects of developmental diet on gene expression, which is necessary for all hypothesised mechanisms that link developmental conditions to late-life health and disease. <strong>Chapter 4</strong> combines phenotypic and transcriptomic approaches to look at the longer term effects of developmental diet in adulthood. Using the same full factorial approach applied in <strong>Chapter 3</strong>, I assessed virgin and mated lifespan and fecundity as well as gene expression at middle and old age. I found that for the most part larval and adult diet exerted independent effects on the phenotype and on gene expression, and thus there was no evidence for Predictive Adaptive Responses (described in <strong>Chapter 1</strong>) operating in Drosophila melanogaster. Rather, the responses followed the silver spoon hypothesis which predicts that the effect of developmental conditions will be similar across adult conditions. Furthermore, in contrast to the beginning of life (<strong>Chapter 3</strong>), adult diet explained considerably more variation in gene expression and phenotypes than larval diet, showing that flies retain extensive plasticity into adulthood, and suggesting that the long-term effects of developmental diet likely reflect the inability or lack of incentive to erase such effects, rather than an adaptive response. I did identify some genes that retain a legacy of developmental diet in their expression into middle and old-age. Many of these genes show no linear correlation with the observed phenotypic responses, however, in both sexes, I identified a cluster of genes whose expression was negatively correlated with the observed lifespan differences and that were enriched with terms related to transcription and translation, particularly with respect to ribosomes. Given several recent studies which show that the down-regulation of ribosomes and other aspects of transcriptional and translational machinery increases lifespan these genes provide promising candidates for mediating the long-term effects of larval diet on lifespan. As these processes are highly conserved across the tree of life our results may be relevant for other species as well, including for humans. In <strong>Chapter 5</strong>, I address the evolutionary, rather than the plastic response to developmental diet by evolving flies on the three different larval diets in combination with selection for early or late age-at-reproduction in a full factorial design. This approach addresses how life histories evolve under different levels of larval acquisition, as well as how individuals cope with potentially competing selection pressures experienced at two different life stages. I found that the two life stages do not act independently but rather interact to determine both developmental time and lifespan. Across all evolutionary developmental diets, selection for later age-at-reproduction increased lifespan, however, the magnitude of the response was dependent on the sex, the evolutionary diet, and the experimental assay conditions. Developmental time from egg-to-adult also showed a similar dependency on both evolutionary diet levels and selection on age-at-reproduction. Given that multiple selection pressures are likely the norm rather than the exception in nature, this finding argues that trade-offs should be considered not only between traits within an organism, but also between adaptive responses to differing selection pressures. Furthermore, because variation in available nutrition is so frequent in natural environments, we argue that this can play a large role in shaping the evolution and diversity of life histories in nature. In the discussion in<strong> Chapter 6</strong> I synthesise the findings across the experimental chapters, and address how they relate to each other. I also discuss some more general points that emerged, such as the potential importance of allometry or tissue specific effects in mediating effects of larval diet, and the difficulty of linking life history phenotypes to variation in gene expression. Finally, I point out future directions suggested by this thesis, as well as the potential insights that can be gained from this work in the context of theories linking developmental conditions to late life health in humans