Towards testable quantitative models for the evolution of caste specific ageing part II: integrating theory and experiment

进化论面临的一个主要挑战是，如何解释群居昆虫的生殖阶层如何进化得比工蚁阶层和类似大小的非群居昆虫的寿命长得多。尽管经典进化衰老理论——突变积累（MA）、拮抗多效性（AP）、一次性体细胞（DS）——的口头论据已被用于定性解释种姓特异性衰老，但经典模型并不适用，因为它们的假设与昆虫社会生物学相冲突。该项目的目的是：(1)为种姓特异性衰老的进化建立一般和特定物种的定量模型，(2)将模型与研究单位（RU）的实验数据相结合，以便严格测试和完善模型，并帮助指导进一步的实验研究。在第一个资助阶段，我们在特定种姓老龄化方面取得了若干概念突破。首先，我们用数学方法证明了经典衰老理论的一个基本原理——自然选择的力量随着年龄的增长而减弱——并不适用于群居昆虫的生殖等级。相反，蚁后生存的选择可以在第一次繁殖的年龄后很长一段时间内继续发挥最大的作用。这可以从“超有机体”的观点来理解，蚁后实际上是“种系”，工蜂的生产相当于“体细胞生长”，而不是繁殖。其次，我们已经证明，种姓之间的选择强度差异确实可以引发蜂后/工蜂寿命比大幅增加的进化，但问题仍然是需要何种机制（MA和/或SA和/或DS和/或其他）的组合来解释在自然界中观察到的定量极端差异。在第一阶段的剩余时间里，我们打算进一步分析这个问题，并开始将这些模型与来自其他RU项目的转录组数据整合起来，通过将表达水平中按种姓的年龄相关性映射到模型中的权衡，这将产生针对种姓特定的衰老谱的第一个定量预测。对于第二阶段，我们有两个主要目标。第一个目标是通过(a)纳入明确可进化的遗传调控网络，从而对我们当前的模型进行更机械的扩展，这些网络是导致种姓内部和种姓之间健康成分之间权衡的资源分配过程的基础；(b)允许种姓专门化程度和种姓数量与种姓特有的衰老表型共同进化。第二个目标是进一步将来自其他RU项目的生命历史、社会和生物信息学数据与模型集成。这将允许对进化衰老模型进行前所未有的定量测试，我们希望这将大大推动进化衰老领域的发展，而不仅仅是我们对社会性昆虫的应用。

A major challenge for evolutionary theory is to explain how in social insects the reproductive caste has evolved a lifespan much greater than that of the worker caste and that of similar-sized non-social insects. Although verbal arguments from classical evolutionary ageing theories – mutation accumulation (MA), antagonistic pleiotropy (AP), disposable soma (DS) – have been used to qualitatively explain caste-specific ageing, the classical models cannot apply because their assumptions clash with social insect biology. The aims of this project are to (1) develop general and species-specific quantitative models for the evolution of caste-specific ageing, and (2) integrate the models with experimental data from the Research Unit (RU), in order to rigorously test and refine the models and to help guide further experimental research. During the 1st funding phase we achieved several conceptual breakthroughs regarding caste-specific ageing. First, we have shown mathematically that one of the bedrock principles of classical ageing theory – the declining force of natural selection with age – does not apply to the social insect reproductive caste. Instead, selection on queen survival can continue to act at maximal strength long after the age of first reproduction. This can be understood from a “superorganism” viewpoint, with queens acting as de facto “germline”, and the production of workers amounting to “somatic growth” rather than reproduction. Secondly, we have shown that the between-caste divergence in the strength of selection can indeed trigger the evolution of strongly increased queen/worker lifespan ratios, but the question remains what combination of mechanisms (MA and/or SA and/or DS and/or others) is required to explain the quantitatively extreme divergence observed in nature.For the remainder of the 1st phase we intend to further analyse this question, and to start integrating the models with transcriptome data from the other RU projects by mapping age-by-caste correlations in expression levels to trade-offs in the models, which will yield the first quantitative predictions for caste-specific ageing profiles. For the 2nd phase we have two main goals. The first goal is to develop more mechanistic extensions of our current models by (a) incorporating explicitly evolvable genetic regulatory networks underlying the resource allocation processes that cause trade-offs between fitness components, both at the within-caste and between-caste levels; and (b) by allowing the degree of caste specialization and the number of castes to co-evolve with caste-specific ageing phenotypes. The second goal is to further integrate life historical, social and bioinformatics data from other RU projects with the models. This should allow an unprecedented degree of quantitative testing of evolutionary ageing models, which we hope will considerably advance the evolutionary ageing field in general, beyond our applications to social insects.