Telomere length variation has been implicated in processes of ecological and evolutionary importance in a wide range of organisms. However, while the temporal component of this variation has been the subject of much research, we do not know yet whether a spatial component exists within species in telomere dynamics. Here, we investigated for the first time whether the biogeographic history of populations, a key driver of eco‐evolutionary processes, can influence telomere dynamics. Based on a one‐year longitudinal common‐garden experiment in a tree frog species we found that co‐specific populations can show striking differences in telomere dynamics, not explained by distinct environmental conditions experienced by individuals through time. Indeed, the populations did not differ in telomere length at the beginning of the study, yet they did after one year under standard conditions. We observed stability of telomere length over time within historically stable populations, but remarkable elongation (31.2% on average) within populations arisen during a recent range expansion. Our results suggest the intriguing scenario that non‐equilibrium processes, such as range expansions, might promote plasticity in the molecular machinery regulating telomere dynamics.

Telomeres are nucleoprotein structures located at the terminal ends of chromosomes that play a major role in maintaining genome stability and in avoiding loss of genetic material (O'Sullivan and Karlseder 2010). In recent times, there has been a burgeoning interest in understanding the fitness consequences of variation in telomere length and dynamics (Monaghan et al. 2018). A core idea is that telomere length might integrate the individual history, thus representing a valuable metric of individual phenotypic quality (Angelier et al. 2019 and references therein). For example, the telomere dynamics may be affected by habitat quality, intraspecific competition, immune function, inbreeding, and have been associated with personality traits and the expression of sexual weapons. This emerging integrative view of telomere biology, raises immediate questions about its potential implications in eco‐evolutionary processes that are prototypical of the integrative phenotype, such as dispersal.

Dispersal is a key process in ecology and evolution, which can influence – and be influenced by – the evolution of virtually all the features of an organism, including behaviour, morphology, physiology and genetics (Clobert et al. 2012, Canestrelli et al. 2016). Here, we ask for the first time whether dispersal‐driven processes of historical biogeographic relevance, such as range expansions, can affect telomere dynamics. We hypothesize that, if individual variation in telomere dynamics translates into heritable variation in the capacity to cope with the new demographic and/or environmental conditions encountered during a range expansion, a spatial structure in telomere dynamics might emerge along the range of a recently expanded population. We explore this hypothesis using the Tyrrhenian tree frog Hyla sarda as study species. This is a small, cryptically coloured amphibian endemic to the Tyrrhenian islands, which colonized the Corsica island (i.e. the northern portion of its current range) from the Sardinia island, during a Late Pleistocene range expansion (Bisconti et al. 2011a, b, Fig. 1A2011b). Using a long‐term longitudinal common garden experiment (12 months) during which animals have been housed under identical conditions, we compared telomere length and its change over time between individuals sampled in the source area in Sardinia and individuals sampled in the expansion range in Corsica (Supporting information).

We found significant effects of geographic area (F = 10.25, p = 0.002), sampling time (F = 21.59, p < 0.001) and of their interaction (F = 6.60, p = 0.014). Post‐hoc analyses showed that tree frogs from both Corsica and Sardinia had similar telomere length at the beginning of the experiment (p = 0.549), while tree frogs from Corsica had longer telomeres than those from Sardinia at the end of the experiment (coeff. estimate ± SE: 0.218 ± 0.057, p = 0.002) (Fig. 1). Telomere length did not change significantly over the experiment in Sardinia tree frogs (p = 0.229), while it increased significantly from the beginning to the end of the experiment in Corsica tree frogs (−0.215 ± 0.050, p = 0.0006; 31.2% elongation on average) (Fig. 1B). Results do not change if four females out of 47 frogs are removed from the models (data not shown). Individual telomere length measured from samples taken at the beginning and at the end of the experiment was significantly repeatable (coefficient of 0.42 with an associated variance of 0.03), meaning that 42% of variance in telomere length during an individual's life could be explained by within‐individual consistency. All frogs included in this study survived until the end of the experiment, indicating that there was no selective bias owing to individuals with shorter telomeres having lower chances of survival.

In recent years, telomere dynamics have been studied in a wide range of species (Monaghan et al. 2018). The species, however, might not be the appropriate unit of analysis. Indeed, our study showed, for the first time, that different co‐specific populations can show striking differences in telomere dynamics, and that these differences can hardly be explained by distinct environmental conditions experienced by individuals through time (Supporting information). We also did not observe any differences between populations in telomere length at the beginning of the study; rather, these differences emerged along a longitudinal common‐garden experiment, indicating that point estimates (e.g. as routinely used when comparing sexes or developmental stages) might provide blurred pictures of interindividual variation in telomere dynamics.

Telomeres shorten with age in most organisms studied to date (Tricola et al. 2018), but not in the Tyrrhenian tree frog, as in some other species (Hoelzl et al. 2016, Spurgin et al. 2017). Our longitudinal data covered a non‐negligible portion of the expected lifespan of an individual tree frog (one out of approximately three years), so that telomere attrition was a plausible expectation. Instead, we observed no differences between time points in the Sardinia population, and substantial lengthening of telomeres in the Corsica population. Since Corsica population was founded during a recent range expansion from Sardinia (Bisconti et al. 2011a, b2011b), the evolution of the observed differences in telomere dynamics between the two populations should have occurred during the expansion process or later. Previous data showed that levels of both genetic diversity and bioclimatic suitability did not differ appreciably between the studied populations (Bisconti et al. 2011a, b2011b; see also Supporting information). Accordingly, founder events, genetic drift and environmental adaptation following the expansion, appear unlikely as drivers of the observed divergence in telomere dynamics. Hence, we suggest that adaptive processes occurred during the range expansion event have promoted the evolution of this geographic pattern in telomere dynamics.

In vertebrates, telomere length is maintained and restored by the enzyme telomerase (Monaghan and Haussmann 2006). Telomerase activity appears to be particularly relevant for the regulation of telomere dynamics in ectotherms (Olsson et al. 2018). Telomere elongation in Corsica but not Sardinia under common garden conditions suggests differential expression (i.e. re‐activation or upregulation) of the telomerase in Corsica in response to these novel conditions. This might imply that the range expansion event promoted plasticity in telomere dynamics, adding perspective to the burgeoning focus on the role of plasticity in evolution (Schwander and Leimar 2011). Since telomere attrition may also be linked to behavioural phenotype (Bateson et al. 2015), it might also be that frogs with a specific behavioural type, related to exploration or boldness, are more common in the sink population (i.e. Corsica; Canestrelli et al. 2016). Whether such increased plasticity is an adaptation to the unpredictability of the novel environments, variation in developmental conditions across populations, life‐history/behavioural strategies, or to the non‐equilibrium demographic dynamics encountered during the expansion process, is a further intriguing subject for future research.

端粒长度的变化与多种生物在生态和进化上的重要性有关。然而，尽管这种变化的时间成分已成为许多研究的主题，但我们尚不知道端粒动力学中物种内是否存在空间成分。在这里，我们首次调查了人口的生物地理历史（生态进化过程的主要驱动力）是否会影响端粒动力学。基于对一个树蛙物种进行的为期一年的纵向公共花园实验，我们发现同种种群的端粒动力学表现出显着差异，而个体随时间经历的独特环境条件并不能解释这些差异。实际上，在研究开始时，这些人群的端粒长度没有差异，但他们在标准条件下工作了一年。我们观察到在历史稳定的种群中端粒长度随时间的稳定性，但是在最近的范围扩展过程中，种群内部出现了显着的伸长（平均31.2％）。我们的研究结果提出了一个有趣的场景，即非平衡过程（例如范围扩展）可能会促进调控端粒动力学的分子机械的可塑性。

端粒是位于染色体末端的核蛋白结构，在维持基因组稳定性和避免遗传物质损失方面起主要作用（O'Sullivan和Karlseder 2010）。近年来，人们对理解端粒长度和动力学变化的适应性后果产生了浓厚的兴趣（Monaghan et al.2018）。核心思想是端粒长度可能整合了个体历史，从而代表了个体表型质量的宝贵指标（Angelier等人，2019年）以及其中的参考文献）。例如，端粒动力学可能会受到栖息地质量，种内竞争，免疫功能，近交的影响，并与人格特质和性武器的表达有关。这种端粒生物学的综合观点，提出了其端粒在生态进化过程中的潜在影响的直接问题，生态进化过程是典型的综合表型，例如扩散。

扩散是生态学和进化的关键过程，它可以影响生物体几乎所有特征的演化，并受其影响，包括行为，形态，生理学和遗传学（Clobert等，2012； Canestrelli等，2016）。）。在此，我们首次询问历史生物地理相关性的分散驱动过程（例如范围扩展）是否会影响端粒动力学。我们假设，如果端粒动力学的个体变化转化为适应范围扩展过程中遇到的新的人口和/或环境条件的能力的遗传变异，端粒动力学的空间结构可能会沿着最近扩展的种群范围出现。我们使用第勒尼安树蛙Hyla sarda探索这一假设作为研究物种。这是第勒尼安群岛特有的，隐秘的彩色两栖动物，在晚更新世范围扩展期间，该群岛定居在撒丁岛的科西嘉岛（即其当前范围的北部）（Bisconti等，2011a，b，图）。 1A 2011b）。使用长期纵向公共花园实验（12个月），在相同条件下圈养动物，我们比较了撒丁岛源区域采样的个体和科西嘉岛扩展范围采样的个体之间的端粒长度及其随时间的变化（支持信息）。

我们发现地理区域（F = 10.25，p = 0.002），采样时间（F = 21.59，p <0.001）及其交互作用（F = 6.60，p = 0.014）的显着影响。事后分析显示，在实验开始时，来自科西嘉岛和撒丁岛的树蛙的端粒长度相似（p = 0.549），而来自科西嘉岛的树蛙在实验结束时的端粒长度比撒丁岛的更长（coeff。估计值±SE：0.218±0.057，p = 0.002）（图1）。撒丁岛树蛙的端粒长度在实验中没有显着变化（p = 0.229），而在科西嘉树蛙中，端粒长度从实验开始到结束时显着增加（−0.215±0.050，p = 0.0006； 31.2％的伸长率平均）（图1B）。如果从模型中删除47只青蛙中的四只雌性，结果不会改变（数据未显示）。在实验开始和结束时从样品中测得的单个端粒长度具有可重复性（系数为0.42，相关方差为0.03），这意味着个体生命期间端粒长度的42％的变化可以用以下公式解释：个体一致性。这项研究中包括的所有青蛙都存活到实验结束，这表明由于端粒较短的个体存活机会较低，因此没有选择偏见。

近年来，已经在广泛的物种中研究了端粒动力学（Monaghan et al.2018）。但是，该物种可能不是适当的分析单位。确实，我们的研究首次表明，不同的同种种群可以显示出端粒动力学的显着差异，而且这些差异很难用个体随时间经历的独特环境条件来解释（支持性信息）。在研究开始时，我们也没有观察到端粒长度之间的任何差异。相反，这些差异沿纵向的共同花园实验出现，表明点估计（例如，在比较性别或发育阶段时通常使用的点估计）可能会提供端粒动力学个体间差异的模糊图片。

迄今为止，在大多数研究过的生物中端粒随着年龄的增长而缩短（Tricola等人，2018），但在第勒尼安树蛙中却不像其他一些物种那样（Hoelzl等人，2016 ; Spurgin等人，2017）。我们的纵向数据涵盖了单个树蛙的预期寿命（大约三年中的一个）的不可忽略的部分，因此端粒磨损是一个合理的期望。取而代之的是，我们在撒丁岛人群中的时间点与科西嘉岛人群中端粒的延长之间没有发现差异。由于科西嘉岛的种群是在最近的撒丁岛范围扩展中建立的（Bisconti等人2011a，b 2011b），这两个种群之间观察到的端粒动力学差异的演化应该在扩展过程中或之后发生。先前的数据表明，研究人群之间遗传多样性和生物气候适应性水平均无明显差异（Bisconti等，2011a，b 2011b；另见支持性信息）。因此，创始人事件，遗传漂移和扩展后的环境适应似乎不太可能成为观察到的端粒动力学差异的驱动因素。因此，我们建议在范围扩展事件期间发生的自适应过程促进了端粒动力学中这种地理模式的演变。

在脊椎动物中，端粒长度由端粒酶维持和恢复（Monaghan and Haussmann 2006）。端粒酶活性似乎是端粒动力学的外温动物的调节（Olsson等。特别相关2018）。在常见的花园条件下，可西嘉岛的端粒延长而不是撒丁岛的端粒伸长表明，应对这些新情况，可西嘉岛的端粒酶差异表达（即重新激活或上调）。这可能意味着范围扩展事件促进了端粒动力学的可塑性，增加了人们对可塑性在进化中的作用的迅速关注的视角（Schwander和Leimar 2011）。由于端粒磨损也可能与行为表型有关（Bateson等人，2015年），也有可能是特定行为类型的青蛙（与探索或大胆相关）在汇人群中更常见（即科西嘉岛; Canestrelli等人2016）。这种增加的可塑性是否适应新环境的不可预测性，不同人群的发展条件变化，生活史/行为策略，或在扩张过程中遇到的非平衡人口动态，是未来另一个令人感兴趣的话题。研究。