Telomere length and dynamics in Astyanax mexicanus cave and surface morphs

Introduction

Cavefishes represent one of the most successful groups of vertebrates that colonized the subterranean world, with more than 350 species from every continent (Niemiller et al., 2019). Subterranean habitats are semi-enclosed ecosystems characterized by specific environmental features, such as the absence of light, stable microclimate, and scarcity of food resources (Culver & Pipan, 2019). These features triggered the evolution of specific traits in species that settled there (Balart-García et al., 2023). Characteristic traits of subterranean species may be morphological (e.g., loss of eyes and/or pigmentation, elongation of appendages), physiological (e.g., lower metabolism, loss of circadian rhythm), or behavioral (e.g., increase in exploratory behavior, lower intraspecific agonistic behavior) (Christiansen, 2012; Lunghi et al., 2023; Mösslacher & Creuzé Des Châtelliers, 1996; Poulson, 1963). A recent study that compiled evidence from the available literature highlighted the higher longevity of subterranean species compared to their closely related aboveground species and hypothesized that this may be considered an additional adaptive trait for subterranean life (Lunghi & Bilandžija, 2022).

A widely used method to predict the lifespan of individuals is to measure telomere length (Bize et al., 2009). Telomeres are repeated sequences of non-coding DNA located at chromosome ends in all vertebrates and most metazoans (Gomes, Shay & Wright, 2010; Monaghan, 2010); their function is to protect chromosomes from degradation and fusion (Armanios & Blackburn, 2012). The shortening of telomeres is one of the main causes of cell senescence (Campisi & D’Addadi Fagagna, 2007; Haussmann, Winkler & Vleck, 2005; Takubo et al., 2010). Each time a cell divides, telomere replication is incomplete, resulting in the loss of part of the telomere sequence and the generation of two daughter cells with shorter telomeres (Campisi & D’Addadi Fagagna, 2007). When telomere length is reduced to a critical length, the cell ceases replication and enters senescence; this mechanism is the basis of aging in many organisms (Blasco, 2007; Hemann et al., 2001). However, having longer telomeres (or higher telomerase activity) does not necessarily facilitate longer lifespan (Gomes et al., 2011; Gorbunova et al., 2014). Environmental exposures and different stressors (e.g., oxidative stress) can increase telomere shortening and drive cells into senescence earlier than is expected for their biological age (Boonekamp et al., 2013; Chatelain, Drobniak & Szulkin, 2020; Epel et al., 2004). Telomere shortening is sometimes counteracted by telomerase activity, which can maintain and extend telomere length, helping to slow overall organismal aging (Anchelin et al., 2011; Blasco, 2007). Furthermore, limited telomere length is necessary to prevent development of cancer (Gomes et al., 2011) as programmed cell death is critical for removal of damaged and potentially tumorigenic cells (Elmore, 2007; Strasser & Vaux, 2020). Therefore, telomere length and organismal longevity is not in a perfect correlation, and the balance between longevity and cancer dictates an optimal interval for each specific animal group.

In this study, we measured average telomere length (hereafter TL) in Mexican tetra, Astyanax mexicanus, a model species in evolutionary biology (Jeffery, 2020). This species has two ecomorphs, one fully adapted to living in caves—cavefish (hereafter CF) and the other living in surface waters—surface fish (hereafter SF). The advantage of using A. mexicanus in evolutionary studies is that comparisons between conspecifics allow us to avoid potential biases due to the intrinsic life history or evolutionary history differences that characterize different species, regardless of how closely related they are (Ficetola et al., 2018; Michaux, Libois & Filippucci, 2005; Verberk, Siepel & Esselink, 2008). The two Astyanax ecomorphs are adapted to living in environments characterized by different typologies and magnitude of stressors; for example, surface fish are subjected to high predator pressure and exposed to UV-radiation, whereas cavefish live in nutrient-limited and possibly hypoxic environment (Culver & Pipan, 2019). Our hypothesis was that the two ecomorphs initially shared similar TL and that this particular trait underwent selective pressure during the process of specialization to different habitats (Culver & Pipan, 2019). For example, UV radiation and high predator pressure are expected to shorten telomeres whereas calorie restriction and hypoxia would have the opposite effect. Therefore, we expect to find divergence in TL between Astyanax CF and SF. If so, it can be an effect of difference in longevity between the two morphs, consistent with the observed higher longevity of cave animals compared to their surface relatives (Lunghi & Bilandžija, 2022). Alternatively, considering that TL shortening is significantly amplified by various stressors individuals are exposed to (Von Zglinicki, 2002), and that aboveground and belowground environments are characterized by different stressors, a difference in telomere length between SF and CF could be the effect of different selective pressure promoted by the two environments.

Materials & Methods

Results

Experiment with caudal fins

The average telomere length (TL) was significantly correlated with the fish ecomorph (df = 1, F = 10.479, P = 0.002), while no significant correlation with fish age (df = 1, F = 0.062, P = 0.774) was observed; CF had shorter TL compared to SF (Fig. 1A).

Experiment with gametes

TL significantly correlated with ecomorph (F_1,38 = 6.99, P = 0.012) and with sex (F_1,38 = 6.18, P = 0.017), while no significant effect was observed for fish age (F_1,38 = 0.1, P = 0.751) and tissue type (F_1,41 = 3.84, P = 0.057), although significance is slightly above the limit set at 0.05 for the latter. The TL was longer in SF compared to CF (Fig. 2A), and in males compared to females (Fig. 2B). Caudal fin TL was significantly correlated with TL in gametes in SF (R = 0.6, P = 0.013), whereas no significant correlation was found between TL of these two tissues in CF (R = 0.36, P = 0.068).

Discussion

In this study, we observed a divergence in TL between the Pachón CF and Rio Choy SF (Figs. 1–2), where CF had shorter TL. This contradicts our prediction that CF could have longer TL than SF based on the expected higher longevity of cave animals (Lunghi & Bilandžija, 2022). However, we also observed no effect of age on telomere length in either of the two morphs. Considering this result and the lack of robust evidence for a divergence in lifespan between Astyanax SF and CF (Riddle et al., 2018; Simon et al., 2017), TL in A. mexicanus could be considered a proxy for somatic redundancy (i.e., the ability of an organisms to buffer exogenous stress) rather than a biomarker for aging (Boonekamp et al., 2013; Sauer et al., 2021).

CF having shorter TL than SF can be attributed to two possible explanations. Compared to subterranean animals, individuals from surface rivers are exposed to a wider variety of biotic and abiotic stressors, including oxidative stress due to faster metabolism, UV radiation, greater environmental fluctuations (e.g., microclimate), and predation risk (Culver & Pipan, 2019). All of these stressors can contribute to the acceleration of telomere shortening (Von Zglinicki, 2002). Therefore, longer telomeres in SF may be an efficient mechanism to counteract these stressors. On the other hand, subterranean organisms exhibit slower growth, reduced metabolism and limited reproduction rates (Bulog et al., 2000; Howarth & Moldovan, 2018; Poulson, 1963). These physiological adjustments correlate with slower telomere attrition. Furthermore, cave animals live in environments where they must cope with prolonged periods of starvation and hypoxia (Bizjak Mali, Sepčić & Bulog, 2013; Lipovšek et al., 2019; Van der Weele & Jeffery, 2022), conditions that may cause stress, but were shown to have a positive effect on telomere length (Iglesias et al., 2019; Wang et al., 2018). Therefore, the relief from some sources of stress present on the surface and the reduction of potential body damage in cavefish, may have contributed to relaxed selection and allowed CF to reduce TL. Alternatively, a direct selection for shorter TL may have occurred when surface ancestors colonized subterranean environments (Boonekamp et al., 2021; Gross, 2012). According to the “thrifty telomere” theory (Eisenberg, 2011), having longer telomeres is costly, and in nutrient-poor environments such as caves, where organisms have fewer resources to devote to their biological activities (Culver & Pipan, 2019) less costly and shorter telomeres could be beneficial and selected for. However, another possibility is that telomere length is controlled by pleiotropic effects of genes selected for a different phenotypic trait (Pathak et al., 2021) as is the case in the evolution of eye degeneration and albinism in Astyanax CF (Bilandžija et al., 2018; Bilandžija et al., 2013; Krishnan et al., 2022; O’Gorman et al., 2021; Yamamoto et al., 2009). Future research will enable assessing which of the evolutionary forces may be responsible for the divergence in TL between Astyanax morphs.

No significant effects of aging on telomere shortening in adult A. mexicanus suggest that telomere maintenance mechanisms may be very efficient in this species. Lost telomere sequences are mainly replaced by telomerase activity (Campisi & D’Addadi Fagagna, 2007; Gomes, Shay & Wright, 2010; Xie et al., 2008). Studies conducted on different fish species, reported that despite the high telomerase activity in different organs, there was a significant decrease in TL with the age, which is a clear indication of the senescence of the organism (Hartmann et al., 2009; Hatakeyama et al., 2008). In short-lived zebrafish, on the other hand, TL does not shorten until 24 months of age, implying that individuals avoid senescence for approximately four-fifths of their lives due to telomerase activity (Anchelin et al., 2011; Lau et al., 2008; Lund et al., 2009). In our model organism, A. mexicanus, we found no evidence of telomere attrition in adult individuals up to 9 years old, which is consistent with the estimated age of the oldest wild-caught A. mexicanus (Simon et al., 2017). Therefore, senescence in this species might not be correlated with TL shortening, but is regulated by other mechanisms (Gruber et al., 2014). However, we cannot rule out the possibility that A. mexicanus does not age significantly within the age range examined in our study (1.5–9 years) and that telomere attrition may occur in older fish like some other signs of senescence (Anchelin et al., 2011; Riddle et al., 2018). Also, our results were obtained from lab-reared fish and may require a careful interpretation (i.e., conditions in laboratory environments differ from those occurring at natural sites). However, this is the easiest and most precise way to obtain exact information on fish age. Furthermore, rearing in the laboratory conditions controls for the effects of differing ecology present in the two environments in nature and is ideal to uncover genetic differences between the two morphs.

The third experiment did confirm once again that SF have longer TL compared to CF (Fig. 2A), but also highlighted a significant difference between sexes (Fig. 2B), where males have longer TL compared to females. This is not surprising, as females usually invest more in the reproduction, which strongly impact the TL (Kokko & Jennions, 2008; Sudyka, 2019). Although the difference in TL between gametes and fin was slightly above the significance threshold (P = 0.05), a trend showing larger TL in the former can be observed (Fig. 2B). This suggests that telomere dynamics not only differs between the two morphs, but also between different ontogenetic stages.

We observed a clear divergence of TL between the fish organs investigated; in both ecomorphs, the shortest TL was observed in the bladder and skin. In bony fish, the bladder regulates buoyancy. It is composed mainly of collagen fibers and generally lacks vessels (Pough, Janis & Heiser, 2013). Therefore, oxidative stress and cell renewal in the bladder are likely to be very limited (Mizushima & Komatsu, 2011; Von Zglinicki, 2002), so reduction of mechanisms that counteract telomere attrition in this organ could reduce maintenance costs (Eisenberg, 2011). At the same time, the skin represents the protective layer that constantly defends organism against external stressors and therefore undergoes high cellular division, which accelerates the shortening of telomeres (Buckingham & Klingelhutz, 2011). On the other hand, other fish organs are involved in various biological processes such as metabolism, locomotion or information processing, and constant cell turnover and production of energy are essential for maintaining the vitality of these organs (Mizushima & Komatsu, 2011; Pough, Janis & Heiser, 2013), which means that the cells composing their tissues are constantly under oxidative stress, one of the major exogenous causes of telomere shortening (Eisenberg, 2011; Von Zglinicki, 2002). Therefore, prolonged TL in these organs may not only help maintain their vitality, but also increase overall survival and longevity of the fish (Boonekamp et al., 2013; Haussmann et al., 2002; Haussmann, Winkler & Vleck, 2005).

The TL of the different organs correlated poorly, showing only six significant correlations for SF and one for CF (Table 1). Of the total significant correlations, only two in SF involved the caudal fin. For these relatively small fish, fin clipping is a more practical and sustainable method of DNA sampling because it does not require sacrificing the fish and can be used in longitudinal studies due to the ability of Astyanax to regenerate fins (Stockdale et al., 2018). Our goal was to establish a non-lethal protocol for examining TL in Astyanax to identify potential variation in telomere attrition over the course of the lifespan of the fish. Unfortunately, TL of the caudal fin did not correlate significantly with other organs (especially in CF) in the populations we studied. However, repeating this experiment in other populations, especially sampled from natural habitats, may increase the possibility of developing a sustainable method for studying TL in Astyanax fish.

Conclusions

Our study provides the first information on the average telomere length in Astyanax mexicanus revealing that Pachón cavefish exhibit shorter telomeres compared to Rio Choy surface fish. Notably, we observed no evidence of telomere attrition as an indicator of senescence in adult fish used in our experiments, ranging in age from 1.5 to 9 years. While the laboratory conditions, marked by a stable environment, consistent food supply, and absence of predation risk (Chin et al., 2018; Chin et al., 2020), may have influenced these results, they also underscore the genetic basis of the observed telomere length differences between CF and SF. Under these favorable conditions, potential telomerase activity, known for mitigating telomere attrition due to environmental stress in the wild (Boonekamp et al., 2013) may have contributed to masking telomere attrition due to aging. Our analyses support the hypothesis that, at least in the laboratory stocks of A. mexicanus used in this study, average telomere length is more indicative of somatic redundancy than of individual lifespan. The less stressful (fewer predators, reduced environmental fluctuations) and longevity-promoting conditions (lack of UV irradiation, caloric restriction, hypoxia) in subterranean habitats may have contributed to cavefish exhibiting shorter telomeres that require less energy for maintenance. Additional analyses involving wild fish and other Astyanax populations will be important to validate and interpret these findings. Our study positions Astyanax mexicanus as a valuable model species for deepening our understanding of telomere dynamics and evolution.

Supplemental Information