Marine mammals consist of cetaceans (mysticetes and odontocetes), sirenians (manatees and dugong), pinnipeds (phocids, otarids, and walrus), marine and sea otters, and the polar bear (Jefferson et al., 1994). These species are not randomly distributed across the world’s oceans, but rather occupy species-specific geographical niches that vary by depth, temperature, and food resources (Jefferson et al., 1994). Many of these species are at risk of extinction (Reynolds et al., 2009), and it has been estimated that nearly three-quarters of all marine mammals experience high levels of human interference, with the most at-risk species inhabiting coastal areas in close proximity to humans (Davidson et al., 2012). Today, a large number of species living in more open waters, which generally have been considered safe from anthropogenic impact, also are being negatively affected by human activities (McCauley et al., 2015). The greatest direct threats to marine mammals are incidental catch (particularly by fisheries), pollution, commercial harvest, hunting, and vessel strikes (Avila et al., 2018). Indirect threats include degradation of habitats and food sources that increase disease susceptibility and reduce reproductive rates, all of which negatively impact population survival (Schipper et al., 2008; Simmonds et al., 2014). Conservation efforts in situ have had mixed results despite their visibility and high value. For example, according to a 2019 report (Valdivia et al., 2019), of 62 marine mammal and sea turtle species, 18 mammal (78%) and six turtle (75%) populations increased after being listing by the United States Endangered Species Act (ESA), while two marine mammal (9%) populations declined and three mammal (13%) and two turtle (25%) populations showed no change after ESA protection. And while some marine mammal species breed well in captivity [e.g., bottlenose dolphin (Tursiops truncatus) (Jaakkola and Willis, 2019), Indo-Pacific bottlenose dolphins (Tursiops aduncus) (Koga et al., 2019)], which can be a hedge against extinction in the wild, most do not. Thus, life history data from captive bred and wild animals are important for risk modeling, although it is limited for most marine mammal species (Davidson et al., 2012; IUCN, 2020).

In general, marine mammals are fairly long-lived (Bourg and Bourg, 2020), but determining age is difficult, particularly for those that live fully underwater. Births and deaths are rarely observed in the wild, so reliable population demographics are lacking for most species. A number of methods for age estimation have been tested in marine mammals, including measures of tissue amino acid racemization (Garde et al., 2007, 2012), fatty acid signatures from blubber, bone mineral density (Bourg and Bourg, 2020), counting dentine growth layer groups (GLGs) in bone or teeth (Read et al., 2018; Bourg and Bourg, 2020), radiographic imaging to determine dentine thickness (Barratclough et al., 2019), measures of telomere length (Olsen et al., 2018; Whittemore et al., 2019) and epigenetic (Polanowski et al., 2014). Estimated lifespans of 211 years in bowhead whale (Balaena mysticetus) determined by aspartic acid racemization in the eye lens (George et al., 1999), 73 years in dugong (Dugong dugon) measured by GLG counts in incisors (Marsh, 1980), and 110 years in the southern hemisphere blue whale (Balaenoptera musculus) and 114 years in the fin whale (Balaenoptera physalus) based on counts of dark and light areas in the core of ear plugs (Ohsumi, 1979) have been reported.

More recently, relative telomere length (rTL) measurements by quantitative real-time polymerase chain reaction (qPCR) have been gaining attention for age estimation (Cawthon, 2002; Hewakapuge et al., 2008; O’Callaghan et al., 2008; Heidinger et al., 2012; Buddhachat et al., 2017; Eastwood et al., 2019; Kaewkool M.K.W. et al., 2020). This method was first used by Cawthon (2002) and based on amplification DNA portion of telomeric repeats TTAGGG that share a conserved sequence of six base 5′-TTAGGG-3′ at the terminal end of vertebrate chromosome. The validity of rTL measures by qPCR has been demonstrated by comparisons with standard methods, such as terminal restriction fragment length (TRF) in humans of known ages (5–94 years), exhibiting a high correlation (R² = 0.68) (O’Callaghan et al., 2008). In zebra fish, rTL decreases with increasing age through telomere attrition (Heidinger et al., 2012). Among bird species of varying longevity, initial rTL was related to lifespan, as well as to lifetime reproductive success (Eastwood et al., 2019). However, Whittemore et al. (2019) has proposed it is not initial telomere length that correlates with longevity across a wide range of mammals, but rather the rate of telomere attrition.

This study investigated relationships and phylogenetic signals between rTL in skin samples of 23 species of marine mammals found along Thailand sea coasts (Andaman Sea and Gulf of Thailand) and measures of body weight and length and degree to which they may be used to estimate maximum life span.

In the present study, our results revealed two significant findings: (i) lifespan and body size appeared to be phylogenetically conserved, especially body length, based on Blomberg’s K and Pagel’s λ = 1; and (ii) lifespan is likely to be connected to body size, both weight and length, but not to median rTL. This information can provide insight into the drivers of conserved traits in the evolutionary process of cetacean species, as well as relationships among life variables that might contribute to longevity.

The phylogeny of 22 cetaceans and traits including lifespan and body sizes (weight and length) gave statistically significant phylogenetic signals using both Blomberg’s K and Pagel’s λ. K and λ-value for body length showed phylogenetic conservatism (K = 1.10 and λ = 1.05) corresponding to evolution processes superior to the BM model (e.g., genetic stabilizing and genetic drift). In addition, LIPA analysis revealed that the genus Balaenoptera showed a considerable hotspot autocorrelation for body length. This was consistent with a study of Slater that explained the independent evolution of mysticete body size might be driven by ocean dynamics like wind-driven upwellings, with high prey density as the primary determinant of efficient foraging of baleen whales (Slater et al., 2017). The ecological prey scape might be an important driver rather than predatory avoidance and niche partitioning (Slater et al., 2017). A phylogenetic signal of λ-value for lifespan and body weight in cetacean species was also significant, but lower than that for body length. This suggested that those two traits might be more rapid diversification than body length causing the relatively variation among the closely related species (Smith et al., 2004; Kamilar and Cooper, 2013). The relative variation in lifespan and body weight were noted in the closely related species of family Delphinidae. However, several studies found similar body size among congeneric species or within the same clade [e.g., primates (Kamilar and Cooper, 2013), mammals (Smith et al., 2004)]. For the study of phylogenetic signals in primate traits, lifespan showed a low phylogenetic signal (Kamilar and Cooper, 2013) in contrast to marine mammal lifespans in the present study, which had a moderate, but significant phylogenetic signal. This might represent a difference in ecological niches between terrestrial and marine mammals affecting the extrinsic mortality risk. A study by Healy et al. (2014) demonstrated that, in addition to the positive correlation between lifespan and body size, there was a striking difference between flying and non-flying vertebrates related to maximum lifespan, as birds with similar body sizes to non-flying vertebrates live longer. Thus, the volant ability and enormous size appear to be important traits for investment in long-term evolution because they reduce the risk of extrinsic mortality (Healy et al., 2014; Slater et al., 2017; Whittemore et al., 2019).

It was not surprising, however, that the rTL in marine mammals was not a good phylogenetic signal due to the presence of high variation of rTL within the same clade, indicating non-directional change or random rTL values through time within mammal marines. For example, the lifespan of K. breviceps (n = 3) was 22 years and had median rTL of 0.23, whereas B. musculus (n = 1) with the longest lifespan (110 years) had a median rTL of 0.03. The rTL variation observed across different species in this study may be due to (i) the limited number of specimens for some species, (ii) the uneven distribution of age classes within species, and/or (iii) inaccurate estimates of true lifespans (Magalhães et al., 2007), (iv) the variation of telomere assays, or (v) the variation of rTL within intra-individual and low repeatability (Olsen et al., 2012). Quality control for rTL estimates using real-time PCR is important in identifying a suitable assay with high consistency, accuracy, and resolution (Olsen et al., 2012). While our findings showed that there were no significant relationships between median rTL and lifespan across marine mammal species, the lifespan of Odontoceti species actually appeared to be inversely related to median rTL. A study of telomeres in 60 different mammals demonstrated that species with long telomeres have shorter lifespans, which is in agreement with marine mammal data (Gomes et al., 2011). It was suggested that median rTL might be not be a major factor in determining lifespan in these groups. Further, a wide variety of bird and mammal species has been studied with no strong correlations found between lifespan and initial telomere length, although there were significant relationships between telomere shortening rate and lifespan of a species (Whittemore et al., 2019). Several other studies also have reported correlations between telomere attrition and a species’ lifespan, including in birds (Haussmann et al., 2003; Boonekamp et al., 2014; Tricola et al., 2018), chimpanzees (Tackney et al., 2014), and cynomolgus monkeys (Lee et al., 2002). However, one study in zebra finches clearly showed that initial telomere length measured at 25 days was reliable for lifespan estimation of individuals (Heidinger et al., 2012).

In the present study, we were not able to measure absolute telomere length, and so had to rely on estimate ages for some cetacean species with limited demographic data (Taylor et al., 2007); i.e., based on five life history parameters obtained from 58 cetacean species. A one-point-in-time measure for rTL was determined from carcasses, contributing to an inability to estimate either initial telomere length or rate of shortening as predictors of longevity.

Relationships between genetically related species were empirically created using a phylogenetic tree annotated with datasets of body index and median rTL. The phylogenetic tree was created based on partial control region sequences of 23 marine mammal species inhabiting Thailand oceans and consisted of a sirenian (D. dugon) and six families of cetaceans, including Balaenopteridae (three species), Ziphiidae (two species), Physeteridae (one species), Kogiidae (two species), Phocoenidae (one species), and Delphinidae (14 species). Thailand accounts for a quarter of cetacean species worldwide (Jefferson et al., 2015), and so represents a highly diverse region for marine mammals. Balaenopteridae members showed a larger body size than the others, but median rTL appeared not to relate. We suggest that each species may undergo specific evolutionary changes with respect to aging in response to their ecological habitat, behavior, and intrinsic factors (heart rates, metabolic rates), which then leads to different median rTL and maximum lifespan relationships.

Measuring telomere shortening in cetaceans is challenging owing to the difficulty of collecting samples from free-ranging animals, and the limited number of species held in captivity. Thus, other traits such as average body weight and body length were used to correlate with maximum lifespan and found that larger species like baleen whales were longer-lived, whereas species of Odontoceti that had a smaller size also had a shorter lifespan. Similarly, P. macrocephalus belonging to Odontoceti has a larger mass and long lifespan (77 years). Moderate correlations (>0.6) between lifespan and log-transformed body weight and/or body length were observed for all species in this study. Our findings further suggest that body size, including body weight and length, is a factor associated with species lifespan in accordance with the results of Magalhães et al. (2007). Ecological factors or the absence of predators likely drive larger animals to have longer lifespans (Kirkwood and Austad, 2000; Magalhães et al., 2007). In addition, higher heart rates are associated with shorter lifespans, in addition to increased risks of cardiovascular mortality through processes involving protein oxidation, free radical production, inflammation, and telomere shortening (Levine, 1997; Zhang and Zhang, 2009). Membrane composition might also be linked to longevity, with longer lifespans being associated with more saturated and monounsaturated fatty acids membrane structures (Hulbert, 2008). More recently, it has been shown that mitochondrial DNA (mtDNA) GC content exhibits significant correlations with maximum lifespan (MLS), and when included as a variable with body mass to predict the MLS is highly predictive (Lehmann et al., 2008, 2013; Toren et al., 2016). MtDNA GC content based on MitoAge database (Toren et al., 2016) might be used as a determinant factor to increase the predictive power to estimate lifespan of marine mammals. Animals such as the bowhead whale (B. mysticetus) and gray whale (Eschrichtius rubustus) are recognized as among the top 1% of longest-lived mammals, possibly due to adaptive mechanisms of cell physiology that support cell survival under extreme environmental conditions, such as DNA maintenance and repair, ubiquitination, apoptosis, immune responses, and insulin signaling (Seim et al., 2014; Keane et al., 2015; Toren et al., 2020). In addition, a study by Yanai et al. (2017) revealed that longevity-associated genes (LAGs), especially Sod2, Sirt1, Mtor, and Rps6 kb1, that have a high number of orthologs are overrepresented across diverse taxa with a high evolutionary distance (yeast, worm, fruit fly, and mouse) and may affect the extension of lifespan.

Finally, longevity may be related to survival tactics and how susceptible species, large and small, are to predation (Ballance, 2018). Dolphins and porpoises with a small body size are likely to be maimed or killed by larger-bodied sharks (Ballance, 2018) and so do not live as long, whereas baleen and larger toothed whales such as P. macrocephalus live longer than smaller dolphins and porpoises because they are less susceptible to predation. Species with a higher likelihood of survival may have evolved to age more slowly (Kirkwood and Austad, 2000).

Our study revealed two significant findings that: (i) lifespan and body length appear to be important traits involved in the evolutionary process of cetacean species, particularly the genus Balaenoptera; and (ii) median rTL of 23 marine mammal species was not correlated with maximum lifespan of individual species, whereas body size (body weight and length) was. This study was limited by a lack of basic information on maximum lifespans of cetacean species retrieved from secondary sources (e.g., text book and online databases), and because they are free-ranging in habitats that are difficult to monitor. However, we attempted to use data from accredited sources as much as possible to compare across species, and provide the first comparative analysis of how life traits may be related to longevity. A one-point-in-time measure of rTL to obtain the median rTL was not a good predictor for longevity, which was not unexpected given the limitations of that measure and findings in other studies. We believe that the enormous size of Balaenoptera species likely reduces the extrinsic mortality risk from environmental pressures (e.g., predators). Finally, our model could be used for lifespan prediction of other marine mammal species based on basic morphometric data.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

Ethical review and approval was not required for the animal study because according to the Animals for Scientific Purposes Act, B.E. 2558 (2015), since a part of this experiment was performed on carcass of stranding marine mammals, no ethical approval was required for this study and confirmed by the Animal Ethics Committee, Faculty of Veterinary Medicine, Chiang Mai University (License number U1006312558).

KB assisted in conducting the experiments, performed the statistical analysis, and data visualization and wrote the manuscript. JB assisted in discussion of the data and edited the manuscript. MK and AP performed the telomere experiments. PK and KK supported the tissue and preliminary data of marine mammals. KN designed and conducted all of the experiments and wrote the manuscript. All authors have read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.