1 | Introduction
Life history is a crucial factor that determines the ecology of animal
species, and each species exhibits a specific life history strategy.
Growth stages can be roughly classified based on life history traits
such as the timing of weaning, maturation, and lifespan. However, the
allocation of resources towards growth, survival, and future
reproduction varies among species (Williams, 1966), and each species has
a different duration for each life stage (e.g. in female Indo-Pacific
bottlenose dolphins (Tursiops aduncus ): calf (pre-weaning):
0–3.5 years, subadult (immature): 3.5–10.3 years, adult (after first
parturition): 10.3–50 years, Kogi et al ., 2004; Kogi, 2013;
Wang, 2018). There have been reports on age-related cessation of
reproductive ability (e.g. post reproductive lifespan in humans
(Homo sapiens ): Levitis & Bingham, 2011; killer whales
(Orcinus orca ): Croft et al ., 2015; Franks et al .,
2016, belugas (Delphinapterus leucas ): Ellis et al .,
2018), and age-related changes in sociality (red deers (Cervus
elaphus ): Albery et al ., 2020). These reports provide evidence
of the changes in resource allocation occurring after maturity,
indicating that information on age is necessary to clarify the life
history of a species, rather than growth stage.
However, longitudinal observation is expensive and time-consuming,
especially for long-lived animals like the Indo-Pacific bottlenose
dolphins which can live about 50 years (Wang, 2018). Therefore, age
estimation methods are essential to efficiently investigate the age
structure of a specific population.
A commonly used method for age estimation in toothed whales
(odontocetes) is counting dental growth layer groups (GLGs) (see Perrin
and Myrick, 1980). This method requires capturing of individuals to
collect dental samples. The invasive nature of measuring dental GLGs,
makes it unsuitable for small or threatened populations. It is also
difficult to estimate age using dental GLGs for populations living in
offshore areas where capture can be difficult and for populations that
are targeted for tourism. Thus, methods of non-invasive age estimation
for toothed whales has developed recently. One is the method using
age-related external appearance changes on the body, including scars and
body cololations (e.g. risso’s dolphins (Grampus griseus ):
Hartman et al., 2015; Indo-Pacific humpback dolphins (Sousa
chinensis ): Guo et al ., 2020). Krzyszczyk & Mann (2012) and
Yagi et al . (2022) described the age-related changes to the
speckle appearance patterns on the Indo-Pacific bottlenose dolphins in
Shark Bay, Australia, and Mikura Island, Japan respectively. Yagiet al. (2023) developed a speckle-based age estimation model that
showed high accuracy (R2 = 0.77, standard deviation
(SD) = 2.58). However, the model is only limited to estimating the ages
between 7.68–21 years due to the spots appearing age and the upper
limit of the age-known individuals.
Aging occurs in many organisms and it leads to various changes at the
tissue and cellular levels (Petralia et al ., 2014). Although
aging is thought to be caused by the combined effects of various factors
(López-Otin et al. , 2013), one of the factors that regulates
aging are epigenetic changes which are dysfunctional systems that
accompany aging at the gene level (Booth & Brunet, 2016). DNA
methylation is an example of an epigenetic changes, in which DNA
methylation rates at CpG sites (cytosine-phosphate-guanine) in specific
gene regions changes with age. Recently, the correlation between DNA
methylation rate and aging has been used to develop an age estimation
method, known as the epigenetic clock. This method was initially
developed for humans (Horvath, 2013) and has since been applied to
several species (e.g. Bechstein’s bats (Myotis bechsteinii ):
Wright et al., 2018; domestic cats (Felis catus ) & snow
leopards (Panthera uncia ): Qi et al ., 2021; brown bears
(Ursus arctos ): Nakamura et al ., 2023). In cetaceans,
Polanowski et al . (2014) was the first to report the application
of DNA extracted from skin samples in humpback whales (Megaptera
novaeangliae ). Following this pioneering study, similar approaches have
been successfully applied to other cetacean species (belugas: Borset al. , 2020; Antarctic minke whales (Balaenoptera
bonaerensis ): Tanabe et al ., 2020; common bottlenose dolphins
(T. truncatus ): Beal et al ., 2019; Indo-Pacific bottlenose
dolphins: Peters et al ., 2022). Capture methods from research
whaling, commercial whaling and hunting are extremely invasive. These
previous studies on cetaceans using epigenetic clock analyses relied on
the samples obtained from invasive methods including commercial whaling
(fin whales (B. physalus ): García‐Vernet et al ., 2021),
whale research program (Antarctic minke whales: Tanabe et al .,
2020), capture and release of wild individuals (common bottlenose
dolphins: Beal et al., 2019), stranded carcasses (belugas: Borset al ., 2021), and the use of rifles and crossbows for biopsy
(humpback whales: Polanowski et al ., 2014; belugas: Bors et
al ., 2021; Indo-Pacific bottlenose dolphins: Peters et al .,
2022). Although biopsy procedures are less invasive compared to capture
methods including whaling and hunting, it remains at a certain level of
invasiveness, particularly for small cetacean species where instances of
mortality have been reported (Weller et al ., 1997; Bearzi, 2000;
Noren & Mocklin, 2012). By using fecal samples, DNA can be collected
non-invasively without the need to touch individuals. However, there is
a limited number of studies based on fecal-sampled epigenetic clocks. To
our knowledge, the only studies to have developed epigenetic clocks
using fecal samples are from Nakano et al . (2019, 2020) which
reported a significant correlation between the methylation rate ofELOVL2 (Elongation of very long chain fatty acids protein 2) and
age in chimpanzees (Pan troglodytes ) and Japanese macaques
(Macaca fuscata ).
To examine the correlation between methylation rate and age, fecal
samples from age-known individuals are required. At our research field,
coastal water off Mikura Island located approximately 200 km south of
Tokyo, Japan, around 160 Indo-Pacific bottlenose dolphins were living
year-round (Kakuda et al ., 2002; Kogi et al ., 2004). Since
1994, longitudinal individual identification surveys using underwater
video data have been conducted around this island (Kogi et al .,
2004). These underwater surveys allow tracking the actual ages of
individuals born after 1994 and collection of fecal samples from
individuals, making the population well-suited for fecal sample-based
age estimation studies.
Here, we investigated the correlation between DNA methylation rate and
age in fecal samples using a low-cost and convenient method called
methylation-sensitive high-resolution melting (MS-HRM) analysis (Wojdacz
& Dobrovic, 2007; Wojdacz et al ., 2008; Tse et al .,
2011). We focused on the genes TET2 (ten eleven translocation 2),GRIA2 (glutamate receptor Ia2/AMPA2), and CDKN2A (cyclin
dependent kinase inhibitor 2A), which reported a correlation between age
and methylation rate in skin samples of a closely related species, the
common bottlenose dolphins (Beal et al., 2019). Furthermore, we
developed an age estimation model using the methylation rates of these
genes. We also assessed the effects of biological factors (sex
differences and female nursing states) on methylation rate because in
humans, various stressors are known to affect the epigenetic clock (e.g.
Lawn et al ., 2018, Marini et al ., 2020) such as increased
frequency of pregnancy causes acceleration in epigenetic age (Ryanet al., 2018). This study aimed to develop a non-invasive age
estimation model using DNA extracted from fecal samples ahead of other
mammals and to contribute to ecological and conservation studies.