Discussion
Parental age often has long-term consequences for offspring phenotype
and fitness, but the mechanisms that mediate these effects remain
insufficiently studied. Previous research suggests that telomeres may be
an important mechanism mediating these effects, but the routes by which
these effects occur are not well understood. The cross-fostering
experiment used here allowed us to distinguish whether any effects of
parental age on offspring telomeres occur as a result of epigenetic like
and/or pre-natal effects vs. post-natal effects. Previous research in
this colony of common gulls has demonstrated that offspring fitness
decreases with increasing parental age: recruitment rate decreases in
common gulls after the 10th breeding year (12-13 years of age)
(Rattiste, 2004), and older gulls allocate less nutrients to their eggs
(Urvik et al., 2018). However, despite these demonstrated effects of
parental age on offspring fitness in this species, we did not find any
evidence of the effect of parental age on offspring telomeres at
hatching or the change in telomere length during post-natal development.
Although parental age has been shown to have both negative and positive
effects on offspring telomere length (Table 4), this has not been found
in all studies. For example, no effect of parental age on offspring
telomere length was found in a study of Soay sheep (Ovis aries )
despite a large sample size (Froy et al., 2017). Although bird studies
have generally found a negative effect of parental age on offspring
telomeres (see for example studies in alpine swifts, (Criscuolo et al.,
2017), and European shags Phalacrocorax aristotelis , (Heidinger
et al., 2016), this is not a universal pattern (for example, older great
reed warbles Acrocephalus arundinaceus produced offspring with
longer telomeres, (Asghar et al., 2015b). Also in humans the effect of
paternal age is positive rather than negative, as older fathers sire
offspring with longer telomeres (Broer et al., 2013). A review article
summarizing these results concluded that while there is no clear pattern
across species, this is unlikely to be explained by statistical noise or
publication bias (Eisenberg, 2014), but rather linked to the specific
biology of each species. There are also several additional factors that
could contribute to these mixed results: whether the study is
experimental or correlative, at what age the offspring are sampled, and
whether it is a longitudinal versus cross-sectional study of parental
age. Accordingly, we cannot rule out that a longitudinal set-up would
have revealed an effect of parental age on offspring telomere results
also in our study system, as the cross-sectional design could have
obscures the within individual patterns. Another possibility is that the
link between parental age and offspirng telomere length in this species
was too weak to be revealed from our sample size, despite relatively
good test power (Table S3). In this case, and also in the case of a
missing link, it is possible that telomere length in common gulls is
indeed relatively buffered from environmental influence.
Telomere length was repeatable across the chick rearing period in our
study, indicating a possible increase during post-natal development, and
not being related to chick growth rate. Telomeres were longer at the
second sampling, when chicks were approximately 11 days old. This period
covered the fastest growth period of gull chicks. While telomeres
shorten during growth in most studied adult vertebrates (which, as of
now, mainly include mammals and some bird species), some vertebrates
show increased telomerase activity during development (reviewed by
Larsson, Rattiste, & Lilleleht, 1997, Monaghan & Haussmann, 2006). It
has even been suggested that in free-living long-lived organisms,
evolution should favour mechanisms that maintain the longest possible
telomeres at the end of the most active growth period (Chan, &
Blackburn, 2004). Studies in the wild have shown that telomeres elongate
in at least some life stages in Seychelles warblers (Acrocephalus
sechellensis (Spurgin et al., 2018)), edible dormouse (Glis glis(Hoelzl et al., 2016)), Soay sheep (Fairlie et al., 2016), Magellanic
penguins (Spheniscus magellanicus (Cerchiara et al., 2017)), and
Atlantic salmon (Salmo salar (McLennan et al., 2018)). However,
when looking at the few studies on species more closely related to
common gulls, telomere shortening during the early growth period has
been reported. For example, in lesser blackābacked gull (Larus
fuscus ), telomeres shortened from hatching to ten-days old (Foote et
al., 2011), and in black-tailed gull chicks (Larus
crassirostris ), telomere attrition was shown for chicks growing with
siblings in the nest (but not singleton chicks (Mizutani, Niizuma, &
Yoda, 2016)). The latter study suggests that, in favourable growing
conditions, telomere attrition might be prevented. However, more results
on closely related species are needed to confirm the possibility of
telomere lengthening during early growth in gulls. Growth rate was
independent of telomere length or telomere dynamics (Figure 3). In
general, a trade-off between rapid growth and telomere maintenance is
expected, due to increased number of cell divisions required to attain
larger size, and/or increased loss of telomere length during each cell
division as a consequence of the conditions required for fast growth
(e.g. higher metabolic rate and ROS production reviewed by Monaghan &
Ozanne (2018)). The current study adds to the increasing number of
studies suggesting a different pattern in long-lived seabirds (Mizutani
et al., 2016; Young et al., 2017). Seabirds are distinguished from most
other species by a long time period between the end of relatively fast
somatic growth and the beginning of reproduction (stretching several
years). More studies applying comparable methodological approaches are
needed for a comparative study, including patterns of growth and
life-history strategies of different species to determine if this
phenomenon of delayed reproduction is causally linked with the lack of
an association between fast growth and telomere shortening in seabirds.
In conclusion, our results suggest that the age of the parents at the
time of offspring conception does not influence offspring telomere
length or the change in telomere length in common gulls. An important
area of future research is to identify other mechanisms that mediate the
long-term effects of parental age on offspring and to better understand
the factors that contribute to the variation in the influence of
parental age on offspring telomeres across species.