Content uploaded by Christopher W Kuzawa
Author content
All content in this area was uploaded by Christopher W Kuzawa on Apr 07, 2018
Content may be subject to copyright.
rstb.royalsocietypublishing.org
Review
Cite this article: Eisenberg DTA, Kuzawa CW.
2018 The paternal age at conception effect on
offspring telomere length: mechanistic,
comparative and adaptive perspectives.
Phil. Trans. R. Soc. B 373: 20160442.
http://dx.doi.org/10.1098/rstb.2016.0442
Accepted: 6 October 2017
One contribution of 19 to a theme issue
‘Understanding diversity in telomere
dynamics’.
Subject Areas:
evolution, genetics, health and disease and
epidemiology
Keywords:
intergenerational inertia, predictive adaptive
response, senescence, disposable soma,
evolutionary biology, plasticity
Author for correspondence:
Dan T. A. Eisenberg
e-mail: dtae@dtae.net
Electronic supplementary material is available
online at https://dx.doi.org/10.6084/m9.
figshare.c.3945181.
The paternal age at conception effect on
offspring telomere length: mechanistic,
comparative and adaptive perspectives
Dan T. A. Eisenberg1and Christopher W. Kuzawa2
1
Department of Anthropology, Center for Studies in Demography and Ecology, University of Washington,
Seattle, WA, USA
2
Department of Anthropology, Institute for Policy Research, Northwestern University, Evanston, IL, USA
DTAE, 0000-0003-0812-1862
Telomeres are repeating DNA found at the ends of chromosomes that, in the
absence of restorative processes, shorten with cell replications and are impli-
cated as a cause of senescence. It appears that sperm telomere length (TL)
increases with age in humans, and as a result offspring of older fathers inherit
longer telomeres. We review possible mechanisms underlying this paternal age
at conception (PAC) effect on TL, including sperm telomere extension due to
telomerase activity, age-dependent changes in the spermatogonial stem cell
population (possibly driven by ‘selfish’ spermatogonia) and non-causal con-
founding. In contrast to the lengthening of TL with PAC, higher maternal
age at conception appears to predict shorter offspring TL in humans.
We review evidence for heterogeneity across species in the PAC effect on TL,
which could relate to differences in statistical power, sperm production rates
or testicular telomerase activity. Finally, we review the hypothesis that the
PAC effect on TL may allow a gradual multi-generational adaptive calibration
of maintenance effort, and reproductive lifespan, to local demographic con-
ditions: descendants of males who reproduced at a later age are likely to find
themselves in an environment where increased maintenance effort, allowing
later reproduction, represents a fitness improving resource allocation.
This article is part of the theme issue ‘Understanding diversity in
telomere dynamics’.
1. Introduction
Telomeres are repeating nucleotide sequences found at the ends of chromosomes
that shorten in dividing cells as they proceed through the cell cycle [1,2]. As a
result, telomeres decrease in length with age in most human tissues [3] and
frequently in somatic tissues of metazoans [4].This shortening can eventually
place limits on further cell replication. Environmental factors, like smoking,
inflammation and infection shorten telomere length (TL), but also have
TL-independent effects on biology. Thus, correlations between TL and health
outcomes should not be assumed to be directly causal. Still, converging
evidence from molecular biology, and experimental and epidemiological studies
strongly suggest that shorter TL can directly influence health, particularly by
impairing immune and cardiovascular function [5– 8].
In contrast to the tendency for TL to decrease in length with age in most
human tissues, sperm TL shows a positive correlation with age [7,9]. In accord-
ance with the fact that DNA carried in sperm contribute half of each offspring’s
autosomal genome, offspring of older fathers have longerTLs [7,10]. This putative
paternal age at conception (PAC) effect on TL is intriguing because it appears to be
a rare case of intergenerational genetic plasticity in which the DNA passed on to
offspring is systematically changed based upon the age at reproduction of one’s
father. This finding, combined with evidence that the PAC effect accumulates
across at least two generations of recent ancestors, has led us to hypothesize
that the PAC effect on TL represents an example of adaptive intergenerational
&2018 The Author(s) Published by the Royal Society. All rights reserved.
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
plasticity whereby an offspring’s investment in telomere-
dependent maintenance effort (e.g. immunity) might be
adjusted based on the average ages of reproduction of recent
male ancestors [7,11,12].
In this review, we explore the mechanistic, comparative and
evolutionary dimensions of the PAC effect on TL. To this end,
we first examine the biology of sperm formation which may
underlie the PAC effect. We then briefly consider evidence
that PAC associations with TL are caused by sperm TL increas-
ing with paternal age and whether a maternal age at
conception (MAC) effect on TL exists in humans. We proceed
to examine what is known about the PAC effect on TL across
species. Because telomeres are a general feature of eukaryotic
nuclear chromosomes [13], cross-species variability can
provide insights into the evolutionary bases and possible func-
tion of TL, including the PAC effect on TL. We conclude our
review by considering whether the PAC effect on TL could
represent an adaptation allowing relatively rapid intergenera-
tional adjustment of somatic maintenance effort—and
perhaps even female reproductive lifespan—based upon
predicted environmental and demographic conditions.
2. Mechanisms leading to PAC effect on TL
in humans
Evidence that telomeres lengthen with age in human sperm,
a finding that runs counter to their behaviour in most cell
lines and tissues, has led to a search to explain these findings.
Explanations fall into two broad and not mutually exclusive
categories. First, there is some evidence that especially high
levels of telomerase in testes could lead to a progressive
lengthening of TL with age in sperm. Second, there may be
selective loss of spermatogonial stem cells with shorter TL
and gains of spermatogonial stem cells with longer TL.
Below, we discuss the logic and empirical evidence for each
of these scenarios to help set the stage for a broader discus-
sion of telomeres from the perspectives of comparative and
evolutionary biology.
(a) Telomerase
Since males need to produce a constant supply of sperm via cell
division, mechanisms must be in place to allow this production
despite the shortening of TL which occurs with each round of
DNA replication. Telomerase, a reverse-transcriptase enzyme
that carries an RNA template molecule coding for the telomere
DNA sequence, is a chief mechanism via which telomeres
are lengthened. Telomerase is generally inactive in postnatal
human somatic tissues, but is active across adult life at high
levels in men’s testes [14], and appears to be critical for contin-
ued sperm production. Spermatogonia are the diploid stem
cells that give rise to a series of daughter cells, including pri-
mary spermatocytes (diploid) and secondary spermatocytes
(haploid) which eventually result in terminally differentiated
spermatozoa (sperm).The RNA component of telomerase is
expressed at high levels in human spermatogonia [15], con-
sistent with TLs being actively maintained via lengthening in
the relevant stem cell population. The telomerase RNA tem-
plate component is also highly expressed in primary and
secondary spermatocytes [15] suggesting that sperm TL exten-
sion with age could be at least partially due to extension of TL
not in the spermatogonia, but in differentiated spermatogonia
descendant cells. In humans, TLs increase between the
spermatogonia and primary spermatocyte stage and then
decrease with further differentiation towards spermatozoa
[16]. However, older men show similar or slightly greater
rates of TL decline between spermatogonia and differentiated
spermatozoa, suggesting that the PAC effect on sperm TL is
probably due to changes in spermatogonial stem cell TL [16].
If spermatogonial telomerase activity is sufficiently high it
could not only maintain TL in spite of continued sperm
production, but also progressively lengthen telomeres with
age [14,15,17– 21].
(b) Selective changes in spermatogonia
Selective survival and/or replication of spermatogonia with
long telomeres is another potentially compelling explanation
for the PAC effect on TL in humans—although one not
necessarily mutually exclusive with that of testicular telomer-
ase activity [22,23]. Evidence for such an effect comes from a
study documenting shifts in the distribution of spermatogo-
nial TL with age. As Kimura et al. [22] pointed out, if
spermatogonia TLs were extended at a uniform rate within
individuals (e.g. via telomerase) this leads to the expectation
that the distribution of sperm TL will shift to the right with
age while maintaining the same shape of the distribution
(figure 1a—Scenario A). Instead, these authors found that
older men not only have longer average sperm TL than
younger men, but that they have relatively fewer short and
more long telomeres in their sperm (figure 1b, consistent
with figure 1a—Scenario B) [22]. One possible explanation
for this change in distribution of sperm TL with age is selec-
tive changes wherein spermatogonia with shorter TL are less
likely to survive or replicate to form daughter spermatogonia
than spermatogonia with longer TL.
Furtherclues to the biology of the PACeffect may come from
examinations of not just the changes in the mean TL of sperm
and offspring with age, but also the changes in TL variance.
While analysis suggests that sperm TL variances increase as a
man ages [22,24], a twin study deploying a novel analysis
suggest that the range of sperm TL transmitted to the next gen-
eration decreases with paternal age (i.e. by comparing the
similarity in TLs between dizygotic twins with different aged
fathers). This apparent contradiction between sperm TL and
the TLs passed on by fathers to their offspring via sperm is unre-
solved. One possible explanation is that the increasing variance
in sperm TL with age is driven by greater within-sperm
(i.e. across chromosome end) variation in sperm TL, even as
between sperm mean TL variation decreases with age. Alterna-
tively, it is possible that there is a selective bias such that only a
subset of the sperm with a lower variance in TL in older men are
capable of fertilizing and producing viable offspring. Indeed,
there is some evidence that sperm functionality varies with TL
[25– 29]. Regardless, future studies should be attentive not just
to changes in mean TL in sperm and offspring with increasing
age, but also to changes in variances and distributions of TL
across chromosomes, sperm cells and offspring.
(c) Is the PAC effect on TL due to selfish
spermatogonial selection?
Related to the selection scenarios outlined above, de novo
germ line mutations that drive a recently described phenom-
enon called selfish spermatogonial selection (SSS) [30,31] could
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
2
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
theoretically also contribute to a progressive lengthening of
average sperm TL with age. SSS is a phenomenon wherein
some spermatogonial stem cells acquire mutations that affect
the phenotype of the spermatogonia in a way that increases
their rate of division to form daughter spermatogonia.
Mutations which cause SSS seem to be in pathways regulating
self-renewal and differentiation of spermatogonial stem cells
[30] and have shared signatures with preneoplastic somatic
mutations [31,32]. SSS explains some autosomal-dominant
disorders that increase in prevalence with PAC [31,32].
Spermatogonial stem cells which are capable of expanding
in a pattern consistent with SSS seem to gain their selective
advantage via signalling pathways including PI3 K (phospha-
tidylinositol-3 kinase)/AKT, and STAT3 (signal transducer and
activator of transcription 3) [33,34]. These same pathways also
play a role in regulating telomerase activity [35– 37], hinting at
the potential for shared biology between SSS and the PAC on
TL—and possibly even that de novo mutations that drive SSS
could increase telomerase activity. Because these highly prolif-
erative cell lines become numerically dominant with time, any
tendency for these cells to express telomerase at higher levels,
and increase sperm TL, could cause average sperm TL to
increase with age.
Despite these shared molecular pathways between SSS and
the PACeffect on TL, it is presently unclear if SSS contribute to
the PAC effect on TL. Available data provide mixed support for
such a role. On the one hand, and as noted above, age-related
shifts in the distribution of sperm TL suggest that the PAC
effect on TL is at least partially driven by selective losses of
spermatogonia with shorter TL and gains in spermatogonia
with longer TL. These findings are potentially consistent with
the selective expansion of long TL spermatogonia described
in the SSS model. However, the SSS model also leads to the
expectation that SSS mutations increase exponentially in fre-
quency with age (see [38]). In contrast, this expectation is not
confirmed for the PAC effect on TL. Instead in humans both
sperm and offspring TL appear to increase linearly, not expo-
nentially, with advancing PAC. In contrast to the nonlinear
changes in SSS, telomerase activity extends telomeres during
the DNA replicative phase (S phase) of the cell cycle in
humans [39], such that a slight but persistent overshooting of
expression of testicular telomerase is expected to lead to a
linear increase in sperm TL with age, which is consistent
with the character of the PAC– TL association as documented
thus far in humans.
This line of investigation suggests multiple areas for future
research. First, while the PAC effect on TL has been observed to
be linear, large and well-powered studies should continue to
examine whether exponential models are a better fit for the
influence of male age on sperm and offspring TL, which
would be consistent with a role for SSS. Second, important evi-
dence for SSS has been demonstrated via examination of the
clustering of mutations within the testes [40]. SSS leads to the
expectation that spermatogonia with SSS mutations will tend
to cluster together as some spermatogonia symmetrically
divide and take over a region of testes [30]. If SSS is driving
the PAC effect on TL we should find that long TL in spermato-
gonia also tends to cluster together. Since mutations which
cause SSS are in pathways which also regulate telomerase
activity, it is possible that these SSS mutations tend to cause
increased telomerase activity and thereby longer telomeres. If
de novo mutations that drive SSS also cause higher telomerase
activity and longer TL then clusters of spermatogonia with
these mutations should also tend to have higher telomerase
activity and longer TL.
3. Is the PAC association with TL causal in
humans?
Despite the plausibility of the above scenarios linking advan-
cing paternal age with sperm telomere lengthening, it remains
possible that the PAC association with TL in humans is not
causal. The most obvious non-causal scenario is one in which
men who inherited longer TLs at birth tend to live healthier
and/or longer lives. Even if sperm TL did not change at all
with male age, we might then see that healthier men with
longer TL are more likely to reproduce at later ages and more
likely to volunteer to donate sperm (used in telomere studies)
at later ages. Because mortality is low in early adulthood and
high in late adulthood, this non-causal scenario leads to the
expectation that the PAC effect should be most apparent, or
A
B
12
young
old
frequency (%/kb)
03020
MW (kb)
MW (kb) 10
2
4
6
8
10
(a)(b)
Figure 1. Hypothetical and empirical changes in distributions of sperm TL with age. (a) Two hypothetical scenarios for shifts in TL distributions in sperm from young
and older donors. Scenario A depicts a uniform shift in the TL distribution towards longer telomeres in sperm of older men (interrupted blue line) from that of
younger men (red line). Scenario B depicts a reconfiguration of the distribution, so that shorter telomeres are disproportionably ‘under-represented’, while longer
telomeres are ‘over-represented’. (b) Frequency distributions of TLs in sperm from eight young (18– 19 years; red triangles) and eight older donors (50 –59 years;
blue circles) based on averaging the within-ejaculate variation in TL from southern blot telomere restriction fragment length analysis across individuals. Adapted
from Kimura et al. [22]. (Online version in colour.)
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
3
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
perhaps only apparent, among offspring born to fathers of
advanced age. Contrary to this expectation, past studies point
to a PAC effect on sperm and offspring TL that is linear and
apparent even when fathers are young [22,41]. If men with
longer TL were more likely to donate sperm samples with
advancing age, than we would expect these donors to not
only have longer sperm TL, but also longer blood TL. Contrary
to this, within the same men, sperm TL shows a cross-sectional
increase with age while blood TL shows a cross-sectional
decrease [42]. Thus, although selection of men with longer TL
is a theoretically possible explanation for the PAC effect on
TL, multiple lines of evidence argue against this explanation.
Factors such as birth order or socio-economic status (SES)
might also be correlated with PAC but affect offspring TL via
pathways independent of changes in sperm TL with paternal
age. For example, older fathers could have higher SES and
being raised in a higher SES household could promote
increased TL in their offspring. If the PAC– TL association
were due to factors such as health, SES, or birth order we
would expect the PAC effect to be attenuated after statistical
adjustment for these factors, and to change across ecological
and cultural contexts where these factors and their inter-
correlations vary. Contrary to this, the PAC association with
offspring TL is reported in all human populations studied,
including in the US, Canada, UK, Denmark and the Philippines
[10,22,41,43]. The PAC effect has been found to not be attenu-
ated appreciably by adjustment for SES, birth order or other
factors [41,44]. Similarly, there is no evidence that the human
PAC effect varies with offspring age [10,22,41,43,45,46] (elec-
tronic supplementary material, table S1) [41], as might be
expected if age-related changes in paternal provisioning ability
led to variable rates of postnatal TL attrition across indivi-
duals born to fathers varying in age (see evidence for this in
European shag: [47]). Finally, our closest living relatives, chim-
panzees, show a similar PAC–TL association despite having
very different mating, rearing (e.g. no paternal care) and
social systems than humans [48]. Collectively, these findings
suggest that these other sources of confounding are unlikely
to account for the PAC– TL association.
4. Is there a maternal age effect on TL in
humans?
Because oocytes are all produced prenatally, while sperm are
continually produced throughout life, it is thought that there
is more potential for TL plasticity with age in sperm than in
oocytes. Nonetheless, studies in humans tend to show a positive
correlation between MAC and offspring TL. One challenge in
assessing a MAC effect on offspring TL comes from the typi-
cally strong age correlation between reproductive partners in
humans. That is, offspring of older mothers may have longer
TL simply because the offspring’s fathers are also older.
Supporting the hypothesis that PAC is of primary impor-
tance in determining offspring TL in humans, controlling for
MAC in regression models tends to lead to a larger PAC
effect than when not controlling for MAC (figure 2a—most
values greater than 1). Comparing the PAC and MAC effects
included together in the same regression model, most human
studies show evidence for a negative association between
MAC and offspring TL (figure 2b—most values less than 0).
Both of these patterns are particularly apparent in larger
studies (farther to the right in both figures), which due to
their greater statistical power are likely to have more reliable
estimates. The Eisenberg et al. [41] study of 295 Filipino
women stands out as an outlier among the studies investi-
gated. Of note, that study focused on a small subset of the
sample (16% of a larger sample) for which data on PAC and
MAC were opportunistically available. In contrast, another
cohort from the same population (the offspring of these
women) with more complete MAC and PAC data (Eisenberg
et al. [41], N¼1711) shows results more consistent with other
0.4
0.6
0.8
1.0
1.2
1.4
1.6
Kimura et al. [22]
Unryn et al. [10]
Kimura et al. [22]
Eisenberg et al. [41]
Arbeev et al. [50]
Arbeev et al. [50]
Eisenberg et al. [41]
Kimura et al. [22]
Broer et al. [46]
44 111 235 295 995 1182 1711 1954 5127
PAC b with MAC in model/PAC b without
MAC in model
N
(a)
–1.0
–0.5
0
0.5
1.0
1.5
Kimura et al. [22]
Unryn et al. [10]
Kimura et al. [22]
Eisenberg et al. [41]
Arbeev et al. [50]
Arbeev et al. [50]
Eisenberg et al. [41]
Kimura et al. [22]
Broer et al. [46]
44 111 235 295 995 1182 1711 1954 5127
MAC b/PAC b (both together in same model)
N
(b)
Figure 2. Comparing the relative strength of the PAC and MAC effects on offspring TL in human studies: (a) assesses how the PAC effect on offspring TL changes
after controlling for the MAC effect. Calculated by examining PAC effect in a regression model including MAC and one not including MAC and then dividing PAC effect
in first model by PAC effect in second. Values greater than 1 indicate that PAC effect increases when controlling for MAC, suggesting that PAC effect is not driven by
MAC. (b) Assesses direction and relative magnitude of MAC effect compared to PAC effect (the ratio of MAC effect/PAC effect). Values greater than 1 indicate a greater
positive MAC effect than PAC effect. Values less than zero indicate MAC effects in opposite direction of PAC effects (i.e. shorter offspring TL with older MAC). Statistics
from Eisenberg et al. [41] has been updated from the original publication based on measurement improvements in the same dataset [49]. Studies with multiple
entries include separate estimates for independent cohorts. (Online version in colour.)
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
4
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
studies. We speculate that biases introduced by opportunistic
sampling might account for the outlier results among the
older cohort in this study.
The tendency towards a negative association between
MAC and offspring TL (figure 2b) might be explained if
oocytes that had fewer mitotic divisions tend to be ovulated
earlier in the lifecycle than those ovulated later—although it
is not clear if this occurs in humans [51]. In cows, it has been
shown that TLs in immature oocytes are shorter in older
cows than in younger ones, providing some support for this
notion [52]. However, in cows TLs elongate during embryo
development up until the blastocyst stage and TL has not
been shown to differ between young and old cows at the blas-
tocyst stage [52], suggestingthat initial differences in oocyte TL
may be obscured or eliminated through TL changes over the
course of development. Because telomere and telomerase
biology vary markedly across species [53,54], it may be that
more of the initial oocyte differences in TL with MAC are
retained with age in humans compared to other species.
Together, available evidence suggests that, at least in
humans, older PAC probably acts to extend offspring TL
while older MAC may slightly decrease offspring TL. This
apparent negative MAC effect on TL could result if oocytes
formed after more mitotic divisions, which will tend to
decrease TL, tend to be ovulated later in life. Evidence in
other species, among which parental ages are less strongly
correlated and can be experimentally manipulated, may
provide more definitive answers.
5. Cross-species evidence
Because telomeres are general features of eukaryotic nuclear
chromosomes [13], additional insights into their function, and
the extent of variability and possible function of the PAC
effect on TL, may be gained by looking at these processes
across species differing in life-history and reproductive strat-
egies. Non-human studies also sometimes have the advantage
of allowing the use of experimental approaches (e.g. selecting
the age of mating), low correlations between parental ages,
and more invasive and precise tissue sampling. High testicular
telomerase levels are probably a general feature of mammals;
indeed, high telomerase activity has been found in testes in
all mammalian species in which it has been examined, includ-
ing 15 rodent species spanning from the small, short-lived
mouse to the long-lived naked mole-rat and large capybara
[54], cats, dogs [55], long-tailed macaques [21] and humans [14].
As summarized in table 1, the PAC effect on TL has only
been studied in eight species. Findings are highly variable
across species, hinting at rapid evolution of telomere/sperm
production biology. Only half of the studied species show stat-
istically significant associations between PAC and offspring/
sperm TL—two of these associations are negative and two are
positive. Although the remaining null findings could reflect
the true nature of the biology of these species, it is important
to also consider the effect of statistical power as a limitation.
All else equal, the greater the range of variation in PAC, the
greater the statistical power to detect relationships. As illus-
trated in supplementary figure 1, assuming the yearly PAC
effect size observed in humans, PAC effects in species with
limited variation of PAC require large sample sizes to detect
an association. For example, samples greater than 1000 and
4000, respectively, would be needed to reliably detect a PAC
of human magnitude in European shagsand Soay sheep. Incon-
trast, assuming the much greater PAC effect found in
chimpanzees, the more manageable sample sizes of 80 and
346 would be sufficient. Conversely, the significant negative
PAC effects found in mice and sand lizards, despite limited
power, imply a very large magnitude effect on TL in order to
be detected in these species (table 1). The negative PAC effects
on TL in mice shows that even high testicular telomerase
activity, as has been documented in this species [54], may not
be sufficient to increase sperm TL with age, or that testicular tel-
omeraseactivity is by itself insufficient to explain the PAC effect.
Building on the idea that the continual need for production
of sperm is a key driver of the PAC effect on TL, Eisenberg et al.
[48] predicted that the PAC effect should be larger in species
with greater sperm production rates. This prediction gains
only limited support in the sparse available data. The chimpan-
zee–human comparison is particularly informative given
the close phylogenetic relationship between the two species.
Chimpanzees experience sperm competition secondary to
Table 1. Paternal age effect on offspring TL across species. Bold values indicate p,0.05.
genus species common name N
PAC s.d.
(years) r
a
p
relative
testis size
b
refs
Lacerta agilis sand lizard 12 1.38
c
20.59 0.041 [56]
Phalacrocorax aristotelis European shag 204 4.38 þ0.43 [47]
Acrocephalus arundinaceus great reed warbler 154 þ0.7 [57]
Mus musculus mouse 12
c
0.39 2
0.05 0.5 [58]
Ovis aries Soay sheep 318 2.17 0.066 0.238 4.9/1.2
d
[59]
c
Macaca fascicularis
e
long-tailed macaque 9 7.8 þNS 2.3 [21]
Homo sapiens Human 144 8.2 0.15 0.03 0.5 [48]
Pan troglodytes chimpanzee 40 6.4 0.42 0.009 1.5 [48]
a
Correlation values if reported, otherwise ‘þ’ indicates positive association and ‘2’ negative association.
b
Ratio of observed mass of testis to that predicted by body mass from [60].
c
Personal communication.
d
Breeding season/non-breeding season calculated from [61].
f
Testicular TL instead of offspring TL.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
5
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
their promiscuous mating system, which has led to a higher
rate of sperm production compared to humans [62,63]. Consist-
ent with an effect of sperm production rate on the magnitude of
the PAC effect on TL, chimpanzees show increases in TL with
each year of delayed paternal conception that are an estimated
six-fold greater than in humans ( p¼0.026) [48]. In contrast,
Soay sheep, which have a much larger relative testes size
(and probably, sperm productive rates) even than chimpan-
zees, lack a discernible PAC effect on TL (table 1). Possibly
explaining this, Soay sheep are seasonal breeders whose rela-
tive testes size reduces considerably in the off season
(table 1), which probably reflects an attenuated sperm pro-
duction rate [64]. Additionally, as noted above, statistical
power to detect a PAC effect on TL may be low in this species.
In sum, while all 19 measured mammalian species show
evidence for high testicular telomerase activity, evidence of
PAC associations with offspring TL show little consistency
across species. Depending on the species, PAC has been
shown to relate to offspring TL in positive or negative direc-
tions, while in others there is no evidence for an association.
Some of the null findings might be a result of low statistical
power. The most consistent pattern is a tendency for catarrhine
primates to have positive PAC effects—as evident from signifi-
cant positive relationships between PAC and offspring TL in
humans and chimpanzees and suggestive evidence for a
similar relationship in a small sample of long-tailed macaques.
6. Does the PAC effect on TL allow adaptive
transmission of plasticity-generated genetic
information across generations?
Thus, evidence for relationships between PAC and offspring
TL are most consistently observed in humans and their close
primate kin. Although it is presently uncertain if there is a
direct causal relationship between PAC and offspring TL in
humans, such an effect is a likely explanation in our view, in
light of the converging evidence reviewed above (§3). Irrespec-
tive of the specific underlying mechanisms, an effect of PAC on
offspring TL would represent an unusual form of intergenera-
tional genetic plasticity in which the DNA passed on to
offspring varies systematically based on the father’s age at
reproduction. Age at reproduction is an event that is of funda-
mental importance from an evolutionary perspective.
Organisms have limited energy and other substrate that can
be devoted to growth, reproduction, maintenance, activity
or storage [65]. According to some models of aging [66], the
optimal allocation of resources varies primarily based on una-
voidable mortality and other environmental circumstances.
When the likelihood of living and reproducing into the
future are high, intensive investment in maintenance effort at
the expense of reduced immediate reproductive expenditure
can be a good fitness strategy. Conversely, when extrinsic mor-
tality risks are high, optimism about the future is reduced,
shifting the balance in favour of current reproductive expendi-
ture and at a cost to maintenance functions. This evolutionary
life-history perspective has lead us to suggest that the PAC
effect on TL may represent an adaptation whereby TLs
passed on to offspring are systematically altered to improve
the fit to the environmental and demographic setting he or
she is likely to encounter across the life course [7]. That is,
males with the ability to adjust the TL of their offspring in
response to their own age of reproduction may have had off-
spring with higher average fitness—thus selecting for this
capability to adjust sperm TL with age.
One obvious limitation to the utility of the PAC effect on TL
as a reliable signal of the expected age at reproduction is the
noise introduced by stochastic processes, such as the effect of
birth order. A father may sire multiple children at different
ages, thus potentially sending different age signals to offspring
facing the same environment. This problem led Eisenberg [7] to
suggest that TL might not only reflect the father’s age at con-
ception, but could, by virtue of the high fidelity of DNA
replication, also reflect the conception ages of multiple gener-
ations of paternal ancestors going back in time. In this
scenario, the TL that an individual receives would reflect not
just one’s own father’s age, but also that of the grandfather
and other more distant male ancestors, thereby providing a
rolling average age of recent reproductive timing in the lineage
(for a related example, see [67]). Consistent with this pre-
diction, a subsequent study showed that the paternal
grandfather’s age at conception of the father predicted the
grandchild’s TL, independent of and additive to, the effect of
the father’s age at conception [41]. Studies of socio-economic
status across many generations show high levels of continuity
within families across multiple countries around the world
[68], suggesting a level of consistency that could make the
PAC effect on TL a good adaptive signal of local conditions.
We also recognize, however, that this same stability in socio-
economic status may complicate efforts to distinguish a direct
PAC effect on offspring TL from other processes like niche
construction, genetic, epigenetic and cultural transmission.
It is important to note that, older fathers transmitting longer
telomeres to their offspring than younger fathers do not imply
that TLs increase with each generation. The testes are formed
from the intermediate mesoderm early in pre-natal develop-
ment [69]. Sperm production in the testis begins at puberty,
but adult-like sperm concentration, morphology and motility
are not evident until some years later [69,70]. This developmen-
tal sequence involves many cell divisions, which probably
allow opportunities for TL shortening due to cell replication
and DNA damage, prior to the onset of any mechanisms that
cause lengthening of TL in spermatogonia with age. It is pre-
sently not clear at what age (if any) a man would need to
reproduce in order to transmit longer telomeresto his offspring
than he himself received. However, all else equal, intergenera-
tional lengthening of sperm TL is made more likely as
the conception ages of one’s father, and his recent paternal
ancestors, is delayed.
(a) What are the potential benefits of inheriting long
telomeres from late-reproducing male ancestors?
If the PAC effect on TL does provide a reliable cue of local
demographic conditions, as we hypothesize, it is of interest
to consider the biological changes that might be calibrated in
response to this information, and the trade-offs probably
involved. As alluded to above, because TL provides a limit
on how many times a cell can replicate, longer TL is expected
to enhance cell proliferation-dependent functions. Because
mounting an immune response often involves cell prolifer-
ation-dependent processes, this leads to the expectation that
longer TL should manifest as more robust cell-proliferation-
dependent components of immunity, and possibly a slowing
of the pace of immune senescence with advancing age.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
6
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
Indeed, experimental and longitudinal observational evidence
in vitro, in mice, and in humans suggests that longer TL pro-
motes resistance to infection, and that these effects are
evident in young adulthood and possibly infancy [7,71– 74].
Longer TL are also likely to reduce the likelihood of athero-
sclerosis by improving vascular maintenance and reducing
the build-up of senescent cells in atherosclerotic plaques
[75,76]. Additionally long TLs probably improve wound
healing, which also relies upon cellular proliferation [77].
Another biological function that depends upon cell prolifer-
ation is gamete formation. As discussed above, in females the
pool of oocytes is established in its entirety during fetal life,
and menopause occurs when the pool of follicles is depleted.
It has been hypothesized that TL could influence the pace of
female reproductive senescence, including late-life fecundity,
by influencing the pool of initial oocytes formed, and their via-
bility as a woman ages [78,79]. There is evidence that TL
measured in blood cells is shorter in women who experience
menopause at an earlier age—although this finding has not
been replicated across all ethnic groups [78– 80]. While links
between TL and female reproduction await additional vali-
dation in humans, they add to the list of candidate cell
replication-dependent processes in offspring that might benefit
as a result of inheriting longer TL from older fathers.
(b) The potential costs of inheriting long telomeres
Given that longer telomeres very probably improve functions
that require cell proliferation, it also seems likely that long tel-
omeres will carry costs. Otherwise we would expect the rapid
evolution and fixation of inherited (i.e. germ line) long telo-
meres to avoid any functional constraints imposed by short
telomeres, and thus no association between TL and fitness
relevant phenotypes in the general population. The most com-
monly cited fitness cost associated with inheriting long TL is an
increased risk of cancer. The theoretical reasons to expect a link
between longer telomeres and cancer risk are clear: because
accumulation of oncogenic mutations is dependent on cell
replication, a cell lineage with shorter TL is less likely to gain
the necessary mutations for cancer development before cell
proliferation stops due to critically short TL [76]. However,
there are also pathways via which longer TL could promote
cancer protection. Short TLs are known to cause chromosomal
instability and fusions which can promote cancer, although it
has been argued that this only occurs in rare pathological con-
ditions and is of limited relevance to the general population
[76]. Longer TL also promotes improved immune function
[7,71– 73], which plays a key role in combating cancer-inducing
pathogens and parasites (e.g. human papillomavirus and
Helicobacter pylori), as well as fighting off cancers once they
begin to develop [7,81].
Empirical results have not yet given a clear resolution to the
TL-cancer question. Although telomerase shows high activity
in the vast majority of cancers and the activation of telomere
maintenance pathways in somatic tissues is likely to be critical
for cancer development, it is less clear what role inheriting
longer telomeres—germ line TL—plays in cancer risk [7].
Results from prospective human studies tend to show that
longer somatic TL is associated with decreased cancer risk or
has no association with cancer at all [82–85]. In contrast, gen-
etic polymorphisms associated with longer TL tend to
predict increased cancer risk—particularly for rarer cancers
[6,84,86,87]. These contrasting results could be due to the fact
that measured somatic TL are confounded by environmental
variables that shorten TLs and increase cancer risks via other
pathways (e.g. smoking), or pleiotropic effects of TL-associated
genetic polymorphisms that increase cancer risk via germ line
TL-independent pathways [84,88,89].
Another factor that reduces the likelihood that cancer has
been a key factor shaping within-species variation in TL in
humans is the fact that in human cancer mortality is very
rare until middle age and mostly occurs in the elderly [7].
Additionally, human cancer is thought to have been much
less prevalent before recent changes coincident with the demo-
graphic transition and industrialization [7]. While humans
engage in extensive intergenerational transfers of resources
that allow us to continue to contribute to inclusive fitness late
in life, the force of selection still declines with age in
humans—such that late-life traits are less shaped by natural
selection than those earlier in life [7]. The presence of a PAC
effect on TL in chimpanzees, among whom many of these fac-
tors do not apply, suggests that the PAC effect on TL that is
documented in human populations very probably predates
the human– chimpanzee split. Given this, other TL-influenced
(i.e. cell proliferation dependent) phenotypes which influence
health earlier in life, such as immunity and wound healing
ability, may be more likely to shape optimal TL.
In the absence of clear costs to longer telomeres, or of
benefits of having shorter TL, Eisenberg proposed the thrifty
telomere hypothesis [7]. The thrifty telomere hypothesis suggests
that longer telomeres promote increased maintenance effort
and thereby health and longevity—but that this comes at ener-
getic costs. Increased investments in long-term maintenance
mean less energy available for current reproductive invest-
ments, a strategy that could reduce fitness in environments
with high extrinsic mortality risks. By the reasoning of this
model then, in low mortality environments, offspring inherit
longer telomeres from multiple recent generations of male
ancestors, who reproduced on average at later ages. This
signal of late-life survival and reproduction in turn facilitates
a shift in resource allocation away from immediate repro-
duction and in favour of enhanced maintenance effort and
lifespan extension—via boosted immunity, wound repair,
and perhaps even oocyte number and longevity. This model
leads to several testable predictions, including that individuals
who inherit longer TL will exhibit evidence for greater main-
tenance expenditures at an expense to reproduction, and that,
controlling for potential confounders, secular trends in PAC
will predict improved immune function in progeny.
7. Conclusion
The putative PAC effect on TL represents a novel type of inter-
generational plasticity. We are not aware of any other such
systematic alteration to DNA sequences that are transmitted
to offspring. As we have reviewed, there is convergent evi-
dence that this PAC effect on TL is due to progressive
lengthening of sperm TL as human males age (as we note, it
is at present less clear how widespread this phenomena is in
other species). The biological mechanisms underlying these
increases in sperm TL are less certain, but could include
some combination of high testicular telomerase activity, selec-
tive loss of spermatogonia with short TL and/or increased
proliferation of spermatogonia with long TL. We also reviewed
parallels between selfish spermatogonial selection and the PAC
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
7
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
effect on TL which might reflect shared biology. Unlike the
plasticity created by continued sperm production with age,
oocytes are produced prenatally and correspondingly there is
limited evidence for a MAC effect on offspring TL. If a MAC
effect exists, it is likely to be negative rather than positive.
The PAC effect on TL has only been studied across eight
species and shows considerable variation, with chimpanzees
showing a positive six-fold greater PAC effect than humans,
other species showing negative PAC effects and several show-
ing no associations at all. We note that statistical power to
detect PAC effects will be limited by a combination of variation
in PAC within a species and study sample size, and that some
past studies are thus underpowered to detect human-sized
PAC effects. In the future, appropriately powered comparative
studies will help clarify the extent, and possible function, of
PAC–TL associations across a wider range of species.
We have suggested that the PAC effect on TL could rep-
resent an adaptive intergenerational signalling mechanism.
Specifically, age at reproduction is a key factor shaping life-
history strategies and having had multiple generations of
recent male ancestors who reproduced at later ages may
indicate that similar demographic conditions are likely to be
experienced in the near future. Longer TLs are postulated
to promote increased maintenance effort, particularly via
improved immune function and wound healing. Maintenance
effort is energetically costly and has a greater fitness payoff in
contexts of lower extrinsic mortality and delayed reproduction.
In addition, there is emerging evidence that inheriting longer
telomeres could also allow increased proliferation of follicles
in female offspring, which might delay the timing of meno-
pause and cessation of reproductive senescence when recent
male ancestors have delayed reproduction themselves.
Other similar adaptive intergenerational signalling mech-
anisms have been suggested [12] but it remains unclear how
common or adaptive such phenomena are [90– 92]. The PAC
effect on TL is noteworthy, and distinct from many examples
of maternal-offspring signalling, for being multi-generational
[41] which may allow for the communication of an integrated
signal reflecting average age at conception among multiple
recent generations of paternal ancestors [12]. Further, the
PAC effect on TL is based in changes to DNA sequences
and thus provides an intergenerational signalling pathway
that is mechanistically well-established and of high fidelity.
Data accessibility. This article has no additional data.
Authors’ contributions. D.T.A.E. led the researching, writing and design of
the article. C.W.K. contributed to refining the ideas, writingand revising.
Competing interests. We have no competing interests.
Funding. We received no funding for this study.
Acknowledgements. Thanks to Neil Metcalfe, Tiffany Pan, Rebecca Ortega,
Robert Tennyson, Peter Rej and an anonymous reviewer for feedback
on an earlier draft of this manuscript; Abraham Aviv, Anne Goriely,
Alexander Hill, and ‘Understanding Diversity in Telomere Dynamics’
conference attendees and organizers, for valuable discussions; Mats
Olsson, Pablo Bermejo-A
´lvarez, Dan Nussey and Hannah Froy for
sharing data for table 1; and Karl Riabowol for sharing data used to
test whether PAC varies by own age in this manuscript.
References
1. Olovnikov AM. 1971 Principle of marginotomy in
template synthesis of polynucleotides. Dokl. Akad.
Nauk SSSR 201, 1496– 1499.
2. Harley CB, Futcher AB, Greider CW. 1990 Telomeres
shorten during ageing of human fibroblasts. Nature
345, 458–460. (doi:10.1038/345458a0)
3. Daniali L, Benetos A, Susser E, Kark JD, Labat C,
Kimura M, Desai K, Granick M, Aviv A. 2013
Telomeres shorten at equivalent rates in somatic
tissues of adults. Nat. Commun. 4, 1597. (doi:10.
1038/ncomms2602)
4. Young AJ. 2018 The role of telomeres in the
mechanisms and evolution of life-history trade-offs
and ageing. Phil. Trans. R. Soc. B 373, 20160452.
(doi:10.1098/rstb.2016.0452)
5. Cawthon RM, Smith KR, O’Brien E, Sivatchenko A,
Kerber RA. 2003 Association between telomere
length in blood and mortality in people aged 60
years or older. Lancet 361, 393 – 395. (doi:10.1016/
S0140-6736(03)12384-7)
6. The Telomeres Mendelian Randomization C. 2017
Association between telomere length and risk of
cancer and non-neoplastic diseases: a mendelian
randomization study. JAMA Oncol. 3, 636. (doi:10.
1001/jamaoncol.2016.5945)
7. Eisenberg DTA. 2011 An evolutionary review of
human telomere biology: the thrifty telomere
hypothesis and notes on potential adaptive paternal
effects. Am. J. Hum. Biol. 23, 149– 167. (doi:10.
1002/ajhb.21127)
8. Aviv A, Kark JD, Susser E. 2015 Telomeres,
atherosclerosis, and human longevity: a causal
hypothesis. Epidemiology 26, 295–299. (doi:10.
1097/EDE.0000000000000280)
9. Allsopp RC, Vaziri H, Patterson C, Goldstein S,
Younglai EV, Futcher AB, Greider CW, Harley CB.
1992 Telomere length predicts replicative
capacity of human fibroblasts. Proc. Natl Acad.
Sci. USA 89, 10 114– 10 118. (doi:10.1073/pnas.89.
21.10114)
10. Unryn BM, Cook LS, Riabowol KT. 2005 Paternal age
is positively linked to telomere length of children.
Aging Cell 4, 97– 101. (doi:10.1111/j.1474-9728.
2005.00144.x)
11. Eisenberg DT, Kuzawa CW. 2013 Commentary: the
evolutionary biology of the paternal age effect on
telomere length. Int. J. Epidemiol. 42, 462–465.
(doi:10.1093/ije/dyt027)
12. Kuzawa CW, Eisenberg DTA. 2014 The Long reach of
history: intergenerational pathways to plasticity in
human lifespan. In Sociality, hierarchy, health
comparative biodemography: a collection of papers
(eds M Weinstein, MA Lane), pp. 65– 94.
Washington, DC: The National Academies Press.
13. Chan SR, Blackburn EH. 2004 Telomeres and
telomerase. Phil. Trans. R. Soc Lond. B 359,
109–121. (doi:10.1098/rstb.2003.1370)
14. Wright WE, Piatyszek MA, Rainey WE, Byrd W, Shay
JW. 1996 Telomerase activity in human germline
and embryonic tissues and cells. Dev. Genet. 18,
173–179. (doi:10.1002/(SICI)1520-6408(1996)
18:2,173::AID-DVG10.3.0.CO;2-3)
15. Yashima K, Maitra A, Rogers BB, Timmons CF, Rathi
A, Pinar H, Wright WE, Shay JW, Gazdar AF. 1998
Expression of the RNA component of telomerase
during human development and differentiation. Cell
Growth Differ. 9, 805– 813.
16. Jorgensen PB, Fedder J, Koelvraa S, Graakjaer J.
2013 Age-dependence of relative telomere
length profiles during spermatogenesis in man.
Maturitas 75, 380–385. (doi:10.1016/j.maturitas.
2013.05.001)
17. Achi MV, Ravindranath N, Dym M. 2000 Telomere
length in male germ cells is inversely correlated
with telomerase activity. Biol. Reprod. 63,
591–598. (doi:10.1095/biolreprod63.2.591)
18. Bekaert S, Derradji H, Baatout S. 2004 Telomere
biology in mammalian germ cells and during
development. Dev. Biol. 274, 15–30. (doi:10.1016/
j.ydbio.2004.06.023)
19. Kim NW et al. 1994 Specific association of human
telomerase activity with immortal cells and cancer.
Science 266, 2011–2015. (doi:10.1126/science.
7605428)
20. Fradiani PA, Ascenzioni F, Lavitrano M, Donini P.
2004 Telomeres and telomerase activity in pig
tissues. Biochimie 86, 7– 12. (doi:10.1016/j.biochi.
2003.11.009)
21. Gardner JP et al. 2007 Telomere dynamics in
macaques and humans. J. Gerontol. A. Biol.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
8
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
Sci. Med. Sci. 62, 367–374. (doi:10.1093/gerona/
62.4.367)
22. Kimura M et al. 2008 Offspring’s leukocyte telomere
length, paternal age, and telomere elongation in
sperm. PLoS Genet. 4, e37. (doi:10.1371/journal.
pgen.0040037)
23. Hjelmborg JB et al. 2015 Paternal age and telomere
length in twins: the germ stem cell selection paradigm.
Aging Cell 14, 701– 703. (doi:10.1111/acel.12334)
24. Antunes DF, Kalmbach K, Wang F, Dracxler R, Seth-
Smith M, Kramer Y, Buldo-Licciardi J, Kohlrausch F,
Keefe D. 2015 A single-cell assay for telomere DNA
content shows increasing telomere length
heterogeneity, as well as increasing mean telomere
length in human spermatozoa with advancing age.
J. Assist. Reprod. Genet. 32, 1685– 1690. (doi:10.
1007/s10815-015-0574-3)
25. Santiso R, Tamayo M, Gosalvez J, Meseguer M,
Garrido N, Fernandez JL. 2010 Swim-up procedure
selects spermatozoa with longer telomere length.
Mutat. Res. 688, 88– 90. (doi:10.1016/j.mrfmmm.
2010.03.003)
26. Thilagavathi J, Kumar M, Mishra S, Venkatesh S,
Kumar R, Dada R. 2013 Analysis of sperm telomere
length in men with idiopathic infertility. Arch.
Gynecol. Obstet. 287, 803–807. (doi:10.1007/
s00404-012-2632-8)
27. Ferlin A, Rampazzo E, Rocca MS, Keppel S, Frigo AC,
De Rossi A, Foresta C. 2013 In young men sperm
telomere length is related to sperm number and
parental age. Hum. Reprod. 28, 3370– 3376.
(doi:10.1093/humrep/det392)
28. Yang Q, Zhao F, Dai S, Zhang N, Zhao W, Bai R, Sun
Y. 2015 Sperm telomere length is positively
associated with the quality of early embryonic
development. Hum. Reprod. 30, 1876– 1881.
(doi:10.1093/humrep/dev144)
29. Yang Q et al. 2015 Processing of semen by density
gradient centrifugation selects spermatozoa with
longer telomeres for assisted reproduction
techniques. Reprod. Biomed. Online 31, 44– 50.
(doi:10.1016/j.rbmo.2015.02.016)
30. Maher GJ, Goriely A, Wilkie AO. 2014 Cellular
evidence for selfish spermatogonial selection in
aged human testes. Andrology 2, 304– 314. (doi:10.
1111/j.2047-2927.2013.00175.x)
31. Goriely A, McGrath JJ, Wilkie AOM, Malaspina D.
2013 ‘Selfish spermatogonial selection’: a novel
mechanism for the association between advanced
paternal age and neurodevelopmental disorders.
Am. J. Psychiatry 170, 599– 608. (doi:10.1176/appi.
ajp.2013.12101352)
32. Rahbari R et al. 2016 Timing, rates and spectra
of human germline mutation. Nat. Genet. 48,
126–133. (doi:10.1038/ng.3469)
33. Choi S-K, Yoon S-R, Calabrese P, Arnheim N. 2012
Positive selection for new disease mutations in the
human germline: evidence from the heritable
cancer syndrome multiple endocrine neoplasia type
2B. PLoS Genet. 8, e1002420. (doi:10.1371/journal.
pgen.1002420)
34. Yoon S-R, Choi S-K, Eboreime J, Gelb, Bruce D,
Calabrese P, Arnheim N. 2013 Age-dependent
germline mosaicism of the most common noonan
syndrome mutation shows the signature of germline
selection. Am. J. Hum. Genet. 92,917 – 926. (doi:10.
1016/j.ajhg.2013.05.001)
35. Heeg S et al. 2011 EGFR overexpression
induces activation of telomerase via PI3 K/AKT-
mediated phosphorylation and transcriptional
regulation through Hif1-alpha in a cellular
model of oral– esophageal carcinogenesis. Cancer
Sci. 102, 351– 360. (doi:10.1111/j.1349-7006.2010.
01796.x)
36. Breitschopf K, Zeiher AM, Dimmeler S. 2001 Pro-
atherogenic factors induce telomerase inactivation in
endothelial cells through an Akt-dependent
mechanism. FEBS Lett. 493, 21– 25. (doi:10.1016/
S0014-5793(01)02272-4)
37. Konnikova L, Simeone MC, Kruger MM, Kotecki M,
Cochran BH. 2005 Signal transducer and activator of
transcription 3 (STAT3) regulates human telomerase
reverse transcriptase (hTERT) expression in human
cancer and primary cells. Cancer Res. 65,
6516–6520. (doi:10.1158/0008-5472.CAN-05-0924)
38. Arnheim N, Calabrese P. 2016 Germline stem cell
competition, mutation hot spots, genetic disorders,
and older fathers. Annu. Rev. Genomics Hum. Genet.
17, 219– 243. (doi:10.1146/annurev-genom-
083115-022656)
39. Tomlinson RL, Ziegler TD, Supakorndej T, Terns RM,
Terns MP. 2006 Cell cycle-regulated trafficking of
human telomerase to telomeres. Mol. Biol. Cell 17,
955–965. (doi:10.1091/mbc.E05-09-0903)
40. Arnheim N, Calabrese P. 2009 Understanding what
determines the frequency and pattern of human
germline mutations. Nat. Rev. Genet. 10, 478– 488.
(doi:10.1038/nrg2529)
41. Eisenberg DT, Hayes MG, Kuzawa CW. 2012 Delayed
paternal age of reproduction in humans is
associated with longer telomeres across two
generations of descendants. Proc. Natl Acad. Sci.
USA 109, 10 251 –10 256. (doi:10.1073/pnas.
1202092109)
42. Aston KI, Hunt SC, Susser E, Kimura M, Factor-Litvak
P, Carrell D, Aviv A. 2012 Divergence of sperm and
leukocyte age-dependent telomere dynamics:
implications for male-driven evolution of telomere
length in humans. Mol. Hum. Reprod. 18, 517– 522.
(doi:10.1093/molehr/gas028)
43. Wojcicki JM, Heyman MB, Elwan D, Lin J, Blackburn
E, Epel E. 2016 Early exclusive breastfeeding is
associated with longer telomeres in Latino
preschool children. Am. J. Clin. Nutr. 104, 397–405.
(doi:10.3945/ajcn.115.115428)
44. Prescott J, Du M, Wong JY, Han J, De Vivo I. 2012
Paternal age at birth is associated with offspring
leukocyte telomere length in the nurses’ health
study. Hum. Reprod. 27, 3622– 3631. (doi:10.1093/
humrep/des314)
45. Factor-Litvak P, Susser E, Kezios K, McKeague I, Kark
JD, Hoffman M, Kimura M, Wapner R, Aviv A. 2016
Leukocyte telomere length in newborns:
implications for the role of telomeres in human
disease. Pediatrics 137, 20153927. (doi:10.1542/
peds.2015-3927)
46. Broer L et al. 2013 Meta-analysis of telomere length
in 19 713 subjects reveals high heritability, stronger
maternal inheritance and a paternal age effect.
Eur. J. Hum. Genet. 21, 1163– 1168. (doi:10.1038/
ejhg.2012.303)
47. Heidinger BJ, Herborn KA, Granroth-Wilding HMV,
Boner W, Burthe S, Newell M, Wanless S, Daunt F,
Monaghan P. 2016 Parental age influences offspring
telomere loss. Funct. Ecol. 30, 1531– 1538. (doi:10.
1111/1365-2435.12630)
48. Eisenberg DTA, Tackney J, Cawthon RM, Cloutier CT,
Hawkes K. 2016 Paternal and grandpaternal ages at
conception and descendant telomere lengths in
chimpanzees and humans. Am. J. Phys. Anthropol.
162, 201–207. (doi:10.1002/ajpa.23109)
49. Eisenberg DT, Kuzawa CW, Hayes MG. 2015 Improving
qPCR telomere length assays: Controlling for well
position effects increases statistical power. Am. J. Hum.
Biol. 27, 570–575. (doi:10.1002/ajhb.22690)
50. Arbeev KG, Hunt SC, Kimura M, Aviv A, Yashin AI.
2011 Leukocyte telomere length, breast cancer risk
in the offspring: the relations with father’s age at
birth. Mech. Ageing Dev. 132, 149– 153. (doi:10.
1016/j.mad.2011.02.004)
51. Segurel L, Wyman MJ, Przeworski M. 2014
Determinants of mutation rate variation in the
human germline. Annu. Rev. Genomics Hum. Genet.
15, 47–70. (doi:10.1146/annurev-genom-031714-
125740)
52. Kin A, Kansaku K, Sumiya M, Itami N, Sirasuna K,
Kuwayama T, Iwata H. 2017 Effect of Aging on
telomere lengths in bovine oocytes and granulosa
cells. J. Mamm. Ova Res. 34, 37–43. (doi:10.1274/
032.034.0101)
53. Gomes NM et al. 2011 Comparative biology of
mammalian telomeres: hypotheses on ancestral
states and the roles of telomeres in longevity
determination. Aging Cell 10, 761– 768. (doi:10.
1111/j.1474-9726.2011.00718.x)
54. Seluanov A, Chen Z, Hine C, Sasahara T, Ribeiro A,
Catania K, Presgraves D, Gorbunova V. 2007
Telomerase activity coevolves with body mass not
lifespan. Aging Cell 6, 45– 52. (doi:10.1111/j.1474-
9726.2006.00262.x)
55. Gorbunova V, Seluanov A. 2009 Coevolution of
telomerase activity and body mass in mammals:
from mice to beavers. Mech. Ageing Dev. 130,3–9.
(doi:10.1016/j.mad.2008.02.008)
56. Olsson M, Pauliny A, Wapstra E, Uller T, Schwartz T,
Blomqvist D. 2011 Sex differences in sand lizard
telomere inheritance: paternal epigenetic effects
increases telomere heritability and offspring
survival. PLoS ONE 6, e17473. (doi:10.1371/journal.
pone.0017473)
57. Asghar M, Bensch S, Tarka M, Hansson B,
Hasselquist D. 2014 Maternal and genetic factors
determine early life telomere length. Proc. R. Soc. B
282, 20142263. (doi:10.1098/rspb.2014.2263)
58. de Frutos C, Lo
´pez-Cardona AP, Fonseca Balvı
´sN,
Laguna-Barraza R, Rizos D, Gutierrez-Ada
´nA,
Bermejo-A
´lvarez P. 2016 Spermatozoa telomeres
determine telomere length in early embryos and
offspring. Reproduction 151,1–7.
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
9
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
59. Froy H et al. 2017 No evidence for parental age
effects on offspring leukocyte telomere length in
free-living Soay sheep. Sci. Rep. 7, 9991. (doi:10.
1038/s41598-017-09861-3)
60. Kenagy G, Trombulak SC. 1986 Size and function of
mammalian testes in relation to body size.
J. Mammal. 67, 1– 22. (doi:10.2307/1380997)
61. Lincoln G, Short R. 1980 Seasonal breeding: nature’s
contraceptive. Recent Prog. Horm. Res. 36, 1– 52.
62. Venn O, Turner I, Mathieson I, de Groot N, Bontrop
R, McVean G. 2014 Strong male bias drives
germline mutation in chimpanzees. Science 344,
1272–1275. (doi:10.1126/science.344.6189.1272)
63. Wong A. 2014 Covariance between testes size and
substitution rates in primates. Mol. Biol. Evol. 31,
1432–1436. (doi:10.1093/molbev/msu091)
64. Johnson L, Varner DD, Roberts ME, Smith TL, Keillor
GE, Scrutchfield WL. 2000 Efficiency of
spermatogenesis: a comparative approach. Anim.
Reprod. Sci. 60– 61, 471 –480. (doi:10.1016/S0378-
4320(00)00108-1)
65. Stearns SC. 1992 The evolution of life histories, xii,
249 p. Oxford, UK: Oxford University Press.
66. Kirkwood TBL, Holliday R. 1979 The evolution of
ageing and longevity. Proc. R. Soc. Lond. B 205,
531–546. (doi:10.1098/rspb.1979.0083)
67. Kuzawa CW. 2005 Fetal origins of developmental
plasticity: are fetal cues reliable predictors of future
nutritional environments? Am. J. Hum. Biol. 17,
5–21. (doi:10.1002/ajhb.20091)
68. Clark G. 2013 What is the true rate of social
mobility? Evidence from the information content of
surnames. In American Economic Association
conference paper. http://citeseerx.ist.psu.edu/
viewdoc/summary?doi=10.1.1.269.6016.
69. Gilbert SF. 2000 Developmental biology, xviii, 749
pp, 6th edn. Sunderland, MA: Sinauer Associates.
70. Kronenberg H, Williams RH. 2008 Williams textbook
of endocrinology, xix, 1911 pp, 11th edn.
Philadelphia, PA: Saunders/Elsevier.
71. Najarro K et al. 2015 Telomere length as an
indicator of the robustness of B- and T-Cell response
to influenza in older adults. J. Infect. Dis. 212,
1261–1269. (doi:10.1093/infdis/jiv202)
72. Helby J, Nordestgaard BG, Benfield T, Bojesen SE.
2017 Shorter leukocyte telomere length is
associated with higher risk of infections: a
prospective study of 75 309 individuals from the
general population. Haematologica 102, 1457–
1465. (doi:10.3324/haematol.2016.161943)
73. Cohen S, Janicki-Deverts D, Turner RB, Casselbrant
ML, Li-Korotky HS, Epel ES, Doyle WJ. 2013
Association between telomere length and
experimentally induced upper respiratory viral
infection in healthy adults. JAMA 309, 699 – 705.
(doi:10.1001/jama.2013.613)
74. Eisenberg DTA, Borja JB, Hayes MG, Kuzawa CW.
2017 Early life infection, but not breastfeeding,
predicts adult blood telomere lengths in the
Philippines. Am. J. Hum. Biol. 29, 22962. (doi:10.
1002/ajhb.22962)
75. Haycock PC, Heydon EE, Kaptoge S, Butterworth AS,
Thompson A, Willeit P. 2014 Leucocyte telomere
length and risk of cardiovascular disease: systematic
review and meta-analysis. BMJ 349, g4227. (doi:10.
1136/bmj.g4227)
76. Aviv A, Anderson JJ, Shay JW. 2017 Mutations,
cancer and the telomere length paradox. Trends in
Cancer 3, 253–258. (doi:10.1016/j.trecan.2017.02.
005)
77. Varela E, Munoz-Lorente MA, Tejera AM, Ortega S,
Blasco MA. 2016 Generation of mice with longer
and better preserved telomeres in the absence of
genetic manipulations. Nat. Commun. 7, 11739.
(doi:10.1038/ncomms11739)
78. Aydos SE, Elhan AH, Tu¨ku¨n A. 2005 Is telomere
length one of the determinants of reproductive life
span? Arch. Gynecol. Obstet. 272, 113– 116. (doi:10.
1007/s00404-004-0690-2)
79. Gray KE, Schiff MA, Fitzpatrick AL, Kimura M, Aviv A,
Starr JR. 2014 Leukocyte telomere length and age
at menopause. Epidemiology 25, 139–146. (doi:10.
1097/EDE.0000000000000017)
80. Shenassa ED, Rossen LM. 2015 Telomere length and
age-at-menopause in the US. Maturitas 82, 215–
221. (doi:10.1016/j.maturitas.2015.07.009)
81. Zhou J, Shen X, Huang J, Hodes RJ, Rosenberg SA,
Robbins PF. 2005 Telomere length of transferred
lymphocytes correlates with in vivo persistence and
tumor regression in melanoma patients receiving
cell transfer therapy. J. Immunol. 175, 7046– 7052.
(doi:10.4049/jimmunol.175.10.7046)
82. Ma H et al. 2011 Shortened telomere length is
associated with increased risk of cancer: a meta-
analysis. PLoS ONE 6, e20466. (doi:10.1371/journal.
pone.0020466)
83. Wentzensen IM, Mirabello L, Pfeiffer RM, Savage
SA. 2011 The association of telomere length and
cancer: a meta-analysis. Cancer Epidemiol. Biomark
Prev. 20, 1238– 1250. (doi:10.1158/1055-9965.EPI-
11-0005)
84. Rode L, Nordestgaard BG, Bojesen SE. 2016
Long telomeres and cancer risk among 95 568
individuals from the general population.
Int. J. Epidemiol. 45, 1634– 1643. (doi:10.1093/
ije/dyw179)
85. Zhu X, Han W, Xue W, Zou Y, Xie C, Du J, Jin G.
2016 The association between telomere length and
cancer risk in population studies. Sci. Rep. 6, 22243.
(doi:10.1038/srep22243)
86. Machiela MJ et al. 2016 Genetically predicted longer
telomere length is associated with increased risk of
B-cell lymphoma subtypes. Hum. Mol. Genet. 25,
1663–1676. (doi:10.1093/hmg/ddw027)
87. Ojha J et al. 2016 Genetic variation associated with
longer telomere length increases risk of chronic
lymphocytic leukemia. Cancer Epidemiol. Biomark
Prev. 26, 1043– 1049. (doi:10.1158/1055-9965.epi-
15-1329)
88. Cong Y, Shay JW. 2008 Actions of human
telomerase beyond telomeres. Cell Res. 18,
725–732. (doi:10.1038/cr.2008.74)
89. Geserick C, Blasco MA. 2006 Novel roles for
telomerase in aging. Mech. Ageing Dev. 127,
579–583. (doi:10.1016/j.mad.2006.01.017)
90. Uller T, Nakagawa S, English S. 2013 Weak evidence
for anticipatory parental effects in plants and
animals. J. Evol. Biol. 26, 2161–2170. (doi:10.1111/
jeb.12212)
91. Bateson P, Gluckman P, Hanson M. 2014 The
biology of developmental plasticity and the
Predictive Adaptive Response hypothesis. J. Physiol.
592, 2357–2368. (doi:10.1113/jphysiol.2014.
271460)
92. Burgess SC, Marshall DJ. 2014 Adaptive parental
effects: the importance of estimating environmental
predictability and offspring fitness appropriately.
Oikos 123, 769– 776. (doi:10.1111/oik.01235)
rstb.royalsocietypublishing.org Phil. Trans. R. Soc. B 373: 20160442
10
on January 16, 2018http://rstb.royalsocietypublishing.org/Downloaded from
