Heritability of telomere length in the Zebra Finch

Abstract

Telomere length predicts survival in birds, and many stressors that presumably reduce fitness have also been linked to telomere length. The response to selection of telomere length will be largely determined by the heritability of this trait; however, little is known about the genetic component of telomere length variation in animals other than humans. Moreover, published heritability estimates of telomere length are based on telomere measurements with techniques that do not distinguish between terminal telomeres, which are susceptible to age and stress, and the interstitial telomeric repeats, which are relatively inert. Heritability estimates that combine interstitial and terminal telomeres are difficult to interpret in species such as birds, where interstitial telomeres are often numerous. We estimated the heritability of terminal telomere length in a captive Zebra Finch population of cross-fostered (half-)siblings using data obtained with an electrophoresis technique that excludes the interstitial repeats from the measurements. We used both a Bayesian quantitative genetic 'animal' model and a frequentist sibling regression approach to estimate heritability. With the animal model, we estimated a high heritability of telomere length (h 2 = 0.99, 95 % credible interval = 0.87-1), but had insufficient statistical power to separate parental and permanent environment effects. The frequentist approach yielded similar heritability estimates, although with large confidence intervals. We used general linear mixed models to disentangle variance components of telomere length. The relative contributions of the individual, mother and father to telomere length variation were statistically indistinguishable at 23-31 %. Chicks were cross-fostered 4-days after hatching, and no effect of rearing nest was found, indicating that any undetected environmental effects exerted their influence prior to, or soon after, hatching. Thus, we conclude that telomere length resemblance between relatives is high and proportional to their relatedness, but we cannot conclusively distinguish between genetic and other forms of inheritance.

Evolutionary ecology of telomeres: a review

“…Although it was an impressive effort, few studies of natural populations find significant heritabilities for almost any trait with such low sample sizes. In stark contrast to these zero heritability estimates, and with similar sample sizes in a laboratory population (66 females, 57 males, and 73 broods), Atema et al .’s 0.99 heritability estimate using TRF was indistinguishable from one (1.0). In another bird species, (king penguins, Aptenodytes patagonicus ), maternal heritability of telomere length (measures using qPCR) was significant between mothers and chicks early in life, but was no longer significant as offspring became older .…”

The environmental and genetic determinants of chick telomere length in Tree Swallows (Tachycineta bicolor)

“…The strong role for genetic effects on chick TL is indicated by both the high estimated h 2 value (0.81) and the fact that both maternal and paternal TL were among the most important predictors of chick TL in the full mixed model (Tables and ). Previous studies offer a wide range of estimates for TL heritability, with the measures that we obtained falling at the high end of this range (Atema et al, ; Dugdale Hannah & Richardson David, ). While these high heritability estimates may in part be due to maternal effects that are expressed prior to hatching, such as egg composition or incubation, the weak effect of the brood enlargement combined with the strong effect of parental TL in the main model suggest that, at least at the age of 12 days old, Tree Swallow TL is determined more by heritable genetic factors than by environmental ones.…”

Mother–offspring and nest-mate resemblance but no heritability in early-life telomere length in white-throated dippers

“…At first sight in contrast to Voillemot et al [19], another study employing crossfostering arrived at a heritability of 0.99 [20]. However, as pointed out by the authors, this estimate might also be upwardly biased due to difficulties in separating genetic and non-genetic sources of variation in their breeding design [20]. A second source of variation might be provided by differences among the methods that have been employed to quantify telomere length (reviewed in [42]), and in particular between the qPCR method employed by most studies so far and which measures the total amount of telomeric sequence, and electrophoresis-based methods that quantify the amount of terminal telomeric sequence only [20].…”