RESULTS
The telomere length of the five wild zebra finches measured using the
in-gel TRF method was similar to that of captive bred birds (range
12.2-13.59kb, mean 12.77kb+/-0.24. Overall, there was no change in
telomere length between day 3 and day 11 of post-hatch development,
however on average Mean telomere length was 0.84 (range: 0.29-2.59, SEM:
0.059) on day 3, and 0.74 (range: 0.44-2.16, SEM: 0.041) on day 11. This
decrease in telomere length was not statistically significant (paired
t-test: t 49 = 1.59, p = 0.116) (Figure 2).
Thirty-three individuals showed a reduction in telomere length between
day 3 and day 11 post-hatch, while 17 were found to have telomeres that
increased in length. Telomere length decreased by an average of 0.28
(range: -0.64 to -0.02, SEM: 0.03), and increased by 0.25 (range:
<0.01 to 1.92, SEM: 0.16).
Our MS-AFLP analyses verified 92 consistent CpG sites between 50 and 500
base pairs. (%) DNAm was 29.2% at day 3 (Standard deviation (SD) =
6.5), and 34.9% at day 11 (SD=8.3). We found a significant, negative
relationship between changes in the level (%) of DNAm across these 92
CpG sites and change in telomere length between day 3 and day 11 (Table
1, Figure 3a). We also found a significant, negative relationship
between change in tarsus length and change in telomere length between
day 3 and day 11 (Table 1, Figure 3b). Tarsus length at day 3, brood
size and post-hatch ambient temperature did not significantly affect
change in telomere length (Table 1). The fixed factors explained 43.73%
(Marginal R2=0.437), while the random effect of natal
nest ID explained <1% of the variance in the data
(Conditional R2=0.441; this value describes the
proportion of variance explained by both the fixed and the random
factor, thus most variance in this model was explained by the fixed
factors) (Table 1)).
In our second LMM we detected a significant, positive association
between clutch size and telomere length at day 3 (i.e., larger clutches
were associated with longer telomeres at day 3 (Table 2)), while percent
DNAm at day 3 and other fixed factors (pre-hatch ambient temperature and
tarsus length at day 3) were not related to telomere length. In this
model, the fixed factors explained 28.57% of the variance in telomere
length at day 3 (Marginal R2=0.285), while the random
effect of natal nest ID explained 40.19% of variance in telomere length
at day 3 (Conditional R2=0.687) (Table 2).
In our third LMM, DNAm at day 11 was not related to telomere length at
day 11, and, similarly none of the other fixed factors (brood size,
temperature and tarsus length at day 11) included in our model appeared
to influence telomere length at day 11 (Table 3). The fixed factors in
our model (Table 3), explained 8.63% of the variance in the data
(Marginal R2=0.086), while the random effect of natal
nest ID explained 59.43% of variance in telomere length at day 11
post-hatch (Conditional R2=0.6807) (Table 3). After
visual inspection of figures 2, 3a and 3b, we identified an individual
with a particularly high telomere length value. We re-ran all our models
after excluding this anomalous data point, however, excluding this data
point did not alter our results significantly. Since there was no
biological reason to omit this individual from the data set it was
included in our final analyses.