Antagonistic pleiotropy and mutation accumulation contribute to age‐related decline in stress response

Abstract

As organisms age, the effectiveness of natural selection weakens, leading to age-related decline in fitness-related traits. The evolution of age-related changes associated with senescence is likely influenced by mutation accumulation (MA) and antagonistic pleiotropy (AP). MA predicts that age-related decline in fitness components is driven by age-specific sets of alleles, nonnegative genetic correlations within trait across age, and an increase in the coefficient of genetic variance. AP predicts that age-related decline in a trait is driven by alleles with positive effects on fitness in young individuals and negative effects in old individuals, and is expected to lead to negative genetic correlations within traits across age. We build on these predictions using an association mapping approach to investigate the change in additive effects of SNPs across age and among traits for multiple stress-response fitness-related traits, including cold stress with and without acclimation and starvation resistance. We found support for both MA and AP theories of aging in the age-related decline in stress tolerance. Our study demonstrates that the evolution of age-related decline in stress tolerance is driven by a combination of alleles that have age-specific additive effects, consistent with MA, as well as nonindependent and antagonistic genetic architectures characteristic of AP.

Dissection of complex, fitness-related traits in multiple Drosophila mapping populations offers insight into the genetic control of stress resistance

“…We similarly compared starvation resistance phenotypes for the DGRP measured in the 420 current study with data generated by Mackay et al (2012) and Everman and Morgan (2018). In 421 our study, the DGRP exhibited considerably higher resistance than in these previous works (F2,532 422 = 1457.5, p < 0.0001; Fig S8).…”

“…Correlations among three DGRP starvation datasets -that from Mackay et al (2012), 305 Everman and Morgan (2018), and the new screen we report here -were examined in a pairwise 306 manner using line means, accounting for multiple comparisons with a Bonferroni-adjusted alpha 307 level. Differences in the overall mean starvation resistance among the three datasets were 308 analyzed with a one-way ANOVA, treating study as a fixed factor.…”

“…Mean female triglyceride level was similar between the pA1 and pA2 332 subpopulations (Tukey's HSD p = 0.75; Fig S3; Table S3) but varied between the pB1 and pB2 333 subpopulations (Tukey's HSD p < 0.0001; Fig S3; Huang et al 2014). We performed GWA analysis on data collected 341 in this study and additionally reanalyzed starvation data from Mackay et al (2012) and starvation 342 of young flies (5 -7 days old) from Everman and Morgan (2018). We additionally assigned the 343 150 DGRP lines that are shared between the three datasets an across-study mean and performed 344 GWA analysis on this summary measure of starvation resistance.…”

“…Since starvation resistance is sensitive to the thermal environment (van 409 Herrewege and David 1997;Karan and Parkash 1998;Hoffmann et al 2005a;410 Bauerfeind et al 2014) and may vary under different photoperiods (Sheeba et al 2000;Xu et al 411 2008; Seay and Thummel 2011), we sought to re-measure starvation resistance for a subset of 412 DSPR RILs at 25°C and with a 12 hour : 12 hour light/dark cycle, conditions that have been used 413 in other starvation studies (e.g. Mackay et al 2012;Everman and Morgan 2018). Overall, 414 starvation resistance of the re-tested RILs was lower in both sexes compared to that measured 415 in the original large-scale starvation assay (effect of assay: F1,136 = 31.60, p < 0.0001; Fig S7A).…”

“…702 703 Different mapping approaches reveal unique genetic architectures for starvation 704 resistance 705 Dissection of a quantitative genetic trait using different approaches can allow greater 706 resolution of the genetic architecture, and provide insight into how alleles unique to different 707 mapping panels contribute to phenotypic variation. To gain this additional understanding we 708 assessed the genetic architecture of starvation resistance in the DGRP using GWA mapping of 709 four starvation resistance datasets: new data collected in this study, data from Mackay et al 710 (2012), data from Everman and Morgan (2018), and a consensus, across-study starvation 711 resistance measure calculated as the mean response across the three starvation datasets (150 712 lines were measured across the three studies). Using an FDR threshold of 20%, between 0 and 12 713 SNPs were associated with starvation resistance in each dataset and sex (Table 3; Table S11).…”