Rapamycin supplementation of Drosophila melanogaster larvae results in less viable adults with smaller cells

Abstract

The intrinsic sources of mortality relate to the ability to meet the metabolic demands of tissue maintenance and repair, ultimately shaping ageing patterns. Anti-ageing mechanisms compete for resources with other functions, including those involved in maintaining functional plasma membranes. Consequently, organisms with smaller cells and more plasma membranes should devote more resources to membrane maintenance, leading to accelerated intrinsic mortality and ageing. To investigate this unexplored trade-off, we reared Drosophila melanogaster larvae on food with or without rapamycin (a TOR pathway inhibitor) to produce small- and large-celled adult flies, respectively, and measured their mortality rates. Males showed higher mortality than females. As expected, small-celled flies (rapamycin) showed higher mortality than their large-celled counterparts (control), but only in early adulthood. Contrary to predictions, the median lifespan was similar between the groups. Rapamycin administered to adults prolongs life; thus, the known direct physiological effects of rapamycin cannot explain our results. Instead, we invoke indirect effects of rapamycin, manifested as reduced cell size, as a driver of increased early mortality. We conclude that cell size differences between organisms and the associated burdens of plasma membrane maintenance costs may be important but overlooked factors influencing mortality patterns in nature.

Keywords: TOR; ageing; cell size; life history; mortality; survivorship.

Figure 1.

Adult phenotypes of Drosophila melanogaster after larval feeding diets with and without rapamycin. (a) Female and male rapamycin-treated flies were characterized by a smaller body size and smaller cells in all studied organs compared to control flies. The effect was significant at p ≤ 0.0002 for thorax length and the size of cells in muscles, wings, ommatidia and Malpighian tubules but only at p = 0.168 for the size of leg cells. (b) Independent of diet, females were characterized by a larger body size and larger cells in all studied organs compared with males (p < 0.0001). Arrows show mean values obtained from Szlachcic et al. [43], estimated with statistical methods from measurements of individuals derived from the same pool of flies as studied here but used to explore orchestration of cell size throughout the body via TOR activity (for detailed statistical results, see table 1 in [43]). Thorax length (mm) was measured as a proxy for body size. Cell size was measured in five (males) or four (females) organs as follows: dorsal longitudinal indirect flight muscle cells in the thorax (µm²) by the mean cross-sectional area of fibres; epidermal cells in the wing (µm²) by the number of trichomes per unit area; ommatidial cells (µm²) by the mean area of ommatidia in the eye; epidermal cells in the leg (µm²) by the number of trichomes per unit length; and Malpighian tubule epithelial cells (µm²; measured only in males) by the number of nuclei/nucleoli per unit area.

Figure 2.

Study design used to raise adult Drosophila melanogaster after larval feeding diets with and without rapamycin. Flies from each of 14 isolines were raised on diets with or without rapamycin (rapamycin versus control flies), using procedures shown here for one of the studied isolines. Using stock isolines, we first produced two generations of flies with controlled mating and larval density (generations 1 and 2), increasing the number of vials with flies for each isoline. The second-generation larvae were raised on either standard food or food enriched with rapamycin. Females and males emerging from the two developmental treatments were then maintained in same-sex groups until death on standard food without rapamycin to compare mortality between developmental treatments. We used a mixed-vial collection approach when collecting flies, sampling flies from as many vials as possible for each isoline. Another set of flies originating from the same experiment was used by Szlachcic et al. [43] to characterize body size and cell size of the studied flies (figure 1).

Figure 3.

Survivorship curves of adult Drosophila melanogaster flies estimated by the KM method. Thin lines indicate survivorship of each isoline, and thick lines indicate survivorship of different combinations of sex and treatment. Flies were raised by larval feeding on diets with or without rapamycin (rapamycin versus control flies), but in adulthood, all flies were fed standard food. Initially, each isoline was represented in each treatment by approximately 90 adult males and 90 adult females. The phenotypic characteristics of flies are shown in figure 1. Panels show survivorship of (a) control and rapamycin females; (b) control and rapamycin males; (c) females and males in the control; (d) females and males in the rapamycin treatment. Small cell size phenotypes (rapamycin-treated flies) are associated with lower survivorship at a young age than large cell size phenotypes (control flies). Females have higher survivorship than males.

Figure 4.

Mortality rates of adult Drosophila melanogaster flies estimated by GAMMs. Thin lines indicate the predicted mortality rates of each isoline, and thick lines indicate the marginal predicted mortality rates. Flies were raised by larval feeding on diets with or without rapamycin (rapamycin versus control flies), but in adulthood, all flies were fed standard food. Initially, each isoline was represented in each treatment by approximately 90 adult males and 90 adult females. The phenotypic characteristics of flies are shown in figure 1. Panels show the mortality rate of (a) control and rapamycin females; (b) control and rapamycin males; (c) females and males in the control; (d) females and males in the rapamycin treatment. All predictions are plotted within observable age ranges. The mortality rate associated with the small cell size phenotype (rapamycin-treated flies) was higher at the beginning of adult life than that of the large cell size phenotype (control flies). Males have a higher mortality rate than females.

Figure 5.

Differences in predicted log marginal mortality rates of adult Drosophila melanogaster flies calculated from GAMM estimates. Differences in log marginal mortality rates (which are affected by compositional changes of isolines due to heterogeneity in frailty among isolines; see supplementary material for details) between (a) control and rapamycin females; (b) control and rapamycin males; (c) females and males in the control; (d) females and males in the rapamycin treatment. The solid line shows differences; dashed lines show 95% confidence intervals; blue and red lines show age ranges for which the differences become significant (at the level of 0.05), negative and positive, respectively. All log marginal differences in mortality rates are plotted within observable age ranges. Flies were raised by larval feeding on diets with or without rapamycin (rapamycin versus control flies), but in adulthood, all flies were fed standard food. Initially, each isoline was represented in each treatment by approximately 90 adult males and 90 adult females. The phenotypic characteristics of flies are shown in figure 1.

Figure 6.

Differences in predicted log mortality rates of adult Drosophila melanogaster flies calculated from GAMM estimates. Differences in conditional log mortality rates (after excluding effects of heterogeneity in frailty of isolines from model predictions) between (a) control and rapamycin females; (b) control and rapamycin males; (c) females and males in the control; (d) females and males in the rapamycin treatment. Solid lines show the predicted differences; dashed lines show 95% confidence intervals; blue lines show age ranges for which the differences become significant (at the level of 0.05). Flies were raised by larval feeding on diets with or without rapamycin (rapamycin versus control flies), but in adulthood, all flies were fed standard food. Initially, each isoline was represented in each treatment by approximately 90 adult males and 90 adult females. The phenotypic characteristics of flies are shown in figure 1.