Target of Rapamycin Drives Unequal Responses to Essential Amino Acid Depletion for Egg Laying in Drosophila Melanogaster

Abstract

Nutrition shapes a broad range of life-history traits, ultimately impacting animal fitness. A key fitness-related trait, female fecundity is well known to change as a function of diet. In particular, the availability of dietary protein is one of the main drivers of egg production, and in the absence of essential amino acids egg laying declines. However, it is unclear whether all essential amino acids have the same impact on phenotypes like fecundity. Using a holidic diet, we fed adult female Drosophila melanogaster diets that contained all necessary nutrients except one of the 10 essential amino acids and assessed the effects on egg production. For most essential amino acids, depleting a single amino acid induced as rapid a decline in egg production as when there were no amino acids in the diet. However, when either methionine or histidine were excluded from the diet, egg production declined more slowly. Next, we tested whether GCN2 and TOR mediated this difference in response across amino acids. While mutations in GCN2 did not eliminate the differences in the rates of decline in egg laying among amino acid drop-out diets, we found that inhibiting TOR signalling caused egg laying to decline rapidly for all drop-out diets. TOR signalling does this by regulating the yolk-forming stages of egg chamber development. Our results suggest that amino acids differ in their ability to induce signalling via the TOR pathway. This is important because if phenotypes differ in sensitivity to individual amino acids, this generates the potential for mismatches between the output of a pathway and the animal's true nutritional status.

Keywords: GCN2; amino acid sensing pathways; egg chamber development; egg laying dynamics; holidic diet; nutrient signalling; target of rapamycin.

FIGURE 1

GCN2 and TOR signalling in the adipose tissue and ovary transmit amino acid availability by affecting distinct processes of egg chamber development. Early stages include stages 1–7, yolk-forming includes stages 8–11, and late includes stages 12–14 of egg chamber development. The question mark indicates that it is unclear through which cell type TOR regulates the yolk-forming stages. TFCs: terminal filament cells, GSC: germline stem cells, FSC: follicle stem cells.

FIGURE 2

Amino acids differ in their effects on egg laying. (A) While in most cases removing individual essential amino acids results in a rapid decline in egg laying at a rate similar to removing all amino acids from the diet, removing valine, histidine, or methionine results in a slower rate of decline. (B) For most diets in which a single amino acid is removed, the total number of eggs that females produce over a 9-day period is similar to females on a diet that contains no amino acids. Females on either histidine, valine, or methionine drop-out diets produce more eggs than those on diets missing any of the other essential amino acids. Diets that have differing letters are significantly different, as determined from post hoc pairwise comparisons derived from a generalised linear model assuming a Poisson distribution.

FIGURE 3

Inhibiting either the GCN2 or Target of Rapamycin pathways alters the impact of diet on egg laying. (A) Changes in the rate of decline in egg laying in response to amino acid availability in the diet between control (wDah, solid circles and lines) and GCN2∆ flies (open circles and dashed lines). (B) Mean number of eggs laid per female in response to amino acid availability between control (wDah, closed circles) and GCN2∆ flies (open circles). (C) Changes in the rate of decline in egg production in response to amino acid availability in the diet between ethanol-treated Red Dahomey flies (solid circles and lines) and rapamycin-treated Red Dahomey flies (open circles and dashed lines). (D) Mean number of eggs laid per female in response to amino acid availability between ethanol-treated Red Dahomey flies (solid circles) and rapamycin-treated flies (open circles). In (B,D), diets that have differing letters are significantly different within a genotype or treatment, as determined from post hoc pairwise comparisons derived from a generalised linear model assuming a Poisson distribution (Supplementary Table S2). Capital letters indicate significance groups across diets for the control flies. Lower cases letters indicate significance groups for the GNC2 mutant (B) or rapamycin-treated (D) flies across diets.

FIGURE 4

Predicting how diets that induce a slow versus fast decline impact egg laying via the TOR pathway. (A) Egg chamber development can be broadly divided into three stages: 1) early stages (stages 1–7), yolk-forming stages (stages 8–10), and late stages (stages 11–14). TOR signalling promotes germline stem cell maintenance, egg chamber growth, and the transition into yolk-forming stages (LaFever et al., 2010). Both TOR and GCN2 regulate oviposition; TOR signalling promotes oviposition while GCN2 signalling inhibits TOR to suppress oviposition (Armstrong et al., 2014). (B) GCN2 mutant females (No GCN2—All AA) should have normal stem cell numbers and should not show repression of oviposition. Relative to wild type females on diets containing all amino acids (All AA), this might result in faster rates of oviposition, reducing the percentage of late stage egg chambers. We would expect ovaries from females on amino acid drop out diets (Slow and Fast Decline) to have reduced percentages of egg chambers in the yolk-forming and late stages relative to ovaries from females on All AA. If this difference is due to differences in the sensitivity of the TOR pathway across amino acids, then we would predict that on the slow decline diets we would see a more gradual decline in the percentage of yolk-forming egg chambers than on the diets that induce fast decline in egg laying. Treatment with rapamycin (low TOR) should reduce or eliminate the differences between slow and fast decline diets.

FIGURE 5

Egg chamber development changes with diet and with reductions in GCN2 and TOR signalling. Ovaries from white Dahomey wDah, (A–C), GNC2 mutant (D–F), red Dahomey fed diets laced with ethanol Ethanol, (G–I), and red Dahomey fed diets laced with rapamycin Rapamycin, (J–L) females. Females were fed either a diet consisting of all amino acids (All AA), (A,D,G,J), diets that induced a slow decline in egg laying (no histidine), (B,E,H,K), or diets that induced a fast decline in egg laying (no arginine), (C,F,I,L). All samples are stained with DAPI to visualise nuclei. In (A), the light teal arrows point to early, the teal arrows point to yolk-forming, and the dark teal arrows point to late stage egg chambers. All pictures were taken after 7 days of diet exposure.

FIGURE 6

The percentage of egg chambers in each category is affected by diet, and these effects are partially mediated by GCN2 signalling. Comparison between the percentage of egg chambers between white Dahomey (wDah) and GCN2∆ flies.

FIGURE 7

The percentage of egg chambers in each category differs among amino acid drop-out diets, and these effects are eliminated when Target of Rapamycin signalling is inhibited using the drug rapamycin. Comparison between the percentage of egg chambers between red Dahomey females treated with ethanol and treated with rapamycin (30 µM).