To further establish a role for Gr63a-expressing neurons in the modulation of lifespan, we expressed the proapoptotic gene reaper using GR63a-GAL4 to ablate the associated neurons. (A) Gr63a mRNA is reduced in the heads of female flies with both Gr63a-GAL4 and UAS-reaper compared to w¹¹¹⁸ control (p = 0.01). Expression was determined using at least three independent RNA extractions of 40 fly heads each. (B) Gr63a>reaper flies are significantly less repulsed by CO₂ compared to control flies, UAS-reaper only, and Gr63a-GAL4 only (p = 0.002, 0.0007, and 0.003 respectively). Data were obtained using a T-maze assay and a stream of ∼4% CO₂. Forty flies were used for each measure, and a minimum of eight measures were made for each genotype. (C) Female Gr63a>reaper flies (green solid line) are longer lived than corresponding controls, including w¹¹¹⁸ wild type (red dashed line), UAS-reaper only (dark gray dashed line), and Gr63a-Gal4 only (orange dashed line) control (p<1×10⁻¹⁵ for comparing Gr63a>reaper to any control line). There are no significant differences among the survivorship curves from the three control lines (p>0.05). Samples sizes are w1118, 292; GAL4/UAS, 286; UAS only, 282; GAL4 only, 286. Error bars represent standard errors of the mean.

(A and B) Survivorship of female flies homozygous for Gr63a¹ loss-of-function (null) allele (green solid lines) and their respective background control (red dashed lines) (A) w¹¹¹⁸ background (n = 249 and 254 for mutant and control flies, respectively), (B) yw background (n = 219 and 216 for mutant and control flies, respectively); p<1×10⁻¹⁵ in both cases. (C) Survivorship of flies from a transgenic rescue experiment. Expression of Gr63a cDNA under the control of its endogenous promoter region using the UAS/GAL4 system restored the natural lifespan of Gr63a¹ mutant flies (compare green solid lines to blue dot-dash line and red dashed line; p = 0.15), whereas overexpression of Gr63a had modest negative effect on lifespan (yellow dot dashed line vs. red dashed lines; p = 0.005). Samples sizes are as follows: control, 230; Gr63a^−/−, 228; overexpression, 230; rescue, 240. (D) Quantitative real-time PCR using RNA extracted from fly heads revealed that Gr63a¹ mutant flies lacked measureable Gr63a mRNA (no amplification was observed). Furthermore, transgenic rescue restored Gr63a mRNA to levels that were indistinguishable from w¹¹¹⁸ control (p = 0.59), and overexpression resulted in a roughly 2-fold increase in Gr63a transcript (but this was not significant; p = 0.32). Forty fly heads were used per RNA extraction, and expression measures were obtained from multiple quantitative PCR reactions each from at least three independent replicate RNA extractions. Error bars represent standard errors of the mean.

(A) Longevity of Gr63a¹ mutant flies was further extended by dietary restriction. Gr63a mutants live significantly longer when fed a low-nutrient diet (5% sugar-yeast diet, green solid line) than when fed a high-nutrient diet (15% sugar-yeast diet; red solid line, p<1×10⁻¹⁵). A similar diet-dependent extension of lifespan was observed in control animals (red vs. green dashed lines; p = 1×10⁻¹⁰). Sample sizes are as follows: fully fed: control, 269; mutant, 270; diet restriction: control, 251; mutant, 235. (B) Of a range of food-based odorants tested, such as mango scent and apple cider vinegar, only odors from live yeast consistently affected lifespan of w¹¹¹⁸ female flies (see Table S2 for additional details). Samples sizes are water, 258; mango, 262; vinegar, 260; and yeast, 244. (C) Yeast odors failed to impact lifespan of Gr63a¹ mutants. In multiple experiments involving control and mutant flies in two independent genetic backgrounds (w¹¹¹⁸ and yw), Gr63a mutants did not exhibit reduced lifespan when exposed to odor from live yeast. In all four experiments, there was a tendency for yeast odor to increase Gr63a¹ mutant lifespan, but the effect was small and statistically significant in only one instance. Killing of the yeast through autoclaving removes the lifespan-shortening effects in wild-type flies. (**p<0.001, *p<0.01; see also Figure S6B). (D) Flies that are homozygous for a null mutation in Or83b have greatly reduced general olfactory capabilities but are able to sense CO₂ normally. The lifespan of Or83b animals was also reduced by live yeast odor (p = 1×10⁻⁸). Samples sizes are: control, 274; yeast-odor supplemented, 279. All odorant-supplemented longevity experiments were carried out using 5% (low nutrient) sugar-yeast medium. Error bars represent standard errors of the mean.

(A) Female Gr63a¹ mutants exhibited higher triglyceride levels than control flies. This was observed in two independent genetic backgrounds, w¹¹¹⁸ and yw. Transgenic rescue of Gr63a restored mutant fat levels to that seen in wild-type flies. Overexpression of Gr63a had no effect on triglyceride levels. For each triglyceride measure, six independent measures of five females each were obtained and measured for triglyceride content. Data were normalized to total protein levels. (B) Gr63a¹ mutant flies (solid green lines) had normal starvation resistance compared to controls (dashed red lines) when placed on 1% agar, despite higher triglyceride levels shown in (A) (p = 0.10). (C) Over the course of 10 d, Gr63a mutants (green) laid roughly 40% more eggs than controls (red). Fifteen 2-d-old virgin females were each mated with two control males and subsequently maintained on 15% yeast-sucrose diet. Flies were moved to new vials daily, and eggs were counted each day for 10 d. The plot is representative of three independent experiments (p<0.0001 for final cumulative egg count). (D) Gr63a mutant flies (solid green line) exhibited higher resistance to 10 mM paraquat feeding than controls (dashed red lines; p = 1×10⁻⁶). Paraquat (10 mM) in a 5% sucrose solution was fed to the flies. (E) Gr63a¹ mutants exhibited normal daily activity patterns (left), overall activity, and overall sleep (right). Activity data are representative of at least three independent experiments using Trikinetics activity monitoring hardware. Each trace is the composite average of 5 d of recording and 16+ animals per group. General trends were confirmed using video-based analysis. Error bars represent standard errors of the mean. ZT, zeitgeber.