The role of anorexia in resistance and tolerance to infections in Drosophila

Abstract

Most infections induce anorexia but its function, if any, remains unclear. Because this response is common among animals, we hypothesized that infection-induced diet restriction might be an adaptive trait that modulates the host's ability to fight infection. Two defense strategies protect hosts against infections: resistance, which is the ability to control pathogen levels, and tolerance, which helps the host endure infection-induced pathology. Here we show that infected fruit flies become anorexic and that diet restriction alters defenses, increasing the fly's tolerance to Salmonella typhimurium infections while decreasing resistance to Listeria monocytogenes. This suggests that attempts to extend lifespan through diet restriction or the manipulation of pathways mimicking this process will have complicated effects on a host's ability to fight infections.

Figure 1. Effect of infection on appetite.

Flies were infected with live or heat killed L. monocytogenes, live S. typhimurium, live E. faecalis, medium as a control, or left unmanipulated. Feeding was monitored by measuring the rate that flies took a meal (A–C) and the volume that they consumed during this meal (D–F). Feeding rate measurements: (A) L. monocytogenes 24 h postinfection; (B) S. typhimurium 24 h postinfection; (C) E. faecalis 24 h postinfection. To measure the volume of food consumed, fed flies were homogenized and the absorbance of an added blue dye was measured 24 h postinfection. (D) L. monocytogenes, (E) S. typhimurium, (F) E. faecalis. Error bars indicate standard error of the mean. Significance for (A–C) was assessed using a Fisher's exact test. Green asterisks represent live bacteria significantly different compared to both unmanipulated and media-injected flies. Blue asterisks represent dead bacteria (L. monocytogenes only) significantly different than both unmanipulated and media-injected flies. Black cross represents live bacteria significantly different from unmanipulated flies only. Pink cross represents live bacteria significantly different from media-injected flies only. Actual p-values are listed in Tables S1, S2, S3. Statistical analysis for (D–F) was done using ANOVA and a Tukey post-test; black asterisk indicates p<0.01 with respect to both unmanipulated and media-injected flies, black cross indicates p<0.05 with respect to unmanipulated flies only, and pink cross indicates p<0.05 with respect to media-injected flies only.

Figure 2. Effect of mutation of the gustatory receptor gr28b on appetite.

Isogenic wild-type and gr28b mutants were assayed for feeding rates (A) and meal volumes (B). Error bars indicate standard error of the mean. Statistical analysis for (A) was done using a Fisher's exact test. Green asterisk indicates Listeria-infected gr28b mutants are significantly different than unmanipulated wild-type flies. Blue asterisks indicate that unmanipulated gr28b mutants are significantly different than unmanipulated wild-type flies. Green crosses indicate that Listeria-infected mutants are significantly different from unmanipulated wild-type flies only. Actual p-values are listed in Table S4. Statistical analysis for (B) was done using ANOVA and a Tukey post-test. Black asterisk indicates p<0.001 compared to unmanipulated wild-type flies.

Figure 3. Effect of mutation of the gustatory receptor gr28b and diet restriction on sensitivity to infections.

Isogenic wild-type and gr28b homozygous mutant flies were challenged with (A) L. monocytogenes (p<0.0001); (B) S. typhimurium (p<0.0001); (C) E. faecalis (p = 0.2779); or (D) medium alone (p<0.0001); and survival rates were measured and compared between flies given the two treatments. Wild-type flies fed on 1× and 0.5× diets, and were challenged with (E) L. monocytogenes (p<0.0001); (F) S. typhimurium (p<0.0001); (G) E. faecalis (p = 0.6053); or (H) medium alone (p<0.0001) and survival rates were measured. Significance was determined by log-rank test. Effects of gr28b mutations and diet restriction in unmanipulated flies on lifespan are shown in Figure S1.

Figure 4. Effect of gr28b mutation and diet restriction on the growth of L. monocytogenes and S. typhimurium.

Isogenic wild-type and gr28b homozygous mutant flies or wild-type flies fed on 1× and 0.5× diets were challenged with pathogens. Because L. monocytogenes infections showed a change in pathogen levels, half of the infected flies were injected with gentamicin to determine the relative abundance of intracellular and extracellular bacteria. Wild-type versus gr28b mutants: (A) L. monocytogenes; (B) S. typhimurium; and (C) E. faecalis. Regular food versus diet restriction: (D) L. monocytogenes; (E) S. typhimurium; (F) E. faecalis. Error bars indicate standard deviation. Statistical analysis was done using an unpaired two-tailed t-test. One asterisk indicates p<0.01 and two asterisks indicates p<0.005.

Figure 5. Effects of diet restriction on immunity when introduced post-eclosure.

Five to 7-d-old adult males flies were collected and placed on a restricted diet or left at a 1× diet 24 h prior to infection. Survival of (A) L. monocytogenes (0.25× compared to 1×, p = 0.0154); (B) S. typhimurium (0.5× compared to 1×, p<0.0001); (C) growth of L. monocytogenes. Because L. monocytogenes infections showed a change in pathogen levels, half of the infected flies were injected with gentamicin to determine the relative abundance of intracellular and extracellular bacteria. Asterisk indicates p = 0.0173 as determined by an unpaired two-tailed t-test.

Figure 6. Effect of anorexia and diet restriction on antimicrobial peptide expression and melanization.

gr28b mutants and diet restricted flies were injected with L. monocytogenes, and antimicrobial peptide transcript levels were monitored at 6 h postinfection by quantitative real-time reverse-transcription PCR. Transcript levels were recorded as the ratio of the antimicrobial peptide transcript divided by a housekeeping transcript (ribosomal protein 15a) and normalized to 1 for unmanipulated wild-type flies. (A) drosomycin; (B) drosocin; (C) attacin. Error bars report standard error of the mean. ANOVA and Tukey tests were performed for statistical analysis and asterisks indicate p<0.05. Melanized spots were recorded in (D) L. monocytogenes and (E) S. typhimurium infections. ANOVA and Tukey test were done for statistical analysis. Asterisks represent p<0.001.