Microbial metabolism disrupts cytokine activity to impact host immune response

Abstract

Host-pathogen interactions are shaped by the metabolic status of both the host and pathogen. The host must regulate metabolism to fuel the immune response, while the pathogen must extract metabolic resources from the host to enable its own survival. In this study, we focus on the metabolic interactions of Mycobacterium abscessus with Drosophila melanogaster. We identify MAB_1132c as an asparagine transporter required for pathogenicity in M. abscessus. We show that this requirement is specifically associated with damage to the host: flies infected with MAB_1132c knockout bacteria, or with wild-type bacteria grown in asparagine-restricted conditions, are longer lived without showing a significant change in bacterial load. This is associated with a reduction in the host innate immune response, demonstrated by the decreased transcription of antimicrobial peptides as well as a significant reduction in the ability of the infection to disrupt systemic insulin signaling. Much of the increase in host survival during infection with asparagine-limited M. abscessus can be attributed to alterations in unpaired cytokine signaling. This demonstrates that asparagine transport in M. abscessus prior to infection is not required for replicative fitness in vivo but does significantly influence the interaction with the host immune responses.

Keywords: Drosophila melanogaster; Mycobacterium abscessus; infection; innate immunity; metabolism.

Fig. 1.

MAB_1132c is important but not essential for growth with asparagine as the sole nitrogen source. Growth of M. abscessus ATCC 19977 (red) and ∆MAB_1132c strains (blue) in various nutrient conditions, measured by optical density at 592 nm. Data represent the mean ± SD of quadruplicate samples and are representative of three independent experiments. Bacteria grown in (A) complete 7H9 media, or minimal media containing 5 mM of (B) aspartate (Asp), (C) asparagine (Asn), (D) glutamate (Glu), or (E) glutamine (Gln) as the sole nitrogen source.

Fig. 2.

Asparagine availability in M. abscessus impacts host survival but not bacterial in vivo growth. (A) Survival and (B) bacterial loads of w¹¹¹⁸ flies infected with M. abscessus ATCC 19977 or ∆MAB_1132c strains grown in complete medium (7H9). (C) Survival and (D) bacterial loads of w¹¹¹⁸ flies infected with M. abscessus ATCC 19977 or ∆MAB_1132c strains grown in Sauton’s minimal medium (SMM) with 5 mM asparagine as the sole nitrogen source. (E) Survival of w¹¹¹⁸ flies infected with M. abscessus ATCC 19977 or ∆MAB_1132c strains grown in Sauton’s minimal media (SMM) with 40 mM asparagine as the sole nitrogen source. (F, G) Expression of transcriptional regulator whiB7 (MAB_3508c) normalised against housekeeping gene rpoB (MAB_3869c) in M. abscessus ATCC 19977 and ∆MAB_1132c strains following growth to log phase in (F) complete medium (7H9) or (G) SMM with 5 mM asparagine. Survival data are pooled from three independent experiments performed with common controls, 123 ≥ n ≥ 68 per group. Asterisks in the legend indicate the P-value compared to flies infected with ATCC 19977, compared using the log-rank test. Bacterial load data are pooled from three independent experiments, n = 8 per strain per condition per independent experiment. Statistical significance was calculated by comparing the mutant strains to ATCC 19977 within the same time point and media using Kruskal–Wallis ANOVAs. Bacterial gene expression data are representative of two independent experiments (n = 8 per strain), and were compared using a Kruskal-Wallis ANOVA. ^∗P < 0.05; ^∗∗P < 0.01; ^∗∗∗P < 0.001; ^∗∗∗∗P < 0.0001.

Fig. 3.

Reduction in antimicrobial peptide transcription by ∆MAB_1132c M. abscessus. RT-qPCRs show transcription of antimicrobial peptides in flies infected with M. abscessus ATCC 19977 (red) or ∆MAB_1132c strains (blue) over the course of an infection. Uninjected (black) and PBS Tween80 0.1% sterile-injury (gray) flies are included as controls. Antimicrobial peptide transcript levels shown are (A) Drosomycin, (B) Metchnikowin, (C) Attacin A, (D) Diptericin, and (E) Drosocin. Stress protein (F) Turandot A is also shown. Data are representative of three independent experiments (N = 12), normalized against RpL4, and were compared using a Kruskal–Wallis ANOVA. Statistical comparisons between M. abscessus ATCC 19977 and the ∆MAB_1132c strains are shown, all statistically significant comparisons are available in SI Appendix, Table S7. Survival curves of (G) w¹¹¹⁸ and (H) imd¹⁰¹⁹¹; spz^∆8−1 flies infected with M. abscessus ATCC 19977 (red) or ∆MAB_1132c-1 (blue), grown in complete 7H9 media. The control group is PBS Tween80 0.1% injected flies (gray). Data represent a single experiment, with at least 40 flies per group. Asterisks in the legend indicate the P-value compared to flies infected with ATCC 19977, compared using the log-rank test. ^∗P < 0.05; ^∗∗P < 0.01; ^∗∗∗P < 0.001; ^∗∗∗∗P < 0.0001.

Fig. 4.

Increased susceptibility of unpaired mutant D. melanogaster to asparagine-limited M. abscessus. Survival curves of (A) w¹¹¹⁸ (control), (B) upd2^∆, (C) upd3^∆, or (D) upd2^∆upd3^∆ flies infected with M. abscessus ATCC 19977 (red) or ∆MAB_1132c strains (blue) grown in complete media (7H9). Control groups are uninfected flies (black) and PBS Tween80 0.1% injected flies (gray). Data represent three independent experiments with a minimum of 20 flies per group per experiment and were analyzed using the log-rank test. Asterisks in legends indicate significant differences compared against ATCC 19977. RT-qPCRs show transcription of antimicrobial peptides in w¹¹¹⁸ and upd2^∆upd3^∆ flies infected with M. abscessus ATCC 19977 (red) or ∆MAB_1132c-1 (blue) at 80 h postinfection. Uninjected (black) and PBS Tween80 0.1% sterile-injury (gray) flies are included as controls. Transcript levels shown are (E) Attacin A, (F) Diptericin, (G) Drosocin, and (H) Turandot A. Data are representative of two independent experiments (N = 8), normalized against RpL4, and were compared using a Kruskal–Wallis ANOVA. ^∗P < 0.05; ^∗∗P < 0.01; ^∗∗∗P < 0.001; ^∗∗∗∗P < 0.0001.

Fig. 5.

Reduction of activated AKT in D. melanogaster by M. abscessus is partially dependent on bacterial asparagine uptake and host upd2 signaling. Systemic levels of phosphorylated Ser505 (pAKT) and total AKT protein of (A) w¹¹¹⁸ and (B) upd2^∆ flies at 48 and 96 h postinfection with M. abscessus ATCC 19977 (red) or ∆MAB_1132c-2 (blue) cultured in complete 7H9 media, measured by western blot. The control group is PBS Tween80 0.1% injected flies (gray). Data represent one independent experiment; n = 3 per strain per genotype. Groups were compared with one-way ANOVA with Tukey’s test. Survival of (C) w¹¹¹⁸; +; ilp2-Gal4/+ (control) and (D) w¹¹¹⁸; tub-Gal80^[ts]/+; ilp2-Gal4/UAS-reaper (IPC-depleted) flies infected with M. abscessus ATCC 19977 (red) or ∆MAB_1132c-1 (blue). Control groups are PBS Tween80 0.1% injected flies (gray). Data represent one experiment, with at least 20 flies per group. Asterisks in the legend indicate the P-value compared to flies infected with ATCC 19977, compared using the log-rank test. RT-qPCRs show transcript levels of (E) PepCK and (F) Thor in w¹¹¹⁸ flies infected with M. abscessus ATCC 19977 (red) or ∆MAB_1132c strains (blue) over the course of an infection. Uninjected (black) and PBS Tween80 0.1% sterile-injury (gray) flies are included as controls. Data are representative of three independent experiments (N = 12), normalized against RpL4, and were compared using a Kruskal–Wallis ANOVA. Statistical comparisons between M. abscessus ATCC 19977 and the ∆MAB_1132c strains are shown. ^∗P < 0.05; ^∗∗P < 0.01; ^∗∗∗P < 0.001; ^∗∗∗∗P < 0.0001.