Microbiome derived acidity protects against microbial invasion in Drosophila

Abstract

Microbial invasions underlie host-microbe interactions that result in microbial pathogenesis and probiotic colonization. While these processes are of broad interest, there are still gaps in our understanding of the barriers to entry and how some microbes overcome them. In this study, we explore the effects of the microbiome on invasions of foreign microbes in Drosophila melanogaster. We demonstrate that gut microbes Lactiplantibacillus plantarum and Acetobacter tropicalis improve survival during invasion of a lethal gut pathogen and lead to a reduction in microbial burden. Using a novel multi-organism interactions assay, we report that L. plantarum inhibits the growth of three invasive Gram-negative bacteria, while A. tropicalis prevents this inhibition. A series of in vitro and in vivo experiments revealed that inhibition by L. plantarum is linked to its ability to acidify both internal and external environments, including culture media, fly food, and the gut itself, while A. tropicalis diminishes the inhibition by quenching acids. We propose that acid produced by the microbiome serves as an important gatekeeper to microbial invasions, as only microbes capable of tolerating acidic environments can colonize the host. The methods described herein will add to the growing breadth of tools to study microbe-microbe interactions in broad contexts.

Figure 1.. Presence of microbiome members reduces host susceptibility to P. entomophila.

A) Scheme describing the generation of gnotobiotic flies. B) Kaplan-Meier survival analysis of female flies infected with P. entomophila (Pe) via feeding (feeding occurred from t=0 to t=1 day); fly microbiome treatments included no bacteria (Axenic), L. plantarum (Lp), A. tropicalis (At), or a combination of Lp and At (Lp/At). The mock infected group shows survival of axenic flies fed LB media; survival of all gnotobiotic conditions fed LB were also recorded but were not significantly different to axenic control and are not shown. Log-rank statistical analyses of each infection condition are compared to the other conditions. Significance is expressed as follows: NS, not significant; **, P ≤ 0.01; ****, P ≤ 0.0001. n=60 flies per control & 110-160 flies per Pe-infected treatment over 3 independent replicates. C) Number of colony forming units (CFUs) of bacteria per fly infected with Pe screened in axenic flies and gnotobiotic flies colonized with Lp, At, and Lp/At. Flies were sacrificed to determine Pe load at 24 hours, 72 hours, and 7 days post feeding infection. Each point represents bacterial load from an individual fly; bars and error bars represent the median and 95% confidence intervals; limit of detection is 2x10² CFUs/Fly. n=9 flies per control & 10-15 flies (depending on availability of living flies) per infected treatment per time point over 3 biological replicates. Statistical significance was determined between flies containing different microbiomes by the Kruskal-Wallis method with Dunn’s multiple comparisons analysis. P-values are represented as follows: *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Figure 2.. Microbiome composition alters the microbial load of invasive organisms during infection.

Number of colony forming units (CFUs) of bacteria per fly infected with Ecc15 (A), and EcN (B). Each invasive organism was screened in axenic flies and gnotobiotic flies colonized with Lp, At, and Lp/At. Bacterial load was determined at 3 hours, 24 hours, and 48 hours post feeding infection. Each point represents bacterial load from an individual fly; bars and error bars represent the median and 95% confidence intervals; limit of detection is 2x10² CFUs/Fly. n=15 flies per treatment per time point over 3 biological replicates. Statistical significance was determined between flies containing different microbiomes by the Kruskal-Wallis method with Dunn’s multiple comparisons analysis. P-values are represented as follows: *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

Figure 3.. In vitro growth assays reveal inhibition of Gram-negative invasive organisms by Lp.

A) Scheme describing the multi-organism interaction assay procedure. B) Multi-organism interaction assays display growth effects of gut microbes Lp and At on invasive organisms Pe, Ecc15, and EcN. Microbiome members were grown in perpendicular streaks for 3 days, and invasive organisms were added to the adjacent quadrants and allowed to grow for an additional day. Zones of inhibition are indicated by dashed lines. C) Co-culture analysis of invasive organisms with microbiome members. Pe, Ecc15, and EcN were grown in mono-culture or in media previously inoculated with Lp, At, or a mixture of both (LpAt). Bars and error bars represent the mean concentration of invasive organisms in CFUs/mL ±SEM. N=3 biological replicates per condition.

Figure 4.. Microbiome-derived shifts in pH contribute to inhibition by Lp.

A) Multi-organism interaction assay showing the effects of Lp and At on EcN growth on media containing the pH indicator bromophenol blue. After three days of growth, a yellow acidic zone appears around Lp, but it tapers off near the At interaction point. When EcN is added, the zone of inhibition overlaps with the acidic region. B) pH measurement of microbiome mono-cultures and co-cultures reveals a sharp acidification of culture media by Lp. C) Density of invasive organisms in media adjusted to pH 4.0 with lactic acid (LA) with or without 24 hours of prior growth with At. Each bar represents mean CFUs/mL ±SEM of 3 biological replicates. D) Microbial concentration of invasive organisms in media buffered to pH 6.0 with phosphate buffer with or without 24 hours of prior growth with Lp. Each bar represents mean CFUs/mL ±SEM for 3 biological replicates. E) Ratio of Ecc15 microbial load in flies infected on buffered vs. standard fly diets, n=15 flies per treatment per time point over three biological replicates. Statistical significance was determined for flies of each microbiome status between standard and buffered diets using the Kruskal-Wallis method with Dunn’s multiple comparisons analysis. P-values are represented as follows: NS, P>0.05; **, P ≤ 0.01. F) The pH of the copper cell region of the intestine was approximated by feeding axenic and gnotobiotic flies food soaked with 2% bromophenol blue. Guts were dissected and imaged immediately. A yellow/green color in the copper cell region indicates an acidified environment (pH < 4.6).

Figure 5.. Microbiome composition alters the chemical environment of fly food.

A) pH analysis of fly food after three days of growth of Lp, At, or Lp/At, with culture media added as a control. Bars represent the mean pH of three biological replicates ±SEM. B) pH analysis of fly food three days after the addition of 40 male flies of different microbiome statuses (Axenic, Lp, At, Lp/At). Bars represent the mean pH ±SEM of three biological replicates. C) Bacterial load of Pe, Ecc15, and EcN present on fly food initially inoculated with microbiome members Lp, At, or Lp/At, with culture media as a control. Each bar represents the mean CFUs/g fly food ±SEM for three biological replicates. D) Pictorial representation of microbe-microbe interactions between microbiome members and Gram-negative invasive bacteria on fly food and during consumption by the host.