Metabolic Cooperation among Commensal Bacteria Supports Drosophila Juvenile Growth under Nutritional Stress

Abstract

The gut microbiota shapes animal growth trajectory in stressful nutritional environments, but the molecular mechanisms behind such physiological benefits remain poorly understood. The gut microbiota is mostly composed of bacteria, which construct metabolic networks among themselves and with the host. Until now, how the metabolic activities of the microbiota contribute to host juvenile growth remains unknown. Here, using Drosophila as a host model, we report that two of its major bacterial partners, Lactobacillus plantarum and Acetobacter pomorum, engage in a beneficial metabolic dialogue that boosts host juvenile growth despite nutritional stress. We pinpoint that lactate, produced by L. plantarum, is utilized by A. pomorum as an additional carbon source, and A. pomorum provides essential amino acids and vitamins to L. plantarum. Such bacterial cross-feeding provisions a set of anabolic metabolites to the host, which may foster host systemic growth despite poor nutrition.

Keywords: Biological Sciences; Microbiology; Microbiome.

Graphical abstract

Figure 1

Bi-Association with Ap^WJL and Lp^NC8 Enhances Commensal-Mediated Benefit on Larval Development (A and B) Developmental timing (time from egg to metamorphosis) on complete holidic diet (HD) of Germ-Free (GF) larvae (gray) or GF larvae inoculated with high dose (10⁷ or 10⁸ CFU) respectively; (A) or low dose (10⁵ CFU); (B) of Ap^WJL and/or Lp^NC8 (Ap^WJL, black; Lp^NC8, green; Ap^WJL:Lp^NC8, purple). D₅₀: Day when 50% of the larvae population has entered metamorphosis. (C) Larval length at every day post-embryogenesis of GF larvae or post-inoculation (Day 1) with 10⁵ CFU of Ap^WJL and/or Lp^NC8 or Ap^WJL mono-associated larvae supplemented with DL-lactate at a final concentration of 0.6 g/L (red). Θ, pupae detected in the vial. (D and E) Microbial load (Ap^WJL, D; Lp^NC8, E) of larvae mono- or bi-associated with 10⁵ CFU of Ap^WJL and/or Lp^NC8. (F and G) Growth in liquid HD (F) and growth rates (G) of Ap^WJL and Lp^NC8 in mono- (plain lines) or cocultures (dashed lines) in liquid HD. Gray always refers to GF, black to Ap^WJL mono-association, green to Lp^NC8 mono-association condition, and purple to Ap^WJL:Lp^NC8 bi-association. Each symbol represents an independent replicate except in (F) where symbols represent the means ± SEM of three biological replicates. Boxplots show minimum, maximum, and median where each point is a biological replicate. Dot plots show mean ± SEM. (A and B) We performed Kruskal-Wallis test followed by uncorrected Dunn's tests to compare each gnotobiotic condition with GF. (D and E) Each point represents a biological replicate comprising the average microbial load of a pool of 10 larvae. We performed Mann-Whitney test to compare microbial loads in mono-association with microbial loads in bi-association for the strain of interest at each time point. (G) We performed Mann-Whitney test to compare the growth rate in monoculture to the growth rate in coculture for the strain of interest. ns: non-significant, ∗: p value<0,05, ∗∗: p value<0.005, ∗∗∗: p value<0.0005, ∗∗∗∗: p value<0.0001. See also Figures S1 and S2.

Figure 2

Ap^WJL Benefits Lp^NC8 via Essential Amino Acid and Vitamins Provision (A–G) Growth curves of Ap^WJL and Lp^NC8 in mono- (plain lines) or cocultures (dotted lines) in liquid holidic diets (HD) lacking Arg (HDΔArg) (A), Cys (HDΔCys) (B), Ile (HDΔIle) (C), Leu (HDΔLeu) (D), Val (HDΔVal) (E), Biotin (HDΔBiotin) (F) or Pantothenate (HDΔPantothenate) (G). Black refers to Ap^WJL, green the Lp^NC8. (H–J) HPLC quantification of Arg, Ile, and Leu in Ap^WJL or Lp^NC8 mono-culture supernatants (black and green lines, respectively) or Ap^WJL:Lp^NC8 coculture (purple line) in HDΔArg, HDΔIle, HDΔLeu, respectively. (A–J) Symbols represent the means ± SEM of three biological replicates. (K) Heatmap representing the mean D₅₀ (day when 50% of the larvae population has entered metamorphosis) of GF larvae (first column) and larvae mono-associated with Ap^WJL or Lp^NC8 or bi-associated with Ap^WJL:Lp^NC8 (columns 2, 3, and 4, respectively). Each row shows D₅₀ in a different version of the HD: complete HD or HDs each lacking a specific nutrient HDΔArg, HDΔIle, HDΔLeu, HDΔVal, HDΔCys, HDΔBiotin, HDΔPantothenate. White color code means that larvae did not reach pupariation.

Figure 3

Lp^NC8-Derived Lactate Benefits Ap^WJL and Enhances Ap^WJL-Mediated Larval Growth Promotion (A and M) Developmental timing of Germ-Free (GF) larvae or GF larvae inoculated with 10⁵ CFU of Ap^WJL (A, M black) or Lp^NC8 (A, green) supplemented with either sterile PBS (A) or the supernatant from a 72-h culture of Lp^NC8 (A, M), Lp^WCSF1 (M, turquoise), or Lp^WCFS1ΔldhDL (M, light green) in complete holidic diet (HD). (B) Developmental timing on HD of GF larvae (gray) or GF larvae inoculated with 10⁵ CFU of Ap^WJL supplemented with either sterile PBS (black) or DL-lactate solutions (red) at inoculation (final concentration in the diet 0.06 or 0.6 g/L). (A, B, and M) Each dot represents an independent biological replicate. Boxplots show minimum, maximum, and median. We performed Kruskal-Wallis test followed by uncorrected Dunn's tests to compare each condition with GF. ns: non-significant, ∗: p value<0,05, ∗∗: p value<0.005, ∗∗∗: p value<0,0005. (C–H) (C) Developmental timing of GF larvae inoculated with 10⁵ CFU of Ap^WJL supplemented at inoculation with either sterile PBS (black) or DL-lactate at final concentration of 0.6 g/L in HDs lacking each an essential amino acid for Drosophila: from left to right, HDΔArg, HDΔHis, HDΔIle, HDΔLeu, HDΔLys, HDΔMet, HDΔPhe, HDΔThr, and HDΔVal. Boxplots show minimum, maximum, and median, and each dot represents an independent biological replicate. Growth curves (D and E) and growth rates (F) of Ap^WJL in liquid HD supplemented (E) or not (D) with DL-lactate solution. D- (dotted line) and L-lactate (dashed line) levels (red) were quantified in both conditions. Growth rates of Ap^WJL in solid HD and HD + DL-lactate with (H) or without (G) larvae. (I–L) Growth curves in liquid HD of Lp^NC8 (green) or Ap^WJL (black) in mono- (I) or coculture (J), or Lp^WCSF1 (K, green) or Lp^WCFS1ΔldhDL (L, dotted green) with the respective D- (dotted line) or L-lactate (dashed line) levels (red). Note the low OD₆₀₀ of Lp^WCFS1ΔldhDL versus Lp^WCSF1 but similar CFU counts (Figures S4A and S4B). Symbols represent the means ± SEM of three biological replicates except for (F)–(H) where each symbol represents an independent replicate ±SEM. See also Figures S1, S3, and S4.

Figure 4

Upon Lactate Consumption Ap^WJL Produces an Amino Acid Cocktail that Enhances the Growth-Promoting Ability of Lp^NC8 (A and B) Net production of essential and non-essential fly amino acids at 24, 48, and 72 h. Net production was calculated from HPLC quantification data by subtracting the amino acid concentration quantified at 0 h from each incremental time point. Conditions included the supernatant of Ap^WJL cultures (inner panels) in complete HD supplemented (B) or not (A) with DL-lactate. Symbols in inner pannels represent the means ± SEM of three biological replicates. Bars represent the means ± SEM of three biological replicates. (C–E) Developmental timing of GF larvae (C) inoculated with 10⁵ CFU of Ap^WJL (D) or 10⁵ CFU of Lp^NC8 (E) supplemented with either sterile PBS, the amino acid mix produced by Ap^WJL in liquid culture at 48 h (+AA mix Ap @48h), the amino acid mix produced by Ap^WJL in liquid culture supplemented with DL-lactate at 24 h (+AA mix Ap + Lactate @24h) or the amino acid mix produced by Ap^WJL in liquid culture supplemented with DL-lactate at 48 h (+AA mix Ap + Lactate @48h). See Table S1 for detailed information on the amino acid mixes. Boxplots show minimum, maximum, and median; each point represents a biological replicate. We performed Kruskal-Wallis test followed by uncorrected Dunn's tests to compare each condition with the PBS-treated condition. ns: non-significant, ∗∗: p value<0.005.

Figure 5

Lactate Utilization by Acetobacter Is Central to Its Physiological Response to Lp^NC8 and Enhanced Benefit on Host Growth (A) Schematic representation of the main metabolic routes of DL-lactate utilization by Acetobacter species. Purple: Fly's essential amino acids. Yellow: Fly's non-essential amino acids. Blue: genes related with lactate consumption. (B) Developmental timing of Germ-Free (GF, gray) larvae or GF larvae inoculated with 10⁵ CFU of A. fabarum^DsW_054 (Af, orange), Lp^NC8 (green), both strains (Af:Lp^NC8, purple), or Af supplemented with the supernatant from 72-h culture of Lp^NC8 (black, filled green) in complete HD. (C) Developmental timing of GF (gray) larvae or GF larvae inoculated with 10⁵ CFU of Af (orange), Af::Tnldh (10B7) (blue) or Af::Tnldh (92G1) (brown) supplemented with sterile PBS, DL-lactate, D-lactate, or L-lactate in complete HD. (D) Developmental timing of GF (gray) larvae or GF larvae inoculated with 10⁵ CFU of Af (orange) or Af (red), Af:Tnals (brown), Af:Tnaldc (brown), Af:Tnpdc (brown) supplemented with DL-lactate in complete HD or complete HD supplemented with 50 μg/mL of kanamycin (GF and Af mutants). Boxplots show minimum, maximum, and median; each point represents a biological replicate. We performed Kruskal-Wallis test followed by uncorrected Dunn's tests to compare each condition with the GF treated condition or the Af condition when indicated. ns: non-significant, ∗: p value<0.05 ∗∗: p value<0.005, ∗∗∗∗: p value<0.0001. See also Figure S5.

Figure 6

Lactate-Dependent Acetobacter Stimulation of Larval Growth Evokes Metabolites Release Enhancing Host Anabolic Metabolism and Resistance to Oxidative Stress (A) Schematic representation of sample preparation for metabolomic analysis. (B) Outsourced untargeted metabolomics and data analysis pipeline. (C) Investigator-driven data analysis and biological interpretation. (D) Venn diagram of the identified metabolites in the three test conditions. Our analysis points to 45 metabolites of interest belonging to all major metabolite families. See Table 1 for a detailed list of metabolites.