The Gut Commensal Microbiome of Drosophila melanogaster Is Modified by the Endosymbiont Wolbachia

Abstract

Endosymbiotic Wolbachia bacteria and the gut microbiome have independently been shown to affect several aspects of insect biology, including reproduction, development, life span, stem cell activity, and resistance to human pathogens, in insect vectors. This work shows that Wolbachia bacteria, which reside mainly in the fly germline, affect the microbial species present in the fly gut in a lab-reared strain. Drosophila melanogaster hosts two main genera of commensal bacteria-Acetobacter and Lactobacillus. Wolbachia-infected flies have significantly reduced titers of Acetobacter. Sampling of the microbiome of axenic flies fed with equal proportions of both bacteria shows that the presence of Wolbachia bacteria is a significant determinant of the composition of the microbiome throughout fly development. However, this effect is host genotype dependent. To investigate the mechanism of microbiome modulation, the effect of Wolbachia bacteria on Imd and reactive oxygen species pathways, the main regulators of immune response in the fly gut, was measured. The presence of Wolbachia bacteria does not induce significant changes in the expression of the genes for the effector molecules in either pathway. Furthermore, microbiome modulation is not due to direct interaction between Wolbachia bacteria and gut microbes. Confocal analysis shows that Wolbachia bacteria are absent from the gut lumen. These results indicate that the mechanistic basis of the modulation of composition of the microbiome by Wolbachia bacteria is more complex than a direct bacterial interaction or the effect of Wolbachia bacteria on fly immunity. The findings reported here highlight the importance of considering the composition of the gut microbiome and host genetic background during Wolbachia-induced phenotypic studies and when formulating microbe-based disease vector control strategies. IMPORTANCEWolbachia bacteria are intracellular bacteria present in the microbiome of a large fraction of insects and parasitic nematodes. They can block mosquitos' ability to transmit several infectious disease-causing pathogens, including Zika, dengue, chikungunya, and West Nile viruses and malaria parasites. Certain extracellular bacteria present in the gut lumen of these insects can also block pathogen transmission. However, our understanding of interactions between Wolbachia and gut bacteria and how they influence each other is limited. Here we show that the presence of Wolbachia strain wMel changes the composition of gut commensal bacteria in the fruit fly. Our findings implicate interactions between bacterial species as a key factor in determining the overall composition of the microbiome and thus reveal new paradigms to consider in the development of disease control strategies.

Keywords: Drosophila microbiome; Wolbachia; gut microbiome; symbiosis.

FIG 1

16S rRNA gene profiling of D. melanogaster shows reduction of Acetobacteraceae levels in Wolbachia-infected adult flies. (A) Grape juice agar plates used for D. mauritiana egg collection. Grape juice acts as a pH indicator, turning pink under acidic conditions and yellow under basic conditions. Over time, the microbial composition of the feces of Wolbachia-infected flies turns plates more yellowish than plates used by Wolbachia-free flies. (B) Schematic of the bacterial 16S rRNA gene. Hypervariable regions V1 to V9 are red. Primers (arrows, 27F and 338R) amplify regions V1 and V2. In BstZ17I restriction enzyme digestion of the total 16S rRNA gene pool, only the Wolbachia rRNA gene between V1 and V2 is selectively digested. (C) Agarose gel image showing the 16S rRNA gene PCR products from the microbiome of D. melanogaster and efficient digestion of the Wolbachia 16S rRNA gene amplicon by the BstZ17I restriction enzyme. The red arrowhead indicates the digested Wolbachia product. The pie charts indicate the percentages of Wolbachia reads before and after BstZ17I digestion. (D) 16S rRNA gene profiles of male and female wMel-free (W−) or -infected D. melanogaster. The proportions of Acetobacteraceae are significantly different under every pair of conditions (P < 0.0001, chi-square test with Yates correction).

FIG 2

Wolbachia bacteria suppress A. pasteurianus across various life stages of D. melanogaster. (A) Schematic of the experimental setup used. Stopwatches indicate sample collection times. (B to E) PCR products obtained with A. pasteurianus and L. plantarum species-specific primers from BstZ17I-digested total genomic DNA from wMel-free (W−) or -infected D. melanogaster. (B) The parental flies are F0 0- to 7-day-old males and females. (C to E) Experiments were done in triplicate, and each sample consisted of five adults (B, D, E) or five larvae (C). Vial numbers are above the gel lanes. (C) F1 unsexed L3 larvae (n = 3), (D) F1 0- to 7-day-old male and female flies (n = 3). (E) F1 7- to 14-day-old male and female flies (n = 3).

FIG 3

Wolbachia infection reduces A. pasteurianus levels in L3 larvae of gnotobiotic flies. (A) Schematic of the experimental setup used. Stopwatches indicate sample collection times. (B, C) qPCR of A. pasteurianus (B) and L. plantarum (C) titers in wMel-infected gnotobiotic flies compared to those in Wolbachia-free (W−) gnotobiotic flies, normalized to 16S rRNA gene levels, in L3 larvae, 0- to 7-day-old adults, and 7- to 14-day-old adults. Bar graphs show mean values (three biological replicates of a sample of five individuals per replicate), and error bars show standard deviations. Asterisks indicate statistically significant differences (P < 0.05, Student t test). (D) Proportion of individual L3 larvae that had no A. pasteurianus in the gut (n = 30, chi-square test; error bars show confidence intervals). (E, F) Box-and-whisker plots of levels of bacteria in L3 larvae (E) and 10-day-old adults (F) (n = 30). Median values are shown next to the boxes, and P values (two-sided Wilcoxon rank sum test) are reported.

FIG 4

A. pasteurianus is absent from Wolbachia-infected (W+) L3 larval guts. (A to F) Composites of z-stack projections of confocal images of gnotobiotic L3 larval midguts. (A to C) Wolbachia-free (W−) guts. (D to F) Wolbachia-infected guts. (A, D) Wolbachia channel. (B, E) A. pasteurianus channel. (C, F) Merged images of Wolbachia (red), A. pasteurianus (green), and DNA (blue).

FIG 5

Wolbachia bacteria are present in gut cells but absent from the lumen. Composite confocal images of the whole midgut and hindgut of Wolbachia-infected (W+) flies (A) and Wolbachia-free (W−) flies (F) and the respective Wolbachia channels (A′ and F′). Magnification (×60) of the midgut (B to D) and hindgut (E) regions shows that Wolbachia bacteria are present intracellularly in gut cells but absent from the lumen of the gut (the lumen is marked by green autofluorescence).

FIG 6

Wolbachia bacterial infection does not affect the expression of Imd pathway components and ROS-producing oxidases. The relative expression of Imd pathway signal transducer imd (A), the transcription factor Rel (A), AMPs (B), Nox (C), and AMPs downstream of JAK/STAT signaling in both axenic and gnotobiotic L3 larval guts in the presence or absence of Wolbachia infection was determined by RT-qPCR. All conditions are normalized to Wolbachia-free (W−) axenic L3 larval guts. Bar graphs show mean values (three biological replicates of 10 larvae each), and error bars show standard deviations. Asterisks indicate statistically significant differences (P < 0.05, Student t test).