Gut symbiont enhances insecticide resistance in a significant pest, the oriental fruit fly Bactrocera dorsalis (Hendel)

Abstract

Background: Symbiotic bacteria affect insect physiology and ecology. They may also mediate insecticide resistance within their hosts and thereby impact pest and vector control practices. Here, we document a novel mechanism of insecticide resistance in which a gut symbiont of the tephritid pest fruit fly Bactrocera dorsalis enhances resistance to the organophosphate insecticide trichlorphon.

Results: We demonstrated that the gut symbiont Citrobacter sp. (CF-BD) plays a key role in the degradation of trichlorphon. Based on a comparative genomics analysis with other Citrobacter species, phosphatase hydrolase genes were identified in CF-BD. These CF-BD genes had higher expression when trichlorphon was present. Bactrocera dorsalis inoculated with isolated CF-BD obtained higher trichlorphon resistance, while antibiotic-treated flies were less resistant confirming the key role of CF-BD in insecticide resistance.

Conclusions: Our findings suggest that symbiont-mediated insecticide resistance can readily develop in B. dorsalis and may represent a more widely relevant insecticide resistance mechanism than previously recognized.

Keywords: Bactrocera dorsalis; Insecticide resistance; Oriental fruit fly; Symbiotic bacteria; Trichlorphon.

Fig. 1

Different symbiotic bacterial communities in the gut of susceptible and resistant strains of Bactrocera dorsalis. a Toxicity regression analyses of susceptible (SS; green line) and resistant (RS; red line) fly responses to trichlorphon. b Non-metric multidimensional scaling (nMDS) plot exhibiting a structural difference (according to the Bray-Curtis index) between the bacterial communities of SS and RS. c A heat map reflecting the mean number of the prominent OTUs in SS and RS. The data were log-transformed before plotting. d A stack map at the genus level for six samples. e Differences in the numbers (mean ± standard error (SE)) of tags that were assigned to Citrobacter between SS and RS. Statistically significant differences at the p < 0.01 level, as evaluated with independent-sample t tests are indicated by two asterisks above the bars

Fig. 2

Citrobacter sp. isolation and identification. a Colony characteristics on brain heart infusion (BHI) agar plates for bacteria that were isolated from the gut of SS and RS flies. b Phylogenetic relationships of the symbiotic CF-BD strain. The red star indicates the trichlorphon-degrading CF-BD strain. A maximum likelihood phylogeny inferred from 1521 aligned nucleotide sites in 16S rDNA gene sequences is presented with bootstrap values. The accession numbers of the 16S rDNA sequences for each bacterium are listed in the square brackets. c The rod-shaped CF-BD strain has blunt, round ends and a red color, as identified by Gram staining. d, e The midgut organization of B. dorsalis and CF-BD localization in the midgut of RS flies. Red signals indicate CF-BD symbionts, whereas blue signals show host insect nuclei. Abbreviations: M, midgut; Mt, Malpighian tubules; H, hindgut; Rp, rectal pads

Fig. 3

Detection of CF-BD in Bactrocera dorsalis from diverse populations. GZ Guangzhou, YF Yunfu, ZJ Zhanjiang, HBI Hebao island, TS Taishan, WZI Weizhou island, ZH Zhuhai, MZ Mengzi, CD Chengdu, SY Sanya, NN Nanning, ZZ1 Zhengzhou, and ZZ2 Zhangzhou. Bars (mean ± SE) labeled with the same letter within each treatment are not significantly different (p > 0.05, Tukey’s test)

Fig. 4

Trichlorphon tolerance and degradation ability of CF-BD and its effects on the drug susceptibility and resistance of flies. a A comparison of the number of CF-BD colony forming units (mean ± SE) between the RS flies and those that were given streptomycin. b Toxicity regression analyses of the RS (red line) and streptomycin-fed RS (green line) fly responses to trichlorphon. c A comparison of CF-BD colony numbers (mean ± SE) between SS flies and CF-BD-fed SS flies. d Toxicity regression analyses of of SS (green line) and CF-BD-fed SS (red line) fly responses to trichlorphon. e Characteristics of bacteria that were cultivated on trichlorphon-enriched plates and normal plates. f Diameters of the bacterial colonies (mean ± SE) cultivated on trichlorphon-enriched plates and normal plates. g The gas chromatograph identification of trichlorphon in the filtrate of mineral media with or without CF-BD cultures. Differences in the trichlorphon concentrations were evaluated by comparing the peak areas (mean ± SE) that were identified by gas chromatography. h The gas chromatography mass spectrometry (GC-MS) identification of the trichlorphon degradation products of CF-BD (the degradation pathway is presented). The significant differences at the p < 0.01 level, as evaluated with independent-sample t tests, are indicated by two asterisks above the bars. ns indicates not significant. Tr trichlorphon-enriched plates, CK control plates

Fig. 5

CF-BD genomic characterization and comparative analysis. a Synteny analysis between the genome of Citrobacter freundii, Citrobacter koseri, Citrobacter rodentium, and CF-BD. b Genomic components comparison between C. freundii, C. koseri, C. rodentium, and CF-BD. c Alignment of the CF-BD genome (circle 1 from the outside in) with C. freundii (circle 2), C. koseri (circle 3), and C. rodentium (circle 4) by applying BLAST and BLAST Ring Image Generator (BRIG). Similarity is symbolized by colored blocks. The more intense the color, the higher the similarity. CF-BD-specific regions are highlighted by black frames labeled with Roman numerals (I–III). Circles 5, 6, and 7 represent the predicted coding DNA sequence (CDS), the ORF of the function-annotated genes, and the phosphatase genes, respectively. Circles 8 and 9 represent the GC skew and GC content, respectively. d The phylogenetic relationship of the different phosphate hydrolase families. Family members with green dots are from CF-BD and others are organophosphorus hydrolase genes of other bacteria. The red trees indicate potential OPH genes in CF-BD (accession numbers for genes are listed in Additional file 8: Table S6)

Fig. 6

Relative expression of the five OPH-like genes stimulated by trichlorphon. Expression of OPH genes was measured against the reference gene recA. Bars (mean ± SE) labeled with the same letter within each treatment are not significantly different (p > 0.05, Tukey’s test)