Wolbachia is a group of intracellular symbiotic bacteria that widely infect arthropods and nematodes. Wolbachia infection can regulate host reproduction with the most common phenotype in insects being cytoplasmic incompatibility (CI), which results in embryonic lethality when uninfected eggs fertilized with sperms from infected males. This suggests that CI-induced defects are mainly in paternal side. However, whether Wolbachia-induced metabolic changes play a role in the mechanism of paternal-linked defects in embryonic development is not known. In the current study, we first use untargeted metabolomics method with LC-MS to explore how Wolbachia infection influences the metabolite profiling of the insect hosts. The untargeted metabolomics revealed 414 potential differential metabolites between Wolbachia-infected and uninfected 1-day-old (1d) male flies. Most of the differential metabolites were significantly up-regulated due to Wolbachia infection. Thirty-four metabolic pathways such as carbohydrate, lipid and amino acid, and vitamin and cofactor metabolism were affected by Wolbachia infection. Then, we applied targeted metabolomics analysis with GC-MS and showed that Wolbachia infection resulted in an increased energy expenditure of the host by regulating glycometabolism and fatty acid catabolism, which was compensated by increased food uptake. Furthermore, overexpressing two acyl-CoA catabolism related genes, Dbi (coding for diazepam-binding inhibitor) or Mcad (coding for medium-chain acyl-CoA dehydrogenase), ubiquitously or specially in testes caused significantly decreased paternal-effect egg hatch rate. Oxidative stress and abnormal mitochondria induced by Wolbachia infection disrupted the formation of sperm nebenkern. These findings provide new insights into mechanisms of Wolbachia-induced paternal defects from metabolic phenotypes.

Wolbachia are among the most successful intracellular bacteria that can infect many arthropod species as well as filarial nematodes. Wolbachia can manipulate the reproduction of their insect hosts to enhance their transmission through host populations. Cytoplasmic incompatibility (CI) is the most common phenotype induced by Wolbachia in insect hosts, which results in embryonic mortality when Wolbachia infected males cross with normal (uninfected) females. This indicates that the impact of Wolbachia infection is mainly on paternal side. Evidence has suggested that the metabolism is strongly linked with male reproduction. We were interested in investigating whether Wolbachia-induced metabolic changes involve in the mechanism of paternal defects in embryogenesis. Here, using comparative metabolomics method, we identified 414 potential differential metabolites related to 34 metabolic pathways between Wolbachia-bearing and Wolbachia-free male Drosophila melanogaster. Then, by targeted metabolomics analysis, gene overexpression, and transmission electron microscopy (TEM) techniques, we found that in Wolbachia infected D. melanogaster, energy metabolism and oxidative stress were enhanced, and the development of mitochondria-derived nebenkern during spermatogenesis was impaired. These may be the main causes of paternal defects induced by Wolbachia.

Citation: Zhang H-B, Cao Z, Qiao J-X, Zhong Z-Q, Pan C-C, Liu C, et al. (2021) Metabolomics provide new insights into mechanisms of Wolbachia-induced paternal defects in Drosophila melanogaster. PLoS Pathog 17(8): e1009859. https://doi.org/10.1371/journal.ppat.1009859

Editor: Elizabeth A. McGraw, Pennsylvania State University, UNITED STATES

Received: January 1, 2021; Accepted: August 1, 2021; Published: August 12, 2021

Copyright: © 2021 Zhang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: This work was supported by the National Natural Science Foundation of China (No. 31872288) to YFW (http://nsfc.gov.cn) and National Key Research and Development Project (2018YFE0110800) to LZ (http://most.gov.cn). The sponsors or funders did not play any roles in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Wolbachia are endosymbiotic bacteria that frequently infect arthropods and nematodes[1,2]. They can influence host reproduction by various strategies. Cytoplasmic incompatibility (CI) is the most common phenotype induced by Wolbachia in insect hosts, which causes lethality of embryos derived from the crosses between Wolbachia infected males and either uninfected females or females carrying a different strain of Wolbachia. However, when the Wolbachia-infected males mate with females infected with the same strain of Wolbachia, their offsprings develop properly[1,3,4]. CI has attracted considerable interest because it can be utilized to control agricultural pests and reduce the spread of insect-borne diseases[5–7]. The fact that Wolbachia-infected male insects caused no or few progenies when mated with uninfected females suggests that Wolbachia might induce paternal defects in embryonic development in insect hosts.

Numerous studies have suggested that Wolbachia infection induces modified sperms. Once these sperms fertilize uninfected eggs, the maternal histone deposition in the male pronucleus is delayed, causing improper chromosome condensation and cell division, and thus resulting in embryonic mortality[8–11]. The presence of Wolbachia in the female can rescue these effects on the paternal chromosomes and restore the embryonic development after fertilization[10,12,13]. These ideas have been formalized into “mod-resc” (modification/rescue) model[14], which was further interpreted by several models, such as “lock-and-key” model, “titration-restitution” model, and the “mistiming” model[15]. All the models exhibit that Wolbachia may deposit or remove some factors in or from host sperms. Recently, researchers have proposed that Wolbachia could deposit some “toxic” material in maturing sperm, such as deubiquitylase or nuclease, which can cause embryonic lethality[3,16,17]. In Drosophila melanogaster, the Wolbachia deubiquitylase CidB can bind to host nuclear import receptor, thus may impair protamine-histone exchange in sperm nucleus after fertilization, which might be the main cause of the defect in maternal histone deposition onto the paternal DNA[18]. On the other hand, sets of studies have identified host factors, including some metabolic enzymes, which could contribute to the phenotypes similar to the paternal defects induced by Wolbachia[19–22]. For instance, iLvE, coding for branched-chain-amino-acid aminotransferase which mediates branched-chain amino acid biosynthesis, exhibited the highest level of down-regulation in Wolbachia-infected Laodelphax striatellus testes versus uninfected testes. Knockdown of iLvE in Wolbachia-free males led to a significant decrease in egg hatch rate[21]. Our previous work on gene expression analyses between the Wolbachia-infected and uninfected 3rd instar larval testes of D. melanogaster showed that significantly changed genes associated with biological process were mainly involved in host metabolism, including the upregulations of genes involved in fatty acid biosynthesis, glycolysis, and glutamate synthesis[19]. Further comparative analysis of proteins from the spermathecae and seminal receptacles of females that mated with infected or uninfected male flies indicated that the largest group of differentially expressed proteins was also highly relevant to host metabolism[20]. Collectively, Wolbachia infection may have a notable effect on the metabolism of insect hosts.

Previous studies have indicated the interaction of metabolism with male reproduction. For example, the testis of Drosophila promotes carbohydrate metabolism within the adjacent intestinal portion via JAK-STAT signaling. In turn, the male intestine secretes citrate to the adjacent testes thus promotes sperm maturation[23]. Furthermore, evidence also reveals that metabolic regulators of lipids, including fatty acids act as key players during Drosophila spermatogenesis. Mutation of DmATPCL encoding the human ATP citrate lyase ortholog results in defects in spindle organization, cytokinesis, and fusome assembly during meiosis of spermatocytes in D. melanogaster[24]. This mutation significantly reduced the levels of fatty acids without affecting total amount of triglycerides. Supplementation with fatty acid-medium to the mutated flies induced significant cytokinesis defects leading to reduction of the frequency of irregular spermatids, which suggested that fatty acids metabolism is closely related to spermatogenesis[24]. In mammals, intratesticular infusion of lactate into the adult cryptorchidic rat testis could apparently improve spermatogenesis, indicating that lack of energy might be a cause of defects in male germ cell development [25]. As an essential metabolite, hormones in insects play an important role in both development and reproduction[26–28]. Our previous work showed that enhancing of juvenile hormone (JH) signaling pathway in male D. melanogaster resulted in significantly decreased male fertility[29]. These suggest that the metabolism is strongly linked with male reproduction, which may contribute to Wolbachia induced paternal defects in their insect hosts. However, it remains unclear how, or whether metabolic changes induced by Wolbachia infection in D. melanogaster impact male reproduction.

In this study, we first performed untargeted metabolomics by Liquid Chromatography-Mass Spectrometry (LC-MS) to screen the potential differential metabolites and metabolic pathways between Wolbachia-infected and uninfected 1-day-old (1d) male flies. Subsequently, targeted metabolomics was also employed to analyze host metabolism such as carbohydrate, lipid metabolism and oxidative stress in Drosophila after Wolbachia infection. Biological assays such as gene expression and overexpression were further conducted to assay potential metabolic mechanisms of paternal defects in Drosophila due to Wolbachia infection. Our results provide new insights regarding modification of male host spermatogenesis by stimulating host metabolism after Wolbachia infection. Given that the effects of Wolbachia infection on insect hosts has broad implications in the medical and agricultural fields, our findings that Wolbachia can enhance the metabolism and thus induce paternal defects in embryonic development in insect hosts may have great importance in reducing transmission of human diseases and economic burden.

All flies were reared on standard cornmeal-yeast-agar medium at 25°C with a photoperiod of 12 h:12 h LD (light:dark) and under non-crowded conditions (approx. 200 eggs per 50 ml vial of media in 150 ml conical flask). The wMel Wolbachia infected D. melanogaster (Brisbane nuclear background with introgressed wMel from YW) [30], designated as wMel, was kindly provided by Professor Scott O’Neill (Monash University, Australia). Wolbachia-uninfected Dmel T flies was subsequently generated by tetracycline treatment following established protocol [31] and confirmed to be Wolbachia-free by PCR using Wolbachia surface protein gene (wsp) primers (S1 Table).

UAS-Dbi (w-/y; pUAST-CG8627-RA; pUAST-CG8627-RA) and UAS-Mcad (w-/y; +; pUAST-CG12262-RA) fly lines were generated as our previously reported[29]. We generated Dbi and Mcad double overexpression line by a sequence of crossing, the detailed protocols were shown in S1A Fig. Dbi-RNAi (Dbi-hp, stock ID:64850) and Mcad-RNAi (Mcad-hp, stock ID:32436) fly lines were obtained from the Bloomington Center (Indiana University). The balancer flies were kindly provided by Professor Yongqing Zhang at Institute of Genetic and Developmental Biology, Chinese Academy of Sciences, Beijing, China. The actGal4 fly line (w-/y; actGal4/cyo; +) was from Professor Shan Jin at Hubei University, Wuhan, China. The bamGal4 flies (w-/y; bamGal4; +) were kindly provided by Professor Dahua Chen at Institute of Zoology, Chinese Academy of Sciences, Beijing, China. All of these transgenic fly lines were not infected by Wolbachia.

Wolbachia-infected bamGal4 flies were generated by crossing the Wolbachia-free bamGal4 males with virgin infected balancer females (w-; Cyo/Sp; Mkrs:Sb /Tm6B). The detailed crossing procedures were shown in S1B Fig.

For untargeted metabolomics analysis with LC-MS, 100 mg of 1d wMel or Dmel T males were collected and immediately frozen with liquid nitrogen, then stored at—80°C. For metabolites extraction, the frozen flies were deposited at—20°C for 30 minutes and then thawed further in a refrigerator at 4°C. About 25 mg thawed flies were mixed with 800 μl refrigerated methanol/water (1:1, v/v) in a 2 ml EP tube. After grinding for 4 min (50 Hz), the samples were kept overnight at—20°C. Then, the samples were centrifuged (30000 g, 4°C) for 20 minutes and supernatant was collected for untargeted metabolomics analysis by LC-MS. Chromatographic separation and mass spectrometry were performed using 2777 C UPLC system (Waters, UK) and Xevo G2-XS QTOF (Waters, UK). Ten microliter sample was injected for chromatographic separationthe, and the gradient program was set as follows: 0–0.1 min, 100% mobile phase solvents A (95% H2O/5% acetonitrile + 0.1% formic acid); 0.1–0.6 min, 100% A-50% A (Replaced by solvents B (95% acetonitrile/5% H2O + 0.1% formic acid)); 0.6–5 min, 50–0% A; 5–8 min, 0% A and 8–10 min, 0% A-100% A. The fractionated samples were directly added in to the mass spectrometer, and the parameters for the MS was as the method previously described by Zhou et al.[32]. In the HPLC-QTOF-MS platform, we applied MSE mode for centroid data collection and there are only MS1 (Level 1) and MS2 (Level 2). In this paper, we analyzed all the metabolomic data at MS2 (Level 2) which have the higher reliability than those at MS1 (Level 1) in MSE mode.

For targeted fatty acid methyl ester (FAME)-based GC-MS assays, 30 frozen flies were mixed with 1 ml of methanol−chloroform (2/1, v/v) containing 5 μl internal standards (50 μM methyl ester of C17:0 and C23:0) in an EP tube. Following grinding (2 minutes) and centrifuging (20187 g, 4°C) three times for 15 minutes each time, the supernatants were combined and collected for further analysis. The methylated long chain fatty acids (LCFA) were dissolved in hexane (50 μl) and analyzed on a Shimadzu 2010 Plus GC-MS spectrometer (Shimadzu Scientific Instruments, Columbia, MD) equipped with a flame ionization detector (FID) and a DB-225 capillary GC column (10 m × 0.1 mm, 0.1 μm, Agilent Technology). Helium was used as the carrier gas and the injection volume was 1 μl. Temperature of injection port and detector was set at 230°C. The programmed column temperature was as follows: the temperature of oven was increased from 55°C to 205°C at a rate of 25°C per min, kept at 205°C for 3 min and then increased to 225°C at a rate of 10°C per min. The temperature was then kept at 225°C for further 3 min. Methylated long chain fatty acids were identified by comparing retention time with a mixture of 37 fatty acid standards. The long chain fatty acids composition was subsequently quantified by comparing integrated peak areas following normalization to the internal standards.

For short chain fatty acids (SCFA)-based GC-MS assays were performed as described by Demehri et al. [33] with minor modifications. Briefly, 30 frozen flies were mixed with 400 μl of 1N HCl and 10 μl of internal standard (2,2-dimethylbutyric acid, 1mg / ml). The samples were ground 10 times (2 minutes each time with a brief break of 1 minute), and then mixed with 400 μl of ether with centrifugation (20187 g, 4°C) for 10 min. Finally, the supernatant was collected and filtered into an autosampler vial for GC-MS measurement on the GCMS-QP2010 Plus GC-MS spectrometer (Shimadzu, Japan) equipped with a CP-FFAP CB capillary GC column (25 m × 0.32 mm, 0.3 μm, Agilent Technology). Helium was used as the carrier gas and the injection volume was 1 μl. The programmed column temperature was as follows: the temperature of injection port and detector was set at 250°C, the temperature of oven was increased from 100°C to 200°C at a rate of 10°C per min. SCFAs were identified by comparing retention time with a mixture of 9 SCFAs standards.

Seven biological replicates were performed for all LC-MS or GC-MS assays.

The levels of glucose, trehalose, glycogen, and triglycerides (TGs) were determined using D-glucose (GOPOD-Format) (Cat. No. K-GLUC, Megazyme, Bray, Ireland), trehalose (Cat. No. E-TREH, Megazyme, Bray, Ireland), glucogen (Cat. No. BC0345, Solarbio, Beijing, China) and triglyceride (Cat. No. BC0625, Solarbio, Beijing, China) assay kits according to the manufacturer’s instructions, respectively. The 3-ml cuvettes and conventional spectrophotometer were used for measuring glucose and trehalose, while 96-well plates and a FLx 800 fluorescence reader (Bio-Tek, Winooski, USA) were used to assay glycogen and triglycerides. Thirty 1d wMel or Dmel T male flies were used for each biological replicate. Three biological replicates were performed for this assay.

For each biological replicate, thirty newly emerged male flies were weighed with a MS105DU semi-micro balances (METTLER TOLED, Switzerland) and then introduced into an empty vial (without food and water). After fasting for 12h and 24h, the body weight loss was recorded. Three biological replicates were performed for the body weight assay.

For each biological replicate, thirty newly emerged male flies were introduced into the feeding apparatus with capillaries (inner diameter: 0.59 mm) containing liquid food [8% (w/v) sucrose + 8% (w/v) autolyzed yeast extract] as described in our previous work[34]. The flies were fed in this apparatus for 24 h. Initial and final heights of food in the capillary were recorded at the end of the experiment. Considering the effect of evaporation on the results, we measured the relative amount of food lost in the experimental group to that in the blank group (without flies). The food uptake of flies in each experimental group was quantified by calculating the relative height reduction of the food in the capillary to the blank group [(Initial height-Final height)experiment—(Initial height—Final height)blank]. The relative food uptake in Wolbachia infected group to uninfected control group was recorded. Three biological replicates were performed for this assay.

Reactive oxygen species (ROS) level was assayed according to our previous work [35]. The levels of superoxide dismutase, glutathione (GSH) were determined by using the micro SOD assay kit (Cat. No. BC0175, Solarbio, Beijing, China), micro reduced GSH assay kit (Cat. No. BC1175, Solarbio, Beijing, China) according to the manufacturer’s instructions. The 96-well plates and the FLx 800 fluorescence reader were used in this assay. Thirty 1d wMel or Dmel T males flies were used for each biological replicate. Three biological replicates were performed for this experiment.

Quantitative reverse transcriptase PCR (qRT-PCR) was performed to investigate the relative gene expression level. For each sample, ten 1d males or 60 pair testes (for verifying the gene expression levels specifically in testes to investigate the functions of these genes in testes) of 1d males were used to extract total RNA using Trizol (Invitrogen). RNA was reverse transcribed to generate cDNA using a High Capacity cDNA Reverse Transcription kit, EasyScript (TransGen), which included DNase to remove the DNA contamination. Specific primers for tested genes were designed based on sequences from flybase database (S1 Table). QPCR was performed using a Miniopticon system (BioRad) with a TransStart Tip Green QPCR SuperMix (TransGen). The qPCR cycling program was 95°C for 2 min, followed by 40 cycles of 95°C for 10 s, 56–60°C (depending on various primers) for 20 s and 72°C for 20 s, a melting curve was constructed from 55°C to 98°C. rp49 (stably expressed in 1-day males and testes, S2A and S2B Fig) was chosen as the housekeeping gene, and relative expression of each gene was calibrated against rp49 using 2-ΔΔCT [ΔΔCT = (CT, Target—CT, rp49)experiment—(CT, Target—CT, rp49)control][36]. For qPCR experiments, we did 3 technical replicates for each biological replicate. Three biological replicates were performed for this experiment.

For each biological replicate, twelve 1d males were arranged to mate with eight 3d Wolbachia-free virgin (Dmel T) females for overnight. The following males were used in this experiment: actGal4(I-: Wolbachia-free)>UAS-Dbi, bamGal4(I-)>UAS-Dbi, actGal4(I-)>UAS-Mcad, bamGal4(I-)>UAS-Mcad, bamGal4(I+: Wolbachia-infected) >Dbi-hp, and bamGal4(I+)>Mcad-hp. In addition, bamGal4(I-)>UAS-Mcad; UAS-Dbi males were designed to mate with wMel females to assay the rescue effects. actGal4(I-)>w-, bamGal4(I-)>w-, and bamGal4(I+)>w- males were used as the corresponding control. The males were then removed and only females were maintained to lay eggs for four days. Eggs were collected and incubated at 25°C and 45–70% relative humidity for about 30 h. Egg hatch rates were determined as the proportion of hatched eggs to total eggs. Three biological replicates were performed for this test.

Thirty pairs of testes were dissected for each sample in PBS. The testes were fixed in 2.5% glutaraldehyde (0.2 M phosphate buffer, pH 7.4) at 4°C overnight, and post-fixed in 1% OsO4 for 1 h. Then, the samples were dehydrated and embedded in Araldite (EMbed 812, China) for ultrathin sections as described by Wu et al. [37]. The samples were observed and photographed using Tecnai G2 20 TWIN 200 KV transmission electron microscope (FEI, USA) and HT-7700 80 KV transmission electron microscope (Hitachi, Japan)

All values are presented as mean ± SEM. Seven biological replicates were performed for LC-MS or GC-MS assays, and three biological replicates were performed for the other tests.

The raw data of LC-MS were converted into CDF format and then imported into Progenesis QI software (version 2.0) to generate visual data, which included retention time (RT), mass-to-charge ratio (m/z) values, and peak intensity. Normalized data of LC-MS were analysed and graphed using the R language package (version 4.0.2 (2020-06-22). ape and ggplot2 were used for Partial least squares-discriminant analysis (PLS-DA); heatmap were used for cluster analysis). The freely accessible database of Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.kegg.jp) was used for identification of fragment ions and annotations of metabolic pathways. The actual spectrograms were matched to the theoretical fragmentation spectrograms to identity the potential differential metabolites, and the reference secondary database was metabolic small molecules. The significance threshold was set at P1.2 or UAS-Dbi;UAS-Mcad (S5E and S5F Fig). In this case, only one copy of UAS-Dbi could be maintained) and assayed the male fertility. The result showed that the egg hatch rate was 29.43%, lower than that in Mcad-overexpressing group (53.17%), but higher than in the bamGal4 (I-) >UAS-Dbi; UAS-Dbi male group (2.83%) (Table 2). This suggests that, relative to Mcad, Dbi may play more important role in male reproduction, or they may play distinct, non-synergistic roles in the phenotype.

Dmel T: uninfected flies; wMel: Wolbachia-infected flies. *P < 0.05.

https://doi.org/10.1371/journal.ppat.1009859.g006

https://doi.org/10.1371/journal.ppat.1009859.t002

To examine the relationship of over-expression of Dbi and Mcad with Wolbachia-induced CI, we used Wolbachia-infected females to cross with Dbi or Mcad overexpressing males and found that Wolbachia-infected females could partially rescue the embryonic lethality defects caused by over-expression of Dbi or Mcad in fly testes, since the egg hatch rates were higher than the corresponding Dmel T female groups (Table 2). We also knocked down Dbi or Mcad in the testes of Wolbachia infected flies (S5G and S5H Fig), and found either Dbi or Mcad knockdown could rescue the CI phenotype caused by Wolbachia. When they mated with Dmel T females, the egg hatch rates were significantly higher than the corresponding control group (CI cross group) (Table 2). These results suggest that Dbi and Mcad might be partially involved in Wolbachia-induced CI.

The above results have demonstrated that Wolbachia is able to enhance the energy metabolism in fly hosts. This infers that Wolbachia may speed up the energy expenditure in D. melanogaster. To verify this prediction, we compared the body weight of 30 male flies with fasting food and water for 12 h or 24 h between Wolbachia infected and uninfected males. The results revealed that after fasting food and water for 12 h, there was no apparent distinction in weight between these two Wolbachia status flies. However, Wolbachia infected flies fasting for 24 h exhibited significantly lighter body weight than the uninfected flies (Fig 7A). The levels of D-glucose and TGs were significantly reduced in Wolbachia infected flies relative to uninfected flies following 24h-fasting (Fig 7B and 7C). Wolbachia infection resulted in a significant increase in food uptake during 24 h after eclosion. The food uptake in Wolbachia-infected fly group was 1.58 times over that in uninfected groups (Fig 7D, P