Lactic acid bacteria modulate the CncC pathway to enhance resistance to β-cypermethrin in the oriental fruit fly

Abstract The gut microbiota of insects has been shown to regulate host detoxification enzymes. However, the potential regulatory mechanisms involved remain unknown. Here, we report that gut bacteria increase insecticide resistance by activating the cap “n” collar isoform-C (CncC) pathway through enzymatically generated reactive oxygen species (ROS) in Bactrocera dorsalis. We demonstrated that Enterococcus casseliflavus and Lactococcus lactis, two lactic acid-producing bacteria, increase the resistance of B. dorsalis to β-cypermethrin by regulating cytochrome P450 (P450) enzymes and α-glutathione S-transferase (GST) activities. These gut symbionts also induced the expression of CncC and muscle aponeurosis fibromatosis. BdCncC knockdown led to a decrease in resistance caused by gut bacteria. Ingestion of the ROS scavenger vitamin C in resistant strain affected the expression of BdCncC/BdKeap1/BdMafK, resulting in reduced P450 and GST activity. Furthermore, feeding with E. casseliflavus or L. lactis showed that BdNOX5 increased ROS production, and BdNOX5 knockdown affected the expression of the BdCncC/BdMafK pathway and detoxification genes. Moreover, lactic acid feeding activated the ROS-associated regulation of P450 and GST activity. Collectively, our findings indicate that symbiotic gut bacteria modulate intestinal detoxification pathways by affecting physiological biochemistry, thus providing new insights into the involvement of insect gut microbes in the development of insecticide resistance.


Introduction
The intricate relationship between insects and their gut microbiota is an extensively studied research topic, underscoring a crucial aspect of insect physiology that inf luences metabolic, developmental, and immune responses [1][2][3].Among these roles, the contribution of the gut microbiota to detoxification processes is important, providing critical insights into insect survival strategies after exposure to plant-derived toxins and synthetic pesticides [4][5][6][7][8].Additionally, various symbiotic bacteria are involved in the detoxification activity of their hosts.For example, the expression of toxin-degrading enzymes such as cytochrome P450 (P450) enzymes and α-glutathione S-transferase (GST) is induced by the gut microbiota in Amplicephalus curtulus [9], Bombyx mori [10], and Anopheles stephensi [11].Therefore, symbiotic gut bacteria may also play a crucial role in increasing insecticide resistance by indirectly modulating the activity of detoxifying enzymes in the host.However, the molecular mechanisms governing these processes remain unclear.
Insecticide resistance is a complex process that involves intricate regulatory pathways.Among these transcription factors, cap "n" collar isoform-C (CncC), which is analogous to mammalian Nrf2, has an important role.CncC and its heterodimer partner, muscle aponeurosis fibromatosis (Maf), regulate the overexpression of detoxification enzymes, transporters involved in insecticide resistance, and elements involved in the antioxidant response to xenobiotics [12,13].CncC gene was overexpressed in insecticide-resistant strains (RSs) of Drosophila melanogaster.Conversely, the silencing of CncC repressed the expression of several detoxification genes, including Cyp6g1, Cyp12d1, Cyp6a2, GstD2, Cyp6a8, and CYP6AB12 [14,15].Keach-like ECH-related protein 1 (Keap1) contains cysteine residues that are sensitive to redox reactions and can chelate CncC under low levels of reactive oxygen species (ROS) [15,16].However, under ROS-related oxidative stress, CncC dissociates from Keap1, translocates to the nucleus, and binds to Maf [12,16].Furthermore, intestinal commensal bacteria, especially Lactobacillus, stimulate the enzymatic production of ROS in epithelial cells to promote cell proliferation [17,18], accelerate recovery from injury [19], and alter epithelial NF-κB signaling [20].This activation of the Nrf2 pathway has salutary effects against exogenous insults to the intestinal epithelium [21].
Bactrocera dorsalis is a prime example of how these mechanisms are important in agricultural contexts [22,23].Given that B. dorsalis is a major pest that adversely affects diverse fruit crops, elucidating the interactions of B. dorsalis with its gut microbiota, particularly how these interactions regulate detoxification pathways and contribute to insecticide resistance, is important for developing effective pest management strategies [6,[24][25][26][27][28][29][30][31].Previous studies have demonstrated that changes in the microbiota composition, such as increases in Acetobacter spp.and Lactobacillus, correlate with insecticide resistance, highlighting the role of gut bacteria in modulating host defense mechanisms against these chemical stressors [32].Nevertheless, the intricacies of how gut commensal bacteria affect the CncC signaling pathway and its downstream effects on detoxification and resistance remain to be fully elucidated.
In this study, we investigated the molecular mechanisms underlying how the gut microbiota regulates insecticide resistance in B. dorsalis, focusing on the activation of the CncC pathway by ROS enzymatically produced by gut bacteria.We specifically explored how Enterococcus casselif lavus and Lactococcus lactis, notable lactic acid (LA) producers, inf luence β-cypermethrin resistance, impact P450 and GST activities, and modulate CncC pathway gene expression.Through an integrative approach that includes BdCncC knockdown and ROS scavenging, we aimed to unravel the complex interactions governing these processes, offering new insights into the dynamic interplay between the gut microbiota and host detoxification mechanisms.This research not only advances our understanding of the biochemical pathways essential for insecticide resistance but also paves the way for innovative pest management strategies that exploit these microbiota-host interactions.

Insect strains and rearing
B. dorsalis susceptible strain (SS) was collected from a field in Qingyuan, Guangdong Province, China, in August 2003 and reared indoors for more than 100 generations without exposure to pesticides.RS of B. dorsalis was obtained by selection after adult exposure to a β-cypermethrin-treated surface over the course of 40 generations.
B. dorsalis flies were reared at 27 ± 1 • C with 75 ± 1% relative humidity under a 14:10-h light:dark photoperiod cycle.Hatched larvae were maintained on an artificial diet mainly consisting of bananas before pupation.Pupae were kept in a plastic bucket with wet sand until adults emerged.Adults were fed an artificial diet consisting of yeast extract and dry sugar mixed at a 1:1 ratio (w/w) [33] and housed in transparent plastic cages.

Bioassay and synergism experiment
The toxicity of β-cypermethrin to different strains was determined with the aid of an 8% sucrose solution with gradient concentrations of the insecticide.Twenty f lies of each strain were fed each concentration.Mortality was recorded after 1 day.Three replicates were prepared for each concentration.Five concentrations of insecticide were used to establish the log-probit lines.
With the abovementioned bioassay, a synergistic experiment was conducted with the P450 inhibitor piperonyl butoxide (PBO) and the GST inhibitor diethyl maleate (DEM).PBO (BP1176, Sigma-Aldrich) and DEM (D97703, Sigma-Aldrich) were dissolved in acetone to final concentrations of 10 mM (PBO) and 100 mg/l (DEM), respectively.A 2 μl solution of the inhibitor was topically applied to the pronotum, with acetone as the control.After 2 h, the fruit f lies treated with PBO or DEM were transferred to a sugar water solution containing different doses of β-cypermethrin for toxicological experiments.The toxicity ratio was calculated by dividing the median lethal concentration (LC 50 ) value of β-cypermethrin alone by the LC 50 of β-cypermethrin with PBO or DEM [34].

RNA preparation and transcriptomic sequencing
Total RNA was extracted from the guts of eight 10-day-old adult SS and RS f lies (females:males = 1:1) as one biological replicate by using TRIzol reagent (Life Technologies, Carlsbad, CA) according to the manufacturer's protocol.Three biological replicates were analyzed.The integrity of the RNA was verified by 1% RNase-free agarose gel electrophoresis, and the RNA concentration was determined by measuring the absorbance of the sample at 260 nm using a spectrophotometer (Thermo NanoDrop 2000c; Santa Clara, CA).After the total RNA was extracted, eukaryotic mRNA was enriched with oligo(dT) beads.Then, the enriched mRNA was fragmented into short fragments using fragmentation buffer and reverse transcribed to cDNA by using the NE Next Ultra RNA Library Prep Kit for Illumina (NEB #7530, New England Biolabs, Ipswich, MA).The purified double-stranded cDNA fragments were end repaired, A bases were added, and the fragments were ligated to Illumina sequencing adapters.The ligation reaction mixture was purified with AMPure XP Beads (1.0X).Ligated fragments were subjected to size selection by agarose gel electrophoresis and polymerase chain reaction (PCR) amplification.The resulting complementary DNA (cDNA) library was sequenced on NovaSeq 6000 (Illumina).The raw reads were deposited into the National Center for Biotechnology Information (NCBI) Sequence Read Archive (SRA) database (PRJNA1027602).

cDNA synthesis and reverse transcription quantitative PCR
RNA was extracted using TRIzol reagent, and the RNA integrity was determined via electrophoresis on a formaldehyde agarose gel.RNA aliquots (1 μg) were reverse transcribed to cDNA using a PrimeScript RT reagent Kit with gDNA Eraser (TaKaRa Bio, Otsu, Japan) according to the manufacturer's instructions.The biosynthesized cDNA was used as a template for reverse transcription quantitative PCR (RT-qPCR), which was conducted on a CFX Connect Real-Time PCR System (Bio-Rad, Hercules, CA) with TB Green Premix Ex Taq II (Tli RNase H Plus) (TaKaRa Bio, Otsu, Japan) and primers designed using the NCBI database (Table S1).The thermal cycling conditions were as follows: 95 • C for 30 s, followed by 40 cycles of 95 • C for 5 s and 60 • C for 34 s.The transcript levels of different genes were quantified using the 2 − CT method.α-Tubulin and RPL32 [35] were used as reference genes for gene expression analysis in B. dorsalis due to their stable expression.

P450 and GST activity assays
The p-nitrophenyl ether-O-demethylase activity of P450 in the gut of B. dorsalis was measured by using p-nitroanisole (p-NA) as a substrate, as previously described [15].The midguts of 20 adult f lies were dissected in ice-cold Phosphate Buffer Saline (PBS) and homogenized with 200 μl of 0.1 M PBS (pH = 7.5) containing 1 mM ethylenediaminetetraacetic acid, 1 mM phenylmethanesulfonyl f luoride, 1 mM phenylthiourea, 1 mM dithiothreitol, and 10% glycerol.The supernatant was then centrifuged at 4 • C for 20 min at 12000 rpm to prepare an enzyme stock solution.Twenty microliters of the supernatant were added to PBS containing 0.2 μM p-NA and 9.6 mM nicotinamide adenine dinucleotide phosphate (NADPH) for the enzyme reaction.After incubation at 30 • C for 30 min, the absorbance at a wavelength of 405 nm was measured using a microplate reader.The P450 activity in the supernatant was calculated by creating a standard curve using the A405 values for different concentrations of p-nitrophenol.The protein concentration in the enzyme supernatant was measured using a Bovine Serum Albumin (BSA) assay kit (Sangon Biotech, Shanghai, China).
A GST activity assay kit (Sangon Biotech, Shanghai, China) was used to measure GST activity in the gut of B. dorsalis.GST catalyzes the binding of glutathione to 1-chloro-2,4-dinitrobenzene, and the binding product has an optical absorption peak at 340 nm.GST activity can be calculated by measuring the rate of increase in the absorbance at 340 nm.We dissected the midguts of 20 adult f lies in ice-cold PBS and then homogenized them with 200 μl of Reagent I (available in the reagent kit).Next, the supernatant was obtained by centrifugation at 12000 rpm at 4 • C for 20 min.The enzymatic reaction was carried out according to the kit manufacturer's instructions, and then, the GST activity of the supernatant was calculated from the protein concentration.

Antibiotic treatment of B. dorsalis and oral ingestion of bacteria
For the RS + axenic f lies, 5-day-old RS B. dorsalis adults were fed 8% sugar water supplemented with 50 μg/ml tetracycline, 100 μg/ml penicillin-streptomycin, 150 μg/ml gentamycin, and 150 μg/ml rifampicin for 4 days.Then, the f lies were treated with sterile water for 1 day to metabolize the antibiotics.Before dissection, the f lies were sterilized by soaking in 70% alcohol for 3 min and then washed three times with sterile PBS.We dissected the intestines in sterile PBS.PCR using universal 16S rDNA gene primers (Table S1) and colony-forming unit (CFU) assays was used to verify the bactericidal efficacy of antibiotics.Then, the guts were dissected to extract DNA and RNA, and P450 and GST activities were measured.
For the RS + axenic+bacteria group, E. casselif lavus, L. lactis, Klebsiella pneumoniae, and Enterobacter hormaechei were isolated from the gut of the RS f lies.The bacteria were grown in de Man-Rogosa-Sharpe medium at 37 • C and 200 rpm for 16 h.The bacterial culture was then pelleted by centrifugation (3000 × g, 15 min), washed twice in sterile PBS, and resuspended in 8% sterile sucrose solution.The RS axenic f lies were fed a bacterial suspension (OD600 = 5) for 3 days.The SS f lies were fed different concentrations of bacterial suspension (OD600 = 0.5, 5, and 10) for 3 days.Then, the guts were dissected to extract DNA and RNA, and the P450 and GST activities were measured.

Gut bacterial DNA sample preparation and 16S rRNA gene amplicon sequencing
The guts of 16 SS or RS f lies were dissected under sterile conditions, and DNA was extracted using an E.Z.N.A. Soil DNA kit (Omega Bio-Tek, Inc., Norcross, GA) according to the manufacturer's instructions.Five biological replicates per strain were analyzed.The 16S rRNA gene V3-V4 region was amplified by the primers 341-F (5 -CCTACGGGNGGCWGCAG-3 ) and 806-R (5 -GGACT ACHVGGGTATCTAAT-3 ).The PCR products were assessed using 2% agarose gel electrophoresis, purified using an AxyPrep DNA gel extraction kit (Axygen Biosciences, Union City, CA), and quantified using an ABI StepOnePlus Real-Time PCR System (Life Technologies, Foster City, USA).The purified amplicons were pooled in equimolar amounts and paired-end sequenced (PE250) on NovaSeq 6000 (Illumina) according to the standard protocols.The raw reads were deposited into the NCBI SRA database (PRJNA1027561).

Quantification of specific gut bacteria by quantitative PCR
The loads of E. casselif lavus and L. lactis in the gut were quantified by real-time PCR using specific primers and normalized to realtime PCR data for the host β-actin gene [36].The primer pairs used in the quantitative PCR analysis are shown in Table S1.Genomic DNA from the intestinal tracts of B. dorsalis was extracted using an EZNA Soil DNA Kit (Omega Bio-Tek, Inc., Norcross, GA) according to the manufacturer's instructions.

Prediction of the binding site for the transcription factors CncC-Maf
The promoter sequence information for detoxification genes (P450s and GSTs) downregulated in the RS f lies after elimination of gut bacteria was obtained from the B. dorsalis genome.A 2000 bp fragment upstream of the transcription start site of each gene was selected as a potential promoter region for prediction of the CncC-Maf binding site by JASPAR [37].

Dual-luciferase reporter gene assay
The open reading frame sequence of BdCncC was amplified using cDNA from B. dorsalis as a template with the primers specified in Table S1.The BdCncC overexpression vector was constructed with the BsmBI and Esp3I restriction enzyme sites on the pcDNA3.1 plasmid.Additionally, the promoter regions of the P450 and GST genes were amplified from B. dorsalis DNA using primers from Table S1 and further utilized to construct promoter vectors on the pGL3-basic plasmid with SacI and HindIII restriction enzyme sites.The internal reference plasmid used was pGL4-TK.
HEK-293 T cells were cultured, and 1 day prior to transfection, the cells were plated in a 48-well plate.Transfection was carried out using Lipofectamine 3000 (L3000015, Invitrogen) reagent; cotransfection of pcDNA3.1(+)BdCncC,pGL3-basic-promoter, and pGL4-TK plasmids at a ratio of 1:1 for the promoter plasmid to the BdCncC overexpression plasmid was performed; and transfection of pGL4-TK, which represented one-tenth of the total system, was performed.Control transfections included the pcDNA3.1,pGL3basic-promoter, and pGL4-TK plasmids.After 48 h of transfection, the Dual-Luciferase Reporter Assay System (E1910, Promega) was used to measure firef ly and renilla luciferase activities.Cell lysis, reagent addition, and readings were conducted according to the manufacturer's instructions.The assay included three biological replicates, and values were recorded for the subsequent analysis.

Vitamin C treatment
Vitamin C (VC) is an antioxidant, and in previous studies, feeding VC to B. dorsalis effectively reduced ROS levels in the gut [36].Ten-day-old adult f lies of RS B. dorsalis were fed 8% sugar water containing 100 mg/ml VC (PHR1008, Sigma-Aldrich) for 24 h.Flies fed 8% sugar water were used as the controls.Then, the dissected guts were subjected to P450 and GST activity analysis, ROS activity assays, and RT-qPCR.

Analysis of ROS activity in vivo
We used a hydrogen peroxide assay kit (Beyotime Biotechnology, Shanghai, China) to measure the production of H 2 O 2 and determine the activity of ROS with the ROS f luorescent probe dihydroethidium (DHE; D7008, Sigma-Aldrich).The midguts were then dissected in cold PBS containing 2 mg/ml of the catalase inhibitor 3-amino-1,2,4-triazole (A8056, Sigma-Aldrich).Twenty dissected guts (females:males = 1:1) were combined into a sample and homogenized in lysis buffer to measure the concentration of H 2 O 2 [38].Other guts were immediately incubated in 2 μM DHE in PBS to assess ROS activity [38].

Measurement of lactate content
The amount of lactate in dissected guts (n = 20, same number of males and females) was determined using a Lactate Assay Kit (Sangon Biotech, Shanghai, China) following the manufacturer's instructions and was normalized to the total protein content.

Measurement of pyruvate content
A pyruvate (PA) content detection kit (Sangon Biotech, Shanghai, China) was used to measure the PA content in the gut of B. dorsalis.PA plays an important pivotal role in linking glucose, fatty acids, and amino acids through acetyl-CoA metabolism.Acetate reacts with 2,4-dinitrophenylhydrazine to produce PA-2,4-dinitrophenylhydrazone, which appears cherry red in alkaline solution.PA in a sample can be detected at a wavelength of 520 nm.Twenty intestines were placed in 200 μl of extraction buffer (provided by the reagent kit) for ice-cold homogenization.The mixture was allowed to stand for 30 min, followed by centrifugation at 8000 × g for 10 min at room temperature.The supernatant was then collected for subsequent analysis following the manufacturer's instructions.

LA feeding
SS and axenic RS adult f lies (with the same number of males and females) were fed DL-LA (69 785, Sigma-Aldrich).The axenic RS f lies were fed 50 μM DL-LA for 3 days.The SS f lies were fed different concentrations of DL-LA (5 μM, 50 μM, 500 μM, 5 mM, or 50 mM) for 2 days.Then, the dissected guts were subjected to P450 and GST activity assays, ROS activity assays and RT-qPCR.

Statistical analysis
The mortality data of the f lies were corrected using Abbott's formula [39], and the LC 50 values (50% killing concentration of f lies) and 95% fiducial limits of the LC 50 for each strain were determined via probit analysis using Statistical Package for the Social Sciences 27.0 software.Two LC 50 values were considered significantly different if their 95% fiducial limits did not overlap [40].Other statistical analyses were performed using Prism 7 (GraphPad Software).Two-tailed independent t-tests were used for unpaired comparisons between two groups of data.For comparisons of three or more sets of data, one-way analysis of variance (ANOVA) was performed, followed by Tukey's multiple comparison test.A value of P < .05 was considered indicative of statistical significance.

Gut symbiotic bacteria regulate resistance to β-cypermethrin in B. dorsalis
To confirm the resistance level of the B. dorsalis SS and RS f lies, we examined the toxicity of ingested β-cypermethrin.The LC 50 values of β-cypermethrin for the SS and RS f lies were 2.0 mg/l and 154.6 mg/l, respectively (Fig. 1A).Moreover, the results of both intestinal transcriptomics and real-time f luorescence quantitative PCR showed that the expression of several cytochrome P450 genes and GST genes was significantly greater in the RS f lies than in the SS f lies (Fig. S1A-C).The corresponding P450 enzyme activity and GST enzyme activity of the RS f lies were also significantly greater than those of the SS f lies (Fig. S1D and E).The P450 inhibitor PBO and the GST inhibitor DEM reduced the LC 50 value of β-cypermethrin in the RS f lies from 151.8 mg/l to 67.9 mg/l (PBO) and 77.5 mg/l (DEM), respectively, with toxicity ratios of 2.2 and 2.0, respectively (Fig. S2).After the elimination of intestinal symbiotic bacteria ( Fig. S3A), the LC 50 value of β-cypermethrin for the RS f lies decreased from 145.8 mg/l to 83.1 mg/l (Fig. 1B).After antibiotic treatment, the expression of the P450 and GST genes in the gut was significantly reduced in the RS f lies, decreasing by 7.7%-70.8%(Fig. 1C).However, no difference was found in the expression of other detoxification genes (Fig. S3B).Moreover, the intestinal P450 activity and GST activity decreased by 60.4% and 63.2%, respectively, in the RS f lies after antibiotic treatment (Fig. 1D and E).

Differences in the gut microbiota community structure between the RS and SS flies
To explore the role of the gut microbiota in regulating insecticide resistance, we conducted 16S rDNA sequencing analysis of the gut microbiota of the SS and RS f lies.We obtained a total of 12 sets of experimental data, and PCA based on weighted UniFrac distances revealed differences in the gut bacterial community structure between the SS and RS f lies (Fig. S4A).The Shannon index, Pielou index, and Simpson index of the gut microbiota in the RS f lies were significantly different from those in the SS f lies (Fig. S4B-D).
In a comparison of the composition and structure of gut bacterial communities between the RS and SS f lies, we observed significant changes in the abundances of multiple gut bacteria in the RS f lies.Specifically, at the genus level, the abundances of the intestinal probiotics Enterococcus and Lactococcus were significantly increased by 7.0-fold and 4.4-fold, respectively, whereas those of Vagococus, Commesalibacter, Stenotrophomonas, Escherichia Shigella, and Providencia were significantly reduced (Fig. 2A).Furthermore, at the species level, we also found a significant increase in the abundances of E. casselif lavus (OTU000005) and L. lactis (OTU000004) (Fig. 2B and C and Fig. S4E).Moreover, the f luorescence quantitative PCR results of the 16S rRNA gene showed that the relative abundances of E. casselif lavus and L. lactis in the RS f lies were significantly greater than those in the SS f lies (Fig. 2D and E).Therefore, we speculated that these two bacteria play crucial roles in the development of βcypermethrin resistance.

L. lactis and E. casseliflavus enhance resistance to β-cypermethrin in B. dorsalis
We further isolated E. casselif lavus and L. lactis from the gut of the RS f lies and orally fed them to the axenic RS f lies (Fig. S5A and B).Toxicity tests revealed that both E. casselif lavus and L. lactis reversed the antibiotic-induced decreases in β-cypermethrin resistance, with the LC 50 values increasing from 79.8 mg/l to 106.1 mg/l (E.casselif lavus) and 102.2 mg/l (L.lactis) (Fig. 3A).However, after oral administration of the intestinal bacteria K. pneumoniae and E. hormaechei (Fig. S6A), the LC 50 of β-cypermethrin did not significantly increase in the axenic RS f lies (Fig. S6B).The P450 and GST genes were significantly activated by E. casselif lavus and L. lactis (Fig. 3B).The intestinal P450 and GST activities were significantly greater in these f lies than in those treated with antibiotics (Fig. 3C and D).To determine whether gut microbes exert a similar impact on SS f lies, we fed SS f lies with 5 OD E. casselif lavus or L. lactis (Fig. S5C and D).Similar to the observations in the RS f lies, E. casselif lavus and L. lactis significantly elevated the P450 and GST activities (Fig. 3E and F).The LC 50 of βcypermethrin for the SS f lies fed E. casselif lavus and L. lactis increased from 2.3 mg/l to 7.1 mg/l and 8.1 mg/l, respectively (Fig. 3G).
Given that E. casselif lavus and L. lactis similarly increase insecticide resistance, we used L. lactis for gradient feeding experiments in SS f lies to investigate the role of the abundance of symbiotic bacteria in the gut microbiota in the development of insecticide resistance.Feeding L. lactis at 0.5 OD and 5 OD induced the expression of intestinal P450 and GST genes (Fig. 3H), confirming that increasing the abundance of L. lactis in the intestine of susceptible f lies promoted insecticide resistance.However, feeding at 10 OD did not induce the expression of detoxification genes (Fig. 3H).Collectively, our data suggest that increases in the abundances of E. casselif lavus and L. lactis in the f ly guts increase their resistance to β-cypermethrin by stimulating P450 and GST activity.

The gut microbiota regulates resistance to β-cypermethrin by activating the CncC pathway
Our previous studies demonstrated that BdCncC belongs to the bZIP superfamily based on its structural domains and is highly expressed in the intestine [33].Members of the Cnc-bZIP transcription factor superfamily can regulate genes involved in enhancing insecticide metabolic resistance (Fig. 4A).In this context, we evaluated the expression of genes in the CncC pathway in SS and RS f lies.We observed significant upregulation of BdCncC and BdMafK expression in the CncC pathway in the intestine of the RS f lies compared to the SS f lies (Fig. 4B).Moreover, the expression of BdKeap1 was significantly downregulated (Fig. 4B).Furthermore, we analyzed the pretranscriptional 2000 bp fragments of the P450 and GST genes in the gut as potential promoter regions for predicting the CncC-Maf binding site of transcription factors.The prediction results revealed the presence of the CncC-Maf binding site in all resistancerelated gene promoters.Specifically, the binding site was located within the following ranges: LOC105222599-CYP6g1 from positions −1721 to −1708, LOC105226935-CYP6v1 from −1552 to −1538, LOC105226035-CYP6d4 from −500 to −486, LOC105233823-CYP6g1 from −1771 to −1757, and LOC105225813-GSTD1 from −1231 to −1217 (Fig. S7A).Using a dual-luciferase reporter gene assay, we further confirmed that overexpression of BdCncC significantly increased the promoter activity of the target P450 and GST genes by 1.5-3.7-foldcompared to that of the pcDNA3.1 control (Fig. S7B).
The RT-qPCR results demonstrated a significant reduction in the expression levels of BdCncC (Fig. 4C) and BdMafK (Fig. 4D) upon elimination of intestinal bacteria in the RS f lies, accompanied by a notable increase in the expression level of BdKeap1 (Fig. 4E).However, supplementation with the gut symbiotic bacteria L. lactis or E. casselif lavus effectively counteracted the downregulation of BdCncC and BdMafK expression resulting from elimination of the gut bacteria (Fig. 4C and D).Additionally, this treatment inhibited the increase in BdKeap1 expression levels (Fig. 4E).
To assess the functionality of BdCncC, we silenced the expression of the BdCncC gene in traditionally fed RS f lies with 71.3% interference efficiency at 72 h (Fig. 4F).Compared to those in the f lies treated with RS-GFP-RNAi, the expression levels of the P450 and GST genes harboring CncC-Maf binding sites were significantly decreased in the intestines of the f lies subjected to RS-BdCcnC-RNAi (Fig. 4H).Correspondingly, there was a significant decrease in the enzyme activities of both P450 and GST (Fig. 4J and L).Additionally, in a separate interference experiment in which BdCncC was used as an off-target control, similar alterations in the levels of the detoxification enzymes P450 and GST were identified at 72 h (Fig. S8).
To verify that intestinal bacteria-mediated insecticide resistance is dependent on the expression of the BdCncC gene, we injected ds-BdCncC into RS f lies that had been fed L. lactis following antibiotic treatment.After BdCncC interference, the enzyme activity and expression levels of the associated P450 and GST genes were significantly lower than those in the ds-GFP control group (Fig. 4G, I, K, and M).This observed pattern was consistent with the results of the toxicity assay (Fig. 4N and O).In conclusion, these findings collectively suggest that L. lactis increases the resistance of B. dorsalis to β-cypermethrin by regulating BdCncC, ultimately activating the expression of P450 and GST.

Gut microbiota-mediated ROS activate the CncC pathway to increase resistance to β-cypermethrin
ROS induce the heterodimeric segregation of CncC from Keap1, leading to the translocation of CncC to the nucleus where it heterodimerizes with Maf.This heterodimer binds to and activates the expression of a CncC/Maf response element located in the promoter of the ectopically responsive gene [13] (Fig. 4A).In our study, the H 2 O 2 content and ROS activity were significantly greater in the gut of the RS f lies than in that of the SS f lies (Fig. 5A and B).After elimination of the intestinal bacteria, the H 2 O 2 content in the RS intestine decreased significantly (Fig. 5C), resulting in a corresponding attenuation of ROS activity (Fig. 5D).However, supplementation with L. lactis or E. casselif lavus individually increased the ROS levels (Fig. 5C and D).Additionally, feeding each of the two bacteria separately to SS f lies also stimulated intestinal ROS activity (Fig. 5E and F).To determine whether the high level of intestinal ROS activity in the gut of the RS f lies activated the CncC pathway, we fed RS f lies VC, an ROS scavenger.After feeding with VC, the intestinal H 2 O 2 content of the RS f lies decreased, and the ROS activity was attenuated compared with that in the f lies fed sugar water ( Fig. 5G and H).This reduction in intestinal ROS activity resulted in a significant downregulation of the ROS-regulated transcription factors BdCncC and BdMafK (Fig. 5I and J) and a corresponding upregulation of Bdkeap1 expression (Fig. 5K).Moreover, the P450 and GST genes, which are transcriptionally regulated by BdCncC/BdMafK, exhibited significant downregulation (Fig. 5L).Consequently, the corresponding ROS scavenging-related P450 and GST enzyme activities were reduced by 76.6% and 47.6%, respectively (Fig. 5M and N).Therefore, we concluded that the P450 activity in the gut of (J) the RS and (K) RS + axenic+ LL strain f lies after BdCncC interference for 72 h; GST activity in the gut of (L) the RS and (M) RS + axenic+LL strain f lies after BdCncC interference for 72 h; effect of BdCncC knockdown on the response of (N) RS f lies and (O) RS + axenic+LL flies to β-cypermethrin; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; Conv, conventionally colonized f lies; Axenic, axenic f lies; LC 50 values were considered significantly different if their fiducial limits did not overlap; Student's t-test was performed for B-M; error bars indicate ± s.e.m.; * * * P < .001,* * P < .01,* P < .05;all results were obtained from at least two independent experiments.(O) VC scavenges ROS in the gut and negatively modulates CncC pathway regulation; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of a 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; RS + VC: RS f lies were fed 8% sugar water containing 100 mg/ml VC for 24 h; SS + LL/EC strain: SS f lies continuously fed 5 OD L. lactis (E.casselif lavus) for 3 days; the H 2 O 2 levels in A and E were normalized to those in the SS controls, those in C were normalized to those in the RS + axenic f lies, and those in G were normalized to those in the RS f lies; the scale bars in B, D, F, and H represent 1000 μm; Student's t-test was performed for A, C, E, G, and I-N; error bars indicate ±s.e.m. * * * P < .001,* * P < .01;all results were obtained from at least two independent experiments.gut microbiota of B. dorsalis activates the BdCncC pathway by producing ROS (Fig. 5O), which in turn regulate insecticide pesticide resistance in B. dorsalis.

BdNOX5 is required for insecticide resistance
The production of ROS induced by the gut microbiota occurs mainly through the NADPH family enzymes DUOX and NOX.We measured the expression levels of BdDUOX and BdNOX5 in the guts of SS and RS f lies.Quantitative PCR revealed that the expression level of BdNOX5 in the gut of the RS f lies was significantly greater than that in the gut of the SS f lies (Fig. 6A), whereas there was no difference in BdDUOX expression (Fig. S9A).Additionally, upon elimination of gut bacteria from the RS f lies, the expression level of BdNOX5 significantly decreased, and single bacterial supplementation with L. lactis or E. casselif lavus restored the expression of BdNOX5 (Fig. 6B).Conversely, K. pneumoniae and E. hormaechei, which do not increase insecticide resistance in B. dorsalis, failed to activate highly expressed BdNOX5 (Fig. S9B).Similar to those of the P450 and GST genes, the expression level of BdNOX5 significantly increased following continuous oral administration of L. lactis at 0.5 OD and 5 OD for 3 days in the SS f lies (Fig. S9C).
To validate the role of BdNOX5 in the control of insecticide resistance in RS flies, we silenced BdNOX5 expression by dsRNA injection.The interference efficiency of the ds-BdNOX5 injection at 72 h was 57.1% in the traditionally fed RS and 27.3% in the axenic RS f lies (Fig. 6C).After BdNOX5 gene expression was significantly downregulated, the intestinal H 2 O 2 content and ROS activity of the RS f lies injected with ds-BdNOX5 were significantly reduced (Fig. 6D and E).With ds-GFP as a control, the expression levels of the transcription factors BdCncC and BdMafK, which are regulated by ROS, in the intestines of both RS and axenic RS f lies injected with ds-BdNOX5 were significantly reduced, and the expression levels of BdKeap1 were significantly increased (Fig. 6F and G).The expression of P450-and GST-related genes restricted by BdC-ncC/MafK transcriptional regulation also decreased significantly (Fig. 6H and I).Moreover, the P450 and GST enzyme activities decreased significantly (Fig. 6J-M).Furthermore, feeding L. lactis after injecting ds-BdNOX5 into axenic f lies did not reverse the reduction in H 2 O 2 levels or ROS activity (Fig. 6D and E).Additionally, the low expression of BdNOX5 did not result in activation of the BdCncC pathway, as indicated by the low expression levels observed (Fig. 6G).This inhibition, in turn, affected the high expression of detoxification genes (Fig. 6I) and the activities of the P450 and GST enzymes (Fig. 6K and M).In another interference experiment involving off-target control with BdNOX5-RNAi, the levels of the detoxification enzymes P450 and GST in the samples at 72 h also exhibited similar changes (Fig. S10).The data indicate that the gut microbiota regulates insecticide resistance by activating BdNOX5-ROS, which subsequently controls the expression and enzymatic activity of BdCncC-associated P450 and GST genes.
In our study, the intestinal lactate content in the RS group was significantly greater than that in the SS group (Fig. 7B).After intestinal bacteria were eliminated with antibiotics, the lactate content also decreased significantly, and the lactate content in the intestine increased after single bacterial supplementation with L. lactis or E. casselif lavus (Fig. 7B).We screened two LDH genes in the genome of B. dorsalis: LOC105231405-LDH and LOC105231406-LDH.The quantitative PCR results showed no difference in the expression of LOC105231406-LDH in the guts of the SS and RS f lies (Fig. S11).However, compared with that in the SS intestine, LOC105231405-LDH was significantly overexpressed in the RS intestine but was downregulated after the elimination of intestinal bacteria (Fig. 7C).Single bacterial supplementation with L. lactis or E. casselif lavus restored the high expression of LOC105231405-LDH (Fig. 7C).Moreover, the increase in lactate levels and the high expression of LOC105231405-LDH contributed to the increase in PA content (Fig. 7D).Furthermore, we effectively inhibited LOC105231405-LDH at 48 h and 72 h after injection of ds-BdLDH-1 and ds-BdLDH-2 (Fig. S12A).The interference efficiency of ds-BdLDH-2 at 72 h reached 50.1% (Fig. 7E).Additionally, the knockdown of LOC105231405-LDH significantly reduced the levels of PA in the gut of the RS f lies (Fig. 7F), the ROS content (Fig. 7G and H), and the enzyme activities of P450 (Fig. 7I and Fig. S12B) and GST (Fig. 7J and Fig. S12C).
After continuously feeding SS f lies with different concentrations of LA for 72 h, we found that both 5 μM and 50 μM LA increased the expression of LOC105231405-LDH.However, 500 μM and 50 mM LA restored the expression of LOC105231405-LDH (Fig. S13A).Moreover, feeding 50 μM LA to axenic RS f lies led to elevated LOC105231405-LDH expression at 48 h (Fig. 7K).LA feeding after antibiotic treatment activated BdNOX5 expression (Fig. 7L) and ROS activity (Fig. 7M and N), replacing L. lactis and E. casselif lavus.Continuous LA feeding in the SS f lies also activated BdNOX5 expression (Fig. S13B) and ROS activity (Fig. S13C).LA-induced ROS activated the BdCncC pathway (Fig. 7O), regulating P450 and GST expression (Fig. 7P) and increasing their enzyme activities (Fig. 7Q and R).Higher P450 and GST enzyme activities were also observed in the SS LA feeding experiment (Fig. S13D and E).These findings identify a central role for LOC105231405-LDH in ROS generation by L. lactis and E. casselif lavus, likely by oxidizing lactate and generating PA and NADH, a substrate for BdNOX5.

Discussion
Numerous studies have shown that symbiotic bacteria of the insect gut regulate the activity of detoxifying enzymes involved in metabolism of host food sources, thereby indirectly increasing host resistance [8][9][10][11].Although symbiotic bacteria and insect resistance are linked, the physiological and biochemical mechanisms underlying this relationship have not been systematically elucidated.In our study, the gene expression and enzyme activity of P450 and GST in the gut were significantly greater in the RS f lies that were chronically exposed to βcypermethrin, which is a broad-spectrum pyrethroid insecticide, than in the SS f lies.We demonstrated that the LAB L. lactis and E. casselif lavus were enriched in the gut of the RS f lies.Further experiments revealed that ROS generated by LAB activated the BdCncC/BdKeap1/BdMafK pathway, which modulates downstream P450 and GST activities and increases insecticide resistance in B. dorsalis.Therefore, our current findings elucidate the pathway by which the gut microbiota indirectly regulates host resistance to insecticides.silencing in (F) the RS and (G) RS + axenic strain f lies; the mRNA levels of the P450 and GST genes in the guts of the BdNOX5 knockdown (H) RS and (I) RS + axenic f lies; P450 activity in the gut of (J) the RS and (K) RS + axenic f lies after BdNOX5 knockdown; GST activity in the gut of (L) the RS and (M) RS + axenic f lies after BdNOX5 knockdown; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of a 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; dsBdNOX5 + LL: after injection of 2 μg dsRNA into the RS + axenic strain, continuous feeding of 5 OD L. lactis was performed for 72 h; Conv, conventionally colonized f lies; Axenic, axenic f lies; Student's t-test was performed for A-D and F-M; error bars indicate ±s.e.m. * * * P < .001,* * P < .01,* P < .05;all results were obtained from at least two independent experiments.
Mutually beneficial symbiotic insect-bacteria relationships help symbiotic systems adapt to pesticide screening pressures and allow continued coevolution [ 49].Thus, the increased resistance to insecticides in our RS f lies may be attributed to coevolution with LAB.Enterococcus has been shown to increase insecticide tolerance in Spodoptera litura [50], B. mori [51], Plutella xylostella [52], and Spodoptera frugiperda [53].Genomic analysis of strains of Enterococcus isolated from S. frugiperda larvae revealed that they possessed exogenous degradation enzyme mechanisms [53].When honeybees (Apis mellifera) and D. melanogaster are exposed to insecticides, the relative abundance of LAB typically increases [32,[54][55][56].In this study, the activities of P450 and GST were inhibited after feeding with mixed antibiotic solutions to clear the intestinal commensal bacteria in B. dorsalis.In contrast, oral administration of L. lactis or E. casselif lavus rescued the activities of P450 and GST and increased the tolerance of the flies to β-cypermethrin.We demonstrated that detoxification does not occur through direct encoding of hydrolases but rather through modulation of the host's physiological and biochemical environment, thereby indirectly increasing the detoxification capacity of the host via intracellular signaling pathways.This regulatory activity is dependent on ROS levels in the gut of B. dorsalis.We found here that the induction of ROS by LAB required BdNOX5 but not BdDUOX.However, DUOX is mainly activated by uracil secreted by pathogenic bacteria and some symbiotic gut bacteria [57].We identified L. lactis-and E. casselif lavus-derived LA as an essential metabolite triggering NOX5-dependent ROS production.Similar results have been obtained in Drosophila, in which LA secreted by L. plantarum can drive ROS production via NOX [42].Moreover, the decrease in the abundance of Vagococus, Commesalibacter, Stenotrophomonas, E. Shigella, and Providencia in the gut of the RS group may be attributed to the increase in ROS, which are important immune factors in the gut [58].
Most previous studies have concluded that insect resistance mechanisms are mainly categorized into metabolic resistance and target resistance [59,60].The detoxification enzymes most closely associated with metabolic resistance include carboxylesterases [61], P450s [62,63], GSTs [64], and aldehyde oxidase [65,66].The transcription factor CncC is a central regulatory factor in the response of insects to xenobiotics, and its actions include induction of the expression of the detoxification genes P450 and GST [12].ROS interact with key signaling molecules to initiate signaling in various cellular processes and play an important role in cellular signaling cascades [67].Previously, numerous studies have demonstrated that chemical pesticideinduced ROS bursts activate detoxification activities related to the CncC/Keap1/Maf pathway [12,13,37].For example, exposure to clothianidin in Bradysia odoriphaga caused ROS accumulation that activated the CncC pathway and involved the P450 genes [68], and the upregulation of resistance genes induced by indoxacarb exposure in S. litura was also predicted to involve a CncC/Maf binding site [37].Moreover, xanthotoxin exposure increased ROS levels and activated the expression of the CncC/MafK pathway and detoxification genes in the larvae of S. litura, which in turn increased tolerance to λ-cypermethrin [69].In our study, the detoxification genes that varied with the level of ROS produced by enteric bacteria were all predicted to have CncC/Maf binding sites.VC feeding experiments showed that changes in ROS levels induced by L. lactis and E. casselif lavus in B. dorsalis also led to corresponding regulation of the BdCncC/BdKeap1/BdMafK pathway and the expression of downstream detoxification genes, as well as enzyme activity.Furthermore, the associated detoxification activity induced by symbiotic bacteria was inhibited after BdCncC knockdown, indicating that BdCncC is indispensable for activating the associated detoxification enzymes due to changes in the intestinal physicochemical environment caused by symbiotic bacteria.These results provide a theoretical basis for the indirect regulation of insect resistance by symbiotic bacteria.
The degree of resistance decreased as the bacterial concentration increased to 10 OD.Likewise, at concentrations greater than 500 μM, LA also reduced ROS activation in the gut.This observation suggests a potential self-protective mechanism of the organism.Excessive ROS cause oxidative stress damage to intestinal epithelial cells, resulting in irreversible damage to cellular components and leading to cell death [70].In the Drosophila intestine, when the immune deficiency pathway, which produces various antimicrobial peptides to regulate the balance of the intestinal microbiota, is blocked, the substantial proliferation and accumulation of L. plantarum lead to an excess of ROS, which ultimately damages intestinal epithelial cells and accelerates their death [42].In addition, when many exogenous pathogens invade the intestinal tract, DUOX regulates ROS, scavenges overexpressed ROS via the immune response through a variety of ROSscavenging enzymes, such as SOD, CAT, and POD, and maintains ROS levels within the threshold for host damage [58].However, ROS are also central regulatory factors for intestinal stem cell (ISC) renewal.When spatiotemporally limited, ROS initiate Jnk signaling and activate calcineurin/CRTC and ERK signaling, as well as the JAK/STAT signaling pathway, leading to downstream ISC-induced intestinal epithelial cell renewal [71].
In summary, our research provides a deeper understanding of the molecular mechanisms by which the insect gut microbiota regulates insecticide resistance through indirect physiological and biochemical reactions.For more effective pest control, innovative control strategies such as the use of nanoantibiotics [72] carrying insecticides can be developed to address the challenge of increasing insecticide resistance by symbiotic bacteria.Furthermore, the screened P450 and GST genes can serve as potential target sites for overcoming insecticide resistance.

Figure 1 .
Figure 1.The elimination of symbiotic gut bacteria reduces resistance to β-cypermethrin in B. dorsalis; (A) susceptibility of the SS and RS f lies to β-cypermethrin; LC 50 , lethal concentration that killed 50% of B. dorsalis adults; 95% FL, 95% fiducial limit of the LC 50 ; (B) susceptibility of the RS and RS + axenic f lies to β-cypermethrin; (C) gut transcriptional responses of LOC105222599-CYP6g1, LOC105233823-CYP6g1, LOC105222603-CYP6g1, LOC105226035-CYP6d4, LOC105226935-CYP6v1, LOC105232220-CYP4ae1, LOC105225813-GSTD1, LOC105229682-GSTT1, and LOC105222427-GSTD7 to the elimination of gut symbiotic bacteria in the RS f lies; (D) P450 activity and (E) GST activity in the gut of B. dorsalis between the RS and RS + axenic strain f lies; (F) the interference efficiency of P450 and GST dsRNA injection after 48 h; (G) changes in the LC 50 of β-cypermethrin after P450 and GST gene interference; the toxicity ratio was calculated by dividing the LC 50 of ds-GFP by that of ds-P450s or ds-GSTs; LC 50 values were considered significantly different if their fiducial limits did not overlap; Student's t-test was performed for C-F; error bars indicate ± s.e.m.; * * * P < .001,* * P < .01,* P < .05,ns indicates P > .05;all results were obtained from at least two independent experiments.

Figure 2 .
Figure 2. Differences in gut symbiotic bacteria between the SS and RS of B. dorsalis; (A) relative genus-level abundance profiles of bacteria in the guts of the SS and RS f lies; relative abundances of the species (B) E. casselif lavus and (C) L. lactis in the SS and RS f lies, as determined by 16S rDNA sequencing; the loads of (D) E. casselif lavus and (E) L. lactis in the guts of the SS and RS f lies; the housekeeping gene β-actin was used as an endogenous control in D and E; Student's t-test was performed for A-E; error bars indicate ± s.e.m.; * * * P < .001,* * P < .01,* P < .05;all results were obtained from at least two independent experiments.

Figure 3 .
Figure 3. Intestinal symbiotic bacteria increase resistance to β-cypermethrin in B. dorsalis; (A) toxicity regression analyses of the RS + axenic, RS + axenic+EC, and RS + axenic+LL f ly responses to β-cypermethrin; (B) gut transcriptional responses of LOC105222599-CYP6g1, LOC105233823-CYP6g1, LOC105226035-CYP6d4, LOC105226935-CYP6v1, and LOC105225813-GSTD7 to elimination of gut symbiotic bacteria followed by separate feeding of the RS f lies with E. casselif lavus and L. lactis; (C) P450 activity and (D) GST activity in the gut of the RS + axenic, RS + axenic+LL and RS + axenic+EC strain f lies; (E) P450 activity and (F) GST activity in the gut of the SS, SS + LL, and SS + EC strain f lies; (G) susceptibility response to β-cypermethrin in the SS, SS + LL, and SS + EC strain f lies; LC 50 values were considered significantly different if their fiducial limits did not overlap; (H) gut transcriptional responses of detoxification genes in the SS f lies after feeding different concentrations (0.5 OD, 5 OD, and 10 OD) of L. lactis (LL) for 3 days; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of a 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; SS + LL/EC: SS flies continuously fed 5 OD L. lactis (E.Casselif lavus) for 3 days; Student's t-test was performed for B-F, and one-way ANOVA followed by Tukey's multiple comparison test was performed for H; error bars indicate ± s.e.m.; * * * P < .001,* * P < .01,* P < .05,ns indicate P > .05;all results were obtained from at least two independent experiments.

Figure 4 .
Figure 4. Gut symbiotic bacteria regulate β-cypermethrin resistance in B. dorsalis by activating the CncC pathway; (A) CncC/Keap1/Maf pathway regulatory model in insects; (B) quantitative PCR of the BdCncC, BdMafK, and BdKeap1 genes in the intestines of the SS and RS f lies; gut transcriptional response of (C) BdCncC, (D) BdMafK, and (E) BdKeap1 in the nonaxenic (RS), RS + axenic, RS + axenic+LL, and RS + axenic+EC strain f lies; gut BdCncC silencing efficiency in B. dorsalis at 72 h postinjection with 2.0 μg of ds-GFP or ds-BdCncC in (F) RS and (G) RS + axenic+LL strain f lies; the data were normalized to the expression levels in the ds-GFP-treated f lies; gut transcriptional response of LOC105222599-CYP6g1, LOC105233823-CYP6g1, LOC105226035-CYP6d4, LOC105226935-CYP6v1, and LOC105225813-GSTD1 to gut BdCncC silencing at 72 h in (H) the RS and (I) RS + axenic+LL strain f lies;P450 activity in the gut of (J) the RS and (K) RS + axenic+ LL strain f lies after BdCncC interference for 72 h; GST activity in the gut of (L) the RS and (M) RS + axenic+LL strain f lies after BdCncC interference for 72 h; effect of BdCncC knockdown on the response of (N) RS f lies and (O) RS + axenic+LL flies to β-cypermethrin; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; Conv, conventionally colonized f lies; Axenic, axenic f lies; LC 50 values were considered significantly different if their fiducial limits did not overlap; Student's t-test was performed for B-M; error bars indicate ± s.e.m.; * * * P < .001,* * P < .01,* P < .05;all results were obtained from at least two independent experiments.

Figure 5 .
Figure 5.The activation of ROS by intestinal commensal bacteria increases the activity of detoxification enzymes in B. dorsalis; ROS activity in the gut of the SS and RS f lies was determined by (A) H 2 O 2 assays and (B) DHE analysis; ROS activity in the gut after elimination of intestinal bacteria and refeeding with E. casselif lavus or L. lactis in the RS f lies was measured by (C) H 2 O 2 assay and (D) DHE analysis; the ROS activity of the gut after continuous feeding of E. casselif lavus or L. lactis for 3 days in the SS f lies was measured by (E) H 2 O 2 assays and (F) DHE analysis; ROS activity in the gut after VC treatment in the RS f lies was measured by (G) H 2 O 2 assays and (H) DHE analysis; gut transcriptional response of the (I) BdCncC, (J) BdMafK, (K) BdKeap1, and (L) P450 or GST genes to VC treatment for 24 h in the RS f lies; (M) P450 and (N) GST activity in the gut of the RS f lies after VC treatment;(O) VC scavenges ROS in the gut and negatively modulates CncC pathway regulation; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of a 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; RS + VC: RS f lies were fed 8% sugar water containing 100 mg/ml VC for 24 h; SS + LL/EC strain: SS f lies continuously fed 5 OD L. lactis (E.casselif lavus) for 3 days; the H 2 O 2 levels in A and E were normalized to those in the SS controls, those in C were normalized to those in the RS + axenic f lies, and those in G were normalized to those in the RS f lies; the scale bars in B, D, F, and H represent 1000 μm; Student's t-test was performed for A, C, E, G, and I-N; error bars indicate ±s.e.m. * * * P < .001,* * P < .01;all results were obtained from at least two independent experiments.

Figure 6 .
Figure 6.Pesticide resistance in B. dorsalis is increased by ROS activation via BdNOX5 of intestinal commensal bacteria; (A) intestinal BdNOX5 gene expression was elevated in the RS group compared to the SS group; (B) gut transcriptional response of BdNOX5 in the RS, RS + axenic, RS + axenic+LL, and RS + axenic+EC strain f lies; (C) gut BdNOX5 silencing efficiency in B. dorsalis at 72 h postinjection of 2.0 μg of ds-GFP or ds-BdNOX5 in the RS and RS + axenic strain f lies; the data were normalized to the expression levels in the ds-BdNOX5-treated f lies; the effects of BdNOX5 silencing on ROS activity in the gut of the RS and RS + axenic strain f lies were measured by (D) H 2 O 2 assays and (E) DHE analysis; the H 2 O 2 levels in D were normalized to those of the ds-GFP controls; the scale bars in E represent 1000 μm; gut transcriptional response of BdCncC, BdMafK, and BdKeap1 to BdNOX5silencing in (F) the RS and (G) RS + axenic strain f lies; the mRNA levels of the P450 and GST genes in the guts of the BdNOX5 knockdown (H) RS and (I) RS + axenic f lies; P450 activity in the gut of (J) the RS and (K) RS + axenic f lies after BdNOX5 knockdown; GST activity in the gut of (L) the RS and (M) RS + axenic f lies after BdNOX5 knockdown; RS + axenic strain: RS strain fed mixed antibiotic and sugar water solution for 4 days; RS + axenic+LL/EC strain: RS + axenic strain f lies fed aseptic water for 24 h followed by continuous feeding of a 5 OD bacterial solution of L. lactis or E. casselif lavus for 3 days; dsBdNOX5 + LL: after injection of 2 μg dsRNA into the RS + axenic strain, continuous feeding of 5 OD L. lactis was performed for 72 h; Conv, conventionally colonized f lies; Axenic, axenic f lies; Student's t-test was performed for A-D and F-M; error bars indicate ±s.e.m. * * * P < .001,* * P < .01,* P < .05;all results were obtained from at least two independent experiments.

Figure 7 .
Figure 7. Lactate regulates detoxification in B. dorsalis through BdLDH; (A) scheme illustrating the potential mechanism of lactate-induced ROS generation in D. melanogaster; (B) levels of lactate in the guts of the SS, RS, RS + axenic, RS + axenic+LL, and RS + axenic+EC flies; N = 20 guts per biological duplication; (C) RT-qPCR showing LOC105231405-LDH expression in the guts of the SS, RS, RS + axenic, RS + axenic+LL, and RS + axenic+EC f lies; (D) PA levels were measured in the guts of f lies under different conditions: SS, RS, RS + axenic, RS + axenic+LL, and RS + axenic+EC; N = 30 guts per biological replicate; (E) LOC105231405-LDH silencing efficiency in B. dorsalis at 72 h postinjection of 2.0 μg of ds-GFP or ds-BdLDH-2 in the gut of RS strain f lies; the data were normalized to the expression levels in the ds-BdLDH-2-treated f lies; (F) PA concentrations were evaluated in the gastrointestinal tracts of the ds-GFP-and ds-BdLDH-2-treated f lies after 72 h.ROS activity in the gut of the ds-GFP-and ds-BdLDH-2-treated f lies after 72 h was measured by (G) H 2 O 2 assays and (H) DHE analysis; (I) P450 and (J) GST activity in the guts of the ds-GFP and ds-BdLDH-2 f lies after 72 h; RT-qPCR showing (K) LOC105231405-LDH and (L) BdNOX5 expression in the guts of the f lies of the indicated genotypes after 50 μM LA feeding for 72 h in the RS + axenic f lies; ROS activity in the guts of the RS + axenic and RS + axenic+LA f lies was measured by (M) H 2 O 2 assays and (N) DHE analysis; (O) mRNA levels of the BdCncC, BdMafK, BdKeap1, (P) P450 and GST genes in the guts of the RS + axenic and RS + axenic+LA f lies; (Q) P450 activity and (R) GST activity in the gut of the RS + axenic and RS + axenic+LA f lies; the H 2 O 2 levels in G and M were normalized to those in the RS + axenic controls; the scale bars in H and N represent 1000 μm; Student's t-test was performed for B-G, I-M, and O-R; error bars indicate ± s.e.m.; * * * P < .001,* * P < .01,* P < .05;all results were obtained from at least two independent experiments.