Acaricidal effect of the antimicrobial metabolite xenocoumacin 1 on spider mite control

Jiaxing Wei ,Hong Yan ,Jie Ren, ,Guangyue Li, # ,Bo Zhang,3#,Xuenong Xu,3#

1 Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, China

2 State Key Laboratory for Biology of Plant Diseases and Insect Pests, Beijing 100193, China

3 Key Laboratory of Natural Enemies Insects, Ministry of Agriculture and Rural Affairs, Beijing 100193, China

Abstract The two-spotted spider mite,Tetranychus urticae Koch,is one of the most harmful pests in many agroecosystems worldwide. To effectively manage this pest,there is an urgent need to develop novel bio-active acaricides that support integrated pest management strategies targeting T.urticae. In this study,we explored the acaricidal effects of xenocoumacin 1 (Xcn1) on T.urticae and its predator Neoseiulus californicus using the highly purified compound.Xcn1 was extracted and purified from the cell-free supernatant of the Xenorhabdus nematophila CB6 mutant constructed by the easy promoter activated compound identification (easyPACId) method. When the concentration of Xcn1 exceeded 100 μg mL–1,the survival rate of spider mite adults declined to below 40% and the fecundity was decreased by 80% at six days post-application. At concentrations of 25 and 50 μg mL–1,Xcn1 significantly impeded spider mite development by inhibiting the molt. However,neither concentration had any adverse effects on the survival or reproduction of the predatory mite N.californicus. The results from laboratory and semi-field experiments consistently demonstrated the effectiveness of the antimicrobial metabolite Xcn1 in controlling pest mites at both the molecular and physiological levels. Our study offers a promising possibility that combines the compatible biocontrol agents of Xcn1 and predatory mites for integrated pest mite control.

Keywords: pest management,predatory mite,Xcn1,morphology,developmental inhibition,transcriptome

1.Introduction

The two-spotted spider mite,TetranychusurticaeKoch,is one of the most successful global pests in multiple agroecosystems since it attacks more than 1,400 plant species across 250 plant families worldwide (Migeonet al.2010). The fast development,high fecundity,and broad feeding habits ofT.urticaeare major factors contributing to its rapid population growth on multiple host plants,including many economically important crops such as vegetables,fruits,cotton,and ornamentals which are closely linked to human health (Rott and Ponsonby 2000;Grbićet al.2011). Despite the implementation of a variety of agronomic,chemical,and biological methods for controllingT.urticae,synthetic acaricides and insecticides remain the predominant and most effective approach at present (Urbanejaet al.2008;Van Leeuwenet al.2010). Acaricides and insecticides are typically effective for rapidly controlling mite populations,particularly in the case of sudden outbreaks. However,the excessive use of these chemical acaricides has raised widespread concerns over their negative impacts on environmental safety and human health (Wanget al.2018). Furthermore,the rapid emergence of resistance to acaricides and insecticides inT.urticaeposes a great challenge (van Leeuwen and Dermauw 2016;Adesanyaet al.2021). Thus,it is urgent to develop integrated management strategies that utilize various control methods in a compatible manner,including natural enemies and other beneficial organisms,to effectively address these issues.

Predatory mites are regarded as the main predators ofT.urticaeon various host plants in natural field settings(Barberet al.2003;Zhenget al.2016;Fonsecaet al.2020). As a selective predator of tetranychid mites,NeoseiuluscalifornicusMcGregor is regarded as one of the most effective biological control agents for the management ofT.urticaeand related spider mites in various crops around the world (Fraulo and Liburd 2007;Songet al.2019). Since it was first explored in southeastern China in the early 2010s (Xuet al.2013),extensive research on the suitable distribution and adaptability ofN.californicusconducted over the past decade has demonstrated its promising application prospects (Songet al.2019;Wanget al.2021). The convenient mass rearing technique for this predator also facilitates its economical and efficient application to controlT.urticaein diverse agricultural ecosystems.However,frequent applications of pesticides could disrupt the interactions between prey and predator by reducing the population size ofN.californicusand impairing their predatory performance (Masuiet al.2018;Silvaet al.2019). Thus,the development of efficient acaricidal pesticides that are compatible with predatory mites is an urgent need for the integrated management of spider mites.

Xenorhabdusspp.,the Gram-negative bacteria symbiotically associated with entomopathogenic nematodesSteinernemaspp.,are known to produce a large number of secondary metabolites with antibacterial,antifungal,nematicidal and insecticidal activities (Bode 2009;Tobiaset al.2018;Ozkanet al.2019;Abebewet al.2022;Gulsenet al.2022). Some of the secondary metabolites have shown great potential for development as new biopesticides that can effectively control plant pathogens in the field (Yanget al.2011;Zhanget al.2019;Donget al.2020). Interestingly,a series of studies evaluated the acaricidal activity of cell-free supernatants(CFS) of differentXenorhabdusstrains against pest mites and predatory mites (Bussamanet al.2006,2012;Namsenaet al.2016;Erogluet al.2019;Nermutet al.2019;Cevizciet al.2020). For example,Incedayiet al.(2021) recently identified xenocoumacins as potent acaricidal compounds using the approach of easy promoter activated compound identification (easyPACId).However,the acaricidal effects were not tested for any purified xenocoumacin,such as xenocoumacin 1 (Xcn1),nor were the corresponding toxicological mechanisms against pest mites revealed.

In this study,we isolated and purified Xcn1 from the cell-free supernatant of theX.nematophilaCB6 mutant strain,and examined the effects of purified Xcn1 onT.urticaeand its predatorN.californicus. This system was examined from three aspects: (1) Determining whether the secondary metabolite Xcn1 could reduce pest mite fitness;(2) measuring the effective dosage of Xcn1 for controlling pest mites under laboratory and semifield conditions,as well as any side effects on predatory mite;and (3) exploring the transcriptional profiles of spider mites in response to Xcn1. This study aims to provide new insights for the integrated management of pest mites.

2.Materials and methods

2.1.Mite rearing

Spider miteT.urticaeand predatory miteN.californicuswere maintained on bean plants in the Laboratory of Predatory Mites,Institute of Plant Protection,Chinese Academy of Agricultural Sciences,for over ten years (Yanet al.2021). All mites were reared at (25±1)°C,(70±5)%RH and a 14 h L:10 h D photoperiod in the climate chambers (RXZ-380D,Ningbo,China).

2.2.Construction of deletion and promoter exchange mutants

The RNA chaperon,hfq,was reported to control the expression of biosynthetic gene clusters through sRNA/mRNA interactions (Tobiaset al.2017). To facilitate the purification of Xcn1 fromX.nematophilaCB6,we applied the easy promoter activated compound identification(easyPACId) approach to construct the mutant for the exclusive production of Xcn1 (Bodeet al.2015,2019).The Δhfqmutant was generated according to a previously reported protocol (Incedayiet al.2021). Two homologous arms,consisting of 1,000 bp fragments upstream and downstream ofhfq,were seamlessly cloned onto a linearized pJQ200SK plasmid. The resulting plasmid,pJQ-hfq,was transformed intoEscherichiacoliS17-l_pir and cultivated in Luria–Bertani (LB) medium at 37°C.Meanwhile,X.nematophilaCB6 was cultured in LB medium at 28°C. After the liquid cultures reached an optical density of 0.6–0.8 (OD600),1 mL of each culture was collected and rinsed twice with fresh LB medium.TheE.coliS17-1_pir andX.nematophilaCB6 were mixed and spotted onto an LB agar plate,which was maintained at 28°C for 18 h. Then the bacterial colonies were washed with liquid LB medium and spread on an LB agar plate containing ampicillin and gentamicin. The single crossover homologous recombination bacteria were confirmed by PCR. Thereafter,we cultured the colonies for 12 h in LB (without NaCl) and spread them on an LB plate containing 5% sucrose. Sucrose resistant colonies were picked into LB and cultured at 28°C. The Δhfqmutant generatedviathe second crossover was confirmed by PCR. ThePBAD-xcnA-Δhfqmutant,in which the native promoter ofxcnAwas replaced by an arabinose-inducible promoter (PBAD) in the Δhfqstrain,was constructed for the nearly exclusive production of Xcn1 in CB6 strain. The upstream and downstream homologous arms of the native promoter ofxcnA,the fragment of thePBADpromoter and linearized pJQ200SK were assembled to construct the plasmid pJQ-PBAD. This plasmid was then used to generate thePBAD-xcnA-Δhfqmutant through homologous recombination following the procedure mentioned above.

2.3.Preparation of cell-free supernatant of the promotor exchange mutant and purification of Xcn1

The promoter exchange mutant ofX.nematophilaCB6 was cultured with LB broth (Bacto tryptone 10 g L–1,yeast extract 5 g L–1,and NaCl 10 g L–1). One colony ofX.nematophilaCB6 mutant was inoculated into 5 mL of LB medium and cultured at 28°C with shaking at 200 r min–1for 12 h,then the culture was transferred into 100 mL of fresh LB medium in a 250 mL conical flask with an initial OD600value of 0.05. We added 1.5% L-arabinose at 8 h,and the cultivation was continued for 40 h. The CFS of the mutant was obtained by centrifugation at 10,000 r min–1for 10 min,and the supernatant was filtered through a 0.22-μm membrane to remove the CB6 cells. Xcn1 was purified from the CFS of the mutant by the following procedure. The CFS containing Xcn1 was adsorbed onto X-5 macroporous resin. The resin was then washed once with deionized water,and eluted with 50% acetone. The resulting eluted fraction was evaporated to yield the crude product,which was subsequently loaded onto a 110-cation exchange chromatography column. It was then subjected to elution with 0.2 mol L–1NH4Cl,and the resulting eluate was evaporated and dissolved in ethanol. This solution was subsequently filtered and evaporated to yield the secondary crude product. To further purify the product,the secondary crude product was passed through a CM-sephadex-C-25 cationic resin column and eluted with 0.1 mol L–1NH4Cl. The resulting sample was then dissolved in ethanol and desalted,and then evaporated to yield the purified Xcn1. The concentration of Xcn1 was determined using high-performance liquid chromatography (HPLC) by comparing it with a standard sample (Donget al.2020). An Agilent C18 column(4.6 mm×150 mm) was utilized for analyzing Xcn1,with the mobile phase consisting of ddH2O and 0.1% formic acid/acetonitrile (70:30).

2.4.Effects of CB6 CFS on the survival and fecundity of T.urticae

To ensure consistent developmental stages,we collected approximately 600 eggs ofT.urticaewithin 24 h and reared them in a plastic box. Once the nymphs from these eggs reached adulthood,30 female mites were moved on a kidney bean leaf which was wrapped with wet cotton in a Petri dish (diameter=50 mm). The lid was covered with a screen (diameter=40 mm) to facilitate moisture evaporation. The CFS was applied to mites using a hand sprayer,at a dosage of 0.5 m per Petri dish. Sterile distilled water and LB medium were used as controls. We monitored the mortality rate ofT.urticaefor seven consecutive days. The same procedure was used to apply CFS to 30 gravid adults ofT.urticaein each Petri dish,and its impact on the mite’s fecundity was evaluated.Daily egg production was recorded in each Petri dish for seven consecutive days. To ensure accurate counts of the adult females and offspring,all gravid females were moved to a new Petri dish every day. The experiment was conducted with six replicates for each treatment.

2.5.Effects of Xcn1 on the survival and fitness of T.urticae

We selected thousands of eggs that were laid within 24 h using the same sampling strategy. When they reached adulthood,we transferred approximately 30 female adults that had emerged within 24 h onto a fresh bean leaf in the Petri dish. Xcn1 was diluted into the concentrations of 5,000,2,500,1,000,500,250,100,50 and 25 μg mL–1,and each concentration was replicated in six Petri dishes with 180T.urticaeindividuals in total. Survival numbers ofT.urticaewere counted over seven consecutive days.

We then selected three concentrations of 25 μg mL–1(the minimum diluted concentration),100 μg mL–1and 250 μg mL–1(the lowest lethal concentration) to examine the effects of Xcn1 on mite development and fecundity. After gravid females had laid eggs for 24 h,we removed the adult females and applied the above three concentrations of Xcn1 to the eggs. The 1-day-old eggs were then moved individually to small arenas. Briefly,the arena was a tiny device with four tightly clipped layers: a transparent acrylic board (30 mm×20 mm×3 mm) with a 10 mm diameter hole in the middle,a bean leaf disc and two pieces of rectangular glass (30 mm×20 mm×1 mm)each at the top and bottom. In each treatment,the life parameters of 50 individuals were individually recorded starting from the egg stage. The conventionally reared and water-sprayedT.urticaewere used as controls. To test the effects of Xcn1 onT.urticaereproduction,we used the same method to treat and maintain the spider mites.

2.6.Attraction or repulsion of Xcn1 to T.urticae

To investigate whether Xcn1 could attract or repelT.urticaein a short distance,we placed two fresh leaves in a clean Petri dish (diameter=50 mm). A filter paper strip (30 mm×3 mm) connected the two leaves,one of which was sprayed with 0.5 mL Xcn1 at a concentration of 25 μg mL–1and the other was sprayed with distilled water.After a 12 h starvation period,ten 1-day-old female mite adults were released in the middle of the paper strip. We recorded the numbers of mites on each of the two leaves after 24 h. This experiment was repeated 10 times.

2.7.Effects of Xcn1 on the survival and fitness of N.californicus

We collected 1-day-old adults of predatory miteN.californicusin the Petri dish,and sprayed them with Xcn1 at concentrations of 100 and 250 μg mL–1. The mites were then transferred individually to an arena.The mortality of 30 female adults in each treatment was monitored for seven successive days. To examine the effect of Xcn1 on the predator’s development,250 eggs ofN.californicuslaid within 24 h were treated with Xcn1 at 25,100 and 250 μg mL–1. About 50 eggs were selected and reared individually in arenas to record the duration of each developmental stage. We also examined the effect of 100 μg mL–1Xcn1 on the fecundity ofN.californicus.The newly emerged adults treated with 100 μg mL–1Xcn1 were allowed to mate for 12 h before gravid females were transferred into individual arenas. The arenas were changed daily to observe egg production for seven successive days.

2.8.Semi-field assay

The experimental arena consisted of a flooded pot(diameter=12 cm) with sterilized soil and two kidney bean plants. Once the plants had grown five leaves,100T.urticaeadults were introduced onto the two bottom leaves of each plant in the experimental arena. Aliquots of Xcn1 (1 mL) at concentrations of 25,50 and 100 μg mL–1were applied to the foliage of each plant using a hand sprayer. Controls were established with both unsprayed and water-sprayed arenas. All treatments were replicated in five arenas. These arenas were maintained in a greenhouse at (24±2)°C and a 14 h L:8 h D photoperiod. We observed leaf appearances daily and took photos at 2,4,7,9,12 and 15 days postapplication (DPA). However,counting spider mites on each leaf was challenging due to the considerable overlap of the generations and the rapid population growth on the control plants. Therefore,we used the chlorophyll content of infested leaves to represent the level of mite infestation,which was measured by a handheld chlorophyll analyzer(LYS-A,Hangzhou Lvbo Instrument Company,China) at 15 DPA. As the control plants had been severely affected by mite infestation at 15 DPA,the semi-field assessment of the Xcn1 treatment was terminated at that time.

2.9.Transcriptional responses of T.urticae to Xcn1

To investigate the molecular responses ofT.urticaeto Xcn1,250 first-day female adults were treated with 250 μg mL–1of Xcn1 and water separately. After 24 h,the samples were collected and immediately frozen in liquid nitrogen. Both treatment and control samples consisted of three biological replicates,each of which included 80 individuals. RNA was extracted by the Qiagen RNeasy mini kit according to the manufacturer’s protocol. The library was conducted in SinoBiocore Company (Beijing,China) and sequenced on an Illumina NextSeq500 instrument by PE150. After removing the adapter and low-quality reads from the raw data,the clean reads were mapped to the referenceT.urticaegenome(GCF_000239435.1) using Hisat2 v2.1.0 (Kimet al.2015),and the aligned reads were assembled with StringTie v1.3.3b (Perteaet al.2015). Fragments per kilobase of transcript sequence per million base pairs (FPKM) were used to quantify the transcript abundance of each gene.Differentially expressed gene (DEG) analyses were performed by comparing treated samples with the control using DESeq2 v1.16.1 (Loveet al.2014). DEGs were identified with a cut-off of 0.05 for Benjamini-Hochberg adjustedP-values and absolute log2fold changes (FC) at 1 (Benjamini and Hochberg 1997). Gene Ontology (GO)enrichment analysis of the DEGs was implemented using the clusterProfiler v3.4.4 in R package,and significant enrichment was defined as those with a correctedP-value≤0.05 (R Core Team 2022).

2.10.Statistical analysis

All analyses except for the survival curves were performed in SPSS 20.0 software. One-way ANOVA with Tukey’s multiple comparisons was used to examine the treatment impacts of CFS and Xcn1 on the fecundity ofT.urticaeandN.californicus,as well as chlorophyll content. The log-rank (Mantel-Cox) test was used to examine survival curves,and nonlinear regression was used to explore the LC50ofT.urticaein Graphpad version 8.3.0. Contingency analyses were used to compare the development success rates in the two mite species. The Chi-square goodness of fit test was used to analyze the selections of predatory mites for either Xcn1 or water. The results were visualized in Graphpad or R v.3.5.1.

3.Results

3.1.Effects of CFS of the CB6 mutant on the survival and fecundity of T.urticae

Both the CFS of the CB6 mutant and LB medium had significantly negative effects on the survival ofT.urticaeadults compared to the water control (χ2=205.7,df=1,P<0.001;χ2=118.8,df=1,P<0.001;Fig.1-A). Treatment with LB medium caused 50% mortality in spider mites after 4 DPA,which may result from the viscosity of the liquid culture medium. Compared with the LB medium,T.urticaeadults treated with CFS were completely dead at 6 DPA,and reproduction was inhibited significantly.The cumulative egg number was only 1.08 at 7 DPA,which was strikingly lower than the values of 19.99 and 15.18 in the groups treated by water and LB,respectively(F(2,15)=148.85,P<0.001;Fig.1-C).

Fig. 1 Survival rate (A) and fecundity (C) of Tetranychus urticae treated with water,Luria-Bertani broth (LB) and cell-free supernatant of CB6 (CB6 CFS),and HPLC chromatograms of the CB6 CFS (B) and purified Xcn1(D),respectively. Error bars are SD. “n”represents the treated mite number. Asterisks indicate significant differences between different treatments on each day (P<0.001).

In order to eliminate the potential influence of the culture medium and other compounds produced by the CB6 strain (Fig.1-B),we constructed the promoter exchange mutant strain ofX.nematophilaCB6 using the easyPACId method,in order to specifically produce Xcn1.We then extracted Xcn1 from the CFS of the mutant to obtain the Xcn1 metabolite at 95% purity (Fig.1-D). We used this purified Xcn1 for the further experiments.

3.2.Effects of Xcn1 on the survival and fitness of T.urticae

We used eight concentrations to examine the lethal effects of Xcn1 onT.urticaefemale adults over one week (Fig.2-A). Higher concentrations of Xcn1 led to greater mortality rates of the mites from 4 DPA. Notably,the concentrations above the threshold of 100 μg mL–1significantly reduced the mite survival rate to less than 40%at 6 DPA,while 25 and 50 μg mL–1had no obvious effects.The LC50of Xcn1 towardsT.urticaewas determined to be 79.47 μg mL–1at 7 DPA. We subsequently chose three concentrations of Xcn1,namely 25,100,and 250 μg mL–1,to evaluate its impact on the development (Fig.2-B)and reproduction ofT.urticae(Fig.2-C). The results revealed that even the lowest concentration of Xcn1 (25 μg mL–1) could impedeT.urticaedevelopment by halting the protonymph stage until death (χ2=102.39,df=4,P<0.001;Fig.2-B and D). When the higher concentrations of Xcn1(100 and 250 μg mL–1) were applied,the fecundity of adults was significantly reduced to 4.81 and 1.58 eggs,respectively,which were notably lower than the controls(21.54 and 20.99) within the first week (F(4,25)=42.83,P<0.001;Fig.2-C). Additionally,female adults were morphologically affected by Xcn1,being dorso-ventrally flattened,with reduced body size,a less intense body color and discoloration in the opisthosoma (Fig.2-E).

Fig. 2 Fitness indices of Tetranychus urticae impacted by Xcn1. A,survival curves of mites treated with different concentrations of Xcn1 for 7 days. B,developmental success rates when the egg stage was treated with Xcn1 for 7 days. The numbers in each column represent successful individuals and total tested individuals,respectively. C,cumulative egg reproduction (mean±SD) per female in 7 successive days after spraying with Xcn1. D,morphological comparison at 9 days post-application (DPA) when eggs were treated with or without Xcn1. E,morphological comparison of adults treated with or without Xcn1 at 6 DPA. Error bars are SD. “n” represents number of individuals per treatment. ＊＊ indicates P<0.01;＊＊＊ indicates P<0.001.

We did not observe any attractive or repellent effects of Xcn1 towardsT.urticaeon bean leaves (Appendix A).Furthermore,the different treatment manners of Xcn1 onT.urticae,i.e.,leaf spray,body contact or combined approaches,demonstrated similar control efficiencies on mite fecundity (Appendix A).

3.3.Side effects of Xcn1 on predatory mite N.californicus

We tested three concentrations of Xcn1 (25,100,and 250 μg mL–1) on mite predators,which showed greater tolerance to this metabolite. For the survival rate,while the application of 100 μg mL–1of Xcn1 caused 50%mortality ofT.urticae,this concentration did not impact the survival ofN.californicus,although 250 μg mL–1was lethal to the predators (χ2=26.18,df=3,P<0.001;Fig.3-A). Similarly,Xcn1 can also inhibit molting during the development ofN.californicus,resulting in developmental rates of 26.67 and 0% when applied at 100 and 250 μg mL–1on eggs (χ2=116.23,df=4,P<0.001;Fig.3-B). However,100 μg mL–1of Xcn1 did not affect either the reproduction (F(2,76)=2.67,P=0.76;Fig.3-C) or the morphology ofN.californicusadults (Fig.3-D).

3.4.Semi-field test

Considering that the concentration of Xcn1 below 100 μg mL–1was compatible with the predator,we selected 25,50 and 100 μg mL–1concentrations of Xcn1 to test the control efficiency for spider mites in the greenhouse. As shown in Fig.4-A-a,the control plants suffered severe damage fromT.urticae,resulting in noticeable chlorosis of the bean leaves from 9 DPA. In contrast,all plants treated with Xcn1 displayed a long-lasting protection against the pest mites,with the most distinct differences observed at 15 DPA(Fig.4-A-f). In addition,the higher concentrations of Xcn1 at 50 and 100 μg mL–1exhibited better protection againstT.urticaecompared to the treatment at the concentration of 25 μg mL–1. The SPAD values also corroborated pest mite control by Xcn1,as the chlorophyll contents of leaves significantly decreased in the untreated bean plants during the experiment (F(4,80)=9.57,P<0.001;Fig.4-B).

Fig. 3 Fitness indices of Neoseiulus californicus impacted by Xcn1. A,survival curves of adults treated with different concentrations of Xcn1 for 7 days. B,developmental success rates when the egg stage was treated with Xcn1 for 7 days. The numbers in each column represent successful individuals and total tested individuals,respectively. C,cumulative egg reproduction (mean±SD) per female in 7 successive days after spraying with Xcn1. D,morphological comparison of predator adults treated with and without Xcn1. Error bars are SD. “n” represents individual numbers per treatment. NS indicates no significance and ＊＊＊ indicates P<0.001.

3.5.Transcriptomic responses of T.urticae to Xcn1

To study the pest mite’s response to Xcn1,we compared the transcriptional profiles of adults with or without Xcn1 treatment after 24 h. Principal coordinate analysis(PCA) showed clear separation of the two groups(Fig.5-A). Xcn1 significantly upregulated the expression of 440 genes,while it downregulated 87 genes ofT.urticae(Fig.5-B;Appendices B and C). Gene set enrichment analysis showed that 10 KEGG pathways were significantly regulated by the Xcn1 treatment,including at least three cytochrome P450 metabolic pathways (Fig.5-C). Among the 11 upregulated P450 genes,four were classified to the CYP2 clan,one to the CYP3 clan,and six to the CYP4 clan (Appendix D),whereas all five downregulated P450 genes were from the CYP2 clan. Two isoforms of CYP18A1 were found with contrary profiles,i.e.,one upregulated and the other downregulated. None of the glutathioneS-transferase(GST) genes increased after Xcn1 treatment,whereas three isoforms of the GST 1-like genes were in the list of downregulated genes. Since the treatment impacts vitamin digestion and absorption of the mite,many genes from biological process (BP) and molecular function(MF) were clustered in the GO enrichment (Appendix E).However,no immunity-related genes were induced under this xenobiotic stress.

4.Discussion

In the current study,we confirmed the ability of metabolite Xcn1 to control pest mites in both laboratory and semifield conditions. Using purified Xcn1,we found that Xcn1 exhibited excellent control effects by accelerating mortality,inhibiting development and reducing fecundity inT.urticae.Among the various metabolic mechanisms of xenobiotic substances,the most common mechanisms involve the increased activities of detoxification enzymes,including P450s,GSTs,and carboxy/cholinesterases (Van Leeuwen and Dermauw 2016). Intriguingly,our results show that Xcn1 inhibited the expression of several P450 and GST genes,suggesting a potential new biodegradation pathway of Xcn1 inT.urticaecompared with other pesticides.The results showed that Xcn1 strongly regulated the expression of CYP2 clan and CYP4 clan genes,but not any immunity-related genes inT.urticae. The results also indicated that Xcn1 worked as a pesticide,rather than an immunosuppressant,to activate the pathways of xenobiotic or drug metabolism. In contrast to insects which rely on the CYP3 clan for insecticide resistance,members of the CYP2 clan inT.urticaeare considered to be responsible for acaricide resistance,in particular the CYP392 family(Dermauwet al.2020;Xuet al.2020;Kurlovset al.2022).Xcn1 reduced the expression of CYP392 inT.urticae,which might indicate its potential for use as an alternative or supplement to conventional acaricides in order to avoid pesticide resistance.

Fig. 4 Dynamics of leaf damage by introducing Tetranychus urticae in a semi-field experiment. A,leaf morphology after spraying with Xcn1. Insets of a,b,c,d and e represent different treatments of: control,water,25 μg mL–1,50 μg mL–1 and 100 μg mL–1.I,II,III,IV,V,VI indicate leaf morphology at 2,4,7,9,12,and 15 days post-application (DPA). f,color contrast of bean leaves at 15 DPA. B,chlorophyll contents of leaves at 15 DPA. Error bars are SD (n=13,18,18,18,18 for each treatment from left to right,respectively).

Fig. 5 Transcriptomic analysis of Tetranychus urticae after treatment with Xcn1. A,principal coordinate analysis (PCA) of the RNA-seq data from T.urticae treated with Xcn1 and control (n=3). B,gene numbers of significantly differentially expressed genes(DEGs) between the groups. C,enriched KEGG pathways in T.urticae affected by Xcn1.

P450 members are also involved in the biosynthetic pathways of ecdysteroids and juvenile hormones,which are responsible for the development and reproduction of insects (Rewitzet al.2006;Iga and Kataoka 2012;Yanget al.2021). In this study,two isoforms of CYP18A1 from P450 clan 2 were significantly affected by Xcn1. This gene was first reported to be involved in molting hormone inactivation by encoding an ecdysteroid 26-hydroxylase inDrosophila(Guittardet al.2011). Knockdown of CYP18A1 led to the mortality of pupae or adults during metamorphosis in the fly and beetle (Guittardet al.2011;Wuet al.2023). Although no direct CYP18A1 functional study has been reported in spider mites,such as study might shed light on the possible mechanism of Xcn1 againstT.urticaeby disturbing the normal biological functions and molting process. Additionally,Xcn1 remarkably and rapidly terminated mite reproduction,as we observed that the reduced egg production occurred from 1 DPA at several concentrations. The significant morphological changes indicated an abnormal reproductive capacity in females,with a 50% decrease in the size of their abdomen. Furthermore,the decline in adult fitness may be attributed to either hormonal imbalances or excessive energy consumption.

Both laboratory and semi-field results showed Xcn1 to be highly effective in controlling pest mites. Its prolonged effectiveness on leaves would allow it to consistently control leaf-feeding pests over multiple generations.However,we must acknowledge the potential adverse effects of high concentrations of Xcn1 on predators,as evidenced by the obvious symptoms of toxicity observed in predatory miteN.californicus. Fortunately,relative to pest mites,the predators showed higher tolerance to Xcn1. According to our results,Xcn1 in the range of 25–100 μg mL–1would not impact the survival or reproduction of predator adults,while it would be lethal to the spider mites. These results are consistent with a previous report,which attributed differences in tolerance to the reduced contact of the mite body with the bacterial metabolites,due to morphological differences such as longer legs,thicker cuticle and the diet of the predatory mites (Cevizciet al.2020). However,it is important to note that concentrations of Xcn1 at or above 100 μg mL–1can result in phytotoxicity symptoms in the bean leaves(Appendix F). Therefore,to avoid injury to non-target species,we recommend using a lower concentration of Xcn1 for pest mite control,such as 25 or 50 μg mL–1.

Xenocoumacins are water-soluble peptide antimicrobial natural products produced byXenorhabdusspecies,and xenocoumacin 1 (Xcn1) and xenocoumacin 2 (Xcn2)are the major bioactive compounds (Reimeret al.2011).Furthermore,several xenocoumacin intermediates were found to be involved in the conversion process from Xcn1 to Xcn2 (Reimeret al.2009). Although both of these Xcns show antimicrobial activity,Xcn1 demonstrated higher antifungal activity. Previous studies reported that Xcn1 exhibits a broad and strong antifungal activity against plant pathogens,such asBotrytiscinerea,Alternariaalternata,RhizoctoniasolaniandPhytophthoraspecies (Huanget al.2005,2006;Yanget al.2011;Zhouet al.2017). One particular study showed that Xcn1 efficiently controlled the phytophthora blight of pepper with a disease reduction rate of 99% at a concentration of 5 μg mL–1(Zhouet al.2017).Coupled with the results presented here,we suppose that Xcn1 could be used as a multi-functional biopesticide to control both plant pests and diseases simultaneously with one application. The long persistence of Xcn1 on leaves could consistently control pest mites through feeding.Meanwhile,the antimicrobial effects of Xcn1 might also inhibit the gut microbiome of the mites,and further influence individual mite performance and population expansion to some extent.

5.Conclusion

Xcn1 was shown to be an attractive bio-acaricide,which effectively controlledT.urticaeby affecting mortality,development and fecundity. The semi-field experiment further corroborated the effectiveness of Xcn1 as a bioacaricide that can be applied in an open environment.Importantly,Xcn1 at a concentration below 100 μg mL–1was compatible with the natural enemy. Therefore,combining Xcn1 with predatory mites would form a synergistic management strategy for pest mite control with a higher efficiency. Moreover,considering the antimicrobial activity of Xcn1 against plant pathogens,we anticipate that an integrated management strategy including Xcn1,predatory mites and other compatible biocontrol agents will provide a new and eco-friendly avenue for controlling plant diseases and pest mites simultaneously.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (32070402) and the Beijing Natural Science Foundation,China (6222052).

Declaration of competing interests

The authors declare that they have no conflict of interest.

Ethical statement

All applicable international,national and institutional guidelines for the care and use of animals were followed.

Appendicesassociated with this paper are available on https://doi.org/10.1016/j.jia.2023.06.008