Nematicidal activities of aureothin and alloaureothin from endophytic bacteria, Streptomyces sp. AE170020, against the pine wood nematode, Bursaphelenchus xylophilus

Min-Kyoung Kang Korea Research Institute of Bioscience and Biotechnology (KRIBB) Jong-Hoon Kim Korea Research Institute of Bioscience and Biotechnology (KRIBB) Min-Jiao Liu Korea Research Institute of Bioscience and Biotechnology (KRIBB) Chun-Zhi Jin Korea Research Institute of Bioscience and Biotechnology (KRIBB) Dong-Jin Park Korea Research Institute of Bioscience and Biotechnology (KRIBB) Junheon Kim National Institute of Forest Science Kwang-Hee Son (  sonkh@kribb.re.kr ) Korea Research Institute of Bioscience and Biotechnology (KRIBB) Chang-Jin Kim Korea Research Institute of Bioscience and Biotechnology (KRIBB)


Introduction
Pine trees are one of the most important tree species in global ecosystems. They are not only a predominant component of natural reserves, parks, and urban ornamental landscapes but are also a valuable source of high-value timber carpentry items. The most serious threat to pine forests worldwide is pine wilt disease (PWD) caused by the pine wood nematode (PWN), Bursaphelenchus xylophilus. B. xylophilus is a quarantined plant parasitic nematode of the Aphelenchoidoidea superfamily and belongs to clade 10 tylenchida 1 .. Native to North America, it was rst introduced to Japan at the beginning of the 20th century 2 . Since then, PWN has spread further to China, South Korea, Portugal, and thereafter to Europe 3,4 . When pine trees are infected with pathogenic PWN, some dynamic host responses are induced by the nematode infection, such as the generation of superoxide anions, the development of vacuoles in ray parenchyma cells to capacity, and a dramatic increase in lipid peroxide levels, resulting in host mortality 5 . PWD, a disease that has reportedly caused signi cant damage to forestry, ecologies of the affected countries, and local economies, is becoming a matter of concern in several continents.
The current means of controlling PWNs rely on chemical nematicides, which are expensive, ine cient and have adverse impacts on the environment and human health 6 . Indeed, the use of bene cial bacteria and bacterial nematicides has spawned intense interest in the development of safer alternatives for PWN management 7,8 . However, many problems remain regarding the bacterial control of nematodes, including the relative scarcity of nematicidal bacteria. Only a few bacteria have been reported to possess activity against nematodes and to show potential for the biological control of nematodes [9][10][11] . Identifying novel nematicidal bacteria has become urgent for the bacterial control of nematodes and is of vital practical and economic signi cance.
Endophytic bacteria that reside in the tissues of living plants and establish a symbiotic relationship with the host are relatively unstudied and potent sources of novel natural products for application in medicine, industry, and agriculture [12][13][14] . Several reports have investigated bacterial endophytes as possible agents against PWNs 15,16 . These bacteria kill nematodes by different mechanisms, including by mobilizing plant defenses and producing an array of secondary metabolites and enzymes.
The community of endophytic bacteria in Pinus spp. has been investigated using several methods, such as cultivation-based methodologies, biochemical characterization of the isolates by BIOLOG phenotypic assay 17 , and molecular approaches [18][19][20] . Strains belonging to the genera Bacillus, Paenibacillus, and Pseudomonas were isolated from several pine tree species, and some of these bacteria were found to produce siderophores 18 or have the ability to x nitrogen 19,20 . However, bacteria associated with Pinus spp. have rarely been explored with regard to their nematicidal activity against B. xylophilus 21 . Therefore, the objectives of this study are (i) to isolate the nematicidal substances from Streptomyces sp. AE170020, an endophytic bacteria isolated from root tissues of pine tree; (ii) to investigate the effects of active substances on B. xylophilus; and (iii) to evaluate the disease control e cacy of the strain as a biocontrol agent in vivo.
MeOH and water for HPLC analysis were purchased from Duksan Pure Chemicals Co., Ltd., Korea. Folin-Ciocalteu's phenol reagent was obtained from Merck KGaA, Darmstadt, Germany. Tris buffer and gallic acid were purchased from Bioneer Corp and MP Biomedicals, Korea, respectively.
Pine tree samples were obtained with permission from the National Institute of Forest Science. We complied with relevant institutional, national and international guidelines and laws in the study of the use of pine samples.
Plant samples were washed with an ultrasonic step (160 W, 30 min) to remove the surface soils and epiphytes. The washed plants were put into sterilized Petri dishes with aseptic lter paper to remove the surface water. After drying, plants were subjected to a seven-step surface sterilization procedure: immersion in 0.1% Tween 20 for 1 min, followed by a 6-min wash in 5% sodium hypochlorite, a 10-min wash in 2.5% sterilized sodium thiosulfate, three washes in sterile water, a 4-min wash in 70% ethanol, ve washes in sterile distilled water, and a nal rinse in 5% sodium bicarbonate for 10 min. Samples were then put into 2 ml sterile tubes and grinded using liquid nitrogen and then put onto TWYE media (0.25g Yeast extract, 0.5g K 2 HPO 4 , agar, 1L DW). After six weeks of inoculation, the isolate was selected and puri ed. Genomic DNA was extracted, and the 16S rRNA gene was ampli ed using polymerase reaction.
The resulting 16S rRNA gene sequence was compared with the curated sequences from EzBioCloud (http://www.ezbiocloud.net/) and a phylogenetic tree was built with the neighbor-joining method.
Acquisition of pine wood nematode (PWD). B. xylophilus were supplied by the National Institute of Forest Science, Seoul, Korea. B. xylophilus were reared on a lawn of Botrytis cinerea cultured on potato dextrose agar (Difco, BD) medium in the dark at 25°C for 7 days; nematodes were then extracted using the Baermann funnel method for bioassay 22 . To obtain bacteria-free PWNs, nematodes were extracted and transferred to Petri dishes containing sterile distilled water to allow egg laying for 3 h. Most eggs adhered to the bottom surface of the Petri dish and were gently washed with distilled water many times to remove the hatched nematodes. Then, 15% H 2 O 2 was added into the Petri dish, which was rinsed for 60 min at 25°C; next, the Petri dish was washed with sterile water three times. Eggs were then resuspended in sterile distilled water and allowed to hatch. The PWN J2 S that were almost synchronized after 24 h at 25°C were obtained and used for mortality assays. The J2 S were transferred onto a culture of B. cinerea on a PDA plate, and J3 S and J4 S /adults were collected at 30 h and 78 h after feeding re-initiation, respectively.
Detecting nematicidal activity of strain Streptomyces sp. AE170020. The strain was inoculated into the GSS broth (20 g glucose; 10 g soluble starch; 25 g soybean meal; 1 g beef extract; 4 g yeast extract; 2 g NaCl; 0.25 g K 2 HPO 4 ; 2 g CaCO 3 , and 1 L DW) for 36 h. Two percent of the seed was transferred to a 500 mL ask containing 50 mL of GSS broth and incubated on a rotary shaker at 150 rpm at 28°C for 1-10 days. The culture of strain AE170020 was then centrifuged at 13,000 rpm for 10 min to separate the broth and cells. The resultant supernatant was aseptically transferred to sterile tubes and used for a nematicidal bioassay, as described by Gao et al. (2016), with some modi cation 23 . Brie y, 50 µL of the supernatant and bacteria-free L2s of B. xylophilus (n = 30-40) specimens were added to each well of a 96-well tissue culture plate. Later, 100 µg mL − 1 streptomycin was also added to each well to inhibit bacterial contamination, and GSS broth was used as a control.
Detecting the stability of nematicidal activity of Streptomyces sp. AE170020 culture. The stability of strain AE170020 was detected following the method of Gao et al. (2016). The supernatant of strain Streptomyces sp. AE170020 was treated with 20 mg mL − 1 of proteinase K at 37°C for 30 min or boiled at 100°C for 5 min. Thereafter, the nematicidal activity of the supernatant was detected. This supernatant was also adjusted to pH values from 1.0 to 11.0 by HCl or NaOH solution and incubated for 2 h; the nematicidal activity of the supernatant was then detected after the pH value was re-adjusted to 7.0.
Fermentation, extraction, and puri cation of nematicidal substances from Streptomyces sp. AE170020 culture. Strain cultures were prepared as described above and centrifuged at 6,000 rpm for 30 min. The collected supernatant was then sequentially extracted with an equivalent volume of n-hexane, CHCl 3 , EtOAc, and n-BuOH. The mycelium of strain AE170020 was extracted with acetone, sonicated for 30 min, and kept overnight. All extractions were performed thrice. Different solvent layers were concentrated in vacuo to obtain the dry extracts. The nematicidal activity of the extracts was determined: 2 µL of the solvent extract of Streptomyces sp. AE170020 was dissolved in DMSO and added to the 96-well plate, containing approximately 40-50 L2s nematodes in 98 µL of sterile water. The nal concentrations of solvent extracts were 0.0625, 0.1250, 0.2500, 0.500, and 1.000 mg mL − 1 , and 2 µL of DMSO was used as a control. Acetone extracts, which showed the highest activity against PWN, were used for further puri cation.
The fractions extracted by acetone were applied on a silica gel column and eluted with a stepwise nhexane/acetone gradient of increasing polarity. Fractions were checked on a TLC plate, and those with similar TLC patterns were pooled for nematicidal activity testing. One of the fractions that showed high nematicidal activity was applied on a Sephadex LH-20 column and eluted with MeOH. Nematicidal activity of subfractions was tested and combined as above. Further, active subfractions were puri ed with preparative TLC plates, and two active compounds were obtained. HPLC with a gradient solvent system of aqueous MeOH, containing 0.04% TFA, as a mobile phase was used to check the purity of the compound.
Different developmental stages of B. xylophilus mortality assay. Two of the most potent puri ed compounds, alloaueohtin and aureothin, were further explored for their effects on the different stages, reproduction, growth, and behavior of B. xylophilus. Different developmental stages of nematode were prepared as previously described, and L2s were exposed to the two compounds at nal concentrations of 0.05, 0.10, 0.20, 0.50, 1.00, 2.00, and 4.00 µg mL − 1 ; L3s and L4s/adults were exposed to concentrations of 0.25, 0.50, 1.00, 2.50, 5.00, and 10.00 µg mL − 1 . The commercial nematicide, abamectin, at the same concentrations and DMSO were used as positive and solvent controls, respectively.
The treated plates were stored in the dark at 25°C, and the mortality of PWNs was recorded after 24-h treatment by visualization under a light microscope (Nikon, SMZ-U). The LC 50 values of active compounds at different life stages were estimated. Mortality was de ned based on the observation of motility, a visibly moving nematode was marked as alive, and nematodes that failed to respond after several touches were marked as dead. The bioassays were performed three times in triplicate. Mortality was calculated according to the following formula: mortality (%) = dead juveniles/total juveniles × 100.
Effects of compounds on the locomotion behavior of B. xylophilus. In liquids, a single movement of the stylet knob forwards and then backwards to its original position was called "thrashing" 24 . To assess the impact of alloaureothin and aureothin on B. xylophilus mortality, thrashing of L3s nematode was assayed. Thrashing rate was scored for 1 min at the following time points: 2, 6, and 24 h in the presence of two compounds at the indicated concentrations (1.0, 2.5 and 5.0 µg mL − 1 ). Control assays were conducted in the presence of DMSO or abamectin, and all assays were conducted at 22°C. Two trials were performed for this experiment, and six numbers of surviving nematode in each treatment were used to test the thrashing behavior.
Nematode population inhibition assay. A population inhibition assay was conducted according to the method described by Cheng et al. (2017) with some modi cation. Brie y, adult nematodes (~ 500) were initially treated with alloaureothin and aureothin at concentrations of 0.5 and 1.0µg mL − 1 , and incubated for 24 h at 22°C. After 24 h, 10 female and male nematodes were randomly selected and placed on a PDA plate fully covered with Botrytis cinerea and allowed to grow. When the B. cinerea has been completely consumed by nematodes in control plates, nematodes were extracted from plates using distilled water. The recovered nematodes were then serially diluted and numbers of nematodes in 100 µL suspensions were counted. The reproduction rates (P f /P i , P f , nal nematode population; P i , initial nematode population) were calculated. The experiment was conducted twice, with three replicates per treatment.
Nematode hatch inhibition assay. Nematode eggs were obtained as above and used for the embryonic lethality tests. Experiments were conducted using established procedures 24 . Egg suspensions were exposed to alloaureothin and aureothin at nal concentrations of 20, 15, 10, 5, 2, 1, 0.5, and 0.1µg mL − 1 . Two microliters of DMSO and abamectin at the same concentration were served as negative and positive, respectively. Plates were incubated at 22°C and nematodes of the L2s stage were scored after 48 h. The hatching rate was calculated according to the following formula: Hatching rate (%, HR) = [juveniles / (eggs + juveniles) × 100]. Each treatment was repeated in three wells, and the experiment was repeated three times.
Egg deposition assay. One-day-old female nematodes in the presence of alloaureothin, aureothin or abamectin (5 ug mL − 1 ) were used in the egg deposition assay according to an established protocol 24 . B. xylophilus (about 100 numbers) were treated with compound for 24 h in the 96-well plates. Approximately 10 male and 10 female nematodes were then selected randomly and transferred to Petri dishes and allowed to lay eggs for 24 h. The eggs laid by a single female were then recorded and the nematodes inside the female body were also imaged. The experiments were performed two times with six repetitions.
Suppression of pine wilt disease under pot condition. Injection of active substances into pine trees was performed to evaluate the control e cacy in vivo, as previously described 25  and Pinus densi ora were transferred to a greenhouse for at least one month prior to the pot test and watered every other day and fertilized if necessary. The average temperature of the greenhouse was 25 ± 5°C. Acetone extracts of strain AE170020 were prepared as described above and dissolved using DPPT solvent (20% DMSO, 20% propylene glycol monomethyl ether, 50% propylene glycol, and 10% Tween-20). The solution of abamectin in DPPT solvent served as a positive control. Holes were made in the trunk of pine trees (average height, 124 cm; average basal diameter, 1.4 cm) by using an electric drill about 5 cm above the ground level. Then, 200 µL of strain extract was injected into holes at different concentrations, and holes were immediately covered with para lm. The nal concentration of the acetone extract was 7.2 mg, 3.6 mg, and 1.8 mg per tree and abamectin was 3.6 mg per tree. Control plants were injected with the DPPT solution. Inoculations of B. xylophilus were performed one week after the injection of chemicals in the trunk. B. xylophilus were reared on B. cinerea and extracted; they were then washed with distilled water at least ve times to remove any fungal hyphae present. Final nematode density in the suspension was adjusted to about 20,000 nematodes per mL. A hole in the trunk of each pine tree was made approximately 20 cm above the soil. A 0.1 mL aliquot of the nematode suspension (containing about 2,000 nematodes: a mixture of adults and juveniles) was pipetted into the holes and the holes were covered with para lm. Control plants were also wounded but distilled water was used instead of the PWN solution. Pine trees were watered three times a week. Three runs with ve replicates for each treatment were conducted. External symptoms of pine trees were visually assessed at 15, 30, 45, and 60 days after the inoculation and results were recorded at 60 days.
Sample collection and processing. Sixty days after inoculation, three pine trees from each treatment in one trial were randomly selected and collected. Plants were cut immediately above the ground. Needles and the twigs were removed, and the stems were chopped into small pieces. Stems were immediately weighed and used for nematode quanti cation. The remaining stems and leaves were stored at − 80°C for chemical analysis.
Nematode quanti cation in inoculated trees. The nematode population was measured using established procedures 26 . Five grams of the stem pieces were weighed and transferred into gauze equipped in the upper part of a 50 mL falcon tube. The plant tissues were then immersed in water for 24 h, and after that the gauze and plants were gently removed from tubes and allowed to precipitate for 12 h. Nematodes were mostly located at the bottom of the falcon tube after the precipitation and the upper water was removed to make a nal volume of 10 mL. Then, 100 µL of the extracted nematodes was transferred into 96-well tissue plates, and live nematodes were counted under a light microscope. Nematode density was expressed as the number of nematodes in the stem per gram of wood. The data of each treatment were generated from three trees and obtained in triplicate.
Chemical analysis. Chemical analyses were performed on leaf tissues (total chlorophyll, total polyphenolics, and lipid peroxidation) and stem tissues (water content). Leaves collected as previously described were used for chlorophyll quanti cation. The extraction and determination of the chlorophyll content were performed as per the method described by 26  The concentration of total soluble phenolics was determined using the protocol described by Ainsworth et al (2007) 27 . Needles were ground with liquid nitrogen; about 100 mg of leaves was extracted with 95% methanol and sonicated for 15 min. Samples were then gently shook and allowed to react overnight in the dark at room temperature. Later, 200 µL of 10% (vol/vol) Folin-Ciocalteu's phenol reagent was added to 100 µL of the methanolic extract. The solution was mixed thoroughly and allowed to stand for 5 min, after which 800 µL of 700 mM Na 2 CO 3 was added, and the tubes were incubated at ambient room temperature for 2 h. Next, 150 µL of each sample solution was loaded on a 96-well microplate and the absorbance was measured at 760 nm with a spectrophotometric microplate reader (VERSAmax™). The total phenolic concentration was calculated from the gallic acid calibration curve. Data were expressed as gallic acid equivalents (GA)/g of extracts averaged from each treatment.
To gain insights about plant oxidative stress and cell damage, lipid peroxidation was measured. The level of lipid peroxidation in tissues was determined in terms of the malondialdehyde (MDA) content by the method of Nunes et al (2015). After homogenization with liquid nitrogen, about 0.1 g of leaf sample was extracted with 10 mL 0.5% thiobarbituric acid in 20% trichloroacetic acid. The sample was then vortexed thoroughly and incubated at 100°C for 30 min. Next, reaction solutions were immediately transferred to ice for 5 min and then centrifuged at 4,000 rpm for 15 min. The absorbance was monitored at 450, 532, and 600 nm in a UV/visible spectrophotometer. Calculation of MDA was based on the following formula: MDA concentration (µmol L − 1 ) = 6.45(A 532 -A 600 ) -0.56A 450 .
Speci c water content was used as a proxy for water stress, which is usually associated with PWD. For water content determination, each wood sample was oven dried for 72 h, and the relative water content was evaluated through the following formula: relative water content (%) = (fresh weight -dry weight) / fresh weight × 100. All the chemical data were obtained from six trees and three samples per tree.
Statistical analysis. Data analyses were performed using SPSS 18.0 software (SPSS Inc.), and the LC 50 values were determined via a probit analysis. Values are expressed as mean ± standard deviation (SD) unless indicated otherwise. The data were compared using analysis of variance (ANOVA) followed by a post-hoc test, as appropriate. Signi cant differences were determined according to thresholds of *p < 0.05; **p < 0.01, and ***p < 0.001. All charts and gures generated in this study were constructed using GraphPad Prism version 8.0.2.

Results
Molecular and phylogenetic analysis. The identi cation of 16S rRNA gene sequence using EzTaxon server revealed that strain AE170020 belongs to the genus Streptomyces and it shared the highest (98.6%) sequence similarity with Streptomyces stramineus. The neighbor-joining phylogenetic trees analysis showed that strain AE170020 lies in a subclade in the tree with species of genera Streptomyces (Fig. 1).
Characterization of nematicidal activity of Streptomyces sp. AE170020. As shown in Fig. 2, the 3-d to 10d fermentation supernatant of strain AE170020 had evident activity against B. xylophilus, with mortality ranging from 62.8 ± 2.5% to 94.2 ± 0.8%. The highest mortality was observed when B. xylophilus was exposure to 5-d and 6-d fermentation broths. The broth of GSS medium led to a modest mortality rate of 6.7 ± 0.6% (Fig. 2).
The nematicidal substances of strain AE170020 were also analyzed for their stability at different pH values. It was found that the supernatant of strain AE170020 exhibited high nematicidal activity within a pH range of 4.0-11.0. However, the nematicidal activity slightly decreased at pH 4.0 and 5.0, and no mortality was observed in strong acidic environments (pH 1.0-3.0) (Fig. 3A).
The supernatant of strain AE170020 was also analyzed for its stability towards protease digestion. After hydrolysis with protease K, the supernatant of strain AE170020 also showed high nematicidal activity with a mortality of 91.6 ± 2.9% against B. xylophilus (Fig. 3B). Therefore, we speculated that strain AE170020 produces some extracellular substances to kill the nematode and the nematicidal substances of the fermentation broth are not protein but could be secondary metabolites instead.
Nematicidal activity of strain AE170020 extracts. The mortality rates of different solvent partitions were assessed as shown in Fig. 4. The extracts of different solvents clearly display a concentration-dependent nematicidal activity against B. xylophilus. The phase extracted by acetone showed the highest nematicidal activity with 89.2 ± 2.2% mortality at a concentration of 1.0 mg mL − 1 , followed by the hexane extract. The extracts of CHCl 3 and EtOAc also exhibited nematicidal activities relative to the control group with a mortality rate of 5.7 ± 0.7%. However, BuOH extract and DW layer had no activity.
Puri cation and identi cation of bioactive compound. 1 H NMR (400 MHz, CDCl 3 ) and 13 C NMR (100 MHz, CDCl 3 ) data for the two compounds are listed in Table 1. On the basis of spectroscopic data and comparison with the values reported in the literature, the two compounds were assigned as alloaureothin and aureothin, respectively (Fig. 5). Effects of alloaureothin and aureothin on different developmental stages of B. xylophilus. Alloaureothin and aureothin showed highly signi cant activity against L2s, L3s, and L4s/adults of B. xylophilus, and as expected, the two compounds concentration-dependently killed B. xylophilus. Among these, the L2s stage is the most sensitive to the two compounds, followed by L3s and L4s/adults stages, and this sensitivity pattern is similar to that observed with the positive control, abamectin (Fig. 6). After 24-h exposure, the LC 50 values of alloaureothin on different life stages of B. xylophilus: (L2s, L3s, and L4s/adults) were 0.83, 1.10, and 1.47 µg mL − 1 , respectively, and the values of aureothin were 0.81, 1.15, and 1.54 µg mL − 1 respectively (Fig. 6A, B, C). Compared with abamectin, both compounds showed higher nematicidal activity against B. xylophilus at all tested life stages, and the two compounds showed similar mortality rates against B. xylophilus.
Upon exposure to the active compounds at a concentration of 5.0 µg mL − 1 , nematodes almost had no movement after 24-h treatment (Fig. 7A).
Effects of alloaureothin and aureothin on the reproductive traits of B. xylophilus. The mycelia of Botrytis cinerea were almost completely consumed by nematodes after 7 days in the non-treated plates and the DMSO-treated group exhibited a similar clearance zone to the non-treated plates (Fig. 8Ai, ii). In contrast, in the plates treated with two active compounds, clearance zones were clearly reduced (Fig. 8Aiii, iv, v, vi).
Alloaureothin and aureothin affected the hatching and fecundity of nematode. The effect of active compounds on reproductive activity was measured in vitro by direct contact. The results suggested that the hatching rate of B. xylophilus was signi cantly inhibited by the test compounds in a dose-dependent manner. Non-treated and solvent controls presented a hatching rate of 92.1 ± 2.1% and 87.5 ± 3.3% respectively, whereas alloaureothin and aureothin at a concentration of 20 µg mL − 1 treated nematode eggs almost failed to hatch. In addition, both compounds exhibited a signi cantly lower hatching rate compared with abamectin (Fig. 9A).
Furthermore, two active compounds also adversely in uenced the fecundity of gravid adults at the concentration of 5 µg mL − 1 . Eggs laid by a single female in non-treatment and DMSO-treated were 12 ± 3 and 12 ± 2, respectively. However, the values were reduced to 3 ± 1, 2 ± 1, and 6 ± 1 by alloaureothin, aureothin, and abamectin (Fig. 9B). Microscopic imaging also revealed the inability of female nematodes to lay eggs, leading to the accumulation of eggs inside nematodes (Fig. 9C).
Suppression of pine wilt disease under pot condition. Successful nematode infection in pine trees is visualized by the presence of needle discoloration, and these visual symptoms were registered to understand and compare the effect of compounds on tree survival. As shown in Table 2, after inoculation with B. xylophilus, 90% of P. densi ora plants in the control group died after 60 days. The injection of acetone extracts of Streptomyces sp. AE170020 and abamectin into pine trees at test concentrations successfully suppressed the development of PWD in pine trees arti cially infected with B. xylophilus.
Remarkably, extracts of strain AE170020 exhibited 100% control e cacy at a concentration of 7.2 mg per tree. Moreover, after 60 days, the nematode population was four to nine times larger in the nematode control group than in the compound-injection groups, with numbers increasing from about 2000 nematodes at inoculation to 3426 ± 576 nematodes per gram of plant tissue ( Table 2). The lowest density of nematodes (350 ± 84) was recovered from pine trees treated with strain extract at the concentration of 7.2 mg per tree.
Variation of chemical traits after inoculation. It was clearly seen that injection of abamectin and acetone extract of strain AE170020 successfully prevented the reductions of water content, total chlorophyll content, and total polyphenolics (Fig. 10). PWN inoculation without acetone extract of strain 680998 signi cantly reduced the stem water content. After inoculation, the pine trees inoculated only with the nematodes showed relatively low water content of 18.6 ± 4.0% (Fig. 10A). From the analysis of Fig. 10B, we can observe that the total chlorophyll content signi cantly decreased in the nematode-control group in comparison with the acetone extract-treated and solvent-control groups. However, the nematodeinoculated plants, either with or without extracts, showed a lower concentration of total chlorophyll than the uninoculated group plants. B. xylophilus resulted in the decrease of total polyphenolics of tissues compared to the solvent-control group (71.0 ± 8.4 mg gallic acid Eq. g − 1 leaf). The synthesis of total polyphenols in those treated with acetone substances was higher than in those inoculated only with the nematodes (Fig. 10C); however, the difference was not signi cant between the trees injected with acetone extract of 680998 at the concentrations of 7.2 and 3.6 mg per tree and those in nematode-control group.
As shown in Fig. 10D, the concentration of MDA in the nematode-control group was greater than in other groups, suggesting a high degree of lipid oxidative damage caused by PWN. Injection with active extracts protected pine trees from this damage. The increasing concentration of malondialdehyde (MDA) in the nematode-inoculation control group suggested a high degree of lipid oxidative damage caused by PWNs, and abamectin and strain extracts protected pine trees from this damage.

Discussion
PWD, which is caused by PWN, is one of the most destructive diseases in trees of the genus Pinus and is responsible for environmental and economic losses around the world. The frequently used nematicides to control the disease are chemical nematicides. However, these chemical nematicides reportedly have deleterious effects on human health and the environment. Biological control of the PWD using microorganisms can be a safe, cost-effective, and e cient method. Endophytes reside with plant tissues without causing any signs of disease or infections and are rich sources of bioactive substances and promising biocontrol agents because of the direct effects of metabolites to induce mortality in nematodes. Many endophytic bacteria are reported to exert nematicidal activity by producing nematicidal toxins [27][28][29] .
In the present study, an endophytic bacteria, Streptomyces sp. AE170020, isolated from pine trees was able to kill B. xylophilus. We further investigated the active substances of this strain and hypothesized that some secondary metabolites may be responsible for the nematicidal activity. Subsequently, the nematicidal substances were puri ed by using bioassay-guided fractionation and two active compounds, alloaureothin and aureothin, were identi ed. Alloaureothin and aureothin, which are polypropionates with a nitro group belonging to the class of polyketides, exhibited high activity against B. xylophilus. Two potent compounds were previously isolated from several actinomycetes and displayed interesting biological activities. Aureothin isolated from the mycelium of Streptomyces thioluteus 30 and Streptomyces sp. MM23 28 was reported to exhibit anti-Helicobacter, antitumor, and antifungal activities 28, 31 . In accordance with our results, previous studies have also shown that aureothin exerts nematicidal activity 30 . Alloaureothin obtained from Streptomyces sp. MM23 was reported to exhibit growth inhibitory effects against human brosarcoma HT1080 cells with an IC 50 value of 30 µM 28 . In our present study, alloaureothin also showed high nematicidal activity against B. xylophilus.
It is important to investigate the mode of action of active compounds in the practice of nematode control, because it can provide useful information for avoiding resistance and choosing appropriate way of delivery. Further research was conducted to evaluate the effects of two potent compounds, alloaureothin and aureothin, on the growth, reproduction, and development of nematodes. The present results showed that both compounds were lethal to all developmental stages of nematode and displayed a lower LC 50 value than the commercial nematicide, abamectin. In addition, the nematode population was signi cantly reduced by two active compounds, which supports that both compounds control the reproductive behavior of nematodes. Eggs hatchability is one of the most important parameters that affects population numbers and it has been suggested that chemicals with high embryonic lethality can be potential nematicides 31,32 .The two compounds obtained in our research both effectively inhibited the hatchability of nematode eggs and can be considered as potential nematicides.
The trunk injection method involves injection of pesticides directly into the tree trunk; the liquids are then transported through the plant's conductive tissues to the site of action. This method has been successfully used in biocontrol of PWD 33,34 . Abamectin, a member of macrocyclic lactones, has been widely used in biocontrol of various kinds of parasitic nematodes, and several studies have shown its ability in controlling PWD 35,36 . In our current study, in vivo experiments under greenhouse conditions using the trunk injection method showed that the extracts of Streptomyces sp. AE170020 and abamectin successfully suppressed the development of PWD. To monitor the multiplication of nematodes in plants with or without the active substances, pine trees in each treatment were harvested and the number of nematodes was calculated. The compounds did not kill all the nematodes but signi cantly reduced the total number of nematodes compared with the nematode-control group (3426 ± 576 nematode per gram stem). These results indicate that these compounds prevented, or at least slowed, the beginning and progression of PWD. Although the two compounds isolated from different Streptomyces sp. than used herein have already been described and their biological activities have been studied, to our best knowledge, this is the rst report exploring the nematicidal activity of alloaureothin and applying strain Streptomyces sp. AE170020 to the biocontrol of PWD under pot conditions.
In addition to the external symptoms, the examination of other characteristics of pine trees after treatments allowed a better assessment of the physiological impact of nematode infection and compound injection. Speci c water content was used as a proxy for water stress, which is usually associated with PWD. A decrease in chlorophyll content was previously described in PWN-inoculated pine seedlings and is regarded as a symptom of the advanced stage of the PWD, which is induced by water de ciency in leaves 37 . Phenolic substances have been considered responsible for the browning of injured or pathogen-infected plant tissues, and PWD may be a result of the production of such compounds 38 . In the current study, acetone extract of strain Streptomyces sp. AE170020 prevented the reductions of water content, total chlorophyll, and total polyphenolics. Lipid peroxidation is an indicator of membrane cellular damage and usually occurs in plants after PWN inoculation as a result of xylem parenchyma cell necrosis and partial destruction of the cortex and phloem 39 . The increasing concentration of malondiadehyde in the nematode-control group suggested a high degree of lipid oxidative damage caused by PWN, and the active substances successfully protected pine trees from this damage.
Therefore, this study explored an endophytic strain, Streptomyces sp. AE170020, and its related compounds for defense against B. xylophilus in vitro and in vivo under greenhouse conditions. These results support the development of endophytic microorganisms as alternatives for the management of PWD. Further investigation is needed to understand the molecular mechanisms responsible for the nematicidal activity. In addition, scale-up, modi cation for this strain, safety evaluation, and risk assessment should be further developed to control B. xylophilus in actual agricultural applications. Figure 1 Phylogenetic analysis showing the relationship between strain Streptomyces sp. AE170020 and the related species. The tree was generated by neighbor-joining method of 16S rRNA gene sequences.

Declarations
Numbers of nodes are bootstrap values based on 1000 resampling.   Nematicidal activity of solvent-partitioned extracts of Streptomyces sp. AE170020 against B. xylophilus after 24 h treatment. All the data are shown as the mean ± SD (n = 9).
Page 23/27 Figure 5 Chemical structure of alloaureothin and aureothin  In uences of alloaureothin, aureothin and abamectin on the locomotor activity of B. xylophilus.
Nematodes were exposure to the compound at the concentration of 5.0 μg mL-1 (A) and 1.0 μg mL-1 (B), respectively. The graphs show the mean ± SD of two trials with six repetitions.