Relative potency of a novel acaricidal compound from Xenorhabdus, a bacterial genus mutualistically associated with entomopathogenic nematodes

Our study aimed to identify the novel acaricidal compound in Xenorhabdus szentirmaii and X. nematophila using the easyPACId approach (easy Promoter Activated Compound Identification). We determined the (1) effects of cell-free supernatant (CFS) obtained from mutant strains against T. urticae females, (2) CFS of the acaricidal bioactive strain of X. nematophila (pCEP_kan_XNC1_1711) against different biological stages of T. urticae, and females of predatory mites, Phytoseiulus persimilis and Neoseiulus californicus, (3) effects of the extracted acaricidal compound on different biological stages of T. urticae, and (4) cytotoxicity of the active substance. The results showed that xenocoumacin produced by X. nematophila was the bioactive acaricidal compound, whereas the acaricidal compound in X. szentirmaii was not determined. The CFS of X. nematophila (pCEP_kan_XNC1_1711) caused 100, 100, 97.3, and 98.1% mortality on larvae, protonymph, deutonymph and adult female of T. urticae at 7 dpa in petri dish experiments; and significantly reduced T. urticae population in pot experiments. However, the same CFS caused less than 36% mortality on the predatory mites at 7dpa. The mortality rates of extracted acaricidal compound (xenocoumacin) on the larva, protonymph, deutonymph and adult female of T. urticae were 100, 100, 97, 96% at 7 dpa. Cytotoxicity assay showed that IC50 value of xenocoumacin extract was 17.71 μg/ml after 48 h. The data of this study showed that xenocoumacin could potentially be used as bio-acaricide in the control of T. urticae; however, its efficacy in field experiments and its phytotoxicity need to be assessed in future.


Rearing of mites.
All mites used in the study are laboratory cultures previously identified based on morphological characteristics by Dr. Ibrahim Cakmak and used in previous studies 22,23 . Tetranychus urticae was obtained from strawberry plants in Aydin, Turkey. Bean plants which reached the 5-6 leaves were brought to the T. urticae rearing room and infested with different biological stages of the pest. The rearing of T. urticae was performed in another climate room with the same features as the plant growth room.
The predatory mites, P. persimilis and N. californicus, were obtained from bean plants in Hatay and strawberry plants in Aydin, respectively 6,23 . They were reared on detached bean leaves infested with all biological stages of T. urticae at 25 ± 1 °C temperature, 70 ± 10% relative humidity and 16 h light conditions in a third climate room. The detached bean leaves were placed on inverted pots in different size of two trays (45 × 32 × 8 cm; 78 × 56 × 18 cm). The trays were filled with water and covered with a plexiglass container to prevent the escape of the mites 26,27 . Three detached bean leaves infested with T. urticae were placed on each inverted pot three times a week to rear the predatory mites.

Identification of acaricidal bioactive compounds. Bacterial sources.
In the study carried out by Eroglu et al. 22 , Xenorhabdus szentirmaii and X. nematophila were determined as the species with the highest acaricidal activity among the CFS obtained from many tested Xenorhabdus and Photorhabdus spp. To identify Generation of deletion and promoter exchange mutants. The easyPACId approach (easy Promoter Activated Compound Identification) was used to identify the acaricidal compound(s) in X. szentirmaii and X. nematophila. The RNA chaperon, hfq, is directly associated with the production of natural products (NPs) as it controls the expression of biosynthetic gene clusters (BGCs) using the sRNA/mRNA interactions 28 . Therefore, ∆hfq mutants were generated in X. szentirmaii and X. nematophila to stop the biosynthesis of NPs. Subsequently activation of desired BGCs (Table 1) in a Δhfq background led to the nearly exclusive production of the corresponding NPs in Xenorhabdus strains following targeted BGC activation using the inducible promoter 24 . These methods can accelerate the identification of bioactive NPs by performing direct bioactivity tests without the need for purification of supernatants containing certain NPs 24,25,29 . Mutant strains of X. szentirmaii and X. nematophila with natural promoter regions replaced with inducible promoter regions were used in our study ( Table 1). The generation of X. szentirmaii Δhfq and X. nematophila Δhfq as well as promotor exchange mutants shown in Table 1 were performed as described by Tobias et al. 28,29 and Bode et al. 24 .
Preparation of bacterial supernatants of mutant strains. The 21-promoter exchange mutant strains (12 X. szentirmaii and 9 X. nematophila) listed in Table 1 were cultivated on LB agar, supplemented with a 50 μg/mL final concentration of kanamycin, and incubated for 48 h at 30°C 30 . A single colony was transferred into 10 ml LB medium, supplemented with a 50 μg/ml final concentration of kanamycin to obtain an overnight culture at 200 rpm and 30 °C. The optical densities of the overnight cultures (10 ml LB) were measured at 600 nm. The final OD of the cultures was adjusted to 0.1 after inoculation 100 ml Nutrient Broth (NB) 30 . For each strain, two flasks were prepared, and the cultures were incubated at 30 °C for 1 h. One of the flasks was induced with 0.2% L-arabinose (Carl Roth), and the other flask was not treated with L-arabinose (non-induced). All induced and non-induced cultures were incubated for 72 h at 200 rpm and 30 °C. The CFS was harvested by centrifugation at 10,000 rpm for 10 min, and the supernatant was filtered through a 0.22 μm millipore filter (Thermo scientific) 10,31 .
Determination of acaricidal compound/s using mutant strains. The effects of induced and non-induced CFS of mutant strains were tested on T. urticae adult females in Petri dishes. Experiments were carried out in a climate room (PG34 − 3 Digitech Ltd., Ankara, Turkey) at 25 ± 1 °C temperature, 70 ± 5% relative humidity and 16 h light conditions. Moistened cotton wool was placed on Petri dishes (15 cm in diameter) first, and then the bean leaf was placed with its bottom face up. The adult females of T. urticae were separately transferred with a fine brush in each Petri dish as 20 individuals. The CFS of mutant strains were sprayed on the leaves with a hand sprayer (2.5 ml/Petri dish). Sterile NB medium in which bacteria were grown was used as the control group. Mortality www.nature.com/scientificreports/ rates of mites were determined in 2, 5 and 7 days after the application (dpa). The experiments were carried out in 20 repetitions and repeated 4 times at different times.
The effect of the supernatant of induced mutant strain responsible from acaricidal activity against different biological stages of Tetranychus urticae. The acaricidal compound that causes high mortality on mites was determined and the gene region responsible for the production of the relevant bioactive compound in X. nematophila (pCEP_kan_XNC1_1711) was induced by L-arabinose, thus enabling the bacterium to produce an acaricidal bioactive compound only as a secondary metabolite. The effects of this CFS against the different biological stages of T. urticae were investigated in Petri dishes and pots.
Petri dish experiments. The effects of CFS of X. nematophila pCEP_kan_XNC1_1711 against different biological stages (egg, larva, protonymph, deutonymph, adult) of T. urticae were detected as previously described. Moistened cotton wool was placed on Petri dishes (15 cm in diameter) first, then the bean leaf was placed with its bottom face up. The egg, larva, protonymph, deutonymph and adult female of T. urticae were separately transferred with a fine brush in each Petri dish as 20 individuals. In order to obtain different biological stages of T. urticae at the same age to be used in the experiments, 25 gravid females of T. urticae were transferred on leaf discs. After 24 h, females were removed from the environment and the eggs remained. In this way, different biological stages (egg, larva, protonymph, deutonymph and adult female) of T. urticae were obtained at the same age. The CFS of mutant strain was sprayed on the leaves with a hand sprayer (2.5 ml/Petri dish). Sterile NB was used as the control group. Mortality rates of mites were determined in 2, 5 and 7 dpa. The experiments were carried out in 20 repetitions and repeated 4 times at different dates. The toxicity of the supernatant of mutant strain on predatory mites. The potential toxic effect of the CFS of X. nematophila pCEP_kan_XNC1_1711 on the predatory mites, P. persimilis and N. californicus, was investigated in Petri dishes at 25 ± 1 °C, 70 ± 5% R.H. and 16 h L:D photoperiod in a climate room. Moistened cotton wool (10 cm diameter) was placed on the Petri dishes (15 cm in diameter), and the gap between the Petri dish and cotton was filled with tap water to prevent the escape of the predatory mites. Adult females of P. persimilis and N. californicus (20 individuals/Petri dish) obtained from the culture were separately transferred with a fine brush on the leaves in the Petri dishes. Bean leaves infested with different biological stages of T. urticae (~ 300 individuals) were brushed onto the leaves at two-day intervals to feed the predatory mites. The CFS of mutant strain was sprayed with a hand sprayer on the leaves, and sterile NB was used as control. The mortality rate of the predatory mites in each Petri dishes was recorded at 2, 5 and 7 dpa. The study was carried out in 20 repetitions and repeated 4 times at different times.

The effects of the acaricidal extract on different biological stages of Tetranychus urticae.
Extraction of the acaricidal bioactive compound was performed as follows: Induced X. nematophila pCEP_kan_XNC1_1711 mutant strain was cultured in LB (6L) with 2% XAD resin at 30 °C for 3 days. The resin was extracted exhaustively with methanol (3 × 2 L) at room temperature. The methanol extract was concentrated under reduced pressure to give extracted compound 24 . The obtained extracted compound was first dissolved in DMSO and prepared as a stock solution with distilled water at a concentration of 208 μg/ml. Different dilutions (100%, 50%, 25%, 12.5%, 6.25% and 3.125%) of this prepared stock solution to determine LC 50 and LC 90 value of extracted compound were applied to T. urticae females in Petri dishes. Then, the activity of the LC 90 value of the acaricidal extracted compound on different biological stages (egg, larva, protonymph, deutonymph and adult female) of T. urticae was determined in Petri dishes. For these studies, moistened cotton wool was placed on Petri dishes (15 cm in diameter) first, then the bean leaf was placed with its bottom face up. 20 adult females of T. urticae were transferred to each petri dish with a fine brush. Different dilutions and the LC 90 value of the extracted compound were sprayed on the leaves with a hand sprayer (2.5 ml / petri dish). Sterile distilled water with DMSO was used as the control group. Mortality rates of mites were determined in 2, 5 and 7 dpa. The experiments were carried out in 10 repetitions and repeated 2 times.
Cytotoxicity of extracted bioactive acaricidal compound. Cytotoxicity assay was conducted using MRC-5 normal human fetal lung fibroblast cell-line. MRC-5 cells were obtained from the cell culture bank of the Turkish Ministry of Agriculture and Forestry (MRC-5 An 1 , HÜKÜK no: 96101701). The cells were maintained in Eagle's Minimum Essential Medium (EMEM) supplemented with 10% fetal bovine serum (Sigma-Aldrich) and 5% penicillin-streptomycin solution. The cells were cultured in tissue culture flask and incubated at 37 °C, 5% carbon dioxide and 96% humidity. The culture medium was replenished in 2 day-intervals. The cytotoxic effects of the extracted acaricidal compound of X. nematophila pCEP_kan_XNC1_1711 were measured in MRC-5 cell line using the MTT method. MRC-5 cells were treated with various concentrations of extracted

Results
Determination of acaricidal compound/s using mutant strains. The experiments conducted with mutant strains showed that xenocoumacin induced strain of X. nematophila (pCEP_kan_XNC1_1711) exhibited the highest acaricidal effect on T. urticae (Fig. 1). When the gene region responsible for the production of xenocoumacin was induced, the mortality rate of mites at 7 dpa was 100%, while the mortality rate of the noninduced xenocoumacin gene was less than 40%. None of the other induced or non-induced mutant strains of X. nematophila caused more than 50% mortality at 7 dpa. There was a statistically significant difference between xenocoumacin and all of the other tested compounds and the control group (F = 16.695, df = 18, P < 0.001) (Fig. 1). On the other hand, induced or non-induced 12 mutant strains of X. szentirmaii displayed acaricidal activity less than 50% (Fig. 2).

The effect of the supernatant of induced mutant strain responsible from acaricidal activity against different biological stages of Tetranychus urticae. Petri dish experiments. The study
showed that the CFS of X. nematophila (pCEP_kan_XNC1_1711) mutant strain had no effect on T. urticae eggs (ovicidal rate was 0%). The mortality rates on larva, protonymph, deutonymph and adult female of T. urticae were 100, 81, 44.9, and 43.1% at 2 dpa (Fig. 3). There was a statistical difference in mortality rates between dif- www.nature.com/scientificreports/ ferent biological stages of T. urticae at 2 dpa, and the highest mortality rate was found in larvae (F = 187,580; P < 0.001). The highest mortality was detected in larvae and protonymphs at 5 and 7 dpa, and the lowest was in deutonymphs and adults (5 dpa F = 24.417, P < 0.001; 7 dpa F = 4.694, P < 0.05; Fig. 3). The mortality rate in all biological stages of T. urticae was over 85% at 5 dpa and over 97% at 7 dpa (Fig. 3).

Discussion
When entomopathogenic nematodes infect an insect host, their symbiotic bacteria produce a wide variety of biologically active compounds with a broad-spectrum activity to protect infected cadaver from opportunistic organisms and scavengers such as ants, crickets, cockroaches, mites etc. 38  www.nature.com/scientificreports/ EPN-infected insect cadavers and on the developing EPN IJs herein [41][42][43][44][45][46] . To protect infected cadaver and developing nematodes from mites, nematode-bacteria complex has to produce bioactive acaricidal compound/s. Accordingly, numerous studies have shown that some species of Xenorhabdus bacteria have acaricidal activity [19][20][21][22][23] . However, none of these studies identified the bioactive acaricidal compound. Therefore, the aim of this study was to establish the acaricidal activities present in X. nematophila supernatant using the easyPACId biotechnological approach. This biotechnological approach allowed us to determine the bioactive compound by activating mutants with inducible promotors of encoding gene clusters and eliminating the background effect of genes of other compounds 24,25 . The experiments conducted with promoter exchange mutant strains showed that xenocoumacin induced strain of X. nematophila (pCEP_kan_XNC1_1711) exhibited the highest acaricidal effect on T. urticae.
Xenocoumacins are benzopyran-1-one (isocoumarin) derivatives first identified by Mclnernery et al. 47 in X. nematophila in two forms. Reimer et al. 15 later discovered 4 additional derivatives of these natural products system from several Xenorhabdus strains and reported that they are synthesized by a hybrid polyketide synthase (PKS)-nonribosomal polypeptide synthetase (NRPS). Both forms have many biological activities such as antifungal, antibacterial, anticancer and anti-ulcer however, xenocoumacin 1 is more biologically active 11,48,49 .
On the other hand, induced or non-induced 12 mutant strains of X. szentirmaii displayed acaricidal activity less than 50%. A large-scale genome and metabolome analysis of 25 Xenorhabdus strains by Tobias et al. 29 revealed that X. szentirmaii DSM 16,338 and US strains do not produce xenocoumacin. So, the acaricidal compound must be a different compound than xenocoumacin. Our collection of mutant strains from X. szentirmaii in our study was limited. Further studies should be conducted with different promotor exchange mutants of X. szentirmaii.
We assessed the acaricidal effects of CFS of X. nematophila (pCEP_kan_XNC1_1711) against all biological stages of an important argonomic pest, T. urticae. First, we showed that the mobile stages of T. urticae were affected at different levels by the CFS of X. nematophila (pCEP_kan_XNC1_1711) mutant strain and xenocoumacin extract. Larval stages were more susceptible compared to adult female in Petri dish experiments, though the mortality rate in all biological stages of T. urticae was over 97% at 7 dpa. Similarly, Eroglu et al. 22 found that female adults were relatively more tolerant to the supernatants of X. nematophila wildtype than larval and nymph stages as the supernatant exhibited 90% mortality on adult females and 98% mortality on the larvae of T. urticae at 7 dpa. The LC 50 values of xenocoumacin extract against T. urticae adult females in our study for 2, 5 and 7 dpa were 60, 26, 21 µg/ml, respectively. Comparatively, Furuya et al. 50 reported that a novel acaricidal compound, pyflubumide, had a LC 50 value of 1.2 mg a.i./L against adult twospotted spider mites. The LC 50 for cyflumetofen against T. urticae female adults as reported in Hayashi et al. 51 was 1.1 mg/L.
Besides petri dish experiments, the results of our pot experiment showed that CFS of X. nematophila (pCEP_ kan_XNC1_1711) mutant strains significantly reduced the T. urticae population. Likewise, Eroglu et al. 22 showed that the supernatants from wildtypes of X. szentirmaii and X. nematophila, singularly and in combination, significantly reduced the T. urticae population in pot experiment.
We also tested the effects of CFS of X. nematophila (pCEP_kan_XNC1_1711) mutant strains against eggs of T. urticae. We found that xenocoumacin had no effect on T. urticae eggs (ovicidal rate was 0%). Generally, mite eggs have been observed to be resistant to acaricide 52 , supernatants of Xenorhabdus and Photorhabdus bacteria 22,23 or infection from entomopathogenic fungi 52 .
Tetranychus urticae is the most resistant species among arthropod pests in the world as it has gained resistance to 96 currently available active ingredients 53 . Hence, predatory mites like P. persimilis and N. californicus are widely used as alternatives to control T. urticae populations. An ideal acaricidal compound should kill T. urticae and have minimum side effects on these predatory mites. Our study also evaluated the toxicity of CFS of X. nematophila (pCEP_kan_XNC1_1711) against the adult females of P. persimilis and N. californicus. Although, the CFS of X. nematophila (pCEP_kan_XNC1_1711) or xenocoumacin XAD extract caused over 90% mortality on the adult female of T. urticae, less than 40% mortality of both predatory mites were affected at 7dpa. Morphological differences between predatory mites and T. urticae have a key role in the different susceptibility of the mites to bacterial supernatants. For instance, T. urticae feeds on treated leaves and have shorter legs compared to predatory mites, their body parts are more in direct contact with applied compounds 23 . Besides this, predatory mites have a thicker cuticle than of T. urticae 54 .
Cytotoxicity assays revealed that xenocoumacin compound is not toxic on human cells when it is used at concentrations < 17.71 μg/ml. Bode et al. 24 tested the effect of aqueous extract of xenocoumacin obtained from X. nematophila on the human microvascular endothelial cell line. Except for toxicity on cell proliferation, xenocoumacin extract displayed very low effect on the cell metabolic activity. Cytotoxicity of xenocoumacin was moderate and leucocyte adhesion to endothelial cell was low. We found that the LC 50 values of xenocoumacin extract against T. urticae adult females in our study ranged between 21-60 µg/ml during the 7 days of assessment. However, this is higher than cytotoxicity against human cells. Future studies should assess the persistence of this compound on plant tissues.
In conclusion, the data of this study showed that xenocoumacins could potentially be used as bio-acaricides in the control of T. urticae at concentrations less than 17 μg/ml, however, the efficacy of xenocoumacin in the field experiment and its phytotoxicity need to be assessed in future.