The entomopathogenic bacterium Xenorhabdus nematophila is mutualistic to a nematode, Steinernema carpocapsae . The infective juveniles (IJs) of S. carpocapsae enter the hemocoel of target insects through natural openings, such as the mouth, anus, or spiracle . In the hemocoel, IJs release X. nematophila from their gut and suppress target insect immunity . After the bacterial growth, the host nematodes begin to grow and reproduce in the insect cadaver after the bacterial septicemia . The multiplied nematodes become the IJs and reassociate with X. nematophila and come out of the insect cadavers to look for other target insects .
Insects defend the bacterial infection with their innate immune responses . Insect immunity is divided into cellular and humoral responses. Cellular immunity is the response mostly exhibited by hemocytes, such as phagocytosis and nodulation in response to bacterial infection . Humoral immunity consists of the chemical reactions of the antimicrobial peptides depending on bacterial cell wall patterns . The acute antibacterial responses are executed by cellular immunity . The successful pathogenity of X. nematophila is partially explained by its inhibitory activity against the insect immunity. Lipopolysaccharide of X. nematophila inhibits phenoloxidase activity and is cytotoxic to insect hemocytes [10, 11]. Indeed, a toxin (α-Xenorhabdolysis) of X. nematophila induces apoptosis in both insect and mammalian cells . In terms of immunosuppression by X. nematophila, Park and Kim  proposed a hypothesis that the bacteria inhibit eicosanoid biosynthesis, because it has been well known that eicosanoids play crucial roles in reducing insect immune responses . Eicosanoids are a group of C20 oxygenated and polyunsaturated fatty acids . Specifically, eicosanoid derived from both cyclooxygenase (COX) and lipoxygenase (LOX) products mediate insect immunity signals to induce both cellular and humoral responses . In eicosanoid biosynthesis, arachidonic acid (AA) release from a phospolipid is a crucial committed step and then is catalyzed by phospholipase A2 (PLA2), because AA is the major substrate for COX and LOX . Both entomopathogenic bacterial groups, Xenorhabdus and Photorhabdus, inhibit PLA2 to suppress eicosanoid biosynthesis and immune responses of target insects . Indeed, Seo et al.  identified seven PLA2 inhibitiors from the bacterial culture broth of X. nematophila.
A “Bt-Plus” insecticide has been developed to enhance the insecticidal activity of Bacillus thuringiensis (Bt) by suppressing target insect immunity by the culture broth of X. nematophila . The synergistic effect of Bt and X. nematophila on the insecticidal activity was initially confirmed by their independent targets of the midgut and the hemocoel, in which the midgut epithelium was disintegrated by Bt and the hemocoel was exposed to X. nematophila . Then, X. nematophila suppresses the target insect immunity against the invasion of Bt and other microbes, which ultimately enhances the Bt pathogenicity [27, 52]. Similar synergism of X. nematophila to Bt was also confirmed in the Mediterranean flour moth, Ephestia kuehniella, in which only Cry toxins with high binding affinity to the insect midgut epithelium were effective to enhance the Bt toxicity . In the meantime, a variant of X. nematophila was found in the successively cultured colonies. Our current variant no longer absorbs bromothymol blue dye and exhibits a low insecticidal activity at hemocoelic injection. Owing to a wellknown phase variation in X. nematophila , the variant was suspected to be a secondary form of X. nematophila. This study compared the primary form (newly isolated bacteria from an entomopathogenic nematode, S. carpocapsae) with the secondary form in terms of pathogenicity, PLA2 inhibitor biosynthesis, and immunosuppression. Then this study developed a highly efficient Bt-Plus insecticide against both Plutella xylostella and Spodoptera exigua, by mixing two effective Bt strains and the culture broth of the primary form of X. nematophila.
Materials and Methods
S. exigua larvae were reared in a laboratory under the conditions of 25 ± 1℃ and 16 h of light : 8 h of darkness. The larvae were fed with artificial diet , whereas adults were fed 10% sucrose solution. The final (fifth) instar larvae were used for bioassay and hemolymph collection in this study. P. xylostella larvae were reared with cabbage at the same rearing environment. The final (fourth) instar larvae were used for the insecticide bioassay.
Isolation of the Primary Form of X. nematophila
Infective nematodes (IJs) of S. carpocapsae Pochon were topically applied on the fifth instar larvae of S. exigua and incubated at 25℃ for 48 h. The infected hosts were surface-sterilized with 70% ethanol, and the first abdominal prolegs were cut by a pair of sterile scissors to collect the exuded hemolymph. The hemolymph was diluted with an equal volume of sterilized water and streaked on tryptic soy agar (Difco, Sparks, MD, USA) to isolate bacteria. The isolated bacteria were cultured in Luria-Bertani medium (LB, Difco) at 25℃ for 48 h. After washing the cultured cells three times with sterilized water by centrifuging the culture medium at 4,000 ×g for 2 min at 4℃, the cells were resuspended in sterilized insect Ringer’s solution  for further experiments. Another primary form of X. nematophila was obtained from the American Type Culture Collection (ATCC 101061; Manassas, VA, USA). The secondary form analyzed in this study was originally isolated from S. carpocapsae  and successively cultured in vitro.
Biolog Identification System
Isolates were prepared according to the manufacturer’s instruction in the OmniLog ID System User Guide (Biolog, Hayward, CA, USA). Each isolate was cultured on tryptic soy broth (TSB) at 28℃ for 48 h. Each cultured isolate was emulsified to the specific density in the inoculation fluid (0.40% sodium chloride, 0.03% Pluronic F-68, and 0.02% Gellan Gum). Each well of the GN Microplate (Biolog) was subsequently inoculated with 150 μl of the bacterial suspension and incubated at 28℃ for 24 h. Each metabolic profile was compared with the appropriate GN Omnilog Biolog database (Biolog) and used to identify the bacterial species.
HPLC Analysis of PLA2 Inhibitor from the Bacterial Culture Broth
For analysis of bacterial metabolites using HPLC, 1 L of X. nematophila was cultured in LB for 48 h at 28℃. After removing bacterial cells by centrifugation at 8,000 rpm for 20 min, the supernatant was mixed with the same volume of hexane in a separate funnel. The organic extract was concentrated using an evaporator. The dried extract was resuspended with methanol and analyzed by HPLC (Waters, Milford, MA, USA). Samples were cleaned with a PTFE syringe filter (Cronus, Churcham, UK). Ten milliliters of the cleaned sample was injected to an HPLC equipped with a C18 column (Deltapak, 15 mm, 300 A, 300 × 7.8 mm). The samples were then separated with a mobile phase of methanol:water (60:40 (v/v)) at a flow rate of 0.5 ml/min for 30 min with a UV detector (Waters, Milford, MA, USA) at 254 nm. The eight bacterial metabolites included oxindole, indole, p-hydroxypropionic acid (PHPP), cyclo-Pro-Tyr (cPY), 4-hydroxyphenylacetic acid (HPA), benzylideneacetone (BZA), Pro-Tyr (PY), and acetylated Phe-Gly-Val (Ac-FGV).
Bioassay of X. nematophila by Hemocoelic Injection
Test bacteria of X. nematophila were cultured as described above. Final instar larvae (fourth instar of P. xylostella and fifth instar of S. exigua) were surface-sterilized with 70% ethanol. The bacterial culture medium (103-105 CFU/ml) was injected into the insects by Hamilton microsyringe (Hamilton, Reno, Nevada, USA). Each treatment consisted of three replicates, and each replicate used 30 insect larvae. Control larvae were injected with sterile Ringer’s solution. Mortality was assessed at 24 h after the bacterial injection. Dead larvae were determined by no voluntary movement after prodding test larvae with a blunt stick.
Bioassay of “Bt-Plus”
Test solutions of Bt-Plus were prepared by mixing Bt and X. nematophila culture broths. Bacterial culture broth (48 h at 28℃ in TSB) of X. nematophila was serially diluted with the uncultured fresh TSB medium containing 1,000 μg/ml of Bt. Bt insecticides used in this study were B. thuringiensis ssp. kurstaki (Btk) and B. thuringiensis ssp. aizawai (Bta) used in the earlier studies [53, 54]. The Bt suspensions were kept at 4℃ for 2 days to form spores. Finally, the spore densities of Btk and Bta were 7.4 × 1010 and 8.7 × 108 CFU/ml, respectively. Cabbage leaves were soaked for 10 min in different Bt-Plus suspensions and dried under darkness for 30 min. Each treatment was replicated three times. Each replication used 10 larvae of P. xylostella (fourth instar) or S. exigua (fifth instar). At every 24 h, the number of the live individuals was counted for 7 days.
Bioassay of “Dual Bt-Plus”
For Dual Bt-Plus, two Bt strains were mixed before adding to the culture broth of X. nematophila. Two Bt strains were mixed with a 4:1 (v/v) ratio of Bta and Btk. The first type of Dual Bt-Plus (“Bt-Plus+oxindole” or “Bt-Plus+BZA”) was prepared by the additional oxindole or BZA at 100 ppm to the Bt-Plus spraying solution (0.5% Bt mixture, 0.4% X. nematophila culture broth, and 0.1% ethanol). The other type of Dual Bt-Plus was prepared with the Bt-Plus spraying solution including a 100-fold high concentration of X. nematophila by addition of the bacterial culture broth (0.5% Bt mixture, 40% X. nematophila culture, and 0.1% ethanol). The bioassay followed the method as described above.
Phenoloxidase (PO) Activity
Hemolymph PO activity was determined using L-3,4-dihydroxyphenylalanine (DOPA) as a substrate . Hemolymph was collected into a 1.5 ml tube by cutting the abdominal proleg. The sample solution consisted of 1 μl of test material with 9 μl of hemolymph. The PO substrate solution prepared had 990 μl of 100 mM phosphate-buffered saline (0.7% NaCl, pH 7.4) containing 20 μg/μl of DOPA in acetone. The PO reaction was initiated by adding the sample solution to the PO substrate solution. The initial absorbance change was monitored at 495 nm in a spectrophotometer.
Hemocyte PLA2 activity was fluorometrically determined with a substrate, the pyrene-labeled phospholipids [1-hexadecanoyl-2-(1-pyrenedecanoyl)-sn-glycerol-3-phosphatidyl choline] in the presence of bovine serum albumin (BSA), as described by Radvanyi et al. . The fluorescent phospholipid was prepared at 0.2 mM concentration using ethanol. For preparation of enzyme source, hemolymph was collected into a 1.5 ml tube containing a few granules of phenylthiourea and centrifuged at 400 ×g for 3 min. The plasma was removed and washed three times with washing buffer (50 mM Tris-HCl, pH 7.0, 100 mM NaCl, 1 mM EDTA). The hemocyte pellet was resuspended in the washing buffer and homogenized by an ultrasonicator (Bandelin Sonoplus, Berlin, Germany) for 10 min at 3 cycles and 75% power. Protein concentrations of hemocyte extracts were measured by the Bradford method  using BSA as a standard. The reaction mixture was prepared in a 96-well microplate by adding 142.5 μl of Tris buffer, 1 μl of 1 M CaCl2, 1 μl of 10% BSA, and 2 μl of 0.2 mM substrate. The sample solution consisted of 1 μl of test material with 1 μl of hemocyte enzyme extract and was incubated for 20 min at room temperature. The PLA2 reaction was initiated by adding the sample solution (2 μl) to the reaction mixture (146.5 μl). The fluorescence intensity was monitored with an Aminco Bowmen Series 2 luminescence spectrometer (FA257; Spectronic Instruments, Madison, WI, USA) using excitation and emission wavelengths of 345 and 394 nm, respectively. The enzyme activity was calculated in pmol/min using a formula provided by Radvanyi et al. .
The nodulation assay was performed by injecting 104 cells of Escherichia coli Top10 (Invitrogen, Carlsbad, CA, USA) in 2 μl, through the abdominal proleg, using a microsyringe, into hemocoel of S. exigua as previously described . After 8 h incubation at room temperature, test insects were dissected and counted in melanized nodules under a stereoscopic microscope (SZX9; Olympus, Tokyo, Japan). For the nodulation inhibition assay, 2 μl of different X. nematophila bacterial concentrations was injected into each larva after E. coli injection, and nodules were counted as described above.
Survival data were transformed by the square root and arcsine method for normalization. Treatment means and variances of the transformed data were analyzed by PROC GLM of the SAS program .
Comparison of Primary and Secondary Forms of X. nematophila in Physiological and Biochemical Characters
A laboratory colony (“Xnk2”) of X. nematophila exhibited a low entomopathogenic activity. A hemocoelic injection of 103 CFU of X. nematophila gave only 60-70% mortality to S. exigua larvae at 24 h. In a previous study, the same dose of X. nematophila would kill 100% of the identical age of S. exigua . To explain this decrease of insecticidal activity of X. nematophila, we needed to compare the laboratory colony with a primary form of X. nematophila. One was to use a primary type strain (“ATCC” strain) of X. nematophila. The other was to isolate a new X. nematophila (“Xnk1”) from the hemolymph of the larvae infected with the nematode IJs. In a similar way to the ATCC strain, Xnk1 exhibited a blue colony on NBTA medium due to absorption of bromothymol blue dye. However, the Xnk2 colony did not show the blue colony. To compare the biochemical properties of carbon source utility of the two forms of X. nematophila, the bacteria were cultured on Biolog plate and the outcomes were compared (Table 1). Xnk1 resulted in 99% identity with respect to the carbon utility characters of the canonical X. nematophila described in the Biolog database system. However, Xnk2 characters showed only 54% identity with those of X. nematophila, even though the 16S rDNA sequence of XnK2 perfectly (100%) matched to that of the X. nematophila K1 strain  originally isolated from the nematode host.
Table 1.1Details of X. nematophila are described in Bergey’s Manual . 2A primary form of X. nematophila isolated from the host nematode, Steinernema carpocapsae. 3A secondary form of X. nematophila after successive in vitro cultures. The Biolog microbial identification system was used to monitor the carbon usage.
Comparison of Primary and Secondary Forms of X. nematophila in PLA2 Inhibitor Production
Eight bacterial secondary metabolites of X. nematophila were reported to inhibit insect PLA2 [24, 54]. All compounds were extracted in hexane extracts (Fig. 1). In both primary forms of Xnk1 and ATCC strains of X. nematophila, all eight metabolites were detected; however, the secondary form of Xnk2 produced the compounds at significantly lower quantities than the two primary forms, and oxindole and BZA were not detected in the secondary form (Table 2). There was little difference between the two primary forms in the production of eight secondary metabolites.
Fig. 1.Chromatograms of three different culture broths of Xenorhabdus nematophila (Xn). “Xn1” and “Xn2” represent primary and secondary forms, respectively. “ATCC” represents an isolate of X. nematophila obtained from the American Type Culture Collection (Manassas, VA, USA). The bacteria (5 × 103 CFU) were inoculated to 1 L of TSB culture media and cultured at 28℃ for 48 h. Hexane extracts of the culture broth were separated with a reverse-phase C18 HPLC column and analyzed with a UV detector as described in Materials and Methods. Peaks are (1) PHPP, (2) PY, (3) Ac-FGV, (4) cPY, (5) oxindole, (6) HPA, (7) indole, and (8) BZA.
Table 2.1”Xnk1” is a primary form of X. nematophila isolated from the host nematode, Steinernema carpocapsae. “Xnk2” is a secondary form of X. nematophila after successive in vitro cultures. “ATCC” is a primary form of X. nematophila obtained from the American Type Culture Collection. 2Different letters following standard deviations indicate significance among means at Type I error = 0.05 (LSD test).
Comparison of Primary and Secondary Forms of X. nematophila in Immunosuppression and Entomopathogenicity
The decrease of PLA2 inhibitor synthesis in the secondary form of X. nematophila suggested its reduced activities against inhibition of the immunity-associated enzymes or a cellular immunity. First, the PLA2 activity of the infected larvae was analyzed in the fat body and hemocytes (Fig. 2A). The larvae infected with Xnk2 showed significantly higher PLA2 activity (F = 15.39; df = 2, 24; p = 0 .0001) , compared with larvae infected with Xnk1 or ATCC strains. Second, PO activity in the hemolymph was significantly higher in larvae infected with Xnk2 than those infected with Xnk1 or ATCC (Fig. 2B). Third, a significantly greater number of hemocyte nodules (F = 21.1; df = 2, 24; p = 0.0001) was formed in the larvae treated with Xnk2 than those of Xnk1 or ATCC (Fig. 2C). This indicated that XnK2 did not inhibit insect immunity as much as the two primary forms did.
Fig. 2.Reduced activity of the secondary form (Xn2) of Xenorhabdus nematophila on three immune-associated responses of Spodoptera exigua, compared with those of two primary forms (Xn1 and ATCC). (A) For the PLA2 enzyme activity assay, fifth instar larvae of S. exigua were injected with different numbers of bacteria. After 20 min, the hemolymph was collected and analyzed for enzyme activity using pyrene-labeled phospholipid. Each treatment was replicated three times. (B) For the phenoloxidase (PO) activity assay, fifth instar larvae were injected with different numbers of bacteria. After 15 min, the hemolymph was collected and assessed in the PO activity using DOPA as a substrate. Each treatment was replicated three times. (C) For the nodulation assay, different numbers of X. nematophila were injected to the larvae along with Escherichia coli (2 × 104 cells/larva). After 8 h at 25℃, the nodules were counted in the hemocoel. Control represents PBS injection with E. coli. Each measurement used five larvae.
Fig. 3.Reduced pathogenicity of the secondary form (Xn2) of Xenorhabdus nematophila on the larvae of Spodoptera exigua, compared with those of the two primary forms (Xn1 and ATCC). Different bacterial numbers were injected to fourth instar Plutella xylostella (A) or fifth instar Spodoptera exigua (B) larvae. After 24 h at 25℃, the mortality was assessed. Each treatment was replicated three times. Each replication used 10 larvae.
The pathogenic activities of the different X. nematophila strains were analyzed by hemocoelic injection (Fig. 3). In both target insects, Xnk2 was significantly lower in pathogenic activity (F = 19.32; df = 2, 24; p = 0.0001) than Xnk1 or ATCC strains in all treated doses in P. xylostella (Fig. 3A) and S. exigua (Fig. 3B). Moreover, the rate of lethal effect of Xnk2 progressed much slowly compared with those of both primary forms in the same infection dose (103 CFU dose).
Primary Form of X. nematophila Significantly Increases Bt Pathogenicity
The culture broth of X. nematophila contains the immunosuppressive metabolites, which enhance Bt pathogenicity . A recent production of a commercial Bt-Plus (Lepkill) containing both X. nematophila and Bt suffered a low control efficacy, presumably due to a low synergistic effect of X. nematophila on Bt pathogenicity. All current data suggested that the low efficacy might be the occurrence of a phase variant of X. nematophila. Clearly, use of the primary form of X. nematophila significantly enhanced Bt efficacy (Fig. 4). The enhanced efficacy was specific to target insects depending on the Bt strain (Fig. 4A). B. thuringiensis subsp. aizawai (Bta) was highly effective only against S. exigua when it was mixed with Xnk1 (Fig. 4B). On the other hand, B. thuringiensis subsp. kurstaki (Btk) was highly effective only against P. xylostella when it was mixed with Xnk1. When both Bt strains (1:1 = Bta: Btk (v/v)) were combined with the culture broth of Xnk1, the mixture was much more effective against both insect species, but much weaker in its efficacy than the single Bt-containing Bt-Plus (Fig. 4C). For example, in P. xylostella, the Bt mixture Bt-Plus (XnK1+Btk+Bta) was more potent than XnK1+Bta, but not as much as XnK1+Btk. Likewise, in S. exigua, the Bt mixture Bt-Plus (XnK1+Btk+Bta) was more potent than XnK1+Btk, but not as much as XnK1+Bta. This raised a further analysis to determine the optimal mixture ratio of both Bt strains. A screening test to determine the Bt ratio in Bt-Plus was conducted (Fig. 5). The most effective mixture was a 4:1 ratio of Bta and Btk to effectively control both lepidopteran species. We called the Bt-Plus using Bt mixture as “Dual Bt-Plus”.
Fig. 4.Enhanced insecticidal effect of Bt-Plus using the primary form of Xenorhabdus nematophila. Two different Bt strains, B. thuringiensis ssp. kurstaki (Btk) and B. thuringiensis ssp. aizawai (Bta), were assessed against fourth instar Plutella xylostella or fifth instar Spodoptera exigua larvae. Each treatment was replicated three times, at which each replication used 10 larvae. Mortality was assessed at 7 days after treatment. (A) Insecticidal activities of Bt alone. (B) Synergistic effect of X. nematophila on Bt toxicity. Bt-Plus (XnK1+Btk or XnK1+Bta) was prepared by mixing two bacterial culture broths in 1:1 ratio. (C) Insecticidal activity of a Bt mixture in Bt-Plus (XnK1+Btk+Bta) on both insect species. Different letters above standard deviation bars indicate significant difference among means at Type I error = 0.05 (LSD test).
Fig. 5.Determination of an optimal mixture ratio of Bt mixture in Bt-Plus to control both lepidopteran species of fourth instar Plutella xylostella and fifth instar Spodoptera exigua larvae. Culture broths of B. thuringiensis ssp. kurstaki (Btk) and B. thuringiensis ssp. aizawai (Bta) were mixed in different ratios and added to the culture broth of Xenorhabdus nematophila (Xn). A total Bt-Plus consisted of Bt:Xn:ethanol = 5:4:1 (v/v). Each treatment was replicated three times, at which each replication used 10 larvae. Mortality was assessed at 7 days after treatment. Different letters above standard deviation bars indicate significant difference among means at Type I error = 0.05 (LSD test).
Addition of BZA or Oxindole Significantly Increases the Insecticidal Activity of Bt-Plus
To further increase the insecticidal activity of Bt-Plus, the effects of the additional bacterial metabolites of X. nematophila were assessed (Fig. 6). Among eight metabolites, BZA and oxindole were the most effective to increase the insecticidal activities in both lepidopteran species.
These two metabolites were then individually mixed with the Bt-Plus (Fig. 7). Both metabolites significantly increased the insecticidal activities of the Bt-Plus against both S. exigua and P. xylostella larvae. To confirm the enhanced insecticidal activity of the Dual Bt-Plus fortified with BZA or oxindole, a pot assay was conducted (Table 3). A complete control efficacy was obtained in both insect pests within 4 days.
Fig. 6.Effects of the additional bacterial metabolites on Bt toxicity to control fourth instar Plutella xylostella or fifth instar Spodoptera exigua larvae. Two different Bt strains, B. thuringiensis ssp. kurstaki (Btk) and B. thuringiensis ssp. aizawai (Bta), were used to test the bacterial metabolites in P. xylostella and S. exigua, respectively. Each treatment was replicated three times, at which each replication used 10 larvae. Mortality was assessed at 7 days after treatment.
Fig. 7.Effects of two bacterial metabolites on Bt-Plus toxicity to control fourth instar Plutella xylostella or fifth instar Spodoptera exigua larvae. Culture broths of B. thuringiensis ssp. kurstaki (Btk) and B. thuringiensis ssp. aizawai (Bta) were mixed at 1:4 (Bta:Btk (v/v)) and added to the culture broth of Xenorhabdus nematophila (Xn). A total Bt-Plus consisted of Bt:Xn:ethanol = 5:4:1 (v/v). Each 100 ppm of oxindole or BZA was added to the spray suspension of the Bt-Plus in Bt-Plus+oxindole or Bt-Plus+BZA. Each treatment was replicated three times, at which each replication used 10 larvae. Mortality was assessed at 3 and 7 days after treatment (DAT).
Table 3.1Average initial density per replication. Three cabbage plants in a pot were infested with late instar larvae. Each treatment was replicated with three pots. Mortality was assessed at 3 and 7 days after treatment (DAT). Different letters represent significant difference in mean survivals of different treatments at Type I error = 0.05 (Multiple mean range test: DMRT).
Dual Bt-Plus Using the Concentrated Culture Broth of X. nematophila
The fact that the addition of BZA or oxindole significantly increased the insecticidal activity of the Dual Bt-Plus suggested that the concentrated culture broth of X. nematophila might be effective because it contained the bacterial metabolites. To test this hypothesis, the culture broth of X. nematophila was concentrated 100-fold and used to prepare the Dual Bt-Plus (Fig. 8). Like the metabolite-fortified Bt-Plus (Bt-Plus+oxindole or Bt-Plus+BZA), Dual Bt-Plus using the concentrated culture broth of X. nematophila significantly increased the insecticidal activity compared with Bt-Plus. Analysis of bacterial metabolites supported the increases of both BZA and oxindole contents in the Dual Bt-Plus along with other metabolites, in which most metabolites were detected at 5-10 times increase in Dual Bt-Plus compared with those of Bt-Plus (Table 4). The far less amounts of the bacterial metabolites in the Dual Bt-Plus, even though it used 100-concentrated X. nematophila, might be the solubility limits of the nonpolar bacterial metabolites in the spraying suspension.
Fig. 8.Development of Dual Bt-Plus by mixing different bacterial culture broths to control fourth instar Plutella xylostella or fifth instar Spodoptera exigua larvae. Culture broths of B. thuringiensis ssp. kurstaki (Btk) and B. thuringiensis ssp. aizawai (Bta) were mixed and added to the culture broth of Xenorhabdus nematophila (Xn). Bt-Plus consisted of a mixture of Bta:Btk:Xn:ethanol = 4:1:4:1 (v/v). Bt-Plus+oxindole or Bt-Plus+BZA was prepared by adding 100 ppm metabolite to the spray suspension of the Bt-Plus. Dual Bt-Plus consisted of a mixture of Bta:Btk:Xn:ethanol = 4:1:4:1 (v/v), in which Xn culture broth was 100-fold concentrated. Each treatment was replicated three times, at which each replication used 10 larvae. Mortality was assessed at 7 days after treatment. Different letters above standard deviation bars indicate significant difference among means at Type I error = 0.05 (LSD test).
Table 4.1Different letters following standard deviations indicate significance among means at Type I error = 0.05 (LSD test).
The significance of this study was to develop a novel biopesticide using Bt-Plus containing Bt and X. nematophila metabolites to simultaneously control two different lepidopteran insect pests P. xylostella and S. exigua, which infest the common crop plants in field. To understand the Bt-Plus efficacy and its development to control both insect targets, Bt pathogenicity and resistance mechanisms need to be described. Bt has been widely used as a biopesticide to control lepidopteran, coleopteran, and dipteran insect pests. Bt Cry toxins are used to breed genetically modified crops, such as Bt-corn, Bt-soybean, and Bt-cotton, which drastically increase in cultivating areas over the world . As like other insecticides, target insects develop resistance against various Bt strains in the field as well as in laboratories . The major resistance mechanism is derived from the interaction between Bt Cry toxins and insect midgut epithelium . When Bt spores are consumed, they germinate and the crystals associated with spores are dissolved in the alkaline juice in the midgut lumen. The protoxin is then cleaved by insect digestive proteases to produce the active, protease-insensitive toxin core protein. The active toxins migrate to the ectoperitrophic space through the peritrophic membrane and reversibly or irreversibly interact with molecules in the epithelial membrane. Aggregation of toxins into oligomers happens and disrupts the epithelial membrane by pore formation, which results in cell lysis called “colloid-osmotic lysis” . The damaged midgut epithelial membranes cause a complete cessation of feeding behavior or a fatal proliferation of Bt and other microorganisms to kill insects .
Bt resistance can be induced by interruption of any of the steps of Bt pathogenicity described above. A Bt resistance may be derived from interruption of toxin activation from the inactive para-crystal form. This resistance mechanism includes overexpression of proteases to degrade the protoxins , sequestering the toxins by precipitation  or coagulation  or trapping in the peritrophic matrix. Modification of the Bt toxin-binding sites in the epithelial membrane reduces or prevents the irreversible binding, which is crucial to Bt toxicity [13, 65]. Furthermore, pore formation can be interfered with or pores can be plugged . Dead midgut cells can be replaced by the increase of stem cell proliferation . Finally, the elevated immune response can effectively defend Bt and other enteric bacterial infection to the hemocoel . The role of the elevated immune responses has been demonstrated by increase of melanization immune response in Ephestia kuehniella and Helicoverpa armigera that exhibited 16-fold and 12-fold greater tolerances to Cry toxins [40, 48]. In S. exigua, the Bt-resistant strain elevated the gene expression associated with immune responses . The significant increase of Bt toxicity in Bt-Plus can be explained by the immunosuppression induced by PLA2-inhibitory metabolites of X. nematophila. All eight metabolites are potent to inhibit PLA2 activity to prevent the production of eicosanoids, which play crucial roles in mediating insect immunity . Furthermore, these eight metabolites are effective to inhibit PO activity, which is crucial to form coagulation of Bt Cry toxins in the midgut lumen . Thus, the metabolites of X. nematophila in Bt-Plus suppress target insect immunity by inhibiting PLA2 and PO to facilitate Bt toxin activation and induce the fatal septicemia in the target insect hemocoel.
A phase variation is well documented in X. nematophila [14, 59]. Initially, Akhurst  demonstrated that X. nematophila produces two colony types on agar media. The primary (phase I) form, isolated from the nematode intestine, is unstable in vitro and produces the secondary (phase II) form, which is well adapted to grow in vitro . The common characters of the primary form include motility, dye absorption, antibiotics production, hemagglutination, and synthesis of the outer membrane protein, OpnB [16, 17, 67]. However, the secondary form characters of X. nematophila are variable depending on strains . Our secondary form was distinct in bromophenol blue dye absorption, insect pathogenicity, and the production of the secondary metabolites, compared with those of the primary form isolated from the nematode S. carpocapsae. The decreased pathogenicity of the secondary form is partially explained by the decrease in the production of the bacterial metabolites, which are potent to inhibit insect immune responses . However, both primary and secondary forms of X. nematophila are equally pathogenic to Galleria mellonella  and functional in the in vitro culture of host nematodes . On the other hand, the decrease of insecticidal activity is observed from a secondary form of X. nematophila possessing a Lrp (leucine-responsive regulatory protein) mutant in Manduca sexta . The variation of the secondary forms of X. nematophila may be explained in its origin of a random mutation of a global regulator [50, 67]. Thus, the percent decrease of Bt-Plus efficacy must be the occurrence of the secondary form of X. nematophila.
With the primary form of X. nematophila, a Dual Bt-Plus was developed to control both P. xylostella and S. exigua. Usually, P. xylostella becomes more susceptible to CryIA (a main Cry toxin of Btk) than Cry1C (a main Cry toxin of Bta), whereas S. exigua becomes more susceptible to Cry1C than Cry1A [39, 68]. In this current study, an optimal ratio of Bta and Btk (4:1 in spore density) was relatively effective to control both insect species. With respect to Bt toxicity, active Cry toxins are concentrated on the midgut membrane receptors, such as alkaline phosphatase (ALP) or aminopeptidase N (APN), and interact with cadherin molecules on the membrane for further cleavage at the N-terminal piece [2, 42]. The cleaved Cry toxins then bind to each other to form an oligomeric prepore structure, which exhibits high binding activity to both ALP and APN [19, 60]. Then ALP and APN facilitate the pore formation of the oligomers on the midgut membrane. The oligomerization of the Cry toxins has been regarded as a crucial step for the toxicity of Bt Cry toxins . Thus, our results of the optimal Bt mixture exhibiting high toxicity to both lepidopteran species suggests a hetero-oligomerization of different Cry toxins, which may further facilitate a cooperative oligomerization between different Cry toxins to form pores in the epithelial membrane of the midgut. In fact, the mixture of Cry1Ab and Cry1Ac exhibited 5-fold synergistic toxicity against Chilo partellus larvae . Similarly, Cry1Aa and Cry1Ac have a synergistic effect against Lymantria dispar larvae, increasing their toxicity by 4.9-fold when the larvae were fed with a mixture of toxins, in which the combination of the two toxins resulted in greater pore formation activity than the individual toxins . Most Bt strains produce more than one type of Cry toxins, suggesting that hetero-oligomerization of different Cry toxins may be favored during evolution of Cry toxins as a mechanism to exploit the target insects for their specific and cooperative toxicity through an optimal pore formation strategy.
To further increase the Bt-Plus efficacy to both insect species, the specific bacterial metabolites were added after selecting the most potent metabolites (BZA and oxindole). The two metabolites possess PLA2/PO inhibition and cytotoxicity . Furthermore, BZA also exhibits an independent oral toxicity by suppressing the digestive efficiency of S. exigua . In addition, the dual Bt-Plus fortified with BZA or oxindole exhibited a complete control efficacy in a pot assay. This suggests that the 100-fold concentrated culture broth of X. nematophila can be an alternative choice to prepare a dual Bt-Plus without adding BZA or oxindole because of the presence of ≈1 ppm of these compounds in the current Bt-Plus formulation. Indeed, the 100-fold concentrated culture broth of X. nematophila effectively increased the Bt-Plus insecticidal activity like the metabolite-fortified Bt-Plus.
These results recapitulate the crucial role of the primary form of X. nematophila to enhance insecticidal activity. With the primary form of X. nematophila, this study developed a novel biopesticide to effectively control both S. exigua and P. xylostella by combining both types of Cry toxins with the concentrated culture broth of X. nematophila to meet the effective concentration of BZA or oxindole.