The enterobacterium Photorhabdus luminescens produces a number of toxins to kill its insect host. By analyzing the genomic sequence of P. luminescens TT01, we found that amino acid sequences encoded by plu1961 and plu1962 showed high similarity to XaxAB binary toxin of Xenorhabuds nematophila, which has both necrotic and apoptotic activities in both insect and mammalian cells in vitro. To evaluate the biological activity of Plu1961/Plu1962, their coding genes were cloned and expressed in Escherichia coli. Both Plu1961 and Plu1962 were expressed as soluble protein in BL21 (DE3) and their mixture caused insect midgut CF-203 cells death via necrosis. Confocal fluorescence microscopy showed that Plu1961/Plu1962 mixture was able to depolymerize microtubule and induce the increase in plasma membrane permeabilization in CF-203 cells. Moreover, co-expression of Plu1961/Plu1962 in the same cytoplasm exhibited cytotoxic effect against mammalian cells (B16, 4T1, and HeLa cells) and injectable activity against Spodoptera exigua larvae. Until now, two types of binary toxins have been identified in P. luminescens, the first type is PirAB and Plu1961/Plu1962 is the second one. The biological role of Plu1961/Plu1962 binary toxin played in the infection process should attract more attention in future.
Photorhabdus luminescens is an entomopathogenic, Gram-negative, bioluminescent bacterium that exists in a state of mutualistic symbiosis with entomopathogenic nematodes of the family Heterorhabditidiae (Ffrench-Constant et al., 2007). Upon entering an insect host, the nematodes release the bacteria directly into the insect hemocoel. Once released into the insect blood system, the bacteria kill their insect host by producing a large number of toxins. Various toxins have been characterized in P. luminescens (Rodou et al., 2010), which can be classified into four major groups: the toxin complexes (Tcs), the ‘makes caterpillars floppy’ (Mcf) toxins, the Photorhabdus insect-related (Pir) proteins, and the Photorhabdus virulence cassettes (PVC). Tc toxins attracted attention from the fact that some of these toxin complexes are highly toxic to insects after oral feeding, suggesting potential as insecticides (Waterfield et al., 2001). Mcf toxins cause damage to the insect midgut after injection into larvae with loss of body turgor and a ‘floppy’ phenotype of the caterpillars (Dowling et al., 2004). Injection of PVCs destroys insect hemocytes, which undergo dramatic actin cytoskeleton condensation (Yang et al., 2006). Pir toxins act as binary proteins. Both PirA and PirB proteins are necessary for insecticidal activity. Injection of either PirA or PirB alone into caterpillars of Galleria is not associated with any mortality, and mixture of individual PirA and PirB preparations exhibits full activity against this insect (Waterfield et al., 2005). Histological examination of Plutella xylostella larvae fed with recombinant Escherichia coli expressing PirA and PirB proteins reveals gross abnormalities of the midgut epithelium, with profound swelling and shedding of the apical membranes (Blackburn et al., 2006). PirAB toxins also show larvicidal activity against mosquito larvae (Aedes aegypti and Aedes albopictus; Ahantarig et al., 2009).
Binary toxins have also been reported in several other bacteria, including Clostridium botulinum C2 toxin, Clostridium difficile toxin (CDT), Clostridium perfringens iota (ι) toxin, Clostridium spiroforme toxin (CST), Bacillus anthracis edema and lethal toxins, as well as the Bacillus cereus vegetative insecticidal proteins (VIP; Barth et al., 2004). Normally, binary toxins consist of binding component and enzymatic component. The binding component recognizes a cell surface receptor and allows the internalization of the enzymatic component into the cytosol, and the enzymatic component catalyzes the reaction and induces the toxicity (Carman et al., 2011).
Recently, a new binary toxin gene xaxAB from Xenorhabdus nematophila, a bacterial species closely related to P. luminescens, was cloned and sequenced. XaxAB toxin exhibited both necrotic and apoptotic activities in both insect and mammalian cells in vitro. Incubations of sheep red blood cells with XaxAB showed that maximum hemolytic activity was obtained with equimolar concentrations of XaxA and XaxB. This binary toxin cannot be classified in any known family of cytotoxins on the basis of amino acid sequences, locus organization, and activity features. The putative hemolysin loci, containing two closely linked genes similar to xaxAB, were also found to be present in the chromosome of Photorhabdus, Pseudomonas, and Yersinia (Vigneux et al., 2007).
Analysis of the genomic sequence of P. luminescens TT01 (Duchaud et al., 2003) revealed that amino acid sequences encoded by plu1961 and plu1962 showed 76.8% and 74.9% similarity to XaxA and XaxB, respectively. To evaluate the biological activity of this potential binary toxin, plu1961 and plu1962 were cloned and expressed in E. coli. Both oral and injectable toxicities of Plu1961/Plu1962 were assayed against insect larvae. Cytotoxic effect of binary toxin was tested against insect midgut CF-203 cells and mammalian cell lines. The possible mechanism for the cytotoxicity of Plu1961/Plu1962 was also discussed.
Materials and methods
Bacterial strains and plasmids
Bacterial strains and plasmids used in this study are listed in Supporting Information, Table S1. Photorhabdus luminescens TT01 was grown in Luria–Bertani (LB) medium at 28 °C, and strains of E. coli were grown in LB medium at 37 °C. Escherichia coli DH5α was used as the host for recombinant DNA cloning. Escherichia coli BL21 (DE3) was used as the host for expression of binary toxin genes. Plasmid pET28a (Novagen) was used as expression vector in BL21 (DE3). Plasmid pETDuet-1 (Novagen) was used as co-expression vector in BL21 (DE3).
Construction of expression plasmids
Total DNA was extracted from P. luminescens TT01 using the alkali lysis method. It was used as template for amplification of plu1961 and plu1962 (GenBank accession no. BX571865). Oligos used in this study were listed in Table S2. Oligo pair Plu1961-F/Plu1961-R was used to amplify plu1961, and Plu1962-F/Plu1962-R was used to amplify plu1962. Both PCR products were double-digested by EcoRI/SalI and cloned into pET28a to generate plasmids pET-plu1961 and pET-plu1962, respectively. For co-expression of Plu1961 and Plu1962 in BL21 (DE3), Co1961-F/Co1961-R and Co1962-F/Co1962-R were used to amplify plu1961 and plu1962, respectively. PCR products of plu1961 and plu1962 were double-digested by PstI/SalI and NdeI/XhoI, respectively, and cloned into pETDuet-1 sequentially to generate co-expression plasmid pET-pluBi. All the plasmids were confirmed by DNA sequencing.
Expression of binary toxins and solubility analysis
Plasmids pET-plu1961, pET-plu1962, and pET-pluBi were transformed into BL21 (DE3), and resultant strains were designated as BL21 (plu1961), BL21 (plu1962), and BL (Bi), respectively. Recombinant strains were grown in LB medium with kanamycin (50 μg mL−1) or ampicillin (100 μg mL−1) at 37 °C to an OD600 between 0.6 and 0.8. Then, isopropyl-beta-d-thiogalactopyranoside (IPTG) was added at a final concentration of 1 mmol L−1. After IPTG induction for 4 h, aliquots of 1 mL bacteria culture were sampled and harvested by centrifugation (10 000 g, 1 min). Pellets were washed three times with distilled water and suspended in 0.1 mL lysis buffer (50 mmol L−1 NaH2PO4, 300 mmol L−1 NaCl, 10 mmol L−1 imidazole, pH 8.0). Then, cells were lysed by sonication and centrifuged at 10 000 g for 2 min. The supernatant was collected, and 10 μL aliquots were taken for SDS-PAGE.
Purification of binary toxin
Soluble binary toxins (Plu1961 and Plu1962) were directly purified on 1-mL HisTrapTM HP prepacked columns (GE Healthcare), using an AKTA Purifier system (GE Healthcare; flow rate 1 mL min−1). The column was equilibrated in His A buffer (20 mmol L−1 sodium phosphate, 0.5 mol L−1 NaCl, 20 mmol L−1 imidazole, pH 7.4). Proteins were eluted using a step gradient up to 0.5 mol L−1 imidazole in His A buffer. Fractions were analyzed by SDS-PAGE, and the protein content of the pools was determined using the Bio-Rad Bradford reagent. The purified proteins were dialyzed against PBS buffer prior to application.
The oral toxicity of binary toxin was tested against Helicoverpa armigera and Spodoptera exigua. The diet of insects was prepared by the method described previously (Hu et al., 2009). Different amounts (15–150 μL) of concentrated supernatant of BL (Bi) lysate were applied to 1-cm3 disks of artificial diet. Treated food blocks were allowed to dry for 20 min. Three first-instar larvae were placed on each food block before incubation at 25 °C for 7 days. The percent mortality of larvae and the weight of surviving larvae were then recorded. Concentrated supernatants of BL21 (DE3) lysate were used as negative control. Solubilized Cry1Ac protein, which is highly toxic against H. armigera larva, was used as a positive control. The bioassay was performed three times on different days.
The injectable toxicity of binary toxin was tested against S. exigua fourth-instar larvae. Different amounts (10, 25, 50 and 100 μL) of concentrated supernatant of untreated or heat-inactivated (incubated at 80 °C for 30 min) BL (Bi) lysate were injected directly into the insect hemocoel from the third abdominal foot. Two repeats of 40 larvae for each amount were used. Concentrated supernatant of BL21 (DE3) lysate was used as a negative control. After injection, larvae were incubated at 28 ± 1 °C on an artificial diet and monitored over 7 days for any deleterious effects. Bioassay was performed three times on different days.
Cell lines and culture conditions
The cell line FPMI-CF-203/2.5 (CF-203), originated from the midgut of the spruce budworm (Choristoneura fumiferana; Lepidoptera, Torticididae), was kindly gifted by Prof. Guido F. Caputo (Natural Resource Canada). CF-203 was cultured in Insect-Xpress medium (BioWhittaker, Cambrex Bioscience, Walkersville, MD) supplemented with 2.5% heat-inactivated fetal bovine serum (FBS; Sigma-Aldrich, Bornem, Belgium) at 27 °C (Vandenborre et al., 2008).
4T1 mouse breast tumor cells, Hep 3B human hepatoma cells, HeLa human cervical cancer cells, and HCT116 human colon cancer cells were purchased from the American Type Culture Collection (ATCC). B16 mouse melanoma cells were obtained from the Cell Bank of Type Culture Collection, Chinese Academy of Sciences. All mammalian cell lines were cultured in RPMI 1640 medium supplemented with L-glutamine (Gibco) and 10% heat-inactivated FBS, 100 U mL−1 penicillin, and 100 μg mL−1 Streptomycin at 37 °C.
Cell viability bioassay
The effect of different concentrations of Plu1961 (0.5–3.0 μmol L−1) alone, Plu1962 (0.5–2.5 μmol L−1) alone, and their mixture (0.2–1.6 μmol L−1) on cell viability was determined against CF-203, 4T1, Hep 3B, HeLa, HCT116, B16 cell lines. Wells of a 96-well microtiter plate were loaded with 100 μL of cell suspension, containing 2.0 × 105 cells mL−1, and exposed to different concentrations of object proteins or deionized water in the control treatment. Cytotoxicity of lysate from BL (Bi) was also tested against 4T1, Hep 3B, HeLa, HCT116, B16 cell lines, and the lysate from BL21 (DE3) was used as a control. The plates were incubated for 2 days at 28 °C. For each concentration, three replicates were performed and each experiment was repeated twice. Cell morphological changes were analyzed by an inverted light microscope (Leica DMIL; Leica Microsystems S.p.A, Milan, Italy). Cell viability was determined by the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay. The absorbance was read at 490 nm using an enzyme-linked immunosorbent assay plate reader.
Confocal fluorescence microscopy
Plu1961/Plu1962-treated CF-203 cells or untreated CF-203 cells were seeded in 6-well microtiter plate containing Insect-Xpress culture medium with a coverslip in the bottom and incubated for 12 h. Then the coverslip was turned upside down on a glass slide containing 0.25 μg mL−1 Mitotrack Red, incubated at the room temperature for 20 min. Cells were then incubated in PBS containing 0.5% Triton X-100 at room temperature for 10 min. After being washed with PBS, cells were incubated at room temperature for 1 h with PBS containing 2% BSA. Then the coverslip was turned upside down on a glass slide containing 5 μg mL−1 Mouse anti-α-Tubulin-Alexa 488 and incubated at room temperature for 1 h. After washing four times, cells were then incubated at room temperature for 20 min in 5 μg mL−1 4′,6-diamino-2-phenylindole dihydrochloride (DAPI).After washing extensively with PBS, a fluorescence quencher was added to seal the tablet. Confocal images were acquired using Zeiss LSM 510 META (Germany) and processed with lsm image browser software and adobe photoshop (CS2). The confocal fluorescence microscopy was performed three times on different days.
Hoechst 33342 and propidium iodide co-staining
Cell viability and nuclear morphology were assessed by the Hoechst 33342 and propidium iodide co-staining method (Yuan et al., 2002). The Apoptosis and Necrosis Assay Kit (Beyotime Institute of Biotechnology, Hai men, China) was used according to the manufacturer's instructions.
DNA fragmentation analysis
CF-203 cells treated with different concentrations of Plu1961/Plu1962 binary toxin and treated with PBS (negative controls) were collected after 24 h. Cells were homogenized by freezing and thawing several times and mixed in DNA extraction buffer (10 mmol mL−1 Tris-HCl, 150 mmol mL−1 NaCl, 10 mmol mL−1 EDTA-NaOH, 0.1% SDS, pH 8.0) on ice. Homogenized cells were treated with 20 mg mL−1 RNase for 30 min at 37 °C. Subsequently, 100 mg mL−1 of proteinase K was added, and cells were incubated at 50 °C for 60 min. DNA samples were extracted using a standard phenol–chloroform extraction method and analyzed by 2% agarose gel.
Both components of binary toxin were expressed as soluble proteins in BL21 (DE3)
To evaluate the biological activity of Plu1961/Plu1962, their encoding genes were cloned and expressed in BL21 (DE3). The theoretical molecular weight (MW) of 342-amino acid protein Plu1961 is 39 kDa, while theoretical MW of Plu1962 which consists of 412 amino acids is 46.5 kDa. After inducing with 1 mmol L−1 IPTG, prominent bands of c. 39 and 46.5 kDa were found in the supernatants of induced cultures of BL21 (plu1961) and BL21 (plu1962), respectively (Fig. 1, lanes 2 and 1), both of which were absent from the supernatants of uninduced cultures (Fig. 1, lane 4 and lane 5). And both bands (c. 39 and 46.5 kDa) were present in the supernatants of induced cultures of BL (Bi; Fig. 1, lane 3). It indicated that both Plu1961 and Plu1962 were expressed as soluble proteins in BL21 (DE3), no matter whether they were separately expressed or co-expressed.
Binary toxin exhibited injectable activity against insect larvae
When Plu1961/Plu1962 was applied by mixing with diet, neither mortality nor growth inhibition of both H. armigera and S. exigua larvae was observed within the tested amounts (15–150 μL) of BL (Bi) lysate. However, injection of 10 μL of supernatant of BL (Bi) lysate resulted in around 42% mortality of S. exigua fourth-instar larvae after 24 h. And the mortality rate rose with the increase in BL (Bi) lysate volume. When 100 μL of concentrated supernatant of BL (Bi) lysate was injected into S. exigua fourth-instar larvae, 97% mortality rate was observed after 24 h (Fig. 2b). When compared with the control group (supernatant of BL21 (DE3) lysate and heat-inactivated supernatant of BL (Bi) lysate), the supernatant of BL (Bi) lysate caused extensive blackening of larvae (Fig. 2a). Blackening of S. exigua larvae suggested that injection of BL (Bi) lysate resulted in the activation of phenoloxidase which was responsible for the synthesis of melanin, a key component in arthropod immunity and wound healing (Li et al., 2008). It demonstrated that Plu1961/Plu1962 had injectable toxicity against tested insect larvae, but no oral toxicity.
Binary toxin causes insect midgut CF-203 cells death via necrosis
MTT assay was performed against insect midgut CF-203 cells to investigate the cytotoxicity of Plu1961/Plu1962. Neither component of binary toxin could affect the growth of CF-203 cells even after 4 days of incubation. In contrast, the mixture of Plu1961/Plu1962 caused a loss of cell viability after 24 h of incubation within the tested concentrations (0.2–1.6 μmol L−1). 0.2 μmol L−1 of binary toxin mixture resulted in 55% loss of cell viability. More than 90% of cells lost viability after treatment with 1.6 μmol L−1 of binary toxin (Fig. 2c).
When compared with control cells (Fig. 3a), CF-203 cells treated with the mixture of Plu1961/Plu1962 showed marked swelling, formation of surface blisters, followed by membrane lysis and dispersal of the cytoplasmic organelles and swollen nuclear contents into the surrounding medium (Fig. 3d). In contrast, individual application of Plu1961 or Plu1962 alone had no morphological effect on CF-203 cells (Fig. 3b and c).
Morphological changes in CF-203 cells exposed to Plu1961/Plu1962 mixture were further investigated by confocal microscope. The control cells and cells treated by Plu1961 alone displayed strong green fluorescence (microtubules) around the nuclei (strong blue fluorescence), the mitochondria (red fluorescence) appeared to be almost evenly distributed in the cytoplasm (Fig. 4a and b). In contrast, cells treated with the mixture of Plu1961/Plu1962 lost virtually all green and red fluorescence and exhibited only blue fluorescence (Fig. 4d). This demonstrated that the mixture of Plu1961/Plu1962 could depolymerize microtubules. Cells treated by Plu1962 alone displayed a dramatic decrease in density of green fluorescence (Fig. 4c). This implied that Plu1962 alone could depolymerize microtubules to a certain extent.
We next investigated the possible mechanisms responsible for the rapid cell death caused by binary toxin using Apoptosis and Necrosis Assay Kit. The intact membrane of live cells excludes charged cationic dyes, such as trypan blue, propidium, or ethidium, and short incubation with these dyes results in selective labeling of dead cells, while live cells show minimal dye uptake. Loss of plasma membrane integrity leading to increased permeability to PI was found to be characteristic of necrosis. Most of the CF-203 cells treated with 0.6 μmol L−1 mixture of Plu1961/Plu1962 showed strong blue fluorescence and red fluorescence. Conversely, weak blue fluorescence and no red fluorescence were detected in control cells (Supporting Information, Fig. S1). Moreover, incubation of CF-203 cells with mixture of Plu1961/Plu1962 (0.6 nM) failed to induce DNA ladder fragmentation, a hallmark of apoptosis, even after incubation for 24 h (data not shown). Taken together, we therefore assumed that Plu1961/Plu1962 exhibited necrotic cytotoxicity in CF-203 cells.
Co-expression of Plu1961/Plu1962 exhibited cytotoxic effect against several mammalian cell lines
Five mammalian cell lines (B16, 4T1, HeLa, Hep 3B, HCT116) were also used to examine the cytotoxicity of binary toxin. Neither Plu1961 nor Plu1962 alone could inhibit the growth of all tested mammalian cell lines. Unexpectedly, the mixture of Plu1961/Plu1962 (1.6 μmol L−1) exhibited no cytotoxic effect on all tested mammalian cell lines (data not shown). We then co-expressed Plu1961 and Plu1962 in BL21 (DE3). Lysate from BL (Bi) exhibited strong cytotoxicity against B16, 4T1, and HeLa cells (Fig. 5). Hep 3B and HCT116 cells were insensitive to BL (Bi) lysate (data not shown).
In the present study, we identified a XaxAB-like binary toxin from P. luminescens, which exhibits cytotoxicity against insect midgut CF-203 cells and some mammalian cell lines. Both Plu1961 and Plu1962 were necessary to restore full cytotoxicity against CF-203 cells. XaxAB and Plu1961/Plu1962 show no homology to any other protein with known function, indicating that they constitute a distinct family of binary toxins (Vigneux et al., 2007).
Photorhabdus luminescens proliferates in the hemolymph before the insect dies and must therefore be able to escape the insect immune response. Cell-mediated immunity comes into play immediately after the insect hemocoel is penetrated by a foreign body (Ribeiro & Brehelin, 2006). It was reported that injection of wild-type E. coli into Manduca sexta resulted in rapid encapsulation of all of the bacteria by the insect hemocytes, completely clearing the infection from the hemocoel. These capsules, or nodules, are formed by complex interactions between different subpopulations of the hemocytes, which result in encapsulation of the bacteria and final melanization of the resulting capsule. Once encapsulated, the trapped bacteria subsequently die (Silva et al., 2002). Cytotoxic factors targeting immunocytes are good candidates for the mediators of immunosuppression (Clarke, 2008). Plu1961/Plu1962 caused death of CF-203 cells via necrosis. Further studies on the necrotic and apoptotic activities of Plu1961/Plu1962 against insect hemocytes will be necessary to elucidate its role in immunosuppression.
Confocal microscopy revealed that Plu1961/Plu1962 caused a notable decrease in cellular tubulin of CF-203 cells. Microtubule, one of the principal components of cytoskeleton, is critical to cell shape, cell movement, intracellular transport of organelles, and the separation of chromosomes during mitosis (Archuleta et al., 2011). As a result, microtubule is a prime target for pathogens and their virulence factors. Mouse macrophages treated with Bacillus anthracis lethal toxin (LT) induced a notable decrease in the level of cellular tubulin and altered stability of the microtubule network (Chandra et al., 2005). Treatment of human colonocytes with Clostridium difficile toxin A resulted in tubulin deacetylation and subsequent microtubule depolymerization (Nam et al., 2010).
Assembly of the two components is essential for binary toxins to exhibit their cytotoxicity (Schleberger et al., 2006). However, the stage at which the assembly of the binary toxin components occurs is debatable. Previous study suggested that intoxication by binary toxins initially involved specific, receptor-mediated binding of ‘B’ component to a targeted cell as monomers that form homoheptamers on the cell surface. The ‘B’ heptamer–receptor complex then acts as a docking platform that subsequently translocates the enzymatic ‘A’ component into the cytosol. Once inside the cytosol, ‘A’ component can inhibit normal cell functions (Barth et al., 2004). It was reported that at low toxin concentrations, complex formation might enhance the efficiency of the binary toxin (Kaiser et al., 2006). Our data demonstrated that when co-expressed in the same cytoplasm, Plu1961 and Plu1962 could interact with each other and form a complex. This could in part explain the observation that Plu1961 and Plu1962 mixed in vitro did not affect the growth of mammalian cells, but while co-expressed in the same cytoplasm, Plu1961/Plu1962 exhibited cytotoxic effect against B16, 4T1, and HeLa cells.
In conclusion, we have identified XaxAB-like binary toxin from P. luminescens TT01, which exhibits highly injectable toxicity against insect larvae. Plu1961/Plu1962 mixture could cause rapid cell necrosis when applied to insect midgut CF-203 cells. However, co-expression in the same cytoplasm is essential for Plu1961/Plu1962 to exhibit cytotoxic activity against mammalian cells. The biological role of Plu1961/Plu1962 in the infection process needs further study.
This investigation was supported by National Natural Science Foundation of China (31200004, 30900037), Specialized Research Fund for the Doctoral Program of Higher Education from the Ministry of Education of China (2011430620005), Scientific Research Fund of Hunan Provincial Education Department (10K041). The authors report no financial or other conflict of interest relevant to the subject of this article.