Functional redundancy of Burkholderia pseudomallei phospholipase C enzymes and their role in virulence

Phospholipase C (PLC) enzymes are key virulence factors in several pathogenic bacteria. Burkholderia pseudomallei, the causative agent of melioidosis, possesses at least three plc genes (plc1, plc2 and plc3). We found that in culture medium plc1 gene expression increased with increasing pH, whilst expression of the plc3 gene was pH (4.5 to 9.0) independent. Expression of the plc2 gene was not detected in culture medium. All three plc genes were expressed during macrophage infection by B. pseudomallei K96243. Comparing B. pseudomallei wild-type with plc mutants revealed that plc2, plc12 or plc123 mutants showed reduced intracellular survival in macrophages and reduced plaque formation in HeLa cells. However, plc1 or plc3 mutants showed no significant differences in plaque formation compared to wild-type bacteria. These findings suggest that Plc2, but not Plc1 or Plc3 are required for infection of host cells. In Galleria mellonella, plc1, plc2 or plc3 mutants were not attenuated compared to the wild-type strain, but multiple plc mutants showed reduced virulence. These findings indicate functional redundancy of the B. pseudomallei phospholipases in virulence.

Additionally, these Plc enzymes appear to play a role in nutrient acquisition 14 . Subsequently, Burtnick et al. 15 showed that B. pseudomallei Plc1 and Plc2 are secreted from the bacterial cell via the type II secretion system (T2SS) in a GspD-dependent manner 15 .
Little is known about the B. pseudomallei Plc3 enzyme. Whole-genome microarrays have revealed that plc3 is up-regulated in vivo, and in hamsters a plc3 mutant shows reduced virulence compared to the wild-type, suggesting that it is required for virulence 16 . However, the mechanisms underlying attenuation are unknown. Dowling et al. 17 reported that Plc3 might be a potential candidate vaccine requiring further study.
In this study, the expression of the plc genes in culture medium and in J774A.1 macrophage-like cells was analyzed using RT-PCR. B. pseudomallei plc1, plc2 or plc3 single mutants and plc12 or plc123 mutants were assessed for virulence in macrophages and in Galleria mellonella larvae. Our work provides new insights into the role of Plc enzymes in the pathogenesis of disease caused by B. pseudomallei.

Results and discussion
Culture pH differentially affects plc1 gene expression. B. pseudomallei is an intracellular bacterium.
After phagocytosis by phagocytes, the bacteria encounter acidic condition within the phagosome, before escaping to survive in the cytosol 4 . To determine the effect of pH on the expression of the plc1-3 genes, B. pseudomallei K96243 was incubated in LB broth which had been adjusted to pH 4.5, 5.0, 7.0, 8.0 or 9.0 before bacterial mRNA was extracted and tested. Using RT-PCR we first showed that expression of B. pseudomallei 23S rRNA was similar at all pH conditions tested. RT-PCR revealed that the level of transcription of the plc1 gene increased between pH 4.5 and 9.0. This result suggests that the pH of bacterial cultures can differentially affect B. pseudomallei plc1 gene expression. In contrast, the level of expression of plc3 was similar at all pH values tested. Expression of the plc2 gene was not detected under any of the conditions tested ( Fig. 1a) (full-length gels are presented in Supplementary Fig. 1). Our finding that plc2 was not expressed in strain K96243 is similar to the results reported by Ooi et al. 18 . However, plc2 is expressed in B. pseudomallei strain 22 18 , and the Plc2 protein was detected in B. pseudomallei strain MSHR668 15 culture supernatant. These findings indicate that plc2 gene expression is straindependent.
The B. pseudomallei plc genes are induced in infected macrophages. We next investigated expression of the plc genes in macrophages. J774A.1 macrophage-like cells were infected with B. pseudomallei K96243. At 2, 4, and 6 h post-infection (p.i.), the cells were lysed, and mRNAs corresponding to the plc gene detected using RT-PCR. The results showed no expression of the plc1, plc2 or plc3 genes at 2 h p.i. The expression of all of the plc genes was detected at 4 h p.i., but the level of expression of plc2 was relatively higher than the levels of expression of plc1 and plc3. At 6 h p.i., the levels of expression of all plc genes increased compared with 4 p.i. (Fig. 1b). Our finding that the expression of plc2 gene was induced in macrophages, even though we could not detect expression in culture medium (Fig. 1a), suggests a role of Plc2 in the infection of macrophages.
We stained infected macrophages for LAMP-1 and B. pseudomallei using probes labelled with Alexa Fluor 488 or Alexa Fluor 568 and nuclei were stained with DAPI (Fig. 1c). This revealed that at 2 h p.i. most B. pseudomallei K96243 were trapped within the phagosome. However, at 4 h p.i, co-localization of B. pseudomallei with lysosomes was rare indicating bacterial escape from phagosomes, and almost all of the B. pseudomallei cells were not associated with the phagosome at 6 h p.i. It is possible that the intracellular environment induces expression of plc1, plc2 and plc3 genes.
Several previous reports show the expression of plc genes of other species of bacteria in host tissues 12,13 . For example, the Mycobacterium tuberculosis plc genes are up-regulated in macrophages, and the Plcs are cytotoxic to mouse macrophages 7,19 . The Clostridium perfringens PLC (α-toxin) is produced in host tissues and can activate the arachidonic acid cascade in cells, with consequent modulation of host immune responses 20 , and the induction of ERK1/2 pathway, resulting in cytotoxicity 21 . B. pseudomallei plc2, but not plc1 or plc3, is required for bacterial survival and replication in macrophages. To provide insight into the role of plc1, plc2 and plc3 genes in virulence, we tested a range of single and multiple mutants. We have previously constructed plc1, plc2 and plc12 mutants 14 and these were included in our study. Additionally, for this study we constructed plc3 single and plc123 mutants by insertion mutagenesis 14 . Mutagenesis of the plc genes was confirmed by Southern blotting (data not shown). J774A.1 macrophage-like cells were infected with the mutants. At 2, 4, 6 and 8 h p.i., the numbers of recoverable plc2, plc12 or plc123 mutants were significantly lower (*P < 0.05, **P < 0.01, ***P < 0.001) than the number of viable wild-type bacteria (Fig. 2). In contrast, there was no significant difference (P > 0.05) in the numbers of viable single plc1 or plc3 mutants compared with the number of viable wild-type bacteria at all tested time points (2, 4, 6 or 8 h p.i.). These findings indicate that plc2 gene was required for B. pseudomallei survival and replication inside the macrophage. Our results also indicate that plc1 and plc3 play no role in survival and replication in macrophages, but we cannot discount the possibility of functional redundancy between these enzymes, which would mask the phenotype associated with the single plc1 and plc3 mutants.
To investigate whether the phenotypes we observed with the plc mutants was due to polar effects on downstream genes we measured expression of the genes downstream of plc1, plc2 or plc3 (bpsl2404, bpsl0337 or bpss0068 respectively) using RT-PCR. B. pseudomallei wild-type and the plc mutants were cultured in LB broth, or extracted from intracellular bacteria after macrophage infection. We demonstrated similar bpsl2404, bpsl0337 and bpss0068 amplicons with mRNA from wild-type or plc1, plc2, or plc3 mutants cultured in LB broth (Fig. 3a) and extracted from intracellular bacteria (Fig. 3b) (full-length gels are presented in Supplementary Fig. 2). These results indicate that the insertional mutation in the plc1, plc2, or plc3 genes did not abolish the expression of downstream genes. However, we cannot discount the possibility that expression of downstream genes was affected.
The defect in intracellular survival of the plc2 mutant is not due to delayed escape from the phagolysosome. The ability of B. pseudomallei to escape from the phagosome thought to be considered to be a mechanism by which the bacteria evade phagocyte killing. We had already showed that the plc1, plc2 and plc3 genes are expressed in macrophages ( Fig. 1c) and that the B. pseudomallei plc2 mutant was defective in survival in macrophages (Fig. 2). Here we investigated whether Plc2 enzyme played a role in escape from the phagolyosome. J774A.1 macrophage-like cells were infected with either B. pseudomallei wild-type, plc2, plc12, or plc123 mutants, and co-localization with lysosomes at 3 h p.i. was investigated by immunostaining with antibodies specific to LAMP-1. A B. pseudomallei bipB mutant, which is known to be delayed in phagosome escape 22,23 , was included as a control in our experiments. As expected, we found that the majority of bipB mutant cells showed delayed escape from phagosome as evidence the increased association with LAMP-1 (86 ± 3.3% association) when compared with the wild-type bacteria (Fig. 4a). The B. pseudomallei plc2, plc12 and plc123 mutants rarely co-localized with LAMP-1 (20.3 ± 1.2%, 20.7 ± 2.3%, and 27.3 ± 3.5% co-localization, respectively; Fig. 4b), similar to the degree of co-localization of the wild-type bacteria with LAMP-1 (17.7 ± 1.5% co-localization). This finding suggests that mutation of the plc genes did not affect escape of B. pseudomallei from phagosome.
B. pseudomallei plc2 mutant shows deficiency in plaque formation. We next measured plaque formation in monolayers of HeLa cells infected with either B. pseudomallei wild-type or the plc1 plc2, plc3, plc12 or plc123 mutants. Plaque-formation reflects the ability of bacteria to invade, survive within and then spread from cell to cell. As shown in Fig. 5, plaque-formation in HeLa cells was significantly reduced in cells infected with the plc2, plc12 or plc123 mutants, compared to the wild-type strain (**P < 0.01, ***P < 0.001, ***P < 0.001, respectively). In contrast, there was no significant reduction in plaque formation after infection with the plc1 or plc3 mutants (P > 0 0.05) (Fig. 5). This result shows correlation with our previous study 14 which showed that plaque-formation efficiency in HeLa cells was significantly reduced after infection with plc2 or plc12 double mutants compared to the wild-type strain. Plaque-formation was restored in a plc2 complemented strain. This finding suggested that the defective phenotype was due to the plc2 gene mutation 14 . However, complementation of plc12 double and plc123 triple mutants was not possible because of restrictions on the use of multiple antibiotic resistance markers in B. pseudomallei. B. pseudomallei wild-type, plc1, plc2, plc3, plc12 or plc123 mutants were used to infect J774A.1 cells at an MOI of 0.5. At 2, 4, 6 or 8 h p.i., the infected cells were lysed and bacteria enumerated after plating onto agar. Error bars represent standard error of mean for data collected from 3 independent experiments. Asterisks indicate significant differences (*P < 0.05, **P < 0.01, and ***P < 0.001, one-way ANOVA followed by Dunnett's post hoc test) between B. pseudomallei wild-type K96243 and mutant strains. The figure was prepared using GraphPad Prism version 7.05 for Windows (www.graph pad.com).
Scientific Reports | (2020) 10:19242 | https://doi.org/10.1038/s41598-020-76186-z www.nature.com/scientificreports/ To assess whether the reduction in plaque formation reflects a reduced ability to adhere to or to invade HeLa cells, we assessed invasion efficiency. There was no significant difference (P > 0 0.05) in the number of culturable intracellular bacteria at 2 h p.i. between wild-type and either plc2, plc12 double, or plc123 triple mutants (Fig. 6). This finding indicates that the absence of the plc 2 gene had no effect of the bacteria to adhere to or invade HeLa cells. Overall, our findings indicate that Plc2 is required for survival and replication of B. pseudomallei in nonphagocytic cells.
The plc12 and plc123 mutants are attenuated in G. mellonella. Our results above showed that Plc2 was required for intracellular survival and replication in host cells. To investigate the roles of Plc1, Plc2 and Plc3 in virulence of B. pseudomallei, a G. mellonella larvae infection model was used 24 .
There was no significant difference (P > 0 0.05) in the survival of larvae infected with the plc1, plc2 or plc3 mutants compared with the wild-type strain (Fig. 7). However, larvae infected with the plc12 or plc123 mutants showed significantly (P = 0.0031, P = 0.0018. respectively) increased survival, compared to larvae infected with the wild-type strain. This finding suggests redundancy of the functions of the phospholipases in virulence of B. pseudomallei. Our finding that there was no significant difference (P > 0 0.05) in the survival of larvae infected with the plc12 and plc123 mutants suggests that plc3 does not contribute to virulence in G. mellonella larvae. Because of the restrictions on the number of antibiotic markers we could introduce into B. pseudomallei, we could not generate plc12 or plc123 complemented mutants further validate our findings.
Our finding contrasts with a previous study where a plc3 mutant was found to 10 4 -fold attenuated Syrian hamsters compared to the wild-type bacterium 16 . This might reflect differences in the immune system between G. mellonella and mammals. G. mellonella possess an innate immune system which involves a cellular immune response mediated by hemocytes, and a humoral immune response orchestrated by antimicrobial peptides 25 , However, they lack the complement system found in mammals and G. mellonella also lacks an adaptive immune system 26   Expression of genes downstream of plc1, plc2 or plc3, assessed using RT-PCR. The mRNA from wild-type, plc1, plc2, plc3, plc12, or plc123 mutants cultured in LB broth (a), or isolated from infected J774A.1 macrophages (b) was extracted before converting to cDNA as outlined in material and methods (full-length gels are presented in Supplementary Fig. 2). The cDNA was amplified using PCR primers specific to the bpsl2404, bpsl0337 or bpss0068 genes which are downstream of plc1, plc2 or plc3, respectively (upper panel). The 16S rRNA (middle panel) and DNase-treated mRNA (lower panel) were included as a normalization control and negative control, respectively.

Conclusion
The genes encoding three B. pseudomallei Plc enzymes are expressed within macrophage-like cells, but at different expression levels. The plc2 gene was expressed in infected macrophages but not in culture medium, suggesting a role in virulence. Our findings suggest that plc2 either alone, or in combination with plc1 and plc3, contributes to growth in host cells and our finding that virulence in G. mellonella was dependent on the inactivation of genes encoding combinations of Plc enzymes, indicates functional redundancy. The data reported in this study provide important new insight into the roles of Plcs in virulence of B. pseudomallei and open new opportunities for further research into the roles on these enzymes in virulence.

Materials and methods
Primers, bacterial strains and cell lines. Primers used in this study are shown in Table 1. Escherichia coli, B. pseudomallei K96243 and the mutant strains were routinely cultured in Luria-Bertani (LB) or trypticase soy medium. B. pseudomallei K96243 plc1, plc2 single and plc12 double mutants were constructed in the previous study 14 . All cultures were typically grown for 24-48 h at 37 °C. Appropriate antibiotics (Sigma-Aldrich) i.e. chloramphenicol 50 µg/mL, kanamycin 400 µg/mL and tetracycline 50 µg/mL were added into the medium if required. All manipulations of B. pseudomallei were approved by the Technical Biosafety Committee (TBC), National Center for Genetic Engineering and Biotechnology (BIOTEC). J774A.1 murine macrophage-like and human epithelial HeLa cells were obtained from the American Type Culture Collection (ATCC) and were cultured in Dulbecco's Modified Eagle medium supplemented with 10% (v/v) heat-inactivated fetal bovine serum (Invitrogen) under a 5% CO 2 atmosphere at 37 °C in a humidified incubator.

RNA preparation and reverse transcription (RT)-PCR analysis. B. pseudomallei K96243 was grown
in LB broth for 6 h before incubation at 37 °C for 15 min in LB broth pH 4.5, 5.0, 6.0, 7.0, 8.0 or 9.0. Total RNA was extracted from 10 8 CFUs B. pseudomallei cultured in each condition using TRIZOL (Invitrogen) according to manufacturer's instructions. The isolated bacterial RNA was then treated with DNase I (Ambion) to remove any genomic DNA contamination before use.
To detect B. pseudomallei genes expression within macrophages, monolayers of J774A.1 murine macrophagelike cells were infected with the bacteria. At the indicated time points, the infected cell monolayers were washed and subsequently lysed with 500 μL of 0.1% Triton X-100 (Sigma-Aldrich) to allow intracellular bacteria released from infected cells. Then, 500 μL of 1 × PBS was added and the intracellular bacterial RNA were extracted using TRIZOL (Invitrogen) according to manufacturer's instructions.
To convert the extracted total RNA to cDNA, SuperScript III First-Strand Synthesis System (Invitrogen) was used. The cDNA was quantified and adjusted so that similar quantities were included in the PCR reactions. The cDNA was amplified using the PCR with primers (Table 1), GoTaq DNA polymerase (Promega) and cycling conditions of 94 °C, 3 min and 30 cycles of 94 °C for 30 s, 50 °C for 1 min, and 72 °C for 45 s, followed by incubation at 72 °C for 5 min. In each PCR experiment, the amplification of 23S rRNA was used as a normalization control. B. pseudomallei plc3 single and plc123 triple mutants. A B. pseudomallei plc3 mutant was constructed by insertion mutagenesis 30 . A 406-bp (nucleotide positions 1052-1457) internal region of the plc3 gene was amplified from B. pseudomallei K96243 genomic DNA with primers PLC88 and PLC89 ( Table 1). The amplified DNA fragment was ligated into EcoRV digested pKNOCK-Cm, a suicide vector 30 to generate recombinant plasmid pVSK3 for insertion mutagenesis. The constructed plc3 mutant was selected on Pseudomonas agar base supplemented with SR103E (Oxoid) and chloramphenicol.

Construction of
To construct the B. pseudomallei plc123 triple mutant, the amplified internal plc3 fragment (nucleotide positions 1052-1457) was ligated into EcoRV digested pKNOCK-Km 30 . This constructed plasmid, designated pVSK4, was introduced into B. pseudomallei plc12 double mutant 14 . The mutants were selected on Pseudomonas agar base (Oxoid) supplemented with SR103E (Oxoid) containing chloramphenicol, kanamycin and tetracycline (Sigma-Aldrich). The plc123 mutant was verified by PCR and Southern blotting. (a) J774A.1 cells were infected with B. pseudomallei wild-type K96243, plc2, plc12, plc123 or bipB mutants for 3 h. Cells were fixed, permeabilized and immunostained with rat anti-LAMP1 and mouse anti-Burkholderia monoclonal antibodies. The y-axis shows the number of bacteria co-localizing with LAMP-1 × 100/total number of intracellular bacteria. Error bars represent standard error of the mean for data collected from 3 independent experiments. Asterisks indicate significant differences (P < 0.01, students t-test) between B. pseudomallei wildtype K96243 and mutants. The figure was prepared using GraphPad Prism version 7.05 for Windows (www. graph pad.com). (b) Representative confocal micrographs of the association of B. pseudomallei with endocytic vesicles in J774A.1 cells. B. pseudomallei wild-type K96243, plc2, plc12 double, plc123 triple or bipB mutants were used to infect J774A.1 cells. At 3 h p.i., the infected macrophages and the bacteria were stained and visualised using confocal microscopy. Macrophage LAMP-1 was stained green with rat monoclonal antibody (1D4B) and Alexa Fluor 488 goat anti-rat IgG antibody, and nuclei were stained blue with DAPI. Bacteria were stained red with mouse anti-B. pseudomallei monoclonal antibody (9D5) and Alexa Fluor 568 goat anti-mouse IgG antibody. White arrows indicated the co-localization of B. pseudomallei and LAMP-1.

Confocal analysis of bacterial co-localization with LAMP-1. The intracellular localizations of B.
pseudomallei wild-type K96243 and the plc mutants in J774A.1 macrophages cells relative to LAMP-1 containing vesicles were investigated according to previously described 33   Virulence in G. mellonella. Virulence in G. mellonella larvae was tested as described previously 24 with some modifications. Larvae between 2-2.5 cm and free of melanization or injury were used in the experiments. To prepare the bacterial culture for infection, the overnight cultures of B. pseudomallei wild-type and the mutants were adjusted to a concentration of 10 4 CFUs per ml in PBS. A 701 N fixed needle syringe (Hamilton, Nevada) was used to inject 10 µl aliquots of the bacterial suspension into the G. mellonella larvae to get the final concentration of 10 2 CFUs. Injections were performed directly into the larval body cavity and groups of 10 larvae were injected with each bacterial strain. Control larvae were injected with PBS. Following injection, larvae were incubated in the dark at 37 °C and the number of dead larvae were recorded at a variety of times post injection. The Galleria mellonella study was approved by the Mahidol University-Institute Animal Care and Use Committee (U1-05763-2559).