Burkholderia pseudomallei Type III Secretion System Cluster 3 ATPase BsaS, a Chemotherapeutic Target for Small-Molecule ATPase Inhibitors

Abstract

Melioidosis is an infectious disease of high mortality for humans and other animal species; it is prevalent in tropical regions worldwide. The pathogenesis of melioidosis depends on the ability of its causative agent, the Gram-negative bacterium Burkholderia pseudomallei, to enter and survive in host cells. B. pseudomallei can escape from the phagosome into the cytosol of phagocytic cells where it replicates and acquires actin-mediated motility, avoiding killing by the autophagy-dependent process, LC3 (microtubule-associated protein light chain 3)-associated phagocytosis (LAP). The type III secretion system cluster 3 (TTSS3) facilitates bacterial escape from phagosomes, although the mechanism has not been fully elucidated. Given the recent identification of small-molecule inhibitors of the TTSS ATPase, we sought to determine the potential of the predicted TTSS3 ATPase, encoded by bsaS, as a target for chemotherapeutic treatment of infection. A B. pseudomallei bsaS deletion mutant was generated and used as a control against which to assess the effect of inhibitor treatment. Infection of RAW 264.7 cells with wild-type bacteria and subsequent treatment with the ATPase inhibitor compound 939 resulted in reduced intracellular bacterial survival, reduced escape from phagosomes, and increased colocalization with both LC3 and the lysosomal marker LAMP1 (lysosome-associated membrane protein 1). These changes were similar to those observed for infection of RAW 264.7 cells with the bsaS deletion mutant. We propose that treatment with the ATPase inhibitor compound 939 decreased intracellular bacterial survival through a reduced ability of bacteria to escape from phagosomes and increased killing via LAP. Therefore, small-molecule inhibitors of the TTSS3 ATPase have potential as therapeutic treatments against melioidosis.

INTRODUCTION

Burkholderia pseudomallei is a Gram-negative, soil-dwelling bacillus. It is the causative agent of melioidosis, an often fatal infection of many animal species and humans that is endemic in tropical and subtropical areas of the world (1, 2). Melioidosis generally presents as a febrile illness with a range of acute or chronic clinical manifestations, although prolonged periods of latency have also been documented (3). The high resistance of B. pseudomallei to a wide spectrum of antibiotics makes therapy of melioidosis problematic and its overall mortality remains high, at ca. 40% in northeast Thailand and 20% in northern Australia (2). As an intracellular pathogen, B. pseudomallei can invade both phagocytic (4) and nonphagocytic (5) cells. After internalization, bacteria can escape from the phagosome into the host cytoplasm. Once in the cytoplasm, B. pseudomallei can replicate and induce actin polymerization at one pole of the bacterium, facilitating intracellular motility (6, 7). This actin-based motility facilitates bacterial spreading into adjacent cells via membrane protrusions, leading to the formation of multinucleated giant cells (MNGC), which have been observed in both cultured cell lines and the tissues of patients (8).

Numerous B. pseudomallei virulence factors have been characterized, including capsule, pili, flagella, lipopolysaccharide (LPS), quorum-sensing molecules, and type III and type VI secretion systems (7, 9, 10). One of the major virulence factors is the type III secretion system cluster 3 (TTSS3; also termed bsa, for Burkholderia secretion apparatus), which mediates the secretion of effector molecules directly into host cells through a membrane-spanning needle (11). B. pseudomallei has three different TTSS clusters, namely, TTSS1 (BPSS1390-1408), TTSS2 (BPSS1613-1629), and TTSS3 (BPSS1520-1554), which are suggested to play roles in the interaction of B. pseudomallei with different hosts (12). Indeed, TTSS1 and TTSS2 are required for infection of tomato plants by B. pseudomallei (13) but not for infection of hamsters (14). Furthermore, we recently reported that TTSS1 plays an important role during infection of BALB/c mice, supporting the idea that TTSS1-mediated pathogenesis is host dependent (15). In contrast, TTSS3, similar to the Inv/Mxi-Spa TTSS of Salmonella and Shigella species (16), is essential for full virulence in both hamsters and mice (14, 17). Several lines of evidence have demonstrated that the TTSS3 facilitates bacterial escape from phagosomes (16, 18), evasion of LC3 (microtubule-associated protein light chain 3)-associated phagocytosis (LAP; an autophagy-related process) (19), and induction of caspase-1-dependent cell death in macrophage cells (20). It was reported that TTSS3 was required for invasion of nonphagocytic cells (21); however, a more recent report, in which a photothermal nanoblade was used to deliver B. thailandensis directly to the cytosol (thereby bypassing the need for endosome escape), concluded that while TTSS3 is indeed required for escape from endosomes, it is not required for invasion (18). However, the authors of the latter report noted that observations made with HEK293 cells may not translate directly to other cell types such as professional phagocytes.

The TTSS3 locus encodes at least 30 proteins (16); the functions of many of these proteins in B. pseudomallei remain mostly uncharacterized. Four TTSS3 secretion apparatus genes, bsaQ (20, 22, 23), bsaU (23, 24), bsaZ (14, 16), and bipD (16, 25) are essential for TTSS3 function and therefore bacterial escape from phagosomes, intracellular survival, and virulence in mice.

Bacterial TTSSs are energized through the hydrolysis of ATP by TTSS-associated ATPases (26, 27), which are proposed to form hexameric ring structures associated with the secretion apparatus at the inner bacterial membrane (28). ATP hydrolysis promotes the initial docking of TTSS substrates to the secretion apparatus, unfolding of effector proteins prior to secretion, and release of effectors from their cognate chaperones (26, 28). Bacterial strains lacking the TTSS ATPase are defective in TTSS apparatus function (28) and assembly (29, 30). The TTSS3 bsaS (BPSS1541) encodes a putative protein of 435 amino acids with a predicted molecular mass of 46.8 kDa. The deduced amino acid sequence of B. pseudomallei BsaS has a high degree of identity to TTSS ATPases from other Gram-negative bacteria, including the B. mallei TTSS3 BsaS (99.8%), B. thailandensis TTSS3 BsaS (95.0%), Salmonella enterica serovar Typhimurium SPI-1 TTSS SpaL/InvC (55.5%), Shigella flexneri Spa47 (53.1%), enteropathogenic Escherichia coli EivC (56.0%), and Yersinia pestis YscN (46.4%).

The structural and functional conservation of TTSS ATPases across many pathogenic bacteria makes them attractive targets for the development of novel antibacterial therapeutics. Recently, several small-molecule ATPase inhibitors were identified that exhibited efficient inhibition of the ATPase activity of Y. pestis YscN and B. mallei BsaS in vitro and blocked the secretion of TTSS effectors by Y. pestis at micromolar concentrations (31). Furthermore, these TTSS ATPase inhibitors were relatively nontoxic to mammalian cells, indicating no significant inhibition of host ATPase proteins, which share low identity and have different active sites from bacterial ATPases (31). Based on the high amino acid identity (only 1 of 436 residues is different) between B. pseudomallei BsaS and B. mallei BsaS (32), we reasoned that one or more of the same group of inhibitors could be effective against B. pseudomallei. Since there are no data available concerning the function of BsaS in vivo, we sought to determine the potential of B. pseudomallei BsaS as a target for inhibition of bacterial intracellular replication. To facilitate this study, a bsaS mutant was first characterized in order to serve as a control against which to assess treatment of cells infected with wild-type bacteria with inhibitor. Our study suggests that the TTSS3 ATPase BsaS is critical for B. pseudomallei virulence and is a putative target for inhibition by small-molecule ATPase inhibitors.

MATERIALS AND METHODS

Ethics statement.

Animal experiments were performed in accordance with the provisions of the “Prevention of Cruelty to Animal Act, 1986,” the “Australian Code of Practice for the Care and Use of Animals for Scientific Purposes, 7th edition, 2004,” and the Monash University Animal Welfare Committee Guidelines and Policies. The protocol was approved by the Monash Animal Research Platform (MARP)-2 Animal Ethics Committee (AEC) of Monash University (AEC number MARP/2011/067: Pathogenesis in Melioidosis).

Bacterial strains and cell culture.

The bacterial strains and plasmids used are listed in Table 1. B. pseudomallei wild-type strain K96243 (33) and mutants were cultured in Luria-Bertani (LB) broth at 37°C. Escherichia coli strain SM10λpir was used as a conjugative donor of the λpir-dependent suicide replicon pDM4 (oriR6K, mobRP4, sacBR, and cat) or its derivatives (34). The mouse macrophage-like cell line RAW 264.7 was obtained from the American Type Culture Collection (Manassas, VA). Cells were maintained at 37°C in 5% CO₂ without antibiotics in RPMI 1640 medium (Invitrogen, Carlsbad, CA), supplemented with 10% (vol/vol) heat-inactivated fetal bovine serum (JRH Biosciences, Lenexa, KS). The RAW 264.7 cell line stably expressing green fluorescent protein (GFP)-LC3 was constructed as described previously (19). The rat anti-LAMP1 (lysosome-associated membrane protein 1) antibody was obtained from the Development Studies Hybridoma Bank (Department of Biological Sciences, The University of Iowa, Iowa City, IA). All chemicals were purchased from Sigma (St. Louis, MO) unless otherwise indicated.

TABLE 1.

Strains and plasmids used in this study

Strain or plasmid	Descriptiona	Source or reference
Strains
B. pseudomallei
K96243	Virulent Thai clinical isolate	33
K96243ΔbsaS::tetA(C)	bsaS deletion mutant; Tetr	This study
K96243ΔbsaS::tetA(C)(pBHR1)	bsaS deletion mutant; Tetr, containing the original pBHR1 plasmid	This study
K96243ΔbsaS::tetA(C) (pBHR1comp)	bsaS deletion mutant; Tetr; containing the pBHR1 plasmid with the inserted bsaS complementation construct	This study
E. coli K-12 SM10λpir	For propagation of λpir-dependent plasmid pDM4. thi-1 thr leu tonA lacY supE recA::RP4-2-Tc::Mu Kmr λ′	34
Plasmids
pDM4	λpir dependent; Cmr; sacBR negative selection	34
pDM4::bsaS::tetA(C)	The pDM4 vector containing the bsaS mutagenesis construct	This study
pBHR1	Mob, rep; Cmr Kanr	14
pBHR1comp	pBHR1 plasmid carrying the bsaS complementation construct	This study

^aCm^r, chloramphenicol resistance; Tet^r, tetracycline resistance; Km^r, kanamycin resistance.

Mutagenesis of bsaS.

All oligonucleotide primers are listed in Table 2. A bsaS deletion mutant was constructed by double-crossover allelic exchange using the λpir-dependent vector pDM4, which carries the negative-selectable marker sacB (34). The primers 1541up5 and 1541up3 were used to amplify a 678-bp fragment flanked by XbaI and BglII restriction sites and encompassing 479 bp upstream of the bsaS coding sequence and the first 199 bp of the coding sequence. The primers 1541do5 and 1541do3 were used to amplify an 850-bp fragment flanked by BglII and XbaI restriction sites and encompassing 128 bp of the 3′ end of the coding sequence and 722 bp of downstream sequence. These two fragments, together with a tetracycline resistance cassette tetA(C) flanked by BglII restriction sites, were introduced by three-way ligation into pBluescript KS phagemid (Stratagene, La Jolla, CA), generating a plasmid containing a bsaS gene cassette in which a tetA gene had been inserted and having a 981-bp internal deletion in bsaS. An XbaI fragment containing the bsaS deletion cassette was released from this plasmid and ligated into XbaI-digested pDM4. This plasmid was then introduced by transformation into E. coli SM10λpir. For mobilization into B. pseudomallei, overnight cultures of E. coli and B. pseudomallei were subcultured, grown to mid-log phase, spotted together on LB agar plates, and grown overnight at 37°C. Transconjugants were subsequently selected on LB agar containing tetracycline (25 μg/ml) and gentamicin (64 μg/ml). Transconjugants were screened for chloramphenicol sensitivity (40 μg/ml), and the identity of the presumptive deletion mutant [K96243 ΔbsaS::tetA(C)] was confirmed by PCR and sequence analysis using primers (Table 2) specific for the tetA(C) gene (5424tet) and the sequence upstream or downstream of bsaS (1541up5out and 1541do3out, respectively).

TABLE 2.

Primers used in this study

Primer	Sequence (5′-3′)	Description
1541up5	GGCCTCTAGAGGAGGTGCTGTCGTTCGGCG	Forward primer upstream of bsaS; specifying an XbaI site
1541up3	GGCCAGATCTTGCTCAGGATCGCGGTGTCG	Reverse primer within bsaS; specifying a BglII site
1541do5	GGCCAGATCTGCGAGAACCCGGAGAACGACG	Forward primer within bsaS; specifying a BglII site
1541do3	GGCCTCTAGAAACCGCAACGCGAACGCCC	Reverse primer downstream of bsaS; specifying an XbaI site
5424tet	GCTGTCGGAATGGACGATAT	Forward primer located 3′ of tetA(C)
1541up5out	TCTGAACCTGGAGCCCGCGA	Forward primer upstream of bsaS and outside the region used to make the mutagenesis construct
1541do3out	CCCACGCCCGCTTCGATACC	Reverse primer downstream of bsaS and outside the region used to make the mutagenesis construct
1541comp5	GGGGAATTCGAGCGAGAGGGTCGCGCGATGAAGGCCG	Forward primer for complementation of bsaS; specifying an EcoRI site
1541comp3	GCGCCATGGTGCCGCGCACGCGCATCAG	Reverse primer for complementation of bsaS; specifying an NcoI site
1540infor	GATGGTTGAACCCATGACCT	Forward primer internal to bsaT (BPSS1540)
1540inrev	AATCGAAGTCGACGATCTGC	Reverse primer internal to bsaT
1539infor	TTGAAGCTGTAGCGCACCTG	Forward primer internal to bsaU (BPSS1539)
1539inrev	CGCGAGGAAAAAGACGAAGC	Reverse primer internal to bsaU

Complementation of the bsaS mutant.

A 1,566-bp fragment, spanning the entire bsaS open reading frame was amplified by PCR with primers 1541comp5 and 1541comp3 (Table 2). This fragment was then ligated into the EcoRI and NcoI sites of pBHR1 (14). The resulting plasmid (pBHR1comp) was introduced into E. coli SM10 λpir and subsequently transferred by conjugation into the bsaS deletion mutant.

RT-PCR.

Total RNA was extracted from mid-log-phase cultures of B. pseudomallei wild-type and bsaS mutant strains using TRIzol reagent (Invitrogen). Reverse transcription-PCR (RT-PCR) was performed using the SuperScript III one-step RT-PCR system (Invitrogen) according to the manufacturer's manual. Amplification of bsaT and bsaU cDNA was performed using specific primer pairs, 1540infor/1540inrev and 1539infor/1539inrev, respectively (Table 2), in a T-Gradient thermal cycler (Biometra, Gottingen, Germany) with the following cycling conditions: 95°C (for 30 s), 56°C (for 30 s), and 72°C (for 60 s) for 30 cycles.

ATPase inhibitors.

The small molecule designated compound 939 (molecular formula: C₂₃H₁₈N₄O₃S₂) used here as a potential TTSS ATPase inhibitor was purchased from Enamine (Kiev, Ukraine [http://www.enamine.net]) as dry powder. A stock solution (50 mM) of the compound was prepared in dimethyl sulfoxide (DMSO) and stored at −20°C in amber glass vials. Serial dilutions of this stock with DMSO were made when required and stored at −20°C, also in amber glass vials. Compound 939 at the final concentration of 10 μM was added to cultured bacteria or RAW 264.7 cells at the indicated time points. The same volume of DMSO was added to the “untreated” control cultures. At this concentration, the compound did not show appreciable cell toxicity against mammalian cells (data not shown), in agreement with previous findings (31).

Bacterial replication assay.

RAW 264.7 cells (seeded at 10⁵ cells/well) in 24-well trays (BD Biosciences, Bedford, MA) were infected with B. pseudomallei at a multiplicity of infection of 10:1 and then incubated at 37°C for 1 h to allow bacterial invasion. Infected cells were washed three times with phosphate-buffered saline (pH 7.2; PBS) and then incubated in fresh medium supplemented with kanamycin (800 μg/ml) and ceftazidime (100 μg/ml) to kill extracellular bacteria. Compound 939 was added at a 10 μM concentration as required either 4 h before or at the time of bacterial addition. At 2, 4, and 6 h postinfection (p.i.), the cells were washed four times with PBS, and intracellular bacteria were released by the addition of 0.1% (vol/vol) Triton X-100. Serial dilutions of each lysate were plated onto LB agar, and the numbers of intracellular bacteria were enumerated by direct colony counts after 48 h of incubation. Bacterial survival was normalized to counts obtained at 2 h p.i. and is presented as the relative survival (%).

Immunoblotting of BopE in bacterial lysates and culture supernatants.

To detect BopE, B. pseudomallei wild-type or bsaS mutant cells were grown to mid-log phase in 10 ml of LB broth at 37°C. Compound 939 was added at a 10 μM concentration when needed. Bacteria were harvested by centrifugation, and culture supernatants were filtered through a 0.22-μm-pore-size filter (Millipore, Billerica, MA) prior to precipitation of proteins by the addition of trichloroacetic acid to a final concentration of 10% (wt/vol). Cell pellets were lysed by the addition of 0.5 ml of B-PERII bacterial protein extraction reagent (Pierce, Rockford, IL), and lysates were processed as described previously (22). Proteins (∼20 μg) recovered from cell lysate or culture supernatants were separated on 12% sodium dodecyl sulfate (SDS)-polyacrylamide gels and transferred to polyvinylidene difluoride membranes (Pall Life Sciences, Ann Arbor, MI). Membranes were probed with rabbit anti-BopE (1:100) antiserum (21), followed by horseradish peroxidase-conjugated donkey anti-rabbit IgG serum (1:5,000; GE Healthcare, Pittsburgh, PA). Signals were detected using SuperSignal West Femto chemiluminescent substrate (Thermo Scientific, Rockford, IL).

Immunofluorescence and fluorescence microscopy.

RAW 264.7 cells expressing GFP-LC3 were cultured in 24-well trays containing 13-mm-diameter glass coverslips (ProSciTech, Thuringowa, Australia). At the indicated time points postinfection (p.i.), the cells were fixed with methanol for 10 min and blocked for 1 h in PBS containing 1% (wt/vol) bovine serum albumin and 0.1% (vol/vol) Triton X-100. Coverslips were incubated with rabbit anti-B. pseudomallei outer membrane serum (19) at a 1:100 dilution. The fluorescently labeled secondary antibody Alexa Fluor 405-conjugated goat anti-rabbit IgG serum (Molecular Probes, Eugene, OR) was used at a 1:250 dilution. LAMP1 was detected with rat anti-LAMP1 antibody at a 1:100 dilution, followed by Alexa Flour 568-conjugated goat anti-rat IgG serum (Molecular Probes) at a 1:250 dilution. Stained cells were washed with PBS, coverslips mounted in Permafluor aqueous mounting medium (Immunotech, Marseille, France), and visualized using an Olympus FV500 confocal laser scanning fluorescence microscope equipped with a 1.2 NA water immersion lens (Olympus 60× UPlanapo). Image analysis and processing was performed using Olympus FluoView Tiempo software (version 5.01) and the public domain software ImageJ 1.36. The intracellular location of bacteria was confirmed by inspection of images in a Z-stack series. For actin staining, RAW 264.7 cells, cultured on coverslips at the indicated time points p.i., were fixed in 3.5% (wt/vol) paraformaldehyde and 0.1% (vol/vol) Triton X-100 for 10 min prior to immunostaining as described above. Stained cells were further incubated for 50 min with Alexa Flour 488-conjugated phalloidin (Molecular Probes) at a 1:500 dilution to visualize cellular actin filaments.

TEM.

B. pseudomallei-infected RAW 264.7 cells and uninfected cells were prepared, fixed, sectioned, and mounted for transmission electron microscopy (TEM) as described previously (25). The sections were visualized at 80 kV on a JEM1011 transmission electron microscope (JEOL, Tokyo, Japan) equipped with a charge-coupled device camera (Gatan, Pleasanton, CA). Images were processed and analyzed using Olympus iTEM software.

Mouse virulence assay.

Female 6- to 8-week-old BALB/c mice were housed under specific-pathogen-free conditions on a 12-h-light/12-h-dark cycle with free access to food and water. For each infection, aliquots of B. pseudomallei wild-type or bsaS mutant bacteria were thawed from −80°C frozen stocks, subcultured in fresh medium containing appropriate antibiotics, and grown for 4 h to an optical density at 600 nm of 0.8 (mid-log growth phase corresponding to 5 × 10⁸ CFU/ml). Groups of seven mice were infected via the intranasal route with 20 μl of wild-type or mutant bacteria at doses of 6 × 10⁶ CFU or 6 × 10⁴ CFU. Viable counts were performed to confirm the inoculation dose. Mice were then monitored five times daily for symptoms of infection for 10 days and euthanized if moribund in accordance with animal ethics requirements. Survival data were displayed as Kaplan-Meier curves, the difference in overall survival was assessed using the Fisher exact test, and the difference in time to death was analyzed using a log-rank Mantel-Cox test. To enumerate bacteria in the spleen, the organ was removed aseptically and homogenized in sterile PBS by passage through a 23G syringe needle. Serial dilutions of tissue homogenates were plated onto LB agar, and colonies were enumerated after 24 h of incubation at 37°C. For the bsaS insertion mutant, the stability of the pDM4 insertion in bsaS was verified by plating of the recovered bacteria from mice infected with the bsaS mutant on LB medium, either containing 50 μg of chloramphenicol/ml or not. Recovered bacteria were further tested for retention of the mutation by PCR.

Statistical analysis.

For quantification studies, at least 100 bacteria were counted for each condition in each experiment, unless otherwise indicated. Values were expressed as means ± the standard errors of the mean (SEM). Differences between groups were analyzed by a two-tailed, two-sample, unequal variance Student t test or by analysis of variance with the Dunn post hoc test where appropriate. A P value of <0.05 was considered statistically significant.

RESULTS AND DISCUSSION

Secretion of the TTSS3 effector BopE is blocked by treatment of wild-type B. pseudomallei with the ATPase inhibitor compound 939.

A small-molecule ATPase inhibitor, compound 939, was recently identified as an efficient inhibitor of both the Y. pestis YscN and the B. mallei BsaS TTSS ATPases in vitro (31). We aimed to obtain data in support of the function of B. pseudomallei BsaS as a TTSS component, which has not been proven experimentally, and also to examine the potential of BsaS as a target for therapeutic intervention. To facilitate this study, a bsaS deletion mutant was constructed and characterized in order to serve as a control against which to assess cells infected with wild-type bacteria and treated with compound 939. To determine the effect of the inhibitor on BsaS function and the secretion of TTSS3 effectors, whole-cell lysates and culture supernatants of wild-type and bsaS deletion mutant bacteria were evaluated by immunoblotting for the expression and secretion of the well-characterized TTSS3 effector BopE (21). BopE was detected in the whole-cell lysate of both the wild type (whether treated with compound 939 or not) and the bsaS mutant strain and in the culture supernatant of the wild-type strain (Fig. 1A, lane 1 and lanes 4 to 6). However, BopE was not detected in the culture supernatant of the bsaS mutant nor in the culture supernatant of the wild-type strain treated with compound 939 (Fig. 1A, lanes 2 and 3, respectively). Since the level of BopE protein detected in the whole-cell lysates was comparable between strains, these results indicate that the loss of bsaS function by either inhibitor treatment or mutation of bsaS can prevent the secretion of BopE into the culture supernatant.

FIG 1.

B. pseudomallei bsaS deletion or treatment of wild-type bacteria with the ATPase inhibitor compound 939 blocks the secretion of the TTSS3 effector BopE. (A) Western blot probed with rabbit anti-BopE antiserum showing the expression and secretion of BopE protein in B. pseudomallei wild-type strain (untreated or treated with compound 939) and bsaS deletion mutant. Lanes 1 and 2 indicate culture supernatants from the wild-type and bsaS mutant, respectively; lane 3 indicates culture supernatant from the wild-type strain treated with compound 939; lanes 4 and 5 indicate whole-cell lysates of the wild type and bsaS deletion mutant, respectively; lane 6 indicates whole-cell lysate from the wild-type strain treated with compound 939. The BopE protein is indicated by the black arrow. (B) Coomassie blue-stained SDS-PAGE gel of proteins derived from culture supernatant of B. pseudomallei wild-type strain (lane 2), bsaS deletion mutant (lane 3), or wild-type strain treated with compound 939 (lane 4). Lane 1, molecular mass markers (kDa).

Loss of BopE secretion has also been observed in strains with mutations in other TTSS3 secretion apparatus genes, including bsaU, bsaQ, or bsaZ (16, 22, 23). The data presented here show that BsaS is essential for the secretion of BopE and that treatment of wild-type bacteria with the TTSS ATPase inhibitor, compound 939, can also block BopE secretion. Thus, BsaS is essential for the delivery of TTSS3 effector proteins and is sensitive to inhibition by the inhibitor compound 939. Notably these results suggest that the BsaS paralogs encoded by the TTSS1 and TTSS2 clusters (14) cannot cross-complement for BsaS function. However, since the three TTSS clusters in B. pseudomallei are involved in infection of different hosts (12), we cannot exclude the possibility that the TTSS1 and/or TTSS2 genes were not expressed or activated during infection of cultured murine cells and therefore that the BsaS paralogs had no opportunity to cross-complement.

Inhibitor treatment or mutation of bsaS reduces intracellular survival of bacteria in RAW 264.7 cells.

Since the TTSS is critical for B. pseudomallei intracellular survival, we hypothesized that the loss of BsaS function by either bsaS mutation or inhibitor treatment would lead to increased susceptibility to killing by mouse macrophages. In order to address this hypothesis, we measured bacterial replication in RAW 264.7 cells infected with wild-type or bsaS mutant bacteria and either treated with inhibitor at the time of bacterial addition or left untreated. Significantly reduced intracellular survival of wild-type bacteria was observed in RAW 264.7 cells treated with inhibitor compound 939 compared to untreated cells (Fig. 2B), the percentage reduction was 53% at 4 h and 77% at 6 h p.i. A similar reduction in survival of bsaS mutant bacteria was observed at 4 (49%) and 6 (78%) h p.i. in untreated RAW 264.7 cells (Fig. 2A). We also assessed the effect of addition of compound 939 4 h prior to infection of RAW 264.7 cells with B. pseudomallei; again, a similar level of reduction in intracellular survival was observed (Fig. 2B). Compound 939 was therefore added at the onset of infection in all subsequent experiments.

FIG 2.

The B. pseudomallei wild-type strain treated with ATPase inhibitor and the bsaS deletion mutant show significantly decreased replication in RAW 264.7 cells. (A) The relative intracellular survival of B. pseudomallei wild-type and bsaS deletion mutant bacteria in RAW 264.7 cells was determined at 2, 4, and 6 h p.i. (B) The relative intracellular survival of B. pseudomallei wild-type bacteria in RAW 264.7 cells, untreated or treated with inhibitor 939 (with or without 4 h of preincubation) was determined at 2, 4, and 6 h p.i. Bacterial survival was normalized to the CFU counts obtained at 2 h p.i. and presented as the relative survival (%). The data represent the means ± the SEM of three separate experiments, each carried out in triplicate. Where shown, an asterisk (*) indicates P < 0.05 relative to the wild-type strain at each time point.

Inhibitor treatment or bsaS mutation reduces efficiency of escape from phagosomes.

Two components of the B. pseudomallei TTSS3 apparatus, the inner-membrane export subunit BsaZ and the needle tip translocator BipD, are essential for bacterial escape from the phagosome, intracellular survival, and virulence in BALB/c mice (16, 17, 25). We therefore investigated whether the impaired intracellular replication of the bsaS mutant and wild-type bacteria treated with the inhibitor compound 939 resulted from a reduced ability to escape from phagosomes. We used TEM to view sections prepared from B. pseudomallei-infected RAW 264.7 cells and determine the proportions of bsaS mutant bacteria or wild-type bacteria treated with compound 939 within phagosomes, autophagosomes, or free in the cytoplasm (Fig. 3A to C). At each of the time points analyzed, bacteria were only identified either within phagosomes or free in the cytoplasm. No bacteria were observed in double membrane autophagosomes, confirming our earlier observations (25) that cytoplasmic bacteria are rarely targeted by canonical autophagy.

FIG 3.

The B. pseudomallei bsaS deletion mutant or wild-type bacteria treated with ATPase inhibitor show reduced efficiency of phagosome escape in RAW 264.7 cells. Transmission electron micrographs of RAW 264.7 cells infected with wild-type bacteria (A), bsaS deletion mutant (B), or wild-type bacteria treated with ATPase inhibitor (C) were examined to determine the intracellular location of individual infecting bacteria. The percentage of bacteria in the phagosomes of infected cells at 2, 4, and 6 h p.i. were determined by direct counting of micrographs (D and E). The data represent the means ± the SEM of three separate experiments (n = 50 bacteria). Where shown, an asterisk (*) indicates P < 0.05 relative to the untreated wild-type strain at each time point.

The proportion of wild-type bacteria found within phagosomes was 71, 46, and 22% at 2, 4, and 6 h p.i., respectively (Fig. 3D). The bsaS deletion mutant bacteria were only observed within phagosomes at 2 h p.i., and even at 4 and 6 h p.i. 99 and 94% of the bacteria, respectively, remained in phagosomes (Fig. 3D). The observation of a small number of free bsaS mutant bacteria at 6 h p.i. suggests that the bsaS mutant can escape from phagosomes at a delayed time point compared to wild-type bacteria, a phenotype described for other TTSS3-defective strains, including a bsaZ mutant (35), a bsaU mutant (23), and a bipD mutant (25). In support of these findings, we observed impaired actin-based motility and MNGC formation in the bsaS mutant (data not shown). The proportions of wild-type bacteria within phagosomes in RAW 264.7 cells infected with the wild-type strain treated with the ATPase inhibitor compound 939 were 97, 85, and 71% at 2, 4, and 6 h p.i., respectively (Fig. 3E), i.e., significantly higher than that observed for infection with wild-type bacteria without 939 treatment. Although treatment with the ATPase inhibitor clearly reduced the ability of B. pseudomallei to escape from phagosomes, it did not reduce escape to the same levels observed for the bsaS mutant, indicating that drug-target interaction does not fully inactivate BsaS activity. Collectively these results show that the TTSS3 ATPase BsaS is required for efficient phagosomal escape and that compound 939 is able to significantly inhibit this function.

Inhibitor treatment or bsaS mutation results in increased colocalization of B. pseudomallei with LC3 and LAMP1.

Since the bsaS deletion mutant and wild-type bacteria treated with the inhibitor compound 939 were unable to escape efficiently from phagosomes, we predicted that these trapped bacteria would be more vulnerable to LC3-associated phagocytosis (LAP) and killing within phagolysosomes. We have shown that B. pseudomallei is subject to LAP in RAW 264.7 cells and that a functional TTSS3 and the effector BopA are required for evasion of this process (19, 25, 36). In LAP the autophagy marker protein LC3 is recruited to phagosomes and stimulates phagosomal maturation, thereby leading to enhanced killing of the bacteria contained therein (37). To test the hypothesis that the decreased intracellular survival of bacteria was associated with LAP, we investigated the colocalization of B. pseudomallei with LC3 and LAMP1 in RAW 264.7 cells (Fig. 4). Both the bsaS mutant and wild-type bacteria treated with inhibitor compound 939 displayed increased levels of colocalization with LAMP1 (Fig. 4B) and GFP-LC3 (Fig. 4C) compared to the wild-type strain at 2, 4, and 6 h p.i. As shown above, the bsaS mutant showed no escape from phagosomes at 2 h p.i., and only 6% of bacteria had escaped at 6 h p.i. (Fig. 3D). Furthermore, we saw no bacteria within double membrane vesicles at any of the analyzed time points. Therefore, the association of bacteria with GFP-LC3 at these time points must result from GFP-LC3 recruitment to bacterium-containing phagosomes rather than bacterial uptake by canonical autophagosomes. We also determined dual colocalization of bacteria with GFP-LC3 and LAMP1 (Fig. 4D), as a measure of the bacteria in LC3-recruited phagosomes fused with lysosomes. The percentage of inhibitor-treated wild-type or bsaS mutant bacteria colocalized with both GFP-LC3 and LAMP1 was also significantly higher than that observed for the wild-type strain not treated with inhibitor at each of the time points tested (Fig. 4D). Taken together, these data indicate that loss of BsaS function either via direct genetic inactivation or treatment with the inhibitor compound 939 results in increased susceptibility of B. pseudomallei to LAP in RAW 264.7 cells, leading to elevated levels of intracellular killing. As noted above, TEM analysis did not identify any intracellular bacteria enclosed within double-membrane autophagosome structures (Fig. 3). These results suggest that inhibition of BsaS with compound 939 confines B. pseudomallei to phagosomes, which in turn leads to increased targeting and killing by LAP.

FIG 4.

Intracellular B. pseudomallei bsaS deletion mutant or wild-type bacteria in RAW 264.7 cells treated with ATPase inhibitor show enhanced colocalization with autophagy marker protein LC3 and the lysosome marker LAMP1. (A) Confocal micrographs of RAW 264.7 cells expressing GFP-LC3 (green) and infected with B. pseudomallei (Bp). Cells were fixed at 2 h p.i., permeabilized, and stained for B. pseudomallei (blue) and LAMP1 (red). Arrows indicate bacteria associated with LC3 that are also within LAMP1-positive vacuoles. Bacterial colocalization with LC3 or LAMP1 was defined by the presence of labeled B. pseudomallei (blue) which were fully overlaid by intense green/red or fully contained within a green/red ring. Scale bar, 5 μm. (B to D) Quantitative analysis of bacterial colocalization with LC3 (C), LAMP1 (B), or both LC3 and LAMP1 (D) in RAW 264.7 cells infected with B. pseudomallei wild-type strain (untreated or treated with compound 939) or the bsaS deletion mutant at 2, 4, and 6 h p.i. The data represent the means ± the SEM of three separate experiments (n = 100 bacteria). Where shown, an asterisk (*) indicates P < 0.05 relative to the wild-type strain at each time point.

The bsaS deletion mutant is highly attenuated for virulence in BALB/c mice.

To determine the importance of bsaS in B. pseudomallei pathogenesis, the bsaS mutant was tested for virulence in the well-established BALB/c mouse model of acute melioidosis (38). Mice were challenged via the intranasal route with either the wild-type or mutant strain at doses of 6 × 10⁶ CFU (Fig. 5A) or 6 × 10⁴ CFU (Fig. 5B) and then monitored for 10 days. All mice infected with 6 × 10⁴ CFU of the wild-type strain showed signs of acute illness within 24 h p.i. and were euthanized by 112 h p.i. Similarly, mice infected with 6 × 10⁶ CFU of the wild-type strain all showed disease signs by 24 h p.i. and were euthanized by 98 h p.i. However, all mice, whether infected with a high or a low dose of bsaS mutant bacteria, showed no signs of disease over the 10 days of the experiment (Fig. 5). These results demonstrate unequivocally that the bsaS deletion mutant is highly attenuated for virulence in BALB/c mice, which is in consistent with the attenuated virulence of two other TTSS3 mutants reported previously (17).

FIG 5.

The B. pseudomallei bsaS deletion mutant is attenuated in mice. The survival of BALB/c mice inoculated via the intranasal route with 6 × 10⁶ CFU (A) or 6 × 10⁴ CFU (B) of B. pseudomallei K96243 wild type (■) or the bsaS deletion mutant (○) was determined (n = 7 per group). The P value indicates the difference between mice infected with different strains analyzed using the log-rank Mantel-Cox test.

Complementation of bsaS function.

We sought to transcomplement the bsaS deletion mutant by introduction of the intact gene on plasmid pBHR1. The strain carrying this construct showed statistically significant reversal of increased colocalization with LC3 and partial reversal of impaired intracellular survival (Fig. 6). However, neither phenotype was restored fully to wild-type levels. We hypothesized that this was due to loss of the pBHR1 plasmid from B. pseudomallei when maintained under the nonselective conditions that follow infection of host cells. To test this, B. pseudomallei carrying the plasmid pBHR1comp (Table 1) was grown in vitro in medium without antibiotic for 6 h (the longest time analyzed in RAW264.7 cell culture experiments). Colonies were then recovered on antibiotic-free plates and patched to determine the percentage of plasmid retention. At this time point, 28% of the colonies had lost the plasmid, indicating that pBHR1 shows low stability in the absence of antibiotic selection. Furthermore, we analyzed the stability of pBHR1 constructs in mice and over 20 h observed 100% plasmid loss (P. Treerat, M. Prescott, B. Adler, R. Devenish, and J. D. Boyce, unpublished data). Therefore, the complemented bsaS mutant was not tested in the mouse infection model.

FIG 6.

Partial complementation of the bsaS deletion mutant. Quantitative analysis of bacterial colocalization with LC3 (A) and the relative bacterial survival (B) in RAW 264.7 cells infected with B. pseudomallei wild-type strain (WT), the bsaS deletion mutant (bsaS), or the complemented strain (bsaS-comp) at 2, 4, and 6 h p.i. was performed. Where shown, an asterisk (*) indicates P < 0.05 relative to the bsaS mutant at each time point.

To provide additional evidence that the phenotypes of the bsaS deletion mutant did not arise from polarity effects on downstream genes, we performed RT-PCR analysis of the transcription of the downstream genes bsaT and bsaU. mRNA was extracted from wild-type and bsaS deletion mutant bacteria and used as the templates for RT-PCR with primer pairs specific for bsaT and bsaU (Table 2). RT-PCR products of the predicted sizes, 203 bp (bsaT) and 861 bp (bsaU), were generated from cDNA derived from both wild-type and bsaS mutant strains (Fig. 7), indicating that the inactivation of bsaS did not disrupt the transcription of the downstream genes. These results indicate that the phenotypes displayed by the deletion mutant most likely arise directly from the inactivation of bsaS, given that the expression of downstream genes, and by inference the production of the corresponding proteins, was unaffected. Furthermore, there is strong correspondence between the phenotypes of the bsaS mutant and the wild-type strain treated with the previously characterized (31) ATPase inhibitor compound 939.

FIG 7.

RT-PCR analysis of bsaT and bsaU gene transcription in B. pseudomallei wild-type and bsaS deletion mutant strains. (A) Schematic diagram showing the genetic organization of bsaSTU as part of TTSS3 cluster and the position of primers (1540f/1540r for bsaT or 1539f/1539r for bsaU) used for RT-PCR. (B) The transcription of bsaT (top panel) or bsaU (lower panel) was analyzed using RT-PCR with specific primers. PCR products of different cDNA/DNA samples: lane 2, bsaS deletion mutant cDNA; lane 3, bsaS deletion mutant cDNA, no RT control; lane 4, wild-type cDNA; lane 5, wild type cDNA, no RT control; lane 6, wild-type genomic DNA control; lane 7, no DNA control. DNA size markers are shown in lane 1.

The small-molecule inhibitor of the TTSS3 ATPase, compound 939, has potential as a therapeutic treatment against melioidosis.

The TTSS of pathogenic bacteria is considered a promising target for the development of anti-infection therapeutics because it is highly conserved in structure and function and has an essential role in virulence (39–41). Alternative treatment strategies are required urgently for many TTSS-containing pathogens, since many, including B. pseudomallei, have developed resistance to multiple antibiotics. Over the last decade many natural product and synthetic compound inhibitors targeting bacterial TTSS have been identified by high-throughput screening or rational design methods (39–41). However, in the majority of cases the TTSS component targeted by the inhibitor remains unknown, and thus very few have been fully investigated for their mechanism of action (42). In the present study, we used compound 939, which has been identified as an efficient inhibitor of Y. pestis YscN and B. mallei BsaS TTSS ATPase in vitro (31). The data presented here strongly suggest this small-molecule TTSS ATPase inhibitor can efficiently inhibit B. pseudomallei BsaS and consequently is able to interfere with the downstream functions of B. pseudomallei TTSS BsaS, including BopE secretion, phagosomal escape, evasion of LAP, and intracellular replication. These findings offer promise that compound 939 (or its derivatives) could be developed for therapeutic application to B. pseudomallei infections, expanding the treatment options for melioidosis.