SseJ Deacylase Activity by Salmonella enterica Serovar Typhimurium Promotes Virulence in Mice
http://www.100md.com
感染与免疫杂志 2005年第10期
Departments of Microbiology Genome Sciences Medicine, University of Washington, Seattle, Washington 98195
Infection, Immunity, Injury, and Repair Program, Hospital for Sick ChildrenDepartment of Molecular and Medical Genetics
University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Salmonella enterica serovar Typhimurium utilizes a type III secretion system (TTSS) encoded on Salmonella pathogenicity island-2 (SPI2) to promote intracellular replication during infection, but little is known about the molecular function of SPI2-translocated effectors and how they contribute to this process. SseJ is a SPI2 TTSS effector protein that is homologous to enzymes called glycerophospholipid-cholesterol acyltransferases and, following translocation, localizes to the Salmonella-containing vacuole and Salmonella-induced filaments. Full virulence requires SseJ, as sseJ null mutants exhibit decreased replication in cultured cells and host tissues. This work demonstrates that SseJ is an enzyme with deacylase activity in vitro and identifies three active-site residues. Catalytic SseJ mutants display wild-type translocation and subcellular localization but fail to complement the virulence defect of an sseJ null mutant. In contrast to the wild type, SseJ catalytic mutants fail to down regulate Salmonella-induced filament formation and fail to restore the sifA null mutant phenotype of loss of phagosomal membrane to sifA sseJ null double mutants, suggesting that wild-type SseJ modifies the vacuolar membrane. This is the first demonstration of an enzymatic activity for a SPI2 effector protein and provides support for the hypothesis that the deacylation of lipids on the Salmonella-containing vacuole membrane is important to bacterial pathogenesis.
INTRODUCTION
Salmonellae are gram-negative bacteria that are capable of infecting a wide range of animals, such as cattle, poultry, snakes, and humans, leading to various diseases (35). Salmonella enterica serovar Typhimurium causes gastroenteritis in humans, while in inbred mice with innate immune defects, it causes severe systemic disease. During systemic infection of mice, serovar Typhimurium survives and replicates intracellularly in a modified phagosomal compartment enriched in cholesterol (11) and efficiently avoids destruction by phagocytes (32).
S. enterica serovar Typhimurium utilizes two type III secretion systems (TTSS) to directly translocate virulence determinants, called effector proteins, from the bacterial cytoplasm into the host cell cytoplasm (21). The TTSS encoded on Salmonella pathogenicity island-1 (SPI1) is expressed upon contact with host cells and translocates SPI1-specific effectors that alter cell signaling, induce membrane ruffling, and disrupt actin polymerization in order to promote bacterial uptake and modulate host inflammatory responses (16). Following internalization, the SPI2 TTSS is induced and translocates a different set of SPI2-specific effector proteins across the phagosomal membrane (20, 36). These effectors inhibit host cell processes, such as NADPH oxidase and iNOS localization (12, 39), which allow serovar Typhimurium to replicate. Moreover, a SPI2 effector was shown to decrease surface presentation of major histocompatibility complex class II molecules in serovar Typhimurium-infected cells (31).
Following translocation, SPI2 effectors localize to different cellular compartments, including the nucleus, cytoplasm, cytoskeleton, Golgi apparatus, and phagosome (15, 19, 22, 23, 25, 26, 28, 34). Such differential localization patterns indicate that effectors manipulate numerous aspects of host cell function. SifA is a well-characterized SPI2 effector that is important for virulence in mice, as sifA null mutants are severely attenuated (37). Following translocation, SifA localizes to the Salmonella-containing vacuole (SCV) (6) and is required for the formation of Salmonella-induced filaments (Sifs) in epithelium-derived cells such as HeLa cells (37). Sifs are highly dynamic structures that appear to be tubular extensions of the SCV and contain lysosomal glycoproteins, such as lysosome-associated membrane protein-1 (LAMP-1) (4, 7). The formation of Sifs requires microtubules, and the timing of Sif appearance in epithelial cells is concomitant with bacterial replication (6, 7, 17). Although Sifs have been analyzed extensively, and their role during infection remains unknown, they likely represent the ability of translocated SPI2 effectors to manipulate vesicular trafficking to and from the SCV and provide an important phenotype that represents Salmonella-induced host cell alterations.
Several other SPI2 effectors, including SifB, SseJ, SopD2, PipB, PipB2, SseF, and SseG, also localize to the SCV and Sif structures in infected cells, suggesting that they function collectively, and possibly redundantly, to exert their effect on the SCV (15, 22, 23, 25, 33). In addition to SifA, the effectors SseF, SseG, and SopD2 have been shown to contribute to the formation of Sif, which highlights the complexity of this phenotype (22, 24). Results from recent studies of SCV and Sif-localizing effectors have supported the hypothesis that they alter the SCV via manipulation of host vesicular traffic; however, there is no biochemical evidence to indicate the molecular mechanism by which this is accomplished.
Among the known SPI2 effector proteins, SseJ is the only effector with a predicted biochemical activity. The amino-terminal 140 amino acids of SseJ are required for its translocation by S. enterica serovar Typhimurium (29), while its carboxy-terminus (residues 140 to 408) has approximately 29% homology to several members of the GDSL family of lipases (38). GDSL lipase activity is dependent on three amino acid residues (Ser, Asp, and His), which compose a catalytic triad and are present in SseJ (see alignment in reference 14). Among GDSL lipases, SseJ has the most conservation of catalytic regions with glycerophospholipid-cholesterol acyltransferase (GCAT) enzymes, which are proteins secreted by Aeromonas and Vibrio species that catalyze the transfer of fatty acid acyl chains from phospholipids to cholesterol at lipid-water interfaces (9, 10, 27).
Studies have shown that SseJ is required for the full virulence of serovar Typhimurium, as sseJ mutants are mildly defective for intracellular replication in cultured cells and have a virulence defect in the mouse infection model (15, 33). Interestingly, SseJ is one of only a few SPI2 effector proteins, including SifA, that lead to a virulence phenotype when deleted, which underscores the importance of this protein to Salmonella pathogenesis. Recent analysis of the dynamics of Sif formation has demonstrated that SseJ down regulates the formation of Sif (4), supporting the hypothesis that SseJ functions to manipulate the SCV and host vesicular trafficking. Additional important phenotypes that implicate SseJ in modifying host processes include destabilization of the SCV membrane of sifA deletion mutants and formation of globular membraneous compartments in cultured cells by ectopic expression of SseJ (33). Given the conserved amino acid similarities between SseJ and other GCAT enzymes and the putative role of SseJ in altering host trafficking processes, we hypothesized that SseJ has deacylase and acyltransferase enzymatic activities and that these activities are important for SseJ function in vivo. This work investigates the potential enzymatic activity of SseJ and tests its role in SseJ-associated phenotypes and the promotion of bacterial virulence in mice.
MATERIALS AND METHODS
Bacterial strains, plasmids, eukaryotic cell lines, and growth conditions. All bacterial strains and plasmids used are listed in Table 1. Salmonella enterica serovar Typhimurium and Escherichia coli strains were grown and maintained in Luria-Bertani (LB) broth or on plates at 37°C with antibiotics added at the following concentrations: ampicillin, 100 or 50 mg/ml; kanamycin, 45 mg/ml; chloramphenicol (Cam), 20 mg/ml; and tetracycline, 10 mg/ml, unless described otherwise. HeLa and RAW264.7 cell lines originated from the ATCC and were maintained as previously described (15).
Plasmid construction. The oligonucleotides used in this study are listed in Table 2. The His-SseJ plasmid for protein purification, pJAF08, was made by amplifying the chromosomal region carrying sseJ from serovar Typhimurium with primers 5' JAF11 and 3' JAF12 by PCR and was cloned into pET-15b (Novagen) using NdeI and BamHI enzymes to create an in-frame amino-terminal fusion between the six-His tag and SseJ. The wild-type SseJ-HA-expressing plasmid used for immunofluorescence and competitive index (CI) complementation studies is pJAF111, which was published previously (15). Plasmids pKF27, pKF28, pKF29, pKF33, pKF35, pKF36, pKF37, and pMBO76 were all produced by site-directed mutagenesis using QuikChange (Stratagene). Primers 5' KF43 and 3' KF44 produced an S151A mutation in plasmids pKF35 and pKF29, with pJAF111 and pJAF08 used as templates, respectively. Similarly, 5' KF49 and 3' KF50 produced a D247N mutation in pKF36 and pKF33, with pJAF111 and pJAF08 used as templates, respectively; 5' KF51 and 3' KF52 produced a H384N mutation in pKF37 and pKF28, with pJAF111 and pJAF08 used as templates, respectively; and 5' KF47 and 3' KF48 produced an S153N mutation in pKF27, with pJAF08 used as a template. The triple mutant SseJ-HA plasmid pMBO76 was produced sequentially using QuikChange. First, the H384N mutation was incorporated onto the D247N-containing SseJ-HA plasmid pKF36 using 5' KF51 and 3' KF52, which created the intermediate double mutant (D247N and H384N) plasmid pMBO74. The double mutant plasmid was then used as a template to incorporate the S151A mutation using primers 5' KF43 and 3' KF44. All plasmids were sequenced to verify the mutations.
sseJ deletion strain construction. The chromosomal deletion of sseJ in serovar Typhimurium was constructed using the Red recombinase system (13). The Camr gene and FLP recognition target (FRT) sites from pKD3 were amplified by primers 5' MBO71 and 3' MBO72, which also contain flanking regions with homology to the 5' and 3' regions of sseJ, respectively. The resulting PCR fragment was purified and transformed into electrocompetent wild-type serovar Typhimurium cells containing pKD46 that were induced with arabinose at 30°C, and recombinants were selected on LB broth-Cam plates. Candidate colonies were verified by PCR. This approach yielded a serovar Typhimurium strain with the sequence encoding amino acids 3 to 406 of sseJ replaced by the FRT-Cam-FRT cassette.
Competitive index assay. Mice were ordered from Charles River Laboratories, Inc., and all experiments were performed with IACUC approval. Virulence phenotypes were tested by competitive index assay as described previously (15). Briefly, 6- to 8-week-old female BALB/c mice were inoculated intraperitoneally with a mixture of 5 x 104 organisms each of two serovar Typhimurium strains for a total of 105 bacteria in a 0.2-ml volume. Each strain was diluted from stationary-phase cultures grown overnight and contained either a chromosomal or stable plasmid-based antibiotic marker to allow the strains to be differentiated. The bacterial inoculum contained approximately equal concentrations of both strain 1 and strain 2 bacteria, and the ratio of strain 1 to strain 2 was confirmed by plating dilutions of the inoculum onto selective media. Forty-eight hours after infection, mice were euthanized by CO2 asphyxiation, the spleens were dissected, and each spleen was homogenized in sterile phosphate-buffered saline using a 2-ml glass Dounce homogenizer. Ratios of strain 1 to strain 2 bacteria in each spleen were calculated from bacterial counts produced by plating aliquots of 1:10 dilutions of homogenized spleen on selective media. The CI was calculated by dividing the ratio of strain 1 to strain 2 bacteria isolated from the spleen by the ratio of strain 1 to strain 2 bacteria inoculated into the mouse. CI results were determined by calculating the means ± standard deviations for at least 10 mice that were infected in three separate experiments. Statistical significance (P value) was determined by a Student two-tailed t test.
Immunofluorescence microscopy. Mammalian cells were seeded onto 12-mm coverslips in wells of 24-well tissue culture dishes in antibiotic-free media at a concentration of either 5 x 104 or 1 x 105 cells for HeLa or RAW264.7 cells, respectively. RAW264.7 cells were infected with stationary-phase serovar Typhimurium, while HeLa cells were infected with bacteria that were back-diluted to 1:100 for 3 h for log-phase growth. Both cell types were infected with bacteria at a multiplicity of infection of 100:1 in antibiotic-free media for 1 h, after which cells were washed with phosphate-buffered saline, and medium containing 10 μg gentamicin/ml was applied to the infected cells for the remainder of the infection. Following infection, cells were fixed with 2% paraformaldehyde in phosphate-buffered saline for at least 1 hour to overnight, washed with phosphate-buffered saline twice, and permeabilized with ice-cold acetone for 20 seconds. Prior to being stained, cells were blocked for 1 hour with a blocking solution composed of phosphate-buffered saline containing 5% donkey serum, 1% bovine serum albumin, 0.2% saponin, and 5 mM EDTA at room temperature. All primary and secondary antibodies were diluted in blocking solution and applied to cells for 1 hour, followed by washes with phosphate-buffered saline. To detect the hemagglutinin (HA) epitope, mouse anti-HA antibodies (Covance) were used at 1:500 and were probed with tetramethyl rhodamine isothiocyanate-conjugated donkey-anti-mouse antibodies (Sigma) at 1:200 as secondary antibodies. In RAW264.7 cells, mouse LAMP-1 was stained with rat anti-LAMP-1-fluorescein isothiocyanate-conjugated antibodies (BD Pharmingen) at a concentration of 5 μg/ml. Human LAMP-1 was stained using a 1:10 dilution of mouse anti-LAMP-1-fluorescein isothiocyanate conjugate (BD Pharmingen). Rabbit antilipopolysaccharide (anti-LPS) antibodies (Difco) were used at a ratio of 1:1,000. Secondary staining was performed using either Cy5- or tetramethyl rhodamine isothiocyanate-conjugated donkey-anti-rabbit secondary antibodies (Jackson) at 1:200. Stained coverslips were mounted onto slides using Vectashield (Vector Laboratories, Inc.) and sealed with nail polish. Slides were examined on an inverted Olympus microscope with a 60x oil immersion lens, and images were collected and deconvolved using SoftWoRx (Applied Precision) and then imported into Photoshop 7.0 (Adobe) to separate color layers.
Sif and LAMP-1 colocalization quantitation. HeLa cells were seeded at a density of 2.5 x 104 or 5 x 104 cells for Sif quantitation or LAMP-1 colocalization experiments, respectively. Cells were infected with serovar Typhimurium at a multiplicity of infection of 325:1 for 1 h, followed by gentamicin protection as described previously (4). At specific time points, the cells were fixed with 2.5% paraformaldehyde, permeabilized, and stained with anti-LPS and anti-human LAMP-1. Percent cells expressing Sifs was calculated by counting 100 infected cells per time point for each strain. Percent bacteria colocalizing with LAMP-1 was calculated by counting the number of LAMP-1-positive Salmonella isolates out of the total number of Salmonella isolates, with at least 100 Salmonella isolates counted per time point for each strain.
Protein purification. Strains containing His tag expression constructs were induced with 1 mM IPTG (isopropyl--D-thiogalactopyranoside) for 4 h prior to collection of cell pellets by centrifugation. The cell pellets were resuspended in cold 50 mM Tris (pH 7.4), and the cells were lysed using a French press. The soluble protein fraction was separated from the cellular debris in the lysate by ultracentrifugation for 1 h at 40,000 rpm. Supernatants were cleaned by passage through a 0.45-μm filter and loaded onto a NiSO4-charged HiTrap column (Amersham) to bind His-tagged protein. Nonspecific proteins were removed by washing the column with wash buffer (20 mM NaPO4, 100 mM imidazole, pH 7.4) prior to elution with elution buffer (20 mM NaPO4, 400 mM imidazole, pH 7.4). Fractions containing His-tagged protein were collected and concentrated with a Centricon-30 concentrating filter (Millipore), and imidazole was removed by dialyzing the protein in a Slide-A-Lyzer (Pierce) against 50 mM Tris (pH 7.2) overnight at 4°C.
Deacylase assay. The PNPB (para-nitrophenyl butyrate) deacylase assay is a modified version of that described previously (5). Briefly, purified enzyme of varied concentrations was added to 1-ml spectrophotometer cuvettes containing the assay reaction mix (65 μl 50 mM PNPB [Sigma] dissolved in 20 mM Tris [pH 7.4] and 3% acetonitrile, 100 μl 200 mM Tris [pH 7.4], and H2O to make 1 ml) and monitored in a Beckman DU 650 spectrophotometer at 400 nm for 30 min, with measurements taken every 30 seconds while the cuvette holder was equilibrated to 37°C. The spectrophotometer values obtained for the reaction mix without enzyme were subtracted from the values for each reaction mixture that contained enzyme at each time point to control for spontaneous PNPB deacylation. The rate (absorbance/time) was calculated using Kaleidagraph (Synergy Software) and was used to determine the micromoles of para-nitrophenol (PNP) formed per second using Beer's law (5).
RESULTS
SseJ has in vitro deacylase activity. The carboxy-terminal region of the SPI2 effector SseJ displays sequence homology to a family of enzymes termed GCATs, which typically exhibit deacylase activity as a first step in the reaction mechanism. To test whether SseJ posseses deacylase activity, we chose to use a soluble colorless substrate, PNPB, that forms a yellow product, PNP, upon removal of a short-chain fatty acid acyl group in a colorimetric assay (5). Amino-terminal His-tagged SseJ was purified from E. coli and tested for deacylase activity using this assay. We found that 5 μg of His-SseJ formed PNP at a rate of 9.16 x 10–4 μmol/second (Fig. 1A), demonstrating that SseJ functions as a deacylase. As a control, native purified Aeromonas salmonicida GCAT protein was analyzed, and 0.5 μg of protein exhibited the same rate of PNP production as 5 μg His-SseJ (data not shown), indicating that the activity of SseJ in this assay is approximately 10-fold lower.
SseJ mutants lack deacylase activity. Based on sequence homology, we hypothesized that the conserved Ser151, Asp247, and His384 residues of SseJ compose a catalytic triad (Fig. 1B). To test whether mutations of the catalytic residues had impaired activity in the PNPB assay, site-directed mutants were designed and purified from E. coli. We made three mutant versions of His-SseJ that contain single amino acid substitutions: S151A, D247N, and H384N. The predicted catalytic Ser151 residue is the middle serine residue in a G-D-S-L-S motif, which is present in all known GCAT enzymes. Since it was unknown whether Ser151 or Ser153 was the catalytic residue, a Ser153A mutant was also constructed. When tested for deacylase activity using the PNPB assay, the His-SseJ-S153A mutant protein produced 9.33 x 10–4 μmol of PNP per second, similar to the rate produced by wild-type SseJ (Fig. 1A). However, mutant SseJ proteins His-SseJ-S151A, His-SseJ-D247N, and His-SseJ-H384N exhibited rates of 2.16 x 10–4, 1.83 x 10–4, and 1.18 x 10–4 μmol PNP per second, respectively, an approximately fivefold reduction in activity compared to that of wild-type His-SseJ (Fig. 1A). These results suggest that Ser151, Asp247, and His384 residues are members of a catalytic triad and that Ser153 is not a catalytic residue. The reduction in catalytic activity was not due to unspecific destabilization of SseJ, as all mutants could be expressed to similar levels and exhibited identical peak fluorescence spectra. Peak fluorescence of wild-type His-SseJ in 50 mM Tris buffer when excited at 280 nm was found to be 335 nm. When assayed in the presence of 7 M urea to induce unfolding, the peak fluorescence shifted to 350 nm, indicating that a conformational change had occurred. The 335-nm peak fluorescence of each of the catalytic mutant proteins in 50 mM Tris buffer was identical to that of wild-type undenatured His-SseJ and indicated that the proteins were equally folded.
Translocated SseJ catalytic mutants maintain SCV localization. Although it was previously shown that ectopic expression of an SseJ-S151V mutant did not alter subcellular distribution from that of the wild type, suggesting that active-site residues do not influence localization (33), it was not known whether active-site mutations would affect the bacterial translocation of SseJ or SCV localization in the context of infection. To test this, HeLa cells were infected with a sseJ mutant expressing wild-type SseJ-HA; an SseJ-S151A-HA, SseJ-D247N-HA, or SseJ-H384N-HA catalytic mutant; or an SseJ-S151A-SsJ-D247N-SsJ-H384N-HA triple mutant. Expression of SseJ-HA from each of these stable, low-copy-number plasmids is under the control of its native promoter. Western blots of infected cell lysates probed with anti-HA antibodies indicated that the amount of translocated SseJ catalytic mutant protein was similar to wild-type SseJ-HA (data not shown), indicating that the catalytic mutations did not affect translocation by the TTSS apparatus. In addition, infected HeLa cells were fixed 10 hours postinfection; immunostained for the HA epitope, human LAMP-1, and LPS; and analyzed by fluorescence microscopy. Wild-type SseJ-HA colocalized with LAMP-1 and LPS, which indicates its localization to the SCV, and colocalized with the LAMP-1-positive Sif structures that radiate away from the SCV (Fig. 2A), confirming previous reports (15, 33). Similarly, we observed that SseJ-S151A-HA, SseJ-D247N-HA, SseJ-H384N-HA, and the SseJ-S151A-D247N-H384N-HA triple mutant localized to the SCV and Sifs like wild-type SseJ-HA (Fig. 2B to D and data not shown, respectively). Sif production was observed in infections with all four of the SseJ-HA mutants, and there were no gross morphological differences between the wild type and the catalytic mutants.
In cultured macrophages, SseJ- and LAMP-1-positive structures have been observed distant from the SCV (15). To ascertain whether SseJ catalytic mutants exhibited the same pattern of macrophage localization and trafficking, RAW264.7 cells were infected with a sseJ mutant expressing either wild-type SseJ-HA, SseJ-S151A-HA, SseJ-D247N-HA, or SseJ-H384N-HA and were immunostained with antibodies to the HA epitope, LPS, and mouse LAMP-1 10 hours postinfection. We observed that SseJ-HA mutants maintained the ability to localize to the SCV and colocalized with LAMP-1-positive compartments distinct from the SCV (Fig. 3). The localization of mutant SseJ in both HeLa and RAW264.7 cells indicates that catalytic residues of SseJ are not required for translocation or localization and demonstrates that the ability to localize to vacuolar membranes is not a consequence of enzymatic activity.
SseJ catalytic mutants fail to complement in vivo sseJ phenotypes. S. enterica serovar Typhimurium strains that lack sifA display two major phenotypes: failure to induce Sif formation and failure to remain within the Salmonella-containing vacuole as determined by LAMP-1 colocalization and electron microscopy (2). Additionally, sifA null mutants replicate in the cytosol, as opposed to the SCV, following loss of the vacuolar membrane (2, 3, 8). SseJ activity is implicated in sifA vacuolar membrane loss because sifA sseJ double mutants remain enclosed in the SCV (33). This phenotype can be exploited to measure SseJ activity in vivo. Specifically, to test whether SseJ catalytic activity plays a role in membrane loss by the sifA mutant, plasmids expressing either wild-type or catalytic mutant SseJ were assayed for the ability to restore the sifA phenotype to sifA sseJ double mutant bacteria by LAMP-1 colocalization. We observed that wild-type and sseJ bacteria maintained a high percentage of colocalization with LAMP-1, while sifA bacteria exhibited a marked decrease in LAMP-1 colocalization, and sifA sseJ LAMP-1 colocalization was significantly higher than with sifA bacteria (Fig. 4), confirming previous results (33). The LAMP-1 colocalization phenotype in cells infected with the sifA sseJ mutant expressing wild-type SseJ was complemented, as these cells exhibited a decrease in LAMP-1 colocalization similar to that of sifA bacteria. However, the LAMP-1 colocalization phenotype of the sifA sseJ mutant expressing the catalytic SseJ-HA triple mutant was not complemented, as these cells had a higher percentage of bacteria that continued to colocalize with LAMP-1, approximately at the same levels as sifA sseJ bacteria (Fig. 4). These results confirm that SseJ enzymatic activity allows sifA bacteria to escape from the SCV and provides strong evidence that the function of SseJ is to modify the SCV membrane.
Recently, we demonstrated that SseJ negatively regulates the formation of Sifs in epithelial cells (4); thus, we hypothesized that the catalytic activity of SseJ was involved in the down regulation of Sifs in infected cells. To test this, Sif formation was analyzed in HeLa cells that were infected with the wild type, the sseJ mutant, or the sseJ mutant expressing either wild-type SseJ-HA or the triple mutant SseJ-S151A-D247N-H384N-HA. Sif formation was quantified at 2-hour intervals between 4 and 10 hours postinfection. We determined that sseJ mutant-infected cells induced 30% more Sif-expressing cells than wild-type-infected cells, confirming previous results (4). HeLa cells infected with the sseJ mutant expressing wild-type SseJ-HA induced Sif expression at levels similar to those of wild-type serovar Typhimurium, indicating that the phenotype was complemented. In contrast, the sseJ mutant expressing triple mutant SseJ-HA failed to down regulate Sif expression, indicating that the ability of SseJ to negatively regulate Sif formation requires enzymatic activity (Fig. 5).
SseJ catalytic activity is required for full virulence in mice. To determine whether the catalytic activity of SseJ is required for serovar Typhimurium virulence in mice, we tested the virulence phenotype of sseJ strains expressing catalytic mutant SseJ-HA by the CI assay (1). First, the sseJ mutant was competed against the wild type in order to confirm CI results from previously published competitions with the sseJ mutant versus the wild type. As shown in Fig. 6, the sseJ mutant exhibits a competitive index defect of 0.56 ± 0.10, similar to CI values of 0.5 that we and others observed previously (15, 33). As a control, sseJ strains containing Kanr and Ampr plasmids were competed, and we verified that plasmid-based antibiotic resistance genes do not influence competitive index ratios, as these strains competed equally (1.0 ± 0.13). The sseJ mutant expressing wild-type SseJ-HA was found to compete equally against the wild type, with a competitive index of 0.90 ± 0.18, suggesting that low-copy-number-SseJ expression by its native promoter can restore virulence to the sseJ mutant. However, when the wild type was competed against the sseJ mutant expressing SseJ-S151A-HA, SseJ-D247N-HA, SseJ-H384N-HA, and SseJ-S151A-D247N-H384N-HA, CI phenotypes of 0.50 ± 0.09, 0.17 ± 0.05, 0.45 ± 0.13, and 0.44 ± 0.06 were observed (Fig. 6). These competitive indices show that SseJ catalytic mutants cannot complement the virulence defect of sseJ and clearly demonstrate that SseJ catalytic activity is required for full virulence in mice.
DISCUSSION
The ability of Salmonella to survive and replicate within mammalian cells is dependent on the SPI2 TTSS; however, the exact molecular contribution of SPI2 TTSS effector proteins remains unknown. In this study, we sought to analyze the potential enzymatic activity of the SPI2 effector SseJ and to determine whether the residues of this protein, which contain homology to GDSL lipases, contribute to the function of SseJ in vivo. Wild-type His-SseJ catalyzed the formation of PNP in the PNPB assay and clearly showed that SseJ has deacylase activity. We found that amino acid substitutions S151A, D247N, and H384N in SseJ led to a fivefold reduction in deacylase activity, suggesting that these residues form a catalytic triad. Strains expressing catalytic mutant SseJ-HA translocated mutant protein into host cells, which displayed wild-type localization to the Salmonella-containing phagosome. However, despite proper localization, SseJ catalytic mutants failed to complement the known cellular phenotypes attributed to SseJ, namely, down regulation of Sifs and destabilization of the SCV in sifA mutants. Most importantly, catalytically inactive SseJ mutants demonstrated a virulence defect in mice.
The deacylase activity exhibited by SseJ is approximately 10-fold lower than the activity of a purified A. salmonicida GCAT protein that is secreted extracellularly by type II secretion. The PNPB substrate is not likely to be physiologic for either protein but was chosen because it is water soluble. Differences in affinities for PNPB could explain the rate difference, and SseJ could have higher affinity for different substrates. However, it is interesting to speculate that the reduced enzymatic activity of SseJ compared to that of other GCAT proteins is physiologic. High levels of SseJ activity could have a deleterious effect on Salmonella survival within macrophages by perhaps leading to destabilization of the SCV membrane and the subsequent release of bacteria into the cytosol. Further studies will be required to identify the biologically relevant substrates of SseJ and whether modest enzyme activity is relevant to its in vivo function.
From our current study, we cannot conclude whether SseJ functions solely as a deacylase or whether it also has acyltransferase activity. The primary sequence indicates that SseJ is more similar to GCAT enzymes than to other GDSL lipases (see alignment in reference 14 and Fig. 1B). Altering the catalytic aspartic acid (D247) in the G-A-N-D-Y motif of SseJ to an asparagine (D247N) converts the motif to a G-A-N-N-Y motif like that found in PlaA, the Legionella lysophospholipase. In SseJ, the D247N mutation, and therefore the altered motif, results in inactive SseJ protein, which suggests that the active site of SseJ requires a conformation more like that of GCAT enzymes than of lysophospholipases or other GDSL lipases.
If SseJ possesses deacylase and acyltransferase activities, the production of esterified cholesterol may contribute to the effect of SseJ on the SCV. Cholesterol is enriched in the phagosomal compartment in serovar Typhimurium-infected cells (7, 11, 18) and is likely important for providing membrane rigidity and forming lipid rafts. Hence, esterified cholesterol in lipid rafts could disrupt cell-signaling platforms. Conversely, lipid and/or cholesterol modifications could inhibit the formation or interaction of molecular complexes on the SCV surface that are required for vesicular trafficking to or from the SCV.
This work supports the hypothesis that SseJ activity induces lipid modifications on the SCV membranes that alter the phagosomal compartments. SseJ likely acts in concert or sequentially with SifA and other effector proteins to promote vacuolar remodeling. It seems that Sif formation reflects an activity of SifA that promotes the fusion of endocytic vesicles with the SCV or inhibits recycling of vesicles away from the SCV. Since SseJ activity negatively regulates this SifA activity, we hypothesize that SseJ activity promotes trafficking toward or away from the SCV that opposes the action of SifA.
Although the exact biochemical activity of SseJ remains to be defined, its localization to the membrane is probably important for positioning SseJ in proximity of the substrate. However, catalytic mutants that exhibit impaired enzymatic activity have wild-type patterns of localization, indicating that enzymatic activity does not influence localization. Since SseJ lacks predicted domains that could explain membrane localization, the lack of delocalization of SseJ catalytic mutants suggests that the mechanism promoting SseJ localization is likely a protein-protein interaction between SseJ and a phagosomal protein. Identification of proteins that interact with SseJ will be important for dissecting subcellular localization from enzymatic activity. Though multiple questions remain to be resolved, these studies establish that SseJ enzymatic activity is important for Salmonella virulence in mice and Sif down regulation by modifying the Salmonella phagosome.
ACKNOWLEDGMENTS
This work was supported by grant AI48683 from the National Institutes of Health, and M.B.O. is supported by Oral Biology Training Grant no. DEO7023. The J. H. Brumell laboratory is supported by grant funding and a New Investigator Award from the Canadian Institutes of Health Research to J.H.B. Infrastructure for the J. H. Brumell laboratory was provided by a New Opportunities Fund from the Canadian Foundation for Innovation and the Ontario Innovation Trust. J.H.B. is a recipient of the Premiers Research Excellence Award from the Ontario Ministry of Economic Development and Trade. C.L.B. is a recipient of a University of Toronto open scholarship, a student tuition bursary from the Hospital for Sick Children, and a Natural Sciences and Engineering Research Council of Canada studentship.
We thank James T. Buckley (University of Victoria, Victoria, British Columbia, Canada) for the generous gift of purified GCAT enzyme, Jeremy A. Freeman for use of his strains and plasmid pJAF08, and especially Marie-Pierre Blanc for her tremendous assistance with the competitive index experiments. We thank members of the S. I. Miller and J. H. Brumell laboratory for providing comments and discussion. We also acknowledge the W. M. Keck Center for Advanced Studies in Neural Signaling (University of Washington) for use of their Deltavision microscope and SoftWoRx software.
Present address: Davis Wright Tremaine, Seattle, WA 98101.
REFERENCES
1. Beuzon, C. R., and D. W. Holden. 2001. Use of mixed infections with Salmonella strains to study virulence genes and their interactions in vivo. Microbes Infect. 3:1345-1352.
2. Beuzon, C. R., S. Meresse, K. E. Unsworth, J. Ruiz-Albert, S. Garvis, S. R. Waterman, T. A. Ryder, E. Boucrot, and D. W. Holden. 2000. Salmonella maintains the integrity of its intracellular vacuole through the action of SifA. EMBO J. 19:3235-3249.
3. Beuzon, C. R., S. P. Salcedo, and D. W. Holden. 2002. Growth and killing of a Salmonella enterica serovar Typhimurium sifA mutant strain in the cytosol of different host cell lines. Microbiology 148:2705-2715.
4. Birmingham, C. L., X. Jiang, M. B. Ohlson, S. I. Miller, and J. H. Brumell. 2005. Salmonella-induced filament formation is a dynamic phenotype induced by rapidly replicating Salmonella enterica serovar Typhimurium in epithelial cells. Infect. Immun. 73:1204-1208.
5. Bonelli, F. S., and A. Jonas. 1989. Reaction of lecithin cholesterol acyltransferase with water-soluble substrates. J. Biol. Chem. 264:14723-14728.
6. Brumell, J. H., D. L. Goosney, and B. B. Finlay. 2002. SifA, a type III secreted effector of Salmonella typhimurium, directs Salmonella-induced filament (Sif) formation along microtubules. Traffic 3:407-415.
7. Brumell, J. H., P. Tang, S. D. Mills, and B. B. Finlay. 2001. Characterization of Salmonella-induced filaments (Sifs) reveals a delayed interaction between Salmonella-containing vacuoles and late endocytic compartments. Traffic 2:643-653.
8. Brumell, J. H., P. Tang, M. L. Zaharik, and B. B. Finlay. 2002. Disruption of the Salmonella-containing vacuole leads to increased replication of Salmonella enterica serovar Typhimurium in the cytosol of epithelial cells. Infect. Immun. 70:3264-3270.
9. Brumlik, M. J., and J. T. Buckley. 1996. Identification of the catalytic triad of the lipase/acyltransferase from Aeromonas hydrophila. J. Bacteriol. 178:2060-2064.
10. Buckley, J. T., L. N. Halasa, and S. MacIntyre. 1982. Purification and partial characterization of a bacterial phospholipid: cholesterol acyltransferase. J. Biol. Chem. 257:3320-3325.
11. Catron, D. M., M. D. Sylvester, Y. Lange, M. Kadekoppala, B. D. Jones, D. M. Monack, S. Falkow, and K. Haldar. 2002. The Salmonella-containing vacuole is a major site of intracellular cholesterol accumulation and recruits the GPI-anchored protein CD55. Cell. Microbiol. 4:315-328.
12. Chakravortty, D., I. Hansen-Wester, and M. Hensel. 2002. Salmonella pathogenicity island 2 mediates protection of intracellular Salmonella from reactive nitrogen intermediates. J. Exp. Med. 195:1155-1166.
13. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645.
14. Flieger, A., B. Neumeister, and N. P. Cianciotto. 2002. Characterization of the gene encoding the major secreted lysophospholipase A of Legionella pneumophila and its role in detoxification of lysophosphatidylcholine. Infect. Immun. 70:6094-6106.
15. Freeman, J. A., M. E. Ohl, and S. I. Miller. 2003. The Salmonella enterica serovar Typhimurium translocated effectors SseJ and SifB are targeted to the Salmonella-containing vacuole. Infect. Immun. 71:418-427.
16. Galan, J. E. 2001. Salmonella interactions with host cells: type III secretion at work. Annu. Rev. Cell Dev. Biol. 17:53-86.
17. Garcia-del Portillo, F., M. B. Zwick, K. Y. Leung, and B. B. Finlay. 1993. Salmonella induces the formation of filamentous structures containing lysosomal membrane glycoproteins in epithelial cells. Proc. Natl. Acad. Sci. USA 90:10544-10548.
18. Garner, M. J., R. D. Hayward, and V. Koronakis. 2002. The Salmonella pathogenicity island 1 secretion system directs cellular cholesterol redistribution during mammalian cell entry and intracellular trafficking. Cell. Microbiol. 4:153-165.
19. Haraga, A., and S. I. Miller. 2003. A Salmonella enterica serovar Typhimurium translocated leucine-rich repeat effector protein inhibits NF-B-dependent gene expression. Infect. Immun. 71:4052-4058.
20. Hensel, M. 2000. Salmonella pathogenicity island 2. Mol. Microbiol. 36:1015-1023.
21. Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol. Mol. Biol. Rev. 62:379-433.
22. Jiang, X., O. W. Rossanese, N. F. Brown, S. Kujat-Choy, J. E. Galan, B. B. Finlay, and J. H. Brumell. 2004. The related effector proteins SopD and SopD2 from Salmonella enterica serovar Typhimurium contribute to virulence during systemic infection of mice. Mol. Microbiol. 54:1186-1198.
23. Knodler, L. A., B. A. Vallance, M. Hensel, D. Jackel, B. B. Finlay, and O. Steele-Mortimer. 2003. Salmonella type III effectors PipB and PipB2 are targeted to detergent-resistant microdomains on internal host cell membranes. Mol. Microbiol. 49:685-704.
24. Kuhle, V., and M. Hensel. 2002. SseF and SseG are translocated effectors of the type III secretion system of Salmonella pathogenicity island 2 that modulate aggregation of endosomal compartments. Cell. Microbiol. 4:813-824.
25. Kuhle, V., D. Jackel, and M. Hensel. 2004. Effector proteins encoded by Salmonella pathogenicity island 2 interfere with the microtubule cytoskeleton after translocation into host cells. Traffic 5:356-370.
26. Kujat Choy, S. L., E. C. Boyle, O. Gal-Mor, D. L. Goode, Y. Valdez, B. A. Vallance, and B. B. Finlay. 2004. SseK1 and SseK2 are novel translocated proteins of Salmonella enterica serovar Typhimurium. Infect. Immun. 72:5115-5125.
27. MacIntyre, S., and J. T. Buckley. 1978. Presence of glycerophospholipid:cholesterol acyltransferase and phospholipase in culture supernatant of Aeromonas hydrophila. J. Bacteriol. 135:402-407.
28. Miao, E. A., M. Brittnacher, A. Haraga, R. L. Jeng, M. D. Welch, and S. I. Miller. 2003. Salmonella effectors translocated across the vacuolar membrane interact with the actin cytoskeleton. Mol. Microbiol. 48:401-415.
29. Miao, E. A., and S. I. Miller. 2000. A conserved amino acid sequence directing intracellular type III secretion by Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 97:7539-7544.
30. Miao, E. A., C. A. Scherer, R. M. Tsolis, R. A. Kingsley, L. G. Adams, A. J. Baumler, and S. I. Miller. 1999. Salmonella typhimurium leucine-rich repeat proteins are targeted to the SPI1 and SPI2 type III secretion systems. Mol. Microbiol. 34:850-864.
31. Mitchell, E. K., P. Mastroeni, A. P. Kelly, and J. Trowsdale. 2004. Inhibition of cell surface MHC class II expression by Salmonella. Eur. J. Immunol. 34:2559-2567.
32. Ohl, M. E., and S. I. Miller. 2001. Salmonella: a model for bacterial pathogenesis. Annu. Rev. Med. 52:259-274.
33. Ruiz-Albert, J., X. J. Yu, C. R. Beuzon, A. N. Blakey, E. E. Galyov, and D. W. Holden. 2002. Complementary activities of SseJ and SifA regulate dynamics of the Salmonella typhimurium vacuolar membrane. Mol. Microbiol. 44:645-661.
34. Salcedo, S. P., and D. W. Holden. 2003. SseG, a virulence protein that targets Salmonella to the Golgi network. EMBO J. 22:5003-5014.
35. Scherer, C. A., and Miller, S. I. 2000. Molecular pathogenesis of salmonellae, p. 266-316. In E. A. Groisman (ed.), Principles of bacterial pathogenesis. Academic Press, San Diego, Calif.
36. Shea, J. E., M. Hensel, C. Gleeson, and D. W. Holden. 1996. Identification of a virulence locus encoding a second type III secretion system in Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 93:2593-2597.
37. Stein, M. A., K. Y. Leung, M. Zwick, F. Garcia-del Portillo, and B. B. Finlay. 1996. Identification of a Salmonella virulence gene required for formation of filamentous structures containing lysosomal membrane glycoproteins within epithelial cells. Mol. Microbiol. 20:151-164.
38. Upton, C., and J. T. Buckley. 1995. A new family of lipolytic enzymes Trends Biochem. Sci. 20:178-179.
39. Vazquez-Torres, A., Y. Xu, J. Jones-Carson, D. W. Holden, S. M. Lucia, M. C. Dinauer, P. Mastroeni, and F. C. Fang. 2000. Salmonella pathogenicity island 2-dependent evasion of the phagocyte NADPH oxidase. Science 287:1655-1658.
40. Wang, R. F., and S. R. Kushner. 1991. Construction of versatile low-copy-number vectors for cloning, sequencing and gene expression in Escherichia coli. Gene 100:195-199.(Maikke B. Ohlson, Kerry F)
Infection, Immunity, Injury, and Repair Program, Hospital for Sick ChildrenDepartment of Molecular and Medical Genetics
University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Salmonella enterica serovar Typhimurium utilizes a type III secretion system (TTSS) encoded on Salmonella pathogenicity island-2 (SPI2) to promote intracellular replication during infection, but little is known about the molecular function of SPI2-translocated effectors and how they contribute to this process. SseJ is a SPI2 TTSS effector protein that is homologous to enzymes called glycerophospholipid-cholesterol acyltransferases and, following translocation, localizes to the Salmonella-containing vacuole and Salmonella-induced filaments. Full virulence requires SseJ, as sseJ null mutants exhibit decreased replication in cultured cells and host tissues. This work demonstrates that SseJ is an enzyme with deacylase activity in vitro and identifies three active-site residues. Catalytic SseJ mutants display wild-type translocation and subcellular localization but fail to complement the virulence defect of an sseJ null mutant. In contrast to the wild type, SseJ catalytic mutants fail to down regulate Salmonella-induced filament formation and fail to restore the sifA null mutant phenotype of loss of phagosomal membrane to sifA sseJ null double mutants, suggesting that wild-type SseJ modifies the vacuolar membrane. This is the first demonstration of an enzymatic activity for a SPI2 effector protein and provides support for the hypothesis that the deacylation of lipids on the Salmonella-containing vacuole membrane is important to bacterial pathogenesis.
INTRODUCTION
Salmonellae are gram-negative bacteria that are capable of infecting a wide range of animals, such as cattle, poultry, snakes, and humans, leading to various diseases (35). Salmonella enterica serovar Typhimurium causes gastroenteritis in humans, while in inbred mice with innate immune defects, it causes severe systemic disease. During systemic infection of mice, serovar Typhimurium survives and replicates intracellularly in a modified phagosomal compartment enriched in cholesterol (11) and efficiently avoids destruction by phagocytes (32).
S. enterica serovar Typhimurium utilizes two type III secretion systems (TTSS) to directly translocate virulence determinants, called effector proteins, from the bacterial cytoplasm into the host cell cytoplasm (21). The TTSS encoded on Salmonella pathogenicity island-1 (SPI1) is expressed upon contact with host cells and translocates SPI1-specific effectors that alter cell signaling, induce membrane ruffling, and disrupt actin polymerization in order to promote bacterial uptake and modulate host inflammatory responses (16). Following internalization, the SPI2 TTSS is induced and translocates a different set of SPI2-specific effector proteins across the phagosomal membrane (20, 36). These effectors inhibit host cell processes, such as NADPH oxidase and iNOS localization (12, 39), which allow serovar Typhimurium to replicate. Moreover, a SPI2 effector was shown to decrease surface presentation of major histocompatibility complex class II molecules in serovar Typhimurium-infected cells (31).
Following translocation, SPI2 effectors localize to different cellular compartments, including the nucleus, cytoplasm, cytoskeleton, Golgi apparatus, and phagosome (15, 19, 22, 23, 25, 26, 28, 34). Such differential localization patterns indicate that effectors manipulate numerous aspects of host cell function. SifA is a well-characterized SPI2 effector that is important for virulence in mice, as sifA null mutants are severely attenuated (37). Following translocation, SifA localizes to the Salmonella-containing vacuole (SCV) (6) and is required for the formation of Salmonella-induced filaments (Sifs) in epithelium-derived cells such as HeLa cells (37). Sifs are highly dynamic structures that appear to be tubular extensions of the SCV and contain lysosomal glycoproteins, such as lysosome-associated membrane protein-1 (LAMP-1) (4, 7). The formation of Sifs requires microtubules, and the timing of Sif appearance in epithelial cells is concomitant with bacterial replication (6, 7, 17). Although Sifs have been analyzed extensively, and their role during infection remains unknown, they likely represent the ability of translocated SPI2 effectors to manipulate vesicular trafficking to and from the SCV and provide an important phenotype that represents Salmonella-induced host cell alterations.
Several other SPI2 effectors, including SifB, SseJ, SopD2, PipB, PipB2, SseF, and SseG, also localize to the SCV and Sif structures in infected cells, suggesting that they function collectively, and possibly redundantly, to exert their effect on the SCV (15, 22, 23, 25, 33). In addition to SifA, the effectors SseF, SseG, and SopD2 have been shown to contribute to the formation of Sif, which highlights the complexity of this phenotype (22, 24). Results from recent studies of SCV and Sif-localizing effectors have supported the hypothesis that they alter the SCV via manipulation of host vesicular traffic; however, there is no biochemical evidence to indicate the molecular mechanism by which this is accomplished.
Among the known SPI2 effector proteins, SseJ is the only effector with a predicted biochemical activity. The amino-terminal 140 amino acids of SseJ are required for its translocation by S. enterica serovar Typhimurium (29), while its carboxy-terminus (residues 140 to 408) has approximately 29% homology to several members of the GDSL family of lipases (38). GDSL lipase activity is dependent on three amino acid residues (Ser, Asp, and His), which compose a catalytic triad and are present in SseJ (see alignment in reference 14). Among GDSL lipases, SseJ has the most conservation of catalytic regions with glycerophospholipid-cholesterol acyltransferase (GCAT) enzymes, which are proteins secreted by Aeromonas and Vibrio species that catalyze the transfer of fatty acid acyl chains from phospholipids to cholesterol at lipid-water interfaces (9, 10, 27).
Studies have shown that SseJ is required for the full virulence of serovar Typhimurium, as sseJ mutants are mildly defective for intracellular replication in cultured cells and have a virulence defect in the mouse infection model (15, 33). Interestingly, SseJ is one of only a few SPI2 effector proteins, including SifA, that lead to a virulence phenotype when deleted, which underscores the importance of this protein to Salmonella pathogenesis. Recent analysis of the dynamics of Sif formation has demonstrated that SseJ down regulates the formation of Sif (4), supporting the hypothesis that SseJ functions to manipulate the SCV and host vesicular trafficking. Additional important phenotypes that implicate SseJ in modifying host processes include destabilization of the SCV membrane of sifA deletion mutants and formation of globular membraneous compartments in cultured cells by ectopic expression of SseJ (33). Given the conserved amino acid similarities between SseJ and other GCAT enzymes and the putative role of SseJ in altering host trafficking processes, we hypothesized that SseJ has deacylase and acyltransferase enzymatic activities and that these activities are important for SseJ function in vivo. This work investigates the potential enzymatic activity of SseJ and tests its role in SseJ-associated phenotypes and the promotion of bacterial virulence in mice.
MATERIALS AND METHODS
Bacterial strains, plasmids, eukaryotic cell lines, and growth conditions. All bacterial strains and plasmids used are listed in Table 1. Salmonella enterica serovar Typhimurium and Escherichia coli strains were grown and maintained in Luria-Bertani (LB) broth or on plates at 37°C with antibiotics added at the following concentrations: ampicillin, 100 or 50 mg/ml; kanamycin, 45 mg/ml; chloramphenicol (Cam), 20 mg/ml; and tetracycline, 10 mg/ml, unless described otherwise. HeLa and RAW264.7 cell lines originated from the ATCC and were maintained as previously described (15).
Plasmid construction. The oligonucleotides used in this study are listed in Table 2. The His-SseJ plasmid for protein purification, pJAF08, was made by amplifying the chromosomal region carrying sseJ from serovar Typhimurium with primers 5' JAF11 and 3' JAF12 by PCR and was cloned into pET-15b (Novagen) using NdeI and BamHI enzymes to create an in-frame amino-terminal fusion between the six-His tag and SseJ. The wild-type SseJ-HA-expressing plasmid used for immunofluorescence and competitive index (CI) complementation studies is pJAF111, which was published previously (15). Plasmids pKF27, pKF28, pKF29, pKF33, pKF35, pKF36, pKF37, and pMBO76 were all produced by site-directed mutagenesis using QuikChange (Stratagene). Primers 5' KF43 and 3' KF44 produced an S151A mutation in plasmids pKF35 and pKF29, with pJAF111 and pJAF08 used as templates, respectively. Similarly, 5' KF49 and 3' KF50 produced a D247N mutation in pKF36 and pKF33, with pJAF111 and pJAF08 used as templates, respectively; 5' KF51 and 3' KF52 produced a H384N mutation in pKF37 and pKF28, with pJAF111 and pJAF08 used as templates, respectively; and 5' KF47 and 3' KF48 produced an S153N mutation in pKF27, with pJAF08 used as a template. The triple mutant SseJ-HA plasmid pMBO76 was produced sequentially using QuikChange. First, the H384N mutation was incorporated onto the D247N-containing SseJ-HA plasmid pKF36 using 5' KF51 and 3' KF52, which created the intermediate double mutant (D247N and H384N) plasmid pMBO74. The double mutant plasmid was then used as a template to incorporate the S151A mutation using primers 5' KF43 and 3' KF44. All plasmids were sequenced to verify the mutations.
sseJ deletion strain construction. The chromosomal deletion of sseJ in serovar Typhimurium was constructed using the Red recombinase system (13). The Camr gene and FLP recognition target (FRT) sites from pKD3 were amplified by primers 5' MBO71 and 3' MBO72, which also contain flanking regions with homology to the 5' and 3' regions of sseJ, respectively. The resulting PCR fragment was purified and transformed into electrocompetent wild-type serovar Typhimurium cells containing pKD46 that were induced with arabinose at 30°C, and recombinants were selected on LB broth-Cam plates. Candidate colonies were verified by PCR. This approach yielded a serovar Typhimurium strain with the sequence encoding amino acids 3 to 406 of sseJ replaced by the FRT-Cam-FRT cassette.
Competitive index assay. Mice were ordered from Charles River Laboratories, Inc., and all experiments were performed with IACUC approval. Virulence phenotypes were tested by competitive index assay as described previously (15). Briefly, 6- to 8-week-old female BALB/c mice were inoculated intraperitoneally with a mixture of 5 x 104 organisms each of two serovar Typhimurium strains for a total of 105 bacteria in a 0.2-ml volume. Each strain was diluted from stationary-phase cultures grown overnight and contained either a chromosomal or stable plasmid-based antibiotic marker to allow the strains to be differentiated. The bacterial inoculum contained approximately equal concentrations of both strain 1 and strain 2 bacteria, and the ratio of strain 1 to strain 2 was confirmed by plating dilutions of the inoculum onto selective media. Forty-eight hours after infection, mice were euthanized by CO2 asphyxiation, the spleens were dissected, and each spleen was homogenized in sterile phosphate-buffered saline using a 2-ml glass Dounce homogenizer. Ratios of strain 1 to strain 2 bacteria in each spleen were calculated from bacterial counts produced by plating aliquots of 1:10 dilutions of homogenized spleen on selective media. The CI was calculated by dividing the ratio of strain 1 to strain 2 bacteria isolated from the spleen by the ratio of strain 1 to strain 2 bacteria inoculated into the mouse. CI results were determined by calculating the means ± standard deviations for at least 10 mice that were infected in three separate experiments. Statistical significance (P value) was determined by a Student two-tailed t test.
Immunofluorescence microscopy. Mammalian cells were seeded onto 12-mm coverslips in wells of 24-well tissue culture dishes in antibiotic-free media at a concentration of either 5 x 104 or 1 x 105 cells for HeLa or RAW264.7 cells, respectively. RAW264.7 cells were infected with stationary-phase serovar Typhimurium, while HeLa cells were infected with bacteria that were back-diluted to 1:100 for 3 h for log-phase growth. Both cell types were infected with bacteria at a multiplicity of infection of 100:1 in antibiotic-free media for 1 h, after which cells were washed with phosphate-buffered saline, and medium containing 10 μg gentamicin/ml was applied to the infected cells for the remainder of the infection. Following infection, cells were fixed with 2% paraformaldehyde in phosphate-buffered saline for at least 1 hour to overnight, washed with phosphate-buffered saline twice, and permeabilized with ice-cold acetone for 20 seconds. Prior to being stained, cells were blocked for 1 hour with a blocking solution composed of phosphate-buffered saline containing 5% donkey serum, 1% bovine serum albumin, 0.2% saponin, and 5 mM EDTA at room temperature. All primary and secondary antibodies were diluted in blocking solution and applied to cells for 1 hour, followed by washes with phosphate-buffered saline. To detect the hemagglutinin (HA) epitope, mouse anti-HA antibodies (Covance) were used at 1:500 and were probed with tetramethyl rhodamine isothiocyanate-conjugated donkey-anti-mouse antibodies (Sigma) at 1:200 as secondary antibodies. In RAW264.7 cells, mouse LAMP-1 was stained with rat anti-LAMP-1-fluorescein isothiocyanate-conjugated antibodies (BD Pharmingen) at a concentration of 5 μg/ml. Human LAMP-1 was stained using a 1:10 dilution of mouse anti-LAMP-1-fluorescein isothiocyanate conjugate (BD Pharmingen). Rabbit antilipopolysaccharide (anti-LPS) antibodies (Difco) were used at a ratio of 1:1,000. Secondary staining was performed using either Cy5- or tetramethyl rhodamine isothiocyanate-conjugated donkey-anti-rabbit secondary antibodies (Jackson) at 1:200. Stained coverslips were mounted onto slides using Vectashield (Vector Laboratories, Inc.) and sealed with nail polish. Slides were examined on an inverted Olympus microscope with a 60x oil immersion lens, and images were collected and deconvolved using SoftWoRx (Applied Precision) and then imported into Photoshop 7.0 (Adobe) to separate color layers.
Sif and LAMP-1 colocalization quantitation. HeLa cells were seeded at a density of 2.5 x 104 or 5 x 104 cells for Sif quantitation or LAMP-1 colocalization experiments, respectively. Cells were infected with serovar Typhimurium at a multiplicity of infection of 325:1 for 1 h, followed by gentamicin protection as described previously (4). At specific time points, the cells were fixed with 2.5% paraformaldehyde, permeabilized, and stained with anti-LPS and anti-human LAMP-1. Percent cells expressing Sifs was calculated by counting 100 infected cells per time point for each strain. Percent bacteria colocalizing with LAMP-1 was calculated by counting the number of LAMP-1-positive Salmonella isolates out of the total number of Salmonella isolates, with at least 100 Salmonella isolates counted per time point for each strain.
Protein purification. Strains containing His tag expression constructs were induced with 1 mM IPTG (isopropyl--D-thiogalactopyranoside) for 4 h prior to collection of cell pellets by centrifugation. The cell pellets were resuspended in cold 50 mM Tris (pH 7.4), and the cells were lysed using a French press. The soluble protein fraction was separated from the cellular debris in the lysate by ultracentrifugation for 1 h at 40,000 rpm. Supernatants were cleaned by passage through a 0.45-μm filter and loaded onto a NiSO4-charged HiTrap column (Amersham) to bind His-tagged protein. Nonspecific proteins were removed by washing the column with wash buffer (20 mM NaPO4, 100 mM imidazole, pH 7.4) prior to elution with elution buffer (20 mM NaPO4, 400 mM imidazole, pH 7.4). Fractions containing His-tagged protein were collected and concentrated with a Centricon-30 concentrating filter (Millipore), and imidazole was removed by dialyzing the protein in a Slide-A-Lyzer (Pierce) against 50 mM Tris (pH 7.2) overnight at 4°C.
Deacylase assay. The PNPB (para-nitrophenyl butyrate) deacylase assay is a modified version of that described previously (5). Briefly, purified enzyme of varied concentrations was added to 1-ml spectrophotometer cuvettes containing the assay reaction mix (65 μl 50 mM PNPB [Sigma] dissolved in 20 mM Tris [pH 7.4] and 3% acetonitrile, 100 μl 200 mM Tris [pH 7.4], and H2O to make 1 ml) and monitored in a Beckman DU 650 spectrophotometer at 400 nm for 30 min, with measurements taken every 30 seconds while the cuvette holder was equilibrated to 37°C. The spectrophotometer values obtained for the reaction mix without enzyme were subtracted from the values for each reaction mixture that contained enzyme at each time point to control for spontaneous PNPB deacylation. The rate (absorbance/time) was calculated using Kaleidagraph (Synergy Software) and was used to determine the micromoles of para-nitrophenol (PNP) formed per second using Beer's law (5).
RESULTS
SseJ has in vitro deacylase activity. The carboxy-terminal region of the SPI2 effector SseJ displays sequence homology to a family of enzymes termed GCATs, which typically exhibit deacylase activity as a first step in the reaction mechanism. To test whether SseJ posseses deacylase activity, we chose to use a soluble colorless substrate, PNPB, that forms a yellow product, PNP, upon removal of a short-chain fatty acid acyl group in a colorimetric assay (5). Amino-terminal His-tagged SseJ was purified from E. coli and tested for deacylase activity using this assay. We found that 5 μg of His-SseJ formed PNP at a rate of 9.16 x 10–4 μmol/second (Fig. 1A), demonstrating that SseJ functions as a deacylase. As a control, native purified Aeromonas salmonicida GCAT protein was analyzed, and 0.5 μg of protein exhibited the same rate of PNP production as 5 μg His-SseJ (data not shown), indicating that the activity of SseJ in this assay is approximately 10-fold lower.
SseJ mutants lack deacylase activity. Based on sequence homology, we hypothesized that the conserved Ser151, Asp247, and His384 residues of SseJ compose a catalytic triad (Fig. 1B). To test whether mutations of the catalytic residues had impaired activity in the PNPB assay, site-directed mutants were designed and purified from E. coli. We made three mutant versions of His-SseJ that contain single amino acid substitutions: S151A, D247N, and H384N. The predicted catalytic Ser151 residue is the middle serine residue in a G-D-S-L-S motif, which is present in all known GCAT enzymes. Since it was unknown whether Ser151 or Ser153 was the catalytic residue, a Ser153A mutant was also constructed. When tested for deacylase activity using the PNPB assay, the His-SseJ-S153A mutant protein produced 9.33 x 10–4 μmol of PNP per second, similar to the rate produced by wild-type SseJ (Fig. 1A). However, mutant SseJ proteins His-SseJ-S151A, His-SseJ-D247N, and His-SseJ-H384N exhibited rates of 2.16 x 10–4, 1.83 x 10–4, and 1.18 x 10–4 μmol PNP per second, respectively, an approximately fivefold reduction in activity compared to that of wild-type His-SseJ (Fig. 1A). These results suggest that Ser151, Asp247, and His384 residues are members of a catalytic triad and that Ser153 is not a catalytic residue. The reduction in catalytic activity was not due to unspecific destabilization of SseJ, as all mutants could be expressed to similar levels and exhibited identical peak fluorescence spectra. Peak fluorescence of wild-type His-SseJ in 50 mM Tris buffer when excited at 280 nm was found to be 335 nm. When assayed in the presence of 7 M urea to induce unfolding, the peak fluorescence shifted to 350 nm, indicating that a conformational change had occurred. The 335-nm peak fluorescence of each of the catalytic mutant proteins in 50 mM Tris buffer was identical to that of wild-type undenatured His-SseJ and indicated that the proteins were equally folded.
Translocated SseJ catalytic mutants maintain SCV localization. Although it was previously shown that ectopic expression of an SseJ-S151V mutant did not alter subcellular distribution from that of the wild type, suggesting that active-site residues do not influence localization (33), it was not known whether active-site mutations would affect the bacterial translocation of SseJ or SCV localization in the context of infection. To test this, HeLa cells were infected with a sseJ mutant expressing wild-type SseJ-HA; an SseJ-S151A-HA, SseJ-D247N-HA, or SseJ-H384N-HA catalytic mutant; or an SseJ-S151A-SsJ-D247N-SsJ-H384N-HA triple mutant. Expression of SseJ-HA from each of these stable, low-copy-number plasmids is under the control of its native promoter. Western blots of infected cell lysates probed with anti-HA antibodies indicated that the amount of translocated SseJ catalytic mutant protein was similar to wild-type SseJ-HA (data not shown), indicating that the catalytic mutations did not affect translocation by the TTSS apparatus. In addition, infected HeLa cells were fixed 10 hours postinfection; immunostained for the HA epitope, human LAMP-1, and LPS; and analyzed by fluorescence microscopy. Wild-type SseJ-HA colocalized with LAMP-1 and LPS, which indicates its localization to the SCV, and colocalized with the LAMP-1-positive Sif structures that radiate away from the SCV (Fig. 2A), confirming previous reports (15, 33). Similarly, we observed that SseJ-S151A-HA, SseJ-D247N-HA, SseJ-H384N-HA, and the SseJ-S151A-D247N-H384N-HA triple mutant localized to the SCV and Sifs like wild-type SseJ-HA (Fig. 2B to D and data not shown, respectively). Sif production was observed in infections with all four of the SseJ-HA mutants, and there were no gross morphological differences between the wild type and the catalytic mutants.
In cultured macrophages, SseJ- and LAMP-1-positive structures have been observed distant from the SCV (15). To ascertain whether SseJ catalytic mutants exhibited the same pattern of macrophage localization and trafficking, RAW264.7 cells were infected with a sseJ mutant expressing either wild-type SseJ-HA, SseJ-S151A-HA, SseJ-D247N-HA, or SseJ-H384N-HA and were immunostained with antibodies to the HA epitope, LPS, and mouse LAMP-1 10 hours postinfection. We observed that SseJ-HA mutants maintained the ability to localize to the SCV and colocalized with LAMP-1-positive compartments distinct from the SCV (Fig. 3). The localization of mutant SseJ in both HeLa and RAW264.7 cells indicates that catalytic residues of SseJ are not required for translocation or localization and demonstrates that the ability to localize to vacuolar membranes is not a consequence of enzymatic activity.
SseJ catalytic mutants fail to complement in vivo sseJ phenotypes. S. enterica serovar Typhimurium strains that lack sifA display two major phenotypes: failure to induce Sif formation and failure to remain within the Salmonella-containing vacuole as determined by LAMP-1 colocalization and electron microscopy (2). Additionally, sifA null mutants replicate in the cytosol, as opposed to the SCV, following loss of the vacuolar membrane (2, 3, 8). SseJ activity is implicated in sifA vacuolar membrane loss because sifA sseJ double mutants remain enclosed in the SCV (33). This phenotype can be exploited to measure SseJ activity in vivo. Specifically, to test whether SseJ catalytic activity plays a role in membrane loss by the sifA mutant, plasmids expressing either wild-type or catalytic mutant SseJ were assayed for the ability to restore the sifA phenotype to sifA sseJ double mutant bacteria by LAMP-1 colocalization. We observed that wild-type and sseJ bacteria maintained a high percentage of colocalization with LAMP-1, while sifA bacteria exhibited a marked decrease in LAMP-1 colocalization, and sifA sseJ LAMP-1 colocalization was significantly higher than with sifA bacteria (Fig. 4), confirming previous results (33). The LAMP-1 colocalization phenotype in cells infected with the sifA sseJ mutant expressing wild-type SseJ was complemented, as these cells exhibited a decrease in LAMP-1 colocalization similar to that of sifA bacteria. However, the LAMP-1 colocalization phenotype of the sifA sseJ mutant expressing the catalytic SseJ-HA triple mutant was not complemented, as these cells had a higher percentage of bacteria that continued to colocalize with LAMP-1, approximately at the same levels as sifA sseJ bacteria (Fig. 4). These results confirm that SseJ enzymatic activity allows sifA bacteria to escape from the SCV and provides strong evidence that the function of SseJ is to modify the SCV membrane.
Recently, we demonstrated that SseJ negatively regulates the formation of Sifs in epithelial cells (4); thus, we hypothesized that the catalytic activity of SseJ was involved in the down regulation of Sifs in infected cells. To test this, Sif formation was analyzed in HeLa cells that were infected with the wild type, the sseJ mutant, or the sseJ mutant expressing either wild-type SseJ-HA or the triple mutant SseJ-S151A-D247N-H384N-HA. Sif formation was quantified at 2-hour intervals between 4 and 10 hours postinfection. We determined that sseJ mutant-infected cells induced 30% more Sif-expressing cells than wild-type-infected cells, confirming previous results (4). HeLa cells infected with the sseJ mutant expressing wild-type SseJ-HA induced Sif expression at levels similar to those of wild-type serovar Typhimurium, indicating that the phenotype was complemented. In contrast, the sseJ mutant expressing triple mutant SseJ-HA failed to down regulate Sif expression, indicating that the ability of SseJ to negatively regulate Sif formation requires enzymatic activity (Fig. 5).
SseJ catalytic activity is required for full virulence in mice. To determine whether the catalytic activity of SseJ is required for serovar Typhimurium virulence in mice, we tested the virulence phenotype of sseJ strains expressing catalytic mutant SseJ-HA by the CI assay (1). First, the sseJ mutant was competed against the wild type in order to confirm CI results from previously published competitions with the sseJ mutant versus the wild type. As shown in Fig. 6, the sseJ mutant exhibits a competitive index defect of 0.56 ± 0.10, similar to CI values of 0.5 that we and others observed previously (15, 33). As a control, sseJ strains containing Kanr and Ampr plasmids were competed, and we verified that plasmid-based antibiotic resistance genes do not influence competitive index ratios, as these strains competed equally (1.0 ± 0.13). The sseJ mutant expressing wild-type SseJ-HA was found to compete equally against the wild type, with a competitive index of 0.90 ± 0.18, suggesting that low-copy-number-SseJ expression by its native promoter can restore virulence to the sseJ mutant. However, when the wild type was competed against the sseJ mutant expressing SseJ-S151A-HA, SseJ-D247N-HA, SseJ-H384N-HA, and SseJ-S151A-D247N-H384N-HA, CI phenotypes of 0.50 ± 0.09, 0.17 ± 0.05, 0.45 ± 0.13, and 0.44 ± 0.06 were observed (Fig. 6). These competitive indices show that SseJ catalytic mutants cannot complement the virulence defect of sseJ and clearly demonstrate that SseJ catalytic activity is required for full virulence in mice.
DISCUSSION
The ability of Salmonella to survive and replicate within mammalian cells is dependent on the SPI2 TTSS; however, the exact molecular contribution of SPI2 TTSS effector proteins remains unknown. In this study, we sought to analyze the potential enzymatic activity of the SPI2 effector SseJ and to determine whether the residues of this protein, which contain homology to GDSL lipases, contribute to the function of SseJ in vivo. Wild-type His-SseJ catalyzed the formation of PNP in the PNPB assay and clearly showed that SseJ has deacylase activity. We found that amino acid substitutions S151A, D247N, and H384N in SseJ led to a fivefold reduction in deacylase activity, suggesting that these residues form a catalytic triad. Strains expressing catalytic mutant SseJ-HA translocated mutant protein into host cells, which displayed wild-type localization to the Salmonella-containing phagosome. However, despite proper localization, SseJ catalytic mutants failed to complement the known cellular phenotypes attributed to SseJ, namely, down regulation of Sifs and destabilization of the SCV in sifA mutants. Most importantly, catalytically inactive SseJ mutants demonstrated a virulence defect in mice.
The deacylase activity exhibited by SseJ is approximately 10-fold lower than the activity of a purified A. salmonicida GCAT protein that is secreted extracellularly by type II secretion. The PNPB substrate is not likely to be physiologic for either protein but was chosen because it is water soluble. Differences in affinities for PNPB could explain the rate difference, and SseJ could have higher affinity for different substrates. However, it is interesting to speculate that the reduced enzymatic activity of SseJ compared to that of other GCAT proteins is physiologic. High levels of SseJ activity could have a deleterious effect on Salmonella survival within macrophages by perhaps leading to destabilization of the SCV membrane and the subsequent release of bacteria into the cytosol. Further studies will be required to identify the biologically relevant substrates of SseJ and whether modest enzyme activity is relevant to its in vivo function.
From our current study, we cannot conclude whether SseJ functions solely as a deacylase or whether it also has acyltransferase activity. The primary sequence indicates that SseJ is more similar to GCAT enzymes than to other GDSL lipases (see alignment in reference 14 and Fig. 1B). Altering the catalytic aspartic acid (D247) in the G-A-N-D-Y motif of SseJ to an asparagine (D247N) converts the motif to a G-A-N-N-Y motif like that found in PlaA, the Legionella lysophospholipase. In SseJ, the D247N mutation, and therefore the altered motif, results in inactive SseJ protein, which suggests that the active site of SseJ requires a conformation more like that of GCAT enzymes than of lysophospholipases or other GDSL lipases.
If SseJ possesses deacylase and acyltransferase activities, the production of esterified cholesterol may contribute to the effect of SseJ on the SCV. Cholesterol is enriched in the phagosomal compartment in serovar Typhimurium-infected cells (7, 11, 18) and is likely important for providing membrane rigidity and forming lipid rafts. Hence, esterified cholesterol in lipid rafts could disrupt cell-signaling platforms. Conversely, lipid and/or cholesterol modifications could inhibit the formation or interaction of molecular complexes on the SCV surface that are required for vesicular trafficking to or from the SCV.
This work supports the hypothesis that SseJ activity induces lipid modifications on the SCV membranes that alter the phagosomal compartments. SseJ likely acts in concert or sequentially with SifA and other effector proteins to promote vacuolar remodeling. It seems that Sif formation reflects an activity of SifA that promotes the fusion of endocytic vesicles with the SCV or inhibits recycling of vesicles away from the SCV. Since SseJ activity negatively regulates this SifA activity, we hypothesize that SseJ activity promotes trafficking toward or away from the SCV that opposes the action of SifA.
Although the exact biochemical activity of SseJ remains to be defined, its localization to the membrane is probably important for positioning SseJ in proximity of the substrate. However, catalytic mutants that exhibit impaired enzymatic activity have wild-type patterns of localization, indicating that enzymatic activity does not influence localization. Since SseJ lacks predicted domains that could explain membrane localization, the lack of delocalization of SseJ catalytic mutants suggests that the mechanism promoting SseJ localization is likely a protein-protein interaction between SseJ and a phagosomal protein. Identification of proteins that interact with SseJ will be important for dissecting subcellular localization from enzymatic activity. Though multiple questions remain to be resolved, these studies establish that SseJ enzymatic activity is important for Salmonella virulence in mice and Sif down regulation by modifying the Salmonella phagosome.
ACKNOWLEDGMENTS
This work was supported by grant AI48683 from the National Institutes of Health, and M.B.O. is supported by Oral Biology Training Grant no. DEO7023. The J. H. Brumell laboratory is supported by grant funding and a New Investigator Award from the Canadian Institutes of Health Research to J.H.B. Infrastructure for the J. H. Brumell laboratory was provided by a New Opportunities Fund from the Canadian Foundation for Innovation and the Ontario Innovation Trust. J.H.B. is a recipient of the Premiers Research Excellence Award from the Ontario Ministry of Economic Development and Trade. C.L.B. is a recipient of a University of Toronto open scholarship, a student tuition bursary from the Hospital for Sick Children, and a Natural Sciences and Engineering Research Council of Canada studentship.
We thank James T. Buckley (University of Victoria, Victoria, British Columbia, Canada) for the generous gift of purified GCAT enzyme, Jeremy A. Freeman for use of his strains and plasmid pJAF08, and especially Marie-Pierre Blanc for her tremendous assistance with the competitive index experiments. We thank members of the S. I. Miller and J. H. Brumell laboratory for providing comments and discussion. We also acknowledge the W. M. Keck Center for Advanced Studies in Neural Signaling (University of Washington) for use of their Deltavision microscope and SoftWoRx software.
Present address: Davis Wright Tremaine, Seattle, WA 98101.
REFERENCES
1. Beuzon, C. R., and D. W. Holden. 2001. Use of mixed infections with Salmonella strains to study virulence genes and their interactions in vivo. Microbes Infect. 3:1345-1352.
2. Beuzon, C. R., S. Meresse, K. E. Unsworth, J. Ruiz-Albert, S. Garvis, S. R. Waterman, T. A. Ryder, E. Boucrot, and D. W. Holden. 2000. Salmonella maintains the integrity of its intracellular vacuole through the action of SifA. EMBO J. 19:3235-3249.
3. Beuzon, C. R., S. P. Salcedo, and D. W. Holden. 2002. Growth and killing of a Salmonella enterica serovar Typhimurium sifA mutant strain in the cytosol of different host cell lines. Microbiology 148:2705-2715.
4. Birmingham, C. L., X. Jiang, M. B. Ohlson, S. I. Miller, and J. H. Brumell. 2005. Salmonella-induced filament formation is a dynamic phenotype induced by rapidly replicating Salmonella enterica serovar Typhimurium in epithelial cells. Infect. Immun. 73:1204-1208.
5. Bonelli, F. S., and A. Jonas. 1989. Reaction of lecithin cholesterol acyltransferase with water-soluble substrates. J. Biol. Chem. 264:14723-14728.
6. Brumell, J. H., D. L. Goosney, and B. B. Finlay. 2002. SifA, a type III secreted effector of Salmonella typhimurium, directs Salmonella-induced filament (Sif) formation along microtubules. Traffic 3:407-415.
7. Brumell, J. H., P. Tang, S. D. Mills, and B. B. Finlay. 2001. Characterization of Salmonella-induced filaments (Sifs) reveals a delayed interaction between Salmonella-containing vacuoles and late endocytic compartments. Traffic 2:643-653.
8. Brumell, J. H., P. Tang, M. L. Zaharik, and B. B. Finlay. 2002. Disruption of the Salmonella-containing vacuole leads to increased replication of Salmonella enterica serovar Typhimurium in the cytosol of epithelial cells. Infect. Immun. 70:3264-3270.
9. Brumlik, M. J., and J. T. Buckley. 1996. Identification of the catalytic triad of the lipase/acyltransferase from Aeromonas hydrophila. J. Bacteriol. 178:2060-2064.
10. Buckley, J. T., L. N. Halasa, and S. MacIntyre. 1982. Purification and partial characterization of a bacterial phospholipid: cholesterol acyltransferase. J. Biol. Chem. 257:3320-3325.
11. Catron, D. M., M. D. Sylvester, Y. Lange, M. Kadekoppala, B. D. Jones, D. M. Monack, S. Falkow, and K. Haldar. 2002. The Salmonella-containing vacuole is a major site of intracellular cholesterol accumulation and recruits the GPI-anchored protein CD55. Cell. Microbiol. 4:315-328.
12. Chakravortty, D., I. Hansen-Wester, and M. Hensel. 2002. Salmonella pathogenicity island 2 mediates protection of intracellular Salmonella from reactive nitrogen intermediates. J. Exp. Med. 195:1155-1166.
13. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645.
14. Flieger, A., B. Neumeister, and N. P. Cianciotto. 2002. Characterization of the gene encoding the major secreted lysophospholipase A of Legionella pneumophila and its role in detoxification of lysophosphatidylcholine. Infect. Immun. 70:6094-6106.
15. Freeman, J. A., M. E. Ohl, and S. I. Miller. 2003. The Salmonella enterica serovar Typhimurium translocated effectors SseJ and SifB are targeted to the Salmonella-containing vacuole. Infect. Immun. 71:418-427.
16. Galan, J. E. 2001. Salmonella interactions with host cells: type III secretion at work. Annu. Rev. Cell Dev. Biol. 17:53-86.
17. Garcia-del Portillo, F., M. B. Zwick, K. Y. Leung, and B. B. Finlay. 1993. Salmonella induces the formation of filamentous structures containing lysosomal membrane glycoproteins in epithelial cells. Proc. Natl. Acad. Sci. USA 90:10544-10548.
18. Garner, M. J., R. D. Hayward, and V. Koronakis. 2002. The Salmonella pathogenicity island 1 secretion system directs cellular cholesterol redistribution during mammalian cell entry and intracellular trafficking. Cell. Microbiol. 4:153-165.
19. Haraga, A., and S. I. Miller. 2003. A Salmonella enterica serovar Typhimurium translocated leucine-rich repeat effector protein inhibits NF-B-dependent gene expression. Infect. Immun. 71:4052-4058.
20. Hensel, M. 2000. Salmonella pathogenicity island 2. Mol. Microbiol. 36:1015-1023.
21. Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol. Mol. Biol. Rev. 62:379-433.
22. Jiang, X., O. W. Rossanese, N. F. Brown, S. Kujat-Choy, J. E. Galan, B. B. Finlay, and J. H. Brumell. 2004. The related effector proteins SopD and SopD2 from Salmonella enterica serovar Typhimurium contribute to virulence during systemic infection of mice. Mol. Microbiol. 54:1186-1198.
23. Knodler, L. A., B. A. Vallance, M. Hensel, D. Jackel, B. B. Finlay, and O. Steele-Mortimer. 2003. Salmonella type III effectors PipB and PipB2 are targeted to detergent-resistant microdomains on internal host cell membranes. Mol. Microbiol. 49:685-704.
24. Kuhle, V., and M. Hensel. 2002. SseF and SseG are translocated effectors of the type III secretion system of Salmonella pathogenicity island 2 that modulate aggregation of endosomal compartments. Cell. Microbiol. 4:813-824.
25. Kuhle, V., D. Jackel, and M. Hensel. 2004. Effector proteins encoded by Salmonella pathogenicity island 2 interfere with the microtubule cytoskeleton after translocation into host cells. Traffic 5:356-370.
26. Kujat Choy, S. L., E. C. Boyle, O. Gal-Mor, D. L. Goode, Y. Valdez, B. A. Vallance, and B. B. Finlay. 2004. SseK1 and SseK2 are novel translocated proteins of Salmonella enterica serovar Typhimurium. Infect. Immun. 72:5115-5125.
27. MacIntyre, S., and J. T. Buckley. 1978. Presence of glycerophospholipid:cholesterol acyltransferase and phospholipase in culture supernatant of Aeromonas hydrophila. J. Bacteriol. 135:402-407.
28. Miao, E. A., M. Brittnacher, A. Haraga, R. L. Jeng, M. D. Welch, and S. I. Miller. 2003. Salmonella effectors translocated across the vacuolar membrane interact with the actin cytoskeleton. Mol. Microbiol. 48:401-415.
29. Miao, E. A., and S. I. Miller. 2000. A conserved amino acid sequence directing intracellular type III secretion by Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 97:7539-7544.
30. Miao, E. A., C. A. Scherer, R. M. Tsolis, R. A. Kingsley, L. G. Adams, A. J. Baumler, and S. I. Miller. 1999. Salmonella typhimurium leucine-rich repeat proteins are targeted to the SPI1 and SPI2 type III secretion systems. Mol. Microbiol. 34:850-864.
31. Mitchell, E. K., P. Mastroeni, A. P. Kelly, and J. Trowsdale. 2004. Inhibition of cell surface MHC class II expression by Salmonella. Eur. J. Immunol. 34:2559-2567.
32. Ohl, M. E., and S. I. Miller. 2001. Salmonella: a model for bacterial pathogenesis. Annu. Rev. Med. 52:259-274.
33. Ruiz-Albert, J., X. J. Yu, C. R. Beuzon, A. N. Blakey, E. E. Galyov, and D. W. Holden. 2002. Complementary activities of SseJ and SifA regulate dynamics of the Salmonella typhimurium vacuolar membrane. Mol. Microbiol. 44:645-661.
34. Salcedo, S. P., and D. W. Holden. 2003. SseG, a virulence protein that targets Salmonella to the Golgi network. EMBO J. 22:5003-5014.
35. Scherer, C. A., and Miller, S. I. 2000. Molecular pathogenesis of salmonellae, p. 266-316. In E. A. Groisman (ed.), Principles of bacterial pathogenesis. Academic Press, San Diego, Calif.
36. Shea, J. E., M. Hensel, C. Gleeson, and D. W. Holden. 1996. Identification of a virulence locus encoding a second type III secretion system in Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 93:2593-2597.
37. Stein, M. A., K. Y. Leung, M. Zwick, F. Garcia-del Portillo, and B. B. Finlay. 1996. Identification of a Salmonella virulence gene required for formation of filamentous structures containing lysosomal membrane glycoproteins within epithelial cells. Mol. Microbiol. 20:151-164.
38. Upton, C., and J. T. Buckley. 1995. A new family of lipolytic enzymes Trends Biochem. Sci. 20:178-179.
39. Vazquez-Torres, A., Y. Xu, J. Jones-Carson, D. W. Holden, S. M. Lucia, M. C. Dinauer, P. Mastroeni, and F. C. Fang. 2000. Salmonella pathogenicity island 2-dependent evasion of the phagocyte NADPH oxidase. Science 287:1655-1658.
40. Wang, R. F., and S. R. Kushner. 1991. Construction of versatile low-copy-number vectors for cloning, sequencing and gene expression in Escherichia coli. Gene 100:195-199.(Maikke B. Ohlson, Kerry F)