Polynucleotide Phosphorylase Negatively Controls spv Virulence Gene Expression in Salmonella enterica
http://www.100md.com
感染与免疫杂志 2006年第2期
Microbiology and Tumor Biology Center, Karolinska Institute, Nobels vg 16, 171 77 Stockholm, Sweden
Wolfson Institute for Biomedical Research, University College London, London WC1E 6BT, United Kingdom
Centre for Molecular Biology and Infection, Department of Infectious Diseases, Imperial College London, Armstrong Road, London SW7 2AZ, United Kingdom
Molecular Microbiology Group, Institute of Food Research, Norwich Research Park, Colney, Norwich, NR4 7UA, United Kingdom
ABSTRACT
Mutational inactivation of the cold-shock-associated exoribonuclease polynucleotide phosphorylase (PNPase; encoded by the pnp gene) in Salmonella enterica serovar Typhimurium was previously shown to enable the bacteria to cause chronic infection and to affect the bacterial replication in BALB/c mice (M. O. Clements et al., Proc. Natl. Acad. Sci. USA 99:8784-8789, 2002). Here, we report that PNPase deficiency results in increased expression of Salmonella plasmid virulence (spv) genes under in vitro growth conditions that allow induction of spv expression. Furthermore, whole-genome microarray-based transcriptome analyses of bacteria growing inside murine macrophage-like J774.A.1 cells revealed six genes as being significantly up-regulated in the PNPase-deficient background, which included spvABC, rtcB, entC, and STM2236. Mutational inactivation of the spvR regulator diminished the increased expression of spv observed in the pnp mutant background, implying that PNPase acts upstream of or at the level of SpvR. Finally, competition experiments revealed that the growth advantage of the pnp mutant in BALB/c mice was dependent on spvR as well. Combined, our results support the idea that in S. enterica PNPase, apart from being a regulator of the cold shock response, also functions in tuning the expression of virulence genes and bacterial fitness during infection.
INTRODUCTION
Members of the genus Salmonella have evolved through the acquisition of several genetic elements that enhance bacterial virulence and enable these bacteria to act as facultative intracellular pathogens (31, 34). These acquired genetic elements include the Salmonella pathogenicity islands (SPIs) (31, 34, 40), the spv virulence gene cluster (33), and selected prophages (25). In addition, Salmonella pathogenicity is dependent on a strict transcriptional regulation of horizontally acquired genetic elements (27, 30, 40, 60, 70, 71). This regulation is coordinated through the bacterial responses to environmental cues experienced during infection (30, 71) and includes the expression of SPI1 genes needed for bacterial invasion (12, 26, 34) and for the induction of proinflammatory responses (37) and the expression of spv and SPI2 genes needed for intracellular survival and replication (10, 30, 34, 56, 57, 71, 77). It has, however, remained enigmatic as to how the imported genetic elements became adapted to the repertoire of preexisting gene regulators. In part, this regulatory compatibility can be explained by the fact that the imported genetic elements rely to a degree on evolutionary conserved regulatory factors for their expression, such as two-component sensor regulatory systems, the RNA-polymerase factors, and nucleoid-associated proteins. These factors then act in concert with specific virulence-associated gene regulatory proteins to achieve the proper induction of virulence genes (71).
Polynucleotide phosphorylase (PNPase; encoded by the pnp gene) belongs to an expanding family of exoribonucleases (7) with homologues identified in eubacteria (22, 29, 79), Drosophila melanogaster (47), plants (44, 87), and even mice and humans (48, 75, 87). The Escherichia coli PNPase participates in RNA degradation (2) and plays a central role in adaptation to growth at low temperature (86). In this context, PNPase assists adaptation to the new environmental situation by specifically degrading mRNAs that code for cold shock proteins (CSPs), a process which is needed for the resumption of bacterial replication after shifting to decreased temperature (86). PNPase is also a crucial component of the RNA degradosome, the multiprotein complex that is responsible in part for mRNA degradation in E. coli (2, 69).
The facultative intracellular pathogen Salmonella enterica serovar Typhimurium causes a systemic infection in mice (8, 53). The murine infection resembles human typhoid fever and involves invasion of the intestinal epithelial cell barrier, subsequent visceral colonization, and replication in phagocytic cells (8, 26, 34, 53). S. enterica serovars Typhimurium and Typhi both contain a gene which is very similar to the E. coli pnp gene (11, 55), and an S. enterica serovar Typhimurium mutant deficient for PNPase exoribonuclease activity shows the expected defect in cold adaptation (11). When grown at 37°C in rich complex medium, the S. enterica serovar Typhimurium pnp mutation does not show alterations in mRNA levels of csp genes but instead expresses increased levels of other mRNA species, in particular those coded for by SPI1 and, to a lesser degree, SPI2 (11). Mutational inactivation of the S. enterica serovar Typhimurium PNPase also results in an altered infection pathogenesis; in contrast to acute systemic infections caused by wild-type (wt) S. enterica serovar Typhimurium, the pnp mutation gains the ability to establish a persistent infection in BALB/c mice (11).
The complexity by which environmental signals regulate the expression of virulence genes in Salmonella is well illustrated by the fact that the bulk of the essential virulence genes are not properly expressed during growth in ordinary complex laboratory medium (71). Therefore, to find out whether virulence genes apart from SPI1 and SPI2 would depend on PNPase for their expression, we have in this report studied the effect of PNPase deficiency on bacterial gene expression under in vitro conditions that simulate selected aspects of the intracellular environment and during actual bacterial replication inside murine macrophage-like J774-A.1 cells. The concomitant results reported here show that the expression of the spv operon is strongly increased in the absence of PNPase under conditions that allow the induction of spv gene expression. In the absence of the transcriptional activator gene spvR, however, no accumulation in spv gene expression was observed with PNPase deficiency, implying that the relaxed expression of spv genes resulting from PNPase deficiency did not relive spv expression from the gene regulator SpvR. Significantly, competition experiments with BALB/c mice showed that the apparent growth advantage of the pnp mutant depended on SpvR as well.
MATERIALS AND METHODS
Bacterial strains, growth media, and plasmids. Salmonella enterica serovar Typhimurium MC1 (SR-11 variant 3181) and the isogenic pnp mutant MC71 are described in Clements et al. (11). Strains were grown in Luria broth (LB) or on Luria agar with antibiotics as appropriate (ampicillin, 100 μg/ml; chloramphenicol, 10 μg/ml; kanamycin, 30 μg/ml; tetracycline, 10 μg/ml). Low-pH minimal medium (MM5.8), as adapted from Kox et al. (42), included 100 mM Bis/Tris (Sigma) buffer (pH 5.8), 0.1% (wt/vol) Casamino Acids, 0.16% (wt/vol) glycerol, and 10 μM MgCl2. Complete cell tissue culture medium (CC) consisted of RPMI 1640 medium (Gibco) supplemented with 10% (wt/vol) fetal bovine serum (Gibco), 2 mM L-glutamine (Sigma), and 20 mM HEPES (Sigma) as previously described (17). To generate phoP, rpoS, or spvR mutants of wt MC1 and the pnp mutant MC71 bacteria, the strains were transduced with P22 int phage lysates (76) containing phoP::Tn10 (23), spvR::Tn5 (72), or rpoS inactivated through insertion with the suicide vector pRR10 (21). A Tn10 tetracycline resistance marker, carried on the virulence plasmid pSLT and shown to be neutral in virulence assays applying BALB/c mice (4), was used to tag bacteria in competition experiments. Plasmid pFF-1 was obtained by cloning the PCR-amplified pnp gene into the T7 RNA-polymerase-dependent expression vector pET21(c) using the primers 5'CGGGATCCCGATGCGAGAAGATCGGGTATT3' and 5'CCGCTCGAGCGGCTCGGCCTGTTCGCTCGC3' with the BamHI and XhoI cloning sites, respectively, in boldface type (11). The fragment thus amplified included 17 bp upstream of the pnp initiation codon, as well as the last codon before the stop signal. In this construct, expression is driven from the plasmid-encoded start codon and ribosomal-binding site. The pET-based vector pHAM-1 encoding the catalytic carboxy-terminal portion of SpvB was available from previous work (67, 83).
Cell culture and infection models. The murine macrophage-like J774-A.1 cells (ATCC TIB67) were grown in RPMI 1640 medium (Gibco) supplemented with 10% (wt/vol) fetal bovine serum (Gibco), 2 mM L-glutamine (Sigma), and 20 mM HEPES (Sigma). To avoid the rapid cytotoxicity mediated by expression of invasion-associated SPI1 genes (9, 50, 63), the bacterial inocula were taken from bacteria grown on agar plates, a condition that does not induce SPI1 expression (16, 71), and opsonized with fresh mouse serum to facilitate bacterial uptake into J774-A.1 cells. To increase the uptake of salmonella, plates were centrifuged at 1,000 x g for 5 min, and this was defined as time zero hour. After 1 h of phagocytosis, extracellular bacteria were killed with 30-μg/ml gentamicin. This medium was replaced after 1 h with medium containing 5-μg/ml gentamicin. Incubations were continued for the time required for each time point tested. The relatively low levels of gentamicin were chosen to avoid any potential effects on intracellular bacterial gene expression by gentamicin (17). Bacteria were released from host cells by hypotonic lysis and enumerated by viable counting on Luria agar plates.
RNA extraction for microarrays. Total bacterial RNA for the microarray experiments was isolated from bacteria propagated in either of three separate growth media. First, for experiments referred to as MM5.8, bacteria were grown aerobically in MM5.8 until they reached optical density at 600 nm (OD600) of 0.5. Second, for CC samples, the bacteria were grown as described in Eriksson et al. (17). The bacterial RNA was in each experiment stabilized in 1% (wt/vol) acidic phenol and 19% (wt/vol) ethanol in water (17, 36, 82). Third, for extraction of bacterial RNA from intracellular bacteria, we refer to a recent description (17). Briefly, for each extraction of Salmonella RNA, a total of 108 J774-A.1 murine macrophage-like cells were seeded in 20 six-well cell culture plates and infected with mouse complement-opsonized S. enterica serovar Typhimurium MC1 or MC71 prepared as described in Eriksson et al. (17), at a multiplicity of infection (MOI) of 100:1 or 10:1 bacteria to cells (17). Eight hours postinfection, the macrophages were lysed for 30 min on ice in 0.1% (wt/vol) sodium dodecyl sulfate (SDS), 1% (wt/vol) acidic phenol, and 19% (wt/vol) ethanol in water, causing lysis of eukaryotic cells but not lysing the intracellular bacteria. The phenol-ethanol mixture acted to stabilize all bacterial RNA (36, 82). Bacteria were isolated by centrifugation (10,000 x g; 30 min) from the host cell lysate obtained from 120 wells of infected J774-A.1 murine macrophage-like cells and pooled. For all experiments, total RNA was prepared using the Promega SV total RNA purification kit. The quality of bacterial RNA samples was analyzed by size fractionation on a microfluid-based system where the fractionated RNA is detected by being stained with an intercalating fluorescent dye (2100 Bioanalyzer; Agilent). This allowed us to simultaneously define the RNA samples with respect to quantity and quality.
Microarray procedures. DNA microarray analysis of gene expression was performed essentially as previously described (17). Briefly, RNA was first reverse transcribed into cDNA and subsequently labeled by random priming with Cy5-dCTP (for labeling protocols, see www.ifr.bbsrc.ac.uk/Safety/Microarrays/#Protocols). Genomic DNA (gDNA) was fluorescently labeled with Cy3-dCTP and used as a reference channel in each experiment. The labeled cDNA and gDNA were mixed together, denatured, and hybridized to the microarray under standard glass coverslips (22 by 22 mm). Hybridization was performed at 65°C overnight, after which the slides were washed twice in 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 sodium citrate) at 65°C for 5 min, followed by washing twice, once with 1x SSC and once with 0.2x SSC for 5 min, each at room temperature. The washed slides were scanned on a GenePix 4000A scanner (Axon Instruments, Inc.). Fluorescent spots and local background intensities were quantified using Genepix Pro software (Axon Instruments, Inc.). All hybridizations were repeated three or four times. Spots showing a reference signal that was lower than the background plus 2 standard deviations or obvious blemishes were excluded from subsequent analyses. The local background was subtracted from spot signals, and fluorescence ratios were calculated. To compensate for unequal dye incorporation or any effect of the amount of template, data centering was performed by bringing the median natural logarithm of the ratios for each group of spots printed by the same pin to zero. The averages from the four data sets from each strain were calculated and compared to each other, and genes that had a >2-fold difference in expression were further analyzed. This new subset of data was analyzed by analysis of variance, and genes for which the difference in expression that had a P value of <0.005 were considered significant.
RNA blot analyses. For Northern blotting, bacteria were grown in MM5.8 to an OD600 of 0.3. Rifampin (200 μg/ml; Sigma) was added, and samples were taken immediately and at defined intervals (0, 2.5, 5, 10, and 15 min). Total RNA was prepared and analyzed for purity as above. Ten micrograms of RNA per lane was separated on denaturing agarose gels and finally transferred onto nylon sheets (Amersham) by capillary blotting. Hybridization conditions were applied as specified by the manufacturer of the membrane (Amersham). Briefly, the membrane was baked for 2 h at 80°C and then prehybridized in 50% (wt/vol) formamide, 4x SSC, 0.2% (wt/vol) SDS, 2.5x Denhardt solution, and 0.05-mg/ml sonicated herring sperm DNA at 42°C for 3 h. Internal fragments of the genes encoding spvR (positions 31630 to 32131, as annotated in the sequenced LT2 genome), spvA (30302 to 30810, as annotated in the sequenced LT2 genome), and 16S rRNA (positions 289339 to 289840, as annotated in the sequenced LT2 genome) were amplified by PCR, labeled with [-32P]dCTP (Amersham) using a random priming labeling kit (Roche), and used as probes. The synthesized probes were added to the prehybridization solution and incubated at constant agitation at 42°C overnight. The membrane was washed twice with 2x SSC and 0.1% (wt/vol) SDS and then twice for 20 min each in 0.1x SSC-0.1% (wt/vol) SDS. A radioisotope imaging system (Phosphoirmager 445SI; Molecular Dynamics) was used to detect radioactivity. Half-lives were calculated using the rate equation for a one-substrate enzyme-catalyzed reaction. For the dot blot experiment, bacteria were grown to OD600 of 0.6 in LB and then transferred to MM5.8 to mimic transition to an intracellular environment. Samples were taken at 1, 1.5, and 2 h and finally overnight, posttransition. Total RNA was prepared and analyzed as above. Twelve micrograms of total RNA per dot was applied onto nylon sheets (Amersham) and UV cross-linked. After this, the same procedure as for Northern blots was used.
Determination of -galactosidase activities. Enzyme activities expressed by lacZ transcriptional constructs were measured according to Miller (59) using 2-nitrophenyl--galactoside (Sigma) as a substrate. The spvR-lacZ (pHUB70), spvA-lacZ (pHUB61), and tlpA-lacZ(pOF1) constructs have been described previously (38, 72, 80). For bacteria grown in MM5.8, the samples used for -galactosidase activity measurements were neutralized with an equal volume of 1 M Tris-HCl buffer, pH 8.0, before addition to the reaction mixture. For comparing -galactosidase activities, we used an unpaired t test with Welch correction.
Immunostaining and immunofluorescence. Murine macrophage-like J774-A.1 cells were infected as described for the microarray experiment (using MOI as indicated) and subsequently fixed in a freshly prepared 4% (wt/vol) formaldehyde solution, prepared as described by Eriksson et al. (18). The fixed cells were subsequentially permeabilized in 0.1% (wt/vol) saponin and stained with antibodies as indicated in a solution with 10% (wt/vol) horse serum and 0.1% (wt/vol) saponin (3). Texas red-conjugated phalloidin was from Molecular Probes. For primary antibodies, we used a 1:200 dilution of the rabbit 156 anti-lysosome-associated membrane glycoprotein 1 (LAMP-1) (3) and a 1:200 dilution of goat anti-Salmonella (CSA-1) (3), while for secondary antibodies we used rhodamine red X-conjugated donkey anti-rabbit and Cy2-conjugated donkey anti-goat antibodies (both diluted 1:200). The samples were mounted using Aqua polymount (Polysciences) and analyzed by confocal microscopy (3).
Assessment of cytotoxicity. Three separate methods were used to score the viability of the macrophage-like J774A.1 cells during infection. (i) The proportion of attached to detached cells was estimated by phase-contrast microscopy. To document these observations, the cells were grown on glass coverslips, fixed, and stained for bacteria and actin after infection as specified above. Thereafter, the density of infected murine macrophage-like J774 cells was monitored by microscopical counting. (ii) As release of lactate dehydrogenase (LDH) is used as a defined measure of SPI1- and SPI2-induced induced cell death (14, 62, 63), we used a cytotoxicity detection kit (Roche) measuring LDH activity in the cell culture supernatants after given time points postinfection. The activities are given as the percentage of LDH activity released in relation to the LDH activity obtained by lysing a comparable amount of uninfected cells with Triton X-100. (iii) Apoptosis was monitored by measurement of nuclear fragmentation (28, 51). For these experiments, infected murine macrophage-like J774-A.1 cells were collected by centrifugation (800 x g; 1 min). This included both the attached cells and the potentially detached cells (in total, approximately 2 x 105 cells). The pelleted cells were washed in cold phosphate-buffered saline. The pellet was then rapidly dissolved in 200 μl of 10 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 1% (wt/vol) Triton X-100 buffer. Thereafter, the suspension was separated into "Triton-insoluble" and "Triton-soluble" fractions by centrifugation (14,000 x g; 5 min). A small proportion of the supernatant was then immediately run on a 1% (wt/vol) agarose gel. Meanwhile, the pellet was solubilized into 100 μl of a 10 mM Tris-HCl-2 mM EDTA-1% (vol/vol) buffer (pH 7.5) and analyzed on a 1% agarose gel. Finally, the rest of the Triton-soluble material was extracted with phenol-chloroform and precipitated with ethanol by using glycogen (Roche) as a carrier. The material was dissolved into 10 μl distilled water, treated with RNase, and analyzed on 1% (wt/vol) agarose gels. Agarose gels included ethidium bromide (0.5 μg/ml; Sigma) to reveal nucleic acids.
Infection experiments. For in vivo infection experiments, we used S. enterica serovar Typhimurium strains MC1 and MC1 spvR mutant (11, 72) tagged with a Tn10 element on pSLT and previously shown to be neutral in mice infection experiments (4) and MC71 and MC71 spvR mutant strains. Infection experiments were carried out as previously described (11, 78). Briefly, female BALB/c mice, 6 to 8 weeks of age, were infected intraperitoneally with a 50:50 mixture of MC1-Tn10 and MC71 bacteria in a total dose of approximately 5 x 103 bacteria per mouse or with a 50:50 mixture of MC1-Tn10-spvR mutant strain and MC71 spvR mutant bacteria, also in a total dose of 5 x 103 bacteria per mouse. Groups of five mice were used for each of two series of experiments (10 mice per competition in total). Three days after infection, mice were sacrificed, and the livers and spleens were removed for enumeration of the total amount of viable bacteria, as well as the percentage of tetracycline-resistant bacteria. To compare the significance in ratios, we applied the two-sided Mann-Whitney U test.
RESULTS
Whole-genome scale probing for bacterial gene expression in vitro reveals a strong association between PNPase and spv gene expression. Our previous transcriptome analyses of Salmonella enterica serovar Typhimurium gene expression in vitro were carried out with bacterial cultures replicating in complex LB medium (11). While these analyses defined a strong association between SPI1 gene expression and PNPase, the growth conditions applied were not optimal for the expression of several virulence genes, such as SPI2 (39) or spv genes (17, 71). To further define the repertoire of virulence genes affected by PNPase, we extended the microarray-based transcriptome analysis to include bacteria grown in deprived medium with a low Mg2+ concentration and a pH of 5.8 (MM5.8). The composition of the medium was primarily designed to mimic the intravacuolar environment in which Salmonella replicates (17, 42) and is known to induce expression of SPI2 (78).
We first defined the transcriptome for the wild-type, PNPase-proficient S. enterica serovar Typhimurium strain MC1 grown in MM5.8 medium. Samples for RNA preparation were taken when the culture entered the transition into the stationary phase (OD600, 0.5). To probe for alterations in gene expression, we used cultures grown in CC as a comparator and applied a false discovery rate of 0.005. Under these conditions, we defined 447 genes as up-regulated and 479 genes as down-regulated during replication in MM5.8 (see Table S1 in the supplemental material). Among the genes strongly induced for expression, we identified representatives of the SPI2 and the spv gene clusters, whereas among the genes strongly suppressed we identified SPI1 and genes for flagellation and motility (see Table S1 in the supplemental material).
Having shown that MM5.8 did induce expression of genes that are normally induced when bacteria are residing in an intracellular compartment (17), we next tested whether PNPase affected the mRNA levels of any of these genes. To probe for the effect of PNPase deficiency, we applied S. enterica serovar Typhimurium strain MC71, which was constructed from MC1 by introducing a STOP codon in pnp by site-directed mutagenesis (11). When MC71 was grown in MM5.8 to the transition to the stationary phase of growth (OD600, 0.5), MC71 showed a relative increase in expression of 37 genes and a relative decrease in the expression of 9 genes compared to wt MC1 (Fig. 1). More specifically, the comparison revealed a strong (7- to 8-fold) relative increase in spv expression in MC71, in addition to an increase (2- to 10-fold) in mRNA levels of genes located in prophages (STM0906, STM0908, STM0914, STM2233 to STM2237, STM2587, STM2616, and STM2620) and of genes encoding proteins involved in stress responses (STM1251 and ibpAB) and DNA repair (mutM and recF). In parallel, we found a PNPase-dependent decrease (two- to threefold) in the expression of genes associated with purine synthesis (carA, purK, trpCD, and purF) and of genes involved in flagellar synthesis (flgBC) (Fig. 1).
mRNA analyses reveal increased levels of spv gene mRNA at PNPase deficiency in vitro. The spv gene cluster is transcribed from two main promoters, one in front of the transcriptional activator protein gene spvR (PspvR) and another in front of spvA (PspvA) directing the expression of Spv effector proteins (13, 43, 52, 81). To verify the association between PNPase and spv expression, we next determined spvA expression by comparing spvA mRNA levels in the wt MC1 PNPase-proficient and the MC71 PNPase-deficient mutant background by carrying out Northern blotting to probe for spvA mRNA during logarithmic growth (OD600, 0.3) in MM5.8 after blocking RNA synthesis with rifampin. This analysis clearly showed expression of spvA mRNA in both MC1 and MC71. The analysis also revealed an overall higher spvA mRNA content and an alternative spvA mRNA pattern in MC71 (Fig. 2A); whereas both MC1 and MC71 showed spvA mRNA of apparently higher molecular weight named A1, MC71 revealed clear accumulation of a second spvA mRNA named A2 that was only weakly detected from MC1 (Fig. 2A and B). When using rifampin to block the mergenc of nascent mRNA species, we could not demonstrate any stabilization of the spvA1 mRNA in MC71 in either of two separate sets of experiments (Fig. 2A and data not shown). In contrast, the spvA1 mRNA appeared more stable in MC1 (Fig. 2A and B), suggesting that this transcript was differentially degraded by PNPase.
Parallel Northern analyses using a 32P-labeled PCR-generated spvR fragment revealed a clear signal for spvR in MC71 but no signal in MC1 (data not shown). This is consistent with the previously reported difficulty in detecting the spvR transcript in S. enterica (1). Nevertheless, these analyses confirmed the increased spv mRNA levels in MC71 implied by microarray analyses.
Transcriptional spv-lacZ fusions reveal PNPase-dependent expression of spv genes. The fact that we could not conclusively demonstrate any increase in the half-life of the spvA mRNA in the PNPase-deficient strain MC71 (Fig. 2A and B) prompted us to test whether an increased transcriptional strength of spv promoters could account for the increased spvA mRNA levels observed with the PNPase-deficient background. The constructs used included the spvA (pHUB61) or spvR (pHUB70) promoter positioned to direct lacZ expression in a pACYC184-based construct (Fig. 3A) (72). As the expression of spvA-lacZ fusions tends to be weak in rich growth medium (84) and indeed confirmed for pHUB61 grown in CC (data not shown), we initiated a series of experiments to compare the expression of -galactosidase from spv-lacZ fusions when the bacteria were grown overnight in MM5.8 in batch cultures (15 to 17 h). Regarding the spvA-lacZ construct pHUB61, the -galactosidase levels obtained when the bacteria were grown overnight in MM5.8 were significantly higher for MC71 than for PNPase-proficient MC1 (Fig. 3B). Likewise, in comparison to MC1/pHUB70, we noted a significant increase in the -galactosidase levels expressed by MC71/pHUB70 grown in MM5.8 (Fig. 3C).
To ensure that the alterations in -galactosidase activities in the PNPase-deficient MC71 background were not due to stabilization of lacZ mRNA or to alterations in the copy number of the operon fusion constructs in MC71, we applied the tlpA-lacZ fusion pOF1 as a control (38). Like the spv genes, tlpA is contained on the virulence plasmid pSLT but was not induced during bacterial replication in MM5.8 (see Table S1 in the supplemental material). Furthermore, pOF1 relied on the same lacZ cartridge as pHUB61 and pHUB70 and cloned with tlpA in the same position in the same cloning vector. The tlpA-lacZ fusion did not show increased but rather a slightly decreased level of activity in the pnp mutant MC71 background when grown in MM5.8 (Fig. 3D), which indeed suggests that the increased -galactosidase activities obtained with spvA-lacZ and spvR-lacZ in MC71 originated from increased transcriptional activity.
The increased expression of spvA in overnight cultures of PNPase-deficient MC71 was also verified by blotting for spvA mRNA. For this experiment, PNPase-proficient MC1 and MC71 were grown in MM5.8, and the bacterial RNA was extracted at early (1 h, 1.5 h, and 2 h) and late (overnight) time points of growth. Dot blot assays applying the PCR-amplified and 32P-labeled spvA gene as a probe revealed no significant signal for spvA mRNA during the very early time points in either MC1 or MC71 (Fig. 2C). Also, we could not detect any significant accumulation of spvA mRNA in the overnight cultures of MC1. This would be consistent with the observation that spvA mRNA becomes induced during logarithmic growth in minimal medium low in magnesium, to become only weakly detectable after prolonged incubation (84). In contrast and in accordance with the LacZ measurements from MC71/pHUB61, PNPase-deficient strain MC71 showed a substantial accumulation of spvA mRNA in MC71 after overnight incubation (Fig. 2C).
We next tested if complementation of the PNPase deficiency in MC71 could restore the spvA-lacZ activity to wild-type levels. Initial attempts using cloned pnp to complement for the pnp mutation in trans were not successful, due to our inability to maintain pnp-containing plasmid constructs in S. enterica serovar Typhimurium. However, when the pnp open reading frame was contained in the pET21(c) vector under the T7 promoter (pFF-1), the constructs could be maintained. The pFF-1 construct caused a reduction in the spvA-lacZ activities in the MC71 pnp mutant background but did not affect lacZ fusion expression in the MC1 wt background (Fig. 3E). A control construct containing the 3'-terminal portion of spvB in the pET32 vector (pHAM-1) (67, 83) did not mediate complementation (Fig. 3E). These observations implied that the T7 promoter shows weak expression in S. enterica serovar Typhimurium, even though the bacterium formally lacks the T7 RNA polymerase and the PNPase deficiency of MC71 can be complemented in trans.
PNPase deficiency does not relieve spvA gene expression from dependency on growth phase, PhoP, RpoS, or SpvR. Expression of spv-lacZ fusions in bacteria grown in low-magnesium minimal medium is reported to be dependent on the bacterial growth phase, the transcription factor SpvR, and the alternative -factor RpoS (84, 85). Furthermore, the conditions prevailing in MM5.8 are likely to activate the PhoP/PhoQ response regulatory system (42), also implicated in spv gene expression (30). Therefore, we set out to test whether the -galactosidase activity expressed by the spvA-lacZ constructs in the PNPase-deficient background MC71 grown in MM5.8 still depended on the bacterial growth phase and the transcription factors PhoP, RpoS, and SpvR.
The growth slopes in MM5.8 appeared similar for both PNPase-proficient MC1 and PNPase-deficient MC71 (Fig. 3F). However, the concomitant expression of -galactosidase became much stronger in MC71/spvA-lacZ when the cultures entered late logarithmic phase of growth (Fig. 3G). In parallel, the spvR-lacZ fusion construct pHUB70 was observed to express a similar growth-phase-dependent induction of -galactosidase activity when the bacteria were grown in MM5.8, albeit the difference in -galactosidase activity between MC1 and MC71 remained lower (data not shown).
To test to what extent the increase in spvA expression that associated with PNPase deficiency during propagation in MM5.8 still depended on the transcription factors PhoP, RpoS, or SpvR implicated in spv gene activation (5, 20, 21, 30, 35, 43, 68, 71, 84, 85), we measured -galactosidase activities encoded by pHUB61 (spvA-lacZ) in phoP, rpoS, and spvR mutants of MC1 and MC71. Inactivation either of phoP, rpoS, or spvR in PNPase-proficient MC1 or PNPase-deficient MC71 resulted in a significant reduction of spvA-lacZ expression in overnight cultures grown in MM5.8 (Fig. 3H). Thus, PNPase deficiency did affect the transcriptional strengths from both the spvA and spvR promoters, yet PNPase deficiency apparently did not evoke or apply any alternative bypass routes for spvA gene activation.
PNPase-proficient and PNPase-deficient bacteria do not differ in their ability to induce cytotoxicity in J774-A.1 cells 8 h postinfection. We have recently developed a method that enables transcriptometric analysis of bacteria growing within host cells (17, 36). However, to generate enough bacterial RNA for microarray analyses, we needed to apply an MOI of 100:1 (17), potentially generating infection-associated cytotoxicity and apoptosis (14, 45, 46, 49, 62, 63). Therefore, before defining and comparing the transcriptomes of PNPase-proficient MC1 and PNPase-deficient MC71 while growth occurred in host cells, we analyzed the murine macrophage-like J774-A.1 cells infected with S. enterica serovar Typhimurium MC1 and MC71 for possible alterations in toxicity or bacterial growth.
Upon infection of J774-A.1 macrophage-like cells with PNPase-proficient MC1 or PNPase-deficient MC71 at an MOI of 100:1, no signs of cytotoxicity or detachment of murine macrophage-like J774-A.1 cells as judged by phase-contrast microscopy were observed at 8 h postinfection (data not shown). Likewise, as followed by immunofluorescence microscopy labeling for Salmonella bacteria and filamentous actin at 10 h postinfection, the density and appearance of the host cells on glass slides were very similar whether the cells were infected with MC1 or MC71 bacteria (Fig. 4A and B). When the infected murine macrophage-like J774-A.1 cells were monitored for the release of lactate dehydrogenase (LDH) as a general indicator of cytotoxicity, the levels of released LDH observed at 8 h postchallenge by application of an MOI of 100:1 or 18 h postchallenge by application of an MOI of 10:1 remained low (Fig. 4C). The LDH levels released by cells infected with MC71 at an MOI of 100:1 were slightly but statistically significantly higher than those released upon infection with MC1. Yet, the LDH levels remained very low compared to the levels reported with invasive SPI1-expressing bacteria (14, 63) and to the levels contained in a comparable number of uninfected cells (Fig. 4C) (63). Similarly, when analyzed on ethidium bromide-stained agarose gels, no detectable DNA fragmentation was apparent at 8 h postinfection when an MOI of 100:1 of noninvasive complement-opsonized bacteria was applied (Fig. 4D). However, if the uninfected cells were allowed to overgrow the cell culture, leading to acidified and crowded cell conditions, DNA fragmentation was indeed detected (Fig. 4D).
In separate infection experiments, immunofluorescence microscopy labeling for intracellular bacteria revealed PNPase-proficient MC1 and PNPase-deficient MC71 bacteria residing in an LAMP-1-positive compartment, with no differences observed in the labeling pattern for the two strains (Fig. 4E to J). Viable counts, performed in parallel from nonfixed and unstained cells 8 h postinfection, revealed a 5.4-fold and 5.0-fold growth index, respectively, for the MC1 wt and MC71 pnp mutation, when the growth yields were compared to intracellular bacterial counts at 2 h postinfection. A repetition of the growth experiment did not provide any apparent alterations in the growth yields (data not shown).
In summary, the host cells appeared intact 8 h postinfection, and no significant differences in cytotoxicity, intracellular localization, or bacterial growth yields were found between cells infected with either PNPase-proficient or PNPase-deficient S. enterica serovar Typhimurium strains.
Intracellular transcriptome of S. enterica serovar Typhimurium SR-11. Having established the infection procedures, we next determined the transcriptome of the PNPase-proficient MC1 strain when replicating in the macrophage-like cells. For these experiments, bacterial RNA was extracted at 8 h postinfection was compared to RNA extracted from CC-grown bacteria. This time point was previously defined to reveal the majority of the gene expression profile alterations upon infection of murine macrophage-like J774-A.1 cells (17). Since we previously determined the transcriptome of S. enterica serovar Typhimurium SL1344 when replication occurs within host cells, we first compared the transcriptomes of intracellular S. enterica serovar Typhimurium of lines SR-11 and SL1344 to define to what extent the pattern of gene expression was dependent on the bacterial strain background. The gene expression profiles of the two lines were consistent: 94.4% of the genes revealed similar expression patterns in both strains after application of a twofold cutoff (false-discovery rate = 0.005) (see Table S1 in the supplemental material) (17). A proportion of the expressional difference reflected the histidine auxotrophy of strain SL1344. Nevertheless, these observations show that the transcriptomes of the two different strains of S. enterica serovar Typhimurium were comparable when the bacteria were replicating within murine macrophage-like J774-A.1 cells.
We then evaluated to what extent the intracellular transcriptome reflected the one obtained when bacteria were replicating in MM5.8. The comparison showed that 42% of the MC1 genes that were up-regulated in murine macrophage-like J774-A.1 cells were also more highly expressed in MM5.8 (see Table S1 in the supplemental material). It is noteworthy that virulence genes within SPI2 (ssaG), SPI3 (mgtB), and the spv locus (spvA) and a gene within SPI5 (pipD) that functionally associates with SPI2 were clearly represented within this category of genes (see Table S1 in the supplemental material). Almost one-third (30.5%) of MC1 genes that were down-regulated in murine macrophage-like J774-A.1 cells were also expressed at lower levels in MM5.8 (see Table S1 in the supplemental material), including genes within SPI1 (hilC) and SPI4 (STM4260) and motility-related (fliC) genes, as well as a gene within SPI5 (pipC) that associates with SPI1 gene expression (see Table S1 in the supplemental material) (40). Thus, while growth in vitro in MM5.8 obviously did not reproduce all aspects of the SCV, the large proportion of genes that showed parallel trends of expression in MM5.8 and during infection of murine macrophage-like J774-A.1 cells confirms that MM5.8 indeed does represent certain crucial aspects of the intravacuolar compartment.
PNPase deficiency affects spv expression not only in vitro but also in murine macrophage-like J774-A.1 cells. Having defined the intracellular transcriptomes for PNPase-proficient S. enterica serovar Typhimurium MC1, we compared the transcriptomes of intracellular MC1 and PNPase-deficient MC71. The gene expression profiles obtained at 8 h postuptake were found to reveal an extremely high degree of similarity. Only six genes (spvABC, entC, rtcB, and STM2236) were up-regulated, whereas the motility-associated cheAW and motB genes were down-regulated in intracellular MC71 compared to intracellular MC1 bacteria (Fig. 5A). These data showed that PNPase specifically affected the expression of a small set of genes during macrophage infection.
We next assessed to what extent inactivation of spvR, the gene encoding a key activator of spvABCD transcription (Fig. 3H) (71, 81), influenced the intracellular transcriptome. As measured by microarray analysis, expression of spvAB and to some extent spvC became significantly reduced in both PNPse-proficient MC1 and PNPase-deficient MC71 upon mutational inactivation of spvR (Fig. 5B). More significantly, the differences observed for spv gene expression between intracellular MC1 and MC71 were not observed in the spvR mutant backgrounds. Surprisingly, inactivation of spvR also affected the increased mRNA levels observed for entC in the PNPase-deficient MC71 (Fig. 5B). This shows that much of the differences observed in the intracellular transcriptomes for PNPase-proficient MC1 and PNPase-deficient MC71 depended on SpvR.
PNPase deficiency associates with a spvR-dependent alteration in virulence in BALB/c mice. Apart from affecting virulence gene expression, PNPase deficiency was previously reported to affect infection pathogenesis and replication of the bacteria in BALB/c mice (11). Here, we followed up this phenomenon by performing in vivo competition experiments between PNPase-proficient MC1 and PNPase-deficient MC71 bacteria in BALB/c mice after intraperitoneal infection. Upon recovery of bacteria from the livers and spleens 3 days postinfection, the MC71 bacteria showed a higher competitive index in both the livers and spleens in two separate sets of experiments (the combined data are shown Fig. 6). As PNPase deficiency was found associated with increased spv gene expression in murine macrophage-like J774-A.1 cells (Fig. 5A) and as the spv genes participate in bacterial replication during the systemic phase of infection (32, 33, 73), we next tested the effect of inactivating spvR on the competitive indexes in mice. Thus, spvR mutants of MC1 and MC71 were competed against each other in BALB/c mice after intraperitoneal challenge. Astonishingly, under these premises MC71 lost its competitive advantage in mice in each of two separate sets of experiments, and in fact the competitive index for MC71 decreased (the combined data are shown in Fig. 6). Therefore, a noticeable proportion of the in vivo alteration in competitive index conferred by PNPase deficiency was dependent on alterations in spvR and SpvR-regulated gene expression.
DISCUSSION
For many serovariants of Salmonella enterica, virulence in the mouse model for systemic salmonellosis is dependent on the bacterial spv gene cluster that promotes bacterial replication in the liver and spleen, as the bacteria reside in macrophages (32, 33). While the exact details by which Spv proteins participate in the disease process are not known, the molecular function has been determined for two of the five Spv proteins: SpvB is an actin-specific mono(ADP-ribosyl)transferase (46, 49, 67, 83), whereas SpvR acts as transcriptional activator for spv expression (5, 43, 54, 68, 80, 81, 84, 85). However, the genetic control of spv gene expression is complex and requires selected environmental cues, as well as the interplay between several transcriptional regulators (71). Expression of spv genes is clearly induced when the bacteria reside within mammalian cells (17, 24, 72) and in vitro when the bacteria are grown in minimal medium mimicking the intravacuolar environment (84, 85). Expression of spv genes can also be achieved, at least to some extent, as the bacteria enter the stationary phase of growth in rich medium (13, 15, 21, 80, 84, 85). Apart from SpvR, the regulatory factors that participate in spv gene induction include the response regulator PhoP (30) the alternative -factor RpoS (15, 21, 41, 71), and integration host factor (52). In addition, the global gene regulator H-NS (65, 71), leucine-responsive regulatory protein (52), and the catabolite repression system (66) have been identified as repressors of spv gene expression.
We have previously demonstrated that mutational inactivation of the pnp gene encoding the exoribonuclease PNPase in Salmonella enterica serovar Typhimurium resulted in apparent increased bacterial replication in the host and in the establishment of chronic carrier states among convalescent mice (11). In parallel, when the bacteria were grown in vitro in rich growth medium, the PNPase-deficient mutant revealed stronger expression of invasion-associated SPI1 genes and to some extent increased expression of SPI2 genes (11), implying that PNPase-activity somehow connected with virulence gene expression. A recent investigation demonstrated that persistent salmonellosis in the mouse involves prolonged bacterial colonization of macrophages (61). Therefore, in the current study we probed the effect of PNPase on S. enterica serovar Typhimurium gene expression while the bacteria were replicating in an in vitro growth medium that partially mimicked the intravacuolar environment (MM5.8) or within murine macrophage-like J774-A.1 cells.
When whole-genome-based microarray technology was applied to define different bacterial transcriptomes, growth in MM5.8 was found to result in decreased expression of SPI1 genes and in a strong increase in the SPI2 and spv gene expression in both the PNPase-proficient MC1 and PNPase-deficient MC71 backgrounds. Somewhat surprisingly, PNPase deficiency associated only with a fairly moderate alteration in the transcriptome when the bacteria were grown in MM5.8. The alteration recorded in the PNPase-deficient MC71 background included genes associated with stress responses, DNA repair, and, significantly, the spv genes (Fig. 1; see Table S1 in the supplemental material).
Yet while Northern blot analyses clearly confirmed an accumulation of spvA mRNA in the PNPase-deficient MC71 background grown in MM5.8, we could not demonstrate any increase in the half-life of the spvA mRNA in MC71 (Fig. 2A and B). In parallel, lacZ fusions probing the transcription of the spvA and spvR promoters also revealed higher -galactosidase levels in the PNPase-deficient MC71 background (Fig. 3B and C). These observations suggest that the increased levels of spvA expression observed at PNPase deficiency, at least partially, could originate from an increased transcription from the spvA promoter. Still, the expression of spvA-lacZ remained dependent on growth phase and on the transcriptional regulators PhoP, RpoS, and SpvR, even in the PNPase-deficient MC71 strain (Fig. 3F to H). This would be in accordance with the previously defined strict dependency of the spvA promoter on RpoS and SpvR (35, 84, 85). Thus, one could envision a model in which spvA mRNA is not directly affected by PNPase. Instead PNPase could, by affecting the amount or activity of accessory transcription factors or regulatory RNAs, control spv expression indirectly. However, the array data did not reveal any significant increase in the mRNAs in MC71 for the activator genes phoP or rpoS or any significant decrease in the mRNAs for H-NS or the leucine-responsive regulatory protein (see Table S1 in the supplemental material). Yet, when we compared the genes altered for their expression at shift to MM5.8, we did find parallels to the PhoPQ-regulon recently revealed for S. enterica serovar Typhimurium grown in magnesium-deprived M9 minimal medium (64). The transcriptomes of MC1 and MC71 grown in MM5.8 showed induction in the expression of the PhoPQ-connected bioA, cysCDJN, mig-3, pagOC, oat, udg, ybaY, and ygaY genes and a repression in the expression of the PhoPQ-connected nrdFH, slyB, and yfbE genes, compared to the transcriptomes expressed in CC (see Table S1 in the supplemental material). Furthermore, it is interesting to note that the promoter region of spvA in particular does share sequences reminiscent of the PhoP-binding site (38). It thus remains quite possible that MM5.8 provokes activation of the PhoPQ regulon, while it may not activate phoPQ expression in itself. Nevetheless, the growth medium applied by Monsieurs et al. (64) to provoke the PhoPQ regulon differs in many respects from MM5.8, which might explain differences in the transcriptome profiles observed.
The transition from CC into murine macrophage-like J774-A.1 cells is known to have a drastic effect on the bacterial transcriptome, and this alteration also includes induction of spv gene expression (17). Therefore, we used the microarray analysis to define differences in the intracellular transcriptomes displayed by the PNPase-proficient MC1 and PNPase-deficient MC71 strains while the bacteria were grown in murine macrophage-like J774-A.1 cells. The comparison of the intracellular transcriptomes of MC1 and MC71 revealed a significantly different expression of only nine genes and included spvA, spvB, and spvC as being up-regulated in MC71 (Fig. 5A). The expression of spvA, spvB, and spvC genes was reduced to equal levels in spvR mutant derivatives of MC1 and MC71 grown in murine macrophage-like J774-A.1 cells (Fig. 5B). Somewhat unexpectedly, introducing the spvR mutation into MC71 also reduced the increase in entC expression conferred by PNPase deficiency. Thus, it remains possible that expression of the isochorismate synthetase EntC is affected by the expression of spv genes. As entC expression was grossly unaffected by spvR in the PNPase-proficient MC1 strain, it is unlikely that entC would directly rely on SpvR for its expression. However, we cannot exclude the possibility that entC expression would respond to an alternative PNPase-suppressed and SpvR-dependent transcriptional activation.
Mutational inactivation of PNPase was originally reported to affect not only SPI1 and SPI2 expression, but also infection pathogenesis in BALB/c mice (11). The pnp mutant MC71 displayed higher bacteria doses in the spleens of the infected mice at early time points after infection; the mutant bacteria showed increased rates of replication in mice and finally were capable of establishing persistent infections (11). As PNPase deficiency clearly caused an increase in spv gene expression when the bacteria were growing in MM5.8 or in murine macrophage-like J774-A.1 cells, we probed the effect of inactivating the spvR gene on bacterial replication in vivo. This was done by comparing the competitive indexes of the pnp mutant when competed against a PNPase-proficient strain in BALB/c mice and by competing corresponding spvR mutants. The results obtained (Fig. 6) clearly showed that the growth advantage conferred in mice by PNPase deficiency depended on spv gene expression. Thus, PNPase seems not only to suppress spv gene expression under conditions that induce spv gene expression, but also to suppress bacterial replication in BALB/c mice.
In many respects, the pattern of regulation of PNPase-dependent expression of virulence in S. enterica genes resembles that of the PNPase-directed cold shock response in E. coli; spv genes, as well as csp genes, are induced by environmental changes. Both classes of genes are needed to adapt the bacteria to new environmental demands, and the mRNA levels for all of these genes are modulated by PNPase. In E. coli, PNPase is not required for the induction of the cold shock response itself but for degradation of the csp mRNAs during adaptation to decreased growth temperature (86). A parallel scenario could be envisaged for the interplay between PNPase and virulence gene expression in Salmonella. According to the gene expression profile data presented here, PNPase is not necessary for the induction of spv virulence gene expression but may function to restrict gene expression once the genes have been induced. Nevertheless, the effect of PNPase on the expression of a given virulence gene may not necessarily reflect a PNPase-mediated degradation of the given mRNA. Instead, PNPase could affect gene expression indirectly through affecting the stability of the mRNA encoding a gene regulatory factor, the stability of regulatory RNA molecules, or the regulation of a factor otherwise needed for proper SPI expression (58). Any such scenario could well explain the increased spvA expression observed here for the PNPase-deficient strain MC71.
Recent studies of Yersinia virulence suggest yet another mechanism for PNPase-mediated regulation of virulence expression. As for Salmonella, the virulence of Yerinia relies on the expression of type III secretion (6). The effector proteins secreted by the Yersina type III secretion system prevent phagocytosis and expression of the oxidative burst (6, 74). Inactivation of Yersinia PNPase caused a delay in the secretion of such virulence proteins to the extent that the mutants expressed decreased survival in murine macrophage-like RAW cells (74). The defect in the expression did not rely on the level of gene expression but rather occurred at the posttranslational level. In parallel, another component of the RNA degradosome, the endoribonuclease RNaseE, has also been implicated in the expression of hilA, a main transcriptional regulatory gene of SPI1 (19).
We propose that PNPase can fine-tune the expression of environmentally induced genes, such as cold-shock or virulence genes, and allow bacteria to successfully adapt to alternating biological niches. Such observations suggest that imported genetic virulence elements also could be subjected to posttranscriptional regulation and that their regulation has become integrated not only into the network of existing gene regulatory proteins, but also to the activity of preexisting ribonucleases.
ACKNOWLEDGMENTS
We are grateful for support from VR Cancer fonden and STINT (M.R.), for a Core Strategic grant from the BBSRC, and for an EU Marie Curie Training site grant (J.C.D.H.). Work in D.W.H.'s laboratory was supported by grants from the Medical Research Council (United Kingdom) and the Wellcome Trust.
Supplemental material for this article may be found at http://iai.asm.org/.
REFERENCES
1. Abe, A., and K. Kawahara. 1995. Transcriptional regulation and promoter sequence of the spvR gene of virulence plasmid pKDSC50 in Salmonella choleraesuis serovar Choleraesuis. FEMS Microbiol. Lett. 129:225-230.
2. Bernstein, J. A., P. H. Lin, S. N. Cohen, and S. Lin-Chao. 2004. Global analysis of Escherichia coli RNA degradosome function using DNA microarrays. Proc. Natl. Acad. Sci. USA 101:2758-2763.
3. 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.
4. Bjrkman, J., M. Rhen, and D. I. Andersson. 1996. Salmonella typhimurium cob mutants are not hypervirulent. FEMS Microbiol. Lett. 139:121-126.
5. Caldwell, A. L., and P. A. Gulig. 1991. The Salmonella typhimurium virulence plasmid encodes a positive regulator of a plasmid-encoded virulence gene. J. Bacteriol. 173:7176-7185.
6. Carniel, E. 2002. Plasmids and pathogenicity islands in Yersinia. Curr. Top. Microbiol. Immunol. 264:89-108.
7. Carpousis, A. J. 2002. The Escherichia coli RNA degradosome: structure, function and relationship in other ribonucleolytic multienzyme complexes. Biochem. Soc. Trans. 30:150-155.
8. Carter, P. B., and F. M. Collins. 1974. The route of enteric infection in normal mice. J. Exp. Med. 139:1189-1203.
9. Chen, L. M., K. Kaniga, and J. E. Galán. 1996. Salmonella spp. are cytotoxic for cultured macrophages. Mol. Microbiol. 21:1101-1115.
10. Clements, M., S. Eriksson, D. Tezcan-Merdol, J. C. Hinton, and M. Rhen. 2001. Virulence gene regulation in Salmonella enterica. Ann. Med. 33:178-185.
11. Clements, M. O., S. Eriksson, A. Thompson, S. Lucchini, J. C. Hinton, S. Normark, and M. Rhen. 2002. Polynucleotide phosphorylase is a global regulator of virulence and persistency in Salmonella enterica. Proc. Natl. Acad. Sci. USA 99:8784-8789.
12. Collazo, C. M., and J. E. Galán. 1997. The invasion-associated type III protein secretion system in Salmonella—a review. Gene 192:51-59.
13. Coynault, C., V. Robbe-Saule, M. Y. Popoff, and F. Norel. 1992. Growth phase and SpvR regulation of Salmonella typhimurium spvABC virulence genes. Microb. Pathog. 13:133-145.
14. Decktrach, D., L. A. Knodler, K. Galbraith, and O. Steele-Mortimer. 2005. The Salmonella SPI1 effector SopB stimulates nitric oxide production long after invasion. Cell. Microbiol. 7:105-113.
15. El-Gedaily, A., G. Paesold, C.-Y. Chen, G. D. Guiney, and M. Krause. 1997. Plasmid virulence gene expression induced by short chain fatty acids in Salmonella dublin: identification of rpoS-dependent and rpoS-independent mechanisms. J. Bacteriol. 179:1409-1412.
16. Eriksson, S., M. Clements, and M. Rhen. 2001. Salmonella as a tenant in phagocytic cells. Res. Adv. Microbiol. 2:87-102.
17. Eriksson, S., S. Lucchini, A. Thompson, M. Rhen, and J. C. Hinton. 2003. Unravelling the biology of macrophage infection by gene expression profiling of intracellular Salmonella enterica. Mol. Microbiol. 47:103-118.
18. Eriksson, S., B. J. Chambers, and M. Rhen. 2003. Nitric oxide production by murine dendritic cells is cytotoxic for intracellular Salmonella typhimurium. Scand. J. Immunol. 58:493-502.
19. Fahlen, T. F., N. Mathur, and B. D. Jones. 2000. Identification and characterization of mutants with increased expression of hilA, the invasion gene transcriptional activator of Salmonella typhimurium. FEMS Immunol. Med. Microbiol. 28:25-35.
20. Fang, F. C., M. Krause, C. Roudier, J. Fierer, and D. G. Guiney. 1991. Growth regulation of a Salmonella plasmid gene essential for virulence. J. Bacteriol. 173:6783-6789.
21. Fang, F. C., S. J. Libby, N. A. Buchmeier, P. C. Loewen, J. Switala, J. Harwood, and D. G. Guiney. 1992. The alternative -factor KatF (RpoS) regulates Salmonella virulence. Proc. Natl. Acad. Sci. USA 89:11978-11982.
22. Favaro, R., and G. Deho. 2003. Polynucleotide phosphorylase-deficient mutants of Pseudomonas putida. J. Bacteriol. 185:5279-5286.
23. Fields, P. I., E. A. Groisman, and F. Heffron. 1989. A Salmonella locus that controls resistance to microbicidal proteins from phagocytic cells. Science 243:1059-1062.
24. Fierer, J., L. Eckmann, F. Fang, C. Pfeifer, B. B. Finlay, and D. G. Guiney. 1993. Expression of the Salmonella virulence plasmid gene spvB in cultured macrophages and non-phagocytic cells. Infect. Immun. 61:5231-5236.
25. Figueroa-Bossi, N., and L. Bossi. 1999. Inducible prophages contribute to Salmonella virulence in mice. Mol. Microbiol. 33:167-176.
26. Galán, J. E., and D. Zhou. 2000. Striking a balance: modulation of the actin cytoskeleton by Salmonella. Proc. Natl. Acad. Sci. USA 97:8754-8761.
27. Garvis, S. G., C. R. Beuzon, and D. W. Holden. 2001. A role for the PhoP/Q regulon in inhibition of fusion between lysosomes and Salmonella-containing vacuoles in macrophages. Cell. Microbiol. 3:731-744.
28. Gerschenson, L. E., and R. J. Rotello. 1992. Apoptosis: a different type of cell death. FASEB J. 6:2450-2455.
29. Goverde, R. L., J. H. Huisin't Veld, J. G. Kusters, and F. R. Mooi. 1998. The psychrotrophic bacterium Yersinia enterocolitica requires expression of pnp, the gene for polynucleotide phosphorylase, for growth at low temperature (5°C). Mol. Microbiol. 28:555-569.
30. Groisman, E. A. 2001. The pleiotropic two-component regulatory system PhoP-PhoQ. J. Bacteriol. 183:1835-1842.
31. Groisman, E. A., and H. Ochman. 1997. How Salmonella became a pathogen. Trends Microbiol. 5:343-349.
32. Gulig, P. A., H. Danbara, D. G. Guiney, A. J. Lax, F. Norel, and M. Rhen. 1993. Molecular analysis of spv virulence genes of the Salmonella virulence plasmid. Mol. Microbiol. 7:825-830.
33. Gulig, P. A., and T. J. Doyle. 1993. The Salmonella typhimurium virulence plasmid increases the growth rate of Salmonellae in mice. Infect. Immun. 61:504-511.
34. Hansen-Wester, I., and M. Hensel. 2001. Salmonella pathogenicity islands encoding type III secretion systems. Microbes Infect. 3:549-559.
35. Heiskanen, P., S. Taira, and M. Rhen. 1994. Role of rpoS in the regulation of spv virulence genes of Salmonella typhimurium. FEMS Microbiol. Lett. 123:125-130.
36. Hinton, J. C., I. Hautefort, S. Eriksson, A. Thompson, and M. Rhen. 2004. Benefits and pitfalls of using microarrays to monitor bacterial gene expression during infection. Curr. Opin. Microbiol. 7:277-282.
37. Hobbie, S., L. M. Chen, R. J. Davis, and J. E. Galán. 1997. Involvement of mitogen-activated protein kinase pathways in the nuclear responses and cytokine production induced by Salmonella typhimurium in cultured intestinal epithelial cells. J. Immunol. 159:5550-5559.
38. Hurme, R., K. D. Berndt, S. J. Normark, and M. Rhen. 1997. A proteinaceous gene regulatory thermometer in Salmonella. Cell 90:55-64.
39. Kim, C. C., and S. Falkow. 2004. Delineation of upstream signaling events in the Salmonella pathogenicity island 2 transcriptional activation pathway. J. Bacteriol. 186:4694-4704.
40. Knodler, L. A., J. Celli, W. D. Hardt, B. A. Vallance, C. Yip, and B. B. Finlay. 2002. Salmonella effectors within a single pathogenicity island are differentially expressed and translocated by separate type III secretion systems. Mol. Microbiol. 43:1089-1103.
41. Kowarz, L., C. Coynault, V. Robbe-Saule, and F. Norel. 1994. The Salmonella typhimurium katF (rpoS) gene: cloning, nucleotide sequence, and regulation of spvR and spvABCD virulence plasmid genes. J. Bacteriol. 176:6852-6860.
42. Kox, L. F., M. M. Wosten, and E. A. Groisman. 2000. A small protein that mediates the activation of a two-component system by another two-component system. EMBO J. 19:1861-1872.
43. Krause, M., C. Roudier, J. Fierer, J. Hardwood, and D. G. Guiney. 1991. Molecular analysis of the virulence locus of the Salmonella dublin plasmid pSDL2. Mol. Microbiol. 5:307-316.
44. Kudla, J., R. Hayes, and W. Gruissem. 1996. Polyadenylation accelerates degradation of chloroplast mRNA. EMBO J. 15:7137-7146.
45. Kurita, A., H. Gotoh, M. Eguchi, N. Okada, S. Matsumura, H. Mastui, H. Danbara, and Y. Kikuchi. 2003. Intracellular expression of the Salmonella virulence plasmid virulence protein SpvB causes apoptotic cell death in eukaryotic cells. Microb. Pathog. 35:43-48.
46. Lesnick, M. L., N. E. Reiner, J. U. Fierer, and D. G. Guiney. 2001. The Salmonella spvB virulence gene encodes an enzyme that ADP-ribosylates actin and destabilizes the cytoskeleton of eukaryotic cells. Mol. Microbiol. 39:1464-1470.
47. Leszczyniecka, M., R. DeSalle, D. C. Kang, and P. B. Fisher. 2004. The origin of polynucleotide phosphorylase domains. Mol. Phylogenet. Evol. 31:123-130.
48. Leszczyniecka, M., D. C. Kang, D. Sarkar, Z. Z. Su, M. Holmes, K. Valerie, and P. B. Fisher. 2002. Identification and cloning of human polynucleotide phosphorylase, hPNPase old-35, in the context of terminal differentiation and cellular senescence. Proc. Natl. Acad. Sci. USA 99:16636-16641.
49. Libby, S. J., M. Lesnick, P. Hasegawa, E. Weidenhammer, and D. G. Guiney. 2000. The Salmonella virulence plasmid spv genes are required for cytopathology in human monocyte-derived macrophages. Cell. Microbiol. 2:49-58.
50. Lundberg, U., U. Vinatzer, D. Berdnik, A. von Gabain, and M. Baccarini. 1999. Growth phase-regulated induction of Salmonella-induced macrophage apoptosis correlates with transient expression of SPI-1 genes. J. Bacteriol. 181:3433-3437.
51. Mangan, D. F., and S. M. Wahl. 1991. Differential regulation of human monocyte programmed cell death (apoptosis) by chemotactic factors and proinflammatory cytokines. J. Immunol. 147:3408-3412.
52. Marshall, D. G., B. J. Sheehan, and C. J. Dorman. 1999. A role for the leucine-responsive regulator protein and integration host factor in the regulation of the Salmonella plasmid virulence (spv) locus in Salmonella typhimurium. Mol. Microbiol. 34:134-145.
53. Mastroeni, P., and M. Sheppard. 2004. Salmonella infections in the mouse model: host resistance factors and in vivo dynamics of bacterial spread and distribution in the tissues. Microb. Infect. 6:398-405.
54. Matsui, H., A. Abe, S. Suzuki, M. Kijima, Y. Tamura, M. Nakamura, K. Kawahara, and H. Danbara. 1993. Molecular mechanism of the regulation of expression of plasmid-encoded mouse bacteremia (mba) genes in Salmonella serovar cholearesuis. Mol. Gen. Genet. 236:219-226.
55. McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L. Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S. Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan, H. Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston, and R. K. Wilson. 2001. Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413:852-856.
56. Meresse, S., K. E. Unsworth, A. Habermann, G. Griffiths, F. Fang, M. J. Martinez-Lorenzo, et al. 2001. Remodelling of the actin cytoskeleton is essential for replication of intravacuolar Salmonella. Cell. Microbiol. 3:567-577.
57. 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.
58. Miki, T., N. Okada, and H. Danbara. 2004. Two periplasmic disulfide oxidoreductases, DsbA and SrgA, target outer membrane protein SpiA, a component of the Salmonella pathogenicity island type III secretion system. J. Biol. Chem. 279:34631-34642.
59. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Press, Cold Spring Harbor, N.Y.
60. Miller, S. I., and J. J. Mekalanos. 1990. Constitutive expression of the phoP regulon attenuates Salmonella virulence and survival within macrophages. J. Bacteriol. 172:2485-2490.
61. Monack, D. M., D. M. Boule, and S. Falkow. 2004. Salmonella typhimurium persists within macrophages in the mesenteric lymph nodes of chronically infected Nramp+/+ mice and can be reactivated by IFN neutralization. J. Exp. Med. 199:231-241.
62. Monack, D. M., C. S. Detweiler, and S. Falkow. 2001. Salmonella pathogenicity island 2-dependent macrophage death is mediated in part by the host cysteine protease caspase-1. Cell. Microbiol. 3:825-837.
63. Monack, D. M., W. W. Navarre, and S. Falkow. 2001. Salmonella-induced macrophage death: the role of caspase-1 in death and inflammation. Microb. Infect. 3:1201-1212.
64. Monsieurs, P., S. de Keersmaecker, W. W. Navarre, M. W. Bader, F. De Smet, M. McLelland, F. C. Fang, B. De Moor, J. Vanderleyden, and K. Marchal. 2005. Comparison of the PhoPQ regulon in Escherichia coli and Salmonella typhimurium. J. Mol. Evol. 60:462-479.
65. O'Byrne, C. P., and C. J. Dorman. 1994. Transcription of the Salmonella typhimurium spv virulence locus is regulated negatively by the nucleoid-associated protein H-NS. FEMS Microbiol. Lett. 121:99-106.
66. O'Byrne, C. P., and C. J. Dorman. 1994. The spv virulence operon of Salmnella typhimurium LT-2 is negatively regulated by the cyclic AMP (cAMP)-cAMP receptor protein system. J. Bacteriol. 176:905-912.
67. Otto, H., D. Tezcan, R. Girish, F. Haag, M. Rhen, and F. Koch-Nolte. 2000. The spvB gene product of the Salmonella enterica virulence plasmid is a mono(ADP-ribosyl)transferase. Mol. Microbiol. 37:1106-1115.
68. Pullinger, G. D., G. D. Baird, G. M. Williamson, and A. J. Lax. 1989. Nucleotide sequence of a plasmid gene involved in the virulence of Salmonella. Nucleic Acids Res. 17:7983.
69. Py, B., C. F. Higgins, H. M. Krisch, and A. J. Caprousis. 1996. A DEAD-box RNA helicase in the Escherichia coli RNA degradosome. Nature 381:169-172.
70. Raupach, B., N. Kurth, K. Pfeffer, and S. H. E. Kaufmann. 2003. Salmonella typhimurium strains carrying independent mutations display similar virulence phenotypes yet are controlled by distinct host defense mechanisms. J. Immunol. 170:6133-6140.
71. Rhen, M., and C. J. Dorman. 2005. Hierarchial gene regulators adapt Salmonella enterica to its host milieus. Int. J. Med. Microbiol. 294:487-502.
72. Rhen, M., P. Riikonen, and S. Taira. 1993. Transcriptional regulation of Salmonella enterica virulence plasmid genes in cultured macrophages. Mol. Microbiol. 10:45-56.
73. Rhen, M., M. Virtanen, and P. H. Mkel. 1989. Localization by insertion mutagenesis of a virulence-associated region on the Salmonella typhimurium 96 kilobase pair plasmid. Microb. Pathog. 6:153-158.
74. Rosenzweig, J. A., G. Weltman, G. V. Plane, and K. Schesser. 2005. Modulation of Yersinia type three secretion system by the S1 domain of polynucleotide phosphorylase. J. Biol. Chem. 280:156-163.
75. Sarkar, D., M. Leszczyniecka, D. C. Kang, I. V. Lebedeva, K. Valerie, K., Dhar, S. Pandita, and P. B. Fisher. 2003. Down-regulation of Myc as a potential target for growth arrest induced by human polynucleotide phosphorylase (hPNPaseold-35) in human melanoma cells. J. Biol. Chem. 278:24542-24551.
76. Schmieger, H. 1972. Phage P22 mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75-78.
77. 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.
78. Shea, J. E., C. R. Beuzon, C. Gleeson, R. Mundy, and D. W. Holden. 1999. Influence of the Salmonella typhimurium pathogenicity island 2 type III secretion system on bacterial growth in the mouse. Infect. Immun. 67:213-219.
79. Symmons, M. F., G. H. Jones, and B. F. Luisi. 2000. A duplicated fold is the structural basis for polynucleotide phosphorylase catalytic activity, processivity, and regulation. Structure Fold. Des. 8:1215-1226.
80. Taira, S., P. Heiskanen, R. Hurme, H. Heikkil, P. Riikonen, and M. Rhen. 1995. Evidence for functional polymorphism of the spvR gene regulating virulence gene expression in Salmonella. Mol. Gen. Genet. 246:437-444.
81. Taira, S., P. Riikonen, H. Saarilahti, S. Sukupolvi, and M. Rhen. 1992. The mkaC virulence gene of the Salmonella serovar Typhimurium 96 kb plasmid encodes a transcriptional activator. Mol. Gen. Genet. 228:381-384.
82. Tedin, K., and U. Blasi. 1996. The RNA chain elongation rate of the lambda late mRNA is unaffected by high levels of ppGpp in the absence of amino acid starvation. J. Biol. Chem. 271:17675-17686.
83. Tezcan-Merdol, D., T. Nyman, U. Lindberg, F. Haag, F. Koch-Nolte, and M. Rhen. 2001. Actin is ADP-ribosylated by the Salmonella enterica virulence-associated protein SpvB. Mol. Microbiol. 39:606-619.
84. Wilson, J. A., T. J. Doyle, and P. A. Gulig. 1997. Exponential-phase expression of spvA of the Salmonella typhimurium virulence plasmid: induction in intracellular salts medium and intracellularly in mice and cultured mammalian cells. Microbiology 143:3827-3839.
85. Wilson, J. A., and P. A. Gulig. 1998. Regulation of the spvR gene of the Salmonella typhimurium virulence plasmid during exponential-phase growth in intracellular salts medium and a t stationary phase in L broth. Microbiology 144:1823-1833.
86. Yamanaka, K., and M. Inouye. 2001. Selective mRNA degradation by polynucleotide phosphorylase in cold shock adaptation in Escherichia coli. J. Bacteriol. 183:2808-2816.
87. Yehudai-Resheff, S., M. Hirsh, and G. Schuster. 2001. Polynucleotide phosphorylase functions as both an exonuclease and a poly(A) polymerase in spinach chloroplasts. Mol. Cell. Biol. 21:5408-5416.(Sofia Eriksson Ygberg, Ma)
Wolfson Institute for Biomedical Research, University College London, London WC1E 6BT, United Kingdom
Centre for Molecular Biology and Infection, Department of Infectious Diseases, Imperial College London, Armstrong Road, London SW7 2AZ, United Kingdom
Molecular Microbiology Group, Institute of Food Research, Norwich Research Park, Colney, Norwich, NR4 7UA, United Kingdom
ABSTRACT
Mutational inactivation of the cold-shock-associated exoribonuclease polynucleotide phosphorylase (PNPase; encoded by the pnp gene) in Salmonella enterica serovar Typhimurium was previously shown to enable the bacteria to cause chronic infection and to affect the bacterial replication in BALB/c mice (M. O. Clements et al., Proc. Natl. Acad. Sci. USA 99:8784-8789, 2002). Here, we report that PNPase deficiency results in increased expression of Salmonella plasmid virulence (spv) genes under in vitro growth conditions that allow induction of spv expression. Furthermore, whole-genome microarray-based transcriptome analyses of bacteria growing inside murine macrophage-like J774.A.1 cells revealed six genes as being significantly up-regulated in the PNPase-deficient background, which included spvABC, rtcB, entC, and STM2236. Mutational inactivation of the spvR regulator diminished the increased expression of spv observed in the pnp mutant background, implying that PNPase acts upstream of or at the level of SpvR. Finally, competition experiments revealed that the growth advantage of the pnp mutant in BALB/c mice was dependent on spvR as well. Combined, our results support the idea that in S. enterica PNPase, apart from being a regulator of the cold shock response, also functions in tuning the expression of virulence genes and bacterial fitness during infection.
INTRODUCTION
Members of the genus Salmonella have evolved through the acquisition of several genetic elements that enhance bacterial virulence and enable these bacteria to act as facultative intracellular pathogens (31, 34). These acquired genetic elements include the Salmonella pathogenicity islands (SPIs) (31, 34, 40), the spv virulence gene cluster (33), and selected prophages (25). In addition, Salmonella pathogenicity is dependent on a strict transcriptional regulation of horizontally acquired genetic elements (27, 30, 40, 60, 70, 71). This regulation is coordinated through the bacterial responses to environmental cues experienced during infection (30, 71) and includes the expression of SPI1 genes needed for bacterial invasion (12, 26, 34) and for the induction of proinflammatory responses (37) and the expression of spv and SPI2 genes needed for intracellular survival and replication (10, 30, 34, 56, 57, 71, 77). It has, however, remained enigmatic as to how the imported genetic elements became adapted to the repertoire of preexisting gene regulators. In part, this regulatory compatibility can be explained by the fact that the imported genetic elements rely to a degree on evolutionary conserved regulatory factors for their expression, such as two-component sensor regulatory systems, the RNA-polymerase factors, and nucleoid-associated proteins. These factors then act in concert with specific virulence-associated gene regulatory proteins to achieve the proper induction of virulence genes (71).
Polynucleotide phosphorylase (PNPase; encoded by the pnp gene) belongs to an expanding family of exoribonucleases (7) with homologues identified in eubacteria (22, 29, 79), Drosophila melanogaster (47), plants (44, 87), and even mice and humans (48, 75, 87). The Escherichia coli PNPase participates in RNA degradation (2) and plays a central role in adaptation to growth at low temperature (86). In this context, PNPase assists adaptation to the new environmental situation by specifically degrading mRNAs that code for cold shock proteins (CSPs), a process which is needed for the resumption of bacterial replication after shifting to decreased temperature (86). PNPase is also a crucial component of the RNA degradosome, the multiprotein complex that is responsible in part for mRNA degradation in E. coli (2, 69).
The facultative intracellular pathogen Salmonella enterica serovar Typhimurium causes a systemic infection in mice (8, 53). The murine infection resembles human typhoid fever and involves invasion of the intestinal epithelial cell barrier, subsequent visceral colonization, and replication in phagocytic cells (8, 26, 34, 53). S. enterica serovars Typhimurium and Typhi both contain a gene which is very similar to the E. coli pnp gene (11, 55), and an S. enterica serovar Typhimurium mutant deficient for PNPase exoribonuclease activity shows the expected defect in cold adaptation (11). When grown at 37°C in rich complex medium, the S. enterica serovar Typhimurium pnp mutation does not show alterations in mRNA levels of csp genes but instead expresses increased levels of other mRNA species, in particular those coded for by SPI1 and, to a lesser degree, SPI2 (11). Mutational inactivation of the S. enterica serovar Typhimurium PNPase also results in an altered infection pathogenesis; in contrast to acute systemic infections caused by wild-type (wt) S. enterica serovar Typhimurium, the pnp mutation gains the ability to establish a persistent infection in BALB/c mice (11).
The complexity by which environmental signals regulate the expression of virulence genes in Salmonella is well illustrated by the fact that the bulk of the essential virulence genes are not properly expressed during growth in ordinary complex laboratory medium (71). Therefore, to find out whether virulence genes apart from SPI1 and SPI2 would depend on PNPase for their expression, we have in this report studied the effect of PNPase deficiency on bacterial gene expression under in vitro conditions that simulate selected aspects of the intracellular environment and during actual bacterial replication inside murine macrophage-like J774-A.1 cells. The concomitant results reported here show that the expression of the spv operon is strongly increased in the absence of PNPase under conditions that allow the induction of spv gene expression. In the absence of the transcriptional activator gene spvR, however, no accumulation in spv gene expression was observed with PNPase deficiency, implying that the relaxed expression of spv genes resulting from PNPase deficiency did not relive spv expression from the gene regulator SpvR. Significantly, competition experiments with BALB/c mice showed that the apparent growth advantage of the pnp mutant depended on SpvR as well.
MATERIALS AND METHODS
Bacterial strains, growth media, and plasmids. Salmonella enterica serovar Typhimurium MC1 (SR-11 variant 3181) and the isogenic pnp mutant MC71 are described in Clements et al. (11). Strains were grown in Luria broth (LB) or on Luria agar with antibiotics as appropriate (ampicillin, 100 μg/ml; chloramphenicol, 10 μg/ml; kanamycin, 30 μg/ml; tetracycline, 10 μg/ml). Low-pH minimal medium (MM5.8), as adapted from Kox et al. (42), included 100 mM Bis/Tris (Sigma) buffer (pH 5.8), 0.1% (wt/vol) Casamino Acids, 0.16% (wt/vol) glycerol, and 10 μM MgCl2. Complete cell tissue culture medium (CC) consisted of RPMI 1640 medium (Gibco) supplemented with 10% (wt/vol) fetal bovine serum (Gibco), 2 mM L-glutamine (Sigma), and 20 mM HEPES (Sigma) as previously described (17). To generate phoP, rpoS, or spvR mutants of wt MC1 and the pnp mutant MC71 bacteria, the strains were transduced with P22 int phage lysates (76) containing phoP::Tn10 (23), spvR::Tn5 (72), or rpoS inactivated through insertion with the suicide vector pRR10 (21). A Tn10 tetracycline resistance marker, carried on the virulence plasmid pSLT and shown to be neutral in virulence assays applying BALB/c mice (4), was used to tag bacteria in competition experiments. Plasmid pFF-1 was obtained by cloning the PCR-amplified pnp gene into the T7 RNA-polymerase-dependent expression vector pET21(c) using the primers 5'CGGGATCCCGATGCGAGAAGATCGGGTATT3' and 5'CCGCTCGAGCGGCTCGGCCTGTTCGCTCGC3' with the BamHI and XhoI cloning sites, respectively, in boldface type (11). The fragment thus amplified included 17 bp upstream of the pnp initiation codon, as well as the last codon before the stop signal. In this construct, expression is driven from the plasmid-encoded start codon and ribosomal-binding site. The pET-based vector pHAM-1 encoding the catalytic carboxy-terminal portion of SpvB was available from previous work (67, 83).
Cell culture and infection models. The murine macrophage-like J774-A.1 cells (ATCC TIB67) were grown in RPMI 1640 medium (Gibco) supplemented with 10% (wt/vol) fetal bovine serum (Gibco), 2 mM L-glutamine (Sigma), and 20 mM HEPES (Sigma). To avoid the rapid cytotoxicity mediated by expression of invasion-associated SPI1 genes (9, 50, 63), the bacterial inocula were taken from bacteria grown on agar plates, a condition that does not induce SPI1 expression (16, 71), and opsonized with fresh mouse serum to facilitate bacterial uptake into J774-A.1 cells. To increase the uptake of salmonella, plates were centrifuged at 1,000 x g for 5 min, and this was defined as time zero hour. After 1 h of phagocytosis, extracellular bacteria were killed with 30-μg/ml gentamicin. This medium was replaced after 1 h with medium containing 5-μg/ml gentamicin. Incubations were continued for the time required for each time point tested. The relatively low levels of gentamicin were chosen to avoid any potential effects on intracellular bacterial gene expression by gentamicin (17). Bacteria were released from host cells by hypotonic lysis and enumerated by viable counting on Luria agar plates.
RNA extraction for microarrays. Total bacterial RNA for the microarray experiments was isolated from bacteria propagated in either of three separate growth media. First, for experiments referred to as MM5.8, bacteria were grown aerobically in MM5.8 until they reached optical density at 600 nm (OD600) of 0.5. Second, for CC samples, the bacteria were grown as described in Eriksson et al. (17). The bacterial RNA was in each experiment stabilized in 1% (wt/vol) acidic phenol and 19% (wt/vol) ethanol in water (17, 36, 82). Third, for extraction of bacterial RNA from intracellular bacteria, we refer to a recent description (17). Briefly, for each extraction of Salmonella RNA, a total of 108 J774-A.1 murine macrophage-like cells were seeded in 20 six-well cell culture plates and infected with mouse complement-opsonized S. enterica serovar Typhimurium MC1 or MC71 prepared as described in Eriksson et al. (17), at a multiplicity of infection (MOI) of 100:1 or 10:1 bacteria to cells (17). Eight hours postinfection, the macrophages were lysed for 30 min on ice in 0.1% (wt/vol) sodium dodecyl sulfate (SDS), 1% (wt/vol) acidic phenol, and 19% (wt/vol) ethanol in water, causing lysis of eukaryotic cells but not lysing the intracellular bacteria. The phenol-ethanol mixture acted to stabilize all bacterial RNA (36, 82). Bacteria were isolated by centrifugation (10,000 x g; 30 min) from the host cell lysate obtained from 120 wells of infected J774-A.1 murine macrophage-like cells and pooled. For all experiments, total RNA was prepared using the Promega SV total RNA purification kit. The quality of bacterial RNA samples was analyzed by size fractionation on a microfluid-based system where the fractionated RNA is detected by being stained with an intercalating fluorescent dye (2100 Bioanalyzer; Agilent). This allowed us to simultaneously define the RNA samples with respect to quantity and quality.
Microarray procedures. DNA microarray analysis of gene expression was performed essentially as previously described (17). Briefly, RNA was first reverse transcribed into cDNA and subsequently labeled by random priming with Cy5-dCTP (for labeling protocols, see www.ifr.bbsrc.ac.uk/Safety/Microarrays/#Protocols). Genomic DNA (gDNA) was fluorescently labeled with Cy3-dCTP and used as a reference channel in each experiment. The labeled cDNA and gDNA were mixed together, denatured, and hybridized to the microarray under standard glass coverslips (22 by 22 mm). Hybridization was performed at 65°C overnight, after which the slides were washed twice in 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 sodium citrate) at 65°C for 5 min, followed by washing twice, once with 1x SSC and once with 0.2x SSC for 5 min, each at room temperature. The washed slides were scanned on a GenePix 4000A scanner (Axon Instruments, Inc.). Fluorescent spots and local background intensities were quantified using Genepix Pro software (Axon Instruments, Inc.). All hybridizations were repeated three or four times. Spots showing a reference signal that was lower than the background plus 2 standard deviations or obvious blemishes were excluded from subsequent analyses. The local background was subtracted from spot signals, and fluorescence ratios were calculated. To compensate for unequal dye incorporation or any effect of the amount of template, data centering was performed by bringing the median natural logarithm of the ratios for each group of spots printed by the same pin to zero. The averages from the four data sets from each strain were calculated and compared to each other, and genes that had a >2-fold difference in expression were further analyzed. This new subset of data was analyzed by analysis of variance, and genes for which the difference in expression that had a P value of <0.005 were considered significant.
RNA blot analyses. For Northern blotting, bacteria were grown in MM5.8 to an OD600 of 0.3. Rifampin (200 μg/ml; Sigma) was added, and samples were taken immediately and at defined intervals (0, 2.5, 5, 10, and 15 min). Total RNA was prepared and analyzed for purity as above. Ten micrograms of RNA per lane was separated on denaturing agarose gels and finally transferred onto nylon sheets (Amersham) by capillary blotting. Hybridization conditions were applied as specified by the manufacturer of the membrane (Amersham). Briefly, the membrane was baked for 2 h at 80°C and then prehybridized in 50% (wt/vol) formamide, 4x SSC, 0.2% (wt/vol) SDS, 2.5x Denhardt solution, and 0.05-mg/ml sonicated herring sperm DNA at 42°C for 3 h. Internal fragments of the genes encoding spvR (positions 31630 to 32131, as annotated in the sequenced LT2 genome), spvA (30302 to 30810, as annotated in the sequenced LT2 genome), and 16S rRNA (positions 289339 to 289840, as annotated in the sequenced LT2 genome) were amplified by PCR, labeled with [-32P]dCTP (Amersham) using a random priming labeling kit (Roche), and used as probes. The synthesized probes were added to the prehybridization solution and incubated at constant agitation at 42°C overnight. The membrane was washed twice with 2x SSC and 0.1% (wt/vol) SDS and then twice for 20 min each in 0.1x SSC-0.1% (wt/vol) SDS. A radioisotope imaging system (Phosphoirmager 445SI; Molecular Dynamics) was used to detect radioactivity. Half-lives were calculated using the rate equation for a one-substrate enzyme-catalyzed reaction. For the dot blot experiment, bacteria were grown to OD600 of 0.6 in LB and then transferred to MM5.8 to mimic transition to an intracellular environment. Samples were taken at 1, 1.5, and 2 h and finally overnight, posttransition. Total RNA was prepared and analyzed as above. Twelve micrograms of total RNA per dot was applied onto nylon sheets (Amersham) and UV cross-linked. After this, the same procedure as for Northern blots was used.
Determination of -galactosidase activities. Enzyme activities expressed by lacZ transcriptional constructs were measured according to Miller (59) using 2-nitrophenyl--galactoside (Sigma) as a substrate. The spvR-lacZ (pHUB70), spvA-lacZ (pHUB61), and tlpA-lacZ(pOF1) constructs have been described previously (38, 72, 80). For bacteria grown in MM5.8, the samples used for -galactosidase activity measurements were neutralized with an equal volume of 1 M Tris-HCl buffer, pH 8.0, before addition to the reaction mixture. For comparing -galactosidase activities, we used an unpaired t test with Welch correction.
Immunostaining and immunofluorescence. Murine macrophage-like J774-A.1 cells were infected as described for the microarray experiment (using MOI as indicated) and subsequently fixed in a freshly prepared 4% (wt/vol) formaldehyde solution, prepared as described by Eriksson et al. (18). The fixed cells were subsequentially permeabilized in 0.1% (wt/vol) saponin and stained with antibodies as indicated in a solution with 10% (wt/vol) horse serum and 0.1% (wt/vol) saponin (3). Texas red-conjugated phalloidin was from Molecular Probes. For primary antibodies, we used a 1:200 dilution of the rabbit 156 anti-lysosome-associated membrane glycoprotein 1 (LAMP-1) (3) and a 1:200 dilution of goat anti-Salmonella (CSA-1) (3), while for secondary antibodies we used rhodamine red X-conjugated donkey anti-rabbit and Cy2-conjugated donkey anti-goat antibodies (both diluted 1:200). The samples were mounted using Aqua polymount (Polysciences) and analyzed by confocal microscopy (3).
Assessment of cytotoxicity. Three separate methods were used to score the viability of the macrophage-like J774A.1 cells during infection. (i) The proportion of attached to detached cells was estimated by phase-contrast microscopy. To document these observations, the cells were grown on glass coverslips, fixed, and stained for bacteria and actin after infection as specified above. Thereafter, the density of infected murine macrophage-like J774 cells was monitored by microscopical counting. (ii) As release of lactate dehydrogenase (LDH) is used as a defined measure of SPI1- and SPI2-induced induced cell death (14, 62, 63), we used a cytotoxicity detection kit (Roche) measuring LDH activity in the cell culture supernatants after given time points postinfection. The activities are given as the percentage of LDH activity released in relation to the LDH activity obtained by lysing a comparable amount of uninfected cells with Triton X-100. (iii) Apoptosis was monitored by measurement of nuclear fragmentation (28, 51). For these experiments, infected murine macrophage-like J774-A.1 cells were collected by centrifugation (800 x g; 1 min). This included both the attached cells and the potentially detached cells (in total, approximately 2 x 105 cells). The pelleted cells were washed in cold phosphate-buffered saline. The pellet was then rapidly dissolved in 200 μl of 10 mM Tris-HCl (pH 7.5), 150 mM NaCl, and 1% (wt/vol) Triton X-100 buffer. Thereafter, the suspension was separated into "Triton-insoluble" and "Triton-soluble" fractions by centrifugation (14,000 x g; 5 min). A small proportion of the supernatant was then immediately run on a 1% (wt/vol) agarose gel. Meanwhile, the pellet was solubilized into 100 μl of a 10 mM Tris-HCl-2 mM EDTA-1% (vol/vol) buffer (pH 7.5) and analyzed on a 1% agarose gel. Finally, the rest of the Triton-soluble material was extracted with phenol-chloroform and precipitated with ethanol by using glycogen (Roche) as a carrier. The material was dissolved into 10 μl distilled water, treated with RNase, and analyzed on 1% (wt/vol) agarose gels. Agarose gels included ethidium bromide (0.5 μg/ml; Sigma) to reveal nucleic acids.
Infection experiments. For in vivo infection experiments, we used S. enterica serovar Typhimurium strains MC1 and MC1 spvR mutant (11, 72) tagged with a Tn10 element on pSLT and previously shown to be neutral in mice infection experiments (4) and MC71 and MC71 spvR mutant strains. Infection experiments were carried out as previously described (11, 78). Briefly, female BALB/c mice, 6 to 8 weeks of age, were infected intraperitoneally with a 50:50 mixture of MC1-Tn10 and MC71 bacteria in a total dose of approximately 5 x 103 bacteria per mouse or with a 50:50 mixture of MC1-Tn10-spvR mutant strain and MC71 spvR mutant bacteria, also in a total dose of 5 x 103 bacteria per mouse. Groups of five mice were used for each of two series of experiments (10 mice per competition in total). Three days after infection, mice were sacrificed, and the livers and spleens were removed for enumeration of the total amount of viable bacteria, as well as the percentage of tetracycline-resistant bacteria. To compare the significance in ratios, we applied the two-sided Mann-Whitney U test.
RESULTS
Whole-genome scale probing for bacterial gene expression in vitro reveals a strong association between PNPase and spv gene expression. Our previous transcriptome analyses of Salmonella enterica serovar Typhimurium gene expression in vitro were carried out with bacterial cultures replicating in complex LB medium (11). While these analyses defined a strong association between SPI1 gene expression and PNPase, the growth conditions applied were not optimal for the expression of several virulence genes, such as SPI2 (39) or spv genes (17, 71). To further define the repertoire of virulence genes affected by PNPase, we extended the microarray-based transcriptome analysis to include bacteria grown in deprived medium with a low Mg2+ concentration and a pH of 5.8 (MM5.8). The composition of the medium was primarily designed to mimic the intravacuolar environment in which Salmonella replicates (17, 42) and is known to induce expression of SPI2 (78).
We first defined the transcriptome for the wild-type, PNPase-proficient S. enterica serovar Typhimurium strain MC1 grown in MM5.8 medium. Samples for RNA preparation were taken when the culture entered the transition into the stationary phase (OD600, 0.5). To probe for alterations in gene expression, we used cultures grown in CC as a comparator and applied a false discovery rate of 0.005. Under these conditions, we defined 447 genes as up-regulated and 479 genes as down-regulated during replication in MM5.8 (see Table S1 in the supplemental material). Among the genes strongly induced for expression, we identified representatives of the SPI2 and the spv gene clusters, whereas among the genes strongly suppressed we identified SPI1 and genes for flagellation and motility (see Table S1 in the supplemental material).
Having shown that MM5.8 did induce expression of genes that are normally induced when bacteria are residing in an intracellular compartment (17), we next tested whether PNPase affected the mRNA levels of any of these genes. To probe for the effect of PNPase deficiency, we applied S. enterica serovar Typhimurium strain MC71, which was constructed from MC1 by introducing a STOP codon in pnp by site-directed mutagenesis (11). When MC71 was grown in MM5.8 to the transition to the stationary phase of growth (OD600, 0.5), MC71 showed a relative increase in expression of 37 genes and a relative decrease in the expression of 9 genes compared to wt MC1 (Fig. 1). More specifically, the comparison revealed a strong (7- to 8-fold) relative increase in spv expression in MC71, in addition to an increase (2- to 10-fold) in mRNA levels of genes located in prophages (STM0906, STM0908, STM0914, STM2233 to STM2237, STM2587, STM2616, and STM2620) and of genes encoding proteins involved in stress responses (STM1251 and ibpAB) and DNA repair (mutM and recF). In parallel, we found a PNPase-dependent decrease (two- to threefold) in the expression of genes associated with purine synthesis (carA, purK, trpCD, and purF) and of genes involved in flagellar synthesis (flgBC) (Fig. 1).
mRNA analyses reveal increased levels of spv gene mRNA at PNPase deficiency in vitro. The spv gene cluster is transcribed from two main promoters, one in front of the transcriptional activator protein gene spvR (PspvR) and another in front of spvA (PspvA) directing the expression of Spv effector proteins (13, 43, 52, 81). To verify the association between PNPase and spv expression, we next determined spvA expression by comparing spvA mRNA levels in the wt MC1 PNPase-proficient and the MC71 PNPase-deficient mutant background by carrying out Northern blotting to probe for spvA mRNA during logarithmic growth (OD600, 0.3) in MM5.8 after blocking RNA synthesis with rifampin. This analysis clearly showed expression of spvA mRNA in both MC1 and MC71. The analysis also revealed an overall higher spvA mRNA content and an alternative spvA mRNA pattern in MC71 (Fig. 2A); whereas both MC1 and MC71 showed spvA mRNA of apparently higher molecular weight named A1, MC71 revealed clear accumulation of a second spvA mRNA named A2 that was only weakly detected from MC1 (Fig. 2A and B). When using rifampin to block the mergenc of nascent mRNA species, we could not demonstrate any stabilization of the spvA1 mRNA in MC71 in either of two separate sets of experiments (Fig. 2A and data not shown). In contrast, the spvA1 mRNA appeared more stable in MC1 (Fig. 2A and B), suggesting that this transcript was differentially degraded by PNPase.
Parallel Northern analyses using a 32P-labeled PCR-generated spvR fragment revealed a clear signal for spvR in MC71 but no signal in MC1 (data not shown). This is consistent with the previously reported difficulty in detecting the spvR transcript in S. enterica (1). Nevertheless, these analyses confirmed the increased spv mRNA levels in MC71 implied by microarray analyses.
Transcriptional spv-lacZ fusions reveal PNPase-dependent expression of spv genes. The fact that we could not conclusively demonstrate any increase in the half-life of the spvA mRNA in the PNPase-deficient strain MC71 (Fig. 2A and B) prompted us to test whether an increased transcriptional strength of spv promoters could account for the increased spvA mRNA levels observed with the PNPase-deficient background. The constructs used included the spvA (pHUB61) or spvR (pHUB70) promoter positioned to direct lacZ expression in a pACYC184-based construct (Fig. 3A) (72). As the expression of spvA-lacZ fusions tends to be weak in rich growth medium (84) and indeed confirmed for pHUB61 grown in CC (data not shown), we initiated a series of experiments to compare the expression of -galactosidase from spv-lacZ fusions when the bacteria were grown overnight in MM5.8 in batch cultures (15 to 17 h). Regarding the spvA-lacZ construct pHUB61, the -galactosidase levels obtained when the bacteria were grown overnight in MM5.8 were significantly higher for MC71 than for PNPase-proficient MC1 (Fig. 3B). Likewise, in comparison to MC1/pHUB70, we noted a significant increase in the -galactosidase levels expressed by MC71/pHUB70 grown in MM5.8 (Fig. 3C).
To ensure that the alterations in -galactosidase activities in the PNPase-deficient MC71 background were not due to stabilization of lacZ mRNA or to alterations in the copy number of the operon fusion constructs in MC71, we applied the tlpA-lacZ fusion pOF1 as a control (38). Like the spv genes, tlpA is contained on the virulence plasmid pSLT but was not induced during bacterial replication in MM5.8 (see Table S1 in the supplemental material). Furthermore, pOF1 relied on the same lacZ cartridge as pHUB61 and pHUB70 and cloned with tlpA in the same position in the same cloning vector. The tlpA-lacZ fusion did not show increased but rather a slightly decreased level of activity in the pnp mutant MC71 background when grown in MM5.8 (Fig. 3D), which indeed suggests that the increased -galactosidase activities obtained with spvA-lacZ and spvR-lacZ in MC71 originated from increased transcriptional activity.
The increased expression of spvA in overnight cultures of PNPase-deficient MC71 was also verified by blotting for spvA mRNA. For this experiment, PNPase-proficient MC1 and MC71 were grown in MM5.8, and the bacterial RNA was extracted at early (1 h, 1.5 h, and 2 h) and late (overnight) time points of growth. Dot blot assays applying the PCR-amplified and 32P-labeled spvA gene as a probe revealed no significant signal for spvA mRNA during the very early time points in either MC1 or MC71 (Fig. 2C). Also, we could not detect any significant accumulation of spvA mRNA in the overnight cultures of MC1. This would be consistent with the observation that spvA mRNA becomes induced during logarithmic growth in minimal medium low in magnesium, to become only weakly detectable after prolonged incubation (84). In contrast and in accordance with the LacZ measurements from MC71/pHUB61, PNPase-deficient strain MC71 showed a substantial accumulation of spvA mRNA in MC71 after overnight incubation (Fig. 2C).
We next tested if complementation of the PNPase deficiency in MC71 could restore the spvA-lacZ activity to wild-type levels. Initial attempts using cloned pnp to complement for the pnp mutation in trans were not successful, due to our inability to maintain pnp-containing plasmid constructs in S. enterica serovar Typhimurium. However, when the pnp open reading frame was contained in the pET21(c) vector under the T7 promoter (pFF-1), the constructs could be maintained. The pFF-1 construct caused a reduction in the spvA-lacZ activities in the MC71 pnp mutant background but did not affect lacZ fusion expression in the MC1 wt background (Fig. 3E). A control construct containing the 3'-terminal portion of spvB in the pET32 vector (pHAM-1) (67, 83) did not mediate complementation (Fig. 3E). These observations implied that the T7 promoter shows weak expression in S. enterica serovar Typhimurium, even though the bacterium formally lacks the T7 RNA polymerase and the PNPase deficiency of MC71 can be complemented in trans.
PNPase deficiency does not relieve spvA gene expression from dependency on growth phase, PhoP, RpoS, or SpvR. Expression of spv-lacZ fusions in bacteria grown in low-magnesium minimal medium is reported to be dependent on the bacterial growth phase, the transcription factor SpvR, and the alternative -factor RpoS (84, 85). Furthermore, the conditions prevailing in MM5.8 are likely to activate the PhoP/PhoQ response regulatory system (42), also implicated in spv gene expression (30). Therefore, we set out to test whether the -galactosidase activity expressed by the spvA-lacZ constructs in the PNPase-deficient background MC71 grown in MM5.8 still depended on the bacterial growth phase and the transcription factors PhoP, RpoS, and SpvR.
The growth slopes in MM5.8 appeared similar for both PNPase-proficient MC1 and PNPase-deficient MC71 (Fig. 3F). However, the concomitant expression of -galactosidase became much stronger in MC71/spvA-lacZ when the cultures entered late logarithmic phase of growth (Fig. 3G). In parallel, the spvR-lacZ fusion construct pHUB70 was observed to express a similar growth-phase-dependent induction of -galactosidase activity when the bacteria were grown in MM5.8, albeit the difference in -galactosidase activity between MC1 and MC71 remained lower (data not shown).
To test to what extent the increase in spvA expression that associated with PNPase deficiency during propagation in MM5.8 still depended on the transcription factors PhoP, RpoS, or SpvR implicated in spv gene activation (5, 20, 21, 30, 35, 43, 68, 71, 84, 85), we measured -galactosidase activities encoded by pHUB61 (spvA-lacZ) in phoP, rpoS, and spvR mutants of MC1 and MC71. Inactivation either of phoP, rpoS, or spvR in PNPase-proficient MC1 or PNPase-deficient MC71 resulted in a significant reduction of spvA-lacZ expression in overnight cultures grown in MM5.8 (Fig. 3H). Thus, PNPase deficiency did affect the transcriptional strengths from both the spvA and spvR promoters, yet PNPase deficiency apparently did not evoke or apply any alternative bypass routes for spvA gene activation.
PNPase-proficient and PNPase-deficient bacteria do not differ in their ability to induce cytotoxicity in J774-A.1 cells 8 h postinfection. We have recently developed a method that enables transcriptometric analysis of bacteria growing within host cells (17, 36). However, to generate enough bacterial RNA for microarray analyses, we needed to apply an MOI of 100:1 (17), potentially generating infection-associated cytotoxicity and apoptosis (14, 45, 46, 49, 62, 63). Therefore, before defining and comparing the transcriptomes of PNPase-proficient MC1 and PNPase-deficient MC71 while growth occurred in host cells, we analyzed the murine macrophage-like J774-A.1 cells infected with S. enterica serovar Typhimurium MC1 and MC71 for possible alterations in toxicity or bacterial growth.
Upon infection of J774-A.1 macrophage-like cells with PNPase-proficient MC1 or PNPase-deficient MC71 at an MOI of 100:1, no signs of cytotoxicity or detachment of murine macrophage-like J774-A.1 cells as judged by phase-contrast microscopy were observed at 8 h postinfection (data not shown). Likewise, as followed by immunofluorescence microscopy labeling for Salmonella bacteria and filamentous actin at 10 h postinfection, the density and appearance of the host cells on glass slides were very similar whether the cells were infected with MC1 or MC71 bacteria (Fig. 4A and B). When the infected murine macrophage-like J774-A.1 cells were monitored for the release of lactate dehydrogenase (LDH) as a general indicator of cytotoxicity, the levels of released LDH observed at 8 h postchallenge by application of an MOI of 100:1 or 18 h postchallenge by application of an MOI of 10:1 remained low (Fig. 4C). The LDH levels released by cells infected with MC71 at an MOI of 100:1 were slightly but statistically significantly higher than those released upon infection with MC1. Yet, the LDH levels remained very low compared to the levels reported with invasive SPI1-expressing bacteria (14, 63) and to the levels contained in a comparable number of uninfected cells (Fig. 4C) (63). Similarly, when analyzed on ethidium bromide-stained agarose gels, no detectable DNA fragmentation was apparent at 8 h postinfection when an MOI of 100:1 of noninvasive complement-opsonized bacteria was applied (Fig. 4D). However, if the uninfected cells were allowed to overgrow the cell culture, leading to acidified and crowded cell conditions, DNA fragmentation was indeed detected (Fig. 4D).
In separate infection experiments, immunofluorescence microscopy labeling for intracellular bacteria revealed PNPase-proficient MC1 and PNPase-deficient MC71 bacteria residing in an LAMP-1-positive compartment, with no differences observed in the labeling pattern for the two strains (Fig. 4E to J). Viable counts, performed in parallel from nonfixed and unstained cells 8 h postinfection, revealed a 5.4-fold and 5.0-fold growth index, respectively, for the MC1 wt and MC71 pnp mutation, when the growth yields were compared to intracellular bacterial counts at 2 h postinfection. A repetition of the growth experiment did not provide any apparent alterations in the growth yields (data not shown).
In summary, the host cells appeared intact 8 h postinfection, and no significant differences in cytotoxicity, intracellular localization, or bacterial growth yields were found between cells infected with either PNPase-proficient or PNPase-deficient S. enterica serovar Typhimurium strains.
Intracellular transcriptome of S. enterica serovar Typhimurium SR-11. Having established the infection procedures, we next determined the transcriptome of the PNPase-proficient MC1 strain when replicating in the macrophage-like cells. For these experiments, bacterial RNA was extracted at 8 h postinfection was compared to RNA extracted from CC-grown bacteria. This time point was previously defined to reveal the majority of the gene expression profile alterations upon infection of murine macrophage-like J774-A.1 cells (17). Since we previously determined the transcriptome of S. enterica serovar Typhimurium SL1344 when replication occurs within host cells, we first compared the transcriptomes of intracellular S. enterica serovar Typhimurium of lines SR-11 and SL1344 to define to what extent the pattern of gene expression was dependent on the bacterial strain background. The gene expression profiles of the two lines were consistent: 94.4% of the genes revealed similar expression patterns in both strains after application of a twofold cutoff (false-discovery rate = 0.005) (see Table S1 in the supplemental material) (17). A proportion of the expressional difference reflected the histidine auxotrophy of strain SL1344. Nevertheless, these observations show that the transcriptomes of the two different strains of S. enterica serovar Typhimurium were comparable when the bacteria were replicating within murine macrophage-like J774-A.1 cells.
We then evaluated to what extent the intracellular transcriptome reflected the one obtained when bacteria were replicating in MM5.8. The comparison showed that 42% of the MC1 genes that were up-regulated in murine macrophage-like J774-A.1 cells were also more highly expressed in MM5.8 (see Table S1 in the supplemental material). It is noteworthy that virulence genes within SPI2 (ssaG), SPI3 (mgtB), and the spv locus (spvA) and a gene within SPI5 (pipD) that functionally associates with SPI2 were clearly represented within this category of genes (see Table S1 in the supplemental material). Almost one-third (30.5%) of MC1 genes that were down-regulated in murine macrophage-like J774-A.1 cells were also expressed at lower levels in MM5.8 (see Table S1 in the supplemental material), including genes within SPI1 (hilC) and SPI4 (STM4260) and motility-related (fliC) genes, as well as a gene within SPI5 (pipC) that associates with SPI1 gene expression (see Table S1 in the supplemental material) (40). Thus, while growth in vitro in MM5.8 obviously did not reproduce all aspects of the SCV, the large proportion of genes that showed parallel trends of expression in MM5.8 and during infection of murine macrophage-like J774-A.1 cells confirms that MM5.8 indeed does represent certain crucial aspects of the intravacuolar compartment.
PNPase deficiency affects spv expression not only in vitro but also in murine macrophage-like J774-A.1 cells. Having defined the intracellular transcriptomes for PNPase-proficient S. enterica serovar Typhimurium MC1, we compared the transcriptomes of intracellular MC1 and PNPase-deficient MC71. The gene expression profiles obtained at 8 h postuptake were found to reveal an extremely high degree of similarity. Only six genes (spvABC, entC, rtcB, and STM2236) were up-regulated, whereas the motility-associated cheAW and motB genes were down-regulated in intracellular MC71 compared to intracellular MC1 bacteria (Fig. 5A). These data showed that PNPase specifically affected the expression of a small set of genes during macrophage infection.
We next assessed to what extent inactivation of spvR, the gene encoding a key activator of spvABCD transcription (Fig. 3H) (71, 81), influenced the intracellular transcriptome. As measured by microarray analysis, expression of spvAB and to some extent spvC became significantly reduced in both PNPse-proficient MC1 and PNPase-deficient MC71 upon mutational inactivation of spvR (Fig. 5B). More significantly, the differences observed for spv gene expression between intracellular MC1 and MC71 were not observed in the spvR mutant backgrounds. Surprisingly, inactivation of spvR also affected the increased mRNA levels observed for entC in the PNPase-deficient MC71 (Fig. 5B). This shows that much of the differences observed in the intracellular transcriptomes for PNPase-proficient MC1 and PNPase-deficient MC71 depended on SpvR.
PNPase deficiency associates with a spvR-dependent alteration in virulence in BALB/c mice. Apart from affecting virulence gene expression, PNPase deficiency was previously reported to affect infection pathogenesis and replication of the bacteria in BALB/c mice (11). Here, we followed up this phenomenon by performing in vivo competition experiments between PNPase-proficient MC1 and PNPase-deficient MC71 bacteria in BALB/c mice after intraperitoneal infection. Upon recovery of bacteria from the livers and spleens 3 days postinfection, the MC71 bacteria showed a higher competitive index in both the livers and spleens in two separate sets of experiments (the combined data are shown Fig. 6). As PNPase deficiency was found associated with increased spv gene expression in murine macrophage-like J774-A.1 cells (Fig. 5A) and as the spv genes participate in bacterial replication during the systemic phase of infection (32, 33, 73), we next tested the effect of inactivating spvR on the competitive indexes in mice. Thus, spvR mutants of MC1 and MC71 were competed against each other in BALB/c mice after intraperitoneal challenge. Astonishingly, under these premises MC71 lost its competitive advantage in mice in each of two separate sets of experiments, and in fact the competitive index for MC71 decreased (the combined data are shown in Fig. 6). Therefore, a noticeable proportion of the in vivo alteration in competitive index conferred by PNPase deficiency was dependent on alterations in spvR and SpvR-regulated gene expression.
DISCUSSION
For many serovariants of Salmonella enterica, virulence in the mouse model for systemic salmonellosis is dependent on the bacterial spv gene cluster that promotes bacterial replication in the liver and spleen, as the bacteria reside in macrophages (32, 33). While the exact details by which Spv proteins participate in the disease process are not known, the molecular function has been determined for two of the five Spv proteins: SpvB is an actin-specific mono(ADP-ribosyl)transferase (46, 49, 67, 83), whereas SpvR acts as transcriptional activator for spv expression (5, 43, 54, 68, 80, 81, 84, 85). However, the genetic control of spv gene expression is complex and requires selected environmental cues, as well as the interplay between several transcriptional regulators (71). Expression of spv genes is clearly induced when the bacteria reside within mammalian cells (17, 24, 72) and in vitro when the bacteria are grown in minimal medium mimicking the intravacuolar environment (84, 85). Expression of spv genes can also be achieved, at least to some extent, as the bacteria enter the stationary phase of growth in rich medium (13, 15, 21, 80, 84, 85). Apart from SpvR, the regulatory factors that participate in spv gene induction include the response regulator PhoP (30) the alternative -factor RpoS (15, 21, 41, 71), and integration host factor (52). In addition, the global gene regulator H-NS (65, 71), leucine-responsive regulatory protein (52), and the catabolite repression system (66) have been identified as repressors of spv gene expression.
We have previously demonstrated that mutational inactivation of the pnp gene encoding the exoribonuclease PNPase in Salmonella enterica serovar Typhimurium resulted in apparent increased bacterial replication in the host and in the establishment of chronic carrier states among convalescent mice (11). In parallel, when the bacteria were grown in vitro in rich growth medium, the PNPase-deficient mutant revealed stronger expression of invasion-associated SPI1 genes and to some extent increased expression of SPI2 genes (11), implying that PNPase-activity somehow connected with virulence gene expression. A recent investigation demonstrated that persistent salmonellosis in the mouse involves prolonged bacterial colonization of macrophages (61). Therefore, in the current study we probed the effect of PNPase on S. enterica serovar Typhimurium gene expression while the bacteria were replicating in an in vitro growth medium that partially mimicked the intravacuolar environment (MM5.8) or within murine macrophage-like J774-A.1 cells.
When whole-genome-based microarray technology was applied to define different bacterial transcriptomes, growth in MM5.8 was found to result in decreased expression of SPI1 genes and in a strong increase in the SPI2 and spv gene expression in both the PNPase-proficient MC1 and PNPase-deficient MC71 backgrounds. Somewhat surprisingly, PNPase deficiency associated only with a fairly moderate alteration in the transcriptome when the bacteria were grown in MM5.8. The alteration recorded in the PNPase-deficient MC71 background included genes associated with stress responses, DNA repair, and, significantly, the spv genes (Fig. 1; see Table S1 in the supplemental material).
Yet while Northern blot analyses clearly confirmed an accumulation of spvA mRNA in the PNPase-deficient MC71 background grown in MM5.8, we could not demonstrate any increase in the half-life of the spvA mRNA in MC71 (Fig. 2A and B). In parallel, lacZ fusions probing the transcription of the spvA and spvR promoters also revealed higher -galactosidase levels in the PNPase-deficient MC71 background (Fig. 3B and C). These observations suggest that the increased levels of spvA expression observed at PNPase deficiency, at least partially, could originate from an increased transcription from the spvA promoter. Still, the expression of spvA-lacZ remained dependent on growth phase and on the transcriptional regulators PhoP, RpoS, and SpvR, even in the PNPase-deficient MC71 strain (Fig. 3F to H). This would be in accordance with the previously defined strict dependency of the spvA promoter on RpoS and SpvR (35, 84, 85). Thus, one could envision a model in which spvA mRNA is not directly affected by PNPase. Instead PNPase could, by affecting the amount or activity of accessory transcription factors or regulatory RNAs, control spv expression indirectly. However, the array data did not reveal any significant increase in the mRNAs in MC71 for the activator genes phoP or rpoS or any significant decrease in the mRNAs for H-NS or the leucine-responsive regulatory protein (see Table S1 in the supplemental material). Yet, when we compared the genes altered for their expression at shift to MM5.8, we did find parallels to the PhoPQ-regulon recently revealed for S. enterica serovar Typhimurium grown in magnesium-deprived M9 minimal medium (64). The transcriptomes of MC1 and MC71 grown in MM5.8 showed induction in the expression of the PhoPQ-connected bioA, cysCDJN, mig-3, pagOC, oat, udg, ybaY, and ygaY genes and a repression in the expression of the PhoPQ-connected nrdFH, slyB, and yfbE genes, compared to the transcriptomes expressed in CC (see Table S1 in the supplemental material). Furthermore, it is interesting to note that the promoter region of spvA in particular does share sequences reminiscent of the PhoP-binding site (38). It thus remains quite possible that MM5.8 provokes activation of the PhoPQ regulon, while it may not activate phoPQ expression in itself. Nevetheless, the growth medium applied by Monsieurs et al. (64) to provoke the PhoPQ regulon differs in many respects from MM5.8, which might explain differences in the transcriptome profiles observed.
The transition from CC into murine macrophage-like J774-A.1 cells is known to have a drastic effect on the bacterial transcriptome, and this alteration also includes induction of spv gene expression (17). Therefore, we used the microarray analysis to define differences in the intracellular transcriptomes displayed by the PNPase-proficient MC1 and PNPase-deficient MC71 strains while the bacteria were grown in murine macrophage-like J774-A.1 cells. The comparison of the intracellular transcriptomes of MC1 and MC71 revealed a significantly different expression of only nine genes and included spvA, spvB, and spvC as being up-regulated in MC71 (Fig. 5A). The expression of spvA, spvB, and spvC genes was reduced to equal levels in spvR mutant derivatives of MC1 and MC71 grown in murine macrophage-like J774-A.1 cells (Fig. 5B). Somewhat unexpectedly, introducing the spvR mutation into MC71 also reduced the increase in entC expression conferred by PNPase deficiency. Thus, it remains possible that expression of the isochorismate synthetase EntC is affected by the expression of spv genes. As entC expression was grossly unaffected by spvR in the PNPase-proficient MC1 strain, it is unlikely that entC would directly rely on SpvR for its expression. However, we cannot exclude the possibility that entC expression would respond to an alternative PNPase-suppressed and SpvR-dependent transcriptional activation.
Mutational inactivation of PNPase was originally reported to affect not only SPI1 and SPI2 expression, but also infection pathogenesis in BALB/c mice (11). The pnp mutant MC71 displayed higher bacteria doses in the spleens of the infected mice at early time points after infection; the mutant bacteria showed increased rates of replication in mice and finally were capable of establishing persistent infections (11). As PNPase deficiency clearly caused an increase in spv gene expression when the bacteria were growing in MM5.8 or in murine macrophage-like J774-A.1 cells, we probed the effect of inactivating the spvR gene on bacterial replication in vivo. This was done by comparing the competitive indexes of the pnp mutant when competed against a PNPase-proficient strain in BALB/c mice and by competing corresponding spvR mutants. The results obtained (Fig. 6) clearly showed that the growth advantage conferred in mice by PNPase deficiency depended on spv gene expression. Thus, PNPase seems not only to suppress spv gene expression under conditions that induce spv gene expression, but also to suppress bacterial replication in BALB/c mice.
In many respects, the pattern of regulation of PNPase-dependent expression of virulence in S. enterica genes resembles that of the PNPase-directed cold shock response in E. coli; spv genes, as well as csp genes, are induced by environmental changes. Both classes of genes are needed to adapt the bacteria to new environmental demands, and the mRNA levels for all of these genes are modulated by PNPase. In E. coli, PNPase is not required for the induction of the cold shock response itself but for degradation of the csp mRNAs during adaptation to decreased growth temperature (86). A parallel scenario could be envisaged for the interplay between PNPase and virulence gene expression in Salmonella. According to the gene expression profile data presented here, PNPase is not necessary for the induction of spv virulence gene expression but may function to restrict gene expression once the genes have been induced. Nevertheless, the effect of PNPase on the expression of a given virulence gene may not necessarily reflect a PNPase-mediated degradation of the given mRNA. Instead, PNPase could affect gene expression indirectly through affecting the stability of the mRNA encoding a gene regulatory factor, the stability of regulatory RNA molecules, or the regulation of a factor otherwise needed for proper SPI expression (58). Any such scenario could well explain the increased spvA expression observed here for the PNPase-deficient strain MC71.
Recent studies of Yersinia virulence suggest yet another mechanism for PNPase-mediated regulation of virulence expression. As for Salmonella, the virulence of Yerinia relies on the expression of type III secretion (6). The effector proteins secreted by the Yersina type III secretion system prevent phagocytosis and expression of the oxidative burst (6, 74). Inactivation of Yersinia PNPase caused a delay in the secretion of such virulence proteins to the extent that the mutants expressed decreased survival in murine macrophage-like RAW cells (74). The defect in the expression did not rely on the level of gene expression but rather occurred at the posttranslational level. In parallel, another component of the RNA degradosome, the endoribonuclease RNaseE, has also been implicated in the expression of hilA, a main transcriptional regulatory gene of SPI1 (19).
We propose that PNPase can fine-tune the expression of environmentally induced genes, such as cold-shock or virulence genes, and allow bacteria to successfully adapt to alternating biological niches. Such observations suggest that imported genetic virulence elements also could be subjected to posttranscriptional regulation and that their regulation has become integrated not only into the network of existing gene regulatory proteins, but also to the activity of preexisting ribonucleases.
ACKNOWLEDGMENTS
We are grateful for support from VR Cancer fonden and STINT (M.R.), for a Core Strategic grant from the BBSRC, and for an EU Marie Curie Training site grant (J.C.D.H.). Work in D.W.H.'s laboratory was supported by grants from the Medical Research Council (United Kingdom) and the Wellcome Trust.
Supplemental material for this article may be found at http://iai.asm.org/.
REFERENCES
1. Abe, A., and K. Kawahara. 1995. Transcriptional regulation and promoter sequence of the spvR gene of virulence plasmid pKDSC50 in Salmonella choleraesuis serovar Choleraesuis. FEMS Microbiol. Lett. 129:225-230.
2. Bernstein, J. A., P. H. Lin, S. N. Cohen, and S. Lin-Chao. 2004. Global analysis of Escherichia coli RNA degradosome function using DNA microarrays. Proc. Natl. Acad. Sci. USA 101:2758-2763.
3. 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.
4. Bjrkman, J., M. Rhen, and D. I. Andersson. 1996. Salmonella typhimurium cob mutants are not hypervirulent. FEMS Microbiol. Lett. 139:121-126.
5. Caldwell, A. L., and P. A. Gulig. 1991. The Salmonella typhimurium virulence plasmid encodes a positive regulator of a plasmid-encoded virulence gene. J. Bacteriol. 173:7176-7185.
6. Carniel, E. 2002. Plasmids and pathogenicity islands in Yersinia. Curr. Top. Microbiol. Immunol. 264:89-108.
7. Carpousis, A. J. 2002. The Escherichia coli RNA degradosome: structure, function and relationship in other ribonucleolytic multienzyme complexes. Biochem. Soc. Trans. 30:150-155.
8. Carter, P. B., and F. M. Collins. 1974. The route of enteric infection in normal mice. J. Exp. Med. 139:1189-1203.
9. Chen, L. M., K. Kaniga, and J. E. Galán. 1996. Salmonella spp. are cytotoxic for cultured macrophages. Mol. Microbiol. 21:1101-1115.
10. Clements, M., S. Eriksson, D. Tezcan-Merdol, J. C. Hinton, and M. Rhen. 2001. Virulence gene regulation in Salmonella enterica. Ann. Med. 33:178-185.
11. Clements, M. O., S. Eriksson, A. Thompson, S. Lucchini, J. C. Hinton, S. Normark, and M. Rhen. 2002. Polynucleotide phosphorylase is a global regulator of virulence and persistency in Salmonella enterica. Proc. Natl. Acad. Sci. USA 99:8784-8789.
12. Collazo, C. M., and J. E. Galán. 1997. The invasion-associated type III protein secretion system in Salmonella—a review. Gene 192:51-59.
13. Coynault, C., V. Robbe-Saule, M. Y. Popoff, and F. Norel. 1992. Growth phase and SpvR regulation of Salmonella typhimurium spvABC virulence genes. Microb. Pathog. 13:133-145.
14. Decktrach, D., L. A. Knodler, K. Galbraith, and O. Steele-Mortimer. 2005. The Salmonella SPI1 effector SopB stimulates nitric oxide production long after invasion. Cell. Microbiol. 7:105-113.
15. El-Gedaily, A., G. Paesold, C.-Y. Chen, G. D. Guiney, and M. Krause. 1997. Plasmid virulence gene expression induced by short chain fatty acids in Salmonella dublin: identification of rpoS-dependent and rpoS-independent mechanisms. J. Bacteriol. 179:1409-1412.
16. Eriksson, S., M. Clements, and M. Rhen. 2001. Salmonella as a tenant in phagocytic cells. Res. Adv. Microbiol. 2:87-102.
17. Eriksson, S., S. Lucchini, A. Thompson, M. Rhen, and J. C. Hinton. 2003. Unravelling the biology of macrophage infection by gene expression profiling of intracellular Salmonella enterica. Mol. Microbiol. 47:103-118.
18. Eriksson, S., B. J. Chambers, and M. Rhen. 2003. Nitric oxide production by murine dendritic cells is cytotoxic for intracellular Salmonella typhimurium. Scand. J. Immunol. 58:493-502.
19. Fahlen, T. F., N. Mathur, and B. D. Jones. 2000. Identification and characterization of mutants with increased expression of hilA, the invasion gene transcriptional activator of Salmonella typhimurium. FEMS Immunol. Med. Microbiol. 28:25-35.
20. Fang, F. C., M. Krause, C. Roudier, J. Fierer, and D. G. Guiney. 1991. Growth regulation of a Salmonella plasmid gene essential for virulence. J. Bacteriol. 173:6783-6789.
21. Fang, F. C., S. J. Libby, N. A. Buchmeier, P. C. Loewen, J. Switala, J. Harwood, and D. G. Guiney. 1992. The alternative -factor KatF (RpoS) regulates Salmonella virulence. Proc. Natl. Acad. Sci. USA 89:11978-11982.
22. Favaro, R., and G. Deho. 2003. Polynucleotide phosphorylase-deficient mutants of Pseudomonas putida. J. Bacteriol. 185:5279-5286.
23. Fields, P. I., E. A. Groisman, and F. Heffron. 1989. A Salmonella locus that controls resistance to microbicidal proteins from phagocytic cells. Science 243:1059-1062.
24. Fierer, J., L. Eckmann, F. Fang, C. Pfeifer, B. B. Finlay, and D. G. Guiney. 1993. Expression of the Salmonella virulence plasmid gene spvB in cultured macrophages and non-phagocytic cells. Infect. Immun. 61:5231-5236.
25. Figueroa-Bossi, N., and L. Bossi. 1999. Inducible prophages contribute to Salmonella virulence in mice. Mol. Microbiol. 33:167-176.
26. Galán, J. E., and D. Zhou. 2000. Striking a balance: modulation of the actin cytoskeleton by Salmonella. Proc. Natl. Acad. Sci. USA 97:8754-8761.
27. Garvis, S. G., C. R. Beuzon, and D. W. Holden. 2001. A role for the PhoP/Q regulon in inhibition of fusion between lysosomes and Salmonella-containing vacuoles in macrophages. Cell. Microbiol. 3:731-744.
28. Gerschenson, L. E., and R. J. Rotello. 1992. Apoptosis: a different type of cell death. FASEB J. 6:2450-2455.
29. Goverde, R. L., J. H. Huisin't Veld, J. G. Kusters, and F. R. Mooi. 1998. The psychrotrophic bacterium Yersinia enterocolitica requires expression of pnp, the gene for polynucleotide phosphorylase, for growth at low temperature (5°C). Mol. Microbiol. 28:555-569.
30. Groisman, E. A. 2001. The pleiotropic two-component regulatory system PhoP-PhoQ. J. Bacteriol. 183:1835-1842.
31. Groisman, E. A., and H. Ochman. 1997. How Salmonella became a pathogen. Trends Microbiol. 5:343-349.
32. Gulig, P. A., H. Danbara, D. G. Guiney, A. J. Lax, F. Norel, and M. Rhen. 1993. Molecular analysis of spv virulence genes of the Salmonella virulence plasmid. Mol. Microbiol. 7:825-830.
33. Gulig, P. A., and T. J. Doyle. 1993. The Salmonella typhimurium virulence plasmid increases the growth rate of Salmonellae in mice. Infect. Immun. 61:504-511.
34. Hansen-Wester, I., and M. Hensel. 2001. Salmonella pathogenicity islands encoding type III secretion systems. Microbes Infect. 3:549-559.
35. Heiskanen, P., S. Taira, and M. Rhen. 1994. Role of rpoS in the regulation of spv virulence genes of Salmonella typhimurium. FEMS Microbiol. Lett. 123:125-130.
36. Hinton, J. C., I. Hautefort, S. Eriksson, A. Thompson, and M. Rhen. 2004. Benefits and pitfalls of using microarrays to monitor bacterial gene expression during infection. Curr. Opin. Microbiol. 7:277-282.
37. Hobbie, S., L. M. Chen, R. J. Davis, and J. E. Galán. 1997. Involvement of mitogen-activated protein kinase pathways in the nuclear responses and cytokine production induced by Salmonella typhimurium in cultured intestinal epithelial cells. J. Immunol. 159:5550-5559.
38. Hurme, R., K. D. Berndt, S. J. Normark, and M. Rhen. 1997. A proteinaceous gene regulatory thermometer in Salmonella. Cell 90:55-64.
39. Kim, C. C., and S. Falkow. 2004. Delineation of upstream signaling events in the Salmonella pathogenicity island 2 transcriptional activation pathway. J. Bacteriol. 186:4694-4704.
40. Knodler, L. A., J. Celli, W. D. Hardt, B. A. Vallance, C. Yip, and B. B. Finlay. 2002. Salmonella effectors within a single pathogenicity island are differentially expressed and translocated by separate type III secretion systems. Mol. Microbiol. 43:1089-1103.
41. Kowarz, L., C. Coynault, V. Robbe-Saule, and F. Norel. 1994. The Salmonella typhimurium katF (rpoS) gene: cloning, nucleotide sequence, and regulation of spvR and spvABCD virulence plasmid genes. J. Bacteriol. 176:6852-6860.
42. Kox, L. F., M. M. Wosten, and E. A. Groisman. 2000. A small protein that mediates the activation of a two-component system by another two-component system. EMBO J. 19:1861-1872.
43. Krause, M., C. Roudier, J. Fierer, J. Hardwood, and D. G. Guiney. 1991. Molecular analysis of the virulence locus of the Salmonella dublin plasmid pSDL2. Mol. Microbiol. 5:307-316.
44. Kudla, J., R. Hayes, and W. Gruissem. 1996. Polyadenylation accelerates degradation of chloroplast mRNA. EMBO J. 15:7137-7146.
45. Kurita, A., H. Gotoh, M. Eguchi, N. Okada, S. Matsumura, H. Mastui, H. Danbara, and Y. Kikuchi. 2003. Intracellular expression of the Salmonella virulence plasmid virulence protein SpvB causes apoptotic cell death in eukaryotic cells. Microb. Pathog. 35:43-48.
46. Lesnick, M. L., N. E. Reiner, J. U. Fierer, and D. G. Guiney. 2001. The Salmonella spvB virulence gene encodes an enzyme that ADP-ribosylates actin and destabilizes the cytoskeleton of eukaryotic cells. Mol. Microbiol. 39:1464-1470.
47. Leszczyniecka, M., R. DeSalle, D. C. Kang, and P. B. Fisher. 2004. The origin of polynucleotide phosphorylase domains. Mol. Phylogenet. Evol. 31:123-130.
48. Leszczyniecka, M., D. C. Kang, D. Sarkar, Z. Z. Su, M. Holmes, K. Valerie, and P. B. Fisher. 2002. Identification and cloning of human polynucleotide phosphorylase, hPNPase old-35, in the context of terminal differentiation and cellular senescence. Proc. Natl. Acad. Sci. USA 99:16636-16641.
49. Libby, S. J., M. Lesnick, P. Hasegawa, E. Weidenhammer, and D. G. Guiney. 2000. The Salmonella virulence plasmid spv genes are required for cytopathology in human monocyte-derived macrophages. Cell. Microbiol. 2:49-58.
50. Lundberg, U., U. Vinatzer, D. Berdnik, A. von Gabain, and M. Baccarini. 1999. Growth phase-regulated induction of Salmonella-induced macrophage apoptosis correlates with transient expression of SPI-1 genes. J. Bacteriol. 181:3433-3437.
51. Mangan, D. F., and S. M. Wahl. 1991. Differential regulation of human monocyte programmed cell death (apoptosis) by chemotactic factors and proinflammatory cytokines. J. Immunol. 147:3408-3412.
52. Marshall, D. G., B. J. Sheehan, and C. J. Dorman. 1999. A role for the leucine-responsive regulator protein and integration host factor in the regulation of the Salmonella plasmid virulence (spv) locus in Salmonella typhimurium. Mol. Microbiol. 34:134-145.
53. Mastroeni, P., and M. Sheppard. 2004. Salmonella infections in the mouse model: host resistance factors and in vivo dynamics of bacterial spread and distribution in the tissues. Microb. Infect. 6:398-405.
54. Matsui, H., A. Abe, S. Suzuki, M. Kijima, Y. Tamura, M. Nakamura, K. Kawahara, and H. Danbara. 1993. Molecular mechanism of the regulation of expression of plasmid-encoded mouse bacteremia (mba) genes in Salmonella serovar cholearesuis. Mol. Gen. Genet. 236:219-226.
55. McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L. Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S. Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan, H. Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston, and R. K. Wilson. 2001. Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413:852-856.
56. Meresse, S., K. E. Unsworth, A. Habermann, G. Griffiths, F. Fang, M. J. Martinez-Lorenzo, et al. 2001. Remodelling of the actin cytoskeleton is essential for replication of intravacuolar Salmonella. Cell. Microbiol. 3:567-577.
57. 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.
58. Miki, T., N. Okada, and H. Danbara. 2004. Two periplasmic disulfide oxidoreductases, DsbA and SrgA, target outer membrane protein SpiA, a component of the Salmonella pathogenicity island type III secretion system. J. Biol. Chem. 279:34631-34642.
59. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Press, Cold Spring Harbor, N.Y.
60. Miller, S. I., and J. J. Mekalanos. 1990. Constitutive expression of the phoP regulon attenuates Salmonella virulence and survival within macrophages. J. Bacteriol. 172:2485-2490.
61. Monack, D. M., D. M. Boule, and S. Falkow. 2004. Salmonella typhimurium persists within macrophages in the mesenteric lymph nodes of chronically infected Nramp+/+ mice and can be reactivated by IFN neutralization. J. Exp. Med. 199:231-241.
62. Monack, D. M., C. S. Detweiler, and S. Falkow. 2001. Salmonella pathogenicity island 2-dependent macrophage death is mediated in part by the host cysteine protease caspase-1. Cell. Microbiol. 3:825-837.
63. Monack, D. M., W. W. Navarre, and S. Falkow. 2001. Salmonella-induced macrophage death: the role of caspase-1 in death and inflammation. Microb. Infect. 3:1201-1212.
64. Monsieurs, P., S. de Keersmaecker, W. W. Navarre, M. W. Bader, F. De Smet, M. McLelland, F. C. Fang, B. De Moor, J. Vanderleyden, and K. Marchal. 2005. Comparison of the PhoPQ regulon in Escherichia coli and Salmonella typhimurium. J. Mol. Evol. 60:462-479.
65. O'Byrne, C. P., and C. J. Dorman. 1994. Transcription of the Salmonella typhimurium spv virulence locus is regulated negatively by the nucleoid-associated protein H-NS. FEMS Microbiol. Lett. 121:99-106.
66. O'Byrne, C. P., and C. J. Dorman. 1994. The spv virulence operon of Salmnella typhimurium LT-2 is negatively regulated by the cyclic AMP (cAMP)-cAMP receptor protein system. J. Bacteriol. 176:905-912.
67. Otto, H., D. Tezcan, R. Girish, F. Haag, M. Rhen, and F. Koch-Nolte. 2000. The spvB gene product of the Salmonella enterica virulence plasmid is a mono(ADP-ribosyl)transferase. Mol. Microbiol. 37:1106-1115.
68. Pullinger, G. D., G. D. Baird, G. M. Williamson, and A. J. Lax. 1989. Nucleotide sequence of a plasmid gene involved in the virulence of Salmonella. Nucleic Acids Res. 17:7983.
69. Py, B., C. F. Higgins, H. M. Krisch, and A. J. Caprousis. 1996. A DEAD-box RNA helicase in the Escherichia coli RNA degradosome. Nature 381:169-172.
70. Raupach, B., N. Kurth, K. Pfeffer, and S. H. E. Kaufmann. 2003. Salmonella typhimurium strains carrying independent mutations display similar virulence phenotypes yet are controlled by distinct host defense mechanisms. J. Immunol. 170:6133-6140.
71. Rhen, M., and C. J. Dorman. 2005. Hierarchial gene regulators adapt Salmonella enterica to its host milieus. Int. J. Med. Microbiol. 294:487-502.
72. Rhen, M., P. Riikonen, and S. Taira. 1993. Transcriptional regulation of Salmonella enterica virulence plasmid genes in cultured macrophages. Mol. Microbiol. 10:45-56.
73. Rhen, M., M. Virtanen, and P. H. Mkel. 1989. Localization by insertion mutagenesis of a virulence-associated region on the Salmonella typhimurium 96 kilobase pair plasmid. Microb. Pathog. 6:153-158.
74. Rosenzweig, J. A., G. Weltman, G. V. Plane, and K. Schesser. 2005. Modulation of Yersinia type three secretion system by the S1 domain of polynucleotide phosphorylase. J. Biol. Chem. 280:156-163.
75. Sarkar, D., M. Leszczyniecka, D. C. Kang, I. V. Lebedeva, K. Valerie, K., Dhar, S. Pandita, and P. B. Fisher. 2003. Down-regulation of Myc as a potential target for growth arrest induced by human polynucleotide phosphorylase (hPNPaseold-35) in human melanoma cells. J. Biol. Chem. 278:24542-24551.
76. Schmieger, H. 1972. Phage P22 mutants with increased or decreased transduction abilities. Mol. Gen. Genet. 119:75-78.
77. 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.
78. Shea, J. E., C. R. Beuzon, C. Gleeson, R. Mundy, and D. W. Holden. 1999. Influence of the Salmonella typhimurium pathogenicity island 2 type III secretion system on bacterial growth in the mouse. Infect. Immun. 67:213-219.
79. Symmons, M. F., G. H. Jones, and B. F. Luisi. 2000. A duplicated fold is the structural basis for polynucleotide phosphorylase catalytic activity, processivity, and regulation. Structure Fold. Des. 8:1215-1226.
80. Taira, S., P. Heiskanen, R. Hurme, H. Heikkil, P. Riikonen, and M. Rhen. 1995. Evidence for functional polymorphism of the spvR gene regulating virulence gene expression in Salmonella. Mol. Gen. Genet. 246:437-444.
81. Taira, S., P. Riikonen, H. Saarilahti, S. Sukupolvi, and M. Rhen. 1992. The mkaC virulence gene of the Salmonella serovar Typhimurium 96 kb plasmid encodes a transcriptional activator. Mol. Gen. Genet. 228:381-384.
82. Tedin, K., and U. Blasi. 1996. The RNA chain elongation rate of the lambda late mRNA is unaffected by high levels of ppGpp in the absence of amino acid starvation. J. Biol. Chem. 271:17675-17686.
83. Tezcan-Merdol, D., T. Nyman, U. Lindberg, F. Haag, F. Koch-Nolte, and M. Rhen. 2001. Actin is ADP-ribosylated by the Salmonella enterica virulence-associated protein SpvB. Mol. Microbiol. 39:606-619.
84. Wilson, J. A., T. J. Doyle, and P. A. Gulig. 1997. Exponential-phase expression of spvA of the Salmonella typhimurium virulence plasmid: induction in intracellular salts medium and intracellularly in mice and cultured mammalian cells. Microbiology 143:3827-3839.
85. Wilson, J. A., and P. A. Gulig. 1998. Regulation of the spvR gene of the Salmonella typhimurium virulence plasmid during exponential-phase growth in intracellular salts medium and a t stationary phase in L broth. Microbiology 144:1823-1833.
86. Yamanaka, K., and M. Inouye. 2001. Selective mRNA degradation by polynucleotide phosphorylase in cold shock adaptation in Escherichia coli. J. Bacteriol. 183:2808-2816.
87. Yehudai-Resheff, S., M. Hirsh, and G. Schuster. 2001. Polynucleotide phosphorylase functions as both an exonuclease and a poly(A) polymerase in spinach chloroplasts. Mol. Cell. Biol. 21:5408-5416.(Sofia Eriksson Ygberg, Ma)