|Home | About | Journals | Submit | Contact Us | Français|
The food-borne pathogenic bacterium Listeria monocytogenes has the potential to adapt to an array of suboptimal growth environments encountered within the host. The pathogen is relatively bile tolerant and has the capacity to survive and grow within both the small intestine and the gallbladder in murine models of oral infection. We have previously demonstrated a role for the principal carnitine transport system of L. monocytogenes (OpuC) in gastrointestinal survival of the pathogen (R. Sleator, J. Wouters, C. G. M. Gahan, T. Abee, and C. Hill, Appl. Environ. Microbiol. 67:2692-2698, 2001). However, the mechanisms by which OpuC, or indeed carnitine, protects the pathogen in this environment are unclear. In the current study, systematic analysis of strains with mutations in osmolyte transporters revealed a role for OpuC in resisting the acute toxicity of bile, with a minor role also played by BetL, a secondary betaine uptake system which also exhibits a low affinity for carnitine. In addition, the toxic effects of bile on wild-type L. monocytogenes cells were ameliorated when carnitine (but not betaine) was added to the medium. lux-promoter fusions to the promoters of the genes encoding the principal osmolyte uptake systems Gbu, BetL, and OpuC and the known bile tolerance system BilE were constructed. Promoter activity for all systems was significantly induced in the presence of bile, with the opuC and bilE promoters exhibiting the highest levels of bile-dependent expression in vitro and the betL and bilE promoters showing the highest expression levels in the intestines of orally inoculated mice. A direct comparison of all osmolyte transporter mutants in a murine oral infection model confirmed a major role for OpuC in intestinal persistence and systemic invasion and a minor role for the BetL transporter in fecal carriage. This study therefore demonstrates a previously unrecognized function for osmolyte uptake systems in bile tolerance in L. monocytogenes.
Listeria monocytogenes is the food-borne agent of the debilitating illness listeriosis in immunocompromised patients and of febrile gastroenteritis in immunocompetent individuals (9, 25). This gram-positive, opportunistic food-borne pathogen has evolved a myriad of sophisticated stress management strategies that allow the detection of environmental stress conditions and the elicitation of appropriate responses to conditions experienced during saprophytic growth in the external environment and, subsequently, during a parasitic life cycle in the mammalian host (5, 9).
Following the ingestion and successful gastric transit of intestinal pathogens, the biological detergent bile represents their most significant challenge, interacting with membrane lipids to reduce membrane integrity and increase permeability (2). The ability to tolerate bile is thus an important factor in promoting pathogen survival and subsequent colonization of the gastrointestinal tract (16). L. monocytogenes is inherently relatively resistant to the deleterious effects of bile (1) and can survive within the mammalian small intestine, as well as the gallbladder of mice and humans (14). This resistance can be ascribed to the presence of a number of bile resistance/detoxification systems, which include an ability to hydrolyze bile via a bile salt hydrolase (BSH) enzyme that is also important for colonization of the gastrointestinal tract and oral infection (2, 8).
The pathogen also expresses a novel bile exclusion system, BilE, which functions by actively excluding bile from the cell. Deleting this system results in a 4 to 5 log reduction in listerial bile tolerance and significantly impairs the ability to infect mice via the oral route (34). In another study, heterologous bilE expression in both Lactococcus lactis and Bifidobacterium breve (species which normally lack this system) significantly improved in vitro bile tolerance and resulted in improved gastrointestinal persistence of the bioengineered strains (36). Significantly, the BilE bile exclusion system shares significant homologies with osmolyte uptake systems in gram-positive organisms (34) and was originally annotated as a glycine betaine/carnitine/choline ABC transporter (11). In addition, BilE is coregulated with osmolyte uptake systems OpuC, BetL, and Gbu by the alternative sigma factor σB (4, 13, 28, 31, 33). However, the system does not appear to play a role in compatible solute uptake or osmotolerance and may represent an evolutionary adaptation of an osmolyte transporter toward a more defined role in bile tolerance.
We have previously created stable deletion mutations in the genes encoding the principal osmolyte transporters in L. monocytogenes (gbu, betL, and opuC). Analysis of the mutants defined dedicated roles for these transporters in the uptake of the compatible solutes betaine (Gbu and BetL) and carnitine (OpuC) (reviewed in reference 33). Furthermore, a significant role for OpuC in the gastrointestinal phase of L. monocytogenes virulence was determined (32, 37). Interestingly, recent work has demonstrated that cloning and heterologous expression of the listerial BetL transporter in the probiotic strain B. breve significantly improved colonization of the murine intestine (29). However, the roles of BetL and Gbu have not been directly tested in the oral murine infection model, and the mechanistic basis underpinning the requirement for OpuC is unclear.
The significant sequence similarity between BilE and other members of the betaine/carnitine/choline transporter family, coupled with the role for OpuC in gastrointestinal persistence, prompted our analysis of the role of osmolyte uptake systems in bile tolerance. Here, we show that Listeria has a requirement for osmolyte uptake systems for the maintenance of full bile tolerance in vitro (in the order OpuC > BetL > Gbu) and that the presence of carnitine contributes significantly to bile tolerance. We used lux tagging of genes to measure promoter expression. This demonstrated that all genes involved in osmolyte uptake are responsive to bile salts, with opuC showing the greatest expression levels in vitro and betL exhibiting the highest expression in vivo during oral murine infection. The data indicate that the uptake of carnitine by OpuC plays a significant role in the bile tolerance and pathogenesis of L. monocytogenes.
The strains, plasmids, and primers used in this study are listed in Tables Tables11 and and2.2. Strains disrupted in one of the BetL (30), Gbu, or OpuC (32) transport systems were used, as well as double and triple mutants constructed against these single mutation backgrounds (37). Since we encountered problems growing the ΔbilE ΔopuC double mutant, this strain was excluded from the experiments. Escherichia coli strain DH10B (Invitrogen, Paisley, United Kingdom) was used as a cloning host during construction of the pPL2lux promoter fusions (see below) and was grown aerobically at 37°C in Luria-Bertani (LB) medium. L. monocytogenes EGD-e and LO28 were grown aerobically at 37°C in brain heart infusion (BHI) medium. Oxoid Ltd. (Basingstoke, Hampshire, United Kingdom) supplied all growth media. For agar plates, BHI medium was solidified with 1.5% agar (Merck). For the survival assays, bovine bile (oxgall, B3883; Sigma) or porcine bile (B8631-100G; Sigma) was added to broth prior to autoclaving. The mutant strains were grown either in BHI broth or in the defined medium (DM) described by Premaratne and coworkers (23). When required, the concentration of bile in the medium was adjusted by the addition of filter-sterilized porcine bile to produce DMB (DM containing 1% bile). Where indicated below, carnitine and glycine betaine (Sigma Chemical Co., St. Louis, MO) were added to DMB as filter-sterilized solutions to a final concentration of 1 mM. For LB medium (pH 7.4), 100 mM morpholinepropanesulfonic acid (MOPS; Sigma) was added to the LB broth, the pH adjusted with NaOH, and the broth autoclaved. Again, bile in the medium was adjusted by the addition of filter-sterilized porcine bile. When appropriate, antibiotics were added to the medium as follows: for E. coli, chloramphenicol (15 μg/ml), kanamycin (50 μg/ml), and erythromycin (250 μg/ml), and for L. monocytogenes, chloramphenicol (7.5 μg/ml), kanamycin (50 μg/ml), and erythromycin (5 μg/ml).
Plasmid DNA was isolated from E. coli by using a Qiagen QIAprep spin Miniprep kit according to the manufacturer's instructions (Qiagen, Crawley, United Kingdom). Genomic DNA was isolated from L. monocytogenes EGD-e by using a Genelute bacterial genomic DNA kit (Sigma, Steinheim, Germany) as recommended by the manufacturer. Transformation of L. monocytogenes was performed as described previously (21). Standard procedures were used for DNA manipulation in E. coli (27). Restriction endonucleases (Roche Diagnostics, Mannheim, Germany), T4 DNA ligase (Roche), and 2× PCR mixture (Promega, Madison, WI) were used as recommended by the manufacturers. Primers were purchased from Sigma Genosys (Haverhill, United Kingdom). PCR products required for cloning were obtained with KOD hot-start high-fidelity DNA polymerase (Merck, Nottingham, United Kingdom) using either 10 ng L. monocytogenes genomic DNA or 50 ng plasmid DNA as the template. PCR was carried out using a Hybaid (Middlesex, United Kingdom) PCR express system.
pORI19 insertional mutagenesis is based on gene disruption followed by a single crossover event, as described by Law et al. (18). Primer pairs F and R for each gene (Table (Table2)2) were used to amplify approximately 500-bp fragments by PCR. The resulting PCR products were digested and ligated to similarly digested pORI19, creating pDW1, pDW2, pDW3, and pDW4 (Table (Table1).1). These resulting plasmid constructs were transformed into E. coli EC101. L. monocytogenes EGD-e containing pVE6007 was then transformed with each construct after they were isolated from E. coli EC101. pORI19 is maintained in Listeria through the use of the temperature-sensitive plasmid pVE6007, which provides the RepA protein in trans. Transformants were selected and grown overnight in 10 ml of broth prewarmed to 42°C (the nonpermissive temperature for pVE6007 replication in Listeria, which selects for events in which pORI19 has integrated into the host chromosome by homologous recombination at the point of homology provided by the cloned DNA) before they were plated onto BHI agar containing 5 μg of erythromycin/ml to select for chromosomal integration and disruption of each gene. Integration was confirmed by PCR with a primer on the chromosome and a primer on the plasmid.
To determine the ability of the strains to survive at lethal bile concentrations, the LO28 strains carrying single and multiple deletions in the osmolyte transport systems (Table (Table1)1) were grown in BHI to stationary phase (~16 h). These cultures were then subcultured (2%) into fresh BHI broth containing 30% (wt/vol) oxgall or 1% (wt/vol) porcine bile. Surviving cells were enumerated by serial dilution in 1/4-strength Ringer's solution and plating onto BHI agar at 20-min intervals.
Studies were also carried out in a DM that was prepared as outlined by Premaratne et al. (23). Listeria monocytogenes LO28 wild type was initially grown to stationary phase in BHI, pelleted, washed, and reinoculated into fresh DM before any experiment using DM was undertaken. This ensured that any intracellular store of the compatible solutes was utilized. Cultures were then grown to stationary phase (shaking at 37°C). Test cultures were exposed to medium containing trimethylglycine (betaine; Sigma Chemical Co., St. Louis, Mo) and/or β-hydroxy-γ-N-trimethyl aminobutyrate (dl-carnitine; Sigma Chemical Co., St. Louis, Mo) that were added to the minimal medium as filter-sterilized solutions with a final concentration of 1 mM.
The luciferase-based reporter system pPL2lux (3) was utilized to investigate the expression of the genes encoding the listerial osmolyte uptake genes in response to bile. The plasmid contains a unique SwaI restriction site that overlaps the start codon of luxA, which allows exact translational fusions of the listerial promoters to the luxABCDE operon. Chromosomal DNA from L. monocytogenes EGD-e was used as the template in PCRs using KOD polymerase (Merck, Nottingham, United Kingdom) with primer pairs as indicated in Table Table2.2. The resulting 0.5-kb PCR products harbored the 500-bp promoter regions immediately upstream of opuC, gbu, betL, and bilE (http://genolist.pasteur.fr/ListiList/). Notably, the start codon of each gene was included at the ultimate 3′ end of the reverse primers. The PopuC, Pgbu, PbetL, and PbilE amplimers were digested with XhoI endonuclease and ligated into pPL2lux digested with endonucleases SwaI and SalI as exact translational fusions upstream of the luxABCDE operon, yielding pPL2lux-PopuC, pPL2lux-Pgbu, pPL2lux-PbetL, and pPL2lux-PbilE, respectively. These plasmids were cloned into E. coli DH10B. Purified plasmids were checked for the correct insert by PCR, digestion, and sequencing by Eurofins MWG Operon to ensure that no mutational event had occurred. Subsequently, pPL2lux, pPL2lux-PopuC, pPL2lux-Pgbu, pPL2lux-PbetL, and pPL2lux-PbilE were integrated into a tRNAArg gene in the chromosome of L. monocytogenes EGD-e by using a site-specific integration system with pPL2 (3). Candidate integrants were checked for site-specific integration by PCR using template DNA obtained directly from colonies by a microwave treatment (3 min at 600 W) and primers PL95 and PL102 (17) to yield a 0.5-kb band, coupled with a chloramphenicol resistance phenotype. Correct inserts were also confirmed by KpnI restriction analysis and promoter sequencing (using a T3 sequencing primer). These constructs allowed the analysis of promoter activity by monitoring bioluminescence activity; therefore, from each transformation, one colony possessing the desired genotype was selected for bioluminescence analysis using a Xenogen IVIS 100 system (Xenogen, Alameda, CA) according to the instructions of the manufacturer.
For the lux promoter assay, strains were grown overnight in BHI at 37°C with shaking. Cultures were diluted 1:50 into 10 ml of fresh medium and grown to mid-exponential phase (optical density at 600 nm of 0.5). They were centrifuged at 7,000 × g for 5 min and washed twice with phosphate-buffered saline (PBS), and the pellet was resuspended in 300 μl of LB medium (pH 7.4) or LB medium (pH 7.4) plus 0.3% porcine bile and immediately visualized using the Xenogen IVIS 100 (binning of 8 for 5 min). Bioluminescence was measured in photons s−1 cm−2 in the Xenogen IVIS 100 (Xenogen, Alameda, CA) using black 96-well plates (Costar).
L. monocytogenes containing pPL2lux-PopuC, pPL2lux-Pgbu, pPL2lux-PbetL, and pPL2lux-PbilE were first cultured in BHI broth containing 7.5 μg/ml chloramphenicol. Cells were then subcultured in fresh BHI broth without antibiotic selection for a total of 100 generations. Vector segregation stability was monitored by plating for isolated colonies every 20 generations, subsequently replica plating 100 colonies onto BHI agar plates with and without 7.5 μg/ml chloramphenicol, and incubating at 37°C for 24 h. The percent loss of the test plasmid in the population was then calculated. A similar procedure was carried out for the plasmids used in the competitive index animal trial.
The plasmid pIMC3kan (22) was used to antibiotic mark L. monocytogenes LO28 wild type and pIMC3ery to mark the deletion strains ΔopuC, Δgbu, ΔbetL, and ΔbilE for a competitive index study. Briefly, purified plasmid preparations were concentrated using pellet paint (Novagen) to a final volume of 5 μl in elution buffer (10 mM Tris-HCl [pH 8.5]) and electroporated into the relevant competent strains. Transformants were selected on BHI agar with either chloramphenicol (7.5 μg/ml) plus kanamycin (50 μg/ml) or chloramphenicol (7.5 μg/ml) plus erythromycin (5 μg/ml). Integration at the correct site was confirmed at the molecular level by PCR using primers PL102 and PL95 (Table (Table22).
L. monocytogenes was inoculated into BHI broth and incubated for 16 h with shaking at 37°C until stationary-phase growth was reached. The cells were recovered by centrifugation (7,000 × g for 5 min) and resuspended in an equal volume of PBS. The optical density at 600 nm of the bacterial suspension was read with a spectrophotometer (Biophotometer; Eppendorf). Appropriate dilutions of the bacterial suspension in sterile PBS were performed to verify bacterial numbers by plate counts on BHI agar, with the appropriate level of antibiotic in each experiment.
Oral inoculation of the 8- to 12-week-old female BALB/c mice with the lux-tagged strains was carried out to examine the role of the promoters in the gastrointestinal stage of listerial infection. Mice received food and water ad libitum until 5 h prior to an intragastric inoculation, at which time food was removed from the cage. Overnight cultures of lux-tagged L. monocytogenes EGD-e::pPL2lux-PopuC, EGD-e::pPL2lux-Pgbu, EGD-e::pPL2lux-PbetL, EGD-e::pPL2lux-PbilE, and L. monocytogenes EGD-e::pPL2lux were used to inoculate animals. The listerial inoculum (5 × 1010 CFU) was introduced (in a total volume of 200 μl) via a 1.5-in., 24-gauge stainless steel gavage needle attached to a 1-ml syringe. Once the desired time had elapsed (75 min postinoculation), the mice were euthanized, the abdominal cavity was aseptically opened, and the intestines removed and viewed under the Xenogen IVIS 100 system with a 5-min exposure and a binning value of 16. The intestines were then added to 5 ml of sterile PBS, homogenized, diluted in sterile PBS, and plated onto BHI agar plates with 7.5 μg/ml chloramphenicol which were subsequently incubated overnight at 37°C. The colonies were counted, and the data were expressed as the mean (± standard error of the mean [SEM]) log10 CFU of L. monocytogenes per organ.
Bacterial infection in the liver and spleen at 3 days postinfection was examined. Intraperitoneal injection was used to bypass the intestinal intraluminal stage of infection, allowing analysis of the role of the promoters in listerial survival following gut transit. For intraperitoneal inoculations, 8- to 12-week-old specific-pathogen-free female BALB/c mice (Harlan, United Kingdom) were inoculated with 2 × 106 CFU in 200 μl of PBS of all strains used for the oral gavage study. Three days postinfection, the mice were sacrificed by cervical dislocation, and individual organs were examined for bioluminescence using a Xenogen IVIS 100 system. Organs were then homogenized in PBS, and serial dilutions were plated onto BHI agar, followed by overnight incubation at 37°C. The resulting colonies were used to calculate the number of bacterial cells per organ.
The role of the deleted genes in virulence was determined by peroral gavage of 8- to 12-week-old BALB/c mice. For peroral inoculations, LO28::pIMC3kan (kanamycin-tagged wild type) and erythromycin-tagged mutant strains (ΔopuC::pIMery, ΔbetL::pIMery, Δgbu::pIMery, and ΔbilE::pIMery) were washed three times with PBS. The wild-type and mutant strains were mixed at a ratio of 1:1. Mice were infected with approximately 3 × 1010 cells (total) suspended in 20 μl of PBS, using a micropipette tip placed immediately behind the incisors. The inoculum was enumerated on BHI agar containing 7.5 μg/ml chloramphenicol plus 50 μg/ml kanamycin or 7.5 μg/ml chloramphenicol plus 5 μg/ml erythromycin. Excretion of viable Listeria was measured in fecal samples collected by scruffing at 24 h postinfection. Stool samples were weighed, homogenized in PBS, diluted, and plated onto BHI containing the relevant antibiotic. At 3 days postinfection, mice were euthanized and the listerial numbers in the liver and spleen were determined by spread plating homogenized samples onto BHI. The organs were homogenized in 5 ml of PBS, serially diluted, and enumerated on BHI agar containing the appropriate concentration of kanamycin or erythromycin. Following enumeration of L. monocytogenes cells in tissue samples, 100 random colonies were tested for the presence of the plasmid by colony PCR using primers PL92 and PL105. The results were interpreted by using the following competitive index (Ci) formula described by Dramsi et al. (7): Ci = output (mutant/wild type)/input (mutant/wild type), where “wild type” represents the number of wild-type bacteria and “mutant” the number of mutant bacteria.
All murine experiments were carried out according to institutional guidelines.
Numerical results were expressed as arithmetic means ± standard deviations of the means. CFU determinations were converted to log10 values, and then the arithmetic means and standard deviations were calculated. Error bars in the figures represent standard deviations. Student's t test was performed to determine the statistical significance.
Previously, we demonstrated a role for the osmolyte uptake system OpuC and a homologous system (BilE) in gastrointestinal infection by L. monocytogenes (32, 34). Here, we employed a set of mutants with deletions in the genes encoding the principal osmolyte uptake systems, OpuC, BetL, and Gbu (35), to establish whether osmolyte uptake may contribute to listerial growth and survival in bile and during the gastrointestinal phase of infection. We first measured the growth characteristics of these mutants under nonstress conditions (no added bile) at 37°C in BHI. No significant difference in growth rate or final optical density was observed for any of the strains (data not shown), indicating normal growth of these mutants under nonstress conditions.
In order to investigate the contribution of the individual genes to the survival of the bacterium at lethal bile concentrations, a series of survival experiments were carried out on the osmolyte uptake mutants in complex broth (BHI) with the addition of either oxgall (30%, wt/vol) or porcine bile (1%, wt/vol) (Fig. 1A and B, respectively). Porcine bile is considered an acceptable substitute for human bile because the salt/cholesterol, phospholipid/cholesterol, and glycine/taurine ratios resemble the composition of human bile (2). We utilized levels of porcine bile (1%, wt/vol) that are lethal for the bacterial species examined and are likely to approximate in vivo levels in microenvironments within the small intestine (e.g., the duodenum) where bile is most concentrated (1, 2). It was observed that the absence of opuC alone rendered L. monocytogenes ~3.5 log (P < 0.001) more sensitive to both oxgall and porcine bile. Following repeated cell viability assays in 30% bovine and 1% porcine bile, the hierarchical arrangement of transporters important in bile tolerance can be represented as OpuC > BetL > Gbu. The survival of ΔopuC null mutants was considerably and consistently lower than that of ΔbetL and Δgbu mutants and was similar to that of a strain with a mutation in the well-defined bile tolerance locus (bilE). This indicates a significant role for OpuC in the bile tolerance of L. monocytogenes. As OpuC primarily transports the compatible solute carnitine, this suggests that efficient carnitine transport is necessary for sustaining cell homeostasis following exposure to bile in vitro.
In order to confirm a role for these loci in bile tolerance, we created separate mutants with mutations in opuC, bilE, gbu, and betL using another method (pORI19 plasmid insertion). While we acknowledge the potential for downstream effects using this mutagenesis approach, examination of the plasmid insertion mutants confirmed a requirement for the selected loci in bile tolerance with the order bilE > opuC > betL > gbu (Fig. (Fig.1C1C).
Subsequent analysis of double and triple mutants lacking combinations of transporters further verifies a significant role for OpuC (and to a lesser extent BetL) in bile tolerance (Fig. (Fig.2).2). All of these mutants are deficient in OpuC, and all demonstrated significantly reduced survival compared to that of the wild type in the presence of lethal concentrations of bile. Deletion mutants lacking both OpuC and BetL demonstrated greater bile sensitivity than mutants lacking OpuC and Gbu, indicating a role for BetL in bile tolerance in this system. Indeed, the mutant with double mutations in OpuC and BetL demonstrated survival rates similar to those of the triple mutant lacking all three transporters. This suggests that the sole uptake of the osmolyte betaine through the glycine betaine transporter Gbu is insufficient to confer protection during bile stress. Another possible explanation for this observation was proposed by Mendum and Smith (20). They demonstrated that BetL, which responded immediately to low concentrations (1 to 2%) of NaCl, was responsible for glycine betaine uptake in response to rapid changes in medium osmolarity, a function which shifts to Gbu after long-term exposure to stress. This may also be true for bile tolerance in vitro. Bile was not completely detrimental for the mutant strains under these conditions, as the BilE system and other survival mechanisms (such as BSH) are still intact in these mutants.
Our initial experiments demonstrated a significant role for the carnitine transporter OpuC in listerial bile tolerance. To determine whether the principal compatible solutes, betaine and carnitine, contribute to the survival of L. monocytogenes in response to bile stress, L. monocytogenes LO28 wild type was exposed to DMB. Carnitine and betaine were added to the DMB to a final concentration of 1 mM. The addition of both carnitine and betaine to the DMB resulted in a survival rate similar to that observed for L. monocytogenes in DM in the absence of bile (Fig. (Fig.3).3). The addition of 1 mM betaine alone did not influence survival in DMB. However, the addition of carnitine resulted in significant amelioration of the toxic effects of bile in our system, with wild-type L. monocytogenes LO28 exhibiting levels of survival equivalent to those seen in the absence of bile. In this experimental system, the addition of carnitine therefore increased survival against porcine bile stress by ~1.5 log. These combined results indicate that OpuC and, more specifically, the carnitine which it transports play a critical role in the ability of L. monocytogenes to survive the extreme cytotoxic effects encountered due to bile stress in vitro. Interestingly, in this model, carnitine was not sufficient to rescue the bile sensitivity of the ΔbilE mutant, indicating that the BilE system is absolutely required for bile tolerance even in the presence of carnitine (data not shown).
The promoter regions of the opuC, gbu, betL, and bilE genes were cloned upstream of a luciferase reporter system in pPL2lux, and the plasmids thus constructed were integrated into the tRNAArg locus of the chromosome of L. monocytogenes EGD-e, allowing the direct assessment of gene expression levels by monitoring bioluminescence levels (3). L. monocytogenes strain EGD-e was utilized for lux fusion studies because the complete genome sequence of this strain was available to inform the selection of promoter regions and primer design (11). The stability of all integrated pPL2lux derivatives in the absence of antibiotic selection pressure was evaluated. Cultures of L. monocytogenes EGD-e::pPL2lux, EGD-e::pPL2lux-PopuC, EGD-e::pPL2lux-Pgbu, EGD-e::pPL2lux-PbetL, and EGD-e::pPL2lux-PbilE were grown for 100 generations in BHI medium without antibiotic selection, after which dilutions were plated every 20 generations to assess the presence of the plasmids in the resulting colonies by scoring for chloramphenicol resistance encoded by the pPL2lux backbone. All 100 colonies tested for each of the four strains at each of the time points were chloramphenicol resistant (data not shown), indicating that the plasmids were stably maintained without antibiotic selection pressure for at least 100 generations under the conditions tested. This high level of stability allowed the use of the luciferase reporter system in animal experiments in which antibiotic selection pressure could not be maintained.
The measurement of gene expression using bacterial luciferase allows for the observation of temporal variations in gene expression and facilitates the monitoring of gene expression in complex environments. We monitored the expression of the selected promoters under elevated bile concentration conditions in order to analyze whether these gene systems are responsive to bile. The growth profiles of the EGD-e control (pPL2lux without cloned promoter) and the EGD-e strains harboring the various luciferase fusions were highly similar in LB medium (pH 7.4) and LB medium (pH 7.4) plus 0.3% porcine bile (data not shown), thus allowing for the direct comparison of reporter gene activity throughout growth. We investigated the levels of gene expression by measuring luminescence from bacterial cells resuspended in medium with or without sublethal concentrations of porcine bile (0.3%, wt/vol). While luminescence was undetectable in medium without bile, the presence of porcine bile stimulated significant luminescence from reporter fusions. The levels of observed luminescence were consistently greatest for opuC and bilE promoter fusions and were in the order opuC = bilE > betL > gbu (Fig. (Fig.4).4). The threshold of detection was 1 × 104 photons s−1. The levels of induction with respect to that of the negative control were calculated to be 83-fold for opuC, 9.8-fold for betL, 86-fold for bilE, and 1-fold for gbu.
We examined the in vivo gene expression in our reporter fusion strains during oral infection of BALB/c mice. Mice orally infected with the lux fusion strains were euthanized 75 min postinfection, and bioluminescence imaging was conducted on the complete intestines (from the stomach to the anus) (Fig. (Fig.55).
The most-significant luminescence levels were observed for the PbetL and PbilE mutants in the small intestine during murine gastrointestinal transit. The levels of luminescence emanating from the intestines of mice inoculated with EGD-e::pPL2lux-PbetL and EGD-e::pPL2lux-PbilE were 6.4 log10 and 5.4 log10 relative light units per intestine, respectively, compared to no detectable luminescence from the negative control (P < 0.001). Figure Figure55 presents images of gastrointestinal tracts that are representative of the results from three independent experiments. Following imaging, homogenization of the intestines revealed similar levels of Listeria bacteria, excluding the possibility of observed differences in expression resulting from differences in gastrointestinal survival.
In order to determine whether genes were expressed as a general response to mammalian infection, intraperitoneal injection was used to bypass the intestinal intraluminal stage of infection, allowing the analysis of the transport systems in listerial survival following gut transit. Three days postinfection, mice were anesthetized with isoflurane, and whole live animals were imaged with the IVIS 100 system. Subsequently, the animals were euthanized and livers and spleens aseptically dissected, and bioluminescent imaging was conducted. This in vivo trial showed no light in the livers and spleens 3 days postinfection, despite significant levels of organ infection (data not shown).
We have previously shown that the OpuC osmolyte transport system is required for full virulence of L. monocytogenes during oral infection of mice (32), and the results of studies with double and triple mutants have suggested that OpuC is the key transporter involved in gastrointestinal infection (34, 37). However, we have not previously directly compared osmolyte uptake mutants in a single study for their ability to infect mice by the oral route. Here, we utilize a previously developed method for tagging Listeria strains prior to competitive index experiments. For our purposes, we tagged the wild-type strain with an integrative plasmid expressing kanamycin resistance, while mutants were tagged with a similar plasmid expressing resistance to erythromycin. Mutant and wild-type strains were checked for site-specific integration of the marker and subjected to the same plasmid stability determinations as described above. Plasmids were integrated at the tRNAArg site and were 100% stable over 100 generations (data not shown).
Animals were inoculated perorally with a 1:1 ratio of the wild type and the appropriate mutant strain. A competitive index value of less than 1 indicated that the mutant strain was out-competed by the wild-type strain.
Gut persistence was monitored by determining fecal carriage after 24 h (Fig. (Fig.6).6). The mutant lacking the betaine uptake system, Gbu, exhibited no significant difference in gut persistence relative to that of the wild-type and was therefore not deemed to play a role in gut persistence under the conditions tested. However, mutants lacking OpuC, BilE, and, to a lesser extent, BetL exhibited significantly reduced levels in feces 24 h postinfection.
Systemic infection by L. monocytogenes follows invasion of the gastrointestinal epithelium and results in listerial multiplication in the primary organs of infection, the liver and spleen. On day 3 postinfection, the levels of the wild-type and the mutant strains were determined for both the liver and the spleen (Fig. (Fig.7).7). The LO28 ΔopuC and LO28 ΔbilE mutant strains reached significantly (P < 0.001) lower levels than the wild type in the liver and spleen of orally infected animals after 3 days. Deletion of gbu or betL had no significant effect upon the infection of liver or spleen following oral infection.
L. monocytogenes is an opportunistic pathogen capable of a free-living existence within the external environment, as well as adaptation to the host following food-borne infection. The passage from the food environment to the gastrointestinal tract elicits an array of adaptive responses that are necessary for transient colonization of the upper gastrointestinal tract prior to systemic infection (5, 9, 12, 35). In particular, tolerance of bile by L. monocytogenes is key to pathogenesis, since bile resistance is necessary for survival within the gastrointestinal tract (2, 8, 35) and the pathogen is known to colonize the gallbladder during infection in the mouse model (14, 15). Passage to the gastrointestinal tract is also postulated to involve osmotic upshift (6), and we and others have previously described the key osmolyte uptake systems of L. monocytogenes that are necessary for osmoadaptation in this organism (33). In our previous analysis of the carnitine transporter (OpuC), we determined that this system plays an important role in gastrointestinal tract survival (32), which we previously attributed to an ability to overcome osmotic challenges in this environment (6). Prompted by sequence and predicted structural similarities between the main listerial bile exclusion system (BilE) and osmolyte transporters, particularly OpuC, we investigated the possibility that osmolyte uptake systems may also play a role in bile tolerance. Using a set of mutants with deletions in the known osmolyte transporters BetL, Gbu, and OpuC, we determined a significant role for OpuC and carnitine transport in resistance to bile.
We determined that a L. monocytogenes mutant lacking the carnitine transport system, OpuC, is exquisitely sensitive to bile stress. In our experiments, the OpuC mutant survived bile stress as poorly as a strain with a mutation in the dedicated bile exclusion system BilE (34) and was reduced by a factor of 4 logs in comparison to the level of the wild type following exposure to bile. A lesser role was determined for the betaine transporter BetL, and overall, the importance of these systems for bile tolerance was determined as ΔbilE ≥ ΔopuC > ΔbetL > Δgbu. The survival of wild-type L. monocytogenes at high bile concentrations was greatly enhanced by the addition of carnitine but not betaine to DMB. This suggests that carnitine is the primary osmolyte mediating the resistance to bile in our system and that the influence of BetL is a reflection of the fact that this osmolyte uptake system can also transport carnitine at low efficiencies (37).
We utilized lux-promoter fusions to analyze promoter expression from genes encoding the major osmolyte uptake systems. All fusion strains were strongly induced in response to bile in vitro, with the promoter of the opuC operon showing the highest levels of expression (equivalent to that of the bilE promoter). In vivo gene expression from lux fusions was determined by visualizing light emission in excised gastrointestinal tracts from orally infected mice. We consistently obtained luminescence signals, indicative of gene expression in this environment, from mice infected with PbetL, PopuC, and PbilE strains but not from mice infected with the Pgbu strain. We did not detect light emission from organs of mice infected systemically (even at high inoculum levels), indicating that gene expression from these promoters is responsive to local conditions encountered within the gastrointestinal tract. Indeed, the osmolarity of the lumen of the gastrointestinal tract has been estimated at 0.3 M NaCl, which is approximately twice that encountered within the circulatory system (6). This elevated osmolarity is likely to represent inducing conditions for the major osmolyte uptake systems which are under the transcriptional control of the alternative sigma factor σB. Significantly, σB also regulates the expression of the main bile resistance mechanisms, BSH and BilE (13). We have previously shown that sigB is upregulated in L. monocytogenes during passage through the gastrointestinal tract (2), and Garner and coworkers (10) have demonstrated that the sigma factor is necessary for gastrointestinal infection in the guinea pig model. It is proposed that σB regulates genes required for survival of gastrointestinal stressors (bsh, bilE, opuC, and arcA), as well as genes required for early infection (inlA), and is therefore a key coordinator of gastrointestinal adaptation (5, 9, 13, 26).
Our previous work has demonstrated a role for OpuC in gastrointestinal infection in mice (32, 37). Here we directly compared a strain with a mutation in opuC with single mutants with mutations in other transporters and in bilE for early gastrointestinal persistence (fecal shedding) and ability to cause invasive disease in orally infected mice. The results confirm a major role for OpuC in mediating gastrointestinal survival (as indicated by fecal shedding) and invasive disease following oral infection and demonstrate that OpuC is as important as BilE during this phase of infection. The data eliminate a direct role for BetL or Gbu in invasive disease; however, deletion of betL had a significant effect upon early fecal colonization. This role in gastrointestinal persistence is supportive of recent data demonstrating that heterologous expression of betL in Bifidobacterium breve can enhance colonization of the gastrointestinal tract in a mouse model (29).
In the murine system, the primary source of carnitine in the gastrointestinal tract is most likely from desquamation of host enterocytes. However, our data may have implications for the survival of Listeria in the gastrointestinal tract of humans or animals that consume diets rich in meats, which are a source of dietary carnitine. While the OpuC mutant exhibits dramatic bile sensitivity in vitro, the effects of the opuC mutation upon in vivo colonization are more subtle. This may suggest that luminal levels of carnitine are low in the murine model, as mouse chow is likely to provide a poor source of dietary carnitine. In addition, gastrointestinal motility ensures that bile is not evenly distributed along the small intestine, and there may be intestinal microenvironments where bacteria are not exposed to high levels of bile. Finally, the composition of murine bile differs significantly from that of porcine or bovine bile with respect to the specific bile acids present (2). We are currently investigating the role of additional/dietary carnitine in protection of L. monocytogenes within the gastrointestinal tract.
In conclusion, we have systematically examined the role of the main osmolyte uptake systems of L. monocytogenes during the gastrointestinal phase of infection. The work confirms a significant role for the principal carnitine transport system (OpuC) during gastrointestinal infection, potentially elicited through enhanced bile tolerance mediated by carnitine. The finding has significant implications, as foods rich in carnitine (including processed ready-to-eat meat products) may potentially enhance the bile tolerance and survival of infective strains within the gastrointestinal environment.
Debbie Watson is funded by Science Foundation Ireland under the Research Frontiers Programme (05/RFP/Gen0021). Roy D. Sleator is a Health Research Board (HRB) Principal Investigator.
We acknowledge the continued financial assistance of the Alimentary Pharmabiotic Centre (APC), funded by Science Foundation Ireland (SFI).
Editor: J. B. Bliska
Published ahead of print on 8 September 2009.