|Home | About | Journals | Submit | Contact Us | Français|
The bacterial pathogen Listeria monocytogenes survives under a myriad of conditions in the outside environment and within the human host where infections can result in severe disease. Bacterial life within the host requires the expression of genes with roles in nutrient acquisition as well as the biosynthesis of bacterial products required to support intracellular growth. A gene product identified as the substrate-binding component of a novel oligopeptide transport system (encoded by lmo0135) was recently shown to be required for L. monocytogenes virulence. Here we demonstrate that lmo0135 encodes a multifunctional protein that is associated with cysteine transport, acid resistance, bacterial membrane integrity, and adherence to host cells. The lmo0135 gene product (designated CtaP, for cysteine transport associated protein) was required for bacterial growth in the presence of low concentrations of cysteine in vitro, but was not required for bacterial replication within the host cytosol. Loss of CtaP increased membrane permeability and acid sensitivity, and reduced bacterial adherence to host cells. ctaP deletion mutants were severely attenuated following intragastric and intravenous inoculation of mice. Taken together, the data presented indicates that CtaP contributes to multiple facets of L. monocytogenes physiology, growth, and survival both inside and outside of animal cells.
The gram-positive bacterium Listeria monocytogenes is a food-borne pathogen that has been responsible for some of the largest and most expensive food recalls in US history that have resulted in thousands of reported illnesses and hundreds of deaths within the past few years (CDC, 1998, 1999, 2004; Cossart, 2007; Gibbons et al., 2006; Mead et al., 2006). The bacterium is a ubiquitous pathogen that survives in a wide array of conditions encountered in the outside environment and inside the human host (Farber and Peterkin, 1991; Fenlon et al., 1996). While L. monocytogenes does not form spores, it is capable of adapting to a range of fluctuations in pH, salt concentration, and temperature; these properties have enabled the organism to contaminate and persist in food sources despite common methods used for preservation (Czuprynski, 2005; Gandhi and Chikindas, 2007). Humans become infected with L. monocytogenes through the ingestion of contaminated food products, resulting in mild gastroenteritis in healthy individuals or developing into more serious conditions such as meningitis and septicemia in immunocompromised individuals, often resulting in death (Drevets and Bronze, 2008; Gray and Killinger, 1966; Vazquez-Boland et al., 2001).
The L. monocytogenes transition from life in the outside environment to life as an intracellular pathogen in the human host involves the transcriptional induction of a number of gene products that are required for host cell invasion, perforation of the phagosomal membrane and escape into the cytosol, bacterial replication, and spread to neighboring host cells (Cossart and Toledo-Arana, 2008; Dussurget, 2008; Freitag, 2006; Gray et al., 2006; Ireton, 2007; Scortti et al., 2007). The expression of many of the gene products required for intracellular bacterial replication and cell-to-cell spread is controlled by the transcriptional regulator PrfA (Freitag, 2006; Gray et al., 2006; Leimeister-Wachter et al., 1990; Mengaud et al., 1991; Scortti et al., 2007). While much attention has focused on the contributions of specific PrfA-dependent virulence factors required for L. monocytogenes host cell invasion, phagosome escape and cell-to-cell spread, less is known about the metabolic requirements of L. monocytogenes either in the outside environment or within host cells. L. monocytogenes is auxotrophic for selected vitamins and amino acids and thus must acquire specific nutrients directly from the host or from the outside environment (Marquis et al., 1993; Premaratne et al., 1991; Tsai and Hodgson, 2003; Welshimer, 1963). In particular, the mechanisms used by the bacterium to scavenge nutrients and the specific biosynthetic pathways required for intracellular replication are not well defined.
The genome of Listeria contains 331 genes encoding different transport proteins, representing approximately 11.6% of all the predicted genes encoded in its genome (Glaser et al., 2001). Approximately 88 of these 331 predicted transport proteins are devoted to carbohydrate transport as part of the phosphoenolpyruvate-dependent phosphotransferase (PTS) system. The large number of carbon transport systems presumably allows bacteria to import carbon sources for survival in diverse niches in the outside environment, whereas intracellular growth appears to exploit alternative (non-PTS) carbon utilization pathways such as those involved in the uptake of phosphorylated sugars and glycerol from the host (Chatterjee et al., 2006; Chico-Calero et al., 2002; Joseph et al., 2006; Joseph et al., 2008; Milohanic et al., 2003). Another host component scavenged by intracellular L. monocytogenes is lipoic acid which is used by the bacterium to generate branched chain fatty acids (Keeney et al., 2009; Keeney et al., 2007). In addition to using host-derived lipoylated peptides and phosphorylated sugars, previous studies have reported that L. monocytogenes preferentially uses intracellular peptides as a source of amino acids (Marquis et al., 1993). Currently, two distinct oligopeptide transport systems have been described in L. monocytogenes: the Dpt and OppA systems (Borezee et al., 2000; Verheul et al., 1995; Verheul et al., 1998; Wouters et al., 2005). The OppA system transports peptides 4-8 amino acids in length, and is required for bacterial growth at low temperatures and for intracellular survival in both macrophages and mice (Borezee et al., 2000). The Dpt system transports smaller peptides and has been shown to contribute to salt stress resistance and virulence in a mouse model of infection (Verheul et al., 1995; Wouters et al., 2005).
Recently, a putative oligopeptide substrate-binding protein encoded by lmo0135 was found to contribute to virulence of L. monocytogenes in a murine model of intravenous infection (Port and Freitag, 2007). Based on homology, the lmo0135 encoded protein belongs to a family of solute binding proteins that are part of ABC transport systems associated with the uptake of peptides (Port and Freitag, 2007; Tam and Saier, 1993). Peptide transport systems have been associated with diverse functions including nutrient acquisition, bacterial quorum sensing, signal transduction, recycling of peptidoglycan fragments, regulation of virulence expression, adherence to eukaryotic cells, alterations in antibiotic susceptibility, and hemolytic activity (Alloing et al., 1998; Cundell et al., 1995; Domenech et al., 2009; Goodell and Higgins, 1987; Lyon and Novick, 2004; Malone et al., 2007; Martin et al., 2000; Wu et al., 2007). We report here that lmo0135 (designated ctaP, for cysteine transport associated protein) encodes a multifunctional protein that is associated with cysteine transport, acid resistance, bacterial membrane integrity, and adherence to host cells. CtaP thus makes multiple contributions to L. monocytogenes growth and survival in environments both inside and outside of mammalian hosts.
lmo0135 is predicted to encode a 524 amino acid gene product that shares significant homology with known oligopeptide-binding proteins of oligopeptide ABC transport systems (Port and Freitag, 2007; Tam and Saier, 1993). Two genes, lmo0136 and lmo0137, encoding putative permease proteins, are located immediately downstream of lmo0135 and presumably function together with the lmo0135 gene product as an ABC transporter (Fig. 1A). The amino acid sequence of Lmo0135 shares 46% and 42% similarity with the oligopeptide binding proteins of L. monocytogenes oligopeptide transport systems OppA and Lmo2569, an uncharacterized putative dipeptide transporter. Based on topology predictions, Lmo0135 contains one transmembrane domain at its N-terminus followed by a large extracellular domain putatively involved in substrate binding (Fig. 1B) (Baumgartner et al., 2007; Tam and Saier, 1993). A consensus sequence for proteolytic cleavage and lipoprotein attachment (LTACGGS) is present within the N-terminal domain (Fig. 1B) consistent with the identification of Lmo0135 as a lipoprotein (Baumgartner et al., 2007). The isolation of Lmo0135 from both culture supernatants and cell membrane fractions suggests that the protein is both cell membrane associated and secreted (Baumgartner et al., 2007; Port and Freitag, 2007; Schaumburg et al., 2004). The lmo0135-0137 gene cluster is highly conserved among all sequenced pathogenic strains of L. monocytogenes as well as the non-pathogenic Listeria innocua.
Previous work examined the contributions of Lmo0135 to bacterial virulence through the construction of a lmo0135 loss of function mutation using a temperature sensitive plasmid insertion vector (Port and Freitag, 2007). As the resulting plasmid insertion mutants were unstable, we sought to more rigorously assess the role of Lmo0135 in Listeria physiology through the construction of a stable in-frame deletion mutant. The coding region of lmo0135 was deleted and replaced with the ermB gene encoding resistance to erythromycin. The resulting Δlmo0135::erm mutation facilitated phage transduction of the gene deletion into different L. monocytogenes genetic backgrounds.
Upon isolation of the Δlmo0135::erm mutant, it was observed that colonies of the mutant strain on BHI agar plates were smaller in size in comparison to wild type colonies (data not shown). An assessment of in vitro growth in BHI indicated that the Δlmo0135::erm mutant exhibited a noticeable growth defect in comparison to the parental wild-type 10403S strain (Fig. 2A). Complementation of the Δlmo0135::erm mutation using a single copy of the lmo0135 gene carried on the integrative plasmid vector pPL2 (Lauer et al., 2002) fully restored growth (Fig. 2A). The doubling time of wild-type and the complemented mutant strain in BHI broth was approximately 40 minutes, whereas the doubling time of the Δlmo0135::erm mutant was approximately 90 minutes, more than twice that of the wild-type strain. Transduction of the Δlmo0135::erm mutation back into the wild type strain background resulted in an identical growth defect for isolated transductants (data not shown). Similar growth defects were also observed for the Δlmo0135::erm mutants in LB, a less complex and more nutrient-limited media, and the defect could also be rescued by complementation (Fig. 2B). Most strikingly, no measurable growth of Δlmo0135::erm was detected when the mutant was inoculated into a minimal defined medium (HTM) (Fig. 2C). HTM supported the growth of wild-type 10403S strain however the growth rate was significantly slower in comparison to bacterial growth to BHI or LB (Fig. 2). The introduction of lmo0135 in single copy on the pPL2 plasmid only partially restored the growth of the Δlmo0135::erm mutant in HTM, suggesting that plasmid-based expression of lmo0135 does not fully recapitulate the expression patterns of the chromosomal copy.
The growth defect observed for strains lacking lmo0135 suggested that the gene product is required for the transport of a component necessary for optimal bacterial replication in broth culture, most notably in minimal media. HTM is a relatively simple medium that contains glucose as a carbon source, methionine and cysteine (0.1 mg/ml final concentration) as amino acid sources, vitamin B, biotin, lipoic acid, MgSO4, Na2HPO4, KH2PO4 and MOPs as a buffering agent (Tsai and Hodgson, 2003). To determine which nutrient component(s) facilitated the growth of the Δlmo0135::erm mutant in BHI and LB versus minimal medium, individual components of LB (yeast extract and tryptone) were added to HTM and assessed for the ability to rescue Δlmo0135::erm mutant growth. The addition of tryptone was found to almost completely restore growth of the Δlmo0135::erm mutant in HTM (Fig. 3A) and it also enhanced the growth of the wild type strain in this medium. Yeast extract was also found to enhance the growth of both the mutant and wild type strains in HTM (data not shown), however this product represents a more complex mixture of substrates. Since tryptone consists of peptide fragments derived from the tryptic digest of casein, tryptone-dependent growth restoration indicated that peptides can at least partially compensate for the growth defect associated with the loss of Lmo0135 in HTM.
Based on the observations that: (1) HTM medium contains only free methionine and cysteine as amino acid sources; and (2) that the growth of the lmo0135 mutant in HTM could be restored by the addition of tryptone, we investigated whether strains lacking Lmo0135 are deficient for the transport of either methionine or cysteine, or for both amino acids. We determined whether a significant increase in the concentration of either methionine or cysteine could rescue mutant growth in HTM by potentially allowing for uptake of the amino acids via an alternate and/or low affinity transport system. Increasing the concentration of free cysteine from 0.1 mg/ml to 0.5 mg/ml or 1.0 mg/ml was found to restore the growth of the Δlmo0135::erm mutant in HTM (Fig. 3B). Increased concentrations of free methionine did not restore mutant growth, nor did increasing the concentrations of other amino acids such as valine or histidine (Fig. 3C and data not shown). The requirement for high concentrations of cysteine could not be compensated for by the addition of a comparable concentration of a reducing agent such as dithiothreitol (data not shown), suggesting that the requirement for cysteine did not reflect a mutant requirement for the presence of a reducing agent. Multiple cysteine transport systems of both high and low affinity have been described in other bacteria including Bacillus subtilis, Legionella pneumophila, E. coli, and Salmonella enterica (Baptist and Kredich, 1977; Berger and Heppel, 1972; Burguiere et al., 2004; Ewann and Hoffman, 2006). Lmo0135 may therefore function as part of a high affinity cysteine transporter to enable bacterial growth in the presence of low concentrations of cysteine, with other transport systems and/or channels able to support the uptake of cysteine only at high concentrations. We therefore chose to designate Lmo0135 as CtaP, for cysteine transport associated protein.
CtaP was required for bacterial growth in broth culture in the presence of low concentrations of free cysteine, however the presence of tryptic peptides (tryptone) was also capable of restoring growth of the mutant in HTM to nearly wild type levels, indicating that peptides can be used as an alternate source of cysteine (Fig. 3). It has been demonstrated that L. monocytogenes preferentially uses peptides over free amino acids when growing in the cytosol of infected host cells (Marquis et al., 1993). We therefore examined the ability of the ΔctaP::erm mutant to replicate within the cytosol of infected J774 mouse macrophage-like cells. J774 cells were infected with wild type or the ΔctaP::erm mutant, and bacterial growth was monitored over 8 hours of infection. Strains lacking CtaP were indistinguishable from the wild-type strain with respect to intracellular growth in J774 cells (Fig. 4). The robust growth of the ΔctaP::erm mutant in this cell line suggests that the cysteine transport defect observed for the mutant in broth culture can be compensated for by available peptides within the host cytosol.
Given that the ΔctaP::erm mutant did not exhibit any growth defect within phagocytic J774 cells, we next examined intracellular growth of the mutant in non-professional phagocytic cells. A rapid assessment of the ability of L. monocytogenes to invade host cells, escape from the phagosome, replicate and spread to adjacent cells can be obtained by measuring the bacterium's ability to form visible zones of cell clearing (or plaques) in monolayers of mouse L2 fibroblast cells (Sun et al., 1990). Plaques formed by the ΔctaP::erm mutant following the infection of L2 cells were similar in size to the plaques formed by the wild-type and complemented mutant strains indicating that the ΔctaP::erm mutant exhibited normal patterns of intracellular growth and cell-to-cell spread (Fig. 5A). However, the absolute number of plaques formed by the mutant per CFU was severely reduced (about 5- to 10-fold fewer plaques in comparison to the wild type or complemented strain) (Fig. 5A). The overall reduction in the number of plaques formed by the mutant is indicative of a reduction in the ability of the bacteria to adhere to and/or invade fibroblast cells.
CtaP's potential role in cell adhesion and/or invasion was further explored using PtK2 epithelial cells. Consistent with the reduction in the number of plaques formed in L2 cells, the number of bacteria recovered after 1 hour of infection of PtK2 cells was approximately one log lower for the ΔctaP::erm mutant compared to wild-type strain (Fig. 5B). While bacterial numbers recovered from cells infected with the ΔctaP::erm mutant remained consistently lower in comparison to wild-type L. monocytogenes, the rate of intracellular bacterial growth was similar between the two strains (Fig. 5B). Visual inspection of the infected monolayers by light microscopy at seven hours post-infection indicated that for cells infected with the wild type strain, nearly 100% of the cells in the monolayer were infected. In contrast, only about 10% of the cells in the monolayer appeared to be infected with mutant bacteria (data not shown). Examination of cells incubated with wild type or mutant strain for shorter time periods prior to the recovery of cell-associated bacteria indicated that the number of bacteria recovered at 15 and 30 minutes post-infection were approximately 1 log lower for the mutant in comparison to wild type strains (Fig. 5C), a result consistent with a cell adhesion defect.
To determine if CtaP has the capacity to function directly as an adhesin, purified N-terminal His-tagged CtaP was covalently coupled to fluorescently labeled latex beads and incubated with monolayers of PtK2 cells for one hour. Control experiments using beads coupled to BSA were carried out in parallel for comparison, and adherent beads associated with cell monolayers were visualized using fluorescent microscopy. Beads coupled to CtaP were found to adhere to PtK2 cell monolayers in numbers that were approximately 10-fold higher than the numbers of adherent beads coupled to BSA (Fig. 6) or beads without coupled protein (data not shown). These data are consistent with an ability of CtaP to directly facilitate bacterial adhesion to host cells.
CtaP thus appears to be a multifunctional protein associated with both cysteine transport and cell adhesion. Previous studies had indicated a role for CtaP in bacterial virulence based on the observation that the inactivation of ctaP (lmo0135) by a plasmid insertion resulted in reduced numbers of bacteria recovered from the livers and spleens of intravenously infected mice (Port and Freitag, 2007). One caveat associated with the previous study was the use of a mutant strain that contained a large temperature sensitive plasmid insertion within ctaP that could have resulted in polar effects on downstream gene expression as well as the potential for loss of the unstable plasmid insertion. Therefore, we re-examined the role of CtaP in bacterial virulence following intravenous inoculation of mice with the stable ΔctaP::erm deletion mutant. Groups of 8- to 10- week old female mice were inoculated through the tail vein with 2 × 104 CFUs of the wild-type, ΔctaP::erm, or the ΔctaP::erm mutant complemented with pPL2-ctaP. Consistent with previous results (Port and Freitag, 2007), a significant defect was observed for the ΔctaP::erm mutant following an intravenous infection of mice and this defect increased in magnitude with time (Fig. 7A). The number of bacterial CFU recovered from both the livers and spleens of mice infected with ΔctaP::erm were approximately 2-4 logs lower than the number recovered from the wild-type or the complemented mutant strain at 72 hours post-infection (Fig. 7A).
Given the ability of CtaP to promote bacterial adhesion to host cells, and because bacterial adhesion would likely be of critical importance within the intestine, we next examined the ability of ΔctaP::erm strains to colonize the intestine and to cause infections in mice following intragastric inoculation. Although mice have generally been considered to be a sub-optimal model for intragastric infection due to the low binding affinity of the L. monocytogenes invasion protein InlA for the mouse E-cadherin receptor (Bonazzi et al., 2009; Lecuit, 2007), the recent isolation by Wollert et al (Wollert et al., 2007) of InlA mutants (InlAm) with increased affinity for mouse E-cadherin has resulted in a highly reproducible intragastric murine model of L. monocytogenes infection. Following the introduction of the inlAm allele into wild type and ΔctaP::erm strains, groups of 6- to 8-week old C57BL/6 mice were orally inoculated with 1 × 108 CFUs of either L. monocytogenes inlAm, inlAm ΔctaP::erm, or the complemented mutant strain. At 72 hours post-infection, mice infected with L. monocytogenes inlAm exhibited symptoms of illness (ruffled appearance and hunched backs), whereas mice infected with L. monocytogenes inlAm ΔctaP::erm appeared completely healthy and showed no signs of illness (data not shown). Examination of the stomach, small intestines, livers, and spleens of infected animals for bacterial CFUs indicated that L. monocytogenes inlAm ΔctaP::erm was severely defective in its ability to colonize all four organs (Fig. 7B). The number of bacteria recovered from the organs of mice infected with the inlAm ΔctaP::erm mutant were at least 3-4 logs lower than the numbers recovered from the organs of mice infected with L. monocytogenes inlAm. Interestingly, and in striking contrast to the intravenous infection data, the plasmid encoded copy of ctaP failed to complement the ΔctaP::erm mutation (Fig. 7B). Analyses of bacteria recovered from animals infected with the complemented mutant strain indicated the almost complete loss of the integrated plasmid from nearly all (>98%) of the recovered bacteria (data not shown). The essentially uniform loss of the normally stable integrated pPL2-ctaP plasmid strongly suggests that the expression of ctaP from the plasmid was detrimental to bacterial fitness or survival at an early stage following intragastric inoculation. Despite the loss of the complementation plasmid, the striking reduction in bacterial CFUs recovered from the stomachs and intestines of animals infected with the ΔctaP::erm mutant suggests a critical role for CtaP in gastric survival and intestinal colonization.
The reduced ability of the ΔctaP::erm mutant to survive and colonize the gastrointestinal tract of mice could reflect reduced bacterial adhesion and/or invasion of intestinal epithelial cells, but could also result from decreased bacterial fitness or survival, particularly in the stomach and the small intestine. We therefore compared growth of the mutant and wild-type strains under conditions of acid (BHI pH 5.5) or osmotic stress (BHI + 5% NaCl). Whereas the addition of 5% NaCl reduced the growth rate of both the wild type and the ΔctaP::erm mutant in BHI medium, both strains reached overall cell densities that were comparable to those observed in BHI broth without additional NaCl (Fig. 8A). However, in contrast to wild type L. monocytogenes, overall growth of the ΔctaP::erm mutant was severely reduced in acidic BHI media (Fig. 8B). The ΔctaP::erm mutant reached a maximum OD600 of 0.5 under acidic conditions, whereas the wild type and complemented mutant strains reached an OD600 of ≥1.0 under the same conditions. Interestingly, examination of the colony forming units of the complemented mutant strain recovered after bacterial growth in acidic conditions indicated that 30-50% of the pPL2-ctaP plasmid complemented mutants no longer retained the pPL2-ctAP plasmid after 24 hours; plasmid loss was not observed to occur during exponential or early stationary growth or under conditions of osmotic stress (data not shown). Taken together, these results indicate that CtaP is required for bacterial resistance to acid stress but that prolonged exposure to acidic media results in the loss of the complementing plasmid. As loss of the pPL2-ctAP plasmid was not observed under other in vitro conditions examined or during iv infection of mice [but was observed following intragastric infection (Fig. 7)], we speculate that the expression of plasmid-encoded ctAP differs from chromosomally encoded ctaP in some way that eventually compromises bacterial fitness under acidic conditions.
In addition to observed differences in acid resistance, ΔctaP::erm strains also exhibited noticeable alterations in bacterial surface hydrophobicity as well as membrane integrity. Both the wild type and complemented mutant strains were capable of binding Congo Red (CR), a planar, hydrophobic, lipophilic diazo dye (Cangelosi et. al., 1999), however the ΔctaP::erm mutant did not bind CR (Fig. 9A). The wild type and complemented mutant strains also exhibited increased binding to untreated polystyrene (a hydrophobic surface) in comparison to the ΔctaP::erm mutant (data not shown). Alterations in mutant strain membrane permeability were also observed based on the ability of the mutant to take up both membrane permeant and impermeant fluorescently labeled nucleic acid stains. The ΔctaP::erm mutant incorporated both a membrane impermeant dye, propidium iodide (PI), and a membrane permeant dye, SYTO9 (Fig. 9B). Of a total of 1184 ΔctaP::erm mutant bacterial cells counted in the ten different fields, 8% incorporated PI alone and 26% incorporated both dyes (thus appearing orange/yellow). In contrast, wild type and complemented mutant strains incorporated SYTO9 with only 1% scored as PI positive out of 1764 wild type and 1638 complemented mutant bacteria counted in ten independent fields (Fig. 9B). Although the ΔctaP::erm mutant incorporated both dyes, the bacterial cells were viable as the number of colony forming units in these cultures were similar to the numbers obtained for the wild type and the complemented mutant strain. Taken together, the increased sensitivity of the ΔctaP::erm mutant to acid stress and the observed alterations in surface and membrane integrity could contribute to a reduction in bacterial survival during transit through the stomach and small intestine as well as potentially reducing bacterial adhesion to host cells, thereby reducing the numbers of bacteria capable of entering the bloodstream.
While mutants lacking CtaP were strikingly attenuated in intragastric mouse models of infection (Fig. 7B), significant attenuation was observed even when the gastrointestinal barrier was bypassed via the injection of bacteria directly into the bloodstream (Fig. 7A). Given that loss of CtaP compromises bacterial membrane integrity, it seemed possible that the virulence defect observed for the ΔctaP::erm mutant following bloodstream infection resulted from an increased susceptibility to macrophage killing. Murine bone marrow-derived macrophages (BMMØ) were therefore isolated and infected with wild type and ΔctaP::erm mutant strains in the presence and absence of IFNγ activation. Equivalent numbers of bacterial CFU's were recovered at all time points for wild-type and ΔctaP::erm mutant strains following the infection of primary BMMØ cells in both the presence and absence of IFNγ (data not shown). These results suggest that the in vivo virulence defect of ΔctaP::erm observed following intravenous infection of mice is not due to increased growth inhibition or killing of the mutant by macrophages.
The disparate environments encountered by L. monocytogenes require physiological flexibility such that the bacterium can adapt and survive when exposed to a wide range of conditions. CtaP represents a multifunctional L. monocytogenes protein that has been adapted to fulfill a variety of roles that serve to enhance bacterial survival. Detectable both at the bacterial surface and as a secreted gene product in bacterial supernatants (Baumgartner et al., 2007; Port and Freitag, 2007; Schaumburg et al., 2004), we demonstrate here that CtaP can be associated with cysteine acquisition, host cell adhesion, acid resistance, bacteria cell surface hydrophobicity and membrane integrity, and L. monocytogenes virulence.
CtaP was found to facilitate bacterial growth in the presence of low concentrations of free cysteine, suggesting that CtaP functions as part of a high-affinity cysteine transport system. The requirement for CtaP can be almost fully eliminated by high concentrations of free cysteine (Fig. 3). These data indicate that L. monocytogenes must be capable of cysteine acquisition through an alternative channel or transport system. Thus far, proteins involved in L. monocytogenes cysteine transport have not been described, however multiple systems involved in the transport of cysteine or cystine (cysteine dipeptide) have been reported in other bacteria (Baptist and Kredich, 1977; Berger and Heppel, 1972; Burguiere et al., 2004; Ewann and Hoffman, 2006). B. subtilis and S. enterica serovar Typhimurium each contain three cystine transporters, and E. coli has two cystine transport systems, one that is specific for cystine and one that is non-specific (Baptist and Kredich, 1977; Berger and Heppel, 1972; Burguiere et al., 2004). Two cysteine-transport systems have been described in L. pneumophila and based on kinetic studies, one system functions as a higher-affinity transporter, whereas the other system has lower affinity and transports cysteine only when concentrations are high (Ewann and Hoffman, 2006). Similar to L. monocytogenes, cysteine is required for L. pneumophila growth in broth culture, but cysteine supplementation is not required for intracellular growth in host cells, indicating the ability of the bacteria to acquire necessary amino acids directly from the host (Wieland et al., 2005). It is possible that multiple cysteine transport systems exist in L. monocytogenes as the genome contains several genes whose predicted products are homologous to both cysteine-binding and cysteine export proteins. High concentrations of free cysteine are not present in the cytosol (de Graaf-Hess et al., 1999), which suggests that the robust growth of the ΔctaP::erm mutant in this environment is due to bacterial use of host-derived peptides. Peptides have been previously implicated as a preferred source of amino acids for cytosolic L. monocytogenes (Marquis et al., 1993).
CtaP was also found to contribute to the adhesion of L. monocytogenes to host cells. Uptake of the ΔctaP::erm mutant was reduced in non-professional phagocytic cells but not in professional phagocytes (Fig. 4 and Fig. 5). We have demonstrated that CtaP can function directly as an adhesin, but it has not yet been determined whether it acts as part of an adhesive complex (with, for example, the downstream encoded permease proteins). In addition to functioning as an adhesin, CtaP's contribution to bacterial surface hydrophobicity and membrane integrity may also serve to enhance cellular adhesion. Based on its role in cysteine transport, CtaP could potentially bind cysteine-containing molecules on the surface of the host cell to promote adhesion prior to bacterial invasion. Preliminary data indicates that the preincubation of wild type L. monocytogenes with free cysteine prior to infection of epithelial cell monolayers reduced bacterial adhesion approximately two-fold (data not shown). Previously, peptide and sugar-binding permeases of both known and unknown substrate specificity have been described to mediate adherence in different Streptococcal species (Cundell et al., 1995; Jenkinson and Easingwood, 1990; Jenkinson, 1992; Kolenbrander et al., 1994; Russell et al., 1992; Sutcliffe et al., 1993). For L. monocytogenes, it is possible that host cell adhesion could be mediated via a bridge formed between a host surface molecule, membrane associated or secreted CtaP, and interaction of CtaP with its associated permeases.
It is not yet clear how CtaP contributes to bacterial membrane integrity or surface hydrophobicity, but preliminary data indicates that the protein is not present in greater abundance in the membrane or supernatant fractions of cells grown in broth culture than most other L. monocytogenes surface and/or secreted proteins (data not shown). Oligopetide permease proteins in V. furnissii, Mycobacterium tuberculosis, and E. coli have been reported to affect sensitivity to antimicrobial peptides and uptake of fluorescent substrates (Domenech et al., 2009; Wu et al., 2007). It is possible that CtaP is required for the stability of a complex of membrane proteins and that in its absence these complexes are unstable or result in improperly gated channels or pores. If so, the isolation of CtaP's protein partners may help to clarify its role in membrane homeostasis.
Alterations in membrane integrity may also contribute to the increased sensitivity of ΔctaP::erm mutants to acid stress or may indicate a role for CtaP in the transport of other components that alleviate acid stress. The OppA system has been shown to be important for survival of L. monocytogenes at low temperatures and it is speculated that the OppA system may be involved in the uptake of peptides that function as cryoprotectants (Borezee et al., 2000). L. monocytogenes possesses a number of systems involved in the uptake or processing of compounds involved in acid, osmostic, and cold stress responses, including gad (glutamate decarboxylase), BetL and Gbu (both glycine betaine transport systems), OpuC (carnitine transport system), and cspD and ltrC which encode proteins involved in cold stress tolerance (Beumer et al., 1994; Chaturongakul et al., 2008; Wemekamp-Kamphuis et al., 2004). Although dedicated transport systems exist for the uptake of protective compounds, it is possible that other transport systems provide some functional redundancy by transporting compounds that also have protective function (such as the previously mentioned example of the peptide transporter OppA and cold resistance). Alternatively, free L-cysteine has been reported to serve as a signal for changes in Salmonella typhimurium swarming motility and antibiotic resistance (Turnbull and Surette, 2008), thus it is possible that CtaP-dependent transport of cysteine could serve as a signal to induce physiological changes in L. monocytogenes that contribute to acid resistance, membrane integrity, or even cell adhesion.
Experimental evidence indicates that the loss of CtaP affects multiple aspects of L. monocytogenes pathogenesis, resulting in significant attenuation of bacterial virulence. A model summarizing the contributions of CtaP identified thus far to various aspects of L. monocytogenes host infection is presented in Fig. 10. With respect to the cysteine acquisition-associated function of CtaP and its potential role in bacterial virulence: while the loss of CtaP does not appear to affect bacterial growth in the cytosol of infected host cells, it is possible the loss limits bacterial extracellular growth in other body sites that may be poor sources for peptide acquisition. Extracellularly replicating L. monocytogenes have been reported in the gall bladder and blood, and may exist within other body sites as well (Drevets, 1999; Drevets et al., 2004; Hardy et al., 2004; Hardy et al., 2009). Loss of CtaP could potentially reduce bacterial colonization in the gastrointestinal tract through reduced bacterial adhesion to epithelial cells, which in turn would reduce bacterial invasion and translocation across the intestinal epithelium. The alterations in acid sensitivity and membrane integrity associated with the loss of CtaP presumably decrease bacterial survival in the stomach and intestine. For infections initiated by intravenous inoculation, bacterial attenuation could reflect increased susceptibility of the mutant strain to PMN-mediated killing, however the mutants were able to survive within primary macrophages. Taken together, the combined multiple functions associated with CtaP would serve to significantly enhance the ability of L. monocytogenes to proliferate and cause disease within an infected host.
L. monocytogenes and E. coli strains used in this study are listed in Table 1. L. monocytogenes 10403S (serotype 1/2a) is streptomycin resistant and was the parent strain used in the construction of all mutants. E. coli alpha-select (Bioline, Boston, MA), One Shot TOP10 (Invitrogen, Carlsbad, CA), and SM10 were used as host strains for maintenance and propagation of recombinant plasmids. L. monocytogenes and E. coli strains were grown at 37°C in brain heart infusion (BHI) media (Difco Laboratories, Detroit, MI) and Luria broth (LB) (Invitrogen Corp., Grand Island, NY). Bacteria containing the temperature-sensitive shuttle vector pKSV7 (Smith and Youngman, 1992) or its recombinant plasmids were maintained in E. coli using 50 μg/ml carbenicillin and 10 μg/ml chloramphenicol in L. monocytogenes. Maintenance of the integration plasmid pPL2 (Lauer et al., 2002) was selected for using 25 μg/ml in E. coli and 7.5 μg/ml in L. monocytogenes. Erythromycin selection of L. monocytogenes containing an Em-marked deletion was used at a concentration of 5 μg/ml. Streptomycin 200 μg/ml was used in selection of L. monocytogenes following bacterial conjugation and isolation from tissue organs of infected mice. Where indicated, Congo red (Sigma-Aldrich, St. Louis, MO) was added to BHI agar at a concentration of 100 μg/ml.
An in-frame deletion of ctaP was generated by cloning 600 base pairs of the immediate upstream and downstream regions of ctaP into the temperature-sensitive shuttle plasmid pKSV7 (Smith and Youngman, 1992), leaving only the translational start and stop of the ctaP open reading frame. The flanking regions were amplified using SOEing PCR with primer pairs 5’AAAACTGCAGGGTGAAAATAGAATTTAT’3 and 5’GAATGATTTTTAATTATTAGGTACCCATCTAACAAAAACTCCCC’3 for amplification of the upstream fragment, and 5’GGGGAGTTTTTGTTAGATGGGTACCTAATAATTAAAAATCATTC’3 and 5’GCGTCTAGAAAGAGTCGACCCCAATCG’3 for generation of the downstream fragment. Letters in bold indicate restriction enzyme site added to primer pairs. Internal primers contain a 20 bp overlap region to facilitate SOEing PCR and also a KpnI restriction site. The 1216 bp SOEing PCR product was digested with PstI and XbaI and ligated into the temperature-sensitive shuttle vector pKSV7 to generate a ctaP deletion construct. The ermB gene encoding erythromycin (Em) resistance and containing its native promoter was PCR amplified from pHY-304 (a kind gift from Dr. Craig Rubens and Dr. Amanda Jones, Seattle Children's Hospital Research Foundation) using primer pairs 5’CGGGGTACCTACAAATCACTTATCAC’3 and 5’GCGGGTACCTTATTTCCTCCCGTTAAATAA’3 and inserted in between the internal KpnI site of the flanking regions in the pKSV7-based ctaP deletion construct described above. The resulting ΔctaP::erm plasmid construct was transformed into the L. monocytogens wild-type strain 10403S. Deletion mutants were selected based on EmR and CmS, and confirmed by PCR amplification of products from L. monocytogenes chromosomal DNA using the erm gene primers in combination with primer pairs, 5’CGGGTAATTGTGTGTAAAAGGTAACG’3 and 5’GGATAGACCTTTCGCATAACCAAA’3, which are located outside the regions used for cloning.
The conjugative plasmid, pPL2, was used for complementation of the ΔctaP::erm mutant (Lauer et al., 2002). This plasmid integrates in single copy into the L. monocytogenes 10403S chromosome at a phage attachment site within the tRNAArg gene following conjugation. For construction of the ΔctaP::erm complementation vector, the entire open reading of ctaP along with the 331 bp region upstream and 70 bp downstream of ctaP was PCR amplified from 10403S genomic DNA using primers 5’GCGGAGCTCAATAATTTTTGGGATTTCAG’3 and 5’GCGCCCGGGATTTATCTCCTATCT’3. The PCR fragment and pPL2 vector were digested with SacI and SmaI and subsequently ligated together and transformed into E. coli SM10. The resulting construct was conjugated into the ΔctaP::erm mutant and transconjugants containing chromosomally-integrated pPL2-ctaP were selected for chloramphenicol resistance.
The inlAm allele, as described by Wollert et al (Wollert et al., 2007), was generated by site directed mutagenesis with overlap extension (SOEing) PCR. Briefly, primer pairs Marq403/405, Marq404/407, Marq406/408 served to amplify DNA from regions upstream and downstream of the mutation sites respectively, using strain 10403S genomic DNA as template. The three PCR products and primer pair Marq403/408 were used for SOEing PCR. The final DNA fragment was digested with Xba1 and cloned into pKSV7, creating plasmid pHEL-913. The inlAm allele was introduced into the appropriate L. monocytogenes strains by allelic exchange. Mutant clones were identified by amplifying the inlAm allele from genomic DNA by PCR using primer pair Marq403/408 and digesting the DNA fragment with Hinf1 and Taq1, as a Hinf1 restriction site was gained by creating the S192N mutation and a Taq1 restriction site was gained by creating the N369S mutation. The strains containing the inlAm allele were designated NF-L1556 (10403S background), NF-L1558 (with ΔctaP::erm) and NF-L1662 (with pPL2-ctaP complemented ΔctaP::erm). The primer sequences are: Marq 403, 5’CAGATCTAGACCAAGTTACAA’3; Marq 404, 5’CAGCAATTAAATTTTGGGAATCAAGTG’3; Marq 405, 5’CACTTGATTCCCAAAATTTAATTGCTG’3; Marq 406: 5’TTTTCTCGAATAACAAGGTAAGTGAC’3; Marq 407: 5’GTCACTTACCTTGTTATTCGAGAAAA’3; and Marq 408: 5’CAGATCTAGAATAGTGACAGGTTGGCTAA’3.
Tranduction of the ΔctaP::erm mutation into different background strains was performed as previously described (Hodgson, 2000; Shetron-Rama et al., 2003). Briefly, high titer U153 phage lysates were prepared from the ΔctaP::erm mutant and mixed with mid-exponential phase L. monocytogenes grown at 30°C in the presence of 10 mM MgSO4 and 10 mM CaCl2. The mixture was incubated at room temperature for 40 minutes and 3 ml of warm LBSA (LB + 0.75% agar + 10 mM MgSO4 and 10 mM CaCl2) was added to the bacteriophage/L. monocytogenes suspension, followed by plating on Em 5 μg/ml BHI agar plates for selection of ΔctaP::erm transductants. Plates were incubated at 37°C and Em-resistant transductants were isolated and confirmed by PCR.
For generation of purified CtaP, the coding sequence of ctaP was PCR amplified from wild type 10403S using primer pairs 5’CACCATGAAGAAATTTTTATTAGTAGCGG’3 and 5’TTATTCCGTTAAATATAATTTAGAAAGATCACG’3 and ligated into a pET100 Directional TOPO Expression vector (Invitrogen Corp., Grand Island, NY), which contains an N-terminal 6x-histidine tag and an isoproyl-β-D-thiogalactopyranoside (IPTG) inducible promoter. Letters in bold indicate the translational start and stop of the ctaP coding sequence. The resulting construct was isolated using E. coli TOP10 cells, followed by transformation into BL21 Star (DE3). An overnight culture containing the expression construct was diluted 1:100 in fresh LB broth and the culture was incubated at 37°C (with shaking) until an optimal density of 0.5 was reached. To induce expression of the CtaP protein, 1 mM IPTG (Inalco, Milano, Italy) was added to the culture and induction was allowed to proceed for 3 to 4 hours. The bacterial cells were recovered by centrifugation followed by sonication with 4 repeated 10 second bursts and 1 minute cooling on ice. The soluble fraction containing the N-His-CtaP protein was collected and purified using the His-Pur Purification Kit (Thermo Scientific, Rockford, IL). Protein concentration was determined using a BCA Protein Assay Kit (Thermo Scientific, Rockford, IL).
HTM minimal media was prepared as described (Tsai and Hodgson, 2003). Where indicated, specific amino acids (Sigma-Aldrich, St. Louis, MO) were added to the HTM media at concentrations of 0.1, 0.5 or 1.0 mg/ml, or supplemented with 0.1g/ml of Bacto-Tryptone (Difco Laboratories, Detroit, MI), or 1 mM or 4 mM dithiothreitol (DTT) (Sigma-Aldrich, St. Louis, MO). For preparation of bacteria samples for HTM growth curves, overnight cultures of strains were grown in BHI at 37°C with shaking. Samples were normalized to OD 600, pelleted, washed 2X in PBS before resuspension in PBS to a final volume equal to the normalized culture volume. Bacterial suspensions were diluted 1:20 in HTM and growth in HTM media containing the various supplements was measured at OD600 at indicated time points.
Bacterial intracellular growth assays in murine J774 macrophage-like cells and Potoroo tridactylis kidney epithelial cells (PtK2) were performed as previously described (Alonzo et al., 2009; Marquis et al., 1995; Mueller and Freitag, 2005; Wong and Freitag, 2004). In brief, monolayers of cells were grown on glass coverslips to confluency and infected with bacterial strains with an MOI of 0.1:1 for J774 cells and 100:1 for PtK2 cells. Thirty minutes post-infection of J774 cells and 1 hour post-infection of PtK2 cells, monolayers were washed 3X in PBS and 5 μg/ml of gentamicin was added to kill extracellular bacteria. At indicated time points, coverslips were removed and lysed in 5 mls of sterile H2O to release intracellular bacteria for enumeration of intracellular growth or were processed for microscopy. For adhesion assays, no gentamicin was added and coverslips were washed in PBS to remove non-adherent bacteria prior to lysis.
For infection of murine bone-marrow derived macrophages (BMMØ), macrophages were isolated from the femurs of mice and maintained as described (Portnoy et al., 1988) and intracellular growth was performed with an MOI of 0.1:1 as described above for J774 cells. Where indicated, 1 ηg/ml of recombinant mouse IFN-γ (Invitrogen Corp., Carlsbad, CA) was added to monolayers of BMMØ 24 hours prior to infection to activate the macrophages.
Plaque assays were conducted as previously described (Sun et al., 1990). Briefly, murine L2 fibroblasts were grown to confluency in 6-well microtiter plates and infected with 20 μl of a normalized 1:20 dilution of overnight culture grown at 37°C in BHI with shaking (MOI 10:1). One hour post-infection, L2 infected monolayers were washed and 5 μg/ml of gentamicin was added to kill extracellular bacteria. Three days post-infection, Neutral Red (Sigma-Aldrich, St. Louis, MO) was added and plaques were visualized and measured using a micrometer (Finescale, Orange County, CA).
His-purified CtaP or Bovine Serum Albumin (BSA) (Fisher Scientific, NJ) was covalently coupled to red fluorescent latex beads (Invitrogen Molecular Probes, Eugene, OR), as described by Lecuit et al (2007). To test the adhesive properties of CtaP- or BSA-coated latex beads, beads were incubated with a monolayer of PtK2 cells grown on glass coverslips at a bead to cell ratio of 1000:1 for one hour at 37°C. Coverslips were washed by dipping 3X in PBS followed by mounting with PermaFluor (Thermo Electron Corp., Pittsburg, PA) onto glass slides. Coverslips containing bound latex beads were visualized on a Leica DM4000B wide-field epifluorescence microscope (Leica Microsystems, Wetzlar, Germany). A total of fifteen different fields were visualized and the number of adherent beads in each field was enumerated.
All procedures were IACUC approved and preformed under the strict guidelines of the BRL animal facilities at the University of Illinois at Chicago. For intravenous tail vein injections, Listeria strains were grown in BHI with the appropriate antibiotics, with shaking at 37°C, to an OD600 of ~0.6. Bacterial cells were pelleted, washed in PBS, and resuspended in PBS to a final concentration of 1 × 105 CFU/ml. Eight- to 10-week old female Swiss Webster mice (Harlan Laboratories, Indianapolis, IN) were infected by tail vein injection with 0.2 ml of bacteria culture containing 2 × 104 CFUs. At 6, 48 and 72 hours post-infection, mice were sacrificed and the livers and spleens were harvested. Organs were placed in 5 ml of sterile water and tissues were homogenized using a Tissue Master 125 (Omni International, Marietta, GA). 10-fold serial dilutions were plated on BHI agar plates containing 200 μg/ml streptomycin (Sigma-Aldrich, St. Louis, MO) for enumeration of bacterial loads in each organ.
For intragastric infections, pHEL-913, the pKSV7 allelic exchange vector containing the inlAS192N-Y369S substitution, was used to introduce the inlA mutation into the 10403S parent strain. These two amino acid substitutions in InlAS192N-Y369S increase the binding affinity of InlA to murine E-cadherin and thereby increase the efficiency and reproducibility of L. monocytogenes oral inoculations in mice (Wollert et al., 2007). Allelic exchange of the inlAS192N-Y369S mutant copy (designated inlAm) with wild-type inlA was confirmed by restriction digest and sequencing using primer pairs 5’CAGATCTAGACCAAGTTACAA’3 and 5’CAGATCTAGAATAGTGACAGGTTGGCTAA’3, which amplify the inlA coding region. The ΔctaP::erm mutation was then introduced the inlAm strain using bacteriophage transduction as described above. This strain was also complemented with the pPL2-ctaP plasmid vector. For oral inoculations, Listeria strains were grown in BHI with the appropriate antibiotics, with shaking at 37°C, to an OD600 of ~0.6. Bacterial cells were pelleted, washed 2X in PBS, and resuspended in PBS to a final concentration of 5 × 108 CFU/ml. Groups of 6- to 8-week old C57BL/6 mice (Harlin Laboratories, Indianapolis, IN) were orally inoculated with 0.2 ml of bacterial culture containing 1 × 108 CFUs of either wild-type 10403S or the ΔctaP::erm mutant and 20 mg CaCO3 using an 18 gauge feeding tube (Solomon Scientific, Plymouth Meeting, PA) attached to a 1 ml syringe. Food was removed 2 - 4 hours prior to infection. At 72 hours post-infection, the mice were sacrificed and the stomach, intestine, livers, and spleens were harvested to determine bacterial CFUs in these organs as described above. Statistical analyses were done using a two-tailed unpaired student t-test (GraphPad V.5.0A).
BHI (Difco laboratories, Detroit, MI) was adjusted to a pH of 5.5 with 1M HCl or supplemented with 5% NaCl to mimic conditions of high osmolarity. Bacterial strains were grown overnight in BHI, normalized and diluted 1:20 in the adjusted BHI media. At indicated time points, bacterial growth was measured at OD600.
For analyses of bacterial surface hydrophobicity, wild type, ΔctaP::erm mutant and complemented mutant strains were streaked onto BHI agar plates containing either 50 or 100 μg/ml of Congo Red (Sigma-Aldrich, St. Louis, MO) and incubated at 37°C for 72 hours. Differences in apparent bacterial hydrophobicity were also measured by assessing bacterial adherence to untreated polystyrene as described by Boruki et al (ref) for biofilm formation in 96-well microtiter plates. Plates were incubated at 37°C for 24, 48, and 72 hours prior to measurement of bound cells by crystal violet staining. Membrane permeability differences between the varying strains were assessed using the LIVE/DEAD BacLight Bacterial Viability Kit (Invitrogen Corp., Carlsbad, CA). This system utilizes a mixture of fluorescently labeled nucleic acid stains that are both membrane impermeant (propidium iodide) and permeant (SYTO9). Live cells with intact membranes will incorporate only SYTO9 and fluoresce green, whereas dead cells or cells with compromised membranes will incorporate PI and fluoresce red. Bacterial cells were grown to mid-exponential phase, 1 mL of culture volume was normalized to OD600 of 0.6, washed 2X in PBS, and resuspended in 1 mL of PBS. Bacterial suspensions were then stained with the nucleic acid dyes as per manufacturer's protocol. Five microliters of stained bacteria were spotted onto to glass coverslips and visualized on a Leica DM4000B wide-field epifluorescence microscope (Leica Microsystems, Wetzlar, Germany). A total of 1764 wild type bacteria and 1638 complemented mutant strain were counted in ten different fields with only 1% of the population incorporating the membrane impermeant propidium iodide (PI) dye. A total of 1184 ΔctaP::erm mutant bacterial cells were counted in the ten different fields with 8% incorporating PI and 26% incorporation both dyes (thus appearing orange/yellow) equaling a total of 37%. Ten different fields were viewed for each strain in three independent experiments.
We thank Dr. Craig Rubens and Dr. Amanda Jones for the kind of pHY304 used as a source of erm, and Francis Alonzo III for assistance in the animal infections. We thank all members of the Freitag lab for their helpful comments and discussions. This work was supported by Public Health Service grants AI41816 (N.E.F) and AI152154 (H. M.) from NIAID and a National Science Foundation Graduate Research Fellowship (NSF-GRF) (G.C.P.). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the funding sources.