Effects of Psa and F1 on the Adhesive and Invasive Interactions of Yer
http://www.100md.com
《感染与免疫杂志》
Department of Pathobiology, University of Pennsylvania School of Veterinary Medicine, 3800 Spruce Street, Philadelphia, Pennsylvania 19104
ABSTRACT
Yersinia pestis, the causative agent of plague, expresses the Psa fimbriae (pH 6 antigen) in vitro and in vivo. To evaluate the potential virulence properties of Psa for pneumonic plague, an Escherichia coli strain expressing Psa was engineered and shown to adhere to three types of human respiratory tract epithelial cells. Psa binding specificity was confirmed with Psa-coated polystyrene beads and by inhibition assays. Individual Y. pestis cells were found to be able to express the capsular antigen fraction 1 (F1) concomitantly with Psa on their surface when analyzed by flow cytometry. To better evaluate the separate effects of F1 and Psa on the adhesive and invasive properties of Y. pestis, isogenic caf (F1 genes), psa, and caf psa mutants were constructed and studied with the three respiratory tract epithelial cells. The psa mutant bound significantly less to all three epithelial cells compared to the parental wild-type strain and the caf and caf psa mutants, indicating that Psa acts as an adhesin for respiratory tract epithelial cells. An antiadhesive effect of F1 was clearly detectable only in the absence of Psa, underlining the dominance of the Psa+ phenotype. Both F1 and Psa inhibited the intracellular uptake of Y. pestis. Thus, F1 inhibits bacterial uptake by inhibiting bacterial adhesion to epithelial cells, whereas Psa seems to block bacterial uptake by interacting with a host receptor that doesn't direct internalization. The caf psa double mutant bound and invaded all three epithelial cell types well, revealing the presence of an undefined adhesin(s) and invasin(s).
INTRODUCTION
Since the last plague pandemic at the end of the 19th century, its bacterial agent, Yersinia pestis, has been maintained in rodents in several Asian, African, and American countries, including the United States (17, 40). Bubonic plague results from the transmission of Y. pestis by flea bites. In contrast, primary pneumonic plague is acquired when a mammalian host inhales particles or aerosols carrying Y. pestis. Although plague is currently not a major public health problem in developed countries and has been suggested to be less contagious than commonly believed (32), the spread of Y. pestis by aerosols could cause a cluster of human cases of primary pneumonic plague with potential amplification of the outbreak (27).
The major adhesins and invasins of enteropathogenic Yersinia pseudotuberculosis and Yersinia enterocolitica (YadA, Ail, and Inv) are not expressed by Y. pestis strains (15, 45, 51). Thus, how Y. pestis attaches to and translocates through the epithelial layer of the respiratory tract to reach deeper tissues and the bloodstream following airborne transmission remains unknown. Interestingly, Y. pestis exhibits an extensive extracellular lifestyle due to the intracellular delivery of several antiphagocytic effector proteins by its type III secretion system (T3SS-1 or Yops regulon) (5-7, 14, 54, 56). Moreover, two antigenic surface structures exported by usher-chaperone proteins characteristic of fimbrial biogenesis systems have been shown to inhibit Y. pestis uptake by macrophages (19, 26). Each of these structures, designated capsular antigen fraction 1 (F1) and pH 6 antigen (Psa), consists of the homopolymeric association of a single-subunit protein (Caf1 and PsaA, respectively) on the bacterial surface (36, 59). PsaA can assemble into distinct thin individual fimbrial strands, bundles of fimbriae associated along their lengths, or structureless aggregates, as shown by electron microscopy (12, 36). The F1 capsular antigen is made of linear fibers of Caf1 subunits that cannot be detected as fimbrial organelles (59), not unlike the Dr adhesins that were designated afimbrial adhesins (2).
Although the expression of redundant antiphagocytic virulence factors suggests that Y. pestis remains and replicates mainly extracellularly, Y. pestis has been reported to invade human epithelial HeLa cells (16) and to be phagocytosed by macrophages, where it can replicate (9, 11, 31, 36, 37, 41, 52). It has been suggested that intracellular survival and replication occur mainly during the early stages of host colonization and invasion (42). Phagocytosed Y. pestis isolates have also been found to have cytotoxic and apoptotic effects on macrophages (22, 60).
Cultured epithelial cell lines have been used previously to investigate the epithelioadhesive property of Y. pestis and the involvement of Psa (16, 58). Here, we determined whether Y. pestis adheres to human respiratory tract epithelial cells and possibly penetrates these cells. Whether and how the F1 and Psa surface structures modulate such interactions were the focus of this study.
MATERIALS AND METHODS
Bacterial strains and constructed plasmids. Escherichia coli strains, Y. pestis strains, and plasmids used in this study are listed in Table 1. E. coli strains were grown overnight at 37°C in LB broth (Lennox) with appropriate antibiotics, when required, using a microbial culture roller drum. Y. pestis strains were grown overnight at 26°C in brain heart infusion (BHI) broth, diluted 1:20 in fresh BHI broth, and cultured overnight at 37°C using the roller drum. Under these conditions, the pH of the spent broth dropped to 6.2 to 6.4, resulting in the expression of both the F1 and Psa structures on the bacterial surface. Some cultures of Y. pestis were prepared in BHI broth buffered with 0.1 M MES (morpholineethanesulfonic acid) (pH 6) and grown overnight at 37°C to ensure optimal pH 6 antigen expression. F1 was also expressed under these growth conditions.
Cloning and expression of Psa and F1. The gene cluster encoding the structural genes of Psa (psaABC) was cloned into pBR322. For this, a DNA fragment starting from the ribosomal binding site upstream of the psaA ATG start codon and ending at the stop codon of psaC was amplified by PCR using genomic DNA from strain KIM6 as template, primers 5'-CTAGCTAGCATTTAATTGCCCAATATCACC-3' and 5'-ACGCGTCGACAAACCTGGCGGGAATTGAACC-3' (restriction sites used are underlined), and the thermostable DNA polymerase blend of the Expand Long Template PCR system (Roche Applied Sciences, Indianapolis, IN). The NheI- and SalI-restricted amplicon was inserted into the corresponding sites of pBR322 to create pCS267, taking advantage of the constitutive tetracycline promoter to drive the expression of Psa. The chloramphenicol resistance gene of pACYC184 was excised as an AhdI-XmnI fragment that was blunted and inserted into the blunted AhdI site of the ampicillin resistance gene of pCS267, resulting in plasmid pCS334. The gene cluster encoding the structural genes for the F1 capsular antigen or fimbriae (caf1M-caf1A-caf1) was cloned into pET16b. For this, a DNA fragment starting from the caf1M ATG start codon and ending at 17 nucleotides downstream of the stop codon of caf1 was amplified by PCR using primers 5'-GCTATTCCGTCTCGCATGATTTTAAATAGATTAAGTACGTT and 5'-ACGCGTCGACTTATCTATATGGATTATTGGTTAGA. The BsmBI- and SalI-restricted amplicon was inserted into the restricted NcoI and XhoI sites of pET16b to create pCS274. The XbaI and BamHI fragment of pCS274 that includes the ribosomal binding site and the F1 genes was subcloned into NheI- and BamHI-restricted pLG339 to create plasmid pCS331. The chloramphenicol resistance gene described above was inserted into the SmaI site of the kanamycin resistance gene of pCS331, resulting in plasmid pCS332. The PCR-cloned DNAs corresponding to the surface-expressed fimbrial proteins Caf1 and PsaA were sequenced and found to correspond to the published sequences for strain KIM, indicating the absence of PCR-mediated mutations. Expression of the Psa and F1 proteins on the surface of E. coli was confirmed by negative staining for transmission electron microscopy (see below) and sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE). The presence of Psa was further demonstrated by hemagglutination with sheep red blood cells.
Construction of Y. pestis mutants. The caf, psa, and caf psa mutants were constructed by allelic exchange, as described previously (20). To construct a Y. pestis psa deletion mutant, two DNA fragments of 1 to 1.3 kb flanking the psaABC locus were amplified by PCR with primer pairs 5'-CTAGTCTAGACAAACAGCAATGAAAGCAAAATCACT-3' and 5'-GGCCGAGCTCCGAGAACAGTCTCCA-3' and 5'-TTCTCGGAGCTCGGCCAGGCATCCGAGATGAC-3' and 5'-CGGGGGTACCTCTGGGTTAATCATCTGATTAC-3', respectively. These two PCR fragments were joined by splicing by overlap extension (SOEing) (24) to produce a 2.3-kb amplicon with a central unique SacI restriction site. The designed flanking restriction sites XbaI and KpnI were used to clone the amplicon in the corresponding sites of pDMS197, generating pLS013. The SacI-restricted kanamycin resistance cassette of pBSL128 (1) was then inserted into the corresponding site in pLS013, generating suicide plasmid pLS014. The same approach was used to engineer a caf1M-caf1A-caf1 deletion mutant. For this, a 2.05-kb PCR product was constructed by SOEing with primers 5'-CTAGGAGCTCCAGGCAAAGAGTTATAATATAT-3', 5'-CGCGGGGCCCGAGCTTAACCTCCT-3', 5'-AAGCTCGGGCCCCGCGTCCATATAGATAATAG-3', and 5'-CGGGGGTACCTTACCTTTACTTCAGAGATATG-3', creating a central unique ApaI restriction site. The 2.05-kb amplicon was inserted into pDMS197, generating plasmid pLS007. An ApaI-digested ampicillin resistance cassette of pBSL159 (1) was inserted into pLS007, generating suicide plasmid pLS009. To delete caf or/and psa genes from Y. pestis, 10 μg of plasmid DNA was electroporated into Y. pestis cells (100 μl; 1011 cells/ml). Transformed cells were allowed to recover for 2 to 4 h in BHI broth at 26°C and then plated onto BHI agar plate with tetracycline (10 μg/ml) to select for clones with integrated plasmids. Clones were subcultured at 26°C for 18 h in NaCl-free LB broth with ampicillin (200 μg/ml) or kanamycin (45 μg/ml) to allow for spontaneous plasmid loss and plated onto NaCl-free LB agar with and without 5% sucrose to select for tetracycline-sensitive and sucrose-resistant clones. For strain DSY50, the resistance gene markers were removed in two steps, using pLS007 and pLS013 sequentially as described above. Isolates were selected for sucrose resistance and screened for susceptibility to the respective antibiotic. All the gene deletions were confirmed by PCR using appropriate primers.
Isolation of F1 and Psa proteins and antibody production. F1 or Psa fimbriae expressed on the bacterial surface were prepared by heat extraction of strain KIM6 or recombinant Psa- or F1-expressing E. coli SE5000, as described previously (25, 29, 30). Briefly, bacteria were pelleted by centrifugation, resuspended in 0.5 mM Tris-HCl (pH 7.4)-75 mM NaCl, and treated at 60°C for 30 min. After centrifugation, ammonium sulfate was added to the supernatant to a 30% final concentration to precipitate the extracted proteins overnight on ice. After centrifugation (12,000 x g, 30 min), the pellets were resuspended in phosphate-buffered saline (PBS), and excess ammonium sulfate was removed by dialysis. The purity of the proteins was confirmed by SDS-PAGE and Coomassie blue staining (>90% pure). An anti-Psa antibody was prepared in rabbits with isolated recombinant Psa protein by using a conventional immunization protocol (Cocalico Biologicals Inc., Reamstown, PA). The serum was adsorbed with E. coli SE5000. The antibody had a specific enzyme-linked immunosorbent assay titer of 105.
Binding and internalization assays. Human cell lines used in this study include type II alveolar epithelial cell line A549 (ATCC CCL185), pharynx pleural effusion carcinoma cell line Detroit 562 (ATCC CCL138), and type I alveolar epithelial cell line WI26 VA4 (ATCC CCL95.1). The cells were routinely maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37°C in a humidified atmosphere of 5% CO2 in air. The cells were grown in 6- or 24-well plates to confluent monolayers and washed three times with PBS, and DMEM without FBS was added. Bacteria grown as described above were washed with PBS, retaken in DMEM without FBS, and added to the cells at a multiplicity of infection (MOI) of 10:1 to 50:1. Preliminary studies showed that the bacteria bound significantly better when bacterial binding was studied in DMEM without FBS (as described above) instead of PBS. The cell monolayers were incubated at 37°C for 1 h before being washed three times with PBS. For the adhesion assays, the epithelial cells were lysed with cold 0.1% Triton X-100 in deionized H2O, and 10-fold-serially-diluted lysates were spread onto agar plates to determine CFU counts. For the internalization assays, the cell monolayers were incubated for an additional 2 h in 1 ml (24-well plate) or 2 ml (six-well plate) medium per well, with 50 μg/ml gentamicin. After three steps of washing with PBS, the cell monolayers were lysed for determining CFU counts as described above. Adherence percentages were calculated as the numbers of cell-associated bacteria divided by the total numbers of inoculated bacteria x 100. Internalization percentages were calculated as the numbers of internal bacteria divided by the numbers of inoculated bacteria x 100. Invasion indices were calculated as the numbers of internal bacteria divided by the numbers of cell-associated bacteria x 100. Student's t test was used to calculate statistical significance.
Coating and binding of beads. Polystyrene beads (1-μm diameter) (Polybead; Polysciences, Inc., Warrington, PA) were washed three times in Tris-HCl buffer (5 mM, pH 7.0) and resuspended in PBS, and 4.7 x 109 beads were mixed with isolated Psa or F1 fimbriae or with BSA (60 μg/ml) in a total volume of 0.5 ml on a rotating wheel at room temperature overnight. The beads were centrifuged (14,000 rpm, 5 min), washed three times in PBS, resuspended in PBS containing 0.1% BSA, and briefly sonicated to disperse bead aggregates. Successful Psa or F1 coating was confirmed by agglutination with the corresponding antibodies. The bead-binding assay (five beads per cell) was undertaken with the A549 cells as described above for the bacterial adhesion assay. Student's t test was used to calculate statistical significance.
Microscopy. Bacterial binding to cells was examined after Giemsa staining by using bright-field microscopy. The binding of polystyrene beads to A549 cells was visualized by phase-contrast microscopy. All images were captured with a Coolsnap digital camera (Photometrics, Tuscon, AZ) mounted onto a Nikon Eclipse E600 microscope with Coolsnap version 1.2.0 software. The number of cell-associated beads was counted for at least 100 cells to calculate means (μ) and standard errors. Negatively stained Psa- and F1-expressing bacteria were examined by electron microscopy (Joel 1010 transmission electron microscope). Images obtained with the Jeol JEM 1010 transmission electron microscope were captured by using an AMT 12-HR-aided Hamamatsu charge-coupled-devise camera (46).
Flow cytometry. Bacteria were grown overnight in BHI broth at 37°C. The bacterial cells were washed once with PBS and labeled with rabbit anti-Psa antibody and monoclonal mouse anti-F1 antibody (Advanced ImmunoChemicals Inc., Long Beach, CA), followed by fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulin G (ICN Biomedicals, Inc., Aurora, Ohio) and phycoerythrin-conjugated F(ab')2 goat anti-mouse immunoglobulin G (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). The labeled bacteria were fixed overnight at 4°C in 10% (vol/vol) neutral buffered formalin in PBS, pH 7.4. The analysis by flow cytometry was undertaken with FACScan and CellQuest Pro software (BD Biosciences, San Jose, CA).
RESULTS
Recombinant Psa and F1 capsular antigen expression. E. coli SE5000 transformed with plasmid pCS267, which carries the psaABC genes, formed negatively stained fimbria-like structures as well as aggregates on the bacterial surface when observed by electron microscopy (data not shown). These structures and the additional aggregates surrounding the bacteria were not seen on strain SE5000 lacking pCS267. Heat extraction at 60°C, conventionally used to release fimbrial proteins from bacteria, identified a major protein of 15 kDa by SDS-PAGE (data not shown), corresponding to the known molecular mass of PsaA, the Psa fimbrial subunit (35). This protein demonstrated the typical properties of fimbrial subunits by remaining associated as a large polymer in 1% SDS when not treated at 100°C. Psa-fimbriated bacteria are known to have hemagglutinating properties (4). Consistent with this, only the Psa-fimbriated E. coli SE5000(pCS267), and not E. coli SE5000, agglutinated sheep erythrocytes (data not shown), confirming the successful cloning and expression of recombinant Psa fimbriae in E. coli. Heat extraction of E. coli SE5000 transformed with plasmid pCS274 or pCS331, which carry the caf1M-caf1A-caf1 genes, identified a major protein of 16 kDa by SDS-PAGE (data not shown), corresponding to the known molecular mass of the F1 protein subunit Caf1 (21, 55). As found with PsaA, Caf1 remained associated as a large polymer in 1% SDS when not treated at 100°C.
Psa-fimbriated E. coli cells bind to human respiratory tract epithelial cells. In order to investigate whether Psa could mediate bacterial adhesion to respiratory tract epithelial cells, the binding of E. coli SE5000(pCS267) (Psa+) was studied with the human type II alveolar epithelial cell line A549. Microscopy showed that the Psa-fimbriated E. coli cells associated in greater numbers with the A549 cells than the nonfimbriated E. coli (Fig. 1A and B). Colony counts determined that the Psa-fimbriated E. coli cells adhered 13 times better to A549 cells than E. coli SE5000(pBR322) (Fig. 1C). Similar, albeit less striking, results were observed with the human pharyngeal epithelial cell line Detroit 562 and type I alveolar epithelial cell line WI26 VA4 (data not shown). Binding specificity was confirmed by inhibiting the adhesive interaction in a dose-dependent manner with isolated Psa fimbriae (50% inhibitory concentration of 8 μg Psa for a bacterial inoculum of 2 x 107 CFU/well of a six-well plate) or a Psa-specific polyclonal antibody (Fig. 1D).
Psa-coated beads bind to human respiratory tract epithelial cells. To further confirm that the Psa fimbriae are adhesins, Psa-coated polystyrene beads were prepared and used to determine their binding properties with the type II alveolar epithelial cell line A549. Other sets of beads were coated with isolated F1 protein or BSA and studied in parallel for comparative purposes. Microscopic examination showed that the Psa-coated beads bound significantly better to A549 cells (mean ± standard error, 2.50 ± 0.13 beads/cell) than the BSA-coated beads (0.80 ± 0.19 beads/cell) or F1-coated microspheres (0.34 ± 0.5 beads/cell) (Fig. 2). Taken together, the data indicated that Psa acts as a Y. pestis adhesin for human type II alveolar epithelial cells.
Psa is not an invasin. Since the Psa fimbriae were shown to increase the association of recombinant E. coli with A549 human type II alveolar epithelial cells, and since Y. pestis has previously been shown to invade epithelial cells (16), the possibility that the Psa fimbriae modulate the uptake of E. coli by A549 cells was investigated. Internalized bacteria were identified by using a gentamicin protection assay. Although the expression of Psa rendered E. coli significantly more adhesive (Fig. 1C), Psa-fimbriated and nonfimbriated E. coli cells were similarly poorly internalized (Fig. 3A), suggesting that Psa is not an invasin. However, significantly more cell-associated nonfimbriated E. coli cells were internalized than cell-associated Psa-fimbriated E. coli cells (Fig. 3B), suggesting that Psa-mediated E. coli binding had an inhibitory effect on the observed low level of E. coli uptake.
Coexpression of F1 and Psa on individual Y. pestis cells. In order to evaluate the role of the Psa fimbriae in the adhesive and invasive properties of Y. pestis, a psa mutant was constructed. Moreover, because the F1 surface structure of Y. pestis might affect the accessibility of Psa to a host cell receptor, both an isogenic caf mutant and an isogenic psa caf double mutant were constructed and studied in parallel. The phenotypes of the wild-type and mutant strains were examined by electron microscopy and SDS-PAGE (Fig. 4). The wild-type strain and two single mutants were typically surrounded by a thick layer of negatively stained material (Fig. 4A to C). In contrast, most double-mutant cells did not have this negatively stained halo (Fig. 4D), suggesting that it represented the staining of PsaA and/or F1 subunit aggregates, as described previously by others (12, 28, 36). This interpretation was consistent with the protein profile of heat-extracted surface proteins from the four strains, as shown by SDS-PAGE (Fig. 4E). Moreover, double-mutant cells with one or a few fimbria-like structures were observed (Fig. 4D, arrow), suggesting the expression of additional surface structures on Y. pestis. The finding that both Caf1 and PsaA were produced by the wild-type strain in the same culture was in agreement with recently published findings (26) indicating that Y. pestis can express both F1 and Psa surface structures together. To ensure that such cultures did not consist of mixtures of cells that exclusively produce only one of the two surface structures, the expression of F1 and Psa on individual Y. pestis cells was investigated by flow cytometry using antigen-specific antibodies for double labeling. As shown with the mutants that served as controls, both antibodies labeled only their respective antigens (Fig. 5). Most importantly, 80% or more of the wild-type cells expressed both F1 and Psa simultaneously, demonstrating that F1 or Psa expression does not exclude the expression of the other antigen on the bacterial surface.
Reduced bacterial adhesion with a Y. pestis psa mutant. The binding properties of the psa or/and caf mutants were investigated with three types of human respiratory tract epithelial cells, namely, pharyngeal (Detroit 562) and alveolar type I (WI26 VA4) and type II (A549) epithelial cells. In general, for all three cell lines, the psa single mutant bound the least (Fig. 6). When complemented in trans with the psa-containing plasmid pCS334, but not with pBR322, the psa mutant regained 82% of the adhesive properties of the parental KIM6 (psa+) strain towards the WI26 VA4 cells, confirming that Psa is an adhesin for respiratory tract epithelial cells. Interestingly, in contrast to the psa mutant but not unlike the caf mutant, the psa caf double mutant bound well to all three cell lines (Fig. 6). Complementing the caf mutant with the caf-containing plasmid pCS332 did not interfere significantly with the strains' adhesive properties, with binding by the parental strain, the mutant strain, and the complemented strain being essentially the same (<15% differences). In contrast, when the double mutant was complemented with the caf-containing plasmid pCS332, the strains' binding properties towards the WI26 VA4 cells were decreased by 72%, mimicking the reduced adhesiveness of the single psa mutant. These results suggested that in addition to Psa, Y. pestis expresses at least one other adhesin whose adhesive property is detectable only in the absence of F1. Binding to WI26 VA4 cells was not changed significantly after complementing the double mutant with the psa-containing plasmid pCS334 (<19% difference). Taken together, the results of the binding assays with recombinant E. coli and the Y. pestis mutants indicated that Psa+ is a dominant adhesive phenotype and that the binding mediated by a proposed new Y. pestis adhesin(s) is a recessive phenotype that becomes apparent in the absence of Psa and F1.
Improved bacterial internalization with a Y. pestis caf psa mutant. To evaluate the role of the Psa adhesin in the cellular uptake of Y. pestis by respiratory tract epithelial cells, the wild-type and psa and caf mutant strains were studied by a standard gentamicin protection assay. The pCD1-encoded proteins, including the antiphagocytic type III effector proteins, were previously shown to have no significant role in the uptake of Y. pestis by HeLa and WI26 VA4 epithelial cells (16). Nevertheless, to avoid any potential confounding effects by the type III secretion system, all the studied strains were derivatives from the pCD1– strain KIM6. In comparison to the wild-type strain, the psa mutant was internalized 89 times better by A549 cells (Fig. 6). Better bacterial uptake was also detected with the other epithelial cells, as shown in Fig. 6. Complementation of the psa mutant with the psa-containing plasmid pCS334 reduced bacterial internalization by the WI26 VA4 cells to nearly the same level (1.4 times) as that observed for KIM6 internalization, confirming that Psa inhibits Y. pestis internalization by respiratory tract epithelial cells. In general, the caf mutant was also better internalized than the parental strain KIM6 (Fig. 6). However, in contrast to the psa mutant and because of the Psa-mediated binding of the caf mutant, the invasion index of the latter strain showed lower values, since this index relates to the intracellular uptake of bound bacteria (Fig. 6). As observed for the pCS334-complemented psa mutant, complementing the caf mutant with the caf-containing plasmid pCS332 significantly reduced bacterial internalization by the WI26 VA4 cells (from 121 to 3.3 times the value observed for KIM6 internalization), confirming that F1 also inhibits Y. pestis internalization by respiratory tract epithelial cells. Interestingly, the strain that was internalized the most efficiently by all three epithelial cells was the caf psa double mutant (Fig. 6). Complementing the double mutant with the psa- or the caf-containing plasmid reduced Y. pestis internalization approximately four to five times compared to the noncomplemented double mutant. Internalization of these complemented bacteria was comparable to the internalization of the respective single mutants. Taken together, these results indicated that both Psa and F1 inhibit the uptake of Y. pestis by respiratory tract epithelial cells. Moreover, the results also suggested that at least one new surface-exposed invasin was uncovered by the removal of F1 and Psa.
Plasmid pPCP1 (Pla) plays a minor role in the adhesion and invasion of A549 cells. Because previous studies suggested that the Pla protein acts as an adhesin for extracellular matrix proteins and endothelial cells (33, 34) and an invasin for HeLa cells and endothelial cells (16, 33), we evaluated the potential role of Pla in the interaction of Y. pestis with A549 cells. To determine whether Pla could be responsible for the adhesive and invasive properties of the Y. pestis caf psa double mutant, we compared this strain with the same strain lacking the Pla-containing plasmid pPCP1. Y. pestis caf psa pPCP1– and Y. pestis caf psa pPCP1+ were grown overnight in BHI broth at 37°C and used in adhesion and invasion assays, as described in Materials and Methods. The absence of pPCP1 resulted in only a partial decrease of bacterial adhesion and internalization, with 56% adhesion (±3% [standard deviation of three independent experiments]) and 64% internalization (±12%) relative to the strain carrying pPCP1 (100%). This result hinted at the presence of an additional unknown surface molecule involved in the adhesive and invasive phenotypes of the caf psa double mutant.
DISCUSSION
Currently, it is not clear how inhaled Y. pestis initiates primary pneumonic plague. Although the interaction of Y. pestis with macrophages has been the subject of many studies (11, 22, 31, 41, 42, 52, 53), little is known about the interaction of Y. pestis with major respiratory tract cells such as type I and type II alveolar epithelial cells lining the air-tissue interface. Since the Yersinia type III secretion system is activated by cell contact (43), and the recently deciphered genome of Y. pestis includes several potential adhesins (18, 39, 49), it is likely that one or more adhesins intervene in the first steps of the pulmonary infectious process. Moreover, Cowan et al. (16) previously showed that Y. pestis invades HeLa cells. In this study, we focused on two major surface structures of Y. pestis, namely, the F1 and Psa fimbriae, and analyzed their role in bacterial binding and uptake by respiratory tract epithelial cells. Adhesion assays with recombinant E. coli or Y. pestis, both expressing Psa, and with a Y. pestis psa mutant showed that the Psa fimbriae act as adhesins for both type I and type II alveolar epithelial cells as well as for pharyngeal epithelial cells. Interestingly, the surface antigen F1, whose coexpression was demonstrated by flow cytometry, did not significantly affect Psa-mediated binding. Removal of both surface structures, as shown with the caf psa double mutant, resulted in an adhesive phenotype, indicating the presence of one or more unknown adhesins, which might be inaccessible when the F1 and/or Psa surface structures are expressed.
Both F1 and Psa, together with effector proteins of the type III secretion system of Y. pestis, have been shown to have antiphagocytic properties with murine macrophages (13, 19, 26). Similarly, F1 or Psa was found here to inhibit the uptake of Y. pestis by the three human respiratory tract epithelial cell lines studied. Although F1 inhibits bacterial uptake by inhibiting both bacterial adhesion and internalization, Psa seems to be more efficient at blocking bacterial uptake by binding to a host receptor. The Psa fimbriae are required for full virulence in the mouse intravenous infection model (35) and accelerate death in mice infected intraperitoneally (3), but it is not known whether Psa is expressed after Y. pestis is inhaled into the respiratory tract and whether it plays a role in pneumonic plague. As we have shown, the presence of F1 does not significantly affect the adhesive property of Psa-expressing Y. pestis. Y. pestis-containing aerosols inhaled by humans will bind better to most of the respiratory tract surface if they express Psa, optimizing the needed cell contact to activate the type III secretion system for the intracellular delivery of effector molecules.
Removal of the F1 and Psa surface structures by using a constructed double mutant resulted in a synergistic effect, with increased cellular uptake of a caf psa double mutant being significantly more effective than the uptake of the single mutants. Taken together, our results can be summarized in a model whereby (i) Psa is an adhesin whose phenotype is dominant over other surface molecules and (ii) Psa and F1 are anti-invasive proteins whose phenotypes are dominant over other undefined adhesins and invasins (Fig. 7). Cowan et al. previously observed that the invasion of HeLa cells by Y. pestis was temperature dependent (16). The invasive property was threefold higher when Y. pestis was grown at 26°C instead of 37°C. This difference was even more obvious (20-fold) in the absence of plasmid pCD1 and its Yops regulon. This result can be linked to the fact that Y. pestis cells grown at 26°C express only little or no Psa and F1, mimicking the phenotype of our caf psa double mutant prepared in Y. pestis KIM6 that lacks pCD1. Interestingly, the adhesive and invasive phenotype of this mutant uncovered the existence of additional interactive surface-exposed molecules on Y. pestis. The identity of a potential new invasin involved in the cellular uptake of the Y. pestis double mutant remains to be determined. Because a Y. pestis pPCP1– strain grown at 26°C was previously shown by Cowan et al. to be significantly less invasive with HeLa cells than its parental pPCP1+ strain, those authors suggested that the plasminogen activator protein Pla, a pPCP1-encoded multifunctional outer membrane protease that is expressed at 26°C and 37°C, could be an epithelial cell invasin (16). This was consistent with the later finding that Pla is involved in the invasion of endothelial cells by Y. pestis (33). Although Pla had been described to mediate adhesion to endothelial cells and extracellular matrix proteins (33, 34), Cowan et al. did not find any effect on Y. pestis adhesion to HeLa cells, whether pPCP1 was present or not. Our results indicated that a caf psa pPCP1– strain was still significantly more invasive than the wild-type strain, suggesting that Pla is not an essential surface molecule responsible for the invasive phenotype of the caf psa double mutant. Cowan et al. also found that the pPCP1– strain still invaded host cells (16), albeit at a reduced level. Moreover, strain Pestoides F, which lacks the pla gene, remained virulent when administered with an aerosol (57), although other factors might have replaced the function of Pla. Thus, the observed invasive property of the caf psa double mutant must have been mediated by another surface molecule that is expressed at 37°C. Such a protein could have escaped previous detection with bacteria grown at 26°C, a method commonly used to minimize F1 expression.
It is most likely that several proteins with redundant adhesive or invasive functions appearing specifically under different environmental conditions are involved in the pathogenesis of pneumonic plague. Besides the F1 and Psa surface structures, a panoply of potential fimbrial and nonfimbrial adhesins and invasins has been found by genome sequencing (18, 39, 49). Demonstrating the expression of these proteins and characterizing their role in the pathogenesis of pneumonic plague in the context of other surface proteins will be helpful for the development of effective vaccines against aerosolized Y. pestis.
ACKNOWLEDGMENTS
We thank Jon Goguen, University of Massachusetts Medical School, Worcester, for generously providing Y. pestis strains and for many constructive discussions and Jeff Weiser for the epithelial cells. We are also grateful to Leonard J. Bello for critical reading of the manuscript.
This work was supported by NIH grant 1R21 AI053343-01A1.
FOOTNOTES
Corresponding author. Mailing address: Department of Pathobiology, University of Pennsylvania School of Veterinary Medicine, 3800 Spruce Street, Philadelphia, PA 19104. Phone: (215) 898-1685. Fax: (215) 898-7887. E-mail: dmschiff@vet.upenn.edu.
F.L., H.C., and E.M.G. are co-first authors who contributed equally to this work.
REFERENCES
1. Alexeyev, M. F., I. N. Shokolenko, and T. P. Croughan. 1995. Improved antibiotic-resistance gene cassettes and omega elements for Escherichia coli vector construction and in vitro deletion/insertion mutagenesis. Gene 160:63-67.
2. Anderson, K. L., J. Billington, D. Pettigrew, E. Cota, P. Simpson, P. Roversi, H. A. Chen, P. Urvil, L. du Merle, P. N. Barlow, M. E. Medof, R. A. Smith, B. Nowicki, C. Le Bouguenec, S. M. Lea, and S. Matthews. 2004. An atomic resolution model for assembly, architecture, and function of the Dr adhesins. Mol. Cell 15:647-657.
3. Ben-Efraim, S., M. Aronson, and L. Bichowsky-Slomnichi. 1961. New antigenic component of Pasteurella pestis formed under specified conditions of pH and temperature. J. Bacteriol. 81:704-714.
4. Bichowsky-Slomnicki, L., and S. Bren-Efraim. 1963. Biological activities in extracts of Pasteurella pestis and their relation to the "pH 6 antigen." J. Bacteriol. 86:101-111.
5. Black, D. S., and J. B. Bliska. 2000. The RhoGAP activity of the Yersinia pseudotuberculosis cytotoxin YopE is required for antiphagocytic function and virulence. Mol. Microbiol. 37:515-527.
6. Bliska, J. B. 2000. Yop effectors of Yersinia spp. and actin rearrangements. Trends Microbiol. 8:205-208.
7. Bliska, J. B., and D. S. Black. 1995. Inhibition of the Fc receptor-mediated oxidative burst in macrophages by the Yersinia pseudotuberculosis tyrosine phosphatase. Infect. Immun. 63:681-685.
8. Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L. Heyneker, and H. W. Boyer. 1977. Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2:95-113.
9. Cavanaugh, D. C., and R. Randall. 1959. The role of multiplication of Pasteurella pestis in mononuclear phagocytes in the pathogenesis of flea-borne plague. J. Immunol. 83:348-363.
10. Chang, A. C., and S. N. Cohen. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J. Bacteriol. 134:1141-1156.
11. Charnetzky, W. T., and W. W. Shuford. 1985. Survival and growth of Yersinia pestis within macrophages and an effect of the loss of the 47-megadalton plasmid on growth in macrophages. Infect. Immun. 47:234-241.
12. Chen, T. H., and S. S. Elberg. 1977. Scanning electron microscopic study of virulent Yersinia pestis and Yersinia pseudotuberculosis type 1. Infect. Immun. 15:972-977.
13. Cornelis, G. R. 2000. Molecular and cell biology aspects of plague. Proc. Natl. Acad. Sci. USA 97:8778-8783.
14. Cornelis, G. R. 2002. The Yersinia Ysc-Yop ‘type III’ weaponry. Nat. Rev. Mol. Cell Biol. 3:742-752.
15. Cornelis, G. R., A. Boland, A. P. Boyd, C. Geuijen, M. Iriarte, C. Neyt, M. P. Sory, and I. Stainier. 1998. The virulence plasmid of Yersinia, an antihost genome. Microbiol. Mol. Biol. Rev. 62:1315-1352.
16. Cowan, C., H. A. Jones, Y. H. Kaya, R. D. Perry, and S. C. Straley. 2000. Invasion of epithelial cells by Yersinia pestis: evidence for a Y. pestis-specific invasin. Infect. Immun. 68:4523-4530.
17. Cully, J. F., Jr., A. M. Barnes, T. J. Quan, and G. Maupin. 1997. Dynamics of plague in a Gunnison's prairie dog colony complex from New Mexico. J Wildl. Dis. 33:706-719.
18. Deng, W., V. Burland, G. Plunkett III, A. Boutin, G. F. Mayhew, P. Liss, N. T. Perna, D. J. Rose, B. Mau, S. Zhou, D. C. Schwartz, J. D. Fetherston, L. E. Lindler, R. R. Brubaker, G. V. Plano, S. C. Straley, K. A. McDonough, M. L. Nilles, J. S. Matson, F. R. Blattner, and R. D. Perry. 2002. Genome sequence of Yersinia pestis KIM. J. Bacteriol. 184:4601-4611.
19. Du, Y., R. Rosqvist, and A. Forsberg. 2002. Role of fraction 1 antigen of Yersinia pestis in inhibition of phagocytosis. Infect. Immun. 70:1453-1460.
20. Edwards, R. A., L. H. Keller, and D. M. Schifferli. 1998. Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene 207:149-157.
21. Galyov, E. E., O. Smirnov, A. V. Karlishev, K. I. Volkovoy, A. I. Denesyuk, I. V. Nazimov, K. S. Rubtsov, V. M. Abramov, S. M. Dalvadyanz, and V. P. Zav'yalov. 1990. Nucleotide sequence of the Yersinia pestis gene encoding F1 antigen and the primary structure of the protein. Putative T and B cell epitopes. FEBS Lett. 277:230-232.
22. Goguen, J. D., W. S. Walker, T. P. Hatch, and J. Yother. 1986. Plasmid-determined cytotoxicity in Yersinia pestis and Yersinia pseudotuberculosis. Infect. Immun. 51:788-794.
23. Goguen, J. D., J. Yother, and S. C. Straley. 1984. Genetic analysis of the low calcium response in Yersinia pestis mu d1(Ap lac) insertion mutants. J. Bacteriol. 160:842-848.
24. Horton, R. M., and L. R. Pease. 1991. Recombination and mutagenesis of DNA sequences using PCR, p. 217-247. In M. J. McPherson (ed.), Directed mutagenesis. IRL Press, Oxford, United Kingdom.
25. Hoschvtzky, H., F. Lottspeich, and K. Jann. 1989. Isolation and characterization of the -galactosyl-1,4--galactosyl-specific adhesin (P adhesin) from fimbriated Escherichia coli. Infect. Immun. 57:76-81.
26. Huang, X. Z., and L. E. Lindler. 2004. The pH 6 antigen is an antiphagocytic factor produced by Yersinia pestis independent of Yersinia outer proteins and capsule antigen. Infect. Immun. 72:7212-7219.
27. Inglesby, T. V., R. Grossman, and T. O'Toole. 2001. A plague on your city: observations from TOPOFF. Clin. Infect. Dis. 32:436-445.
28. Iriarte, M., J. C. Vanooteghem, I. Delor, R. Diaz, S. Knutton, and G. R. Cornelis. 1993. The Myf fibrillae of Yersinia enterocolitica. Mol. Microbiol. 9:507-520.
29. Jacob-Dubuisson, F., J. Heuser, K. Dodson, S. Normark, and S. Hultgren. 1993. Initiation of assembly and association of the structural elements of a bacterial pilus depend on two specialized tip proteins. EMBO J. 12:837-847.
30. Jacobs, A. A. C., and F. K. de Graaf. 1985. Production of K88, K99 and F41 fibrillae in relation to growth phase, and a rapid procedure for adhesin purification. FEMS Microbiol. Lett. 26:15-19.
31. Janssen, W. A., and M. J. Surgalla. 1969. Plague bacillus: survival within host phagocytes. Science 163:950-952.
32. Kool, J. L. 2005. Risk of person-to-person transmission of pneumonic plague. Clin. Infect. Dis. 40:1166-1172.
33. Lahteenmaki, K., M. Kukkonen, and T. K. Korhonen. 2001. The Pla surface protease/adhesin of Yersinia pestis mediates bacterial invasion into human endothelial cells. FEBS Lett. 504:69-72.
34. Lhteenmki, K., R. Virkola, A. Saren, L. Emdy, and T. K. Korhonen. 1998. Expression of plasminogen activator Pla of Yersinia pestis enhances bacterial attachment to the mammalian extracellular matrix. Infect. Immun. 66:5755-5762.
35. Lindler, L. E., M. S. Klempner, and S. C. Straley. 1990. Yersinia pestis pH 6 antigen: genetic, biochemical, and virulence characterization of a protein involved in the pathogenesis of bubonic plague. Infect. Immun. 58:2569-2577.
36. Lindler, L. E., and B. D. Tall. 1993. Yersinia pestis pH6 antigen forms fimbriae and is induced by intracellular association with macrophages. Mol. Microbiol. 8:311-324.
37. Lukaszewski, R. A., D. J. Kenny, R. Taylor, D. G. Rees, M. G. Hartley, and P. C. Oyston. 2005. Pathogenesis of Yersinia pestis infection in BALB/c mice: effects on host macrophages and neutrophils. Infect. Immun. 73:7142-7150.
38. Manoil, C., and J. Beckwith. 1985. TnphoA: a transposon probe for protein export signals. Proc. Natl. Acad. Sci. USA 82:8129-8133.
39. Parkhill, J., B. W. Wren, N. R. Thomson, R. W. Titball, M. T. Holden, M. B. Prentice, M. Sebaihia, K. D. James, C. Churcher, K. L. Mungall, S. Baker, D. Basham, S. D. Bentley, K. Brooks, A. M. Cerdeno-Tarraga, T. Chillingworth, A. Cronin, R. M. Davies, P. Davis, G. Dougan, T. Feltwell, N. Hamlin, S. Holroyd, K. Jagels, A. V. Karlyshev, S. Leather, S. Moule, P. C. Oyston, M. Quail, K. Rutherford, M. Simmonds, J. Skelton, K. Stevens, S. Whitehead, and B. G. Barrell. 2001. Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413:523-527.
40. Perry, R. D., and J. D. Fetherston. 1997. Yersinia pestis—etiologic agent of plague. Clin. Microbiol. Rev. 10:35-66.
41. Pujol, C., and J. B. Bliska. 2003. The ability to replicate in macrophages is conserved between Yersinia pestis and Yersinia pseudotuberculosis. Infect. Immun. 71:5892-5899.
42. Pujol, C., and J. B. Bliska. 2005. Turning Yersinia pathogenesis outside in: subversion of macrophage function by intracellular yersiniae. Clin. Immunol. 114:216-226.
43. Rosqvist, R., K. E. Magnusson, and H. Wolf-Watz. 1994. Target cell contact triggers expression and polarized transfer of Yersinia YopE cytotoxin into mammalian cells. EMBO J. 13:964-972.
44. Silhavy, T. J., M. L. Berman, and L. W. Enquist. 1984. Experiments with gene fusion. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
45. Simonet, M., B. Riot, N. Fortineau, and P. Berche. 1996. Invasin production by Yersinia pestis is abolished by insertion of an IS200-like element within the inv gene. Infect. Immun. 64:375-379.
46. Smith, R. M., and L. Jarett. 1993. Electron microscopic immunocytochemical approaches to the localization of ligands, receptors, tranducers, and transporters, p. 227-264. In F. De Pablo, C. G. Scanes, and B. D. Weintraub (ed.), Handbook of endocrine research techniques. Academic Press, San Diego, Calif.
47. Sodeinde, O. A., A. K. Sample, R. R. Brubaker, and J. D. Goguen. 1988. Plasminogen activator/coagulase gene of Yersinia pestis is responsible for degradation of plasmid-encoded outer membrane proteins. Infect. Immun. 56:2749-2752.
48. Sodeinde, O. A., Y. V. Subrahmanyam, K. Stark, T. Quan, Y. Bao, and J. D. Goguen. 1992. A surface protease and the invasive character of plague. Science 258:1004-1007.
49. Song, Y., Z. Tong, J. Wang, L. Wang, Z. Guo, Y. Han, J. Zhang, D. Pei, D. Zhou, H. Qin, X. Pang, J. Zhai, M. Li, B. Cui, Z. Qi, L. Jin, R. Dai, F. Chen, S. Li, C. Ye, Z. Du, W. Lin, J. Yu, H. Yang, P. Huang, and R. Yang. 2004. Complete genome sequence of Yersinia pestis strain 91001, an isolate avirulent to humans. DNA Res. 11:179-197.
50. Stoker, N. G., N. F. Fairweather, and B. G. Spratt. 1982. Versatile low-copy-number plasmid vectors for cloning in Escherichia coli. Gene 18:335-341.
51. Straley, S. C. 1993. Adhesins in Yersinia pestis. Trends Microbiol. 1:285-286.
52. Straley, S. C., and P. A. Harmon. 1984. Growth in mouse peritoneal macrophages of Yersinia pestis lacking established virulence determinants. Infect. Immun. 45:649-654.
53. Straley, S. C., and P. A. Harmon. 1984. Yersinia pestis grows within phagolysosomes in mouse peritoneal macrophages. Infect. Immun. 45:655-659.
54. Straley, S. C., and R. D. Perry. 1995. Environmental modulation of gene expression and pathogenesis in Yersinia. Trends Microbiol. 3:310-317.
55. Titball, R. W., A. M. Howells, P. C. Oyston, and E. D. Williamson. 1997. Expression of the Yersinia pestis capsular antigen (F1 antigen) on the surface of an aroA mutant of Salmonella typhimurium induces high levels of protection against plague. Infect. Immun. 65:1926-1930.
56. Viboud, G. I., and J. B. Bliska. 2005. Yersinia outer proteins: role in modulation of host cell signalling responses and pathogenesis. Annu. Rev. Microbiol. 59:69-89.
57. Worsham, P. L., and C. Roy. 2003. Pestoides F, a Yersinia pestis strain lacking plasminogen activator, is virulent by the aerosol route. Adv. Exp. Med. Biol. 529:129-131.
58. Yang, Y., J. J. Merriam, J. P. Mueller, and R. R. Isberg. 1996. The psa locus is responsible for thermoinducible binding of Yersinia pseudotuberculosis to cultured cells. Infect. Immun. 64:2483-2489.
59. Zavialov, A. V., J. Berglund, A. F. Pudney, L. J. Fooks, T. M. Ibrahim, S. MacIntyre, and S. D. Knight. 2003. Structure and biogenesis of the capsular F1 antigen from Yersinia pestis: preserved folding energy drives fiber formation. Cell 113:587-596.
60. Zhang, Y., and J. B. Bliska. 2005. Role of macrophage apoptosis in the pathogenesis of Yersinia. Curr. Top. Microbiol. Immunol. 289:151-173.(Fengzhi Liu, Huaiqing Chen, Estela M. Ga)
ABSTRACT
Yersinia pestis, the causative agent of plague, expresses the Psa fimbriae (pH 6 antigen) in vitro and in vivo. To evaluate the potential virulence properties of Psa for pneumonic plague, an Escherichia coli strain expressing Psa was engineered and shown to adhere to three types of human respiratory tract epithelial cells. Psa binding specificity was confirmed with Psa-coated polystyrene beads and by inhibition assays. Individual Y. pestis cells were found to be able to express the capsular antigen fraction 1 (F1) concomitantly with Psa on their surface when analyzed by flow cytometry. To better evaluate the separate effects of F1 and Psa on the adhesive and invasive properties of Y. pestis, isogenic caf (F1 genes), psa, and caf psa mutants were constructed and studied with the three respiratory tract epithelial cells. The psa mutant bound significantly less to all three epithelial cells compared to the parental wild-type strain and the caf and caf psa mutants, indicating that Psa acts as an adhesin for respiratory tract epithelial cells. An antiadhesive effect of F1 was clearly detectable only in the absence of Psa, underlining the dominance of the Psa+ phenotype. Both F1 and Psa inhibited the intracellular uptake of Y. pestis. Thus, F1 inhibits bacterial uptake by inhibiting bacterial adhesion to epithelial cells, whereas Psa seems to block bacterial uptake by interacting with a host receptor that doesn't direct internalization. The caf psa double mutant bound and invaded all three epithelial cell types well, revealing the presence of an undefined adhesin(s) and invasin(s).
INTRODUCTION
Since the last plague pandemic at the end of the 19th century, its bacterial agent, Yersinia pestis, has been maintained in rodents in several Asian, African, and American countries, including the United States (17, 40). Bubonic plague results from the transmission of Y. pestis by flea bites. In contrast, primary pneumonic plague is acquired when a mammalian host inhales particles or aerosols carrying Y. pestis. Although plague is currently not a major public health problem in developed countries and has been suggested to be less contagious than commonly believed (32), the spread of Y. pestis by aerosols could cause a cluster of human cases of primary pneumonic plague with potential amplification of the outbreak (27).
The major adhesins and invasins of enteropathogenic Yersinia pseudotuberculosis and Yersinia enterocolitica (YadA, Ail, and Inv) are not expressed by Y. pestis strains (15, 45, 51). Thus, how Y. pestis attaches to and translocates through the epithelial layer of the respiratory tract to reach deeper tissues and the bloodstream following airborne transmission remains unknown. Interestingly, Y. pestis exhibits an extensive extracellular lifestyle due to the intracellular delivery of several antiphagocytic effector proteins by its type III secretion system (T3SS-1 or Yops regulon) (5-7, 14, 54, 56). Moreover, two antigenic surface structures exported by usher-chaperone proteins characteristic of fimbrial biogenesis systems have been shown to inhibit Y. pestis uptake by macrophages (19, 26). Each of these structures, designated capsular antigen fraction 1 (F1) and pH 6 antigen (Psa), consists of the homopolymeric association of a single-subunit protein (Caf1 and PsaA, respectively) on the bacterial surface (36, 59). PsaA can assemble into distinct thin individual fimbrial strands, bundles of fimbriae associated along their lengths, or structureless aggregates, as shown by electron microscopy (12, 36). The F1 capsular antigen is made of linear fibers of Caf1 subunits that cannot be detected as fimbrial organelles (59), not unlike the Dr adhesins that were designated afimbrial adhesins (2).
Although the expression of redundant antiphagocytic virulence factors suggests that Y. pestis remains and replicates mainly extracellularly, Y. pestis has been reported to invade human epithelial HeLa cells (16) and to be phagocytosed by macrophages, where it can replicate (9, 11, 31, 36, 37, 41, 52). It has been suggested that intracellular survival and replication occur mainly during the early stages of host colonization and invasion (42). Phagocytosed Y. pestis isolates have also been found to have cytotoxic and apoptotic effects on macrophages (22, 60).
Cultured epithelial cell lines have been used previously to investigate the epithelioadhesive property of Y. pestis and the involvement of Psa (16, 58). Here, we determined whether Y. pestis adheres to human respiratory tract epithelial cells and possibly penetrates these cells. Whether and how the F1 and Psa surface structures modulate such interactions were the focus of this study.
MATERIALS AND METHODS
Bacterial strains and constructed plasmids. Escherichia coli strains, Y. pestis strains, and plasmids used in this study are listed in Table 1. E. coli strains were grown overnight at 37°C in LB broth (Lennox) with appropriate antibiotics, when required, using a microbial culture roller drum. Y. pestis strains were grown overnight at 26°C in brain heart infusion (BHI) broth, diluted 1:20 in fresh BHI broth, and cultured overnight at 37°C using the roller drum. Under these conditions, the pH of the spent broth dropped to 6.2 to 6.4, resulting in the expression of both the F1 and Psa structures on the bacterial surface. Some cultures of Y. pestis were prepared in BHI broth buffered with 0.1 M MES (morpholineethanesulfonic acid) (pH 6) and grown overnight at 37°C to ensure optimal pH 6 antigen expression. F1 was also expressed under these growth conditions.
Cloning and expression of Psa and F1. The gene cluster encoding the structural genes of Psa (psaABC) was cloned into pBR322. For this, a DNA fragment starting from the ribosomal binding site upstream of the psaA ATG start codon and ending at the stop codon of psaC was amplified by PCR using genomic DNA from strain KIM6 as template, primers 5'-CTAGCTAGCATTTAATTGCCCAATATCACC-3' and 5'-ACGCGTCGACAAACCTGGCGGGAATTGAACC-3' (restriction sites used are underlined), and the thermostable DNA polymerase blend of the Expand Long Template PCR system (Roche Applied Sciences, Indianapolis, IN). The NheI- and SalI-restricted amplicon was inserted into the corresponding sites of pBR322 to create pCS267, taking advantage of the constitutive tetracycline promoter to drive the expression of Psa. The chloramphenicol resistance gene of pACYC184 was excised as an AhdI-XmnI fragment that was blunted and inserted into the blunted AhdI site of the ampicillin resistance gene of pCS267, resulting in plasmid pCS334. The gene cluster encoding the structural genes for the F1 capsular antigen or fimbriae (caf1M-caf1A-caf1) was cloned into pET16b. For this, a DNA fragment starting from the caf1M ATG start codon and ending at 17 nucleotides downstream of the stop codon of caf1 was amplified by PCR using primers 5'-GCTATTCCGTCTCGCATGATTTTAAATAGATTAAGTACGTT and 5'-ACGCGTCGACTTATCTATATGGATTATTGGTTAGA. The BsmBI- and SalI-restricted amplicon was inserted into the restricted NcoI and XhoI sites of pET16b to create pCS274. The XbaI and BamHI fragment of pCS274 that includes the ribosomal binding site and the F1 genes was subcloned into NheI- and BamHI-restricted pLG339 to create plasmid pCS331. The chloramphenicol resistance gene described above was inserted into the SmaI site of the kanamycin resistance gene of pCS331, resulting in plasmid pCS332. The PCR-cloned DNAs corresponding to the surface-expressed fimbrial proteins Caf1 and PsaA were sequenced and found to correspond to the published sequences for strain KIM, indicating the absence of PCR-mediated mutations. Expression of the Psa and F1 proteins on the surface of E. coli was confirmed by negative staining for transmission electron microscopy (see below) and sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE). The presence of Psa was further demonstrated by hemagglutination with sheep red blood cells.
Construction of Y. pestis mutants. The caf, psa, and caf psa mutants were constructed by allelic exchange, as described previously (20). To construct a Y. pestis psa deletion mutant, two DNA fragments of 1 to 1.3 kb flanking the psaABC locus were amplified by PCR with primer pairs 5'-CTAGTCTAGACAAACAGCAATGAAAGCAAAATCACT-3' and 5'-GGCCGAGCTCCGAGAACAGTCTCCA-3' and 5'-TTCTCGGAGCTCGGCCAGGCATCCGAGATGAC-3' and 5'-CGGGGGTACCTCTGGGTTAATCATCTGATTAC-3', respectively. These two PCR fragments were joined by splicing by overlap extension (SOEing) (24) to produce a 2.3-kb amplicon with a central unique SacI restriction site. The designed flanking restriction sites XbaI and KpnI were used to clone the amplicon in the corresponding sites of pDMS197, generating pLS013. The SacI-restricted kanamycin resistance cassette of pBSL128 (1) was then inserted into the corresponding site in pLS013, generating suicide plasmid pLS014. The same approach was used to engineer a caf1M-caf1A-caf1 deletion mutant. For this, a 2.05-kb PCR product was constructed by SOEing with primers 5'-CTAGGAGCTCCAGGCAAAGAGTTATAATATAT-3', 5'-CGCGGGGCCCGAGCTTAACCTCCT-3', 5'-AAGCTCGGGCCCCGCGTCCATATAGATAATAG-3', and 5'-CGGGGGTACCTTACCTTTACTTCAGAGATATG-3', creating a central unique ApaI restriction site. The 2.05-kb amplicon was inserted into pDMS197, generating plasmid pLS007. An ApaI-digested ampicillin resistance cassette of pBSL159 (1) was inserted into pLS007, generating suicide plasmid pLS009. To delete caf or/and psa genes from Y. pestis, 10 μg of plasmid DNA was electroporated into Y. pestis cells (100 μl; 1011 cells/ml). Transformed cells were allowed to recover for 2 to 4 h in BHI broth at 26°C and then plated onto BHI agar plate with tetracycline (10 μg/ml) to select for clones with integrated plasmids. Clones were subcultured at 26°C for 18 h in NaCl-free LB broth with ampicillin (200 μg/ml) or kanamycin (45 μg/ml) to allow for spontaneous plasmid loss and plated onto NaCl-free LB agar with and without 5% sucrose to select for tetracycline-sensitive and sucrose-resistant clones. For strain DSY50, the resistance gene markers were removed in two steps, using pLS007 and pLS013 sequentially as described above. Isolates were selected for sucrose resistance and screened for susceptibility to the respective antibiotic. All the gene deletions were confirmed by PCR using appropriate primers.
Isolation of F1 and Psa proteins and antibody production. F1 or Psa fimbriae expressed on the bacterial surface were prepared by heat extraction of strain KIM6 or recombinant Psa- or F1-expressing E. coli SE5000, as described previously (25, 29, 30). Briefly, bacteria were pelleted by centrifugation, resuspended in 0.5 mM Tris-HCl (pH 7.4)-75 mM NaCl, and treated at 60°C for 30 min. After centrifugation, ammonium sulfate was added to the supernatant to a 30% final concentration to precipitate the extracted proteins overnight on ice. After centrifugation (12,000 x g, 30 min), the pellets were resuspended in phosphate-buffered saline (PBS), and excess ammonium sulfate was removed by dialysis. The purity of the proteins was confirmed by SDS-PAGE and Coomassie blue staining (>90% pure). An anti-Psa antibody was prepared in rabbits with isolated recombinant Psa protein by using a conventional immunization protocol (Cocalico Biologicals Inc., Reamstown, PA). The serum was adsorbed with E. coli SE5000. The antibody had a specific enzyme-linked immunosorbent assay titer of 105.
Binding and internalization assays. Human cell lines used in this study include type II alveolar epithelial cell line A549 (ATCC CCL185), pharynx pleural effusion carcinoma cell line Detroit 562 (ATCC CCL138), and type I alveolar epithelial cell line WI26 VA4 (ATCC CCL95.1). The cells were routinely maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37°C in a humidified atmosphere of 5% CO2 in air. The cells were grown in 6- or 24-well plates to confluent monolayers and washed three times with PBS, and DMEM without FBS was added. Bacteria grown as described above were washed with PBS, retaken in DMEM without FBS, and added to the cells at a multiplicity of infection (MOI) of 10:1 to 50:1. Preliminary studies showed that the bacteria bound significantly better when bacterial binding was studied in DMEM without FBS (as described above) instead of PBS. The cell monolayers were incubated at 37°C for 1 h before being washed three times with PBS. For the adhesion assays, the epithelial cells were lysed with cold 0.1% Triton X-100 in deionized H2O, and 10-fold-serially-diluted lysates were spread onto agar plates to determine CFU counts. For the internalization assays, the cell monolayers were incubated for an additional 2 h in 1 ml (24-well plate) or 2 ml (six-well plate) medium per well, with 50 μg/ml gentamicin. After three steps of washing with PBS, the cell monolayers were lysed for determining CFU counts as described above. Adherence percentages were calculated as the numbers of cell-associated bacteria divided by the total numbers of inoculated bacteria x 100. Internalization percentages were calculated as the numbers of internal bacteria divided by the numbers of inoculated bacteria x 100. Invasion indices were calculated as the numbers of internal bacteria divided by the numbers of cell-associated bacteria x 100. Student's t test was used to calculate statistical significance.
Coating and binding of beads. Polystyrene beads (1-μm diameter) (Polybead; Polysciences, Inc., Warrington, PA) were washed three times in Tris-HCl buffer (5 mM, pH 7.0) and resuspended in PBS, and 4.7 x 109 beads were mixed with isolated Psa or F1 fimbriae or with BSA (60 μg/ml) in a total volume of 0.5 ml on a rotating wheel at room temperature overnight. The beads were centrifuged (14,000 rpm, 5 min), washed three times in PBS, resuspended in PBS containing 0.1% BSA, and briefly sonicated to disperse bead aggregates. Successful Psa or F1 coating was confirmed by agglutination with the corresponding antibodies. The bead-binding assay (five beads per cell) was undertaken with the A549 cells as described above for the bacterial adhesion assay. Student's t test was used to calculate statistical significance.
Microscopy. Bacterial binding to cells was examined after Giemsa staining by using bright-field microscopy. The binding of polystyrene beads to A549 cells was visualized by phase-contrast microscopy. All images were captured with a Coolsnap digital camera (Photometrics, Tuscon, AZ) mounted onto a Nikon Eclipse E600 microscope with Coolsnap version 1.2.0 software. The number of cell-associated beads was counted for at least 100 cells to calculate means (μ) and standard errors. Negatively stained Psa- and F1-expressing bacteria were examined by electron microscopy (Joel 1010 transmission electron microscope). Images obtained with the Jeol JEM 1010 transmission electron microscope were captured by using an AMT 12-HR-aided Hamamatsu charge-coupled-devise camera (46).
Flow cytometry. Bacteria were grown overnight in BHI broth at 37°C. The bacterial cells were washed once with PBS and labeled with rabbit anti-Psa antibody and monoclonal mouse anti-F1 antibody (Advanced ImmunoChemicals Inc., Long Beach, CA), followed by fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulin G (ICN Biomedicals, Inc., Aurora, Ohio) and phycoerythrin-conjugated F(ab')2 goat anti-mouse immunoglobulin G (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). The labeled bacteria were fixed overnight at 4°C in 10% (vol/vol) neutral buffered formalin in PBS, pH 7.4. The analysis by flow cytometry was undertaken with FACScan and CellQuest Pro software (BD Biosciences, San Jose, CA).
RESULTS
Recombinant Psa and F1 capsular antigen expression. E. coli SE5000 transformed with plasmid pCS267, which carries the psaABC genes, formed negatively stained fimbria-like structures as well as aggregates on the bacterial surface when observed by electron microscopy (data not shown). These structures and the additional aggregates surrounding the bacteria were not seen on strain SE5000 lacking pCS267. Heat extraction at 60°C, conventionally used to release fimbrial proteins from bacteria, identified a major protein of 15 kDa by SDS-PAGE (data not shown), corresponding to the known molecular mass of PsaA, the Psa fimbrial subunit (35). This protein demonstrated the typical properties of fimbrial subunits by remaining associated as a large polymer in 1% SDS when not treated at 100°C. Psa-fimbriated bacteria are known to have hemagglutinating properties (4). Consistent with this, only the Psa-fimbriated E. coli SE5000(pCS267), and not E. coli SE5000, agglutinated sheep erythrocytes (data not shown), confirming the successful cloning and expression of recombinant Psa fimbriae in E. coli. Heat extraction of E. coli SE5000 transformed with plasmid pCS274 or pCS331, which carry the caf1M-caf1A-caf1 genes, identified a major protein of 16 kDa by SDS-PAGE (data not shown), corresponding to the known molecular mass of the F1 protein subunit Caf1 (21, 55). As found with PsaA, Caf1 remained associated as a large polymer in 1% SDS when not treated at 100°C.
Psa-fimbriated E. coli cells bind to human respiratory tract epithelial cells. In order to investigate whether Psa could mediate bacterial adhesion to respiratory tract epithelial cells, the binding of E. coli SE5000(pCS267) (Psa+) was studied with the human type II alveolar epithelial cell line A549. Microscopy showed that the Psa-fimbriated E. coli cells associated in greater numbers with the A549 cells than the nonfimbriated E. coli (Fig. 1A and B). Colony counts determined that the Psa-fimbriated E. coli cells adhered 13 times better to A549 cells than E. coli SE5000(pBR322) (Fig. 1C). Similar, albeit less striking, results were observed with the human pharyngeal epithelial cell line Detroit 562 and type I alveolar epithelial cell line WI26 VA4 (data not shown). Binding specificity was confirmed by inhibiting the adhesive interaction in a dose-dependent manner with isolated Psa fimbriae (50% inhibitory concentration of 8 μg Psa for a bacterial inoculum of 2 x 107 CFU/well of a six-well plate) or a Psa-specific polyclonal antibody (Fig. 1D).
Psa-coated beads bind to human respiratory tract epithelial cells. To further confirm that the Psa fimbriae are adhesins, Psa-coated polystyrene beads were prepared and used to determine their binding properties with the type II alveolar epithelial cell line A549. Other sets of beads were coated with isolated F1 protein or BSA and studied in parallel for comparative purposes. Microscopic examination showed that the Psa-coated beads bound significantly better to A549 cells (mean ± standard error, 2.50 ± 0.13 beads/cell) than the BSA-coated beads (0.80 ± 0.19 beads/cell) or F1-coated microspheres (0.34 ± 0.5 beads/cell) (Fig. 2). Taken together, the data indicated that Psa acts as a Y. pestis adhesin for human type II alveolar epithelial cells.
Psa is not an invasin. Since the Psa fimbriae were shown to increase the association of recombinant E. coli with A549 human type II alveolar epithelial cells, and since Y. pestis has previously been shown to invade epithelial cells (16), the possibility that the Psa fimbriae modulate the uptake of E. coli by A549 cells was investigated. Internalized bacteria were identified by using a gentamicin protection assay. Although the expression of Psa rendered E. coli significantly more adhesive (Fig. 1C), Psa-fimbriated and nonfimbriated E. coli cells were similarly poorly internalized (Fig. 3A), suggesting that Psa is not an invasin. However, significantly more cell-associated nonfimbriated E. coli cells were internalized than cell-associated Psa-fimbriated E. coli cells (Fig. 3B), suggesting that Psa-mediated E. coli binding had an inhibitory effect on the observed low level of E. coli uptake.
Coexpression of F1 and Psa on individual Y. pestis cells. In order to evaluate the role of the Psa fimbriae in the adhesive and invasive properties of Y. pestis, a psa mutant was constructed. Moreover, because the F1 surface structure of Y. pestis might affect the accessibility of Psa to a host cell receptor, both an isogenic caf mutant and an isogenic psa caf double mutant were constructed and studied in parallel. The phenotypes of the wild-type and mutant strains were examined by electron microscopy and SDS-PAGE (Fig. 4). The wild-type strain and two single mutants were typically surrounded by a thick layer of negatively stained material (Fig. 4A to C). In contrast, most double-mutant cells did not have this negatively stained halo (Fig. 4D), suggesting that it represented the staining of PsaA and/or F1 subunit aggregates, as described previously by others (12, 28, 36). This interpretation was consistent with the protein profile of heat-extracted surface proteins from the four strains, as shown by SDS-PAGE (Fig. 4E). Moreover, double-mutant cells with one or a few fimbria-like structures were observed (Fig. 4D, arrow), suggesting the expression of additional surface structures on Y. pestis. The finding that both Caf1 and PsaA were produced by the wild-type strain in the same culture was in agreement with recently published findings (26) indicating that Y. pestis can express both F1 and Psa surface structures together. To ensure that such cultures did not consist of mixtures of cells that exclusively produce only one of the two surface structures, the expression of F1 and Psa on individual Y. pestis cells was investigated by flow cytometry using antigen-specific antibodies for double labeling. As shown with the mutants that served as controls, both antibodies labeled only their respective antigens (Fig. 5). Most importantly, 80% or more of the wild-type cells expressed both F1 and Psa simultaneously, demonstrating that F1 or Psa expression does not exclude the expression of the other antigen on the bacterial surface.
Reduced bacterial adhesion with a Y. pestis psa mutant. The binding properties of the psa or/and caf mutants were investigated with three types of human respiratory tract epithelial cells, namely, pharyngeal (Detroit 562) and alveolar type I (WI26 VA4) and type II (A549) epithelial cells. In general, for all three cell lines, the psa single mutant bound the least (Fig. 6). When complemented in trans with the psa-containing plasmid pCS334, but not with pBR322, the psa mutant regained 82% of the adhesive properties of the parental KIM6 (psa+) strain towards the WI26 VA4 cells, confirming that Psa is an adhesin for respiratory tract epithelial cells. Interestingly, in contrast to the psa mutant but not unlike the caf mutant, the psa caf double mutant bound well to all three cell lines (Fig. 6). Complementing the caf mutant with the caf-containing plasmid pCS332 did not interfere significantly with the strains' adhesive properties, with binding by the parental strain, the mutant strain, and the complemented strain being essentially the same (<15% differences). In contrast, when the double mutant was complemented with the caf-containing plasmid pCS332, the strains' binding properties towards the WI26 VA4 cells were decreased by 72%, mimicking the reduced adhesiveness of the single psa mutant. These results suggested that in addition to Psa, Y. pestis expresses at least one other adhesin whose adhesive property is detectable only in the absence of F1. Binding to WI26 VA4 cells was not changed significantly after complementing the double mutant with the psa-containing plasmid pCS334 (<19% difference). Taken together, the results of the binding assays with recombinant E. coli and the Y. pestis mutants indicated that Psa+ is a dominant adhesive phenotype and that the binding mediated by a proposed new Y. pestis adhesin(s) is a recessive phenotype that becomes apparent in the absence of Psa and F1.
Improved bacterial internalization with a Y. pestis caf psa mutant. To evaluate the role of the Psa adhesin in the cellular uptake of Y. pestis by respiratory tract epithelial cells, the wild-type and psa and caf mutant strains were studied by a standard gentamicin protection assay. The pCD1-encoded proteins, including the antiphagocytic type III effector proteins, were previously shown to have no significant role in the uptake of Y. pestis by HeLa and WI26 VA4 epithelial cells (16). Nevertheless, to avoid any potential confounding effects by the type III secretion system, all the studied strains were derivatives from the pCD1– strain KIM6. In comparison to the wild-type strain, the psa mutant was internalized 89 times better by A549 cells (Fig. 6). Better bacterial uptake was also detected with the other epithelial cells, as shown in Fig. 6. Complementation of the psa mutant with the psa-containing plasmid pCS334 reduced bacterial internalization by the WI26 VA4 cells to nearly the same level (1.4 times) as that observed for KIM6 internalization, confirming that Psa inhibits Y. pestis internalization by respiratory tract epithelial cells. In general, the caf mutant was also better internalized than the parental strain KIM6 (Fig. 6). However, in contrast to the psa mutant and because of the Psa-mediated binding of the caf mutant, the invasion index of the latter strain showed lower values, since this index relates to the intracellular uptake of bound bacteria (Fig. 6). As observed for the pCS334-complemented psa mutant, complementing the caf mutant with the caf-containing plasmid pCS332 significantly reduced bacterial internalization by the WI26 VA4 cells (from 121 to 3.3 times the value observed for KIM6 internalization), confirming that F1 also inhibits Y. pestis internalization by respiratory tract epithelial cells. Interestingly, the strain that was internalized the most efficiently by all three epithelial cells was the caf psa double mutant (Fig. 6). Complementing the double mutant with the psa- or the caf-containing plasmid reduced Y. pestis internalization approximately four to five times compared to the noncomplemented double mutant. Internalization of these complemented bacteria was comparable to the internalization of the respective single mutants. Taken together, these results indicated that both Psa and F1 inhibit the uptake of Y. pestis by respiratory tract epithelial cells. Moreover, the results also suggested that at least one new surface-exposed invasin was uncovered by the removal of F1 and Psa.
Plasmid pPCP1 (Pla) plays a minor role in the adhesion and invasion of A549 cells. Because previous studies suggested that the Pla protein acts as an adhesin for extracellular matrix proteins and endothelial cells (33, 34) and an invasin for HeLa cells and endothelial cells (16, 33), we evaluated the potential role of Pla in the interaction of Y. pestis with A549 cells. To determine whether Pla could be responsible for the adhesive and invasive properties of the Y. pestis caf psa double mutant, we compared this strain with the same strain lacking the Pla-containing plasmid pPCP1. Y. pestis caf psa pPCP1– and Y. pestis caf psa pPCP1+ were grown overnight in BHI broth at 37°C and used in adhesion and invasion assays, as described in Materials and Methods. The absence of pPCP1 resulted in only a partial decrease of bacterial adhesion and internalization, with 56% adhesion (±3% [standard deviation of three independent experiments]) and 64% internalization (±12%) relative to the strain carrying pPCP1 (100%). This result hinted at the presence of an additional unknown surface molecule involved in the adhesive and invasive phenotypes of the caf psa double mutant.
DISCUSSION
Currently, it is not clear how inhaled Y. pestis initiates primary pneumonic plague. Although the interaction of Y. pestis with macrophages has been the subject of many studies (11, 22, 31, 41, 42, 52, 53), little is known about the interaction of Y. pestis with major respiratory tract cells such as type I and type II alveolar epithelial cells lining the air-tissue interface. Since the Yersinia type III secretion system is activated by cell contact (43), and the recently deciphered genome of Y. pestis includes several potential adhesins (18, 39, 49), it is likely that one or more adhesins intervene in the first steps of the pulmonary infectious process. Moreover, Cowan et al. (16) previously showed that Y. pestis invades HeLa cells. In this study, we focused on two major surface structures of Y. pestis, namely, the F1 and Psa fimbriae, and analyzed their role in bacterial binding and uptake by respiratory tract epithelial cells. Adhesion assays with recombinant E. coli or Y. pestis, both expressing Psa, and with a Y. pestis psa mutant showed that the Psa fimbriae act as adhesins for both type I and type II alveolar epithelial cells as well as for pharyngeal epithelial cells. Interestingly, the surface antigen F1, whose coexpression was demonstrated by flow cytometry, did not significantly affect Psa-mediated binding. Removal of both surface structures, as shown with the caf psa double mutant, resulted in an adhesive phenotype, indicating the presence of one or more unknown adhesins, which might be inaccessible when the F1 and/or Psa surface structures are expressed.
Both F1 and Psa, together with effector proteins of the type III secretion system of Y. pestis, have been shown to have antiphagocytic properties with murine macrophages (13, 19, 26). Similarly, F1 or Psa was found here to inhibit the uptake of Y. pestis by the three human respiratory tract epithelial cell lines studied. Although F1 inhibits bacterial uptake by inhibiting both bacterial adhesion and internalization, Psa seems to be more efficient at blocking bacterial uptake by binding to a host receptor. The Psa fimbriae are required for full virulence in the mouse intravenous infection model (35) and accelerate death in mice infected intraperitoneally (3), but it is not known whether Psa is expressed after Y. pestis is inhaled into the respiratory tract and whether it plays a role in pneumonic plague. As we have shown, the presence of F1 does not significantly affect the adhesive property of Psa-expressing Y. pestis. Y. pestis-containing aerosols inhaled by humans will bind better to most of the respiratory tract surface if they express Psa, optimizing the needed cell contact to activate the type III secretion system for the intracellular delivery of effector molecules.
Removal of the F1 and Psa surface structures by using a constructed double mutant resulted in a synergistic effect, with increased cellular uptake of a caf psa double mutant being significantly more effective than the uptake of the single mutants. Taken together, our results can be summarized in a model whereby (i) Psa is an adhesin whose phenotype is dominant over other surface molecules and (ii) Psa and F1 are anti-invasive proteins whose phenotypes are dominant over other undefined adhesins and invasins (Fig. 7). Cowan et al. previously observed that the invasion of HeLa cells by Y. pestis was temperature dependent (16). The invasive property was threefold higher when Y. pestis was grown at 26°C instead of 37°C. This difference was even more obvious (20-fold) in the absence of plasmid pCD1 and its Yops regulon. This result can be linked to the fact that Y. pestis cells grown at 26°C express only little or no Psa and F1, mimicking the phenotype of our caf psa double mutant prepared in Y. pestis KIM6 that lacks pCD1. Interestingly, the adhesive and invasive phenotype of this mutant uncovered the existence of additional interactive surface-exposed molecules on Y. pestis. The identity of a potential new invasin involved in the cellular uptake of the Y. pestis double mutant remains to be determined. Because a Y. pestis pPCP1– strain grown at 26°C was previously shown by Cowan et al. to be significantly less invasive with HeLa cells than its parental pPCP1+ strain, those authors suggested that the plasminogen activator protein Pla, a pPCP1-encoded multifunctional outer membrane protease that is expressed at 26°C and 37°C, could be an epithelial cell invasin (16). This was consistent with the later finding that Pla is involved in the invasion of endothelial cells by Y. pestis (33). Although Pla had been described to mediate adhesion to endothelial cells and extracellular matrix proteins (33, 34), Cowan et al. did not find any effect on Y. pestis adhesion to HeLa cells, whether pPCP1 was present or not. Our results indicated that a caf psa pPCP1– strain was still significantly more invasive than the wild-type strain, suggesting that Pla is not an essential surface molecule responsible for the invasive phenotype of the caf psa double mutant. Cowan et al. also found that the pPCP1– strain still invaded host cells (16), albeit at a reduced level. Moreover, strain Pestoides F, which lacks the pla gene, remained virulent when administered with an aerosol (57), although other factors might have replaced the function of Pla. Thus, the observed invasive property of the caf psa double mutant must have been mediated by another surface molecule that is expressed at 37°C. Such a protein could have escaped previous detection with bacteria grown at 26°C, a method commonly used to minimize F1 expression.
It is most likely that several proteins with redundant adhesive or invasive functions appearing specifically under different environmental conditions are involved in the pathogenesis of pneumonic plague. Besides the F1 and Psa surface structures, a panoply of potential fimbrial and nonfimbrial adhesins and invasins has been found by genome sequencing (18, 39, 49). Demonstrating the expression of these proteins and characterizing their role in the pathogenesis of pneumonic plague in the context of other surface proteins will be helpful for the development of effective vaccines against aerosolized Y. pestis.
ACKNOWLEDGMENTS
We thank Jon Goguen, University of Massachusetts Medical School, Worcester, for generously providing Y. pestis strains and for many constructive discussions and Jeff Weiser for the epithelial cells. We are also grateful to Leonard J. Bello for critical reading of the manuscript.
This work was supported by NIH grant 1R21 AI053343-01A1.
FOOTNOTES
Corresponding author. Mailing address: Department of Pathobiology, University of Pennsylvania School of Veterinary Medicine, 3800 Spruce Street, Philadelphia, PA 19104. Phone: (215) 898-1685. Fax: (215) 898-7887. E-mail: dmschiff@vet.upenn.edu.
F.L., H.C., and E.M.G. are co-first authors who contributed equally to this work.
REFERENCES
1. Alexeyev, M. F., I. N. Shokolenko, and T. P. Croughan. 1995. Improved antibiotic-resistance gene cassettes and omega elements for Escherichia coli vector construction and in vitro deletion/insertion mutagenesis. Gene 160:63-67.
2. Anderson, K. L., J. Billington, D. Pettigrew, E. Cota, P. Simpson, P. Roversi, H. A. Chen, P. Urvil, L. du Merle, P. N. Barlow, M. E. Medof, R. A. Smith, B. Nowicki, C. Le Bouguenec, S. M. Lea, and S. Matthews. 2004. An atomic resolution model for assembly, architecture, and function of the Dr adhesins. Mol. Cell 15:647-657.
3. Ben-Efraim, S., M. Aronson, and L. Bichowsky-Slomnichi. 1961. New antigenic component of Pasteurella pestis formed under specified conditions of pH and temperature. J. Bacteriol. 81:704-714.
4. Bichowsky-Slomnicki, L., and S. Bren-Efraim. 1963. Biological activities in extracts of Pasteurella pestis and their relation to the "pH 6 antigen." J. Bacteriol. 86:101-111.
5. Black, D. S., and J. B. Bliska. 2000. The RhoGAP activity of the Yersinia pseudotuberculosis cytotoxin YopE is required for antiphagocytic function and virulence. Mol. Microbiol. 37:515-527.
6. Bliska, J. B. 2000. Yop effectors of Yersinia spp. and actin rearrangements. Trends Microbiol. 8:205-208.
7. Bliska, J. B., and D. S. Black. 1995. Inhibition of the Fc receptor-mediated oxidative burst in macrophages by the Yersinia pseudotuberculosis tyrosine phosphatase. Infect. Immun. 63:681-685.
8. Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L. Heyneker, and H. W. Boyer. 1977. Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2:95-113.
9. Cavanaugh, D. C., and R. Randall. 1959. The role of multiplication of Pasteurella pestis in mononuclear phagocytes in the pathogenesis of flea-borne plague. J. Immunol. 83:348-363.
10. Chang, A. C., and S. N. Cohen. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J. Bacteriol. 134:1141-1156.
11. Charnetzky, W. T., and W. W. Shuford. 1985. Survival and growth of Yersinia pestis within macrophages and an effect of the loss of the 47-megadalton plasmid on growth in macrophages. Infect. Immun. 47:234-241.
12. Chen, T. H., and S. S. Elberg. 1977. Scanning electron microscopic study of virulent Yersinia pestis and Yersinia pseudotuberculosis type 1. Infect. Immun. 15:972-977.
13. Cornelis, G. R. 2000. Molecular and cell biology aspects of plague. Proc. Natl. Acad. Sci. USA 97:8778-8783.
14. Cornelis, G. R. 2002. The Yersinia Ysc-Yop ‘type III’ weaponry. Nat. Rev. Mol. Cell Biol. 3:742-752.
15. Cornelis, G. R., A. Boland, A. P. Boyd, C. Geuijen, M. Iriarte, C. Neyt, M. P. Sory, and I. Stainier. 1998. The virulence plasmid of Yersinia, an antihost genome. Microbiol. Mol. Biol. Rev. 62:1315-1352.
16. Cowan, C., H. A. Jones, Y. H. Kaya, R. D. Perry, and S. C. Straley. 2000. Invasion of epithelial cells by Yersinia pestis: evidence for a Y. pestis-specific invasin. Infect. Immun. 68:4523-4530.
17. Cully, J. F., Jr., A. M. Barnes, T. J. Quan, and G. Maupin. 1997. Dynamics of plague in a Gunnison's prairie dog colony complex from New Mexico. J Wildl. Dis. 33:706-719.
18. Deng, W., V. Burland, G. Plunkett III, A. Boutin, G. F. Mayhew, P. Liss, N. T. Perna, D. J. Rose, B. Mau, S. Zhou, D. C. Schwartz, J. D. Fetherston, L. E. Lindler, R. R. Brubaker, G. V. Plano, S. C. Straley, K. A. McDonough, M. L. Nilles, J. S. Matson, F. R. Blattner, and R. D. Perry. 2002. Genome sequence of Yersinia pestis KIM. J. Bacteriol. 184:4601-4611.
19. Du, Y., R. Rosqvist, and A. Forsberg. 2002. Role of fraction 1 antigen of Yersinia pestis in inhibition of phagocytosis. Infect. Immun. 70:1453-1460.
20. Edwards, R. A., L. H. Keller, and D. M. Schifferli. 1998. Improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene 207:149-157.
21. Galyov, E. E., O. Smirnov, A. V. Karlishev, K. I. Volkovoy, A. I. Denesyuk, I. V. Nazimov, K. S. Rubtsov, V. M. Abramov, S. M. Dalvadyanz, and V. P. Zav'yalov. 1990. Nucleotide sequence of the Yersinia pestis gene encoding F1 antigen and the primary structure of the protein. Putative T and B cell epitopes. FEBS Lett. 277:230-232.
22. Goguen, J. D., W. S. Walker, T. P. Hatch, and J. Yother. 1986. Plasmid-determined cytotoxicity in Yersinia pestis and Yersinia pseudotuberculosis. Infect. Immun. 51:788-794.
23. Goguen, J. D., J. Yother, and S. C. Straley. 1984. Genetic analysis of the low calcium response in Yersinia pestis mu d1(Ap lac) insertion mutants. J. Bacteriol. 160:842-848.
24. Horton, R. M., and L. R. Pease. 1991. Recombination and mutagenesis of DNA sequences using PCR, p. 217-247. In M. J. McPherson (ed.), Directed mutagenesis. IRL Press, Oxford, United Kingdom.
25. Hoschvtzky, H., F. Lottspeich, and K. Jann. 1989. Isolation and characterization of the -galactosyl-1,4--galactosyl-specific adhesin (P adhesin) from fimbriated Escherichia coli. Infect. Immun. 57:76-81.
26. Huang, X. Z., and L. E. Lindler. 2004. The pH 6 antigen is an antiphagocytic factor produced by Yersinia pestis independent of Yersinia outer proteins and capsule antigen. Infect. Immun. 72:7212-7219.
27. Inglesby, T. V., R. Grossman, and T. O'Toole. 2001. A plague on your city: observations from TOPOFF. Clin. Infect. Dis. 32:436-445.
28. Iriarte, M., J. C. Vanooteghem, I. Delor, R. Diaz, S. Knutton, and G. R. Cornelis. 1993. The Myf fibrillae of Yersinia enterocolitica. Mol. Microbiol. 9:507-520.
29. Jacob-Dubuisson, F., J. Heuser, K. Dodson, S. Normark, and S. Hultgren. 1993. Initiation of assembly and association of the structural elements of a bacterial pilus depend on two specialized tip proteins. EMBO J. 12:837-847.
30. Jacobs, A. A. C., and F. K. de Graaf. 1985. Production of K88, K99 and F41 fibrillae in relation to growth phase, and a rapid procedure for adhesin purification. FEMS Microbiol. Lett. 26:15-19.
31. Janssen, W. A., and M. J. Surgalla. 1969. Plague bacillus: survival within host phagocytes. Science 163:950-952.
32. Kool, J. L. 2005. Risk of person-to-person transmission of pneumonic plague. Clin. Infect. Dis. 40:1166-1172.
33. Lahteenmaki, K., M. Kukkonen, and T. K. Korhonen. 2001. The Pla surface protease/adhesin of Yersinia pestis mediates bacterial invasion into human endothelial cells. FEBS Lett. 504:69-72.
34. Lhteenmki, K., R. Virkola, A. Saren, L. Emdy, and T. K. Korhonen. 1998. Expression of plasminogen activator Pla of Yersinia pestis enhances bacterial attachment to the mammalian extracellular matrix. Infect. Immun. 66:5755-5762.
35. Lindler, L. E., M. S. Klempner, and S. C. Straley. 1990. Yersinia pestis pH 6 antigen: genetic, biochemical, and virulence characterization of a protein involved in the pathogenesis of bubonic plague. Infect. Immun. 58:2569-2577.
36. Lindler, L. E., and B. D. Tall. 1993. Yersinia pestis pH6 antigen forms fimbriae and is induced by intracellular association with macrophages. Mol. Microbiol. 8:311-324.
37. Lukaszewski, R. A., D. J. Kenny, R. Taylor, D. G. Rees, M. G. Hartley, and P. C. Oyston. 2005. Pathogenesis of Yersinia pestis infection in BALB/c mice: effects on host macrophages and neutrophils. Infect. Immun. 73:7142-7150.
38. Manoil, C., and J. Beckwith. 1985. TnphoA: a transposon probe for protein export signals. Proc. Natl. Acad. Sci. USA 82:8129-8133.
39. Parkhill, J., B. W. Wren, N. R. Thomson, R. W. Titball, M. T. Holden, M. B. Prentice, M. Sebaihia, K. D. James, C. Churcher, K. L. Mungall, S. Baker, D. Basham, S. D. Bentley, K. Brooks, A. M. Cerdeno-Tarraga, T. Chillingworth, A. Cronin, R. M. Davies, P. Davis, G. Dougan, T. Feltwell, N. Hamlin, S. Holroyd, K. Jagels, A. V. Karlyshev, S. Leather, S. Moule, P. C. Oyston, M. Quail, K. Rutherford, M. Simmonds, J. Skelton, K. Stevens, S. Whitehead, and B. G. Barrell. 2001. Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413:523-527.
40. Perry, R. D., and J. D. Fetherston. 1997. Yersinia pestis—etiologic agent of plague. Clin. Microbiol. Rev. 10:35-66.
41. Pujol, C., and J. B. Bliska. 2003. The ability to replicate in macrophages is conserved between Yersinia pestis and Yersinia pseudotuberculosis. Infect. Immun. 71:5892-5899.
42. Pujol, C., and J. B. Bliska. 2005. Turning Yersinia pathogenesis outside in: subversion of macrophage function by intracellular yersiniae. Clin. Immunol. 114:216-226.
43. Rosqvist, R., K. E. Magnusson, and H. Wolf-Watz. 1994. Target cell contact triggers expression and polarized transfer of Yersinia YopE cytotoxin into mammalian cells. EMBO J. 13:964-972.
44. Silhavy, T. J., M. L. Berman, and L. W. Enquist. 1984. Experiments with gene fusion. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
45. Simonet, M., B. Riot, N. Fortineau, and P. Berche. 1996. Invasin production by Yersinia pestis is abolished by insertion of an IS200-like element within the inv gene. Infect. Immun. 64:375-379.
46. Smith, R. M., and L. Jarett. 1993. Electron microscopic immunocytochemical approaches to the localization of ligands, receptors, tranducers, and transporters, p. 227-264. In F. De Pablo, C. G. Scanes, and B. D. Weintraub (ed.), Handbook of endocrine research techniques. Academic Press, San Diego, Calif.
47. Sodeinde, O. A., A. K. Sample, R. R. Brubaker, and J. D. Goguen. 1988. Plasminogen activator/coagulase gene of Yersinia pestis is responsible for degradation of plasmid-encoded outer membrane proteins. Infect. Immun. 56:2749-2752.
48. Sodeinde, O. A., Y. V. Subrahmanyam, K. Stark, T. Quan, Y. Bao, and J. D. Goguen. 1992. A surface protease and the invasive character of plague. Science 258:1004-1007.
49. Song, Y., Z. Tong, J. Wang, L. Wang, Z. Guo, Y. Han, J. Zhang, D. Pei, D. Zhou, H. Qin, X. Pang, J. Zhai, M. Li, B. Cui, Z. Qi, L. Jin, R. Dai, F. Chen, S. Li, C. Ye, Z. Du, W. Lin, J. Yu, H. Yang, P. Huang, and R. Yang. 2004. Complete genome sequence of Yersinia pestis strain 91001, an isolate avirulent to humans. DNA Res. 11:179-197.
50. Stoker, N. G., N. F. Fairweather, and B. G. Spratt. 1982. Versatile low-copy-number plasmid vectors for cloning in Escherichia coli. Gene 18:335-341.
51. Straley, S. C. 1993. Adhesins in Yersinia pestis. Trends Microbiol. 1:285-286.
52. Straley, S. C., and P. A. Harmon. 1984. Growth in mouse peritoneal macrophages of Yersinia pestis lacking established virulence determinants. Infect. Immun. 45:649-654.
53. Straley, S. C., and P. A. Harmon. 1984. Yersinia pestis grows within phagolysosomes in mouse peritoneal macrophages. Infect. Immun. 45:655-659.
54. Straley, S. C., and R. D. Perry. 1995. Environmental modulation of gene expression and pathogenesis in Yersinia. Trends Microbiol. 3:310-317.
55. Titball, R. W., A. M. Howells, P. C. Oyston, and E. D. Williamson. 1997. Expression of the Yersinia pestis capsular antigen (F1 antigen) on the surface of an aroA mutant of Salmonella typhimurium induces high levels of protection against plague. Infect. Immun. 65:1926-1930.
56. Viboud, G. I., and J. B. Bliska. 2005. Yersinia outer proteins: role in modulation of host cell signalling responses and pathogenesis. Annu. Rev. Microbiol. 59:69-89.
57. Worsham, P. L., and C. Roy. 2003. Pestoides F, a Yersinia pestis strain lacking plasminogen activator, is virulent by the aerosol route. Adv. Exp. Med. Biol. 529:129-131.
58. Yang, Y., J. J. Merriam, J. P. Mueller, and R. R. Isberg. 1996. The psa locus is responsible for thermoinducible binding of Yersinia pseudotuberculosis to cultured cells. Infect. Immun. 64:2483-2489.
59. Zavialov, A. V., J. Berglund, A. F. Pudney, L. J. Fooks, T. M. Ibrahim, S. MacIntyre, and S. D. Knight. 2003. Structure and biogenesis of the capsular F1 antigen from Yersinia pestis: preserved folding energy drives fiber formation. Cell 113:587-596.
60. Zhang, Y., and J. B. Bliska. 2005. Role of macrophage apoptosis in the pathogenesis of Yersinia. Curr. Top. Microbiol. Immunol. 289:151-173.(Fengzhi Liu, Huaiqing Chen, Estela M. Ga)