Evaluation of O-antigen inactivation on Pla activity and virulence of Yersinia pseudotuberculosis harbouring the pPla plasmid

Flavie Pouillot1, Anne Derbise1, Maini Kukkonen2, Jeannine Foulon1, Timo K. Korhonen2 and Elisabeth Carniel1

1 Yersinia Research Unit, Institut Pasteur, 28 rue du Dr Roux, 75724 Paris Cedex 15, France
2 General Microbiology, Faculty of Biosciences, FIN-00014 University of Helsinki, Finland

Correspondence
Elisabeth Carniel
carniel2{at}pasteur.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Yersinia pestis is a species that emerged recently from Yersinia pseudotuberculosis and gained an exceptional pathogenicity potential. Among the major genetic differences between the plague bacillus and its ancestor is the acquisition of the pPla plasmid, which has been associated with the increased virulence of Y. pestis. In a previous study, introduction of pPla into Y. pseudotuberculosis did not lead to any modification of the virulence of the host bacterium. However, it was subsequently demonstrated that the presence of smooth lipopolysaccharide (LPS) inhibits the activity of Pla. In this study, pPla was introduced into a Y. pseudotuberculosis strain expressing smooth LPS, and into a variant in which a mutation that abrogates the formation of O-antigen (O-Ag) repeats (as in natural isolates of Y. pestis) was generated. It was found that in both strains, Pla was synthesized, exported to the bacterial membrane and processed as in Y. pestis. However, the ability of Pla to activate plasminogen was weak and observed only at 37 °C in the smooth strain, while this activity was similar to that of Y. pestis and expressed at both 28 and 37 °C in the O-Ag mutant strain. Similarly, Pla-mediated inactivation of the antiprotease {alpha}2-antiplasmin was not detected in the smooth Y. pseudotuberculosis strain grown at 28 °C, but was expressed at both temperatures in the O-Ag mutant strain. Despite the more efficient activity of Pla, the Y. pseudotuberculosis O-Ag mutant strain exhibited a lower pathogenicity upon subcutaneous infection of mice. The results thus indicate that, although abrogation of O side chain synthesis in a Y. pseudotuberculosis strain harbouring pPla potentiates the two proteolytic activities of Pla, this is not sufficient to confer to Y. pseudotuberculosis a higher pathogenicity potential. These results also suggest that acquisition of pPla may not have been sufficient to confer an immediate higher pathogenic potential to the ancestor Y. pestis strain.


Abbreviations: {alpha}2AP, {alpha}2-antiplasmin; ECM, extracellular matrix; LPS, lipopolysaccharide; O-Ag, O-antigen; sc, subcutaneous; Tmp, trimethoprim


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Yersinia pestis has caused at least three waves of pandemic plague of devastating magnitude. The bacterium still persists endemically in many countries of Africa, Asia and America and is responsible for local outbreaks (World Health Organization, 2003). This Gram-negative bacterium is transmitted from rodent to rodent, and from rodent to human by fleabites. After injection into the dermis of a new host, Y. pestis rapidly disseminates to the draining lymph node, the spleen and the liver, and produces an often fatal systemic infection.

Y. pestis is a highly uniform species, which emerged about 1500 to 20 000 years ago from Yersinia pseudotuberculosis (Achtman et al., 1999), probably from a serotype O : 1b strain (Skurnik et al., 2000). Despite their very close genetic relationship, the two species have distinct epidemiological and clinical features, and a drastically different pathogenic potential. The molecular bases for their difference in virulence are still not understood. Comparison of the genome sequences of representatives of the two species revealed an extensive genetic loss that has probably been more important than acquisition of new genes in the evolution of Y. pestis (Chain et al., 2004). Nonetheless, 32 Y. pestis chromosomal genes and two Y. pestis-specific plasmids have been acquired by Y. pestis since its divergence from Y. pseudotuberculosis. The two Y. pestis-specific plasmids are a 101 kb plasmid called pFra (or pMT1) and a 9·6 kb plasmid called pPla (or pPst or pPCP1).

pFra encodes a phospholipase D (Hinnebusch et al., 2000; Rudolph et al., 1999), which promotes Y. pestis survival in and colonization of the flea midgut (Hinnebusch et al., 2002). The fact that a pFra-cured Y. pestis is still fully virulent for mice (Friedlander et al., 1995) and African green monkeys (Davis et al., 1996), indicates that the main property of this plasmid is not to enhance Y. pestis virulence but to support its flea-borne transmission.

The pPla plasmid encodes at least four proteins (Sodeinde & Goguen, 1988): the bacteriocin pesticin (Pst), its immunity protein (Pim), an IS100 transposase and a plasminogen activator (Pla). Activation of plasminogen by Pla damages the extracellular matrix (ECM) of cultured human epithelial cells (Lähteenmäki et al., 1998). Plasminogen is a circulating precursor of the serine protease plasmin, whose main physiological function is to degrade fibrin clots and ECMs, thus facilitating cell migration (Plow et al., 1999; Saksela, 1985). Pla not only increases plasmin activity by cleaving plasminogen, but also inactivates the antiprotease {alpha}2-antiplasmin ({alpha}2AP), which is the main inhibitor of plasmin (Kukkonen et al., 2001). Pla has been shown to play a role in Y. pestis virulence. Some pla mutants of Y. pestis were severely attenuated (one millionfold) after subcutaneous (sc) infection of mice (Sodeinde et al., 1992). It has been proposed that Pla facilitates Y. pestis dissemination from its site of inoculation by cleaving fibrin deposits that trap the organisms, and by reducing the chemoattraction of inflammatory cells (Sodeinde et al., 1992). However, in another study, no difference in the bacterial proliferation and local inflammatory response induced by wild-type Y. pestis or the pla mutant strain was observed at the site of injection, but a lower number of pla mutants at more distant sites, such as the draining lymph node and spleen, was noted (Welkos et al., 1997). Furthermore, a few Y. pestis isolates lacking pPla were found to be fully virulent (Kutyrev et al., 1989; Samoilova et al., 1996; Welkos et al., 1997), indicating that the role of pPla in Y. pestis pathogenesis remains to be clarified.

To determine whether acquisition of pPla may have been sufficient to confer an increased pathogenicity potential to the Y. pestis ancestor, Kutyrev et al. (1999) introduced this plasmid into Y. pseudotuberculosis to evaluate the pathogenicity of the recombinant strains for mice infected subcutaneously. Although Pla was produced and expressed on the cell surface, no significant modification of the LD50 of the pPla harbouring Y. pseudotuberculosis recombinant strains, as compared to the wild-type strain, was observed, suggesting that the presence of this plasmid is not sufficient to confer additional pathogenicity potential to the host strain. However, it was subsequently demonstrated that O-antigen (O-Ag) repeats in the lipopolysaccharide (LPS) of Y. pseudotuberculosis sterically inhibit plasminogen activation by Pla (Kukkonen et al., 2004). The lack of Pla activity due to the presence of O-Ag repeats on the surface of Y. pseudotuberculosis could thus have accounted for the previously observed absence of increased pathogenicity of the pPla harbouring Y. pseudotuberculosis strains.

In this study, a mutation that abrogates the formation of O-Ag repeats (as in natural isolates of Y. pestis) was introduced into the chromosome of a pPla-harbouring Y. pseudotuberculosis strain. In this background, the ability of Pla to activate plasminogen and to degrade {alpha}2AP was investigated, and the impact of pPla on the virulence of the recombinant strain for mice was evaluated.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and growth conditions.
The bacterial strains and plasmids used in this study are listed in Table 1. Bacteria were grown in Luria–Bertani (LB), brain–heart infusion or Müller-Hinton (MH) broth, and on LB, LB-Sac (without NaCl and supplemented with 10 % sucrose) or MH agar. Yersinia strains were grown at 28 or 37 °C, and Escherichia coli strains at 37 °C. Chloramphenicol (25 µg ml–1), kanamycin (30 µg ml–1) or trimethoprim (Tmp; 100 µg ml–1) were added to the media as necessary. Selection for Tmp resistance was done in MH broth. All experiments involving virulent Y. pestis CO92 were performed in a BL3 laboratory.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids

 
DNA and gene manipulation.
Genomic and plasmid DNA were extracted with the Isoquick nucleic acid extraction kit, and the Qiagen Plasmid Maxi Kit, respectively. Mutagenesis of chromosomal target genes and insertion of antibiotic-resistance genes onto the plasmid were performed as described previously (Derbise et al., 2003). Recombinant plasmid DNA and PCR fragments were introduced into E. coli by transformation or into Yersinia by electroporation, as described by Conchas & Carniel (1990).

PCR.
The primers used in this study are listed in Table 2. PCRs were performed with 1 unit of Taq polymerase (Roche) or 1 unit of a 3 : 1 mixture of Taq and Pfu (Stratagene) polymerases in the supplier's buffer. PCR amplification reaction mixtures contained 10 µM each primer and 1 mM dNTPs. The PCR program involved one step at 95 °C for 5 min, followed by 30 cycles of amplification of three steps at (i) 95 °C for 30 s, (ii) 55 °C for 30 s and (iii) 72 °C for 1–3 min, depending on the fragment length. PCR products were maintained at 72 °C for 5 min, electrophoresed in 1 % agarose gels, and stained with ethidium bromide. Amplification of the kan cassette with long flanking homologous regions of the Yersinia target DNA was done with the PCR program described by Derbise et al. (2003).


View this table:
[in this window]
[in a new window]
 
Table 2. Primers used for PCR amplification

Lower-case letters correspond to sequences located on the kan (primer pair OKf1/OKr2) or dfr (primer pair 258A/258B) antibiotic-resistance genes.

 
Generation of mutant strains.
Labelling of pPla with a dfr cassette encoding resistance to Tmp was carried out following the (short flanking homologous regions) SFH-PCR procedure (Derbise et al., 2003). Briefly, the dfr locus was amplified by PCR with the primer pair 258A/258B, using pGP704N-dfr (Lesic & Carniel, 2005) as a template. The amplification product consisted of the entire dfr gene and promoter region, flanked by 55 bp homologous to an intergenic region on pPla (I-Region). This PCR product was introduced into Y. pestis CO92 (pKOBEG-sacB) by electroporation. Colonies in which allelic exchange between the linear PCR product and the plasmid target sequence had occurred were selected on agar plates containing Tmp. Correct insertion of the dfr cassette was checked by PCR with primers 260B/260C, which are located on each side of the inserted fragment (Fig. 1a).



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 1. Genetic maps of pPla, and the LPS chromosomal locus with the sites of allelic exchange between the target gene and an antibiotic-resistance cassette. (a) pPla with the insertion of a Tmp-resistance cassette in the I-Region. (b) The O-Ag gene cluster with the replacement of the gmd and fcl loci by a kan-resistance cassette. Arrows indicate the location and direction of the primers used to check proper allelic exchanges.

 
Introduction of pPla into E. coli LE392 and Y. pseudotuberculosis IP32953 was done by electroporation and mutants were selected on Tmp plates. The integrity of the plasmid was checked by digestion with BamHI, yielding the expected 9·9 kb linear plasmid fragment.

To obtain a Y. pseudotuberculosis strain that does not synthesize O side chains, primer pairs OKF/OKf1 and OKR/OKr2 were used to amplify the region encompassing the gmd and fcl genes, and primer pair 136A/136B to amplify the kan cassette (Table 2). Following the (long flanking homologous regions) LFH-PCR procedure (Derbise et al., 2003), a linear PCR fragment composed of the kan gene with its promoter region, and flanked by {approx}550 bp DNA homologous to the regions located on each side of gmd and fcl, was generated. This fragment was introduced by electroporation into strain IP32953(pKOBEG-sacB) and correct allelic exchange between the PCR fragment and the chromosomal gene was checked by PCR with primer pairs OKup/167, OKdown/166 and OKup/OKdown (Table 2, Fig. 1b). Y. pseudotuberculosis IP32953 clones with the appropriate O-Ag mutation and cured of pKOBEG-sacB were selected on LB-Sac agar plates.

The IP32953(pPla) O-Ag variant was obtained by introducing pPla-Tmp into IP32953 O-Ag by electroporation. Presence of a plasmid with the appropriate size was checked by digestion of the plasmid extract with BamHI.

Protein analysis.
To determine whether Pla is present in bacterial cell-envelope preparations, bacteria grown overnight were centrifuged, and cell suspensions were adjusted to an OD600 1·2 in PBS. Bacteria were sonicated four times for 30 s over crushed ice. Unbroken cells were removed by low speed centrifugation (1800 g), and the cell envelopes were pelleted by centrifugation at 15 000 g for 10 min at 4 °C. Total membrane proteins were separated by SDS-PAGE in 12·5 % gels, the proteins were transferred onto PVDF membranes (Hybond), and the Pla polypeptides were detected using 1 : 500 dilutions of primary anti-His6-Pla antisera (Kukkonen et al., 2001) and 1 : 10 000 dilutions of secondary anti-rabbit-IgG alkaline-phosphatase conjugate. Bound antibodies were detected with the phosphatase substrate (ECL Plus Kit, Amersham).

Extraction and analysis of LPS.
Bacteria grown overnight at 37 or 25 °C on LB plates were collected and resuspended in 2 ml PBS. The bacterial suspension was adjusted to OD420 0·4 and subjected to proteolysis for 2 h at 60 °C by the addition of 40 µg proteinase K ml–1. An aliquot (1·5 ml) of this suspension was centrifuged for 3 min at 1 200 g, and the pellet was resuspended in 50 µl lysis buffer [2 % SDS, 4 % {beta}-mercaptoethanol, 10 % (v/v) glycerol, 1 M Tris-HCl pH 6·8, bromphenol blue] and heated at 100 °C for 10 min. Ten microlitres of this extract was subjected to SDS-PAGE in a 15 % polyacrylamide gel and transferred onto PVDF membranes. The LPS was detected using 1 : 1000 dilution of a primary rabbit anti-LPS antiserum raised against Y. pseudotuberculosis V2812/79 (kindly provided by Dr M. Skurnik, University of Helsinki, Helsinki, Finland), and 1 : 10 000 dilution of secondary anti-rabbit IgG alkaline-phosphatase conjugate. Bound antibodies were detected with the phosphatase substrate.

Plasminogen activation and {alpha}2AP degradation.
Kinetic measurement of plasminogen activation was performed as described by Kukkonen et al. (2001), by incubating 2x108 bacteria, 4 µg human Glu-plasminogen (American Diagnostica) and the chromogenic plasmin substrate S-2251 (Val-Leu-Lys-p-nitroaniline dihydrochloride; Chromogenix) in a total volume of 200 µl at 37 °C. Breakdown of the chromogenic substrate was measured in a microtitre-plate reader at A405. The A405 values were determined at time intervals of 15 min. The level of {alpha}2AP inactivation was measured as described by Lähteenmäki et al. (2005). Briefly, 5 µg {alpha}2AP ml–1 was incubated overnight at 37 °C with 109 bacteria in 105 µl PBS. Gentamicin (1 µg ml–1) was added to prevent bacterial growth. After the addition of 100 µl PBS, bacteria were removed by centrifugation and 168 µl supernatant was transferred into a well of a microtitre plate. Human plasmin (2·5 µg ml–1; Sigma) and the plasmin substrate S-2251 (0·45 mM; Chromogenix) were added, and plasmin activity was measured at A405 as the breakdown of the substrate after 90 min incubation at 37 °C. The control for {alpha}2AP activity was produced by incubating {alpha}2AP with PBS instead of bacteria, and plasmin activity was checked in the absence of bacteria and {alpha}2AP. All measurements were carried out in duplicate and the experiments were repeated twice.

Animal infections.
Prior to infection, the presence in each Yersinia strain of known unstable genetic elements, such as the plasmid(s) and the high-pathogenicity island, was systematically verified by PCR with primer pairs located on these various elements: 18/19 (irp2), 37A/37B (caf1), 159A/159B (pla) and 160A/160B (yopM) (Table 2). Groups of five 5-week-old OF1 female mice (Iffa Credo) were inoculated subcutaneously with 0·1 ml bacterial suspension. Lethality was recorded daily for 3 weeks. The LD50 was determined according to the method of Reed & Muench (1935). To determine whether pPla-Tmp is stably maintained in Y. pseudotuberculosis during growth in vivo, three mice were infected subcutaneously with 108 c.f.u. of the IP32953(pPla) recombinant strain. Moribund animals were euthanized, their spleens were removed aseptically, crushed in saline and streaked onto LB plates. Individual colonies (100) were spotted onto MH and MH-Tmp agar plates. The percentage of TmpR colonies were recorded. All experiments were performed in a biosafety level 3 animal facility.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Construction of Y. pseudotuberculosis recombinant strains carrying pPla-Tmp and/or a mutation in O-Ag biosynthesis genes
In order to introduce the pPla plasmid into Y. pseudotuberculosis IP32953, the Y. pestis CO92 pPla was first labelled with a dfr cassette inserted at position 2071–2091 on the plasmid sequence (Parkhill et al., 2001). This insertion site, designated I-Region, was chosen because it is located between the IS100 coding sequence and the replication locus (Fig. 1a), in an apparently non-coding region. To ensure that this insertion did not affect Y. pestis virulence, Y. pestis CO92 strains carrying either the natural pPla or the labelled pPla-Tmp were injected subcutaneously into mice and their virulence was compared. The LD50 of the Y. pestis strains harbouring the wild-type pPla (29 c.f.u.) and the labelled pPla-Tmp (32 c.f.u.) were similar, indicating that the presence of the antibiotic-resistance cassette did not affect pPla-virulence-associated properties. In addition to pPla, Y. pestis harbours two other resident plasmids: pFra and pYV. In order to purify the pPla-Tmp molecules present in Y. pestis CO92, the plasmid content of this strain was extracted and introduced into E. coli LE392 by electroporation. Plasmid extracts of TmpR E. coli were checked and subsequently used to introduce pPla-Tmp into Y. pseudotuberculosis IP32953 by electroporation, yielding IP32953(pPla).

Since the presence of O-Ag repeats was found to sterically inhibit plasminogen activation by Pla (Kukkonen et al., 2004), a Y. pseudotuberculosis derivative with a mutation in the O-Ag biosynthesis pathway was constructed. For this purpose, most of the gmd and fcl genes which encode a GDP-mannose-4,6-dehydratase and a GDP-fucose synthetase, respectively, were replaced by a kan cassette (Fig. 1b). These two genes were chosen because they are naturally inactivated in the O-Ag gene cluster of Y. pestis (Skurnik et al., 2000), and they act at an early step of O-side-chain assembly (in the biosynthesis of the repeat unit sugar precursors). Absence of O-Ag synthesis in the resulting IP32953 O-Ag mutant strain was evidenced by immunoblotting with an anti-LPS antibody. While the wild-type strain synthesized O side chains at 25 °C, the O-Ag mutant strain was unable to do so (Fig. 2). pPla-Tmp was subsequently introduced into IP32953 O-Ag, yielding IP32953(pPla) O-Ag.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 2. Western blot of the LPS extract of the Y. pseudotuberculosis wild-type strain IP32953 and the O-Ag mutant strain grown at either 25 or 37 °C, detected with anti-LPS antibodies.

 
Presence of Pla in Y. pseudotuberculosis membranes
Immunoblotting of bacterial membrane proteins with anti-His6-Pla antibodies revealed the presence of Pla in the cell envelopes of both IP32953(pPla) and IP32953(pPla) O-Ag, in amounts similar to those observed in Y. pestis, and its absence in strains that do not carry pPla (Fig. 3). The presence of Pla in membranes was detected in bacteria grown at either 28 °C (data not shown) or 37 °C (Fig. 3). Pla is known to undergo C-terminal autoprocessing in Y. pestis, leading to two major forms designated {alpha}-Pla and {beta}-Pla (Kukkonen et al., 2001). Similar autoprocessed forms of Pla were detected in the recombinant Y. pseudotuberculosis strains harbouring pPla (Fig. 3). Therefore, as observed by Kutyrev et al. (1999), when pPla is introduced into Y. pseudotuberculosis, Pla is synthesized, exported to the bacterial membrane and processed as in Y. pestis.



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 3. Western blot of cell envelope proteins from Yersinia strains grown at 37 °C, detected with anti-His6-Pla antisera. {alpha}-Pla is the mature form of the protein, {beta}-Pla is autoprocessed at a single site at its C-terminus. Numbers on the left indicate the size of the molecular mass markers (kDa). M, Size marker.

 
Plasminogen activation by the various Y. pseudotuberculosis mutants
As expected, in the absence of pPla, Y. pseudotuberculosis strains (wild-type and the O-Ag derivative) did not exhibit any plasminogen-activation activity, at either 28 or 37 °C (Fig. 4). In contrast, the recombinant IP32953(pPla) strain was able to activate plasminogen. However, this was observed only when the bacteria were grown at 37 °C, and the activity was much lower than that observed in Y. pestis (Fig. 4). In the absence of O-Ag, plasminogen activation by IP32953(pPla) O-Ag was much more efficient, and almost reached the levels detected in wild-type Y. pestis. This activity was observed when recombinant bacteria were grown at either 28 or 37 °C (Fig. 4). These data are in accordance with previous results indicating that expression of O-Ag prevents plasminogen activation by Pla (Kukkonen et al., 2004), and they show that a mutation preventing the formation of O-Ag repeats in Y. pseudotuberculosis allows a Pla activity similar to that of Y. pestis.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 4. Plasminogen-activation activity of Y. pestis and Y. pseudotuberculosis strains grown at 28 or 37 °C. Bacteria were incubated with human plasminogen and a chromogenic plasmin substrate. Breakdown of the chromogenic substrate was measured as an A405 value, at time intervals of 15 min. The experiments were performed twice in duplicate and the SDs are indicated by bars. {blacklozenge}, Y. pestis; {bullet}, Y. pseudotuberculosis IP32953; {circ}, IP32953(pPla); {square}, IP32953(pPla) O-Ag; {blacksquare}, IP32953 O-Ag.

 
{alpha}2AP degradation by Y. pestis and the various Y. pseudotuberculosis mutant strains
The ability of Pla to inactivate {alpha}2AP was evaluated by incubating the various Y. pestis and Y. pseudotuberculosis strains with {alpha}2AP and testing the capacity of the {alpha}2AP remaining in the supernatant to inhibit plasmin. In the absence of pPla, the Y. pseudotuberculosis supernatant did not prevent plasmin inhibition by {alpha}2AP, at either 28 or 37 °C (Fig. 5). In the presence of pPla, a plasmin activity was detected, indicating that {alpha}2AP was inactivated by IP32953(pPla). This inactivation was as efficient as that of Y. pestis, but only at a temperature of 37 °C (Fig. 5). In contrast, in the O-Ag mutant of Y. pseudotuberculosis, inactivation of {alpha}2AP was observed at both 28 and 37 °C, and was similar to that exhibited by Y. pestis (Fig. 5). These results demonstrate that, as for plasminogen activation, Pla-mediated degradation of {alpha}2AP is inhibited in the presence of O-Ag repeats in Y. pseudotuberculosis.



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 5. {alpha}2AP inactivation by Y. pestis and Y. pseudotuberculosis IP32953 (Y.PST) strains grown at 28 or 37 °C. Bacteria were incubated overnight with {alpha}2AP, and the supernatant was incubated for 90 min at 37 °C with human plasmin and the plasmin substrate. Plasmin activity was measured as breakdown of the substrate at A405. The positive control for plasmin activity was incubation without bacteria and with no {alpha}2AP added. The positive control for {alpha}2AP activity was incubation without bacteria. Vertical bars represent the SDs of experiments performed twice in duplicate.

 
Virulence of the various Y. pseudotuberculosis mutants for mice
In order to determine whether acquisition of pPla might have an impact on Y. pseudotuberculosis virulence, tenfold dilutions of the wild-type strain IP32953, and of the three derivatives (pPla, O-Ag and pPla O-Ag), were injected subcutaneously into mice in groups of 5, and mortality was recorded daily for 21 days. Although not natural for Y. pseudotuberculosis, the sc route of infection was used in order to mimic the Y. pestis mode of transmission, and determine whether acquisition of pPla facilitates Y. pseudotuberculosis dissemination from its site of inoculation and enhances mouse lethality. We noted that mouse lethality did not correlate with the number of bacteria when low doses of Y. pseudotuberculosis were inoculated (i.e. a 1000-fold increase in bacterial numbers did not lead to an increase in animal death). However, a correlation between bacterial numbers and mouse lethality was observed for bacterial doses >=105 c.f.u. This is in marked contrast to Y. pestis for which a good association between the number of bacteria injected and animal death is observed even at low doses. In order to compare the virulence of the various Y. pseudotuberculosis derivatives upon sc injection into mice, bacterial inocula of >=105 c.f.u. were thus used. Strain IP32953(pPla) had an LD50 (2·1x105 c.f.u.) of the same order of magnitude as that of the wild-type strain (6·1x105 c.f.u.), thus confirming that introduction of pPla into Y. pseudotuberculosis does not significantly enhance its virulence for mice. To rule out the possibility that pPla-Tmp was lost by Y. pseudotuberculosis during growth in vivo, the spleens of IP32953(pPla)-infected mice were removed and processed, and the resultant percentage of TmpR colonies was determined. One hundred per cent of the colonies from three different animals were TmpR, indicating that pPla was stably maintained during the infectious process. The O-Ag mutant strain exhibited a {approx}10-fold increase in LD50 (5·8x106 c.f.u.), suggesting that O-side-chain expression plays some role in Y. pseudotuberculosis virulence. The IP32953(pPla) O-Ag mutant strain had the higher LD50 (1·4x107 c.f.u.), which was {approx}23-fold higher than that of the wild-type strain. Analysis of the kinetics of mouse lethality also indicated that animal death was delayed for both the IP32953 O-Ag and IP32953(pPla) O-Ag mutant strains (Fig. 6). Therefore, the absence of O-Ag expression alters the virulence of Y. pseudotuberculosis and this defect is not compensated by the presence of pPla.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 6. Kinetics of mouse lethality upon sc injection of 107 c.f.u. of wild-type and three derivatives of Y. pseudotuberculosis IP32953. Five mice were infected for each strain, and the number of mice still alive was recorded daily. {blacklozenge}, IP32953; {bullet}, IP32953(pPla); {triangleup}, IP32953 O-Ag; {circ}, IP32953(pPla) O-Ag.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Y. pestis is a clonal species which emerged fewer than 20 000 years ago from Y. pseudotuberculosis (Achtman et al., 1999). During its short evolution, Y. pestis has acquired a pathogenic potential quite exceptional in the bacterial kingdom. However, the molecular bases for this extreme pathogenicity are still not understood. One of the major genetic differences between the plague bacillus and its recent ancestor Y. pseudotuberculosis is the acquisition of two plasmids, pFra and pPla. While the prominent function of pFra appears to be to enhance transmission of Y. pestis by fleas, pPla seems to play a role in the virulence of this species (Sodeinde et al., 1992). Of the four proteins encoded by this plasmid, only Pla was shown to be important for Y. pestis virulence (Sodeinde et al., 1992). The major function attributed to this outer-membrane protein is to activate plasminogen and to destroy the main control system for plasmin proteolysis. The uncontrolled proteolysis created can be targeted to the ECM and/or fibrin, potentially enhancing bacterial migration (reviewed by Korhonen et al., 2004).

Whether or not acquisition of pPla during Y. pestis evolution has been sufficient to confer an increased pathogenicity to this species was a key question. In an attempt to answer it, Kutyrev et al. (1999) introduced pPla into Y. pseudotuberculosis. They observed that Pla was expressed at the surface of the host strains and was processed to a smaller form designated {beta}-Pla, as is known to be the case in Y. pestis (Kukkonen et al., 2001; Sodeinde & Goguen, 1988). In agreement with these results, we found in this study that introduction of pPla into Y. pseudotuberculosis led to the synthesis of Pla, to levels similar to those of Y. pestis, to its export to the bacterial membrane, and to its autoprocessing. When we looked at the plasminogen activation activity of Pla, we observed that this activity was detected in cells grown at 37 °C but not in those grown at 28 °C. In contrast to Kutyrev et al. (1999), we found that plasminogen activation at 37 °C was much lower than that of Y. pestis. These results may be explained by the recent demonstration that Pla requires rough LPS to activate plasminogen, but is inhibited in the presence of O-Ag repeats (Kukkonen et al., 2004).

Y. pseudotuberculosis is known to exhibit temperature-dependent O-Ag expression (Bengoechea et al., 1998). At 28 °C, O-Ag repeats are present on the conserved core oligosaccharide and could thus prevent plasminogen activation by IP32953(pPla). At 37 °C, the LPS is rough and could allow functional Pla molecules to be present on the surface of the recombinant bacterium. The decreased plasminogen activation in IP32953(pPla), as compared to Y. pestis, might be attributed to a reduced number, but not a complete absence, of O side chains on the LPS of Y. pseudotuberculosis grown at 37 °C. This hypothesis is reinforced by our observation that in Y. pestis, which naturally harbours a completely rough LPS because of mutations in genes involved in O-side-chain assembly (Skurnik et al., 2000), plasminogen activation activity is similar in bacteria grown at 28 and 37 °C. Further demonstration that the O-Ag repeats are responsible for the absence of plasminogen activation by IP32953(pPla) grown at 28 °C, and for a low activity in cells grown at 37 °C, was obtained by introducing a mutation that abrogates the formation of O-Ag repeats (as in natural isolates of Y. pestis) in the chromosome of IP32953(pPla). The IP32953(pPla) O-Ag mutant strain exhibited a temperature-independent plasminogen-activation activity that was much higher than that of the parental strain IP32953(pPla), and was close to that found in Y. pestis. Altogether these results demonstrate that a mutation that abrogates the synthesis of the O side chains in Y. pseudotuberculosis harbouring pPla allows a plasminogen-activation activity similar to that of Y. pestis, at both 28 and 37 °C.

In addition to increasing plasmin production by plasminogen activation, Pla also overcomes inhibition of plasmin by the serine protease inhibitor {alpha}2AP (Kukkonen et al., 2001). In the human circulation, unbound plasmin is rapidly inactivated by plasma antiproteases, the main one being {alpha}2AP. We found that {alpha}2AP inactivation was effective in Y. pseudotuberculosis IP32953(pPla) grown at 37 °C, but was not detectable in bacteria grown at 28 °C. In contrast, {alpha}2AP degradation was similar in Y. pestis grown at 28 and 37 °C. These results again suggest a role for the LPS O side chains on the temperature-dependent capacity of Pla to inactivate {alpha}2AP in Y. pseudotuberculosis. The demonstration that an O-Ag derivative of IP32953(pPla) acquired the capacity to degrade {alpha}2AP to the same extent as Y. pestis when grown at either 28 or 37 °C confirmed this hypothesis.

Therefore, Pla-mediated increased production of plasmin, resulting from both plasminogen activation and {alpha}2AP degradation, is prevented at 28 °C in Y. pseudotuberculosis because of the presence of LPS O side chains, but is observable at 37 °C, a temperature at which O-Ag repeats are reduced or absent in vitro. However, this inhibition of O-side-chain formation at 37 °C does not appear to occur in vivo. This is demonstrated by the fact that during human infections with Y. pseudotuberculosis, high titres of antibodies directed against the O-Ag repeats of the various serotypes of this species are detected (Chung et al., 1997; Schmidt, 1965; Sizaret & Mollaret, 1968; Splino et al., 1969). The in vivo production of a smooth LPS may thus prevent the activity of Pla when a Y. pseudotuberculosis strain harbouring pPla infects its host. This could potentially explain why no difference in virulence was observed by Kutyrev et al. (1999) between wild-type Y. pseudotuberculosis and smooth pPla-harbouring derivatives. We thus used an O-Ag mutant of IP32953(pPla), which was shown to exhibit plasminogen-activation and {alpha}2AP-degradation activities similar to those of Y. pestis, to compare its virulence to that of the wild-type strain and of the IP32953(pPla) and IP32953 O-Ag variants. First, our results confirmed the absence of a difference in mouse lethality between the pPla-harbouring and non-harbouring Y. pseudotuberculosis smooth strains. Second, they indicated that absence of O-Ag expression results in a slight decrease in Y. pseudotuberculosis pathogenicity. The role of LPS in Y. pseudotuberculosis virulence was not demonstrated until now, but was indirectly suggested by results of in vivo gene expression technologies (Darwin, 2004). Our data further suggest that O-side-chain production is necessary for the expression of full virulence in Y. pseudotuberculosis, at least by the sc route of infection. Third, our results showed that introduction of pPla into the IP32953 O-Ag variant does not increase the pathogenicity of Y. pseudotuberculosis. On the contrary, the presence of pPla further decreased the virulence of the IP32953 O-Ag mutant strain and increased the time to death of the animals.

Altogether, our results indicate that in vitro, plasminogen-activation and {alpha}2AP-degradation activities can reach levels similar to those of Y. pestis in a Y. pseudotuberculosis variant harbouring pPla and lacking O-side-chain repeats. However, the efficient activity of Pla is not sufficient to confer a higher pathogenic potential to the Y. pseudotuberculosis host strain. The deleterious effect of a rough LPS on Y. pseudotuberculosis pathogenicity demonstrated in this study, may compensate the virulence-promoting effect of pPla. Furthermore, due to its protease activity, Pla may degrade proteins important for Y. pseudotuberculosis pathogenicity. This has been shown for YadA (Kutyrev et al., 1999) and may be true for additional surface-exposed virulence factors. Although pPla may be necessary for the full virulence of most Y. pestis strains, this plasmid may not be sufficient by itself to explain the difference in the virulence of Y. pestis and Y. pseudotuberculosis. Additional genetic changes (mutations, deletions and/or gene acquisition) that occurred in Y. pestis upon its divergence from Y. pseudotuberculosis most likely play a critical role in the virulence of the plague bacillus. Therefore, acquisition of pPla by the Y. pestis ancestor may not have been sufficient to confer an immediate higher pathogenic potential to the host strain. Interplay between the acquired Pla protease and other bacterial components may have gradually led to subtle modifications of the Y. pestis ancestor's genetic background, and finally to the emergence of the highly pathogenic plague bacillus.


   ACKNOWLEDGEMENTS
 
F. P. was the recipient of a grant from the French ‘Ministère de la Recherche et de la Technologie’. T. K. K. was supported by the Academy of Finland (grant numbers 105824, 211300, and the Microbes and Man Research Programme). We thank Corinne Fayolle for her help during some animal experiments.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Achtman, M., Zurth, K., Morelli, C., Torrea, G., Guiyoule, A. & Carniel, E. (1999). Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc Natl Acad Sci U S A 96, 14043–14048.[Abstract/Free Full Text]

Bengoechea, J. A., Brandenburg, K., Seydel, U., Diaz, R. & Moriyon, I. (1998). Yersinia pseudotuberculosis and Yersinia pestis show increased outer membrane permeability to hydrophobic agents which correlates with lipopolysaccharide acyl-chain fluidity. Microbiology 144, 1517–1526.[Medline]

Chain, P. S., Carniel, E., Larimer, F. W. & 20 other authors (2004). Insights into the evolution of Yersinia pestis through whole-genome comparison with Yersinia pseudotuberculosis. Proc Natl Acad Sci U S A 101, 13826–13831.[Abstract/Free Full Text]

Chung, S.-I., Kim, W.-Y., Kim, K.-J. & Choi, C.-S. (1997). Seropositive rate to Yersinia enterocolitica-pseudotuberculosis complex among patients with acute gastroenteritis in children, 1994-1997. J Korean Soc Microbiol 32, 701–715.

Conchas, R. F. & Carniel, E. (1990). A highly efficient electroporation system for transformation of Yersinia. Gene 87, 133–137.[CrossRef][Medline]

Darwin, A. (2004). Identification of Yersinia genes expressed during host infection. In Yersinia, Molecular and Cellular Biology, pp. 149–168. Edited by E. Carniel & B. J. Hinnebusch. Norfolk: Horizon Bioscience.

Davis, K. J., Fritz, D. L., Pitt, M. L., Welkos, S. L., Worsham, P. L. & Friedlander, A. M. (1996). Pathology of experimental pneumonic plague produced by fraction 1-positive and fraction 1-negative Yersinia pestis in African green monkeys (Cercopithecus aethiops). Arch Pathol Lab Med 120, 156–163.[Medline]

Derbise, A., Lesic, B., Dacheux, D., Ghigo, J. M. & Carniel, E. (2003). A rapid and simple method for inactivating chromosomal genes in Yersinia. FEMS Immunol Med Microbiol 38, 113–116.[CrossRef][Medline]

Friedlander, A. M., Welkos, S. L., Worsham, P. L., Andrews, G. P., Heath, D. G., Anderson, G. W., Pitt, M. L. M., Estep, J. & Davis, K. (1995). Relationship between virulence and immunity as revealed in recent studies of the F1 capsule of Yersinia pestis. Clin Infect Dis 21, S178–S181.[Medline]

Hinnebusch, J., Cherepanov, P., Du, Y., Rudolph, A., Dixon, J. D., Schwan, T. & Forsberg, A. (2000). Murine toxin of Yersinia pestis shows phospholipase D activity but is not required for virulence in mice. Int J Med Microbiol 290, 483–487.[Medline]

Hinnebusch, B. J., Rudolph, A. E., Cherepanov, P., Dixon, J. E., Schwan, T. G. & Forsberg, A. (2002). Role of Yersinia murine toxin in survival of Yersinia pestis in the midgut of the flea vector. Science 296, 733–735.[Abstract/Free Full Text]

Korhonen, T. K., Kukkonen, M., Virkola, R., Lång, H., Suomalainen, M., Kyllonen, P. & Lähteenmäki, K. (2004). The plasminogen activator Pla of Yersinia pestis: localized proteolysis and systemic spread. In Yersinia, Molecular and Cellular Biology, pp. 349–362. Edited by E. Carniel & B. J. Hinnebusch. Norfolk: Horizon Bioscience.

Kukkonen, M., Lähteenmäki, K., Suomalainen, M., Kalkkinen, N., Emödy, L., Lång, H. & Korhonen, T. K. (2001). Protein regions important for plasminogen activation and inactivation of {alpha}2-antiplasmin in the surface protease Pla of Yersinia pestis. Mol Microbiol 40, 1097–1111.[CrossRef][Medline]

Kukkonen, M., Suomalainen, M., Kyllonen, P., Lähteenmäki, K., Lång, H., Virkola, R., Helander, I. M., Holst, O. & Korhonen, T. K. (2004). Lack of O-antigen is essential for plasminogen activation by Yersinia pestis and Salmonella enterica. Mol Microbiol 51, 215–225.[CrossRef][Medline]

Kutyrev, V. V., Filippov, A. A., Shavina, N. & Protsenko, O. A. (1989). Genetic analysis and simulation of the virulence of Yersinia pestis. Mol Gen Mikrobiol Virusol 8, 42–47.[Medline]

Kutyrev, V., Mehigh, R. J., Motin, V. L., Pokrovskaya, M. S., Smirnov, G. B. & Brubaker, R. R. (1999). Expression of the plague plasminogen activator in Yersinia pseudotuberculosis and Escherichia coli. Infect Immun 67, 1359–1367.[Abstract/Free Full Text]

Lähteenmäki, K., Virkola, R., Saren, A., Emödy, L. & Korhonen, T. K. (1998). Expression of plasminogen activator Pla of Yersinia pestis enhances bacterial attachment to the mammalian extracellular matrix. Infect Immun 66, 5755–5762.[Abstract/Free Full Text]

Lähteenmäki, K., Kyllonen, P., Partanen, L. & Korhonen, T. K. (2005). Antiprotease inactivation by Salmonella enterica released from infected macrophages. Cell Microbiol 7, 529–538.[CrossRef][Medline]

Lesic, B. & Carniel, E. (2005). Horizontal transfer of the high-pathogenicity island of Yersinia pseudotuberculosis. J Bacteriol 187, 3352–3358.[Abstract/Free Full Text]

Parkhill, J., Wren, B. W., Thomson, N. R. & 32 other authors (2001). Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413, 523–527.[CrossRef][Medline]

Plow, E. F., Ploplis, V. A., Carmeliet, P. & Collen, D. (1999). Plasminogen and cell migration in vivo. Fibrinol Proteol 13, 49–53.[CrossRef]

Reed, L. J. & Muench, H. (1935). A simple method of estimation fifty per cent endpoints. Am J Hyg 27, 493–497.

Rudolph, A. E., Stuckey, J. A., Zhao, Y., Matthews, H. R., Patton, W. A., Moss, J. & Dixon, J. E. (1999). Expression, characterization, and mutagenesis of the Yersinia pestis murine toxin, a phospholipase D superfamily member. J Biol Chem 274, 11824–11831.[Abstract/Free Full Text]

Saksela, O. (1985). Plasminogen activation and regulation of pericellular proteolysis. Biochim Biophys Acta 823, 35–65.[CrossRef][Medline]

Samoilova, S. V., Samoilova, L. V., Yezhov, I. N., Drozdov, I. G. & Anisimov, A. P. (1996). Virulence of pPst(+) and pPst(-) strains of Yersinia pestis for guinea-pigs. J Med Microbiol 45, 440–444.[Abstract]

Schmidt, V. J. (1965). Untersuchungen über den fluoreszenzserologischen Nachweis von Pasteurella pseudotuberculosis-Antikörpenrn. Pathol Microbiol 28, 21–25.

Sizaret, P. & Mollaret, H. H. (1968). Sensibilité et spécificité comparées des réactions d'agglutination et d'immunofluorescence indirecte vis-à-vis du bacille de Malassez et Vignal "Pasteurella pseudotuberculosis". In International Symposium on Pseudotuberculosis, pp. 197–206. Edited by R. H. Regamey, A. de Barbieri, W. Hennessen, D. Ikié, F. T. Perkins & J. Ungar. Basel: S. Karger.

Skurnik, M., Peippo, A. & Ervela, E. (2000). Characterization of the O-antigen gene clusters of Yersinia pseudotuberculosis and the cryptic O-antigen gene cluster of Yersinia pestis shows that the plague bacillus is most closely related to and has evolved from Yersinia pseudotuberculosis serotype O : 1b. Mol Microbiol 37, 316–330.[CrossRef][Medline]

Sodeinde, O. A. & Goguen, J. D. (1988). Genetic analysis of the 9·5-kilobase virulence plasmid of Yersinia pestis. Infect Immun 56, 2743–2748.[Medline]

Sodeinde, O. A., Subrahmanyam, Y. V. B. K., Stark, K., Quan, T., Bao, Y. D. & Goguen, J. D. (1992). A surface protease and the invasive character of plague. Science 258, 1004–1007.[Medline]

Splino, M., Kyntera, F., Peychl, L. & Kotrlik, J. (1969). Ein beitrag zur serologie und epidemiologie der humanen Pasteurella-Pseudotuberkulose. Pathol Anat Inst Med Fak der Karlssuniversität 209, 204–211.

Welkos, S. L., Friedlander, A. M. & Davis, K. J. (1997). Studies on the role of plasminogen activator in systemic infection by virulent Yersinia pestis strain C092. Microb Pathog 23, 211–223.[CrossRef][Medline]

World Health Organization (2003). Human plague in 2000 and 2001. Wkly Epidemiol Rec 78, 130–135.[Medline]

Received 16 June 2005; revised 10 August 2005; accepted 26 August 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Pouillot, F.
Articles by Carniel, E.
PubMed
PubMed Citation
Articles by Pouillot, F.
Articles by Carniel, E.
Agricola
Articles by Pouillot, F.
Articles by Carniel, E.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.