HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplementary figures 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 Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Hares, M. C. Articles by Waterfield, N. Search for Related Content PubMed PubMed Citation Articles by Hares, M. C. Articles by Waterfield, N. Agricola Articles by Hares, M. C. Articles by Waterfield, N.

The Yersinia pseudotuberculosis and Yersinia pestis toxin complex is active against cultured mammalian cells

¹ University of Bath, Department of Biology and Biochemistry, Claverton Down, Bath BA2 7AY, UK
² University of Exeter in Cornwall, Department of Biosciences, Tremough Campus, Penryn, Cornwall TR10 9EZ, UK
³ London School of Hygiene and Tropical Medicine, London WC1E 7HT, UK
⁴ CNRS-UPR9022, Institut de Biologie Moléculaire et Cellulaire, 15 rue René Descartes, 67084 Strasbourg Cedex, France

Correspondence
Stewart Hinchliffe
S.J.Hinchliffe{at}exeter.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The toxin complex (Tc) genes were first identified in the insect pathogen Photorhabdus luminescens and encode 1 MDa protein complexes which are toxic to insect pests. Subsequent genome sequencing projects have revealed the presence of tc orthologues in a range of bacterial pathogens known to be associated with insects. Interestingly, members of the mammalian-pathogenic yersiniae have also been shown to encode Tc orthologues. Studies in Yersinia enterocolitica have shown that divergent tc loci either encode insect-active toxins or play a role in colonization of the gut in gastroenteritis models of rats. So far little is known about the activity of the Tc proteins in the other mammalian-pathogenic yersiniae. Here we present work to suggest that Tc proteins in Yersinia pseudotuberculosis and Yersinia pestis are not insecticidal toxins but have evolved for mammalian pathogenicity. We show that Tc is secreted by Y. pseudotuberculosis strain IP32953 during growth in media at 28 °C and 37 °C. We also demonstrate that oral toxicity of strain IP32953 to Manduca sexta larvae is not due to Tc expression and that lysates of Escherichia coli BL21 expressing the Yersinia Tc proteins are not toxic to Sf9 insect cells but are toxic to cultured mammalian cell lines. Cell lysates of E. coli BL21 expressing the Y. pseudotuberculosis Tc proteins caused actin ruffles, vacuoles and multi-nucleation in cultured human gut cells (Caco-2); similar morphology was observed after application of a lysate of E. coli BL21 expressing the Y. pestis Tc proteins to mouse fibroblast NIH3T3 cells, but not Caco-2 cells. Finally, transient expression of the individual Tc proteins in Caco-2 and NIH3T3 cell lines reproduced the actin and nuclear rearrangement observed with the topical applications. Together these results add weight to the growing hypothesis that the Tc proteins in Y. pseudotuberculosis and Y. pestis have been adapted for mammalian pathogenicity. We further conclude that Tc proteins from Y. pseudotuberculosis and Y. pestis display differential mammalian cell specificity in their toxicity.

A supplementary figure showing sequence alignments of Tc proteins is available with the online version of this paper.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The high molecular mass (1 MDa) toxin complexes (Tc's) are dominant secreted virulence factors of the insect pathogen and nematode symbiont Photorhabdus luminescens (Bowen et al., 1998; Bowen & Ensign, 1998). They were originally described in P. luminescens subsp. akhursti strain W14, where they were shown to comprise four native complexes encoded by the loci tca, tcb, tcc and tcd, with the individual genes assigned according to their order within each locus, e.g. tcaA, tcaB, tcaC, etc. (Bowen et al., 1998).

Comparative sequence analysis has revealed the four loci to have a high degree of similarity that can be grouped into three genetic components: the tcaAB/tcdA-like or [A] components, the tcaC/tcdB-like or [B] components and the tccC-like or [C] components (ffrench-Constant & Waterfield, 2006; Waterfield et al., 2001b). The Tc's have previously been demonstrated to possess both oral and injectable toxicity to lepidopteran caterpillar pests (Blackburn et al., 1998; Bowen et al., 1998). However, the precise mechanism of how they elicit this toxic response remains to be determined. There is evidence to suggest that the Tc's cause massive cell death of the insect midgut; when the Tca complex was fed to first-instar caterpillar larvae of the model insect host Manduca sexta, cessation of feeding occurred within 24 h, followed by death after several days. Histopathological examination of the larvae showed that the columnar cells of the anterior midgut swell apically, extruding large cytoplasmic vesicles 3 h post-feeding. After 6 h, apical ‘blebs’ containing nuclei and vacuoles also occur. The disintegration of the epithelium is complete after 12 h (Blackburn et al., 1998; Silva et al., 2002).

Subsequently, heterologous expression of the individual Tc proteins in Escherichia coli revealed that the TcdA subunit (an [A]) alone was capable of demonstrating toxicity to M. sexta (Waterfield et al., 2001a) but that it was significantly less toxic than the native complex. In addition, transgenic expression of tcdA in Arabidopsis thaliana also produced toxicity, although again this was not associated with the high levels of oral activity of the native protein complexes (Liu et al., 2003). Subsequent work by Waterfield et al. (2005) and Sergeant et al. (2003) in Xenorhabdus nematophilus showed that when heterologously expressed, in order for high toxicity to be established not only must all three subunits be present but the [BC] subunit genes must be co-expressed in the same E. coli bacterial cytoplasm.

Tc orthologues have since been found in other Gram-negative bacteria including Serratia entomophila and X. nematophilus (Hurst et al., 2000; Sergeant et al., 2003) and have also been found in the Gram-positive bacterium Paenibacillus (Enright & Griffin, 2005). This implies that the Tc's may be important in insect pathogens. Interestingly, they have also been found in several Yersinia species which are not known to be insect pathogens (Parkhill et al., 2001; Hinchliffe et al., 2003; Chain et al., 2004; Tennant et al., 2005; Bresolin et al., 2006a).

Three species of the genus Yersinia are pathogenic to humans. Yersinia pestis, transmitted by a flea, is the causative agent of bubonic plague, and Yersinia pseudotuberculosis and Yersinia enterocolitica, which are intestinal pathogens, cause symptoms such as diarrhoea, fever and abdominal pain (reviewed by Wren, 2003). Interestingly, although Y. pestis and Y. pseudotuberculosis cause different diseases, they share 97 % amino acid sequence identity. Indeed, Y. pestis is believed to have evolved from Y. pseudotuberculosis as little as 1500–20 000 years ago whereas Y. enterocolitica is far more distantly related (Achtman et al., 1999).

For a comprehensive discussion on the evolution of the Yersinia Tc toxins see Waterfield et al. (2007); however, we will discuss these genes briefly below. To date, tc coding sequences have only been identified in a very small percentage of Y. enterocolitica strains. Of these, Y. enterocolitica W22703 has orthologues to tcaA, tcaB, tcaC and tccC, which are located on a pathogenicity island, TcPAI, and are toxic to M. sexta. Insertion knockouts of tcaA resulted in a loss of activity, indicating that this gene is directly involved in the observed insecticidal toxicity. The tc genes of Y. enterocolitica W22703 were expressed 4.6-fold higher when the bacteria were grown at 10 °C compared to 37 °C (Bresolin et al., 2006b), which suggests that Y. enterocolitica may use the Tc proteins in the soil environment or that the bacterium is hosted by an insect during its life cycle. Conversely, divergent tc gene orthologues identified in Y. enterocolitica strain T83 were not expressed at low temperature and tc gene knockouts resulted in an inability to colonize the rat gut or invade CHO cells (Tennant et al., 2005). These findings suggest that different strains of Y. enterocolitica may use Tc proteins either in the soil environment against invertebrates or in the mammalian gut. The tc genes in these two strains are significantly divergent in gene sequence and operon structure (Waterfield et al., 2007), suggesting that different Tc's are specific to different hosts.

Unlike Y. enterocolitica where a few strains have independently acquired pathogenicity islands containing tc orthologoues, a single tc locus appears to be present in all Y. pseudotuberculosis and Y. pestis strains. The genome sequence of Y. pestis strain CO92 first revealed orthologues of tcaA, tcaB, tcaC and two copies of tccC with a further two orthologues of tccC located elsewhere in the genome. This locus has since been identified in all strains of Y. pestis sequenced so far or by comparative genomics using microarray analysis (Deng et al., 2002; Hinchliffe et al., 2003; Song et al., 2004). Similarly, microarray analysis and sequencing of Y. pseudotuberculosis strains has shown that this locus is also present (Hinchliffe et al., 2003; Chain et al., 2004). However, in Y. pseudotuberculosis only one copy of tccC is associated with this locus (Chain et al., 2004). The amino acid sequences of Y. pestis and Y. pseudotuberculosis Tc proteins are approximately 97 % identical, with between 35 and 70 % identity to the P. luminescens Tc's. It has been suggested that Y. pseudotuberculosis Tc's may be insecticidal toxins and that the divergence observed between Y. pseudotuberculosis and Y. pestis evolved to enable Y. pestis to persist within the rat flea Xenopsylla cheopis (Parkhill et al., 2001; Wren, 2003; Waterfield et al., 2004). However, while Y. pseudotuberculosis is able to colonize the midgut of the rat flea and cause chronic infection, displaying symptoms such as diarrhoea, the heterologously expressed Y. pseudotuberculosis TcaATcaB [A] and TcaCTccC [BC] were found to lack toxicity to the flea. Furthermore, knockouts of these tc genes in Y. pseudotuberculosis strains did not diminish flea toxicity (Erickson et al., 2006, 2007). This adds weight to the hypothesis that the Tc's in Y. pestis and Y. pseudotuberculosis are not insecticidal but may instead be mammalian toxins. The Tc toxins of Y. pestis have been suggested to be secreted via a type III secretion system indicative of immunomodulatory toxins although no functional data supporting this hypothesis have been published (Gendlina et al., 2007).

Here we present what is believed to be the first investigation into the biological role of the tc genes of Y. pseudotuberculosis and Y. pestis. Our data provide evidence that the Tc's in Y. pseudotuberculosis and Y. pestis have no demonstrable insecticidal activity but show differential toxicity to cultured mammalian cells, indicating a role in mammalian pathogenicity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains and growth media.
Y. pseudotuberculosis strains IP32953 and YPIII pIB1 were obtained from E. Carniel (Institute Pasteur, Paris, France) and H. Wolf-Watz (Umeå University, Umeå, Sweden) respectively. Other Y. pseudotuberculosis strains used for sequencing were obtained from H. Fukushima (Shimane Prefectural Institute of Public Health and Environmental Science, Matsue, Japan). All strains were maintained on CIN agar supplemented with Yersinia selective supplement (Oxoid) and cultures were grown in LB broth with appropriate antibiotics unless otherwise stated.

Cloning and sequencing of Y. pseudotuberculosis Tc coding sequences.
Full-length tcaA, tcaB and tcaC coding sequences were PCR amplified from six different strains of Y. pseudotuberculosis and the Y. pestis KIM⁺ strain. Primers TcaA-Myc, TcaB-Myc and TcaC-Myc were used to amplify the coding sequences in conjunction with TcaA-STOP, TcaB-STOP and TcaC-STOP respectively (Table 1). PCRs were performed using Accuprime pfx DNA polymerase (Invitrogen) to ensure high fidelity and were cloned into the CMV-promoter-based eukaryotic expression vector RK5myc using BamHI and XbaI, for later transfection into the Caucasian colon adenocarcinoma (Caco-2) cell line (Fogh et al., 1977). Cloned products were then sequenced using the ABI Big Dye Terminator v3.1 cycle sequencing kit and an ABI 3730 DNA analyser.

Heterologous expression of Tc components.
Expression and assays were performed essentially as described previously (Waterfield et al., 2005). Briefly, glycerol stocks were streaked onto LB agar plates containing 100 µg ampicillin (Amp) ml^–1. Colonies were picked and grown in 5 ml LB containing 0.4 % glucose and Amp overnight at 30 °C, 300 r.p.m. Then 0.5 ml of cells was harvested and used to inoculate 50 ml LB+Amp and grown at 37 °C, 250 r.p.m. to OD₆₀₀ 0.7. Cultures were induced with 0.4 % arabinose and cooled to 15 °C, at which temperature they were grown for 24 h. Cells were harvested by centrifugation at 4000 r.p.m., 4 °C. The cell pellet was resuspended in 10 ml sterile 1x PBS and physically lysed by sonication (45 mA, 10 s on/10 s off for a total 2 min on ice using a Branson 450 digital sonifier fitted with a tapered probe). Lysates were analysed by 6 % SDS-PAGE to confirm the presence of the expressed Tc proteins. These same lysates were subsequently split in two: one half was used in insect oral bioassays, and the other for cytosolic preparation and in topical bioassays as described below.

Western blotting to determine expression and secretion of TcaCTccC [BC].
Y. pseudotuberculosis strain IP32953 was used to inoculate 5 ml LB broth at 28 °C overnight with shaking. Then 0.5 ml was harvested and used to inoculate 50 ml LB broth and grown at 37 °C, 28 °C or 10 °C for 72 h. One millilitre of cell suspension was harvested and the OD₆₀₀ adjusted to 1.5. The pellet was then harvested and resuspended in 0.05 ml SDS loading buffer (50 mM Tris/HCl, pH 6.8, 4 %, w/v, SDS, 0.2 % w/v bromophenol blue, 20 % glycerol and 5 % β-mercaptoethanol). The remaining cell suspension was harvested and supernatant was collected. The supernatant was subsequently concentrated using an Amicon Ultra 15 centrifugal concentrator (Millipore) and total protein determined using Bradford assay according to the manufacturer's instructions (Sigma) and adjusted to 4 mg ml^–1. Equal volumes of whole-cell extract and concentrated supernatant were subjected to 6 % SDS-PAGE, then transferred to Immunoblot PVDF membrane (Bio-Rad). The membrane was incubated with an anti-P. luminescens TcdB rabbit polyclonal antiserum raised against the C-terminal peptide RRTSTGVPNYDETDC, followed by an anti-rabbit IgG alkaline-phosphatase-conjugated antibody (Sigma) and visualized with the BCIP/NBT liquid substrate system (Sigma).

Expression levels of the heterologously expressed Y. pseudotuberculosis TcaCTccC (pBAD30-YptbBC), Y. pestis TcaCTccC (pBAD30-YpeBC) and P. luminescens TcdBTccC (pBAD30-tcdBtccC) were also determined by Western blotting.

Generation of tcaAB : : Km insertion mutants.
The mutation method was a variation on the -Red recombinase method (Datsenko & Wanner, 2000) and utilized plasmid pAJD434, which has been optimized for efficiency in Yersinia ssp. (Maxson & Darwin, 2004). Plasmids and primers used for the mutant generation are listed in Table 1. Gene-specific primers were designed to introduce a kanamycin resistance cassette into the intended coding sequence. These 70-mer oligonucleotides consisted of 50 bases of gene-specific sequence at the 5' end of the primer whilst the 3' end consisted of 20 bases designed to amplify the kanamycin cassette from the pUC4K plasmid. After optimization, the single PCR product was purified using a Qiagen PCR Purification kit and digested with DpnI to remove the plasmid template.

Wild-type strains were electroporated with the -Red plasmid pAJD434 and grown at 28 °C in LB broth containing 0.8 % (w/v) L-arabinose to maintain a high level of expression of the bacteriophage red-gam genes. Once cells had reached OD₆₀₀ 0.7, they were incubated on ice for 20 min before being washed twice in ice-cold H₂O and resuspended with the 50 µl purified PCR product. Cells were then electroporated, transferred to a bijou bottle containing 500 µl LB+0.8 % (w/v) L-arabinose and incubated overnight at 28 °C before being plated onto selective plates. Mutations were verified by PCR and Southern blotting. Retention of the virulence plasmid pYV was confirmed by PCR and by colony appearance on CRMOX plates (Riley & Toma, 1989). pAJD434 is thermosensitive and thus retention of the plasmid was selected against by growth of mutants at 37 °C in LB broth containing 2.5 mM CaCl₂ to prevent the loss of the virulence plasmid pYV. Complementation was performed by transformation of the pBAD30-YptbA plasmid in these mutant strains.

Oral toxicity of Y. pseudotuberculosis strains to M. sexta.
Y. pseudotuberculosis wild-type and insertion mutants were grown in 5 ml LB at 28 °C overnight with shaking. An aliquot (0.5 ml) was harvested and used to inoculate 50 ml of the same medium and grown at 28 °C for 72 h (OD₆₀₀=1.5). Aliquots (200 µl; 2x10⁸ cells) of these whole cultures were then topically applied to 1 cm³ blocks of artificial wheatgerm M. sexta diet. Treated food blocks were allowed to dry at room temperature for 20 min before two neonate larvae were applied to each food block and incubated at 25 °C for 7 days to allow the larvae to feed. After this time the larvae were weighed and scored for mortality.

Similarly lysates from 50 ml cultures of BL21 cells expressing pBAD30-YptbA, pBAD30-YptbBC, pBAD30-tcdA or pBAD30-tcdBtccC were prepared as described above. Equal volumes of pBAD30-YptbA and pBAD30-YptbBC lysates were mixed as were equal volumes of pBAD30-tcdA and pBAD30-tcdBtccC lysates. These were then incubated at room temperature for 20 min before 200 µl aliquots were applied to food blocks as above.

Topical application to cultured cell lines.
Lysates prepared as described above were centrifuged at 20 000 r.p.m., 4 °C for 20 min (Beckman Coulter) in sterile 50 ml Oakridge centrifuge tubes. Supernatants were collected. Concentration was determined by SDS-PAGE and Bradford assay (Sigma), and samples were adjusted to 4 mg ml^–1. Caco-2 cells were seeded into six-well plates containing borosilicate glass coverslips (BDH) and grown for 24 h in 2 ml Dulbecco's Modified Eagle's Medium (DMEM) (Invitrogen) supplemented with 10 % fetal calf serum, 1 % non-essential amino acids (Sigma) and 1 % penicillin and streptomycin at 37 °C until cells were confluent. A 250 µl aliquot of BL21 TcaATcaB lysate was combined with an equal volume of TcaCTccC lysate (TcdBTccC for P. luminescens) and incubated at room temperature for 20 min. This mixture was then diluted with 500 µl serum-free DMEM and added to cells refreshed in 1 ml DMEM before incubation for 24 h at 37 °C with 95 % air/5 % carbon dioxide (v/v). After incubation, cells were washed with PBS, fixed with 4 % paraformaldehyde and permeabilized with 0.02 % Tween 20. Cells were washed three times with PBS before being incubated with either 50 µg FITC-phalloidin ml^–1 or 50 µg TRITC-phalloidin ml^–1 (Sigma) for 1 h at room temperature in the dark, followed by 10 min incubation with 0.12 mg Hoechst 33258 DAPI stain (Sigma) ml^–1. After washing twice with PBS and twice with dH₂O, coverslips were mounted in Dako Cytomation fluorescent mounting medium, examined and photographed with an Olympus BX61 microscope. These conditions were similar for the NIH3T3 mouse fibroblast cell line except fetal calf serum was replaced with calf serum. Sf9 (Invitrogen) cells were treated as above except cells were incubated in Grace's medium with amino acids, yeastolate and hydrosylate (Invitrogen) and supplemented with 25 µg gentamicin (Sigma) ml^–1 and incubated at 28 °C in air.

In situ apoptosis detection.
Heterologously expressed BL21 Tc lysates were applied to Caco-2 cells as described above. After the 24 h incubation at 37 °C, cells were stained via TdT-mediated dUTP nick end label (TUNEL) following the manufacturer's instructions (Roche Applied Science), followed by Hoechst 33258 DAPI stain, before being mounted and visualized as described.

Mammalian cell transfection.
Caco-2 cells were seeded into six-well plates containing borosilicate glass coverslips (BDH) and grown for 24 h in 2 ml DMEM (Invitrogen) supplemented with 10 % fetal calf serum, 1 % non-essential amino acids (Sigma) and 1 % penicillin and streptomycin and grown at 37 °C, 95 % air/5 % carbon dioxide (v/v). These conditions were similar for the NIH3T3 mouse fibroblast cell line except that fetal calf serum was replaced with calf serum. The RK5myc mammalian expression vector containing the Tc fusion and an N-terminal Myc epitope tag (see ‘Cloning and sequencing of Y. pseudotuberculosis Tc coding sequences’) were co-transfected with a GFP-fusion construct, using GeneJuice Transfection reagent (Novagen), according to the manufacturer's protocol. The GFP-actin construct chosen was the pEGFP-actin, a mammalian expression vector expressing the EGFP-actin fusion protein of the human codon optimized variant green fluorescent protein and the gene encoding human cytoplasmic β-actin (Clontech). After 24 h the samples were fixed with 4 % paraformaldehyde (w/v) in PBS, permeabilized with 0.2 % Triton X-100 and blocked with four drops of Image-iT FX signal enhancer (Molecular Probes). Cells were labelled with monoclonal anti-Myc primary antibody (Invitrogen), followed with Alexa Fluor 594 goat anti-mouse dye-labelled secondary antibody (Molecular Probes) to detect the Myc protein. The cell nucleus was visualized by incubating in Hoechst 33258 DAPI stain (Sigma), at a final concentration of 0.12 mg ml^–1 in PBS for 10 min at room temperature in the dark. Slides were washed twice in PBS followed by two washes in water. Coverslips were mounted in Dako Cytomation fluorescent mounting medium and examined and photographed with an Olympus BX61 microscope.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Sequence alignment of Yersinia Tc proteins
In order to investigate the genetic diversity of tc genes in different serotypes of Y. pseudotuberculosis, we sequenced paralogues from several strains, as described. Sequence alignments of TcaA, TcaB and TcaC revealed a reasonable degree of conservation (see Supplementary Fig. S1a, b and c, available with the online version of this paper). However, we do see differences in the level of conservation for the different proteins of the complex. TcaC is the most conserved of the three protein sequences, showing 96 % identity between even the most diverse orthologues. The TcaA proteins show a higher degree of variation, which appears to be mainly in the C-terminal regions and may suggest some selective diversification. TcaB shows the highest degree of diversification, especially in the region between amino acids 201 and 426.

Expression and secretion of the Tc
Previous RT-PCR results have demonstrated that the Tc components are preferentially transcribed at 30–37 °C. In order to show that the proteins are actually produced at these temperatures and are secreted by wild-type Y. pseudotuberculosis, we performed SDS-PAGE and Western blot analysis using a cross-reactive rabbit antiserum raised against the TcdB component of P. luminescens (Fig. 1a). Expression and secretion of the Tc was observed at 30 °C and 37 °C, with the complex being present in whole-cell extracts and supernatants. These data demonstrate that the proteins are secreted during growth in normal media and the size of the observed protein suggests that the secreted complex contains all the Tc components as expected.

View larger version (92K): [in this window] [in a new window]   Fig. 1. Expression and secretion of Y. pseudotuberculosis TcaCTccC. (a) Western blot of Y. pseudotuberculosis wild-type strain IP32953 grown at 15 °C, 20 °C, 28 °C or 37 °C. Equal total protein concentrations as determined by Bradford assay of whole-cell pellet and concentrated supernatant were analysed. Samples were loaded as follows: M, standard size marker; 1, 15 °C whole-cell extract; 2, 15 °C supernatant; 3, 20 °C whole-cell extract; 4, 20 °C supernatant; 5, 28 °C whole-cell extract; 6, 28 °C supernatant; 7, 37 °C whole-cell extract; 8, 37 °C supernatant; C, BL21 pBAD30-YptbBC control. (b) Western blot showing levels of expression from E. coli BL21 transformed with (1) pBAD30-YpeBC, (2) pBAD30-YptbBC and (3) pBAD30-tcdBtccC.

Oral toxicity of Y. pseudotuberculosis to M. sexta
Oral toxicity to M. sexta has previously been observed for Y. pseudotuberculosis strain IP32953 but not for other Y. pseudotuberculosis strains tested (Waterfield et al., 2007) In order to determine if this toxicity could be attributed to Tc proteins, a tcaAB : : Km insertion mutant was generated in strain IP32953 and in the non-orally toxic strain YPIII pIB1. Wild-type and mutant strains were then fed to M. sexta neonates as described. As can be seen in Fig. 2(a), there was no observable difference in oral toxicity between the wild-type strains and their knockouts. Contrary to previous findings, no increase in mortality of caterpillars was observed for Y. pseudotuberculosis IP32953 wild-type (5 %) or YPIII pIB1 wild-type (10 %). However, there is a small increase in mortality for both the Y. pseudotuberculosis IP32953 tcaAB : : Km mutant (15 %) and YPIII pIB1 tcaAB : : Km mutant (20 %) when compared to the negative control (10 %).

View larger version (35K): [in this window] [in a new window]   Fig. 2. Effect of Tc's on insects. (a) Oral bioassay of wild-type Y. pseudotuberculosis strains IP32953 and YPIII pIB1 and their respective tcaAB : : Km knockouts. The histogram bars show the relative weight gain of M. sexta neonate larvae fed with cell lysates added to artificial diet. Weight gain of different treatments is expressed relative to a pBAD30 BL21 (negative control; always the value of 1) and standard errors of each mean are shown. Columns: 1, BL21 pBAD30 control; 2, BL21 lysates containing P. luminescens [A]+[BC]; 3, BL21 lysates containing Y. pseudotuberculosis [A]+[BC]; 4, YPIII pIB1; 5, YPIII pIB1 tcaAB : : Km; 6, IP32953; 7, IP32953 tcaAB : : Km. The black line shows the percentage of dead caterpillars. (b) Fluorescence micrographs of Sf9 insect cells topically treated with cytosolic preparations of A, BL21 pBAD30 control; B, BL21 lysates containing P. luminescens [A]+[BC]; C, BL21 lysates containing Y. pestis [A]+[BC]; or D, BL21 lysates containing Y. pseudotuberculosis [A]+[BC]. Cells were stained for F-actin with FITC-phalloidin, and for nuclei with Hoechst 33258. Arrows indicate typical cells observed (A, C) or apoptotic nuclei with heavily condensed actin (B). (c) Graph representing percentage of cells with actin condensation (white bars) and apoptotic nuclei (grey bars), where n=minimum of 300 cells. Columns: 1, BL21 pBAD30 control; 2, BL21 lysates containing Y. pseudotuberculosis [A]+[BC]; 3, BL21 lysates containing Y. pestis [A]+[BC]; 4, BL21 lysates containing P. luminescens [A]+[BC].

Toxicity of Y. pseudotuberculosis Tc's to Spodoptera fragiperala insect cell line Sf9
Lysates from E. coli BL21 expressing the [A] and [BC] components from Y. pseudotuberculosis and Y. pestis were compared to lysates from E. coli BL21 expressing the [A] and [BC] components from P. luminescens when topically applied to the insect cell line Sf9 over a 24 h period. Cells were stained for F-actin and cell nuclei and were examined using fluorescence microscopy. Marked actin condensation (86 %) and apoptotic nuclei (94 %) were observed when BL21 lysates containing P. luminescens Tc proteins were topically applied to cells, whereas limited effects were observed for BL21 lysates containing Y. pseudotuberculosis (34 % and 36 %) and Y. pestis (22 % and 27 %) Tc proteins (Fig. 2b and c). These were comparable to the vector control (28 % and 27 % respectively).

Effect of Y. pseudotuberculosis wild-type and tcaAB : : Km mutants topically applied to the Caco-2 cell line
Incubation of wild-type Y. pseudotuberculosis strains IP32953 and YPIII pIB1 with Caco-2 cells produced similar phenotypic results to those observed for the lysates of BL21 cells expressing Y. pseudotuberculosis Tc proteins (Fig. 3). Vacuolation was observed in 26.5 % and 29 % of cells for wild-type Y. pseudotuberculosis strains IP32953 and YPIII pIB1 respectively, whereas the cytosolic preparations produced vacuolation in 29 % of Caco-2 cells whilst only 4.4 % of control cells demonstrated this phenotype. Interestingly, the Y. pseudotuberculosis strains IP32953 and YPIII pIB1 tcaAB : : Km mutants produced significantly less vacuolation than wild-type (6.6 % and 11.8 %), which indicates that the Tc's are involved in the cellular toxicity exhibited by Y. pseudotuberculosis.

View larger version (99K): [in this window] [in a new window]   Fig. 3. Fluorescence micrographs of Caco-2 cells topically treated with (a, b) BL21 lysates containing Y. pseudotuberculosis [A]+[BC], (c) BL21 lysates containing P. luminescens [A]+[BC], (d) BL21 lysates containing Y. pestis [A]+[BC], (e) IP32953, (f) IP32953 tcaAB : : Km, (g) YPIII pIB1, (h) YPIII pIB1 tcaAB : : Km or (j) BL21 pBAD30 control. Cells were stained for F-actin with FITC-phalloidin or TRITC-phalloidin, and for nuclei with Hoechst 33258. Arrows indicate membrane ruffles (a), actin vacuoles (b), multinucleation (e) and actin condensation (g) typically seen in Caco-2 cells treated with Y. pseudotuberculosis containing TC's or heterologously expressed Y. pseudotuberculosis [A]+[BC]. (h) Punctuate actin seen with cells treated with any Y. pseudotuberculosis strain. (i) Graph representing percentage of cells with vacuoles/membrane ruffles (white bars) and nuclei fragmentation (grey bars), where n=minimum of 300 cells. Columns: 1, BL21 pBAD30 control; 2, BL21 lysates containing Y. pseudotuberculosis [A]+[BC]; 3, BL21 lysates containing Y. pestis [A]+[BC]; 4, BL21 lysates containing P. luminescens [A]+[BC].

View larger version (46K): [in this window] [in a new window]   Fig. 4. Fluorescence micrographs of NIH3T3 cells topically treated with (a) BL21 lysates containing Y. pseudotuberculosis [A]+[BC], (b) BL21 lysates containing P. luminescens [A]+[BC], (c) BL21 lysates containing Y. pestis [A]+[BC], or (d) BL21 pBAD30 control. Cells were stained for F-actin with FITC-phalloidin, and for nuclei with Hoechst 33258. Arrows indicate punctuate actin (red arrows; b), or actin condensation (white arrows; b and c). (e) Graph representing percentage of actin condensation (white bars) and nuclear fragmentation (grey bars), where n=minimum of 300 cells. Columns: 1, BL21 pBAD30 control; 2, BL21 lysates containing P. luminescens [A]+[BC]; 3, BL21 lysates containing Y. pseudotuberculosis [A]+[BC]; 4, BL21 lysates containing Y. pestis [A]+[BC].

View larger version (85K): [in this window] [in a new window]   Fig. 5. TUNEL staining of intoxicated cells to measure apoptosis levels. Cells were TUNEL stained to determine nuclear strand breaks, which indicate apoptosis (a, b and d). The nuclei were co-stained with Hoechst 33258 DAPI stain. (a) Negative control, (b) BL21 lysates containing Y. pseudotuberculosis [A]+[BC] on Caco-2 cells; (d) BL21 lysates containing Y. pestis [A]+[BC] on NIH3T3 cells. Arrows indicate co-stained cells indicative of apoptotic and TUNEL-positive cells. However, co-staining is not seen in all cases, suggesting a combination of necrosis and apoptosis. (c) Graph showing that 26 % of cells were TUNEL positive for Y. pseudotuberculosis, compared to 0.8 % in the negative control and 46 % cells in the positive control. For Y. pestis, 19.4 % of cells were TUNEL positive compared to 2.9 % in the negative control and 75 % in the positive control. White bars, Caco-2 cells; grey bars, NH3T3 cells.

After 24 h the Myc-tagged Y. pseudotuberculosis TcaA protein accumulated in the cytoplasm and caused ‘membrane ruffling’ in the actin cytoskeleton of 18.7 % cells (control 9.3 %). However, most cells (64.3 %) showed a heavily condensed actin cytoskeleton (control 7.3 %) with 12 % showing nuclear fragmentation (Fig. 6b, f, j and n), which was not observed in cells transfected with pEGFP-actin alone (0 %) (Fig. 6a, e, i and m). These two phenotypes may represent early- and late-stage intoxication effects, as transfection and 24 h incubation does not give synchronous cell effects. Transient expression of TcaB produced a very distinctive phenotype (Fig. 6c, g, k and o). Heavy actin condensation was observed in 78.8 % of cells, with membrane ruffling/vacuolation observed at the edges of the actin cytoskeleton in the remainder of cells (21 %). The Myc-tag accumulated in the cytoplasm. More interestingly, multinucleation was observed in 66.7 % of cells. Transient expression of TcaC also produced a highly distinctive phenotype, with large actin vacuoles and nuclear fragmentation (28 % of cells had vacuoles compared to 2 % observed in the controls) (Fig. 6d, h, l and p), whilst the cMyc-tagged protein accumulated in the nucleus. Non-synchronous effects were again observed, with 53.5 % of cells showing heavy condensation (7.3 % in controls). Reassuringly, the cytotoxic effects seen with tcaA, tcaB and tcaC resembled the effects seen by topical application of the BL21 lysates containing the [A]+[BC] complex.

View larger version (88K): [in this window] [in a new window]   Fig. 6. Intracellular transient expression of tc toxin genes from Y. pseudotuberculosis IP32953 in Caco-2 cells. (a–d) Actin visualization with pEGFP. (e–h) Detection of cMyc-proteins, visualized with Alexa Fluor 594 goat anti-mouse dye-labelled secondary antibody (Molecular Probes). (i–l) Nuclei were visualized with Hoechst 33258 DAPI stain. (m–p) Merge. (a, e, i and m) Control cells transfected with pEGFP alone. Control shows normal actin and nuclear morphology with no cMyc co-localization. (b, f, j and n) Cells transfected with pRK5myc-tcaA and pEGFP-actin. tcaA shows membrane ruffling which results in actin condensation (arrows; b). (f) RK5myc-tcaA tag shows localization to the cytoplasm. Nuclei are normal (j). Note the intense yellow staining corresponding to co-localization of myc-tagged tcaA and pEGFP-actin (n). (c, g, k and o) Cells transfected with pRK5myc-tcaB and pGEFactin. tcaB expression results in membrane ruffling/vacuoles and actin condensation (c and o). Interestingly, as observed in topical applications, multinucleation is observed in 70 % of cells (k and o). The RK5myc-tcaB appears to co-localize with the cytoplasm, although this is difficult to determine in heavily condensed cells. (d, h, l and p) Cells transfected with pRK5myc-tcaC and pEGFP-actin. RK5myc-tcaC shows large vacuoles (white arrow; d and h). Here the RK5myc-tcaC tag co-localizes to the nucleus (h), and causes nuclear aberrations (l and p).

View larger version (90K): [in this window] [in a new window]   Fig. 7. Intracellular transient expression of tc toxin genes from Y. pestis KIM+ in NIH3T3. (a–d) Actin visualization with pEGFP. (e–h) Detection of cMyc-proteins, visualized with Alexa Fluor 594 goat anti-mouse dye-labelled secondary antibody (Molecular Probes). (i–l) Nuclei were visualized with Hoechst 33258 DAPI stain. (m–p) Merge. (a, e, i and m) Control cells transfected with pEGFP alone. Control shows normal actin and nuclear morphology with no cMyc co-localization. (b, f, j and n) Cells transfected with pRK5myc-tcaA and pEGFP-actin. tcaA shows actin condensation (b and n) and nuclear fragmentation (j and n). RK5myc-tcaA tag shows localization to the cytoplasm (f). Note the intense yellow staining corresponding to co-localization of myc-tagged tcaA and pEGFP-actin (n). (o, g, k and o) Cells transfected with pRK5myc-tcaB and pGEFactin: condensation is observed (c and o) as well as apoptotic nuclei (o and k; arrows). The RK5myc-tcaB appears to co-localize with the cytoplasm, although this is difficult to determine in heavily condensed cells. (d, h, l and p) Cells transfected with pRK5myc-tcaC and pEGFP-actin. RK5myc-tcaC shows membrane ruffling (white arrows) and actin condensation (yellow arrow; d and p). Here the RK5myc-tcaC tag shows localization to the nucleus (h).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The recent increase in sequencing of bacterial genomes has led to the discovery of tc orthologues in a multitude of species with no known association with insects, including Pseudomonas and Burkholderia spp. Genome sequencing and microarray analysis has also revealed the presence of tc orthologues in all strains of the closely related mammalian pathogens Y. pestis and Y. pseudotuberculosis (Hinchliffe et al., 2003; Parkhill et al., 2001). Neither of these species is known to be toxic to insects and Y. pestis actually requires an insect vector for its transmission. Infected fleas do eventually die from colonization by Y. pestis, though this is thought to be due to starvation caused by the blockage of the proventriculus preventing blood meal from reaching the stomach rather than any direct toxic effect. Interestingly, microarray analysis revealed a level of divergence in the all of the tc genes between Y. pestis and Y. pseudotuberculosis (Hinchliffe et al., 2003; Parkhill et al., 2001). In X. nematophilus, differing insect host specificities have been traced to two paralogous copies of the [A] component (XptA1 and XptA2), suggesting that relatively minor differences in [A] contribute to toxin specificity. This led to the hypothesis that the Tc's of Y. pseudotuberculosis might be insecticidal toxins that had then diversified during the evolution of Y. pestis in order to prevent toxicity in the flea. Y. pseudotuberculosis is often found in soil, where it could have the opportunity to interact with insect larvae. In order to determine the extent of the divergence we PCR amplified and cloned the individual tc genes from several Y. pseudotuberculosis strains of diverse origin and serotype (Table 1) and compared them to published Y. pestis gene sequences. Protein sequence alignment (Supplementary Fig. S1a) confirmed the conservation of TcaA across a range of Y. pseudotuberculosis serogroups. Differences observed were mainly localized to the C-termini of these proteins, suggesting that they are under a diversification selection. TcaC was also highly conserved (Supplementary Fig. S1c), whilst TcaB was least conserved (Supplementary Fig. S1b), particularly within the region between amino acids 201 and 426. The diversity within these sequences suggests either diversity between strains or selective pressure for diversification such as the evasion of the mammalian immune system. As yet little is known about the structural and functional regions of Tc proteins in any species; thus we can only speculate as to the basis of this diversity.

We have previously shown that wild-type strains representing five different serogroups of Y. pseudotuberculosis were not orally toxic to M. sexta larvae (Waterfield et al., 2007), with only strain IP32953 causing any reduction in larval weight gain. These results suggested that either: (i) Y. pseudotuberculosis does not encode orally active insecticidal Tc toxins and some other factor is responsible for any toxicity observed, (ii) insect-active Tc toxins are encoded but not expressed, or (iii) the Tc proteins are insect active but not specifically toxic to M. sexta. In order to determine if the slight oral toxicity observed for strain IP32953 could be attributed to Tc expression, a tcaAB : : Km insertion knockout was constructed for this strain and also for strain YPIII pIB1. As shown in Fig. 2, no differences in oral toxicity were observed between the wild-type and the respective insertion knockout strains. In order to further rule out any toxicity of the Tc proteins to M. sexta we cloned the appropriate coding sequences and expressed them in the BL21 strain of E. coli. Cytosol preparations of expressing cells were fed to neonate larvae as described. These heterologously expressed Y. pseudotuberculosis Tc proteins did show limited oral toxicity against M. sexta neonates (Fig. 2); however, analysis by two-tailed t-test showed that any toxicity seen was not significant (P values are as follows: [A]=0.9064, [BC]=0.9064, [A]+[BC]=0.1346), unlike P. luminescens [A] (0.0017) [BC] (0.0087) and [A]+[BC] (<0.0001). In addition, we also demonstrated that heterologously expressed Y. pestis [A]+[BC] (0.3127) is not toxic to M. sexta (data not shown). These results were despite equal concentrations of proteins being fed to M. sexta as determined by Western blot.

Further evidence that the Yersinia Tc toxins are not insecticidal comes from their lack of toxicity towards cultured insect cells. When cytosolic preparations were topically applied to the S. fragiperala insect cell line Sf9 only heterologously expressed P. luminescens [A]+[BC] showed marked rearrangement of the actin cytoskeleton and nuclear apoptosis (Fig. 2). Both heterologously expressed Y. pseudotuberculosis and Y. pestis [A]+[BC] were comparable to the control. This apparent lack of insecticidal toxicity is further backed up by the fact that Y. pseudotuberculosis Tc gene knockouts show similar toxicity towards the flea as wild-type bacteria (Erickson et al., 2006, 2007; Hinnebusch et al., 2002).

Western blots carried out with Y. pseudotuberculosis wild-type strain IP32953 grown at temperatures varying from 15 °C to 37 °C demonstrate the expression and secretion of these toxins at 28 °C and 37 °C, whereas there is only a low level of expression at 20 °C or 15 °C (Fig. 1). Gels were loaded with similar amounts of total protein and therefore the cell supernatant from cultures grown at 15 °C had to be concentrated far more than other samples in order to achieve this, which may account for the presence of a band in the 15 °C supernatant. Indeed, previous RT-PCR results indicated that there was no expression at 15 °C and thus the Tc complex present in this culture may be an artefact from the initial starter culture, which was grown at 28 °C. Most importantly these Western blots demonstrate that the toxin complex of strain IP32953 is secreted during growth in normal media at all temperatures. This indicates a mechanism independent of the plasmid-borne type III secretion system, which has been previously reported to be responsible for secretion of this complex and is only expressed at 37 °C (Gendlina et al., 2007). In certain tropical climates there is the potential for bacterial interactions with insects to occur at environmental temperatures of 30 °C or more. Therefore there is the possibility that the Y. pseudotuberculosis and Y. pestis Tc proteins are specifically toxic to certain insect species which are as yet unidentified. This seems unlikely in the case of Y. pestis, which is not known to survive in the environment and only colonizes mammalian hosts and its flea vector. Thus a more likely scenario is that the Tc toxins have evolved to become toxic to the mammalian host.

In order to investigate this hypothesis, lysates from BL21 cells containing Y. pseudotuberculosis [A]+[BC] and Y. pestis [A]+[BC] were topically applied to the mammalian cell lines Caco-2 and NIH3T3. As shown in Fig. 4, Y. pseudotuberculosis [A]+[BC] caused dramatic rearrangement of the actin cytoskeleton of Caco-2 cells, causing membrane ruffles and vacuoles similar to the effects described for CNF1 (Boquet, 2001; Fiorentini et al., 1988). Interestingly, approximately 7 % of cells were also multinucleate, compared to 2 % of controls. Surprisingly, no toxic effects were seen with the Y. pseudotuberculosis Tc's when applied to NIH3T3 cells. Conversely lysates of BL21 cells containing Y. pestis [A]+[BC] showed limited cytopathic effects on Caco-2 cells but produced dramatic rearrangement of the actin cytoskeleton and nuclear fragmentation in NIH3T3 cells (Fig. 5). These effects on NIH3T3 cells were also seen with lysates of BL21 cells containing P. luminescens [A]+[BC] which is perhaps surprising because P. luminescens is not a mammalian pathogen and the Tc's are thought to act on insect midgut. This suggests that the Tc's of P. luminescens have a general toxic activity and therefore the evolution of mammalian-specific Tc proteins in Yersinia ssp. is not a huge evolutionary step. As previously discussed, host specificity has been attributed to the [A] component of the complex. Unlike the TcdA1, XptA1 and SepA [A] components, which are single long polypeptides (designated the tcd configuration), the [A] components of Y. pseudotuberculosis and Y. pestis are split into two polypeptides, encoded by the tcaA and tcaB genes (designated the tca configuration). In P. luminescens the TcaB protein contains an RDG motif at amino acid position 557–560 which is a QGD motif in Y. pseudotuberculosis, and in Y. pestis the equivalent homology position is 570–572. The presence of RDG motifs has led to speculation that a host integrin may be the toxin receptor on the cell surface. The differences between P. luminescens TcaB and Y. pseudotuberculosis TcaB could be attributed to this; however, Y. pestis TcaB also has QDG and exerts its effects on different cells to the Y. pseudotuberculosis Tc proteins. Thus it can be assumed that although this motif may contribute to binding, it does not determine cell specificity. Rather, the high level of divergence seen in the N-terminal region of TcaB (Supplementary Fig. S1b) could contribute to the specificity observed. Further analysis of the effects of the Tc's on different cell types and structure–function characterization of TcaB will be required to fully understand the cell specificity seen here.

Delineation of the phenotypic effects was determined by transient expression of individual Y. pseudotuberculosis and Y. pestis TcaA, TcaB and TcaC proteins in Caco-2 or NIH3T3 cells (Figs 6 and 7). This confirmed that the Tc's are capable of causing significant rearrangement of the actin cytoskeleton. TcaA localized to the cytoplasm and induced actin condensation with membrane ruffling (Y. pseudotuberculosis) and nuclear fragmentation (Y. pestis). This observation is in contrast to the P. luminescens tcaA gene, which, when expressed in NIH-3T3 cells, had no obvious adverse effect on the cells' F-actin distribution (Waterfield et al., 2005). These results suggest TcaA alone can be toxic, which has been previously reported for Y. enterocolitica strain W22703, where TcaA alone is responsible for oral toxicity to M. sexta (the tcaB gene contains a frameshift). Transient expression of TcaB also caused heavy actin condensation with the protein again accumulating in the cytoplasm (Figs 6 and 7). More interestingly, multinucleation was observed in Caco-2 cells expressing the Y. pseudotuberculosis TcaB protein (Fig. 6). Multinucleation has only previously been described with CNF1, which also causes actin rearrangement (Caprioli et al., 1983; Fiorentini et al., 1988). These phenotypic effects of CNF1 can be attributed to the permanent activation of Rho-GTPases (Fabbri et al., 2002; Fiorentini et al., 1997). Further work is now being carried out to determine if the Tc's have a similar mode of action. Transient expression of Y. pseudotuberculosis TcaC also produced a very unusual phenotype involving nuclear fragmentation and extensive vacuolation of the cytoplasm (Fig. 6d, h, l and p). In addition, the cMyc-tagged TcaC appeared to accumulate in the cell nucleus. Interestingly, the transient expression of the N terminus of the Photorhabdus tcdB (an orthologue of tcaC) in NIH3T3 fibroblasts also localized in the cell nucleus, suggesting that this family of proteins represent nuclear-active toxins. The N-terminal domains of TcaC [B] and TcdB [B] share a high percentage identity to the Salmonella plasmid-borne virulence factor, SpvB. In SpvB, this domain is responsible for export across the outer membrane of the Gram-negative cell, suggesting this domain may have general membrane-crossing abilities. The large cytoplasmic vacuoles observed (Fig. 6d and p) are similar to those produced by the Helicobacter pylori VacA toxin and the CNF1 toxin of E. coli. The vacuoles produced by CNF1 intoxication originate from excessive pinocytosis caused by the inappropriate activation of Rho, Rac and Cdc42 (Boquet, 2001; Fabbri et al., 2002; Falzano et al., 1993), whereas VacA vacuoles originate from late endosomal compartments (Papini et al., 1994). Further work will be required to determine the origin of the Y. pseudotuberculosis TcaC-derived cytoplasmic vacuoles. Expression of TcaC from Y. pestis did not cause vacuoles but did cause massive actin condensation (Fig. 7d, h, l and p). Interestingly, it too accumulated at the nuclear periplasm in what look like vesicles, supporting the theory that this domain is involved in export.

In conclusion, we have shown that Y. pseudotuberculosis (and Y. pestis) Tc proteins appear to have no apparent oral toxicity against the model insect M. sexta and are only transcribed at temperatures relevant to expression in the mammalian host, and that these proteins are secreted. In addition, we have demonstrated that the Tc proteins from Y. pseudotuberculosis, Y. pestis and P. luminescens have the ability to cause significant effects on cultured mammalian cells and our results suggest cell specificity. Finally, intracellular transient expression of different tc subunit genes in the cytoplasm of mammalian cells reproduces the phenotypes seen by topical application of the protein and suggests an unusual mode of action. Taken together, these observations suggest that the Y. pseudotuberculosis tc genes have been adapted for activity against the mammalian gut whereas Y. pestis has been adapted to other mammalian cell types, while those of P. luminescens are adapted for the insect gut, but do show toxicity to mammalian cells, indicating a common mechanism.

	ACKNOWLEDGEMENTS

 
We are grateful to Professor Stuart Reynolds and Sandra Barnes for help with the Manduca sexta bioassays, and allowing us to perform them at the University of Bath, and to Professor Richard Titball for useful discussion.

Edited by: T. J. Mitchell

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Achtman, M., Zurth, K., Morelli, G., 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]

Blackburn, M., Golubeva, E., Bowen, D. & ffrench-Constant, R. H. (1998). A novel insecticidal toxin from Photorhabdus luminescens: histopathological effects of toxin complex A (Tca) on the midgut of Manduca sexta. Appl Environ Microbiol 64, 3036–3041.[Abstract/Free Full Text]

Boquet, P. (2001). The cytotoxic necrotizing factor 1 (CNF1) from Escherichia coli. Toxicon 39, 1673–1680.[Medline]

Bowen, D. J. & Ensign, J. C. (1998). Purification and characterization of a high molecular weight insecticidal protein complex produced by the entomopathogenic bacterium Photorhabdus luminescens. Appl Environ Microbiol 64, 3029–3035.[Abstract/Free Full Text]

Bowen, D., Rocheleau, T. A., Blackburn, M., Andreev, O., Golubeva, E., Bhartia, R. & ffrench-Constant, R. H. (1998). Insecticidal toxins from the bacterium Photorhabdus luminescens. Science 280, 2129–2132.[Abstract/Free Full Text]

Bresolin, G., Morgan, J. A., Ilgen, D., Scherer, S. & Fuchs, T. M. (2006a). Low temperature-induced insecticidal activity of Yersinia enterocolitica. Mol Microbiol 59, 503–512.[CrossRef][Medline]

Bresolin, G., Neuhaus, K., Scherer, S. & Fuchs, T. M. (2006b). Transcriptional analysis of long-term adaptation of Yersinia enterocolitica to low-temperature growth. J Bacteriol 188, 2945–2958.[Abstract/Free Full Text]

Caprioli, A., Falbo, V., Roda, L. G., Ruggeri, F. M. & Zona, C. (1983). Partial purification and characterization of an Escherichia coli toxic factor that induces morphological cell alterations. Infect Immun 39, 1300–1306.[Abstract/Free Full Text]

Chain, P. S., Carniel, E., Larimer, F. W., Lamerdin, J., Stoutland, P. O., Regala, W. M., Georgescu, A. M., Vergez, L. M., Land, M. L. & 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]

Datsenko, K. A. & Wanner, B. L. (2000). One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 97, 6640–6645.[Abstract/Free Full Text]

Deng, W., Burland, V., Plunkett, G., III, Boutin, A., Mayhew, G. F., Liss, P., Perna, N. T., Rose, D. J., Mau, B. & other authors (2002). Genome sequence of Yersinia pestis KIM. J Bacteriol 184, 4601–4611.[Abstract/Free Full Text]

Enright, M. R. & Griffin, C. T. (2005). Effects of Paenibacillus on the entomopathogenic nematode Heterorhabditis megidis. J Invertebr Pathol 88, 40–48.[CrossRef][Medline]

Erickson, D. L., Jarrett, C. O., Wren, B. W. & Hinnebusch, B. J. (2006). Serotype differences and lack of biofilm formation characterize Yersinia pseudotuberculosis infection of the Xenopsylla cheopis flea vector of Yersinia pestis. J Bacteriol 188, 1113–1119.[Abstract/Free Full Text]

Erickson, D. L., Waterfield, N. R., Vadyvaloo, V., Long, D., Fischer, E. R., ffrench-Constant, R. & Hinnebusch, B. J. (2007). Acute oral toxicity of Yersinia pseudotuberculosis to fleas: implications for the evolution of vector-borne transmission of plague. Cell Microbiol 9, 2658–2666.[CrossRef][Medline]

Fabbri, A., Falzano, L., Travaglione, S., Stringaro, A., Malorni, W., Fais, S. & Fiorentini, C. (2002). Rho-activating Escherichia coli cytotoxic necrotizing factor 1: macropinocytosis of apoptotic bodies in human epithelial cells. Int J Med Microbiol 291, 551–554.[CrossRef][Medline]

Falzano, L., Fiorentini, C., Boquet, P. & Donelli, G. (1993). Interaction of Escherichia coli cytotoxic necrotizing factor type 1 (CNF1) with cultured cells. Cytotechnology 11 (Suppl. 1), S56–S58.[CrossRef][Medline]

ffrench-Constant, R. & Waterfield, N. (2006). An ABC guide to the bacterial toxin complexes. Adv Appl Microbiol 58, 169–183.[Medline]

Fiorentini, C., Arancia, G., Caprioli, A., Falbo, V., Ruggeri, F. M. & Donelli, G. (1988). Cytoskeletal changes induced in HEp-2 cells by the cytotoxic necrotizing factor of Escherichia coli. Toxicon 26, 1047–1056.[Medline]

Fiorentini, C., Fabbri, A., Matarrese, P., Falzano, L., Boquet, P. & Malorni, W. (1997). Hindrance of apoptosis and phagocytic behaviour induced by Escherichia coli cytotoxic necrotizing factor 1: two related activities in epithelial cells. Biochem Biophys Res Commun 241, 341–346.[CrossRef][Medline]

Fogh, J., Wright, W. C. & Loveless, J. D. (1977). Absence of HeLa cell contamination in 169 cell lines derived from human tumors. J Natl Cancer Inst 58, 209–214.[Medline]

Gendlina, I., Held, K. G., Schesser Bartra, S., Gallis, B. M., Doneanu, C. E., Goodlett, D. R., Plano, G. V. & Collins, C. M. (2007). Identification and type III dependent secretion of the Yersinia pestis insecticidal-like proteins. Mol Microbiol 64, 1214–1227.[CrossRef][Medline]

Guzman, L.-M., Belin, D., Carson, M. J. & Beckwith, J. (1995). Tight regulation, modulation and high level expression by vectors containing the arabinose pBAD promoter. J Bacteriol 177, 4121–4130.[Abstract/Free Full Text]

Hinchliffe, S. J., Isherwood, K. E., Stabler, R. A., Prentice, M. B., Rakin, A., Nichols, R. A., Oyston, P. C., Hinds, J., Titball, R. W. & Wren, B. W. (2003). Application of DNA microarrays to study the evolutionary genomics of Yersinia pestis and Yersinia pseudotuberculosis. Genome Res 13, 2018–2029.[Abstract/Free Full Text]

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]

Hurst, M. R., Glare, T. R., Jackson, T. A. & Ronson, C. W. (2000). Plasmid-located pathogenicity determinants of Serratia entomophila, the causal agent of amber disease of grass grub, show similarity to the insecticidal toxins of Photorhabdus luminescens. J Bacteriol 182, 5127–5138.[Abstract/Free Full Text]

Hurst, M. R., Jones, S. M., Tan, B. & Jackson, T. A. (2007). Induced expression of the Serratia entomophila Sep proteins shows activity towards the larvae of the New Zealand grass grub Costelytra zealandica. FEMS Microbiol Lett 275, 160–167.[CrossRef][Medline]

Liu, D., Burton, S., Glancy, T., Li, Z. S., Hampton, R., Meade, T. & Merlo, D. J. (2003). Insect resistance conferred by 283-kDa Photorhabdus luminescens protein TcdA in Arabidopsis thaliana. Nat Biotechnol 21, 1222–1228.[CrossRef][Medline]

Maxson, M. E. & Darwin, A. J. (2004). Identification of inducers of the Yersinia enterocolitica phage shock protein system and comparison to the regulation of the RpoE and Cpx extracytoplasmic stress responses. J Bacteriol 186, 4199–4208.[Abstract/Free Full Text]

Papini, E., de Bernard, M., Milia, E., Bugnoli, M., Zerial, M., Rappuoli, R. & Montecucco, C. (1994). Cellular vacuoles induced by Helicobacter pylori originate from late endosomal compartments. Proc Natl Acad Sci U S A 91, 9720–9724.[Abstract/Free Full Text]

Parkhill, J., Wren, B. W., Thomson, N. R., Titball, R. W., Holden, M. T., Prentice, M. B., Sebaihia, M., James, K. D., Churcher, C. & other authors (2001). Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413, 523–527.[CrossRef][Medline]

Riley, G. & Toma, S. (1989). Detection of pathogenic Yersinia enterocolitica by using congo red-magnesium oxalate agar medium. J Clin Microbiol 27, 213–214.[Abstract/Free Full Text]

Sergeant, M., Jarrett, P., Ousley, M. & Morgan, J. A. (2003). Interactions of insecticidal toxin gene products from Xenorhabdus nematophilus PMFI296. Appl Environ Microbiol 69, 3344–3349.[Abstract/Free Full Text]

Silva, C. P., Waterfield, N. R., Daborn, P. J., Dean, P., Chilver, T., Au, C. P., Sharma, S., Potter, U., Reynolds, S. E. & ffrench-Constant, R. H. (2002). Bacterial infection of a model insect: Photorhabdus luminescens and Manduca sexta. Cell Microbiol 4, 329–339.[CrossRef][Medline]

Song, Y., Tong, Z., Wang, J., Wang, L., Guo, Z., Han, Y., Zhang, J., Pei, D., Zhou, D. & other authors (2004). Complete genome sequence of Yersinia pestis strain 91001, an isolate avirulent to humans. DNA Res 11, 179–197.[Abstract]

Taylor, L. A. & Rose, R. E. (1988). A correction in the nucleotide sequence of the Tn903 kanamycin resistance determinant in pUC4K. Nucleic Acids Res 16, 358[Free Full Text]

Tennant, S. M., Skinner, N. A., Joe, A. & Robins-Browne, R. M. (2005). Homologues of insecticidal toxin complex genes in Yersinia enterocolitica biotype 1A and their contribution to virulence. Infect Immun 73, 6860–6867.[Abstract/Free Full Text]

Waterfield, N., Dowling, A., Sharma, S., Daborn, P. J., Potter, U. & ffrench-Constant, R. H. (2001a). Oral toxicity of Photorhabdus luminescens W14 toxin complexes in Escherichia coli. Appl Environ Microbiol 67, 5017–5024.[Abstract/Free Full Text]

Waterfield, N. R., Bowen, D. J., Fetherston, J. D., Perry, R. D. & ffrench-Constant, R. H. (2001b). The tc genes of Photorhabdus: a growing family. Trends Microbiol 9, 185–191.[CrossRef][Medline]

Waterfield, N. R., Wren, B. W. & Ffrench-Constant, R. H. (2004). Invertebrates as a source of emerging human pathogens. Nat Rev Microbiol 2, 833–841.[CrossRef][Medline]

Waterfield, N., Hares, M., Dowling, A. & ffrench-Constant, R. (2005). Potentiation and cellular phenotypes of the insecticidal toxin complexes of Photorhabdus bacteria. Cell Microbiol 7, 373–382.[CrossRef][Medline]

Waterfield, N., Hares, M., Hinchliffe, S., Wren, B. & ffrench-Constant, R. (2007). The insect toxin complex of Yersinia. Adv Exp Med Biol 603, 247–257.[CrossRef][Medline]

Wren, B. W. (2003). The yersiniae – a model genus to study the rapid evolution of bacterial pathogens. Nat Rev Microbiol 1, 55–64.[CrossRef][Medline]

Received 13 March 2008; revised 27 June 2008; accepted 22 July 2008.

This article has been cited by other articles:

				D. Zhou and R. Yang Molecular Darwinian Evolution of Virulence in Yersinia pestis Infect. Immun., June 1, 2009; 77(6): 2242 - 2250. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplementary figures 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 Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Hares, M. C. Articles by Waterfield, N. Search for Related Content PubMed PubMed Citation Articles by Hares, M. C. Articles by Waterfield, N. Agricola Articles by Hares, M. C. Articles by Waterfield, N.