Enteroaggregative Escherichia coli plasmid-encoded toxin

1005

part of

Future Microbiol. (2010) 5(7), 1005–101310.2217/FMB.10.69 © 2010 Future Medicine Ltd ISSN 1746-0913

Futu

re M

icro

bio

log

y

Enteroaggregative Escherichia coli (EAEC) is a predominant cause of persistent diarrhea in the developing world where infection has been associated with malnourishment and growth retardation. In addition, EAEC is the predomi-nant E. coli pathotype that causes diarrhea in the developed world, and is second only to Campylobacter spp. as a cause of bacteria-medi-ated diarrhea [1]. The ability of this pathotype to mediate diarrhea was clearly established when a volunteer study demonstrated that EAEC strain 042 elicited diarrhea in the majority of volun-teers [2]. Thus, EAEC has been significantly associated with: endemic diarrhea in infants in developing and industrialized nations, par-ticularly persistent diarrhea; persistent diarrhea in HIV-positive patients; traveller’s diarrhea; food/water-borne outbreaks; and sporadic cases of diarrhea [3]. EAEC is defined by its charac-teristic aggregative adherence to HEp-2 cells in culture. The pathogenesis of EAEC diarrhea is not completely defined; however, two promi-nent pathogenic features have been described: formation of a thick mucus gel on the intestinal mucosa, and mucosal damage, apparently via the production of cytotoxin(s) [3].

Histopathologic alterations of the intestinal epi-thelium in patients and animal models infected with EAEC have been reported [2,4]. Similar his-tological alterations have been observed in autopsy samples of the ileum from children who died as a consequence of persistent diarrhea associated with EAEC infection [5]. It has been shown that a 104 kDa EAEC protein, termed plasmid-encoded toxin (Pet), is required for EAEC-induced dam-age to human intestinal mucosa [6–8]. Pet is a member of the autotransporter class of pro-teins secreted by Gram-negative bacteria. Once

secreted, Pet interacts with intestinal epithelial cells to cause enterotoxic and cytotoxic effects, leading to extrusion of intestinal epithelial cells. This article discusses Pet biological activities, from its mechanism of secretion to its mechanism of action on the epithelial cells.

Pet secretionPet is a member of the autotransporter class of secreted proteins and together with Tsh, EspP, EspC, Pic, SigA, Hbp, Sat and SepA proteins comprises the serine protease autotransporter of Enterobacteriaceae (SPATE) subfamily. The defining feature of autotransporters is their self-contained secretion system [7,9]. Thus, the DNA sequence of cloned pet (3885 bp in length) reveals a single gene coding for a 1295-amino acid protein with a predicted molecular mass of 140 kDa and a calculated isoelectric point of 6.71 [7]. This Pet pre-cursor protein contains three functional domains: the signal sequence, the passenger domain and the translocation unit (Figure 1A). The signal sequence is present at the N-terminal end of the protein and allows targeting of the protein to the inner mem-brane for its further export into the periplasm [10].

The passenger domain confers the diverse effector functions of Pet. The translocation unit (also called the b-domain), located at the C-terminal end of the protein, consists of a short linker region with an a-helical secondary structure and a b-core that adopts a b-barrel tertiary structure when embed-ded in the outer membrane [11,12], and facilitates translocation of the passenger domain through the outer membrane (Figure 1B) [9].

Several features of the autotransporter family are evident within the predicted pet gene product: ana lysis of the predicted Pet amino acid sequence reveals the presence of

Enteroaggregative Escherichia coli plasmid-encoded toxin

Fernando Navarro-GarciaDepartment of Cell Biology, Centro de Investigación y de Estudios Avanzados del IPN (CINVESTAV-IPN), Ap. Postal 14–740, 07000 México DF, México n Tel.: +52 555 747 3990 n Fax: +52 555 747 3393 n [email protected]

Plasmid-encoded toxin (Pet) is secreted by enteroaggregative Escherichia coli (EAEC), a pathotype of diarrhogenic E. coli. EAEC infection is an important cause of diarrhea in outbreak and nonoutbreak settings in developing and developed countries. EAEC secretes Pet by using the type V secretion system. Mature secreted Pet is a serine protease and its eukaryotic target is the actin-binding protein a-fodrin. When Pet cleaves a-fodrin in the target cell cytosol, the organization of the actin cytoskeleton is disrupted. The loss of actin filament structure results in cell rounding and detachment from the substratum. This article summarizes the long trip of Pet during its biogenesis, its interaction with epithelial cells, intracellular trafficking and mechanism of action.

Keywords

n autotransporter n bacterial toxin n cell trafficking n cytoskeleton n enteroaggregative E. coli n epithelial cells n serine protease n type V secretion system

Spe

cia

l Rep

ort

Future Microbiol. (2010) 5(7)1006 future science group

Special Report Navarro-Garcia

Pet

EAEC

β-barrelExtracellular spaceOM

SecIM

Periplasm

Bacterial cytoplasm

Signalsequence Passenger domain

Translocating unit

Serine proteaseactive site(GDSGSP)

Signalpeptidase Outer membrane protease

Role in virulence

Antiparallel β-sheet structure

Clathrin-mediatedendocytosis

Fusion with early endosomeEndoplasmic

reticulum

Golgi apparatus

Actin

FodrinSec61ptranslocon

Nucleus

Figure 1. Structure, secretion mechanism, intracellular trafficking and targeting of Pet. (A) Structure of Pet as an autotransporter protein. (B) Processing of Pet for its secretion through the IM and OM (type V secretion system). (C) Uptake and trafficking of Pet from the eukaryotic plasma membrane to the cytosol by retrograde intracellular transport, and its interaction with a target protein, fodrin. EAEC: Enteroaggregative Escherichia coli; IM: Inner membrane; OM: Outer membrane; Pet: Plasmid-encoded toxin.

www.futuremedicine.com 1007future science group

Enteroaggregative Escherichia coli plasmid-encoded toxin Special Report

a putative Walker A box nucleotide bind-ing motif (G281IIGNGK), which has been described for a number of other members of this class, though a function for these motifs has not yet been shown [13–16]; and a serine pro-tease motif (GDSGSP) has been found in sev-eral of the closest homologs of Pet – in Pet, the sequence was determined to be G256DSGSGV [7]. Although this site has been shown to be involved in proteolysis in Haemophilus and Neisseria autotransporters [17,18], in Pet this motif is not involved in autoproteolysis as in the autotransporters of members of the family Enterobacteriaceae [19]. In silico ana lysis of the deduced amino acid sequence of Pet and the N-terminal amino acid sequence of secreted Pet indicate that the protein possesses the characteristics of a signal sequence [7]. This signal is unusually long for E. coli but similar to those predicted for other autotransporters, with positively charged amino acids followed by a hydrophobic region, and a signal pepti-dase cleavage site between residues A52 and A53 (A53NMDISKAWARD) [7]. Despite several investigations, the function of the extended signal peptide region (ESPR) remains obscure and surprisingly it has not been proven that the ESPR is actually synthesized as part of the signal sequence. However, it has been shown that the DNA region encoding the ESPR of Pet is transcribed and translated, indicating that an unusually long 52 amino acid signal sequence is produced. Furthermore, the ESPR is present only in Gram-negative bacterial proteins origi-nating from the class b- and g-proteobacteria, and, more particularly, only in proteins secreted via the type V secretion pathway: autotrans-porters, TpsA exoproteins of the two-partner system and trimeric autotransporters [20]. Moreover, evidence shows that, for Pet, this ESPR in the signal sequence directs protein translocation across the inner membrane in a novel post-translational fashion. Mutational ana lysis suggests that the ESPR delays inner membrane translocation by adopting a par-ticular conformation, or by interacting with a cytoplasmic or inner membrane co-factor, prior to inner membrane translocation [21].

In silico ana lysis of the amino acid sequence of Pet and site-directed mutagenesis indicate that Pet undergoes a second post-translational pro-cessing step, namely, cleavage of the passenger domain from the b-domain. This latter region of the protein forms a b-barrel structure in the outer membrane through which the passenger domain of the protein passes and the cleavage

of the passenger domain occurs during this step. In Pet, this proteolytic process occurs between N1018 and N1019 [22]. Structural predictions of the b-domain of Pet, from the N1019 cleavage site to the terminal phenylalanine residue, consists of at least 14 membrane-spanning amphipathic b-strands interrupted by large external loops and generally short periplasmic loops, span-ning amino acid positions 1032 to 1295 of the Pet precursor [7]. Pet is secreted efficiently by EAEC and it is possible to detect Pet after 2 h, as determined by densitometry alongside an EAEC growth curve [8].

An important step during the secretion of proteins is the displacement of autotransporter proteins into the periplasmic space during their secretion, including whether accessory periplasmic proteins are involved in translo-cation to the external milieu. Recent studies have demonstrated the participation of peri-plasmic factors such as DegP, FkpA, Skp and SurA in the secretion of the autotransporter proteins IcsA and EspP [23,24]; EspP is mem-ber of the SPATE family like Pet. However, these periplasmic factors are not determin-ing factors in the secretion of IcsA and EspP by Shigella flexneri and EPEC, respectively. Recently, it was shown that a host-specific factor is necessary for eff icient folding of Pet. It was reported that two variants of Pet are secreted by a laboratory strain of E. coli, which was transformed with the cloned pet gene (pCEFN1). Biophysical ana lysis and cell-based toxicity assays demonstrated that only one of the two variants was in a folded, active conformation. The misfolded variant was not produced by a pathogenic strain of EAEC and did not result from protein overproduction in the laboratory strain of E. coli. Furthermore, EAEC transformed with pCEFN1 is able to overproduce Pet and all the Pet protein is in a folded conformation. These data suggest a host-specific factor is required for efficient fold-ing of Pet [25]. These accessory factors could be chaperones that help to maintain an unfolded and protected passenger domain since it has to pass through the b-barrel, which is thought to be around 2 nm in diameter. In addition, chaperones must be needed for inserting the b-barrel into the outer membrane. Thus, many accessory factors might be needed, questioning the autotransporter nickname for this family of proteins. The quest for discovering these chaperones will be interesting in relation to possible pharmacological targets for blocking the type V secretion system.



Pet delivery to epithelial cellsOnce secreted, soluble Pet must interact with the epithelial cells in order to reach its cellu-lar target. However, recently, it was found that other bacterial factors are needed for the effi-cient delivery of Pet on the plasma membrane. Interestingly, Pet toxin is efficiently delivered to epithelial cells by the bacteria, since it is inter-nalized into epithelial cells infected with EAEC at similar concentrations to those obtained by using 37 µg/ml purified Pet protein. These data suggest that other factor(s) could be helping in Pet delivery. Indeed, unlike the wild-type strain, Pet is not internalized when the epithelial cells were infected with HB101(pCEFN1), where pet is cloned in a laboratory strain of E. coli that does not have the same adhesion factors as the wild-type strain. In addition, this clone will be unable to secrete 37 µg/ml of Pet protein during 4 h of infection [8].

Pet binds to the epithelial plasma membrane at 4°C but only damages cells after the tem-perature is raised to 37°C for at least 2 h [19,26]. Endocytosis is blocked at 4°C, so the tempera-ture requirement for productive intoxication indicats that Pet does not act at the cell surface. Binding to the plasma membrane at 4°C further suggests that a possible specific surface receptor is recognized by Pet, although this receptor has yet to be identified. Indirect data support the notion that a receptor for Pet is needed on the plasma membrane of epithelial cells based on a study showing that Pet is internalized into the epithelial cells by clathrin-mediated endocytosis [26], and it is well known that this mechanism is triggered by receptor-mediated endocytosis; the ligand–receptor interaction is a requisite to form the clathrin-coated vesicles.

Pet uptake & trafficking into epithelial cells

Pet is internalized via clathrin-mediated endo-cytosis, demonstrated by studies showing that its internalization is inhibited by monodansyl-cadaverine and sucrose, but not by filipin or methyl-b-cyclodextrin, which are drugs that interfere with protein entry via a clathrin-inde-pendent pathway. In addition, Pet was immuno-precipitated by anticlathrin antibodies, but not by anticaveolin antibodies. Moreover, siRNA, designed to knock down clathrin gene expres-sion in HEp-2 cells, prevented Pet internaliza-tion, and thereby the Pet-induced cytotoxic effect. However, the use of siRNA to knock down caveolin expression had no effect on Pet internalization, and the cytotoxic effect was

clearly observed. Together, these data indicate that Pet secreted by EAEC binds to the cell sur-face via an unknown receptor, to be taken up by clathrin-mediated endocytosis, and exert its toxic effect in the cytoplasm (Figure 1C) [26].

Efficient Pet internalization into HEp-2 cells is affected by a tyrosine kinase inhibitor, genis-tein, suggesting that phosphorylation events in tyrosine are involved in Pet endocytosis induc-tion [27]. Perhaps genistein interferes with Pet endocytosis by altering the structural orga-nization of clathrin-coated pits, since inhibi-tion of tyrosine kinase activity blocks clathrin redistribution in the cellular periphery [28]. Another hypothesis is that the Pet receptor has tyrosine kinase activity, as is the case for epider-mal growth factor receptor, insulin receptor or albumin receptor, in which it has been deter-mined that binding of the ligand to its mem-brane receptor stimulates its phosphorylation into tyrosine, leading to active endocytosis by either clathrin- or caveolae-dependent mecha-nisms [29]. An interesting unpublished finding that supports Pet endocytosis by clathrin-coated pits is that Pet endocytosis is inhibited by alter-ing either the actin or microtubules cytoskel-eton, which demonstrates the participation of these cytoskeletal elements in the function of clathrin-coated pits and trafficking of proteins. Actin cytoskeleton is involved in vesicle forma-tion and clathrin-dependent endocytosis [30]. On the other hand, the role of microtubules in endocytosis has been extensively studied and it is clear that late steps of endocytosis depend on an integrated network of microtubules [31].

Many plant and bacterial toxins use the eukaryotic secretory pathway to enter the host cell cytoplasm [32,33]. These toxins have an AB structure that consists of a catalytic A moiety and a receptor-binding B moiety. Most of these AB toxins have intracellular targets and undergo intracellular trafficking. Pet is not an AB toxin, yet preliminary studies suggested that it could follow an AB toxin trafficking pathway from the cell surface to the endoplasmic reticulum (ER) and from the ER to the cytosol [22]. Such transport does not involve a KDEL (or RDEL) retrieval motif since Pet’s amino acid sequence does not contain a retrieval motif. To better char-acterize the intracellular trafficking and trans-location routes of Pet, Navarro-Garcia and col-leagues used confocal microscopy to document Pet transport from the early endosome to the Golgi apparatus and from the Golgi apparatus to the ER. Pet associates with the Sec61p trans-locon in the ER before it moves into the cytosol



as an intact, 104-kDa protein (Figure 1C). This translocation process contrasts with the export of other ER-translocating toxins, in which only the catalytic A subunit of the AB toxin enters the cytosol. Functional assays confirmed an ER exit site for Pet, since, as with intoxication with AB toxins, Pet intoxication is inhibited in a subset of mutant Chinese hamster ovary cell lines with aberrant activity in the ER-associated degrada-tion (ERAD) pathway of ER-to-cytosol translo-cation, but not by endosomal alkalization. This was the first report to demonstrate cell surface-to-ER trafficking and ER-to-cytosol translocation of a bacterial non-AB autotransporter toxin [27].

Epithelial cell damage & intracellular target of Pet

Originally, a 108-kDa protein secreted by EAEC strains was identified as a heat-labile entero-toxin [6]. This protein and another protein of 116 kDa were recognized by sera from patients in a Mexican outbreak of EAEC diarrhea. A number of findings suggest that the 108-kDa protein is an enterotoxin: fractions contain-ing both the 108-kDa protein and a distinct 116-kDa protein produce rises in short-circuit current (Isc), whereas a fraction from an EAEC strain producing only the 116-kDa protein does not; polyclonal antibodies raised against the 108-kDa protein abolish enterotoxic activity in a dose-dependent fashion, whereas anti-116-kDa protein antibodies have no effect; a 108-kDa protein-encoding subclone from the pAA plas-mid induces increases in Isc; a pAA-cured EAEC strain does not. The 108-kDa toxin induces not only enterotoxic effects but also tissue damage; these effects correlated with a fall in electric resistance values [6]. The cloning of 108-kDa protein revealed that the gene coding for a 140-kDa precursor, which is then processed to give a 104-kDa protein that was called Pet due to its plasmid localization in EAEC. Thus, it was confirmed that Pet increases Isc and decreases electrical resistance of rat jejunum mounted in the Ussing chamber, effects that are accompa-nied by mucosal damage, exfoliation of cells and development of crypt abscesses [7].

Since Pet causes a fall in tissue resistance and damage to the tissue when examined under light microscopy [6], it was hypothesized that Pet could be an enterotoxin that elicits cytopathic effects on intestinal epithelial cells. Indeed, Pet is able to intoxicate epithelial cells (HEp-2 and HT29 cells) after 2 h at 37°C as detected by light microscopy; these effects are characterized by time- and dose-dependent cell elongation

followed by rounding and ultimately release from the substratum. However, only 10 min of exposure to Pet, followed by incubation for another 2 h at 37°C, is sufficient to elicit the same morphologic changes. Thus, Pet appears to be a cytoskeleton-altering toxin, since it induces contraction of the cytoskeleton, loss of actin stress fibers and release of focal contacts in HEp-2 and HT29/C1 cell monolayers, followed by complete cell rounding and detachment [19]. Changes in the cytoskeleton of intestinal epi-thelial cells have been associated with dimin-ished resistance of intestinal epithelia [34,35]. Pet protein also induces proteolysis in zymogram gels, and preincubation with the serine protease inhibitor phenylmethylsulfonyl fluoride results in complete abrogation of casein proteolysis. Interestingly, Pet cytotoxicity and enterotoxicity depend on Pet serine protease activity, since both effects are inhibited by phenylmethylsulfonyl fluoride and are not induced by Pet S260I, which is mutated in the catalytic serine and thereby lacks in vitro protease activity [19]. As mentioned above, Pet enters the eukaryotic cell and traf-ficking through the vesicular system appears to be required for the induction of cytopathic effects. Moreover, the Pet serine protease motif is the main requisite for the cytopathic effects, because internalization assays have shown that Pet and mutant Pet S260I are found inside epi-thelial cells, but that only native Pet produces cytopathic effects [22]. All these data suggest an intracellular target for Pet.

Pet degrades erythroid spectrin [36]. Using purified erythrocyte membranes treated with Pet, degradation of a- and b-spectrin chains was observed; this effect was dose- and time-dependent, and a 120-kDa protein fraction was observed as a breakdown product. Spectrin degradation and production of the 120-kDa subproduct were confirmed using specific anti-bodies against the a- and b-spectrin chains. The spectrin degradation caused by Pet is related to the Pet serine protease motif, as spectrin deg-radation by Pet toxin was inhibited by anti-Pet antibodies and by PMSF. A site-directed Pet mutant, which had been shown to abolish the enterotoxic and cytotoxic effects of Pet was unable to degrade spectrin in erythrocyte mem-branes or purified spectrin [36]. Even though these findings were performed with erythroid spectrin, they were important since Pet deliv-ery within the intestine suggests that Pet may degrade epithelial fodrin (which is a homolog protein found in nonerythroid cells) as its intra-cellular target. Indeed, it was found to be an



intracellular target for Pet. Pet is able to cleave fodrin in vivo (Figure 1C); Pet-treated HEp-2 cells reveal intracellular redistribution of fodrin after 2 h of incubation. After 3 h, when Pet had pro-duced a cytoskeletal effects, almost all the fodrin was found in intracellular aggregates, which appeared to be located inside blebs. Fodrin deg-radation was confirmed in in vitro experiments, using a recombinant glutathione S-transferase (GST)-fusion protein, representing repeat units 8–14 of human fetal brain fodrin (109 kDa). Pet generated two breakdown subproducts of 37 and 72 kDa. This proteolytic activity was time-dependent and was observed during the first 10 min until its complete degradation after 2 h. Sequencing of the 37-kDa fragment, which did not have the GST in its N-terminus, dem-onstrated that the Pet cleavage site is localized within fodrin’s 11th repetitive unit, in helix C between M1198 and V1199, and inside the calmod-ulin-binding domain. The change of these amino acids by site-directed mutagenesis pre-vented fodrin degradation by Pet. A mutant in the serine protease motif of Pet was also unable to generate fodrin redistribution in epithelial cells and to cleave recombinant fodrin–GST. The cleavage site in recombinant fodrin occurs within repeat 11, suggesting that the breakdown subproducts in the whole natural molecule are of 138 and 146 kDa. These data were confirmed by detection of a subproduct of 153 kDa from epithelial a-fodrin of Pet-treated HEp-2 cells by using western blot [37]. This was the first report showing cleavage of a-fodrin by a bacte-rial protease. Cleavage occurs in the middle of the calmodulin-binding domain, which leads to cytoskeleton disruption. Furthermore, intracel-lular expression of mature Pet toxin (comprising only the passenger domain of the Pet precursor) in the cytoplasm of HEp-2 cells by using mam-malian expression vectors, was accompanied by condensation of the fodrin cytoskeleton. These studies corroborate an intracellular site of action for the Pet toxin, further implicating a role for fodrin in Pet intoxication [38].

Cytoskeleton contraction due to a-fodrin cleavage by Pet may explain previous obser-vations showing cell damage by EAEC. The enterotoxic effects produced by Pet [6] could be caused by the disruption of the membrane skeleton because a-fodrin was found to form a macromolecular complex with epithelial sodium channels [39], and epithelial channels mediate the entry of Na from the luminal fluid into cells during the first stage of elec-trogenic transepithelial Na transport across

Na-reabsorbing epithelia [40]; this explains the diarrheal pathogenesis caused by EAEC. Thus, cytoskeletal disruption and cell detachment by Pet are due to a-fodrin degradation and this explains the damages caused by EAEC strains in intestinal necropsy of Mexican children [5], in vitro organ culture model [41], in T84 cultured cells [42], or those detected when Pet is directly used in HEp-2 and HT29 cultured cells [19], in vitro organ culture [43] or intestinal preparation mounted in Ussing chambers [6].

So far, it is clear that purified Pet follows a sequence of events in order to damage the epi-thelial cells. However, all these events have been deduced from research performed in in vitro sys-tems using purified Pet (37 µg/ml) and have not yet been shown to occur during infection of epi-thelial cells by EAEC. Recently, Navarro-Garcia and colleagues reported that the secretion of Pet by EAEC during epithelial cell infections is reg-ulated at the transcriptional level, since secretion is inhibited in eukaryotic cell culture medium, although Pet is efficiently secreted in the same medium supplemented with tryptone. Inefficient secretion of Pet by EAEC in Dulbecco’s modi-fied Eagle medium (DMEM) prevents cell detachment, whereas efficient Pet secretion in DMEM/tryptone increases cell detachment in a HEp-2 cell adherence assay. Interestingly, Pet toxin is efficiently delivered to epithelial cells, since it is internalized into epithelial cells infected with EAEC at similar concentrations to those obtained by using 37 µg/ml purified Pet protein. In addition, Pet is not internalized when the epithelial cells are infected with a pet clone in a laboratory strain, HB101(pCEFN1), unlike the wild-type strain, which has a high adherence capability. Interestingly, there is a correlation between Pet secretion by EAEC, the internal-ization of Pet into epithelial cells, cell detach-ment and cell death in EAEC-infected cells. The ratio between live and dead cells decreases in cells treated with wild-type EAEC in compari-son with cells treated with an isogenic mutant in the pet gene, whereas the effects are restored by complementing the mutant with the pet gene. All these data indicate that Pet is an important virulence factor in the pathogenesis of EAEC infection [8].

Future perspectiveThe cellular microbiology approach to under-stand the role of bacterial effectors and tox-ins on host components allow us to integrate several host-cell targets that can be translated into accurate understanding of the mechanism



by which these virulence factors cause disease. Many toxins and effector proteins have been detected and many more will be detected in this era of genetics, genomics and proteomics efforts, thereby the integrated knowledge from cell biologists and microbiologist will be fun-damental. In this context, Pet is an interesting virulence factor, since it is a toxin secreted by an E. coli pathotype whose mechanism of patho-genesis is still unknown, but it is recognized as a cause of persistent diarrhea in the develop-ing world where infection has been associated with malnutrition and growth retardation. Furthermore, EAEC is the predominant cause of diarrhea induced by the different E. coli pathotypes in the developed world, and is sec-ond only to Campylobacter spp. as a cause of bacteria-mediated diarrhea [1].

In addition, Pet is secreted by a recently described mechanism of secretion, the type V secretion system, also known as the autotrans-porter family. The autotransporter secretion mechanism is the most common mechanism for the secretion of virulence factors across the outer membrane from pathogenic Gram-negative bacteria [9]. These virulence factors are associated with a wide variety of diseases caused by pathogenic Gram-negative bacteria, and they play a variety of roles in pathogenesis. In addition, autotransporters have attracted biotechnological and biomedical interest for protein display on bacterial cell surfaces. Despite their importance, the details of many events in autotransporter biogenesis (e.g., chap-erone function in the periplasm; mechanism of b-barrel insertion into the outer membrane; translocation pathway across the outer mem-brane; proteolytic release of the mature pro-tein from the outer membrane) remain unre-solved. In addition, despite the relatedness of autotransporter family members, the activity of autotransporters is diverse and each one of them needs to be characterized as an individual virulence factor in the context of each auto-transporter-producing pathogen.

Another unique feature of Pet is its intracel-lular trafficking mechanism. Pet travels from the cell surface to the ER before exploiting ERAD to enter the cytosol where it damages the actin cytoskeleton. Many AB toxins also exploit ERAD to enter the target cell cytosol. For these toxins, the catalytic A subunit dissociates from the cell-binding B subunit in the ER. The disso-ciated A subunit then unfolds and consequently activates ERAD, an endogenous quality-control system that removes misfolded proteins from the ER by exporting them to the cytosol through Sec61p and/or Derlin-1 protein-conducting channels. Pet, the only non-AB toxin known to use the ERAD translocation mechanism, does not exhibit the arginine-over-lysine amino acid bias seen in the A subunits of ER-translocating toxins [5,27]. Furthermore, Pet activation of the ERAD system does not appear to involve sub-stantial unfolding of the toxin. Thus, under-standing Pet intracellular trafficking allows us to explore new mechanisms of toxin trafficking as well as to understand the cell biology of gen-eral mechanisms of protein trafficking through research on intracellular factors and host chap-erones involved in these processes. After gain-ing access to the epithelial cytosol, Pet targets the actin-binding protein fodrin (also known as spectrin in erythroid cells) [37]. However, the exact mechanism of cytoskeleton damage to cause epithelial cell death is unclear.

Executive summaryn Plasmid-encoded toxin (Pet) is an autotransporter protein secreted by the type V secretion system and member of the serine protease

autotransporter family of Enterobacteriaceae.n Pet is endocytosed by clathrin-coated vesicles after its binding to eukaryotic plasma membrane. n Pet follows vesicular trafficking, including endosomes, the Golgi apparatus and the endoplasmic reticulum (ER). n From the ER, Pet is translocated to the cytosol using the Sec61 translocon and the ER-associated degradation system.n In the cytosol, Pet interacts with fodrin, producing a proteolytic cleavage in this molecule, leading to cytoskeleton disruption.n Fodrin degradation correlates with epithelial cell damage caused by Pet, including actin cytoskeleton retraction, loss of the stress fibers,

cell rounding with membrane blebs, and cell detachment from the substrate.

AcknowledgementThe author would like to thank Paul Ugalde for the artistic work on the preparation of Figure 1.

Financial & competing interests disclosureThe data from the author’s laboratory were generated with the support of Consejo Nacional de Ciencia y Tecnología (Conacyt, Mexico). The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.



BibliographyPapers of special note have been highlighted as:n of interest

1. Wilson W, Scott RB, Pinto A, Robertson MA: Intractable diarrhea in a newborn infant: microvillous inclusion disease. Can. J. Gastroenterol. 15(1), 61–64 (2001).

2. Nataro JP, Deng Y, Cookson S et al.: Heterogeneity of enteroaggregative Escherichia coli virulence demonstrated in volunteers. J. Infect. Dis. 171(2), 465–468 (1995).

3. Harrington SM, Dudley EG, Nataro JP: Pathogenesis of enteroaggregative Escherichia coli infection. FEMS Microbiol. Lett. 254(1), 12–18 (2006).

4. Tzipori S, Montanaro J, Robins-Browne RM, Vial P, Gibson R, Levine MM: Studies with enteroaggregative Escherichia coli in the gnotobiotic piglet gastroenteritis model. Infect. Immun. 60(12), 5302–5306 (1992).

5. Eslava C, Villaseca J, Morales R, Navarro A, Cravioto A: Identification of a protein with toxigenic activity produced by enteroaggregative Escherichia coli. Presented at: 93rd General Meeting of the American Society for Microbiology. Washington, DC, 23–27 May 1993.

6. Navarro-Garcia F, Eslava C, Villaseca JM et al.: In vitro effects of a high-molecular-weight heat-labile enterotoxin from enteroaggregative Escherichia coli. Infect. Immun. 66(7), 3149–3154 (1998).

n First paper showing the identification of plasmid-encoded toxin (Pet) as a toxin secreted by enteroaggregative Escherichia coli (EAEC).

7. Eslava C, Navarro-Garcia F, Czeczulin JR, Henderson IR, Cravioto A, Nataro JP: Pet, an autotransporter enterotoxin from enteroaggregative Escherichia coli. Infect. Immun. 66(7), 3155–3163 (1998).

n Cloning and characterization of Pet as an autotransporter toxin.

8. Betancourt-Sanchez M, Navarro-Garcia F: Pet secretion, internalization and induction of cell death during infection of epithelial cells by enteroaggregative Escherichia coli. Microbiology 155(Pt 9), 2895–2906 (2009).

n Characterization of the role of Pet during the infection of culture cells with a Pet-producing EAEC, as opposed to using a purified Pet protein.

9. Henderson IR, Navarro-Garcia F, Desvaux M, Fernandez RC, Ala’aldeen D: Type V protein secretion pathway: the autotransporter story. Microbiol. Mol. Biol. Rev. 68(4), 692–744 (2004).

n The most complete review on autotransporter proteins.

10. Henderson IR, Navarro-Garcia F, Nataro JP: The great escape: structure and function of the autotransporter proteins. Trends Microbiol. 6(9), 370–378 (1998).

11. Maurer J, Jose J, Meyer TF: Characterization of the essential transport function of the Aida-I autotransporter and evidence supporting structural predictions. J. Bacteriol. 181(22), 7014–7020 (1999).

12. Oliver DC, Huang G, Fernandez RC: Identification of secretion determinants of the Bordetella pertussis Brka autotransporter. J. Bacteriol. 185(2), 489–495 (2003).

13. Brunder W, Schmidt H, Karch H: EspP, a novel extracellular serine protease of enterohaemorrhagic Escherichia coli o157:H7 cleaves human coagulation factor V. Mol. Microbiol. 24(4), 767–778 (1997).

14. Provence Dl, Curtiss R 3rd: Isolation and characterization of a gene involved in hemagglutination by an avian pathogenic Escherichia coli strain. Infect. Immun. 62(4), 1369–1380 (1994).

15. Schmitt W, Haas R: Genetic ana lysis of the Helicobacter pylori vacuolating cytotoxin: Structural similarities with the IgA protease type of exported protein. Mol. Microbiol. 12(2), 307–319 (1994).

16. St Geme JW 3rd, Cutter D, Barenkamp SJ: Characterization of the genetic locus encoding Haemophilus influenzae type b surface fibrils. J. Bacteriol. 178(21), 6281–6287 (1996).

17. Bachovchin WW, Plaut AG, Flentke GR, Lynch M, Kettner Ca: Inhibition of IgA1 proteinases from Neisseria gonorrhoeae and Hemophilus influenzae by peptide prolyl boronic acids. J. Biol. Chem. 265(7), 3738–3743 (1990).

18. Hendrixson DR, De La Morena Ml, Stathopoulos C, St Geme JW 3rd: Structural determinants of processing and secretion of the Haemophilus influenzae Hap protein. Mol. Microbiol. 26(3), 505–518 (1997).

19. Navarro-Garcia F, Sears C, Eslava C, Cravioto A, Nataro JP: Cytoskeletal effects induced by Pet, the serine protease enterotoxin of enteroaggregative Escherichia coli. Infect. Immun. 67(5), 2184–2192 (1999).

n Evidence of Pet as a cytoskeleton-altering toxin and the role of Pet serine protease motif.

20. Desvaux M, Cooper LM, Filenko NA et al.: The unusual extended signal peptide region of the type V secretion system is phylogenetically restricted. FEMS Microbiol. Lett. 264(1), 22–30 (2006).

21. Desvaux M, Scott-Tucker A, Turner SM et al.: A conserved extended signal peptide region directs posttranslational protein translocation via a novel mechanism. Microbiology 153(Pt 1), 59–70 (2007).

22. Navarro-Garcia F, Canizalez-Roman A, Luna J, Sears C, Nataro JP: Plasmid-encoded toxin of enteroaggregative Escherichia coli is internalized by epithelial cells. Infect. Immun. 69(2), 1053–1060 (2001).

23. Purdy GE, Fisher CR, Payne SM: IcsA surface presentation in Shigella flexneri requires the periplasmic chaperones DegP, Skp, and SurA. J. Bacteriol. 189(15), 5566–5573 (2007).

24. Ruiz-Perez F, Henderson IR, Leyton DL, Rossiter AE, Zhang Y, Nataro JP: Roles of periplasmic chaperone proteins in the biogenesis of serine protease autotransporters of Enterobacteriaceae. J. Bacteriol. 191(21), 6571–6583 (2009).

25. Nemec KN, Scaglione P, Navarro-Garcia F, Huerta J, Tatulian SA, Teter K: A host-specific factor is necessary for efficient folding of the autotransporter plasmid-encoded toxin. Biochimie 92(2), 171–717 (2010).

26. Navarro-Garcia F, Canizalez-Roman A, Vidal JE, Salazar MI: Intoxication of epithelial cells by plasmid-encoded toxin requires clathrin-mediated endocytosis. Microbiology 153(Pt 9), 2828–2838 (2007).

27. Navarro-Garcia F, Canizalez-Roman A, Burlingame KE, Teter K, Vidal JE: Pet, a non-AB toxin, is transported and translocated into epithelial cells by a retrograde trafficking pathway. Infect. Immun. 75(5), 2101–2109 (2007).

n First report documenting the cell surface-to-endoplasmic reticulum and endoplasmic reticulum-to-cytosol trafficking of a non-AB bacterial toxin.

28. Wilde A, Beattie EC, Lem L et al.: EGF receptor signaling stimulates src kinase phosphorylation of clathrin, influencing clathrin redistribution and egf uptake. Cell 96(5), 677–687 (1999).

29. Tiruppathi C, Song W, Bergenfeldt M, Sass P, Malik AB: Gp60 activation mediates albumin transcytosis in endothelial cells by tyrosine kinase-dependent pathway. J. Biol. Chem. 272(41), 25968–25975 (1997).

30. Lamaze C, Schmid SL: Recruitment of epidermal growth factor receptors into coated pits requires their activated tyrosine kinase. J. Cell Biol. 129(1), 47–54 (1995).

31. Matteoni R, Kreis TE: Translocation and clustering of endosomes and lysosomes depends on microtubules. J. Cell Biol. 105(3), 1253–1265 (1987).



32. Lord JM, Roberts LM: Toxin entry: retrograde transport through the secretory pathway. J. Cell Biol. 140(4), 733–736 (1998).

33. Sandvig K, Van Deurs B: Delivery into cells: Lessons learned from plant and bacterial toxins. Gene Ther 12(11), 865–872 (2005).

34. Hecht G, Koutsouris A, Pothoulakis C, Lamont JT, Madara JL: Clostridium difficile toxin B disrupts the barrier function of T84 monolayers. Gastroenterology 102(2), 416–423 (1992).

35. Hecht G, Pothoulakis C, Lamont JT, Madara JL: Clostridium difficile toxin A perturbs cytoskeletal structure and tight junction permeability of cultured human intestinal epithelial monolayers. J. Clin. Invest. 82(5), 1516–1524 (1988).

36. Villaseca JM, Navarro-Garcia F, Mendoza-Hernandez G, Nataro JP, Cravioto A, Eslava C: Pet toxin from enteroaggregative Escherichia coli produces cellular damage associated with fodrin disruption. Infect. Immun. 68(10), 5920–5927 (2000).

37. Canizalez-Roman A, Navarro-Garcia F: Fodrin cam-binding domain cleavage by Pet from enteroaggregative Escherichia coli leads to actin cytoskeletal disruption. Mol. Microbiol. 48(4), 947–958 (2003).

n Identification of an epithelial target protein for Pet and characterization of the action mechanism.

38. Sui BQ, Dutta PR, Nataro JP: Intracellular expression of the plasmid-encoded toxin from enteroaggregative Escherichia coli. Infect. Immun. 71(9), 5364–5370 (2003).

39. Zuckerman JB, Chen X, Jacobs JD, Hu B, Kleyman TR, Smith PR: Association of the epithelial sodium channel with Apx and a-spectrin in A6 renal epithelial cells. J. Biol. Chem. 274(33), 23286–23295 (1999).

40. Garty H, Palmer LG: Epithelial sodium channels: function, structure, and regulation. Physiol. Rev. 77(2), 359–396 (1997).

41. Hicks S, Candy DC, Phillips AD: Adhesion of enteroaggregative Escherichia coli to pediatric intestinal mucosa in vitro. Infect. Immun. 64(11), 4751–4760 (1996).

42. Nataro JP, Hicks S, Phillips AD, Vial PA, Sears CL: T84 cells in culture as a model for enteroaggregative Escherichia coli pathogenesis. Infect. Immun. 64(11), 4761–4768 (1996).

43. Henderson IR, Hicks S, Navarro-Garcia F, Elias WP, Philips AD, Nataro JP: Involvement of the enteroaggregative Escherichia coli plasmid-encoded toxin in causing human intestinal damage. Infect. Immun. 67(10), 5338–5344 (1999).

n Demonstrates the role of Pet in intestinal damage using in vitro organ culture of infant human intestine.

Documents

Enteroaggregative Escherichia coli plasmid-encoded toxin