*These authors contributed equally to this work.Clostridium difficile infection is the leading cause of healthcare associated diarrhoea in Europe and North America 1, 2 . During infection, C. difficile produces two key virulence determinants, toxin A and toxin B. Experiments with purified toxins have suggested that toxin A alone is able to evoke the symptoms of C. difficile infection, but toxin B is unable to do so unless it is mixed with toxin A, or there is prior damage to the gut mucosa 3 . However, a recent study suggested that toxin B is essential for C. difficile virulence and that a strain producing toxin A alone was avirulent 4 . This creates a paradox over the individual importance of toxin A and toxin B. Here we show that isogenic mutants of C. difficile producing either toxin A or toxin B alone can cause fulminant disease in the hamster model of infection. By using a gene knock-out system 5, 6 to permanently inactivate the toxin genes, we found that C. difficile producing either one or both toxins displayed cytotoxic activity in vitro, which translated directly into virulence in vivo.Furthermore, by constructing the first ever double mutant strain of C. difficile, in which both toxin genes were inactivated, we were able to completely attenuate virulence. Our findings re-establish the importance of both toxin A and toxin B and highlight the need to continue considering both toxins in the development of diagnostic tests and effective counter-measures against C. difficile.2 Toxin A and toxin B both catalyse the glucosylation, and hence inactivation, of Rho-GTPases; small regulatory proteins of the eukaryotic actin cell cytoskeleton. This leads to disorganisation of the cell cytoskeleton and cell death 7 . The toxin genes, tcdA and tcdB, are situated on the C. difficile chromosome in a 19.6 kilobase pathogenicity locus (PaLoc), along with the three accessory genes, tcdC, tcdR and tcdE (Fig. 1a). To address the individual importance of toxin A and toxin B, we used the ClosTron gene knock-out system 6 to inactivate the toxin genes of C. difficile. This system inactivates genes by inserting an intron into the protein-encoding DNA sequence of a gene, thus resulting in a truncated and non-functional protein. The intron sequence itself encompasses an erythromycin resistance determinant which permits selective isolation of mutants. Furthermore, it has been shown experimentally that the insertions are completely stable, meaning that inactivation of a gene is permanent 5 .Using the ClosTron system, we targeted insertions to tcdA and tcdB at nucleotide positions 1584 and 1511, respectively (Fig. 1a). In both cases, this placed the intron within DNA sequence encoding the toxin catalytic domain. Three separate isogenic The genotype of each toxin mutant was characterised by PCR and DNA sequence analysis to confirm the exact location of each intron insertion made (data not shown).Southern blot analysis of EcoRV-digested genomic DNA samples, using an intron-3 specific probe, confirmed that the A -B + and A + B -mutants ...

Successful infection by enteric bacterial pathogens depends on the ability of the bacteria to colonise the gut, replicate in host tissues and disseminate to other hosts. Pathogens such as Salmonella, Shigella and enteropathogenic and enterohaemorrhagic E. coli (EPEC and EHEC), utilise a type III secretion system (T3SS) to deliver virulence effector proteins into host cells during infection that promote colonisation and interfere with antimicrobial host responses 1-3. Here we report that the T3SS effector NleB1 from EPEC binds to host cell death domain containing proteins and thereby inhibits death receptor signalling. Protein interaction studies identified FADD, TRADD and RIPK1 as binding partners of NleB1. NleB1 expressed ectopically or injected by the bacterial T3SS prevented Fas ligand or TNF-induced formation of the canonical death inducing signalling complex (DISC) and proteolytic activation of caspase-8, an essential step in death receptor induced apoptosis. This inhibition depended on the N-GlcNAc transferase activity of NleB1, which specifically modified Arg117 in the death domain of FADD. The importance of the death receptor apoptotic pathway to host defence was demonstrated using mice deficient in the FAS signalling pathway, which showed delayed clearance of the EPEC-like mouse pathogen Citrobacter rodentium and reversion to virulence of an nleB mutant. The activity of NleB suggests that EPEC and other attaching and effacing (A/E) pathogens antagonise death receptor induced apoptosis of infected cells, thereby blocking a major antimicrobial host response.

Bacteria have mechanisms to export proteins for diverse purposes, including colonization of hosts and pathogenesis. A small number of archetypal bacterial secretion machines have been found in several groups of bacteria and mediate a fundamentally distinct secretion process. Perhaps erroneously, proteins called 'autotransporters' have long been thought to be one of these protein secretion systems. Mounting evidence suggests that autotransporters might be substrates to be secreted, not an autonomous transporter system. We have discovered a new translocation and assembly module (TAM) that promotes efficient secretion of autotransporters in proteobacteria. Functional analysis of the TAM in Citrobacter rodentium, Salmonella enterica and Escherichia coli showed that it consists of an Omp85-family protein, TamA, in the outer membrane and TamB in the inner membrane of diverse bacterial species. The discovery of the TAM provides a new target for the development of therapies to inhibit colonization by bacterial pathogens.

Many bacterial pathogens utilize a type III secretion system to deliver multiple effector proteins into host cells. Here we found that the type III effectors, NleE from enteropathogenic E. coli (EPEC) and OspZ from Shigella, blocked translocation of the p65 subunit of the transcription factor, NF-κB, to the host cell nucleus. NF-κB inhibition by NleE was associated with decreased IL-8 expression in EPEC-infected intestinal epithelial cells. Ectopically expressed NleE also blocked nuclear translocation of p65 and c-Rel, but not p50 or STAT1/2. NleE homologues from other attaching and effacing pathogens as well OspZ from Shigella flexneri 6 and Shigella boydii, also inhibited NF-κB activation and p65 nuclear import; however, a truncated form of OspZ from S. flexneri 2a that carries a 36 amino acid deletion at the C-terminus had no inhibitory activity. We determined that the C-termini of NleE and full length OspZ were functionally interchangeable and identified a six amino acid motif, IDSY(M/I)K, that was important for both NleE- and OspZ-mediated inhibition of NF-κB activity. We also established that NleB, encoded directly upstream from NleE, suppressed NF-κB activation. Whereas NleE inhibited both TNFα and IL-1β stimulated p65 nuclear translocation and IκB degradation, NleB inhibited the TNFα pathway only. Neither NleE nor NleB inhibited AP-1 activation, suggesting that the modulatory activity of the effectors was specific for NF-κB signaling. Overall our data show that EPEC and Shigella have evolved similar T3SS-dependent means to manipulate host inflammatory pathways by interfering with the activation of selected host transcriptional regulators.

Clostridium difficile causes a potentially fatal diarrheal disease through the production of its principal virulence factors, toxin A and toxin B. The tcdC gene is thought to encode a negative regulator of toxin production. Therefore, increased toxin production, and hence increased virulence, is often inferred in strains with an aberrant tcdC genotype. This report describes the first allele exchange system for precise genetic manipulation of C. difficile, using the codA gene of Escherichia coli as a heterologous counterselection marker. It was used to systematically restore the ⌬117 frameshift mutation and the 18-nucleotide deletion that occur naturally in the tcdC gene of C. difficile R20291 (PCR ribotype 027). In addition, the naturally intact tcdC gene of C. difficile 630 (PCR ribotype 012) was deleted and then subsequently restored with a silent nucleotide substitution, or "watermark," so the resulting strain was distinguishable from the wild type. Intriguingly, there was no association between the tcdC genotype and toxin production in either C. difficile R20291 or C. difficile 630. Therefore, an aberrant tcdC genotype does not provide a broadly applicable rationale for the perceived notion that PCR ribotype 027 strains are "high-level" toxin producers. This may well explain why several studies have reported that an aberrant tcdC gene does not predict increased toxin production or, indeed, increased virulence.

Clostridium difficile is a major cause of healthcare-associated infection and inflicts a considerable financial burden on healthcare systems worldwide. Disease symptoms range from self-limiting diarrhoea to fatal pseudomembranous colitis. Whilst C. difficile has two major virulence factors, toxin A and B, it is generally accepted that other virulence components of the bacterium contribute to disease. C. difficile colonises the gut of humans and animals and hence the processes of adherence and colonisation are essential for disease onset. Previously it has been suggested that flagella might be implicated in colonisation. Here we tested this hypothesis by comparing flagellated parental strains to strains in which flagella genes were inactivated using ClosTron technology. Our focus was on a UK-outbreak, PCR-ribotype 027 (B1/NAP1) strain, R20291. We compared the flagellated wild-type to a mutant with a paralyzed flagellum and also to mutants (fliC, fliD and flgE) that no longer produce flagella in vitro and in vivo. Our results with R20291 provide the first strong evidence that by disabling the motor of the flagellum, the structural components of the flagellum rather than active motility, is needed for adherence and colonisation of the intestinal epithelium during infection. Comparison to published data on 630Δerm and our own data on that strain revealed major differences between the strains: the R20291 flagellar mutants adhered less than the parental strain in vitro, whereas we saw the opposite in 630Δerm. We also showed that flagella and motility are not needed for successful colonisation in vivo using strain 630Δerm. Finally we demonstrated that in strain R20291, flagella do play a role in colonisation and adherence and that there are striking differences between C. difficile strains. The latter emphasises the overriding need to characterize more than just one strain before drawing general conclusions concerning specific mechanisms of pathogenesis.

Clostridium difficile infection is the main cause of healthcare-acquired diarrhea in the developed world. In addition to the main virulence factors toxin A and B, epidemic, PCR Ribotype 027 strains, such as R20291, produce a third toxin, CDT. To develop effective medical countermeasures, it is important to understand the importance of each toxin. Accordingly, we created all possible combinations of isogenic toxin mutants of R20291 and assessed their virulence. We demonstrated that either toxin A or toxin B alone can cause fulminant disease in the hamster infection model and present tantalizing data that C. difficile toxin may also contribute to virulence.

Attaching and effacing (A/E) pathogens are a significant cause of gastrointestinal illness in humans and animals. All A/E pathogens carry a large pathogenicity island, termed the locus for enterocyte effacement (LEE), which encodes a type III secretion system that translocates several effector proteins into host cells. To identify novel virulence determinants in A/E pathogens, we performed a signature-tagged mutagenesis screen in C57BL/6 mice by using the mouse A/E pathogen Citrobacter rodentium. Five hundred seventy-six derivatives of C. rodentium were tested in pools of 12 mutants. One attenuated mutant carried a transposon insertion in nleB, which encodes a putative effector of the LEE-encoded type III secretion system (T3SS). nleB is present in a genomic pathogenicity island that also encodes another putative effector, NleE, immediately downstream. Using translational fusions with ␤-lactamase (TEM-1), we showed that both NleB and NleE were translocated into host cells by the LEE-encoded T3SS of enteropathogenic Escherichia coli. In addition, deletion of the gene encoding NleB in C. rodentium resulted in reduced colonization of mice in single infections and reduced colonic hyperplasia. In contrast, the deletion of other non-LEE-encoded effector genes in C. rodentium, nleC, nleD, or nleE, had no effect on host colonization or disease. These results suggest that nleB encodes an important virulence determinant of A/E pathogens.