Sophisticated genetic tools to modify essential biological processes at the molecular level are pivotal in elucidating the molecular pathogenesis of Clostridium difficile, a major cause of healthcare associated disease. Here we have developed an efficient procedure for making precise alterations to the C. difficile genome by pyrE-based allelic exchange. The robustness and reliability of the method was demonstrated through the creation of in-frame deletions in three genes (spo0A, cwp84, and mtlD) in the non-epidemic strain 630Δerm and two genes (spo0A and cwp84) in the epidemic PCR Ribotype 027 strain, R20291. The system is reliant on the initial creation of a pyrE deletion mutant, using Allele Coupled Exchange (ACE), that is auxotrophic for uracil and resistant to fluoroorotic acid (FOA). This enables the subsequent modification of target genes by allelic exchange using a heterologous pyrE allele from Clostridium sporogenes as a counter-/negative-selection marker in the presence of FOA. Following modification of the target gene, the strain created is rapidly returned to uracil prototrophy using ACE, allowing mutant phenotypes to be characterised in a PyrE proficient background. Crucially, wild-type copies of the inactivated gene may be introduced into the genome using ACE concomitant with correction of the pyrE allele. This allows complementation studies to be undertaken at an appropriate gene dosage, as opposed to the use of multicopy autonomous plasmids. The rapidity of the ‘correction’ method (5–7 days) makes pyrE − strains attractive hosts for mutagenesis studies.

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.

No abstract

Clostridium difficile is a major cause of antibiotic induced diarrhea worldwide, responsible for significant annual mortalities and represents a considerable economic burden on healthcare systems. The two main C. difficile virulence factors are toxins A and B. Isogenic toxin B mutants of 2 independently isolated erythromycin-sensitive derivatives (630E and 630Δerm) of strain 630 were previously shown to exhibit substantively different phenotypes. Compared to 630, strain 630E and its progeny grow slower, achieve lower final cell densities, exhibit a reduced capacity for spore-formation, produce lower levels of toxin and are less virulent in the hamster infection model. By the same measures, strain 630Δerm and its derivatives more closely mirror the behavior of 630. Genome sequencing revealed that 630Δerm had acquired 7 unique Single Nucleotide Polymorphisms (SNPs) compared to 630 and 630E, while 630E had 9 SNPs and a DNA inversion not found in the other 2 strains. The relatively large number of mutations meant that the identification of those responsible for the altered properties of 630E was not possible, despite the restoration of 3 mutations to wildtype by allelic exchange and comparative RNAseq analysis of all 3 strains. The latter analysis revealed large differences in gene expression between the 3 strains, explaining in part why no single SNP could restore the phenotypic differences. Our findings suggest that strain 630Δerm should be favored over 630E as a surrogate for 630 in genetic-based studies. They also underline the importance of effective strain curation and the need to genome re-sequence master seed banks wherever possible.

Forty-eight Staphylococcus aureus isolates collected from a young, healthy, Irish university student population from 1995 to 2004 were screened for 16 enterotoxin (SE) and enterotoxin-like (SEl) genes (sea–see, seg–sei, selj–selo, selq, selu), and for the toxic shock toxin syndrome toxin-1 gene, tst. All of the isolates harboured at least one SE or SEl gene and 66.7 % possessed a classical SE gene (sea, seb, sec), the commonest being the seb gene. Most of the isolates (85.4 %) had a complete egc locus (selo, selm, sei, seln, seg). The intergenic sei–seln region of the egc locus was typed by PCR-RFLP in 34 isolates, 15 possessing pseudogenes ψent1 and ψent2 and 19 having the selu gene. The seh and sell genes, the selk–selq gene combination, and the tst gene were each found in <15 % of isolates. The agr genotype distribution was agr type III, 37.5 %; agr type I, 35.4 %; agr type II, 25 %; and agr type IV, 2.1 %. There was no association between SE–SEl genotype and agr type. All tst gene-positive isolates were of agr type III and harboured a classical SE gene. Multiple locus, variable number tandem repeat analysis (MLVA) produced 47 different patterns. While the sdr locus was present in all isolates, half of them lacked one or two of the sdr gene amplimers. Twenty isolates harboured the bbp gene, its presence being associated with agr type III, but not with the SE–SEl gene profile. The agr types of isolates were associated with MLVA subclusters. Selective MLST analysis revealed seven novel sequence types and a new aroE allele. Five clonal clusters (CCs), including CCs comprising major pandemic clones CC30, CC5 and CC22 and minor lineages CC6 and CC9, and three singletons were identified.

The egc locus of Staphylococus aureus harbours two enterotoxin genes (seg and sei) and three enterotoxin-like genes (selm, seln and selo). Between the sei and seln genes are located two pseudogenes, ψent1 and ψent2, or the selu or selu v gene. While these two alternative sei–seln intergenic regions can be distinguished by PCR, to date, DNA sequencing has been the only confirmatory option because of the very high degree of sequence similarity between egc loci bearing the pseudogenes and the selu or selu v gene. In silico restriction enzyme digestion of genomic regions encompassing the egc locus from the 3′ end of the sei gene through the 5′ first quarter of the seln gene allowed pseudogene- and selu- or selu v-bearing egc loci to be distinguished by PCR-RFLP. Experimental application of these findings demonstrated that endonuclease HindIII cleaved PCR amplimers bearing pseudogenes but not those with a selu or selu v gene, while selu- or selu v-bearing amplimers were susceptible to cleavage by endonuclease HphI, but not by endonuclease HindIII. The restriction enzyme BccI cleaved selu- or selu v-harbouring amplimers at a unique restriction site created by their signature 15 bp insertion compared with pseudogene-bearing amplimers, thereby allowing distinction of these egc loci. PCR-RFLP analysis using these restriction enzymes provides a rapid, easy to interpret alternative to DNA sequencing for verification of PCR findings on the nature of an egc locus type, and can also be used for the primary identification of the intergenic sei–seln egc locus type.

Twenty genes encoding enterotoxin and enterotoxin-like proteins have been described in Staphylococcus aureus strains. Five of these occur commonly in the enterotoxin gene cluster (egc: selo, selm, sei, seln and seg). In the sei-seln intergenic region, two pseudogenes, yent1 and yent2, can be present or an additional gene designated selu or a variant selu v . Whilst frequencies of loci bearing pseudogenes (egc1) or the selu gene (egc2) have been reported, the distinction between selu-bearing and selu v -bearing (egc3) loci has rarely been made. A PCR-RFLP procedure involving cleavage of the sei-seln intergenic region by restriction endonuclease BbvI or TseI was developed that allowed differentiation of selu + and selu v + loci. In addition, PCR primers were designed to yield a 203 bp amplimer for sequencing of a selu or selu v intragenic region, which encompassed ten signature nucleotide differences. A total of 43 egc + human nasal isolates and 53 egc + bovine, ovine, caprine, leporine and gallinaceous isolates were egc typed and agr typed. None of the animal isolates was of agr type III. A total of 12 out of 17 egc3 + human nasal isolates were of agr type III, the other 5 being agr type I. On the basis of representative multilocus sequence typing, agr type III/egc3 + strains belonged to CC30. Human nasal isolates bearing an egc1 locus were distributed evenly across agr types I, II and III. Only two nasal isolates had an egc2 locus. All 14 agr type IV isolates, only 1 of which was of human origin, possessed an egc2 locus. The agr IV nasal isolate was fusidic acid sensitive and was found to be ST123 (CC121). There were strong associations between bovine, leporine and gallinaceous S. aureus clonal types and egc locus types. The PCR-RFLP procedure was used to screen an additional 45 S. aureus isolates from dogs, cats, rats, pigs and horses for egc locus types. Of these, 33 were egc " . Six equine isolates were selu + . One canine and three porcine isolates possessed pseudogenes yent1 and yent2. One porcine and one canine isolate each had the selu v gene. Putative relationships between disease-causing propensity and egc type need (re-)evaluation.

Comparative analysis of the Clostridium difficile BI/NAP1/027 strain R20291 and ClosTron-derived ermB mutants in the hamster infection model are compromised by the clindamycin susceptibility of the parent. Mutants can appear more virulent. We have rectified this anomaly by genome engineering. The variant created (CRG20291) represents an ideal control strain for virulence assays of ClosTron mutants.