No abstract

Clostridium difficile is a major nosocomial pathogen and the main causative agent of antibiotic-associated diarrhoea. The organism produces two potent toxins, A and B, which are its major virulence factors. These are chromosomally encoded on a region termed the pathogenicity locus (PaLoc), which also contains regulatory genes, and is absent in non-toxigenic strains. Here we show that the PaLoc can be transferred from the toxin-producing strain, 630Δerm, to three non-toxigenic strains of different ribotypes. One of the transconjugants is shown by cytotoxicity assay to produce toxin B at a similar level to the donor strain, demonstrating that a toxigenic C. difficile strain is capable of converting a non-toxigenic strain to a toxin producer by horizontal gene transfer. This has implications for the treatment of C. difficile infections, as non-toxigenic strains are being tested as treatments in clinical trials.

The DNA sequence of the regulatory region and the structural gene, nrfA, for cytochrome c552 of Escherichia coli K-12 have been reported. We have now established that nrfA is the first gene in a seven-gene operon, designated the nrf operon, at least five of which are essential for formate-dependent nitrite reduction to ammonia. This operon terminates just upstream of the previously sequenced gltP gene encoding a sodium-independent, glutamate and aspartate transporter. Expression of lac fused to nrfA, nrfE or nrfG is regulated by oxygen repression, FNR-dependent anaerobic induction, nitrite induction and nitrate repression during anaerobic growth, exactly as previously reported for the nrfA promoter. In contrast, expression of the gltP-lac fusion was FNR-independent. The open reading frame immediately downstream of nrfA encodes NrfB, a hydrophilic, penta-haem cytochrome c with an M(r) of 20,714. The structure of the N-terminal region is typical of a signal peptide for a periplasmic protein: cleavage at the putative signal peptide cleavage site, Ala-26, would result in a periplasmic cytochrome with a molecular mass of 18 kDa. The NrfC polypeptide, M(r) 24,567, contains 16 cysteine residues arranged in four clusters typical of the CooF super-family of non-haem iron-sulphur proteins. The NrfD sequence predicts a 318-residue hydrophobic protein with a distribution of acidic and basic amino acids which suggests that NrfD is an integral transmembrane protein with loops in both the periplasm and the cytoplasm. Proteins most similar to NrfD include the PsrC subunit of polysulphide reductase from Wolinella, but, as seven of the 10 most similar proteins are NADH-ubiquinone oxidoreductases, we propose that NrfD participates in the transfer of electrons from the quinone pool into the terminal components of the Nrf pathway. NrfE, M(r) 60,851, is predicted to be another hydrophobic, integral membrane protein homologous to the CdI1 protein of Rhodobacter capsulatus, which has been implicated in the assembly of periplasmic c-type cytochromes. The sequence of the 127 residue NrfF polypeptide, M(r) 14,522, is strikingly similar to the CcI2 protein of R. capsulatus, especially in the putative haem-binding motif, RCPQCQNQN.(ABSTRACT TRUNCATED AT 400 WORDS)