Sequence and phylogenetic analysis of the gene for surface layer protein, slpA, from 14 PCR ribotypes of Clostridium difficile.

Clostridium difficile is the commonest cause of antibiotic-associated diarrhoea, with the hospitalized elderly being at particular risk. The organism makes a crystalline surface protein layer (S-layer), encoded by the slpA gene, the product of which is cleaved to give two mature peptides which associate to form the layer. The larger peptide (high molecular weight; HMW), derived from the C-terminal portion of the precursor, is relatively conserved, whereas the smaller peptide (low molecular weight; LMW), derived from the N-terminal portion of the precursor, is a dominant antigen which substantially forms the basis for serotyping of isolates. PCR ribotyping is a more discriminatory typing method, based on the intergenic rRNA. We obtained the sequence for slpA and some flanking DNA from a collection of C. difficile strains of 14 ribotypes isolated from elderly patients. Sequences from different ribotypes were compared with one another and with published sequences. Sequences from C. difficile ribotypes 046 and 092 were identical. Sequences from ribotype pairs 005 and 054, 012 and 046/092, 014 and 066 and 031 and 094 differed by 1-3 nt in the slpA gene. There were ultimately nine ribotypes or groups of ribotypes with very different slpA sequences, particularly in the region encoding the LMW peptide. The sequence from ribotype 002 was very different from previously published sequences. The DNA segment sequenced included the 5' 315 bp of a secA homologue, encoding a putative transport protein required for peptide secretion across the plasma membrane. The amino acid sequences of the predicted HMW peptides were aligned and a neighbour-joining tree was produced using 10,000 bootstrap replicates. The predicted SecA N-terminal region was similarly analysed. For both SlpA and SecA, a strong association was found between ribotypes 012, 046/092, 017, 031 and 094. Ribotypes 001 and 078 formed part of this clade for SlpA but not SecA, indicating independent evolution for slpA and secA, presumably because they come under different selection pressures.