Macromolecular assembly and secretion across the bacterial cell envelope: type II protein secretion systems

A decade ago, Pugsley and colleagues reported the existence of a large region of Klebsiella DNA, distinct from the Klebsiella gene encoding pullulanase, which was necessary for secretion of this enzyme to the cell surface in Escherichia coli (d'Enfert et al., 1987a,b). The pul genes it contained proved to be the tip of an iceberg. The sequences reported before 1992 (d'Enfert et al., 1987a,b; d'Enfert & Pugsley, 1989; Pugsley & Reyss, 1990; Reyss & Pugsley, 1990) included only one gene (pulD) that matched any sequence in the data base; a 220 amino acid residue segment of PulD was 32% identical with a portion of the filamentous phage-encoded protein, pIV. But by the time the sequence of the 18.8 kb DNA fragment that contained the pul genes had been completed (Possot et al., 1992), reports of sets of homologous genes in several species of Gram-negative plant and animal pathogens had appeared. For the most part, these gene clusters were cloned by their ability to complement mutants that produced, but failed to secrete, proteins normally found in the extracellular milieu; when tested, the mutants showed reduced pathogenicity or were totally avirulent. The secreted proteins included hydrolytic enzymes such as cellulase and pectinase from plant pathogens, and proteases and toxins from animal pathogens. The multi-gene family necessary for secretion of these enzymes is now known as the type II system or the main terminal branch (MTB) of the general secretion pathway (GSP). As summarized by Pugsley et al. (1997), the current tally includes type II systems from Klebsiella oxytoca (pul), Erwinia chrysanthemi and carotovora (out), Xanthomonas campestris (xps), Pseudomonas aeruginosa (xcp), Aeromonas hydrophila (exe), and Vibrio cholerae (eps). A second type II system (sps) necessary for deposition of the S-layer on the cell surface in A. hydrophila is more similar to the X. campestris than A. hydrophila genes (Thomas & Trust, 1995). The biggest surprise has been the discovery of a complete set of type II secretion genes in E. coli K12. The E. coli genes are not expressed under normal growth conditions, and a search is underway to find inducing conditions and secretion substrates (Francetic & Pugsley, 1996). Impressive progress has already been made in defining components of the pathway. What remains to be understood in mechanistic detail is how this protein secretion system functions.