The integral inner membrane component XcpS of the Pseudomonas aeruginosa type II secretion system: structure, function and localisation

Type II secretion systems (T2SS) of Gram-negative bacteria are multi-component apparatuses that secrete proteins into the extracellular milieu. These systems have homology to type IV pili, competence systems of Gram-positive bacteria and archaeal flagella. Although the systems are only involved in the translocation step over the outer membrane, most components are located in or attached to the inner membrane. Here we describe our studies into the main T2SS of Pseudomonas aeruginosa, the Xcp system. This system consists of 12 proteins, named XcpA and XcpP to XcpZ. Of these, only XcpQ is located in the outer membrane. It is thought to form a pore in this membrane through which other proteins are secreted. The components XcpT to XcpX, named pseudopilins, are believed to form a pilus-like structure, called pseudopilus, that pushes the secreted proteins through the pore. The pseudopilins are matured by the prepilin peptidase XcpA. The pseudopilus is believed to be assembled on top an inner membrane platform, consisting of the inner membrane proteins XcpS, XcpY and XcpZ and the cytoplasmic protein XcpR. Of these, XcpR presumably generates the energy required for the secretion process. XcpY and XcpZ mutually stabilise each other, and XcpY docks XcpR to the inner membrane. A last protein, XcpP, is believed to link XcpQ in the outer membrane to the inner membrane platform. We have studied the inner membrane protein XcpS. This is a highly conserved protein, of which the function is still unknown. XcpS has three membrane spanning domains. Its N terminus is located in the cytoplasm, whereas the C terminus is located in the periplasm. Investigations into its multimeric state showed that it preferentially trimerises. Furthermore, indications for higher oligomeric forms were obtained. Purified negatively stained XcpS was visualised with electron microscopy as ring-like structures, which seemed to contain a central cavity. XcpS was also demonstrated to interact with the major pseudopilin XcpT. When these two proteins were simultaneously introduced in an artificial lipid bilayer, evidence was found for pore-formation by XcpS. When XcpS and XcpT were co-produced in Escherichia coli, this induced a growth defect and a Psp response; both indications for stress. This further demonstrated the putative pore-forming capacities of XcpS upon interaction with XcpT. Additional experiments indicated that an XcpR-XcpY-XcpZ complex can block the XcpS pore, which explains how pore formation by XcpS does normally not lead to toxic effects in vivo. Other experiments showed that XcpS is N-terminally processed. This procedure was found essential for the functioning of the T2SS. A region in the cytoplasmic N-terminal domain important for this event was identified. This domain is also found in many homologues of XcpS, suggesting that processing of XcpS-like proteins occurs in all T2SSs. A last experimental chapter provides evidence for the polar localisation of the Xcp system and type II secretion. Altogether these results have been integrated into a new model for type II secretion, which basically explains the reason for pore-formation by multimeric XcpS. Thus, our results provide novel insights in the functioning of T2SSs