Delineation and mutational analysis of the Yersinia pseudotuberculosis YopE domains which mediate translocation across bacterial and eukaryotic cellular membranes.

Pathogenic yersiniae deliver a number of different effector molecules, which are referred to as Yops, into the cytosol of eukaryotic cells via a type III secretion system. To identify the regions of YopE from Yersinia pseudotuberculosis that are necessary for its translocation across the bacterial and eukaryotic cellular membranes, we constructed a series of hybrid genes which consisted of various amounts of yopE fused to the adenylate cyclase-encoding domain of the cyclolysin gene (cyaA) of Bordetella pertussis. By assaying intact cells for adenylate cyclase activity, we show that a YopE-Cya protein containing just the 11 amino-terminal residues of YopE is efficiently exported to the exterior surface of the bacterial cell. Single amino acid replacements of the first seven YopE residues significantly decreased the amount of reporter protein detected on the cell surface, suggesting that the extreme amino-terminal region of YopE is recognized by the secretion machinery. As has recently been shown for the Y. enterocolitica YopE protein (M.-P. Sory, A. Boland, I. Lambermont, and G. R. Cornelis, Proc. Natl. Acad. Sci. USA 92:11998-12002, 1995), we found that export to the cell surface was not sufficient for YopE-Cya proteins to be delivered into the eukaryotic cytoplasm. For traversing the HeLa cell membrane, at least 49 yopE-encoded residues were required. Replacement of leucine 43 of YopE with glycine severely affected the delivery of the reporter protein into HeLa cells. Surprisingly, export from the bacterial cell was also not sufficient for YopE-Cya proteins to be released from the bacterial cell surface into the culture supernatant. At least 75 residues of YopE were required to detect activity of the corresponding reporter protein in the culture supernatant, suggesting that a release domain exists in this region of YopE. We also show that the chaperone-like protein YerA required at least 75 YopE residues to form a stable complex in vitro with YopE-Cya proteins and, furthermore, that YerA is not required to target YopE-Cya proteins to the secretion complex. Taken together, our results suggest that traversing the bacterial and eukaryotic membranes occurs by separate processes that recognize distinct domains of YopE and that these processes are not dependent on YerA activity.