Cytotoxic necrotizing factor 1 (CNF1), a Rho GTPase-activating bacterial toxin, has been shown to contribute to invasion by meningitis-causing Escherichia coli K1 of human brain microvascular endothelial cells (HBMEC), which constitute the blood-brain barrier. However, CNF1 is a cytosolic protein and it remains unclear how its secretion occurs, contributing to E. coli invasion of HBMEC. To investigate the genetic requirement for CNF1 secretion in E. coli K1 strain RS218, we performed mini-Tn5 in vitro mutagenesis and constructed a transposon mutant library of strain NBC, in which b-lactamase was fused to the C-terminus of CNF1 in the chromosome of strain RS218. We identified a transposon mutant (NBC-1E6) that exhibited reduced b-lactamase activity in its culture supernatant and had the transposon inserted into the ygfZ gene. When ygfZ was deleted from the genome of strain RS218 (DygfZ), the translocation of CNF1 into HBMEC was impaired. Subcellular localization analysis of CNF1 demonstrated that YgfZ, a periplasmic protein, contributes to secretion of CNF1 into outer-membrane vesicles (OMVs). The DygfZ mutant was significantly defective in invasion of HBMEC compared to the parent E. coli K1 strain. The defects of the DygfZ mutant in CNF1 secretion into OMVs and translocation into HBMEC as well as invasion of HBMEC were abrogated by complementation with ygfZ. Taken together, our findings demonstrate that YgfZ contributes to CNF1 secretion into OMVs in meningitis-causing E. coli K1.
INTRODUCTIONCNF1, the paradigm of the Rho GTPase-activating bacterial toxins (Boquet, 2001;Knust & Schmidt, 2010;Lemonnier et al., 2007), is frequently associated with uropathogenic and meningitis-causing Escherichia coli strains (Khan et al., 2002;Landraud, et al., 2000). CNF1 activates Rho GTPases by deamidation of glutamine 63, converting it into glutamic acid, thus inhibiting GTPhydrolysing activity, resulting in constitutive activation of Rho GTPase. CNF1 has been shown to contribute to polymerization of F-actin, increased formation of stress fibres and phagocytosis (Caprioli et al., 1983;Falzano et al., 1993;Flatau et al., 1997;Schmidt et al., 1997; VouretCraviari et al., 1999). CNF1-like toxins have also been found in other bacteria, including the dermonecrotic toxin of Bordetella spp. and the CNFc from Yersinia pseudotuberculosis (Knust & Schmidt, 2010;Kume et al., 1986;Lockman et al., 2002).We have shown that CNF1 contributes to invasion by E. coli K1 of human brain microvascular endothelial cells (HBMEC) and penetration into the brain via the interaction with its receptor, 37 laminin receptor precursor (37LRP)/67 laminin receptor (67LR) (Chung et al., 2003;Khan et al., 2002;Kim et al., 2005). CNF1 is a cytoplasmic protein and its secretion is, therefore, a strategy utilized by meningitis-causing E. coli K1 to invade the blood-brain barrier (Khan et al., 2002;Kim, 2003Kim, , 2008. However, it remains incompletely understood how CNF1 secretion occurs. No typical signal peptide is found in the CNF1 sequence. Recent studies have ...