Premium
Secretion signal of the filamentous haemagglutinin, a model two‐partner secretion substrate
Author(s) -
Hodak Hélène,
Clantin Bernard,
Willery Eve,
Villeret Vincent,
Locht Camille,
JacobDubuisson Françoise
Publication year - 2006
Publication title -
molecular microbiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.857
H-Index - 247
eISSN - 1365-2958
pISSN - 0950-382X
DOI - 10.1111/j.1365-2958.2006.05242.x
Subject(s) - periplasmic space , secretion , biology , bacterial outer membrane , signal peptide , virulence , bordetella pertussis , microbiology and biotechnology , bacterial adhesin , secretory protein , transport protein , transporter , biochemistry , peptide sequence , bacteria , genetics , escherichia coli , gene
Summary The sorting of proteins to their proper subcellular compartment requires specific addressing signals that mediate interactions with ad hoc transport machineries. In Gram‐negative bacteria, the widespread two‐partner secretion (TPS) pathway is dedicated to the secretion of large, mostly virulence‐related proteins. The secreted TpsA proteins carry a characteristic 250‐residue‐long N‐terminal ‘TPS domain’ essential for secretion, while their TpsB transporters are pore‐forming proteins that specifically recognize their respective TpsA partners and mediate their translocation across the outer membrane. However, the nature of the secretion signal has not been elucidated yet. The whooping cough agent Bordetella pertussis secretes its major adhesin filamentous haemagglutinin (FHA) via the TpsB transporter FhaC. In this work, we show specific interactions between an N‐terminal fragment of FHA containing the TPS domain and FhaC by using two different techniques, an overlay assay and a pull‐down of the complex. FhaC recognizes only non‐native conformations of the TPS domain, corroborating the model that in vivo , periplasmic FHA is not yet folded. By generating single amino acid substitutions, we have identified interaction determinants forming the secretion signal. They are found unexpectedly far into the TPS domain and include both conserved and variable residues, which most likely explains the specificity of the TpsA–TpsB interaction. The N‐terminal domain of FhaC is involved in the FHA–FhaC interaction, in agreement with its proposed function and periplasmic localization.