Identification of New In Vivo TonB-FepA Rendezvous Sites

Abstract

The TonB system of Gram-negative bacteria uses the protonmotive force of the cytoplasmic membrane to energize active transport of large or scarce nutrients across the outer membrane by means of customized beta-barrels known as TonB-dependent transporters (TBDTs). The lumen of each TBDT is occluded by an amino-terminal domain, called the cork, which must be displaced for transport of nutrients or translocation of the large protein toxins that parasitize the system. A complex of cytoplasmic membrane proteins consisting of TonB, ExbB and ExbD harnesses the protonmotive force that TonB transmits to the TBDT. The specifics of this energy transformation are a source of continuing interest. The amino terminal domain of a TBDT contains a region called the TonB box, that is essential for the reception of energy from TonB. This domain is the only identified site of in vivo interaction between the TBDT and TonB, occurring through a non-essential region centered on TonB residue Q160. Because TonB binds to TBDTs whether or not it is active or even intact, the mechanism and extent of cork movement in vivo has been challenging to discover. In this study, we used in vivo disulfide crosslinking between eight engineered Cys residues in Escherichia coli TonB and 42 Cys substitutions in the TBDT FepA, including the TonB box, to identify novel sites of interaction in vivo. The TonB Cys substitutions in the core of an essential carboxy terminal amphipathic helix (residues 199-216) were compared to TonB Q160C interactions. Functionality of the in vivo interactions was established when the presence of the inactive TonB H20A mutation inhibited them. A previously unknown functional interaction between the hydrophilic face of the amphipathic helix and the FepA TonB box was identified. Interaction of Q160C with the FepA TonB box appeared to be less functionally important. The two different parts of TonB also differed in their interactions with the FepA cork and barrel turns. While the TonB amphipathic helix Cys residues interacted only with Cys residues on the periplasmic face of the FepA cork, TonB Q160C interacted with buried Cys substitutions within the FepA cork, the first such interactions seen with any TBDT. Both sets of interactions required active TonB. Taken together, these data suggest a model where the amphipathic helix binds to the TonB box, causing the mechanically weak domain of the FepA cork to dip sufficiently into the periplasmic space for interaction with the TonB Q160 region, which is an interaction that does not occur if the TonB box is deleted. The TonB amphipathic helix also interacted with periplasmic turns between FepA β-strands in vivo supporting a surveillance mechanism where TonB searched for TBDTs on the periplasmic face of the outer membrane.

Studies on theEscherichia coliExbD Transmembrane Domain, Residue L132, and an Inhibitory Cyclic Peptide

“…While in vivo studies of the TonB system have admittedly not always been interpreted correctly [see (39) subsequently corrected in (94)], it is also worth noting that structural studies have failed to correctly represent in vivo conformations of TonB carboxy terminal homodimers (41) and important interactions between the TonB carboxy terminus and TBDTs FhuA and BtuB proteins (40). Structural studies have also missed interactions between the TonB carboxy terminal amphipathic helix and TBDT FepA (18). Additionally, Cys substitutions that formed ExbD disulfide-linked homodimers cannot be modeled on the ExbD NMR structure without significant distortion (36).…”

“…Energized ExbD correctly configures TonB for productive interaction with FepA. Stage IV is binding of a monomeric periplasmic domain of TonB to FepA, such that the siderophore enterochelin is actively transported across the outer membrane (18). After a transport event, H20, a key structural residue in the transmembrane domain, is somehow required for re-establishment of TonB periplasmic domain homodimers in Stage I (12, 26, 41).…”

“…After a transport event, H20, a key structural residue in the transmembrane domain, is somehow required for re-establishment of TonB periplasmic domain homodimers in Stage I (12, 26, 41). The green circles represent that all stages of the TonB energy transduction cycle involve the participation of the essential carboxy terminal amphipathic helix (residues 199-216) in TonB (18). In this model, the cellular ratios of 14 ExbB: 4ExbD: 2TonB leads to an excess of 2 ExbDs that are not associated with TonB (11).…”

“…Structurally, the TBDT b-barrels consist of 22 strands, with a lumen that is occluded by an amino-terminal globular domain, called the cork or hatch. In vivo, the periplasmically exposed surface of the TBDT appears to signal that ligand is bound via a conformational change that allows for productive interaction with TonB (13)(14)(15)(16)(17)(18). The cork domain moves in vivo during contact with TonB (18).…”

“…In vivo, the periplasmically exposed surface of the TBDT appears to signal that ligand is bound via a conformational change that allows for productive interaction with TonB (13)(14)(15)(16)(17)(18). The cork domain moves in vivo during contact with TonB (18).…”