ATP disrupts lipid binding equilibrium to drive retrograde transport critical for bacterial outer membrane asymmetry

Abstract

The hallmark of the Gram-negative bacterial envelope is the presence of the outer membrane (OM). The OM is asymmetric, comprising lipopolysaccharides (LPS) in the outer leaflet and phospholipids (PLs) in the inner leaflet; this critical feature confers permeability barrier function against external insults, including antibiotics. To maintain OM lipid asymmetry, the OmpC-Mla system is believed to remove aberrantly localized PLs from the OM and transport them to the inner membrane (IM). Key to the system in driving lipid trafficking is the MlaFEDB ABC transporter complex in the IM, but mechanistic details, including transport directionality, remain enigmatic. Here, we develop a sensitive point-to-point in vitro lipid transfer assay that allows direct tracking of [14C]-labelled PLs between the periplasmic chaperone MlaC and MlaFEDB reconstituted into nanodiscs. We reveal that MlaC spontaneously transfers PLs to the IM transporter in an MlaD-dependent manner that can be further enhanced by coupled ATP hydrolysis. In addition, we show that MlaD is important for modulating productive coupling between ATP hydrolysis and such retrograde PL transfer. We further demonstrate that spontaneous PL transfer also occurs from MlaFEDB to MlaC, but such anterograde movement is instead abolished by ATP hydrolysis. Our work uncovers a model where PLs reversibly partition between two lipid binding sites in MlaC and MlaFEDB, and ATP binding and/or hydrolysis shift this equilibrium to ultimately drive retrograde PL transport by the OmpC-Mla system. These mechanistic insights will inform future efforts towards discovering new antibiotics against Gram-negative pathogens.

Of zones, bridges and chaperones – phospholipid transport in bacterial outer membrane assembly and homeostasis

“…2d) [34,[45][46][47][48]. This expanded PL-binding cavity likely facilitates spontaneous PL transfer from MlaC to MlaFEDB, an activity that has recently been observed in vitro, and can further be enhanced with ATP binding/hydrolysis [35]. As part of the initial controversy on transport directionality, spontaneous PL transfer from MlaFEDB to MlaC is also known to occur in vitro [33,49], yet we now know that ATP binding/ hydrolysis abolishes such movement [35].…”

“…A brief controversy for the transport directionality of the OmpC-Mla system arose from observations that spontaneous lipid 'anterograde' transfer can occur from the MlaFEDB complex to MlaC in vitro [33]. However, this suggestion was quickly debunked as ATP-dependent retrograde lipid transfer was clearly demonstrated in more recent reconstitution studies [34,35]. Consistently, removing the Mla system in A. baumannii did not significantly disrupt anterograde PL transport as judged by OM vesicle formation [36].…”

“…It is now understood that PL exchange between MlaC and MlaFEDB at the IM is spontaneous yet reversible, and ATP binding/hydrolysis drives the ultimate transfer of PLs into the IM, thus powering overall retrograde transport from the OM (Fig. 2a) [35].…”

Molecular mechanism of phospholipid transport at the bacterial outer membrane interface

“…Many recent biochemical and structural studies have provided detailed insights into ATP-dependent PL transfer steps at the IM (19, 20, 23-30). In particular, we now know that when PL-bound MlaC arrives at the IM, it can spontaneously transfer the lipid molecule to the binding cavity of MlaFEDB (21). Even though this initial process is reversible (20, 21, 25), ATP binding/hydrolysis by the MlaFEDB complex ultimately catalyzes the transfer of PLs into the IM, driving overall retrograde transport from the OM (20, 21).…”

“…In particular, we now know that when PL-bound MlaC arrives at the IM, it can spontaneously transfer the lipid molecule to the binding cavity of MlaFEDB (21). Even though this initial process is reversible (20, 21, 25), ATP binding/hydrolysis by the MlaFEDB complex ultimately catalyzes the transfer of PLs into the IM, driving overall retrograde transport from the OM (20, 21). How ATP binding/hydrolysis may be coupled to PL transport in the MlaFEDB complex can be partly inferred from recently solved structures (20, 26-30), though detailed mechanistic understanding is still limited.…”