Exploring the Dynamics of Proton-Driven Multidrug Transporter
By: Carol A. Rouzer, VICB Communications
Published: December 16, 2013
Studies of the molecular motions of the multidrug transporter LmrP reveal how proton translocation drives substrate transport.
Antibiotic resistant bacteria (Figure 1) are a rapidly increasing threat to global public health. One mechanism of resistance for many bacteria is the expression of multidrug transporters. These versatile membrane proteins are notable for their ability to recognize a wide range of molecules with diverse structures and transport them out of the cell. Understanding how these proteins work is an important step in our ability to combat drug-resistant bacteria. Thus, Vanderbilt Institute of Chemical Biology member Hassane Mchaourab, his collaborator Cédric Govaerts (Université Libre de Bruxelles, Belgium), and their laboratories launched an investigation of the transport mechanism of the L. lactis LmrP mutlidrug transporter [M. Masureel, et al. (2013) Nat. Chem. Biol., published online Dec. 8, DOI: 10.1038/nchembio.1408, copyright (2013)].
Figure 1. . Low-temperature electron micrograph of a cluster of E.coli bacteria. E. coli is one of many bacteria exhibiting multidrug resistance mediated, in part, through multidrug transporters. Photo by Eric Erbe and Christopher Pooley, U.S. Department of Agriculture, obtained from Wikimedia Commons, public domain.
The LmrP transporter is a member of the major facilitator superfamily (MFS) of transporters that couples translocation of ions down a concentration gradient with transport of sugars, amino acids, and nucleosides, along with a range of organic and inorganic ions, including many drugs. MFS proteins are characterized by a common structure comprising an N-terminal and a C-terminal bundle of six α-helices each. The two bundles enclose a central binding site through which the substrate is transported, and the transport process is believed to occur via movement of the bundles relative to each other. Existing crystal structures of MFS proteins suggest at least three conformations, one in which the binding site is open toward the interior of the cell (inward-open), one in which it is open to the exterior of the cell (outward-open), and an occluded conformation, in which both ends of the binding site are closed. Some prior models of transporter function have proposed that the two bundles of helices are rigid and that they rock back and forth in order to open and close the binding site at the opposing ends. However, recent work with the MFS lactose permease (LacY) has suggested a more detailed model describing a flexible protein that switches from outward-open to inward-open conformations. This work inspired the Mchaourab and Govaerts labs to explore the inner workings of LmrP.
The only currently available crystal structure for an MFS multidrug transporter is for the E. coli EmrD protein. The structure reveals the characteristic dual 6-helix bundles surrounding a central hydrophobic binding site in an occluded conformation. The investigators used this structure as a template to derive a molecular model for LmrP. They also used crystal structures of LacY and FucP (an MFS fructose transporter) to derive models for LmrP in an inward-open and outward-open conformation, respectively (Figure 2). The models helped the researchers to identify amino acid residues at key positions in each 6-helix bundle at both the extracellular and intracellular faces of the protein. They expressed a number of LmrP proteins in which a pair of these residues had been mutated to cysteines, and then they reacted the mutant proteins with a sulfur-sensitive spin-label probe (Figure 3). After verifying the functionality of each protein by its ability to transport the substrate Hoechst 33342 in lipid vesicles, the investigators used double electron-electron resonance (DEER) spectroscopy to measure the distance between the pairs of labeled cysteines. With these tools, they could monitor the movement of the two 6-helix bundles relative to each other at either the extracellular or the intracellular face of the protein under various conditions.
Figure 2. Ribbon diagrams of the models of LmrP based on LacY (a), EmrD (b), and FucP (c). Image reprinted by permission from Macmillan Publishers Ltd: [M. Masureel, et al. (2013) Nat. Chem. Biol., published online Dec. 8, DOI: 10.1038/nchembio.1408, copyright (2013)].
Figure 3. Diagrammatic representation of the location of spin labels used for DEER measurements. The 12 helices of the LmrP protein are shown as rods, with the N-terminal and C-terminal helices in green and tan, respectively. The blue spheres show the locations of amino acids that were mutated (in pairs) to cysteine for spin labeling, and the lines show the distances measured by DEER between selected pairs. (a) View from the side with the extracellular face on top. (b) View looking down on the extracellular face. (c) View looking down on the intracellular face. Image reprinted by permission from Macmillan Publishers Ltd: [M. Masureel, et al. (2013) Nat. Chem. Biol., published online Dec. 8, DOI: 10.1038/nchembio.1408, copyright (2013)].
L. lactis bacteria face an extracellular environment that is more acidic (pH ≤ 6.5) than their cytoplasm (pH ≅7). LmrP uses this pH gradient to power transport of its substrate molecules from the inside to the outside of the cell. The transporter’s reliance on protein transport suggested that its conformation would likely be sensitive to pH, so the investigators first tested this hypothesis. Using molecules spin-labeled at the exterior face, the DEER results indicated that, at pH 8, nearly all of the LmrP molecules were in the same conformation. Switching to pH 5 resulted in a shift of some of the molecules to a conformation in which the two 6-helix bundles were closer together, suggesting that high pH favored the outward-open conformation, while a more acidic environment led to closing of the binding site at the extracellular face (outward-closed). In general, the opposite was true for proteins labeled at the intracellular surface of the protein, where low pH favored an inward-open conformation, with closing of the intracellular end of the binding site (inward-closed) in the more basic environment. The data also showed that the two ends of the protein could move independently of each other, refuting the idea of a rigid structure.
Prior data had identified five acidic residues, Asp68, Asp128, Asp142, Asp235, and Glu327, that play an important role in proton and substrate transport in LmrP. The investigators mutated each of these residues to either asparagine or glutamine to mimic a permanently protonated state. They discovered that the D68N mutation stabilized the outward-closed, inward-open conformation of LmrP, while E327Q stabilized the outward-open conformation and had little effect at the intracellular surface. These results confirmed that LmrP’s conformational state is sensitive to the protonation of individual amino acids.
Binding of Hoechst33342 to LmrP caused a narrowing of distance distributions for all pairs of residues involving transmembrane helix 8. The rigidity of this normally flexible helix increased substantially upon Hoechst33342 binding, suggesting that it is intimately involved in the protein-substrate interaction. An equally notable effect of substrate binding was the stabilization of LmrP in the outward-open conformation.
With their accumulated data in hand, the Mchaourab and Govaerts groups proposed a working model for proton-driven substrate transport by LmrP (Figure 4). They noted that it is important that the transporter not translocate protons in the absence of substrate, as this would simply dissipate the proton gradient without coupling it to a useful function. Thus, their data indicating that acidic pH favors the outward-closed conformation makes sense. However, at the more alkaline intracellular pH, the protein is also expected to be inward-closed. Consequently, the model proposes that resting LmrP adopts an occluded conformation and that substrate enters the binding site from within the membrane through an opening between the helices. Once bound, the substrate’s position is stabilized by Glu327, which is located along the binding surface within the membrane. Substrate binding then shifts the protein to an outward-open conformation, allowing the entry of protons and the protonation of Glu327. These changes facilitate the release of the substrate into the extracellular space. The protons that have entered the binding site are translocated down the chain of acidic residues until they reach Asp68, which is located at the intracellular surface. Protonation of this residue leads to a switch to the inward-open conformation, while release of the substrate favors return to the outward-closed conformation. Finally, deprotonation of Asp68 allows a return to the inward-closed conformation, and the protein is ready for entry of a new substrate molecule.
Figure 4. Proposed model of substrate transport by LmrP. (i) The resting protein is in the occluded conformation. Substrate (green hexagon) enters from the membrane at the intracellular end of the molecule. (ii) The substrate moves toward the extracellular surface and (iii) induces opening at the extracellular face, allowing protons to enter and (iv) bind Glu327. The substrate is released into the extracellular space. (v) Protons are transferred to Asp68, leading to (vi) an opening at the intracellular space with proton release. The deprotonated protein returns to the resting state (i). Image reprinted by permission from Macmillan Publishers Ltd: [M. Masureel, et al. (2013) Nat. Chem. Biol., published online Dec. 8, DOI: 10.1038/nchembio.1408, copyright (2013)].
This simple model explains how LmrP can transport a wide range of molecular structures in concert with proton translocation. It provides a solid framework upon which to build a better understanding of multidrug transporters and may lead to new approaches to block their function in order to combat the growing problem of bacterial multidrug resistance.