- Email: [email protected]
The multidrug resistance efflux complex, EmrAB from Escherichia coli forms a dimer in vitro
Biochemical and Biophysical Research Communications 380 (2009) 338–342
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
The multidrug resistance efflux complex, EmrAB from Escherichia coli forms a dimer in vitro Mikio Tanabe a,b,1, Gerda Szakonyi c, Katherine A. Brown b, Peter J.F. Henderson c, Jon Nield d, Bernadette Byrne a,* a
Division of Molecular Biosciences, Imperial College London, Exhibition Road, London SW7 2AZ, UK Division of Cell and Molecular Biology, Imperial College London, Exhibition Road, London SW7 2AZ, UK c Institute of Membrane and Systems Biology, University of Leeds, Leeds LS2 9JT, UK d School of Biological and Chemical Sciences, Queen Mary University of London, Mile End, London E1 4NS, UK b
a r t i c l e
i n f o
Article history: Received 9 January 2009 Available online 24 January 2009
Keywords: Tripartite efflux system Multidrug resistance EmrAB–TolC Single particle electron microscopy Oligomer Membrane transport
a b s t r a c t Tripartite efflux systems are responsible for the export of toxins across both the inner and outer membranes of Gram negative bacteria. Previous work has indicated that EmrAB–TolC from Escherichia coli is such a tripartite system, comprised of EmrB an MFS transporter, EmrA, a membrane fusion protein and TolC, an outer membrane channel. The whole complex is predicted to form a continuous channel allowing direct export from the cytoplasm to the exterior of the cell. Little is known, however, about the interactions between the individual components of this system. Reconstitution of EmrA + EmrB resulted in co-elution of the two proteins from a gel filtration column indicating formation of the EmrAB complex. Electron microscopic single particle analysis of the reconstituted EmrAB complex revealed the presence of particles approximately 240 140 Å, likely to correspond to two EmrAB dimers in a back-toback arrangement, suggesting the dimeric EmrAB form is the physiological state contrasting with the trimeric arrangement of the AcrAB–TolC system. Ó 2009 Elsevier Inc. All rights reserved.
Bacteria have evolved sophisticated systems for the removal of toxic molecules [1]. These efflux systems contribute to bacterial multidrug resistance, an increasing problem in the treatment of infectious diseases. One such transport system found in Gram negative bacteria is the three component membrane efflux complex that provides a continuous channel exporting a wide range of toxic compounds, including antibiotics, across both the inner (IM) and outer (OM) bacterial membranes [2]. These so-called tripartite efflux systems are comprised of an IM protein, usually a resistance nodulation division (RND) transporter or a major facilitator superfamily (MFS) transporter [1,2], which couples the energy stored in a cation gradient to the transport of molecules across the membrane, and an OM channel. The interactions between these two components are believed to be mediated by a periplasmically located protein usually referred to as a membrane fusion protein (MFP). The MFP is typically anchored to the inner membrane via a single transmembrane (TM) domain, or an N-terminal lipid group [3,4]. Whilst some of these tripartite efflux systems including AcrAB– TolC [5–7] and HlyBD–TolC [8,9] have been extensively studied, lit* Corresponding author. Fax: +44 207 594 3022. E-mail address: [email protected] (B. Byrne). 1 Present address: Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37212-6600, USA. 0006-291X/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2009.01.081
tle is known about the functionally related EmrAB–TolC. The EmrAB components were first identified in Escherichia coli more than a decade ago and shown to confer resistance to hydrophobic toxins such as carbonyl cyanide m-chlorophenyl-hydrazone (CCCP) [10]. Preliminary sequence analyses predicted that EmrB is a membrane protein with 14 suggested TM domains and homology to the MFS transporters, while EmrA has a large soluble C-terminal domain with a single N-terminal TM domain and homology to the MFP, HlyD. Based on homology to the HlyBD–TolC system, it was suggested that together with the outer membrane channel TolC, EmrAB forms a tripartite efflux system [11]. Information about tripartite efflux systems can be inferred from the high resolution structures of AcrB [6,12], MexA [13,14] and TolC [15]. The crystal structure of AcrB was the first example of an RND transporter and revealed a trimeric arrangement. Each monomer is comprised of 12 transmembrane domains and a large soluble region which projects roughly 70 Å into the periplasm. AcrA is the associated MFP, for which a partial structure is available [16] while TolC is the outer membrane channel whose structure [15] also revealed a trimeric arrangement. TolC is composed of an outer membrane b-barrel and a cylinder of coiled helices which projects 100 Å into the periplasm. Based on the individual structures it was postulated that AcrB would dock directly with TolC [5], although cross-linking studies suggest that interactions between the two are dependent on AcrA [17]. Being the predominant
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
multidrug efflux protein, AcrB is of major biological interest and much effort has been expended on elucidating the molecular mechanisms of action. Recent structural studies identified five separate drug binding sites within AcrB [12,18]. It is interesting to note that like AcrAB–TolC, EmrAB–TolC is constitutively expressed and the two systems appear to have overlapping substrate profiles [19,20]. It is unclear why E. coli and other Gram negative bacteria express both systems. In addition another question remains as to whether, given the overlapping range of substrates between AcrAB–TolC and EmrAB–TolC, EmrB also contains multiple drug binding sites. Despite the similarity in substrate profiles EmrB differs significantly from AcrB, in that it is an MFS transporter, rather than an RND transporter, lacking the large periplasmically located soluble domain suggested to be involved in the interactions between AcrAB and TolC. Therefore it is likely that both the mode of interaction between the three components and the export mechanisms are different in EmrAB–TolC. Using a composite of the available structures it has been possible to model the interactions between the separate components of a tripartite efflux system [13,14,21] and suggest a possible oligomeric state for the MFPs. Recent cross-linking studies combined with modelling have postulated that the interaction between TolC and AcrA is mediated via a coiled-coil interface [22]. However, the precise nature of the interactions between the separate components of any tripartite system has yet to be revealed and the oligomeric state of the complete complexes has yet to be determined. This study aimed to obtain a reliable estimate of the stoichiometry of the EmrAB complex and compare this with information known about other well-studied systems. The data revealed that the physiological form for the EmrAB complex is likely to be of a dimeric nature. Materials and methods Cloning and expression of EmrA and EmrB. The genes coding for EmrA and EmrB were amplified directly from E. coli cells. EmrA was cloned into pET24a (Novagen) and EmrB was cloned into a modified form of pTTQ18 [23,24] containing a C-terminal streptagII [24]. E. coli strains BL21(DE3) (Novagen) and C43(DE3) (Avidis) were used for protein expression. Maintenance and growth of these E. coli strains were achieved by culturing the bacteria in Luria broth (LB). Carbenicillin (100 lg/ml) and kanamycin (50 lg/ml) were used throughout all stages of growth to maintain plasmid integrity. For protein expression, cells harbouring the appropriate vector construct were grown at 37 °C to an OD600 0.4. Expression was induced by the addition of isopropyl-b-D-thiogalactopyranoside (IPTG) to a final concentration of 0.5 mM for both vector constructs. Growth continued for 3 h following induction at 30 °C. Purification of EmrA and EmrB. The preparation of the cell membranes was performed as described previously [25]. The membranes from 10 L EmrA expression cultures were suspended in 120 ml buffer A [50 mM Tris–HCl (pH 7.5), 400 mM NaCl] supplemented with 1% (w/v) dodecyl maltopyranoside (DDM), 0.5 mM PMSF and a protease inhibitor cocktail (Roche) and incubated for 30 min at 4 °C. After centrifugation at 100,000 g for 1 h the supernatant containing the solubilized proteins was incubated with Co2+ metal affinity resin (Stratagene) for 2 h. The resin was washed with five column volumes of buffer B [50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 0.05% DDM] supplemented with 30 mM imidazole followed by 10 column volumes of buffer B supplemented with 100 mM imidazole. The bound proteins were eluted with buffer B supplemented with 400 mM imidazole. EmrB was purified using a similar protocol. The isolated solubilized fraction was incubated
339
with Strep-Tactin resin (IBA) for 1 h at 4 °C. The resin was washed with 20 column volumes of buffer C (Buffer B supplemented with 1 mM EDTA) and the bound EmrB protein eluted with buffer C supplemented with 5 mM D-desthiobiotin. All proteins were concentrated and washed through a 30 or 50 kDa molecular weight cut off filter (Millipore). Protein purity was confirmed by EZ-blue (Sigma–Aldrich) stained SDS–PAGE gels. Protein concentration was determined by the Bradford method (Bio-Rad) using bovine serum albumin as a protein standard. SDS–PAGE and Western blotting analysis. Protein fractions were separated on 10% NuPAGE gels (Invitrogen) containing 0.1% SDS and then transferred to a PVDF membrane which was probed with an anti-His-tag antibody at 1:4000 dilution (Dianova) followed by goat anti-mouse IgG conjugated HRP (Dako) at 1:8000 dilution for detection of EmrA while anti-strep-tagII conjugated HRP (IBA) at 1:5000 dilution was used for detection of EmrB. Labelled proteins were detected using the ECL plus kit (Amersham Biosciences). Reconstitution of the EmrAB complex. EmrAB was reconstituted into liposomes by a detergent dilution method similar to that previously described [26,27]. A total of 8 mg purified concentrated EmrB and EmrA in 50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 0.1% DDM were mixed in EmrA:EmrB ratio of 3:1. The protein solution was then mixed with dissolved phospholipids in a ratio of 1:4 (1% phosphatidylcholine, Avanti Polar Lipids) in 50 mM Tris–HCl (pH 7.5), 200 mM NaCl. The sample was sonicated for 5 min (VCX500, Sonics) and then incubated on ice for 1 h followed by centrifugation at 20,000g, for 30 min. The concentrated supernatant (200 ll; approximately 3–4 mg) was applied to a Superose 6 (10/ 30) gel filtration column (Amersham BioSciences) equilibrated with 50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 0.1% DDM. Electron microscopy and single particle analyses. EmrAB complex was applied to carbon-coated copper 300 mesh grids and negatively stained with 2% uranyl acetate. Various dilutions were screened to achieve as even a distribution of protein complexes, as single particles, as possible. Imaging was performed at room temperature using a Philips CM100 electron microscope operating at 100 kV. Micrographs were recorded that displayed the first minima of their Fourier power spectra to be in the 21–23 Å range. No correction was made for the contrast transfer function, given this minima value and the presence of negative stain. Datasets were compiled using the automatic particle selection procedures of ‘boxer’, a module of the EMAN software package (v1.7) [28], at a sampling frequency of 2.44 Å/pixel. All subsequent image processing was performed using the Imagic-5 environment (Image Science GmbH, Berlin). Reference-free alignment, followed by multi-variate statistical analyses, allowed for initial 2D class averages to be identified which were then iteratively refined resulting in the final averages shown [29]. Results Expression and purification of EmrA and EmrB The proteins were purified using one-step affinity chromatography, yielding 0.4 mg pure EmrA and 0.2 mg pure EmrB/L LB culture, respectively, (Fig. 1). The identities of the purified proteins were confirmed by N-terminal sequencing. EmrA + EmrB can be reconstituted into a stable EmrAB complex in vitro The EmrA + EmrB complex was reconstituted into liposomes and then re-exchanged into detergent micelles on a gel filtration column, a step which also separates the EmrAB complex from free components. Following gel filtration, both EmrA and EmrB were
340
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
Fig. 1. Purification of EmrA and EmrB. SDS–PAGE analysis of the purified proteins; EmrA (lane 1) and EmrB (lane 2).
eluted from the column in the same fractions, as assessed by both Western blot using anti-His-tag antibody to detect EmrA and antistrep-tagII conjugated HRP to identify EmrB and EZ-stained SDS– PAGE analysis (Fig. 2A and B). The EmrAB complex found in fractions 2 and 3 was subjected to EM analysis. EM analysis of the EmrAB complex EmrAB complexes were negatively stained, imaged (Fig. 3A) and digitised to yield a dataset of 8700 particles. Reference-free alignment and classification resolved particles into two distinct populations based on overall size. Each subpopulation was treated de novo with iterative multi-reference alignments improving their respective signal-to-noise ratios. The subpopulation of larger sized complexes contained 1800 particles (circled; Fig. 3A), revealing averages dimeric in nature with dimensions of 240 140 Å (Fig. 3C). Other views appeared to consistently maintain a further dimeric symmetry, with a possible height of 130 Å. The preliminary three-dimensional (3D) reconstruction, shown in surface rendered view in Fig. 3D, was calculated by angular reconstitution using 60 characteristic 2D averages that were chosen for their overall quality and diverse orientation. Fourier shell correlation was used to estimate that the overall resolution achieved was, conservatively, in the region of 30 Å (see Supplementary Fig. panel B) and the reconstruction was Gaussian filtered accordingly. However, a missing cone of angular information was seen towards the centre of the C2 Eulerian map (see Supplementary Fig. panel A) hence caution must be exercised when interpreting the side view orientation. The volume of the reconstruction, assuming a density of 0.844 Da/Å3 at a threshold of three sigma, gave a calculated molecular mass of 268 kDa. The remaining 6900 particles were much smaller (Fig. 3F), having dimensions of 100 80 Å, with a double cylinder appearance merging at one end. The third characteristic average of this panel (Fig. 3F) has a modelled overlay, to scale, of two MexA subunits (PDB accession code 1VF7 [13]) arranged by visual inspection along the longest axis as a demonstration of overall sizing. The similar scale does give rise to the possibility that these averages (Fig. 3F) are dimeric EmrA. As a further control, a preparation of EmrB was imaged, but no significantly sized particles were observed (Fig. 3B) compared to those seen in Fig. 3C and F.
Fig. 2. Reconstitution of the EmrAB complex. Following reconstitution into liposomes EmrA + EmrB was loaded onto a gel filtration column. The elution profile is shown in (A). The initial large peak corresponds to large EmrA aggregates while the third smaller peak corresponds to free EmrA. The second peak contains the EmrAB complex (representative fractions, labelled 1, 2, 3, 4 shown for clarity). SDS– PAGE and Western blot using both anti-His and anti-Strep antibodies was performed on all fractions eluted from the column (B). Fractions 2 and 3 were analysed by electron microscopy.
Discussion The precise nature of the interactions between the individual components of tripartite efflux systems is not fully understood. Here we have investigated the interactions involved in the relatively unstudied EmrAB–TolC system. It proved possible to express EmrA and EmrB separately and reconstitute the EmrAB complex in the absence of TolC as seen previously for the related HlyB–HlyD complex [8]. The reconstituted EmrAB preparation was subjected to negative stain electron microscopic analysis as a means to obtain an estimate of the stoichiometry of the complex and its overall spatial size. The data consistently gave rise to particles with maximum dimensions of 240 140 130 Å. Similarly solved structures [30] with these dimensions are also typically in the range of 250– 500 kDa. These data also strongly suggest that each of the two main domains (Fig. 3C) are in themselves a dimeric EmrAB complex i.e., 2 (2 (42 kDa EmrA + 56 kDa EmrB)); 392 kDa in total, which would place the particles, observed here in three dimensions with an estimated mass of 268 kDa, to be in the same range. This result differs from the stoichiometry predicted for the functionally related AcrAB–TolC complex. The crystal structures of AcrB and TolC indicate a trimeric arrangement for both proteins [6,15]. It had been assumed in these papers for modelling purposes that a complete tripartite complex would be made up of three molecules of TolC, three molecules of the inner membrane protein and either 3, 6, 9 or 12 molecules of the MFP [13,14,31]. Previous EM analysis had suggested that AcrA only forms a dimer [32] while analysis of the relative cellular levels of the components of the MexAB–OprM efflux system indicated a hexameric arrangement
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
341
‘‘doughnut” shaped structure was obtained of AcrA by electron crystallography [31] indicating that these particles are likely to correspond to dimeric EmrA. This is supported by an earlier biochemical study [4] suggesting that EmrA forms dimers, with the interaction between the monomers mediated via a leucine zipper domain located in the N-terminal TM domain of the protein. This study has shown that the EmrAB–TolC tripartite efflux system differs considerably from the functionally related AcrAB–TolC in terms of the physiological oligomeric arrangement of the complex. Furthermore, this study has also produced protein suitable for further structural and functional studies, essential for a fuller understanding of drug transport mechanisms in this system. Acknowledgments M.T. was supported by the Department of Science and Technology Laboratory, Porton Down, UK, G.S. and P.J.F.H. were supported by the EU European Membrane Protein Consortium (E-MeP, contrct LSHG-CT-2004-504601); J.N. currently holds a Royal Society University Research Fellowship. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2009.01.081. References
Fig. 3. Electron micrograph and single particle averaged images of negatively stained EmrAB and EmrA. (A) Typical region from micrograph with complexes attributed to EmrAB (circled) and background EmrA (arrowed). (B) Typical region of negatively stained EmrB complexes. (C) Characteristic single particle average views, in different orientations, attributed to EmrAB. The first two frames show particles that have been imaged whilst sitting face-up or face-down on the copper grid. (D) In the same orientation as (C) a C2 symmetry-imposed 3D map calculated from 60 characteristic views chosen with as diverse range of angles as possible, although some angular information was seen to be missing in the side view orientation, then Gaussian filtered conservatively down to 30 Å resolution (see Supplementary Fig. panel A and B). (E) Reprojections through the 3D map, again using angles to ensure the same orientation as in (C). (F) 2D averages calculated from a small subset of particles observed within the same sample, possibly representative of EmrA alone. The third average of this panel has been overlaid with a known crystal structure of MexA (red; 1VF7.pdb), modelled here to scale and in a putative dimeric form, to illustrate the similar size of our data and the MexA structure. Scale bars in A and B represent 100 nm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
for the MFP, MexA [33]. It should be noted that in the EM data collected for our study, only particles with a twofold symmetry were observed, again strongly indicating that the EmrAB complex is likely to form a dimer. Furthermore, as all possible single particles were extracted from the micrographs used and, through image classification, a second subpopulation of much smaller particles was revealed of 100 80 Å (Fig. 3F), it is highly likely that excess protein present in the preparation is consistently dimeric in form. Indeed, these smaller particles can be compared favourably in terms of both size and shape to the high resolution X-ray structure of the related protein, MexA [13,14]. Fig. 3F shows an overlay of the structure of MexA on one of the small particles illustrating the similarity in size and shape. In addition, a similar low resolution
[1] M. Putman, H.W. van Veen, W. Konings, Molecular properties of bacterial multidrug transporters, Microbiol. Mol. Biol. Rev. 4 (2000) 672–693. [2] I.T. Paulsen, M.H. Brown, R.A. Skerry, Proton-dependent multidrug efflux systems, Microbiol. Rev. 60 (1996) 575–608. [3] A. Sharff, C. Fanutti, J. Shyi, C. Calladine, B. Luisi, The role of the TolC family in protein transport and multidrug efflux, Eur. J. Biochem. 268 (2001) 5011–5026. [4] M.I. Borges-Walmsley, J. Beauchamp, S.M. Kelly, K. Jumel, D. Candlish, S.E. Harding, N.C. Price, A.R. Walmsley, Identification of oligomerisation and drug binding domains of the membrane fusion proteins EmrA, J. Biol. Chem. 278 (2003) 12903–12912. [5] S. Murakami, A. Yamaguchi, Multidrug exporting secondary transporters, Curr. Opin. Struct. Biol. 13 (2003) (2003) 443–452. [6] S. Murakami, R. Nakashima, E. Yamashita, A. Yamaguchi, Crystal structure of bacterial multidrug efflux transporter AcrB, Nature 419 (2002) 587–593. [7] E.B. Tikhonova, H.I. Zgurskaya, AcrA, AcrB, TolC of Escherichia coli form a stable intermembrane multidrug efflux complex, J. Biol. Chem. 279 (2004) 32116– 32124. [8] T. Thanabalu, E. Koronakis, C. Hughes, V. Koronakis, Substrate induced assembly of a contiguous channel for protein export from E. Coli: reversible bridging of an inner-membrane translocase to an outer membrane exit pore, EMBO J. 17 (1998) 6487–6496. [9] J. Zaitseva, S. Jenewan, C. Oswald, T. Jumpertz, I.B. Holland, L. Schmitt, A molecular understanding of the catalytic cycle of the nucleotide-binding domain of the ABC transporter HlyB, Biochem. Soc. Trans. 33 (2005) 990–995. [10] O. Lomovskaya, K. Lewis, Emr, an Escherichia coli locus for multidrug resistance, Proc. Natl. Acad. Sci. USA 89 (1992) 8938–8942. [11] K. Lewis, Translocases: a bacterial tunnel for drugs and proteins, Curr. Biol. 10 (2000) R678–R681. [12] E.W. Yu, G. McDermott, H.I. Zgurskaya, H. Nikaido, D.E. Koshland Jr., Structural basis of multiple drug-binding capacity of the AcrB multidrug efflux pump, Science 300 (2003) 976–980. [13] H. Akama, T. Matsuura, S. Kashiwagi, H. Yoneymara, S. Narita, T. Tsukihara, A. Nakagawa, T. Nakae, Crystal structure of the membrane fusion proteins, MexA, of the multidrug transpoter in Pseudomonas aeruginosa, J. Biol. Chem. 279 (2004) (2004) 25939–25942. [14] M. Higgins, E. Bokma, E. Koronakis, C. Hughes, V. Koronakis, Structure of the periplasmic component of a bacterial drug efflux pump, Proc. Natl. Acad. Sci. USA 101 (2004) 9994–9999. [15] V. Koronakis, A. Sharff, E. Koronakis, B. Luisi, C. Hughes, Crystal structure of the bacterial membrane protein TolC central to multidrug efflux and protein export, Nature 405 (2000) 914–919. [16] J. Mikolosko, K. Bobyk, H.I. Zgurskaya, P. Ghosh, Conformational flexibility in the multidrug efflux system protein AcrA, Structure 14 (2006) 577–587. [17] T. Touze, J. Eswaran, E. Bokma, E. Koronakis, C. Hughes, V. Koronakis, Interactions underlying assembly of the Escherichia coli AcrAB–TolC multidrug efflux system, Mol. Microbiol. 53 (2004) 697–706. [18] E.W. Yu, J.R. Aires, G. McDermott, H. Nikaido, A periplasmic drug-binding site of the AcrB multidrug efflux pump: a crystallographic and site-directed mutagenesis study, J. Bacteriol. 187 (2005) 6804–6815.
342
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
[19] C.A. Elkins, L.B. Mullis, Substrate competition studies using whole-cell accumulation assays with the major tripartite multidrug efflux pumps of Escherichia coli, Antimicrob. Agents Chemother. 51 (2007) 923–929. [20] C.A. Elkins, L.B. Mullis, Mammalian steroid hormones are substrates for the major RND- and MFS-type tripartite efflux pumps of Escherichia coli, J. Bacteriol. 188 (2006) 1191–1195. [21] J. Eswaran, E. Koronakis, M.K. Higgins, C. Hughes, V. Koronakis, Three’s company: component structures bring a closer view of tripartite drug efflux pumps, Curr. Opin. Struct. Biol. 14 (2004) 741–747. [22] S. Lobedanz, E. Bokma, M.F. Symmons, E. Koronakis, C. Hughes, V. Koronakis, A periplasmic coilecoil interface underlying TolC recruitment and the assembly of bacterial drug efflux pumps, Proc. Natl. Acad. Sci. USA 14 (2007) 4612–4617. [23] M. Saidijam, G. Psakis, J.L. Clough, J. Meuller, S. Suzuki, C.J. Hoyle, S.L. Palmer, S.M. Morrison, M.K. Pos, R.C. Essenberg, M.C.J. Maiden, A. Abu-Bakr, S.G. Baumberg, M.J. Stark, A. Ward, J. O’Reilly, N.G. Rutherford, M.K. Phillips-Jones, P.J.F. Henderson, Collection and characterisation of bacterial membrane proteins, FEBS Lett. 555 (2003) 170–175. [24] S. Suzuki, P.J.F. Henderson, The hydantoin transport protein from Microbacterium liquefaciens, J. Bacteriol. 188 (2006) 3329–3336. [25] L. Waskell, Expression and purification of the amphipathic form of rabbit cytochrome b5 in Escherichia coli, in: B.S. Selinsky (Ed.), Membrane Protein Protocols: Expression, Purification, and Characterization, Humana Press, New Jersey, 2003, pp. 3–9.
[26] H.I. Zgurskaya, H. Nikaido, Bypassing the periplasm: reconstitution of the AcrAB multidrug efflux pump of Escherichia coli, Proc. Natl. Acad. Sci. USA 96 (1999) 7190–7195. [27] P. Curnow, M. Lorch, K. Charalambous, P.J. Booth, The reconstitution and activity of the small multidrug transporter EmrE is modulated by non-bilayer lipid composition, J. Mol. Biol. 343 (2004) 213–222. [28] S.J. Ludtke, D.H. Chen, J.L. Song, D.T. Chuang, W. Chiu, Seeing GroEL at 6 Å resolution by single particle electron cryomicroscopy, Structure 12 (2004) 1129–1136. [29] M. van Heel, G. Harauz, E.V. Orlova, A new generation of the IMAGIC image processing system, J. Struct. Biol. 116 (1996) 17–24. [30] J. Nield, E.V. Orlova, E.P. Morris, B. Gowen, M. van Heel, J. Barber, 3D map of the plant photosystem II supercomplex obtained by cryoelectron microscopy and single particle analysis, Nat. Struct. Biol. 7 (2000) 44–47. [31] J. Fernandez-Recio, F. Walas, L. Federici, J. Venkatesh Pratap, V.N. Bavro, R.N. Miguel, K. Mizuguchi, B. Luisi, A model of transmembrane drug-efflux pump from Gram-negative bacteria, FEBS Lett. 578 (2004) 5–9. [32] A.J. Avila-Sakar, S. Misaghi, E.M. Wilson-Kubalek, K.H. Downing, H. Zgurskaya, H. Nikaido, E. Nogales, Lipid-layer crystallization and preliminary threedimensional structural analysis of AcrA, the periplasmic component of a bacterial multidrug efflux pump, J. Struct. Biol. 136 (2001) 81–88. [33] S. Narita, S. Eda, E. Yoshihara, T. Nakae, Linkage of the efflux-pump expression level with substrate extrusion rate in the MexAB–OprM efflux pump of Pseudomonas aeruginosa, Biochem. Biophys. Res. Commun. 308 (2003) 922–926.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
The multidrug resistance efflux complex, EmrAB from Escherichia coli forms a dimer in vitro Mikio Tanabe a,b,1, Gerda Szakonyi c, Katherine A. Brown b, Peter J.F. Henderson c, Jon Nield d, Bernadette Byrne a,* a
Division of Molecular Biosciences, Imperial College London, Exhibition Road, London SW7 2AZ, UK Division of Cell and Molecular Biology, Imperial College London, Exhibition Road, London SW7 2AZ, UK c Institute of Membrane and Systems Biology, University of Leeds, Leeds LS2 9JT, UK d School of Biological and Chemical Sciences, Queen Mary University of London, Mile End, London E1 4NS, UK b
a r t i c l e
i n f o
Article history: Received 9 January 2009 Available online 24 January 2009
Keywords: Tripartite efflux system Multidrug resistance EmrAB–TolC Single particle electron microscopy Oligomer Membrane transport
a b s t r a c t Tripartite efflux systems are responsible for the export of toxins across both the inner and outer membranes of Gram negative bacteria. Previous work has indicated that EmrAB–TolC from Escherichia coli is such a tripartite system, comprised of EmrB an MFS transporter, EmrA, a membrane fusion protein and TolC, an outer membrane channel. The whole complex is predicted to form a continuous channel allowing direct export from the cytoplasm to the exterior of the cell. Little is known, however, about the interactions between the individual components of this system. Reconstitution of EmrA + EmrB resulted in co-elution of the two proteins from a gel filtration column indicating formation of the EmrAB complex. Electron microscopic single particle analysis of the reconstituted EmrAB complex revealed the presence of particles approximately 240 140 Å, likely to correspond to two EmrAB dimers in a back-toback arrangement, suggesting the dimeric EmrAB form is the physiological state contrasting with the trimeric arrangement of the AcrAB–TolC system. Ó 2009 Elsevier Inc. All rights reserved.
Bacteria have evolved sophisticated systems for the removal of toxic molecules [1]. These efflux systems contribute to bacterial multidrug resistance, an increasing problem in the treatment of infectious diseases. One such transport system found in Gram negative bacteria is the three component membrane efflux complex that provides a continuous channel exporting a wide range of toxic compounds, including antibiotics, across both the inner (IM) and outer (OM) bacterial membranes [2]. These so-called tripartite efflux systems are comprised of an IM protein, usually a resistance nodulation division (RND) transporter or a major facilitator superfamily (MFS) transporter [1,2], which couples the energy stored in a cation gradient to the transport of molecules across the membrane, and an OM channel. The interactions between these two components are believed to be mediated by a periplasmically located protein usually referred to as a membrane fusion protein (MFP). The MFP is typically anchored to the inner membrane via a single transmembrane (TM) domain, or an N-terminal lipid group [3,4]. Whilst some of these tripartite efflux systems including AcrAB– TolC [5–7] and HlyBD–TolC [8,9] have been extensively studied, lit* Corresponding author. Fax: +44 207 594 3022. E-mail address: [email protected] (B. Byrne). 1 Present address: Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37212-6600, USA. 0006-291X/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2009.01.081
tle is known about the functionally related EmrAB–TolC. The EmrAB components were first identified in Escherichia coli more than a decade ago and shown to confer resistance to hydrophobic toxins such as carbonyl cyanide m-chlorophenyl-hydrazone (CCCP) [10]. Preliminary sequence analyses predicted that EmrB is a membrane protein with 14 suggested TM domains and homology to the MFS transporters, while EmrA has a large soluble C-terminal domain with a single N-terminal TM domain and homology to the MFP, HlyD. Based on homology to the HlyBD–TolC system, it was suggested that together with the outer membrane channel TolC, EmrAB forms a tripartite efflux system [11]. Information about tripartite efflux systems can be inferred from the high resolution structures of AcrB [6,12], MexA [13,14] and TolC [15]. The crystal structure of AcrB was the first example of an RND transporter and revealed a trimeric arrangement. Each monomer is comprised of 12 transmembrane domains and a large soluble region which projects roughly 70 Å into the periplasm. AcrA is the associated MFP, for which a partial structure is available [16] while TolC is the outer membrane channel whose structure [15] also revealed a trimeric arrangement. TolC is composed of an outer membrane b-barrel and a cylinder of coiled helices which projects 100 Å into the periplasm. Based on the individual structures it was postulated that AcrB would dock directly with TolC [5], although cross-linking studies suggest that interactions between the two are dependent on AcrA [17]. Being the predominant
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
multidrug efflux protein, AcrB is of major biological interest and much effort has been expended on elucidating the molecular mechanisms of action. Recent structural studies identified five separate drug binding sites within AcrB [12,18]. It is interesting to note that like AcrAB–TolC, EmrAB–TolC is constitutively expressed and the two systems appear to have overlapping substrate profiles [19,20]. It is unclear why E. coli and other Gram negative bacteria express both systems. In addition another question remains as to whether, given the overlapping range of substrates between AcrAB–TolC and EmrAB–TolC, EmrB also contains multiple drug binding sites. Despite the similarity in substrate profiles EmrB differs significantly from AcrB, in that it is an MFS transporter, rather than an RND transporter, lacking the large periplasmically located soluble domain suggested to be involved in the interactions between AcrAB and TolC. Therefore it is likely that both the mode of interaction between the three components and the export mechanisms are different in EmrAB–TolC. Using a composite of the available structures it has been possible to model the interactions between the separate components of a tripartite efflux system [13,14,21] and suggest a possible oligomeric state for the MFPs. Recent cross-linking studies combined with modelling have postulated that the interaction between TolC and AcrA is mediated via a coiled-coil interface [22]. However, the precise nature of the interactions between the separate components of any tripartite system has yet to be revealed and the oligomeric state of the complete complexes has yet to be determined. This study aimed to obtain a reliable estimate of the stoichiometry of the EmrAB complex and compare this with information known about other well-studied systems. The data revealed that the physiological form for the EmrAB complex is likely to be of a dimeric nature. Materials and methods Cloning and expression of EmrA and EmrB. The genes coding for EmrA and EmrB were amplified directly from E. coli cells. EmrA was cloned into pET24a (Novagen) and EmrB was cloned into a modified form of pTTQ18 [23,24] containing a C-terminal streptagII [24]. E. coli strains BL21(DE3) (Novagen) and C43(DE3) (Avidis) were used for protein expression. Maintenance and growth of these E. coli strains were achieved by culturing the bacteria in Luria broth (LB). Carbenicillin (100 lg/ml) and kanamycin (50 lg/ml) were used throughout all stages of growth to maintain plasmid integrity. For protein expression, cells harbouring the appropriate vector construct were grown at 37 °C to an OD600 0.4. Expression was induced by the addition of isopropyl-b-D-thiogalactopyranoside (IPTG) to a final concentration of 0.5 mM for both vector constructs. Growth continued for 3 h following induction at 30 °C. Purification of EmrA and EmrB. The preparation of the cell membranes was performed as described previously [25]. The membranes from 10 L EmrA expression cultures were suspended in 120 ml buffer A [50 mM Tris–HCl (pH 7.5), 400 mM NaCl] supplemented with 1% (w/v) dodecyl maltopyranoside (DDM), 0.5 mM PMSF and a protease inhibitor cocktail (Roche) and incubated for 30 min at 4 °C. After centrifugation at 100,000 g for 1 h the supernatant containing the solubilized proteins was incubated with Co2+ metal affinity resin (Stratagene) for 2 h. The resin was washed with five column volumes of buffer B [50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 0.05% DDM] supplemented with 30 mM imidazole followed by 10 column volumes of buffer B supplemented with 100 mM imidazole. The bound proteins were eluted with buffer B supplemented with 400 mM imidazole. EmrB was purified using a similar protocol. The isolated solubilized fraction was incubated
339
with Strep-Tactin resin (IBA) for 1 h at 4 °C. The resin was washed with 20 column volumes of buffer C (Buffer B supplemented with 1 mM EDTA) and the bound EmrB protein eluted with buffer C supplemented with 5 mM D-desthiobiotin. All proteins were concentrated and washed through a 30 or 50 kDa molecular weight cut off filter (Millipore). Protein purity was confirmed by EZ-blue (Sigma–Aldrich) stained SDS–PAGE gels. Protein concentration was determined by the Bradford method (Bio-Rad) using bovine serum albumin as a protein standard. SDS–PAGE and Western blotting analysis. Protein fractions were separated on 10% NuPAGE gels (Invitrogen) containing 0.1% SDS and then transferred to a PVDF membrane which was probed with an anti-His-tag antibody at 1:4000 dilution (Dianova) followed by goat anti-mouse IgG conjugated HRP (Dako) at 1:8000 dilution for detection of EmrA while anti-strep-tagII conjugated HRP (IBA) at 1:5000 dilution was used for detection of EmrB. Labelled proteins were detected using the ECL plus kit (Amersham Biosciences). Reconstitution of the EmrAB complex. EmrAB was reconstituted into liposomes by a detergent dilution method similar to that previously described [26,27]. A total of 8 mg purified concentrated EmrB and EmrA in 50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 0.1% DDM were mixed in EmrA:EmrB ratio of 3:1. The protein solution was then mixed with dissolved phospholipids in a ratio of 1:4 (1% phosphatidylcholine, Avanti Polar Lipids) in 50 mM Tris–HCl (pH 7.5), 200 mM NaCl. The sample was sonicated for 5 min (VCX500, Sonics) and then incubated on ice for 1 h followed by centrifugation at 20,000g, for 30 min. The concentrated supernatant (200 ll; approximately 3–4 mg) was applied to a Superose 6 (10/ 30) gel filtration column (Amersham BioSciences) equilibrated with 50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 0.1% DDM. Electron microscopy and single particle analyses. EmrAB complex was applied to carbon-coated copper 300 mesh grids and negatively stained with 2% uranyl acetate. Various dilutions were screened to achieve as even a distribution of protein complexes, as single particles, as possible. Imaging was performed at room temperature using a Philips CM100 electron microscope operating at 100 kV. Micrographs were recorded that displayed the first minima of their Fourier power spectra to be in the 21–23 Å range. No correction was made for the contrast transfer function, given this minima value and the presence of negative stain. Datasets were compiled using the automatic particle selection procedures of ‘boxer’, a module of the EMAN software package (v1.7) [28], at a sampling frequency of 2.44 Å/pixel. All subsequent image processing was performed using the Imagic-5 environment (Image Science GmbH, Berlin). Reference-free alignment, followed by multi-variate statistical analyses, allowed for initial 2D class averages to be identified which were then iteratively refined resulting in the final averages shown [29]. Results Expression and purification of EmrA and EmrB The proteins were purified using one-step affinity chromatography, yielding 0.4 mg pure EmrA and 0.2 mg pure EmrB/L LB culture, respectively, (Fig. 1). The identities of the purified proteins were confirmed by N-terminal sequencing. EmrA + EmrB can be reconstituted into a stable EmrAB complex in vitro The EmrA + EmrB complex was reconstituted into liposomes and then re-exchanged into detergent micelles on a gel filtration column, a step which also separates the EmrAB complex from free components. Following gel filtration, both EmrA and EmrB were
340
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
Fig. 1. Purification of EmrA and EmrB. SDS–PAGE analysis of the purified proteins; EmrA (lane 1) and EmrB (lane 2).
eluted from the column in the same fractions, as assessed by both Western blot using anti-His-tag antibody to detect EmrA and antistrep-tagII conjugated HRP to identify EmrB and EZ-stained SDS– PAGE analysis (Fig. 2A and B). The EmrAB complex found in fractions 2 and 3 was subjected to EM analysis. EM analysis of the EmrAB complex EmrAB complexes were negatively stained, imaged (Fig. 3A) and digitised to yield a dataset of 8700 particles. Reference-free alignment and classification resolved particles into two distinct populations based on overall size. Each subpopulation was treated de novo with iterative multi-reference alignments improving their respective signal-to-noise ratios. The subpopulation of larger sized complexes contained 1800 particles (circled; Fig. 3A), revealing averages dimeric in nature with dimensions of 240 140 Å (Fig. 3C). Other views appeared to consistently maintain a further dimeric symmetry, with a possible height of 130 Å. The preliminary three-dimensional (3D) reconstruction, shown in surface rendered view in Fig. 3D, was calculated by angular reconstitution using 60 characteristic 2D averages that were chosen for their overall quality and diverse orientation. Fourier shell correlation was used to estimate that the overall resolution achieved was, conservatively, in the region of 30 Å (see Supplementary Fig. panel B) and the reconstruction was Gaussian filtered accordingly. However, a missing cone of angular information was seen towards the centre of the C2 Eulerian map (see Supplementary Fig. panel A) hence caution must be exercised when interpreting the side view orientation. The volume of the reconstruction, assuming a density of 0.844 Da/Å3 at a threshold of three sigma, gave a calculated molecular mass of 268 kDa. The remaining 6900 particles were much smaller (Fig. 3F), having dimensions of 100 80 Å, with a double cylinder appearance merging at one end. The third characteristic average of this panel (Fig. 3F) has a modelled overlay, to scale, of two MexA subunits (PDB accession code 1VF7 [13]) arranged by visual inspection along the longest axis as a demonstration of overall sizing. The similar scale does give rise to the possibility that these averages (Fig. 3F) are dimeric EmrA. As a further control, a preparation of EmrB was imaged, but no significantly sized particles were observed (Fig. 3B) compared to those seen in Fig. 3C and F.
Fig. 2. Reconstitution of the EmrAB complex. Following reconstitution into liposomes EmrA + EmrB was loaded onto a gel filtration column. The elution profile is shown in (A). The initial large peak corresponds to large EmrA aggregates while the third smaller peak corresponds to free EmrA. The second peak contains the EmrAB complex (representative fractions, labelled 1, 2, 3, 4 shown for clarity). SDS– PAGE and Western blot using both anti-His and anti-Strep antibodies was performed on all fractions eluted from the column (B). Fractions 2 and 3 were analysed by electron microscopy.
Discussion The precise nature of the interactions between the individual components of tripartite efflux systems is not fully understood. Here we have investigated the interactions involved in the relatively unstudied EmrAB–TolC system. It proved possible to express EmrA and EmrB separately and reconstitute the EmrAB complex in the absence of TolC as seen previously for the related HlyB–HlyD complex [8]. The reconstituted EmrAB preparation was subjected to negative stain electron microscopic analysis as a means to obtain an estimate of the stoichiometry of the complex and its overall spatial size. The data consistently gave rise to particles with maximum dimensions of 240 140 130 Å. Similarly solved structures [30] with these dimensions are also typically in the range of 250– 500 kDa. These data also strongly suggest that each of the two main domains (Fig. 3C) are in themselves a dimeric EmrAB complex i.e., 2 (2 (42 kDa EmrA + 56 kDa EmrB)); 392 kDa in total, which would place the particles, observed here in three dimensions with an estimated mass of 268 kDa, to be in the same range. This result differs from the stoichiometry predicted for the functionally related AcrAB–TolC complex. The crystal structures of AcrB and TolC indicate a trimeric arrangement for both proteins [6,15]. It had been assumed in these papers for modelling purposes that a complete tripartite complex would be made up of three molecules of TolC, three molecules of the inner membrane protein and either 3, 6, 9 or 12 molecules of the MFP [13,14,31]. Previous EM analysis had suggested that AcrA only forms a dimer [32] while analysis of the relative cellular levels of the components of the MexAB–OprM efflux system indicated a hexameric arrangement
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
341
‘‘doughnut” shaped structure was obtained of AcrA by electron crystallography [31] indicating that these particles are likely to correspond to dimeric EmrA. This is supported by an earlier biochemical study [4] suggesting that EmrA forms dimers, with the interaction between the monomers mediated via a leucine zipper domain located in the N-terminal TM domain of the protein. This study has shown that the EmrAB–TolC tripartite efflux system differs considerably from the functionally related AcrAB–TolC in terms of the physiological oligomeric arrangement of the complex. Furthermore, this study has also produced protein suitable for further structural and functional studies, essential for a fuller understanding of drug transport mechanisms in this system. Acknowledgments M.T. was supported by the Department of Science and Technology Laboratory, Porton Down, UK, G.S. and P.J.F.H. were supported by the EU European Membrane Protein Consortium (E-MeP, contrct LSHG-CT-2004-504601); J.N. currently holds a Royal Society University Research Fellowship. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2009.01.081. References
Fig. 3. Electron micrograph and single particle averaged images of negatively stained EmrAB and EmrA. (A) Typical region from micrograph with complexes attributed to EmrAB (circled) and background EmrA (arrowed). (B) Typical region of negatively stained EmrB complexes. (C) Characteristic single particle average views, in different orientations, attributed to EmrAB. The first two frames show particles that have been imaged whilst sitting face-up or face-down on the copper grid. (D) In the same orientation as (C) a C2 symmetry-imposed 3D map calculated from 60 characteristic views chosen with as diverse range of angles as possible, although some angular information was seen to be missing in the side view orientation, then Gaussian filtered conservatively down to 30 Å resolution (see Supplementary Fig. panel A and B). (E) Reprojections through the 3D map, again using angles to ensure the same orientation as in (C). (F) 2D averages calculated from a small subset of particles observed within the same sample, possibly representative of EmrA alone. The third average of this panel has been overlaid with a known crystal structure of MexA (red; 1VF7.pdb), modelled here to scale and in a putative dimeric form, to illustrate the similar size of our data and the MexA structure. Scale bars in A and B represent 100 nm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
for the MFP, MexA [33]. It should be noted that in the EM data collected for our study, only particles with a twofold symmetry were observed, again strongly indicating that the EmrAB complex is likely to form a dimer. Furthermore, as all possible single particles were extracted from the micrographs used and, through image classification, a second subpopulation of much smaller particles was revealed of 100 80 Å (Fig. 3F), it is highly likely that excess protein present in the preparation is consistently dimeric in form. Indeed, these smaller particles can be compared favourably in terms of both size and shape to the high resolution X-ray structure of the related protein, MexA [13,14]. Fig. 3F shows an overlay of the structure of MexA on one of the small particles illustrating the similarity in size and shape. In addition, a similar low resolution
[1] M. Putman, H.W. van Veen, W. Konings, Molecular properties of bacterial multidrug transporters, Microbiol. Mol. Biol. Rev. 4 (2000) 672–693. [2] I.T. Paulsen, M.H. Brown, R.A. Skerry, Proton-dependent multidrug efflux systems, Microbiol. Rev. 60 (1996) 575–608. [3] A. Sharff, C. Fanutti, J. Shyi, C. Calladine, B. Luisi, The role of the TolC family in protein transport and multidrug efflux, Eur. J. Biochem. 268 (2001) 5011–5026. [4] M.I. Borges-Walmsley, J. Beauchamp, S.M. Kelly, K. Jumel, D. Candlish, S.E. Harding, N.C. Price, A.R. Walmsley, Identification of oligomerisation and drug binding domains of the membrane fusion proteins EmrA, J. Biol. Chem. 278 (2003) 12903–12912. [5] S. Murakami, A. Yamaguchi, Multidrug exporting secondary transporters, Curr. Opin. Struct. Biol. 13 (2003) (2003) 443–452. [6] S. Murakami, R. Nakashima, E. Yamashita, A. Yamaguchi, Crystal structure of bacterial multidrug efflux transporter AcrB, Nature 419 (2002) 587–593. [7] E.B. Tikhonova, H.I. Zgurskaya, AcrA, AcrB, TolC of Escherichia coli form a stable intermembrane multidrug efflux complex, J. Biol. Chem. 279 (2004) 32116– 32124. [8] T. Thanabalu, E. Koronakis, C. Hughes, V. Koronakis, Substrate induced assembly of a contiguous channel for protein export from E. Coli: reversible bridging of an inner-membrane translocase to an outer membrane exit pore, EMBO J. 17 (1998) 6487–6496. [9] J. Zaitseva, S. Jenewan, C. Oswald, T. Jumpertz, I.B. Holland, L. Schmitt, A molecular understanding of the catalytic cycle of the nucleotide-binding domain of the ABC transporter HlyB, Biochem. Soc. Trans. 33 (2005) 990–995. [10] O. Lomovskaya, K. Lewis, Emr, an Escherichia coli locus for multidrug resistance, Proc. Natl. Acad. Sci. USA 89 (1992) 8938–8942. [11] K. Lewis, Translocases: a bacterial tunnel for drugs and proteins, Curr. Biol. 10 (2000) R678–R681. [12] E.W. Yu, G. McDermott, H.I. Zgurskaya, H. Nikaido, D.E. Koshland Jr., Structural basis of multiple drug-binding capacity of the AcrB multidrug efflux pump, Science 300 (2003) 976–980. [13] H. Akama, T. Matsuura, S. Kashiwagi, H. Yoneymara, S. Narita, T. Tsukihara, A. Nakagawa, T. Nakae, Crystal structure of the membrane fusion proteins, MexA, of the multidrug transpoter in Pseudomonas aeruginosa, J. Biol. Chem. 279 (2004) (2004) 25939–25942. [14] M. Higgins, E. Bokma, E. Koronakis, C. Hughes, V. Koronakis, Structure of the periplasmic component of a bacterial drug efflux pump, Proc. Natl. Acad. Sci. USA 101 (2004) 9994–9999. [15] V. Koronakis, A. Sharff, E. Koronakis, B. Luisi, C. Hughes, Crystal structure of the bacterial membrane protein TolC central to multidrug efflux and protein export, Nature 405 (2000) 914–919. [16] J. Mikolosko, K. Bobyk, H.I. Zgurskaya, P. Ghosh, Conformational flexibility in the multidrug efflux system protein AcrA, Structure 14 (2006) 577–587. [17] T. Touze, J. Eswaran, E. Bokma, E. Koronakis, C. Hughes, V. Koronakis, Interactions underlying assembly of the Escherichia coli AcrAB–TolC multidrug efflux system, Mol. Microbiol. 53 (2004) 697–706. [18] E.W. Yu, J.R. Aires, G. McDermott, H. Nikaido, A periplasmic drug-binding site of the AcrB multidrug efflux pump: a crystallographic and site-directed mutagenesis study, J. Bacteriol. 187 (2005) 6804–6815.
342
M. Tanabe et al. / Biochemical and Biophysical Research Communications 380 (2009) 338–342
[19] C.A. Elkins, L.B. Mullis, Substrate competition studies using whole-cell accumulation assays with the major tripartite multidrug efflux pumps of Escherichia coli, Antimicrob. Agents Chemother. 51 (2007) 923–929. [20] C.A. Elkins, L.B. Mullis, Mammalian steroid hormones are substrates for the major RND- and MFS-type tripartite efflux pumps of Escherichia coli, J. Bacteriol. 188 (2006) 1191–1195. [21] J. Eswaran, E. Koronakis, M.K. Higgins, C. Hughes, V. Koronakis, Three’s company: component structures bring a closer view of tripartite drug efflux pumps, Curr. Opin. Struct. Biol. 14 (2004) 741–747. [22] S. Lobedanz, E. Bokma, M.F. Symmons, E. Koronakis, C. Hughes, V. Koronakis, A periplasmic coilecoil interface underlying TolC recruitment and the assembly of bacterial drug efflux pumps, Proc. Natl. Acad. Sci. USA 14 (2007) 4612–4617. [23] M. Saidijam, G. Psakis, J.L. Clough, J. Meuller, S. Suzuki, C.J. Hoyle, S.L. Palmer, S.M. Morrison, M.K. Pos, R.C. Essenberg, M.C.J. Maiden, A. Abu-Bakr, S.G. Baumberg, M.J. Stark, A. Ward, J. O’Reilly, N.G. Rutherford, M.K. Phillips-Jones, P.J.F. Henderson, Collection and characterisation of bacterial membrane proteins, FEBS Lett. 555 (2003) 170–175. [24] S. Suzuki, P.J.F. Henderson, The hydantoin transport protein from Microbacterium liquefaciens, J. Bacteriol. 188 (2006) 3329–3336. [25] L. Waskell, Expression and purification of the amphipathic form of rabbit cytochrome b5 in Escherichia coli, in: B.S. Selinsky (Ed.), Membrane Protein Protocols: Expression, Purification, and Characterization, Humana Press, New Jersey, 2003, pp. 3–9.
[26] H.I. Zgurskaya, H. Nikaido, Bypassing the periplasm: reconstitution of the AcrAB multidrug efflux pump of Escherichia coli, Proc. Natl. Acad. Sci. USA 96 (1999) 7190–7195. [27] P. Curnow, M. Lorch, K. Charalambous, P.J. Booth, The reconstitution and activity of the small multidrug transporter EmrE is modulated by non-bilayer lipid composition, J. Mol. Biol. 343 (2004) 213–222. [28] S.J. Ludtke, D.H. Chen, J.L. Song, D.T. Chuang, W. Chiu, Seeing GroEL at 6 Å resolution by single particle electron cryomicroscopy, Structure 12 (2004) 1129–1136. [29] M. van Heel, G. Harauz, E.V. Orlova, A new generation of the IMAGIC image processing system, J. Struct. Biol. 116 (1996) 17–24. [30] J. Nield, E.V. Orlova, E.P. Morris, B. Gowen, M. van Heel, J. Barber, 3D map of the plant photosystem II supercomplex obtained by cryoelectron microscopy and single particle analysis, Nat. Struct. Biol. 7 (2000) 44–47. [31] J. Fernandez-Recio, F. Walas, L. Federici, J. Venkatesh Pratap, V.N. Bavro, R.N. Miguel, K. Mizuguchi, B. Luisi, A model of transmembrane drug-efflux pump from Gram-negative bacteria, FEBS Lett. 578 (2004) 5–9. [32] A.J. Avila-Sakar, S. Misaghi, E.M. Wilson-Kubalek, K.H. Downing, H. Zgurskaya, H. Nikaido, E. Nogales, Lipid-layer crystallization and preliminary threedimensional structural analysis of AcrA, the periplasmic component of a bacterial multidrug efflux pump, J. Struct. Biol. 136 (2001) 81–88. [33] S. Narita, S. Eda, E. Yoshihara, T. Nakae, Linkage of the efflux-pump expression level with substrate extrusion rate in the MexAB–OprM efflux pump of Pseudomonas aeruginosa, Biochem. Biophys. Res. Commun. 308 (2003) 922–926.