- Email: [email protected]
Studying subunit–subunit interactions in a bacterial ABC transporterby in vitro assembly
Biochimica et Biophysica Acta 1798 (2010) 1250–1253
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a m e m
Studying subunit–subunit interactions in a bacterial ABC transporterby in vitro assembly Viola Eckey 1, Heidi Landmesser, Erwin Schneider ⁎ Institut für Biologie, Physiologie der Mikroorganismen, Humboldt Universität zu Berlin, Chausseestr. 117, D-10115 Berlin, Germany
a r t i c l e
i n f o
Article history: Received 4 January 2010 Received in revised form 17 February 2010 Accepted 1 March 2010 Available online 10 March 2010 Keywords: ABC transporter ArtJ-(MP)2 Q loop EAA loop In vitro assembly Geobacillus stearothermophilus
a b s t r a c t The thermostable arginine ABC transporter of Geobacillus stearothermophilus consists of a solute binding protein, ArtJ; a transmembrane subunit, ArtM; and a nucleotide-binding subunit, ArtP. An ArtM/His6–ArtP complex was functionally assembled from separately purified subunits as demonstrated by assaying stimulation of its ATPase activity by arginine-loaded ArtJ in proteoliposomes. Studying in vitro assembly with variants carrying mutations in the conserved Q loop and/or the EAA loop of ArtP and ArtM, respectively, confirmed the predicted roles of both motifs in intersubunit signaling and physical interaction, respectively. In vitro assembly is considered a useful tool for investigating assembly defects of ABC transporters caused by mutations. © 2010 Elsevier B.V. All rights reserved.
1. Introduction ABC transporters are found in all living organisms and couple the translocation of diverse substrates encompassing small ions to proteins across the cell membrane to the free energy of ATP hydrolysis [1,2]. ABC transporters share a common architectural organization comprising two pore-forming transmembrane domains/subunits (TMDs) and two nucleotide binding domains/subunits (NBDs). A large subfamily of ABC transporters found exclusively in prokaryotes allows the uptake of nutrients essential for proliferation and survival of the cells. Canonical importers require an additional extracellular substrate receptor (or binding protein) for function [3]. Structural and biochemical data of isolated NBDs and complete transporters suggest that alternating access of the translocation pore to the intracellular and extracellular space achieves a net transport of substrate. Current models imply that the alternation of the TMDs between an outwardfacing and an inward-facing conformation is energized by the NBDs' catalytic cycle. The latter comprises NBD dimer closure upon ATP binding, hydrolysis of ATP in the closed conformer and reopening towards a semi-closed, ADP-bound state. In ABC importers, the
Abbreviations: ABC, ATP-binding cassette; DDM, n-dodecyl-β-D-maltopyranoside; DS, decanoylsucrose; His-tag, hexahistidine tag; IPTG, isopropyl-β-D-thio-galactopyranoside; PMSF, phenylmethylsulfonylfluoride ⁎ Corresponding author. Tel.: +49 30 2093 8121; fax: +49 30 20938126. E-mail address: [email protected] (E. Schneider). 1 Present address: Bundesanstalt für Materialprüfung, Unter den Eichen 87, D-12205 Berlin, Germany. 0005-2736/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2010.03.001
interaction of liganded substrate binding proteins with the TMDs is considered to trigger complete closure of the ATP-bound NBD dimers (reviewed in Ref. [4]). X-ray structures of fully assembled transporters show that the conserved Q loops of NBDs are in close proximity to the EAA loops (also termed L loops or ‘coupling helices’) of the pore-forming subunits [5]. These structural findings are in line with genetic and biochemical studies [6–8] that also provided evidence for both motifs to be involved in intersubunit communication. However, dissecting their structural and/or catalytic functions by biochemical means is often hampered by the fact that purification of transporter variants carrying mutations in these motifs results only in a subpopulation of protein complexes that hide their possible role in assembly. On the other hand, quantitative analysis of assembly defects caused by mutations in subcellular fractions is rather unreliable since mostly based on immunoblots. Thus, investigating the structural and/or catalytic significance of amino acid residues requires an alternative tool. Here, we describe an in vitro assembly protocol allowing the identification of residues critical for subunit–subunit association. To this end, we have employed the small and thermostable ABC transporter for positively charged amino acids, ArtJ-(MP)2 of Geobacillus stearothermophilus. The transporter is composed of a solute binding protein, ArtJ, and a homodimer each of the membraneintegral subunit, ArtM, and the ATPase, ArtP [9]. Structures of ArtJ in complex with different ligands [10] and of ArtP with bound nucleotides have been solved (PDB IDs: 2OUK, 2OLK, 2OLJ, 2QOH, 3C4J, and 3C41). Our results show that an EAA loop mutant is defective in assembly while a Q loop mutant is impaired in transport activity although normally assembled.
V. Eckey et al. / Biochimica et Biophysica Acta 1798 (2010) 1250–1253
2. Materials and methods 2.1. Overproduction and purification of ArtM The artM gene was amplified from plasmid pRF2 (artMP on pQE60) [9] and cloned into expression vector pQE60 (Qiagen, Germany) providing a C-terminal His-tag, resulting in plasmid pVE15. Cells of Escherichia coli strain Rosetta2 (pLacI, pVE15) were grown in TB medium [11], supplemented with ampicillin and chloramphenicol at 30 °C to an OD650 of 0.8 –1.0. Expression of artM was induced by IPTG (5 mM) and growth continued at 22 °C overnight. After cell disrupture, ArtM was solubilized from the membrane fraction in 50 mM Tris–HCl, pH 8, 5% (v/v) glycerol, 0.1 mM PMSF, 1.2% DS. After ultracentrifugation, the supernatant was diluted with twice the volume of 50 mM MOPS/KOH (pH 7.5), 0.3 M NaCl, 5% (v/v) glycerol, 0.1 mM PMSF, 0.05% DDM, 4 mM 2-mercaptoethanol, 5 mM imidazole and incubated with a Co2+-loaded affinity matrix (TALON, Clontech) for 1 h at 4 °C. Subsequently, the matrix was washed with the same buffer in the presence of 10 mM imidazole. ArtM was eluted by supplementing the buffer with 0.25 M imidazole. ArtM-containing fractions were pooled, concentrated by a centrifugal device (VIVASpin, YM 10, Millipore), and subsequently passed through a PD10 desalting column (GE Healthcare) to remove imidazole. 2.2. Overproduction and purification of ArtP The artP gene was subcloned from plasmid pRF2 [9] into expression vector pET15 providing an N-terminal His-tag (Novagen, Germany). The resulting plasmid pVE6 was transferred into E. coli strain Rosetta (pLysS) (Stratagene). Cells were grown in LB (Luria Bertani) medium at 37 °C to A650 = 0.5, IPTG (0.5 mM) was added to induce artP expression, and growth continued for 4 h. After disintegration of cells and ultracentrifugation, the soluble fraction containing ArtP was applied to a TALON resin previously equilibrated with 50 mM MOPS/KOH, pH7.5, 0,15 M NaCl, 5% (v/v) glycerol, 10 mM ATP, 4 mM 2-mercaptoethanol, 0.1 mM PMSF. After removal of unbound material, ArtP was eluted with 0.25 M imidazole. ArtPcontaining fractions were further treated as described for ArtM (2.1). 2.3. Overproduction and purification of ArtJ ArtJ was overproduced and purified as described in Ref. [10]. Briefly, cells of E. coli strain Rosetta (pLysS) transformed with pSN1 (artJ on expression vector pET15b) were grown in LB (Luria Bertani) medium supplemented with ampicillin and chloramphenicol at 37 °C. At an OD650 of 0.5, overexpression was induced by adding 0.5 mM IPTG, and growth was continued for 4 h. Cells were harvested; resuspended in 50 mM MOPS/KOH, pH 7.5, 0.15 M NaCl, 0.1 mM PMSF; and disintegrated by passage through a French Press. After ultracentrifugation, the supernatant containing His6-tagged ArtJ was applied to metal affinity chromatography using a TALON resin and purified essentially as described for ArtP (2.2). 2.4. Overproduction and purification of Art(MP)2 The Art(MP)2 complex was purified by making use of a His6-tag engineered to the carboxyl terminus of ArtP as described in [9] with minor modifications. E. coli strain Rosetta (pLacI) transformed with pRF2 was grown in TB-ampicillin/chloramphenicol to an OD650 of 1 prior to the addition of 0.5 mM IPTG for induction of gene expression. For purification of Art(MP)2, the membrane fraction obtained after cell disruption and ultracentrifugation (1 h at 200,000 × g) was resuspended in 50 mM MOPS/KOH, pH 7.5, 5% (v/v) glycerol, 0.1 mM PMSF and solubilized with 1.2% DS. After ultracentrifugation (30 min at 200,000 × g), the supernatant was adjusted to 10 mM ATP,
1251
0.3 M NaCl, 4 mM 2-mercaptoethanol and purified on a TALON matrix in the presence of 0.05% DDM. Pooled Art(MP)2-containing fractions were further treated as described for ArtM (2.1). 2.5. Assembly of ArtMP complex from isolated subunits His6-ArtP was incubated with ArtM, which was previously treated with thrombin (Thrombin-Clean Cleavage Kit from Sigma, Taufkirchen, Germany) to cleave the his-tag, at an equimolar ratio in 50 mM MOPS/ KOH (pH 7.5), 5% glycerol, 0.3 M NaCl, 0.01% DDM for 12 h at 4 °C. Subsequently, the mixture was incubated with a TALON matrix for 1 h at 4 °C under gentle shaking and poured into a disposable column. The flow-through, containing tagless ArtM, was collected, concentrated, and analyzed by SDS-PAGE to confirm removal of the tag. 2.6. Preparation of proteoliposomes Incorporation of complex variants into liposomes prepared from G. stearothermophilus total lipids in the presence or absence of arginine-loaded ArtJ was carried out as described in Ref. [10]. Briefly, lipids (20 mg) were dried under a stream of nitrogen, slowly redissolved in 1 ml 50 mM MOPS/KOH, pH 7.5, containing 1% octylß-D-glucopyranoside, and sonicated for 15 min. Subsequently, Art (MP)2 variants (50 µg), ArtJ (230 µg), and L-arginine (1 mM, final) were added to 125 µl of the lipid–detergent mixture, resulting in a final volume of 300 µl. Proteoliposomes were formed by removal of detergent by adsorption to Biobeads (100 mg; BioRad, München) at 4 °C overnight. After replacing the beads with a new batch, incubation continued for 2 h. Then the mixture was centrifuged for 1 min at 10,000 × g to pellet the beads and subsequently, proteoliposomes were recovered by ultracentifugation for 30 min at 220,000 × g, resuspended in 50 mM MOPS/KOH, pH 7.5, and assayed for ATPase activity. As controls, proteoliposomes were prepared by the same procedure but omitting ArtJ variantsL-arginine. 2.7. Site-directed mutagenesis Plasmids pVE65 and pVE66, expressing artM(G127P) and artP (N87G), respectively, were constructed by using plasmids pVE15 and pVE6 as templates and Stratagene's QuikChange kit according to the manufacturer's instructions. 2.8. Analytical procedures ATPase activity was assayed as described in Ref. [12]. Assay temperature was 55 and60 °C, respectively, for ArtP and Art(MP)2. SDS-PAGE and protein determination was carried out as in Ref. [13]. Protein amount of proteoliposomes was based on quantitation of ArtP in SDS gels. 3. Results 3.1. Purification and assembly of His6-ArtP and ArtM His6-ArtP was purified to near homogeneity from the cytosolic fraction of E. coli strain Rosetta (pVE6, pLysS) by Co2+ affinity chromatography (Fig. 1, lane 3). The average yield of purified protein was 40 mg per 1 l culture. His6-ArtP exhibited a specific ATPase activity at 55 °C (the ambient temperature of G. stearothermophilus) of 0.03–0.04 µmol phosphate/min/mg, which fits the range of 0.01 to 1.65 µmol phosphate/min/mg reported for NBDs of other ABC transporters [14]. The membrane-integral subunit, ArtM-His6, was overproduced in E. coli strain Rosetta2 (pVE15, pLacI) and solubilized from the membrane fraction by 1.2% DS. Purification was achieved by metal affinity chromatography in the presence of 0.05% DDM, similarly to
1252
V. Eckey et al. / Biochimica et Biophysica Acta 1798 (2010) 1250–1253
Fig. 1. SDS-PAGE of purified ArtP and ArtM subunits. ArtM-His6 and His6-ArtP were purified as described in Materials and methods. ArtM-His6 was subsequently treated with thrombin to cleave the hexahistidine tag. Lanes: 1, molecular mass standards; 2, ArtM; 3, His6-ArtP; 4, in vivo assembled Art(MP-His6)2.
ArtP (Fig. 1A, lane 2). Routinely, about 1.2 mg of purified protein was obtained from six liters of culture. To test the capability of the separately purified subunits to assemble a functional Art(MP)2 complex, tag-less ArtM was incubated with His6-ArtP at an equimolar ratio and subsequently subjected to metal affinity chromatography on a TALON matrix. Analysis of column fractions by SDS-PAGE revealed virtually no protein in the flowthrough or the wash fractions. ArtM co-eluted exclusively with His6ArtP, thereby suggesting that an association of the proteins had occurred (Fig. 2, panel A). Quantitation of His6-ArtP and ArtM from the peak fraction (Fig. 2, panel A, lane 8) by densitometric scanning revealed a ratio of His6-ArtP/ArtM of 1.7, which is in excellent agreement with that obtained from an in vivo assembled complex (1.6; Fig. 1, lane 4) (please note that the theoretical ratio of 1 is not found probably due to less efficient staining of hydrophobic ArtM). Subsequent analysis of concentrated ArtMP-containing fractions by gel filtration further demonstrated stable complex formation between both proteins (not shown). Furthermore, co-elution was neither affected by the addition of substrate-loaded binding protein (ArtJ/arginine), ATP or phospholipids to the assembly mixture (not shown). Next, we assayed ATPase activity of the co-eluted His6-ArtP/ArtM complex in proteoliposomes as a means to proof functional association. It is well established that binding protein-dependent ABC transporters, including the ArtJ-Art(MP)2 complex [9,10], exhibit low intrinsic ATPase
Fig. 2. Co-elution of ArtM and His6-ArtP variants from a Co2+ affinity matrix. Experiments were performed as described in Materials and methods. Lanes: 1, ArtM variant; 2, His6-ArtP variant; 3, molecular mass standards (25 and 20 kDa; see legend to Fig. 1); 4, assembly mixtures containing indicated ArtM and His6-ArtP variants; 5 and 6, wash fractions 1 and 10; 7–11, elution fractions.
Fig. 3. Time dependence of ATPase activity of in vivo- and in vitro- assembled Art(MP)2 wild-type complexes. Proteoliposomes containing Art(MP)2 were prepared in the presence (closed symbols) or absence (open symbols) of arginine-loaded ArtJ and assayed for ATPase activity as described in Materials and methods. Symbols: circles, Art (MP-His6)2 (in vivo); squares, ArtM + His6-ArtP; triangles, ArtM + His6-ArtP assembled in the presence of 2 mM ATP.
activity in proteoliposomes that is stimulated severalfold by the cognate substrate-loaded receptor [3]. This observation is generally taken as evidence for coupling of ATP hydrolysis to substrate translocation. As shown in Fig. 3, ATPase activity of the co-eluted complex was negligible in the absence of ArtJ/arginine but was stimulated severalfold in its presence, similar to in vivo-assembled Art(MP)2. Moreover, an 85% inhibition of the ATPase activity was observed in the presence of 0.5 mM ortho-vanadate, a specific inhibitor of ABC transporters [1] (not shown). Fig. 3 also shows that the presence of 2 mM ATP during the assembly reaction had no significant effect on ATPase activity. Together, these results clearly indicate that separately purified His6ArtP and ArtM are competent to assemble a fully functional transport complex in vitro. 3.2. Co-elution as a means to study the effect of mutations on complex assembly The above results prompted us to test the potential use of the established protocol as a tool to characterize the effect of mutations on the assembly process. To address this question, we chose residue Asn-87 from the Q loop of ArtP and Gly-127 from the EAA loop of ArtM for mutagenesis to proline and glycine, respectively. Residues from both motifs have been implicated in inter-domain signaling and/or physical interaction between subunits [4,5]. However, mutations affecting these interactions have been studied only in a few complete transporters, including the maltose transporter of E. coli/Salmonella [6–8,15] and the ferric hydroxamate transporter of E. coli [16]. The respective ArtP and ArtM variants were purified and analyzed for complex formation as established for the wild-type proteins. As shown in Fig. 2B, an ArtM/His6–ArtP(N87G) complex co-eluted with a P/M ratio of 1.7, very similar to wild type. Stability of the complex was confirmed by gel filtration (not shown). ATPase activity of separately purified His6-ArtP (N87G) was 0.03 µmol Pi/min/mg and thus in good agreement with that of wild-type His6-ArtP. However, ArtJ/arginine-stimulated ATPase activity of the assembled ArtM/His6–ArtP(N87G) complex in proteoliposomes was severely impaired (Fig. 4). These results indicate that the Q loop mutation is likely to affect signaling between ArtM and ArtP but not complex formation. Moreover, our finding is consistent with a positional change of Asn-87 upon ATP hydrolysis as observed in the crystal structures (compare pdb 3C41 and 2QOH). In contrast, the ArtM(G127P) variant displayed a strong defect in forming a complex with His6-ArtP as revealed by substoichiometric coelution from the TALON matrix (P/M ratio = 4.3; Fig. 2C). Further
V. Eckey et al. / Biochimica et Biophysica Acta 1798 (2010) 1250–1253
1253
Acknowledgment This work was supported by the Deutsche Forschungsgemeinschaft (SFB 449, TP B14). References
Fig. 4. ATPase activities of in vivo- and in vitro-assembled wild-type and mutant ArtMP complexes. Proteoliposomes were prepared as in legend to Fig. 3. Basal activities recorded in the absence of ArtJ/arginine were subtracted. SE mean (n ≥ 3).
ore, no complex formation was observed when both variants were combined (Fig. 2D). Consequently, no ATPase activity could be detected (Fig. 4).
4. Discussion In this communication, we demonstrate that ArtP and ArtM can be individually overproduced, purified, and functionally assembled to a complex displaying ArtJ/arginine-stimulated ATPase activity in proteoliposomes highly similar to in vivo-assembled Art(MP)2. Functional re-assembly after overproduction of intact complexes and subsequent dissociation of NBDs by treatment with urea was first reported for the histidine [17] and maltose [12,18] transporters. More recently, Horn et al. [19] demonstrated in vitro assembly of the glycine betaine transporter OpuA from purified subunits in detergent solution, while reconstitution of the LolCDE exporter involved in lipoprotein sorting to the outer membrane was studied in proteoliposomes [20]. In contrast to our study, the stability of the latter was highly dependent on phospholipids or ATP. Unlike these reports, we show here for the first time that in vitro assembly provides a means to study the role of individual amino acid residues on subunit–subunit interactions. Variants carrying any mutation can be assayed for assembly or signaling defects by co-elution and subsequent functional analysis in proteoliposomes. As proof of principle, we have confirmed the proposed roles of residues from the Q and EAA loops, respectively, of ArtP and ArtM. Moreover, the successful overexpression of ArtM and ArtP will be of further use for structural investigations of the Art (MP)2 transporter, e.g., by solid-state NMR. Such experiments are underway.
[1] E.B. Holland, S. Cole, K. Kuchler, C.F. Higgins (Eds.), ABC proteins: from bacteria to man, Elsevier, Amsterdam, 2003. [2] C.F. Higgins, ABC transporters: from microorganisms to man, Annu. Rev. Cell Biol. 8 (1992) 67–113. [3] E. Schneider, Import of solutes by ABC transporters—the maltose and other systems, in: E.B. Holland, S. Cole, K. Kuchler, C. Higgins (Eds.), ABC proteins: From bacteria to man, Elsevier, Amsterdam, 2003, pp. 157–185. [4] M.L. Oldham, A.L. Davidson, J. Chen, Structural insights into ABC transporter mechanism, Curr. Opin. Struct. Biol. 18 (2008) 726–733. [5] A.L. Davidson, E. Dassa, C. Orelle, J. Chen, Structure, function, and evolution of ATPbinding cassette systems, Microbiol. Mol. Biol. Rev. 72 (2008) 317–364. [6] M. Mourez, M. Hofnung, E. Dassa, Subunit interaction in ABC transporters. A conserved sequence in hydrophobic membrane proteins of periplasmic permeases define sites of interaction with the helical domain of ABC subunits, EMBO J. 16 (1997) 3066–3077. [7] S. Hunke, M. Mourez, M. Jéhanno, E. Dassa, E. Schneider, ATP modulates subunit– subunit interactions in an ATP-binding-cassette transporter (MalFGK2) determined by site-directed chemical cross-linking, J. Biol. Chem. 275 (2000) 15526–15534. [8] M.L. Daus, M. Grote, P. Müller, M. Doebber, A. Herrmann, H.-J. Steinhoff, E. Dassa, E. Schneider, ATP-driven MalK dimer closure and re-opening and conformational changes of the ‘EAA’ motifs are crucial for function of the maltose ATP-binding cassette transporter (MalFGK2), J. Biol. Chem. 282 (2007) 22387–22396. [9] R. Fleischer, A. Wengner, F. Scheffel, H. Landmesser, E. Schneider, Identification of a gene cluster encoding an arginine ATP-binding-cassette transporter in the genome of the thermophilic gram-positive bacterium Geobacillus stearothermophilus DSMZ 13240, Microbiology 151 (2005) 835–840. [10] A. Vahedi-Faridi, V. Eckey, F. Scheffel, C. Alings, H. Landmesser, E. Schneider, W. Saenger, Crystal structures and mutational analysis of the arginine-, lysine-, histidine-binding protein ArtJ from Geobacillus stearothermophilus. Implications for interactions of ArtJ with its cognate ATP-binding cassette transporter, Art (MP)2, J. Mol. Biol. 375 (2008) 448–459. [11] K.D. Tartoff, C.A. Hobbs, Improved media for growing plasmid and cosmid clones, Bethesda Research Labs Focus, vol. 9, 1987, p. 12. [12] H. Landmesser, A. Stein, B. Blüschke, M. Brinkmann, S. Hunke, E. Schneider, Largescale purification, dissociation and functional reassembly of the maltose ATP-binding cassette transporter (MalFGK2) of Salmonella typhimurium, Biochim. Biophys. Acta 1565 (2002) 64–72. [13] M.L. Daus, H. Landmesser, A. Schlosser, P. Müller, A. Herrmann, E. Schneider, ATP induces conformational changes of periplasmic loop regions of the maltose ATP-binding cassette transporter, J. Biol. Chem. 281 (2006) 3856–3865. [14] I.B. Holland, M.A. Blight, ABC-ATPases, adaptable energy generators fuelling transmembrane movement of a variety of molecules in organisms from bacteria to humans, J. Mol. Biol. 293 (1999) 381–399. [15] M.I. Tapia, M. Mourez, M. Hofnung, E. Dassa, Structure–function study of MalF protein by random mutagenesis, J. Bacteriol. 181 (1999) 2267–2272. [16] W. Köster, B. Böhm, Point mutations in 2 conserved glycine residues within the integral membrane protein FhuB affect iron(III)hydroxamate transport, Mol. Gen. Genet. 232 (1992) 399–407. [17] P.-Q. Liu, G.F.-L. Ames, In vitro disassembly and reassembly of an ABC transporter, the histidine permease, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 3495–3500. [18] S. Sharma, J.A. Davis, T. Ayvaz, B. Traxler, A.L. Davidson, Functional reassembly of the Escherichia coli maltose transporter following purification of a MalF-MalG subassembly, J. Bacteriol. 187 (2005) 2908–2911. [19] C. Horn, E. Bremer, L. Schmitt, Functional overexpression and in vitro re-association of OpuA, an osmotically regulated ABC-transport complex from Bacillus subtilis, FEBS Lett. 579 (2005) 5765–5768. [20] K. Kanamaru, N. Taniguchi, S. Miyamoto, S.-I. Narita, H. Tokuda, Complete reconstitution of an ATP-binding cassette transporter LolCDE complex from separately isolated subunits, FEBS J. 274 (2007) 3034–3043.
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a m e m
Studying subunit–subunit interactions in a bacterial ABC transporterby in vitro assembly Viola Eckey 1, Heidi Landmesser, Erwin Schneider ⁎ Institut für Biologie, Physiologie der Mikroorganismen, Humboldt Universität zu Berlin, Chausseestr. 117, D-10115 Berlin, Germany
a r t i c l e
i n f o
Article history: Received 4 January 2010 Received in revised form 17 February 2010 Accepted 1 March 2010 Available online 10 March 2010 Keywords: ABC transporter ArtJ-(MP)2 Q loop EAA loop In vitro assembly Geobacillus stearothermophilus
a b s t r a c t The thermostable arginine ABC transporter of Geobacillus stearothermophilus consists of a solute binding protein, ArtJ; a transmembrane subunit, ArtM; and a nucleotide-binding subunit, ArtP. An ArtM/His6–ArtP complex was functionally assembled from separately purified subunits as demonstrated by assaying stimulation of its ATPase activity by arginine-loaded ArtJ in proteoliposomes. Studying in vitro assembly with variants carrying mutations in the conserved Q loop and/or the EAA loop of ArtP and ArtM, respectively, confirmed the predicted roles of both motifs in intersubunit signaling and physical interaction, respectively. In vitro assembly is considered a useful tool for investigating assembly defects of ABC transporters caused by mutations. © 2010 Elsevier B.V. All rights reserved.
1. Introduction ABC transporters are found in all living organisms and couple the translocation of diverse substrates encompassing small ions to proteins across the cell membrane to the free energy of ATP hydrolysis [1,2]. ABC transporters share a common architectural organization comprising two pore-forming transmembrane domains/subunits (TMDs) and two nucleotide binding domains/subunits (NBDs). A large subfamily of ABC transporters found exclusively in prokaryotes allows the uptake of nutrients essential for proliferation and survival of the cells. Canonical importers require an additional extracellular substrate receptor (or binding protein) for function [3]. Structural and biochemical data of isolated NBDs and complete transporters suggest that alternating access of the translocation pore to the intracellular and extracellular space achieves a net transport of substrate. Current models imply that the alternation of the TMDs between an outwardfacing and an inward-facing conformation is energized by the NBDs' catalytic cycle. The latter comprises NBD dimer closure upon ATP binding, hydrolysis of ATP in the closed conformer and reopening towards a semi-closed, ADP-bound state. In ABC importers, the
Abbreviations: ABC, ATP-binding cassette; DDM, n-dodecyl-β-D-maltopyranoside; DS, decanoylsucrose; His-tag, hexahistidine tag; IPTG, isopropyl-β-D-thio-galactopyranoside; PMSF, phenylmethylsulfonylfluoride ⁎ Corresponding author. Tel.: +49 30 2093 8121; fax: +49 30 20938126. E-mail address: [email protected] (E. Schneider). 1 Present address: Bundesanstalt für Materialprüfung, Unter den Eichen 87, D-12205 Berlin, Germany. 0005-2736/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2010.03.001
interaction of liganded substrate binding proteins with the TMDs is considered to trigger complete closure of the ATP-bound NBD dimers (reviewed in Ref. [4]). X-ray structures of fully assembled transporters show that the conserved Q loops of NBDs are in close proximity to the EAA loops (also termed L loops or ‘coupling helices’) of the pore-forming subunits [5]. These structural findings are in line with genetic and biochemical studies [6–8] that also provided evidence for both motifs to be involved in intersubunit communication. However, dissecting their structural and/or catalytic functions by biochemical means is often hampered by the fact that purification of transporter variants carrying mutations in these motifs results only in a subpopulation of protein complexes that hide their possible role in assembly. On the other hand, quantitative analysis of assembly defects caused by mutations in subcellular fractions is rather unreliable since mostly based on immunoblots. Thus, investigating the structural and/or catalytic significance of amino acid residues requires an alternative tool. Here, we describe an in vitro assembly protocol allowing the identification of residues critical for subunit–subunit association. To this end, we have employed the small and thermostable ABC transporter for positively charged amino acids, ArtJ-(MP)2 of Geobacillus stearothermophilus. The transporter is composed of a solute binding protein, ArtJ, and a homodimer each of the membraneintegral subunit, ArtM, and the ATPase, ArtP [9]. Structures of ArtJ in complex with different ligands [10] and of ArtP with bound nucleotides have been solved (PDB IDs: 2OUK, 2OLK, 2OLJ, 2QOH, 3C4J, and 3C41). Our results show that an EAA loop mutant is defective in assembly while a Q loop mutant is impaired in transport activity although normally assembled.
V. Eckey et al. / Biochimica et Biophysica Acta 1798 (2010) 1250–1253
2. Materials and methods 2.1. Overproduction and purification of ArtM The artM gene was amplified from plasmid pRF2 (artMP on pQE60) [9] and cloned into expression vector pQE60 (Qiagen, Germany) providing a C-terminal His-tag, resulting in plasmid pVE15. Cells of Escherichia coli strain Rosetta2 (pLacI, pVE15) were grown in TB medium [11], supplemented with ampicillin and chloramphenicol at 30 °C to an OD650 of 0.8 –1.0. Expression of artM was induced by IPTG (5 mM) and growth continued at 22 °C overnight. After cell disrupture, ArtM was solubilized from the membrane fraction in 50 mM Tris–HCl, pH 8, 5% (v/v) glycerol, 0.1 mM PMSF, 1.2% DS. After ultracentrifugation, the supernatant was diluted with twice the volume of 50 mM MOPS/KOH (pH 7.5), 0.3 M NaCl, 5% (v/v) glycerol, 0.1 mM PMSF, 0.05% DDM, 4 mM 2-mercaptoethanol, 5 mM imidazole and incubated with a Co2+-loaded affinity matrix (TALON, Clontech) for 1 h at 4 °C. Subsequently, the matrix was washed with the same buffer in the presence of 10 mM imidazole. ArtM was eluted by supplementing the buffer with 0.25 M imidazole. ArtM-containing fractions were pooled, concentrated by a centrifugal device (VIVASpin, YM 10, Millipore), and subsequently passed through a PD10 desalting column (GE Healthcare) to remove imidazole. 2.2. Overproduction and purification of ArtP The artP gene was subcloned from plasmid pRF2 [9] into expression vector pET15 providing an N-terminal His-tag (Novagen, Germany). The resulting plasmid pVE6 was transferred into E. coli strain Rosetta (pLysS) (Stratagene). Cells were grown in LB (Luria Bertani) medium at 37 °C to A650 = 0.5, IPTG (0.5 mM) was added to induce artP expression, and growth continued for 4 h. After disintegration of cells and ultracentrifugation, the soluble fraction containing ArtP was applied to a TALON resin previously equilibrated with 50 mM MOPS/KOH, pH7.5, 0,15 M NaCl, 5% (v/v) glycerol, 10 mM ATP, 4 mM 2-mercaptoethanol, 0.1 mM PMSF. After removal of unbound material, ArtP was eluted with 0.25 M imidazole. ArtPcontaining fractions were further treated as described for ArtM (2.1). 2.3. Overproduction and purification of ArtJ ArtJ was overproduced and purified as described in Ref. [10]. Briefly, cells of E. coli strain Rosetta (pLysS) transformed with pSN1 (artJ on expression vector pET15b) were grown in LB (Luria Bertani) medium supplemented with ampicillin and chloramphenicol at 37 °C. At an OD650 of 0.5, overexpression was induced by adding 0.5 mM IPTG, and growth was continued for 4 h. Cells were harvested; resuspended in 50 mM MOPS/KOH, pH 7.5, 0.15 M NaCl, 0.1 mM PMSF; and disintegrated by passage through a French Press. After ultracentrifugation, the supernatant containing His6-tagged ArtJ was applied to metal affinity chromatography using a TALON resin and purified essentially as described for ArtP (2.2). 2.4. Overproduction and purification of Art(MP)2 The Art(MP)2 complex was purified by making use of a His6-tag engineered to the carboxyl terminus of ArtP as described in [9] with minor modifications. E. coli strain Rosetta (pLacI) transformed with pRF2 was grown in TB-ampicillin/chloramphenicol to an OD650 of 1 prior to the addition of 0.5 mM IPTG for induction of gene expression. For purification of Art(MP)2, the membrane fraction obtained after cell disruption and ultracentrifugation (1 h at 200,000 × g) was resuspended in 50 mM MOPS/KOH, pH 7.5, 5% (v/v) glycerol, 0.1 mM PMSF and solubilized with 1.2% DS. After ultracentrifugation (30 min at 200,000 × g), the supernatant was adjusted to 10 mM ATP,
1251
0.3 M NaCl, 4 mM 2-mercaptoethanol and purified on a TALON matrix in the presence of 0.05% DDM. Pooled Art(MP)2-containing fractions were further treated as described for ArtM (2.1). 2.5. Assembly of ArtMP complex from isolated subunits His6-ArtP was incubated with ArtM, which was previously treated with thrombin (Thrombin-Clean Cleavage Kit from Sigma, Taufkirchen, Germany) to cleave the his-tag, at an equimolar ratio in 50 mM MOPS/ KOH (pH 7.5), 5% glycerol, 0.3 M NaCl, 0.01% DDM for 12 h at 4 °C. Subsequently, the mixture was incubated with a TALON matrix for 1 h at 4 °C under gentle shaking and poured into a disposable column. The flow-through, containing tagless ArtM, was collected, concentrated, and analyzed by SDS-PAGE to confirm removal of the tag. 2.6. Preparation of proteoliposomes Incorporation of complex variants into liposomes prepared from G. stearothermophilus total lipids in the presence or absence of arginine-loaded ArtJ was carried out as described in Ref. [10]. Briefly, lipids (20 mg) were dried under a stream of nitrogen, slowly redissolved in 1 ml 50 mM MOPS/KOH, pH 7.5, containing 1% octylß-D-glucopyranoside, and sonicated for 15 min. Subsequently, Art (MP)2 variants (50 µg), ArtJ (230 µg), and L-arginine (1 mM, final) were added to 125 µl of the lipid–detergent mixture, resulting in a final volume of 300 µl. Proteoliposomes were formed by removal of detergent by adsorption to Biobeads (100 mg; BioRad, München) at 4 °C overnight. After replacing the beads with a new batch, incubation continued for 2 h. Then the mixture was centrifuged for 1 min at 10,000 × g to pellet the beads and subsequently, proteoliposomes were recovered by ultracentifugation for 30 min at 220,000 × g, resuspended in 50 mM MOPS/KOH, pH 7.5, and assayed for ATPase activity. As controls, proteoliposomes were prepared by the same procedure but omitting ArtJ variantsL-arginine. 2.7. Site-directed mutagenesis Plasmids pVE65 and pVE66, expressing artM(G127P) and artP (N87G), respectively, were constructed by using plasmids pVE15 and pVE6 as templates and Stratagene's QuikChange kit according to the manufacturer's instructions. 2.8. Analytical procedures ATPase activity was assayed as described in Ref. [12]. Assay temperature was 55 and60 °C, respectively, for ArtP and Art(MP)2. SDS-PAGE and protein determination was carried out as in Ref. [13]. Protein amount of proteoliposomes was based on quantitation of ArtP in SDS gels. 3. Results 3.1. Purification and assembly of His6-ArtP and ArtM His6-ArtP was purified to near homogeneity from the cytosolic fraction of E. coli strain Rosetta (pVE6, pLysS) by Co2+ affinity chromatography (Fig. 1, lane 3). The average yield of purified protein was 40 mg per 1 l culture. His6-ArtP exhibited a specific ATPase activity at 55 °C (the ambient temperature of G. stearothermophilus) of 0.03–0.04 µmol phosphate/min/mg, which fits the range of 0.01 to 1.65 µmol phosphate/min/mg reported for NBDs of other ABC transporters [14]. The membrane-integral subunit, ArtM-His6, was overproduced in E. coli strain Rosetta2 (pVE15, pLacI) and solubilized from the membrane fraction by 1.2% DS. Purification was achieved by metal affinity chromatography in the presence of 0.05% DDM, similarly to
1252
V. Eckey et al. / Biochimica et Biophysica Acta 1798 (2010) 1250–1253
Fig. 1. SDS-PAGE of purified ArtP and ArtM subunits. ArtM-His6 and His6-ArtP were purified as described in Materials and methods. ArtM-His6 was subsequently treated with thrombin to cleave the hexahistidine tag. Lanes: 1, molecular mass standards; 2, ArtM; 3, His6-ArtP; 4, in vivo assembled Art(MP-His6)2.
ArtP (Fig. 1A, lane 2). Routinely, about 1.2 mg of purified protein was obtained from six liters of culture. To test the capability of the separately purified subunits to assemble a functional Art(MP)2 complex, tag-less ArtM was incubated with His6-ArtP at an equimolar ratio and subsequently subjected to metal affinity chromatography on a TALON matrix. Analysis of column fractions by SDS-PAGE revealed virtually no protein in the flowthrough or the wash fractions. ArtM co-eluted exclusively with His6ArtP, thereby suggesting that an association of the proteins had occurred (Fig. 2, panel A). Quantitation of His6-ArtP and ArtM from the peak fraction (Fig. 2, panel A, lane 8) by densitometric scanning revealed a ratio of His6-ArtP/ArtM of 1.7, which is in excellent agreement with that obtained from an in vivo assembled complex (1.6; Fig. 1, lane 4) (please note that the theoretical ratio of 1 is not found probably due to less efficient staining of hydrophobic ArtM). Subsequent analysis of concentrated ArtMP-containing fractions by gel filtration further demonstrated stable complex formation between both proteins (not shown). Furthermore, co-elution was neither affected by the addition of substrate-loaded binding protein (ArtJ/arginine), ATP or phospholipids to the assembly mixture (not shown). Next, we assayed ATPase activity of the co-eluted His6-ArtP/ArtM complex in proteoliposomes as a means to proof functional association. It is well established that binding protein-dependent ABC transporters, including the ArtJ-Art(MP)2 complex [9,10], exhibit low intrinsic ATPase
Fig. 2. Co-elution of ArtM and His6-ArtP variants from a Co2+ affinity matrix. Experiments were performed as described in Materials and methods. Lanes: 1, ArtM variant; 2, His6-ArtP variant; 3, molecular mass standards (25 and 20 kDa; see legend to Fig. 1); 4, assembly mixtures containing indicated ArtM and His6-ArtP variants; 5 and 6, wash fractions 1 and 10; 7–11, elution fractions.
Fig. 3. Time dependence of ATPase activity of in vivo- and in vitro- assembled Art(MP)2 wild-type complexes. Proteoliposomes containing Art(MP)2 were prepared in the presence (closed symbols) or absence (open symbols) of arginine-loaded ArtJ and assayed for ATPase activity as described in Materials and methods. Symbols: circles, Art (MP-His6)2 (in vivo); squares, ArtM + His6-ArtP; triangles, ArtM + His6-ArtP assembled in the presence of 2 mM ATP.
activity in proteoliposomes that is stimulated severalfold by the cognate substrate-loaded receptor [3]. This observation is generally taken as evidence for coupling of ATP hydrolysis to substrate translocation. As shown in Fig. 3, ATPase activity of the co-eluted complex was negligible in the absence of ArtJ/arginine but was stimulated severalfold in its presence, similar to in vivo-assembled Art(MP)2. Moreover, an 85% inhibition of the ATPase activity was observed in the presence of 0.5 mM ortho-vanadate, a specific inhibitor of ABC transporters [1] (not shown). Fig. 3 also shows that the presence of 2 mM ATP during the assembly reaction had no significant effect on ATPase activity. Together, these results clearly indicate that separately purified His6ArtP and ArtM are competent to assemble a fully functional transport complex in vitro. 3.2. Co-elution as a means to study the effect of mutations on complex assembly The above results prompted us to test the potential use of the established protocol as a tool to characterize the effect of mutations on the assembly process. To address this question, we chose residue Asn-87 from the Q loop of ArtP and Gly-127 from the EAA loop of ArtM for mutagenesis to proline and glycine, respectively. Residues from both motifs have been implicated in inter-domain signaling and/or physical interaction between subunits [4,5]. However, mutations affecting these interactions have been studied only in a few complete transporters, including the maltose transporter of E. coli/Salmonella [6–8,15] and the ferric hydroxamate transporter of E. coli [16]. The respective ArtP and ArtM variants were purified and analyzed for complex formation as established for the wild-type proteins. As shown in Fig. 2B, an ArtM/His6–ArtP(N87G) complex co-eluted with a P/M ratio of 1.7, very similar to wild type. Stability of the complex was confirmed by gel filtration (not shown). ATPase activity of separately purified His6-ArtP (N87G) was 0.03 µmol Pi/min/mg and thus in good agreement with that of wild-type His6-ArtP. However, ArtJ/arginine-stimulated ATPase activity of the assembled ArtM/His6–ArtP(N87G) complex in proteoliposomes was severely impaired (Fig. 4). These results indicate that the Q loop mutation is likely to affect signaling between ArtM and ArtP but not complex formation. Moreover, our finding is consistent with a positional change of Asn-87 upon ATP hydrolysis as observed in the crystal structures (compare pdb 3C41 and 2QOH). In contrast, the ArtM(G127P) variant displayed a strong defect in forming a complex with His6-ArtP as revealed by substoichiometric coelution from the TALON matrix (P/M ratio = 4.3; Fig. 2C). Further
V. Eckey et al. / Biochimica et Biophysica Acta 1798 (2010) 1250–1253
1253
Acknowledgment This work was supported by the Deutsche Forschungsgemeinschaft (SFB 449, TP B14). References
Fig. 4. ATPase activities of in vivo- and in vitro-assembled wild-type and mutant ArtMP complexes. Proteoliposomes were prepared as in legend to Fig. 3. Basal activities recorded in the absence of ArtJ/arginine were subtracted. SE mean (n ≥ 3).
ore, no complex formation was observed when both variants were combined (Fig. 2D). Consequently, no ATPase activity could be detected (Fig. 4).
4. Discussion In this communication, we demonstrate that ArtP and ArtM can be individually overproduced, purified, and functionally assembled to a complex displaying ArtJ/arginine-stimulated ATPase activity in proteoliposomes highly similar to in vivo-assembled Art(MP)2. Functional re-assembly after overproduction of intact complexes and subsequent dissociation of NBDs by treatment with urea was first reported for the histidine [17] and maltose [12,18] transporters. More recently, Horn et al. [19] demonstrated in vitro assembly of the glycine betaine transporter OpuA from purified subunits in detergent solution, while reconstitution of the LolCDE exporter involved in lipoprotein sorting to the outer membrane was studied in proteoliposomes [20]. In contrast to our study, the stability of the latter was highly dependent on phospholipids or ATP. Unlike these reports, we show here for the first time that in vitro assembly provides a means to study the role of individual amino acid residues on subunit–subunit interactions. Variants carrying any mutation can be assayed for assembly or signaling defects by co-elution and subsequent functional analysis in proteoliposomes. As proof of principle, we have confirmed the proposed roles of residues from the Q and EAA loops, respectively, of ArtP and ArtM. Moreover, the successful overexpression of ArtM and ArtP will be of further use for structural investigations of the Art (MP)2 transporter, e.g., by solid-state NMR. Such experiments are underway.
[1] E.B. Holland, S. Cole, K. Kuchler, C.F. Higgins (Eds.), ABC proteins: from bacteria to man, Elsevier, Amsterdam, 2003. [2] C.F. Higgins, ABC transporters: from microorganisms to man, Annu. Rev. Cell Biol. 8 (1992) 67–113. [3] E. Schneider, Import of solutes by ABC transporters—the maltose and other systems, in: E.B. Holland, S. Cole, K. Kuchler, C. Higgins (Eds.), ABC proteins: From bacteria to man, Elsevier, Amsterdam, 2003, pp. 157–185. [4] M.L. Oldham, A.L. Davidson, J. Chen, Structural insights into ABC transporter mechanism, Curr. Opin. Struct. Biol. 18 (2008) 726–733. [5] A.L. Davidson, E. Dassa, C. Orelle, J. Chen, Structure, function, and evolution of ATPbinding cassette systems, Microbiol. Mol. Biol. Rev. 72 (2008) 317–364. [6] M. Mourez, M. Hofnung, E. Dassa, Subunit interaction in ABC transporters. A conserved sequence in hydrophobic membrane proteins of periplasmic permeases define sites of interaction with the helical domain of ABC subunits, EMBO J. 16 (1997) 3066–3077. [7] S. Hunke, M. Mourez, M. Jéhanno, E. Dassa, E. Schneider, ATP modulates subunit– subunit interactions in an ATP-binding-cassette transporter (MalFGK2) determined by site-directed chemical cross-linking, J. Biol. Chem. 275 (2000) 15526–15534. [8] M.L. Daus, M. Grote, P. Müller, M. Doebber, A. Herrmann, H.-J. Steinhoff, E. Dassa, E. Schneider, ATP-driven MalK dimer closure and re-opening and conformational changes of the ‘EAA’ motifs are crucial for function of the maltose ATP-binding cassette transporter (MalFGK2), J. Biol. Chem. 282 (2007) 22387–22396. [9] R. Fleischer, A. Wengner, F. Scheffel, H. Landmesser, E. Schneider, Identification of a gene cluster encoding an arginine ATP-binding-cassette transporter in the genome of the thermophilic gram-positive bacterium Geobacillus stearothermophilus DSMZ 13240, Microbiology 151 (2005) 835–840. [10] A. Vahedi-Faridi, V. Eckey, F. Scheffel, C. Alings, H. Landmesser, E. Schneider, W. Saenger, Crystal structures and mutational analysis of the arginine-, lysine-, histidine-binding protein ArtJ from Geobacillus stearothermophilus. Implications for interactions of ArtJ with its cognate ATP-binding cassette transporter, Art (MP)2, J. Mol. Biol. 375 (2008) 448–459. [11] K.D. Tartoff, C.A. Hobbs, Improved media for growing plasmid and cosmid clones, Bethesda Research Labs Focus, vol. 9, 1987, p. 12. [12] H. Landmesser, A. Stein, B. Blüschke, M. Brinkmann, S. Hunke, E. Schneider, Largescale purification, dissociation and functional reassembly of the maltose ATP-binding cassette transporter (MalFGK2) of Salmonella typhimurium, Biochim. Biophys. Acta 1565 (2002) 64–72. [13] M.L. Daus, H. Landmesser, A. Schlosser, P. Müller, A. Herrmann, E. Schneider, ATP induces conformational changes of periplasmic loop regions of the maltose ATP-binding cassette transporter, J. Biol. Chem. 281 (2006) 3856–3865. [14] I.B. Holland, M.A. Blight, ABC-ATPases, adaptable energy generators fuelling transmembrane movement of a variety of molecules in organisms from bacteria to humans, J. Mol. Biol. 293 (1999) 381–399. [15] M.I. Tapia, M. Mourez, M. Hofnung, E. Dassa, Structure–function study of MalF protein by random mutagenesis, J. Bacteriol. 181 (1999) 2267–2272. [16] W. Köster, B. Böhm, Point mutations in 2 conserved glycine residues within the integral membrane protein FhuB affect iron(III)hydroxamate transport, Mol. Gen. Genet. 232 (1992) 399–407. [17] P.-Q. Liu, G.F.-L. Ames, In vitro disassembly and reassembly of an ABC transporter, the histidine permease, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 3495–3500. [18] S. Sharma, J.A. Davis, T. Ayvaz, B. Traxler, A.L. Davidson, Functional reassembly of the Escherichia coli maltose transporter following purification of a MalF-MalG subassembly, J. Bacteriol. 187 (2005) 2908–2911. [19] C. Horn, E. Bremer, L. Schmitt, Functional overexpression and in vitro re-association of OpuA, an osmotically regulated ABC-transport complex from Bacillus subtilis, FEBS Lett. 579 (2005) 5765–5768. [20] K. Kanamaru, N. Taniguchi, S. Miyamoto, S.-I. Narita, H. Tokuda, Complete reconstitution of an ATP-binding cassette transporter LolCDE complex from separately isolated subunits, FEBS J. 274 (2007) 3034–3043.