Exploiting Hydrophobicity for Efficient Production of Transmembrane

Subscriber access provided by FLORIDA ATLANTIC UNIV

Technical Note

Exploiting hydrophobicity for efficient production of transmembrane helices for structure determination by NMR spectroscopy Katrine Bugge, Helena Steinocher, Andrew J. Brooks, Kresten Lindorff-Larsen, and Birthe B. Kragelund Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b02365 • Publication Date (Web): 26 Aug 2015 Downloaded from http://pubs.acs.org on August 28, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Exploiting hydrophobicity for efficient production of transmembrane helices for structure determination by NMR spectroscopy Katrine Bugge1, Helena Steinocher1, Andrew J. Brooks2,3, Kresten Lindorff-Larsen1, Birthe B. Kragelund1* 1

Structural Biology and NMR Laboratory, Department of Biology, University of Copenhagen, Ole Maaløes Vej 5, DK-2200 Copenhagen N, Denmark. 2 The University of Queensland Diamantina Institute, The University of Queensland, Translational Research Institute, Qld 4072, Australia. 3 The University of Queensland, Institute for Molecular Bioscience, Qld 4072, Australia. ABSTRACT: Despite the biological and pharmaceutical significance of membrane proteins, their tertiary structures constitute less than 3% of known structures. One of the major obstacles for initiating structural studies of membrane proteins by NMR spectroscopy is the generation of high amounts of isotope-labeled protein. In this work we have exploited the hydrophobic nature of membrane proteins to develop a simple and efficient production scheme for isotope-labeled single-pass transmembrane domains (TMDs) with or without intrinsically disordered regions. We have evaluated the applicability and limitations of the strategy using seven membrane protein variants that differ in their overall hydrophobicity and length, and show a recovery for suitable variants of >70%. The developed production scheme is cost-efficient and easy to implement, and has the potential to facilitate an increase in the number of structures of single-pass TMDs, which are difficult to solve by other means. Membrane proteins represent close to one-third of the proteome1,2 and are responsible for vital processes through their actions as e.g. receptors, enzymes, transporters, and channels. Due to these key roles, they are associated with many pathologies and are targets for more than 50% of all pharmaceutical drugs.3 Despite their significance, the tertiary structures of membrane proteins remain understudied. Currently, integral membrane proteins constitute only 2.2% of the RCSB Protein Data Bank (PDB) repository4 and of these ~70% are of prokaryotic origin.5 This significant under-representation constitutes a critical bottleneck in the molecular understanding of many important biological systems, and predominantly originates from the methodological challenges associated with structural studies of membrane proteins.6 Significant progress in structure determination of membrane proteins by NMR spectroscopy has occurred within the last decade,5 but the growth in available structures remains hampered by three major obstacles. First, the establishment of a recombinant production protocol allowing for isotope labeling as well as a high yield is considered a major bottleneck.7 Secondly, their subsequent purification is complicated by the necessity of lipids or detergents to keep the protein in solution, which may interfere with conventional purification methods. Finally, because of the size-limitation of solution state NMR spectroscopy, the selection of a native-fold-supporting membrane mimetics is further restricted. The level of complexity

increase further when a membrane domain is coupled to a folded, water-soluble domain, not only increasing the total size of the complex, but also requiring conditions supporting solubility and correct folding of both domains. For these reasons, a divide-and-conquer approach is often evoked. Allthough such approaches may not always preserve the native structure and thus should be considered carefully, several studies have found that transmembrane domains (TMDs) may preserve fold and/or functionality even when separated from their adjacent soluble- or membrane-embedded domains.8-10⁠ Current production strategies for high yields of isotopelabeled membrane proteins typically rely on purification schemes originally developed for water-soluble proteins. Often purification is achieved by utilizing hexa-histidine tags, which may result in structural perturbations or introduce nonnative interactions with charged membrane mimetics. Alternatively, reverse-phase HPLC is attempted, which often results in low yields due to irreversible adsorption to the hydrophobic column material. In this work, we have exploited a characteristic feature of membrane proteins, their high hydrophobicity, to develop a simple, efficient, and non-tag based production scheme for single-pass transmembrane (TM) proteins. We adopted the divide-and-conquer approach to study the membrane-embedded domains either in isolation or with intrinsically disordered (ID) regions, which are often present in eukaryotic proteins. The core principle of the production strategy

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

is a simple two hour chloroform:methanol (CHCl3:MeOH) extraction, utilizing the characteristic hydrophobicity of the target. With this scheme, seven days and approximately 10 hours of active work is required to go from plasmid DNA to an NMR-ready sample (Figure 1, A). In addition, the protocol requires only simple remedies that are standard equipment in most laboratories. The production scheme has been successfully applied to obtain NMR samples of sufficient quality to solve the structure of the TMD of the human prolactin receptor (hPRLR) in DHPC micelles.11 In the present paper, we present the development and details of the strategy and demonstrate its general applicability by examining its potential and limitations on different eukaryotic single-pass TMD variants of low sequence identity. EXPERIMENTAL SECTION The following is a short version of the production scheme presented in Figure 1, A. A more detailed protocol is provided in the SI.

Page 2 of 7

The plasmid DNA was transformed into competent E. coli BL21(DE3) cells and expressed in either unlabeled, 15N, or 13 C-, 15N-labeled M9 media added 100 µg/ml ampicillin. The harvested cells were resuspended in lysis buffer (40 mL/L culture 25% (w/v) sucrose, 5 mM EDTA, 1xPBS buffer (pH 7.4), 1 mM PMSF), and sonicated on ice. Subsequently the IBs were harvested by centrifugation (20000g, 25 min, 4°C). This cycle of resuspension, sonication, and centrifugation was repeated three times. The resulting IBs were resuspended in 50 mM Tris-HCl buffer, harvested by centrifugation (20000g, 20 min, 4°C), solubilized in 12 mL (pr. L culture) 1.5% (w/v) sarkosyl, 100 mM DTT, 20 mM Tris-HCl buffer (pH 7.4) and incubated at room temperature (RT) with gentle agitation for 3 hours. Insoluble material was removed by centrifugation (12000g, 20 min, 4 °C). The supernatant was dialyzed in a 6-8 kDa MWCO dialysis tube at 4°C against 0.5% (w/v) sarkosyl, 10 mM NaCl, 50 mM Tris-HCl buffer (pH 7.4) until complete removal of DTT. To release the GST from the target sequence, 30-40 units of thrombin pr. mg fusion protein was added to the

Figure 1. Overview of the production scheme. A) Overview of steps from plasmid DNA to NMR-ready sample with the presented production scheme. Black bars represent active work time, while grey bars represent inactive time (bar size do not accurately scale with time). B) SDS-PAGE of samples from different steps in the production scheme. Top: EPOR-TMD purification. Note that EPOR-TMD stains poorly. Bottom: EPOR-TMD-ICD1 purification. LMW is “Low molecular weight ladder (Amersham)”, f. is “fraction”, 1 is “GST-hEPORTMD”, 2 is “GST”, 3 is “hEPOR-TMD”, 4 is “GST-hEPOR-TMD-ICD1”, and 5 is “hEPOR-TMD-ICD1”. C) Far-UV CD spectrum of hEPOR-TMD in 100 times molar excess of DPC, 50 mM NaCl, and 20 mM Na2HPO4/NaH2PO4 buffer (pH 7.2). D) Illustration of steps in the CHCl3:MeOH extraction protocol.

2


Page 3 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


dialyzed solution, followed by 24-48 hours of incubation at RT. When complete digestion had been confirmed by SDSPAGE, the solution was lyophilized, followed by resuspension in milliQ water (200 µL/mL of original solution). This solution was divided into batches of 50 µL, each of which was added to 750 µL of a 1:2 CHCl3:MeOH solution and mixed well. The solution was centrifuged (14000g, 2 min, 4°C), resulting in separation in three layers. The top aqueous layer was carefully removed. Subsequently, 500 µL of MeOH was added to the remaining solution, followed by thorough mixing. The mixture was incubated on ice for 20 min, followed by centrifugation (16000g, 40 min, 4°C). The supernatant containing the target protein was transferred to a glass vial, and the organic solvent evaporated under a stream of N2. RESULTS AND DISCUSSION The membrane proteins used in this work were the hPRLR, the human growth hormone receptor (hGHR), and the human erythropoietin receptor (hEPOR), which are all members of the homodimeric group 1 of the class I cytokine receptors.12 These are single-pass TM proteins (Figure S1), which share an overall similar topology consisting of a folded extracellular domain (ECD), an ID intracellular domain (ICD), and a nonconserved TMD connecting the ECD and ICD.12,13 The TMDs of hPRLR, hGHR, and hEPOR have sequence identities of less than 23% (EMBOSS Needle14,15), and are thus excellent representatives of diverse eukaryotic TMDs. To assess the potentials and limitations of the production strategy, three TMD variants of different lengths and grand average of hydropathy (GRAVY) scores16 were designed (Figure S2). For each receptor, a variant with the bioinformatically predicted TMD17 and 11-13 additional residues were designed as the

primary TMD variant (termed hPRLR-TMD, hEPOR-TMD, and hGHR-TMD). In addition, a TMD-ICD1 variant was constructed for the hEPOR and the hPRLR containing an additional 26 and 30 residues C-terminal to the TMD, respectively, of the ICDs, as well as a TMD-ICD2 variant of hGHR and hPRLR with respectively 62 and 70 additional ICD residues. The TMDs, the TMD-ICD1s, and the TMD-ICD2s shared less than 33% sequence identity (Table S1). The DNA sequences encoding the seven TMD-variants were cloned into pGEX-4T-1 vectors (Amersham, GE Healthcare) with the target sequences positioned downstream of a Glutathione S-transferase (GST)-carrier protein intervened by a thrombin cleavage site (Figure 2). The fusion of the target sequences with GST was essential for successful overexpression, as no expression could be detected if it was omitted (data not shown). The seven plasmids entered the production scheme (Figure 1, A), starting from transformation into E. coli BL21(DE3) cells followed by recombinant overexpression in M9-media. All expressions presented were done at 37 °C, but for some constructs, lowering of the expression temperature in combination with increased expression time enhanced the yield (data not shown). Analysis of the cell lysates by SDS-PAGE revealed that the fusion proteins were in the insoluble pellet (Figure 1, B), most likely as inclusion bodies (IBs). This is advantageous since the target protein typically accounts for 80-95% of the IB material18⁠, and they thus provide a first, crude purification step, as well as protection against degradation. The IBs were solubilized using DTT and the detergent sarkosyl, effectively bringing more than 90% of the fusion proteins into solution. Following thorough dialysis, the misfolded GST was released by thrombin digestion. The activity of thrombin appeared to be largely conserved in sar-

Figure 2. Design of expression constructs and primary structure of targets. Top: Schematic representation of the expressed fusion proteins resulting from cloning the target sequences into the pGEX-4T-1 vector. An N-terminal GST protein (light grey oval) was used as carrier protein during recombinant expression in E. coli, followed by a thrombin recognition sequence (LVPR-GS) for subsequent release. Thrombin digestion leaves a “Gly-Ser” sequence N-terminally in the target sequence (dark grey frame). Bottom: Target sequences cloned into the pGEX-4T-1 vector following the thrombin recognition sequence. The predicted TMDs are highlighted in bold letters; numbering is without signal peptides.

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 7

Figure 3. 1H,15N-HSQC spectra of five TMD-variants purified using the presented production strategy. From left to right: hEPORTMD, hPRLR-TMD, hGHR-TMD, hEPOR-TMD-ICD1, hPRLR-TMD-ICD1. The spectra were acquired at 37°C and pH 7.2, with all five proteins reconstituted in DHPC micelles.

kosyl, but the digestion efficiency varied (30-40 units of thrombin/mg fusion protein to obtain ~90% cleavage). The final purification step consisted of a simple two-hour CHCl3:MeOH extraction procedure (Figure 1, D), which was based on the widely applied CHCl3:MeOH precipitation procedure originally presented by Wessel and Flügge.19 The ability of some proteins to stay soluble in organic solvents has already been utilized in plant proteomics to prepare samples for mass spectrometry20, and as part of more gentle protein purification schemes. In approaches tailored for expression and purification of the subunit C of F1FO ATP synthase21 and EmrE,22,23 the proteins were purified from E. coli membranes by protocols consisting of organic extraction followed by chromatographic steps. Furthermore, organic extraction has been applied in a production scheme developed for prokaryote TMDs, where the overall strategy was to purify the proteins by Ni2+-NTA affinity chromatography, after which a CHCl3/MeOH/CH3COOH precipitation was utilized to remove a released tag-protein.24 The production scheme presented in the present paper takes a shorter and simpler route that omits

chromotographic steps and can be applied to eukaryotic TMDs with or without ID regions. Here, due to the robust nature of the targets and their expression into IBs where they are isolated from other membrane proteins, they are purified in a single CHCl3:MeOH extraction step. The classical CHCl3:MeOH extraction protocol has been optimized in terms of solvent ratios, centrifugation speed, and incubation times to efficiently keep the target TMD-variants in solution while all other proteins precipitated (Figure S3). During the procedure, the mixtures were phase separated into a top H2O/MeOH layer, a middle layer of precipitated constituents, and a bottom CHCl3 layer. This enabled further isolation of the target from the majority of nonprecipitated, but mainly water-soluble molecules found in the H2O/MeOH layer, including sarkosyl. Finally, the organic solvent containing the target protein was separated from the pellet, after which it was evaporated under a stream of N2. The yield and purity of the seven target proteins were evaluated by SDS-PAGE, a 2D-Quant kit (GE Healthcare, Piscataway, NJ), and NMR spectroscopy. Due to low expression,

Table 1. Physical properties of the TMD variants and obtained yields. Summary of GRAVY scores,16 number of residues, fusion protein yields from 1 L cultures, target protein yields, and the recoveries of the seven TMD variants subjected to the production scheme. Recovery is the percentage of target protein recollected after the CHCl3:MeOH extraction.

hEPOR-TMD hPRLR-TMD hGHR-TMD hPRLR-TMD-ICD1 hEPOR-TMD-ICD1 hPRLR-TMD-ICD2 hGHR-TMD-ICD2

GRAVY score

No. of residues

Yield of fusion protein [nmol]

Yield of target protein [nmol]

1.55 1.49 1.01 0.76 0.67 0.29 0.28

38 37 38 67 63 97 99

750 +/-140 1280 +/-140 420 +/-30 360 +/-100 540 +/-40 580 +/-70 N.A.

~800¤ 1200 +/-90 440 +/-30 260 +/-3 400 +/-50 120* +/-1 N.A.

Recovery >90%¤ >90% >90% 70% 70% 20%* N.A.

N.A.: Estimation not available. *: No 1H,15N-HSQC spectrum was recorded. ¤ : Quantification estimated by 1H,15N-HSQC peak volumes, see details in SI.

4


Page 5 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


yield estimations were not conducted for the hGHR-TMDICD2. The resulting estimated yields of the remaining six variants from 1 L cultures are presented in Table 1, and the 1 H,15N-HSQC NMR fingerprint spectra of five of the variants in DHPC micelles are shown in Figure 3. Due to low yield, no 1 H,15N-HSQC spectrum was recorded of hPRLR-TMD-ICD2. A high recovery of the target fusion protein from expression to lyophilization was obtained for all variants, but the recovery success after the CHCl3:MeOH extraction varied. For all TMD variants, a high recovery above 90% was obtained after extraction, while the recovery of the TMD-ICD1 variants was reasonable at approximately 70% (Table 1). For unknown reasons the 2D-Quant Kit did not provide meaningful results for the hEPOR-TMD. Furthermore, as the hEPOR-TMD has no Trp and Tyr residues the protein was not detectable with UV absorption at 280 nm. From comparisons of peak volumes in the 1 H,15N-HSQC spectra of the hEPOR-TMD-ICD1 and hEPORTMD, an ~800 nmol yield was estimated for hEPOR-TMD (details in SI), suggesting a similar recovery percentage as for the other TMD variants. For the TMD-ICD2s, the CHCl3:MeOH extraction was not as effective providing a low yield of 20% or less (Table 1). Since the basic principle of the purification step is precipitation by organic solvent of all proteins except the target protein, a certain level of distinct tolerance to organic solvent of the target protein is needed. Thus, from the poor recoveries of the TMD-ICD2s, it appears that a GRAVY score below 0.3 renders the CHCl3:MeOH extraction ineffective, and we therefore recommend a GRAVY score of

at least 0.5 to ensure a reasonable yield. The assessments of the developed CHCl3:MeOH extraction strategy showed that for proteins of similar length and hydrophobicity as the TMD and TMD-ICD1 variants, the target membrane protein can be extracted from a complex protein mixture with high recovery. Since recoveries are high from expression to purified target, this also implies that the best (and for some proteins, only) way to increase yields further is to improve the initial expression levels. As a higher amount of MeOH may be more suitable for proteins of a less hydrophobic nature, yields could possibly be increased by changing the ratio of the CHCl3:MeOH, which here was kept fixed at 1:2. The yields obtained also suggested a higher recovery of shorter proteins (Table 1), since the recovery of hGHR-TMD was higher than expected judged from the GRAVY score alone. The applicability of the organic solvent extraction to larger proteins with a high GRAVY score was not evaluated, but it is likely feasible given earlier solubilizations of multi-pass membrane proteins in organic solvents.21,22 It should, however, be noted that refolding of multi-pass membrane proteins from IBs may not be straightforward, and may result in nonnative or even aggregated structures. In such cases, the previously described and more gentle approaches involving purification from E. coli membranes may be better alternatives.21,22,23 We next evaluated the applicability of the strategy to produce proteins suitable for structural studies by assessing the produced proteins’ ability to adopt their predicted secondary structure and provide high quality NMR data. Using hEPOR-

Figure 4. Screening of membrane mimetics for hEPOR-TMD using 1H,15N-HSQC fingerprint spectra. Spectra of 0.5 mM hEPORTMD in 50 mM NaCl, 20 mM Na2HPO4/NaH2PO4 buffer (pH 7.2) and 10% D2O (black peaks) or 95% D2O (red peaks) reconstituted in different detergents or lipids. A) 80 mM DPC micelles, B) 80 mM SDS micelles, C) 20 mM sarkosyl micelles, D) 10% (w/v) POPC/POPS/DHPC bicelles (q=0.3) (only in 10% D2O), E) 200 mM LPPG micelles (only in 10% D2O), F) 200 mM LMPG micelles, G) 200 mM DDM micelles, H) 70 mM DHPC micelles. 5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TMD as example, the far-UV CD spectrum in DPC micelles (Figure 1, C) showed the characteristic local minima at 208 and 222 nm indicative of α-helical structure. In addition, a screening procedure for selection of an appropriate membrane mimetics for solution state NMR studies was conducted. The dried hEPOR-TMD was solubilized in various solvents, and a set of 1H,15N-HSQC spectra were acquired at 37°C, with a protein concentration of ~0.5 mM, and 10% D2O (Figure 4, black peaks) or 95% D2O (Figure 4, red peaks). Although the majority of the solvents readily solubilized the TMD, a high degree of variability in the spectral quality was immediately apparent. The detergents DHPC, DPC, SDS, and sarkosyl resulted in 1H,15N-HSQC spectra of the highest quality (Figure 4, A, B, C, H), providing well-dispersed peaks with narrow line widths and more than 95% of the expected peaks visible. The 1H,15N-HSQC spectra acquired in 95% D2O (Figure 4, red peaks) revealed backbone amides that were protected from water exchange by either hydrogen bonding or hydrophobic shielding,25,26 and thereby gave additional organizational information on the TMD-solvent complex to be utilized in the selection process. Based on these results, we found that the hEPOR-TMD solubilized in DPC micelles was most suitable for structural studies with solution state NMR spectroscopy. Our procedure for protein production is, however, also suitable for structural studies with solid state NMR spectroscopy or crystallography, which are both methods that are not limited by size and therefore allow the study of larger protein-lipid complexes. In summary, we have shown that isotope-labeled TMDs of eukaryotic origin can be produced through a simple, costefficient, and effective production scheme by exploiting their characteristic hydrophobicity. The strategy provides high yields and purity, and is generally applicable to proteins of similar length and hydrophobicity as the TMD and TMDICD1 variants. The tested TMD variants were in this work all of eukaryotic origin, but we expect the strategy to be equally applicable to prokaryotic proteins. Since the purification is based on a CHCl3:MeOH extraction the limitations on the strategy appear to be mainly dictated by the tolerance of the target protein to organic solvent. As production of high yields of pure and isotope-labeled protein is one of the major hurdles in structural biology of membrane proteins, this production strategy could pave the way for an increase in the number of structures of single-pass membrane proteins, which are difficult to solve by other means.

ASSOCIATED CONTENT Supporting Information Materials, experimental details, Figures S1-S3, and Table S1 as mentioned in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

Page 6 of 7

AUTHOR INFORMATION Corresponding Author *[email protected]

ACKNOWLEDGMENT This work was supported by the Lundbeck Foundation (to B.B.K) and the Novo Nordisk Foundation (to B.B.K). A.J.B. was supported by NHMRC grants (1084797, 1025088, 1025082). We thank Signe A. Sjørup for excellent technical assistance.

REFERENCES (1) Wallin, E.; von Heijne, G. Protein Sci. 1998, 7, 1029–1038. (2) Goffeau, A.; Barrell, B. G.; Bussey, H.; Davis, R. W.; Dujon, B.; Feldmann, H.; Galibert, F.; Hoheisel, J. D.; Jacq, C.; Johnston, M.; Louis, H.; Mewes, H. W.; Murakami, Y.; Philippsen, P.; Tettelin, H.; Oliver, S. G. Science 1996, 274, 546, 563–567. (3) Overington, J. P.; Al-Lazikani, B.; Hopkins, A. L. Nat. Rev. Drug Discov. 2006, 5, 993–996. (4) Tusnády, G. E.; Dosztányi, Z.; Simon, I. Bioinformatics 2004, 20, 2964–2972. (5) Kim, H.; Howell, S. Prog. Nucl. Magn. Reson. Spectrosc. 2009, 55, 335–360. (6) Rösner, H. I.; Kragelund, B. B. Compr. Physiol. 2012, 2, 1491– 1539. (7) Bill, R. M.; Henderson, P. J. F.; Iwata, S.; Kunji, E. R. S.; Michel, H.; Neutze, R.; Newstead, S.; Poolman, B.; Tate, C. G.; Vogel, H. Nat. Biotechnol. 2011, 29, 335–340. (8) Manni, S.; Mineev, K. S.; Usmanova, D.; Lyukmanova, E. N.; Shulepko, M. A.; Kirpichnikov, M. P.; Winter, J.; Matkovic, M.; Deupi, X.; Arseniev, A. S. Structure 2014, 22, 1–13. (9) Montal, M.; Oblatt-Montal, M.; Bühler, L. K.; Iwamoto, T.; Tomich, J. M. J. Biol. Chem. 1993, 268, 14601–14607. (10) Katragadda, M.; Alderfer, J. L.; Yeagle, P. L. Biophys. J. 2001, 81, 1029–1036. (11) Bugge, K.; Papeleo, E.; Haxholm, G. W.; Dagil, R.; Olsen, J. G.; Lindorff-Larsen, K.; Kragelund, B. B. in preparation (12) Liongue, C.; Ward, A. C. BMC Evol. Biol. 2007, 7. (13) Haxholm, G. W.; Nikolajsen, L. F.; Olsen, J. G.; Fredsted, J.; Larsen, F. H.; Goffin, V.; Pedersen, S. F.; Brooks, A. J.; Waters, M. J.; Kragelund, B. B. Biochem. J. 2015, 468, 495506. (14) Needleman, S. B.; Wunsch, C. D. J. Mol. Biol. 1970, 48, 443– 453. (15) Rice, P.; Longden, I.; Bleasby, A.; Trends Genet. 2000, 16, 276– 277. (16) Kyte, J.; Doolittle, R. F. J. Mol. Biol. 1982, 157, 105–132. (17) UniProt Consortium, Nucleic Acids Res. 2014, 43, D204–D212. (18) Baneyx, F.; Mujacic, M. Nat. Biotechnol. 2004, 22, 1399–1408. (19) Wessel, D.; Flügge, U. I. Anal. Biochem. 1984, 138, 141–143. (20) Vertommen, A.; Panis, B.; Swennen, R.; Carpentier, S. C. Planta 2010, 231, 1113–1125. (21) Fillingame, R. H. J. Biol. Chem. 1976, 251, 6630–6637. (22) Yerushalmi, H.; Lebendiker, M.; Schuldiner, S. J. Biol. Chem. 1995, 270, 6856–6863. (23) Schwaiger, M.; Lebendiker, M.; Yerushalmi, H.; Coles, M.; Gröger, A.; Schwarz, C.; Schuldiner, S.; Kessler, H. Eur. J. Biochem. 1998, 254, 610–618´9. (24) Hu, J.; Qin, H.; Li, C.; Sharma, M.; Cross, T. A.; Gao, F. P. Protein Sci. 2007, 16, 2153–2165. (25) Kragelund, B. B.; Knudsen, J.; Poulsen, F. M. J. Mol. Biol. 1995, 250, 695–706. (26) Spyroacopoulos, L.; O’Neil, J. D. J. Am. Chem. Soc. 1994, 116, 1395–1402.

6


Page 7 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Keywords: membrane proteins, NMR spectroscopy, purification, single-pass receptors, isotope-labeling.

7


Exploiting Hydrophobicity for Efficient Production of Transmembrane

Recommend Documents