Oriented Reconstitution of the Full-Length KcsA Potassium Channel in

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Letter pubs.acs.org/JPCL

Oriented Reconstitution of the Full-Length KcsA Potassium Channel in a Lipid Bilayer for AFM Imaging Ayumi Sumino,†,‡ Takayuki Uchihashi,§,∥ and Shigetoshi Oiki*,‡ †

PRESTO, Japan Science and Technology Agency (JST), 4-1-8 Honcho, Kawaguchi-shi, Saitama 332-0012, Japan Department of Molecular Physiology and Biophysics, Faculty of Medical Sciences, University of Fukui, 23-3 Matsuokashimoaizuki, Yoshida-gun, Fukui 910-1193, Japan § Department of Physics, Kanazawa University, Kakuma-machi, Kanazawa 920-1192, Japan ∥ Bio-AFM Frontier Research Center, Kanazawa 920-1192, Japan ‡

S Supporting Information *

ABSTRACT: Here, we have developed a method of oriented reconstitution of the KcsA potassium channel amenable to high-resolution AFM imaging. The solubilized full-length KcsA channels with histidine-tagged (His-tag) C-terminal ends were attached to a Ni2+-coated mica surface, and then detergent-destabilized liposomes were added to fill the interchannel space. AFM revealed that the membraneembedded KcsA channels were oriented with their extracellular faces upward, seen as a tetrameric square shape. This orientation was corroborated by the visible binding of a peptide scorpion toxin, agitoxin-2. To observe the cytoplasmic side of the channel, a His-tag was inserted into the extracellular loop, and the oppositely oriented channels provided wholly different images. In either orientation, the channels were individually dispersed at acidic pH, whereas they were self-assembled at neutral pH, indicating that the oriented channels are allowed to diffuse in the membrane. This method is readily applicable to membrane proteins in general for AFM imaging.

M

proteins, but the modifications cause surface roughness, which often deteriorates the high-resolution imaging. Here, we have developed an oriented reconstitution method for AFM imaging of the KcsA potassium channel.30 Previously, a channel-embedded membrane was prepared by (i) incorporating the channel proteins into the preformed membrane, (ii) extending channel-reconstituted liposomes onto a mica surface, or (iii) fusing channel-reconstituted liposomes to a preformed membrane. By contrast, here, solubilized KcsA channels histidine-tagged (His-tagged) at the C-terminal end were first attached to a Ni2+-modified mica surface, and the lipids were then provided from liposomes to fill the space between the channel molecules. AFM showed oriented KcsA channels embedded in the membrane with their extracellular side upward (the top view). To orient the channel oppositely and view the channel from the cytoplasmic side (the bottom view) faces a problem: the KcsA channel contains two transmembrane helices, and both the N- and C-termini are located on the cytoplasmic side. Accordingly, a His-tag was introduced into the extracellular loop, and the bottom view was thereby attained.

embrane proteins perform their function on the membrane, where various types of lipids surround the proteins in an inhomogeneous manner.1,2 To investigate these complicated systems, experiments under controlled lipid compositions are necessary, 3−5 and various membrane reconstitution methods incorporating purified membrane proteins into lipid bilayers of defined membrane compositions have been developed.6−10 However, the benefit of a certain controllability of the reconstitution methods, such as in liposomes and planar lipid bilayers, has the trade-off of the undesired random orientation of the reconstituted proteins. Oppositely oriented channels may interfere with the activities of the correctly oriented ones, which never occurs in native membranes. Thus, establishing methods for oriented reconstitution is crucially important, and various methods have been developed. For instance, (i) membrane proteins have been inserted into preformed liposomes,11−13 (ii) beads have been used as membrane protein adsorbents,14,15 and (iii) membrane proteins have been attached to a chemically modified surface.16−18 These reconstitution methods must be optimized for the relevant measurement techniques. To elucidate the structure and dynamics of membraneembedded proteins at the single-molecule level, AFM has been applied.19−29 Technically, membranes are formed on an atomistically flat mica surface for high-resolution recording. For oriented reconstitution, the mica surface was subjected to chemical modifications for the attachment of membrane © 2017 American Chemical Society

Received: December 29, 2016 Accepted: January 31, 2017 Published: January 31, 2017 785

DOI: 10.1021/acs.jpclett.6b03058 J. Phys. Chem. Lett. 2017, 8, 785−793

Letter

The Journal of Physical Chemistry Letters

Figure 1. Schematic illustration of the oriented reconstitution of the KcsA channel on a mica surface. (A) Freshly cleaved mica surface. (B) Ni2+ adsorption onto a mica surface. (C) Specific adsorption of the solubilized KcsA channel onto Ni2+-coated mica via the C-terminal His-tag. (D) Application of DDM-destabilized liposome onto the KcsA-immobilized surface. The destabilized bilayer fills the interspace between the KcsA channels. (E) Rinsing the liposomes and DDM. The KcsA channels are reconstituted into the membrane attached to the mica surface with their extracellular side up.

Figure 2. Specific absorption of the detergent-solubilized, His-tagged KcsA channel onto Ni2+-coated mica. (A) KcsA channels adsorbed onto Ni2+coated mica. (B) No channels were adsorbed onto the mica surface without Ni2+ coating. (C−E) Low pH (pH 4.0), imidazole (600 mM), and EDTA (100 mM) inhibited the specific adsorption of the His-tagged KcsA channels onto the Ni2+-coated mica. (F) Summary of the number of particles shown in the AFM images in (A−E). The KcsA concentration used was 1 μg/mL. Images were taken in 10 mM HEPES (pH 7.5) and 300 mM KCl. Scale bars: 400 nm.

Experimental Methods section), and optimization of the experimental conditions and the outcomes of each procedure are presented in subsequent sections. We used a KcsA channel His-tagged at the C-terminal end because the C-termini are exposed at one end of the homotetrameric channel37 and attachment of the channel through the C-termini rendered the channel uprightly oriented on a surface.38,39 First, a freshly cleaved mica surface was treated with NiCl2 solution (0.5 M) for 5 min and then rinsed with pure water. Second, the detergent (n-dodecyl-β-Dmaltoside; DDM)-solubilized KcsA channels (C-terminal Histagged) were attached to a Ni2+-coated mica surface. Third, to form a bilayer in the interchannel space, the lipids were added as detergent (DDM)-destabilized liposomes. After rinsing away the liposomes and detergents, a flat membrane with the oriented channels was formed. The channels could be detached from the surface by either acidifying the solution or adding a metal chelator. In the following sections, each step is evaluated using AFM imaging.

The KcsA potassium channel is a pH-dependent channel, undergoing opening conformational changes at acidic pH.31−34 Previously, we reported the supramolecular assembly and dispersion of the KcsA channel in the membrane, and clustering and dispersion occurred in a gating-dependent manner.35 In these previous studies, cytoplasmic domain-deleted channels reconstituted into liposomes were ruptured and adsorbed onto a mica surface.36 In the present study, the full-length KcsA channels were embedded in the membrane in either orientation, and AFM produced distinct top and bottom views of the KcsA channel. The images of clustering and dispersion under different pH conditions indicate that the oriented channels are allowed to diffuse in the membrane. These results showed that this simple oriented reconstitution method is readily applicable for other membrane proteins in general. We developed a simple oriented reconstitution method for the KcsA potassium channel for AFM imaging. The overall procedures are presented here (Figure 1; see details in the 786


Letter


Figure 3. Formation of a supported lipid bilayer and reconstitution of the KcsA channel using DDM-destabilized liposomes. (A) Light scattering of lipid solutions. Asolectin was dispersed in an aqueous solution (50 μg/mL), and absorbance at 230 nm was measured at different DDM concentrations. Arrows a−c indicate the plots with 0.0, 0.12, and 1.2 mM of DDM. (B−D) AFM images of the surface-attached KcsA channel after treatment with DDM-destabilized liposomes. The DDM concentration increased from 0.0, 0.12, and 1.2 mM in B−D, respectively. Height profiles from the mica surface along white lines are shown below the AFM images. The KcsA concentration was 1 μg/mL. Images were taken in 10 mM HEPES (pH7.5) and 300 mM KCl. Scale bars: 400 nm. (E−G) Schematic illustrations of plausible treatment processes with DDM-destabilized liposomes at different DDM concentrations relevant to panels B−D. Without DDM, the bilayer islands did not grow to fill the surface (E). DDMdestabilized liposomes nearly filled the interchannel space (F). At high DDM concentrations, micelles were associated with the attached channels but did not fill the interchannel space (G).

The first attachment step of the KcsA channel is critical for oriented reconstitution, and the specificity of KcsA channel adsorption onto mica was evaluated by AFM. Figure 2A shows an AFM image of the solubilized KcsA-attached surface. Many particles of similar size were dispersed randomly on the surface. The height distribution of the particles in Figure 2A shows a single peak with a height of 6.1 ± 1.0 nm (n = 807) (Figure S1). The histogram of the equivalent disk diameter of these particles also showed a single peak (d = 13.6 ± 4.6 nm (n = 807), Figure S1), suggesting that the channels are dispersed individually. The diameter was larger than expected, probably because of the absorbed DDM molecules on the hydrophobic surface of the channel (Figure 1C) and the tip convolution effect of the AFM tip. When the same solubilized KcsA was applied onto bare mica without Ni2+-coating, no particles were observed (Figure 2B),

indicating that the KcsA channels seen in Figure 2A were specifically adsorbed onto the mica surface via interactions between the His-tag and Ni2+. Furthermore, the addition of inhibitors of the His-tag−Ni 2+ interaction (acidic pH, imidazole, and EDTA for Figure 2C−E, respectively) in the KcsA-containing solution also showed remarkable inhibition of the adsorption. When the ionic strength of the solution was decreased, nonspecific binding increased. The density of the adsorbed particles in these experimental conditions is summarized in Figure 2F. The surface density of the channels is readily controlled by changing the concentration of the KcsA channel in the solution (Figure S2). When a bare coverglass or nitrilotriacetic acid (NTA)modified mica was used, highly specific adsorption of the Histagged channel could not be observed (Figure S3), and considerable amounts of nonspecific adsorption were observed. 787


Letter

The Journal of Physical Chemistry Letters The surface structure of the bare mica has been examined and is suitable for holding Ni2+;40−43 thus, the Ni2+-treated mica surface is optimal for the specific interaction with His-tagged molecules. Lipids were added to fill the space between the KcsA channels adsorbed onto the mica surface. When liposomes without DDM were added, the bilayer membrane formed islands, which did not grow to cover the entire surface (Figure 3B,E). Most of the adsorbed channels were left isolated without being embedded in the membrane. To fill the space, we used DDM-destabilized liposomes, and the optimal concentration of DDM for the membrane filling was examined. The DDM-destabilized liposomes were formed with a certain concentration of DDM added to the lipid solution at the liposome preparation step, and the size of the liposomes or micelles was qualitatively evaluated by light scattering (see the Experimental Methods section). The critical micelle concentration (CMC) of DDM is 0.17 mM, and the liposomes were intact below 0.08 mM DDM (Figure 3A). At 0.12 mM DDM, liposomes still exist, called destabilized liposomes. When 0.12 mM DDM-destabilized liposomes was used (Figure 3C), 90% of the surface was covered by the lipid bilayer, and the oriented channels were successfully reconstituted in the membrane. The height profile shows a constant thickness of ∼4 nm with numerous protrusions from the membrane surface (Figure 3F). When the DDM concentration was increased to 1.2 mM, many small disks were formed (Figure 3D,G), but the bilayer did not cover the entire surface. The bilayer coverage of these samples is summarized in Figure S4. Regardless of the channel density, the bilayers filled the interchannel space (Figure S2). Figure 4A shows time-lapse AFM images of the bilayer filling process (see also video). Initially (−14 s), the KcsA channels were adsorbed on the mica surface. After addition of the DDMdestabilized liposomes (0 s), bilayer gradually filled the interchannel space (13−668 s). Gaps between the adsorbed channels (dark brown) were completely filled at 668 s and are seen as a flat surface. The time course of the percent coverage is shown in Figure S5. Finally, the DDM and liposomes were washed out, and membrane defects (seen as dark brown) appeared (13−498 s) (Figure 4B). The size of the defects did not increase at 13 s from the start of the wash, indicating that DDM removal from the membrane was completed immediately after the onset of the wash. The reconstituted channels with their extracellular side up were observed (top view) at neutral pH, when the channels were in the closed state (Figure 5A). The channels were clustered, and several individual channels with a square shape were resolved within each cluster (Figure 5E). Similar clusters were observed after treatment with EDTA, which chelates Ni2+ and detaches the channel from the mica surface. These results indicate that even without the chelator, the randomly dispersed channels were diffusible and joined the neighboring channels to form clusters in the membrane. The KcsA channels in the open state were observed by changing the solution pH to acidic (Figure 5B). The channels were completely dispersed to individual channels. The highmagnification image indicates a square shape with a central depression (Figure 5F inset). The height profile along a green line in Figure 5B indicates that the channels protruded from the membrane only slightly. This result is in contrast to the profile at neutral pH, in which 4 nm protrusions are observed. Figure 6 shows the distribution of the height from the mica surface, which represents the longitudinal length of the channel. The

Figure 4. Time-lapse snapshots of the addition of DDM-destabilized liposomes (A) and rinsing out the DDM (B). (A) To the KcsAadsorbed surface (−14 s), the destabilized liposomes were added at 0 s, and the interchannel space was filled over time (13−668 s). (B) DDM was washed out from the DDM-containing bilayer (−26 s) by replacement with DDM-free buffer at 0 s, and membrane defects appeared due to the DDM removal. The images were taken in 10 mM HEPES (pH 7.5), 200 mM KCl (−13 s of A, and 7−730 s of B), and 10 mM HEPES (pH 7.5), 200 mM KCl, 50 μg/mL asolectin, and 0.12 mM DDM (13−668 s of A and −26 s of B). The KcsA concentration was 1 μg/mL. Scale bars: 400 nm.

longitudinal length of the clustered channels was 9 nm, whereas the length of the dispersed channels was 4.5 nm. The substantial shortening of the longitudinal length of the fulllength KcsA channels implies that, in addition to slight shortening of the transmembrane domain,36 the cytoplasmic domain dramatically changed in conformation at acidic pH. To confirm the extracellular side up orientation, we used a scorpion toxin,44 agitoxin-2 (AgTx2).45,46 AgTx2 is a peptide toxin of 38 amino acid residues, and its conformation was determined by NMR.47,48 AgTx2 docks on the extracellular surface of various types of potassium channels, and for the 788


Letter


Figure 5. Membrane-embedded KcsA channels with extracellular (A, B, E, F) and intracellular (C, D, G. H) side up orientations at neutral and acidic pH. Height profiles along green solid lines are shown below the AFM images. The KcsA concentration was 10 μg/mL. The phospholipid used was DOPC. Images were taken at pH 7.5 (10 mM HEPES, 200 mM KCl) for (A, E, C, G) and at pH 4.0 (10 mM succinic acid, 200 mM KCl) for (B, D, F, H). Scale bars: 400 nm for (A−D), 50 nm for (E−H), and 3 nm for (inset in F).

and fixing the His-tag there causes the channel to stand on a surface in the upright orientation.38 The His-tag without a linker was introduced between residues 54 and 55 (see the Experimenta Methods section for details). The same reconstitution procedure was conducted as that for the Cterminal His-tagged channels. The channels were successfully attached to the mica surface. An overall view of the AFM image for the intracellular side up channels (bottom view) differed substantially from the top view. At neutral pH (Figure 5C,G), the KcsA channels were clustered, and each channel was not resolved as a square shape. The height profile shows a variable length of protrusion from the membrane, with an average longitudinal length of 7 nm (Figure 6). At acidic pH (Figure 5D,H), the membrane surface was coarse and noisy, suggesting that channels diffuse rapidly in the membrane surface. Each individual channel, if captured, had an ambiguous contour, and the height profile was dissipated. These results suggest that the cytoplasmic domain is fluffy and the channels diffuse rapidly in the membrane. In this study, with the oriented reconstitution of the fulllength KcsA channels, the individually dispersed channels were clearly resolved at acidic pH (Figure 5F), where Ni2+−His-tag interactions are abolished. Moreover, clustering occurred at neutral pH, and several individual channels were resolved in a cluster. Thus, we conclude that the transitions between clustering and dispersion occurred for the full-length KcsA channels in either orientation. The pH dependency of the clustering−dispersion was examined by counting the number of isolated channels in a defined area. Compared to fully isolated channels at pH 4 (Figure 5F), no individual channels were resolved at pH 7

Figure 6. Height distribution of membrane-embedded KcsA channels. The height was measured from the mica surface. Extracellular side up channels (top view) at neutral (black circle) and acidic (open circle) pH and intracellular side up channels (bottom view) at neutral pH (gray circle) are shown.

KcsA channel, a few residues were mutated for tight binding (see the Experimental Methods section).46 The top view of the KcsA channels in the presence of AgTx2 (200 nM) is shown in Figure 7. A round-shaped bulky substance was bound at the center of the square-shaped channel. These images are consistent with the predicted structure of the AgTx2-bound KcsA channel, explicitly confirming the extracellular side up orientation. To directly observe the cytoplasmic side of the channel, the channels need to be oppositely oriented. However, the KcsA channel has two transmembrane helices, and consequently, both the N- and C-terminals are located on the cytoplasmic side. Thus, the His-tag was inserted into the extracellular loop. Our previous studies demonstrated that the tip of the extracellular loop is most exposed at the molecular surface,37 789


Letter


probability measured by the single-channel current recordings (pKa = 4.37).49 This result supports that the clustering− dispersion is intimately related to the gating status of the KcsA channel. We have established an oriented reconstitution method suitable for high-resolution AFM imaging, and the top and bottom views of the full-length KcsA channel were revealed. Rather than incorporating the channels into the preformed membrane, first the solubilized channels were attached to the Ni2+-coated mica surface, and then the membrane was filled between the interchannel space. There are critical factors in this method. First, treatment of the fresh mica surface with Ni2+ minimizes the roughness of the surface, allowing highresolution imaging. Earlier studies revealed that the flat mica surface is recessed slightly below the surface where Ni2+ sits, and the surface flatness is maintained.50 Second, the use of DDM-destabilized liposomes, in which the DDM concentration was optimized, allowed infilling of the space between the oriented channels and coverage of the surface. Destabilized liposomes have been used,51 but use of which for the space filling has not been reported. Third, for the oppositely oriented reconstitution, an issue specific to the KcsA channel, such that the N- and C-termini were located on the same side of the membrane, was circumvented by insertion of the His-tag into the extracellular loop. The AFM images of the oriented reconstitution channels provided an unprecedented dynamic structure of the KcsA channel in the membrane. In the top view, the tetrameric square shape was clearly resolved, and the extracellular side up orientation was confirmed by AgTx2 binding. The longitudinal length of the channel was 9 nm at neutral pH but was shortened to 4.5 nm at acidic pH. The clustered channels in the closed state may bear the mechanically rigid cytoplasmic domain by their own right or by lateral packing of the channels. The crystal structure of the closed KcsA channel at neutral pH (PDB code: 3EFF) shows that its longitudinal length is 11 nm.52 The difference in length between the AFM image (9 nm) and the crystal structure (11 nm) is likely due to the flexibility of the cytoplasmic domain. The square shape was less clear in the clustered channels than that in the dispersed channels, probably because of jostling of the channels in the cluster. At acidic pH, the longitudinal length of the AFM image (4.5 nm) suggests that the cytoplasmic domain is no longer bulky and the length is nearly equal to the length of the transmembrane domain. In the open structure, it is likely that the cytoplasmic domain is softly packed or unpacked and is pushed against the mica surface as the cantilever scans over the reconstituted channels. This flexibility of the cytoplasmic domain differs substantially from that of the bundled helices in the crystal structure in the open conformation (PDB code: 3PJS),53 confirming the concern that the cytoplasmic domain is bundled in the crystal structure by the C-terminal directed antibody and otherwise flexible. Enhanced flexibility of the cytoplasmic domain at acidic pH is also supported by the bottom view. The protrusion from the membrane surface is clearly resolved at neutral pH, and the shorter longitudinal length relative to that of the top view suggests greater freedom of motion for the cytoplasmic domain in the free space. At acidic pH, the noisy surface suggests that the cytoplasmic domain is highly flexible and also suggests the diffusion of the channel on the membrane. In this study, we found that the full-length KcsA channel in both orientations exhibited clustering dispersion at different pH

Figure 7. AFM images (A) and schematic illustration (B) of the KcsA channel in the extracellular side up orientation without and with AgTx2. The KcsA mutant (Q58A/T61S/R64D) was reconstituted into the DMPC bilayer. Images were taken in 10 mM HEPES (pH 7.5), 200 nM AgTx2, and 300 mM KCl. Scale bars: 3 nm.

(Figure 5E). The transitions between clustering and dispersion occurred within several minutes,35 and the number of isolated channels in 90000 nm2 were counted over 10 min after the solution exchange. At pH 4, the number of the isolated channel was ∼500, which decreased as the pH increased (Figure 8). This pH dependency, having a rough estimate of the midpoint at around pH 4.5, is similar to that for the pH-dependent open

Figure 8. pH dependency of the channel dispersion. The pH of the recording buffer was changed successively, and the number of isolated channels on a fixed area (90000 nm2) of the channel-reconstituted membrane was counted. Recording buffers (10 mM HEPES, 10 mM MES, 10 mM succinic acid with 200 mM KCl) were adjusted at different pHs (pH 7.5, 6.0, 5.3, 5.0, and 4.0). The KcsA channels were reconstituted in a DOPC bilayer with the extracellular surface up orientation. 790


Letter


constant was 0.1 N/m. All of the images were taken in buffer solution (details are in the figure caption). For time-lapse imaging, the buffer solution was replaced by simultaneous injection and suction, which was completed within 3 s. To image the AgTx2 binding, home-built high-speed AFM was used. The cantilever used was AC7 with an EBD tip, whose resonant frequency in water was approximately 700 kHz.

values. At neutral pH, the channels are clustered, whether or not the reconstituted membrane is treated with divalent cation chelator. This result indicates that even though the channels are tethered to the mica through the His-tag, the channels can diffuse on the membrane. Compared to our previous results for cytoplasmic domain-deleted channels, it is likely that the cytoplasmic domain does not interfere with the clustering− dispersion dynamics. The pH-dependent clustering was evaluated (Figure 8), which suggests intimate coupling between the gating conformational change and the clustering− dispersion dynamics. Future studies will elucidate the underlying mechanism. Throughout the procedures of the oriented reconstituted method, each procedure is simple. Thus, this method is readily applicable to membrane proteins, in general, embedded in various types of membranes for AFM imaging.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.6b03058. Histograms of the maximum height and the mean diameter of the KcsA channels adsorbed onto mica imaged in Figure 2A, AFM images indicating that the supported bilayer can form on the KcsA-adsorbed substrate with high KcsA density, AFM images of nonspecific adsorption on a bare coverglass and NTAmodified mica, effect of DDM concentration on the coverage of supported bilayers on the mica surface, time course of bilayer filling, and light scattering of destabilized liposome (PDF) Movie of membrane filling (AVI)

■

EXPERIMENTAL METHODS Materials. The gene for the KcsA channel with the hexa-His-tag at the C-terminal end or in the extracellular loop (between residues 54 and 55) was inserted in the pET29c vector and expressed in E. coli BL21(DE3)pLysS cells. For the AgTx2 binding experiment, triple mutations (Q57A/T61S/R64D) were introduced into the extracellular docking surface of the KcsA channel to allow high-affinity binding of AgTx2. Protein expression and purification were performed as previously described.37 Asolectin (SIGMA), 1,2-dioleoyl-sn-glycero-3phosphocholine (DOPC; Avanti Polar Lipids), and 1,2dimyristoyl-sn-glycero-3-phosphocholine (DMPC; Avanti Polar Lipids) were used as phospholipids, and n-dodecyl-β-Dmaltopyranoside (DDM) was used as the detergent. Preparation of DDM-Destabilized Liposome. Phospholipids were dissolved in chloroform and dried in vacuo for at least 2 h in a glass tube. The lipid film at the bottom was hydrated by buffer solution (10 mM HEPES [pH 7.5], 200 mM KCl) at room temperature to form a liposome solution. DDM was added (1.2 mM) to the liposome solution, and the liposomes were solubilized. This co-micellar solution of DDM and lipids was diluted with DDM-free buffer solution to a final DDM concentration of 0.12 mM, generating DDM-destabilized liposomes. The absorption spectra of the DDM-destabilized liposome solution were measured to check the light scattering of liposomes (Figure S6). Oriented Reconstitution of KcsA Channels on Mica. First, 0.5 M NiCl2 solution was placed on freshly cleaved mica for 5 min to coat the mica surface with Ni2+ ions, and then, the surface was rinsed with pure water. Second, 1−10 μg/mL of KcsA channel solution (10 mM HEPES [pH 7.5], 2 M KCl, 1.2 mM DDM, 1−6 mM imidazole) was deposited on the surface for 5 min to attach the His-tagged KcsA channels to the Ni2+-coated mica via a Ni2+−His-tag interaction. Third, the surface was rinsed with buffer solution (10 mM HEPES [pH 7.5], 2 M KCl, 1.2 mM DDM). Fourth, 50 μg/mL DDM-destabilized liposome suspended in buffer solution was deposited and incubated for ∼10 min to form a supported bilayer. Fifth, the remaining DDM-destabilized liposomes and DDM molecules in the supported bilayers were removed by rinsing the substrate with 3 mL of DDM-free buffer (10 mM HEPES [pH 7.5], 200−300 mM KCl). AFM Observation. AFM imaging was performed using Cypher (Asylum Research) with AC mode. The cantilever used was BL-AC40TS with a tip radius of ∼7 nm, the resonant frequency was 25 kHz in buffer solution, and the spring

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: +81-776-61-8306. ORCID

Shigetoshi Oiki: 0000-0002-8438-6750 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are grateful to Masayuki Iwamoto (Univ. Fukui) for fruitful discussions. A.S. thanks Mariko Yamatake (Univ. of Fukui) for experimental assistance. A.S. acknowledges PRESTO (JST), the Shiseido Female Researcher Science Grant, Grantin-Aid for Young Scientists (B) (26870233) and Challenging Exploratory Research (16K15178) for funding, and S.O. acknowledges JSPS KAKENHI (No. 26253014) and Grantin-Aid for Scientific Research on Innovative Areas (16H00759).

■

REFERENCES

(1) Nicolson, G. L. The Fluid - Mosaic Model of Membrane Structure: Still Relevant to Understanding the Structure, Function and Dynamics of Biological Membranes after More than 40 Years. Biochim. Biophys. Acta, Biomembr. 2014, 1838, 1451−1466. (2) Engelman, D. M. Membranes Are More Mosaic than Fluid. Nature 2005, 438, 578−580. (3) Lee, A. G. How Lipids Affect the Activities of Integral Membrane Proteins. Biochim. Biophys. Acta, Biomembr. 2004, 1666, 62−87. (4) Iwamoto, M.; Oiki, S. Amphipathic Antenna of an Inward Rectifier K+ Channel Responds to Changes in the Inner Membrane Leaflet. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 749−754. (5) Oiki, S. Channel Function Reconstitution and Re-Animation: A Single-Channel Strategy in the Postcrystal Age. J. Physiol. 2015, 593, 2553−2573. (6) Paternostre, M. T.; Roux, M.; Rigaud, J. L. Mechanisms of Membrane Protein Insertion into Liposomes during Reconstitution Procedures Involving the Use of Detergents. 1. Solubilization of Large Unilamellar Liposomes (Prepared by Reverse-Phase Evaporation) by Triton X-100, Octyl Glucoside, and Sodium. Biochemistry 1988, 27, 2668−2677.

791


Letter


(25) Mari, S. a; Pessoa, J.; Altieri, S.; Hensen, U.; Thomas, L.; Morais-Cabral, J. H.; Müller, D. J. Gating of the MlotiK1 Potassium Channel Involves Large Rearrangements of the Cyclic NucleotideBinding Domains. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 20802− 20807. (26) Milhiet, P.-E.; Gubellini, F.; Berquand, A.; Dosset, P.; Rigaud, J.L.; Le Grimellec, C.; Lévy, D. High-Resolution AFM of Membrane Proteins Directly Incorporated at High Density in Planar Lipid Bilayer. Biophys. J. 2006, 91, 3268−3275. (27) Muller, D. J. AFM: A Nanotool in Membrane Biology. Biochemistry 2008, 47, 7986−7998. (28) Sturgis, J. N.; Niederman, R. a. Atomic Force Microscopy Reveals Multiple Patterns of Antenna Organization in Purple Bacteria: Implications for Energy Transduction Mechanisms and Membrane Modeling. Photosynth. Res. 2008, 95, 269−278. (29) Scheuring, S.; Rigaud, J.-L.; Sturgis, J. N. Variable LH2 Stoichiometry and Core Clustering in Native Membranes of Rhodospirillum Photometricum. EMBO J. 2004, 23, 4127−4133. (30) Zhou, Y.; Morais-Cabral, J.; Kaufman, A.; MacKinnon, R. Chemistry of Ion Coordination and Hydration Revealed by a K+ Channel-Fab Complex at 2.0 Å Resolution. Nature 2001, 414, 43−48. (31) Thompson, A. N.; Posson, D. J.; Parsa, P. V.; Nimigean, C. M. Molecular Mechanism of pH Sensing in KcsA Potassium Channels. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 6900−6905. (32) Cuello, L. G.; Cortes, D. M.; Jogini, V.; Sompornpisut, A.; Perozo, E. A Molecular Mechanism for Proton-Dependent Gating in KcsA. FEBS Lett. 2010, 584, 1126−1132. (33) Cortes, D. M.; Cuello, L. G.; Perozo, E. Molecular Architecture of Full-Length KcsA: Role of Cytoplasmic Domains in Ion Permeation and Activation Gating. J. Gen. Physiol. 2001, 117, 165−180. (34) Hirano, M.; Takeuchi, Y.; Aoki, T.; Yanagida, T.; Ide, T. Rearrangements in the KcsA Cytoplasmic Domain Underlie Its Gating. J. Biol. Chem. 2010, 285, 3777−3783. (35) Sumino, A.; Yamamoto, D.; Iwamoto, M.; Dewa, T.; Oiki, S. Gating-Associated Clustering-Dispersion Dynamics of the KcsA Potassium Channel in a Lipid Membrane. J. Phys. Chem. Lett. 2014, 5, 578−584. (36) Sumino, A.; Sumikama, T.; Iwamoto, M.; Dewa, T.; Oiki, S. The Open Gate Structure of the Membrane-Embedded KcsA Potassium Channel Viewed From the Cytoplasmic Side. Sci. Rep. 2013, 3, 1063. (37) Iwamoto, M.; Shimizu, H.; Inoue, F.; Konno, T.; Sasaki, Y. C.; Oiki, S. Surface Structure and Its Dynamic Rearrangements of the KcsA Potassium Channel upon Gating and Tetrabutylammonium Blocking. J. Biol. Chem. 2006, 281, 28379−28386. (38) Shimizu, H.; Iwamoto, M.; Konno, T.; Nihei, A.; Sasaki, Y. C.; Oiki, S. Global Twisting Motion of Single Molecular KcsA Potassium Channel upon Gating. Cell 2008, 132, 67−78. (39) Yamakata, A.; Shimizu, H.; Oiki, S. Surface-Enhanced IR Absorption Spectroscopy of the KcsA Potassium Channel upon Application of an Electric Field. Phys. Chem. Chem. Phys. 2015, 17, 21104−21111. (40) Hansma, H. G.; Laney, D. E. DNA Binding to Mica Correlates with Cationic Radius: Assay by Atomic Force Microscopy. Biophys. J. 1996, 70, 1933−1939. (41) Hansma, H. G.; Laney, D. E.; Bezanilla, M.; Sinsheimer, R. L.; Hansma, P. K. Applications for Atomic Force Microscopy of DNA. Biophys. J. 1995, 68, 1672−1677. (42) Hsueh, C.; Chen, H.; Gimzewski, J. K.; Reed, J.; Abdel-Fattah, T. M. Localized Nanoscopic Surface Measurements of NickelModified Mica for Single-Molecule DNA Sequence Sampling. ACS Appl. Mater. Interfaces 2010, 2, 3249−3256. (43) Piétrement, O.; Pastré, D.; Fusil, S.; Jeusset, J.; David, M.-O.; Landousy, F.; Hamon, L.; Zozime, A.; Le Cam, E. Reversible Binding of DNA on NiCl 2 -Treated Mica by Varying the Ionic Strength. Langmuir 2003, 19, 2536−2539. (44) Miller, C.; Moczydlowski, E.; Latorre, R.; Phillips, M. Charybdotoxin, a Protein Inhibitor of Single Ca2+-Activated K+ Channels from Mammalian Skeletal Muscle. Nature 1985, 313, 316− 318.

(7) Rigaud, J.-L.; Lévy, D. Reconstitution of Membrane Proteins into Liposomes. Methods Enzymol. 2003, 372, 65−86. (8) Rigaud, J.-L.; Levy, D.; Mosser, G.; Lambert, O. Detergent Removal by Non-Polar Polystyrene Beads Applications to Membrane Protein Reconstitution and Two-Dimensional Crystallization. Eur. Biophys. J. 1998, 27, 305−319. (9) Rigaud, J. L.; Paternostre, M. T.; Bluzat, A. Bluzat, a. Mechanisms of Membrane Protein Insertion into Liposomes during Reconstitution Procedures Involving the Use of Detergents. 2. Incorporation of the Light-Driven Proton Pump Bacteriorhodopsin. Biochemistry 1988, 27, 2677−2688. (10) Knol, J.; Sjollema, K.; Poolman, B. Detergent-Mediated Reconstitution of Membrane Proteins. Biochemistry 1998, 37, 16410−16415. (11) Veenhoff, L.; Knol, J.; Liang, W. J.; Henderson, P. J. F.; Leblanc, G.; Poolman, B. Unidirectional Reconstitution into DetergentDestabilized Liposomes of the Purified Lactose Transport System of Streptococcus Thermophilus. J. Biol. Chem. 1996, 271, 15358−15366. (12) Yanagisawa, M.; Iwamoto, M.; Kato, A.; Yoshikawa, K.; Oiki, S. Oriented Reconstitution of a Membrane Protein in a Giant Unilamellar Vesicle: Experimental Verification with the Potassium Channel KcsA. J. Am. Chem. Soc. 2011, 133, 11774−11779. (13) Dezi, M.; Di Cicco, A.; Bassereau, P.; Lévy, D. DetergentMediated Incorporation of Transmembrane Proteins in Giant Unilamellar Vesicles with Controlled Physiological Contents. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 7276−7281. (14) Yokogawa, M.; Takeuchi, K.; Shimada, I. Bead-Linked Proteoliposomes: A Reconstitution Method for NMR Analyses of Membrane Protein-Ligand Interactions. J. Am. Chem. Soc. 2005, 127, 12021−12027. (15) Trépout, S.; Mornet, S.; Benabdelhak, H.; Ducruix, A.; Brisson, A. R.; Lambert, O. Membrane Protein Selectively Oriented on Solid Support and Reconstituted into a Lipid Membrane. Langmuir 2007, 23, 2647−2654. (16) Ataka, K.; Richter, B.; Heberle, J. Orientational Control of the Physiological Reaction of Cytochrome c Oxidase Tethered to a Gold Electrode. J. Phys. Chem. B 2006, 110, 9339−9347. (17) Giess, F.; Friedrich, M. G.; Heberle, J.; Naumann, R. L.; Knoll, W. The Protein-Tethered Lipid Bilayer: A Novel Mimic of the Biological Membrane. Biophys. J. 2004, 87, 3213−3220. (18) Ataka, K.; Stripp, S. T.; Heberle, J. Surface-Enhanced Infrared Absorption Spectroscopy (SEIRAS) to Probe Monolayers of Membrane Proteins. Biochim. Biophys. Acta, Biomembr. 2013, 1828, 2283−2293. (19) Scheuring, S.; Ringler, P.; Borgnia, M.; Stahlberg, H.; Müller, D. J.; Agre, P.; Engel, A. High Resolution AFM Topographs of the Escherichia Coli Water Channel Aquaporin Z. EMBO J. 1999, 18, 4981−4987. (20) Gonçalves, R. P.; Busselez, J.; Lévy, D.; Seguin, J.; Scheuring, S. Membrane Insertion of Rhodopseudomonas Acidophila Light Harvesting Complex 2 Investigated by High Resolution AFM. J. Struct. Biol. 2005, 149, 79−86. (21) Sumino, A.; Dewa, T.; Noji, T.; Nakano, Y.; Watanabe, N.; Hildner, R.; Bö sch, N.; Kö hler, J.; Nango, M. Influence of Phospholipid Composition on Self-Assembly and Energy-Transfer Efficiency in Networks of Light-Harvesting 2 Complexes. J. Phys. Chem. B 2013, 117, 10395−10404. (22) Casuso, I.; Khao, J.; Chami, M.; Paul-Gilloteaux, P.; Husain, M.; Duneau, J.-P.; Stahlberg, H.; Sturgis, J. N.; Scheuring, S. Characterization of the Motion of Membrane Proteins Using High-Speed Atomic Force Microscopy. Nat. Nanotechnol. 2012, 7, 525−529. (23) Kowal, J.; Chami, M.; Baumgartner, P.; Arheit, M.; Chiu, P.-L.; Rangl, M.; Scheuring, S.; Schröder, G. F.; Nimigean, C. M.; Stahlberg, H. Ligand-Induced Structural Changes in the Cyclic NucleotideModulated Potassium Channel MloK1. Nat. Commun. 2014, 5, 3106. (24) Grage, S. L.; Keleshian, A. M.; Turdzeladze, T.; Battle, A. R.; Tay, W. C.; May, R. P.; Holt, S. a; Contera, S. A.; Haertlein, M.; Moulin, M.; et al. Bilayer-Mediated Clustering and Functional Interaction of MscL Channels. Biophys. J. 2011, 100, 1252−1260. 792


Letter

The Journal of Physical Chemistry Letters (45) Garcia, M. L.; Garcia-calvo, M.; Hidalgo, L. P.; Lee, A.; Mackinnon, R. Purification and Characterization of Three Inhibitors of Voltage-Dependent K+ Channels from Leiurus Quinquest riatus var. Hebraeus Venom. Biochemistry 1994, 33, 6834−6839. (46) MacKinnon, R. Structural Conservation in Prokaryotic and Eukaryotic Potassium Channels. Science (Washington, DC, U. S.) 1998, 280, 106−109. (47) Takeuchi, K.; Yokogawa, M.; Matsuda, T.; Sugai, M.; Kawano, S.; Kohno, T.; Nakamura, H.; Takahashi, H.; Shimada, I. Structural Basis of the KcsA K+ Channel and Agitoxin2 Pore-Blocking Toxin Interaction by Using the Transferred Cross-Saturation Method. Structure 2003, 11, 1381−1392. (48) Lange, A.; Giller, K.; Hornig, S.; Martin-Eauclaire, M.-F.; Pongs, O.; Becker, S.; Baldus, M. Toxin-Induced Conformational Changes in a Potassium Channel Revealed by Solid-State NMR. Nature 2006, 440, 959−962. (49) Chakrapani, S.; Cordero-Morales, J. F.; Perozo, E. A Quantitative Description of KcsA Gating II: Single-Channel Currents. J. Gen. Physiol. 2007, 130, 479−496. (50) Heath, M. D.; Henderson, B.; Perkin, S. Ion-Specific Effects on the Interaction between Fibronectin and Negatively Charged Mica Surfaces. Langmuir 2010, 26, 5304−5308. (51) Lu, S.; Zgurskaya, H. I. MacA, a Periplasmic Membrane Fusion Protein of the Macrolide Transporter MacAB-TolC, Binds Lipopolysaccharide Core Specifically and with High Affinity. J. Bacteriol. 2013, 195, 4865−4872. (52) Uysal, S.; Vásquez, V.; Tereshko, V.; Esaki, K.; Fellouse, F.; Sidhu, S.; Koide, S.; Perozo, E.; Kossiakoff, A. Crystal Structure of Full-Length KcsA in Its Closed Conformation. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 6644−6649. (53) Uysal, S.; Cuello, L. G.; Cortes, D. M.; Koide, S.; Kossiakoff, A. A.; Perozo, E. Mechanism of Activation Gating in the Full-Length KcsA K+ Channel. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 11896− 11899.

793


Oriented Reconstitution of the Full-Length KcsA Potassium Channel in

Recommend Documents