Three-Dimensional Structure of a Human Connexin26 Gap Junction Channel Reveals a Plug in the Vestibule

Home , Connexin, Connexon, Gap junction, Membrane channel

Three-dimensional structure of a human connexin26 gap junction channel reveals a plug in the vestibule

Atsunori Oshima†‡, Kazutoshi Tani†, Yoko Hiroaki†‡, Yoshinori Fujiyoshi†‡§¶, and Gina E. Sosinsky¶ʈ

†Department of Biophysics, Faculty of Science, Kyoto University, Oiwake, Kitashirakawa, Sakyo-ku, Kyoto 606-8502, Japan; ʈNational Center for Microscopy and Imaging Research, Department of Neurosciences, University of California at San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0608; ‡Core Research for Evolution Science and Technology (CREST), Japan Science and Technology Agency (JST), Oiwake, Kitashirakawa, Sakyo-ku, Kyoto 606-8502, Japan; and §Japan Biological Information Research Center (JBIRC), National Institute of Advanced Industrial Science and Technology (AIST), 2-41-6, Aomi, Koto-ku, Tokyo 135-0064, Japan

Communicated by David J. DeRosier, Brandeis University, Waltham, MA, April 24, 2007 (received for review January 29, 2007) Connexin molecules form intercellular membrane channels facili- that the relatively unordered carboxyl terminus of larger iso- tating electronic coupling and the passage of small molecules forms interferes with the necessary tight packing in a crystal. In between adjoining cells. Connexin26 (Cx26) is the second smallest addition, we used a site-specific mutant of human Met-34, member of the gap junction protein family, and mutations in Cx26 hCx26M34A, because this mutant expresses in baculovirus in- cause certain hereditary human diseases such as skin disorders and fected Sf9 cells at higher quantities than wild-type Cx26-infected hearing loss. Here, we report the electron crystallographic struc- cells. The hCx26M34A mutant is a single-site mutation at the ture of a human Cx26 mutant (M34A). Although crystallization same position as the hCx26M34T mutant, which can cause trials used hemichannel preparations, the density map revealed prelingual nonsyndromic hereditary deafness (11). Although not that two hemichannels redocked at their extracellular surfaces into as well characterized as hCx26M34T (12), this single-point full intercellular channels. These orthorhombic crystals contained mutation of a Met to Ala decreases dye coupling in exogenously two sets of symmetry-related intercellular channels within three transfected HeLa cells and forms structures indistinguishable lipid bilayers. The 3D map shows a prominent density in the pore from wild-type gap junctions (13). We succeeded in making 2D of each hemichannel. This density contacts the innermost helices of the surrounding connexin subunits at the bottom of the vestibule. crystals of the hCx26M34A suitable for cryoelectron crystallog- The density map suggests that physical blocking may play an raphy. The 3D structure of hCx26M34A gap junction channels important role that underlies gap junction channel regulation. Our reveals a prominent density in the pore of each hemichannel, structure allows us to suggest that the two docked hemichannels suggesting that the channel is blocked by a physical obstruction. can be independent and may regulate their activity autonomously with a plug in the vestibule. Results Two-Dimensional Crystallization of Connexin26 Complexes. connexin channels ͉ electron crystallography ͉ intercellular hCx26M34A gap junction channels were expressed in and iso- communication ͉ membrane protein structure ͉ two-dimensional crystals lated from Sf9 insect cells [see supporting information (SI) Fig. 7A]. Hemichannels (connexons) were isolated by affinity puri- ap junctions contain intercellular communication channels fication using a C-terminal hexa-histidine tag (SI Fig. 7B). Gthat allow a wide variety of solutes with different sizes to be Purified hemichannels were mixed with the lipid dioleoylphos- transferred between the cytoplasm of adjacent cells. These phatidylcholine (DOPC) at a lipid-to-protein ratio of 1 (wt/wt). solutes include ions, metabolites, nucleotides, peptides, and Reconstitution into lipid bilayers by dialysis produced 2D crys- secondary messengers. Gap junction channels have critical roles tals Ͼ1 ␮m in diameter (SI Fig. 8A). Although the purified in many biologically important processes including cardiac de- hemichannel is hexameric, the 2D arrays obtained by dialysis velopment, fertility, the immune system, and electrical signaling showed an orthorhombic crystal lattice (SI Fig. 8B). in the nervous system (1). The diversely expressed connexin26 We recorded images of the hCx26M34A crystals embedded in (Cx26) is the second-smallest member of the conserved mam- 0.05–1% tannic acid, 2–40% trehalose, or the combination of malian gap junction protein family. Hereditary mutations in them. Computed diffraction patterns of one of the best 0° images human Cx26 cause macroscopic symptoms such as certain skin showed reflections to a resolution of Ϸ11Å(SI Fig. 8C). After disorders and nonsyndromic and syndromic deafness (2). Early electron microscopic studies suggested that closure of image processing to correct crystal distortions, the resolution the gap junction channel occurs by rotating all six subunits in improved to 7 Å (SI Fig. 8D). Image processing revealed that the ϭ each hemichannel (3, 4). More recently, a 3D structure of a crystals had p22121 symmetry and unit cell parameters of a ϭ ␥ ϭ truncated form of Cx43 reported on the arrangement and the 112.4 Å, b 111.2 Å, and 90°. assignments of the four transmembrane ␣-helical bundle in each connexin (5, 6). Electrophysiological experiments have shown Author contributions: A.O. and Y.F. designed research; A.O., K.T., Y.H., Y.F., and G.E.S. that the voltage sensor involves charged residues in the connexin performed research; A.O., K.T., Y.H., Y.F., and G.E.S. analyzed data; and A.O., Y.F., and N terminus and that it appears to face the aqueous pore (7, 8), G.E.S. wrote the paper. and studies of hemichannel conductivity favor ‘‘an individual The authors declare no conﬂict of interest. subunit gating model’’ (9). In addition, the selectivity of gap Freely available online through the PNAS open access option. junctions for permeation of small molecules under Ϸ1 kDa (also Abbreviations: Cx: connexin, Cx26M34A: connexin26 site-speciﬁc mutant Met34Ala. known as ‘‘permselectivity’’) depends on the connexin isoform. Data deposition: The cryoEM structure reported in this paper has been deposited in the Each isoform has unique physiological responses to ions, phos- Macromolecular Structure Database (MSD), www.ebi.ac.uk/msd-srv/emsearch/index.html phorylation, and pH. Permselectivity is hypothesized to occur (accession no. EMD-1341). independently from voltage gating, implying that gap junction ¶To whom correspondence may be addressed. E-mail: [email protected] or yoshi@ channels possess multiple gating mechanisms (10). em.biophys.kyoto-u.ac.jp. In this study, we focus on the structure of Cx26 gap junction This article contains supporting information online at www.pnas.org/cgi/content/full/ channels, because the short cytoplasmic tail of Cx26 makes it 0703704104/DC1. more amenable for the formation of 2D crystals. It is believed © 2007 by The National Academy of Sciences of the USA

10034–10039 ͉ PNAS ͉ June 12, 2007 ͉ vol. 104 ͉ no. 24 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0703704104 Downloaded by guest on September 28, 2021 BIOPHYSICS

Fig. 1. Three-dimensional structure of Cx26 orthorhombic crystals.(A) Molecular packing of Cx26 in the 2D crystal. The gap junction channels are incorporated in three lipid bilayers (Mem-1–Mem-3) with 21 symmetry along Mem-2. (B) View of the Cx26 density map perpendicular to the membrane plane. The three membranes, indicated by gray bars, surround two extracellular gap regions. The map is contoured at 1.0␴ (light blue) and 2.4␴ (wheat color) above the mean density. The transmembrane ␣-helical ribbon model (6) is docked into the density for one of the hemichannels. The four helices are color-coded as in Fig. 2. Two helices D make contact with adjacent gap junction channels (red arrows). (Scale bar, 40 Å.) (C and D) Forty-angstrom-thick sections through the density map parallel to the membrane plane, showing protein embedded in membranes Mem-1 (identical to Mem-3) (C) and Mem-2 (D). Tail ends of two helices D are indicated by red arrows as in B. (Scale bars, 40 Å.)

Three-Dimensional Structure Determination and Organization of that the crystals have a thickness of Ϸ240 Å and contain three hCx26M34A Orthorhombic Crystals. To determine a 3D structure, lipid bilayers (labeled Mem-1, Mem-2, and Mem-3 in Fig. 1). we collected images of samples tilted up to 45° and combined Remarkably, the map also shows that the hemichannels re- them to produce a density map at a resolution of 10 Å in the docked through their extracellular surfaces, forming complete membrane plane and 14.1 Å normal to the membrane plane (Fig. gap junction channels (Fig. 1 A and B). This is consistent with 1, SI Table 1, and SI Fig. 8E). A side view of the 3D map reveals published results proposing extensive hydrophobic surfaces in

Oshima et al. PNAS ͉ June 12, 2007 ͉ vol. 104 ͉ no. 24 ͉ 10035 Downloaded by guest on September 28, 2021 Fig. 2. Structural details of the Cx26 gap junction. The map is contoured as in Fig. 1. (Inset) Twenty-angstrom-thick section perpendicular to the membrane plane through the density map of a hemichannel in Mem-2. This section corresponds to the region enclosed by the white lines shown in A. The arrowhead points to the large density in the pore. The inner cytoplasmic protrusions (white arrows) extend from the cytoplasmic ends of helices B and C. (A–C) Thirty-angstrom-thick slabs through the density map corresponding to the position of the lines shown in Inset. The four helices are labeled A (cyan, AЈ), B (green, BЈ), C (yellow), and D (pink) as in the original Cx43 structure (5). The arrowhead and white arrows represent the plug and the inner cytoplasmic protrusions, respectively,asinInset.

the gap region (14). In bilayers Mem-1 and Mem-3, the ture made it possible to unambiguously dock the proposed hemichannels show poorer density than in Mem-2 (Fig. 1 C and ribbon model of the transmembrane domain (PDB accession no. D), presumably because of variability in the molecular packing 1TXH) with slight modifications (Fig. 2). A comparison of the because of the large lipid areas between individual hemichannels transmembrane cylinder models for Cx43 (6) with our Cx26 and/or the flexibility of the cytoplasmic domains of the connexin structure (Fig. 3) show that the dimensions of the channel, the subunits. The cytoplasmic structures in Mem-1 and Mem-3 may size of the pore constriction, and the positions of the helices are also be deformed by their contact with the carbon film to which all essentially the same. It should be noted that Cx26 is part of the crystals are adsorbed in the sample preparation procedure the ␤-subgroup of the connexin family, whereas Cx43 is a for cryoelectron microscopy. By contrast, the hemichannels in member of the ␣-subgroup, yet the overall structure of these two Mem-2 are protected from any forces such as the surface tension connexin isoforms is very similar. Small shifts in the positions upon specimen drying and mechanical interactions with the and tilts of these superimposed helices are more likely because carbon film. Therefore, the structural features of the hemichan- of the different crystal forms used in the structure analysis for nels in Mem-2 should be the most accurate and, in particular, Cx43 and Cx26 (hexagonal versus orthorhombic) than differ- preserve the structure of the flexible cytoplasmic domains of the ences between the isoforms. connexins. Thus, the following description of the gap junction There are three protrusions on the cytoplasmic surface for each structure is based on the hemichannels in Mem-2 unless noted four-helix bundle, AЈ,BЈ, C, and D (Figs. 2 and 4). The cytoplasmic otherwise. extension of helix D makes contact with the one from the adjacent Each connexon in a gap junction membrane channel contains gap junction channel, stabilizing the crystal packing (red arrows in four transmembrane segments, usually referred to as M1, M2, Fig. 1 B and D). Four-helix bundles are connected by a bridge-like M3, and M4 (15). The combination of our x–y resolution of 10 density forming the largest of the three cytoplasmic protrusions Å and the well defined cytoplasmic and transmembrane struc- (white arrows in Fig. 2; and see Fig. 4). This new density could

10036 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0703704104 Oshima et al. Downloaded by guest on September 28, 2021 a plug physically blocks the channel within the membrane. Density for the plug is also observed in the pores of the hemichannels in membranes Mem-1 and Mem-3 (Fig. 1 B and C). Each hemichannel has its own plug, conferring on it the ability to gate its pore autonomously. It is possible that the transjunctional voltage sensor and the physical gate reside exclusively within a single hemichannel (7, 17). The connection of the plug to the top of the channel wall is not perfectly resolved, presumably because of flexibility of the linker (Fig. 2 Inset). We applied a B-factor of Ϫ700 to the map to enhance the higher-resolution weak amplitude data. This map shows that the plug consists of six ␣-helix-like features and that it has four protrusions, implying connections from the plug (Fig. 5). It is not surprising that only four protrusions rather than six appear from the plug because 6-fold symmetry is lost in this orthorhombic crystal form. These features strongly suggest that the plug is formed by either the N- or C-terminal helices of Cx26.

Fig. 3. Superimposition of the transmembrane helices of Cx26 (yellow) and Discussion Cx43 (red) (6). The arrangement of the helices differs slightly between the two Key findings of this study are that hydrophobic interactions at the connexins, but all of the corresponding helices more or less overlap with each extracellular domains drive hemichannels to redock into do- other. The channel dimensions and the pore diameters are approximately the same in gap junctions formed by Cx26 and Cx43. decameric channels, the transmembrane domain structure is fairly conserved between ␣- and ␤-connexins and different crystal forms, and each hemichannel has its own plug in its represent part of the cytoplasmic loop of Cx26, because this domain vestibule. The crystal form shown here is unprecedented in the caps the top of helices B and C. The cytoplasmic loop has a larger gap junction structure literature because it contains three mem- number of amino acid residues than the N or C terminus, and thus branes and two sets of symmetry-related intercellular channels. should have a larger mass density. This implies that the connexin This could occur only because removal of the detergent from polypeptide boundary may be across adjacent four-helix bundles. hemichannels causes the hydrophobic extracellular surfaces to However, flexible loops are often invisible in crystallographic maps. be exposed. Rather than have these surfaces face an aqueous The two loops in the extracellular gap domain are not clearly environment, hemichannels redocked into an intercellular chan- separated (Fig. 2C) and are postulated to form a differently nel. Comparison of the shape, size, and arrangements of the organized subdomain (16). At this time, the resolution is insufficient transmembrane helices in the hCx26M34A structure with the to assign these ␣-helical segments and the polypeptide boundary to truncation mutant of Cx43 show that the architecture of the gap specific sequences within the transmembrane domain. junction channel is conserved between these two isoforms. Without the resolution to unambiguously trace the primary The Prominent Pore Density in the Vestibule. The 3D map shows a sequence of the connection from the plug to the surrounding density in the center of the pore (white arrowheads in Figs. 2 and channel wall, we can only speculate that the plug arises from one 4). Because this density is observed in both projection maps of of the termini, in particular, the N terminus. tannic acid- and trehalose-embedded specimens (data not shown), it is most likely that the plug is part of the connexin and Possible Candidates for Components of the Plug. Our Cx26 gap junction crystal structure shows that the channel vestibule is not an artifact of the embedding agent. This plug density is blocked by a physical obstruction we call the ‘‘plug.’’ Whether clearly visible even in a map contoured at 2.4␴ (wheat-colored this structure reflects a functionally closed nonpermeant channel contour in Fig. 2). The plug is located inside of the membrane or an aberrant mutant channel remains to be determined. layer and forms contacts with the surrounding channel wall, However, it is unlikely that a single amino acid change by itself which, at the constricted part of the vestibule, is formed by the would give rise to such a prominent feature. We have obtained innermost helices C (Fig. 2B). This density strongly suggests that several 0° projection micrographs of wild-type 2D crystals whose crystallinity is not as good as crystals of the hCx26M34A mutant. These 2D reconstructions have also revealed the density in the

pore (data not shown). This last finding strongly implies that the BIOPHYSICS physical channel closure may be regulated independently in each of the hemichannels and that this plug may make a channel impermeant to larger molecules such as those used in our previous dye-transfer studies (13). The likely candidates for the plug and its connector are the N terminus, the cytoplasmic loop, and the C terminus of Cx26. The N terminus is the most probable candidate of the three. A wealth of electrophysiological studies has established that the transjunctional voltage sensor resides in the N terminus and that the residues sensing changes in the voltage field are located in the vestibule of the pore (7–9, 18). It has also been proposed that movement of the charges in the N terminus initiate channel gating (7, 19). This notion is supported by an NMR solution Fig. 4. Stereoview of the surface structure of Cx26 perpendicular to the structure of an N-terminal peptide of Cx26 (20), although it has membrane plane. The cytoplasmic protrusions are clearly deﬁned. The map is not yet been shown definitively that the N terminus forms the contoured at 1.0␴ (solid surface) and at 2.4␴ (wheat-colored mesh) above the physical assembly. We propose that the plug, which is located mean density. Helices B (green) and C (yellow) are color coded as in Fig. 2. within the pore region, and is therefore ideally located to detect

Oshima et al. PNAS ͉ June 12, 2007 ͉ vol. 104 ͉ no. 24 ͉ 10037 Downloaded by guest on September 28, 2021 Fig. 5. Stereo top view of the Cx26 density map to which a B-factor of Ϫ700 was applied. The B-factor was applied to enhance the amplitudes of the high-resolution reﬂections, revealing six ␣-helix-like features in the plug density and protrusions that probably reﬂect the loops connecting the plug to the surrounding channel wall. The four helices are color-coded A (cyan, AЈ), B (green, BЈ), C (yellow), and D (pink) as in Fig. 2.

the transjunctional voltage field, contains the N termini of the The Plug Creates a Blocked State of Cx26 Gap Junction Channel. Since connexin subunits. This is different from voltage-sensitive Kϩ or the M34T substitution of hCx26 was reported to be the cause of Naϩ channels, where the S4 domain is the voltage sensor but nonsyndromic hearing loss (11), the functional role of the does not form an assembled plug (21). It should be noted that position 34 has been studied, and it has been suggested that sequence analysis has shown that only N termini have very M34T leads to a constriction of the channel pore (12). In our similar lengths and a highly conserved sequence, but both the work, it is conceivable that the plug in the pore is locked into cytoplasmic loops and the C termini show variable lengths and place by the hCx26M34A mutation because the physical obstruc- less sequence conservation (22). This sequence conservation of tion is consistent with the decreased permeability of the the N terminus might, in turn, result in conservation of the plug hCx26M34A mutation (13). In this case, the mutation at the M34 feature and closing mechanism within the connexin channel site in M1 could affect the movement of the M1 helix in each family. connexin in each hemichannel, thus resulting in a change of the The C terminus is the next most likely candidate for the plug. plug position that makes the channels nonfunctional or poorly The Cx26 construct in this work has a hexa-histidine tag with a functional. thrombin digestion recognition sequence linker that results in It should be noted that not only have we used a permeability Ϸ30 aa residues assigned to the C-terminal tail and makes it mutant for this study but also that we have used crystallization slightly longer than the N-terminal tail (21 aa residues). It has conditions (low pH, aminosulfonate buffer, carbenoxolone, and also been proposed that the C terminus and the part of cyto- high Ca2ϩ and Mg2ϩ) that generally promote a closed Cx26 plasmic loop is involved in pH regulation of gap junction gating, channel. The remaining pore openings at the constriction site are which is referred to as ‘‘particle-receptor’’ model (23), although Ͻ8 Å in diameter (Fig. 2B), indicating that these openings are there is no direct evidence that these domains interact within the too small for fully hydrated ions to pass (29). If our structure of vestibule. Whereas Cx26 channels do gate at low pH (24), this the blocked state resembles a functionally closed state because model has not been demonstrated for Cx26 channels because of of those factors in the crystallization condition, it allows us to the very short C-terminal tail. Another mechanism has been suggest a plug gating mechanism. Specifically, the two interact- proposed for pH gating of Cx26 hemichannels (25, 26), although ing Cx26 hemichannels can gate their channel directly and it has been reported that addition of tags to the Cx26 C terminus autonomously with a plug (Fig. 6). In this model, a Cx26 gap eliminates the ligand binding that promotes pH-induced channel junction is open only when both hemichannels release their plugs closure (27, **). Furthermore, we found that connexins without toward the cytoplasmic side. Once open, the pore can immedi- the C-terminal tag never crystallized, and removal of the tag ately conduct large molecules such as peptides with a molecular after crystallization destroyed the crystals (data not shown). It is mass of up to 1.8 kDa (30), because the large pore size (Ϸ15 Å more likely that the end of helix D corresponds to the hexa- histidine tag that serves to stabilize crystal formation rather than forming plug. The idea that the C terminus extends from helix D is consistent with the assignment of helix D as M4 as suggested (6, 28). The last candidate for the plug would be the cytoplasmic loop of Cx26, however, the number of amino acids (Ϸ35 residues) and requirement that this part of the sequence makes a loop cannot account for connecting densities from the plug (Fig. 5). Previ- ously published work has suggested that the N terminus and cytoplasmic loop of Cx26 may interact directly (18). Considering that even the largest cytoplasmic protrusion could be too small to cover the length of the cytoplasmic loop (Figs. 2 and 4), it is Fig. 6. Hypothesized plug gating mechanism of gap junctions. Each possible for the N terminus and the cytoplasmic loop to coop- hemichannel (green) can regulate its channel activity autonomously. The gap erate with each other, resulting in forming the plug. junction is open only when the plugs (red) in both hemichannels are displaced from the channel constriction formed by the innermost helices C (yellow) toward the cytoplasmic side. The ﬂexible connections of the plug with the **Tao, L., Harris, A. L. (2005) Biophys J 88:201a (abstr.). channel are shown as red dashed lines.

10038 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0703704104 Oshima et al. Downloaded by guest on September 28, 2021 in the most constricted region) virtually does not change during Sample Preparation. Samples were negatively stained with 2% the gating cycle. Given that our structure and others (15) have uranyl acetate for screening of 2D crystals. For cryoelectron indicated that the cytoplasmic domains of connexin are very microscopy, crystals were mixed with dialysis buffer containing labile, the movement of the plug could also be regulated by 0.05–1% tannic acid, 2–40% trehalose, or their combination and interactions with other cytoplasmic connexin domains, which applied to molybdenum grids covered with a thin carbon film. might account for the diversity of channel properties between The different concentration of tannic acid or trehalose did not different connexins. affect the plug in the pore because all of the 0° projections In either case, the fact that each hemichannel has its own plug obtained from any preparations showed the plug density as well (data not shown). After removal of excess liquid, the grid was supports the argument that the relationships between two op- blotted with filter paper and plunged into liquid nitrogen. posing hemichannels can be independent of one another. This notion well reconciles the observation of functional conductive Cryoelectron Microscopy and Structure Determination. Frozen spec- hemichannels such as Cx46, Cx56, and Cx32 chimera designated imens were transferred into a JEM-3000SFF electron micro- Cx32*Cx43E1 (31–33). Because the current Cx43 structure does scope (JEOL, Tokyo, Japan) operated at 300 kV and equipped not have any plug density in the pore (5, 6), further studies are with a field emission gun and a superfluid helium stage (34). The necessary to verify whether the plug structure can be generalized specimens were cooled to a temperature of 4K, and images of to all connexins. Structures at higher resolution of the open and specimens tilted up to 45° were recorded on SO-163 film (Kodak, closed states will provide more structural detail and will be Rochester, NY) at a magnification of ϫ60,000 with an electron needed to fully understand the functional role of a plug and dose of 25 electrons per Å2. The quality of images was checked gating mechanisms in these widely expressed channels. by optical diffraction, and selected images were digitized with a SCAI scanner (Zeiss, Thornwood, NY) by using a step size of 7 ␮ Materials and Methods m. Lattice distortions and the contrast-transfer function were corrected with the MRC package (35–37). The final density map Protein Purification and 2D Crystallization. hCx26M34A hemichan- is based on 254 images that were combined to generate a merged nels were purified as described (13), with the minor modification phase and amplitude data set. The 3D density map was visualized that the protein was eluted with 300 mM L-histidine instead of with the program Pymol (http://pymol.sourceforge.net/). 300 mM imidazole. Purified hCx26M34A hemichannels were mixed with decyl maltoside-solubilized DOPC (Avanti Polar This work is supported by Grants-in Aid for Specially Promoted Re- Lipids, Alabaster, AL) at a lipid-to-protein ratio of 1. The search, Grant-in Aid for 21st Century Centers of Excellence, Kyoto mixture was dialyzed against 10 mM Mes (pH 5.8), 100 mM University, Japan Science and Technology Corporation and New Energy NaCl, 50 mM MgCl , 5 mM CaCl ,2mMDTT,100␮M and Industrial Technology Development Organization (to Y.F.), Na- 2 2 tional Institutes of Health Grant GM065937 (to G.E.S.), and National carbenoxolone (Sigma, St. Louis, MO), 0.005% NaN3, and 1% Science Foundation Grants MCB-0543934 (to G.E.S.) and RR04050 (to glycerol. Dialysis was performed at 20°C for the first 24 h, 37°C Mark Ellisman, National Center for Microscopy and Imaging Research, for the next 96 h, and 20°C for the final 24 h. University of California at San Diego).

1. Harris AL (2001) Q Rev Biophys 34:325–472. 21. Jan LY, Jan YN (1992) Annu Rev Physiol 54:537–555. 2. Kelsell DP, Di WL, Houseman MJ (2001) Am J Hum Genet 68:559–568. 22. Cruciani V, Mikalsen SO (2006) Cell Mol Life Sci 63:1125–1140. 3. Unwin PNT, Zampighi G (1980) Nature 283:545–549. 23. Delmar M, Coombs W, Sorgen P, Duffy HS, Taffet SM (2004) Cardiovasc Res 4. Unwin PNT, Ennis PD (1984) Nature 307:609–613. 62:268–275. 5. Unger VM, Kumar NM, Gilula NB, Yeager M (1999) Science 283:1176–1180. 24. Stergiopoulos K, Alvarado JL, Mastroianni M, Ek-Vitorin JF, Taffet SM, 6. Fleishman SJ, Unger VM, Yeager M, Ben-Tal N (2004) Mol Cell 15:879–888. Delmar M (1999) Circ Res 84:1144–1155. 7. Verselis VK, Ginter CS, Bargiello TA (1994) Nature 368:348–351. 25. Bevans CG, Harris AL (1999) J Biol Chem 274:3711–3719. 8. Purnick PE, Oh S, Abrams CK, Verselis VK, Bargiello TA (2000) Biophys J 26. Yu J, Bippes CA, Hand GM, Muller DM, Sosinsky GE (2007) J Biol Chem 79:2403–2415. 282:8895–8904. 9. Oh S, Abrams CK, Verselis VK, Bargiello TA (2000) J Gen Physiol 116:13–31. 27. Tao L, Harris AL (2007) Mol Pharmacol 71:570–579. 10. Peracchia C (2004) Biochim Biophys Acta 1662:61–80. 28. Skerrett IM, Aronowitz J, Shin JH, Cymes G, Kasperek E, Cao FL, Nicholson 11. Kelsell DP, Dunlop J, Stevens HP, Lench NJ, Liang JN, Parry G, Mueller RF, BJ (2002) J Cell Biol 159:349–360. Leigh IM (1997) Nature 387:80–83. 29. Miyazawa A, Fujiyoshi Y, Unwin N (2003) Nature 423:949–955. 12. Bicego M, Beltramello M, Melchionda S, Carella M, Piazza V, Zelante L, 30. Neijssen J, Herberts C, Drijfhout JW, Reits E, Janssen L, Neefjes J (2005) Bukauskas FF, Arslan E, Cama E, Pantano S, et al. (2006) Hum Mol Genet Nature 434:83–88. 15:2569–2587. 31. Paul DL, Ebihara L, Takemoto LJ, Swenson KI, Goodenough DA (1991) J Cell 13. Oshima A, Doi T, Mitsuoka K, Maeda S, Fujiyoshi Y (2003) J Biol Chem Biol 115:1077–1089.

278:1807–1816. BIOPHYSICS 14. Ghoshroy S, Goodenough GA, Sosinsky GE (1995) J Memb Biol 146:15–28. 32. Ebihara L, Berthoud VM, Beyer EC (1995) Biophys J 68:1796–1803. 15. Sosinsky GE, Nicholson BJ (2005) Biochim Biophys Acta 1711:99–125. 33. Pfahnl A, Zhou XW, Werner R, Dahl G (1997) Pflu¨gers Arch 433:733– 16. Foote CI, Zhou L, Zhu X, Nicholson BJ (1998) J Cell Biol 140:1187–1197. 779. 17. Harris AL, Spray DC, Bennett MV (1981) J Gen Physiol 77:95–117. 34. Fujiyoshi Y, Mizusaki T, Morikawa K, Yamagishi H, Aoki Y, Kihara H, Harada 18. Oh S, Rubin JB, Bennett MVL, Verselis VK, Bargiello TA (1999) J Gen Physiol Y (1991) Ultramicroscopy 38:241–251. 114:339–364. 35. Henderson R, Baldwin JM, Downing KH, Lepault J, Zemiln F (1986) 19. Oh S, Rivkin S, Tang Q, Verselis VK, Bargiello TA (2004) Biophys J Ultramicroscopy 19:147–178. 87:912–928. 36. Henderson R, Baldwin JM, Ceska TA, Zemlin F, Beckmann E, Downing KH 20. Purnick PE, Benjamin DC, Verselis VK, Bargiello TA, Dowd TL (2000) Arch (1990) J Mol Biol 213:899–929. Biochem Biophys 381:181–190. 37. Crowther RA, Henderson R, Smith JM (1996) J Struct Biol 116:9–16.

Oshima et al. PNAS ͉ June 12, 2007 ͉ vol. 104 ͉ no. 24 ͉ 10039 Downloaded by guest on September 28, 2021