Molecular Membrane Biology, November /December 2005; 22(6): 485 /496
Structural and dynamic studies of the g-M4 trans-membrane domain of the nicotinic acetylcholine receptor
P. T. F. WILLIAMSON1, G. ZANDOMENEGHI1, F. J. BARRANTES2, A. WATTS3, & B. H. MEIER1 1
Physical Chemistry, ETH-Zurich, Switzerland, 2UNESCO Chair of Biophysics and Molecular Neurobiology and Instituto de Investigaciones Bioquı´micas de Bahia Blanca, Argentina, and 3Biomembrane Structure Unit, Biochemistry Department, University of Oxford, Oxford, UK (Received 10 June 2005; and in revised form 4 August 2005)
Abstract A structural characterization of a synthetic peptide corresponding to the fourth transmembrane domain (M4-TMD) of the g -subunit of the nicotinic acetylcholine receptor from Torpedo californica has been undertaken. Solid-state NMR and CD spectroscopy studies indicate that upon reconstitution into lipid vesicles or magnetically aligned lipid bilayers, the synthetic M4-TMD adopts a linear a -helical conformation with the helix aligned within 158 of the membrane normal. Furthermore, analysis of the motional averaging of anisotropic interactions present in the solid-state NMR spectra of the reconstituted peptide, indicate that the dynamics of the peptide within the bilayer are highly sensitive to the phase adopted by the lipid bilayer, providing an insight into how the interaction of lipids with this domain may play a important role in the modulation of this receptor by its lipid environment.
Keywords: Nicotinic acetylcholine receptor, magnetically aligned phases, ether-linked bicelles, cross-polarization magic-angle spinning solid-state NMR
Introduction The nicotinic acetylcholine receptor (nAcChoR) represents one of the best characterized members of the pentameric ligand gated ion channel family of receptors, which includes the GABA receptor, serotonin receptor (5-HT3) and glycine receptors. Muscle and electric organ nicotinic acetylcholine receptors are composed of five homologous subunits (a2bgd ). Each subunit is composed of a large Nterminal extracellular domain of /200 amino acids followed by four hydrophobic domains of 20 /30 residues in length (M1 /M4) linked by hydrophilic loops of varying length and ends with a short Cterminal extracellular domain. It is generally accepted that these four hydrophobic domains correspond to the transmembrane domains of the receptor (Barrantes 2004). Currently the most comprehensive structural characterization of this class of receptors is a series of cryoelectron microscopy studies of AcChoR containing tubules embedded in ice (Miyazawa et al. 2003; Unwin 2000). The most recent of these studies provides insight into the morphology of this
protein, particularly of its extracellular ligand-binding domain, and insights into the overall structural of the transmembrane domain (TMD) down to 0.4 nm resolution which tentatively permits the identification of secondary structural elements. The best characterized TMD is M2, which constitutes the inner ring, devoid of contacts with the membrane lipid (for review see Barrantes 2003). Site directed mutagenesis data combined with patch-clamp electrophysiology and with photoaffinity labelling with non-competitive channel blockers indicate that M2 lines the channel lumen and adopts a a -helical conformation (Changeux & Edelstein 1998). NMR studies of the d -M2-TMD indicate that the domain is inserted into the bilayer at an angle of /128 with respect to the bilayer normal and adopts a purely a helical conformation with no-kinks (Opella et al. 1999). These observations are largely in agreement with the cryoelectron microscopy data for this TMD (Miyazawa et al. 2003; Unwin 2000). The other three transmembrane domains are less well characterized both structurally and functionally. M1 and M3 constitute an intermediate ring, in contact with the M2 ring on the inner surface and
Correspondence: P. T. F. Williamson, CNRS(FRE2446)/ULP Strasbourg, ISIS, 8 rue Gaspard Monge, F-67000 Strasbourg, France. Tel: /33 3 90 24 51 52. E-mail:
[email protected] ISSN 0968-7688 print/ISSN 1464-5203 online # 2005 Taylor & Francis DOI: 10.1080/09687860500370653
486
P. T. F. Williamson et al.
with the M4 ring on the outside surface (Barrantes 2003). The cysteine scanning accessibility method has been applied to the study of the M1-TMD, revealing that the residues on this transmembrane domain in the b subunit contribute to the channel lining (Zhang & Karlin 1997; Zhang & Karlin 1998), although these observations do not agree with those from other studies (Miyazawa et al. 2003). Chemical labelling studies employing photoactivatable hydrophobic probes indicate that the M3 and M4-TMD are both exposed to the lipid surface (Blanton & Cohen 1992; Blanton & Cohen 1994). These studies show that the observed periodicity in labelling is consistent with an a -helical structure, in agreement with deuterium exchange FTIR (Corbin et al. 1998) and fluorescence studies (Barrantes et al. 2000) performed on both the intact TMD of the whole nAcChoR and individual TMDs. NMR studies of the a -M3-TMD corroborate these observations, indicating a totally helical structure (Lugovskoy et al. 1998). These results are now largely supported by the latest electron microscopy studies of the intact receptor (Miyazawa et al. 2003; Unwin 2005). An understanding of the structural and functional interactions between these TMDs and the surrounding lipid environment is important for a comprehension of both structural and functional aspects of the receptors. The lipid /protein interface has been proposed as the site of action of local anaesthetics (Fraser et al. 1990; Mantipragada et al. 2003; Marsh & Barrantes 1978; Watts 1991), general volatile anaesthetics and short chain alcohols (Campagna et al. 2003; Miller 2002), whilst others have indicated how the function of the nAcChoR can be modulated by its lipid environment (Baenziger et al. 2000; Barrantes 1993; Sunshine & McNamee 1994). Here the structure and dynamics of the M4-TMD are characterized with the goal of understanding its structure and how it interacts with its surrounding lipid environment. The M4-TMD of the g-subunit has been synthesized with NMR active isotopes incorporated at selective sites. Solid-state NMR experiments have been performed on the M4TMD reconstituted into lipid vesicles and a binary lipid mixture of 1,2,-Di-O-tetradecyl-SN-glycero-3phosphocholine (DoMPC) and 1,2,-Di-O-hexylSN-glycero-3-phosphocholine (DoHPC). Above a critical temperature this binary lipid system forms a bicellar phase composed of liquid-crystalline lamellar-like structures which spontaneously align within the magnetic field (Sanders & Schwonek 1992) due to the diamagnetic anisotropy of the lipids. DoMPC and DoHPC ether-linked lipids were used instead of ester-linked lipids both to improve sample stability and to facilitate the interpretation of the carbon-13
NMR spectra. Magic-angle spinning spectra of both lipid vesicles and bicellar systems have been obtained, and the observed chemical shifts exploited to predict the secondary structure of the peptide. Through the analysis of the anisotropic interactions observable in the NMR spectra of the M4-TMD reconstituted into these systems, it has been possible to analyze both the secondary structure and the orientation of the peptide with respect to the membrane normal. Furthermore, analysis of the spectra in both the gel and liquid crystalline phase reveals significant changes in peptide dynamics. Materials and methods Synthesis of the g -M4 TMD of the nicotinic acetylcholine receptor A peptide composed of the sequence N-Asp-LysAla-Cys-Phe-Trp-Ile-(D3)Ala8-Leu-Leu(1-13C)Leu11-Phe-Ser-Ile-(15N)Gly15-Thr-Leu-AlaIle-Phe-Leu-Thr-(2-13C)Gly23-His-Phe-Asn-GlnVal-C (corresponding to Asp464 to Val492 in the intact g-subunit) was prepared using conventional FMOC synthesis (NSR Centre, Nijmegen, Netherlands). All labelled amino acids were purchased from Cambridge Isotopes Ltd. (MA, USA) and protected using conventional FMOC chemistry (Jones 1997). The resulting peptide was deemed to be over 90% pure as determined by analytical HPLC and mass spectroscopy. Reconstitution of g -M4 TMD into DoMPC vesicles and DoMPC:DoHPC bicelles Bilayer vesicles containing the M4-TMD of the gsubunit of the nicotinic acetylcholine receptor were prepared by co-solubilizing peptide ( /1 mg) and DoMPC at a molar ratio of 1:50 in a chloroform/ methanol mixture (1:1). All lipids were purchased from Avanti Polar Lipids (Alabaster, AL) and used without further purification. Following solubilization, a thin lipid protein film was prepared by rotary evaporation and left under high vacuum (10 8 bar) overnight to remove trace quantities of organic solvent. The film was resuspended in excess water at room temperature and subjected to five cycles of freeze thawing. The sample was then lyophilised and the resulting powder resuspended in buffer (10 mM TRIS, 1 mM EDTA, 10 mM NaCl, pH 7.2) at a concentration of 50% (w/v) and subjected to five cycles of freeze thawing and vortexing. Bicelles containing the M4-TMD were prepared by co-solubilizing the peptide (/1mg) and DoMPC at a molar ratio of 1:50 in a chloroform/methanol mixture (1:1). A lipid-protein film was prepared as described above. The DoMPC/M4 film was then
Structure and dynamics of the nAChRs 4th TMD hydrated to a concentration of 30% (w/v) with buffer. A solution of DoHPC was added to the resulting vesicles such that the final molar ratio of DoMPC:DoHPC:M4 was 50:16:1 and hydrated at 20% (w/v). The sample was then subjected to three cycles of freeze thawing. Prior to the NMR studies, all samples were transferred into a 6 mm Chemagnetics rotor and the top sealed to prevent dehydration during the course of the experiments. Circular dichroism spectroscopy Circular Dichroism spectra were recorded on a Jasco /500 at 358C. Measurements were performed on the M4-TMD in TFE at a concentration of 0.28 mg ml 1 with a cell path length of 0.05 cm with 1 nm resolution and a scan speed of 10 nm min 1. Spectra of the M4-TMD in bicellar solution were obtained at a concentration of 0.89 mg ml 1 and were measured in a cell with a path-length of 0.01 cm, 2 nm resolution and a scan speed of 20 nm min 1. The spectra were analysed using the deconvolution algorithm Contill in the range of 250 /190 nm using basis set 5 to ascertain the secondary structure composition of the M4-TMD in TFE and in lipid bicelles (Provencher & Glockner 1981; Vanstokim et al. 1990). Solid-state NMR measurements All NMR measurements on the reconstituted M4TMD were performed on a Varian Infinity/ 500 spectrometer equipped with 6 mm double resonance MAS-NMR probe. Deuterium NMR measurements were made using a standard quadrupolar echo sequence using 5 ms 908 pulses with a 50 ms interpulse delay and a 500 ms recycle time (Davis et al. 1976). Deuterium spectra shown are typically the result of 128 k acquisitions. Phosphorous-31 NMR measurements were acquired using direct excitation and high-power proton decoupling. Phosphorous-31 908 pulse lengths were 6 ms and the proton decoupling field amplitude was 60 kHz, spectra shown are the result of 512 acquisitions. Carbon-13 spectra were acquired with direct excitation and high-power proton decoupling; carbon-13 908 pulse lengths were typically 5 ms with decoupling fields of 60 kHz. Cross polarization spectra of both carbon-13 and nitrogen-15 were performed using a adiabatic sweep of the proton spin-lock field through the Hartmann /Hahn condition (Hediger et al. 1994) with central carbon-13 and nitrogen-15 spin lock fields of 50 and 35 kHz respectively. For all experiments employing cross-polarization and decoupling a recycle time of 3.5 sec was employed to minimize sample heating. Typically 32 k and 64 k acquisitions
487
were made for carbon-13 and nitrogen-15 spectra respectively. Carbon-13 and nitrogen-15 chemical shielding tensors for the lyophilized peptide were measured on a Bruker Avance 600 spectrometer. Phosphorous-31 chemical shifts are reported relative to an 85% H3PO4 solution at 258C and carbon-13 chemical shifts relative to external TMS. Nitrogen15 chemical shifts are reported relative to liquid ammonia whose frequency was calculated relative to the proton resonance frequency of TMS using a ratio, F, of 0.101329144 (Live et al. 1984). All spectra are referenced with chemical shifts; however the anisotropies are reported as chemical shielding. Results and discussion Orientation of bicelles containing the g-M4-TMD The M4-TMD has been incorporated into bicelles composed of DoMPC and DoHPC (Cavagnero et al. 1999, Ottiger & Bax 1999) at a molar ratio of 3:1 at 20% (w/v) to induce macroscopic ordering of the system. To determine the range of temperatures over which the DoMPC/DoHPC bicelles adopted a macroscopically ordered phase, the chemical shift of the two 31P resonances (see Figure 1A and B) was followed as a function of temperature (Figure 2). Below 258C the 31P spectra of the DoMPC/DoHPC mixture give rise to a single resonance corresponding to the isotropic shift, /0.75 ppm, of both species (the similarity of the isotropic shifts of DoMPC and DoHPC precludes the resolution of both species). Above 258C, an upfield shift is observed as the two species adopt a macroscopically ordered phase where the resonance lines observed correspond to the lipids orienting with their long axis perpendicular to the magnetic field (Sanders & Schwonek 1992). This upfield shift for both the DoMPC and DoHPC resonances continues as the temperature rises. A similar profile is observed for the DoMPC/DoHPC
Figure 1. Proton decoupled phosphorous-31 NMR spectra of DoMPC/DoHPC bicelles (q /3, 20% w/v, 358C) (A) and with the M4-TMD incorporated at a lipid to protein ratio of 50:1 (B).
488
P. T. F. Williamson et al.
Figure 2. Variation of phosphorous-31 chemical shift as a function of temperature for both DoMPC/DoHPC and DoMPC/DoHPC/M4 bicelles. Phosphorous-31 chemical shifts for bicelles in the absence and presence of M4 are indicated. In absence of M4, DoMPC (") and DoHPC(2) and in the presence of M4 DoMPC(m) and DoHPC(k).
bicelles containing the M4-TMD. In both cases, the observed transition from isotropically tumbling bicelles to oriented bicelles is similar to that observed for bicelles prepared from DoMPC/DoHPC (q /3) at 5% w/v (Ottiger & Bax 1999) indicating that the level of hydration and incorporation of the M4TMD, in this instance, do not significantly effect the ability of the system to undergo macroscopic ordering above the phase transition. Figure 1A shows 31P NMR spectra of DoMPC:DoHPC bicelles without peptide at a temperature of 358C. Qualitatively these spectra show good agreement with 31P spectra observed for comparable ester linked lipids (Sanders & Schwonek 1992), with two resonance lines at /3.129/0.05 ppm and /8.899/ 0.05 ppm arising from DoHPC and DoMPC respectively. The spectrum is consistent with a bicelle director that is perpendicular to the magnetic field. Analysis of the 31P spectrum (Zandomeneghi et al. 2003) indicates that in the absence of M4TMD the observed chemical shielding anisotropy (CSA) of the DoHPC and DoMPC is /4.89/0.1 ppm and /16.49/0.1 ppm, and the distribution of the two components about the bicelles director, m, is 589/0.18 (considering statistical errors only). The spectrum of DoMPC/DoHPC bicelles containing the M4-TMD is shown in Figure 1B, and consists of two resonances at /4.569/0.05 and / 11.069/0.05 ppm for DoHPC and DoMPC respectively. Analysis of the observed lineshapes indicates a mosaic spread of 3.79/0.18 with a CSA of /8.09/ 0.1 ppm and /21.19/0.1 ppm for DoHPC and DoMPC respectively. Notably, in the presence of
M4-TMD, a larger 31P CSA is observed compared to that for the DoMPC/DoHPC bicelles alone. The increase in CSA observed in the bicelle samples may arise from several parameters including enhanced ordering of the bicellar phase through the incorporation of the peptide (Bolze et al. 2000) or a reduction in the mobility of the lipid headgroup through the interaction of the peptide with the surrounding lipids (Belohorcova et al. 2000). Similar increases in 31P CSA were also observed in spectra of vesicles prepared from DoMPC in the presence and absence of the M4-TMD (data not shown) where significant changes in macroscopic ordering cannot contribute to the observed CSA. This suggests that the increased CSA observed arises from an overall reduction in lipid headgroup mobility through the interaction of the peptide with the surrounding lipid environment. This appears in contrast to recent studies of the M1-TMD of the nAcChoR in DMPC bilayers where the incorporation led to a increase in headgroup mobility, as observed by a reduction in the 31P CSA, due to the disruption on the hydrogen bonding network between the lipid headgroups (de Planque et al. 2004). To compare the results obtained in vesicle and bicellar systems we must take into account the additional rapid (/10 kHz) motional averaging about the bicelle director normal that is not present in the vesicle system. This can be estimated by comparing the CSA observed for the phosphate group within the lipid of the bicellar sample with that observed for DoMPC vesicles, and is characterized by the order parameter Sbic. For M4-TMD containing DoMPC/DoHPC bicelles at 358C Sbic is found to be 0.819/0.07. CD spectroscopy of the g -M4 TMD CD spectra of the M4-TMD in trifluoroethanol and in an aqueous solution of bicelles at the same lipid to protein ratio employed for all NMR experiments are shown in Figure 3. The spectrum of the M4-TMD in TFE is characteristic of a a -helical peptide with a pronounced maximum at 190 nm and double minima at 208 and 220 nm. Although these features are not as pronounced in the spectrum of the M4TMD reconstituted into bicelles, this spectrum is consistent with a predominantly a -helical conformation. Multi-component analysis of these spectra indicates that over 50% of the peptide adopts a a helical conformation with the remaining residues showing alternative secondary structures including random coil and b structures (both strand and turn). This is somewhat lower than observed in the recent electron microscopic studies of the intact receptor (Miyazawa et al. 2003) where the secondary
Structure and dynamics of the nAChRs 4th TMD
489
the introduced labels are sufficiently well spaced that dipolar interactions between the labelled sites and interactions between the labelled 13C atoms and directly bonded 14N (maximally 700 Hz) are sufficiently small that they are not discernable on the lineshapes measured here. NMR analysis of the M4-TMD reconstituted into DoMPC vesicles
Figure 3. CD spectra of the M4-TMD dissolved in TFE (solid line) and reconstituted bicelles (dotted line). Both spectra recorded at 358C.
structure observed sequence corresponding to the peptide studied here is almost entirely helical. However, this would be sufficient for the formation of a single transmembrane helix with the addition extra-membranous domains accounting for the addition random-coil and b -sheet structures. Presumably, these discrepancies reflect the additional ordering which may arise from additional contacts present in the intact receptor. NMR characterization of lyophilized M4-TMD In order to describe the dynamics and orientation of the peptide in the bilayer it is necessary to characterize the static anisotropic interactions arising at each of the labelled sites. Using a combination of slow MAS spectra (vr B/CSA) and static 2H, 13C and 15N NMR measurements of the lyophilized peptide (data not shown) the anisotropy, asymmetry and isotropic values for each of the labelled sites were obtained and are presented in Table I. The observed isotropic shifts and anisotropies are consistent with previously published values for lyophilised peptides (Hartzell et al. 1987, Stark et al. 1983). In the following analyses we have neglected dipolar interactions between low-gamma nuclei as Table I. Characterization of the anisotropic interactions present at the labelled sites within the lyophilised M4-TMD. Labelled Site
Isotropic Shift
2
n.d.a 176.6 ppm 107.0 ppm 43.2 ppm
H3-Ala 1-13C-Leu 15 N-Gly 2-13C-Gly
Anisotropy 40.09/0.5 kHz /859/5 ppm 1109/5 ppm n.d.a
Asymmetry 0.00b 0.659/0.08 0.179/0.06 n.d.a
n.d., not determined; bAssumed to be zero due to rapid rotation of the methyl group. a
To probe the dynamics of the M4-TMD as a function of lipid phase, the M4-TMD was reconstituted into DoMPC vesicles and 2H, 13C and 15N spectra were acquired both above and below the gel to liquid-crystalline phase transition temperature (Figure 4). The 13C spectra of the M4-TMD reconstituted into vesicles both below and above the phase transition are shown in Figure 4A and D respectively. The slow MAS spectrum in Figure 4A permits the assignment of the peaks at 43.3 ppm and 176.5 ppm to the centre-band of 2-13C-Gly15 and 1-13CLeu11 respectively. From the sideband family observed for the 1-13C-Leu the chemical shielding anisotropy is estimated to be 759/10ppm and the asymmetry 0.89/0.2, comparable with those observed for a lyophilized sample (see Table I). The static 13C proton decoupled spectrum above the phase transition is shown in Figure 4D. The intense resonance in the carbonyl region is attributed to the 1-13C-Leu label. The line shape is characteristic of an axially symmetric lineshape spread between 169 and 190 ppm. Analysis of this lineshape is well described by an axially symmetric CSA with an anisotropy of /14.09/1.0 ppm, indicative of extensive motional averaging of the CSA. Other intense resonances in the spectrum arise from the natural abundance carbon-13 present in the lipids. The MAS 15N spectrum of the 15N-Gly16 of the M4TMD reconstituted into DoMPC vesicles below and above the phase transition are shown in Figure 4B and E, respectively. Analysis of this sideband family, characterizes the 15N CSA with an anisotropy of / 969/5 ppm assuming an asymmetry parameter of h /0. Above the phase transition only a slight reduction in the anisotropy to /869/5 ppm is observed. The deuterium spectra of the D3-Ala8 of the M4TMD reconstituted into DoMPC vesicles in the gel and liquid crystalline phase are shown in Figure 4C and F respectively. The spectrum in Figure 4C is characterized by a Pake pattern with a splitting of 37.6 kHz and a narrow resonance attributed to the residual D2O in the sample. The broader component is consistent with a deuterated methyl group undergoing rapid rotation about the C-CD3
490
P. T. F. Williamson et al.
Figure 4. Spectra showing the effect of the lipid phase on the dynamics observed by the M4-TMD. Spectra of the M4-TMD reconstituted into DoMPC vesicles below (58C) (A /C) and above (358) (D /F) the gel to liquid crystalline phase transition temperature. Proton decoupled 13C spectrum, with magic-angle spinning, vr /1.86kHz (A) and static (D). Cross polarization 15N spectra vr /1.735 kHz at 58C (B) and vr /1.700 at 358C (E). Deuterium quadrupolar echo lineshapes at 58C (C) and 358C (F).
bond, with further averaging indicative of additional motional processes absent. Above the phase transition the spectrum (Figure 4F) is characterized by a single rather narrow resonance. The resonance line contains, besides the sharp resonance from D2O, a broader component with a width of 3 / 4 kHz.
NMR analysis of the M4-TMD reconstituted into DoMPC:DoHPC bicelles The isotropic and in the macroscopically aligned 13 C, 15N and deuterium NMR spectra of the M4TMD reconstituted into bicelles are shown in Figure 5. In the isotropic carbon-13 spectrum, obtained under MAS (Figure 5A), two resonances at 176.5
Figure 5. Spectra showing effect of macroscopic orientation of bicelles on the solid state NMR spectra of the M4-TMD. Isotropic spectra (A /C) and macroscopically aligned (D /F). 13C and 15N isotropic spectra acquired with magic angle spinning, 2H isotropic spectra acquired below the phase transition of the bicelles (58C). Oriented lineshapes acquired on static samples at 358C (asterisks indicate the resonances arising from the carbon-13 labelled sites within the peptide).
Structure and dynamics of the nAChRs 4th TMD ppm and 43.3 ppm are clearly assignable to 1-13CLeu11 and 2-13C-Gly23 respectively and are the same as those observed for M4-TMD in vesicles. The foot on the upfield side of the resonance arising from 1-13C-Leu11 is attributed to the natural abundance 13 C in carbonyl groups within the peptide backbone (with an expected intensity of /25% of the labelled 1-13C-Leu11 resonance). Notably, the narrow distribution of resonances observed for the other carbonyl groups, despite the differing amino acids present, suggests that the other carbonyl groups may adopt a secondary structure similar to that observed for the labelled carbonyl site. In the static spectrum (Figure 5D), the carbonyl resonance shifts upfield to 170.2 ppm due to the residual CSA. From this shift, the anisotropy in the bicelle frame can be calculated to be /12.69/0.5 ppm. Correcting this for the additional motional averaging present in the bicelle sample (Sbic), a chemical shielding anisotropy of / 15.59/1.8 ppm is obtained, assuming an axially symmetric tensor. The magic-angle-spinning 15N spectrum of the M4-TMD reconstituted into bicelles is shown in Figure 5B and features a single resonance at 107 ppm. In the macroscopically aligned sample, the resonance shifts upfield to 70 ppm (Figure 5E). The CSA corrected for the bicelle order parameter is /919/12 ppm. The isotropic deuterium spectrum of M4-TMD in bicelles is shown in Figure 5C and was obtained from samples below the phase transition where the bicelles adopt an isotropic phase. On the basis of the chemical shift dispersion, the relative spectral intensity and the changes observed upon ordering the sample by increasing the temperature it is possible to assign many of the resonances in the isotropic deuterium spectrum. The main resonance at 0 kHz is attributed to residual D2O in the sample (spectra references to pure D2O) whilst resonances at /138 Hz ( /1.79 ppm) and /300 Hz ( /3.90 ppm) arise from residual deuterium at the g -position in the choline headgroup and the aliphatic chains respectively (Sixl & Watts 1983). In addition, a broad spectral feature is observed beneath the narrow resonances which is characterized by a broad Lorentzian lineshape of width 2300 Hz and can be tentatively assigned to the deuterated methyl group of the M4 in the sample. Upon raising the temperature to 358C, where the bicelles spontaneously align in the applied field, significant changes are observed in many of the resonances (Figure 5F). The peak assigned to residual D2O is split into a doublet with a splitting of 76 Hz, consistent with partial ordering of the water due to the interaction of the D2O with the oriented bicelles (Ottiger & Bax 1999). Additional
491
splittings of 950 Hz and 20 kHz which are observed in the spectra and are clearly assignable to the natural abundance lipid in the sample. Significant intensity is also apparent between /6.0 kHz and 6.0 kHz. Some of this intensity presumably arises from other natural abundance deuterons in the lipid head group and chain region. However, comparing the intensity observed with that expected from natural abundance deuterons in the lipids it appears that intensity must also arise from other components within the system. Thus we tentatively assign this intensity to the D3-Ala8 labelled M4-TMD with the broad lineshape possibly arising from motions occurring on the intermediate timescale (ms regime) (Sharpe et al. 2002). Analysis of backbone isotropic chemical shifts Analysis of 13C and 15N isotropic chemical shifts is used routinely to map secondary structural elements in proteins (Kameda & Ando 1997; Saito 1986; Wishart & Sykes 1994). Isotropic chemical shifts have been observed for 15N-Gly16 at 107 ppm and 176.5 and 43.4 ppm for 1-13C-Leu11 and 2-13CGly23 respectively. Comparison of these chemical shifts with the secondary chemical shift index (Wishart & Sykes 1994), suggests that all three residues are in a a -helical conformation. Notably very little change is observed in the isotropic chemical shifts below the phase transition, indicating that the peptide maintains its conformation. Analysis of dynamics and peptide orientation It has been shown that peptides in the bilayer may observe fast rotations about the bilayer normal and/ or about the peptide long axis (Cornell et al. 1988; Lewis et al. 1985; Marassi et al. 1999; Nicholson et al. 1991; Separovic et al. 1993; Smith et al. 1994; Yamaguchi et al. 2001; Yamaguchi et al. 2002). Most of the membrane peptides described in literature rapidly rotate about the bilayer normal. In the case where the peptide undergoes rapid rotation about the bilayer normal the averaging of the anisotropic interactions will depend on the Euler angles (aPD, bPD, gPD) which describe the transformation of the anisotropy from the PAS through the molecular frame (MF) to the director frame (DF), and will be characteristic of both the conformation of the peptide and its orientation with respect to the bilayer (Figure 6). In contrast, rapid rotation of the peptide about its long axis, the motional averaging observed will depend solely on the Euler angles (aPM, bPM, gPM) which describe the orientation of the PAS with respect to the molecular frame (in this case defined as along the peptide long axis) and
492
P. T. F. Williamson et al. are characteristic of the secondary structure. In the case of rapid motions about both the peptide long axis and the bilayer normal, the scaling observed will depend on the product of both scaling processes. In a bicelle sample, where the bilayer normal is aligned perpendicular to the magnetic field, the absence of molecular motion of the M4-TMD about the bilayer normal would result in a spectrum characteristic of a cylindrical distribution unless all the anisotropic interactions studied were axially symmetric and aligned parallel to the bilayer normal. The presence of narrow resonances at all three sites (Figure 5) in the bicelle sample has led to an interpretation of the data in which fast rotation of the M4-TMD about the bilayer normal is assumed. This assumption is also supported by the observation that the motions averaging the NMR interactions affecting the M4-TMD spectra depend strongly of the phase of the lipids, becoming slow when the sample is in the gel phase (Figure 4) and resulting in NMR spectra which are very similar to the ones of the lyophilized peptide. To interpret the scaling of the observed anisotropic interactions, a simplified motional model is initially assumed whereby the anisotropy is averaged by rapid rotational diffusion (on a timescale faster than 10 ms) about a single axis of motional averaging. Under these conditions the observed anisotropy is scaled according to the equation (Mehring 1983): d d [(3cos2 b 1)hcos2asin2 b] 2
Figure 6. The relationship between the anisotropic interaction present in the M4-TMD and how they are related to the morphology of the bilayer, together with the motional processes that can lead to the averaging of the anisotropic interactions. The initial transformation (VPM) transforms the PAS to the molecular frame (MF), in this case defined as collinear with the helix long axis. The second transformation (VMB) determines the orientation of the peptide with respect to the bilayer normal (bicelle frame, BF). In the case of vesicles the bicellar frame is defined as collinear with the director frame (DF). For bicelles, fluctuations between the bicelle normal and the director are accounted for by the transformation (VMB) and gives rise to the order parameter Sbic. The final transformation between the director and the lab frame (VDL) is characterized by the order in the sample. In the case of vesicles VDL is characterized by a powder distribution. In bicelles VDL is characterized by a Lorentzian distribution about the director normal. This figure is reproduced in colour in Molecular Membrane Biology online.
where a and b are the Euler angles that define the orientation of the tensor with respect to a axis of motional averaging, d is the chemical shielding anisotropy, h is the asymmetry of the shielding tensor and d is the motional averaged anisotropy of the axially symmetric tensor. Using the anisotropy and asymmetry parameters determined from the analysis of the spectra of the lyophilized peptide, a range of possible values for a and b can be obtained which are consistent with the observed scaling of the anisotropic interactions. In both the lyophilized M4TMD and the peptide reconstituted into DoMPC vesicle in the gel phase, the 13C-chemical shielding anisotropy of 1-13C-Leu11 is characterized by a tensor with d / /859/5 ppm and h /0.65. In vesicles in the liquid crystalline phase and in the bicelles the chemical shielding anisotropy is averaged to /15 ppm. The range of Euler angles consistent with such a scaling is shown in Figure 7A. Similarly the range of Euler angles a and b which can bring about a scaling of the 15N-chemical shielding
Structure and dynamics of the nAChRs 4th TMD
Figure 7. Contour plots showing how the chemical shielding anisotropy for the carbonyl (A) and amide (B) tensor averaged as a function of the Euler angles a and b which relate them to the axis of motional averaging. The contours plotted indicate the values for the averaged chemical shielding anisotropy (in ppm) which are consistent with the experimentally determined values for the carbonyl (A) and amide (B) tensors. Additional contours show the expected deviation in Euler angles upon varying the observed chemical shielding anisotropy by 9/5 ppm (A) and 9/10 ppm (B). The values plotted are calculated according to equation 1 assuming a static carbonyl tensor with an anisotropy of /85 ppm and asymmetry of 0.65 (A) and a amide tensor of /90 ppm and asymmetry of 0.17 (B). The areas highlighted indicate the areas where the Euler angles would relate the PAS of the chemical shielding anisotropy to the axis of motional averaging consistent with an a -helix.
anisotropy from a tensor of d / /1109/5 ppm, h / 0.179/0.16 observed in the lyophilized solid to the /869/5 ppm in the liquid crystalline vesicles and the bicelles are shown in Figure 7B. From studies of model compounds, the orientation of the principal axis of the 13C and 15N chemical shielding tensors with respect to a function group and thus the molecular frame is known (Hartzell et al. 1987; Stark et al. 1983). Although the transformation of carbonyl shielding tensor to the molecular
493
frame defined as the helix long axis shows some sequence specificity, it is typically characterized by the Euler angles 658 5/aPM 5/908 and 778 5/bPM 5/ 908 (Lewis et al. 1985; Smith et al. 1994; Yamaguchi et al. 2001). Similarly, the chemical shielding of the nitrogen amide has been reported to be oriented at 08 5/aPM 5/308 and 1485/bPM 5/258 (Hartzell et al. 1987; Lee et al. 1998; Separovic et al. 1993; Wu et al. 1995) with respect to the helix long axis. Comparison of the Euler angles that relate the PAS to a MF aligned along a helix long axis with the Euler angles that relate the PAS of the anisotropic interactions to the axis of motional averaging (region highlighted in Figure 7A and B) indicates that the data can be well explained with a single rotation axis collinear with the a -helical long axis. In the case of the deuterium label, the degree of motional averaging is only dependent of the angle bPM since the quadrupolar interaction of the methyl deuterons is axially symmetric. However it is clearly seen in the deuterium spectrum of the M4-TMD in the liquid crystalline phases (Figure 4F and 5F) that the line shape does not have the characteristic Pake pattern (as described above). The relatively narrow line width ( /3 /4 kHz) indicates that the methyl group of D3-Ala8 is oriented with respect to the averaging axis at an angle close to the magic angle (b /54.78), consistent with the angle of 548 observed between the C-CD3 axis of the alanine and the peptide long axis, assuming a a -helical conformation. In both the vesicle and bicelle samples the observed scaling of the anisotropic interactions is consistent with a linear a -helix undergoing rapid rotation about the helix long axis. As the observed scaling is explained by the this single motional process we can conclude that the peptide must be aligned approximately parallel to the bilayer normal as additional rotation about an axis tilted with respect to the bilayer normal would introduce additional scaling processes which cannot be rationalized with the observed experimental data. The uncertainty in the Euler angles which define the orientation of the PAS of the 1-13C and 15N chemical shielding anisotropies (about 10 /158) with respect to the molecular frame limits the definition of the tilt angle between the membrane normal and the a -helix axis. Thus, the experimental data presented is compatible with the helix axis being tilted with respect to the bilayer normal by any angle within the uncertainty on bPM in this case 158. As the scaling of the anisotropic interactions observed at each of the sites share a common axis of motional averaging it is concluded that between Ala8 to Gly15 there is no significant bending or kinking of the a -helix. This is in agreement with the recent
494
P. T. F. Williamson et al.
cryoelectron microscopy studies of the intact receptor (Miyazawa et al. 2003) which shows little deformation of the helix until after Gly15 where the helix bends and the tilt of the peptide increases slightly. Comparing these results with fluorescence studies (Antollini et al., 2005, this issue) we find that a helix tilt (bMD) of 158 would result in a helix chain tilt parameter of 1.03 (HCTP /1/cos(bMD)). Assuming the bilayer thickness of DoMPC mimicks that of DMPC (Nagle & Tristran-Nagle 2000), such an angle would result in the complete burial of the hydrophobic domain part of the M4-TMD within the hydrophobic core of the DoMPC bilayer in agreement with the fluorescence studies. The absence of dynamics in the vesicle samples below the phase transition prohibits our characterization of the orientation of the M4-TMD in the gel phase. Thus, we have been unable to draw comparisons with fluorescence spectroscopy studies which indicate a decrease in helix tilt due to the increased bilayer thickness in the gel phase (Antollini et al. 2005, this issue). Similarly, the absence of a gel phase for bicelles composed of DoMPC and DoHPC precludes our determination of the orientation of the M4-TMD in the gel phase. The differences in observed scaling of the anisotropic interactions between the gel and liquid crystalline phase also allow the characterization, of both dynamics and quaternary structure of the M4TMD with respect to the lipid bilayer. The significant motion on the microsecond timescale observed by the M4-TMD reconstituted into a liquid crystalline bilayer, suggests that in DoMPC the peptide has little propensity to aggregate. This is in agreement with recent fluorescence studies of the same synthetic peptide reconstituted into POPC bilayers, which also show that the M4-TMD exists in a predominantly monomeric form (de Almeida et al. 2004). This contrasts with earlier studies of the M2TMD where higher oligomeric pore forming structures were identified using both solid state NMR and electrophysiological studies (Opella et al. 1999). In contrast to the highly dynamic system observed for the M4-TMD in the liquid crystalline phase, in the gel phase the motional processes occurring on the microsecond timescale are significantly repressed. Although many plausible explanations exist for the reduction in overall mobility, the most likely would be the formation of higher oligomers of the M4-TMD or the restriction of peptide motion through the interaction of the peptide with the lipid environment. Complementary fluorescence experiments have shown that in the absence of cholesterol, the M4-TMD shows little tendency to oligomerize
(de Almeida et al. 2004). On the basis of these observations it is proposed that in the gel phase the interaction between the M4-TMD and its surrounding environment lipid is sufficiently enhanced such that the rate of peptide rotation is significantly reduced. However, from the studies presented here we are unable to determine the nature of these interactions. These observations are in agreement with previous studies of transmembrane peptides whereby entry into the gel-phase is sufficient to hinder the rate of peptide rotation (Smith et al. 1989, Smith et al. 1994). The increased interactions observed between M4-TMD and the surrounding lipids observed in the gel phase are accompanied by a change in the tilt of the M4-TMD within the lipid bilayer (see Antollini et al. 2005, this issue). This suggests that changes in both peptide orientation and lipid phase may play an important role in determining how the nAChR is modulated by its lipid environment. Conclusion These studies provide information about both the structure of the M4-TMD domain of the nAcChoR and how it interacts with the lipid environment. Analysis of isotropic chemical shifts and CD spectra indicate that the peptide adopts a predominantly a helical conformation along the segment which includes the three residues Leu11, Gly15 and Gly23. An analysis of the averaging of the anisotropic interactions in the solid state NMR spectra in liquid crystalline bilayers is corroborative of the presence of a linear a -helix spanning the region Ala8 to Gly23. This further suggests that the non-helical regions identified by CD-spectroscopy lie at the termini of the peptide where it exits the lipid bilayer. An analysis of the motional averaging observed in both DoMPC vesicles and DoMPC/DoHPC bicelles indicates that this linear a -helix is aligned within 158 of the membrane normal. Analysis of peptide rotation in both the gel and liquid crystalline phase clearly indicates that in the gel phase the interactions between the peptide and the surrounding lipids are sufficient to reduce the rate of peptide rotation such that it has negligible effects of the anisotropies observed in the solid state NMR spectra (B/1 kHz). These results are also reflected in the increased 31P CSA observed for the lipid headgroup in the liquid crystalline phase when the M4-TMD is present, which suggests interactions between the M4-TMD and the surrounding lipids plays a role in enhancing order in the headgroup region.
Structure and dynamics of the nAChRs 4th TMD Acknowledgements We are grateful to Marco Tomaselli, Matthias Ernst, Gerhard Gro¨bner and Boyan Bonev for many helpful discussions during the course of this work and Maurizio Zandomeneghi for his assistance with the CD spectroscopy. This work was supported in part by grants from the Universidad Nacional del Sur, the Agencia Nacional de Promocio´n Cientı´fica (FONCYT), Argentina, and FIRCA 1-RO3TOW1225-01 (NIH) to F.J.B., from HEFCE, BBSRC and MRC to A.W. and the Swiss National Science Foundation to P.T.F.W., G.Z. and B.H.M. References Baenziger JE, Morris ML, Darsaut TE, Ryan SE. 2000. Effect of membrane lipid composition on the conformational equilibria of the nicotinic acetylcholine receptor. J Biol Chem 275:777 / 784. Barrantes FJ. 1993. Structural and functional crosstalk between acetylcholine-receptor and its membrane environment. Mol Neurobiol 6:463 /482. Barrantes FJ. 2003. Transmembrane modulation of nicotinic acetylcholine receptor function. Curr Opin Drug Discov Dev 6:620 /632. Barrantes FJ. 2004. Structural basis for lipid modulation of nicotinic acetylcholine receptor function. Brain Res Reviews 47:71 /95. Barrantes FJ, Antollini S, Blanton MP, Prieto M. 2000. Topography of nicotinic acetylcholine receptor membrane embedded domains. J Biol Chem 275:37333 /37339. Belohorcova K, Qian J, Davis JH. 2000. Molecular dynamics and H-2-NMR study of the influence of an amphiphilic peptide on membrane order and dynamics. Biophys J 79:3201 /3216. Blanton MP, Cohen JB. 1994. Identifying the lipid /protein interface of the torpedo nicotinic acetylcholine-receptor / secondary structure implications. Biochemistry 33:2859 /2872. Blanton MP, Cohen JB. 1992. Mapping the lipid-exposed regions in the torpedo-californica nicotinic acetylcholine-receptor. Biochemistry 31:3738 /3750. Bolze J, Fujisawa T, Nagao T, Norisada K, Saito H, Naito A. 2000. Small angle X-ray scattering and 31P NMR studies on the phase behaviour of phospholipid bilayered mixed micelles. Chem Phys Lett 329:215 /220. Campagna JA, Miller KW, Forman SA. 2003. Mechanisms of action of inhaled anesthetics. N Engl J Med 348:2110 /2124. Cavagnero S, Jane Dyson H, Wright PE. 1999. Improved low pH bicelle system for orienting macromolecules over a wide temperature range. J Biomol NMR 13:387 /391. Changeux JP, Edelstein SJ. 1998. Allosteric receptors after 30 years. Neuron 21:959 /980. Corbin J, Methot N, Wang HH, Baenziger JE, Blanton MP. 1998. Secondary structure analysis of individual transmembrane segments of the nicotinic acetylcholine receptor by circular dichroism and Fourier transform infrared spectroscopy. J Biol Chem 273:771 /777. Cornell BA, Separovic F, Baldassi AJ, Smith R. 1988. Conformation and orientation of gramacidin-A in oriented phospholipid bilayers measured by solid state NMR. Biophys J 53:67 /76. Davis JH, Jeffrey KR, Bloom MM, Valic MI, Higgs TP. 1976. Quadrupolar echo deuterium magnetic resonance spectroscopy in ordered hydrocarbon chains. Chem Phys Lett 42:390 /394. de Almeida RFM, Loura LMS, Prieto M, Watts A, Fedorov A, Barrantes FJ. 2004. Cholesterol modulates the organization of
495
the gamma M4 transmembrane domain of the muscle nicotinic acetylcholine receptor. Biophys J 86:2261 /2272. de Planque MRR, Rijkers DTS, Liskamp RMJ, Separovic F. 2004. The alpha-M1 transmembrane segment of the nicotinic acetylcholine receptor interacts strongly with model membranes. Mag Reson Chem 42:148 /154. Fraser DM, Louro SRW, Horvath LI, Miller KW, Watts A. 1990. A study of the effect of general-anesthetics on lipid protein interactions in acetylcholine-receptor enriched membranes from torpedo-nobiliana using nitroxide spin-labels. Biochemistry 29:2664 /2669. Hartzell CJ, Pratum TK, Drobny GP. 1987. Mutual orientation of three magnetic tensors in polycrystalline dipeptide by dipolemodulated 15N chemical shift spectroscopy. J Chem Phys 87:4324 /4331. Hartzell CJ, Whitfield M, Oas TG, Drobny GP. 1987. Determination of 15N and 13C chemical shift tensors of L-[13C]alanyl-L-[15N]-alanine from dipolar coupled powder patterns. J Am Chem Soc 109:5966 /5969. Hediger S, Meier BH, Kurur ND, Bodenhausen G, Ernst RR. 1994. Nmr cross-polarization by Adiabatic Passage through the Hartmann /Hahn condition (APHH). Chem Phys Lett 223:283 /288. Jones J. 1997. Amino acid and peptide synthesis. Oxford: Oxford University Press. Kameda T, Ando I. 1997. The relationship between the helical conformation and 13C NMR chemical shift of amino acid residue carbonyl carbons of polypeptides in the solid state. J Mol Struct 412:197 /203. Lee DK, Wittebort RJ, Ramamoorthy A. 1998. Characterization of N-15 chemical shift and H-1-N-15 dipolar coupling interactions in a peptide bond of a uniaxially oriented and polycrystalline samples by one-dimensional dipolar chemical shift solid state NMR spectroscopy. J Am Chem Soc 120:8868 /8874. Lewis BA, Harbison GS, Herzfeld J, Griffin RG. 1985. NMR structural analysis of a membrane protein / bacteriorhodopsin peptide backbone orientation and motion. Biochemistry 24:4671 /4679. Live DH, Davis DG, Agosta WC, Cowburn D. 1984. Long-range hydrogen bond mediated effects in peptides-N-15 NMR study of gramacidin-S in water and organic solvents. J Am Chem Soc 106:1939 /1941. Lugovskoy AA, Maslennikov IV, Utkin YN, Tsetlin VI, Cohen JB, Arseniev A. 1998. Spatial structure of the M3 transmembrane segment of the nicotinic acetylcholine receptor alpha-subunit. Eur J Biochem 255:455 /461. Mantipragada SBL, Horvath LI, Arias HR, Schwarzmann G, Sandhoff K, Barrantes FJ, Marsh D. 2003. Lipid protein interactions and teh effect of local anaesthetics in acetylcholine receptor rich membranes from Torpedo marmorata electric organ. Biochemistry 42:9167 /9175. Marassi FM, Opella SJ, Juvvadi P, Merrifield RB. 1999. Orientation of cecropin A helices in phospholipid bilayers determined by solid-state NMR spectroscopy. Biophys J 77:3152 /3155. Marsh D, Barrantes FJ. 1978. Immobilized lipid in acetylcholine receptor-rich membranes from torpedo-marmorata. Proc Natl Acad Sci USA 75:4329 /4333. Mehring M. 1983. Principles of high-resolution NMR in solids. Berlin: Springer-Verlag. Miller KW. 2002. The nature of sites of general anaesthetic action. Br J Anaesthes 89:17 /21. Miyazawa A, Fujiyoshi Y, Unwin N. 2003. Structure and gating mechanism of the acetylcholine receptor pore. Nature 423:949 /955. Nagle JF, Tristran-Nagle S. 2000. Structure of lipid bilayers. Biochim Biophys Acta 1469:159 /195.
496
P. T. F. Williamson et al.
Nicholson LK, Teng Q, Cross TA. 1991. Solid state nuclear magnetic resonance derived model for dynamics in the polypeptide backbone of the gramacidin-A channel. J Mol Biol 218:621 /637. Opella SJ, Marassi FM, Gesell JJ, Valente AP, Kim Y, OblattMontal M, Montal M. 1999. Structures of the M2 channellining segments from nicotinic acetylcholine and NMDA receptors by NMR spectroscopy. Nature Struct Biol 6:374 / 379. Ottiger M, Bax A. 1999. Bicelle-based liquid crystals for NMR measurement of dipolar couplings at acidic and basic pH values. J Biomol NMR 13:187 /191. Provencher SW, Glockner J. 1981. Estimation of globular protein secondary structure from circular dichroism. Biochemistry 20:33 /37. Saito H. 1986. Conformation-dependent C-13 chemical-shifts-a new means of conformational characterization as obtained by high-resolution solid-state C-13 Nmr. Mag Res Chem 24:835 / 852. Sanders CR, Schwonek JP. 1992. Characterization of magnetically orientable bilayers in mixtures of dihexanoylphosphatidylcholine and dimyristoylphosphatidylcholine by solid state NMR. Biochemistry 31:8898 /8905. Separovic F, Pax R, Cornell BA. 1993. NMR Order parameter analysis of a peptide plane aligned in a lyotropic liquid crystal. Mol Phys 78:357 /369. Separovic F, Pax R, Cornell BA. 1993. NMR order parameter analysis of a peptide plane aligned in the lyotropic liquid crystal. Mol Phys 78:357 /369. Sharpe S, Barber KR, Grant CWM, Goodyear D, Morrow MR. 2002. Organization of model helical peptides in lipid bilayers: Insight into the behaviour of single span protein transmembrane domains. Biophys J 83:345 /358. Sixl F, Watts A. 1983. Headgroup interactions in mixed phospholipid bilayers. Proc Natl Acad Sci USA 80:1613 /1615. Smith R, Separovic F, Milne TJ, Whittaker A, Bennett FM, Cornell BA, Makriyannis A. 1994. Structure and orientation of pore-forming peptide, mellitin, in lipid bilayers. J Mol Biol 241:456 /466. Smith R, Thomas DE, Separovic F, Atkins AR, Cornell BA. 1989. Determination of the structure of a membrane incorporated ion channel. Biophys J 56:307 /314. Stark RE, Jelinski LW, Ruben DJ, Torchia DA, Griffin RG. 1983. 13C chemical shift and 13C-15N dipolar tensors for the
peptide bond [1-13C]glycyl[15N]glycine.HCl.H20. J Mag Res 55:266 /273. Sunshine C, McNamee MG. 1994. Lipid modulation of nioctinic acetylcholine receptor function: the role of membrane lipid composition and fluidity. Biochim Biophys Acta 1191:59 /64. Unwin N. 2000. The Croonian Lecture 2000. Nicotinic acetylcholine receptor and the structural basis of fast synaptic transmission. Phil Trans Roy Soc London Ser B. Biol Sci 355:1813 /1829. Unwin N. 2005. Refined structure of the nicotinic acetylcholine receptor at 4 angstrom resolution. J Mol Biol 346:967 /989. Vanstokim IHM, Spoelder HJW, Bloemendal M, VanGrondelle R, Groen FCA. 1990. Estimation of protein secondary structure and error analysis from circula-dichroism spectra. Anal Biochem 191:110 /118. Watts A. 1991. Magnetic resonance studies of lipid /protein interfaces and lipophilic molecule partitioning. Ann NY Acad Sci 625:653 /669. Wishart DS, Sykes BD. 1994. The C-13 chemical-shift index /a simple method for the identification of protein secondary structure using C-13 chemical-shift data. Journal of Biomolecular NMR 4:171 /180. Wu CH, Ramamoorthy A, Gierasch LM, Opella SJ. 1995. Simultaneous characterization of the amide H-1 chemical shift,H-1-N-15 dipolar, and N-15 chemical-shift interaction tensors in a peptide-bond by 3-dimensional solid-state NMRspectroscopy. J Am Chem Soc 117:6148 /6149. Yamaguchi S, Hong T, Waring A, Lehrer RI, Hong M. 2002. Solid state NMR investigations of peptide /lipid interaction and orientation of ss-sheet antimicrobial peptide, protegrin. Biochemistry 41:9852 /9862. Yamaguchi S, Huster D, Tack B, Kearney WR, Lehrer RI, Waring AJ, Hong M. 2001. Orientation and dynamics of an antimicrobial peptide by solid-state NMR. Biophys J 80:2421. Zandomeneghi G, Tomaselli M, Williamson PTF, Meier BH. 2003. NMR of bicelles: Orientation and mosaic spread of the liquid crystal director under sample rotation. J Biomol NMR 25:113 /123. Zhang H, Karlin A. 1998. Contribution of the beta subunit M2 segment to the ion conducting pathway of the acetylcholine receptor. Biochemistry 37:7952 /2964. Zhang H, Karlin A. 1997. Identification of acetylcholine receptor channel-lining residues in the M1 segment of the beta-subunit. Biochemistry 36:15856 /15864.
