nicotinic acetylcholine receptor-lipid interactions

Molecular Membrane Biology, 2002, 19, 277±284

Lipid matters: nicotinic acetylcholine receptor-lipid interactions (Review) Francisco J. Barrantes*

Instituto de Investigaciones BioquõÂmicas de BahõÂa Blanca and UNESCO Chair of Biophysics and Molecular Neurobiology, BZW8000 BahõÂa Blanca, Argentina. Summary Ligand-gate d ion channels mediate fast intercellular communication in response to endogenous neurotransmitters. The nicotinic acetylcholine receptor (AChR) is the archetype molecule in the superfamily of these membrane proteins. Early electron spin resonance studies led to the discovery of a lipid fraction in direct contact with the AChR, with rotational dynamics 50-fold slower than those of the bulk lipids. This AChR-vicinal lipid region has since been postulated to be a possible site of lipid modulation of receptor function . The polarity and molecular dynamics of solvent dipolesÐmainly waterÐof AChR-vicinal lipids in the membrane have been studied with Laurdan extrinsic fluorescence, and FoÈrster-type resonance energy transfer (FRET) was introduced to characterize the receptor-associate d lipid microenvironment. FRET enabled one to discriminate between the bulk lipid and the AChR-vicinal lipid. The latter is in a liquid-ordered phase and exhibits a higher degree of order than the bulk bilayer lipid. Changes in FRET efficiency induced by fatty acids, phospholipids and cholesterol also led to the identification of discrete sites for these lipids on the AChR protein. After delineating the topograph y of the AChR membrane-embedded domains with fluorescence methods, sites for steroids are being explored with site-directed mutagenesis and patch-clam p recording. Pyrene-labelled Cys residues in aM1, aM4, gM1 and gM4 transmembrane regions were found to lie in a shallow position. For M4 segments, this is in agreement with a canonical linear a-helix; for M1, it is necessary to postulate a substantial amount of non-helical structure, and/or of kinks, to rationalize the shallow location of Cys residues. Mutations of Thr 422, a residue close to the extracellular-facin g membrane hemilayer in aM4, affect the steroid modulation of AChR function , suggestin g its involvement in steroid-AChR interactions.

Keywords: Cholinergic receptor, membrane, lipid ± protein interactions, fluorescence, ion channels.

Introduction The ligand-gated ion channels (LGIC) are a set of membrane proteins coded by a few hundred genes so far identified (Barrantes 1998). Within the LGIC superfamily, the nicotinic acetylcholine receptor (AChR) and the subtype 3 of the serotonin (5-HT3) receptor comprise two families of cationselective channels, whilst glycine and gamma aminobutyric acid type A (GABAA ) receptors are anion-selective channels. The several genes coding for AChR sub-units exhibit amino acid sequence homology and presumably higher-order structural motifs. The a1 , b1, g, and d sub-units are present in foetal skeletal muscle, BC3H-1 muscle cells and electric fish electrocytes and 12 different types of vertebrate *To whom correspondence should be addressed. e-mail: [email protected]

neuronal AChR sub-units: a2 ± a10 and b2 ± b4 have been isolated from the central nervous system. Signal transduction in the cholinergic synapse results from relatively fast and similarly simple mechanistic steps: binding of the neurotransmitter followed by conformational transitions in the receptor proteins that lead to changes in the ionic permeability of the post-synaptic membrane. Upon binding, acetylcholine (ACh) initiates a conformational change in this protein that triggers the transient opening of its intrinsic cation-specific channel across the post-synaptic membrane. At the molecular level, this is accomplished by the concerted action of an array of four different but highly homologous AChR sub-units in the stoichiometry a2bg(or e)d (Lindstrom et al. 1979, Raftery et al. 1980, Karlin 2002). Each AChR sub-unit contains four putative hydrophobic segments, 20 ± 30 amino acids in length, referred to as M1 ± M4 and proposed to be membrane-spanning segments (figure 1). The M2 segment from each sub-unit is thought to contribute structurally to form the ion channel proper. The M4 segment was proposed as the most likely candidate to be exposed to the bilayer lipid (Giraudat et al. 1985, Tobimatsu et al. 1987). M1 and M3 effectively incorporate membranepartitioning photoactivatable probes (Blanton and Wang 1991, Blanton and Cohen 1992, 1994) and are likely to be exposed to the lipid phase as well. It is usually accepted that the four hydrophobic segments M1 ± M4 correspond to regions of the protein fully embedded in the membrane, and are referred to as transmembrane (TM) domains (Chavez and Hall 1992, Blanton and Cohen 1994). As to their secondary structure, the original postulation of an all (four)-helix bundle (Noda et al. 1983) has been challenged by the results of cryoelectron microscopy of frozen AChR tubules (Unwin 1995) and computer-aided molecular modelling indicating that the dimensions of the AChR TM region are not compatible with a pentameric four-helix bundle (Ortells et al. 1998). Only a very limited number of high-resolution structures of membrane proteins have been obtained to date using classical X-ray crystallography methods, mainly because of the difficulty of obtaining large well-ordered three-dimensional crystals. Two-dimensional arrays adequate for electron diffraction techniques are available for a few membrane proteins. The crystal structure of a water-soluble AChbinding protein from a snail has appeared very recently (Brejc et al. 2001) (figure 1). This soluble AChR-like protein is released into the synaptic region in a mollusc cholinergic synapse in response to the natural agonist ACh. The protein traps ACh and, thus, suppresses synaptic transmission. The release of this decoy glial receptor protein into the synaptic cleft provides a novel mechanism by which glial cells can modulate the efficacy of cholinergic neurotransmission. The structure of this protein, highly homologous to the watersoluble extracellular domain of the AChR protein proper (Brejc et al. 2001), provides the first high-resolution views of the putative region of the AChR involved in agonist

Molecular Membrane Biology ISSN 0968-7688 print/ISSN 1464-5203 online # 2002 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/09687680210166226

278

F. J. Barrantes

Figure 1. Molecular models of the actual crystal structure of the ACh-binding protein obtained by Brejc et al. (2001) using X-ray diffraction (AChBP, upper panel), and the hypothetical AChR transmembrane region (lower panel). The primary structure of the AChBP is highly homologous to the extracellular region of the AChR, and, thus, could provide a useful framework to learn about the ligand-binding domain of the AChR. The water-soluble AChBP is released into the synaptic region in response to the natural agonist, ACh, and has been postulated to act as a decoy molecule to suppress synaptic transmission. Atomic coordinates of the AChBP are deposited at the Brookhaven Protein Data Bank with accession number 1I9B. The two 3-D models were drawn using the software WebLab Viewer from Molecular Simulations, Inc. The lower panel shows a hypothetical representation of the AChR transmembrane region (coordinates were obtained from Dr Marcelo Ortells, see also Ortells et al. (1997)). The model reproduces faithfully the overall shape and dimensions of Unwin’s (1995) medium-resolution electron microscopy images of the AChR transmembrane domain.

recognition, the first step in the cascade leading to channel opening in the cholinergic synapse. This exciting development is bound to lead to new insights into the mechanisms of fast signal transduction in general. Meanwhile, in the absence of detailed information from X-ray or electron crystallographic data, drawing structural-functional correlations of many important channel-forming membrane proteins has relied on the powerful combination of site-directed mutagenesis and patch-clamp electrophysiology and various spectroscopic techniques applied to membrane fragments and purified, reconstituted proteins. Site-directed mutagenesis data combined with patchclamp electrophysiology and results from photoaffinity labelling with non-competitive channel blockers support the notion that the M2 domain lines the walls of the pore and are

indicative of a-helical periodicity in the residues exposed to the lumen of the AChR channel (Changeux and Edelstein 1998). NMR studies of the M2 segment of the d sub-unit (Opella et al. 1999) indicate that this domain is inserted in the bilayer at an angle of 128 relative to the membrane normal, in a totally a-helical configuration. A synthetic peptide corresponding to the Torpedo aM2 segment in chloroform:methanol containing LiClO4 also adopts a totally a-helical configuration (Pashkov et al. 1999). Cryoelectron microscopy has so far revealed a relatively featureless appearance of the other putative TM domains (M1, M3 and M4). A large portion of this AChR region is postulated to be arranged in the form of a b-barrel outside the central rim of M2 channel-forming rods (Unwin 1995). This interpretation is in contrast with photoaffinity labelling

Lipid-acetylcholine receptor crosstalk

279

studies, in which the observed periodicity of the lipid-exposed residues in M4 and M3 is consistent with an a-helical pattern (Blanton and Cohen 1992, 1994, Blanton et al. 1998) and with deuterium-exchange Fourier transform infrared spectroscopy studies which indicate a predominantly a-helical structure in the AChR TM region (Baenziger and MeÂthot 1996). In addition, secondary structure analysis (CD and Fourier transform infrared spectroscopy) of isolated and lipidreconstituted TM AChR peptides indicates a-helical structure for M2, M3, and M4 segments (Corbin et al. 1998a). Furthermore, a synthetic peptide corresponding to the aM3 segment of Torpedo AChR exhibited a totally a-helical 1 structure by 2-dimensional H-NMR spectroscopy (Lugovskoy et al. 1998) and a recent NMR study of a synthetic gM4 peptide is also compatible with an a-helical secondary structure (Williamson et al. 2000). Another area of research in which protein-AChR interactions might be involved is that of the stabilization of receptor aggregates or `clusters’. Prior to innervation, the muscle AChR is uniformly distributed on the cell surface. In the course of synaptogenesis, the AChR becomes clustered first in the form of `hot spots’ and, finally, upon establishment of the neuromuscular junction, receptor molecules adopt the high-density packing characteristic of the adult motor endplate synapse. It is not known whether the different organizational modes of the AChR protein are accompanied by large-scale modifications of their lipid environment. It is known, however, that the closely associated lipid belt region surrounding the AChR protein in the electrocyte is qualitatively different in terms of molecular mobility (Marsh and Barrantes 1978), and the lipid composition of the synaptic region has been reported to differ from that of the rest of the membrane in muscle cells (Bloch and Morrow 1989). This review evaluates the impact of some biophysical studies of the lipid-AChR interactions on current knowledge of structural and dynamic properties of the AChR molecule in its membrane environment.

for the correct functioning of the AChR, but the details of the interaction were still obscure.

Early studies on AChR lipid microenvironment define AChR-vicinal lipid

Laurdan GP in living cells: physical state of membrane lipids correlate with energetics of channel gating

Early electron spin resonance (ESR) studies by Marsh and Barrantes (1978) and Marsh et al. (1981) made apparent the occurrence of direct interactions between lipid and AChR in its native membrane environment. This series of studies demonstrated that the protein-vicinal or `annular’ lipids were relatively immobile with respect to the rest of the membrane lipids, a finding confirmed by other laboratories (Rousselet et al. 1979, Ellena et al. 1983) using AChR reconstituted into lipid systems. The need to include sterols and certain phospholipids to preserve AChR functionality in reconstituted systems was subsequently demonstrated (Epstein and Racker 1978). The different contributions of phospholipid and sterol could be established in various studies in vitro (Criado et al. 1982, 1984, Ochoa et al. 1983) and the minimal number of lipid molecules (*45) per AChR was ascertained in ESR experiments (Ellena et al. 1983, Jones et al. 1988). AChRassociated (annular) lipids were thus shown to be important

The characterization of Laurdan thermotropic behaviour was subsequently extended to living mammalian cells expressing endogenous or heterologous AChR. Interestingly, the differences in physical properties of cell membranes measured by Laurdan GP in a variety of cells, reflecting the molecular dynamics of water molecules, could be correlated with the energetic changes in the AChR ion channel occurring as a function of temperature, as measured in single-channel recordings (Zanello et al. 1996). Laurdan expends energy in solvent (water) reorientation, as evidenced in the red shift of its emission spectrum. The decrease in Laurdan GP upon increasing the temperature reflects the increase in water diffusion into the membrane. Water diffusion is facilitated by the increased thermal-induced disorder in the bilayer lipid. The higher AChR channel conductance upon increasing the temperature is a manifestation of the augmented ion and water permeability in the AChR channel or `pore’ region as was observed in single-channel recordings with the patch-

Laurdan extrinsic fluorescence studies of the AChR lipid microenvironment The so-called Generalized Polarization (`GP’) (Parasassi et al. 1991, Zanello et al. 1996) of the fluorescent probe Laurdan (6dodecanoyl-2-dimethylamino naphthalene) has been used to learn about the dynamics of the AChR and some physical properties of the protein-vicinal lipid (Antollini et al. 1996, Antollini and Barrantes 1998) (figure 2). Towards this end, a hitherto unexploited property of Laurdan was introduced, namely its ability to act as a FoÈrster-type resonance energy transfer (FRET) acceptor of tryptophan emission. Laurdan is a particularly advantageous fluorescent probe; its spectral properties are extremely sensitive to the polarity and to the molecular dynamics of dipoles in its environment. This is due to dipolar relaxation processes that are reflected as relatively large spectral shifts. Laurdan GP reports on the polarity and dynamics of the solvent dipoles in its surroundings. The main dipoles sensed by Laurdan in the membrane are water molecules. When no relaxation occurs, high GP values result, indicative of low water content in the hydrophilic/hydrophobic interface region of the membrane. In particular, the temperature dependence of Laurdan GP has been utilized to describe and to quantify the phase states of membrane systems. Moreover, by studying the behavior of Laurdan fluorescence under different excitation/emission wavelengths, it has been possible to characterize qualitatively the phase state of native and artificial membrane systems (Parasassi et al. 1991). Studies of Laurdan GP showed that the AChR-lipid interface region, i.e. the AChR-vicinal lipid, exhibits a higher rigidity than the rest of the membrane bilayer, although a single thermotropic phase with the characteristics of the socalled liquid-ordered phase defines the AChR-rich postsynaptic membrane of fish electrocytes (Antollini et al. 1996).

280

F. J. Barrantes

Figure 2. Molecular model (left) of the environmentally sensitive fluorescent probe Laurdan (6-dodecanoyl-2-dimethylamino naphthalene) and rationale of the so-called General Polarization (GP) of Laurdan (Parasassi et al. 1991), given by GP = (IB - IR)/ (IB + IR). The simple algorithm combines the Laurdan fluorescence intensities at the blue (IB, ca. 440 nm) and red (IR, ca. 490 nm) edges of the emission spectrum, respectively. Spectra in pure gel and fluid phases are shown in blue and red, respectively, in the on-line version of this paper, and in black and grey in the printed Journal. The environmentally sensitive probe Laurdan senses the solvent (water) dipoles in the membrane, yielding high GP values in the gel phase and low GP values in the fluid phase. One has introduced the use of Laurdan in FRET experiments, exploiting its ability to form an ideal acceptor of intrinsic protein fluorescence (Antollini et al. 1996, Antollini and Barrantes 1998). This enabled one to explore the physical state of the lipids in the immediate vicinity of the receptor protein (Antollini et al. 1996, Zanello et al. 1996, Antollini and Barrantes 1998). GP values observed under energy transfer conditions exhibit higher absolute values than those obtained by direct excitation of the probe, indicating the lower polarity and higher `rigidity’ of the lipids in the microenvironment of the AChR.

clamp technique. This study (Zanello et al. 1996) indicates that AChR channel kinetics depend not only on intrinsic properties of the AChR protein but also on the physical state of the membrane in which the receptor is embedded. Lipids modulate the fluidity of the AChR microenvironment Studies from Fong and McNamee (1987) and Sunshine and McNamee (1994) show that reconstituted AChR is sensitive to the bulk fluidity of the membrane, as measured by ion flux experiments. When Laurdan GP was measured in Torpedo native AChR membrane by direct excitation or under FRET conditions in the presence of exogenous lipids, GP, and, by inference, the `fluidity’ and order of the membrane were found to diminish upon addition of oleic acid and DOPC, and not to vary significantly upon addition of cholesterol hemi-

succinate, indicating an increase in the polarity of the single, ordered-liquid lipid phase in the two former cases (Antollini and Barrantes 1998). Complementary information about the bulk lipid order was obtained from measurements of fluorescence anisotropy of DPH and two of its derivatives. The membrane order diminished in the presence of oleic acid and DOPC. The location of Laurdan was determined using the parallax method of Chattopadhyay and London (1987). Their method is based on the relative position of a fluorescence probe embedded in the membrane and its quenching by probes having nitroxide spin labels at different positions along their acyl chains. The parallax determination is accomplished by pairwise comparison of quenching parameters with different pairwise combinations of the PC analogues with spin labels in carbons 7, 10 and 12. Using this approach, Laurdan was found to lie at *10 AÊ from the centre of the bilayer, i.e. at a

Lipid-acetylcholine receptor crosstalk depth of *5 AÊ from the lipid ± water interface (Antollini and Barrantes 1998). In previous work from the laboratory, a minimum donor ± acceptor distance of 14+1 AÊ was calculated for the Laurdan-AChR pair in the Torpedo native membrane (Antollini et al. 1996), in agreement with the overall dimensions of the AChR and its transmembrane region in particular (Barrantes 1998). Lipid sites at the lipid-AChR interface Fatty acids (Andreasen et al. 1979, Andreasen and McNamee 1980, Bouzat and Barrantes 1993a), phospholipids containing short-chain fatty acids (Braun and Haydon 1991), and steroids (Bouzat and Barrantes 1993b, 1993c, 1996) have been shown to exert effects on AChR function. Fatty acids and dioleylphosphatidylcholine decrease GP of bulk and lipid belt-regions in the native AChR membrane, whereas a cholesterol ester does not. This is in agreement with fluorescence quenching studies with brominated lipids that identified annular sites for phospholipids and nonannular sites for cholesterol, both accessible to fatty acids (Jones and McNamee 1988, Narayanaswami and McNamee 1993). Subsequently, a new strategy was introduced for determining lipid-AChR interactions in AChR-rich membranes (Antollini and Barrantes 1998). The approach consists in measuring the decrease in the energy transfer efficiency (E) between the intrinsic fluorescence of AChR-rich membranes and Laurdan, induced by different lipids. E between AChR (donor) and Laurdan (acceptor) was calculated using equations (1) and (2). According to FoÈrster’s (1948) theory, E is given by: E ˆ Ro6 =…Ro6 ‡ r 6 †

… 1†

where r is the intermolecular distance and Ro is a constant parameter for each donor ± acceptor pair, defined as the distance at which E is 50%. E can also be calculated as: E ˆ 1 ¡ …f=fD † º 1 ¡ …I=ID †

… 2†

where f and fD are the fluorescence quantum yields of donor in the presence and absence of the acceptor, respectively, and I and ID are the corresponding emission intensities. FRET efficiency was found to decrease upon addition of exogenous lipids, which displace Laurdan molecules from the AChRmicroenvironment. The maximal decrease in E resulting from the addition of a fatty acid (18:1) amounted to *60%, whereas cholesterol or phospholipid reduced E by 35% and 25%, respectively. The sum of the decreases caused by DOPC and CHS equaled that obtained in the presence of 18:1 alone. From this series of experiments, one reached the conclusion that there are independent sites for phospholipid and sterol, both accessible to fatty acid, in the vicinity of the AChR (Antollini and Barrantes 1998). Topography of AChR membrane-embedded domains One has recently prepared transmembrane (TM) peptides from purified Torpedo californica AChR protein by controlled enzymatic digestion, and derivatized with N-(1-pyrenyl)maleimide (PM), purified and reconstituted these peptides into

281

asolectin liposomes (Barrantes et al. 2000). PM labelling occurred in cysteine residues in aM1, aM4, gM1 and gM4. The labelled cysteine residues are in all cases exposed to the surrounding lipid and located at the `cytoplasmic’ end of the AChR TM region, in fact very close to the bilayer ± water interface. The topography of the pyrene-labelled Cys residues with respect to the membrane was experimentally determined by differential fluorescence quenching with spin-labelled derivatives of fatty acids, phosphatidylcholine and the steroids cholestane and androstane. Stern-Volmer plots of whole AChR quenching by spin-labelled lipid analogues showed no deviation from linearity. Stearic acid and androstane spin label derivatives were the most effective quenchers of the pyrene fluorescence of whole AChR and derived TM peptides. In the case of spin-labelled stearic acid derivatives, the 5-SASL isomer quenched more effectively than the 7SASL and the 12-SASL analogues, indicating a shallow location of the pyrene-labelled Cys residues. In fact, the quenching studies were compatible with the occurrence of shallow positions for all labelled Cys residues in the AChR and derived TM peptides. In the case of gM4 (see figure 3) and, by inference in other M4 segments of the AChR, this is compatible with a linear a-helical structure (Barrantes et al. 2000). In the case of aM1, `classical’ models 230 locate Cys at the centre of the bilayer in an extended ahelical structure. This is not compatible with the shallow 230 location of pyrene-labelled Cys emerging from the experimental results. The transmembrane topography of M1 can be explained on the basis of the presence of a substantial amount of non-helical structure, and/or of kinks along M1. In a mixed a-helix/b-sheet model of the AChR (Ortells and Lunt 1996), aM1 was constructed as a three-strand b-sheet interrupted by short loops generated by searching in the database of known structures for an appropriate backbone conformation. The proline residues themselves cannot be found within a b-strand, so they were positioned in the loops. The same model can be extended to gM1, having a proline 229 230 residue (Pro ) immediately adjacent to Cys and two 222 244 other (Pro and Pro ) at the end of the TM region. Thus, the conserved proline residues in the M1 segments of the AChR might introduce `curls’ or kinks in a manner analogous to that recently reported for one of the transmembrane + segments of a K channel (del Camino et al. 2000). In the case of the gM1 fragment there appears to be no detailed information on its topography in the membrane as yet, but the occurrence of helix-interrupting proline residues is a striking feature of M1 in the AChR and all members of the rapid ligand-gated ion channel superfamily. Putative steroid sites at the AChR transmembrane region Different steroids affect AChR channel kinetics in a qualitatively similar manner, albeit to different extents, following an inverse relationship between their lipophilicity and inhibitory effect (Garbus et al. 2001). One general conclusion from studies on steroids is that these compounds do not affect the membrane in a global and unspecific manner, but rather that

282

F. J. Barrantes

422

Figure 3. Molecular stick model of the mouse AChR aM4 transmembrane region outlining the position of Thr , close to the extracellular side of the membrane, as a putative target of steroid action (left). Complementary site-directed mutagenesis and electrophysiological data (Garbus et al. 2002) lend support to this suggestion. The surface potential projection of aM4 is shown on the right, with darker red and blue levels in the on-line version of this paper, and darker levels in the printed Journal, corresponding to higher values of calculated surface potentials. The molecular 422 surface displays a crevice or cavity at the level of Thr (arrows). The hypothetical model of the aM4 was drawn using the theoretical modelling coordinates employed in Ortells et al. (1997) and the software WebLab Viewer (surface potential, solvent-accessibility rendering mode) from Molecular Simulations, Inc.

the effect exerted by each depends on the degree of access to sites at the AChR-lipid interface (Barrantes et al. 2000). Photoaffinity labelling experiments with hydrophobic probes such as TID (Blanton and Wang 1991, Blanton and Cohen 1992, 1994), the cholesterol analogue azido-cholesterol (Corbin et al. 1988b), the steroid derivative promegestone (Blanton et al. 1999) and a phosphatidylserine arylazido derivative (Blanton and Wang 1991) all support the view that the transmembrane segment M4 is in contact with membrane lipids. In Torpedo AChR all these probes labelled amino acid residues in all four sub-units, principally in transmembrane domains M1, M3 and M4 (Blanton and Wang 1991, Blanton and Cohen 1992, 1994, Corbin et al. 1998, Blanton et al. 1999). The periodicity of the labelling in aM4 suggests an a-helical secondary structure (Blanton and Cohen 1992, 1994). More recent fluorescent spectroscopy studies from the laboratory are compatible with an a-helical structure in aM4 and gM4, whereas aM1 and possibly gM1

possess non-helical structure (Barrantes et al. 2000, Barrantes 2001). In a photoaffinity study with a synthetic progestin, 17,21dimethyl-19-nor-pregn-4,9-diene-3,20-dione (Promegestone) probe, labelling was observed in some of the Torpedo 412 AChR residues labelled with TID, predominantly in Cys (Blanton et al. 1999). Given the steroidal nature of the probe, 412 these results suggested that Cys could be involved in the steroid-AChR interaction in general. Based on the photoaffinity labelling of a412, this position was mutated to different residues to explore the properties of side-chain volume, hydrophobicity and charge on AChR-steroid interactions (Garbus et al. 2002). An alanine ± substituted quadruple mutant of four putative lipid-exposed residues in aM4 (Leu411 , Met415 , Cys418 and Thr422 ) exhibited less inhibition by steroid than that observed in wild-type AChR. When one dissected the quadruple mutant into four individual 422 alanine-substituted receptors, it was found that the Thr

Lipid-acetylcholine receptor crosstalk mutant AChR behaved like the quadruple mutant, whereas the other three were indistinguishable from the wild-type. It 422 was concluded that Thr , a residue close to the extracellular-facing membrane hemilayer in aM4, has direct bearing on the changes in hydrocortisone sensitivity and propose its involvement in the steroid-AChR interaction site 422 (Garbus et al. 2002). Figure 3 shows the position of Thr in aM4, close to the extracellular side of the membrane, as a putative target of steroid action.

Acknowledgements The experimental work quoted in this review was supported by grants from the Universidad Nacional del Sur, the Agencia Nacional de PromocioÂn CientõÂfica, Argentina, FIRCA 1-RO3-TW01225-01 and Antorchas/ British Council to F.J.B.

References Andreasen, T.J. and McNamee, M.G.,1980, Inhibition of ion permeability control properties of acetylcholine receptor from Torpedo californica by long-chain fatty acids. Biochemistry, 19, 4719 ± 4726. Andreasen, T.J., Doerge, D.R. and McNamee, M.G., 1979, Effects of phospholipase A2 on the binding and ion permeability control properties of the acetylcholine receptor. Archives of Biochemistry and Biophysics, 194, 468 ± 480. Antollini, S.S. and Barrantes, F.J., 1998, Disclosure of discrete sites for different lipids at the protein-lipid interface in native acetylcholine receptor-rich membrane. Biochemistry, 37, 16653 ± 16662. Antollini, S.S., Soto, M.A., Bonini de Romanelli, I., Gutierrez-Merino, C., Sotomayor, P. and Barrantes, F.J., 1996, Physical state of bulk and protein-associated lipid in nicotinic acetylcholine receptor-rich membrane studied by Laurdan generalized polarization and fluorescence energy transfer. Biophysics Journal, 70, 1275 ± 1284. Baenziger, J. E. and MeÂthot, N., 1996, Fourier transform infrared and hydrogen/deuterium exchange reveal an exchange-resistant core of a-helical peptide hydrogens in the nicotinic acetylcholine receptor. Journal of Biological Chemistry, 270, 29129 ± 29137. Barrantes, F.J., 1998, The Nicotinic Acetylcholine Receptor: Current Views and Future Trends, (Berlin/Heidelberg: Springer Verlag and Georgetown, TX: Landes Publishing Co.). Barrantes, F. J., 2001, Fluorescence studies of the acetylcholine receptor: Structure and dynamics in the membrane environment. Journal of Fluorescence, 11, 273 ± 285. Barrantes, F.J., Antollini, S.S.,Blanton, M.P. and Prieto, M., 2000, Topography of nicotinic acetylcholine receptor membraneembedded domains. Journal of Biological Chemistry, 275, 37333 ± 37339. Barrantes, F.J., Antollini, S.S., Bouzat, C.B., Garbus, I. and Massol, R., 2000, Nongenomic effects of steroids on the nicotinic acetylcholine receptor. Kidney International, 57, 1382 ± 1389. Blanton, M.P. and Cohen, J.B., 1992, Mapping the lipid-exposed regions in the Torpedo californica nicotinic acetylcholine receptor. Biochemistry, 31, 3738 ± 3750. Blanton, M.P. and Cohen. J.B., 1994, Identifying the lipid-protein interface of the Torpedo nicotinic acetylcholine receptor: secondary structure implications. Biochemistry, 33, 2859 ± 2872. Blanton, M.P. and Wang, H.H., 1991, Localization of regions of the Torpedo californica nicotinic acetylcholine receptor labeled with an aryl azide derivative of phosphatidylserine. Biochimica et Biophysica Acta, 1067, 1 ± 8.

283

Blanton, M.P., McCardy, E.A., Huggins, A. and Parikh, D., 1998, Probing the structure of the nicotinic acetylcholine receptor with 125 the hydrophobic photoreactive probes [ I]TID-BE and 125 [ I]TIDPC/16. Biochemistry, 37, 14545 ± 14555. Blanton, M.P., Xie, Y., Dangott, L.J. and Cohen, J.B., 1999, The steroid promegestone is a noncompetitive antagonist of the Torpedo nicotinic acetylcholine receptor that interacts with the lipid-protein interface. Molecular Pharmacology, 55, 269 ± 278. Bloch, R.J. and Morrow, J.S., 1989, An unusual b-spectrin associated with clustered acetylcholine receptors. Journal of Cell Biology, 108, 481 ± 494. Bouzat, C.B. and Barrantes, F.J., 1993a, Effects of long-chain fatty acids on the channel activity of the nicotinic acetylcholine receptor. Receptors and Channels, 1, 251 ± 258. Bouzat, C.B. and Barrantes, F.J., 1993b, Hydrocortisone and 11desoxycortisone modify acetylcholine receptor channel gating. NeuroReport, 4, 143 ± 146. Bouzat, C.B. and Barrantes, F.J., 1993c, Acute exposure of nicotinic acetylcholine receptor to the synthetic glucocorticoid dexamethasone alters single-channel gating properties. Molecular Neuropharmacy, 3, 109 ± 116. Bouzat, C.B. and Barrantes, F.J., 1996, Modulation of muscle nicotinic acetylcholine receptors by the glucocorticoid hydrocortisone. Possible allosteric mechanism of channel blockade. Journal of Biological Chemistry, 271, 25835 ± 25841. Braun, M.S. and Haydon, D.A., 1991, The effects of short-chain phospholipids on the acetylcholine-activated ion channel. Pflugers Archives, 418, 62 ± 67. Brejc, K., van Dijk, W.J., Klaassen, R.V., Schuurmans, M., van der Oost, J., Smit, A.B. and Sixma, T.K., 2001, Crystal structure of an ACh-binding protein reveals the ligand-binding domain of nicotinic receptors. Nature, 411, 269 ± 276. Changeux, J.-P. and Edelstein, S.J., 1998, Allosteric receptors after 30 years. Neuron, 21, 959 ± 980. Chattopadhyay, A. and London, E., 1987, Parallax method for direct measurement of membrane penetration depth utilizing fluorescence quenching by spin-labeled phospholipids. Biochemistry, 26, 39 ± 45. Chavez, R.A. and Hall, Z.W., 1992, Expression of fusion proteins of the nicotinic acetylcholine receptor from mammalian muscle identifies the membrane-spanning regions in the a and d subunits. Journal of Cell Biology, 116, 385 ± 393. Corbin, J., MeÂthot, N., Wang, H.H., Baenziger, J.E. and Blanton, M.P., 1998a, Secondary structure analysis of individual transmembrane segments of the nicotinic acetylcholine receptor by circular dichroism and fourier transform infrared spectroscopy. Journal of Biological Chemistry, 273, 771 ± 777. Corbin, J., Wang, H.H., Blanton, M.P., 1998b, Identifying the cholesterol binding domain in the nicotinic acetylcholine 125 receptor with [ I]azido-cholesterol. Biochimica et Biophysica Acta, 1414, 65 ± 74. Criado, M., Eibl, H. and Barrantes, F.J., 1982, Effects of lipids on acetylcholine receptor. Essential need of cholesterol for the maintenance of agonist-induced state transitions in lipid vesicles. Biochemistry, 21, 3622 ± 3629. Criado, M., Eibl, H. and Barrantes, F.J., 1984, Functional properties of the acetylcholine receptor incorporated in model lipid membranes. Differential effects of chain length and head group of phospholipids on receptor affinity states and receptormediated ion translocation. Journal of Biological Chemistry, 259, 9188 ± 9198. del Camino, D., Holmgren, M., Liu, Y. and Yellen, G., 2000, Blocker + protection in the pore of a voltage-gated K channel and its structural implications. Nature, 403, 321 ± 325. Ellena, J.F., Blazing, M.A. and McNamee. M.G., 1983, Lipid-protein interactions in reconstituted membranes containing acetylcholine receptor. Biochemistry, 22, 5523 ± 5535. Epstein, M. and Racker, E., 1978, Reconstitution of carbamylcholinedependent sodium ion flux and desensitization of the acetylcholine receptor from Torpedo californica. Journal of Biological Chemistry, 253, 6660 ± 6662.

284

F. J. Barrantes

Fong, T.M. and McNamee, M.G., 1987, Stabilization of acetylcholine receptor secondary structure by cholesterol and negatively charged phospholipids in membranes. Biochemistry, 26, 3871 ± 3880. FoÈrster, Th., 1948, Intermolecular energy migration and fluorescence. Annals of Physics (Leipzig), 2, 55 ± 75. Garbus, I., Bouzat, C. and Barrantes, F.J., 2001, Steroids differentially inhibit the nicotinic acetylcholine receptor. NeuroReport, 12, 227 ± 231. Garbus, I., Roccamo, A.M. and Barrantes, F. J., 2002, Identification of threonine 422 in transmembrane domain aM4 of the nicotinic acetylcholine receptor as a possible site of interaction with hydrocortisone. Neuropharmacology, 43, 65 ± 73. Giraudat, J., Montecucco, C., Brisson, R. and Changeux, J.-P. , 1985, Transmembrane topology of acetylcholine receptor subunits probed with photoreactive phospholipids. Biochemistry, 24, 3121 ± 3127. Jones, O.T. and McNamee, M.G., 1988, Annular and nonannular binding sites for cholesterol associated with the nicotinic acetylcholine receptor. Biochemistry, 27, 2364 ± 2374. Jones, O.T., Eubanks, J.H., Earnest, J.P. and McNamee, M.G., 1988, A minimum number of lipids are required to support the functional properties of the nicotinic acetylcholine receptor. Biochemistry, 27, 3733 ± 3742. Karlin, A., 2002, Emerging structure of the nicotinic acetylcholine receptors. Nature Reviews in Neuroscience, 3, 102 ± 114. Lindstrom, J., Merlie, J. P. and Yogeeswaram, G., 1979, Biochemical properties of acteylcholine receptor subunits from Torpedo californica. Biochemistry, 18, 4465 ± 4470. Lugovskoy, A.A., Maslennikov, I.V., Utkin, Y.N., Tsetlin, V.I., Cohen, J.B. and Arseniev, A.S., 1998, Spatial structure of the M3 transmembrane segment of the nicotinic acetylcholine receptor a subunit. European Journal of Biochemistry, 255, 455 ± 461. Marsh, D. and Barrantes, F.J., 1978, Immobilized lipid in acetylcholine receptor-rich membranes from T. marmorata. Proceedings of the National Academy of Sciences (USA), 75, 4329 ± 4333. Marsh, D., Watts, A. and Barrantes, F.J., 1981, Phospholipid chain immobilization and steroid rotational immobilization in acetylcholine receptor-rich membranes from Torpedo marmorata. Biochimica et Biophysica Acta, 645, 97 ± 101. Narayanaswami, V. and McNamee, M.G., 1993, Protein-lipid interactions and Torpedo californica nicotinic acetylcholine receptor function. 2. Membrane fluidity and ligand-mediated alteration in the accessibility of gamma subunit cysteine residues to cholesterol. Biochemistry, 32, 12420 ± 12427. Noda, M., Takahashi, H., Tanabe, T., Toyosato, M., Kikyotani, S., Hirose, T., Asai, M., Takashima, H., Inayama, S., Miyata, T. and Numa, S., 1983, Primary structure of a-subunit precursor of Torpedo californica acetylcholine receptor deduced from cDNA sequence. Nature, 299, 793 ± 797. Ochoa, E.L., Dalziel, A.W. and McNamee, M.G., 1983, Reconstitution of acetylcholine receptor function in lipid vesicles of defined composition. Biochimica et Biophysica Acta, 727, 151 ± 162.

Opella, S.J., Marassi, F.M., Gesell, J.J., Valente, A.P., Kim, Y., Oblatt-Montal, M. and Montal, M., 1999, Structures of the M2 channel-lining segments from nicotinic acetylcholine and NMDA receptors by NMR spectroscopy. Nature Structural Biology, 6, 374 ± 379. Ortells, M.O. and Lunt, G.G., 1996, A mixed helix-b-sheet model for the transmembrane region of the nicotinic acetylcholine receptor. Protein Engineering., 9, 51 ± 59. Ortells, M.O., Barrantes, G.E. and Barrantes, F.J.,1998, Molecular modelling of the nicotinic acetylcholine receptor. In The nicotinic acetylcholine receptor: Current views and future trends, F.J. Barrantes, ed. (Berlin/Heidelberg: Springer Verlag and Georgetown: Landes Publishing Co.), pp. 85 ± 108. Ortells, M.O., Barrantes, G.E., Wood, C., Lunt, G.G. and Barrantes, F.J., 1997, Molecular modelling of the nicotinic acetylcholine receptor transmembrane region in the open state. Protein Engineering, 10, 511 ± 517. Parasassi, T., De Stasio, G., Ravagnan, G., Rusch, R.M. and Gratton,E., 1991, Quantitation of lipid phases in phospholipid vesicles by the generalized polarization of Laurdan fluorescence. Biophysics Journal, 60, 179 ± 189. Pashkov, V.S., Maslenikov, I.V., Tchikin, L.D., Efremov, R.G., Ivanov, V.T. and Arseniev, A.S., 1999, Spatial structure of the M2 transmembrane segment of the nicotinic acetylcholine receptor alpha-subunit. FEBS Letters, 457, 117 ± 121. Raftery, M. A., Hunkapiller, M. W. , Strader, C. D. and Hood, L. E. 1980, Acetylcholine receptor: complex of homologous subunits. Science, 208, 1454 ± 1457. Rousselet, A., Devaux, P.F. and Wirtz, K.W., 1979, Free fatty acids and esters can be immobilized by receptor rich membranes from Torpedo marmorata but not phospholipid acyl chains. Biochemical and Biophysical Research Communications, 90, 871 ± 877. Sunshine, C. and McNamee, M.G., 1994, Lipid modulation of nicotinic acetylcholine receptor function: the role of membrane lipid composition and fluidity. Biochimica et Biophysica Acta, 1191, 59 ± 64. Tobimatsu, T. , Fujita, Y., Fukuda, K., Tanaka, K., Mori, Y., Konno, T., Mishina, M. and Numa, S., 1987, Effects of substitution of putative transmembrane segments on nicotinic receptor function. FEBS letters, 222, 56 ± 62. Unwin, N., 1995, Acetylcholine receptor channel imaged in the open state. Nature, 373, 37 ± 43. Williamson, P.F., Bonev, B., Barrantes, F.J. and Watts, A., 2000, Structural characterization of the M4 transmembrane domain of the acetylcholine receptor: an NMR study. Biophysics Journal, 78, 147A. Zanello, L.P., Aztiria, E., Antollini, S.S. and Barrantes, F.J., 1996, Nicotinic acetylcholine receptor channels are influenced by the physical state of their membrane environment. Biophysical Journal, 70, 2155 ± 2164. Received 24 April 2002, and in revised form 29 May 2002