Influence of Lipids on nAChR Structure

Structural knowledge of nAChR and other LGIC has been greatly hindered by the absence of a crystal structure. The most informative structure is from electron micrographs of the tightly packed arrays of nAChR in tubular membranes isolated from the electric organ of Torpedo electric rays (Toyoshima and Unwin 1988; Unwin 1995; Miyazawa et al. 2003). These studies indicate that the transmembrane segments (M1-M4) are basically a helices, although Ml seems to have a distorted helical structure, probably due to the presence of a proline residue. These data have been confirmed through NMR and other spectroscopic studies of the peptides corresponding to the different transmembrane segments of the protein or with biochemical approaches such as photolabelling, protein modification and site-directed mutagenesis of the entire protein (Karlin et al. 1986; Akabas et al. 1994; Blanton and Cohen 1994; Blanton et al. 1998; Corbin et al. 1998; Lugovskoy et al. 1998; Opella et al. 1999; Pashkov et al. 1999; Williamson et al. 2004; Barrantes et al. 2000; Tamamizu et al. 2000; Cruz-Martin et al. 2001; Guzman et al. 2003; Ortiz-Acevedo et al. 2004; Santiago et al. 2004). Finally, site-directed mutagenesis and NMR experiments proposed that M3 contains a mixture of a helix and 3i0-helix (Lugovskoy et al. 1998; Guzman et al. 2003). The secondary structure of the entire protein has also been studied through spectroscopic techniques such as Raman, FT-IR or CD. The calculated a-helix content ranges from 20 to 43%, p sheet content from 29 to 48%, and non-ordered structure from 20 to 28% (Moore et al. 1974; Mielke and Wallace 1988; Yager et al. 1984; Fong and McNamee 1987; Butler and McNamee 1993; Methot et al. 1994; Castresana et al. 1992). On the other hand, theoretical predictions estimate 44% a-helix and 27% P-sheet (Finer-Moore and Stroud 1984). The lack of concordance of these data is probably due to the low sensitivity of CD to detect beta structures, the diversity of the FT-IR quantification methods, and finally to the different conditions used to reconstitute the protein.

As stated in the previous section, lipid surrounding nAChR modulates protein function, possibly through changes in the general properties of the membrane or by direct binding to the protein. This modulation should be caused by some effect on the protein conformation, probably acting on one or several of the transmembrane domains that are in direct contact with lipids (Ml, M3 and M4). In fact, site-directed mutagenesis studies report many examples of residues located at the lipid-protein interface whose mutation to tryptophan dramatically affects protein function (Tamamizu et al. 2000; Guzman et al. 2003; Ortiz-Acevedo et al. 2004; Santiago et al. 2004). In order to detect the possible structural changes associated with lipid nAChR modulation, different spectroscopic studies have been done, mainly through analysis of the FT-IR amide I' band. Lipid membranes

Fig. 8.3 Infrared spectra (1800-1505 cm"1) of nAChR reconstituted in different lipid vesicles at 20 and 70°C. Spectra show lipid carbonyl,

1800 1750 1700 1650 1600 1550 Wave number, cm"1

Fig. 8.3 Infrared spectra (1800-1505 cm"1) of nAChR reconstituted in different lipid vesicles at 20 and 70°C. Spectra show lipid carbonyl, amide I', amide II and tyrosine bands

1800 1750 1700 1650 1600 1550 Wave number, cm"1

where nAChR is fully functional, typically those containing PA and cholesterol, are those where the protein presents a higher a-helical content relative to non-ordered structure, whereas nAChR in lipid membranes where it is less active, such as those with only zwitterionic phospholipids, shows a larger non-ordered structure concomitant with a decrease in the a-helix. Meanwhile, the P-sheet content remains basically unchanged (Fig. 8.3).

These FT-IR experiments were done after submitting the samples to a D20 for H20 exchange. This process can be followed through the FT-IR amide II band, which diminishes with the H-D exchange, and depends on the accessibility of the different amino acids to the solvent, so it reports information about the tertiary protein structure (Hvidt and Nielsen 1966; Pershina and Hvidt 1974). Once the equilibrium for the H-D exchange is reached, the remaining amide II band was quantified, showing that those samples with a larger remaining amide II, that is, with the lower HtoD exchange, are precisely those with less non-ordered structure. Furthermore, if those samples are constantly heated this amide II band disappears, showing a sigmoidal behaviour. From these curves, it is possible to calculate a temperature (Tm) for the collapse of the tertiary structure. Again it is observed that samples with a higher content of ordered secondary structure are those with a higher Tm. These results support nAChR structure data calculations from amide I' analysis, since those samples with more ordered secondary structure, typically a-helical, are those more resistant to H-D exchange (authors' submitted manuscript). The mechanism by which anionic lipids, especially PA, stabilize the a-helix structure of nAChR is not clear, although some authors have pointed to an interaction between the dipole from the a-helical structure of nAChR and the PA phosphates (Hoi et al. 1978; Sali et al. 1988). In addition, some work has been done with model peptides that also detect stabilization of a-helical structure by anionic phospholipids (Liu and Deber 1997). However, other authors have reported an increment in p-sheet content upon addition of anionic phospholipids (Butler and McNamee 1993; Fong and McNamee 1987), although these studies were done in a region of the infrared spectrum where bands are rather weak and its assignment to secondary structure is not clear. It has also been reported that cholesterol favours an increment in a-helical structure (FernandezBallester et al. 1994; Fong and McNamee 1987; Butler and McNamee 1993) and even the p sheet (Fernandez-Ballester et al. 1994). To explain this result it has been postulated that the rigid sterol ring, oriented parallel to the receptor axis, may localize in between helices at the lipid-protein interface, causing their stabilization (Fong and McNamee 1987).

Among the different regions conforming to the nAChR overall structure, the Ml transmembrane segment may be a good candidate to be modulated by lipids (Williamson et al. 2004; dePlanque et al. 2004). NMR studies of this transmembrane segment reconstituted in lipids show that some portions adopt an a-helical conformation but that the presence of a proline located in the middle of the segment significantly disrupts the a-helical structure. In fact, a proline and about four surrounding residues typically form a kink in the transmembrane stretch with an angle that can vary between 5° and 60°, and these hinge regions are thought to play a key role in membrane proteins because of their expected inherent flexibility (Cordes et al. 2002; Arshava et al. 2002). Furthermore, conformational studies as a function of the lipid environment suggest that the degree of helicity in this region strongly depends on the lipid environment, and that Ml orders DMPC acyl chains and interact more favourably with cholesterol containing PC bilayers, mimicking several aspects of the effect of the entire nAChR on model membranes (de Planque et al. 200). This flexibility could be maintained in the entire protein as transmembrane segments in nAChR are loosely packed, and Ml, M3 and M4 are largely separated by water-filled cavities from the inner ring of M2 helices (Mi-yazawa et al. 2003). These results, together with the observed Ml labelling from both hydrophobic and hydrophilic probes (Blanton and Cohen 1994; Karlin et al. 1986), and the close proximity between Ml and M2, suggest that the conformational flexibility around the proline in the Ml-transmembrane maybe important for the modulation of channel gating by the lipid environment and by other molecules which partition into the lipid bilayer, such as general anaesthetics.

Opposite to these results are those using ATR spectroscopy (Ryan et al. 1996; Baenziger et al. 2000). These authors find very faint variations in the amide I' band for nAChR reconstituted in different lipid vesicles, only detectable after band deconvolution. They propose that these little variations are caused by the different rate ofH-D exchange for the different samples, due to subtle variations in the protein dynamics that do not involve changes in the secondary structure. These variations may be those causing the lipid modulation on nAChR function. This could explain the small differences they find in the amide I' band, but not the large differences referred to above, as there is evidence showing that large variations in H-D exchange do not cause significant changes in the quantification of secondary structure motifs from the amide I' band (authors' submitted manuscript). One reason to explain why these authors do not detect large variations in the amide I' band could be the fact that nAChR-containing samples are submitted to a drying cycle accompanied by a long period of rehydration (up to three days) before doing the FT-IR experiments. The consequences of this process have not been tested, since no functional experiments were done after these treatments. It is possible then, that nAChR is in a desensitized-like state, independently of the lipids where it is reconstituted. In the same sense, cholesterol has been proposed to modulate nAChR function without varying protein structure. To do so, it would localize in the spaces between different nAChR subunits facilitating the sliding between them, so making possible the conformational changes necessaries for channel function (Corbin et al. 1998).

nAChR, like other ligand-gated ion channels, after binding of the corresponding agonist, suffers a conformational change in transmembrane segments, probably a rotational movement, destabilizing the hydrophobic girdle that forms the channel gate and thus allowing ions to pass through the pore. The mechanisms and pathway to transform the energy of ligand binding at the extracytoplasmic domain of the protein into the movements of the M2 segments are largely unknown. Electrostatic and hydrophobic interactions have been proposed between the M2-M3 loop and other loops in the agonist-binding domain as being responsible for this transmission (Kash et al. 2003; Miyazawa et al. 2003). The action of phospholipids to modulate nAChR function should interfere with either the transmission between ligand-binding domain towards transmembrane segments or the subsequent transmembrane movement. Possibility to do that is causing the protein to enter into a non-active conformation, through changes in the secondary and/or tertiary nAChR structure. Considering the above results, zwitterionic lipids may stabilize a conformation with less ordered a-helical content that would impede some of the steps that allow the protein function. By contrast, anionic lipids and cholesterol would stabilize a more compact conformation able to transmit movements of the binding domain towards the transmembrane domain.

Was this article helpful?

0 0
Healthy Chemistry For Optimal Health

Healthy Chemistry For Optimal Health

Thousands Have Used Chemicals To Improve Their Medical Condition. This Book Is one Of The Most Valuable Resources In The World When It Comes To Chemicals. Not All Chemicals Are Harmful For Your Body – Find Out Those That Helps To Maintain Your Health.

Get My Free Ebook

Post a comment