unimpaired as compared to that observed in biological membranes. A few examples will suffice to
illustrate the kind of evidence that has been gathered, as well as some caveats. For other specific cases,
the reader is referred to Table 5.1 and the references given therein.
BR, an archaebacterial light-driven proton pump (see Chap. 1, § 1.6.1), accomplishes its entire
photocycle after being trapped in A8-35 (Pocanschi et al. 2006a, b; Gohon et al. 2008; Charvolin et al.
2009; Dahmane et al. 2013) or in NAPols (Bazzacco et al. 2012). Compared to the kinetics observed in
the purple membrane, the first steps of the photocycle are accelerated whether BR is solubilized in
detergent or trapped in A8-35. For the detergent-solubilized state, this effect has been attributed to a
conformational relaxation that brings Asp-85, the first proton acceptor, closer to the Schiff base donor
(Milder et al. 1991). The same is probably true of the APol-trapped state (Dahmane et al. 2013). The
last steps of the photocycle feature similar kinetics in APol and in PM, whereas in detergent solution
the return to the fundamental state is slower. However, late kinetics similar to those in detergent
solution are observed for BR refolded in A8-35 in the absence of lipids (Dahmane et al. 2013). In
keeping with earlier observations on partially delipidated and relipidated purple membrane (Joshi et al.
1998), these data strongly suggest that the restoration of native-like protein/lipid interactions upon
transferring BR from detergent to APol is responsible for the recovery of PM-like late kinetics
(Dahmane et al. 2013).
The nicotinic acetylcholine receptor (nAChR) is a ligand-gated channel of the pLGIC family.
Located in the postsynaptic membrane at neuronal synapses and the neuromuscular junction, it
transduces a presynaptic chemical signal, the release of acetylcholine in the synaptic cleft, into an
electrical one, due to the permeabilization to cations of the postsynaptic membrane (see Chap. 1,
§ 1.6.2). The allosteric equilibrium between the resting and desensitized states of the nAChR that exists
in the absence of the neurotransmitter is strongly perturbed when receptor-rich postsynaptic
membranes from the electric organ of Torpedo marmorata are solubilized in detergent (Changeux
et al. 1980; compare Fig. 5.25A, B with Fig. 5.25C, D). Membrane-like equilibrium and transition
kinetics are recovered when a CHAPS-solubilized preparation is supplemented with A8-35 and diluted
below the CMC of CHAPS (Martinez et al. 2002; Fig. 5.25E, F). Lipids are known to modulate the
function of the nAChR (Chap. 1, § 1.6.2; see daCosta and Baenziger 2009, Baenziger et al. 2015,
Hénault et al. 2015, and references therein). As for BR, one possible explanation for the return to
membrane-like allosteric properties upon transfer to APols could therefore be that this transfer allows
lipids to rebind to critical allosteric sites at the TM surface of the receptor. Because detergents may also
perturb allosteric equilibria by binding to the TM domain of the nAChR, much as anesthetics do (see
Forman et al. 2015), it is equally possible, however, that the return to a membrane-like allosteric
equilibrium be due to the dissociation of the detergent, whether it is replaced by lipids or by APol. Both
factors can come into play, of course, if dilution of the detergent allows lipids to rebind to sites from
which they had been displaced.
Eight G protein-coupled receptors (GPCRs) have been studied in more or less detail after
trapping or folding in one or the other APol (Table 5.1; for reviews, see Banères et al. 2011; Mary
et al. 2014). Functional characterization is often limited to measuring the specificity and affinity of
ligand binding. Such is the case for the leukotriene BLT1, serotonin 5-HT 4(a) , and cannabinoid CB1
receptors, all of them expressed as inclusion bodies and folded by transfer from SDS to A8-35
(Dahmane et al. 2009; Banères et al. 2011; Mary et al. 2014). In the case of the leukotriene BLT2
receptor, expressed and folded in the same way, it was observed that G protein activation and arrestin
recruitment were inhibited or slowed down by A8-35, but not by NAPols. Conformational equilibria of
the A8-35-trapped BLT2 receptor were characterized by ligand binding and by NMR, and the receptorbound conformation of the ligand, leukotriene B 4 , established by NMR (Dahmane et al. 2009; Catoire
et al. 2010a, 2011; Banères et al. 2011; Mary et al. 2014). The ghrelin GHS-R1a receptor was folded in
NAPols and ligand binding, conformational transitions, G protein activation, and arrestin recruitment
characterized (Bazzacco et al. 2012). Interestingly, the ghrelin receptor folded in APol presents the
5.4 Functionality of Amphipol-Trapped Membrane Proteins
287
illustrate the kind of evidence that has been gathered, as well as some caveats. For other specific cases,
the reader is referred to Table 5.1 and the references given therein.
BR, an archaebacterial light-driven proton pump (see Chap. 1, § 1.6.1), accomplishes its entire
photocycle after being trapped in A8-35 (Pocanschi et al. 2006a, b; Gohon et al. 2008; Charvolin et al.
2009; Dahmane et al. 2013) or in NAPols (Bazzacco et al. 2012). Compared to the kinetics observed in
the purple membrane, the first steps of the photocycle are accelerated whether BR is solubilized in
detergent or trapped in A8-35. For the detergent-solubilized state, this effect has been attributed to a
conformational relaxation that brings Asp-85, the first proton acceptor, closer to the Schiff base donor
(Milder et al. 1991). The same is probably true of the APol-trapped state (Dahmane et al. 2013). The
last steps of the photocycle feature similar kinetics in APol and in PM, whereas in detergent solution
the return to the fundamental state is slower. However, late kinetics similar to those in detergent
solution are observed for BR refolded in A8-35 in the absence of lipids (Dahmane et al. 2013). In
keeping with earlier observations on partially delipidated and relipidated purple membrane (Joshi et al.
1998), these data strongly suggest that the restoration of native-like protein/lipid interactions upon
transferring BR from detergent to APol is responsible for the recovery of PM-like late kinetics
(Dahmane et al. 2013).
The nicotinic acetylcholine receptor (nAChR) is a ligand-gated channel of the pLGIC family.
Located in the postsynaptic membrane at neuronal synapses and the neuromuscular junction, it
transduces a presynaptic chemical signal, the release of acetylcholine in the synaptic cleft, into an
electrical one, due to the permeabilization to cations of the postsynaptic membrane (see Chap. 1,
§ 1.6.2). The allosteric equilibrium between the resting and desensitized states of the nAChR that exists
in the absence of the neurotransmitter is strongly perturbed when receptor-rich postsynaptic
membranes from the electric organ of Torpedo marmorata are solubilized in detergent (Changeux
et al. 1980; compare Fig. 5.25A, B with Fig. 5.25C, D). Membrane-like equilibrium and transition
kinetics are recovered when a CHAPS-solubilized preparation is supplemented with A8-35 and diluted
below the CMC of CHAPS (Martinez et al. 2002; Fig. 5.25E, F). Lipids are known to modulate the
function of the nAChR (Chap. 1, § 1.6.2; see daCosta and Baenziger 2009, Baenziger et al. 2015,
Hénault et al. 2015, and references therein). As for BR, one possible explanation for the return to
membrane-like allosteric properties upon transfer to APols could therefore be that this transfer allows
lipids to rebind to critical allosteric sites at the TM surface of the receptor. Because detergents may also
perturb allosteric equilibria by binding to the TM domain of the nAChR, much as anesthetics do (see
Forman et al. 2015), it is equally possible, however, that the return to a membrane-like allosteric
equilibrium be due to the dissociation of the detergent, whether it is replaced by lipids or by APol. Both
factors can come into play, of course, if dilution of the detergent allows lipids to rebind to sites from
which they had been displaced.
Eight G protein-coupled receptors (GPCRs) have been studied in more or less detail after
trapping or folding in one or the other APol (Table 5.1; for reviews, see Banères et al. 2011; Mary
et al. 2014). Functional characterization is often limited to measuring the specificity and affinity of
ligand binding. Such is the case for the leukotriene BLT1, serotonin 5-HT 4(a) , and cannabinoid CB1
receptors, all of them expressed as inclusion bodies and folded by transfer from SDS to A8-35
(Dahmane et al. 2009; Banères et al. 2011; Mary et al. 2014). In the case of the leukotriene BLT2
receptor, expressed and folded in the same way, it was observed that G protein activation and arrestin
recruitment were inhibited or slowed down by A8-35, but not by NAPols. Conformational equilibria of
the A8-35-trapped BLT2 receptor were characterized by ligand binding and by NMR, and the receptorbound conformation of the ligand, leukotriene B 4 , established by NMR (Dahmane et al. 2009; Catoire
et al. 2010a, 2011; Banères et al. 2011; Mary et al. 2014). The ghrelin GHS-R1a receptor was folded in
NAPols and ligand binding, conformational transitions, G protein activation, and arrestin recruitment
characterized (Bazzacco et al. 2012). Interestingly, the ghrelin receptor folded in APol presents the
5.4 Functionality of Amphipol-Trapped Membrane Proteins
287