n-alkyl chains associated with a given α-helical MP is plotted as a function of the square root of the
number of TM helices in the protein, which can be taken to be roughly proportional to their membraneexposed surface. Whereas it is obvious that this is only a very crude measure of their hydrophobic TM
surface, a more or less linear relationship should be expected. The considerable scatter of the data
clearly suggests imprecise measurements of APol binding. More systematic measurements under
better controlled conditions would be highly desirable. A standard procedure is proposed in § 5.13,
Protocol 5.2.
Despite this caveat, it seems clear that there is a tendency for less APol n-alkyl chains than DDM
ones to bind to a given MP (Fig. 5.13), even though these chains are shorter in APols than in DDM and
could, a priori, cover less of the protein’s hydrophobic surface. This suggests a different organization
of the alkyl chains at the surface of the protein. Indeed, according to NMR data, the octyl chains of
A8-35 appear to lie sidewise at the surface of the protein, thereby occupying a larger surface than if
they were more or less perpendicular to it (Catoire et al. 2009; Planchard et al. 2014; Fig. 5.14).
Furthermore, both NMR data (Planchard et al. 2014; Fig. 5.14) and MD data (Perlmutter et al. 2014;
Fig. 5.15) indicate that, in the case of A8-35, part of the protein surface is in contact with isopropyl
chains (Fig. 5.15), which may contribute to explain why less n-alkyl chains are required to shield this
surface from water than in the case of DDM.
5.3.1.2 Lipid Binding
Quantitative data about the presence of lipids in MP/APol complexes are scarce. The most detailed
analysis is that of BR/A8-35 complexes obtained by solubilizing purple membrane fragments in OTG,
centrifuging, supplementing the supernatant with A8-35, and eliminating OTG by adsorption onto
Bio-Beads. BR/A8-35 complexes were then separated from free APol particles by centrifugation onto
sucrose gradients, so as to eliminate those lipids that would not be associated to the complexes. Thinlayer chromatography analyses revealed no difference between lipids extracted from PM or from
purified BR/A8-35 complexes and were similar to literature profiles (Corcelli et al. 2000). Phosphate
determination performed on two different preparations of BR/APol complexes indicated the presence
of 5.2 and 4.5 moles of phospholipid per mole of BR, to be compared with 4.6 mol/mol in PM. These
values are within experimental error of each other and of those previously reported for PM: 6.4, 5, and
4.3 mol/mol (see Renner et al. 2005, and references therein). This indicates that BR/APol complexes
retain – or, more likely, rebind – all or almost all of the lipids originally present in the purple
membrane, i.e. ~0.38 g lipid per g BR (Gohon et al. 2008).
It is interesting to view these data in the context of more recent experiments, in which BR was
either trapped from OTG-solubilized PM, as in the above experiments, or refolded in A8-35 either from
SDS-solubilized PM – that is in the presence of PM lipids – or from bacterio-opsin (BO; the
apoprotein) that had been delipidated in organic solvents before being transferred to SDS (Dahmane
et al. 2013). Photocycle measurements show a clear difference between, on the one hand,
OTG-solubilized BR and BR refolded in A8-35 in the absence of lipids and, on the other hand, BR
trapped in A8-35 from OTG-solubilized PM or BR refolded in A8-35 in the presence of PM lipids
(Dahmane et al. 2013). The simplest interpretation of these data is that, in OTG, lipids are separated
from BR, or some critical lipids do not interact with it as they do in PM, whereas, after transfer to
A8-35, they rebind to BR. Note that, even though A8-35-trapped BR appears to rebind a full
complement of lipids, those do not suffice to form a complete annulus. Indeed, in the purple membrane,
BR molecules are assembled into trimers. The trimers are separated from each other by lipids, and their
central cavity filled with lipids, but individual BR molecules in a trimer interact both via protein/lipid
and protein/protein contacts (Grigorieff et al. 1996; cf. Chap. 3, Fig. 3.12C). The membrane therefore
does not contain enough lipids to form a complete annulus around each BR monomer.
5.3 Composition, Organization, Dynamics, and Solution Properties of Membrane. . .
275
number of TM helices in the protein, which can be taken to be roughly proportional to their membraneexposed surface. Whereas it is obvious that this is only a very crude measure of their hydrophobic TM
surface, a more or less linear relationship should be expected. The considerable scatter of the data
clearly suggests imprecise measurements of APol binding. More systematic measurements under
better controlled conditions would be highly desirable. A standard procedure is proposed in § 5.13,
Protocol 5.2.
Despite this caveat, it seems clear that there is a tendency for less APol n-alkyl chains than DDM
ones to bind to a given MP (Fig. 5.13), even though these chains are shorter in APols than in DDM and
could, a priori, cover less of the protein’s hydrophobic surface. This suggests a different organization
of the alkyl chains at the surface of the protein. Indeed, according to NMR data, the octyl chains of
A8-35 appear to lie sidewise at the surface of the protein, thereby occupying a larger surface than if
they were more or less perpendicular to it (Catoire et al. 2009; Planchard et al. 2014; Fig. 5.14).
Furthermore, both NMR data (Planchard et al. 2014; Fig. 5.14) and MD data (Perlmutter et al. 2014;
Fig. 5.15) indicate that, in the case of A8-35, part of the protein surface is in contact with isopropyl
chains (Fig. 5.15), which may contribute to explain why less n-alkyl chains are required to shield this
surface from water than in the case of DDM.
5.3.1.2 Lipid Binding
Quantitative data about the presence of lipids in MP/APol complexes are scarce. The most detailed
analysis is that of BR/A8-35 complexes obtained by solubilizing purple membrane fragments in OTG,
centrifuging, supplementing the supernatant with A8-35, and eliminating OTG by adsorption onto
Bio-Beads. BR/A8-35 complexes were then separated from free APol particles by centrifugation onto
sucrose gradients, so as to eliminate those lipids that would not be associated to the complexes. Thinlayer chromatography analyses revealed no difference between lipids extracted from PM or from
purified BR/A8-35 complexes and were similar to literature profiles (Corcelli et al. 2000). Phosphate
determination performed on two different preparations of BR/APol complexes indicated the presence
of 5.2 and 4.5 moles of phospholipid per mole of BR, to be compared with 4.6 mol/mol in PM. These
values are within experimental error of each other and of those previously reported for PM: 6.4, 5, and
4.3 mol/mol (see Renner et al. 2005, and references therein). This indicates that BR/APol complexes
retain – or, more likely, rebind – all or almost all of the lipids originally present in the purple
membrane, i.e. ~0.38 g lipid per g BR (Gohon et al. 2008).
It is interesting to view these data in the context of more recent experiments, in which BR was
either trapped from OTG-solubilized PM, as in the above experiments, or refolded in A8-35 either from
SDS-solubilized PM – that is in the presence of PM lipids – or from bacterio-opsin (BO; the
apoprotein) that had been delipidated in organic solvents before being transferred to SDS (Dahmane
et al. 2013). Photocycle measurements show a clear difference between, on the one hand,
OTG-solubilized BR and BR refolded in A8-35 in the absence of lipids and, on the other hand, BR
trapped in A8-35 from OTG-solubilized PM or BR refolded in A8-35 in the presence of PM lipids
(Dahmane et al. 2013). The simplest interpretation of these data is that, in OTG, lipids are separated
from BR, or some critical lipids do not interact with it as they do in PM, whereas, after transfer to
A8-35, they rebind to BR. Note that, even though A8-35-trapped BR appears to rebind a full
complement of lipids, those do not suffice to form a complete annulus. Indeed, in the purple membrane,
BR molecules are assembled into trimers. The trimers are separated from each other by lipids, and their
central cavity filled with lipids, but individual BR molecules in a trimer interact both via protein/lipid
and protein/protein contacts (Grigorieff et al. 1996; cf. Chap. 3, Fig. 3.12C). The membrane therefore
does not contain enough lipids to form a complete annulus around each BR monomer.
5.3 Composition, Organization, Dynamics, and Solution Properties of Membrane. . .
275