accommodated by small movements of the APol alkyl chains may not be strongly affected, whereas
those requiring rearrangements of the polyacrylate backbone would (Popot et al. 2003, 2011; Picard
et al. 2006). The MD data lead to a more nuanced view, inasmuch as (i) all movements appear to be
damped, whatever their length scale, even though large-scale movements tend to be more strongly
damped than small-scale ones (Fig. 5.38A); this may suggest that reasoning in terms of increased
overall surface viscosity of the surfactant, as observed when comparing the polar surface of pure A8-35
particles with that of detergent micelles (Perlmutter et al. 2011; see Chap. 4, § 4.3.1.2.4), may be more
appropriate than dissecting the molecular movements of the polymer; and (ii) the effect propagates to
the extramembrane loops, even though those are not in contact with the polymer, or only minimally so
(§ 5.3.3, Fig. 5.15B). This opens the possibility that functional effects may be indirect. Whatever the
mechanism, it does not seem shocking that SERCA1a, the enzymatic cycle of which requires ample
TM and extramembrane rearrangements (see Chap. 1, § 1.6.3), be detectably inhibited, whereas BR
and the nAChR, whose transconformations are subtler (ibid., § 1.6.1 and 1.6.2), show no obvious
evidence of perturbation.
It is worth noting that, on the basis of coarse-grained MD simulations, a “straightjacket effect”
similar to the “Gulliver effect” has been postulated to account for the inhibition experienced by
SERCA1a upon insertion into too thin lipid bilayers, such as those formed by diC 14:1 PC (Sonntag
et al. 2011). These effects are reminiscent of the slowing down of dynamics and functional blockade of
proteins by partial dehydration (Zaccai 2004). It is interesting to note that the stabilizing effect of some
mutations on GPCRs seems to involve a stiffening of the TM region, thermostable mutants showing
less relative TM helix movements than their respective wild-type receptors (Vaidehi et al. 2016). By
totally different routes, trapping with APols and selecting thermostabilizing mutations may reach the
same end, that of restricting conformational excursions that lead to denaturation.
Clearly, the degree of functional perturbation experienced by a number of MPs with different
structures and functions will have to be studied in greater detail before a general picture can emerge.
The OmpX/A8-35 simulations report on the structural fluctuations of a small compact protein around a
single equilibrium structure and not on large conformational changes in a complex multi-domain
protein such as SERCA1a. One should also note that, if the rate-limiting step in a functional cycle does
not depend on transconformations occurring in the TM region, APols are unlikely to affect it. It should
also be kept in mind that there is now good evidence that some of the functional differences observed
between detergent-solubilized and APol-trapped MPs are due to lipid rebinding upon transfer to APols,
as is clearly the case for the photocycle of BR (Dahmane et al. 2013), and is suspected of the allosteric
equilibria of the nAChR (Martinez et al. 2002) (see § 5.5). Lipid rebinding is clearly a stabilizing
factor, whether in the presence of detergents (Chap. 2, § 2.4.1) or in APols (this chapter, § 5.5). It may
suffice, in some cases, to account for the stabilization observed upon transfer from detergent solution to
APols, without invoking their effects on dynamics. Lipid rebinding, however, cannot explain all
stabilizing effects of APols. In the case of SERCA1a, it cannot explain the observed correlation
between stabilization and inhibition: given that delipidation is known to diminish the activity of the
pump (de Foresta et al. 1989; Lund et al. 1989), lipid rebinding upon transfer from detergent to APols
ought to be stimulatory, not inhibitory. In the case of BR, the protein is more stable in A8-35 in the
total absence of lipids than it is in detergent solution with its whole quota of purple membrane lipids
present (Dahmane et al. 2013; see Fig. 5.27). This is compatible with the APol-trapped protein
experiencing a higher free energy barrier to unfolding even in the absence of lipids.
5.6 Membrane Protein Dynamics and the Effects of Amphipols on Stability and Function
305
those requiring rearrangements of the polyacrylate backbone would (Popot et al. 2003, 2011; Picard
et al. 2006). The MD data lead to a more nuanced view, inasmuch as (i) all movements appear to be
damped, whatever their length scale, even though large-scale movements tend to be more strongly
damped than small-scale ones (Fig. 5.38A); this may suggest that reasoning in terms of increased
overall surface viscosity of the surfactant, as observed when comparing the polar surface of pure A8-35
particles with that of detergent micelles (Perlmutter et al. 2011; see Chap. 4, § 4.3.1.2.4), may be more
appropriate than dissecting the molecular movements of the polymer; and (ii) the effect propagates to
the extramembrane loops, even though those are not in contact with the polymer, or only minimally so
(§ 5.3.3, Fig. 5.15B). This opens the possibility that functional effects may be indirect. Whatever the
mechanism, it does not seem shocking that SERCA1a, the enzymatic cycle of which requires ample
TM and extramembrane rearrangements (see Chap. 1, § 1.6.3), be detectably inhibited, whereas BR
and the nAChR, whose transconformations are subtler (ibid., § 1.6.1 and 1.6.2), show no obvious
evidence of perturbation.
It is worth noting that, on the basis of coarse-grained MD simulations, a “straightjacket effect”
similar to the “Gulliver effect” has been postulated to account for the inhibition experienced by
SERCA1a upon insertion into too thin lipid bilayers, such as those formed by diC 14:1 PC (Sonntag
et al. 2011). These effects are reminiscent of the slowing down of dynamics and functional blockade of
proteins by partial dehydration (Zaccai 2004). It is interesting to note that the stabilizing effect of some
mutations on GPCRs seems to involve a stiffening of the TM region, thermostable mutants showing
less relative TM helix movements than their respective wild-type receptors (Vaidehi et al. 2016). By
totally different routes, trapping with APols and selecting thermostabilizing mutations may reach the
same end, that of restricting conformational excursions that lead to denaturation.
Clearly, the degree of functional perturbation experienced by a number of MPs with different
structures and functions will have to be studied in greater detail before a general picture can emerge.
The OmpX/A8-35 simulations report on the structural fluctuations of a small compact protein around a
single equilibrium structure and not on large conformational changes in a complex multi-domain
protein such as SERCA1a. One should also note that, if the rate-limiting step in a functional cycle does
not depend on transconformations occurring in the TM region, APols are unlikely to affect it. It should
also be kept in mind that there is now good evidence that some of the functional differences observed
between detergent-solubilized and APol-trapped MPs are due to lipid rebinding upon transfer to APols,
as is clearly the case for the photocycle of BR (Dahmane et al. 2013), and is suspected of the allosteric
equilibria of the nAChR (Martinez et al. 2002) (see § 5.5). Lipid rebinding is clearly a stabilizing
factor, whether in the presence of detergents (Chap. 2, § 2.4.1) or in APols (this chapter, § 5.5). It may
suffice, in some cases, to account for the stabilization observed upon transfer from detergent solution to
APols, without invoking their effects on dynamics. Lipid rebinding, however, cannot explain all
stabilizing effects of APols. In the case of SERCA1a, it cannot explain the observed correlation
between stabilization and inhibition: given that delipidation is known to diminish the activity of the
pump (de Foresta et al. 1989; Lund et al. 1989), lipid rebinding upon transfer from detergent to APols
ought to be stimulatory, not inhibitory. In the case of BR, the protein is more stable in A8-35 in the
total absence of lipids than it is in detergent solution with its whole quota of purple membrane lipids
present (Dahmane et al. 2013; see Fig. 5.27). This is compatible with the APol-trapped protein
experiencing a higher free energy barrier to unfolding even in the absence of lipids.
5.6 Membrane Protein Dynamics and the Effects of Amphipols on Stability and Function
305