points provided by proline and glycine residues. As a result, the equilibrium structure is polygonal
rather than perfectly circular (Figs. 3.8 and 3.9).
A major asset of NDs being their ability to reproduce a bilayer environment, the state of the lipids
has been the object of particular attention, using such techniques as NMR, molecular dynamics (MD),
differential scanning calorimetry (DSC), laurdan fluorescence measurements, and SAXS (see e.g.
Shaw et al. 2004; Denisov et al. 2005; Shih et al. 2005). The major conclusions can be summarized as
follows. All experimental and MD data converge to describe the lipid component as a patch of bilayer.
This is the major conclusion, the nuances that can be brought to this picture being second-order
refinements. The bilayer nature of the lipid patch is supported experimentally, in particular, by the
observation of phase transitions analogous to those observed with pure lipid bilayers, even though
slightly higher transition temperatures and a lesser cooperativity indicate a degree of perturbation
(Shaw et al. 2004; Denisov et al. 2005; Kijac et al. 2010). The latter is attributed (i) to the fact that a
large number of lipids (see Fig. 3.12) are in contact with MSPs, and do not take part in the transition,
and (ii) to the sensitivity of MSPs themselves to temperature changes (Denisov et al. 2005). As
expected, phase transitions become sharper as the size of the NDs grows, more lipids taking part in
them (Grinkova et al. 2010). There are other differences between those lipids that interact with MSPs
and those at the center of the ND: the former tend to occupy a larger area per molecule and, therefore,
Fig. 3.9 Experimental determination of the structure of the membrane scaffold proteins (MSPs)
stabilizing a dimyristoylphosphatidylcholine (DMPC) nanodisc (ND). The structure was obtained by a
combination of solution NMR, electron paramagnetic resonance (double electron-electron resonance
(DEER)-derived distance restraints), and transmission electron microscopy data. (A, B) Top (A) and
side (B) views of ten superimposed conformers of the MSPs. Two antiparallel MSPΔH5 molecules
(a shortened version of apoA-I lacking the non-lipid-binding residues 1–54 as well as residues 121–142,
which are proposed to form helix 5; cf. Fig. 3.7) encircle a DMPC bilayer patch (not shown) to form a
structure with twofold symmetry. The nine helices of each monomer are individually color-coded and
labeled. Pro or Gly residues (yellow) usually separate individual helices. (C) van der Waals surface
representation of the MSPs cut in half to reveal the lipid-oriented interior, with hydrophobic residues
colored in green. (D) A view of the solvent-exposed outside surface (From Bibow et al. 2017, # 2017
Macmillan Publishers Limited, Nature Structural Molecular Biology. All rights reserved).
110
3 Alternatives to Detergents for Handling Membrane Proteins in Aqueous Solutions
rather than perfectly circular (Figs. 3.8 and 3.9).
A major asset of NDs being their ability to reproduce a bilayer environment, the state of the lipids
has been the object of particular attention, using such techniques as NMR, molecular dynamics (MD),
differential scanning calorimetry (DSC), laurdan fluorescence measurements, and SAXS (see e.g.
Shaw et al. 2004; Denisov et al. 2005; Shih et al. 2005). The major conclusions can be summarized as
follows. All experimental and MD data converge to describe the lipid component as a patch of bilayer.
This is the major conclusion, the nuances that can be brought to this picture being second-order
refinements. The bilayer nature of the lipid patch is supported experimentally, in particular, by the
observation of phase transitions analogous to those observed with pure lipid bilayers, even though
slightly higher transition temperatures and a lesser cooperativity indicate a degree of perturbation
(Shaw et al. 2004; Denisov et al. 2005; Kijac et al. 2010). The latter is attributed (i) to the fact that a
large number of lipids (see Fig. 3.12) are in contact with MSPs, and do not take part in the transition,
and (ii) to the sensitivity of MSPs themselves to temperature changes (Denisov et al. 2005). As
expected, phase transitions become sharper as the size of the NDs grows, more lipids taking part in
them (Grinkova et al. 2010). There are other differences between those lipids that interact with MSPs
and those at the center of the ND: the former tend to occupy a larger area per molecule and, therefore,
Fig. 3.9 Experimental determination of the structure of the membrane scaffold proteins (MSPs)
stabilizing a dimyristoylphosphatidylcholine (DMPC) nanodisc (ND). The structure was obtained by a
combination of solution NMR, electron paramagnetic resonance (double electron-electron resonance
(DEER)-derived distance restraints), and transmission electron microscopy data. (A, B) Top (A) and
side (B) views of ten superimposed conformers of the MSPs. Two antiparallel MSPΔH5 molecules
(a shortened version of apoA-I lacking the non-lipid-binding residues 1–54 as well as residues 121–142,
which are proposed to form helix 5; cf. Fig. 3.7) encircle a DMPC bilayer patch (not shown) to form a
structure with twofold symmetry. The nine helices of each monomer are individually color-coded and
labeled. Pro or Gly residues (yellow) usually separate individual helices. (C) van der Waals surface
representation of the MSPs cut in half to reveal the lipid-oriented interior, with hydrophobic residues
colored in green. (D) A view of the solvent-exposed outside surface (From Bibow et al. 2017, # 2017
Macmillan Publishers Limited, Nature Structural Molecular Biology. All rights reserved).
110
3 Alternatives to Detergents for Handling Membrane Proteins in Aqueous Solutions