Folding was observed to be very efficient, with yields of ~100% for OmpA and ~90% for FomA
(Fig. 6.9A, B; Table 6.1, lines 19–20). It was completed within ~7 h for OmpA and ~24 h for FomA.
Several criteria were used to assess that the two proteins had achieved their native state:
(i) SDS-PAGE (Fig. 6.9A, B);
(ii) Protection of the folded β-barrels against proteolysis (ibid.);
(iii) Far-UV circular dichroism (CD) spectroscopy (Fig. 6.9C, D);
(iv) Functional studies, that is single-channel conductance recordings after transfer of A8-35refolded OmpA and FomA to black lipid films (Pocanschi et al. 2006).
In these experiments, the first three criteria did not reveal any differences between the detergentand A8-35-refolded forms. However, in single-channel measurements of the conductance of the pores
formed upon incorporation into black lipid bilayers, both OmpA and FomA initially displayed smaller
conductance when inserted from complexes with A8-35 rather than from detergent-refolded forms.
This difference was traced to an asymmetrical distribution of A8-35 in the black lipid films. When
A8-35 was present at equal concentrations on both sides of the film, the same conductance was
observed as with the native or detergent-folded proteins (Pocanschi et al. 2006; see Chap. 5, § 5.7,
Fig. 5.34). The spontaneous transfer of refolded OmpA and FomA from A8-35 to black lipid bilayers is
consistent with the transfer, observed earlier, of the α-helical MP DAGK from another APol, OAPA20 (similar to A8-75; see Chap. 4, Fig. 4.1), into lipid bilayers of 1-palmitoyl-2-oleoyl-phosphatidylcholine (multilamellar vesicles) in a functionally active form (Nagy et al. 2001; see Chap. 5, § 5.7). The
effect of the asymmetrical distribution of A8-35 on conductance levels is consistent with the adsorption of APols onto either artificial lipid vesicles (reviewed in Marie et al. 2014) or living cells (Popot
et al. 2011; see Chap. 15). A direct interaction of the membrane-adsorbed APol with the proteins is
possible, but seems rather unlikely, because both entropic considerations and observations made in
lipid mesophases (Polovinkin et al. 2014; see Chap. 11) rather suggest that APol and protein dissociate
one from another upon insertion into a lipid membrane (see Chap. 5, § 5.7). A physical effect seems
more likely, such as could result from an APol-induced asymmetry in surface charge and/or lateral
pressure. This would readily explain why the conductance levels go back to normal upon adding APol
to the other side of the film.
6.3.1.2.2 Folding of tOmpA, OmpT, and PagP
Folding of a β-barrel MP has also been demonstrated with sulfonated APols (SAPols; see Chap. 4,
Fig. 4.1), using a similar protocol (Dahmane et al. 2011). After 2 days of incubation, the genetically
engineered TM domain of OmpA (tOmpA), isolated after expression into inclusion bodies, folded to
yields approaching 100% (Table 6.1, line 21), as determined by the shift of its electrophoretic mobility
from ~16 kDa for the unfolded form to ~19 kDa for the folded one.
OmpT, an integral outer membrane protease with ten TM β-strands, and PagP, an eight-stranded
β-barrel MP that catalyzes the transfer of a palmitate chain from a phospholipid to lipid A, have been
folded from their unfolded forms in 8 M urea by supplementing them with A8-35 at a 1:5 protein/APol
mass ratio and dialyzing away the urea (Leney et al. 2012). Using electrospray ionization mass
spectrometry coupled with ion mobility spectrometry (ESI-IM-MS; see Chap. 14), yields were
shown to reach ~100% for OmpT and ~40% for PagP (Table 6.1, lines 22–23, and Fig. 6.11D, E).
Folding was confirmed by electrophoretic mobility measurements (Fig. 6.10A, B), by far-UV CD
spectroscopy, and by functional studies (Leney et al. 2012). SEC showed a single narrow peak for
OmpT/A8-35 complexes, indicating the presence of a homogeneous single species of OmpT, whereas
a broader peak was observed after folding of PagP, indicating the presence of a mixture of folded and
unfolded species, consistent with the ~60% folding yield observed by SDS-PAGE and ESI-IMS-MS
(Leney et al. 2012).
348
6 Amphipol-Assisted Folding of Membrane Proteins
(Fig. 6.9A, B; Table 6.1, lines 19–20). It was completed within ~7 h for OmpA and ~24 h for FomA.
Several criteria were used to assess that the two proteins had achieved their native state:
(i) SDS-PAGE (Fig. 6.9A, B);
(ii) Protection of the folded β-barrels against proteolysis (ibid.);
(iii) Far-UV circular dichroism (CD) spectroscopy (Fig. 6.9C, D);
(iv) Functional studies, that is single-channel conductance recordings after transfer of A8-35refolded OmpA and FomA to black lipid films (Pocanschi et al. 2006).
In these experiments, the first three criteria did not reveal any differences between the detergentand A8-35-refolded forms. However, in single-channel measurements of the conductance of the pores
formed upon incorporation into black lipid bilayers, both OmpA and FomA initially displayed smaller
conductance when inserted from complexes with A8-35 rather than from detergent-refolded forms.
This difference was traced to an asymmetrical distribution of A8-35 in the black lipid films. When
A8-35 was present at equal concentrations on both sides of the film, the same conductance was
observed as with the native or detergent-folded proteins (Pocanschi et al. 2006; see Chap. 5, § 5.7,
Fig. 5.34). The spontaneous transfer of refolded OmpA and FomA from A8-35 to black lipid bilayers is
consistent with the transfer, observed earlier, of the α-helical MP DAGK from another APol, OAPA20 (similar to A8-75; see Chap. 4, Fig. 4.1), into lipid bilayers of 1-palmitoyl-2-oleoyl-phosphatidylcholine (multilamellar vesicles) in a functionally active form (Nagy et al. 2001; see Chap. 5, § 5.7). The
effect of the asymmetrical distribution of A8-35 on conductance levels is consistent with the adsorption of APols onto either artificial lipid vesicles (reviewed in Marie et al. 2014) or living cells (Popot
et al. 2011; see Chap. 15). A direct interaction of the membrane-adsorbed APol with the proteins is
possible, but seems rather unlikely, because both entropic considerations and observations made in
lipid mesophases (Polovinkin et al. 2014; see Chap. 11) rather suggest that APol and protein dissociate
one from another upon insertion into a lipid membrane (see Chap. 5, § 5.7). A physical effect seems
more likely, such as could result from an APol-induced asymmetry in surface charge and/or lateral
pressure. This would readily explain why the conductance levels go back to normal upon adding APol
to the other side of the film.
6.3.1.2.2 Folding of tOmpA, OmpT, and PagP
Folding of a β-barrel MP has also been demonstrated with sulfonated APols (SAPols; see Chap. 4,
Fig. 4.1), using a similar protocol (Dahmane et al. 2011). After 2 days of incubation, the genetically
engineered TM domain of OmpA (tOmpA), isolated after expression into inclusion bodies, folded to
yields approaching 100% (Table 6.1, line 21), as determined by the shift of its electrophoretic mobility
from ~16 kDa for the unfolded form to ~19 kDa for the folded one.
OmpT, an integral outer membrane protease with ten TM β-strands, and PagP, an eight-stranded
β-barrel MP that catalyzes the transfer of a palmitate chain from a phospholipid to lipid A, have been
folded from their unfolded forms in 8 M urea by supplementing them with A8-35 at a 1:5 protein/APol
mass ratio and dialyzing away the urea (Leney et al. 2012). Using electrospray ionization mass
spectrometry coupled with ion mobility spectrometry (ESI-IM-MS; see Chap. 14), yields were
shown to reach ~100% for OmpT and ~40% for PagP (Table 6.1, lines 22–23, and Fig. 6.11D, E).
Folding was confirmed by electrophoretic mobility measurements (Fig. 6.10A, B), by far-UV CD
spectroscopy, and by functional studies (Leney et al. 2012). SEC showed a single narrow peak for
OmpT/A8-35 complexes, indicating the presence of a homogeneous single species of OmpT, whereas
a broader peak was observed after folding of PagP, indicating the presence of a mixture of folded and
unfolded species, consistent with the ~60% folding yield observed by SDS-PAGE and ESI-IMS-MS
(Leney et al. 2012).
348
6 Amphipol-Assisted Folding of Membrane Proteins