• Study 12.12: TRPM1, the transient receptor potential melastatin-1, solved in A8-35 to a
resolution of 22 Å at FSC ¼ 0.143 (Agosto et al. 2014).
• Study 12.14: the CsgG/CsgE complex involved in the formation of bacterial curli, solved in
A8-35 to a resolution of 24 Å at FSC ¼ 0.5 (Goyal et al. 2014).
• Study 12.17: the ABCB1 P-glycoprotein (PgP), an ABC drug transporter, solved in SMALPs
at a resolution of ~35 Å (Gulati et al. 2014).
• Study 12.18: the TRPV2 thermosensor channel, solved in A8-35 at ~15-Å resolution (Fan
et al. 2014).
• Study 12.23: the structure of E. coli holotranslocon was deduced from an ensemble of
biophysical data, including 14–15-Å resolution cryo-EM images of the translocon stabilized
with A8-35 (Botte et al. 2016).
• Study 12.24: the structure of the zebrafish STRA6 receptor, which mediates the internalization of retinol, solved to 3.9-Å resolution following trapping with A8-35 (Chen et al. 2016).
• Study 12.26: the structure of an engineered AcrAB-TolC complex, trapped in A8-35, solved
to 8.2-Å resolution (Jeong et al. 2016); the authors show a comparison of particle distribution
illustrating the crowding at the edge of the holes in detergent solution as compared to a much
more even distribution in A8-35.
• Study 12.27: the resolution of A8-35-stabilized respirasome was pushed from 19 Å (Study
12.5; Althoff et al. 2011) to ~9 Å (Sousa et al. 2016).
• Study 12.29: TRPV2 is regulated by temperature; by ligands, such as probenecid and
cannabinoids; and by lipids. Following transfer from a DDM/lipid mixture to A8-35, cryoEM single-particle images were analyzed to a resolution of ~4 Å (Zubcevic et al. 2016).
Putative bound lipids are identified.
• Study 12.30: cyclic-nucleotide-gated channels belong to the voltage-gated ion channel
superfamily, but their activities are controlled by intracellular cyclic nucleotides. The structure of a cyclic-nucleotide-gated channel trapped in A8-35 has been solved to 3.5-Å resolution (Li et al. 2017b).
• Study 12.31: the structure of the TM pore formed by the Streptococcus toxin pneumolysin
was solved to 4.5-Å resolution following extraction from liposomes with Cymal-6 and
transfer to A8-35 (van Pee et al. 2017).
• Study 12.32: the bacterial efflux pump AcrABZ-TolC was transferred from DDM to A8-35
and its structure in the presence or absence of an inhibitor solved, respectively, to 3.6-Å and
6.5-Å resolution (Wang et al. 2017).
• Study 12.33: the structure of yet another member of the TRP channel family, polycystin-2,
was solved to 4.2–4.3-Å resolution after transfer from LMNG/cholesterol hemisuccinate
solution to A8-35 (Wilkes et al. 2017).
Two further high-resolution studies of APol-trapped MPs have appeared, that of γ-secretase, to
3.4–4.5 Å (Study 12.15; Lu et al. 2014; Bai et al. 2015b,c), and that of the TRPA1 channel (Study
12.21; three structures at 4.24, 3.9, and 4.7 Å, depending on the ligand bound; Paulsen et al. 2015).
They are briefly described in Table 12.1 but each of them deserves a special comment here.
The four components of the human γ-secretase were co-expressed in mammalian HEK293F cells
(for a recent review about co-expression of multisubunit complexes, see Zorman et al. 2015). The
complex was extracted and purified in CHAPSO and transferred to digitonin. However, the complexes
in digitonin yielded poorly exploitable cryo-EM images (Fig. 12.24A). The complexes were therefore
transferred to A8-35 using Bio-Beads and free APol removed by SEC, resulting in significant
improvement (Fig. 12.24B). Combined with the use of a direct-electron detector in single-electron
12.3 The Use of Amphipols for Membrane Protein Electron Microscopy Studies
569
resolution of 22 Å at FSC ¼ 0.143 (Agosto et al. 2014).
• Study 12.14: the CsgG/CsgE complex involved in the formation of bacterial curli, solved in
A8-35 to a resolution of 24 Å at FSC ¼ 0.5 (Goyal et al. 2014).
• Study 12.17: the ABCB1 P-glycoprotein (PgP), an ABC drug transporter, solved in SMALPs
at a resolution of ~35 Å (Gulati et al. 2014).
• Study 12.18: the TRPV2 thermosensor channel, solved in A8-35 at ~15-Å resolution (Fan
et al. 2014).
• Study 12.23: the structure of E. coli holotranslocon was deduced from an ensemble of
biophysical data, including 14–15-Å resolution cryo-EM images of the translocon stabilized
with A8-35 (Botte et al. 2016).
• Study 12.24: the structure of the zebrafish STRA6 receptor, which mediates the internalization of retinol, solved to 3.9-Å resolution following trapping with A8-35 (Chen et al. 2016).
• Study 12.26: the structure of an engineered AcrAB-TolC complex, trapped in A8-35, solved
to 8.2-Å resolution (Jeong et al. 2016); the authors show a comparison of particle distribution
illustrating the crowding at the edge of the holes in detergent solution as compared to a much
more even distribution in A8-35.
• Study 12.27: the resolution of A8-35-stabilized respirasome was pushed from 19 Å (Study
12.5; Althoff et al. 2011) to ~9 Å (Sousa et al. 2016).
• Study 12.29: TRPV2 is regulated by temperature; by ligands, such as probenecid and
cannabinoids; and by lipids. Following transfer from a DDM/lipid mixture to A8-35, cryoEM single-particle images were analyzed to a resolution of ~4 Å (Zubcevic et al. 2016).
Putative bound lipids are identified.
• Study 12.30: cyclic-nucleotide-gated channels belong to the voltage-gated ion channel
superfamily, but their activities are controlled by intracellular cyclic nucleotides. The structure of a cyclic-nucleotide-gated channel trapped in A8-35 has been solved to 3.5-Å resolution (Li et al. 2017b).
• Study 12.31: the structure of the TM pore formed by the Streptococcus toxin pneumolysin
was solved to 4.5-Å resolution following extraction from liposomes with Cymal-6 and
transfer to A8-35 (van Pee et al. 2017).
• Study 12.32: the bacterial efflux pump AcrABZ-TolC was transferred from DDM to A8-35
and its structure in the presence or absence of an inhibitor solved, respectively, to 3.6-Å and
6.5-Å resolution (Wang et al. 2017).
• Study 12.33: the structure of yet another member of the TRP channel family, polycystin-2,
was solved to 4.2–4.3-Å resolution after transfer from LMNG/cholesterol hemisuccinate
solution to A8-35 (Wilkes et al. 2017).
Two further high-resolution studies of APol-trapped MPs have appeared, that of γ-secretase, to
3.4–4.5 Å (Study 12.15; Lu et al. 2014; Bai et al. 2015b,c), and that of the TRPA1 channel (Study
12.21; three structures at 4.24, 3.9, and 4.7 Å, depending on the ligand bound; Paulsen et al. 2015).
They are briefly described in Table 12.1 but each of them deserves a special comment here.
The four components of the human γ-secretase were co-expressed in mammalian HEK293F cells
(for a recent review about co-expression of multisubunit complexes, see Zorman et al. 2015). The
complex was extracted and purified in CHAPSO and transferred to digitonin. However, the complexes
in digitonin yielded poorly exploitable cryo-EM images (Fig. 12.24A). The complexes were therefore
transferred to A8-35 using Bio-Beads and free APol removed by SEC, resulting in significant
improvement (Fig. 12.24B). Combined with the use of a direct-electron detector in single-electron
12.3 The Use of Amphipols for Membrane Protein Electron Microscopy Studies
569