purification, more than 130 proteins where found to be associated to NADPH-dependent cytochrome
P450 oxidoreductase, including two cytochrome P450 enzymes specifically involved in the synthesis
of dhurrin, a compound that releases cyanide as a defense against predators. Functional data support
the existence of a metabolon (a complex in which metabolites are channeled from one enzyme to the
next) involving these three P450s, all of them bitopic MPs, and a soluble partner (Laursen et al. 2016).
It will be very interesting to examine more in depth which complexes can be extracted intact under
which conditions, which not, and why.
It has been reported that there seems to be a limit to the size of the complexes that can be
encapsulated in SMALPs (Lee et al. 2016). This is rather unexpected, given that, by varying the ratio of
SMA to lipids, it is possible to form very large discs (~90–100 nm in diameter) (Li et al. 2015a; Zhang
et al. 2015b).
SMA extracts MPs from native membranes along with a complement of lipids, another very
interesting characteristic that will be discussed in § 5.3.1.2.
A diisobutylene/maleic acid copolymer (DIBMA) has recently been shown to exhibit similar
performances to SMA in solubilizing phospholipids, stabilizing an integral membrane enzyme, and
extracting MPs from biomembranes, with advantages linked to the absence of phenyl rings: no
absorbance in near-UV and less perturbation of the lipids (Oluwole et al. 2017).
5.2.2.2 Can Membrane Proteins Be Directly Solubilized Using Polyacrylate-Based
Amphipols?
In the initial stages of the development of conventional APols (A8-35 and other polyacrylate-based
APols), attempts were made to use them to directly extract MPs from native membranes, namely the
purple membrane from Halobacterium salinarum and thylakoid membranes from Chlamydomonas
reinhardtii, two membranes whose protein content is very high. They showed very little efficiency,
leading to the current approach of extracting the target protein with a detergent and then transferring it
to APols. Over the years, however, a couple of exceptions were encountered. Using A8-35, it was
observed that, according to immunoblots, close to half of the FGK 2 maltose transporter could be
extracted directly from E. coli plasma membranes, that is, about as much as could be solubilized by
DDM under the same conditions (M. Zoonens and H.A. Shuman, unpublished data quoted in Popot
et al. 2003). Similarly, the human insulin receptor was totally extracted by A8-35 from the Chinese
hamster ovary cells where it had been overexpressed (Gérard Crémel and colleagues, unpublished data
quoted in Popot et al. 2003). A summary of these two sets of experiments is provided in Box 5.1.
Preliminary studies with C. reinhardtii thylakoid membranes using mixtures of APols and detergent in
various proportions showed that photosynthetic complexes can be extracted to various degrees
(Bazzacco 2009), possibly a way to gently extract fragile supercomplexes. Whole proteomes, including MPs, have been extracted from human cells using a high concentration of A8-35 combined with
sonication (Ning et al. 2013), but no functional studies were carried out (see Chap. 14, § 14.4).
Box 5.1 Direct Extraction of Membrane Proteins by A8-35
In Popot et al. (2003), it was briefly mentioned that two sets of preliminary experiments indicated
that certain MPs (or MPs from certain membranes) can be directly extracted with A8-35. Some
details are provided below.
B5.1.1. Direct extraction of insulin receptors from Chinese hamster ovary (CHO) cells
In order to test whether the insulin receptor expressed in CHO cells and solubilized with Triton
X-100 (TX-100) can be transferred to A8-35 under a functional form, the TX-100-solubilized
receptor was immobilized on a wheat-germ agglutinin (WGA) column (Leray et al. 1993), washed,
and eluted by 0.3 M acetylglucosamine in the presence either of CHAPS or of A8-35. The receptor
was observed to elute faster in the presence of A8-35 than of CHAPS. Determination of the protein
content and radioactive insulin binding in the eluate indicated that the specific activity was twice
5.2 Forming Membrane Protein/Amphipol Complexes
261
P450 oxidoreductase, including two cytochrome P450 enzymes specifically involved in the synthesis
of dhurrin, a compound that releases cyanide as a defense against predators. Functional data support
the existence of a metabolon (a complex in which metabolites are channeled from one enzyme to the
next) involving these three P450s, all of them bitopic MPs, and a soluble partner (Laursen et al. 2016).
It will be very interesting to examine more in depth which complexes can be extracted intact under
which conditions, which not, and why.
It has been reported that there seems to be a limit to the size of the complexes that can be
encapsulated in SMALPs (Lee et al. 2016). This is rather unexpected, given that, by varying the ratio of
SMA to lipids, it is possible to form very large discs (~90–100 nm in diameter) (Li et al. 2015a; Zhang
et al. 2015b).
SMA extracts MPs from native membranes along with a complement of lipids, another very
interesting characteristic that will be discussed in § 5.3.1.2.
A diisobutylene/maleic acid copolymer (DIBMA) has recently been shown to exhibit similar
performances to SMA in solubilizing phospholipids, stabilizing an integral membrane enzyme, and
extracting MPs from biomembranes, with advantages linked to the absence of phenyl rings: no
absorbance in near-UV and less perturbation of the lipids (Oluwole et al. 2017).
5.2.2.2 Can Membrane Proteins Be Directly Solubilized Using Polyacrylate-Based
Amphipols?
In the initial stages of the development of conventional APols (A8-35 and other polyacrylate-based
APols), attempts were made to use them to directly extract MPs from native membranes, namely the
purple membrane from Halobacterium salinarum and thylakoid membranes from Chlamydomonas
reinhardtii, two membranes whose protein content is very high. They showed very little efficiency,
leading to the current approach of extracting the target protein with a detergent and then transferring it
to APols. Over the years, however, a couple of exceptions were encountered. Using A8-35, it was
observed that, according to immunoblots, close to half of the FGK 2 maltose transporter could be
extracted directly from E. coli plasma membranes, that is, about as much as could be solubilized by
DDM under the same conditions (M. Zoonens and H.A. Shuman, unpublished data quoted in Popot
et al. 2003). Similarly, the human insulin receptor was totally extracted by A8-35 from the Chinese
hamster ovary cells where it had been overexpressed (Gérard Crémel and colleagues, unpublished data
quoted in Popot et al. 2003). A summary of these two sets of experiments is provided in Box 5.1.
Preliminary studies with C. reinhardtii thylakoid membranes using mixtures of APols and detergent in
various proportions showed that photosynthetic complexes can be extracted to various degrees
(Bazzacco 2009), possibly a way to gently extract fragile supercomplexes. Whole proteomes, including MPs, have been extracted from human cells using a high concentration of A8-35 combined with
sonication (Ning et al. 2013), but no functional studies were carried out (see Chap. 14, § 14.4).
Box 5.1 Direct Extraction of Membrane Proteins by A8-35
In Popot et al. (2003), it was briefly mentioned that two sets of preliminary experiments indicated
that certain MPs (or MPs from certain membranes) can be directly extracted with A8-35. Some
details are provided below.
B5.1.1. Direct extraction of insulin receptors from Chinese hamster ovary (CHO) cells
In order to test whether the insulin receptor expressed in CHO cells and solubilized with Triton
X-100 (TX-100) can be transferred to A8-35 under a functional form, the TX-100-solubilized
receptor was immobilized on a wheat-germ agglutinin (WGA) column (Leray et al. 1993), washed,
and eluted by 0.3 M acetylglucosamine in the presence either of CHAPS or of A8-35. The receptor
was observed to elute faster in the presence of A8-35 than of CHAPS. Determination of the protein
content and radioactive insulin binding in the eluate indicated that the specific activity was twice
5.2 Forming Membrane Protein/Amphipol Complexes
261