(ii) As illustrated by the delivery of retinal after BO has folded (Dahmane et al. 2013) and by the
exchange of monomers between dimers of the TM α-helix of glycophorin A (GpA) (Stangl
et al. 2014), being trapped with APols does not prevent a MP from interacting and
associating with a molecule delivered by another APol particle.
(iii) By increasing the MP/APol ratio, APol-trapped MPs can be induced to interact with one
another (Zoonens et al. 2007; Gohon et al. 2008; Arunmanee et al. 2014).
(iv) There are indications that MPs folded in A8-35 can dimerize. Whereas no strong, direct
demonstration has been provided yet, this is suggested, as regards GPCRs, by the fact that
there is an optimum protein/A8-35 ratio for the BLT1 leukotriene receptor to fold
(Dahmane et al. 2009). As discussed in Box 6.2, in the absence of lipids or any other
cofactor that could become diluted, the most straightforward interpretation of this observation is that newly folded receptors become stabilized by dimerization, which occurs less
efficiently in the presence of too large an excess of APol. SEC analysis of the solution
behavior of the BLT1, BLT2, CB1, and 5-HT 4(a) receptors folded in A8-35 does indeed
suggest partial dimerization (Dahmane et al. 2009). Similarly, as noted above, dimers of the
TM α-helix of GpA can exchange monomers after being trapped in A8-35 (Stangl et al.
2014). This strongly suggests that monomers of GpA folded in APols should be able to
dimerize.
One of the main challenges in folding oligomeric MPs is that, in most cases, unassembled
monomers can be expected to be only marginally stable. Conditions must be found that favor their
folding and assembly while discouraging the formation of improper intermolecular interactions. Given
that it can be done in detergent or mixed lipid/detergent micelles (Fig. 6.4B), there is every reason to
hope that APols will make it easier. It is likely that, in most cases, simultaneous (re)folding of all
subunits will turn out to be the best strategy. In cases where unassembled monomers have very
different stabilities, however, alternative routes could be experimented with, such as folding or
expressing the most unstable subunit in the presence of its already folded partner(s).
6.4
Protocol 6.1. Amphipol-Assisted Folding of Membrane Proteins
APols have proven to be very helpful in folding MPs expressed as inclusion bodies in E. coli, like class
A GPCRs or porins (Pocanschi et al. 2006, 2013; Dahmane et al. 2009, 2011, 2013; Banères et al.
2011; Bazzacco et al. 2012) (Fig. 6.3). The protocol used for α-helical MPs is derived from one
initially developed to refold BR in lipids (see Popot et al. 1987, in which many useful practical details
can be found). It is quite simple but requires some optimization regarding the quantity of APol to add
and the presence or not of lipids. For variants and the effect of various modifications to this protocol,
see Dahmane et al. (2013). Comments are in italics and preceded with a pointing hand ( ).
6.4.1
Solubilization and Purification of MPs in Denaturing Conditions
Inclusion bodies are aggregates principally comprising misfolded forms of the protein of interest,
but they can also contain some DNA and other bacterial proteins. They need to be solubilized and
purified in denaturing conditions. For α-helical MPs, the denaturing agent is usually SDS, whereas for
β-barrel MPs it is urea. Purification is most often carried out by affinity chromatography. For
instance, if the protein of interest is fused to a polyhistidine tag, purification can be carried out on
354
6 Amphipol-Assisted Folding of Membrane Proteins
exchange of monomers between dimers of the TM α-helix of glycophorin A (GpA) (Stangl
et al. 2014), being trapped with APols does not prevent a MP from interacting and
associating with a molecule delivered by another APol particle.
(iii) By increasing the MP/APol ratio, APol-trapped MPs can be induced to interact with one
another (Zoonens et al. 2007; Gohon et al. 2008; Arunmanee et al. 2014).
(iv) There are indications that MPs folded in A8-35 can dimerize. Whereas no strong, direct
demonstration has been provided yet, this is suggested, as regards GPCRs, by the fact that
there is an optimum protein/A8-35 ratio for the BLT1 leukotriene receptor to fold
(Dahmane et al. 2009). As discussed in Box 6.2, in the absence of lipids or any other
cofactor that could become diluted, the most straightforward interpretation of this observation is that newly folded receptors become stabilized by dimerization, which occurs less
efficiently in the presence of too large an excess of APol. SEC analysis of the solution
behavior of the BLT1, BLT2, CB1, and 5-HT 4(a) receptors folded in A8-35 does indeed
suggest partial dimerization (Dahmane et al. 2009). Similarly, as noted above, dimers of the
TM α-helix of GpA can exchange monomers after being trapped in A8-35 (Stangl et al.
2014). This strongly suggests that monomers of GpA folded in APols should be able to
dimerize.
One of the main challenges in folding oligomeric MPs is that, in most cases, unassembled
monomers can be expected to be only marginally stable. Conditions must be found that favor their
folding and assembly while discouraging the formation of improper intermolecular interactions. Given
that it can be done in detergent or mixed lipid/detergent micelles (Fig. 6.4B), there is every reason to
hope that APols will make it easier. It is likely that, in most cases, simultaneous (re)folding of all
subunits will turn out to be the best strategy. In cases where unassembled monomers have very
different stabilities, however, alternative routes could be experimented with, such as folding or
expressing the most unstable subunit in the presence of its already folded partner(s).
6.4
Protocol 6.1. Amphipol-Assisted Folding of Membrane Proteins
APols have proven to be very helpful in folding MPs expressed as inclusion bodies in E. coli, like class
A GPCRs or porins (Pocanschi et al. 2006, 2013; Dahmane et al. 2009, 2011, 2013; Banères et al.
2011; Bazzacco et al. 2012) (Fig. 6.3). The protocol used for α-helical MPs is derived from one
initially developed to refold BR in lipids (see Popot et al. 1987, in which many useful practical details
can be found). It is quite simple but requires some optimization regarding the quantity of APol to add
and the presence or not of lipids. For variants and the effect of various modifications to this protocol,
see Dahmane et al. (2013). Comments are in italics and preceded with a pointing hand ( ).
6.4.1
Solubilization and Purification of MPs in Denaturing Conditions
Inclusion bodies are aggregates principally comprising misfolded forms of the protein of interest,
but they can also contain some DNA and other bacterial proteins. They need to be solubilized and
purified in denaturing conditions. For α-helical MPs, the denaturing agent is usually SDS, whereas for
β-barrel MPs it is urea. Purification is most often carried out by affinity chromatography. For
instance, if the protein of interest is fused to a polyhistidine tag, purification can be carried out on
354
6 Amphipol-Assisted Folding of Membrane Proteins