question of whether APols can contribute usefully to MP crystallography, they will be presented
somewhat more in detail below (§ 11.2.2).
A critically important factor has been the genetic engineering and biochemical manipulation of
MPs, so as to make them more amenable to crystallization. Schematically, this engineering has taken
two main paths:
• Modifying the sequence of the protein so as (i) to eliminate as much flexibility as possible –
e.g. by removing mobile loops and tails and separating domains that tend to move one with
respect to the other; an approach that is classical with soluble proteins – and (ii) to make the
target MPs more stable by selecting detergent-resistant mutants (see e.g. Zhou and Bowie
2000; Bowie 2001; Magnani et al. 2008; Sarkar et al. 2008; Serrano-Vega et al. 2008; Shibata
et al. 2009; Tate 2010; Miller and Tate 2011; Lluis et al. 2013; Egloff et al. 2014; Vaidehi
et al. 2016, and discussion in Chap. 2, § 2.5.2).
• Decorating the target protein with water-soluble domains in order to (i) increase the hydrophilic surface available for establishing crystal contacts, (ii) move the proteins farther one
from another in the crystal, which can limit interference by the detergent belts, (iii) stabilize
the protein, and (iv) possibly select among conformational states. There are two ways to
introduce these domains:
– One is to co-crystallize the target protein along with soluble binders, such as natural or
engineered antibody F ab or F v fragments or camelid single-chain antibody fragments
(nanobodies) (see e.g. Ostermeier et al. 1995, 1997; Hunte et al. 2000; Zhou et al. 2001;
Uysal et al. 2009; Hibbs and Gouaux 2011; Rasmussen et al. 2011; see, below, Fig. 11.4).
Alternatively, one can resort to designed artificial proteins such as DARPins (Binz et al.
2003; Sennhauser and Grütter 2008; Boersma and Plückthun 2011; Jost and Plückthun
2014; Plückthun 2015) or αReps (Urvoas et al. 2012; Ferrandez et al. 2014).
– Another is to insert genetically a soluble domain, such as T4 lysozyme, into the sequence
of the target protein itself, e.g. by substituting it for a flexible loop (see e.g. Cherezov et al.
2007; Rosenbaum et al. 2007 and, below, Fig. 11.3).
• A third approach, which requires no tinkering with the target protein but can be combined
with it, is to screen orthologs from many species in order to identify those that express well
and are particularly stable (see e.g. Love et al. 2010).
As for any protein, it is of course good politics to crystallize a fragile and/or flexible MP along
with a high-affinity ligand, whose binding will stabilize it and shift the conformational equilibrium
toward a single state.
11.2.2 The Formation of Membrane Protein Crystals
In order to form a clear view of the advantages and disadvantages APols can present for MP
crystallization, it is necessary to examine the structure of MP crystals and the various media in
which they can form. The first MP crystals were obtained in aqueous solutions, by inducing
MP/detergent or MP/lipid/detergent complexes to associate into 3D crystals, at first glance much
like soluble proteins do (§ 11.2.2.1). A second, more recent approach is to induce the formation of the
crystals within 3D lipidic phases, in which MPs diffuse not as soluble complexes but as membraneembedded entities (§ 11.2.2.2).
500
11 Amphipols and Membrane Protein Crystallization
somewhat more in detail below (§ 11.2.2).
A critically important factor has been the genetic engineering and biochemical manipulation of
MPs, so as to make them more amenable to crystallization. Schematically, this engineering has taken
two main paths:
• Modifying the sequence of the protein so as (i) to eliminate as much flexibility as possible –
e.g. by removing mobile loops and tails and separating domains that tend to move one with
respect to the other; an approach that is classical with soluble proteins – and (ii) to make the
target MPs more stable by selecting detergent-resistant mutants (see e.g. Zhou and Bowie
2000; Bowie 2001; Magnani et al. 2008; Sarkar et al. 2008; Serrano-Vega et al. 2008; Shibata
et al. 2009; Tate 2010; Miller and Tate 2011; Lluis et al. 2013; Egloff et al. 2014; Vaidehi
et al. 2016, and discussion in Chap. 2, § 2.5.2).
• Decorating the target protein with water-soluble domains in order to (i) increase the hydrophilic surface available for establishing crystal contacts, (ii) move the proteins farther one
from another in the crystal, which can limit interference by the detergent belts, (iii) stabilize
the protein, and (iv) possibly select among conformational states. There are two ways to
introduce these domains:
– One is to co-crystallize the target protein along with soluble binders, such as natural or
engineered antibody F ab or F v fragments or camelid single-chain antibody fragments
(nanobodies) (see e.g. Ostermeier et al. 1995, 1997; Hunte et al. 2000; Zhou et al. 2001;
Uysal et al. 2009; Hibbs and Gouaux 2011; Rasmussen et al. 2011; see, below, Fig. 11.4).
Alternatively, one can resort to designed artificial proteins such as DARPins (Binz et al.
2003; Sennhauser and Grütter 2008; Boersma and Plückthun 2011; Jost and Plückthun
2014; Plückthun 2015) or αReps (Urvoas et al. 2012; Ferrandez et al. 2014).
– Another is to insert genetically a soluble domain, such as T4 lysozyme, into the sequence
of the target protein itself, e.g. by substituting it for a flexible loop (see e.g. Cherezov et al.
2007; Rosenbaum et al. 2007 and, below, Fig. 11.3).
• A third approach, which requires no tinkering with the target protein but can be combined
with it, is to screen orthologs from many species in order to identify those that express well
and are particularly stable (see e.g. Love et al. 2010).
As for any protein, it is of course good politics to crystallize a fragile and/or flexible MP along
with a high-affinity ligand, whose binding will stabilize it and shift the conformational equilibrium
toward a single state.
11.2.2 The Formation of Membrane Protein Crystals
In order to form a clear view of the advantages and disadvantages APols can present for MP
crystallization, it is necessary to examine the structure of MP crystals and the various media in
which they can form. The first MP crystals were obtained in aqueous solutions, by inducing
MP/detergent or MP/lipid/detergent complexes to associate into 3D crystals, at first glance much
like soluble proteins do (§ 11.2.2.1). A second, more recent approach is to induce the formation of the
crystals within 3D lipidic phases, in which MPs diffuse not as soluble complexes but as membraneembedded entities (§ 11.2.2.2).
500
11 Amphipols and Membrane Protein Crystallization