CHAPTER THREE

Membrane Preparation
and Solubilization
Ankita Roy1
Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
1
Corresponding author: e-mail address: amr161@pitt.edu

Contents
1. Introduction 46
2. Membrane Preparation 47
2.1 Choice of buffer 47
2.2 Cell lysis 48
2.3 Membrane fractionation 48
2.4 Methodology 49
3. Solubilization 50
3.1 Types of detergents 51
3.2 Choice of detergent 52
3.3 Stability in detergent 52
3.4 Alternative solubilization strategy 53
3.5 Presolubilization screen 53
3.6 Methodology 54
References 55

Abstract
Membrane proteins play an essential role in several biological processes like ion trans-
port, signal transduction, and electron transfer to name a few. For structural and func-
tional studies of integral membrane proteins, it is critically important to isolate proteins
from the membrane using biological detergents. Detergents disrupt the native lipid
components of the native membrane and encase the membrane protein in an unnat-
ural environment in aqueous solution. However, a particular membrane protein is best
solubilized in a specific detergent; therefore, screening for the optimal detergent is
essential. Apart from keeping the membrane protein monodispered in solution, the
detergent has to be compatible with downstream processes to isolate and characterize
a membrane protein. Over the past several years, a number of membrane proteins have
been successfully isolated for structural and functional studies that allowed an outline of
general strategies for isolating a novel membrane protein of interest.

Methods in Enzymology, Volume 557 # 2015 Elsevier Inc. 45

ISSN 0076-6879 All rights reserved.
http://dx.doi.org/10.1016/bs.mie.2014.11.044
46 Ankita Roy

1. INTRODUCTION
Cell membranes act as a physiological barrier separating the external
environment from the internal milieu. Membranes are rich in integral mem-
brane proteins that play diverse role in cellular processes ranging from trans-
port of nutrients, ions, electron transfer, and signal transduction to several
other processes. Additionally, cells are packed with internal membranes that
are rich in membrane proteins. As approximately 70% of current day med-
ications are targeted against membrane proteins, it is essential to elucidate the
structure of these proteins. Though 30% of the human genome codes for
integral membrane proteins, only 30,000 structures deposited in the protein
data bank, less than 1%, represent membrane protein. The paucity of the
deposited structures can be attributed to the amphipathic nature of these
proteins.
There are several limitations in purifying membrane proteins for struc-
tural and functional studies. Firstly, membrane proteins are present in
minute amounts in native tissues. However for structural studies, milligram
amounts of protein must be isolated. To overcome this problem, recombi-
nant proteins are overexpressed in heterologous host organisms. Although
Escherichia coli has been used extensively to overexpress and purify membrane
proteins, other bacterial hosts such as Lactococcus lactis, yeast expression sys-
tems such as Pichia pastoris and Saccharomyces cerevisiae, and insect and mam-
malian expression systems have been routinely used for the purpose
(Lundstrom et al., 2006; Morello et al., 2008; Sahdev, Khattar, & Saini,
2008; Shukla, Reinhart, & Michel, 2006). To heterologously express and
purify a membrane protein, systematic screening should be performed
which involves expression analysis using a variety of host organisms, expres-
sion vectors, and expression conditions (Surade et al., 2006).
Often overexpression of membrane proteins can be problematic; as large
quantities of the protein are expressed using a strong promoter, due to lack of
membranes, the expressed proteins fail to fold properly. To overcome this
problem, membrane-rich expression hosts like Rhodobacter sphaeroides have
been developed. These photosynthetic organisms are genetically engineered
to provide excess membrane surface area in comparison to typical expression
hosts for harboring overexpressed membrane proteins (Laible, Scott,
Henry, & Hanson, 2004; Roy, Shukla, Haase, & Michel, 2008).
Secondly, membrane proteins must be solubilized and purified from the
membranes using small amphipathic molecules called detergents. Several
Membrane Preparation and Solubilization 47

types of detergents have been used to isolate membrane proteins; however, it

is important to select a detergent that not only maximally extracts the protein
from the membrane but also stabilizes it. The detergent used to solubilize a
membrane protein must also be compatible with the purification strategy
and crystallization. Often detergent mixtures have been employed in the
purification and crystallization stages to improve crystal quality for structural
studies.
In recent years, high-throughput platforms like JCSG, MePNet, EMeP,
etc., have generated structure determination pipelines which involve work-
ing with larger set of proteins. Large numbers of membrane proteins are
processed in parallel; this is advantageous from the standpoint of optimiza-
tion of strategies involved in the structure determination (Lundstrom, 2006).
This chapter describes a general protocol for membrane preparation and
then shows how to screen for the most suitable detergent to solubilize a
protein.

2. MEMBRANE PREPARATION
Membrane preparation is the first step in the isolation and purification
of a membrane protein. Membrane preparation involves disruption of the
cell membrane using mechanical methods and differential centrifugation,
which helps to isolate the membranes from soluble proteins, nuclear mate-
rial, and other internal membranes. Membrane preparation is an important
step in purifying a membrane protein, as isolation of membranes, in itself, is a
purification step. Soluble proteins and other membrane-associated proteins
(like outer membrane proteins, mitochondria associated proteins, etc.) are
removed from the membrane fraction. Moreover, a good membrane prep-
aration ensures reproducibility amongst purification preps, which is a pre-
requisite for successful crystallization trials.

2.1. Choice of buffer

Membranes are prepared by breaking open cells in a suitable buffer such as
Tris, HEPES, and phosphate buffer. The pH of the buffer is critical as it plays
a role in protein solubility and stability. Additionally, buffers also affect
binding and elution of a protein from a column matrix during purification.
Care should also be taken to check if the properties of the buffer affect the
enzymatic activity of a membrane protein. As detergents are required in sub-
sequent steps of membrane protein isolation, the ionic strength of the buffer
should be compatible with the detergent used for solubilization and
48 Ankita Roy

purification. Postcell disruption, several proteases are released from the cell
that degrades the target protein thereby reducing the yield during the iso-
lation steps. Protease inhibitors such as Complete Protease Inhibitor Cock-
tails should be added to the buffer in each step of the protein isolation
process. Reducing agents like dithiothreitol should be present in the buffer
to prevent oxidation of disulfide bonds in a protein.

2.2. Cell lysis

Based on the source material, cells are lysed using a variety of methods. Cells
with tough walls are broken using mechanical force such as ultrasonication,
homogenization involving French Press, Microfluidizer, or mechanical lysis
using glass beads-based agitation and nitrogen cavitation method. For
smaller volumes of cells, osmotic shock, repeated cycles of freeze thaw or
enzymatic lysis using lysozyme or zymolase can be employed (Table 1).

2.3. Membrane fractionation

Postcell lysis, membranes are fractionated using differential centrifugation.
Membranes are composed of lipid–protein complexes, which have a specific

Table 1 Cell lysis methods for large-scale membrane preparation

Lysis method Apparatus Advantages Disadvantages Cells lysed
Ultrasonication Sonicator Easy to use, Ear protection Wide
(20–50 kHz in applied to required, variety of
short pulses of various cell prevent cell types
5–15 s) types generation of
heat
Homogenization French Press: Effective in Prone to Wide
18,000–30,000 psi small volumes clogging variety of
cell types
Microfluidizer: Effective for Cooling Wide
20,000–30,000 psi larger volume, required variety of
reproducible cell types
Mechanical Beads-based Effective in Not Mammalian
agitation small volumes reproducible, cells
heat generated
Nitrogen cell Not effective Care in Mammalian
cavitation for cells with handling cells
tough wall
Membrane Preparation and Solubilization 49

density; this allows them to be separated from soluble proteins and other
organelles like mitochondria and nuclei using centrifugation techniques.
Sucrose density centrifugation can be employed to separate different types
of membranes. The inner membranes like the intracytoplasmic membranes
of R. sphaeroides that houses the photosynthetic proteins, such as the reaction
center and the light harvesting complexes can be separated from the cell
membrane using sucrose density gradient (Rafferty & Clayton, 1997).

2.4. Methodology
Described below is a general approach to prepare membrane fragments for
subsequent solubilization and purification of a membrane protein from
E. coli.
Materials
1. Source material
2. Low-speed centrifuge
3. Ultracentrifuge
4. Instrument for lysis
5. Bicinchoninic acid (BCA) protein estimation kit
6. Spectrophotometer (UV/Vis)
Methods
1. E. coli cells were harvested when OD600 of about 1.5 was reached.
2. Washed bacterial cell pellet (5 g) was resuspended in 10 ml of ice-cold
breaking buffer (20 mM HEPES, 100 mM NaCl, 1 mM phenylmethyl
sulfonyl fluoride, EDTA-free protease inhibitors).
3. Lysozyme was added at a final concentration of 1 mg/ml.
4. Cells were disrupted by passage through French Press (18,000 psi pres-
sure) at a flow rate of 10 ml/min at 4 °C. The process was repeated
thrice.
5. Unbroken cells, nuclear material and cell debris were removed by low-
speed centrifugation at 5000  g, at 4 °C for 10 min.
6. The supernatant was subjected to ultracentrifugation at 100,000  g at
4 °C for 60 min.
7. The pellet was resuspended in 10 ml of 20 mM HEPES, 100 mM NaCl,
1 mM phenylmethyl sulfonyl fluoride, EDTA-free protease inhibitors,
and ultracentrifuged again at 100,000  g at 4 °C for 30 min.
8. The washed membrane pellet was resuspended in 1 ml of 20 mM
HEPES, 100 mM NaCl, 1 mM phenylmethyl sulfonyl fluoride,
EDTA-free protease inhibitors. Smaller aliquots were flash-frozen and
50 Ankita Roy

stored at 80 °C. A small fraction of the membrane suspension was

diluted and used to determine the protein concentration using BCA pro-
tein quantification method.
Notes
1. Yeast cells have very rigid cell walls that are difficult to break. Vigorous
vortexing of yeast cells mixed with 0.5 mm glass beads for 30 min result
in efficient cell lysis (André et al., 2006).
2. Outer membranes from E. coli can be isolated by two methods. The first
one involves the use of density gradient centrifugation using a 45–70%
sucrose gradient. The second method involves the use of an extraction
strategy using 0.5% Sarkosyl NL97 for 30 min at 28 °C followed by
ultracentrifugation at 10,000  g for 60 min. The outer membrane is
present in the pellet (Bachhawat & Ghosh, 1987).
3. To isolate internal membranes, like intracytoplasmic membranes of
R. sphaeroides rate-zonal centrifugation can be employed to separate
the cell membrane from the ICMs by layering the cell extract on a
5–35% (w/w) sucrose gradient in 1 mM Tris pH 7.5, 1 mM EDTA.
4. BCA method is an efficient method for estimation of membrane protein
concentration as the assay is less sensitive to the detergents and buffer
components. However, other assays like a modified method based on
Lowry protein estimation assay can also be used. This method involves
precipitating proteins from a detergent solubilized solution with
trichloroacetic acid and deoxycholate prior to protein estimation.

3. SOLUBILIZATION
Membrane proteins must be purified to near homogeneity for struc-
tural or functional studies. Integral membrane proteins must be extracted
from the native lipid environment of the membrane in a functional form
using detergents. Detergents are amphipathic molecules; they contain a
polar head group at one end and long hydrocarbon chain at the other
end. In solution, at low concentration, detergents are present as monomers;
however, with increase in the concentration of the detergent above a Crit-
ical Micellar Concentration (CMC), the detergent from micelles. Micelles
are ordered thermodynamically stable spherical structures with polar head
groups facing the exterior, while hydrophobic tails toward the interior. Dur-
ing solubilization, membrane proteins partition from the lipid-rich native
membrane to the detergent micelles. The detergent molecules surround
the hydrophobic regions of the membrane protein, mimicking the lipid
Membrane Preparation and Solubilization 51

bilayer. Usually, a suitable detergent to protein ratio is employed to solubi-

lize an integral membrane protein, which is determined empirically.

3.1. Types of detergents

Detergents are amphipathic in nature, but structurally diverse. They are pri-
marily characterized by the CMC, which is the optimal concentration of the
detergent when micelles start to form. In order words, CMC is the minimal
concentration of detergent that should be present in extraction buffers to
keep the protein monodispersed in solution. CMC of a particular detergent
also depends on several other factors like temperature of solubilization,
buffer pH, and ionic strength. Detergents are classified based on either
the hydrophilic head group or the hydrophobic tail group. Based on the
head group, there are four categories of detergents—nonionic, anionic, cat-
ionic, and zwitterionic. Nonionic detergents are most commonly used for
solubilization. Nonionic polyethylene detergents are characterized by the
presence of a polymeric chain (O–CH2–CH2)N–OH. This class of detergent
is milder and is commonly used to extract peripheral membrane protein.
These mild detergents help retain lipids bound to the protein, thereby
retaining catalytic activity of the protein. Common detergents in this cate-
gory include, C12E9 (Thesit), Brij series, Triton X-100, Tween 20, etc.
Most common detergents used in solubilizing integral membrane proteins
are sugar detergents, which are considered mild. Sugar detergents have a
hydrophobic part linked to a carbohydrate moiety like glucoside, maltoside
via a glycosidic bond. Dodecyl maltoside (DDM), also referred to as
laurylmaltoside (LM), is considered one of the best detergents for purifica-
tion and crystallization of membrane proteins. Digitonin is a detergent
milder than DDM and it preserves mild interactions between membrane
integral proteins, thus used extensively in protein interaction studies. How-
ever, digitonin cannot effectively solubilize membrane proteins, it is used in
combination with other detergents for extraction. Ionic detergents are harsh
than nonionic detergents; however, anionic detergents like cholate and
deoxycholate have been used for extraction of mitochondrial respiratory
proteins (Hatefi, 1978). However, such detergents can solubilize proteins
at a higher pH. Anionic detergents like bile salts can be used at a lower
pH. Zwitterionic detergents have both a positive and negative group.
Lauryldimethylammoniumoxide (LDAO), a zwitterionic detergent, has
two charged groups at higher pH. LDAO has been successfully used to
purify and crystallize reaction center proteins of photosynthetic bacteria,
52 Ankita Roy

some of the earliest structures of membrane proteins solved (Deisenhofer,

Epp, Miki, Huber, & Michel, 1984).
Detergents have also been classified based on the size and flexibility of the
tail. Longer the tail length, greater is the chance of denaturation. Triton
series detergents are the most flexible while bile salts such as CHAPS and
CHAPSO fall in the rigid category.

3.2. Choice of detergent

Choice of a suitable detergent involves selection of the detergent that most
efficiently solubilizes a protein of interest. Firstly, the detergent should pre-
serve the functionality of the target protein. Secondly, the most suitable deter-
gent should have the highest efficiency of extraction in comparison to other
detergents. Thirdly, the detergent should be compatible with the downstream
processes like purification and crystallization. Many times, membrane pro-
teins are solubilized in a detergent; however, in the later purification stages
the detergent is replaced by a second detergent that can promote crystal lattice
formation for structural studies. Detergent exchange can be performed when
the protein is bound to the column matrix; alternatively, dialysis can be
employed which is a time-consuming method. Several studies have shown
that the intrinsic lipids are bound tightly to membrane proteins to retain their
activity. Therefore, care should be taken to choose suitable detergents that
will not delipidate a membrane protein (Palsdottir & Hunte, 2004).

3.3. Stability in detergent

Long-term stability of the purified membrane protein can be assessed by tech-
niques such as size-exclusion chromatography (SEC), SDS-polyacrylamide
gel electrophoresis (SDS-PAGE), circular dichroism (CD), and differential
scanning calorimetry (DSC). However, prior to the purification, some
techniques can be employed to determine the stability of a membrane protein
in a particular detergent. To determine the stability of a protein in a given
detergent, a method was designed whereby the target membrane protein
was tagged with Green Fluorescent Protein (GFP) at the N- or the
C-terminus. Detergent solubilized extracts of cells transfected or transformed
with the GFP fusion constructs were directly loaded on an SEC column and
the stability of the protein was analyzed based on GFP fluorescence. By this
method, the stability of a GFP fusion protein can be determined in an
unpurified heterogeneous sample (Kawate & Gouaux, 2006). In case of pro-
teins like G protein-coupled receptors (GPCRs), radioligand-binding assays
Membrane Preparation and Solubilization 53

can be performed using the different detergents under varying conditions

(Magnani, Shibata, Serrano-Vega, & Tate, 2008; Serrano-Vega, Magnani,
Shibata, & Tate, 2008).
3.4. Alternative solubilization strategy
To maximally extract a membrane protein, detergents are used in excess.
Often, excess detergent in solution obscures biophysical measurements.
In the past decade, several attempts have been made to improve solubiliza-
tion of membrane protein using alternative compounds. One such approach
involves the use of amphipols such as PMAL-B-100, which are detergent-
like amphipathic polymers. The advantage of using amphipols over deter-
gents is that the amphipols are used in stoichiometric quantities and not
in excess quantities like detergents; additionally, they are capable of preserv-
ing the function of most membrane proteins tested so far (Popot et al., 2003).
However, till date there are no data on the use of amphipols in crystallization
of a membrane protein.
3.5. Presolubilization screen
The best approach to select a suitable detergent to solubilize and purify a new
target protein is to employ a presolubilization screen. A presolubilization
screen involves a systematic approach to screen for some detergents
(Table 2), which have been successfully used for functional solubilization
of several membrane proteins. Apart from choosing the right detergent it
is also important to identify the suitable solubilization conditions such as
buffer pH, temperature, and time of solubilization.
Table 2 Commonly used detergents in membrane protein research
Detergent Type Molecular weight CMC (mM)
DDM (LM) Nonionic 511 0.2
DM Nonionic 483 0.2
OG Nonionic 308 15
NG Nonionic 306.4 6.5
C12E8 Nonionic 540 <0.1
LDAO Zwitterionic 229 1
CHAPS Zwitterionic 615 3–10
Fos12 Ionic 351.5 1.5
Fos14 Ionic 395 0.12
54 Ankita Roy

The detergent that maximally extracts the membrane protein leaving lit-
tle or no protein in the pellet fraction can be considered suitable for purifi-
cation and crystallization. At this point, it is important to assess the stability of
the protein. However, in most cases, testing stability before purification is
not possible. In such cases, stability of the protein can be tested only after
purification. Once the conditions of solubilization are standardized using
the presolubilization screen, large-scale solubilization can be performed
by scaling up the amount of the membranes and detergent using the most
suitable solubilization conditions.

3.6. Methodology
Materials
1. Isolated membranes
2. Ultracentrifuge
3. Detergent stocks (stored at 20 °C)
4. BCA protein estimation kit
5. Spectrophotometer (UV/Vis)
6. SDS-polyacrylamide gel
7. Western blot apparatus
8. PVDF/nitrocellulose membrane
9. Protein molecular weight marker
Methods
1. Membrane pellet was resuspended in suitable buffer (20 mM HEPES,
100 mM NaCl, pH 7.4) with protease inhibitors at a total concentration
of 10 mg/ml in 500 μl total volume. Detergents are used at a final con-
centration of 1% except for octylglucoside (OG), which is 2% (generally
two times CMC).
2. The samples are incubated in rotating shaker at 4, 37 °C, or room tem-
perature for 2 h.
3. The solubilized membranes are centrifuged at 100,000  g at 4 °C for 1 h.
4. The supernatant is loaded on an SDS-PAGE and analyzed as described
previously; the amount of protein solubilized is quantified based on the
unsolubilized control protein used (Roy et al., 2008).
Notes
1. Following presolubilization screening, detergent titration is performed
to ensure reproducible solubilization. This can be accomplished in
two steps: in the first step, different amount of membrane sample
2.5–10 mg/ml is taken and solubilized with the same amount of
Membrane Preparation and Solubilization 55

detergent under previously determined optimal conditions. The amount

solubilized is detected using SDS-PAGE analysis and quantification.
Similarly, different detergent concentration can also be tested using
the same amount of starting material. These initial experiments provide
sufficient information for maintaining reproducibility of solubilization
from batch to batch. Based on all these parameters, large-scale solubili-
zation can be performed prior to purification.
2. At this stage, the activity of the target protein is also tested to identify the
detergent in which the protein is functionally stable. In case of GPCRs,
radioligand-binding assays can be performed in different detergents. The
advantage of this assay is that it can be performed using picomolar
amounts of unpurified receptor proteins. The method is reproducible
and easy to perform. It has been previously shown that ligand binding
can also increase the stability of the GPCR during solubilization. How-
ever, in case of other membrane proteins such as transporters, such assays
cannot be used, the stability of the protein can be determined by recon-
stitution in lipid vesicles only after purification.
3. GPCRs contain characteristic cholesterol binding clefts, which were
identified based on the crystal structure of β1 and β2 adrenergic receptors
(Hanson et al., 2008). Cholesterol analogs such as 0.2% (w/v) cholesterol
hemisuccinate (CHS) are added to stabilize the protein of interest (Lopez
et al., 2008). Dynamic light scattering and small-angle X-ray scattering
have shown that CHS forms bicelle-like architecture when mixed with
DDM (Thompson et al., 2011).
4. Integral lipids are required for stability of some membrane proteins;
longer solubilization time should be avoided as the proteins might get
delipidated.

REFERENCES
André, N., Cherouati, N., Prual, C., Steffan, T., Zeder-Lutz, G., Magnin, T., et al. (2006).
Enhancing functional production of G protein-coupled receptors in Pichia pastoris to
levels required for structural studies via a single expression screen. Protein Science,
15(5), 1115–1126.
Bachhawat, A. K., & Ghosh, S. (1987). Isolation and characterization of outer membrane
proteins of Azospirillum brasilense. Journal of General Microbiology, 133, 1751–1758.
Deisenhofer, J., Epp, O., Miki, K., Huber, R., & Michel, H. (1984). X-ray structure analysis
of a membrane protein complex. Electron density map at 3 A resolution and a model of
the chromophores of the photosynthetic reaction center from Rhodopseudomonas viridis.
Journal of Molecular Biology, 180(2), 385–398.
Hanson, M. A., Cherezov, V., Griffith, M. T., Roth, C. B., Jaakola, V. P., Chien, E. Y., et al.
(2008). A specific cholesterol binding site is established by the 2.8 A structure of the
human beta2-adrenergic receptor. Structure, 16(6), 897–905.
56 Ankita Roy

Hatefi, Y. (1978). Preparation and properties of dihydroubiquinone: Cytochrome c oxido-

reductase (complex III). Methods in Enzymology, 53, 35–40.
Kawate, T., & Gouaux, E. (2006). Fluorescence-detection size-exclusion chromatography
for precrystallization screening of integral membrane proteins. Structure, 14(4), 673–681.
Laible, P. D., Scott, H. N., Henry, L., & Hanson, D. K. (2004). Towards higher-throughput
membrane protein production for structural genomics initiatives. Journal of Structural and
Functional Genomics, 5(1–2), 167–172.
Lopez, J. J., Shukla, A. K., Reinhart, C., Schwalbe, H., Michel, H., & Glaubitz, C. (2008).
The structure of the neuropeptide bradykinin bound to the human G-protein coupled
receptor bradykinin B2 as determined by solid-state NMR spectroscopy. Angewandte
Chemie (International Ed. in English), 47(9), 1668–1671.
Lundstrom, K. (2006). Structural genomics for membrane proteins. Cellular and Molecular Life
Science, 63(22), 2597–2607.
Lundstrom, K., Wagner, R., Reinhart, C., Desmyter, A., Cherouati, N., Magnin, T., et al.
(2006). Structural genomics on membrane proteins: Comparison of more than 100
GPCRs in 3 expression systems. Journal of Structural and Functional Genomics, 2, 77–91.
Magnani, F., Shibata, Y., Serrano-Vega, M. J., & Tate, C. G. (2008). Co-evolving stability
and conformational homogeneity of the human adenosine A2a receptor. Proceedings of the
National Academy of Sciences of the United States of America, 105(31), 10744–10749.
Morello, E., Bermúdez-Humarán, L. G., Llull, D., Solé, V., Miraglio, N., Langella, P., et al.
(2008). Lactococcus lactis, an efficient cell factory for recombinant protein production and
secretion. Journal of Molecular Microbiology and Biotechnology, 14(1–3), 48–58.
Palsdottir, H., & Hunte, C. (2004). Lipids in membrane protein structures. Biochimica et
Biophysica Acta, 1666(1–2), 2–18.
Popot, J. L., Berry, E. A., Charvolin, D., Creuzenet, C., Ebel, C., Engelman, D. M., et al.
(2003). Amphipols: Polymeric surfactants for membrane biology research. Cellular and
Molecular Life Sciences, 60(8), 1559–1574.
Rafferty, C. N., & Clayton, R. K. (1997). The orientations of reaction center transition
moments in the chromatophore membrane of Rhodopseudomonas sphareroides, bases on
new linear dichroism and photoselection measurements. Biochimica et Biophysica Acta,
546(2), 189–206.
Roy, A., Shukla, A. K., Haase, W., & Michel, H. (2008). Employing Rhodobacter sphaeroides
to functionally express and purify human G protein-coupled receptors. Biological Chem-
istry, 389(1), 69–78.
Sahdev, S., Khattar, S. K., & Saini, K. S. (2008). Production of active eukaryotic proteins
through bacterial expression systems: A review of the existing biotechnology strategies.
Molecular and Cellular Biochemistry, 307(1–2), 249–264.
Serrano-Vega, M. J., Magnani, F., Shibata, Y., & Tate, C. G. (2008). Conformational
thermostabilization of the beta1-adrenergic receptor in a detergent-resistant form.
Proceedings of the National Academy of Sciences of the United States of America, 105(3), 877–882.
Shukla, A. K., Reinhart, C., & Michel, H. (2006). Comparative analysis of the human angio-
tensin II type 1a receptor heterologously produced in insect cells and mammalian cells.
Biochemical and Biophysical Research Communications, 349(1), 6–14.
Surade, S., Klein, M., Stolt-Bergner, P. C., Muenke, C., Roy, A., & Michel, H. (2006).
Comparative analysis and “expression space” coverage of the production of prokaryotic
membrane proteins for structural genomics. Protein Science, 15(9), 2178–2189.
Thompson, A. A., Liu, J. J., Chun, E., Wacker, D., Wu, H., Cherezov, V., et al. (2011).
GPCR stabilization using the bicelle-like architecture of mixed sterol-detergent
micelles. Methods, 55, 310–317.