Fluorescence spectroscopy studies of HEK293 cells expressing DOR-G i1αfusion [604260]

Fluorescence spectroscopy studies of HEK293 cells expressing DOR-G i1αfusion
protein; the effect of cholesterol depletion
Jana Brejchováa,1, Jan Sýkorab,1, Kate řina Dlouháa, Lenka Roubalováa, Pavel Osta šova,
Miroslava Vo šahlíkováa, Martin Hofb, Petr Svobodaa,⁎
aInstitute of Physiology, Academy of Sciences of the Czech Republic, Víde ňská 1083, 142 20 Prague 4, Czech Republic
bJ. Heyrovsky Institute of Physical Chemistry of the ASCR, v. v. i., Dolej škova 2155/3, 182 23 Prague 8, Czech Republic
abstract article info
Article history:
Received 1 December 2010
Received in revised form 4 August 2011Accepted 8 August 2011Available online 16 August 2011
Keywords:Plasma membrane
Cholesterol depletion
Fluorescence spectroscopy studiesHydrophobic membrane interiorδ-opioid receptor (DOR)
G protein coupling±PTXBiophysical studies of fluorescence anisotropy of DPH and Laurdan generalized polarization were performed
in plasma membranes (PM) isolated from control and cholesterol-depleted HEK293 cells stably expressing
pertussis toxin (PTX)-insensitive DOR-G i1α(Cys351–Ile351) fusion protein. PM isolated from control, PTX-
untreated, cells were compared with PM isolated from PTX-treated cells. Results from both types of PMindicated that i) hydrophobic membrane interior was made more accessible to water molecules and more
chaotically organized in cholesterol-depleted samples, ii) cholesterol depletion resulted in an overall increase
in surface area of membrane, membrane fluidity, and mobility of its constituents.
Analysis of DOR-G
i1αcoupling in PTX-treated and PTX-untreated cells indicated that cholesterol depletion
did not alter the agonist binding site of DOR (B maxand K d) but the ability of DOR agonist DADLE to activate G
proteins was markedly impaired. In PTX-untreated membranes, EC 50for DADLE-stimulated [35S]GTP γS
binding was shifted by one order of magnitude to the right: from 4.3±1.2×10−9M to 2.2±1.3×10−8Mi n
control and cholesterol-depleted membrane samples, respectively. In PTX-treated membranes, EC 50was
shifted from 4.5±1.1×10−9M to 2.8±1.4×10−8M.
In summary, the perturbation of optimum PM organization by cholesterol depletion deteriorates functionalcoupling of DOR to covalently bound G
i1αas well as endogenously expressed PTX-sensitive G proteins of
Gi/Gofamily while receptor ligand binding site is unchanged. The biophysical state of hydrophobic plasma
(cell) membrane interior should be regarded as regulatory factor of DOR-signaling cascade.
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
There is a growing body of evidence indicating that cholesterol-
and sphingolipid-enriched membrane domains/rafts play a substan-
tial role in trans-membrane signaling through G protein-coupled
receptors [1–10]. These membrane structures, characterized by high
content of cholesterol, saturated phospholipids, glycolipids and
sphingomyelin, have been proposed as lipid platforms capable toharbor and con fine a number of different signaling molecules,
including trimeric G proteins and GPCR [11]. Caveolin-containing
rafts, caveolae , represent the main cellular constituent responsible for
transport of extracellular cholesterol into the cell and intracellular
membrane traf fic of this lipid molecule [12–14]. Membrane domains/
rafts as well as caveolae are degraded by cholesterol depletion induced
by cyclodextrins [15,16] .
We have previously reported that δ-opioid agonist DADLE exhibits
high ef ficacy when activating DOR-G i1α(Cys351–Ile351) fusion protein
in membrane domains/rafts when compared with the “bulk of plasma
membranes ”representing the majority of PM fragments [17].E x p o s u r e
of HEK293 cell membrane to cholesterol-depleting agent β-cyclodextrin
(β-CDX) was shown to attenuate the TRH-R induced signaling in intact
HEK293 cells detected as calcium response [18].T h ed o s e –response
curves of calcium response were shifted by three orders of magnitude to
the right; similar decrease in sensitivity of agonist response was
observed in isolated membranes by high-af finity [35S]GTP γSb i n d i n g
assay used for determination of trimeric G protein activity. Membrane
domains were degraded when isolated as detergent-resistant, low-
density membrane fragments from cholesterol-depleted cells [19].
Therefore, in this work we wanted to compare the effect of cholesterolBiochimica et Biophysica Acta 1808 (2011) 2819 –2829
Abbreviations: β-CDX, β-cyclodextrin; DADLE, D-Ala2-D-Leu5 enkephalin; DMEM,
Dulbecco's modi fied Eagle's medium; DOR, δ-opioid receptor; DOR-G i1αcells, HEK293
cells stably expressing DOR-G i1α(Cys351–Ile351) fusion protein; DPH, 1,6-diphenyl-
1,3,5-hexatriene; DRMs, detergent-resistant membrane domains; GPCR, G protein-coupled receptor; G proteins, heterotrimeric guanine nucleotide-binding regulatoryproteins; G
i/Go, G proteins inhibiting adenylyl cyclase activity in pertussis toxin-
sensitive manner; HEK, human embryonic kidney; PBS, phosphate-buffered saline; PM,plasma (cell) membrane; PMSF, phenylmethylsulfonyl fluoride; PNS, post-nuclear
fraction; PTX, pertussis toxin; TBS, Tris-buffered saline; TRH, thyrotropin-releasing
hormone; TRH-R, thyrotropin-releasing hormone receptor; TCSPC, Time Correlated
Single Photon Counting
⁎Corresponding author. Tel.: +420 241062478; fax: +420 24106 2488.
E-mail address: svobodap@biomed.cas.cz (P. Svoboda).
1Both authors contributed equally.
0005-2736/$ –see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbamem.2011.08.010
Contents lists available at SciVerse ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbamem

depletion on functional coupling between DOR and G i1αwithin fusion
protein (PTX-insensitive) with that between DOR and endogenously
expressed G protein of G i/Gofamily (PTX-sensitive).
In the first part of our present study, the effect of cholesterol
depletion induced by increasing concentrations of β-CDX on
physico-chemical properties of isolated plasma (cell) membranes
was determined. We wanted to analyze in details the perturbationof integrity of membrane structure caused by cholesterol depletion.
Biophysical state of the isolated P M interior was studied by steady-
state and time-resolved fluorescence anisotropy of hydrophobic
membrane probe DPH. The membrane interface was analyzed by
Laurdan generalized polarizati on. In the second part of our work,
the effect of cholesterol depletion on the ligand binding charac-
teristics and the coupling ef ficiency of DOR to the cognate G
proteins was compared in PM isolated from PTX-untreated and
PTX-treated cells. In the third part of our work, the purity of PM
preparation was analyzed by detailed marker analysis of fractions
collected from the whole Percoll
Rgradient. PM preparation used
for DOR analysis contained 81% of available receptor agonist
binding sites, i.e., of all receptors recovered in the gradient.
Therefore, this preparation was fully relevant for determination
of the in fluence of cholesterol depletion on DOR-G protein coupling
described in this work. The amount of membrane fragments of
Golgi origin was close to zero; co-localization of PM fragments with
minor portion of early endosomes re flected the known association
of Rab5 (used as a marker of these vesicles) with cytoplasmic face
of PM [20].
2. Methods
2.1. Chemicals
Complete protease inhibitor cocktail was purchased from Roche
Diagnostic, Mannheim, Germany (cat. no. 1697498). Fluorescent
probes 1,6-diphenyl-1,3,5-hexatriene (D-202) and Laurdan (D-250)
were from Invitrogen. All other chemicals were of higher purity
available.
2.2. Cell culture
HEK293 cells stably expressing DOR-G i1αfusion protein were
cultivated in DMEM supplemented with 8% newborn calf serum and
2 mM (0.292 g/l) L-glutamine in humidi fied 5% CO 2–95% air atmo-
sphere at 37 °C as described by [17,21] . Geneticin (800 μg/ml) was
included in the course of cell cultivation. The cells were grown to 60 –
80% con fluency before harvesting.
2.3. Treatment of HEK293 cells with pertussis toxin and β-cyclodextrin
Cells were incubated with (+PTX) or without ( −PTX) pertussis
toxin (25 ng/ml) for 24 h and, subsequently, they were either treated
(+β-CDX) or untreated ( −β-CDX) with 10 mM β-cyclodextrin in
serum-free DMEM for 60 min at 37 °C. Replenishment of cholesterol
was carried out by incubation of β-CDX-treated cells in serum-free
DMEM supplemented with 2.4 g/l of “water soluble cholesterol ”
(Sigma, C4951) for 2 h at 37 °C. The final concentration of cholesterol
in incubation medium was 0.25 mM. Harvesting of cells was
performed by centrifugation for 10 min at 1800 rpm. The cell
sediment was snap frozen in liquid nitrogen and stored at −80 °C
until use. Intracellular potassium (PBFI-AM, Invitrogen P-1267) and
sodium (SBFI-AM, Invitrogen S-1264) concentrations as well as
overall permeability/integrity of the cell membrane (trypan blue)
was unchanged when exposed 10 mM β-CDX for 2 h.2.4. Isolation of plasma (cell) membranes in PercollRgradient
The cell sediment was homogenized in 250 mM sucrose, 20 mM
Tris–HCl, 1 mM EDTA, pH 7.6 plus fresh 1 mM PMSF and complete
protease inhibitors cocktail (STE medium) in loosely- fitting, Te flon-
glass homogenizer for 7 min at 1800 rpm. The cell homogenate was
centrifuged for 7 min at 3500 rpm (1000 × g). In this way, cell debris
and nuclear fraction remaining in the sediment were separated from
post-nuclear supernatant (PNS). 3 ml of PNS was applied on top of
20 ml of 30% v/v PercollRin thick polycarbonate Beckman Ti70 tubes.
Centrifugation for 1 h at 27000 rpm (65,000 × g) resulted in separa-
tion of the two clearly visible layers. The upper layer represented the
plasma membrane fraction (PM); the lower layer was composed from
mitochondria and lysosomes. The upper layer was diluted 1:4 in
distilled water and centrifuged in Beckman Ti70 rotor for 2 h at
50,000 rpm (160,000 × g). The PM sediment was removed from the
compact, gel-like sediment of PercollR, re-homogenized by hand in
small volume of 50 mM Tris –HCl, 3 mM MgCl 2, 1 mM EDTA, pH 7.4
(TME medium), snap frozen in liquid nitrogen and stored at −80 °C
until use. To verify the purity of PM preparation, fractions 1 –22 (1 ml)
were collected from top to bottom of the centrifuge tube after
separation of the upper and lower bands in PercollRgradient and
frozen in liquid nitrogen. Subsequently, these fractions were sub-
jected to analysis of receptor content, its ability to activate the cognate
G proteins and distribution of plasma membrane, Golgi and
endosomal markers: Na, K-ATPase, G βsubunit protein, caveolin-1,
β-galactosidase, trans-Golgi membrane protein (t-GMP) and Rab5
(see below for more details).
2.5. Steady-state fluorescence anisotropy of DPH
PercollR-purified PM were labeled with DPH by the fast addition
(under mixing) of 1 mM DPH in freshly distilled acetone to the
membrane suspension (0.1 mg protein/ml; 1 μMfinal concentration);
after 30 min at 25 °C, which were allowed to ensure the optimum
incorporation of the probe into the membrane interior [22], the
anisotropy of DPH fluorescence was measured at Ex 365 nm/Em
425 nm wavelengths. Under these conditions, the fluorescence
intensity of the membrane-bound DPH was ≈500× higher than that
of the unbound, free probe in aqueous medium; light scatteringproblems could be therefore omitted. Steady-state fluorescence
anisotropy, r
DPH, was calculated according to the formula: r DPH=
(Ivv−Ivh)/(I vv+2I vh) according to Refs. [23,24] .
2.6. The time-resolved fluorescence and dynamic depolarization of DPH
Fluorescence lifetime and polarization experiments were per-
formed in a time correlated single photon counting (TCSPC)
spectro fluorometer IBH 5000 U equipped with a cooled Hamamatsu
R3809U-50 microchannel plate photomultiplier detector as described
before [25].
The anisotropy data were analyzed according to the “wobble in
cone ”model introduced by [26]. Wobbling diffusion constant D wand
S-order parameter were calculated according to Eqs. (1) and (2) .
S=r∞ðȚ
r0ðȚ/C18/C19
1=2;ð1Ț
Dw=σs
ϕ; ð2Ț
where r(0), and r(∞) stand for limiting and residual anisotropy, ϕis
the rotational correlation time and σsis the relaxation time which is a
function of the S-order parameter and has been determined according
to[26,27] .2820 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

2.7. Laurdan generalized polarization
Incorporation of fluorescent probe Laurdan into isolated HEK293
cell membranes was performed by addition of 0.4 mM stock solution
(in methanol) to membrane suspension followed by incubation for
30 min at 40 °C under mixing to 10 μMfinal concentration of this
probe. Recording of emission and excitation spectra was carried out at25 °C and 40 °C, respectively. The emission scans were measured with
the excitation wavelengths set to 340 nm and 410 nm, respectively;
the excitation spectra were determined at the emission wavelengths
set to 440 nm and 490 nm. The generalized polarization (GP) spectra
were determined from intensities of Laurdan fluorescence according
to Refs. [25,28] .
The excitation and emission GP spectrum slopes were obtained by
the linear fitting of corresponding GP spectra in the 340 –400 nm and
440 –490 nm range as described before [25,29,30] .
2.8. [
3H]DADLE (DOR agonist) binding; saturation binding study
DOR ( δ-opioid receptors) were characterized by saturation
binding study with agonist [3H]DADLE according to Refs. [17,21] .
Membranes were incubated with increasing concentrations of this
radioligand (0.15 –20 nM) in 0.1 ml of 75 mM Tris –HCl, pH 7.4,
12.5 mM MgCl 2, 1 mM EDTA for 60 min at 30 °C; 10 μg of membrane
protein was added per assay. The bound and free radioactivity was
separated by filtration through Whatman GF/B filters in Brandel cell
harvester. Filters were washed 3× with 3 ml of ice-cold 10× diluted
incubation buffer and placed in 4 ml of scintillation cocktail(CytoScint, ICN). Radioactivity remaining on filters was determined
after 16 h at laboratory temperature. The non-speci fic binding was
defined as that remaining in the presence of 10 μM non-radioactive
DADLE. Data were analyzed by GraphPad Prizm4 ;K
dand B maxvalues
were calculated according to the method of the least-squares by
fitting the data with rectangular hyperbola. Distribution of agonist
binding sites of DOR along the PercollRgradient was determined in
aliquots of fractions 1 –22 (1 ml each) collected from top to bottom of
the centrifuge tube by “one-point assay ”using 2.0 nM [3H]DADLE.
2.9. DADLE-stimulated, high-af finity [35S]GTP γS binding
Membranes were incubated with (total binding, B t) or without
(basal binding, B basal) increasing concentrations of DOR agonist
DADLE [0.1 –20 nM] in final volume of 100 μl of reaction mix
containing 20 mM HEPES, pH 7.4, 3 mM MgCl 2, 100 mM NaCl, 2 μM
GDP, 0.2 mM ascorbate and [35S]GTP γS (about 100 –200,000 dpm per
assay) for 30 min at 30 °C. The binding reaction was terminated by
dilution with 3 ml of ice-cold 20 mM HEPES, pH 7.4, 3 mM MgCl 2and
filtration through Whatman GF/C filters on Brandel cell harvester.
Radioactivity remaining on the filters was determined by liquid
scintillation using BioScint cocktail. Non-speci fic binding was deter-
mined in parallel assays containing 10 μM unlabelled GTP γS. Data
were analyzed by GraphPad Prizm4 and the EC 50and B maxvalues
calculated according to the method of the least-squares by fitting the
data with rectangular hyperbola.
When screening the PercollRdensity gradient pro files, [35S]GTP γS
binding assay was performed in the absence (basal) or presence of
constant concentration of 10−5M DADLE. Distribution of high-af finity
[35S]GTP γS binding±DADLE was determined in aliquots of fractions
1–22 collected from top to bottom of the centrifuge tube.
2.10. Na, K-ATPase
Plasma membrane density of Na, K-ATPase (E.C. 3.6.1.3) was
determined by [3H]ouabain binding assay according to Ref. [31].
Constant volume aliquots (50 μl) of gradient fractions were incubated
with 20 nM [3H]ouabain in total volume of 0.45 ml of 5 mM NaHPO 4,5 mM MgCl 2, 50 mM Tris –HCl, pH 7.6 for 90 min at 30 °C. The binding
reaction was terminated by dilution with 5 ml of ice-cold buffer and
filtration through Whatman GF/B filters. Filters were washed 3× with
5 ml of cold buffer, dried overnight at laboratory temperature and the
radioactivity determined by liquid scintillation. Non-speci fic binding
was determined in the presence of 1 μM unlabelled ouabain.
2.11. SDS-PAGE and immunoblotting
Distribution of plasma membrane, Golgi and endosomal markers
in PercollRgradient was analyzed by NuPAGE system (Invitrogen)
according to the manufacturer's protocol ( http://tools.invitrogen.
com/content/sfs/manuals/nupage_tech_man.pdf ) instructions. The
constant volume aliquots of gradient fractions were analyzed.
Molecular mass determinations were based on pre-stained
molecular mass markers (Sigma, SDS 7B). After SDS-PAGE, proteins
were transferred to nitrocellulose and blocked for 1 h at room
temperature in 5% (w/v) low-fat milk in TBS-Tween buffer [10 mM
Tris–HCl, pH 8.0, 150 mM NaCl, 0.1% (v/v) Tween 20]. The primary
antibodies were added in TBS-Tween containing 1% (w/v) low-fat
milk and incubated for at least 2 h.
The following antibodies were used: sc-28800 [Na, K-ATPase, α-
subunit], sc-378 [G βsubunit protein, T20], sc-21726 [trans-Golgi
membrane protein, t-GMP], sc-22291-R [ β-1,4-Gal-T5 (C-17)], sc-598
[Rab5B, A-20], all from Santa Cruz. Caveolin-1 was identi fied by
C13630 from Transduction Lab. The primary antibody was then
removed and the blot washed extensively (3×10 min) in TBS-Tween.
Secondary antibodies (donkey anti-rabbit IgG conjugated with horse-
radish peroxidase) were diluted in TBS-Tween containing 1% (w/v)
low-fat milk, applied for 1 h and after three 10 min washes the blots
were developed by ECL technique using Super Signal West Dura
(Pierce) as substrate. The developed blots were scanned with an
imaging densitometer ScanJett 5370C (HP) and quanti fied by Aida
Image Analyzer v. 3.28 (Ray test).
3. Results
3.1. Studies of hydrophobic membrane interior by steady-state
fluorescence anisotropy of DPH
Hydrophobic membrane probe 1,6-diphenyl-1,3,5-hexatriene
(DPH) was intensively studied in the past and it has been shown
that the anisotropy of its fluorescence is closely related to fluidity/
microviscosity of biological membranes: the increase in membrane
fluidity was re flected in decrease of r DPHand vice versa [23,24,32 –35].
In our experiments, the effect of β-CDX on r DPHwas used as a measure
of the effect of cholesterol depletion on the hydrophobic plasma
membrane interior.
In the first set of experiments, r DPHwas compared in plasma
membranes (PM) isolated from control and β-CDX-treated cells which
were not exposed to PTX. The values of r DPHwere not signi ficantly
different between the two types of membranes when measured at
laboratory temperature of 20 °C ( Fig. 1 A; NS), however, when
measured at 40 °C, the value of DPH anisotropy in PM isolated from
β-CDX-treated cells was signi ficantly lower (r DPH=0.114±0.007)
than in PM isolated from control cells (r DPH=0.151±0.005). The
effect of cholesterol depletion was fully reversible as DPH anisotropy
was not different in control and cholesterol-replenished PM
(rDPH=0.156±0.005; NS). PM isolated from cholesterol-enriched
cells exhibited higher values of DPH anisotropy (r DPH=0.184±0.014)
than those prepared from control cells. Thus, r DPHreflected readily the
change of “membrane fluidity ”which was increased in PM with lower
cholesterol level. Environmental temperature of living cells in tissue
culture medium was 37 °C. Therefore, measurement of r DPHat 40 °C
corresponded better to natural conditions of PM in intact cells than
measurement at 20 °C. Decrease of temperature of measurement from2821 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

45 to 20 °C was re flected in continuous increase of r DPHvalues with no
evidence of phase transitions. Arrhenius plots of data collected from
both PTX-untreated and PTX-treated cells were represented by
straight lines with correlation coef ficients close to 1 (r2=0.98 and
0.99, respectively) and no indication of breaks between 20 and 45 °C
(Table 1 ).
In the second set of experiments ( Fig. 1 B), PM isolated from PTX-
untreated and PTX-treated cells were compared. Both types of
membranes were incubated with increasing concentrations of β-
CDX (0, 2.5, 5 and 10 mM) for 60 min at 25 °C and labeled with DPH as
described in Methods . Data shown in Fig. 1 Bc o n firmed those
described in previous paragraph: r DPHwas decreased from 0.214±
0.001 to 0.190±0.002 (in PTX-untreated) or from 0.184±0.002 to
0.158±0.003 in PM isolated from PTX-treated cells. Thus, the results
obtained by analysis of direct effect of β-CDX on the membraneindicated clearly that cholesterol depletion caused signi ficant fluidi-
zation of PM. Membranes isolated from PTX-treated cells exhibited
significantly lower values of r DPH in the whole range of β-CDX
concentrations what may be explained by toxic effect of this inhibitor
on overall cell metabolism. The total protein content present in 106
PTX-treated cells represented about 70% of that detected in control,PTX-untreated cells.
3.2. Studies of hydrophobic plasma membrane interior by dynamic
depolarization of DPH fluorescence
More detailed view of the effect of cholesterol depletion on PM
organization was obtained by determination of the lifetime values of
DPHfluorescence, analysis of the time-resolved DPH anisotropy decay
and by application of the “wobble in cone ”model for interpretation of
the measured data [26,36,37] . These measurements were performed
in membranes isolated from PTX-untreated cells. As shown in Fig. 2Table 1
The effect of temperature on DPH anisotropy. PM isolated from control, PTX-untreatedcells, and PTX-treated cells were labeled with DPH and r
DPHmeasured at increasing
temperatures. Five minutes was allowed to achieve equilibrium at higher temperature.
Numbers represent the average r DPH±SEM of 3 independent experiments. Arrhenius
plots represented the straight lines with correlation coef ficients r2=0.98 and 0.99,
respectively. T, absolute temperature.
t (°C) r DPH ln (r DPH)1 03/T
−PTX
15 0.253±0.005 −1.373 3.470
20 0.231±0.005 −1.465 3.411
25 0.224±0.005 −1.497 3.354
30 0.202±0.005 −1.600 3.299
35 0.189±0.005 −1.669 3.245
40 0.168±0.004 −1.784 3.193
45 0.162±0.005 −1.822 3.143
+PTX
15 0.225±0.004 −1.492 3.470
20 0.221±0.004 −1.510 3.411
25 0.192±0.006 −1.649 3.354
30 0.183±0.006 −1.701 3.299
35 0.163±0.005 −1.816 3.245
40 0.151±0.004 −1.888 3.193
45 0.138±0.007 −1.981 3.14320°C 40°C0.000.050.100.150.200.25control
β-cdx
β-cdx+chol
cholNS
***
NSDPH anisotropy
0.0 2.5 5.0 7.5 10.00.150.160.170.180.190.200.210.22
PTX-treatedPTX-untreated
β-CDX (mM)DPH anisotropyA
BSteady-state DPH anisotropy
β-CDX exposure of intact cells
Direct effect of β-CDX on isolated PM
Fig. 1. Steady-state fluorescence anisotropy of DPH in isolated plasma membranes
A: The steady-state fluorescence anisotropy of DPH (r DPH) was compared in
plasma membranes isolated from control (open), cholesterol-depleted (full),
cholesterol-replenished (gray) and cholesterol-enriched (dashed) HEK293 cellsexpressing DOR-G
i1αfusion protein at 20 °C and 40 °C. Data represent average of
3 experiments±S.E.M. The difference between the different types of membraneswas compared by the Students t-test; *, **, ***, represent the signi ficant difference,
pb0.05, p b0.01, p b0.001, respectively. NS, not signi ficant. B: The direct effect of
β-CDX on r
DPHwas measured in PM isolated from PTX-untreated ( ○) and PTX-
treated (•) cells which were exposed to increasing concentrations of β-CDX as
described in Methods . Data represent average of 3 experiments±S.E.M.
0 24 68 1 00,0250,0300,035
0,550,600,650,70
β-CDX concentration (mM)(D)w (ns-1)
S-order parameter Time-resolved parameters of DPH fluorescence;
wobbling constant Dw and S-order parameter
Fig. 2. Time-resolved parameters of DPH fluorescence; wobbling constant Dwand S-
order parameter The decay of anisotropy of DPH fluorescence in ns range was
measured in PM isolated from PTX-untreated cells and analyzed as described inMethods .(•), D w, wobbling diffusion constant; ( ○), S-order parameter. The data
shown represent the average of three experiments±S.E.M.2822 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

and Table 2 , the average lifetime of DPH dye was decreased as the
concentration of β-CDX was increased; rotational correlation time ϕ
was unchanged. The values of both limiting ( r0) and residual ( r∞)
anisotropy were signi ficantly decreased. The S-order parameter
decreased continuously with the increasing β-cyclodextrin concen-
tration; this decrease proceeded in parallel with the increase of
wobbling diffusion constant Dw. Thus, the lower friction in the DPH
microenvironment and higher rate of the DPH rotation was noticed.
3.3. Studies of membrane interface by Laurdan generalized polarization
The polar membrane –water interface of HEK293 cell membrane
was studied in the next part of our work by Laurdan generalized
polarization [28]. PM isolated from PTX-untreated cells were used
again to avoid the toxic effects of this inhibitor on membrane
structure. As illustrated in Fig. 3 , addition of increasing concentrations
of cholesterol-depleting agent β-CDX caused the continuous decrease
offluorescence intensity at 440 nm ( blue shoulder of Laurdan
emission spectrum) and increase of intensity at 490 nm ( redshoulder
of Laurdan fluorescence). The value of generalized polarization
excGP370(Fig. 4 ) was decreased by increasing β-CDX concentration
from 0.29 to 0.09 (at 25 °C) or from 0.12 to –0.02 (at 40 °C).Further information was obtained by calculation of the slopes of
both the GP emission (GP emS) and excitation (GP excS) spectra [29,38] ,
which are depicted in Fig. 5 . This type of analysis indicated that the
cholesterol depletion did not cause clearly detectable phase transi-
tions, neither there was direct evidence for coexistence of the gel and
liquid crystalline phase within the PM vesicles used in this study. The
emission GP slope was raised monotonously with the increasing β-
CDX concentration, in parallel with decrease of excitation GP slope.
This type of symmetrical, reciprocal pro files of GP slopes, which were
detected at both 25 °C and 40 °C indicated that β-CDX-effect on the
cell membrane proceeded as an increase in hydration and mobility of
membrane constituents in the polar head-group region.Table 2
The in fluence of increasing concentrations of β-CDX on parameters of the time-resolved
DPHfluorescence. Parameters of the time-resolved DPH fluorescence were obtained by
fitting and further analysis of the time-resolved anisotropy decays recorded for DPH
incorporated into PM treated with increasing concentrations of β-CDX. PM were
isolated from PTX-untreated cells. τstands for the average fluorescence lifetime, r 0for
the limiting anisotropy at time zero, r ∞for the residual anisotropy, ϕfor the rotational
correlation time, S for the S-order parameter and Dwrepresents the wobbling diffusion
constant.
0 2.5 mM 5 mM 10 mM
τ 9.61 9.37 8.99 8.41
r0 0.295 0.282 0.280 0.279
r∞ 0.121 0.105 0.096 0.095
ϕ 4.4 4.7 4.4 4.3
S 0.68 0.64 0.61 0.58D
w 0.028 0.030 0.032 0.037
400 450 500 5500.00.20.40.60.81.0Relative Intensity (a.u.)
wavelength (nm)Changes of Laurdan fluorescence induced by
normalized steady-state emission spectra increasing concentrations of -CDX; area β
Fig. 3. Changes of Laurdan fluorescence induced by increasing concentrations of β-CDX;
area normalized steady-state emission spectra. Incorporation of Laurdan probe in PMisolated from control, PTX-untreated HEK293 cells was performed as described inMethods . Area normalized emission spectra were recorded at 25 °C. The arrows
indicate the change of Laurdan fluorescence upon β-CDX addition. The data represent
the average of 3 experiments±S.E.M.02468 1 00.00.10.20.3exc GP370
β-CDX concentration (mM)Dependence of Laurdan generalized polarization
excGP370 on the -CDX concentration β
Fig. 4. Dependence of Laurdan generalized polarization on the β-CDX concentration.
The value of excGP370was calculated as described in Methods . The data were recorded
at 25 °C (full) and 40 °C (open symbols) and represent the average of 3 experiments±S.E.M.
02468 1 0-2-101234
-2-101234GPexcS (103 nm-1), 25oC, 40oC
GPemS (103 nm-1), 25oC, 40oC
β-CDX concentration (mM) The slopes of the excitation
and emission Laurdan GP
Fig. 5. The slopes of the excitation and emission Laurdan GP spectra. The effect of
increasing concentrations of β-CDX on slopes of excitation (GP excS; open symbols) and
emission (GP emS; full symbols) Laurdan GP spectra were measured at 25 °C (circles) or
40 °C (triangles) in PM isolated from PTX-untreated cells. The excitation and emissionGP spectrum slopes were obtained by the linear fitting of corresponding GP spectra in
the range of 340 –400 nm and 440 –490 nm, respectively. The data represent the
average of 3 experiments±SEM.2823 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

3.4. Receptor agonist binding assays
The effect of cholesterol depletion on the ligand binding
parameters of DOR-G i1αfusion protein was assessed by saturation
binding study with DOR agonist [3H]DADLE. Speci fic binding of this
radioligand was determined in membranes isolated from both PTX-
untreated± β-CDX and PTX-treated± β-CDX HEK293 cells stably
expressing this protein (compare upper and lower panel in Fig. 6 ).
As described in Methods , PTX-treated cells were exposed to pertussis
toxin prior to membrane isolation in the presence or absence of β-
CDX in order to prevent activation of PTX-sensitive, endogenously
expressed G proteins of G i/Gofamily.
Cholesterol depletion did not change the maximum number of
binding sites (B max) nor the af finity (K d)o f[3H]DADLE binding
reaction in membranes isolated from PTX-untreated ( Fig. 6 A) as well
as PTX-treated cells ( Fig. 6 B). The difference between β-CDX-
unexposed ( ○) and β-CDX-exposed ( ●) membrane samples isolated
from both types of cells was not signi ficant. Thus, depletion ofcholesterol from the HEK293 cell membrane did not interfere with
agonist binding site of DOR-G i1α(Cys351–Ile351) fusion protein.
3.5. DOR-G i1αcoupling
Determination of intrinsic ef ficiency of DOR was performed by
analysis of high-af finity [35S]GTP γS binding in membranes isolated
from both PTX-untreated and PTX-treated cells ( Fig. 7 ). In the latter
type of membranes, the set of endogenously expressed G proteins of
Gi/Gofamily was blocked by incubation of intact cells with PTX. In
PTX-untreated cells ( Fig. 7 A), cholesterol depletion lowered the basal
level of [35S]GTP γS binding (from 0.72 to 0.48 pmol mg−1), but the
maximum DADLE-stimulated binding expressed as the net-increment
of agonist stimulation ( Δ) was not signi ficantly different in β-CDX-
treated and untreated membrane samples.
Analysis of the dose-response curves in wide range of agonist
concentrations (2×10−9–1×10−4M) indicated that EC 50value of
DADLE response was shifted by one order of magnitude to the right
[from 4.3±1.2×10−9M to 2.2±1.3×10−8M in control and[3H]DADLE
Saturation binding curves
0 2 4 6 8 10 12 14 160.00.51.01.52.0
Kd Bmax
(nM) (pmol.mg-1)
Control 1.2 0.2 1.8 0.1
β-CDX 1.3 0.2 1.9 0.1
[3H]DADLE (nM)
[3H]DADLE (nM)pmol.mg-1pmol.mg-1
0 2 4 6 8 10 12 14 160.00.51.01.52.0
Kd Bmax
(nM) (pmol.mg-1)
PTX 1.0 0.2 1.5 0.1
PTX+ β-CDX 1.2 0.2 1.5 0.1Untreated β-CDX
PTX β-CDXA
B
Fig. 6. Agonist binding assay of DOR. DOR agonist [3H]DADLE was used for characterization
of ligand binding site of DOR-G i1αmolecule in PM isolated from PTX-untreated (A) and
PTX-treated (B) cells as described in Methods . The data shown represent the averaged
saturation binding curves of 3 experiments; K dand B maxvalues±S.E.M were calculated by
GraphPad Prizm4 .[35S]GTP γS binding
Dose-response curves
00.00.20.40.60.81.01.21.41.61.8
-10 -9 -8 -7 -6 -5 -4 -3Control
EC50=4.3 ±1.2 x10-9M
β-CDX
EC50=2.2 ±1.3 x10-8M
log [DADLE] (M)
00.00.20.40.60.81.01.21.41.61.8
-10 -9 -8 -7 -6 -5 -4 -3PTX
EC50=4.5 ±1.1 x10-9M
PTX+β-CDX
EC50=2.8 ±1.4 x10-8M
log [DADLE] (M)A
BUntreated β-CDX
PTX β-CDXpmol.mg-1pmol.mg-1
Fig. 7. High-af finity [35S]GTP γS binding. Stimulation of trimeric G proteins by DOR
agonist DADLE was measured by high-af finity binding of nonhydrolysable analog of
GTP, [35S]GTP γS. (A), membranes isolated from PTX-untreated cells; (B), membranes
isolated from PTX-treated cells. The data represent the averaged dose –response curves
of 3 experiments; EC 50and B maxvalues±S.E.M of were calculated by GraphPad Prizm4 .2824 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

[3H]DADLE
123456789101112131415161718192021220.00.10.20.30.40.50.60.70.80.9
fractionpmol.ml-1[35S]GTP γS binding
123456789101112131415161718192021220.000.040.080.120.160.200.240.280.320.36
Basal
DADLE 10-4M
fraction[3H]Ouabain
123456789101112131415161718192021220.01.02.03.04.05.06.07.08.0
fraction
123456789101112131415161718192021220.000.050.100.150.200.250.300.350.400.45
fractionpmol.mg-1
pmol.ml-1pmol.mg-1
pmol.ml-1pmol.mg-1
123456789101112131415161718192021220.000.020.040.060.080.100.120.140.160.18 Basal
DADLE 10-4 M
fraction123456789101112131415161718192021220.00.51.01.52.02.53.03.54.0
fractionACE
D B F
Fig. 8. Density gradient pro files of plasma membrane markers; radioligand binding studies. Fractions 1 –22 (1 ml each) were collected from top to bottom of the centrifuge tube and used in radioligand binding assays. (A, B) Distribution of
receptor binding sites was determined by DOR agonist [3H]DADLE. (C,D) High-af finity [35S]GTP γS binding was used as an estimate of functional ef ficiency of DOR when activating G proteins. (E, F) [3H]ouabain binding was used to determine
the distribution of plasma membrane marker Na, K-ATPase. Data represent the typical gradient pro files.2825 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

cholesterol-depleted membrane samples, respectively]. This indicat-
ed marked decrease in af finity (potency) of DADLE response. This
decrease was also detected in membranes isolated from PTX-treated
cells ( Fig. 7 B) as EC 50value was increased by order of magnitude
[from 4.5±1.1×10−9M to 2.8±1.4×10−8M]. In this type of cells,
the basal level of binding was unchanged by β-CDX treatment.
Results of [35S]GTP γS binding assays thus indicated that the
affinity of DADLE response was severely deteriorated in the case of
covalently bound G i1αprotein as well as endogenously expressed
Gp r o t e i n so fG i/Gofamily. Thus, covalent attachment of G i1αto
DOR did not protect G i1αagainst membrane damage caused by
cholesterol depletion; both types of G proteins were affected in
negative way by β-CDX-induced change of PM organization.
When combined with receptor analysis, it may be concluded that
the deleterious effect of cholesterol depletion on DOR-induced
signaling cascade described in Figs. 6 and 7 was essentially “non-
specific”in the terms of classical “receptor pharmacology ”based on
specificprotein-protein interactions between a given type or subtype
of receptor molecule and a given type of G protein αsubunits.
3.6. Quantitative recovery and veri fication of purity of PM preparation
Radioligand binding studies with DOR agonist [3H]DADLE were
also used for the determination of quantitative recovery and
verification of purity of PM preparation resolved in PercollRgradient(upper layer; compare with Methods ). Data shown in Fig. 8 (left
panels) indicated the maximum [3H]DADLE binding in fraction 9. PM
fractions 6 –12 representing the upper layer in PercollRgradient
contained by far the highest binding level of this radioligand. The 81%
of total receptor sites recovered in the whole gradient (1 –22) were
present in PM peak. Furthermore, the total number of [3H]DADLE
binding sites in PM fractions was 10× higher than in “mitochondrial ”
band which was represented by fractions 15 –20. DOR agonist binding
to“mitochondria ”was small and represented 8% of total binding
recovered in the whole gradient. The same result was obtained by
functional analysis of receptor ef ficiency by determination of DADLE-
stimulated [35S]GTP γS binding along the density gradient ( Fig. 8 ,
middle panels). The highest agonist-stimulated [35S]GTP γS binding
was detected in PM fractions 6 –12 (maximum in 9) and the sum of
this binding was 10× higher than in “mitochondria ”(frs. 15 –20).
Data obtained by [3H]DADLE and [35S]GTP γS binding assays were
verified by analysis of prototypical plasma membrane marker, Na, K-
ATPase ( Fig. 8 , right panels). Density gradient pro files of binding of
selective inhibitor of Na, K-ATPase, [3H]ouabain, were identical with
those obtained by receptor binding assays. The highest binding level
of [3H]ouabain was detected again in fractions 8 –9 and 80% of sites
present in the whole gradient (1 –22) were recovered in PM (frs. 6 –
12). Please notice the identity of receptor (80%) and Na, K-ATPase
recovery (81%). Furthermore, the total [3H]ouabain binding in PM
fractions was 8x higher than in “mitochondria ”(frs. 15 –20).
Na,K-ATPase
123456789101112131415160510152025303540
fractionarbitrary units
12345678910111213141516012345678910Gβ
fractionarbitrary unitscaveolin -1
1234567891011121314151605101520253035404550
fractionarbitrary units
β-1,4-Gal-T5
123456789101112131415160.02.55.07.510.012.515.017.5
fractionarbitrary unitst-GMP
123456789101112131415160510152025303540
fractionarbitrary unitsRab 5B
1234567891011121314151605101520253035
fractionarbitrary unitsAB C
EF D
Fig. 9. Density gradient pro files of plasma membrane, Golgi and endosomal markers; immunoblot analysis. Distribution of Na, K-ATPase (A), G β-subunit protein (B), caveolin-1
(C),β-galactosidase [ β-1,4-Gal-T5 (C-17)-R] (D), trans-Golgi membrane protein, t-GMP (E), and early endosomal marker rab 5 (F) was determined by quantitative
immunoblotting with speci fic antibodies as described in Methods . Data represent the typical gradient pro files.2826 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

Identical distribution of Na, K-ATPase molecules along the PercollR
gradient was obtained by immunoblot analysis with antibodies
oriented against the α-subunit of this enzyme ( Fig. 9 A). The highest
immunoblot signal was again noticed in fractions 8 and 9. Co-
localization of Na, K-ATPase with other plasma membrane proteins
and distribution of membrane fragments of Golgi and endosomal origin
in density gradient was analyzed in the last part of our work byimmunoblotting with speci fic antibodies. Distribution of G βsubunit
protein and caveolin-1 was very similar with that of Na, K-ATPase:
maximum immunoblot signals were detected in fraction 9 ( Fig. 9 Ba n d
C). Contrarily, Golgi markers β-galactosidase and t-GMP were localized
in the top most area of Percoll
Rgradient in fractions 1 –5(Fig. 9 Da n dE ) .
The overlap with PM was very small as the content of both β-1, 4-Gal-
T5 (C-17)-R and t-GMP decreased sharply towards higher densities.
The amount of β-galactosidase and t-GMP in fraction 9 was
undetectable or close to zero and no detectable signal of both Golgi
markers was present in fractions 10 –16. Thus, the density gradient
distribution of Golgi membranes was completely different from that of
Na, K-ATPase, G βsubunit and caveolin-1 which were clearly co-
localized in PM (compare Figs. 9 Da n dEw i t h 9A, B and C).
The upper fractions of density gradient were also enriched in
endosomal marker Rab 5 ( Fig. 9 F); however, when proceeding from
the top to bottom of centrifuge tube, the amount of this protein was
clearly decreased. The overlap of Rab5-containing membrane frag-
ments with PM may be interpreted as natural consequence of an
association of a portion of early/recycling endosomes with cytoplas-
mic face of plasma membrane [20,39] .
4. Discussion
Cholesterol is an essential constitutive component of eucaryotic
plasma (cell) membrane and plays the crucial role in membrane
organization, dynamics and function. It was also considered as the
major building component of membrane domains/ rafts and caveolae ,
a specialized PM compartments characterized by resistance todetergent-solubilisation at 0 °C [1–10,12 –15]. The lipid composition
of these structures is different from that of bulk of plasma membrane
which is composed mainly from the bilayer forming lipids phospha-
tidylcholine, phosphatidylethanolamine and phosphatidylserine. The
phospholipids are present in membrane domains as well, however,
the spectrum of domain lipids is more complex and is represented,
besides cholesterol and phospholipids, by large portion of glycolipids
and sphingolipids [1–10].
Depletion of cholesterol leads to disruption of membrane domains
[14–16], attenuation of trans-membrane signaling induced by
activated GPCR [15,18,19,40] and, as shown in measurement of
steady-state fluorescence anisotropy of DPH ( Fig. 1 ), it also results in
an increase of “overall ”membrane fluidity. However, the magnitude
ofβ-CDX-induced change of r
DPHin DOR-G i1αcells was much smaller
than that induced by low concentrations of non-ionic detergents Brij-
58 and Triton X-100 (data not shown). More detailed analysis of the
effect of cholesterol depletion on PM organization, revealed by
recording the time-resolved DPH anisotropy decays ( Fig. 2 and
Table 2 ) indicated that rotational correlation time ϕwas unchanged
by drop of cholesterol level, however, both limiting ( r0) and residual
(r∞) anisotropy were decreased. The S-order parameter characterizing
the degree of the static order in the region of aliphatic fatty acid chains
was decreased as β-CDX concentration was increased. The decrease of
static order proceeded in parallel with the increase of wobbling
diffusion constant D w, a parameter characterizing the dynamics of
DPH motion within the membrane. Thus, cholesterol depletion was
reflected in lower friction in the DPH microenvironment and higher
rate of the DPH rotation ( Fig. 2 ).
As the decay of DPH fluorescence is polarity dependent [36,37] , i.e.,
when increasing polarity of surrounding medium the lifetime of
membrane-bound DPH gets shorter, we could conclude that hydrophobicmembrane interior became more accessible to surrounding polar phase
(water). Moreover, as the difference in the lifetime values between β-
CDX-untreated and 10 mM β-CDX-treated membranes ( τfl=9.6 and
8.4 ns, respectively; Table 2 ) was relatively small, it may be assumed that
no dramatic change of membrane structure like micelle formation or
phase transition occurred. This co nclusion was supported by analysis of
temperature-dependence of r DPHindicating no breaks in Arrhenius plots
between 20 and 45 °C ( Table 1 ).
Lipid organization in the polar head-group region of membrane
bilayer was studied by generalized polarization of Laurdan fluores-
cence [25,28,38] . The solvato-chromic properties of this dye were
found to lead to red-shift of its emission spectra in hydrated liquid
crystalline phase compared to the rigid phase. The values of the
generalized polarization (GP) calculated from the fluorescence
intensities at either the excitation or emission wavelengths yielded
information about the phase state of the lipid microenvironment of
this probe. In arti ficial membranes composed from lipids of de fined
composition, GP reached the value of approximately 0.6 and −0.2 in
the gel and liquid crystalline phase, respectively. In PM isolated from
DOR-G i1αcells, the magnitude of change of Laurdan excGP370values
induced by β-CDX [from 0.29 to 0.09 (at 25 °C) or from 0.12 to −0.02
(at 40 °C); Fig. 4 ] was relatively small. Furthermore, the slopes of both
GP emission (GP emS) and excitation (GP excS) spectra were altered in
parallel suggesting that the cholesterol depletion did not cause a
clearly de fined phase transitions nor there was evidence for
coexistence of the gel and liquid crystalline phases within the PM
vesicles used in this study.
When interpreting the data obtained by steady-state and time-
resolved DPH anisotropy ( Figs. 1 and 2 ) and Laurdan GP ( Figs. 3 –5)i n
PM isolated from DOR-G i1αcells and considering changes in polarity
of membrane environment together with alternations in mobility,
lateral packing, water content (hydration) and various structural
states of membrane lipid molecules (observed in arti ficial membranes
of defined composition) such as gel, liquid crystalline, liquid ordered
(Lo) or liquid disordered (L d) states and phase transitions among
them, we can conclude that data presented in this work do not bring
direct evidence for phase transitions but cannot exclude their
existence in PM used in this study. The reason why not, is based on
complex chemical composition of lipid molecules and presence of
integral and peripheral proteins in natural PM. Furthermore, PM donot represent homogeneous membrane, but are laterally organized in
different compartments denominated as membrane domains and
caveolae containing the high amount of cholesterol [2,10,41 –43].
Cholesterol depletion by β-CDX degrades these structures and causes
redistribution of receptors within plasma membrane with well
documented functional consequences depending on the cell type
[44–49]. The final outcome of the effect of cholesterol depletion (of β-
CDX treatment) is therefore dif ficult to predict as it depends on the
cell type and represents inhibition as well as activation of GPCR-
induced signaling cascades. More speci fically, cholesterol depletion by
methyl- β-CDX was reported to attenuate DOR-stimulated [
35S]GTP γS
binding in neuronal cells but enhance it in non-neuronal cells [50].
Studies of DOR in parallel samples as those used in physico-
chemical analysis of PM structure and dynamics were surprisingly
clear-cut ( Figs. 6 and 7 ). Cholesterol depletion did not exert any
detectable change of agonist binding to DOR, but coupling ef ficiency
was severely attenuated regardless whether oriented towards
covalently bound G i1αor endogenously expressed PTX-sensitive G
proteins of G i/Gofamily. This is rather surprising finding as covalent
attachment of G protein to the receptor anticipates more firm
coupling which should not be easily diminished by minor alternation
of membrane structure documented in Figs. 1 –5. Thus, our data
indicate a signi ficant effect of “non-speci fic”perturbation of plasma
membrane structure (and dynamics) by cholesterol depletion on
coupling between DOR and the cognate PTX-sensitive G proteins.
Receptor ligand binding site is not affected. This finding appears to2827 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

correspond to/fall in line with previous literature data indicating no
speci fic cholesterol “recognition site ”in DOR protein molecule [51].
Relevance of our PM preparation for studies of DOR function and
biophysical characterization of PM structure and dynamics was
shown by detailed marker analysis of all fractions (1 –22) collected
from PercollRgradient. Quantitative radioligand binding analysis
indicated that the dominant portion of DOR and Na, K-ATPasemolecules (80%) was co-localized in fractions 6 –12 ( Fig. 8 ). Very
similar data were obtained by immunoblot analysis of distribution of
α-subunit of Na K-ATPase, G βsubunit protein and caveolin-1 ( Fig. 9 A,
B and C). Therefore, these fractions contained the representative
preparation of PM. Contamination by membrane fragments of Golgi
origin was close to zero ( Fig. 9 D and E). It is reasonable to assume that
co-localization of small portion of early endosomal marker Rab 5 with
PM ( Fig. 9 F) represents vesicles associated with inner face of PM [20].
5. Conclusions
Biophysical studies of puri fied plasma membranes isolated from
cholesterol-depleted HEK293 cells stably expressing DOR-G
i1α
(Cys351–Ile351) fusion protein indicated a signi ficant change of both
structural and dynamic parameters of fluorescence of membrane
probes DPH and Laurdan. Results obtained may be interpreted as
evidence for the existence of two different membrane environments:
i) hydrophobic membrane interior represented by aliphatic chains of
fatty acids which was made more accessible to water molecules and
more chaotically organized (by cholesterol depletion), ii) surface area
of membrane represented by polar head-groups of lipid molecules
which was enlarged. Furthermore, cholesterol depletion was re flected
in an increase of overall PM fluidity and increase in mobility of
hydrophobic membrane constituents.
Analysis of receptor ligand binding and coupling of DOR to G
proteins indicated that cholesterol depletion did not affect the
receptor agonist binding site but severely deteriorated the ability of
DOR to transmit the signal further down-stream, i.e., to activate the
cognate G proteins regardless whether covalently bound within
fusion protein or endogenously present in HEK293 cells.
Our data indicate a signi ficant effect of “non-speci fic”alternation
of PM organization by cholesterol depletion on functional coupling
between DOR and the cognate PTX-sensitive G proteins. Receptor
ligand binding site was not affected.
Acknowledgements
This work was supported by projects LC 06063 and LC 554 of the
Ministry of Education of the Czech Republic, Grant Agency of the
Czech Republic (305/08/H037 and P208/10/1090) and by the
Academy of Sciences of the Czech Republic (AV0Z50110509).
References
[1] S.N. Ahmed, D.A. Brown, E. London, On the origin of sphingolipid/cholesterol-rich
detergent-insoluble cell membranes: physiological concentrations of cholesterol
and sphingolipid induce formation of a detergent-insoluble, liquid-ordered lipid
phase in model membranes, Biochemistry 36 (1997) 10944 –10953.
[2] R.G. Anderson, The caveolae membrane system, Annu. Rev. Biochem. 67 (1998)
199 –225.
[3] D.A. Brown, E. London, Structure of detergent-resistant membrane domains: does
phase separation occur in biological membranes? Biochem. Biophys. Res.
Commun. 240 (1997) 1 –7.
[4] D.A. Brown, E. London, Structure and function of sphingolipid- and cholesterol-
rich membrane rafts, J. Biol. Chem. 275 (2000) 17221 –17224.
[5] R.E. Brown, Sphingolipid organization in biomembranes: what physical studies of
model membranes reveal, J. Cell Sci. 111 (1998) 1 –9.
[6] T. Harder, K. Simons, Caveolae, DIGs, and the dynamics of sphingolipid-cholesterol
microdomains, Curr. Opin. Cell Biol. 9 (1997) 534 –542.
[7] B. Chini, M. Parenti, G-protein-coupled receptors, cholesterol and palmitoylation:
facts about fats, J. Mol. Endocrinol. 42 (2009) 371 –379.[8] R.S. Ostrom, P.A. Insel, The evolving role of lipid rafts and caveolae in G protein-
coupled receptor signaling: implications for molecular pharmacology, Br. J.
Pharmacol. 143 (2004) 235 –245.
[9] L.J. Pike, Lipid rafts: heterogeneity on the high seas, Biochem. J. 378 (2004)
281 –292.
[10] K. Simons, E. Ikonen, Functional rafts in cell membranes, Nature 387 (1997)
569 –572.
[11] P. Oh, J.E. Schnitzer, Segregation of heterotrimeric G proteins in cell surface
microdomains. G(q) binds caveolin to concentrate in caveolae, whereas G(i) and
G(s) target lipid rafts by default, Mol. Biol. Cell 12 (2001) 685 –698.
[12] F.R. Max field, G. van Meer, Cholesterol, the central lipid of mammalian cells, Curr.
Opin. Cell Biol. 22 (2010) 422 –429.
[13] C.J. Fielding, A. Bist, P.E. Fielding, Caveolin mRNA levels are up-regulated by free
cholesterol and down-regulated by oxysterols in fibroblast monolayers, Proc. Natl.
Acad. Sci. U.S.A. 94 (1997) 3753 –3758.
[14] K. Dreja, M. Voldstedlund, J. Vinten, J. Tranum-Jensen, P. Hellstrand, K. Swärd,
Cholesterol depletion disrupts caveolae and differentially impairs agonist-
induced arterial contraction, Arteri oscler. Thromb. Vasc. Biol. 22 (2002)
1267 –1272.
[15] C. Prendergast, J. Quayle, T. Burdyga, S. Wray, Cholesterol depletion alters
coronary artery myocyte Ca(2+) signaling in a stimulus-speci fic manner, Cell
Calcium 47 (2010) 84 –91.
[16] R. Zidovetzki, I. Levitan, Use of cyclodextrins to manipulate plasma membrane
cholesterol content: evidence, misconceptions and control strategies, Biochim.
Biophys. Acta 1768 (2007) 1311 –1324.
[17] L. Bourova, A. Kostrnova, L. Hejnova, Z. Moravcova, H.E. Moon, J. Novotny, G.
Milligan, P. Svoboda, delta-Opioid receptors exhibit high ef ficiency when
activating trimeric G proteins in membrane domains, J. Neurochem. 85 (2003)
34–49.
[18] P. Ostasov, J. Krusek, D. Durchankova, P. Svoboda, J. Novotny, Ca(2+) responses
to thyrotropin-releasing hormone and angiotensin II: the role of plasma
membrane integrity and effect of G(11)alpha protein overexpression on
homologous and heterologous desensiti zation, Cell Biochem. Funct. 26 (2008)
264 –274.
[19] P. Ostasov, L. Bourova, L. Hejnova, J. Novotny, P. Svoboda, Disruption of the plasma
membrane integrity by cholesterol depletion impairs effectiveness of TRH
receptor-mediated signal transduction via G(q)/G(11)alpha proteins, J. Recept.Signal Transduct. Res. 27 (2007) 335 –
352.
[20] J.P. Gorvel, P. Chavrier, M. Zerial, J. Gruenberg, rab5 controls early endosome
fusion in vitro, Cell 64 (1991) 915 –925.
[21] H.E. Moon, D.S. Bahia, A. Cavalli, M. Hoffmann, G. Milligan, Control of the
efficiency of agonist-induced information transfer and stability of the ternary
complex containing the delta opioid receptor and the alpha subunit of G(i1) by
mutation of a receptor/G protein contact interface, Neuropharmacology 41
(2001) 321 –330.
[22] J. Svobodova, P. Svoboda, Cytoplasmic membrane fluidity measurements on intact
living cells of Bacillus subtilis by fluorescence anisotropy of 1,6-diphenyl-1,3,5-
hexatriene, Folia Microbiol. 33 (1988) 1 –9.
[23] M. Shinitzky, Y. Barenholz, Dynamics of the hydrocarbon layer in liposomes of
lecithin and sphingomyelin containing dicetylphosphate, J. Biol. Chem. 249
(1974) 2652 –2657.
[24] M. Shinitzky, Y. Barenholz, Fluidity parameters of lipid regions determined by
fluorescence polarization, Biochim. Biophys. Acta 515 (1978) 367 –394.
[25] J. Sykora, L. Bourova, M. Hof, P. Svoboda, The effect of detergents on trimeric G-
protein activity in isolated plasma membranes from rat brain cortex: correlation
with studies of DPH and Laurdan fluorescence, Biochim. Biophys. Acta Biomembr.
1788 (2009) 324 –332.
[26] S. Kawato, K. Kinosita, A. Ikegami, Dynamic structure of lipid bilayers studied by
nanosecond fluorescence techniques, Biochemistry 16 (1977) 2319 –2324.
[27] K. Kinosita, A. Ikegami, S. Kawato, On the wobbling-in-cone analysis of
fluorescence anisotropy decay, Biophys. J. 37 (1982) 461 –464.
[28] T. Parasassi, G. De Stasio, G. Ravagnan, R.M. Rusch, E. Gratton, Quantitation of lipid
phases in phospholipid vesicles by the generalized polarization of Laurdan
fluorescence, Biophys. J. 60 (1991) 179 –189.
[29] A.A. Vallejo, J.B. Velazquez, M.S. Fernandez, Lateral organization of mixed,
two-phosphatidylcholine liposomes as investigated by GPS, the slope ofLaurdan generalized polarization spectra, Arch. Biochem. Biophys. 466 (2007)
145 –154.
[30] J.B. Velazquez, M.S. Fernandez, GPS, the slope of Laurdan generalized polarization
spectra, in the study of phospholipid lateral organization and Escherichia coli lipid
phases, Arch. Biochem. Biophys. 455 (2006) 163 –174.
[31] P. Svoboda, E. Amler, J. Teisinger, Different sensitivity of ATP+Mg+Na (I) and
Pi+Mg (II) dependent types of ouabain binding to phospholipase A2, J. Membr.
Biol. 104 (1988) 211 –221.
[32] D. Chapman, Biomembrane fluidity: the concept and its development, Membrane
Fluidity in Biology, vol. 2, Academic Press, 1983, pp. 5 –42.
[33] M. Shinitzky, M. Shinitzky, Physiology of Membrane Fluidity, Vol. 1, CRC Press,
Boca Raton, 1984.
[34] M. Shinitzky, M. Inbar, Difference in microviscosity induced by different
cholesterol levels in the surface membrane lipid layer of normal lymphocytes
and malignant lymphoma cells, J. Mol. Biol. 85 (1974) 603 –615.
[35] M. Shinitzky, M. Inbar, Microviscosity parameters and protein mobility in
biological membranes, Biochim. Biophys. Acta 433 (1976) 133 –149.
[36] R. Fiorini, E. Gratton, G. Curatola, Effect of cholesterol on membrane microheter-
ogeneity: a study using 1,6-diphenyl-1,3,5-hexatriene fluorescence lifetime
distributions, Biochim. Biophys. Acta 1006 (1989) 198 –202.2828 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829

[37] R. Fiorini, M. Valentino, S. Wang, M. Glaser, E. Gratton, Fluorescence lifetime
distributions of 1,6-diphenyl-1,3,5-hexatriene in phospholipid vesicles, Biochem-
istry 26 (1987) 3864 –3870.
[38] T. Parasassi, M. Di Stefano, M. Loiero, G. Ravagnan, E. Gratton, In fluence of
cholesterol on phospholipid bilayers phase domains as detected by Laurdan
fluorescence, Biophys. J. 66 (1994) 120 –132.
[39] A. Radhakrishnan, J.L. Goldstein, J.G. McDonald, M.S. Brown, Switch-like control of
SREBP-2 transport triggered by small changes in ER cholesterol: a delicatebalance, Cell Metab. 8 (2008) 512 –521.
[40] V. Rudajev, J. Novotny, L. Hejnova, G. Milligan, P. Svoboda, Dominant portion of
thyrotropin-releasing hormone rece ptor is excluded from lipid domains.
Detergent-resistant and detergent-sensitive pools of TRH receptor and G(q)
alpha/G(11)alpha protein, J. Biochem. 138 (2005) 111 –125.
[41] R.R. Neubig, Membrane organization in G-protein mechanisms, FASEB J. 8 (1994)
939 –946.
[42] J.E. Schnitzer, D.P. McIntosh, A.M. Dvorak, J. Liu, P. Oh, Separation of caveolae from
associated microdomains of GPI-anchored proteins, Science 269 (1995)
1435 –1439.
[43] J.A. Allen, R.A. Halverson-Tamboli, M.M. Rasenick, Lipid raft microdomains and
neutrotransmitter signalling, Nat. Rev. Neurosci. 8 (2007) 128 –140.
[44] W.F. de Weerd, L.M. Leeb-Lundberg, Bradykinin sequesters B2 bradykinin
receptors and the receptor-coupled Galpha subunits Galphaq and Galphai in
caveolae in DDT1 MF-2 smooth muscle cells, J. Biol. Chem. 272 (1997)
17858 –17866.[45] O. Feron, T.W. Smith, T. Michel, R.A. Kelly, Dynamic targeting of the agonist-
stimulated m2 muscarinic acetylcholine receptor to caveolae in cardiac myocytes,
J. Biol. Chem. 272 (1997) 17744 –17748.
[46] V.O. Rybin, X. Xu, M.P. Lisanti, S.F. Steinberg, Differential targeting of beta-
adrenergic receptor subtypes and adenylyl cyclase to cardiomyocyte caveolae. A
mechanism to functionally regulate the cAMP signaling pathway, J. Biol. Chem.
275 (2000) 41447 –41457.
[47] R.D. Lasley, P. Narayan, A. Uittenbogaard, E.J. Smart, Activated cardiac adenosine
A(1) receptors translocate out of caveolae, J. Biol. Chem. 275 (2000) 4417 –4421.
[ 4 8 ]C .D e s s y ,R . A .K e l l y ,J . L .B a l l i g a n d ,O .F e r o n ,D y n a m i nm e d i a t e sc a v e o l a r
sequestration of muscarinic cholinergic receptors and alteration in NO signalling,
EMBO J. 19 (2000) 4272 –4280.
[49] A. De Luca, M. Sargiacomo, A. Puca, G. Sgaramella, P. De Paolis, G. Frati, C. Morisco,
B. Trimarco, M. Volpe, G. Condorelli, Characterization of caveolae from rat heart:
localization of postreceptor signal transduction molecules and their rearrange-
ment after norepinephrine stimulation, J. Cell. Biochem. 77 (2000) 529 –539.
[50] P. Huang, W. Xu, S.I. Yoon, C. Chen, P.L. Chong, L.Y. Liu-Chen, Cholesterol reduction
by methyl-beta-cyclodextrin attenuates the delta opioid receptor-mediatedsignaling in neuronal cells but enhances it in non-neuronal cells, Biochem.
Pharmacol. 73 (2007) 534 –549.
[51] M.A. Hanson, V. Cherezov, M.T. Grif fith, C.B. Roth, V.P. Jaakola, E.Y. Chien, J.
Velasquez, P. Kuhn, R.C. Stevens, A speci fic cholesterol binding site is established
by the 2.8 A structure of the human beta2-adrenergic receptor, Structure 16
(2008) 897 –905.2829 J. Brejchová et al. / Biochimica et Biophysica Acta 1808 (2011) 2819 –2829