| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Laboratoire Plantes-Microbe-Environnement, UMR INRA 1088/CNRS 5184, and
UMR 866 Lipides, Nutrition, Cancer, Equipe Physiopathologie des Dyslipidémies, Université de Bourgogne, Dijon, France;
Laboratoire de Biogenèse Membranaire, UMR 5200 CNRS-Université Victor Segalen Bordeaux 2, Bordeaux, France; and
Laboratoire de Génie des Procédés Microbiologiques et Alimentaires, Ensbana, Dijon, France
1Correspondence: Laboratoire Plantes-Microbe-Environnement, UMR INRA 1088/CNRS 5184/Université de Bourgogne, 17 rue Sully, BP 86510, 21065 Dijon cedex, France. E-mail: patricia.gerbeau-pissot{at}dijon.inra.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Key Words: cyclodextrin • detergent-resistant membranes • membrane lateral organization • microdomains • tobacco • spectrofluorimetry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. Indeed, cholesterol insertion in the bilayer is known to regulate lipid chain order and modify the thermotropic phase transition between liquid-disordered (ld) and solid-ordered phases by inducing an intermediate liquid-ordered (lo) phase (2)
. This lo phase combines a high rotational or translational mobility and a high conformational order in the lipid acyl chain. Consequently, cholesterol has the distinctive feature of increasing bilayer permeability and optimizing the mechanical properties of the PM (1
, 3)
while maintaining a liquid form. In addition, biophysical and biochemical studies have shown that cholesterol has the ability to laterally organize the bilayer components in specialized microdomains, or lateral heterogeneities, such as lipid rafts (4
, 5)
. These domains provide for lateral compartmentalization of PM proteins and thereby create a dynamic scaffold to organize cellular processes and synchronize efficiency and specificity of cellular responses (6
, 7)
. The lipid rafts are small, highly dynamic platforms, enriched in sphingolipids and cholesterol, with a high content in saturated acyl chains, and are mainly located in the outer exoplasmic layer (8)
. In animal PM, the presence of cholesterol is crucial for microdomain formation either by forming reversible condensed complexes of defined stochiometry with phospholipids, by ordering phospholipid chains to avoid contact with the polar environment, or by its preferential location at the boundary of sphingomyelin-rich regions (see refs. 2
, 9
for reviews). The physical reality of lipid rafts is thought to be depicted by the lo phase. Numerous studies, most using the cyclic oligosaccharide methyl-β-cyclodextrin (MβCD) to lower the cholesterol content of living animal cells, have provided evidence that the functional integrity of lipid rafts, or their biochemical counterparts detergent-resistant membranes (DRMs), depends on the presence of cholesterol (10
11
12)
.
In contrast to fungal or mammalian cells, where ergosterol or cholesterol are, respectively, the major sterol, higher plant cells contain a vast array of sterols. For example, 61 sterols and pentacyclic triterpenes have been identified in maize seedlings (13)
. This complex sterol mixture, mainly located in the PM, is an interesting characteristic of plants (14)
. Phytosterols differ from cholesterol by an extra alkyl group on the side chain in the C24 position (Fig. 1A
). The 
5 sterols having an ethyl group are represented by β-sitosterol and stigmasterol. Stigmasterol additionally contains a double bond in the C22 position and is by far the most abundant plant sterol. The proportion of other sterols such as campesterol is genetically defined (15)
. In addition, phytosterols can be conjugated with sugar (generally glucose), which, in turn, can be acylated by fatty acids to form steryl glycosides (SGs) and acyl-steryl glycosides (ASGs), respectively. The composition of sterol mixtures resulting from this variability in plant lipids could suggest a regulation of the structural and functional properties of PM and have consequences on plant physiology. For example, a correct ratio of β- sitosterol to campesterol is imperative for normal plant growth and sexual reproduction (16)
, and it varies during development despite being genetically defined (17)
. Furthermore, stigmasterol and sitosterol seem to be more efficient than ergosterol in extending the temperature range for functioning of membrane-associated biological processes (18)
. As PM phospholipid contents remain highly conserved between plant species, regulation of phytosterol composition appears to be a crucial adaptative process (18
, 19)
.
|
In plants, investigations about the presence of PM microdomains are very recent and limited to a small number of publications. A few years ago, Peskan et al. (20)
reported the first isolation of Triton X-100-insoluble fractions from tobacco PM. Mongrand et al. (21)
subsequently provided a detailed analysis of the lipid composition of such DRM fractions and reported an enrichment in sphingolipid, phytosterols, and steryl-conjugates compared with the whole PM. Characterization of their protein content has demonstrated a significant increase also in the relative abundance of proteins related to signaling and to response to biotic and abiotic stress, suggesting a role as a signaling platform (22)
. Similar observations have been concerning DRMs prepared from Arabidopsis thaliana callus (23)
and from Medicago truncatula roots (24)
. Although plant DRMs have been biochemically characterized, their relation to physical characterization of lipid domains in the plant cell PM has not yet been established. Moreover, the membrane-ordering effect of phytosterols has only been shown in model membranes (18
, 25)
. Therefore, the role of phytosterols in the organization of the plant PM needs to be clarified.
The aim of the present work, therefore, was to analyze the involvement of phytosterols in the lateral heterogeneity and organization of the plant PM. We report for the first time the capacity of MβCD to deplete sterols in isolated plant PM purified from BY2 tobacco cells. The influence of this modification in the phospholipid-to-sterol ratio on the phase behavior of the PM was monitored using steady-state fluorescence spectroscopy of ampiphilic markers of membrane organization: Laurdan (26
27
28
29)
, DPH and TMA-DPH (30)
, and di-4-ANEPPDHQ (31)
. Finally, the effect of sterol depletion on the isolation of DRMs was investigated.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Chemicals were purchased as follows: SG and ASG from Matreya (Pleasant Gap, PA, USA); 2-hydroxy-3-(N,N-dimethyl-N-hydroxyethyl)ammoniopropyl]-4-[β-[2-(din-butylamino)-6- napthyl]vinyl]pyridinium dibromide (di-4-ANEPPDHQ) from Molecular Probes Inc. (Eugene, OR, USA); 1,6-diphenyl-1, 3,5-hexatriene (DPH), 1-(4-trimethylammoniumphenyl)-6- phenyl-1,3,5-hexatriene (TMA-DPH), 2-dimethylamino-6-lauroylnaphtalene (Laurdan) and all others from Sigma-Aldrich Corp. (St. Louis, MO, USA).
Solvents, dimethylsulfoxide (DMSO) and tetrahydrofuran (THF), were of spectroscopic grade.
Preparation of highly enriched BY2 cell PM fractions
All steps were performed at 4°C. Cells (100 g) were collected by filtration, frozen in liquid N2, and homogenized in grinding medium [50 mM Tris-MES, pH 8.0; 500 mM sucrose; 20 mM EDTA; 10 mM dithiothreitol (DTT); 1 mM PMSF]. The homogenate was centrifuged at 16,000 g for 20 min. The supernatant was collected, filtered through 2 successive meshes (63 and 38 µm) and centrifuged at 96,000 g for 35 min. The pelleted microsomal fraction was purified by partitioning in an aqueous 2-phase system (polyethylene glycol 3350:dextran T-500, 6.6% each) to obtain the PM-enriched fraction, according to Mongrand et al. (21)
.
MβCD treatment
PM fractions were incubated 30 min at room temperature with 20 mM MβCD in buffered conditions (20 mM Tris, pH 7.6; 150 mM NaCl; 1 mM PMSF) and in the presence of a magnetic stirrer.
Sterol characterization
Total lipids were extracted by the method of Bligh and Dyer (32)
. Briefly, PM fractions (100 µg) were transferred to glass tubes and resuspended in a solution consisting of 900 µl methanol, 500 µl chloroform, and 400 µl of KCl (0.37 M) supplemented with epicoprostanol (5β-cholestan-3
-ol, 100 µl in methanol) as the internal recovery standard for quantification. After agitation for 2 h, 1.5 ml of chloroform was added, and total lipids were saponified by heating 1 h at 80°C with 1 ml ethanol and 100 µl KOH (11 N). The nonsaponifiable fraction containing total sterols were extracted by hexane and transformed into the trimethylsilyl derivatives. Analytical gas chromatography was carried out on Chrompack chromatograph equipped with a flame ionization detector and a VF-5 capillary column (15 m x 0.32 mm i.d.; FactorFour; Varian, Palo Alto, CA, USA), using the following operating conditions: 3.5 ml/min carrier gas helium flow; 260 and 300°C oven and detector temperature, respectively. Sterol species were identified by comparison of their retention time with those of standard solutions. Peak areas for each sterol were normalized against the epicoprostanol standard.
Protein and phospholipid quantification
Protein concentrations were determined by the Bradford assay or, to avoid sodium dodecyl sulfate interference, according to Schaffner and Weissman (33)
, using bovine serum albumin as a standard. Total phospholipid concentrations were determined by ammonium ferrothiocyanate coloration (34)
using phosphatidylcholine as the standard.
Protein separation
After solubilization in a Laemmli buffer, 30 µg PM proteins (treated or not with MβCD) were separated in a 10% polyacrylamide gel, fixed overnight, and revealed by silver nitrate staining.
Purification of detergent-resistant membrane fractions
PM fractions were resuspended in buffer containing 50 mM Tris-HCl, pH 7.4; 3 mM EDTA; and 1 mM DTT and treated with 1% Triton X-100 (w/v) for 30 min on ice with very gentle shaking every 10 min. Solubilized membranes were placed at the bottom of a centrifuge tube, mixed with 60% sucrose (w/w) to reach a final concentration of 48% (w/w), and overlaid with a discontinuous sucrose gradient (40, 35, 30, and 20%, w/w). After centrifugation for 20 h at 110,000 g, a ring of Triton X-100-insoluble membranes was observed at the 30–35% interface. Fractions of 1 ml were collected from this interface, and proteins were precipitated successively by 20 and 15% trichloroacetic acid (w/v), 1 h on ice, and cold acetone. Pellets were suspended in a minimal volume of Laemmli buffer, boiled 10 min, placed on ice 5 min, shaken overnight at room temperature, and then quantified.
Western blot experiments
After solubilization in Laemmli buffer, 10 µg of proteins was separated in a 10% polyacrylamide gel (1 h, 20 mA/gel) and then electroblotted onto a polyvinylidene difluoride membrane (20 min, 15 V; Trans-Blot SD semidry transfer cell; Bio-Rad Laboratories, Hercules, CA, USA) in transfer buffer (192 mM glycine, 20% methanol). Membranes were blocked overnight at 4°C in TBS (250 mM NaCl, 50 mM Tris, pH 7.6)-Tween 20 (0.05%) supplemented with 5% (w/v) powdered milk, washed 3 times for 7 min in TBS-Tween 20 (0.05%), and incubated 1 h in primary antibody, NtrbohD diluted 1/600 in TBS-Tween 20 (0.05%) (21)
. Blots were rinsed 3 times for 7 min and then incubated for 1 h with horseradish peroxidase-conjugated anti-mouse immunoglobulin G (IgG; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). After washing 3 times for 7 min in TBS, detection was performed as described in the ECL Western blot kit (Amersham Bioscience, Little Chalfont, UK).
Lipid analysis
Lipids were extracted and purified from the various fractions, according to Bligh and Dyer (32)
. Polar and neutral lipids were separated by monodimensional HP-TLC using the solvent systems described previously (24)
. Lipids of PM fractions were detected after staining with copper acetate (sterol-containing and phospholipid compounds were identified by their blue and brown color, respectively) and quantified from HP-TLC plates by densitometric scanning (35)
from 3 independent biological replicates. Lipids were identified by comparing their migration time with those of standards.
Fluorescence spectroscopy
Fluorescence measurements were performed on a Fluorolog-3 FL3–211 spectrometer (Jobin-Yvon, Horiba Group; Edison, NJ, USA) in the T-format with one double-monochromator for excitation and 2 single-monochromators for emission light. Detection is ensured by 2 photomultipliers. The light source is a xenon arc lamp. This spectrophotometer is equipped with movable excitation and emission polarizers. All fluorescence signals were recorded with emission and excitation bandwidths of 1 nm and 5 nm, respectively, except for anisotropy experiments (2 and 5 nm), and systematically corrected from light scattering of an unlabeled sample. Integration time was 1 s for spectra and anisotropy profile acquisition and 3 s for other temperature-dependent traces. All data acquisition was performed with the Datamax software (Jobin-Yvon/Thermo Galactic, Inc., Salem, NH, USA). Samples were stirred and equilibrated in a temperature-controlled chamber using a thermoelectric Peltier junction (Wavelength Electronics Inc., Bozeman, MT, USA). Experiments were done using a 10 mm special optic glass path cuvette filled with 2.5 ml of PM adjusted to a lipid concentration of 0.04 g/L in buffer (10 mM Tris-MES, pH 7.5; 1 mM EDTA; 250 mM sucrose). The probe-to-lipid ratio was estimated between 1:200 and 1:300, depending on the probe. All the temperature-dependent measurements were performed with a heating rate of 2°C/min.
Laurdan spectrofluorimetry
PM was labeled with 1.7 µl 0.4 mM Laurdan stock solution (DMSO) added in the cuvette and incubated at 4°C during 15 min in the dark. The excitation wavelength was 360 nm for Laurdan emission spectra. The generalized polarization excitation (GPex) was derived from the equation:
 | (1) |
Fluorescence polarization experiments
DPH and TMA-DPH labeling was done by adding 2.6 µl 0.25 mM stock solution (THF) in the cuvette, with a 15 min incubation at 4°C in the dark. Fluorescence anisotropy (r) was calculated using the equation:
 | (2) |
Spectrofluorimetry using the di-4-ANEPPDHQ probe
A volume of 0.7 µl di-4-ANEPPDHQ 1.5 mM stock solution (DMSO) was used to label samples. Emission spectra and temperature profiles of the ratio of intensities at 660 nm and 550 nm (I660/I550) were performed after a 15 min incubation at 4°C in the dark, with an excitation wavelength of 480 nm.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
. Gas chromatography (GC) to identify and quantify phytosterols gave a sterol-to-protein ratio of 0.16 ± 0.02 (w/w, n=14), which is in the commonly observed range.
Data obtained from GC profiles are illustrated in Fig. 1B
. The PM of 7-day-old tobacco suspension cells contains stigmasterol, campesterol, and sitosterol, which together constitute more than 95% of the membrane sterol mixture. Sitosterol is often the most abundant phytosterol in plant cells, although its level reached only 32% in the BY2 cell PM. This result indicates a slight subenrichment in tobacco suspension cells, as previously observed in tobacco leaves (21)
. Stigmasterol and campesterol were found at similar levels, 34 and 27%, respectively, of the total sterol mixture. The proportion of total 24-ethyl sterols, including isofucosterol (5%), reached 71% of the total sterol mixture, in agreement with previous experiments, suggesting a ratio of more than 60% in PM (36)
. Cholesterol is classically poorly represented in plant cells, and its BY2 cell PM level was not above the background.
Effect of MβCD on PM sterol content
Highly enriched PM fractions were treated with increasing concentrations of MβCD (0–20 mM) during incubation times varying from 0 to 40 min (data not shown). Data obtained from GC experiments showed that sterol depletion was maximum (50%) for 20 mM and 30 min (Fig. 2A
). The sterols remaining in the PM pellet seemed inaccessible to MβCD because longer incubation times failed to extract them (data not shown).
|
Phytosterol diversity in PM raised the question of sterol solubilization specificity of MβCD. In mammalian cells, cholesterol is widely depleted but without other sterols to compete with it. Under our conditions, depletion of campesterol, β-sitosterol, and isofucosterol reached 66, 54, and 47%, respectively, whereas stigmasterol seemed to be more resistant to MβCD treatment, with only 35% depletion (Fig. 2B
). The relative amount of sterols solubilized by MβCD in the aqueous supernatant was quite consistent with that found in the pellet of depleted PM (data not shown). This result suggests that differential extractability by the solvent used for purification and analysis can be excluded. Moreover, because the less abundant of the 3 major sterols, campesterol, was the most efficiently depleted, sterol availability did not seem to influence the level of sterol extraction.
Sterol distribution was changed by MβCD treatment as revealed by the GC analysis presented in Fig. 2B
. Compared with control PM, stigmasterol was still the major PM phytosterol, increasing from 34 to 45% of the total sterol mixture. Campesterol content was strongly reduced from 27 to 18%, whereas β-sitosterol was slightly affected (from 32 to 28%). No significant effect on isofucosterol content was observed.
Sterol specificity of MβCD
To check for nonspecific effects of MβCD, we measured the quantity of PM proteins and phospholipids after treatment. An incubation of 30 min at room temperature, either with MβCD or control buffer, induced no loss of proteins or total phospholipids associated with the membrane fraction (Fig. 3A
). The global amount of other PM components was also not affected by MβCD, confirming its specificity for sterols.
|
To further investigate possible qualitative modifications of PM content induced by MβCD treatment, BY2 cell PM proteins and lipids were separately analyzed. Gel electrophoresis revealed similar profiles for PM obtained in control, buffer-incubated, or MβCD-treated conditions (Fig. 3B
). Quantitative analysis by HP-TLC showed an equal representation of each lipid component (Fig. 3C
), except for free sterols (depleted as expected from GC data). Also, the amounts of SG and ASG were not modified by MβCD treatment (Fig. 3C
), with a constant ratio of SG:ASG in treated and control PM fractions (0.72±0.2).
Alteration of DRMs by MβCD treatment
Membrane fractions enriched in sterols and assumed as biochemical counterparts of microdomains can be isolated on the basis of their resistance to solubilization by nonionic detergents such as Triton X-100 at low temperature. To determine whether they are disrupted by removal of membrane-free sterols, the classical procedure of DRM isolation (solubilization at 4°C by Triton X-100, followed by centrifugation on a sucrose gradient) was used for PM treated or not with MβCD (30 min, 20 mM). The separation gradient was highly reproducible, and proteins were extensively (ca. 80%) recovered in the first 5 fractions corresponding to the solubilized membrane (Fig. 4A
). In control experiments, the amount of proteins recovered in the DRM fraction was around 7% (w/w) of the final quantity retrieved and was localized in the subfraction 8 as an opalescent ring of insoluble material. When PM was treated with MβCD, no DRM ring was observed in the sucrose density gradient. Moreover, no significant amount of proteins was detected in fractions other than 1 to 5 (Fig. 4A
). Because sterol depletion induced by MβCD was checked in all experiments, these data clearly indicate that the capacity to isolate DRMs from PM depends on the sterol content of the membrane.
|
Using antibodies against the DRM-enriched protein, NtrbohD (21)
, to follow localization on the gradient after MβCD treatment of a protein normally trapped in sterol-enriched domains, NtrbohD was mainly detected in subfraction 8, suggesting its enrichment in DRM (Fig. 4B
). MβCD treatment of PM modified NtrbohD repartition with a shift of the signal down to TX100-soluble subfractions at the bottom of the gradient (Fig. 4B
), suggesting that sterol depletion prevents location of the protein in a membrane fraction exhibiting particular physical properties.
Fluorescence anisotropy of DPH and TMA-DPH
Consequences of the MβCD treatment of PM isolated from BY2 cells were investigated by steady-state fluorescence anisotropy measurements of the membrane probes DPH and TMA-DPH, located in the bilayer core and at the interfacial region, respectively. Fluorescence anisotropy is correlated to the rotational diffusion of membrane-embedded probes and is sensitive to local viscosity. The anisotropy values of DPH gradually decreased as temperature increased for BY2 PM treated or not with MβCD, from 0.28 at 4°C to 0.15 at 49°C (Fig. 5A
). In the case of TMA-DPH, the amplitude of the decrease was smaller, ranging from 0.32 to 0.25. Anisotropy values obtained with TMA-DPH were systematically higher, whatever the temperature. These differences in behavior simply reflect the different location of the probes in the bilayer, as has been observed in various membrane systems (37
, 38)
. For both probes, the temperature dependence of anisotropy is rather linear and no marked change in the slopes of these profiles was observed, indicating a lack of thermotropic phase transition. The MβCD treatment of membranes resulted in a slight increase in slopes of temperature profiles for DPH and TMA-DPH. Linear regression gave slope factors of (–2.92±0.05/°C) x 10–3 vs. (–2.98±0.05/°C) x 10–3 for DPH, and (–1.31±0.02/°C) x 10–3 vs. (–1.46±0.02/°C) x 10–3 for TMA-DPH, for control and treated PM, respectively. Thus, MβCD treatment did not have a significant effect on membrane fluidity as probed by DPH and only a slight influence as monitored by TMA-DPH.
|
Monitoring of the generalized polarization excitation of Laurdan
Steady-state Laurdan fluorescence spectroscopy was performed on BY2 isolated PM, with or without MβCD treatment, between 4°C and 49°C. The normalized emission spectra of Laurdan in PM exhibited a maximum at 430 nm at low temperatures. An additional band at 490 nm appeared at higher temperatures (Fig. 5B
). Such temperature-dependent spectral changes have been initially reported for phospholipid bilayers in the liquid-crystal phase (26)
and in various membrane systems (39
40
41)
. For MβCD-treated PM, emission spectra of Laurdan indicated an increase in the ratio of intensities at 490 to 430 nm (Fig. 5B
). This increase is greatest at the highest temperature.
The spectral changes in the emission spectrum of Laurdan are quantified by the GPex function (Eq. 1)
, which was defined by analogy to classical polarization (26)
. In our measurements, GPex values varied from 0.65 to 0.20 in the 4–49°C range. This is in accordance with various studies utilizing cholesterol-containing lipid-mixture bilayers as model membranes (42
, 43)
. According to these models (28)
, the BY2 PM would be mainly constituted of gel-like domains below 20°C (GPex>0.55), and liquid-crystal domains, still partly ordered, would progressively appear above this temperature (0.55>GPex>0.25).
The temperature dependence of GPex values was used to monitor the thermotropic behavior of BY2 PM (Fig. 5C
). Typically, natural membranes do not exhibit the sharp phase transitions observed in phospholipid bilayers (39
, 44)
. Instead, a gradual decrease in GPex with temperature is observed. Here, GPex decreased in a nonlinear manner, with a marked plateau at low temperatures, indicating a thermotropic phase behavior (39
, 40)
. This decrease was more pronounced for MβCD-treated PM. The GPex data were fitted to a linear regression for data points above 25°C to determine valuable slope factors (28)
that reflect the cooperativity of the transition. Slopes of (–1.12±0.02/°C) x 10–2 and (–1.27±0.01/°C) x 10–2 were found for control and MβCD-treated membranes, respectively, in very good agreement with the temperature dependency of GPex in artificial and cell membranes (28)
. Thus, MβCD treatment led to an increase in the cooperativity index of the thermotropic transition.
The wavelength dependence of GPex provides information about the phase states of the membrane, basically, the gel, liquid-crystal, or coexisting phases. Briefly, this wavelength-dependence depends on the occurrence of a photoselection process originating from the particular features of Laurdan excitation spectra in phospholipid bilayers (44)
. The profiles of the GPex spectra (Fig. 5D
) were not noticeably modified by MβCD treatment at the tested temperatures, but a global decrease in GPex values was observed, whatever the wavelength, with temperature increase on the one hand, and with MβCD treatment on the other, except at 4°C, at which GPex was slightly higher for treated membranes. At a low temperature (4°C), the GPex spectra did not show significant wavelength dependence. This result suggests the predominance of an ordered gel-like organization at low temperatures, confirmed by the slightly higher disorder noticed for nontreated PM. At 22°C, the GPex spectra displayed a minor wavelength-dependent decrease, which became statistically significant at 40°C. This result supports the conclusion that Laurdan probes an environment in the BY2 PM that contains lo-like domains at temperatures above 22°C.
Dual wavelength ratiometric fluorescence analysis of di-4-ANEPPDHQ
This fluorescent dye was recently described as a promising probe of membrane domains (31)
, its emission spectrum being sensitive to local cholesterol concentration and lipid phase. The emission spectra of di-4-ANEPPDHQ within BY2 cell PM were investigated in the 4–49°C temperature range. The emission peak wavelength varied between 580 nm and 595 nm, depending on temperature and MβCD treatment (Fig. 6A
). These wavelength values are comparable to those observed by Jin et al. (31)
for phospholipid/cholesterol mixed bilayers and neutrophil cells. The normalized emission spectra showed a continuous red-shift with temperature increase, from 580 nm (4°C) to 590 nm (40°C). Similar temperature-dependent red-shift has been observed for phospholipid bilayers in liquid-crystal phase (31)
. When PM was treated with MβCD, the emission spectrum of the dye clearly red-shifted and broadened (Fig. 6A
). To quantify these spectral changes, a dual-wavelength ratiometric scheme was employed. We divided the emission intensity at 660 nm and 550 nm and plotted the ratio against the temperature (Fig. 6B
) to assess the thermotropic behavior of PM. These plots exhibited gradually increasing temperature dependence and followed a rather nonlinear relation. To determine slope factors of the linear part of the curves, data were fitted to a linear regression function above 10°C. This resulted in an increase in the slope factor from (0.85±0.04/°C) x 10–2 to (1.07±0.09/°C) x 10–2 on MβCD treatment. Seemingly, di-4-ANEPPDHQ senses that the MβCD treatment affects the apparent thermotropic transition essentially through an increase in the cooperativity index. Another major consequence of the MβCD treatment was the pronounced positive shift in intensity ratio, which varied from 0.08 (4°C) to 0.17 (49°C). This positive shift was also significant at 4°C, a temperature at which the PM would be probably in a gel-like phase, even after MβCD treatment.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
, or bilayers, by nuclear magnetic resonance (18
, 46)
, neutron diffraction (47)
, fluorescence spectroscopy (25
, 48
, 49)
, differential scanning calorimetry (50)
, and small-angle X-ray scattering (51)
. Despite efforts to improve biomimetic model systems and the information subsequently collected about sterol/phospholipid interactions, an approach involving phytosterol manipulation in plant PM has not been previously reported.
In animal cells, cyclodextrins are well known to affect cholesterol content (52)
, and they have also been used to manipulate ergosterol in yeast (53)
. To the best of our knowledge, the only reports of a phytosterol-cyclodextrin interaction are those in which the cyclic oligosaccharide was tested as a vehicle for sterol incorporation or in substitution experiments (37
, 54)
. In the present work, the efficiency of MβCD to remove phytosterols from plant cell-isolated PM is demonstrated for the first time. The estimated sterol depletion of 50% is consistent with data reporting MβCD-cholesterol depletion levels from mammalian cell membranes (39)
.
It is important to note that MβCD does not affect either phospholipids or plant-specific steryl-conjugates, perhaps because of the polar head created by the sugar moiety, preventing access of MβCD to the membrane-embedded sterol moiety. The modifications observed for MβCD-treated PM organization can thus be attributed to changes in PM free-sterol levels and/or composition.
Phase behavior of BY2 PM
When phase properties of BY2 PM were monitored in the 4–49°C temperature range using 4 fluorophores, the Laurdan GPex temperature profiles clearly established a thermotropic phase behavior, with a break in the profile around 20°C, characteristic of a gel-to-liquid phase transition. The liquid phase is characterized by a gradual decrease in membrane order with temperature increase, which strongly suggests the presence of lo-like structures, whereas wavelength dependence of GPex identified a gel-like phase at low temperatures. Earlier GPex Laurdan experiments on garden pea PM (55)
displayed a comparable thermotropic behavior. Temperature profiles obtained with the intensity ratio of the di-4-ANEPPDHQ dye similarly showed thermotropic behavior, with a break at low temperatures followed by a rather linear increase in the intensity ratio. The pioneering work on the characterization of this dye as a membrane probe (31)
evidenced a linear relation with temperature for 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) liposomes in a single liquid-crystal phase, whereas it showed phase transition for 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC):cholesterol (30:70) liposomes in the 20–60°C temperature range. This was identified as a lo-to-ld phase transition, even though for many authors (e.g., ref. 48
) this transition is rather gel-to-liquid crystal. The temperature plots we report, which are the first ever obtained for an isolated PM system with di-4-ANEPPDHQ, are consistent with Laurdan data and show this dye to be a good reporter of natural membrane organization because comparable absolute slope factors were found for Laurdan (1.12) and di-4-ANEPPDHQ (0.85). Slope factor is related to the cooperativity of phase transition and is representative of membrane heterogeneity. The rather linear thermotropic behavior of fluorescence anisotropy of DPH and TMA-DPH inserted in BY2 PM illustrates the absence of phase transition. Interestingly, similar trends have been observed with other plant PM systems, such as N. tabacum (56)
and corn roots (57)
, with comparable sterol-to-protein ratios. Nevertheless, as has already been proposed, DPH and TMA-DPH anisotropy may not be as efficient as Laurdan in revealing phase behavior in natural membranes (39)
, possibly because of the presence of proteins. In addition, light scattering in turbid media can be a problem in fluorescence anisotropy, which is not met with Laurdan GPex (39)
. Consequently, we were able to observe several BY2 PM phases: gel-like or liquid crystal with distinct fluidity and order properties.
Phytosterols modulate plant PM lateral organization
The change in relative sterol content (from 0.16 to 0.08 w/w lipid ratio) following MβCD treatment influenced the phase properties of PM differently, depending on the probe utilized. A relatively slight effect on DPH and TMA-DPH rotational diffusion was observed by fluorescence anisotropy. This tendency is consistent with phytosterol depletion, according to a temperature-dependence study of TMA-DPH fluorescence anisotropy in DPPC-sterol vesicles (48)
. Indeed, anisotropy values reflect both structural (order) and dynamic (fluidity) membrane properties (29
, 42)
, whereas only structural consequences, (i.e., decreased acyl chain order) are expected from sterol depletion. The Laurdan GPex temperature profiles were, on the contrary, highly influenced by MβCD treatment. For temperatures higher than 20°C, GPex decreases corresponded to an increase in acyl-chain disorder, as described for cholesterol-phospholipid bilayers (43)
and bovine hippocampal membranes depleted with MβCD (39)
. Previous Laurdan GPex experiments on liposomes (49)
have already shown the ability of various phytosterols to order model membranes. Moreover, the increase in cooperativity index of the GPex temperature plots indicated a correlation of sterol depletion with an alteration in membrane lateral organization, that is, with a decrease in liquid-phase heterogeneities (39)
. Based on the abundant literature that documents the relation between bilayer organization and Laurdan spectroscopic behavior in various membrane systems (26
27
28
29
, 44)
, our data support the conclusion that at room temperature, PM contains lo-like domains that are partially lost after sterol depletion, thus accounting for the diminution in membrane heterogeneities. Similar alteration in temperature profile after sterol depletion was found for di-4-ANEPPDHQ, with an increase in cooperativity index. This result is concomitant with a transition involving lo-like and ld-like domains (31)
and emphasizes the potential of di-4-ANEPPDHQ to characterize sterol-dependent phase changes in natural membranes. In agreement with this, sitosterol has already been reported to be able to promote formation of small-size highly ordered domains (18)
. The consistently positive intensity ratio shift at low temperatures, corresponding to a gel-like state in which sterols should have a disordering effect, could directly reflect the difference in sterol contents between MβCD-treated and control PM. Hence, our results demonstrate that sterol depletion in BY2 PM leads to a diminution in overall liquid-phase heterogeneity that is correlated with disruption of phytosterol-rich lo-like domains.
These conclusions were corroborated by DRM fraction purification experiments. Numerous studies on animal and fungi membranes have previously shown that functional domains of the PM can be isolated as a DRM fraction on a sucrose gradient after solubilization in Triton X-100 (58)
. Such detergent-insoluble fractions were recently reported for the PM of higher plants: tobacco (21)
, A. thaliana (23)
, and M. truncatula (24)
. One major finding of our study is that phytosterol depletion leads to removal of the floating DRM fraction from a sucrose density gradient. In comparison with mammalian cells, the extent of depletion we found, although partial, is consistent with a major perturbation of DRM properties (59
60
61)
. Moreover, cholesterol depletion of Chinese hamster ovary cells induces formation of large, stable, and more fluid domains associated with partial solubilization of DRM compounds (62)
. MβCD-mediated cholesterol depletion of 60–70% has been found to induce a redistribution of a GPI-anchored protein from DRM to Triton-soluble fractions in melanoma cells (60)
. Such loss of raft markers from the low-density region of a sucrose gradient has also been noted in leukemia cells (61)
or PM isolated from human lymphoblastic leukemia T cells (59)
. Likewise, the tobacco raft marker NtrbohD was lost from the low-density region of the sucrose gradient after MβCD of PM. This result clearly indicates the predominant role of phytosterols in the lateral segregation of proteins within the PM. Furthermore, the association of proteins to the PM of tobacco cells is not dependent on the sterol-dependent organization of PM because the global proteic composition is preserved after MβCD treatment. Also, the major side effects of MβCD treatment observed in living cells, that is perturbations in cell functions such as cytoskeleton actin networks or clathrin endocytosis (9)
, are absent from purified PM. Thus, phytosterol depletion certainly causes disappearance of DRM from isolated PM, and our results clearly support the role of phytosterols in the lateral organization of the PM of BY2 cells.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Evidence for a relation among 1) DRM, 2) ordered sterol-rich domains, and 3) lipid rafts is convincing, although this depends on the detergent extraction method used (62
63
64
65)
. Thus, in the PM of BY2 cells, phytosterols would be key compounds to structure lipid rafts by ordering acyl chain of lipids and to promote them as platforms of membrane protein recruitment. Fluorescence microscopy investigations on living cells should be helpful to further characterize these microdomains.
Moreover, removal of sterol-dependent DRM fractions does not totally abolish the complexity of PM, and lateral heterogeneities are still present, which evokes a crucial role for phytosterols in structuring membrane by DRM-independent levels of organization. Indeed, the diversity in phytosterols retrieved from plant cell PM raises the question of the coexistence of lipid domains of varying composition and with distinct roles, the sterol composition of the PM being crucial for physiological functioning. Loading of PM with exogenous phytosterols should be informative about the role of each analog.
|
ACKNOWLEDGMENTS |
|---|
Received for publication May 13, 2008. Accepted for publication July 10, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
-alkylsterols and 24-β-alkylsterols in tracheophytes. Phytochemistry 16,683-686[CrossRef]This article has been cited by other articles:
|
|
 |
|
  M. Sharma, Z. Sasvari, and P. D. Nagy Inhibition of Sterol Biosynthesis Reduces Tombusvirus Replication in Yeast and Plants J. Virol., March 1, 2010; 84(5): 2270 - 2281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Bouvier-Nave, A. Berna, A. Noiriel, V. Compagnon, A. S. Carlsson, A. Banas, S. Stymne, and H. Schaller Involvement of the Phospholipid Sterol Acyltransferase1 in Plant Sterol Homeostasis and Leaf Senescence Plant Physiology, January 1, 2010; 152(1): 107 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Stanislas, D. Bouyssie, M. Rossignol, S. Vesa, J. Fromentin, J. Morel, C. Pichereaux, B. Monsarrat, and F. Simon-Plas Quantitative Proteomics Reveals a Dynamic Association of Proteins to Detergent-resistant Membranes upon Elicitor Signaling in Tobacco Mol. Cell. Proteomics, September 1, 2009; 8(9): 2186 - 2198. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Raffaele, E. Bayer, D. Lafarge, S. Cluzet, S. German Retana, T. Boubekeur, N. Leborgne-Castel, J.-P. Carde, J. Lherminier, E. Noirot, et al. Remorin, a Solanaceae Protein Resident in Membrane Rafts and Plasmodesmata, Impairs Potato virus X Movement PLANT CELL, May 1, 2009; 21(5): 1541 - 1555. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
