Effect of oxidative stress on plasma membrane fluidity of THP-1 ...

32 downloads 63899 Views 767KB Size Report
THP · Fig. 7. Effects of the azo initiator AAPH on lipid peroxidation and membrane fl ..... means of Adobe Photoshop 7.0. All GP ...... This might allow the mac-.
Biochimica et Biophysica Acta 1828 (2013) 357–364

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbamem

Effect of oxidative stress on plasma membrane fluidity of THP-1 induced macrophages Carlos de la Haba a, b, José R. Palacio a, Paz Martínez a, Antoni Morros b,⁎ a

Unitat d'Immunologia, Institut de Biotecnologia i de Biomedicina (IBB), Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Barcelona, Spain Unitat de Biofísica, Departament de Bioquímica i de Biologia Molecular and Centre d'Estudis en Biofísica (CEB), Facultat de Medicina, Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Barcelona, Spain

b

a r t i c l e

i n f o

Article history: Received 28 March 2012 Received in revised form 26 July 2012 Accepted 17 August 2012 Available online 23 August 2012 Keywords: Oxidative stress Macrophage Membrane fluidity Laurdan Two-photon microscopy Lipid raft

a b s t r a c t Plasma membrane is one of the preferential targets of reactive oxygen species which cause lipid peroxidation. This process modifies membrane properties such as membrane fluidity, a very important physical feature known to modulate membrane protein localization and function. The aim of this study is to evaluate the effect of oxidative stress on plasma membrane fluidity regionalization of single living THP-1 macrophages. These cells were oxidized with H2O2 at different concentrations, and plasma membrane fluidity was analyzed by two-photon microscopy in combination with the environment-sensitive probe Laurdan. Results show a significant H2O2 concentration dependent increase in the frequency of rigid lipid regions, mainly attributable to lipid rafts, at the expense of the intermediate fluidity regions. A novel statistical analysis evaluated changes in size and number of lipid raft domains under oxidative stress conditions, as lipid rafts are platforms aiding cell signaling and are thought to have relevant roles in macrophage functions. It is shown that H2O2 causes an increase in the number, but not the size, of raft domains. As macrophages are highly resistant to H2O2, these new raft domains might be involved in cell survival pathways. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Membrane fluidity is a very important physical feature of biomembranes as it is involved in cell functions such as regulation of enzyme activity, permeability, transport of nutrients, lateral motion of membrane constituents and osmotic stability of cells [1]. One of the ideas, on the biological membranes, which has been consolidating in recent years is that the membrane protein structure and function can be regulated and modulated by the composition, structure and dynamics, of its host lipid environment [2–4]. It has been recently described how the particular biophysical properties of membrane lipid bilayer may regulate the localization of receptors in the plasma membrane and the potential functional consequences in signal transduction [5]. It is now generally accepted that plasma membranes contain highly dynamic (fluctuant) nanoscale microdomains, commonly called lipid rafts. These microdomains are enriched in sphingolipids

Abbreviations: AAPH, 2,2′‐azobis(2‐amidinopropane) dihydrochloride; DMSO, dimethylsulphoxide; DPPC, dipalmitoylphosphatidylcholine; FBS, fetal bovine serum; GP, generalized polarization; GUVs, giant unilamellar vesicles; ld, liquid disordered; lo, liquid ordered; MTT, 3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazolium bromide; PMA, phorbol‐12‐myristate‐13‐acetate; ROS, reactive oxygen species; THP‐1, human acute monocytic leukemia cell line ⁎ Corresponding author. Tel.: +34 93 581 1177; fax: +34 93 581 1907. E-mail addresses: [email protected] (C. de la Haba), [email protected] (J.R. Palacio), [email protected] (P. Martínez), [email protected] (A. Morros). 0005-2736/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbamem.2012.08.013

and cholesterol and are associated to a subset of membrane proteins with an important role in cell signal transduction and membrane traffic [6–8]. Lipids within the plasma membrane are one of the preferential targets of reactive oxygen species (ROS) which cause lipid peroxidation. In particular, polyunsaturated phospholipids are an extremely vulnerable target, due to the susceptibility of its chains to lipid peroxidation. This process disturbs the bilayer structure, modifies membrane properties such as membrane fluidity, alters the physiological functions of cell membranes and contributes to cell membrane damage [9,10]. Altered membrane fluidity might also affect membrane protein function by modifying its lipid microenvironment and interactions [2–5]. Several reports have described a decrease of membrane fluidity in different cell membranes as a consequence of lipid peroxidation [11–13]. Although macrophages were originally recognized as major immune effector cells of the innate immune response, it is now appreciated that they also play other important roles in the maintenance of tissue homeostasis, and are involved in a variety of pathological conditions including cancer [14] or multi-factorial inflammatory diseases [15]. In atherosclerosis, inflammatory reactions induced by ROS contribute to increase atherosclerotic lesions [16]. In granulocytes, ROS and reactive nitrogen species (RNS) have a dual function: they function as signaling molecules, modulators of protein and lipid kinases and phosphatases, receptors, ion channels, and transcription factors [17]. It has been recently demonstrated that, during the immune

358

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364

and 5% CO2, in the presence of 0.16 μM phorbol 12-myristate 13-acetate (PMA). After macrophage induction, cells were left in fresh medium for 24 h before H2O2 was added. Cell to cell interaction produces an increase of rigid domains as it was described by Gaus [33]. Therefore our aim was to achieve an optimal cellular density (less 50% confluent) to minimize cell to cell contact, but enough to obtain cells for our study. Our results showed that for our study the optimal cellular concentration was 4 × 10 5 cells/plate.

response, macrophages suppress T cell responses by producing ROS leading to the subsequent induction of T regulatory cells in a ROS-dependent manner [18]. During an inflammatory response, ROS and RNS modulate phagocytosis, secretion, gene expression, and apoptosis [19]. Many recent reports show the pivotal role of dynamic remodeling of membrane microdomains in some of the relevant functions of immune cells, such as antigen recognition or the signaling across the cell [20–22]. In the case of the immune system, it has been suggested that oxidation of membrane lipids may serve as a critical step in innate and adaptive immune signaling [22]. Nevertheless, the effect of oxidative stress on macrophage plasma membrane fluidity remains poorly studied. Several imaging methods are presently used for investigating cell membrane lipid dynamics as recently reviewed [8,23,24]. Many of these techniques use fluorescent dyes sensitive to the packing of its lipid microenvironment, thus allowing to analyze membrane fluidity [25]. One of the more suitable fluorescent probes to study the regionalization of membrane fluidity is Laurdan (6-dodecanoyl2-dimethylamino-naphthalene), which exhibits an emission spectral maximum shift from 440 nm to 490 nm in the transition from a gel phase to a liquid phase [26,27]. By using a multiphoton confocal microscope it is possible to excite the fluorophore with two photons at double the wavelength causing it to fluoresce. This technique, known as two-photon microscopy, allows obtaining high resolution images of Laurdan-stained single cells, in vivo, which enables visualizing the distribution of membrane fluidity in different domains within the plasma membrane [26–32]. The aim of the present study is to evaluate the fluidity regionalization in the plasma membrane of single macrophages, in vivo, and the effects of oxidative stress, by using two-photon microscopy. We will use the monocytic line THP-1 induced to macrophages with phorbol-12-myristate-13-acetate (PMA).

Laurdan was dissolved to a concentration of 2 mM in DMSO and stored at − 20 °C at the dark. For membrane labeling, cells were incubated in medium without FBS at a final concentration of 5 μM Laurdan for 30–60 min at 37 °C and 5% CO2 with agitation.

2. Materials and methods

2.6. Two-photon microscopy technique

2.1. Liposome preparation

Membrane fluidity was evaluated with a multiphoton scanning confocal microscope Leica TCS-SP5 (Leica Microsystems, Heidelberg GmbH), after incubation with Laurdan, at the Servei de Microscòpia (UAB). For macrophages a 63× oil immersion objective lens and a 1.4 numerical aperture were used. Images were obtained at a resolution of 512× 512 pixels and at a scan speed of 400 Hz with LEICA LAS AF software. After fluorescence excitation with the multiphoton laser at 800 nm, the two-photon microscope captures two simultaneous emission images, with wavelengths ranging from 400–460 nm and 470–530 nm. Emission intensities from every image pixel were introduced into the generalized polarization (GP) equation (Eq. (1)) providing a final GP value, which is a measure of membrane fluidity:

Appropriate amounts of dipalmitoylphosphatidylcholine (DPPC, Avanti Polar Lipids (Alabaster, AL)) and of the membrane fluidity probe Laurdan (6-dodecanoyl-2-dimethylamino-naphthalene, Invitrogen) were dissolved in chloroform. The solvent was evaporated under an oxygenfree N2 stream and the resulting film was maintained under high vacuum overnight to remove organic solvent traces. Giant vesicles were obtained by gently applying PBS, pH 7.4, at a temperature 10 °C above the DPPC phase transition (Tm = 41 °C) on the film, to a final DPPC concentration of 2.5 mM (0.5 mg/ml) and Laurdan concentration of 5 μM. The sample was incubated at this temperature in resting conditions and in the dark for 24 h. This method rends mainly giant unilamellar vesicles (GUVs) of about 20 μm of mean diameter, although some oligolamellar vesicles could also be observed. For the fluorescence measurements by confocal microscope only GUV were selected. 2.2. Monocyte cell line culture Cells were maintained in the Servei de Cultius Cel·lulars, Producció d'Anticossos i Citometria (UAB). The human monocytic cell line THP-1 was cultured in RPMI medium 1640, 10% Fetal Bovine Serum (FBS) with GlutaMax (Invitrogen) and without phenol red. Cells were maintained at 37 °C in 5% CO2. All experiments were performed within the cell passages 20–50. 2.3. Induction of THP-1 monocyte cells to macrophages THP-1 monocytes were induced to macrophages in 35 mm MatTek glass bottom dishes at 4 × 10 5 cells/plate for 72 h at 37 °C

2.4. Analysis of THP-1 cell viability under oxidative stress conditions To analyze the effects of H2O2 on membrane fluidity, we first determined the highest H2O2 concentration allowing acceptable levels of viability. Macrophages were induced in 96 micro well plates for viability analysis. Cells were then washed three times with PBS. H2O2 in fresh medium without FBS was added to final concentrations ranging from 0.1 mM to 1 M. After 24 h at 37 °C and 5% CO2 cells were washed and viability was assessed by MTT assay (EZ4U Cell Proliferation Assay, Biomedica Gruppe, Wein, Austria). This assay measures the reduction of MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) to formazan (1-(4,5-dimethylthiazol-2-yl)-3,5-diphenylformazan) catalyzed by mitochondrial dehydrogenase in functional mitochondria. Results showed that the H2O2 concentrations suitable for our study ranged from 0.5 mM to 2.0 mM, which maintained cell viability above 80%. 2.5. Laurdan-staining of macrophage membranes

GP ¼

Ið400460Þ  G  Ið470530Þ Ið400460Þ þ G  Ið470530Þ

ð1Þ

where G is a correction factor for the microscope being used, which was calculated by using Eq. (2): G¼

GPtheo þGPtheo  GPexp 1  GPexp GPtheo  GPexp GPtheo þGPexp  1

ð2Þ

where GPtheo is the GP theoretical value of a standard solution of 5 μM Laurdan in DMSO, which has a known value (GPtheo = 0.207), whereas GPexp is the GP value of the same solution measured in our confocal microscope [34]. GP values range from − 1, corresponding to the highest fluidity, to + 1 for the lowest fluidity. GP images, therefore, show membrane fluidity distributions across the cell membrane providing an excellent tool for membrane dynamics visualization [35].

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364

2.7. Image analysis

3. Results

A WiT 8.3 imaging software (Dalsa Digital Imaging, Canada) was used and adapted for the image processing to obtain a GP value for each pixel. All calculations were carried out in floating point format and all images were first converted to 8-bit unsigned format [35]. Images were then processed with a custom made sub-program. Background values were set to zero: equation 1 denominator was converted to a binary image with background values set to zero and nonbackground values set to one, then the binary image was multiplied by the GP image [31]. Images were then pseudo colored by means of Adobe Photoshop 7.0. All GP images were corrected using the G factor previously obtained from a 5 μM Laurdan solution in DMSO for each experiment [35]. GP distributions were obtained from the histograms of the GP images.

3.1. Standardization of two-photon microscopy

2.8. Assessment of lipid peroxidation Lipid peroxidation in macrophage membranes was assessed by using C11-BODIPY581/591 (4,4-difluoro-5-(4-phenyl-1,3-butadienyl)-4bora-3a,4a-diaza-s-indacene-3-undecanoic acid, Invitrogen), a fluorescent fatty acid analogue which incorporates into membranes. Upon oxidation, both the excitation and emission fluorescence spectra of the dye shift to shorter wavelengths [36,37]. A 2 mM stock solution of C11-BODIPY581/591 was prepared by dissolving the probe powder in dimethylsulfoxide (DMSO) and then it was stored at −20 °C in the dark. For probe incorporation, macrophages were placed in 96-well plates (2.5 × 105 cells·mL−1), incubated in medium without FBS at a final concentration of 10 μM C11-BODIPY581/591 for 45 min at 37 °C and 5% CO2, and finally washed three times with PBS. H2O2 was then added to final concentrations ranging from 0.1 mM to 5 mM, and cells were incubated for 30 min and 24 h at 37 °C. Control cells were incubated in the same conditions without H2O2. A multi-well spectrofluorometer (Victor3, Perkin-Elmer) was used to measure the emissions corresponding to the oxidized (green fluorescence: λexcitation, 488 nm; λemission, 520 nm) and non‐ oxidized (red fluorescence: λexcitation, 545 nm; λemission, 590 nm) states of the probe. Oxidation of C11-BODIPY 581/591 was estimated by calculating the ratio between green fluorescence (oxidized) and total fluorescence (oxidized plus non‐oxidized), in order to normalize for probe uptake and distribution into cellular membranes [38]. Lipid peroxidation values were calculated from three different experiments, each experiment composed of triplactes. Lipid peroxidation induced by hydrogen peroxide was compared to the effect of the free radical initiator 2,2´-azobis(2-amidinopropane) dihydrochloride (AAPH) (Sigma-Aldrich). AAPH has been used successfully to study the actions of free radicals upon cell membrane lipid peroxidation [37,39,40]. To assess the azo compound oxidizing effect, 5 mM AAPH solution in medium without FBS was added to macrophages loaded with C11-BODIPY 581/591, as described above, in 96-well plates. The plates were incubated for 30 min and 24 h, at 37 °C. Oxidation of C11-BODIPY 581/591 was estimated by calculating the ratio between green fluorescence and total fluorescence.

359

Two-photon microscopy standardization was performed using model membranes of DPPC stained with Laurdan, at different temperatures, to follow the drastic fluidity changes produced in the bilayers throughout the gel to liquid-crystalline phase transition. Fig. 1A–C shows the Laurdan generalized polarization (GP) images of DPPC GUVs at temperatures corresponding to the fluid phase (50 °C) (Fig. 1A), to the phase transition temperature (42 °C) (Fig. 1B) and to the gel phase (30 °C) (Fig. 1C). GP images (Fig. 1A–C) were pseudocolored with an arbitrary color palette (Fig. 1D), which shows extreme fluidity (GP=–1) in purple color and extreme rigidity (GP=+1) in red color. Coherently, liquid crystalline phase liposome (Fig. 1A) appears with predominant blue color, liposome near the phase transition temperature is shown in yellow-green color (Fig. 1B), and gel phase liposome (Fig. 1C) shows orange-red color. GP images were transformed to GP distribution graphs (Fig. 1E) to evaluate liposome membrane fluidity distribution at each temperature. In addition to the GP distribution graphs shown in Fig. 1E, we obtained a series of GP histograms for DPPC at different temperatures ranging 30 °C to 50 °C. The GP values for the peaks of the resulting histograms were plotted as a function of temperature (Fig. 2). As it can be observed in Fig. 2, the GP peaks vs temperature plot describes the DPPC gel to liquid-crystalline phase transition with a Tm value near 43 °C.

3.2. Membrane fluidity distribution of THP-1 macrophages under oxidative stress induced by H2O2 For the evaluation of the effect of oxidative stress on macrophage membrane fluidity, THP-1 macrophages were incubated with H2O2,

2.9. Statistics Statistical analyses were performed by using SigmaPlot for Windows 11.0 (Systat Software, Inc.). In membrane fluidity studies, comparisons between control and oxidation treatments were performed by one-way analysis of variance based on ranks (Kruskal–Wallis) followed by post-hoc Dunn's test. For H2O2 experiments, the sample size, n, was 70 to 94. For azo compound experiments, the sample size, n, was 55 to 61. For all lipid peroxidation measures with C11-BODIPY 581/59, comparisons between control and oxidation treatments were performed using a Student's t-test. Statistical significance was indicated by p b 0.05.

Fig. 1. Validation of the technique two-photon microscopy with Laurdan-labeled DPPC GUVs: GP images of DPPC giant unilamellar vesicles (GUVs) stained with 5 μM Laurdan, in PBS at 50 °C (fluid phase) (A), 42 °C (near the gel to liquid phase transition) (B) and 30 °C (gel phase) (C). Scale bar, 10 μm. GP images were pseudocolored with an arbitrary color palette (D). (E) Experimental (dots) and fitted (continuous lines) normalized GP frequency distribution curves corresponding to the three images presented in (A–C). GP values range from −1 to +1, −1 being very fluid lipid domains and +1 very rigid lipid domains. For the fitting procedure we have used Gaussian or Weibul functions for asymmetric curves.

360

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364

Fig. 2. Temperature dependence of GP maxima for DPPC GUVs: dots represent the GP peak values of GP histograms at different temperatures (from 30 °C to 50 °C), as those shown in Fig. 1E.

they were stained with Laurdan, and membrane fluidity distribution was then analyzed by two-photon microscopy. For every treatment and for the control 70 to 94 cells were individually analyzed. Fig. 3A–D shows GP images of macrophages, at 37 °C, in control conditions (A) and in oxidizing conditions, in 0.5 mM H2O2 (B), 1 mM H2O2 (C) and 2 mM H2O2 (D), pseudocolored with an arbitrary color palette (E). It can be detected a substantial enrichment in high GP areas orange-red colored in the macrophage surface, corresponding to rigid domains, as H2O2 concentration increases. GP frequency distributions, at 37 °C, of control and treated macrophages, which presented interesting differences when compared, are shown in Fig. 3F. Two vertical lines have been drawn in Fig. 3F crossing the two GP points with common frequencies (GP ~ −0.28 and GP ~ + 0.22), to help defining three zones with different behavior in the GP scale. For a better observation of these frequency changes, GP frequency distribution values of control macrophages were subtracted from the corresponding values of each GP distribution of H2O2 treated macrophages (Fig. 3F). The resulting frequency difference curves are shown in Fig. 3G where the three above defined GP zones can also been observed. As it can be clearly seen, oxidation induces an increase of frequency difference in both ends of the GP scale (from − 1.00 to − 0.28, and from + 0.22 to + 1.00), whereas middle GP values (from −0.28 to + 0.22) decrease in frequency. In order to statistically analyze the frequency changes for every single cell, the area under the frequency distribution curve (Fig. 3F) of these three GP zones was measured and compared. The results of this analysis are shown in Fig. 4. In the intermediate and positive GP zones, significant differences (p b 0.05) were found between the control and each oxidation treatment. In summary, GP values above + 0.22 show a significant increase of frequency for oxidized macrophages. These frequency differences increase with H2O2 concentration and occur at the expense of a significant frequency decrease in the GP interval between −0.28 and +0.22. This means that more rigid regions are generated in macrophage plasma membrane as a consequence of oxidation. It is interesting to attribute the above described GP frequency changes to the different kinds of lipid phases, as it is accepted for lipid bilayers of living cell membranes. The GP values corresponding to these different lipid phases have been estimated by comparison with membrane models showing compositions similar to those of cell membranes [35,41,42]. GP values above +0.55 and below −0.05 represent membranes in gel and fluid phase, respectively [35,41]. GP values

Fig. 3. GP images and GP frequency distributions of control and oxidized Laurdanstained THP-1 macrophages: THP-1 monocytes were induced in MatTek plates with 0.16 μM PMA and incubated for 72 hours. THP-1 macrophages were washed with phenol free media and 0.0 mM H2O2 (control cells) (A), 0.5 mM H2O2 (B), 1 mM H2O2 (C) and 2 mM H2O2 (D) was added. Scale bar, 10 μm. After a 24 hour incubation period TPH-1 macrophages were stained with 5 μM Laurdan in serum free media. Images were collected (n = 70 to 94) at 37 °C in a two-photon microscope, GP images were calculated and then pseudocolored with an arbitrary color palette (E) (see Materials and methods). GP images were transformed to GP frequency distribution curves and normalized (sum = 100) (F): dots show experimental data and continuous lines show best fitting curves to the experimental data using Gaussian functions or Weibul functions for asymmetric curves. GP frequency distribution values of control macrophages were subtracted from the corresponding values of each GP distribution of H2O2 treated macrophages. The resulting frequency difference distribution curves are shown in G. Symbols indicate values of the resulting frequency differences and smoothed lines were obtained using a Savitsky–Golay algorithm.

approximately between +0.25 and +0.55 have been attributed to liquid-ordered (lo) or lipid raft domains whereas GP values between −0.05 and +0.25 would correspond to the surrounding nonraft regions or liquid-disordered (ld) phase [28,33]. We further investigated if raft domains, as platforms for efficient signaling, could increase in number or in size. Therefore, a novel software was developed in our laboratory to statistically analyze these two parameters from the data obtained for every single cell. The mean and SEM obtained for the size of domains with +0.25b GP b +0.55 (lipid rafts) and for the size of domains with GP >+0.55 (gel phase), for control or oxidized macrophage are shown in Fig. 5. No significant changes were found in the size of raft domains or in the size of gel phase domains as a consequence of H2O2 macrophage oxidation. However, the number of both lipid raft and gel phase domains significantly increased as the H2O2 concentration rises (Fig. 5).

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364

361

Fig. 4. Statistical analysis of the area of the three GP regions: The mean values of the area under the distribution curve for the three GP regions, defined in Fig. 3F, were plotted as a function of hydrogen peroxide concentration. Error bars correspond to SEM. Statistical analysis showed significant differences (p b 0.05) between control and treated macrophages in the mean area of both intermediate and rigid regions.

3.3. Lipid peroxidation induced by H2O2 In order to assess macrophage membrane lipid peroxidation in our experimental conditions: incubation for 24 h in the presence of 0.5 to 2.0 mM H2O2, in which cell viability is above 80%, oxidation of the probe C11-BODIPY 581/591 was determined. It is well known that this fluorescent probe is a fatty acid analogue which incorporates into membranes and allows the evaluation of lipid peroxidation in vivo [43]. Lipid peroxidation increases in a H2O2 concentration depending manner as it is shown in Fig. 6. At the H2O2 concentration range used in our GP study (Fig. 3) a significant increase in lipid peroxidation, as compared to the control cells, is detected. Lipid peroxidation was also measured at 30 min, but no significant changes were detected. 3.4. Effects of free radical initiator AAPH on macrophage lipid peroxidation and membrane fluidity The free radical initiator AAPH was used as a standard for lipid peroxidation. It is known that azo compounds generate free radicals at a constant rate by their spontaneous thermal decomposition and, through a chain mechanism, they induce biochemical modifications in the biological molecules [39]. As can be seen in Fig. 7A, after 30 min incubation of macrophages with AAPH, there was a significant increase of C11-BODIPY581/591 peroxidation as compared to control cells. The azo compound also induced changes in the membrane fluidity distribution (Fig. 7B–C); a significant (p b 0.05) frequency increase in lipid rafts (+0.25 b GP b +0.55) and gel phase domains (GP >+0.55) was obtained, at the expense of intermediate and fluid lipid domains. The results obtained with the azo compound AAPH indicate membrane lipid peroxidation and this is related to the appearance of rigid domains. 4. Discussion In order to validate the technique of two-photon microscopy in our laboratory, we first analyzed giant liposomes of DPPC. The GP images and the corresponding histograms, shown in Fig. 1, as well as the

Fig. 5. Statistical analysis of the size and the number of rigid membrane domains: The mean size, in pixels, and the mean domain number of the raft (+0.25 b GP b +0.55) and gel phase domains (GP > +0.55) were plotted as a function of hydrogen peroxide concentration. Error bars correspond to SEM. Statistical analysis showed significant differences (p b 0.05) between control and treated macrophages in the mean domain number of both raft and gel phase domains.

transition curve in Fig. 2, were in agreement with the results previously obtained for DPPC GUVs by other groups [29,44]. We also measured, by a spectrofluorometric method described by Parasassi et al. [45], the GP values at different temperatures of Laurdan stained DPPC multilamellar liposome suspensions. The mean GP values (0.6 at 30 °C, − 0.2 at 50 °C and 0.2 at 41 °C) matched to those obtained in our laboratory by two-photon microscopy. Our GP distributions for control THP-1 macrophages (Fig. 3) were similar to those obtained by Gaus et al. [46]. Nevertheless, small differences can be observed that could be attributed to the different staining method used and/or to the different temperatures at which the images were acquired. Our Laurdan-staining method has some advantages in preserving the membranes, as it did not include any fixing or pre-treatment of the cells. Concerning the acquisition temperature, our experiments were performed at physiological temperature, 37 °C, whereas other authors perform the experiments at 22 °C. Our GP images have been pseudocolored with an arbitrary palette of colors which may be different from the arbitrary colors chosen in other reported studies. In our study we show that macrophages treated with hydrogen peroxide undergo a concentration dependent significant increase in the frequency of both rigid lipid domains: gel phase and raft domains. It has been shown that lipid rafts are particularly enriched in certain areas of the macrophage plasma membrane like filopodia and cell to cell contacts [35]. Moreover, in the last years several authors have suggested a relevant role of rigid lipid domains in some of the functions of macrophages like motility and phagocytosis [21,46] or during

362

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364

Fig. 6. Macrophage membrane lipid peroxidation induced by H2O2. THP-1 macrophages were stained with the lipid analog C11-BODIPY581/591 for 45 min at 37 °C and treated with H2O2 for 24 h at 37 °C. Bars indicate the mean percentage values of C11-BODIPY oxidation. Error bars correspond to SEM. Statistical analysis showed significant differences between control and treated macrophages: p b 0.05 for concentrations ranging from 0.5 mM H2O2 to 2.0 mM H2O2 and pb 0.001 for 5.0 mM H2O2.

macrophage activation [47,48]. Interestingly, the oxidation-induced frequency increase, described in the present study, is placed in the high GP zone which corresponds to raft/gel phase lipid domains. The implication of ROS in lipid raft formation in the plasma membrane of other types of cells has been previously described by using raft isolation and visualization techniques [49,50]. An increase in the frequency of a given GP population, as observed here for oxidized macrophages, may reflect, as suggested by Gaus et al. [35], an increase in the corresponding mean domain size or an increase in the abundance of this kind of domains. As two-photon microscopy allows visualizing the distribution of lipid domains, an accurate quantification of the size and the number of every type of domains can be obtained from the original experimental data. To perform this analysis, we devised in our laboratory suitable software capable of analyzing every type of lipid domain from the GP images obtained for every single cell. This new tool allows calculating for first time the size and the number of every type of lipid domain. This analysis can be applied directly on the two-photon microscopy experimental data without changing any parameter in the microscope and it might allow comparing the results obtained in different laboratories. After the above mentioned analysis, we clearly demonstrate that both raft and gel phase frequency increase, under oxidizing conditions, are attributable to a significant augment in the number of domains (Fig. 5B), whereas the mean size of these rigid domains remains at the same value of the control cells (Fig. 5A), i.e. approximately 170 nm× 170 nm for raft-like domains and 120 nm× 120 nm for gel phase domains. The increase in the number of raft domains may be interpreted either as a consequence of a de novo condensation of domains with intermediate GP values, or as a coalescence of undetectable preexisting rafts below the resolution of the microscope [33]. Our results exclude clustering of detectable domains, as their size did not augment. Nevertheless, two-photon microscopy, in our experimental conditions, is only able to distinguish domains above 100 nm× 100 nm; therefore smaller domains remain undetectable. Clustering or coalescence could then occur in oxidizing conditions between these small undetectable domains, giving rise to larger detectable domains which would account for the increase in visible domain number. With this hypothesis, the significant decrease observed in the frequency of intermediate GP values (Fig. 3F–G) would correspond to the disappearing of these undetectable rigid domains surrounded by more fluid lipid domains.

Fig. 7. Effects of the azo initiator AAPH on lipid peroxidation and membrane fluidity. THP-1 macrophages were stained with the lipid analog C11-BODIPY581/591 for 45 min at 37 °C. AAPH was added to a final concentration of 0.0 mM (control cells), and of 5.0 mM, for 30 min at 37 °C and lipid peroxidation was evaluated (A). To analyze the effect of the azo compound AAPH on membrane fluidity, macrophages were incubated for 30 min with 0.0 mM (control cells) and 5.0 mM AAPH, and stained with 5 μM Laurdan in serum free media. Two-photon microscopy images were collected and transformed to GP frequency distribution curves. Statistical analysis of the area under the curve of the GP region from +0.25 to +1.0 showed significant differences (p b 0.05) between control and AAPH treated macrophages (B). The GP frequency distribution curve of control macrophages was subtracted from the GP distribution curve of AAPH treated macrophages similarly to the H2O2 treatment. The resulting frequency difference distribution curve is shown (C).

The azo compound effect on macrophage membrane lipid peroxidation (Fig. 7A) was comparable to the effect obtained in the presence of 0.5 to 2.0 mM H2O2 (Fig. 6). Interestingly, changes in membrane fluidity distribution due to the azo compound (Fig. 7B–C) are also comparable to the changes induced by 2 mM H2O2 (Fig. 3G), that is, a similar frequency increase in lipid raft and gel phase domains is obtained. From our data, it seems that there is a direct relationship between lipid peroxidation induced by H2O2 and reduced membrane fluidity in macrophages. A difference is that AAPH reaches lipid peroxidation changes in 30 min, while H2O2 needs 24 h to reach a similar effect, although similar changes are obtained both in lipid peroxidation and membrane fluidity distribution. In this work we have not investigated other mechanisms or signaling pathways that, triggered by the presence of H2O2, might contribute to changes in membrane fluidity.

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364

At the cellular level, oxidant injury elicits a wide spectrum of responses ranging from proliferation to cell death, depending of the cell type, exposure time or dose of the oxidant agent. Great differences exist between the responses of different cell types to a determinate oxidative injury. Recent studies have shown that, depending on H2O2 concentration, endothelial cells can undergo proliferation and survival or apoptosis [49,51]. In our oxidative conditions, for all H2O2 concentrations (0.5–2.0 mM), THP-1 macrophage viability was above 80%; this is consistent with a high resistance of macrophages to oxidative conditions, as compared to lymphocytes (data not shown). Lipid rafts would provide an efficient means to respond to external stimuli, including oxidants which may activate survival or apoptotic signaling pathways [49,52]. Recent data suggests an important role of ceramide, the precursor of important bioactive sphingolipids, in lipid rafts formation [53,54]. Moreover, it has been demonstrated that, in macrophage and other cells, ceramide is overexpressed under oxidative stress conditions [47,55]. There is increasing evidence suggesting that ceramide and derivatives can regulate cell proliferation, inflammation and apoptosis [56]; all these signaling pathways are closely related with oxidative stress conditions and implicate lipid raft mechanisms. Therefore it is not surprising that when cells are placed under oxidative stress, more lipid rafts domains appear within the membrane as a consequence of signaling pathways being activated. In conclusion, in this paper we have clearly demonstrated that oxidative stress modifies the physicochemical properties of the lipid bilayer, that is, membrane becomes more rigid and increase the number of raft domains. This would allow an efficient functional "decision" of the cell that could be, among others, a survival pathway or an apoptotic pathway. In our experiments macrophages show a high viability that would match with a survival pathway. This might allow the macrophage to proliferate, to be activated and to secrete inflammatory cytokines. Acknowledgements We are grateful to Pau Coma for his help in adapting the software to the GP image processing. C.H. was funded with a predoctoral fellowship from the UAB. This work was supported by a grant from the Network of Excellence LSHM-CT-2004-512040 (EMBIC) to P.M.

[13]

[14] [15] [16] [17]

[18]

[19]

[20]

[21]

[22]

[23]

[24] [25]

[26]

[27]

[28]

[29]

[30]

[31]

References [32] [1] J. Kaur, S.N. Sanyal, Alterations in membrane fluidity and dynamics in experimental colon cancer and its chemoprevention by diclofenac, Mol Cell Biochem 341 (2010) 99–108. [2] J.A. Lundbaek, Regulation of membrane protein function by lipid bilayer elasticity —a single molecule technology to measure the bilayer properties experienced by an embedded protein, J Phys Condens Matter 18 (2006) S1305–S1344. [3] T.S. Tillman, M. Cascio, Effects of membrane lipids on ion channel structure and function, Cell Biochem Biophys 38 (2003) 161–190. [4] U. Coskun, K. Simons, Cell membranes: the lipid perspective, Structure 19 (2011) 1543–1548. [5] P.V. Escriba, P.B. Wedegaertner, F.M. Goni, O. Vogler, Lipid–protein interactions in GPCR-associated signaling, Biochim Biophys Acta 1768 (2007) 836–852. [6] U. Coskun, K. Simons, Membrane rafting: from apical sorting to phase segregation, FEBS Lett 584 (2010) 1685–1693. [7] L.J. Pike, The challenge of lipid rafts, J Lipid Res 50 (Suppl.) (2009) S323–S328. [8] K. Jacobson, O.G. Mouritsen, R.G. Anderson, Lipid rafts: at a crossroad between cell biology and physics, Nat Cell Biol 9 (2007) 7–14. [9] A. Catala, Lipid peroxidation of membrane phospholipids generates hydroxyalkenals and oxidized phospholipids active in physiological and/or pathological conditions, Chem Phys Lipids 157 (2009) 1–11. [10] A. Catala, Lipid peroxidation modifies the picture of membranes from the “Fluid Mosaic Model” to the “Lipid Whisker Model”, Biochimie 94 (2012) 101–109. [11] P. Kaplan, M. Doval, Z. Majerova, J. Lehotsky, P. Racay, Iron-induced lipid peroxidation and protein modification in endoplasmic reticulum membranes, Protection by stobadine, Int J Biochem Cell Biol 32 (2000) 539–547. [12] R. Solans, C. Motta, R. Sola, A.E. La Ville, J. Lima, P. Simeon, N. Montella, L. Armadans-Gil, V. Fonollosa, M. Vilardell, Abnormalities of erythrocyte membrane

[33]

[34] [35]

[36]

[37]

[38] [39] [40] [41] [42]

363

fluidity, lipid composition, and lipid peroxidation in systemic sclerosis: evidence of free radical-mediated injury, Arthritis Rheum 43 (2000) 894–900. M. Benderitter, L. Vincent-Genod, J.P. Pouget, P. Voisin, The cell membrane as a biosensor of oxidative stress induced by radiation exposure: a multiparameter investigation, Radiat Res 159 (2003) 471–483. H.W. Wang, J.A. Joyce, Alternative activation of tumor-associated macrophages by IL-4: priming for protumoral functions, Cell Cycle 9 (2010) 4824–4835. A.N. Seneviratne, B. Sivagurunathan, C. Monaco, Toll-like receptors and macrophage activation in atherosclerosis, Clin Chim Acta 413 (2012) 3–14. P. Montuschi, P.J. Barnes, L.J. Roberts II, Isoprostanes: markers and mediators of oxidative stress, Faseb J 18 (2004) 1791–1800. E. Stachowska, M. Baskiewicz-Masiuk, V. Dziedziejko, I. Gutowska, I. Baranowska-Bosiacka, M. Marchlewicz, B. Dolegowska, B. Wiszniewska, B. Machalinski, D. Chlubek, Conjugated linoleic acid increases intracellular ROS synthesis and oxygenation of arachidonic acid in macrophages, Nutrition 24 (2008) 187–199. M.D. Kraaij, S.W. van der Kooij, M.E. Reinders, K. Koekkoek, T.J. Rabelink, C. van Kooten, K.A. Gelderman, Dexamethasone increases ROS production and T cell suppressive capacity by anti-inflammatory macrophages, Mol Immunol 49 (2011) 549–557. L. Fialkow, Y. Wang, G.P. Downey, Reactive oxygen and nitrogen species as signaling molecules regulating neutrophil function, Free Radic Biol Med 42 (2007) 153–164. A. Szoor, J. Szollosi, G. Vereb, Rafts and the battleships of defense: the multifaceted microdomains for positive and negative signals in immune cells, Immunol Lett 130 (2010) 2–12. A. Magenau, C. Benzing, N. Proschogo, A.S. Don, L. Hejazi, D. Karunakaran, W. Jessup, K. Gaus, Phagocytosis of IgG-coated polystyrene beads by macrophages induces and requires high membrane order, Traffic 12 (2011) 1730–1743. M.B. Fessler, J.S. Parks, Intracellular lipid flux and membrane microdomains as organizing principles in inflammatory cell signaling, J Immunol 187 (2011) 1529–1535. D.M. Owen, K. Gaus, A.I. Magee, M. Cebecauer, Dynamic organization of lymphocyte plasma membrane: lessons from advanced imaging methods, Immunology 131 (2010) 1–8. D. Marguet, P.F. Lenne, H. Rigneault, H.T. He, Dynamics in the plasma membrane: how to combine fluidity and order, Embo J 25 (2006) 3446–3457. A.P. Demchenko, Y. Mely, G. Duportail, A.S. Klymchenko, Monitoring biophysical properties of lipid membranes by environment-sensitive fluorescent probes, Biophys J 96 (2009) 3461–3470. T. Parasassi, E. Gratton, W.M. Yu, P. Wilson, M. Levi, Two-photon fluorescence microscopy of laurdan generalized polarization domains in model and natural membranes, Biophys J 72 (1997) 2413–2429. L.A. Bagatolli, To see or not to see: lateral organization of biological membranes and fluorescence microscopy, Biochim Biophys Acta 1758 (2006) 1541–1556. L.A. Bagatolli, E. Gratton, Two photon fluorescence microscopy of coexisting lipid domains in giant unilamellar vesicles of binary phospholipid mixtures, Biophys J 78 (2000) 290–305. L.A. Bagatolli, E. Gratton, Two-photon fluorescence microscopy observation of shape changes at the phase transition in phospholipid giant unilamellar vesicles, Biophys J 77 (1999) 2090–2101. L.A. Bagatolli, S.A. Sanchez, T. Hazlett, E. Gratton, Giant vesicles, Laurdan, and two‐ photon fluorescence microscopy: evidence of lipid lateral separation in bilayers, Methods Enzymol 360 (2003) 481–500. K. Gaus, S. Le Lay, N. Balasubramanian, M.A. Schwartz, Integrin-mediated adhesion regulates membrane order, J Cell Biol 174 (2006) 725–734. W. Yu, P.T. So, T. French, E. Gratton, Fluorescence generalized polarization of cell membranes: a two-photon scanning microscopy approach, Biophys J 70 (1996) 626–636. K. Gaus, E. Chklovskaia, B. Fazekas de St Groth, W. Jessup, T. Harder, Condensation of the plasma membrane at the site of T lymphocyte activation, J Cell Biol 171 (2005) 121–131. K. Gaus, T. Zech, T. Harder, Visualizing membrane microdomains by Laurdan 2-photon microscopy, Mol Membr Biol 23 (2006) 41–48. K. Gaus, E. Gratton, E.P. Kable, A.S. Jones, I. Gelissen, L. Kritharides, W. Jessup, Visualizing lipid structure and raft domains in living cells with two-photon microscopy, Proc Natl Acad Sci U S A 100 (2003) 15554–15559. E.H. Pap, G.P. Drummen, V.J. Winter, T.W. Kooij, P. Rijken, K.W. Wirtz, J.A. Op den Kamp, W.J. Hage, J.A. Post, Ratio-fluorescence microscopy of lipid oxidation in living cells using C11-BODIPY(581/591), FEBS Lett 453 (1999) 278–282. G.P. Drummen, L.C. van Liebergen, J.A. Op den Kamp, J.A. Post, C11-BODIPY(581/591), an oxidation-sensitive fluorescent lipid peroxidation probe: (micro)spectroscopic characterization and validation of methodology, Free Radic Biol Med 33 (2002) 473–490. E.H. Pap, G.P. Drummen, J.A. Post, P.J. Rijken, K.W. Wirtz, Fluorescent fatty acid to monitor reactive oxygen in single cells, Methods Enzymol 319 (2000) 603–612. E. Niki, Free radical initiators as source of water- or lipid-soluble peroxyl radicals, Methods Enzymol 186 (1990) 100–108. E. Schnitzer, I. Pinchuk, D. Lichtenberg, Peroxidation of liposomal lipids, Eur Biophys J 36 (2007) 499–515. C. Dietrich, L.A. Bagatolli, Z.N. Volovyk, N.L. Thompson, M. Levi, K. Jacobson, E. Gratton, Lipid rafts reconstituted in model membranes, Biophys J 80 (2001) 1417–1428. L.A. Bagatolli, T. Parasassi, E. Gratton, Giant phospholipid vesicles: comparison among the whole lipid sample characteristics using different preparation

364

[43]

[44]

[45]

[46]

[47] [48]

C. de la Haba et al. / Biochimica et Biophysica Acta 1828 (2013) 357–364 methods: a two photon fluorescence microscopy study, Chem Phys Lipids 105 (2000) 135–147. A.E. Dominguez-Rebolledo, F. Martinez-Pastor, M.R. Fernandez-Santos, E. del Olmo, A. Bisbal, J.L. Ros-Santaella, J.J. Garde, Comparison of the TBARS assay and BODIPY C11 probes for assessing lipid peroxidation in red deer spermatozoa, Reprod Domest Anim 45 e360-368. S.A. Sanchez, M.A. Tricerri, G. Gunther, E. Gratton, Laurdan generalized polarization: from cuvette to microscope, in: A.a.D. Méndez-Vilas, J. (Eds.), Modern Research and Educational Topics in Microscopy, 2007, pp. T. Parasassi, G. De Stasio, A. d'Ubaldo, E. Gratton, Phase fluctuation in phospholipid membranes revealed by Laurdan fluorescence, Biophys J 57 (1990) 1179–1186. K. Gaus, M. Rodriguez, K.R. Ruberu, I. Gelissen, T.M. Sloane, L. Kritharides, W. Jessup, Domain-specific lipid distribution in macrophage plasma membranes, J Lipid Res 46 (2005) 1526–1538. J. Cuschieri, R.V. Maier, Oxidative stress, lipid rafts, and macrophage reprogramming, Antioxid Redox Signal 9 (2007) 1485–1497. S. Olsson, R. Sundler, The role of lipid rafts in LPS-induced signaling in a macrophage cell line, Mol Immunol 43 (2006) 607–612.

[49] B. Yang, T.N. Oo, V. Rizzo, Lipid rafts mediate H2O2 prosurvival effects in cultured endothelial cells, Faseb J 20 (2006) 1501–1503. [50] S.P. Lu, M.H. Lin Feng, H.L. Huang, Y.C. Huang, W.I. Tsou, M.Z. Lai, Reactive oxygen species promote raft formation in T lymphocytes, Free Radic Biol Med 42 (2007) 936–944. [51] J.L. Martindale, N.J. Holbrook, Cellular response to oxidative stress: signaling for suicide and survival, J Cell Physiol 192 (2002) 1–15. [52] K.S. George, S. Wu, Lipid raft: A floating island of death or survival, Toxicol Appl Pharmacol (2011). [53] A.E. Cremesti, F. Goñi, R. Kolesnick, Role of sphingomyelinase and ceramide in modulating rafts: do biophysical properties determinate biologic outcome? FEBS Lett 531 (2002) 47–53. [54] I. Johnston, L.J. Johnston, Ceramide promotes restructuring of model raft membranes, Langmuir 22 (2006) 11284–11289. [55] D. Martin, M. Salinas, N. Fujita, T. Tsuruo, A. Cuadrado, Ceramide and reactive oxygen species generated by H2O2 induce caspase-3-independent degradation of Akt/protein kinase B, J Biol Chem 277 (2002) 42943–42952. [56] L. Arana, P. Gangoiti, A. Ouro, M. Trueba, A. Gomez-Munoz, Ceramide and ceramide 1-phosphate in health and disease, Lipids Health Dis 9 (2010) 1–12.