Biochimica et Biophysica Acta 1808 (2011) 2965–2972

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a m e m

Photo-dynamic induction of oxidative stress within cholesterol-containing membranes: Shape transitions and permeabilization Rachid Kerdous a, Julien Heuvingh b, Stéphanie Bonneau a,⁎ a b

Université Pierre et Marie Curie, ANBioΦ, FRE3207 CNRS, 4 place Jussieu, Paris, France Université Paris Diderot, PMMH, UMR7636 CNRS/ ESPCI/Université Pierre et Marie Curie, 10 rue Vauquelin, Paris, France

a r t i c l e

i n f o

Article history: Received 26 March 2011 Received in revised form 15 July 2011 Accepted 2 August 2011 Available online 8 August 2011 Keywords: Photosensitizer Cholesterol Curvature Permeabilization Photochemical Internalization Liposome

a b s t r a c t Photochemical internalization is a drug delivery technology employing a photo-destabilization of the endosomes and the photo-controlled release of endocyted macromolecules into the cytosol. This effect is based on the ability of some photosensitizers to interact with endosomal membranes and to photo-induce damages leading to its breakdown. The permeabilization efficiency is not quantitatively related to the importance of the damages, but to their asymmetric repartition within the leaflets. Using unilamellar vesicles and a chlorin, we studied the effect of the membrane's cholesterol content on its photo-permeabilization. First, the affinity of the chlorin for membranes was studied. Then, we asymmetrically oxidized the membranes. For DOPC/CHOL GUVs, we observed different shape transitions, in accordance with an increase followed by a decrease of the membrane effective curvature. These modifications are delayed by the cholesterol. Finally, the photo-permeabilization of GUVs occurs, corresponding to a pore formation due to the membrane tension, resulting from vesicles buddings. Cholesterol-rich GUVs permeabilization occurs after a lag, and is less important. These results are interpreted regarding both (i) the cholesterol-induced tightening of the lipids, its consequences on physical parameters of the membrane and on oxidation rate and (ii) the suggested ability of cholesterol to flip rapidly and then to relax the differential density-based stress accumulated during membrane bending. © 2011 Elsevier B.V. All rights reserved.

1. Introduction The oxidation of unsaturated lipids plays a central role, both in biological functions and in pathogenesis of many diseases [1–4]. Oxidation alters the properties of biomembranes, including their fluidity and permeability [5,6]. These changes, in turn, affect the activity of membrane proteins, including membrane-bound enzymes, receptors and transport proteins [7–9]. The photochemical induction of oxidation is an effective and controlled way of inducing oxidation processes [10]. It is based on the ability of certain molecules, the photosensitizers, to generate Reactive Oxygen Species (ROS) upon light irradiation. The ROS react with surrounding biomolecules, generating photo-damages. This property, together with the preferential retention of certain photosensitizers by tumors as compared to normal surrounding tissues, has found an application in an anti-tumoral therapy, the Photodynamic Therapy (PDT) [11]. Indeed, a huge level of such light-induced damages leads to the death of the targeted cells. More recently, photosensitizer-induced

⁎ Corresponding author at: Laboratoire ANBioPhy-FRE3207, Université Pierre et Marie Curie, Case courrier 138, 4 place Jussieu, 75005 Paris, France. Tel.: + 33 1 40 79 36 97; fax: + 33 1 40 79 37 05. E-mail address: [email protected] (S. Bonneau). 0005-2736/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2011.08.002

lipid oxidation has been used to improve the delivery of macromolecular therapeutic agents to their intracellular targets by an approach called Photochemical Internalization (PCI) [12]. After the uptake by endocytosis, the degradation of the macromolecules in lysosomes is greatly reduced by the photodynamic destabilization of the endocytic vesicles membrane, increasing their biological activity. This approach demonstrated a huge and controlled increase of the drug efficacy, both in vitro and in vivo [13,14]. Because of their very short lifetime [15], photo-induced ROS have a limited diffusion length (10–20 nm), which imply a close association of photosensitizers with the targeted location [15–18]. For tetrapyrrole photosensitizers, the ability to cross membranes is governed by the charge of their lateral chains [19–23]. The chlorin e6 (Ce6), a second generation photosensitizer, is not able to cross the biological membranes [24,25]. Consequently, it interacts only with the monolayer in contact with the photosensitizer solution, i.e. labels the membrane asymmetrically. Recently, we thus labeled model membranes, Giant Unilamellar Vesicles (GUVs) composed of dioleoylphosphatidylcholine (DOPC), which is an unsaturated lipid. Under light-induced oxidation, we observed shape transitions and permeabilization of the GUVs. Our results demonstrated that, under light-induced asymmetric modifications, the curvature of the membrane is strongly modified. The budding induced by the curvature is responsible for a stretching of the membrane up to its lysis tension and its subsequent permeabilization. The key role of the asymmetry in this

2966

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

permeabilization of the membrane, related to its bending stress, was highlighted [26]. Biophysical studies of membranes, including experimental observations of vesicle shape transitions, led to the area difference elasticity (ADE) model [27]. This model successfully predicts the shape transformations of a vesicle, when its curvature or area to volume ratio is modified. The effective spontaneous curvature of the membrane used in this model comes from two distinct contributions: one due to the preferred packing of individual lipids and the other to the mismatch of the area of a monolayer compared to the other. When present in the outer leaflets of a membrane, lipids with cone shape will create a mismatch between the area of the hydrophobic and hydrophilic regions in the leaflet, inducing a positive spontaneous curvature. Likewise, an increase in the preferred area of lipids in the outer layer will induce an area difference between the two leaflets and increase the effective spontaneous curvature of the membrane. In pure phospholipidic membranes, the diffusion of lipids from one leaflet to the other (flip-flop) is very slow, and cannot equilibrate the area difference between leaflets. This is however not the case for a biological membrane with a more complex composition. In particular, a variety of amphiphilic molecules, such as cholesterol (0.1 to 0.4 of molar fraction [28,29]), are able to flip rapidly from a leaflet to the other [30–32]. This rapid flip-flop has been suggested to be an efficient pathway for the relaxation of the stress due to the leaflet area difference accumulated during membrane bending [33]. Moreover, the cholesterol is known to be an agent promoting membrane order. Most of the physical and functional membrane properties depend on cholesterol and its interaction with the other lipid species [28,29,34,35]. For dioleoylphosphatidylcholine (DOPC), this interaction can be described by the “umbrella model” and increases the membrane cohesion and mechanical stiffness [36]. It rigidifies the bilayers, reducing the free volume available for diffusion and decreasing the rate of lateral diffusion. The solubility limit of cholesterol in DOPC bilayers has been measured to be 0.67, corresponding to a regular distribution of the cholesterol molecules in the lipid bilayer up to solubility limit [37– 39]. The cholesterol-induced tightening of the packing of lipids in the bilayers through its effect on the fluidity, viscosity and lateral diffusion, consequently decrease the membrane permeability [40], and also affects the rate of peroxidation of liposomal lipids. It is however impossible to predict the effect of membrane fluidity on the rate of its peroxidation [41]. Recently, GUV composed of DOPC/CHOL in 1:1 ratio have been reported to fluctuate following oxidative stress [42]. Here, we photochemicaly induce lipid oxidation in DOPC/CHOL-GUVs to address the effect of cholesterol content in the dynamics of a photooxidative membrane and on its photo-permeabilization. The photosensitizer used, the Ce6, allows an asymmetric targeting of the external bilayer of the membrane (called “targeted leaflet”) and a fine control of the location of the oxidation [26]. Our results are discussed in relation with the technology of photochemical internalization, PCI. 2. Materials and Methods 2.1. Chemicals All chemicals were purchased from Sigma (USA), except dioleoylphosphatidylcholine (DOPC), dipalmitoylphosphocholine (DPPC) and cholesterol (CHOL) from Avanti Polar Lipids (USA), and chlorin e6 from Frontier Scientific (USA). Chlorin stock solution (5 mM) was prepared in ethanol and kept at −18 °C. The experimental Ce6 aqueous solutions were prepared, used without delay and handled in the dark. The osmolarity of the solutions was checked with an osmometer (Löser Messtechnik, Germany).

chloroform. Large Unilamellar Vesicles (LUVs) were prepared by extrusion method. After evaporation of chloroform, lipids were dispersed in phosphate buffer pH 7.4 by vortexing. The liposome suspension was extruded 8–10 times through a stack of two polycarbonate membrane filters (Poretics, Livermore, CA) with pores of 50, 100 or 200 nm using an extruder device (Avanti Polar Lipids, USA). Giant Unilamellar Vesicles (GUVs) with an average diameter of 10–20 μm were formed by the electroformation method [43] as reported previously [26]. Lipids mixtures in chloroform were deposited on ITO-covered glass plates. A chamber was made from two such glass plates and a Teflon spacer of 4 mm and the solvent was dried in vacuum. The chamber was then filled with a solution of 300 mM sucrose and an AC field was applied between the plates for 4 h. For experiments, the GUVs were mixed with a 300 mM glucose solution. The density difference between sucrose and glucose caused the GUVs to sediment to the bottom of the chamber. The difference in optical index between sucrose inside and glucose outside the vesicle allowed phase contrast microscopy observation. For DPPC containing GUVs, the preparation and experiments were performed at 60 °C, up to the transition temperature of the lipid mixture. 2.3. Steady-state interaction For the steady-state study of the interaction of the Ce6 with LUV, fluorescence spectra were measured with an Aminco Bowman Series 2 spectrofluorimeter (Edison, NJ, USA). Recording was generally started 2 min after the preparation of the solutions under study. Liposome solutions were prepared at different concentrations of lipids. 10 μl of 10 μM Ce6 solution were added to 2 ml of vesicle preparation and the fluorescence spectra were recorded. In order to correct the spectra for small differences in Ce6 concentration arising from experimental inaccuracy, 20 μl of Triton-X100 were added after measurement leading to disruption of vesicles and solubilization of all Ce6 in the Triton micelles. The spectra were normalized accordingly. Data thus obtained were treated as described elsewhere [24,44,45]. The global binding constant, KB, was derived from changes in the fluorescence signal at a wavelength corresponding to the maximum of fluorescence emission of Ce6 incorporated into the membrane. We used the previously derived relationship: F = F0 +

ðF∞ −F0 Þ · KB · ½lip! 1 + KB · ½lip!

ð1Þ

where F0, F∞ and F are respectively the fluorescence intensities corresponding to zero, total and intermediate incorporation of Ce6 into vesicles. Lipids being in large excess, the saturation of the bilayer is far to be reached and it can be assumed that Ce6 binding does not affect the properties of the model membrane. Then, regardless of the number of Ce6 molecules incorporated into a vesicle, [lip] was assumed to be equivalent to the total lipid concentration added, i.e. the sum of the cholesterol and DOPC concentrations. 2.4. Asymmetric labeling of GUVs Giant Vesicles were asymmetrically labeled with Ce6 into the outer leaflet as reported in previous work [26]. The Ce6 concentration used was 25 μM. 2.5. Observation and illumination

2.2. Vesicles formation Lipids mixtures with various contents (DOPC/CHOL 10:0, 9:1, 8:2, 7:3 and 5:5 mol/mol or DPPC/CHOL 10:0, 9:1, 8:2) were made in

GUVs were observed under an inverted microscope. The instrumental set-up was based on a Nikon Eclipse TE 300 DV inverted microscope equipped with a high numerical aperture phase oil

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

objective (CFI Plan apochromat DM 60 n.a.: 1.4, Nikon, France). Illumination was provided by a 120 W Metal halide lamps with a 465– 495 nm bandpass filter. Neutral density filters (NDx8) were used to reduce illumination level. Image acquisition (100 ms integration time) was performed with a CCD camera (CoolSNAP HQ², Roper Scientific, France). Data acquisition was performed with Metamorph software, supplied by Universal Imaging Corporation (Roper Scientific, France). Images were analyzed using Image J (NIH, USA). For each experimental condition, the images of a hundred vesicles were analyzed. When photo-oxidizing the membrane, the illumination was kept constant from the starting time. As compared to ref [26], the illumination was less intense, which allowed the observation of intermediate steps in vesicle photo-damage.

3. Results 3.1. Photosensitizer-membrane interaction The fluorescence emission spectra of Ce6 (5× 10−8 M) in solution and in presence of various lipid concentration ([lip] = 0 to 1.2× 10−3 M) are given in Fig. 1 (inset). The excitation wavelength was 410 nm. Upon liposomes addition the intensity of the fluorescence emission is increased significantly and is shifted from 660 to 668 nm. These spectral changes are characteristic of the transfer of the chlorin from an aqueous to a hydrophobic environment [24]. The plot of the fluorescence intensity at 668 nm versus the lipid concentration shows a saturation shape (see Fig.1). The binding constant value, KB, was determined by fitting the experimental data with Eq. (1). KB increases by more than a factor 2.5 for DOPC membrane from 0% to 50% of cholesterol content (Table 1). Such an increase of the affinity is consistent with the data of the literature on photosensitizers-membranes interactions [46]. For DPPC and DPPC/CHOL membranes, the KB values are of same order of magnitude than for DOPC and DOPC/CHOL liposomes, and the same effect of cholesterol was observed (Table 2). In order to estimate the influence of membrane curvature on the affinity values experimentally measured, the interaction of Ce6 with lipid unilamellar vesicles of various sizes has been studied: the extrusion was performed using polycarbonate membrane filters with pores of 50, 100 or 200 nm. In accordance with the literature [44],

Fig. 1. Interaction of Ce6 with lipid large unilamellar vesicles. Typical evolution of the fluorescence emission of Ce6 upon incorporation into LUV (here, LUVs are made with pure DOPC, extruded through pores of 100 nm, and [DOPC] is 0 to 1.2 × 10−3 M); Ce6 concentration is 5 × 10−8 M. The excitation wavelength is set at 410 nm and the fluorescence intensity is recorded at 668 nm. Eq. (1) is used to fit experimental data and yields the binding constant KB. The inset shows the red shift of Ce6 fluorescence peak upon incorporation into the membrane.

2967

Table 1 Ce6 binding constants to DOPC/CHOL vesicles. DOPC:Chol

10:0

9:1

8:2

7:3

5:5

KB (M−1)

(6.0 ± 0.7) × 103

(6.5 ± 0.9) × 103

(7.5 ± 0.4) × 103

(1.1 ± 0.1) × 104

(1.6 ± 0.1) × 104

equilibrium constants are almost independent of the vesicle size when expressed in terms of phospholipid concentration (see Fig. S1). 3.2. Photo-induced shape transitions As the Ce6 do not flip-flop at the timescale involved in our experiments [24,25], only the external leaflet is labeled. Under light irradiation, the Ce6 molecules generate ROS and, after a few to two hundred second of latency, the Ce6-labeled DOPC containing vesicles exhibit shape transitions. The average sequence of events is: fluctuations during tens of seconds followed by transitions from oblate to prolate ellipsoids (N95% of the vesicles), external buddings (detected for 90% of the vesicles) and, tens of seconds second later, internal buddings of vesicles (detected for 92% of the vesicles) (Fig. 2, top). Vesicles with no Ce6 and not illuminated ones showed none of these transitions. These results are in good accordance with first an increased and then a decreased spontaneous curvature of the external leaflet (see Discussion). The average number of buds per vesicle is not influenced by the cholesterol content and is 3.46 ± 0.40. However, the above described sequence is clearly slowed by increasing the cholesterol content of the membrane. The timings of the external buddings and internal ones are given in Fig. 3 for each percentage of cholesterol. The moment of the buddings is delayed when the membrane is composed of more than 30% of cholesterol, indicating the influence of the cholesterol on the photo-induced shape transitions. 3.3. Relation between physical responses of the GUVs to photo-oxidation and chemical scenarios A brief summary of the most commonly postulated sequence of reactions triggered by irradiation of photosensitizers in presence of oxygen within a lipid membrane is: under light activation, the photosensitive molecule produces free radicals (Type I photooxidation) and/or singlet oxygen (Type II photo-oxidation). Within the core of the membrane, the C―C double bonds have been shown to be privileged site of these ROS reactions. Both DOPC and cholesterol are potentially chemical targets for the photo-oxidative reactions. In order to clarify the role of the oxidative products of these two compounds affecting shape transitions, we replaced DOPC by the saturated phospholipid DPPC: experiments were performed on DPPC/CHOLGUVs 10:0, 9:1 and 8:2 (Fig. 4, middle). DPPC is insensitive to photosensitization, and the observed responses will then strictly be induced by the cholesterol products. Sequences presented in the middle of Fig. 4 are typical: no effect was detected, except gentle fluctuation and area increase of the membrane around 500 s for the most important cholesterol content (20% CHOL). This area increase, appearing very late during the experiment, is attributed to the cholesterol oxidation. No permeabilization was detected over 60 min of irradiation. As aforementioned, photo-oxidation may occur through type I and/or type II reaction. To attempt to prove the oxidative nature of the

Table 2 Ce6 binding constants to DPPC/CHOL vesicles. DPPC:Chol

10:0

9:1

8:2

KB (M−1)

(4.6 ± 0.8) × 103

(7.7 ± 1.2) × 103

(1.2 ± 0.2) × 104

2968

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

the phase contrast constant. However, after a long irradiation time (200 s), modest alterations of the state of the DOPC/CHOL membranes became apparent for a significant proportion of the GUVs, suggesting that these CHOL-rich vesicles are more susceptible to the photooxidation, in particular of type I, than the others. 3.4. Photo-controlled permeabilization The ability of Ce6 to permeabilize lipid membrane was estimated from the exchange, by diffusion, of the sucrose entrapped inside the GUV and the glucose in the surrounding medium, leading to the fading of the contrast between inside and outside medium [47]. Typically, after the shape transitions of the GUV, the contrast fades away, according with an exponential decrease, until there is no more difference between the vesicle and the surrounding solution (Fig. 2, bottom). The sudden start of permeabilization suggests that the opening of a pore is responsible for permeabilization, in good accordance with the pore opening scenario commonly suggested for the photo-induced membrane permeabilization [26,48,49], the processes begin when the alteration of the membrane are quantitatively sufficient to involve an hydrophilic mismatch (pre-pore) within the membrane. Depending on the hydrophobicity/hydrophilicity balance of the buffer solute, the diffusion starts at T0 and is, at this time, described by: dcin PA c ; =− V in dt

Fig. 2. Irradiation of the Ce6-labeled DOPC containing vesicles. (top) Video sequence showing the typical morphological changes of Ce6-labeled DOPC-GUV under light irradiation (top: phase contrast microscopy; bottom: fluorescence of the Ce6). The morphology transitions observed denote first an increase and then a decrease of the membrane's equivalent curvature (see text). Irradiation starts at time t = 0. Images shown at 5, 15, 50, 100 and 240 s. Scale bar = 10 μm. (bottom) Typical evolution of the contrast between the inside and outside of a photosensitized GUV. The decay of the contrast corresponds to the equilibration between the inside and the outside medium. The experimental points are fitted by a decreasing exponential according to Eq. (3), from which a starting and a characteristic time of permeabilization are extracted.

light-induced membrane responses and to discriminate which chemical mechanism was dominant, we used sodium azide as quencher of singlet oxygen. DOPC, DOPC/CHOL, DPPC and DPPC/CHOL vesicles were formed and labeled with Ce6 in aqueous solution containing 50 mM of sodium azide. As depicted in the bottom of Fig. 4, exposure to light has no or minor effect on the membranes. More precisely, no responses were registered, the GUVs remain stable and

where cin is the instant concentration of sucrose inside the vesicle, P is the membrane permeability, A the membrane area, V the volume of the vesicle and t is time. Because the external volume is huge in relation to the total internal medium (i.e. the volume entrapped in the vesicles), we assumed the external sucrose concentration to remain constant, equal to zero. The contrast experimental data were fitted accordingly with:  € t−T0 ; cin = c0 exp − τ

Starting time of the buddings, T (s)

180

T outward

160

T inward

140 120 100 80 60 40 20 0 0

10

20

30

40

50

Mol% Cholesterol Fig. 3. Effect of the cholesterol content on the photo-induced sequence showing the typical morphological changes. Starting time of the inward buddings (squares) and of the outward ones (diamonds). Data correspond to the measures on 100–120 vesicles for each cholesterol content.

ð3Þ

where c0 is the initial concentration of sucrose inside the vesicle, T0 the initial time of permeabilization and τ the characteristic time of the diffusion. For each experiment, starting and characteristic times were extracted, according with Eq. 3. The mono-exponential fit is in very good agreement with our experimental data and, according with Eq. (2), the permeability has been deduced from τ according with: P=

200

ð2Þ

V τ×A

ð4Þ

The starting times and the photo-induced permeability have been plotted as functions of the vesicle radius and show no radiusdependence, for all the DOPC/CHOL compositions. However, as expected, increasing the cholesterol content of the membrane increases the delay before its permeation and decreases its photoinduced permeabilization (Fig. 5). The facts that permeabilization occurs suddenly, after 100–300 s of exposition, and is characterized by an exponential decay in sucrose concentration, are both in good agreement with the opening of a pore scenario. If the pore is unique, its diameter can be inferred from the permeability of the membrane [26]. The diameters of the photoopening pores have been calculated and are between 1.2 nm and 21.2 nm. Moreover, for each cholesterol content, the distribution function of the permeability (and of the corresponding pore diameter) has been determined (Fig. 5). As expected, the probability density function of the membrane permeability obeys to a log-normal distribution [50].

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

2969

Fig. 4. Physical responses of the Ce6-lebelled membranes to irradiation and chemical scenarios. (top) Irradiation of Ce6-labeled DOPC-GUV, showing the typical sequence of events under photosensitization. As mentioned in the text, this sequence is delay by the presence of cholesterol within the membrane. Scale bar = 10 μm. (middle) Irradiation of Ce6 labeled DPPC containing vesicles. DPPC is insensitive to light exposure and no effect was detected, except gentle fluctuation and area increase of the membrane around 500 s for 20% cholesterol containing membrane. No permeabilization was detected. Scale bar = 10 μm. (bottom) Irradiation of Ce6-labeled GUVs containing DOPC or DPPC, in presence of sodium azide. Morphological transition, membrane area variations as well as permeabilization are suppressed, except brief fluctuation around 200 s for DOPC/CHOL 8/2 GUV. Scale bar = 10 μm.

To verify that the permeabilization was mainly due to the oxidation of the lipid unsaturation, we compared the photosensitization effect on DOPC and DPPC. As shown in Fig. 4, contrary to the DOPC-GUVs, the DPPC-GUVs did not permeabilize during our experiments (extended experiments have been performed during 1 h). 4. Discussion 4.1. Photo-oxidation products As a photosensitizer, the Ce6 interacts very efficiently with light to produce reactive species (singlet oxygen and radicals) from its longlife triplet state. Its quantum yield for 1O2 production is important in hydrophobic environments (around 0.65 in ethanol [51]). The non-enzymatic oxidative products of cholesterol are 7ketocholesterol and 7β-hydroxycholesterol, the latter being the major photo-oxidation product [42]. Type I oxidation generates directly 7α- an 7β-hydroxycholesterol, whereas in type II, peroxida-

tion of cholesterol by singlet oxygen produces primarily a C-5 oxygenated molecule which may be later rearranged giving 7αhydroperoxycholesterol. In both cases, 7α-hydroperoxycholesterol is progressively epimerized into 7β-hydroperoxycholesterol. Such products induce changes in the physicochemical properties of membrane's leaflets [52]), in particular the molecular packing, as reported for a Langmuir monolayer with oxidized cholesterol [53]. However, the flip-flop of sterols is very rapid and, as previously mentioned, would be an efficient pathway for the relaxation of the curvature due to area difference between the leaflets [33]. In unsaturated lipids, the primary products of the photo-induced oxidation are peroxides [54]. The lipid peroxidation is a radical chain reaction leading to the formation of intermediate hydroperoxide. Remaining in the hydrocarbon region of the membrane bilayer, hydroperoxyl group should drastically change the membrane architecture by increasing the cross-section area of its lipid tails. The packing parameter of the peroxide lipid would be above one and the relaxed area of the corresponding leaflet would be increased. Moreover, because of

2970

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

350

brane, measured via the affinity constant. According with Eq. (1) and Refs. [24,44,45], concentration of Ce6 in the membrane is:

300

½Ce6!b =

200 150 100

0.4 0.2 0

50

0

100 200 T0 (s)

300

0

# vesicles

Permeability (m.s-1)

2.5E-08 2E-08

0.4 0.2 0 1

1.5E-08

3 5 Log (Px109)

1E-08 5E-09 0 0

10

20

30

40

50

60

Mol% Cholesterol Fig. 5. Effect of the cholesterol content on photo-induced permeabilization of the DOPCmembrane. Starting time (T0) of permeabilization and photo-induced permeability for 100–120 vesicles for each cholesterol content. Typical distribution functions are given (insets).

the high unstability of those primary products in presence of any trace of transition metal, in our microscopy experiments they will subsequently decay in secondary products. The photo-oxidation of the DOPC results in the cleavage of the carbon chain near the initial position of the double bound, so-called cleaved-lipids [55,56], inducing an important modification of the leaflet geometry as compared to DOPC. They correspond to a packing parameter below one and to a decrease of the relaxed area in the leaflet in which they are present. Additionally, they are known to favor pore formation [57]. Here, it should be noted that the suggested sequence of events (an increased and a subsequent decrease of the relaxed area in the targeted leaflet) is in very good accordance with the typical shape transition sequence that we observed (Fig. 2, top). The presence of hydroperoxide in the outer leaflet would increase the curvature of the membrane, inducing oblate to prolate transition and external budding, while the subsequent formation of cleaved lipids in the outer lipid layer would decrease the membrane curvature, inducing internal budding. The increase in the membrane area due to the presence of peroxide [58] will favor both the transitions to external and internal budding [59], whereas the reduction of membrane area due to the cleavage of lipids will bring the vesicle shape closer to a sphere, favoring its tension and eventual lysis and permeabilization [26]. These phenomena can then be attributed to the oxidation of the DOPC, even if the cholesterol peroxidation may improve the phase of increasing area of the membrane. As a matter of fact, with DPPC, neither shape transitions nor permeabilization of the membrane were observed, except only for Chol-rich GUVs (containing more than 20% of cholesterol) where an increase of area was indeed detected after 500 s of irradiation.

KB · ½lip! ½Ce6! 1 + KB · ½lip!

ð5Þ

where [Ce6] is the total Ce6 concentration. In our experiments, the typical lipid concentration is ~10−5 to 10−6 M, and thus KB.[Lip] b b 1 The concentration of bound Ce6 is therefore proportional to KB and the permeabilization effect has been normalized accordingly. The relative photo-induced permeability is given in Fig. 6. As expected, because the cholesterol induces physical changes of the membrane (in particular its fluidity, viscosity packing parameters and the lateral diffusion of lipids), it modifies the relative permeabilization-efficacy. Indeed, the fluidity of the bilayers affects the rate of peroxidation [41]: the presence of cholesterol slows down the initial rate of peroxidation but enhances its maximal rate. Moreover, cholesterol and its oxidation products within the membrane involve an increase of its lysis tension [35]. In accordance with this complex sum of processes, our results (Fig. 6) demonstrate a significant and non-linear decrease of the relative permeabilization efficacy when the cholesterol content increases. Previously, we established a model linking the membrane tension and its permeabilization [26]. We demonstrated that, under lightinduced asymmetric oxidation, the curvature of the targeted-leaflet is strongly decreased, inducing buds of micron and sub-micron size and hence the stretching of the membrane up to its lysis tension and its subsequent permeabilization. In giant vesicles containing cholesterol, we observe an increased delay for the shape transitions leading to budding. Both the tightening of the packing of lipids, its consequences on oxidation rate and the flip-flop of cholesterol are moderating the photo-induced asymmetry between the two leaflets and reducing the subsequent spontaneous curvature of the membrane induced by the photo-oxidation of lipids. The increase in general membrane area should however not be affected by the presence of cholesterol, and will still favor the formation of internal and external budding. We also observed a significant delay in the lysis and permeation of the cholesterol-rich membrane. This increased delay can be interpreted as a reduced oxidation due to packing, the relaxation of asymmetry by the flip-flop of cholesterol and the increased lysis tension necessary to open a pore in cholesterol rich membrane. 4.3. Cholesterol and PCI effect The photo-chemical internalization of macromolecules in cells is an efficient method to increase their cellular activity, both in vitro and in vivo [13,14]. It is based on the photo-induced permeabilization of the membranes of endosomes. Indeed, the uptake of most of the 1.0

Relative permeability

# vesicles

T0 (s)

250

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0

4.2. Relative photo-permeabilization efficacy and shape transitions The oxidation is photo-chemically initiated via the Ce6, and the permeabilization is proportional to its concentration in the mem-

10

20

30

40

50

60

Mol% Cholesterol Fig. 6. Ce6-membrane affinity and photo-induced permeabilization: Effect of cholesterol content within DOPC-GUVs. Relative photo-induced permeability, normalized by the Ce6 labeling of the membrane.

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

macromolecules by cells occurs by endocytosis, leading to their sequestration in the endosomal compartment. The PCI consists to photo-induce the release of the macromolecules from this compartment to improve their capability to interact with their cellular targets. This reduces strongly their degradation in lysosomes and then increases their biological activity. Furthermore, it makes possible to photo-control this activity. The delay between the uptake and the light-irradiation is of course of major importance, because it defines the compartment where the release occurs: a too early irradiation will be limiting for the uptake duration, but a too late one will trigger a release after the beginning of the lysosomal degradation of the compound. In this context, the role of the cholesterol content of the targeted-membrane on the PCI effect has to be taken into consideration. Plasma membranes are rich in cholesterol (cholesterol/phospholipids ratio around one) and resist to mechanical stress; the early endosomes are similar to plasma membranes, but on maturation to late endosomes there is a decrease of the cholesterol content (cholesterol/phospholipids ratio around 0.5) [29]. Their response to PCI would then be more important, and such parameter, together with the integrity of the macromolecule of interest through its endosomal pathway, has to be considered to determine the lightirradiation moment. 5. Conclusion The asymmetrical shape transitions observed in photosensitized GUVs reveal changes in their membrane spontaneous curvature. Their subsequent photo-induced permeabilization was also characterized. These phenomena are strongly influenced by the cholesterol content of the membrane. This point has been correlated with the tightening of the lipids, its consequences on physical parameters of the membrane and on oxidation rate and with the suggested ability of cholesterol to flip rapidly and then to relax the differential densitybased stress accumulated during membrane bending. These findings might shed a new light on some membrane permeation phenomenon involved in biomedicine photodynamic approaches, in particular photochemical internalization (PCI) and in cell oxidative stress. Supplementary materials related to this article can be found online at doi:10.1016/j.bbamem.2011.08.002. Acknowledgments The microscopy experiments were performed by using the apparatus installed at CEMIM, a division of the National Museum of Natural History (MNHN) and we thank Dr. Marc Gèze and Dr. Marc Dellinger for the use of this experimental set-up. The authors also thank Pr. R. Santus for useful discussions.

[9]

[10] [11] [12]

[13]

[14]

[15] [16]

[17] [18]

[19]

[20]

[21]

[22] [23]

[24]

[25]

[26] [27] [28] [29] [30]

[31]

References [1] A.W. Girotti, T. Kriska, Role of lipid hydroperoxides in photo-oxidative stress signaling, Antioxid Redox Signal 6 (2004) 301–310. [2] V. Ruipérez, F. Darios, B. Davletov, Alpha-synuclein, lipids and Parkinson's disease, Prog Lipid Res 49 (2010) 420–428. [3] I.R. Rodriguez, I.M. Larrayoz, Cholesterol oxidation in the retina: implications of 7KCh formation in chronic inflammation and age-related macular degeneration, J Lipid Res 51 (2011) 2847–2862. [4] N.A. Strobel, R.G. Fassett, S.A. Marsh, J.S. Coombes, Oxidative stress biomarkers as predictors of cardiovascular disease, Int J Cardiol 147 (2011) 191–201. [5] A. Sevanian, F. Ursini, Lipid peroxidation in membranes and low-density lipoproteins: similarities and differences, Free Radic Biol Med 29 (2000) 306–311. [6] G. van Ginkel, A. Sevanian, Lipid peroxidation-induced membrane structural alterations, Meth Enzymol 233 (1994) 273–288. [7] P. Filipe, A.M. Silva, R.S. Seixas, D.C. Pinto, A. Santos, L.K. Patterson, J.N. Silva, J.A. Cavaleiro, J.P. Freitas, J.C. Maziere, R. Santus, P. Morliere, The alkyl chain length of 3-alkyl-3′,4′,5,7-tetrahydroxyflavones modulates effective inhibition of oxidative damage in biological systems: illustration with LDL, red blood cells and human skin keratinocytes, Biochem Pharmacol 77 (2009) 957–964. [8] A. Hernandez-Hernandez, M.N. Garabatos, M.C. Rodriguez, M.L. Vidal, A. LopezRevuelta, J.I. Sanchez-Gallego, M. Llanillo, J. Sanchez-Yague, Structural characteristics of a lipid peroxidation product, trans-2-nonenal, that favour inhibition of

[32] [33]

[34] [35] [36] [37]

[38]

[39]

[40]

2971

membrane-associated phosphotyrosine phosphatase activity, Biochim Biophys Acta 1726 (2005) 317–325. V.E. Kagan, G.G. Borisenko, Y.Y. Tyurina, V.A. Tyurin, J. Jiang, A.I. Potapovich, V. Kini, A.A. Amoscato, Y. Fujii, Oxidative lipidomics of apoptosis: redox catalytic interactions of cytochrome c with cardiolipin and phosphatidylserine, Free Radic Biol Med 37 (2004) 1963–1985. N. Paillous, S. Fery-Forgues, Interest of photochemical methods for induction of lipid peroxidation, Biochimie 76 (1994) 355–368. T.J. Dougherty, C.J. Gomer, B.W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan, Q. Peng, Photodynamic therapy, J Natl Cancer Inst 90 (1998) 889–905. K. Berg, P.K. Selbo, L. Prasmickaite, T.E. Tjelle, K. Sandvig, J. Moan, G. Gaudernack, O. Fodstad, S. Kjolsrud, H. Anholt, G.H. Rodal, S.K. Rodal, A. Hogset, Photochemical internalization: a novel technology for delivery of macromolecules into cytosol, Cancer Res 59 (1999) 1180–1183. P.K. Selbo, G. Sivam, O. Fodstad, K. Sandvig, K. Berg, In vivo documentation of photochemical internalization, a novel approach to site specific cancer therapy, Int J Cancer 92 (2001) 761–766. P.K. Selbo, A. Weyergang, A. Hogset, O.J. Norum, M.B. Berstad, M. Vikdal, K. Berg, Photochemical internalization provides time- and space-controlled endolysosomal escape of therapeutic molecules, J Control Release 148 (2010) 2–12. M.K. Kuimova, G. Yahioglu, P.R. Ogilby, Singlet oxygen in a cell: spatially dependent lifetimes and quenching rate constants, J Am Chem Soc 131 (2009) 332–340. J. Dairou, C. Vever-Bizet, D. Brault, Interaction of sulfonated anionic porphyrins with HIV glycoprotein gp120: photodamages revealed by inhibition of antibody binding to V3 and C5 domains, Antivir Res 61 (2004) 37–47. J. Moan, On the diffusion length of singlet oxygen in cells and tissues, J Photochem Photobiol 6 (1990) 343–347. A. Hogset, L. Prasmickaite, P.K. Selbo, M. Hellum, B.O. Engesaeter, A. Bonsted, K. Berg, Photochemical internalisation in drug and gene delivery, Adv Drug Deliv Rev 56 (2004) 95–115. S. Bonneau, C. Vever-Bizet, Tetrapyrrole photosensitisers, determinants of subcellular localisation and mechanisms of photodynamic processes in therapeutic approaches, Expert Opin Ther Pat 18 (2008) 1025–1040. S. Bonneau, C. Vever-Bizet, H. Mojzisova, D. Brault, Tetrapyrrole-photosensitizers vectorization and plasma LDL: a physico-chemical approach, Int J Pharm 344 (2007) 78–87. H. Mojzisova, S. Bonneau, D. Brault, Structural and physico-chemical determinants of the interactions of macrocyclic photosensitizers with cells, Eur Biophys J 36 (2007) 943–953. M. Marzorati, P. Bigler, M. Vermathen, Interactions between selected photosensitizers and model membranes: an NMR classification, Biochim Biophys Acta (2011). M. Vermathen, M. Marzorati, P. Vermathen, P. Bigler, pH-dependent distribution of chlorin e6 derivatives across phospholipid bilayers probed by NMR spectroscopy, Langmuir 26 (2010) 11085–11094. H. Mojzisova, S. Bonneau, C. Vever-Bizet, D. Brault, The pH-dependent distribution of the photosensitizer chlorin e6 among plasma proteins and membranes: a physico-chemical approach, Biochim Biophys Acta 1768 (2007) 366–374. H. Mojzisova, S. Bonneau, C. Vever-Bizet, D. Brault, Cellular uptake and subcellular distribution of chlorin e6 as functions of pH and interactions with membranes and lipoproteins, Biochim Biophys Acta 1768 (2007) 2748–2756. J. Heuvingh, S. Bonneau, Asymmetric oxidation of giant vesicles triggers curvatureassociated shape transition and permeabilization, Biophys J 97 (2009) 2904–2912. L. Miao, U. Seifert, M. Wortis, H.G. Döbereiner, Budding transitions of fluid-bilayer vesicles: the effect of area-difference elasticity, Phys Rev E 49 (1994) 5389–5407. B. Mesmin, F.R. Maxfield, Intracellular sterol dynamics, Biochim Biophys Acta 1791 (2009) 636–645. G. van Meer, D.R. Voelker, G.W. Feigenson, Membrane lipids: where they are and how they behave, Nat Rev Mol Cell Biol 9 (2008) 112–124. L.M. Estronca, M.J. Moreno, W.L. Vaz, Kinetics and thermodynamics of the association of dehydroergosterol with lipid bilayer membranes, Biophys J 93 (2007) 4244–4253. J.A. Hamilton, Fast flip-flop of cholesterol and fatty acids in membranes: implications for membrane transport proteins, Curr Opin Lipidol 14 (2003) 263–271. T.L. Steck, J. Ye, Y. Lange, Probing red cell membrane cholesterol movement with cyclodextrin, Biophys J 83 (2002) 2118–2125. R.J. Bruckner, S.S. Mansy, A. Ricardo, L. Mahadevan, J.W. Szostak, Flip-flop-induced relaxation of bending energy: implications for membrane remodeling, Biophys J 97 (2009) 3113–3122. Z. Chen, R.P. Rand, The influence of cholesterol on phospholipid membrane curvature and bending elasticity, Biophys J 73 (1997) 267–276. D. Needham, R.S. Nunn, Elastic deformation and failure of lipid bilayer membranes containing cholesterol, Biophys J 58 (1990) 997–1009. J. Huang, G.W. Feigenson, A microscopic interaction model of maximum solubility of cholesterol in lipid bilayers, Biophys J 76 (1999) 2142–2157. J. Huang, J.T. Buboltz, G.W. Feigenson, Maximum solubility of cholesterol in phosphatidylcholine and phosphatidylethanolamine bilayers, Biochim Biophys Acta 1417 (1999) 89–100. R.F. de Almeida, J. Borst, A. Fedorov, M. Prieto, A.J. Visser, Complexity of lipid domains and rafts in giant unilamellar vesicles revealed by combining imaging and microscopic and macroscopic time-resolved fluorescence, Biophys J 93 (2007) 539–553. A. Parker, K. Miles, K.H. Cheng, J. Huang, Lateral distribution of cholesterol in dioleoylphosphatidylcholine lipid bilayers: cholesterol–phospholipid interactions at high cholesterol limit, Biophys J 86 (2004) 1532–1544. W.V. Kraske, D.B. Mountcastle, Effects of cholesterol and temperature on the permeability of dimyristoylphosphatidylcholine bilayers near the chain melting phase transition, Biochim Biophys Acta 1514 (2001) 159–164.

2972

R. Kerdous et al. / Biochimica et Biophysica Acta 1808 (2011) 2965–2972

[41] E. Schnitzer, I. Pinchuk, A. Bor, A. Leikin-Frenkel, D. Lichtenberg, Oxidation of liposomal cholesterol and its effect on phospholipid peroxidation, Chem Phys Lipids 146 (2007) 43–53. [42] T. Yoda, M.C. Vestergaard, Y. Akazawa-Ogawa, Y. Yoshida, T. Hamada, M. Takagi, Dynamic response of a cholesterol-containing model membrane to oxidative stress, Chem Lett 39 (2010) 1273–1274. [43] M. Angelova, D.S. Dimitrov, A mechanism of liposome electroformation, In Trends in colloid and interface science II, 1988, pp. 59–67. [44] K. Kuzelova, D. Brault, Kinetic and equilibrium studies of porphyrin interactions with unilamellar lipidic vesicles, Biochemistry 33 (1994) 9447–9459. [45] S. Bonneau, C. Vever-Bizet, P. Morliere, J.C. Maziere, D. Brault, Equilibrium and kinetic studies of the interactions of a porphyrin with low-density lipoproteins, Biophys J 83 (2002) 3470–3481. [46] A. Santos, A.M. Rodrigues, A.J. Sobral, P.V. Monsanto, W.L. Vaz, M.J. Moreno, Early events in photodynamic therapy: chemical and physical changes in a POPC: cholesterol bilayer due to hematoporphyrin IX-mediated photosensitization, Photochem Photobiol 85 (2009) 1409–1417. [47] N. Rodriguez, J. Heuvingh, F. Pincet, S. Cribier, Indirect evidence of submicroscopic pores in giant unilamellar vesicles, Biochim Biophys Acta 1724 (2005) 281–287. [48] E.A. Kotova, A.V. Kuzevanov, A.A. Pashkovskaya, Y.N. Antonenko, Selective permeabilization of lipid membranes by photodynamic action via formation of hydrophobic defects or pre-pores, Biochim Biophys Acta 1808 (2011) 2252–2257. [49] A. Pashkovskaya, E. Kotova, Y. Zorlu, F. Dumoulin, V. Ahsen, I. Agapov, Y. Antonenko, Light-triggered liposomal release: membrane permeabilization by photodynamic action, Langmuir 26 (2010) 5726–5733.

[50] J.T. Saari, J.S. Beck, Probability density function of the red cell membrane permeability coefficient, Biophys J 14 (1974) 33–45. [51] E. Zenkevich, E. Sagun, V. Knyukshto, A. Shulga, A. Mironov, O. Efremova, R. Bonnett, S. Songca, M. Kassem, Photophysical and photochemical properties of potential porphyrin and chlorin photosensitizers for pdt, J Photochem Photobiol B 33 (1996) 171–180. [52] D.H. Kim, J.A. Frangos, Effects of amyloid beta-peptides on the lysis tension of lipid bilayer vesicles containing oxysterols, Biophys J 95 (2008) 620–628. [53] J.M. Kauffman, P.W. Westerman, M.C. Carey, Fluorocholesterols, in contrast to hydroxycholesterols, exhibit interfacial properties similar to cholesterol, J Lipid Res 41 (2000) 991–1003. [54] A.P. Castano, T.N. Demidova, M.R. Hamblin, Mechanisms in photodynamic therapy: part two—cellular signaling, cell metabolism and modes of cell death, Photodiagn Photodyn Ther 2 (2005) 1–23. [55] G.R. Buettner, B.A. Jurkiewicz, Catalytic metals, ascorbate and free radicals: combinations to avoid, Radiat Res 145 (1996) 532–541. [56] M.R. Domingues, A. Reis, P. Domingues, Mass spectrometry analysis of oxidized phospholipids, Chem Phys Lipids 156 (2008) 1–12. [57] E. Karatekin, O. Sandre, H. Guitouni, N. Borghi, P.H. Puech, F. Brochard-Wyart, Cascades of transient pores in giant vesicles: line tension and transport, Biophys J 84 (2003) 1734–1749. [58] K.A. Riske, T.P. Sudbrack, N.L. Archilha, A.F. Uchoa, A.P. Schroder, C.M. Marques, M.S. Baptista, R. Itri, Giant vesicles under oxidative stress induced by a membraneanchored photosensitizer, Biophys J 97 (2009) 1362–1370. [59] H.G. Dobereiner, E. Evans, M. Kraus, U. Seifert, M. Wortis, Mapping vesicle shapes into the phase diagram: a comparison of experiment and theory, Phys Rev E 55 (1997) 4458–4474.