Interaction of curcumin with lipid monolayers and liposomal bilayers

Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Interaction of curcumin with lipid monolayers and liposomal bilayers Anna Karewicz a,∗ , Dorota Bielska a , Barbara Gzyl-Malcher a,b , Mariusz Kepczynski a , Radosław Lach c , Maria Nowakowska a a b c

Nanotechnology of Polymers and Biomaterials Group, Physical Chemistry Department, Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland Physical Chemistry of Surfaces Group, Physical Chemistry Department, Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Kraków, Poland Department of Advanced Ceramics, Faculty of Materials Science and Ceramics, AGH – University of Science and Technology, Al. Mickiewicza 30, 30-059 Krakow, Poland

a r t i c l e

i n f o

Article history: Received 31 March 2011 Accepted 24 June 2011 Available online 1 July 2011 Keywords: Curcumin Liposomes Langmuir monolayer Fluorescence

a b s t r a c t Curcumin shows huge potential as an anticancer and anti-inflammatory agent. However, to achieve a satisfactory bioavailability and stability of this compound, its liposomal form is preferable. Our detailed studies on the curcumin interaction with lipid membranes are aimed to obtain better understanding of the mechanism and eventually to improve the efficiency of curcumin delivery to cells. Egg yolk phosphatidylcholine (EYPC) one-component monolayers and bilayers, as well as mixed systems containing additionally dihexadecyl phosphate (DHP) and cholesterol, were studied. Curcumin binding constant to EYPC liposomes was determined based on two different methods: UV/Vis absorption and fluorescence measurements to be 4.26 × 104 M−1 and 3.79 × 104 M−1 , respectively. The fluorescence quenching experiment revealed that curcumin locates in the hydrophobic region of EYPC liposomal bilayer. It was shown that curcumin impacts the size and stability of the liposomal carriers significantly. Loaded into the EYPC/DPH/cholesterol liposomal bilayer curcumin stabilizes the system proportionally to its content, while the EYPC/DPH system is destabilized upon drug loading. The three-component lipid composition of the liposome seems to be the most promising system for curcumin delivery. An interaction of free and liposomal curcumin with EYPC and mixed monolayers was also studied using Langmuir balance measurements. Monolayer systems were treated as a simple model of cell membrane. Condensing effect of curcumin on EYPC and EYPC/DHP monolayers and loosening influence on EYPC/DHP/chol ones were observed. It was also demonstrated that curcumin-loaded EYPC liposomes are more stable upon interaction with the model lipid membrane than the unloaded ones. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Curcumin (curc), a natural polyphenol derived from a traditional herbal remedy – Curcuma longa, shows a variety of desired pharmacological properties, including anti-inflammatory, anti-cancer and anti-oxidant activities. The pleiotropic role of this compound includes the inhibition of transcription factors (NF-␬B and AP-1) [1], enzymes (COX-2, MMPs) [2], cell cycle arrest (cyclin D1) [3] and proliferation (EGFR and Akt). Various studies carried out in the past years confirmed its strong potential to inhibit the growth of various cancer cells [4]. Anticancer properties of that compound involve modulation of several cell signaling pathways at multiple levels, resulting in alterations in gene expression, cell cycle inhibition or apoptosis [5]. Recently, curcumin was shown to possess a therapeutic potential in the treatment of colon cancer in tumors that are resistant to conventional chemotherapy [6]. In addition, this natural polyphenol is also an inhibitor of vascular hyper-

∗ Corresponding author. Tel.: +48 12 663 2020; fax: +48 12 634 05 15. E-mail address: [email protected] (A. Karewicz). 0927-7765/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2011.06.037

permeability following haemorrhagic shock thanks to its strong anti-oxidant properties [7] and inhibits cyclooxygenase-2 (COX2) gene expression, which is responsible for the inflammation and, hence, acts as an anti-inflammatory agent [8]. Unfortunately the high potential of curcumin for treatment of cancer and chronic inflammation is hindered by some of its disadvantageous properties. Since that hydrophobic compound is only sparingly soluble in water, its bioavailability is poor. In addition, it also exhibits very low stability under neutral pH conditions, although in blood the degradation process is somewhat slower than in buffer and serumfree medium [9]. Both problems (low bioavailability and stability) have to be addressed before curcumin may be applied as an efficient drug. The liposomal formulation of curcumin was suggested as a solution [10]. Liposomes show a potential as an effective nanocarrier allowing for both stabilization of the compound in physiological pH and increasing its solubility in aqueous environment. On the other hand, phosphatidylcholine (EYPC) is one of the main constituents of the cell membrane and PC liposomes are widely used as its model. Therefore, thorough studies on the curcumin impact on EYPC mono- and bilayer systems are of great importance for the better

232

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

understanding of the mechanism and improving the efficiency of curcumin delivery to cells. So far, the studies on liposomeencapsulated curcumin have been presented in several papers. Most of them are focused almost exclusively on the biological studies of the formulation effectiveness in the anti-cancer treatment [10,11]. Not much attention was given to the curcumin–liposome interactions, although there are singular works dedicated to the binding properties of curcumin [12] and its thinning effect on lipid membranes [13]. The main aim of current work was to perform a detailed study on the curcumin interactions with lipid mono- and bilayers. Two main issues related to the curcumin delivery, namely optimization of the carrier performance and the interaction of the free and liposomebound curcumin with cell membranes were addressed. First, the affinity of the dye to liposomes and its location in their bilayer was determined using spectroscopic methods. DLS measurements and titration with lysis-inducing surfactant, Triton X-100, allowed to determine the effect of curcumin on the size and stability of the liposomal carriers. Next, the changes evoked by introduction of curcumin into the lipid monolayer, used in this study as a simple model of cell membrane, were described. The interactions of the free and liposomal curcumin with monolayer were also followed using Langmuir balance. Finally, the influence of two other lipids on liposomal carrier, as well as EYPC monolayer properties, was established. Bearing negative charge dihexadecyl phosphate (DHP) and cholesterol (chol), known for modulating membranes fluidity and permeability were chosen for those studies.

2. Materials and methods 2.1. Materials Curcumin (≥94% curcuminoid content, ≥80% curcumin), dihexadecyl phosphate (DHP), cholesterol (chol, 99%), 2bromohexadecanoic acid (99%), 11-bromoundecanoic acid (99%), 16-bromohexadecanoic acid (99%) and Sephadex G-50 (20–50 ␮m) were obtained from Sigma–Aldrich. l-␣-Phosphatidylcholine type XIII-E from egg yolk (EYPC, 99%, solution of 100 mg/ml in ethanol) was obtained from Sigma Chemical Co. (St. Louis, MO). Triton X-100, methanol (spectroscopic grade) and chloroform (spectroscopic grade) were purchased from POCH (Gliwice, Poland).

2.3. Preparation of giant liposomes (GUV) containing curcumin 10 ␮l of curcumin solution (200 ␮M in ethanol) was mixed with 14 ␮l of fructose (Fru) solution (19 mM in ethanol) and added to 50 ␮l solution of EYPC (4 mg/ml of lipid in ethanol) in a 10-ml glass test tube. The mixture was vortexed for about 5 min and then solvent was evaporated under a gentle stream of nitrogen. The dry lipid film was hydrated with 1 ml of 0.01 M NaCl aqueous solution and left for 24 h at room temperature. As a result the giant liposomes were formed. Images of curcumin-labeled GUV were acquired with a Nikon inverted microscope Ti-E with a confocal system Nikon A1 using a Plan Apo 100×/1.4 Oil DIC objective and a 405 nm diode laser for excitation. Images were acquired at a resolution of 2048 × 2048. Fluorescence spectra of curc embedded into liposomes were collected using a 32-channel spectral detector. 2.4. Absorption and emission spectra measurements UV measurements were carried out at 25 ◦ C with a HewlettPackard 8452A diode-array spectrophotometer equipped with an HP 89090A Peltier temperature control accessory using a quartz cuvette of 1 cm optical path length. Fluorescence emission spectra were measured using Fluorescence Spectrometer LS55 Perkin Elmer precisely at 25 ◦ C. 2.5. Determination of the liposome-binding constants A spectroscopic titration technique was used to determine the binding constants (Kb ) of the curcumin chromophores to lipid vesicles. Details on that technique were described previously [14]. Fluorescence of curcumin in aqueous solution is very weak, but its intensity increases considerably in the presence of liposomes. The changes in curcumin fluorescence intensity were followed as a function of lipid concentration. After each addition of an aliquot of liposomal solution the system was equilibrated and steady-state emission spectrum of curcumin chromophore was recorded. The incubation time required to reach equilibrium was determined experimentally, all experiments were done in PBS buffer (pH 7.4). The liposome-binding constant for curcumin was calculated using the previously derived Eq. (1) [15] using the set of fluorescence spectra for the lipid concentration range of 0–1.74 × 10−4 M. F0 + Fcomplex Kb [L]

2.2. Preparation of unloaded and curcumin-loaded liposomes

F=

Liposomes were prepared by hydration of the dry lipid film, followed by sonication. 50 ␮l of the lipid was placed in the 10 ml glass bottle. For EYPC liposomes 100 mg/ml egg yolk phosphatidylcholine solution in ethanol was used. For EYPC/DHP vesicles 5 mg/ml dihexadecyl phosphate solution in chloroform was mixed with 100 mg/ml EYPC solution to obtain 1:0.125 weight ratio of both lipids. For EYCP/DHP/chol liposomes, DHP (5 mg/ml) and cholesterol (3.5 mg/ml) were dissolved in chloroform and added to the l-␣-phosphatidylcholine (100 mg/ml) solution in ethanol in the weight ratio EYPC:DHP:chol = 1:0.125:0.0525. For all curcumincontaining liposomes 200 ␮l of curcumin solution in chloroform (200 ␮M) was then added to the lipid solution. All components were thoroughly mixed and all the solvents were evaporated under a gentle stream of gaseous nitrogen to complete dryness. The dry lipid film was then hydrated with 2 ml of 1 mM PBS (phosphate buffered saline, pH 7.4) and vortexed. The resulting vesicle dispersion was subjected to sonication using ultrasound bath (5 min) and ultrasonic probe sonicator (20 kHz Sonics Vibracell CV18, 20 min, amplitude 40%, pulsar 4, at 5 ◦ C). Titanium particles from the sonicator horn were removed by centrifugation.

where F is curcumin fluorescence intensity measured in the presence of the lipid at concentration [L], F0 is curcumin fluorescence intensity measured in the absence of the lipid, Fcomplex is an asymptotic value of fluorescence intensity at complete of curcumin and Kb is the binding constant to liposomes given in [M−1 ]. The binding constant (Ka ) of curcumin molecules to liposomal bilayer was determined also by the spectrophotometric measurements. The changes in absorbance of the solution containing curcumin at the concentration [curc] ranging from 3 × 10−6 to 2 × 10−5 M and liposomes at constant concentration [L]0 = 3.3 mM were followed in PBS at pH 7.4. Ka was then calculated according to Benesi–Hildebrand equation [16].

1 + Kb [L]

1 1 1 1 = × + [curc] A lε[L]0 Ka lε[L]0

(1)

(2)

where A = A − A0 is a difference between the absorbance A at any point of the binding process and the initial absorbance A0 in the absence of curcumin, ε is the differential extinction coefficient and l is the optical path length.

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

233

2.6. Fluorescence quenching by brominated fatty acids A set of samples was prepared for each brominated fatty acid. Each sample contained sensitizer (15% of the brominated fatty acid), EYPC and curcumin (in the same ratio as in point 2.2). The mixture was vortexed for about 5 min and then solvents were evaporated under a gentle stream of nitrogen. The dry lipid film was hydrated with 2 ml of 0.001 M PBS pH 7.4 and vortexed. The resulting vesicle dispersion was subjected to sonication using ultrasound bath (5 min) and ultrasonic probe sonicator (20 kHz Sonics Vibracell CV18, 20 min, amplitude (40%), pulsar 4, at 5 ◦ C). The vesicle suspension was then centrifuged at 6000 × g for 5 min. 2.7. Liposome size and zeta-potential A Malvern Nano ZS light-scattering apparatus (Malvern Instrument Ltd., Worcestershire, UK) was used for dynamic light scattering (DLS) and zeta potential measurements. The timedependent autocorrelation function of the photocurrent was acquired every 10 s, with 15 acquisitions for each run. The sample of solutions was illuminated by a 633-nm laser, and the intensity of light scattered at an angle of 173◦ was measured by an avalanche photodiode. The Z-averaged hydrodynamic mean diameters (dZ ), polydispersity (PDI) and distribution profiles of the samples were calculated using the software provided by Malvern. The zeta potential of liposomes was measured using the technique of laser Doppler velocimetry (LDV). 2.8. Determination of the liposomes’ stability Stability of the liposomes was evaluated by titrating the liposome dispersion with 5% solution of Triton X-100 using spectrophotometric detection. Disruption of the vesicles decreased the optical density measured at  = 320 nm. 2.9. Langmuir balance experiments Surface pressure–molecular area isotherms were recorded with a NIMA (Coventry, UK) Langmuir trough (total area = 300 cm2 ). Surface pressure was measured at an accuracy of ±0.1 mN/m using a Wilhelmy plate made of filter paper (ashless Whatman Chr1) connected to an electrobalance. The subphase surface was cleaned repeatedly by sweeping the barrier slowly between the maximum and minimum area positions and aspirating the surface until no change in surface pressure was detectable between the “open” and “closed” positions. The substances were spread from chloroform solutions onto the water surface with a Hamilton microsyringe. After spreading, the monolayers were left to equilibrate for 10 min before the compression was started with a barrier speed of 20 cm2 /min. In experiments with liposomal suspension, the monolayer was compressed to a pressure of 5 mN/m. After a period of equilibration of ∼15 min, 50 ␮l of liposomal solution was spread on the subphase surface and the increase in surface pressure was recorded during 1 h. All experiments were done at 20 ◦ C. The subphase temperature was controlled thermostatically to within 0.1 ◦ C by a circulating water system.

Fig. 1. (A) Confocal microscopic image of the GUV labeled with curcumin. The bar shown in the pictures represents 20 ␮m. (B) Fluorescence spectra of curcumin embedded into liposomes (exc = 405 nm) and dissolved in different organic solvents (exc = 420 nm).

Fig. 1A shows a micrograph of liposomes with entrapped curcumin at the curc/lipid ratio of 0.00075. The liposomes treated with curcumin exhibit strong fluorescence after excitation at 405 nm. That is a clear evidence that the curc chromophore is embedded into the lipid bilayer. The fluorescence spectrum of curcumin entrapped in EYPC bilayer is shown in Fig. 1B together with its spectra of curc in organic solvents of different polarity. As can be seen, the spectrum in liposomes is relatively broad compare to that in the organic solvents. Thus the compound is located at various depth in the membrane, but mostly in the polar part of membrane. EYPC membrane shows high polarity, comparable to acetonitrile and methanol up to a depth of approximately C10 and C12 [17]. 3.2. Liposome-binding constant of curcumin

3. Results and discussion 3.1. Curcumin entrapment in PC liposomes Curcumin was entrapped in the liposomal bilayer by its introduction into the dry lipid film during the preparation procedure. The success of the entrapment was directly confirmed by visualization of the curcumin-labeled GUV using the confocal microscope.

Affinity of curcumin to enter the lipid membrane is quantitatively described by binding constant, Kb . The value of Kb to EYPC liposomes was determined in PBS buffer at pH 7.4. Fig. 2 shows the fluorescence emission spectra of the curcumin solution containing increasing concentrations of EYPC liposomes, while the curcumin concentration was kept constant. The gradual increase in the fluorescence intensity at em = 502 nm was observed upon the lipid

234

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

Fig. 4. Fluorescence spectra of liposomal curcumin in the presence of fatty acids bromated in various positions. Fig. 2. Fluorescence spectra of curcumin (ccurc = 10 ␮M, exc = 450 nm) in PBS buffer pH 7.4 in the presence of the increasing lipid concentration [L]. The insert shows the fluorescence intensity at 502 nm as a function of lipid concentration.

addition, confirming the effective binding of curcumin to the liposomal vesicles. The dependence of the fluorescence intensity in the maximum on the lipid concentration is given as the insert. The value of Kb for curcumin was then determined from this plot to be equal to (3.79 ± 0.24) × 104 M−1 . As the second method to determine the binding constant of curcumin to the EYPC membrane, UV/Vis absorption spectrometry was used. To describe the one-to-one binding process between the host and guest molecules using the absorbance measurements the Benesi–Hildebrand method is typically employed [15,18]. In Fig. 3 the set of absorption spectra of the PC liposome solutions containing various concentrations of curc is presented, while the insert shows the double reciprocal plot, 1/A versus 1/[C], and the obtained fit. The value of the binding constant (Ka ) was calculated from this plot based on Eq. (2) to be (4.26 ± 0.12) × 104 M−1 . The values found using both methods are very close and show a very good agreement with that reported in the literature for liposomes containing cholesterol [12].

3.3. Curcumin location studies by fluorescence quenching Information on vertical localization of the molecule inside the lipid membrane is very important. Comparison of the extent of fluorescence quenching by two lipid-bound quenchers that are located at known, different vertical depths in the bilayer can provide information regarding the vertical depth of the fluorophore in a bilayer [19–22]. Brominated derivatives of carboxylic acids, such as 2-bromohexadecanoic acid (2-Br), 16-bromohexadecanoic acid (16-Br), and 11-bromoundecanoic acid (11-Br) were used in our studies. They acted as fluorescence quenchers active only at a certain depths of the bilayer. 2-Br has quenching moieties that are located in the interface between the lipid chains and polar region of the bilayer. The quenching groups of the other two of the quenchers used are immersed deeper in the membrane, in the hydrophobic chain region. Liposomes containing curcumin and one of the quenchers at the [lipid]:[quencher] ratio of 85:15 were prepared. The reduction of the fluorescence intensity compared to that obtained for the liposomes without quencher was measured. The results are shown in Fig. 4. The largest quenching effect was observed for 11-Br. These findings suggest that curcumin is localized in the hydrophobic acyl chain region, close to the glycerol group of the lipid molecules. 3.4. Loading capacity of the liposomal carriers and their stability As was demonstrated above, curcumin locates preferentially in the lipid bilayer of the liposome. Therefore the loading capacity of carrier is directly related to its interaction with the lipid components of the vesicle. Using DLS method and spectrophotometric titration the correlation of the extent of curcumin loading with the vesicle’s size and stability was investigated. Table 1 presents the changes in the average diameter of the PC liposomes containing various amounts of curcumin. Up to 0.84 Table 1 The values of the mean hydrodynamic diameter (dZ ) and polydispersity (PDI) for PC liposomes unloaded and loaded with various amounts of curcumin at pH 7.4.

Fig. 3. Absorption spectra of the liposome dispersion in the presence of curcumin in the concentration range of 0–2 × 10−5 M. The insert shows 1/A as a function of 1/[curc] and the fitted line.

Weight % of curcumin in formulation

dZ [nm]

– 0.14 0.28 0.42 0.56 0.84 1.12

51.94 50.54 48.27 48.39 48.06 43.53 52.49

± ± ± ± ± ± ±

PDI 0.87 0.28 0.98 0.63 0.81 0.87 0.37

0.262 0.281 0.247 0.241 0.255 0.212 0.234

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

235

w/w % of the dye content the size of the vesicle decreases with the increasing amount of curcumin, confirming its condensing effect on the liposomal bilayer, observed also by the Langmuir balance measurements (see below). Also the loading capacity of curcumin in EYPC liposomes was found to be 0.84 w/w %, as for the higher amounts added to the formulation a precipitation, accompanied by an increase in the average size occurs, most probably due to the appearance of aggregates as a result of destabilization of the system. 3.5. Langmuir monolayer studies As was already mentioned, the bilayer properties of the liposomal carrier can significantly alter its interactions with biological environment, including the cell membranes and, as a result, modify significantly the pharmacokinetics and efficacy of the drug delivery system. Lipid monolayers are suitable as membrane target models for studying liposomal drug carriers [23]. Hernandez-Borrell et al. have shown that a simple lipid monolayer – liposome model can be used to simulate experimentally the observed interactions between drug-loaded lipid vesicles and cell surfaces. Such interactions can be investigated by spreading a lipid/drug and lipid/drug delivery system mixture over a subphase to form a monolayer in a Langmuir trough. The monolayer is then compressed, and the surface pressure () – molecular area (A) isotherms of the pure lipid and the lipid with drug or drug delivery system are compared [24]. In our studies interactions between the free or entrapped in liposome curcumin and EYPC monolayer were investigated. In the first step, the –A isotherms were recorded for EYPC, curcumin and mixed EYPC–curcumin (0.5 wt.%) monolayers (Fig. 5A). The isotherm for EYPC demonstrates the phase behaviour characteristic of a liquid phase, as a consequence of the lipid composition, representing a mixture of molecules containing various fatty acids. The presence of unsaturated chains in the molecules prevents them from close packing at the interface [25]. The initial increase in the surface pressure upon film compression occurs at about 140 A˚ 2 /molecule, and the pressure rises steadily as the molecular area decreases. The EYPC monolayer can be compressed to nearly 43 mN/m before its collapse. Curcumin, on the other hand, forms a more expanded monolayer characterized by a less steep isotherm, which collapses at about 10.4 mN/m. A less condensed character of that monolayer is also demonstrated by the lower values of the compression moduli CS −1 presented in Fig. 5B. The surface compressibility modulus of the monolayer is calculated from surface pressure and area per molecule data according to the following equation [26]: CS−1

 d 

= −A

dA

(3) T

where A is the area per molecule and (d/dA)T is the surface pressure change with area under isothermal conditions. High CS −1 values suggest a closely packed film that requires a small area change for attaining lower surface tensions (higher surface pressures). When curcumin is added to EYPC, the average cross sectional area per molecule in the mixed monolayer decreases. The isotherm of mixed monolayer appears between the isotherms of pure components and its shape resembles that of pure PC monolayer. That is not surprising, taking into consideration that the mixed monolayer contains mostly EYPC (0.5 wt.% of curcumin). However, as one can see in Fig. 5B, the mixed monolayer is more condensed than the pure ones. The mixed monolayer attains the maximum value of compressibility modulus (CS −1 max ) of 105 mN/m, whereas CS −1 max value obtained for pure EYPC monolayer is only 93 mN/m. The increase of CS −1 max values suggests a condensing effect of curcumin on PC monolayer.

Fig. 5. (A) Surface pressure versus mean area per molecule for EYPC, curcumin and mixed EYPC–curcumin (0.5 wt.%) monolayers spread on water, at 20 ◦ C. (B) Compressibility modulus versus surface pressure for EYPC, curcumin and mixed EYPC–curcumin (0.5 wt.%) monolayers.

Interactions in mixed monolayers can be also studied from the point of view of the miscibility between their components. If a simple mixture of the two components is assumed to form ideally mixed monolayer or are immiscible, then, at a fixed surface pressure, the area occupied by a molecule in the mixed film, Aid can be estimated as [27]: Aid = A1 x1 + A2 x2

(4)

where x1 and x2 are the molar fractions of the components in the mixture, and A1 and A2 are the molecular areas occupied by individual components in the one-component layer. The deviation of the experimentally determined molecular area in the mixed film (A12 ) from the value assuming ideal mixing (Aid ) is attributed to increased packing or expansion. To characterize the mixing behaviour of curcumin and EYPC at various surface pressures, values of the average molecular area for mixed monolayer (A12 ) are compared with those calculated for the case of ideal miscibility of components (Aid ) at three different surface pressures (3, 6, and 9 mN/m) (see Table 2). One can see that the areas available for the lipid molecules in the mixed monolayer are smaller than the calculated ones, thus the condensing effect of curcumin on EYPC monolayer is further confirmed. This finding is in agreement with the results obtained by Ramamoorthy et al. [28]. Based on solid-state NMR and differential scanning calorimetry measurements they have suggested that curcumin molecules insert into the liposomal bilayer in a manner analogous to that of cholesterol. It was also found that curcumin

236

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

Fig. 6. Changes in the surface pressure with time after spreading of 50 ␮l of a liposomal suspension on a monolayer compressed to 5 mN/m.

at low concentrations has a strong ordering effect on membrane structure. Andersen et al. [29] have already shown in their work that curcumin is a general modifier of lipid bilayer material properties and may exert some of its effects on a diverse range of membrane proteins through a bilayer-mediated mechanism These findings support our expectations that curcumin present in the liposome bilayer could, similarly to cholesterol [30], change the vesicle stability and their circulation lifetimes in a body fluids. The next step of the experiment was designed to verify if the surface properties of EYPC monolayer change upon spreading of the liposomal suspension on the monolayer compressed to 5 mN/m. Changes in the surface pressure with time were recorded, the time being set to zero at the beginning of the spreading procedure. Fig. 6 shows the temporal evolution of the surface pressure after deposition of the unloaded and of the curcumin-loaded EYPC liposomes on the EYPC monolayer (curves 1 and 2, respectively). The increase in surface pressure is related to the disruption of the liposomes, yielding surface active molecules that build into the EYPC monolayer. Unloaded vesicles build in the EYPC monolayer easily, changing effectively surface pressure of the monolayer. The intercalation process advances gradually over the whole 60 min period of time. In the case of curcumin-loaded liposomes their intercalation is significantly hindered, most probably due to the condensed nature of the vesicle’s bilayer leading to its stabilization. Only small increase of the surface pressure was observed. The surface pressure initial steeper increase for the first few minutes after deposition was followed by the plateau over the next 50 min. This suggests Table 2 Comparison of the experimentally recorded and calculated mean area per molecule for various monolayers containing 0.5 wt.% of curcumin at various surface pressures. Surface pressure [mN/m]

Calculated area [A˚ 2 /molecule]

Experimental area [A˚ 2 /molecule]

EYPC monolayer containing 0.5 wt.% of curcumin 105.6 3 119.9 6 111.3 97.3 104.4 91.4 9 EYPC/DHP monolayer containing 0.5 wt.% of curcumin 105.4 101.3 3 92.7 6 97.0 9 90.7 86.3 EYPC/DHP/chol monolayer containing 0.5 wt.% of curcumin 3 103.6 114.6 94.9 106.9 6 100.2 9 88.4

Fig. 7. (A) Surface pressure versus mean area per molecule for EYPC monolayers, before and after spreading of 50 ␮l of a liposomal suspension on a EYPC monolayer, compressed to 5 mN/m. (B) Compressibility modulus versus surface pressure for EYPC monolayers, before and after spreading of 50 ␮l of a liposomal suspension on a EYPC monolayer, compressed to 5 mN/m.

that only a small part of the spread liposomes was disrupted soon after deposition. Curve 3 in Fig. 6 shows the surface pressure changes of the curcumin/EYPC mixed monolayer upon deposition of the unloaded EYPC liposomes. Its shape is very similar to curve 2 indicating that the condensing influence of curcumin manifests itself similarly in the flat monolayers as in the small curved bilayers of liposomes by stabilization of the surface and hindering the interactions with other membranes. After 1 h since the liposomal suspension spreading, the mixed monolayer thus obtained was compressed up to collapse pressure and its –A isotherm was recorded and compared with that for pure EYPC monolayer (see Fig. 7A). The dashed curve represents the pure EYPC monolayer being, therefore, essentially the same as that of Fig. 5A, with no liposomes in the system. The shape of isotherm after spreading of EYPC/curcumin liposomes differs slightly from the isotherm of pure EYPC monolayer, being a little steeper. That confirms the previous conclusion that curcumin-loaded liposomes built into EYPC monolayer only to a small extent. This is also confirmed by the data presented in Fig. 7B, where the compressibility modulus values (CS −1 ) are plotted against the surface pressure . The CS −1 values calculated for EYPC monolayer after liposome

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

237

Table 3 The values of the mean hydrodynamic diameter (dZ ) and polydispersity (PDI) for the mixed liposomes. Weight % of curcumin in formulation PC/DPH 0 0.101 0.202 0.404 PC/DPH/chol 0 0.096 0.193 0.386

Fig. 8. (A) Surface pressure versus mean area per molecule for PC, PC/DPH and PC/DPH–curcumin monolayers spread on water, at 20 ◦ C. (B) Surface pressure versus mean area per molecule for EYPC, EYPC/DPH/chol and EYPC/DPH/cholesterol–curcumin monolayers spread on water, at 20 ◦ C.

spreading are a little larger than those obtained for pure EYPC monolayer. The maximum value of CS −1 increases from 93 mN/m (for pure EYPC monolayer) to 102 mN/m (for EYPC monolayer after liposome spreading). 3.6. Influence of DHP and chol on the monolayers and bilayers interaction with curcumin Very often, instead of pure liposomes, the mixed composition of the lipids is used to change the vesicle loading capacity, bilayer permeability or surface charge. One of the reasons may be to decrease the drug leaking by stiffening the bilayer. Another – to modify the bilayer effective charge allowing for the additional stabilization by polyelectrolyte coating. Here we discuss the effect of the two lipids: dihexadecyl phosphate (DHP) and cholesterol. DHP adopts the anionic form at neutral pH, as the pKa value of its head group is in the range of 5.8–6.3 [31]. Therefore, the presence of DHP allows to obtain negatively charged liposomal carriers suitable for additional stabilization using polycation coating. Cholesterol, on the other hand, was chosen for further studies as another typical liposome constituent. It is interesting due to its confirmed stiffening

dZ [nm]

PDI

48.78 68.48 62.56 60.81

± ± ± ±

0.53 2.96 0.53 1.43

0.261 0.458 0.538 0.387

64.24 63.70 78.07 80.64

± ± ± ±

0.57 1.18 0.40 0.84

0.218 0.230 0.262 0.252

effect on EYPC bilayers and frequent application as a component limiting drug leakage. Introduction of DHP to the EYPC membrane does not change significantly its interaction with curcumin as demonstrated in Fig. 8A. The isotherms show similar profile as those for EYPC monolayer and areas available for the lipid molecules in the mixed monolayer are also smaller than the calculated ones, confirming the condensing effect of curcumin on EYPC/DPH mixed monolayer. Addition of cholesterol to the EYPC/DHP monolayer has a dramatic influence on the monolayer interaction with curcumin. In this system introduction of curcumin to EYPC/DHP/cholesterol monolayer causes relaxation of the structure, as confirmed by the change in the isotherm profile (Fig. 8B) and by the comparison of the experimental and calculated mean area per molecule for mixed EYPC/DHP/cholesterol–curcumin monolayers (Table 2). This behaviour of the monolayer can be explained by the fact that cholesterol exhibits strong ordering effect on the monolayer. Therefore introduction of curcumin disturbs this highly packed arrangement causing some disorder that manifests itself as loosening of the structure. In the next step the results obtained by studying the interactions in monolayers were compared with DLS measurements for liposomes of the same compositions (Table 3). Introducing DHP as an additional component of the bilayer results in the vesicles of significantly smaller diameter. This corresponds well with the condensing effect of DHP presence in EYPC monolayers. When curcumin is added the particle size increases. That is specific for the liposome structure. Curcumin introduces some stiffness to the bilayer and enforces the grow of the vesicle diameter. For the increasing amounts of curcumin again its condensing influence prevails and the liposome sizes gradually decrease. For EYPC/DHP/cholesterol systems it is the presence of cholesterol that is mainly responsible for the higher stiffness of the bilayer. For this reason the unloaded vesicles show larger sizes comparing to EYPC ones. When curcumin is added in lower amounts no significant effect is observed, as it builds in the already quite relaxed lipid structure. However, for the higher curcumin contents further loosening of the bilayer can be observed, again in accordance with the disordering effects observed in monolayers (Table 2). Finally, the stability studies were done for both liposomal systems of mixed composition, as it constitutes one of the most important parameters of the liposome as a drug carrier. A nonionic surfactant Triton X-100 causes disintegration of the liposomal bilayer. 5% Triton X-100 solution was used to titrate the dispersions of the liposomes loaded with various amounts of curcumin. The progress of the solubilization process was followed by measuring the optical density of the dispersion at  = 320 nm. At low surfactant concentrations its molecules adsorb onto the vesicle surface [32], what leads to the temporary grow of the liposome size and to increase in the optical density. After exceeding a certain critical concentration the surfactant induces lyses of the liposomes. The

238

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239

the interactions between the curcumin-loaded liposomes and EYPC monolayer show that curcumin-loaded EYPC liposomes are more stable upon interaction with the model lipid membrane than the unloaded ones. Further experiments using titration method delivered the additional proof for that assumption. Finally, the influence of DHP and cholesterol on the liposomal curcumin stability and the nanocarrier size was analyzed. Unlike for the monolayers, in vesicle systems an important role of the stiffness induced by the presence of both curcumin and cholesterol was confirmed. Curcumin loaded into the EYPC/DPH/cholesterol liposomal bilayer stabilizes the system proportionally to its content while the EYPC/DPH system is destabilized upon the drug loading. The three-component lipid composition of the liposome seems to be the most promising for further studies, necessary to establish the nanocarrier stability in vivo, as well as its circulation lifetimes and efficiency of the drug delivery. Fig. 9. Changes in optical density at  = 320 nm observed for the unloaded and curcumin-loaded PC/DHP liposomal dispersions upon their titration with 5% Triton X-100 at pH 7.4.

Acknowledgements We thank the Polish Ministry of Science and Higher Education for the financial support in the form of the grant NN 209 118937. Project operated within the Foundation for Polish Science Team Programme co-financed by the EU European Regional Development Fund, PolyMed, TEAM/2008-2/6. The research was carried out with the equipment purchased thanks to the financial support of the European Regional Development Fund in the framework of the Polish Innovation Economy Operational Program (contract no. POIG.02.01.00-12-023/08). References

Fig. 10. Changes in optical density at  = 320 nm observed for the unloaded and curcumin-loaded PC/DHP/cholesterol liposomal dispersions upon their titration with 5% Triton X-100 at pH 7.4.

smaller fragments of the bilayer undergo further transformation to the mixed micelles, which results in a decrease of dispersion turbidity [33,34]. In the case of EYPC/DPH vesicles curcumin loaded into the bilayer destabilizes the system proportionally to its content (Fig. 9). This is a serious disadvantage since a compromise between the loading efficiency and stability has to be made. In the case of EYPC/DHP/cholesterol liposomes the opposite relationship was registered (Fig. 10), constituting an obvious advantage over other systems studied. 4. Conclusions Our studies have shown that curcumin, a potential anti-cancer and anti-inflammatory drug, binds effectively to the liposomal bilayer. The measured binding constant is very high. Fluorescence quenching measurements showed that curcumin locates preferentially in the hydrophobic acyl chain region, close to the glycerol group of lipid molecules. Condensing effect of curcumin on EYPC and EYPC/DHP monolayer and loosening influence on EYPC/DHP/cholesterol were observed. An ordering/disordering effect of curcumin suggests that its presence in the liposome bilayer could change significantly the nanocarrier stability. The studies on

[1] S. Shishodia, T. Singh, M.M. Chaturvedi, Modulation of transcription factors by curcumin, Adv. Exp. Med. Biol. 595 (2007) 127. [2] P. Yoysungnoen, P. Wirachwong, P. Bhattarakosol, H. Niimi, S. Patumraj, Effects of curcumin on tumor angiogenesis and biomarkers, COX-2 and VEGF, in hepatocellular carcinoma cell-implanted nude mice, Clin. Hemorheol. Micro. 34 (2006) 109. [3] R.P. Sahu, S. Batra, S.K. Srivastava, Activation of ATM/Chk1 by curcumin causes cell cycle arrest and apoptosis in human pancreatic cancer cells, Brit. J. Cancer 100 (2009) 1425. [4] S.M. Plummer, K.A. Holloway, M.M. Manson, R.J. Munks, A. Kaptein, S. Farrow, L. Howells, Inhibition of cyclo-oxygenase 2 expression in colon cells by the chemopreventive agent curcumin involves inhibition of NF-kappaB activation via the NIK/IKK signalling complex, Oncogene 18 (1999) 6013. [5] D. Karunagaran, R. Rashmi, T.R. Kumar, Induction of apoptosis by curcumin and its implications for cancer therapy, Curr. Cancer Drug Targets 5 (2005) 117. [6] J.L. Watson, R. Hill, P.B. Yaffe, A. Greenshields, M. Walsh, P.W. Lee, C.A. Giacomantonio, D.W. Hoskin, Curcumin causes superoxide anion production and p53-independent apoptosis in human colon cancer cells, Cancer Lett. 297 (2010) 1. [7] B. Tharakan, F.A. Hunter, W.R. Smythe, E.W. Childs, Curcumin inhibits reactive oxygen species formation and vascular hyperpermeability following haemorrhagic shock, Clin. Exp. Pharmacol. Physiol. 37 (2010) 939. [8] F. Zhang, N.K. Altorki, J.R. Mestre, K. Subbaramaiah, A.J. Dannenberg, Curcumin inhibits cyclooxygenase-2 transcription in bile acid- and phorbol ester-treated human gastrointestinal epithelial cells, Carcinogenesis 20 (1999) 445. [9] Y.-J. Wang, M.-H. Pan, A.-L. Cheng, L.-I. Lin, Y.-S. Ho, C.-Y. Hsieh, J.-K. Lin, Stability of curcumin in buffer solutions and characterization of its degradation products, J. Pharm. Biomed. Anal. 15 (1997) 1867–1876. [10] L. Li, F.S. Braiteh, R. Kurzrock, Liposome-encapsulated curcumin: in vitro and in vivo effects on proliferation, apoptosis, signaling, and angiogenesis, Cancer 104 (2005) 1322. [11] N.K. Narayanan, D. Nargi, C. Randolph, B.A. Narayanan, Liposome encapsulation of curcumin and resveratrol in combination reduces prostate cancer incidence in PTEN knockout mice, Int. J. Cancer 125 (2009) 1. [12] A. Kunwar, A. Barik, R. Pandey, K.I. Priyadarsini, Transport of liposomal and albumin loaded curcumin to living cells: an absorption and fluorescence spectroscopic study, Biochim. Biophys. Acta Gen. Subj. 1760 (2006) 1513. [13] Y. Sun, Ch.-Ch. Lee, W.-Ch. Hung, F.-Y. Chen, M.-T. Lee, H.W. Huang, The bound states of amphipathic drugs in lipid bilayers: study of curcumin, Biophys. J. 95 (2008) 2318. [14] H. Benesi, J. Hildebrand, A spectrophotometric investigation of the interaction of iodine with aromatic hydrocarbons, J. Am. Chem. Soc. 71 (1949) 2703. ˛ ´ [15] M. Kepczy nski, R.P. Pandian, K.M. Smith, B. Ehrenberg, Do liposome-binding constants of porphyrins correlate with their measured and predicted partitioning between octanol and water? Photochem. Photobiol. 76 (2002) 127.

A. Karewicz et al. / Colloids and Surfaces B: Biointerfaces 88 (2011) 231–239 [16] O. Exner, Calculating equilibrium constant from spectral data: reliability of the Benesi–Hildebrand method and its modifications, Chemom. Intell. Lab. Syst. 39 (1997) 85. [17] W.K. Subczynski, A. Wisniewska, J.-J. Yin, J.S. Hyde, A. Kusumi, Hydrophobic barriers of lipid bilayer membranes formed by reduction of water penetration by alkyl chain unsaturation and cholesterol, Biochemistry 33 (1994) 7670. ˛ ´ nski, A. Karewicz, A. Górnicki, M. Nowakowska, Interactions of [18] M. Kepczy porphyrin covalently attached to poly(methacrylic acid) with liposomal membranes, J. Phys. Chem. B 109 (2005) 1289. [19] E. Blatt, W.H. Sawyer, Depth-dependent fluorescent quenching in micelles and membranes, Biochim. Biophys. Acta Biomembr. 822 (1985) 43. [20] J.H. Van Esch, M.C. Feiters, A.M. Peters, J.M. Nolte, UV–Vis, fluorescence, and EPR studies of porphyrins in bilayers of dioctadecyldimethylammonium surfactants, J. Phys. Chem. 98 (1994) 5541. [21] R.F. Steiner, Topics in fluorescence spectroscopy, in: J.R. Lakowicz (Ed.), Principles, vol. 2, Plenum, New York, 1991 (Chapter 1). ˛ ´ [22] M. Kepczy nski, M. Kumorek, M. Stepniewski, T. Róg, B. Kozik, D. Jamróz, J. Bednar, M. Nowakowska, Behavior of 2,6-bis(decyloxy)naphthalene inside lipid bilayer, J. Phys. Chem. B 114 (2010) 15483. [23] J. Hernandez-Borrell, F. Mas, J. Puy, A theoretical approach to describe monolayer-liposome lipid interaction, Biophys. Chem. 36 (1990) 47. [24] Ch. Peetla, A. Stine, V. Labhasetwar, Biophysical interactions with model lipid membranes: applications in drug discovery and drug delivery, Mol. Pharm. 6 (2009) 1264.

239

[25] B. Gzyl, M. Filek, A. Dudek, Biophysical interactions with model lipid membranes: Applications in drug discovery and drug delivery, J. Colloid Interface Sci. 269 (2004) 153. [26] J.T. Davies, E.K. Rideal, Interfacial Phenomena, Academic Press, New York, 1963. [27] K.S. Birdi, Lipid and Biopolymer Monolayers at Liquid Interfaces, John Wiley & Sons, New York, 1989. [28] J. Barry, M. Fritz, J.R. Brender, P.E.S. Smith, D-K. Lee, A. Ramamoorthy, Determining the effects of lipophilic drugs on membrane structure by solid-state NMR spectroscopy: the case of the antioxidant curcumin, J. Am. Chem. Soc. 131 (2009) 4490. [29] H.I. Ingolfsson, R.E. Koeppe, O.S. Andersen, Curcumin is a modulator of bilayer material properties, Biochemistry 46 (2007) 10384. [30] S.C. Semple, A. Chonn, P.R. Cullis, Influence of cholesterol on the association of plasma proteins with liposomes, Biochemistry 35 (1996) 2521. [31] R.A. Mortara, F.H. Quina, H. Chaimovich, Formation of closed vesicles from a simple phosphate diester. Preparation and some properties of vesicles of dihexadecyl phosphate, Biochem. Biophys. Res. Commun. 81 (1978) 1080. [32] J. Gustafsson, G. Oradd, M. Almgren, Disintegration of thelecithin lamellar phase by cationic surfactants, Langmuir 13 (1997) 6956. [33] A. De la Maza, J.L. Parra, Vesicle–micelle structural transitions of phospholipid bilayers and sodium dodecyl sulfate, Langmuir 11 (1995) 2435. [34] A.A. Yaroslavov, E.A. Kiseliova, O.Yu. Udalykh, V.A. Kabanov, Integrity of mixed liposomes contacting a polycation depends on the negatively charged lipid content, Langmuir 14 (1998) 5160.