Article pubs.acs.org/molecularpharmaceutics
Idebenone Loaded Solid Lipid Nanoparticles Interact with Biomembrane Models: Calorimetric Evidence Lucia Montenegro,* Sara Ottimo, Giovanni Puglisi, Francesco Castelli, and Maria Grazia Sarpietro Department of Drug Sciences, University of Catania, V.le A. Doria 6, 95125 Catania, Italy ABSTRACT: The knowledge of the interactions between solid lipid nanoparticles (SLN) and cell membranes is important to develop effective carrier systems for drug delivery applications. Loading idebenone (IDE), an antioxidant drug useful in the treatment of neurodegenerative diseases, into SLN improves IDE antioxidant activity in in vitro biological studies, but the mechanism by which IDE permeation through the blood−brain barrier (BBB) occurs are still unclear. Therefore, in this research, unloaded and IDE loaded SLN interaction with biomembrane models, consisting of dimyristoylphosphatidylcholine multilamellar vesicles (MLV), were studied by differential scanning calorimetry (DSC). In the experiments performed, unloaded and IDE loaded SLN where incubated with the biomembrane models and their interactions were evaluated through the variations in their calorimetric curves. The results of our DSC studies indicated that the SLN under investigation were able to go inside the phospholipid bilayers with a likely localization in the outer bilayers of the MLV from where they moved toward the inner layers by increasing the contact time between SLN and MLV. Furthermore, IDE loaded SLN were able to release IDE into the biomembrane model, thus facilitating IDE penetration into the bilayers while free IDE showed only a low ability to interact with this model of biomembranes. Our results suggest that these SLN could be regarded as a promising drug delivery system to improve IDE bioavailability and antioxidant activity. KEYWORDS: idebenone, solid lipid nanoparticles, differential scanning calorimetry, biomembrane interactions, MLV liposomes
■
peroxidation,4 IDE was found to induce protection from neuronal damage such as stroke, intoxication, and memory loss, and it also improves recovery after brain trauma.7 Preliminary testing in humans reported that IDE can be regarded as a safe treatment for Friedreich’s ataxia, exhibiting a positive effect on cardiac hypertrophy and neurological function.8 To improve drug efficacy in the treatment of brain diseases, many vesicular and particulate carriers have been investigated and the feasibility of using solid lipid nanoparticles (SLN) as colloidal carriers for drug delivery to the brain has been widely reported.9,10 Due to their lipophilicity and their small particle size, SLN could reach the central nervous system (CNS), overcoming the blood−brain barrier (BBB) by endocytosis or transcytosis in the endothelial cells, or they could permeate the tight junctions between endothelial cells.11,12 Furthermore, retention of SLN in brain blood capillaries with absorption into capillary walls could occur giving rise to higher drug concentration gradients, which, in turn, could lead to increased drug transport across the BBB.11,12 Recently, we have developed a novel technique to prepare SLN using the phase inversion temperature (PIT) method, obtaining SLN with small particle size, narrow size distribution, good stability, and good loading capacity.13 These SLN were
INTRODUCTION In recent years, a great deal of attention has been focused on the role of oxidative stress in pathological conditions. Oxygen reactive species (ROS) can cause both structural and functional damage to cell membranes, due to free radical attacks of the polyunsaturated fatty acid in the biomembrane. Furthermore, when impairment of the mitochondrial respiratory chain occurs, mitochondria may become the main source of ROS.1,2 As widely reported in the literature,3 antioxidant drugs can block ROS production, thus limiting oxidative cell damage. Idebenone (IDE, Figure 1) is a synthetic drug whose
Figure 1. Chemical structure of IDE.
antioxidant activity is due to its structural analogy with coenzyme Q10, a natural antioxidant of cell membranes and a basic component of the mitochondrial electronic transport chain. IDE potent antioxidant activity is mainly due to its ability to inhibit lipid peroxidation (LPO), and to protect cell and mitochondrial membranes from oxidative damage.4 IDE has been proposed to be beneficial in the treatment of neurodegenerative diseases, such as Parkinson's syndrome, Alzheimer's disease (AD), cerebral ischemia, and physiological brain aging processes.5,6 Owing to its ability to inhibit lipid © 2012 American Chemical Society
Received: Revised: Accepted: Published: 2534
March 19, 2012 July 5, 2012 August 3, 2012 August 15, 2012 dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
loaded with IDE, and preliminary in vitro biological tests pointed out that they did not cause any cytotoxic effect and showed better antioxidant properties compared to free IDE when assessed in primary cultures of rat astrocytes.13 Furthermore, the results of in vitro transport studies, performed across MDCKII-MDR1 cell monolayers, chosen as an in vitro model of the BBB, using markers of the paracellular pathway and the transcellular pathway, pointed out that IDE loaded SLN allowed IDE permeation via a transcellular pathway.14 The processes involved in SLN uptake by cells are very complex, and the cell membranes are very intricate structures; as a consequence the biophysical interactions between SLN and biomembrane are difficult to study and understand. Therefore, biomimetic model cell membranes are widely used to study such interactions. In this work, we evaluated the interaction between biomembrane models, represented by multilamellar vesicles (MLV) made of dimyristoylphosphatidylcholine (DMPC), and SLN to get information on the mechanism by which SLN interact with biological membranes and IDE permeation occurs. DMPC, widely used in studying the interaction of bioactive compounds and biomembrane models,15 was chosen for its transition temperature (about 24 °C), that allows the uptake experiments to be carried out at 37 °C, a temperature similar to the physiological one, at which the bilayers are in a disordered liquid crystalline state more favorable to interact with drug. The differential scanning calorimetry technique, which can reveal the effect of “stranger” molecules on the MLV phospholipid bilayers,16,17 was used to study such interactions as well as for SLN characterization.
Table 1. Composition (% w/w) of Unloaded and IDE Loaded SLNa SLN A A1 B B1 C C1 a
ceteth-20
isoceteth-20
oleth-20
GO
CP
7.5 7.5
3.5 3.5 4.4 4.4 3.7 3.7
7 7 7 7 7 7
10.6 10.6 8.7 8.7
IDE 0.7 0.7 0.7
GO = glyceryl oleate; CP = cetyl palmitate. Buffer: q.s. to 100% w/w.
(PIT) method as previously reported.14 The aqueous phase was a phosphate buffer (pH 7.4) consisting of (a) NaCl 4.60 g/L, phosphate monobasic 0.64 g/L and phosphate dibasic 5.0 g/L for the preparation of SLN A (containing isoceteth-20/glyceryl oleate) and (b) NaCl 3.0 g/L, phosphate monobasic 0.60 g/L and phosphate dibasic 6.40 g/L for the preparation of SLN B and C (containing ceteth-20/glyceryl oleate and oleth-20/ glyceryl oleate, respectively). Both buffers contained 0.35% w/ w imidazolidinyl urea and 0.05% w/w methylchloroisothiazolinone and methylisothiazolinone as preservatives. The aqueous phase and the oil phase (consisting of cetyl palmitate, emulsifiers and 0.7% w/w IDE for loaded SLN), were separately heated at ∼90 °C; then the aqueous phase was added drop by drop, at constant temperature and under agitation, to the oil phase. The resulting mixture was cooled to room temperature under slow and continuous stirring. At the phase inversion temperature (PIT) the turbid mixture turned into clear. The PIT temperature was determined using a conductivity meter model 525 (Crison, Milan, Italy), which measured an electric conductivity change when the inversion from w/o to o/w system occurred. No degradation of IDE occurred under these conditions, as confirmed by TLC analysis. A pH-meter model Basic 20 (Crison, Milan, Italy) was used to measure pH values of SLN samples while osmolarity was determined using an osmometer Osmomat, model 030-D (Gonotec, Berlin, Germany). The instrument was previously calibrated with a sample of saline (300 mOsm/kg). Size and Zeta Potential Measurements of SLN. SLN mean particle size and size distribution were determined by dynamic light scattering (DLS) using a Zetasizer NanoZS (ZEN 3600, Malvern, Germany). Samples were measured after dilution (1:1, sample/double distilled water) at 25 °C and adjusted to the temperature 2 min prior to the measurement. The autocorrelation functions were analyzed using the DTS v 5.1 software provided by Malvern. Measurements were performed in triplicate with 20 runs each, and the calculated mean values were used. The determination of the ζ-potential was performed using the technique of laser Doppler velocimetry using a Zetasizer NanoZS after dilution with KCl 1 mM (pH 7.0), as previously reported.18 Transmission Electron Microscopy (TEM). For negativestaining electron microscopy, 5 μL of SLN samples were placed on a 200-mesh Formvar copper grid (TAAB Laboratories Equipment, U.K.) and allowed to be adsorbed. Then, the surplus was removed by filter paper. A drop of 2% (w/v) aqueous solution of uranyl acetate was added over 2 min. After the removal of the surplus, the sample was dried at room conditions before the SLN was imaged with a transmission electron microscope (model JEM 2010, Jeol, Peabody, MA, USA) operating at an acceleration voltage of 200 kV.
■
EXPERIMENTAL SECTION Materials. Idebenone (IDE) was obtained from Wyeth Lederle (Italy). Polyoxyethylene-20-isohexadecyl ether (Arlasolve 200 L, isoceth-20) was a kind gift of Bregaglio (Italy). Glyceryl oleate (Tegin O, GO) was obtained from Th. Goldschmidt Ag (Italy). Cetyl palmitate (Cutina CP, CP) was purchased from Cognis Spa Care Chemicals (Italy). Polyoxyethylene-20-cetyl ether (Brij 58, ceteth-20) was supplied by Fluka (Italy). Polyoxyethylene-20-oleyl ether (Brij 98, oleth-20) was supplied by Sigma Aldrich (USA). Methylchloroisothiazolinone and methylisothiazolinone (Kathon CG) and imidazolidinyl urea were kindly supplied by Sinerga (Italy). 1,2-Dimyristoyl-sn-glycero-3-phosphatidylcholine (purity = 99%) was supplied by Genzyme Pharmaceuticals (Liestal, Switzerland). Methanol and water used in the HPLC procedures were of LC grade and were bought from Merck (Germany). All other reagents were of analytical grade, and they were used as supplied. Calorimetric analyses were performed by a Mettler Toledo STARe system (Switzerland) equipped with a DSC-822e calorimetric cell, and Mettler TA-STARe software was used. The sensitivity was automatically chosen as the maximum possible by the calorimetric system, and the reference pan was filled with the same buffer present in sample under investigation. The calorimetric system was calibrated, in transition temperature and enthalpy changes, following the procedure of the DSC 822 Mettler TA STARe instrument, by using indium, stearic acid, and cyclohexane. Preparation of SLN. SLN whose composition is reported in Table 1 were prepared using the phase inversion temperature 2535
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
Stability Tests. Samples of SLN were stored in airtight jars, and then kept in the dark at room temperature and at 37 °C for two months, separately. Particle size and polydispersity index of the samples were measured at fixed time intervals (24 h, one week, two weeks, three weeks, one month, and two months) after their preparation. SLN Differential Scanning Calorimetry Analysis. 100 μL of SLN was taken from the bulk, put into the calorimetric pan, hermetically sealed and submitted to analysis as follows: (i) a heating scan from 5 to 65 °C (2 °C/min); (ii) a cooling scan from 65 to 5 °C (4 °C/min); for at least three times. Multilamellar Vesicle Preparation. MLV were prepared with and without IDE. Solutions of DMPC and IDE were prepared in chloroform/methanol (1:1, v:v), separately. Aliquots of the DMPC solution were distributed in glass tubes to have the same amount of DMPC (0.010325 mmol) in each tube. Aliquots of the IDE solution were distributed in the same tubes to have the following molar fraction of IDE with respect to DMPC: 0.00, 0.03, 0.06, and 0.12. The solvents were evaporated under a nitrogen stream, and the obtained films were freeze-dried to remove eventual traces of solvents. Depending on the use of MLV, a different amount of bidistilled water was added to the films: (i) MLV used to study the interaction with IDE were added with bidistilled H2O to have 0.007375 mmol of DMPC in 120 μL; (ii) MLV used to study the interaction with SLN were added with bidistilled H2O to have 0.007375 mmol of DMPC in 94 μL. The samples were kept 1 min at 37 °C and vortexed 1 min, for three times, and kept at 37 °C 60 min. MLV Differential Scanning Calorimetry Characterization. 120 μL of the MLV sample was put into the calorimetric pan, hermetically closed and submitted to the following cycles: a heating scan from 5 to 37 °C (2 °C/min); a cooling scan from 37 to 5 °C (4 °C/min); for at least three times. Differential Scanning Calorimetry Analysis of the Interaction between SLN and MLV. Twenty-six microliters of SLN was put into the calorimetric pan where 94 μL of MLV was added. The calorimetric pan was closed and submitted to the following scans: (i) a heating scan from 5 to 55 °C (2 °C/ min); a cooling scan from 55 to 37 °C (4 °C/min); (iii) an isotherm scan of 60 min at 37 °C (a temperature which is very close to the physiological temperature and at which the MLV are in a disordered liquid crystalline state favorable to drug uptake and interaction); (iv) a cooling scan from 37 to 5 °C (4 °C/min). This procedure was repeated for at least eight times. Differential Scanning Calorimetry Analysis of the Uptake of IDE by the MLV. An amount of IDE corresponding to 0.06 molar fraction with respect to DMPC was weighed in the calorimetric pan, and 120 μL of MLV was added. The calorimetric pan was closed and submitted to the same calorimetric scans reported in Differential Scanning Calorimetry Analysis of the Interaction between SLN and MLV.
Figure 2. TEM picture of SLN prepared by the PIT method. This image was obtained from unloaded SLN A.
All SLN showed physiological pH (7.24−7.34) and osmolarity (296−326 mOsm/kg) values, a mean particle size ranging from 33 to 44 nm, a single peak in size distribution, and ζ potential values from +3 to −3 mV (Table 2). Table 2. Particle Size (Size ± SD), Polydispersity Index (Poly ± SD), ζ Potential (ζ ± SD), Phase Inversion Temperature (PIT) Values, pH and Osmolarity (Osm) of Unloaded and IDE Loaded SLN SLN A A1 B B1 C C1
size ± SD (nm) 40 39 44 36 34 35
± ± ± ± ± ±
1 1 1 0 1 0
poly ± SD 0.37 0.31 0.39 0.24 0.26 0.25
± ± ± ± ± ±
0.01 0.02 0.06 0.01 0.01 0.01
ζ ± SD (mV) +2 +2 +3 +2 −2 −3
± ± ± ± ± ±
1 0 0 1 1 1
PIT (°C)
pH
Osm (mOsm/ kg)
80 80 80 78 78 78
7.24 7.27 7.34 7.33 7.31 7.32
326 318 301 296 298 305
IDE loading capacity of the SLN under investigation has been previously determined19 and was SLN A 0.7% w/w, SLN B and SLN C 1.1% w/w. SLN C particle size was lower compared to SLN A and B, likely due to a different packing of the surfactant and cosurfactant molecules at the interface owing to the different structure of the primary surfactants: isoceteth20 has a branched acyl chain while oleth-20 and ceteth-20 have linear acyl chains, unsaturated and saturated respectively. Surfactant lipophilicity did not seem to play a significant role in determining particle size since all the surfactants showed similar hydrophilic lipophilic balance (HLB) values (oleth-20 15.3, isoceteth-20 15.5, ceteth-20 15.7) and no correlation was observed between SLN particle size and surfactant HLB. As shown in Table 2, IDE incorporation into SLN B caused a slight decrease of particle size while no significant effect was observed upon IDE addition to SLN A and C. Previous studies performed loading coenzyme Q10 into SLN showed that this compound was in part homogeneously dispersed within the SLN matrix and in part arranged in separate nanoaggregates.20 Since IDE is structurally related to coenzyme Q10, a different status of IDE into the solid lipid matrix could be expected, leading to different interactions with the surfactant layer and then to different changes of particle size. These interactions could depend on surfactant lipophilicity and/or structure. IDE
■
RESULTS AND DISCUSSION SLN Characterization and Stability. Physicochemical properties of unloaded and IDE loaded SLN were similar to those previously reported.14 TEM analyses of the SLN under investigation showed spherical particles with no evident sign of aggregation (Figure 2). All the tested SLN provided similar images; therefore we reported only one picture as an example. 2536
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
being a lipophilic active agent (Log P 3.49, calculated using Advanced Chemistry Development Software Solaris V. 4.67), its ability to penetrate the tail group region of the surfactant layer, owing to hydrophobic interactions, could affect SLN curvature radius at different extent. Therefore, a lower IDE interaction with the least lipophilic surfactant (ceteth-20) could be expected with no change of particle size. On the contrary, our results showed a decrease of particle size upon addition of IDE to SLN B, thus suggesting that in our experiments the structure of the surfactant may play an important role in determining surfactant/active agent interactions. As reported in the literature,21 the HLB temperature (or PIT temperature) is predictive of emulsion-based formulation stability. Therefore, formulations prepared with surfactants showing similar HLB could be expected to have similar PIT values and similar stability. PIT values of all the SLN investigated in this study were very close, but SLN stability was different. As shown in Figure 3, during storage at room
The calorimetric behavior of cetyl palmitate (CP) and that of the SLN investigated (unloaded SLN A) is shown in Figure 4.
Figure 4. Calorimetric curves, in heating mode, of unloaded SLN A and cetyl palmitate.
The calorimetric curve of CP bulk exhibited a broad peak at about 39 °C and a main peak at about 50.50 °C while that of unloaded SLN A exhibited a shaped peak at about 42 °C and a shoulder at 38 °C. Similar behaviors were observed for all other unloaded and IDE loaded SLN. The melting peak of loaded or unloaded SLN occurred at a temperature about 12−14 °C lower than the bulk CP, suggesting that CP located in the core of the SLN was in the solid state. As reported in the literature,25 these results confirmed that solid lipid nanoparticles were prepared. We used DSC to characterize unloaded and loaded SLN as well as to study their interaction with biomembrane models. Calorimetric thermograms of unloaded and loaded SLN are shown in Figure 5. Unloaded SLN are characterized by a large shoulder at about 32 °C, a main peak at 42 °C and a very small shoulder at 46.50 °C. The loading of IDE into these SLN produces marked modification in the SLN calorimetric thermograms. With regard to SLN A and SLN B, the presence of IDE causes the decrease of the large shoulder at low temperature and the small shoulder at high temperatures turns into a well evident peak. To ascertain that the small peak obtained at high temperature can be attributed to the presence in the investigated sample of IDE loaded into SLN, we carried out three different cross-check experiments. In particular, DSC runs were made on IDE as solid, hydrated IDE and IDE which underwent the same procedure used for the SLN preparation. Solid IDE showed a well-defined melting peak at 53.19 °C, whereas IDE hydration shifted the peak at 45.73 °C. On the contrary, the thermogram of IDE subjected to the same procedure used for SLN preparation did not show any melting peak but a flat line in the investigated temperature range (data not shown). From these cross-check experiments we can conclude that the peak at 46.50 °C can be assigned to the SLN included IDE. IDE loaded SLN C (Figure 5) show a small shoulder at 32 °C, and two peaks centered at 40.99 and 42.57, respectively. These results suggest that IDE is incorporated in the SLN
Figure 3. Particle size of unloaded SLN A-C and IDE loaded SLN A1C1 during storage at rt for 2 months.
temperature for two months, the size of IDE loaded SLN A increased, thus suggesting the occurrence of aggregation phenomena leading to the growing of mean size. Since differences in ζ-potential values were too small to explain the particle increase of SLN A1, the lower stability of IDE loaded SLN A could be attributed to the structure of the surfactant used to prepare these SLN. Having isoceteth-20, the surfactant used to prepare SLN A, a branched acyl chain, a lower intercalation of IDE between the tail group region of the surfactant layer could occur, resulting in a looser packing of the surfactant layer and an increase of aggregation phenomena. Stability tests, performed at room temperature and 37 °C, did not show any significant change of pH and osmolarity values (data not shown). During storage at 37 °C, less stability in terms of particle size for all the formulations tested was observed (data not shown). As reported in the literature,22 less stability at higher temperature could be expected due to an increase of energy into the system leading to particle aggregation and size growth. Differential Scanning Calorimetry Analysis. Structural alterations of materials are accompanied by heat exchanges, e.g., uptake of heat during melting or emission of heat during crystallization. Differential scanning calorimetry is able to measure these heat exchanges during controlled temperature programs and allows one to draw conclusions on the structural properties of a sample. As reported in the literature,22,23 DSC analysis can be used to determine the physical state of the core lipid in SLN. The nanocrystalline size of the lipids in the SLN is thought to be responsible for the lowering of the melting peak temperature of the lipid core compared to that of bulk lipid.24 2537
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
Figure 6. Calorimetric curves, in heating mode, of unloaded SLC C put in contact with IDE.
Figure 5. Calorimetric curves, in heating mode, of unloaded and IDE loaded SLN.
structure and causes the split of the main peak into two peaks. The presence of IDE in these SLN was confirmed by a simple experiment. In fact, solid IDE was weighed in the calorimetric pan, 100 μL of unloaded SLN C was added and the sample was submitted to the same procedure used to analyzed unloaded and loaded SLN. The thermograms, shown in Figure 6, are compared with those of unloaded SLN C and IDE loaded SLN C. It is clearly evident that the thermograms of SLN put in contact with IDE do not change and are quite different from that of loaded SLN proving that the double peak present in loaded SLN is due to the presence of IDE inside the SLN; in addition it proves that the simple contact does not permit the incorporation of IDE into SLN. At this point, we wanted to evaluate the interactions of IDE with the MLV. Therefore, we prepared MLV containing three different amounts of IDE: the same contained in the SLN (molar fraction 0.06 with respect to MLV), half (molar fraction 0.03) and double (molar fraction 0.12), and submitted them to calorimetric analysis. The thermograms, shown in Figure 7, indicate that IDE affects the behavior of MLV: the pretransition peak is lost, and the main peak shifts toward lower temperature and broadens. The effect is dependent on the molar fraction of IDE: the greater the molar fraction, the bigger the effect. At molar fraction 0.12, IDE causes a phase separation meaning that regions rich of IDE and regions poor of IDE exist in the MLV. These results are in agreement with those observed by other research groups.26,27 Then, we evaluated the ability of SLN to act as carrier for IDE and to permit its entry into the cells, studying the interaction of unloaded SLN and IDE loaded SLN with biomembrane models made of DMPC MLV. In
Figure 7. Calorimetric curves, in heating mode, of MLV prepared without and with different molar fractions of IDE.
particular, SLN were put in contact with MLV at 37 °C and the samples were analyzed immediately after the contact and at hourly intervals. The interaction between SLN and MLV was assessed through the changes of SLN and MLV thermograms. In Figure 8A−C, the thermograms are compared with those of the two samples put in contact (unloaded SLN and MLV). In the MLV calorimetric thermogram, two signals exist which are associated with two phase transitions: the pretransition, at about 17 °C, and the main transition, at about 25 °C. The 2538
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
related to unloaded SLN. In the second scan thermogram, the pretransition disappears and the main peak of MLV splits into two peaks. It is clear evidence of the entry of the SLN into the MLV and that SLN do not localize uniformly into MLV but probably localize in the outer bilayers of MLV; the peak related to unloaded SLN remains unchanged, suggesting that SLN maintain their structure. In subsequent scans, the two peaks of MLV merge in a unique broad peak at a temperature lower than that of the first scan. The signal related to SLN does not undergo major changes but only loses the shoulder. It indicates that SLN goes into the MLV and that, as the incubation time increases, SLN moves from the outer bilayers to the inner bilayers of MLV maintaining almost unaltered structure. A similar behavior is observed for MLV and unloaded SLN B (Figure 8B). The MLV pretransition peak disappears; the MLV main peak splits into a shoulder and a peak which, as the incubation time passes, form one peak at low temperature; the signal related to SLN remains unchanged; then also unloaded SLN B penetrate among the MLV phospholipid bilayers retaining completely their integrity. SLN C maintain almost unaltered structure and produce deep change in the MLV signals (Figure 8C); in fact, the pretransition peak is lost from the second scan; the main peak is preceded by a shoulder, which disappears while the main peak moves toward lower temperature. It indicates that also SLN C can enter the MLV. Then, we proceeded to analyze the interaction between MLV and IDE loaded SLN, and the resulting thermograms were compared with those of MLV, IDE loaded SLN, IDE loaded MLV and unloaded SLN (Figure 9A−C). In the signal belonging to MLV, a large shoulder and a peak are present: the shoulder is relative to SLN containing MLV whereas the peak is relative to MLV which do not yet contain SLN. As the incubation time increases, the shoulder disappears whereas the main peak shifts toward lower temperature and broadens. In the signal belonging to SLN, the broad shoulder at about 32 °C decreases and, then, disappears, and the peak at about 42.40 °C remains unchanged; interestingly the peak at about 46 °C, due to the presence of IDE in the SLN, is absent. These results indicate that loaded SLN go inside the MLV releasing IDE. This is confirmed by another experiment in which IDE loaded SLN were closed in the calorimetric pan and submitted to the same analysis used to study the interaction MLV/loaded SLN but without MLV. The thermograms (data not shown) remained unchanged; in particular, with regard to loaded SLN A and SLN B, the peak at 46 °C is always present; with regard to loaded SLN C the peaks at 41 and 43 °C, respectively, are always present, meaning that SLN retain IDE and release the drug only inside the MLV. The comparison of the calorimetric signals of MLV in the presence of unloaded and IDE loaded SLN emphasizes a different behavior; in fact, when MLV are incubated with unloaded SLN the main peak splits into two peaks but when MLV are put in contact with IDE loaded SLN the peak at lower temperature is substituted by a large shoulder: it can indicate that (i) loaded SLN distribute easily and more quickly in the MLV with respect to unloaded SLN or (ii) SLN go into the MLV where they quickly release IDE, the effect of which is in addition to that of SLN. The second hypothesis is more plausible as SLN release IDE immediately after the contact with MLV (the peak at about 46 °C of loaded SLN is lost as soon as after the contact). In the next step, we wanted to evaluate the uptake of IDE by the MLV without the SLN. For this reason, MLV were put in contact with IDE (molar fraction 0.06) in the calorimetric pan and
Figure 8. Calorimetric curves, in heating mode, of MLV put in contact with (A) unloaded SLN A, (B) unloaded SLN B and (C) unloaded SLN C. For comparison calorimetric curves of the samples put in contact (MLV and unloaded SLN) are shown.
transitions separate three distinct phases: lamellar gel, rippled gel, and liquid crystalline phase.28,29 With regard to the experiment with unloaded SLN A (Figure 8A), the first thermogram (recorded immediately after the contact) shows a small peak at about 17 °C, related to the MLV pretransition, a main peak at about 25 °C, related to the MLV transition, a shoulder at about 32 °C and a big peak at about 42.50 °C, both 2539
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
which interacts with the bilayers. If the total uptake of the drug by MLV occurs, the resulting thermogram should be similar to that of the MLV formed with the same 0.06 molar fraction of IDE. Therefore, the thermograms shown in Figure 10 were
Figure 10. Calorimetric curves, in heating mode, of MLV put in contact with IDE (0.06 molar fraction) at increasing time of incubation. The curves are compared with that of MLV prepared with IDE at 0.06 molar fraction.
compared with that of MLV formed with 0.06 molar fraction of IDE (IDE loaded MLV). These thermograms indicate that IDE has little effect on the MLV: the drug produces only a decrease of the pretransition peak and a light shift of the main peak toward lower temperature than that of MLV without IDE. No curves similar to IDE loaded MLV (0.06 molar fraction) thermogram were obtained, meaning that only a small amount of IDE is taken up by MLV. The results of our research on the interactions between IDE loaded SLN and a biological membrane model pointed out that SLN could be able to penetrate into the biomembrane, thus facilitating IDE permeation into the biomembrane itself. Furthermore, the different surfactants used to obtain these SLN did not seem to affect significantly their ability to interact with the model of biomembrane tested. These findings support the results of previous in vitro transport studies, performed on a model of BBB, suggesting that IDE loaded SLN allowed IDE permeation via a transcellular pathway rather than via a paracellular pathway.14 Therefore, SLN interactions with biomembranes could account for their ability to allow IDE penetration into cell membranes improving its uptake by the cells and, hence, its bioavailability.
■
Figure 9. Calorimetric curves, in heating mode, of MLV put in contact with (A) IDE loaded SLN A, (B) IDE loaded SLN B and (C) IDE loaded SLN C. For comparison calorimetric curves of the samples put in contact (MLV and IDE loaded SLN) and of unloaded SLN are shown.
CONCLUSION The interaction of solid lipid nanoparticles, unloaded and loaded with idebenone, with biomembrane models was evaluated by differential scanning calorimetry, with the aim to have information on the interaction of SLN with cell membranes. Our results put in evidence the ability of the SLN under investigation to penetrate into the phospholipid bilayers of MLV, used as model of biological membranes. Our DSC studies provided clear evidence of the entry of the SLN into the phospholipid bilayer and of a likely localization of these SLN in the outer bilayers of MLV. As the time of contact
submitted to the same calorimetric analysis used to study the interaction between MLV and SLN. To evaluate IDE uptake by MLV bilayers, MLV loaded with 0.06 molar fraction of IDE were used as reference. Since IDE could be taken up by the MLV bilayers, the eventual changes of the thermotropic parameters of MLV could be related to the amount of drug 2540
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
Molecular Pharmaceutics
Article
dynamic structure of lipid model membranes. Biochim. Biophys. Acta 2011, 1808, 55−64. (16) Sarpietro, M. C.; Spatafora, C.; Tringali, C.; Micieli, D.; Castelli, F. Interaction of resveratrol and its trimethyl and triacetyl derivatives with biomembrane models studied by Differential Scanning Calorimetry. J. Agric. Food Chem. 2007, 55, 3720−3728. (17) Castelli, F.; Sarpietro, M. G.; Micieli, D.; Ottimo, S.; Pitarresi, G.; Tripodo, G.; Carlisi, B.; Giammona, G. Differential Scanning Calorimetry Study on Drug Release from an Inulin-based Hydrogel and its Interaction with a Biomembrane Model: pH and Loading Effect. Eur. J. Pharm. Sci. 2008, 35, 76−85. (18) Lopedota, A.; Trapani, A.; Cutrignelli, A.; Chiarantini, L.; Pantucci, E.; Curci, R.; Manuali, E.; Trapani, G. The Use of Eudragit® RS 100/Cyclodextrin Nanoparticles for the Transmucosal Administration of Glutathione. Eur. J. Pharm. Biopharm. 2009, 72, 509−520. (19) Montenegro, L.; Sinico, C.; Castangia, I.; Carbone, C.; Puglisi, G. Idebenone-loaded solid lipid nanoparticles for drug delivery to the skin: In vitro evaluation. Int. J. Pharm. 2012, 434 (1−2), 169−174. (20) Wissing, S. A.; Muller, R. H.; Manthei, L.; Mayer, C. Structural Characterization of Q10-Loaded Solid Lipid Nanoparticles by NMR Spectroscopy. Pharm. Res. 2004, 21, 400−405. (21) Izquierdo, P.; Feng, J.; Esquena, J.; Tadros, T. F.; Dederen, J. C.; Garcia, M. J.; Azemar, N.; Solans, C. The Influence of Surfactant Mixing Ratio on Nano-Emulsion Formation by the PIT Method. J. Colloid Interface Sci. 2005, 285, 388−394. (22) Mehnert, W.; Mäder, K. Solid Lipid Nanoparticles. Production, Characterization and Applications. Adv. Drug Delivery Rev. 2001, 47, 165−196. (23) Muëller, R. H.; Maëder, K.; Gohla, S. Solid Lipid Nanoparticles (SLN) for Controlled Drug Deliverya Review of the State of the Art. Eur. J. Pharm. Biopharm. 2000, 50, 161−177. (24) Westesen, K.; Bunjes, H. Do Nanoparticles Prepared from Lipids Solid at Room Temperature always Possess a Solid Lipid Matrix? Int. J. Pharm. 1995, 115, 129−131. (25) Lee, M. K.; Lim, S. J.; Kim, C. K. Preparation, Characterization and in Vitro Ccytotoxicity of Paclitaxel-Loaded Sterically Stabilized Solid Lipid Nanoparticles. Biomaterials 2007, 28, 2137−2146. (26) Fresta, M.; Ventura, C. A.; Mezzasalma, E.; Puglisi, G. A Calorimetric Study on the Idebenone−Phospholipid Membrane Interaction. Int. J. Pharm. 1998, 163, 133−143. (27) Pignatello, R.; Intravaia, V. D.; Puglisi, G. A Calorimetric Evaluation of the Interaction of Amphiphilic Prodrugs of Idebenone with a Biomembrane Model. J. Colloid Interface Sci. 2006, 299, 626− 635. (28) Walde, P. Preparation of Vesicles (Liposomes). In Encyclopedia of Nanoscience and Nanotechnology; Nalwa, H .S., Ed.; American Scientific Publishers: Stevenson Ranch, 2004; pp 43−79. (29) Yi, Z.; Nagao, M.; Bossev, D. P. Effect of charged lidocaine on static and dynamic properties of model bio-membranes. Biophys. Chem. 2012, 160, 20−27.
between SLN and phospholipid bilayers increased, SLN moved from the outer bilayers to the inner bilayers, maintaining almost unchanged their structure. Loading IDE into these SLN facilitated IDE penetration into the bilayers while free IDE showed only a low ability to interact with this model of biomembranes. These results suggest that these SLN could be regarded as a promising carrier to improve IDE penetration into biological membranes, thus improving its bioavailability and its antioxidant activity.
■
AUTHOR INFORMATION
Corresponding Author
*Department of Drug Sciences, University of Catania, V.le A. Doria 6, 95125 Catania, Italy. Phone: + 39 095 738 4010. Fax: + 39 095 738 4211. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
REFERENCES
(1) Fato, R.; Bergamini, C.; Leoni, S.; Lenaz, G. Mitochondrial Production of Reactive Oxygen Species: Role of complex I and Quinine Analogues. BioFactors 2008, 32, 31−39. (2) Imada, I.; Sato, E. F.; Kira, Y.; Inoue, M. Effect of CoQ Homologues on Reactive Oxygen Generation by Mitochondria. BioFactors 2008, 32, 41−48. (3) Leonarduzzi, G.; Sottero, B.; Poli, G. Targeting tissue oxidative damage by means of cell signaling modulators: The antioxidant concept revisited. Pharmacol. Ther. 2010, 128, 336−374. (4) Abdel Bakya, N. A.; Zaidib, Z. F.; Fatania, A. J.; Sayed-Ahmeda, M. M.; Yaqub, H. Nitric oxide pros and cons: The role of l-arginine, a nitric oxide precursor, and idebenone, a coenzyme-Q analogue in ameliorating cerebral hypoxia in rat. Brain Res. Bull. 2010, 83, 49−56. (5) Uttara, B.; Singh, A. V.; Zamboni, P.; Mahajan, R. T. Oxidative Stress and Neurodegenerative Diseases: A Review of Upstream and Downstream Antioxidant Therapeutic Options. Curr. Neuropharmacol. 2009, 7, 65−74. (6) Schols, L.; Meyer, C. H.; Schmid, G.; Wihelms, I.; Przuntek, H. J. Therapeutic Strategies in Friedreich’s Ataxia. J. Neural Transm., Suppl. 2004, 68, 135−145. (7) Schwarz J.; Weisspapir, M. Idebenone composition for the treatment of neurological disorders U.S. Patent 20100129431, May 27, 2010. (8) Di Prospero, N. A.; Baker, A.; Jeffries, N.; Fischbeck, K. H. Neurological Effects of High-Dose Idebenone in Patients with Friedreich’s Ataxia: a Randomised, Placebo-Controlled Trial. Lancet Neurol. 2007, 6, 878−86. (9) Blasi, P.; Giovagnoli, S.; Schoubben, A.; Ricci, M.; Rossi, C. Solid Lipid Nanoparticles for Targeted Brain Drug Delivery. Adv. Drug Delivery Rev. 2007, 59, 454−477. (10) Kuo, Y. C.; Chen, H. H. Effect of Electromagnetic Field on Endocytosis of Cationic Solid Lipid Nanoparticles by Human BrainMicrovascular Endothelial Cells. J. Drug Targeting 2010, 18, 447−456. (11) Kaur, I. P.; Bhandari, R.; Bhandari, S.; Kakkar, V. Potential of Solid Lipid Nanoparticles in Brain Targeting. J. Controlled Release 2008, 127, 97−109. (12) Joshi, M. D.; Muller, R. H. Lipid Nanoparticles for Parenteral Delivery of Actives. Eur. J. Pharm. Biopharm. 2009, 71, 161−172. (13) Montenegro, L.; Campisi, A.; Sarpietro, M. G.; Carbone, C.; Acquaviva, R.; Raciti, G.; Puglisi, G. In Vitro Evaluation of Idebenoneloaded Solid Lipid Nanoparticles for Drug Delivery to the Brain. Drug Dev. Ind. Pharm. 2011, 37, 737−746. (14) Montenegro, L.; Trapani, A.; Latrofa, A.; Puglisi, G. In Vitro Evaluation on a Model of Blood Brain Barrier of Idebenone-loaded Solid Lipid Nanoparticles. J. Nanosci. Nanotechnol. 2012, 12, 330−337. (15) Basso, L. G. M.; Rodrigues, R. Z.; Naal, R. M. Z. G.; CostaFilho, A. G. Effects of the antimalarial drug primaquine on the 2541
dx.doi.org/10.1021/mp300149w | Mol. Pharmaceutics 2012, 9, 2534−2541
