Sequestration of amitriptyline by liposomes

Journal of Colloid and Interface Science 300 (2006) 7–19 www.elsevier.com/locate/jcis

Sequestration of amitriptyline by liposomes Marissa S. Fallon, Anuj Chauhan ∗ Department of Chemical Engineering, University of Florida, Gainesville, FL 32611, USA Received 7 October 2005; accepted 27 March 2006 Available online 3 April 2006

Abstract We study the uptake of amitriptyline, which is a common cause of overdose-related fatalities, in aqueous solutions by 1,2-dimyristoyl-snglycero-3-phosphocholine (DMPC) liposomes and liposomes composed of a mixture of DMPC and 1,2-dioleoyl-sn-glycero-3-[phospho-rac(1-glycerol)] (DOPG) lipids. The effect of drug concentration, liposomal charge, pH, salt, and protein presence on the drug uptake is investigated using two different methodologies, a precipitation and a centrifugation method. Furthermore, the time scale of the drug uptake is studied through qualitative observations at high pH and through conductivity measurements at neutral pH and found to be <5 s. The results of the quantitative studies show that the fractional drug uptake decreases with increasing drug concentration, and for a given concentration it increases with the pH and decreases in the presence of salt. We find that a larger amount of drug is sequestered by negatively charged liposomes (those containing DOPG) than liposomes with no net charge (DMPC). We speculate that the mechanism of drug uptake is due to both electrostatic interactions as well as hydrophobic effects. The fractional uptake by DMPC:DOPG in a 70:30 ratio is as high as 95% in water and about 90% in physiological buffer. The fractional uptake is also measured in presence of 2% (w/w) bovine serum albumin (BSA), which is approximately the protein concentration in the intercellular fluid. In presence of protein the fractional uptakes by 70:30 DMPC:DOPG liposomes and 50:50 DMPC:DOPG liposomes are 82 and 90%, respectively, at 125 µM drug amitriptyline. In the absence of liposomes, 67% of the drug is taken up by the protein in a 2% (w/w) BSA, 125 µM amitriptyline solution. Thus, addition of 50:50 DMPC:DOPG liposomes reduces the free drug concentration by a factor of about 3.5, making them attractive candidates for drug detoxification. © 2006 Elsevier Inc. All rights reserved. Keywords: Liposomes; Drug; Amitriptyline; DMPC; DOPG; Absorption; Adsorption; pH; Protein; Albumin; Drug overdose treatment

1. Introduction Nanoparticles, such as liposomes, can potentially separate compounds from aqueous solutions by either adsorption on the surface or absorption into the particle. Such separations can be important in a variety of applications, such as removal of toxic substances from the blood stream. Their small size and large surface area make nanoparticles particularly advantageous for drug overdose treatment because they can flow through even the smallest capillaries and have the potential to rapidly sequester large quantities of drug. Nanoparticles that are able to sequester drug in vivo can reduce the free drug concentration in the body and therefore, are possible drug overdose treatments [1,2]. * Corresponding author. Fax: +1 352 392 9513.

E-mail address: [email protected] (A. Chauhan). 0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2006.03.065

Nanoparticles such as microemulsions [1], core–shell nanoparticles [3], microgels, liposomes [4,5], and nanotubes may be used for detoxification. Varshney et al. have used Pluronic-based oil-in-water microemulsions to reduce the free concentration of bupivacaine, a local anesthetic drug, and have found that these microemulsions can extract up to 90% of bupivacaine from normal saline solution [1]. Further studies have demonstrated the ability of these microemulsions to sequester drug in blood and whole animal systems. Underhill et al. have synthesized oil-in-water microemulsions with a polysiloxane/silicate shell at the surface of the oil [3]. These nanocapsules have been shown to sequester the drugs quinoline and bupivacaine from saline, but with less efficiency than the microemulsion used by Varshney et al. Thus, the potential of nanoparticles to reduce free drug concentrations has been demonstrated. Herein, we investigate the potential of liposomes, composed of the lipids 1,2-dimyristoyl-sn-gly-

8

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

cero-3-phosphocholine (DMPC) and 1,2-dioleoyl-sn-glycero3-[phospho-rac-(1-glycerol)] (DOPG), at sequestering the antidepressant drug, amitriptyline. Liposomes may offer significant advantages over other types of nanoparticles when used as drug overdose treatment for several reasons. Liposomes have been widely investigated for drug delivery applications [6–10] and are already currently being used on the market for some of these applications [11]. Thus, liposomes are proven to be biocompatible systems. The bilayer configuration of the lipids in the liposome is similar in structure to that of the cell membrane, making liposomes less recognizable by the body’s immune system. Furthermore, liposomes may be sterically stabilized by coating with inert hydrophilic polymers, allowing for even longer circulation time in vivo, and targeted interactions are possible with attachment of ligands to the liposome’s surface [6,9,10,12]. In this paper, we demonstrate the ability of liposomes to sequester drug molecules. We aim to understand the mechanism of drug sequestration by liposomes, determine the effect of pH and charge on drug uptake by liposomes, and quantify the amount of drug liposomes sequester. We study the separation of amitriptyline, which is a common cause of overdose-related fatalities, from aqueous solutions by dimyristoyl phosphatidylcholine (DMPC) liposomes and by liposomes composed of DMPC and 1,2-dioleoyl-sn-glycero-3-[phosphorac-(1-glycerol)] (DOPG). The DMPC liposomes have no net charge under physiological conditions, whereas the liposomes containing DOPG have a net negative surface charge. The structures of the DMPC and DOPG lipids and the drug amitriptyline are given in Fig. 1.

Some studies on the potential of liposomes for drug overdose treatment have been done by other researchers [4,5]. Petrikovics et al. have used sterically stabilized liposomes containing a water soluble enzyme, phosphotriesterase, to absorb and chemically degrade the toxin, paraoxon [4]. In their study, they showed that administering the encapsulated enzyme prior to administering parathion eliminated gross toxic side effects in mice; however, these experiments involved a specific enzyme– drug reaction, which may not work for other drug types. Deo et al. have investigated liposomes composed of phosphatidylcholine and phosphatidic acid in a 1:1 ratio for their potential to sequester amitriptyline [5]. They have demonstrated the capability of liposomes to sequester amitriptyline in the pH range of 6 to 7.4 and have suggested that amitriptyline may partition into the bilayer of the liposomes. While our study is based on the same ideas as those of Deo et al. in many regards, there are a number of differences. We use liposomes composed of DMPC and also a mixture of DMPC and DOPG lipids, which allows us to investigate the effect of liposome surface charge on the drug sequestration, and we also investigate drug uptake at a wide range of pH values. Furthermore, Deo et al. focused on measurements at drug concentrations higher than 1 mM whereas we have explored concentrations in the range of 60–500 µM. It is shown later in this paper that the drug uptake by liposomes is substantially higher at smaller drug concentrations, and thus, the results for partition coefficients reported by Deo et al. are considerably smaller than those corresponding to the results reported here. Our study also investigates the time scale of drug uptake, both through visual observation at high pH and through conductivity experiments, which was not determined

Fig. 1. Structures of the lipids used to make the liposomes as well as the drug used. (a) 1,2-Dimyristoyl-sn-glycero-3-phosphocholine (DMPC) lipid; (b) 1,2-dioleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (sodium salt) (DOPG) lipid; (c) the drug amitriptyline.

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

by Deo et al. Additionally, Deo et al. did not investigate the effect of proteins on the drug uptake by liposomes, and herein, we have examined drug uptake by liposomes in the presence of the plasma protein albumin. There are other differences in our study compared to those of Deo et al., and these differences become apparent in subsequent sections of this paper. Fisar [13] has investigated the mechanisms of uptake of tricyclic depressants, such as amitriptyline, by lipid bilayers. Their studies have some similarities with those reported here, but there are also a number of differences. First, they studied lipids other than those used in our study. Second, they used large multilamellar vesicles (MLVs) whereas we have use small unilamellar vesicles in this study. Third, we show that the time scales for uptake at high pH are extremely rapid, whereas they reported a time scale of about 30 min for equilibration, which was presumably due to the fact that they used MLVs. Fourth, the systems studied by them had uptake values of only about 60–70% which are not sufficient for detoxification applications, and the systems we have studied have uptakes as high as 90–95%. Fifth, while Fisar has measured the concentration dependence of drug uptake for the drug imipramine, they have not reported the concentration dependence of the drug uptake for the drug amitriptyline, which we report herein.

9

sonicator (G112SP1 Special Ultrasonic Cleaner, Avanti Polar Lipids, Inc.) at room temperature to form lipid vesicles. More deionized water was then added, such that the lipid concentration became 4 mg/mL, and the lipid suspension was sonicated using a probe sonicator (Fisher Scientific Sonic Dismembrator Model 100) for 40 min at room temperature to reduce the vesicle size. The liposomes were filtered using a 0.45-µm filter, and then sized by Brookhaven particle size analyzer to obtain the effective diameter (d65 ), which was about 40–45 nm. 2.3. Preparation of buffers Solutions of 100 mM pH 9 buffer were prepared by dissolving bicine in water such that a 111.11 mM concentration was obtained. The solution was then titrated to pH 9 with 1 M NaOH and diluted with water to 100 mM bicine. Solutions of 100 mM pH 10.7 were prepared by preparing an aqueous solution of 46.3 mol% sodium monobasic monohydrate and 53.7 mol% sodium tribasic dodecahydrate. Solutions of 100 mM pH 12 were prepared by dissolving sodium phosphate monobasic monohydrate and sodium phosphate tribasic dodecahydrate in water. The solution was then titrated to pH 12 with 1 M NaOH and diluted with water to 100 mM.

2. Materials and methods

2.4. Drug uptake experiments

2.1. Materials

Below we describe two different methods to quantify drug uptake by the liposomes. The first of these methods, referred to as the precipitation method, provides visual evidence of drug uptake and also allows us to estimate the time scale for drug sequestration; however, it can only provide quantitative information on drug uptake at the solubility limit of the drug at high pH. The second method of quantification, referred to as the centrifugation method, is not restricted to the solubility limit of the drug or to high pH, and thus, this method is used to measure the drug uptake at concentrations other than the solubility limit and at lower pH values. This method cannot determine the time scale of uptake if it is rapid because the centrifugation step takes about 10–20 min. We note that in each case, amitriptyline is added to the solutions in the hydrochloride form (AHCl), which immediately ionizes to the charged form of amitriptyline (AH+ ) once in solution. The AH+ form then establishes equilibrium with the hydrophobic base form (A), as dictated by the pH of the solution. The fraction of the base form near neutral pH is less than 1% near neutral pH and increases to 99% near pH 11.5.

Methanol, chloroform, bicine, sodium hydroxide, Dulbecco’s phosphate buffered saline (PBS) without calcium chloride and magnesium chloride, bovine serum albumin (BSA), and amitriptyline hydrochloride were purchased from Sigma Aldrich. Sodium monobasic monohydrate, sodium tribasic dodecahydrate, and 13 mm, 0.45-µm nylon filters were purchased from Fisher Scientific. The lipids 1,2-dimyristoyl-snglycero-3-phosphocholine (DMPC), in powder form, and 1,2dioleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (sodium salt) (DOPG), dissolved in chloroform (25 mg/mL), were purchased from Avanti Polar Lipids, Inc. 2.2. Liposome preparation Liposomes containing DMPC lipid (4 mg/mL) as well as liposomes containing a mixture of DMPC and DOPG (4 mg/mL) were prepared using a reverse phase evaporation procedure. For the liposomes containing only DMPC, the lipid was dissolved in a 9:1 mixture (by volume) of chloroform:methanol, such that a 10 mg/mL concentration was obtained. For the liposomes containing both DMPC and DOPG, the DMPC was first dissolved in a 9:1 mixture of chloroform:methanol such that a 10 mg/mL concentration was obtained, and DOPG (25 mg/mL chloroform) was then added such that a 70:30 molar ratio of DMPC:DOPG was obtained. The organic solvent was then evaporated under a stream of nitrogen. After an even and uniformly dried lipid film was obtained, the dried lipid layer was hydrated in deionized water, such that the lipid concentration was 40 mg/mL, and the hydrated lipid was sonicated in a bath

2.5. Visual evidence of uptake and quantification of uptake at the solubility limit by the precipitation method As mentioned above, amitriptyline exists in both a charged, hydrophilic form (AH+ ) and an uncharged hydrophobic form (A). The pKa value of amitriptyline is 9.4, and at high pH amitriptyline exists mostly in the hydrophobic form. Since the hydrophobic form of the drug in relatively insoluble in water, at high pH the drug will form a precipitate in aqueous solutions, causing these solutions to become cloudy in appearance. We

10

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

have performed experiments in which drug was mixed in water above the solubility limit in a high pH buffer and liposomes were then added to these cloudy dispersions. Depending on the concentration and the amount of liposomes added, in some cases, the dispersions changed from cloudy to clear in appearance after the addition of liposomes. This visual observation provides direct evidence of drug uptake by liposomes and also a measure of the time required for the uptake. Some drug dispersions stayed cloudy even after liposome addition because the amount of un-sequestered drug was above the solubility limit of the drug. In order to obtain quantitative measurement of drug uptake from these experiments, after complete precipitation (ca. 20 h), the drug dispersions were filtered with a 0.45-µm filter to remove the precipitated drug. The absorbances of the filtered solutions, which were visually clear, were measured with a UV–vis spectrophotometer over the wavelength range of 270 to 290 nm. Control solutions were also prepared by adding water, rather than liposomes, to the high pH drug solutions, and the absorbances of these solutions were measured using same methodology. The absorbances of the high pH buffers and the absorbances of the liposomes in the buffers were also measured. The experiments described above were conducted at pH 9, 10.7, and 12 with 9% and 18% (by volume) DMPC liposome and at pH 10.7 using 9% and 18% (by volume) 70:30 DMPC:DOPG liposomes. It should be noted that the original liposome formulations contain 4 mg lipid/mL, as described above, and thus, the solutions containing 9 and 18% liposomes by volume have concentrations of 0.36 and 0.72 mg lipid/mL, respectively. In these experiments, aqueous solutions of amitriptyline were added to high pH buffer, followed by the addition of either liposomes or water (for control). The samples were prepared such that the amitriptyline concentrations were 0.23, 0.45, 0.92, and 1.82 mM amitriptyline and contained either 9 or 18% liposomes (or water in the case of the control) by volume. 2.6. Uptake measurements by the centrifugation method Liposomes were added to solutions of amitriptyline, such that the volume of the liposomal dispersion (containing 4 mg lipid/mL, as described above) was 9% of the total solution volume (giving a final concentration of 0.36 mg lipid/mL). The solutions were then ultracentrifuged at 3000 rpm for 10 min in a vial that contained a YM30 filter (30,000 molecular weight cutoff). Light scattering was used to verify that the liposomes were not present in the filtrate. To minimize the effect of any leaching components from the filter, all of the filters were rinsed with DI water at 3000 rpm for 20 min prior to their use in these experiments. The concentration of the drug in the filtrate (free drug concentration), was measured using either HPLC or UV–vis spectrophotometry, and the fraction of drug uptake by the liposomes was then calculated. Experiments were conducted at neutral pH, in pH 7.4 buffer, in pH 10.7 buffer, and at high pH with the pH adjusted by addition of NaOH. Additionally, some experiments were conducted in the presence of albumin and in the presence of NaCl.

2.7. Conductivity measurements The conductivities of aqueous solutions of amitriptyline over the concentration range of 31.25 to 500 µM were measured using a Oaklon CON110 Conductivity Meter. Additionally, conductivity measurements were made for amitriptyline solutions in the presence of 9% DMPC liposomes and 9% 70:30 DMPC:DOPG liposomes. 3. Results and discussion 3.1. Determination of time scale of uptake 3.1.1. Changes in visual appearance after liposome addition Fig. 2 shows a sequence of images that show the transition in the visual appearance of a cloudy drug dispersion of 0.45 mM amitriptyline at pH 12 after the addition of 18% DMPC liposomes. The cloudy dispersion turns clear approximately 5 s after addition of liposomes. This clearly demonstrates that the liposomes sequester the drug rapidly at high pH. Not all of the dispersions became clear upon addition of liposomes, and Table 1 summarizes which dispersions turned clear and which remained cloudy upon addition of liposomes. For the 18% liposomes concentrations, the 0.23 and 0.45 mM drug dispersions turned clear after DMPC liposome addition at all pH values tested and after 70:30 DMPC:DOPG liposome addition at pH 10.7. For the 9% liposome concentrations, the 0.23 mM drug dispersion turned clear after DMPC liposome addition at all pH values tested, whereas both the 0.23 and 0.45 mM drug dispersions turned clear after 70:30 DMPC:DOPG liposome addition at pH 10.7. This suggests that the 70:30 DMPC:DOPG liposomes are sequestering more drug than the DMPC liposomes. 3.1.2. Conductivity measurements The conductivity of aqueous solutions of amitriptyline was measured after the addition of 9% DMPC liposomes, after the addition of 9% 70:30 DMPC:DOPG liposomes, and in the absence of liposomes. Upon the addition of liposomes to the drug solutions, the conductivity immediately changed and remained independent of time after the change. This implies that the sequestration of drug by liposomes is rapid, as was visually seen in the high pH experiments described above. To understand the interaction between the liposome and the drug, we calculated the difference () between (1) the sum of the conductivity from the 9% liposomal dispersion and the drug solution of a given concentration and (2) the conductivity of the liposome–drug solution with the same liposome type as the 9% liposomal dispersion and the same drug concentration. This difference is shown in Tables 2 and 3 for the DMPC and 70:30 DMPC:DOPG liposomes, respectively. This difference was positive for all the concentrations studied, which implies that there is some uptake of the drug by the liposome. It should be noted that the conductivity of the drug solution mainly arises from the high mobility chloride ions and the contribution from the acid form (AH+ ) is small. Thus, uptake of the drug by the liposomes is not expected to make significant changes in the conductivity. This explains

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

(a)

(b)

(c)

(d)

11

Fig. 2. Qualitative look at drug uptake by liposomes. Both vials contain a solution of amitriptyline at pH 12. The vial on the left is the control solution, to which water (18% by volume) has been added. DMPC liposomes (18% by volume) are added to the vial on the right. The amitriptyline concentration after liposome or water addition is 0.45 mM. Initially, both solutions are cloudy in appearance. The water has been added to the control solution prior to the pictures being taken, and no visible change in appearance was observed. A change in appearance from cloudy to clear is observed in the vial on the right after DMPC liposomes are added. The pictures show the solutions: (a) 0.9 s, (b) 2.1 s, (c) 3.5 s, and (d) 5.4 s after liposome addition.

Table 1 Summary of the appearance of each solution in the drug uptake experiments after liposome addition DMPC pH 9

70:30 DMPC:DOPG pH 10.7 pH 12

pH 9

9% 0.27 mM amitriptyline liposomes 0.45 mM amitriptyline 0.92 mM amitriptyline 1.82 mM amitriptyline

Clear Cloudy Cloudy Cloudy

Clear Cloudy Cloudy Cloudy

Clear Cloudy Cloudy Cloudy

Clear Clear Cloudy Cloudy

18% 0.27 mM amitriptyline liposomes 0.45 mM amitriptyline 0.92 mM amitriptyline 1.82 mM amitriptyline

Clear Clear Cloudy Cloudy

Clear Clear Cloudy Cloudy

Clear Clear Cloudy Cloudy

Clear Clear Cloudy Cloudy

why the difference  is small. It is difficult to isolate the contributions to conductivity from each of the ions present in the solutions, and thus, we have not used the conductivity measurements to obtain quantitative information about the amount of drug uptake by the liposomes using these conductivity measurements. These measurements have, however, confirmed that there is some interaction between the liposomes and the drug and that the uptake of the drug by the liposomes occurs rapidly, even near neutral pHs.

Table 2 Difference between (1) the sum of the conductivities of a 9% DMPC liposomal dispersion in water and an aqueous amitriptyline solution at a given concentration and (2) the conductivity of a solution containing 9% DMPC liposomes and amitriptyline at the same concentration Amitriptyline concentration (µM)

 (µS)

31 62 123 247 496

1.09 1.01 1.01 1.29 1.51

3.2. Quantitative analysis 3.2.1. Precipitation method In our quantitative analysis of the data obtained from the drug uptake experiments using the precipitation method, we assume that the absorbances of each component in the solution (i.e., buffer, liposome, and drug) are additive. To check the validity of this assumption, we measured the absorbances of the drug and that of liposomal dispersions at various concentrations, and then measured the absorbances after adding liposomes to a drug solution. The measured absorbances were compared with the calculated absorbances assuming additivity

12

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Table 3 Difference between (1) the sum of the conductivities of a 9% 70:30 DMPC: DOPG liposomal dispersion in water and an aqueous amitriptyline solution at a given concentration and (2) the conductivity of a solution containing 9% 70:30 DMPC:DOPG liposomes and amitriptyline at the same concentration Amitriptyline concentration (µM)

 (µS)

31 155 279 526

0.20 0.92 8.78 28.54

Fig. 3. Check of the absorbance additivity assumption. Absorbance vs wavelength for DMPC liposomes in water (9 and 18% by volume), 0.23 mM amitriptyline in water, 0.23 mM amitriptyline in water containing DMPC liposomes (9 and 18% by volume), and the calculated absorbance values of the 0.23 mM amitriptyline in water containing DMPC liposomes (9 and 18% by volume).

and the results of two such comparisons are shown in Fig. 3. It can be seen in this figure that our calculated predictions for the absorbance values of the 0.23 mM amitriptyline solutions containing 9 and 18% DMPC liposomes matched the measured values. Thus, our assumption that absorbances are additive is valid. 3.2.1.1. Absorbance results for control solutions The controls contain drug at concentrations above the solubility limit, and thus, the drug in excess of the solubility limit precipitates and is removed during filtration. In the filtrate the uncharged hydrophobic form (A) must be at the solubility limit and the charged form (AH+ ) must be at the equilibrium concentration, which is dictated only by the ionization equilibrium of the drug. Thus, the concentration of both the charged and the uncharged forms should be independent of the initial drug concentration and should only depend on the pH. Accordingly, the absorbance of the filtrates of the controls should be independent of the starting drug concentration. These results are shown in Fig. 4 and demonstrate that, within experimental error, the absorbances of the controls are independent of the initial drug concentration. Some differences in the absorbances in controls at different concentrations may be attributed to the fact that some small precipitate particles may have passed through the filter. The differences between the absorbances from different concentrations are larger at higher pHs because the actual absorbances are ex-

tremely small and the experiments are thus susceptible to larger errors. However at these pH values the absorbances from the experiments were significantly larger than those of the controls and thus, errors in measurement of the control absorbances lead to negligible errors in estimation of drug uptake. By subtracting the absorbance of the buffer from the absorbance of the control, and then by using the calibration curve for the drug, the amount of drug in the filtrate can be computed. Based on such a calculation the filtrate from the pH 9 control contains 0.223 mM drug. In a separate experiment the solubility limit of the drug was determined to be 0.21 mM, which, within experimental error, matches the drug concentration in the filtrate. This supports our assertion that the drug concentration in the filtrate is at the solubility limit. Attempts to measure solubility limit at higher pH values did not produce reliable results because of the extremely small solubility values. 3.2.1.2. Absorbance results for the solutions after liposome addition Below we first describe the absorbance results for drug solutions containing both types of liposomes (DMPC and 70:30 DMPC:DOPG) and then quantify the amount of drug sequestered by the liposomes. DMPC liposomes Figs. 5–7 show the absorbances for the samples containing 9 and 18% DMPC liposomes at pH values 9, 10.7, and 12, respectively. For each pH value tested, the 0.23 mM cloudy amitriptyline dispersion turned clear after 9% liposome addition, and the 0.45, 0.92, and 1.82 mM amitriptyline dispersions remained cloudy. At the 18% DMPC liposome concentration, the 0.23 and 0.45 mM dispersions turned clear after liposome addition, while the 0.92 and 1.82 mM dispersions remained cloudy. It can be seen in Figs. 5–7 for each pH and DMPC liposome concentration that all of the dispersions that remained cloudy had approximately the same absorbance values after filtration. This was expected because the total amount of drug in each of these filtered solutions is the same and equal to sum of the drug in the bulk, which is at the solubility limit, and that on/in the liposomes, which is at the equilibrium value. The results also show that the filtrates from the clear solutions had an absorbance value less than the absorbances of the filtrates from the systems that remained cloudy, which was also expected because the concentration of drug in the bulk for solutions that turned clear is less than the solubility limit. Furthermore, for each pH value at the 18% DMPC liposome concentration, the 0.45 mM solution had approximately twice the absorbance of the 0.23 mM drug solution, once the absorbance of the liposomes in the buffer was subtracted. Since there is twice the amount of drug in the 0.45 mM solution as there is in the 0.23 mM solution, this finding was expected. DMPC:DOPG liposomes Fig. 8 shows the absorbances of the samples containing 9 and 18% 70:30 DMPC:DOPG liposomes at pH 10.7. In this case, the 0.23 and 0.45 mM cloudy dispersions turned clear at both the 9 and 18% liposome concentrations. Similar trends were found with these liposomes as were found in the DMPC liposomes, with the cloudy dispersions having approximately the same absorbance values af-

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

13

Fig. 5. Absorbance vs wavelength for the drug solutions at pH 9 containing (a) 9% DMPC liposomes and (b) 18% DMPC liposomes. Data are expressed as mean ± SD with n = 3.

Table 4 Fraction of drug uptake at different pH values and DMPC liposome concentrations, calculated at λ = 285 nm

Fig. 4. Absorbance vs wavelength for the control solutions: (a) pH 9 control, (b) pH 10.7 control, (c) pH 12 control. Data are expressed as mean ± SD with n = 3.

ter filtration and the 0.45 mM solutions having approximately twice the absorbance of the 0.23 mM solutions after subtracting the absorbance of the liposomes in the buffer. The liposomes containing DOPG sequestered more drug than those containing only DMPC. This was evident from the visual observation that the 9% liposome 0.45 mM amitriptyline solution turned clear when 70:30 DMPC:DOPG liposomes were added, but remained cloudy when DMPC liposomes were added. Additionally, even though the absorbances of the DMPC liposomes were about the same as the absorbance values of the 70:30 DMPC:DOPG liposomes, the absorbance values of the filtered solutions containing 70:30 DMPC:DOPG liposomes were higher than the absorbance values of the filtered solutions containing the DMPC liposomes at the same pH. 3.2.1.3. Quantitative evaluation of drug uptake by liposomes As shown above, it is reasonable to assume that the absorbances of each component in the solutions (liposomes, drug, and buffer) are additive, and thus, by subtracting the contribution from the lipids, the buffer and the control (which accounts

9% liposomes 18% liposomes

pH 9

pH 10.7

pH 12

0.71 0.81

0.94 0.98

0.92 0.91

for the amount of drug in the bulk), the absorbance from the drug sequestered by the liposomes can be determined. The absorbance of the drug sequestered by the liposomes for pH values 9, 10.7, and 12 for DMPC liposome concentrations of 9 and 18% liposome by volume are shown in Fig. 9. After determining the absorbance of the sequestered drug, the amount of drug taken up by the liposomes can be determined by using the absorbance spectrum of the drug. Additionally, the fraction of drug sequestered by the liposome (f) at each pH and liposome concentration can be calculated as the ratio of the drug present sequestered by the liposome and the total drug present in the solution system after filtration. The results for the fractional uptake as calculated at λ = 285 nm are shown in Table 4. As is shown in Table 4, approximately 70 to almost 100% of the drug present in the solution is sequestered by the liposomes. The absorbance of the drug sequestered by the liposome at the solubility limit of the drug at pH 10.7 is also found for 9 and 18% 70:30 DMPC:DOPG liposomes and is shown in Fig. 10, along with the absorbance of the drug sequestered by the liposome at the solubility limit of the drug at pH 10.7

14

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Fig. 6. Absorbance vs wavelength for the drug solutions at pH 10.7 containing (a) 9% DMPC liposomes and (b) 18% DMPC liposomes. Data are expressed as mean ± SD with n = 3.

Fig. 7. Absorbance vs wavelength for the drug solutions at pH 12 containing (a) 9% DMPC liposomes and (b) 18% DMPC liposomes. Data are expressed as mean ± SD with n = 3.

for the 9 and 18% DMPC liposomes. The 70:30 DMPC:DOPG solutions have higher absorbances than those solutions containing DMPC liposomes, and we conclude that more drug is sequestered by the liposomes containing DOPG. Additionally, we calculate the fraction of drug uptake for the 70:30 DMPC:DOPG liposomes at pH 10.7 for λ = 285 nm. This value is 0.99 for both the 9 and 18% liposome concentrations, and thus, practically all of the drug present in the solution is sequestered by the liposome. 3.2.2. Centrifugation method The results of the precipitation method described above clearly demonstrate that liposomes can sequester a large amount of drug very rapidly; however, the above method can only be used to measure the amount of drug sequestered by the liposomes at high pH and at the solubility limit of the drug. The uptake can be measured under other conditions by centrifuging the liposome–drug solutions as described above, and then measuring the drug concentration in the filtrate. We refer to this method as the centrifugation method. Below we report the drug uptake measurements obtained by using the centrifugation method. These experiments focused on studying the effect of drug concentration, protein addition, pH, and salt concentration on drug uptake. 3.2.2.1. Effect of concentration on drug uptake In the results reported below, the fraction of drug sequestered by the liposomes is determined by calculating the ratio of the drug con-

Fig. 8. Absorbance vs wavelength for the experiment solutions at pH 10.7 with 70:30 DMPC:DOPG liposomes at (a) 9% liposomes by volume and (b) 18% liposomes by volume. Data are expressed as mean ± SD with n = 3.

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Fig. 9. Absorbance of drug sequestered by the liposome at the solubility limit of the drug for three different pH values (9, 10.7, and 12) and two different DMPC liposome concentrations (9 and 18% by volume). Data are expressed as mean ± SD with n = 3.

15

Fig. 11. Fractional drug uptake by 9% DMPC liposomes vs initial amitriptyline concentration in aqueous drug solutions (n = 2–4).

Fig. 12. Fractional drug uptake by 9% 70:30 DMPC:DOPG liposomes vs initial amitriptyline concentration in aqueous drug solutions (n = 2–4 for day 1 data, n = 1 in day 2 and day 3 data).

Fig. 10. Absorbance of drug sequestered by the liposome at the solubility limit at pH 10.7 for two different DMPC and 70:30 DMPC:DOPG liposome concentrations (9 and 18% by volume). Data are expressed as mean ± SD with n = 3.

centration in the filtrate and the concentration in the original drug solution. Drug uptake in water 9% DMPC The fractional drug uptake by 9% DMPC liposomes is plotted as a function of the initial drug concentration in Fig. 11. The fractional uptake decreases on increasing the concentration, which is expected because of saturation effects; however, the fractional uptake levels off above a concentration of about 200 µM. This behavior is also evident in other results shown below. The uptake is only about 30% at 62.5 µM, which is the smallest concentration in the study. This small amount of uptake implies that DMPC liposomes may not be suitable for drug overdose treatment; however, it is noted though that the uptake shows an increasing trend on reducing the concentration. Thus, the uptake may be significantly higher at the physiological concentrations of about 1 µM. It is also noted that these experiments were not buffered, and thus, the pH of the solutions may decrease with increasing drug concentration due to conversion of the acid form to the base form. However, we expect this change in pH to be minimal because the fraction of the base form is minimal near neutral pH.

9% DMPC:DOPG The fractional drug uptake by 9% 70:30 DMPC:DOPG liposomes is plotted as a function of the initial drug concentration in Fig. 12. The three curves in this figure correspond to fractional uptake measured at different times. The day 1 data corresponds to data measured immediately after addition of liposomes to the drug solution, and the day 2 and day 3 results correspond to data acquired at about 24 and 48 h after mixing the liposomes and the drug solution. It is noted that since the centrifugation and measurement steps take about 30 min, the day 1 data corresponds to time of about 30 min after the liposome addition. The fractional uptake curves for the three days almost overlap which shows that the equilibration between the liposomes and the drug occurs within the first 30 min. This result agrees with the inference drawn above on the basis of the conductivity measurements that the drug uptake by the liposomes is rapid. As with the DMPC liposomes, the fractional uptake by the 70:30 DMPC:DOPG liposomes decreases with increasing drug concentration, and the fractional uptake levels off above a concentration of about 300 µM. The uptake is about 40% at 500 µM and increases to about 98% at 100 µM, which is the smallest concentration in the study. The 70:30 DMPC:DOPG liposomes are thus clearly superior to the DMPC liposomes for drug overdose treatment. The increase in the uptake is perhaps due to electrostatic interaction between the negatively charged DOPG lipids and the positively charged drug molecules. It is also noted that these experiments were not buffered, and thus, the pH of

16

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Fig. 13. Fractional drug uptake by 9% DMPC liposomes vs initial amitriptyline concentration in pH 10.7 buffer (n = 1) Note: The square indicates data obtained using the precipitation method.

the solution may minimally decrease with increasing drug concentration. Deo et al. [5] have also studied the uptake of amitriptyline by liposomes in water. They used liposomes composed of a 1:1 ratio of phosphatidylcholine (PC) and phosphatidic acid (PA) and measured an uptake of about 70% at 1 mM amitriptyline, which was the smallest drug concentration in their study. In our study, the uptake decreases substantially with increasing concentration and decreases from about 95% at 62.5 µM to about 40% at 500 µM for 9% 70:30 DMPC:DOPG liposomes. Thus, we would expect an uptake 40% at 1 mM. The uptakes in Deo et al. experiments at similar concentrations were higher than our DMPC:DOPG results because the lipid loading in experiments of Deo et al. was approximately twice the lipid loading that was used in our experiments. Drug uptake in buffer 9% DMPC in pH 10.7 buffer The experimental results reported above show that in the studies conducted at pH values near 7, the DMPC liposomes sequester less than 30% of the drug; however, the experimental results from the precipitation method using UV–vis spectroscopy showed 70–95% uptake by DMPC liposomes at higher pH values. It was thus decided to conduct centrifugation experiments in a pH 10.7 buffer to measure the drug uptake by 9% DMPC liposomes, and these results are shown in Fig. 13. At this pH the fractional uptake is about 95% and is relatively independent of the drug concentration for the concentration regime studied. These results match the results obtained earlier by the precipitation method, which the uptake was determined to be 94% at the solubility limit of the drug. The reasons for the significant differences between drug uptakes at pH 7 and 10.7 will be discussed below. 9% 70:30 DMPC:DOPG in pH 7.4 buffer The uptake by 9% 70:30 DMPC:DOPG liposomes in pH 7.4 buffer (PBS) is similar to that in water, as is seen in Fig. 14. The uptake decreases with increasing concentration, and at 65 µM, which is the smallest concentration explored here, the fractional uptake is about 90%, which, within experimental error, matches

Fig. 14. Fractional drug uptake by 9% 70:30 DMPC:DOPG liposomes vs initial amitriptyline concentration in pH 7.4 buffer (n = 2).

the uptake in water. The uptake levels off at about 60% which is slightly higher than that in water. Effect of salt concentration on uptake by 70:30 DMPC: DOPG liposomes Due to the charge on the DOPG lipids, the drug uptake by the 70:30 DMPC:DOPG liposomes is expected to be partially driven by electrostatic interactions, which are impacted by the salt concentration. In order to determine the role of electrostatic interactions, it was decided to measure the fractional uptake in water after addition of 1 M salt and compare this uptake with that in water. The results, shown in Fig. 15, demonstrate that the presence of salt reduces the drug uptake, but the effect is relatively minor at higher concentrations. 3.2.2.2. Effect of pH on drug uptake The results described above already demonstrate that the drug uptake by the liposomes is affected by pH. Further evidence on the effect of pH is provided in Fig. 16 in which the fractional uptake is plotted as a function of pH at a initial drug concentration of 62.5 µM for 9% DMPC and an initial drug concentration of 250 µM for the 9% 70:30 DMPC:DOPG liposomes. The fractional uptake increases with increasing pH, and the increase is more drastic for the DMPC liposomes. The DMPC lipids carry a net zero charge under physiological conditions but near the pH of 11.25, which is the pKa of the amino group on the DMPC, these lipids become negatively charged [13]. Thus at high pH, electrostatic interaction could be driving the uptake; however, at pH values larger than 11, only a negligible fraction of the drug is in the charged form. This suggests that the increase in uptake at higher pH values is due to the partitioning of the uncharged drug into the lipid bilayer. The increase in uptake for 70:30 DMPC:DOPG liposomes can also be attributed to the same reason. The drug uptake values determined by the centrifugation method at high pH are comparable to the drug uptake values determined using the precipitation method. Fisar et al. have also investigated the pH dependence of amitriptyline uptake, as well as the uptakes of other antidepressant drugs, by liposomes and found trends similar to those obtained here, such that the drug uptake increased with increasing pH [13]. They found uptakes as high as about 40, 20, 60, and 50% at high pH (around pH 12) for liposomes composed of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS), and phosphatidylinositol (PI), respec-

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Fig. 15. Fractional drug uptake by 9% 70:30 DMPC:DOPG liposomes vs initial amitriptyline concentration in water and with the addition of 1 M salt (n = 2–3).

tively. For liposomes composed of a mixture of PC and PI lipids, uptakes as high 70% were achieved at high pH. Our study shows higher uptake values at pH 12 than those found by Fisar et al.; however, Fisar et al. have used multilamellar vesicles as opposed to the small unilamellar vesicles used in our study and conducted their study at 5 nM amitriptyline with about half the lipid loading as was used in our study. 3.3. Mechanism of drug uptake The following are the main features of the amitriptyline uptake by DMPC and 70:30 DMPC:DOPG liposomes: (1) The uptake is very rapid for both types of liposomes, and occurs on a time scale too short to be measured by conductivity or by visual means. (2) The fractional uptake decreases with an increase in drug concentration and then levels off. (3) The fractional uptake for 70:30 DMPC:DOPG decreases on addition of salt but the decrease is minor in the high concentration range where the uptake has leveled off. (4) The fractional uptake increases with an increase in pH for both types of liposomes, but the increase is more significant for the DMPC liposomes. Based on the observations that the uptake by 70:30 DMPC: DOPG liposomes in unaffected by salt concentration in the ‘plateau’ region, i.e., the region where the uptake has leveled off, we attribute the uptake in this region to partitioning of the drug into the bilayer, with its charged group pointing outwards. This mechanism is identical to that postulated by Deo et al. [5]. The observation that the uptake in the small concentration range depends on the salt concentration suggests that the increase in uptake as the concentration decreases below about 300 µM is driven largely by electrostatic adsorption of the drug on the surface. Thus, the reduction in uptake in the small concentration region occurs due to saturation of the available surface area. We expect that the uptake due to partitioning into the bilayer will also saturate and there will be another region at higher concentrations in which the fractional uptake will further decline from the value in the plateau region to zero. The uptake for DMPC

17

Fig. 16. Fractional drug uptake by 9% DMPC for an initial amitriptyline concentration of 62.5 µM and 9% 70:30 DMPC:DOPG liposomes for an initial amitriptyline concentration of 125 µM vs pH, where the pH was adjusted using NaOH (n = 2–3).

liposomes is small near neutral pH because of the lack of electrostatic interactions. Finally, the effect of pH suggests that as the pH increases, there is a larger fraction of the uncharged drug, and thus, the uptake is primarily governed by uptake of the uncharged drug into the annulus. Accordingly, the uptake becomes very large and does not show any sign of saturation for concentrations as large as 300 µM. These mechanisms also explain the high values of drug uptake obtained in the high pH experiments using the precipitation method. 3.4. Effectiveness of liposomes under physiological conditions Since we propose to use liposomes as drug overdose treatments, it is important to determine their efficacy at drug sequestration under physiological conditions in vivo. There are two major differences between the conditions under which our experiments were performed and the conditions in vivo. The first difference is the presence of plasma proteins and other blood components, and the second difference is that the experiments reported here were conducted at concentrations higher than 50 µM, whereas the concentrations experienced in vivo will be less than 1 µM. The results shown above demonstrate that the uptake increases with a reduction in drug concentration, and thus, the values reported here are the lower bounds on the expected uptake at physiological concentrations. The presence of plasma proteins is expected to have a significant effect on the drug uptake by the liposomes. Blood contains about 4% albumin, 2% fibrinogen, and 1% globulins, giving a total plasma protein concentration of about 7% [14]. The intercellular fluid that surrounds the cells in the tissue contains only about 2% protein. In blood, about 95% of amitriptyline is bound to the proteins, and this will impact amitriptyline uptake by liposomes. We investigate the effect of presence of plasma proteins on drug sequestration by liposomes below. 3.4.1. Drug sequestration by liposomes in the presence of protein The fraction of drug sequestered by BSA at an initial drug concentration of 125 µM in PBS was measured using centrifugation and HPLC techniques at a 2% by weight concentration of BSA, which corresponds to the amount of protein present in

18

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Table 5 Fractional drug uptake in the presence of 2% (w/w) albumin for an initial drug concentration of 125 µM in PBS in the absence of liposomes, in the presence of 9% by volume liposomes composed of a 70:30 molar ratio of DMPC:DOPG, and in the presence of 9% by volume liposomes composed of a 50:50 molar ratio of DMPC:DOPG Liposome type

Fraction drug uptake

No liposome 9% 70:30 DMPC:DOPG 9% 50:50 DMPC:DOPG

0.67 0.82 0.90

the intercellular fluid in the tissue. The results are presented in Table 5, along with the results in the presence of 9% by volume 70:30 DMPC:DOPG liposomes and in the presence of 9% by volume 50:50 DMPC:DOPG liposomes (liposomes composed of a 50:50 molar ratio of DMPC:DOPG). As is shown in Table 5, about 67% of the drug was sequestered by the protein, and the uptake increased to 82% and 90% in the presence of 70:30 and 50:50 DMPC:DOPG liposomes, respectively. It is noted that the fractional uptake mentioned here is the total amount of drug sequestered by either liposomes or protein, not purely the uptake by liposomes. It was expected that the 50:50 DMPC:DOPG liposomes would have a higher uptake than the 70:30 DMPC:DOPG liposomes due to the higher amount of negative charge present on these liposomes, which could cause increased electrostatic interactions between the drug and the liposome. The results show that by increasing the amount of negative charge on the surface of the liposome by 20%, the free amount of drug is decreased by about 8% in the presence of 2% albumin. As shown earlier, the uptake by liposomes in the absence of protein at 125 µM was about 65% by 9% 70:30 DMPC:DOPG liposomes in PBS in the absence of protein, and it reduces to 15% (= 82% − 67%) of the total amount in the presence of 2% protein. This reduction is caused by the fact that liposomes have to compete with proteins for binding the drug. It is also noted that addition of the 50:50 DMPC:DOPG liposomes to the 2% protein + 125 µM drug mixture reduces the free drug concentration from 35 to 10%. This reduction by a factor of 3.5 can lead to significant beneficial effects for an overdosed patient. 4. Conclusions In this paper we explore the possibility of using liposomes for treatment of overdoses of the antidepressant drug, amitriptyline. Experiments are performed by using two different approaches to quantify the drug uptake by the liposomes as a function of concentration, pH, salt concentration, and lipid loading. Measurements of conductivity after addition of liposomes to drug solutions show that the uptake of drug is very rapid. The same inference can be drawn on the basis of the experiments at high pH at which the cloudy drug solution becomes clear immediately after liposome addition. The experiments conducted using the precipitation method show visual evidence of drug uptake, prove that the uptake is rapid, show that the uptake scales linearly with the liposome loading, and also provide quantitative data on the amount of drug sequestered by liposomes at

high pH and at the solubility limit of the drug. These experiments are useful for preliminary screening of the nanoparticles, but these are not sufficient to understand the mechanism of drug uptake and to quantify drug uptake under physiological concentrations. These experiments are supplemented with centrifugation experiments in which the free solution in a liposome–drug mixture is drawn through a filter, and the free concentration of drug is measured in the filtrate to determine the drug uptake as a function of concentration, pH and salt concentration. In these studies two types of liposomes were investigated: (1) DMPC liposomes, containing no net charge under physiological conditions and (2) 70:30 DMPC:DOPG liposomes, containing a negative surface charge. From the high pH studies, we determine that at the solubility limit, DMPC liposomes sequester approximately 70–100% of the drug present in the solution, while the liposomes containing DOPG sequester approximately all the drug present in the solution. The drug uptake decreases with an increase in concentration and levels off at a concentration of about 300 µM. The drug uptake increases with an increase in pH and the uptake for 70:30 DMPC:DOPG liposomes reduces with addition of salt. It is speculated that the mechanism of uptake is partitioning into the annulus for concentrations above 300 µM and below that there is additional uptake due to electrostatic binding of the drug to the charged liposome surface. It is important to determine the efficacy of the liposomes at drug sequestration in the presence of complex media such as plasma, where the liposomes have to compete with the plasma proteins for drug binding. Herein, we have determined drug uptake by liposomes in the presence of 2% albumin, which is the protein concentration in the intercellular tissue fluid. We found that at a 2% by weight albumin concentration, the addition of 50:50 DMPC:DOPG liposomes decreases the amount of free drug by a factor of about 3.5. More studies should be done to examine the uptake of drug by liposomes at higher protein concentrations such as those present in the blood and also in the presence of other blood and cellular components. Additionally, it needs to be determined whether these liposomes cross the capillary walls and enter the tissues, because the liposomes will be more effective at sequestering drug inside the tissue than in the blood because of the lower protein concentration. However, inside the tissue the liposomes will need to compete with the cellular components, in particular the ion channels, and these interactions need to be studied further. While other types of liposomes are currently in use for drug delivery applications, it is important that the safety of DMPC and DMPC:DOPG liposomes is established before use as drug overdose treatments. Thus, future work should include in vitro and in vivo safety analysis tests. To determine the efficacy of these liposomal systems in vivo, animal studies should be also performed. While further in vitro and in vivo experiments need to be performed to gauge the effectiveness of the DMPC and DMPC:DOPG liposomes for use in treatment of amitriptyline drug overdoses, the results of this study clearly demonstrate the significant potential of these systems, particularly in light of the fact that liposomes are highly biocompatible and are already being used for other drug delivery applications.

M.S. Fallon, A. Chauhan / Journal of Colloid and Interface Science 300 (2006) 7–19

Acknowledgments The authors acknowledge the financial support of the Engineering Research Center (ERC) for Particle Science and Technology at the University of Florida, The National Science Foundation (NSF Grant EEC-94-02989), and the Industrial Partners of the ERC for support of this research. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Science Foundation. References [1] M. Varshney, T.E. Morey, D.O. Shah, J.A. Flint, B.M. Moudgil, C.N. Seubert, D.M. Dennis, J. Am. Chem. Soc. 126 (2004) 5108–5112. [2] T.E. Morey, M. Varshney, J.A. Flint, S. Rajasekaran, D.O. Shah, D.M. Dennis, Nano Lett. 4 (2004) 757–759.

19

[3] R.S. Underhill, A.V. Jovanovic, S.R. Carino, M. Varshney, D.O. Shah, D.M. Dennis, T.E. Morey, R.S. Duran, Chem. Mater. 14 (2002) 4919– 4925. [4] I. Petrikovics, K. Hong, G. Omburo, Q.Z. Hu, L. Pei, W.D. McGuinn, D. Sylvester, C. Tamulinas, D. Papahadjopoulos, J.C. Jaszberenyi, J.L. Way, Toxicol. Appl. Pharmacol. 156 (1999) 56–63. [5] N. Deo, T. Somasundaran, P. Somasundaran, Colloids Surf. B 34 (2004) 155–159. [6] A. Sharma, U.S. Sharma, Int. J. Pharm. 154 (1997) 123–140. [7] Y. Rui, S. Wang, P.S. Low, D.H. Thompson, J. Am. Chem. Soc. 120 (1998) 11213–11218. [8] J.N. Weinstein, L.D. Leserman, Pharmac. Ther. 24 (1984) 207–233. [9] D.D. Lasic, Trends Biotechnol. 16 (1998) 307–321. [10] D.D. Lasic, D. Papahadjopoulos, Science 267 (1995) 1275–1276. [11] D.D. Lasic, D. Papahadjopoulos, Medical Applications of Liposomes, Elsevier, Amsterdam, 1998. [12] M. Ostro, Liposomes from Biophysics to Therapeutics, Dekker, New York, 1987. [13] Z. Fisar, Gen. Physiol. Biophys. 24 (2005) 161–180. [14] A.C. Guyton, Textbook of Medical Physiology, Saunders, Philadelphia, 1986.