On the Stability of Liposomes and Catansomes in Aqueous Alcohol

The physical stability of liposomal cosolvent systems and the role cosolvents played in the liposome stability have been less systematically studied i...
1 downloads 13 Views 529KB Size
Download PDF
Langmuir 2008, 24, 1695-1700

1695

On the Stability of Liposomes and Catansomes in Aqueous Alcohol Solutions Yu-Min Yang,* Kuo-Chang Wu, Zheng-Lin Huang, and Chien-Hsiang Chang Department of Chemical Engineering, National Cheng Kung UniVersity, Tainan 701, Taiwan ReceiVed June 25, 2007. In Final Form: October 11, 2007 In this work, a systematic study of effects of three cosolvents (methanol, ethanol, and 1-propanol) on the stability of liposomes formed from soybean phosphatidylcholine (PL-90) by a semispontaneous process was carried out. The experimental results revealed that significant enhancement of PL-90 liposome stability could be achieved by cosolvent addition with suitable amounts. A similar phenomenon was also observed for catansomes formed from ion-pair amphiphiles (IPAs) as demonstrated by decyltrimethylammonium dodecyl sulfate (DeTMA-DS) with a comparatively high concentration of 5 mM. In general, with continued increase in the cosolvent concentration, the stability of liposomes and catansomes first increased, reached a maximum, and then decreased. Furthermore, it was realized that the cosolvent effects on the liposome stability were very similar to those on the catansome stability, which might be also explained by the mechanism proposed on the basis of the viewpoint of a mixed solvent dielectric constant.

Introduction

* Corresponding author. Tel: (+) 886-6-2757575, ext. 62633. Fax: (+) 886-6-2344496. E-mail: [email protected].

propanol, was often used to prevent the self-organization of amphiphiles. The intuitive argument and previous findings, which showed that surfactants did not self-organize in ethanol, led them to reexamine the results of Huang and co-workers. A subtle distinction exists in the literature between mixed cationic-anionic surfactants, where each surfactant is accompanied by its own counterion, and ion-pair amphiphiles (IPAs), in which the counterions are removed, leaving two amphiphilic ions oppositely charged.4,10 Vesicle formation from IPAs, however, is usually made possible only by a classic mechanical dispersion process through the preparation of thin films in a manner similar to that for liposome formation from phospholipids.11-14 Unfortunately, catansomes formed from IPAs by the mechanical dispersion process usually showed only a short-term physical stability.15,16 To overcome this shortcoming, various approaches have been employed to enhance the stability of catansomes.17 Effects of four homologous cosolvents (methanol, ethanol, 1-propanol, and 1-butanol) on the spontaneous formation of catansomes from eight 1:1 cationic-anionic mixed surfactants, alkyltrimethylammonium bromides-sodium alkyl sulfates (CmN(CH3)3Br-CnSO4Na; m ) 8, 10, 12, 14; n ) 12, 14), at a total surfactant concentration of 10 mM have been systematically studied by Yu et al.18 The experimental results indicated that the vesicle formability of mixed cationic-anionic surfactant systems strongly depended on the kind and amount of cosolvents. Four types of cosolvent effects have been classified, as shown in Figure 1. Among them, type 2 and type 3 cosolvent effects would serve the purpose of promoting catansome formation and have been exemplified by mixed C10N(CH3)3Br-C12SO4Na, C10N(CH3)3-

(1) Regev, O.; Khan, A. J. Colloid Interface Sci. 1996, 182, 95-109. (2) Filipovic-Vincekovic, N.; Bujan, M.; Smit, I.; Tusek-Bozic, L.; Stefanic, I. J. Colloid Interface Sci. 1998, 201, 59-70. (3) Villeneuve, M.; Kaneshina, S.; Arotono, M. J. Colloid Interface Sci. 2003, 262, 227-234. (4) Tondre, C.; Caillet, C. AdV. Colloid Interface Sci. 2001, 93, 115-134 and references therein. (5) Wu, K.-C.; Huang, Z.-L.; Yang, Y.-M.; Chang, C.-H.; Chou, T.-H. Colloids Surf. A 2007, 302, 599-607. (6) Huang, J.-B.; Zhu, B.-Y.; Zhao, G.-X.; Zhang, Z.-Y. Langmuir 1997, 13, 5759-5761. (7) Zhang, X.-R.; Huang, J.-B.; Mao, M.; Tang, S.-H.; Zhu, B.-Y. Colloid Polym. Sci. 2001, 279, 1245-1249. (8) Wang, C.-Z.; Tang, S.-H.; Huang, J.-B.; Zhang, X.-R.; Fu, H.-L. Colloid Polym. Sci. 2002, 280, 770-774. (9) Zana, R.; Michels, B. Langmuir 1998, 14, 6599-6602.

(10) Marques, E. F.; Regev, O.; Khan, A.; Lindman, B. AdV. Colloid Interface Sci. 2003, 100-102, 83-104 and references therein. (11) Liposomes: A Practical Approach; New, R. R. C., Ed.; Oxford University Press: New York, 1990. (12) Liposomes: From Physics to Applications; Lasic, D., Ed.; Elsevier: New York, 1993. (13) Vesicles; Rosoff, M., Ed.; Marcel Dekker: New York, 1996. (14) Synthetic Surfactant Vesicles: Niosomes and Other Non-phospholipid Vesicular Systems; Uchegbu, I. F., Ed.; Harwood Academic: Australia, 2000. (15) Fukuda, H.; Kawata, K.; Okuda, H. J. Am. Chem. Soc. 1990, 112, 16351637. (16) Chien, C.-L.; Yeh, S.-J.; Yang, Y.-M.; Chang, C.-H.; Maa, J.-R. J. Chin. Colloid Interface Soc. 2002, 24, 31-45. (17) Yeh, S.-J.; Yang, Y.-M.; Chang, C.-H. Langmuir 2005, 21, 6179-6184. (18) Yu, W.-Y.; Yang, Y.-M.; Chang, C.-H. Langmuir 2005, 21, 6185-6193.

Liposomes, which are commonly used as drug delivery carriers, are usually constituted by double-chained amphiphiles, such as phospholipids. However, a great deal of work has been done to demonstrate the possibility of forming vesicles from singlechained surfactants under specific conditions. This is the case for mixtures of cationic and anionic surfactants, whose association through the interactions of their polar headgroups can mimic the double-chained structures of phospholipids.1-4 The term “catanionic vesicle” (or catansome) is commonly accepted to describe the vesicular system formed from mixed cationic-anionic surfactants.4,5 Vesicle formation of mixed cationic-anionic surfactant systems is usually through a spontaneous process. However, mixed cationic-anionic surfactants, especially 1:1 mixtures, easily precipitate in aqueous solutions. In addition, the spontaneous formation of catansomes is always accompanied by other surfactant aggregates, such as micelles or liquid crystals. These problems greatly limit the advances in the applications of catansomes. In fact, a series of investigations have been performed by Huang and co-workers6-8 to study the stability of vesicles formed in mixtures of cationic and anionic surfactants with the addition of organic solvents. Their results revealed that organic solvent addition might provide the surfactant mixtures with outstanding vesicle-forming capability, especially for easily precipitated mixed surfactant systems. However, Zana and Michels9 have argued that ethanol, as well as methanol and

10.1021/la701882d CCC: $40.75 © 2008 American Chemical Society Published on Web 01/29/2008

1696 Langmuir, Vol. 24, No. 5, 2008

Yang et al.

Figure 1. Schematic representation of different types of cosolvent effects on the formability of vesicles by various 1:1 cationic-anionic mixed surfactants in aqueous alcohol solutions (After Yu et al.18).

Br-C14SO4Na, and C12N(CH3)3Br-C12SO4Na surfactants. Furthermore, the effectiveness of alcohols on the catansome formation enhancement increased in the order 1-butanol > 1-propanol > ethanol > methanol. On the other hand, four IPAs derived from the paring of alkyltrimethylammonium chlorides and sodium alkyl sulfates have been used to prepare catansomes in water upon a classic mechanical dispersion process through the formation of thin films by Yeh et al.17 In the study, short-chained alcoholssincluding methanol, ethanol, 1-propanol, and 1-butanolswere added as cosolvents at various concentrations, and their effects on the ensuing catansome stability were systematically studied. The experimental results indicated that catansomes formed from one of the IPAs, i.e., dodecyltrimethylammonium dodecyl sulfate (DTMA-DS), could be successfully stabilized by the addition of appropriate amounts of 1-propanol and 1-butanol. Maximum lifetimes of more than 1 year and 132 days for the catansomes formed in 5 vol % 1-butanol and 15 vol % 1-propanol solutions, respectively, were observed, demonstrating that an approach for stabilizing catansomes formed from IPAs became available by means of cosolvent addition. Furthermore, the physical stability of catansomes was found to be strongly dependent on the cosolvent concentration. In general, the catansome stability first increased with increasing the cosolvent concentration, reached a maximum at a specific cosolvent concentration, and thereafter decreased with further increase in the cosolvent concentration. The physical stability of liposomal cosolvent systems and the role cosolvents played in the liposome stability have been less systematically studied in the literature. In addition, it seems that there might be a strong correlation between the cosolvent effects on the liposome stability and those on the catansome stability. A theoretical explanation that can delineate the general trend of cosolvent effects and elucidate the possible roles of cosolvents in the stability of both liposome and catansome is, therefore, most desirable. In this work, a semispontaneous process was used to prepare liposomes and catansomes in water with the aid of cosolvent addition.5 The physical stability of both liposome and catansome was then examined and explained within the framework of the cosolvent effects. Experimental Section Materials. Soybean phosphatidylcholine (Phospholipon 90, PL90) containing 95 ( 3 % phosphatidylcholine was a gift from Nattermannallee Phospholipid GmbH, Germany. The anionic surfactant sodium dodecyl sulfate (SDS) (99% pure) and the cationic

Figure 2. Initial size distributions of liposomes formed from 0.3 wt % PL-90 in aqueous (a) methanol, (b) ethanol, and (c) 1-propanol solutions by the semispontaneous process. surfactant decyltrimethylammonium bromide (DeTMAB) (98% pure) were supplied by Sigma and Fluka, respectively. The cosolvents methanol (99.8% pure), ethanol (99.8% pure), and 1-propanol (99.5% pure) were purchased from Aldrich. All chemicals were used as received without further purification. All experiments were conducted with pure water that was passed through a Milli-Q Plus purification system (Millipore) with a resistivity of 18.2 MΩ cm. Preparation of the Ion-Pair Amphiphile. The IPA will precipitate out when cationic and anionic surfactants in aqueous solutions of sufficiently high concentrations are allowed to react with each other. In this work, such a precipitate, decyltrimethylammonium dodecyl sulfate (DeTMA-DS), was prepared by mixing equal volumes (500 mL) of 20 mM solutions of DeTMAB and SDS. A concentration of 20 mM is well beyond the critical micelle concentrations (cmc’s) of these surfactants. After standing for 1 h, the precipitate was separated from the solution by repeated centrifuging and washing. It was then dried for 36 h under vacuum and ground into fine powders for further studies. The IPA DeTMA-DS was thereby obtained by pairing of two oppositely charged surfactants. The resulting pure IPA can be subjected to elemental analysis and determination of mass and infrared spectra.16 There was enough evidence to prove that the pseudo-double-chained amphiphilic compound contained amphiphilic cation and amphiphilic anion in an equimolar ratio.

Stability of Liposomes and Catansomes

Langmuir, Vol. 24, No. 5, 2008 1697

Figure 3. TEM image of the liposomes formed from 0.3 wt % PL-90 in 30 vol % ethanol solution observed by the negative staining technique. Vesicle Preparations by a Semispontaneous Process. A very simple semispontaneous process was used for preparing vesicles in water with the aid of cosolvent addition.5 In a typical experiment of vesicle preparation, a suitable amount of PL-90 or IPA was dissolved in the alcohol. Water was then added and the solution sample was passed through a homogenizer (Ultra-Turrax T-25, IKA) at 11 000 rpm for 3 min at room temperature in a sealed container specially designed for this vesicle preparation process. The composition of mixed solvents was represented by the cosolvent volume percentage. Vesicle Preparations by a Classic Mechanical Dispersion Process. PL-90 vesicles were also prepared by the classic mechanical dispersion process through the formation of thin films, which is frequently used for preparing liposomes. An appropriate amount of PL-90 in 10 mL of chloroform was dried to form thin films on the surfaces of glass beads in a test vessel by nitrogen blowing. The resulting PL-90 thin films were then hydrated with 60 mL of water or aqueous alcohol solution while vortexing for 10 min and sonicating for 1 h. The translucent vesicle solution was then extruded at 60 °C under nitrogen pressure through 0.4 and 0.2 µm nuclepore polycarbonate membranes (Osmonics) sequentially. An extruder obtained from Lipex Biomembranes, Inc., was used in the extrusion process. After extrusion, the vesicle solution was let to stand for 1 h before any experimental study was performed. Size Measurements and Physical Stability Evaluation of Vesicles. For the vesicle formation processes, typical vesicles with mean sizes of hundreds of nanometers were expected. It is obvious that the variation of vesicle size with time is a critical test for the physical stability of vesicles. Therefore, the vesicle stability was evaluated by measuring the vesicle size and size distribution as a function of time using a commercial particle size analyzer (model 3000HS, Malvern, UK) with the considerations of physical property changes of the samples. During the measurement, the count rate, i.e., the sample scattering intensity, was also provided. It is worthy to note that the count rate determined for pure water is about 0.6 kcps (kilocount per second). Medium stable count rates are deemed to be the manifestation of the presence of vesicles in a large enough amount. In this work, a criterion of count rate for the existence of an appreciable amount of vesicles was set to be g50 kcps. A transmission electron microscope (model H-7500, Hitachi) was used to obtain the vesicle images with the negative-staining technique.

For the sample preparation, a few drops of vesicle dispersions were applied to a carbon-coated Cu grid and dried. A drop of uranyl acetate-ethanol solution was then added as the staining agent.

Results and Discussion It is noteworthy that vesicles could not be formed in pure water from PL-90 and the IPA studied in this work by the semispontaneous process. This is apparently due to the extremely low solubilities of PL-90 and the IPA in water. Figure 2 indicates the initial size distributions of PL-90 vesicles and the corresponding count rates of size analyses, which were measured immediately following the liposome preparation by the semispontaneous process, for 0.3 wt % PL-90 in various aqueous alcohol solutions. As shown in Figure 2a, mean liposome sizes around 100 nm were found in the 20, 30, 50, and 60 vol % methanol solutions. Proper values of size and count rate suggested the existence of liposomes in these solutions. In addition, the solutions actually appeared bluish to the eye. For the 70 vol % methanol solution, however, a double-peaked size distribution of the vesicles was found. Furthermore, no aggregate formed in the 80 vol % methanol solution, as suggested by a count rate of near zero obtained in the size measurement. Under this condition, the solution appeared clear to the eye. Similar results for 0.3 wt % PL-90 in aqueous solutions of ethanol and 1-propanol with various concentrations were demonstrated in parts b and c, respectively, of Figure 2. A typical TEM image for the liposomes formed from 0.3 wt % PL-90 in 30 vol % ethanol solution is demonstrated in Figure 3. In contrast with the unformability of PL-90 vesicles in pure water or mixed solvents with low concentrations of alcohols, significant enhancement of liposome stability was found in some mixed solvents with appropriate alcohol concentrations. The data in Figure 4 suggest that 50 vol % methanol, 30 vol % ethanol, and 15 vol % 1-propanol can provide the PL-90 liposome lifetimes of 207, 231, and 217 days, correspondingly. It should be noted that the upper and lower limits of an accurate size measurement by the particle size analyzer used in this work are 3000 and 3

1698 Langmuir, Vol. 24, No. 5, 2008

Figure 4. Time-dependent size distributions of liposomes formed from 0.3 wt % PL-90 in aqueous solutions of (a) 50 vol % methanol, (b) 30 vol % ethanol, and (c) 15 vol % 1-propanol by the semispontaneous process. The arrow indicates the time when the vesicle size exceeded the upper limit of an accurate size measurement.

nm, respectively. The arrow in Figure 4, hence, indicates the time when the vesicle size exceeded the upper limit of an accurate size measurement and, therefore, represents the lifetime of the PL-90 liposomes. As shown in Figure 5, all of the three cosolvents could significantly promote the formation of PL-90 liposomes by the semispontaneous process. Generally, the liposome lifetime increased with increasing the cosolvent concentration at the beginning, reached a maximum at a specific cosolvent concentration, and then decreased with further increasing the cosolvent concentration. The liposome stability eventually became negligible at the cosolvent concentrations higher than a specific value. Obviously, cosolvent effect type 3 (as shown in Figure 1) for the enhancement of PL-90 vesicle stability by means of cosolvent addition in water can be identified. It can be noted that the lowest cosolvent concentration that resulted in noticeable liposome stability decreased in the order C3OH < C2OH < C1OH. This is also revealed in Figure 5 by the cosolvent concentrations at the left-hand sides of the shaded areas, which cover the cosolvent concentration ranges with liposome lifetimes

Yang et al.

Figure 5. Cosolvent effects of (a) methanol, (b) ethanol, and (c) 1-propanol on the physical stability of liposomes formed from 0.3 wt % PL-90 by the semispontaneous process.

g1 week. Moreover, the amount of cosolvent being too large to provide noticeable liposome stability also decreased in the order C3OH < C2OH < C1OH. This is also revealed by the cosolvent concentrations at the right-hand sides of the shaded areas indicated in Figure 5. Since a classic mechanical dispersion process through the formation of thin lipid films is generally applied to prepare liposomes, the cosolvent effects on the PL-90 liposome stability of using the mechanical dispersion process were also examined. It appeared that similar cosolvent effects as demonstrated in Figure 5 on the PL-90 vesicle stability were also found for the three cosolvents in the classic mechanical dispersion process (Figure 6). Apparently, cosolvent effects work on the liposome stability by both semispontaneous and classic mechanical dispersion processes. It is noteworthy that the cosolvent concentration at the left-hand sides of the shaded areas in Figure 6 is zero, indicating that liposomes were found to still be stable enough in pure water. Cosolvent effect type 2 rather than type 3 (as shown in Figure 1) for the enhancement of PL-90 vesicle

Stability of Liposomes and Catansomes

Langmuir, Vol. 24, No. 5, 2008 1699

Figure 7. (a) Time-dependent size distributions of catansomes formed from 5 mM DeTMA-DS in 20 vol % ethanol by the semispontaneous process. The arrow indicates the time when the vesicle size exceeded the upper limit of an accurate size measurement. (b) Cosolvent effects of ethanol on the physical stability of catansomes formed from 5 mM DeTMA-DS by the semispontaneous process. Table 1. Cosolvent Concentration Ranges in Which the Effective Enhancement of Vesicle Formation from PL-90 and IPAs by the Semispontaneous Process Can Be Detected concn of cosolvents, vol % (dielectric constant) double-chained amphiphiles

concn

methanol

ethanol

1-propanol

PL-90

0.3 wt %

Figure 6. Cosolvent effects of (a) methanol, (b) ethanol, and (c) 1-propanol on the physical stability of liposomes formed from 0.3 wt % PL-90 by the classic mechanical dispersion process.

DeTMA-DS

1 mMa

5.2-74.4 (76-47) 1.7-38.3 (78-63) -

1.4-19.2 (78-65) 3-18 (77-66) -

stability by means of cosolvent addition in water can then be identified. This is different from those shown in Figure 5 and is probably due to the different preparation processes and PL-90 dissolution procedures. To provide further evidence on the similarity between the cosolvent effects on the PL-90 liposome stability and those on the catansome stability, the catansome stability of DeTMA-DS with a comparatively high concentration of 5 mM in the presence of ethanol by the semispontaneous process was also investigated in this work. As shown in Figure 7a, it seemed that 20 vol % ethanol could provide the catansomes formed from 5 mM DeTMA-DS with a lifetime of 168 days. Moreover, the similar cosolvent effects on the formation of DeTMA-DS catansomes are also confirmed in Figure 7b. For PL-90 and three cosolvents studied in this work, Table 1 summarizes the results regarding the vesicle stability by the semispontaneous process. The applicability of short-chained alcohols as cosolvents in the semispontaneous process of catansome formation with 1 mM IPAs has been studied,5 and

DeTMA-TS

1 mMa

1.7-44.2 (78-55) 3.3-29.3 (77-64) 8-38 (74-58) 9.3-22.6 (74-68)

DTMA-DS

1 mMa

5 mM

12.5-38 (75-63)

0.2-14.3 (78-71) 8-17 (72-68)

a The data for three IPAs, DeTMA-DS, DeTMA-TS (decyltrimethylammonium tetradecyl sulfate), and DTMA-DS (dodecyltrimethylammonium dodecyl sulfate), at 1 mM were reported by Wu et al.5

some of the data are also listed in Table 1 for the purpose of comparison. It can be realized from Table 1 that the cosolvent effects on the PL-90 liposome stability were similar to those on the catansome stability. It is noteworthy that methanol and 1-propanol in addition to ethanol were also used as cosolvents to affect the stability of liposomes formed from PL-90. A similar trend of cosolvent effect, however, was always found for these three cosolvents. For the catansomes formed from IPAs by the classic mechanical dispersion process17 and the spontaneous formation process,18 an explanation based on the viewpoint of medium dielectric constant has been proposed for the cosolvent effects. When a

1700 Langmuir, Vol. 24, No. 5, 2008

short-chained alcohol with a low dielectric constant is added in water, the dielectric constant of the aqueous mixture is expected to be lower than that of pure water. The lower dielectric constant or polarity of the mixed solvents would enhance the solvophobic characteristic of the lipid or IPA polar headgroups. However, the hydrophobic attraction between the hydrocarbon chains of lipids or IPAs would be deteriorated due to the presence of alcohols, weakening the solvophobic effect of the lipid or IPA alkyl chains. The opposing influences of cosolvents on the solvophobic effects of PL-90 or IPA molecules thus result in a maximum in the cosolvent effects on the stability of PL-90 liposome or catansome, as shown in Figures 5-7. It is worthy to note that, by further increasing the cosolvent concentration, further lowering of the liquid dielectric constant finally caused complete dissolution of PL-90 or IPA molecules. In the whole ranges of cosolvent concentration, therefore, three regimessenhanced vesicular state, diminished vesicular state, and free molecule statesare available regarding the formation and stability of liposomes or catansomes. In addition, it is noteworthy that ethanol, 1-propanol, and 2-propanol were used as cosolvents by Huang and co-workers6-8 in the spontaneous formation of vesicles from mixtures of cationic and anionic surfactants. An explanation based on the medium dielectric effect was also suggested. The corresponding dielectric constant ranges for the effective cosolvent concentration ranges reported in Table 1 were evaluated19,20 and are also listed in the table. It appeared that the stability differences of liposome and catansome in the presence of various alcohols with different concentrations may not be totally explained only by taking the dielectric constant influence into account. Other possible causes have been discussed elsewhere.17 For example, the alcohols may enter the assemblies and reside in the polar headgroup layer and then affect the apparent critical packing parameters of the amphiphiles, causing changes in the solubility and vesicle stability of the amphiphiles with the (19) Perry’s Chemical Engineers’ Handbook, 6th ed.; Perry, R. H., Green, D. W., Eds.; McGraw-Hill: New York, 1984. (20) Yilmaz, H. Turk. J. Phys. 2002, 26, 243-246.

Yang et al.

extent depending on the hydrocarbon chain length of the alcohols. More work is apparently needed to clarify the mechanisms of the cosolvent effects further.

Conclusions Effects of three cosolvents (methanol, ethanol, and 1-propanol) on the physical stability of liposomes formed from PL-90 by the semispontaneous process were systematically studied. The cosolvent effects on the PL-90 liposome stability were also compared with those on the IPA catansome stability. Some meaningful conclusions can be drawn from the study as follows. First, significant enhancement of physical stability was found for liposomes formed from PL-90 in water with the aid of cosolvent addition. In general, liposome stability at first increased with increasing the cosolvent concentration, reached a maximum at a specific cosolvent concentration, and thereafter decreased with further increasing the cosolvent concentration. The liposome formation was eventually unavailable at cosolvent concentrations higher than a specific value. Furthermore, the cosolvent effects were also effective in the physical stability enhancement of liposomes formed by the classic mechanical dispersion process through the formation of thin films. Second, the similarity between the cosolvent effects on the physical stability of PL-90 liposomes and those on the physical stability of the IPA catansomes was realized. Third, the effectiveness of promoting PL-90 vesicle formation by cosolvents increased in the order 1-propanol > ethanol > methanol. Finally, it seems that an explanation of cosolvent effects based on the medium dielectric constant may be still applicable for the PL-90 vesicle formation by the preparation processes but may have to take the other mechanisms into account. Acknowledgment. This work was supported by the National Science Council of Taiwan through Grants NSC 93-2214-E006-006 and NSC94-2214-E-006-020. LA701882D