EUROPEAN JOURNAL OF DRUG METABOUSM AND PHARMACOKINETICS, 1994, No 3, pp. 257-269
Surfactant Systems: Microemulsions and Vesicles as Vehicles for Drug Delivery M JAYNE LAWRENCE Department of Pharmacy, King's College London, Chelsea Campus, London England
Keywords: Microemulsions, Vesicles, Drug delivery
SUMMARY Although surfactants have been widely used as pharmaceutical adjuvants for many years, it is only relatively recently that their phase structures have been seriously considered as drug delivery vehicles per se. This review highlights the work to date investigating the potential of microemulsions as drug carlers and also reports on preliminary studies performed on the use of vesicles formed from nonionic surfactants,
-. -. Surfactant Molecules
Lamellar Phase
Figure 1:
Spherical Micelles
Rod-shaped Micelles
Reverse Hexagonal Phase
Hexagonal Phase
Reverse Micelles
Most commonly encountered surfactant aggregates in aqueous solution. Copied from [1].
Please send reprints requests to: Dr M. Jayne Lawrance, Lecturer in Pharmacy, Department of Phannacy King's College London, University of London, Manzera Road, London SW3 6LX, England.
Bur. J. Drug Metab. Pharmacokinet., 1994, No 3
258
Table 1. Equilibrium Phase Structures Encountered In Oil-Water-Surfactant Systems
Micelles/Swollen MicellesJMicroemulsions Reverse Micelles/Reverse Micelles/Reverse Microemulsions
Table 2. Use Of Equilibrium Surfactant Phase Structures In Drug Delivery
Surfactant System
Binary Systems
Swollen
Polymer-like Reverse Micelles
Micelles
Solution - most routes
Cubic
Sustained release - oral, subcutaneous, intravenous and topical
Hexagonal
Sustained release - most routes, particularily topical
Lamellar
Sustained release - most routes, particularily topical
Cubic/Reverse Cubic Phases Hexagonal/Reverse Hexagonal Oil and/or Water Swollen Lamellar Phases In addition to the above one phase structures, a
range of two and three phase systems are observed, in which one of the above exists in equilibrium with an excess of oil/and/or an excess of water.
Table 3. Uses Of Non-Equilibrium Surfactant Structures
System
Possible Use
Vesicles
Solution and sustained release - most routes, except oral
Reverse vesicles
Solution and sustained relase - most routes, except intravenous
(Macro)emulsions
Solution - most routes
Multiple emulsions
Sustained release - most routes except intravenous
Self-emulsifying Solution systems NB All systems have the potential to protect labile compounds
INTRODUCTION Surfactants are amphiphilic molecules in that they contain both a hydrophilic/water soluble and hydrophobieJoil soluble region. When dispersed in water surfactant molecules self-associate to form a wide variety of equilibrium phase structures in which the hydrophobic chains are removed from contact with water. The most commonly encountered aggregates are illus-
Possible Use
Ternary Systems Microemulsions
Solution - most routes
Reverse microemulsions
Solution and sustained release-excepting intravenous
Polymer-like reverse Topical drug delivery micelles Cubic Phase (Microemulsions gels, ringing gels, transparent-oil-water gels, isotropic gols)
Sustained release - most routes, particularly topical not intravenous
NB All systems have the potential to protect labile compounds trated in Figure 1. The aggregate formed by a particular surfactant is dependent upon the surfactant structure and concentration [2J. Not surprisingly the variety of possible equilibrium phase structures is further increased upon the addition of oil as seen in Table 1. This is partly because a number of reverse structures are formed in which the hydrophilic head groups are expressed on the interior of the aggregate. Again the type of aggregate formed is dependent upon both the structure and relative proportions of the various components. As each of the phase structures mentioned above form spontaneously and therefore are thermodynamically stable, they are potentially very valuable as drug delivery vehicles (Table 2). In addition to these equilibrium phase structures, there are also a number of non-equilibrium surfactant structures that have potential as drug carriers, for example vesicles (Table 3). The present paper reviews the pharmaceutical potential exhibited by microemulsions and vesicles.
259
Microemulsions and vesicles
MICROEMULSIONS DeFinition Microemulsions are fluid, transparent. thermodynamically stable oil and water systems, stabilized by a surfactant usually in conjunction with a cosurfactant, which may be a short chain alcohol. amine or other weakly amphiphilic molecule [3J. Figure 2 shows a theoretical pseudo-ternary phase diagram indicating the possible range of microemulsion compositions. At very high surfactant concentrations liquid crystalline phases are formed. While microemulsions may occur over a wide range of oil and water compositions, in the majority of systems microemulsions exist only over a narrow range of concentrations, although in a number of systems two distinct microemulsion areas are observed. surfactant
water
Figure 2:
2/3 phase region
...;:~
oil
Theoretical pseudo-ternary phase diagram
At low oil or water volume fractions, microemulsions are generally considered to be a dispersion of either oil or water droplets stabilized by an interfacial film of surfactant and cosurfactant (Figure 3). Indeed droplets are the most commonly encountered type of microemulsion microstructure. However under the conditions of low dispersed phase, there is a considerable amount of debate as to what differentiates a micelle from a microemulsion [5,6,7J. This distinction is complicated by the fact that in most cases the transformation between a micelle and microemulsion appears to be gradual with no obvious changes in physico-chemical properties. Indeed some workers believe that there is no difference in nature between a microemulsion and a solute-swollen micelle [8J and that the term micellar solution should not be used to denote systems in which three components are present but should instead be applied only to systems containing just surfactant in solvent Others argue however that the name microemulsion should be reserved for systems that contain more than a few percent of the dispersed phase [9J and that there is a distinction between swollen micelles (which have no core) and microemulsions (which exhibit a distinct core
composed of disperse phase) [IOJ. In the present article no attempt is made to distingiush between a microemulsion and a swollen micelle. At intermediate oil and water concentrations it is obviously not possible for the dispersion to consist of dispersed droplets. In these cases it is thought that a bicontinuous structure exists [11J in which the water and oil domains are separated by a regular or topologically chaotic continuous amphiphile-rich interfacial layer (Figure 3). Frequently the transition from an oil-in-water to a water-in-oil microemulsion occurs via a bicontinuous type structure, although in some systems it may occur via a liquid crystalline lamellar phase. In terms of their microstructure microemulsions are therefore very complex systems. Indeed it has been shown that in microemulsions that exist over a wide range of compositions several different types of structure may be present [12J. Evidence appears to suggest however that when only a small amount of one phase is present it is reasonable to assume that a droplet structure exists, although the amount of internal phase that can be considered small is dependent upon the composition and chemical structure of the components of the microemulsion. At intermediate dispersed phase volumes a bicontinuous structure is more probable [13J. It is also important to remember that whatever the microstructure, microemulsions are dynamic systems in which the interface is continuously and spontaneously fluctuating [14J.
a)
water
d)
Figure 3:
Possible structures in microemulsion systems. Droplet structures; a) oil-in-water (Ll) and b) water-in-oil (L2). Two extremes of biocontinuous; c) cubically symmetrical and d) completely random. Roproduced from [3J.
Eur. J. Drug Metab. Pharmacokinet., 1994, No 3
260
Table 4 Pharmaceutical Advantages Of Microemulsions General Advantages
General Advantages Ease of preparation Clarity Stability Ability to be filtered Vehicle for drugs of different Iipophilicities in the same system Low viscosity, no pain upon injection Specific Advantages Water-In-Oil Protection of water soluble drugs Sustained release of water soluble material Increased bioavailability Oil-In-Water Increased solubility of lipophilic drugs Increased bioavailability Bicontinuous Concentrated formulation of both oil and water soluble drug
Pharmaceutical Applications Over the last ten years microemulsions have attracted a considerable amount of interest as potential drug delivery systems [15]. Part of this interest stems from their clarity, ease of preparation and indefinite stability. These properties combined with their potential for incorporating a range of drugs of varying Iipophilicity are just a few of the features that make microemulsions very attractive for the purposes of formulation. Table 4 lists some of the advantages of microemulsions cited in the literature, while Table 5 illustrates the range of studies that have been reported using microemulsions for the purposes of drug delivery. Although microemulsions possess a number of very important advantages as vehicles for a wide variety of delivery routes, no formulations have yet found their way to the clinic. One of the main problems encountered has been the type of surfactant and/or cosurfactant used to stabilize the system. By far the majority of work reported has involved the use of toxic short chain alcohol or amine cosurfactants, such as butanol and butylamine. This is because most surfactants will not
Table 5. Pharmaceutical Investigations Using Microemulsions
Transdermal delivery of hydrophilic drugs using w/o microemulsions and lipophilic drugs using o/w microemulsions Oral delivery of peptides using w/o microemulsions Intramuscular formulation of peptides using w/o microemulsions Occular delivery of a drug in o/w microemulsion w/o = water - in - oil o/w = oil - in - water form microemulsions without the presence of a cosurfactant. There are notable exceptions however such as the double chain ionic surfactants sodium di-2-ethylhexyl sulphosuccinate (AOI) and dioctadecylammonium bromide (DDAB), and single chain nonionic surfactants, such as the n-alkyl polyoxyethylene ethers. In addition to the problem of toxicity, the presence of a cosurfactant also causes problems, because these forms of microemulsion are frequently destroyed upon dilution. This is because the cosurfactant generally partitions to all three of the various microemulsion phases, (ie the interfacial region and the water and oil phases). So that after dilution some of the cosurfactant moves from the interfacial region into the continuous phase, to restore the equilibrium, thereby destroying the integrity of the microemulsion. The removal of cosurfactants from microemulsions intended for pharmaceutical use would obviously be a big advantage. Another serious problem from a formulation point of view is that most studies to date have investigated the formation of microemulsions using aliphatic and aromatic oils, such as hexane and benzene, which are obviously not appropriate for pharmaceutical formulation. Furthermore the ionic surfactants frequently used are themselves toxic and so are equally unsuitable for the purposes of pharmaceutical formulation [16]. Recently, because of the interest in these systems as potential delivery vehicles a number of studies have reported the use of less toxic nonionic surfactants [17,18] or the biocompatible zwitterionic surfactant, lecithin [19,20,21,22]. The use of nonionic surfactants such as the n-alkyl polyoxyethylene ethers is particularly attractive because it is possible to produce microemulsions without the need for the presence of a cosurfactant [23]. Although it is possible to produce water-in-oil microemulsions over a very limited range of composi-
Microemulsions and vesicles
tions from lecithin alone [24], there is a requirement for the presence of a cosurfactant in order to produce microemulsions over a wide area [3]. In an attempt to avoid the use of toxic alcohol or amine cosurfactants some workers have used pharmaceutically acceptable polyhydric alcohols, such as sorbitol [18]. Although these compounds can be considered as cosurfactants they do not exert their effect at the interfacial layer but instead act by decreasing the effective hydrophile - lipophile - balance of the surfactant by reducing the solubility of the head group in the aqueous phase [25]. Unfortunately, while these cosurfactants are non-toxic, any microemulsions containing them are not dilutable.
A recent (successful) attempt at overcoming the joint problems of toxicity and dilutability has examined the use of nonionic surfactants as cosurfactants in lecithinbased microemulsions [26]. Because of their strongly amphiphilic nature the nonionic surfactants are localized predominantly in the interfacial layer and so even when the microemulsion is diluted, the droplet integrity is maintained. It is obvious from the above discussion that for pharmaceutical use, no matter what type of microemulsion is required, only nonionic or zwitterionic surfactants should be considered as these are less toxic and less affected by pH, or the presence of salts etc. [16]. As mentioned, the use of a nonionic surfactant has the additional advantage of possibly avoiding the need for a cosurfactant.
261
bined with physico-chemical studies on the systems) in an attempt to achieve this goal. Until such a time as it is possible to predict which oils and surfactants are needed to produce a particular microemulsion, the pharmaceutical formulator will have to resort to the use of the observations present in the literature. Some useful guidance is presented in an early paper by Schulman et al [32]. These authors maintain that there are three essential conditions necessary for the formation of microemulsions, namely; l.
The production of a transient negative interfacial tension at the oil/water interface.
2. The formation of a highly fluid interfacial sur-
factant film. 3. The penetration and association of the molecules
of the oil phase with the interfacial surfactant film. Although these observations, especially the first, are generally considered to be an over simplification they do provide a useful starting point for the formulation of microemulsions. In the following discussions the applications of these principles are illustrated with particular emphasis on points of pharmaceutical importance.
Transient Negative Interfacial Tension It is now recognized that one of the conditions for microemulsion formation is a very small, rather than a transient negative, interfacial tension. This is rarely achieved by the use of a single surfactant, almost always necessitating the addition of a cosurfactant. The
Formulation If microemulsions are to be used as vehicles for
pharmaceuticals there is a need for some simple rules to aid in the choice of the constituents used. Unfortunately however there are no simple expressions (either theoretically or empirically derived) that adequately predict the structure and likely properties of these systems [27]. Most of the theories presented are too complex to be of use in formulation because they require the determination of a number of unknown parameters. Furthermore it is worth noting that, while the use of both the HLB of the surfactant [28] and the critical packing parameter (CPP) [29] as proposed by Israelachvili and coworkers [30] have been advocated, they are too simplistic to be of any great use in microemulsion formulation [23]. The ideal situation for the formulator would be the ability to predict which surfactants to use, or even synthesize, in order to produce a microemulsion. Work is currently underway in our laboratory (using molecular modelling [31] com-
Table 6. Ability Of CI8:IElO And Cl8ElO To Form 3-Component Oil-In-Water Microemulsions
Oil Examined
ClsEIO
Soybean Oil
y
N
Isopropyl Myristate
y
N
Heptane
y
N
Y
=
does form microemulsions
N
=
does not form microemulsions
The surfactants are abbreviated by the general formula CnEm where n is the number of carbons the aluyl in chain and m is the average number of ethylene oxide groups present 18: 1 represents anoleyl chain.
Eur. J. Drug Metab. Pharmacokinet, 1994, No 3
262
presence of a short chain alcohol for example can reduce the interfacial tension from about 10 mN/m (for a system containing just surfactant) to a value less than 10-2 mN/m. Exceptions to this rule are provided by nonionic surfactants which, at their phase inversion temperature (PTI), also exhibit very low interfacial tensions, of the order of 10-1 to 10-2 mN/m [33]. This explains why nonionic surfactants at their PIT frequently form microemulsions over a wide range of compositions. Generally at temperatures below the PIT any
a) heptane
microemulsion formed by the nonionic surfactant will be of the oil-in-water type, while at temperatures above the PIT the microemulsion will be water dispersed in oil. It is important to note that the PIT of a system depends on the oil, the surfactant(s) and any additives present.
Fluid Interfacial Film The requirement for the formation of a fluid interfacial film is usually achieved by the addition of a cosurfactant. For example without the presence of a cosurfactant, lecithin produces a water-in-oil microemulsion over only a very limited range of concentrations, due in part to the tendency of lecithin to form highly rigid films [34]. The presence of a short chain cosurfactant allows the interfacial film sufficient flexi-
a) oleic acid
water
75
surfactantmixture (lecithin/butanol 1.94:J)
50
b) hexadecane
waterL
water
75
' oleic acid
50 b) octanoic acid
c) octadecane
water
75
50
Figure 4: Partial triangular phase diagram for oil-inwater microemulsions formed with C12ElO, water and a) heptane, b) hexadecane and c) octadecane at 298K. On the absicssa surfactant concentration is increasing in the direction left to right and on the ordinate oil concetration is increasing from bottom to top.
waterL-
...3octanoicacid
Figure 5: Phase diagram for lecithin/butanol/water and a) oleic acid and b) octanoic acid at room temperature and lecithin/butanol mixing ratio of 1.94:1. L1 is the oil-in-water microemulsion region, L2 is the water-in-oil microemulsion region and LC is a liquid crystalline phase.
Microemulsions and vesicles
bility to take up the different curvatures required to form microemulsions over a wide range of compositions [3].
In the case of nonionic surfactants the need for a fluid interfacial film can be clearly seen in Table 6 which shows how the introduction of a cis double bond into the long (CIS) hydrocarbon chain of an n-alkyl polyoxyethylene ether surfactant fluidized the interfacial film sufficiently to allow the formation of oil-in-water microemulsions at 29SK [23]. The corresponding saturated chain surfactant does not form an oil-in-water microemulsion at the experimental temperature, but if the temperature is raised to 310K the hydrophobic chains become fluidized and enable the microemulsion to form (data not shown). One obvious implication of the need for a fluid interfacial film is that the surfactants used should not possess very long hydrophobic chains, or alternately, if a long hydrophobic chain surfactant is required (see below) they should contain sufficient fluidizing groups (eg unsaturated bonds) to ensure that the hydrophobic chains are in a liquid-like state. As a rough guide to whether the chains are in a fluid-like state, the hydrophobe should not have a melting point above the temperature at which the microemulsions are being used.
Association of the Oil with the Interfacial Film The need for the oil to associate with the interfacial film of the microemulsion means that the size of the oil molecules is important in determining whether a microemulsion is formed. As a rule, greater solubilization is achieved with smaller oils, with no microemulsions formed if the chain length of the oil is too long. The effect of chain length is depicted clearly in Figure 4 which shows the extent of the oil-in-water microemulsion area exhibited by C12EIO* decreases with increasing oil molecule length [23]. Interestingly the longer (more fluid) alkyl chain surfactant ClS:IElO* appears to be much less sensitive to the size of the oil incorporated [23]. The size of the oil can also have an influence on the formation of a water-in-oil microemulsion as shown in Figure 5. Here the effect is demonstrated by the change from a medium chain fatty acid (octanoic acid) to a long chain fatty acid (oleic acid) [35]. On the basis of these observations therefore it is *
263
concluded that in order to maximize both the solubilization and the area of existence of a microemulsion, it is necessary to use short chain oils. However from the pharmaceutical point of view the situation is not that simple. In addition to considering the toxicity of the various components, other factors must be borne in mind, such as drug loading, and the effect of the drug on microemulsion stability.
Specific Pharmaceutical Considerations It has been shown that with the n-alkyl polyoxyethylene ethers, the increase in area of existence of oil-in-water microemulsions afforded by the smaller oils may be counter productive in terms of the amount of drug incorporated. It appears that these oils significantly penetrate the interfacial surfactant monolayer, altering deleteriously, one of the main sites of solubilization in the microemulsion, namely the relatively dehydrated polyoxyethylene chains closest to the hydrophobic core [36,37]. If solubilization in oil-in-water microemulsions is important therefore, it maybe more advantageous to use larger oils which are less likely to penetrate the interfacial surfactant layer, and consequently will not dilute the polyoxyethylene region near the core. It will be apparent, therefore, that in most cases it is necessary to strike a comprimise, so that the amount of oil solubilized is balanced against any disruption of the interfacial surfactant layer.
The amount of drug incorporated into a microemulsion depends upon its relative solubility in the various components of the system. Recent work has shown that unless the drug is very soluble in the oil forming the dispersed phase of an oil-in-water microemulsion, the amount of drug incorporated will be no more than that observed in the corresponding micellar solution [17]. In this context it is worthwhile noting that drugs generally show an improved solubility in polar oils such as medium and long chain trigylcerides when compared to nonpolar oils such as hexane. Polar oils should therefore be considered first as the dispersed phase for oil-in-water microemulsions if solubilization is important. However it must be realized that the oil cannot be too polar or else a microemulsion will not be formed. Triacetin for example has a very high aqueous solubility and would not therefore be suitable as a dispersed phase. The effect of an incorporated drug on the stability of a microemulsion is very dependent upon the drug
The surfacbants are abbreviated by the general formula CnEm where n is the number of carbons in the alkyl chain and m is the average number of the ethylene oxide groups present.
Bur. J. Drug Metab. Pharmacokinet, 1994, No 3
264
a) water
surfactant mixture (Iccithin{butanoll:l)
Table 7 Formulation of Microemulsions
Component
isopropylmyrislate
water
b) lO%w/w sodium salicylate solution surfactant mb:ture (lecithin/butanoll:1)
1O%w/W aqueoussodium salicytatesolution
Figure 6:
Possible Consequence
Polar oil
May not allow the formation of a microemulsion
Low molecular weight oil
May alter solubilization of drug in the interfacial region
Cosurfactant
Microemulsions containing cosurfactant may not be dilutable
Long alkyl chain surfactants
May not produce microemulsions over an appropriate temperature range
Surface active drug
May alter area of existence and cause phase changes upon release
Presence of electrolytes, buffer and other additives
May affect stability and alter area of microemulsion formation
isopropyl myristete
Pseudo-ternary phase diagram of quateranry systems containing lecithin/butanol/isoproyl myristate and a) water and b) lO%w/w aqueous sodium salicylate solution at room temperature and a lecithin/butanol mixing ratio of 1:1.
structure. For example, work has shown that it is possible to incorporate lipophilic drugs such as testosterone enthanate at levels greater than 6 wt% in a 2 wt% soybean oil-20 wt% Brij 96 microemulsion without affecting stability [17]. Unfortunately this may not be the situation when surface active drugs are incorporated. Sodium salicylate, a weakly surface active agent, has been shown to significantly alter the phase diagram of a number of lecithin-based microemulsions (Figure 6) [38]. This effect has important implications for drug delivery because this phase change may concevably be reversed as the surface active drug is released and, consequently, this phase reversal may lead to a change in the pattern of drug release with time. Although this effect may potentially be a problem it can. under certain conditions, be used to advantage. For example it has been shown that it is possible to produce a controlled release formulation from a reverse mieroemulsion containing drug which, upon contact with biofluids, transforms into a liquid crystalline system, which controls the rate of drug release [39].
It is important also to remember that the presence of buffers etc., may affect the microemulsion region. It is well known that a large number of electrolytes alter (usually by depressing) the PIT of nonionic surfactants such as the polyoxyethylene ethers [16]. The presence of polyhydric alcohols such as glycerol, xylitol and sorbitol [37] can also decrease the PIT. The presence of electrolytes and other additives may therefore have a considerable affect on the microemulsion. It is consequently important to determine the area(s) of microemulsion existence in the presence of any electrolytes and other additives that may be present under the intended conditions of use of the microemulsion. If the PIT of the final formulation is close to the operating temperature range of the microemulsion the formulation will be very temperature sensitive. If, however, the PIT is more than about 30 degrees above the maximium operating temperature the microemulsion will be fairly insensitive to temperature fluctuations.
In summary, for pharmaceutical purposes it is therefore probably best to use a fairly large polar oil such as a medium chain triglyceride, especially for oil-inwater microemulsions. As a consequence therefore the surfactants used are required to possess long alkyl chains in order to incorporate these oils. By necessity these long hydrophobic chain surfactants must contain a number of unsaturated bonds or other fluidizing groups to ensure their ability to produce microemul-
265
Microemulsions and vesicles
Table 8. Problems Encountered With Coating Liposomes With Polyoxyethylene Chains
Method of Coating
Problem
Table 9 Type of Aggregate Formed By 2CxEy Surfactants
Micelles
Vesicles
Incorporation of modified phospholipid
Too much added destroys vesicles
2CsE16
2Cl4El6
2CIOEl6
2Cl6El6
Incorporation of surfactants
Too much added destroys vesicles
2CIZEl6
2CISEl6
Physically coating preformed liposomes
Leakage of vesicle contents and possible destruction of vesicle
2Cl6EIZ
Chemically coupling to pre-formed vesicles
Increased complexity of production
2CISEIZ
The double alkyl chain surfactanes are abbreviared by the general formula 2CnEm where n is the number of carbons in the alkyl chain and m is the average number of ethylene oxicle groups. Sea figure 7 for the moleculare structure of the surfactants.
sions.
Conclusion From the above discussion it is obvious that, if certain principles are borne in mind during formulation (see Table 7) microemulsions possess considerable potential as vehicles for the delivery of pharmaceuticals.
VESICLES For over twenty years phospholipid vesicles (liposomes) have been heralded as an almost universal drug delivery system [40]. However only a handful of liposome preparations, such as Ambisone® and Amphocil® (injectable liposomal formulations of amphotericin B) have reached the market place. This lack of commercial exploitation is due, at least in part, to the inherently poor in vitro and in vivo stability of liposomes. For example, upon standing liposomes, particularly those made solely from phospholipid, revert back to the planar bilayers from which they originate. In vivo liposomes are quickly removed from the systemic circulation, following opsonization by specific plasma proteins, by the macrophages of the reticuloendothelial system. in particular the Kupffer cells of the liver. Although these problems were realized fairly early on in the development of liposomal formulations, it is only recently that significant progress has been made towards producing "stable" formulations [41]. The most successful attempts at improving stability have resulted from the alteration of surface characteristics [41]. This is because both in vitro and in vivo vesicle stability are largely a result of their surface
properties [41]. The beneficial changes in surface properties have been achieved by the (partial) covering of the vesicle surface with a range of hydrophilic moieties such as the glycolipid GMI [42]. Although a range of hydrophilic groups have been investigated, the most encouraging results have been achieved by attaching to the surface, long polymeric ethylene oxide chains. It is generally considered that the long hydrophilic polyoxyethylene stabilizes the vesicle by producing a steric barrier [43]. This steric barrier acts to reduce both vesicle-vesicle contact, thereby increasing vesicle in vitro stabilty and, the ability of opsonins (plasma proteins) to coat the vesicles, a necessary pre-requistite for removal by the macrophages of the liver. The use of a steric barrier produced by long hydrophilic chains to stabilise colloidal particles is not new and has been used for many years to stabilize emulsion particles [44,45].
One of the problems with covering the vesicle surface with long hydrophilic chains is the method of attachment of the polymer chains as can be seen in
Figure 7: Molecular structure of the novel nonionic surfactants (2CnEm). See the legend to Table 9 for explanation of the nomenclature used for the novel surfactant
Bur. J. Drug Metab. Pharmacokinet., 1994, No 3
266
U~--------------I
1500 , . . - - - - - - - - - - _ . ,
--0-
Liposome
--+- Liposome and drug
-
1000
e.5
DSPC
----fr- 2C lsEzo
1.5
vesicles
2C18Ezo vesicles and drug
1\1
....CIl N
500
....-..,.--4
o~-..,..-..,.:;0 o 20 '0
-
50
Ti:ne IOCys.
Figure 8:
Comparison between stability with respect to size of 2C1SE12 vesicles and lipoomes prepared from DSPC (distearoylphosphatidylcholine) stored at 277K. Size determined by photon correlation spectroscopy at 298K. See the legend to Table 9 for explanation of the nomenclature used for the novel surfactant.
Table 8. For a number of years we have been exploiting an alternative approach wherein we have synthesized novel double chain nonionic surfactants that contain as their head group a long polyoxyethylene chain [46,47,48,49]. The general structure of the nonionic surfactants is given in Figure 7. Preparing vesicles from these nonionic surfactants circumvents the problems experienced with the other methods of covering the vesicle surface. In addition this approach has several other advantages including the ability to synthesize a surfactant to produce a vesicle of the desired characteristics, for example size, and within certain limits (see below) the length of the chains coating the vesicles. Vesicles can be produced from these novel surfactants by those methods commonly used to prepare phospholipid containing vesicles [49]. In addition the vesicles are able to incorporate both dicetylphosphate and cholesterol into their structure [49]. Furthermore, because of the good solvent properties of polyoxyethylene glycol. the possibility of an improved encapsulation of water-insoluble materials exists [43,49]. Finally producing 'sterically stabilized' vesicles from these synthetic surfactants should significantly reduce production costs. One significant difference between these and other types of 'sterically stabilized' vesicles is that in order to produce a vesicle, the hydrophilic head group of the
time (h)
Figure 9:
Comparison of the uptake of 2C18E20 vesicles and dipalmitoylphosphatidylcholine/cholesterol/dicetylphosphate (ratio 9:9:2) liposomes into the mouse macrophage cell line (J774) infected with the leishmania mexicana mexicana parasite. Error bars represent standard deviation. See the legend to Table 9 For explanation of the nomenclature used for the novel surfactant.
surfactant should not be too long (or the hydrocarbon chain too short) or else micelles, and not vesicles, are formed (Table 9) [50]. As a result of this balance the length of the hydrophilic chains forming the steric barrier must be reduced in size when compared to those used in other sterically stabilized vesicle formulations. However while it is well known that the lengths of the polymer chains are important in determining stability, it is also realized that the coating density is also critical [44,45]. At present the precise relationship between chain length and density of coverage for vesicle stability is not known. What is known however, is that the vesicles produced by these novel surfactants exhibit a considerably improved in vitro stability when compared to vesicles
267
Microemulsions and vesicles
produced largely from phospholipid. For example the improved stability with respect to vesicle size seen upon changing the hydrophilic head moeity from a phosphorylcholine group to a polyoxyethylene glycol chain is shown in Figure 8. A similar increase in in vitro stability was seen when long chain polyoxyethylene derivatives of phosphatidylethanolamine were incorporated into phospholipid vesicles [43J. The vesicles produced from the novel surfactants also exhibit a significantly reduced uptake by a murine macrophagelike cell line (1774) (Figure 9). This reduced uptake was thought to be a result of the sterically stabilized surface of the nonionic surfactant vesicles preventing adsorption of plasma proteins onto its surface and hence reducing the subsequent uptake into the phagocytic cells. Both these studies demonstrate the potential of novel vesicles as delivery vehicles. As these surfactants are novel no toxicity studies have yet been performed on them, but vesicles formed from these surfactants exhibited no obvious toxic effects on the 1774 cell line. It has been recently realized that vesicle size is an important factor in determining in vitro stability as
small vesicles are retained in the systemic circulation longer and as a consequnece can be taken up by non-macrophage cells. Interestingly the vesicles produced by these novel surfactants are smaller than those observed with phospholipid [49,50,51J. This obviously may have important consequences in the use of vesicles produced by these novel surfactants.
Conclusion By appropriate formulation it is possible to produce vesicles that exhibit sufficient in vitro and in vivo stability for many drug delivery applications. The use of novel surfactants allows the possibility of tailoring vesicle characteristics to their intended use.
ACKNOWLEDGEMENTS The author would like to thank C. Malcolmson, R. Aboofazeli, G. Musana, N. Patel. S. Chauhan and S.M. Lawrence for their invaluable help in obtaining the data.
268
Eur. J. Drug Metab. Pharmacokinet., 1994, No 3
REFERENCES 1. Lawrence, M.J. (1994) Non-ionic surfactant systems: their use in drug delivery. Chem, Soc. Rev. Inpress. 2. Mitchell, n.r., Tiddy, O.J.T., Waring, L., Bostock, T., McDonald, M.P. (1983) Phase behaviour of polyoxyethylene surfactant with water. Mesophase structures and partial miscibility (cloud points) J. Chem. Soc. Trans. I, 79, 975-1000. 3.
Aboofazeli, R, Lawrence, C.B., Wicks, S.R, Lawrence, MJ. (1994) Investigations into the formation and characterization of phospholipid microemulsions. III. Pseudoternary phase diagrams of systems containing waterlecithin-isopropyl myristate and either an alkanoic acid, amine, alkanediol, polyethylene glycol alkyl ether or alcohol as cosurfaetant. Int, J. Pharm. III, 63-72.
4. Langevin, D. (1986) Recent advances in the physics of microemulsions. Physica Scripta. T3, 252-258. 5. Rosano, H.L. (1984) Introduction. In H.L. Rosano & M. Clausse (eds) Microemulsions systems. Marcel Dekker, New York.
19. Aboofazeli, R, Lawrence, M.J. (1993) Investigations into the formation and characterization of phospholipid microemulsions. I. Pseudo-ternary phase diagrams of systems containing water-Iecithin-alcohol-ispropyl myristate, Int. J. Pharm., 93, 161-175. 20. Aboofazeli, R, Lawrence, M.J. (1994) Investigations into the formation and characterization of phospholipid microemulsions. II. Pseudo-ternary phase diagrams of systems containing water-Iecithin-ispropyl myristate and alcohol: Influence of purity of lecithin. Int, J. Pharm., 106, 51-61. 21. Attwood, D., Mallon, c., Taylor, CJ. (1992) Phase studies on oil-in-water phospholipid microemulsions. Int. J. Pharm., 84, R5-R8. 22. Shinoda, K., Araki, M., Sadaghiani, A., Khan, A., Lindman, B. (1991) Lecithin-based microemulsions: Phase Behaviour and microemulsions. J. Phys. Chern., 95, 989-993.
6. Danielsson, I., Lindman, B. (1981) The definition of microemulsion. Colloids Surf., 3, 391-392.
23. Malcolmson, c., Lawrence, M.J. (1994) Three-component nonionic oil-in-water microemulsions using polyoxyethylene ether surfactants. Colloids Surf. B. Accepted
7. Friberg, S.E. (1983) Comments on "the definition of microemulsion". Colloids Surf., 4, 201.
24. Schurtenberger, P., Peng, Q., Leser, M.E., Luisi, P.-L. (1993) Structure and phase behaviour of lecithin-based
Chevalier, Y., Zemb, T. (1990) The structure of micelles and microemulsions. Rep. Prog. Phys., 53, 279-371.
rnicroemulsions: A study of chain length dependence. J. Colloid Interface Sci., 156, 43-51.
9. Friberg, S.E. (1983) Microemulsions. Progr. Colloid Polym. Sci., 68, 41-47.
25. Eicke, H-P. (1979) On the cosurfaetant concept. J. Colloid Interface Sci., 68, 440- 450.
10. Siano, D.B. (1983) The swollen micelle-microemulsion transition. J. Colloid Interface Sci., 93, 1-7.
26. Mehta, C., Thomas, M., Lawrence, M.J. (1994) Lecithin based oil-in-water microemulsions. J. Pharm. Pharmacol. Accepted.
8.
11. Scriven, L.R (1977) Equilibrium bicontinuous structures. In K.L. Mittal (00) Micellization, solubilization and microemulsions. Vol 2., Plenum Press, New York. 12. Rushforth, D.S., Sanchez-Rubio, M., Santos-Vidals, LM., Wormuth, K.R, Kaler, E.W., Cuevas, R., Puig, lE. (1986) Structural study of one-phase microemulsions. J. Phys. Chern. 90. 6668-6673. 13. Davis, H.T., Bodet, J.P., Scriven, L.E., Miller, W.O. (1989)
Microstructure and transport in midrange mieroemulsions. Physica A, 157, 470-481. 14. Lam, A.C., Schechter, RS. (1987) The theory of diffusion in microemulsion. 1 Colloid Interface Sci., 120, 56-63. 15. Attwood, D. (1994) Microemulsions. In J. Kreuter (00) Colloidal Drug Delivery Systems. Marcel Dekker, New York. 16. Attwood, D., Florence, A.T. (1983) Surfactant systems. Their Chemistry, Pharmacy and Biology. Chapman and Hall, London. 17. Malcolmson, C, Lawrence, M.J. (1993) A comparison of the incorporation of steroids into nonionic micellar and microemulsion systems. J. Phann. Pharmacol., 45, 141-143. 18. Attwood, D., Mallon, C., Kstistis, 0., Taylor, CJ. (1989) A study on factors influencing the droplet size in oilin-water microemulsions. Int. J. Pharm., 52, 165-171.
27. Weincek, J.M., Qutubuddin, S. (1991) Solubilization in nonionic microemulsions. Colloids Surf., 54, 1-23. 28. Calfors, J., Blute, I., Schmidt, V. (1991) Lidocaine in microemulsion - a dermal delivery system. 1. Disp. Sci. Tech., 12, 467-482. 29. Mitchell, DJ., Ninham, B.W. (1981) Micelles, vesicles and microemulsions. J. Chem. Soc. Paraday Trans. 2, 77, 601-629. 30. Israelchvili, J.N., Mitchell, DJ., Ninham, B.W. (1976)
Theory of self assembly of hydrocarbon amphiphiles into micelles and bilayers. J. Chern. Soc. Paraday Trans. 2, 72, 1525-1567. 31. Lawrence, S.M., Lawrence, M.J., Barlow, D.J. (1991) VESICA: Comuter graphics modelling of lipid vesicles. J. Mol. Graphics, 9, 218-225. 32. Schulman, J.H., Stoeckenius, W., Prince, L.M. (1959) Mechansim of formation and structure of microemulsions by electron microscopy. J. Phys. Chern., 63, 1677-1680. 33. Aveyard, R, Binks, B.P., Flectcher, P.D.I. (1989) Interfacial tensions and aggregate structure in C12Es/oil/water microemulsion systems. Langmuir, 5, 1210-1217. &5 34. Binks, B.P., Meunier, J., Langevin, D. (1989) Characteristic sizes, film rigidity and interfacial tensions in
Microemulsions and vesicles
microemulsion systems. Progr. Colloid Polym. Sci., 79, 208-213. 35. Patel, N., Thomas, M., Aboofazeli, R & Lawrence, M.J. (1994) Personnel communication. 36. Elwonhy, P.H., Patel, M.S. (1982) Demonstration of maximium solubilization in a polyoxyethylene alkyl ether
series of nonionic surfactants, 1. Phann. Pharmacol., 34, 543-546. 37. Malcolmson, C.A. (1992) The physico-chemical properties of nonionic oil-in-water microemulsions. PhD Thesis, University of London. 38. Musana, G., Aboofazeli, R & Lawrence, MJ. (1994) Personnel communication. 39. Muller-Goymann, C.c., Hamann, H.-J. (1993) Sustained release from reverse micellar solutions by phase transformations into lamellar liquid crystals. J. Control. ReI., 23, 165-174. 40. Gregoriadis, G. (1983) In G. Gregoriadis, (ed) Liposome Technology. CRC Press Boca Raton, Florida 41. Waddle, M.e., Lasic, D.O. (1992). Sterically stabilized liposomes, Biochim. Biophys. Acta., 1113, 171-199. 42. Park, Y.S., Huang, L. (1993) Effect of chemically modified GMI and analogs of GMI on liposome circulation time: evidence supporting the dysopsonin hypothesis. Biochim. Biophys. Acta, ))66, 105-))4. 43. Blume, G., Cevc, G. (1993) Molecular mechansim of the lipid vesicle longevity in vivo. Biochim. Biophys. Acta.,
269
))46, 157-168. 44. Napper, D., (1984) In J.W. Goodwin, (ed) Stabilisation of Colloidal Dispersions. Royal Society of Chemistry,
London. 45. Tomlinson, E. (1987) Theory and practice of site-specific drug delivery. Adv. Drug. Del. Rev. I, 87-198. 46. Chauhan, S., Lawrence, M.J. (1989) The preparation of polyoxyethylene containing non-ionic vesicles. 1. Pharm. Pharmacol., 41(Suppl) 6P. 47. Lawrence, S.M., Tipple, S.c., Barlow, D.J., Lawrence, novel nonionic surfactant M.J. (1991). Structure of vesicles. Pharm. Res., 7(Suppl) 7291. 48. Lawrence, MJ., Lawrence, S.M., Chauhan, S. & Barlow, OJ. (1994) Synthesis and aggregation properties of dialkyl polyoxyethylene glycerol ethers. Submitted. 49. Chauhan, S.K. (1993) The synthesis and characterization of vesicle forming nonionic surfaetants. phD Thesis, University of London. 50. Lawrence, S.M. (1992) Characterization and computer modelling of nonvel nonionic surfactant vesicles. PhD Thesis, University of London. 51. Roberts, C.J., Davios, M.e., Jackson, D.E., Tandler, S.1.B., Lawrence, S.M., Lawrence, M.1., Chauhan., S., Barlow, D.J. (1991) Imaging of surfactant vesicles by scanning tunnelling microscopy. 1. Pharm. Phannaca., 43 (Suppl) 1368.