WO1985004578A1 - Steroidal liposomes - Google Patents

Abstract

Methods and compositions for the preparation of lipid vesicles, the bilayers of which comprise a salt form of an organic acid derivative of a sterol such as the tris-salt form of a sterol hemisuccinate. The method is rapid, efficient, and does not require the use of organic solvents. The sterol hemisuccinate vesicles may be used to entrap compounds which are insoluble in aqueous solutions. Such preparations are especially useful for entrapping bioactive agents of limited solubility thus enabling administration $i(in vivo).

Description

STEROIDAL LIPOSOMES 1. FIELD OF THE INVENTION

The present invention relates to methods and compositions for the entrapment of compounds in liposomes composed of salt forms of organic acid derivatives of sterols that are capable of forming bilayers.

Sterols such as cholesterol or other lipids, to which a hydrophilic moiety such as a salt form of an organi acid is attached, can be used to prepare suspensions of multilamellar or small unilamellar vesicles. The sterol liposomes of the present invention may be prepared with or without the use of organic solvents. These vesicles may entrap water-soluble compounds, partially water-soluble compounds, and water-insoluble compounds.

The sterol vesicles described herein are _t articularly useful for the entrapment of biologically active compounds or pharmaceutical compounds which can be administered _in_ vivo. Alternatively, the sterol liposomes of the present invention may be used _in_ vitro. For instance, the cholesterol hemisuccinate liposomes described herein may be used _in_ vitro in divalent cation-dependent assay systems.

2. BACKGROUND OF THE INVENTION

2.1. LIPOSOMES

Liposomes are completely closed bilayer membranes containing an encapsulated aqueous phase. Liposomes may be any variety of multilamellar vesicles (onion-like structure characterized by concentric membrane bilayers each separate by an aqueous layer ) or unilamellar vesicles (possessing a single membrane bi layer ) .

Two parameters of liposome preparations are ς funct ions of vesicle size and lip id concentration:

(1) Captured volume, defined as the volume e rnclosed by a given amount of lipid, is expressed as units of liters entrapped per mole of total lipid (1 rnol^" ). The captured volume depends upon the radius of the liposomes which in

•_jQ turn is affected by the lipid composition of the vesicles and the ionic composition of the medium. (2) Encapsulation efficiency, defined as the fraction of the aqueous compartment sequestered by the bilayers, is expressed as a percentage. The encapsulation efficiency is directly

15 proportional to the lipid concentration? when more lipid is present, more solute can be sequestered within liposomes. (See Deamer and ϋster, 1983, Liposome Preparation: Methods and Mechanisms, in Liposomes, ed. M. Ostro, Marcel Dekker, Inc., NY, pp. 27-51.)

20

The original method for liposome preparation (Bangha et al., 1965, J. Mol. Biol. 13.: 238-252) involved suspending phospholipids in an organic solvent which was then evaporated to dryness leaving a waxy deposit of

25 phospholipid on the reaction vessel. Then an appropriate amount of aqueous phase was added, the mixture was allowed to "swell," and the resulting liposomes which consisted of multilamellar vesicles (hereinafter referred to as MLVs) were dispersed by mechanical means. The structure of the

30 resulting membrane bilayer is such that the hydrophobic (non-polar) "tails" of the lipid orient toward the center of the bilayer while the hydrophilic (polar) "heads" orient towards the aqueous phase. This technique provided the basis for the development of the small sonicated unilamellar

35 vesicles (hereinafter referred to as SUVs) described by Pa ahadjopoulos and Miller (1967, Biochim. Biophys. Acta. 135: 624-638). Both MLVs and SUVs, however, have limitations as model systems.

In attempts to increase captured volume or encapsulation efficiency a number of methods for the preparation of. liposomes comprising physpholipid bilayers have been developed; however, all methods require the use o organic solvents. Some of these methods are briefly described below.

An effort to increase the encapsulation efficiency involved first forming liposome precursors or micelles, i.e., vesicles containing an aqueous phase surrounded by a monolayer of lipid molecules oriented so that the polar hea groups are directed towards the aqueous phase. Liposome precursors are formed by adding the aqueous solution to be encapsulated to a solution of polar lipid in an organic solvent and sonicating. The liposome precursors are then emulsified in a second aqueous phase in the presence of excess lipid and evaporated. The resultant liposomes, consisting of an aqueous phase encapsulated by a lipid bilayer are dispersed in aqueous phase (see U.S. Patent No. 4,224,179 issued September 23, 1980.to M. Schneider).

In another attempt to maximize the encapsulation efficiency, Papahadjopoulos (U.S. Patent No. 4,235,871 issued November 25, 1980) describes a "reverse-phase evaporation process" for making oligolamellar lipid vesicles also known as reverse-phase evaporation vesicles (hereinafter referred to as REVs) . According to this procedure, the aqueous material to be encapsulated is added to a mixture of polar lipid in an organic solvent. Then a homogeneous water-in-oil type of emulsion is formed and the organic solvent is evaporated until a gel is formed. The gel is then converted to a suspension by dispersing the gel-like mixture in an aqueous media. The REVs produced consist mostly of unilamellar vesicles (large unilamellar vesicles, or LUVs) and some oligolamellar vesicles which are characterized by only a few concentric bilayers with a large internal aqueous space.

Much has been written regarding the possibilities of using liposomes for drug delivery systems. See, for example, the disclosures in U.S. Patent No. 3,993,754 issued on November 23, 1976, to Yeuh-Erh Rahman and Elizabeth A. Cerny, and U.S. Patent No. 4,145,410 issued on March 20, 1979, to Barry D. Sears. In a liposome drug delivery syste the medicament is entrapped during liposome formation and then administered to the patient to be treated. The medicament may be soluble in water or in a non-polar solvent. Typical of such disclosures are U.S. Patent No. 4,235,871 issued November 25, 1980, to Papahadjopoulos and Szoka and U.S. Patent No. 4,224,179, issued September 23, 1980 to M. Schneider. When preparing liposomes for use in vivo it would be advantageous (1) to eliminate the necessit of using organic solvents during the preparation of liposomes; and (2) to maximize the encapsulation efficiency and captured volume so that a greater volume and concentration of the entrapped material can be delivered pe dose.

2.2. WATER-SOLUBLE STEROLS

A variety of sterols and their water soluble derivatives have been used for cosmetic, pharmaceutical and diagnostic purposes. Of the water soluble sterols, for example, branched fatty acid cholesterol esters, steroid esters and PEG-phytosterols have been used in cosmetic preparations (European Patent Application No. 28,456; U.S. Patent No. 4,393,044; and Schrader, Drug and Cosmetic Industry, September 1983, p.33 and October 1983, p.46). Thakkar and Kuehn (1969, J. Pharm. Sci. 58(7): 850-852) disclose the solubilization of steroid hormones using aqueous solutions of steroidal non-ionic surfactants. specifically ethoxylated cholesterol (i.e., PEG-cholesterol at a concentration of 1-5%. However, the effectiveness or utility of the solubilized steroid hormones .in. vivo was not demonstrated. A number of water soluble cholesterols have been prepared and used as water-soluble standards for the determination of cholesterol levels in biological fluids (U.S. Patent No. 3,859,047; U.S. Patent No. 4,040,784; U.S. Patent No. 4,042,330; U.S. Patent No. 4,183,847; U.S. Paten No. 4,189,400; and U.S. Patent No. 4,224,229). Shinitzky e al. (1979, Proc Natl . Acad. Sci. USA 76:5313-5316) incubated tumor cells in tissue culture medium containing a low concentration of cholesterol and cholesteryl hemisuccinate. Incorporation of cholesterol or cholesteryl hemisuccinate into the cell membrane decreased membrane fluidity and increased membrane-lipid microviscosity.

Cholesterol and other sterols, have also been incorporated into phospholipid liposome membranes in order to alter the physical properties of the lipid bilayers. Fo example, in a recent abstract, Ellens et al. (1984, Biophys J. _45_: 70a) discuss the effect of H on the stability of lipid vesicles composed of phosphatidylethanolamine and cholesteryl hemisuccinate. In fact, Brockerhoff and Ramsammy (1982, biochim. Biophys. Acta. 691:227-232) reported that bilayers can be constructed which consist entirely of cholesterol, provided a stabilizing hydrophilic anchor is supplied. Multilamellar and unilamellar cholesterol liposomes were prepared in a conventional manne described above evaporating to dryness the cholesterol derivatives (i.e. , cholesterol-phosphocholine, cholesterol- polyethylene glycol, or cholesterol-SO.) dispersed in an organic solvent leaving a lipid film deposited in the reaction vessel. The lipid films were sonicated under 2mA water using an ultrasonic homogenizer with a microtip. Formation of multilamellar vesicles required 10 minutes sonication, whereas formation of small unilamellar vesicles required 4 hours of sonication. The resulting suspensions of multilamellar liposomes were milky whereas the suspensions of unilamellar liposomes were transparent.

However, the ability to efficiently entrap bioactive agents in sterol vesicles which are suitable for administation iji. vivo to provide for the administration of higher doses of water-soluble agents and to facilitate the administration of water-insoluble agents has not heretofore been explored.

3. SUMMARY OF THE INVENTION

The present invention involves methods and compositions for the entrapment of various compounds in liposomes, the bilayers of which comprise salt forms of organic acid derivatives of sterols. Entrapment of a compound is defined herein as the encapsulation of a water- soluble compound in the aqueous compartment of the liposome or the entrapment of a water-insoluble compound within the sterol bilayer. The sterol liposomes of the present invention are particularly useful for the administration of the entrapped compound .in. vivo, in which case a biocompatible salt form of an organic acid derivative of a sterol should be used to prepare the liposomes. In fact, for .in. vivo administration, the tris(hydroxymethyl)aminomethane salt (tris-salt) form of organic acid derivatives of sterols are particularly useful as the vesicle bilayer ingredient.

The method for preparing the sterol vesicles involves adding to an aqueous buffer a salt form of an organic acid derivative of a sterol capable of forming closed bilayers in an amount, sufficient to form completely closed bilayers which entrap an aqueous compartment. A suspension of multilamellar vesicles is formed by shaking the mixture. The formation of vesicles is facilitated if the aqueous buffer also contains the counterion of the salt in solution. Furthermore, if the dissociated salt form of the organic acid derivative of a sterol is negatively charged at neutral pH, the aqueous buffer should be* essentially free of divalent or multivalent cations.

Similarly, when the dissociated salt form of the organic acid derivative of a sterol is positively charged at neutra pH, the aqueous buffer should be essentially free of multivalent anions. The application of energy to the suspension, e.g. , sonication, or extrusion of the vesicles through a French pressure cell (French Press) or through a porous filter of the appropriate pore size, will convert th multilamellar sterol vesicles to unilamellar vesicles.

In order to entrap a water-soluble compound, a partially water-soluble compound or a water-insoluble compound in the sterol vesicles of the present invention, a number of approaches are possible. Compounds which either partition into the sterol bilayers (e.g. , water-insoluble compounds) or water-soluble compounds may be added to the aqueous phase before formation of the vesicles in order to entrap the agent within the vesicles during formation. Alternatively, compounds which are water-insoluble or lipid soluble may be added to the suspension of sterol vesicles after the vesicles are formed, in which case the compound partitions into the sterol bilayers. In another embodiment, a water-soluble compound and the salt-form of an organic acid derivative of a sterol may be added to an organic solvent so that both are solubilized (co-solubilized) . The organic solvent may then be evaporated leaving a film containing a homogeneous distribution of the water-insoluble compound and the sterol derivative. Sterol liposomes entrapping the water-insoluble compounds are formed when an aqueous buffer is added to the film with shaking.

The sterol liposomes of the present invention are particularly advantageous when used to entrap water- insoluble bioactive agents or those that are sparingly -8- soluble in water . This enables the administration ^n. vivo of water-insoluble drugs ; and furthermore , it allows for th administration .in. vivo of high concentrations of the water insoluble compounds , because it allows alteration of the

* dose : volume ratio . The sterol vesicles of the present invention offer similar advantages when used to entrap wate soluble bioactive agents . In addition , the sterol vesicles of the present invention may be used in diagnostic assays i vitro.

The present invention affords a number of advantages in that the sterol vesicles:

(1) are formed easily and rapidly;

(2) have high encapsulation efficiencies as compared with phospholipid MLVs;

(3) do not require the use of organic solvents fo their preparation (although the sterol vesicles of the present invention can be prepared using organic solvents) ; and

(4) can entrap a bioactive or pharmaceutical agent, which when administered .in. vivo, is released and metabolized. The fate of the entrapped agent .in. vivo depends upon the mode of administration.

4. BRIEF DESCRIPTION OF THE FIGURES'

FIG. 1 graphically demonstrates the inverse relationship of the captured solute (chromium) and the concentration of cholesterol hemisuccinate used to prepare the multilamellar liposomes. FIG . 2 represents the X-ray diffraction patterns obta ined for four diff erent CHS-MLV preparat ions .

FIG. 3 represents the electron spin resonance dat for CHS-multilamellar vesicles and EPC-muItilamellar vesicles.

FIG. 4 graphically demonstrates the swelling profiles of cholesterol hemisuccinate liposomes and egg phosphatidylcholine liposomes in aqueous buffers of various tonicity.

FIG. 5 graphically illustrates the effectiveness o indomethacin entrapped in cholesterol hemisuccinate liposomes in reducing joint swelling when administered intramuscularly.

FIG. 6 represents the_ organ distribution of 14C- diazepam administered intravenously in mice either unencapsulated (free) or encapsulated in CHS-SUVs.

FIG. 7 A, B and C, represent the organ distributio of Chromium administered intravenously in mice either, encapsulated in CHS-MLVs (FIG. 7A) or encapsulated in EPC- SPLVs (FIG. 7B) or unencapsulated (FIG. 7C).

5. DETAILED DESCRIPTION OF THE INVENTION

The present invention describes methods and compositions for the entrapment of water-soluble, partially water-soluble, or water-insoluble compounds in liposomes, the bilayers or which comprise salt forms of organic acid derivatives of sterols that are capable of forming closed bilayers. Accordingly, the sterol liposomes of the present invention can be prepared to (1) entrap a water-soluble compound in the aqueous compartment; or (2) entrap a water- insoluble compound which partitions into the sterol bi layers; or (3) both entrap a water-soluble compound and entrap a water-insoluble compound in one liposome preparation.

Any salt form of an organic acid derivative of a sterol which is capable of forming completely closed bilayers in aqueous solutions ( i.e. , liposomes) may be used in the practice of the invention. The suitability of a particular salt-form of an organic acid derivative of a sterol depends upon its ability to sequester a water-solubl compound so that the compound is not in contact with the outside environment.

To ^"determine definitively that entrapment within the aqueous compartment of any liposome has occurred, the following criteria have been established (See Sessa and Weissmann, 1970, J. Biol. Chem. 245: 3295): (a) there must be a clear separation of free from sequestered compound as assayed by gel filtration; (b) there must be no hydrophobic or charge-charge interaction between the outermost vesicle bilayer and the entrapped compound since this may result in a failure to achieve separation of the free compound from the liposomes by molecular sieving, thereby artificially increasing the apparent sequestration or encapsulation efficiency. To eliminate this possibility it must be shown that the water-soluble compound added to a suspension of previously formed liposomes does not coelute with preformed liposomes; (c) disruption of gel-filtered liposomes by use of detergents or other membrane perturbants must induce a shift in the gel filtration pattern of the sequestered molecule from a position coincident with the liposome peak to one that coelutes with the free molecule.

Generally any sterol which can be modified by the attachment of an organic acid may be used in the practice o the present invention. For example, such sterols include but are not limited to cholesterol, vitamin D, phytosterols ( including but not limited to sitosterol, campesterol, stigmasterol, and the like), steroid hormones, and the like

Organic acids which can be used to derivatize the sterols include but are not limited to the carboxylic acids dicarboxylic acids, polycarboxylic acids, hydroxy acids, amino acids and polyamino acids. Because the salt forms increase the water solubility of organic acids, any organic acid may be used to derivatize the sterols; however an advantage may be obtained if the organic acid moiety itself is water. soluble. Such water-soluble organic acid moieties include but are not limited to water-soluble aliphatic carboxylic acids such as acetic, propionic, butyric, valeri acids and the like (N.B. , up to four-carbon acids are miscible with water; the five-carbon free acid is partly soluble and the longer chain free acids are virtually insoluble); water-soluble aliphatic dicarboxylic acids such as malonic, succinic, glutaric, adipic, pimelic, maleic and the like (N.B., the shorter chains are appreciably more soluble in water; borderline solubility in water occurs at C_g to C₇) ; and water-insoluble aromatic dicarboxylic acids such as hemimellitic, trimesic, succinimide, and the like; polycarboxylic acids; water-soluble hydroxy acids such as glycolic, lactic, mandelic, glyceric, malic, tartaric, citric, and the like (N.B. , α-hydroxy acids containing a branched chain attached to the α-carbon of the carbonyl group would be less susceptible to hydrolysis and, therefore, advantageous in the practice of the present invention); and any of the amino acids and polyamino acids.

The organic acid can be linked to an hydroxyl group of the sterol via an ester or an ether bond using conventional methods (see, for example, U.S. Patent No. 3,859,047; U.S. Patent No. 4,040,784; U.S. Patent No. 4,042,330; U.S. Patent No. 4,183,847; and U.S. Patent No. 4,189,400). The salt forms of the derivatized sterols can be prepared by dissolving both the organic acid derivative of the sterol and the counterion of the salt (e.q., the fre base of the salt) in an appropriate volatile solvent, and removing the solvent by evaporation or a similar techique leaving a residue which consists of the salt form of the organic acid derivative of the sterol. Counterions that ma be used include, but are not limited to, tris, 2-amino-2- methyl-l,3-propanediol, 2-aminoethanol, bis-tris propane, triethanolamine, and the like to form the corresponding salt. In fact, the free base of an ionizable bioactive agent such as miconazole free base and the like may be used as the counterion.

The sterol liposomes of the present invention may be prepared by adding to an aqueous phase a salt form of an organic acid derivative of a sterol capable of forming bilayers so that the derivatized sterol is present in an amount sufficient to form vesicles (i.e., completely closed bilayers containing an entrapped aqueous compartment). The preparation is then shaken until a milky suspension of multilamellar sterol vesicles is formed. In the preferred embodiment, the aqueous phase should contain the salt in solution to facilitate vesicle formation. Furthermore, if the dissociated salt form of the organic acid derivative of the sterol is negatively charged at neutral pH, the aqueous buffer should be essentially free of multivalent cations. Similarly, when the dissociated salt form of the organic acid derivative is positively charged at neutral pH, the aqueous buffer should be essentially free of multivalent anions.

In complete contrast to reported methods for multilamellar vesicle formation (e.g., phospholipid vesicle or the cholesterol liposomes of Brockerhoff and Ramsammy, 1982, Biochim. Biophys. Acta. 691: 227-232), the method for the formation of the sterol multilamellar vesicles of the present invention does not require the use of organic solvents. Furthermore, unlike the method of Brockerhoff an Ramsammy (supra) sonication is not necessary to form the sterol multilamellar vesicles. In fact, sonciation of the milky suspension of sterol multilamellar vesicles of the present invention, or the use of a French press (SLM-Aminco

Urbana, ILL) followed by sonication may be used to convert the milky suspension of multilamellar sterol vesicles to a clear suspension of unilamellar sterol vesicles. Similarly multiple extrusions of the multilamellar sterol vesicles at moderate pressures through a filter have a pore size of equal to or less than 100 nm in diameter can be employed to obtain unilamellar sterol vesicles. This extrusion technique is described in detail in co-pending application

Serial No. 622,690 filed June 20, 1984 by Cullis et al. for

"Extrusion Technique for Producing Unilamellar Vesicles" which is incorporated by reference herein.

As previously explained, the tris-salt form of any organic acid derivative of a sterol may be advantageously used in the practice of the present invention. For example the tris-salt form of a sterol hemi-dicarboxylic acid such as a sterol hemisuccinate or a mixture of sterol hemisuccinates are particularly useful for forming the vesicle bilayers of the steroidal liposomes to be administered in vivo. For instance, when using cholesterol hemisuccinate, 2.5 to 700 ymoles of the tris-salt form may be added to 2.0 mi aqueous buffer containing Tris-HCl (tris(hydroxymethyl)aminomethane hydrochloride) in order to form vesicles; in this case the aqueous buffer should be essentially free of divalent or multivalent cations.

According to the present invention, the entrapment of water-soluble compounds, water-insoluble compounds, or sparingly soluble compounds in liposomes composed of the salt form of organic acid derivatives of sterols may be accomplished in a number of ways: (1) A water-insoluble compound can be added to a suspension of sterol liposomes (either multilamellar sterol vesicles or unilamellar sterol vesicles), which were prepared as described above using an appropriate salt form of an organic acid derivative of a sterol. The compound is entrapped in the liposomes because it partitions into the sterol bilayers. This embodiment may be conveniently carried out as follows: the water-insoluble compound may be dissolved in an appropriate organic solvent which is then evaporated leaving a film or residue of the compound. When an aqueous suspension of previously formed sterol liposomes is added to the residue, the residue will be entrapped in the bilayers of the sterol liposomes. In the preferred embodiment, unilamellar sterol vesicles should be used; if multilamellar sterol vesicles are used, the water-insoluble compound may be entrapped in the outermost bilayers of the liposomes, leaving the innermost bilayers unaltered resulting in a unnecessary use of the sterol ingredient.

(2) A water-insoluble compound and the salt form of an organic acid derivative of a sterol can both be co- solubilized in an organic solvent which is then evaporated off leaving a film comprising a homogeneous distribution of the water-insoluble compound and the sterol derivative. A suspension of multilamellar sterol vesicles containing the entrapped compound is formed with an aqueous phase is added to the film with shaking. The multilamellar vesicles may be converted to unilamellar vesicles as previously described.

(3) A water-soluble compound or a water-insoluble compound can be entrapped in the sterol liposomes by adding the compound to the aqueous phase which is used in the preparation of the sterol vesicles; i.e., the compound can be added to the aqueous phase before or simultaneously with the addition of the salt form of an organic acid derivative of a sterol. In this case, a water-insoluble compound becomes entrapped when it partitions into the bilayers during vesicle formation; whereas a water-soluble compound becomes entrapped in the aqueous compartment of the sterol vesicles during vesicle formation. In either case, the multilamellar vesicles can be converted to unilamellar vesicles as previously described.

(4) If the bioactive agent is ionizable, the free base of the bioactive agent may be used as the counterion t prepare the salt form of the organic acid derivative of a sterol. The sterol liposomes may be prepared by any of the methods previously described herein using the bioactive agent-salt form of the organic acid derivative of the sterol. For example, the free base of miconazole, an anti- fungal compound, may be used to make the salt derivatives i this embodiment of the present invention.

Using any of the four method described above, both a water-soluble compound and a water-insoluble compound may be entrapped in one sterol liposome preparation.

According to the methods described above for the entrapment of water-insoluble compounds using the sterol vesicles of the present invention, it is not required that the vesicles remain intact once a water-insoluble compound partitions into the bilayers. In fact, it is conceivable that once the compound partitions into the bilayers the vesicles will be disturbed or disrupted leading to the leakage or release or aqueous entrapped compounds. Althoug these "leaky" vesicles could be used to deliver the entrapped water-insoluble compound, they should not be used to encapsulate or deliver a water-soluble substance.

According to one embodiment of the present invention, sterol liposomes are prepared using the tris-salt form of cholesterol hemisuccinate as follows: 4.5 to 200 g of the tris-salt form of cholesterol hemisuccinate is added per ml or aqueous buffer containing 0.01 M Tris-HCl, 0.14 M NaCl. The mixture is shaken and a milky suspension of cholesterol hemisuccinate multilamellar vesicles forms. The vesicles may be pelleted by centrifugation and washed repeatedly with aqueous buffer. The suspension of cholesterol hemisuccinate multilamellar vesicles (CHS-MLVs) may be sonicated (e.g. in a bath-type sonicator) in order to form cholesterol hemisuccinate small unilamellar vesicles (CHS-SUVs). Alternatively, the CHS-MLVs may be passed through a French pressure cell (a French Press) at 40,000 psi or the CHS-MLVs may be passed through two 100 nm Nucleopore (TM) filters at 300-400 pa in order to form CHS- SUVs. The cholesterol hemisuccinate vesicles (whether MLVs or SUVs) are unstable in the presence of divalent cations; i.e. upon exposure to divalent cations the entrapped aqueous compartment and water-soluble compounds are released. Thus, the aqueous medium used in the preparation or during storage of the vesicles should be essentially free of divalent cations.

The compounds which are entrapped according to the method of the present invention may be used in various ways. For example, if the compound is a bioactive agent, the sterol liposome entrapped compound may be administered in vivo. This facilitates the .in. vivo delivery of bioactive agents which are normally insoluble or sparingly soluble in aqueous solutions. Entrapment in liposomes composed of the salt form of organic acid derivatives of sterols enables ease in the administration of such insoluble compounds at a higher dose:volume ratio. In fact, the sterol vesicles of the present invention are particularly advantageously used in vivo because the vesicles may be used to entrap one or more bioactive agents for delivery .in. vivo*. Furthermore, the vesicles of the present invention offer an advantage over conventional lipid vesicles or liposomes when used in vivo because they can be prepared without using organic solvents. The fate of the entrapped agent in vivo depends upon the route or mode of administration. For instance. when the sterol liposome entrapped agent is administered intravenously the cle arance of the agent .in. vivo follows a pathway different from that of non-entrapped agent or that of an agent entrapped in convent ional liposomes composed of phospholipids ( i . e . , MLVs , SUVs , REVs , LUVs ) . On the other hand , intramuscul ar administrat ion of the sterol liposome entrapped agent resu l ts in a sustained re lease of the agent in vivo .

Virtually any bioactive agent can be entrapped within the sterol liposomes of the present invention. Such agents include but are not limited to antibacterial agents, antiviral agents, antifungal agents, anti-parasitic agents, tumo icidal agents, anti-metabolites, polypeptides, peptides, proteins, toxins, enzymes, hormones, neurotransmitters-, glycoproteins, lipoproteins, immunoglobulins, immunomodulators, vasodilators, dyes, radiolabels, radio-opaque compounds, fluorescent compounds, receptor binding molecules, anti-inflammatories, antiglaucomic agents, mydriatic compounds, local anesthetics, narcotics, vitamins, nucleic acids, polynucleotides, etc. The entrapment of two or more compounds simultaneously may be especially desirable where such compounds produce complementary or synergistic effects

The sterol liposome entrapped agent may be administered .in. vivo by any suitable route including but no limited to: inoculation or injection (e. ., intravenous, intraperitoneal, intramuscular, subcutaneous, intra-aural, intraarticular,intra-mammary, and the like), topical application (e.g. , on areas such as eyes, skin, in ears or on afflictions such as wounds and burns), and by absorption through epithelial or mucocutaneous linings (e.g., nasal, oral, vaginal, rectal, gastrointestinal mucosa, and the like). In another example of their use, the sterol liposome entrapped compound may be incorporated into a broa range of materials including but not limited to other lipid vesicles or liposomes, gels, oils, emulsions and the like.

For instance, the suspension of sterol liposomes containing the entrapped compound may be added to the aqueous phase as an ingredient in any type of liposome preparation (e. ., phospholipid MLVs, SUVs, LUVs, REVs, and others). This allows for the entrapment of the compound in the phospholipid liposomes.

Other uses, depending upon the particular- properties of a preparation, may be envisioned by those skilled in the art. For example, because of their divalent cation sensitivity, the cholesterol hemisuccinate liposomes of the present invention may be made to entrap indicator dyes which are sensitive to divalent cation sensitive for use in colorimetric diagnostic assays .in. vitro.

The following examples are given for purposes of illustration and not by way of limiting the scope of the invention.

6. EXAMPLE: CHOLESTEROL HEMISUCCINATE

LIPOSOMES ENTRAPPING WATER-SOLUBLE COMPOUNDS

The following subsections describe the preparation of cholesterol hemisuccinate vesicles which entrap arsenazo III, inulin or chromium. Parameters such as the encapsulation efficiency and the captured volume are assessed; the calcium dependent instability of the cholesterol vesicles is demonstrated. Freeze-etch electron microscopy. X-ray diffraction and electron spin resonance o the cholesterol hemisuccinate vesicles are also described. 6.1. LIPOSOMES PREPARED USING VARIOUS SALT FORMS OF CHOLESTEROL HEMISUCCINATE

The following subsections describe the preparation of CHS liposomes using various salt forms of cholesterol hemisuccinate.

In all examples involving the tris-salt form of

__* cholesterol hemisuccinate (hereinafter referred to as tris- salt CHS) the tris-salt CHS was either purchased from Sigma Biochemicals, St. Louis, MO, and used without purification or synthesized as follows: 30 ml of a 3.3 molar solution o Tris base was added to 1.5 liters of a 67 M molar solution of cholesterol hydrogen succinate (ICN, Cleveland, Ohio) in ether. The resulting solution was rotoevaporated to a wet residue and lyophilized for 12 hours. The resulting tris- salt CHS was recrystallized three times from ethyl acetate. Residual ethyl acetate was removed by heating to 56°C under vacuum (0.1 mm Hg) .

6.1.1. TRIS-SALT CHOLESTEROL HEMISUCCINATE-MLVs

Tris-salt CHS (54 mg) was added to a 1 ml solution of arsenazo III (4.5 M, final concentration) in 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl. A milky suspension of CHS- MLVs was formed by mechanical shaking. The CHS-MLVs were pelleted by centrifugation at 10,000 x g for 15 minutes, an the resulting pellet was washed three times using 10 ml 0.0 M Tris-HCl (pH 7.3), 0.14 M NaCl. The resulting pellet was red in color indicating entrapment of the arsenazo III.

6.1.2. 2-AMINO-2-METHYL-1,3-PROPANEDIOL CHOLESTEROL HEMISUCCINATE-MLVs

The 2-amino-2-methyl-l,3-propanediol salt of cholesterol hemisuccinate (50 mg) was added to a 1 mil solution of arsenazo III (4.5 mM, final concentration) in 0.01 M 2-amino-2-methyl-l,3-propanediol-HCl (pH 7.3), 0.07 KC1, 0.07 M NaCl. The suspension of CHS-MLVs was formed by mixing vortically with glass beads. The CHS-MLVs were pelleted by centrifugation at 10,000 x g for 15 minutes, and the resulting pellet was washed three times as described in Section 6.1.1. The resulting pellet was in color indicating entrapment of the arsenazo III.

6.1.3 2-AMIN0ETHAN0L CHOLESTEROL HEMISUCCINATE-MLVs

The 2-aminoethanol salt of cholesterol hemisuccinate (50 mg) was added to a 1 mi solution of arsenazo III (4.5 mM, final concentration) in 0.01 M 2- aminoethanol-HCl (pH 7.3), 0.07 M KC1, 0.07 M NaCl. The suspension of CHS-MLVs was formed by mixing vortically with glass beads. The CHS-MLVs were pelleted by centrifugation at 10,000 x g for 15 minutes, and the resulting pellet was washed three times as described in Section 6.1.1. The resulting pellet was red in color indicating entrapment of the arsenazo III.

6.1.4. BIS-TRIS-PROPANE CHOLESTEROL HEMISUCCINATE-MLVs

The bis-tris-propane salt of cholesterol hemisuccinate (50 mg) was added to a 1 ml solution of arsenazo III (4.5 mM, final concentration) in 0.01 M bis- tris-p o ane-HCl (pH 7.3), 0.07 M KC1, 0.07 M NaCl. The suspension of CHS-MLVs was formed by mixing vortically with glass beads. Ttie CHS-MLVs were pelleted by centrifugation at 10,000 x g for 15 minutes, and the resulting pellet was washed three times as described in Section 6.1.1. The resulting pellet was red in color indicating entrapment of the arsenazo III. 6.1.5. TRIETHANOLAMINE CHOLESTEROL HEMISUCCINATE-MLVs

The triethanolamine salt of cholesterol hemisuccinate (50 mg) was added to a 1 ml solution of arsenazo III (4.5 mM, final concentration) in 0.01 M triethanolamine-HCL (pH 7.3), 0.07 M KC1, 0.07 M NaCl. The suspension of CHS-MLVs was formed by mixing vortically with glass beads. The CHS-MLVs were pelleted by centrifugation at 10,000 x g for 15 minutes, and the resulting pellet was washed three times as described in Section 6.1.1. The resulting pellet was red in color indicating entrapment of the arsenazo III.

6.1.6. MICONAZOLE CHOLESTEROL HEMISUCCINATE-MLVs

The free base of miconazole was prepared as follows: an aqueous solution of NaOH was titrated into a suspensian of miconazole-nitrate in ether. The ether phase was collected and the ether was evaporated leaving an oil comprising the miconazole free base. The oil was then adde to ethanol containing cholesterol hydrogen succinate. The ethanol was evaporated leaving a film comprising the salt form of miconazole cholesterol hemisuccinate. Then a salin solution was added to the film. After extensive vortical mixing, vesicles were observed in the solution.

6.1.7. CHOLESTEROL HEMISUCCINATE-SUVs PREPARED BY SONICATION

CHS-MLVS were prepared as described in Sections 6 . 1. 1. , 6. 1.2. , 6 . 1.3. , 6. 1.4. , and 6. 1.5. , except that the arsenazo II I was omitted . Each f inal pellet of vesicles wa resuspended in 2 ml of the buf fer in which it was prepared and sonicated in a bath sonicator until the milky suspensio turned clear indicating the conversion of CHS-MLVs to CHS- SUVs . 6.1.8. CHOLESTEROL HEMISUCCINATE-SUVs

PREPARED BY EXTRUSION TECHNIQUES

CHS was dispersed in 10 M HEPES, 150 mM NaCl (pH 7.5) at a concentration of 100 mg/mϋ. This material was extruded 10 times through a 30 nm nucleopore filter resulting in CHS-SUVs.

6.2. ENTRAPMENT OF INULIN IN

CHOLESTEROL HEMISUCCINATE MLVs

Cholesterol hemisuccinate multilamellar vesicles (CHS-MLVs) incorporating H-Inulin as the entrapped agent were prepared as follows: H-Inulin (1.0 mCi/mi, New England Nuclear, Boston, MA) was dissolved in 2 mil 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl. Then 40 mg of tris-salt CH was added to the solution and the resulting mixture was mechanically dispersed by shaking. A milky suspension formed indicating the formation of multilamellar vesicles. The suspension was allowed to stand undisturbed for 2 hours at which time the suspension was diluted to a final volume of 10 mil using 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl. The radioactivity of a lOμil aliquot was determined to be 24,625 cpm/10 y il by adding the aliquot to 10 mil scintillation flui (40 g Omnifluor (New England Nuclear, Boston, MA) 6 toluene, 4 ethylene glycol monoethyl ether) and assaying radioactivity using a Beckmann L6800 liquid scintillation counter with windows set at 0.400. Radioactivity in counts per minute (cpm) was converted to disintergrations per minute (dpm) by applying the H# method of quench correction (Horrock, D.L. The Number Concept, Beckman Instruments, 1977). The CHS-MLVs were then pelleted by centrifugation a 10,000 x g for 15 minutes. The resulting pellet was washed three times by resuspending the pellet in 10 ml 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl and repelletting by centrifugation at 10,000 x g for 15 minutes. The washed pellet of vesicles was resuspended in 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl to a final volume of 10 mi_? the radioactivity of a 10 y Jl aliquot was determined to be 3,442 cpm/10 μi. Therefore, a total of approximately 14% of the starting H-Inulin was entrapped in the CHS-MLVs.

6.2.1. ENCAPSULATION EFFICIENCY OF INULIN

IN CHOLESTEROL HEMISUCCINATE-MLVs AND EGG PHOSPHATIDYLCHOLINE-MLVS

The encapsulation efficiencies of inulin entrapped in MLVs comprising varying concentrations of cholesterol hemisuccinate were compared to encapsulation efficiencies o inulin entrapped in MLVs comprising varing concentrations o egg phosphatidylcholine . (N.B . , encapsulation efficiency for any liposome is defined as the fraction of aqueous compartment sequestered by bilayers; and is expressed as a percentage, see Section 2.1 supra.)

Multilamellar vesicles composed of either egg phosphatidylcholine (EPC) or tris-salt CHS were prepared using identical protocols in order to compare encapsulation efficiencies. Accordingly, tris-salt CHS at a concentratio of 40, 80, 160, 320 or 400 mg in 2.0 mi 0.01 M Tris-HCl (pH

7.3), 0.14 M NaCl buffer with 5 μi ³H-inulin (217.0 mCi/mg) was mixed vortically and allowed to stand for 2 hours,

3 forming CHS-MLVs with H-mulin as the entrapped compound.

An additional 3 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer was added to the suspension which was left at room temperature overnight. Then, approximately 3.0 mi 0.01 M

Tris-HCl (pH 7.3), 0.14 M NaCl buffer was added to bring th total volume up to 10 mi.

Multilamellar vesicles composed of egg phosphatidylcholine (EPC-MLVs) (Avanti, Birmingham, AL) wer prepared according to the following protocol: 40, 80, 160, 320 or 400 mg/mi EPC was suspended in sufficient cholorofor to completely dissolve the phospholipid. The choloroform was evaporated to dryness leaving a waxy deposit on the tes tube. Then 2.0 mi of 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer with 5 μi H-Inulin (217.0 mCi/mg) was added, the mixture allowed to "swell", and the resulting EPC-MLVs were dispersed by extensive vortical mixing. An additional 3 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer was added to the suspension which was left at room, temperature overnight. Then, the mixture was brought up to a total volume of 10 mi with 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer.

The encapsulation efficiency of H- inulin by the CHS-MLVs and EPC-MLVs was determined as follows: The radioactivity in the 20 yi aliquot of each initial mixture of ingredients was determined by scintillation counting as previously described. After formation, the liposomes were pelleted by centrifuging the suspension for 10-20 minutes at 10,000 x g, each pellet was washed four times in 10 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer, and resuspended to a final volume of 10 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer. The radioactivity of a 20 ul aliquot of this final washed sample was determined. The fraction of the initial radioactivity measured in this final sample represented the H-inulin entrapped in the lipid vesicles.

As illustrated in Table I, an increase in encapsulation efficiency is proportional to an increase in CHS concentration but, more importantly, CHS-MLVs made using 20-200 mg/ i CHS demonstrate a higher encapsulation efficiency for inulin than do EPC-MLVs made using the same concentration of phospholipid. T A B L E I

COMPARISON OF ENCAPSULATION EFFICIENCIES OF INULIN IN PHOSPHOLIPID VESICLES AND CHOLESTEROL HEMISUCCINATE VESICLES

3

Concentration of % H-Inulin Entrapped

Lipid (mg/mi) EPC-MLVs^a CHS-MLVs

20 2 10

40 4 14

80 5 29

1 16600 8 38

200 11 60

a Egg phosphatidylcholine multilamellar vesicles, b Cholesterol hemisuccinate multilamellar vesicles.

In order to determine whether the encapsulation efficiency of CHS-MLVs was influenced by the amount of time the CHS was in contact with free 3 H-Inulm in aqueous buffer, CHS-MLVs were prepared as follows: either 80 or 30 mg tris-salt CHS was mixed vortically in 20 mi 0.01 M Tris- HCl, 0.14 M NaCl buffer containing 10 μi ³H-Inulin (217 mCi/mg specific radioactivity), thus forming CHS-MLVs using a concentration of 40 and 150 mg/mi CHS respectively.

Five samples at each of the two lipid concentrations were prepared and the CHS-MLV suspensions were allowed to stand at room temperature in the 2.0 mi of 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer. At the following time intervals: 0, 15, 30, 60, and 120 minutes, the samples were brought up to 10 mi with the same buffer. An initial 10 y i aliquot of each sample was removed for scintillation counting as previously described. The sample were then centrifuged at 10,000 x g for 10 minutes, and eac pellet was washed four times in 10 mi buffer. The final pellet was suspended in buffer to a final volume of 10 mi, and the radioactivity of a 10 aliquot of each final sample was compared to that of the initial sample at each time point. The results demonstrated no significant difference in encapsulation efficiency for the five time points for either concentration of lipid tested. This indicates that entrapment of about 12% or 20% of the initial 3 H-Inulin added to the preparation occurred regardless of contact time at 40 or 150 mg/mi CHS respectively. This demonstrates that unlike conventional MLVs prepared using egg phosphatidylcholine, no "swelling time" is required in the preparation of CHS-MLVs.

6.3. ENTRAPMENT OF INULIN IN _ CHOLESTEROL HEMISUCCI ATE-SUVs

Small unilamellar vesicles composed of cholesterol

3 hemisuccinate (CHS-SUVs) containing H-Inulin as the entrapped agent were prepared as follows: 100 μi of 1.0

3 mCi/mi H-Inulin (New England Nuclear, Boston MA) was dissolved in 2.5 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl to which either 100 or 200 mg tris-salt CHS was added. After mixing voritically with glass beads, the mixture was drawn off from the beads in the pipette, and sonicated to clarity, i .e. for approximately 2 hours. Clearing of the suspension indicates a transition of CHS-MLVs to small unilamellar vesicles. The final concentrations of CHS in the CHS-SUV suspensions was 40 mg/mi and 80 mg/mi, respectively.

In order to demonstrate inulin entrapment (see Section 5 supra), the CHS-SUVs were separated from unentrapped inulin by gel filtration as follows: each liposome suspension was applied separately to a Bio-Gel A- 15m, 100-200 mesh agarose column (Bio-Rad Laboratories, Richmond, CA) with an operating range of 40,000 to • 15,000,000 daltons molecular weight, equilibrated and calibrated with 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer. Then, 1 mi fractions which eluted from the column were collected and the radioactivity of a 10 aliquot of each fraction was determined as previously described. A clear separation of free from sequestered inulin was obtained by the gel filtration thus indicating the entrapment of the inulin in the CHS-SUVs. This analysis indicated that about 1% of the inulin was entrapped in the

CHS-SUVs.

6.4. ENTRAPMENT OF CHROMIUM IN

CHOLESTEROL HEMISUCCINATE MLVs

Cholesterol hemisuccinate multilamellar vesicles incorporating 51Chromium as the entrapped agent were prepared as follows: 15.0, 40.0, 65.8, 100.0, 175.0, 263.2,

526.4 or 658.0 ymoles of tris-salt CHS was added to 5 mi

0.01 M Tris-HCl, 0.14 M NaCl, pH 7.3, containing trace amounts of Chromium New England Nuclear, Boston, MA) and allowed to stand at room temperature for 2 hours, resulting in a suspension of CHS-MLVs contai .ni.ng 51Chromium entrapped

6.4.1. ENCAPSULATION EFFICIENCY OF CHROMIUM IN CHOLESTEROL HEMISSUCCINATE-MLVs

In order to determine the encapsulation efficiency of the CHS-MLVs prepared in Section 6.4 samples of each preparation were pipetted into dialysis bags (Thomas Scientific, Catalog No. 3787-D22, molecular weight cut off of 12,000 daltons) that had been boiled three times in distilled water. The samples in the dialysis bags were initially counted in a gamma counter (TmAnalytic, model no. 1191). The samples were then dialyzed for 20 hours against the same 0.01M Tris-HCl, 0.14 M NaCl pH 7.3, -buffer in retentate:dialysate ratios of greater than 1:150, respectively; the dialysate was changed every 2 hours for the first 6 hours. Encapsulation efficiency was determined by computing the percentage of initial counts retained. As indicated in Table II, an increase in encapsulation efficiency is proportional to an increase in CHS concentration.

TABLE II

ENCAPSULATION EFFICIENCY OF CHROMIUM IN CHOLESTEROL HEMISUCCINATE VESICLES

Concentration of ,.,

CHS (μmol) % Chromium Entrapped

15.0 14.79

40.0 15.20

65.8 15.09

100.0 16.10

165.0 20.13

263.2 27.90

526.4 40.74

658.0 48.03

6.4.2. CAPTURED VOLUME IN CHOLESTEROL

HEMISUCCINATE-MLVs: CHROMIUM ENTRAPMENT AND CHOLESTEROL HEMISUCCINATE CONCENTRATION

The captured volume of the CHS vesicles prepared a described in Section 6.4.1. was determined for each concentration of cholesterol hemisuccinate by computing the captured solute using the following calculation:

% entrapment x initial aqueous volume μmol CHS

The data illustrated in FIG. 1 indicate that less chromium/mole of lipid is entrapped as the concentration of tris-salt CHS is increased. Thus, although an increase in the encapsulation efficiency is proportional to the increas in lipid concentration, the captured solute decreases as th lipid concentration increases. The number of trials per point are indicated in parenthesis next to each point. 6.5. ULTRASTRUCTURE OF CHOLESTEROL HEMISUCCINATE LIPOSOMES

Samples of CHS-MLVs and CHS-SUVs, prepared using the tris-salt of cholesterol hemisuccinate as described in Section 6.1. except that the inulin was omitted, were prepared for freeze-etch electron microscopy (for freeze- etch method See Pfenninger et al. , 1975, J. Cell Biol. 65: 15-28) .

Electron microscopy of the CHS-MLV freeze-etched preparation revealed discrete units bounded by at lease one lamella, i.e., liposomes or lipid vesicles. There was a vast heterogeneity of size of the CHS-MLVs, ranging from 80 to 10,000 nm in diameter.

The larger vesicles could be categorized into a number of classes including: those with one or a few outer lamellae, and those with many lamellae. Most of the larger vesicles had substantial areas inside with a grainy appearance, possibly indicating the aqueous chambers. In many instances, small vesicles or groups of small vesicles were apparent, more or less free inside the larger vesicles sometimes four or five layers deep. This apparent "nesting is commonly observed in conventional liposomes made with a negatively charged phospholipid.

Occasionally, zones of closely apposed lamellae could be seen. These appear similar in all respects to the lamellae of conventional phospholipid MLVs.

The sonicated samples examined also contained many small spheroids ranging from 50 nm to 500 nm. These vesicles probably compare to SUVs made by sonication of phospholipid MLVs. As the smaller CHS-vesicles did not cleave, it was impossible to discern the structure of the interior or their component lamellae. The CHS vesicles which were extruded through the 30 nm filter (10 times) vesicles with an average size of about

65 nm were observed. This contrasts with CHS vesicles prepared by the French press procedure which were extremely _c small (average diameter of 25 nm or less).

6.6. X-RAY DIFFRACTION ANALYSIS OF CHOLESTEROL HEMISUCCINATE LIPOSOMES

X-ray diffraction of various CHS-MLV preparations 0 was performed using the 2-dimensional image-intensified X- ray detector apparatus described elsewhere (Gruner, S.M. 1977, PhD thesis, Princeton University, Princeton NJ 09540 USA; Reynolds, Geo. T., Milch, J.R. and Gruner, S.M, 1978, Rev. Sci. Instr. 49:1241-1249; Tilcock, C.P.T., Bally, M.B., 5 Farren, S.B., Cullis,_. P. R. and Gruner, S.M., 1984, Biochem, 23: 2696-2703). X-ray repeat spacings are expressed as +. o

0.5 A. CHS dispersions were held in 1.5 mm glass X-ray capillaries sealed with epoxy plugs. Specimens were hydrated either gently or vigorously. For gentle hydration, 0 the buffer was layered via a syringe onto dry CHS in the bottom of the X-ray capillary. The capillary was then momentarily centrifuged in a table top centrifuge to eliminate air bubbles from the lipid water paste. The capillaries were then sealed and allowed to equilibrate for 5 at least 4 hours at 5°C. Vigorous hydration was accomplished by vortexing dry CHS with buffer and two glass mixing beads in a test tube. An aliquot was then transfere to an X-ray capillary.

0 X-ray diffraction demonstrated that hydrated CHS forms multilamellar structures. FIG. 2a and 2b show the lo angle diffraction which resulted from gently hydrated specimens composed of 68.9% and 59.1%, CHS, respectively, b weight. Up to four equally spaced orders of diffraction ar o

35 visible, consistent with multilamellar arrays of 68.1 A and o

79.8 A, repeats, respectively. The orders were sharp and well resolved, indicating that the lattice contained very little disorder. These concentrated CHS specimens were of uniform paste-like appearance with no visible excess buffe consistent with the fact that the repeat specing increased as the aqueous content increased. At very high aqueous concentrations, gently hydrated CHS specimens exhibited a clearly visible pool of excess buffer solution on top of t hydrated lipid. The diffraction from such a sample (20.2% CHS by total weight) is shown in FIG. 2c The broadening o the higher angle diffraction peaks is indicative of considerable disorder in the lattice. The disorder in the lattice made a definitive lattice assignment difficult, but if a lamellar fit was made, as indicated in FIG. 2c, the o repeat was about 86 A, suggestive of large aqueous spaces between the lipid bilayers.

If, instead of using gentle hydration, a 20.7% CHS specimen was prepared by mixing the dry lipid vortically with the buffer, then the specimen which resulted had a uniform milky appearance. As shown in FIG. 2d, the low angle diffraction exhibited a broad band of scatter with little evidence of a sharply defined lattice. A similar diffraction signature would be expected from a multilamella system in which the interlamellar aqueous width varied widely. The x-ray diffraction of the dilute CHS dispersion is most consistently interpreted as arising from a multilamellar system in which the interlamellar forces are weak. For other lipid systems, such as dilute egg phosphatidylcholine dispersions, the x-ray diffraction pattern indicates a sharply defined lamellar lattice which is, by weight, mostly lipid (Rand, 1981 Annu. Rev. Biophys. Bioeng. 10: 277-314). This well-defined lattice repeat is result of a relatively sharp minimum in the lattice potential as a function of the lipid layer separation. If the potential vs. distance curve has only a shallow well, then one expects weak interlamellar forces and considerable lattice disorder. This appears to be the case with CHS vesicles. o

The specimen of FIG. 2d has about an 86 A repeat, o as compared to the 68.1 A repeat of FIG. 2a. This indicates that in the presence of excess buffer, CHS liposomes have a large water to lipid ratio. Similar results are observed with other charged lipid systems (Rand, 1981 supra).

6.7. ELECTRON SPIN RESONANCE ANALYSIS

OF CHOLESTEROL HEMISUCCINATE LIPOSOMES

Multilamellar liposomes made of egg phosphatidylcholine (EPC) (Avanti Polar Lipids, Birmingham, AL) were spin labelled and compared to similarly labeled tris-salt CHS-MLVs prepared essentially as described previously. In the case of EPC MLVs, 1 mole percent of either 5, 7, 9, 10, 12 or 16 doxylstearate (Molecular Probes, Junction City, OR) was added to 40 mg lipid in chloroform and the resulting solution dried to a thin film by rotary evaporation. Then 2 ml of Tris-HCl buffer was used to hydrate this film by vortexing until the film was completely suspended. The resulting EPC-MLVs were washed twice prior to spectroscopy.

In the case of CHS-MLVs, 1 mole % of the appropriate spin label in ethanol was dried to a thin film on the side of a test tube to which was added 40 mg of tris-salt CHS powder and 2 mi of Tris-HCl buffer. This suspension was vortexed and the resulting liposomes washed twice. All electron spin resonance experiments were carrie out with an IBM Instruments ER100D ESR spectrometer. The order parameter (S) was calculated as described elsewhere (Griffith and Jist in. Spin Labelling, Berliner, L.J. (ed.). Academic Press, N.Y. 1976). FIG. 3 shows the order parameter profiles of CHS- MLVs and EPC-MLVs as determined by spin labelling these preparations with either 5, 6, 7, 9, 10, 12 or 16 doxylsterate. For EPC bilayers, the order parameter decreases with increasing carbon number into the bilayer, a has been previously reported. The supra-molecular structur of CHS bilayers is markedly different: not only is the bilayer dramatically more rigid than the EPC bilayer, but CHS systems actually exhibit an increase in order from the 50th to the 90th carbon, indicative of an entirely differen physical and chemical bilayer structure than has previously been reported.

6.8. ISOTONIC SWELLING OF CHOLESTEROL HEMISUCCINATE LIPOSOMES

In the following series of experiments the isotoni swelling behavior of cholesterol hemisuccinate and phospholipid multilamellar vesicles was compared:

(1) CHS-MLVs were prepared as described in Section

6.1. using 40 mg tris-salt CHS in 2.0 mi 0.01 M Tris-HCl, 0.1 M KC1 buffer.

(2) EPC-MLVs were prepared as described in Section 6.2.1. using 51.8 mg EPC in 2.0 mi 0.01 M Tris-HCl, 0.1 M

KC1 buffer.

( 3 ) Multilame llar vesicles with a lipid bilayer composed of EPC and egg phosphat idic acid ( EPA) were prepared using the method described in Section 6 . 2.1 . for

EPC-MLV preparat ion using 41. 1 mg EPC and 9.79 mg EPA in 2. mi 0.01 M Tris-HCl , 0. 1 M KC1 buf fer . The resulting MLVs ( EPC : EPA-MLVs ) comprised EPC : EPA in a 8 : 2 molar ratio respective ly. After vortical mixing, each suspension of MLVs

(i.e., the CHS-MLVs, the EPC-MLVs and the EPC:EPA-MLVs) was allowed to stand at room temperature for two hours in the preparatory buffer. A 20 μi aliquot of each liposome preparation was then added to 1.0 mi of a series of 0.01 M

Tris-HCl buffers with KCl contrations ranging from 0.055M to

0.5 M. After equilibration for one-half hour, light scattering was determined by measuring absorbance of the samples at a wavelength of 550 nm.

Results are graphically illustrated in FIG. 4 in which absorbance is plotted against the inverse of the KCl concentrations of the media to which the vesicles are exposed. Increased absorbance indicates swelling of lipid vesicles. Curves B and D demonstrate that, as expected the phospholipid MLVs (i.e. , EPC-MLVs, and EPC:EPA-MLVs) behave as ideal osmometers. Curve C, however, indicates that although the CHS-MLVs behave as closed vesicular structures, they exhibit non-ideal behavior in hypo- and hypertonic media. This behavior, indicated in FIG. 4, is quite different than that observed in the cholesterol liposomes o Brockerhoff and Ramsammy (1982, Biochim. Biophys. Acta. 691: 227-232).

7. EXAMPLE: CHOLESTEROL HEMISUCCINATE LIPOSOMES ENTRAPPING SPARINGLY SOLUBLE COMPOUNDS

The entrapment in CHS-liposomes of compounds that are sparingly soluble in water is demonstrated for bovine growth hormone, insulin and tylosin.

7.1. BOVINE GROWTH HORMONE ENTRAPPED IN CHOLESTEROL HEMISUCCINATE-SUVs

Bovine growth hormone (BGH), a simple protein composed of a single chain of approximately 191 amino acids, is partially water soluble. The normal solubility is 1 to 1.5 mg/mi at pH 8.0. BGH precipitates in organic solvents such as chloroform.

CHS-MLVs were prepared as described in Section 6.2 using 27 mg tris-salt CHS in 1.0 mi 0.01 M Tris-HCl (pH 7.4), 0.14 M NaCl buffer. The CHS-MLV preparation was sonicated to clarity to form sonicated CHS-SUVs and then either 5, 10, 15, 25, 30 or 166 mg BGH (Eli Lilly & Co., Indianapolis, IN) was added to separate aliquots of the sonicated CHS-SUV suspension. The suspensions were extensively mixed vortically, resulting in partitioning of the protein into the CHS-SUV bilayers. The sonicated CHS- SUV suspensions were visually observed for the presence of precipitate at 1, 2, and 21 days. No precipitate was observed, indicating that the bovine growth hormone remaine entrapped in the CHS liposomes at all concentrations tested for 21 days at room temperature.

7.2. INSULIN ENTRAPPED IN CHOLESTEROL HEMISUCCINATE-SUVs

Zinc-insulin, a polypeptide hormone, although readily soluble in dilute acid or alkali, is practically insoluble in aqueous phases from pH 4.5 to 7.0. In fact th tendency of insulin solutions to form macroaggregates is an obstacle in the development of long-term insulin delivery systems.

CHS-MLVs were prepared as described in Section 6.2 using 27 mg tris-salt CHS in 1.0 mi 0.01 M Tris-HCl (pH 7.4), 0.14 M NaCl buffer. The CHS-MLV preparation was sonicated to carity to form sonicated CHS-SUVs and up to 47 mg zinc-insulin powder (Bovine Pancreatic Insulin, Sigma Chemical Co., St. Louis, MO) was added to the CHS-SUV suspension. The suspension was extensively mixed vortically, resulting in partitioning of the insulin into the CHS-SUV bilayers. The sonicated CHS-SUVs were visually observed for th e presence of precipitate at 1, 2, and 21 days. No precipitate was observed indicating that insulin at a concentration of 5 mg/mi remains entrapped for at least 21 days at room temperature. Insulin entrapment occurs more rapidly at 37 °C.

7.3 TYLOSIN ENTRAPPED IN CHOLESTEROL HEMISUCCINATE-SUVs

Tylosin is an antibiotic that is soluble in water at 25 °C at 5 mg/mi, and is also soluble in lower alcohols, esters and ketones, chlorinated hydrocarbons, benzene, and ether.

Small unilamellar vesicles were prepared as follows: 100 mg tylosin-base (Eli Lilly & Co.,

Indianapolis, IN) and 200 mg tris-salt CHS were mixed vortically in 4 mi phosphate buffered saline (pH 7.4). The resulting milky suspension of CHS-MLVs was sonicated with a probe tip sonicator for 15 minutes. (As a precautionary measure, an ice bath was placed around the test tube to kee the temperature of the mixture down.) The mixture was then placed into a bath sonicator for 1 hour and 45 minutes.

After the 2 hour sonication period, the CHS-SUV entrapped tylosin was separated from the suspension by centrifugation of the suspension at 10,000 x g for 10 minutes, forming a small pellet and an opalescent supernatant. The sonicated CHS-SUVs in the supernatant wer visually compared to a suspension of 100 mg tylosin base (Eli Lilly & Co., Indianapolis, IN) added to the same volum of water. A precipitate formed in the suspension of tylosi base, however no precipitate formed in the CHS-SUV tylosin preparation. Thus the tylosin appeared to remain entrapped for at least 48 hours. 8. EXAMPLE: THE USE OF CHOLESTEROL HEMISUCCINATE LIPOSOMES TO ENTRAP LIPID SOLUBLE COMPOUNDS

The entrapment of lipid soluble bioactive agents i demonstrated for indomethacin and diazepa .

8.1. INDOMETHACIN ENTRAPPED IN CHOLESTEROL HEMISUCCINATE-MLVs

Indomethacin, a prostaglandin inhibitor, is practically insoluble in water. The free acid of indomethacin is soluble in ehtanol, ether, acetone, and castor oil.

CHS-MLVs incorporating varying amounts of indomethacin as bioactive agent were prepared as follows: i a round-bottom flask, 27 mg tris-salt CHS, 1-5 mg indomethacin, and 10 μ i C-indomethacin (22.0 mCi/mmol , Ne England Nuclear, Boston, MA) were combined. Sufficient methanol to dissolve all components was added. The mixture was then rotoevaporated to form a thin film on the vessel, and vacuum desiccated overnight to insure removal of all th methanol. Then CHS-MLVs were formed by adding 1.0 mi 0.01 Tris-HCl (pH 7.3), 0.14 M NaCl buffer to each flask. The suspensions were mixed vortically with glass beads, and allowed to stand undisturbed for 2 hours.

After 2 hours, the relative amount of indomethacin entrapped in the CHS-MLVs was determined as follows: 9.0 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer was added to each sample and the mixture was centrifuged for 10-20 minutes at 10,000 x g. The resulting pellet was washed three times in 10 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer, and was suspended in a final volume of 1.0 mi 0.01 Tris-HCl (pH 7.3), 0.14 M NaCl buffer. A "standard" was prepared in the same manner as the samples except that only radiolabeled indomethacin was added to the initial mixture ( i.e. , the 1 to 5 mg indomethacin was omitted from the CHS- MLV standard preparation). The radioactivity contained in filtered 20 y aliquots from each sample were counted in 10 mi scintillation fluid. Comparison of the "standard" with the samples containing various concentrations of indomethacin permitted dertermi ation of the percentage of indomethacin entrapped. Results are presented in Table III

T B L E III

ENTRAPMENT OF INDOMETHACIN IN CHS-MLVs

Concentration , .

Indomethacin % C-Indomethacin

(mg/mi) Entrapped in CHS-MLVs i 78

2 70

3 ^' 37

4 28

5 34

Results indicate that up to 78% of the indomethacin can be entrapped in CHS-MLVs .

8.1.2. ULTRASTRUCTURE OF CHOLESTEROL HEMISUCCINATE VESICLES CONTAINING INDOMETHACIN

In order to determine whether the entrapped indomethacin altered the membrane vesicles, the CHS-MLVs prepared in the presence of indomethacin (see Section 8.1.1.) were processed as previously described for freeze- etch electron microscopy. In freeze-etch electron microscopy, under low magnification, the "empty" CHS-MLVs wer indistinguishable from those that have indomethacin included. That is, there was no obvious feature that one can discern to be unique to one or the other. At high magnification, however, the "bilayer" of the CHS-MLVs containing indomethacin is distinctive. Because indomethacin is a water-insoluble drug and is soluble in ehtanol, ether, acetone, and other non-polar solvents, it can be expected that the indomethacin, in the presence of lipids, would be arranged such that it was sequestered from the water. Examinations- of the bilayers seen by electron microscopy indicated that the thickness of the bilayers varies in the cross fracture. This suggested that the indomethacin was indeed distributed in the lipid portion of the bilayers, such that it appeared to give an added thickness and a very non-uniform configuration; that is, th thickness varied as one bilayer was traced along a fracture line. This effect is presumably special for those drugs whose solubilities are such that they sequester in the lipi portion of the bilayers.

8.2. DIAZEPAM ENTRAPPED IN CHOLESTEROL HEMI SUCCINATE-SUVs

Diazepam, a sedative or tranquilizer ( i.e. , Valiu ) is soluble in chloroform, dimethylf ormamide, benzene, acetone and alcohol; it is only slightly soluble in water.

CHS-SUVs incorporating diazepam were prepared as follows: 2, 3, 4, or 5 mg diazepam was added to a test tube containing 5 μi H-diazepam (76.7 Ci/mmol, New England Nuclear, Boston, MA). Sufficient methanol was added to eac tube to dissolve the drug (maximum 2 mi methanol). The mixture was then rotoevaporated to a thin film, and desiccated overnight under vacuum to insure removal of all the methanol. The dried film was resuspended in 1 mi of a suspension of sonicated CHS-SUVs (prepared as described below using 50, 100, or 200 mg/mi CHS), mixed vortically using glass beads and filtered using 0.22 μm Millipore filters (Millipore Corp., New York, NY).

The CHS-SUVs were prepared according to the following protocol: 50, 100, or 200 mg tris-salt CHS was vortically mixed with 1.0 mi 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer. Using a probe sonicator, the mixture was sonicated to an optical density of about 0.40 measured at a wavelength of 550 nm (i.e, a "clear" solution), and then centrifuged at 1,000 x g to remove any titanium that might have come off the tip of the sonicator probe. The suspension of CHS-SUVs was then decanted from the tube.

The relative amount of diazepam that could be entrapped by the sonicated CHS-SUVs was determined by 0 comparison of the samples with a "standard" preparation. The "standard" preparation was prepared in the same manner as the samples except that only radiolabeled diazepam was added (i.e, the 2 to 5 mg diazepam was omitted from the CHS-SUV standard preparation. In either case, the

15 radioactivity contained in 10 μi aliquots of the filtered suspensions were counted in 10 mi scintillation fluid. Results are presented in Table IV.

20

T A B L E IV

ENTRAPMENT OF DIAZEPAM IN SONICATED CHS-VESICLES Concentration % 3H-Diazepam Entrapped

" Diazepam Concentration CHS (mg/mi

(mg/mi) 50_ _100_ 20

2 86 100 100

3 79 95 89

4 58 100 100

5 53 100 100 30

Results indicate that both 100 and 200 mg/mi CHS entrap 100% of the highest concentration of diazepam used (5 mg/mi) . Since both 100 and 200 mg/mi CHS sufficiently ³⁵ entrap the diazepam, the 100 mg/mi represents a more ideal concentration of CHS for entrapment of diazepam. 9. EXAMPLE: THE USE OF CHOLESTEROL HEMISUCCINATE LIPOSOMES TO DETERMINE AMINOGLYCOSIDE CONCENTRATION IN SERUM

It was observed that a relatively low concentratio of CHS-vesicles (less than 1 μg/mi CHS) strongly agglutinated red blood cells (RBC) from a suspension in phosphate buffered saline (PBS). Since CHS-vesicles are precipitated by Ca and other cations such as aminoglycoside antibiotics, the CHS-vesicles can be used to determine the concentration of aminoglycoside antibiotics i sera by:

(a) determining the dilution of serum containing antibiotic at which a fixed amount of CHS- vesicles are precipitated; and

(b) Determining by hemagglutination-titration the concentration of the remaining free vesicles which are not precipitated after addition to the antibiotic-containing serum.

In both situations, the exact amount of antibiotic coul be established using a comparison with a standard curve derived from known concentrations of antibiotic. The basic experiments are described below:

Accordingly, 24 μi of gentamycin sulfate in PBS (1 mg/mi) was serially diluted in serum (i.e. , 25 ui aliquots of serum). Then a 24 μi aliquot of CHS-unilamellar vesicle prepared by sonicating tris-salt CHS at a concentration of 25 mg/mi in 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer at pH 7.4, was added to each sample.

After about 10 minutes at room temperature, the turbidity of each mixture was recorded. Only mixtures containing 50 μg or more gentamycin showed a visible precipitation indicating that the CHS vesicles interacted with the gentamycin.

Next, the precipitated material was pelleted and 5 the supernatant was used to determine the concentration of the CHS-SUVs by hemagglutination of chicken red blood cells. The hemagglutination assay was done in 96 U-shaped microwel plates by serially diluting the vesicle suspensions to 50 y PBS and then adding in each well 40 l of 0.5% RBC in PBS. •JO After 60 minutes at 4°C, hemagglutination was observed with an inverted mirror. The control (RBC in the absence of CHS vesicles) showed no hemagglutination. All samples containing CHS strongly hemagglutinated the RBC with the exception of those which, in the previous experiment, were

15 turbid. This result indicated that CHS vesicles interacted with gentamycin such that relatively fewer vesicles were available for hemagglutination in comparison with the control suspension containing only CHS.

20 10. EXAMPLE: IN VIVO ADMINISTRATION OF CHOLESTEROL HEMISUCCINATE LIPOSOMES

The following subsections describe methods and compositions for the n vivo administration of bioactive agents using the cholesterol hemisuccinate liposomes of the 25 present invention. The clinical effectiveness of the entrapped bioactive agent is determined, and drug distribution within selected organs is traced where appropriate.

30 10.1. TREATMENT OF JOINT ARTHRITIS USING INDOMETHACIN

ENTRAPPED IN CHOLESTEROL HEMISUCCINATE-MLVs

Male white New Zealand rabbits (2 to 2.5 kg) were immunized intradermally, twice at two-week intervals, with mi of 20 mg/mi bovine serum albumin (BSA) (Miles

35

Laboratories, Elkhart, IN) emulsified in complete Freunds adjuvant. On the third week, the rabbits received a single intra-articular injection of 10 mg BSA in 1.0 mi of saline into the right knee joint to initiate joint arthritis. Lef knee joints served as controls. The diameter of the joints was measured using a Fowler dial caliper, sensitive to 0.01 mm. The BSA-injected joints swell and typically measure 3 to 4 mm larger than control joints. On the fourth week, th rabbits received another intra-articular injection of BSA i saline to initiate joint inflammation.

CHS-MLVs were prepared as described in Section 8.1 using 270.0 mg CHS and 10 mg indomethacin, resulting in final concentration of 1.0 mg/mi indomethacin. Three days following induction of the inflammation, BSA-injected animals received a single intramuscular injection of the 1 mg/mi indomethacin entrapped in CHS-vesicles (total dose 1 mg/animal). Joint swelling was measured for another ten days.

Results illustrated graphically in FIG. 5, indicat that the indomethacin entrapped in CHS-MLVs was effective i reducing joint swelling when administered intramuscularly.

10.2. IN VIVO ADMINISTRATION OF DIAZEPAM ENTRAPPED IN CHOLESTEROL HEMISUCCINATE SUVs

Mice were ^'inoculated intravenously or intramuscularly with 500 μg/kg body weight of diazepam entrapped in CHS-SUVs prepared as described in Section 8.2 The diazepam had a sedative effect on the mice (the mice fell asleep after inoculation) indicating the retention of activity of the entrapped drug. 10.2.1. ORGAN DISTRIBUTION AFTER INTRAVENOUS INOCULATION

CHS-SUVs containing entrapped diazepam were prepared as described in Section 8.2. After sonication in bath sonicator, the absorbance of the suspension measured a a wavelength of 550 nm was 0.370. Then a 44.2 μi aliquot o the CHS-SUV suspension was added into a glass test tube, onto which 40 μCi of C-diazepam (specific activity is equal to 181 μCi/mg supplied as 100 μCi/mi in ethanol) had been dried down under nitrogen. After mixing vortically f five minutes, 1.282 mi of 0.01 M Tris-HCl (pH 7.3), 0.14 M NaCl buffer was added to the solution; resulting in a thirty-fold dilution of the suspension, yielding a therapeutic dose for the mice of 0.167 mg/μi diazepam.

A 0.1 mi aliquot of the CHS-SUV-¹ C-diazepam suspension was injected into the tail vein of conscious, restrained 35 mg Swiss-Webster mice. Control mice were similarly inoculated with an equivalent does of unentrapped C-diazepam. At 1, 2, or 5 hours post-injection, the mice were sacrificed by cervical dislocation and the internal organs (kidney, lung, spleen, liver, gut, brain, heart, pancreas, and fat) and a blood sample were removed. The organs were weighed and a small sample (20-40 μg) of each was digested and decolorized according to the method of Mahim and Kohberg (1966, Analytical Biochem. .16_: 500). Th samples were then dark-adapted for five days to allow chemi luminescence to subside before the radioactivity was measured.

The results of this experiment are shown in FIG.6. In the mice inoculated with the CHS-SUV entrapped diazepam, the drug does not accumulate in the spleen, indicating that the diazepam entrapped in CHS-SUVs does not behave like phospholipid liposome-entrapped drug when administered intravenously in vivo. 10.3 IN VIVO ADMINISTRATION OF CHROMIUM

ENTRAPPED IN CHOLESTEROL HEMISUCCINATE-MLV

In order to determine whether CHS vesicles remain intact when administered .in. vivo, the organ distribution o a free aqueous marker as compared to that of the aqueous marker entrapped in CHS vesicles was determined after intravenous injection in mice. To this end, the organ distribution of unentrapped Chromium ( Cr), Chromium entrapped in CHS-MLVs and 51Chromium entrapped in phospholipid vesicles were compared. The following protocol were followed:

(a) Unencapsulated 51Cr. 51Cr is supplied as

Cr0 = in sterile 0.9% saline (New England Nuclear, Newton MA). The free 51Cr injectate was made by diluting 100 μi

⁵¹Cr in 0.9% saline to 1.5 mi total volume with 0.01 M

Tris-HCl, pH 7.3, 0.14 M NaCl, 5% dextrose. Sixteen 40 g male Swiss Webster mice each received a 0.-1 mi (about

700 ,000 cpm ) intrave nous inj ection via the tail ve in .

(b) 51Cr n CHS-MLVs were prepared by dissolving dry tris-salt CHS powder in 0.01 M Tris-HCl, pH 7.3, 0.14 M NaCl, 5% dextrose containing a trace amount of 51Cr (

in sterile 0.9% saline). The mixture was vortically mixed and briefly then sonicated for 30 minutes. After pelleting by centrifugation and washing three times as previously described, the final pellet was resuspended to a final concentration of 10 mg/mi CHS, and sized at a 1 micron diameter by Nicomp quasielastic light scattering analysis. Twelve 40 g male Swiss Webster mice each received a 0.1 mi

(about 120,000 cpm) intravenous injection via the tail vein

(c) 51Cr in EPC-SPLVs were prepared by the genera procedure described in detail in copending application Serial No. 476,496, filed March 24, 1983 by Lenk et al. entitled "Stable Plurilamellar Vesicles, Their Preparation and Use", which is incorporated by reference herein. To this end, 5 mi batches were prepared by dissolving 65 mg eg phosphatidylcholice (EPC) in chloroform and drying down the EPC to form a film. The film was resuspended in 10 mi ethe and 0.3 mi ⁵¹Crθ2 n 0.9% saline (pH 8) was added. The mixture was then emulsified by sonication while concurrentl evaporating the ether under N₂ gas. The resulting stable plurilamellar vesicles (SPLVs) were resuspended in 5 mi 0.0 M Tris-HCl, pH 7.3, 0.14 M NaCl, 5% dextrose and pelleted b centrifugation. The pellet was washed three times to remov unentrapped 51Chromium and the final pellet was resuspended in 5 mi 0.01 M Tris-HCl, pH 7.3, 0.14 M NaCl, 5% dextrose.

The final concentration of EPC was 13 mg/mi. Twelve 40 g male Swiss Webster mice each received a 0.1 mi (about

100,000 cpm) intravenous injection via the tail vein.

At the end of 1, 2, 5, and 24 hours 3 mice from each group treated with a liposome preparation and 4 mice from the group treated with unencapsulated 51Chromium were sacrificed by cervical dislocation. The organs were removed, rinsed with 0.9% saline, weighed and counted as previously described to determine the % dose and % dose/gra remaining in each organ tested.

The results, shown in FIG. 7 demonstrate that unencapsulated chromium (see FIG. 7C) is excreted rapidly, and does not concentrate in any of the organs tested. EPC and CHS encapsulated chromium remains at measurable levels 24 hours after injection, indicating that CHS vesicles, lik EPC-SPLVs, remain intact in vivo. Moreover, EPC-SPLVs (13 mg/mi) which have a mean diameter betrween 0.5 and 1.0 microns, accumulate in the liver, lung, and spleen (see FIG 7B); i.e., the typical pattern of liposome distrubution. Equimolar CHS vesicles accumulate primarily in the liver, and to a much lesser extent, in the lung and spleen (see FIG. 7A) . Indicating a difference in the distribution pattern of the two liposome preparatins in vivo. 10.4. IN VIVO ADMINISTRATION OF HUMAN GROWTH HORMONE ENTRAPPED IN-CHOLESTEROL HEMISUCCINATE MLVs

CHS-multilamellar vesicles incorporating 125I-hum growth hormone (HGH, New England Nuclear, Boston, MA) were prepared as follows: tris-salt CHS was added to 0.01 M

Tris-HCl, 0.14 M NaCl buffer (pH 7.4) containing 1 μCi/ i

125 I-HGH (New England Nuclear, Boston, MA) to yield a fina concentration 25 mg/mi tris-salt CHS. The suspension was mixed vortically with glass beads. The resulting CHS-MLVs eennttrraappppeedd 1100%% ooff tthhee ¹ 1²2⁵5II--HHGGHH aass dde. termined by a comparison of the initial radioactive counts .

A group of 12 female Swiss-Webster mice were injected intramuscularly in the hind limb with 0.5 mi of th CHS-MLV entrapped 125 I-HGH suspension. A control group of 12 mice were inj ected with 0.5 mi free 125 I-HGH in .01 M

Tris-HCl, 0.14 M NaCl buffer. Each animal in both groups received approximately 35,000 cpm/animal. At periodic intervals post-injection, the mice were sacrificed, the hin limb dissected, and percent total radioactivity remaining was determined. The data in Table V demonstrate substantia increase in retention of 125I-HGH when incorporated into

CHS-MLVs. Thus, when injected intramuscularly, the CHS- liposome entrapped drug is released in a sustained fashion.

T A B L E V

RETENTION,QF ADMINISTERED MLV-ENTRAPPED 'I-HUMAN GROWTH HORMONE

% Radioactivity Remaining in Limb

Inoculum Time (Hours)

(N = 12 animals) 24. ___=. 168

Control³ 5 0.7 0.2 0.3 ^iZDI-HGH

125 -I-HGH in 56 37 29 22

CHS-vesicles

Animals received 0.5 mi 1 "25^''I.-Human Growth Hormone in 0.01 M Tris-HCl, 0.14 M NaCl. Animals received 0.5 mi 125 I-Human Growth Hormone in

CHS-vesicles (25 mg/mi CHS) in 0.01 M Tris-HCl, 0.14 M

NaCl.

Claims

WHAT IS CLAIMED IS:

1. A method for administering a compound jLn. vivo comprising, administering to a host the compound entrapped in a liposome, the bilayers of which comprise a salt form o an organic acid derivative of a sterol.

2. The method according to claim 1 in which the liposomes comprise multilamellar vesicles.

3. The method according to claim 1 in which the liposomes comprise unilamellar vesicles.

4. The method according to claim 1 in which the bilayers comprise a tris(hydroxymethyl) aminomethane salt form of an organic acid derivative of a sterol.

5. The method according to claim 4 in which the liposomes comprise multilamellar vesicles.

6. The method according to claim 4 in which the liposomes comprise unilamellar vesicles.

7. The method according to claim 1 in which the bilayers comprise a salt form of a carboxylic acid derivative of a sterol, a salt form of a dicarboxylic acid derivative of a sterol, or a salt form of a polycarboxylic acid derivative of a sterol.

8. The method according to claim 7 in which the liposomes comprise multilamellar vesicles.

9. The method according to claim 7 in which the liposomes comprise unilamellar vesicles.

10. The method according to claim 7 in which the carboxylic acid comprises an aliphatic carboxylic acid. -so¬ il. The method according to claim 10 in which the aliphatic carboxylic acid contains up to five carbon atoms.

12. The method according to claim 10 in which the aliphatic carboxylic acid contains more than five carbon atoms.

13. The method according to claim 7 in which the carboxylic acid comprises an aromatic carboxylic acid.

14. the method according to claim 7 in which the dicarboxylic acid comprises an aliphatic dicarboxylic acid.

15. The method according to claim 14 in which the aliphatic dicarboxylic acid contains up to seven carbon atoms.

16. The method according to claim 15 in which the aliphatic dicarboxylic acid is succinate.

17. The method according to claim 14 in which the aliphatic dicarboxylic acid contains more than seven carbon atoms.

18. The method according to claim 7 in which the dicarboxylic acid comprises an aromatic dicarboxylic acid.

19. The method according to claim 1 in which the bilayers comprise a salt form of an hydroxy acid derivative of a sterol.

• 20. The method according to claim 19 in which the liposomes comprise multilamellar vesicles.

21. The method according to claim 19 in which the liposomes comprise unilamellar vesicles. 22. The method according to claim 19 in which the hydroxy acid comprises an α hydroxy acid.

23. The method according to claim 1 in which the b rilayers comprise a salt form of an amino acid derivative o a sterol or a salt form of a polyamino acid derivative of a sterol.

24. The method according to claim 23 in which the liposomes comprise multilamellar vesicles.

25. The method according to claim 23 in which the liposomes comprise unilamellar vesicles.

26. The method according to claim 1 in which the bilayers comprise a salt form of an organic acid derivative of a cholesterol.

27. The method according to claim 26 in which the liposomes comprise multilamellar vesicles.

28. The method according to claim 26 in which the liposomes comprise unilamellar vesicles.

29. The method according to claim 1 in which the bilayers comprise a salt form of an organic acid derivative of a vitamin D.

30. The method according to claim 29 in which the liposomes comprise multilamellar vesicles.

31. The method according to claim 29 in which the liposomes comprise unilamellar vesicles.

32. The method according to claim 1 in which the bilayers comprise a salt form of an organic acid derivative of a phytosterol. 33. The method according to claim 32 in which the liposomes comprise multilamellar vesicles.

34. The method according to claim 32 in which the liposomes comprise unilamellar vesicles.

35. The method according to claim 1 in which the bilayers comprise a salt form of an organic acid derivative of a steroid hormone.

36. The method according to claim 35 in which the liposomes comprise multilamellar vesicles.

37. The method according to claim 35 in which the liposomes comprise unilamellar vesicles.

38. The method according to claim 1 in which the bilayers comprise the tris(hydroxymethyl)aminomethane salt form of a hemisuccinate derivative of cholesterol.

39. The method according to claim 38 in which the liposomes comprise multilamellar vesicles.

40. The method according to claim 38 in which the liposomes comprise unilamellar vesicles.

41. The method according to claim 1 or 38 in which the compound comprises a bioactive agent.

42. The method according to claim 41 in which the bioactive agent is selected from the group consisting of: antibacterial, antifungal, antiviral, and antiparasitic compounds.

43. The method according to claim 42 in which the antibacterial compound is tylosin. 44. The method according to claim 42 in which the antifungal compound is miconazole.

45. The method according to claim 41 in which the bioactive agent comprises a cell receptor binding compound.

46. The method according to claim 41 i i which the bioactive agent is selected from the group consisting of: hormones, neurotransmitters, tumoricidal compounds, growth factors, and toxins.

47. The method according to claim 46 in which the bioactive agent comprises growth hormone.

48. The method according to claim 47 in which the growth hormone comprises bovine growth hormone.

49. The method according to claim 47 in which the growth hormone comprises human growth hormone.

50. The method according to claim 46 in which the bioactive agent comprises insulin.

51. The method according to claim 41 in which the bioactive agent is selected from the group consisting of: peptides, proteins, glycoproteins, and lipoproteins.

52. The method according to claim 51 in which the bioactive agent is an immunoglobulin.

53. The method according to claim 41 in which the bioactive agent comprises an immunomodulating compound.

54. The method according to claim 41 in which the compound is selected from the group consisting of catalysts and enzymes. 55. The method according to claim 1 or 38 in whic the bioactive agent is selected from the group consisting of: dyes, radiolabels, radio-opaque compounds, and fluorescent compounds.

56. The method according to claim 55 in which the dye is arsenazo III.

57. The method according to claim 41 in which the bioactive agent is selected from the group consisting of: anti-inflammatory, antiglaucomic, mydriatic, analgesic and anesthetic compounds.

58. The method according to claim 57 in which the bioactive agent comprises indomethacin.

59. The method according to claim 41 in which the bioactive agent comprises a narcotic.

60. The method according to claim 41 in which the bioactive agent comprises a psychoactive or anxiolytic agent.

61. The method according to claim 60 in which the bioactive agent comprises diazepam.

62. The method according to claim 41 in which the bioactive agent comprises a vitamin.

63. The method according to claim 62 in which the vitamin comprises α-tocopherol.

64. The method according to claim 41 in which the bioactive agent is selected from the group consisting of: monosaccharides, disaccharides, and polysaccharides. 65. The method according to claim 41 in which the route of administration is intravenous.

66. The method according to claim 41 in which the route of administration is intraperitoneal.

67. The method according to claim 41 for sustaine release in which the route of administration is intramuscular.

68. A method- for the preparation of sterol liposomes, comprising adding to an aqueous phase a salt for of an organic acid derivative of a sterol capable of formin closed bilayers in an amount sufficient to form completely closed vesicles, and. shaking the mixture unitl a milky suspension of multilamellar vesicles is formed.

69. The method according to claim 68 further comprising sonicating the milky suspension to form a suspension of unilamellar vesicles.

70. The method according to claim 68 or 69 in which the bilayers comprise a tris(hydroxymethyl)aminomethane salt form of an organic aci derivative of a sterol.

71. The method according to claim 68 or 69 in which the bilayers comprise a 2-amino-2-methyl-l,3- propanediol salt form of an organic acid derivative of a sterol.

72. The method according to claim 68 or 69 in which the bilayers comprise a 2-aminoethanol salt form of a organic acid derivative of a sterol. 73. The method according to claim 68 or 69 in which the bilayers comprise a bis-tris-propane salt form of an organic acid derivative of a sterol.

74. The method according to claim 68 or 69 in which the bilayers comprise a triethanolamine salt form of an organic acid derivative of a sterol.

75. The method according to claim 68 or 69 in which the bilayers comprise a miconazole salt form or an organic acid derivative of a sterol.

76. The method according to claim 68 or 69 in which the bilayers comprise a salt form of a carboxylic acid derivative of a sterol, a salt form of a dicarboxylic acid derivative of a sterol, or a salt form of a polycarboxylic acid derivative of a sterol.

77. The method according to claim 76 in which the dicarboxylic acid is succinate.

78. The method according to claim 68 or 69 in which the bilayers comprise a salt form of an organic acid derivative of a cholesterol.

79. The method according to claim 68 or 69 in which the bilayers comprise the tris(hydroxmethyl)aminomethane salt form of a hemisuccinate derivative of cholesterol.

80. The method according to claim 79 in which the aqueous phase is essentially free of divalent cations.

81. The method according to claim 79 in which the cholesterol hemisuccinate is present in a range of about 4.5 mg to about 200 mg for each mi of aqueous phase. 82. The method according to claim 79 further comprising the addition of a tris-salt to the aqueous phas

83. The method according to claim 68 or 69 furthe comprising the addition of a compound to be entrapped in th sterol vesicles to the aqueous phase.

84. The method according to claim 83 in which the compound is water-soluble.

86. A method for the entrapment of a water- insoluble compound in sterol vesicles comprising:

(a) dissolving the water-insoluble compound in an organic solvent;

(b) removing the organic solvent leaving a film comprising the water-insoluble compound; and

(c) adding the suspension of sterol vesicles of claim 68 or 69 to the film.

87. A method for the entrapment of a water- insoluble compound in sterol vesicles comprising:

(a) co-solubilizing in an organic solvent, the water-insoluble compound and a salt form of a organic acid derivative of sterol capable of forming bilayers in an aqueous phase;

(b) evaporating the organic solvent leaving a fil comprising the compound and the salt form of an organic derivative of a sterol;

(c) adding an aqueous phase to the film and shaking to form a milky suspension of completely closed multilamellar vesicles.

88. The method according to claim 87 further comprising sonicating the milky suspension to form unilamellar vesicles. 89. Steroidal liposomes comprising completely closed bilayers comprising a salt form of an organic acid derivative of a sterol.

c 90. Steroidal liposomes according to claim 89 in which the liposomes comprise multilamellar vesicles.

91. Steroidal liposomes according to claim 89 in which the liposomes comprise unilamellar vesicles. 0

92. Steroidal liposomes according to claim 89 in which the bilayers comprise a tris(hydroxymethyl)aminomethane salt form of an organic aci derivative of a sterol. 5

93. Steroidal liposomes according to claim 89 in whcih the bilayers comprise a l-amino-l-methyl-1,3- propanediol salt form of an organic acid derivative of a sterol. 0

94. Steroidal liposomes according to claim 89 in which the bilayers comprise a 2-aminoethanol salt form of a organic acid derivative of a sterol.

5 95. Steroidal liposomes according to claim 89 in which the bilayers comprise a bis-tris-propane salt form of an organic acid derivative of a sterol.

96. Steroidal liposomes according to claim 89 in which the bilayers comprise a triethanolamine salt form of 0 an organic acid derivative of a sterol.

97. Steroidal liposomes according to claim 89 in which the salt is derived from an ionizable bioactive agent 5

98. Steroidal liposomes according to claim 97 in which the ionizable bioactive agent is micronazole. 99. Steroidal liposomes according to claim 89 in which the bilayers comprise a salt form of a carboxylic aci derivative of a sterol, a salt form of a dicarboxylic acid of a sterol, or a salt form of a polycarboxylic acid derivative of a sterol.

100. Steroidal liposomes according to claim 89 in which the bilayers comprise a salt form of an organic acid derivative of cholesterol.

101. Steroidal liposomes according to claim 89 in which the bilayers comprise a tris(hydroxymethyl)- aminomethane salt form of a hemisuccinate derivative of cholesterol.

102. A method for the detection of a cation in serum comprising:

« (a) mixing the steroidal liposomes of claim 101 with serum; and

(b) detecting any precipitate which may form.

103. A method according to claim 102 further comprising:

(c) mixing the steroidal liposomes of claim 101 with a known concentration of the cation;

(d) detecting any precipitate which may form; and

(e) comparing the formation of precipitate in ste (d) with the formation of any precipitate in claim 102 step (b)

wherein steps (c) through (e) may be performed before, during or after steps (a) and (b) of claim 102.

104. The method according to claim 102 or 103 in which the cation is an aminoglycoside antibiotic. 105. A method for the detection of a cation in serum comprising:

(a) mixing the steroidal liposomes of claim 101 with serum; ?

(b) removing any precipitate which may form;

(c) mixing an aliquot of the precipitate-free serum with a known concentration of red blood cells;

(d) detecting whether hemagglutination of the red blood cells has occurred.

106. The method according to claim 105 further comprising:

(e) mixing the steroidal liposomes of claim 101 with serum containing a known concentration of the cation;

(f) removing any precipitate which may form;

(g) mixing an aliquot of the precipitate-f ee serum with a known concentration of red blood cells;

(h) ^' detecting whether hemagglutination of the red blood cells has occurred; and

(i) comparing the hemagglutination reaction of step (h) with the hemagglutination reaction of claim 105 step (d)

wherein steps (e) through (i) may be performed before during or after steps (a) through (d) of claim 105.

107. The method according to claim 105 or 106 in which the cation is an aminoglycoside antibiotic.