CA2117332A1 - Glycosylated steroid acid derivatives for transport across biological membranes and process for making same - Google Patents

Abstract

Novel glycosylated steroid derivatives for facilitating the transport of compounds across biological membranes, either in admixture or as conjugates, are disclosed. A novel process for efficient synthesis of these glycosylated steroid derivatives, using activated glycosyl sulfoxide intermediates is also provided.

Description

`:VO93/11772 PCT/US92/10778 GLYCOSYLATED STEROID DERIVATIVES FOR TRANSPORT ACROSS
BIOLOGICAL MEMBRANES AND PROCESS FOR MAKING SAME
This application is a continuation-in-part of Application Serial No. 07/806,985, filed December 13, 1991, the disclosure of which is hereby incorporated by referencq. This invention was made with Government support unde~r Grant No. N0014-91-J-1230, awarded by Office of Naval Research. The Government has certain rights in this invention.
BACKGROUND OF THE INVENTION
The present invention is generally directed to novel glycosylated steroid derivatives for facilitating the transport of molecules across 15 biological membranes and the blood-brain barrier. The -~
invention is further directed to a novel glycosylation process for the efficient synthesis of these glycosylated steroid derivatives.
To elicit the desired biological response, a molecule of diagnostic, prophylactic, or therapeutic interest ~termed herein "therapeutically-sign~ificant-molecule" or ~therapeutically-significant-compound~]
must be available in an effective concentration at its site of action. Many factors determine the concentration of a therapeutically-significant-compound that ultimately reaches the site of action, including the amount administered, and the extent and rate of the compound's absorption, distribution, biotransformation, and excretion. [Goodman and Gilman, 30 The Pharmacological Basis of Ther~peutics, 6th Edition, MacMillan Publishing Co., Inc., New York, 1980, pp. 1-39]. The foregoing factors may, in turn, be influenced by the route chosen for administration of the th~rapeutically-significant-compound.

CA211733~

The most common routes of administration of therapeutically-significant-compounds are parenteral (intravenous, subcutaneous, and intramuscular) and enteral (oral ingestion), although methods to administer therapeutically-significant-compounds across the skin or mucosa.(oral, nasal, rectal, vaginal, etc.) also are known. Parenteral methods are considered to be extremely effective in general, allowing for rapid increases in blood levels of a wide range of therapeutically-significant-compounds.
Parenteral methods are advantageous in that they circumvent first-passage hepatic metabolism. However, parenteral administration of a therapeutically-significant-compound can cause pain, irritation, possible tissue damage over the long term, and carries a potential risk of infection. In addition, parenteral methods frequently are inconvenient, partic~larly those that are restricted to trained medical personnel (e.g., intravenous methods).
Enteral methods are more convenient than parenteral methods, and generally are more economical and acceptable to the recipients. However, orally administered, therapeutically-significant-compounds may be inefficiently absorbed and the time from - .
2S ingestion to absorption may prohibit effective use in emergency situations. Moreover, many therapeutically-significant-compounds cannot be orally administered as they are destroyed, prior to reaching their site of action, by the digestive enzymes, acid, and surface-active lipids in the gut. Other therapeutically-significant-compounds are subject to extensive, first-passage hepatic metabolism, rendering them ineffective following oral administration.
Non-parenteral methods which circumvent problems associated with instability of drug preparations in CA2 1 17332 :~

the gut and first-passage hepatic metabolism long have ~`-been sought. Administration via transdermal, oral mucosal, rectal, and nasal routes are among the alternatives which have been explored. Such alternatives further include administering the therapeutically-significant-compound orally, but encapsulated in a protective delivery system designed to extrude the contents at a predetermined point in the lower gastrointestinal tract. However, the lo efficacy of these alternative drug delivery methods often is limited by poor absorption of the therapeutically-significant-compounds at the site of delivery or application. Effective strategies to enhance absorption of therapeutically-significant-molecules across cell membranes could enhance theefficacy of many known drug preparations which are poorly absorbed regardless of the method of administration. Such strategies to enhance trans-membrane absorption could be particularly useful for therapeutically-significant-compounds that are administered across the skin and mucosal tissues, including mucosal tissues of the gastrointestinal, genitou~inary, and respiratory tracts.
The basic structural unit of biological membranes is a phospholipid bilayer, in which are embedded proteins ~f various size and composition. The surfaces of the phospholipid bilayer, which project into the aqueous cellular environment, are formed by the hydrophilic heads of the phospholipids; the interior, by the fatty acyl hydrophobic tails. The membrane proteins may be involved in transport processes and also may serve as receptors in cellular regulatory mechanisms.
Natural mechanisms for traversal of biological membranes include passive diffusion, facilitated diffusion, active transport, receptor-mediated endocytosis and pinocytosis. Passive diffusion works best for small molecules which are lipid-soluble.
However, biological membranes are essentially impermeable to most water-soluble molecules, such as nucleosides, amino acids~ proteins, and other hydrophilic, therapeutically-significant-molecules.
Such molecules enter cells via some type of carrier-mediated transport system in which specific entities facilitate traversal of the membrane. Natural carriers for facilitating traversal of the membrane are of limited utility, however, as such carriers will accept substrates of only a predetermined molecular configuration. Many therapeutically-significant-~5 compounds are not efficiently absorbed because they - are neither lipophilic enough to cross cell membranes by passive diffusion nor recognized by the natural transport systems.
Strategies to enhance the uptake of therapeutically-significant-molecules across biological membranes have been investigated previously and fall into two broad categories. The first category includes all strategies in which the structure of the therapeutically-significant-compound is changed, either by making the compound more lipophilic itself, or by conjugating the compound to other entities known to interact with phospholipid membranes. The common goal of these strategies has been to increase passive diffusion across the membrane by lowering the energy barrier to diffusion and/or by increasing the local concentration of the compound at the membrane surface. Also included in the first category is a strategy for conjugating the therapeutically-significant-compound to entities known to interact with transport machinery embedded in the ^`YV0 93/11772 P~-r/US92/10778 biological membranes, the goal being to take advantage of the transport machinery (either active or facilitated transport or receptor-mediated endocytosis) to increase delivery of the compound across the membrane.
Many investigators are studying the feasibility of increasing the efficacy of hydrophilic compounds by conjugating these compounds to entities known to interact with phospholipid membranes. Among the techniques reported are utilization of oligonucleotide-cholesterol conjugates [Letsinger ~L
et al. "Cholesteryl-conjugated oligonucleotides:
Synthesis, properties, and activity as inhibitors of replication of human immunodeficiency virus in cell culture." Proc. Natl . Acad. Sci . USA 86: 6553-6556 (September 1989); Stein CA et al. "Mode of Action of 5'-Linked Cholesteryl Phosphorothioate Oligodeoxynucleotides in Inhibiting Syncytia Formation and Infection by HIV-1 and HIV-2 in Vitro."
20 Biochemist~y 30: 2439-2444 (1991)~.
Targeting molecules to the brain requires traversal of the blood-brain barrier -- a capillary-including system, with unique morphological characteristics, which acts as a system-wide cell~lar membrane separating the brain interstitial space from the blood. Like biolo~ical membranes, the blood-brain barri~r is relatively impermeable to many hydrophilic, therapeutically-significant-compounds. Among the strategies which have been, developed for targeting compounds to the brain are direct delivery by invasive procedures, intra-arterial infusion of hypertonic substances, and conversion of hydrophilic compounds to lipid-soluble entities. Recent attempts at facilitated transport, as described in United States Patent No. 4,902,505, involve coupling a hydrophilic WO93/11772 PCT/US92/107~-peptide of interest to a peptide carrier which, by itself, is capable of traversing the barrier via receptor-mediated transcytosis.
The second category of strategies to enhance uptake includes those in which the therapeutically-significant-compounds are administered to specific body surfaces as admixtures with other molecules which are known to permeabilize membranes. For example, several investigators have attempted to mix insulin with adjuvants, such as bile salts, which might enhance nasal insulin absorption. [Hirai et al., Int.
J. Pharmaceutics 9: 165-184 (1981); Hirai et al.
Diabetes 27: 296-199 (1978); British Patent No.
1,527,506; U.S. Patent No. 4,153,689; and Pontiroli et al. Br. Med. J. 28~: 303-386 (19~2)]. EP 0 444 778 describes the use of alkyl saccharides to enhance the penetration of topically applied drugs across mucus-covered epithelial tissues in general, and the corneal epithelium, in particular. U.S. Patent No. 4,865,848 to Cheng et al., issued September 12, 1989, discloses the use of sucrose esters, particularly sucr~se monolaurate, for enhancing the transdermal flux of transdermally-delivered drugs. U.S. Patent No.
4,746,508 to Carey et al J issued May 24, 1988, reports the use of fusidic acid and cephalosporin derivatives to increase the permeability of human and animal body surfaces to drugs.
The glycosylated steroid derivatives of the present invention are known to interact with phospholipid membranes, thereby enhancing the penetration of therapeutically-significant-compounds through such membranes, including biological membranes. Like some of the previously used adjuvants and enhancers (e.g., cholic acid and fusidic acid derivatives) the novel derivatives of the present ~-~093~11772 PCT/US92/10778 invention are amphiphilic in a facial sense. However, the novel steroid derivatives of the present invention have significantly different structures in that they are glycosylated on their hydrophilic surfaces, a feature not shared by any of the previously-known, facially-amphiphilic steroids. The present inventors have discovered that glycosylation on the hydrophilic surfaces significantly changes both the solubility properties of the steroids and the manner in which they associate. Many of these glycosylated steroids have been shown by the inventors to be more effective that the parent, non-glycosylated, steroids, in permeabilizing both artificial and biological membranes. The novel glycosylated steroid derivatives of the present invention, therefore, may be used to increase the delivery of therapeutically-significant-compounds across cell membranes, either in admixture with the compounds or as conjugates to the compounds.
Prior to the present invention, no method existed for synthesizing all of the glycosylated steroid derivatives of the present invention. Many glycosylation reactions using thioglycosides have been reported. [Ferrier RJ et al. "A Potentially Versatile Synthesis of Glycosides." Carbohydrate-Research 27: 55-61 (19733; Garegg PJ et al. "A
reinvestigation of glycosidation reactions using 1-thioglycosides as glycosyl donors and thiophilic cations as promoters." Carbohydrate Research 116:
162-5 (1983); Nicolaou KC et al. "A Mild and General Method for the Synthesis of O-Glycosides." J Am Chem Soc 105: 2430-2434 (1983); Lonn H. "Synthesis of a tri- and a hepta-saccharide which contain ~-L-fucopyranosyl groups and are part of the complex type of carbohydrate moiety of glycoproteins."
Carbohydrate Research ~39: 105-113 (1985); Andersson F

et al. "Synthesis of 1,2-cis-linked glycosides using dimethyl(methylthio)sulfonium triflate as promoter and thioglycosides as glycosyl donors." Tetrahedron Letters pp. 3919-3922 (1986); Brown DS et al.
"Preparation of cyclic ether acetals from 2-benzenesulphonyl derivatives: a new mild glycosidation procedure." Tetrahedron Letters 29/38: 4873-4876 (198~); Ito Y et al. "Benzeneselenenyl triflate as a promoter of thioglycosides: a new method for O-glycosylation using thioglycosides." TetrahedronLetters pp. 1061-4 (1988); Dasgupta F. et al. "Alkyl sulfonyl triflate as activator in the thioglycoside-mediated formation of ~-glycosidic linkages during oligosaccharide synthesis." Carbohydrate Research 177:
c13-c17 (1988)]. However, none of these reported methods teach the use of a glycosyl sulfoxide as a glycosylating agent.
Utilization of an activated glycosyl sulfoxide intermediate in a process for glycosylating steroids, previously has been reported by the inventors in J.
Am. Chem. Soc. 111: 6881-2 (1989), the content of which is hereby incorporated by reference. However, the reported method represents only preliminary results on the glycosylation of steroids of the -Formula (I). More specifically, furtherexperimentation in the series has revealed unique reaction conditions which are necessary to achieve the efficient and stereo-selective synthesis of glycosylated compounds of the ~ormula (I). The reaction solvent used plays a critical role in the stereoselectivity of glycosylation. Using a non-polar, aprotic solvent increases selectivity for alpha (~) glycosidic bond formation while the use of a polar, aprotic solvent such as propionitrile increases selectivity for beta (~) glycosidic bond formation.

lA 2 ~ 1 7332 ~093/11772 PCT/US92tlO778 g The type of sulfoxide used in the glycosylation reaction also affects the outcome of the reaction.
For example, it is vital to use the para-methoxy phenyl sulfoxide as the leaving group in the novel process described herein to obtain good yields of beta (~) selectivity in the glycosidic bond formation. The yield of the glycosylation reaction yielding alpha (a) or beta t~) glycosidic linkages also may be increased by the use of less than one equivalent of triflic anhydride in the glycosylation process.
Finally, the protecting groups on the glycosyl donor also have an impact on the stereochemical course of the glycosylation reaction. When the protecting group used on the glycosyl donor is pivaloyl, only lS beta (~) glycosidic bonds are formed in the glycosylation process, regardless of whether an aprotic, non-polar solvent or an aprotic, polar solvent is used for the reaction. The above factors taken together indicate that one skilled in the art could not have practiced the invention without the detailed further experimentation provided herein.

SUMMARY OF THE INVENTION
The present invention is generally directed-to novel, facially-amphiphilic, glycosylated steroid derivatives which have been found to be soluble in both aqueous and membrane~ e environments. These unique solubility properties permit the glycosylated steroid derivatives to facilitate the transport of other molecules across biological membranes and the blood brain barrier. It is, therefore, contemplated that the glycosylated steroid derivatives of the present invention can be used, either in admixture with the therapeutically-significant-molecules or by being conjugated to such molecules, to enhance WO93/11772 PCT/US92/107~

delivery of the molecules across body surfaces including, but not limited to, the buccal, sublingual t conjunctival,.rectal, gastric, intestinal, endometrial, cervical, vaginal, or colonic e~ithelium;
5 the oropharynx, ear canal, respiratory tract, nasopharynx, urethra, urinary bladder, and tympanic membrane. Alternatively, the glycosylated steroid derivatives of the present invention may be administered in admixture with the glycosylated steroid derivative/therapeutically-significant-molecule conjugate [hereinafter referred to as the "derivative-compound-conjugate"~ to further enhance facilitation of trans-surface and trans-membrane transport.
S ~t is further contemplated that the novel glycosylated steroids of the present invention may be used for the delivery of antiviral agents, systemic insecticides, and herbicides, across plant surfaces;
and, for the delivery of contact insecticides and miticides, across arthropod surfaces~
A novel process for obtaining these novel, facially-amphiphilic, glycosylated steroid derivatives and other glycosylated steroids is also disclosed.
'Of particular interest are the steroid derivatives of the general formula (I) C~21 1 733~
~093/11772 PCT/US92/10778 ~4 "..

R2 ¦

A' ~ ~R3 (I) wherein A is O, O~, oR6~ NR7R8, N3, NHCo~7~ OCOAr o-C-oR9, OCOR9, NCH2C6H5;

Ar is phenyl or phenyl substituted with 1-3 groups l selected from the group consisting of halogen, C~-C~2 alkyl or Cl-C3 alkoxy;
a is a single bond in the alpha or beta configuration with the proviso that when A=O, a is a double bond;
R1 is H which is cis or trans to R2;
R2 is CH
R3 is H, OH or oR6;
R4 is H, OH or OR~;
R5 is ~O2R1, CH2OR9, CONH2, CoNHR7, CoNR7R8, C--S--Rlr CH2S (O) p-S~RI
CH2NH2, CH2NHR7, CH~NR7R~, CH~-S~O)p-S-RI, R6 is a monosaccharide where the glycosidic linkage at the anomeric carbon atom in said monosaccharide is alpha or beta or is an oligosaccharide of 2-l0 monosaccharides where the glycosidic linkage at any of the anomeric carbon atoms in each 3s C A2 1 1 733~

- ~2 -monosaccharide residue of the oligosaccharide is independently alpha or beta;
R7 and R8, independently are H, C,-C4 alkyl, C3-C7 cycloalkyl, C4-C~o alkylcycloalkyl, phenyl, benzyl, or, taken together are (CH2)~, where f=3-6;
R9 is H or, C~-C3 alkyl;
R~ is H, C1-C10 alkyl, C~-C10 alkenyl, C,-C10 alkynyl, C6H~
or CH2C6H5;
monosaccharide is a protected or deprotected hexose or deoxyhexose selected from the group consisting of D- or L-allose, D- or L-altrose, D- or L-glucose, D- or L-mannose, D- or L-gulose, D- or L-idose, D- or L-galactose, and D- or L-talose or a protected or deprotected furanose or deoxyfuranose selected from the group consisting of D- or L-ribose, D- or L-arabinose, D- or L-xylose and D- or L-lyxose where said protecting groups for the hydroxy groups of said hexoses or furanoses are selected from the group consisting of benzyl, pivaloyl, trimethylsilyl, tert-butyldimethylsilyl, tert-butyldiphenylsilyl, tri-isopropylsilyl, acetyl, tetrahydropyranyl, benzoyl, Cl-C3 alkyl, isopropylidene, benzylidene, ~2-methoxyethoxy)methyl, orthoester, para- --methoxybenzyl and allyl;
p is 0, l or 2;
n is 0, l or 2;
or a pharmaceutically suitable salt thereof.
Of further interest are conjugates of a therapeutically-significant-compound linked to a compound of Formula (I) through any of Rs~ where said therapeutically-significant-compound includes, but is not limited to, anti-bacterials such as polyene antibiotics (erythromycin), beta-lactam antibiotics, and peptide-based or steroidal antibiotics; anti-`~VO93/11772 PCT/US92/10778 fungal agents such as lO-thiastearic acid and 24-thiacholestanol; peptides or proteins, such as regulatory factors, enzymes, antibodies, hormones, and toxins; nucleotides, nucleosides and nucleic acids;
and saccharides; and wherein the A and B rings of the steroidal residue of said compounds of Formula (I) may be cis or trans to one another; and the O glycosidic linkage at C-7 and C-12 may be in the alpha or beta configuration, each independently of the other. Also provided are methods for facilitating the transport of any therapeutically-significant-compound across a biological membrane, either in admixture with a glycosylated steroid derivative of the present invention or in the form of a derivative-compound-~5 conjugate. Alternatively, a method is provided forfurther enhancing trans-membrane transport of the derivative-compound-conjugate by administering the derivative-compound-conjugate in admixture with a glycosylated steroid derivative of the present invention, which may be either the same as, or different from, the derivative of the conjugate.
Also provided are pharmaceutical compositions containing (l) effective amount of a compound of Formula (I) and a suitable pharmaceutical carrier, (2) an effective amount of a compound of Formula ~I), an effective amount of a therspeutically-æignificant compound, and a suitable pharmaceutical carrier; (3) an effective amount of derivative-compound-conjugate and a suitable pharmaceutical carrier; or (4) an effective amount of a compound of Formula (I), an effective amount of derivative-compound-conjugate, and a suitable pharmaceutical carrier.
The invention is further directed to a novel process (Scheme I) for the efficient synthesis of glycosylated steroid derivatives of the Formula (I) C A2 1 1 73 3~

which comprise: (a) reacting a protected glycoside, which is prepared by standard methods for conversion of a sugar well known to those of ordinary skill in the art, where the oxygen atoms at all positions of the sugar except the anomeric position are protected with the same or different groups, with (b) an -S-R
entity under standard conditions, where R is C,-CIo alkyl, pyridyl, furyl, thienyl, phenyl or phenyl substituted with 1-3 groups selected from the group comprising halogen, Cl-C3 alkyl, NO2, C~-C3 alkoxy, to yield a protected thio-glycoside which is further reacted with (c) meta-chloroperoxybenzoic acid to yield the corresponding sulfoxide derivative and (d) converted to an activated glycosylating agent intermediate using a triflate-containing compound, such as triflic anhydride, methyl triflate or trimethylsilyl triflate and contacting said activated glycosylating agent with (e) a steroid (in which any oxygens which are not to be glycosylated have been protected by standard methods) in the presence of 2,6-di-tert-butyl-4-methylpyridine in toluene, fôr formation of alpha, ~lpha glycosidic linkages, or in propionitrile, for the formation of beta, beta glyaosidic linkages thereby yielding a protected--glycosylated steroid which is then deprotected byremoving the protecting groups by ~f) standard procedures to yield glycosylated steroids of the Formula (I). The oxygen-protecting groups utilized may be either electron-withdrawing groups ~uch as esters; or electron-donating groups, such as ethers, including alkyl, silyl, phenyl or benzyl ethers.
Howe~er, if a pivaloyl ester is the protecting group used, the resulting glycosidic linkage that is formed is always ~,~ regardless of the solvent used for the reaction. The resulting compounds of the invention C~21 ~ 7332 ~093/11772 PCT/US92/10778 may be characterized by proton NMR, Ct3-NM~, high resolution mass spectroscopy, X-ray crystallography and thin layer chromatography.
Also provided is a process for synthesis of the novel derivative-compound-conjugates of the present invention.
Preferred for their ability to permeabilize biological membranes are those compounds of Formula (I) where:

A is OH, oCoR9, OCOC6Hs, OCOC~5-pOMe, NH2;
a is a single bond;
R3 is oR6;
R4 is oR6;
Rs is CQ2R1, CoNR7R8;
R6 is a monosaccha~ide where the glycosidic linkage at the anomeric carbon atom in said monosaccharide is alpha or beta;
R10 is H or C,-C10 alkyl;
monosaccharide is a protected or deprotected hexose such as D- or L-glucose where the protecting groups are benzyl or pivaloyl.
Preferred for their ability to permeabilize biological membranes are:
(a) 3a-O-benzoyl-trans-5,lO-bis-~ 7,l2-glucosyl cholic acid methyl ester;
(b) 3a-hydroxy-cis-5, 10-bis-a,~-7,12-glucosyl cholic acid;
(c) 3~-hydroxy-cis-5, 10-bis-a,~-7,12-glucosyl cholic acid methyl ester;
(d) 3~-hydroxy-cis-5, 10-bis-a,a-7,12 -glucosyl-25-tryptopha~yl cholic acid;
(e) 3a-ethylcarbonate-cis-5,l0-bis-a,a-7,l2-glucosyl cholic ~cid methyl ester;

WOg3/11772 PCTJUS92/1077' (f) 3~-O-benzoyl-cis~5,10-bis-~,~-7,12-glucosyl cholic acid methyl ester;
(g) 3~-O-p-methoxybenzoyl-cis-5,10-bis~ -7,12-glucosyl cholic acid methyl ester;
(h) 3~-O-benzoyl-cis-5, 10-bis-~, ~-7,12-glucosyl cholic acid methyl ester;
(i) 3~-hydroxy-cis-5,10-bis-~,~-7,12-glucosyl cholic acid;
(j) 3~-O-benzoyl-trans-5,10-bis-~ 7,12-glucosyl cholic acid methyl ester;
tk) 3~-hydroxy-trans-5,10-bis-~ 7,12 glucosyl cholic acid.
Particularly preferred is Compound G above, 3~-O-p-methoxybenzoyl-cis-5,10-~is-~,~-7,12-glucosyl cholic acid methyl ester and its acid form tI8 THIB ~ CORR~CT
DESCRIPTIO~???], 3~-O-p-methoxybenzoyl-cis-5,10-bis-a, ~-7,12-glucosyl cholic acid.
DESCRIPTION OF THE DRAWINGS
Figure 1. A graph depicting the enhancing effect of CME, a novel glycosylated steroid derivative of the present invention, on the efficacy of thiastearic acid, an antifungal agent.
Figure 2. A graph depicting the enhancing effect of CME, a novel glycosylated steroid derivative of- the present invention, on the efficacy of thiacholestanol, an antifungal agent.
Figure 3. A graph depicting the lack of an snhancing effect of CDE, the non-glycosylated version of CME, on the efficacy of thiacholestanol, an antifungal agent.

DETAILED DESCRIPTION OF THE INVENTI~
The introduction of molecules of dia~nostic, prophylactic, or therapeutic interest across body surfaces and/or into cells requires the transversal of ~A2 1 1 7332 `VO93/11772 P~T/US92/10778 one or more semi-permaable biological membranes. The compounds of this invention are useful in permeabilizing biological membranes, thereby assisting body surface and/or membrane transversal of therapeutically-significant-compounds. In one embodiment, the therapeutically-significant-compound is administered in admixture with a glycosylated steroid derivative of the present invention. In another embodiment, trans-surface and/or trans-membrane transport is facilitated by administering thetherapeutically-significant-compound in the form of a derivative-compound-conjugate in which the compound of interest is conjugated to the glycosylated steroid, i.e., Rs is linked to a therapeutically-significant-~5 compound. Alternatively, the derivative-compound-conjugate may be administered in admixture with a novel glycosylated steroid derivative of the present invention, which may be either the same as, or different from, the derivative of the conjugate.
The novel glycosylated steroid derivatives of the present invention may be expected to enhance the therapeutic efficacy of a wide variety of compounds.
As a result, many therapeutic applications for the compounds of the present invention may be -cont~mplated. ~embrane permeable therapeutic agents could be used in the treatment of a wide variety of illnesses such as AIDS and other chronic viral infections, cancer, bacterial and fungal infections, and metabolic diseases such as lupus, diabetes and rheumatoid arthritis.
The ahility of the novel glycosylated steroid derivatives of the present invention to interact with, and/or permeabilize, biological membraneæ, is believed to result from the compounds' facial amphiphilicity.
The glycosylated surface of the derivatives is CA2 1 l 7332 W093/11772 PCT/US92/107~-hydrophilic; the non-glycosylated surface, hydrophobic. This facially amphiphilic structure confers unusual properties on the molecules, including an ability to self-associate in both hydrophobic and hydrophilic environments, and to organize at amphiphilic interfaces. Some of the glycosylated steroid derivatives of the present invention have now been shown, by the inventors, to crystallize in layers, with alternating hydrophobic and hydrophilic layers. The non-glycosylated, parent steroid compounds, although possessing some facial amphiphilicity, do not crystallize in register and in organized layers ~ike the glycosylated steroids. In addition, the solubility properties of the glycosylated steroid derivatives of the present invention differ substantially from those of the parent compounds. More particularly, the novel glycosylated steroid derivatives of the present invention, while more soluble than the parent compounds in an aqueous environment are, unexpectedly, not significantly less soluble in an organic environment. Based on these observations, the inventors belie~e that the novel glycosylated steroid derivatives of the present invention permeabilize membranes by self-associating to form small, rever~e micelles, with their hydrophobic surfaces exposed to the lipids within the membranes. These rever~e micelles may function as water-filled pores, allowing therapeutically-significant-compounds to pass through, or the presence of these reverse micelles in the membrane may perturb membrane order.
Additionally, the compounds of the present invention facilitate the transport of protons or other ions such as Ca 12, Na~ or K~ across biological --~093/11772 PCT/US92/10778 membranes, indicating their use as potential antifungal or antibiotic agents.
The derivative-compound-conjugate of the present invention can be used in vivo, as a component of a pharmaceutical composition in a manner similar to more conventional therapeutic agents. Administration of the derivative-compound-conjugate to an individual with a chronic viral infection may inactivate the virus or the derivative-compound-conjugate may contain an antisense oligonucleotide sequence which is inhibitory to viral gene or oncogene activity. For the individual with a genetic defect, the therapeutically-significant-compound can be a protein which supplements a missing or defective protein.
lS The derivative-compound-conjugate may be administere~ as a pharmaceutical composition via a variety of routes, including subcutaneous, intravenous, intramuscular, intrasternal, intranasal and intracranial injection or infusion. The pharmaceutical composition also may be administered topically or ~ia inhalation.
More specifically, the compounds of thi~
invention,,including the compounds of Formula ~I) and the derivative-compound-conjugates, can ~e -administered to treat chronic viral infections such asAIDS (Acquired Immune Deficiency Syndrome) or herpes simplex; autoimmune diseases such as lupus, rheumatoid arthritis; diabetes, cystic fibrosis, growth hormone deficiencies; and cancer, by any means that produces contact of the active agents with the appropriate site of action in a mammal. They can be administered by any conventional means available for use in conjunction with pharmaceuticals, either as individual therapeutic agents or in a combination of therapeutic agents. They can be administered alone, but are W093/11772 PCT/US92~ 107?^

generally administered with a pharmaceutical carrier selected on the basis of the chosen route of administration and standard pharmaceutical practice.
The dosage administered will, of course, vary depending upon known factors such as the pharmacodynamic characteristics of the particular agent, and its mode and route of administration; age, health, and weight of the recipient; nature and extent of symptoms; kind of concurrent treatment, frequency of treatment, and the effect desired. Usually a daily dosage of therapeutically-significant-compound can be about 0.1 to 100 milligrams per kilogram of body weight. Ordinarily 0.5 to 50, and preferably 1 to 10 milligrams per kilogram per day given in divided ~5 doses 1 to 6 times a day or in sustained release form is effective to obtain desired results.
Dosage forms (compositions) suitable for internal administration contain from about 1 milligram to about 500 milligrams of therapeutically-significant-compound 20 per unit. In these pharmaceutical compositions the therapeutically-significant-compound ordinarily will ~e present in an amount of about 0.5-95~ by weight based on the total weight of ~he composition.
The compositions can be administered orally in solid dosage forms, such as capsu~es, tablets, and powders, or in liquid dosage forms, such as elixirs, syrups, and suspensions. The compositions also can be administered parenterally, in sterile liquid dosage forms, by inhalation in the form of a nasal spray or lung inhaler, or topically as an ointment, cream or lotion.
Gelatin capsules additionally may contain powdered carriers, such as lactose, sucrose, mannitol, starch, cellulose derivatives, magnesium stearate, stearic acid, and the like. Similar diluents can be C~21 1 7332 - 21 ~

used to make compressed tablets. Both tablets and capsules can be manufactured as sustained release products to provide for continuous release of therapeutically-significant-compound over a period of hours. Compressed tablets can be sugar coated or film coated to mask any unpleasant taste and protect the tablet from the atmosphere, or enteric coated for selective disintegration in the gastrointestinal tract.
Liquid dosage forms for oral administration can contain coloring and flavoring to increase patient acceptance.
In general, water, a suitable oil, saline, aqueous dextrose (glucose), and related sugar solutions and glycols such as propylene glycol or polyethylene glycols are suitable carriers for parenteral solutions. Solutions for parenteral administration additionally may contain suitable stabilizing agen~s, and if necessary, buffer substances. Antioxidizing agents such as sodium ~isulfite, sodium sulfite, or ascorbic acid either alone or combined are suitable stabilizing agents.
Also used are citric acid and its salts and sodium ~DTA. In addition, parenteral solutions can contain preservatives, such as benzalkonium chloride, methyl or propyl-paraben, and chlorobutanol~

SYNTHESIS
~ .
The compounds of Formula (I) can be prepared according to the process shown in Scheme I.

1. m~CPBA

R0 2.Tf20,-78 C RO
~ O 3.Nucleophlle,Base R0 R0 ~ RO ~
RO SPh R0 Nuc s SCHEME I
A protected thioglycoside is oxidized with m-chloroperoxybenzoic acid under standard conditions to yield the corresponding sulfoxide. Triflic anhydride (Aldrich) is then added to a solution of the protected glycosyl sulfoxide in toluene at -78C
followed by the addition of an acid scavenger such as 2,6-di-tert-butyl-4-methyl pyridine (Aldrich Chemical Co.) in toluene and the nucleophile dissolved in toluene at -78C. After stirring for 15-30 minutes, the reaction was removed from the ~old bath and stirred for an additional l0 minutes and quenched by pouring the mixture into aqueous sodium bicarbonate and the protected adduct was isolated by chromatography. Deprotection of the adduct under s~andard conditions yields compounds of the Formula ~I). The appropriate thioglycoside is obtained via standard protection of a selected sugar followed by thioglycoside formation according to methods descri~ed above. Via this method, bis-glycosylation of a steroid derivative of the Formula (I) where R3 and R4 are OH selectively produces ~,~ glycosidic linkages with the glycosyl donor, except where the protectin~
grcu~ used is pivaloyl, in which case only ~,~
glycosidic linkages are formed regardless of th~
solvent used for the reaction.
Alternatively, the protected glycosyl sulfoxide, nucleophile and pyridine base are dissolved in propionitrile at -78OC, followed by the addition of triflic anhydride at -78C and the product is isolated as described above. Via this method, glycosylation of Ca21 1 7332 ~093~11772 PCT/US92/10778 a steroid derivative of the Formula (I) where R3 and R4 are OH selectively produces ~,~ glycosidic linkages with the glycosyl donor. It is vital to use the p-methoxy phenyl sulfoxide as the leaving group in the above process to obtain the ~,~ selectivity in the glycosylation.
The compounds of this invention and their preparation are illustrated further in the following examples. All temperatures are in degrees Centigrade and parts and percentages by weight. In these Examples, unless otherwise indicated, the reactions were performed under an atmosphere of dry argon;
"isolation by extraction" refers to the liquid-liquid extraction of a water containing mixture with an indicated solvent, followed by drying the organic phase over sodium sulfate, filtering, and evaporating the solvent under reduced pressure; chromatography refers to the method of medium pressure column chromatography described by W. C. Still, et al., Journal of Organic Chem., 43: 2923 (1978).

Example l Part A: Perbenzylated-3~-Qthylcarbo~ate-cis-5,10-bis-~ gluco~yl cholic a~id methyl e~t~r~
A l00 ml round bottom flask containing a Teflon~
stir bar is flame dried and cooled to -78C
(acetone/dry ice bath) under argon. 2,3,4,6-tetra-O-benzyl glucose sulfoxide (2.97 g, 4.57 mmol, ~.0 eq.), C3 ethylcarbonate cholic acid (0.563 g, 1.14 mmol, l.0 eq.) and 2,6-di-tert-butyl-4-methylpyridine (0.936g, 4.57 mmol, 4.0 eq.) are each dried by azeotroping each separately three times with toluene (l5.0 ml~. Triflic anhydride (824 ~l, 4.57 mmol, 4.0 eq.) is added to the glycosyl sulfoxide dissolved in toluene (5.0 ml) at -78C. To this mixture is then 2 1 1 733~
WO93/11772 PCT/US92/107~-added the pyridine base in toluene (5.0 ml). After five minutes, the cholic acid derivative, dissolved in methylene chloride ~1.0 ml) and toluene ~5.0 ml). is added. The reaction is allowed to stir at -78C for thirty minutes and then removed from the dry ice bath.
After ten minutes, the reaction is quenched by the addition of saturated sodium bicarbonate and the product was isolated by extraction with methylene chloride and purified by flash chromatography on lo silica gel to provide the title compound (60%) as an oil, RF = - 3 (20% ether/CH2C12) .

Part B: 3~-ethylcarbonate-cis-5,10-bis-~glucosyl cholic acid methyl e~ter.
~5 Palladium hydroxide (0.030 g, 15% by weight) is added to a mixture of the product of Part A ~0.220 g, 0.014 mmol, 1.0 eq.) dissolved in benzene (4.0 ml) and methanol (32.0 ml) at room temperature. The mixture is hydrogenated at 50 psi for 48 hours. The product is filtered throuqh Celite~ (diatomaceous silica, Johns-Manville Corp.) under nitrogen. The so~lvent was evaporated and the oil was flash chromatographed with 10~ methanol/methylene chloride. To remove the silica gel that dissolves under elution conditions, the product is run through on a reverse phase LH-20 column using methanol as an eluent. The solvent is evaporated to yield the title compound (65%) as a white powder, RF = 0-3 (15% Me0H/CH2C12), NMR (CDC13 500 MhZ) ~:5.04 (m, Lh, anomeric ~-H), 4.82 (m, Lh, anomeric ~-H).

~A 2 1 1 73 -~
--'WO93/11772 PCT/~S92/10778 Example 2 3~-benzoyl-cis-5,10-bis-~,~-glucosyl cholic acid methyl ester.
2,3,4,6-tetra-O-benzyl p-methoxy glucose S sulfoxide (1.012 g, 1.45 mmol, 4.0 eq.~, C3-0-benzoyl cholic acid methyles~er (0.191 g, 0.364 mmol, 1.0 eq.
and 2,6-di-tert-butyl-4 methyl pyridine (0.179 g, O.874 mmol, 2.4 eq.) are azeotroped together three times from toluene (20 ml). After removing the toluene under reduced pressure for the last time, t,he mixture is dissolved in freshly distilled propionitrile and cooled under argon in a dry ice/
acetone bath at -78C. Triflic anhydride t244 ~l, 1.45 mmol, 4.0 eq.) is added and the reaction mixture is stirred at -78C for 40 minutes. The reaction vessel is removed from the ice bath and stirred for an additional 10 minutes. The reaction is quenched by pouring it into saturated sodium bicarbonate and the product is isolated by extraction with methylene chloride and purified by flash chromatography on silica gel. Catalytic hydrogenation to remove the benzyl protecting groups is accomplished as described above to yield the title compound (60~) as an oil, RF =
0.3 (15% Me~H~CH2Cl2~, NMR (CDCl3 500 Mhz) ~:4.36 ¢d, lH, J=7.~2Hz, anDmeric ~H), 4.37 (dl lH, J=7.92Hz, anomeric ~-H).
The compounds of Example 1 and 2 and compounds which were prepared or could be prepared following procedures analogous to those outlined above are shown in Table I.

CA21 17;33~
WO 93~11772 PCl'/US92/107?' o a) a) ~ I ~ a) o o o o o ~ s o In In U~ U~ In In o o o o o o U O O ~

O O O O O O

> ~ ~ ~ _ ~ ~

Y ~ * I O O O 0 ~0 0 ~I t~) ~ ) U ~.) O O
~
~r ~ ~' ~3 ~ ~ o o o o o o ~r~, :~ X 3~ X

s ~
~1 0 0 0 0=0 0 ~: OC Or Q=~) X¦ D ~ ~ ~

(~42 1 1 7332 WO 93/11772 PCI/US9t/10778 ~;l o o o o o o o o o o c~ o o o *
t O O O O O O O O O O O O O O
S 1 O U U ~ U U U ~ U

c) o o o o o o o o o o o o o 0 ~ a) 0 ~ o o o o o o o o o o o o o o p~ u u u o u c) u u o u ~ ~ o u - o o o o o o o ~ o o o o o o ~d ~ N ~

~ p~ ~ p~ w ~
~:1 0 0 0 V ~ O O
U S O X ~ O--U O = U
o o o o o o o z o o o o o o Xl ~ ~ ._ O ~1 ~ ~ ~ Ir) ~ I~ ~o ~ o CA21 1733~
WO ~3/1 1772 P~/USg2/107'- :

~;1 o o o o ~ o o o o o o o o V ~ V C~ C~ V

a) c o o o o o o o o o o o o o o O O o O O o O O O O O O o ~ ~ 0 ~ ~ 0 ~
q~ ~ o o o o o o o o o o o o o ~; u v o t~

o o o o o o o o o o o o o ~

~¢ P~ U C~ U

U

~ ~ ~ s ~ ~
r~ r I r I r~ ~ ~ r~ ~
O O O C~ O O O O O OO O O

Xl ~ ~ ~ ~ Lt) ~C) r` c~ o~ o , W¦ N ~ ~ t~

--"WO 93/11772 PCI/US92/1077 X
P~l O O O O O O O
V V

C
~1 0 0 0 0 0 0 0 U O O c~ V t~ ~
o o o o o o o o o o o o o o o o ~n p~ ~ J~ O
U U t) U ~ ~

- o ~ o ~ o ~ o ~ o ~ o ~ ~ ~ o i3 ~ ~1 ~ v ~) v o t~

~ ~ ~ ~ ~l w ~ ~
~CI O O O O O O O O
o - v o =u o - ~ o =u o =c~ o =~ o =cJ c~
o o o '~ o o o o Xl ~ l~ ~ I` oo C~ o CA2l 17332 WO 93/11772 P~/lJS92/1077' ~1 o o o o o o o o o o o C~ O V C~ U C~

~ ~ _ _.

~, ~ ~ n o o o o O O O O O O O
52 Q Q ~2 ~ Q Q ~
I
o O o o o o o o o o o ,~

~ ~ In ~ o ~ o l:C o o o o o o o o u ,Q R ~ R Q ,Q ~

ooo~ oooo o o o ~ ~ v o t~ u c~
E~

W

S .C ~ ~ ~ S ~ ~ ~
~1 0 0 0 0 0 0 0 0 0 0 0 U U U V U o=V 0=0 o--- v o=~
o o o o o o o o o o o Xl ~ ~ ~ ~ o ~ c~

`"~WO 93/11772 PCl/US92/1077X

~ .~
l_ ~ o X' _ --tJ ~ _ _ , o :s:
u .

~ , .,.~.,.~ r h S ~ ~
-~ O ~1 B ~0 lo ~5 O O
Q .c s \ \ u~
t5`Q o ~ o--~-' I ~ , I
~ o 1/ ' ~ I11 il 11 11 11 11 ,1 o a) Q ~ o , I
u~ ~ Q ~ o r ~ I
I
Il 11 11 11 C: ~
~n u~
* * ~: U ~ .-~

PCI /U~;92/1077 ~1 _ N

t) ~s h o o OCO OD '' ~ o ~
~ .~
~ ~ o~
o o ~ -E~ --W N
~r d' 10 Ul S:~ O O O
In LS~ U~ In .. .. .. ..

O O O O
O O O O

L~ a a c ,~ V C~
~: ____ Z ........

~WO93/11772 PCT/US92/10778 Use The compounds of the i~vention have been shown to interact with, and permeabilize, biological membranes and to enhance the efficacy of antibiotics and antifungal agents on living cells. Since the compounds of the invention have been shown to permeabilize membranes, and the compounds themselves have no effect on cell growth at the concentrations used, it is assumed that the enhanced efficacy is related to increased delivery of the therapeutically-~ignificant-compounds to the cells.
The utility of the compounds for permeabilizing membranes was demonstrated using an assay (Hoyt, D.W~, et al. Biochemistry, Vol. 30, 10155 (l99l)) in which a fluorescein derivative is encapsulated at self-quenching concentrations inside vesicles. An increase in fluorescent intensity upon addition of a test compound indicates leakage of the fluorescein derivative out of the vesicle and therefore implies a disruption of the membrane. Tha compounds of the present invention induced a rapid and significant increase in fluorescent intensity at very low concentrations (0.05mM-0.5mM~, indicating phospholipid membrane permeabilization.
In addition, both light scattering and turbidity measurements on vesicles treated with selected glycosylated steroid derivatives (at concentrations which induce lO0~ leakage of carboxyfluorescein) showed that the average size of the vesicles was not significantly different from that of untreated vesicles. Moreover, electron micrographs of vesicles treated with selected glycosylated steroid derivatives (at concentrations which induce 100% leakage of carboxyfluorescein) did not show significant changes in morphology relative to untreated vesicles. The W O 93/11772 PC~r/US92/107'^ - 34 -glycosylated steroid derivatives of the present invention, therefore, permeabilize membranes without destroying the vesicles or inducing extensive fusion.
The inventors believe, based on NMR studies of aggregation in solution and also on crystallographic evidence, that the glycosylated steroids of the present invention self-associate and insert into membranes in an associated form, and that membrane permeabilization is related to this process. Although the pure phospholipid vesicles used in this assay do not have the complexity of biological membranes, the inventors have shown that compounds which work well in this assay also enhance the action of therapeutically-significant-compounds (e.g., antibacterial agents and antifungal agents) on living cells. This finding supports the concept that the ability of the glycosylated steroid derivatives to interact with phospholipid bilayers is related to the ability of the derivatives to enhance therapeutic efficacy. It further indicates that the carboxyfluorescein assay is a reasonable initial model system for identifying potential candidates for the permeabilization of biological membranes.
A variation of the above assay (V.E. Carmichael et al. J. Amer. Chem. Soc., Vol. III, 767 (1989) ? was employed to determine whether the compounds make the membranes permeable to protons at extremely low concentrations (O.OlmM-0.005mM~. For this assay, the fluorescein derivative was encapsulated inside vesicles at non-quenching concentrations in a Ph 6.5 buffer. The vesicles were diluted into a Ph 5.5 buffer and a compound of Formula (I) was added at a concentration lower than the concentration required to make the membranes permeable to the fluorescein derivative. After addition of compounds of the - 35 ~

Formula (I), the fluorescent intensity decreased, indicating ~hat the membrane had become permeable to protons.
The utility of the glycosylated steroid derivatives of the invention for permeabilizing phospholipid membranes suggested the usefulness of the derivatives for enhancing the permeability of cell membranes, which are composed in large part of phospholipids and other lipids, to therapeutically-significant-molecules. This use was demonstrated in assays testing the efficacy of two different antifungal agents for killing Crithidia fasciculata.
The use further was demonstrated in assays testing the efficacy of erythromycin for killing E . Coli ATCC
25922 cells.
ASSAY I: Leakage of C~rboxyfluorescein from V~si~le~
To a 25 mL round bottom flask 20.5 mg egg yolk (Sigma,.average MW 770.4) dissolved in CHCl3/MeOH, 5~0 mg phosphatidyl glycerol (Sigma, MW ?72) dissolved in CHCl3/MeOH, and 12.7 mg repurified cholesterol (Aldrich, MW 386.66) were added. The molar ratio of egg yolk; phosphatidyl glycerol:cholesterol was 4:1:5 (6~ ~moles total lipid). The solvent was removed on a rotary evaporatox. The dried lipid mixture was then put under argon and 3 mL freshly distilled diethyl ether was added. After the lipid had redissolved, 1 mL of carboxyfluorescein dissolved in water (pH
adjusted to 7.4) was added to a concentration of 180 mM (the concentration of carboxyfluorescein was determined by W ; the extinction coefficient at pH 7.4 is 5.6 X 104; ~max=492). The lipid mixture containing carboxyfluorescein was sonicated under argon in a bath type sonicator at 5-15C for 15-30 minutes. The mixture was then placed on the rotary evaporator and the organic solvent was removed. To separate the CA21 1 73-~2 car~xyfluorescein-loaded vesicles from unencapsulated carboxyfluorescein, the remaining aqueous ve-~icle mixture was loaded on a Sephadex ~-25 column equilibrated with 145 mM NaCl/l0 mM Hepes at pH 7.4.
The carboxyfluorescein-loaded vesicles eluted in the first fraction after the void volume while the unencapsulated carboxyfluorescein remained on the column. The purified vesicles were diluted with 145 mM NaCl/l0 mM Hepes buffer (pH 7.4) until the fluorescent intensity of the vesicle mixture measured approximately l0.
Because the carboxyfluorescein is encapsulated at self-quenching concentrations in the vesicles, an increase in fluorescent intensity over time indicates that the fluorophore is leaking out of the vesicles into the buffer. 5~ Triton Xl00 was added in 50 ~L
MeOH to a sample of the vesicle solution to determine the maximum possible fluorescent increase (Triton Xl00 is a nonionic detergent that at the high concentration use,d breaks vesicles by solubilizing the lipids). The ability of each glycosylated steroid to induce the release of carboxyfluorescein from the vesicles was determined by monitoring the increase in fluorescent intensity upon addition of glycosteroid. For each experiment, 50 ~L of glycosteroid in methanol (initial concentrations ranged from 0~6145 to 2.458 mM) was added to the cuvette and the fluorescent intensity followed over lO minutes. A control in which 50 ~L
pure methanol was added showed that methanol alone does not cause a significant increase in fluorescent intensity. However, several of the glycosteroids efficiently permeabilized vesicle membranes at very low concentrations, permitting the carboxyfluorescein to leak out into the buffer. The results are summarized in Table II.

C~21 1 7332 -~WO93/11772 P~T/US9~/10778 If the concentrations required to induce significant ~i e., >50%) leakage are taken as a measure of efficacy, then compounds 7, 8, and 11, are the most effective glycosylated steroids tested for permeabilizing phospholipid membranes in this assay.
Compounds 7 and 8 have a cis A/B ring junction and two ~-linked glucose sugars attached to the hydrophilic face of the molecule. Compound 11 also has two ~-linked glucose sugars attached to the hydrophilic face of the molecule. Cholic acid, deoxycholic acid, and chenodeoxycholic acid, compounds known to permeabilize biological membranes in other uses ~Gordon GS et al.
Proc. Nat'l . Acad. s~i . USA 82: 7419-7423 (1985)) also permeabilize membranes in this assay, although at much higher concentrations than many of the compounds of the present invention. From these observations, it may be concluded that glycosylation changes the chemical properties of the -~teroids, making them more efficient at permeabilizing membranes.

TABLE II
_ EX CONCENTRATION % increase in _(mM)* Fluoresc2nce Cholic Acid 0.117 o ~.341 S9.1 Methyl Cholate 0.117 25.4 Chenodeoxycholic 0.117 17.7 acid 1.17 80.9 Triton-X 100 4.04 100 1.17 46.4 Q.117 18.6 Deoxycholic Acid 0.117 ` 0 1.17 82.7 1 0.117 o 2 0.117 10 WO~3/1~772 PCT/US92/1077 TAsLE I I
EX CONCENTR~TION % increase in (mM)* Fluorescence 3 2.34 o 4 0.117 0 0.117 57.3 7 0.117 8~.1 8 0.117 89.1 g 0.117 24.5 0.117 0 11 0.117 g8 13 0.117 o * Final concentration after dilution.

ASSAY II: Proton Tr~nsport ~cross Lipid M~mbr~e~
This assay was used to judge the ability of protons to pass across vesicle membranes treated with glycosteroids. Vesicles loaded with carboxyfluorescein at non-self quenching con~entrations were prepared exactly as described above except that the carboxyfluorescein was added to the lipid mixture in 1 mL water (pH ~.5) at a concentration of 1 m~l. After sonication under argon and rotary evaporation to remQve th~ diethyl ether t the carboxyfluorescein-loaded vesicles were purified on a Sephadex-G75 column as described above. The ~oncentration of the vesicle solution after purification on the G-25 column was adjusted until the fluorescent intensity equaled 100 af`ter 100-fold dilution into 80 mM NaClt5 mM Hepes buffer at pH 5.5.
A 100-fold dilution of the vesicle stock into pH 5.5 buffer was made immediately before each experiment and 1 mL of the diluted solution was put in a cuvette. To evaluate the ability of the glycosteroids to facilitate transport of protons CA21 1733~
-~WO93/11772 PCT/US92/10778 across the lipid bilayer, 50 ~L of a 0.245 M solution of each qlycosteroid in methanol was added to the 1 mL
vesicle solution in a fluorescence cuvette and the change in fluorescent intensity was monitored over a period of 10 minutes. A significant decrease in fluorescence indicates that the glycosteroid in question facilitates the transport of protons across the mem~rane. This assay is based on the fact that the fluorescent intensity of carboxyfluorescein is much greater at pH 6.5 than at pH 5.5. If vesicles prepared at pH 6.5 are diluted into a buffer at pH

5.5, the fluorescent intensity will drop over time as the pH gradient across the membrane collapses.
As a control, 50 ~L pure MeOH was added and the fluorescent intensity was found not to change significantly. Addition of MeOH at low concentrations therefore does not make the vesicles permeable to protons. The results are summarized in Table III.

TABLE III
EX Concentration % Decrease in (mM)*_ Fluorescence Triton-X 100 4.04 100 0.0116 2.43 -Gramicidin 0.00579 87.2 0.000~79 81.6 Cholic Acid 0.0116 1.0 Methyl Cholate 0.0116 5.4 Chenodeoxycholic 0.0116 8.2 Acid Deoxycholic Acid 0.0116 5.39 1 0.0116 7.6 0.00579 4.3 2 0.0116 8.6 0.00579 1.7 `r CA21 1 77~~
WOg3/11772 PCT/US~2/1077 TABLE III
3 ~.0116 35.4 0.00579 21~0 4 0.0116 12.3 0~0579 7.89 ~.0116 26.1 0.00579 19~4 7 0.0116 19.8 0.00579 15.2 8 0.~116 32.2 0.00579 20.6 9 0.0116 43~0 0.00579 ~7.4 11 0.0116 22~0 0.00585 14.7 13 0~116 70.6 0.00579 35.2 0O000~79 2.8 * Final concentration after dilution~ ~
A~8AY III: The ~tibiotic ~ficacy of Erythro~ycin With and Without E~hancer~
Erythromycîn is an antibiotic whose efficac~ is known to be increased by compounds that permeabilize cell membranes (Kubesch P. et al. Biochemistry 2~:
2139-2149 (1987)). The efficacy of erythromycin~ in the presence of novel glycosylated steroid deri~atives o~ the present invention/ was evaluated in a plate assay. Briefly, DH2 cells [a mutant strain of E. coli K-12, developed at Cold Spring Harbor Laboratories]
grown in culture broth to an optical density [O.D.] of about 0.5 were mixed with 2.5 mL melted top agar ~Top agar preparation: lO grams tryptone (DIFC0), 5 grams yeast extract (DIFC0), 10 grams NaCl, 7 grams agar ~093/11772 PCT/US92/10778 (DIFCO) and l mL l M NaOH dissolved in one liter of pure water and autoclaved for 25 minutes] and then poured onto agar plates [agar plate preparation: l0 grams tryptone, 5 grams yeast, lO grams NaCl, 15 grams agar, and l mL lM NaOH dissolved in one liter pure water, autoclaved and cooled]. After cooling for 15-30 minutes, each plate was divided into a grid and 4 ~1 of a test solution containing erythromycin [0.5 mM
or l.O mM] in methanol, or erythromycin plus test compound [20 ~M] in methanol, was spotted on each section of the grid. The plates were incubated for sixteen (16) hours at 37 C and then examined for zones of inhibition (i.e., clear areas in sections of the grid where the test solution inhibited bacterial cell growth). Each section of the grid was scored.
The section of the grid containing erythromycin alone at l.O mM concentration was used as a standard for evaluating efficacy, with the other sections scored relative to this. The results, summarized in Table IV
below show that 3~-O-p-methoxybenzoyl-cis-5,l0-bis-~,~-7,12-glucosyl cholic acid methyl ester ~hereinafter referred to as "CME"] is the best enhancer in this assay. Of the non-glycosylated, bile acid derivatives used in this assay, only deoxycholic acid and its sodium salt showed any effect.
Chenodeoxycholic acid and cholic acid and its salts did not have a detectable effect on the antibiotic efficacy of erythromycin in this assay.
Interestingly, deoxycholic acid salts also have been shown to be more effective than chenodeoxycholic acid salts and cholic acid salts in enhancing the uptake of insulin through nasal membranes (Gordon GS et al.
Proc. Nat'l . Acad. sci . USA, 82: 7419-7423 (1985)).

WO 93/11772 PCI/US92/1077"

TABLE :CV
.. ... __ - -- . . - ..
COMPO~ 2 om2s ) ERYT~OMYCIN ~ mM ) EFFECT
. . . I
Cholic Acid 1.0 mM ___ Cholic Acid 0.5 mM ___ Sodium Cholate 1.0 mM ___ Sodium Cholate 0.5 mM ___ Methyl Cholate 1.0 mM ___ Methyl Cholate 0.5 mM ___ Chenodeoxycholic Acid 1.0 mM ___ _ Chenodeoxycholic Acid o.5 mN ___ - .. ~. . _ Deoxycholic Acid 1.0 mM +
Deoxycholic Acid 0.5 mM +
Sodium Deoxycholate 1.0 mM +
Sodium Deoxycholate 0.5 mM ~
CME 1.0 mM +++
CME O.5 mM +++
_ , 3~-0-benzoyl-trans- 1.~ mM +
5,10-bis-~,~-7,12-glucosyl cholic acid methyl ester [BTME]
, , _ BTME O.5 mM +
3~-oH-cis-s ~ lO--bis-a~, ~ 1. 0 mM +
glucosyl cholic acid K~
. _ 3~-OH-cis-5,10-bis-~,~ 0.5 mM +
glucosyl cholic acid X+
_ .
~ erythromycin alone at 1.0 mM (baseline) and all lesser effects ~ : enhancement relative to baseline +++ : significant enhancement relative to baseline The above plate assay was repeated using lower concentrations of CME and comparing its efficacy as an enhancer to that of the non-glycosylated parent, 3~-0-p-methoxybenzoyl-cis-5,10-cholic acid methyl ester [hereinafter referred to as "CDE"]. The results, CA2 1 1 7 33~
- ~VO93/11772 PCT/US92/10778 summarized in Table V below, show that while CME acts as an enhancer at very low concentrations, the non-glycosylated parent compound does not function as an enhancer. This demonstrates that the sugars are critical for enhancing effect.
TABLE V
_ _ _ ~ . , ¦COMPOUND ~mM) ERY~HRONYCIN (mN) EFFECT
¦1.0 mM CDE 0.1 mM ___ I . ..... ~
¦0.1 mM CDE 0.1 mM ___ I . - . ___ ._ ¦O.1 mM CME 0.1 mM ~
I -- . _ ¦O.1 mM CME 0.01 mM
I _ _ -¦0.01 mM CME 0.01 mM + r __ ¦0.001 mM CME 0.01 mM +
l ~
l0.001 mM CME O.001 mM ___ . . ... __~
no detectable clearlng (zone of lnhlbltlon ~ : visible clearing A88AY IV: EFFICACY OF ANTIFUNGAL ABEN$S ON P~OT020A
WITH AND WITHOUT ADDED G~YCOSYLATED 8TEROID

CME, identified in both Assay I described above (compound 8 in the carboxyfluorescein assay) and in Assays II and III described above, as a good membrane permeabilizing agent, was tested for its ability ~o enhance the efficacy of two different antifungal agents on the protozoan Crithidia fasciculata. The ability of the non-glycosylated parent steroid to enhance efficacy was also studied. The studies were carried o~t as described in Pascal RA et al.
Biochemistry 22: 171-178 (1983) and Rahman MD et al.
J. Med. Chem. 31: 1656-1659 (1988). Briefly, flasks containing 25 mL of growth medium ~Preparation: 1.5 grams sucrose, 0.5 grams yeast extract, 0.4 grams tryptone and 0.25 mL triethanolamine dissolved in 100 mL water and pH ad~usted to 8.0 with 10 M HCl.

WO93/11772 PCT/US9~/1077' Autoclave. After cooling, add lO0 ~L hemin (SIGMA) (2 mg hemin/l mL O.l N NaOH) and 20 mg. streptomycin sulfate ~SIGMA)] and the antifungal agent and/or the glycosylated or non-glycosylated steroid derivatives were inoculated with aliquots of C. fasci~ul~ta ( 250 ~L of culture containing approximately l X 106 - l X
107 cells) ~Preparation of culture: C. fasciculata in glycerol added to culture medium and grown, with shaking, for three (3) days at 2 6 C; then stored at 0 to 4 C]. The cultures were incubated, with shaking, at 25 C and growth was monitored by changes in absorbance at 535 nm (relative to the uninoculated medium).
Two different antifungal agents were used in the assays. The first was lO-thiastearic acid (lO-TSA;
see Pahman MD et al. J. Med. Chem. 31: 1656-1~59), which has an IC50 of lO ~M; the second was ~4-thiacholestanol (24-TC; see Rahman MD et al. J. Lipid Research 29: 1543-1548 (1988); Rahman MD and Pascal RA. J. Biol. Chem. 265: 4989-4996 (1990)), which has an IC50 of 0.32 ~M. The results, depicted in Figures l, 2, and 3, demonstrate that the presence of CME
enhances the efficacy of lO~-TSA dramatically, allowing it to be used in lO- to lO0- fold lower concentr~ions than otherwise necessary to achieve 50~ inhibition of growth [Fig. 1]~ The presence of CME also was shown to enhance the efficacy of 24-TC [Fig. 2]. The non-glycosylated parent steroid (CDE~ was not observed to act as an enhancer in this assay [Fig. 3~.
ASSAY V: EFFICACY OF DERIVATIVE-COMPO~ND~CONJ~GATE ON
THE PROTOZOA CRI~IDIA FASCICULAT~
A novel glycosylated steroid derivative of Formula (l) is conjugated to a therapeutically-significant-compound by methods known in the art for coupling an acid group to an amine. The ability of the derivative-compound-conjugate to inhibit the ~093/11772 PCT/US92/10778 growth of Crithidia fasciculata is evaluated as described in Pascal RA et al. Biochemistry 22: 171-178 (1983) and Rahman MD et al. J. Med. Chem. 31: 1656-l65s (1988). Briefly, flasks containing 25 mL growth medium alone, growth medium plus 24-TC at 0.32 ~M
concentration ~the IC50 level), and growth medium plus the derivative-compound-conjugate at 0.32 ~M
concentration are inoculated with aliquots of C.
fasciculata (250 ~L of culture containing approximately 1 X 106 - 1 X 107 cells). The cultures are incubated with shaking at 25 C and growth is monitored by changes in absorbance at 535 nm (relative to the uninoculated medium). Enhanced efficacy of the derivative-compound-conjugate relative to the non-conjugated therapeutically-significant-compound would be reflected in a lower rate of growth (i.e., lower absorbance over time). The ICs~ level of the derivative-compound-conjugate can be measured by repeating the experiments with different concentrations of derivative~compound-conjugate to define the concentration which causes a 50% inhibition of growth relative to the culture containing C.
fasciculata alone.
In another set of experiments, the flasks of growth medium contain derivative-compound-conjugate at its IC50 value, as defined in the above experiments, plus a glycosylated steroid of the present invention, such as CME, which is known to increase the efficacy of 24-TC when not conjugated [hereinaftex referred to as "the enhancer"]. The enhancer is present at the following ratios relative to the derivative-compound-conjugate: O:l, O.l~ l, lO:l, lOO:l, lOOO:l, or any concentration in between. The medium is inoculated with aliquots of C. fasciculata as described above and growth is monitored by ~hanges in C A2 1 1 7 3 3 ?
WO93/11772 P~T/US92/107~^

the absorbance at 535 nm relative to the uninoculated medium. Increased efficacy of the derivative-compound-conjugate in the presence of the enhancer is reflected in a lower rate of growth relative to the derivative-compound-conjugate alone. The optimum ratio of enhancer:derivative-compound-conjugate is defined as that ratio which gives the lowest rate of growth.

Claims (51)

What is claimed is:

1. A compound having the formula:

(I) wherein A is O, OH, OR6, NR7R8, N3, NHCOR7, OCOAr o-?-OR9, OCOR9, NCH2C4H5;

Ar is phenyl or phenyl substituted with 1-3 groups selected from the group consisting of halogen, C1-C12 alkyl or C1-C3 alkoxy;
a is a single bond in the alpha or beta configuration with the proviso that when A=O, a is a double bond;
R1 is H which is cis or trans to R2;
R2 is CH3;
R3 is H, OH or OR6;
R4 is H, OH or OR6;
R5 is CO2R10, CH2OR9, CONH2, CONHR7, CONR7R8, ?-S-R10, CH2S(O)p-S-R10 CH2NH2, CH2NHR7, CH2NR7R8, CH2-S(O)p-S-R10;

R6 is a monosaccharide where the glycosidic linkage at the anomeric carbon atom in said monosaccharide is alpha or beta or is an oligosaccharide of 2-10 monosaccharides where the glycosidic linkage at any of the anomeric carbon atoms in each monosaccharide residue of the oligosaccharide is independently alpha or beta;
R7 and R8, independently are H,C1-C4 alkyl, C3-C7 cycloalkyl, C4-C10 alkylcycloalkyl, phenyl, benzyl, or, taken together are (CH2)f, where f=3-6;
R9 is H or, C1-C3 alkyl;
R10 is H, C1-C10 alkyl, C1-C10 alkenyl, C1-C10 alkynyl, C6H5 or CH2C6H5;
monosaccharide comprises a protected or deprotected hexose or deoxyhexose selected from the group consisting of D-allose, L-allose, D-altrose, L-altrose, D-glucose, L-glucose, D-mannose, L-mannose, D-gulose, L-gulose, D-idose, L-idose, D-galactose, L-galactose, D-talose and L-talose, or a protected or deprotected furanose or deoxyfuranose selected from the group consisting of D-ribose, L-ribose, D-arabinose, L-arabinose, D-xylose, L-xylose, D-lyxose, and L-lyxose, where said protecting groups for the hydroxy groups of said hexoses or furanoses are selected from the group consisting of benzyl, pivaloyl, trimethylsilyl, tert-butyldimethylsilyl, tert-butyldiphenylsilyl, tri-isopropylsilyl, acetyl, tetrahydropyranyl, benzoyl, C1-C3 alkyl, isopropylidene, benzylidene, (2-methoxyethoxy)methyl, orthoester, para-methoxybenzyl and allyl;
p is 0, 1 or 2;
n is 0, 1 or 2;
or a pharmaceutically suitable salt thereof.

2. A compound of claim 1 wherein A is OH, O?OR9, OCOC6H5, OCOC6H5-pOMe, NH2;
a is a single bond;
R3 is OR6;
R4 is OR6;
R5 is CO2R10, CONR7R8;
R6 is a monosaccharide where the glycosidic linkage at the anomeric carbon atom in said monosaccharide is alpha or beta;
R10 is H or C1-C10 alkyl;
monosaccharide is a protected or deprotected hexose such as D-glucose where the protecting groups are benzyl or pivaloyl.

3. The compound of claim 1 which is 3.alpha.-0-benzoyl-trans-5,10-bis-.beta.-.beta.-7,12-glucosyl cholic acid methyl ester.

4. The compound of claim 1 which is 3.alpha.-hydroxy-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid.

5. The compound of claim 1 which is 3.alpha.-hydroxy-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid methyl ester.

6. The compound of claim 1 which is 3.alpha.-hydroxy-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl-25-tryptophanyl cholic acid.

7. The compound of claim 1 which is 3.alpha.-ethylcarbonate-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid methyl ester.

8. The compound of claim 1 which is 3.alpha.-0-benzoyl-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid methyl ester.

9. The compound of claim 1 which is 3.alpha.-0-p-methoxybenzoyl-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid methyl ester.

10. The compound of claim 1 which is 3.alpha.-0-p-methoxybenzoyl-cis-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid.

11. The compound of claim 1 which is 3.alpha.-0-benzoyl-cis-5,10-bis-.beta.,.beta.-7,12-glucosyl cholic acid methyl ester.

12. The compound of claim 1 which is 3.alpha.-hydroxy-cis-5,10-bis-.beta.,.beta.-7,12-glucosyl cholic acid.

13. The compound of claim 1 which is 3.alpha.-0-benzoyl-trans-5,10-bis-.alpha.,.alpha.-7,12-glucosyl cholic acid methyl ester.

14. The compound of claim 1 which is 3.alpha.-hydroxy-trans-5,10-bis-.beta.,.beta.-7,12 glucosyl cholic acid.

15. A derivative-compound-conjugate comprising a compound of claim 1 linked to a therapeutically-significant-compound.

16. A derivative-compound-conjugate comprising a compound of claim 2 linked to a therapeutically-significant-compound.

17. A derivative-compound-conjugate comprising a compound of claim 9 linked to a therapeutically-significant-compound.

18. A derivative-compound-conjugate comprising a compound of claim 10 linked to a therapeutically-significant-compound.

19. The derivative-compound-conjugate of claim 18 wherein the therapeutically-significant-compound is an antifungal agent.

20. The derivative-compound-conjugate of claim 19 wherein the antifungal agent is 24 thiacholestanol.

21. A pharmaceutical composition comprising an effective amount of a compound of claim 1 and a suitable pharmaceutical carrier.

22. A pharmaceutical composition comprising an effective amount of a compound of claim 2 and a suitable pharmaceutical carrier.

23. A pharmaceutical composition comprising an effective amount of a compound of claim 9 and a suitable pharmaceutical carrier.

24. A pharmaceutical composition comprising an effective amount of a compound of claim 10 and a suitable pharmaceutical carrier.

25. A pharmaceutical composition comprising an effective amount of a compound of claim 1, an effective amount of a therapeutically-significant-compound, and a suitable pharmaceutical carrier.

26. A pharmaceutical composition comprising an effective amount of a compound of claim 2, an effective amount of a therapeutically-significant-compound, and a suitable pharmaceutical carrier.

27. A pharmaceutical composition comprising an effective amount of a compound of claim 9, an effective amount of a therapeutically-significant-compound, and a suitable pharmaceutical carrier.

28. A pharmaceutical composition comprising an effective amount of a compound of claim 10, an effective amount of a therapeutically-significant-compound, and a suitable pharmaceutical carrier.

29. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 15 and a suitable pharmaceutical carrier.

WO 93/11772 PCT/US92/107??

30. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 16 and a suitable pharmaceutical carrier.

31. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 17 and a suitable pharmaceutical carrier.

32. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 18 and a suitable pharmaceutical carrier.

33. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 15, a compound of claim 1, and a suitable pharmaceutical carrier.

34. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 16, a compound of claim 1, and a suitable pharmaceutical carrier.

35. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 17, a compound of claim 1, and a suitable pharmaceutical carrier.

36. A pharmaceutical composition comprising an effective amount of a derivative-compound-conjugate of claim 17, a compound of claim 1, and a suitable pharmaceutical carrier.

37. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a compound of claim 1.

38. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a compound of claim 2.

39. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a compound of claim 9.

40. The method of claim 38 wherein the therapeutically-significant-compound is an antibacterial agent.

41. The method of claim 40 wherein the antibacterial agent is erythromycin.

42. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a compound of claim 10.

43. A method of transporting a therapeutically significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 15.

44. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 16.

45. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an-effective amount of a derivative-compound-conjugate of claim 17.

46. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 18.

47. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 15 and an effective amount of a compound of claim 1.

48. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 16 and an effective amount of a compound of claim 1.

49. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 17 and an effective amount of a compound of claim 1.

50. A method of transporting a therapeutically-significant-compound across a biological membrane in a mammal which comprises administering to said mammal an effective amount of a derivative-compound-conjugate of claim 18 and an effective amount of a compound of claim 1.

51. A process for the preparation of a compound of claim 1 which comprises:
(a) reacting a protected glycoside, where the oxygen atoms at all positions of the sugar except the anomeric position are protected with the same or different groups selected from the group including esters and ethers such as alkyl, silyl, phenyl, or benzyl, with (b) an S-R entity under standard conditions where R is C1-C10 alkyl, pyridyl, furyl, thienyl, phenyl substituted with 1-3 groups selected from the group comprising halogen, C1-C3 alkyl, NO2, C1-C3 alkoxy, to yield a protected thio-glycoside which is further reacted with (c) meta-chloroperoxybenzoic acid to yield the corresponding sulfoxide derivative and (d) converted to an activated glycosylating agent intermediate with a triflate-containing compound, such as triflic anhydride, methyl triflate, or trimethylsilyltriflate at -78°C and contacting said activated glycosylating agent with (e) a steroid (in which any oxygens which are not to be glycosylated have been protected by standard methods) in the presence of 2,6-di-tert-butyl-4-methylpyridine in toluene, for formation of .alpha.,.alpha. glycoside linkages, or in propionitrile, for the formation of .beta.,.beta. linkages which is then (f) deprotected following standard procedures to yield glycosylated steroids of the formula (I).