EP1212096A2

EP1212096A2 - Drug conjugates and methods of designing the same

Info

Publication number: EP1212096A2
Application number: EP00959512A
Authority: EP
Inventors: Sydney Brenner; Philip Goelet; Joseph Stackhouse; Steven M. Millward
Original assignee: Individual
Current assignee: Individual
Priority date: 1999-08-26
Filing date: 2000-08-28
Publication date: 2002-06-12
Also published as: CA2382202A1; AU7082400A; JP2003507439A; WO2001013958A2; WO2001013958A3

Abstract

The invention relates to drug conjugates and methods of their design. One embodiment of the invention is directed to a method of designing vector-linker-pharmacophore ('VLP') conjugates that is generally applicable to a wide variety of vectors, linkers, and pharmacophores. The invention also encompasses a method of improving the delivery of a pharmacophore to a patient, as well as a method of improving the therapeutic efficacy of a pharmacophore and a method of decreasing the toxicity of a pharmacophore. A method of increasing the concentration of a pharmacophore in a cell is further encompassed by the invention.

Description

DRUG CONJUGATES AND METHODS OF DESIGNING THE SAME

This application claims the benefit of United States provisional application nos. 60/150,765 and 60/150,894, both filed August 26, 1999, and United States provisional application nos. 60/184,411 and 60/184,412, both filed February 23, 2000, all of which are incoφorated herein by reference.

1. FIELD OF THE INVENTION

The invention relates to methods for designing drug conjugates, methods of improving the delivery of drugs, methods of improving the therapeutic indices of drugs, methods of decreasing the toxicities of drugs, and drug conjugates and compositions comprising the conjugates.

2. BACKGROUND OF THE INVENTION

The therapeutic effectiveness, or therapeutic index, of a drug can depend on a variety of factors which include its ability to reach its intended site of action. For example, a drug may provide a therapeutic effect when delivered to a specific site within the body of a patient, but cause unpleasant or harmful side-effects when delivered to other sites. The preferred site of action of a drug is referred to as its target (e.g., a tissue, organ, cell, receptor, enzyme, or endogenous signaling molecule). Because the accurate delivery of a drug to its target may increase its efficacy and reduce undesired side effects, targeted drug delivery is one method by which the safety and efficacy of a drug can be improved. Attempts at targeted drug delivery include the use of different routes of administration (e.g., topical instead of oral delivery). It has further been attempted by attaching pharmacologically active compounds, or "pharmacophores," to moieties that have an affinity for the organs or tissues to which the pharmacophores are preferably delivered. Examples of targeted pharmacophores, which are sometimes referred to as "drug conjugates," may be found in U.S. patent nos.: 5,466,681; 5,502,037; 5,723,589; 5,739,287; 5,827,819; and 5,869,465.

Until now, targeted pharmacophores have been difficult to design, in part because the scientific community lacked a systematic and rational method for their design. The design of small targeted pharmacophores (e.g., those with molecular weights of less than about 2000 daltons) has been particularly difficult, since the moieties that are attached to the pharmacophores, and which presumably increase the specificity with which the pharmacophores are delivered, can themselves have undesirable pharmacological effects. A need therefore exists for a predictable methodology that can be used to facilitate the design o f targeted pharmacophores .

3. SUMMARY OF THE INVENTION

This invention is directed, in part, to a method of designing compounds which are referred to herein as "vector-linker-pharmacophore" or "VLP" conjugates, and which comprise three components: a "vector," a "linker," and a "pharmacophore."

As discussed in detail herein, each VLP conjugate comprises at least one pharmacophore, which exerts a pharmacological effect by interacting with a "pharmacophore target" (e.g., a receptor or an enzyme). Each VLP conjugate further comprises at least one vector, which has an affinity for a "vector target," which is different from the pharmacophore target but which is either located in close proximity to the pharmacophore target, or can easily travel (e.g., via the bloodstream or by diffusion) to it. The vector(s) and pharmacophore(s) of each conjugate are connected by covalent attachment to at least one linker.

In addition to methods of designing VLP conjugates, the invention encompasses a method of improving the delivery of a pharmacophore to a patient, a method of improving the therapeutic efficacy of a pharmacophore, and a method of decreasing the toxicity of a pharmacophore. Further encompassed by the invention is a method of increasing the concentration of a pharmacophore in or on a cell.

The methods disclosed herein are facilitated by the use of computers and automated screening devices. The invention therefore encompasses an electronic device for the design of VLP conjugates. The invention also encompasses methods of managing and manipulating data or information concerning vectors, linkers, and pharmacophores such as approved drugs, drugs in clinical development, and drugs that failed during clinical development. Such methods encompass the use of existing software to search, mix, and match databases of existing data and information. The invention further encompasses the generation of databases or packages of information useful for the design of VLP conjugates. The invention also encompasses VLP conjugates. Preferred VLP conjugates of the invention have molecular weights of less than about 2000, more preferably less than about 1500, and most preferably less than about 1000 daltons.

3.1. DEFINITIONS

As used herein, and unless otherwise indicated, the term "small" when used to describe a molecule (e.g., a VLP conjugate) means that the molecule has a molecular weight of less than about 2000, preferably less than about 1500, and most preferably less than about 1000 daltons. As used herein, the term "therapeutic index" refers to the ratio of the concentration at which a pharmacophore exerts an undesired effect to the concentration at which it exerts a desired effect. A higher therapeutic index provides a greater margin of safety than a lower therapeutic index, and is therefore desirable.

As used herein, the term "toxicity index" of a compound is the concentration at which an undesired effect of the compound is, or is estimated to be, no longer tolerable by a typical patient.

As used herein, the abbreviation "MIC₅₀" refers to the minimum concentration of a compound that will kill or inhibit the growth of 50 percent of test cultures of a microbial strain. As used herein, the abbreviation "IC₅₀" refers to the concentration of an antimicrobial compound that will displace 50 percent of labeled compound bound to a receptor. The abbreviation also encompasses the concentration of an antimicrobial compound that will inhibit the binding of 50 percent of a labeled compound that binds to the same receptor. As used herein, the term "patient" means a plant or animal suffering or likely to suffer from a disease or condition. Examples of animals include, but are not limited to, vertebrates such as mammals (e.g., humans), reptiles, birds, and fish.

Unless otherwise indicated, the term "target" as used herein refers to a biological entity upon which a compound or chemical moiety (e.g., a pharmacophore) acts or to which a compound (e.g., a vector) has an affinity. Examples of targets include, but not limited to, organs, tissues, cells, cellular parts (e.g., ribosomes, surface receptors, and DNA), proteins, peptides, polypeptides, and nucleotides. Unless otherwise indicated, the term "affect," when used herein to describe the interaction between a compound or chemical moiety and a target, means to alter the behavior, physical properties, or structure of the target. For example, pharmacophores typically affect their targets by associating (e.g. , ligand-receptor binding) with them. Unless otherwise indicated, the term "associate," as used herein to describe the interaction between a compound or chemical moiety and a target, means to bind to the target. Examples of binding include, but are not limited to, covalent bonding, hydrogen bonding (e.g., hybridization), hydrophilic/hydrophobic interactions, ligand-receptor bonding, and Van der Waals interactions. Unless otherwise indicated, the term "affinity for," as used herein to describe the characteristics of a molecule or chemical moiety with regard to target, means that the molecule or chemical moiety will associate with the target when the two are contacted. Such association can be measured by techniques known in the art and expressed, for example, with a rate constant or binding constant. Typically, a molecule or chemical moiety that has an affinity for a target will bind to the target (e.g. , by ligand-receptor or enzyme-substrate interaction) with a dissociation constant of less than about 10^"6, more preferably less than about 10^"8, and most preferably less than about 10^"10 molar. It is to be understood, however, that the affinity of a molecule or chemical moiety for a given target may be lowered because of factors present in the target's natural environment such as, but not limited to, diffusion within a cell or across cell membranes, such that in vivo affinity is less than that which would be measured if the molecule or chemical moiety were contacted with a target removed from its natural environment.

Unless otherwise indicated, the term "associated," when used to describe the relationship between a biological entity (e.g., a receptor, protein, or enzyme) with a disease or condition, means that the inhibition, destruction, modification, production, or accumulation of the biological entity causes or aggravates the disease or condition, or causes or aggravates a symptom thereof.

Unless otherwise indicated, the phrase "likely to be located near," as used herein to describe the relationship between a first target and a second target, means that there is a probability of at least about 50, 60, 70, 80, 90, 95, or 99 percent that the first target is located within a distance of less than about 10^"4, 10^"5, 10^"6, or 10^"7 meters from the second target. If a first and/or second target are mobile within a patient or biological system, the phrase means that there is a probability of at least about 50, 60, 70, 80, 90, 95, or 99 percent that the first target will pass within a distance of less than about 10^'4, 10^"5, 10^"6, or 10^"7 meters from the second target.

3.2. BRIEF DESCRIPTION OF THE DRAWINGS

Aspects and advantages of the invention are further apparent from the following drawings, in which:

FIG. 1 is a schematic diagram of a VLP conjugate;

FIG. 2 is a schematic diagram of the mode of action of a VLP conjugate; FIG. 3 is a table of fungal pharmacophores;

FIG. 4 is a table of fungal vectors;

FIG. 5 is a table of fungal linkers;

FIG. 6 is a flow chart of the process by which a VLP conjugate may be designed;

FIG. 7 provides the structures of some fluorescent probes; FIG. 8 is a table of antifungal ligands;

FIG. 9 provides the chemical structure of sordarin;

FIG. 10 provides the chemical structure of fluconazole;

FIG. 11 is a representation of a sordarin-linker-fluconazole conjugate;

FIG. 12 provides a synthetic scheme for the preparation of a sordarin-linker- fluconazole conjugate;

FIGS. 13a-i provide examples of tables used in the design of VLP conjugates with antibiotic properties;

FIG. 14 is a diagram of bio tin-penicillin VLP conjugates;

FIG. 15 provides structures of conjugates of tetracycline and trimethoprim with a mechansim-based inhibitor of β-galactosidase;

FIG. 16 provides the chemical structure of a kirromycin-trimethoprim conjugate;

FIG. 17 provides the chemical structure of a kirromycin-tetracycline conjugate;

FIGS. 18-20 provide a synthetic scheme for the preparation of biotin-penicillin conjugates; FIG. 21 shows the expected inhibition of microbial growth caused by an antibiotic

VLP conjugate as compared to inhibition due only to the unconjugated antibiotic pharmacophore; FIG. 22 provide a synthetic scheme for the preparation of modified trimethoprim pharmacophores ;

FIG. 23-25 provide a synthetic scheme for the preparation of modified tetracycline pharmacophores ; FIG. 26 provides a synthetic scheme for the preparation of a mechanism-based inhibitor of β-galactosidase, which can be used as a vector;

FIG. 27-38 provide a synthetic scheme for the preparation of kirromycin derivatives useful as vectors;

FIG. 39 provides a comparison of kirromycin fragments; and FIG. 40 provides a description of relationships that can in the preparation of databases and the design of VLP conjugates.

4. DETAILED DESCRIPTION OF THE INVENTION

The invention encompasses a rational and systematic method that can be used to lower the cost and decrease the time associated with the development of vector-linker- pharmacophore ("VLP") conjugates. A representation of a VLP conjugate is shown in FIG. 1. The invention also encompasses methods of improving the delivery, increasing the therapeutic efficacy, and decreasing the toxicity of a class of pharmacophores, a pharmacophore, or a drug delivery molecule. VLP conjugates are also encompassed by the invention that can be used to treat and/or prevent diseases and conditions in plants and animals, including vertebrates such as mammals (e.g., humans), reptiles, birds, and fish. The invention is based, at least in part, on the use of a vector that has an affinity for a target that is different from the target of a given pharmacophore, and which can exist in a concentration and/or location sufficient to concentrate the VLP conjugate near the pharmacophore target. It is believed that the use of such vector/target interactions in the targeted delivery of pharmacophores is novel, and has not been previously reported.

A first embodiment of the invention encompasses a method of designing a VLP conjugate for use in the treatment or prevention of a disease or condition in a patient, which comprises: selecting a pharmacophore that can affect a first target associated with the disease or condition, and that has a first affinity for the first target; selecting a vector that has a second affinity for a second target likely to be located near the first target; and selecting a linker to which the first pharmacophore and the first vector can both be covalently attached to provide the VLP conjugate; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in the patient is greater than that of the first target.

A preferred method of this embodiment further comprises covalently attaching the pharmacophore and vector to the linker to provide a conjugate, and testing the ability of the conjugate to affect the first target.

In another preferred method of this embodiment, the pharmacophore is selected using at least one criterion selected from the group which includes, but is not limited to: mode of action; target of action (e.g., location in the body, cellular or intracellular location, concentration, and molecules that react with it); molecular weight; solubility; types and/or severities of adverse effects; therapeutic index; chemical stability; presence of chemically reactive (and preferably modifiable) moieties; and structure-activity relationship data.

This method can be used to design a plurality of VLP conjugates, which can then be ranked by at least one pharmacological or chemical characteristic. Examples of such characteristics include, but are not limited to: affinity for the first target (i.e., the pharmacophore target); affinity for the second target (i.e., the vector target); mechanism- directed inhibition of a specific enzyme; modification of a specific enzyme; ability to inhibit or modify enzyme production, DNA or RNA synthesis, or signal transduction; chemical stability (e.g., ability to withstand cleavage under certain conditions); physiological concentration of the first target; physiological concentration of the second target; enzyme kinetic constants of the first and/or second targets; diffusion characteristics (e.g., ability to enter particular types of cells); solubility; estimated systemic concentration when administered to a patient under specific conditions; resistance to metabolic degradation; and estimated systemic clearance and metabolism when administered to a patient under specific conditions.

A second embodiment of the invention encompasses a method of designing a VLP conjugate of a pharmacophore having a first target, and a first affinity for the first target, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; and selecting a linker to which the first pharmacophore and the first vector can both be covalently attached to provide the VLP conjugate; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in a patient to whom the VLP conjugate could be administered is greater than that of the first target.

A third embodiment of the invention encompasses a method of improving the delivery of a pharmacophore to a first target located in or on a cell, wherein the pharmacophore has a first affinity for the first target, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; selecting a linker; covalently binding the pharmacophore and the vector to the linker to provide a conjugate; testing the ability of the conjugate to concentrate near the first target; and repeating the process with a different vector if the ability of the conjugate to concentrate near the first target is less than the ability of the pharmacophore alone to concentrate near the first target; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in or on the cell is greater than that of the first target.

In a preferred method of this embodiment, information provided by the testing is used in the selection of a different vector if the process is repeated.

A related method encompassed by this embodiment is a method of targeting a pharmacophore having an affinity for a first target in vivo, which comprises: chemically linking the pharmacophore and a vector to a linker to provide a VLP conjugate; and administering the VLP conjugate to a host; wherein the vector can associate with a second target with a dissociation constant of less than about 10^"6, the second target is different from the first target, and the second target is located within 10^"4 meters of the first target. In a preferred method of this embodiment, the second target is present in the host in a concentration of greater than 10 times that of the first target.

A fourth embodiment of the invention encompasses a method of improving the therapeutic index of a pharmacophore having a first target, a first affinity for the first target, and a first therapeutic index, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; selecting a linker; covalently binding the pharmacophore and the vector to the linker to provide a conjugate which has a second therapeutic index; testing the conjugate to determine the second therapeutic index; and repeating the process if the second therapeutic index is less than the first therapeutic index; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in a patient to whom the VLP conjugate could be administered is greater than that of the first target.

In a specific method encompassed by this embodiment, the pharmacophore has a poor first affinity and an acceptable toxicity index. Another specific method encompassed by this embodiment is a method of decreasing the toxicity of a pharmacophore having a first affinity for a first target and a first toxicity index, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; selecting a linker; covalently binding the pharmacophore and the vector to the linker to provide a conjugate which has a second toxicity index; testing the conjugate to determine the second toxicity index; and repeating the process if the second toxicity index is greater than the first toxicity index; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in a patient to whom the VLP conjugate could be administered is greater than that of the first target. A fifth embodiment of the invention encompasses a method for treating or preventing a disease which comprises the systemic (e.g., oral or parenteral) or local (e.g., topical) administration of a VLP conjugate to a patient in need of such treatment or prevention, which comprises: a pharmacophore that has a first affinity for a target associated with the disease; a vector that has a second affinity for a second target likely to be located near the first target in the patient; and a linker covalently linking the pharmacophore and the vector; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in the patient is greater than that of the first target. In a preferred method of this embodiment, the disease is fungal or bacterial infection. Each embodiment of the invention that is directed to a method encompasses a number of preferred methods. In preferred methods of the invention, the linker is selected using at least one criterion selected from the group which includes, but is not limited to: chemical stability under physiological conditions (e.g., those typically surrounding the first and second targets); types and number of reactive moieties; metabolic stability; solubility (e.g., hydrophobicity); length; and flexibility.

In other preferred methods of the invention, each vector is selected using at least one criterion selected from the group which includes, but is not limited to: pharmacological effects; and lack of affinity for the first target. Preferred vectors bind to their targets with high affinity, but bind to the pharmacophore target with low affinity. A preferred vector binds to its target with a binding constant that is greater than about 10, 100, or 1000 times the binding constant that describes the vector's affinity for the pharmacophore target.

In additional preferred methods, the vector and linker are selected using information obtained by screening a plurality of vector- linker conjugates for their affinities for the second target. Examples of such vector- linker conjugates include radioactively-labeled vector-linker conjugates and vector-linker conjugates attached to probe molecules (e.g., fluorescent probes). Vector-linker conjugates can also be screened prior to the selection of their components using combinatorial chemistry techniques such as those disclosed herein.

Similarly, in methods wherein a pharmacophore and linker are selected, both can be selected using information obtained by screening a plurality of pharmacophore-linker conjugates for their affinities for the first target.

In other preferred methods of the invention, the VLP conjugate has a molecular weight of less than about 2000, more preferably less than about 1500, and most preferably less than about 1000 daltons.

In other preferred methods of the invention, the second target (i.e., vector target) is located in a cell containing the first target (i.e., pharmacophore target), and the concentration of the second target in that cell is greater than about 10^"5, more preferably greater than about 10^~4, and most preferably greater than about 10^"3 molar.

A sixth embodiment of the invention encompasses an electronic device for the design of VLP conjugates which comprises: a searchable database containing information about the pharmacological, chemical, and/or physical properties of vectors, linkers, and pharmacophores; an input device configured to receive instructions from a user; an output device configured to display at least one string of text data; memory configured to store the database; and a processor configured to rank data entries in the database (e.g., rank a plurality of pharmacophores, a plurality of linkers, and/or a plurality of vectors), and to identify a pharmacophore, a linker, and a vector that can be bound together to form a VLP conjugate. A seventh embodiment of the invention encompasses a VLP conjugate that can be used in the treatment or prevention of a disease or condition in a patient, which comprises: a pharmacophore moiety having an affinity for a first target; a linker moiety; and a vector moiety having an affinity for one or more second targets; wherein the pharmacophore and vector moieties are covalently attached to the linker, each second target is likely to be located near the first target in a typical patient suffering or likely to suffer from the disease or condition, and the first target is not the same as any of the second targets.

Preferred VLP conjugates have a molecular weight of less than about 2000, more preferably less than about 1500, and most preferably less than about 1000 daltons. In a specific embodiment, the second target is not associated with the disease or condition to be treated, but the first target is so associated. However, both first and second targets can be associated with a disease or condition if they otherwise meet the criteria described herein.

In other preferred VLP conjugates, the covalent bonds attaching the vector to the linker and the linker to the pharmacophore do not cleave under the physiological conditions surrounding the first target.

Specific VLP conjugates of the invention comprise vector moieties derived from one of the following molecules: sordarin; biotin; and kirromycin. Specific VLP conjugates of the invention comprise pharmacophore moieties derived from one of the following molecules: fluconazole; penicillin; trimethoprim; and tetracycline. In a more specific VLP conjugate, the vector is sordarin and the pharmacophore is an antifungal preferably of the conazole class. Even more specific conjugates of the invention include, but are not limited to: sordarin-linker-fluconazole conjugates; biotin-linker-penicillin conjugates; kirromycin-linker-trimethoprim conjugates; and kirromycin-linker- tetracyline conjugates. 4.1. METHODS OF DESIGNING CONJUGATES

Unlike methods used in the past, the methods of this invention are based on a unique systematic and rational approach to the design of VLP conjugates and drug targeting. Advantageously, methods of the invention are not limited to specific types of vectors, linkers, or pharmacophores, and can be used to design VLP conjugates useful in the treatment or prevention of a wide variety of diseases and conditions. Methods of the invention can further be used to develop conjugates of small pharmacophores and small vectors.

The methods of this invention, and the novel properties of the compositions disclosed herein, are based on the ability of some chemical moieties, referred to herein as "vectors," to interact with target macromolecules in such a way that pharmacophores attached to the vectors and administered to a patient will be concentrated near those macromolecules. As shown in FIG. 2, this can improve the delivery of pharmacophores to their sites of action -or targets- if those sites of action are proximate to the vector targets.

In a general method of the invention, information relevant to a plurality of potential vectors, pharmacophores, and linkers is assembled in what is referred to herein as a "database" or "table". The term "table" can be used to refer to a part of a database. The database can be stored in the memory of a computer, on magnetic media (e.g., floppy disks, hard disks, and tapes), on optical media (e.g., CD-ROMS), or even on paper. As used herein, the terms "database" and "table" therefore refer to a collection of information which is not limited to a particular method of storage, manipulation, or organization. However, a preferred database contains data that is divisible into three subsets, each of which is related to pharmacophores, linkers, or vectors. Preferably, at least one, and more preferably at least two of the three subsets of data contain information for a plurality of compounds or chemical moieties. For example, a database can contain one subset of information concerning physical properties, chemical properties, pharmacological (including pharmacokinetic, metabolic, toxicity, potency, receptor binding, and enzyme inhibitory) properties, and/or physiological properties of one pharmacophore, another subset containing similar information for a plurality of linkers (e.g., at least about 5, 10, 20, 50, 100, or 500), and a third subset containing similar information for a plurality of vectors (e.g., at least about 5, 10, 20, 50, 100, or 500). Using algorithms and criteria disclosed herein, the information within the database is organized or sorted such that one or more combinations of pharmacophore, linker, and vector are obtained. Each of these combinations represents the components of one VLP conjugate, the estimated physical, chemical, and/or physiological properties of which are preferably improved with regard to the pharmacophore(s) alone, and which can be used to rank potential VLP conjugates according to a particular preference.

The information that is assembled into a database is preferably obtained prior to application of sorting algorithms. However, as VLP conjugates designed by methods of the invention are tested, the information those tests provide can be used to augment and/or modify the database in such a way that information obtained from it can be used to predict the properties of potential VLP conjugates with more accuracy.

Advantageously, databases already in existence can be used in methods of invention. Examples of commercially available databases include, but are not limited to: the Merck Index^®; the Merck Manual^®; the Physicians' Desk Reference^®; online service databases such as those available from Lexis^®, Westlaw^®, Dialog^® (e.g., Medline^®), and Chemical Abstracts Service^®; and governmental registries and databases such as those maintained by or on the behalf of the Food and Drug Administration, the United States Patent and Trademark Office, the Japanese Patent and Trademark Office; the European Patent and Trademark Office; and the United States Pharmacopea. However, these and other sources of information known to those skilled in the art are preferably used to assemble a modifiable database of the invention.

As described in more detail herein, a general database that can be used to design VLP conjugates contains information about a variety of pharmacophores, including their ability to treat certain diseases and their sites and mechanisms of action. The database further contains information about a variety of vectors, including their targets (e.g., cells, parts of cells, or proteins to which they have a particular affinity) and pharmacological effects, as well as information about a variety of linkers, such as their chemical stabilities, solubilities, and lengths. Such information is used in the methods described herein to select a particular combination of pharmacophore, linker, and vector that when covalently attached may provide a VLP conjugate with novel and improved properties.

More particular databases can be generated and analyzed according to the methods disclosed herein. For example, a database may contain information about only one pharmacophore, and the method used to analyze it may focus only on ways of more accurately targeting its delivery. Alternatively, a database and method can focus on VLP conjugates of a particular vector, given the concentration and location of its target in a particular group of patients (e.g., humans suffering from a particular disease or condition).

4.1.1. DATABASE CONSTRUCTION

A database of the invention comprises information about at least one pharmacophore, at least one vector, and at least one linker. Examples of such information are provided below.

4.1.1.1. Pharmacophores and Their Targets

A pharmacophore is a molecule or chemical moiety having an affinity for a particular target and thus active against a biological condition or a disease state. Pharmacophores include, but are not limited to, pharmaceutical compounds recognized by the medical and scientific communities as safe and effective (e.g., compounds disclosed in the Physicians' Desk Reference^®), as well as compounds that have adverse effects that outweigh their therapeutic effects. Examples of such compounds include those that have been rejected as drug candidates during clinical trials and those that have few or no adverse effects but which interact poorly with their targets.

Examples of pharmacophores that can be including in a database of the invention include, but are not limited to, antibiotics, anti-tumor agents, angiogenesis inhibitors, antivirals, antifungals, molecules that bind tightly to metabolites, inhibitors or activators that bind to macromolecular receptors, neural receptor agonists or antagonists, transcription factors inhibitors or activators, enzyme inhibitors or activators, inhibitors or activators of binding interactions between or catalytic activities of proteins in cell signaling pathways; protein synthesis inhibitors, ionophores, antigens, and molecules that modify DNA or RNA.

Specific pharmacophores include, but are not limited to, penicillin, tetracycline, trimethoprim, and those listed in FIG. 3. Additional pharmacophores are disclosed by U.S. patent nos. 4,873,088; 5,190,969; 5,466,681; 5,795,909, each of which is incoφorated herein by reference. Potential pharmacophores further include: cytotoxic phenolic compounds, such as those disclosed by U.S. patent no. 5,639,737, which is incoφorated herein by reference; antibodies, such as those disclosed in U.S. patent no. 5,824,805, which is incoφorated herein by reference; psychotropic, neurotropic, and neurological agents, such as those disclosed by U.S. patent no. 5,827,819, which is incoφorated herein by reference; bradykinin analogs, such as those disclosed by U.S. patent no. 5,863,899, which is incoφorated herein by reference; lysomotropic moieties, intracellular polymerizing moieties, protein sorting signals or sequences, conditional membrane binding peptides and bi- or multi-valent receptor cross linking moieties, such as those disclosed by U.S. patent no. 5,869,465, which is incoφorated herein by reference; and chimeric fused proteins, such as those disclosed by U.S. patent no. 5,871,753, which is incoφorated herein by reference.

In a database of the invention, information about each pharmacophore includes information about its preferred site(s) of action (i.e., its target(s)). Examples of such information include, but are not limited to, the location, concentration, and structure of the target, other drugs that interact with it, macromolecules that interact with it, and other members its family (e.g., receptors of similar structure or function).

Because the role played by a pharmacophore target in the severity, onset, or symptoms of a disease or condition is typically known, pharmacophores can be correlated to disease states. Disease states can be classified as those caused by microorganisms or as those associated with mammalian cells or quality of life states. Classes of microorganisms include prokaryotes, single cell eukaryotes, multicellular eukaryotes, and viruses. Disease conditions of mammalian cells, or quality of life states, can be classified by cell types and locations. Thus, a database used in a method of the invention may contain information regarding the organism that causes the disease of interest, or the types and locations of the cells associated with the disease.

Pharmacophore information stored in a database can further include: the molecular structure of the pharmacophore; the minimum concentration at which it is typically effective; structure-activity relationship data; the MIC₅₀ of the pharmacophore; binding and/or kinetic constants for the pharmacophore's association with its target; the mechanism or mode of the pharmacophore's action; its therapeutic effects; and its adverse effects. Structure-activity data can, for example, include the structures of derivatives of the pharmacophore, together with the minimum concentration at which each of those derivatives exhibits a desired or undesired effect (e.g., the therapeutic index of each derivative), and the binding and/or kinetic constant of each for its target. Structural modifications that consisted of addition of long chain substituents to some part of the pharmacophore may also be of interest, as they can provide information relevant to the selection of a linker, as described elsewhere herein.

A table of pharmacophore targets can be prepared that corresponds to the pharmacophores chosen to combat a disease state of interest. Pharmacophore targets can be classified as, for example: receptors for, or enzymes that process, known pharmacophores; as undesired or unintended targets of known pharmacophores; or as macromolecules or structures that have a function which, if modified, can cause a pharmacological effect. Information about each target can include: its cellular or extracellular abundance; its cellular or extracellular location; the location of the cell types that express it; the location of the target within a particular organism; and the cellular or physiological function of the target. Additional information for targets that are proteins can include their amino acid sequences and 3-dimensional structures. A database may also include the identities of small and large molecules that associate with each target in its native environment.

Pharmacophore targets can thus be characterized in any number of ways, and can be indexed, classified, or arranged by any of those characteristics. Examples of ways by which a pharmacophore target can be classified or identified include, but are not limited to: a. whether it is a target for known pharmacophores or compounds which, if modified, can have a pharmacological effect (e.g., prodrugs); b. pharmacophores with which it interacts; c. its cellular or physiological function(s); d. its cellular or extracellular location; e. the location of the cell types that express it; f. if the target is a protein, its molecular properties, which in turn can be determined or characterized by: i. its amino acid sequence; ii. its 3-dimensional structure; iii. the identities of small molecules that bind to it, or which are substrates for it, in its native environment; and/or iv. the identities of large molecules (e.g., proteins) that bind to it, or which are substrates for it, in its native environment; g. the identities of other small molecules that bind to it or are substrates for it; and h. the molecular receptors that are localized in the vicinity of it and vectors for such receptors.

4.1.1.2. Vectors and Their Targets As discussed elsewhere herein, vectors are compounds or moieties which have an affinity for targets ("vector targets") located in the vicinity of pharmacophore targets. Preferably, the interaction between a vector and its target will cause few physiological effects, or will cause effects that augment the therapeutic effects of the pharmacophore. There are many potential vector targets within, or in the vicinity of, bacteria, fungi, mammalian cells, and virally infected cells. In order for a VLP conjugate to function desirably, the target of the vector it comprises must either be present at a sufficiently high concentration or associate with the vector with a sufficiently high binding or good kinetic constants such that the increase in the local concentration of the pharmacophore due to the vector/vector target interaction is sufficient to improve the activity of the pharmacophore.

Vector targets can be abundant proteins (preferably comprising more than one percent of the mass of a target cell), with binding sites that have very low dissociation constants for the vector. They can also be enzymes that process the vector as a substrate. Alternatively, vector targets can be cellular systems that increase the local concentration of the conjugate by other means. Whatever the vector target, it must be located in close proximity to the cellular or molecular target upon which the pharmacophore preferably acts (i.e., the pharmacophore target).

Other vector targets include, but are not limited to, macromolecular receptors for small molecules, enzymes for which mechanism-based inhibitors are known, polymerases, and synthetases. Proteins that are uniquely found in specific cells (e.g., diseased or infected cells) are also potential vector targets. Vector targets further include specific and abundant protein, DNA, RNA, or polysaccharide products of a given pathogen, such as structural proteins of a pathogen, or abundant enzymes of a pathogen or of a diseased cell or tissue, or protein products of a disease. They also include, but are not limited to: bacterial ribosomes, ribosomal RNAs, tRNAs, and ribosomal proteins; peptidyl prolyl cis-trans isomerases or rotamases; structural proteins of viruses, such as the nucleoproteins, and coat proteins; abundant proteins of specific eukaryotic cells, such as steroid hormone receptors and the receptors for retinols and vitamin D; and cleavage enzymes. Potential vector targets also include antibiotic resistance elements, including those which have genes carried on plasmids such as the aminoglycoside phosphokinases and acetylases, bacterial beta lactamases, and bacterial transamidases. Vectors are selected to achieve preferential targeting or concentration at vector targets. A suitable vector may bind with high affinity to a protein receptor or be a substrate for an enzyme, for example a polymerase. Such vector targets are associated with pathogens or have a distinct cellular distribution in organisms suffering from certain diseases. Vectors can also be small molecules that bind with high affinity to particular macromolecules. New small molecules with the sole characteristic that they bind with high affinity to some macromolecular target or are the substrate for some enzyme may also be used as vectors. Similarly, small molecules that are known to accumulate in, on, or in the vicinity of cells or intercellular or intracellular structures, by other means may be used as the vectors in VLP conjugates. Specific vectors that can be included in a database of the invention are listed in

FIG. 4. Other vectors are disclosed by U.S. patent nos. 5,466,681; 5,639,737; 5,723,859; 5,739,287; 5,824,805; 5,827,819; 5,837,690; 5,863,899; 5,869,465; and 5,871,753, each of which is incoφorated herein by reference.

Methods and conjugates of the invention are premised on the interactions between small molecules and target macromolecules. A typical database of the invention will thus contain information about ligands and their receptors or substrates and their corresponding enzymes. Because many small molecules that interact with macromolecules have been identified on the basis of their pharmacological activity, most initial entries in a database will be pharmacophores and their receptors. However, the data used to provide a database of the invention need not be limited to intermolecular interactions that lead to a pharmacological effect; those that do not can simply be characterized as vectors and their targets. A table of ligands and receptors or substrates and enzymes which can be used as vector information within a database can include information such as, but are not limited to: the identity of the macromolecule to which a small molecule binds or the enzyme for which it is a substrate; the cellular location and concentration of such a macromolecule; the molecular structure of the small molecule; structure activity relationship data regarding the interaction between the small molecule and its target receptor or enzyme; and reaction catalyzed by the enzyme. Vector information can also include the pharmacological effects of each potential vector, its pharmacologically effective concentration, its mode of action, and its side effects. Of particular interest is structure activity data that can be used to determine what effects, if any, the attachment of a linker to a vector will have on its affinity for its target.

Vectors and their targets can thus be characterized in any number of ways, including, but not limited to: a. the identity of the primary target (e.g., protein receptor, enzyme, or another macromolecule) with which the vector interacts; b. the position of the target (e.g. , in relation to a particular pharmacophore target); c. molecular properties such as, but not limited to, structure and molecular weight; d. the pharmacological effects of each vector or its metabolites; e. the binding constant of each vector for its target (e.g., kinetic constants with regard to a target enzyme); and fi structure-activity relationship data.

A table of vector targets associated with each target microorganism or cell type can be classified, for example, according to their abundance and/or subcellular or extracellular location.

If vector targets are proteins, a typical table of a database of the invention comprises at least one of the following pieces of information for each target: a. its cellular or extracellular abundance; b. its cellular or extracellular location; c. in the case of multicellular organisms, the location within the organism; d. its cellular or physiological function; e. its molecular properties, which can be characterized or determined using information such as, but not limited to: i. its amino acid sequence; ii. its 3-dimensional structure; iii. small molecules that bind to it in its native environment, and their binding constants; and iv. large molecules (e.g., macromolecules) that bind to it in its native environment; and f. the identities of other small molecules that bind to it.

If vector targets are enzymes, a typical table of a database of the invention comprises at least one of the following pieces of information for each target: a. reaction(s) catalyzed by the enzyme; b. its cellular or extracellular abundance and location; c. in the case of multicellular organisms, its location within the organism; d. the cellular or physiological function of the enzyme; e. its molecular properties, which can be characterized or determined using information such as, but not limited to: i. its amino acid sequence; ii. its 3-dimensional structure; iii. the small molecules that are substrates for, or inhibitors of, the enzyme in its native environment, and their binding or kinetic constants; and iv. large molecules (e.g., macromolecules) that bind or interact with it in its native environment; and f. other small molecules that are known to interact with the enzyme.

From such information, a table can be formed that contains information about vectors known to bind to the vector targets of interest. In a preferred database of the invention, the pharmacological effects -if any- exerted by the vectors when bound to their targets are further included. Consequently, the vector information stored in a database can include data such as, but not limited to: the identity of each vector target (e.g., receptor or enzyme); its function; its structure; its molecular weight; its binding or kinetic constants with regard to its target; structure activity relationship data; and its pharmacological effects. A preferred database of the invention contains a table of vectors that encompasses at least one piece of the following information: a. small molecule(s) (i.e., vectors) known to interact with vector targets. For targets associated with microorganisms, many of these will be known antibiotics. b. the function of each vector i. if the vector is a molecule that binds to a receptor protein, it should be noted whether the receptor is activated or inhibited by the action of the vector; and ii. if the vector is a molecule that interacts with an enzyme, it should be noted whether it is a substrate for the enzyme, an inhibitor for the enzyme, or as a mechanism based inhibitor of the enzyme; c. the molecular properties of the vector, such as: i. molecular structure; ii. molecular weight; iii. binding constant of the vector for its receptor; and/or iv. kinetic or inhibition constants for the vector target if this is an enzyme; d. available structure activity relationship data such as, but not limited to, binding constants or inhibition constants of each structure; and e. the pharmacological effects of a vector.

4.1.1.3. Linkers

A linker is a chemical moiety that connects a pharmacophore to a vector. A typical database of the invention contains information about at least one linker that can be used to design a VLP conjugate having desired pharmacological, chemical and/or physical properties.

The information about each linker that can be included in a database includes, but is not limited to, the functional groups (or "end groups") it contains, chain structure and repeating units in the chain, multiplicity of end groups, range of reasonable chain lengths, chain charge and hydrophobicity, metabolism associated with the linker, whether the linker can be cleaved under metabolic conditions, conformational rigidity; and commercial availability.

A linker can be chosen on the basis of its length, and thus the spacing allowed between the vector and pharmacophore in the conjugate; the solubility characteristics that it will impart to the conjugate; the degree to which it will affect the ability of the conjugate to gain access to the interior of a cell or compartments therein; the degree to which it will affect systemic clearance of the conjugate; the degree to which it will affect distribution of the conjugate; the degree to which it will affect the metabolism of the conjugate; conformational flexibility or rigidity; and commercial availability or ease of synthesis.

Typical linkers of the invention can be of a length of from about 10 to greater than about 50, 75, 100, 150, or 200 angstroms. Linkers can be also be charged, polar, or non- polar, depending on, for example, the desired solubility properties of the VLP conjugate. Linkers can also be prepared which contain hydrophobic chains, or ring structures that limit the conformation or freedom of movement of the conjugates that contain them. Preferably, linkers are selected such that they will remain intact under the conditions with which the conjugates that contain them will interact. Preferred linkers also will not diminish the activity of the pharmacophore or the simultaneous interaction of the pharmacophore with its target and the vector its target. Preferred linkers also permit conjugates to gain access to the interior or interior compartments of a cell.

Examples of linkers include, but are not limited to: polyethylene glycol based linkers; polyethylene enimine based linkers; linear alkane based linkers; and combinations thereof. They can include, but are not limited to, those having chains joining the end groups that are, or are combinations of, linear alkanes, linear alkenes, alkynes, di- and multi-substituted phenyl rings, di- and multi-substituted napthalene rings, di- and multi-substituted cyclohexyl rings, multi- (e.g., tri- and di-) substituted decalin rings, fused aryl or alkyl rings of varying conformational rigidity, linear alkylamines, linear alkyl alcohols, polyalkylamines, polyalkylethers, and polyalkylthioethers. The reactive -or end- groups of each linker can be the same or different. Examples of end groups include, but are not limited to, amines, imines, amine oxides, hydrazines, alcohols, thiols, azo compounds, ethers, thioethers, sulfoxides, sulfones, sulfonamides, sulfonyl esters, phosphate esters, phosphines, methylenes, methines, carboxyl amides, carboxyl esters, and imidates. Specific linkers are shown in FIG. 5. Other examples are disclosed by U.S. patent nos. 4,810,784; 5,034,514; 5,466,681; 5,502,037; 5,723,589; 5,739,287; 5,795,909; 5,824,805; 5,827,819; 5,863,899; 5,869,465; and 5,871,753, each of which is incoφorated herein by reference.

A database of the invention thus contains a table of linker data containing at least one piece of the following information for each linker: a. reactive end groups; b. chain structure and repeating units in chain; c. multiplicity of reactive end groups; d. chain length; e. chain charge and hydrophobicity; and f metabolism and chemical stability under in vivo conditions.

4.1.2. CONJUGATE DESIGN

It is possible to design VLP conjugates rapidly and efficiently once a database has been formed as described herein. In a general method of the invention directed to the design of conjugates suitable for the treatment of a particular disease, pharmacophores are identified that may be therapeutically effective in its treatment. Based on the location of the pharmacophore target, one or more candidate vector targets are identified that are typically located within close proximity to it. Vectors are then identified that bind to those targets, and are ranked using relationships such as those provided herein.

Combinations of vectors, linkers, and pharmacophores are then selected and ranked according to relationships known in the art or described herein. For example, quantitative relationships can be used to estimate the solution or systemic concentration at which the VLP conjugate can exhibit activity similar to that of the unconjugated pharmacophore. Such relationships may take into consideration: the local concentration of a vector target; the dissociation constant of a particular vector from that target, or, if the target is an enzyme, the rate at which it processes the vector; and the minimum effective concentration needed by the pharmacophore for it to exhibit activity. From this information, only those vectors that are estimated to concentrate a particular conjugate to a level sufficient for its particular pharmacophore to exhibit activity at its target are selected. Suitable linkers are then chosen that should not interfere with solubility or access of the conjugate to its site of action. Other design issues can then be addressed, such as systemic distribution, stability to metabolic degradation, rate of systemic clearance, ease of synthesis, and commercial availability of starting materials. A flow chart of a design method of the invention is shown in FIG. 6.

4.1.2.1. Pharmacophore Sorting

In a general method of the invention, the object of which is the design a VLP conjugate useful in the treatment or prevention of a disease or condition, the table of pharmacophores within a database is preferably sorted by at least one criterion. For example, compounds or moieties suitable for use as pharmacophores in a conjugate can be selected from the database as follows: a. Pharmacophores with the lowest molecular weight are ranked above those with higher molecular weight; b. Pharmacophores, which when modified by the attachment of a long-chain substituent still affect their targets, are ranked above those which do not; and c. Those pharmacophores which exhibit activity at low concentrations are ranked above those which require higher concentrations.

4.1.2.2. Vector Sorting The vectors in the database are also preferably sorted or ranked to identify those which are most likely to assist in the targeted delivery of the pharmacophore(s) identified according to Section 4.1.2.1 above.

Vectors are typically ranked according to their suitability in delivering a given pharmacophore to its target such that the concentration of pharmacophore necessary for it to provide a therapeutic or prophylactic effect is decreased. This can be done by using quantitative relationships disclosed herein and information such as, but not limited to: the concentration of a vector target; and the dissociation constant of the vector from its target, or the kinetic constants of the vector for its target if that target is an enzyme. This ranking can further be made with regard to the minimum concentration at which the pharmacophore can exhibit its desired activity.

Suitable vector candidates can be selected, organized, or ranked as follows: a. Vectors can be ranked according to the concentration of their targets in the cellular compartment being targeted. Penicillins and cephalosporins, for example, can be ranked according to the concentration of penicillin binding proteins in the pathogen targeted. Target concentrations should be modified by the number of binding sites available on the target; b. Vectors that bind to receptors can be ranked according to their dissociation constants for their targets. Vectors with low dissociation constants are ranked above those with higher dissociation constants; c. Vectors that are substrates for enzymes can be ranked according to their kinetic constants for their targets. Vectors favorable kinetic constants are ranked above those with less favorable constants; d. Vectors which when modified by the attachment of a long-chain substituent still bind to or affect their targets, are ranked above those which do not; e. Vectors can also be ranked in a quantitative or semi-quantitative manner using the quantitative relationships described below in Example 3 and knowledge of the dissociation constant of the pharmacophore from its receptor or the MIC₅₀ of the pharmacophore. A threshold value for these numbers can be used, or the values can be taken from a representative set of pharmacophores that are likely to be used with the vector (because the vector target and pharmacophore target are adjacent to one another). This information is reported in the literature; f. Vectors with the lowest molecular weight should be ranked above those with higher molecular weight; and g. Vectors with undesirable pharmacological effects of their own should be ranked lower than those having no effect or beneficial effects.

Additional information can also be used to rank potential vectors. For example, and as addressed in more detail herein, potential vectors can be bound to probes (e.g., fluorescent moieties) that are representative of pharmacophores, but which can be used to test the affinity of a vector conjugate for the vector's target. Examples of probes are known in the art, and include those shown in FIG. 7.

4.1.2.3. Vector-Conjugate Pairing Once promising vectors and pharmacophores have been identified, they are matched to form pairs using quantitative relationships disclosed herein. First, and as described above, vectors and pharmacophores can be scored according to how well they may function in an actual conjugate. They are then paired such that the vector target of each conjugate is concentrated in the proximity of the pharmacophore target. Other criteria may also be used to determine optimal pharmacophore/vector pairs. For example, the binding or kinetic constants of each vector with its target can be taken into account. In addition, if the vector/vector target interaction is known to yield pharmacological effects, those effects must be compatible with those of the pharmacophore.

Once a pharmacophore/vector pair has been identified, the database is analyzed for suitable linkers that can be used to join the pair. During the actual synthesis of a VLP conjugate, linkage is accomplished by covalently coupling a functional group of the pharmacophore to a functional group bound to one end of a linker, and coupling a functional group on the other end of the linker to a functional group of the vector.

Linkers are consequently selected from the database with reference to their functional groups. Linkers are further selected based on their length, flexibility, chemical stability, and solubility.

For example, linkers can be chosen to be as small as possible if their influence on the properties of the conjugate are a concern. They can also be chosen on the basis of charge, polarity, and hydrophobicity. Several different linkers can be selected for each vector-pharmacophore pair. Further, although pharmacophores and vectors are typically joined by a single linker, they may also be joined to one another by more than one linker, thereby forming a ring containing both the pharmacophore and vector. The use of multiple linkers of same or different structures and sizes allows the formation of VLP conjugates that comprise two or more pharmacophores attached to one vector, two or more vectors attached to one pharmacophore, or clusters or rings of pharmacophores and vectors. Multiple linkers further allow the formation of compounds of the structure X-VLP-Y, wherein VLP is a conjugate of the invention and X and Y are the same or different, and are moieties (e.g., a non-cleavable or cleavable small or large molecule such as polyethylene glycol) or substrates (e.g., a surface or insoluble polymer) useful in the delivery of the VLP to a patient.

The pharmacophore, vector, and linker are selected to maintain the small size of the assembled conjugate, which is preferably less than about 2000 daltons, more preferably less than about 1500 daltons, and most preferably less than about 1000 daltons. The small size of the conjugates can facilitate their entry into target cells or organisms by diffusion or by the same mechanisms as are used by the pharmacophore alone.

Additionally, smaller conjugates may be absorbed (e.g., by the intestine or following oral administration) at a greater rate than larger conjugates.

Combinations of candidate vectors and pharmacophores are preferably selected as follows: a. Vectors and pharmacophores can be paired according to proximity of their receptors (i.e., those associated with a particular pathogen or in a particular cellular compartment). The site of action and probably the mechanism of action, of the pharmacophore can be identified directly from the tables or the literature, or by experiment or conjecture based on characteristics of known analogues. This information can be used to search the database for vector targets with the same cellular or extracellular location as the site of action of the pharmacophore. Ideally the concentration of a vector target should be at least ten times greater than the MIC₅₀ of the unconjugated pharmacophore or the dissociation constant of the pharmacophore from the pharmacophore target. Preferably, the concentration is also at least time times greater than the concentration of the pharmacophore target, b. Structure activity relationships for both vectors and pharmacophores can be examined. Structure activity relationship data can include the structures of derivatives of pharmacologically active molecules, together with the concentrations at which they exhibit activity. Structures that contain moderate- to long-chain substituents attached to some part of them, but which still exhibit activity or affinity can be identified when present, since the substituents may provide information relevant to the attachment of linkers, c. Linkers of a charge and polarity similar to those of the vector and pharmacophore are preferably chosen. Linkers are further selected which have end groups that can be covalently bound to the attachment sites on the pharmacophore and vector.

4.1.2.4. Vector-Linker-Pharmacophore Grouping

Methods of the invention need not be used to design a single VLP conjugate. Indeed, preferred methods allow the identification of groups of potential conjugates which can subsequently be tested. For example, designs can include different possible linkers, different linker attachment sites on the pharmacophores, different vectors, different linker attachment sites on the vectors, and variations in the length and composition of the linker chain. Since the linker can be selected to make beneficial contributions to the solubility, resistance to degradation, systemic clearance, and tissue penetration of the conjugate, it can be expected that many of the candidates can have adequate solubility, and that for some, the VLP conjugate can have the same access to the pharmacophore and vector targets as the pharmacophores and vectors do alone. Linker chain length can be varied to ensure that it does not interfere with the pharmacophore/pharrnacophore target and vector/vector target interactions.

There are different approaches to acquiring the information necessary for the design of VLP conjugates. Data preferably used during the design include structure- activity relationships and linkage chemistry. Such data can be obtained from the literature or by experiment. Data can also be estimated from published 3-dimensional structures of vector and pharmacophore targets. Additional information may be gathered by constructing test conjugates with reporter groups in place of the vector or the pharmacophore. These can be used for preliminary screening of vector-linker and pharmacophore-linker combinations to determine which are suitable for use together in conjugates. Structure-activity relationship data allows the determination of the locations on the vector and pharmacophore to which linkers can be attached without interfering with their interactions with their targets. For example, the part of a pharmacophore that interacts with its target cannot be modified by incoφoration of large functional groups, since such groups can diminish the activity of the pharmacophore. On the other hand, those portions of a pharmacophore that do not interact with its target may support the attachment of linkers. In many cases, this information has been reported in the literature in connection with attempts to identify structural modifications that improve the interactions of a small molecule with a macromolecular receptor or enzyme. In the process of gathering such information, numbers of structural variants of the small molecule are often prepared, and the interaction of each of these with the macromolecular receptor or enzyme is determined. Frequently, such variants include those with short to moderately long chain substituents, which can serve as models for linkers.

In some cases, structure activity data is not available. But if, for example, the three dimensional structure of a target has been reported, it might be possible to estimate where on a particular pharmacophore or vector a linker could be attached without substantially affecting its activity or affinity. Alternatively, routine experimentation can be conducted using techniques known in the art.

4.1.2.4.1. Experimental Determination of

Vector-Linker and Pharmacophore-Linker Sub-conjugate Properties As discussed elsewhere herein, experiments using labels or reporter groups can be used to select the best combinations of vectors, linkers, and pharmacophores. For example, labels and reporter groups can be used to determine optimal vector-linker conjugates which have a desired affinity for a specific vector target, and which can be attached to a large number of pharmacophores with targets proximate to that vector target. Candidate pharmacophores and vectors can be conjugated to different linkers and tested to determine which combinations do not diminish desired interactions. In particular, test conjugates can be prepared which consist of a vector, a linker, and a model pharmacophore or a reporter group. To measure the degree of interaction of the test conjugate with the vector target, assays can be performed wherein a labeled, but unconjugated, vector is allowed to interact with its target in the presence of varying concentrations of a test conjugate. For example, the vector can be labeled with a radioactive isotope that will not alter its structure or perturb its interaction with the vector target. The concentration at which the test conjugate inhibits the interaction between the labeled vector and the target can then be determined.

Assays can be performed on isolated preparations of vector target. These can be direct measurements of inhibition of vector binding to receptors or of enzymatic activity that processes the vector. For vectors that bind to receptors, the degree of inhibition of vector binding by the test conjugate can be assayed by measuring diminution of binding of the labeled vector to its target caused by the presence of varying concentrations of the test conjugate. For vectors that are substrates for enzymes, the inhibition constant of the test conjugate can be determined by measuring the diminution in the rate of turnover of labeled vector caused by the presence of varying concentrations of the test conjugate. Assays can also be performed in vivo on whole cells or tissue preparations to determine if the test conjugate can gain access to the vector target.

Combinations of a vector and reporter group with various linkers can also be tested to determine the identity of linkers that do not interfere with the vector's interaction with its target. Assays can directly measure the accumulation of the test conjugate in a cell or tissue sample at various concentrations. Reporter groups, which are not themselves pharmacophores, can be chosen to span the range of molecular characteristics of potential pharmacophores; e.g., size, charge, polarity, and hydrophobicity. This can assure that the test conjugates accurately model the behavior of real conjugates.

For targets that can be prepared in a purified form, a combinatorial approach to identifying good vector-linker combinations may also be used. A combinatorial synthesis can be used to produce a library of test conjugates in which the structure of the linker is varied, with the test conjugates being attached through their non-vector end to the pixels of a charged-coupled device. The target macromolecule can be labeled with a fluorescent probe, and its binding to individual pixels can be determined. This procedure can be used to identify vector-linker-probe combinations in which the linker does not interfere with the vector/vector target interaction.

In sum, suitable vector-linker and pharmacophore-linker combinations can be determined as follows: a. For vectors: i. A reporter group or model pharmacophore can be selected by the experimenter. The group can be similar to a proposed pharmacophore, if this had been chosen. ii. Each proposed linker can be joined at the pharmacophore end to the model pharmacophore or reporter group, and at the vector end to the vector to produce a test conjugate. iii. Inhibition of interaction between labeled vector and the vector target by the test conjugate can be measured. iv. Alternately, interaction between the test conjugate and the vector target can be measured directly through use of the reporter group. b. For pharmacophores: i. A reporter group or a model vector can be selected by the experimenter. The group can be similar to a proposed vector. ii. Each proposed linker can be joined at the vector end to the reporter group or model vector, and at the pharmacophore end to the pharmacophore. iii. Inhibition of interaction between labeled pharmacophore and the pharmacophore target by the test conjugate can be measured, iv. Alternately, interaction between the test conjugate and the pharmacophore target can be measured directly through use of the reporter group. c. Combinations: i. The results of steps a and b can be examined for a number of test conjugates; and ii. vector-linker or pharmacophore-linker pairs which exhibit normal or enhanced performance are selected.

4.1.3. TESTING

VLP conjugates designed by the approach described above can be prepared by known techniques and screened for desired characteristics. Such screening can provide information such as whether a conjugate is adequately soluble and whether it can indeed concentrate a pharmacophore at its site of action. Further testing of successful candidates can address systemic distribution, resistance to metabolic degradation, and systemic clearance. All such information can then be entered into a database of the invention, thereby facilitating the more accurate predication of suitable VLP conjugates. Conjugates are preferably tested using methods well known in the art, and those described above in Section 4.1.2.4.

4.2. VLP CONJUGATES VLP conjugates designed by methods of the invention can be used to treat or prevent innumerable diseases and conditions caused by, for example, viruses, bacteria, mycoplasmas, fungi, protists, parasites, or prions. The target of the pharmacophore of a conjugate is typically contained in, or associated with, such causes.

Examples of viruses include, but are not limited to, HBV (Hepatitis B Virus), HIV (Human Immunodeficiency Virus), HCV (Hepatitis C Virus), and influenza. Examples of bacteria include, but are not limited to, Tuberculosis, Streptococci, Chlamydia, Borrelia, Haemophilus, Neisseria, Heliobacter, Shingella, Pasteurella, Coxiella, Mycobacteria, Salmonella, Fusobecteria, Camphlobacteria, and Staphylococci. Examples of fungi include, but are not limited to, Candida; Cryptococci; Histoplasma; Sporothrix; Trichophyton; Microsporum; and Epidermon. Examples of prions include, but are not limited to, those causing or associated with, Kreuzfeldt Jacob Disease, scrapie, and Alzheimers disease. Specific disease that can be treated or prevented by the conjugates of the invention include, but are not limited to, cancer, heart disease, neurodegenerative disease, HIV, and diabetes. Quality of life states that can be improved by the conjugates of the invention include, but are not limited to, memory loss, balding, obesity, impotence, and aging.

Preferred VLP conjugates of the invention are small, and of the form shown in FIG. 1. Conjugates of the invention include conjugates of pharmacophores selected from the group that includes, but is not limited to: antibiotics; antibacterials; antimycoplasmals; antivirals; antifungals; antiprotozoals; molecules active against single celled eukaryotes; molecules active against parasites; molecules that bind tightly to metabolites; inhibitors or activators of binding to macromolecular receptors, including antagonists or agonists of neural receptors and inhibitors or activators of transcription factors; enzyme inhibitors or activators; inhibitors or activators of binding interactions between macromolecules; inhibitors or activators of binding interactions between or catalytic activities of the proteins in cell signaling pathways; nucleic acid polymerase inhibitors; protein synthesis inhibitors; protease inhibitors or activators; kinase and phosphatase inhibitors or activators; glycosylation inhibitors; dihydrofolate reductase inhibitors; ionophores; nucleic acid mutagens; nucleic acid alkylating agents; nucleic acid cleavage agents; and other molecules which modify DNA.

Conjugates of the invention include conjugates of vectors selected from the group that includes, but is not limited to those that interact with: polymerases; transcriptases; ribosomes; proteins involved with protein folding or other chaperones; structural proteins of viruses; abundant proteins of specific eukaryotic cells; antibiotic resistance elements; and enzymes, such as bacterial transamidases, that function in the formation the bacterial cell wall. Conjugates of the invention also include conjugates of vectors that are mechanism directed inhibitors that result in a covalent modification of the target enzyme with covalent binding of the vector to it or that result in covalent binding of the vector to structures in the vicinity of the pharmacophore target.

The linkers used to connect the pharmacophore and vector moieties of a VLP conjugate of the invention include those molecules which comprise a polyethylene glycol, polyethylene enimine, or linear alkane moiety; molecules which comprise an end- group selected from the group consisting of: amines, imines, amine oxides, hydrazines, azo compounds, ethers, thioethers, sulfoxides, sulfonamides, sulfonyl esters, phosphate esters, phosphines, methylenes, methines, carboxyl amides, carboxyl esters, and imidates; and molecules having chains joining the end groups that are, or are combinations of, linear alkanes, linear alkenes, alkynes, di- and multi- substituted phenyl rings, di- and multi- substituted napthalene rings, di- and multi-substituted cyclohexyl rings, di- and multi- substituted decalin rings, di- and multi- substituted heteroaromatic rings, fused aryl, heteroaromatic or alkyl rings of varying conformational rigidity, linear alkylamines, linear alkyl alcohols, polyalkylamines including polyethylene enimine based linkers, polyalkylethers including polyethylene glycol based linkers, and polyalkylthioethers.

Specific VLP conjugates of the invention are described in the following examples.

5. EXAMPLES

5.1. EXAMPLE 1: QUANTITATIVE ANALYSIS Various relationships known to those skilled in the art can be used to rank and select the pharmacophores, vectors, and linkers contained in a database of the invention. Examples of such relationships are provided in FIG. 40. 5.2. EXAMPLE 2: DESIGN OF AN ANTIFUNGAL VLP CONJUGATE

This is an example of the design and preparation of a VLP conjugate useful in the treatment of fungal infections. The vector in this conjugate is sordarin, and the pharmacophore is fluconazole.

Step 1 : Building a database i. Choose the Disease Condition

This example is based on a desire to treat or prevent fungal infection. In particular, a VLP conjugate was desired that is active against fungi, which are the disease organisms in this case.

ii. Make an Initial Table of Ligands and Receptors or Substrates and Enzymes Relating to Fungal Infection The ligand table is shown in FIG. 8.

iii. Make a Table of Pharmacophores That Are Used to Treat Fungal Infection

The table of Fungal Pharmacophores is shown in FIG. 3.

iv. Make a Table of Pharmacophore Targets. Enter in this Table the Pharmacophore Targets Which Correspond to the Pharmacophores Chosen to Combat the Disease State

Fungal targets are shown in the table in FIG. 3.

v. Make a Table of Vector Targets Associated with Fungi. Classify

These According to Their Abundance and According to Their Subcellular or Extracellular Location

Vector targets are shown in the table in FIG. 4.

vi. Make a Table of Vectors. In this Table, Enter Vectors That Are Known to Bind to the Vector Targets

Vectors are shown in the table in FIG. 4.

vii. Make a Table of Linkers A table of linkers is provided in FIG. 4.

Step 2: Making Selections from the Database i. Vector and Vector Target

Choice of the vector target: Fungi are chosen as the demonstration microorganisms. The fungal protein Elongation Factor 2 (EF2) is chosen as the vector target. Elongation Factor 2 functions as part of the ribosomal protein synthesis machinery. It is present at around the same high copy number as the ribosome within the cell. The protein is present in the cytosol of the cell. Although all eukaryotes contain proteins with the function of EF2, experimental work with the sordarin family of antibiotics demonstrated that these small molecules interact specifically with the fungal protein. See, e.g., B. Tse, J. M. Balkovec, C. M. Blazey, M-J. Hsu, J. Neilsen, and D. Schmatz, Bioorganic and Medicinal Chemistry Letters 8:2269-2272 (1998). Choice of the vector: A modified version of sordarin is chosen as the test vector.

The structure of sordarin is shown in FIG. 9. Sordarin reportedly interacts specifically with EF2. As such, its site of interaction is the fungal cytosol. It exhibits antifungal activity at concentrations between micromolar and nanomolar. Binding of sordarin to EF2 inhibits fungal protein synthesis. Sufficient structure activity relationship data exist to indicate that a long aliphatic chain can be attached to the sordarin system without disrupting its antifungal effect, and this implies that binding to EF2 will not be disrupted. Sordarin is a weakly polar molecule, and probably gains access to its site of action by diffusion through the fungal cell wall. The molecule is not expected to be highly water soluble. The molecular weight of sordaricin (sordarin with its sugar moiety removed) is 332 daltons.

Sordarin derivatization: Methodology in the literature describes the removal of the sugar moiety from sordarin and the attachment of a long aliphatic chain. Molecules that are so modified retain antifungal activity. The experimental plan for modification of sordarin, to produce thiopropyl and aminopropyl sordarin, is described in the experimental section in step 4. These structural modifications will allow facile attachment of linker molecules with a high probability that the attached linkers will not interfere with the molecule's interaction with fungal EF2. Linking derivatized sordarin to probe molecules and model pharmacophores: Candidate linkers are obtained from commercial sources. FIG. 5 shows a table of representative linkers. Candidate probes are obtained from commercial sources. FIG. 7 shows a diagram of representative fluorescent probes. Linkers are chosen that should not limit the aqueous solubility of the VLP conjugate or its ability to diffuse into the fungal cytosol. Probe molecules (model pharmacophores) are either small fluorescent molecules or small molecules that incoφorate a radioisotope. These are obtained from commercial sources. The experimental plan for construction of model VLP conjugates using sordarin, linkers, and a probe molecule is described below in step 4. Assays of activity of model VLP conjugates of sordarin and probe molecules are conducted as described below in step 4.

Building and testing the conjugate: Assays to determine antifungal activity of the combination of the sordarin vector, with a linker and a probe molecule are described below in step 4. Exhibition of antifungal activity, although in no way essential for the action of sordarin as a vector, would indicate that the sordarin conjugates retained the ability to bind to fungal EF2. Assays to determine the concentrating effect of the combination of the sordarin vector with a linker and a probe molecule are described below in step 4.

ii. Pharmacophore

Choice of the pharmacophore and the pharmacophore target: Fluconazole, an azole antifungal, was chosen as the test pharmacophore. FIG. 10 shows the structure of fluconazole. Fluconazole is a competitive inhibitor of the fungal enzyme lanosterol 14C demethylase. Fluconazole reportedly binds to the enzyme lanosterol 14C demethylase at concentrations in the submicromolar range. Inhibition of this enzyme disrupts the synthesis of ergosterol, a steroid that is essential for the formation of fungal cell membrane. Inhibition also causes a harmful buildup of intermediates that lie on the fungal synthetic pathway to ergosterol. Fungal lanosterol 14C demethylase, the pharmacophore target, is located on the endoplasmic reticulum membrane, presumably on the cytosolic side of the membrane. Sufficient structure activity relationship data exist to indicate that considerable structural modifications can be made to the azole system without disrupting its antifungal effect, or binding to lanosterol 14C demethylase. Fluconazole is a moderately polar molecule, and probably gains access to its site of action by diffusion through the fungal cell wall. The molecule is significantly water soluble. The molecular weight of fluconazole is 306 daltons.

Fluconazole synthesis and derivatization: Methodology in the literature demonstrates the synthesis of fluconazole and similar azole antifungals. It also indicates that azole antifungals can have considerable structural elaboration, presumably including addition of a linker chain, while retaining antifungal activity. The experimental plan for modification of fluconazole, to prepare molecules with thiobutyl and aminobutyl sidechains, is described in the experimental section. These structural modifications will allow facile attachment of linker molecules with a high probability that the attached linkers will not interfere with the molecule's antifungal activity.

Linking modified fluconazole to probe molecules: Candidate linkers are obtained from commercial sources. FIG. 5 shows a table of representative linkers. FIG. 7 shows a diagram of representative fluorescent probes (model pharmacophores), which are small fluorescent molecules. Probes may also be small molecules that incoφorate a radioisotope. These are obtained from commercial sources. The experimental plan for attachment of linkers and probe molecules to fluconazole is described in step 4.

Assays to determine antifungal activity of the combination of the fluconazole vector, with a linker and a probe molecule are described in step 4. Assays to determine the concentrating effect of the combination of the fluconazole vector, with a linker and a probe molecule are also described in step 4.

Step 3: Designing VLP Conjugates

VLP conjugates of sordarin with fluconazole: FIG. 5 shows a table of representative linkers. The experimental plan for the synthesis of conjugates of sordarin with fluconazole is described in step 4.

Assays to determine antifungal activity of VLP conjugates consisting of the combination of the sordarin vector, a linker, and the fluconazole pharmacophore, are conducted as described step 4. The structure of one of these is shown in FIG. 11.

Step 4: Building and Testing the VLP Conjugate

FIG. 12 details the construction of the sordarin-linker- fluconazole conjugates, the components of which have been selected above in steps 1-3 of this example. Two different lengths of linker are proposed. The "compound/scheme" notation used herein refers to the steps of the synthesis as shown in FIG. 12, which also details construction of test (or surrogate) conjugates with a dansyl group as a flourescent probe attached to either the sordarin vector or fluconazole pharmacophore. Again, two different lengths of linker are proposed. The experiments and assay methods following the construction scheme are proposed to test the sordarin-linker-fluconazole (VLP) construct and the surrogate conjugates.

Sordarin (compound 1. scheme 1): This material is prepared by fermentation of Sordaria araneosa as reported in D. Hauser and H. P. Sigg, Helvetica Chimica Acta 54:1178-1190 (1971).

Sordaricin paramethoxybenzyl ester (compound 2. scheme 1): This material is prepared by the method reported by B. Tse, J. M. Balkovec, C. M. Blazey, M-J. Hsu, J. Neilsen, and D. Schmatz, Bioorganic and Medicinal Chemistry Letters 8:2269-2272 (1998).

Iodopropyl sordarin paramethoxybenzyl ester (compound 3. scheme 1): This compound is prepared by a method analogous to the alkylation procedure of Tse, et. al. Thus, sordaricin paramethoxybenzyl ester (compound 2, scheme 1) (904 mg, 2 mmol) is dissolved in dimethylformamide (30 ml) and diiodopropane (5.9 g, 20 mmol) and sodium hydride (240 mg of a 60% dispersion, 6 mmol) is added. The mixture is stirred overnight at room temperature. Ethyl ether (100 ml) is added and the solution extracted with water (3x, 100 ml). The solvent is removed on the rotary evaporator and the residue dried under vacuum. The residue is dissolved in a minimal volume of ethyl acetate-pentane and chromatographed on silica gel with chloroform-ethyl acetate-pentane as the eluent. The solvent is removed on the rotary evaporator and the product dried under vacuum. Mercaptopropylsordarin paramethoxybenzyl ester (compound 4. scheme I). Iodopropylsordarin paramethoxybenzyl ester (compound 3. scheme 11: (1.24 g, 2 mmol) is dissolved in methanol-ethyl acetate and sodium sulfide (1.56 g, 20 mmol) dissolved in a minimal volume of methanol-water or tetrabutylammonium bisulfide (5.5 g, 20mmol) dissolved in methanol-ethyl acetate is added. The reaction is stirred overnight at room temperature. Ethyl ether (100 ml) is added and the solution extracted with water (3x, 100 ml). The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude product may be used directly in the next step or may be dissolved in a minimal volume of ethyl acetate-pentane and chromatographed on silica gel with chloroform-ethyl acetate-pentane as the eluent. The solvent is then removed on the rotary evaporator and the product dried under vacuum. Mercaptopropylsordarin (compound 5. scheme 1): This compound is prepared by a method analogous to the procedure of Tse, et. al. Thus, mercaptopropylsordarin paramethoxybenzyl ester (compound 4, scheme 1) (0.53 g, 1 mmol) is dissolved in methanol (90 ml) and a hydrogenation catalyst (for example, palladium hydroxide on charcoal) is added. The mixture is stirred under hydrogen (using a hydrogen balloon). Progress of the reaction is followed by thin layer chromatography on silica gel. When the hydrogenation is complete, the reaction mixture is filtered, the solvent removed on the rotary evaporator and the product dried under vacuum. It may be purified by chromatography on silica gel or used directly.

Alternately, mercaptopropylsordarin paramethoxybenzyl ester (0.53 g, 1 mmol) is dissolved in water-acetone (100 ml) and hydrochloric acid (12 N, 10 ml) is added. The reaction mixture is stirred at room temperature or is heated, and the progress of the reaction followed by thin layer chromatography. When the reaction is complete, the product is dissolved in methylene chloride (50 ml) then extracted with water (3x, 100 ml). It is purified by chromatography on silica gel with chloroform-ethyl acetate-pentane as the eluent.

Aminopropylsordarin paramethoxybenzyl ester (compound 1. scheme 2). Iodopropyl sordarin paramethoxybenzyl ester (compound 3. scheme 1): (1.24 g, 2 mmol) is dissolved in ether and a solution of ammonia dissolved in ether (about 0.34 g, 20 mmol of ammonia) is added. The progress of the reaction is followed by thin layer chromatography. When the reaction is complete, The reaction mixture is extracted with water (3x, 100 ml). The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude product may be used directly in the next step or may be dissolved in a minimal volume of ethyl acetate-pentane and chromatographed on silica gel with chloroform-ethyl acetate-pentane as the eluent. The solvent is then removed on the rotary evaporator and the product dried under vacuum.

Aminopropylsordarin (compound 2, scheme 2): The paramethoxybenzyl ester is removed from aminopropylsordarin paramethoxybenzyl ester (compound 1, scheme 2) by the methods used for mercaptopropylsordarin (compound 5, scheme 1) above.

5-(4-Methoxybenzylthio valeronitrile (compound 1. scheme 3): This compound is prepared by dissolving 5-bromovaleronitrile (16.2 g, 100 mmol) in chloroform (200 ml). The solution is cooled in an ice bath and a solution of triethyl amine (10.1 g, 100 mmol) and 4-methoxybenzylmercaptan (15.4 g, 100 mmol) dissolved in chloroform (200 ml) is slowly added. When the addition is complete, the reaction is stirred and allowed to warm to room temperature. Progress of the reaction is followed by thin layer chromatography. When the reaction is complete, the solvent is removed on the rotary evaporator and the residue dried under vacuum. It is resuspended in ethyl ether, and suction filtered. The solvent is removed on the rotary evaporator and the residue dried under high vacuum. The crude material is purified by chromatography on silica gel.

5-(4-Methoxybenzylthio)-l-imido valeric acid ethyl ester hvdrochloride (compound 2. scheme 3): This compound is prepared by an adaptation of the method of Reynaud and Moreau. Thus 5-(4-methoxybenzylthio)valeronitrile (compound 1, scheme 3) (23.5 g. 100 mmol) is dissolved in toluene (125 ml) and methanol (23 ml) is added. The solution is cooled in an ice bath and hydrogen chloride gas is added at atmospheric pressure over 3 hours. The reaction mixture is stirred under hydrogen chloride overnight at room temperature. Dry toluene (100 ml) is then added, then the most of the solvent, and any residual methanol is removed on the rotary evaporator. Ethyl acetate or ethyl ether is then added, and the reaction mixture cooled on ice. The resulting precipitate is collected by suction filtration, and washed with cold, dry toluene. The crude 5-(4-methoxybenzyl)thio)-l-imidovaleric acid ethyl ester is used without further purification.

5-(4-MethoxybenzylthioVl-imidovaleric acid formyl hydrazide (compound 3. scheme 3): This compound is prepared by an adaptation of the method of P. Westermann, H. Paul, and G. Hilgetag, Chemische Berichte 97:528-532 (1964). Thus, crude 5-(4-methoxybenzylthio)-l-imidovaleric acid ethyl ester hydrochloride (compound 2, scheme 3) is suspended in dry toluene or ether (300 ml) and triethyl amine (10.1 g, 100 mmol) and formyl hydrazine (6.0 g, 100 mmol) are added. The reaction mixture is stirred at room temperature and progress of the reaction is followed by thin layer chromatography. When the reaction is complete, the reaction mixture is filtered, then the filtrate washed with water (3x, 100 ml). The organic solution is dried over magnesium sulfate, and the solvent removed on the rotary evaporator. The residue is dried under vacuum and the product is purified by chromatography on silica gel with pentane-ethyl acetate-chloroform as the eluent.

3-(4-(4-Methoxybenzylthio butyl - 2.4-triazaole (compound 4, scheme 3): This compound is prepared by an adaptation of the method of Westermann. Thus,

5-(4-methoxybenzylthio)-l-imidovaleric acid formyl hydrazide (compound 3, scheme 3) (14.75 g, 50 mmol) is added to toluene (500 ml) and aniline (12.5 g). The reaction mixture is heated under reflux and progress of the reaction followed by thin layer chromatography. W en the reaction is complete, the solvent is removed on the rotary evaporator. The residue is purified by chromatography on silica gel with pentane-ethyl acetate-chloroform as the eluent.

5 -Amino val eronitrile : This compound is prepared by dissolving 5-bromovaleronitrile (16.2 g, 100 mmol) in chloroform (200 ml). The solution is cooled in an ice bath and a solution of anhydrous ammonia (17.0 g, 1.0 mol) dissolved in chloroform (200 ml) is slowly added. When the addition is complete, the reaction is stirred and allowed to warm to room temperature. Progress of the reaction is followed by thin layer chromatography. When the reaction is complete, the solvent is removed on the rotary evaporator and the residue dried under vacuum. It is resuspended in ethyl ether, and suction filtered. The solvent is removed on the rotary evaporator and the residue dried under high vacuum. The crude material may be used without further purification, or purified by chromatography on silica gel.

5-(Carbobenzyloxyamino)valeronitrile (compound 1. scheme 4): This compound is prepared by dissolving 5-aminovaleronitrile (9.8 g, 100 mmol) in chloroform (200 ml). The solution is cooled in an ice bath and triethyl amine (10.1 g, 100 mmol) is added. This is followed by the slow addition of a solution of carbobenzyloxy chloride (17.0 g, 100 mmol) dissolved in chloroform (200 ml). When the addition is complete, the reaction is stirred and allowed to warm to room temperature and progress is followed by thin layer chromatography. When the reaction is complete, the reaction mixture is washed with water (3x, 100 ml) and the organic layer dried over magnesium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude material is purified by chromatography on silica gel.

5-(CarbobenzyloxyaminoVl-imidovaleric acid ethyl ester (compound 2, scheme 4): This compound is prepared from 5-(carbobenzyloxyamino)valeronitrile (compound 1, scheme 4) by an adaptation of the method of P. Reynaud and R. C. Moreau, Bull Soc. Chim. France 2997-2999 (1964), which is described for the preparation of 5-(4-methoxybenzylthio)-l-imidovaleric acid ethyl ester hydrochloride (compound 2, scheme 3).

5-(Carbobenzyloxyamino)-l-imidovaleric acid formyl hydrazide (compound 3, scheme 4): This compound is prepared from 5-(carbobenzyloxyamino)-l-imidovaleric acid ethyl ester (compound 2, scheme 4) by an adaptation of the method of Westermann which is described for the preparation of 5 -(4-methoxybenzylthio)-l-imido valeric acid formyl hydrazide (compound 3, scheme 3).

3-(4-(Carbobenzyloxyamino)butylV1.2.4-trizaole (compound 4. scheme 4): This compound is prepared from 5-(carbobenzyloxyamino)-l-imidovaleric acid formyl hydrazide (compound 3, scheme 4) by an adaptation of the method of Westermann which is described for the preparation of 3-(4-(4-methoxybenzylthio)butyl)-l,2,4-triazaole (compound 4, scheme 3). l-[2-(2,4-DifluorophenylV2.3-epoxypropyll-lH-1.2.4-triazole methylsulfonate (compound 1. scheme 5) is prepared by the method disclosed by K. Richardson in United States patent 4,404,216, which is incoφorated herein by reference.

2-(2.4-DifluorophenylVl-(lH-1.2.4-triazol-l-ylV3-(3-(4-(4-methoxybenzylthioVb utyl)-lH-l,2,4-triazol-l-yl)propan-2-ol (compound 1, scheme 6): This compound is prepared by a method analogous to that reported by Richardson. Thus, l-[2-(2,4-difluorophenyl)-2,3-epoxypropyl]-lH-l, 2,4-tri azole methylsulfonate (compound 1, scheme 5) (4.75 g, 20 mmol) is dissolved in dimethylformamide (35ml) and 3-(4-(4-methoxybenzylthio)butyl)-l,2,4-triazaole (compound 4, scheme 3) (6.93 g, 25 mmol) and anhydrous potassium carbonate (9.1 g, 66 mmol) are added. The reaction mixture is heated to 90 degrees with stiπing. The progress of the reaction is followed by thin layer chromatography on silica gel. When the reaction is complete, the reaction mixture is cooled to room temperature. Water (180 ml) is added and the mixture extracted twice with chloroform (2x, 60 ml). The combined chloroform extracts are washed with water (2x, 100 ml), dried over magnesium sulfate, filtered, and the solvent removed on the rotary evaporator. The product is purified by chromatography on silica gel with pentane-ethyl acetate-chloroform as the eluent. 2-(2.4-DifluorophenylVl-(lH-1.2.4-triazol-l-ylV3-(3- (4-thiobutylVlH-1.2.4-triazol-l-yl')propan-2-ol (compound 2. scheme 6): This compound is prepared by dissolving 2-(2,4-difluorophenyl)-l-(lH-l,2,4-triazol-l-y_)-3-(3-(4- (4-methoxybenzylthio)butyl)-lH-l,2,4-triazol-l-yl)propan-2-ol (compound 1, scheme 6) (5.14 g, 10 mmol) in ice cold trifluoroacetic acid (100 ml) to which Hg(OAc)₂ (3.21 g, 10 mmol) is added. The reaction mixture is stirred on ice for 1 hour, then mercaptoethanol (3.12 g, 40 mmol) is added. The solvent is removed on the rotary evaporator, and the residue is dried under vacuum. The product is purified by chromatography on silica gel with pentane-ethyl acetate-chloroform as the eluent. Alternately the 2-(2,4-difluorophenyl)-l-(lH-l,2,4-triazol-l-yl)-3-(3-(4-(4- methoxybenzylthio)butyl)-lH-l,2,4-triazol-l-yl)propan-2-ol (compound 1, scheme 6) is dissolved in trifluoroacetic acid and the solution refluxed. Progress of the reaction is followed by thin layer chromatography. When the reaction is complete, the solvent is removed on the rotary evaporator and the residue dried under vacuum. The product is purified by chromatography on silica gel with pentane-ethyl acetate-chloroform as the eluent.

2-(2.4-DifluorophenylV 1 -( 1 H- 1.2.4-triazol- 1 - yl -3 -(3 -(4-carbobenzyloxyaminobut yD-lH-l.Σ^-triazol-l-vDpropan-Σ-ol (compound 3, scheme 6): This compound is prepared by a method analogous to that reported by Richardson. Thus, l-[2-(2,4-difluorophenyl)-2,3-epoxypropyl]-lH-l,2,4-triazole methylsulfonate

(compound 1, scheme 5) (4.75 g, 20 mmol) is dissolved in dimethylformamide (35ml) and 3-(4-(carbobenzyloxyamino)butyl)-l,2,4-trizaole (compound 4, scheme 4) (6.85 g, 25 mmol) and anhydrous potassium carbonate (9.1 g, 66 mmol) is added. The reaction mixture is heated to 90 degrees with stirring. The progress of the reaction is followed by thin layer chromatography on silica gel. When the reaction is complete, the reaction mixture is cooled to room temperature. Water (180 ml) is added the mixture extracted twice with chloroform (2x, 60 ml). The combined chloroform extracts are washed with water (2x, 100 ml), dried over magnesium sulfate, filtered, and the solvent removed on the rotary evaporator. The product is purified by chromatography on silica gel with pentane-ethyl acetate-chloroform as the eluent.

2-(2.4-DifluorophenylVl-(lH-1.2.4-triazol-l-ylV3-(3-(4-aminobutvn-lH-1.2.4-tri azol-l-yl)propan-2-ol (compound 4. scheme 6): This compound is prepared from 2-(2,4-difluorophenyl)- 1 -( 1 H- 1 ,2,4-triazol- 1 -yl)-3-(3 -(4-carbobenzyloxyaminobutyl)- 1 H- l,2,4-triazol-l-yl)propan-2-ol (compound 3, scheme 6) (5.11 g, 10 mmol) which is dissolved in ethanol (100 ml). Palladium on carbon (0.5 g) is added and the reaction is placed under hydrogen, with stirring. The progress of the reaction is followed by thin layer chromatography. When the reaction is complete, the reaction mixture is filtered, the filtrate washed with ethanol, and the solvent removed from the combined ethanol washings on the rotary evaporator. The residue is purified by chromatography on silica gel with pentane ethyl acetate-chloroform as the eluent.

Alternately, 2-(2,4-difluorophenyl)- 1 -( 1 H- 1 ,2,4-triazol- 1 -yl)-3 -(3 -(4- carbobenzyloxyaminobutyl)-lH-l,2,4-triazol-l-yl)propan-2-ol (compound 3, scheme 6) (5.11 g, 10 mmol) is dissolved in acetonitrile (100 ml) and chlorotrimethylsilane (1.09 g, 10 mmol) is added. The reaction is followed by thin layer chromatography. When complete, the reaction mixture is washed with water (3x, 100 ml), and dried over magnesium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The residue is purified by chromatography on silica gel with pentane ethyl acetate-chloroform as the eluent.

2-(2.4-Difluorophenyl)-l-(lH-L2,4-triazol-l-vπ-3-(3-carbobenzyloxyaminometh yl-lH-l,2,4-triazol-l-yl)propan-2-ol (compound 5. scheme 6): This compound is prepared from l-[2-(2,4-difluorophenyl)-2,3-epoxypropyl]-lH-l,2,4-triazole methylsulfonate (compound 1, scheme 5) by a method analogous to that reported by Richardson and described for

2-(2,4-difluorophenyl)- 1 -( 1 H- 1 ,2,4-triazol- 1 -yl)-3 -(3 -(4-carbobenzyloxyaminobutyl)- 1 H- l,2,4-triazol-l-yl)propan-2-ol (compound 3, scheme 6). The starting triazole, 3-carbobenzyloxyaminomethyl-l,2,4-trizaole, is prepared by the method of Westermann. 2-(2.4-DifluorophenylVl-(lH-1.2.4-triazol-l-vD-3-(3-aminomethyl-lH-1.2.4-triaz ol-l-yl)propan-2-ol (compound 6. scheme 6): This compound is prepared from 2-(2,4-difluorophenyl)- 1 -( 1 H- 1 ,2,4-triazol- 1 -yl)-3 -(3 -carbobenzyloxyaminomethyl- 1 H- 1 , 2,4-triazol-l-yl)propan-2-ol (compound 5, scheme 6) by the method described for 2-(2,4-difluorophenyl)- 1 -( 1 H- 1 ,2,4-triazol- 1 -yl)-3 -(3 -(4-aminobutyl)- 1 H- 1 ,2,4-triazol- 1 -y l)propan-2-ol (compound 4, scheme 6).

Conjugate of mercaptopropylsordarin with 2-(2.4-difluorophenyl^')- 1 -( 1 H- 1.2.4-triazol- 1 -ylV3 -(3-(4-aminobutvD- 1 H- 1 ,2.4-triazol- 1 - yr)propan-2-ol (compound 1. scheme 7 A, scheme 7B1: This compound is prepared from mercaptopropylsordarin (compound 5, scheme 1) and 2-(2,4-difluorophenyl)-l- ( 1 H- 1 ,2,4-triazol- 1 -yl)-3 -(3 -(4-aminobutyl)- 1 H- 1 ,2,4-triazol- 1 -yl)propan-2-ol (compound 4, scheme 6). Thus succinimidyl 6-[6-(((iodoacetyl)amino)hexanoyl)amino]hexanoate (SIAXX, 0.51 g, 1.0 mmol) or succinimidyl 6-((iodoacetyl)amino)hexanoate (SIAX, 0.39 g, 1.0 mmol) is dissolved in dry methylene chloride (20 ml) and mercaptopropylsordarin (compound 5, scheme 1) (0.37 g, 1.0 mmol) is added. The reaction mixture is stirred at room temperature and its progress is followed by thin layer chromatography. When the condensation is completed, 2-(2,4-difluorophenyι)-l-(lH-l,2,4-triazol-l-yι)-3- (3-(4-aminobutyl)-lH-l,2,4-triazol-l-yl)propan-2-ol (0.38 g, 1.0 mmol) is added. The reaction is again stiπed at room temperature and its progress followed by thin layer chromatography. When the reaction is complete, the solvent is removed on the rotary evaporator and the residue dried under vacuum. The residue is purified by high pressure liquid chromatography. Conjugate of mercaptopropylsordarin with dansyl chloride (compound 1. scheme

8A. scheme 8B): This compound is prepared from mercaptopropylsordarin (compound 5, scheme 1) and dansyl chloride. Thus, dansyl chloride (2.69 g, 10 mmol) is dissolved in dry methylene chloride (100 ml) and 1,4-diaminobutane (8.8 g, 100 mmol) is added. The reaction mixture is stirred at room temperature and its progress is followed by thin layer chromatography. When the reaction is complete, the reaction mixture is washed with aqueous sodium hydroxide (0.5 N), then water. It is dried over sodium sulfate, then filtered. The solvent is removed on the rotary evaporator, and the residue dried under vacuum. The crude material is used without further purification. Succinimidyl 6-[6-(((iodoacetyl)amino)hexanoyl)amino]hexanoate (SIAXX, 0.51 g, 1.0 mmol) or succinimidyl 6-((iodoacetyl)amino)hexanoate (SIAX, 0.39 g, 1.0 mmol) is dissolved in dry methylene chloride (10 ml) and mercaptopropylsordarin (compound 5, scheme 1) (0.37 g, 1.0 mmol) is added. The reaction mixture is stirred at room temperature and its progress is followed by thin layer chromatography. When the condensation is completed, the reaction product of dansyl chloride and 1,4-diaminobutane is dissolved in methylene chloride (10 ml) and one tenth of this material (1.0 mmol) is added. The reaction is again stiπed at room temperature and its progress followed by thin layer chromatography. When the reaction is complete, the solvent is removed on the rotary evaporator and the residue dried under vacuum. The residue is purified by high pressure liquid chromatography.

Conjugate of mercaptobutylfluconzole with dansyl chloride (compound 1. scheme 9 A. scheme 9B This compound is prepared from (2-(2,4-difluorophenyl)- 1 -( 1 H- 1 ,2,4-triazol- 1 -yl)-3 -(3 -(4-thiobutyl)- 1 H- 1 ,2,4-triazol- 1 -yl)propan-2-ol)

(compound 2, scheme 6) and dansyl chloride. Thus, dansyl chloride (2.69 g, 10 mmol) is dissolved in dry methylene chloride (100 ml) and 1,4-diaminobutane (8.8 g, 100 mmol) is added. The reaction mixture is stirred at room temperature and its progress is followed by thin layer chromatography. When the reaction is complete, the reaction mixture is washed with aqueous sodium hydroxide (0.5 N), then water. It is dried over sodium sulfate, then filtered. The solvent is removed on the rotary evaporator, and the residue dried under vacuum. The crude material is used without further purification. Succinimidyl 6-[6-(((iodoacetyl)amino)hexanoyl)amino]hexanoate (SIAXX, 0.51 g, 1.0 mmol) or succinimidyl 6-((iodoacetyl)amino)hexanoate (SIAX, 0.39 g, 1.0 mmol) is dissolved in dry methylene chloride (20 ml) and 2-(2,4-difluorophenyl)-l-(lH-l,2,4-triazol- l-yl)-3-(3-(4-thiobutyl)-lH-l,2,4-triazol-l-yl)propan-2-ol (compound 2, scheme 6) (0.39 g, 1.0 mmol) is added. The reaction mixture is stiπed at room temperature and its progress is followed by thin layer chromatography. When the condensation is completed, the reaction product of dansyl chloride and 1,4-diaminobutane is dissolved in methylene chloride (10 ml) and one tenth of this material (1.0 mmol) is added. The reaction is again stiπed at room temperature and its progress followed by thin layer chromatography. When the reaction is complete, the solvent is removed on the rotary evaporator and the residue dried under vacuum. The residue is purified by high performance liquid chromatography.

5.3. EXAMPLE 3: DESIGN OF ANTIBACTERIAL VLP CONJUGATES

The following is an example of the design of VLP conjugates which contain vector moieties that bind to targets not usually present in high concentrations in bacteria. However, bacteria can be modified by placing the genes for the vector targets on plasmids such that the target proteins are expressed in the bacteria at a high concentration. This approach therefore demonstrates that VLP conjugates can be designed which will target diseased cells to the exclusion of non-diseased cells. Construction of Antibacterial Vector/Pharmacophore Tables

Creation of Ligand Tables Information on bacterial protein ligands is gathered from the literature. This information preferably includes: molecular weight, receptor(s), mode of inhibition, association constant (Ka), IC50, binding sites, and MIC₅₀. This information is organized into a tabular format. An example is given in FIGS. 13a-d.

Creation of Ligand Receptor Tables Information on ligands that bind to bacterial proteins is gathered from the literature. This information preferably includes: species, receptor concentration, molecular weight (daltons), cellular compartment, solubility, natural substrate, function, and three-dimensional structure. An example is given in FIG. 13e.

Creation of Vector and Pharmacophore Tables Once the ligand data has been collected, ligands are separated into vectors and pharmacophores. For a ligand to be considered a good vector, it must have (1) a high receptor concentration in the cellular compartment being targeted and (2) a high association constant for its receptor. To incoφorate the contribution of these factors in determining the suitability of ligands as vectors, a concentrating factor is calculated for each ligand. The concentrating factor is determined by first multiplying the number of receptor molecules in the cell by the number of ligand binding sites on the receptor to give the number of ligand binding sites per cell. This number is then converted to micromoles and divided by the estimated volume of a bacterial cell in liters to give the concentration of receptor sites in micromolar. The binding site concentration is then multiplied by the ligand' s association constant for the site to generate a concentrating factor. Ligands with concentrating factors above 50 are designated vectors. Ligands with concentrating factors below 50 are designated pharmacophores. An example of this process is given in FIG. 13f.

The ligands are then ranked according to concentrating factor. Ligands with high concentrating factors will efficiently concentrate a wider range of pharmacophores than ligands with low concentrating factors. However, it is possible that a ligand with a low concentrating factor can serve as a vector for a pharmacophore with a relatively high association constant or low MIC₅₀. For the puφoses of this example, all ligands with concentrating factors above 50 are designated as vectors, and all ligands with concentrating factors below 50 as designated as pharmacophores. An example vector table is given in FIG. 13g. An example pharmacophore table is given in FIG. 13h.

Determination of Candidate Antibacterial VLP Pairs

Criteria for Vector/Pharmacophore Pairs Selection (Bacterial Cytosol), either

1 > 100 Kvd / ([VTt] - 10 Kpd) or ([VTt] - 10 * MIC₅₀) / Kvd > 100

Convert dissociation constants to association constants:

KvaKpa[VTt] - 10 Kva > 100 Kpa or Kva[VTt] - 10KvaMIC₅₀ > 100 Rearranging:

Kva [VTt] > 100 + ((10 Kva) / Kpa) or Kva[VTt] > 100 + 10KvaMIC₅₀ These equations permit the construction of a simple search algorithm to test Vector/Pharmacophore pairs whose binding affinities and target concentrations satisfy the inequalities. The above equations assume that [VTt] > 10 [PTt]. Thus, to target ribosomal pharmacophores, one must have a vector target whose concentration is on the order often times the pharmacaphore target concentration. In this case, EF-Tu is present in roughly 10 times the concentration of the bacterial ribosome. Kirromycin, which binds with high affinity to Ef-Tu (Ka = 4 xlO⁶ M^"1) is selected from the vector table. The above set of inequalities is then used to determine the viability of various kiπomycin/pharmacophore pairs. An example is given in FIG. 13i.

This table gives the possible vector/pharmacophore pairs that satisfy the inequalities given above. A similar set of computations can be performed for other promising cytosolic vectors. After possible V/P pairs are determined, SAR data is collected for each pharmacophore, with special attention to pharmacophores whose activity is unaffected by the attachment of a long chain. Following this, promising V/P pairs are ranked according to the combined molecular weights of the vector and pharmacophore. Those with the lowest combined molecular weight are then investigated as potential VLP conjugates.

Examples of VLP conjugates are shown in FIG. 14-17, and discussed in detail below. These conjugates are designed by first selecting a disease state (bacterial infection), and then pharmacophores that are reportedly active against it. The selected anti-bacterials include: penicillin, tetracycline and trimethoprim. For each pharmacophore, sufficient structure activity data was present to allow a good estimate of one or more locations on the pharmacophore molecule to which a linker can be attached without significant disruption to the activity. These sites of action are in the bacterial periplasmic space for penicillin and in the bacterial cytosol for tetracycline and trimethoprim. Hence, two proteins are chosen as model bacterial vector targets, one that can be expressed in the periplasm and one that can be intracellularly expressed, since interaction with these targets can thus result in accumulation of the respective antibiotics at their sites of action. Although bacteria contain unique targets, the genes for these model target proteins can be introduced on plasmids, and the level of expression of the proteins can be controlled. In general, vector targets with high levels of expression are desirable, and in these systems, expression can be adjusted to high levels.

The vectors for these targets are then chosen. Based on the known properties of the protein avidin, the tightly binding molecule biotin can be used as a vector. Based on the known properties of the cytosolic protein, beta galactosidase, a substrate, which is a galactose based mechanism based inhibitor, can be used as a vector. Biotin binds to avidin with a dissociation constant of around 10^"14 molar. Sufficient data exists to indicate that the mechanism based beta galactosidase inhibitor chosen can be processed at a rate resulting in a change in the bound inhibitor concentration of at least 10^"9 molar per second. This can result in a pharmacophore being covalently bound, and thus concentrated, within a bacterial cell to an effective level within the length of the bacterial cell cycle. For both vectors, sufficient structure activity relationship data exists to allow a good estimate of locations on the vector molecules to which a linker can be bound without disturbing their activity significantly.

The pharmacophores are then paired with their respective vectors. For each pharmacophore antibiotic, the minimum concentration at which the antibiotic can exert an effect is known. Using the quantitative information disclosed herein and this number, the expected concentration of the vector target, and the dissociation constant of the vector from its target or the expected rate of enzyme processing of its substrate, it is possible to estimate the solution concentration of the VLP conjugate at which the antibiotic pharmacophore can be concentrated to an effective level at its site of action. This is substantially lower than that required by the unconjugated pharmacophore molecule. Only selection of the linkers remains. These are selected to have polarity and solubility similar to that of the vectors and pharmacophores. This should not limit the solubility of the conjugate or its access to the cellular compartment containing the vector target and pharmacophore target. Linker lengths of 15 to 20 bonds are selected. Linker length may need to be optimized, however, to allow the interaction of the VLP conjugate with both the vector target and pharmacophore target. This is achieved by routine methods known to those skilled in the art.

5.4. EXAMPLE 4: SYNTHESIS OF BIOTIN/PENICILLIN

CONJUGATE

Synthesis of biotin-penicillin conjugates (penicillin as a pharmacophore and biotin as a vector) is shown in FIG. 18. As in the case of all synthesis disclosed herein, the purity and identity of compounds can be determined using methods known in the art including, but not limited to, nuclear magnetic resonance (NMR) spectroscopy and mass spectral analysis. Similarly, all solvents and standard reagents are analytical reagent grade, available from commercial suppliers.

6-(5-oxopentyl)aminopenicillanic acid (compound 1 FIG. 18): (+) 6-Aminopenicillanic acid (1.08 g, 5 mmol) is dissolved in water-methanol (50 miililiters), with the pH adjusted to 8-9 with aqueous sodium bicarbonate. An excess of freshly distilled glutardialdehyde (10.0 g, 100 mmol), will then be added. The mixture is stirred for 1 hour at room temperature, then the solvent is removed on a rotary evaporator, and the residue dried under vacuum, washed with methylene chloride, and redried under vacuum. This material is used in the next step without further purification. 6-(5-biocvtinhydrazidopentyl^")aminopenicillanic (compound 2 FIG. 18): The material is prepared by the general methods in Greg T. Hermanson, Bioconjugate Techniques (Academic Press, New York: 1996) ("Hermanson"). The crude 6-(5- oxopentyl)aminopenicillanic acid (compound 1 FIG. 18), from the above reaction, is dissolved in water, and the pH is adjusted to 8-9 with aqueous sodium bicarbonate. To this is added biocytin-LC-hydrazide (1.93 g, 5 mmol). The mixture is stirred for 1 hour at room temperature, then cooled in an ice bath. Working in a fume hood, a slight excess of sodium cyanoborohydride solution (3 mmol) in methanol is added. The mixture is stiπed at 0 degrees for 10 minutes, then excess sodium cyanoborohydride destroyed by addition of acetic acid- water, until the pH is around 7 and evolution of gas has ceased. The solvent is removed on a rotary evaporator, and the residue dried under vacuum. This material is purified by high performance liquid chromatography on silica gel with ethyl acetate- methylene chloride-triethyl amine as the eluant or by reverse phase high performance liquid chromatography with water-acetonitrile as the eluant, or by ion exchange chromatography using a sulfate or carboxylate ion exchange resin.

6-(5-(6-(biotinyl)-aminocaproyl)-hvdrazidopentyl)-aminopenicillanic (compound 3 FIG. 18): The material is prepared by the general methods in Hermanson. The crude 6- (5-oxopentyl)aminopenicillanic acid (compound 1 FIG. 18), from the above reaction, is dissolved in water, and the pH is adjusted to 8-9 with aqueous sodium bicarbonate. To this is added biotin-LC -hydrazide (1.86 g, 5 mmol). The mixture is stirred for 1 hour at room temperature, then cooled to around 0 degrees in an ice bath. Working in a fume hood, a slight excess of sodium cyanoborohydride solution (3 mmol) in methanol is added. The mixture is stirred at 0 degrees for 10 minutes, then excess sodium cyanoborohydride destroyed by addition of acetic acid-water, until the pH is around 7 and evolution of gas has ceased. The solvent is removed on a rotary evaporator, and the residue dried under vacuum. This material is purified by high performance liquid chromatography on silica gel with ethyl acetate-methylene chloride-triethyl amine as the eluant or by reverse phase high performance liquid chromatography with water- acetonitrile as the eluant, or by ion exchange chromatography using a sulfate or carboxylate ion exchange resin.

6-(6-(biotinyl)-aminocaproyl)aminopenicillanic acid (6(LC- biotin)aminopenicillanic acid (compound 4 FIG. 19): The material is prepared by the general methods in Hermanson. (+) 6-Aminopenicillanic acid (1.08 g, 5 mmol) is dissolved in water-methanol (50 milliliters), with the pH adjusted to 8-9 with aqueous sodium bicarbonate. NHS-LC-biotin (2.28 g, 5 mmol) dissolved in methanol is added and the mixture is stirred for one hour or until thin layer chromatography shows the reaction to be complete. The pH is maintained at around 8 by additions of sodium bicarbonate solution. The solvent is removed under vacuum. Alternately, to a solution of NHS-LC- biotin (2.28 g, 5 mmol) dissolved in chloroform is added solution of anhydrous triethyl amine (0.5 g, 5 mmol) and (+) 6-aminopenicillanic acid, (1.08 g, 5 mmol), in chloroform. The solution is stiπed for one hour or until thin layer chromatography shows the reaction to be complete, and the solvent is then removed under vacuum. The residue from these reactions is washed with 1% aqueous acetic acid, then water, then dried under vacuum. The product is purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile as the eluant, or by high performance liquid chromatography on silica gel with methylene chloride-ethyl acetate-tri ethyl amine as the eluant, or by high performance ion exchange liquid chromatography with a sulfate or carboxylate ion exchange resin.

6-(4-(2-(6-(biotinyl)-aminohexylamino)-aceto)-thiobutyrylimidato)- aminopenicillanic acid (compound 5 FIG. 20): The material is prepared by the general methods in Hermanson. To a solution of (+) 6-aminopenicillanic acid (1.08 g, 5 mmol) in methanol-water is added Traut's reagent (0.69 g, 5 mmol). The reaction is stirred at room temperature for 1 hour, then N-iodoacetyl-N-biotinylhexylenediamine (iodoacetyl-LC- biotin, 2.55 g, 5 mmol), dissolved in methanol, is added and the reaction is stiπed for an additional hour or until thin layer chromatography shows the reaction to be complete. The solvent is then removed under vacuum, and the product purified by high performance liquid chromatography on silica gel with chloroform-ethyl acetate-triethyl amine as the eluant or reverse phase high performance liquid chromatography on silica gel with a linear gradient of water-acetonitrile as the eluant, or by high performance ion exchange liquid chromatography with a sulfate or carboxylate ion exchange resin.

5.5. EXAMPLE 5: ACTIVITY OF BIOTIN/PENICILLIN CONJUGATES

Biotin-penicillin conjugates may be tested for antibacterial activity against strains of E. coli K 12 containing cloned chicken avidin, that will either be expressed and secreted into the periplasmic space or is delivered to the outer cell membrane as a fusion protein with the transmembrane portion of the aspartate receptor. In this test system, avidin is the vector receptor, biotin is the vector, and penicillin is the antibiotic pharmacophore. The gene for avidin is carried by a plasmid and expression of the protein to a high concentration is induced by materials added to the growth medium. Testing of inhibition of bacterial growth is performed with a series of increasing concentrations of the biotin-penicillin conjugate, alone in the growth medium. In parallel, a series of increasing concentrations of an unconjugated penicillin derivative, ampicillin, is also be tested. A control experiment is also be performed, in which a sufficiently high concentration of free biotin is added to the medium to saturate all avidin expressed by the bacteria, in the presence of the biotin penicillin conjugates. Again, the degree of bacterial growth is measured in the presence of a series of increasing concentrations of the biotin- penicillin conjugate, and a parallel measurement is performed with a series of increasing concentrations of the unconjugated penicillin derivative, ampicillin. Additionally, the conjugates is tested against a strain of E. coli K 12 that does not express cloned chicken avidin. Finally, the biotin-penicillin conjugates is pre-treated with beta lactamase, then tested as above. Again, in both cases, a parallel experiment is performed with ampicillin. In each case, bacterial growth in the presence of increasing concentrations of the conjugate or of unconjugated antibiotic is measured, and plotted as the curves seen in the accompanying graph (FIG. 21).

For the E. coli strains that express avidin, unconjugated penicillins (such as ampicillin) should inhibit bacterial growth at the expected concentration range (around 1- 10 ug/ml). The biotin-penicillin conjugates should inhibit bacterial growth at a significantly lower concentration range (10 to 100 fold lower). In the presence of saturating concentrations of biotin or using a strain of E. coli K 12 that does not express cloned chicken avidin, the biotin-penicillin conjugates should inhibit bacterial growth at around the same concentration range as the unconjugated penicillins. When pretreated with beta lactamase, neither the biotin-penicillin nor the unconjugated penicillins should show any inhibition of growth.

5.6. EXAMPLE 6: SYNTHESIS OF A TRIMETHOPRIM DERIVATIVE AS A PHARMACOPHORE

2,6-Dimethoxy-4-((N.N-dimethylamino)methyl)phenol hydrochloride (Compound 1 FIG. 22): This material is prepared by the method of Barbara Roth, Justina Z. Strelitz, and Barbara S. Rauckman, J. Med. Chem. 23:379-84 (1980). Thus, 2,6-dimethylphenol

(92 g, 0.6 mmol) is slowly added to a mixture of 2 N hydrochloric acid (315 ml, 0.63 mmol), 25% aqueous dimethylamine (135 g, 0.75 mmol), and 37% formaldehyde (81 g, 1 mol). The reaction is exothermic, and the reaction mixture will turn puφle. Additional dimethylamine solution (45 g, 0.25 mmol) is added and the mixture allowed to stand overnight at room temperature. The solvent is removed under vacuum, and the resultant tan solid washed well with diethyl ether then dissolved in ethanol (1.5 liters) and crystallized. This yields the purified product (around 122 g, 82%); melting point is estimated to be about 226-227 degrees.

2.4-Diamino-5-(3,5-dimethoxy-4-hvdroxybenzyl)pyrimidine (compound 2 FIG. 22): This material is prepared by the method of Roth, Strelitz, et. al. Thus, to a solution of 2,4-diaminopyrimidine (22 g, 0.2 mol) and 2,6-dimethoxy-4-((N,N-dimethylamino)- methyl)phenol hydrochloride (compound 1 FIG. 22, 49.5 g, 0.2 mol) in ethylene glycol (300 ml) is added sodium hydroxide (11 g, 0.203 mol). The mixture is heated under nitrogen to 150-160 degrees with stirring for 3 hours. The reaction mixture is cooled and the solvent removed under vacuum. The residual oil is washed with water, then acetone. This should yield a tan precipitate. This is recrystallized from dimethylformamide to yield white plates (around 23 g, 42%,). Melting point should be 265-267 degrees 2.4-Diamino-5-(3.5-dimethoxy-4-(3-chloropropoxy)benzyl)pyrimidine (compound 3 FIG. 22): This material is prepared by the method of Barbara Roth, Edward Aig, Barbara S. Rauckman, Justina Z. Strelitz, Arthur P. Phillips Robert Ferone, S. R. M. Bushby, and Carl W. Sigel, J. Med. Chem. 24:913-941 (1981). Thus, 2,4-diamino-5-(3,5- dimethoxy-4-hydroxybenzyl)pyrimidine (compound 2 FIG. 22, 3.5 g, 12.7 mmol) is dissolved under nitrogen in dry dimethylsulfoxide (50 ml) by warming to 50 degrees. After cooling the solution, sodium methoxide (0.69 g, 12.7 mmol) is added under nitrogen and when this has been dissolved, 1,3-dichloropropane (1.44 g, 12.7 mmol) is added, under nitrogen. The reaction mixture is allowed to stand at room temperature under nitrogen for 4 hours, then neutralized with acetic acid. The solvent is removed under vacuum and the residue washed with aqueous sodium hydroxide solution, then water. The product is dried under vacuum, and optionally purified by high performance liquid chromatography. Melting point should be 163-164 degrees.

2,4-Diamino-5-(3,5-dimethoxy-4-(3-bromopropoxy)-benzyl)pyrimidine (compound 4 FIG. 22): The compound is prepared by the method above, using 1,3- dibromopropane (2.57 g, 12.7 mmol). 2.4-Diamino-5-(3.5-dimethoxy-4-(3-mercaptopropoxy)-benzyl)pyrimidine

(compound 5 FIG. 22): To 2,4-diamino-5-(3,5-dimethoxy-4-(3-chloropropoxy))- benzyl)pyrimidine (compound 3 FIG. 22, 1.06 g, 3 mmol) or 2,4-diamino-5-(3,5- dimethoxy-4-(3-bromopropoxy))-benzyl)pyrimidine (compound 4 FIG. 17, 1.19 g, 3 mmol), dissolved in ethanol or dimethylformamide (50 ml) is added sodium sulfide (2.34 g, 30 mmol) dissolved the same solvent. The mixture is allowed to stand at room temperature for several hours, then the solvent removed under vacuum. The product is washed with water, dried under vacuum, and can be purified by high performance liquid chromatography.

5.7. EXAMPLE 7: SYNTHESIS OF TETRACYCLINE

DERIVATIVES (BACTERIAL RIBOSOME INHIBITORS) AS PHARMACOPHORES

13-(3-Mercaptopropylthio)-5-hydroxy-6-deoxytetracycline (compound 1 FIG. 23): This compound is prepared by a method similar to that of Levy. Thus a mixture of methacycline hydrochloride (3.0 g, 6.2 mmol) and 2,2'-azobisisobutyronitrile (AIBN, 0.25 g) and 1,3-propanethiol (6.7 g, 62 mmol) in ethanol (100 ml) is heated under nitrogen at reflux for 6 hours. The reaction mixture is cooled, filtered, and concentrated to one-fifth volume under vacuum. The crude product is precipitated by the addition of cold ethyl ether, then washed with ethyl ether. It is then dissolved in hot water, at pH 5.0, and extracted into chloroform. The solvent is removed, and the product treated with activated charcoal in methanol. The product is further purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile. 13-(4-Mercapto-2,3-dihvdroxybutylthio)-5-hvdroxy-6-deoxytetracycline (compound 2 FIG. 23): This compound is prepared as above, using dithiothreitol (4.79 g, 31 mmol).

9-Nitro-5-hydroxy-6-deoxytetracvcline (compound 3 FIG. 24): This compound is prepared by the method of Barden, T.C., et al, J. Med. Chem. 37:3205-3211 (1994). Thus, doxycycline hydrochloride (14.5 g, 30.1 mmol) is slowly added to concentrated sulfuric acid (50 ml). After gas evolution has stopped, the orange solution is slowly dripped into ice cold ethyl ether (1 liter) and the resulting hydrosulfate salt collected by filtration, washed with ether and dried under nitrogen. The orange powder is dissolved in sulfuric acid (70 ml) at room temperature and sodium nitrate (4.0g, 47.0 mmol) is added over a 2 minute period. The reaction is stiπed for 3 hours then dripped into ice-cooled, stirred ethyl ether (2 liters). The precipitate is collected by filtration. The precipitate is washed with ether, and air dried to give an orange powder. The crude material is used without further purification. 9-Amino-5-hvdroxy-6-deoxytetracycline (compound 4 FIG. 24): This compound is prepared by the method of Barden, et. al. Thus, crude 9-nitro-5-hydroxy-6- deoxtetracycline (compound 3 FIG. 24) from the preparation above is dissolved in a small amount of methanol and filtered into a 500 milliliter Paar hydrogenation bottle. The solution is diluted to 200 ml with methanol, and 10% Pd on charcoal (1.0 g) is added. The system is charged with hydrogen to a pressure of 50 psi and the bottle shaken for 2 hours. The catalyst is removed by filtration and the filtrate diluted to 300 ml with methanol, then dripped into stirred ethyl ether to give a light tan powder (around 16 g). The product is purified by high performance liquid chromatography using a Hamilton PRP-1 column and a linear gradient of water-acetonitrile with added 0.1% trifluoroacetic acid. 9-Bromoacetamido-5-hvdroxy-6-deoxytetracvcιine (compound 5 FIG. 24): This compound is prepared by the method of Barden, et. al. Thus 9-amino-5-hydroxy-6- deoxytetracycline (compound 4 FIG. 24, 3.43 g, 7.47mmol) is mixed in N- methylpyrrolidin-2-one (40 ml) with sodium bicarbonate (2.0g, 23.8 mmol) at room temperature. Bromoacetyl bromide (0.75 ml, 8.62 mmol) is added and the solution stiπed for 30 minutes. Additional bromoacetyl bromide (0.4 ml, 4.6 mmol) and sodium bicarbonate (1.0 g, 11.9 mmol) is added, and the stirring continued for 2 hours. The reaction mixture is then filtered into stiπed ethyl ether (1.5 liters) and the crude product (around 4 g) is collected as a yellow powder by filtration. The product is purified by high performance liquid chromatography using a Hamilton PRP-1 column and a linear gradient of water-acetonitrile with added 0.1 % trifluoroacetic acid.

9-(4-Thiobutyrlimidato)-5-hvdroxy-6-deoxytetracycline (compound 6 FIG. 24): To 9-bromoacetamido-5-hydroxy-6-deoxytetracycline (compound 5 FIG. 24, 0.58 g, 1.0 mmol) dissolved in chloroform or methanol is added an excess of anhydrous ammonia. The solution is allowed to stand for 1 hour at room temperature, then the solvent is removed under vacuum. The residue is washed with acetone-water, to remove the ammonium bromide. Then it is dried under vacuum. Alternatively, the material is flash chromatographed on silica gel with chloroform-mefhanol as the eluant and the solvent removed under vacuum. The material is then dissolved in water or methanol, and 2- iminothiolane (Traut's reagent, 0.30 g, 2.2 mmol) is added. The solution is stirred at room temperature for 1 hour and the solvent removed under vacuum. The crude material is washed with acetone then ethyl ether and dried under vacuum. The product may be purified by high performance liquid chromatography using a Hamilton PRP-1 column and a linear gradient of water-acetonitrile with added 0.1% trifluoroacetic acid.

9-(4-Aminobutylamino)-5-hvdroxy-6-deoxytetracycline (compound 7 FIG. 24): To 9-bromoacetamido-5-hydroxy-6-deoxytetracycline,(compound 5 FIG. 24, 0.58 g, 1.0 mmol) dissolved in methanol or chloroform is added 1,4 diaminobutane (0.88 g, 10 mmol). The solution is allowed to stand for 1 hour at room temperature, then the solvent removed under vacuum. The residue is washed with acetone-water, then dried under vacuum. Alternatively, the material is flash chromatographed on silica gel with chloroform-methanol as the eluant and the solvent removed under vacuum. The product may be purified by high performance liquid chromatography using a Hamilton PRP-1 column and a linear gradient of water-acetonitrile with added 0.1 % trifluoroacetic acid.

N-Hydroxymethylmaleimide (compound 8 FIG. 25): This material is prepared by the method in Martell, M.J., et al, J. Med. Chem. 10:359-363 (1967) (referencing P. O. Tawney, P.O., et al, J.O.C 26:15 (1961). Thus, to a mixture of maleimide (0.97 g, 10 mmol) and 40% aqueous formaldehyde (0.81 ml, 11 mmol) is added aqueous sodium hydroxide (0.3 ml to adjust the pH to ca. 7.5). The solid should dissolve and the solution is allowed to stand at room temperature for 3 hours. The solvent is then removed under vacuum to leave an oil which can be crystallized. The product may be further purified by sublimation.

7-Succinimidomethyl-6-demethyl-6-deoxytetracycline (compound 9 FIG. 25): This material is prepared by the method in Martell et. al Thus, a solution of 453 mg (1.0 mmol) of 6-demethyl-6-deoxytetracycline hydrochloride in 96% sulfuric acid (5ml) is treated with N-hydroxymethylmaleimide (compound 8 FIG. 25, 190 mg, 1.5 mmol). The solution is stiπed at room temperature for 35 minutes, then poured into dry ether (150 ml). The resulting precipitate is collected by filtration, washed well with dry ether and dried under vacuum. The precipitate is dissolved in water (18 ml) and the pH adjusted to 5.0 with 1 N sodium hydroxide. The resulting precipitate is filtered, washed with two portions of water (1 ml each) and dried under vacuum. The product may be purified by reverse phase high performance liquid chromatography using a linear gradient of water- acetonitrile.

7-(3-(3-Mercaptoρropyl)thio)succinimidomethyl-6-demethyl-6-deoxytetracvcline (compound 10 FIG. 25): To a solution of 7-succinimidomethyl-6-demethyl-6- deoxytetracycline(compound 9 FIG. 25, 0.26 g, 0.5 mmol) in methanol-chloroform (50 ml) is added 1,3-propanedithiol (0.54 g, 5.0 mmol). The solution is stiπed for 3 hours at room temperature or until thin layer chromatography shows the reaction to be complete, and the solvent is removed under vacuum. The residue is washed with acetone and dried under vacuum. The material is purified by reverse phase high performance liquid chromatography using a linear gradient of water-acetonitrile.

7-(3-(4-Mercapto-2.3-hvdroxybutyl)thio)succinimidomethyl-6-demethyl-6- deoxytetracycline (compound 11 FIG. 25): To a solution of 7-succinimidomethyl-6- demethyl-6-deoxytetracycline (compound 9 FIG. 25, 0.26 g, 0.5 mmol) in methanol- chloroform (50 ml) is added dithiothreitol (0.77 g, 5.0 mmol). The solution is stiπed for 3 hours at room temperature or until thin layer chromatography shows the reaction to be complete, and the solvent is removed under vacuum. The residue is washed with acetone, and dried under vacuum. The residue is purified by reverse phase high performance liquid chromatography using a linear gradient of water-acetonitrile.

5.8. EXAMPLE 8: SYNTHESIS OF A BETA GALACTOSIDASE

MECHANISM BASED INHIBITOR (MBP AS A VECTOR l-Bromo-2.3.4.6-tetraacetyl-beta-D-galactose (compound 1 FIG. 26): This material is prepared by a method analogous to that of Halazy, S., et al, Bioorganic Chem.

18:330-344 (1990). Thus, to 1,2,3,4,6-pentaacetyl-β-D-galactose (compound 1 FIG. 26, 28.1 g, 72 mmol) is added ice cold 33% hydrobromic acid (275 ml). The reaction mixture is stiπed at room temperature for 3 hours then extracted with chloroform (3 x 150 ml). The combined extracts is washed with cold water several times (until the pH reaches 6-7), dried over sodium sulfate, then filtered. The solvent is removed under vacuum to give a brown viscous oil (around 29 g) which is used without further purification. l-(2-Oxomethyl-4-nitrophenoxy)- 2.3.4,6-tetraacetyl-β-D-galactose (compound 2

FIG. 26): This material is prepared by the method of Janda, K.D., et al, Science 275:945- 948 (1997) according to the method of Halazy, S., et al, Bioorganic Chem. 18:330-344 (1990). Thus, a solution of 5-nitrosalicylaldehyde (7.0 g, 42 mmol) in dichloromethane is vigorously stirred at room temperature with 5% sodium hydroxide (70 ml) and tetramethylammonium bromide (2.26 g, 7 mmol). To this stiπed mixture is added a solution of l-bromo-2,3,4,6-tetraacetyl-beta-D-galactose (compound 1 FIG. 26, 11.5 g, 27.9 mmol) in dichloromethane (20 ml). Stirring is continued at room temperature for 40 hours. The two phases is then separated and the organic layer washed with 5% aqueous sodium hydroxide (2 x 20 ml), then with water several times. It is then dried over sodium sulfate, filtered, and the solvent removed under vacuum. The crude product (around 14g) is chromatographed on silica gel with 7:3 petroleum ether: ethyl ether as the eluant, This should yield the purified product (around 7.8g, 55%). l-(2-Difluoromethyl-4-nitrophenoxy)-2.3,4.6-tetraacetyl-β-D-galactose

(compound 3 FIG. 26): This material is prepared by the method of Janda et. al. according to the method of S. Halazy et. al. Thus, to a solution of l-(2-oxomefhyl-4-nitrophenoxy)- 2,3,4,6-tetraacetyl-β-D-galactose (compound 2 FIG. 26, 1.0 g, 2.0 mmol) in anhydrous dichloromethane (4 ml) at 0 degrees is slowly added diethylaminosulfurtrifluoride (DAST, 1 ml). The reaction mixture is stirred at 20 degrees for 18 hours, quenched slowly at 0 degrees with methanol (1 ml) and the solvent is evaporated under vacuum. The residue is chromatographed on silica gel with 8:2 petroleum ether:ethyl acetate as the eluant, This should yield the purified product (around 0.99 g, 95%). l-(2-Difluoromethyl-4-aminophenoxy)-2,3.4.6-tetraacetyl-β-D-galactose (compound 4 FIG. 26): This material is prepared by the method of Janda et. al. Thus 1- (2-difluoromethyl-4-nitrophenoxy)-2,3,4,6-tetraacetyl-β-D-galactose (compound 3 FIG. 22, 1.04 g, 2 mmol) is dissolved in ethyl acetate and loaded into a 500 milliliter Paar hydrogenation bottle. To this solution is added 10% Pd on charcoal (0.15 g). The system is charged with hydrogen to a pressure of 50 psi and the bottle is shaken for 2 hours. The catalyst is removed by filtration and the solvent removed under vacuum to give a residue. The residue is chromatographed on silica gel with petroleum ether: ethyl acetate as the eluant. This should yield the purified product (around 0.92 g, 95%). l-(2-Difluoromethyl-4-aminophenoxy)-β-D-galactose (compound 5 FIG. 26): This material is prepared by the method of Janda et.al. according to the method of S. Halazy et. al. Thus, l-(2-difluoromethyl-4-aminophenoxy)-2,3,4,6-tetraacetyl-β-D- galactose (compound 4 FIG. 26, 0.86 g, 1.75 mmol) is dissolved at 20 degrees in methanol (15 ml) to which sodium methoxide (13 mg) had been added. The reaction mixture is allowed to stand for 2 hours, then neutralized with IN hydrochloric acid (0.2 ml) and filtered. The solvent is removed from the filtrate under vacuum. This should yield the product (around 0.54 g, 95%). The material may be further purified by reverse phase chromatography with a linear gradient of water-acetonitrile as the eluant. l-(2-Difluoromethyl-4-(6-(6-(((iodoacetyl)amino)-hexanoyl)amino)hexanoyl)- amidophenoxy)-β-D-galactose (compound 6 FIG. 26): l-(2-Difluoromethyl-4- aminophenoxy)-β-D-galactose (compound 5 FIG. 26, 0.32 g, 1 mmol) is dissolved in methanol-water, or it is dissolved in 100 mm sodium bicarbonate solution, and methanol is added. Succinimidyl 6-(6-(((iodoacetyl)amino)-hexanoyl)amino)hexanoate (0.39 g, 1 mmol) is then added, dissolved in methanol or ethanol. The solution is stiπed at room temperature, and the progress of the reaction monitored by thin layer chromatography. The solvent is removed under vacuum, and the residue dried under vacuum. The material may be purified by reverse phase chromatography with a linear gradient of water- acetonitrile as the eluant. l-(2-Difluoromethyl-4-(6-(6-(((iodoacetyl)amino)-hexanoyl)amino)hexanoyl)- amidophenoxy)-2,3 A6-tetraacetyl-β-D-ga1actose (compound 7 FIG. 26): l-(2- Difluoromethyl-4-aminophenoxy)-2,3,4,6-tetraacetyl-β-D-galactose (compound 4 FIG. 22, 0.52 g, 1 mmol) is dissolved in chloroform. Succinimidyl 6-(6-(((iodoacetyl)amino)- hexanoyl)amino)hexanoate (0.39 g, lmmol) is then added. The solution is stiπed at room temperature, and the progress of the reaction is monitored by thin layer chromatography. The solvent is removed under vacuum, and the residue dried under vacuum. The material is purified by chromatography on silica gel with petroleum ether-chloroform as the eluant.

5.9. EXAMPLE 9: SYNTHESIS OF MBI/BGM CONJUGATES Trimethprim-beta galactosidase MBI and tetracycline-beta galactosidase MBI conjugates can be used as pharmacophores, to which can be attached beta galactosidase mechanism (BGM) based inhibitors as a vector to form VLP conjugates of the invention.

The conjugation of l-(2-difluoromethyl-4-(6-(6-(((iodoacetyl)amino)- hexanoyl)amino)hexanoyl)amidophenoxy)-β-D-galactose (compound 6 FIG. 26) with thiol or amino derivatives of tetracyclines (compound 2 FIG. 25, compound 6 FIG. 24, compound 7 FIG. 24, compound 10 FIG. 25, compound 11 FIG. 25) or trimethoprim (compound 5 FIG. 22) can be accomplished by the following procedure. l-(2-Difluoromethyl-4-(6-(6-(((iodoacetyl)amino)-hexanoyl)amino)hexanoyl)- amidophenoxy)-β-D-galactose (0.72 g, lmmol) is dissolved in methanol or methanol- water or dimethylethylene glycol and the tetracycline thiol derivative, tetracycline amino derivative, or trimethomprim thiol derivative is immediately added, dissolved in the appropriate solvent. The reaction mixture is stiπed at room temperature, typically for 1 hour, and the progress of the reaction is monitored by thin layer chromatography. The solvent is removed under vacuum with minimal heating, and the residue is dried under vacuum. The product is purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile as the eluant, or high performance ion exchange liquid chromatography using a sulfate or carboxylate ion exchange resin. The crude starting materials may also be used in the coupling reaction.

Conjugation l-(2-Difluoromethyl-4-(6-(6-(((iodoacetyl)amino)-hexanoyl)amino)- hexanoyl)-amidophenoxy)-2,3,4,6-tetraacetyl-β-D-galactose (compound 7 FIG. 26) with thiol derivatives of tetracyclines (compound 2 FIG. 23, compound 6 FIG. 24, compound 10 FIG. 25, compound 11 FIG. 25) or trimethoprim (compound 5 FIG. 22) can be accomplished by the following procedure. l-(2-Difluoromethyl-4-(6-(6-(((iodoacetyl)amino)-hexanoyl)amino)hexanoyl)- amidophenoxy)-2,3,4,6-tetraacetyl-β-D-galactose (0.89 g, lmmol) is dissolved in chloroform or methanol or dimethylethylene glycol and the tetracycline thiol derivative or trimethomprim thiol derivative is immediately added, dissolved in the appropriate solvent. The reaction mixture is stiπed at room temperature, typically for 1 hour, and the progress of the reaction is monitored by thin layer chromatography. The solvent is removed under vacuum with minimal heating, and the residue is dried under vacuum. The product may be purified by high performance liquid chromatography on silica gel, reverse phase high performance liquid chromatography with a linear gradient of water- acetonitrile as the eluant, or high performance ion exchange liquid chromatography using a sulfate or carboxylate ion exchange resin. The product may also be deacylated, as below, before purification. The crude starting materials may also be used in the coupling reaction. The deacylation of l-(2-Difluoromethyl-4-(6-(6-(((iodoacetyl)amino)- hexanoyl)amino)hexanoyl)amidophenoxy)-2,3,4,6-tetraacetyl-β-D-galactose (compound 7 FIG. 26) conjugates with thiol or amino derivatives of tetracyclines (compound 2 FIG. 23, compound 6 FIG. 24, compound 10 FIG. 25, compound 11 FIG. 25) or trimethoprim (compound 5 FIG. 22) is done as follows. The material from the above coupling reaction is dissolved in a small volume methanol or methanol-water to which sodium methoxide (around 0.15 equivalents) had been added. The reaction mixture is allowed to stand for 2 hours, and the pH checked periodically, and adjusted to remain basic, as necessary, by the addition of further sodium methoxide. The reaction is then neutralized with IN hydrochloric acid and filtered. The solvent is removed from the filtrate under vacuum to yield the crude product, which can be purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile as the eluant, or high performance ion exchange liquid chromatography using a sulfate or carboxylate ion exchange resin.

5.10. EXAMPLE 10: PROPERTIES OF MBI/BGM CONJUGATES Conjugates of a mechanism based inhibitor (or MBI) of beta-galactosidase and tetracycline or trimethoprim may be tested for antibacterial activity against strains of E. coli K 12 containing cloned beta galactosidase, which is expressed in soluble form in the bacterial cytosol. In this test system, the beta galactosidase is the vector receptor, the MBI galactose analog is the vector, and tetracycline or trimethoprim is the antibiotic pharmacophore. The gene for beta galactosidase is carried by a plasmid and expression of the enzyme to a high concentration is induced by materials added to the growth medium. Testing of inhibition of bacterial growth is performed with a series of increasing concentrations of the MBI galactose analog-antibiotic conjugate, alone in the growth medium. In parallel, a series of concentrations of the unconjugated antibiotic tetracycline or triemthoprim is also be tested. A control experiment is also performed, in which a sufficiently high concentration of lactose, or of a galactose analog that is not conjugated to an antibiotic, is added to the medium in order to saturate all beta galactosidase expressed by the bacteria, in the presence of the vector-antibiotic conjugates. Again, the degree of bacterial growth is measured in the presence of a series of increasing concentrations of the MBI galactose analog-antibiotic conjugate, and in parallel, a series of increasing concentrations of the unconjugated antibiotic tetracycline or trimethoprim is tested. Additionally, the conjugates is tested against a strain of E. coli K 12 that does not express any beta galactosidase, and in parallel, the unconjugated antibiotic tetracycline or trimethoprim is also be tested. In each case, bacterial growth in the presence of increasing concentrations of the conjugate, or of the unconjugated antibiotic, is measured, and plotted as curves. Calculated curves are shown in FIG. 21. For the E. coli strains that express beta galactosidase, unconjugated antibiotics should inhibit bacterial growth at the expected concentration ranges (around 1-10 ug/ml for tetracycline and 0.01-0.1 ug/ml for trimethoprim). The MBI galactose analog- antibiotic conjugates should inhibit bacterial growth at a significantly lower concentration range (10 to 100 fold lower). In the presence of saturating concentrations of lactose, or of a galactose analog that is not conjugated to antibiotic, or using a strain of E. coli K 12 that does not express any beta galactosidase, the MBI galactose analog-antibiotic conjugates should inhibit bacterial growth at around the same concentration range as the respective unconjugated antibiotics.

5.11. EXAMPLE 11: MEASURING ANTIBIOTIC ACTIVITY The antibiotic activity of each VLP conjugate is performed upon several representative, non-pathogenic gram positive and gram negative strains of bacteria using the following methods. For compounds expected to have enhanced antibiotic activity against actual pathogens, these measurements can constitute a preliminary screening procedure. Bacteria, stored as frozen cultures, is streaked onto LB agar plates, containing any necessary antibiotics, and is grown at 37 degrees in an incubator. On the day before a measurement, individual colonies is picked and used to inoculate LB medium (5 or 25 ml) containing any necessary antibiotics or supplemental nutrients, in culture tubes or Erlenmeyer flasks. These are shaken overnight at 37 degrees to produce saturated cultures. On the day of the measurement, aliquots of these cultures (0.5 ml to 1 ml) is used to inoculate fresh medium containing any necessary antibiotics or supplemental nutrients, in culture tubes or shake flasks, and the culture grown to an absorbance of 0.5 (lcm pathlength cuvette) at 550 nm. This log phase growth culture is then used as the inoculum in the following procedures. For some experiments the cultures may also be allowed to grow to saturation, and used as an inoculum. Method A

Serial dilutions of the antibiotic or test substance are made in water or the appropriate organic solvent. To produce the final antibiotic or test substance concentrations, aliquots of these dilutions are added to culture tubes containing LB media (5 or 10 ml) containing any necessary additional nutrients or activators or inhibitors of antibiotic or test substance activity, as required by the experiment. The tubes are inoculated with aliquots (50 to 100 ul) of the bacterial culture, and shaken in an incubator, usually at 37 degrees. At intervals of several hours, for a period of up to several days, they are scored visually for bacterial growth. Additionally, aliquots may be withdrawn and the absorbance at 550 nm determined, for a more precise spectrophotometric determination of the cell concentration.

Method B

Serial dilutions of the antibiotic or test substance are made in water or the appropriate organic solvent. Molten LB agar is prepared, containing any necessary additional nutrients or activators or inhibitors of antibiotic or test substance activity, as required by the experiment. It is cooled below 60 degrees and aliquoted into pre-warmed culture tubes. To each tube is added an aliquot of one of the set of serial dilutions of the antibiotic or test substance. As soon as the aliquot has been added to a tube, the contents of the tube is mixed rapidly with a vortex mixer, and poured into a petri dish. This is covered, let stand to allow the agar to harden, and dried briefly in an incubator. Aliquots of the bacterial culture (50 to 100 ul), prepared as above, are spread onto the medium in each petrie dish, and these are covered and placed in an incubator. After 24 hours or more, the number of bacterial colonies growing on each plate are counted. To determine the degree of cell killing, as opposed to growth stasis, aliquots of a growing or saturated bacterial culture are treated with concentrations of antibiotic or test substance sufficient to halt growth. After an incubation period, the cells from each aliquot of culture are collected by centrifugation, washed twice in fresh LB medium (without antibiotic), and resuspended in fresh LB medium (without antibiotic, 5 or 10 ml). Aliquots of these suspensions are then spread onto fresh LB plates, that also have no added antibiotic. As a control for this experiment, a portion of the original culture, to which no antibiotic or test substance had been added, is taken. The cells is treated in the same manner as the antibiotic or test substance treated cells. The plates are incubated at 37 degrees and after 24 hours or more, the number of bacterial colonies growing on each plate is counted.

Data Inteφretation

The experimental data is inteφreted as follows. The growth of E.coli should be inhibited by the unconjugated antibiotic at the concentration expected for each antibiotic. The antibiotics and bacteria thus should behave normally. The growth of E.coli should be inhibited by significantly lower concentrations of the conjugates because the vector should cause the conjugate molecule, and thus the pharmacophore, to accumulate in the bacteria to a concentration that can inhibit growth or kill the bacteria, when the concentration of conjugate in the medium was sufficient for significant amounts of the vector to become bound to the vector receptor. This is the effect expected for vector reagents. When there is no vector receptor in the bacteria or when the vector receptor is saturated by a molecule which is not the conjugate, the conjugate cannot accumulate in the bacteria to a concentration higher than it can have in the suπounding medium because it cannot bind to the vector receptor. Under these conditions, the concentration at which the conjugate exerts an effect can be about the same as the concentration at which the pharmacophore alone is effective. Finally, when the antibiotic portion (pharmacophore) of the conjugate molecule is converted to an inactive form, the conjugate molecule is expected to show no inhibition of bacterial growth, demonstrating that the vector and linker portions of the conjugate are devoid of antibiotic activity.

Thus, vector reagents are expected to exhibit the pharmacological activity of the unconjugated pharmacophore, but at a concentration much lower than that needed by the unconjugated pharmacophore.

5.12. EXAMPLE 12: ELONGATION FACTOR BINDING MOLECULE. KIRROMYCIN

In this example, the bacterial elongation factor EF Tu, which is present at the high concentration of around 10^"4 molar in the bacterial cytoplasm, is selected as the vector target. The vector selected is the antibiotic kirromycin, which binds to this elongation factor with a dissociation constant of around 10^"6 molar. Sufficient structure activity relationship data is present for kiπomycin to allow the design of modified molecules that should bind to the EF Tu elongation factor about as well as kiπomycin itself. Kirromycin reportedly gains access to the cytosol of gram positive bacteria by diffusion. As a vector in a conjugate, it can be expected to concentrate pharmacophore antibiotics in the bacterial cytosol to their effective concentration, but with a solution concentration of the conjugate around 100 fold lower than that required by the unconjugated pharmacophore molecule.

With this information, and given a disease state caused by gram positive bacteria, the kirromycin -EF Tu vector- vector receptor pair can be a candidate for use in a VLP conjugate that can enhance the concentration of many pharmacophores in the bacterial cytosol of gram positive bacteria. For a disease state caused by bacterial infection, potential pharmacophores could be the antibiotics trimethoprim and tetracycline. Conjugates of kiπomycin with these antibiotics may therefore be useful in the treatment of bacterial infections.

5.13. EXAMPLE 13: SYNTHESIS OF KIRROMYCIN

DERIVATIVES AS VECTORS

1-N-Desmethylgoldinamine (compound 1 FIG. 27) and Goldinonic Acid

(compound 2 FIG. 27): 1-N-Desmethylgoldinamine is prepared as its formate salt by the method of Tavecchia, P., et al, Journal of Antibiotics 49:1249-57 (1996). Thus, kiπomycin, (20g, 25.12 mmol) dissolved in dioxane (200 ml) and formic acid (99%, 50 ml) is stirred at 350 for 22 hours. After cooling, the mixture is concentrated to half its volume, then ethyl ether (250 ml) is added. The solvent is decanted and saved and the gummy residue stiπed in ethyl ether/ethyl acetate (1 :1 v/v). The solid material obtained is washed with ethyl ether and dried at room temperature to give 1-N- desmethylgoldinamine as the formic acid salt (22 mmol, 87%).

The goldinonic acid is recovered by a modification of the procedure of Maehr, H, et. al, Canadian Journal of Chemistry 58:501-26 (1980). The decanted solvent and the collected washings are combined, and the volatile materials removed on the rotary evaporator. The residue is dried under vacuum. It is then washed with hexane-ethyl ether. The residue is dissolved in a minimal volume of methanol and purified by reverse phase chromatography on silanized silica gel or by reverse phase high performance liquid chromatography with a linear gradient of chloroform-methanol or chloroform-acetonitrile as the eluant. The goldinonic acid recovered by this procedure may be used without further purification.

As shown in FIG. 28, the modified goldinonic acid or goldinonic acid analog (1 mmol), with its hydroxyl groups protected as silyl ethers, is dissolved in dry chloroform (20 ml) and EDC (l-(3-(dimethylamino)propyl)-3-ethylcarbodiimide) (155) (155 mg, 1 mmol) or DCC (dicyclohexylcarbodiimide) (206) (206 mg, 1 mmol) is added. The reaction mixture is stiπed at room temperature for 1 hour, then imidazole (68) (68 mg, lmmol) is added. The reaction mixture is again stiπed for an hour, then a solution of goldinamine formate (compound 1 FIG. 27) (546 mg, 1 mmol) and triethylamine (101 mg, 1 mmol) dissolved in dry chloroform is added. The reaction mixture is stirred overnight. Then ethyl ether is added, and the resulting precipitate washed with ethyl ether and ethyl acetate. The residue is purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile or acetone as the eluant. Tris-O-triisopropylsilylgoldinonic Acid (compound 1 FIG. 29): Purified goldinonic acid lactone (compound 2 FIG. 27) (314) (9.42 g, 30 mmol), from the formic acid cleavage of kiπomycin, is dissolved in tetrahydrofuran-chloroform and triethylamine (101) (12.1 g, 120 mmol) is added. To this is added triisopropylsilyl chloride (193.5) (23.2 g, 120 mmol). The reaction mixture is analyzed periodically by thin layer chromatography until it shows no further change. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The product is purified by flash chromatography on silica gel.

Aminoboronation Product of Tris-O-triisopropylsilylgoldinonic Acid (compound 2 FIG. 29): Purified tris-O-triisopropylsilylgoldinonic acid (compound 1 FIG. 29) (785) (2.355 g, 3 mmol) is dissolved in dry diglyme (20 ml) under nitrogen. The flask is cooled in an ice bath and a solution of 9-borabicyclononane (9-BBN ) (3 mmol) or thexylborane (3 mmol) in tetrahydrofuran is added. The reaction mixture is stiπed at room temperature. After 3 hours, a solution of hydroxylamine-O-sulfonic acid (396 mg, 3.5 mmol) in dry diglyme (20 ml) is added. The reaction mixture is heated slowly, and the disappearance of the boronated adduct is monitored by thin layer chromatography on silica gel. After a maximum of 3 hours at 100 degrees, the reaction mixture is cooled, water (20 ml) is added, and the pH is adjusted to 4 with hydrochloric acid (IN) (measured by pH paper). The aqueous phase is washed with ether, then made neutral with sodium hydroxide solution (IN), saturated with sodium chloride, and extracted with chloroform. The chloroform solution is dried over sodium sulfate, and the solvent is removed on the rotary evaporator. The residue is dried under vacuum. The crude product can be used directly in the next step or purified by chromatography, although the silyl ether protecting group is likely to have been partially hydrolyzed during the workup, leading to inhomogenous material, all of which should be usable product.

Addition of (Gamma- thiobutyrolactone to the Amine Adduct of the Olefin Cleavage Product of Tris-O-triisopropylsilyl Goldinonic Acid (compound 4 FIG. 29): The aminoboronation product of tris-O-triisopropylsilylgoldinonic acid (compound 2 FIG. 29) (801) (1.602 g, 2 mmol) is dissolved in chloroform and (gamma- thiobutyrolactone (102) (204 mg, 2 mmol) and triethylamine (101) (202 mg, 2 mmol) is added. The reaction mixture is stiπed at room temperature, until thin layer chromatography indicates that all of the starting material has reacted. The reaction mixture is washed with water, and dried over sodium sulfate. The solvent is removed on the rotary evaporator. The crude material is fully resilylated using the procedure for tris- O-triisopropylsilyl-goldinonic acid (compound 1 FIG. 29). An aqueous workup should be sufficient to hydrolyze any silyl carboxyate ester or silyl thioether formed. The resulting material is purified by high performance liquid chromatography on silica gel. Ozono lysis olefin cleavage product of tris-O-triisopropylsilyl goldinonic acid (compound 1 FIG. 30): Purified tris-O-triisopropylsilylgoldinonic acid (compound 1

FIG. 29) (785) (785 mg, 1 mmol) is dissolved in methylene chloride (20 ml) in a 100 ml round bottom flask equipped with a gas inlet, outlet, thermometer, and magnetic stiπing bar and placed in a fume hood. The solution is cooled in to around -30 degrees by partial immersion of the flask in a dry ice- acetone bath and a stream of oxygen containing ozone gas (from an ozone generator) is bubbled into it with stirring. During the reaction, the temperature is allowed to fall to around -60 degrees. The progress of the reaction mixture is monitored by thin layer chromatography. When all of the starting material has reacted, the reaction mixture is purged with a stream of dry nitrogen to evaporate excess ozone. Then dimethylsulfide (62) (0.124 g, 2 mmol) in methanol (5 ml) is added at dry ice temperatures, and the reaction mixture is allowed to warm to -10 degrees. It is stiπed at this temperature for an hour, then placed in an ice bath and stiπed for an hour, then stiπed at room temperature for an hour. The solvent is removed on the rotary evaporator, and the residue dissolved in methylene chloride. The methylene chloride solution is washed with water, and dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude material may be used without purification or purified by chromatography on silica gel. Osmium tetroxide olefin cleavage product of tris-O-triisopropylsilyl goldinonic acid (1 FIG. 30): Purified tris-O-triisopropylsilylgoldinonic acid (compound 1 FIG. 29) (785) (785 mg, 1 mmol) is dissolved in dioxane (30 ml). To this is added osmium tetroxide (26 mg, 0.1 mmol). The solution is stirred in the dark for 15 minutes, to allow the osmate ester to form. The solution is then diluted with 10 ml of deionized water, and to this, over several hours, in the dark, is added a solution of sodium periodate (504 mg, 2.3 mmol) dissolved in 15 ml of deionized water. The addition is performed in the dark and the progress of the reaction mixture is monitored by thin layer chromatography. When all of the starting material has reacted, chloroform (40 ml) is added and the reaction mixture is suction filtered through a sintered glass funnel. The salt filtrate is washed with methylene chloride. The combined filtrate and washings is washed with a solution of sodium sulfide (10%) in saturated aqueous sodium chloride, until the washings become free from precipitate or color. Then they is washed with saturated aqueous sodium chloride, and dried over sodium sulfate. The solvent is removed on the rotary evaporator, and the residue dried under vacuum. The crude material is used without further purification in the next step.

Reductive amination of olefin cleavage product of tris-O-triisopropylsilyl goldinonic acid (compound 2 FIG. 30): The aldehyde from the oxidative olefin cleavage of tris-O-triisopropylsilylgoldinonic acid (compound 1 FIG. 30) (747) (3.73g, 5 mmol) is dissolved in methanol, and concentrated ammonia solution is added to 5% of the total volume. The reaction mixture is placed in a fume hood and allowed to stand for one hour. Aqueous hydrochloric acid (IN) is added until the pH of an aliquot, when added to a 10 fold excess of water, is around 9 (measured by pH paper). Then sodium cyanoborohydride (63) (315 mg, 5 mmol) is added, and the reaction mixture is stiπed for an additional hour. Excess cyanoborohydride is destroyed by acidification of the reaction mixture with hydrochloric acid. The reaction mixture is degassed under aspirator vacuum in the fume hood, then the solvent is removed on the rotary evaporator, and the residue dissolved in chloroform. Saturated sodium chloride solution is added, the pH is adjusted to 7 (measured by pH paper), and the chloroform solution is washed with saturated aqueous sodium chloride. The solution is dried over sodium sulfate and the solvent removed on the rotary evaporator. The product may be purified by chromatography on silica gel or used directly in the next step. Addition product of (gamma-thiobutyrolactone to the amine adduct of the olefin cleavage product of tris-O-triisopropylsilyl goldinonic acid (compound 3 FIG. 30), (compound 4 FIG. 30): The amination product of the aldehyde from the olefin cleavage of tris-O-triisopropylsilylgoldinonic acid (compound 2 FIG. 30) (746) (1.492 g, 2 mmol) is dissolved in chloroform and (gamma-thiobutyrolactone (102) (204 mg, 2 mmol) and triethylamine (101) (202 mg, 2 mmol) is added. The reaction mixture is stirred at room temperature, until thin layer chromatography indicates that all of the starting material has reacted. The reaction mixture is washed with water, and dried over sodium sulfate. The solvent is removed on the rotary evaporator. The crude material is fully resilylated using the procedure for tris-O-triisopropylsilylgoldinonic acid (compound 1 FIG. 29). An aqueous workup should be sufficient to hydrolyze any silyl carboxyate ester or silyl thioether formed. The resulting material is purified by high performance liquid chromatography on silica gel.

N-Triisopropylsilyl-4-piperidone (compound 1 FIG. 31): Anhydrous 4-piperidone (99) (9.9 g, 100 mmol) is distilled from the hydrate under vacuum (0.1mm Hg), using a short path still, and collected in a tared receiver cooled in a dry ice bath. ^• The flask is then weighed. To its contents is added, for 9.9 g of distillate, a chilled solution of triethyl amine (101) (10.1 g, 100 mmol) and triisopropylsilyl chloride (193.5) (19.35 g, 100 mmol) in chloroform (200 ml). The mixture is stiπed in the receiving flask and allowed to warm to room temperature. The solvent is removed on the rotary evaporator and the product is purified by flash chromatography on silica gel.

4-Aceto-4-hvdroxypiperidine (compound 2 FIG. 31): N-triisopropylsilyl-4- piperidone (compound 1 FIG. 31) (256) (12.8 g, 50 mmol) is dissolved in dry tetrahydrofuran (200 ml), chilled in a dry ice bath. To this is added a mixture of tert- butylacetate (116) (5.8 g, 50 mmol) and fresh potassium tert-butoxide (112) (6.7 g, 60 mmol) in the same solvent, with rapid stirring. Stύring is continued and the solution is allowed to warm to room temperature. The disappearance of starting material is monitored by thin layer chromatography. When the reaction mixture is complete, water is added, the pH is adjusted to 9-10 (measured by pH paper) with aqueous hydrochloric acid (1 N), and the solvent is removed on the rotary evaporator. The residue is dissolved in chloroform, washed with water, then dried over sodium sulfate. The solvent is again removed under vacuum, and the product purified by chromatography on silica gel. The resulting material is warmed in aqueous-methanolic hydrochloric acid (I N) and sodium fluoride until thin layer chromatography indicates that the tert-butyl ester and the triisopropylsilylamine have been hydrolyzed, then all solvents is removed on the rotary evaporator, and the residue is dried under vacuum. This material is purified by chromatography on silica gel or by reverse phase chromatography on silanized silica gel. N-(4-Thiobutyryl)-4-aceto-4-hvdroxypiperidine (compound 3 FIG. 31): 4-Aceto-

4-hydroxypiperidine (compound 2 FIG. 31) (159) (7.95g, 50 mmol) is dissolved in dry chloroform (200 ml) and dry triethylamine (101) (5.05g, 50 mmol) is added. Then (gamma-thiobutyrolactone (102) (5.1 g, 50 mmol) is added. The progress of the reaction mixture is monitored by thin layer chromatography. When the starting material has reacted, water is added. The pH is adjusted to 3 (measured by pH paper) with hydrochloric acid (IN) and the chloroform solution is washed with this, then water. It is then extracted into sodium bicarbonate solution (0.1 N) and washed with ether. The aqueous layer is acidified with hydrochloric acid (IN) the product extracted into chloroform. The product is purified by chromatography on silica gel. N-(4-Thiobutyryl)-4-aceto-4-triisopropylsiloxypiperidine (compound 4 FIG. 31):

N-(4-Thiobutyryl)-4-aceto-4-hydroxypiperidine (compound 3 FIG. 31) is converted to the silyl ether by the same procedure used for tris-O-triisopropylsilylgoldinonic acid (compound 1 FIG. 29). An aqueous workup should be sufficient to hydrolyze any silyl carboxyate ester or silyl thioether formed. The resulting material is purified by chromatography on silica gel.

4-Dimethylaceto-4-hydroxypiperidine (compound 5 FIG. 31): This material is prepared from N-triisopropylsilyl-4-piperidone (compound 1 FIG. 31) and tert- butylvalerate by the procedure for 4-aceto-4-hydroxypiperidine (compound 2 FIG. 31) above. N-(4-Thiobutyryl)-4-dimethylaceto-4-hvdroxypiperidine (compound 6 FIG. 31):

This material is prepared from 4-dimethylaceto-4-hydroxypiperidine (compound 5 FIG. 31) by the procedure for N-(4-thiobutyryl)-4-aceto-4-hydroxypiperidine (compound 3 FIG. 31) above.

N-(4-Thiobutyryl)-4-dimethylaceto-4-triisopropylsiloxypiperidine (compound 7 FIG. 31): This material is prepared from N-(4-thiobutyryl)-4-dimethylaceto-4- hydroxypiperidine (compound 6 FIG. 31) by the procedure N-(4-thiobutyryl)-4-aceto-4- triisopropylsiloxypiperidine (compound 4 FIG. 31) above.

2-(Triisopropylsilyloxy)acetic acid methyl ester (compound 1 FIG. 32): Methylglycolate (90)(18 g, 0.2 mol) is dissolved in chloroform (200 ml) and treated with triispropylsilyl chloride (193.5)(38.7 g, 0.2 mol) and triethylamine (101) (20.2 g, 0.2 mol). The reaction mixture is monitored by thin layer chromatography. When the starting material has reacted, the solvent is removed under vacuum and the residue purified by flash chromatography on silica gel.

Bis-L5-triisopropylsilyloxy-2.4-dioxo-3.3-(S.S-cylo-l,2-dithioethyl)pentane (compound 2 FIG. 32): 2-(Triisopropylsilyloxy)acetic acid methyl ester (compound 1 FIG. 32) (247) (12.35 g, 0.05 mol) is dissolved in dry tetrahydrofuran (100 ml) under nitrogen in a 1 liter round bottom flask. This is cooled in a dry ice-acetone bath. In a second flask, 1,3-dithiolane (5.3 g, 0.05 mol) is dissolved in dry tetrahydrofuran (50 ml). This is also be cooled in a dry ice bath, under dry nitrogen. To this is added butyl lithium solution (0.053 mole butyl lithium in hexane) by means of a dry syringe. The solution is allowed to warm to around -40 degrees and stiπed for 2 hours while keeping it at this temperature. It is then added by syringe to the 2-(triisopropylsilyloxy)acetic acid methyl ester. The reaction mixture is stiπed for 2 hours at -78 degrees, then allowed to warm to room temperature and stiπed for an additional 1/2 hour. The reaction mixture is again cooled to around -40 degrees, and a second portion of butyl lithium solution (0.053 mol) is added. The reaction mixture is stiπed at this temperature for 2 hours, and then cooled to -78 degrees. A second portion of 2-(triisopropylsilyloxy)acetic acid methyl ester (12.35 g, 0.05 mol) is added. The reaction mixture is stirred for 2 hours at -78 degrees, then allowed to warm to room temperature, and stirred for an additional half hour. Water (400 ml) is then added to the reaction mixture, and sufficient chloroform is added to allow the formation of two phases. The reaction mixture is extracted with chloroform and the extracts is combined. The chloroform layer is washed with water, then dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The product is purified by column chromatography on silica gel.

1.5-Dihvdroxy-2,4-dioxo-3.3-(S.S-cylo-l,2-dithioethyl)pentane (compound 3 FIG. 32): Bis-l,5-triisopropylsilyloxy-2,4-dioxo-3,3-(S,S-cylo-l,2-dithioethyl)pentane (compound 2 FIG. 32) (536) (10.72 g, 20 mmol) is dissolved in methanol and treated with aqueous sodium fluoride (42) (4.2 g, 100 mmol). The reaction mixture is stiπed for 1 hour, then the solvent is removed on the rotary evaporator, and the residue dried under vacuum. The material is purified by chromatography on silica gel.

Bis-1.5-trifluoromethylsulfonato-2,4-dioxo-3.3-(S,S-cylo-1.2-dithioethyl)pentane (compound 4 FIG. 32): l,5-Dihydroxy-2,4-dioxo-3,3-(S,S-cylo-l,2-dithioethyl)pentane (compound 3 FIG. 32) (222) (4.44 g, 20 mmol) is dissolved in dry ether and slowly added to a solution of dry triethylamine (101) (4.04 g, 40 mmol) and trifluoromethylsulfonyl chloride (168.5) (6.75 g, 40 mmol) also dissolved in dry ether. The reaction mixture is stiπed at room temperature for 1 hour, then it is suction filtered through a glass fritted suction funnel to remove most of the triethylammonium chloride. The solvent is removed on the rotary evaporator and the residue dried thoroughly under vacuum. The crude material is used directly in the next step.

3.5-Dioxo-4.4-(S.S-cylo-1.2-dithioethyl)piperidine (compound 5 FIG. 32): Bis- 1 ,5-trifluoromethylsulfonato-2,4-dioxo-3,3-(S,S-cylo-l ,2-dithioethyl)pentane (compound 4 FIG. 32) (386) (3.86 g, 10 mmol) is dissolved in 1 liter of dry chloroform, in a dry 3 liter round bottom flask, under nitrogen. Anhydrous ammonia is condensed in a second, tared, round bottom flask, placed in a dry ice bath. This is then weighed, and dry chloroform, chilled in a dry ice bath, is added by means of a dry syringe chilled in powdered dry ice. An aliquot of this stock solution containing 10 mmol of ammonia is added to 1 liter of chilled, dry chloroform, then this is added to the flask containing the bis-1 ,5-trifluoromethylsulfonato-2,4-dioxo-3,3-(S,S-cylo-l ,2-dithioethyl)pentane. The reaction mixture is allowed to warm to room temperature and allowed to stand overnight. The next day, an aliquot is removed, concentrated, and assayed by thin layer chromatography. Should any starting material remain, additional ammonia is condensed, diluted, and 10 mmol of this is added directly by syringe to the rapidly stirred, chilled reaction mixture. When the reaction mixture is complete, the solvent is removed on the rotary evaporator, and the residue dried under vacuum. The residue is redissolved in chloroform, washed with saturated sodium chloride solution containing sodium hydroxide (0.1N), then with saturated sodium chloride solution. Then it is dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue again dried under vacuum. The product is purified by chromatography on silica gel or by crystallization of the hydrochloride salt.

3.5-Dihvdroxy-4,4-(S.S-cylo-l,2-dithioethyl)piperidine (compound 6 FIG. 32): 3,5-Dioxo-4,4-(S,S-cylo-l,2-dithioethyl)piperidine (compound 5 FIG. 32) (203) (2.03 g, 10 mmol) is dissolved in methanol (50 ml) and sodium borohydride (37) (740 mg, 20 mmol) is added. The reaction mixture is stiπed overnight. Aqueous hydrochloric acid is then added to destroy the excess borohydride. When gas evolution has ceased, the pH is adjusted to 9-10 (measured by pH paper) and the solvent is removed on the rotary evaporator. The residue is dissolved in chloroform, and washed with water saturated with sodium chloride. The solvent is again removed on the rotary evaporator and the crude material purified chromatography on silica gel or by crystallization of the hydrochloride salt. A similar reduction may be performed using lithium tris-sec-butylborohydride as the reducing agent.

N-Benzyl-bis-3,5-O-benzyl-4,4-(S.S-cylo- 2-dithioethyl)piperidine (compound 7 FIG. 32): 3,5-Dihydroxy-4,4-(S,S-cylo-l,2-dithioethyl)piperidine (compound 6 FIG. 32) (207) (2.07g, 10 mmol) is dissolved in anhydrous tetrahydrofuran or dioxane and sodium hydride (24) (790 mg, 33 mmol) or sodamide (39) (1.287 g, 33 mmol) is added, under dry nitrogen. Benzyl bromide (171) (5.64 g, 33 mmol) is then added slowly, with rapid stirring. The reaction mixture is stiπed at room temperature, and monitored by thin layer chromatography. After an hour, additional benzyl bromide (0.56 g, 3.3 mmol) and sodium hydride (79 mg, 3.3 mol) or sodamide (129 mg, 3.3 mol), is added, and stirring is continued. The process is continued until no further change is apparent. The reaction mixture is neutralized by the addition of hydrochloric acid (IN) and the solvent is removed on the rotary evaporator. The product is dissolved in chloroform and washed with aqueous sodium hydroxide (0.1 N), then water. The solution is dried over sodium sulfate, and the solvent removed on the rotary evaporator. The residue is purified by chromatography on silica gel or by crystallization of the hydrochloride salt.

N-Benzyl-bis-3.5-O-benzyl-4-piperidone (compound 8 FIG. 32): N-Benzyl-bis- 3,5-O-benzyl-4,4-(S,S-cylo-l,2-dithioethyl)piperidine (compound 7 FIG. 32) (477) (4.8 g, 10 mmol) is dissolved in a minimum volume of tetrahydrofuran. This is dropped over a period of 15 minutes into a rapidly stiπed mixture of red mercuric oxide (224) (4.48 g, 20 mmol) and boron trifluoride etherate (20 mmol) in 15% aqueous tetrahydrofuran (50 ml) under nitrogen. Stirring is continued for one half hour after the addition is complete. Ethyl ether (100 ml) is then added, and the precipitated salts removed by suction filtration through a sintered glass funnel. The ether layer is washed with saturated sodium bicarbonate solution then saturated sodium chloride solution. The ether solution is dried over sodium sulfate and the solvent removed on the rotary evaporator. The crude material is purified by column chromatography on silica gel or by crystallization of the hydrochloride salt.

N-Benzyl-bis-3,5-O-benzyl-4-aceto-4-hvdroxypiperidine (compound 9 FIG. 32): N-Benzyl-bis-3,5-O-benzyl-4-piperidone (compound 8 FIG. 32)(401) (4.01 g, 10 mmol) is dissolved in dry dioxane or tetrahydrofuran, chilled in an ice bath. To this is added a mixture of tert-butylacetate (116) (1.28 g, 11 mmol) and fresh potassium tert-butoxide (112) (1.23 g, 11 mmol) in the same solvent, with rapid stirring. Stirring is continued and the solution is allowed to warm to room temperature. The disappearance of starting material is monitored by thin layer chromatography. When the reaction mixture is complete, water is added, the pH is adjusted to 9-10 (measured by pH paper) with aqueous hydrochloric acid (1 N), and the solvent is removed on the rotary evaporator. The residue is dissolved in chloroform, washed with water, then dried over sodium sulfate. The solvent is again be removed under vacuum, and the product purified by chromatography on silica gel. The resulting material is dissolved in aqueous-methanolic hydrochloric acid (I N) and warmed until thin layer chromatography indicates that the tert-butyl ester has been hydrolyzed, then all solvents is removed on the rotary evaporator, and the residue is dried under vacuum. This material is dissolved in chloroform, extracted into aqueous sodium hydroxide (0.1N), and washed with toluene. Then the aqueous solution is neutralized with hydrochloric acid (IN), and the product extracted into chloroform. It is dried over sodium sulfate and the solvent is removed on the rotary evaporator. The residue is dried under vacuum. The crude material may be purified by chromatography on silica gel, by crystallization of the hydrochloride salt, or it may be used directly. 4-Aceto-3A5-trihvdroxypiperidine (compound 10 FIG. 32): Sodium (23) (1.4 g, 60 mmol) is placed in a dry, 3-neck 300 ml round bottom flask, under dry nitrogen and equipped with a magnetic stiπing bar, in a fume hood. The flask is placed in a dry ice bath and anhydrous ammonia (around 50 ml) is condensed into it. Into this solution is slowly dripped a solution of N-benzyl-bis-3,5-O-benzyl-4-aceto-4-hydroxypiperidine (compound 9 FIG. 32) (517) (5.17 g, 10 mmol) dissolved in tetrahdyrofuran and tert- butanol. The reaction mixture is rapidly stirred for an hour at dry ice temperatures. The progress of the reaction mixture is followed by quenching aliquots of the reaction mixture in water, neutralizing them with hydrochloric acid (0.1 N), saturating them with sodium chloride, extracting them into chloroform, then analyzing the material by thin layer chromatography. When no further change is apparent, the reaction mixture is allowed to warm to room temperature with continued rapid stirring. After the evaporation of the ammonia, the solution is stiπed at room temperature for an additional hour, then any remaining sodium is destroyed by the slow addition of ethanol. After hydrogen evolution has ceased, water is cautiously added, and the aqueous solution is extracted with toluene. It is then neutralized by the addition of aqueous hydrochloric acid (0.1 N), saturated with sodium chloride, and extracted into chloroform. The chloroform solution is washed with saturated sodium chloride solution, then dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The product may be purified by chromatography on silica gel, reverse phase chromatography on silanized silica gel, or crystallization of the hydrochloride salt.

N-(4-Thiobutyryl)-4-aceto-3.4,5-trihydroxypiperidine (compound 11 FIG. 32): 4- Aceto-3,4,5-trihydroxypiperidine (compound 10 FIG. 32) (191) (1.91g, 10 mmol) is dissolved in dry chloroform (200 ml) and dry triethylamine (101) (l.Olg, 10 mmol) is added. Then (gamma-thiobutyrolactone (102) (1.02 g, 10 mmol) is added. The progress of the reaction mixture is monitored by thin layer chromatography. When the starting material has reacted, saturated sodium chloride solution is added. The pH is adjusted to 3 (measured by pH paper) with hydrochloric acid (IN) and the chloroform solution is washed with this, then more saturated sodium chloride solution. It is then extracted into sodium bicarbonate solution (0.1 N) and washed with ether. The aqueous layer is acidified with hydrochloric acid (IN) and saturated with sodium chloride, and the product is extracted into chloroform. The product is purified by chromatography on silica gel. N-(4-Thiobutyryl)-4-aceto-3.4.5-tris-triisopropylsiloxypiperidine (compound 12 FIG. 32): The hydroxyl groups in N-(4-thiobutyryl)-4-aceto-3,4,5-trihydroxypiperidine (compound 11 FIG. 32) is converted to silyl ethers with triisopropylsilyl chloride by the same procedure used for tris-O-triisopropylsilylgoldinonic acid (compound 1 FIG. 29). An aqueous workup should be sufficient to hydrolyze any silyl carboxyate ester or silyl thioether formed. The resulting material is purified by chromatography on silica gel.

N-Benzyl-bis-3.5-O-benzyl-4-dimethylaceto-4-hvdroxypiperidine (compound 13 FIG. 32): This material is prepared from N-benzyl-bis-3,5-O-benzyl-4-piperidone (compound 8 FIG. 32) and tert-butylvalerate by the procedure for N-benzyl-bis-3,5-O- benzyl-4-aceto-4-hydroxypiperidine (compound 9 FIG. 32) above.

4-Dimethylaceto-3,4,5-trihvdroxypiperidine (compound 14 FIG. 32): This material is prepared N-benzyl-bis-3,5-O-benzyl-4-dimethylaceto-4-hydroxypiperidine (compound 13 FIG. 32) by the procedure 4-aceto-3,4,5-trihydroxypiperidine (compound 10 FIG. 32) above. N-(4-Thiobutyryl)-4-dimethylaceto-3.4.5-trihvdroxypiperidine (compound 15

FIG. 32): This material is prepared from 4-dimethylaceto-3,4,5-trihydroxypiperidine (compound 14 FIG. 32) and (gamma-thiobutyrolactone by the procedure for N-(4- thiobutyryl)-4-aceto-3,4,5-trihydroxypiperidine (compound 11 FIG. 32) above.

N-(4-Thiobutyryl)-4-dimethylaceto-3.4.5-tris-triisopropylsiloxypiperidine (compound 16 FIG. 32): This material is prepared from N-(4-thiobutyryl)-4- dimethylaceto-3,4,5-trihydroxypiperidine (compound 15 FIG. 32) and triisopropylsilyl chloride by the procedure for N-(4-thiobutyryl)-4-aceto-3,4,5-tris- triisopropylsiloxypiperidine (compound 12 FIG. 32) above.

Procedure for coupling goldinamine to anilines through a urea (carbonyl) linkage: As shown in FIG. 33, 1-N-desmethylgoldinamine formate (compound 1 FIG. 27) (546 mg, 1 mmol) is dissolved in dry chloroform (10 ml). Triethylamine (101) (101 mg, 1 mmol) is added, followed by carbonyldiimadazole (162) (810 mg, 5 mmol), also dissolved in chloroform (10 ml). The reaction mixture is stirred for 1 hour at room temperature, then the solvent is removed on the rotary evaporator. The residue is washed with dry ethyl ether and dry ethyl acetate, to remove the unreacted carbonyldiimadazole, then it is dried under vacuum. The residue is dissolved in dry chloroform and the aniline derivative (1 mmol) is added. The reaction mixture is stiπed for an additional hour. It is then extracted with water and dried over sodium sulfate. The solvent is removed under vacuum to yield a crude residue that is washed with ethyl ether and ethyl acetate. This is purified by reverse phase high performance liquid chromatography on silica gel with a linear gradient of water-acetonitrile or acetone as the eluant. 2-Aminoresorcinol (compound 1 FIG. 34): 2-Nitrororesorcinol (155) (31 g, 0.2 mol) is dissolved in methanol and placed in a hydrogenation bottle. To this is added 10% Pd on charcoal (1 gm). The bottle is sealed, evacuated, and hydrogen gas is applied at a pressure of 50 psi. The reaction mixture is shaken at room temperature until no further hydrogen uptake is apparent. The reaction mixture is removed, filtered, and the solvent removed by evaporation to yield a solid. The crude material is used without further purification.

2-Acetylaminoresorcinol (compound 2 FIG. 34): The crude 2-aminoresorcinol (compound 1 FIG. 34) (125) (12.5 g, 0.1 mol) is dissolved in chloroform (200 ml), and triethyl amine (101) (10.1 g, 0.1 mmol) is added. To this is added acetic anhydride (102) (10.2 g, 0.1 mol). The reaction mixture is stiπed for 3 hours, then the solvent is removed on the rotary evaporator. The residue is dissolved in chloroform and washed with hydrochloric acid (0.1 N) once, then with water. The chloroform solution is dried over sodium sulfate. The solvent is again removed on the rotary evaporator and the residue dried under vacuum. The crude material is used without further purification. N.N-dimethyl-(3,5-dihvdroxy-4-acetylamino)-benzylamine (compound 3 FIG.

34): The crude 2-acetylaminoresorcinol (compound 2 FIG. 34) (167) (8.35g, 50 mmol) is carefully mixed with 37% aqueous formaldehyde (8.1 ml, 100 mmol) and 33% aqueous dimethylamine (13.5 ml, 100 mmol). The reaction mixture is allowed to stand at room temperature and is periodically analyzed by thin layer chromatography. It is heated, as necessary, until all of the 2-acetylaminoresorcinol has reacted. The volatile materials are then removed under vacuum. The residue is dissolved in chloroform and water is added. The pH is adjusted to 9 (measured by pH paper), and the chloroform layer is washed with this, then again with water. The product is then extracted into aqueous hydrochloric acid (0.1 N). The aqueous layer is washed with chloroform then the pH is adjusted to 9-10 (measured by pH paper) by the addition of aqueous sodium hydroxide (0.1 N) and the product extracted into chloroform. The solution is dried over sodium sulfate, the solvent is removed on the rotary evaporator, and the residue dried under vacuum. The product is purified by chromatography on silica gel or by crystallization of the hydrochloride salt. 3-(3.5-dihvdroxy-4-acetylamino)-thiobenzyl-l-thiopropane (compound 4 FIG. 34): The crude N,N-dimethyl-(3,5-dihydroxy-4-acetylamino)-benzylamine (compound 3 FIG. 34) (224) (5.6 g, 25 mmol) is dissolved in dimethylformamide (50 ml) in a round bottom flask equipped with a reflux condenser, and a water filled gas-bubbler trap. To this, 1,3-dithiopropane (108) (21.6 g, 200 mmol) is added. The reaction mixture is heated in an oil bath to 150 degrees. At intervals, aliquots of the water in the gas-bubbler trap is removed and titrated to determine the amount of dimethylamine evolved by the reaction. Aliquots of the reaction mixture also is removed and assayed by thin layer chromatography (silica gel plates, methylene chloride-hexane as the eluant) to follow the loss of N,N-dimethyl-(3,5-dihydroxy-4-acetylamino)-benzylamine. When dimethylamine evolution has ceased, the reaction mixture is cooled and the solvents removed under vacuum. The residue is purified by chromatography on silica gel. 4-(3.5-dihvdroxy-4-amino)-thiobenzyl-l-thiopropane (compound 5 FIG. 34): To a solution of 4-(3,5-dihydroxy-4-acetylamino)-thiobenzyl-l-thiopropane (compound 4 FIG. 34) (303) (3.03 g, 10 mmol) in methanol (50 ml) is added hydrazine (28) (0.56 g, 20 mmol). The solution is allowed to stand overnight. The solvent is then removed on the rotary evaporator and the residue dissolved in chloroform. It is washed with water, dried over sodium sulfate, and the solvent again removed on the rotary evaporator. The residue is purified by chromatography on silica gel.

2-Thiobenzylethylamine (compound 1 FIG. 35): 2-Thioethylamine (76) (15.2 g, 0.2 mol) is dissolved in chloroform (200 ml) in a round bottom flask. Into the flask is placed finely ground, anhydrous sodium carbonate (106) (21.2 g, 0.2 mol). The suspension is rapidly stiπed and a solution of benzyl bromide (171) (34.2 g, 0.2 mol) in chloroform (200 ml) is added dropwise. The reaction mixture is assayed by thin layer chromatography and stining is continued until all of the benzyl bromide has reacted. The reaction mixture is then suction filtered and added to water in a separatory funnel. The pH of the water layer is adjusted to 9-10 (checked by pH paper) and the chloroform layer is washed once with this, then twice with water. It is dried over sodium sulfate. The solvent is removed on the rotary evaporator, and the residue dried under vacuum. The crude material is used without further purification. N-methyl-2-thiobenzylethylamine (compound 2 FIG. 35): 2- Thiobenzylethylamine (compound 1 FIG. 35) (167) (33.4 g, 0.2 mol) and freshly distilled formaldehyde (30) (9 g, 0.20 mol) is mixed. A minimum volume of absolute ethanol may be added to assist dissolution. To this is added titanium (IV) isopropoxide (0.25 mol). The mixture is stirred for an hour under nitrogen in a fume hood at room temperature. The solution is diluted with absolute ethanol (200 ml), then sodium cyanoborohydride (63) (8.2 g, 0.13 mmol) is carefully added and the reaction mixture is stiπed overnight. Water (20 ml) is then added, and the resulting inorganic precipitate is removed by suction filtration through a glass fritted funnel. The precipitate is washed with absolute ethanol, and the volatile materials is removed from the combined filtrate and washings under vacuum. The crude material is dissolved in chloroform, filtered again, washed with sodium hydroxide solution (0.1 N), extracted into hydrochloric acid solution (0.1 N), then washed with chloroform. The aqueous solution is again made basic with sodium hydroxide, and the product is extracted into chloroform. The crude material may be purified by flash chromatography on silica gel or by fractional crystallization of the hydrochloride salt

N-methyl-N-(2-thiobenzylethyl)-3,5-dihvdroxy-4-acetylaminobenzylamine (compound 3 FIG. 35): A portion of 2-acetylaminoresorcinol (compound 1 FIG. 34) (167) (8.35g, 50 mmol) is carefully mixed with 37% aqueous formaldehyde (8.1 ml, 100 mmol) and N-methyl-2-thiobenzylethylamine (compound 2 FIG. 35) (181) (9.05 g, 50 mmol). The reaction mixture is allowed stand to at room temperature and is periodically analyzed by thin layer chromatography. It is heated, as necessary, until all of the 2-acetylaminoresorcinol has reacted. The volatile materials is then removed under vacuum. The residue is dissolved in chloroform, washed with water, extracted into aqueous hydrochloric acid (0.1 N), and washed with chloroform. The pH of the aqueous layer is adjusted to 10 (measured by pH paper) by the addition of aqueous sodium hydroxide (0.1 N) and the product is extracted into chloroform. The solvent is removed under vacuum and the residue purified by chromatography on silica gel or fractional crystallization of the hydrochloride salt. N-Methyl-N-(2-thioethyl)-3.5-dihydroxy-4-acetylaminobenzylamine (compound 4

FIG. 35): Sodium (23) (2.3 g, 100 mmol) is placed in a dry, 3-neck 500 ml round bottom flask, equipped with a magnetic stirring bar, under dry nitrogen, in a fume hood. The flask is placed in a dry ice bath and anhydrous ammonia (around 100 ml) is condensed into it. Into this solution is slowly dripped a solution of N-methyl-N-(2-thiobenzylethyl)-3,5- dihydroxy-4-acetylaminobenzylamine (compound 3 FIG. 35) (344) (17.2 g, 50 mmol) dissolved in tetrahydrofuran and tert-butanol. The reaction mixture is rapidly stiπed for an hour at dry ice temperatures. The progress of the reaction mixture is followed by quenching aliquots of the reaction mixture in water, neutralizing them with hydrochloric acid (0.1 N), saturating them with sodium chloride, extracting them into chloroform, then analyzing material by thin layer chromatography. When all of the starting material has reacted, the reaction mixture is allowed to warm to room temperature with continued rapid stining. After the evaporation of the ammonia, the solution is stirred at room temperature for an additional hour, then any remaining sodium is destroyed by the slow addition of ethanol. After hydrogen evolution has ceased, methanol, then water is cautiously added with rapid stirring. The pH is adjusted to 9 (measured by pH paper) by the addition of hydrochloric acid (IN). The reaction mixture is extracted into chloroform, then washed with water. The product is extracted into hydrochloric acid (0.1 N), washed with chloroform, then the pH of the aqueous solution is adjusted to 9-10 (checked with pH paper) with sodium hydroxide solution (0.1N) and the product is extracted into chloroform. The chloroform solution is dried over sodium sulfate, and the solvent removed on the rotary evaporator. The residue is dried under vacuum. It may be purified by chromatography on silica gel or by crystallization of the hydrochloride salt.

N-Methyl-N-(2-thioethyl)-3.5-dihydroxy-4-aminobenzylamine (compound 5 FIG. 35): N-Methyl-N-(2-thioethyl)-3,5-dihydroxy-4-acetylaminobenzylamine (compound 4 FIG. 32) (254) (6.35 g, 25 mmol) is dissolved in methanol (100 ml) and hydrazine (32) (3.2 g, 100 mmol) is added. The reaction mixture is allowed to stir overnight, then the solvent is removed on the rotary evaporator. The residue is dissolved in chloroform, washed with water, then dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue chromatographed on silica gel with hexane-chloroform- triethylamine as the eluant.

N-Methyltaurine (compound 6 FIG. 35): In a 500 ml, 3 neck round bottom flask, taurine (125) (25.0 g, 0.2 mol) is dissolved in a minimum volume of methanol.

Triethylamine (101) (1.0 g, 0.01 mmol) and freshly distilled formaldehyde (30) (6.6 g, 0.22 mol) is added. The mixture is stiπed in a fume hood at room temperature for an hour, then sodium cyanoborohydride (63) (8.2 g, 0.13 mmol) in methanol (100 ml) is carefully added using an addition funnel. The reaction mixture is stined for an additional hour. The excess cyanoborohydride is destroyed by acidification with hydrochloric acid (0.1 N) in the fume hood, and the reaction mixture briefly degassed under an aspirator vacuum in the fume hood. The solvent is then removed on the rotary evaporator. The crude residue is dissolved in chloroform, washed with saturated sodium chloride-sodium bicarbonate (0.1 N) solution, then dried over sodium sulfate. It may be purified by flash chromatography on silica gel or by fractional crystallization of the hydrochloride salt. N-methyl-N-(2-sulfonoethyl)-3.5-dihydroxy-4-acetylaminobenzylamine (compound 7 FIG. 35): A portion of 2-acetylaminoresorcinol (compound 1 FIG. 34)

(167) (8.35 g, 50 mmol) is carefully mixed with 37% aqueous formaldehyde (8.1 ml, 100 mmol) and N-methyltaurine (compound 6 FIG. 32) (139) (6.95 g, 50 mmol). The reaction mixture is allowed stand to at room temperature and is periodically analyzed by thin layer chromatography. It is heated, as necessary, until all of the 2-acetylaminoresorcinol has reacted. The volatile materials is then removed under vacuum. The residue is dissolved in chloroform, water is added and the pH adjusted to neutrality (checked by pH paper). The aqueous layer is saturated with sodium chloride, and the reaction mixture is washed with saturated sodium chloride, then extracted into aqueous hydrochloric acid (0.1 N), and washed with ethyl acetate. The aqueous layer is made neutral (to pH 7 measured by pH paper) by the addition of aqueous sodium hydroxide (0.1 N). It is saturated with sodium chloride, and the product extracted into chloroform. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude product may be purified by chromatography on silica gel, reverse phase chromatography on silanized silica gel, or by fractional crystallization. N-Methyl-N-(2-thioethyl)-3 ,5-dihydroxy-4-acetylaminobenzylamine (compound 4

FIG. 35): N-Methyl-N-(2-sulfonoethyl)-3,5-dihydroxy-4-acetylaminobenzylamine (compound 7 FIG. 32) (346) (8.65 g, 25 mmol) is dissolved in anhydrous diglyme (100 ml) under nitrogen in a dry, 3 neck round bottom flask. The flask is cooled in an ice bath and the solution rapidly stiπed. Lithium aluminum hydride (38) (0.95 g, 25 mmol) is added in small portions with rapid stirring. Sti ing is continued and the solution is allowed to warm to room temperature. Progress of the reaction mixture is monitored by thin layer chromatography. When all of the starting material has reacted, the excess lithium aluminum hydride is destroyed by cautious addition of methanol, then water. The pH is adjusted to around 9 (measured by pH paper) and chloroform and sufficient water to form two phases is added. The aqueous layer is extracted with chloroform. The chloroform extract is washed with water and dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude product may be purified by chromatography on silica gel or by fractional crystallization of the hydrochloride salt.

N-methyl-N-(2-thiobenzylethyl)-4-acetylaminobenzylamine (compound 8 FIG. 35): This material is prepared from N-methyl-2-thiobenzylethylamine (compound 2 FIG. 35) and acetylaniline by the method used to prepare N-methyl-N-(2-thiobenzylethyl)-3,5- dihydroxy-4-acetylaminobenzylamine (compound 3 FIG. 35).

N-methyl-N-(2-thioethyl)-4-aminobenzylamine (compound 10 FIG. 35): N- Methyl-N-(2-thiobenzylethyl)-4-acetylaminobenzylamine (compound 8 FIG. 35) (312) (15.6 g, 50 mmol) is dissolved in methanol (100 ml) and hydrazine (32) (3.2 g, 100 mmol) is added. The reaction mixture is allowed to stir overnight, then the solvent is removed on the rotary evaporator. The residue is dissolved in chloroform, washed with water, then dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue thoroughly dried under vacuum. The benzyl group of the crude material is then cleaved by the method used to prepare N-methyl-N-(2-thioethyl)-3,5-dihydroxy-4- acetylaminobenzylamine (compound 4 FIG. 35) above.

N-Methyl-N-(2-sulfonoethyl)-4-acetylaminobenzylamine (compound 11 FIG. 35): This material is prepared from N-methyltaurine (compound 6 FIG. 35) and acetylaniline by the method used to prepare N-methyl-N-(2-sulfonoethyl)-3,5-dihydroxy-4- acetylaminobenzylamine (compound 7 FIG. 35) above. N-Methyl-N-(2-thioethyl)-4-acetylaminobenzylamine (compound 9 FIG. 35):

This material is prepared from N-methyl-N-(2-sulfonoethyl)-4-acetylaminobenzylamine (compound 11 FIG. 35) by the reaction used with N-methyl-N-(2-sulfonoethyl)-3,5- dihydroxy-4-acetylaminobenzylamine (compound 7 FIG. 35) above.

Hydrazone of kiπomycin with 3-nitro-4-hvdrazidophenylthioethanol (compound 1 FIG. 36): Hydrazones of krπomycin are prepared by the method of Chinali, G, et al, Bollettino Societa Italiana Biologia Sperimentale 57:1706-12 (1981). Thus, kiπomycin sodium salt (830) (0.83 g, 1 mmol) is dissolved in a minimal volume of absolute ethanol. 3-Nitro-4-hydrazidophenylthioethanol (compound 5 FIG. 37) (0.22 g, 1 mmol), dissolved in the same solvent, is then added. The reaction mixture is stiπed at room temperature overnight, then the solvent removed under vacuum. The residue is washed with ethyl ether and ethyl acetate, then dried under vacuum. The product may purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile or acetone as the eluant, or it may be crystallized.

3-nitro-4-aminophenylethanol (compound 1 FIG. 37): 4-Aminophenylethanol (151) (30.2 g, 0.2 mol) is dissolved in dichlorobenzene or nitrobenzene (200 ml) and concentrated nitric acid (63) (0.44 mol) is dripped in slowly, with rapid stirring. The reaction mixture is cooled in an ice bath, as necessary. The reaction mixture is allowed to stir, then ice water is added. The water layer is separated and saved and the reaction mixture is extracted twice with water. The combined aqueous extracts is made basic with sodium hydroxide (IN) and the product is extracted into chloroform. This is washed twice with water and dried over sodium sulfate. The solvent is removed on the rotary evaporator, and the residue dried under vacuum. The product may be purified by chromatography on silica gel or crystallization of the hydrochloride salt.

3-nitro-4-hydrazidophenylethanol (compound 2 FIG. 37): Methylsulfonyl chloride (114.5) (22.9 g, 0.2 mol) is dissolved in dry chloroform (200 ml). To this, with cooling and rapid stirring, is added dry triethylamine (101) (20.2 g, 0.2 mol). This mixture is slowly added to a cooled, rapidly stirred solution of hydroxylamine hydrochloride (69.5) (13.9 g, 0.2 mol) also dissolved in chloroform (200 ml). The reaction mixture is stiπed at room temperature for 1 hour, then slowly added to a solution of 3-nitro-4-aminophenylethanol (compound 1 FIG. 37) (182) (36.4 g, 0.2 mmol) and triethylamine (101) (40.4 g, 0.4 mol), dissolved in dry chloroform (200 ml). The reaction mixture is allowed to stir overnight. Water is added to the reaction mixture and the pH is adjusted to alkaline, as necessary. The chloroform layer is washed with water and dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude product may be purified by column chromatography, or used directly in the next step. 3-nitro-4-acetylhydrazidophenylethanol (compound 3 FIG. 37): The crude 3- nitro-4-hydrazidophenylethanol (compound 2 FIG. 34) is dissolved in chloroform (200 ml) and dry triethylamine (101) (20.2 g, 0.2 mmol) is added. Acetic anhydride (102) (20.4 g, 0.2 mmol) is dripped into this solution, with rapid stining and cooling. The reaction mixture is stined for one hour. Water is added, the pH is adjusted to 5-6 with hydrochloric acid (0.1N), and the chloroform layer is washed twice with hydrochloric acid (0.1N), then with water. It is then dried over sodium sulfate, the solvent is removed on the rotary evaporator, and the residue dried under vacuum. The product is purified by column chromatography on silica gel or by crystallization.

3-nitro-4-acetylhydrazidophenylthioethanol (compound 4 FIG. 37): 3-Nitro-4- acetylhydrazidophenylethanol (compound 3 FIG. 34) (255) (25.5 g, 0.1 mol) and dry triethylamine (101) (10.1 g, 0.1 mol) is dissolved in chloroform (200 ml). The solution is rapidly stiπed and methyl sulfonyl chloride (114.5) (11.45 g, 0.1 mol) in dry chloroform (50 ml) is added by means of a dropping funnel. The reaction mixture is stiπed for one hour, then finely powdered solid sodium sulfide (78) (39 g, 0.5 mol), a solution of sodium sulfide (39 g, 0.5 mol) in a minimal volume of water, or tetrabutylammonium sulfide (242) (24.2 g, 0.1 mol) is added. The reaction mixture is stiπed overnight, then washed with water. The chloroform solution is dried over sodium sulfate, then the solvent is removed on the rotary evaporator. The residue is dried under vacuum. The crude product may be purified by flash chromatography or used directly in the next step.

3-Nitro-4-hydrazidophenylthioethanol (compound 5 FIG. 37): The crude 3-nitro- 4-acetylhydrazidophenylthioethanol (compound 4 FIG. 37) is dissolved in ethanol and hydrazine, (32) (16 g, 0.5 mol) is added. The reaction mixture is stiπed at room temperature overnight. It is then washed with water. The solution is dried over sodium sulfate, the solvent removed on the rotary evaporator, and the residue dried under vacuum. The product may be purified by chromatography on silica gel or by crystallization of the hydrochloride salt. 3-nitro-4-aminophenylacetic acid (compound 6 FIG. 37): 4-Aminophenylacetic acid (165) (33.0 g, 0.2 mol) is dissolved in dichlorobenzene or nitrobenzene (200 ml) and concentrated nitric acid (63) (0.44 mol) is dripped in slowly, with rapid stirring. The reaction mixture is cooled in an ice bath, as necessary. The reaction mixture is allowed to stir, then ice water is added. The water layer is separated and the reaction mixture extracted once again with water. The combined aqueous extracts is made neutral with sodium hydroxide (IN), saturated with sodium chloride, and the product is extracted into chloroform. This is washed twice with saturated sodium chloride solution and dried over sodium sulfate. The solvent is removed on the rotary evaporator, and the residue dried under vacuum. The product may be purified by chromatography on silica gel or crystallization.

3-nitro-4-hvdrazidophenylacetic acid (compound 7 FIG. 37): Methylsulfonyl chloride (114.5) (22.9 g, 0.2 mol) is dissolved in dry chloroform (200 ml). To this, with cooling and rapid stirring, is added dry triethylamine (101) (20.2 g, 0.2 mol). This mixture is slowly added to a cooled, rapidly stiπed solution of hydroxylamine hydrochloride (69.5) (13.9 g, 0.2 mol) also dissolved in chloroform (200 ml). The reaction mixture is stiπed at room temperature for one hour, then slowly added to a solution of 3-nitro-4-aminophenylacetic acid (compound 6 FIG. 37) (182) (36.4 g,

0.2mmoι) and triethylamine (101) (40.4 g, 0.4 mol), dissolved in dry chloroform (200 ml). The reaction mixture is allowed to stir overnight. Water is added and the pH adjusted to neutrality, as necessary, then the water is saturated with sodium chloride. The solution is washed with saturated sodium chloride solution and dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The crude product may be purified by chromatography or by crystallization.

Sulfonate esters of kiπomycin are prepared by the method of Maehr, H, et al, Journal of Antibiotics 32:361-367 (1978). Thus, kinomycin sodium salt, (2.2g, 2.65 mmol) is dissolved in dimethylformamide (25 ml) and cooled in an ice bath. The sulfonyl chloride (3 mmol) is added and the reaction mixture is stiπed vigorously for 30 minutes. After this, the reaction mixture is quenched by adding it to a well stiπed mixture of saturated sodium bicarbonate (100 ml) and chloroform (25 ml). After 20 minutes of stirring, the mixture is transfeπed to a separatory funnel containing chloroform (25 ml) and the phases is equilibrated. The aqueous layer is discarded and the chloroform layer is washed once with saturated sodium bicarbonate and twice with water, then dried over sodium sulfate. The solution is concentrated on a rotary evaporator at 55 degrees to yield an oily residue that is dissolved in acetone and purified by reverse phase liquid chromatography (on a Sephadex LH-20 column with acetone as the eluant). 4-Maleimidomethylenephenylsulfonic acid (compound 1 FIG. 38): 4- Aminomethylenephenylsulfonic acid (187) (18.7 g, 0.1 mol) is dissolved in chloroform and triethylamine (101) (30.3 g, 0.3 mol) is added. The solution is cooled in an ice bath. Maleyldichloride (153) (15.3 g, 0.1 mol) is added dropwise. The solution is stiπed for an hour then allowed to warm to room temperature. Ice water is added, and acidified with hydrochloric acid (IN). The chloroform layer is washed with hydrochloric acid (0.1 N), then with water, then it is dried over sodium sulfate. The solvent is removed on the rotary evaporator and the residue dried under vacuum. The product may be purified by reverse phase chromatography on silanized silica gel or by crystallization.

4-Maleimidomethylenephenylsulfonyl chloride (compound 2 FIG. 38): 4- Maleimidomethylenephenylsulfonic acid (compound 1 FIG. 38) (269) (26.9 g, 0.1 mol) is dissolved in chloroform and sulfonyl chloride (119) (23.8 g, 0.2 mol) is added. The reaction mixture is refluxed until gas evolution has ceased, then the volatile materials is removed on the rotary evaporator. The residue is dissolved in chloroform, and the chloroform again removed on the rotary evaporator. The residue is dried under vacuum. It may be purified by crystallization.

5.14. EXAMPLE 14: CONJUGATION OF KIRROMYCIN. 3-NITRO-4-HYDRAZIDOPHENYLTHIOETHANOL. AND

TETRACYCLINE DERIVATIVE

The amino derivative of tetracycline (compound 7 FIG. 24) (0.59 g, lmmol) is dissolved in methanol, methanol-chloroform, or dimethylethylene glycol and succinimidyl 6-[(iodoacetyl)-amino]hexanoate (0.40 g, 0.1 mmol) is added. The reaction mixture is stiπed at room temperature, and the progress of the reaction mixture is monitored by thin layer chromatography. After the reaction mixture is complete, the hydrazone of kiπomycin and 3-nitro-4-hydrazidophenylthioethanol (compound 1 FIG. 36) (0.97 g, lmmol) is added and the reaction stirred further at room temperature. Progress of the reaction will again be monitored by thin layer chromatography. When the reaction mixture is complete, the solvent is removed under vacuum with minimal heating, and the residue is dried under vacuum. The product is purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile as the eluant, or it may be crystallized. The crude kiπomycin hydrazone may also be used in the coupling reaction.

5.15. EXAMPLE 15: CONJUGATION OF KIRROMYCIN.

3-NITRO-4-HYDRAZIDOPHENYLTHIOETHANOL. AND TRIMETHOPRIM OR TETRACYCLINE DERIVATIVES The thiol compound, such as the trimethoprim derivative (compound 5 FIG. 22) (0.35 g, 0.1 mmol), is dissolved in chloroform or chloroform-methanol and N,N'- hexamethylene-bis(iodoaetamide) (3.52 g, 1 mmol) added. The reaction mixture is stiπed at room temperature, and the progress of the reaction mixture is monitored by thin layer chromatography. When the reaction mixture is complete, the solvent is removed under vacuum with minimal heating, and the residue is dried under vacuum. The residue is purified from excess N,N'-hexamethylene-bis(iodoaetamide) by chromatography. To this material is added the hydrazone of krrromycin and 3-nitro-4-hydrazidophenylthioethanol (compound 1 FIG. 36) (0.97 g, lmmol) which is dissolved in methanol- dimethylformamide. The reaction mixture is stiπed at room temperature, and the progress of the reaction mixture is monitored by thin layer chromatography. When the reaction mixture is complete, the solvent is removed under vacuum with minimal heating, and the residue is dried under vacuum. The product is purified by reverse phase high performance liquid chromatography with a linear gradient of water-acetonitrile as the eluant, or it may be crystallized. The crude kiπomycin hydrazone may also be used in the coupling reaction.

5.16. EXAMPLE 16: DETERMINING THE DEGREE OF

CELLULAR ACCUMULATION OF VLP CONJUGATES Testing of the degree of cellular accumulation of model conjugates of a vector with a probe molecule is performed with representative, non-pathogenic microorganisms using the following methods. These measurements constitute a preliminary screening procedure.

Serial dilutions of conjugate are made in water or the appropriate organic solvent. Aliquots of these are mixed with portions of freshly grown yeast or fungal cultures (5-10 ml) that have been placed in culture tubes. The tubes are shaken in an incubator. At intervals of 10 minutes to several hours, aliquots of the cultures are withdrawn. The cellular material is collected by centrifugation, washed with media, then recentrifuged. The resulting pellet is analyzed for the presence of conjugate molecule. As a control, the above procedure is performed with an excess of unconjugated vector molecule that is added to the culture medium at the same time as the conjugate of the vector with the probe. Another control uses serial dilutions of the probe molecule alone or in combination with the linker. For analysis of model VLP conjugates with fluorescent probes, the washed cellular pellet is suspended in water or culture medium. The suspension is illuminated by ultraviolet light and the degree of fluorescence estimated by visual inspection. Aliquots of the suspension may also be analyzed in a fluorimeter, for a more precise spectrophotometric determination of the degree of fluorescence.

For radioactive probes, the cellular pellet is suspended in distilled water, and a portion of the material aliquoted into scintillation fluid and dispersed. Radioactivity (¹⁴C, ³H, or ³⁵S) in the resulting suspension is determined by scintillation counting.

The cells may also be lysed, and fractionated, and the presence of the probe determined in various cellular fractions.

Fungi or yeast, taken from stored, frozen cultures, are grown in culture tubes or shake flasks at 25 °C. Aliquots of around 10⁵ colony forming units per milliliter is then used as inocula for the assays. Cultures are incubated at 25 °C, for up to 48 hours with yeasts and 120 hours with filamentous fungi, to obtain an adequate cell density. Using the methods described above, the degree of accumulation of fluorescent or radiolabeled model conjugates of a vector, a linker, and a probe molecule can be determined at a series of concentrations of the conjugate. In parallel, a series of concentrations of the conjugate plus excess free vector is also tested, and the results graphed.

The embodiments of the invention described above are intended to be merely exemplary, and those skilled in the art will recognize, or will be able to ascertain using no more than routine experimentation, numerous equivalents of the specific materials, procedures, and devices described herein. All such equivalents are considered to be within the scope of the invention and are encompassed by the appended claims.

Claims

THE CLAIMSWhat is claimed is:

1. A method of designing a VLP conj ugate for use in the treatment or prevention of a disease or condition in a patient, which comprises: selecting a pharmacophore that can affect a first target associated with the disease or condition, and that has a first affinity for the first target; selecting a vector that has a second affinity for a second target likely to be located near the first target; and selecting a linker to which the first pharmacophore and the first vector can both be covalently attached to provide the VLP conjugate; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in the patient is greater than that of the first target.

2. The method of claim 1 which further comprises covalently binding the pharmacophore and vector to the linker to provide a conjugate, and testing the ability of the conjugate to affect the first target.

3. The method of claim 1 wherein the pharmacophore is selected using at least one criterion selected from the group which includes, but is not limited to: mode of action; target of action; molecular weight; solubility; types and/or severities of adverse effects; therapeutic index; chemical stability; presence of chemically reactive moieties; and structure-activity relationship data.

4. A method of designing a VLP conjugate of a pharmacophore having a first target, and a first affinity for the first target, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; and selecting a linker to which the first pharmacophore and the first vector can both be covalently attached to provide the VLP conjugate; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in a patient to whom the VLP conjugate could be administered is greater than that of the first target.

5. The method of claim 4 which further comprises covalently binding the pharmacophore and vector to the linker to provide a conjugate, and testing the ability of the conjugate to affect the first target.

6. A method of improving the delivery of a pharmacophore to a first target located in, on, or near a cell, wherein the pharmacophore has a first affinity for the first target, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; selecting a linker; covalently binding the pharmacophore and the vector to the linker to provide a conjugate; testing the ability of the conjugate to concentrate near the first target; and repeating the process with a different vector if the ability of the conjugate to concentrate near the first target is less than the ability of the pharmacophore alone to concentrate near the first target; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in, on, or near the cell is greater than that of the first target.

7. The method of claim 6 wherein information provided by the testing is used in the selection of a different vector if the process is repeated.

8. A method of targeting a pharmacophore having an affinity for a first target in vivo, which comprises: chemically linking the pharmacophore and a vector to a linker to provide a VLP conjugate; and administering the VLP conjugate to a host; wherein the vector can associate with a second target with a dissociation constant of less than about 10^"6, the second target is different from the first target, and the second target is located within 10^"4 meters of the first target.

9. The method of claim 8 wherein the second target is present in the host in a concentration of greater than 10 times that of the first target.

10. A method of improving the therapeutic index of a pharmacophore having a first target, a first affinity for the first target, and a first therapeutic index, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; selecting a linker; covalently binding the pharmacophore and the vector to the linker to provide a conjugate which has a second therapeutic index; testing the conjugate to determine the second therapeutic index; and repeating the process if the second therapeutic index is less than the first therapeutic index; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in a patient to whom the VLP conjugate could be administered is greater than that of the first target.

11. The method of claim 10 wherein information provided by the testing is used in the selection of a different vector if the process is repeated.

12. A method of decreasing the toxicity of a pharmacophore having a first affinity for a first target and a first toxicity index, which comprises: selecting a vector that has a second affinity for a second target likely to be located near the first target; selecting a linker; covalently binding the pharmacophore and the vector to the linker to provide a conjugate which has a second toxicity index; testing the conjugate to determine the second toxicity index; and repeating the process if the second toxicity index is greater than the first toxicity index; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in a patient to whom the VLP conjugate could be administered is greater than that of the first target.

13. A method for treating or preventing a disease which comprises the systemic or local administration of a VLP conjugate to a patient in need of such treatment or prevention, which comprises: a pharmacophore that has a first affinity for a target associated with the disease; a vector that has a second affinity for a second target likely to be located near the first target in the patient; and a linker covalently linking the pharmacophore and the vector; wherein the first target is not the same as the second target, and either the second affinity is greater than the first affinity or the concentration of the second target in the patient is greater than that of the first target.

14. The method of claim 13 wherein the disease is fungal or bacterial infection.

15. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the VLP conjugate has a molecular weight of less that about 2000 daltons.

16. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the linker is selected from a plurality of linkers which are ranked by at least one criterion selected from the group consisting of: chemical stability under physiological conditions; types and number of reactive moieties; metabolic stability; solubility; length; and flexibility.

17. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the vectors is selected according to at least one criterion selected from the group consisting of: its pharmacological effects; and its lack of affinity for the first target.

18. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the vector that is selected binds to the first target with a first binding constant and binds to the second target with a second binding constant, wherein the second binding constant is greater than about 10 times that of the first binding constant.

19. The method of claim 18 wherein the second binding constant is greater than about 100 times that of the first binding constant.

20. The method of claim 19 wherein the second binding constant is greater than about 1000 times that of the first binding constant.

21. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the vector and linker are selected using information obtained by screening a plurality of vector-linker conjugates for their affinities for the second target.

22. The method of claim 21 wherein the vector-linker conjugates are radioactively-labeled.

23. The method of claim 21 wherein the vector- linker conjugates are attached to probe molecules.

24. The method of claim 21 wherein the vector- linker conjugates are screened using combinatorial chemistry techniques.

25. The method of claim 1 wherein the pharmacophore and linker are selected using information obtained by screening a plurality of pharmacophore-linker conjugates for their affinities for the first target.

26. The method of claim 25 wherein the pharmacophore-linker conjugates are attached to probe molecules.

27. The method of claim 25 wherein the pharmacophore-linker conjugates are screened using combinatorial chemistry techniques.

28. The method of claim 25 wherein the pharmacophore and linker are selected using information obtained by screening a plurality of pharmacophore-linker conjugates for their affinities for the first target.

29. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the second target is located in a cell containing the first target, and the concentration of the second target in that cell is greater than about 10^"5 molar.

30. The method of claim 29 wherein the concentration of the second target in the cell is greater than about 10^"4 molar.

31. The method of claim 30 wherein the concentration of the second target in the cell is greater than about 10^"3 molar.

32. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the pharmacophore is selected from the group consisting of: antibiotics; antibacterials; antimycoplasmals; antivirals; antifungals; antiprotozoals; molecules active against single celled eukaryotes; molecules active against parasites; molecules that bind tightly to metabolites; inhibitors or activators of binding to macromolecular receptors, including antagonists or agonists of neural receptors and inhibitors or activators of transcription factors; enzyme inhibitors or activators; inhibitors or activators of binding interactions between macromolecules; inhibitors or activators of binding interactions between or catalytic activities of the proteins in cell signaling pathways; nucleic acid polymerase inhibitors; protein synthesis inhibitors; protease inhibitors or activators; kinase and phosphatase inhibitors or activators; glycosylation inhibitors; dihydrofolate reductase inhibitors; ionophores; nucleic acid mutagens; nucleic acid alkylating agents; nucleic acid cleavage agents; and other molecules which modify DNA.

33. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the linker is selected from the group consisting of: molecules which comprise a polyethylene glycol, polyethylene enimine, or linear alkane moiety; molecules which comprise an end-group selected from the group consisting of: amines, imines, amine oxides, hydrazines, azo compounds, ethers, thioethers, sulfoxides, sulfonamides, sulfonyl esters, phosphate esters, phosphines, methylenes, methines, carboxyl amides, carboxyl esters, and imidates; and molecules having chains joining the end groups that are, or are combinations of, linear alkanes, linear alkenes, alkynes, di- and multi-substituted phenyl rings, di- and multi-substituted napthalene rings, di- and multi-substituted cyclohexyl rings, di- and multi- substituted decalin rings, di- and multi- substituted heteroaromatic rings, fused aryl, heteroaromatic and alkyl rings of varying conformational rigidity, linear alkylamines, linear alkyl alcohols, polyalkylamines including polyethylene enimine based linkers, polyalkylethers, and polyalkylthioethers.

34. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the second target is a specific and abundant protein or polysaccharide product of a pathogen.

35. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the second target is selected from the group consisting of: polymerases; transcriptases; ribosomes; proteins involved with protein folding or other chaperones; structural proteins of viruses; abundant proteins of specific eukaryotic cells; enzymes; and antibiotic resistance elements.

36. The method of claim 1, 4, 6, 8, 10, 12, or 13 wherein the linker is selected so that its length is sufficient to allow the pharmacophore to interact with the first target and the vector to interact with the second target at the same time.

37. A VLP conjugate that can be used in the treatment or prevention of a disease or condition in a patient, which comprises: a pharmacophore moiety having an affinity for a first target; a linker moiety; and a vector moiety having an affinity for one or more second targets; wherein the pharmacophore and vector moieties are covalently attached to the linker, each second target is likely to be located near the first target in a typical patient suffering or likely to suffer from the disease or condition, and the first target is not the same as any of the second targets.

38. The VLP conjugate of claim 37 wherein the covalent bonds attaching the vector to the linker and the linker to the pharmacophore do not cleave under the physiological conditions suπounding the first target.

39. The VLP conjugate of claim 37 wherein the vector moiety is derived from a compound selected from the group consisting of: sordarin; biotin; and kiπomycin.

40. The VLP conjugate of claim 37 wherein the pharmacophore moiety is derived from a compound selected from the group consisting of: fluconazole; penicillin; trimethoprim; and tetracycline.

41. A sordarin- linker-fluconazole VLP conjugate.

42. A biotin- linker-penicillin VLP conjugate.

43. A kiπomycin- linker-trimefhoprim VLP conjugate.

44. A kiπomycin-linker-tetracyline VLP conjugate.

45. An automated device for designing a VLP conjugate, which comprises: an input means operably linked to a central processing unit; an output means operably linked to the central processing unit; and a memory or storage means operably linked to the central processing unit and containing a database comprised of physical, pharmacological, chemical, or structural information about a least one vector, linker, pharmacophore, vector target, and pharmacophore target; wherein: the central processing unit is configured for the selection from the database of a pharmacophore with a first affinity for a first target, a linker, and a vector with a second affinity for a second target; the first target is not the same as the second target; and either the second affinity is greater than the first affinity or the concentration of the second target in the patient is greater than that of the first target.