AU2007258550B2 - Liposome-mediated native chemical ligation - Google Patents

Abstract

Native chemical ligation of hydrophobic reactants in a lipid phase, using the chemoselective process of native chemical ligation (NCL).

Description

WO 2007/146070 PCT/US2007/013431 LIPOSOME-MEDIATED NATIVE CHEMICAL LIGATION 5 This application claims the benefit of U.S. Provisional Application Serial No. 60/811,882, filed June 8, 2006; and is a continuation-in-part of International Application No. PCT/US2007/000158, filed January 3, 2007, which claims the benefit of U.S. Provisional Application Serial No. 60/755,881, filed January 3, 2006, U.S. Provisional Application Serial No. 60/796,769, filed May 2, 2006, 10 and U.S. Provisional Application Serial No. 60/809,272, filed May 30, 2006, all of which are incorporated herein by reference in their entireties. STATEMENT OF GOVERNMENT RIGHTS This invention was made with government support under a grant from 15 the National Cancer Institute of the National Institutes of Health, Grant No. RO1 CA88986. The U.S. Government has certain rights in this invention. BACKGROUND OF THE INVENTION Native chemical ligation (NCL) is a chemo-selective reaction that occurs 20 at physiological pH between an N-terminal cysteine residue and a C-terminal peptide thioester (Dawson et al., Science 1994, 266, 776-779; Dawson et al., Annu. Rev. Biochem. 2000, 69, 923-960; Yeo et al., Chem.-Eur. J. 2004, 10, 4664-4672). In the first step of ligation, a reversible trans-thioesterification takes place between the C-terminal thioester and the sulfhydryl group from the 25 N-tenninal cysteine residue. The ligated peptide thioester then undergoes a rapid, irreversible and spontaneous intramolecular S-+N shift, generating the thermodynamically favored native aide bond at the ligation junction. NCL occurs uniquely at an N-terminal cysteine residue regardless of the presence of any additional internal cysteine residues and, as this ligation method is 30 compatible with both carbohydrates and peptides, provides access to glycopeptides. The applicability of NCL is reduced when peptide segments are poorly soluble in aqueous buffer. Since NCL is usually performed in aqueous buffers, 1 WO 2007/146070 PCT/US2007/013431 this can present complications when one of the reactants to be ligated has hydrophobic character. Recently, some researchers have attempted to use native chemical ligation to link selected reactants to membrane-spanning domain fragments of transmembrane proteins. Otaka et al. covalently linked two 5 membrane-embedded transmembrane peptide domains at a ligation site that was situated in the hydrophilic extracellular loop region (Chem Commun., 2004, 1722-1723). Hunter et al. attached a small soluble peptide to the end of a transmembrane peptide embedded in a cubic lipidic phase matrix (Bioconjugate Chem., 2004, 15:3; U.S. Pat. Publ. 20030018169, published Jan. 23, 2003). 10 There remains, however, a need for reliable processes for chemical ligation of a wide variety of hydrophobic molecules including compounds that contain lipid and/or carbohydrate moieties. 15 SUMMARY OF THE INVENTION The invention provides a method for native chemical ligation (NCL) of hydrophobic reactants in a lipid phase to produce a multicomponent ligation product. The reactants are embedded or solubilized within a lipidic structure such as a lipid monolayer, lipid bilayer, a liposome, a micelle, a film, an 20 emulsion, matrix, or a gel. The lipid structure is typically formed from nonpolar, hydrophobic and/or amphipathic components, such as phospholipids. Preferably, the thioester and cysteine moieties that are involved in the chemoselective reaction are positioned within the lipid phase such that the ligation reaction takes place within the lipidic structure. 25 In one embodiment of the method of the invention, one or more first and second hydrophobic reactants are initially mixed with one or more lipid phase components. The first hydrophobic reactant includes an N-terminal cysteine residue, and the second hydrophobic reactant includes a thioester. The lipid phase components are nonpolar, hydrophobic and/or amphipathic molecules that 30 are capable of forming a lipidic structure. The mixture is subjected to conditions effective to form a lipidic structure in which the first and second reactants are embedded. The first and second reactants are then subjected to conditions effective to allow ligation of the first reactant and the second reactant to yield a multicomponent compound comprising the first and second reactant. Optionally, 2 WO 2007/146070 PCT/US2007/013431 one or both of the first and second reactants is not a transmembrane protein or membrane-spanning fragment thereof. Another embodiment of the method of the invention utilizes a preformed lipidic structure. The first and second hydrophobic reactants are contacted with 5 a preformed lipidic structure under conditions to allow ligation of the first reactant and the second reactant to yield a multicomponent compound comprising the first and second reactant. The method of the invention optionally further includes contacting the resulting (first) multicomponent compound with at least one third hydrophobic 10 reactant within a lipid structure under conditions to allow ligation of the multicomponent compound and the third reactant, to yield a second, further multicomponent compound comprising the first, second and third reactants. Preferably, prior to or concurrent with ligation, the first multicomponent compound and the third reactant are solubilized within a lipidic structure to 15 facilitate ligation of the first multicomponent to the third reactant. Preferably, the linkage reaction takes place in the lipid phase, within the lipidic structure, rather than at the interface between the lipidic structure and the external aqueous environment. The use of an initiator compound, such as a thiol, to catalyze the ligation 20 is optional. The ligation is readily performed in the absence of an initiator compound. Optionally, in any method of the invention, one or more of the first, second or third hydrophobic reactants are not transmembrane proteins or membrane-spanning fragments thereof. 25 One example of a compound that can be produced by the method of the invention is a multicomponent vaccine. The reactants used in the ligation reaction can, for example, take the form of vaccine components such as a carbohydrate component, a peptide component, a lipid component, or conjugates or combinations thereof. A multicomponent vaccine can be synthesized, for 30 example, from lipopeptide thioester, peptide and glycopeptide reactants (Fig. 1) using the method of the invention. These reactants can be advantageously designed or selected to include desired antigenic or immunogenic features, such as T-epitopes or B-epitopes. A reactant that includes a T-epitope may be, for example, a peptide, glycopeptide, or lipopeptide. A reactant that includes a B 3 WO 2007/146070 PCT/US2007/013431 epitope may be, for example, a carbohydrate-containing compound. The B epitope can be from a microorganism such as a virus, e.g., human immunodeficiency virus or hepatitis C virus, or from a bacterium, a fungus, or a protozoan. The B-epitope can be one that is overexpressed on a cancer cell. The 5 carbohydrate may be a self-antigen, such as a MUC-1 glycopeptide. A carbohydrate reactant useful in vaccine synthesis can include a glycoconjugate selected from the group consisting of a glycosylated protein, a glycosylated peptide, a glycosylated lipid, a glycosylated amino acid, a DNA and an RNA. A lipid reactant useful in vaccine synthesis can, for example, include a lipopeptide 10 adjuvant. One example of a suitable lipid reactant is a compound that includes a Toll-like receptor (TLR) ligand, such as Pam 3 Cys or Pam 3 CysSK,, wherein n = 0, 1, 2, 3, 4 or 5, preferably Pam 3 CysSK 4 . Unless otherwise specified, "a," "an," "the," and "at least one" are used interchangeably and mean one or more than one. 15 Abbreviations: Cha, cyclohexylalanine; DIPEA, N,N diisopropyl ethylamine; DMF, dimethylformamide; DPC, dodecylphosphocholine; DTT, dithiothreitol; EDT, 1, 2-ethanedithiol; EDTA, ethylenediaminetetraacetic acid; Fmoc, fluorenylmethoxycarbonyl; HATU, 0 (7-azabenzotriazol-1-yl)-N,N,N',N'-tetramethyl-uronium hexafluorophosphate; 20 HBTU, 2-(1 H-benzotriazole-l -yl)- 1,3,3,3 -tetramethylaminium hexafluorophosphate; HOAt, I -hydroxy-7-azabenzotriazole; HOBt, N hydroxybenzotraizole; NCL, native chemical ligation; NMP, N methylpyrrolidone; PyBOP, benzotraizole- I -yl-oxy-tris-pyrrolidino phosphonium hexafluorophosphate; SPPS, solid phase peptide synthesis; TCEP, 25 Tris[2-carboxyethyl]phosphine; TFA, trifluoroacetic acid; THF, tetrahydrofuran; TIS, triisopropylsilane. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows a general schematic of (a) exemplary reactants; and (b) a 30 three-component vaccine synthesized from those reactants using liposome mediated chemical ligation. Figure 2 shows an exemplary three-component glycolipopeptide vaccine 1. 4 WO 2007/146070 PCT/US2007/013431 Figure 3 shows synthesis of an exemplary three-component glycolipopeptide vaccine 7 using native chemical ligation (NCL) (Scheme 1). Figure 4 shows preparation of 10 from 5 and 8 (Scheme 2). Figure 5 shows preparation of 11 and 12 from 3, 6, 8 and 9 (Scheme 3). 5 Figure 6 shows synthesis of cys-glycopeptide 3; (a) SPPS using Fmoc chemistry, coupling with HBTU/HOBt (Knorr et al., Tetrahedron. Lett. 1989, 30, 1927-1930) in the presence of DIPEA in NMP; (b) 17', HATU/HOAt, DIPEA, DMF, overnight; (c) TFA (94.0%), water (2.5%), EDT (2.5%), TIS (1%); (d) 5% aqueous hydrazine, excess of DTT (Scheme 4). 10 Figure 7 shows synthesis of cys(Acm)-athioester peptide 2 using the alkanesulfonamide "safety-catch" linker. (a) SPPS using Fmoc-chemistry, coupling with HBTU/HOBt in the presence of DIPEA in NMP; (b) ICH 2 CN, DIPEA, NMP, 24 hr; (c) BnSH, Na-Thiophenate,THF, 24 hr; (d) Reagent B (TFA (88%), Phenol (5%), H 2 0 (5%), TIS (2 %)), 4 hr (Scheme 5). 15 Figure 8 shows synthesis of lipopeptide athioester 6 using the alkanesulfonamide "safety-catch" linker. (a) SPPS using Fmoc-chemistry, coupling with HBTU/HOBt in the presence of DIPEA in NMP; (b) Manual coupling of Pam 2 Cys-OH (Metzger et al., Int. J. Pro. Pep. Res. 1991, 38, 545 554), PyBOP, HOBt in the presence of DIPEA in DMF; (c) 20 % Piperidine in 20 DMF; (d) Coupling of Palmitic acid, PyBOP, HOBt in the presence of DIPEA in DMF; (e) ICH 2 CN, DIPEA, NMP, 24 hr; (f) BnSH, Na-Thiophenate,THF, 24 hr; (g) Reagent B (TFA (88%), Phenol (5%), H 2 0 (5%), TIS (2%)), 4 hr (Scheme 6). Figure 9 shows synthesis of lipidated amino acid athioester 8 using the 25 alkanesulfonamide "safety-catch" linker. (a) i. Manual coupling of Fmoc lipidated amino acid with PyBOP/HOBt in the presence of DIPEA in DMF; ii. 20% Piperidine in DMF; iii. Manual Coupling of Fmoc-Gly-OH with PyBOP/HOBt in the presence of DIPEA in DMF; iv. 20% Piperidine in DMF; v. Manual coupling of Fmoc-lipidated amino acid (Toth et al. Liebigs Ann. Chem. 30 1990, 1175-1183; Helv. Chim. Acta. 1997, 80, 1280-1300) with PyBOP/HOBt in the presence of DIPEA in DMF; vi. 20% Piperidine in DMF; vii. 10% Ac 2 0, 5% DIPEA in NMP for 10 min; (b) ICH 2 CN, DIPEA, NMP, 24 hr; (c) BnSH, Na-Thiophenate,THF, 24 hr (Scheme 7). 5 WO 2007/146070 PCT/US2007/013431 Figure 10 shows sequential native chemical ligation of 7 or 10, (a) 6 M Gn-HCI, 200 mM sodium phosphate buffer (pH 7.5), thiophenol 4% (final v/v); (b) 200 mM Sodium Phosphate buffer, pH 7.5, DPC, tris(carboxyethyl)phosphine (2% w/v), EDTA (0.1% w/v), sonication (1 min), 5 extrusion and then, Sodium 2-mercapto-ethanesulfonate (2% w/v); (c) Hg(OAc) 2 , 10% aq HOAc, DTT (Scheme 8). Figure 11 shows synthesis of Cys(Acm)- athioester 9 using the alkanesulfonamide "safety-catch" linker. (a) SPPS using Fmoc-chemistry, coupling with HBTU/HOBt in the presence of DIPEA in NMP; (b) ICH 2 CN, 10 DIPEA, NMP, 24 hr; (c) BnSH, Na-Thiophenate,THF, 24 hr; (d) Reagent B (TFA (88%), Phenol (5%), H 2 0 (5%), TIS (2%)), 4 hr (Scheme 9). Figure 12 Sequential native chemical ligation, (a) 200 mM Sodium Phosphate buffer, pH 7.5, DPC, tris(carboxyethyl)phosphine (2% w/v), EDTA 15 (0.1 O/ w/v), sonication ( 1 min), extrusion and then, Sodium 2-mercapto ethanesulfonate (2% w/v); (b) Hg(OAc) 2 , 10% aq HOAc, DTT (Scheme 10). Figure 13 shows liposome-mediated native chemical ligation of glycolipopeptide 37 from 38 and 39 in the absence of thiol initiator. 20 DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS Native chemical ligation (NCL) has proven inefficient or ineffective in the ligation of reactants that are poorly soluble in aqueous buffer. Such reactants include hydrophobic, nonpolar, or amphipathic compounds that tend not to dissolve well in water or have a low affinity for water. In the present invention, 25 hydrophobic reactants are incorporated within liposomes, micelles, or other lipophilic structures, thereby allowing NCL of the hydrophobic reactants to proceed. The method of the invention provides a novel method for ligation of first and second hydrophobic, nonpolar or amphipathic reactants. Unless otherwise 30 indicated, the term "hydrophobic reactant" as used herein is inclusive of nonpolar and amphipathic reactants. Preferred hydrophobic reactants include lipophilic peptides, lipopeptides, glycopeptides, glycolipopeptides, lipidated amino acids and glycosylated amino acids. One of the hydrophobic reactants 6 WO 2007/146070 PCT/US2007/013431 includes a thiol, preferably a terminal cysteine residue, more preferably an N terminal cysteine reside, and the other hydrophobic reactant includes a thioester, preferably a C-terminal thioester. The first and second reactants are contacted with a lipid or lipidic structure, e.g., a membrane, under conditions to allow 5 native chemical ligation of the first reactant and the second reactant to yield a multicomponent compound comprising the first and second reactant. In one embodiment, the reactants are contacted with the components of the lipid structure prior to formation of the lipid structure. The resulting mixture is then subjected to physical or chemical conditions so as to allow the formation 10 of a lipidic structure, such as a bilayer, monolayer, micelle, liposome, film, emulsion, matrix or gel. Methods for making lipid bilayers, monolayers, liposomes, micelles, films, matrices, gels and emulsions are well known to the art, and the invention is not intended to be limited by the method for making the lipidic structure. 15 In another embodiment, the reactants are contacted with a preformed lipid structure, then the mixture is subjected to physical or chemical conditions so as to allow for the solubilization of the reactants in the lipidic phase. For example, the mixture can be shaken, sonicated, heated or the like to fully solubilize the reactants. 20 Optionally, the ligation reaction is initiated with an initiator or catalyst. Preferred initiators are sulfur-containing compounds such as thiols, including thiophenol, substituted thiophenols such as 4-carboxylmethylthiophenol, thiophenol/benzyl mercaptan, 2-mercaptoethanesulfonate, or sodium-2 mercaptoethane sulfonate. However, it has been found (see Example II) that the 25 ligation reaction proceeds within the lipidic structure even without the addition of a catalyst. Thus, the ligation method of the invention can be practiced with or without a catalyst. If a catalyst is used, the ligation reaction is preferably not initiated until both reactants are solubilized in the lipid phase; i.e., neither reactant remains in 30 the aqueous phase. The ligation reaction preferably takes place within the lipid phase, as discussed in more detail below. The present invention utilizes lipid solubilization is used to facilitate native chemical ligation involving hydrophobic, nonpolar, or amphipathic reactants. Lipids are examples of hydrophobic compounds. Glycolipids, 7 WO 2007/146070 PCT/US2007/013431 glycopeptides, and phospholipids are examples of amphipathic compounds. Amphipathic compounds contain both hydrophobic and hydrophilic parts. The word amphipathic is used interchangeably with the word amphiphilic. Further, as noted above, unless otherwise indicated the term "hydrophobic reactant" as 5 used herein is inclusive of nonpolar and amphipathic reactants. Most hydrophobic, nonpolar and amphipathic reactants are lipophilic, tending to dissolve in, having a strong affinity for, or readily mixing with lipids or other substances of low polarity. Lipophilic reactants are preferred for use in the method of the invention. 10 In the present invention, native chemical ligation takes place in a lipid phase, preferably within a lipidic structure. The molecular components of a lipid phase may be ordered or disordered. The lipidic structure can be a planar or sheet-type structure; it can take the form of a closed structure, such as a sphere; it can constitute a lipid or lipophilic emulsion, film, matrix or gel; or it can take a 15 more complex form, such as a cubic lipidic phase (Hunter et al., Bioconjugate Chem., 2004, 15:3; U.S. Pat. Publ. 20030018169, published Jan. 23, 2003). The lipidic structure can take the form of a monolayer (e.g., a spherical monolayer structure such as a micelle), a bilayer (e.g., a spherical bilayer structure such as a liposome) or it can include additional layers. The lipidic structure is also 20 referred to herein as a membrane or lipidic membrane. The lipidic structure may be formed from one or more types of naturally occurring or synthetic nonpolar, hydrophobic or amphipathic molecules, such as amphipathic detergents, phospholipids, glycolipids, sterols such as cholesterols, synthetic amphipathic polymers and the like. It should be understood that the 25 invention is not limited by the composition of the lipidic structure. Suitable phospholipids include, without limitation, naturally occurring or synthetic phospholipids, including derivatized forms thereof. Common phospholipids suitable for use in forming the lipidic structure include phosphatidylcholine (lecithin) (PC), phosphatidylglycerol (PG), phosphatidic acid (PA), 30 diphosphatidylglycerol(cardiolipin), phosphatidyl-inositol (PI), phosphatidylethanolamine (PE), phosphatidylserine (PS), sphingolipids such as sphingomyelin, and their analogs and derivatives as well as their lysophospholipid counterparts in which one of the acyl substituents is missing. Phospholipid derivatives can have, for example, one or more saturated acyl 8 WO 2007/146070 PCT/US2007/013431 groups, unsaturated acyl groups, or mixed acyl groups. Furthermore, derivatizations at the acyl groups of the phospholipid can be symmetric or asymmetric (such as POPC, 1-palmitoyl-2-oleoyl phosphatidylcholine). Additional exemplary components of the lipidic structure include, without 5 limitation, dodecylphosphocholine and phosphocholine. Optionally phospholipids and other membrane components can be derivatized with polyethylene glycol (PEGylated) or other polymers. Examples of phosphatidylcholines for use in preparation of the lipidic structure include DOPC, dioleoylphosphatidylcholine; DEPC, 10 dierucoylphosphatidylcholine; DDPC, didecanoylphosphatidylcholine; DLPC, dilauroylphosphatidylcholine; DMPC, dimyristoylphosphatidylcholine; DPPC, dipalmitoylphosphatidylcholine; DSPC, distearoylphosphatidylcholine; and DLoPC, dilinoleoyl phosphatidylcholine. Examples of phosphatidylglycerols for said use include DLPG, dilauroyl phosphatidylglycerol; DMPG, dimyristoyl 15 phosphatidylglycerol; DPPG, dipalmitoyl phosphatidylglycerol; DSPG, distearoyl phosphatidylglycerol; DOPG, dioleoyl phosphatidylglycerol; and DEPG, dierucoyl phosphatidylglycerol. Examples of phosphatidic acids include DLPA, dilauroyl phosphatidic acid; DMPA, dimyristoyl phosphatidicacid; DPPA, dipalmitoyl phosphatidicacid; and DSPA, distearoyl phosphatidicacid. 20 Examples of phosphatidylethanolamines include DLPE, dilauroyl phosphatidylethanolamine; DMPE, dimyristoyl phosphatidylethanolamine; DPPE, dipalmitoyl phosphatidylethanolamine; DSPE, distearoyl phosphatidylethanolamine; DOPE, dioleoyl phosphatidylethanolamine; and DEPE, dierucoyl phosphatidylethanolamine. Examples of phosphatidylserines 25 include DLPS, dilauroyl phosphatidylserine; DPPS, dipalmitoyl phosphatidylserine; DMPS, dimyristoyl phosphatidylserine; DSPS, distearoyl phosphatidylserine; and DOPS, dioleoyl phosphatidylserine. An example of a sphingomyelin derivative is dihidrosphingomyelin. In a preferred embodiment, NCL is performed in the presence of a 30 micelle (a vesicle formed from a lipid monolayer) or a liposome (a vesicle formed from a lipid bilayer). It should be understood that the term "liposome mediated" ligation, as used herein, is intended to include ligations that are mediated by liposomes (bilayers), micelles (monolayers) or other lipid structures such as films, emulsions, gels and matrices. 9 WO 2007/146070 PCT/US2007/013431 Advantageously, the liposome, micelle or other lipidic structure within which the ligation is performed can be used as a delivery vehicle for administration of a therapeutic ligation product to a patient in need thereof. It was observed that reaction rates of liposome-mediated NCL are 5 substantially higher than traditional reaction conditions, resulting in improved yields. Without intending to be bound by theory, NCL in the presence of a lipid structure such as a liposome or micelle is believed to reduce nonspecific aggregation of the hydrophobic reactants and provide better access to the hydrophobic reactant for ligation. The ligation reactions described in Example I 10 take place in the lipid environment, and the relative high reaction rate of the liposome-mediated NCL is likely due to a relatively high local concentration of reactants. In a particularly preferred embodiment of the method of the invention, the ligation of the reactants takes place within the lipidic phase, e.g., the lipid 15 bilayer or monolayer, as opposed to at the interface between the membrane structure and the external, aqueous solution. More specifically, the functional groups involved in the ligation reaction, e.g., the thioester and the cysteine, are solubilized within the lipid phase. When the reaction takes place within the lipidic phase, both the thioester of the first reactant and the sulthydryl group 20 from the N-terminal cysteine residue of the second reactant are embedded within the membrane structure, in contrast to surface ligation as shown, for example, in Otaka et al. (Chem Commun., 2004, 1722-1723). The method of the invention is useful for native chemical ligation using one or more hydrophobic or lipophilic reactants, without limitation. The method 25 is particularly useful in methods involving the ligation of one or more biomolecules, such as hydrophobic peptides, lipids, phospholipids, steroids, triglycerides, glycopeptides, lipopeptides, and glycolipopeptides. In a particularly preferred embodiment, the method is used to synthesize lipidated carbohydrates, such as lipidated glycopeptides as exemplified in Example I. 30 Lipidated carbohydrates, including lipidated glycopeptides, that are synthesized according to the method of the invention (see Fig. I for a general synthetic scheme) can be useful as vaccines, as further described in international patent application PCT/US2007/000158, filed January 3, 2007, and Buskas et al., Angew. Chem., Int. Ed. 2005, 44, 5985-5988. 10 WO 2007/146070 PCT/US2007/013431 Optionally, one or more of the reactants is derivatized prior to ligation so as to add a C-terminal thioester and/or an N-terminal cysteine residue, as needed, in order to facilitate the native chemical ligation reaction. In a preferred embodiment, the method of the invention is used to 5 synthesize a compound that contains one or more carbohydrate components, one or more peptide components, and/or one or more lipid components. The individual components or "building blocks" to be assembled into a multi component compound using the method of the invention can be chemically, enzymatically or biologically synthesized, without limitation, and may include 10 one or more protecting groups that can be removed during a later step in a multi step synthesis. A carbohydrate component that is chemically synthesized can, for example, contain an acetyl ester that is subsequently removed prior to or during the process of liposome-mediated native chemical ligation. The method of the invention can be used in a single step to synthesize a compound 15 containing two or more components, or it can be used in multiple steps to synthesize a compound containing three or more components. Examples of suitable carbohydrate components include oligosaccharides, polysaccharides and monosaccharides, and glycosylated biomolecules (glycoconjugates) such as glycoproteins, glycopeptides, glycolipids, 20 glycosylated amino acids, DNA, or RNA. Glycosylated peptides (glycopeptides) and glycosylated amino acids, which contain one or more carbohydrate moieties as well as a peptide or amino acid, are particularly preferred as the carbohydrate component of the ligation product. An example of a glycopeptide is CD52, which is expressed on virtually all human lymphocytes 25 and believed to play an important role in the human immune system. An example of a glycosylated amino acid is the Tn antigen. It should be understood that when the carbohydrate component is a glycopeptide, the peptide part of the glycopeptide optionally includes a T-epitope and thus may serve as a peptide component of the glycol ipopeptide. 30 The carbohydrate component of the ligation product, if present, includes a carbohydrate that contains one or more saccharide monomers. For example, the carbohydrate can include a monosaccharide, a disaccharide or a trisaccharide; it can include an oligosaccharide or a polysaccharide. An oligosaccharide is an oligomeric saccharide that contains two or more 11 WO 2007/146070 PCT/US2007/013431 saccharides and is characterized by a well-defined structure. A well-defined structure is characterized by the particular identity, order, linkage positions (including branch points), and linkage stereochemistry (a, P) of the monomers, and as a result has a defined molecular weight and composition. An 5 oligosaccharide typically contains about 2 to about 20 or more saccharide monomers. A polysaccharide, on the other hand, is a polymeric saccharide that does not have a well defined structure; the identity, order, linkage positions (including brand points) and/or linkage stereochemistry can vary from molecule to molecule. Polysaccharides typically contain a larger number of monomeric 10 components than oligosaccharides and thus have higher molecular weights. The term "glycan" as used herein is inclusive of both oligosaccharides and polysaccharides,.and includes both branched and unbranched polymers. When the carbohydrate component contains a carbohydrate that has three or more saccharide monomers, the carbohydrate can be a linear chain or it can be a 15 branched chain. In a preferred embodiment, the carbohydrate component contains less than about 15 saccharide monomers; more preferably in contains less than about 10 saccharide monomers. The carbohydrate component of the glycolipopeptide includes a carbohydrate that contains a B-epitope. The B-epitope can be a naturally 20 occurring epitope or a non-naturally occurring epitope. Preferably, two or more saccharide monomers of the carbohydrate interact to form a conformational epitope that serves as the B-epitope. A B-epitope is an epitope recognized by a B cell. Any antigenic carbohydrate that contains a B-epitope can be used as the carbohydrate component, without limitation. 25 In one embodiment, the carbohydrate component contains all or part of a self-antigen. Self-antigens are antigens that are normally present in an animal's body. They can be regarded as "self-molecules," e.g., the molecules present in or on the animal's cells, or proteins like insulin that circulate in the animal's blood. An example of a self-antigen is a carbohydrate-containing component derived 30 from a cancer cell of the animal, such as a tumor-associated carbohydrate antigen (TACA). Typically, such self-antigens exhibit low immunogenicity. Examples include tumor-related carbohydrate B-epitope such as Le' antigen (a cancer related tetrasaccharide; e.g., Fucax(1,2)-Galp(1,4)-[Fuca(1,3)]-GaINAc); 12 WO 2007/146070 PCT/US2007/013431 Globo-H antigen (e.g., Fuca( 1,2)-Galp(1,3)-GaNAcp(1,3)-Galai(1,4) Galp(1,4)-Glu); T antigen (e.g., Gal (l,3)-GalNAca-O-Ser/Thr); STn antigen (sialyl Tn, e.g., NeuAca(2,6)-GalNAcat-O-Ser/Thr); and Tn antigen (e.g., a GalNAc-O-Ser/Thr). Another example of a self-antigen is a glycopeptide 5 derived from the tandem repeat of the breast tumor-associated MUC-1 of human polymorphic epithelial mucin (PEM), an epithelial mucin (Baldus et al., Crit. Rev. Clin. Lab. Sci., 41(2):189-231 (2004)). A MUC-1 glycopeptide comprises at least one Tn and/or sialyl Tn (sialyl a-6 GaINAc, or "STn") epitope; preferably linked to a threonine (T-Tn or T-STn). 10 Structures of exemplary tumor-associated carbohydrate antigens (TACA) that can be used as a component of the glycolipopeptide include, without limitation, the structures shown below. Tn. STn. and TF antigens Tn3. STn3. and TF3 M UC-I with Tn. STn. and TF HOOR HO OR HONOR H OR RO RO AHN H1N)COOH NH N COOH TSAPDTRPAPO RNH2 O OR NHAC 7RHR" ROR' E- OH R-H I.R=H.RH 4.R-H.R'=H RO OHH.. HCO 2.R- R- H 5.R- OH R1- AcHN OH OH O HO HO, OH HO OOH HO OH AcHH AcHH \ 9.R-H H00 Ho OH HO OH H 3.R-H.R- HO &R1HR- H4040 15 It should be noted that the Tn, STn, and TF structures shown in above (monomeric, trimeric, clustered) are all shown with a threonine residue. The corresponding serine analogues are also suitable structures. In the case of Tn3, 20 STn3, TF3 and their respective clusters, all possible homo- and hetero-analogues with differences in the threonine/serine composition of the backbone are included. 13 WO 2007/146070 PCT/US2007/013431 L and SLcise wi, Md SLewig' Lcnis'*actose and SLcwis'-Iatose OH OH HO OH H OH HOH HNAc HO O O&HH OH O H H R HO HOAH OH OH OH H HO O OO 5ROH MRH OR-H O.OH OOH OHH OH AcHH=Q1 HOHCO O H COOH OH AcHN4 M AcH O OHN OH The KH-I mliga O HHO O ~OH ~ OH 0 OH O OH VO t O O eO 'C OHO o N 0 HOO N0 H 0 IA HO OH HlNAC HO OH HNAC OH OH 'OH N OH HO HO HO 0 HO OH In another embodiment, the carbohydrate component contains all or part 5 of a carbohydrate antigen (typically a glycan) from a microorganism, preferably a pathogenic microorganism, such as a virus (e.g., a carbohydrate present on gpl20, a glycoprotein derived from the HIV virus), a Gram-negative or Gram positive bacterium (e.g., a carbohydrate derived from Haemophilus influenzae, Streptococcus pneumoniae, or Neisseria meningitides), a fungus (e.g., a 1 ,3-p 10 linked glucan) a parasitic protozoan (e.g., a GPI-anchor found in protozoan parasites such as Leishmania and Trypanosoma brucei), or a helminth. Preferably, the microorganism is a pathogenic microorganism. An exemplary glycan from viral pathogens, Man9 from HIV-1 gpl20, is shown below. 15 Man9 from HIV-1 gp120 Man(ct1-2)-Man(at1-6),, Man(cz1-6) Man(c1-2)-Man(a1-3) Man(p1-4)-GlcNAc(p1-4)-GcNAc(p1) Man(al-2)-Man(ol-2)-Man(ol-3) NH 2 Exemplary HIV carbohydrate and glycopeptide antigens are described in 20 Wang et al. (Current Opinion in Drug Disc. & Develop., 9(2): 194-206 (2006)) 14 WO 2007/146070 PCT/US2007/013431 and Danishefsky (Top. Curr Chem 2007, 267: 109-141), and include both naturally occurring HIV carbohydrates and glycopeptides, as well as synthetic carbohydrates and glycopeptides based on naturally occurring HIV carbohydrates and glycopeptides. 5 Exemplary HCV carbohydrate and glycopeptide antigens are described in Koppel et al. Cellular Microbiology 2005; 7(2):157-165 and Goffard et al. J. of Virology 2005;79(13):8400-8409, and include both naturally occurring HCV carbohydrates and glycopeptides, as well as synthetic carbohydrates and glycopeptides based on naturally occurring HCV carbohydrates and 10 glycopeptides. Exemplary glycans from bacterial pathogens are shown in below. Haemophilus influenzae Type b, CPS repeating unit HO-O -OH -OH O OH -OHO -O-I- 0 OH n Group B type III Streptococcus OH AcOH H OOH OH COOH OH HO OH OH H 00 15 HO HO OH NHAc OH OH 15 WO 2007/146070 PCT/US2007/013431 Cell-wall Carbohydrates from &76llus omthracis, Antes. Sterne. Paxiewr OH HO Z HOOH O HO I 0L; OH HO -U.- NHAc OH 0 ~HO HO .AdfH 0 '0HO T2 HO %N OH OH HO0 NHAr. 0 goJ4 HO HOu HO H HO\-, OH OHHO OH Franciseffa tularends, Con: region and 0-side chain repeating uanit HO HOA H 0 HC) HCO 0 t;NH HO 26 HLRO HO HO Hi ;HO- _ H '~ 0 O AcHNcPO NH, H-r1O 0I

HO

0

HO

1 O Repeating unit of O-chtrin HO'A.HO C o-ligosaccharide I H(R =extension point) AcHN I_ OCn Ro- attlachuncrrt 0-ntain 01" ACHN HO)Z OH 0-ipid A Burkhlrodaria psetedonrullei, exo-polysaccaridc HO OH I-tO HO OH OH HO 0 L O C O HOO 5 OH Exemplary glycans ftrm protozoan pathogens are shown below. 10 Plasmodium falsiparum, Malaria parasite ekhanolarnine phosphate 9 Leishinarcrn species antigenic glycart Man Ga 4 1 1~ 4...~ J 4 Mal - -c- MaPO ~I - Mart -"P- Ma G a a IcNH2 InosinlIPO4a-R 16 WO 2007/146070 PCT/US2007/013431 Trypanozoma cruzi Neu5Ac 2 .J3- Gal Gal GlcNAc-SerrThr GlcNAc-Serfrhr Gal 6 Gal - Gal 6 Gal Gal GIcNAc-ScrfThr GlcNAc-Ser/hr Gal Gal l Gal -L-- Gal 6 5 An exemplary glycan from a fungal pathogen is shown below. Cryptococcus neoformans Capsular polysaccharide GlepA Xyl p (+/-) Xyl p a Man a Man a 1 Man a 4 4 Xyl p (+/-) Xyl p (+/-) 10 An exemplary glycan from helminth pathogen is shown below. Schistosona Fuc Fuc Gal -R4 GIcNAc ^2al GGcNAc n 15 It will be appreciated by one of skill in the art that while numerous antigenic carbohydrate structures are known, many more exist, since only a small fraction of the antigenic or immunogenic carbohydrates have been identified thus far. Examples of the many carbohydrate antigens discovered thus 17 WO 2007/146070 PCT/US2007/013431 far can be found in Kuberan et al., Curr. Org. Chem, 4, 653-677 (2000); Ouerfelli et al., Expert Rev. Vaccines 4(5):677-685 (2005); Hakomori et al., Chem. Biol. 4, 97-104 (1997); Hakomori, Acta Anat. 161, 79-90 (1998); "The use of carbohydrate antigens for the preparation of vaccines for therapy in breast 5 cancer," Drugs of Today 38(11):759-768 (2002); Mendonca-Previato et al., Cuff Opin. Struct. Biol. 15(5):499-505 (2005); Jones, Anais da Academia Brasileira de Ciencias 77(2):293-324 (2005); Goldblatt, J. Med. Microbiol. 47(7):563-567 (1998); Diekman et al., Immunol. Rev., 171: 203-211, 1999; Nyame et al., Arch. Biochem. Biophys., 426 (2): 182-200, 2004; Pier, Expert Rev. Vaccines, 4 (5): 10 645-656, 2005; Vliegenthart, FEBS Lett., 580 (12): 2945-2950, Sp. Iss., 2006; Ada et al., Clin. Microbiol. Inf., 9 (2): 79-85, 2003; Fox et al., J. Microbiol. Meth., 54 (2): 143-152, 2003; Barber et al., J. Reprod. Immunol., 46 (2): 103 124, 2000; and Sorensen, Persp. Drug Disc. Design, 5: 154-160, 1996. Any antigenic carbohydrate derived from a mammal or from an infectious organism 15 can be used as a carbohydrate component, without limitation. The peptide component, if present in the ligation product, can be any peptide-containing structure, and can contain naturally occurring and/or non naturally occurring amino acids and/or amino acid analogs (e.g., D-amino acids). The peptide component advantageously may include a T-epitope, preferably a 20 helper T epitope. Preferably the peptide component contains less than about 20 amino acids and/or amino acid analogs. Examples of peptide components include the universal helper T peptide, QYIKANSKFIGITEL ("QYI") (SEQ ID NO: 1), the universal helper T peptide YAFKYARHANVGRNAFELFL ("YAF") (SEQ ID NO:2), the murine helper T peptide KLFAVWKITYKDT 25 ("KLF") (SEQ ID NO:3) derived from polio virus, and pan-DR binding (PADRE) peptides (PCT WO 95/07707; Alexander et al., Immunity 1:751-761 (1994); Alexander et al., J. Immunol. 2000 Feb 1;164(3):1625-33; U.S Pat. No. 6,413,935 (Sette et al., July 2, 2002)). Preferred immunogenic peptide components for use in a glycolipopeptide 30 ligation product include universal (degenerate or "promiscuous") helper T-cell peptides, which are peptides that are immunogenic in individuals of many major histocompatibility complex (MHC) haplotypes. Numerous universal helper T cell peptide structures are known; however, it should be understood that additional universal T-epitopes, including some with similar or even higher 18 WO 2007/146070 PCT/US2007/013431 potency, will be identified in the future, and such peptides are well-suited for use as the peptide component. Exemplary T-cell peptides for use in the glycolipopeptide include, without limitation: 5 Synthetic, nonnatural PADRE peptide, DAla-Lys-Cha-Val-Ala-Ala-Trp Thr-Leu-Lys-Ala-Ala-DAla, including all the analogues described by J Alexander et al. in Immunity, Vol. 1, 751-761, 1994; Peptides derived from tetanus toxin, e.g., (TT830-843) QYIKANSKFIGITEL (SEQ ID NO:1), (TT1084-1099) VSIDKFRIFCKANPK 10 (SEQ ID NO:4), (TT1 174-1189) LKFIIKRYTPNNEIDS (SEQ ID NO:5), (TT1064-1079) IREDNNITLKLDRCNN (SEQ ID NO:6), and (TT947-967) FNNFTVSFWLRVPKVSASHLE (SEQ ID NO:7); Peptides derived from polio virus, e.g., KLFAVWKITYKDT (SEQ ID NO:3); 15 Peptides derived from Neisseria meningitidis, e.g., YAFKYARHANVGRNAFELFL (SEQ ID NO:2); and Peptides derived from P. falsiparum CSP, e.g., EKKIAKMEKASSVFNVNN (SEQ ID NO:8) The peptide component of a glycolipopeptide ligation product may 20 contain a T-epitope. A T-epitope is an epitope recognized by a T cell. The T epitope can elicit a CD4+ response, thereby stimulating the production of helper T cells; and/or it can elicit a CD8+ response, thereby stimulating the production of cytotoxic lymphocytes. Preferably, the T-epitope is an epitope that stimulates the production of helper T cells (i.e., a helper T-cell epitope or Th-epitope), 25 which in turn make possible a humoral response to the B-epitope supplied by the carbohydrate component. It should be understood that a glycolipopeptide ligation product can contain multiple T-epitopes, which may be the same or different. Further, T epitopes may be present on the carbohydrate component and/or the lipid 30 component (e.g., in embodiments that include glycopeptides and/or glycolipids as the carbohydrate and/or lipid components) in addition to, or in place of, the peptide component. In one embodiment, the B-epitopes and the T-epitopes are homologous; that is, they are derived from the same organism. For example, in a 19 WO 2007/146070 PCT/US2007/013431 glycolipopeptide suitable for use as a vaccine against a microbial pathogen, the T-epitope in addition to the B-epitope may be epitopes that are present in the microbial pathogen. In another embodiment, the B-epitopes and the T-epitopes are heterologous; that is, they are not derived from the same organism. For 5 example, a glycolipopeptide suitable for use as an anti-cancer vaccine may have a B-cell epitope from a cancer cell, but a T-cell epitope from a bacterium or virus. The lipid component, if present in the ligation product, can be any lipid containing component, such as a lipopeptide, fatty acid, phospholipid, steroid, or 10 a lipidated amino acids and glycolipids such as Lipid A derivatives. In some embodiments, the lipid component is non-antigenic; that is, it does not elicit antibodies directed against specific regions of the lipid component. However, the lipid component may and preferably does serve as an immunoadjuvant. The lipid component can serve as a carrier or delivery system for the multi-epitopic 15 glycolipopeptide. It assists with incorporation of the glycolipopeptide into a vesicle such as a liposome or micelle to facilitate delivery of the glycolipopeptide to a target cell, and it enhances uptake by target cells, such as dendritic cells. Further, the lipid component stimulates the production of cytokines. 20 One class of preferred lipid components for use in the ligation product comprises molecular ligands of the various Toll-like receptors (TLRs). There are many known subclasses of Toll-like receptors (e.g., TLRI, TLR2, TRL3, TLR4, TLR5, TLR6, TLR7, TLR8, TLR9, TLRI0, TLRI 1, TLRI2, TLRl 3, TLR14, TLRI5 and TLR16). See Roach et al., PNAS 2005,102:9577-9582, for 25 a review of the relationships between and evolution of Toll-like receptors; and Duin et al., TRENDS Immunol., 2006, 27:49-55, for a discussion of TLR signaling in vaccination. Particularly preferred are lipid components that interact with TLR2 and TLR4. TLR2 is involved in the recognition of a wide array of microbial molecules from Gram-positive and Gram-negative bacteria, as well as 30 mycoplasma and yeast. TLR2 ligands include lipoglycans, lipopolysaccharides, lipoteichoic acids and peptidoglycans. TLR4 recognizes Gram-negative lipopolysaccharide (LPS) and lipid A, its toxic moiety. TLR ligands are widely available commercially, for example from Apotech and InvivoGen. Preferably, 20 WO 2007/146070 PCT/US2007/013431 the lipid component is a TLR ligand that facilitates uptake of the glycolipopeptide by antigen presenting cells. Suitable lipids for use as the lipid component of a ligation product include PamCys-type lipid structures, such as those derived from Pam 3 Cys (S 5 [(R) -2, 3-dipalmitoyloxy-propyl]-N-palmitoyl-(R) - cysteine) and Pam 2 Cys (S [(R) -2, 3-dipalmitoyloxy-propyl]-(R) - cysteine), which lacks the N-palmitoyl group of Pam 3 Cys. Pam 3 Cys and Pam 2 Cys are derived from the immunologically active N-terminal sequence of the principal lipoprotein of Escherichia coli. This class of lipids also includes Pam 3 CysSK 4 (N-palmitoyl-S 10 [(R)-2,3-bis(palmitoyloxy)-propyl]-(R)-cysteinyl-(S)-sery-(S)-lysine-(S)-lysine (S)-lysine-(S)-lysyne) and Pam 2 CysSK 4 (S-[(R)-2,3-bis(palmitoyloxy)-propyl] (R)-cysteinyl-(S)-seryl-(S)-lysine-(S)-lysine-(S)-lysine-(S)-lysyne), which lacks the N-palmitoyl group of Pam 3 CysSK4; it should be understood that the number of lysines in these structures can be 0, 1, 2, 3, 4, 5 or more (i.e., K,, where n = 0, 15 1, 2, 3, 4, 5 or more). Another preferred class of lipids includes Lipid A (LpA) type lipids, such as Lipid As derived from E. coli, S. typhimurium and Neisseria meningitidis. The Lipid As can be attached to the carbohydrate component (containing a B epitope) of the glycolipopeptide and/or to the peptide component (containing a 20 T-epitope) through a linker that is connected, for example, to the anomeric center or anomeric phosphate, the C-4' phosphate or the C-6' position. The phosphates can be modified, for example, to include one or more phosphate ethanolamine diesters. Exemplary Lipid A derivatives are described in, for example, Caroff et al., Microbes Infect. 4, 915-926 (2002); Raetz et al., Annu. 25 Rev. Biochem. 71, 635-700 (2002); and Dixon et al., J. Dent. Res. 84, 584-595 (2005). Advantageously, the method of the invention allows multiple-component compounds to be synthesized using a modular approach. For example, first and second components can be ligated using liposome-mediated native chemical 30 ligation to yield a two-component ligation product. The two-component ligation product is then used as a reactant in a second round of liposome-mediated native chemical ligation with a third component to yield a three-component ligation product. This allows a modular approach to be used to screen for, or synthesize, various vaccines or vaccine candidates. An array of B- and T-epitopes and 21 WO 2007/146070 PCT/US2007/013431 lipopeptides can be made available, including two-component modules that include, for example, selected B- and T- epitopes, or a selected T-epitope and a selected lipopeptide adjuvant. Then, custom compounds can be built by combining the desired modules. The method the invention, liposome-mediated 5 native chemical ligation, can be used to synthesize two-component modules and/or the final compound. Alternatively or additionally, liposome-mediated native chemical ligation can be precede or succeed other ligation methods in a multiple step synthesis to produce the final multi-component compound. A modular approach is attractive because it provides greater synthetic flexibility 10 than linear synthesis. Each building block can be used for the preparation of several different target compounds. EXAMPLES 15 The present invention is illustrated by the following examples. It is to be understood that the particular examples, materials, amounts, and procedures are to be interpreted broadly in accordance with the scope and spirit of the invention as set forth herein. 20 Example 1. Synthesis of a Three-Component Vaccine using Liposome-Mediated Native Chemical Ligation 25 Recently, we demonstrated (Buskas et al., Angew. Chem., Int. Ed. 2005, 44, 5985-5988) that the three-component vaccine candidate 1 (Fig. 2) composed of the tumor-associated Tn-antigen (Springer, Science 1984, 224, 1198-1206; Kagan et al., Cancer Immunol. Immunother. 2005, 54, 424-430; Toyokuni et al., J. Am. Chem. Soc. 1994, 116, 395-396), the peptide T-epitope 30 YAFKYARHANVGRNAFELFL (SEQ ID NO:2; YAF) (Wiertz et al., J. Exp. Med. 1992, 176, 79-88), and the lipopeptide S-[(R)-2,3-dipalmitoyloxy-propyl] N-palmitoyl-(R)-cysteine (Pam 3 Cys) (Spohn et al., Vaccine 2004, 22, 2494 2499; Metzger et al., J. Med. Chem. 1991, 34, 1969-197) can elicit IgG antibody 22 WO 2007/146070 PCT/US2007/013431 responses. This finding was significant because it had been difficult to elicit relevant immune responses against tumor-associated carbohydrates (Kuduk et al., J.Am.Chem.Soc. 1998, 120, 12474-12485; Danishefsky et al., Angew.Chem. Int. Ed. 2000, 39, 836-863). 5 To optimize the immunological properties of a three-component vaccine, a synthetic methodology was required, which would allow a convenient assembly of a number of B- and T- epitopes and lipopeptide adjuvants into a range of vaccine candidates. During our investigation, we discovered that liposome mediated native chemical ligation (NCL) is a useful approach that greatly 10 increases the reaction rates and yields of ligations of sparingly soluble peptide reactants (Ingale et al., Org Lett. 2006 Dec 7;8(25):5785-8; supplementary information is available electronically on the worldwide web at http: //pubs.acs.org/subscribe/j oumals/orlef7/suppinfo/o1062423x/ol062423xsi2006 11 07_021934.pdf). Importantly, for the first time the new approach makes it 15 possible to employ lipidated peptides in NCL. The methodology is also attractive for NCL of lipophilic peptides, which usually give low yields of products under classical reaction conditions. Compound 7, which is composed of the tumor-associated glycopeptide derived from MUC-1 (Snijdewint et al., Int. J. Cancer 2001, 93, 97-106) the 20 well-documented T-cell epitope YAFKYARHANVGRNAFELFL (SEQ ID NO:2; YAF), and the lipopeptide Pam 3 CysSK 4 , was selected as a synthetic target. It was envisaged that this compound could be prepared from building blocks 2, 3, and 6 by sequential NCL. Thus, NCL between the cysteine moiety of 3 and the thioester of 2 should link the B- and T-epitopes. Next, removal of 25 the S-acetamidomethyl (Acm) protecting group (Veber et al., J. Am. Chem. Soc. 1972, 94, 5456-5461) of the N-terminal cysteine of the ligation product should reveal a free cysteine thiol, which can then be ligated with the thioester of 6 to give required adduct 7. MUC-1 epitope 3 was assembled by automated solid-phase peptide 30 synthesis (SPPS) using Fmoc protected amino acids and NFmocThr(a-AcO 3

-D

GaINAc)OH (Tn antigen; Cato et al., J.Carb.Chem. 2005, 24, 503-516) on a Rink amide linker resin. After the assembly, the glycopeptide was cleaved from the solid support by treatment with TFA (94.0%), water (2.5%), ethanedithiol 23 WO 2007/146070 PCT/US2007/013431 (2.5%) and TIS (1%). Next, the acetyl esters of the saccharide moiety were cleaved by treatment of 5% aqueous hydrazine in the presence of DTT to give glycopeptide 3. Peptide thioester 2 was synthesized on a sulfonamide "safety-catch" 5 linker (Kenner et al., J. Chem. Soc. D-Chem. Commun. 1971, 636; Shin et al., J. Am. Chem. Soc. 1999, 121, 11684-11689; Ingenito et al., J. Am. Chem. Soc. 1999, 121, 11369-11374). Cleavage of the fully assembled peptide from the resin was accomplished by a two-step procedure entailing alkylation of the sulfonamide with iodoacetonitrile followed by treatment with benzyl mercaptan 10 to give a protected peptide having a C-terminal thioester. The acid sensitive protecting groups of the peptide were removed by treatment with reagent B (TFA, phenol, water and TIS; 88/5/5/2) to give 2. This compound is equipped with an N-terminal cysteine residue carrying the orthogonal Acm thiol protecting group, which is stable under conventional side-chain deprotection with TFA but 15 can be cleaved using Hg(II) or Ag(I), or oxidatively by using 12. Finally, Pam 3 CysSK 4 a-thioester 6 was synthesized similar to the preparation of compound 2. Having building blocks 2, 3, and 6 at hand, attention was focused on the preparation of glycolipopeptide 7 by sequential NCL (Scheme 1, Fig. 3). The 20 ligation of 2 with 3 was performed under standard conditions using a phosphate buffer (pH 7.5) containing 6 M of guanidinium-hydrochloride. The ligation was catalyzed by the addition of 4% thiophenol (v/v) (Dawson et al., J. Am. Chem. Soc. 1997, 119, 4325-4329) and the progress of the reaction monitored by LC/MS. The reaction was rather sluggish and after a reaction time of 18 hours 25 partial conversion of 2 and 3 into 4 and some hydrolysis of the thioester was observed. Purification by semi-preparative RP-H PLC gave 4 in a yield of 48%. Next, the Acm group of 4 was removed using mercury(Il) acetate to give glycopeptide 5, containing a free sulfhydryl moiety. Unfortunately, a second NCL of compound 5 with the thioester 6 in a phosphate buffer containing 6 M 30 guanidinium-hydrochloride and thiophenol did not provide target compound 7. The failure of this reaction is probably due to the poor solubility of 6. Addition of detergents such as SDS (Valiyaveetil et al., J. Am. Chem. Soc. 2002, 124, 9113-9120) and DPC (Clayton et al., Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 24 WO 2007/146070 PCT/US2007/013431 4764-4769), at ambient and elevated reaction temperatures (40-50*C) did not improve the ligation. Furthermore, the use of alternative catalysts such as a mixture of sodium thiophenate and thiophenol or sodium 2-mercaptoethane sulfonate did not lead to product formation. Attempts to perform the ligation in a 5 phosphate buffer containing 8 M urea and use of trifluoroethanol as a reaction solvent also led to failure. We envisaged that the incorporation of compounds 5 and 6 into liposomes would facilitate solubilization (Hunter et al., Bioconj. Chem. 2004, 15, 437-440; Otaka et al., Chem. Commun. 2004, 1722-1723) and hence increase 10 the rate of ligation. Thus, a film of dodecylphosphocholine, thiol 5, and thioester 6 was hydrated by incubation at 37*C for 4 hours in a phosphate buffer (pH 7.5) in the presence of carboxyethyl phosphine and EDTA. The latter two reagents were added to suppress disulfide formation. The mixture was ultra-sonicated for 1 minute and the resulting vesicles were sized to I pm by passing through a 15 polycarbonate membrane filter. The ligation was catalyzed by the addition of sodium 2-mercaptoethane sulfonate (Grogan et al., J. Am. Chem. Soc. 2005, 127, 14383-14387) and, surprisingly, after a reaction time of 2 hours, LC-MS showed completion of the reaction. After purification by RP-HPLC over a C-4 column, compound 7 was obtained in a high yield of 83%. The use of thiophenol as a 20 catalyst resulted in a significantly slower reaction rate and after 4 hours the reaction had proceeded to only ~ 60% completion. After a reaction time of 16 hours, LC-MS revealed sigificant hydrolysis of palmitoyl esters. Encouraged by the successful preparation of 7, attention was again on the synthesis of glycopeptide 4 this time using the new methodology. The 25 preparation of this compound by traditional NCL was relatively low yielding due to the poor solubility of 2 in a phosphate buffer containing 6 M guanidinium hydrochloride. It was envisaged that incorporation of 2 and 3 into liposomes would increase the solubility and hence a higher yield of product may be expected. Thus, a liposomal preparation of peptide 2 and glycopeptide 3 was 30 prepared using the conditions employed for the preparation of 7. The ligation was catalyzed by the addition of sodium 2-mercaptoethane sulfonate and, after a reaction time of 2 hours, the product was purified by RP-HPLC to give 4 in an excellent yield of 78%. 25 WO 2007/146070 PCT/US2007/013431 Interestingly, no product formation was observed when a solution of 3 was added to a liposomal preparation of 2 using sodium 2-mercaptoethane sulfonate as the promoter (compound 3 has reasonable solubility in phosphate buffer). The results of these experiments indicate that NCL takes place within 5 the lipid environment of the liposome and not at the water-liposome interface. To examine the utility of the approach, compounds 10 (Scheme 2; Fig. 4), 11, and 12 (Scheme 3; Fig. 5), which differ in (glyco)peptide and lipid composition, were prepared by sequential liposome- mediated NCL starting from building blocks 2, 3, 6, 8, and 9. Thus, glycolipopeptide 10 could easily be 10 obtained by ligation of 5, which was prepared from compounds 2 and 3 with thioester 8.. Derivatives 11 and 12 were prepared by ligation of 3 with 9 to give glycopeptide 13, which after removal of the Acm group (--14) was ligated with thioesters 6 or 8, respectively. In each liposome-mediated NCL the thioester was consumed within 2 hours as determined by LC-MS, and after purification by 15 semi-preparative RP-HPLC the glycopeptides or glycolipopeptides were obtained in high yield. Previously, Kochendoerfer and co-workers (Hunter et al., Bioconj. Chem. 2004, 15, 437-440) performed a NCL between a synthetic hydrophobic polypeptide incorporated into a cubic lipidic phase and a tetrapeptide, which was 20 added to the membrane preparation. This mode of ligation is different from the approach described here because only one of the two reactants is incorporated into the membrane. Furthermore, Otaka and coworkers (Otaka et al., Chem. Commun. 2004, 1722-1723) reported that lipid bilayer assisted NCL between a thioester and an N-terminal cysteine peptide can successfully be used for the 25 synthesis of membrane protein segments possessing two transmembrane regions and one extracellular domain. In this approach, peptides were embedded in a palmitoyloleoyl phosphatidylcholine membrane and the reaction was catalyzed by the addition of thiophenol. The results of our study demonstrate that incorporation of a lipophilic 30 (lipo)peptide thioester and an N-terminal cysteine glycopeptide into DPC liposomes facilitates NCL to afford a range of glycopeptides and glycolipopeptides. Surprisingly, the new approach is not limited to peptides that have a trans- and an extra cellular domain. Furthermore, it was found that 2 mercaptoethane sulfonate is a more effective catalyst compared to thiophenol. In 26 WO 2007/146070 PCT/US2007/013431 this respect, it was observed that the liposome- mediated NCLs were completed within 2 hours, which is remarkably fast for the type of substrates employed. The high reaction rate can probably be attributed to a concentration effect in the liposomes. 5 In conclusion, we have developed a novel approach for native chemical ligation by the entrapment of reactants in liposomes. The new methodology is particularly suited for the synthesis of lipophilic (glyco)peptides of biological importance (Guo et al., Med. Res. Rev. 2005, 25, 655-678; Buskas et al., Glycobiology, 2006, 16, 113R-136R; Dube et al., Nat. Rev. Drug Disc. 2005, 4, 10 477-488; Doores et al., Chem. Eur. J. 2006, 12, 656-665; Macmillan et al., Angew. Chem. Int. Ed. 2004, 43, 1355-1359; Dziadek et al., Angew.Chem. Int. Ed. 2005, 44, 7624-7630). For example, it allows the synthesis of a range of three-component vaccine candidates by a modular approach using an array of B and T-epitopes and lipopeptide adjuvants. A modular approach is attractive 15 because it provides greater synthetic flexibility than linear synthesis. In this respect, each building block can be used for the preparation of several different target compounds. Furthermore, compared to conventional linear SPPS, a block synthetic approach will minimize by-product build-up in the growing peptide chain. In this respect, the DT sequence of the MUC-1 glycopeptide is prone to 20 aspartimide formation (Mergler et al., J. Pept. Sci. 2003, 9, 518-526) which can occur at each coupling step. In a convergent block synthesis, the individual building blocks can be purified by RP-HPLC and characterized by NMR and MS prior to assembly, providing a sound basis for highly pure final products. 25 Materials and Methods Reagents and general experimental procedures: Amino acid derivatives and resins were purchased from NovaBioChem and Applied Biosystems; DMF from EM Science; and NMP from Applied Biosystems. Dodecyl phosphocholine was obtained from Avanti Polar Lipids. All other 30 chemical reagents were purchased form Aldrich, Acros, Alfa Aesar and Fischer and used without further purification. All solvents employed were reagent grade. Reverse Phase HPLC was performed on an Agilent 1100 series system equipped with an autosampler, UV-detector and fraction-collector. RP-HPLC was carried 27 WO 2007/146070 PCT/US2007/013431 out by using a Zorbax Eclipse C8 analytical column (5 urm, 4.6 x 150 mm) at a flow rate of 1 ml/min, a semi-preparative CS column (5 pm, 25 x 250 mm) at a flow rate of 4 ml/min, a Synchropak C4 analytical column (5 pm, 4.6 x 100 mm) at a flow rate of 1 ml/min and a Vydac C4 semi preparative column (5 im, 4.6 x 5 250 mm) at a flow rate of 2 ml/min. All runs used linear gradients of 0-95% solvent B in A over a 40 min. period unless otherwise specified. (A = 0.1% TFA in water, B= 0.1% TFA in acetonitrile). MALDI-ToF mass spectra were recorded on a ABI 4700 proteomic analyzer. General methods for Solid-Phase Peptide Synthesis (SPPS): Peptides 10 were synthesized by established protocols on a Applied Biosystems, ABI 433A peptide synthesizer equipped with UV-detector using Na-Fmoc-protected amino acids and 2-(1H-bezotriazole-1-yl)-oxy-1,1,3,3-tetramethyl hexafluorophosphate (HBTU)/1 -Hydroxybenzotriazole (HOBt) as the activating reagents. Single coupling steps were performed with conditional capping. The coupling of the 15 glycosylated amino acid N"-Fmoc-Thr-(Ac 3 -a-D-GalNAc) and N-Fmoc-R-(2,3 bis(palmitoyloxy)-(2R-propyl)-(R)-cysteine were carried out manually. The manual couplings were monitored by standard Kaiser test. Synthesis of Cys-MUCI glycopeptide (20): The synthesis of Cys glycopeptide (3) is shown in Scheme 4 (Fig. 6). SPPS was performed on a Rink 20 amide linker resin (0.1 mmol) as described above. Side chain protection was as follows: N"-Fmoc-Arg (2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl), Na-Fmoc-Asp(O-tert-butyl), N"-Fmoc-Cys(Trt), N-Fmoc-Ser(tert.-butyl), Na_ Fmoc-Thr(tert.-butyl). The first four amino acids, Arg-Pro-Ala-Pro were coupled on the peptide synthesizer using a standard protocol. After the completion of the 25 synthesis, a manual coupling was carried out using Na-Fmoc-Thr-(AcO 3

-X-D

GalNAc) (0.4 mmol, 268 mg), with PyBOP (0.4 mmol, 208 mg), HOBt (0.4 mmol, 55 mg) and DIPEA (0.4 mmol, 70 pl) in DMF for 12 hrs. The coupling reaction was monitored by standard Kaiser test. The resin was washed with DMF (6 ml) and DCM (6 ml), and resubjected to the same coupling conditions to 30 ensure complete coupling. The glycopeptide was then elongated on peptide synthesizer. The resin was thoroughly washed with DMF (6 ml), DCM (6 ml) and McOH (6 ml) and dried in vacuo to constant weight. The resin was then swelled in DCM (5 ml) for 1 hr. After which it was treated with 94 % TFA, 28 WO 2007/146070 PCT/US2007/013431 2.5% water, 2.5% EDT and 1% TIS (10 ml) for 2 hr at room temperature. The resin was filtered and washed with neat TFA (2 ml). The filtrate was then concentrated in vacuo approximately 1/3 of its original volume. The peptide was then precipitated using diethyl ether (0"C) and recovered by centrifugation at 5 3000 rpm for 15 min. The crude glycopeptide was purified by RP-HPLC on a semi-preparative C-18 reversed phase column using a linear gradient of 0-95% solvent B in A over a period of 40 min., and lyophilization of the appropriate fractions afforded 20 (90% based on resin loading capacity). MALDI-ToF MS: observed, 1443.8918Da; calculated, 1443.5371Da. 10 Deacetylation of Cys-MUCI-glycopeptide (3): The glycopeptide 20 (5 mg, 3.4 pmol) was treated with 5% aqueous hydrazine (2 ml) containing excess of DTT (12 mg), the reaction was monitored by MALDI-ToF MS. After standing for 1 hr at room temperature, the crude product was purified by RP HPLC on a semi-preparative C-18 reversed phase column using a linear gradient 15 of 0-95% solvent B in A over a period of 40 min., to afford after lyophilization compound 3 (4.0 mg, 88%). MALDI-ToF MS: observed, 1317.9580Da; calculated, 1317.42711Da. Synthesis of C (Acm)YAFKYARHANVGRNAFELFLGCOSBn (2): The synthesis of Cys(Acm)-thioester peptide (2) is shown in Scheme 5 (Fig. 7). 20 The synthesis of Acm protected peptide thioester was carried out on preloaded H-Gly-sulfamylbutyryl Novasyn TG resin (0.1 mmol) as described in the general methods section for peptide synthesis. The following side chain protection was employed: N 0 -Fmoc-Arg(2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl), Na-Fmoc-Asn(Trt), Na-Fmoc-Cys(Acm), Na-Glu(O-tert.-butyl), Glu(O-tert.

25 butyl), Na-His(Trt), Na-Fmoc-Lys(Boc), N"-Fmoc-Thr(tert.-butyl), N"-Fmoc Tyr(tert -butyl). Activation and cleavage. The resin bound peptide was washed thoroughly with DCM (10 ml) and N-methyl-2-pyrrolidone (NMP) until the swelling was complete (1 hr). The resin was then treated with DIPEA (0.5 ml, 3 30 mmol), iodoacetonitrile (0.36 ml, 5 mmol) in NMP (6 ml). Before addition, iodoacetonitrile (0.36 ml) was filtered through a plug of basic alumina. The resin was then agitated under the exclusion of light for 24 hrs, filtered and then washed with NMP (20 ml), DCM (20 ml) and THF (20 ml). The activated N 29 WO 2007/146070 PCT/US2007/013431 acyl sulfonamide resin was swollen in DCM (5 ml), drained and then transferred to a 50 ml round bottom flask. To the resin-containing flask was added THF (4 ml) and benzyl mercaptan (0.64 ml, 5 mmol), and sodium thiophenate (27 mg, 0.2 mmol). After agitation for 24 hrs, the resin was filtered and washed with 5 DMF (3 ml). The combined filtrate and washings were collected and concentrated in vacuo. The crude peptide was triturated with tert-butyl methyl ether (O*C) (60 ml). Side chain deprotection: The protected peptide was treated with of reagent B (5 ml, (TFA 88%, phenol 5%, H 2 0 5%, TIS 2%)) for 6 hrs at room 10 temperature. The TFA solution was then added drop wise to a screw cap centrifuge tube containing ice cold tert-butyl methyl ether (40 ml) and the resulting suspension was left overnight at 4*C, after which the precipitate was collected by centrifugation at 3000 rpm (20 min), and after the decanting of the ether the peptide precipitate was resuspended in ice cold tert-butyl methyl ether 15 (40 ml) and the process of washing was repeated twice. The crude peptide was purified by semi preparative C-8 reversed phase column using a linear gradient of 0-95% solvent B in A over a period of 40 min., and lyophilization of the appropriate fractions afforded 2 in good yield (79% based on resin loading capacity). MALDI-ToF MS: observed, [M+Na] 2748.2439Da; calculated, 20 [M+Na] 2748.1584Da. Synthesis of lipopeptide thioester (6). The chemical synthesis of lipopeptide thioester (6) is shown in Scheme 6 (Fig. 8). The synthesis of 6 was carried out on a H-Gly-sulfamylbutyryl Novasyn TG resin (0.1 mmol) as described in the general methods. After coupling of the first five amino acids, the 25 remaining steps were performed manually. N-Fmoc-R-(2,3-bis (palmitoyloxy) (2R-propyl)-(R)-cysteine (267 mg, 0.3 mmol) was dissolved in DMF (5 ml) and PyBOP (156.12 mg, 0.3 mmol), HOBt (40 mg, 0.3 mmol) and DIPEA (67 pl, 0.4 mmol) were added. After premixing for 2 min, the mixture was added to the resin. The coupling reaction was monitored by the Kaiser test. Upon completion 30 of the coupling, the N-Fmoc group was cleaved using 20% piperidine in DMF (6 ml). Palmitic acid (77 mg, 0.3 mmol) was coupled to the free amine as described above using PyBOP (156.12 mg, 0.3 mmol), HOBt (40 mg, 0.3 mmol) and DIPEA (67 pl, 0.4 mmol) in DMF. The resin was thoroughly washed with DMF 30 WO 2007/146070 PCT/US2007/013431 (10 ml), DCM (10 ml) and MeOH (10 ml) and then dried in vacuo. Side chain deprotection was carried out by using the method described for peptide 2. The crude peptide was purified by HPLC on a semi preparative C-4 reversed phase column using a linear gradient of 0-95% solvent B in A over a 40 min., and the 5 appropriate fractions were lyophilized to afford 6 (65% based on resin loading capacity). MALDI-ToF MS: observed, [M+Na] 1695.2335Da; calculated, [M+Na] 1695.4714Da. Synthesis of lipopeptide thioester (8). The chemical synthesis of lipidated amino acid thioester (8) is shown in Scheme 7 (Fig. 9). The synthesis 10 of 8 was carried out on a H-Gly-sulfamylbutyryl Novasyn TG resin (0.1 mmol) by a manual procedure. N-a-Fmoc-Gly-OH (90 mg, 0.3 mmol) was dissolved in DMF (5 ml) and PyBOP (156.12 mg, 0.3 mmol), HOBt (40 mg, 0.3 mmol) and DIPEA (67 pl, 0.4 mmol) were added. After standing for 2 min, the mixture was added to the resin. The coupling reaction was monitored by Kaiser test. Upon 15 completion of the coupling, the N-Fmoc group was cleaved using 20% piperidine in DMF (6 ml). N-a-Fmoc-Lipidated amino acid 31 (139.57 mg, 0.3 mmol) was coupled to the free amine of the resulting product as described above using PyBOP (156.12 mg, 0.3 mmol), HOBt (40 mg, 0.3 mmol) and DIPEA (67 pl, 0.4 mmol) in DMF. This cycle was repeated twice. Finally, the N-Fmoc 20 group was cleaved using 20% piperidine in DMF (6 ml) and acetylated using 5 ml of 10% Ac 2 0, 5% DIPEA in NMP for 10 min. The resin was thoroughly washed with DMF (10 ml), DCM (10 ml) and.MeOH (10 ml) and dried in vacuo. The product was cleaved from the resin by using the method described for peptide 2. The crude peptide was purified by HPLC on a semi preparative C 25 4 reversed phase column using a linear gradient of 0-95% solvent B in A over a period of 40 min., and the appropriate fractions were lyophilized to afford 8 (69% based on resin loading capacity). MALDI-ToF MS: observed, [M+Na] 753.4871Da; calculated, [M+Na] 753.5067Da. Synthesis of C(Acm)KLFAVWKITYKDTGCOSBn (9): Cys (Acm) 30 T-epitope thioester (9) is shown in Scheme 9 (Fig. 11). The synthesis of Acm protected peptide thioester was carried out on preloaded H-Gly-sulfamylbutyryl Novasyn TG resin (0.1 mmol) as described in the general methods section for peptide synthesis. Side chain protection was as follows: Na-Fmoc-Asp(O-tert.

31 WO 2007/146070 PCT/US2007/013431 butyl), Na-Fmoc-Cys(Acm), Na-Fmoc-Lys(Boc), Na-Thr(tert.-butyl), Na-Fmoc Tyr(tert.-butyl). Activation, cleavage and side chain deprotection was performed by the method described for compound 2. The crude peptide was purified by semi preparative C-8 reversed phase column using a linear gradient of 0-95% of 5 solvent B over A over period of 40 min., and lyophilization of the appropriate fractions afforded 9 in good yield (74% based on resin loading capacity). MALDI-ToF MS: observed, [M+Na] 1972.1240Da; calculated, [M+Na] 1973.3716Da. Ligation between 2 and 3 to give 5. Method A. The peptide thioester 2 10 (10 mg, 3.6 pmol) and peptide 3 (7.24 mg, 5.5 Vtmol) were dissolved in 6 M Gn.HCI, 200 mM sodium phosphate (pH 7.5) as 1:1.5 ratios to obtain final concentration of 1 mM. The ligation was started by the addition- of 4% thiophenol (300 pl). The ligation reaction was carried out in an incubator at 37"C and the progress of the reaction was periodically monitored by RP-HPLC 15 and LC-MS. After a reaction time of 18 hrs, the reaction was diluted with 2 mercaptoethanol in ligation buffer (3 ml). The resulting mixture was then purified by C-8 semi-preparative reversed phase column using linear gradients of 0-95% solvent B in A over 40 min., and the appropriate fractions were collected and lyophilized to give 4 (6.7 mg, 48%) . The Acm protecting group of the 20 ligated product was removed by dissolving the glycopeptide in 10% aq. AcOH (2 ml) (pH 4.0) followed by the treatment of Hg (II) acetate (8.18 mg) for 30 min., the reaction was quenched by addition of DTT (5.27 mg). The Acm deprotected product was purified by semi-preparative RP-HPLC using a water/acetonitrile gradient to yield 5 (5.7 mg, 87%). MALDI-ToF MS: observed, 25 3847.6615 Da, calculated, 3847.3031 Da. Method B. The peptide thioester 2 (2 mg, 0.73 pimol) and peptide 3 (1.44 mg, 1.1 pLmol), and dodecyl phosphocholine (1.5 mg, 4.4 pmol) were dissolved in a mixture of trifluoroethanol and CHC1 3 (2.5 ml/ 2.5 ml). The solvents were removed under reduced pressure to give a lipid/peptide film on the surface of the 30 round bottom flask. The lipid/peptide film was hydrated for 4 hours at 37*C using 200 mM phosphate buffer (pH 7.5, 2 ml) in the presence of tris(carboxyethyl)phosphine (2% w/v) and EDTA (0.1% w/v). The mixture was ultrasonicated for 1 min. The peptide/lipid suspension was extruded through 1.0 32 WO 2007/146070 PCT/US2007/013431 pm polycarbonate membranes (Whatman, Nucleopore, Track-Etch Membrane) at 50*C to obtain uniform vesicles. To the vesicle suspension was added sodium 2-mercaptoethane sulfonate (2% w/v) to initiate the ligation reaction. The reaction was carried out in an incubator at 37"C and was complete within 2 5 hours. The reaction was then diluted with 2-mercaptoethanol in ligation buffer (2 ml). The resulting mixture was purified by RP-HPLC on a semi-preparative C-8 reversed phase column using a linear gradient of 0-95% solvent B in A over a 40 min., and the fraction possessing the expected product as determined by MALDI-ToF were collected and lyophilized to give 4 (2.2 mg, 78%). The Acm 10 protecting group of the ligated product was removed by dissolving the glycopeptide in 10% aq. AcOH (2 ml) (pH 4.0) followed by the treatment of Hg(II)acetate (2.7 mg) for 30 min., the reaction was quenched by addition of DTT (1.7 mg). The Acm deprotected product was purified by semi-preparative RP-HPLC using a water/acetonitrile gradient to yield 5 (1.9 mg, 89%). MALDI 15 ToF MS: observed, 3847.6015 Da, calculated, 3847.303IDa. Sequential native chemical ligation (7 or 10) is shown in Scheme 8 (Fig. 10). Ligation between 5 and 6 to give 7: The peptide 5 (3.0 mg, 0.77 pmol) and peptide thioester 6 (1.96 mg, 1.1 ptmol) was subjected to ligation reaction 20 conditions as described in method B. The progress of the reaction was periodically monitored by MALDI-ToF which showed that the reaction was complete within 2 hours. The crude peptide was purified by semi preparative C 4 reversed phase column using a linear gradient of 0-95% solvent B in A over a 40 min., and lyophilization of the appropriate fractions afforded 7 (3.5 mg, 25 83%). MALDI-ToF MS: observed, 5392.9712Da, calculated, 5392.0171 Da. Ligation between 5 and 8 to give 10: The peptide 5 (2 mg, 0.51 pmol) and peptide thioester 8 (0.53 mg, 0.72 pmol) was subjected to ligation reaction conditions as described in method B. The progress of the reaction was periodically monitored by MALDI-ToF which showed that the reaction was 30 complete within 2 hours. The crude peptide was purified by semi preparative C 4 reversed phase column using a linear gradient of 0-95% solvent B in A over a 40 min., and lyophilization of the appropriate fractions afforded 10 (1.7 mg, 78%). MALDI-ToF MS: observed, 4454.0313Da, calculated, 4454.1791Da. 33 WO 2007/146070 PCT/US2007/013431 Ligation between 3 and 9 to give 14: The peptide 3 (5.6 mg, 4.3 pimol) and peptide thioester 9 (6.0 mg, 3.0 tmol) was subjected to ligation reaction conditions as described in method B. The progress of the reaction was periodically monitored by MALDI-ToF which showed that the most of 5 conversion within 2 hours. The resulting reaction mixture was purified by using RP-HPLC on a semi-preparative C-8 reversed phase column using linear gradients of 0-95% solvent B in A over a 40 min., the fraction possessing the expected mass were collected and lyophilized to give 13 (7.4 mg, 79%). The Acm protecting group of the ligated product was removed by dissolving the 10 glycopeptide in 10% aq. AcOH (2 ml) (pH 4.0) followed by the treatment of Hg(II)acetate (11.5 mg) for 30 min. After which the reaction was quenched by addition of DTT (7.4 mg). The Acm deprotected product was purified by semi preparative RP-HPLC using a water/acetonitnile gradient to yield 14 (5.6 mg, 77%). MALDI-ToF MS: observed, 3073.7275Da, calculated, 3072.5129Da. 15 Sequential native chemical ligation (11 or 12) is shown in Scheme 10 (Fig. 12). Ligation between 14 and 6 to give 11: The peptide 14 (1.5 mg, 0.48 pmol) and peptide thioester 6 (0.98 mg, 0.58 pimol) was subjected to ligation reaction conditions as described in method B. The progress of the reaction was 20 periodically monitored by MALDI-ToF and the reaction was complete within 2 hours. The crude peptide was purified by semi preparative C-4 reversed phase column using a linear gradient of 0-95% solvent B in A over a 40 min., and lyophilization of the appropriate fractions afforded 11 (1.8 mg, 85%). MALDI ToF MS: observed, 4622.3549Da, calculated, 4621.7785Da. 25 Ligation between 14 and 8 to give 12: The peptide 14 (3.081 mg, 1 pmol) and peptide thioester 8 (1.1 mg, 1.5 pmol) was subjected to ligation reaction conditions as described in method B. The progress of the reaction was periodically monitored by MALDI-ToF and the reaction was complete within 2 hours. The crude peptide was purified by semi preparative C-8 reversed phase 30 column using a linear gradient of 0-95% solvent B in A over a 40 min., and lyophilization of the appropriate fractions afforded 12 (2.6 mg, 73%). MALDI ToF MS: observed, 3679.6072Da, calculated, 3679.3928Da. 34 WO 2007/146070 PCT/US2007/013431 Example II. Liposome-Mediated Native Chemical Ligation in the Presence or Absence of Thiol Initiator 5 Liposome-mediated native chemical ligation between peptide thioester 38 and glycopeptide 39 having a N-terminal cysteine resulted into formation of glyco(lipo)peptide 37 (Fig. 13). This reaction was carried out in the presence and absence of catalyst required for ligation reaction such as 2-mercaptoethane sulfonate or thiophenol. Surprisingly, this reaction gave similar results, 10 indicating that the thiol initiator is not required under liposome-mediated native chemical ligation. Thus, the liposome mediated native chemical ligation can be performed in the presence or absence of thiol initiator or catalyst. The complete disclosures of all patents, patent applications including 15 provisional patent applications, and publications, and electronically available material (e.g., GenBank amino acid and nucleotide sequence submissions) cited herein are incorporated by reference. The foregoing detailed description and examples have been provided for clarity of understanding only. No unnecessary limitations are to be understood therefrom. The invention is not limited to the 20 exact details shown and described; many variations will be apparent to one skilled in the art and are intended to be included within the invention defined by the claims. 35

Claims (24)

1. A method for making a multicomponent compound comprising: mixing at least one first hydrophobic reactant comprising an N-terminal cysteine residue, at least one second hydrophobic reactant comprising a thioester, and a nonpolar, hydrophobic or amphipathic molecule capable of forming a lipidic structure; subjecting the mixture to conditions effective to form a lipidic structure in which the first and second reactants are embedded; and subjecting first and second reactants to conditions effective to allow ligation of the first reactant and the second reactant to yield a multicomponent compound comprising the first and second reactant, wherein the ligation reaction takes place within the lipidic structure; and wherein at least one of the first and second reactants is not a transmembrane protein or membrane-spanning fragment thereof.

2. A method for making a multicomponent compound comprising: providing at least one first hydrophobic reactant comprising an N-terminal cysteine residue, and at least one second hydrophobic reactant comprising a thioester; and contacting the first and second reactants with a preformed lipidic structure under conditions effective to allow solubilization of the reactants in the lipid phase; and subjecting first and second reactants to conditions effective to allow ligation of the first reactant and the second reactant to yield a multicomponent compound comprising the first and second reactant, wherein the ligation reaction takes place within the lipidic structure.

3. The method of claim 2 wherein at least one of the first and second reactant is not a transmembrane protein or membrane-spanning fragment thereof.

4. The method of any of the preceding claims wherein neither the first nor the second reactant is a transmembrane protein or membrane-spanning fragment thereof. 36

5. The method of any of the preceding claims further comprising contacting the multicomponent compound with at least one third hydrophobic reactant within a lipid structure under conditions to allow ligation of the multicomponent compound and the third reactant, to yield a multicomponent compound comprising the first, second and third reactants.

6. The method of claim 5 comprising solubilizing the multicomponent compound and the third reactant within a lipidic structure to facilitate ligation of the multicomponent to the third reactant.

7. The method of any of the preceding claims wherein the lipidic structure is selected from the group consisting of a lipid monolayer, lipid bilayer, a liposome, a micelle, a film, an emulsion, a matrix and a gel.

8. The method of any of the preceding claims wherein the ligation reaction does not take place at the interface between the lipidic structure and the external aqueous environment.

9. The method of any of the preceding claims further comprising contacting the lipidic structure with an initiator compound to catalyze the ligation.

10. The method of claim any of claims 1 to 8 wherein the ligation is performed in the absence of an initiator compound.

11. The method of any of the preceding claims wherein the lipid structure comprises an amphipathic molecule.

12. The method of any of the preceding claims wherein each of the reactants is independently selected from the group consisting of a carbohydrate component, a peptide component, a lipid component, or conjugate of any of them.

13. The method of any of the preceding claims wherein at least one reactant comprises a T-epitope. 37

14. The method of any of the preceding claims wherein at least one reactant comprises a B-epitope.

15. The method of claim 13 wherein the B-epitope is from a microorganism selected from the group consisting of a virus, a bacterium, a fungus, and a protozoan.

16. The method of claim 15 wherein the microorganism is a human immunodeficiency virus or a hepatitis C virus.

17. The method of claim 14 wherein the B-epitope is overexpressed on a cancer cell.

18. The method of claim 12 wherein the carbohydrate component comprises a self-antigen.

19. The method of claim 18 wherein the self-antigen comprises a MUC-1 glycopeptide.

20. The method of any of the preceding claims wherein at least one reactant comprises a glycoconjugate selected from the group consisting of a glycosylated protein, a glycosylated peptide, a glycosylated lipid, a glycosylated amino acid, a DNA and an RNA.

21. The method of any of the preceding claims wherein at least one reactant comprises a lipopeptide adjuvant.

22. The method of claim 12 wherein the lipid component comprises a Toll-like receptor (TLR) ligand.

23. The method of claim 22 wherein the Toll-like receptor ligand comprises Pam 3 Cys or Pam 3 CysSKn, wherein n = 0, 1, 2, 3, 4 or 5.

24. The method of claim 12 wherein the lipid component comprises Pam 3 CysSK 4 . 38