SYNTHETIC ARCHAEAL GLYCOLIPID ADJUVANTS

FIELD OF THE INVENTION

This invention relates to the preparation of synthetic archaeal polar lipids 5 whose structures are selected to achieve a desired immune system activating activity.

BACKGROUND OF THE INVENTION

Archaeal polar lipids are novel to the domain of life Archaea and are

10 characterized as isoprenoid ether lipids of opposite sn-2,3 stereochemistry (12).

Archaeosome vaccine adjuvants comprising natural lipid mixtures extracted from an archaeon have previously been disclosed. Such adjuvants alleviate the need for effective induction of humoral (Th2), cell-mediated (ThI), and particularly CD8 ⁺ cytotoxic T cell responses (CTL) to an antigen (25, 28). The disclosed lipids are

15 restricted to the polar lipids extracted from archaeal biomass.

Best long-term adjuvant activity and memory responses occurred with archaeosomes prepared from the total polar lipids (TPL) of Methanobrevibacter smithii and Thermoplasma acidophilum (16), both of which have challenges in their preparation. M. smithii is an obligate anaerobe, requiring specialized medium for

20 growth, including toxic sulfides and potentially flammable, explosive levels (80%) of hydrogen gas. Methanogens such as M. smithii must be protected from the lethal effects of oxygen in air. Tliermoplasma and Sulfolobus TPLs consist of about 90% caldarchaeol membrane-spanning lipids that hydrate to form archaeosomes at low recoveries, in our experience of about 10-20%. Purification of lipids from lipid

25 extracts is uncertain, tedious and costly. A more efficient, cost-effective method to produce archaeal polar lipids would be a great advantage.

Further, use of archaeal TPL natural mixtures limits the adjuvant composition to only those lipids, and in the proportions, that are extracted from archaeal species. This approach is likely to achieve stable archaeosomes but may not be

30 optimal for a selected application. The lipid composition of the vaccine may theoretically determine whether protection occurs or not, based on the type of

SUBSTITUTE SHEET (RULE 26)

immune response (MHC class I or II presentation- Fig. IA, or systemic versus mucosal) that is generated to the antigen.

For human applications the lipid mixture used for archaeosome formation needs to be defined and reproducibly produced from each batch of biomass 5 grown and extracted. Recovery of mixtures of often 10 or more different polar lipids in each TPL extract is typical. Production benefits, including more control of head group in/out orientation on the archaeosome surface, as well as ease of obtaining regulatory approval for human use of defined and simplified compositions, may be anticipated.

10 Natural archaeal core lipids are predominantly of two types; namely, archaeol and its dimer called caldarchaeol (Fig. IB). Both of these core lipids or their analogues may be synthesized chemically (3, 21). However, chemical synthesis of archaeal core lipids is complex and must consider the problems of producing mixed stereoisomers (methyl groups of archaeal isoprenoid chains are R) and of

15 generating unwanted chemical by-products. In addition, the archaeal sn-2,3 stereochemistry must be adhered to.

DESCRIPTION OF THE PRIOR ART

Friesleben et al (8) have grown Thermoplasma acidophilum and obtained a 20 caldarchaeol lipid core. The 2 free hydroxyl groups on the caldarchaeol were reacted to form carboxylic acid groups, then dicarboxylic acid chloride, then reacted with dimethylaminopropylamine and finally with dimethylsulfate. The result was a positively charged caldarchaeol capable of binding DNA by charge- charge interactions, for the purpose of transfecting cells.

25 Benvegnu et al (2) prepared a synthetic tetraether lipid by linking 2 glycerols by C15 saturated polymethylene chains through a cyclopentane group and attaching lactose or phosphatidylcholine groups to the 2 free hydroxyls. This synthetic tetraether lipid differs considerably in structure from archaeal caldarchaeols with unpredictable changes in biological responses. These authors 30 report the relative ability of these synthetic tetraether lipids to form liposomes, and suggest their stability properties may be useful for oral drug delivery.

Three archaeal lipids found in Halobacterium salinarum have been synthesized for studies on the light driven proton pump of purple membrane (32). These were O-(β-D-glucopyranosyl 3-sulfate)-(l-6)-O-α-D-mannopyranosyl-(l- 2)-O-α-D-glucopyranosyl-(l-l)-2,3-di-O-phytanyl-5«-glycero l, and O-β-D- 5 galactopyranosyl-(l-6)-O-α-D-mannopyranosyl-(l-2)-O-α-D-gl ucopyranosyl-(l- l)-2,3-di-O-phytanyl- ₁ s«-glycerol, and O-β-D-glucopyranosyl-(l-6)-O-α-D- mannopyranosyl-(l-2)-O-α-D-glucopyranosyl-(l-l)-2,3-di-O-ph ytanyl- ₁ sn- glycerol. The complexity of these lipids makes them unattractive for synthesis as vaccine adjuvants, and additionally the TPL from H. salinarum which contains 10 these lipids is unsatisfactory in generating only short-term CTL responses (16).

In another study on the physical chemistry of membranes a l,2-diphytanyl-3- O-β-D-glucosyl-^n-glycerol was made (4).

Finally, synthesis has been reported for a series of l,3-di-O-alkyl-2-O-(β- glycosyl)glycerols (18). These glycolipids differ from archaeal lipids and from 15 the synthetic glycolipids reported herein by linkage of the head group specifically to the sn-2 of the glycerol with phytanyl chains at sn- 1,3.

In none of the cited literature above is there speculation or suggestion of antigen delivery for vaccine applications. Indeed, in the Benvegnu et al report (2) it may be expected that too much stability would teach away from adjuvant properties by

20 preventing antigen release in vivo. Further, adjuvant activity has never been demonstrated previously for any synthetic archaeal lipid.

In Sprott et al (23) it is stated 'a striking decline in adjuvant activity was observed upon incorporation of 36 mol% of the uncharged lipid DGAs,' where DGAs is biologically purified β-D-Glc-(l,6)-β-D-Glc-archaeol. This finding 25 teaches away from expectation that glyco-archaeols and specifically β-D-Glc-

(l,6)-β-D-Glc-archaeol or β-D-Glc-(l,6)-β-D-Glc-caldarchaeols would promote adjuvant activity.

SUMMARY OF THE INVENTION

30 Archaeal lipid cores including archaeol and caldarchaeol are obtained by methanolic-HCl hydrolysis of the polar lipids extracted from archaea such as

Halobacterium salinarum or Thermoplasma acidophilum, respectively. The source of archaeal core lipids is preferred to be those archaea that have predominantly or wholly saturated, acid-labile core lipids to avoid instability. Carbohydrate polar head groups or anionic polar head groups of desired type, 5 number, linkage and configuration are then chemically coupled to the archaeal lipid cores, preferably at the free sn-1 hydroxyls of the lipid cores. Such synthetic polar lipids may then be used in the preparation of vaccines formulated as archaeosomes designed to serve as stable antigen carriers that target delivery to antigen-presenting cells, resulting in the appropriate balance of systemic 10 versus mucosal immunity, and CD8 ⁺ T cell and CD4 ⁺ T cell responses.

A first object of the invention is to provide for a polar synthetic lipid. The polar synthetic lipid may be derived from archaeal lipid cores isolated from archaeal cells and then modified to add carbohydrate polar head groups and/or 15 anionic polar head groups.

A second object of the invention is to provide for an archaeosome lipid vesicle composed of polar synthetic lipids. Preferably the archaeosome should include synthetic lipids that comprise lipids with carbohydrate groups and lipids with 20 anionic groups, or lipids with both carbohydrate and anionic groups. The polar synthetic lipid may act as a stabilizer, or an additional lipid, such as a conventional synthetic lipid, may be added.

A third object of the invention is to provide for a vaccine comprising an 25 archaeosome adjuvant as described above, with an antigen. The vaccine may be designed, for example, to elicit an immune response such as a protective CD8 ⁺ or CD4 ⁺ T cell response or a mucosal response.

A first aspect of the invention provides for a polar synthetic lipid comprising at 30 least one carbohydrate or anionic group linked by covalent bonding to at least one free hydroxyl group of an archaeal lipid core.

A second aspect of the invention provides for an arachaeosome comprising at least one polar synthetic lipid comprising at least one carbohydrate or anionic group linked by covalent bonding to at least one free hydroxyl group of an archaeal lipid 5 core.

A third aspect of the invention provides for a vaccine comprising an adjuvant and an antigen, the adjuvant comprising an archaeosome comprising at least one polar synthetic lipid comprising at least one carbohydrate or anionic group linked 10 by covalent bonding to at least one free hydroxyl group of an archaeal lipid core.

A further aspect of the invention provides for the use of a vaccine for the promotion of an immune response, the vaccine comprising an adjuvant and an antigen, the adjuvant comprising an archaeosome comprising at least one polar

15 synthetic lipid comprising at least one carbohydrate or anionic group linked by covalent bonding to at least one free hydroxyl group of an archaeal lipid core.

A further aspect of the invention provides for a method for producing a polar synthetic lipid comprising at least one carbohydrate or anionic group linked by 20 covalent bonding to at least one free hydroxyl group of an archaeal lipid core, the method comprising the steps of isolating an archaeal lipid core from archaeal cells, and processing the archaeal lipid core to add the at least one carbohydrate or anionic group.

25 A further aspect of the invention provides for a method for producing an archaeosome as described above, comprising the steps of isolating archaeal lipid core molecules from archaeal cells, processing the archaeal lipid core molecules to add the at least one carbohydrate or anionic group, adding at least one stabilizing lipid, and allowing an archaeosome to form.

30

BRIEF DESCRIPTION OF THE DRAWINGS

Figure IA illustrates two immune response pathways in an antigen-presenting cell, both beginning by phagocytosis of an antigen carried in an archaeosome vesicle. CD4 ⁺ T cell responses require that antigen (Ag) be released in the 5 phagolysosome for proteolysis and MHC class II presentation of peptides. Cross presentation (dashed line) of exogenous antigens is unexpected, and requires that antigen (Ag) be translocated to the cytosol for MHC class I presentation of peptides to CD8 ⁺ T cells.

10 Figure IB illustrates the molecular structures of archaeol and caldarchaeol core lipids. Notable features of the structures include ether linkages to isopranoid fully saturated chains of constant C-20 or C-40 length and opposite sn-2,3 stereochemistry of chain attachment to the glycerol backbone(s).

15 Figure 2 illustrates synthesis strategies for (A) α-D-Glc-(l,4)-α-D-Glc-(l,4)-β-

D-GIc-(I, l)-archaeol, (B) β-D-Glc-(l,6)-β-D-Glc-archaeol, and (C) α-mannose^ ₅ )-archaeols.

Figure 3 shows CTL responses in splenocytes of mice immunized

20 subcutaneously at 0 and 3 weeks with archaeosomes consisting of 15 μg OVA entrapped in either maltotriose-archaeol (maltotriose-A), mannotriose-A (α-D-

Man-( 1 ,2)-α-D-Man-( 1 ,2)-α-D-Man-( 1 , 1 )-archaeol), mannotetraose-archaeol (α-D-

Man-( 1 ,2)-α-D-Man-( 1 ,2)-α-D-Man-( 1 ,2)-α-D-Man-( 1 , 1 )-archaeol), or archaetidylglycerol (AG). Spleens were taken 7 weeks following the first

25 immunization. Comparatively low adjuvant activity is seen for archaeosomes composed of AG/chol (80/20, mol%). Choi is cholesterol. Splenic CTL responses are shown for antigen specific EG.7 targets (panel A) and non-specific

EL-4 targets (panel B) (2 mice/group). EL-4 is the negative control target cells that do not express the MHC class I specific peptide (SIINFEKL) of OVA. Naive mice

30 were non-immunized controls.

Figure 4 shows CTL responses in splenocytes of mice (A), and anti-OVA antibody responses in sera of mice (B) immunized subcutaneously at 0 and 3 weeks with 15 μg OVA entrapped in liposomes composed of DPPG/chol (80/20, mol%), DPPG/DPPS/chol (60/20/20, mol%), and archaeosomes consisting of 5 GlcrA/DPPG/chol (15/65/20, mol%) or Glc ₂ -A/DPPS/DPPG/chol (15/20/45/20, mol%). GIc ₂ -A is synthetic β-D-Glc-(l,6)-β-D-Glc-archaeol. Liposomes induced comparatively very low CTL responses compared to archaeosomes showing that the active ingredient is synthetic GIc ₂ -A. Responses for non-specific EL-4 targets (data not shown) were below 5%. CTL assays were conducted 6 weeks from first 10 injection, and antibody assays at 6 weeks (open bars) and 11 weeks (hatched bars) post first injection.

Figure 5 illustrates that adjuvant activity of synthetic OVA-archaeosomes depends on the mol% composition of synthetic Glc ₂ -archaeol (β-D-Glc-(l,6)-β-

15 D-Glc-archaeol). Cholesterol was held constant in all archaeosomes at 20mol%, and Glc ₂ -archaeol varied from 5 to 60mol%. DPPG comprised the remaining mol% of each archaeosome type. Also shown is comparatively lower adjuvant activity of mice immunized with MaIi ₄ - A/DPPG/chol (45/35/20, mol%), where MaIi ₄ -A is synthetic α-D-Man-(l,2)-α-D-Man-(l,2)-α-D-Man-(l,2)-α-D-Man-

20 (l,l)-archaeol. Each subcutaneous injection given on 0 and 3 weeks consisted of synthetic archaeosomes with 15 μg OVA entrapped. Data represent CTL (CD8 ⁺ T cell) responses measured in spleen cells taken 7 weeks post first injection where effector to target ratios (E:T) are 3.3:1 (panel A) and 11:1 (panel B). All EL-4 non-specific targets were lysed at below 1.9% (data not shown).

25

Figure 6 illustrates the molecular structures of a series of synthetic disaccharide- archaeols.

Figure 7 shows the adjuvant activity induced in mice immunized with an antigen

30 (15 μg OVA) entrapped in synthetic archaeosomes consisting of various synthetic di- and trisaccharide-archaeols (see Figure 6) /DPPG/cholesterol

(25/55/20, mol%). Immunizations were subcutaneous injections given at 0 and 3 weeks. Splenocytes were taken for CD8 ⁺ T CTL assays 6 weeks and 12 weeks post first injection. A, CTL 6 weeks; B, CTL 12 weeks; C, Elispot assay at 6 weeks comparing isomaltose-A to gentiobiose-A, and maltotriose-A to 5 gentiotriose-A; D, Anti-OVA antibody titres in blood, taken 6 weeks post first injection, n=4 mice. Elispot assay in C panel measures the numbers of IFN-γ secreting antigen-specific CD8 ⁺ T cells. Production of IFN-γ is another functional read-out of CD8 ⁺ T cell activation.

10 Figure 8 compares the CD8 ⁺ T cell adjuvant activity in mice similarly immunized with OVA-archaeosomes consisting of β-D-Glc-(l,6)-β-D-Glc- archaeol/DPPG/chol (35/35/30, mol%) and β-D-Glc-(l,6)-α-D-Glc- archaeol/DPPG/chol (35/35/30, mol%). Spleens were removed for the Elispot assays on week 6, post first injection.

15

Figure 9 shows the structures for several anionic archaeols as well as DPPS and DPPG used in synthetic glyco-archaeosome formulations.

Figure 1OA shows the antigen specific CD8 ⁺ T cell responses based on IFN-γ

20 production (Elispot) in animals injected with the protein antigen entrapped in various synthetic archaeosomes comprised of antigen and gentiotriose archaeol

(β-D-Glc-(l,6)-β-D-Glc-(l,6)-β-D-Glc-archaeol)/anionic lipid/cholesterol

(35/35/30, mol%). Structures for the anionic lipids listed on the x-axis are shown in Figure 9. Mice were injected subcutaneously on 0 and 3 weeks with 15 μg

25 OVA entrapped in the various synthetic archaeosome compositions. Elispot assays were done using spleen cells taken on weeks 6 and 8, post first injection.

Non-specific controls (no SIINFEKL peptide added) were below detection, except at 6 weeks with AG and AGP-CH ₃ where small blank values were subtracted.

30

Figure 1OB shows anti-OVA antibody responses (IgG + IgM) in blood taken 9 weeks post first injection for mice in Figure 1OA.

Figure 11 illustrates the molecular structures of a selection of polar caldarchaeol 5 lipids.

Figure 12 illustrates antigen-specific CD8 ⁺ T cell responses generated in animals immunized with archaeosomes comprised of a single caldarchaeol bipolar lipid and antigen. Strong adjuvant activity is shown for the gentiobiose 10 caldarchaetidylinositol archaeosome that contrary to expectation (10) exceeds that of the gentiobiose caldarchaetidylserine archaeosome. The assay was performed on splenic cells taken 6 weeks post first injection.

Figure 13 illustrates protection against LM-OVA challenge after vaccination with

15 OVA-adjuvant combinations. C57BL/6J mice were immunized on day 0 and 21 with 15 ug OVA entrapped in various synthetic archaeosomes composed of the respective synthetic glyco-archaeol/DPPG/chol (25/55/20, mol%). At 8 weeks, mice were challenged with Iθ4 intravenous dose of Listeria monocytogenes

(LM)-OVA construct. Mice were euthanized 3 days later. Mean ± SEM of

20 splenic colony forming units (CFU) is shown for each group. Mice previously vaccinated with OVA in synthetic archaeosomes showed ~ 2 log 10 protection compared to non-vaccinated naϊve mice.

DETAILED DESCRIPTION OF THE INVENTION

25

Archaeal core lipids are isolated from archaeal cells, following which polar synthetic lipids are synthesized by attaching various groups, including carbohydrate or anionic groups, to the archaeal core lipids. Stable archaeosomes are prepared from these polar synthetic lipids by including anionic and stabilizing

30 features. Incorporating a protein antigen into these archaeosome compositions and injecting them into mice resulted in an unexpected, strong adjuvant activity

for select compositions in immunized mice. The choice of the carbohydrate group in terms of type, number of residues, linkages and configurations were important in determining adjuvant activity. This approach avoids the possibility of encountering the presence of immune system depressing lipids, or immune 5 inactive lipids that occur in TPL extracts, and allows synthesis of potent archaeal lipids that are not found naturally in archaea.

DEFINITIONS

An archaeal lipid refers to a polar lipid common to the Domain Archaea 10 typified by isoprenoid chains in ether linkage to the sn-2,3 carbons of the glycerol backbone.

Archaeal core lipids are most commonly 2,3-di-O-sn-diphytanylglycerol (archaeol), and 2,2 ^l ,3,3'-tetra-O-dibiphytanyl-sn-diglycerol (caldarchaeol).

Synthetic archaeal lipids or polar synthetic lipids refer to core lipid 15 precursors either derived from Archaeal lipids by hydrolysis or made by chemical synthesis, conjugated to at least one new head group.

Archaeol phospholipids are referred to using archaetidyl, for example, AG, archaetidylglycerol; AS, archaetidylserine.

Conventional lipids refer to the lipids common to the Domains Bacteria 20 and Eukarya. This includes polar lipids typified by fatty acyl chains in ester linkage to the sn-1,2 carbons of the glycerol backbone, and neutral lipids such as cholesterol. Conventional phospholipids are referred to in the usual way, for example, DPPG, dipalmitoylphosphatidylglycerol; DPPS, dipalmitoylphosphatidylserine.

25 Archaeosomes refer to closed lipid vesicles that contain any amount of synthetic archaeal lipid(s).

Liposomes are lipid vesicles composed entirely of conventional lipids. APCs, antigen presenting cells CTL, cytotoxic T lymphocyte

30 TPL, total polar lipids obtained by extraction from a specific archaeon

OVA, ovalbumin

10

AgQTf, silver trifluoromethansulfonate (tπfiate) NIS, N-iodosuccmimide animal, as referred to herein humans are included glyco, sugar and carbohydrate are used interchangeably 5

In contrast to expectation, the archaeal core lipid chemically coupled with carbohydrate head groups such as β-D-Glc-(l,6)-β-D-Glc-, and certain other carbohydrate head groups not known to exist naturally in archaea, can be induced to form lipid vesicles that promote a strong CTL response in animals to an 10 entrapped antigen. Further, it is shown that the archaeol portion of an archaeal lipid is insufficient in itself to promote adjuvant activity. Thus, archaeal lipids can be categorized as either poor adjuvants or strong adjuvants based on the head groups they possess.

15 According to one aspect of the invention synthetic chemistry is used to prepare polar synthetic lipids by linking defined carbohydrate head groups (or anionic groups such as phosphoglycerol) to the free sn-1 hydroxyl group of stable archaeal core lipids. Archaeal biomass is the preferred source of the archaeal core lipid, where the archaeal source is chosen to give ease of growth

20 and also a high yield of the desired core lipid product. Halobacterium sahnarum is chosen as a good source of archaeol because it is an aerobe easily grown using inexpensive media, and produces only saturated polar lipids for increased stability. Further, it has only one core lipid structure (archaeol), and requires high concentrations of NaCl for growth simplifying the maintenance of pure cultures.

25 Further, the cells lyse in water, making lipid extraction from the cell debris very efficient. As is the case for all archaea, there is no pathogenicity associated with this environmental bacterium that lacks endotoxin and other virulence factors (9). Presence of only one core lipid results in high yields, requiring only simple purification steps.

30

11

In another aspect of the invention caldarchaeol, or any other core lipid, is prepared based on selecting the appropriate archaeon in which the desired core lipid is abundant. For caldarchaeol isolation an archaeon such as Thermoplasma may be preferred compared to Sulfolobus, to avoid mixtures of caldarchaeols and nonitol- 5 caldarchaeols found in Sulfolobus. Core lipids may be synthesized chemically to achieve the same structures, providing the same stereochemistry of archaeal core lipids is preserved.

In another aspect of the invention novel archaeal lipids not known to exist in nature are synthesized. Further, mixtures of synthetic archaeal lipids are used in the

10 proportions necessary to optimise the desired adjuvant activity, namely CD4 ⁺ T cell,

CD8 ⁺ T cell, or mucosal immunity, of the archaeosomes so formed.

In a further aspect of this invention saturated archaeal lipids with isopranoid chains and ether linkages to the glycerol backbone are used for the synthesis of polar synthetic lipids to give the distinct advantage of chemical stability and allow

15 the chemical coupling and de-blocking steps used in the synthesis to proceed without generating undesired products. In comparison, lipids from sources other than archaea are characterized by unsaturation in their fatty acids that are linked by relatively unstable ester bonds to the glycerol backbone. Archaeal species that have unsaturation in their polar lipids, such as Haloferax volcanii, or acid-sensitive 3-

20 hydroxylated isoprenoid chains (24) are generally avoided (26). A further reason for use of archaeal core lipids is that the stereochemistry of archaeal lipids is sn-2,3 versus sn-1,2 for glycerolipids of Bacteria and Eukarya. Although it is not proven, this stereochemistry may be critical for adjuvant activity, as the adjuvant activity of polar glycerolipids of Bacteria and Eukarya is low compared to the glycerolipids of

25 Archaea.

In another aspect of the invention an antigen that may be a surface molecule or epitope (such as an immunodominant amino acid sequence of a protein) expressed by a pathogen or cancer cell, is entrapped in archaeosomes comprised of polar head groups linked synthetically to archaeol/caldarchaeol to form a protective vaccine.

30 According to another aspect of the invention the issue of difficulty in hydrating and forming archaeosomes from caldarchaeol polar lipid mixtures, as

12

well as from other lipids where the problem occurs, may be solved by constructing an archaeosome of, for example, (1) only archaeol synthetic lipids as the vaccine adjuvant, (2) a mixture of synthetic archaeol/caldarchaeol glycolipids with synthetic anionic archaeol/caldarchaeol lipids or commercially 5 available lipids including DPPG and/or stabilizing cholesterol, or (3) a single synthetic polar caldarchaeol structure with targeting glyco group on one end and anionic group on the other and that is synthesized to be conducive to stable archaeosome formation. These are intended as representative examples only, and it is understood that other possible combinations of archaeal lipids may be

10 possible.

Yet another aspect of the invention provides for a method to elicit an antigen specific, protective MHC class I restricted cytotoxic T cell response (CD8 ⁺ T cell response) and an antigen specific MHC class II response (CD4 ⁺ T cell response) in an animal, and/or a mucosal response, wherein the synthetic archaeosome

15 vaccine composition formulated with antigen is administered to the animal.

Synthetic archaeols - synthesis and adjuvant activity

Strong CTL adjuvant activity is found for OVA-archaeosomes comprising a synthetic glyco-archaeol lipid and antigen. The first synthetic lipids synthesized 20 (Fig. 2) and tested were: α-D-Man-(l,2)- α-D-Man-(l,2)- α-D-Man-(l,l)-archaeol; α-D-Man-(l,2)- α-D-Man-(l,2)- α-D-Man-(l,2)- α-D-Man-(l,l)-archaeol; and α-D-Glc-(l,4)-α-D-Glc-(l,4)-β-D-Glc-(l,l)-archaeol (Fig. 3). To adjuvant the

CTL response, four mannose residues were preferred to three. However, three

25 glucose residues gave best adjuvant activity. This higher activity for α-glucose residues was surprising as macrophages have a well-known mannose receptor to better promote phagocytosis. However, phagocytosis of glyco-archaeal lipids via a mannose receptor, or any other receptor, has not been shown to date. Dectin-1, the recently discovered β-glucan receptor of APCs, is excluded as a mechanism

30 to explain the adjuvant properties of synthetic glyco-archaeols, as that receptor is specific to polysaccharides of at least 10 to 11 glucose residues and has

13

specificity 'exclusively' for β-l,3-linkages (19). Thus, synthetic glyco-archaeols interact with APCs and adjuvant by a new and unexpected mechanism.

Archaeosomes composed of archaetidylglycerol (AG, the archaeal form of phosphatidylglycerol) and cholesterol (80/20, mol%) with antigen (OVA)

5 entrapped, were tested as adjuvants (Fig. 3). These had little ability to serve as adjuvants in animals, showing that the archaeol lipid moiety was insufficient to function as a strong adjuvant per se, without an appropriate head group.

Archaeosomes did not form readily upon attempts to hydrate shorter chain synthetic α-D-Man-archaeol; α-D-Man-(l,2)-α-D-Man-archaeol; or β-D-Glc-(l,6)-

10 β-D-Glc-archaeol. This difficulty in forming lipid vesicles, and a problem of aggregation noted for the archaeosomes in Fig. 3, was overcome by inclusion of an anionic lipid (DPPG) and cholesterol or DPPG/DPPS/cholesterol lipids. This solution also provided a means of testing various synthetic glyco-archaeols for adjuvant activity. DPPG/cholesterol (80/20, mol%) or DPPG/DPPS/cholesterol

15 (60/20/20, mol%) liposomes had little ability to induce a CTL response or an antibody response in animals to the entrapped antigen (Fig. 4). Instability to prolonged storage with the possible loss of entrapped antigen (and therefore loss of adjuvant activity) was ruled out as the mechanism for poor adjuvant activity in the DPPG, DPPS and cholesterol liposomes (Table 10). Incorporation of only 15 mol%

20 of synthetic β-D-Glc-(l,6)-β-D-Glc-(l,l)-archaeol to either of these liposome formulations resulted in a dramatic increase in CTL and antibody responses in mice (Fig. 4A,B)- This is unanticipated based on previous data (13) showing that the adjuvant activity of archaeal lipids dramatically diminished when mixed with non- archaeal lipids (DMPC/DMPG), and data (23) showing a diminished adjuvant

25 activity with incorporation of biologically purified β-D-Glc-(l,6)-β-D-Glc- archaeol into archaeosomes.

Presence of a relationship was explored between adjuvant activity and the mol% of the active ingredient, β-D-Glc-(l,6)-β-D-Glc-archaeol, mixed with DPPG/cholesterol. Cholesterol was maintained constant at 20mol% with DPPG

30 added as required to balance increasing amounts of β-D-Glc-(l,6)-β-D-Glc- archaeol. Surprisingly, liposome formation upon hydration of DPPG/cholesterol

14

(80/20, mol%) was improved by including β-D-Glc-(l,6)-β-D-Glc-archaeol up to 35mol%. Hydration became more difficult at 60mol% D-GIc-(1, 6)-β-D-Glc- archaeol, and archaeosomes with 60mol% of the synthetic glyco-archaeol increased in size during the removal of unentrapped antigen indicating instability 5 (Table 11). Microscopic examination revealed that in contrast to the other preparations shown in Table 11 with exception of DPPG/Glc ₂ -A (65/35, mol%), archaeosomes containing 60mol% D-GIc-(1, 6)-β-D-Glc-archaeol began to convert from vesicles to non-vesicular structures within several days at 4 ⁰ C. The loading of an antigen that was achieved in these archaeosome types is also shown

10 in Table 11. Immunized mice developed CD8 ⁺ T cell immune responses that were best at 15-45mol% Glc ₂ -archaeol content (Fig. 5).

The need for cholesterol, and the optimal mol% cholesterol needed to achieve stability of archaeosomes containing 35mol% D-GIc-(1, 6)-β-D-Glc-archaeol was explored. Cholesterol was varied in each preparation from 0, 10, 20, 30 and

15 45mol%. DPPG made up the balance of each preparation. Lipids were mixed in solvent and archaeosomes were loaded with the antigen OVA as described in methods. Archaeosomes formed in all cases upon hydrating at 35°C. However, when stored at refrigeration temperatures (4-6 ⁰ C) these synthetic archaeosomes with no cholesterol were unstable, converting to amorphous lipid debris and

20 needle shaped crystals. This instability was avoided by inclusion of 10mol% or more cholesterol. As cholesterol was increased to 45mol% some instability became evident, as seen from the increase in size of this preparation (Table 11). The optimal cholesterol was therefore in the range of >10 to <45 mol%.

Various disaccharide head groups were coupled to archaeol to form a series of

25 new synthetic archaeal glycolipids (Fig. 6). These were formulated with antigen for animal trials, with lipids comprised of the various disaccharide-archaeols / DPPG / cholesterol (25/55/20, mol%). Average diameters and antigen loading properties are shown in Table 11. CTL activities in animals show importance for both carbohydrates to be glucose with 1-6 linkage preferred to 1-4 in the shorter term

30 (Fig. 7A, 6 weeks). In the longer term (Fig. 7B, 12 weeks) CTL adjuvant activity was strong also for lactose-archaeol and melibiose-archaeol where sugars are linked

15

gal-glc-archaeol. The preferred linkage configuration between the two glucose units was compared in Fig. 7C using synthetic isomaltose-archaeol (α-(l,6)) and gentiobiose-archaeol (β-(l,6)). CD8 ⁺ T cell adjuvant activity was best in archaeosomes containing the synthetic diglucose-archaeol with β-(l,6) linkage, and 5 this preference for a β-glc-glc linkage is seen also by comparison of gentiotriose-A and maltotriose-A (Fig. 7C). hi contrast, antibody responses in mice were quite strong for isomaltose-A (α-(l,6)) archaeosomes, although preference was still for the β-(l,6) linkage (Fig.7D).

Archaeosomes composed of synthetic β-D-Glc-(l,6)-β-D-Glc-(l,6)-β-D-Glc-

10 (l,l)-archaeol/DPPG/cholesterol (35:45:20, mol%) were also stable and entrapped the antigen OVA similarly to the gentiobiose-archaeol archaeosomes (Table 11). Generally, three sugar moities coupled to archaeol were preferred to two in terms of hydration ease and immune responses achieved. In one example, the Elispot assay revealed an antigen specific CD8 ⁺ T cell response 6 weeks from first injection in

15 mice immunized with antigen-containing gentiotriose-A/DPPG/cholesterol

(25/55/20, mol%) to be 2 to 5 times higher than with gentiobiose- A/DPPG/cholesterol (25/55/20, mol%).

It is recognized that multiple receptor engagement on APCs may be achieved by including more than one type of glyco head group in an archaeosome adjuvant; for

20 example, mannotriose- or mannotetraose archaeal lipid may target the macrophage mannose receptor (Fig. 3) combined with another glyco synthetic lipid with different targeting specificity, or archaetidylserine to target the phosphatidylserine receptor on dendritic cells.

25 Linkage α or β to the lipid

Gentiobiose was linked in α and β configurations to the archaeol moiety (Fig. 6) to determine the influence of this aspect on adjuvant activity. Elispot assays revealed that both linkage configurations were active to adjuvant a CD8 ⁺ T cell response (Fig. 8). Anti-OVA antibody titers in sera of mice (4/group) at 6 weeks 30 post first injection were also comparable (12,216 ± 8078 for β-linked and 10,126 ±

4310 for α-linked).

16

Nature of the anionic lipid

An anionic lipid moiety improved the formulation and stability of synthetic- archaeol archaeosomes, so this aspect of the invention was explored in relation to 5 adjuvant activity. Various anionic lipids (Fig. 9) formed stable archaeosomes that retained entrapped antigen when combined with synthetic gentiotriose-archaeol and cholesterol (Table 11). The degree of adjuvant activity of the various synthetic archaeosomes was influenced considerably by the nature of the anionic lipid (Fig. 10A). Best activity was noted for SQDG and archaetidylglycerols (AG and AGP-

10 CH ₃ ). The antigen-specific CD8 ⁺ T cell response of synthetic gentiotriose-archaeol archaeosomes containing DPPG was transient, in contrast to that with archaeal anionic lipids wherein activity tended to increase from 6 to 8 weeks. Incorporating DPPE as the anionic lipid resulted in an Elispot showing 27% of the activity found for DPPG at 6 weeks (data not shown). These results show the importance and

15 advantage to generate longer-term responses by use of synthetic archaeal lipids (to provide the anionic charge) versus other conventional lipids, with the exception of SQDG.

The nature of the anionic lipid that was incorporated into the synthetic archaeosomes also impacted on the antigen-specific antibody responses (Fig. 10B).

20 Fig. 1OA and B show that the composition of the synthetic lipids used to make the synthetic archaeosomes can be used to direct the immune response towards either MHC class I or MHC class II immune responses, or to achieve high responses for both. For example, an archaetidylinositol favoured the antibody CD4 ⁺ T cell response over the CD8 ⁺ T cell response, and archaetidylglycerols as the anionic

25 lipid resulted in high responses for both. DPPE at 35mol% resulted in relatively low CD8 T cell activity, but was equivalent in antibody response to archaeosomes containing archaetidylglycerol (data not shown).

Replacement of cholesterol by synthetic polar caldarchaeols

30 For stability and adjuvant activity synthetic archaeosomes preferably include at least three elements; namely, an appropriate synthetic glyco-moiety, an anionic

17

moiety, and a stabilizer. One or more of these elements may be embodied in a single molecule. Stability of the synthetic glyco-archaeols may be achieved, for example, using cholesterol. Because cholesterol may itself oxidize and be undesirable in a product for human use, the possibility was explored to bypass 5 the need for cholesterol by use of membrane stabilizing bipolar caldarchaeols.

As a caldarchaeol bipolar lipid may be synthesized with all three elements; namely, one glyco head group, one anionic head group, and stability feature due to the caldarchaeol core lipid structure, archaeosomes were made using two such lipids (gentiobiose-caldarchaetidylinositol, and gentiobiose-

10 caldarchaetidylserine) (Fig. 11). These archaeosomes comprised a single lipid combined with antigen and formed in good yield from each lipid, contrary to expectation based on inefficient archaeosome formation from the caldarchaeol- rich TPLs of Thermoplasma and Sufolobus. The average diameters and antigen loadings are shown in Table 11. It may be appreciated that combination of

15 synthetic polar archaeol with synthetic polar caldarchaeol lipids may be used in the synthetic archaeosome formulation with antigen. Further, it may be appreciated that the 3 criteria listed above may be achieved to form stable archaeosomes with adjuvant activity by mixing two novel synthetic lipids, the first synthesized from caldarchaeol with 2 immuno-active head groups

20 (gentiobiose example in Fig. 11), and the second with two head groups bearing an anionic charge (Fig. 11). The advantage to coupling the same head group to both free hydroxyl moieties of a caldarchaeol is to simplify synthesis. Similarly, it can be recognized that an archaeol may be chemically coupled to either a glyco group or anionic group and combined with a caldarchaeol synthesized to provide

25 glyco or anionic groups to form an archaeosome adjuvant.

Archaeosomes composed of a single bipolar caldarchaeol served as strong CD8 ⁺ T cell adjuvants (Fig. 12). The data indicate that the anionic head group should preferably be phosphoinositol compared to phosphoserine and data in Fig. 10 further show a preference for phosphoglycerol versus either phosphoinositol

30 or phosphoserine. The anti OVA antibody response in blood titre was from

16,151-17,960 for the GIc ₂ -C-PI archaeosome compared to only 2,592-2,600 for

18

the GIc ₂ -C-PS archaeosome, again indicating importance for the proper selection of anionic head group and supporting the observation of higher antibody responses with archaeal lipids containing a phosphoinositol group.

5 Mucosal responses

Secretory slgA was found in faecal extracts of mice immunized with both caldarchaeol archaeosomes, although best responses occurred for phosphoinositol over phosphoserine as the anionic head group (Table 12). Anionic lipids giving activity in the gentiotriose-A compositions were DPPE, AG and AS. These 10 results revealed that a slgA mucosal response in mice to an antigen depended on the lipid composition of the synthetic archaeosome adjuvant.

Up-regulation of co-stimulatory molecules

Macrophages J774A.1 were incubated for 48h with 25 μg of archaeosomes

15 prior to staining and analysis by flow cytometry (Table 13). Compared to control cultures receiving either no stimulation or treatment with DPPG/chol (80/20, mol%) liposomes, expression of co-stimulatory molecules (CD80) on APCs were only up-regulated after exposure to Glc ₂ -archaeol/DPPG/Chol (45/35/20, mol%) and Maα _t -archaeol/DPPG/Chol (45/35/20, mol%) archaeosomes. Immuno-

20 activity was clearly ascribed to the synthetic glyco-archaeal lipid.

Protective vaccines against cancer

Table 14 shows that the vaccine protective effect achieved against a Bl 6 melanoma correlates with the magnitude of the immune response generated to

25 the antigen OVA entrapped in various synthetic archaeosome adjuvants.

Immune response to these synthetic archaeosome adjuvants was documented earlier in Fig. 7.

Protective vaccines against an infectious agent

30 Mice vaccinated with synthetic archaeosomes containing an appropriate antigen were protected from infection (Fig. 13). Best protection occurred with gentiotriose-

19

A archaeosomes, but maltotriose-A also caused striking protection. The lower CD8 ⁺ T cell response observed for this latter archaeosome previously appears to be in part due to the shorter period from first injection to assay in Fig. 1C. In this example (Fig. 13), the antigen-specific CD8 ⁺ T cell activity was measured in blood by tetramer assay and found to be about 30% as high in the maltotriose-A group compared to the gentiotriose-A group (data not shown). Isomaltose-A archaeosomes were less protective and produced lower CD8 ⁺ T cell responses (Fig. 7C).

20

MATERIALS AND METHODS Archaeal core lipids

Halobacterium salinarum ATCC 33170 was grown aerobically and the

5 biomass extracted with chloroform/methanol/water to obtain the total lipids. TPL was obtained as the acetone insoluble lipids (29). This source was chosen to provide saturated archaeol as the sole lipid product following hydrolysis. In one instance 3g of TPL was added to a 500-ml round bottom flask and the solvent removed. To the dried lipids 150 ml of 2.5% methanolic-HCL was added and

10 reflux continued at 64-65 ⁰ C for 4h while stirring magnetically. In another instance, equally good results were obtained by reflux for Ih followed by a second Ih reflux of the residue with fresh methanolic-HCl. Archaeol in the methanolic-HCl was partitioned into petroleum ether by mixing methanolic

HCl/water/petroleum ether (30-65 ⁰ C fraction) in the ratio 93ml/9.3ml/93ml. The

15 ether was evaporated to yield the archaeol as light yellow oil. In some cases a further purification step was conducted. Silica gel G (Merck) was poured into a column (bed 20cm x 1.8cm) in hexane. The archaeol fraction was loaded in hexane. Any neutral lipids present were eluted with hexane prior to recovering pure archaeol by elution with hexane/ethyl acetate = 9: 1 (v/v). The yield was

20 41- 58 % (wt basis) of starting TPL.

Thermoplasma acidophilum (ATTC culture 27658) was grown and lipids extracted as previously described (30). Methanolic-HCl hydrolysis and ether partitioning to recover caldarchaeol was done as described for archaeol, with yields of about 56% of the starting TPL (wt basis). 25

Synthetic glyco-archaeol synthesis

Glyco-archaeols were synthesized as illustrated (Fig. 2) for α-D-Glc-(l,4)-α-D- Glc-(l,4)-β-D-Glc-(l,l)-archaeol and β-D-Glc-(l,6)-β-D-Glc-archaeol. Sequential addition of mannose residues to archaeol is shown also. A series of disaccharides 30 attached to archaeol were made.

21

The oligomeric mannose structures were synthesized starting from known 2- O-acetyl protected monosaccharide donor which was prepared following Douglas et al. (6). Thus, archaeol was sequentially glycosylated to provide Man α-linked to archaeol, deacetylated and then either re-glycosylated or hydrogenated to 5 produce 1 to 4 additional αl,2-linked mannoses. To prepare glucose linked structures the commercially available maltotriose [α-Glc-(l,4)-α-Glc-(l,4)-β-Glc- OH] was first peracetylated. The resulting peracetate was selectively deacetylated at the anomeric position and converted into its known trichloroacetimidate derivative (17, 18) and the archaeol glycosylated followed by deacetylation. To

10 make a β-Glc-(l,6)-Glc building block the known 4,6-phenylboronated thioglycoside (5) was deboronated to the 4,6-diol. The diol was coupled with itself following the procedure developed by Huang et al. (11) and after acetylation and purification to give a disaccharide donor. This donor was used to directly glycosylate archaeol or to glycosylate the 4,6-diol of β-Glc-archaeol

15 prepared by glycosylation followed by deboronation with the original GIc phenylboronate donor. Removal of the acetyl and benzoyl protecting groups yielded GIc ₂ - and Glc ₃ -oligomers of β-Glc-(l,6)-Glc attached β to archaeol. The disaccharide archaeol derivatives derived from maltose, iso-maltose, lactose, cellobiose and meliobiose were made by preparing the known 1-phenylthio

20 derivatives (1, 20, 31) and glycosylating the archaeol using NIS/AgOTf or

NIS/BF ₃ /TFE ₂ conditions followed by removal of the acetyl protecting groups. The α-linked isomer of gentiobiose α-D-Glc-(l,6)-β-D-Glc-archaeol was made by coupling a glucose donor with a cleavable silyl protecting group at 0-6 under conditions optimized to form the α-anomer. The resulting monomer was

25 desilylated and glycosylated under standard conditions to give the protected disaccharide that was deprotected under standard conditions. Gentiobiose β- linked to both hydroxyls of caldarchaeol was made using the GIc ₂ donor. All compounds were characterized by ¹ H and ¹³ C NMR including ¹ H- ¹ H COSY and ¹³ C- ¹ H COSY correlation experiments. As well ID and 2D TOCSY experiments

30 and ID or 2D NOESY or ROESY experiments were used to confirm connectivities and make assignments as necessary. Also, positive ion MALDI

22

MS of lipid containing species gave the expected ions typically (M+Na) ⁺ and often (M+K) ⁺ as well.

Procedure A

5 (2R)-2, 3-Bis[(3R, 7R, 1 IR)S, 7, 11,15-tetramethylhexadecyloxyJpropan-l-yl 2-0- acetyl-3,4, 6-tri-O-benzyl-a-D-mannopyranoside (Manj-A)

To a mixture of (2R)-2,3-Bis[(3R,7R,l lR)-3,7,l l,15- tetramethylhexadecyloxy]propan-l-ol (archaeol) (720mg; l.lmmol), 2-O-acetyl- 3,4,6-tri-O-benzyl-α-D-mannopyranosyl trichloroacetimidate (1.4g; 2.2mmol)

10 and molecular sieves 4A° (3g) was added CH ₂ Cl ₂ (15ml). After stirring at R.T. under an argon atmosphere for Ih, triethylsilyltrifluoromethanesulfonate (25 μL; O.l lmmol) was added and the stirring continued for 40 min. The reaction was quenched with diisopropylethylamine (lOOμL). The whole reaction was adsorbed on silica gel and then purified by silica gel chromatography eluting

15 with hexanes:ethyl acetate 9:1 to yield pure product as a viscous oil (0.58g; 47%) plus some mixed fractions. Procedure B

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 3,4,6- tri-O-benzyl-a-D-mannopyranoside (Mani-B)

20 Mani-A (0.58g; 0.51mmol) was dissolved in a mixture of dry methanol

(1OmL) and CH ₂ Cl ₂ (2mL). Then IM NaOCH ₃ (0.5mL) was added and the stirring continued for 4h. The reaction mixture was diluted with CH ₂ Cl ₂ (15OmL) and washed 2x with NH ₄ Cl _aq followed by saturated NaCl _aq . After drying with Na ₂ SO ₄ , filtration and evaporation the residue was purified by

25 column chromatography on silica gel eluting with hexanes:ethyl acetate 5:1 to yield pure compound as a viscous oil (520mg; 93%).

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 2-O- (2-O-acetyl-3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O-benzyl-a-D- 30 mannopyranoside (Mari ₂ -A)

23

Man ₂ -A was prepared from Man _! -B using procedure A and purified by silica gel eluting with hexanes:ethyl acetate 9:1 followed by hexanes:ethyl acetate 85:15.

5 (2R)-2, 3-Bis[(3R, 7R, 1 IR)S, 7, 11,15-tetramethylhexadecyloxyJpropan-l-yl 2-0-

(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O-benzyl-a-D- mannopyranoside (Mari ₂ -B)

Man ₂ -B was prepared from Man ₂ -A using procedure B and purified by silica gel eluting with hexanes: ethyl acetate 85:15. 10

(2R)-2, 3-Bis[(3R, 7R, 1 lR)-3, 7, 11, 15-tetramethylhexadecyloxy]propan-l-yl 2-0- (2-O-acetyl-3, 4, 6-tή-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a- D-mannopyranosyl)-3, 4, 6-tri-O-benzyl a-D-mannopyranoside (Man^-A)

Man ₃ -A was prepared from Man ₂ -B using procedure A and purified by silica 15 gel eluting with hexanes : ethyl acetate 85 : 15.

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11 ,15-tetramethylhexadecyloxyJpropan-l-yl 2-0- (3, 4, 6-tri-0-benzyl-a-D-mannopyranosyl)-2-0-(3, 4, 6-tri-O-benzyl-a-D- mannopyranosyl)-3, 4, 6-tri-O-benzyl-a-D-mannopyranoside (Mans-B) 20 Man ₃ -B was prepared from Man ₃ -A using procedure B and purified by silica gel eluting with hexanes:ethyl acetate 85:15.

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 2-0- (2-O-acetyl-3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a- 25 D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O- benzyl a-D-mannopyranoside (Man4-A)

Mari4-A was prepared from Man ₃ -B using procedure A and purified by silica gel eluting with hexanes: ethyl acetate 85:15.

30 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl -2-0-

3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl-2-O-(3, 4, 6-tri-O-benzyl-a-D-

24

mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O- benzyl-a-D-mannopyranoside (Mari4-B)

MaH ₄ -B was prepared from MaIi ₄ -A using procedure B and purified by silica gel eluting with hexanes:ethyl acetate 85:15. 5

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 2-0- (2-O-acetyl-3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a- D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O- benzyl a-D-mannopyranoside (Mari4-A)

10 Maru-A was prepared from Man ₃ -B using procedure A and purified by silica gel eluting with hexanes:ethyl acetate 85:15.

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-J-yl -2-0- 3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl-2-O-(3, 4, 6-tri-O-benzyl-a-D- 15 mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O- benzyl-a-D-mannopyranoside (Mari ₄ -B)

MaTi ₄ -B was prepared from MaIi ₄ -A using procedure B and purified by silica gel eluting with hexanes:ethyl acetate 85:15.

20 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 2-0-

(2-O-acetyl-3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a- D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6- tri-O-benzyl-a-D-mannopyranosyl)-3, 4, 6-tri-O-benzyl a-D-mannopyranoside (ManyA)

25 Man ₅ -A was prepared from MaIi ₄ -B using procedure A and purified by silica gel eluting with hexanes:ethyl acetate 85:15.

(2R)-2, 3-Bis[(3R, 7 R, 1 lR)-3, 7, 1 l,15-tetramethylhexadecyloxy]propan-l-yl -2-0- (3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D- 30 mannopyranosyl)-2-O-(3, 4, 6-tri-O-benzyl-a-D-mannopyranosyl)-2-O-(3, 4, 6-tri-

25

O-benzyl-a-D-mannopyranosyl)-3,4,6-tri-O-benzyl-a-D-mannopyr anoside (Man ₅ -B)

Man ₅ -B was prepared from Man ₅ -A using procedure B and purified by silica gel eluting with hexanes:ethyl acetate 75:25. 5

Procedure C

(2R)-2, 3-Bis[(3R, 7R, 1 IR)S, 7, 11, 15-tetramethylhexadecyloxy]propan-l-yl a-D- mannopyranoside (Manj-C)

Mani-B (lOOmg; 0.051mmol) was dissolved in ethyl acetate (1OmL) and after

10 purging with argon Pd(OH) ₂ /C (Pearlman's catalyst) (150mg) was added and the mixture hydrogenated using a Parr apparatus at 50 p.s.i. of H ₂ with shaking for

64h. The catalyst was removed by filtration through a bed of celite and was well washed with ethyl acetate and methanol. The combined filtrates were evaporated and then purified by silica gel chromatography eluting with ethyl

15 actetate:methanol:water 7:1 :1 to yield a waxy solid (62mg, 83%).

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 2-0- (a-D-mannopyranosyl) a-D-mannopyranoside (Mari2-C)

Man ₂ -C was prepared from Man ₂ -B using procedure C. 20

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 2-O-

(a-D-mannopyranosyl)-2-O-(a-D-mannopyranosyl)-a-D-mannopy ranoside

(MaH ₃ -C)

Man ₃ -C was prepared from Man ₃ -B using procedure C. 25

(2R)-2, 3-Bis[(3R, 7R, 1 lR)-3, 7, 11, 15-tetramethylhexadecyloxyJpropan-l-yl 2-0- (a-D-mannopyranosyl)-2-O-(a-D-mannopyranosyl)-2-O-(a-D-manno pyranosyl)- a-D-mannopyranoside (Man4-C)

Maαj-C was prepared from MaIi ₄ -B using procedure C. 30

26

(2R)-2,3-Bis[(3R, 7RJlR)S ₁ 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 2-0- (a-D-mannopyranosyl)-2-O-(a-D-mannopyranosyl)-2-O-(a-D-manno pyranosyl)- a-D-mannopyranoside (Man^-C)

Man ₅ -C was prepared from Mans-B using procedure C. 5

Ethyl 6-O-(4, 6-di-O-acetyl-2, 3-di-O-benzoyl-β-D-glucopyranosyl)-4-O-acetyl- 2,3-di-O-benzoyl-β-D-gluco-l-thiopyranoside (Glc ₂ donor)

To ethyl 2,3-di-O-benzoyl-β-D-gluco-l-thiopyranoside (3. Og; 6.93mmol) [prepared by deboronylation with IRA743 resin see GIc ₁ -A' ¹ H NMR CDCl ₃ 7.93

10 d 4 (J _H , _H 6.8) Bz ₀ , 7.49 m 2 Bz^, 7.35 m 4 Bz _m , 5.42 m (2, H-2, H-3), 4.73 d (1, ha ⁹ - ⁸ _> H" ¹ )' ⁴ - ^{00 dd} (1. ^J 5,6 3.4, J ₆₆ . 12.1, H-6), 3.96 brt (1, J ₃ , ₄ =J ₄ , ₅ 9.4, H-4), 3.87 dd (1, J ₅₎₆ . 4.9, H-6 ¹ ), 3.60 ddd (1, H-5), 3.17 brs (OH), 2.73 m (2, SCH ₂ ), 1.25 t (3, J _HH 7.2, SCH ₂ CH ₃ )] and molecular sieves 4A° (3g) was added CH ₂ Cl ₂ (3OmL) and the mixture cooled in a dry ice acetone bath (bath T about -78 ⁰ C)

15 under an atmosphere of argon. To this mixture was added p- toluenesulfenylchloride (667μL; 4.6mmol) and silver triflluoromethanesulfonate (1.188g; 4.6mmol) and the mixture stirred at this T for Ih. The mixture was then transferred to a dry ice acetonitrile bath (bath T about -45 ⁰ C) and the mixture stirred for 1.75h. The reaction was quenched with diisopropylethylamine (ImL),

20 filtered, followed by rinsing with CH ₂ Cl ₂ and CH ₂ Cl ₂ :ethyl acetate 50:50, followed by concentration of the combined filtrates. The residue was purified by chromatography on silica gel eluting with ethyl acetate :hexanes: CH ₂ Cl ₂ 6:3:1 to yield a viscous oil (1.55g). The product is the lowest R _f spot of the two most prevalent products. This oil was dissolved in pyridine (15mL) and cooled in an

25 ice bath under an atmosphere of argon, and acetic anhydride (7.5mL) was added.

The mixture was left to stir and warm to R.T. overnight. After evaporation the residue was purified by medium pressure liquid chromatography on silica gel eluting with hexanes:ethyl acetate:CH ₂ Cl ₂ 6:3:1 to yield an amorphous white solid (1.2g; 37%); ¹ H NMR CDCl ₃ 7.93 m (8 Bz ₀ , 7.49), m (4 Bz _p ), 7.35 m (8

30 Bz _w ), 5.65 t (1, J ₃ , ₄ 9.4, H-3 ¹ ), 5.59 t (1, J ₃ , ₄ 9.5, H-3 ^π ), 5.44 dd (1, J _2;3 10.0, H-

2"), 5.37 t (1, J _4i5 9.7, H-4"), 5.33 dd (1, J ₂₎₃ 10.0, H-2 ¹ ), 5.08 t (1, J _4,5 9.7, H-4 ¹ ),

27

4.89 d (1, Ji ₎₂ 7.9, H-I ¹¹ ), 4.57 d (1, Ji _,2 10.0, H-I ¹ ), 4.37 dd (1, J _5>6 5.0, J ₆₆ . 12.3, H-6 ¹¹ ), 4.21 dd (1, J ₅₎₆ . 2.1, H-6 ¹ "), 3.95 brd (1, H-6 ¹ ), 3.88 ddd (1, H-5 ¹¹ ), 3.82 m (1, H-5 ¹ ), 3.75 dd (1, J _5>6 . 7.3, J _6>6 . 11.1, H-6' ¹ ), 2.51 m (2, SCH ₂ ), 2.14, 1.94, 1.92 3 x s (3, Ac CH ₃ ), 1.05 t (3, J _HH 7.2, SCH ₂ CH ₃ ); ¹³ C NMR CDCl ₃ 170.7, 169.7, 5 169.4 (3 x Ac C-O), 165.8, 165.6, 165.1 (2) (4 x Bz C=O), 133.4, 133.35,

133.28, 133.2 (4 x Bz _p ), 129.8 (Bz _m ), 129.2, 129.1, 128.80, 128.76 (4 x Bz^), 128.4 (Bz ₀ ), 101.0 (C-I ¹¹ ), 83.3 (C-I ¹ ), 77.7 (C-5 ¹ ), 74.1 (C-3 ¹ ), 73.0 (C-3 ¹¹ ), 72.2 (C-5"), 71.6 (C-2 ^π ), 70.4 (C-4 ¹ ), 69.1 (C-4 ^π ), 68.41 (C-6 ¹ ), 68.36 (C-4 ¹ ), 61.9 (C- 6"), 24.0 (SCH ₂ ), 20.8, 20.54, 20.52 (3, Ac CH ₃ ), 14.7 (SCH ₂ CH ₃ ); MALDI MS 10 (M+Na) ⁺ 951.4, 967.3 (M+K) ⁺ .

(2R)-2, 3-Bis[(3R, 7R, 1 IR)S, 7, 1 l,15-tetramethylhexadecyloxy]propan-l-yl 6-O- (4, 6-di-O-acetyl-2, 3-di-O-benzoyl-β-D-glucopyranosyl)-4-O-acetyl-2, 3-di-O- benzoyl-β-D-glucopyranoside (Glc ₂ -A)

15 To GIc ₂ donor (236mg, 0.255mmol) and molecular sieves 4A° was added

CH ₂ Cl ₂ (2mL) and the mixture cooled in a dry ice methanol bath (bath T about - 6O ⁰ C) under an atmosphere of argon. To this mixture was added p- toluenesulfenylchloride (37μL; 0.255mmol) and silver trifluoromethanesulfonate (65mg; 0.255mmol) and the mixture stirred at this T for Ih. To this mixture was

20 added archaeol (l l lmg; 0.17mmol) in CH ₂ Cl ₂ (1.5mL). After lOmin the bath was removed and the temperature allowed to rise. After 40min the reaction was quenched with diisopropylethylamine (0.ImL), filtered, followed by rinsing with CH ₂ Cl ₂ and CH ₂ Cl ₂ :ethyl acetate 50:50, followed by concentration of the combined filtrates. The residue was purified by chromatography on silica gel

25 eluting with hexanes:ethyl acetate:CH ₂ Cl ₂ 7:2:1 to yield pure GIc ₂ -A as a viscous oil (161mg; 64%).

Modified Procedure B

30 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-0-

(β-D-glucopyranosyl)-β-D-glucopyranoside (Glc ₂ -A)

28

GIc ₂ -A (150mg; 0.099mmol) was dissolved in dry methanol (1OmL) and CH ₂ Cl ₂ (5mL). To this solution was added IM NaOCH ₃ (0.75mL) and stirring continued for 16h at R.T. The mixture was cooled in an ice bath and neutralized with RexynH ⁺ that had been washed with water and methanol. The solids were 5 removed by filtration and washed with methanol. The combined filtrates were evaporated to yield pure GIc ₂ -B (95 mg; 98%) (23).

(2R)-2,3-Bis[(3R, 7R ₁ IlR)S, 7,11, 15-tetramethylhexadecyloxy]propan-l-yl 2,3- di-O-benzoyl-β-D-glucopyranoside (Glci-A ')

10 Ethyl 2,3-di-O-benzoyl-4,6-phenylboranyl-β-D-gluco-l-thiopyranosi de (5)

(125mg; 0.24mmol) was activated and reacted with archaeal (105mg; O.lόmmol) as described above for GIc ₂ A. The crude product was treated with IRA-743 resin (about 1Og) that had been soaked and rinsed extensively with acetonitrile in acetonitrile (about 25mL) by shaking overnight. The resin was removed by

15 filtration, rinsed with CH ₂ Cl ₂ and acetonitrile, and the combined filtrates evaporated to dryness. The residue was purified by silica gel chromatography eluting with hexanes:ethyl acetate: CH ₂ Cl ₂ 7:2:1 to yield pure Glci-A' as a viscous oil (67mg; 41 %).

20 (2R)-2,3-Bis[(3R, 7R, 1 lR)-3, 7, 11 ,15-tetramethylhexadecyloxy]propan-l-yl 6-0-

(4, 6-di-O-acetyl-2, 3-di-O-benzoyl-β-D-glucopyranosyl)-6O-(4-O-acetyl-2, 3-di- O-benzoyl-β-D-glucopyranosyl)-2,3-di-O-benzoyl-β-D-glucopy ranoside (Glc$-A) GIc ₂ donor (92mg, 0.099mmol) was activated as for GIc ₂ -A above and reacted with Glci-A' (67mg; 0.066mmol) for 2h in a dry ice acetonitrile bath (bath T

25 about -45°C). The residue was purified by silica gel chromatography eluting with hexanes:ethyl acetate:CH ₂ Cl ₂ 7:2:1 to yield pure GIc ₃ -A as a waxy solid (90mg; 72%).

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-0- 30 (β-D-glucopyranosyl)-60-(β-D-glucopyranosyl)-β-D-glucopyr anoside (Glcj-B)

29

The acyl groups were removed from GIc ₃ -A (80mg; 0.042mmol) using modified procedure B. TLC analysis of the product indicated an unidentified impurity so the product was purified by preparative TLC eluting with

CHCl ₃ :CH ₃ OH:CH ₃ COOH:H ₂ O 85:22.5:10:4 to yield pure GIc ₃ -B as a waxy

5 solid (40mg; 83%).

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11 ,15-tetramethylhexadecyloxyJpropan-l-yl 4-0- (2, 3, 4, 6-tetra-O-a-D-glucopyranosyl)-4O-(2, 3, 6-tri-O-acetyl-a-D- glucopyranosyl)-2, 3, 6-tri-O-acetyl-β-D-glucopyranoside (GlcsM-A)

10 To known (17, 18) 4-O-(2,3,4,6-tetra-O-α-D-glucopyranosyl)-4-O-(2,3,6-tri-

O-acetyl-α-D-glucopyranosyl)-2,3,6-tri-O-acetyl-β-D-glucop yranosyl trichloroacetimidate (280mg; 0.29mmol), 4 molecular sieves (300mg) and archaeol (78mg; 0.12mmol) was added CH ₂ Cl ₂ (3mL) and the mixture stirred for Ih at R.T. under an atmosphere of argon. To this was added

15 triethylsilyltrifluoromethanesulfonate (3μL; 0.013mmol) and the mixture stirred for 40min when TLC in hexanes:ethyl acetate 1 :1 R _f =O.5 indicated the reaction was complete. The reaction was quenched with diisopropylethylamine (lOμL), filtered with rinsing with CH ₂ Cl ₂ . The combined filtrates were concentrated and the residue purified by silica gel chromatography eluting with hexanes: ethyl

20 acetate 2:1 to yield pure GIc ₂ M-A as a waxy solid (99mg; 53 %).

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 4-O- (a-D-glucopyranosyl)-4O-(a-D-glucopyranosyl)-β-D-glucopyran oside (Glc$M- B)

25 GIc ₃ M-A was deacetylated following modified procedure B and purifed by silica gel chromatography eluting with CHC1 ₃ :CH ₃ OH:H ₂ O 10:3:0.3 to yield pure GIc ₃ M-B.

EthyU, 3, 4-tri-O-benzyl-6-O-(t-butyldiphenylsilyl)- a/β-D-gluco-1 30 -thiopyranoside (1)

30

Ethyl 2,3,4-tri-O-benzyl-α /β-D-gluco-l-thiopyranoside (7) (1.13 g, 2.21 mmol) was dissolved in CH ₂ Cl ₂ (15 niL) and cooled in an ice bath under an argon atmosphere. To this was added imidazole (468 mg, 4.5 eq.) followed by t- butyldiphenylchlorosilane (0.894 mL, 1.5 eq) and the mixture allowed to stir and 5 to warm to room temperature over 16 h. The reaction mixture was concentrated and the residue purified by silica gel chromatography eluting with CH ₂ Cl ₂ followed by 1% t-butylmethylether in CH ₂ Cl ₂ to yield 1 as a α/β (1 :0.46) mixture: ¹ H NMR CDCl ₃ 7.76-7.65 m (ArH), 7.41-7.24 m (ArH), 7.15 m (ArH), 5.45 d (Ji, ₂ 5.3, H-l ^α ), 4.94 - 4.58 m (BnCH ₂ ), 4.48 d (J _u 9.7, H-l ^β ), 4.13 m (H-

10 5 ^α ), 3.92-3.80 m (H-2 ^α , H-3 ^α , H-6 ^α , H-6' ^α , H-6 ^β , H-6' ^β ), 3.75 t (J _4>5 9.1, H-4 ^β ),

3.67 t (J ₃ , ₄ 8.7, H-3 ^β ), 3.57 t (J _3>4 9.8, J ₄₎₅ 9.9, H-4 ^α ), 3.47 dd (J ₂₃ 9.4, H-2 ^β ), 3.37 m (H-5 ^β ), 2.74 m (CH ₂ S"), 2.52 (CH ₂ S ^β ), 1.32 t (J 7.6, CH ₃ CH ₂ S"), 1.25 (J 7.6, CH ₃ CH ₂ S ^β ), 1.04, 1.02 2xs (CH ₃ /-butyl); ¹³ C NMR CDCl ₃ 138.7, 138.3, 138.0 (3 x Bn,/), 138.4, 138.14, 138.10 (3 x Bn,/),135.9, 135.8, 135.6 (Ph^), 133.6, 133.2

15 (Ph/), 134.8, 133.1 (Ph,/),129.5 (m ArC), 128.5-127.5 (ArC), 86.7 (C-3 ^β ), 84.4

(C-l ^β ), 82.7 (C-3 ^α ), 82.2 (C-l ^α ), 81.9 (C-2 ^β ), 80.0 (C-2 ^α ), 79.9 (C-5 ^β ), 77.72 (C- 4 ^β ), 77.69 (C-4 ^α ), 75.91, 75.86, 75.5, 75.15, 75.1, 72.3 (BnCH ₂ ), 71.9 (C-5 ^α ), 63.0 (C-6 ^α ), 62.8 (C-6 ^β ), 34.7, 31.6 (CCH ₃ f-butyl), 26.8 (CH ₃ f-butyl), 24.3 (CH ₂ S ^β ), 23.2 (CH ₂ S"), 15.1 (CH ₂ CH ₃ S ^β ), 14.6 (CH ₂ CH ₃ S"); MALDI MS

20 (M+H) ⁺ 733.5, (M+Na) ⁺ 755.5.

(2R)-2, 3-Bis [(3R, 7 R, 1 lR)-3, 7, 11, 15-tetramethylhexadecyloxy]propan-l-yl 2, 3, 4- tri-O-benzyl-6-O-(\-butyldiphenylsilyl)-a-D-glucopyranoside (2) and (2R)-2,3- Bis [(3R, 7R, 1 lR)-3, 7,11,15-tetramethylhexadecyloxy]propan-l-yl 2, 3, 4-tri-O-

25 benzyl- a-D-glucopyranoside (3)

Archaeol (140 mg; 0.21 mmol) and (1, 236 mg, 1.5 eq.) were dissolved in CH ₂ Cl ₂ (2.6 mL) and cooled in an ice bath under an argon atmosphere along with powdered 4 A molecular sieves (300 mg). To this was added N-selenophenyl phthalimide (116 mg, 1.8 eq.) followed by trifluoromethanesulfonic acid (32 μL,

30 1.6 eq.) and the stirring continued for 3 h. The reaction was quenched with

31

diisopropylethylamine (excess) and the mixture filtered and rinsed with CH ₂ Cl ₂ . The organic layers were concentrated and the residue purified by silica gel chromatography eluting with CH ₂ Cl ₂ : cyc/oC ₆ Hi ₂ : t-butylmethylether 49:49:2 to yield a mixture of 2 and the α-anomer of 1. This mixture was dissolved in 5 tetrahydrofuran (4 mL) and a IM solution of tetrabutylammonium fluoride in

THF (400 μL) was added and the resulting mixture heated at 50 ⁰ C for 16 h under an atmosphere of argon. The reaction mixture was concentrated and the residue purified by silica gel chromatography eluting with CH ₂ Cl ₂ : cyc/oC ₆ Hi ₂ : t- butylmethylether 48:48:4 to yield 3 (108 mg; 48%). A small amount of 2 was

10 repurified by preparative TLC CH ₂ Cl ₂ : cyc/oC ₆ H ₁₂ : r-butylmethylether 48:48:4 for an analytical sample. ¹ H NMR CDCl ₃ 7.69 brd (2 Ph ₀ ), 7.62 brd (2 Ph ₀ ), 7.41- 7.24 m (16 ArH), 7.16 m (4 Ph _m ), 4.97 - 4.62 m (6, BnCH ₂ ), 4.88 d (1, J _1;2 4.8, H-I), 4.00 brt (1, H-3), 3.88 m (2, H-6, H-6'), 3.76 m (1, H-5), 3.67 - 3.49 m (9, H-2, H-4, CH ₂ O, CH ₂ O, CHO, OCH ₂ ), 3.43 brt (2, OCH ₂ ) 1.62 - 1.48 m (6, CH,

15 CH ₂ ), 1.38 - 1.22 m (42, CH, CH ₂ ), 0.88 - 0.78 m (30, CH ₃ ); MALDI MS

(M+Na) ⁺ 1346.0, (M+K) ⁺ 1361.9. 3 ¹ H NMR CDCl ₃ 7.40 -7.24 m (15 ArH), 4.97 d (1, J _H ,H 11-8, BnCH ₂ ), 4.89 d (1, J _H , _H H-I, BnCH ₂ ), 4.81 d (1, J _HjH 11.8, BnCH ₂ ), 4.80 d (1, J _u 3.8, H-I), 4.71 q (2, BnCH ₂ ), 4.65 d (1, J _H,H H-I, BnCH ₂ ), 4.00 brt (1, J _3>4 9.1, H-3), 3.81 dd (1, J ₅ , ₆ 2.3, J ₆ , ₆ - 12.3, H-6), 3.76 m (1,

20 H-5), 3.69 m (1, H-6'), 3.67 - 3.45 m (9, CH ₂ O, CH ₂ O, CHO, OCH ₂ , H-4, H-2),

1.62 - 1.48 m (6, CH, CH ₂ ), 1.38 - 1.22 m (42, CH, CH ₂ ), 0.88 - 0.78 m (30, CH ₃ ); ¹³ C NMR CDCl ₃ 138.8, 138.34, 138.31 (3 x Bn _ψ ), 128.4-127.6 (ArC), 97.2 (C-I), 81.9 (C-3), 80.2 (C-2), 77.8 (CHO), 77.2 (C-4), 76.7, 75.6, 72.9 (3 x BnCH ₂ ), 70.8 (C-5, OCH ₂ ), 70.6 (OCH ₂ ), 69.0 (CH ₂ O), 67.9 (CH ₂ O), 61.9 (C-6),

25 39.4, 37.6, 37.52, 37.47, 37.44, 37.29, 37.2, 36.7 (CH ₂ ), 32.8, 30.0, 29.8, 28.0

(CH), 24.8, 24.5, 24.4 (CH ₂ ), 19.75, 19.68 (CH ₃ ); MALDI MS (M+Na) ⁺ 1084.87.

(2R)-2,3-Bis[(3R, 7RJlR)-S, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-0- 30 [2, 3, 4, 6-tetra-O-benzoyl-β-D-glucopyranosyl]-2, 3, 4-tή-O-benzyl- a-D- glucopyranoside (4)

32

Alcohol (3, 51 mg; 0.047 mmol) and 2,3,4,6-tetra-O-benzoyl-α-D- glucopyranosyl trichloroacetimidate (52 mg; 1.5 eq) were dissolved in CH ₂ Cl ₂ (2.5 mL) along with powdered 4 A molecular sieves (100 mg) and the mixture cooled in an ice bath under an atmosphere of argon. To this mixture BF ₃ etherate 5 (6 μL, 1 eq.) was added and the stirring continued for 2h. The reaction was quenched with excess diisopropylethylamine, filtered and concentrated. The residue was purified by silica gel chromatography eluting with 2.5 % acetone in toluene to yield 4 (17 mg; 22 %). ¹ H NMR CDCl ₃ 7.92 brd (2, J _H , _H 7.2, Bz ₀ ), 7.89 m (4, Bz ₀ ), 7.81 brd (2, J _H>H 7.2, Bz ₀ ), 7.53 -7.16 m (25 ArH), 7.00 m (2,

10 ArH), 5.87 brt (1, J _3>4 9.7, H-3 ¹¹ ), 5.66 brt (1, J ₄ , ₅ 10.2, H-4 ¹¹ ), 5.60 brt (1, J _2j3 9.7,

H-2 ¹¹ ), 4.86 d (1, J _H , _H 10.8, BnCH ₂ ), 4.78 d (1, J _u 3.5, H-I ¹ ), 4.76 d (1, J _u 7.9, H-I ¹¹ ), 4.63 m (4, BnCH ₂ , H-6°), 4.51 dd (1, J ₅₎₆ > 5.0, J ₆₆ - 12.0, H-6' ^π ), 4.42 d (1, J _H , _H 11-3, BnCH ₂ ), 4.25 d (1, J _H;H 11.3, BnCH ₂ ), 4.17 brd (1, H-6 ¹ ), 4.06 m (1, H-5°), 3.88 brt (1, J ₂ , ₃ 9.4, J _3>4 9.0, H-3 ¹ ), 3.79 m (2, H-5 ¹ , H6' ¹ ), 3.57 m (5,

15 CH ₂ O, CHO, OCH ₂ ), 3.44 m (6, CH ₂ O, OCH ₂ , H-4 ¹ , H-2 ¹ ), 1.61 - 1.48 m (6,

CH, CH ₂ ), 1.38 - 1.01 m (42, CH, CH ₂ ), 0.88 - 0.81 m (30, CH ₃ ); ¹³ C NMR CDCl ₃ 166.1, 165.8, 165.1, 164.9 (4 x C=O Bz), 138.9, 138.5, 138.4 (3 x Bn _(> ), 133.4, 133.2, 133.1, 133.0 (4 x Bz _p ), 129.8 - 127.2 (ArC), 101.3 (C-I ¹¹ ), 97.1 (C- I ¹ ), 81.7 (C-3 ¹ ), 79.9 (C-2 ¹ ), 77.8 (CHO), 77.2 (C-4 ¹ ), 75.3, 74.4 (2 x BnCH ₂ ),

20 72.9 (C-3 ^π ), 72.7 (BnCH ₂ ), 72.2 (C-5 ¹¹ ), 71.8 (C-2 ¹¹ ), 70.7 (CH ₂ O), 70.0 (OCH ₂ ),

69.8 (C-4 ¹¹ ), 69.3 (C-5 ¹ ), 69.0 (OCH ₂ ), 68.1 (C-6 ¹ ), 67.7 (CH ₂ O), 63.3 (C-6 ¹¹ ), 39.4, 37.6, 37.45, 37.42, 37.3, 37.2, 36.7 (CH ₂ ), 32.8, 30.0, 29.8, 28.0 (CH), 24.8, 24.5, 24.4 (CH ₂ ), 22.7, 22.6, 19.73, 19.71, 19.67 (CH ₃ ); MALDI MS (M+Na) ⁺ 1686.0.

25

(2R)-2,3-Bis[(3R, 7R ₁ IlR)S, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-O- β-D-glucopyranosyl- a-D-glucopyranoside (Gentα)

Disaccharide (4, 17 mg; 0.010 mmol) was first treated using modified procedure B to yield (2R)-2,3-Bis[(3R, 7R,llR)-3, 7, 11,15-

30 tetramethylhexadecyloxyjpropan-l-yl 6-O-β-D-glucopyranosyl-2, 3, 4-tri-O- benzyl- a-D-glucopyranoside (5) partial ¹ H NMR CDCl ₃ 4.70 d (1, J _u 3.5, H-I ¹ )

33

and 4.28 d (1, Ji _;2 7.6, H-I ¹¹ ) which was then treated with procedure C to yield Gentα (8 mg; 82%).

Procedure D-I 5 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-0-

[2,3,4,6-tetra-O-acetyl-a-D-glucopyranosyl]-2,3,4-tri-O-a cetyl-β-D- glucopyranoside (iMalt-A)

Archaeol (85 mg; 0.13 mmol) and phenyl 6-O-[2,3,4,6-tetra-O-acetyl-α-D- glucopyranosyl]-2,3,4-tri-O-acetyl-β-D-gluco-l-thiopyranosi de (153 mg, 1.5 10 eq.) (1, 31) were dissolved in CH ₂ Cl ₂ (3.0 mL) and cooled in an ice bath under an argon atmosphere along with powdered 4 A molecular sieves (200 mg). To this mixture was added N-iodosucinimide (74 mg, 2.5 eq.) followed by silver trifluoromethanesulfonate (33 mg, 1.0 eq.) and the stirring continued for 0.5 h.

The reaction was quenched with diisopropylethylamine (excess) and the mixture 15 filtered and rinsed with CH ₂ Cl ₂ . The organic layers were concentrated and the residue purified by silica gel chromatography eluting with hexanes: ethyl acetate: CH ₂ Cl ₂ 6:3:1 to yield the peracetylated disaccharide β-linked to archaeol, iMalt-A, (19 mg, 15%).

20 Procedure D-2

Archaeol (51 mg; 0.078 mmol) and (SPh donor, 79 mg, 1.5 eq.) were dissolved in CH ₂ Cl ₂ (1.5 mL) and cooled in an ice bath under an argon atmosphere. To this solution was added N-iodosucinimide (74 mg, 2.5 eq.) followed by a 0.25 M solution Of BF ₃ etherate and trifluoroethanol (20) (128 μL,

25 0.5 eq.; made from trifluoroethanol (386 μL) in CH ₂ Cl ₂ (5.0 mL) which was cooled in a dry ice acetonitrile bath to which was added BF ₃ etherate (316 μL) followed by treatment under vacuum about 5 torr for 20 minutes) and the stirring continued for 1 h. The reaction was quenched with aqueous NaHCO ₃ followed by aqueous Na ₂ S ₂ O ₃ and the mixture filtered and rinsed with CH ₂ Cl ₂ . The

30 organic layers were concentrated and the residue purified by silica gel

34

chromatography eluting with hexanes:ethyl acetate:CH ₂ Cl ₂ 6.5:2.5:1 to yield iMalt (41 mg, 42%).

5 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11 ,15-tetramethylhexadecyloxyJpropan-l-yl 4-0-

[2, 3, 4, 6-tetra-O-acetyl- a-D-glucopyranosyl] -2, 3, 4-tri-O-acetyl-β-D- glucopyranoside (MaIt-A)

MaIt-A was prepared using either procedure D-I or D-2

10 (2R)-2, 3-Bis[(3R, 7R, 1 lR)-3, 7, 11 ,15-tetramethylhexadecyloxyJpropan-l-yl 4-0-

[2, 3, 4, 6-tetra-O-acetyl- β-D-glucopyranosyl] -2, 3, 4-tri-O-acetyl-β-D- glucopyranoside (Cello- A)

Cello- A was prepared using either procedure D-I or D-2

15 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 4-0-

[2, 3, 4, 6-tetra-O-acetyl- β-D-galactopyranosyl] -2, 3, 4-tri-O-acetyl-β-D- glucopyranoside (Lac-A)

Cello-A was prepared using either procedure D-I or D-2

20 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-0-

[2,3,4,6-tetra-O-acetyl-a-D-galactopyranosyl]-2,3,4-tri-O -acetyl-β-D- glucopyranoside (MeIo-A)

MeIo-A was prepared using either procedure D-I or D-2

25 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 6-0 [- a-D-glucopyranosyl] - β-D-glucopyranoside (iMalt-B) iMalt-B was prepared from iMalt- A using modified Procedure B.

(2R)-2,3-Bis[(3R, 7R,l 1R)-3, 7,11 ,15-tetramethylhexadecyloxyJpropan-l-yl 4-0- 30 a-D-glucopyranosyl-β-D-glucopyranoside (MaIt-A)

MaIt-B was prepared using from MaIt-A using modified Procedure B.

35

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 4-0- [2, 3, 4, ό-tetra-O-acetyll-β-D-glucopyranosyl-β-D-glucopyranoside (Cello- A)

Cello-B was prepared from Cello- A using modified procedure B. 5

(2R)-2,3-Bis[(3R, 7R,llR)-3, 7,11,15-tetramethylhexadecyloxyJpropan-l-yl 4-0- β-D-galactopyranosyl-β-D-glucopyranoside (Lac-B)

Lac-B was prepared from Lac- A using modified Procedure B.

10 (2R)-2,3-Bis[(3R, 7R,llR)-3, 7,ll,15-tetramethylhexadecyloxy]propan-l-yl 6-0- a-D-galactopyranosyl-β-D-glucopyranoside (MeIo-B)

MeIo-B was prepared from MeIo-A using modified Procedure B.

cold 6-0-β-D-glucopyranosyl-β-D-glucopyranoside (Gent2cald -A)

15 This was made using GIc ₂ donor and caldarchaeol using procedure D-2 with double the amount of all reagents except the lipid to account for the extra hydroxyl.

cald 6-O-β-D-glucopyranosyl-β-D-glucopyranoside (Gent2cald -A) 20 Gent2cald-B was prepared from Gent2cald-A using modified Procedure B.

Synthetic anionic archaeal lipid synthesis

Archaeol was obtained as before from Halobacterium polar lipids and combined with a phospho-L-serine head group protected as its benzyl ester and 25 carbobenzoxy carbamate, according to the procedure in (33) to form archaetidylserine (AS).

NMR and MS Tables

¹ H and ¹³ C NMR of 1-5, Mani _-5 -A, Man _1-5 -B, Glci ¹ , GIc ₂ -A, GIc ₃ -A, GIc ₃ M-

30 A, iMalt-A, MaIt-A, Lac-A, MeIo-A, Cello-A and Gent2Cald-A were obtained in

CDCl ₃ solution (referenced to residual CHCl ₃ at 7.26 ppm ¹ H and 77.0 ppm

36

central resonance ¹³ C) whereas those of Man _1-5 -C, GIc ₂ -B, GIc ₃ -B, GIc ₃ M-B, iMalt-B, MaIt-B, Lac-B, MeIo-B, Cello-B and Gentα were obtained in 1:1 (v:v) solutions of CD ₃ OD:CDC1 ₃ (referenced to residual CHD ₂ OD at 3.31 ppm ¹ H and 49.15 ppm central resonance ¹³ C). Gent2Cald-B spectra were obtained in 1 :4 5 ( ^v:v ) solutions of CD ₃ OD:CDC1 ₃ with the same reference as 1 :1. Chemical shifts are in ppm and coupling constants in Hz. NMR was performed on either a Varian 400 MHz or 200 MHz spectrometers. The NMR data are compiled in Tables 1-8. Table 9 contains the MALDI MS data.

10 Source of other lipids

The following lipids were purified in biologically pure form by thin-layer chromatography of total polar lipid extracts. Sulfonoquinovosyl diacylglycerol was from Marinococcus halophilus (22), archaetidylinositol and archaetidylglycerol from Methanosarcina mazei, and

15 archaetidylglycerophosphate-methyl (AGP-CH ₃ ) from Halobacterium salinarum.

Archaeosome/liposome formulation

Archaeosomes/liposomes were prepared by hydrating 20-3 Omg lipids at 4O ⁰ C in 2ml of PBS buffer (1OmM sodium phosphate, 16OmM NaCl, pH 7.1) or water

20 with (Ag-loaded) or without (Ag-free) the test antigen OVA dissolved at lOmg/ml. In some cases, cholesterol (Sigma), DPPS, DPPE, or DPPG (Avanti Polar Lipids) polar lipids were mixed in chloroform/methanol with the synthetic glyco-archaeols. These were dried to remove all traces of solvent and hydrated in PBS or water, as above. The size of the vesicles in the preparations was

25 decreased by sonication in a sonic water bath (Fisher Scientific) at 4O ⁰ C. Antigen not entrapped was removed by centrifugation and washing. Quantification of antigen loading was done by SDS polyacrylamide gel electrophoresis as described and based on salt corrected dry weights (27). Average diameters were determined by particle size analysis using a 5 mW He/Ne laser (Nicomp 370).

37

Mice

To determine adjuvant activity, various synthetic archaeosomes with OVA entrapped (OVA-archaeosomes) were used to immunize female C57BL/6 mice on days 0 and 21 (6-8 weeks old on first injection). Injections were subcutaneous 5 at the tail base with 0.1ml PBS containing 15μg OVA entrapped in 0.2-0.63mg lipids. Blood samples were collected from the tail vein for anti-OVA IgG/IgM antibody titration done by ELISA (13). In some cases faecal extracts were assayed for IgA (and IgG) to assess mucosal responses following systemic immunizations. Fresh faeces (about 30 mg dry weight) were collected from 4

10 mice/group, and extracted by adding 0.5 ml of phosphate buffered saline (pH 7.0) containing 10% fetal calf serum and 0.1% sodium azide. Homogenized samples were centrifuged and Elisa measured anti OVA antibody in supernatants. Spleens were collected to determine CTL activity using the Cr^l -assay with specific and non-specific targets EG.7 and EL-4, respectively (14). hi Elispot assays antigen-

15 specific CD8 ⁺ T cell activity was measured in splenic cells from immunized mice by determining the number of SIINFEKL stimulated IFN-γ-secreting cells per 5 x 10 ⁵ splenic cells (15).

A skin melanoma solid tumor model was used in mice to evaluate protection achieved upon vaccination with synthetic archaeosomes. Mice were vaccinated

20 subcutaneously with 15 g OVA or synthetic archaeosome compositions containing 15 g OVA at 0 and 3 weeks. Archaeosome lipid compositions were synthetic archaeal lipid/DPPG/cholesterol (25/55/20 mol%). At week 7 post first injection 5 x 10 ⁶ B16OVA cells were injected subcutaneously in the shaved lower dorsal area. The time was recorded when solid tumours became evident.

25

38

Table 1. H NMR data of Man^-archaeol compounds - Sugar and protecting groups (n.d. not determined).

O O -4

K)

O\ -4

O

H

O

> κ>

O O ;4

O O O Cλ

W O

40

O O -4

K)

O\ -4

O

H

O

> κ>

O O ;4

O O O Cλ

W

O

42

O O -4

K) Ul o -4\

UJ

O

H

O

> κ>

O O -4

O O O Ul

W

O

ro ^" ro ro C

C-

CS VO r—S

O CS OO OO ON CS oo ro _-< Tt ro

¹ r-~ r~ C- C- C-

Tf Tt Tf Tf od OO C-

CS CrS~-

CS CS

*—1

■ rs ^"

OO ro CS Άro OO ro oo OO r

CS~ ro

OO ^* ro CS ^^ ro _T -oH

' '

CS oo" ro CS VO (N T-H

VO O m ro _^ H VO O Tt ro —I

OO OO OO oo oo oo on OO OO OO OO

CO ro ro CS ro ro ro ro ro ro ro ro

■

CS

I CS i

O

I ■

>n in

ON VO C- CS CS ro co Tf CS oo OO Tt ON _T -H O

OO σ\ CS CS CS O\ ON ON ON oo oo ON CS ro ro

VO vo VO VO VO VO VO VO VO VO VO VO VO VD VD VD

r O

ON Ό T-oH O VO C-- r~ VO ro OO OO C- ro VO VD m

T-H Tt Tf Tf Tt Tf Tf Tt Tt Tf Tt Tf Tf c— c- t— c- r^ C- C- C- C- c— C- C- C- C-

CS ro Ό o> OO OO C- ro c— VD vo Tf O OO VO

Tt OO OO !—I __( c- c- CS T-H T-H CS ON OO oo

VD VD VO r~- C-- t— C- C- C- C- C— VD VO VD

Tt vo c— σ\ ro ro ro ro cs ro Tf VO O O σ\ O T-H OO Os ON c- ON ON ON ON O 1—I CS CS

00 C- C- C- r— C- c— C- C- OO r- c— C-

ro C-

ON co VO Tf CS CS Tf OO U~ι 1 1 ON OO

Tt OO ON ON O in OO in OO ON ON c- O

VD C- c- P^ r~ C- C- VO C- C- C- VD oo C- r-

c- ON VO C- ON ON O ON σ O VO _ Cλ OO O Tf OO O O __; ON CS

O O ON O O OO O O ON oo O O ON

^ ^H ON r-H σ\ ^r ~~ ^l r—• O

ON ON T-H T-H T-H O O

ON - ^H

U <;

J-I I μ-4

1—H U

1—I I m

|-| »-H μ-H r-H T-H |-| T-H

U Cd Cj ca

2 2 i m C a 9

44

45

Table 3, ¹ H NMR data of glucose com ounds - Su ar and rotectin rou s

4--

O

H

O

> κ>

O O -4

O O O Ul

W

O

O O -4

K) Ul o -4\

4-.

OO

O

H

O

> κ>

O O -4

O O O Ul

W

O

49

Compound H-I (Ju) H-2 (J ₁₃ ) H-3 J( ₃ , ₄ ) H-4 (J _4i5 ) H-5 JG.6) H-6 (Js.6-) H-6' I

(M Ac(CH ₃ ) ^ O residue O O

MaIt-A GIc ¹ 4.58 d (8.0) 4.83 brt (9.5) 5.26 brt (9.0) 3.99 brt (9.0) 3.66 m 4.46 dd (2.4) 4.23 m (11.9) 2.13, 2.09, 2.04^ 2.01, 2.00, 1.99K (6) cj

GIc" 5.40 d (4.2) 4.86 dd (9.3) 5.35 brt (9.8) 5.04 brt (9.7) 3.95 m 4.23 m (3.0) 4.03 dd (12.5)

MaIt-B GIc ¹ 4.26 d (7.9) 3.28 m (9.7) 3.61 brt (9.1) 3.54 brt (8.9) 3.31 m 3.83 m 3.83 m

GIc" 5.10 d (3.8) 3.46 dd (9.9) 3.62 brt (9.6) 3.26 brt (9.1) 3.66 m 3.69 m 3.82 m iMalt-A GIc' 4.59 d (7.0) 4.94 dd (9.7) 5.20 brt (9.3) 5.07 brt (9.4) 3.65 m (4.6) 3.75 dd 3.65 (10.9) 2.12, 2.09, 2.05, 2.03, 2.02, 2.00 (6)

GIc" 5.12 d (3.8) 4.86 dd (10.3) 5.44 brt (9.6) 5.05 brt (9.7) 4.07 m (4.4) 4.26 dd 4.07 m (12.6) iMalt-B GIc ¹ 4.30 d (7.9) 3.24 brt (9.1) 3.38 brt (9.1) 3.50 m 3.41 m (3.5) 4.03 dd 3.64 m (10.8)

GIc 4.82 d (3.4) 3.37 dd (10.0) 3.62 brt (9.9) 3.32 m 3.64 m (2.3) 3.78 dd (5.3) 3.69 dd (11.4)

Cello-A GIc ¹ 4.52 d (8.3) 4.92 m (9.5) 5.15 brt (9.3) 3.76 brt 3.55 m 4.48 (4.9) 4.08 dd (12.0) 2.11, 2.07, 2.015, 2.011, 2.C

O (6), 1.97

GIc" 4.49 d (8.1) 4.92 m (9.9) 5.13 brt (9.5) 5.05 brt 3.64 m (3.9) 4.36 dd 4.03 brd (12.2)

Cello-B GIc ¹ 4.29 d (7.8) 3.29 dd (8.8) 3.53 brt (9.1) 3.57 m 3.37 m 3.85 m 3.85 m

GIc" 4.39 d (7.9) 3.26 dd (8.5) 3.36 brt (9.1) 3.38 brt (9.4) 3.34 m 3.87 brd 3.67 brd (10.8)

Lac-A GIc ¹ 4.54 d (8.0) 4.90 dd (9.6) 5.18 brt (9.2) 3.79 brt (9.6) 3.54 m 4.47 brd 4.09 m 2.15, 2.11, 2.06, 2.04 (6), 2.03, 1.96

GaI" 4.47 d (8.0) 5.10 dd (10.5) 4.95 dd (3.4) 5.35 brd 3.86 m 4.09 m 4.09 m

Lac-B GIc ¹ 4.29 d (7.8) 3.29 m 3.57 m 3.57 m 3.38 m 3.85 m 3.85 m

Gal" 4.34 d (7.8) 3.57 m 3.48 m (3.5) 3.83 brd 3.63 m (7.2) 3.79 dd (4.4) 3.70 (11.9)

MeIo-A GIc ¹ 4.58 d (8.0) 4.94 brt (8.9) 5.19 brt (10.0) 5.12 m 3.58 m (4.0) 3.74 dd 3.58 m (10.8) 2.132, 2.128, 2.05, 2.04, 2.03p3 2.00, 1.98 H

Gal" 5.17 d (4.1) 5.10 dd (10.4) 5.33 dd (3.2) 5.45 brt (2.6) 4.22 m 4.08 brd 4.08 brd O

MeIo-B GIc ¹ 4.30 d (7.9) 3.24 dd (9.1) 3.38 brt (8.6) 3.47 m 3.47 m (2.9) 4.05 dd 3.63 m (10.6) O

Gal" 4.87 d (3.0) 3.73 m 3.91 m 3.73 m 3.85 brt 3.73 m 3.73 m O -4

Gentα GIc ¹ 4.78 d (3.5) 3.40 dd (10.0) 3.63 brt (9.5) 3.47 brt (9.8) 3.70 m 4.08 brd 3.79 dd (10.3) O O

GIc ¹¹ 4.31 d (7.8) 3.26 m 3.38 m 3.32 m 3.26 m (2.3) 3.85 dd (4.9) 3.69 m (11.8) Cλ O

Table 6 C NMR Data of Dissachaπdes - Su ars and Protectin Grou s

52

Table 7. ¹ H NMR data of the core lipid (archaeol) in synthetic archaeol-compounds (Glycl-3 is H on glycerol carbons sn-1-3; Phyl-2 is H on Cl Cl', C2 or CT; CH, CH ₂ , CH ₃ re resent combined H si nals from these rou s of iso ranoid chains.

O O -4

K) Ul o -4\

Ln

O

H

O

> κ>

O O -4

O O O Ul W

O

54

_I B _t -

_I C _e

B o-

S ε 8 8

IOε VO OεO (Nε IO IOε VO O B T 8f

OO oo OO [^- OO OO OO OO OO OO

O O O O O O Lac O O O O

CO VO CO (N CN O

O O O O O O O O O

La _c O

I I I I I I I I I I

O O O Os O O O

Tf co TJ- Tj- CO CO Tf r C^O Tf Tf

-—1 _l Me

_bd M _lre

Ef 8 Ef f 8 8 8 ^" 8 Ef

OO CN OεO E oo <Nε «—< i—εI Ef

OS ,_, O Ef Oε O Efs O O •o O B O EOf rε-H

IO co O <N Ό CO O CO VO CO O CO VO CO O CO IO CO r—I

G _e n

r ∑-Hf rε-H Ef g 8

VO Oε r—I OεO O EOf ε VO OεO (N Ef O Ef r- OεO Ef Ef 8

IO (N (N CO O EOf

CNε

\o Ό Tf CO O CO O CO IO CO O CO IO CO «o CO IO CO

T-H

S 8

∞ CεO Tεf Oε Oεs ε 8 8 8

CO CN

Tt Tf Tf Tt Tf Tf r O» Tf Tf "O

CO CO CO CO CO CO CO co CO

S 8 8 6 8

VεO Oε IOε Oε Tf co oo CO (Nε

VO O VO O VO O VO "O VO VO co CO CO CO CO CO CO CO CO CO t ^»

Tf

CO

,-εH Ef 8 8

CεO (N Tεf Oε Ό Ef Oεs T Sf C EOf CO O (N

Tf O Tj- TT IO Tf IO VO rε

O- Tf "O O IO co CO CO CO CO CO CO CO CO CO CO CO CO

a 8 8 8

VεO CεO IOε Oε Tf CO OO CεO CεO

VO O VD IO VO IO VO IO VO VO co CO CO CO CO CO CO CO CO CO

Ef Ef

IO Ef Ef S 8

(N ε VεO CN oεo Ό o Eof Ef g Ef Ef f 8 Ef 8

Oε VO Tf (N CO VO oεo r E-H CO CO VO

Os VO OO O Os Ό OO >o OO VD OO O Os VD oo O OS VO r- Tf co co CO CO CO co co CO CO CO CO CO co CO CO CO CO CO CO CO

CQ S < < m

O < CQ ό ό 8

W a

13 O

55

Table 8. C NMR Data of Li ids in Archaeol Com ounds

O O -4

K) Ul o -4\

ON

O

H

O

> κ>

O O -4

O O O Ul

W

O

57

58

59

60

Table 9. M.S. data for archaeol compounds.

61

62

Table 10. Retention of antigen (OVA) in liposomes*

*Leakage of OVA from duplicate 0.1 -ml aliquots of liposomes was determined following storage for 6 months at 4°C. Samples were centrifuged at 202,400 x g (R _ave ) for 1 h. OVA present in supernatant and pellet fractions were quantified by density of the Coomassie stained OVA band following SDS PAGE. Choi = cholesterol.

63

Table 11. Preparation of OVA-liposomes and OVA-archaeosomes containing s nthetic l co-archaeol li ids

*Glc ₂ -A is β-D-Glc-(l,6)-β-D-Glc-archaeol; GIc ₂ -C-PS is gentiobiose caldarchaetidylserine; GIc ₂ -C-PI is gentiobiose caldarchaetidylinositol; MaIi ₄ -A is α-D-Man-(l,2)-α-D-Man-(l,2)-α-D-Man-(l,2)-α-D-Man-(l,l)- archaeol; GIc ₃ -A

64

is β-D-Glc-(l,6)-β-D-Glc-(l,6)-β-D-Glc-(l,l)-archaeol; AS, archaetidylserine; GIc ₂ -C-GIc ₂ , gentiobiose-caldarchaeol-gentiobiose (see Figs. 6 and 11). Table 12. Mucosal responses in mice immunized with various adjuvant compositions containing entrapped antigen (OVA).*

*C57BL/6 mice were immunized subcutaneously at 0 and 3 weeks. Faecal extracts from faeces collected 6 weeks post first injection were assayed for anti OVA antibodies (IgA diluted 1:1; IgG diluted 1:5). Relative amounts of antibody

10 are shown by Elisa absorbance, and normalized based on 100 mg dry weight of faeces. Negative control values (blank values) for faecal extracts from non- immunized mice are shown. Vaccine compositions were the same as in Figures 10 and 12. TPL archaeosomes from Methanobrevibacter smithii show relatively low IgA response.

65

Table 13. Up-regulation of co-stimulatory molecules on APCs exposed to synthetic glyco-archaeol archaeosomes.*

*Macrophages J774A.1 cultures were incubated in RPMI + 8% faetal bovine serum medium containing no addition (naive), lOμg lipopolysaccharide (LPS) from E. coli as a positive activator, 25 μg liposomes or 25 μg archaeosomes. Cultures were incubated for 48h, except for LPS that was incubated 24h, prior to staining for presence of the co-stimulatory molecule CD80 using anti-CD80-PE

10 (Phycoerythrin). Fluorescence intensity was measured using flow cytometry. Data are acquired from 30,000 events for each sample and the fluorescence intensity of each sample is indicated.

66

Table 14. Protection against skin melanoma in C57BL/6 mice vaccinated with synthetic archaeosomes containing an antigen expressed by the melanoma cells.*

* The numbers of mice that developed a solid tumor (5 mice/group) are shown at various times (days) from subcutaneous injection of tumor cells. ** OVA represents the antigen with no adjuvant as a control.

67

Reference List

1. Agnihotri, G., P. Tiwari, and A. K. Misra. 2005. One-pot synthesis of per-O- acetylated thioglycosides from unprotected reducing sugars. Carbohydr.Res.

5 340:1393-1396.

2. Benvegnu, T., G. Rethore, M. Brard, W. Richter, and D. Plusquellec. 2005. Archaeosomes based on novel synthetic tetraether-type lipids for the development of oral delivery systems. Chem.Commun.(Camb.) 5536-5538.

3. Berkowitz, W. F. and Y. Wu. 1997. Synthesis of archaebacterial lipid C20 10 chirons. Tetrahedron Letters 38:8141-8144.

4. Blocher, D., R. Gutermann, B. Henkel, and K. Ring. 1990. Physicochemical characterization of tetraether lipids from Thermoplasma acidophilum. V. Evidence for the existence of a metastable state in lipids with acyclic hydrocarbon chains. Biochim.Biophys.Acta 1024:54-60.

15 5. Cross, G. G. and D. M. Whitfield. 1998. Simplifying Oligosaccharide Synthesis: Boronate Diesters as Cleavable Protecting Groups. SYNLETT. 487-488.

6. Douglas, S. P., D. M. Whitfield, and J. J. Krepinsky. 1995. Polymer-supported solution synthesis of oligosaccharides using a novel versatile linker for the synthesis of D-mannopentaose, a structural unit of D-mannans of pathogenic

20 yeasts. J.Am.Chem.Soc. 117:2116-2117.

7. Eichler, E., F. Yan, J. Sealy, and D. M. Whitfield. 2001. 1-Methyl I ¹ - cyclopropylmethyl- an acid labile O-protecting group for polymer-supported oligosaccharide synthesis. Tetrahedron 57:6679-6693.

8. Freisleben, H. J., Antonopoulos, E., Balakirev, M., Balakirev, L., Hartmann, K., 25 and Gropp, F. Tetraether lipid derivatives and liposomes and lipid agglomerates containing tetraether lipid derivatives, and use thereof. [US 6,316,260]. 1999. Ref Type: Patent

68

9. Gophna, U., R. L. Charlebois, and W. F. Doolittle. 2004. Have archaeal genes contributed to bacterial virulence? Trends Microbiol. 12:213-219.

10. Gurnani, K., J. Kennedy, S. Sad, G. D. Sprott, and L. Krishnan. 2004. Phosphatidylserine receptor-mediated recognition of archaeosome adjuvant

5 promotes endocytosis and MHC class I cross-presentation of the entrapped antigen by phagosome-to-cytosol transport and classical processing. J.Immunol. 173:566-578.

11. Huang, X., L. Huang, H. Wang, and X.-S. Ye. 2004. Iterative One-Pot Oligosaccharide Synthesis. Angew Chem.Int.Ed. 43:5221-5224.

10 12. Kates, M. 1992. Archaebacterial lipids: structure, biosynthesis and function. Biochem.Soc.Symp. 58:51-72.

13. Krishnan, L., C. J. Dicaire, G. B. Patel, and G. D. Sprott. 2000. Archaeosome vaccine adjuvants induce strong humoral, cell-mediated, and memory responses: comparison to conventional liposomes and alum. Infect.Immun.

15 68:54-63.

14. Krishnan, L., S. Sad, G. B. Patel, and G. D. Sprott. 2000. Archaeosomes induce long-term CD8 ⁺ cytotoxic T cell response to entrapped soluble protein by the exogenous cytosolic pathway, in the absence of CD4 ⁺ T cell help. J.Immunol. 165:5177-5185.

20 15. Krishnan, L., S. Sad, G. B. Patel, and G. D. Sprott. 2003. Archaeosomes induce enhanced cytotoxic T lymphocyte responses to entrapped soluble protein in the absence of interleukin 12 and protect against tumor challenge. Cancer Res. 63:2526-2534.

16. Krishnan, L. and G. D. Sprott. 2003. Archaeosomes as self-adjuvanting delivery 25 systems for cancer vaccines. Journal of Drug Targeting 11 :515-524.

17. Minamikawa, H. and H. Masakatsu. 1997. Phase behavior of synthetic Phytanyl-Chained Glycolipid/water systems. Langmuir 13:2564-2571.

69

18. Minamikawa, H., T. Murakami, and H. Masakatsu. 1994. Synthesis of 1,3-di-O- alkyl-2-O-(β-glycosyl)glycerols bearing oligosaccharides as hydrophilic groups. Chem.Phys.Lipids 72:111-118.

19. Palma, A. S., T. Feizi, Y. Zhang, M. S. Stoll, A. M. Lawson, E. Diaz-

5 Rodriguez, M. A. Campanero-Rhodes, J. Costa, S. Gordon, G. D. Brown, and

W. Chai. 2006. Ligands for the beta-Glucan Receptor, Dectin-1, Assigned Using "Designer" Microarrays of Oligosaccharide Probes (Neoglycolipids) Generated from Glucan Polysaccharides. J.Biol. Chem. 281 :5771-5779.

20. Prakash, G. K. S., T. Matthew, E. R. Marinez, P. M. Esteves, G. Rasul, and G.

10 O. Olah. 2006. BF ₃ 2CF ₃ CH ₂ OH (BF ₃ 2TFE), an effcient superacidic caralyst for some organic synthetic transformations J.Org.Chem. 71 :3952-3958.

21. Raguse, B., P. N. Culshaw, J. K. Prashar, and K. Raval. 2000. The synthesis of archaebacterial lipid analogues. Tetrahedron Letters 41 :2971-2974.

22. Sprott, G. D., L. Bakouche, and K. Rajagopal. 2006. Identification of 15 sulfoquinovosyl diacylglycerol as a major polar lipid in Marinococcus halophilus and Salinicoccus hispanicus and substitution with phosphatidylglycerol. Can. J.Microbiol. 52:209-219.

23. Sprott, G. D., J. Brisson, C. J. Dicaire, A. K. Pelletier, L. A. Deschatelets, L. Krishnan, and G. B. Patel. 1999. A structural comparison of the total polar

20 lipids from the human archaea Methanobrevibacter smithii and

Methanosphaera stadtmanae and its relevance to the adjuvant activities of their liposomes. Biochim.Biophys.Acta 1440:275-288.

24. Sprott, G. D., I. Ekiel, and C. Dicaire. 1990. Novel, acid-labile, hydroxydiether lipid cores in methanogenic bacteria. J.Biol.Chem. 265:13735-13740.

25 25. Sprott, G. D., Krishnan, L., Conlan, J. W., Omri, A., and Patel, G. B.

Archaeosomes as adjuvants and carriers for acellular vaccines to induce cytotoxic T lymphocyte (CTL) responses. [WO 01/26683 A3]. 2001.

70

Ref Type: Patent

26. Sprott, G. D., S. Larocque, N. Cadotte, C. J. Dicaire, M. McGee, and J. R. Brisson. 2003. Novel polar lipids of halophilic eubacterium Planococcus H8 and archaeon Haloferax volcanii. Biochim.Biophys.Acta 1633:179-188.

5 27. Sprott, G. D., G. B. Patel, and L. Krishnan. 2003. Archaeobacterial ether lipid liposomes as vaccine adjuvants. Methods Enzymol. 373:155-172.

28. Sprott, G. D., Patel, G. B., Makabi-Panzu, B., and Tolson, D. L. Archaeosomes, archaeosomes containing coenzyme QlO, and other types of liposomes containing coenzyme QlO as adjuvants and as delivery vehicles. PCT/CA

10 96/00835[WO 97/22333], 1-57. 1997.

Ref Type: Patent

29. Sprott, G. D., S. Sad, L. P. Fleming, C. J. Dicaire, G. B. Patel, and L. Krishnan. 2003. Archaeosomes varying in lipid composition differ in receptor-mediated endocytosis and differentially adjuvant immune responses to entrapped protein.

15 Archaea 1:151-164.

30. Swain, M., J. R. Brisson, G. D. Sprott, F. P. Cooper, and G. B. Patel. 1997. Identification of beta-L-gulose as the sugar moiety of the main polar lipid Thermoplasma acidophilum. Biochim.Biophys.Acta 1345:56-64.

31. Tropper, F. D., F. O. Andersson, C. Grande-Maitre, and R. Roy. 1991.

20 Stereospecific synthesis of 1 ,2-trans-phenylthio-β-D-disaccharides under phase transfer catalysis. Synthesis 734-736.

32. van Boeckel, C. A. A., P. Westerduin, and J. H. van Boom. 1984. Synthesis of two purple-membrane glycolipids and the glycolipid sulfate O-(β-D- glucopyranosyl-3-sulfate)-(l-6)-0-α-D-mannopyranosyl-(l-2)- 0-α-D-

25 glucopyranosyl-(l-l)-2,3-di-O-phytanyl-£H-glycerol. Carbohydrate Research

133:219-234.

71

33. Yan, H. and H. J. Jennings. 2006. Synthesis of mono- and di-sialophospholipids via the H-phosphonate approach. Can.J.Chem. 84:540-545.

72