Synthesis, gp120 binding and anti-HIV activity of fatty acid esters of 1,1-linked disaccharides

Abstract

Inspired by the anti-human immunodeficiency virus (HIV) activity of analogues of β-galactosylceramide (GalCer), a set of mono- and di- saccharide fatty acid esters were designed as GalCer mimetics and their binding to the V3 loop peptide of HIV-1 and anti-HIV activity evaluated. 1,1-linked Gal-Man and Glu-Man disaccharides with an ester on the Man subunit bound the V3 loop peptide and inhibited HIV infectivity in single round infection assays with the TZM-bl cell line. IC₅₀'s were in the 50 μM range with no toxicity to the cells at concentrations up to 200 μM. These compounds appear to inhibit virus entry at early steps in viral infection since they were inactive if added post viral entry. Although these compounds were found to bind to the V3 loop peptide of gp120, it is not clear that this interaction is responsible for their anti-HIV activity because the relative binding affinity of closely related analogues did not correlate with their antiviral behavior. The low cytotoxicity of these 1,1-linked disaccharide fatty acid esters, combined with the easy accessibility to structurally diverse analogues make these molecules attractive leads for new topical anti-viral agents.

1. Introduction

While the fusion process mediated by gp41 per se remains well studied and targeted, the events preceding gp41 mediated fusion including gp120 binding to receptor and coreceptor present potential new targets for drug development.^i,ii Besides the well characterized receptor and coreceptor for HIV several cofactors are involved in facilitating gp120 binding to cells.^iii,iv Membrane glycosphingolipids (GSL) including GalCer 1 constitute one well-known group of cofactors^v,vi (Figure 1). In this vein, GSL analogues have attracted interest as potential inhibitors of HIV infectivity.^{ii,vii,viii,ix,x} We have previously synthesized and tested the gp120 binding and antiviral activity of GalCer analogues with simple ceramide substitutes, against both X4 and R5 tropic virus.^xi Some of the lead compounds, represented by 2, bound gp120 similar to GalCer and showed significant activity against HIV envelope glycoprotein (Env)-mediated fusion. The compounds act, presumably as GalCer mimetics, at an early step in HIV Env binding that precedes CD4 binding as determined by kinetic and temperature arrested state experiments. These GalCer analogues also showed inhibitory activity against vesicular stomatitis virus (VSV) pseudotype virus that was similar to other antiviral compounds like cyanovirin and suramin.^xii,xiii The further development of these compounds holds promise towards new therapies that can be used both as prophylaxis and post exposure treatment as the compounds act at an early step in the HIV cycle. Towards this end, with the aim of identifying new GalCer mimetics with anti-HIV activity, we designed a wider set of mono- and di- saccharide-derived lipids and evaluated their binding to the V3 loop peptide of gp120 and their HIV inhibitory activity. These results are discussed herein.

Figure 1. GalCer and analogues.

Our analog design is based on existing SAR data that suggests the Gal and hydrophobic residues of GalCer account for a significant part of the overall binding to gp120,^xiv and the central polar segment that connects the two regions acts as a scaffold that controls the spatial positioning of the sugar and hydrophobic regions.^xv,xvi That our first generation analogues (e.g. 2), in which the ceramide moiety in GalCer was replaced with simple hydrocarbon residues, showed binding properties comparable to GalCer is consistent with this model. To get insight into the conformational and spatial requirements of the ceramide residue in GalCer, for the purpose of developing more effective GalCer antagonists, we argued that a monosaccharide could be used as a replacement for polar lipid head of GalCer. Indeed, it has been suggested that the GlcNHAc subunit in Galβ1-4GlcNAcβ1 is structurally analogous to this segment of GalCer and may mimic its binding to gp120.^xvii In a similar vein, we argued that β-Gal disaccharides such as 3, 4 and 5, with one or two fatty acid esters in the mannose segment, may also act as GalCer mimetics (Figure 2). While the gp120 binding of disaccharide glycolipids have been previously examined, these structures were conceived as mimetics of disaccharide GSL's, as opposed to the monosaccharide GalCer.^x,xvi We opted for a 1,1-linked disaccharide template because of the well-defined conformational bias in the intersaccharide torsions of such frameworks,^xviii and a mannose subunit, because of synthetic considerations relating to stereoselective glycoside bond construction and alcohol group differentiation. The β-Glu-disaccharide 6, the α-Gal-disaccharide 9, and the α -Man monosaccharides 7 and 8 were designed to probe the recognition specificity for the β-Gal motif. The Man-monosaccharide 7 was also of interest because the relative orientation of the 6-O-palmitoyl ester and the 2-OH group was similar to the positioning of the 4-OH and the anomeric lipid chain Gal-monosaccharide 2a, the reference compound from the earlier study.

Figure 2. Proposed GalCer mimetics.

2. Results and Discussion

2.1 Binding to V3 loop peptide of gp120

The binding of 3 - 8 to a synthetic V3 loop peptide, RIQRGPGRAFVTIGK, corresponding to the R15K peptide, from gp120 of HIV-1 IIIb isolate was evaluated. We and others have previously used the R15K peptide as a model for the glycolipid binding domain of gp120.^xi,xix The change in surface pressure (Δπ) at the air-water interface of individual glycolipid monolayers of the test compounds, on exposure to an aqueous solution of the R15K peptide (10 μM) was measured. Increase in surface pressure is associated with integration of the peptide into the glycolipid monolayer. The critical pressure of insertion (CPI), calculated by extrapolation for a null increase in surface pressure is often used as a measure of binding affinity. All the compounds tested interacted with the R15K V3 peptide with CPIs in the range of 18 - 49 mN/m (Figure 3). These values were comparable to that measured for GalCer (CPI = 22-25 mN/m).^xi The likeness of 3 - 6 to, and the similarity of their CPIs to GalCer, are consistent with our hypothesis that these disaccharides mimic GalCer in their binding to the V3 peptide, but in the absence of additional data this conclusion is very speculative. It is also noteworthy that the monosaccharides 7 and 8 appear to bind appreciably more strongly to the V3 peptide than the disaccharides. However, give the structural disparity between the mono- and di- saccharide frameworks, the relative activity of the two groups of analogues does not provide any definitive SAR insight because the two sets of analogues may be binding to different domains on the peptide. Nevertheless the possibility that 3 - 8 could be interacting with the GalCer domain on the V3 peptide, and earlier findings that such compounds may inhibit the attachment of HIV to host cells, the next step was to examine their anti-HIV.

Figure 3. Binding of glycolipid analogues to the V3 loop peptide.

Each analog was spread on the surface of a water subphase at various values of the initial surface pressure. A stable monomolecular film was formed after evaporation of the solvent. The synthetic V3 peptide was then added in the subphase at a final concentration of 10 μM. The maximal surface pressure increase (Δπ_max) was determined after 2 to 3 h of incubation, i.e. when equilibrium was reached. The data were analyzed with the Filmware 2.5 program (Kibron, Inc.)

2.2 Anti HIV activity

The anti-HIV activity of the synthetic glycolipids was evaluated in a virus infection assay using the TZM-bl indicator cell line infection with CXCR4 tropic HIV (NL-Lai).^xx The β-Gal-disaccharide 4 with the O-3 ester on the mannose residue, the β-Glu-disaccharide 6 with the O-6 ester and the β-Gal-disaccharide 9 with the O-6 ester, showed appreciable activity with IC₅₀'s of 75, 53 and 48 μM respectively, which was comparable to the activity of the previously assayed monosaccharide 2a (44 μM, Figure 4, Table 1). The β -Gal-disaccharide 3 with the ester at O-6 of the mannose ring, its di-O-ester 5, and the monosaccharides 7 and 8 showed no significant activity up to 100 μM. Thus the data from the binding and anti-HIV assays indicate that V3 loop binding by itself is not a reliable predictor of anti-HIV activity. This situation may arise because the binding assay measures overall binding but anti-HIV activity requires binding to a specific domain of the peptide. Therefore, the binding data does not necessarily correlate with anti-HIV activity. Within the context of GalCer mimicry, it may be contended that the disaccharides interact with the GalCer binding domain on the V3 peptide but the monosaccharides bind other domains, so that the relative binding affinity of the mono- and di- saccharide analogues is not connected to their anti-HIV activity. However, this argument does not explain why the disaccharide 3 (which is very similar in structure to 4 and 6), and to a lesser extent 5, are active in the binding assay but not in the anti-HIV assay. One explanation for this apparent anomaly is that glycolipid presentation in the two assays are different,^vii so that structural factors that control the morphology of the active glycolipid species conflict with those that affect glycolipid-peptide recognition, thereby skewing the data. Of course, an alternative and arguably more obvious, explanation for the contrasting behavior of individual test compounds in the two assays is that the mechanism for anti-viral activity is not at all connected to V3 loop binding. Notwithstanding the mechanism for anti-viral activity, the data from the anti-HIV assay suggests that a disaccharide framework (vs. monosaccharide) and one ester residue (vs. two), are important for anti-viral activity. That the β-galactoside 4, the β-glucoside 6 and the α-galactoside 9 were active but the β-galactoside 3 was not, also suggests that the location of the ester is important, and that the optimal position for the ester may vary with the structure of disaccharide framework.

Figure 4. Anti HIV activity of glycolipid analogues.

TZM-bl cells were infected with X4 HIV NL-Lai in the presence of serial dilutions of the compounds. Infectivity was determined 24 h later by measuring luciferase activity and normalized to control wells infected with virus in the absence of drug. Pooled data from three independent experiments was used to fit curves using Sigma Plot software.

Table 1. IC₅₀'s of glycolipid analogues.

IC₅₀'s were determined from the data in Figure 4 for compounds with IC₅₀'s of less than 100 μM

Compound	IC50 μM
2a	43.51 ± 1.63
3	>100
4	75.99 ± 6.16
5	>100
6	52.51± 1.93
7	>100
8	>100
9	48.35 ± 1.74
GML	>100

We also examined the activity of glycerol monolaurate (GML, 1-O-lauryl-glycerol) in the TZM-b1/HIV assay. GML shows low micromolar activity against two other envelope viruses herpes-simplex virus 2 and bacteriophage φ6.^xxi Since our active glycolipids and GML are both fatty acid esters, an obvious question was whether the glycolipids and GML inhibit HIV infectivity by a similar mechanism. However, GML was not active at concentrations of up to 100 μM. Parenthetically, it should be noted that GML has been shown to prevent simian immunodeficiency virus (SIV) in a rhesus macaque model.^xxii The low activity observed for GML in our cellular assay suggests that the activity observed for GML in the animal model does not result from inhibition of HIV entry. Indeed the results in the animal studies have been linked to immunological effects. That the three disaccharide analogues 4, 6 and 9 were active but the monosaccharide esters and GML were not, suggests a specificity for a disaccharide scaffold with a single mono fatty acid ester.

2.3 Cytotoxicity assays

Cytotoxicity was determined for test compounds against TZM-bl cells over a 24 h incubation period using a MTS dye reduction assay.^xxii None of the compounds were toxic up to a 200 μM dose.

2.4 Pre and Post Infection assays

To determine whether the compounds affected the viral entry process or events post viral entry, simple time of addition experiments were conducted.^xi TZM-bl cells were exposed to the virus in the presence of the compounds (100 μM) or the compounds were added 4 h after infection with the virus. We also used T20^xxiv, an inhibitor of gp41 mediated fusion and AMD 3100^xxv, a CXCR4 antagonist that inhibits HIV Env mediated fusion and viral entry as controls in these experiments. All active compounds inhibited virus infection if added at the same time as the virus (Figure 5). However addition of the compounds 4 h post-infection failed to show significant inhibition. The same was true for the entry inhibitors T20 and AMD3100. The absence of activity when the compounds were added 4 h post-infection also suggest that they did not have a direct effect on cell viability or function and hence the inhibition is at the step of initial virus cell interaction. This mode of action appears to be similar to 2a^xi but contrasts with the suggestion that a 3-O-fatty acid ester of trehalose (α, α-1,1-Glu-Glu) that is structurally similar to 4, prevents influenza proliferation by inhibiting viral replication.^xxvi,xxvii

Figure 5. Glycolipid analogues show activity pre-infection only.

TZM-bl cells were infected with NL-Lai virus. Compounds 100 μM were added either prior to infection (Pre infection) or 4 h post infection. Infectivity was determined 24 h later by measuring luciferase activity. AMD and T20 controls are used at 2 μM. Data are mean+- SD of triplicate observations.

2.5 Critical micelle concentration (CMC)

The CMC's of selected analogues were measured in order to determine the physical state of the glycolipids at the concentrations used in the cellular assays. Values were obtained by tensiometer measurements of solutions of glycolipids in water, as previously described.^xi The CMC is taken as the concentration of glycolipid that does not induce any further decrease in surface tension. This data was obtained in triplicate: 3 (9.8, 10 and 10.5 μM); 4 (12, 12 and 12.5 μM); 6 (10, 10.5, and 12 μM); 7 (13, 13.5 and 13.7 μM). Thus at their active concentrations analogues 4 and 6, exist as micelles, but this does preclude the possibility that the active species is monomeric. The CMC's obtained for 3, 4 and 6 were in the range observed for mono-fatty acid esters of related 1,1-linked disaccharides.^xxviii,xxix The diester 8 was difficult to study; a regular decrease in surface tension from 72.8 mN/m (pure water) to 0 mN/m (glycolipid film collapse) in the concentration range of 0.25–5 μM was observed. One interpretation of the results is that 8 forms precipitates rather than micelles in water above a concentration of 0.5 μM.

2.6 Synthesis

The synthesis of the disaccharide lipids 3 - 6 started with assembly of the 1,1-linked precursors 13β and 14β (Scheme 1). Initial glycosidation reactions using 2-O-acetylated Gal donors containing different alcohol protecting and glycosyl activating groups with Man acceptors gave low yields of disaccharide products. A solution to this problem, albeit not a stereoselective one, was found in the methyl triflate promoted glycosidation of thioglycoside Man donor 12 with Gal and Glu acceptors 10^xxx and 11^xxxi, respectively. The phenylthio glycoside 12 was prepared from the tetra-O-acetate derivative following the procedure used for the corresponding ethylthio glycoside.^xxxii,xxxiii The reactions of 12 with 10 and 11 afforded in each case an approximately 40% yield of a 1:1 mixture of anomers, α:β disaccharides 13α:β and 14 α:β. The stereochemistry at the anomeric position of the Gal and Glu rings was deduced from vicinal J_H,H values (13α/β: J_1,2 (Gal) = ca. 3.5/8.0 Hz; 14α/β: J_1,2(Glu) = 3.8/8.0 Hz). The α-configuration at C-1 for the Man subunit was assigned on the basis of J_C,H values (13α/β: J_1,1 (Man) = 175/172; 14α/β: J_1,1 (Man) = 175/172 Hz.^xxxiv

Scheme 1. Glycoside synthesis.

The separated anomers from the glycosidation reactions and the known mannoside 21^xxxv were next transformed to the target compounds 3 - 9 through established alcohol protecting group sequences (Scheme 2). These reactions were not optimized in all cases. Representative transformations are described. Acetate hydrolysis on 13β followed by perbenzylation of the resulting tetraol provided the tetra-O-benzyl ether derivative. The 4,6-O-isopropylidene in this product was selectively cleaved under acidic conditions to provide diol 15. DCC mediated esterification of 15 with one equivalent of palmitic acid provided monoester 16 as the major product, and with excess acid, the diester 17. Individual hydrogenolysis of 16 and 17 afforded the mono- and di-O-acylated disaccharide targets 3 and 5. For the 3-O-ester 4, 15 was first transformed to diol 18 via standard protecting group transformations. Selective acylation^xxxvi of the equatorial alcohol in 18 followed by hydrogenolysis of the product provided 4. The β,α-1,1-Glu-Man 6 and the αα-1,1-Gal-Man disaccharide 9 were prepared from 14β and 13α respectively following the reaction sequence used for transformation of 13β to 3. For the monosaccharide esters 7 and 8, 21 was first converted to the 2,3-O-isopropylidene-4-O-palmitoyl- or 2,3-O-isopropylidene-4,6-di-O-palmitoyl- derivatives following procedures used for the analogous disaccharides 16 and 17. Acetal hydrolysis in these materials led to 7 and 8 respectively.

Scheme 2. Esterification and protecting group sequences.

3. Summary

1,1-Linked Gal-Man and Glu-Man disaccharides with a long chain fatty acid ester on the mannose residue were found to inhibit HIV-infectivity in the 50 μM range, with no cytotoxicity at concentrations up to 200 μM. These compounds appear to inhibit virus entry at early steps in viral infection since they were inactive if added post viral entry. Their structural likeness to GalCer and the observation that they bind the V3 loop peptide of HIV-1 and inhibit HIV infectivity fits with the notion their anti-HIV activity results from their ability to bind the GalCer domain of the V3 peptide. However, in the absence of other data this conclusion is very tentative. That closely related disaccharide analogues showed similar binding to the V3 peptide but did not inhibit HIV infectivity suggests that factors other than V3 loop binding may impact on anti-HIV activity, or may point to completely different mechanisms. In this context it is noteworthy that the viruscidal activity of fatty acid derivatives of related disaccharide structures and other small molecule amphiphilic lipids has been attributed to surfactant-like, membrane-perturbing properties.^{xxi,xxxvii,xxxviii,xxxix,xl,xli} Notwithstanding the mechanistic basis for their anti-HIV activity, the preliminary structure activity data obtained from this investigation suggests that a disaccharide-type framework and the position of acylation are important for activity. Athough the trehalose-type fatty acid esters reported here have similar anti-HIV potency to other small molecule lipids, their low cytotoxicity and other reports that trehalose-type disaccharide esters have broader viruscidal activity, combined with the easy accessibility to structurally diverse analogues, make them attractive leads for drug development.^{xlii,xliii,xliv}

4. Experimental

4.1 Surface pressure measurements

The surface pressure of glycolipid monolayers was measured with a fully automated microtensiometer (μTROUGH SX; Kibron, Inc., Helsinki, Finland). The apparatus allowed the real-time recording of the kinetics of interaction of a soluble ligand with the monomolecular film, using a set of specially designed Teflon troughs. All experiments were carried out in a controlled atmosphere at 20°C ± 1°C. Monomolecular films of test compounds were spread on pure water subphases (volume of 800 μL) from hexane-chloroform-ethanol solution as described previously.^xlv After spreading of the film, 5 min was allowed for solvent evaporation. To measure the interaction of the V3 peptide with the individual monolayers, the peptide was injected in the subphase with a 10 μL Hamilton syringe, and pressure increases (Δπ) produced were recorded until reaching a stable value. The experiment was repeated at different values of the initial surface pressure (πi) of the monolayer. The data were analyzed with the Filmware 2.5 program (Kibron, Inc.). The results are expressed as the variations of Δπ as a function of πi for the test compounds. The accuracy of the system under the experimental conditions was ± 0.25 mN/m for surface pressure.

4.2 Antiviral assays

Testing of antiviral activity of test compounds was done as described previously.^xi Briefly, virus stocks were prepared by transfection of 293T cells with NL-Lai full length infectious clone of HIV using Ex gen 500 transfection reagent (Fermentas, Glen Burnie, MD. Virus supernatant was collected 48 h post transfection, cleared of cellular components by centrifugation, aliquoted, stored at −70 °C. Virus titers were determined in TZM-bl cells prior to use. For determination of antiviral activity TZM-b1 cells were seeded in 96 well plates at 2×10⁴ cells per well and allowed to adhere overnight. The subsequent day cells were infected with HIV virus in the presence of 20 μg/mL DEAE dextran and test compounds at various concentrations. Virus infection was determined 24 h post infection as luciferase activity using Brite Lite Luciferase substrate (Perkin Elmer). Percent infection was calculated based on control wells infected with virus in the absence of any compound. Normalized data was fit using Sigma Plot software and IC₅₀ calculated based on the curve fit. Pooled data from at least three independent experiments was used for IC₅₀ determination for each compound.

4.3 Cytotoxicity assays

Cytotoxicity of compounds was determined by a 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) dye reduction assay. Briefly TZM-bl cells were seeded in 96 well plates and incubated overnight. Subsequent day serial dilutions of compounds starting at 200 μM were added to the cells. The cells were incubated for another 24 h before addition of the MTS reagent (Cell titer Aqueos one solution, Promega). Reduction of the MTS dye was determined by measuring OD at 490 nM and normalized to cells incubated with media only.

4.4 Pre and Post Infection assays

To determine the step at which the compounds showed activity a pre/post infection time of addition assay was conducted. TZM-bl cells were infected with NL-Lai virus. The compounds (100 μM) were either added prior to infection (pre infection) or 4 h post infection. For the post infection the uninfected virus was removed by washing with PBS followed by addition of fresh media containing the compounds. For controls T20 and AMD3100 were used at 2 μM conc.

4.5 CMC Measurements

Stock solutions of glycolipids were prepared in hexane: chloroform:ethanol (11:5:4, v/v/v) and inJected in water with a Hamilton microsyringe (dilution 1:1000). The surface tension was continuously recorded with the Kibron microtensiometer. Below the CMC, a drop in surface tension was recorded after each glycolipid injection. Increasing the concentration of the added glycolipid resulted in a linear decrease in surface tension. The CMC was determined as the concentration of glycolipid above which there was no further decrease in surface tension.

4.6 Synthesis - General

Unless otherwise stated, all reactions were carried out under a nitrogen atmosphere in oven-dried glassware using standard syringe and septa technique. ¹H and ¹³C NMR spectra were obtained on a Bruker 500 MHz spectrometer. Chemical shifts are relative to the deuterated solvent peak and are in parts per million (ppm). ¹³C NMR peaks are listed to the nearest 0.1 ppm. When two or more signals occur as distinct paks within 0.1 ppm, the number of these signals are indicated in parenthesis. Assignments for selected nuclei were determined from ¹H COSY experiments. Thin layer chromatography (TLC) was done on 0.25 mm thick precoated silica gel HF₂₅₄ aluminum sheets. Chromatograms were observed under UV (short and long wavelength) light, and were visualized by heating plates that were dipped in a solution of ammonium (VI) molybdate tetrahydrate (12.5 g) and cerium (IV) sulfate tetrahydrate (5.0 g) in 10% aqueous sulfuric acid (500 mL). Flash column chromatography (FCC) was performed using silica gel 60 (230-400 mesh) and employed a stepwise solvent polarity gradient, correlated with TLC mobility. Solvents were purified by standard procedures or used from commercial sources as appropriate.

4.7 2,3,4,6-Tetra-O-acetyl-α/β-D-galactopyranosyl-(1→1)-2,3:4,6-di-O-isopropylidene-α-D-mannopyranoside 13α/β

A mixture of mannose donor 12^xxxii,xxxiii (4.20 g, 11.9 mmol), galactose acceptor 10^xxx (4.10 g, 11.8 mmol), freshly activated, powdered, 4Å molecular sieves, 2,6-di-tert-butyl-4-methylpyridine (DTBMP) (17.3 g, 83.3 mmol) and dry CH₂Cl₂ (50 mL) was stirred for 30 min, at rt. The mixture was then cooled to 0 °C and MeOTf (8.1 mL, 77.4 mmol) was slowly added. After stirring for an additional 20 h the reaction was quenched with triethylamine (3 mL). The mixture was diluted with CH₂Cl₂, filtered and concentrated in vacuo. FCC of the residue afforded 13α (1.59g, 23%) and 13β (1.26 g, 18%) as white amorphous solids. For 13α: R_f = 0.32 (30% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 1.41 (s, 3H), 1.42 (s, 3H), 1.52 (s, 3H), 1.57 (s, 3H), 2.03 (s, 3H), 2.06 (s, 3H), 2.14 (s, 3H), 2.17 (s, 3H), 3.50 (m, 1H), 3.74 (d, 1H, J = 8.1 Hz), 3.78 (m, 2H), 4.08 (m, 1H), 4.14 (m, 1H), 4.20 (d, 1H, J = 3.4 Hz), 4.29 (t, 1H, J = 6.7 Hz), 5.21 (dd, 1H, J = 3.7, 10.9 Hz), 5.30 (s, 1H), 5.36 (dd, 1H, J = 3.15, 10.9 Hz), 5.49 (apparent d, 2H, J = 3.5 Hz, H-1(gal), H-4(gal)); ¹³C NMR (CDCl₃) δ 19.3, 21.2 (3 signals), 26.7, 28.6, 29.6, 61.9, 62.2, 62.6, 67.6, 67.8, 67.9, 68.2, 73.0, 75.2, 76.1, 91.8 (J_{H1, C1} (gal) = 172 Hz), 94.2 (J_{H1, C1} (man) = 175 Hz), 100.3, 110.4, 170.6, 170.7, 170.8, 170.9; HRMS (ESI) calcd. for (M+Na)⁺ C₂₆H₃₈NaO₁₅ 613.2103, found 613.2110. For 13β: R_f = 0.18 (30% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 1.37 (s, 3H), 1.43 (s, 3H), 1.53 (s, 3H), 1.57 (s, 3H), 2.01 (s, 3H), 2.07 (s, 3H), 2.09 (s, 3H), 2.20 (s, 3H), 3.71 (t, 1H, J = 9.81 Hz), 3.76 (d, 1H, J = 7.9 Hz), 3.82 (m, 1H), 3.89 (m, 1H), 3.97 (t, 1H, J = 6.9 Hz), 4.11 (m, 2H), 4.20 (m, 1H), 4.70 (d, 1H, J = 8.0 Hz, H-1 (gal)), 5.21 (s, 1H), 5.24 (dd, 1H, J = 8.1, 10.5 Hz), 5.41 (d, 1H, J = 4.4); ¹³C NMR (CDCl₃) δ 18.9, 20.7, 20.8, 20.9, 21.0, 26.4, 28.3, 29.2, 61.7, 62.0, 62.3, 67.2, 69.2, 71.0, 71.4, 72.5, 75.0, 75.6, 99.8 (J_{H1, C1} (man) = 172 Hz), 99.9 (J_{H1, C1} (gal) = 154 Hz), 100.4, 109.9; HRMS (ESI) calcd. for (M+Na)⁺ C₂₆H₃₈NaO₁₅ 613.2103, found 613.2097.

4.8 2,3,4,6-Tetra-O-acetyl-α/β-D-glucopyranosyl-(1→1)-2,3:4,6-di-O-isopropylidene-α-D-mannopyranoside 14α/β

Disaccharides 14α/β were prepared from mannose donor 12 (415 mg, 1.18 mmol), and glucose acceptor 11^xxxi (373 mg, 1.07 mmol) via the procedure described for 13α/β. For 14α: (130 mg, 21%); R_f = 0.45 (40% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 1.42 (s, 3H), 1.43 (s, 3H), 1.58 (s, 6H), 2.05 (s, 3H), 2.06 (s, 3H), 3.52 (m, 1H), 3.76 (m, 3H), 4.08 (m, 1H), 4.13 (m, 2H), 4.24 (m, 2H), 4.30 (dd, 1H, J = 4.2, 4.3 Hz), 4.97 (dd, 1H, J = 3.9, 10.3 Hz), 5.13 (t, 1H, J = 9 Hz), 5.30 (s, 1H), 5.47 (d, 1H, J = 3.8 Hz, H1(glu)), 5.50 (t, 1H, J = 9.9 Hz); ¹³C NMR (CDCl₃) δ 18.9, 20.7, 20.8, 20.9, 26.4, 28.3, 29.3, 61.8, 61.9, 62.3, 68.3, 68.5, 70.0, 70.3, 72.6, 74.9, 75.7, 91.0 (J_{H1, C1} (glu) = 172 Hz), 93.9 (J_{H1, C1} (man) = 175 Hz), 100.0, 110.1, 169.7, 170.2, 170.4, 170.8; HRMS (ESI) calcd. for (M+Na)⁺ C₂₆H₃₈NaO₁₅ 613.2103, found 613.2095. For 14β: (124 mg, 20%); R_f = 0.39 (40% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 1.37 (s, 3H), 1.43 (s, 3H), 1.52 (s, 3H), 1.57 (s, 3H), 2.03 (s, 3H), 2.05 (s, 3H), 2.08 (s, 3H), 2.12 (s, 3H), 3.74 (m, 4H), 3.89 (dd, 1H, J = 5.0, 10.1 Hz), 4.12 (d, 1H, J = 5.7 Hz), 4.19 (m, 3H), 4.74 (d, 1H, J = 8.0 Hz, H-1(glu)), 5.01(d, 1H, J = 9.6 Hz), 5.05 (t, 1H, J = 9.7 Hz), 5.23 (m, 2H); ¹³C NMR (CDCl₃) δ 18.9, 20.8 (two signals), 26.4, 28.4, 29.1, 61.6, 62.4, 62.5, 68.4, 71.6, 72.4, 72.6, 72.9, 75.0, 75.6, 99.7 (J_{H1, C1} (man) = 172 Hz), 99.8 (J_{H1, C1} (glu) = 152 Hz), 99.9, 109.9, 169.5, 169.6, 170.4, 170.9; HRMS (ESI) calcd. for (M+Na)⁺ C₂₆H₃₈NaO₁₅ 613.2103, found 613.2101.

4.9 2,3,4,6-Tetra-O-benzyl-β-D-galactopyranosyl-(1→1)-2,3-O-isopropylidene-α-D-mannopyranoside 15

Tetra-O-acetate 13β (0.514 g, 0.87 mmol) was treated with NaOMe (100 mg) in dry MeOH (10 mL) under N₂ for 15 min. The mixture was then neutralized with methanolic HCl and concentrated under reduced pressure. The crude product was dried in vacuo and taken up in dry THF (15 mL). TBAI (0.032 g, 0.087 mmol) and NaH (0.348 g, 60% dispersion in mineral oil, 8.7 mmol) were added to the solution at 0 °C, under nitrogen. After 45 min at this temperature, BnBr was added dropwise to the mixture and stirring continued at rt for 15 h. The reaction was then cooled to 0 °C, quenched with MeOH, diluted with water and extracted with ether. The organic phase was dried (Na₂SO₄), filtered and concentrated in vacuo. FCC of the residue provided the tetra-O-benzylated derivative as a clear oil (0.498 g, 73%): R_f = 0.53 (30% EtOAc/petroleum ether). To a portion of this material (100 mg, 0.129 mmol), in dry MeOH (5 mL) was added p-TsOH (12 mg, 0.064 mmol) at rt. The mixture was stirred for 15 min, then neutralized with Et₃N and concentrated in vacuo. FCC of the residue afforded diol 15 (60 mg, 63%) as a white solid. R_f = 0.65 (70 % EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 1.37 (s, 3H), 1.53 (s, 3H), 3.39 (m, 1H), 3.56 (dd, 1H, J = 2.9, 6.9 Hz), 3.61 (m, 3H), 3.79 (dd, 1H, J = 3.1, 8.7 Hz), 3.83 (d, 1H, J = 2.8 Hz), 3.87 (t, 1H, J = 7.9 Hz), 4.01 (m, 1H), 4.06 (d, 1H, J = 5.7 Hz), 4.14 (m, 2H), 4.39 (d, 1H, J = 11.8 Hz), 4.48 (d, 1H, J = 11.7 Hz), 4.61 (d, 1H, J = 6.5 Hz), 4.64 (d, 1H, J = 2.6 Hz), 4.75 (d, 2H, J = 2.9 Hz), 4.80 (d, 1H, J = 11.2 Hz), 4.87 (d, 1H, J = 11.2 Hz), 4.94 (d, 1H, J = 11.7 Hz), 5.29 (s, 1H), 7.33 (m, 20 H); ¹³C NMR (CDCl₃) δ 26.2, 28.0, 62.4, 69.3, 70.1, 70.6, 73.0, 73.3, 73.5, 73.8, 74.4, 75.3, 75.4, 78.5, 79.4, 82.5, 99.0, 102.2, 109.6, 127.6 (two signals), 127.7, 127.9, 128.0, 128.1, 128.2, 128.3, 128.4 (three signals), 137.4, 138.2, 138.3, 138.5; HRMS (ESI) calcd. for (M+Na)⁺ C₄₃H₅₀NaO₁₁ 765.3245, found 765.3243.

4.10 2,3,4,6-Tetra-O-benzyl-β-D-galactopyranosyl-(1→1)-2,3-O-isopropylidene-6-O-palmitate-α-D-mannopyranoside 16

To a solution of 15 (118 mg, 0.160 mmol) and palmitic acid (45 mg, 0.176 mmol), in dry benzene (4 mL) was added DCC (36 mg, 0.176 mmol) and DMAP (5 mg, 0.04 mmol) at 0 °C. The reaction was stirred for 18 h at rt, then diluted with ether and filtered through a bed of Celite. The filtrate was concentrated in vacuo and the residue purified by FCC to give 16 (100 mg, 86% brsm): R_f = 0.4 (30% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 0.9 (t, 3H, J = 6.8, 7.0 Hz), 1.27 (bs, 25H), 1.36 (s, 3H), 1.51 (s, 3H), 1.62 (m, 2H), 2.35(t, 2H, J = 7.5 Hz), 3.15 (bs, 1H), 3.46 (m, 2H), 3.55 (m, 2H), 3.60 (t, 1H, J = 6.2 Hz), 3.85 (t, 1H, J = 7.9 Hz), 3.89 (d, 1H, J = 2.4 Hz), 4.00 (m, 1H), 4.06 (m, 1H), 4.15 (t, 1H, J = 5.9 Hz), 4.38 (d, 1H, J = 11.6 Hz), 4.43 (d, 1H, J = 11.6 Hz), 4.57 (m, 2H), 4.63 (d, 1H, J = 11.7 Hz), 4.74 (s, 1H), 4.79 (d, 1H, J = 11.2 Hz), 4.86 (d, 1H, J = 11.2 Hz), 4.95 (d, 1H, J = 11.7 Hz), 5.30 (s, 1H), 7.34 (m, 20H); ¹³C NMR (CDCl₃) δ 14.3, 22.9, 25.3, 26.4, 28.3, 29.4, 29.5 (two signals), 29.6, 29.8 (two signals), 29.9, 32.1, 34.4, 62.6, 68.5, 69.3, 69.4, 73.2, 73.4, 73.7, 74.0, 74.7, 75.4, 75.7, 77.8, 79.7, 82.8, 99.6, 102.7, 109.7, 127.7 (three signals), 128.0, 128.1, 128.2, 128.4, 128.6 (three signals), 137.8, 138.4, 138.6, 138.7, 175.4.

4.11 2,3,4,6-Tetra-O-benzyl-β-D-galactopyranosyl-(1→1)-2,3-O-isopropylidene-4,6-di-O-palmitate-α-D-mannopyranoside 17

Diol 15 (60 mg, 0.081 mmol) was treated as described in section 4.10, with excess palmitic acid (52 mg, 0.2 mmol). The diester 17 (68 mg, 70%) was obtained as a white solid: R_f = 0.82 (25 % EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 0.92 (s, 6H), δ 1.29 (broad s, 48H), δ 1.60 (s, 4H), 2.30 (s, 4H), 3.56 (m, 4H), 3.86 (dd, 1H, J = 7.8, 9.6 Hz), 3.93 (s, 1H), 4.02 (d, 2H, J = 13.7 Hz), 4.19 (d, 2H, J = 8.4 Hz), 4.28 (d, 1H, J = 12.4 Hz), 4.40 (d, 1H, J = 11.7 Hz), 4.48 (d, 1H, J = 11.7 Hz), 4.59 (d, 1H, J = 9.5 Hz), 4.64 (d, 1H, J = 13.4 Hz), 4.74 (s, 1H), 4.79 (d, 1H, J = 13.1 Hz), 4.88 (d, 1H, J = 13.1 Hz), 4.94 (d, 1H, J = 13.3 Hz), 5.18 (t, 1H, J = 10.2 Hz), 5.35 (s, 1H), 7.32 (m, 20H); ¹³C NMR (CDCl₃) δ 14.3, 22.9, 25.0, 26.8, 27.9, 29.3, 29.4, 29.5 (two signals), 29.7 (two signals), 29.8, 29.9 two signals), 32.1, 34.3, 34.4, 61.5, 67.0, 68.5, 68.8, 70.8, 73.2, 73.4, 73.7, 73.9, 74.9, 75.7 (two signals), 76.5, 79.7, 82.9, 99.3, 102.9, 110.3, 127.7, 127.9 (two signals), 128.1, 128.4, 128.6 (two signals), 137.9, 138.4, 138.6, 138.7, 172.5, 173.8; HRMS (ESI) calcd for (M+Na)⁺ C₇₅H1₁₀NaO₁₃ 1241.7389, found 1241.7389.

4.12 2,3,4,6-Tetra-O-benzyl-β-D-galactopyranosyl-(1→1)-4.6-O-benzyl-α-D-mannopyranoside 18

Diol 15 (61mg, 0.082 mmol) was subjected to the standard benzylation procedure described in section 4.9. FCC of the crude product yielded the di-O-benzylated deriative (71 mg, 0.08 mmol), which was taken up in dry MeOH (5 mL) and treated with CSA (100 mg, 0.43 mmol). The mixture was stirred for 30 min, quenched with Et₃N and concentrated in vacuo. FCC of the residue afforded 18 (55 mg, 75% two steps) as a clear oil: R_f = 0.55 (60% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 3.56 (m, 5H), 3.73 (dd, 1H, J = 3.15, 3.05 Hz), 3.81 (m, 2H), 3.88 (m, 2H), 3.97 (dd, 1H, J = 3.40, 3.50 Hz), 4.12 (m, 1H), 4.37 (s, 2H), 4.40 (s, 1H), 4.59 (m, 4H), 4.60 (d, 1H, J = 11.1 Hz), 4.73 (s, 2H), 4.78 (d, 1H, J = 11.1 Hz), 4.82 (d, 1H, J = 11.1 Hz), 4.94 (d, 1H, J = 11.7 Hz), 7.24 – 7.39 (m, 30H); ¹³C NMR (CDCl₃) δ 68.7, 68.9, 71.1, 71.5, 71.7, 73.1, 73.6 two signals, 73.8, 74.6, 74.8, 75.6, 75.8, 79.7, 82.7, 101.5, 103.5, 127.7, 127.8 (three signals), 127.9 (three signals), 128.0, 128.1, 128.4 9 (two signals), 128.5, 128.6 (three signals), 138.1, 138.4, 138.6, 138.7, 138.8.

4.13 2,3,4,6-Tetra-O-benzyl-β-D-galactopyranosyl-(1→1)-4.6-O-benzyl-3-O-palmitate-α-D-mannopyranoside 19

To a solution of 18 (80 mg, 0.09 mmol) and palmitic acid (26 mg, 0.099 mmol) in dry benzene was added DCC (20 mg, 0.099 mmol) and DMAP (2 mg, 0.018 mmol) at 0 °C. The reaction was stirred for 18 h at rt, then diluted with ether, filtered through a bed of Celite and concentrated in vacuo. FCC of the residue yielded 19 (74 mg, 80% brsm) as a white solid. R_f = 0.30 (30% EtOAc/petroleum ether). ¹H NMR (CDCl₃) δ 0.90 (t, 3H, J = 6.8 Hz), 1.27 (m, 24H), 1.61 (m, 2H), 2.28 (m, 2H), 3.54 (m, 6H), 3.71 (dd, 1H, J = 3.5, 11.5), 3.89 (m, 2H), 4.03 (m, 2H), 4.23 (m, 1H), 4.35 (m, 3H), 4.47 (d, 1H, J = 11.3 Hz), 4.56 (m, 1H), 4.62 (m, 2H), 4.73 (m, 2H), 4.83 (s, 1H), 4.95 (d, 1H, J = 12.0 Hz), 5.11 (d, 1H, J = 1.5 Hz), 5.34 (dd, 1H, J = 3.2, 9.7), 7.34 (m, 35H); ¹³C NMR (CDCl₃) δ 14.1, 22.7, 24.8, 24.9, 25.5, 29.1, 29.2, 29.3, 29.4, 29.6 (three signals), 29.7, 31.9, 33.7, 34.4, 68.2, 68.7, 69.5, 71.7, 72.7, 72.9, 73.3, 73.6, 73.9, 74.5, 75.5, 79.3, 82.3, 101.4, 103.6, 127.4 (two signals), 127.5 9 (three signals), 127.6, 127.8 (two signals), 128.0, 128.2 (two signals), 128.3 (three signals), 128.4, 137.9, 138.3, 138.4, 138.5, 138.6, 172.6; HRMS (ESI) calcd for (M+NH₄)⁺ C₇₀H₉₂NO₁₂ 1138.6614, found 1138.6606.

4.14 2,3,4,6-Tetra-O-benzyl-β-D-glucopyranosyl-(1→1)-2,3-O-isopropylidene-α-D-mannopyranoside 20

Protected disaccharide 14β was transformed to 20 (58 mg, 75% brsm) following the three step procedure described for the conversion of 13β to 15 (section 4.9) For 20: white solid; R_f = 0.18 (40% EtOAc/petroleum ether); ¹H NMR (CDCl₃) δ 1.37 (s, 3H), 1.54 (s, 3H), 2.22 (d, 1H, J = 4.4 Hz), 2.55 (dd, 1H, J = 5.7, 7.5 Hz), 3.48 (m, 2H), 3.56 (m, 2H), 3.61-3.70 (m, 4H), 3.78-3.82 (m, 1H), 3.95-3.99 (m, 1H), 4.04 (d, 1H, J = 5.8 Hz), 4.14 (m, 1H), 4.51 (m, 3H), 4.67 (d, 1H, J = 7.9 Hz), 4.83 (m, 4H), 4.92 (d, 1H, J = 10.9 Hz), 5.30 (s, 1H), 7.33 (m, 20H); ¹³C NMR (CDCl₃) δ 26.4, 28.2, 62.6, 69.2, 70.3, 71.1, 73.6, 74.9, 75.2, 75.4, 75.6, 75.9, 78.0, 78.7, 82.4, 85.0, 99.4, 102.5, 109.9, 127.9, 128.0, 128.1 (two signals), 128.2, 128.3, 128.6 (two signals), 128 .7, 137.7, 138.0, 138.4, 138.5; HRMS (ESI) calcd for (M+Na)⁺ C₄₃H₅₀NaO₁₁ 765.3245, found 765.3238.

4.15 β-D-Galactopyranosyl-(1→1)-6-O-palmitate-α-D-mannopyranoside 3

A solution of 16 (100 mg, 0.1 mmol) and p-TsOH (100 mg, 0.52 mmol) in MeOH (5 mL) was stirred for 1 h. The mixture was then quenched with saturated NaHCO₃ and extracted with CH₂Cl₂. The organic phase was dried (Na₂SO₄), filtered and concentrated in vacuo. FCC of the residue afforded the corresponding triol (70 mg, 73%). A portion of the triol (30 mg, 0.032 mmol), 10% Pd/C (40 mg) and HCOOH (0.1 mL) was stirred under H₂ atmosphere for 20 h. The mixture was then filtered through a bed of Celite and concentrated in vacuo. FCC of the residue gave 3 (9 mg, 50%): R_f = 0.36 (20% MeOH/EtOAc); ¹H NMR (CD₃OD) δ 0.89 (t, 3H, J = 6.8 Hz), 1.28 (m, 28H), 1.61 (m, 2H), 2.36 (t, 2H, J = 7.5 Hz), 3.46 (dd, 1H, J = 3.0, 9.7 Hz), 3.53 (m, 2H), 3.71 (m, 5H), 3.85 (d, 1H, J = 2.7 Hz), 3.95 (s, 1H), 4.05 (m, 1H), 4.25 (dd, 1H, J = 4.9, 11.95 Hz), 4.31 (d, 1H, J = 11.2 Hz), 4.41 (d, 1H, J = 7.7 Hz), 5.02 (s, 1H); ¹³C NMR (CD₃OD) δ 14.6, 23.9, 26.2, 30.4, 30.6, 30.8, 30.9 (two signals), 33.2, 35.1, 62.2, 64.9, 68.5, 70.1, 71.6, 72.4, 72.7, 72.8, 75.1, 77.0, 103.2, 104.4, 175.9; HRMS (ESI) calcd for (M+Na)⁺ C₂₈H₅₂NaO₁₂ 603.3351, found 603.3352.

4.16 β-D-Galactopyranosyl-(1→1)-3-O-palmitate-α-D-mannopyranoside 4

The hexa-O-benzylether 19 (54 mg, 0.048 mmol) was subjected to the hydrogenolysis procedure described for 3. FCC of the crude product provided 4 (22 mg, 79%): R_f = 0.58 (20% MeOH/CH₂Cl₂); ¹H NMR (CD₃OD) δ 0.90 (t, 3H, J = 6.8 Hz), 1.16 (m, 5H), 1.29 (bs, 26H), 1.63 (m, 4H), 1.71 (m, 2H), 1.85 (m, 2H), 2.41 (t, 2H, J = 7.45 Hz), 3.46 (m, 2H), 3.56 (m, 1H), 3.65 (m, 3H), 3.78 (m, 3H), 3.88 (d, 1H, J = 10.2 Hz), 4.10 (m, 2H), 4.44 (d, 1H, J = 7.85 Hz), 5.04 (dd, 1H, J = 3.2, 9.8 Hz), 5.06 (s, 1H); ¹³C NMR (CD₃OD) δ 14.6, 23.9, 26.1, 26.2, 26.9, 30.3, 30.6, 30.8, 30.9 (two signals), 33.2, 34.9, 35.2, 63.0, 63.1, 66.3, 69.7, 70.5, 72.5, 75.0, 75.3, 75.6, 77.4, 102.9, 104.7, 175.5; HRMS (ESI) calcd for C₁₂H₂₂NaO₁₁ (M+Na)⁺ 603.3351, found 603.3351.

4.17 β-D-Galactopyranosyl-(1→1)-4,6-di-O-palmitate-α-D-mannopyranoside 5

The diester 17 (72 mg, 0.06 mmol) was dissolved in a mixture of dry MeOH (3 mL) and dry CH₂Cl₂ (1 mL). CSA (25 mg, 0.11 mmol) was then added. The reaction mixture was stirred for 1 h, then neutralized with Et₃N and evaporated under reduced pressure. FCC of the residue gave the derived diol (32 mg, 46%): white solid. R_f = 0.54 (35% EtOAc/petroleum ether). This material (32 mg, 0.027 mmol) was subjected to the hydrogenolysis procedure described for 3. FCC of the product afforded 5 (13 mg, 60%): white solid; R_f = 0.32 (10% MeOH/EtOAc); ¹H NMR (CD₃OD) δ 0.90 (t, 6H, J = 6.8 Hz), 1.31 (s, 48H), 1.61 (m, 4H), 2.35 (m, 4H), 3.48 (dd, 1H, J = 3.3, 9.7 Hz), 3.54 (m, 2H), 3.72 (d, 2H, J = 6.3 Hz), δ 3.85 (d, 1H, J = 3.3 Hz), 3.89 (dd, 1H, J = 3.3, 9.8 Hz), 4.01 (m, 1H), 4.05 (m, 1H), 4.21 (m, 1H), 4.43 (d, 1H, J = 7.7 Hz), 5.07 (s, 1H), 5.24 (t, 1H, J = 9.8 Hz); ¹³C NMR δ14.6, 23.9, 26.0, 26.1, 30.3, 30.4, 30.6, 30.7, 30.8 (two signals), 30.9, 31.0, 33.2, 35.0, 35.2, 62.4, 63.8, 70.1, 70.2, 70.6, 71.7, 72.7, 75.1, 77.1, 103.3, 104.7, 174.9, 175.5; HRMS (ESI) calcd for (M+Na)⁺ C₄₄H₈₂NaO₁₃ 841.5641, found 841.5648.

4.18 β-D-Glucopyranosyl-(1→1)-6-O-palmitate-α-D-mannopyranoside 6

Diol 20 was transformed to 6 (9 mg, 28% brsm) following the three step procedure described for the conversion of 15 to 3 (sections 4.9, 4.10, 4.15). For 6: R_f = 0.37 (20% MeOH/EtOAc); ¹HNMR (CD₃OD) δ 0.92 (t, 3H, J = 6.8 Hz), 1.20 (t, 3H, J = 7.0 Hz), 1.31 (bs, 24H), 1.64 (m, 2H), 2.23 (t, 2H, J = 7.4 Hz), 3.23 (t, 1H, J = 8.0 Hz), 3.33 (m, 1H), 3.50 (q, 2H, J = 7.0 Hz), 3.70 (m, 3H), 3.87 (dd, 1H, J = 2.2, 11.9 Hz), 3.98 (m, 1H), 4.10 (m, 1H), 4.29 (dd, 1H, J = 5.2, 11.9 Hz), 4.36 (dd, 1H, J = 2.2, 12.0 Hz), 4.48 (d, 1H, J = 8.0 Hz), 5.03 (s, 1H); ¹³C NMR (CD₃OD) δ 12.1, 13.6, 21.9, 24.2, 28.4, 28.6, 28.8, 28.9, 31.2, 33.1, 60.7, 62.9, 65.0, 66.5, 69.3, 60.6, 70.4, 70.8, 73.3, 76.2, 76.4, 101.1, 101.7, 173.9; HRMS (ESI) calcd for C₂₈H₅₂NaO₁₂ (M+Na)⁺ 603.3351, found 603.3344.

4.19 Methyl-6-O-palmitate-α-D-mannopyranoside 7

Compound 7 was prepared from 21^xxxv following procedures used in the conversion of 13β to 3. For 7: R_f = 0.60 (EtOAc); ¹H NMR (CDCl₃) δ 0.88 (t, 3H, J = 6.9 Hz), 1.26 (bs, 24H), 1.65 (m, 2H), 2.28 (bs, 1H), 2.41 (t, 2H, J = 7.4 Hz), 3.39 (s, 3H), 3.56 (t, 1H, J = 9.6 Hz), 3.69 (m, 1H), 3.83 (dd, 1H, J = 3.3, 9.3 Hz), 3.97 (s, 1H), 4.21 (dd, 1H, J = 2.0, 12.4 Hz), 4.66 (dd, 1H, J = 3.8, 12.4 Hz), 4.77 (s, 1H); ¹³C NMR (CDCl₃) δ 14.3, 22.9, 25.1, 29.3, 29.4, 29.5, 29.6, 29.8 (two signals), 29.9, 32.1, 34.4, 55.3, 63.3, 67.7, 70.4, 70.5, 71.4, 101.0, 175.5; HRMS (ESI) calcd. for (M+Na)⁺ C₂₃H₄₄NaO₇ 455.2979, found 455.2986.

4.20 Methyl-4,6-di-O-palmitate-α-D-mannopyranoside 8

Compound 8 was prepared from 21 following procedures used in the conversion of 13β to 5. For 8: R_f = 0.6 (80% EtOAc/petroleum ether); ¹H NMR δ 0.90 (t, 6H, J = 6.8 Hz), 1.27 (bs, 48H), 1.64 (m, 4H), 2.36 (m, 4H), 2.72 (bs, 1H), 3.12 (bs, 1H), 3.41 (s, 3H), 3.90 (m, 2H), 3.96 (s, 1H), 4.19 (dd, 1H, J = 2.2, 12.1 Hz), 4.28 (dd, 1H, J = 5.5, 12.1 Hz), 4.81 (s, 1H), 5.02 (t, 1H, J = 9.8 Hz). ¹³C NMR δ 14.3, 22.9, 25.0 (two signals), 29.2, 29.3, 29.4, 29.5 (two signals), 29.6, 29.7, 29.8 (three signals), 29.9, 32.1, 34.3, 34.5, 55.3, 62.7, 68.0, 70.1, 70.6, 70.7, 100.7, 173.8, 175.0; HRMS (ESI) calcd for C₃₉H₇₄NaO₈ (M + Na)⁺ 693.5276, found 693.5278.

4.21 α-D-Galactopyranosyl-(1→1)-6-O-palmitate-α-D-mannopyranoside 9

Disaccharide 13α was transformed to 9 (60 mg, 13.2% bsrm) following the procedures described for the conversion of 13β to 3. For 9: R_f = 0.2 (20% MeOH/EtOAc); ¹H NMR (CD₃OD) δ 0.81 (t, 3H, J = 6.9 Hz), 1.21 (bs, 23H), 1.52 (m, 2H), 2.26 (t, 2H, J = 7.5 Hz), 3.21 (m, 1H), 3.61 (m, 4H), 3.74 (m, 4H), 3.82 (d, 1H, J = 2.5 Hz), 3.87 (m, 1H), 4.13 (dd, 1H, J = 5.9, 11.8 Hz), 4.29 (bd, 1H, J = 11.7 Hz), 4.99 (s, 1H), 5.03 (d, 1H, J = 3.7 Hz); ¹³C NMR (CD₃OD) δ 14.5, 23.8, 26.1, 30.4, 30.6 (two signals), 30.7, 30.9 (two signals), 33.2, 35.1, 62.8, 64.8, 68.7, 69.7, 71.1, 71.3, 72.2, 72.5, 73.3, 95.5, 96.7, 175.7; HRMS (ESI) calcd for C₂₈H₅₆NO₁₂ (M+NH₄)⁺, 589.3799 found 589.3797.