11.74,111.75,111.76, and 111.77 Scheme 33 The tetrasaccharide synthesis.
regioisomer (15%). No acyl migration was observed when the pivaloyl ester was substituted for the corresponding acetate. Sulfation of the 4-OH, saponification, and hydrogenation furnished III.75 in 70% yield. When the regioselective sulfation of the tetraol III.80 was attempted for the chondroitin 6-sulfate tetrasaccharide, it proceeded slowly, yielding a single product after 2 days that was identified as the tetrasulfated derivative III.77. Therefore, the 6-O-sulfated derivative was obtained stepwise by acetylation of III.79 followed by hydrogenation of the two 6-O-benzyl ethers to afford the corresponding diol. Conventional sulfation and saponification afforded III.76 in 82% yield over three steps.
Chondroitin differs from hyaluronan in two ways: the presence of sulfate esters and the configuration of the amino sugar (i.e., d-galactosamine instead of d-glucos-amine). Since d-galactosamine is a rare and expensive starting material, it is usually prepared by the azidonitration of d-galactal as demonstrated in the earlier preparations of chondroitin and dermatan fragments. An alternate way of accessing d-ga-lactosamine is by inversion of C4 in d-glucosamine, and indeed such conversions have been reported for both monomers  and neutral disaccharides  containing d-glucosamine. Coutant and Jacquinet used this approach to access saccharides con taining uronic acid moieties and reported the preparation of chondroitin 4-O-sulfate trisaccharide III.81 from the hyaluronic acid trisaccharide III.82 (Scheme 34) .
A trisaccharide derivative, in which the central d-glucosamine residue is flanked by two d-glucuronic acid residues, was chosen to test the validity of the strategy. Construction of the protected hyaluronic acid trisaccharide was achieved from the following monosaccharides: III.83, III.84, and III.85. The glucuronic acid precursor III.85 was glycosylated with glucosamine moiety III.84 by using N-io-dosuccinamide (NIS) and trimethylsilyl triflate in dichloromethane to afford the corresponding disaccharide III.86 in 90% yield. In general, uronic esters are poor nu-cleophiles when glycosylation is at the C4 position. Presumably the ester moiety significantly reduces the nucleophilicity of the 4-OH, and as a result, necessitates the use of a d-glucose unit, where the C6 is selectively oxidized after coupling.
Conversion of disaccharide III.86 into glycosyl acceptor III.87 was achieved in 92% yield by removal of the chloroacetyl ester with thiourea in pyridine-ethanol. Condensation of acceptor III.87 with imidate III.83 in the presence of trimethylsilyl triflate afforded the crystalline trisaccharide III.88 in 92% yield. Removal of the isopropylidene with aqueous acid followed by selective benzoylation of C6 gave III.82 (90%, two steps).
Inversion of the configuration at C4 was carried out by treatment of III.82 with triflic anhydride in pyridine to form the 4-O-triflyl derivative followed by reaction with tetrabutylammonium nitrite, a reagent known to give the epi-hydroxyl analog , to afford the d-galacto product III.89 in 87% yield (Scheme 35). Transformation of the N-trichloroacetyl group to the acetamide was carried out with tribu-tylstannane and azoisobutyronitrile  to give the crystalline acetamide III.90 in 92% yield. Sulfation of the free hydroxyl with the sulfur trioxide-trimethylamine complex gave 93% of III.91, which was saponified with sodium hydroxide in aqueous methanol to afford the target chondroitin 4-O-sulfate trisaccharide III.81 in 87% yield.
While the syntheses of derivatives of the chondroitin sulfate disaccharides dominate the existing literature, Jacquinet and coworkers have reported the successful syntheses of the reducing disaccharides of chondroitin 4- and 6-sulfates on a multigram scale . The synthetic sequence (Scheme 36) utilizes a silver triflate mediated glycosylation between bromide III.92 and benzyl glycoside III.93 to afford the disaccharide III.94 in 70% yield. After treatment with hot aqueous acetic acid, diol III.95 was obtained in 87% yield. The strategic choice of functionality in this common intermediate minimizes the number of postglycosylation transformations. Regioselective benzoylation of diol III.95 was achieved in 93% yield by treatment with benzoyl cyanide in pyridine. Consequent O-sulfation with the sulfur trioxide-trimethylamine complex followed by ion exchange chromatography afforded the sodium salt III.96 in 90% yield. Saponification of III.96 with lithium hydroperoxide and methanolic sodium hydroxide provided the disodium salt III.97 in 83% yield, which was subsequently subjected to hydrogenolysis to afford the chondroitin 4-sulfate disaccharide III.98 in 97% yield. Preparation of chondroitin 6-sulfate from the common intermediate III.95 was achieved by regioselective sulfation at C6 with the sulfer trioxide-trimethylamine complex. The resulting monosulfated disaccharide III.99 was isolated in 90% yield. Saponification to the disodium salt III.100 occurred in 82% yield, and subsequent hydrogenolysis provided the chondroitin 6-sulfate di-saccharide III.101 in 96% yield.
Scheme 34 The use of C4 inversion in the preparation of the chondroitin sulfate trimer.
Was this article helpful?