Epoxidation and Bis-hydroxylation of C-Phenyl-Δ2,3-glyco-pyranosides.

Epoxidation and cis-hydroxylation of C-phenyl-Δ2,3-glycopyranosides have been carried out with a view to developing C-aryl glycoside synthesis. Epoxidation of (2,3- dideoxy-d-erythro-hex-2-enopyranosyl)benzene and (6-O-tert-butyldimethylsilyl-2,3- dideoxy-d-erythro-hex-2-enopyranosyl)benzene gave predominantly the allo-adducts whatever the configuration at the anomeric center. Epoxidation of (4,6-di-O-tert-butyl-dimethylsilyl-2,3-dideoxy-d-erythro-hex-2-enopyranosyl)benzene gave the manno- and allo-adducts in a 89:11 and 40:60 ratios for the α- and β-anomers, respectively. Hydroxylation of α-C-phenyl-Δ2,3-glycopyranosides using OsO4 afforded the manno-adduct only, whatever the substituents at positions 4 and 6, whereas hydroxylation of (2,3-dideoxy- β-d-erythro-hex-2-enopyranosyl)benzene and (4,6-di-O-tert-butyldimethylsilyl-2,3- dideoxy- β-d-erythro-hex-2-enopyranosyl)benzene gave the manno- and allo-adducts in 25:75 and 80:20 ratios, respectively.


Introduction
There is a great interest in C-aryl glycosides. This is due to their occurrence in many natural products possessing important medicinal and therapeutic properties [1][2][3], as well as to their use as valuable chiral building blocks [4][5][6]. However, it is important to obtain the two anomers of these C-aryl glycopyranosides with the highest stereoselectivity. Several synthetic methods are now available for the stereoselective and even stereospecific preparation of C-aryl glycopyranosides possessing a double bond in the 2,3-position [7][8][9][10][11][12][13][14]. It is to be noted that these unsaturated glycosides are useful precursors of the corresponding saturated C-aryl glycosides by simple functionalization of this unsaturation. Whereas epoxidation or hydroxylation of 2,3-dideoxy-hex-2-pyranosides has been well studied [15], there have not been any such systematic studies in the case of C-glycopyranosides, although some examples of cis-hydroxylation of C-aryl glycopyranosides appeared in the literature [14,16]. We report in this paper some stereochemical aspects concerning the epoxidation and cishydroxylation of both anomers of some C-phenyl-∆ 2,3 -glycopyranosides (Scheme 1).

Results and Discussion
The different C-phenyl-∆ 2,3 -glycopyranosides having the αand the β-configuration used in this study are shown in Scheme 1. The synthesis of bis-silylated 2,3-unsaturated-C-phenyl glycopyranosides 1a and 2a has already been described [12]. The deprotected unsaturated C-phenyl glycopyranosides 1b and 2b were obtained by a simple desilylation of compounds 1a and 2a, whereas monosilylation of 1b and 2b with tert-butyldimethylsilyl chloride afforded compounds 3a and 3b, respectively.
We studied first the epoxidation of these substrates using as the epoxidation reagent m-chloroperbenzoic acid in CHCl 3 at 50 °C for 24 h. Epoxidation of bis-silylated pseudo-glucal 1a possessing the α-configuration gave in 79% yield a 11:89 mixture of the α−alloand α−manno-epoxides 4a and 5a, which were separated by column chromatography, whereas the non-protectected pseudo-glucal 1b gave only the α−allo isomer 4b, albeit in a quite low yield (20%) (Scheme 2); even prolonged reaction times did not improve this yield, degradation products being observed in these cases. Epoxidation of the monosilylated pseudo-glucal 3a (Scheme 2) gave the reverse selectivity to that observed for the bis-silylated compound 1a; a 86:14 mixture of the α−alloand α-manno-epoxides 6 and 7 was now obtained in 70% yield, the α-allo-epoxide being now predominant. These results are quite different from those observed in the epoxidation of alkyl 2,3-dideoxy-hex-2-pyranosides [15].
The allo or manno configuration of those epoxides was established through comparison of the coupling constants J 1,2 and J 3,4 [15,17,18], these two values being always higher for the α-allo than for the α-manno derivative. Effectively no H-1/H-2 or H-3/H-4 couplings were observed for compounds 5a and 7, whilst compounds 4a and 6 showed J 1,2 = 3.6 and 3. Hz, respectively. The observed stereoselectivities could be rationalized by assuming that the epoxidation of compound 1a is under steric control, the peracid attacking on the β face because of the increased steric requirements on α face imparted by the two substituents at C-1 and C-4, whereas the epoxidation of compounds 1b and 3a was reversed, due to the cis-directing influence of the allylic hydroxyl group at position 4. The epoxidation was then extended to the unsaturated β-phenyl glycopyranosides (Scheme 3). The non-protected unsaturated glycoside 2b and the monosilylated compound 3b gave the unique β-alloepoxides 8b and 10 in 86 and 50% yield, respectively. This stereospecific epoxidation could again be attributed to the cis-directing effect of the hydroxyl function at position 4. Conversely, epoxidation of the bis-silylated β-anomer 2a gave a 60:40 mixture of the β-allo and β-manno-epoxides 8a and 9a in 69% yield; this lack of stereoselectivity was probably due to similar crowding of the two faces of the double bond of this compound.

Scheme
The allo and manno-configurations were assigned from the NMR spectra; whereas H-1 appears as a singlet for all the compounds, the value of J 3,4 is characteristic, this value being 0 for the β-manno configuration and varying from 1.5 to 2. The cis-hydroxylation was then examined using osmium tetroxide and N-methylmorpholine oxide as the re-oxidant. The bis-silylated and bis-hydroxy pseudo-glucals 1a and 1b afforded exclusively α-D-phenyl-mannopyranosides 11 and 12 in 70 and 43% chemical yield, respectively, after acetylation in the last case (Scheme 4). This very high stereoselectivity could be explained, as for alkyl 2,3-dideoxyα-D-hex-enopyranosides, by the approach of the reactant on the less sterically crowded face of the Cglycoside [14][15][16]19]. Application of the cis-hydroxylation process to the β-anomers 2a and 2b gave a mixture of Cphenyl β-mannoand allo-pyranosides (Scheme 5). The bis-O-silylated compound 2a gave in 70% yield a 80:20 ratio of β-manno-pyranoside 13 and β-allo-pyranoside 14, which could not be separated, while unprotected 2b gave in 75% yield after acetylation a 25:75 ratio of β-manno-pyranoside 15 and β-allo-pyranoside 16, which were not separated. This difference in stereoselectivity could be explained by the presence of the crowded Me 2 Bu t SiO group for 2a versus the OH group for 2b.
Configuration assignments for the compounds obtained by bis-hydroxylation were made on the basis of simple 1 H-NMR analyses and by comparison with previously described compounds.

Conclusions
We have shown that epoxidation and cis-hydroxylation of both anomers of C-phenyl-∆ 2,3 -glycopyranosides are highly selective, the selectivity depending mostly on the substituent at position 4. Epoxidation of C-phenyl-∆ 2,3 -glycopyranosides having a free hydroxyl group at position 4 afforded predominantly, if not only, the allo-epoxide, whatever the anomer used. When the hydroxyl function at position 4 was protected as a tert-butyldimethylsilyl ether, the α-anomer gave predominantly the manno-epoxide, when the β-anomer afforded a 60:40 mixture of the two-adducts. Cis-dihydroxylation of C-phenyl-∆ 2,3 -α-glycopyranosides afforded the manno-adduct as the unique compound. For the cisdihydroxylation of C-phenyl-∆ 2,3 -β-glycopyranosides, the presence of the free hydroxyl group at C-4 caused formation of the allo-adduct as the major isomer, when the allo-adduct was obtained predominantly when the hydroxyl function was protected with a tert-butyldimethylsilyl group

General
Solvents were purified by standard methods and dried if necessary. Melting points (uncorrected) were determined with a capillary melting point apparatus Büchi SMP-20. Optical rotations were recorded using a Perkin-Elmer 241 polarimeter. Thin-layer chromatography was performed using Merck silica gel 60 F 254 precoated aluminium plates, 0.2 mm thickness. Column chromatography was performed on silica gel (Merck 60, 70-230 mesh). NMR spectra were recorded on a Bruker 300 MHz spectrometer (operating at 300.13 MHz for 1 H, and 75.01 MHz for 13 C).

Preparation of compounds 1a,b.
A solution of the unsaturated bis-silylated C-phenyl glycopyranoside 1a (or 2a) (3 g, 7.7 mmol) [12], and NBu 4 F.3H 2 O (2.43 g, 7.7 mmol) in THF (50 mL) was stirred at rt for 2 h. After evaporation of the solvent, CH 2 Cl 2 was added (50 mL), and the solution was washed with brine. Evaporation of the solvent under reduced pressure gave a residue that was purified by column chromatography on silica using petroleum ether/ethyl acetate as the eluent to gave compound 1b (or 2b).

Preparation of compounds 3.
To a solution of the unsaturated C-phenyl glycopyranoside 1b (or 2b) (1 g, 4.8 mmol), imidazole (32 mg, 05 mmol) and triethylamine (1 mL, 6.7 mmol) in CH 2 Cl 2 (8 mL) maintained at rt was added a solution of tert-BuMe 2 SiCl (990 mg, 6.6 mmol) in CH 2 Cl 2 (10 mL). After being stirred at rt for 24 h, the solution was poured into cold water (10 mL), and the mixture was extracted with CH 2 Cl 2 . (3x10 mL). After evaporation of the solvent under reduced pressure, the residue was purified by column chromatography on silica using petroleum ether/diethyl ether as the eluent to gave compound 3a (or 3b).  General procedure for the epoxidation.

(6-O-tert-Butyldimethylsilyl-2,3-dideoxy-α-D-erythro-hex-2-enopyranosyl)benzene
A solution of the unsaturated carbohydrate (0.23 mmol) and m-CPBA (1.17 g, 0.69 mmol) in CHCl 3 (10 mL) was stirred at 50 °C for 24 h. The solution was neutralized with a saturated aqueous solution of NaHCO 3 , the organic phase was separated, and the aqueous phase was extracted with CHCl 3 (2x10 mL). The organic phases were dried over Na 2 SO 4 . After evaporation of the solvent under reduced pressure, the residue was purified by column chromatography on silica using the appropriate eluent.