O RO O 1 R = H 2 R = SO3H O HO HO CO2H OH O O O 3 O RCO2 RCO2 CO2Me X RCO2 O AcO AcO CO2Me AcO O RO RO CO2Me RO R O O O 4 R = Me, X = Br 6 R = Me, X = I 8 R = But, X = Br 7 R = b-OAc 10 R = a/b-OH 11a R = a-OC 11b R = b-OC ( NH)CCl3 ( NH)CCl3 5 R = Ac 9 R = ButCO O HO HO NaO2C OH O OH CO2Na 12 370 J. CHEM. RESEARCH (S), 1997 J. Chem. Research (S), 1997, 370–371† Intermediates for Glucuronide Synthesis: 7-Hydroxycoumarin Glucuronide† Richard T. Brown,a Feodor Scheinmannb and Andrew V.Stachulski*b aDepartment of Chemistry, The University of Manchester, Oxford Rd., Manchester M13 9PL, UK bSalford Ultrafine Chemicals and Research Ltd., Synergy House, Guildhall Close, Manchester Science Park, Manchester M15 6SY, UK A convenient synthesis of the important metabolite 7-hydroxycoumarin glucuronide 3 is presented, including a first report of the b-imidate 11b, together with new preparations of the iodosugar 6 and a-imidate 11a: stability data on 11a and 11b are also given, and the importance of carefully controlled hydrolysis in the last step of the preparation of 3 is emphasised. Coumarin is nowadays a ubiquitous additive to many foods and cosmetic products:1 it is therefore important that its in vivo metabolic products should be available as analytical standards.The major phase 1 metabolic process in humans is cytochrome P-450 mediated hydroxylation to 7-hydroxycoumarin 1; subsequent conjugation steps lead to the finally eliminated products, namely the sulfate 2 and glucuronide 3.Indeed, the production of 3 is regarded as a valuable test of the metabolic viability of liver slices.2 We are aware of only one literature synthesis of 3, namely phase-transfer-catalysed reaction of 1 with the bromosugar 4 followed by removal of protecting groups (6.5% overall).2 We now describe a greatly improved synthesis of 3 using imidate methodology3 together with new chemistry of the glucuronate intermediates, especially a first report of a b-imidate in the acyl glucuronate series.Under a variety of conditions,2,4,5 reaction of 4 with 1 gave no better than 10% of the conjugate 5. The iodosugar 6, conveniently prepared in fair yield directly from the b-tetraester 74 by treatment with BF3 .Et2O and KI in acetonitrile at reflux, rather than by halogen exchange on 4,6 gave a faster reaction but no better yield. Undoubtedly the low stabilities of 4 and 6 under the reaction conditions are a major problem, and indeed the pivaloyl bromosugar 8 described by Snatzke7 gave a 30% yield of the corresponding conjugate 9 using the lithium phenolate method.5 However, it then proved impossible to remove the protecting groups without irreversibly opening the lactone.Finally, trimethylsilyl trifluoromethanesulfonate- mediated reaction8 of 7 with 1 gave no reaction in dichloromethane and only 7-acetoxycoumarin in acetonitrile. Such ‘acyl transfer’ is thought to arise from an orthoester intermediate9 and in our experience is rather facile with acyl glucuronates in MeCN, making this a poor glucuronidation solvent.However, the imidate method worked very well. The required hemiacetal 10 has been obtained in various ways10–12 from the bromosugar 4 or tetraester 7; condensation of 10 (a 5:1 a:b mixture) with 7-hydroxycoumarin under Mitsunobu conditions13 gave a modest (20%) yield of 5. This is generally a viable procedure for relatively acidic phenols.It was known in the glucose series that base-catalysed reaction of hemiacetals like 10 with trichloroacetonitrile could be tuned to give either the a- or b-imidate through the operation of what Schmidt has termed the ‘kinetic’ anomeric effect.14,15 Previously only the a-anomer 11a was known in the acyl glucuronate series:16 here, using K2CO3 (rather than NaH as previously) in CH2Cl2, it was obtained in 81% recrystallised yield. Using a still weaker base, however, significant amounts of 11b were detected (TLC, NMR); brief treatment with NaHCO3 in MeCN gave 11b as the major product, and it could be isolated in 30% yield free of 11a by virtue of its lower solubility.Concern expressed over the likely stabilityof imidates such as 11a appears exaggerated.16a Crystalline 11a is stable for months at 20 °C under desiccation and shows no significant degradation over three weeks in NMR-grade CDCl3 solution. Anomer 11b is much less stable, and after a few days in CDCl3 is largely decomposed, giving mainly 10 with other more complex products.Using BF3 .Et2O catalysis in CH2Cl2, 11a and freshly prepared 11b couple equally well to 7-hydroxycoumarin: 5 has been thus obtained in 61% yield. Neither ZnBr2 nor MgBr2, which are effective catalysts in similar glucose examples,17 work here. Finally, we note that the hydrolysis of 5 to give 3 requires great care, for even a slight excess of base leads to irreversible lactone opening, giving 12 (NMR shows J 13.5 Hz, consistent with an E-olefin).It is safest to treat 5 with Na2CO3 in aqueous MeOH at 20 °C for 3 h, followed by acidification to pH 6; highly pure 3 can then be isolated at its sodium salt in 70% yield. This procedure also lessens the risk of generating any 4,5-didehydroglucuronide, a known elimination by-product from such hydrolyses.18 *To receive any correspondence (e-mail: info@ultrfine.u-net.com). †This is a Short Paper as defined in the Instructions for Authors, Section 5.0 [see J.Chem. Research (S), 1997, Issue 1]; there is therefore no corresponding material in J. Chem. Research (M).J. CHEM. RESEARCH (S), 1997 371 Experimental For general procedures, see the previous paper from these laboratories.19 Methyl 1-Deoxy-1-iodo-2,3,4-tri-O-acetyl-b-D-glucopyranuronate 6.·Methyl 1,2,3,4-tetra-O-acetyl-b-D-glucopyranuronate 74 (0.38 g, 1 mmol) was stirred and heated under argon with potassium iodide (0.33 g, 2 mmol) and boron trifluoride–diethyl ether (0.23 cm3) in anhydrous acetonitrile (5 cm3) at gentle reflux for 2 h.The cooled mixture was poured onto ice-cold satd. aq. NaHCO3 and extracted with ethyl acetate (2Å), then the combined organic phases were washed with 5% aq. sodium thiosulfate, satd. aq. NaHCO3 and water. Evaporation gave an orange gum (0.335 g) which was chromatographed, eluting with 10% ethyl acetate–hexane, then 20% and finally 50%: appropriate fractions were pooled and evaporated to give the iodosugar 6 (0.145 g, 32%) as a white solid, mp 108–110 °C (decomp.) (Found: C, 35.2; H, 3.9.C13H17IO9 requires C, 35.1; H, 3.8%); dH 2.10, 2.11, 2.14 (9 H, 3 s, 3ÅCH3CO), 3.81 (3 H, s, CH3O), 4.28 (1 H, dd, 2-H), 4.42 (1 H, d, 5-H), 5.32, 5.57 (2 H, 2 d, 3-H+4-H) and 7.07 (1 H, d, J 4, 1-H); m/z (CI; NH3) 462 (MNH4 +). Methyl [1-(2-Oxo-2H-benzopyran-7-yl)-2,3,4-tri-O-pivaloyl-b-Dglucopyran] uronate 9.·Methyl 1-bromo-1-deoxy-2,3,4-tri-O-pivaloyl- b-D-glucopyranuronate7 8 (0.45 g, 0.86 mmol) was added to a stirred solution of 7-hydroxycoumarin 1 (0.16 g, 1 mmol) in methanol (2 cm3) containing lithium hydroxide hydrate (0.042 g, 1 mmol).Tetra-n-butylammonium bromide (0.050 g) was added; after 7 days at 20 °C, TLC (EtOAc–hexane, 1:1) showed appreciable conversion into a less polar spot, and treatment with sodium iodide and/or heating to 50 °C had little further effect. The mixture was diluted with ethyl acetate, washed with 10% aq.sodium carbonate (2Å), 10% aq. sodium thiosulfate and water and evaporated to a residue (0.370 g) which was chromatographed, eluting with 25% ethyl acetate –hexane, then 33%. Evaporation of appropriate fractions gave the ester 9 (0.154 g, 30%), mp 187–189 °C (Found: C, 61.8; H, 6.8. C31H40O12 requires C, 61.6; H, 6.6%); dH 1.15–1.25 (27 H, 2 s, 3ÅMe3C), 3.79 (3 H, s, CH3O), 4.31 (1 H, d, 5p-H), 5.25 (1 H, d, 1p-H), 5.35–5.55 (3 H, m, 2p-H+3p-H+4p-H), 6.36 (1 H, d, 3-H), 6.90–7.00 (2 H, m, 5-H+6-H), 7.45 (1 H, d, 8-H) and 7.70 (1 H, d, 4-H); m/z (CI; NH3) 622 (MNH4 +). Methyl 2,3,4-Tri-O-acetyl-1-O-(trichloroacetimidoyl)-a-D-glucopyranuronate 11a.·A solution of methyl 1-hydroxy-2,3,4-tri- O-acetyl-a,b-D-glucopyranuronate 1010–12 (1.67 g, 5 mmol) in anhydrous dichloromethane (25 cm3) and trichloroacetonitrile (3.6 cm3) was stirred at 20 °C with potassium carbonate (3.8 g).After 16 h the reaction mixture was filtered through a silica pad, eluting with diethyl ether, then 1:1 ethyl acetate–hexane.Appropriate fractions were pooled and evaporated, then the residue was recrystallised from ethyl acetate–hexane to give the imidate 11a (1.93 g, 81% in two crops), mp 109–110 °C (lit.,16a 108 °C). Methyl 2,3,4-Tri-O-acetyl-1-O-(trichloroacetimidoyl)-b-D-glucopyranuronate 11b.·A solution of the 1-hydroxysugar 10 (0.84 g, 2.5 mmol) in HPLC-grade acetonitrile (10 cm3) and trichloroacetonitrile (2.50 cm3) was stirred with sodium hydrogen carbonate (0.42 g, 5 mmol) over 4 Å molecular sieves at 20 °C. After 3 h the mixture was stored at 0 °C for 1 h, then diluted with diethyl ether and washed with water.The aqueous phase was again extracted with diethyl ether, then the combined extracts were washed with water and evaporated to a yellow gum which was triturated with diethyl ether–hexane; the resulting white crystals were isolated and recrystallised from ethyl acetate–hexane to give the imidate 11b (0.36 g, 30% in two crops), mp 153–154 °C (Found: C, 37.9; H, 3.8; N, 2.8.C10H15Cl3NO10 requires C, 37.6; H, 3.8; N, 2.9%); dH 2.08, 2.10 (9 H, 2 s, 3ÅCH3CO), 3.80 (3 H, s, CH3O), 4.31 (1 H, m, 5-H), 5.35–5.45 (3 H, m, 2-H+3-H+4-H), 5.98 (1 H, d, J 7.5, 1-H) and 8.82 (1 H, br s, NH). The spectrum was complicated by virtual coupling, but irradiation at d 5.4 collapsed the 5-H to a singlet. Methyl [1-(2-Oxo-2H-1-benzopyran-7-yl)-2,3,4-tri-O-acetyl-b-Dglucopyran] uronate 5.·Boron trifluoride–diethyl ether (0.030 cm3) was added at µ15 °C to a stirred suspension of 7-hydroxycoumarin 1 (0.16 g, 1 mmol) and imidate 11a (0.48 g, 1 mmol) in dichloromethane (5 cm3) which had previously been stirred under argon at 20 °C for 2 h over freshly baked 4 Å molecular sieves. After 1.25 h, when the temperature had been allowed to rise to +5 °C, the mixture was re-cooled to µ10 °C and further catalyst (0.030 ml) was added, stirring being then continued for 16 h while the mixture regained ambient temperature. Ethyl acetate (25 cm3) was added, then the solution was washed with satd. aq.NaHCO3 (3Å), water and evaporated. Chromatography on silica, eluting with from 30 to 70% ethyl acetate in hexane, afforded on evaporation of appropriate fractions a semi-solid which on crystallisation from methanol gave in two crops the ester 5 (0.292 g, 61%), mp 192–193 °C (lit.,2 185.5–187.5 °C). 1-(2-Oxo-2H-1-benzopyran-7-yl)-b-D-glucopyranosiduronic Acid 3.·A suspension of ester 5 (0.400 g, 0.84 mmol) in methanol (12 cm3) was stirred at 20 °C with a solution of anhydrous Na2CO3 (0.238 g, 2.24 mmol) in water (5 cm3).After 5 h water (6 cm3) was added, followed by addition of glacial AcOH to pH 6.2, then the product was precipitated by addition of ethanol (25 cm3). The resultant solid was filtered off and recrystallised from aq. EtOH to afford the glucuronide 3 as its sodium salt (0.21 g, 70%), mp 295–296 °C (decomp.) [lit.,2 282–287 °C (decomp.)].We are most grateful to Drs S. Mayalarp and K. W. Lumbard for skilled experimental assistance. Received, 9th June 1997; Accepted, 25th June 1997 Paper E/7/03997B References 1 R. O’Kennedy and R. D. Thornes, Coumarins: Biology, Applications and Mode of Action, Wiley, Chichester, 1997. 2 J. S. Walsh, J. E. Patanella, K. A. Halm and K. L. Facchine, Drug Metabl. Dispos., 1995, 23, 869. 3 R. T. Brown, S. P. Mayalarp, A. T. McGown and J. A. Hadfield, J. Chem. Res.(S), 1993, 496 and references cited therein. 4 G. N. Bollenback, J. W. Long, D. G. Benjamin and J. A. Lindquist, J. Am. Chem. Soc., 1955, 77, 3310. 5 B. Berrang, C. E. Twine, G. L. Hennessee and F. I. Carroll, Synth. Commun., 1975, 5, 231. 6 P. M. Collins and R. J. Ferrier, Monosaccharides, Wiley, Chichester, 1995. 7 J. Vlahov and G. Snatzke, Liebigs Ann. Chem., 1983, 570. 8 T. Ogawa, K. Beppu and S. Nakabayashi, Carbohydr. Res., 1981, 93, C6. 9 Ref. 6, pp. 157–159. 10 N. Pravdic and D. Keglevic, J. Chem. Soc., 1964, 4633. 11 L. F. Tietze and R. Seele, Carbohydr. Res., 1986, 148, 349. 12 A. Nudelman, J. Herzig, H. E. Gottlieb, E. Keinan and J. Sterling, Carbohydr. Res., 1987, 162, 145. 13 G. T. Badman, D. V. S. Green and M. Voyle, J. Organomet. Chem., 1990, 388, 117. 14 R. R. Schmidt, J. Michel and M. Roos, Liebigs Ann. Chem., 1984, 1343. 15 R. R. Schmidt and W. Kinzy, Adv. Carbohydr. Chem. Biochem., 1994, 50, 21 (review). 16 (a) B. Fischer, A. Nudelman, M. Ruse, J. Herzig, H. E. Gottlieb and E. Keinan, J. Org. Chem., 1984, 49, 4988; (b) J. C. Jacquinet, Carbohydr. Res., 1990, 199, 153; (c) P. N. Rao, A. M. Rodriguez and D. W. Miller, J. Steroid Biochem., 1986, 25, 417. 17 F. J. Urban, B. S. Moore and R. Breitenbach, Tetrahedron Lett., 1990, 31, 4421. 18 M. Adamczyk, Y.-Y. Chen and J. R. Fishpaugh, Org. Prep. Proced. Int., 1992, 24, 546. 19 A. V. Stachulski, D. E. Nichols and F. Scheinmann, J. Chem. Res. (S), 1996, 30.