Nucleosides. Part 5. Isolation and characterization of the stable cyclic adducts, (5R,6S)- and (5S,6S)-bromo-O6,5prime;-cyclo-5,6-dihydrouridines in the bromination of 2prime;,3prime;-O-isopropylideneuridine withN-bromosuccinimide

摘要

J. CHEM. SOC. PERKIN TRANS. I 1988 Nucleosides. Part 5.lsquo; Isolation and Characterization of the Stable Cyclic Adducts, (5R.6S)-and (5S,6S)-Bromo-06,5lsquo;-cyclo-5,6-dihydrouridines in the Bromination of 2rsquo;,3rsquo;-O-lsopropylideneuridine with N-Bromosuccinimide Kosaku Hirota,rdquo; Tetsuo Tomishi, Magoichi Sako, and Yoshifumi Maki Gifu Pharmaceutical University, Mitahora -higashi, Gifu 502,Japan Bromination of 2rsquo;,3rsquo;-0-isopropylideneuridine (1) with N-bromosuccinimide in chloroform containing acetic acid gave two diastereoisomeric cyclic adducts, (5R,6S) -and (5S,6S) -brorn0-0~,5rsquo;-cyclo- 5,6-dihydro-2rsquo;,3rsquo;-0-isopropylideneuridines (4a) and (4b), whose structures were determined on the basis of their chemical reactivities and lsquo;H n.m.r. spectral results. The cyclic adducts (4a) and (4b) formed an equilibrium mixture under acidic conditions (4a)/(4b) = 9:111, while under neutral and basic conditions both adducts were converted into 5- bromo-2lsquo;,3rsquo;-0- isopropylideneuridine (2).Electrophilic substitution at the 5-position of uridines has been intermediate, a 5,6-dihydro adduct (B), has often been isolated.rsquo; extensively investigated for the purpose of the synthesis of On the other hand, previous studies have shown that electro- biologically active 5-substituted uridines.2 Two types of mech- philic substitution, e.g. base-catalyzed hydroxymethylation and anism have been proposed to explain the electrophilic hydrogen exchange, at the 5-position of 2rsquo;,3rsquo;-O-isopropyl- substitution reaction (Scheme l).rsquo; One is a mechanism ideneuridine (1) proceeds by the latter mechanism and is 0 R R R (A I /0 ,I+ 0 0-yrsquo;Y *x R R (8) Scheme 1.analogous to that for aromatic electrophilic substitution, remarkably accelerated by the intramolecular participation of a involving the transient formation of a CF complex (A) which 5rsquo;-sugar hydroxy group.rsquo; However, the intermediacy of such an could be stabilized by electron donation from the ring intramolecular cyclic adduct (C)for electrophilic substitution nitrogen (N-1). In another mechanism, nucleophilic addition is less well established. lo* at the 6-position occurs before electrophilic attack at the 5-In the course of our study l2 on the bromination t of 2rsquo;,3lsquo;-position to give a 5,6-dihydro adduct (B) as a key 0-isopropylideneuridine (l), we found the formation of two intermediate.For example, bromination of uridines in aprotic solvents, e.g. acetic anhydride and N,N-dimethylformamide (DMF),5 probably occurs by the former mechanism and in aqueous6 and alcoholic solutionrsquo; the latter mechanism appears to prevail with the addition of an hydroxylic function (e.g. water or alcohol) to the 6-position followed by attack of a bromonium ion (Br+) at the 5-position. For a reaction proceeding via the latter mechanism, a stable t The halogenation of uridine derivatives has been extensivelyE (C Iin~estigated.~However, only few studies on the bromination of 2lsquo;,3rsquo;-0-isopropylideneuridine (1) have been undertaken. Table. Coupling constants for compounds (9,(6a)P and (6b)rdquo; J(HH)/Hz (5) (6a)/6SC (6b)/6R 595 17.1 17.1 17.7 596 9.0 8.9 4.9 5.6 5.6 2.8 4lsquo;3lsquo; 2.6 2.4 3.7 0 0.7 3.7 5rsquo;3rsquo; 12.8 13.0 - lsquo;Cited from the data reported by Cadet et al.15 In CDC1,.In D,O. stable cyclic adducts, 5-brom0-0~,5rsquo;-cycl0-5,6-dihydro-2rsquo;,3rsquo;-0-isopropylideneuridines (4a) and (4b) upon treatment of the uridine (1) with N-bromosuccinimide (NBS) in chloroform containing acetic acid. Here, we describe the isolation and structural elucidation of the cyclic adducts (4a) and (4b), and their chemical reactivities. Results and Discussion Treatment of 2rsquo;,3rsquo;-O-isopropylideneuridine(1) with an equi- molar proportion of NBS in chloroform at 40lsquo;C for 20 h gave 5-bromo-2rsquo;,3rsquo;-0-isopropylideneuridine(2) and 5,Sdibromo-06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-~-isopropylideneuridine(3)in 3 1 and 21 yield, respectively. The structures of (2) and (3) were confirmed by comparison with authentic samples.rsquo; 2*1 In sharp contrast, bromination with NBS in chloroform containing acetic acid chloroform-acetic acid, 10: 1 (v/v) gave a 1 :1 mixture of two diastereoisomeric brominated compounds in 5404 yield.Fractional recrystallization of the mixture from methanol allowed isolation of products (4a) (m.p. 21 3--2 15 ldquo;C, -K~(M~OH)-78.8rdquo;)-and (4b) (m.p. 232--235 ldquo;C, n a: NBS in CHC13 b: NBS. AcOH in CHCL, (la) J. CHEM. SOC. PERKIN TRANS. I 1988 The photo-debromination l4 of (4a) and (4b) by irradiation in DMF gave the same product, 06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-0- isopropylideneuridine (5).l4 Irradiation of 5,5-dibromo-06,5rsquo;- cyclo-2rsquo;,3rsquo;-O-isopropylideneuridine(3) in DMF gave quanti- tatively a mixture of (4a) and (4b) in the ratio of 52:48.These results indicate that the C-6 of (4a) and (4b) adopts the same configuration. The configuration of (5)was reasonably deduced by a comparison of its lsquo;H n.m.r. spectrum with those of (6s)-and (6R)-06,5rsquo;-cyclo-2rsquo;-deoxy-5,6-dihydrouridines(6a) and (6b),the stereochemistry of which has been firmly established by Cadet et al.rsquo;rsquo; As shown in the Table, the coupling constants of protons around the C-6 of (5)are closely similar to those of the 6s-isomer (6a). Therefore, all compounds (4a), (4b), (5), and (3)* have an S-configuration at the 6-position.Irradiation of the 6-H signal in a nuclear Overhauser effect (n.0.e.) experiment for (4a) and (4b) led to enhancement of the 5-H signal (25.5 and 5.8, respectively), clearly indicating that the configuration of protons at C-5 and C-6 is cis for (4a) and trans for (4b). Both (4a) and (4b) were converted with ease into 5-bromo-2rsquo;,3rsquo;-0- isopropylideneuridine (2) upon warming in methanol. Half-lives for the consumption of (4a) and (4b) in methanol at 28 ldquo;C were 1.5and 8 h, respectively. This result supports the view that the SR,6S-isomer (4a) is able to adopt a more favourable con- figuration at the 5-and 6-positions than the 5S76S-isomer (4b) and thus undergo trans-elimination: this is in agreement with the conclusion obtained above by the n.0.e.experiment. In the bromination of the uridine (l),the fb06,5rsquo;-cyclic bond (S-configuration) formed at the 6-position exclusively to give the cis-and trans-cyclo adducts (4a) and (4b). This has an interesting mechanistic implication in comparison with the reaction of thymidine with N-iodosuccinimide rsquo; since the latter gives a mixture of two trans-addition products resulting from both X-and p-06,5rsquo;-cyclic bond formation. Two plausible mechanisms (paths a and b) for the formation 0 + 0 + Scheme 2. a,(MeOH) -49.5rsquo;). Microanalytical results and sDectral data for these indicated that they were two of four pamp;sible diastereoisomeric 5-bromo-06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-~-iso-propylideneuridines, the absolute configurations of which were determined as (5R,6S)- and (5S,6S)-on the basis of the follow- ing facts.of (4a) and (4b) can be formulated as shown in Scheme 3. A mechanism initiated by addition of a bromine radical was omitted because the NBS-carboxylic acid system is used as * The absolute configuration at the 6-position of (3) has been equivocal.rsquo; * 2229J. CHEM. SOC. PERKIN TRANS. I 1988 0 1) have been proposed l7 as possible intermediates in the ring contraction of 5-halogeno-2rsquo;,3rsquo;-O-isopropylideneuridinesinto H~r.go(51 0 0 Hdi.r.(6a) (6b1 the source of bromonium ion.16 Path a involves 1,4-addition initiated by protonation at the 4-carbonyl oxygen leading to an intermediate D. Path b involves concerted 1,2-trans-addition to give (4b)and its subsequent epimerisation to (4a). Path a seems to be a more favourable mechanism than path b, by which the concurrent formation of (4a) and (4b) is not reasonably explained (videinfra). f Scheme 3.As described above (4a) and (4b) were slowly converted into the 5-bromouridine (2) in methanol at room temperature. Upon treatment of (4a) and (4b) with sodium methoxide in methanol or with 1~ sodium hydroxide the Sbromouridine (2) was im- mediately formed. 5-Halogeno-5,6-dihydro-2rsquo;,3rsquo;-O-isopropyl-idene-06,5rsquo;-cyclouridines (C: E = halogeno group, in Scheme the imidazole nucleoside by sodium hydroxide and in the formation of 2rsquo;,3rsquo;-O-is0propylidene-0~,5rsquo;-cyclouridineby the reaction of 5-halogeno-2rsquo;,3rsquo;-0-isopropylideneuridineswith sodium methoxide.The formation of the 5-bromouridine (2) from the cyclic adducts (4a) and (4b), corresponding to the proposed intermediate (C)in the above reactions, is contrary to our expectation. On the other hand, the cyclic adducts (4a) and (4b) underwent slow interconversion under acidic conditions, i.e. independent treatment of (4a) and (4b) in chloroform containing a small amount of acetic acid at 60 ldquo;C gave an equilibrium mixture of (4a) and (4b) (9: 11) after ca. 150 h. The slow interconversion of (4a) and (4b) under acidic conditions could occur oia enol- ization and suggests that path a is more favourable than path b for the formation of (4a) and (4b) as mentioned previously. Experimental M.p.s were determined on a Yanagimoto melting-point appa- ratus and are uncorrected.lsquo;H N.m.r. spectra were determined with a JEOL TNM-GX270 n.m.r. spectrometer using tetra- methylsilane as an internal standard. Chemical shifts are reported in p.p.m. (6) and Jvalues in Hz. Optical rotations were obtained with a JASCO DIP-4 automatic polarimeter. Mass spectra were taken on a JEOL JMS-D300 machine operating at 70 eV. Elemental analysis was carried out at the Microanalytical Laboratory of our University. Column chromatography was carried out on silica gel (Wako gel (2-300). Bromination of 2rsquo;,3rsquo;-O-Isopropylideneuridine (1) with N-Bromosuccinimide (NBS).-A suspension of 2rsquo;,3rsquo;-O-iso-propylideneuridine (1) (244 mg, 0.86 mmol) and NBS (178 mg, 1 mmol) in CHC1, (10 ml) was heated at 40 ldquo;C for 20 h to afford three products and starting material on t.1.c.analysis. The resulting precipitate was filtered off and the filtrate diluted with CHCl,. The solution was washed with saturated aqueous NaHCO, and water, evaporated under reduced pressure, and the residue chromatographed on a silica gel column eluting with CHC1,-MeOH (50: 1). The faster moving fraction gave 5,5-dibromo-06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-0-isopropylidene-uridine (3) (78 mg, 2173, which was identical with an authentic sample.rsquo; The later fraction gave 5-bromo-2rsquo;,3rsquo;-0-isopropylidene-uridine (2) (98 mg, 31), which was identical with an authentic sample.rsquo; (5R,6S)-5-Bromo-06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-0-isopropyl-ideneuridine (4a) and (5S,6S)-5-Bromo-06,5rsquo;-cyclo-5,6-dihydro-2lsquo;,3rsquo;-O-isopropylideneuridine(4b).-AcOH (1 ml) was added to a suspension of 2rsquo;,3rsquo;-O-isopropylideneuridine(1) (284 mg, 1 mmol) and NBS (214 mg, 1.2 mmol) in CHCl, (10 ml) and the mixture was stirred for 4 h at room temperature.The resulting precipitate was filtered off and the filtrate was diluted with CHCl,. The solution was washed with a saturated aqueous NaHCO, and water and evaporated under reduced pressure to afford a 1 :1 mixture of (4a) and (4b) (195 mg, 54). Fractional recrystallization of the mixture from methanol gave (4a) (faster), m.p. 21 3-2 15ldquo;C(Found: C, 39.8; H,4.1; N, 7.7. C I 2HI ,BrN,06 requires C, 39.69; H,4.16; N, 7.71); 6,(CDCl,) 7.47 (1 H, br, HN3),6.19(1H,s,lrsquo;-H),5.07(1H,d,J7.69Hz,6-H),4.75(1H, d, J 5.99 Hz, 2rsquo;-H), 4.71 (1 H, d, J5.99 Hz, 3rsquo;-H), 4.56 (1 H, d, J 2.56 Hz, 4rsquo;-H), 4.53 (1 H, d, J7.69 Hz, 5-H), 4.15 (1 H, d, J 12.82 Hz, 5rsquo;-H), 3.78 (1 H, dd, J 12.82 and 2.56 Hz, 5rsquo;-H), 1.53 (3 H, s, CH,), and 1.34 (3 H,s, CH,);m/z 362 (M+);z,, -78.8rdquo; (c 1 in CH,OH).The later fractional precipitate gave (4b), m.p. 232-235 ldquo;C (Found: C, 39.5; H, 4.05; N, 7.75. C,,H,,BrN,O, requires C, 39.69; H, 4.16; N, 7.71); G,(CDCl,) 7.34 (1 H, br, HN3), 6.40 (1 H, s, 1rsquo;-H), 4.93-4.90 (2 H, m, 6-H and 2rsquo;-H), 4.80 (1 H, d, J 5.56 Hz, 3rsquo;-H), 4.51 (1 H, d, J 1.71 Hz, 4rsquo;-H), 4.49 (1 H, br, 5-H), 4.24 (1 H, d, J 12.83 Hz, 5rsquo;-H), 3.82 (1 H, dd, J 12.83 and 1.71 Hz, 5rsquo;-H), 1.53 (3 H, s, CH,), and 1.35 (3 H, s, CH,); m/z 362 (Mrsquo;);ID-49.5rdquo; (c 1 in CH,OH).Isolation of (6s)-06,5rsquo;-Cyclo-5,6-dihydro-2rsquo;,3rsquo;-O-isopropyl-ideneuridine (5)by Irradiation of Compound (4a).-A solution of compound (4a) (50.4 mg, 0.14 mmol) in DMF (100 ml) was irradiated for 24 h with a 400-W high-pressure mercury lamp under an N, stream. The mixture was evaporated under reduced pressure and the residue was chromatographed on a silica gel column eluting with CHC1,-MeOH (10: 1). The faster fractions gave compound (5) (30 mg, 76), m.p. 174-177 ldquo;C lit.,I4 m.p. 156160 ldquo;C (decomp.) (Found: C, 50.95; H, 5.75; N, 9.85. C12Hl,N,06 requires C, 50.70; H, 5.67; N, 9.86); G,(CDCl,) 7.36 (1 H, br, HNrsquo;), 6.21 (1 H, s, 1rsquo;-H), 5.07 (1 H, dd, J 5.56 and 8.98 Hz, 6-H), 4.72 (2 H, s, 2rsquo;-H and 3rsquo;-H), 4.52 (1 H, d, J2.57 Hz, 4rsquo;-H), 4.04 (1 H, d, J 12.82 Hz, 5rsquo;-H), 3.73 (1 H, dd, J 2.57 and 12.82 Hz, 5rsquo;-H), 2.98 (1 H, dd, J 5.56 and 17.09 Hz, 5-H), 2.76 (1 H, dd, J8.98 and 17.09 Hz, 5-H), 1.53 (3 H, s, CH,), and 1.34 (3 H, s, CH,); m/z 284 (Mrsquo;); .ID -50.0rdquo; (c 1 in CH,OH).The later fractions gave 5-bromo-2rsquo;,3rsquo;-0-isopropylidene-uridine (2) (8 mg, 16), which is identical with an authentic sample. J. CHEM. SOC. PERKIN TRANS. I 1988 prepared from Na (6.4 mg, 0.28 mmol) in absolute MeOH (10 ml) was refluxed for 5 min. The mixture was neutralized with Amberlite CG-50 (H +)and evaporated under reduced pressure. Recrystallisation of the residue from EtOH gave compound (2) (90mg, 90), which is identical with the product prepared above. Reaction of Compound (4b) with Na0Me.-A suspension of compound (4b) (100 mg, 0.28 mmol) in methanolic NaOMe prepared from Na (6.4 mg, 0.28 mmol) in absolute MeOH (10 ml) was refluxed for 5 min.The mixture was neutralized with Amberlite CG-50 (H+) and evaporated under reduced pressure. Recrystallisation of the residue from EtOH gave compound (2) (94 mg, 9404), which is identical with the product prepared above. Conversion of Compounds (4a) and (4b) into Compound (2) in l~ NaOH Solution.-Stirring of a solution of (4a) and (4b) (2mg) in 1M aqueous NaOH (2 ml) at room temperature for 5 min resulted in formation of compound (2), which was confirmed by t.1.c. comparison with the product prepared above. Epimerisation Ofrsquo;Compounds (4a) and (4b) in CDCI, in the Presence of AcOH as Monitored hjj N.m.r.-Compound (4a) (ca.1 mg) was dissolved in CDCI, (1 ml) in an n.m.r. tube with tetramethylsilane as an internal standard and AcOH (1 drop) was added to it. The peak height of the 1rsquo;-H signals at 6.19 (4a) and 6.40 (4b) were measured over a period of 168 h at 60 ldquo;C.The ratio of (4a): (4b) reached 91.5 :8.5 after 6.7 h, 78.8 :22.2 after 25.5 h, 66.3:33.7 after 44 h, 56.7:43.3 after 68 h, 50.9:49.1 after 91 h, 45.1 :54.9 after 139 h, and 43.8: 56.2 after 168 h. Compound (4b) (ca. 1 mg) was dissolved in CDCI, (1 ml) in Irradiation of Compound (4b).-A solution of compound (4b) an n.m.r. tube with tetramethylsilane as an internal standard (3.5 mg, 0.0096 mmol) in DMF (10 ml) was irradiated for 18 h and AcOH (1 drop) was added.The peak height of the 1rsquo;-H with a 400 W high-pressure mercury lamp under an N, stream. signals at 6.40 (4b) and 6.19 (4a) were measured over a period The mixture was evaporated under reduced pressure to give the of 188 h at 60 ldquo;C. The ratio of (4b):(4a) reached 98.3 :1.7 after residue, whose lsquo;H n.m.r. spectrum in CDCI, was completely 1 h, 94.1:5.9 after 4 h, 90.8:9.2 after 8 h, 82.1: 17.9 after 24 h, identical with that of (6S)-06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-O-iso-73.0:27.0 after 47 h, 70.9:29.1 after 53.7 h, 65.5:34.5 after 71 h, propylideneuridine (5) obtained above. 63.2:36.8 after 93 h, 60.6: 39.4 after 1 17 h, and 56.6:43.4 after 188 h. Irradiation of (6S)-5,5-Dibromo-06,5rsquo;-cyclo-5,6-dihydro-2rsquo;,3rsquo;-0-isopropylideneuridine (3).-A solution of compound (3) (5.1 mg, 0.012 mmol) in DMF (10 ml) was irradiated for 3 h with a 400-W high-pressure mercury lamp under an N, stream.The solvent was removed under reduced pressure to give a 52:48 mixture of (4a) and (4b), whose ratio was estimated by inspecting the lsquo;H n.m.r. spectrum of the mixture in CDCl,. Conversion of Compound (4a) into Compound (2) in CD,OD as Monitored by N.rn.r.-Compound (4a) (3.5 mg, 0.0097 mmol) was dissolved in CD,OD (1 ml) in an n.m.r. tube with tetra- methylsilane as an internal standard. The peak height of the 1rsquo;-H signal at 6.11 (4a) and the 6-H signal at 8.33 (2) were measured at 28 ldquo;C. The ratio of (4a):(2) reached 97.5 :2.5 after 0.08 h, 63.5:36.5 after 1 h, 38.8:61.2 after 2 h, 27.7:72.3 after 2.75 h,23.4:76.6after3.33 h, 15.1:84.9after4.33h,and6.0:94.0 after 6.5 h.Conversion of Compound (4b) into Compound (2) in CD,OD as Monitored by N.m.r.-Compound (4b) (3.5mg, 0.0097 mmol) was dissolved in CD,OD (1 ml) in an n.m.r. tube with tetra- methylsilane as an internal standard. The peak height of the 1rsquo;-H signal at 6.27 (4b) and the 6-H signal at 8.33 (2) were measured at 28 ldquo;C. The ratio of (4b):(2)reached 91.7:8.3 after 0.5 h, 86.4:13.6 after 1.25 h, 80.9:19.1 after 2.17 h, 65.6:34.4 after 4 h, 63.5 :36.5 after 5 h, 53.5 :46.5 after 7 h, and 45.9: 54.1 after 10 h. Reaction of Compound (4a) with Na0Me.-A suspension of compound (4a) (100 mg, 0.28 mmol) in methanolic NaOMe References 1 Part 4, K.Hirota, T. Tomishi, and Y. Maki, Chem. Pharm. Bull., 1988, 36, 1298. This paper is also considered as Part 59 of a series entitled Pyrimidines. 2 For examples, see: (a) B. R. Baker, T. J. Schwan, and D. V. Santi, J. Med. Chem., 1966, 9, 66; (b) M. J. Robins and S. R. Naik, J. Am. Chem. Soc., 1971, 93, 5277; (c) T. Nagamachi, J.-L. Fourrey, P. F. Torrence, J. A. Waters, and B. Witkop, J. Med. Chem., 1974, 17,403; (d) S. S. Jones, C. B. Reese, and A. Ubasawa, Synthesis, 1982, 259; (e) C. B. Reese and Y. S. Sanghvi, J. Chem. Soc., Chem. Commun., 1983, 877; ibid., 1984, 62. 3 For a review, see: T. K. Bradshaw and D. W. Hutchinson, Chem. Soc. Reu., 1977, 6, 43. 4 D. W. Visser, in lsquo;Synthetic Procedure in Nucleic Acid Chemistry,rsquo; eds. W.W. Zorbach and R. S. Tipson, Wiley, New York, 1968, vol. 1, p. 409. 5 J. Duval and J. P. Ebel, Bull. Soc. Chim. Bid, 1964, 46, 1059. 6 (a) T. K. Fukuhara and D. W. Visser, J. Biol. Chem., 1951, 190, 95; (h) P. A. Levene and F. B. La Forge, Chem. Ber., 1912, 45, 608; (c) R. E. Beltz and D. W. Visser, J. Am. Chem. SOC.,1955, 77, 736. 7 Y. S. Wang, Photochem. and Photobiol., 1962, 1, 37. 8 (a)R. Duschinsky, T. Gabriel, W. Tautz, A. Nussbaum, M. Hoffer, E. Grunberg, J. H. Burchenal, and J. J. Fox, J. Med. Chem., 1967,10,47; (h)L. Szabo, T. I. Kalman, and T. J. Bardos, J. Org. Chem., 1970,35, 1434; (c) G. W. M. Visser, S. Boele, B. W. v. Halteren, G. H. J. N. Knops, J. D. M. Herscheid, G. A. Brinkman, and A. Hoekstra, ihid., 1986, 51, 1466. 9 A. L. Pogolotti and D. V. Santi, in lsquo;Bio-organic Chemistry,rsquo; vol. 1, ed. E. E. van Tamelen, Academic Press, New York, 1977, 1, p. 277. 10 M. Marton-Meresz, J. Kuszmann, and I. Pelczer, Tetrahedron, 1983, 39, 275. J. CHEM. SOC. PERKIN TRANS. I 1988 2231 11 D. Lipkin and J. A. Rabi, J. Am. Chem. SOC.,1971,93, 3309. 16 (a)H. 0.House, lsquo;Modern Synthetic Reactions,rsquo; W. A. Benjamin, Inc., 12 M. Sako, T. Saito, K. Kameyama, K. Hirota, and Y. Maki, Synthesis, Menlo Park, California, 2nd edn., 1972, p. 432; (b) P. E. Sonnet, J. 1987, 829. Org. Chem., 1979, 45, 154. 13 N. K. Kochetkov, E. I. Budovskii, V. N. Shibaev, G. I. Yeliseeva, M. A. 17 B. A. Otter, E. A. Falco, and J. J. Fox, J. Org. Chem., 1969,34, 1390. Grachev, and V. P. Demushkin, Tetrahedron, 1963, 19, 1207. 14 J.-L. Fourrey and P. Jouin, Tetrahedron Lett., 1977, 3393. 15 J. Cadet, L.-S. Kan, and S. Y. Wang, J. Am. Chem. SOC.,1978, 100, 6715. Received 8th October 1987; Paper 711794