摘要:

440 J.C.S. Perkin I Design of the Substrate for Oxidative Phenol Coupling. An Efficient Synthesis of the c-Homoerythrinan Skeleton By Edward McDonald and Apichart Suksamrarn, University Chemical Laboratory, Lensfield Road, Cam-bridge CB2 1EW Attention is drawn to the importance of conformational interactions in the cyclisation step of intramolecular phenolic coupling reactions. Careful design of the substrate molecule to minimise these interactions should result in better yields for this important strategy of natural product synthesis. The hypothesis is supported by the contrast in the behaviour of the 1-phenethylquinoline (5) and the 1-phenylacetamidoquinoline (8) : when these diphenols were oxidised under identical conditions, (5) gave a complex intractable mixture whereas (8) gave in 67% yield the dienone (23) with the c-homoerythrinan skeleton.OXIDATIVEcoupling of phenols is a key step in living oxidised in the opium poppy to salutaridine (Z),and a systems for generating rather complex structures from well defined sequence of straightforward reduction, comparatively simple ones2 Thus reticuline (1) is tautomerism, and hydrolysis reactions results in the Preliminary communications, E. McDonald and A. Suksam-* See ' Oxidative Phenol Coupling,' ed. A. R. Battersby and ram, Tetrahedron Letters, 1975, 4421, 4425. W. I. Taylor, Dekker, New York, 1969. biosynthesis of morphine (3).3 In principle, oxidation of phenols should be a powerful synthetic tool in vitro _____) HO HO (3 1 and much effort has been spent on biomimetic synthesis of a wide variety of natural products along these lines.4 Unfortunately, yields are usually poor * [though 0.002% for the oxidation of reticuline (1) to salutaridine (2) is exceptionally low], but the recent introduction of new reagents for n~n-phenolic,~monophenolic and di-phenolic * oxidative couplings [particularly with VOCl,, VOF,, Tl(O,CCF,),] coupled with a renewed emphasis 441 on N-protection (as >N+BH3,7” )NC02Et,8c )NCH0,6a-C or >NCOCF,)~CJ.~~~offers considerable hope for the future.In this paper we draw attention to a third factor,l the removal of conformational inter- actions in the transition state for cyclisation, by suitable design of the substrate molecule.c-Homoerysodienone (4) was required in connection with synthetic and biosynthetic studies on the Schell-hammera alkaloids. We reasoned that oxidative cyclis- ation of the diphenol (5) should afford the dienone (a), but inspection of molecular models revealed a set ofM:zvMeO OH0 (4 1 (5) 0 (61 three 1,3-diaxial interactions between the substituents at the marked atoms in the intermediate (6). We * In vivo the transformation (1) +(2) probably involves the enzyme-bound radicals (i) and (ii) held in a conformation which ensures that coupling is exclusively ortho,para to generate (iii) which can easily aromatise to (2). We emphasise that the radical (i 1 (i) is not only extensively delocalised but also sterically hindered survive unchanged until the second electron-transfer step [cf.W.A. See (a) D. H. R. Barton, G. W. Kirby, W. Steglich, G. M. Thomas, T. A. Dobson, and H. Ramuz, J. Chem. Soc., 1965,2423; (b) A. R. Battersby, D. M. Foulkes, and R. Binks, ibid.,p. 3323; (c) G. Blaschke, H. I. Parker, and H. Rapport, J. Amer. Chem. SOC.,1967, 89, 1540; (d) A. R. Rattersby, J. A. Martin, and E. Brockmann-Hanssen, J. Chem. SOC.(C),1967,1785,and references cited therein. For a recent review see T. Kametani, Bio-organic Chenz., 1974, 3, 430. D. H. 13. Barton, D. S. Bhakuni, R. James, and G. W. Kirby,J. Chem. SOC.(C), 1967, 128. (a) S. M. Kupchan, A. J. Liepa, V. Kameswaran, and R. F. Bryan, J. Amer. Chem. SOC.,1973, 95, 6861; (b) S. M. Kupchan, V.Kameswaran, J. T. Lynn, D. K. Williams, and A. J. Liepa,ibid., 1975, 97, 5622; (c) S. M. Kupchan, 0. P. Dhingra, C. K. (ii) (iii) (by the enzyme) and should therefore be sufficiently stable to Waters, J. Chem. Sac. (B),1971, 20261. Kim, and V. Kamsewaran, J. Org. Chem., 1976,41,4047; (d)A. S. Kende and L. S. Liebeskind, J. Amer. Chem. s’oc., 1976, 98, 267; (e) R. E. Damon, R. H. Schlessinger, and J.. F. Blount, J. Org. Chem., 1976, 41, 3772; (f)A. McKillop, A. G. Turrefl, and E. C. Taylor, J. Org. Chevn., 1977, 42, 764. (a) M. A. Schwartz, B. F. Rose, and B. Vishnuvajjala, J. Amer. Chem. Soc., 1973, 95, 612; (b) U. Palmquist, A. Nilsson, V. D. Parker, and A. Ronlbn, ibid., 1976, 98, 2571; (c) S. M. Kupchan, 0.P. Dhingra, and C. K. Kim, J.Org. Chem., 1976, 41,4049. (a) M. A. Schwartz, R. A. Holton, and S. W. Scott, J. Amer. Chem. Soc., 1969,91, 2800; (b) M. A. Schwartzknd R. A. Holton, ibid., 1970, 92, 1090; (c) M. A. Schwartz and I. S. Mami, ibid. 1975, 97, 1239. E. McDonald and A. Suksamrarn, preceding paper. expected that these interactions would raise the energy of the transition state for cyclisation and therefore favour intermolecular coupling leading eventually to polymeric products. (The majority of oxidative phenol couplings are reported to give a low yield of monomeric products together with intractable unidentified poly-mer.) However, if one of the sp3-hybridized carbon atoms C-9 and -11 were replaced by an sp2-hybridized atom, two of the three troublesome interactions would be Me0 \ NO2 Me0 NO2 (9)R = Me (10)R = Me (12)R = H (11) R = H J.C.S.Perkin I hydrogenation in the presence of HCl, and benzylation of (15) gave the protected lactam (16) in 80% overall yield from the aldehyde (12). Methylation of the sodium salt of (16) in dimethylformamide by methyl iodide gave the expected N-methyl derivative, but attempted alkylation using the phenethyl mesylate (17) resulted in elimination rather than substitution and only the styrene (18) was isolated. Although other routes to the amide H (15)R = H (16)R = CH2Ph (7)can be envisaged we turned instead to the synthesis of the isomeric amide (8). Reduction of the lactam (16) with LiAlH, gave the tetrahydroquinoline (19), and this was condensed with the phenylacetic acid l3 (20) using dicyclohexylcarbodi- imide in CH2C1, at 0-5 "C to produce the protected amide (21).Reduction of (21) by LiAlH, gave the amine (22), and hydrogenolysis of (21) and (22) afforded the crystalline diphenolic substrates (8) and (5)required for comparative oxidation studies. The partition coefficients of the diphenolic amide (8) between equal volumes of chloroform and aqueous 5% NaHCO, and Na,CO, were ca. 30 and 1, respectively. HOW2 mOS02Me Me0 Ar ___) Ar-Hocco2HMe0 \ $-o-I (13 1 0-(14) removed. We therefore decided to prepare the dipheno- lic amides (7) and (8) and the corresponding amine (5) and to compare their behaviour in reactions with one- electron oxidants. Selective demethylation of 6-nitroveratraldehyde lo (9) was achieved in ca.70% overall yield via alkaline hydrolysis of the ethylene acetal (10) to (11); de-protection of (11) gave 6-nitroisovanillin (12). This highly effective sequence was devised because (a) nitration of isovanillin itself gives a mixture of isomers ;l2 and (b) alkaline treatment of 6-nitroveratraldehyde (9), or the corresponding cinnamic acid, gave complex mixtures of products. Knoevenagel condensation of 6- nitroisovanillin (12) with malonic acid gave the cinnamic acid (13). Massive reversible shifts (ca. 90 nm) in the long wavelength U.V. absorptions of (ll),(12), and (13) on adding alkali proved that demethylation had occurred specifically para to the nitro-group [clearly via the inter- mediate (14)].Reductive cyclisation of the nitrocinnamic acid (13) to dihydroisoquinolone (15) was achieved by catalytic lo J. T. CassadayandM. T. Bogert, J. Amer. Chem. SOC.,1939,61,2461. l1 C-f. M. Greenwood and R. Robinson, J. Chem. SOC.,1932, 1370. phcHzomMe0 \ H (19) phcH20mMe0 \ Px Ar = Me0Ar Oxidations were therefore carried out using potassium ferricyanide in the two-phase system chloroform-l2 R. Pschorr and W. Storer, Ber., 1902, 35, 4393. l3 A. R. Battersby, R. Binks, R. F. Francis, D. McCaldin, and H. Ramuz, J. Chem. SOC.,1964, 3600. aqueous 5% NaHCO,, since intermolecular coupling (leading eventually to polymer) should be minimised when the diphenol concentration in the aqueous layer is small.These conditions proved remarkably successful and the dienone Zactam (23) was obtained reproducibly in high yield (up to 677,). The structure (23) is fully supported by analytical and spectral data (and by the interconversions described in the accompanying paper) but two points deserve special comment. First the appearance of the two aromatic proton signals as singlets in the lH n.m.r. spectrum reveals that the dienone (23) has been formed by $ara,$ara-coupling, and the isomeric dienone (24) could not be detected in the crude reaction product by t.1.c. or n.m.r. The coupling step is therefore highly regiospecific with the least hindered product being formed exclusively. Secondly, a double doublet (1 H) at 6 4.60 has been assigned to one of the protons at C-9.Molecular models reveal that in the rigid dienone (23)this proton is near to the amide carbonyl group in the deshielding region, accounting for its exceptionally low-field signal. Oxidation of the free base (5) under conditions exactly similar to those used for the amide (8) gave an exceedingly complex mixture of coloured products, and m Me0 \ Meom Ii M eOvMeos?0 no dienone was clearly identified. Nor was any improve- ment noted when iron(m) chloride was used in attempted oxidation of (5). DISCUSSION While the superiority of the amide (8) over the cor- responding amine (5)as a substrate for oxidative phenolic coupling is exceptionally clear-cut, the mechanistic interpretation of the results remains for the present less certain, because we are studying a multistep reaction.A Possible mechanistic pathway is outlined (Scheme) to provide a basis for comparing the two reactions. The deprotonation and oxidation steps (a and b) are dependent respectively on the pK, and the redox potential of ring B, and these values will vary according to the nature of Y. It is possible therefore that the different results found for (5) and (8) may originate in these primary steps." However, if an intermediate suitable for cyclisation can be generated, then its behaviour will depend on the nature of Y, and for the conformational reasons described in the introduction * Different results may also arise because (5) and the inter- mediate derived from it have an unprotected basic nitrogen, whereas (8) has the same nitrogen ' protected ' as an amide.However, successful oxidations on tertiary bases under similar conditions have been reported.14 443 the ratio of cyclisation (step c) to intermolecular reaction should be more favourable when Y = 0. Furthermore, r(5) a or _____) I 0-0 0' (4) or (23) SCHEME this argument holds regardless of the mechanistic nature of the oxidative cyclisation. t Work is in progress to evaluate further the importance of conformational factors in oxidative phenolic coupling and to apply the principle to the synthesis of natural products. EXPERIMENTAL For general directions see preceding paper.g 3,4-Dirnethoxy-6-nitrobenzaldehyde(9).-To concentrated nitric acid (s.g. 1.42; 30 ml), cooled to 5-10 "C in an ice- bath, was added in small portions finely pulverised vera-traldehyde (3.0 g) during 5 min with vigorous stirring. After stirring at room temperature for an additional 1 h the mixture was poured into chipped ice (ca.300 g), and the pale yellow solid was collected, washed with water, and dried in a desiccator; yield 3.60g (94.476).Recrystallis-ation from ethanol gave pale yellow needles, m.p. 134-135" (Iit.,lo 132-133"). This compound was sensitive to light (Found: C, 50.8; H, 4.0; N, 6.8. Calc. for CgHghTO,: C, 51.2; H, 4.3; N, 6.65%); v,,,. 1690, 1607, 1575, and 1332 cm-l; Amax. 222 (8410), 252 (10 329), 261 (10 604), 312 (4 387), and 340 nm (4 662); 6 4.00 (2 x 3 H,s, 2 x CH,O), 7.38(1 H, s, ArH), 7.58 (1 H, s, ,4rH), and 10.36 (1 H, s, CHO), m/e 211 (base peak), 181, 164, and 136; m* 155.26 (211 -+ 181). 3,4-Dimethoxy-6-nitrobenzaldehydeEthylene -4cetal ( 10).-A mixture of the nitro-aldehyde (9) (9.6 g) ethylene glycol Cyclisations which involve pairing of radical cations, radical substitution, or electrophilic substitutions should also be more favourable when Y = 0.l4 E.g. A. R. Battersby, R. B. Bradbury, R. B. Herbert, M. H. G. Munro, and R. Ramage, J.C.S. Perkin I, 1974, 1394. (4 g) and toluene-p-sulphonic acid (250 mg) in benzene (60 ml) was heated at reflux for 135 min and the water produced was removed azeotropically (Dean-Stark appara- tus). The solvent was evaporated off and the crude product recrystallised from ethyl acetate to give the acetal (10) (10.4g, 89.6y0),m.p.123-124" (Found: C, 51.5;H, 5.2; N,5.45. C11Hl,N06 requires C, 51.75;H, 5.15; N, 5.5%); vmax. 1 612,1 580, and 1 335 cm-l; AInax. 220 (10 491), 240 (11 079),and 295 nni (4 234); 6 3.90 (3 H, s, CH,O), 3.94 (3H, s, CH,O), 4.03 (4 H, s, OCH,CH,O), 6.45(1 H, s, OCHO), 7.24(1 H, s, H-2), and 7.54(1 H, s, H-5); m/e255 (M+),238,and 208 (base peak). 3-Hydroxy-4-~nethoxy-6-nitrobenzaldehydeEthylene Acetal (1 1).-Aqueous 20% potassium hydroxide (470 ml) was added to the acetal (10)(25g), dissolved in hot dioxan (75 ml), and the mixture was heated under reflux for 45 h. The clear reddish-brown solution was cooled in an ice-bath and carefully neutralised (to slightly acidic) ; yellow crystals of the product separated.The acetal (11) was collected, washed with water, and dried in a desiccator; yield 20.2g (85.40//,),m.p. 129-130" (from acetone-benzene) (Found; C, 49.65;H, 4.7;N, 5.7. C,,H,,NO, requires C, 49.8;H, 4.6;N, 5.8y0),vmax. (Nujol) 3 275, 1 582,and 1337 cm-l; A,,,, 246 (9 286), 286 (4 561), and 336 (4 643); Amax, (OH-) 270 (8 349),and 426nm (12 667); 6[(CD,),SO] 3.84(3H, s, CH,O), 3.92 (4 H, s, OCH,CH,O), 6.32(1 H, s, OCHO), 7.13(1 H, s, H-2), 7.55(1 H, s,H-5), and 10.45br (1 H, m, OH) ; rn/e 241 (Mt),224,and 194(base peak). 3-Hydroxy-4-nzethoxy- 6-nitrobenzaldehyde (12).-The acetal (11) (3.4g) was dissolved in acetone (30ml) and aqueous 10% hydrochloric acid (20ml) was added.The stirred solution was warmed on a water-bath (55-60 "C) for 35 min. The acetone was evaporated off; the yellow crystals were collected, washed with water, and dried in a desiccator to give the aldehyde (12) (2.55 g, 91.7%). Re-crystallisation from acetone-water gave yellow needles, m.p. 186-187" (Found: C, 48.95;H, 3.65;N,7.1. C,H,-NO,requires C, 48.75;H, 3.6;IS,7.11%); v,,,, (Xujol) 3 220, 1 694,1 615,1 570,and 1 370cm-l; A,,,. 222 (9 980), 261 (11 507), 310 (5 470),and 336 (5 282); Amax. (OH-) 235 (12 215), 293 (9 243), and 438 nrn (12 639); 6[(CTl,),SO] 3.97(3H,s,CH,O), 7.24(1 H, s,ArH), 7.68(1 H, s, ArH), and 10.20(1 H, s,CHO); nz/e 197 (hf'), 167 (base peak), 150, and 122;m* 141.57 (197- 167).3-Hydroxy-4-nzethoxy-6-nitro-trans-cinnarnzcAcid ( 13).-A mixture of the aldehyde (12)(1.90 g), and malonic acid (1.90g) in pyridine (4ml) and piperidine (10drops) was heated under reflux for 105min. The pyridine and piper- idine were removed in vacuo and the crystals which appeared when the solution was concentrated were digested with acetone-water, collected, washed with acetone-water, and dried in a desiccator to give the cinnarnic acid (13) (2.15 g, 93.3%). Recrystallisation from acetone-water gave yellow crystals, m.p. 256-257.5" (decomp.) (Found : C, 50.05; H, 3.8;N, 6.05. Cl,H,N06 requires C, 50.2; H, 3.8; N,5.95%); vmax. (Sujol) 3600, 3 520, 3 360, 2 500-2 800br, 1 695, 1 635, 1575, and 1370 cm-l; Amax. 230 (11 121), 273 (20 689), and 345 (6 457);A,,,, (OH-) 246 (14 888), 296 (12 317), and 432 nm (12 915); 6[(CD3),SOj 3.90 (3H, s, CH,O), 6.26(1 H, d, J 15 Hz, transolefinic), 7.12 (1 H, s,H-2), 7.75 (1 H,s,H-5),and 7.95(1 H, d, J 15Hz, trans olefi nic) .6-Hydroxy-7-methoxy-3,4-dihydroguinolin-( 1H)-one2 (15).-The cinnamic acid (13) (6.0 g) dissolved in hot ethanol (250 ml) was allowed to cool. Concentrated J.C.S. Perkin I hydrochloric acid (2drops) and 10yo palladium-charcoal (600mg) were added before hydrogenation (Parr hydro- genator) at 45"C and 45lb in-, for 2411. The mixture was filtered through Celite, and the residue washed with hot methanol. The filtrate was concentrated in vacuo; crystals slowly separated. The product was collected and washed with cold ethanol; yield 3.80g.The combined filtrate was concentrated and a small portion of ether added, and the mixture was kept in a refrigerator; thus more crystals were obtained. The total yield of lactarn (15) was 4.28g (88.2%). After recrystallisation from methanol it melted at 211-213" (Found: C, 60.75;H, 6.0; N, 6.95. Cl,Hll-N0,,0.25H20requires C, 60.75;H, 5.85; N,7.1%); vmnx. (Nujol) 3 565, 3 200br, 1 652,and 1 630 cm-l; Amax. 220 (14 769), 266 (8 876),and 292 nm (5 818); 6(C,D,N) 2.72 (4H, m,CH,), 3.72 (3 H, s,CH,O), 6.76(1 H, s,ArH), and 10.64br(1 H, s,exchanged with D,O, KH); m/e 193 (AT+), 178, 165, and 150; nz* 141.06 (193- 165)and 126.40 (178-150). 6-Benzyloxy-7-rnethoxy-3,4-dihydroquinolin-2( 1H)-one (16).-A mixture of the quinolone (15) (1.20g), benzyl chloride ( 1.20g ; excess), and anhydrous potassium car-bonate (800 mg) in dry methanol (10 nil) was heated under reflux for 5 h.The mixture was filtered, the residue washed with hot methanol, and the combined filtrate evaporated. The crude product was redissolved in chloro- form; the solution was washed with aqueous potassium carbonate and water, and dried. Evaporation and re-crystallisation from methanol gave the lactam (16) as crystals (1.70g, 96.6%), m.p. 169--170.5' (Found: C, 71.4; H, 6.1; N,5.0. C1,H,,NO3,0.25H20 requires C, 70.95;H, 6.1;N,4.85y0),vmax, 3 410, 3 210br, 1 681,and 1 625an-'; Amx. 220 (21 315), 265 (10 554), and 292infl nm (6 332); 6 2.58 (2H, m, CH,CO), 2.82 (2 H, ca. t, J 6Hz, ArCH,), 3.82 (3 H, s, CH,O), 5.04 (2 H, s, PhCH,), 6.40 (1 H, s,H-8), 6.67(1 H, s, H-5), 7.36(5 H, m,Ph), and 9.07br (1 H, s, NH); m/e 283 (M+),192 (base peak), and 122.Methylation of the Lactanz (16).-A solution of the lactam (16) (28.3 mg) in dry NN-dimethylacetamide (2ml) was added to a suspension of sodium hydride (200mg of 50% dispersion in oil) in 1 : 1 benzene-dimethylacetamide (5ml). After warming at 55-60 "C for 20min the mixture was cooled to 5 'C and methyl iodide (0.6 ml) was added. The mixture was stirred at 20 "C for 2h, then diluted with benzene. The aqueous layer was separated and extracted with ethyl acetate, and the combined organic extract was washed with water and evaporated to give the N-methyl derivative, m.p.94-96'; m/e 297, 220, 206 (base), and 91;vmax. 1 655 cm-l; A,,,, 263and 292nm; 6 2.68 (4H, m, 2 x CH,), 3.32 (3 H,s, NMe), 3.88(3 H, s, OMe), 5.09 (2H, s, PhCH,O), 6.58 and 6.72(each 1 H, s, 2 x ArH), and 7.36(5 H, m, Ph). Attempted A lkylation of the Lactanz ( 16).-Alkylation was attempted exactly as described above but with the phen- ethyl mesylate (17) (33 mg) in place of methyl iodide. The major product (13mg) was isolated by preparative t.1.c. on silica (elution with 1yo methanol-chloroform) and identified by its spectra as the styrene (18); m/e 240, 149, 121, and 91 (base); v,,,,. 1628cm-l; Amax, 261 and 293sh nm; 6 3.88 (3H,s,OMe),5.16(2H,s,PhCH20),5.09(lH,d,JlOHz), 5.52(1 H,d, J 17Hz), and 6.60 (1 H, dd, J 10and 17Hz) (HA,HB,Hc respectively of ArH&CHAHB with HB and Hc trans), ca.6.9(3H, s, 3 ArH), and 7.4(5 H, s,Ph). 3-BenzyZoxy-4-nzethoxyphenethy2 Mesylate ( 17).-Methane- sulphonyl chloride (0.5 nil) was added to a solution of 3-benzyloxy-4-methoxyphenethyl alcohol l3 (45 mg) in dry pyridine (0.5 ml). After stirring for 2 h at 20 "C the mixture was partitioned between water and 2 : 1 ethyl acetate-ether, and the organic phase was washed with N-HC1 and water, dried, and evaporated to give the mesylate (17) (51 mg, 89y0), m.p. 106-107" (from dichloromethane- hexane) (Found: C, 60.75; H, 6.15. Cl,H,,O,S requires C, 60.7; H, 5.95%); m/e 336, 245, and 91 (base); v,,,. 1360 crn-'; A,,,,. 231 and 280 nm; 6 2.76 (3 H, s, MeSO,), 2.93 and 4.32 (each 2 H, t, J 6 Hz, ArCH,CH,SO,), 3.85 (3 H, s, OMe), 5.14 (2 H, s, PhCH,O), 6.80 (3 H, m, 3ArH), and 7.38 (5 H, m, Ph).6-Benzyloxy-7-methoxy-1,2,3,4-tetrahydroquinolane( 19).-To a suspension of an excess of lithium aluminium hydride (350 mg) in dry tetrahydrofuran (5 ml) was added dropwise the quinolone (16) (900 mg) in tetrahydrofuran (20 ml) over 15min. The mixture was stirred at room temperature for another 1 h and the excess of hydride was destroyed by careful addition of cold saturated aqueous Rochelle salt. The mixture was filtered, the residue washed thoroughly with hot chloroform, and the combined filtrate evaporated. The residual syrup was redissolved in chloroform, washed with water, and dried. Evaporation and crystallisation from dichloromethane-ether gave two crops of the tetra-hydroquinoline (19) (745 mg, 87.1%) as crystals, m.p.105-106" (Found: C, 75.5; H, 7.3; N, 5.0. C,,H,,NO, requires C, 75.8; H, 7.1; N, 5.2%); v,,,. 3400br,w, 2 930, 2840, and 1620 cm-l; A,,,. 220 (21 315), 265 (10 554), and 292infl nm (6 332); 6 1.90 (2 H, m, CH,), 2.62 (2 H, t, J 6.5 Hz, ArCH,), 3.22 (2 H, t, J 6.5 Hz, CH,N), 3.37br (1H, NH), 3.75 (3 H, s, CH,O), 4.97 (2 H, s, PhCH,), 6.08 (1H, s, H-8), 6.54 (1 H, s, H-5), and 7.32 (5 H, m, Ph); m/e 269 (M+),178 (base peak), 150, 122, and 91; m* 126.40 (178 +150) and 83.62 (178- 122). 6-Benzyloxy-1-(3-Benzyloxy-4-methoxyphenylacetyl)-7-methoxy-1,2,3,4-tetrahydroquinoline(21) .-To a stirred solu- tion of the amine (19) (1.35 g) and the acid l3 (20) (1.60 g) in dry dichloromethane (25 ml), cooled in an ice-bath, was added NN'-dicyclohexylcarbodi-imide (DCC) ( 1.09 g) in dry dichloromethane (3 ml), and the mixture was stirred for + h, then at room temperature overnight.The crystals of dicyclohexylurea were filtered off (1.17 g); the filtrate was diluted with more dichloromethane, washed with aqueous potassium carbonate to get rid of the excess of starting acid, back-washed with water, and dried. The solution was concentrated to ca. 6-7 ml and kept in an ice-bath; more urea (6-7 mg) separated. The filtered solution was diluted with hexane and kept in a refrigerator; crystals (2.39 g, 91yo) separated. Recrystallisation from dichloromethane-ether gave needles of the amide (2 l), m.p.112.5-113.5" (Found: C, 75.4; H, 6.2; N, 2.5. C,,H,,NO, requires C, 75.7; H, 6.35; N, 2.7%); vmaX. 1 639 cm-l; A,,, 250 infl (12 785) and 286 nm (7 935); 6 1.80 (2 H, m, CH,), 2.48 (2 H, t, J 6.5 Hz, ArCH,CH,), 3.76 (2 H, partially superimposed with signals at 3.77-3.84, CH,N), 3.74 and 3.77 (2 H + 3 H, each s, ArCH,CON and CH,O), 3.84 (3 H, s, CH,O), 5.10 (2 x 2 H, s, 2 x PhCH,O), 6.72 (5 H, m, ArH and Ar'H), and 7.36 (2 x 5H, m, 2 x Ph); m/e 523 (M'), 432, 228, 227, and 91 (base peak) ; m* 356.83 (523 432). 6-Hydroxy-1-(3-Hydroxy-4-methoxyphenylacetyl)-7-methoxy-1,2,3,4-tetrahydroquinoline(8).-The benzyloxy-amide (21) (1.20 g) was dissolved in chloroform (3 ml) in a long-neck flask (50 ml), and methanol (16 ml), hydrogen chloride-methanol (4 ml), and 10% palladium-charcoal (200 mg) were added.The compound was hydrogenated at atmospheric pressure and at room temperature until the theoretical volume of hydrogen was consumed (1-lfr h). The mixture was filtered through Celite and the residue washed with methanol. Evaporation gave the fihenolic amide (8)as a semisolid (0.786 g), m.p. 64-67" (Found: C, 64.7; H, 6.2; N, 4.05. C1,H2,N0,,+H,0 requires C, 64.75; H, 6.25; N, 4.0%); vnBx. 3 540, 3 300br, 1 639, and 1595 cm-1; A,,,,, 253 (9 331) and 287 nm (7 498); 6 1.85 (2 H, m, CH,), 2.55 (2 H, t, J 6.5 Hz, ArCH,CH,), 3.75 (2 H, partially superimposed with signals at 3.71, CH,N), 3.71br (2 H + 3 H, s, ArCH,CON and CH,O), 3.80 (3 H, s, CH,O), 5.69br (20 H, m, exchanged with D,O), and 6.71 (5 H, m, ArH and Ar'H); m/e 343 (M+),207, 179 (base peak), and 137.Phenolic Oxidation of the Difihenolic Amide (8).-A degassed solution of the diphenolic amide (8) (450 mg) in chloroform (20 ml) was injected through a serum cap into degassed, two-phase chloroform-aq. 5% sodium hydrogen carbonate (150 : 150 ml) containing potassium ferricyanide (1.08 g), and the mixture was shaken at room temperature under nitrogen for 24 h (during which time samples of the chloroform layer were withdrawn for t.1.c.). More ferri- cyanide (432 mg) in water (3 ml) was added and the mixture shaken for another 1 h. Water (150 ml) was added, and the chloroform layer was separated. The aqueous layer was extracted twice with chloroform (2 x 100 ml) and the combined extracts were washed with water and dried.The solution was concentrated in vacuo and the syrup chromato- graphed on silica gel (14 g; 50% dichloromethane-hexane changing to chloroform) to give, after recrystallisation from dichloromethane-ether the dienone (23) as dichloromethane solvate (303 mg, 67.7y0), m.p. 135-137" (Found: C, 63.05; H, 5.35; C1, 6.0; N, 3.95. Cl,Hl,NO,,~CH,C1, requires C, 62.8; H, 5.3; C1, 6.4; N, 3.8%) (after prolonged drying at 72-80', m.p. 229-231"); vrnax. 3 530, 1 680, 1 645, and 1 621 cm-l; A&,. 239 (19 931) and 282 nm (3 927); 6 1.80-2.60br (5 H, m, [CH,],CH,N), 3.55 and 3.72 (2 x 3 H, each s, 2 x CH,O, 3.66br (2 H, s, ArCH,O), 4.60 (dd, J 13 and 6.5 Hz, H-9P), 5.72 (1 H, s), 6.44 (2 H, s,) and 6.72 (1H, s) (2 x dienone-H and 2 x ArH), and 6.06 (1H, s, exchanged with D,O, OH); m/e 341 (M+,base peak), 313, 298, 282, and 270; m* 287.29 (341 313) and 244.63 (298 270).Note. When 6 and 4 mol. equiv. of ferricyanide were employed, the reaction went to completion in 25-35 min, but yields were only 35-40y0. 6-Benzyloxy-1-( 3-benzyloxy-4-methoxyfihenethyl)-7-methoxy-1,2,3,4-tetrahydroquinoline(22).-To a stirred sus- pension of lithium aluminium hydride (300 mg) in dry tetrahydrofuran (10 ml) was added dropwise a solution of the dibenzyloxy-amide (21) (700 mg) in dry tetrahydrofuran (15 ml) over 15 min. The mixture was stirred at room temperature for another 1 h and the excess of hydride destroyed by addition of cold saturated aqueous Rochelle salt.The mixture was filtered, the residue was washed thoroughly with hot chloroform, and the combined filtrates were evaporated. The residual syrup was redissolved in chloroform, washed with water, and dried. Evaporation and recrystallisation from dichloromethane-ether gave di-benzyloxy-amine (22) as crystals (505 mg, 74.1%), m.p. 100-102" (Found: C, 77.65; H, 6.8; N, 2.75. C33H35-NO, requires C, 77.75; H, 6.9; N, 2.75%); vmaX. 1 615 and 1586 cm-l; A,,, 264 and 315 nm; 6 1.85 (2 H, m, CH,), 2.64 (2 H, t, J 6 Hz, ArCH,[CH,],N), 2.76 (2 H, t, J 7 Hz, Ar'CH2CH,N), 3.12 (2 H, t, J 6 Hz, Ar[CH,],CH,N), 3.41 (2 H, t, J 7 Hz, Ar'CH,CH,N), 3.76 (3 H, s, CH30), 3.82 (3 H, s, CH,O), 4.98 (2 H, s, PhCH,O), 5.06 (2 H, s, PhCH,O), 6.17 (1 H, s, ArH), 6.56 (1 H, s, ArH), 6.76 (3 H, m, ArH), and 7.36 (2 x 5 H, m, 2 x Ph) (on irradiation at 1.85, the t at 2.64 and t at 3.12 each collapsed to s; on irradiation at 3.12, the m at 1.85 and t at 2.64 and 3.12 each collapsed to s; on irradiation at 3.12, the m at 1.85 changed to t; on irradiation at 2.76, the t at 3.41 collapsed to s; on irradiation at 3.41, the t at 2.76 collapsed to s); m/e 509 (M+),418, and 91 (base peak).6-Hydroxy-1-(3-Jzydroxy-4-metlaoxyphenethyl)-7-methoxy-1,2,3,4-tetrahydroqui~oZ~~e(5).-The dibenzyloxy-aniine (22) (200 mg) was dissolved in dichloromethane (0.5 ml), and methanol (3 mi), hydrogen chloride-saturated methanol (0.5 ml), and 10% palladium-charcoal (50 mg) were added. J.C.S. Perkin I The mixture was hydrogenated at 1 atm and room temper- ature until the theoretical volume of hydrogen was con-sumed, then filtered through Celite into methanolic hydrogen chloride (0.3 ml). Evaporation gave the diphenolic amine (5) as the hydrochloride salt (98 mg, 75.8% crude yield). After recrystallisation from methanol-ether the amine hydrochloride melted at 179-182"; urnax. 3 320br, 2 540br, 1625, and 1592 cm-'; Amax. 228 and 293 nm; 6[(CD3)2-SO] 1.9-3.7 (lOH,ni, CH,), 3.73br (2 x 3H, s, 2 x CH,O), and 6.6-7.0 (5 H, ni, ArH and Ar'H); rn/e 329 (M') and 192 (base peak); m* 112.05 (329-192). Note. The free ainine (5)is unstable; attempted oxidation as described above for the corresponding aniide (8) gave only complex mixtures of coloured materials which were not further examined. [7/1400 Received, 1st August, 19771

ISSN:1472-7781    

DOI:10.1039/P19780000440

出版商:RSC    

年代:1978

数据来源: RSC