Ttm hc dro n Vo l. 32, p p 749 to 751. Pa g a m w Pre ss 1976. Printe d in G re a t Bda m EXTRACTIVES zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONM OF MILLETIA AURICULATA-III N, zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFE ~INH~, Ii. KHA N, S. K. KA PO O R and A. ZA MA N* ~~trnent of Research in Unani Medicine and ~p~tment (Recekedin of Chemistry, Aligarh Muslim University, Aligarh, India zyxwvutsrq UK 16 October 1975: Acceptedfurpublicalion 27 October 1975) Ahstraet-Two new pyranoisoflavones, auriculasin and isoauriculasin were isolated from M. auricuioto zyxwvutsrqponmlkjihgfe and assigned structures 1 and 3 respectively. The structure assigned earlier to isoauriculatin has been revised to 4. The occurrence of the rotenoid summatrol and the pyranoisoflavones auriculatin, auriculin and isoauriculatin in M. au~c~fa?u roots was reported earlier.“’ From the benzene extract of the leaves of this plant two new isoflavones, designated as a~iculasin and isoauriculasin have now been isolated and assigned structures 1 and 3 on the basis of data presented. Further, a reinvestigation of isoauriculatin during the course of this work makes it necessary to revise its structure to 4. Elementary analysis and mass spectrum of 1 agree with the motecular formula CzSH2,,0h.The similarity of this compound with auricufatin is apparent from its spectral characteristics; UV absorption at 295 nm, IR bands at 3250 and 1650cm-’ (chelated -OH, >=O) and NMR I signals at 8.54,&33, 8.21 (-C-CH, of the chromene ring / CHI and the prenyl side chain) and 2.157 fisoflavone proton). Though the aromatic region of the spectrum differs from that of auriculatin the remaining protons of the chromene ring and the side chain resonate at about the same values and have the multiplicities required by structure 1. The chelated OH gives a sharp singlet at -4.07 but the resonances of the two remaining OH protons merge with the signals of the aromatic protons. The three phenolic OH groups are evident from the signals of three Me groups in the NMR spectra of the acetate and methyl ether. In order to distinguish between the linear or angular fusion of the pyran ring the NMR spectra of auriculasin and its triacetate were compared. The comparison revealed shifts of +0*21and -0.157 in the positions of the doublets of protons 4” and 3” as required by structure 1. The presence of an ion m/e (215)arising from the fragment 2 confirms that the side chain is attached to ring A, and hence must be located at the only remaining site C-8. The substitution in ring B can not be inferred from the complicate pattern stretching from 2-58 to Z-707 in the NMR spectrum of the triacetate and integ~ting for three protons. A 2’,4’-substitution pattern is ruled out as it would make the compound identical with auriculatin. Evidence for a vicinal d~y~oxy system was obtained by conversion to the diphenyl methylen~ioxy derivative with diphenyl ~chloromethane. On this basis, the most likely substitution pattern 3’,4’, was confirmed by oxidation of the trimethyl ether to veratric acid. The mass spectrum of isoauriculasin gave the molecular formula C&,,06. The spectrum differs from that of auriculasin most notably in that the peak due to the molecular ion is barely discernible and is followed by a strong M+‘-68 peak indicative of the loss of a C5H8 fragment, a characteristic feature of the mass spectra of W-dimethyl ally1 ethers of pheno1s.3The presence of an O-ally1 grouping was further substantiated by the NMR spectrum which showed the methyiene protons as a doublet at 5347. Isoau~culasin thus has the same spectral features as isoauriculatin. Since like the latter it failed to give a homogeneous acetate, the NMR spectrum did not distinguish between the linear and angular pyranoisoflavone structures. In order to establish the position of the side chain as well as to get conclusive evidence regarding the nature of the pyranoisoflavone nucleus, the 2’-monomethyl ether (3a) was subjected to allylic rearrangement under conditions employed by Murray CI~1,~This gave, after hydrolysis of the butyrate, an oily product which though homogeneous on TLC showed in its NMR spectrum contamination with traces of butyric acid. As required by structure 3b for this compound the NMR spectrum showed the aromatic protons of ring B as a broad singlet at 2.947 and the vinylic protons of the aa-dimethyl ally1 group as quartet and multiplet of an ABX system. The meta relationship of the B ring protons in the rearranged product is clearer in the spectrum of the derived diacetate (3~) in which the broad singlet is resolved into two doublets at 3.06 and 2.97~ (J = 2 Hz). Comparison of the positions of the 3”,4”protons of the chromene ring in 3b and 3c showed shifts of +0*2l and -@I97 respectively thus confirming structure 3 for isoauriculasin. 750 N. MINHAI et al. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA b: R,=CH,, R*=R,=H,R= c: R,=CHs, R,=R,=Ac, R= Since Claisen rearranaement under the conditions 6Hs); side chain (8.33,3Hs, 8*21,3Hs,6*63.2Hd,4.86, IHm); ArH employed gave fairly good yields, zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA it was decided to apply (3.08-3~26.3Hm). Trimethyl auricula.sin. Prepared hy refluxing auriculasin (lg), it to isoauriculatin. The structure assigned initially to this compound was based on the evidence of the Gibb’s test freshly distilled Me2S0, (3 ml), dry acetone (50 ml) and anhyd which is of a controversial nature. Rearrangement of &CO, (2g) for 60 hr. Crystaltisation from EtOH afforded trimethyl auriculasin (0.8g) m.p. 12&21”.(Found: C. 7268: H, isoauriculatin monomethyl ether 4a gave a mixture of two 6.57; C,H,O, requires; C, 72.71; H, 654%). NMR: 2.18 (IHs, components which on resolution by preparative TLC, 2-H); chromene ring (3.28, IHd, 4”-H. 4.33. IHd, 3”-H, J = IOHz, afforded the major component as a crystalline product, 8*54,6Hs);side chain t8:32,3Hs. 8.18,3Hs, 6.56,2Hd, 4.82, IHm): m.p. 120”.The minor component, representing about 20% (6.17, 6-15, 6.13, 3’,4’,5,-OCHz). Triaceryt auriculasin. Auricutasin (0.5 g), Ac,O (20ml) and of the product ratio, could not be sufficiently purified for a m.p. The NMR spectrum of the product, m.p. 120”, fused AcONa (3 g) refluxed for 3 hr and worked up. Crystallization showed three singlets of aromatic protons confirming the from EtOH gave needles (0.55g) m.p. 174-76”.(Found: C, 67.92; substitution pattern of ring B depicted in 4b. Acetylation H, 546; C,IH3009requires: C, 68.12; H, 5.53%).NMR: 2.14 (IHs. 2-H); chromene ring (3-52, lHd, 4”-H, 4.27, IHd, Y-H, J = IOHz, of this gave a homogeneous acetate though again in 8*52,6Hs); side chain (8.3, 3Hs, 8.10, 3Hs, 6.53, 2Hd, 5.0, IHm); amounts insufficient for crystallisation. Comparison of (7.58. 3Hs, 7.74, 6Hs, 5,3’,4’-OAc). the NMR spectra of 4b and its acetate revealed Diphenvl methyfenedioxy auriculosin. Auriculasin (100mg) unexpectedly shifts in the position of the chromene dichlorodiphenyl methane (0.05ml) were heated on a metal bath at protons the direction and magnitude of which is not in 185”for 5 min. The mixture in benzene was passed through a small accord with the angular structure assigned earlier, and this column of silica gel and the product crystallised from EtOH, needles (80 mg), m.p. 180”.(Found: C. 77.56; H, 5.61. C,,H,,O, must therefore be replaced by 4. requires: C, 77.60; H. 563%). KMnO, oxidofion of trimethyl ouricukuin. Auriculasin trimethyl ether (400mg) in acetone (50ml) was refluxed with a small amount of powdered KMnO+ After decolourisation further small amounts of KMnO, were added till the colour persisted. The solvent was removed under reduced pressure, the residue taken up in boiling water, filtered and acidtfied. Repeated crystallisation from water afforded veratric acid (20 mg) m.p. 17678’. Isonuriclrlasin 3 crystallised from EtOH as colourless needles m.p. 134-35”;A::::“’ 225 (inn, 295nm; vzi:“‘, 3400. 3250. 1640, 1620, 1380, 136Ocm-‘; M' , m/e 420 (2%), CZTHZ40n; NMR: 4: R=R,=H, Rz= (60mc); 2.14 (IHs, 2-H); -2.37 (IHs, 5-OH); chromene ring (3.22. lHd, 4”-H, 4.34, IHd, 3”-H, J = lOHz, 84& 6Hs); side chain (8.16. 6H broad singlet, 5.34. 2Hd. 446, IHm); ArH (3.6, IHs, a: R=H, R,=CH,, Rz= 8-H. 2.84-2.97. 3Hm). AIlylic rearrungemenf of 3a. fsoauriculasin (60 mg) was treated with CH,N2 to give the monomethyl eher (40 mg) m.p. 105”.A b: R2 = H, R, = CH,, R = mixture of the monomethyl ether, butyric anhydride (0.5 ml) and N,N-diethylaniline (0.5 ml) was heated under N2 on an oil bath for 8 hr while the temp was maintained at l9tl? 5”. The mixture was EXPERIMENTAL poured into water and worked up. The bulyrate was taken up in The m.ps were taken on a Kofler block and are uncorrected. EtOH and treated with 1% ale NaOH on a water bath for I min. neutralised with HCI. extracted with ether. TLC of the resulting NMR spectra were determined with HA-100 instrument in CDCL with TMS as internal standard. Analyses were carried out by the oily product showed a single spot under UV which gave a positive ferric colouration. NMR: 2.15 (1Hs. 2-H); -4.0 (lHs, 5-OH) Australian Microanalytical Service, Melhourne. Isolation. The benzene extract of air dried M. atiricul~lo leaves chromene ring (3.35, IHd, 4”-H, 4.45, lHd, 3”-H, J = 10HZ, 8.55, (2 kg) was passed through a silica gel column (4OOg),using light 6Hs). side chain (8.30, 6Hs. 5.0. 2Hm, 451, IHq, J = 10 and petroleum, benzene, EtOAc and their mixtures for elution. 16.5 Hz): ArH (3.7. IHs, 8-H, 294 broad singlet); 6.18 (3Hs, lsoauriculasin along with traces of isoauriculatin was eluted by 3’-OMe). Acetate of 3b was prepared as in case of auriculasin. NMR: 2.23 light petroleum-benzene, and purified by preparative TLC which gave isoauriculasin (75 mg). Auriculasin (8 g) was obtained from (IHs. 2-H); chromene ring (3.56, 1Hd. 4”-H, 4.26, IHd, 3”-H. benzene and benzene-EtOAc fractions and purified by repeated J = IOHz, 8.59. 6Hs); side chain (8.5, 6Hs, 5.0, 2Hm, 4.03, tHq. J = IO and 16.5Hz); ArH (3.3, IHs, bH, 3.06, IHd, 297, IHd. crystahisations. Auricufusin-I. Yeltow needles (EtOH) m.p. 176178”. (Found: J = 2 Hz); 6.21 (3Hs, 3’-OMe); 7.56 (6Hs, 5,4’-OAc). Allylic reorrungemenf of 4a. Monomethyl isoauriculatin (50 mg) C,71*74;H,5.74:C,,H,,O,requires: C,71*41;H,5.74%)M’ ,m/e 420; h ::” 225(inf), 295nm; v,,,.*(Nujol), 3420.3250,1650,1620, was rearranged by the same procedure, crystatlised from EtOH 1385, l370cm-‘; NMR: 2.15 (IHs, 2-H); -4.0 (IHs, 5-OH); (15mg) m.p. 120”,M' ,434; NMR: (60 mc) 2.2 (lHs, 2-H); -3.34 chromene ring (3.31, IHd, 4’-H, 4.42, IHd, 3”-H, J = 10Hz; 8.54, (IHs, j-OH) chromene ring (3.28, IHd, 4”-H, 440, IHd, 3”-H, J-3 Extractives of M. auriculafa-HI J = IOHz, 860,6Hs); side chain (8*56,6Hs,5G4,2Hm, 4.1. IHq); ArH (364, lHs, 8-H, 2.7, IHs, 6’-H, 3.38, 3’-H); 6.26 (3Hs, Z’-OMe). Acetale of 4h. NMR (60mc) 2.24 (lHs, 2-H); chromene ring (3.50, IHd. 4”-H, 4-26, IHd. 3”-H, 1 = IOHz. 8.6.6Hs); side chain (8.56,6Hs, 5.08, ?Hm, 4W IHq); 6.28 (3Hs, ?‘-OMe); 7.60 (6Hs, 5,4’-OAc); Arlj (3.43, IHs, 8-H, 3.30, IHs, 3’-H, 2.76, IHs, 6’-H). Acknnvledgrments-The authors are grateful to Mr. K. C. Sahni, F.R.I.. Dehra Dun for his help in the collection and identification 751 of the plant material and the CCRIMH, Govt. of India, for financial assistance. REFERFXES ‘M. Shabbir, A. Zaman, L. Crombie. B. Tuck and D. A. Whiting, /. Chem. sot. (C), 1899(1968). ‘M. Shabbir, A. Zaman, Tel&e&on 26, 5041 (1970). ‘E. Ritchie, W. C. Taylor and J. S. Shanon. Tetrahedron Leffers No. 23, 1437(1964). ‘R. D. H. Murray and M. M. Ballantyne, Terrahedron 26. w7 (1970).