"Science, Faculty of"@en . "Chemistry, Department of"@en . "DSpace"@en . "UBCV"@en . "Young, Robert N."@en . "2011-05-02T23:24:37Z"@en . "1971"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "This thesis describes investigations on the natural products found in Thamnosma montana Torr. and Frem., commonly known as turpentine broom.\r\nIn Part I is described the isolation and identification of fifteen constituents of Thamnosma montana. Of those compounds already known to occur in the plant, alloimperatorin methyl ether (7) and isopimpinellin (2) were found in the shoots and leaves, while \u00CE\u00B2-sitosterol (6), thamnosin (8) and the alkaloids N-methylacridone (4), skimmianine (5a) and \u00C6\u00B4-fagarine (5b) were isolated from the roots. A number of coumarins were recognized for the first time in this plant. Umbelliprenin (83) and a new natural product, alloimperatorin methyl ether epoxide (97) were isolated from the shoots and leaves, while the coumarins, isoimpera-torin (87), phellopterin (96), psoralen (68), bergapten (67) and xantho-toxin (55) were obtained from the roots. Finally a new coumarin which we have named thamnosmin was isolated from the roots and on the basis of chemical and spectral data was assigned the structure 6-(l',2'-epoxy-3'-methyl-3-butenyl)-7-methoxycoumarin (90).\r\nPart II presents a discussion of degradative sequences developed for the furanocoumarins, isopimpinellin (2) and alloimperatorin methyl ether (7) and for the coumarin umbelliprenin (83). These sequences were developed in order to be able to gain information about the distribution of radioactivity in these compounds made available from subsequent biosynthetic studies.\r\nIn Part III is described some biosynthetic experiments performed\r\nwith Thamnosma montana plants. In preliminary studies it was shown that\r\nD,L-phenylalanine-[3-\u00C2\u00B9\u00E2\u0081\u00B4C] (14) serves as an efficient precursor of the coumarin constituents in the shoots and that the rate of biosynthesis of\r\nthese natural products was quite rapid. In studies on the roots, cinnamic\r\nacid-[2-\u00C2\u00B9\u00E2\u0081\u00B4C] was shown to incorporate into the monomeric coumarin\r\nconstituents but all attempts to observe positive incorporation into the\r\ndimeric coumarin thamnosin (8) were fruitless.\r\nMevalonic acid (57) was found to be a very poor precursor of all\r\ncoumarins found in Thamnosma montana shoots except in the case of\r\numbelliprenin (83) where reasonable incorporations were observed.\r\nMevalonic acid-[2-\u00C2\u00B3H] lactone (78) was shown to incorporate essentially\r\nexclusively into the farnesyl ether side chain of umbelliprenin.\r\nMevalonic acid-[2-\u00C2\u00B3H] lactone (78) was shown not to incorporate into\r\nisopimpinellin (2) while mevalonic acid-[3R,4R-4-\u00C2\u00B3H, 3S,4S-4-\u00C2\u00B3H] lactone (78) did provide radioactive 2. This latter material was shown, by means of specific degradation, to have essentially all the activity at the 7-position.\r\nSodium acetate-[2-\u00C2\u00B9\u00E2\u0081\u00B4C] was shown to incorporate considerably more\r\nefficiently into isopimpinellin (2) and alloimperatorin methyl ether (7)\r\nthan did mevalonic acid. By means of specific degradations it was shown\r\nthat sodium acetate-[2-\u00C2\u00B9\u00E2\u0081\u00B4C] provided isopimpinellin (2) labeled at C(4), C(7) and the methoxyl carbon atoms. Specific degradations revealed that the activity present in alloimperatorin methyl ether (7) was present, in part, in the C\u00E2\u0082\u0085 side chain and the methoxyl group. These results suggest that acetate is acting both as a C-l and as a C-2 source in this plant. Possible explanations are presented."@en . "https://circle.library.ubc.ca/rest/handle/2429/34209?expand=metadata"@en . "STUDIES ON THE CONSTITUENTS OF THAMNOSMA MONTANA BY ROBERT N. YOUNG B.Sc. Ronors, University of V i c t o r i a , 1967 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of CHEMISTRY \u00E2\u0080\u00A2 T*Te accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October, 1971 In presenting this thesis in part ia l fulfilment of the requirements for an advanced degree at the University of Br i t i sh Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of CL$\u00C2\u00A3^n ( ....\u00E2\u0080\u00A2. 1 DISCUSS! ON CP ART II .... \u00E2\u0080\u00A2.. . . \u00C2\u00BB ' \u00E2\u0080\u00A2 34 EXpEREMENTAL CPART H ... ^... ^ ........ \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 93 DISCUSSION (PART II) \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 EXPERIMENTAL (PART II) \u00E2\u0080\u00A2 156 DISCUSSION (PART III) \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 192 EXPERIMENTAL (PART III) 220 BIBLIOGRAPHY ........ \u00E2\u0080\u00A2 245 LIST OF FIGURES Figure Page 1 Isomerizations Involving Coumarin (11) 6 2 The Biogenesis of Aromatic Compounds via Shikimic Acid (25) 9,10 3 The Biosynthesis of Coumarin (11) 13 4 The Biosynthesis of Umbelliferone (45) 15 5 Haworth's Proposed Biosynthesis of Umbelliferone (45) 17 6 Proposed Mechanism of Lactone Formation in Umbelliferone (45) 18 7 Proposed Biosynthesis of Herniarin (38) in Lavandula 19 8 Biosynthesis of Farnesol (61b) 21 9 Proposed Mechanism of Alkylation of Phenols 10 Degradation of Pimpinellin (65) 27 11 Proposed Scheme for Furanocoumarin Biosynthesis i n Pimpinella magna 28 12 Alternative Pathway of Furanocoumarin Biosynthesis in Pimpinella magna 30 13 The Formative Route to Marmesin (66) 32 14 Extraction Scheme for Isolation of Natural Products From Thamnosma montana Shoots 36 15 Extraction Scheme for the Isolation of Natural Products From Thamnosma montana Roots 39 16 Purification of Fractions 3 and 4 41 17 The uv Spectra of Coumarin (11) , 4-Methylcoumarin and Umbelliferone (45) 44 18 The uv Spectra of 5-Geranyloxypsoralen and 8-Geranyloxy-psoralen 44 19 I r Spectrum of a T y p i c a l Coumarin 45 20 The nmr Spectrum of Coumarin (11) 47 21 The nmr Spectrum of Isopimpinellin (2) 47 22 Mass Spectrum of Coumarin (11) 49 23 Mass Spectrum of Herniarin (38) 49 24 Proposed Fragmentation of Coumarin (11) and Herniarin (38) i n the Mass Spectrometer 49 25 Mass Spectrum of 6,7-Dimethoxycoumarin 50 26 Proposed Fragmentation of Osthol (81) i n the Mass Spectrometer 51 27 Proposed Fragmentation of Dihydroosthol (82) i n the Mass Spectrometer 52 28 Mass Spectrum of Alloimperatorin methyl ether (7) 53 29 Mass Spectrum of Alloimperatorin methyl ether d i o l (3) .. 53 30 Proposed Fragmentation of Alloimperatorin methyl ether (7) and D i o l (3) \u00E2\u0080\u00A2 . . 53 31 Proposed Fragmentation of Xanthotoxin (55) i n the Mass Spectrometer 54 32 Nmr Spectrum of Umbelliprenin (83) 56 33 Mass Spectrum of Umbelliprenin (83) 58 34 Proposed Fragmentation of Unknown 1 (Umbelliprenin ^83)) i n the Mass Spectrometer 59 35 Synthesis of Umbelliprenin (83) . 60 36 Nmr Spectrum of Isoimperatorin (87) 62 f 37 Uv Spectrum, of Thamnosmin (90) 65 38 I r Spectrum of Thamnosmin (90) 66 39 Nmr Spectrum of Thamnosmin (90) 67 40 Double Irradiation Experiment on Thamnosmin (90) (Irradiating at T8.25) .. 69 41 Double Irradiation Experiment on Thamnosmin (90) (Irradiating at T4.85 and at x5.93) 72 42 Double Irradiation Experiment on Thamnosmin (90) (Irradiating at xl.ll) 73 43 Mass Spectrum of Thamnosmin (90) 75 44 Proposed Fragmentation of Thamnosmin (90) in the Mass Spectrometer 76 45 Nmr Spectrum of Hydrolysis Product of Thamnosmin (93) 78 46 Mass Spectrum of Hydrolysis Product of Thamnosmin (93) .... 80 47 Proposed Fragmentation of Hydrolysis Product of Thamnosmin (93) 81 48 Nmr Spectrum of Alloimperatorin methyl ether epoxide (97) 85 49 Mass Spectrum.of Alloimperatorin methyl ether epoxide (97) 86 50 Proposed Fragmentation of Alloimperatorin methyl ether epoxide (.97) in the Mass Spectrometer 87 51 Nmr Spectrum of 6-Formyl-7-hydroxy-5,8-dimethoxycoumarin (99) 119 52 Proposed Mechanism of the Dakin Reaction 121 53 Proposed Mechanism of the Acid Catalyzed Dakin Reaction ... 122 54 Nmr Spectrum of 6-Hydroxy-5,7,8-trimethoxycoumarin (105) .. 124 55 Nmr Spectrum of l,3-Diformyl-4,6-dihydroxy-2,5-dimethoxy-benzene (106) 126 56 Nmr Spectrum of 1,3-Diformyl-2,4,5,6-tetramethoxybenzene (107) ... 128 57 Nmr Spectrum of 1,3-Diacetoxy-2,4,5,6-tetramethoxy-benzene (108b) 132 58 Degradations of Isopimpinellin (2) 134,212 59 Nmr Spectrum of 5-(2'-hydroxyethyl)-8-methoxypsoralen (113) .... 137 60 Nmr Spectrum of 5-(2'-acetoxy-3'-hydroxy-3'-methylbutyl)-6-formyl-7-hydroxy-8-methoxycoumarin (115) 144 61 Nmr Spectrum of 5-(2'-acetoxy-3'-hydroxy-3,-methylbutyl)-6-acetoxy-7,8-dimethoxycoumarin. (117b) 146 62 Nmr Spectrum of l-(2'-acetoxy-3'-hydroxy-3,-methylbutyl)-2,6-diformyl-3,5-dihydroxy-4-methoxybenzene (118) 148 63 Degradations of Alloimperatorin methyl ether (7) 150,216 14 64 Incorporation of D,L-Phenylalanine-[3- C] into Coumarins of Thamnosma montana Shoots Versus Time 195 65 Degradation of Radioactive Umbelliprenin (83) 205 66 The Glyoxylate Cycle (in Part) 214 67 Proposed Incorporation of \"Active Formate\" into C(4) of Isopimpinellin (2) 214 LIST OF TABLES 1 Constituents of Thamnosma montana 4 14 2 Incorporation of D,L-Phenylalanine-[3- C] into Coumarins of Thamnosma montana Shoots 194 14 3 Incorporation of Cinnamic acid-[2- C] into Thamnosma montana : 197 14 4 Incorporation of Cinnamic acid-[2- C] into Thamnosma montana Root 200 14 5 Incorporation of Mevalonic acid-[2- C] into Monomeric Coumarins in Thamnosma montana \u00E2\u0080\u00A2 202 3 6 Incorporation of D,L-Mevalonic acid-[2- H] lactone (78) into Thamnosma montana . . . 203 3 3 7 Incorporation of Mevalonic acid-[3R,4R-4- H, 3S,4S-4- H] lactone (78) into Thamnosma montana 208 14 8 Incorporation of Sodium Acetate-[2- C] into Thamnosma montana Shoots 210 9 Degradation of Active Isopimpinellin (Experiments 18 and 19) . . 211 10 Degradation of Active Alloimperatorin methyl ether (97) (Experiments 18 and 20) 216 14 11 Incorporation of D,L-Phenylalanine-[3- C] into Thamnosma montana Shoots . 223 14 12a;b Incorporation of Cinnamic acid-[2- C] into Thamnosma montana 226 14 13a,b Incorporation of Cinnamic acid-[2- C] into Thamnosma montana Root 229 Table Page 14 14a,b Incorporation of D,L-mevalonic acid\u00E2\u0080\u0094[2- C] into Thamnosma montana \u00E2\u0080\u00A2 231 3 15a,b Incorporation of D,L-mevalonic acid-[2- H] lactone (78) int o Coumarins i n Young Thamnosma montana Plants 232 3 3 16a,b Incorporation of Mevalonic acid-[3R,4R-4- H,, 3S,4S-4- H] lactone (78) into Thamnosma montana 235 14 17a,b Incorporation of Sodium Acetate-[2 C] into Thamnosma montana Shoots 238 ACKNOWLEDGEMENT S It i s my sincere pleasure to thank Dr. James P. Kutney. His guidance and optimism throughout the course of this research have made this thesis possible. 1 would also like to thank Mr. Ashok K. Verma for his 3 P collaboration with me in this research, and also the other members of the group, past and present, for helpful discussions and suggestions. Thanks are due to Dr. Philip J. Salisbury for his patience and expertise in propagating the plants, and also for many helpful suggestions, and also to Dr. D.L. Dreyer for his gracious g i f t of many coumarin samples and for his aid in collecting the plants used in these studies. Special thanks are due to my wife for her perseverance during the preparation of this thesis and for her excellent typing of the manuscript. Receipt of an H.R. MacMillan Family Fellowship is gratefully acknowledged. DEDICATION To V i v i e n INTRODUCTION The o r g a n i c c h e m i c a l c o n s t i t u e n t s found i n p l a n t s have a l w a y s o f f e r e d a c h a l l e n g e t o c h e m i s t s ; t o d e t e r m i n e t h e i r s t r u c t u r e , and t o chemotaxonomists; t o c o r r e l a t e t h e s t r u c t u r a l f i n d i n g s w i t h t h o s e r e p o r t e d f o r o t h e r p l a n t s p e c i e s . Of i n c r e a s i n g i n t e r e s t t o c h e m i s t s and b i o l o g i s t s a l i k e , i s t h e i n v e s t i g a t i o n o f b i o s y n t h e t i c pathways and p r e c u r s o r s u t i l i z e d by p l a n t s t o s y n t h e s i z e t h e s e n a t u r a l p r o d u c t s . The p l a n t s o f t h e Rutaceae f a m i l y ( o f w h i c h t h e C i t r u s genus i s perh a p s t h e b e s t known) a r e m o s t l y t r e e s and s h r u b s , w i d e l y d i s t r i b u t e d i n t r o p i c a l and s u b t r o p i c a l h a b i t a t s . Many s p e c i e s a r e r i c h i n e s s e n t i a l o i l s ( f o u n d i n s p e c i e s i n a t l e a s t 4 of t h e 7 s u b f a m i l i e s o f Rutaceae) and t h e f a m i l y i s n o t e d f o r t h e p r e s e n c e o f b e n z e n o i d compounds; c o u m a r i n s , f l a v a n o i d s and b e n z e n o i d a l k a l o i d s ( e . g . f u r o q u i n o l i n e , a c r i d i n e a l k a l o i d s and o t h e r s ) 1 . P r i c e 1 s u r v e y e d t h e l i t e r a t u r e on 38 g e n e r a o f Rutaceae and n o t e d t h a t f u r o q u i n o l i n e a l k a l o i d s were r e p o r t e d i n 31 o f t h e s e . I n f a c t , he n o t e d t h a t Rutaceae s p e c i e s were t h e s o l e n a t u r a l s o u r c e o f f u r o q u i n o l i n e , a c r i d i n e and i n d o l o q u i n a z o l i n e a l k a l o i d s . F u r t h e r m o r e , P r i c e 1 f o u n d t h a t o f 79 g e n e r a i n 27 f a m i l i e s s t u d i e d and found t o c o n t a i n c o u m a r i n s , 18 g e n e r a b e l o n g e d t o R u t a c e a e , 18 t o U m b e l l i f e r a e w i t h t h e r e s t d i s t r i b u t e d s p a r s e l y among t h e r e m a i n i n g 25 f a m i l i e s o f p l a n t s . He n o t e d t h a t no coumarins had been found i n p l a n t s of f a m i l i e s s u c h as Z y g o p h y l l a c e a e w h i c h a r e c l o s e l y r e l a t e d t o R u t a c e a e , some s p e c i e s of w h i c h were b o t a n i c a l l y c o n f u s e d w i t h R u t a c e a e . T h i s chemotaxonomic e v i d e n c e s e r v e d t o c o r r o b o r a t e the taxonomic a s s i g n m e n t s . Geissman z directed a t t e n t i o n to the wide occurrence of al k y l a t e d flavanoid compounds i n Rutaceae, c a l l i n g t h i s a \"notable example of the capacity f o r a c l o s e l y a l l i e d group of plants to perform a s i n g l e synthetic r e a c t i o n . \" P r i c e 1 states \"the capacity to add an isopentane unit to a wide v a r i e t y of molecular types appears to be a biochemical c h a r a c t e r i s t i c of the family.\" Thamnosma montana Torr. and Frem. '(Rutaceae)known as the turpentine broom, i s a shrub found i n desert mesas and slopes, p a r t i c u l a r l y i n the southwestern United States. The plant has been reported to be used by native Indians for medicinal purposes 3. Bennet and Bonner 4 5 when studying the t o x i c i t y of aqueous extracts of the leaves of desert plants found Thamnosma montana was the most toxic as judged by the response of tomato plants. The crude extract caused the death of young tomato plants i n concentrations as low as 1 mg/ml, with i n seven days. Muller and M u l l e r 5 noted however that Thamnosma montana was highly hospitable to shrub dependent herbs. They stated \"the mere presence of material which when extracted i s t o x i c to other plants may not n e c e s s a r i l y be of any e c o l o g i c a l s i g n i f i c a n c e \" and concluded that the toxins i n Thamnosma montana were i n e f f e c t u a l f a c t ors i n competition between plants. Bennet and Bonner 4 i s o l a t e d three c r y s t a l l i n e compounds from Thamnosma montana, and i d e n t i f i e d two of them as known coumarins, byakan-g e l i c i n (1) and i s o p i m p i n e l l i n (2). Both showed plant growth i n h i b i t i n g properties. The structure of the t h i r d and most toxic component was elucidated by Dreyer 5 and found to be 5-(3'-methyl-2',3'-d:Lhydroxybutanyl)-8-methoxy-psoralen (3) hereafter known as alloimperatorin methyl ether d i o l . HO HO \ 2 3 1 Dreyer developed an improved e x t r a c t i o n scheme and by chromatography of the acetone extract of Thamnosma montana on alumina was able to i s o l a t e not only the three previously i s o l a t e d coumarins but s i x other components as w e l l . These components were i d e n t i f i e d as three known a l k a l o i d s ; N-methylacridone (4) , skimmianine (5a) ahd':,Y\"~fagarine (5b) ; g - s i t o s t e r o l (6); a known furocoumarin, a l l o i m p e r a t o r i n methyl ether (7) and an unknown compound, thamnosin. This was the f i r s t report of N-methylacridone (4), the parent member of the a c r i d i n e a l k a l o i d s , occuring as a na t u r a l product. Dreyer proposed a p a r t i a l s tructure f o r thamnosin based on preliminary evidence, which proved to be i n c o r r e c t . Inaba and Kutney 7' 8 , working i n our l a b o r a t o r i e s were able to determine the structure of thamnosin as the novel dimeric system 8, a heretofore unknown coumarin. 4 5a; R=0CH 6 5b; R=H 7 \u00C2\u00A7 Dreyer has stated that no limonoids (commonly found in Citrus species)9 could be found in Thamnosma montana. He also could find no trace of the epoxide of alloimperatorin methyl ether6, which he had occasion to prepare during the synthesis of alloimperatorin methyl ether diol (3). The known chemical constituents of Thamnosma montana can be summarized as in Table 1. Table 1. Constituents of Thamnosma montana Name Formula m.p. Reference 8-sitosterol (6) C 2 9 H 5 0\u00C2\u00B0 137-9\u00C2\u00B0 6 alloimperatorin methyl ether (7) C17H16O4 108-10\u00C2\u00B0 6 isopimpinellin (2) C11H10O5 148-9\u00C2\u00B0 4,6 thamnosin (8) C 3 0 H 2 8 \u00C2\u00B0 6 244-6\u00C2\u00B0 6,7,8 N-methylacridone (4) C^HnON 202-3\u00C2\u00B0 6,10 skimmianine (5a) C l t t H 1 303N 173-5\u00C2\u00B0 11,12,13 y-fagarine (5b) C13H11O3N 140-2\u00C2\u00B0 11,12,13 byakangelicin (1) C17H18O7 105-7\u00C2\u00B0 4,6 alloimperatorin methyl ether diol (3) Ci7Hi806 174-6\u00C2\u00B0 4,6 Thus Thamnosma montana, as a Rutaceous plant, fits well in the chemotaxanomic patterns noted by Price 1 in that there are present in the plant, furoquinoline alkaloids, an acridine alkaloid and various prenylated coumarins. Thamnosma montana has been placed in the Rutoideae subfamily of Rutaceae-14. It is interesting to note that the novel coumarin thamnosin (8) has recently been found in two other plants of this subfamily; in Ruta pinnata L. f i l . 1 5 , found in the Canary Islands, and in Zanthoxylum dominianum Merr. and Perry 1 6 , found in Australia. Also found in Zanthoxylum domihianum were two monomeric coumarins which possibly would serve as biogenetic precursors of thamnosin (8); suberosin (9), a known coumarin, and suberenol (10), a new coumarin. The co-occurrence of several types of coumarins in Thamnosma montana, including the interesting dimeric system thamnosin (8), appeared to offer a genuine opportunity to study the biosynthesis of coumarins. 'As biosyn-thetic studies require that efficient isolation and purification techniques of the plant constituents be developed, preliminary work was initiated for this purpose. During this preliminary investigation it became apparent that Thamnosma montana contained a considerable number of isolable constituents in addition to those identified by Dreyer6. Thus to gain a more complete appreciation of the plant system and to determine i f these unknown components might offer further opportunities for biosynthetic experiments, an exhaustive isolation and structure elucidation study was carried out. This work is presented in part I of this thesis. Biosynthesis of Coumarins The coumarins form a diverse and quite widely distributed class of natural products. Well over 100 are known17 and most are found in higher plants 1 8 . It is notable that with few exceptions, such as coumarin (11) i tself , natural coumarins are oxygenated in the 7-position (i.e. para to the position of side chain attachment). 13 Figure 1. Isomerizations Involving Coumarin (11) Coumarin (11) is a lactone of cis-o-cinnamic acid (12a). The lactone ring opens easily on heating with alkali and closes spontaneously upon acidification. If the salt 12b is treated with mercuric oxide a cls-trans isomerization takes place to yield, on workup of the reaction, jo-coumaric acid (13). Isomerization of 13 with ultraviolet light yields coumarin ( l l ) 1 9 . Before the availability of radioactively labeled precursors made i t possible to follow biosynthetic processes with precision, postulates were derived from observed structural regularities within groups of natural products. Robinson20 regarded flavones, chalcones and related compounds as containing a basic three unit group designated Cg-C^-C^; a n d coumarins and related compounds as containing two units, C,-C_. These same two units 0 3 are present in phenylalanine (14) and tyrosine (15) but in earlier speculations these were not considered as direct precursors but rather as related compounds. The nature of the origin of the unit was rather vague21. These amino acids (14 and 15) were in fact, found to be key intermediates, when isotopic labelling studies provided direct evidence. Before discussing the elaboration of these intermediates i t is instructive to consider their biogenesis. The Biosynthesis of the C^-C. Unit The phenylpropane unit in many natural products is now considered to be biosynthesized from cyclohexane derivatives arising from cyclization of carbohydrates. Davis and co-workers 2 2 , 2 3 , 2 4 established the biogenetic 14 15 route to phenylalanine ( l a t e r modified to include chorismic acid) as i s represented i n Figure 2, Using Escherichia c o l i mutants, i n some very elegant research, they established that, i n micro-organisms at l e a s t , the condensation of phospho-enol pyruvate (16) with D-erythrose-4-phosphate (17) y i e l d s shikimic acid (25), v i a dehydroquinic (22) and dehydroshikimic acid (23). Their work has been very w e l l reviewed by S p r i n s o n 2 7 . Phospho-enol pyruvate (16) and D-erythrose-4-phosphate (17) (independent intermediates i n the metabolism of D-glucose) are condensed i n what may be considered as a concerted reaction between two phosphate esters leading to the formation of orthophosphate and 3-deoxy-D-arabino-heptulosonic acid-7-phosphate (18). This hypothesis was supported by i s o l a t i o n of 18 from the crude b a c t e r i a l extracts and by synthesis of 18. The synthetic and natural 18 were i d e n t i c a l and synthetic 18 was converted q u a n t i t a t i v e l y to 5-dehydroquinic acid (22) by the b a c t e r i a l extracts. The enzymic condensation of phospho-enol pyruvate and D-erythrose-4- phosphate (17) i s formulated as an attack by a n u c l e o p h i l i c group on the enzyme (symbolized by OH ) on the enol phosphate, concerted with attack by carbon 3 of phospho-enol pyruvate (16) on the aldehyde carbon of 17 and protonation of the aldehyde carbonyl by an a c i d i c group. This re s u l t s i n release of orthophosphate or of a tr a n s i e n t , phosphorylated enzyme and the open chain 18. Thus the reaction can be considered to resemble the chemical a l d o l condensation and the enzymic aldolose condensation. 5-Dehydroquinic acid (22) i s dehydrated by a s p e c i f i c enzyme to give 5- ^dehydroshikimic acid (23) . The enzyme responsible has been p a r t i a l l y Cont. Figure 2. The Biogenesis of Aromatic Compounds v i a Shikimic Acid (25) p u r i f i e d from c e l l free extracts and was found to be highly s p e c i f i c for thi s reaction. The reduction of 5-dehydroshikimic acid C23) then leads to shikimic acid (25). The enzyme responsible (5-dehydroshikimate reductase) has been studied and nicotinamide-adenine dinucleotide phosphate (NADPH)was found to be required as a cofactor f o r reduction. Shikimic acid (25) i s then converted to 5-phosphoshikimic acid (26) v i a phosphate transfer from adenosine-5-triphosphate (ATP) and intermediate 26 then reacts with phospho-enol pyruvate to give the 3-enol pyruvate ether of phosphoshikimic acid (27). A s t e r e o s p e c i f i c , trans-1,4-elimination of phosphate follows u t i l i z i n g the enzyme chorismate synthetase to y i e l d chorimic acid ( 2 8 ) 2 8 . The key intermediate 28 then rearranges i n a reaction reminiscent of the Claisen rearrangement to give prephenic acid C29). Decarboxylation with accompanying expulsion of a hydroxyl group from C(4) y i e l d s pheriylpyruvie.acid (30)., Both the acid catalyzed and enzyme catalyzed aromatization of 29 may be v i s u a l i z e d to be i n i t i a t e d by e l e c t r o p h i l i c attack on the hydroxyl group. Prephenic acid C29) has also been shown to be converted (by enzymes) to p_-hydroxyphenylpyruvic acid (33). NAD i s required, suggesting conversion may occur v i a oxidation of the C(4) hydroxyl group, followed by decarboxy-l a t i o n . Transamination of phenylpyruvic acid then leads to phenylalanine Q_4). The deamination of 14 by phenylalanine ammonia-lysase y i e l d s cinnamic acid ( 3 1 ) 2 9 . S i m i l a r l y transamination of _p_-hydroxyphenylpyruvic acid (33) y i e l d s tyrosine (15) and subsequent deamination leads to g_-hydroxycinnamic acid (34). Tyrosine (15) can also be formed by hydroxy1-ation of phenylalanine (14) and p-hydroxycinnamic acid (34) can s i m i l a r l y be;formed from cinnamic acid (31) but these are minor r e a c t i o n s 3 0 . Most of the work on the shikimic acid route to aromatic compounds was done with micro-organisms Q_. e o l i -mutants). The s i t u a t i o n i n plants i s more obscure mainly because of the i n a v a i l a b i l i t y of s u i t a b l e biochemical mutants. However studies that have been done i n plants suggest the pathway i s s i m i l a r i f not i d e n t i c a l to that elucidated i n micro-organisms 2 9. One possible d i f f e r e n c e may be that plants prefer quinic acid (24a) to shikimic acid (25) as a key intermediate. I t i s p o s s i b l e that quinic acid (24a) can be dehydrated d i r e c t l y to shikimic acid (25) or that 24a i s converted to 5-phosphoshikimic acid (26) v i a 5^-phosphoquinic acid (24b) . The Biosynthesis of Coumarin (11) Coumarin (11) has been found to be biosynthesized from phenylalanine 0-4) derived v i a the shikimic acid pathway. Brown 3 1, by incorporating labeled compounds into coumarin (11) i n sweet grass (Hjefochloe odprata) has found that o-coumaric acid (13) and cinnamic acid (31) are u t i l i z e d inost e f f i c i e n t l y , while phenylalanine (14) and shikimic acid (25) are l e s s e f f i c i e n t l y u t i l i z e d , with acetate and s a l i c y l i c acid being incorporated at very low l e v e l s . Cinnamic acid (31) i s converted to o-coumaric acid (13) by a mechanism found i i i M e l i lotus species (sweet clover) to be co n t r o l l e d by a s p e c i f i c gene 3 3. This ortho-hydroxylation reaction appears to be a key factor i n a plant's a b i l i t y to synthesize coumarin (11). o-Coumaric acid (13) i s next converted to _-coumaryl glucoside (35) and 35 has been shown to be a very good precursor of coumarin (11), having been found i n a l l species containing 11 thus f a r examined 3 1. The glucoside 35 i s isomerized to the cis-isomer, coumarinyl glucoside (36) which may cyclize to form coumarin (11). 37 11 36 Figure 3. The Biosynthesis of Coumarin (11) . Evidence suggests that in Mel i lotus\u00E2\u0080\u009E coumarin exists in the intact plant cel l as the cis-glucoside 36 and that when the cells are disrupted an enzyme, g-glucosidase, specifically hydrolyzes the cis-glucoside (36) releasing coumarin ( l l ) 3 5 . It has been suggested that free 11 which probably exists in some plants, may be formed by a route involving ortho-hydroxylation of cis-cinnamic acid (37) 3 6. The isomerization of o-coumaryl glucoside (35) to coumarinyl glucoside (36) in sweet clover has been shown to be photochemically induced 3 7 ' 3 8 . Biosynthesis of 7-0xygenated Coumarins As mentioned previously, natural coumarins with very few exceptions contain a 7-oxygen functionality. Some simpler 7-oxygenated coumarins are herniarin (38), aesculetin (39) and scopoletin (40). The corresponding cinnamic acid, _-methoxycinnamic acid (41), caffeic acid (42) and ferulic acid (43) are also naturally occurring, thereby suggesting the existence of a biogenetic relationship between these compounds. A body of evidence supports this hypothesis3 1. COOH 41 42 43 The co-occurrence of coumarin (11) and 7-oxygenated coumarin is rare. However, Brown39 observed that coumarin (11) and herniarin (38) do occur together in lavender'(Lavandula officinalis). He found that such precursors as glucose and phenylalanine (14) were incorporated equally well into both, as was cinnamic acid (31), while oxygenated cinnamic acids showed striking differences. For example, o-coumaric acid (13) and its glucoside 36 were utilized selectively for coumarin biosynthesis while j_-hydroxycin-namic acid (34) and trans-2-glucosyloxy-4-methoxycinnamic acid (44) were utilized selectively for herniarin (38) biosynthesis. This result suggested COOH that the biosynthetic pathways to these coumarins were common in the early stages but diverged at the trans-cinnamic acid (31) stage. At this stage, ortho-hydroxylation would lead to coumarin (11) and _p_ara-hydroxylation to 7-oxygenated coumarins. As with coumarin (11), herniarin (38) is found in lavender in the bound state (as a glucoside) and, as with 11, intermediary glucosides appear to be important. Studies of the biosynthesis of umbelliferone (45) in Hydrangea macrophylla1*0,41 were in general agreement with findings in, lavender and these studies allowed the proposal for umbelliferone biosynthesis as represented in Figure 4. Figure 4. The Biosynthesis of Umbelliferone (45) As with coumarin (11) and h e r n i a r i n ( 3 8 ) , umbelliferone (45) i s found i n the,bound s t a t e 4 2 , as skimmin (49) and cis-2,4-di-g-D-glucosyloxycinnamic acid (48) . The conversion 49-^ -45 may only occur when plant c e l l s are disrupted. Reactions 47->-48 and 48-*49 are assumed to be f a s t as these intermediates cannot be i s o l a t e d , although the conversion of jpj^hydroxycinnamic acid (34) to umbellic acid (46) has been detected by trapping experiments 4 0. Edwards and S t o k e r 3 8 have shown that i n lavender, the t r a n s - c i s isomerization reaction i n h e r n i a r i n (38) biosynthesis i s e n t i r e l y photo-chemical with no isomerase involved. They consider that t h i s s i t u a t i o n i s applicable to a l l plant coumarin biosynthesis. However Ourisson and co-workers 4 3 f e e l the t r a n s - c i s isomerization i n scopoletin (40) formation i n tobacco tissue cultures i s not purely photochemical. Thus there seems to be some doubt about the general v a l i d i t y of Edwards and Stocker's statement. Two further aspects of the biosynthetic pathways have been studied; the method of l a c t o n i z a t i o n and the stage at which methylation occurs i n such coumarins as h e r n i a r i n C38) and scopoletin ( 4 0 ) . Haworth 4 4 i n 1942 suggested that the lactone ring of coumarins could be formed i n an oxidative c y c l i z a t i o n . Figure 5. Haworth's Proposed Biosynthesis of Umbelliferone (45) Kenner and co-workers45 studied the biosynthesis of the coumarin antibiotic novobiocin (50) formed by Streptomyces niveus and found the carboxyl oxygen atoms of tyrosine (34) served as a source of the H3C0 50 heterocyclic coumarin oxygen atom. They postulated an oxidative cyclization with an intermediate spirolactone to explain their results. Others 4 6 * 4 7 suggested such a mechanism might also operate in higher plants (see Figure 6.) 45 Figure 6. Proposed Mechanism of Lactone Formation in Umbelliferone (45) Brown 3 1 feels that since the coumarins in question exist as glucosides of coumeric acid, this situation would not be in keeping with the spiro-lactone postulate. Austin and Meyers 4 1 have shown that synthetic radioactively labeled spirolactone (51) w i l l not undergo this reaction in vivo. Evidence that, i n herniarin (38) biosynthesis, methylation of the para-hydroxyl group occurs prior to lactone formation 3 9 also tends to disagree with the spirolactone mechanism. The stage at which 0-methylation occurs i s uncertain. Brown 3 9 found j_-methoxycinnamic acid (41) a much better precursor of herniarin (38) than was umbellic acid (46), which suggests that methylation occurs prior to ortho-hydroxylation, but trapping experiments failed to detect 41 in Lavandula 4 8. Brown31 feels that this plant can u t i l i z e 41 but the main pathway is by way of umbellic acid (46). He proposed that poor incorporation of 46 could be explained i f j__-hydroxycinnamic acid (34) proceeded to trans-2-glucosyloxy-4-hydroxycinnamic acid (52) via enzyme substrate complex (X) with ortho-hydroxylation and glucosylation both occuring without leaving the enzyme surface. He considered the next step to be methylation to trans-2-glucosyloxy-4-methoxycinnamic acid (53), (which is known to incorporate into herniarin (38)) followed by isomerization of the double bond to give the bound form of herniarin, 54. (See Figure 7.) 52 53 54 Figure 7. Proposed Biosynthesis of Herniarin (38) in Lavendula Ourisson4 3 however suggests that methylation in scopoletin (40) biosynthesis, in tobacco tissue cultures, occurs prior to ortho-hydroxylation, as supported by incorporation of ferulic acid (43) into 40. However, he failed to find evidence of the o-hydroxylated ferulic acid intermediate and suggests that cyclization may occur via radical coupling without intermediary ortho-hydroxylation. Thus the o v e r a l l p i c t u r e of the correct sequence of events i s s t i l l somewhat confused. Alkylated Coumarins L i t t l e information i s a v a i l a b l e as to the o r i g i n of a l k y l side chains often found i n n a t u r a l coumarins. Floss et a l 4 9 have presented evidence which shows that formate or the S-methyl group of methionine may act as a carbon source for the methoxy! group i n ixanthotoxin (55). 0CH 3 55 I t i s generally assumed that 0-methyl groups are derived from the \" C - l pool\" i n the plant with methionine as a methylating agent. Recent 14 work i n our l a b o r a t o r i e s has shown that glycine-[2- C] can act as an e f f i c i e n t source of carbon for the methoxyl groups of i s o p i m p i n e l l i n (2) and a l l o i m p e r a t o r i n methyl ether (7). Glycine i s known to be an e f f i c i e n t source of carbon f o r the \" C - l p o o l \" 5 0 . Many coumarins contain isoprene side chains. These side chains are generally assumed to be terpenoid i n nature and are thought to be derived from mevalonic acid (57). The terpenes are w e l l known to a r i s e from a c e t y l coenzyme-A (56) v i a 3-R-mevalonic acid (57) with A 3-isopentenyl-pyrophosphate (58) and dimethylallylpyrophosphate (59) as key intermediates. These pyrophosphate d e r i v a t i v e s and t h e i r higher homolpgs'\"'gerg.nyi. pyrophosphate (58) and f a r n e s y l pyrophosphate (61a) are considered to be the key alkylating agents in the C- or 0- alkylation of coumarins 5 7. (The biogenesis of farnesol (61b) i s pictured in Figure 8.) SCoA 56 ^ ^ S C o A \u00C2\u00B0 0 A A SCoA SCoA H00C 57. OH V H a 0 SCoA OPP H00C ^OPP OPP, 58 59 OPP 61a; R=Pyrophosphate 61b; R=H Figure 8. Biosynthesis of Farnesol (61b) 60 ,0PP OPP Although l i t t l e or no direct evidence is available to substantiate this \u00E2\u0080\u009E hypothesis in coumarin biosynthesis, experiments utilizing radioactively r labeled mevalonic acid (57) have shown the two extra carbons in furano-coumarins to be mevalonate derived 5 2 (see section on the biosynthesis of furanocoumarins). Other experiments have shown mevalonic acid (57) to be incorporated into the prenyl side chains of some natural phenols and quinones5-1. Recently Hamada and Chubachi53 have reported that mevalonic acid (57) is utilized in forming the isoprene side chain of retenone (62). They found 45% of radioactivity of retenone (62) at C(7') and 33% at C(8') 14 when mevalonic acid-[2- C] was incorporated into Defris el l iptica. Kunesch and Polonsky54 have shown isoleucine to be the precursor of the tigloyl side chain of the 4-phenylcoumarin calophyllolide (63). They 14 could not obtain significant incorporation of mevalonic acid-[2- C] which they expected would be a precursor of the pyran\"ring. ' In this coumarin however, the benzene portion of the coumarin nucleus is acetate derived and thus this natural product is not necessarily representative of shikimate derived coumarins. 63 Alkylations are assumed to occur via phenolic oxygen atom attack with elimination of some leaving group (e.g. pyrophosphate) or by phenol activated ortho or para attack with elimination of the leaving group 5 1 (Figure 9.) Direct attack on the carbon bearing pyrophosphate should be preferred (by chemical analogy) due to the steric effects of the methyl groups. This appears to hold i n the biological system, as alkylated phenols of type a and c are much more common as natural products than those of type b and d. These side chains may be elaborated further through oxidations and further cyclization to yield more complex systems (e.g. thamnosin (8)). Geissman 5 5 has suggested senecioyl coenzyme-A (64) as a possible acylating agent. The corresponding acid i s known to occur naturally. SCoA 0 The Biosynthesis of Furanocoumarins The fused furan ring is found commonly in natural benzenoid compounds and is a particularly common substituent of benzenoid compounds found in plants of the Umbe l l i f e r aeand Rutaceae families. Natural coumarins containing a fused furan ring are numerous and they invariably have the ring attached at the 6,7- or 7,8- position of the benzene portion of the molecule (e.g. isopimpinellin (2) and pimpinellin (65)). The two extra furan carbon atoms have been suggested by Seshadri et a l 5 6 to arise from a degraded isopentyl side chain. The observed co-occurrence of furanocoumarins with coumarins bearing C(6) or C(8) isopentyl side chains supported this proposal. Furanocoumarins also co-occur with such compounds as marmesin (66), with the entire side chain intact, further supporting this proposal 5 7. 66 Caporale, Breccio and Rodighiero^8 have studied the biosynthesis of the furanocoumarins bergapten (67) and psoralen (68) in Ficus carica. They found mevalonic a c i d - T 2 ~ C ] ( 5 7 ) a c e t a t e - T 2 - H ] , tyrosine-[2- Cj (15) , 3 3 IU- Rj^-tyrosine C15) , and J2,3^ K ] - s u c c i n i c acid were incorporated i n t o the furanocoumarins but no degradations were performed to determine the p o s i t i o n s of the l a b e l s . They report however, that mevalonic acid (57) was u t i l i z e d more than t h i r t y times more e f f i c i e n t l y than acetate. The 14 incorporation of mevalonic a c i d - I 2 - C](57) i n t o these furanocoumarins i s d i f f i c u l t to r e c o n c i l e with the proposed formation of the furan r i n g v i a a marmesin type intermediate, as C(2) of mevalonic acid (57) should be l o s t with the degradation of the extra three carbon atoms i f Sheshadri's mechanism i s followed. Floss and Mothes 5 2 o f f e r e d the f i r s t d i r e c t evidence as to the o r i g i n of furanocoumarins i n Pimpinella magna. By incorporating r a d i o a c t i v e l y labeled precursors i n t o the plant, they demonstrated that the coumarin nucleus, C^-C^ p o r t i o n , of sphondin (69) i s derived from cinnamic acid (27). Umbelliferone (48) was found to be a f a r better precursor of Pimpinella furanocoumarins than ..was coumarin (11) 5 9 , indicating that para-hydfbxyration* 14 occurs prior to ortho-hydroxylation. By incorporating mevalonic acid-[4- C] (57) into Pimpinella they were able to isolate radioactive pimpinellin (65) and degradation provided good evidence of specific incorporation of C(4) of 57 into the 7-position of pimpinellin (65) (Figure 10.) 5 2 . 37.6 dpm/mmol 4.0-+ 0.9 dpm/mmol 65 Figure 10. Degradation of Pimpinellin (65) Floss and Mothes also presented evidence which they interpreted as indicating that hydroxylations and methylations of umbelliferone (48), to give the required oxygen substitution patterns, occurred prior to prenylation (with the 7-hydroxyl group protected presumably as a glycoside) 5 2. Thus the formation of the furan ring would be a late step in the biosynthetic pathway (Figure 11). Floss and Mothes52 presented evidence which showed, 14 on incorporation of cinnamic acid-[l- C] into Pimpinella magna^ similar specific activities in pimpinellin (65) and isopimpinellin (2), and the same specific activities for both bergapten (67) and isobergapten (74). They felt that these results indicated that these isomeric pairs of furanocoumarins came from common intermediates, 70 and 73. Hydroxylation of 70, for example, could give 71 which would rearrange to 72, thus giving precursors for both pimpinellin (65) and isopimpinellin (2). However in later work, Floss and Paikert 6 0 tested this proposed scheme and found evidence which did not support prenylation as a late 3 step in the biosynthetic pathway. They found [CH~- H]-scopoletin (AO) was not incorporated preferentially into any one furanocoumarin in Pimpinella magna as they would expect i f prenylation occurred later in the pathway, but instead i t appeared that 40 underwent demethylation with the methyl carbon entering the \" C - l pool\" and subsequently being utilized in appropriate methylations of the furanocoumarins undergoing biosynthesis. They proposed an alternative scheme for furanocoumarin biosynthesis, pictured in Figure 12. They feel that prenylation prior to subsequent hydroxylations would also explain the almost exclusive occurrence of 6,7- and 7,8-furanocoumarins in nature. This opinion is supported by Steck and co-workers61. Figure 12. Alternative Pathway of Furanocoumarin Biosynthesis in Pimpinella magna Steck, El-Dakhanky and Brown61 have recently demonstrated the intermediacy of marmesin (66) (ih Ruta graveolens) in the biosynthesis of psoralen (68), bergapten (67) and xanthotoxin (55) and the intermediacy 55 of columbianetln (76) (in Heracleum lanatum) in the biosynthesis of angelicin C77), isobergapten (74) and pimpinellin (65). Via trapping 74 65 experiments they demonstrated the conversion of umbelliferone-12- C](48) to marmesin (66) and columianetin (76) and by direct incorporation of these intermediates (tritiated) established their conversion to the appropriate furanocoumarins. Little is known of the actual mechanisms of furan ring formation. Steck and Brown61 envision the formation of marmesin (66) as depicted in Figure 13. 66 Figure 13. The Formative Route to Marmesin (66) The nature of the reactions leading on to psoralen (681 remains unclear. Brown62 has recently attempted to test some of the proposed pathways to furanocoumarins, utilizing parsnips (Pastinaca sativa L.) which contain the furanocoumarins bergapten (67), imperatorin (79), isopimpinellin ( 2 ) , and xanthotoxin (55). In his incorporation studies with 5-methoxy-7-14 hydroxycoumarin-[2- C](80) Brown could only conclude that there is no evidence to indicate that 80 is a precursor of bergapten (67). From incorpor-ations of mevalonate-[2- 1 4C]\u00E2\u0080\u00A2and-[5- 1 4C] (78), sodium a c e t a t e - f l - 1 ^ ] and-[2- 1 4C] and umbelliferone-[2- C], Brown showed that mevalonate (78) and acetate were very poorly utilized relative to umbelliferone (48), and that in general acetate was a better precursor than mevalonate (78). He reports 14 that mevalonate-[2- C](78) was incorporated into a l l the furanocoumarins 14 to approximately the same extent as the 5- C compound. These results seem to raise doubts as to the role of mevalonate (78) in furanocoumarin formation, at least in parsnips. He suggests i t may be necessary to consider other possible origins of the 3-methylbutanoid moiety in marmesin (66) and thus also the furan ring carbons in furanocoumarins. The seemingly general confusion as to the nature of prenylation in coumarin biosynthesis, and the presence in Thamnosma montana of a number of interesting prenylated coumarins and furanocoumarins, lead us to consider further experiments which might serve to help clarify some of the many questions that remain unanswered. Such experiments wi l l be discussed in Parts II and III of this thesis. DISCUSSION (PART I) Structural Studies on Coumarins From Thamnosma montana Torr. and Frem. This section w i l l discuss the isolation and identification of eight coumarins, not.previously identified as constituents of Thamnosma montana Torr. and Frem. Six of these proved to be known compounds and therefore w i l l be discussed only briefly. On the other hand two of the compounds isolated had not been previously reported as natural products and therefore a more detailed discussion w i l l be provided. Thamnosma montana uti l i z e d in these studies was obtained from a region in the Mojave Desert of California through the kind cooperation of Dr. D.L. Dreyer of San Francisco State College. Large woody plants were thus obtained. A few of these could be successfully transplanted and continued to grow while others died and were used in these structural studies. A supply of seeds was also obtained and could be successfully propagated by Dr. Phil Salsbury of our department. As noted earlier, Dreyer 6 isolated nine compounds from Thamnosma montana including five coumarins. The extraction procedure, in our case, was carried out essentially as reported by Dreyer 6 and is outlined in Figure 14. The preliminary intent of this work was to become familiar with an efficient isolation and purification of the components isolated by Dreyer, in order to obtain quantities of the desired coumarins for subsequent use in deyeloping degradation procedures and for dilution of radioactive compounds isolated in biosynthetic experiments. However 'the isolation of some previously unidentified components prompted these structural studies. Hydroponic feeding of radioactive precursors to the cut sho'ots of Thamnosma montana appeared to be the easiest and most efficient method of incorporation and with this in mind the constituents of the shoots were f i r s t investigated. Mature Thamnosma montana shoots, leaves and seeds were mechanically ground to a coarse powder and then extracted in a Soxhlet apparatus with acetone. The extract was concentrated and the residue was treated with chloroform. The chloroform soluble portion (representing 10% of the dry plant weight) was fi l t e r e d and concentrated to a small volume, and then chromatographed on alumina with fractions taken as shown in Figure 14. The fractions thus obtained were evaluated by appropriate comparison on thin layer chromatography (tic) with authentic samples of the compounds previously isolated by Dreyer 6. Major constituents which did not correspond to the authentic samples were labeled as unknowns. Fraction C appeared to be mainly one component which exhibited purple fluorescence under ultraviolet (uv) light although considerable quantities of non-polar waxy materials which exhibited no uv fluorescence were also noted on t i c . The main component was easily separated, u t i l i z i n g preparative t i c , to yield unknown I as a colourless o i l which crystallized on standing CO.03% of the dry plant weight). Recrystallization provided pure unknown I, mp 61-62\u00C2\u00B0. The structure elucidation of this unknown, and a l l others subsequently isolated, w i l l be discussed after the entire isolation procedure has been presented. Fraction D was observed to be a complex mixture when examined by t i c . The least polar component labeled unknown II (bright yellow spot; uv) did ground shoots Acetone, Soxhlet crude extract Chloroform chloroform soluble portion Column Chromatography Fraction Solvent Compounds l i g h t petroleum ether waxes 25% benzene i n pet. ether 5.0% benzene i n pet. ether waxes 70% benzene i n pet ether benzene unknown I unknown II benzene u n i d e n t i f i e d components E F G benzene chloroform methanol alloimperatorin methyl ether, unknowns I I , III and IV 25% chloroform i n benzene i s o p i m p i n e l l i n unknown V, chlorophylls di o l ? Figure 14. Extraction Scheme for I s o l a t i o n of Natural Products From ' Thamnosma montana Shoots not correspond to any of the previously i d e n t i f i e d compounds. Although i t was apparently the major component i n f r a c t i o n D, i t proved d i f f i c u l t to separate from the other minor components. It was subsequently also i s o l a t e d from the roots of Thamnosma montana at which time i t could be obtained more e a s i l y i n pure form. Fr a c t i o n E was likewise complex and t i c examination indicated the presence of unknown II (in small amounts) and another component (yellow brown spot; uv) which corresponded to alloimperatorin methyl ether (7). Two s l i g h t l y more polar components re f e r r e d to as unknown III (purple spot; uv) and unknown IV (dark brown spot; uv) were also noted on the chromatoplate. Neither of these substances corresponded to authentic samples of the known constituents of Thamnosma montana. These components (except alloimperatorin methyl ether (7)) were present i n small quantities and subsequent examination of the roots indicated they could be more p r o f i t a b l y i s o l a t e d from t h i s portion of the plant. However preparative t i c allowed i s o l a t i o n of a compound, mp 113-114\u00C2\u00B0, (0.15% of the dry plant weight), which on comparison with an authentic sample, was shown to be alloimperatorin methyl ether (7). Fract i o n F was c r y s t a l l i n e and when examined by t i c was observed to be l a r g e l y i s o p i m p i n e l l i n (2). Direct c r y s t a l l i z a t i o n of f r a c t i o n F yielded the pure sample, mp 149-151\u00C2\u00B0, which proved i d e n t i c a l with an authentic sample of i s o p i m p i n e l l i n (2) (0.2% of the dry plant weight). Also present i n f r a c t i o n F were small quantities of alloimperatorin methyl ether (7), unknown I I I and unknown IV. Fract i o n G was eluted from the column as a green (uv) band. T i c i n v e s t i g a t i o n revealed the major component as a green (uv) spot. This component did not correspond to any of the known constituents of Thamnosma montana and thus was denoted unknown V. The compound was accompanied by chlorophylls but these were e a s i l y removed on preparative t i c . Unknown V, i s o l a t e d from s i l i c a gel chromatoplates as an o i l , c r y s t a l l i z e d on standing. R e c r y s t a l l i z a t i o n y i e l d e d pure unknown V, mp 101-103\u00C2\u00B0, representing 0.3% of the dry plant weight. Fract i o n H consisted of a small quantity of a complex mixture with no discernable major component. T i c comparison suggested that some alloimperatorin methyl ether d i o l (3) might be present i n the mixture but when the appropriate band corresponding i n retention time (R^) to 3 was i s o l a t e d by preparative t i c , the very small quantity of material obtained could not be induced to c r y s t a l l i z e . T i c examination revealed that t h i s material was s t i l l quite complex. It thus appears that 3 i s present at best as only a very minor component of the shoots. Careful t i c examination of the above f r a c t i o n s f a i l e d to reveal the presence of e i t h e r thamnosin (8), byakangelicin (1) or any of the a l k a l o i d s i s o l a t e d by Dreyer 6. It was noted that Dreyer had extracted the e n t i r e a e r i a l portion of the plant i n c l u d i n g the root.crown. Thus the root portion of the plant was also investigated. The dried root material was ground to a coarse powder and extracted with acetone i n a Soxhlet apparatus. The acetone extract was evaporated to dryness and the residue was treated with chloroform. The chloroform soluble portion was f i l t e r e d and concentrated to y i e l d an o i l y residue (7,5% of the dry plant weight). This residue was chromatographed on alumina and f r a c t i o n s taken as represented i n Figure 15. ground roots Acetone Soxhlet crude extract Chloroform chloroform soluble portion Column Chromatography Fract i o n Solvent l i g h t petroleum ether 25% benzene i n pet. ether 33% benzene i n pet. ether 50% benzene i n pet. ether 66% benzene i n pet.- ether benzene benzene 20% chloroform i n benzene 25% chloroform i n benzene 33% chloroform i n benzene 50% chloroform i n benzene chloroform 20% acetone i n chloroform 50% acetone i n chloroform acetone methanol Compounds waxes, unpolar materials waxes, 3 - s i t o s t e r o l alloimperatorin methyl ether, y-fagarine, unknowns alloimperatorin methyl ether, y-fagarine, unknowns N-methyl acridone, thamnosin. -skimmianine thamnosin thamnosin (and diol?) polar materials Figure 15. Extract-ion.Scheme-For the- I s o l a t i o n of Natural Products From Thamnosma sontana Roots Direct c r y s t a l l i z a t i o n of f r a c t i o n 2 yielded a pure compound, mp 134-137\u00C2\u00B0, which on comparison with an authentic sample proved to be i d e n t i c a l with B - s i t o s t e r o l (6), (0.008% of the dry plant weight). Fractions 3 and 4 were observed to be complex mixtures (with many common components) when examined by t i c . Consequently these two fra c t i o n s were combined f o r further chromatographic separation (see Figure 16). T i c examination of f r a c t i o n 5 indicated that i t was composed l a r g e l y of a component ex h i b i t i n g an intense blue fluorescence under uv l i g h t , and a retention time on t i c i d e n t i c a l with an authentic sample of N-methylacridone (4). Pale blue (uv) spots corresponding to thamnosin (8) and skimmianine (5a) were also observed but i n seemingly much smaller amounts. Direct c r y s t a l l i z a t i o n of f r a c t i o n 5 yie l d e d a pure compound, mp 199-202\u00C2\u00B0. This component proved to be i d e n t i c a l with an authentic sample of N-methylacridbne (4) and constituted at least 0.08% of the dry root weight. Preparative t i c of the mother liq u o r s from f r a c t i o n E allowed separation of bands corresponding to thamnosin (8) and skimmianine (5a). The residues corresponding to skimmianine (5a) were c r y s t a l l i z e d to y i e l d a small quantity (0.005% of the dry root weight) of a pure compound mp 171-174\u00C2\u00B0, which upon comparison with an authentic sample of skimmianine (5a) proved to be i d e n t i c a l . Fractions 6 and 7 were noted by t i c to contain thamnosin (8). They were both exposed to preparative t i c and the appropriate bands i s o l a t e d to y i e l d , a f t e r c r y s t a l l i z a t i o n , a pure compound, mp 243-246\u00C2\u00B0 (y 0.05% of the dry root weight). This compound was i d e n t i c a l with an authentic sample of thamnosin (8). Attempts to i s o l a t e alloimperatorin methyl ether d i o l (3) from f r a c t i o n 7 were f r u i t l e s s (although a spot corresponding to 3 was apparent on the t i c plate but i n seemingly very small q u a n t i t i e s ) . Comparison of the various f r a c t i o n s by t i c with an authentic sample of 3 and byakangelicin (1) indicated no appreciable quantities of these compounds were present i n the processed root extract. A discussion of the further p u r i f i c a t i o n of fr a c t i o n s 3 and 4, mentioned e a r l i e r , i s now i n order. Figure 16 outlines the procedure employed for this purpose. Fractions 3 and 4 Column Chromatography Fraction Solvent l i g h t petroleum ether 25% benzene i n pet. ether 33% benzene i n pet. ether 50% benzene i n pet. ether 66% benzene i n pet. ether benzene 20% chloroform i n benzene 20% chloroform i n benzene 33% chloroform i n benzene 50% chloroform i n benzene chloroform acetone methanol Cbiiip bunds waxes unknown I I alloimperatorin methyl ether, unknowns I I I , IV and VI unknowns I I I , IV, VI and VII unknown VIII N-methyl acridone y-fagarine polar materials Figure 16. P u r i f i c a t i o n of Fractions 3 and 4 F r a c t i o n b, when examined by t i c , was seen to be a complex mixture with unknown I I , previously observed i n the shoots extract, as a major component. Preparative t i c and c r y s t a l l i z a t i o n of the resultant material, provided unknown II as a c r y s t a l l i n e substance. This compound, obtained i n i t i a l l y as white needles, melted at 97-98\u00C2\u00B0, then changed c r y s t a l form (plates) and melted again at 105-108\u00C2\u00B0. Unknown II constituted 0.01% of the dry root weight. Fract i o n c was a complex mixture but e f f i c i e n t separation of the components could be achieved with preparative t i c . The least polar component (brown band; uv) proved to be alloimperatorin methyl ether (7); while unknown III (purple band; uv), unknown IV (dark brown band; uv) and f i n a l l y unknown VI (pale blue band; uv) were separated as progressively more polar compounds. R e c r y s t a l l i z a t i o n of unknown VI yielded a pure compound, mp 163.5-164.5\u00C2\u00B0, c o n s t i t u t i n g 0.008% of the dry root weight. Fract i o n d' was also a complex mixture and could be separated by preparative t i c to y i e l d (in order of increasing p o l a r i t y ) , alloimperatorin methyl ether (7), unknown I I I , unknown IV, and another component, unknown VII (yellow band; uv). The appropriate f r a c t i o n s corresponding to unknown III as obtained from f r a c t i o n s c\" and d , were combined and c r y s t a l l i z e d to y i e l d the pure compound, mp 101-104\u00C2\u00B0 (0.015% of the dry root weight). In a s i m i l a r manner the various f r a c t i o n s of unknown IV were combined and c r y s t a l l i z e d to provide pure IV, mp 96-98\u00C2\u00B0 (0.025% of the dry root weight). The f r a c t i o n s of unknown VII were combined and c r y s t a l l i z e d to y i e l d the pure compound, mp 186.5-188.5\u00C2\u00B0 (0.030% of the dry root weight). Fraction e was separated by preparative t i c and unknown VIII (yellow brown band; uv) was i s o l a t e d . C r y s t a l l i z a t i o n of the l a t t e r y i e l d e d pure unknown VIII, mp 146-147\u00C2\u00B0 (0.01% of the dry root weight). F r a c t i o n f when examined by t i c was observed to contain N-methyl-acridone (4) and y-fagarine (5b) as major components. Preparative t i c allowed i s o l a t i o n of 5b (pale blue band; uv) and c r y s t a l l i z a t i o n of the l a t t e r provided 5b i n c r y s t a l l i n e form, mp 142.0-143.5\u00C2\u00B0 (0.1% of the dry root weight). This compound and an authentic sample of y-fagarine (5b) were i d e n t i c a l . Isopimpinellin (2), unknown I and unknown V were not evident i n any of the f r a c t i o n s from the chromatography of the root extract. Spectroscopy and the I d e n t i f i c a t i o n of Coumarins Before'discussing the structure e l u c i d a t i o n of the unknown components, i t i s pertinent to consider some of the spectroscopic properties which are generally common to the coumarins. Coumarins have been studied extensively from a spectroscopic point of view and the fund of knowledge thus a v a i l a b l e i n the l i t e r a t u r e serves to greatly f a c i l i t a t e the i d e n t i -f i c a t i o n of these compounds. U l t r a v i o l e t Spectroscopy (uv) U l t r a v i o l e t absorption spectroscopy i s an important t o o l i n s t r u c t u r a l determinations, p a r t i c u l a r l y where one i s dealing with aromatic and phenolic natural products. Innumerable studies have been made r e l a t i n g structure and s u b s t i t u t i o n to uv spectra. The coumarin series i s one area where t h i s t o o l has proven to be p a r t i c u l a r l y u s e f u l . The coumarin o o CO 2i V\ v / o.oi 200 250 300 Figure 17. The uv Spectra of Coumarin (11) ( ) 6 \ 4-Methyl coumarin ( and umbelliferone (45) ( \u00E2\u0080\u0094 \u00E2\u0080\u00A2 \u00E2\u0080\u0094 ) 6 6 . 1.0 -s o C O m 0.0H 200 250 Figure 18. The uv Spectra of 5-Geranyloxypsoralen (-8-Geranyloxypsoralen (\u00E2\u0080\u00A2 ) 6 8 300 \u00E2\u0080\u00A2 ) 6 7 and (benzopyrone) chromophore has a characteristic uv spectrum (Figure 17) which is modified and complicated according to substitution. Oxygen substitution on the benzene portion of the chromophore produces distinct changes in the uv spectrum while alkyl substitution has l i t t l e or no effect 6 3 (see Figure 17) unless conjugation is extended into the side chain. Thus the uv spectrum of a coumarin can offer good evidence as to the degree and position of oxygen substitution. Furanocoumarins have more complex but equally characteristic spectra (Figure 18). Examples are well documented in the various published uv spectra catalogues. Infrared Spectroscopy (ir) Although infrared spectroscopy is useful in determining the presence or absence of numerous structural groups in a coumarin molecule, i t is the a-pyrone ring common to a l l coumarins that is of most interest here. The ct-pyrone ring normally produces three characteristic adsorptions in the i r spectrum (e.g. Figure 19); 1740-1720 cm\"1 (C=0), 1650-1600 and 1590-1560 cm 1 (C=C)69. The presence of these bands serves to distinguish coumarins from similar compounds such as y-pyrones and chromones. C M - ' M I C R O N S Figure 19. Ir Spectrum of a Typical Coumarin70 Nuclear Magnetic Resonance Spectroscopy (nmr) Nuclear magnetic resonance spectroscopy i s particularly useful i n the structure elucidation of coumarins. The spectra are generally well resolved allowing faci l e measurement of coupling constants (e.g. Figure 20 and 21). The protons on carbons 3 and 4 of the pyrone ring couple to an extent C9.4-10.0 Hz) indicative of their cis relationship 7 1 and these signals normally occur at chemical shifts similar to those for the corresponding protons ih oftho-coumaric acid (13) 7 2. These doublets are a characteristic feature of coumarins thereby allowing f a c i l e recognition of a coumarin system in an unknown natural product. In substituted coumarins, observation of the coupling constants between the benzenoid protons can prove useful i n determining the substitution pattern, as ortho, meta and para coupling constants between protons on a benzene ring are characteristically different and therefore distinguishable 7 3. In furanocoumarins, in addition to the H(3), H(4) doublets, the furan proton absorptions are normally evident as another pair of doublets with coupling constants generally of the order 2-3 Hz. Whereas the H(4) signal occurs invariably at lower f i e l d than that due to H(3), the H(6) proton adsorbs at higher f i e l d than the H(7) signal 7\" 4 > 7 5. H l// 8 0 H Figure 20. The nmr Spectrum of Coumarin ( l l ) 7 6 Figure 21. The nmr Spectrum of Isopimpinellin (2) Long range spin-spin couplings have been observed i n coumarins when protons on adjacent rings are linked through f i v e bonds i n the character-i s t i c \"zig-zag p a t t e r n \" 7 7 (e.g. H(4) and H(8) i n 68). The couplings are small (0.5-1.0 Hz) and are dependent upon s u b s t i t u t i o n to some extent. Such couplings are usually manifested i n normal coumarin nmr spectra as a shortening of one of the doublets of a p a i r such as for H(3) and H(4), and can only be observed accurately under high r e s o l u t i o n . Such couplings can be used as evidence f o r lack of s u b s t i t u t i o n at positions such as H(8) i n 68 and i t s analogs. Solvent e f f e c t s on nmr signals of coumarins have been s t u d i e d 7 8 and appear to be useful i n assignments of adsorptions which would be otherwise ambiguous. Mass Spectrometry Mass spectrometry has developed r a p i d l y i n recent years as a s t r u c t u r a l t o o l and has proved p a r t i c u l a r l y h e l p f u l i n the coumarin s e r i e s . Coumarins have been studied mass spectrometrically by several G r o u p s 7 9 - 8 3 and the published r e s u l t s reveal that the coumarin ring system fragments i n the mass spectrometer with f a c i l e loss of carbon monoxide to give stable ions (e.g. Figure 22). The loss of methyl r a d i c a l s from methoxyl groups ( a l k y l r a d i c a l s from O-alkylated coumarins) i s f a c i l e as i s the 8-cleavage of side chains bonded to the aromatic r i n g . The process i s normally v i s u a l i z e d as occurring with ri n g expansion to y i e l d tropylium type ions. It i s to be noted that i n the following discussion the s t r u c t u r a l representations of fragment ions and the mechanistic proposals f o r t h e i r formation i n the mass spectrometer are simply reasonable r a t i o n a l i z a t i o n s which are consistent with the mass peaks observed and earn credence mainly OCX ISO 148 1-2) \u00E2\u0080\u00A2I33U2) | 103 CH.O' 100 ISO \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 l l Figure 22. Mass Spectrum of Coumarin ( l l ) 8 4 Figure 23. Mass Spectrum of Herniarin (38) 8 2 Figure 24. Proposed Fragmentation of Coumarin (11) and Herniarin (38) in the Mass Spectrometer 7 9 through the predictions they allow. Although the elemental compositions of the ions have for the most part been determined by high resolution mass spectrometry, the actual structures and mechanisms of formation are in doubt and considerable investigation would be required before definitive statements could be made. Coumarin (11) and herniarin (38) lose CO(M-28) and 2CO(M-56) from the pyrone ring (Figure 22 and 23). Herniarin (38) also fragments with facile loss of methyl radical as depicted in Figure 24 7 9> 8 2. The presence of additional methoxyl groups further complicates the situation allowing several combinations of CO and CH\u00E2\u0080\u009E loss (e.g. in Figure 25 loss of methyl radical to give M-15 peak occurs in the preference to CO loss). 60 ' 40 20 O 206(>el1 191 1x2.5 H, I63U2.5V _ J i _ I J l . . Ji i l ii J . 207 U2) CH,0. CH. 100 150 m/t Figure 25. Mass Spectrum of 6,7-Dimethoxycoumarin 82 Barnes and Occolowitz79 have reported an interesting fragmentation of osthol (81) as formalized in Figure 26. The mass spectrum of dihydroosthol (82) did not show similar side chain methyl fission as the conjugation stabilization observed in osthol (81) could not occur. Instead, loss of a C^H_ radical to give a tropylium type ion was observed (Figure 27). F i g u r e 26. P r o p o s e d F r a g m e n t a t i o n of O s t h o l (81) i n t h e Mass S p e c t r o m e t e r Figure 27. Proposed Fragmentation of Dihydroosthol (82) in the Mass Spectrometer Similar fragmentations have been observed by Kutney and co-workers8 an example of which is the mass spectrometric fragmentation of alloimperatorin methyl ether (7) and the corresponding diol (3) (Figures 28, 29 and 30.) 100 ' \u00E2\u0080\u00A2 0 \u00E2\u0080\u00A2 eo 40 20 0 284(>4) 2(9 ISO 200 Figure 28. Mass Spectrum of Alloimperatorih.methyl ether ( 7 ) 8 2 100 80 i \u00C2\u00A3 \u00C2\u00AB 0 229I.2)-I 186 __4.d_.ll_..-. 1.1- \u00E2\u0080\u00A2- \u00E2\u0080\u00A2- - \u00E2\u0080\u0094\u00E2\u0080\u00A2 -. ISO 200 m/t JI8U2) Figure 29. Mass Spectrum of Alloimperatorin methyl ether diol ( 3 ) 8 2 O C H m /e 318 m /e 229 + C16 H13\u00C2\u00B04 m /e 241 Figure 30. Proposed Fragmentation of Alloimperatorin methyl either (7) 8 2 and Diol C3) 8 2 The mass spectrum of alloimperatorin methyl ether diol (3) does not show the strong M-15 peak but instead a very strong peak at m/e 229 for the fragment produced by cleavage 8 to the aromatic ring, as conjugation has been removed from the side chain. Barnes and Occolowitz79 have stated that furanocoumarins follow the coumarin pattern of loss of the carbonyl group to give benzofuran type ions. The remainder of the heterocyclic oxygens are lost as carbon monoxide yielding ions of lower abundance. They have shown that when i t is possible to write quinoid structures for ions derived by loss of a methyl radical, there i s no fission between the aromatic ring and the ether oxygen atom. Thus compounds such as xanthotoxin (55) do not readily lose the remaining two oxygen atoms after the carbonyl i s lost (Figure 31). Figure 31. Proposed Fragmentation of Xanthotoxin (55) in the Mass Spectrometer 7 9 From this brief discussion i t i s evident that coumarins exhibit very characteristic spectral properties thereby allowing the identification of the basic coumarin skeleton with considerable certainty by spectroscopic methods. It w i l l be seen i n the following discussion that spectroscopy has played a s i g n i f i c a n t r o l e i n determining the structures of the various components of Thamnosma montana. Unknown I The molecular formula of unknown I (mp 61-62\u00C2\u00B0) as determined by elemental analysis and high r e s o l u t i o n mass spectrometry was ^24^30^3\" Me OH The uv spectrum of I (A 206.5, 251 (sh) and 323 nm) was e s s e n t i a l l y r max ' 3 i d e n t i c a l to the uv data reported f o r 7-methqxy-4-methylcoumarin 8 5 thus suggesting the presence i n I of the same 7-alkoxycoumarin chromophore. The i r spectrum of I contained adsorptions at 1716 and 1610 cm 1 with a weak band at 1580 cm 1 i n d i c a t i n g the presence of the ct-pyrone system. A strong band at 835 cm 1 was suggestive of the presence of a t r i s u b s t i t u t e d double bond 8 5 or perhaps i f considered i n conjunction, with a medium i n t e n s i t y band at 895 cm could suggest the presence of a 1,2,4-trisub-s t i t u t e d benzenoid system 8 7. The nmr spectrum of I (Figure 32) proved p a r t i c u l a r l y informative. Two doublets appearing at x2.47 and T3.85 (each with J=9.6 Hz) were assigned to H(4) and H(3) r e s p e c t i v e l y of a coumarin system. Also observed i n the aromatic region of the spectrum was a broadened doublet centered at x2.72 (J=9.5 HZ) assigned to H(5) and a two proton mu l t i p l e t at x3.20 assigned to H(6) and H(8) of a coumarin system. These s p e c t r a l dataware therefore conclusive evidence of a 7-oxygenated coumarin. The remaining signals were assigned to protons on a C^ ,. side chain attached v i a the ethereal oxygen atom i n the 7-position. In the o l e f i n i c region a t r i p l e t at T4.56 (J-6 Hz) and a two-proton m u l t i p l e t at x4.93 suggested the presence of Figure 32. Nmr Spectrum of Umbelliprenin (83) three double bonds which, along with the seven degrees of unsaturation involved with the coumarin ri n g system would s a t i s f y the ten degrees of unsaturation required by the molecular formula. A two proton doublet at x5.44 (J=6 H Z ) was assigned to a methylene group which was adjacent to both the ethereal oxygen atom and the side chain double bond bearing the o l e f i n i c proton absorbing at .4.56. - At higher f i e l d were absorptions assigned as a methylene envelope (T7.8 to 8.2) i n t e g r a t i n g for eight protons and three s i n g l e t s at T8.26 (3H), 8.36 (3H) and 8.43 (6H) assigned to four methyl groups. The chemical s h i f t s of the methylene and methyl group signals indicated that these protons were a l l y l i c . In summary, the nmr evidence strongly suggested that unknown I was a 7-oxygenated coumarin with a farnesyl ether side chain. The mass spectrum of I (Figure 33) thoroughly supported the s t r u c t u r a l assignment. The parent peak at m/e 366 was weak as has been found for a l k y l ether coumarins by Kutney and co-workers 8 2. The base peak'at m/e 162 could be assigned to the fragment where the e n t i r e side chain was removed to give an umbelliferone type ion. The corresponding side chain was considered to be the fragment responsible for a strong peak at m/e 204. Peaks at /e 134, 106 and 105 are compatible with the fragmentation pattern of the umbelliferone ion as shown i n Figure 34. The data allowed the assignment of structure 83 for unknown I. The structure i s that of a known coumarin 8 8 umbelliprenin ( 8 3 ) . The a l l trans configuration of the double bonds has been assigned as recently proved for 83 by Bates, Shaube and Soucek 8 9. m 83 /e. 366 + \u00E2\u0080\u00A2 m /e 162 m /e 204 m /e 134 + \u00E2\u0080\u00A2 C7H50 + m /e 105 C-,H,0 7 6 + m /e 106 Figure 34. Proposed Fragmentation of Unknown I (Umbelliprenin (83)) in the Mass Spectrometer The spectral evidence was considered adequate to prove the identity of unknown I as umbelliprenin (83). However since an authentic sample could not be readily obtained, and as in later work it became necessary to obtain more 83 than could be conveniently isolated from the plant, a synthesis was undertaken. Synthesis of Umbelliprenin (83) (Figure 35) The commercial farnesol obtained for this purpose was exposed to vapor phase chromatography (vpc). This investigation revealed that this sample was a mixture of three isomers in accord with the results of Bates et a l \u00C2\u00B0 . The respective retention times were in accord with the mixture; cis-cis (6%), cis-trans and trans-cis (34%) and trans-trans (60%). Preparative vpc allowed isolation of pure trans-trans-farnesol (61b). The nmr spectrum of the isolated material showed two methyl peaks at T8.33 and 8.40 of almost equal intensity as reported for the trans-trans isomer9 0. Umbelliferone (45) was purified by sublimation and the sodium salt prepared by reaction with an equimolar amount of sodium ethoxide in ethanol. The salt thus obtained was reacted, in dimethylformamide, with farnesyl bromide (prepared from pure trans-trans alcohol by the action of phosphorus tribromide in ether) to give umbelliprenin (83) mp 57-59\u00C2\u00B0 , in 44% yield, identical with the natural material referred to as unknown I. + -Figure 35. Synthesis of Umbelliprenin (83) Elemental analysis and high resolution mass spectrometry of unknown II supported a molecular formula, C^H^O^,. The uv spectrum of unknown II (*^\u00C2\u00B0H 219, 249, 257.5, 267 and 308 nm) suggested that the compound was a 5-oxygenated furanocoumarin derivative as the spectrum was essen-t ia l ly identical to the reported spectrum of 5-geranyloxypsoralen (86a)67 86a; R=geranyl 86b; R=H The i r spectrum contained bands at 1718, 1612 and 1582 cm 1 characteristic of the ct-pyrone ring and a band at 819 cm 1 suggestive of a trisubstituted double bond, although adsorptions in this region have been assigned to the furan system in furanocoumarins91. The nmr spectrum of II (Figure 36) was particularly informative. Doublets at T1.90 (J=10.0 Hz) and at T3.81 (J=10.0 Hz) were assigned to H(4) and H(3) respectively of a coumarin system. A doublet at T2.48 (J=2.4 Hz) and a multiplet at T3.10 were assigned to H(7) and H(6) respectively, the furan protons of a furanocoumarin. A broadened signal at T2.92 was assigned to H(8) of the system revealed in 86. Under high resolution the signal assigned to H(4) was revealed to be a doublet of doublets (J=10.0 Hz and 0.5 Hz). The broadened signal at T2.92 was resolved as an overlapping doublet of doublets (J=0.5 Hz and 1.0 Hz). Furthermore the signal assigned to H(6) was resolved as a doublet of doublets (J=1.0 Hz and 2.45 Hz). These extra splittings can be assigned as due to long range coupling through the aromatic ring in the familiar \"zig-zag pattern\" 7 7. Figure 36. Nmr Spectrum of Isoimperatorin (87) The higher f i e l d s i g n a l s i n the nmr spectrum were assigned to protons associated with a side chain joined through the ethereal oxygen at p o s i t i o n 5. The molecular formula of II requires that the side chain possess f i v e carbon atoms with one degree of unsaturation ( i . e . C-H^). Thus the broad t r i p l e t at T4.51 (J=3.5 H Z ) was assigned to an o l e f i n i c proton on a t r i s u b s t i t u t e d double bond and coupled to an adjacent -methylene group which absorbs as a doublet (J=3.5 Hz) at T5.13. The deshielded p o s i t i o n of t h i s methylene absorption requires that i t be adjacent to the ethereal oxygen atom. Two v i n y l methyl s i n g l e t s at T8.22 and 8.30 completed the spectrum. These data require, that unknown II have structure 87, 5-(3 f-methyl-2'-butenyloxy)-psoralen. The mass spectrum of I I o f f e r e d further confirmation of the proposed structure. Thus the molecular ion peak at m / e 270 was weak, while the base peak at m / e 69 could be assigned to the fragment formed by cleavage of the side chain at the s i t e of the ethereal oxygen atom. The remainder of the spectrum was e s s e n t i a l l y i d e n t i c a l to that of bergaptol ( 8 6 b ) 8 2 . Thus the peak at m / e 202 corresponds to cleavage of the side chain to y i e l d a b e r g a p t o l - l i k e ion and subsequent fragments a r i s e from the breakdown of t h i s ion. The structure 87 i s that of a known furanocoumarin, i s o i m p e r a t o r i n 9 2 . An authentic sample was not r e a d i l y a v a i l a b l e but i t i s f e l t that the spectroscopic evidence provides conclusive proof of structure. 87 Elemental analysis and high resolution mass spectrometry of unknown III established the molecular formula as C, _H,,0,. The presence of the 15 14 4 7-alkoxycoumarin chromophore was indicated by the uv spectrum (Figure 37) MeOH (X m a x 227.5, 253(sh), 297(sh) and 327nm) which was essentially identical with the uv spectrum of suberosin (9) 9 3 . The i r spectrum of unknown III (Figure 38) contained the characteristic coumarin absorptions at 1722, 1610 and 1552 cm \ In addition, absorptions at 3075 and 903 cm ^ suggested the presence of a terminal methylene group8 6. The nmr spectrum (Figure 39) again proved to be very instructive. The aromatic region of the spectrum contained the characteristic coumarin doublets at T2.42 (J - 9 . 4 Hz, H(4)) and T3.97 (J=9.4 Hz, H(3)). Also in the aromatic region a broadened singlet at T2.77 and a sharp singlet at x3.25 were readily assigned to H(5) and H(8) respectively. In conjunction with the uv and i r evidence,these data allowed the proposal that unknown III possesses a 6-substituted-7-alkoxycoumarin system. The presence of the coumarin doublets rules out substitution in either the 3- or the 4-position and the lack of significant splitting of the other two aromatic proton signals requires a para and thus 5,8-disposition of these two protons. A sharp three proton singlet at T6.13 was assigned to an aromatic methoxyl group .which is of necessity placed in the 7-position. 200 250 300 350 nanometers Figure 37. Uv Spectrum of Thamnosmin (90) ON CT-. 3 0 0 0 W A V E N U M B E R ( C M ' 1 2 0 0 1 0 0 0 W A V E N U M B E R ( C M ' ) Figure 38. Ir Spectrum of Thamnosmin (90) It follows that the remaining nmr signals must be assigned to protons involved with a side chain attached to the 6-position of the coumarin system. Thus one proton multiplets at x4.85 and 4.97 were assigned as o l e f i n i c proton signals and a three proton multiplet at x8.25 was t e n t a t i v e l y assigned to a v i n y l methyl group. A p a i r of apparent doublets at x5.93 (J=2 Hz) and 6.80 (J=2 Hz) completed the spectrum. Considering the molecular formula of unknown I I I , i t i s evident that the side chain i n the 6-position can be represented by the p a r t i a l formula Cr^HyO. Also, the molecular formula demands nine degrees of unsaturation i n the molecule of which seven degrees can be assigned to the coumarin system, thereby leaving two s i t e s of unsaturation i n the side chain. With t h i s i n mind high r e s o l u t i o n nmr spectra were obtained and spin decoupling studies were c a r r i e d out. Under high r e s o l u t i o n (Figure 39) the methyl mu l t i p l e t at x8.25 was resolved as a doublet of doublets (J=1.5 and 0.9 Hz). The o l e f i n i c proton m u l t i p l e t at x4.85 remained complex with at least seven broad l i n e s being apparent (half-height width of t h i s peak was 4 Hz). The higher f i e l d o l e f i n i c m ultiplet was resolved as a symmetrical quintet (J=1.5 Hz). I r r a d i a t i o n at the frequency of the methyl s i g n a l caused the collapse of the o l e f i n i c quintet to a doublet (J=1.5 Hz) while the lower f i e l d o l e f i n i c m u l t i p l e t sharpened to an apparent doublet of doublets (Figure 40). It was thus evident that the side chain double bond was dis u b s t i t u t e d with one of the substituents being a methyl group. The o l e f i n i c protons are coupled to each other with a coupling constant of 1.5 Hz and therefore must be geminal 7 1. Jackman 9 4 has noted that spin-spin couplings across,-prppene 1 I 250 I T T4.85 i . l . l )\u00E2\u0080\u0094-T\u00E2\u0080\u0094-\u00E2\u0080\u0094\u00E2\u0080\u00941 I I I . 1 I 1 1' \u00E2\u0080\u00A2 I T8,25 >-H> 0 CPS VO double bonds generally follow the relationship J cis > J trans by 0.3 to 0.6 Hz. Thus the olefinic proton absorbing at x4.85 with a coupling constant of 0.9 Hz to the methyl group would be situated in a trans relationship to the methyl group while the cis proton at x4.97 reveals a coupling to the methyl group of 1>5 Hz. On this basis a further partial structure of unknown III could be assigned as shown below. The small additional s p l i t t i n g evident in the x4.85 olefinic multiplet would appear to be due to small coupling with a trans a l l y l i c proton. The residual fragment, C^^O, containing one degree of unsaturation which s t i l l remains to be assigned, must be placed between the isopropyl-idene group and the aromatic ring. Only four basic arrangements of these atoms need be considered; an enol, an a-methylene ketone, an a-methyne aldehyde, or an epoxide. The enol possibility can be quickly eliminated as there are no exchangeable proton signals in the nmr spectrum. Also, an enolic group would provide further conjugation with the coumarin system and thus change the uv spectrum significantly from that observed. The aldehyde function can be eliminated as there are no characteristic low f i e l d signals in the nmr spectrum. A ketonic carbonyl adjacent to the aromatic ring would affect the uv spectrum and therefore can be eliminated from serious consideration. The remaining p o s s i b i l i t i e s are portrayed in structures 88, 89 and 90. Structure 88 is not entirely consistent with, the nmr spectrum since the benzylic methylene signal would be expected to occur as a low field singlet. On the other hand structures 89 or 90 could f i t the data well, although the latter would be preferred on biogenetic grounds. Further examination of the nmr spectrum of III under high resolution and spin decoupling experiments allowed the assignment of structure 90 to unknown III. The apparent doublets at x5.93 and 6.80 were assigned to the epoxide protons. Furthermore, high resolution revealed the x5.93 signal to be a doublet of doublets (J=2.0 Hz and 0.65 Hz) while the higher field doublet was somewhat broadened (J=2.0 Hz). Intuitively the additional splitting in the x5.93 signal would suggest that this signal is due to the proton which is involved in al lyl ie coupling with H(4'a). On the other hand one would also expect the benzylic proton to absorb at lower field than the al lyl ie epoxide proton. This paradox was resolved by double irradiation experiments (Figures 41 and 42). Coupling between the epoxide protons was confirmed since when the signal at x5.93 was irradiated the doublet at x6.80 collapsed to a singlet. When the olefinic proton H(4'a) signal at x4.85 was irradiated, the secondary splitting in the x5.93 signal remained unchanged, while the broadened doublet at x6.80 sharpened noticeably. Examination of the aromatic region of the spectrum under high resolution shed some l i g h t on the nature of the observed couplings. The s i n g l e t at x2.77 assigned to H(5) was observed to be quite broad and when t h i s s i g n a l was i r r a d i a t e d the doublet of doublets at x5.93 collapsed to a doublet (J=2.0 Hz).. Thus t h i s s p l i t t i n g (0.65 Hz) was due to long range coupling between H(5) and H ( l ' ) . Such benzylic coupling i s not unknown and Jackman and Sternhell state that the coupling of a benzylic proton to the ortho proton i s generally i n the range 0.6-0.9 Hz with coupling to the meta proton being considerably s m a l l e r 9 5 . The coupling constant of 2.0 Hz observed for the epoxide protons i s i n d i c a t i v e of a trans r e l a t i o n s h i p between these p r o t o n s 9 6 . Thus the structure of unknown III i s that represented byf.structure 90. The mass spectrum of III (Figure 43) was i n f u l l agreement with the proposed structure. In addition to the molecular ion peak at m/e 258, the base peak at m / e 229 (C-^H^O^ as determined by high r e s o l u t i o n mass spectrometry) represents a loss of CHO from the molecular ion. This fragmentation i s best explained by a rearrangement (Figure 44) which has 9 7 been proposed by D j e r a s s i et a l . for a l i p h a t i c epoxides.' The resultant ion 91 would be highly s t a b i l i z e d thereby explaining i t s abundance. Peaks at m / e 214, 189, 159 and 131 are best explained by the sequence pictured i n Figure 42, a sequence previously reported, i n part, for prenylated coumarins 7 9. 23 Unknown III was o p t i c a l l y active ([ct] D (CHC13) = -17.3\u00C2\u00B0) and the ORD curve showed a minimum at 219 nm ([; uv X ^ e 0 H (e) 206.5 ' I Max (29,600), 251 (sh)(2500), 323 (15,700); nmr (100 MHz) i n CDC13, TMS lock, 2.47 (IH, doublet, J-9.5 Hz, H(4) of coumarin), 2.72 (IH, broadened doublet, J=9.5 Hz, H(5) of coumarin), 3.20 (2H, m u l t i p l e t , H(6) and H(8) of coumarin), 3.85 (IH, doublet, J=9.6.Hz, H(3) of coumarin), 4.56 (IH, broad t r i p l e t , J-6 Hz, o l e p h i n i c ) , 4.93 (2H, broad m u l t i p l e t , o l e p h i n i c ) , 5.44 (2H, doublet, J=6.0 Hz, 0-CH--CH-C), 7.8 to 8.2 (8H, methylene envelope), 8.26, 8.36 (6H, two s i n g l e t s , two v i n y l CH'), 8.43 (6H, s i n g l e t , two v i n y l CH_); mass spectrum m/e 366 (M), 204 (M-162), 162 (M-204), 134 , 106', 3:05. Anal. Calcd. for C ^ f t ^ ^ : C, 78.65; H, 8.25. Found C, 78.52; H, 8.10. Synthesis of Umbelliprenin (83) Purification of trans, trans-farnesol (61b) 9 0 Analysis of a sample of commercial farnesol by glc (column : 30% carbowax 20M, on 60/80 mesh chromosorb W, 10 f t x 3/8 in, helium flow rate 172 m\"'\"/min, 210\u00C2\u00B0) revealed three major components at retention times 9 min (6%), 11 min (34%) and 12 min (60%). Preparative glc on the same column (at 230\u00C2\u00B0) and isolation of the last half of the last peak, allowed isolation of trans1, trans-f arnesol (61b) (205 mg, retention time 11 min on the same column, flow rate 172 ml/min, 235\u00C2\u00B0); nmr (60'MHz) in C0C13, 4.60 (IH, broad t r i p l e t , J=3.3 Hz, CH2-CH=CR2), 4.88 (2H, broad multiplet, two CH=CR2), 5.95 (2H, doublet, J=3.3 Hz, 0-CH_2-CH=CR2), 7.6-8.2 (9H, methylene envelope and Oil), 8.33, 8.40 (12, two singlets of almost equal intensity, four vinyl C I 3 ) . Trans, trans-farnesyl bromide (84) 8 9 Trans, trans-farnesol (61b) (205 mg, 0.925 mmol) was dissolved in dry ether (10 ml) and cooled to -78\u00C2\u00B0. Phosphorus tribromide (0.096 ml; 0.960 mmol) was added slowly, the reaction allowed to come to room temperature and stirred at room temperature for 6 hours. The reaction mixture was washed with 5% aqueous sodium bicarbonate (20 ml), dried over anhydrous sodium sulfate and fil t e r e d . The resultant solution was used in the next r e a c t i o n . Umbelliprenin (83) P r a c t i c a l grade umbelliferone (45) was p u r i f i e d by sublimation (160-165\u00C2\u00B0 at 0.1 mm). Sodium metal (25' mg; .1.09 mmol)'was-'dissolved i n absolute ethanol (5 ml) and a s o l u t i o n of sublimed 48 (162 mg; 1.00 mmol) i n absolute ethanol (5 ml) was added. The solvent was evaporated i n vacuo y i e l d i n g the white s a l t (85). The l a t t e r was dissolved i n dry dimethyl formamide (10 ml) and to t h i s s o l u t i o n was added the ethereal s o l u t i o n of farnesyl bromide (84) (0.925 mmol, t h e o r e t i c a l ) from the previous reacti o n . Most of the ether was evaporated i i i vacuo and the resultant mixture was s t i r r e d at room temperature under nitrogen f o r 14 hours. Water (25 ml) was then added and the s o l u t i o n extracted with hexane (3 x 50 ml), the hexane phase was dried over anhydrous sodium s u l f a t e and evaporated to y i e l d a colourless o i l y residue (260 mg). This residue was chromatographed on two preparative t i c plates e l u t i n g with solvent A and the purple (uv) band at R^ coincident with unknown I was i s o l a t e d as a colourless o i l (200 mg). C r y s t a l l i z a t i o n of t h i s material from l i g h t petroleum ether afforded umbelliprenin (83) as white needles (150 mg; 44% y i e l d ) mp 58-59\u00C2\u00B0; mixed mp with unknown I, 58-60\u00C2\u00B0. Infrared and uv spectra i d e n t i c a l with those of unknown I. Anal. Calcd. f o r C^H^O^ : C, 78.65; H, 8.25. Found: C, 78.42; H, 8.10. Properties of Unknown II (isbimperatbrih (87)) Unknown II from f r a c t i o n b was c r y s t a l l i z e d from e t h y l acetate-hexane as needles mp 97.0-98.0\u00C2\u00B0 reforming plates mp 105-108\u00C2\u00B0 ( l i t . 9 3 : m p 109\u00C2\u00B0); i r (KBr) 1718, 1612, 1.582 (a-pyrone), 819 (RCH=CR2 or furan of furano-MeOH coumarin); uv A -(e) 219 (27,800), 249 (18,800), 257.5 (16,700), TUcLX 267 (15,900), 308 (14,100); nmr (100 KEzj i n CDC1 , \"TMS lock, 1.90 (IH, doublet of doublets, J=10.0 Hz and 0.5 Hz, H(4) of furanocoumarin), 2.48 (IR, doublet, J=2.4 Hz, H(7) of furanocoumarin), 2.92 (IH, doublet of doublets, J \u00E2\u0080\u00A2'. 0.5 Hz and ^ 1.0 Hz, H(8) of furanocoumarin, 3.10 (IH, doublet of doublets, J=1.0 Hz and 2.45 Hz, H(6) of furanocoumarin), 3.81 (IH, doublet, J=10.0 Hz., H(3) of furanocoumarin), 4.51 (IH, broad t r i p l e t , J=3.5 Hz, CH2-CK=CR2), 5.13 (2K, doublet, J=3.5.Hz, 0-CH2-CH=C), 8.22, 8.30 (6H, two s i n g l e t s , C=C(CH 3) 2); mass spectrum m / e 270 (M), 202 (M-68), 174, 145, 118, 89, 69. Anal. Calcd. for C^H.,.0. : C, 71.08; H, 5.22. Found: C, 70.90; H, 5.04. 16 14 4 High r e s o l u t i o n molecular weight determination. Calcd. for C-^H^O^ : . 270.089. Found: 270.090. Properties of Unknown I I I (thamnosmin (90)) Unknown III from f r a c t i o n s c and d c r y s t a l l i z e d from et h y l acetate-hexane as colourless p l a t e s , mp 101-104\u00C2\u00B0; .[a-]^3 (CHC13)= - 17.3\u00C2\u00B0; 0RD (MeOH, c', 0.00106) []2ig - 101,000\u00C2\u00B0; i r (KBr) 1722, 161.0', L55'2 (a-pyrone), 3075, M P O H 903 (terminal methylene); uv A (e) 227.5 (20,100), 253 (sh) (6,420), iricix 297 (sh) (7,500), 327 (13,300); nmr (100 MHz) i n CDC13, TMS lock, 2.42 (IH, doublet, J=9.4 Hz, H(4) of coumarin), 2.77 (IH, broad s i n g l e t , H(5) of coumarin), 3.25 (IH, s i n g l e t , H(8) of coumarin), 4.85 (IH, m u l t i p l e t , width at h a l f height 4 Hz, |^C=C^ H3) , 4.97 (IH, quintet, J-1.5 Hz, it R H _CH \u00C2\u00A7^C=C^ 3), 5.93 (IH, doublet of doublets^ J=2.0 Hz and 0.65 Hz, ben z y l i c epoxide proton), 6.13 (3H, s i n g l e t , aromatic OCR^), 6.80 (IH, broadened doublet, J=2.0 Hz a l l y l i c epoxide proton), 8.25 (3H, doublet of doublets, J=1.5 Hz and 0.9 Hz, ^C^RCH^) ; double i r r a d i a t i o n nmr; i r r a d i a t e at 8.25, 4.85 multiplet becomes apparent doublet of doublets, 4.97 quintet becomes doublet (J=1.5 Hz); i r r a d i a t e at 5.93, 6.80 doublet becomes s i n g l e t ; i r r a d i a t e at 2.77, 5.93 doublet of doublets becomes doublet (J=2.0 Hz); mass spectrum m/e 258 (M), 229 (M-29), 214, 189, 159, 131. Anal. Calcd. f o r C 1 5 H 1 4 0 4 : C, 69.75; H, 5.46. Found: C, 69.93; H, 5.71. High r e s o l u t i o n molecular weight determination. Calcd. for C^H^O^ : 258.089. Found: 258.089. Calcd. for C-.H.-.O. : 229.086. Found 229.087. 14 13 3 Acid Hydrolysis of Thamnosmih (90) i n Methanol- 9 8 . Thamnosmin (20.1 mg; 0.081 mmol) was dissolved i n methanol (3 ml) and to this s o l u t i o n was added 10% aqueous s u l f u r i c acid (2.2 ml). The s o l u t i o n was gently refluxed f o r 4 hours, the reaction mixture poured into water (20 ml) and extracted with chloroform (3 x 20 ml). The chloro-form extract was washed with water (10 ml), dried over anhydrous sodium s u l f a t e , f i l t e r e d and evaporated to dryness. T i c examination of the extract revealed three components with a less polar purple (uv) spot as apparently the major component. Preparative t i c e l u t i n g with solvent B allowed i s o l a t i o n of t h i s component (8.4 mg) as an o i l y s o l i d which c r y s t a l l i z e d from benzene-hexane as colourless needles, mp 120-123\u00C2\u00B0. R e c r y s t a l l i z a t i o n from the same solvent y i e l d e d the major hydrolysis product (93) (4.0 mg) as colourless needles, mp 121.5-124.0\u00C2\u00B0; i r (KBr) 3495 (OH), MeOH 3080, 905 (terminal methylene), 1727, 1616, 1562 (a-pyrone); uv A (e) 228.5 (16,100), 243 (sh) (5,630), 252 (4,425), 297.5 (sh) (7,450), 326 (12,800); nmr (100 MHz) i n CDC13 (region 5.0 to 7.5 expanded under high r e s o l u t i o n v i a time averaging, 14 scans and the region 5.0 to 6.1 expanded under high r e s o l u t i o n v i a time averaging, 11 scans) TMS lock, 2.37 (IH, doublet, J=9.5 Hz, H(4) of coumarin), 2.58 (IH, s i n g l e t , H(5) of coumarin), 3.25 (IH, s i n g l e t , H(8) of coumarin), 3.76 (IH, doublet, J=9.5 Hz, H(3) of H CH coumarin), 5.27 (IH, apparent quintet, J=l.5 Hz, \u00E2\u0080\u0094^ C=CC,, 3), 5.31, 5.32 H K (2H, superimposed, doublet of doublets, J=6.4 Hz, and ^ 0.6 Hz, ben z y l i c H CH proton and mul t i p l e t , C^=CC^ 3), 5.9.6 (IH, broad doublet b ecoming sharp ti K doublet on addition of D 20, J=6.6 Hz, C=CR-CH-0H), 6.16 (3H, s i n g l e t , aromatic 0CH 3), 6.73 (3H, s i n g l e t , a l i p h a t i c 0CH 3), 7.16 (IH, doublet disappearing on addition of D 20, J=3.2 Hz, R2CH-0H), 8.24 (3H, mul t i p l e t , v i n y l CH 3); mass spectrum m/e 290 (M) 258 (M-42), 219 (M-71), 189,159, 131. High r e s o l u t i o n molecular weight determination: Calcd. f o r C,,H.lo0r.: l b l o J 290.115. Found: 290.114. Properties of Unknown LV (Phellopterin ( 9 6 ) ) Unknown IV from fract i o n s c and d c r y s t a l l i z e d from ethyl acetate-hexane as coarse c r y s t a l s mp 92-94\u00C2\u00B0. Three r e c r y s t a l l i z a t i o n s from the same solvent raised the melting point to 96-98\u00C2\u00B0 ( l i t . 10'\u00C2\u00B0'mp 102\u00C2\u00B0), mixed mp with authentic p h e l l o p t e r i n ^(96)- (obtained from Dr. D.L. Dreyer) 95-98\u00C2\u00B0; M P O H i r (KBr) 1730, 1606, 1593 (a-pyrone) 820 (RCH=CRj; uv r (e)223 (25,800), z max 241 (14,600), 248 (14,600), 268.5 (16,700), 312 (11,500); nmr (100 MHz) i n CDC13, TMS lock, 1.95 (IH, doublet, J=10 Hz, H(3) of furanocoumarin), 2.45 (IH, doublet, J=2.2 Hz, H(7) of furanocoumarin), 3.10 (IH, doublet, J=2.2 Hz, H(6) of furanocoumarin), 3.79 (IH, doublet, J=10 Hz, H(4) of furanocoumarin), 4.44 (IH, broad t r i p l e t , J=7 Hz, CH2~CH=CR2), 5.20 (2H, doublet, J=7 Hz, 0CH2-CH=C), 5.88 (3H, s i n g l e t , aromatic 0CH 3), 8.29, 832 (6H, two s i n g l e t s , two v i n y l C H ) ; mass spectrum m/e 300 (M), 285 (M-15), 270 (M-30), 232 (M-68), 217, 189, 161, 133. Anal. Calcd. f o r O1-JH.1,0C : C, 67.99; H, 5.37. Found: C, 67.98; H, 5.36. 17 16 5 Properties of Unknown V (Alloimperatorin Methyl Ether Epoxide (97)) Unknown V from f r a c t i o n G c r y s t a l l i z e d from benzene-hexane as coarse crystals, mp 101-103\u00C2\u00B0. Careful r e c r y s t a l l i z a t i o n from the same solvent y i e l d e d an a n a l y t i c a l sample, mp 103-104\u00C2\u00B0 ( l i t . 6 mp 105-106.5\u00C2\u00B0); [ a ] 2 3 = .< + 1\u00C2\u00B0 ; i r (KBr) 1717 (C=0), 1588 (C=C); uv A M e 0 H (e) 221 (25,200), i> max 244 (sh) (19,900), 251.5 (22,800),. 266 (20,350), 305 (13,000); nmr (100 MHz) i n CDCly TMS lock, 1.82 (IH, doublet, J=10 Hz, H(4) of furanocoumarin), 2.33 (IH, doublet, J=2 Hz, H(7) of furanocoumarin), 3.11 (IH, doublet, J=2 Hz, H(6) of furanocoumarin), 3.64 (IH, doublet, J=10 Hz, H(3) of furanocoumarin), 5.78 (3H, s i n g l e t , aromatic OCH^), 6.5-7.2 (3H, ABC m u l t i p l e t , J =14.5 Hz, J . =3 Hz, J , =7.5 Hz; b e n z y l i c AB AC BC methylene and epoxide proton); 8.51, 8.70 (6H, two s i n g l e t s , C*^C(CH 3) ^ ) ; mass spectrum m/e 300 (M), 285 (M-15), 271 (M-29), 257 (M-43), 229 (M-71), 201, 199, 186, 171, 158. Anal. Calcd. f o r C.-H^O,. : C, 67.99; H, 5.37. Found: C, 67.91: H, 5.48. 17 16 5 High r e s o l u t i o n molecular weight determination. Calcd. for C^H.^,0,. : 300.100. Found: 300.102. Synthesis of Alloimperatdrin Methyl Ether Epoxide ( 9 7 ) 6 A s o l u t i o n of alloimperatorin methyl ether (7) (53 mg; 0.187 mmol) i n chloroform (5 ml) was cooled i n an i c e bath and to i t was added a s o l u t i o n of m-chloroperbenzoic acid (36 mg; 0.204 mmol) i n chloroform (5 ml). The reaction mixture was maintained at 0\u00C2\u00B0 with s t i r r i n g f o r 5 hours at which time the s o l u t i o n was washed with 5% aqueous sodium carbonate (20 ml), then with water (10 ml), dried over anhydrous sodium s u l f a t e and the solvent reduced to ^ 0.5 ml. The r e s u l t i n g s o l u t i o n was chromatographed on a preparative t i c p l a t e , e l u t i n g with solvent B, and the major, green (uv), band i s o l a t e d by extraction with chloroform. The residue (47 mg; 84% y i e l d ) was c r y s t a l l i z e d from benzene-hexane to y i e l d coarse white c r y s t a l s of 67, mp 103-104\u00C2\u00B0, mixed mp with unknown V 5114-116\u00C2\u00B0, i n f r a r e d spectrum superimposable of that of unknown V. Properties of unknown VI (psoralene (68)) Unknown VI from f r a c t i o n c and-d.crystallized from ethyl acetate-hexane as colourless needles, mp 163.5-164.5\u00C2\u00B0 ( l i t . 0 2 mp 1 6 3 \u00C2\u00B0 ) , mixed mp with authentic sample of psoralene (66) (obtained from Dr. D.L. Dreyer) 163-164.5\u00C2\u00B0, i r (KBr) 1714, 1630, 1574 (a-pyrone); uv A M e 0 H (e) 211 r y max (17,900), 240 (23,800), 245 (24,600), 289 (11,000), 325 (6,660); nmr (100 MHz) i n CDC13, TMS lock, 2.27 (IH, doublet of doublets, J=9.5 Hz and 0.5 Hz, H(4) of furanocoumarin), 2.37, 2.39 (2H, res p e c t i v e l y , doublet, J=2 Hz, H(7) of furanocoumarin and s i n g l e t , H(5) of furano-coumarin), 2.59 (IH, m u l t i p l e t , H(8) of furanocoumarin), 3.22 (1H, doublet of doublets, J=2.0Hz and.1.0 Hz,-H(6) of furanocoumarin); mass spectrum m/e 186 (M>, 158 (M-28), 130 (M-56), 102 (M-84). Anal. Calcd. f o r C 1 1H,O n : C, 70:97; H, 3.25. Found: C, 70.65; H, 3.15. 11 6 3 -Infrared spectrum of authentic psoralene (68) superimposable on that of unknown VI. Properties of Unknown VII (Bergapten (67)) Unknown VII from f r a c t i o n d was c r y s t a l l i z e d from et h y l acetate-1 0 3 hexane as colourless needles, mp 186.5-188.5\u00C2\u00B0 ( l i t . - mp 191\u00C2\u00B0); mixed mp with authentic bergapten (65) (obtained from Dr. D.L. Dreyer) 186-188\u00C2\u00B0, MeOH i r (KBr) 1726, 1623, 1580 (a-pyrone); uv X (e) 221 (21,400), 248 (17,800), 257.5 (16,200), 267 (17,500), 309 (14,700); nmr (100 MHz) i n CDC13, TMS lock, 1.91 (IH, doublet of doublets, J=9.75 Hz and 0.6 Hz, H(4) of furanocoumarin), 2.47 (IH, doublet, J=2.5 Hz, H(7) of furano-coumarin), 2.92 (IH, mu l t i p l e t , H(8) of furanocoumarin), 3.05 (IH, doublet of doublets, 3=2.5 Hz and 1.0 Hz, H(6) of -furanocoumarin) 3.80 (IH, doublet, J=9.75 Hz, H(3) of furanocoumarin), 5 . 8 0 (3H, s i n g l e t , aromatic 0CH_3); mass spectrum m/e 216 (M), 201 (M-15) , 188 (M-28), 173, 145. Anal. Calcd. f o r C ^ H ^ : C, 66.67; H, 3.73. Found: C, 66.57; H, 3.80. Infrared spectrum of authentic 6.7 superimposable with that of unknown VII. Properties of Unknown VIII (Xanthotoxin ( 5 5 ) ) Unknown VIII from column f r a c t i o n e c r y s t a l l i z e d from e t h y l acetate-hexane as s l i g h t l y blue tinged needles mp 146-147\u00C2\u00B0 (litW^mp 146\u00C2\u00B0) , mixed mp with authentic xanthotoxin (74) (obtained from Dr. D.L.1 Dreyer) 140-142\u00C2\u00B0; M P O T T i r (KBr) 1710, 1615, 1582 (a-pyrone); uv \ ( E) 218 (20,700), 247 (20,700), 261 (sh) (13,200), 300.5 (11,450); nmr (100 MHz) i n CDC1 3, TMS lock, 2.27 (IH, doublet, J=9.75 Hz, H(4) of furanocoumarin), 2.34 (IH, doublet, J=2.3Hz, H(7) of furanocoumarin), 2.69 (IH, s i n g l e t , H(5) of furanocoumarin), 3.21 (IH, doublet, J=2.3 Hz, H(6) of furanocoumarin), 3.68 (IH, doublet, J=9.75 Hz, H(3) of furanocoumarin), 5.75 (3H, s i n g l e t , aromatic 0CH_3); mass spectrum m/e 216 (M) , 201 (M-15), 188 (M-28), 173 (M-43), 1 4 5 . (M-71). Anal. Calcd. f o r C ^ H ^ : C, 66.67; H, 3.73. Found: C, 66.31; H, 3.46. Infrared spectrum df authentic 5 5 superimposable on that of unknown VIII. DISCUSSION (PART II) Degradations of Coumarins From Thamnosma montana For biosynthetic investigations, u t i l i z i n g radioactively labeled precursors, to yield truly significant information, i t is necessary to degrade the molecules under study and determine the sites where the radioactivity resides. Many coumarins were isolated and their structures determined prior to the advent of spectroscopic techniques. The chemical literature contains numerous examples of structure elucidations of simple coumarins i n which degradations played a key role. From the present point of view the exhaustive degradative approach to structure elucidation appears tedious and unnecessary, due to the availability of physical techniques. However such work in the past had the added benefit that, in many instances, the chemistry of a compound under study was developed in considerable detail even before the structure was known. Such classical degradation reactions proved to be particularly useful in the degradations of coumarins from Thamnosma montana presented in this section. The coumarins, isopimpinellin (2), alloimperatorin methyl ether (7) and umbelliprenin (83) were selected for degradative study as they were present in reasonable quantities in the shoots of Thamnosma montana, were easily isolated and purified and they are representative examples of three types of coumarins. Isopimpinellin (2) is a relatively simple furano-coumarin whose chemistry can be correlated with that of other such furanocoumarins. Thus, although l i t t l e has been reported of the degradative chemistry of isopimpinellin (2) i t s e l f , related compounds such as bergapten (67), have been studied i n detail. Alloimperatorin methyl ether (7) i s also a furanocoumarin but with, the added complication of a C_ a l k y l side chain. Thus these two compounds o f f e r the p o s s i b i l i t y of studying both the mode of furan r i n g formation i n furanocoumarin biosynthesis and the purportedly r e l a t e d o r i g i n of isoprenoid side chains. As both mevalonic acid (57) and acetate have been previously implicated as precursors of such coumarins (see Introduction), umbelliprenin (83) could serve as an i n t e r n a l standard i n that i t s farnesol ether side chain i s almost c e r t a i n l y mevalonate derived (and therefore also acetate derived). Thus by demonstrating s p e c i f i c incorporation of such precursors i n t o 83, the f a c t that the precursors have been absorbed and u t i l i z e d by the plant can be established. With this i n mind, a series of degradations were developed with the important proviso that the reactions must be applicable to the small quantities of components that would be a v a i l a b l e from the radioactive incorporation experiments. Degradation of Isopimpinellin (2) Isopimpinellin (2), as an abundant component of Thamnosma montana shoots was selected for study of the biosynthesis of the furan r i n g i n furanocoumarins. As'previously, mentioned, \"Floss and Mothes 5 2 have found s p e c i f i c incorporation of C(4) of mevalonic acid (37) into C(7) of p i m p i n e l l i n (65). The generality of the conclusions offered by this r e s u l t was perhaps c a l l e d into question by the results of Rodighiero and co-workers 5 8, and more recently by Brown 6 2 who has .found acetate to be a much more e f f i c i e n t precursor of furanocoumarins than was mevalonic acid (57). Their findings that C(2) r a d i o a c t i v e l y labeled mevalonic acid (57) incorporated into simple furanocoumarins as effici e n t l y as did the C(4) labeled material served to further confuse the issue. To attempt to cl a r i f y these uncertainties, i t was decided than an intensive degradation procedure to determine the distribution of radioactivity in isopimpinellin (2) was needed. Floss and Mothes 5 2 found that ozonolysis of pimpinellin (65) yielded 2,4-dif omtyl-5, 6-dimethoxy-l, 3-dihydroxybenzene in poor yie l d . Previous workers 1 0 5 found that by ozonolysis, furanocoumarins could be converted to phenolic aldehydes where the furan ring had undergone degradation in preference to the pyrone ring. For isopimpinellin (2), the ozonolysis reaction appeared to offer a good poss i b i l i t y . It was expected that by u t i l i z i n g a controlled ozonolysis, the furan ring of 2 could be specifically degraded to yield the phenolic aldehyde (99) thus allowing determination of the percentage of the radioactivity which might reside in the 7-position. To accomplish such a selective ozonolysis OCH3 OCH3 2 99 i t was necessary to be able to add to the compound an accurately determined amount of ozone. This was achieved by saturating a volume of acetic acid with ozone. Aliquots were then removed and ozone concentration was determined by titr a t i o n of the iodine produced when the aliquot was reacted with aqueous potassium iodide. Ozonolysis of isopimpinellin (2) with a 60-70% molar excess of ozone, followed by redu c t i o n of the ozonide w i t h z i n c dust, y i e l d e d a product mixture whose major components were the s t a r t i n g m a t e r i a l and a more po l a r y e l l o w compound. I s o l a t i o n of the y e l l o w component by p r e p a r a t i v e t i c proved u n s a t i s f a c t o r y as recovery from the chromatoplates was poor (presumably due to a i r o x i d a t i o n of the compound on the s i l i c a g e l s u r f a c e ) . The r e a c t i o n mixture was i n s t e a d chromatographed on a s i l i c a g e l column. Although complete s e p a r a t i o n of 2 and the y e l l o w compound could not be achieved, the more p o l a r contaminents were thus removed. F r a c t i o n a l c r y s t a l l i z a t i o n of the f r a c t i o n s c o n t a i n i n g the y e l l o w compound afforded t h i s compound i n a pure s t a t e , mp 214-216\u00C2\u00B0 ( i n 25% y i e l d ) . On the b a s i s of the f o l l o w i n g data t h i s compound was c h a r a c t e r i z e d as 6-formyl-7-hydroxy-5,8-dimethoxycoumarin ( 9 9 ) . The molecular formula _ 1 2 _ l l 0 ^ 6 ' was e s t a b l i s h e d by elemental a n a l y s i s and hi g h r e s o l u t i o n Me OH mass spectrometry. The uv spectrum of 99 (A 275 nm) revealed the phenol i c nature of the compound, as, on a d d i t i o n of a l k a l i the spectrum MeOH showed a marked bathochromic s h i f t (A (+NaOH) 238, 269, 299 and 360 max nm). A c i d i f i c a t i o n of the uv sample produced a complex spectrum MeOH (A (+HC1) 208, 226, 263 and 320 nm). The i r spectrum of 99 i n d i c a t e d max v that the phenol was s t r o n g l y hydrogen bonded to the adjacent aldehyde carbonyl group. Thus no d i s c e r n a b l e h y d r o x y l absorption was i n evidence. In the carbonyl region of the spectrum, two absorptions were evident at 1758 and 1730 cm \ but an absorption at 1640 cm was considered to be due to the aldehyde carbonyl. This assignment was confirmed when 99 was methylated. The i r spectrum of the trimethoxy product (100) r e t a i n e d the two absorptions i n the carbonyl region but the 1640 cm absorption had s h i f t e d to 1689 cm \ r e f l e c t i n g the l o s s of hydrogen bonding. The nmr spectrum of 99 (Figure 51) fully confirmed the structure of this compound. Thus the spectrum was similar to that of isopimpinellin (2) but the signals due to the furan protons in the starting material were now absent and instead low f i e l d singlets at x - 0 . 2 3 and at x-2-.03 (disappearing on addition of D2O) were readily assigned to the aldehyde and phenol protons respectively. It was next of interest to determine the proportion of the radio-activity in isopimpinellin (2) which resided in the 6-position. The phenolic aldehyde (99) was f e l t to be the obvious starting material for such a degradation and i t was apparent that some oxidative procedure would be necessary. As phenols are generally unstable to oxidative condition, 99 was methylated with, methyl iodide and potassium carbonate in acetone. In various t r i a l reactions i t was found that maximum yields were obtained i f on completion, the reaction mixture was acidified and the products isolated by extraction. U t i l i z i n g such a work up procedure the main reaction product (colourless spot, uv) was isolated and purified by preparative t i c . Crystallization yielded pure 6-formyl-5,7,8-trimeth-oxycoumarin ( 1 0 0 ) , mp 1 5 2 . 5 - 1 5 4 . 0 \u00C2\u00B0(in 73% yield). This compound was again fully characterized by means of elemental and spectral analysis. The nmr spectrum which was particularly informative was fully consistent with the structure and as such was very similar to that of 9 9 , except there were now three methoxyl group singlets and the phenolic proton signal was absent. It was i n i t i a l l y decided that an obvious method for removal of the 6-formyl group would involve an oxidation of the aldehyde group to the acid function. Decarboxylation of this acid would then provide 5 , 7 , 8 -trimethoxycoumarin thus allowing calculation of the radioactivity present in the formyl group. However for such a scheme to be useful i t must proceed in good yield and provide a product which is easily purified. A comparative shortage of isopimpinellin (2) and the methylated product (100) provided some problems in that i t became necessary to develop the oxidation procedure with very small quantities of starting materials. Oxidation with Jones reagenti106 proved unsuccessful as starting material was recovered i f molar quantities (or slight excesses) of the reagent were used, while intractable mixtures were produced i f a large excess of the reagent was used. Oxidation of 100 with basic potassium permanganate, a method employed in the reported oxidation of 1,3-diformyl-2,4,5,6-tetramethoxybenzene to the corresponding d iac id 1 0 7 , yielded only the starting aldehyde (100), with no obvious oxidation occurring. Similarly, treatment of 100 with silver oxide yielded only the starting compound. The removal of the formyl group from 100 was achieved by application of an interesting reaction reported by Schonberg et a l 1 0 8 in their studies on the chemistry of bergapten (67). Utilizing a modification of the Dakin reaction 1 0 9 , they treated 6-formy 1-5, 7-dimethoxycoumarin (102) with hydrogen peroxide and s u l f u r i c a c i d and achieved good y i e l d s of the corresponding phenol (103). In the o r i g i n a l Dakin reaction, ortho- and para-hydroxybenzaldehydes were found to be converted to the corresponding diphenols when treated with basic hydrogen peroxide. The mechanism of the r e a c t i o n i s thought to involve a process s i m i l a r to the B a e y e r - V i l l i g e r rearrangement (Figure 52). D a k i n 1 0 9 suggested that a quinoid intermediate Figure 52. Proposed Mechanism for the Dakin Reaction was responsible for t h i s unusual r e a c t i o n as meta-hydroxybenzaldehydes and compounds with alkoxy groups instead of phenols, do not undergo t h i s conversion. In these instances conversion to the \"normal\" benzoic acid products occurs. In a c i d i c media the r e a c t i o n would be expected to follow a s i m i l a r mechanism (Figure 53) . Schbnberg e_t a l ^ u o also Figure 53. Proposed Mechanism for the Acid Catalyzed Dakin Reaction reported the conversion of the phenolic aldehyde (100), derived from bergapten (67), to the corresponding diphenol (104) under b a s i c Dakin conditions. I t was f e l t that the expected p o l a r i t y of the diphenol product produced under basic conditions could contribute to problems i n handling such a compound i n a small s c a l e r e a c t i o n and i t was considered to be more expedient to u t i l i z e the acid catalyzed reaction. Thus the methylated aldehyde (100) was treated under conditions as described by Schbnberg e_t _al. and the residue from the reaction mixture was chromatographed on preparative t i c . The major component thus i s o l a t e d was sublimed to y i e l d pure 6-hydroxy-5,7,8-trimethoxycoumarin (105), 100 105 mp 198.5-199.5\u00C2\u00B0 (77% yield). This compound gave analytical and spectral data fully consistent with the assigned structure. In the i r spectrum of 105, absorptions at 1720, 1611 and 1570 cm ^ indicated the continued presence of the coumarin system and an absorption at 3525 cm ^ could be MeOH assigned to the phenolic hydroxyl group. The uv spectrum (X 208 and 308 nm) exhibited a bathochromic shift on addition of MeOH base (X (+ NaOH) 248 and 317 nm) and the neutral spectrum was max r regenerated on acidification of the uv sample. This behaviour is fully consistent with the phenolic character of 105. The nmr spectrum of 105 (Figure 54) confirmed the structure. Thus the spectrum was essentially unchanged from that of the aldehyde (100) except that the aldehyde signal was absent and in its place was a broad one proton signal at T4.38 (disappearing on addition of D20), readily assigned to the phenolic proton. It was next of interest to devise a method which would allow determination of the distribution of radioactivity in the pyrone portion of isopimpinellin (2). Treatment of furanocoumarins with large excesses of ozone is known to cause degradation of both the furan and pyrone r ings 1 0 7 . Such a reaction on isopimpinellin (2) would be expected to yield 1,3-diformyl-4,6-dihydroxy-2,5-dimethoxybenzene (106). Thus i s o p i m p i n e l l i n (2) was ozonized i n the manner described by Hegarty and 107 Lahey . C r y s t a l l i z a t i o n of the product mixture y i e l d e d a pure c r y s t a l l i n e compound, mp 162-164\u00C2\u00B0 ( i n 3,5%/ y i e l d ) and subsequent chara c t e r i z a t i o n established i t to be the expected product, 106. Thus elemental analysis and high r e s o l u t i o n mass spectrometry supported the MeOH molecular f ormulaC.--H, n0,, and the uv spectrum (A. 258 and 325 (sh) 10 10 6 max nm) exhibited the expected bathochromic s h i f t on addition of base MeOH (A (+ NaOH) -302 and 358 nm). A c i d i f i c a t i o n did not regenerate the max \u00E2\u0080\u00A2 M P O H neutral spectrum ( A i i t : u n (+ HC1) 213, 243 (sh) and .268 run) . In the i r max spectrum of 106, no hydroxyl absorption was obvious and the aldehyde carbonyl absorption occurred as a broad peak at 1625 cm This data suggests that the aldehydic carbonyls and the adjacent hydroxyl groups are highly c h e l a t e d 1 1 0 . The nmr spectrum (Figure 55) was e n t i r e l y consistent with the structure 106. Thus two methoxyl s i n g l e t s at x5.94 and 6.13 were the only signals observed i n the T0\u00E2\u0080\u009410 range of the spectrum. Two proton s i n g l e t s at low f i e l d could be assigned to the equivalent aldehyde protons (at T -0.08) and to the equivalent phenol protons (at T -2.44; disappearing on addition of D\u00E2\u0080\u009E0). The s i m p l i c i t y of the spectrum r e f l e c t s the symmetry i n the molecule. Thus this reaction allows determination of the r a d i o a c t i v i t y associated with the 2- and 3- positions of i s o p i m p i n e l l i n (2) and the y i e l d i s s u f f i c i e n t to allow operation on a 40-50 mg scale. It was next of i n t e r e s t to obtain a degradative procedure which would allow determination of the r a d i o a c t i v i t y associated with the 4-position of i s o p i m p i n e l l i n (2). Removal of the aldehyde carbons from 106 would allow such a determination. I t was considered that a Dakin-type reaction on the product derived from methylation of 106 would o f f e r the best chance of success on the small scale that would be required for such a r e a c t i o n to be u s e f u l . Thus 106 was methylated OCH CHO -RO 'OR OCH 3 OCH 3 108a; R=H 108b; R=C0CH under conditions as previously employed and 1,3-diformy1-2,4,5,6-tetra-methoxybenzene (107) (mp 49-50\u00C2\u00B0) was i s o l a t e d , a f t e r workup, v i a preparative t i c i n 86% y i e l d . This compound had s p e c t r a l and a n a l y t i c a l properties completely consistent with the assigned s t r u c t u r e . P a r t i c u l a r l y notable was the nmr spectrum (Figure 56) which consisted of only four s i g n a l s ; at x-0.37, a two proton s i g n l e t assigned to the two equivalent aldehyde protons, at x 5.96, a s i x proton s i n g l e t assigned to two equivalent methoxyl groups, and at x 6.09 and 6.15, two three proton s i n g l e t s assigned to two non-equivalent methoxyl groups. This data, i n conjunction with consistent a n a l y t i c a l and mass s p e c t r a l evidence, must be considered conclusive proof that the methylation product had the structure 107. I t i s i n t e r e s t i n g that Hegarty et a l 1 0 7 ' report a melting point of 109-111\u00C2\u00B0 f o r t h i s compound. In consideration of this c o n f l i c t i n g value i t perhaps should be noted that when 106 was methylated under conditions where the r e f l u x period was maintained for a considerably longer time than was employed i n the successful reaction j u s t described, a by-500 I 350 I 100 I 50 'offset 300 Hz PPM(x) ' - r-i\u00E2\u0080\u0094 r - r1 ' H \" T \u00C2\u00BB T \" \" 0 C H 3 O H C - s ^ X ^ C H O H 3C OCH, OCH\u00E2\u0080\u009E 107 >-H> 0 Ml r F i g u r e 56. Nmr Spectrum o f 1 , 3 - d i f o r m y l - 2 , 4 , 5 , 6 - t e t r a m e t h o x y b e n z e n e (107) product i n which one of the aldehyde groups of 107 had apparently condensed with acetone ( i n an a l d o l condensation), was obtained as a major constituent of the reaction mixture. Also notable was that i f the methylation reaction was worked up without a c i d i f i c a t i o n of the reaction mixture, a rather complex product mixture was obtained. The nmr spectrum of t h i s mixture showed signals i n the region of x6.6 suggestive of an a l i p h a t i c methoxyl group. In attempting to r a t i o n a l i z e the presence of these signals i t was considered that hemiacetals or acetals were being formed i n the reaction. This consideration gained considerable credence when i t was found that i f this mixture was treated with hydro-c h l o r i c acid the pure methylated dialdehyde (107) was produced and the nmr spectrum of this product mixture no longer contained the higher f i e l d methoxyl s i g n a l s . Such by-products as these might explain the discrepancy i n the melting points but as these compounds were not p u r i f i e d one cannot be d e f i n i t i v e on this point. The treatment of the methylated dialdehyde (107) i n a c e t i c acid with s u l f u r i c acid and hydrogen peroxide for 16 hours i n the cold ( i . e . under conditions as described by Schonberg et a l ' 1 0 8 ) y i e l d e d only a complex mixture of highly coloured products. I t was evident that i f the diphenol (108a) was being formed i n the reaction, i t was decomposing under the reaction conditions. In order to reduce the suspected decomposition, the reaction time was reduced d r a s t i c a l l y , less hydrogen peroxide and a nitrogen atmosphere was used. Thus 107 was treated i n a c e t i c acid with s u l f u r i c acid and hydrogen peroxide at 0\u00C2\u00B0 for only 15 minutes. The reaction mixture was then quicklyrwprked up i n the cold. The nmr spectrum of the product mixture revealed that e s s e n t i a l l y a l l the aldehyde had reacted i n this short time. A s i n g l e t at T1 . 6 2 was taken to i n d i c a t e that the mixture contained some formate esters such as have been proposed to be intermediates i n the Dakin reaction. A very broad s i g n a l at x4.7 suggested that some phenol was also present. The methoxyl region of the spectrum was quite complex. Thus i t was apparent that the reaction had proceeded but hydrolysis of the intermediate formate esters was incomplete. Examination of the reaction mixture by t i c indicated that at l e a s t three components were present i n the mixture. At the \"base l i n e \" ; a spot which was observed to q u i c k l y become dark purple was considered to be the diphenol (108a). The less polar spots became purple slowly on exposure to a i r and were considered to be the diformate and monoformate derivatives of 108a. Due to the apparent i n s t a b i l i t y of i08a i t was decided that i t would be more expedient to trap the diphenol (108a) as the diacetate d e r i v a t i v e (108b). A common method f o r the a c e t y l a t i o n of phenols i s to quench the phenolate anion (produced by reaction with aqueous base) with a c e t i c anhydride. Applying this method to the trapping of the diphenol (108a) i t was expected that treatment of the mixture of formate esters with aqueous base would serve to hydrolyze the esters and thus produce the dianion of 108a which could then be trapped with a c e t i c anhydride. However when this procedure was applied to the reaction mixture from the previous reaction a complex product mixture was obtained, although some signals assignable to acetate groups were observed i n the nmr spectrum of this product mixture. Considering the apparent i n s t a b i l i t y of the diphenol (108a) to hydrolysis conditions, a more rigorously controlled method of formate ester hydrolysis was devised. I t was expected that a strong nucleophile such as methyllithium could be u t i l i z e d to e f f e c t rapid and complete transformation of the formate esters to the d i l i t h i o s a l t of diphenol (108a).\" Under s t r i c t l y anhydrous conditions the s a l t would be expected to be p r e c i p i t a t e d from the organic solvent and thus, as a s o l i d , perhaps i t would be less prone to decomposition. Quenching such a reaction mixture with a c e t i c anhydride would then afford the diacetate (108b). Thus the dialdehyde (107) was treated i n a c e t i c a c i d with hydrogen peroxide and s u l f u r i c acid at 0\u00C2\u00B0. In t h i s case nitrogen was bubbled > through the mixture to remove oxygen from the en t i r e system. In previous t r i a l experiments decomposition had been noted to be accompanied by the reaction mixture becoming a dark orange colour. With degassing of the solvents and by keeping the reaction under a nitrogen atmosphere the reaction remained e s s e n t i a l l y colourless over the 20 minute course of the reaction. The reaction was worked up quickly i n the cold as before. The product mixture was dissolved i n anhydrous ether and treated with excess methyllithium. As expected, a p r e c i p i t a t e formed and a f t e r treatment with a c e t i c anhydride (with some pyridine added to ensure complete a c e t y l a t i o n ) , workup of t h i s complex mixture yielded a near quantitative y i e l d of t i c pure l,3-diacetoxy-2,4,5,6-tetramethoxybenzene (108b) (97% y i e l d from preparative t i c ) as a colourless o i l . An a n a l y t i c a l sample of 108b was obtained by preparative t i c and m i c r o d i s t i l l a t i o n . A f t e r such p u r i f i c a t i o n , 108b could be induced to c r y s t a l l i z e , mp 57-58\u00C2\u00B0. As expected, the nmr spectrum (Figure 57) of t h i s compound was very simple with three signals i n the methoxyl region (at T6.14 and 6.28 for two non-equivalent methoxyl groups and at x6.17 for two equivalent methoxyl groups), and a s i n g l e acetate s i g n a l (at x7.68 for two equivalent acetate groups). F i g u r e 57. Nmr Spectrum of 1 , 3 - d i a c e t o x y - 2 , 4 ,5,6-tetrame::hoxybenzene (108b) Thus by comparison of the molar activity of 108b with (that of the dialdehyde (106), the activity associated with the 4- and 6- positions of isopimpinellin (2) could be deduced. Since the activity of the 6-position can be determined, the percentage of radioactivity residing at the 4-position is thus determinable. It was finally of interest to determine the percentage of the radioactivity of isopimpinellin (2) which might be associated with the two methoxyl groups. To accomplish this, isopimpinellin (2) was demethylated by refluxing with hydroiodic a c i d 1 1 1 . The liberated 2 109 110 methyl iodide was swept from the reaction mixture with a stream of nitrogen and was trapped as tetramethylammonium iodide (109). This compound could be isolated in 82% yield. Since 109 is a salt, purification by normal means of tic and crystallization was difficult . Thus after scintillation counting of 109, treatment with picric acid afforded the picrate derivative (110) (mp 323-325\u00C2\u00B0 ) . Elemental analysis was consistent with the molecular formula C, _H.. . 0-,N. . 10 14 7 4 Thus by the series of degradations summarized in Figure 58, isopimpinellin (2) could be degraded to determine the radioactivity Figure 58. Degradations of I s o p i m p i n e l l i n (2) associated with, a l l the carbon atoms attached to the benzene po r t i o n of the molecule. Degradations of Alloimperatorin Methyl Ether (7) Alloimperatorin methyl ether (7) contains both a furan r i n g system and a d i m e t h y l a l l y l s i d e chain. Mevalonic acid (57) and acetate have been suggested as precursors for these portions of s i m i l a r coumarins (see Introduction) and i t was of i n t e r e s t to determine the s p e c i f i c i t y of incorporation of such precursors into 7. Thus a seri e s of degradations were devised which would allow determination of the d i s t r i b u t i o n of r a d i o a c t i v i t y i n a l l o i m p e r a t o r i n methyl ether (7) i s o l a t e d from Thamnosma montana i n the course of b i o s y n t h e t i c experiments. To gain information as to the distribution of radioactivity in the prenyl side chain of 7, a cleavage reaction was indicated. It was felt that although 7 has three double bonds which would be reactive to ozone, the partial aromatic character of both the furan and pyrone rings might allow selective ozonization of the side chain double bond. Thus alloimperatorin methyl ether (7) was treated with a 1.5 molar equivalent of ozone in acetic acid and the resultant ozonide was reductively cleaved by zinc dust. Steam distil lation of the resulting mixture into 2,4-dinitropheny-hydrazine reagent (2,4-DNP) allowed isolation of acetone-2,4-DNP (111a) in 43% yield. The isolated 111a was identical with a sample prepared from acetone (mp, mmp and t ic) . The non-volatile portion of the reaction mixture was extracted and preparative tic allowed isolation of unreacted 7 (41% recovery) and a.less polar component which proved to be the expected aldehyde (112) (53% yield). Although 112 could be crystallized (mp 195-200\u00C2\u00B0) and had uv and nmr spectra consistent with the expected product, a satisfactory elemental analysis could not be obtained. The compound appeared to be unstable to air and thus i t was decided to reduce 112 to the corresponding alcohol. Thus the alde-hyde (112) was treated with sodium borohydride and the alcohol (113) was isolated (in 88% yield), mp 167-169\u00C2\u00B0. This compound gave consistent a n a l y t i c a l and mass s p e c t r a l data and the uv spectrum (A 6 220, 245(sh),' max .251, 265 and 306 nm)\"was' e s s e n t i a l l y unchanged from that of alloimperatorin methyl ether (7) confirming the continued presence of the 8-alkoxyfurano-coumarin chromophore. The i r spectrum contained the expected hydroxyl absorption at 3450 cm 1 . The nmr spectrum (Figure 59) completely confirmed the s t r u c t u r a l assignment. Thus the aromatic region contained the furan and pyrone proton signals i n the f a m i l i a r pattern of doublets confirming that the furanocoumarin skeleton was unchanged. The 8-methoxyl group absorbed as the expected s i n g l e t at T5.79. A broad two proton t r i p l e t at x6.07 (J=6 Hz) which sharpened considerably on addition of D^ O was assigned to the methylene group bearing the hydroxyl group, and a sharp two proton t r i p l e t at 6.75 (J=6 Hz) was r e a d i l y assigned to the be n z y l i c methylene group. ' A broad one proton absorption at x8.38 (disappearing on addition of D 20) was assigned to the hydroxyl proton. This degradation, while giving the required products, was found to have some serious drawbacks when i t was subsequently performed on radioactive alloimperatorin methyl ether (7). The most s i g n i f i c a n t problem was that 7 proved to be very d i f f i c u l t to obtain radiochemically pure. I f c r y s t a l l i z a t i o n was performed to e f f e c t maximum pu r i t y , losses i n the mother liquors were large, but i f c r y s t a l l i z a t i o n was performed with recovery, of 7 as the -prime concern, p u r i f i c a t i o n was not effected. Also the highly coloured nature of acetone-2,4-DNP (111a) made s c i n t i l l a t i o n counting of t h i s material inaccurate when low le v e l s of r a d i o a c t i v i t y were present. In considering an a l t e r n a t i v e scheme, i t must be pointed out that another aim i n degrading alloimperatorin methyl ether (7) was to determine I I I I i i i I i \u00E2\u0080\u00A2 i i i I I I I I I ! _ J L i \u00E2\u0080\u0094 I \u00E2\u0080\u0094 I \u00E2\u0080\u0094 l _ J \u00E2\u0080\u0094 I 1_J\u00E2\u0080\u00941\u00E2\u0080\u0094L_l_l\u00E2\u0080\u0094L_i SOOO 2500 \u00E2\u0080\u00A2\u00E2\u0080\u009El . i i t l i i . i l t i i \u00E2\u0080\u00A2 I i i i i I i i i i I i t Ii I I l l I l I I I I I I 1 I I I I 1 I. o , - PPMCT: Figure 59. Nmr Spectrum of 5-(2'-hydroxyethyl)-8-methoxypsoralen (113) the r a d i o a c t i v i t y associated with the furan r i n g and with the a-pyrone r i n g . As i t has been shown that ozone p r e f e r e n t i a l l y attacks the side chain double bond and i t was apparent that i f an ozonolysis procedure was to be used to cleave the furan and pyrone rings, i t would be necessary to f i r s t modify the side chain double bond to make i t r e s i s t a n t to ozonolysis. In preliminary studies with the aim of deactivating the side chain, hydrogenatibn was-considered but t h i s . r e a c t i o n was found to reduce both the side chain and furan double bonds, with l i t t l e obvious s e l e c t i v i t y . The s e l e c t i v e reduction was attempted with diimide (which 112 i s reported to s e l e c t i v e l y attack non-polarized double bonds) but no reaction was observed. A common intermediate which could be used f o r cleavage of the side chain as w e l l as for degradation of the furan and pyrone rings was obviously desirable. Alloimperatorin methyl ether d i o l (3) was found to be the i d e a l intermediate. Dreyer 6 had previously converted alloimperatorin methyl ether (7) to the d i o l (3) v i a the epoxide (97) i n good o v e r a l l y i e l d . U t i l i z i n g an authentic sample of 3, i t was found to c r y s t a l l i z e very w e l l with l i t t l e loss i n the mother liq u o r s and with apparently good p u r i f i c a t i o n being achieved. Treatment of 3 with p e r i o d i c a c i d would be expected to y i e l d acetone and the aldehyde (112) as did the s e l e c t i v e ozonolysis reaction. Also the side chain of 3 should be r e s i s t a n t to ozonolysis. Thus alloimperatorin methyl ether (7) was converted to the epoxide (97) ( i n 84% y i e l d ) and the epoxide (97) was then treated with o x a l i c acid to obtain the d i o l (3) ( i n 70% y i e l d ) . This compound was i d e n t i c a l with an authentic sample of alloimperatorin methyl ether d i o l (3). As noted e a r l i e r , the 2,4-DNP de r i v a t i v e of acetone (111a) was not w e l l H3 . 112 suited to scintillation counting and i t was felt that a better derivative could be found. Acetone forms a colourless derivative when reacted with p-bromobenzenesulfonylhydrazide in aqueous acetic a c i d 1 1 0 . 111b This hydrazone is soluble in common organic solvents, stable, easily crystallized and has a substantial molecular weight, thus allowing trapping of acetone on a very small scale. The main deficiency of this d e r i v a t i v e was that i t was very strongly adsorbed on t i c supports. However i t was found that good recoveries could be achieved i f alumina t i c plates were used and the band was removed from the plate before the solvent could evaporate. Thus alloimperatorin methyl ether d i o l (3) was treated with p e r i o d i c acid and the reaction mixture was steam d i s t i l l e d into p_-bromobenzene-sulfonylhydrazide reagent. The hydrazine mixture was extracted with acetone-free chloroform to y i e l d a mixture which when subjected to prepar-ative t i c , y i e l d e d acetone j>-bromobenzenesulfonylhydrazone (111b) ( i n 85% y i e l d ) . From the n o n - v o l a t i l e portion of the reaction mixture, extraction y i e l d e d a residue which was reduced with sodium borohydride. The product mixture was chromatographed on preparative t i c and-the component at corresponding to the alcohol (113) was i s o l a t e d . However t h i s material proved to be d i f f i c u l t to c r y s t a l l i z e and although the nmr spectrum was e s s e n t i a l l y i n keeping with the expected product, the i r spectrum had an absorption at 2220 cm \ which was strongly suggestive of a B-H group i n the m o l e c u l e 1 1 \ Also, the melting point of the compound (y 210\u00C2\u00B0) was considerably higher than that of the alcohol (113). I t was found that i f this compound was treated with hydrochloric acid, the resultant material c r y s t a l l i z e d e a z i l y and had the correct melting point and i r spectrum. Re-examination of the reaction- sequence revealed that a f t e r the cleavage of the d i o l (3), the aldehyde (112) was the major product, but as steam d i s t i l l a t i o n was c a r r i e d out, t h i s compound was converted p a r t i a l l y to a less polar compound (blue spot, uv) . Apparently t h i s decomposition was slow and i f steam d i s t i l l a t i o n was c a r r i e d out for a short time, only a small portion of the aldehyde was decomposed. I t was decided that rather than inv e s t i g a t e the exact nature of this unwanted reaction, i t would be more p r o f i t a b l e to devise a method of removing acetone from the reaction mixture without heating to such an extreme temperature. In view of the above,- a d d i t i o n a l work was desirable and the following method which was the most e a s i l y reproducible, was f i n a l l y developed. Alloimperatorin methyl ether d i o l (3) was treated with p e r i o d i c acid and a f t e r the reaction was complete, acetone was removed from the reaction mixture and into a so l u t i o n of p-bromobenzenesulfpnylhydrazide reagent with a strong stream of nitrogen passing over the gently heated mixture. In t h i s .way no noticeable decomposition of the aldehyde (112) took place (as judged by t i c ) and a f t e r i s o l a t i o n of the crude product, reduction with sodium borohydride gave the alcohol (113) i n 47% y i e l d . Work up of the hydrazone mixture y i e l d e d acetone p-bromobenzenesulfonylhydrazone (111b) i n 35% y i e l d . U t i l i z i n g a blank experiment with i d e n t i c a l work up and acetone free solvents, very l i t t l e acetone was detected i n the system i t s e l f . While the y i e l d of 111b was less than could be obtained by steam d i s t i l l a t i o n , the quantity was s u f f i c i e n t to allow radioactive determina-t i o n when ,the reaction was performed on a 30-50 mg scale. As noted previously i t was desired to develop degradations which would allow determination of d i s t r i b u t i o n of r a d i o a c t i v i t y i n the furan portion of alloimperatorin methyl ether (7). It was expected that by u t i l i z i n g a c o n t r o l l e d ozonolysis, alloimperatorin methyl ether d i o l (3) could be degraded to the corresponding phenolic aldehyde as was done i n the case of i s o p i m p i n e l l i n (2). However, i t was noted that the d i o l (3) i s quite a.polar compound and one might expect that the ozonolysis product would be even-more polar, possibly creating problems i n i t s i s o l a t i o n and ch a r a c t e r i z a t i o n . As 3 was known to form a monoacetate 6 i t was f e l t that the acetylated product would serve as a more appropriate compound f o r ozonolysis. Thus 3 was treated with a c e t i c anhydride i n py r i d i n e and alloi m p e r a t o r i n methyl ether d i o l monoacetate (114) was i s o l a t e d by preparative t i c i n 86% y i e l d . OCH-: OCH, OCH, 3 114 115 The monoacetate (114) had melting point and nmr properties i n keeping with those reported by Dreyer 6. The monoacetate (114) was treated with a s l i g h t excess of ozone i n g l a c i a l a c e t i c acid and a f t e r reduction of the ozonide with zinc workup of the mixture y i e l d e d a residue which when analyzed by t i c was observed to be a mixture with one compound appearing as a b r i g h t yellow spot ( v i s i b l e and uv) and some of the s t a r t i n g m a t e r i a l as a major component. The two components exhibited very s i m i l a r R^ values i n sev e r a l solvent systems, making separation v i a preparative t i c d i f f i c u l t . However separation was e a s i l y achieved by f i r s t i s o l a t i n g the two components together, by preparative t i c and then part i o n i n g the mixture between chloroform and aqueous base. The non-base soluble p o r t i o n was examined by t i c and observed to contain some of the s t a r t i n g acetate and a less polar component (dark blue, uv). This material was not examined further. The base soluble material was i s o l a t e d by a c i d i f i c a t i o n and e x t r a c t i o n . Examination of t h i s material by t i c revealed the bright yellow ( v i s i b l e and uv) component as the major constituent. C r y s t a l l i z a t i o n of t h i s material proved to be d i f f i c u l t but once achieved, r e c r y s t a l l i z a t i o n was not d i f f i c u l t . Thus t h i s compound could be i s o l a t e d ( i n 27% y i e l d ) and was f u l l y characterized as the expected phenolic aldehyde (115), mp 162-164\u00C2\u00B0. The uv spectrum MeOH. MeOH (X 273 and 345 nm) became quite complex on addition of base (A max max (+ NaOH) 236, 267.5, 300 and 357 mm) and a c i d i f i c a t i o n did not regenerate MeOH the neutral spectrum (X (+ HC1) 205, 263 and 326 nm). Elemental , max analysis and high r e s o l u t i o n mass spectrometry supported the molecular formula C H 0 . The nmr spectrum (Figure 60) completely confirmed the l o ZU o structure of 115. Thus the spectrum was e s s e n t i a l l y unchanged from that of the s t a r t i n g compound6 except that the furan proton signals were absent and i n t h e i r place were a s i n g l e t at x-0.39, assigned to the aldehyde proton, and a broad s i g n a l at T-2.37 (disappearing on addi t i o n of D\u00E2\u0080\u009E0) assigned to the phenolic proton. Various t r i a l reactions where ozone concentration was varied revealed that optimum y i e l d of 115 was obtained when an excess of about 33% ozone was used. I f the reaction residue was c r y s t a l l i z e d d i r e c t l y a f t e r reduction of the ozonide, about 25% of the s t a r t i n g monoacetate 114 could be recovered. Thus the above'reaction y i e l d s a product which w i l l allow determination of the r a d i o a c t i v i t y i n the 7-position of alloimperatorin methyl ether (7). I t was next desired to be able to determine what r a d i o a c t i v i t y might reside i n 6-position of alloimperatorin methyl ether (7). This information could be obtained by removal of.the aldehyde carbon atom of 115 i n the manner as found successful i n the degradation of i s o p i m p i n e l l i n (2). For thi s purpose the phenolic aldehyde (115) was methylated i n the normal manner and the methylated coumarin (116), mp 116-118\u00C2\u00B0 was i s o l a t e d i n 80% yield. The nmr spectrum of 116 revealed immediately that the correct product had been obtained. Thus the phenol proton signal was absent and two methoxyl signals were now evident. This material was treated in the cold with hydrogen peroxide, sulfuric acid and acetic acid. Chromatography of the reaction mixture on preparative t i c provided a major band which was isolated i n 93% yi e l d . The nmr spectrum of this material revealed that the desired phenol (117a) had been obtained. Thus the spectrum was very similar to that of the starting aldehyde (116) but the aldehyde signal was absent and instead a signal at T3.90 (disappearing on addition of D2O) was evident. This compound however could not be induced to crystallize. Thus 117a was treated with acetic anhydride and pyridine and was converted to the derivative 117b (in 44% overall yield). This material was highly crystalline, mp 143-144\u00C2\u00B0,and had analytical and spectral properties completely consistent with the expected product. Thus the nmr spectrum (Figure 61) now contained signals for both an aromatic acetyl group at xl.61 and for an aliphatic acetyl group at x8.17. To allow determination of the radioactivity associated with the pyrone portion of alloimperatorin methyl ether ( 7 ) , i t was felt that the procedures devised for the degradation of isopimpinellin (2) would be equally applicable in this case. It was therefore expected that 114 118 treatment of the diol acetate (.lib) with excess ozone would yield the corresponding diphenolic dialdehyde (118). As this compound has lost carbons (2) and (3) of the pyrone ring relative to the monoaldehyde (115), the radioactivity associated with these two positions can be calculated. Thus the diol acetate (114) was ozonized in the manner described by Hegarty and Lahey* 0 7. The residue after workup was crystallized to yield a compound which was fully characterized as the expected 118, mp 188-190\u00C2\u00B0 (in 54% yield). Elemental analysis and high resolution mass spectrometry supported the molecular formula ,U~r.0c,. The uv spectrum lo _0 8 exhibited the acid and base shifts now familiar for phenolic aromatic aldehydes. The nmr spectrum of 118 (Figure 62) fully confirmed the structural assignment. Thus the signals assigned to protons on the side chain and for the methoxyl group were essentially unchanged. The remaining absorptions were two proton singlets at x-0.28 and at -2.82 (disappearing on addition of D\u00E2\u0080\u009E0) readily assigned to the aldehyde and phenol protons scoo I 2500 1000 sio 50 i i f I ' > \u00E2\u0080\u00A2 ' I I ' \u00E2\u0080\u00A2 I I I I I 'l I , . . I 1 , '. , I I OHC. HO' OH ^^\u00E2\u0080\u00A2OCOCL .CHO *0H OCH, 118 -~yv\u00E2\u0080\u0094\u00E2\u0080\u0094\u00E2\u0080\u00A2\u00E2\u0080\u0094 offset 300 Hz I i I ' : I ' I I t I t ' i i I i i i i 1 1 1 1 1 1 1 1 1 1 1 PPM(T) I I ' I I I I I I i \u00E2\u0080\u00A2 i i i i i i i i i i i i i i i i i . i i i i ). , , i -i i i > - H \u00C2\u00BB H i I I I I I I I nl 10 I 00 ! respectively. The yield of this reaction allows one to perform this conversion effectively on a 20 mg scale (of diol acetate (114)). To determine the radioactivity associated with the 4-position of alloimperatorin methyl ether (7), a sequence of reactions similar to those performed successfully on isopimpinellin (2) was attempted. 0CH3 0CH3 0CH3 118, 119 Thus the dialdehyde (118) was methylated by standard procedures and the product (mp 79 .5-80 .5 \u00C2\u00B0 ) gave spectral and analytical data fully consistent with the expected product (119). However when this material was treated under the conditions as developed in the degradation of isopimpinellin (2), only complex mixtures of coloured products could be obtained. It was evident that even under the stringent conditions of the reaction , the resultant diphenol (or perhaps the intermediate diformate ester) was decomposing as fast as i t was being formed. Thus attempts to affect this conversion were abandoned. Finally, to determine the radioactivity associated with the methoxyl group in alloimperatorin methyl ether (7), a demethylation, under the same conditions as described for the demethylation of isopimpinellin (2), was performed. Thus the alcohol (113) derived from the cleavage of the diol (3) was treated with refluxing hydrogen iodide and the methyl iodide thus formed was trapped as tetramethylammonium iodide (109) (in 81% yield). Conversion to the picrate derivative (110) was carried out as before. It should be noted that, these degradations as summarized in Figure 63 for alloimperatorin methyl (7) are equally applicable to the degradation of alloimperatorin methyl ether epoxide (97). Degradations of Umbelliprenin (83) As mentioned previously, the distribution of radioactivity ih umbelliprenin (83) was considered to be a measure of the specificity of the incorporation of mevalonic acid (57). Thus i f mevalonic acid (57) was utilized intact with no significant degradation to smaller subunits the radioactivity in umbelliprenin (83) should be located exclusively in the farnesyl side chain with a predictable distribution. To determine the total amount of radioactivity associated with the entire side chain, and with the umbelliferone portion of the molecule as well, an acid catalyzed hydrolysis was performed. Kariyone and 115 Matsumo found that the coumarin auraptene (the geranyl ether of umbelliferone) would undergo efficient hydrolysis in hot glacial acetic acid to yield umbelliferone (45) and geranyl acetate. However i f the hydrolysis was carried out in acetic acid with a small amount of sulfuric acid added, geranyl acetate was not obtained owing to polymerization. This latter situation prevailed even i f the reaction was carried out entirely at room temperature. Thus umbelliprenin (83) was refluxed with glacial acetic acid and the residue after removal of the solvent was chromatographed on a preparative tic plate. Two components were thus isolated, a polar compound which proved to be umbelliferone (45) (72% yield, identified by comparison with an authentic sample) and a non-polar band isolated as a colourless o i l . The oily component did not correspond in polarity on tic with farnesyl acetate (61c) (prepared from farnesol by standard methods). The i r spectrum of the oily component 0 83 45 showed no hydroxyl or carbonyl absorptions. Analysis of this material by vpc revealed i t to be a complex mixture. Indeed when farnesyl acetate was treated under the reaction conditions the resultant oily product was shown to be a similar complex mixture. Thus i t was apparent that under the reaction conditions, farnesyl acetate produced in the reaction was undergoing a series of reactions. Considering the difficulty in characterizing such a mixture and as the radioactivity present in the umbelliferone (45) derived from the reaction would allow calculation of the radioactivity in the side chain by difference, i t was decided to abandon attempts to isolate the entire side chain. In order to gain some information as to the distribution of radio-activity (if any was found) in the umbelliferone portion of 83, umbel-liferone (45) was fused with potassium hydroxide at high temperature (this method was reported to convert 45 to resorcinol) 1 1 6 . However examination of the reaction product by tic revealed that very l i t t l e resorcinol was produced. The major product (a more polar blue spot, uv) could be purified by sublimation of the reaction mixture, yield a compound (in 47% yield) shown to be 2,4-dihydroxybenzoic acid (120) by nmr, mp and elemental analysis. The compound was identical with an authentic sample which was readily available commercially. Thus this reaction allows determination of the radioactivity present\" in the 2- and 3-positions of umbelliprenin (83). Under the reaction conditions i t was evident that decarboxylation did not readily occur. As i t was also of interest to determine the distribution of radio-activity in the farnesyl ether side chain, a suitable degradation scheme was devised. Caldwell and Jon e s 1 1 7 have reported the isolation of both acetone and levulinaldehyde (121) as 2,4-DNP derivatives, from 7-methoxy-5-geranyloxycoumarin (122) by ozonolysis and steam d i s t i l l a t i o n of the reaction mixture into 2,4-DNP reagent, although no yields were reported. With, farnesol (612) as a model compound, the reaction was carried out according to the above procedure. By collecting fractions from the steam d i s t i l l a t i o n , the f i r s t few fractions were found to contain considerable quantities of acetone-2,4-DNP (111a) while the later fractions contained a red solid which was found to be insoluble in a l l common solvents. This material was crystallized from a nitrobenzene-ethanol mixture to yiel d a pure compound, mp 233-235\u00C2\u00B0, as reported for the bis-2,4-DNP derivative of levulinaldehyde (123). The yield, however, was very poor (y 1%) making this procedure impractical for small scale reactions. Levulinaldehyde (121) as a 1,4-dicarbonyl compound would be expected to be quite unstable under the drastic workup conditions of the above 122 111a; R=2,4-DNP 121; R=0 123; R=2,4-DNP reaction, which, might contribute to the poor y i e l d observed. Thus a less d r a s t i c workup of the reaction was considered. Henne and P e r i l s t e i n have described -a reductive workup for ozonization reactions u t i l i z i n g c a t a l y t i c hydrogenation of the r e s u l t i n g ozonides to provide the desired carbonyl products. This method offered a reaction medium which would be e n t i r e l y n eutral. Thus farnesol (61b) was ozonized exhaustively i n eth y l acetate and the mixture was then immediately hydrogenated over palladium on calcium carbonate i n a high pressure Parr hydrogenator for 22 hours. The r e s u l t i n g mixture was quickly f i l t e r e d into a s o l u t i o n of 2,4-DNP reagent i n methanolic hydrogen chloride and the orange coloured c r y s t a l s of 123 p r e c i p i t a t e d . R e c r y s t a l l i z a t i o n y i e l d e d pure 123, mp 238-240\u00C2\u00B0 (10% y i e l d ) . ' As acetone-2,4-DNP would be also produced i n this reaction i t was desirable to i s o l a t e this compound as we l l . However for such an i s o l a t i o n to be meaningful the solvents used i n the reaction and workup must be acetone-free. However methanol which had been d i s t i l l e d from iodine and aqueous base (to remove acetone as iodoform) proved unsuitable for the reaction as i t was found that the 2,4-DNP reagent would not dissolve s a t i s f a c t o r i l y i n the acetone-free methanolic hydrogen chloride, apparently due to the considerable water content of this methanol. Despite several attempts u t i l i z i n g acetone-free methanol containing various amounts of water, only normal \"reagent grade\" methanol gave s a t i s f a c t o r y r e s u l t s . However optimum conditions for the i s o l a t i o n of 123 from farnesol were found from a serie s of experiments. Thus hydrogenation was car r i e d out for a r e l a t i v e l y short time (1.5 hours) and levulinaldehyde-bis-2,4-DNP could be i s o l a t e d i n 19% y i e l d . When the ozonolysis was performed with umbelliprenin (83) under optimum conditions of time and concentration, levulinaldehyde bis-2,4-DNP 123; R=2,4-DNP 83 (123) was isolated i n 26% y i e l d (mp 241-242.5\u00C2\u00B0). An authentic sample of 123 was prepared from 2-methylfuran according to the method of W i l s o n 1 1 5 and proved to be identical to the isolated material. As mentioned previously, i t was desired to isolate the terminal three carbon system of the farnesyl ether side chain of umbelliprenin (83) as acetone in an ozonolysis reaction. Unfortunately at this time the supply of inactive umbelliprenin (83) was very low and i t was f e l t this material could be better used to dilute the active compound to allow degradations to be performed. Thus without the required material for preliminary t r i a l s of this reaction i t was necessary to abandon this degradation. It is hoped that when more 83 becomes available this degradation w i l l be worked out by other workers. EXPERIMENTAL (PART n ) For the general experimental information, see page 93, Methanol was made acetone-free by distillation for iodine in aqueous potassium hydroxide 1 2 0 3 . Chloroform was made acetone-free by flushing through a column of celite impregnated with 2,4-dinitrophenyihydrazine and the elutant was then distilled 120b # Unless otherwise noted, tic was performed on s i l ica gel G (acc. to Stahl). Preparation of a Standard Solution of Ozone in Glacial Acetic Acid Glacial acetic acid was placed in a flask equipped with a bubbler and ozone enriched oxygen was allowed to bubble through the solution for 1 hour at room temperature, at which time the solution had a definite blue tinge. The bubbler was then removed and the flask was tightly stoppered. Aliquots (20 ml) of this solution were added to a solution of potassium iodide lg) in water 0- 20 ml) and the iodine which was liberated was titrated with a standard solution of sodium thiosulfate with starch as indicator. Two aliquots were taken and the results averaged. The sodium thiosulfate solution was standardized by titration of the iodine released on addition of an aliquot of 0.100 N potassium dichromate. The solution was kept at pH 1 and starch was used as indicator. In a typical experiment, glacial acetic acid was saturated with ozone as described above and two aliquots (20 ml) were removed and added individually to aqueous solutions of potassium iodide (lg per flask in two flasks). The iodine liberated was titrated with 0.0125 N sodium thiosulfate solution requiring respectively 17.1 and 16.8 ml to reach the end point. Thus the average of these two values (16.95 ml) required that ozone concentration at room temperature be 0.106 mmol per 20 ml g l a c i a l a c e t i c acid. 6- Formyl-7-hydroxy-5,8-dimethoxycoumarin (99) Isopimpinellin (2) .(57 mg; 0.232 mmol) was treated with ozone saturated g l a c i a l a c e t i c acid (80 ml; 0.40 mmol ozone) and the mixture was s t i r r e d at room temperature, with l i g h t excluded, for 4 hours. Zinc dust (400 mg) was then added and s t i r r i n g was continued for a further 1.5 hours. The mixture was then f i l t e r e d and the solvent was removed i n vacuo. The residue (^ 60 mg) was dissolved i n a chloroform-methanol mixture and was chromatographed on s i l i c a gel (6 g) The \u00E2\u0080\u00A2 fr a c t i o n s eluted with benzene and benzene\u00E2\u0080\u0094chloroform contained i s o p i m p i n e l l i n (2) and a more polar compound (yellow spot; uv and v i s i b l e ) . These f r a c t i o n s were combined (53.5 mg) and c r y s t a l l i z e d from acetone to y i e l d 6-formyl-7- hydroxy-5,8-dimethoxycoumarin (99) (15 mg; 25.9% y i e l d ) , mp 214-216\u00C2\u00B0; i \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 i r (KBr) 1758, 1730, 1625, 1592 (a-pyrone), 1640 (CfO, aldehyde); MeOH MPOH uv X (e) 275 (27,100); uv X u (_) (+ NaOH) 238 (19,200), 269 (16,600), _n.__x nicix 299 (12,900), 360 (14,200); uv AM E 0 H (e) (+ HC1) 208 (29,000), 226 (sh) UlcLX (15,600), 263 (12,800), 320 (15,600); nmr (100 MHz) i n CDC1-, TMS lock, -2.03 (IH, s i n g l e t , disappears on addition of D\u00E2\u0080\u009E0, phenolic OH), -0.23 (IH, s i n g l e t , aromatic CHO), 2.17 (IH, doublet, J=10 Hz, H(4) of coumarin), 3.73 (IH, doublet, J=10 Hz, H(3) of coumarin) , '6.00, 6.02 (6'H, two s i n g l e t s , two aromatic 0CH_3); mass spectrum m/e 250 (M), 235 (M-15) , 221 (M-29) , 207 and 179. Anal.. Calcd. for C^H^O, : C, 57.61; K, 4.03. Found : C, 57.38; 1/ 1U o H, 4.07. High r e s o l u t i o n molecular weight determination. Calcd. f o r C.-H-.O, : 250.048. Found: 250.046. 12 10 6 6-Fbrmyl-5 >7,8-trimethoxycoumarin (100) To a s o l u t i o n of 6-formyl-7-hydroxy-5,8-dimethoxycoumarin (99) (24 mg; 0.096 mmol) i n acetone (20 ml) was added anhydrous potassium . carbonate (lg) and methyl iodide (2 ml). The mixture was refluxed for 1.5 hours, s t i r r e d a further 1 hour at room temperature then water (20 ml) was added. The s o l u t i o n was a c i d i f i e d with concentrated hydrochloric acid, extracted with chloroform (3 x 20 ml); the chloroform extracts were washed with water (20 m l ) x dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a c r y s t a l l i n e residue (25.8 mg), which was observed to be e s s e n t i a l l y one compound on t i c . C r y s t a l l i z a t i o n from ethyl acetate y i e l d e d 6-formyl-5,7,8-trimeth-oxycoumarin (100) (18.4 mg; 73% y i e l d ) as white needles, mp 152.5-154\u00C2\u00B0; MeOH i r (KBr) 1745, 1720, 1610, 1578 (a-pyrone), 1689 (HC=0); uv X ( e ) Ttl 3.X 206 (20,300), 267 (14,600), 300 (sh). (10,200); nmr (100 MHz) i n CDC13, TMS lock, -0.37 (IH, s i n g l e t , CHO), 2.03 (IH, doublet, J=10 Hz, H(4) of coumarin), 3.65 (IH, doublet, J=10 Hz, H(3) of coumarin), 5.91, 6.02, 6.08 (9H, three s i n g l e t s , three aromatic OCH^); mass spectrum m/e 264 (M, base peak), 249 (M-15), 235 (M-29), 221. Anal. Calcd. f o r C ^ H ^ O ^ C, 59.09; H, 4.58. Found: C, 58.91; H, 4.73. High r e s o l u t i o n molecular weight determination. Calcd. f or C. -H.. o0, : 13 12 6 264.063. Found: 264.063. Jones Oxidation of 6-F6rmy1-5,7,8-trimethoxycoumarin (100) a) To a so l u t i o n of 6-formyl-5,7,8-trimethoxycoumarin (100) (6.7 mg; 0.025 mmol) i n acetone (2 ml) was added a standard chromic acid 10 6 s o l u t i o n (0.04 mmol) and the reagents were s t i r r e d at 0\u00C2\u00B0 f o r 1 hour. The reaction was warmed to room temperature and further chromic acid s o l u t i o n (0.15 mmol) was added over 5 hours u n t i l the orange colour p e r s i s t e d . The excess reagent was destroyed with isopropanol and water (25 ml) was added. The mixture was extracted with chloroform (5 x 20 ml) and et h y l acetate (30 ml), the extracts were combined and washed with water (25 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (7.2 mg) which was pa r t i t i o n e d between chloroform and aqueous potassium carbonate s o l u t i o n (10 ml). The chloroform f r a c t i o n (4 mg) consisted mainly of unreacted s t a r t i n g material (by t i c ) . The aqueous layer was a c i d i f i e d with d i l u t e hydrochloric acid and extracted with chloroform (3 x 15 ml). The extract was washed with water (10 ml), dried over anhydrous sodium s u l f a t e and the solvent removed under reduced pressure to y i e l d a residue (1.5 mg) containing a polar component and some unreacted s t a r t i n g material (by t i c ) . b) To a s o l u t i o n of 6-formyl-5,7,8-trimethoxycoumarin (100) (7.5 mg; 0.028 mmol) i n acetone (2 ml) was added standard chromic acid s o l u t i o n 1 0 8 (0.56 mmol) and the reagents were s t i r r e d at room temperature for 15 hours. The excess reagent was destroyed with isopropanol, water (10 ml) was added and the mixture was extracted with chloroform (3 x 20 ml). The chloroform extract was washed with water (10 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (9.8 mg) which, contained polar and non-polar components (by t i c ) . Partioning of this residue between chloroform and aqueous potassium carbonate y i e l d e d an a c i d i c f r a c t i o n (2.0 mg) and a non-acidic f r a c t i o n (7.5 mg). The non-acidic f r a c t i o n did not contain the s t a r t i n g material (by t i c ) and had a complex nmr spectrum. The a c i d i c f r a c t i o n was too small to investigate profitably\"., Attempted Oxidation of 100 With Potassium Permanganate 6-Formy1-5,7,8-trimethoxycoumarin (100) (7.5 mg; 0.028 mmol) was suspended i n water (2 ml) and potassium hydroxide (15 mg) was added and dissolved with s t i r r i n g . Aqueous potassium permanganate (1.8 ml; 0.28 mmol) was added and the reaction mixture was s t i r r e d vigorously at 50\u00C2\u00B0 for 2 hours during which time the s o l u t i o n became brown and a l l s o l i d dissolved. The mixture was f i l t e r e d to remove manganese dioxide, the f i l t e r e d s o l i d was treated with b o i l i n g water and f i l t e r e d again. The f i l t r a t e s were combined, brought to pH. 2 with concentrated hydrochloric acid and extracted with chloroform (3 x 15 ml) and e t h y l acetate (20 ml). The extracts.:were combined, washed with water (20 ml) , dried over anhydrous sodium s u l f a t e and the solvent removed,under reduced pressure to y i e l d a residue (8.0 mg), containing a polar component as a major constituent and some s t a r t i n g material (100) (by t i c ) . On r e - e l u t i n g the t i c plate a f t e r standing, the polar component appeared to be transformed into a component with retention time on t i c i d e n t i c a l with the s t a r t i n g aldehyde (100). C r y s t a l l i z a t i o n of the residue y i e l d e d 100 (4.1 mg), mp]52.5-1 54\u00C2\u00B0, and examination of the mother l i q u o r s revealed more 100 remained. Attempted Oxidation of 6-f ormyl-5, 7, 8-trimethoxycoumarin (100) With S i l v e r Oxide 1 2- 1' To a s t i r r e d s o l u t i o n of 6-formyl-5,7,8-trimethoxycoumarin (100) (7.2 mg; 0.027 mmol) i n ethanol (1.5 ml) was added a s o l u t i o n of s i l v e r n i t r a t e (10.2 mg; 0.055 mmol) i n water (0.5 ml) and a f t e r mixing, sodium hydroxide (5.0 mg; 0.110 mmol) i n water (0.5 ml) was added dropwise over 15 minutes. S t i r r i n g was continued for two hours. The s o l u t i o n was then f i l t e r e d , d i l u t e d with water (5 ml), washed with chloroform and ether, a c i d i f i e d with concentrated hydrochloric acid and extracted with chloroform (3 x 15 ml). The chloroform extract was dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (6.7 mg) which contained a polar component and a component corresponding to 100 (by t i c ) . A f t e r drying and r e - e l u t i o n of the t i c plate, the polar component displayed t i c c h a r a c t e r i s t i c s i d e n t i c a l with the s t a r t i n g material (100). 6-Hydroxy-5,7,8-trimethbxycoumarin (105) 6-Formy1-5,7,8-trimethoxycoumarin (100) (16.8 mg; 0.064 mmol) was dissolved i n g l a c i a l a c e t i c acid (1.5 ml) and cooled i n an i c e bath u n t i l i t began to s o l i d i f y . An i c e - c o l d mixture of 30% hydrogen peroxide (0.20 ml) and 50% s u l f u r i c acid (0.75 ml) was added and the reaction mixture was allowed to stand i n a r e f r i g e r a t o r f o r 20 h o u r s 1 0 8 : The mixture was then poured i n t o cold b r i n e (20 ml)\u00E2\u0080\u00A2 The s o l u t i o n was extracted with chloroform (3 x 20 ml) and the extract was.washed with brine (20 ml), dried over anhydrous sodium s u l f a t e and the solvent removed under reduced pressure to y i e l d a residue (15.4 mg) shown to be e s s e n t i a l l y one component by t i c . Preparative t i c (eluting with solvent A), followed by sublimation of the isolated material (140\u00C2\u00B0, 0.02 mm), allowed isolation of 6-hydroxy-5,7,8-trimethoxycoumarin (105) (12.3 mg; 77%:yield). Crystallization from 95% ethanol yielded pure 104 as fine needles, mp 198.5-199.5\u00C2\u00B0; i r (KBr) MeOH 3525 (OH), 1720, 1611, 1570 (a-pyrone); uv A (e) 208 (24,500), 308 n i 3 . x (10,350); uv AM E 0 H (+.NaOH) 248 (33,800), 317 (21,600); uv AM E 0 H (+ HC1) max K > y j m a x 209 (22,500), 308 (10,100); nmr (100 MHz) i n CDC13> TMS lock, 2.06 (IH, doublet, J=9.5 Hz, H(4\"> of coumarin), 3.70 (IH,, doublet, J=9.5 Hz, H(3)' of coumarin),4.38 (IH,'broad, disappearing on addition of D 2 O , phenolic OH), 5..93, 6.04, 6.06 (9H,~ three.singlets, three aromatic OCH^)-mass- spectrum m/e 252 (M, base peak), 237 (M-15), 209,.181, 153. Anal. Calcd. for C..H,o0. : C, 57.14, H, 4.80. Found: C, 57.24; H, 4.90. l z X Z D High resolution molecular weight determination. Calcd. for ^ ^2^12^6 '' 252.063. Found: 252.063. l,3-Diformyl-4,6-dihydroxy-2,5-dimethoxybenzene (106) Isopimpinellin (2) (52 mg; 0.211 mmol),,dissolved in g l a c i a l a c e t i c acid (5 ml) and eth y l acetate (2 ml),, was subjected to an ozone enriched stream of oxygen at 0\u00C2\u00B0 for 0.5 hours. F i l t r a t i o n of the reaction mixture and removal of the solvent i n vacuo y i e l d e d an orange o i l which was treated with hot chloroform. The chloroform soluble portion (57 mg) was c r y s t a l l i z e d from e t h y l acetate to y i e l d l,3-diformyl-4,6-dihydroxy-2,5-dimethoxybenzene (106) (17 mg; 35% y i e l d ) . R e c r y s t a l l i z a t i o n from ethyl, acetate y i e l d e d an a n a l y t i c a l sample of 106; mp 162-164\u00C2\u00B0; i r (KBr) MPOH MeOH 1625 (C=0, K-bonded); uv \ (e) 258 (30,280), 325 (sh) (6080); uv A TT13.X uia.x (e) (+ NaOE) 302 (49,800), 358 (9,450); uv A (e) (+ HC1) 213 (11,600), max 243 (sh.) (6,800), 268 (23,100); nmr (100 MHz) i n CDC13,. TMS lock, -2.44 (2H, s i n g l e t , disappearing on addition of V^Q, two equivalent phenolic OH) -0.08 (2H, s i n g l e t , two equivalent CHO), 5.94, 6.13 (6H, two s i n g l e t s , two aromatic OCH^); mass spectrum m / c 226 (M, base peak), 211 (M-15), 198 (M-28), 183, 165, 152. Anal.- Calcd. for 'C..1 nOr :.C, 53.10; tt, 4.46. Found: C, 53.06, H, 4.45. 1U 1U D I y High r e s o l u t i o n molecular weight determination. Calcd. for C. .H,n0, : 1U 1U o 226.048. Found: 226.048. 1,3-Diformyl-2,4,5,6-tetramethpxybenzene (107) l,3-Diformyl-4,6-dihydroxy-2,5-dimethoxybenzene (106) (8.5 mg; 0.0335 mmol) was dissolved i n acetone (5 ml) and refluxed with potassium carbonate (500 mg) and methyl iodide (2 ml) for 4 hours. After cooling, water (10 ml) was added and the mixture was a c i d i f i e d with concentrated hydrochloric acid and then extracted with chloroform (3 x 15 ml). Af t e r washing with water (10 ml) and drying over anhydrous sodium s u l f a t e , the chloroform was removed under reduced pressure to y i e l d a residue (9.5 mg) which was chromatographed by preparative t i c ( e l u t i n g with solvent A) to y i e l d 1,3-diformyl-2,4,5,6-tetramethoxybenzene (107) (8.2 mg; 86% y i e l d ) C r y s t a l l i z a t i o n from anhydrous' ether-hexane y i e l d e d 107 as white needles, mp 49-50\u00C2\u00B0 ( l i t . 1 0 7 mp 109-111\u00C2\u00B0); i r (KBr) 1686 (C=0, aldehyde); u v ^ \u00C2\u00B0 H (e) 214 (14,100), 257 (14,900), 257 (14,900), 314 (2770); nmr (100 MHz) i h CDC13, TMS lock, -0.37 (2tt, s i n g l e t , two CHO), 5.96 (6H, s i n g l e t , two 0Ctt3) , 6.09, 6.15 (6tt,,,two s i n g l e t s , two 0CH3) ; mass spectrum m / e 254 (M, base peak), 239 (M-15), 225 (M-29), 211. Anal. Calcd. f or C^H^Og : C, 56.69; tt, 5.55. Found: C, 56.58; H, 5.48. Unsuccessful Methylation of 1,3-diformy 1-4,6-dihydroxy-2, 5r-dlmeth.oxy- benzehe (106) l,3-diformyl-4,6-dihydroxy-2,5-diinethoxybenzene (-106) (12 mg; 0.0531 mmol) was dissolved i n acetone (5 ml) and refluxed with potassium carbonate (0.5 g) and methyl iodide (2 ml) for 22 hours. A f t e r cooling water (10 ml) was added and s t i r r i n g continued 15 minutes. The mixture was then a c i d i f i e d with concentrated hydrochloric acid and extracted with chloroform (3 x 15 ml). The chloroform extract was washed with water, dried over anhydrous sodium s u l f a t e and the solvent was concentrated under reduced pressure and the resultant s o l u t i o n was chromatographed on preparative t i c (solvent A). Two bands were i s o l a t e d ; the less polar band corresponded to 107 (3.5 mg) and the more polar band (8.5 mg) was also i s o l a t e d , nmr (60 MHz) (of more polar compound) i n CDC1\u00E2\u0080\u009E \u00E2\u0080\u00940.22 (IH, s i n g l e t , CHO), 4.50 (IH, doublet of doublets, J=4 Hz and 9 Hz, CH 2-CH(0H)(R)), 5.99, 6.08, 6.17, 6.20 (12H, four s i n g l e t s , four aromatic 0CH_3), 6.5-7.5 (3H, mu l t i p l e t , CH 2-CH(0H)), 7.85 (3H, s i n g l e t , CH3\"C=0) . Hemiacetal or Acetal Intermediates i n the Methylation of 1,3-Diforml- 4,6-dihydroxy-2,5-dimethoxybenzerie (106) l,3-Diformyl-4,6-dihydroxy-2,5-dimethoxybenzene (106) (29 mg; 0.127 mmol) was dissolved i n acetone (3 ml) and refluxed with potassium carbonate (0.5 g) and methyl iodide (2 ml) for 4 hours. The mixture was then f i l t e r e d and the solvent removed under reduced pressure. The mixture i s o l a t e d was chromatographed on preparative t i c (solvent B), el u t i n g three times and the band at R_ corresponding to 107 (20 mg) was i s o l a t e d ; nmr (60 MHz) i n CDC1 , -0.22 C-'IH, s i n g l e t , CHO), 6.0 -6.3 (^\u00C2\u00BB12H, many peaks, aromatic OCE_), 6.55 \u00E2\u0080\u00A2\u00E2\u0080\u00A2 (y 6H, s i n g l e t , a l i p h a t i c OCH_). This material was dissolved i n acetone (1 ml) and water (1 ml) and treated with concentrated hydrochloric acid (15 drops). A f t e r s t i r r i n g 15 minutes the mixture was extracted with chloroform (3 x 10 ml); the extract was washed with water (10 ml), dried over anhydrous sodium s u l f a t e and the solvent removed under reduced pressure. The residue (17 mg) had nmr spectrum and melting point\u00E2\u0080\u00A2(49-50\u00C2\u00B0) consistent with 107. Treatment of l,3-Diformyl-2,4,5,6-tetramethoxybenzene (107) With S u l f u r i c Acid and Hydrogen Peroxide 1\u00C2\u00B0 8' a) l,3-Diformyl-2,4,5,6-tetramethoxybenzene (107) (18 mg; 0.078 mmol) was dissolved i n g l a c i a l a c e t i c acid (3 ml) and cooled i n an i c e bath. To the cooling s o l u t i o n was added an i c e cold mixture of 30% hydrogen peroxide (0.4 ml; 3.52 mM) and 50% s u l f u r i c acid (1.5 ml) and this mixture was allowed to stand at 4\u00C2\u00B0 for 16 hours. The orange reaction mixture was then poured into water (25 ml) and the s o l u t i o n was brought to ^ pH 6 with sodium carbonate. The s o l u t i o n was extracted with chloroform (3 x 20 ml) and ether (3 x 20 ml), then the aqueous layer was made quite a c i d i c with concentrated hydrochloric acid and again extracted with chloroform (2 x 40 ml). In each case the organic extracts were washed with water, dried over anhydrous sodium s u l f a t e and the solvent removed i n vacuo. The f i r s t chloroform extract (10 mg) was quite orange i n colour and appeared to contain no discernable major component when examined by t i c . The ether extract (^ 1 mg)' was not investigated. The second chloroform extract (4 mg) appeared to be mostly \"base l i n e \" material by t i c . b) l,3-Diformyl-2,4,5,6-tetramethoxybenzene (107) (28 mg; 0.11 mmol) was dissolved i n g l a c i a l a c e t i c acid (2 ml), 50% s u l f u r i c a c i d (0.5 ml) was added and the mixture was cooled to 0\u00C2\u00B0 i n an i c e bath. On addition of the acid the mixture was observed to become yellow. 30% hydrogen peroxide (0.05 ml; 0.4 mmol) was added and the mixture was s t i r r e d under nitrogen for 15 minutes, during which time no noticeable colour change occurred. The mixture was then quickly poured into i c e water (10 ml) and extracted with chloroform (3 x 15 ml). The chloroform extract was washed with brine (10 ml), dried over anhydrous sodium s u l f a t e and the solvent removed i n vacuo to y i e l d a s l i g h t l y orange o i l y residue (29.4 mg), nmr (60 MHz) i n CDC13, TMS lock, 1.62 ( s i n g l e t , aromatic -OC(O)H), 4.7 (broad, OH, phenol), 6.0 -6.2 (many peaks, OCH^). The reaction mixture (29.4 mg) was transferred to a 25 ml f l a s k i n chloroform and the solvent was removed i n vacuo. 2% Sodium hydroxide s o l u t i o n (5 ml; 2.5 mmol) was added and the mixture was\" s t i r r e d overnight. A c e t i c anhydride (2 ml) was added and s t i r r i n g was continued for 0.5 ' ' hours more (the s o l u t i o n turned purple on addition of the ac e t i c anhydride). The mixture was ne u t r a l i z e d with sodium carbonate and extracted with chloroform (3 x 15 ml), The aqueous layer was a c i d i f i e d with concentrated hydrochloric acid, saturated with sodium chloride and extracted with chloroform (3 x 15 ml). Both chloroform extracts were washed with brine (10 ml each), dried over anhydrous sodium s u l f a t e , and the solvent removed i n vacuo to y i e l d two o i l y residues (10.0 and 9.2 mg respectively) which when examined by t i c appeared to contain s i m i l a r components; a polar purple spot and a less polar colourless spot. The combined residues were further treated with a c e t i c anhydride (1 ml) and pyridine (1 ml) for 2 hours and a f t e r removal of the solvents i n vacuo, the residue (19 mg) was shown to be mainly a non-polar component which became purple on the t i c plate on prolonged exposure to a i r . The mixture was chromatographed on t i c and the major band (6 mg) was i s o l a t e d . The nmr spectrum (60 MHz i n CDC1-) contained a.signal i n d i c a t i n g the presence of a small amount of aldehyde, at x-0.24. The methoxyl region was very complex and a peak at x7.7 suggested some acetate was present i n the mixture. 1,3-Diacetoxy-2,4,5,6-tetramethoxybenzene (108b) l,3-Diformyl-2,4,5,6-tetramethoxybenzene (107) (12.5 mg; 0.04 mmol) was dissolved i n g l a c i a l a c e t i c acid (1 ml), 50% s u l f u r i c acid (0.25 ml) was added and the mixture was cooled to 0\u00C2\u00B0 i n an i c e bath. The s o l u t i o n became pale yellow. With the system closed to a i r , nitrogen was bubbled through the s o l u t i o n for 10 minutes and then 30% hydrogen peroxide (0.03 ml; 0.26 mmol) was added and the s o l u t i o n was allowed to stand at 0\u00C2\u00B0 with nitrogen passage continued for 20 minutes, during which time no discernable colour change took place.' The mixture was then quickly poured into i c e cold brine (10 ml) and extracted with chloroform (3 x 15 ml). The chloroform extract was washed with brine (2 x 10 ml), dried over anhydrous sodium s u l f a t e and the solvent removed i n vacuo. The residue was dissolved i n anhydrous ether (5 ml) and transferred to a dry f l a s k and the s o l u t i o n was cooled i n an i c e bath. Methyllithium (2.1 M i n ether, 0.25 ml, 0.505 mmol). was added with s t i r r i n g and a f i n e white p r e c i p i t a t e was observed to form immediately. The mixture was allowed to come to room temperature and s t i r r i n g was continued f o r 5 minutes. Acetic anhydride (.0.5 ml) and pyridine (1 ml) were then added and s t i r r i n g was continued for 3 hours. Water (5 ml) was added and the mixture was extracted with chloroform (3 x 15 ml). The chloroform extract was washed with brine (10 ml), dried over anhydrous sodium s u l f a t e and the solvent removed i n vacuo to y i e l d a colourless o i l (16 mg). Preparative t i c (eluting with solvent A) allowed i s o l a t i o n of 1,3-diformyl-2,4,5,6-tetramethoxybenzene (108b) (14.9 mg, 97% y i e l d ) as a colourless o i l bp ^ 110\u00C2\u00B0 at 0.5 mm. M i c r o d i s t i l l a t i o n provided an a n a l y t i c a l sample. C r y s t a l l i z a t i o n from anhydrous ether - hexane yielded 108b as colourless MeOH plates , mp 57-58\u00C2\u00B0; i r (film) 1740 (aromatic acetate, C=0); uv ^ m a x (O 202.5 (40,000), 272 (992); nmr (100 MHz) i n CDC13, TMS lock, 6.14 (3H, s i n g l e t , aromatic OCH^), 6.17 (6H, s i n g l e t , two aromatic OCH^), 6.28 (3H, s i n g l e t , aromatic OCH^), 7.68 (6H, s i n g l e t , two aromatic 0C(0)CH 3; mass spectrum m/e 314 (M), 272 (M-42), 230 (M-84), 215. Anal. Calcd. for C i y H 1 o 0 o : C, 53.50; H, 4.77. Found: C, 53.77; H, 4.89. IH l o o High r e s o l u t i o n molecular weight determination. Calcd. for C .H 1 o0 o : IH l o o 314.098. Found: 314.100. Demethylation of Isopimpinellin ( 2 ) 1 1 1 Isopimpinellin (2) (20.8 mg; 0.084 mmol) was placed i n a three necked f l a s k with hydroiodic acid (3 ml). The apparatus was set up such that one i n l e t was f i t t e d with a nitrogen j e t which blew nitrogen over the surface of the l i q u i d . In another neck was placed a condensor, connected to a trap (containing red phosphorous (1 gm) and cadmium s u l f a t e (1 gm) i n water (100 ml)), i n turn connected to a U-tube containing anhydrous calcium carbonate and then to two successive traps containing 10% trimethylamine i n methanol (100 ml each), kept at -78\u00C2\u00B0. The reaction mixture was refluxed for 3 hours while nitrogen was passed through the system. .The contents of the last two traps were combined and the solvent removed under reduced pressure to yield tetramethylammonium iodide (109) which was washed with dry methanol (28.8 mg; 82.5% yield). Tetramethylammonium picrate (110) Tetramethylammonium iodide (109) was dissolved in water (y 0.2 ml) and a hot, saturated solution of picric acid in water (1 ml) was added. The mixture was allowed to cool, whence a red, precipitate formed. The supernatant liquid was removed and the precipitate was crystallized from methanol to yield tetramethylammonium picrate (110) as yellow needles mp 327-328\u00C2\u00B0 ( l i t . 1 1 1 mp 312-313\u00C2\u00B0 ) . Anal. Calcd. for C-^H^N^O. : C, 39.74; H, 4.67; N, 18.54. Found: C, 39.88, H, 4.45; N, 18.26. Selective Ozonolysis of Alloimperatorin methyl ether (7) Alloimperatorin methyl ether (7) (100 mg; 0.352 mmol) was treated with ozone saturated glacial acetic acid (100 ml; 0.530 mmol ozone) was stirred for five hours at room temperature. Zinc dust (200 mg) was added and stirring continued for 12 hours more. The solution was filtered to remove zinc and the filtrate was steam disti l led for 10 minutes into 2,4-dinitrophenylhydrazine reagent (2,4-dinitrophenyl-hydrazine (500 mg) in concentrated sulfuric acid (10 ml) and diluted to 60 ml with water). The 2,4-dinitrophenylhydrazine solution was extracted with benzene (3 x 30 ml), the extract was washed with water, dried over anhydrous sodium sulfate, concentrated to a small volume and chromatographed on Woelm acid washed alumina (Activity I, 30 g). The fractions eluted with benzene contained only acetone 2,4-dinitrophenyl-hydrazine (111a) (by t i c ) . Crystallization pf the residues of these fractions from methanol yielded 111a C36 mg; 43% yield) as yellow needles, mp 123-125\u00C2\u00B0 (lit. : 1. 2.? mp 128\u00C2\u00B0), mixed mp with 111a prepared from acetone, 123-125\u00C2\u00B0. The non-volatile portion of the reaction mixture was extracted with chloroform (3 x 20 ml). The extracts were dried over anhydrous sodium sulfate and the solution was concentrated to a small volume and was applied to a preparative t i c plate. Elution with solvent B allowed isolation of alloimperatorin methyl ether (7) (41 mg; 41% recovery) and a less polar band (yellow, uv), the expected aldehyde (112) (45.5 mg; 53% yield). Crystallization from ethyl acetate yielded 112 as an MeOH amorphous reddish powder, mp 195-200\u00C2\u00B0; uv 219, 245 (sh) , 250, 265, 307; nmr (100 MHz) in CDCl3-DMS0-d6, TMS lock, 0.21 (IH, multiplet, CHO), 1.85 (IH, doublet, J=10 Hz, H(4) of furanocoumarin), 2.04 (IH, doublet, J=2 Hz, H(7) of furanocoumarin), 2.89 (IH, doublet, J=2 Hz, H(6) of furanocoumarin), 3.67 (IH, doublet, J=10 Hz, H(3) of furanocoumarin), 5.64 (2H, doublet, J=Hz, CH^CHO), 5.84 (3H, singlet, aromatic 0CH3) ; mass spectrum m/e 258 (M), 229 (base peak, M-15), 214, 201, 186, 158. 5-(2'-Hydroxyethyl)-8-methoxypsoralen (113) To a solution of the aldehyde (112) (19.8 mg; 0.077 mmol) i n a methanol-chloroform mixture (5 ml) at 0\u00C2\u00B0C, was added dropwise to an ice cold solution of sodium borohydride (100 mg; 2.6 mmol) in methanol (5 ml). The mixture was stirred at 0\u00C2\u00B0 for 2 hours at which time, water (5 ml) was added to destroy the excess reagent. The solution was extracted with chloroform (5-x 15 ml), the extract was dried over anhydrous sodium sulfate and the solvent was removed under reduced pressure to yield 5-(2'-hydroxyethyl)-8-methoxypsoralen (.113) (17.5 mg; 88% y i e l d ) . C r y s t a l l i z a t i o n from e t h y l acetate followed by sublimation (145\u00C2\u00B0, 0.03 mm) y i e l d e d an a n a l y t i c a l sample of 113, mp 167.0-169.0\u00C2\u00B0; i r (KBr) 3450 (OH), MeOH 1704, 1690, 1585 (a-pyrone); uv A (e) 220 (24,600)., 245 (sh) (18,600), max 251 (20,800), 265 (17,500), 306 (12,770); nmr (100 MHz) i n CDC13, TMS lock, 1.93 (IE, doublet, J=10 Ez, E(4) of furanocoumarin), 2.33 (IE, doublet, J=2 Ez, E(7) of furanocoumarin), 3.12 (IE, doublet, J=2 Ez, E(6) of furanocoumarin), 3.76 (IE, doublet, J=10 Ez, E(3) of furanocoumarin), 5.79 (3E, s i n g l e t , aromatic OCE^), 6.07 (2H, broad t r i p l e t becoming a sharp t r i p l e t on addition of D 20, J=6 Hz, CH 2-CE 2-0E), 6.75 (2H, t r i p l e t J=6 Hz, CH_2-CE2-0E), 8.38 (IH, broad, disappearing on addition of D 20, CE 2-CE 2-0E); mass spectrum m/e 260 (M), 242 (M-18), 229 (base peak, M-31), 214 (M-46), 201, 186, 158. Anal. Calcd. for C 1 4 H 1 2 \u00C2\u00B0 5 : C ' 6 4 - 6 2 ' H ' 4.62. Found: C, 64.66; E, 4.61. Eigh r e s o l u t i o n molecular weight determination. Calcd. f o r ^2.^12^5 * 260.068. Found: 260.069. C a t a l y t i c Eydrogenation of Alloimperatorin methyl ether (7) Alloimperatorin methyl ether (7) (30.2 mg; 0.106 mmol) was dissolved i n 95% ethanol (10 ml) and placed i n a pressure compensating dropping funnel and attached to a f l a s k containing 5% palladium on charcoal (10 mg) i n 95% ethanol (10 ml). The system was placed i n a hydrogen atmosphere and allowed to equibrate for 1 hour. The s o l u t i o n of 7 was then added to the c a t a l y s t and s t i r r e d while hydrogen uptake was observed. Eydrogen uptake appeared to be smooth u n t i l about.2 ml (^ 0.8 molar equivalents) of hydrogen appeared to have been taken up. The ca t a l y s t was then f i l t e r e d o f f and the solvent was removed from the reaction mixture to y i e l d an o i l y residue (31 mg) which was one spot (blue, uv) on t i c , s l i g h t l y less polar than the s t a r t i n g m aterial.' An nmr spectrum revealed t h i s compound to be tetrahydroalloimperatorin methyl ether; nmr (60 MHz) i n CDC13, 2.17 (IH, doublet, J=10 Hz, H(4) of.coumarin), 3.75 (IH, doublet, J=10 Hz, H(3) of coumarin), 5.23 (2H, t r i p l e t , J=9 Hz, CH^-CH.-O of dihydrobenzofuran), 5.95 (3H, s i n g l e t , aromatic 0CH\u00E2\u0080\u009E), 6.78 (2H, t r i p l e t , J=9 Hz, CH2-CH2-0 of dihydrobenzofuran), 7.30 (2H, apparent t r i p l e t , J=8 Hz, aromatic-CH_2-CH2) , 8.0-8.9 (y 5H, methylene envelope plus impurity), 9.0 (6H, doublet, J=6 Hz, CH(CH_ 3) 2). Attempted Reduction of Alloimperatorin methyl ether (7) With Diimide a) Alloimperatorin methyl ether (7) (30 mg; 0.105 mmol) and dipotassium azodicarboxylate (prepared from azodicarboxamide by the method of T h i e l e 1 2 3 ) (50 mg; 0.260 mmol) were dissolved i n dry pyridine (5 ml) and a c e t i c ac i d (34 mg; 0.54 mmol) i n dry pyridine (5 ml) was added over 2 h o u r s 1 1 2 . The reaction was s t i r r e d under nitrogen for 22 hours at room temperature. Water (0.5 ml) was then added, the solvent was removed i n vacuo and the residue was taken up i n chloroform. An nmr spectrum of the soluble material revealed that no reduction had taken place. b) Alloimperatorin methyl ether (7) (30 mg; 0.105 mmol) was dissolved i n a methanol-water mixture (10:1) (10 ml) and dipotassium azodicarboxy-l a t e (250 mg; 1.30 mmol) was ad d e d 1 2 4 . The mixture was s t i r r e d one hour at room temperature, the solvent was removed under reduced pressure and the residue was taken up i n chloroform. An nmr spectrum revealed only s t a r t i n g material, 7, was present. leaf 173 ommitted in page numbering 5-(2',3'-dihydroxy-3'-methylbutyl)-8-methoxypsoralen. (3) (Alloimperatorin. methyl ether d i o l ) Alloimperatorin methyl ether epoxide (97) (100 mg; 0.333 mmol) (prepared from alloimperatorin methyl ether (7) as described previously) was refluxed with 5% aqueous o x a l i c acid (30 ml) for 1 hour. The s o l u t i o n was allowed to cool and was then extracted with chloroform (3 x 30 ml). The chloroform extract was washed with 5% sodium carbonate s o l u t i o n (20 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (103.3 mg) which was chromatographed on a preparative t i c p l a t e ( e l u t i n g with solvent B) to y i e l d the s t a r t i n g epoxide (97) (21.3 mg)\"and alloimperatorin methyl ether d i o l (3)-(74mg, 70% y i e l d ) . ' C r y s t a l l i z a t i o n . o f ; 3 from ethyl acetate yielded a .pure sample, mp 176-177\u00C2\u00B0 ( l i t . 6 mp 174-176\u00C2\u00B0),'mixed'mp with authentic 3 (obtained from Dr. D.L. Dreyer), 176-177\u00C2\u00B0 Acetone \u00C2\u00A3-bromobenzenesulfonylhydrazone (111b) jp_-Bromobenzenesulfonylhydrazide (125 mg; 0.498 mmol) was dissolved i n g l a c i a l a c e t i c (10 ml) and the s o l u t i o n was d i l u t e d with water (10 ml). Excess acetone was added, the s o l u t i o n s t i r r e d , then allowed to stand for 5 minutes. The mixture was then extracted with chloroform, washed with water, dried over anhydrous sodium s u l f a t e and the solvent removed under reduced pressure. The residue was c r y s t a l l i z e d from chloroform-petroleum j ether to y i e l d acetone \u00C2\u00A3-bromobenzenesulf onylhydrazone (111b) , as colourless needles, mp 146-148\u00C2\u00B0 ( l i t . 1 1 3 mp 145-146\u00C2\u00B0); i r (KBr) 3220 (NH), MeOH 1342, 1180 (RSO N); uv X (e) 235 (14,000); nmr (100 MHz) i n CDC1\u00E2\u0080\u009E, \u00C2\u00AB- XTlcLX Jt TMS lock, 2.28 (4H, .AB mul t i p l e t , para-disubstituted benzene), 3.22 (IE, broad, disappears on addition of D 20, NE) , 8.10, 8.22 (6E, two s i n g l e t , NHN=C(Q1 3) 2 c i s and trans); mass spectrum m/e 292, 290 (M), 221, 219 (M-71), 205, 203 (M-87) , 157, 155 (M-135), 71 (base peak). Anal. Calcd. for C ^ ^ S O ^ r : C, 37.1; H, 3.8; N, 9.62. Found: C, 36.86; E, 3.88; N, 9.40. P u r i f i c a t i o n of Acetone p_-broinobenzenesulfonylhydrazone (111b) Pure acetone p_-bromobenzenesulfonylhydrazone (111b) (18.5 mg) was placed on a s i l i c a gel preparative t i c plate and eluted with hexane-ether (1-3), the band corresponding to 111b was scraped from the pl a t e and extracted with chloroform and methanol. Recovery of 111b was 6 mg. On an alumina preparative t i c p l a t e was placed pure 111b (15 mg) and e l u t i o n was c a r r i e d out with chloroform-methanol (95:5). Removal of the appropriate band and extraction with chloroform and methanol y i e l d e d 111b (8.5 mg). Pure 111b (15 mg) was mixed with jv-bromobenzenesulfonylhydrazide (200 mg) and chromatographed on an alumina column ( a c t i v i t y IV; 15 g). E l u t i o n was sta r t e d with benzene. The fr a c t i o n s eluted with chloroform-methanol contained both the reagent and 111b with l i t t l e separation apparently achieved. Pure 111b (17.7 mg) was chromatographed on an alumina preparative t i c p l a t e , e l u t i n g with methanol-chloroform (1:20). The appropriate band was quickly removed from the plate before the solvent could evaporate and extracted with chloroform and methanol to y i e l d 111b (15.7 mg). Reaction of 5-(2',3'-dihydroxy-3'-methylbutyl)-8-methoxypsor^ With p e r i o d i c acid a) 5-(2',3'-dihydroxy-3'-methylbutyl)-8-methoxypsoralen (3) (20 mg; 0.063 mmol) was dissolved i n methanol (2 ml) (acetone-free) and p e r i o d i c acid (45 mg; 0.197 mmol) i n water (2 ml) was added.- The mixture was s t i r r e d for 12 hours at room temperature, then the reaction mixture was steam d i s t i l l e d into p_-bromobenzenesulfonylhydrazide reagent (p-bromo-benzenesulf onylhydrazide (125 mg) i n g l a c i a l a c e t i c acid (10 ml), d i l u t e d with water (10 ml)). The hydrazone mixture was allowed to stand f o r 30 minutes and then was extracted with acetone-free chloroform (3 x 30 ml). The chloroform extract was washed with water (20 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a white residue (129 mg) which when separated by preparative t i c on alumina (chloroform-methanol, 20:1) yi e l d e d acetone p-bromobenzenesulfonyl-hydrazone (111b) (15.6 mg; 85% y i e l d ) , i d e n t i c a l with 111b prepared from acetone. The n o n - v o l a t i l e portion of the reaction mixture was d i l u t e d with water (10 ml) and extracted with chloroform (6 x 15 ml). The chloroform extract was washed with water (15 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (^ 20 mg) which was dissolved i n a chloroform-methanol mixture (5 ml) . and cooled to 0\u00C2\u00B0 i n an i c e bath. Sodium borohydride (200 mg; 5.2 mmol) i n i c e - c o l d methanol (5 ml) was added and the mixture was s t i r r e d 2 hours at i c e temperature. Water (10 ml) was added, the mixture was concentrated under reduced pressure and the aqueous s o l u t i o n was extracted with chloro-form (3 x 15 ml). The chloroform extract was washed with water (10 ml) dried over anhydrous sodium sulfate and the solvent was removed under reduced pressure to yield a residue (17.6 mg) which was subjected to chromatography on preparative tic (eluting with ethyl acetate-chloroform, 2:1). The band corresponding to 5-(2'-hydroxyethyl)-8-methoxypsoralen (113) was isolated (12.0 mg). This material resisted crystallization but some powdery material was obtained from ethyl acetate, mp ^ 210\u00C2\u00B0 ; i r (CH.C1.) 2220 (B-H?): nmr (60 MHz) in CDC1\u00E2\u0080\u009E as observed previously for 113. This material (12.0 mg) was dissolved in chloroform-methanol (5 ml) and hydrochloric acid (2 molar, 10 drops) was added. The mixture was shaken and allowed to stand 15 minutes. Water (10 ml) was added and the mixture was extracted with chloroform (3 x 15 ml). The chloroform extract was washed with water (10 ml), dried over anhydrous sodium sulfate.and the solvent was removed under reduced pressure to yield a residue which crystallized from ethyl acetate as white needles, mp 155-150\u00C2\u00B0 , tic and i r properties as observed for 113. b) 5-(2',3'-dihydroxy-3'-methylbutyl)-8-methoxypsoralen (3) (48.8 mg; 0.153 mmol) was dissolved in acetone-free methanol (5 ml), periodic acid (100 mg; 0.437 mmol) in water (5 ml) was added and the mixture was stirred for 1 hour. The reaction mixture was then heated to ^ 30\u00C2\u00B0 in a water bath, with stirring, while a stream of nitrogen was blown over the surface of the solution and the effluent gases were bubbled through a solution of _p_-bromobenzenesulfonylhydrazide (125 mg) in glacial acetic acid (10 ml) and water (10 ml), for 0.5 hours. The hydrazone mixture was extracted with acetone-free chloroform (3 x 15 ml). The chloroform extract was washed with water (15 ml), dried over anhydrous sodium sulfate and the solvent was removed under reduced pressure to yield a white residue (110 mg) from which, acetone p-bromobenzenesulfonylhydrazone (111b) (15 mg; 35% y i e l d ) was i s o l a t e d by chromatography on an alumina preparative t i c plate ( e l u t i n g with chloroform-methanol, 20:1). The non - v o l a t i l e portion of the reaction mixture was extracted with chloroform (3 x 20 ml), the extracts were washed with sodium bicarbonate, dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (43.5 mg) which was dissolved i n methanol-chloroform (5 ml), cooled to 0\u00C2\u00B0. To i t was added sodium boro-hydride (100 mg; 2.6 mmol) i n i c e - c o l d methanol (5 ml). The mixture was s t i r r e d at 0\u00C2\u00B0 for 0.5 hours then water (10 ml) was added and the mixture was extracted with chloroform (3 x 20 ml). The chloroform extract was washed with water, dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue which was chromato-graphed on preparative t i c (solvent B) to y i e l d 5-(2'-hydroxyethyl)-8-methoxycoumarin (113) (19.0 mg; _\"4.7-2Lyield) . C r y s t a l l i z a t i o n from et h y l acetate y i e l d e d needles, mp 167-169\u00C2\u00B0, 113 by t i c and mixed mp (167-169\u00C2\u00B0). 5-(2'-acetoxy-3'-hydroxy-3'-methylbutyl)-8-methoxypsoralen (114) 6 5-(2',3'-dihydroxy-3'-methylbutyl)-8-methoxypsoralen (3) (101 mg; 0.315 mmol) was dissolved i n pyridine (5 ml) and a c e t i c anhydride (0.5 ml) and the mixture was s t i r r e d at room temperature f o r 12 hours. The solvents were removed i n vacuo and the residue was r e c r y s t a l l i z e d from ethanol to y i e l d 5-(2'-acetoxy-3'-hydroxy-3 1-methylbutyl)-8-methoxypsoralen (114) (97.9 mg; 86% yield), .mp 184-186 ( l i t . 6 mp 183-185\u00C2\u00B0); nmr (60 MHz) i n CDC13, 1.94 (IE, doublet, J=10 Ez, E(4) of furanocoumarin), 2.35 (IH, doublet, J=2HzV E(7.) of furanocoumarin), 3.12 (IH, doublet, J=2 Hz, H(6) of furanocoumarin), 3.64 (IH, doublet, J=10 Hz, H(3) of furanocoumarin), 5.00 (IH, doublet of doublets, J=4 Hz and 10 Hz, CH^-CHR-OCOCH^), 5.73 (3H, s i n g l e t , aromatic 0CH_3) , 6.65 (2H, AB of ABX mu l t i p l e t , J =15 Hz, J A =10 Hz, J'=b Hz, CHo-CH(R)-0) , 7.98 (IH, broad, disappearing AB AX BX \u00E2\u0080\u0094 Z on addition of D.O, OH), 8.23 (3H, s i n g l e t , OCOCH^), 8.64 (6H, s i n g l e t , -C(OH)(CH 3) 2). 5-(2'-acetoxy-3'-hydroxy-3'-methylbutyl)-6-formyl-7-hydroxy-8-methoxy- coumarin (115) The monoacetate (114) (250 mg; 0.69 mmol) i n g l a c i a l a c e t i c a c i d (5 ml) was s t i r r e d with ozone saturated g l a c i a l a c e t i c acid (200 ml; 0.780 mmol ozone) f or 3 hours at room temperature. Zinc dust (400 mg) was then added and s t i r r i n g was continued f o r 2 hours more. The s o l u t i o n was f i l t e r e d to remove zinc and solvent was removed i n vacuo to y i e l d a residue, which was treated with hot chloroform. The chloroform s o l u t i o n was f i l t e r e d , concentrated and chromotographed by preparative t i c ( e l u t i n g with chloroform-ethyl acetate, 2:1). The major yellow (uv) band which appeared to contain some s t a r t i n g material was thus i s o l a t e d . This material was p a r t i t i o n e d between chloroform (30 ml) and 1% aqueous potassium hydroxide s o l u t i o n (30 ml). The non-base soluble material (^ 100 mg) appeared to contain some s t a r t i n g acetate (114) and a less polar component (blue spot; uv) but was not examined further. The aqueous layer was a c i d i f i e d and extracted with chloroform (4 x 15 ml) and e t h y l acetate (2 x 15 ml). The extracts were combined, washed with water (20 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (151 mg) which was c r y s t a l l i z e d from et h y l acetate to y i e l d 5-(2'-acetoxy-3'-hydroxy-3'-methyl-butyl)-6-formyl-7-hydroxy-8-methoxycoumarin (115) (69 mg; 27.3% y i e l d ) , mp 162-164\u00C2\u00B0; i r (KBr) 3520 (OH), 1745 (C=0), 1730, 1625, 1595 (a-pyrone); MeOH MeOH uv A (e) 273 (25,800), 345 (sh) (7,940); uv \ (e) (+ NaOH) 236 max > max MeOH (20,900), 267.5 (14,450), 300 (11,550), 357 (11,950); uv A u e u (e) (+ HC1) max 205 (44,100), 263 (10,150), 326 (13,700); nmr (100 MHz) i n CDC13, TMS locks-2.37 (IH, broad, disappears on addition of D2O, phenolic OH), -0.39 (1H,, s i n g l e t , CHO), 2.00 (IH, doublet, J=10 Hz, H(4) of coumarin), 3.70 (IH, doublet, J=10 Hz, H(3) of coumarin), 5.04 (IH, doublet of doublets, J=4 Hz and 10 Hz, CH 2-CH(R)-0C0CH 3), 6.02 (3H, s i n g l e t , aromatic 0CH 3), 6.3-6.8 (2H, AB of ABX mul t i p l e t , J . =15 Hz, J A =10 Hz, =4 Hz, CH -A B AX. BA \u00E2\u0080\u0094Z CH(R)-O), 7.94 (IH, broad, disappears on addition of D 20,0H), 8.29 (3H, singlet.-OCOCH^), 8.68 (6H, s i n g l e t , -C(OH) (CH 3) 2; mass spectrum m/e 364 (M), 346 (M-18), 245 (base peak, M-119). Anal. Calcd. for C, o H o \u00E2\u0080\u009E 0 o : C, 59.34; H, 5.49. Found: C, 59.59; H, 5.51. l o zu o High r e s o l u t i o n molecular weight determination. Calcd. for C. 0H\u00E2\u0080\u009E 0 : l o ZU o 364.116. Found: 364.114.. -5-(2 '-acetoxy-3 '-hydroxy-3 '-methylbutyl)-6-formyl-.7 ,8-dimethoxycoumarin (116) The phenolic aldehyde (115) (23 mg; 0.063 mmol) from the previous reaction was dissolved i n acetone (7 ml) and potassium carbonate (1.1 g) and methyl iodide (2.2 ml) were added. The mixture was refluxed f o r 40 minutes, allowed to cool, then water (10 ml) was added and the s o l u t i o n was - a c i d i f i e d with\" .concentrated hydrp.chloric acid.' The so l u t i o n was / extracted with chloroform (3 x 15 ml), the extract was dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a residue (35 mg) which was observed to be e s s e n t i a l l y one component by t i c . Preparative t i c (solvent A) allowed i s o l a t i o n of pure 5\u00E2\u0080\u0094(2 1^acetoxy-3'-hydroxy-3'-methylbutyl)-6-formyl-7,8-dimethoxy-coumarin (116) (19 mg; 80% y i e l d ) , which c r y s t a l l i z e d from anhydrous ether - hexane as colourless p l a t e s , mp 116-118\u00C2\u00B0; i r (KBr) ^ 3500 (OH), 1735 (C=0, acetate and pyrone), 1675 (C=0, aldehyde); uv (e) 219 (16,900), 266 (23,600), 305 (sh) (9,400); nmr (100 MHz) i n CDC1-, TMS lock, -0.62 (IH, s i n g l e t , CHO), 1.93 (IH, doublet J=10 Hz, H(4) of coumarin), 3.66 (IH, doublet, J=10 Hz, H(3) of coumarin), 4.96 (IH, doublet of doublets, J=3 Hz, CH-CHX-OOCOCH^), 5.97, 6.05 (6H, two s i n g l e t s , two aromatic 0CH~), 6.18 (IH, doublet of doublets, J=3 Hz and 14 Hz, HCH-CH(R)-O), 6.81 (IH, doublet of doublets, J=10 Hz and 14 Hz, HCH-CH(R)-O), 8.1 (IH, broad, disappears on addition of D-O, OH), 8.34 (3H, s i n g l e t , 0C0CH-) , 8.66, 8.74 (6H, two s i n g l e t s , C(0H) (CH.) _ ) ; mass spectrum m/e 378 (M), 318 (M-60), 303, 277, 259 (base peak, M-119), 245. Anal. Calcd. f o r C i nH o.0 Q : C, 60.31; H, 5.86. Found: C, 60.00; H, xy LL O 5.61. 5-(2 '-Acetoxy-3 '-hydroxy-3 '-methylbutyl)-6-acetoxy-7,8-dimethoxycoumarin (117b) Compound 116 (20 mg; 0.053 mmol) from the previous reaction, was dissolved i n g l a c i a l a c e t i c acid (2 ml) and the s o l u t i o n was cooled i n an i c e bath u n t i l the l i q u i d began to s o l i d i f y . An i c e cold mixture of 50% s u l f u r i c acid (0.75 ml) and 30% hydrogen peroxide (0.20 ml; 1.77 mmol) was then added and the s o l u t i o n was s t i r r e d at i c e temperature for a few minutes. The mixture was placed i n a r e f r i g e r a t o r and allowed to stand at 4\u00C2\u00B0 f o r 18 hours. Water (10 ml) was, then added and the mixture was extracted with chloroform (3 x 15 ml). The chloroform extract was washed with brine (10 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed i n vacuo to y i e l d a residue (25 mg) which was chromatographed on a preparative t i c plate (solvent B). Extraction of the major band from the plate y i e l d e d 5-(2'-acetoxy-3'-hydrpxy-3 ,-methylbutyl)-6-hydroxy-7,8-dimethoxycoumarin (117a) (18 mg; 93% y i e l d ) ; which r e s i s t e d c r y s t a l -l i z a t i o n ; nmr (60 MHz) i n CDC\u00C2\u00B1 3 2.01 (IH, doublet, J=10 Hz, H(4) of coumarin), 3.63 (IH, doublet, J=10 Hz,.H(3) of coumarin), 3.86 (IH, broad, disappears on addition of D_0, phenolic OH), 4.95 (IH, doublet of doublets, J=3.5 Hz and 9 Hz, CH 2CH(R)0C0CH 3), 5.90, 5.98 (6H, two s i n g l e t s , two aromatic 0CH 3), 6.5-7.1 (2H, AB of ABX m u l t i p l e t , CH 2CH(R)0-), 7.98 (IH, broad, disappears on addition of D^O, OH), 8.18 (3H, s i n g l e t , OCOCH^), 8.67 (6H, s i n g l e t , C(0H) (CH_3)2) . This material was treated with a c e t i c anhydride (1 ml) i n pyridine (2 ml) for 12 hours. Removal of the solvents i n vacuo y i e l d e d an o i l y residue which c r y s t a l l i z e d on standing. Recrystal-l i z a t i o n from et h y l acetate y i e l d e d 5-(2'-acetoxy-3-hydroxy-3'-methylbutyl)-6-acetoxy-7,8-dimethoxycoumarin (117b) (9.5 mg; 44% o v e r a l l y i e l d ) , mp 143-144\u00C2\u00B0; i r (KBr) 1770 (aromatic acetate C=0), 1740 ( a l i p h a t i c acetate MeOH C=0), 1705, 1592 (a-pyrone); uv X (e) 208 (36,200), 248 (sh) (4,600), max 303 (11,800); nmr (100 MHz) i n CDC13, TMS lock, 2.05 (IH, doublet, J=10 Hz, H(4) of coumarin), 3.65 (IH, doublet, J=10 Hz, H(3) of coumarin), 5.04 (IH, doublet of doublets, J=5 Hz and 8 Hz, CH 2CH(R)0C0CH 3), 6.01, 6.06 (6H, two s i n g l e t s , two aromatic 0CH_3) , 6.99 (2H, AB of ABX m u l t i p l e t , <- 16 Hz, J M = 8 Hz, J B X=5 Hz, CH 2-CH(R)-0-)y 7.67 (3H, s i n g l e t , aromatic 0C0CH_3) , 8.17 (4H, becoming 3H on addition of D 20, s i n g l e t , a l i p h a t i c OCOCH- and OH), 8.74 (6H, s i n g l e t , C(OH)(CH-)\u00E2\u0080\u009E); mass spectrum m/e 408 (M), 366 (M-42), 351 (M-57), 348 (M-60), 306 (M-102), 288, 273 (base peak, C 1 5 H 1 3 0 5 ) , 235, 149. Anal. Calcd. f o r C^H^O- : C, 58.82; H, 5.92. Found: C, 59.08; H, 5.88. High r e s o l u t i o n molecular weight determination. Calcd. f o r ^20^24^9 : 408.142. Found: 408.145. l-(2'-acetoxy-3'-hydroxy-3'-methylbutyl)-2,6-diformyl-3,5-dihydroxy-4- methoxybenzene (118) The monoacetate (114) (100 mg; 0.278 mmol) was dissolved i n g l a c i a l a c e t i c acid (5 ml) and eth y l acetate (2 ml) and the s o l u t i o n was subjected to a stream of ozone enriched oxygen at -78\u00C2\u00B0 f o r 0.5 hours. Zinc dust (150 mg) was then added and the mixture was s t i r r e d at room temperature for a further 0.5 hours. The s o l u t i o n was then f i l t e r e d and the solvents were removed i n vacuo to y i e l d a residue which was treated with hot chloroform (50 ml). The chloroform s o l u t i o n was f i l t e r e d and the solvent was removed under reduced pressure to y i e l d an o i l y residue (115 mg). C r y s t a l l i z a t i o n of th i s material from e t h y l acetate afforded l - ( 2 ' -acetpxy-3 1-hydroxy-3 1-methylbutyl)-2,6-diformyl-3,5-dihydroxy-4-methoxy-benzene (118) (51.6 mg; 54% y i e l d ) as pl a t e s , mp 188-190\u00C2\u00B0; i r (KBr) 3535 (OH), 1735 (C=0, acetate), 1630 (C=0, aldehyde); uv AM E 0 H (e) 270 (32,400); max MeOH uv X e u (e) (+ NaOH) 266 (sh) (14,700), 292.5 (25,300), 340 (9,330); m_L--MeOH uv X (+ HC1) 224 (10,200), .270 (19,200); nmr (100 MHz) i n CDC1\u00E2\u0080\u009E, TMS ITI_L_- j lock, -2.82,(2H, s i n g l e t , disappears on addition of -\u00C2\u00BB0, two equivalent phenolic OH), -0.28 (2H, s i n g l e t , two equivalent CHO), 5.00 (IH, doublet of doublets, J=4 Hz and 10 flz, CH2CH(R)OCOCH3), 6.09 (3H, s i n g l e t , aromatic 0CH\u00E2\u0080\u009E) , 6.36 (2H, AB of ABX m u l t i p l e t , J . =15 Hz, J. =10 Hz, \u00E2\u0080\u00943 AD AX JL =4 Hz), 8.04 (IH, broad, disappears on addition of D.O, OH), 8.16 BX z \u00E2\u0080\u0094 (3H, singlet,0C0CH_ 3) , 8.70 (6H, s i n g l e t , C(0H) (CH^) 2) ; mass spectrum m/e 340 (M) 322 (M-18), 298, 280, 262, 221 (base peak). Anal. Calcd. for 0,,H0_00 : C, 56.47; H, 5.92. Found: C, 56.24; 5.91. \u00C2\u00B1 0 zu o High r e s o l u t i o n molecular weight determination. Calcd. for C.^Hor.0o : lo z(J o 340.116. Found: 340.120. l-(2'-acetoxy-3'-hydrpxy-3'-methylbutyl)-2,6-diformyl-3,4,5-trimethoxy- benzene (119) Compound 118 (from the previous reaction) (22.7 mg; 0.0628 mmol) was dissolved i n acetone (5 ml) and to i t was added potassium carbonate (0.5 g) and methyl iodide (2 ml). The mixture was refluxed f or 1.5 hours, then water (10 ml) was added and the mixture was s t i r r e d 10 minutes more. The mixture was then a c i d i f i e d with concentrated hydrochloric acid and extracted with chloroform (3 x 15 ml). The chloroform extract was washed with saturated brine (10 ml), dried over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure. The resultant residue was chromatographed on preparative t i c (solvent A) and the major band, l - ( 2 ' -acetoxy-3'-hydroxy-3 '-methylbutyl)-2 ,\"6-dif ormyl-3,4,5-trimethoxybenzene ;-(119) (19.3 mg; 81.5% y i e l d ) was thus i s o l a t e d . C r y s t a l l i z a t i o n from anhydrous ether - hexane yi e l d e d pure 119 as plates, mp 79.5-80.5\u00C2\u00B0; nmr (100 MHz) i n CDC1 3 > TMS lock, -0.39 (2H, s i n g l e t , two equivalent CHO), 5.00 (IH, doublet of doublets, J=5 Hz-and 10 Hz, CH 2CH(R)0C0CH 3), 6.00 (6H, s i n g l e t , two equivalent aromatic 0CH 3), 6.15 (3H, s i n g l e t , aromatic 0CH 3), 6.2-6.6 (2H, AB of ABX m u l t i p l e t , CH 2CH(R)0C0CR 3), 7.90 (IH, broad, d i s a p p e a r s on a d d i t i o n o f D 20, OH), 8.22 (3H, s i n g l e t , -OCOGHL^), 8.70, 8.74 (6H, two s i n g l e t s , C ( 0 H ) ( C H 3 ) 2 ) . A n a l . C a l c d . f o r C ^ H ^ O g :.C, 58.69; H, 6.59. Found: C, 58.63; H, 6.34. Treatment o f 119 w i t h h y d r o g e n p r o x i d e ' a n d S u l f u r i c a c i d . A s o l u t i o n o f 119 (19 mg; 0.0516 mmol) i n g l a c i a l a c e t i c a c i d ( 1 ml) was c o o l e d i n an i c e b a t h and 50% s u l f u r i c a c i d (0.25 ml) was added and the m i x t u r e was c o o l e d t o 0\u00C2\u00B0 w h i l e n i t r o g e n was b u b b l e d t h r o u g h t h e s o l u t i o n t o remove oxygen f r o m t h e sy s t e m . A f t e r 10 m i n u t e s o f s u c h t r e a t m e n t 30% h y d r o g e n p e r o x i d e (0.03 m l ; 0.265 mmol) was added and t h e n i t r o g e n f l o w was c o n t i n u e d f o r 15 m i n u t e s a t 0\u00C2\u00B0 a t w h i c h time t h e s o l u t i o n began t o t a k e on an orange c o l o u r . The m i x t u r e was t h e n poured i n t o i c e c o l d b r i n e (10 ml) and e x t r a c t e d w i t h c h l o r o f o r m (4 x 10 m l ) . The c h l o r o f o r m e x t r a c t was washed w i t h b r i n e (10 m l ) , d r i e d o v e r anhydrous sodium s u l f a t e and t h e s o l v e n t was removed under r e d u c e d p r e s s u r e t o y i e l d an orange o i l w h i c h was d r i e d i n vacuo. T h i s m a t e r i a l was t h e n d i s s o l v e d i n anhydrous e t h e r (10 m l ) , m e t h y l l i t h i u m (0.50 m l ; 1.05 mmol) was added and t h e m i x t u r e was s t i r r e d a t 0\u00C2\u00B0 f o r 15 m i n u t e s . A c e t i c a n h y d r i d e ( 1 ml) and p y r i d i n e (2 ml) was t h e n added and t h e m i x t u r e was s t i r r e d a t room t e m p e r a t u r e f o r 4 h o u r s . Water (10 ml) was t h e n added and t h e s o l u t i o n was e x t r a c t e d w i t h c h l o r o f o r m (3 x 15 m l ) . The c h l o r o f o r m e x t r a c t was washed w i t h b r i n e (10 m l ) , d r i e d o v e r anhydrous s o d i u m s u l f a t e and the s o l v e n t was removed under r e d u c e d p r e s s u r e to y i e l d an orange c o l o u r e d r e s i d u e w h i c h appeared t o c o n t a i n 4 components by t i c . P r e p a r a t i v e t i c ( s o l v e n t B) a l l o w e d s e p a r a t i o n o f two a p p a r e n t l y major bands but which constituted only ^ 4 mg and were thus examined no further. Demethylation of 5-(2'-hydroxyethyl)-8-methoxypsoralen (113) The alcohol (115) (27 mg; 0.104 mmol) was refluxed with hydroiodic acid (5 ml) and the liberated methyl iodide was trapped in a manner as described for the demethylation of isopimpinellin (2). Tetramethylammonium iodide (109) (17 mg; 81% yield) was isolated and converted to its picrate (110), mp 327-328\u00C2\u00B0, as previously described. Acid Catalyzed Hydrolysis of Umbelliprenin (83) Umbelliprenin (83) (39.5 mg; 0.108 mmol) was dissolved in glacial acetic acid (4 ml) and the solution was refluxed for 8 hours. After cooling, water (10 ml), was added and the solution was extracted with ether (4 x 20 ml). The ether extract was washed with water (20 ml), dried over anhydrous sodium sulfate and the solvent was removed under reduced pressure. The residue (41 mg) was chromatographed on preparative tic (eluting with chloroform-ethyl acetate, 1:2). The more polar band (blue; uv) was isolated to yield umbelliferone (45) (12.6 mg; 72% yield). Crystallization from ethyl acetate yielded pure 45, mp 230-231\u00C2\u00B0 ( l i t . 1 1 7 mp 2 3 2 \u00C2\u00B0 ) , mixed up with authentic umbelliferone (45), 230-231\u00C2\u00B0. The less polar band (dark, uv indicator) was isolated from the tic plate as a colourless o i l (6 mg), i r (film) 2950 (C=C). Analysis by glc (column: 20% SE 30, on 60/80 mesh chromosorb W, 1 /4\" x 10', helium flow rate 100 ml/ min, 1 6 5 \u00C2\u00B0 ) , 8 distinct but overlapping peaks between retention times 2 to Farnesol acetate (61c; R=C0CH3) Commercially obtained farnesol (61b) was dissolved i n pyridine (2 ml), a c e t i c anhydride (2 ml) was added and the mixture was s t i r r e d at room temperature for 12 hours. The solvents were removed i n vacuo and the residue was chromatographed on preparative t i c ( e l u t i n g with hexane -anhydrous ether, 10:1). The major band was i s o l a t e d as a colourless o i l , i r (film) 2950 (CH), 1740 (C=0), 1440 (C=C); nmr (60 MHz) i n CDC13, 4.8 (3H, m u l t i p l e t , CH=C), 5.21 (2H, doublet, J=7.5 Hz, 0-CH2\"CH=C), 7.98 (11 H, methylene envelope and O^COO-) , 8,31, 8.40 (12H, four v i n y l methyl groups). S o l v o l y s i s of Farnesol acetate (61c; R=-C0CH3). Farnesol acetate (61c) was dissolved i n g l a c i a l a c e t i c acid and refluxed for 12 hours. Water was added and the mixture was extracted with ether. The ether extract was washed with water, dried over anhydrous sodium s u l f a t e , and the solvent was removed under reduced pressure to y i e l d a colourless o i l . T i c examination revealed that the major component of t h i s o i l had an i d e n t i c a l retention time on t i c as did that material i s o l a t e d from the hydrolysis of umbelliprenin (83); i r (film) 2950 (CH), 1450 (C=C); g l c analysis (column; 20% SE 30, on 60-80 mesh chromosorb W, 1/4\" x 10', 100 ml/min, 165\u00C2\u00B0), 6 overlapping peaks between retention times 2.5 to 15 mins,, pattern e s s e n t i a l l y i d e n t i c a l with that observed i n hydrolysis product from umbelliprenin (83). 2,4-Dihydroxybenzpic acid (120) Umbelliferone (45) (18.0 mg; 0.111 mmol) was added to a s o l u t i o n of potassium hydroxide (260 mg) i n water (20 drops) and the mixture was heated i n a n i c k e l c r u c i b l e under a nitrogen stream, at 280\u00C2\u00B0 f o r 75 m i n u t e s 1 1 6 . A f t e r cooling, the s o l u t i o n was neut r a l i z e d with d i l u t e s u l f u r i c acid, water (20 ml) was added and the mixture was extracted with ether (6 x 35 ml); The ether extract was d r i e d over anhydrous sodium s u l f a t e and the solvent was removed under reduced pressure to y i e l d a white s o l i d (15 mg). Sublimation of th i s material (170\u00C2\u00B0; 0.01 mm) afforded 2,4-dihydroxybenzoic acid (7.5 mg; 47% y i e l d ) , mp 230\u00C2\u00B0 ( l i t . 1 2 2 mp 235-236\u00C2\u00B0); nmr (60 MHz) i n CDC1- - DMSO-dg, 1.7 (3H, broad s i n g l e t , disappears on addition of D-0, COOH and two phenolic OH), 2.32 (IH, doublet, J=9 Hz, H(6)), 3.70 (2H, m u l t i p l e t , H(3) and H(5)); mixed mp with authentic 2,4-dihydroxybenzoic acid (120) ( a v a i l a b l e commercially), 230\u00C2\u00B0; t i c properties i d e n t i c a l . Anal.' Calcd. f o r C,H,0. : C, 54.55; H, 3.92. Found: C, 54.66; H, 3.92. / o 4 Ozonolysis of Farnesol (61b) Farnesol (61b) (500 mg; 2.24 mmol) was dissolved i n g l a c i a l a c e t i c acid (10 ml) and subjected to a stream of ozone enriched oxygen at room temperature, for 5 hours. Water (10 ml) was then added and the mixture was heated i n a steam bath for 10 minutes. The s o l u t i o n was then steam d i s t i l l e d into 2,4-dinitrophenylhydrazine reagent (2,4-dinitrophenyl-hydrazine (500 mg) i n concentrated s u l f u r i c acid (10 ml) and d i l u t e d to 60 ml with water). The steam d i s t i l l a t e was c o l l e c t e d i n 1 ml fra c t i o n s i n i n d i v i d u a l vessels containing ^ 1 ml of reagent. The i n d i v i d u a l f r a c t i o n s were f i l t e r e d and the s o l i d s thus c o l l e c t e d were examined by t i c . The f i r s t f r a c t i o n s contained acetone 2,4-DNP (111a), l a t e r f r a c t i o n s contained both 111a and a s l i g h t l y more polar d e r i v a t i v e , and the l a s t (eighth) f r a c t i o n contained a red s o l i d which would not dissolve i n methanol. This compound was dissolved i n a small amount of hot nitrobenzene and one part ethanol was added. On cooling, a red s o l i d p r e c i p i t a t e d , levulinaldehyde bis-2,4-dinitrophenylhydrazone (123), (20 mg, < 1% y i e l d ) , mp 233-235\u00C2\u00B0 ( l i t . 1 1 7 mp 233\u00C2\u00B0). Ozonolysis of Farnesol ( 6 1 b ) 1 1 8 a) Farnesol (61b) (250 mg; 1.12 mmol) was dissolved i n eth y l acetate (15 ml) and subjected to a stream of ozone enriched oxygen, at -78\u00C2\u00B0 fo r 0.6 hours. The reaction mixture was allowed to warm to room temperature and then was transferred to a pressure hydrogenation b o t t l e , methanol (15 ml) was added and the s o l u t i o n was hydrogenated over 5% palladium on calcium carbonate (50 mg) at 45 p s i i n a Parr High Pressure Hydrogenator for 22 hours. The resultant mixture was immediately-f i l t e r e d i n t o a fresh l y prepared s o l u t i o n of 2,4-dinitrophenylhydrazine reagent (prepared by d i s s o l v i n g 2,4-dinitrophenylhydrazine (2g) i n methanol (50 ml) saturated with hydrogen c h l o r i d e ) . Immediately a yellow p r e c i p i t a t e formed and was f i l t e r e d o f f (130 mg; 10% y i e l d ) . This material, mp 233-235\u00C2\u00B0, was r e c r y s t a l l i z e d from dime thyIformamide to y i e l d pure levulinaldehyde bls-2,4-dinitrophenylhydrazone (123), mp 238-240\u00C2\u00B0 ( l i t . 1 1 7 mp ,233\u00C2\u00B0). b) Farnesol (61b) (26 mg; 0.117 mmol) was dissolved i n ethyl acetate (2 ml), cooled to -78\u00C2\u00B0 and subjected to a stream of ozone'enriched oxygen for 1 hour. The reaction mixture was transferred to a Parr Hydrogenator apparatus, methanol (2 ml) was added and the mixture was hydrogenated over 5% palladium on calcium carbonate, at 45 p s i , for 1.5 hours. The \" resultant s o l u t i o n was immediately f i l t e r e d into 2,4-dinitropheny1-hydrazine reagent (prepared from 2,4-dinitrophenylhydrazine (0.5 g) dissolved i n methanolic hydrogen chloride (3 ml)). The resultant yellow p r e c i p i t a t e (25 mg; 19% y i e l d ) , was r e c r y s t a l l i z e d from dimethylformamide to y i e l d pure 123, mp 238-240\u00C2\u00B0 ( l i t . 1 1 7 mp 233\u00C2\u00B0). Preparation of levulinaldehyde bis-2,4-dinitrophenylhydrazone ( 1 2 3 ) 1 1 9 1 2-Methylfuran (sylvin) (lg) was dissolved i n a solu t i o n of 2,4-dinitrophenylhydrazine (2g) i n aqueous hydrochloric acid (2M; 500 ml) and the mixture was s t i r r e d at room temperature f o r 3 days. The gummy orange p r e c i p i t a t e (1.2 g) was f i l t e r e d o f f , washed with water and hot ethanol and then r e c r y s t a l l i z e d from dimethylformamide to y i e l d l e v u l i n -aldehyde bis-2,i4-dinitrophenylhydrazone (123) as orange needles, mp 245-246\u00C2\u00B0 ( l i t . 1 1 7 mp 233\u00C2\u00B0). Anal. Calcd. for C, _H..,0oNo : C, 44.3; H, 3.5; N, 24.2. Found: C, 44.21; 1/ l o o o H, 3.41; N, 24.34. Mixed mp with 123 from the ozonolysis of farnesol (61b), 240-241\u00C2\u00B0 Levulinaldehyde bis-2,4-dinitrophenylhydrazone (123) from Umbelliprenin (83) Umbelliprenin (83) (24 mg; 0.066 mmol) was dissolved i n eth y l acetate (1 ml) and subjected to a stream of ozone enriched oxygen,.at -78\u00C2\u00B0, f o r I. 5 hours. The reaction mixture was allowed to warm to room temperature, then was transferred to a Parr Hydrogenator f l a s k . The ozonization vessel was rinsed with methanol (1 ml) and t h i s was added to the s o l u t i o n for hydrogenation. 5% Palladium on calcium carbonate (50 mg) was added and the mixture was hydrogenated at 45 p s i for 1.5 hours. The reaction mixture was f i l t e r e d immediately into 2,4-dinitropheny-hydrazine reagent (2,4-dinitropheny-hydrazine (300 mg) i n f r e s h l y prepared methanolic hydrogen chloride (1 ml)). The yellow p r e c i p i t a t e which formed immediately was c o l l e c t e d by f i l t r a t i o n (45.8 mg) and was r e c r y s t a l l i z e d from dimethylformamide and was washed with methanol to y i e l d levulinaldehyde bis-2,4-dinitrophenylhydrazone (123) (19/ mg; 26% y i e l d ) , mp 241-242.5\u00C2\u00B0, mixed mp with authentic (prepared) 123, 242-244\u00C2\u00B0, ( l i t . 1 1 7 mp 233\u00C2\u00B0). Anal. Calcd. f o r C. ..H., ,N o0 o : C, 44.3; H, 35; N, 24.2. Found: C, 44.48; 1/ l o o o H, 3.62 : N, 23.95. DISCUSSION PART III Biosynthetic Studies on Coumarins From Thamnosma montana As has been noted i n the introduction i n t h i s t h e s i s , several questions remain to be answered i n regard to the biosynthesis of furano-coumarins. Of p a r t i c u l a r i n t e r e s t i s the question as to what r o l e mevalonate (78) plays i n the biosynthesis of the \"extra\" furan atoms i n these compounds. Also i t i s of i n t e r e s t to know i f mevalonate (78) i s the precursor of the C_ side chains present i n so many coumarins. The reports of Caporale et^ a l 5 8 and of Brown 6 2 seemed to i n d i c a t e that the incorporation of mevalonate (78) into furanocoumarins was non-specific and Brown 6 2 has shown that acetate i s a better precursor of the furano-coumarins of Pastinaca s a t i v a than i s mevalonate (78). However, Floss and Mothes 5 2 have reported s p e c i f i c incorporation of C(4) of mevalonic acid (57) into p i m p i n e l l i n (65) i n Pimpinella magna. Thus one object of t h i s work was to more accurately define, through s p e c i f i c degradations as described i n the previous section, the r o l e of mevalonate (78) and acetate i n the biosynthesis of furanocoumarins of Thamnosma montana. However, before discussing the studies performed i n t h i s regard, i t i s pertinent to discuss some preliminary work i n which some other aspects of coumarin biosynthesis were explored. These preliminary studies were i n s t i t u t e d with several objectives i n mind; to become more f a m i l i a r with the plant system under conditions of feeding precursors to small quantities of plant, to attempt to gain information as to the biosynthetic i n t e r r e l a t i o n s h i p s between the coumarins of Thamnosma montana, to confirm that biosynthesis of these coumarins was occurring on a regular and measurable basis i n the plant and, f i n a l l y , i t was hoped that some information could be obtained with respect to the biogenesis of the novel dimeric coumarin thamnosin (8). In an attempt to gain information as to the biogenetic i n t e r r e l a -tionships between the coumarins of Thamnosma montana an experiment was performed i n which f i v e samples of shoots from a s i n g l e mature Thamnosma 14 montana plant were each allowed to incorporate D,L-phenylalanine-[3- C] (14) under i d e n t i c a l growing conditions. U t i l i z i n g the hydroponic method for purposes of incorporation, the shoots were allowed to grow for d i f f e r e n t times and each sample was worked up e s s e n t i a l l y as described previously. Thus the chloroform soluble portion of the acetone extracts was chromatographed on alumina columns and the resultant f r a c t i o n s were analyzed for r a d i o a c t i v i t y . The column f r a c t i o n s were chromatographed on prepared s i l i c a g el s t r i p s which were then analyzed with a Nuclear-Chicago t i c s t r i p counter. The l a t t e r instrument, equipped with an integrator and print-out recorder allowed quantitative determination of the r a d i o a c t i v i t y associated with the spots on the t i c s t r i p s . Suitable solvent systems were developed to allow separation and determination of the r a d i o a c t i v i t y associated with isoimperatorin (87), a l l o i m p e r a t o r i n methyl ether (7), thamnosmin (90), i s o p i m p i n e l l i n (2) and alloim p e r a t o r i n methyl ether epoxide (97). The r e s u l t s are given i n Table 2 and represented g r a p h i c a l l y i n Figure 64. It should be noted before discussing these r e s u l t s that as the plant samples were a l l from the same plant, there can be no argument as to d i f f e r i n g age or condition of the plant samples. Furthermore, each experiment u t i l i z e d several shoots to allow compensation for any di f f e r e n c e i n the v i a b i l i t y of the samples Table 2. Incorporation of D,L-Phenylalanine-[3- C] into coumarins of Thamnosma montana shoots Experi-ment No. Feeding Time (hrs) Activity Fed* (dpm) Weight of Plant .<8> % Incorporation isoimpera-torin (87) alloimpera-torin methyl ether (7). thamnosin (90) isopimpin-e l l i n (2) alloimpera-torin methyl ether epoxide (17) 1 24 6.75xl06 1.34 0.039 0.188 0.056 0.754 0.052 2 48 6.98xl06 1.45 -+ - t 0.087 0.546 0.656 3 72 6.79xl06 1.45 0.034 0.048 0.027 0.320 0.215 4 120 7.10xl06 1.40 0.011 0.043 0.009 0.038 0.045 5 168 7.04x106 1.48 0.056 0.068 0.024 0.115 0.088 * the activity fed has been corrected for radioactivity recovered outside the plant t the samples of isoimperatorin and alloimperatorin methyl ether from this experiment could not be satisfactorily separated to allow determination of the incorporation into each coumarin. The combined value i s 0.238%. hours Figure 64. Incorporation of D,L-phenylalanine-[3- C] into coumarins of Thamnosma montana Shoots Versus Time d u r i n g t h e e x p e r i m e n t s . I t i s a p p a r e n t from t h e s e r e s u l t s t h a t t h e coumarins s t u d i e d i n t h e s e e x p e r i m e n t s i n c o r p o r a t e p h e n y l a l a n i n e (14) r a p i d l y w i t h maximum i n c o r p o r a t i o n a f t e r a b out 48 h o u r s ( e x c e p t f o r i s o p i m p i n e l l i n (2) w h i c h r e a c h e s an a p p a r e n t maximum a f t e r 24 h o u r s ) . The i n c o r p o r a t i o n l e v e l s t h e n d e c r e a s e r a p i d l y . The s i g n i f i c a n c e o f t h e i n c r e a s e i n i n c o r p o r a t i o n between 5 and 7 days i s q u e s t i o n a b l e . I t i s t h e r e f o r e r a t h e r e v i d e n t t h a t t h e co u m a r i n s s t u d i e d i n c o r p o r a t e p h e n y l a l a n i n e (14) a t a p p r o x i m -a t e l y t h e same r a t e w i t h t h e o n l y a p p a r e n t d i f f e r e n c e b e i n g t h a t i s o p i m p i n e l l i n (2) r e a c h e s a maximum i n c o r p o r a t i o n i n a s h o r t e r p e r i o d o f t i m e t h a n do t h e o t h e r s . A n o t h e r s t r i k i n g p o i n t i s t h a t t h e degree o f i n c o r p o r a t i o n o f r a d i o a c t i v i t y i n t o t h e s e c o u m a r i n s i s e s s e n t i a l l y r e l a t i v e t h e t h e abundance of t h e c o u m a r i n s i n t h e s h o o t s . Thus i s o p i m p i n e l l i n (2) and a l l o i m p e r a t o r i n m e t h y l e t h e r e p o x i d e (97) b e i n g t h e most abundant c o u m a r i n c o n s t i t u e n t s , i n c o r p o r a t e t o t h e g r e a t e s t d e g r e e . I n o t h e r w ords, t h e s p e c i f i c a c t i v i t y (dpm/mmole) of t h e coumarins would be e s s e n t i a l l y t h e same. These' d a t a t e n d t o s u g g e s t t h a t no one c o u m a r i n i s b e i n g b i o s y n t h e s i z e d a t t h e expense o f a n o t h e r . I t had been hoped t h a t some i n t e r r e l a t i o n s h i p s would be e v i d e n t i n t h e s e r e s u l t s . Thus t h e o b v i o u s s i m i l a r i t y between a l l o i m p e r a t o r i n m e t h y l e t h e r (7) and t h e e p o x i d e (97) r a i s e s s p e c u l a t i o n t h a t one may be t h e p r o g e n i t o r o f t h e o t h e r . However i f t h i s was t h e c a s e one would e x p e c t t h a t t h e i n c o r p o r a t i o n o f p h e n y l a l a n i n e (14) would p r o c e e d f i r s t t o p r o v i d e r a d i o a c t i v e 7 and t h e l a t t e r would t h e n t r a n s f e r i t s a c t i v i t y t o 97. Thus t h e s e compounds would be e x p e c t e d t o r e a c h maximum i n c o r p o r a -t i o n v a l u e s a t d i f f e r e n t t i m e s . A l t h o u g h no such r e l a t i o n s h i p i s o b v i o u s from t h e above d a t a , a d e f i n i t e l a g ~ i n i n c o r p o r a t i o n i n t o 97 i s observed a f t e r one day., while alloimperatorin methyl ether (7) incor-porates r a p i d l y i n t h i s time period. Thus such a r e l a t i o n s h i p .could, well exist but not be obvious oh the time seal-- of t h i s experiment. These data do serve to show that the more abundant coumarins i n Thamnosma montana shoots are being biosynthesized i n the shoots and that the turnover rate i s rather r a p i d . It was next of i n t e r e s t to obtain incorporation of labeled precursors into the dimeric coumarin thamnosin (8). As i t i s known that t h i s compound i s present only i n the roots i t was necessary to administer the precursor to whole plants with a substantial root structure. In experiment 6 (Table 3) cinnamic acid-[2- 1^Cj (31) was administered to a whole mature plant, by the wick method (into a major shoot) and a f t e r 72 hours the roots (including the root crown) and shoots were worked up separately. During the incorporation time i t was observed by means of a Geiger counter that the r a d i o a c t i v i t y entering from the wick at the base of a major stem of the plant, was being transported upwards into the leaves and higher shoots of the plant. Table 3. Incorporation of Cinnamic acid-[2 C] into Thamnosma montana (Experiment 6) A c t i v i t y Weight Plant (dpm) ( g ) % Incorporation \u00E2\u0080\u009E .. , alloimpera- . . alloimpera-. N-methyl- ^ . xsopim- . thamnosin . , J t o r i n . r . t o r m acridone p i n e l l i n , , \u00E2\u0080\u00A2 (8) //s methyl ether r . . methyl ether K } (7) u ; epoxide (97) 138.12 (roots) 6.94xl0 7 15.20 (shoots) i n a c t i v e * i n a c t i v e * ; <0.0003V 0.033+ 0.132* 0.458+ * r a d i o a c t i v i t y determined by s c i n t i l l a t i o n counting technique t r a d i o a c t i v i t y determined by t i c s t r i p counting technique It was thus evident that the precursor which had been administered to a major stem of the plant had been transported only upwards into the shoots and not into the roots. It was thus apparent that i f biosynthesis of coumarins was being carried out in the roots, the precursor was not getting to the site of synthesis. The lack of detectable activity in alloimperatorin methyl ether (7) from the roots, tends to suggest that the transport of coumarins from the shoots to the roots is probably a slow process. It was evidently necessary to be able to administer the precursor directly to the root structure. To this end, i t was felt that i f thamnosin (8) was present in the roots of small young plants then the precursors could easily be administered to the smaller flexible roots by the hydroponic method. Thus such plants, which had been grown from seeds, were subjected to extraction and the extract was chromatographed in the mormal manner. Examination of the column fractions by tic revealed that no detectable amounts of thamnosin (8) or any of the alkaloids found in mature roots were present. In brief, the constitution of the small plants approximated closely that of the shoots of the older established plants. 14 The administration of cinnamic acid-[2- C] to the roots of Thamnosma montana was achieved by cutting a small sub-root from a large growing plant. The root was carefully removed so as not to damage the small rootlets and after coating the main portion of the root with wax the 14 rootlets were submerged in a solution containing cinnamic acid-[2- C] and the root was allowed to take up the solution for a period of eight days. The root material was worked up in the normal manner, except that before extraction, the ground root was diluted with a quantity of inactive ground root material to increase the yield of extract and subsequently increase the quantity of coumarins that could be isolated. The results of this experiment are given in Table 4. It i s evident from these results that the incorporation of cinnamic 14 acid-[2- C] into thamnosin (8) was extremely small whereas significant incorporation was observed with other, simpler coumarins. Thus, i t was apparent that meaningful data could not be obtained for thamnosin (8) by this method. The absence of significant amounts of thamnosin (8) i n the young plants suggests that the biosynthesis of this compound may be very slow. One can speculate that thamnosin (8) is a by-product of coumarin biosynthesis which i s stored in the roots and accumulates in appreciable quantities only in the older plants. It i s quite unfortunate that significant incorporation could not be obtained as a previously developed degradation procedure for determining the distribution of radioactivity i n each half of the thamnosin (8) molecule i s available from the structural studies of Kutney et a l 7 ' 8 . If the molecule was produced in a dimerization reaction as previously described, one would expect an equal distribution of radioactivity in the two coumarin portions of the molecule. In the next series of experiments, attention was focused on the role of mevalonate (78) in the biosynthesis of the monomeric coumarins in Thamnosma montana. It was hoped that by incorporating mevalonate (78) into isopimpinellin (2) and alloimperatorin methyl ether (7) i t would be possible to confirm or refute some of the conflicting evidence which has been published previously. In a series of preliminary experiments, 14 mevalonic acid-[2- C] was administered to young Thamnosma montana plants which had been grown from seeds. The precursor was fed by the Table 4. Incorporation of Cinnamic acid-[2- C] into Thamnosma montana root ... .. Weight Activity \u00C2\u00B0 (dpm) ( g ) % Incorporation N-methyl- c . . . alloimperatorin , . , Y - r f a g a r m e thamnosin thamnosmin ,^ -; , bergapten xanthotoxin acridone ( 5 b ) ( g ) ( 9 Q ) methyl^ether ^ 7 5.91 2.50x10 diluted to 30.91 <0.0008 inactive <0.0004 <0.002t 0.008 0.015 0.009 * activity corrected for radioactivity isolated outside the plant t insufficient material to obtain constant activity hydroponic method to the roots of these plants and a f t e r the desired feeding time the plants were worked up and the components were i s o l a t e d i n the usual manner. The pure coumarins thus i s o l a t e d were d i l u t e d with the i n a c t i v e compounds, c r y s t a l l i z e d to constant a c t i v i t y and the radio-a c t i v i t y determined by the S c i n t i l l a t i o n counting method. The r e s u l t s are depicted i n Table 5. It i s immediately evident from these r e s u l t s that, while mevalonic 14 acid-[2- C] i s being u t i l i z e d with considerable e f f i c i e n c y i n the biosynthesis of umbelliprenin (83), such i s not the case for the other coumarins. Except for umbelliprenin (83), where constant a c t i v i t y was achieved a f t e r f i v e c r y s t a l l i z a t i o n s , the amount of r a d i o a c t i v i t y present i n the coumarins was not s u f f i c i e n t to allow p u r i f i c a t i o n to constant a c t i v i t y ( i . e . d i l u t i o n of the coumarins i s o l a t e d with s u f f i c i e n t i n a c t i v e material to allow the necessary further c r y s t a l l i z a t i o n , would have also d i l u t e d the observed r a d i o a c t i v i t y to l e v e l s below which accurate counting was impossible). Thus the incorporation of mevalonic 14 acid-[2- C] into these coumarins i s at best very low, a r e s u l t also obtained by other w o r k e r s 5 3 ' 6 2 . However, the presence of s i g n i f i c a n t incorporations into umbelliprenin (83) indicates that the precursor, i s being u t i l i z e d by the plant i n the biosynthesis of t h i s coumarin. I t was apparent that i f meaningful values were to be obtained from the incorporation of C(2) labeled mevalonic acid into coumarins other than umbelliprenin (83), then a greater amount of r a d i o a c t i v i t y would have to be fed. Examination of the r e s u l t s i n Table 5 leads only to the conclusion that i n experiment 8, a feeding time of 2 days would appear to be inadequate to a t t a i n maximum incorporation. Incorporation into umbelliprenin (83) appears to be at a maximum af t e r 10 days but the Table 5. Incorporation of Mevalonic acid-[2- C] into monomeric coumarins in Thamnosma montana Experi-ment No. Feeding Time (days) Activity Fed (dpm) Weight of Plant (g) % Incorporation** umbelliprenin (83) alloimpera-torin methyl ether (7) isopimpin-el l in (2) alloimpera-torin methyl ether epoxide (17) thamnosmin (90) 8 2 8.74xl07* 15 0.019 <0.0003 \u00E2\u0080\u0094 <0.00004 - \u00E2\u0080\u0094 9 4 4.43xl07* 35 0.072 <0.0004 <0.000'4 \u00E2\u0080\u0094 <0.0015 10 7 9.OlxlO7* 16 0.045 <0.0006 <0.0013 <0.0006- <0.0023 11 10 2.47xl08* 10 0.160 <0.0009 hydroxypyruvate Figure 67. Proposed Incorporation of \"Active Formate\" into C(4) of Isopimpinellin (2) Although these proposals would explain the observed d i s t r i b u t i o n of a c t i v i t y i n i s o p i m p i n e l l i n (2), they must be considered as conjecture at t h i s moment as acetate i s a key precursor of many plant metabolites, any number of which could enter into coumarin biosynthesis. The r e s u l t s do however o f f e r some explanation of the more e f f i c i e n t incorporation of acetate r e l a t i v e to mevalonate (78) into i s o p i m p i n e l l i n (2). It i s notable that i f one considers only the incorporation of acetate into the furan portion of i s o p i m p i n e l l i n (2), the l e v e l of incorporation i s a c t u a l l y lower i n experiments'18 and .19 than.was observed for mevalonate (78) i n experiments 16 and 17. As noted previously, s i g n i f i c a n t incorporation was also achieved into alloimperatorin methyl ether (7) i n experiments 18 and 20. In these experiments 7 was converted to the d i o l (3) i n the usual manner and then p u r i f i e d to constant r a d i o a c t i v i t y . Unfortunately .the .radioactivity present i n the d i o l (3) a f t e r constant a c t i v i t y was achieved was not s u f f i c i e n t (in experiment 18) to perform a complete degradation. As i t was expected that the furanocoumarin portion of the molecule would be biosynthesized i n a manner s i m i l a r to i s o p i m p i n e l l i n (2), a t t e n t i o n was directed p r i m a r i l y to the d i m e t h y l a l l y l side chain. The r e s u l t s of these degradations are presented i n Table 10 .(and Figure 63). These experiments show that acetate i s indeed.a precursor of the d i m e t h y l a l l y l side chain of 7. The incorporation into the terminal three carbon atoms of the side chain i s considerable (48-62%) and as the side chain can be thought of asrbeing formed from three acetate units ( i f i t were mevalonate (78) derived) one could reasonably conclude that ^25% of the o r i g i n a l a c t i v i t y of 7 should reside i n the remaining two carbon atoms of the side chain. With >v 6% of the r a d i o a c t i v i t y i n the methoxyl Table 10. Degradation of Active Alloimperatorin methyl ether (7) (Experiments 18 and 20) Compound Specific Activity (% of alloimperatorin methyl ether (7)) Experiment 18 Experiment 20 Trial a Trial b alloimperatorin methyl ether diol (3) 5-(21-hydroxyethyl)-8-methoxypsoralen (113) acetone jj-bromobenzene-sulfonylhydrazone (111b) tetramethylammonium iodide (109) Method 1 Method 2 5- (21 -acetoxy-3' -rhydroxy-3' -methylbutyl)-8-methoxy-psoralen (114) 5-(2'-acet oxy-3'-hydroxy-3'-methylbutyl)-6-formyl-7-hydroxy-8-methoxy coumarin (115) 5-(2'-acetoxy-31-hydroxy-3'-methylbutyl)-6-acetoxy-7,8-dimethoxycoumarin(117b) 1-(2'-acetoxy-3'-hydroxy-3'-methylbutyl)-2,6-diformyl-3,5-dihydroxy-4-methoxybenzene (118) 7,02x10 dpm/ mmol (100) 3.86x10 (55) 3.38x10 (48) 4.95x10 (7)* 2.29x10 dpm/ mmol (100) 1.022x10 (43) 1.49x10 (62.7) 1.07x10 (4.5)* 1.605x10 dpm/ mmol (100) 1.44x10 (89.6) 1.39xl04 (87.0) * insufficient 109 obtained to convert to picrate group, this suggests that only 10-20% of the original activity of 7 resides in the furanocoumarin portion of the molecule, presumably in the 4- and 7- positions. The incorporation of acetate into 7 would appear Figure 63. Degradations of Alloimperatorin methyl ether (7) to be considerably greater than has been observed for mevalonate (78). However the s i g n i f i c a n c e of t h i s increase i s unclear. That the side chain i s acetate derived would appear to be i n keeping with mevalonate serving as the v e h i c l e for incorporation but such compounds as senecioyl coenzyme-A (64) cannot be ruled out as possible precursors. It i s hoped that future experiments w i l l provide more d e f i n i t i v e information i n t h i s regard. 0 64 In summary, i t has been shown.that mevalonic acid (57) can indeed serve as a s p e c i f i c precursor of the furan carbon atoms i n furanocoumarins. Unlike the reports of Brown 6 2 and Caporale et a l 5 8 , C(2) labeled mevalonate (78) was not observed to be incorporated into i s o p i m p i n e l l i n (2) while the C(4) labeled compound was incorporated i n a low but s p e c i f i c manner. In addition the r e s u l t s from the feeding of sodium 14 acetate-[2- C] i n d i c a t e that although the t o t a l incorporation of t h i s precursor into i s o p i m p i n e l l i n (2) i s greater than was observed with 3 mevalonate-[4- H] (78), the incorporation into the furan portion of the molecule i s at a comparable l e v e l . Furthermore the incorporation i s i n a manner compatible with the sequence; acetate mevalonate (78) -* i s o -p i m p i n e l l i n (2). The r e s u l t s r e i n f o r c e the statement, made e a r l i e r , that for incorporation studies to y i e l d t r u l y meaningful information, degradations must be performed to determine the d i s t r i b u t i o n of the l a b e l s i n the molecules under study. Thus, although one might have reasonably assumed that i f acetate were to incorporate into simple furanocoumarins, i t would incorporate e x c l u s i v e l y into the furan portion of the molecule, this has been shown to be an invalid assumption. The mechanism by which acetate incorporates into C(4) and the methoxyl groups of isopimpinellin (2) is unclear, but such incorporation has definitely been observed. With respect to alloimperatorin methyl ether (7) i t has been shown that C(2) of acetate can serve as a source of the methoxyl group. The dimethylallyl side chain has been shown to be acetate derived and although incorporation levels are extremely low, i t seems likely to be mevalonate derived as well. As more workers report the results of attempting to incorporate mevalonic acid (57) into alkylated phenolic compounds i t is becoming evident that disappointingly poor incorporation-levels are to be expected and that such poor incorporations do not necessarily reflect the non-utilization of the precursor. In our case, work is now in progress with tissue culture samples of Thamnosma montana in the hope that this medium will prove more conducive to efficient utilization of mevalonate (78). EXPERIMENTAL (PART III) For general experimental information see page 93. Ra d i o a c t i v i t y was measured with a Nuclear Chicago Mark .1 Model 6860 Liqui d S c i n t i l l a t i o n counter i n counts per minute (cpm). The r a d i o a c t i v i t y of a sample i n d i s i n t e g r a t i o n s per minute (dpm) was , subsequently calculated using the counting e f f i c i e n c y which was determined (unless otherwise noted) f o r each sample by the external standard t e c h n i q u e 1 2 5 u t i l i s i n g the b u i l t - i n barium-133 gamma source. The organic l i q u i d s c i n t i l l a t i o n s o l u t i o n used with the counter was made up of the following components: toluene (1 1), 2,5-diphenyloxazole (4 g) and 1,4-bis[2-(5-phenyloxazoly)Jbenzene (0.05 g). In p r a c t i c e , a sample was dissolved i n benzene (1 ml) or i n methanol (1 ml) i f the compound' was not s u f f i c i e n t l y soluble i n benzene, i n a counting v i a l . The volume was then made up to 15 ml with the above s c i n t i l l a t o r s o l u t i o n . When counting samples which were water soluble an aqueous s c i n t i l l a t o r s o l u t i o n was u t i l i z e d , made up of the following components: toluene (385 ml), dioxane (385 ml), methanol (230 ml), napthalene (80 g), 2,5-diphenyloxazole (5 g) and 1,4-bis[2-(5-phenyloxazoly)Jbenzene (0.0625 g). In pr a c t i c e a sample was dissolved i n water (as required) and methanol (1 ml) i n the counting v i a l . The so l u t i o n was made up to 15 ml with the aqueous s c i n t i l l a t o r s o l u t i o n . For each sample counted, the background was determined for the counting v i a l to be used by f i l l i n g the v i a l with the appropriate solvent and s c i n t i l l a t o r s o l u t i o n and counting (3 x 40 min. or 3 x 100 min.). The di f f e r e n c e i n the cpm between the background count and the sample count was used f o r subsequent c a l c u l a t i o n s . Unless otherwise noted r a d i o a c t i v i t y was determined by s c i n t i l l a t i o n counting with organic s c i n t i l l a t o r s o l u t i o n . Deviations from these normal counting procedures w i l l be discussed i n the s p e c i f i c instances i n which they a r i s e . The r a d i o a c t i v i t y of labeled compounds absorbed on t i c plates was measured using a Nuclear-Chicago Actigraph Scanner, Model 1036, connected to a Nuclear-Chicago Model 8416 pen recorder and a Nuclear-Chicago, Model 8437 d i g i t a l print-out recorder. The Thamnosma montana plants used i n t h i s study were c o l l e c t e d i n summer as seeds and mature plants from the north-facing slopes of small h i l l s i n the v i c i n i t y of Joshua Tree National Monument, i n the Mojave Desert area of southern C a l i f o r n i a (with the kind cooperation of Dr. D.L. Dreyer). Some seeds could be propagated by Dr. P. Salisbury of. our department. Of the mature plants c o l l e c t e d some could be su c c e s s f u l l y transplanted and continued to grow. Small mature Thamnosma montana plants (2-4 years old) were obtained from Molecular Biochemical Corporation, Tempe, Arizona. I s o l a t i o n of Constituents of Thamriosma montana The i s o l a t i o n s of components i n the following experiments were performed e s s e n t i a l l y as described previously (see Part I ) , except that the procedures were scaled according to the wet weight of the plant i n each experiment. Feeding Experiments 1 , 2 , 3 , 4 and 5 In these experiments, the cut shoots of mature Thamnosma montana 14 plants were fed D,L-phenylalanine-[3- C] (14) (obtained from New England Nuclear Corp., Boston, Mass.) The cut shoots were divided into q f i v e samples of approximately equal weight. The precursor (2.22x10 dpm, 37.4 mg) was dissolved in distil led water ( 50 ml) and an aliquot was removed, weighed and the radioactivity per unit weight of solution determined (4.97x10 dpm/g). Five aliquots were then weighed out from this solution (^ 1.5 g each) and each was placed in a separate test tube. The five plant samples were then introduced into these test tubes and allowed to absorb the precursor. The plants were kept under continuous fluorescent illumination.- As the tubes became dry the solution was replaced with Hoaglands nutrient mixture 1 2 6 and the plants were allowed to grow for the required time period. The plants were then worked up in the normal manner, except the plant samples were cut into small pieces and ground with sand in a mortar and pestle. As in the normal procedure the plant extract was chromatographed on alumina. The fractions obtained were placed on Eastman chromagram strips (3 x 20 cm) and chromatographed to effect separation of the individual components. In the case of fractions containing isoimperatorin (87), alloimperatorin methyl ether (7) and thamnosmin (90), the strips were eluted three times with hexane-anhydrous ether (10:1). Fractions containing isopimpinellin (2) and alloimperatorin methyl ether epoxide (97) were chromatographed eluting with ethyl acetate-chloroform (1:1) (solvent B). The tic strips were then analyzed for radioactivity utilizing the Nuclear-Chicago \"strip counter\" and the activity of each spot was calculated. In cases where compounds were present in more than one fraction the values were totalled and the incorporation thus calculated. The experimental details and results are presented in Table 11. Table 11. Incorporation of D,L-Phenylalanine-[3- C] into Thamnosma montana Shoots To t a l A c t i v i t y Isolated (dpm) [% Incorporation] Experi-ment No. Feeding Time (hrs) A c t i v i t y Fed* (dpm) Weight Fed (mg) Plant Weight (g) isoimpera t o r i n (87) alloimpera-t o r i n methyl ether (7) thamnosmin (90) isopimpin-e l l i n (2) alloimpera-t o r i n methyl ether epoxide (97) 1 24 6.75xl0 6 1.1 1.34 2.64xl03 [0.039] 1.27xl0 4 [0.188] 3.79xl0 3 [0.056] 5.08xl0 4 [0.754] 3.39xl0 4 [0.502] 2 48 6.98xl0 6 1.2 1.45 - t - t 6.80xl03 [0.087] 3.81xl0 4 [0.546] 4.58xl0 4 [0.656] 72 6.79xl0 6 1.1 1.45 2.29xl03 3.26xl0 3 1.80xl0 3 2.18xl0 4 1.46xl0 4 .J [0.034] [0.048] [0.027] [0.320] [0.215] 4 120 7.10xl0 6 1.2 1.40 7.62xl02 [0.011] 3.06xl0 3 [0.043] 6.36xl0 2 [0.009] 2.7xl0 3 [0.038] 3.22xl0 3 [0.045] 168 7.04xl0 6 1.2 1.48 3.96xl03 4.78xl0 3 1.68xl0 2 8.12xl0 3 6.17xl0 3 [0.056] [0.068 [0.024] [0.115] [0.088] * corrected for a c t i v i t y i s o l a t e d outside the plant t the samples of isoimperatorin and alloimperatorin methyl ether from t h i s experiment could not be s a t i s f a c t o r i l y separated \u00C2\u00A3 0 allow determination of the incorporation into each coumarin. The combined value i s 1.66x10 dpm or 0.238% Feeding Experiment 6 In t h i s experiment cinnamic a c i d - I 2 - C] (obtained from International Chemical and Nuclear Corp. of C i t y of Industry, C a l i f o r n i a ) was adminis-14 tered to a whole mature Thamnosma montana plant. Cinnamic acid-[2- C] g (5.5x10 dpm), obtained as the free a c i d , was dissolved i n water (25 ml) and an a l i q u o t was removed, weighed and the r a d i o a c t i v i t y per u n i t weight of s o l u t i o n determined (2.16x10^ dpm/g). A weighed a l i q u o t was then removed (3.2098 g, 6.94xl0 7 dpm) and reduced to dryness under a stream of nitrogen. The r e s u l t i n g s o l i d was dissolved i n d i s t i l l e d water (0.5 ml) containing sodium bicarbonate (0.5 mg) and t h i s s o l u t i o n was administered to the plant by the cotton wick method. This method entailed the threading of a cotton s t r i n g through the base of a major stem of the plant. The intertwined ends of the wick were placed i n a small v i a l and the v i a l was f i l l e d with the precursor s o l u t i o n . When the s o l u t i o n had been absorbed the o r i g i n a l container of the precursor so l u t i o n was washed with water (0.5 ml) and t h i s washing was allowed to be absorbed by the plant. This procedure was repeated throughout the time of the experiment (3 days). The plant was kept under continuous fluorescent i l l u m i n a t i o n during t h i s time. During the course of the experiment, r a d i o a c t i v i t y could be followed as i t was transported up into the main shoot by means of a Geiger counter. Af t e r 3 days there appeared to be some decrease i n a c t i v i t y i n the shoots measured i n t h i s manner. The plant was then divided into shoots and root (including root crown) sections and each portion was worked up i n the manner described previously. Thamnosin (8), N-methyl acridone (5) and alloimperatorin methyl ether (7) were i s o l a t e d from the roots. Alloimperatorin methyl ether (7), isopimp-i n e l l i n (2) and alloimperatorin methyl ether epoxide (97) were i s o l a t e d from the shoots. The experimental details are given in Tables 12a and 12b. Isolation of the Constituents of Young Thamnosma montana Plants Whole Thamnosma montana plants (wet weight 35 g) which had been grown from seeds (y 16 months old) were ground in a Waring blender to a coarse powder and extracted with acetone (500 ml) in a Soxhlet extractor for 4 hours. The acetone extract was reduced to dryness under reduced pressure to yield a residue (1.13 g) which was treated with hot chloroform (100 ml). The chloroform soluble portion was filtered and the solvent was removed under reduced pressure to yield a residue (0.8 g) which was preadsorbed on alumina (neutral, 3 g, Activity IV). The preadsorbed material was placed on top of a column of alumina (neutral, 27 g, Activity IV) which had been made up in petroleum ether and elution was begun immediately. Compounds isolated were crystallized from solvents as described previously. The fractions eluted with petroleum ether and petroleum ether-benzene contained waxy material (y 127 mg) and were not examined further. Also eluted with petroleum ether-benzene was a fraction (36 mg) noted as a purple (uv) band on the column. Preparative tic on this fraction allowed isolation of umbelliprenin (83) (15.5 mg), mp 62-63\u00C2\u00B0 . Also noted in this fraction was a yellow (uv) band correspond-ing to isoimperatorin (87) but this material was not isolated. Later fractions eluted with petroleum ether-benzene and benzene were noted (by tic) to contain alloimperatorin methyl ether (7), thamnosmin.(90) and phellopterin (91). Preparative tic on these combined fractions (48 mg) anhydrous ether-hexane (1:1) (solvent A) allowed isolation of alloimpera-torin methyl ether (7) (10.5 mg) mp 108-110\u00C2\u00B0, thamnosmin (90) (14.5 mg), mp 101-104\u00C2\u00B0 and phellopterin (96) (6.1 mg), mp 85-96\u00C2\u00B0 (one crystallization). A c t i v i t y Fed (dpm) S p e c i f i c Weight Fed (mg) Wet Weight of plant Weight of Compound Isolated (mg) A c t i v i t y Fed (dpm/mmol) thamnosin (8) N-methyl- . , . ^ ., J alloimperatorin acridone ., \u00E2\u0080\u00A2/ ^ methyl ether (7) is o p i m p i n e l l i n (2) alloim p e r a t o r i n methyl ether epoxide (97) 4.44xl0 1 0 138.12 (roots) 12.2 23.4 23.0 - -6.94xl0 7 0.23 15.20 (shoots) - 12.2 25.2 30.2 Table 12b. Sp e c i f i c A c t i v i t y Isolated (dpm/mmol) [% Incorporation] a ] _ l \u00E2\u0080\u00A2 t \u00E2\u0080\u00A2 th 1 alloimp e r a t o r i n thamnosin (8) N-methylacridone (4) ether (7) i s o p i m p i n e l l i n (2) methyl ether epoxide (97) roots i n a c t i v e i n a c t i v e <1.66x10 [<0.0003] shoots - - 5.38xl0 5 [0.033]* 8.90xl0 5 [0.132] 3.16xl0 6 [0.458]* * r a d i o a c t i v i t y determined by the \" s t r i p counting technique\" The fractions eluted with benzene were combined (35.3 mg) and chromato-graphed on preparative tic (solvent B) allowing isolation of isopimpin-e l l in (2) (28 mg), mp 150-151\u00C2\u00B0. The fractions eluted with benzene-chloroform were combined (46 mg) and chromatographed on preparative tic (solvent B) allowing isolation of alloimperatorin methyl ether epoxide (97) (32 mg), mp 101-103\u00C2\u00B0. Examination of a l l the fractions and mother liquors by tic did not reveal the presence of thamnosin (8), any of the alkaloids previously isolated from the roots or alloimperatorin methyl ether diol (3). Tic evidence did suggest the presence of bergapten (67), xanthotoxin (55) and possibly psoralen (68) but the quantity of these -materials was small. Feeding Experiment 7 For this experiment a branch root from a mature growing Thamnosma montana plant was carefully freed of earth so as not to damage the fine rootlets. The branch root was then removed from the plant and the main portion of the root (including most of the cut end) was coated with wax to aid in retention of moisture. The rootlets were then immersed in an 14 7 aqueous solution of cinnamic acid-[2- C] (3.75x10 dpm, from an approp-riate weighed aliquot of the precursor solution utilized in experiment 6, diluted to 20 ml). Due to the volume of solution (necessary to cover .\u00E2\u0080\u00A2\u00E2\u0080\u00A2 the rootlets) the root was allowed to absorb the solution over a period ! of eight days. After this time, some solution s t i l l remained in the \u00E2\u0080\u00A2 vessel and this solution was counted to determine what residual activity \-remained. This value (1.16xl07 dpm/mmol) was subtracted from the original activity to determine the corrected activity fed. The root and wax coating (5.91 g) was then ground to a powder in a Waring blender and to this material was added inactive ground root (25 g). This mixture was worked up e s s e n t i a l l y as described previously except that i n t h i s case a second chromatography column was not required. Due to the l e s s e r amount of root m a t e r i a l (than i n the d e s c r i p t i o n of i s o l a t i o n of components of Thamnosma montana root described i n part I of the thesis) a l l the components could be i s o l a t e d by preparative t i c a f t e r only one column chromatography. The experimental d e t a i l s are presented i n Tables 13a and 13b. Feeding Experiments 8, 9, 10, 11 and 12 14 In these experiments mevalonic acid-I2- C] (57) was administered to whole young Thamnosma montana plants (y 16 months old) by the hydroponic 14 method.' In experiments 8 to 11, mevalonic acid-[2- C] was administered 14 as the sodium s a l t . In these experiments mevalonic acid-[2- C] lactone 9 (78) 0--11x10 idpm) (obtained from New England Nuclear Corp. of Boston, Mass.) was treated with aqueous sodium carbonate (10.5 mg i n 1 ml of water). The s o l u t i o n was then transferred to a volumetric f l a s k and made up to 25 ml with d i s t i l l e d water. Aliquots were withdrawn as desired 14 8 for the experiments. In experiment 12, mevalonic acid-[2- Cl (5.5x10 dpmX was administered as the dibenzylethylenediamine s a l t as obtained from New England Nuclear Corp. The s a l t was dissolved i n d i s t i l l e d water, transferred to a 'volumetric f l a s k and made up to 25 ml with water. Aliquots were withdrawn as required. In each experiment the young plants were c a r e f u l l y uprooted so as to cause minimum damage to the f i n e r o o t l e t s and the roots were placed i n test tubes containing the precursor i n s o l u t i o n (y 1 ml per test tube). Several plants (3 to 6)were used i n each experiment and the precursor s o l u t i o n was d i s t r i b u t e d equally among the test tubes to be used. Af t e r the plants had absorbed the precursor s o l u t i o n , the o r i g i n a l container of the precursor s o l u t i o n was washed with d i s t i l l e d water (y 3 ml) and the Table 13a. Incorporation of Cinnamic acid-[2- C] into Thamnosma Montana Root A \u00C2\u00AB. \u00E2\u0080\u00A2 Specific T \u00E2\u0080\u00A2 . _ Weight Activity , . . _ Weight % _, \u00E2\u0080\u009E J Activity \u00C2\u00B0 of Jed: Fed , F e d Plant ( d P m ) (dpm/mmol) ( m g ) (g) Compound Isolated (mg) N-methyl- ,. . tham- tham- a H \u00C2\u00B0 i m P e r a . \u00E2\u0080\u009E Y-faearine . . torin bergapten xanthotoxin acridone , \u00C2\u00B0 N nosin nosmm .. / ^ T \ / C C . (5b) ^ ^ 9 0 ^ methyl^ether (67) (55) 7 10 5 ' 9 1 2.50x10 4.40x10 0.08 (diluted 8.4 32.1 16.5 5.0 20 13.3 13.6 to 30.91) * corrected for activity recovered outside the plant Table 13b. Specific Activity Isolated (dpm/mmol) {% Incorporation] N-methyl-acridone (4) y-fagarine (5b) thamnosin (8) thamnosmin (90) a \u00C2\u00B1 l \u00C2\u00B0 i r o p ? E a t o ; r i n bergapten (67) methyl ether(7) xanthotoxin (55) <4.76xl03 [<0.0008] inactive <3.4xl03 2.84xl04 2.90xl04 5.92xl04 [<0.0004] [0.002] [0.008] [0.015] 3.41xl04 [0.009] plants were allowed to absorb the washings. This procedure was continued throughout the course of the experiment. In each experiment the plants were worked up as previously described for the young whole plants. The experimental details are given in Tables 14a and 14b. As necessary, the compounds isolated were diluted with the corresponding inactive compound to allow the necessary crystallizations to be performed. Feeding Experiments 13, 14 and 15 3 In these experiments D,L-mevalonic acid-{-2- H] lactone (obtained from Amersham/Searle Corporation of Des Plaines, Illinois) was adminis-tered to young whole plants (24 months old) by the hydroponic technique (feeding directly into the roots). The precursor, obtained in benzene solution was reduced to dryness under a nitrogen stream and dissolved in disti l led water (10 ml). Aliquots were then withdrawn as required and used as such. In each case the plants were allowed to grow for 14 days under illumination of fluorescent lights for 18 hours a day. After the precursor had been absorbed, the plants were sustained with disti l led water. Workup of the plants was as previously described (for young whole plants) except that after the plants had been ground to a powder,it was diluted with inactive ground shoot material to increase the yield of coumarins. In each case only isopimpinellin (2), alloimperatorin methyl ether (7) and umbelliprenin (83) were isolated. Dilution of the coumarins isolated with inactive coumarins was performed as necessary to obtain quantities which could be crystallized to constant activity or to complete inactivity and where applicable, to yield sufficient coumarin of constant activity to perform degradations. Alloimperatorin methyl ether (7) in each case was converted to its diol (3) by the previously described Experi-ment No. Activity* Fed (dpm) Specific Activity Fed (dpm/mmol) Weight Fed 469 (1964). 41. A.J. Austin and M.B. Meyers, Phytochem., / 4 , 245 (1965). 42. A.J. Austin and M.B. Meyers, Phytochem., /;, 255 (1965). 43. B. F r i t i g , L. H i r t h and G. Ourisson,'Phytochem., _9, 1963 (1970). 44. R.D. Haworth, J . Chem. S o c , 1942, 448. 45. CA. Bunton, G.W. Kenner, ,M.J.T. Robinson and B.R. Webster, Tetrahedron, 19, 1001 (1963). 46. H. Grisebach and W.D. 0111s, Experientla, 17, 4 (1961). 47. A.I. Scott, P.A. Dodson, F. McCapra and M.B. Meyers, J . Am. Chem. Soc., 85, 3702 (1963). 48. S.A. Brown, Canad. J . Biochem.,43, 199 (1965). 49. H.G. F l o s s , H. Guenther and L.A. Hawiger, Phytochem., j$, 585 (1969). 50. S.P.J. Shah and L.J. Rogers, Biochem. J . , 114, 395 (1969). 51. W.D. O l l i s and 1.0. Sutherland i n Chemistry of Natural Phenolic Compounds, Ed. by W.D. O l l i s , Pergamon Press, 1961, Chapter 4 and references c i t e d therein. 52. H.G. Floss and U. Mothes, Phytochem., _5> 161 (1966). 53. M. Hamada and M. Chubachi, Agr. B i o l . Chem.,33, 793 (1969). 54. G. Kunesch and J . Polonsky, Phytochem. , 8^, 1221 (1969). 55. T.A. Geissman i n Biosynthesis of Natural Compounds, Ed. by P. Bernfeld, Pergamon Press, Oxford, 1963, Chapter 12. 56. R. Aneja, S.K. Mukerjee and T.R. Seshadri, Tetrahedron, /4, 256 (1958). 57. E.A. Abu-Mustafa and M.B.E. Fayez, J . Org. Chem.,26, 161 (1961). 58. G. Caporale, A. Breccio and G. Rodighiero, Prepn. Bio-Med. Appl. Labeled Mol., Proc. Symp., Venice, p. 103 (1964). 59. H.G. Floss and U. Mothes, Z. Naturforsch, 19b, 770 (1964). 60. H.G. Floss and H. Paikert, Phytochem., 8, 589 (1969). 61. W. Steck, M. El-Dakhakny and S.A. Brown, Tet. L e t t s . , 4805 (1969). 62. S.A. Brown, Phytochem., _9, 2471 (1970). 63. B.K. Ganguly and P. Bagchi, J . Org. Chem., 21, 1415 (1956). 64. The Sadtler Standard Spectra, U l t r a V i o l e t Spect r a , The Sadtler Research Laboratories, P h i l a d e l p h i a , 1968, uv // 492. 65. Reference 64, uv # 6723. 66. Reference 64, uv # 6712. 67. Reference 64, uv # 3221. 68. Reference 64, uv # 3222. 69. K o j i Nakanishi, Infrared Absorption Spectroscopy, Holden Day Inc., San Francisco, 1962, p. 52. 70. The Sadtler Standard Spectra, Standard Infrared, The Sadtler Research Laboratories, Phi l a d e l p h i a , 1962, i r // 20151. 71. L.M. Jackman and S. S t e r n h e l l , Applications of Nuclear Magnetic Resonance Spectroscopy i n Organic Chemistry, 2nd E d i t i o n , Pergamon Press, Oxford, 1969, p. 278. 72. S.S. Dharmatu and G. G o v i l , Proc. Indian Acad. S c i . , 56., 71 (1962). 73. Reference 71, p. 306. 74. T.J. Batterham and J.A. Lamberton, A u s t r a l . J . Chem., 17, 1305 (1964). 75. E.A. Abu-Mustafa and M.B.E. Fayez, Canad. J . Chem.,45, 325 (1967). 76. NMR Spectra Catalog, Vol. I, compiled by N.S. Bhacca, L.F. Johnson \u00E2\u0080\u00A2 and J.N. Shoolery, Varian Associates, Palo A l t o , 1962, V. 225. 77. E.V. Lassak and J.T. Pinhey, J . Chem. S o c , 2000 .(1967), 78. R. Grigg, J.A. Knight and P. Roffey, Tetrahedron, 22, 3301 (1966). 79. C.S. Barnes and F.L. Occolowitz, A u s t r a l . J . Chem., 17, 975 (1969). 80. A. P e l t e r , P. Stainton, A.P. Johnson and M. Barber, J . Hetero. Chem., 2, 256 (1965). 81. R.A.W. Johnstone, B.J. M i l l a r d , F.M. Dean and A.W. H i l l , J . Chem. Soc. , '.1712 (1966). 82. J.P. Kutney, G. Eignedorf, T. Inaba and D.L. Dreyer, Org. Mass S p e c , .5, 249 (1971), and unpublished r e s u l t s . 83. N.S. Wulfson, V.I. Z a r e t s k i i and V.G. Zyakoon, Isv. Akad. Naiik SSR, Ser. Khim., 2215 (1963). 84. H. Dudzikiewicz, C. Dje r a s s i and D.H. Williams, Structure E l u c i d a t i o n of Natural Products by Mass Spectrometry, Vol. I I , Holden-Day, Inc., San Francisco, 1964, p. 255. 85. Reference 64, uv // 15575. 86. Reference 69, p. 24. 87. Reference 69, p. 27. 88. E. Spath and F. Vierhappen, Ber.,71, 1667 (1938). 89. R.B. Bates, J.H. Shauble and M. Soucek,\" Tet. Letts.,,1683 (1963). 90. R.B. Bates, D.M. Gale and B.J.' Gruner, 'JT'Org. Chem., _28, 1086 (19 6 3 ) . 91. L.H. Briggs and L.D. Colebrook, J . Chem. S o c , .2458 (1960). 92. Reference 18, p. 201. 93. F.E. King?-J.R. Housley and T.J. King, J. Chem. Soc. , 1392 (1954). 94. Reference 71, p. 316. 95. Reference 71, p. 330. 96. JrM. Lehn and J r J . R i e h l , Molec. Phys., 8_, 33 (1964). 97. P. Brown, J . Kossanyi and C. Dj e r a s s i , Tetrahedron, J22_, Suppl. 8, 241 ( 1 9 6 6 ) . 98. P.W. Chow, A.W. D u f f i e l d and P.R. J e f f e r i e s , A u s t r a l . J . Chem., 19, 483 ( 1 9 6 6 ) . 99. W. Steck, B.K. Bailey, J.P. Shyluk and O.L. -Gamborg, Phytochem., JL0, 191 (197 1 ) . 100. T. Noguchi and M. Kawanami, Ber. dtsch. Chem. Ges. , _71_, 344, 1428 ' .. (1938). 101. H.N. Khastgir, P.C. Duttagapta and P.Sengupta, Ind. J . Appl. Chem., 22, 82 (1959). 102. Reference 18, p. 199. 103. Reference 18, p. 200. 104. Reference 18, p. 202. 105. M.E. Brokke and B.E. Christensen, J . Org. Chem., 23, 589 (1 9 5 8 ) . 106. K. Bowden, J.M. Heibron, E.R.H. Jones and B.C.L. Weedon, J . Chem. S o c ; \u00E2\u0080\u00A2 '39 (1946). 107. M.P. Hegarty and F.N. Lahey, A u s t r a l . J.' Chem., 9_, 120 (1 9 5 6 ) . ~ 108. A. S c h 5 n b e r g , N. Badran and N.A. Starkowskyi, J . Am. Chem. Soc., 77, 5390 (1955). 109. H.D. Dakin, Am. Chem. J . , 42, 477 (1909). 110. Reference 69, p. 30. 111. R.N. Gupta and I.D. Spenser, Canad. J . Chem.,43, 133 (1965). 112. E.E. van Tamelen, R.S. Dewey, M.F, Lease and W.H. P i r k l e , J . Am. Chem. Soc., 83, 4302 (1961). 113. R.J.W. Cremlyn, J . Chem. S o c , 1229 (1966). 114. A.D. Cross, Introduction to P r a c t i c a l Infra-red Spectroscopy, Butterworths, London, 1964, p. 80. 115. T. Kariyone and T. Matsuno, Pharm. B u l l . Japan., JL, 119 (1953). 116. D.C. A l l p o r t and J.D. Bu'lock, J . Chem. S o c , ^54 (1960.). 117. A.G. Caldwell and E.R.H. Jones, J . Chem. S o c , 540 (1945). 118. R.L. Henne and W.L. P e r i l s t e i n , J . Chem. S o c , ;,218'3 (1943). 119. C L . Wilson, J . Am. Chem. S o c , 70;, 1313 (1948). 120a. D.D. P e r r i n , W.L.F. Armarego and D.R. P e r r i n , P u r i f i c a t i o n of Laboratory Chemicals, Pergamon Press, Oxford, 1966, p. 110. 120b. Reference 120a p. 200. 121. E. Campaigne and W.M. LeSuer, Org. Syn. C o l l . V o l . , 919 (1963). 122. Handbook of Chemistry and Physics, 49th E d i t i o n , Ed. by R.C Weast, The Chemical Rubber Co., Cleveland, 1968. 123. J . Th i e l e , Ann., 271, 127 (1892). 124. E.J. Corey, W.L. Mock and D.J. Pasto, Tet. L e t t s . , -347 (1961). 125. Mark 1 Liqu i d S c i n t i l l a t i o n Systems Inst r u c t i o n Manual, Section 1, Nuclear-Chicago Corporation, Des Plaines, I l l i n o i s . (1966). 126. D.R. Hoagland, J . Agr. Res., 18, 73 (1919). 127. P. Karlson, Introduction to Modern Biochemistry, 2nd Ed., Academic Press, New York, 1965, p. 211, and references c i t e d therein. 128. Reference 127, Chap. VIII. "@en . "Thesis/Dissertation"@en . "10.14288/1.0060044"@en . "eng"@en . "Chemistry"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Studies on the constituents of Thamnosma montana"@en . "Text"@en . "http://hdl.handle.net/2429/34209"@en .