(3. “26%? ‘..—.- 3‘. ' a: -. , . “J v qua . .- v >1, ... , r.» mo M I . .4 ~ i ‘ y.“ 21"“ b . ‘fihd! “‘5 ‘ 3:" . i‘ér“fi:~g§:1&3zi ‘“ “ ““x‘ofiéifi 533;"- ‘l ' IN a, {1“ a ' {1: ‘ i. Q“ ' 3 ' L321- zi a} . m ’- . ‘ ~, ‘Am'i'h‘a:. . :3 z ‘9‘ f g9 “‘55: i w * 6‘3”; a“ w 3" THESis l/W W/l ° all/l Ill/ll llllll’lllllll 3 1293 01405 1969 This is to certify that the thesis entitled Biologically Active Extracts From Magnolia spp. and Compounds From Magnolia salicifolia presented by Mark Allen Kelm has been accepted towards fulfillment of the requirements for 14.5. degree in Horticulture ,/ /Z I & Major professor Date Mrs; 8, 1996 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIERARY Michigan State University PLACE ll RETURN BOX to remove this checkout from your record. TO AVOID FINES Mum on or More dd. duo. DATE DUE DATE DUE DATE DUE MSU In An Affiflnutlvo ActlonlEquDl Opportunity Institulon W ””1 —7#fi _ _ _ BIOLOGICALLY ACTIVE EXTRACTS FROM IMAGNOLIA SPP. AND COMPOUNDS FROM IMAGNOLIA SALICIFOLIA By Mark Allen Kelm A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Horticulture 1996 ABSTRACT BIOLOGICALLY ACTIVE EXTRACTS FROM M4GNOLL4 SPP. AND COMPOUNDS FROM lilAGNOLIA SALICIFOLIA By Mark Allen Kelm Antimicrobial, nematicidal, insecticidal, and anticancer bioassays were done on extracts from Magnolia salicifolia Maxim. , Magnolia denudata Desr. , Magnolia kobus var. stellata Black, and Magnolia kobus var. loebneria Spong. Bioassays included mosquito larvae (Aedes egiptii), gypsy moth larvae (Lymantria dispar), and nematodes (Panagrellus redivivus Goody and Caenorhabditis elegans Dought). Anticancer bioassays were performed using mutated yeast, Saccharomyces cerevisiae . Most extracts possessed mosquitocidal activity. Extracts from M. steIIata and M. denudata exhibited growth inhibition on L. dispar whereas extracts of M salicifolia and M stellata had anticancer activity. Bioassay-directed work led to the isolation of six and the identification of one mosquitocidal compounds fromM salicifolia. Geranial plus neral, trans-anethole, methyl eugenol, iso-methyl eugenol, myn'sticin, and costunolide, gave 100% mortality on 4th instar larvae at concentrations ranging from 15 to 100 ppm. Parthenolide, identified by I-IPLC, was not mosquitocidal. Compounds were bioassayed for anticancer and growth inhibitory activity. Costunolide, parthenolide, and myristicin were found to have anti- cancer activity. 70c PM 7m PM iv ACKNOWLEDGMENTS I’d like to thank my major advisor, Muraleedharan Nair, for his advice and encouragement throughout this entire project and my guidance committee members, Drs. John Kelly, Robert Schutzki, and Daniel Herms for the assistance they were able to ofi‘er. A special thanks goes to Dr. Amitabh Chandra for his help in the protocols of the Bioactive Natural Products Laboratory and assistance in spectral interpretation. Dr. James Nitao deserves thanks for his expertise with the insect bioassays. Without these two individuals, the duration of this thesis would have been greatly extended. I’m also very fortunate to have met so many good people in this department especially the members of my lab, Yu-Chen Chang, Jennifer Miles, Di Zhang, Geofi“ Roth, Haibo Wang, and Andy Erickson. I’m also gratefiil for the understanding and help my friend, Sarah Breitkreutz so kindly offered. I would finally like to thank my parents for believing in me and always being there no matter how tough things got. TABLE OF CONTENTS LIST OF TABLES ........................................................................................................ vii LIST OF FIGURES ........................................................................................................ x LIST OF ABBREVIATIONS ....................................................................................... xii LIST OF APPENDICES ............................................................................................... xii CHAPTER I -— Introduction .......................................................................................... 1 CHAPTER II — Literature Review Pharmacological Studies ...................................................................................... 4 Pesticidal, ecological, agricultural, and related studies ........................................ 16 CHAPTER HI — Isolation, Purification, Identification and Quantification of Biologically Active Compounds fi'om Magnolia salicifolia Abstract ............................................................................................................. 19 Introduction ....................................................................................................... 21 Materials and Methods ...................................................................................... 22 General Experimental Procedures ...................................................................... 22 HPLC Analyses ................................................................................................. 22 Circular Dichroism Analyses .............................................................................. 23 vi Geranial (1) and Neral (2) .................................................................................. 23 T rans-anethole (3), methyl eugenol (4), iso-methyl eugenol (5), and Myristicin (6) ..................................................................................................... 24 Costunolide (7) .............. 26 W Quantification of ISO-methyl eugenol (5), Costunolide (7), and Parthenolide (8) .27 Circular Dichroism (CD) of Costunolide (7) and Parthenolide (8) ...................... 28 Results and Discussion ................................................................ , ...................... 28 CHAPTER IV —- Biologically Active Extracts fi'om Magnolia .spp. and Compounds from Magnolia salicifolia Abstract ............................................................................................................. 42 Introduction ....................................................................................................... 44 Materials and Methods ...................................................................................... 45 Plant Materials ................................................................................................... 45 Extraction .......................................................................................................... 45 Bioassays ........................................................................................................... 45 Statistical Analyses of Gypsy Moth Data ............................................................ 49 Results and Discussion ....................................................................................... 49 CHAPTER V — Summary and Conclusion ................................................................... 65 BIBLIOGRAPHY ................................................................................................... . ...... 68 APPENDICES ......................................................... W ..................................................... 76 vii . LIST OF TABLES Table 3.1 Yield of compounds (1-5 and 7) from dried M saliczfolia plant parts ............ 34 Table 4.1 Dry matter content for Magnolia spp. plant parts .......................................... 50 Table 4.2 The weight of extracts from various Magnolia spp. plant parts ...................... 51 Table 4.3 Mortality of A. egyptii for hexane extracts at 250 ppm after 24 h .................. 53 Table 4.4 Mortality of A. emtii for EtOAc extracts at 250 ppm after 24 h .................. 54 Table 4.5 Average weights of L. dispar larvae after six days. Treatments for pure compounds were bioassayed at 100 ppm ....................................................... 55 Table 4.6 Average weights of L. dispar larvae after six days. Treatments for crude Magnolia spp. extracts and pure compounds were bioassayed at 250 and 100 ppm, respectively ........................................................................................... 56 Table 4.7 Average weights of L. dispar larvae after six days. Treatments for crude Magnolia spp. extracts and pure compounds were bioassayed at 250 and 100 ppm, respectively ........................................................................................... 57 Table 4.8 Average weights of L. dispar larvae after six days. Treatments for crude Magnolia spp. extracts were bioassayed at 250 ppm ...................................... 58 Table 4.9 Average weights of L. dispar larvae after six days. Treatments for pure compounds were bioassayed at 100 ppm ....................................................... 59 viii Table 4.10 Mosquitocidal activities of pure compounds fromM salicifolia on 4th instar A. egyptii after 24 h ........................................................................... 61 Table 4.11 Anti-cancer activity of citral (l and 2), trans-anethole (3), methyl eugenol (4), iso-methyl eugenol (5), myristicin (6), costunolide (7), parthenolide (8), camptothecin, and etoposide ....................................................................... 64 ix LIST OF FIGURES Figure 3.1 A representative HPLC chromatogram of M salicifolia fruit extract showing iso-methyleugenol (5), costunolide (7), and parthenolide (8) in absorption units, AU. Fruit extracts and pure costunolide, parthenolide, and iso-methyl eugenol were analyzed on a capcell pak C-18 (5 pm, 4.6 x 250 mm) column (Dychrome). The mobile phase ACNszO (80:20 v/v) was used under isocratic conditions at a flow rate of 0.5 ml-min'l. Injection volumes were 10 ul. Data was collected at 217 nm and 222 nm for costunolide and parthenolide, respectively ........................................................................................ 37 Figure 3.2 Percent yield of iso-methyl eugenol (5), costunolide (7), and parthenolide (8) fromM. salicifolia fruit extracts .................................................................................... 38 Figure 3.3 CD spectrum of costunolide, 7 (l mg-ml") at 220 nm shows a molar elipticity, e at a maximum 1.155E+02 mdeg. At 262 run an extremum at 7 .005E+00 mdeg was observed for costunolide ............................................................................................... 40 Figure 3.4 CD spectrum of parthenolide, 8 (l mg-ml“) at 208 nm shows a molar elipticity, e at a extremum at 2.942E+01 mdeg ............................................................................. 41 BNPL CHCl, CD CDCl, DEPT rel.int. SPP- YPD LIST OF ABBREVIATIONS Acetonitrile Bioactive Natural Products Laboratory Chloroform Circular dichroism Deuterated chloroform Distortionless Enhancement by Polarization Transfer Dimethyl sulfoxide ' Electron impact ionization mass spectrometer Ethyl acetate Molar elipticity Fast atom bombardment mass spectroscopy High performance liquid chromatography Nuclear magnetic resonance Photodiode array detector Potato dextrose agar Thin layer chromatography Vacuum liquid chromatography Proton nuclear magnetic resonance 13Carbon nuclear magnetic resonance Chemical shifts Doublet of doublet Coupling constant Methanol Melting point Mass spectroscopy Molecular weight Mass-to-charge ratio Minimum Inhibitory Concentration Relative intensity specie species Yeast Potato Dextrose Agar ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooo 0000000000000000000000000000000000000000000000 oooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo 000000000000000000000000000000000000000000000 ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooo xi APPENDIX I APPENDIX II APPENDIX III APPENDIX IV APPENDIX V APPENDIX VI APPENDIX VII APPENDIX VIII APPENDIX IX APPENDIX X APPENDD( XI LIST OF APPENDICES lHNMR - citral (1 and 2) ......................................... 76 l3CNMR - citral (1 and 2) ....................................... 77 lHNMR - trans-anethole (3) .................................... 78 lI-INMR - methyl eugenol (4) .................................. 79 13CNMR - methyl eugenol (4) ................................. 80 lI-INMR - iso-methyl eugenol (5) ............................. 81 13CNMR - iso-methyl eugenol (5) ............................ 82 lIINMR - myristicin (6) ................ ........................... 83 lHNMR - costunolide (7) ......................................... 84 l3Cvam - costunolide (7) ........................................ 85 Gypsy moth caterpillar diet ...................................... 86 xii CHAPTER I Introduction The genus Magnolia has played a multi-dimensional role in the history of man. When one speaks of magnolias, images of beautiful trees with large showy flagrant flowers and attractive vegetation are visualized. For these reasons and the fact that magnolias are generally pest-fi'ee, they find their place in many ornamental plantings throughout the world. The campus of Michigan State University contains an abundance of magnolia species. Also, it is known for its occasional use as a source of lumber (Callaway, 1994). Species such as Magnolia grandiflora L., Magnolia virginiana L., and Magnolia acuminata L. have been harvested for lumber in North America. The wood of Magnolia hwoleuca Siebold and Zuccarini is prized in Japan for use in making utensils, tools, and toys. Magnolia spp. were used in traditional medicines in China as far back as 1083 BC. Interestingly, these traditional medicines are still available and used today in China. There are three entries in the Chinese Pharmacopoeia for Magnolia spp. (Tang and Eisenbrand, 1992). “Xinyi”, Flos Magnolia is the dried buds of Magnolia biondii Pamp., Magnolia denudata Desr., or Magnolia sprengeri Pamp. Xinyi is used to treat colds and nasal catarrh. “Houpohua”, Flos Magnoliae, used as a stomachic, is the dried buds of 2 Magnolia ofliCinaIis Rehd. et. Wils. or Magnolia oflicinalis var. biloba Rehd. et. Wils. Finally, the dry stern, branch, and root of M officinalis orM officinalis var. biloba known as “Houpo”, Cortex Magnoliae oflicinalis is used as a stomachic and antihistamine. In addition to the three entries, there are a number of other articles in the literature mentioning Magnolia spp. as traditional Asian medicines. “Shin-I” (also called “Hsin-I” or “Hsenyi”), a Japanese folk medicine, used to cure rhinitis and nasosinuitis, is from the dried flower buds of Magnolia saliczfolia Maxim or Magnolia kobus DC. Native North Americans as well as Asians have and still use some Magnolia spp. for treatment of a variety of ailments (Moerman, 1986). Among the tribes using indigenous Magnolia App. have been the Cherokee, Choctaw, Kosati, Houma, Iroquois and Rappahannock. The bark of M acuminata was used by the Cherokee and Iroquois to treat stomach cramps and tooth aches. The Iroquois also used this species to treat venereal diseases. The Houma used a decoction of leaves and stems of M virginiana as a febrifuge and a cold remedy. Other species had similar uses by difi‘erent tribes. The Rappahannock were the only tribe to use Magnolia as a hallucinogen. In order to achieve this affect, they reportedly inhaled the leaves or bark of M virginiana contained within cupped hands. It is noteworthy to point out that none of these Magnolia spp. used by Native Americans have been analyzed chemically for their purported activities. The bark of M acuminata, M viginiana, and M tripetala were first listed as official drugs in the 1787 United States Pharmacopia (USP). They were used to treat malaria, rheumatism, gout, and respiratory ailments (Culbreth, 1927). However, with the advent of more readily available drugs, the 3 use of magnolia bark declined, and the genus has not appeared in North American pharrnacopoeias since 1900. Much of the driving force on magnolia research has been its long history of use in folk medicine. In addition to the exploration of magnolias for pharmacologically active compounds, there has been some interest in secondary metabolites possessing pesticidal and allelopathic activities. Many biologically active compounds have been isolated from Magnolia .spp. as explained in Chapter II. It is our hypothesis that other Magnolia spp. also have the potential to generate novel bioactive compounds for human and agricultural pest management. The purpose of the present work is to investigate the presence of novel bioactive compounds inMagnolia spp. which were not previously studied. These studies will help lead to a greater understanding of the chemistry of Magnolia spp. as well as the potential for new pharmaceutical and pesticidal compounds. M salicifolia, M denudaia, M kobus var. stellata and M kobus var. loebneria were selected for phytochemical analyses for this project. Only the pharmacologically active compounds are reported fi'om M salicifolia and to a lesser extent, M denudata and M kobus var. stellata (M stellata). M kobus var. loebneria has not been investigated for bioactive compounds In this study, extracts from all species will be examined for activity against fungal and bacterial pathogens, mosquito larvae, gypsy moth larvae, nematodes, and topoisomeraes I and II (anti-cancer assay). CHAPTER II Literature Review This chapter reviews papers dealing with a veritable potpourri of bio-active compounds derived from various Magnolia spp. Neolignans, lignans, phenyl propanoids, sesquiterpenes, and alkaloids are among the compounds with reported biological activity. Neolignans, lignans, and phenyl propanoids are derived fi'om the shikimic acid pathway. Sequiterpenes and other terpenes are derived from the mevalonic acid pathway. Alkaloids on the other hand, are hypothesized to come from either biosynthetic pathway depending on whether the alkaloid contains a terpene structure or a phenyl propanoid structure. Pharmacological studies One of the earliest studies conducted on magnolia was part of a search for plants having anti-tumor principles (W iedhopf et al., 1973). From the petroleum ether leaf extract of M grandiflora L., Wiedhopf and coworkers isolated parthenolide, a sesquiterpene lactone that showed inhibitory activity against human epiderrnoid carcinoma (skin cancer). The exocyclic double bond adjacent to the carbonyl group of the lactone ring of sesquiterpene lactones, was demonstrated as the most important fiinctionality for parthenolide cytotoxicity (Lee et al., 1971). A more recent study conducted on the root bark of Magnolia denudata Desr. has yielded compounds cytotoxic to P388 leukemia cells (Funayama et al., 1995). The active constituents , costunolide (1.3 ug-ml") and parthenolide (0.5 ug-rnl") were isolated from the CHCl3 extract of the dried root bark. Other less cytotoxic compounds were isolated as well. Among them were a phenylpropanoid, trans-isomyristicin and two lignans, sesamin and kobusin. Costunolide had been isolated previously fiomM grandiflora (El- Feraly, 1979). The lignan kobusin, isolated earlier from Magnolia campbellii and M mutabilis (Talapatra et al., 1975), was also later found in Magnolia steIIata along with sesamin (Iida et al., 1983). O OCH sesamin 6 The anti-tumor activities for neolignans of M oflicinalis also were explored (Konoshima, 1990). Magnolol , honokiol, and monoterpenylmagnolol exhibited inhibitory efl‘ects on Epstein-Barr virus (human herpes 'virus) activation. Among these compounds, magnolol displayed the highest anti-tumor activity. Honokiol was isolated originally from the MeOH extract of the bark of M obovata Thunb. (Fujita et al., 1972). / O O OH monoterpenylmagnolol honokiol / Baek et a1. (1992), during the course of screening anti-oxidants fiom plants, found a neolignan from the bark of M officinalis. 5, 5'-di-2-propenyl-2-hydroxy-3, 2', 3'- trimethoxy-l-l'-biphenyl shown below, was found to have anti-oxidant activity similar to 2, 6-di-tert-butyl-4-methylphenol (BHT) or 3-tert-butyl-4-hydroxyanisole (BHA). Anti- oxidants are compounds that inactivate free radicles in the body. Free radicles can promote the grth of cancerous cells by initiating spontaneous mitosis. H,co HO OCH, OCH, 5, 5' -di-2-propenyl-2-hydroxy-3, 2', 3' -trimethoxy-l, 1'-biphenyl Furthermore, phenolic anti-oxidants in wine have demonstrated the ability to inhibit human low density lipid (LDL) oxidation in vitro (Frankel et al., 1995). Frankel mentions other studies suggesting that oxidation of LDLs may play a major role in atherosclerosis. Kimura and coworkers (1983) examined the crude alkaloidal fi'actions derived from “Shin-I” (a traditional Sino-Japanese medicine) for the presence of neuromuscular blocking activity. From the dried buds of M salicifolia, three alkaloids, d-coclaurine, d- reticulin, and yuzirine were found to reduce acetylcholine-induced twitching of fi'og skeletal muscle. P1300 H,C0 H,co \ O NH O N... O HO HO HO N O W O O HO H,oo HO yuzirine d—coclaurine ‘ d-reticuline 8 The active component of the purported anti-allergy activity of Shin-I was identified as magnosalicin (T surga et al., 1984). Magnosalicirr, a neolignan, was found in the CHC13 extract of M salicifolia buds. OCH, OCH, OCH, magnosalicin The anti inflammatory effect of Shin-I was found to be supported by magnosalin’s and magnoshinin’s ability to inhibit angiogenesis and granuloma formation (Kimura et al., 1990). These lignans were isolated originally fromM salicifolia (Kikuchi et al., 1983). magnosalin magnoshinin In a later study, Kimura and workers (1989) demonstrated the negative inotropic effects of (+)-R-coclaurine and (+)-S-reticulin as well as their antagonistic actions to 'ihigenamine, the cardiotonic principle of aconite root (Aconitum 519.). In essence, (+)-R- 9 coclaurine and (+)-S-reticulin act as muscle relaxants and to prevent muscle contractions induced by ihigenamine. H co \ H co 3 3 HO HO HCO 3 (+)-R-coclaurine (+)-S-rcticulcnc :l: higcnaminc The traditional Chinese drug, Xinyi, was examined for the presence of platelet activating factor (PFA) antagonists or simply, blood anti-coagulants (Pan et al., 1987). Six PFA antagonists were identified from the methylene chloride extract of M biondii. The PFA antagonists were pinoresinol dimethyl ether, magnolin, liroresinol-B-dimethyl ether, fargesin, demethoxyaschantin and aschantin. HIIII- HIIIH R ,. I 0 II 0 HJCO liroresinol-B-dimethyl ether: R= -OCH3 , R1: -OCI-I3 magnolin: R= -OCI-I3 , R1: ~I-I pinoresinol dimethyl ether: R= -H, R1: -H All of these were lignans with similar structures. The first four PFA antagonists were isolated previously from the flower buds of M fargesii (Kakisawa et al., 1972). More recently, pinoresinol dimethyl ether was isolated fromM saulangiana (Abdallah, 1993). 10 \_ O demethoxyaschantin Calcium ions are involved in the regulation of muscle contractions. Compounds that interfere with or diminish muscle contractions via Ca++ antagonism, may be considered muscle relaxants. Chen et a1. (1988) isolated three Ca” antagonistic principles fi’om “Hsin-I”. Three neolignans; fargesone A, fargesone B, and denudatin B were purified from the CHCl3 extract of the flower bud from M fargesii Cheng. Denudatin B 11 has also been isolated from fresh leaf extract of M liliflora using CHCl3 (Iida and Ito, 1983) OCH. OCH. H’C H,C0 H’C H,co Q... 00-- 0 W0 H 0 W0 0 H i 0 {83830119 A fagecone B WCZKV ECO—i >" 0 0 11,00 dcnudatinB Three compounds possessing neurotrophic activity were characterized from the root bark of M obovata Thunb. (Fukuyama et al., 1989, 1990, 1992). Three novel sesquiterpene neolignans, eudesobovatol A, clovane magnolol, and caryolanemagnolol not only augment neurite sprouting, but also greatly enhance choline acetyl transferase activity. Choline acetyl transferase catalyzes the reaction between acetyl-CoA and choline to yield the neurotransmitter, acetylcholine, which triggers muscle contraction (V oet and Voet, 1990). These results support the alleged use of Magnolia spp. as a teatment for neurosis. 12 cudesbovatol A clovanemagnolol caryolanemagnolol In another article, Fukuyama et al. (1993) reported a novel trilignan, magnolianin. This unusual molecule was found to possess potent 5-lipoxygenase-inhibitory activity. Lipoxygenase is an enzyme responsible for converting arachidonate into leukotrienes (V oet and Voet, 1990). Leukotrienes are local hormones involved in many of the same functions as prostaglandins. Inflammatory responses, production of pain and fever, regulation of blood pressure, and induction of blood clotting are some of their functions. 13 The significance of these findings support earlier reports of Magnolia spp. being used effectively to treat rhinitis and nasosinuitis. OH ° 0 / O o 0 0‘10 \ magnolianin \ Medicines containing the M obovata bark have been prescribed as stomachic, diuretic, and anti-emetic treatments. Both magnolol and honokiol were determined to be anti-emetic in fi'ogs (Kawai et al., 1993). Another lignan, pinoresinol, originally found in the dried hits of Forsythia suspensa, was found to inhibit cyclic adenosine monophosphate (c-AMP) phosphodiesterase (Nikaido et al., 1981). Later, pinoresinol was isolated from Magnolia spp. c-AMP phosphodiesterase is responsible for degradation of c-AMP into AMP. c- AMP is a second messenger in the action of many hormones (Stryer, 1988). Inhibitors of c-AMP phosphodiesterase act synergistically with hormones that use c-AMP as a second messenger. For example, the action of the hormone vassopressin (a peptide hormone that 14 stimulates water resorption) is enhanced when c-AMP phosphodiesterase is inhibited. A toxicological study of the methanolic extract of M grandiflora heartwood led to the identification of menisperine (N-methyl isocorydine), a phenolic quaternary alkaloid (Rao and Davis, 1982). Toxicity of this MeOH extract was ultimately due to the neuromuscular blocking action of the alkaloid. Davis (1981) isolated this alkaloid and two antibacterial compounds from M grandiflora. Parthenolide, from the yellowed leaves of M gandiflora exihibited antibiotic activity against Bacillus subtilis at an MIC of 3 ug-ml". Anonaine, demonstrated antibacterial activity at 100 rng-rnl‘l against Staphylococcus aureus, Mycobacterium smegmatis, and Candida albicans. Davis also stated that anonaine can employ a hypotensive effect in mice and rabbits and was shown to be an inhibitor of dopaminergic response. Also, the toxicity of anonaine HCl was demonstrated in mice at 200 mg-kg”. A related compound, liriodenine, had similar antibacterial activity to anonaine. H,CD H,C0 HO HCO anonaine liriodenine 15 Quaternary alkaloidal compounds associated with ganglionic-blocking activity have been identified in Magnolia extracts (Davis, 1981). Magnocurarine, magnoflorine, salicifoline, and d-tubocurarine have demonstrated this effect in frogs. CH 3 i 3 +Nl. _ CH3 . . salicifoline magnoflorrne magnocurarrne H,C0\ d-tubocurarine The ether and water extracts of M obovata bark were found respectively to depress activity and induce a quick paralysis of respiration on mice and chicks (W atanabe et al., 1973). In the same study, the ether extract was found also to be anticonvulsant. 16 Pesticidal, ecological, agricultural, and related studies Phenolic constituents of M grandiflora L., namely magnolol, honokiol, and 3, 5'- diallyl-2'-hydroxy-4-methoxy-biphenyl were found to have superior antibacterial activity against Bacillus subtilis, Staphylococcus aureus, and Mycobacterium smegatis when compared to streptomycin sulfate (Clarke et al., 1981). Additionally, magnolol and honokiol were found to be moderately active against Candida albicans, Saccharomyces cerevisiae, Aspergillus niger and strongly active against T richophyton mentagrophytes when compared to amphotericin B. 3, 5'-diallyl-2'-hydroxy-4-methoxy—biphenyl demonstrated milder activity against S. cerevisae and strong inhibition of growth against T. mentagrophytes when compared to amphotericin B. Growth inhibitory and antimicrobial activities were demonstrated by cyclocolorenone, a sesquiterpene ketone that was isolated from the leaves of M grandiflora (Jacyno et al., 1991). Cyclocolorenone inhibited the growth of etiolated wheat coleoptiles completely and was comparable to morphactin, juglone, and abscisic acid at 10'3 M and 10" M, respectively. Cyclocolorenone inhibited only 58% whereas morphactin, juglone, and abscisic acid inhibited 38, 100, and 100%, respectively. cyclocolorenone 17 4, 4'-diallyl-2, 3'-dihydroxybiphenyl ether (“biphenyl ether”), 3, 5'-diallyl-2'- hydroxy-4'-methoxybiphenyl, and magnolol from leaves of M virginiana L. displayed antifimgal (C. albicans, Aspergillusflavus, Gloesporum sp, and Rhizoctonia 5p), antibacterial (Streptococcus aureus, Staphylococcus epidermis, and Escherichia coli), mosquitocidal (Aedes egyptii), and crustaceacidal (Artemia salina) activities (N itao et al., 1991). Additionally, allelochemicals (biphenyl ether and magnolol) of M virginiana OH OH \ / / \ 4, 4'-diallyl-2, 3'-dihydroxybiphenyl ether 3, 5'-di311y1-2'-hydme-4'-m¢thoxybiph¢nyl leaves were found to be toxic to Papilio palamedes and reduced survival rates of Papilio troilus (Nitao et al., 1992). 3, 5'-dially]-2'-hydroxy-4'-methoxybiphenyl was previously isolated from the seeds of M grandrflora (El-Feraly and Li, 1978). Two silkworm (Bombix mori L.) grth inhibitors were isolated from the leaves of M kobus DC. The structures of the two active compounds were elucidated as lignans, sesamin and kobusin. A novel ligan of the same skeletal form was isoalatd from the buds of M fargesii. (+) epimagnolin A was isolated by bioassay directed fractionation from the CHZCl2 soluble portion of the MeOH fi'uit extraction and was found to have growth inhibitory activity against larvae of Drosophila melanogaster (Miyazawa et al., 1994). 18 H,CD H,C0 The synergistic activity of sesamin, pinoresinol, and eudesamin with pyrethrum insecticides was reported (Haller et al., 1941). However, these compounds were not isolated from Magnolia spp. until much later (Fukayama et al., 1995, Kikuchi et al., 1982, Kakisawa et al., 1970, Iida et al., 1983). eudesmln pinoresinol CHAPTER HI Isolation, Purification, Identification, and Quantification of Biologically Active Compounds from Magnolia salicifolia. Abstract Bioassay-directed isolation and purification of Magnolia salicifolia extracts yielded six mosquitocidal compounds. Their structures were determined by lI-INMR, l3CNMR, and MS determinations. The pleasant-smelling monoterpenes, gerarrial and neral, collectively known as citral, were isolated from hexane extract of the bark. T rans- anethole and methyl eugenol, both phenylpropanoids, were isolated from hexane extract of the leaves. Myristicin, a methylenedioxy phenyl compound, was first detected as an additional compound in the lHNMR spectrum of trans-anethole. The positional isomer of methyl eugenol, iso-methyl eugenol, was isolated from the hexane extract of leaves, fruits, and flowers. It also was obtained from the EtOAc extract of fruits. Costunolide, a sesquiterpene lactone was obtained from the hexane and EtOAc extracts of the hits of M salicifolia for the first time. Iso-methyl eugenol, costunolide, and an additional related sesquiterpene lactone, parthenolide, were detected and quantified in M salicifoliafor the first time. Circular dichroisms (CD) of costunolide and parthenolide also are presented for the first time. The CD of costunolide and parthenolide indicated that they are dextro (+) 19 20 and levo (-), respectively. Yreld of a geranial and neral mixture was 0.08%. Yield of compounds isolated fi'om leaves were 0.02, 0.25, and 0.43% for trans-anethole, methyl eugenol, and iso- methyl eugenol, respectively. Yield of iso-methyl eugenol and costunolide from fruits were 0.92 and 0.60%, respectively. From flowers, 0.24% of iso-methyl eugenol was isolated. Yields were calculated on a dry-weight basis. 21 Introduction Previously, compounds isolated from M salicifolia have been those associated primarily with pharmacological activities. In particular, compounds with anti-allergy and anti-inflammatory actions have been identified such as magnosalicin, magnoshinin, and magnosalin (T surga et al., 1984, Kimura et al., 1990, Kikuchi et al., 1983). The activity of these compounds supported the purported use of M salicifolia buds as a treatment for rhinitis, sinusitus, and nasosinusitus. The alkaloidal fraction of M salicifolia yielded d- reticuline, d-coclaurine, and yuzirine which were found to possess neuromuscular blocking action (Kimura et al., 1983). Quaternary alkaloidal compounds associated with ganglionic-blocking activity have been identified in Magnolia extracts (Davis, 1981). Magnocurarine, magnoflorine, salicifoline, and d-tubocurarine have demonstrated this effect in fi'ogs. To a lesser extent, pesticidal compounds isolated from Magnolia spp. having been reported. 4, 4'-diallyl-2, 3'-dihydroxybiphenyl ether (“biphenyl ether”), 3, 5'-diallyl-2'- hydroxy-4'-methoxybiphenyl, and magnolol from leaves of M virginiana L. displayed antifungal (Candida albicans, Aspergillusflavus, Gloesporum .51)., and Rhizoctonia 5p), antibacterial (Streptococcus aureus, Staphylococcus epidermis, and Escherichia coli), mosquitocidal (Aedes egyptii), and crustaceacidal (Artemia salina) activities (Nitao et al., 1991) In chapter IV, hexane extracts fiom leaves, stems, flowers, and fruits and EtOAc extracts fromM salicifolia were found to be 100% lethal against mosquito larvae. There are no reports of mosquitocidal constituents inM salicifolia. 22 Materials and Methods General Experimental Procedures 1H and 13CNMR spectra were recorded on Varian VXR 300 and 500 MHz 'spectrometers and were in CDCl3 solutions. EIMS and FABMS were recorded on JEOL JMS-AXSOS and JEOL JMS-HXl 10 mass spectrometers. The melting point, were recorded on a Thomas Model 40 micro hot-stage apparatus and are not corrected. Purification and subsequent isolation of bioactive constituents fi'omM salicifolia was accomplished by TLC (Analtech 20x20 cm silica gel GF-TLC plates with fluorescent . indicator) for stems, leaves, flowers, and hits. Detection of spots/bands was accomplished by visualizing under a UV lamp at 254 nm and developed by spraying with 10 % HZSO4 followed by heating for 5 min at 120°C. Initial purifications were performed by vacuum liquid chromatography (VLC) or column chromatography (CC) over silica gel (Analtech Silica Gel 60A pore size, 35-75 pm particle size). HPLC Analyses Fruit extracts and pure costunolide, parthenolide, and iso-methyl eugenol were analyzed on a capcell pak C-18 (5 pm, 4.6 x 250 mm) column (Dychrome). The mobile phase ACN:HZO (80:20 v/v) was used under isocratic conditions at a flow rate of 0.5 ml-min". The samples were filtered through a 0.22 p PTFE filter (Scientific Resources Inc), prior to the injection. The injections (10 pl) were performed by the Waters 717 autosampler. A Waters 991 photo diode array detector (PDA), (Millipore Corporation, Milford, Massachusetts) was used to collect data at 217 nm and 222 nm for costunolide and parthenolide, respectively. The calibrations and quantifications were carried out using 23 Waters Millennium 2010 Chromatography Manager GPC software, version 2.0 (Millipore Corporation, Waters Chromatography division, Milford, Massachusettes). Circular Dichroism Analyses Pure costunolide and parthenolide at 1mg-25ml'1 in MeOH were analyzed on JASCO J-710 71CD-ORD spectropolarimeter (Jasco Incorporated, Japan). The molar elipticities were spectra were plotted on 7475A Hewlett Packard plotter (Hewlett Packard Corporation, Palo Alto, California). Nitrogen (99.99%) was produced by a nitrogen generator model NG-150 (Peak Scientific, Chicago, Illinois) at a rate of 20 liters-min". Geranial (l) and neral (2) Initial purification of the mosquitocidal hexane extract from the stem (4 g) was carried out by VLC on silica gel using solvent systems 4:1 hexane-ether, CHCl3, 4:1 CHCl3-MeOH, and MeOH, respectively, to yield four fractions, A-D. Fraction B (740 mg) was found to be mosquitocidal. Repeated purification by TLC (25:1, petroleum ether-acetone) gave an oily compound, citral (80.2 mg; Rf 0.19). The presence of geranial (l) and neral (2) was confirmed in this oil by NMR studies (Appendices I and H for respective lI-INMR and l3CNMR spectra of citral). Compound 1, geranial(C10H,60, MW 152); lI-INMR: 6 9.98 (1H, d, J=8.1 Hz, CH0), 5.86 (1H, d, J=8.1 Hz, H-2), 2.56 (2H, t, J=7.5 Hz, H-4), 2.17 (2H, m, H-S), 5.05 (1H, m, H-6), 1.66 (3H, s, H-8), 1.58 (3H, s, H-9), 2.14 (3H, s, H-10); 13CNMR: 6 191.13 (C-l), 127.22 (C-2), 163.70 (C-3), 40.43 (C-4), 25.56 (C-5), 122.39 (C-6), 132.73 (C—7), 17.53 (C-8), 17.40 (C-9), 25.46 (C-10). Compound 2, neral (CmeO, MW 152); lHNMR: 5 9.87 (1H, d, J=8.1 Hz, T ram pressr used V (160 n C b)" 24 CHO), 5.86 (1H, d, J=8. l, H-2), 2.56 (2H, t, J=7.5 Hz, H-4), 2.17 (2H, m, H-5), 5.05 (2H, m, H-6), 1.66 (3H, s, H-8), 1.58 (3H, s, H-9), 2.14 (3H, s, H-lO); l3CNMR: 5 190.62 (C-l), 128.47 (C-2), 163.70 (C-3), 32.41 (C-4), 26.86 (C-S), 122.09 (C-6), 132.73 (C-7), 17.53 (C-8), 17.40 (C-9), 24.88 (C-10). T runs-anethole (3), methyl eugenol (4), iso-methyl eugenol (5), and myristicin (6) Purification of the mosquitocidal hexane extract fiom leaves (1.1 g) by medium pressure column chromatography on silica gel gave six fiactions, A-E. Solvent systems used were 4:1 and 1:1 hexane-acetone, acetone, and MeOH, respectively. Fractions B (160 mg) and C (478 mg) were found to be mosquitocidal. Purification of fractions B and C by TLC (50:1, hexane-acetone) gave compounds 3 (6.3 mg; Rf 0.42), 4, (72.2 mg; Rf 0.14) and 5, (123.7 mg; Rf 0.13). The presence of myristicin (6) was confirmed by 1HNMR in a sample of 5. Myristicin however, was not isolated in the pure form fi'om the leaves of M salicifolia. See Appendices HI, IV, VI, and VIII for lHNMR spectra of compounds 3, 4, 5, and 6, respectively. Appendices V and VH contain l3CNMR spectra for compounds 4 and 5, repectively. Compound 3, trans-anethole (CmI-IHO, MW 148); lHNMR: 6 6.83 (1H, dd, J=8.62 Hz, H-2,6), 7.25 (II-I, dd, J=8.6 Hz, H-3,5), 6.34 (1H, d, J=15.8, vinyl, H-7), 6.09 (II-I, m, vinyl, H-8), 1.86 (3H, d, J=6.6Hz, H-9), 3.80 (3H, s, OCH3). Compound 4, methyl eugenol; 1,2 dimethoxy-4-(2-propenyl) benzene (CuHqu, MW 178); EIMS: m/z (%) 178 (100), 163 (100), 147 (100), 135 (60), 115 (43), 103 (92), 91 (90), 77 (44), 65 (30), 51 (23); lHNMR: 6 6.71 (1H, s, H—3), 6.73 (1H, d, J=10.94, H-6), 6.81 (1H, d, J=7.95, H-S), 3.34 (2H, d, J=6.85, H-7), 5.96 (1H, m, H-8), 25 5.07 (2H, d, H—9), 3.87 (3H, s, -OCH3), 3.86 (3H, s, OCH3); l3CNMR: 6 137.55 (C-4), 132.49 (C-3), 147.22 (C-2), 148.74 (C-l), 115.45 (C-6), 120.34 (C-5), 39.65 (C-7), 111.17 (C-8), 111.10 (C-9), 55.79 (C-10), 55.65 (C-1 1). Compound 5, iso-methyl eugenol; 1,2 dimethoxy-4-(1-propenyl) benzene (C,,H,,o,, MW 178); EIMS: m/z (%) 178 (100), 163 (100), 147 (24), 135 (30), 115 (15), 107 (61), 91 (53), 79 (18), 65 (13), 51 (9); lI-INMR: 6 6.81 (1H, s, H-3), 6.84 (1H, d, H—6), 6.89 (1H, d, H-5), 6.34 (1H, d, J=15.78, H-7), 6.10 (1H, m, H—8), 1.87 (3H, d, J=6.56, H-9), 3.89 (3H, s, -OCH3), 3.87 (3H, s, OCH3); l3’CNMR: 6 130.98 (C-4), 130.44 (C-3), 147.98 (C-2), 148.81 (C-l), 118.48 (C-6), 123.64 (C-S), 111.00 (C-7), 108.28 (C-8), 18.227 (C-9), 55.75 (C-10), 55.61 (C-ll). Compound 6, myristicin; l-methoxy-2-3-methylenedioxy-S(2-propenyl) benzene (CuHuO3, MW 192.2); lI-INMR: 6 6.38 (1H, s, H—4), 6.35 (II-I, s, H-6), 5.93 (2H, s, O- CI-I,-O), 5.91 (II-I, m, H-8), 5.10 (2H, d, J=7.5, H-9), 3.88 (3H, s, -OCH3), 3.29 (2H, d, J=6.6, H-7). Compound 5 was isolated also from the fi'uits and flowers of M salicifolia. Purification of hexane extract of flowers (1 .26 g) by VLC over silica gel using 4:1 hexane- acetone, CHCl3, 4:1 CHCl3-MeOH, and MeOH solvent systems, yielded four fiactions, A- D, respectively. Fractions A-C (367.8, 587.3, and 58.2 mg, respectively) were mosquitocidal. Repeated purification of fraction B by TLC (15:1 hexane-acetone) yielded compound 5 (100.8 mg). Purification of the EtOAc extract of fruits (1 .26 g) by VLC over silica gel using EtOAc, 4:1 EtOAc-acetone, CHCl3' 3:1 CHC13-MeOH, and MeOH solvent systems yielded four fractions A-D, respectively. Fraction A (757.8 mg) was mosquitocidal. 26 Purification of fraction A by TLC (25:1 hexane-acetone) gave four bands, I-IV. Further purification of band 111 by TLC (40:1 hexane-acetone) afforded compound 5 (39.2 mg). 1H NMR confirmed the identity of compound 5. Purification of hexane extract of flowers (1.38 g) by VLC gave four fi'actions A—D. The active fiaction, B was purified further by TLC (15:1 hexane-acetone) and yielded compound 5 (108.1 mg). The identity of compound 5 was confirmed by 1H NMR spectra. Costunolide (7) Purification of the hexane extract of M salicifolia fruit (1.26 g) by VLC over silica gel using 4:1 hexane-acetone, CHCl3, 4:1 CHCl3-MeOI-I, and MeOH solvent systems yielded four fractions, A-D, respectively. Fractions A—C (367.8, 587.3, and 58.2 mg, respectively) were equally mosquitocidal. Repeated purification of fraction B by TLC (15:1 hexane-acetone) yielded compound 7, (63.5 mg; Rf 0.14) as a colorless crystalline solid (Appendix IX and X for the lHNMR and l3CNMR spectra of compound 7, respectively). Compound 7, costunolide; 1 (10), 4 (l 1) (B)-germacratriene-12, 6- olide (C15H2002, MW 232); mp. 101-106° C; EIMS: m/z (%) 232 (56), 217 (24), 204 (8), 189 (11), 175 (15), 161 (14), 150 (20), 135 (15), 123 (57), 109 (73), 91 (39), 81 (100), 67 (23), 53 (53); lI-INMR: 6 6.26 (1H, d, J=3.35, H-13'), 5.52 (1H, d, J=3.07, H—13), 4.83 (1H, br dd, J=6.42, 7.81, H-l), 4.73 (1H, br d, J=10.05, H-5), 4.56 (1H, t, J=9.34, 9.36, H-6), 2.55 (1H, t, J=8.23, 7.68, H-7), 2.45 (II-I, dd, J=8.51, 7.82, H—3'), 1.97-2.35 (6H, m, H-2', H-3, H-8, H-9/9'), 1.68 (1H, m, H-2), 1.67 (3H, s, H-lS), 1.41 (3H, s, H-14). l3CNMR; 6 169.96 (012), 140.97 (01 1), 139.57 (04), 136.44 (010), 126.76 (C-5), 27 126.54 (C-l), 119.14 (C-13), 81.40 (C-6), 49.90 (C-7), 40.48 (C-3), 38.51 (C-9), 27.53 (C-2), 25.69 (C-8), 16.84 (C-15), 15.60 (014). Purification of the EtOAc extract (1.26 g) by VLC over silica gel using EtOAc, 4:1 EtOAc-acetone, CHC13, 3:1 CHCls-MeOI-I, and MeOH solvent systems yielded four fractions A-D, respectively. Fraction A (757.8 mg) was mosquitocidal. Further purification of fi'action A by TLC (25:1 hexane-acetone) gave four bands, I-IV. Repeated purification of band IV by TLC (10:1 hexane-EtOAc) yielded costunolide 7 (31.9 mg). The structure for compound 7 was confirmed by lHNMR, l3CNMR, DEPT, and MS. Quantification of Iso-methyl eugenol (5), Costunolide (7), and Parthenolide (8). Pure costunolide (7) and parthenolide (8), were. used as standards for quantifying these compounds in three replicated M salicifolia fruit extracts. A set of mixed standards of 7 and 8 containing 1mg of each was dissolved in ACN (1 ml), and was used as a the stock solution to generate calibration curves. Serial dilutions of these stock solutions were made to obtain solutions containing 125, 62.5, 31.25, 15.63, 7.81 and 3.91 ugml" of ACN. Pure iso-methyl eugenol, compound 5, was used as a standard for quantifying this compound inM salicifolia fi’uit extracts. Calibration curves of compound 5, 7, and 8, then were generated from their HPLC data by Millennium 2010 (Millipore Corporation, Milford, Massachusettes) using the following equation: y = A + Bx; where y = response calculated by Millennium software for peaks at 217 nm, A = y - intercept of the calibration curve, x = component amount ranging from 125 to 3.91 ugml“. For compound 5, y = response was calculated for peaks at 222 nm. The lyophilized and milled fruits were extracted exhaustively in triplicate, (1.11, 28 1.08, and 1.06 g, respectively) with MeOH at room temperature to afford three extracts A-C (501.6, 502.2, and 462.2 mg, respectively). These extracts were dissolved in ACN (20 mg-ml"). The resultant solutions were filtered through 0.22 pm filters and analyzed (10 111) by HPLC. All samples were quantified in triplicate. HPLC analyses of hit extracts showed three major peaks of interest in the PDA chromatogram and were confirmed to be iso-methyl eugenol (5), costunolide (7), and parthenolide (8) at retention times 8.49, 9.49, and 7.38 min, respectively (Figure 3.1). Circular Dichroism (CD) of Costunolide (7) and Parthenolide (8) Pure costunolide and parthenolide were dissolved in MeOH separately (1 mg-ZS ml") and determined their CDs under the following conditions: scan mode (wavelength), bandwidth (1.0 nm), sensitivity (100 mdeg), response (2 s), wavelength range (3 50-200 nm), step resolution (1 nm/data), scan speed (50 nm-min"), and accumulation (2) (Figure 3.3 and 3.4). Results and Discussion The monoterpene citral, which contains geranial (1) and neral (2) and the phenylpropanoid trans-anethole (3) were isolated previously from the leaves of M kobus and M salicifolia respectively (Fujita, 1955). The spectral data of geranial and neral, both geometric isomers of citral, was confirmed from 1H and l3CNMR data. The ratio of geranial to neral was approx. 9:1 in hexane extract of stems. The combined spectra of l and 2 gave identical 1H and l3CNMR chemical shift values, with the exception of the aldehydic protons and carbonyl carbons. The aldehydic proton of 2 is more shielded in the cis position than in the trans 29 geranial (l) neral (2) position and therefore, is shifted upfield. The aldehydic proton of l, on the other hand, is more deshielded than the corresponding proton in 2 because of its close proximity to the double bond at the C-2 and C-3 positions, thus appearing downfield. The doublet observed at 6 5.86 is characteristic of a vinylic proton whose signal is split due to the adjacent aldehydic proton. The multiplet signal observed at 6 5.05 is also in the characteristic absorption region of olefinic protons. However, it suggested the presence of a neighboring methylene group, and a long-range coupling with another proton at 6 2.56. This is indicative of CH2 protons adjacent to a double bond. A multiplet at 6 2.17 was assigned the protons on carbon 5. Singlets observed at 5 1.66, 1.58, and 2.14 and integrated for three protons each were assigned to three olefinic groups. Based on the l3CNMR chemical shifts, this molecule contains ten carbons one of which is a carbonyl. Furthermore, according to the lHNMR spectrum, this carbonyl belongs-to an aldehyde functionality. Peaks at 6 127.22, 122.39, and 132.73 were assigned to olefinic carbons. Aldehydic 1H and 13CNMR for l and 2 were 6 9.98, 191.13 and 9.87, 190.62, respectively. This confirmed the presence of a mixture of two geometric isomers of this 30 monoterpene aldehyde. It finally was concluded that l and 2 were trans and cis form, respectively. Therefore, based on these spectral data the structures were elucidated as l and 2. The lHNMR spectrum of 3 showed three distinctive features. It gave two sets of doublets in the aromatic region at 6 6.83 and 7.25. By Pople notation (Silverstein et al., 1991), this is a classic AA’XX’ system or an AB system which is indicative of para disubstituted aromatic ring. The coupling constants, J Ax was 8.60 Hz. The sharp singlet at D 3.80 is typical of an aromatic methoxy group. The presence of benzylic olefinic proton is indicated by a doublet at 6 6.34. The coupling constant of this proton was J,,,=15.8 Hz and confirmed a trans orientation. The multiplet observed at 6 6.09 was assigned to another olefinic proton on C8 The doublet at 51.86 was indicative of a terminal methyl group. CH, trans-anethole (3) The lHNMR spectra of 4 was similar to 3 in most respects, since both contained aromatic protons, methoxylated aromatic protons, and an unsaturated side chain. The most striking difference between the two spectra was the presence of an additional 31 methoxy group. Two intense singlets for these OMe groups appeared at 63.86 and 3.87. Expansion of the aromatic region revealed the presence of two doublets (much like that of trans-anethole) appearing at 66.81 for C-5 and at 66.73 for C-6. The proton at C-6 is shielded by the electron-donating ability of an adjacent methoxy group. Coupling constants for aromatic protons at C-5 and C—6 were 7.95 and 1094111, respectively, and indicated that these protons are ortho to one another. A singlet, observed at 66.71 was assigned to the proton at C-3. A doublet at 63.34, integrated for two protons, was assigned to the benzylic protons in compound 4. The olefinic proton, appearing as a multiplet at 65.96, was assigned to the C-8 proton in compound 4. The methylene protons were observed as a doublet at 65.07. l3CNMR data supported aromaticity, two olefinic carbons, one methylene carbon, and two oxygenated carbons. ‘The molecular weight was deduced from the above information and finally confirmed by the molecular methyl eugenol (4) iso-methyl eugenol (5) ion at m/z 178 observed in the MS spectrum. Spectral data on compound 5 showed 32 similiarities to compounds 3 and 4. But 5 was similar in every respect to 4 with only one exception; a vinyl group was positioned in 5 where the allyl group was in 4. Compound 6, the last phenyl propanoid identified, displayed a lI-INMR spectrum most closely resembling that of compound 4. A key difl‘erence was a the singlet at 65.93 due to the protons of a methylenedioxy phenyl moiety. Another interesting feature observed in the spectrum was the presence of the only singlet at 67.26 in the aromatic region. This was due to two magnetically equivalent meta protons. Furthermore, only one methoxy group was observed at 63.88. Based on this data, the methoxy group at C- 1, the methylene dioxy moiety at C-2 and C-3, and the allyl firnctio‘nality at C-5, the structure of the compound was determined to be myristicin (6). myristicin (6) The presence of the phenylpropanoids, methyl eugenol (4), iso-methyl eugenol (5), and myristicin (6) were detected earlier by GC-MS analysis of the CHCl3 extract of M salicifolia buds (Tsurga et al., 1984). O-methyl eugenol was isolated from the bark of M oflicinalis (Baek et al., 1992). Methyl eugenol was detected by GC-MS in the essential oils of M liliflora leaves, branchlets, and flower buds (Fujita, 1989). 33 The spectral data for compound 7, costunolide, was identical to 1H and l3CNMR data published earlier (Ming et al., 1989). Rao and workers (1960) originally isolated costunolide from costus root oil of the costus plant, Saussurea lappa Clarke (Asteraceae). Costunolide also has been found in other Magnoliaceae plants, such as the leaves of M grandiflora L. (Castar‘ieda-Acosta et al., 1995), the root bark of Michelia champaca (Jacobsson, 1995) and the bark of Michelia longzflora Blume (Likhitwitayawuid et al., 1988). This is the first report of costunolide fiom M salicifolia. The yields of compounds 1-5 and 7 isolated from respective dried plant parts are shown in Table 3.1. The leaves contained the greatest number of active compounds (i.e., 3, 4, and 5) followed by fruits, flowers, and stems. The fruits however, contained the most active compounds by weight. The structural features (i.e., a gerrnacrane sesquiterpene lactone) of 7 arose from lHNMlL l3CNMR, DEPT, and mass spectral data. Key components of the lI-INMR spectrum of 7 included: two methyl singlets at 6 1.67 and 1.41, two sets of doublets produced by exocyclic methylene protons at 6 6.62 (H-13') and 5.52 (H-13), and a triplet at 6 4.56 (H-6) for lactonic protons. A broad doublet for H—S was observed at 6 4.73 (J=10 Hz) due to a CH2 coupling. In the olefinic region, a broad doublet of a doublet was observed at 6 4.83 and assigned to C-1. The l3CNMR spectrum of 7 contained 15 carbons with several in the olefinic region, thereby implying the possibility of a sesquiterpene with unsaturations. Also, a peak at 6169.96 (C-12) was indicative of a carbonyl group of an ester or a lactone. Moreover, the presence of a lactone seemed more 34 Table 3.1 Yield of compounds (1-5 and 7) isolated from dried M salicifolia plant parts. Plant Part 1+2 3 4 5 7 % % % % % stems 0.08 - - - - leaves - 0.02 0.25 0.43 - fruits - - - 0.92 0.60 flowers - - - 0.24 - 35 plausible, based on the triplet at 6 4.56 in the lHNMR spectrum. The peak at 6 119.14 (013) is representative of a terminal methylene carbon. It was concluded fi'om these data that the partial structure of 7 consisted of a five-membered lactone ring with an exocyclic double bond. Furthermore, based on the remaining spectral data, the lactone was found to be fused to a larger ten-membered ring. Elucidation of the structure of the remainder of the molecule was achieved by DEPT NMR experiments. Two CH carbons at 6 81.40 and 49.90 were assigned to C-6 and C-7, respectively. The C-6 carbon was deshielded, primarily due to the lactone ring and the adjacent olefinic moiety at C-4 and C- 5. Peaks appearing at 6 40.48 and 27.53 represented CH2 carbons at C-3 and C-2, respectively. Another CH and tertiary olefinic carbons were observed at 6 126.54 and 136.44, respectively, for C-1 and C-10. Differences in chemical shifts for these carbons can be explained by the same rationale used for C-4 and C-5 carbons. The peaks at 6 38.51 (C-9) and 25.69 (C-8) were confirmed to be CH2 carbons. Two peaks observed in DEPT at 6 16.84 (C-15) and 15.60 (C-14) which were assigned to methyl carbons. In the lHNMR spectrum, H-2', H-3, H-8/8', H-9/9' appeared as multiplets at 6 1.97-2.35. costunolide (7) 36 A doublet of a doublet at 6 2.45 (H-3') can be explained by the vicinal coupling. The proton at C-7 appeared as a multiplet at 6 2.55 by coupling with protons on C-6 and C-8. The H-2 proton appeared as a multiplet due to the C-1 and C-3 proton couplings. MS of the compound gave the molecular ion at m/z 232. HPLC analyses of the fruit extracts of M salicifolia led to the quantification of iso- methyl eugenol, costunolide and its related sesquiterpene lactone, parthenolide, compound 8 (Figure 3.1 and 3.2). Parthenolide in M salicifolia extracts was quantified and confirmed by HPLC analyses, using standard parthenolide. The ratio of costunolide to parthenolide was roughly 10:1, as indicated by HPLC quantification. Furthermore, this is the first report of parthenolide in M salicifolia. Parthenolide was isolated previously fromM grandiflora L. and M denudata (W iedhopf et al., 1973 and Funayama et al., 1995) Circular dichromism (CD) is the differential absorbtion of left and right circularly polarized light (cpl) by a non-racemic sample (Eliel, 1994). Cpl results fi'om the filtering of electromagnetic radiation, so that the tip of its electric vector moves in a helical fashion along whose axis radiation arises. In addition to being non-racemic, the sample must contain two types of chromophores to obtain a CD spectra. Chromophores which are inherently achiral by symmetry such as the carbonyl, carboxyl groups, or carbon-carbon double bonds contain at least one plane of symmetry when considered without substituents. Other chromophore types that need to be considered are those that are inherently chiral. Costunolide and parthenolide contain both classes of chromophores. The CD for 37 l 10.00 Minutes Figure 3.1 A representative HPLC chromatogram of M salicifolia fiuit extract showing iso-methyl eugenol (5), costunolide (7), and parthenolide (8) in absorption units, AU. Fruit extracts and pure costunolide, parthenolide, and iso-methyl eugenol were analyzed on a capcell pak C-18 (5 pm, 4.6 x 250 mm) column (Dychrome). The mobile phase ACN2H20 (80:20 v/v) was used under isocratic conditions at a flow rate of 0.5 mlmin". Injection volumes were 10 pl. Data were collected at 217 nm and 222 nm for costunolide and parthenolide, respectively. 38 40 weight mglg 7 8 compound Figure 3.2 Yield (mg°g") of iso-methyl eugenol (5), costunolide (7), and parthenolide (8) from M salicifolia fi'uit extracts. 39 costunolide and parthenolide indicate + and - absorptions for their respective structures (Figure 3.3 and 3.4, respectively). The source of these absorptions are the result of cumulative electronic or vibrational transitions associated with the chirotopic chromophores in parthenolide and costunolide, which causes right and left cpl to be absorbed differentially. Therefore, the absorbances A L at A R and AA = A L -AR are measures of CD. Furthermore, since the molar concentrations are known, AA = Aecl where c is the molar concentration (molliter’l) and I is the path length (cm), therefore, 6L- eR = A6. 6L and 6R are the molar absorption coeffecients for left and right cpl, respectively. Subsequently, A6 defines the sign of the CD. It can be concluded by the above relationships that costunolide, giving a predominately + CD spectrum, absorbs more left cpl, thus giving largely a positive molar elipticity as well. The molar elipticities of costunolide were 1.155E+02 mdeg (220 nm) and -7.815E+00 mdeg (262 nm) for the maximum and extremunr, respectively. Parthenolide, giving a predominately - CD spectrum, absorbs more right cpl, and therefore, has mainly a negative molar elipticity. The molar elipticity of parthenolide was at the extremum of -2.942E+01 mdeg (208 nm). A literature search revealed that CD data for costunolide and parthenolide are not reported. This is the first report of the CD for costunolide and parthenolide. 4O 1.200E+°2 I I I I I I I l I I I lfi' r I I I I I I I I If} I I I I I I I I I I I I I I I'I I I l—'1 I I I Ij-T“f—l_I-[_l'_I—T_l—1_I'—f‘ p .1 L CD [mdeg] ’ 1 _2.o°°E+011111111IIILLIIIIIIIIIIIIIILIIIIIIIIIIIIInrrrinxnrirnnririrnlnr 200 .0 "1. [ma] 325 .0 Figure 3.3 CD spectrum of costunolide, 7 (1 mg-ml") at 220 nm shows a molar elipticity, e at a maximum 1.155E+02 mdeg. At 262 nm an extremum at 7.005E+00 mdeg was observed for costunolide. 41 1.749E+oo I I I I I I I I I I I I I I I I I I IIIIIIIIIIIIITIIIIIIIIIIIjlfiIIIl’IIIIIIIIT I - -1 I. 1 b q l' . P d p -1 " .1 ” 1 . 1 CD , . [mdeg] q l’ .. P d L I p Q . d l. _3 OOOE+O1 1 1 1 1 J_1__.L L [LIMA L LLHM j I l A Ll I 1 L LJ I 1 1 l 1 I 1 1 1 L1 1 1 A a I 1 l L 1 l LkLi l a A ‘ ‘ I I 1 200 . O NL. [nml 325 . 0 Figure 3.4 CD spectrum of parthenolide, 8 (1 mg-ml") at 208 nm shows a molar elipticity, e at a extremum at 2.942E+01 mdeg. CHAPTER IV Biologically Active Extracts from Magnolia spp. and Compounds from Magnolia salicifolia Abstract Preliminary antimicrobial, nematicidal, mosquitocidal and anti-cancer bioassays were carried out for leaf, stem, flower, and fi'uit extracts of ornamental Magnolia salicifolia, Magnolia denudata ‘Yulan ’, Magnolia kobus var. stellata ‘Star ’, and Magnolia kobus var. Ioebneria ‘Merrill ’. Crude extracts were bioassayed against microbial plant and human pathogens Additional bioassays included mosquito larvae, Aedes aegptii; gypsy moth larvae, Lymantria dispar; forest tent moth larvae, Malacosoma distria; and nematodes, Panagrellus redivivus and Caenorhabditis elegans. Anti-cancer bioassays were conducted on mutant Saccharomyces cerevisae strains to determine the presence of topoisomerase I or H poisons. Antimicrobial or nematicidal activities were absent for crude extracts at 250ug per spot on solid agar plates and 250 ppm in solution assays, respectively. Many extracts possessed larvacidal activity at 250 ppm against mosquito, A. egwtii. Extracts of M kobus var. stellata and M denudata demonstrated significant growth inhibition on gypsy moth larvae, L. dispar. Minimum inhibitory concentrations (MIC) for mosquitocidal compounds from M 42 43 salicifolia were determined. Geranial and neral (l and 2), isolated fi'om the bark resulted in 100% mortality at 100 ppm in 24 h. T rans-anethole (3), isolated from the leaves exhibited 100% mortality at 20 ppm in 24 h. 1, 2-dirnethoxy-4-(2-propenyl) benzene (4) isolated fi'om leaves and 1, 2-dimethoxy-4-(l-propenyl) benzene (5) isolated fi'om leaves, fruits, and flowers, resulted in 100% mortality at 60 and 80 ppm, respectively, in 24 h. Costunolide (7), isolated from fruits, killed all A. aegjptii larvae at 15 ppm. Parthenolide (8), was not mosquitocidal when tested at 50 ppm. However, costunolide and parthenolide were found to be effective topoisomerase I poisons at 15 ug spot on solid agar plates lawned with test organism. Also, some of the extracts were active as top I and or top H poisons. 44 Introduction As mentioned earlier, many phytochemical studies carried out on Magnolia spp. examined predominately pharmacological activities. A wide variety of compounds and activities associated with them have been identified. The anti-cancer activity of the sesquiterpenes, costunolide and parthenolide has been well documented (Funayama et al., 1995 and Medhopf et al., 1973). The alkaloids, (+)-R-coclaurine and (+)-S-reticu1in, were found to possess negative inotropic effects (Kimura et al., 1989). Six lignans, derived from the buds of M biondii were identified as blood anticoagulents (Pan et al., 1987). Other notable compounds possessing neurotrophic activity, were characterized fi'om the root bark of M obovata Thunb. (Fukuyama et al., 1989, 1990, 1992). The three novel sesquiterpene neolignans, eudesobovatol A, clovane magnolol, and caryolanemagnolol were resposible for the neurotrophic activity. The phytochemical literature of Magnolia spp. relative to pesticidal activities, as mentioned before, are minimal. Yet, some studies have been done in this area. Phenolic constituents of M grandiflora L., namely magnolol, honokiol, and 3, 5'-diallyl-2'-hydroxy- 4-methoxy-biphenyl were found to have superior antibacterial activity against Bacillus subtilis, Staphylococcus aureus, and Mycobacterium smegatis when compared to streptomycin sulfate (Clarke et al., 1981). Additionally, magnolol and honokiol were found to be moderately active against Candida albicans, Saccharomyces cerevisiae, Aspergillus niger and strongly active against T richophyton mentagrophytes when compared to amphotericin B. 3, 5'-diallyl—2'-hydroxy-4-methoxy-biphenyl demonstrated milder activity against S. cerevisae and strong inhibition of growth against T. 45 mentagrophytes when compared to amphotericin B. Materials and Methods Plant Materials Stems of M salicifolia were collected fi'om trees growing on the campus of Michigan State University (East Lansing, Michigan) in February 1994. Stems, leaves, flowers and fruits of M salicifolia, M denudata, M kobus var. stellata, and M kobus var. Ioebneria were collected during the summer of 1994. Fresh weights were recorded for each of the plant parts. Large stems and fi'uits were cut into smaller pieces in order to facilitate eficient freeze drying. All plant parts were lyophilized at 5 °C under vacuum for at least 24 h or until the plant material was dried. Following the dry weight determination, the dried plant parts were milled and stored at -20°C until extraction. Table 4.1 reports percent dry matter for respective plant and plant parts. Extraction Separate extractions were conducted for each plant part. Dried ground plant materials were placed in an extraction column. All plant parts were extracted sequentially with hexane, EtOAc, and MeOH. The plant materials were soaked in solvent for at least 24 h with one exchange of fresh solvent after 12 h. Solvents were removed in vacuo and stored at -20°C until bioassay and purification. Table 4.2 indicates the yield of extract generated from respective Magnolia spp. plant parts. Bioassays Microbial bioassays. The rrricroorganisms assayed were: Hyphales; Botrytis spp., Aspergillusflavus (MSU strain), F usarium oxysporum (MSU-SM-1322), F. monilzforme 46 (MSU-SM-1323), Melanconiales; Gloesporum spp., Agonomycetales; Rhizoctonia spp. (MSU strain), Candida albicans (MSU strain), Bacteria; Staphylococcus epidermidis (ATCC 25923), Streptococcus aureus (MSU strain), Escherichia coli (ATCC 25922) and nematodes; Panagrellus redivivus Goody and Caenorhabditis elegans. Bacteria, yeast, and fimgi, respectively, were grown on Emmons, YMG, PDA agar media and maintained in the Bioactive Natural Products Laboratory (BNPL). These assays were conducted according to published methods (Nair et al., 1989). Samples were prepared by dissolving a known amount of the extract into an appropriate volume of DMSO such that the final concentration was 250ug-20ul". The microbial organisms were lawned onto petri dishes containing the appropriate agar media. Aliquots of 20 ul of test samples were spotted, respectively, onto lawned plates. Plates were allowed to incubate at 26°C for approximately 5-7 days. Thereafter, plates were analyzed for the presence of inhibition zones. None of the Magnolia spp. tested extracts possessed anti-microbial activity. Mosquitocidal bioassays. Fourth instar mosquito larvae, Aedes aegypti L (Culicidae), were reared in the BNPL from eggs. Eggs were obtained from Dr. Alexander Raikel, Department of Entomology, Michigan State University. At least 10 larvae were placed in 980 pl of degassed distilled H20 and 20 pl of DMSO containing test extracts or purified compounds. Resulting concentrations were 250 ppm for crude extracts. Initial concentrations for purified compounds were 100 ppm which then were diluted serially to determine MIC. The tube containing the control larvae received 20 ul of DMSO. Treatments and controls were left at room temperature. There were three replications per treatment. The number of dead larvae were recorded at 2, 4, 6, and 24 h intervals. The 47 tube containing the control larvae received 20 pl of DMSO alone, and the number of dead larvae were recorded as in the case of the test compounds (Nair et al., 1989 and Nitao et al., 1992). Many of the extracts had mosquitocidal activityiat 250 ppm. MICs for compounds 1+2, 3, 4, 5, and 6, were 100, 20, 60, 80, and 15 ppm, respectively. Compound 7 was not mosquitocidal when tested at 50 ppm. Nematicidal bioassays. These assays were performed on free-living nematodes Panagrellus redivivus and Caenorhabditis elegans fi'om cultures maintained in the BNPL. Test samples were prepared by dissolving a known weight of extract into DMSO such that the concentration was 12.5pg-2pl". An aliquot of 48 pl of media containing 30-50 nematodes at various developmental stages, were transferred aseptically from stock solutions into each sterile well (0.7 cm diameter X 1.0 cm deep) of a 96-well Corning flat- bottomed tissue culture plate. 2 pl of the test solution was added to the 48 pl thereby bringing the final concentration to 12.5pg-50pl'l or 250 ppm. Plates were covered and placed in a sealed container at high humidity. Plates were monitored for 2, 24, and 48 h intervals. Gwsy moth caterpillar bioassay. Gypsy caterpillar eggs were obtained from The Forest Pest Management Institute, Sault Ste. Marie, Ontario Canada via Dr. Daniel A. Herms of the Department of Entomology, Michigan State University. Eggs were stored under refiigeration until needed. Eggs were hatched in a growth chamber at ambient temperature with 15 h photoperiod. After 2-3 days, larvae were moving freely. Crude extracts and purified compounds were dissolved in DMSO to give a concentration of 1250 pg-25 pl‘l and 500 pg-25pl“, respectively. 25 pl of stock test solutions were mixed W 48 completely with 845 mg of dry diet mix (see Appendix XI for recipe). Agar held at 50°C, then was added until total diet weighed 5 g. The final concentration of test extracts and compounds were 250 and 100 ppm, respectively. The warm agar was mixed thoroughly with the dry portion of diet and then poured into 15 disposable polystyrene conicle cups (3.5 ml capacity, Sarstedt). The freshly poured diets were allowed to set 30 min in order for excess moisture to evaporate. Thereafter, larvae were added to the vials. For the gypsy moth assay, each vial received 1 larva. There were a total of 48 extracts with 15 replications per treatment. Afier six days, the larvae were weighed, these weights were averaged and then compared with the control weight average. Data were analyzed using Dunnet’s test, where all means are compared with a control. Anti-cancer bioassays. Saccharomyces cerevisae mutant cell cultures of IN 3 94, JN394t1, and JN3 94t2_, were supplied by Dr. John Nrtiss of St. Jude Children’s Hospital, Memphis, Tennessee and maintained in our laboratory. JN3 94 is hypersensitive to topoisomerase I poisons, while IN 3 94tl is isogenic to JN 3 94 except for the deletion of the top I gene, therefore showing a lack of response to topoisomerase I poisons. JN3 945,, carries top H gene which is resistant to topoisomerase H poisons, but responds to topoisomerase I poisons. These organisms were cultured in petri dishes containing 20 ml YPDA media. Plates were spotted with crude extracts, pure mosquitocides (1-6), parthenolide (7), and camptothecin and etoposide as positive controls. Plates were incubated at 26°C for one week. Camptothecin, a top I poison, is inhibitory to JN3 94 and JN394t2,,. Etoposide, a top II poison, is active against JN349 and JN394t1. MICs were determined for pure compounds by assaying a range of serial dilutions beginning at 250 49 p320 pl’l. Statistical Analyses of Gypsy Caterpillar Data The statistical analyses of the data concerning the gypsy moth larvae bioassay involved analysis of variance (ANOVA) for a completely randomized design (CRD) followed by Dunnett’s test for each experiment. The AN OVA were done to get a preliminary feel if the treatments had any efl‘ect on the test organism. Dunnett’s method was chosen because we are interested in determining whether the mean of the control group is significantly different than each of the means of the treatments. Due to the time involved in this bioassay, treatments were split into six separate experiments. Experimental treatments consisted of crude extracts and/or pure compounds. Geraniol and nerol (Aldrich Chemical, Milwaukee, Wisconsin) the respective corresponding alcohols of compounds 1 and 2 also were included in the study. Results and Discussion Leaves, stems, flowers and fiuits of M salicifolia, M kobus var. stellata ‘Star’, M denudata ‘Yulan’, and Mkobus var. Ioebneria were extracted sequentially with organic solvents. Dry weights for the various Magnolia spp. plant parts are shown in Table 4.1. Generally, the dry weights for the same plant part for different species were nearly equivalent. Extract weights for all plant parts fi'om each Magnolia spp. are shown in Table 4.2. The yields of MeOH extracts in all instances were larger than hexane and EtOAc extracts combined. Preliminary bioassays were performed on these extracts for their antifiingal, antibacterial, mosquitocidal, and nematicidal activities. Other preliminary bioassays were 50 Table 4.1 Dry ‘matter content for Magnolia spp. plant parts. Plant species % dry matter M salicifolia leaves 32. 9 stems 44.5 flowers 10.3 fruits 19.8 M kobus var. stellata leaves 41.8 stems 46.6 fruits 8.9 flowers 18.5 M denudata leaves 7.9 stems 38.9 fruits 9.9 flowers 21.0 M kobus var. ‘ Ioebneria leaves 28.7 stems W 48.9 fruits 11.6 flowers 20.0 Table 4.2 The weight of extracts from various Magnolia spp. plant parts. 51 Plant species Dry weight of Hexane extract EtOAc extract MeOH extract plant material E E E L— M salicifolia leaves 100.27 3.77 2.51 11.88 stems 100.00 4.07 0.57 11.80 flowers 50.15 1.55 0.50 15.42 fruits 120.27 12.66 3.16 23.68 M kobus var. stellata leaves 5.82 0.23 0.09 0.70 stems 100.26 2.98 1.76 5.67 flowers 20.00 1.96 0.18 4.59 fruits 2.83 0.07 0.04 0.27 M denudata leaves 10.49 0.58 0.27 0.94 stems 100.15 1.48 0.87 5.34 flowers 20.00 0.32 0.38 4.53 fruits 2.51 0.09 0.04 0.21 M kobus var. Ioebneria leaves 10.76 0.50 0.30 1.14 stems 100.00 3.34 0.89 8.16 flowers 6.34 0.22 0.18 2.35 fruits 4.33 0.25 0.30 0.64 52 conducted to determine the presence of growth inhibitors for gypsy moth larvae and the topoisomerase I and H poisons. Hexane (Table 4.3) and EtOAc extracts (Table 4.4) from all Magnolia spp. were found to be lethal to the larvae of A. egyptii at 250 ppm. The MeOH extracts displayed only minor mosquitocidal activity. All extracts of M kobus var. stellata and fruit extracts of M denudata demonstrated growth inhibition against L dispar (Table 4.8). No other extracts had shown any significant reduction in weight (Tables 4.5-4.7 and 4.9). Preliminary anti-cancer bioassays indicated the presence of active components in stem, leaf, and fruit hexane extracts of M salicifolia. Leaf EtOAc extracts of M salicifolia and hexane and EtOAc extracts of M stellata also were found to exhibit anticancer activity. It was concluded that compounds contained in these extracts were topoisomerase I poisons, since. activity was observed for IN 3 94 and JN3 94“. None of the extracts showed any antimicrobial or nematicidal activities. Mosquitocidal activities were evaluated for compounds 1-5, 7 and 8 (Table 4.10). Compound 6, myristicin, was reported earlier to be mosquitocidal at 25 ppm and firngicidal against Cladosporium cucumerinum at 20 pg (Marston et al., 1995). Myristicin also has exhibited toxicity towards fruit flies, D. melanogastar at an LD50 of 0.34 mg and against mice at an LD,0 200 mg-kg'l (Lichtenstein et al., 1974). In the same study, myristicin was found to act synergistically at 2% w/v with 0.1% w/v pyrethrins against house flies, Musca domestica. Synergisms also were observed when a mixture of myristicin and parathin, a carbaryl, were tested against fruit flies. Berenbaum and Neal (1985) explained that the mode of action of methylenedioxyphenyl compounds like myristicin, is through inhibition of the activity of mixed function oxidases (MFO), the 53 Table 4.3 Mortality of A. egyptii for hexane extracts of Magnolia plant parts at 250 ppm after 24 h. ‘ Percent Mortality“ Magnolia spp. leaves stems flowers fruits % % % % M salicifolia 100 100 100 100 M kobus var. 100 56.8:hl3.0 NA 100 stellata M denudata 100 85.6:h10.6 100 100 M kobus var. 100 16.7:17.0 100 100 Ioebneria NA=not active at 250 ppm ‘DMSO controls had 100% survival 54 Table 4.4 Mortality of A. egyptii for EtOAc extracts of Magnolia plant parts at 250 ppm after 24 h. Percent Mortality‘ Magnolia spp. leaves stems flowers fruits % “/- % % M salicifolia 40 $21.6 NA NA 100 M kobus var. 100 22.1 35.6 NA 100 stellata M denudata 87.9 $4.3 93.6 $4.5 96.7 $4.7 97.0 $4.3 M kobus var. Sl.7$ 13.2 23.3 $4.7 100 56.7 $12.5 Ioebneria NA=not active at 250 ppm ‘DMSO controls had 100% survival 55 Table 4.5 Average weights of L. dispar larvae after six days. Treatments for pure compounds were bioassayed at 100 ppm. DMSO was used for the control. treatment weight (mg) trans -anethole (3) 9.9, - citral (1 & 2) 11.8, - geraniol 11.3, - nerol 11.2, - control 9 *Data was analyzed using Dunnett's Test, P _<_ 0.01 + Significant reduction in weight - No significant reduction in weight 56 Table 4.6 Average weights of L. dispar larvae after six days. Treatments for crude Magnolia spp. extracts and pure compounds were bioassayed at 250 ppm and 100 ppm, respectively. treatment M salicifolia stem hexane extract stem EtOAc extract stem MeOH extract leaf hexane extract leaf EtOAc extract leaf MeOH extract flower hexane extract flower EtOAc extract flower MeOH extract fi'uit hexane extract fruit EtOAc extract fruit MeOH extract M kobus var. Ioebneria stern hexane extract stem EtOAc extract stem MeOH extract Compounds costunolide (7) iso-methyleugenol (5) methyleugenol (4) control weight (mg) 9, - 11, - 11.3, - 10.4, - 8.3, - 9, - 15.4, - 10, - 10.2, - 11.2, - 8.3, - 11.3, - 11.2, - 12.1, - 9.3, - 10.8, - 10.8, — 9.1, - 9.5 *Data was analyzed using Dunnett's Test, P 5 0.01 + Significant reduction in weight - No significant reduction in weight 57 Table 4.7 Average weights of L. dispar larvae after six days. Treatments for crude Magnolia spp. extracts and pure compounds were bioassayed at 250 ppm and 100 ppm, respectively. treatment weight (mg) M denudata flower hexane extract 9.2, - flower EtOAc extract 8.2, - flower MeOH extract 8, - stem hexane extract 7, - stern EtOAc extract 8.2, - stem MeOH extract 9.1, - leaf hexane extract 10.2, - leaf EtOAc extract 6.1, «- leaf MeOH extract 9.2, - M kobus var. Ioebneria fi'uit hexane extract 11.1, - fi'uit EtOAc extract 8.4, - fruit MeOH extract 9, - flower hexane extract 10.5, - flower EtOAc extract 9.2, - flower MeOH extract 10.4, - leaf hexane extract 9.6, - leaf EtOAc extract 8.3, - leaf MeOH extract 9.1, - control 9.6 *Data was analyzed using Dunnett's Test, P 5 0.01 + Significant reduction in weight - No significant reduction in weight 58 Table 4.8 Average weights of L. dispar larvae after six days. Treatments for crude Magnolia spp. extracts were bioassayed at 250 ppm. treatments weight (mg) M denudata fruit hexane extract 9.9, + fruit EtOAc extract 9.3, + fi'uit MeOH extract 10.1, + M kobus var. stellata leaf hexane extract 9.2, + leaf EtOAc extract 6.8, + leaf MeOH extract 7.3, + stem hexane extract 10.2, + stem EtOAc extract 8.2, + stem MeOH extract 7.9, + flower hexane extract 9.8, + flower EtOAc extract 10.3, + flower MeOH extract 9.6, + fiuit hexane extract 9.3, + fiuit EtOAc extract 14.1, + fruit MeOH extract 17.8, - control 20.1 *Data was analyzed using Dunnett's Test, P 5 0.01 + Significant reduction in weight - No significant reduction in weight 59 Table 4.9 Average weights of L. dispar larvae after six days. Treatments for pure compounds were bioassayed at 100 ppm. treatments weight (mg) myristicin (6) 14, - parthenolide (8) 15.7, - control 15.6 *Data was analyzed using Dunnett's Test, P < 0.01 + Significant reduction in weight - No significant reduction in weight 60 enzymes that metabolize many lipophilic xenobiotics in both vertebrates and invertebrates. MFOs are the primary means by which insects detoxify naturally occurring plant toxins. They found in their study that at as little as 0.1% myristicin in an artificial diet increased the toxicity of xanthotoxin to corn earworrn, Heliothis zea. Mode of action, of the other compounds (1-5, 7, 8) of the this study have not been determined. However, observations regarding lethality of compounds relative to structural differences may be inferred. Phenylpropanoids, 3, 4, 5, and 6 had activities ranging from 20 to 80 ppm concentrations. Among them, compound 3 having a methoxy group para to the benzylic double bound, was the most potent mosquitocide. Addition of another methoxy group in compound 5 gave reduced mosquitocidal activity. However, compound 4 was more active than 5, due to the presence of an allyl group instead of the benzylic double bond. Compound 6, more active than 4 or 5, may derive its toxicity for the reasons mentioned above. When applied in 20 mg quantities, compounds 1 and 2 collectively, were reported to have antifirngal activity against E. coli, B. subtilis, and Staphylococcus aureus (Onawani et al., 1984). Citral also has demonstrated phototoxicity to cabbage loopers, T richoplusia ni at concentrations of 300 ppm in an artificial diet (Green and Berenbaum, 1995) 61 Table 4.10 Mosquitocidal activities of pure compounds from M salicifolia on 4th instarA. emu? after 24 h. Compounds LDw0 (ppm)? 1 + 2 100 3 20 4 60 5 80 6 25" 7 15 8 NA “Reported earlier by Marston et al., 1995 NA=not active at 50 ppm ‘I’DMSO controls had 100% survival. 62 The introduction of allyl groups onto aromatic rings can increase antibacterial activity of some phenolics (Bae et al., 1986). However, extrapolation of this observation with that of the present data, may be spurious. Compound 7, the most active of all the compounds isolated, gave an MIC of 15 ppm. The exocyclic double bond and carbonyl group were demonstrated to be the most important factors responsible for the cytotoxicity among sesquiterpene lactones (Lee et al., 1971). Interestingly, parthenolide (8) did not show any mosquitocidal activity when assayed at 50 ppm. These data suggest that the presence of a double bond rather than an epoxide at carbons 4 and 5 in costunolide is required for mosquitocidal activity. However, information regarding the structural activity relationships with respect to insecticidal activity for these compounds is lacking. The F tests for the AN OVA of first and third gypsy moth bioassay indicated no overall significance among the mean weights. Furthermore, according to Dunnett’s method, none of the treatment mean weights differed significantly from the control mean weights (Tables 4.5 and 4.7). The F test for the AN OVA of second gypsy moth bioassay demonstrated an overall significance at P5001 among the mean weights. However, Dunnett’s test had shown that none of the treatment mean weights were significantly difl‘erent than that of the control mean weights (Table 4.6). The F test for the ANOVA of the fourth gypsy moth bioassay indicated an overall significance at P5 0.01 among the mean weights. Additional support came from Dunnett’s test where it was shown that 14 out of 16 treatment mean weights were significantly lower than the weight of the control mean weights (Table 4.8). The last gypsy moth bioassay showed no significant reduction in weight after six days when parthenolide and myristicin were tested at 100 ppm (Table 63 4.9). In conclusion, all treatment extracts of M kobus var. stellata and fruit extracts of M denudata significantly reduced the weights of the larvae when tested at 250 ppm. Compounds with mosquitocidal activities had little or no efl‘ect on gypsy moth larvae weight when at 100 ppm. Only compounds 6, 7 and 8 showed activity against JN3 94 and JN3 940,, indicating inhibitory activity against topoisomerase I (Table 4.11). Compounds 7 and 8 were equally active at 15 pg per spot, whereas 6 was only active at 250 pg per spot . Results for 7 and 8 support earlier findings expressed by Lee and coworkers (1971). Topoisomerase I and II are enzymes that change the DNA linkers by catalyzing a three- step process. This involves the cleavage of one (topoisomerase I) or both (topoisomerase H) strands of DNA, movement of a segment of DNA through this break, and then rescaling the DNA break (Stryer, 1988). The lack of antibacterial or antifungal activities of these extracts suggests that the phytochemistry of northern North American Magnolia spp. are quite different than those grown in the southern United States. 64 Table 4.11 Anti-cancer activity of citral (l and 2), trans-anethole (3), methyl eugenol (4), isomethyl eugenol (5), myristicin (6), costunolide (7), parthenolide (8), camptothecin and etoposide. Compounds pg/spot ‘ S. cerevisae JN394 JN394 t, JN394 t” 1 +2 250 - - - 3 250 - — - 4 250 - - - 5 250 - - - 6 250 +, 1.7 cm - +, 1.7 cm 7 15 +, l.1$0.1 cm - +, l.l$0.1 cm 8 15 +, 1.1$0.1 cm - +, 1.1$0.1 cm ‘camptothecin 25 +, 2.6$0.2 cm - +, 2.6$0.2 cm ‘etoposide 25 +, 2.5$0.1 cm +, 2.5 cm - + active, zone of inhibition diameter - not active ‘standards purchased from Sigma Chemical Company CHAPTERV Summary and Conclusion Leaves, stems, flowers, and fiuits of four Magnolia spp. were extracted and preliminary bioassays were performed on these extracts at 250 ppm concentrations to test for the presence of antifungal, antibacterial, mosquitocidal, and nematicidal activities. Bioassays also were conducted on all extracts on gypsy moth larvae, L. dispar to test for growth inhibition. Anti-cancer bioassays were evaluated for all crude extracts which utilized mutant S. cerevisae strains. All crude extracts were found to be mosquitocidal on the larvae of A. egwtii. Extracts of M kobus var. stellata and M denudata were found to significantly reduce the grth of L. dispar larvae. However, crude solvent extracts from these Magnolia spp. did not show any antimicrobial or nematicidal activities when tested on Hyphales; Bonytis spp., Aspergillus spp., F usarium oxysporum, E moniliforme, Melanconiales; Gloesporum spp., Agonomycetales; Rhizoctonia spp., Candida albicans, Bacteria; Staphylococcus spp., Streptococcus spp., Escherichia coli and nematodes; Panagrellus redivivus Goody and Caenorhabditis elegans at 250 ppm concentrations However, some crude extracts exhibited moderate anti-cancer activity. Through the process of bioassay directed fi'actionation six mosquitocidal compounds were isolated fi'om the various plant parts of M salicifolia; citral (1 and 2), trans-anethole (3), methyl eugenol (4), iso-methyl eugenol (5), and costunolide (7). 65 66 Myristicin (6) was detected by lHNMR along with anethole in leaf extracts of M salicifolia. HPLC analysis of the fruits of M salicifolia led to the identification and quantification of the sesquiterpene lactone, parthenolide. Costunolide and iso—methyl eugenol also were quantified in an HPLC analysis of the fi'uit extracts. Structures of compounds 1-7 were confirmed by lHNMR, l3CNMR, and MS experiments (Chapter HI). Mosquitocidal and anticancer activities were evaluated for compounds 1-8 (Chapter IV). The most potent mosquitocide was found to be the sesquiterpene lactone, costunolide. The MIC for this compound was 15 ppm. Parthenolide was not active against mosquito larvae when tested at 50 ppm. Citral, trans-anethole, methyl eugenol, iso-methyl eugenol, and myristicin had mosquitocidal activities at 100, 20, 60, 80, and 25 ppm, respectively. Costunolide and parthenolide displayed anti-cancer activity and their MICs were 15 pg. Myristicin also demonstrated anticancer activity at 250 pg. Mode of action of these compounds for their insecticidal activity has not been determined. However, observations regarding lethality of compounds relative to structural differences may be inferred. Phenylpropanoids, 3, 4, 5, and 6 had activity concentrations ranging from 20 to 80 ppm. Among them, compound 3, having a methoxy group para to the benzylic double bound, was the most potent mosquitocide. Addition of another methoxy group in compound 5 gave reduced mosquitocidal activity. However, compound 4 was more active than 5, due to the presence of an allyl group instead of the benzylic double bond. Compound 6 with a methylenedioxy functionality, was the second most potent mosquitocide. Its activity was quite similar to that of 3 and may be due to its allyl functionality. 67 The introduction of allyl groups onto aromatic rings increased antibacterial activity of some phenols and biphenols (Bae et al., 1986). However, extrapolation of this observation with that of the present data, requires additional research. Compound 7, the most active of all compounds isolated, had an MIC of 15 ppm. Exocyclic double bonds and carbonyl groups were the important factors responsible for cytotoxicity among sesquiterpene lactones (Lee et al., 1971). Parthenolide showed no mosquitocidal activity when bioassayed at 50 ppm. This data suggests that the presence of a double bond rather than an epoxide at carbons 4 and 5 in costunolide is required for mosquitocidal activity. However, information regarding the structure activity relationships with respect to insecticidal activity for these compounds is lacking. The work contained herein has yielded known compounds with new activities. These compounds were reported previously. Additionally, compound 7, parthenolide, has been reported for the first time in M salicifolia Similarly, this is the first report of spectra for CD for parthenolide and costunolide. Abda Bae Back Bere' Calla Cast: Cher Clarl BIBLIOGRAPHY Abdallah, O. M. (1993) Lignans in the flower buds of Magnolia saulangiana. Phytochemistry. 34, 1185-1187. Bae, K., Koo, S., and Seo, W. (1986) KoreanJ. Pharmacogn. 17, 85-90. Back, N. 1., Kim, H., Lee, Y. H., Park, J. D., Kang, K. S., and Kim, S. I. (1992) A new dehydrodieugenol from Magnolia officinalis. Planta Med. 58, 566-568. Berenbaum, M. and Neal, J. J. (1985) Synergism between myristicin and xanthotoxin, a naturally cooccurring plant toxicant. J. Chem. Ecol. 11, 1349-1358. Callaway, D. J. (1994) The world of Magnolias. Timber Press. Portland, Oregon. Castaneda-Acosta, J ., Cain, A. W., Fischer, N. H., and Knopf, F. C. (1995) Extraction of bioactive sesquiterpenelactones from Magnolia grandiflora using supercritical carbon dioxide and near-critical propane. J. Agric. Food Chem. 43, 63-68. Chen, C. C., Chen, H. T., Huang, Y. L., Chen, Y. P., and Hsu, H. Y. (1988) On the Ca++ antagonistic principles of the flower buds of Magnolia fargesii . Planta Med. 43 8-440. Clark, A M., El-Feraly, F. S., and Li, W.-S. (1981) Antimicrobial activity of phenolic constitiuents of Magnolia grandiflora L. J. Phann. Sci. 70, 951-952. 68 CUlblt Dans. iii-Fe: El-Fe Fran] Fujit Fujir F1110; 69 Culbreth, D. (1927) Materia medica and pharmacology, 7th ed. Philadelphia, PA: Lee and Febiger. Davis, T. L. (1981) PhD thesis; Chemistry of Magnolia grandiflora. Dept. of Medicinal Chemistry, College of Pharmacy, University of Florida, Gainsville FL 32610. El-Feraly, F. S., Chan, Y.-M., and Capiton, G. A. (1979) Isolation and characterization of peroxycostunolide (V erlotorin) and peroxyparthenolide from Magnolia grandiflora. Carbon-13 nuclear magnetic resonance spectroscopy of costunolide and related compounds. J.. Org. Chem. 44, 3952-3957. El-Feraly, F. S. And Li, W.-S. (1978) Phenolic constituents of Magnolia grandiflora L. seeds. Lloydia. 41, 442-449. Frankel, E. N., Waterhouse, A. L., and Teissedre, P. L. (1995) Principle phenolic phytochemicals in selected california wines and their antioxidant activity in inhibiting oxidation of human low-density lipids. J. Agric. Food Chem. 43, 890-894. Fujita, Y. (1955) The relation between Magnolia kobus DC. And M. salicifolia Maxim. Viewed from the constituents of essential oils. J. Jap. Bot. 30, 28-31. Fujita, M., Itokawa, H., and Sashida, Y. (1972) Honkiol, a new phenolic compound isolated fiom the bark of Magnolia obovata. Chem. Pharm. Bull. 20, 212-213. Fukuyama, Y., Otoshi, Y., Miyoshi, K., Hasegawa, N., Kan, Y., and Kodama, M. (1993) Structure of magnolianin, a novel trilignan possessing potent S-lipoxygenase inhibitory activity. Tetrahedron Lett. 34, 1051-1054. 70 Fukuyama, Y., Otoshi, Y., Miyoshi, K., Nakamura, K., Kodama, M., and Nagasawa, M., Hasegawa, T., Okazaki, H., and Sugawara, M. ( 1992) Neurotrophic sesquiterpene neolignans from Magnolia obovata: Structure and neurotrophic activity. Tetrahedron 48, 377-392. Fukuyama, Y., Otoshi, Y., and Kodama, M. (1990) Structure of clovanemagnolol, a novel neurotrophic sesquiterpene fi'om Magnolia obovata. Tetrahedron Lett. 31, 1477-1480. Fukuyama, Y., Otoshi, Y., and Kodama, M. (1989) Novel neurotrophic sesquiterpene neolignans fi'om Magnolia obovata. Tetrahedron Lett. 30, 5907-5910. Funayama, S., Adachi, M., Aoyagi, T., and Nozoe, S. (1995) Cytocidal principles of Magnolia denudata. J. Pharmacog. 33, 21-24. Haller, H. L., LaForge, F. B., and Sullivan, W. N. (1942) Some compounds related to sesamin: Their structures and their synergistic effect with pyrethrum insecticides. J. Org. Chem. 7, 185-188. Haller, H. L., McGovern, E. R., Goodhue, L. D., and Sullivan, W. N. (1942) The synergistic action of sesarrrin with pyrethrum insecticides. J. Org. Chem. 7, 183- 184. Iida, T., Noro, Y., and Ito, K. (1983) Magnostellin A and B, novel lignans from Magnolia stellata. Phytochemistry. 22, 211-213. Iida, T. and Ito, K. (1983) Four phenolic neolignans from Magnolia liliflora. Phytochemistry. 22, 763-766. ,5... 1w,li,li ' lacyno, J. l amir Foo Kahsawa, ] fart KMikado, lrol Kiwi. r MC IQ110a. A, net Kikuchi, 1 am new 11] 7 1 Ito, K., Iida, T., and Kobayashi, T. (1984) Guaiane sesquiterpenes from Magnolia watsonii. Phytochemistry. 23, 188-190. Jacobsson, U., Kumar, V., and Sarninathan, S. (1995) Sesquiterpene lactones fi'om Mchelia champaca. Phytochemistry. 39, 83 9-843. Jacyno, J. M. Montemurro, N., Bates, A D., and Cutler, H. G. (1991) Phytoxic and antimicrobial propertiesof cycolorenone from Magnolia grandiflora L. J. Agric. Food Chem. 39, 1166-1168. Kakisawa, H., Chen, Y. P., and H80. (1972) Lignans in the flower buds of Magnolia fargesii. Phytochemistry. 11, 2289-2293. Kamikado, T., Chang, C.-F., Murakoshi, S., Sakurai, A., and Tamura, S. (1975) Isolation and structure elucidation of growth inhibitors on silkworm larvae fi'om Magnolia kobus DC. Agr. Biol. Chem. 39, 833-836. Kawai, T., Kinoshita, K., Koyama, K., and Takahashi, K. (1994) Anti-emetic principles of Magnolia obovata and Zingiber oflicinalis Rhizome. Planta Med. 60, 17-20. Kijjoa, A., Pinto, M. M. M., Tantisewie, B., and Herz, W. (1989) A biphenyl type neolignan and a biphenyl ether fiom Magnolia henryi. Phytochemistry. 28, 1284-1286. Kikuchi, T., Kadota, S., Yanada, K., Tanaka, K., Watanabe, K., Yoshizaki, M., Yokoi, T., and Shingu, T. (1983) Isolation and structure of magnosalin and magnoshinin, new neolignans from Magnolia salicifolia Maxim. Chem. Pharm. Bull. 31, 1112-1114. 72 Kimura, 1., Chui, L., Fujitani, K., Kikuchi, T., and Kimura, M. (1989) Inotropic efl‘ectss of ($)-higenarnine and its chemically related components, (+)-R-coclaurine and (+)-S-reticuline, contained in the traditional Sino-Japanese medicines “Bushi” and “Shin-i” in isolated guinea pig papillary muscle. Jap. J. Pharmacol. 50, 75-78. Kimura, I., Kimura, M., Yoshizaki, Yanada, K., Kadota, S., and Kikuchi, T. (1983) Neuromuscular blocking action of alkaloids from Japanese crude drug “Shin-i” (Flos Magnoliae) in fi'og skeletal muscle. Planta Med. 48, 43-47. Kimura, M., Kobayashi, S., Luo, B., and Kimura I. (1990) Selective inhibition by magnosalicin and magnoshin, compounds fi'om “Shin-i” (Flos magnoliae), of adjuvant-induced angiogenesis and granuloma formation in the mouse pouch. Int. Arch. Allergy Appl. Immunol. 93, 365-370. Konoshima, T., Takasaki, M., Kozuka, Takuda, H., Nishino, H., Iwashima, Haruna, M., Ito, K., and Tanabe, M. (1990) Inhibitory effects on epstein-barr virus activation and anti-tumor promoting activities of neolignans from Magnolia oflicinalis. Planta Med. 56, 653. Lee, K. H., Huang, E. S., Piandosi, C., Pagano, J ., et al. (1971) Cancer Res. 31, 1649. Lichtenstein, E. P., Liang, T. T., Schulz, K. R., Schones, H. K., and Carter, G. T. (1974) Insecticidal ansd synergistic components isolated from dill plants. J. Agr. Food Chem. 22, 658-664. Likhitwitayawuid, K., et a1. (1988) Constituents of the bark of Michelia longrflora. J. Sci. Soc. Thailand. 14, 73-76. 73 Marston, A, Hostettmann, K., and Msonthi, J. D. (1995) Isolation of antifimgal and larvacidal constituents of Diplolophium buchanani by centrifugal partition chromatography. J. Nat. Prod 58, 128-130. Ming, C.W., Mayer, R, Zirnmerrnann, H., and Riicker, G. (1989) A non-oxidized melampolide and other gerrnacranolides from Aristolochia yunnanesis. Phytochemistry. 28, 3233-3234. Moerman, D. E. (1986) Medicinal Plants of Native America, Vol.1, 281-282. Ann Arbor, Michigan. Nrkaido, T., Ohmata, T., Kinoshita, T., Sankawa, U., Nishibe, S., and Hisada, S. Chem. Pharm. Bull. 29, 3586-3592. Miyazawa, M., Ishikawa, Y., Kasahara, H., Yamanaka, J.-I., and Kameoka, H. (1994) An insect growth inhibitory lignan from the flower buds of Magnolia fargesii. Phytochemistry. 35, 61 1-613 . Nitao, J. K., Johnson, K. S., Scriber, M., and Nair, M. (1992) Magnolia virginiana Neolignan compounds as chemical barriers to swallowtail butterfly host use. J. Chem. Ecol. 18, 1661-1671. Nitao, J. K., Nair, M., Thorogood, D. L., Johnson, K. S., and Scriber, M. (1991) i Bioactive neolignans from the leaves of Magnolia virginiana. Phytochemistry. 30, 2193-2195. Onawunmi, G. 0., Yisak, W., and Ogunlana, E. O. (1984) Antimicrobial constituents in the essential oil of Cymbopogon citratus. J. Ethnopharm. 12, 279-286. 74 Pan, J.-X., Hensens, O. D., Zink, D. L., Chang, M. N., Hwang, S.-B. (1987) Lignans with platelet activating factor antagonist activity fi'om Magnolia biondii. Phytochemistry. 26, 1377-1379. Rao, K. V. and Davis, T. L. (1982) Constituents of Magnolia grandiflora. III. Toxic principles of the wood. J. Nat. Prod. 45, 283-287. Rao, A S., Kelkar, G. R., and Bhattacharya, S. C. (1960) Terpenoids—XXI. The structure of costunolide, a new sesquiterpene lactone from costus root oil. Tetrahedron. 9, 275-283. Robles, M., Aregullin, M., West, J., and Rodriguez, E. (1995) Recent Studies on the Zoopharmacognosy, Pharmacology, and Neurotoxicology of Sesquiterpene Lactones. Panto Med. 61, 199-203. Stryer, L. (1988) Biochemistry, 3rd ed. pp 977-978. W. H. Freeman and Company, New York. Talapatra, B., Mukhopadhyay, P., and Dutta, L. N. (1975) Alkaloids from Magnolia campbellii and Magnolia mutabilis. Phytochemistry. 14, 589-590. Tang, W. and Eisenbrand, G., (1992) Chinese Drugs of Plant Origin. Chemistry, Pharmacology, and the use in Traditional and Modern Medicines, 63 9—646. Springer-Verlag Berlin Heidelberg. Tsurga, T., Ebizuka, Y., Nakajima, J ., Chun, Y.-T., Noguchi, H., Iitaka, Y., and Sankawa, U. (1984) Isolation of a new neolignan, magnsalicin, from Magnolia salicifolia. Tetrahedron Lett. 25, 4129-4132. WVoet, D. and Voet, J. G. (1990) Biochemistry. 658. John Wiley and Sons New York. 75 Watanabe, K. Goto, Y., and Yoshimoto, K. (1973) Central Depressaant Effects of the Extracts of Magnolia Cortex. Chem. Pharm. Bull. Vol. 21, No. 8, 1700-1708. Wiedhopf, R M., Young, M., Bianchi, E., and Cole, J. R (1973) Tumor inhibitory agent from Magnolia grandiflora (Magnoliaceae) I: Parthenolide. J. Pharmaceut. Sci. 62,345. Woodland, D. W. (1991) Contempory Plant Systematics. 145. Prentice Hall, New Jersey 07632. APPENDICES 76 APPENDIX I lHl‘lMR - citral (l and 2) Multan. l l Ilrfrilrfi PP“, 2 77 APPENDIX II 13CNMR - citral (l and 2) Tfllll[[IIIIllIlllllllIIIIIIIIIIIIIIIIIIIIlI—THIIIIIIIIIIIIIIIIIIIIIjIIIllTIIllrl—IrlllllrrIlllllIl[.IIl PPm 20 40 6O 80 100 120 140 160 180 200 78 APPENDIX III lHNMR - trans-anethole (3) 7i Jill I l l T I I I I I I {fir I I I I 1 I l I I r T I i I I I I I 1 I I l I U 7 I l I I I I I I . I I PPm 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 79 APPENDIX IV 1131““ - methyl eugenol (4) III II I I I I I I I I I I I I I I FI I I I I I I I I I I I I I F! I I I I Iil I I I I I I I PPm 3.5 4 5.0 6 6.5 80 APPENDIX V l"’CNMR - methyl eugenol (5) , WWW 81 APPENDIX VI lIEINMR - iso-methyl eugenol (5) in. I I I Ifi I I I I I I I I T I I I I I I ppm 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 I I I I 82 APPENDIX VII l"CNMR- iso-methyl eugenol (5) a3 IIIITIIrTIIIIIIIIfiIrTIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIDIIIIII PP“ 140 130 120 110 100 90 80 70 60 50 40 30 150 83 APPENDIX VIII 1HNMR - myristicin (6) a — 3‘. i '- N M I I 4 M i 84 APPENDIX VIIII lHNIVIR - costunolide (7) Wm JWMUL - i 85 APPENDIX X l“CNMLR - costunolide (7) an on 2. 8 8 2: o3 3H 8H 8H _~______._.__hkp_...____r.._pp.__pb.__pr_p__~._P~_____.»__._.»_F..__.__L_gpprL______~_._»_.p_r iii—11%.}; iii] J. ...1.fi!u-III~II 86 APPENDIX XI Gypsy moth catapillar diet Gypsy Moth, Lymanlria dispar dry diet ingredients: wheat germ 36 g casein 7.5 g . wesson salts 2.4 g sorbic acid 0.6 g methyl paraben 0.3 g vitamin mix (Hoflinan-LaRoche) 3.0 g HICHIGQN STnTE UNIV. LIBRARIES II”WWII“HIHIIHIIWIIHIIWIIIWIIIIWIIIWHI 31293014051969