INFORMATION TO USERS
This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.
The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.
1.The sign or “target” for pages apparently lacking from the document photographed is “ Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of “sectioning” the material has been followed. It is customary to begin filming at the upper left hand comer of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again beginning below the first row and continuing on until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.
University; Micrdnlms International 300 N. Zeeb Road Ann Arbor, Ml 46106 8300380
Wilton, John Howard
STUDIES OF NATURAL PRODUCTS FROM LIRIODENDRON TULIPIFERA, SIMMONDSIA CAUFORNICA AND HARDWOOD BARK COMPOST. I. NATURALLY-OCCURRING CYTOTOXIC AND ANTIFF.F.DANT SESQUITERPENE LACTONES OF LIRIODENDRON TULIPIFERA. II. ISOLATION, IDENTIFICATION AND ABSOLUTE STRUCTURE ELUCIDATION OF COMPOUNDS FROM SIMMONDSIA CALIFORNICA. III. ISOLATION AND STRUCTURE ELUCIDATION OF ANTIFUNGAL COMPOUNDS PRESENT IN HARDWOOD BARK COMPOST
The Ohio Stale University PH.D. 1982
University Microfilms International 300 N. Zeeb Read, Ann Arbor, MI 48106 PLEASE NOTE:
In alt cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .
1. Glossy photographs or pages ______2. Colored illustrations, paper or print _____ 3. Photographs with dark background _____ 4. Illustrations are poor copy ______5. Pages with black marks, not original copy ______6. Print shows through as there is text on both sides of page ______7. Indistinct, broken or small print on several pages
8. Print exceeds margin requirements ______9. Tightly bound copy with print lost in spine ______10. Computer printout pages with indistinct print _____ 11. Page(s) ______lacking when material received, and not available from school or author. 12. Page(s) ______seem to be missing in numbering only as text follows. 13. Two pages numbered ______. Text follows. 14. Curling and wrinkled pages ______
15. Other______
University Microfilms International STUDIES OF NATURAL PRODUCTS FROM LIRIODENDRON TULIPIFERA, SIMMONDSIA CALIFORNICA AND HARDWOOD BARK COMPOST.
I. NATURALLY-OCCURRING CYTOTOXIC AND ANTIFEEDANT SESQUITERPENE LACTONES OF LIRIODENDRON TULIPIFERA.
I I . ISOLATION, IDENTIFICATION AND ABSOLUTE STRUCTURE ELUCIDATION OF COMPOUNDS FROM SIMMONDSIA CALIFORNICA.
I I I . ISOLATION AND STRUCTURE ELUCIDATION OF ANTIFUNGAL COMPOUNDS PRESENT IN HARDWOOD BARK COMPOST.
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree of Doctor of Philosophy in the Graduate
School of The Ohio State U n iversity
By
John Howard W ilton, B.A *******
The Ohio State University
1982
Reading Committee: Approved by
Dr. Raymond W. Doskotch
Dr. Robert W. Brueggemeier
Dr. Duane D. M iller C ollege of Pharmacy Dr. Larry W. Robertson This work is dedicated to my w ife, Joanne, for her constant support and understanding throughout.
ii ACKNOWLEDGEMENTS
The author wishes to extend his sincere appreciation to his adviser, Dr. Raymond W. Doskotch, for his encouragement, excellent guidance and highly contagious enthusiasm.
The author also wishes to thank Dr. Jack L. Beal, Dr.
Harry A. Hoitink, Dr. Larry W. Robertson, and Dr, Jeffrey W.
Williams for their encouragement and support. Special
thanks are extended to Mr. John W. Fowble for his endless help and many valuable suggestions.
The author acknowledges financial support by the
Department of Health, Education and Welfare in the form of
the National Research Service Award.
The author also wishes to express his appreciation, and
best wishes to Mr. Timothy A. Driscoll, whose support, and
enthusiasm were endless during the completion of this work.
And last, but not least, I thank my loving wife, Joanne,
whose constant encouragement, understanding and sacrifices
have all helped to make this work a reality. Finally, I
thank my parents for their support and the opportunity to
obtain a college education.
ii i VITA
June 21,1952 Born - Buffalo, N.Y., U.S.A.
1974 B.A. (Chemistry - Major) Walsh College Canton, Ohio
1976 Graduate Teaching Associate Division of Pharmacognosy and Natural Products Chemistry College of Pharmacy Ohio State University
1981 Graduate Research Associate National Research Service Award College of Pharmacy Ohio State University
PUBLICATIONS
"Tulirinol, an Antifeedant Sesquiterpene Lactone for ,’ e Gypsy Moth Larvae from Liriodendron t u lip ife r a ." J. Org. Chem. , 1980, 45, 1441.
FIELDS OF STUDY
Major Field: Pharmacognosy and Natural Products Chemistry
iv TABLE OF CONTENTS
PAGE ACKNOWLEDGEMENTS...... i i i
VITA...... iv
LIST OF TABLES...... v ii
LIST OF FIGURES...... xi
GENERAL INTRODUCTION...... 1
PART I: NATURALLY-OCCURRING CYTOTOXIC AND ANTIFEEDANT SESQUITERPENE LACTONES OF LIRIODENDRON TULIPIFERA...... 2
Chapter 1: T u lirin ol, an Antifeedant Sesqui terpene Lactone for the Gypsy Moth Larvae from Lirjodendron tulipifera L ...... 3
Discussion and Results ...... U
Exper im ental ...... 8
Chapter 2: Sesquiterpenes of Lirjodendron tulipifera. Isolation and Characterization of Six Sesqui- terpene Lactones from the Leaves and Root Bark Possessing Cytotoxic and/or Antifeedant Activity 2U
Discussion and Results ...... 30
Ex per im ental ...... 51
Chapter 3^ Isolation and Id en tification of Novel Cyclization Products of Lipiferolide, the Major Sesquiterpene of Lirjodendron tu lip ife r a ...... 75
Discussion and Results ...... 77
Ex per im ental ...... 110
v LIST OF REFERENCES...... 155
PART II: ISOLATION, IDENTIFICATION AND ABSOLUTE STRUCTURE ELUCIDATION OF COMPOUNDS FROM SIMMONDSIA CALIFORNICA...... 160
Chapter 1: The Isolation and Structural Determination of Three New Cyanoglycosides and Other Constituents from Simmondsia c a lifo r n ic a ...... 161
Discussion and Results ...... 164
Experimental ...... 205 Chapter 2: Determination of the Absolute Stereochemistry of Simmondsin and Related Cyanoglycosides...... 269
Discussion and Re su its ...... 271
Experimental ...... 292
LIST OF REFERENCES...... 318
PART III: ISOLATION AND STRUCTURE ELUCIDATION OF ANTIF UNGAL COMPOUNDS PRESENT IN HARDWOOD BARK COMPOST...... 321
Introduction ...... 322
Discussion and Results ...... 323
Experimental ...... 3^8
LIST OF REFERENCES...... 362
APPENDIX...... 364
v i LIST OF TABLES
TABLE Page
1 : 1 - 1 NMR Spectra of Tulirinol and Derivatives.... 6
1 : 1-2 Torsion Angles (deg) for Tulirinol Acetate 7
1:1-3 Lactone Ring Torsion Angles (deg) in Rep resentative Sesquiterpene Lactones ...... 7
1: 1-4 Tulirinol Acetate Crystal Data ...... 9
1:1-5 Final Atomic Coordinates for Tulirinol Acetate ...... 9
1 : 2-1 Sesquiterpene Lactones of Lirjodendron
tulipifera: Source, Activity and Class ...... 27
1 : 2-2 Sesquiterpene Lactones of Lirjodendron
tuliplfera: Physical Constants and
Lactone Stereochemistry ...... 28
1:2-3 Physical Constants for a -L ir iodenolide (I_) ...... 34 1:2-4 Physical Constants for £-Lir iodenol ide (II_).... 35
1 :2-5 Physical Constants for 11,13-Dehydro-
lanuginolide (III) ...... 37
1 : 2 - 6 NMR, Double Irradiation and NOE
Difference Results (300MHz; dg-Acetone)
for /3-Cyclol ipi ferol ide (XIII) ...... 44
vii NMR and Double Irradiation ResultsC300MHz ,
CDCl^) for fl-Cyclolipiferolide(XIII) ...... 45 NMR Data for the [6:2:0] Bicyclodecane
Compounds (300MHz, CDCl^, TMS) ...... 84 13C NMR Data for the [6:2:0]
Bicyclodecane Compounds ...... 88
NMR Data for BF^* Et20 Cyclization
Products (300MHz, CDCl^, TMS)...... 101
^ 3C NMR Data for BF3*Et20 Cyclization
Products ...... 105 Relative Percentages of Products Obtained from the Cyclization of Lipiferolide ...... 109 and ^3C NMR Data for Simmondsia
Glycosides IA-ID ...... 16?
^ 3C NMR Data for 1 -0-Methyl-/3-D-
GlucopyranosideCITA) ...... 171 1H (300MHz, CDC13, IMS) and 13C NMR
(20MHz, CDC13, TMS) Data for Permethyl- s immondsin ( I_I) ...... 175 NMR Data for the Peracetylated
Glycosides IIIA-D ...... 177
Double Irradiation Experiments for
6-0-Demethylsimmondsin Hexaacetate(IIIC) ...... 179 ^ 3C Data for Acetylated Glycosides(IIIA-D)... 180
v ii i 11:1-7 SFORD Results for Simmondsin
PentaacetateC 1IIA) ...... 181
11:1-8 NMR Data for Simmondsin Aglycones IVA-E 188
11:1-9 Double Irradiation Experiments for
Simmondsin Aglycone(IVA) ...... 189
11:1-10 13C NMR Data for Aglycones IVA-E ...... 193
11:1-11 ^ 3C NMR Data for Simmondsin Aglycone( IVA) ...... 194
11:1-12 13C NMR Data for Aglycone IVD ...... 195
11:1-13 Physical Data for the 0-Methyl Ethers of
Inositol and Myoinositol ...... 203
11:2-1 and "* 3C NMR Resonances for
Permethyl Dasycarponin (III) ...... 273 11:2-2 and ^ 3C NMR Resonances for
Dimethyl Dasycarponilide(_IV_) ...... 274
11:2-3 A Comparison of the Physical Properties
of Simmondsin Pentaacetate(VII) and
Isosimmondsin Pentaacetate(VIII) ...... 278
11:2-4 1H NMR (300MHz, CDC13) Data for
Simmondsin Pentaacetate(VII) and
Isosimmomdsin Pentaacetate(V III) ...... 279
11:2-5 13C NMR (20MHz, CDC13) Data for
Simmondsin Pentaacetate(VII) and
Isosimmondsin Fentaacetate(VIII) ...... 280
11:2-6 A Comparison of the Physical Properties
of Simmondsin Agl ycone(I^) and
ix Isosimmondsin Aglycone(DC) ...... 282
11:2-7 NMR (300MHz, d^-acetone) Data for Simmondsin Aglycone(I^)
and Isosimmondsin Aglycone(^X) ...... 283
11: 2-8 1H NMR Data (90MHz, CDC13> for Simmondsin Aglycone Benzoate(X) ...... 286
I I I - 1 Bioassay Results of Crude Bark Compost
Partition Fractions ...... 326
III-2 Spore Germination Assay Results for the
90% MeOH Partition Fractions ...... 333
III-3 Weights and Spore Germination Results for the P recipitates from the Hexane:Chloroform
Solubles (Fraction A )...... 333
III-4 Physical Constants for Compound A ...... 334
III-5 Physical Constants for Compound B ...... 340 III-6 Physical Constants for Cimpound C ...... 343
x LIST OF FIGURES
FIGURE Page 1:1-1 ORTEP Stereoview of Tulirinol Acetate ...... 5
1:1-2 IR Spectrum (CHCl^) of T u lir in o l(l) ...... 10
1:1-3 CD Spectrum (MeOH) of T u lir in o l(l) ...... 11
1:1-4 NMR Spectrum (Acetone-d^) T u lir in o l(l) ...... 12
1:1-5 MS Spectrum (El) of Tulirinol(l) ...... 13
1:1-6 IR Spectrum (CHCl^) of Tulirinol Acetate(2).... 14
I : 1-7 CD Spectrum (MeOH) of Tulirinol A cetate(2) 15
1:1-8 NMR Spectrum (CDCl^) of Tulirinol Acetate(2) 16
1:1-9 MS Spectrum (El) of Tulirinol A cetate(2) ...... 17
1:1-10 IR Spectrum (CH2C12) of Deacetyltulirinol(5)... 18 1:1-11 NMR Spectrum (Acetone-d^) of Deacetyl-
tulirinol(5) ...... 19
1:1-12 MS Spectrum (El) of Deacetyltulirinol(5) ...... 20
1:1-13 IR Spectrum (CHCl^) of T u lirin ol-a- Phenylbutyrate( 6 ) ...... 21
1:1-14 "’H NMR Spectrum (CDCl^) of T u lirin ol-a-
Phenylbutyrate( 6 ) ...... 22
1:1-15 MS Spectrum (El) of Tulirinol-a-
Phenylbutyrate( 6 ) ...... 23
x i Sesquiterpene Lactones Found in
Liriodendron tulipi fera L ...... 26
Ii* tulipifera Rootbark Extraction and
Partitioning Scheme ...... 32
Chromatographic Isolation of
a-, /?-, and y- Lir iodenol ide ...... 33 Epoxidation of Germacranolides ...... 38
Biogenetic Interrelationships of
L. tulipifera Sesquiterpene Lactones ...... 50
IR Spectrum CCHCl^) of a-Liriodenolide(I^) ...... 59 NMR Spectrum (CDCl^) of a-Liriodenolide( 1^). . 60
IR Spectrum (CHCl^) of 3-Liriodenolide( I_I) ...... 61
NMR Spectrum (CDCl^) of fi-L iriodenolide(II? . 62
IR Spectrum CCHCl^) of
11,13-Dehydrolanuginolide(III 1 ...... 63 ^HMR Spectrum (CDCl^) of
11 * 13-Dehydrolanuginolide(III) ...... 6H
^ 3c NMR Spectrum CCDCl^) of
11, 1 3-Dehydrolanuginol ide(I_I_I) ...... 65
CD Spectrum (MeOH) of
II, 13-Dehydrolanuginolide(III) ...... 66
IR Spectrum (CHCl^) of
Dihydrochrysanolide(IXA) ...... 67 ^HMR Spectrum (CDCl^) of
Dihydrochrysanolide(IXA) ...... 68
xii 13C NMR Spectrum (CDCl^) of
Dihydrochrysanolide(IXA) ...... 69 CD Spectrum (MeOH) of
Dihydrochrysanolide(IXA) ...... 70
IR Spectrum (CHCl^) of
fl-Cyclolipiferolide(XIII) ...... 71 1H NMR Spectrum CCDCl^) of
g-Cyclolipiferolide(XIII) ...... 72
^3c NMR Spectrum (CDCl^) of
|?-Cyclolipiferolide(XIII) ...... 73 CD Spectrum (MeOH) of
fl-Cyclolipiferolide(XIII) ...... 74
Cyclization Reactions of Known
Sesquiterpene Lactones ...... 78
Cyclization Products of Syn and Anti
Conformations ...... 80
Acid Catalyzed Cyclization Products from
Lipi ferolide ...... 81
Formation of [2:6:0] Bicyclodecane Systems 83
NMR Spectrum (CDCl^) of L ip ifer o lid e .. . 85
NOE Difference Results for Compound X ...... 89
Products Formed on Epoxide Ring Openings.. 92
Variable Temperature Studies for
Compound XII ...... 93 Guaianolide Formation from Lipiferolide... 98
xiii 1:3-10 Transformation of Lipiferolide to
Simsiolide Acetate ...... 108
1:3-11 13C NMR Spectrum (CDC13) of L ip ife r o lid e ...... 123
1:3-12 IR Spectrum (CHCl^) of Compound VTI ...... 124
1:3-13 NMR Spectrum (CDCl^) of Compound V II ...... 125
1:3-14 13C NMR Spectrum (CDCl^) of Compound VII ...... 126
1:3-15 IR Spectrum (CHCl^) of Compound V III ...... 127 1:3-16 NMR Spectrum (dg-acetone) of
Compound V III ...... 128
1:3-17 13C NMR Spectrum (CDC13 > of Compound V III ...... 129
1:3-18 IR Spectrum (CHC13) of Compound JX ...... 130
1:3-19 1H NMR Spectrum (CDC13 ) of Compound IX ...... 131
1:3-20 13C NMR Spectrum (CDC13) of Compound IX ...... 132
1:3-21 IR Spectrum (CHC13) 0f Compound X^...... 133
1:3-22 1H NMR Spectrum (CDC13 ) of Compound X ...... 134
1:3-23 13C NMR Spectrum (CDCI3 ) of Compound X ...... 135
1:3-24 IR Spectrum (CHClj) of Compound XI_ ...... 136
1:3-25 1H NMR Spectrum (C^d^; 74°) of Compound )LI 137
1:3-26 ’ 3C NMR Spectrum (OgDg; 70°) of Compound X_I.. . . 138
1:3-27 IR Spectrum (CHC13) of Compound XII ...... 139
1:3-28 NMR Spectrum (dg-acetone) of Compound XII... 140
1:3-29 ^ cmR Spectrum (CDC13 ) of Compound XII ...... 141
1:3-30 IR Spectrum (CHC13> of Compound X III ...... 142
1:3-31 ^ NMR Spectrum (CHC13) of Compound X III ...... 143
1:3-32 CD Spectrum (MeOH) of Compound XIV ...... 144
xiv 1:3-33 IR Spectrum (CHCl^) of Compound XIV ...... 145
1 : 3-31* 1H NMR Spectrum (d6-acetone) of Compound X II-• • 146
1:3-35 IR Spectrum (CHCl^) of Compound XV ...... 147
1:3-36 NMR Spectrum (CDC13 ) of Compound XV ...... 148
1:3-37 ^ NMR Spectrum (d^-acetone) of Compound XV^. . . 149
1:3-38 IR Spectrum (CHCX^) of Compound XVI ...... 150
1:3-39 1H NMR Spectrum (CDC13 ) of Compound XVI ...... 151
1:3-40 IR Spectrum (CHC13 ) of Compound XVII ...... 152
1:3-41 1H NMR Spectrum (CDC13 ) of Compound XVII ...... 153
1:3-42 13C NMR Spectrum (CHC13 ) of Compound XVII ...... 154
11:1-1 Droplet Counter-Current Chromatographic
Separation of Simmondsin Gylco3ides ...... 165
11:1-2 Analysis of H-7ax Multiplicity in the
NMR Spectrum...... 186
11:1-3 Analysis of H-7eq Multiplicity in the
1H NMR Spectrum...... 187
11:1-4 Nuclear Overhauser Experiment of
Aglycone IVA: C-5 Methoxy ...... 191
11:1-5 Nuclear Overhauser Experiment of
Aglycone IVA: C-6 Methoxy ...... 192
11:1-6 Graphical Representation of Aglycone
Carbon Absorptions ...... 197
11:1-7 Graphical Representation of Carbon Absorptions
in the Peracetylated Derivatives ...... 199
11:1-8 Graphical Representation of the
xv Glycoside Carbon Absorptions...... 200
Summary of Simmondsia Glycosides and their Derivatives ...... 204
IR Spectrum (KBr) of Simmondsin(_IA) ...... 225
1H NMR Spectrum (D2o) of Simmondsin( IA). 226
^3c NMR Spectrum (D2o) of Simmondsin( IA) 227 IR Spectrum (KBr) of
5-0-Demethylsimmondsin( 113) ...... 228
NMR Spectrum (D20) of
5-0-Demethylsimmondsin(LB) ...... 229 1^c NMR Spectrum (D20) of
5-0-Demethylsimmondsin( IjJ)...... 230
IR Spectrum (KBr) of
6-0-Demethylsimmondsin(2£ ) ...... 231 NMR Spectrum (D20) of
6-0-Demethylsimmondsin(I£ )...... 232
3c NMR Spectrum (D20) of
6-0-Demethylsimmondsin(^£)...... 233 IR Spectrum (KBr) of
5, 6-Di-0-demethylsimmond sin (_ID) ...... 2 3 4
1H NMR Spectrum (D2o) of
5 , 6-Di-O-demethylsimmondsin( H)) ...... 235
13c NMR Spectrum (D20) of
5, 6-Di-Q-demethylsimmond sin (_I_D)...... 236
IR Spectrum (CHC1^) of
xv i Permethylsimmondsin( I_I) ...... 237
11:1-23 1H NMR Spectrum (CDC13) of
Permethylsimmondsin(1^) ...... 238
11:1-24 13C NMR Spectrum (CDC13 ) of
Permethylsimmondsin( IJ^)...... 239
11:1-25 IR Spectrum (CHC13 ) of Simmondsin Pen taaceta te (III A) ...... 240
11:1-26 NMR Spectrum (CDC13) of
Simmondsin Pentaacetate( IIIA) ...... 241
11:1-27 13C NMR Spectrum (CDC13 ) of
Simmondsin Pentaacetate(IIIA) ...... 242
11:1-28 IR Spectrum (CHC13 ) of
5-0-Demethylsimmondsin HexaacetateC H i) ...... 243
11:1-29 NMR Spectrum (CDC13) of
5-0-Demethylsimmondsin Hexaacetate( U^) ...... 244
11:1-30 13C NMR Spectrum (CDC13 ) of
5-0-Demethylsimmondsin HexaacetateC HJ) ...... 245
11:1-31 IR Spectrum (CHC13) of
6-0-Demethylsimmondsin Hexaacetate(IIIC) ...... 246
11:1-32 1H NMR Spectrum (CDClj) of
6-0-Demethylsimmondsin Hexaacetate(IIIC) ...... 247
11:1-33 13C NMR Spectrum (CDC13 ) of
6-0-Demethylsimmondsin Hexaacetate(IIIC) ...... 248
11:1-34 IR Spectrum (CHC13) of 5,6-Di-O-
demethylsimmondsin Heptaacetate(IIID) ...... 249
xv ii IM -35 1H NMR Spectrum (CDCl^) of 5,6-Di-O-
demethylsimmondsin Heptaacetate(HID) ...... 250
1:1-36 NMR Spectrum CCDCl^) of 5 , 6-Di-O-
demethylsimmondsin Heptaacetate(I IIP) ...... 251
1:1-37 IR Spectrum (CHCl^) of
Simmondsin Aglycone (IVA ) ...... 252
1:1-38 NMR Spectrum (d^-acetone) of
Simmondsin Agl ycone (IVA ) ...... 253
1:1-39 13C NMR Spectrum (d^-acetone) of
Simmondsin AglyconeC IVA ) ...... 254
I : 1-40 IR Spectrum (KBr) of
5-0-Demethylsimmondsin AglyconeC IVB) ...... 255
1:1-41 NMR Spectrum (d^-acetone) of 5-0-Demethylsimmondsin Aglycone( IVB) ...... 256
1:1-42 13C NMR Spectrum (d^-acetone) of
5-0-Demethylsimmondsin Aglycone( IVB) ...... 257
1:1-43 IR (KBr) Spectrum of
6-0-Demethylsimmondsin AglyconeC IVC) ...... 258
1:1-44 NMR Spectrum (dg-acetone) of
6-0-Demethylsimmondsin AglyconeC IVC) ...... 259
1:1-45 13q NMR Spectrum (dg_acetone) of
6-0-Demethylsimmondsin AglyconeC IVC) ...... 260
I: 1-46 IR Spectrum (KBr) of
5 ,6-Di-O-demethylsimmondsin AglyconeCIVD) ...... 261
1:1-47 1H NMR Spectrum (d^-acetone) of
xv iii 5 ,6-Di-O-demethylsimmondsin AglyconeCIVD) 262
13c NMR Spectrum (dg-acetone) of
5» 6-Di-0-demethylsimmondsin AglyconeCIVD) 263
IR Spectrum (CHCl^) of Methylsimmondsin
AglyconeCIVE) ...... 264
^H NMR Spectrum Cd^-acetone) of
Methylsimmondsin AglyconeCIVE) ...... 265
!3C NMR Spectrum (d^-acetone) of
Methylsimmondsin AglyconeCIVE) ...... 266
IR Spectrum (KBr) of (+ )-P in ito l(VIA) . .. . 267
1H NMR Spectrum CD2O) of
C+)-PinitolCIVA) ...... 268
Conversion of Simmondsin and Dasycarponin to Dimethyl Dasycarponilide ...... 270
Isomerizations of Simmondsin(V.) and
Simmondsin PentaacetateCVII) ...... 276
UV and CD Spectra for Simmondsin
Aglycone Benzoate(XI_) ...... 287
CD Spectra of Simmondsin Aglycone(I^) and Isosimmondsin Aglycone(_I)t) ...... 289
IR Spectrum CCHCl^) of Permethyl dasycar ponin ...... , ...... 302
1H NMR Spectrum (CDCl^) of Permethyl dasycarpon in ...... 303
^3c NMR Spectrum (CDCl^) of Permethyl
xix dasycarponin ...... 304
11:2-8 IR Spectrum (CHCl^) of Dimethyl
dasycarponillde ...... 308
11:2-9 NMR Spectrum (dg-acetone) of Dimethyl
dasycarponillde ...... 306
11:2-10 13C NMR Spectrum (dg_acetone) of Dimethyl
dasycarponillde ...... 307
11:2-11 CD Spectrum (MeOH) of Dimethyl
dasycarponillde...... 308
11:2-12 IR Spectrum (CHCl^) of Isosimmondsin
Pentaacetate ...... 309
11:2-13 1H NMR Spectrum (CDC13 ) of
Isosimmondsin Pentaacetate ...... 310
11:2-14 '3c NMR Spectrum (CDCl^ of
Isosimmondsin Pentaacetate ...... 311
11:2-15 CD Spectrum (MeOH) of Isosimmondsin
Pentaacetate ...... 312
11:2-16 IR Spectrum (CHCl^) of Isosimmondsin
Aglycone ...... 313
11:2-17 1H NMR Spectrum (dg_acetone) of Isosimmondsin Aglycone...... 314
11:2-18 NMR Spectrum (dg_acetone) of
Isosimmondsin Aglycone ...... 315
11:2-19 IR Spectrum (CHCl^) of Simmondsin
Aglycone Benzoate ...... 316
xx 1H NMR Spectrum (CDCl^) of
Simmondsin Aglycone Benzoate ...... 317
Extraction and Partitioning Scheme for
Composted Hardwood Bark ...... 325
Chromatographic Fractionation of Hexane
Soluble Materials ...... 330
Partitioning and Chromatographic
Fractionation of 90% MeOH Solubles ...... 332
^H NMR Spectrum (CDCl^) of Myristic Acid 338
Proposed Biosynthetic Pathway for
C.jg Cutinic Acids...... 346
IR Spectrum (CHCl^) for Compound A...... 354
NMR (CDCl^) Spectrum for Compound A.. 355
^NMR Spectrum (CDCl^) for Compound A. 356
IR Spectrum (CHCl^) for Compound B .....'. 357
NMR (CDCl^) Spectrum for Compound B .. 358
^C NMR Spectrum (CDCl^) for Compound B. 359 IR Spectrum (CHCl^) for Compound C ...... 360
NMR (CDCl^) Spectrum for Compound C.. 361
xxi GENERAL INTRODUCTION
The work cited herein describes the isolation, structure elucidation and screening of a variety of substances possessing different activities. It consists of three parts with subsequent chapters therein. Part one pertains to the sesquiterpene lactone constituents of Lir jodendron tulipifera which possess both antineoplastic and Gypsy Moth feeding-deterrent activity. Part two deals with the isolation of new cyanoglycosides from Simmondsia californica and their absolute stereochemical determination. Part three describes the isolation of antifungal compounds from hardwood bark compost which are active against the agricultural pest, Phytophthora.
The active constituents were isolated from the plants by extraction with 95% ethanol, solvent-partitioning of the resulting residues to obtain initial fractionation, followed by a variety of chromatographic techniques to obtain pure compounds. Once obtained, a detailed an alytical study was undertaken, using a variety of chemical and spectroscopic techniques to determine their structures and absolute stereochemistry.
1 PART I: NATURALLY-OCCURRING CYTOTOXIC AND ANTIFEEDANT
SESQUITERPENE LACTONES OF LIRIODENDRON TULIPIFERA.
2 Chapter 1: T ulirinol, an Antifeedant Sesquiterpene Lactone
for the Gypsy Moth Larvae from Llrlodendron tulipifera.
A study for the structural elucidation of the sesquiterpene lactone, tulirinol, has been published in the
Journal of Organic Chemistry, Volume 45, page 1441, 1980. A reprint of this publication appears hereafter.
3 [Reprinted front (hi Journal of Orfaaie Chonmtr?* 4.1441 (19601.1 Copyright ♦ I960 qr in* A ntttku Choinfcal Society and reprinted by ptrmteuon of the copyright o*iwr<
Tulirinol, an Antifeedant Sesquiterpene Lactone for the Gypsy Moth Larvae from Liriodendron tulipifera
Raymond W Doekotch,* Edward H. Fairchild,1* Chin-Teh Huang,'* and John H. Wilton Diuidon of Phorm a cofn 09y and.Vaturat Product! Chemiatry, ColIff* of Pharmacy, 77m Ohio Stott L1 mot ruty. CoJumfaua. Ohio 43210 Mark A Be no and Gary G. Chriatoph* Department of Chemiatry. Collett of Mathematicaland Phyaicot Sciences, 77m Ohio State Univtrut>, Mtinigi, O h io 4 3 2 1 0 Sactiotd September 7,1979
The iaoiatian and structure determination at tulirinol (1), an antifeedant for the typay moth larvae from the laavea of fAriodtndrtmtu lip ifera L., a n reported- Tulirinol ie the first recofniaed traru-4,ct4-9-cydod*cadien( sesquitetpans, detail* of which an given for it* characterize Don fay spectral met hod* including X-ray crysiallugraphy. 111* sbeohite *tenoch*mi*tiy was determined fay the Homo partial naolutioo method. Tatridin A wa* identified a* dewcotyitulirinol IS).
The kora of Liriodendron tulipifera L (Magnoliaceae), biological activity at each stage of purification .1 The commonly known u the tulip poplar, gave an ethanolic molecular formula, C^H-tO*, was established by elemental extract that inhibited the feeding of gypay moth larvae. analysis and outs* spectrometry. The 1R spectrum showed Lymantria ditpar I..3 Systematic fractionation of the absorptions for hydroxyl, two carbonyls, a lactone < 1768 extract hat already yielded three feeding-deterrent ses cm'1), and an ester (1738 cm-1), while the UV spectrum quiterpene lactones: lipiferolide, epitulipinolide diepoxide .3 had a low-wavelength peak at X—. 208 nm (log t 4.33) and peroxyferolide.* This report ia on the isolation and typical of an a-methylene >-lactone. The ‘H NMR spec characterization of a fourth antifeedant. tulirinol 41). by trum eras moat informative, exhibiting two weaitly split spectral and chemical methods, including X-ray crystal olefinic methyl peaks at b 1.81 and 1.91. and a three-proton lography. singlet a t 4 2.05 assignable to an acetate methyl. At tow Tulirinol (1), mp 204-6 *C, wa* isolated by extensive field, a pair of doublet* with additional fine splitting iJ m 0.6 H i) were located a t a 5.72 W *ll Hz) and 6.07 (J flf • 3.4 Ha), which are characteristic of y-lactone a-methy lene proton*. Double- resonance experiments allowed for the ordering of all functional groups except the hydroxyl. Irradiation at either of the two exocyclic olefinic proton frequencies caused collapse of the geminal coupling (0.6 Hi) in the i:*1 ■«*,■* • ** other and a change in a multiple! at b 3.08 < Ht> from a g-,*1***, ae triple triplet to a split triplet- These protons. H., Hb. and a^.e^aH could be arranged aa in A Irradiation at a 3.08 col- te'.enCrtfOCO, e*.*c chromatogTaphy of a partition fraction by monitoring
ill tel Department of Pharaacotofy, Caaa Waatara Raaarva Univer se^, Cleveland. OH. (fa) American Cyanaotid Co.. Bound Brook. NJ (S It- W. Doskotch. T. M. ODtlL n d P A. Godwin. frunrwv £nto- maH, a SeS 11*77). (ItlL W. DmkMcfa. S L. Kaatr. Jr. C D Knitted. end T. S. ElFeraly. PhyrecAamtalry, 14, 7S9 11*7*1. (SI TW.rino( dinwi* * H |*i(ka*t feeding inhjbilorv Ktivity-1 At (4) 1L W. Doaknch. F. S. B-Ftialy. E. R FaindaJd. and C.-T, Huan*. i at U sad ISO e t/w L feeding « w *# la d S3W, w p w . J. O p. Chenu, 0, 1614 (1*77).
4 5
1442 J. Org. Chtm., Vol. 45. No. 8, 1980 Doskotch e te l
CM O0 » .CM 001 p a t
COS) C(4) oa
Figure 1. ORTXP itereoview at tulirinol acetate viewed on tht S face. The thermal ellipsoid* art drawn at the M% level. Hydrogen atom* an omitted for clarity, and only on* of the two possible confotmstioiu of Ihs disordered scetoay substituent at C(l) it I lapsed the H, and Hb double doublets to doublets and derivative 2 was satisfactory, and resulted in the structure simplified two additional patterns. The double doublet shown in Figure 1. Before the X-ray results were exam at 4 4.75 <•/ * 8.7, 10.1 Hz) wa* reduced to a doublet (J ined in greater detail, it was immediately evident from the w ld.i Hr), as was another at 4 5.51 iJ » 9.5,10.4 Hz), with derived structure that one of the double bonds of tulirinol loss of the 9.5-Hz coupling. These protons, H 4 and H,, it cis and bean an allylic hydroxyl group. Also, the ring were assigned from chemical shift positions (vide infra) to conformation was found to be boal-chair and can be carbons bearing the ether oxygens of the lactone and the stylized as in 3 with the H15) to H(9) protons pseudoaxiafly aster, respectively. In turn, successive irradiation at 4 4.78 disposed—an arrangement originally indicated by the and 5.51 identified the neighboring protons Hf and as proton coupling values. broadened doublets located at 5.32 and 4.89 ppm, re spectively. and changed them to broadened singlets. The signal broadening was shown to be due to allylic coupling with olefinic methyls. The methyl at 4 1.81 interacted with Hr 14 5.32), and the one at 4 1.91 with H, (4 4.89). A somewhat broadened double triplet at 4 4.44 remained unaffected by the double-irradiation experiments and was considered to be the proton on a hydroxyl-bearing carbon. D ,0 exchange sharpened the pattern and also eliminated a one-pro ton doublet at 4 3.87, which, inridenlly. reverted to a singlet upon irradiation a t 4 4.44. In addition, ace- ty lit ion of tulirinol (1) produced the acetate 2 whose *H NMR spectrum showed a downfield shift of the 4.44-ppm proton to 5.5 ppm, as well as a second acetate peak at 4 The X-ray results provided only the relative stereo 1.98. chemistry for tulirinol acetate ( 21, and the molecule could Since the molecular formula of tulirinol (1) requires accordingly alto fit the heliangolide configuration 4. The seven double-bond equivalents, six of which are accounted absolute stereochemistry wa* subsequently determined by fur by the unsaturated lactone, two olefins, and and ace application of the Horesu procedure for establishment of tate. the one remaining must be for a ring system. Fur the configuration at the alcoholic carbon* For a cloetly thermore, all of the oxygens were assigned and no aliphatic related example, eurecurvin acetate, which bear* a sec methyls were indicated, thereby requiring a ten-membered ondary alcohol flanked by a cis olefin and a methylene, this alicychc ring- The olefins would be at positions 1 and 6 . technique yielded conclusive resuits.' Esteriflcation of rather than 1 and 5 as in a normal germacrane ring. The tulirinol (I) with optically inactive a-phenyl butyric an alcoholic function must reside on one of the three carbons hydride resulted in recovery of l-)-«-phenylbutyric acid not indicated by the double-resonance experiments, but in an optical yield of 5611. This requires an 5 configu the exact location remained uncertain, although an allylic ration for the arrangement of substituents at Cfl) and as position was favored from the width of the 'H NMR absolute stereochemical structure for tulirinol a* in 1. In pattern for the proton on the alcoholic carbon. Some keeping with the convention' of numbering the germacrane indication of relative stereochemistry or disposition of the ring with reference to placing the lactone carbon [C(1D) substituents could be inferred from the 'H coupling substituent 3 at C(7) |H(7j it u], tulirinol (I) is. therefor*, constants—all 9 Hz or higher, agreeing with a trans a 7 A- lactone but not of a standard subgroup.* arrangement of the relevant protons (Ht through H,); but lack of knowledge about the ring conformation precludes an established proof. Unavailability of sufficient quantities *6) In A. H otmu in “^lrn p F u n i t m n i i aftd M ftbok*, of tulirinol tl) prevented a chemical solution to the re V ii j , H- B Kanin. Ed„ THmim. Stuttgart, 1977, p 51. jbk W. H an u d maining structural problems (geometric and relative H. K u u . J- 0r$. Oitm„ X . 216 UM7>. <7i W, H tf i K. da G rw u. R M u n r, and J- F- Blount, J Qrr> Cht*b* stereochemistry) as well as to location of the hydroxyl; thus 42, 119761, an X-ray crystallographic study was initiated. Tulirinol <61 D. R o ftn , G. P. Mom, tftd S. NtidJa* J. C h tm Sot-, Chtm. Com* 11) itself did not give suitable crystals, but the acetate nun., 142 (1972V 6
Tulirinol J. O f- Chtm., Vol. 45, So. 9, 1900 1443 In ssejgning the lH NMR chemical ahift* for the proton of tulirinol (I) end it* derivative*, the detignation of the ' l a proton on the carbon bearing the ether oxygen of the I i | «u^e^5 5 lactone, to which other* would be related, wa* uncertain. 1* « m I I I i ll 31 90 9 Generally, the high-field poeition ii associated with the O qo Q o ok e lactonk group, while the corresponding proton on an hm ci *•« Oi •« ci « ester-containing carbon i* at lower field, with olefinic ge ometry playing an important role.1* Since the double m im. « m • m k k bonds of tulirinol ( 1) are not in the usual biogenet ic pos .□ rD A X Ok I1* 04 itions, the unequivocal location of the lactonic end ester- « Ok Ok wri related proton* becomes *11 the more important. For ex “* ample, in sesquiterpene lactones with -t and 9 double bonds da. the ester-related proton is located at higher field than ^ 5? n w ? those of the lactone.’ To settle the case for tulirinol (II, m -i e* ■W -1 *! — « w 00 >v> S c s c alkaline hydrolysis and relactoniration under acidic con PW *4 ditions afforded a diol, deacetyl tulirinol Fischer, sad L Bernal, (Vpc .Vo,-;. Acad. Sc,. L’-S.A , TV 14VI11973)], but « 2 s d the tc4eioac oh reded Their point li m il. taken if (he correct adoption ef subenup naaaa ia te be atrictly applied. Tuliriaoi dost not fit into any p e e . sot can it be called a tertnacrenolida. linn thia name i* I— ead we the tnana-1 (lOlZrana rytlodotadianoe. tla. i 11 if t a t ihai fermacranolidea be retained a* the generic name for all cyclodecadiene sesquiterpene lactoaea and that coatunolidce be introduced (or the b io fiatlii ally conect trana-l' (Q> trant-4 c a m With thia chans*, the n® to not fitting into the distinct croupe would be called fermecra- i prefixed with the appropriate atereorhrrmetry and numerical 1 (e.g.. tulirinol ia then a troru-4.rt4-9-xennacrannlidtl- (101 Cermaerwtolidei (double bond* at li I')) and 4 bnth train I fallow the pattern, eaamplaa of which can be (cund in R IV. Doakntrh and F. S. Ei-Feraly J Ort Chem.. 33. 1919 (1970). and H- Tada and K. Takeda. 41 Okeaa. A e a . Bull., 34. 667 (197S), an do melampohdea (cie- If l-o.tra/w-I I Ei (bran la M. H. Fischer. d A, Witay. D L. Perry, and K. D Hae«*l*. J. H i b f . C ham . 41.3936 (197S).whileh*l ianfnlidaa 11 rana-1 f 10 l.cie-4 ] show to the raverae feature: W H -n and I. Wehlbent. J. Or*. (Them.. 33. ,4At (1973), and M. Holub and Z. S a n k . C d lrri. O reh . Ckria. C o aru n ., 42. 1063 (1977). ( I ll H- Yoabioka. W. Ranold. and T. J. Mabry. J. Chtm. Sot. D, 143 avrai. (13) M_ H. P. Guy. G. A. Sim. and D. N. J. White. J. Chem. Sort. A rh u t Tnana l. 1917 (1973). Sw % (13) F. Shefieadeh and N R Bhadana. Phytochemistry. 13, SS7 (1*73); J. Orf. Chem , 37. 374 <19731. 1444 J. Org. Chun.. Vol. 45, So. 8. 1980 Doskotch *t *1 Table n. T onion An gin (dag) for Tulirinol A oU ta (1 )p atoms angle •tom* ingle C(a>-c(2)-c(i)-C(io) 489.8 aa>-oa>-c(i 2 >-o< 2 ) -170.0 C(8)-CO>-C(10)-C<9) -123.1 C(8 )-C{7 >-C(l 1 >-C( 12) -20.8 c(i)-c(io)-c(*)-c(8) r-0.3 C(8>-0(7)-C(ll>-C(13) + 157.9 Q10)-C(9)-C(8)-C(7) + 122.4 C(8)-C(7 )-C( 11 V"C(12) -143.9 C(9)-C(8)-C(7)-C<6> — 91.2 C(6H^7)-C(11>-C(13) +157.9 C(8)-C(7)-C(6)-C{8) + 49 0 C(7)-C(H)-C(12)-0(1) + 8.7 C(T)-C(e)-C(S)-C(4) -114.4 C(7)-C(ll)-C(12)-0(2) -172.9 C(6)-C(8)-C(4)-C(3) + 181.3 C(13)-C(ll>-C(12)-0(1) -170.3 «8)-C(4)-C(3)-C(2) -110.7 C(8H5(7)-C(6)-0(3) + 167.8 C(*>-C(3>-C(2>-Cd) + 42.3 C (ll )-C(7 )-C(6>-C( 5) + 184.1 C(2>-C(3)-C(4)-C( IS) + 84.2 C(ll)-C(7 HC(6)-0(3) -75.1 C(3K:(2>-C(1)-CH5)' + 172.9 0(3)-C(6)-C(5)*C(4) + 130.5 qjK(2)-c(i)-0(S) -174.1 C(8)-C(5)-C(4)-C(18) -13.1 C(2)-C<1H3(10)-C(14) + 58.6 C(7)-C(8)-0(3)-C( 18) + 155.6 O(8)-C(l>-C(10)-C(9) + 121.3 C(5)-C(8)-Of 3)-C( 16) -86.9 OfSr-CdHCdOMXO) + 115.2 C(8)-0(3)-C( 16)-C( 17) + 178.5 O(8)-C(l)-C(10)-C(14) - 59.0 0(6>-O(3)-C(l6)-O(4) -2.8 (XB)'-C( 1 >-C( 10)-C(14) -65.1 C(2hC(l>-0(5)-C(18)' +95.8 C(16}-C(10>-C(9)-C(S) -179.4 C(2)-C(1HX5)-C(18) + 146.0 0( 10)-C(9)-C(8 )-0( 1) -120.8 C(1HX5)-C(16H:(19) + 173.6 C(9)-C{8)-C(7)-C(ll) + 143.5 C(l)-0(5)'-C(18)-0(6)' + 2.7 C(1K(IK<7W(«) + 150.5 CdOHTdHXSI'-Cda)' -141.0 C(9)-C<8MX1)-C(12) -148.8 C(10)-C(l)-O(5)-C(18) -89.2 C(7)-C(8)-Otl)-C(ia) -21.8 0(l)-O(5)-C(18)-O(8) -8.5 * Torsion ingles for the I*cion* ring a n (two in Table V.
Figure 1 *bow» the configuration of the molecule u Table UL Lactone Ring Torsion Analea (das) In found in the X-ray itructur* determination, with the Rapran ntatlr* Sesquiterpene Lactone* correct absolute stereochemistry- Well-locaiixed tran* and (onion da double bonda are present at the 4 and 9 positions, eagle coetnno- aupsfor- me Lam- tulirinol reapectivaly, each bearing one methyl substituent. The identifier lide1* toon in'* pod in"’ acetate1 conformation of the cyclodecadiene ring places these u , -8.6 + 4 + 8.6 + 8.5 methyl group* in the anti configuration with CU5) d and ‘•'t -9,7 + 9 + 19.1 + 8.2 C(14) a- The placement of the double bonda and the Irani —, -24.9 + 14 + 35.0 + 25.2 lactone fusion at C(7)-C(8) apparently relieve* some strain +J. + 38.6 + ias -75.8 -91.2 in the ten-membered ring and permits it to adopt the sign of (-) (+) ( + ) (+) chair-boat conformation more representative of the he- Cotton liangolide *114 than of the melampolide*.* The reduction affect in ring strain ia moat apparent in the torsion angle* of the * This work. For tuilrinol, the torsion *n|l*s arc identi double bond* (Table 111, which a t 1° (cis) and 19* (tran*) fied at follows: +>,, C18)-0(1>-C<12I-C it) ; -j,, C( 13)- are significantly less than the 8 ° and 24* observed in C(Il)“C(12)-0<2); u ,, C(H)-C(7>-C(8>-Oll); and +j„ melampodin.* The torsion angle of the tran* olefin linkage C(6)-C(7)-C(»>-C(9). in 2 can be resolved into a 6* bend (away from ring center) and a 13* twist about the bond axis .15 Thus, the cyclo by crystal packing contacts. Although Cit) must certainly decadiene ring in 3, though qualitatively more open than also posses* two alternative positions, it was not possible in moat germacradiene*. where bends and twists in the to refute two separate sites (as was done for 0(5), C(IS), olefin linkages of this magnitude are common ,1517 still and 0 (6): vide infra) because they appear to be separated reflects significant strain from transannular non bonded by lee* than 0.1 A- Each of the acetoxy substituents, in interactions (Table 4 in supplementary material1*). The cluding the half-populatad acetoxy groups, is planar within interior angles around the ring deviate from ideal values, experimental error. There being no hydrogen bond donor* another indicator of ring strain. Although the C~C and in the molecule, the crystal packing is dominated entirely C -0 distance* in the molecule a n in general normal when by van der Waal* contact*. the hybridisation of the involved atoms is taken into ac The endocyclic torsion angles of the 7 -lactone ring are count, the average of the aliphatic C-C bond* is, at 1.496 quit* different from those found in other sesquiterpene A, somewhat shorter than usual. This is perhaps a con lactones (Table HD- Curiously, the values for tulirinol sequence of inductive shortening at the carbons bearing acetate (2 ) are almost perfectly negatively correlated with oxygen substituents. those for costunolide.1’ 'H m acetoxy substituent at C (1 > is disordered, with about The CD spectrum of tulirinol (I) shows three Cotton 63% of the molecules possessing one of the two discrete effect n u tin u which appear at 231,236. and 211 nm. The conformations. The Cl 19) methyl group is immobilized highest wavelength maximum with (t)f 4-120* was assigned the n — *■• transition of the lactons cnrbonyL” The
(14) (*) M- NiahUtew*. K_ Kaniv*. A Tekabatekt. H. 0 » h » . Y. To- adw. and I Mitt*. r t i n M n i , 12, 7001 lb) A. T. M cPhailand (19) M. J. BevOL P. J. Cos. F D. Crsdwitk. M. H. P. Guy. G. A Sim. K. D. O n u . J. Liter*. Sot.. Prrkm Tnm . 2. 578 (l»7S>. and D. S. J. Whila. Arts Ov»telio*+. Stct. B. JJ. 3303 11976). (IB)OiHtv*ku*iatncalculated by tha methoddF.K. Winklerstid 110) The true transition nerelettlth would be le*a than the observed J. D. DuniU, J. Mot. Bioi.. S*. 1*9 119111- 3S1 so, lira the intense positive maximum at 33t> nm in th* rummation (14) H. Keyema snd Y Mieutwa-Ttukud*. J . Chem, Sot.. Ptrkin amb the weaker pnetUre peak at the hi after travelenfth rtaulta in the Trout. t St« 11977); D. N. J Wbile. Htl, Chin. At10 . S*. 1347 11973). apparent etaatmunt of the latter bein* thifted to a hi*iter waveieniih. (IT) M. A. Beoo. Ph.D. Tbeeia, The Ohiu State (Jnirenity. 1979. Thelactonic a - r- traoailtoa ia Bormally found betweea 340 and (15) Sea patrurapk at end at paper nsardin* auppieneniary malarial. T u lirin ol J. O ft. Chem., Vol. 45, No. 8. I960 1441 empirical rul* of GoiiMun ind co-worker*21 carractly A 5-mg sample of deecetyltulirinol (5) wa* treated with Ae*0 predict* • tran* lacton* closed to C-8 for tulirinol. Alao, in pyridine to give 5.5 mg of crystalline tulirinol acetate (2) the lactone torghm angles listed in Table 111 for tulirinol identified by direct comparison (melting point, t*]t* IR, and 'H acetate (2) ahow that the chirality of both the C K - O O NMR) with a sample prepared from tulirinol (1). and unite art positive with respective Herean Esteriflcalion of Tuliriael (1). A 10-mg (0.DS3 C(lI)-C(7)-C(8>-0 mmol) sample of I and 26.1 mg (0.084 mind) of a-phettylbutyrie value* w, +8.2" and u, +25.2*,11 and that according to the anhydride were stirred in Z5 mL of pyridine for 38 h at ambient empirical rule of Beecham” a positive Cotton effect should temperature. After 1 mL of H,0 was added, furring waa continued be observed for the n — »* transition of the lactone. This, for 6 h, and the solution wa* further diluted arith HgO (3 mL) indeed, is the case. Tulirinol (1), therefore, conforms to and extracted with EtjO. The EtjO layer waa extracted suc both the Geiesman and Beecham rules. Since the intera cessively with H,0 (3x3 mL), 5% NtHCO, (3 X 10 mL). and tomic distances1* between the 4 and 9 double bonds sppear HyO (4X5 mLl. All of the aqueous Layers * m combined, too large for effective overlapping of orbital* to show acidified with 1 N HjSO, to pH 2, and extracted with CHCl, The transannular conjugation, the CD maxima at 236 and 211 HrO-washed CHCl, phase on evaporation gave 15.4 mg of w- nm were assigned t o r — r " transitions of the lactone and pheny (butyric acid as a colorless oil (on* spot on TLC), |a l“ a •12-9* (c 0.513, CsHs). eoetttponding to an optical yield of 56%. olefin groups, respectively. A neutral EtfO fraction contained 16.4 mg of oar-spot materiel [TLC Rf 0.83 with hexane-EiOAc (4:61] crystallised from i- Experimental Section 4 PrI0-h*xana a* rosettes of I: mp 149-150 *C after softening at Isolation of Tulirinol (I). The ethanolic rnidus (1.5 kg) 147 *C; [ot^D +58* (c 0.45, MeOH): IR (CHCl,) r _ 3030 (Ar), obtained from percolation of 6.7 kg of L. tuliptfm leave* «a* 1770 (lactone), 1740 (double intensity, ester), 1665 (olefin), 1605 partitioned between CHCti sad HjO, end the CHCL solubles were end 1495 cm*1 (Ar); UV 206 nm (log 14.39) and aromatic fine further divided between heiene end 10% aqueous MeOH ss structure at 251 (shoulder, 254). 256 (2-47), and 264 (2J5); mass already described.' Tbs residue (239 g) from the MeOH phase spectrum (El), m /t 452.2207 (25 %; C-,H„0, require* 452£199), was taken up in 30% aqueous MeOH and extracted successively 393 (13. M - AcO). 306 (5). 289 (9), 246 (12). 229 (25). 201 (7), with heisns-CHClj <4:1 end 1:1) to give 73 g of material from 164 (8), 146 (12). 119 (97, PhCHEt). 91 (100), 77 (It, Ph), and the last solvent, which was chromatographed in sequence through 43 (47. Ac). the following m lnnis Silicic acid (500 g, Mallinckrodt) containing DescetyltuUrine) (5). Tulirinol (16 mg) and 53 mg of NeOH 5% HjO was eluted with CHCL and 1%. 2%. and 5% MeOH in in 3 mL of H.0 were stirrod under N, for 2 h. After eejdificatioo CHCl* The 2% and 5% MeOH in CHCl, rrtiaue (6.4 g) wee with 0 1 N HCI and stirring for 30 min, the sohrticm m saturated separated on 120 g of silicic acid 12-5% H,0> with C,H,-CHC1, with NaC) and extracted with EtOAc. The EtOAc phase wa* (1:1 and 3:17) as eluentt. The last solvent gave *.5 1 oi material washed with 5% NeKCO, and HjO Evaporation of solvent left from which flavnnoid eonstituteat* were removed by filtration 133 mg of a whit* solid (ftf 033 on TLC with besane-EtOAc (4 A) through a Sephedex LH-20 column (2Mg> using MeOH as solvent that was chromaiafraphedce) 1 g of silica gel with the TLC solvent The effluent residue (3.06 g), rachromatogrepbed on silicic acid ayitem to give 6.5 mg of diol 5 which crystallised from hexsne- (300 g, 2.5% HjO) with 1.3% MrOH in CHCl,, gave * fraction EtOAc to give 4.9 mg of fin* need)**: mp 165.5-6.5 *C: [a]Hp from which a total of 94 mg of tulirinol (I) wa* obtained, in ■ yield -46* (c 0.36, MeOH): CD (ID W SteckUa. T. G. Waddell. and T. A C iw sm . T’etrehedrsn. (241 The density af tulirinol it inunndiate htiwete that af I* M, 2397 (19)01, propanol (0.725 | cm'1) sod water tl.O t cm''); there were ineufftoeot (22) A T. McPhail rod C- A Siau T.trohekron, I t . 1751 (1*72). crystals lo carry out t precise density msetunaitnt. (23) A F. Bsediam. Tetrahedron. 21. 5542 (1972) (25) tV R, Busing end H. A Levy. Acta Cryeishogr., 14,180 (1V57). 144C J- O rt. Che/u., Voi. 43, No. 8. 1990 D nskotch a t xL T a b le tV. TMMitol Aeetst* Cry«*l Dnta by direct methods * and refined by standard leaat-tquarr* and farmuls Fourier techniques7’ to final disarmament indices: AlFI “ 0.064. naol » t 848.40 g/mol Mef*) - (1009. and GOF “ 1.30* for all 1634 unique intensities. Hydrogen atoms ware all located and their contribution* includad apnc* (roup « , (C,1, No. 4) la the calculations, but their parameter* wrre not rerinsd. Th* r_i>4 * 1.336 1/cm* •“ 4.4 7 0 (2 )* • « T.* (r thermal parameters for all the nonhydrogen atoms were allowed b ■ 7.9661 (8) A *- 80.037(7)* to taflo* anisatropically. A secondary extinction correction waa C - 6 .7 9 8 7 ( 5 ) A V - 944.5 (2) A * included.* Th* acetate group was found to be disordered: T * 20 (1)*C howssst, the affected atoms were tu/Ticiently removed from each •(Mo IB 1-0.71069 A other (ca. M A to 1.0 A) that th* two conformations refined A (M o K i)-0.9789 cm-1 successfully with approximately equal refined populations “ 0.98S (53%/47%). The final difference map contained a Taw peak* of * < -» r„-) ■ 0.954 ca ( 1 3 0 */A' with a general noise level of *0 20 */ AJ. The final data collected for 4.0* c !» < 50. +10- 250 300 nm -30- + 120- +80 +40- -50- So 270 280 290nm -40- - 70- Figure 1:1-3 CD Spectrum (MeOH) of T u lirin o l{1). 90 MHz x j i - i J. -i... t- j. J l Lt LJ — l___I I I L j i I i i_ 6 5 4 3 2 Figure 1:1-4 PMR Spectrum (Acetone-dg) Tulirinol(1). REL. INTENSITY in o in o o 0 i u e :- MSFigure 1:1-5 Spectrum Tulirinol(1). (El) of 100 /E M 200 j i 'i i i i—i l l i n 300 ~ i ~r i 400 r~ PftC'Mr !«A*lMISStON iu e :- IRFigure Spectrum 1:1-6 (CHC13) Tulirinol of Acetate(2). J700 SJ ■ 4 p ' j 4 W 4 [ H B 4 # I W A V tL H G T H * NUiUt CM** t i*U U fN »*V uoo 15 250 3Q0nm 20 100 40 50 60 290 300 80 Figure 1:1-7 CD Spectrum (MeOH) of Tulirinol A cetate(2). 90 MHz JUUU. _ A k _ j v L l lLl lllI i.i i i I n i i1. lllj_Ll u j 1 i i llI j i i i Llu l! iii i 1 i i i i I i iii I mil j i i i I Figure 1:1-8 PMR Spectrum (CDC13) of Tulirinol Acetate(2). REL. INTENSITY ID ID ID o o o r- C\i i u e :- MSFigure 1:1-9 Spectrum Tulirinol (El) of Acetate(2). 100 200 M/E 300 FftCENl flA M SM IlttO * ISO j __I. i u e :-0 IRFigure Spectrum 1:1-10 (CHjClj) Deacetyltulirinol of (5). MB i r ■i V A V E IN Q IH ^ I AItIR ' f O WAVINtMICR Hi H- m 90 MHz t i l i i i l7 I h h I i i i l 1 l l i l I i h i L l l h L l l i i 1L h lI i illLl l h I i . l l 1 i l i . L .i i i I . t t t 1 ! , . . j 8 6 5 4 3 2 Figure 1:1-11 PMR Spectrum (Acetone-dg) of Deacetyltulirinol(5). VO REL. INTENSITY 0 1 o Figure MS 1:1-12 Spectrum Deacetyltulirinol(5). (El) of M/E 300 400 o to fEaCtnr T«AN)MlSS*ON 1 1 1 — “ -f- I - t H - r .. ;F f T iue :-3 R pcrm CC3 o Tultrinol-a-Phenylbutyrate( (CHC13) Spectrum IR of 1:1-13 Figure ■ 1 E ; AMI C WAVMMIU t a 6 ). *0 90 MH; ■1 1 1 i l l 1 1 I h I I I L i I i I U I J I li.l l Li h I li m l i LI i I n I I I i l l il I I I . | , i t i 1 i l i l 1 t i i t I . ■ . i I ■ . . . I 8 7 6 5 4 3 2 1 0 Figure 1:1-14 PMR Spectrum (CDCl^j Qf Tulirinol-e-Phenylbutyrate(6). w NJ REL. INTENSITY Figure MS 1:1-15 Spectrum Tulirinol-a-Phenylbutyrate(6). (El) of lO 0 1 O o o 100 M/E 200 300 u> M Chapter 2: Sesquiterpenes of Liriodendron. Isolation and Characterization of Six Sesquiterpene Lactones from the Leaves and Root Bark Possessing Cytotoxic and/or Antifeeding Activity. INTRODUCTION Liriondendron tulipifera L., is commonly known as the tulip tree, tulip-poplar or yellow poplar. Itfs genus contains a single species, and belongs to the family Magnoliaceae, far removed from the true poplars. The straight-grained wood is resistant to splitting and is used in furniture, interiors, shingles, boats and implements^. It is found growing in woods from Massachusetts, Ontario and southern Vermont to central Florida, Louisiana, and eastern Arkansas. It is characterized as a tall, straight tree often reaching heights of 50 to 100 feet. The bark is light grey, and the leaves are 6-10 inches in diameter, 4-pointed, fan-lobed, and hairless. When crushed, the buds and leaves emit a lingering, spicy-aromatic fragrance. The flowers are generally seen from May to June, and are large, orange and tulip-like in appearance . Previous investigations into the constituents of the tulip poplar have led to the isolation of a variety of 24 25 compounds including: myo-inositol^, liriodendritol, the dimethyl ether of myo-inositol®, lignans**-®, such as liriodendrin, a di-a-D-glycoside of the lignan lirioresinol1*, the cyanogenic compounds? taxiphyllin and triglochinin, sterols®, carotenoids^, and several aporphine alkaloids10-1® such as liriodenine and glaucine, examples which contribute to the yellow color of the heartwood. In addition, Doskotch and coworkers isolated nine sesquiterpene lactones which showed cytotoxic activity in KB cell culture assay, and/or antifeeding activity against Gypsy moth larvae (Lymantrla dispar The structures of these compounds are shown in Figure 1:2-1. Structural similarities, physical constants, and biological activities are listed in Tables 1:2-1 and 1:2-2. Investigations related to the isolation and structure elucidation of sesquiterpene lactones have increased substantially in the last decade for two main reasons: f i r s t , the sesquiterpene lactones have been su ccessfu lly employed as chemotaxonomical markers, and, secondly, they possess a variety of biological activities representing a potential source of therapeutic agents. In the most recent review on the biogenesis and chemistry of the sesquiterpene lactones, Fischer2? estimated that there had been 924 naturally-occurring lactones isolated by the end of 1977. These compounds are divided 2 6 10 OAc ,OAc .14 COSTUN0LIDE tulipinolide EPITULIPINOLIDE HOO OAc ,OAc .OAc EPITULIPINOLIDE PEROXYFEROLIDELIPIFEROLIDE DIEPOXIDE HO OAc OAc —O OAc EPITULIPDIENOLIDE Y-LIRIODENOLIDE TULIRINOL Figure 1:2-1 Sesquiterpene Lactones Found in Liriodendron tulipifera L. 27 Table 1:2-1 Sesquiterpene Lactones of Liriodendron tulipifera: Source, Activity and Class. Compound Name Source Biological Activity 3 Class*1 Cytotoxic Antifeeding Costunolide root bark 0.26 -— G Tulipinolide root bark 0.46 ------G Epitulipinolide root bark 2. 10 ------G Lipiferolide leaves 0. 16 X G Epitulipinolide d iepox ide leaves 0. 34 XG Peroxyferolide leaves ------XG Epitulipdienolide root bark inact . ------El Y-Liriodenolide root bark 4. 10 ------Eu Tulirinol 1eave3 ------53% Mc aCytoxicity was tested in KB c e lls through the Cancer Chemo- therapy National Service Center following the protocol in Cancer Chemother . Reports, 1962, 25, 22. Antifeed ing activity against Gypsy moth larvae was determined following guidelines in Environ. Entomol., 1977, 6 , 563. C ytotoxicity is reported as the ED^q(^g/ml), and antifeeding activity as the percent feeding inhibition at 250 4g/ml; X indicates men tion of activity without supporting data. bAbbreviations are: G-germacranolide, El-elemanolide Eu-eudesmanolide, M- melampolide. °Not a true melampolide and is more appropriately designated as a trans-4-cis-9-germac ranolide. 28 Table 1:2-2 Sesquiterpene Lactones of Liriodendron tulipifera: Physical Constants and Lactone Stereochemistry. Name MP Lactone CD Costunolide 107° + 1 2Sf* 6 , 7-t [0 ]262 -66OO ^ ^ 220+^ ^ ® t 000 Tulipinolide 6 , 7-t 118° +249° Cd]264-4> 780 [^]221+121,000 Epitulipinolide 92° +76° 6 , 7-t [0 ]264 - 7 , 180 222+^ * 000 Lipi ferolide 119° + 12^° 6 , 7-t ------ Epitulipinolide d iepox ide 215° -56° 6 , 7-t Perox yferolide 190° +30° 6 , 7-t ^ ^ 257"^» 000 «300 Epitulipdienolide 135° + 6° 6 , 7-t ------ Y-Liriodenolide 180° -4° 6 , 7-t ------ Tulir inol 206° 8 -51° 7» -t [0 ] 281 +120 C0 ]236“6»S00 ta 3 21 -|-67, 000 29 into twelve sk eleta l types, some of which have further subgroups. It has been well established from biosynthetic studies on steroids and triterpenoids, and the work of Ruzicka and Hendrickson, that the sesquiterpene lactones derive from trans, trans-farnesyl, or nerolidyl pyrophosphate via the mevalonate-isopentenyl pyrophosphate pathway*^. In a continuing investigation of the biological activity and phytochemical composition of Liriodendron tulipifera, five additional sesquiterpenes of the eudesmanolide, germacranolide and guaianolide skeletal types have been isolated: two from the root bark and three from the leaves. Their isolation and structure proof are presented herein. DISCUSSION AND RESULTS The 10% aqueous methanol partition fraction of the ethanol extract obtained from the leaves and root bark of Liriodendron tulipifera (tulip poplar) was previously shown to possess antitumor activity against KB tumor cells in tissue culture assay, as well as antifeeding activity for Gypsy Moth larvae (Lymantria dispar L .) 2®, 29. This activity has been attributed to the sesquiterpene lactone constituents justifying a further investigation into the members of th i 3 c la s 3 . Reapplication of the basic isolation scheme^ to the 10% aqueous methanol fraction obtained from various collections of L. tulipifera resulted in five additional sesquiterpene lactones: two from the root bark and three from the leaves. These include: a-lirio d en - nolidet I_) , p-liriodenolide(I_I) , 11, 13-dehydrolanuginolide (III) , dihydrochrysanolide(IXA) , and £-cyclolipifer- olide(XII). y-Liriodenolide( TV) has been previously isolated from the root bark of L. tulipifera2^. Its structure was confirmed by synthesis from the acid-catalyzed cyclization of e p itu lip in o lid e -1,10-epoxide. The la tte r was produced from the epoxidation of epitulipinolide with one equivalent 30 31 HO .OAc ,OAc 1lm-CPBA 2) H epitulipinolide liriodenolides of m-chloroperbenzoic acid. Two additional compounds, a-(Jt), and /3-liriodenolideCI_I) were also produced during the reaction, and have now been isolated from the root bark in addition to y-liriodenolide(TV). The fractionation and chromatographic results are depicted in Figures 1:2-2 and 1:2-3. The physical data for both natural and synthetic materials were shown to be identical (Tables 11:2-3 and 11:2-4). The stereochemistry of and at the six asymmetric centers was determined as follows: (a)centers C-6 , -7, and -8 in the products must be identical with that in e p itu lip in o lid e , a compound of known absolute stereochemistry; (b)H-5 must be alpha by nature of the large coupling constants (J.5 ^ = = 11-0) corresponding to a dihedral angle of 180° indicating a trans-diaxial relationship between proton pairs 5-6, and 6-7; also reference to a naturally-occurring trans- 6 , 7-eudesmanolide with a 5-/3 proton could not be found in the litera tu re; and k* tulipifera root bark (Miss. July 1973) 400gms percolation with 95% EtOH 49gms viscous oil 4 7 .8gms partit ioned between chloroform/water chloroform solubles inter face water solubles (38-5gms) ( 1. 6 gms) (6 . Igms) hexane/90% MeOH partitioning hexane 90% MeOH solubles solubles ( 36 . 5gms) ( 1 . 4gms) benzene/70% MeOH partitioning benzene solubles inter face 70% MeOH solubles (22. 8gms) ( 0 .75gms) ( 12.7gms) Figure 1:2-2 L. tulipifera Root Bark Extraction and Partition Scheme. 33 Benzene solubles red gold o il 22. 4gms 220gms silicic acid containing 13$ water solvents: 25% isopropyl ether/hexane, isopropyl ether, and 2,5,10, and 20% MeOH/isopropyl ether 20 fractions A — 0. 04 B - 0. 14 C — 0 . 16 D - 0 . 15 E - 3.57 F - 0. 49 G - 0.45 H - 1.88 I - 0. 94 Fractions I and J J — 0.73 ( 1. 6 gms) K - 2 . 18 L — 0 . 30 85 gms silicic acid M - 0 . 82 benzene:chloroform( 1:1 N — 0. 32 chloroform, 1 and 2% 0 - 1. 63 MeOH/chloroform P — 1.99 r Q — 0.81 11 Fract ions R - 2. 67 A - 38 S — 2.21 B -■ 179 T - 0.40 C -- 167 gms ■ D - 63 E -■ 42 F - 54 y-liriodenolide G - 262 y+ a-liriodenolide H - 217 a-liriodenolide I - 216 a+ /i-lir iodenol ide J ■ - 105 j8-lir iodenol ide K • 306 mg Figure 1:2-3 Chromatographic Isolation of a-, (3-, and y-Liriodenolide. 34 Table 1:2-3 Physical Constants for a-LiriodenolideC!_) Constant Isolated Synthetic mp 72-3° 74-5° tlca Rf=0.29 Rf=0.29 - + 42° (c=0.45,MeOH) IR (CHC13) 3610,3500, 3620,3520, 1773,1745, — ,1740, 1681, — 1680,1605, 1210- 1250 , 1210-1250, 1010 *H NMR 6 . 16 (d) 6.1 9 (d, H-13) ( cdci3 ) 5.73 £ dt) 5 .7 6 (m, H-8) 5.44(d) 5 . 46(d, H-13) 5 . 37(m) 5 .40(brm, H-3) 4 .4 4(dd) 4.4 7 (t , H-6 ) 3. 68 (brm; 3.69(brt, H-1 ; dd with D«0) sharpens with D20 ) 2.7 5 (dq) 2 .7 8 (dq, H-7) 2 . 06(a) 2. 07 (s , OAc ) 1.89(brs) 1.89(b rs, H-15) 1.40(br ; 1.80(brs , OH; lo st with D«0) lo st in D2O) 1.06(s ) 1.07(s , H-14) MS m/e 263.1281 (0 . 7%, M+) aMeOH:iPr20 :CHCl3 (8:41:41) HO 35 Table 1:2-4 Physical Constants for 0 -Liriodenolide( ) . Constant Isolated Synthetic mp 119- 20° 119- 20° t lc ° Rf = 0. 14 Rf=0.14 +52°at 23°C +53°at 25°C (c= 0.365,MeOH) (c=0.44, MeOH) IR 3610,3510, 3610,3510, ( CHCI3 ) 1775, 1748, 1770,1740, 1683,1660, 1675,1650, 1210- 1250 , 1210- 1250 , 955,905 *h nmr 6 . 16 (d) 6 . 18(d, H-13) ( CDC13) 5 . 72(dt) 5 .74(m, H-8) 5 . 45(d) 5 .4 7 (d, H-13) 5 •02(brs) 5 .0 4 (brs) 4 .9 5 (brs) 4 . 97(b rs, H-15) 4 .5 3 (dd) 4 .5 5 (t , H-6 ) 3 .5 1 (m; 3.53(dd, H-1 dd with D20) sharpens with D2O) 2 .8 1 (dq) 2.81(dq, H-7) 2.06(a) 2 .0 7 (s , OAc) 1. 67 (brs ; 1.67(s, OH; lost with D9O) lost with D2O) 0 . 97(a) 0.98(3, H-15) MS ------m/e 306.1461 (0.1%, M+) aMeOH:iPr20:CHCl3 (14:193:193) HO 36 (c)the position and pattern for C-l was similar to equivalent protons in / 3-cyclopyrethrosin^. Ludalbin, a C-1 a-hydroxy eudesmanolide, possesses a different position and pattern for its C-1 proton^. Thus, epitulipinolide-1, 10- epoxide serves as a logical biosynthetic precursor for the 1iriodenolide isomers. However, a previous search*^ failed to reveal its presence suggesting that it is not an intermediate, or that its steady state concentration was too low for detection at that time. Similar procedures used in the isolation of 1^ and LI from the root bark were applied to the ethanolic extracts of the leaves from L. tulipifera. Repetitive chromatography furnished a crystalline material identified as 11, 13-dehydrolanuginolide(III) , a compound previously % isolated from Michelia lanuginosa (Magnoliaceae)^ , but new to this source. The physical data for compound III is presented in Table I : 2-5. This material has been characterized as the C-8 epiraer of 1ipiferolide( V) . The latter compound occurs in the leaves of L. tulipifera^ . Epoxidation of III with excess m-chloroperbenzoic acid gave a diepoxide (VII) which was identical with that obtained from tulipinol ideC^V) under similar conditions^. This confirmed the stereochemistry of carbons 6 , 7 and 8 as well as the nature of the fifth oxygen (Figure 1:2-4)^. The trans stereochemistry of the 6,7-lactone was further substantiated from CD measurements. A negative Cotton 37 Table 1:2-5 Physical Constants for 11, 13-Dehydrolanuginolide(III). Constant L. tulipifera M. lanuginosa mp 168° 168° tic Rf=0.3 ----- iPr2 0:CHCl3 (2: 3) [a] U -98° -97° (c = 0.32, MeOH) {c=0.74, MeOH) uv 21 5nm 215nm Clog « = 4.04) (loge=3.82) IR CHC13 KBr 1769, 1740, 1779, 1733, 1656, 1240, 1647, 1233, 893, 830, 901, 821, 812 806 ^ H NMR CDCI3 CDCI3 H-13 6.38,d(3.5,0.6) 6 .37,d H-13 5.75,d(3.2,0.6) 5.75,d H-15.28,brdd(4.5,10.4) 5 . 30,m H-8 4.54,dt(8.9,4.1,4.1) 4.59,m H-6 4.26,dd(6.7,9.2) 4.25,dd H-7 3.24,m(3 * 2,3-5, 3.27,m 4.1,6.7) H-5 2.64,d(9.2) 2. 63 ,d OAc 2. 03,s 2. 00, s H-14 1.80,s 1.81 ,s H-15 1.28,s 1.28,s MS 306 ( 1. 2%,M+) 306 (0.5%,M+) CEI) 246(25%) 246(65%) 203(12%) 203(22%) 188(46%) 188(100%) OAc 38 OAc OAc m-CPBA m-CPBA O OAc OAc VIII VII OAc Costunolide Figure 1:2-4 Epoxidations of Germacrano1 ides . 39 effect is observed at 260nm ([03-340) consistent with the empirical rule of Geissman and coworkers^. The location of the epoxide at carbons 4 and 5 was confirmed by isolation of the 1,10-epoxide (VIII)1*^ from the reaction mixture of tulipinolidefIV). This compound had different physical constants from that of III. The solution conformation of tulipinolide(LV) has been related to costunol ide^® by comparison of the n-rr* transitions in the CD resulting from exciton splitting of the transannular double bonds^1*. The conformation of costunolide has been determined from X-ray analysis of the silver adduct and Nuclear Overhauser Experiments. As a consequence, of the all-chair conformation, epoxide formation will result in the configuration at all four epoxide centers (ie. 1, 4, 5 and 10) thereby establishing the stereochemistry of the single epoxide in III. A second compound isolated from the leaf extract is dihydrochrysanolide(1XA), the third new naturally-occurring sesquiterpene from JL. tulipifera to be reported herein. Dihydrochrysanolide(IXA) crystallizes as needles with a melting point of 163 - 3-5 °, and an optical rotation of +9°(c=0.091, MeOH). The IR spectrum contains absorptions at 3595 and 3495cm -1 indicative of a hydroxyl group. The intense absorptions at 1765 and 1739cm-1 indicate the presence of a y-lactone and acetate functionality, respectively. The UV spectrum exhibits only end absorption, and the CD contains two Cotton effects: one at 252nm <[63-3970) and the other at 220nm([03+13,280). The mass spectrum had a molecular ion at m/e^ 306 corresponding to a molecular formula of C 17h2205 which calculates for seven double bond equivalents. A review of carbon and proton NMR absorptions, and the results of decoupling experiments^® led to the following proposed structure (IXB) having unknown stereochemistry at C-1. H0 IXB A i u C . I OAc OAc This compound appeared to be closely related to chrysanolide (X), a known compound from previous studies on Chrysanthemum cinerariaefollum^ . If this is true, then oxidation of the alcohol with Mn0 2 should give chrysanolide(X) directly. The product isolated from the reaction mixture was identical in all respects with the naturally-occurring material thereby establishing the absolute and relative stereochemistry of all carbons except C-1. Application of the Horeau procedure was used to determine the configuration at the alcholic carbon^. Esterification of (IXB) with optically inactive 41 a-phenylbutyric anhydride resulted in the recovery of C-) a-phenylbutyric acid having an optical yield of 20%. Therefore, C-1 must have the S configuration resulting in an a-oriented hydroxyl group. This was further confirmed using the NOE difference technique. A degassed sample in CDCl^ was used for the experiment. On irradiation of H-1, Nuclear Overhauser Enhancements were observed for the C-4 pseudoaxial methyl group (4.1%), and the hydroxyl proton (4.5%). 4.5 % /■~N OH 4 . 1% CH Although the enhancement of the C-4 methyl is small, it does indicate that the C-1 proton is oriented toward the j3-face. If a-oriented, enhancements of H-5, H- 9ax and possibly H-7 would be expected. Synthesis of dihydrochrysanolide(IXA) has been accomplished^ by photooxygenation of laurenobiolide(XI) with visible light, and methylene blue as sensitizer^y ’3®. This suggests a feasible mechanism for its biosynthetic 42 origin if laurenobiolide tissue. An investigation of the 90% MeOH solubles of tulipifera was successful in isolating 1 aurenobiolide()Q) in a 0.006% y ie ld ^ . It is possible that an even higher yield of XI^ may be found in the root bark of L. tulipifera since the less-oxygenated germacranolides (costunolide, epitulipinolide and tulipinolide) predominate there. The last compound to be discussed is that of j3-cyclolipiferolide(XII), a new sesquiterpene lactone, and the first guaianolide to be isolated from L. tulipifera. Repetitive chromatography of the hexane:chloroform partition fraction resulted in 153mg of fl-cyclolipiferolide(XII) in an overall yield of 0.0023% from the dried leaves. This material crystallized from diethyl ether as needles; mp 161-2° and [a ]jj-l01°* The UV spectrum displayed end absorption at 210nm (log e = 3*97; MeOH). Infrared peaks were present at 3590 and 3500cm- ^ (0-H stretch), 1772cm- ^ (a ,£-unsaturated-y-lactone) , 1743cm”^ (acetate CsO stretch) , 1670 and 907cm“^ (C=CH2, lactone), and I645cm""^ (C=CH2 ). The electron impact mass spectrum showed a molecular ion at m/e 306 (0.1%). Chemical analysis gave the molecular formula C17H22°5 *1/2H20 corresponding to seven double bond equivalents. The CD spectrum showed negative Cotton effects at 260 (shld, [(93-1100) and 2l7nm ([03-20,400). A comparison of the proton-noise decoupled and off- resonance decoupled spectra showed a total of 17 carbons 43 (guaianolide skeleton plus one acetate). Two singlets appeared at 170.3 and I69.0ppra suggesting two carbonyls. Four carbons appeared in the sp^ region (110-115ppm) as two singlets and two triplets corresponding to two exocyclic methylenes. The singlet at 79.8ppm, and two doublets at 77.9 and 66 . 6ppm represented oxygenated centers. The rest of the spectrum (20-60ppm) displayed three doublets, three triplets and two quartets. The HMR spectra was determined on a 300MHz instrument in CDC1^ and dg-acetone (Tables 1:2-6 and 1:2-7). A number of characteristic structural features were immediately apparent in the deuteroacetone spectrum. A low field pair of doublets appears at 6.14 and 5.56ppm characteristic of the exocyclic C — 13 protons of the y-lactone. Two broadened singlets at 5.14 and 4.96ppm are assignable to a second exocyclic methylene which lacks the 3Hz allylic coupling characteristic of the y-lactone system. The C -6 proton is present as a double doublet at 4.U7ppm with two large coupling constants indicating a trans-diaxial relationship. Also, one acetate methyl at 1.99ppm and one quaternary methyl at 1. 33ppm are present. The problem concerning the exact placement of atoms was attacked by double irradiation experiments. Coupling of this information with that obtained from NMR, NOE, IR, and CD measurements led to the following proposed structure. 44 Table 1:2-6 H NMR, Double Irradiation and WOE Difference Results (300MHz, dfi-Acetone) for j3-Cyclol ipi ferol ide(XII) . 'h 5^, Mult. Comments 1 3. 17,m J(1,5)=11.5 collapse to brt; H-5 irrad'd 10.5% enhancement; H-5 irrad'd 2,3 1.7-1.9 undiscerned patterns 5 2. 39,dd J(1,5)=11.5; J(5,6 )=11.5 11.5% enhancement; H-l irrad'd 6 4.47,dd J<5,6 ) = 11.5; J (6 ,7) = 9- 1 6.1% enhancement; H-15 irrad'd 7 3•30,dq J(6,7)=9.1; J(7 ;8)=2. 1; J(7,13) = 3-5; J (7 ,13) = 2.9 collapse to dt; H-13B irrad'd collapse to brdt; H-8 irrad'd partial collapse; H-6 irrad'd 10. 2% enhancement; H-5 irrad'd 8 5.53,dt J (7 ,8 ) = 2.i; J(8,9)=3.6; J(8,9)=4.3 9eq 2. 70,dd J(8,9 ) = 4.3; J(9,9)=14.3 collapse to brd; H-8 irrad'd 9ax 2. 4,dd J C8,95-3- 6 ; J(9,9)-14.3 partially buried under H-5 collapse to brd; H-8 irrad'd 13A 6.14,d J(7, 13) = 3.5 13B 5.56,d J( 7 ,13)~2.9 1 4A 5. 14,brs 14B 4.96,brs 15 1. 33,s 3.0% enhancement; H-6 irrad'd Ac 1.99,s OH 3.38,brs lost in D2o 45 Table 1:2 -7 ’H NMR and Double Irradiation Results (300MHz , CDC1^) for fi-Cyclolipiferolide(XII). ’h 5h» Mult. Comments 1 3.0 Buried under H-7; J(1,5)=11-9 partial collapse; H-5 irrad'd 2,3 1. 8- 2.0 undiscerned pattern 5 2. 36 ,dd J(1,5) = 11.9; J (56 ,)=11.5 collapse to d; H-1 irrad’d 6 4.50 ,dd J(5,6)=11.5; J6 (,7) = 9.1 parial collapse; H-5 irrad'd 7 3.04 buried under H-1 J(6 , 7)=9• 1; J (7,8) = 2. 3; J(7,13) = 3.5; J (7, 13> = 3-1 collapse to a dt; H-13 irrad'd 8 5.52,dt J(7,8)=2.3; J(8,9>=3.2; J(8,9 )=3 .8 collapse to a dd; H-7 irrad'd 9eq 2.82,dd J(8,9) = 3 •8; J(9,9)=14.3 9ax 2. 22,dd J(8,9)=3.2; J(9,9)=14.3 13A 6.31,d J (7, 13) = 3-5 collapse to s; H-7 irrad'd 13B 5.58,d J(7, 13) = 3 - 1 collapse to s; H-7 irrad'd 14A 5 . 12,brs 1 4B 4.93,brs 15 1.37,s Ac 2.04,s 46 10 OAc CH» * is 3 OH This is closely related to the known cyclodecene epoxide, lipiferolide(V), and may be biosynthetically derived from the latter through an intramolecular cyclization, especially since they co-occur in the tulip p op lar^ , Cyclization reactions of sesquiterpenes using Lewis acids or mineral acids are well-documented PCI ’J P.P ’ 40 . Cyclodecadiene systems such as epitulipinolide, tulipinolide and costunolide normally yield eudesmanolides. However, when applied to monoepoxides such as dihydroparthenolide and 11i13-dehydrolanuginolide, either eudesmanolides or guaianolides are obtained depending on the original location of the epoxide. When lipiferolide V_ was treated with thionyl chloride in benzene:ether (1:1) at RT for two hours under a Wg atmosphere, /3-cyclolipiferolide was isolated from the reaction mixture and found to be identical with the natural product in all respects. ^7 OAc OAc XIII Since the absolute stereochemistry of V is known, carbons 4, 6 , 7 and 8 in /3-c yc lol ipi ferol id e(XII) must have the same configuration as those in V leaving only carbons 1 and 5 undetermined. Examination of the coupling constants for protons 1, 5 and 6 indicates that most probably a trans- diaxial relationship exists between H-5 and its neighboring protons. I C I H C10 O H However, large coupling between adjacent centers may result from either a trans-diaxial or eclipsed arrangement of the vicinal protons as derived from the Karplus equation. 48 Reference to similar compounds in the literature showed that H-5 had large coupling constants in both 10, 5 a (trans- junction) ^^ ^ and 1a, 5 a (cis-junction) **** > systems. This question was most easily resolved by application of the NOE difference technique. Irradiation of H-6 caused an 8.0% enhancement of the C-4 methyl group, whereas irradiation of H-5 produced a 10.5% enhancement of H -1 and a 10.2% enhancement of H-7. Thus, H-1 and H-5 must both be alpha thereby completing the structure-proof for fl-cyclolipi ferolide(XII). OAc OH XIII In summary, fifteen sesquiterpene lactones have been isolated from L tulipifera which show possible biosynthetic interrelationships (Figure 1:2-5). This includes nine germacranolides, three eudesmanolides, one elemanolide, one guaianolide, and a trans-4-cis-9-germacranolide. The latter four classes have been shown to possibly derive from the germacranolides via acid-catalyzed cyclizations, Cope rearrangements, or singlet oxygen additions. All of the compounds possess a ^rans-a,/ 3-unsaturated-y-lactone; twelv being of the 6,7-type. And finally, although as many as fourteen different esters have been found among the sesquiterpene lactones, only the acetate ester is observed in this series and always located at either C -6 or C-8. 5 0 L = leaves OAc HOO MO OAc OAc Figure 1:2-5 Biogenetic Interrelationships of J-. tul ipi f era Sesquiterpene Lactones. EXPERIMENTAL* Isolation of q-Liriodenolide(I) and j3-Liriodenolide(II) . Finely ground dried root bark (400gms) of L. tulipifera (received from Mississippi in Oct. 1973) was extracted with 5% aqueous EtOH to yield 49gms of residue. Partitioning in the usual manner^, gave 36*5gms of 10% aqueous MeOH soluble material. When further partitioned against 30% aqueous MeOH and benzene, a 22. 8gm fraction of benzene-soluble material was obtained. Chromatography over 200gms of s ilic ic acid ( 13% H20) using i-P^O: Hex (1: 1) , i-Pr 20, and 2 , 5, 10 and 20% MeOH in i-Pr2o produced a total of 20 fractions. Fractions 9 and 10 (1.6gms) were combined and rechromato graphed over 85gms of silic ic acid using benzene-CHCl^( 1: 1), CHCl^, and 1 and 2% MeOH in CHCl^ as solvent to give a 217mg fraction. Further chromatography over 5gms of silica gel G ( i-Pr2o :CHCl^:MeOH (72:24:4)) produced a 94mg fraction which crystallized to give a-liriodenolide( 1^) . Fractions 11, 12 and 13 were combined and further separated on 170gms of s ilic ic acid. Elution with CHCl^ and 0.5% MeOH in CHCl^ gave a 771mg fraction which was further *See Appendix on p .364 51 52 purified on 70gms of silica gel G (elution with Et20:benzene:CHCl^C0.3:1i 1])» 5gms of Sephadex LH-20 (elution with MeOH) to remove phenolic compounds, and 25gms of silica gel G impregnated with AgN0^(5%) using Et20 as solvent. A fraction was obtained from which 0-liriodenolide( I_I) was obtained as flat triangular crystals from Et20 :i-Pr20 . Both compounds obtained were identical by mp, tic, [a 3D, IR and 1H NMR with the a- and yg-isomers obtained from the cyclization of epitulipinolide- 1, 10-epoxide^^. Isolation of 11, 13-Dehydrolanuginolide(III) . On ethanolic extraction of 1.424kg of Liriodendron tulipifera leaves (obtained from Mississippi in October of 1973), I82gms of residue was obtained and partitioned in a manner already described 1 to yield 58gms of 10% aqueous methanol soluble material. This was chromatographed over 1.4kg of s ilic ic acid (Mallinckrodt) using CHCl^, and 5 , 10 and 20% MeOH in CHC1 ^ as solvents to produce a 6.043gm fraction. Further chromatography over 200gms of silica gel with benzene:2-butanone ( 9: 1) as solvent, furnished a 0.627gm fraction. Trituration with diethyl ether-hexane produced crystalline 11,13-dehydrolanuginolide(III), identical by mp, [<*]D, IR, UV, ^ NMR, and MS with that reported for I I I , previously isolated from Michelia 53 lanuginosa32: refer to Table 1:2-5. The following are reported to complete the physical data for this compound: ^3C NMR (20 MHz, CDCl^) 5^ 170.2 and 169.0 (2s, CC12) and Ac), 133*9 C(10)), 127.4 (d, C(1 )) , 125.2 (t, 0(13)), 80.1 (d, C( 8)), 72.5 (d, C(6 )) , 66.5 0(7)), 47.3 (t, 0(9)), 35-9 (t, 0(2 )), 24.3 (t, 0(3)), 21.0 (q, Ac), 18.3 (q, 0(14)), 17.3ppm (q, 0(15)); CD (c=3.268x10“3M, MeOH) [ 0 3 ^ 0, [0328o "92» ^ ]270 "2U5» C0326O -337, [0]25O - 222, [0 ]242 -61, [0 ]23O -3060, C0322o -7650, [0 ]215 -3672. Isolation of Dihydrochrysanolide(IXA). The ethanolic residue (922gms) obtained from the percolation of 6.4 kg of L. tulipifera leaves was partitioned in a manner already described^ to yield 10gms of hexane:chloroform(1: 1) soluble material [obtained from 20.7% (43gms) of the total 10% aqueous MeOH fraction of 209gms]. Chromatography of this material over 350gms of silic ic acid(Mallinckrodt) using 1% MeOH in CHCl^ as solvent gave a 1 .30gm fraction. Further separation on a column of 80gms of silic ic acid using 3% 2-butanone in benzene as eluent gave a 590mg fraction. This crystallized from ether-hexane to give 123mg of pure dihydrochrysanolide(IXA) after six developments with EtOAc-Hex(2:3) on preparative layers. Overall yield from the dried leaves was 0.00928%. Dihydrochrysanolide(IXA) crystallizes from diethyl ether- hexane as colorless needles; mp 163 - 163 . 5 °; [a]D +8. 6° H-bonded OH), 1765(lactone C=0), 1739Cacetate C=0), 1 660(C=C) , 1225 C br , C-0 ); UV *MAJ£ 210nm (log«= 4.23, eOH); CD (c = 2.974x10~ 3M, MeOH) [0)295 0, [0)27O -1890, ld)252 "3970, f^^240 -1410, [0)235 01 [0^230 +7740, [0)220 +13280, [0)2io +4040; 13C NMR (22MHz, Me2C0-d6) 6C 169.9 and 169.4 (2s, Ac and C(12)), 148.1 (s, C<4)), 138.9 and 138.2 (2s, COO) and C( 1 1)) , 127.8 (d, C(5)), 123.8 (t, C£13>), 114.2 (t, C( 1 4)) , 79.3 (d, C( 8)) , 73.2 (d, C (6 )) , 70.1 (d, CO)), 49.5 (d, C(7)), 42.2 (t, C( 9)), 35.2 (t, C(2)), 31.7 (t, C(3>), 20.9 (q, Ac), 17.4ppm (q, C(15)); mass spectrum (El), m/e 306. 1474(0.42%, M+ C17H22O5 requires 306. 1467), 288(0.7%, M-H 20), 2620.7%, M-CH2C0), 24603%, M-AcOH), 228(55%, M-AcOH, -HgO), 213(22%), 200(18%), 163(11%), 135(13%), 121( 11%), 95(21%), 91(35%), 4300,0%). Horeau Ester ification of Dihydrochrysanolide(IXA) . A I3*8mg (0.045mmol) sample of IXA and 34.5mg (O.lllmmol) of a-phenylbutyric anhydride were stirred in 2. 5 ml of pyridine for 26 hours at ambient temperature under a nitrogen atmosphere. After addition of 2.5ml of H2o and stirring for 2 hours, the solution was evaporated to 55 dryness. The residue was taken up in 5ml of 5 % NaHCO^ and extracted with Et2o (4x5ml). The combined EtgO was washed with 5% NaHCO^ ( 1x 5 ml) and then H2O (3x4ml) t i l l neutral. All of the aqueous layers were combined, acidified with 1H H2SOjj to pH 2, and extracted with CHCl^. On evaporation of the water-washed chloroform phase, 24.5mg of a-phenylbutyric acid was obtained as a colorless oil Cone spot on tic) , [alp -3•65°(csO.489, c 6 H6), corresponding to an optical yield of 20%36 [calculations are based on 10. 8mg(0. 035 mm) of product since 3. 6mg of starting material was isolated after chromatography on silica gel 60 (230-400 mesh) using 60% EtOAc in hexane as solvent]. DihydrochrysanolideCIXA) Oxidation to Chrysanolide(X). DihydrochrysanolideC 11. 7mg) was dissolved in 1.5ml of CH2Cl2 and added to a column of 1. 5gm of Mn02 (Winthrop; N 264 DO): Celite(1: 2) packed in CH ^^. After one hour the column was eluted with 25 ml of spectrograde acetone which gave 11.6mg of white crystals on evaporation. Crystalli zation from CHCl^-MeOH gave 8.0 mg of chrysanolide(X) as fine needle-rosettes. This material was identical (mp, [a ]D, IR, UV, PMR, and MS) to naturally-occurring chrysanolide(X) previously isolated from Chrysanthemum cinerariaefol ium^. 56 Isolation of fl-CyclolipiferoIide(XII). The hexane-chloroform(1: 1) partition fraction of 73gms, obtained in a manner already described ^8 from leaves collected in Mississippi in September, 1975, was chromatographed through the following sequence of columns: s ilic ic acid(500gms, 70-230 mesh, Mallinckrodt, 5% H^O) using CHCl^ and 1, 2 and 5% MeOH in CHCl^ as solvents. A 6.3gm residue was eluted with 2% MeOH in CHCl^ and rechromatographed on silic ic acid (100gms, 2.5% H 20 ) with benzene-CHCl^t1:1 and 1:3), CHCI3, and 1, 2 and 5% MeOH in CHC1^ as solvents. A 3.61gm fraction was eluted with CHCl^ and further chromatographed over 52gm of Sephadex LH-20 using MeOH as solvent to remove phenolic compounds. Final chromatography over 120gms of Silica Gel 60 PF-254 using 46% EtOAc in hexane as solvent produced a fraction from which a total of 153 mg of fl-cyclolipiferolide(XII) was obtained in an overall yield of 0. 0023% from the dried leaves. Cyclolipiferolide(XII) crystallizes from ether as needles; mp 161-162°; ta^D -101°(c=0.5, MeOH); IR(CHCl^) VMAX 3590COH), 3500(br, H-bonded OH), 1772(0=0, lactone), 1743(C=0, acetate), 1670(C=CH2, lactone), 1645(C=CH2) » 1250-1210(0-0); UV *MAX 210nm (log e = 3-97, MeOH); CD Cc = 4. 706 x 10- 3m( MeOH), 0, [0]2g0 -159, m 270 - 861» ^^260 -1100, [03255 -1 158, [0] 25q -1302, [^32145 -1758, [0]24o -2762, 101225 -1^662, [03217 -20400- [^]210 "1i*025; ^H NMR(300MHz, Me^-CO-d^ and CDCl^) refer to Table 1:2-6 and 57 1:2-7; 13C NMR(22 MHz, CDCl^, 170.3 and 169-0(2s, Ac and C(12), respectively), 141.8(s, C(10)>, 134.8(s, C(11)), 122.0(t, C(13)), 110.2(t, C(14)) , 79-8(s, C(4)), 77.9(d, C(6 )), 6 6.6(d, C(8 )), 56.4(d, C(5))? 50.2(d, C<1)>, 44.2(d, C(7)), 43.7(t, C(9 ) ) , 40.4(t, C(3 )) , 26.4(t, C(2))f 24.2(q, C(15)), 20.9ppm (q, Ac); mass spectrum (El) m/e 306(0.1%, M+), 288(1.4%, M-H20 ) t 246(8%, M-AcOH), 228(29%, M-AcOH -H20), 213(11%). Analysis. Calc'd for ^^2^20: C* 6J4*?4; H, 7.35%. Found: C 64.99; H 7.18%. Cyclization of Lipiferolide(V) to fl-Cyclolipiferolide(XII). To a stirred solution of 1ipiferolide(2.5gms) in 55 ml benzene:ether(55:25) at RT under a N2 atmosphere, 2.0ml of S0C12 was added and.stirring maintained for an additional 2 hours. The reaction was quenched by addition of 20ml of 3% aqueous NaHCO^ and stirring until gas evolution was complete. The benzene layer was extracted with a total of 100ml of 3% aqueous NaHCO^, and 75ml of H20. The combined aqueous layers were extracted thrice with benzene, and the benzene layer backwashed with H2o. On evaporation of the organic layer, 2.42gms of a dark gold viscous oil was obtained which was flash chromatographed1*^ over 110gms of silica gel 60 (EM, 40-63^m) utilizing 35, 60 and 80% EtOAc in hexane as solvent. This produced a total of 15 fractions. Fraction 12, 228mg, 58 crystallized from EtOAc:Hex to give 113mg of fi-cyclolipiferolide(XII) which was identical with the 1 naturally occurring material by mp, co-tlc, [«Od, IR, UV, H NMR, CD, and MS. T ftAM SM IlltON i u e :- IHFigure Spectrum 1:2-6 (CHC13) a-LiriodenolideU) of VO ui 90 MHz n ______i -111II 1111 i. i i 11 l.i 11) 11111.1111111111 ix lIli llI 111111 m 1. 11 ■ I ■ 111111, 11,,,, i,, 7 6 5 4 3 2 1 PPM figure 1:2-7 H NMR Spectrum (CDC1,) of a-Liriadenolide (I) . O' o rttCLNT TlANfcMLSfclON i u e :- IR SpectrumFigure 1:2-8 (CHCl^) of ff-Liriodenolide(II). 9 0 MHz J 5 £ . A w J L _ a J NL_ jJLjl LLi Li-1 LlI LI I I I I I I I I I I I I I I I I I I I I I I I II 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 I I I 1 I I I I I I I I I I PPM figure 1:2-9 1H NMR Spectrum (CDC13) of fi-Liriodenolide(£1) o u> WMYtNUMICt c tr* iue :-0 RSetu (Hl) f 11-eyrlnqnld(I). of 11,13-Dehydrolanuqinolide(III) (CHCl^) 1:2-10 Spectrum Figure IR O ltim tN V U 90 MHz Figure X:2—11 NMR Spectrum (CDCl^) of 11, 13-Dehydrolanuginolide(III) . 2 0 MHz nmp i iiiiinjriiiumjirminn mm iii|ininin|nimni|imiiiii[ii»iirir]nMiiiiiniii ii ni|ininnT|i i iiiiinji iifrTTn|iiiimTi|ni mni|ii iiinmm urn i|iminn|inmnn 111111171 100 . ISO 100 50 PPM figure 1:2-12 NMR Spectrum (CDCl^) of 11,13-Dehydrolanuqinolide(III). cr* 6 6 -In - 2 - 1 -3- O X -4- - 6 - -7- 220 2 0 0 230 260 2 9 0 nm Figure 1:2-13 CD Spectrum (MeOH) of 11, 13-Dehydrolanuginolide(III) . PftCCNT /o •0 vr-m. Figure X: IR Spectrum 2-14 (CHCl^) Dihydrochrysanolide(IXA) of . f TTT i? if 2 . ni: w i w» n NUMll CM lfl M U tN V A W mo -1 uoo r r i r no* «D0 I,-, - ■ , TI ~ Itfij w ■ -a o\ 9 0 MHz J__ _ / v V _ I 1 i I—» L i J *- x 1 j 7 I i i i i I 2-15 1H NMR Spectrum (CDC13) of Dihydrochrysanolide(IXA) . cn Figure I co 20 MHz iiii[iiniMi^iiiin iii|iiiin n i|n iiiiiiniiiiiiin |iiiiiiiii|iiin iin[im ii>rniiiin iii|iTiiim r|iin iiiiniiin iiii]iiinnri[iiiiinir]i)i)in ii|iimTiii|iin iiiii[uiiMiii|iiinTrn|iTn'iiiii 300 iso 100 so PPM Figure 1:2-16 13„ C NMR Spectrum (CDCl-j) of Dihydrochrysanolide (IXA) o iue :-7 D pcrm MO) f Dihydrochrysanolide (MeOH) of CD Spectrum 1:2-17 Figure [6] x io -4 12 8 - - 220 M ortA 13300' + 240 260 70 tttCfNT TtANiJrtllStON i u e :-8 IRFigure Spectrum 1:2-18 (CHC13) of ff-Cyclolipiferolide (XIII) H 14 300 MHz H 19 H-7 H #i M®j AJL ■I— t — I— 1 t I 1 . I . I I 1 1 1 J.mI.. I . J—I -L-, <,„L. L..I. I I ‘ I I I I ■ 1 L ■ i • ■ 1 i_i, 6 6 4 3 2 Figure 1:2-19 1ti NMR Spectrum (CDC13) of fl-Cyclolipiferolide (XIII) . i\> 20 MHz |injininimiiiiijniiinnjiiiiii>njiiiiimi|iMnmqHiimii|iniiiMi[)tiinMi|niiiiiinnii)ini|MiiiiiiniiimiiTpiiniii[ninini)iininmimrnTTjmmin 150 100 50 PPM Figure 1:2-20 NMR Spectrum (CDCl.) of fl-Cyclolipiferolide(XIII). —j U) 74 -1100 'Ixn M 260 290 230 -20400 [i$] of j8-Cydol ipi f erol ide (XI11) . (XI11) ide erol f ipi of j8-Cydol figure 1:2-21 CD Spectrum (MeOH) figure 1:2-21 Spectrum CD — 10 -5 — -5 -15 - -20 01 x [e] Chapter 3: Isolation and Id en tification of Novel Cyclization Products of Lipiferolide, the Major Sesquiterpene of Liriodendron tulipifera. INTRODUCTION In a continuing phytochemical investigation of the sesquiterpene content of Liriodendron tulipi fera ( Magnoliaceae), commonly known as the tu lip poplar, a new sesquiterpene lactone had been isolated from the leaves. Initial work showed that the proposed structure for /3-cyclolipiferolide( I_I) was closely related to the cyclodecene epoxide, 1ipiferolide(I_), the major sesquiterpene lactone isolated from the leaves. As part of the study concerning its relative and absolute structure elucidation, it was necessary to establish the stereochemistry at the carbon bearing the hydroxyl, the ring junction, and the a,£j-unsaturated lactone. Treatment of lip ife r o lid e with thionyl chloride produced £-c yclol ipiferol ide (I_I) from the acid-catalyzed cyclization of _I. Thus, it would be possible that £-cyclol ipi f erol ide is biogenetically derived from lipiferolide since they co occur in j.. tulipifera . 75 76 14 OAc 10 OAc 13 CH 15 However, in addition to g-cyclolipiferoiide(II), numerous other products Ccyclized and uncyclized) were isolated. Their structure proof, and a comparison of the products formed under different conditions are presented herein. These compounds are of interest relative to their structural novelty, spectroscopic identification, and mechanism of formation. In addition, they offer clues to minor sesquiterpenes in L. tu lip ifer a which may be- biogenetically derived from 1ipiferolideCI). DISCUSSION AND RESULTS Cyclization reactions of the sesquiterpene lactones using Lewis or mineral acids are well-documented in the lite r a tu r e ^ * ^2, 34, 35, 40^ Cyclodecadiene systems such as epitulipinolide(III), normally yield eudesmanolides from a Markovnikov-type attack of C-5 on C-10 with the generation of an intermediate carbonium ion (Figure 11:3-1). However, when applied to the epoxide systems, either eudesmanolides, or guaianolides are obtained depending on the location of the original epoxide. The mechanism involved in the opening of dihydroparthenolide(y), or 11,13-dehydrolanuginolide (VI) is of the 3^2 type with attack of C-l on the secondary C-5 center. However, in the 1,10-epoxide, such as pyrethrosin( TV), attack is of C-5 on C-10, a tertiary center. The la tter s t i l l proceeds through an SH2 mechanism, but now has a high degree of S^i character. A change in the direction of ring opening in the ambident epoxide substrate is known to be dependent upon the reaction conditions (i.e. acid v_s base). The reason for the preferred attack at C-5 in the 4,5-epoxide is unclear in view of the fact that acidic reagents have been used in each situ ation . 77 78 H OAc OAc soci 2 HO -O TsOH OAC OAc IV BF? EtOEt r OH OAC OAc Figure 1:3-1 Cyclization Reactions o£ Known Sesquiterpene Lactones. 79 Therefore, both types of attack may occur in the 4,5-epoxide under the proper conditions, although some products may be minor. The fact that dihydroparthenolide(V) and 11 , 13-dehydrolanuginolide(V_I) are structurally related to lipiferolide indicated that at least the alpha isomer should be obtained from the cyclization reaction. The solution conformation of lipiferolide( 1^) is known to be of the syn- chair type and which should theoretically yield three possible products when cyclized: the a:(3:Y double bond isomers in a ratio of 1:3:2 (Figure 1:3-2). When lipiferolide(I_) was treated with thionyl chloride in benzene:ether at room temperatureunder a nitrogen atmosphere for one hour, not only were the a- and fi- cyclolipiferolide isomers obtained, but also a variety of cyclized and uncyclized products (Figure 1:3-3) which inclu de: a) three [6:2:0] bicyclodecane systems b) a chlorohydrin germacranolide c) two cyclodecadiene germacranolides, and d) three isomeric guaianolides The compounds are listed with their respective yields and in order of polarity r ela tiv e to an EtOAc:hexane tic system. The major product is compound XIII which is consistent with that predicted from the results of dihydropartheno- lide(V) and 1 1 , 1 3-dehydrolanuginolide(VI^). Unlike the Me 80 Me SYN Me H OH Me Me Me ANTI OH OAc OAc Me OH Me AcO Me Figure 1:3-2 Cyclization Products of Syn and Anti Sesquiterpene Lactones. Me OAc OAc Vtl 0.85% VIII 8.6% OAc OAc .OAc M« Me X 5.1% XI 9.8% XII 4.5% M* Me ,C> OAc OAc OAc « • i)H XIV -8.5% Figure 1:3-3 Acid Catalyzed Cyclization Products from Lipiferolide. 82 previously cited literatu re reactions in which yields were generally low (10 to 40%), these products accounted for 82% of the mixture. Most of the compounds are quite labile and break down to an insoluble polymer on continued handling; a typical property of the sesquiterpene lactone systems. The major product, compound X III, would not c r y sta lliz e and decomposes upon solvent evaporation above 5° with or without N2. It is only stable in solution at 0°C. In addition, most of the cyclization products in the literature were isolated using argentation chromatography (i.e. AgNO^ adsorbed onto silica gel). These compounds were isolated using only s ilic a gel since they rapidly decomposed on silv er nitrate tic plates which may account for the low yield cited in the literature. It is believed that all of these compounds are unreported, and the [6:2:03 bicyclodecane system is novel and unknown to the class of sesquiterpene lactones either chemically or biogenetically. Formation of the [6:2:0] bicyclodecane systems can be rationalized by a nucleophilic attack of C-1 on C-4 as opposed to C-5 which leads to the guaianolide se r ie s. This is similar to that occurring in the illustrated 1,10-epoxide systems. Formation of a carbonium ion and subsequent loss of a proton from C-1 or C-14, or attack by chloride gives rise to the isomeric mixture (Figure 1:3-^). Compound _I)£, the most abundant of the [6:2:0] bicyclodecane systems, showed a molecular ion at 3^2. This CH OAc CH tOAc CH OH +CI CH iOAc CH CH CH OH OH OH VII Figure 1:3-4 Formation of £6:2:03 Bicyclodecane Systems. 00 (JO 84 Table 1:3-1 MMR Data for the [6:2:0] Bicyclodecane Compounds (300MHz, CDCl^, TMS). V II IX X 1 'm ™ “ 2 .9 7 , dd , 1H 2 .8 8 , dd (8 .3 ,1 0 .7 ) (8.9) 2 2 .5 7 ,dd 2 .3 9 ,dd 1.9-2. 1 ,m (8.4) (9.2,1 0.7) 3 2.1 0 ,m 1 . 7-1 .9,m 1.6-2. 1 ,m 1.87,m 5 3 -73,d 3■57,d 3-70,d (9-9) (9.9) (9.9) 6 4 .63,dd 4 .4 9 ,dd 4.59,dd (6 .6 ,9 -9 ) (6 .6 ,9 .9 ) (7 .3 ,9 .9 ) 7 3•22,m 3•49,m 3.31 ,m (2 .9 ,3 .3 , (2 .9 ,3 .3 , (2 .6 ,2 .9 , 3 .7 ,6 .6 ) 3 .4 ,6 .6 ) 3 .3 ,7 .3 ) 8 5. 3,dq 5. 77 ,dq 5.38,dq (3 -3 ,6 .3 , (3 -4 ,5 .5 ,' (2 .6 ,4 .4 , 11.4) 10. 5) 9.1) 9ax 2 .2 7 ,dd 2 .2 1 ,dd 2 .6 5 ,dd (6 .3 ,1 2 .8 ) (5.5,15. 1) <9.1, 14.3) 9eq 2.3 8 ,dd 2 .0 8 ,dd 2 .4 9 ,dd (11.4, 12.8) (10.5, 15. 1) (4 .4 ,1 4 .3 ) 13c 6 .3 5 ,d 6 .3 9 ,d 6.36 , d (3-7) (3-3) (3.3) 13t 5 .5 9 ,d 5 .7 1 ,d 5 .6 5 ,d (2.9) (2.9) (2. 9) 14 1.60,dd 1.70,s 5.05 , s (1.8) 4 .8 4 ,s 15 1,36,s 1. 31,s 1.22, s Ac 2.00, s 1.98,5 2.01,s OH 4.81 ,brs 4 .7 8 ,brs 2 .6 3 ,br 90 MHz -OAc H -IS H -14 H 131 mac u u v _ 1 I 1 > t I t I I 1 1 t 1 1 » 1 I I I I I > I- 1 - * I I.II I 1 II 1 I.Lll I I I H I I I I I I I I I I I I I 111 d LI I I I 6 5 4 3 2 1 o PPM Figure 1:3-5 H NMR Spectrum (CDC13) of Lipiferolide. was supported by the chemical ionization mass spectrum showing peaks at 343 and 345 in a 3:1 ratio consistent with the presence of chlorine. In the NMR spectrum (Table 1:3-1). a downfield pair of doublets is seen with allylic coupling similar to that observed for lipiferolide (Figure 1:3-5) indicating the presence of the C-13 and C-7 protons. Proton seven appears at 3-49ppm as a doublet of tr ip le ts , and H-8 is seen as a doublet of quartets at 5.77ppm. The latter represents the X portion of an ABX pattern with the C-9 protons corresponding to the AB portion appearing as a pair of upfield double doublets. These collapse to a distinct pair of doublets in a double irradiation experiment of H-8 at 300MHz. The fact that the residual patterns are doublets indicates geminal coupling alone and the lack of any vicinal protons at C-10. Therefore the C-9 protons must be next to a quaternary center. As mentioned, K-7 is coupled to the allylic C-13 protons and must also be coupled to H-6 (4.49ppm) and H-8 (5.77ppm). This has been verified through decoupling experiments. Since H-5 is a doublet and is coupled to H-6, it must be either 90°(£=0) to a second proton or next to a quarternary center. Since n e i t h e r methyl appears as a doublet, this rules out the former possibility placing it next to a quarternary center. This also accounts for the high field methyl singlet which has an 48 appropriate chemical sh ift for a bridgehead methyl group * The low field methyl (1.70ppm) is too sharp to be olefinic 87 but its chemical shift is indicative of a methyl group on a carbon bearing a halogen (1.6ppm) or oxygenated substituent C1 . 4ppm) . The remaining signal, an important feature in the proposed structure, is the double doublet at 2.97 which integrates for one proton and is assigned to H-1. As expected from its isolated position in the cyclobutane ring, there was no effect on any of the downfield protons on irradiation of H-1. The only changes were seen in the 1.8-2.5ppm region. All of the above results are consistent with the proposed structure, and the absolute stereochemistry of carbons 5, 6, 7 and 8 are known from 1ipiferolide. The carbon spectrum (Table 1:3-2) shows two singlets at 170ppm for the two carbonyl carbons and only two absorptions in the olefinic region: a singlet and a triplet corresponding to carbons 11 and 13* In the region from 70-85ppm there are four absorptions indicative of carbons containing an oxygen or halogen substituent: three doublets for C5, C6, and C8, and a sin g let for C-10. Other absorptions include three quartets (20-33ppm), and three tr ip le ts at 20.9, 33.0, amd 43.8ppm (C-9, C-3 and C-2, respectively). Doublets for C-7 and C-1 are found at 44.2 and 50.3pptn, resp ectively, whereas C-4 appears at M5.3ppm as a sin g le t. 88 Table 1:3-2 NMR Data for the [6:2:0] Bicyclodecane Compounds3 . VIIb IX xb 1 141.5,s 50.3,d 4 4 .7 ,d ( J=130Hz) 2 3 5 .0 ,t 4 3 .8 ,t 41 . 5 , t (Jsl42Hz) 3 3 0 .8 ,t 3 3 .0 ,t 33•0,t ( J = 145Hz) 4 47.3,s 45.3,s 43.7,3 5 80 .2 ,d 83. 1, d 8 3 .4,d 6 79- 1,d 7 7 .8 ,d 79. 1,d 7 44.8,d 44.2,d 45.5,d 8 69 .7 ,d 6 9 .6,d 70. 8,d 9 23. 8,t 20.9,t 23.7,t 10 125.3,s 7 0 .0 ,s 142.6,5 1 1 135-4,s 134.8,s 136.4,s 12 168.8,5 168.5,s 168.3,s 13 123.3,t 124. 1, t 123.2,t 14 18.4,q 32.8,q 116.5,t 15 18. 1,q 13.9,q 14.5,q C=0 170.5,s 169.7,s 169.9,s CH3 21. 1,q 2 0 .8 ,q 2 0 .8 ,s aSpectra obbtained at 20MHz in CDCl^. ^Assignments are substantiated by SFORD experiments. 7% 23% H-1 300 MHz H-5 H-7 91(364) 84 26 1 L I 1______|______|------1------1------1------1------L J I 1 L PPM Figure 1:3-6 NOE Difference Results for Compound X. 90 The stereochemistry of the substituents at carbons 1, 4 and 10 has been determined to be a, £ , and /3 (Me) , respectively, as determined by NOE experiments. Irradiation of H-1 caused a 23% enhancement of H-7 and a 7% enhancement of H-5 (Figure 1:3-6). Irradiation of H-15, however, caused Nuclear Overhauser enhancements of H-6, H-9 and H-14. Compound X, which contains an exocyclic double bond, is very similar to the compound just described. Identical patterns are seen for the 5, 6, 7, 8 and 13 protons. The acetate and bridgehead methyl are still present but the lowfield C-14 methyl is gone. The appearance of two broadened sin g lets at 4.84 and 5.05ppm which integrate for one proton each is typical of non-conjugated exocyclic methylene protons. The presence of this functional group is verified by a 1645cm**1 peak in the IF spectrum; conjugated 1 exocyclic methylenes usually appear at 1665-1675cm“ . Also, the 1^C NMR spectrum contains an additional triplet and singlet in the olefinic region at 116.5 and 142.6ppm. The [6:2:0] bicyclodecane system containing an endocyclic double bond at carbons 1 and 10 again had similar absorptions for the 5, 6, 7, 8 and 13 protons. The 1H NMR spectrum of compound VII contained an olefinic methyl group at 1.6Gppm (brt) as well as the usual acetate and bridgehead methyl absorptions. There were no olefinic proton absorptions in the 1H NMR spectrum other than the exocyclic methylene protons of the lactone. However, the 1^C NMR 91 3pectrum contained a tr ip le t (123*3ppm) and three sin g lets (141.5t 135.4 and 125.3ppro) in the o le fin ic region. The only way to satisfy these requirements is to have a tetra- substituted olefin at carbons 1 and 10. This is confirmed by the loss of the absorption for H-1 at 3.0ppm as seen in compound IX and X. The next group of compounds isolated were formed by direct opening of the epoxide with concomitant loss of vicinal proton or attack by chloride (Figure 1:3-7). Their structures were most e a sily elucidated by comparison of the NMR spectra to that of 1 ipiferol ide(1^) . Compound XI_ belongs to the unique class of sesquiter penes which have more than one conformer at room temperature. The NMR at room temperaturein either or CDCl^ produces very broad patterns. At -30°, however, we find two distinct sets of peaks. The conformers are in a 46:54 ratio having a free energy difference of 77cal/mole. Other examples of multi isomers are laurenobiolide (5 : 4: 3: 1) ^9 1 50 ( isabellin (10:7)-^ and neolinderal lactone (8:2)^. The latter has a potential barrier of 600cal/mole indicating a d efin ite preference for one conformer over the other (8:2) which is not the case in the present example. Heating Xl_ to 70° causes the signals to coalesce giving rise to an averaged set of absorptions (Figure 1:3-8). It is now possible to observe that the acetate and olefinic methyl signals are still present and that one methyl is CH OAc OAc CH ?CI* OAc OAc OAc CH OH VIII XII Figure 1:3-7 Products Formed on Epoxide Ring Opening. OJ w r i ju______A j «w wn j v __ a A—^ ^ L jLfLjjL JUJL. r w A i i L nci. I 1 ll IllJ I I 1 1 I I I I I I I I I I I I LI I 1 I I I I I I I I I 111 I I I I I I I I II I I I L I l l J l L l l J IJ I I I I II J.I 7 114 771 1 1 1 * Figure 1:3-8 Variable Temperature Studies for Compound XI_ vO UJ absent. The exocyclic protons are still present with allylic coupling to H-7. However, in addition, there is long range coupling of H-13c to H-6 as determined from double irradiation experiments. When proton six was irradiated, H-5 collapsed to a singlet, H-7 to a doublet of tr ip le ts and H-13c to a doublet. Likewise, when H-13c was irradiated, a double doublet was seen for H-6. Such long range coupling through five bonds is a common phenomenon observed with conjugated systems, homoallylic protons, uft acetylenes, and small ring systems0. Approximate changes in the conformation of th is molecule can be deduced from the spectrum obtained at 243°* coupling constants for 6A and 6B are almost identical as are those of 5A and 5B. Also, a large change in coupling of H-6 to H-7 isn't observed in either conformer, indicating that the C-5, C-6, C-7 system is rather rigid in comparison to the rest of the molecule. The large chemical shift difference of H-6 is explained by the positioning of the acetate. The acetate substituent is normally in a pseudoaxial position. If it were to move away from H-6, the proton absorption of the latter would move further upfield in the m in o r conformer as observed. At the same time, however, the acetate approaches the H-13t proton, causing the latter to shift further downfield as does the acetate itself from the increasing anisotopric effect of the double bond. The H-13c proton, because of its isolated location on 95 the other side of the double bond, is little affected. A large change is also observed for the C-10 methyl group (0.2ppm dow nfield). Further studies in d ifferen t solvents, use of shift reagents, NOE experiments, and variable temperature CD experiments would be necessary in order to fully elucidate the two conformations. A second product of direct epoxide opening is compound XII. The NMR (dg-acetone) of XII showed the downfield pair of doublets for the C-13 protons and two olefinic methyls as broadened multiplets at 1.78 and 1.69ppm CH-15 and H-14, resp ectiv ely ). The assignment of these peaks was established through decoupling experiments by irradiating the two o le fin ic protons at 5.92 and 5.28ppm (H-1 and H-3, respectively). The additional splitting of the olefinic methyl results from coupling with the diallylic C-2 protons at 2.8ppm. Also, H-6 appears at 4.43ppm as a doublet, and H-8 at 5.39ppm as a doublet of triplets. proton five (brd, 5.09ppm) which is coupled to the hydroxyl proton, collapses to a 3inglet on addition of D20. Since both H-5 and H-6 appear as doublets, they must be conformationally oriented at approximately right angles to each other. In the 1^C NMR, compound XII exhibited six absorptions in the olefinic region: three sin g lets (1*40. 5, 138. 5 and 130.5ppm), two doublets, (13*4. *4 and 122.0ppm), and a triplet (122.0ppm). NOE studies gave a 12% enhancement of H-3 when H-15 was irradiated, indicating a c is relationship between the two. 96 Also, a 13X enhancement of H-6 was observed when H-14 was irradiated, demonstrating that they are on the sam e side of the molecule. Although naturally-occurring compounds have been isolated with double bonds at positions 1 and 4, none could be found in the literature at 1(10) and 3. Compound VIII is a chlorohydrin formed by attack of chloride at the epoxide through an SN2 mechanism. The mass spectrum (El) contained a molecular ion at 342 and showed a loss of HC1 [m/£ 306 (3%)]* Peak ratios between m/e 342:344 (2.5:1) and 282:284 (2.7:1; loss of HOAc) verified the presence of chlorine. The 1H NMR contained a broadened 3H singlet at 1.85ppm typical of the olefinic H-14. The acetate is present at 1.98ppm and the H-15 methyl at 1.59ppm (ind icative of a methyl on a carbon bearing an oxygenated or halogen substituent. Proton six is a doublet at 4.69ppm indicating a lack of coupling to H-5. The H-5 and hydroxyl proton form an AB pattern at 80MHz (dg-acetone) which collapses to a 1H singlet at 4.41ppm on addition of D20. The 13C NMR contains only four olefinic carbons: two singlets, a doublet and a triplet. The region from 74-80ppm contains four absorptions. Three are doublets assigned to carbons 5, 6 and 8. The fourth is a singlet and assigned to C-4 which bears the methyl and chlorine substituents (79.3ppm). The orientation of the chlorine is most likely beta resulting from backside attack at C-4 of the epoxide. Since the conformation of the flexible cyclodecene ring is 97 not known, a technique like NOE could not be used to establish the conformation at C-4. Hence the configuration at C-4 remains open. Also, attempts to recyclize to the epoxide failed (aq. NaOH, in acetone, and KgCO^ in MeOH). This negative evidence could support the ^-orientation since backside attack to the chlorine would be d iff ic u lt to achieve for the hydroxyl group and displacement by free alkoxide or other nucleophile may be more competitive. The uncertainty of the ring conformation precludes any further statement about the results of these experiments. Returning to the initial reaction, a series of guaianolides are obtained if attack occurs at C-5 rather than at C-4 which resulted in the [6:2:0] bicyclodecane systems (Figure 1:3-9). Compound Ij[ was identified as /3-cyclol ipi ferol ide by comparison of all physical constants to those of the naturally-occurring compound. Compound XIV contained similar patterns for protons 5 , 6 , 7, 8, 13c, 13t, 15 and the Ac. The major difference was the disappearance of the two broad sin g lets at 5.11 and 4.92ppm, assigned to the exocyclic H-14 protons in the /3-cyclolipiferolide, and the appearance of a methyl singlet at 1.71ppm. The chemical ionization mass spectrum contained an M+H peak at m/£ 343 indicating the presence of chlorine. The final compound, guaianolide XIII, also contained similar patterns for protons 6 , 7, 8, 13c, 13t , 15 and the OAc CH OAc CH + CI -H CH OAc OAc OAc CH CH CH OH OH OH XIV XIII Figure 1:3-9 Guaianolide Formation from Lipiferolide. VD CO acetate. Differences here again involved the loss of the two H-14 exocyclic protons and replacement with a 3H methyl. However, the peak at 1 . 66 ppm is broadened and indicative of an olefinic methyl. Initial evidence suggests that this is a similar situation to that observed for the 1(10)-endocyclic [6:2:0] bicyclodecene compound (VII) . Thus, H-5 appears as a broadened doublet at 2.79ppm and found to be coupled to H -6 having a coupling constant of similar magnitude as seen in /9-cyclolipiferolide. The broadening of the H-5 pattern is the result of long range coupling to the C-10 methyl as verified by decoupling experiments. The nine products obtained from the cyclization of 1 ipiferolide(_I) were found to be the same by tic comparison (nine standards and both reaction mixtures) whether freshly distilled or reagent grade thionyl chloride was used as a catalyst. Treatment of lipiferolide(I^) with dry HC1 in benzene:Et20 (1:1) at room temperaturefor 12.5 hours gave six products which were isolated and found to be identical with those produced by the treatment of 1 ipi ferol ide ( 1^) with thionyl chloride. Comparison was by mp, tic behavior, IR and NMR. The yields are listed in Table 1:3-5. These products ( 93% yield), however, do not include any of the [ 6 : 2: 0] bicyclodecane systems which constitute 17% of the total material recovered from the SOCI 2 catalyzed reaction. It is apparent from the large difference in reaction times 100 ( 1hr v_s 12hrs) that it is not only the HC1 present in undistilled SOCI2 that catalyzes the reaction, but the SOCI 2 itself. As a result, thionyl chloride <1) significantly increases the rate of the reaction, ( 2) decreases the yield of the major product (XIII) by 48%, and (3) increases the i character of the reaction giving rise to additional cyclization products. Since thionyl chloride profoundly influenced the course of the cyclization with respect to HC1, the question was raised as to the nature and ratio of the products formed under the influence of BF3-Et20 . Lipiferol ideU) was treated with BF^*Et20 at 0°C for 2.5 hours in an inert atmosphere. The reaction mixture gave crystalline fluorohydrin(XV) as the major product (total yield 68 . 8%). CH ,OAc 4 The H NMR spectrum appeared to be similar to that of the chlorohydrin obtained from the S0C1 2 reaction. However, the presence of the fluorine at C-4 increased the multiplicity of the surrounding protons since it has a nuclear spin of 1/2. The C-4 methyl group appears at 1.38ppm as a doublet 101 Table I;3 _3 NMR Data for BF^'ET^O Cyclization Products (300MHz, CDCl^, TMS)a . XV XVI XVII 1 5 .6 3 ,dd 5 . 02,brd ^ ™ ™ (7 .4 ,7 .4 ) (5.9) 2 2 . 2- 2 .4 ,m ----- 2. 34,brt (1,7.7) 3 ------2 .57,d t (3- 1,7-7) 4 ----- 2.7 3 ,ra —- (3 .4 ,7 .4 ) 5 4 .3 8 ,dd ----- 5 .7 6 ,dd (7 .4 ,1 1 .8 ) (2. 2, 1) 6 4 .5 6 ,dd 5 .22 , d 5 .2 8 ,dd (<3, 2. 2) (4.8) (2. 2, 10, 2) 7 3.19,m 3.66 ,m 2 .9 0 ,dq (< 7.5,2.2, (2 . 3, 2. 6 , (2 .4 ,3 .2 , 2. 6 ,<3) 3 .3 ,4 .8) 3. 6 , 10. 2) 8 5 .3 4 ,m 5 .4 6 ,m 5 .7 4 ,ddd (<3,3. 1, (2. 3, 2.4, (2 .4 ,3 -7 , 3.4) 3-7) 4.8) 9A 2 .4 5 ,dd 2. 58 ,dd 1. 90,ddd (3 .1 ,1 4 .1 ) (3-7,14.3) ( 4 .8 ,4 . 8, 15.5) 102 Table 1:3-3 (continued) 9B 2.6 0 ,dd 2.26,dd 2.04,ddd (3.4,14. 1) (2.4 ,14.3) (3 .7 ,3 .7 , 15.5) 10 — — * — — — 2.4 2 ,m (3. 7,4. 8, 7.7) 13c 6 .34,d 6.38,d 6.22 , d (2. 6 ) (3.3) (3-6) 13t 5 .7 1 ,d 5.75, d 5 .5 0 ,d ( 2. 2) (2. 6 ) (3*2) 14 1.78,brs 1.83,brs 1. 21,d (7.7) 15 1 * 39,d 1.3 3 ,d 2 .1 8 ,s (22.4) (7.4) Ac 2.03,3 2.06,s 2.04 ,s HO 2. 17,d ...... (7 .4 ;lo st in D20 ) aCoupling constants are given in parenthesis in Hz. 103 with a coupling constant of 22.5Hz. Proton five appears as a double doublet at 4.37ppm which is coupled to both the hydroxyl qh=7. 9Hz) and the fluorine U 5 (p= 11. 4Hz). Addition of D^O causes collapse of this pattern to a doublet. The C-6 proton appears as a triplet-like double doublet with coupling of 2.9Hz to both H-7 and the fluorine. Other features of the NMR spectrum are virtually identical with that of the chlorohydrin (VIII). Further verification of the presence of the fluorine was obtained from the NMR spectrum in which C-4 appears as a doublet (J^=171Hz) in the broad band decoupled spectrum. Also, carbons 3, 5 and 6 are seen as doublets due to long range coupling (36-0, 77.1, 75-Oppm, respectively). In addition, a peak at m/e 326 is observed in the electron impact mass spectrum consistent with the presence of fluorine. A second compound obtained after chromatography of the mother liquor on 55gms of s ilic a (30% EtOAc in hexane) gave epitulipinolide(III) , a known germacranolide isolated from the rootbark of Liriodendron tulipifera. The compound was obtained in 1. 0% yield and was identified by comparison to 104 an authentic sample (mp, c o -tlc , IR and NMR). Continued elution with 30% EtOAc in hexane gave compound XVI as needles from EtOAc in hexane. This compound was found to contain an additional carbonyl peak in the IR spectrum at 1713cm- ^ characteristic of a ketone. The NMR spectrum contained similar absorptions for the 1, 7, 8, 9, 13, 15 and acetyl protons as compared to 1 ipi ferol ide(_I). These assignments and those following were verified with double irradiation experiments. The C -6 proton appears as a deshielded doublet at 5.22ppm y=4.8Hz) resulting from the adjacent ketone. The C-4 methyl group is a doublet at 1.33ppm CJjj^15 = 7. 4Hz) which requires a vicin al proton H-4. Irradiation of H-4 (2.73ppm) caused collapse of H-15 to a singlet. Likewise, irradiation of H-15 sharpened the C-4 proton. The ^C NMR data is presented in Table 1:3-4. Further elution of the original column with 50% EtOAc in hexane gave 59mg of the fluorohydrin(XV) followed by a new 14 CH ,OAc OAc OAc 15 "0 13 CH XIX 105 Table I:3~4 NMR Data for BF^*Et 20 Cyclization Products XV XVII XVI 1 130. 0,d 145.4,s 131.6 ,d 2 21. 0, t 32. 5 , t 36 . 0,t 3 36 . 0,dm 4 1 .9,t 2 4 .7 ,t (d,24Hz)a 4 99.d 20 7 .3 ,s 43.0,d ( d,117Hz) 5 77. 1,dd 126 . 2,d ------(d ,12.9Hz)a 6 7 5 .0,dd 7 5 .6 ,d 7 5 .6 ,d Cd,10.8Hz)a 7 4 9 .2 ,d 50 . 2,d 4 6 .8 ,d 8 7 8 .8 ,d 68 . 5 ,d 7 3 .6 ,d 9 43.9,d 3 5 .9 ,t 43 .2 ,t 10 136 . 1 ,s 36 . 6 ,d 134.9,3 11 135. 1,s 135.7,s 135.4,s 12 170.7,3 169.3,s 168 . 4,s 13 123.9,t 124.3,t 124. 1,t 14 19.3,q 19- 1,q 1 9 . 0 , q 15 1 9•9,d q 30. 0,q 15.3,q ( d,24Hz) Ac,CsO 170.7,3 170. 1,3 169. 8, s Ac,CH3 21. 0,q 21. 0, s 21. 0,s aPattern observed in broad band decoupled spectrum. 106 xanthanolide (seco-guaianolide) sesquiterpene lac tone(XVII). This compound is formed by opening of the epoxide at C-5, generation of a ketone by breaking the bond and an olefin at C-1 to C-5 with a hydride sh ift from C-1 to C-10, The IR spectrum contains a third carbonyl peak at 1720cm"^ corresponding to the ketone which is verified by a 207* 2ppm absorption in the 13C NMR. The mass spectrum, (El) shows a molecular ion at m/e 306. The NMR spectrum (300MHz, CDCl^) contains doublets for the exocyclic protons of the y-lactone at 6.22 ^ 3c = 3* 6Hz) and 5.50ppm (J.7 ( i 3t = 3 * 2Hz) . These protons are coupled to H-7 at 2. 90ppm and collapse to singlets on irradiation of the latter. In addition, H -6 (5.28ppm, dd) and H-8 (5.7*1, ddd) collapsed to a sin g let and a double doublet, resp ectively. Proton five (5.76ppm, dd) couples to H-6 (2.2Hz), H-2 and H-9- The C-10 methyl signal is split (£=7.4Hz) into a doublet which requires a vicinal proton. An unusually lowfield 3H singlet is observed at 2. l 8ppm which is consistent with the presence of a methyl ketone and confirmed by the formation of a yellow precipitate in an iodoform test. The 13C NMR contains three carbonyl carbons and a di- and tri-substituted double bond: [145.4(3), 135.7(s), 125.2(d) and 120.6 (t)ppm]. Two oxygenated carbons appear as doublets for C-6 and C-8 (76.9 and 75.6ppm) in addition to the three methyl quartets (30*0, 21.0 and I9.1ppm). Thus, the above information is consistent with the structure for a 107 member of the small class of naturally-occurring seco- guaianol ides^. Separation of other products obtained from the reaction mixture led to the isolation of the double conformer (XI) , g-cycloliplferolide(XIII) and 0-cyclolipiferolide( IJ) in 1.3, 6.2 and 2.9% y ield s, respectively. These compounds were shown to be identical with those obtained from the 1 SOCI2 catalyzed reaction by up, co-tlc, IR and H NMR. In summary, lipiferolide readily transforms into a variety of cyclized and uncyclized products under different reaction conditions (Table 1:3-5). The ease with which the transformation occurs readily explains why the reaction sequence outlined for the synthesis of simsiolide acetate in* Figure 1:3-10 was unable to be completed. Mild enough conditions could not be found to prevent the formation of numerous side products in any attempt to remove the protecting groups. A major product (excluding halogenated species) in all reactions was a -cyclolipiferolide, a very labile compound which rapidly decomposes at RT. This may serve a 3 a possible explanation for its not being isolated as a natural product from L. tulipifera along with the 0 -isomer found in the leaves. Thus, evaporation and further fractionation of the plant extract without knowledge of its extreme sensitivity would lead to its total destruction. 108 PAc OH 1)OH 2) H+ P.G. Lipi ferolide o- z O-Z OH OH COO" Ac20 pyr O-Z Cleave P.G. 2) H .COOH OAc Simsiolide Acetate PROTECTING GROUPS ( P.G.) 1) CH3OCH2CH2OCH2-CI 2) t-Bu, di-Me-Si-Cl Fig ure 1:3-10 Transformation of Lipiferolide to Simsiolide Acetate. 109 Table 1:3-5 Relative Percentages of Products Obtained from the Cyclization of Lipiferolide3. S0Cl2b dry HClc BF • Et20 VII 1 - - VIII 9 8 - IX 11 - - X 5 - - XI 10 4 1 XII 5 6 - XIII 26 61 6 XIV 9 6 - II 9 9 3 XV - - 69 XVI -- 3 III - - 1 XVII 7 Percentages are based on total isolated yield (meq). bThe products are the same using distilled S0C1 2 although the rela tiv e proportions may vary. Comparison was made by t ic . Composition was similar (tic) except for two polar minor side products when two drops of concentrated HC1 was used in place of the dry gas. EXPERIMENTAL* Cyclization of Liplferollde CJt) with Thionyl Chloride. Lipiferolide (2.5gms, 8.17meq) was dissolved in 55ml of benzene and 25ml Et20 and stirred in an Ng atmosphere. To th is was added 2 . 0ml of SOCI2 and the solution stirred at room temperaturefor two hours during which time it became dark gold in color. Twenty ml of 3% NaHCO^ were added and stirring maintained until gas evolution was complete. The benzene layer was extracted with a total of 100ml of 3% NaHCO^ and then washed with H2O (3*25ml). The combined aqueous layer was extracted three times with 20ml portions of benzene. After washing the combined benzene fractions with water (2x) , evaporation gave 2.42gms of a dark gold viscous oil. The residue was chromatographed over 110gms of silica gel 60 (230-400 mesh) using flash chromatography (2 . 3x 63 cm bed; flow= 7 . 3ml/min) and eluting with 35 , 60 and 80% EtOAc in hexane (500ml each), and 500ml of EtOAc. This allowed collection of 15 fractions combined on the basis of tic (Group A). n See Appendix on p. 364 110 111 Compound VII. Fraction A4 gave four additional fractions when chromatographed over 30gms of silica gel 60 (230-400 mesh) using 4* MeOH in CHCl^ as solvent. Fractions 1 and 2 crystallized from EtOAc/hexane to give 18mg of compound VII: mp 153-154°, tic Rf=0.27, 6 % CH3CN in CHC13; visualization, blue with p-anisaldehyde; 1H NMR (CDCl^), refer to Table 1:3-1; 13C NMR (CDC13), refer to Table 1:3-2; MS (El), m/e 306.1472(13*, M+; C 17h2205 requires 306.1467), 264 (2%, M- C2H20 ), 246 (33*, M-AcOH), 231 ( 6 *), 213 (16*), 200 (18*), 188 (24*), 176 (44*), 161 (16*), 126 (18*), 95 (40*), and 43 ( 100*). Compound VIII. Fraction A5 and A6 were chromatographed over 55gms of silica gel 60 (230-400 mesh) using 42* EtOAc in hexane as solvent ( 6 .4ml/min). This gave four fractions (Group B). Fractions B2 and B3 crystallized from EtOAc/hexane to give a total of 125mg of needles from EtOAc/hexane and identified as the chlorohydrin (VIII) : 197-7.5° (Et 2o :hexane); tic Rfz0.36, EtOAcihexane (1:1) v isu alizatio n , v io le t with p- anisaldehyde; [a ]D- 1 77°( c = 0 . 24,MeOH) , IR (CHCl^, (O-H), 3030 ( C-C —H), 1772 (C=0), 1748 (C=0), 1670 (C=C), 1260-1210 (C-O-C), 1103 (2° ale, C-0), and 895cm"1 (C=CH 2 rock); UV (MeOH), AMAX=210nm (log 6=4.18); CD ( c= 3. 24x 10' 3M, MeOH) [0 ]3OO 0» t012Q2+2960’ [^ 2 4 9 °* and C *]2 12 "35700 1h 112 NMR (d6 -acetone, 90MHz), 6.11 5.75 Cd, J13t ) 7=2.2, H-13t), 5.75 Cm, H-1), 5-34 (dt, 9A= J3 ^ gg=3-18, Jjj (g= 1 • 7, H—8), 4.68 (d, _Jgj7=3*18, H—6 ), 4.43 and 4.34 CAB q, J=7.04, H-5 and O-H; collapses to a singlet on addition of D 20 ), 3*34 dq, jJg 7=3*18, J.7 | 8=1*7 * H-7), 1.98 Cs, -Ac), 1.85 (d, 3H, J=0.9, H-14), and 1. 59ppm ( 3 , 3H, H-15); 13C NMR (CDC13, 20MHz), 8C 170.0 (s, C-12 and Ac), 136.1 Cs, C-10), 135.5 Cs, C-11), 129.7 Cd, C-1), 123,6 Ct, C-13), 79-7 Cd, C- 8), 79.5 (d, C-5), 79-3 Cs, C-4), 74.6 Cd, C-6 ), 40.4 Cd, C-7), 43.8 (t, C-2), 40.3 Ct, C-3) 25.5 Cq, C-15), 22.7 Ct, C-9), 21.1 Cq, Ac), and 19.3ppm Cq, C-14); mass spectrum CEI), ra/e 342.1238 <0.9%, M C 17H2305 C1 requires 342.1234), 246 ( 11%), 177 C7%),117 C8%), 81 C17%) and 43 <100%) . Compound IX. The mother liquor from fraction B3 was chromatographed over 55gms of s ilic a gel 60 <230-400 mesh) using 20% EtOAc in CHC13 as solvent which gave five fractions CGroup C). Fraction C3 crystallized from EtOAc/hexane to give 150mg of crystals which were id en tified as compound ^X: mp 154-550 CEtOAC/hexane), tic Rf=0.25, 20%EtOAc in CHC13; visualization, blue with p-anisaldehyde; IR CCHClg), vmax 3495 CO-H), 3020 ( C=C-H), 1772 CC=0), 1748 1:3-1; NMR CCDC13), refer to Table 1:3-2; mass spectrum 113 (El), m/e 324.1139 (2%, M-H20; C-jgH^O^C, requires 324.1129), 282 (3*), 266 (4%), 264 (14*), 246 (7*), 229 (7%), 188 (9%), 97 (16%), 71 (14%), and 43 ( 100%). Compound X. When the mother liquor of C3, and fractions C4 and C5 were combined and chromatographed over 65 gms of silica gel 60 (230-400 mesh, 6 % H20, flow=2. 9ml/min) using 20% EtOAc in CHCl^ as solvent, three fractions were obtained. The third fraction was rechromatographed twice more over silica gel 60 which led to the isolation of 75 mg of crystalline material (EtOAc in CHCl^) and id en tified as compound X: mp 122-3°» tic Rf=0.19, 15% EtOAc in hexane; visualization, purple with p-anisaldehyde; NMR (CDCl^), refer to Table 1:3-1; NMR (CDCl^)t refer to Table 1:3-2; mass spectrum (E l), m/e 306 (1.1%, M), 264 (29%, M-C2H20 ), 246 (23%, M-AcOH), 188(25%), 176 (80%), 147 (58%), 129 (52%), 91 (100%), 71 (90%); (Cl) 307 (7%, M+H). Compound XI. Compound XI was isolated from fraction A7 (396mg) by flash chromatography over 55gms of silica gel 60 [230-400 mesh using 42% EtOAc in hexane ( 6 ml/min) which gave five fractions. The third and fourth fractions were recombined and chromatographed over a second column of silica (230-400 mesh, 6 % H2Q, 45% EtOAc in hexane) which gave 145mg of cry sta llin e XI from fractions 4 and 5 (275mg t o t a l). The 11 u material crysta llized as fine needles from EtOAc/hexane: mp 107*5-8°; tic Rf=o.43, 60% EtOAc in hexane; visualization, navy blue with p-anisaldehyde, or red with 10% f^SOjj in Et20; [o]D-2U 6°tc=0.155, MeOH); IR (CHC13>, vMAX 3530 (0-H ), 3020 CCsC-H), 1775 (C=0), 1747 (C=0), 1672 (C = C) , 1645 (C=C), 1240-1210 (C-O-C), 947 and 920 (=CH 2 rock), and 820cm"1 (C=C-H); UV (MeOH), Amax 210nm ( l o g s =4.07); CD ( c=5. 065x10 3 , MeOH), [033 [tf]200 -131000; 1H NMR (C6 d6 , 80MHz, 70°C), 8 H 6.15(dd, —6 , 13=0*5 * 2I7, 13=1. 9, H-13c)» 5*17 (ddd, p a rtia lly buried, 3=2*6, ^3 g^=6 . 1, J3 ^gg=6 .2, H—8), 5*15 (d, 113= 1 * 5, H-13t), 502 (m, H-1), 4.81 (brd, 2H, H-15A and H-15B), 4.55 (brdd, i 6j13c = 0.5, J.6,7=1*2’ i s ,6 = 8*6 » H"6 ) * 3*62 Cbrd» J5 >6 =8. 6 , H-5), 2.67 (hx, H-7), 1*69 (C6 D6, 20MHz, 70°C ), SQ 169*5 (s, Ac); 169*1 (s, C-12); 148.0 (s, c-10); 138.0 (s, c - i 1 ); 131*7 (s, C-1); 130.3 (s, c-4 ); 122.5 ( t, C-13); 117.1 (t, C-1 5); 80.5 (d, C-5); 79*8 (d, C-6); 76.8 (d, C-8); 44.2 (d, C-7); 43*8 (t, C-2); 28.0 9q, C-3); 27*7 (t, C-9 > ,* 20.2 (q, CH3-C0) ; and 17.0 C-14); mass spectrum (E l), m/£ 306. 1461 (2%, M, £ ^ ^22®$ requires 306.1467), 264 (2%, M-C2H20 ), 246 (34%, M-AcOH), 228 (7%), 217 ( 6 %), 177 (9.5%), 131 (11%), 93 ( 10%), and 43 ( 100%). 115 Compound XII. From fraction A7, 78mg cry sta llized as needles and were identified as compound XII: mp 174.5-5.5° (Et^o/hexane): tic R{*s0.25, 50% EtOAc in hexane; blue with p-anisaldehyde; [ot]p -163°(c=0.195, MeOH); IR H), 3020 (C=C-H>, 1770 < C-0) t 1748 (C=0), 1668 (C=C), 1240-1210 (C-0-C), 1091 (2° OH and C-O-C) and 952cm“1 (C=CH2); UV (MeOH), JLmaX 210nm (log £= 4.28, end absorption); CD ( c= 9 . 0 6 9 x 10_ 3 M, MeOH), C# ]270 +580, [ * ] 262 °» £*]250“3100 (shld), and [0]2qq _8°500; NMR (d^-acetone, 90MHz), 6.13 (d, i 7( 13C=2.8, H-13c), 5.75 (d, J7 >13t=2.4, H-13t), 5.92 (brt, J1(2A= J.it 28*8-4, H-1), 5.39 (dt, J7 t8=1.6, —8,9A = —8 , 9B = 3 * 4 » H-8 ) , 5-28 (m, H-3), 5-09 (brd, J.5 (oh = 8*3, H-5), 4.43 Cd, J6j7 = 3.7, H-6 ), 3-78 (d, i. 5 f0H~8 *3 ’ °"H)» 3.38 (dq, m), 2.81 (brt, 2H, H-2), 2.73 (brd, 2H, H-9), 2.05 (s, Ac), 1.77 (m, H-5), and 1.69ppm (brs, H-14); 13C NMR (dg-acetone, 20MHz), Sq 169.9 (s, Ac), 169.9 (s, C-12), 140.5 (s, C-4)a, 138.5 (s, C-10)a, 134.4 (s, C-11), 130.5 (d, C-1), 122.2 ( t , C-13)» 122.0 (d, C-3), 82.5 (d, C-5), 79-3 (d, C- 8), 74.3 (d, C-6 ), 49.0 (d, C-7), 43-0 (t, C-9), 27.7 (t, C-2), 20.6 (q, CH3C0), 18.5 (q, C-14)b, and I8.2ppm (q, C-15)b--[a,b these pairs may be interchanged, all other assignments are based on SFORD experiments]; mass spectrum (El), m/£ 306.1458 (1.3% M, C17H220^ requires 306.1467), 288 (2%), 264 (2%), 246 (16%, M-AcOH), 150 (13%), 117 (32%), 84 (30%), 43 (100%); (Cl), m/e 307 (1.5%). 116 a-Cycloliplferollde(XIII). Combination of fraction A10 with those of previous columns having a similar composition gave 853mg of material that was chromatographed over lOOgms of silica gel 60 (230-400 mesh, +6%H20) using 50% EtOAc in hexane as solvent, Ten fractions were collected and the material in fraction four was found to be pure a -c y c lo lip ife r o lid e (X III), non crystalline oil; tic Rf=0.19, 50% EtOAc in hexane; visualization, blue with p-anisaldehyde; IR (CHCl^), ‘'max 3590 (0-H), 3500 (0-H), 3010 (C=C-H), 1775 (C=0), 1740 (C=0), 1675 (CsC), 1260-1210 (C-0-C), 910 (C=CH2); ^ NMR (dg-acetone, 90MHz), SH 6.09 (d, J? ,3=3. 5 , H-13c), 5.61 (dt, J.y(g=1*6, jI8 , 9A = 3*2* iL8(9B=^*5)» 5.52 (d, .^ 7^13 = 3. 2, H-13t) , 4.36 (dd, J5 f6 =10.5, i6,7=10*2' H"6>t 3.20 (dhx, —7,8 " 1*^» —7,1 3 t=^*2 * — 7 , 13c = 3*5» ji6 j7=10. 2), 2.80 (brd, -5,6=10*5» h~5)i 1.96 (s, OAc), 1.63 (brs, H-14), and 1.31ppm (s, H-15). Compound XIV. Fraction A10 crystallized directly from EtOAc/MeOH at 0°C to give 47mg of the chloroquaianolide XIV: mp 148-9° (EtOAc/hexane); t ic Rj.s0.31; 60% EtOAc in hexane; visualization lavender with p-anisaldehyde; [tt]D-48° ( c=0.118, MeOH); IR (CHC13), 3595 (0-H) , 3520 (O-H), 1775 (C=0), 1748 (C=0), 1670 (C=C) , 1240-1210 (C-O-C), 1178 (C-0, 3°alc) , 950 (C = CH2) and 815cm” 1 (C = C-H); UV (MeOH), \ MAX 212 117 (log E =3,86 ); CD (0* 3. 45x1O'^M; MeOH), [* ]2go 0, [ * ] 255 -1190, [tf ] 230 0, Ctf]210 -2320; ^ NMR (CDCl^ gOMHz), 8 H 6.31 (d, .£7 ^ 13*3 ■ 5, H— 13o), 5.64 (dt, J.yig=2.9i J7J13*3*5, H-13c) , 5.64 (dt, J7t8=2.9, I8,9A=1**9' i8,9B=l*-9> H-8),3*18 H-15); mass spectrum (E l), m/e 329 (0.5%), 327.1006 ( 1%, M- CH3; C16 H20) 5 requires 327.0999), 307 (1% M-Cl 286 (4%), 247 (26%), 229 (40%), 204 ( 6 %), 190 (14%), 144 (14%), 92 (10%), 43 ( 100%). tf-Cyclolipiferolide(II). Fraction A2 crystallized to give 113mg of needles from EtOAc/hexane which were id en tified as tf-cyclolip iferolid e ( I_I). This was compared to an authentic sample from L,. tu llp ifer a by mp, c o -tlc , C“]D, IR, UV, 1H NMR and MS. HC1 Cyclization of Lipiferolide(I). A solution of 12ml of benzene:ET 20 (1:1) containing 203mg of lip ife r o lid e was purged with dry HC1 gas for ten minutes. The HC1 was generated by dropwise addition of concentrated H2S0j| onto solid NaCl and passing the gas through a H2S0^ bubbler. The solution was stirred at room temperatureunder a positive pressure of HC1 (stopcocks were closed off) for twelve hours. Tic aliquots were taken every four hours. The solution was transferred to a 50ml erlenmeyer using CHCl^ and evaporated to a small volume 118 without heating. This was diluted with 25ml of CHCl^ and washed with H20 (3x15ml). The combined H20 layers were extracted with CHCl^ and backwashed with H 20. Evaporation of the total CHCl^ extract gave 227mg of pale yellow oil. This was chromatographed on three 20x40cm silica gel G prep plates (0.75mm). The plates were developed three times with 15% EtOAc in hexane and once with 30% EtOAc in hexane. Each plate was scraped into seven bands as determined by short wave UV and edge spraying with p-anisaldehyde followed by gentle heating. Fraction 4 (143mg) was further chromatograped over 25gms of silica gel 60 (230-400 mesh) and eluting with 25% EtOAc in hexane. This led to the isolation of 34mg of starting material (_I, 17%), and compounds VIII, XI, and XIII. Further purification of fractions 5 and 6 (55 mg) over 25gms of silica gel 60 using 35% EtOAc in hexane led to the isolation of compounds XII, XIV and XV. These compounds were compared to those originally obtained from the SOClg-catalyzed reaction by mp, co-tlc, IR and NMR. Also NMR spectra obtained on the mother liquors of these compounds showed no evidence for the presence of the [ 6 :2:0] bicyclodecane compounds. Treatment of Lipiferolide(I) with Distilled Thlonyl Chloride. Thionyl chloride (Fisher; reagent grade) was distilled over quinoline and boiled linseed oil by the standard 119 procedure. As the thionyl chloride was distilling, 0.16ml was taken from the receiver flask (through a septum) and added to 201mg of lipiferolide dissolved in 6 ml of Na-dried benzene:EtgO (2:1). The reaction mixture was stirred vigorously under a dry argon atmosphere for 60 minutes. It was then quenched by addition of 3% NaHCO^ ( 10ml) and stirring for 10 minutes. This was diluted with an additional 15ml of NaHCO^ and extracted with CHCl^ (3x25ml). Washing the organic layer with H2o and evaporation gave a yellow-green oil (no weight since a-cyclolipiferolide decomposes). This was compared by tic to all of the compounds isolated from the un distilled S0C1 2 reaction as well as the crude mixture. Development with 35% EtOAc in hexane ( 2x) or CH^CNiCHCl^ (2:23) showed no difference between the products produced. C ycli 2ation of Lipiferolide(I) with BF^» EtgO. L ipiferolide (I.JJ^gms, 4.71meq) was dissolved in 150ml of dry Et2o at 0°C with stirrin g under an N2 atmosphere and 6 ml of d is tille d BF^'EtgO was added. The solution was stirred for two and one half hours at 0°, 50ml of was added, and after ten minutes the ether layer was evaporated at reduced pressure. The remaining aqueous layer was extracted with CHCl^ (4x30ml) and the CHCl^ washed with HgO. Evaporation of the CHCl^ layer gave 1,83gms of yellow residue which crystallized on trituration with 30% EtOAc in 120 hexane to give 1.047gms of cry sta ls (prisms) id en tified a 3 the fluorohydrin (XV): mp 194.5-5° (from CHCl^; turns purple on melting); tic Rf=o.l4, 40% EtOAc in hexane; [a]D-l66°(c=0.295;Me0H); IR (CHClj) , 358 O (0-H), 3490 (0-H), 3020 ( CsC-H), 1710 (C=0), 1747 (C=0), 1668 (C=C), 1200-1260 (C-0) and 1100cm' 1 (C-0, and OH); ^ NMR (CDC13), refer to Table 1:3—3» NMR (CDC13)I refer to Table 1:3-4; MS (E l), m^e 326. 1537 (0.5%, M; re9 ui-r ®3 326.1529), 306 (0.5%), 284 (2%), 266 (26%, M-AcOH), 246 (28%), 109 ( 8%), 81 (11%), 43 ( 100%). The mother liquor (0.88gm) from several c ry sta lliz a tio n s of the fluorohydrin was chromatographed over 55 gms of silica gel 60 (230-400 mesh) using 30% EtOAc in hexane as solvent. Twenty fractions were obtained (Group A). Fraction A1 (13.6mg) was chromatographed on a prep-tlc plate. The plate was developed once in 25% EtOAc in hexane and twice in 10% EtOAc in hexane. Elution of the scraped band (located by short wave UV) with EtOAc gave a pale yellow o il which crystallized from EtOAc/hexane to give 8.2mg of needles. This material was identical with an authentic sample of e p itu lip in o lid e (I I I ) by mp (91-2°), co- tic (Rf =0.51; 45% EtOAc in h xune), IR and 1H NMR. Fractions A3 and A4 (4 1mg) cry sta llized from EtOAc/hexane to give 24mg of compound XVI as needles: rap 128.5-9°i tic Rj.s0.49, 45% EtOAc in hexane; visualization, pink with p-anisaldehyde; [“Jq - 1 4 o° ( c= 0 . 185, MeOH), IR 121 (CHCl3 ) f *MAX 3030 (C=C-H>, 1773 (C*0), 1748 (C=0), 1712 (CsO), 1665 (C=C), and 1260-121Ocm"1 CC-O-C); 1H NMR CCDCl^), refer to Table 1:3-3; NMR (CDCl^) refer to Table 1:3-4; MS (El), m/e 306.1474 (1.6% M; C17H2205 requires 306.1467), 246 (17%), 218 ( 6 %), 148 (9%), 98 (9%), 81 (18%), 43 ( 100%). Fraction A10 cry sta llized from EtOAc/hexane to give 34mg of needles which were identified as compound XVII: mp 101-101.5°, tic Rf=0.35, 1% MeOH in CHCI3 ; visualization pink with p-anisaldehyde; [«]jj-52°(c=o. 115, MeOH); IR (CHC13), vmax 3020 ( C=C-H); 1770 (C-0), 1740 (C=O0, 1720 (C =0,shl), 1670 (C=C), 1250-1210 (C-O-C), and 815cm-1 (C=C- H); UV (MeOH)J 1H NMR (CDC13>, refer to Table 1:3-3; 13C NMR (CDCl3)f refer to Table 1:3-4; MS (El), m/e 264 (1.2%), 248.1418 (29%, M-C 2H202 , C15 H2Q03 requires 248.1412), 228 (3%), 188 (40%), 91 (11%), 77 (8%), and 43 ( 100%). When fraction A8, A9 and the mother liquor of A10 were combined (215mg) and chromatographed over 55 gms of silica gel 60 (230-400 mesh; 6 % H20) using 0.5% MeOH in CHC13 as eluent, five fractions were obtained (Group B). Fraction B2 gave 56mg more of compound XVII. Fraction B3 crystallized from EtOAc/hexane to give 13.5mg of needles which were identified as compound XI by a comparison of the mp, tic behavior, IR and 1H NMR, Fraction 16A crystallized to give 25mg of compound IJ^ from EtOAc/hexane. This material was compared to an authentic sample of tf-cyclolipiferolide by mp (161-2°), tic, IR and 1H NMR. Compound XIII was isolated from fractions A11, A12 and A13 after chromatography over 25gms of silica gel 60 (230-400 mesh) using 35% EtOAc in hexane as eluent. A 37.5mg pure sample of g -c y c lo llp ife r o lid e (X III) was obtained and compared to that obtained from the S0C1 2 catalyzed reaction by tic, IR and 1H NMR. 20 MHz Wt/ p T T T T T n >i|ri>iim i |r T T T T T i p p | ■ * iT T T T T T J I r i i r i i 111 * ...... ^ p i ■ ■ 11 ■ rtJTTTTTTTTTTTl I I'l 11 11 1111 II tTTTyTTTTTTTTTj I ' M I I f r T T J T T n I T l 1 1 1 n T T T T T f T |T r 11 n T T T T n 1111 IT11 Tprrr 150100 5 0 PPM 13 Figure 1:3-11 C NMR Spectrum (CDC13) of Lipiferolide. Figure 1:3-12 IR Spectrum (CHC13) of Compound VII. 124 C4-Me Cin“ Me CDCI, ^ J\_ jU m IWJJLa 300MHz -J > I i i i I i H -1 3 c H -131 H - 8 H - 6 H - 5 H - 7 H-2* H-9* H-3* r v i - n - j i rn i i i i i j li t i i i i i i | i i i i r i i i i | i i i i p t t t t | it i t i i > i < j t h t 't t i r n 6 5 4 3 2 1 PPM Figure 1:3-13 NMR Spectrum (CDClj) of Compound VII. 20 MHz [iiirmn|fi'imnnnminiifiriim'i|HMinn|iimnMiiTtnnrT|iiimrTHnniim|irMimi[iniinn|iiim'iinI f 11 1 I I M III IT J I I I I I | r u t i t IV 11»Tf T l T n i | J H I H 1 M J H I I I I I I I 150 100 13 Figure 1:3-14 C NMR Spectrum (CDCl^of Compound VII, T*AN9«HftS+ONM iue :-5 IR Spectrum (CHCl^)Figure 1:3-15 Compound of VIII. J J > WX)t> CM-1 M C i E B M U N f V A W uoo 700 O D • 90 MHz ■ * J * * 1 I I—I—i— -i-J~li«Ailii**liii—> t » t t t I i1__|—*--! i i 6 5 4 3 2 1 PPM Figure 1:3-16 NMR Spectrum (d^-acetone) of Compound VIII. 20 MHz 'njHimiiniuFHiinfmum [um » rwrmiTminiiiii>>iifum u]iiiirmT]inn I mriivi iiviti iif ri 150 50 P PM Figure 1:3-17 13C NHR Spectrum (CDC13) 0f Compound VIII. JO Tl**SMIl»lOM m iu e :-8 IR Spectrum (CHC13)Figure 1:3-18 Compound of D(. m ■ M V iN U M lf l l lf M U iN V M ■ C P IM m "* m SOO tn MC 90 MHz Figure 1:3-19 H NMR Spectrum (CDCU) of Compound IX 20 MHz |rn iiiiii|n iTiiiii]im iiiii|t)rrn iin tn tiiiri|iiiiiiiii|in in iii|in iiini|nniiiir|Miiiin i|iMtiMii|iiim iri|tiiim ii]iimirii|iiiiiiTii|iiiiiiiii|niiiiiM|iiliimn ...... |jin .im PPM 150 100 5 0 13 132 Figure 1:3-20 C NMR Spectrum (CDC13) of Compound IX. nANlttlsttOM iu e :-1 IR Spectrum {CHClj)Figure 1:3-21 Compound of X. 133 9 0 MHz 1 _ J 1 1 1 1 1 I I 1 I I 1 I 1 1 I I I I I I I I I 1 I 1 I I I 1 I ! 1 1 1 I I I I I I L I I I I ■ I I I I ! I 1 I I 1 I I I I I L L I PPM Figure 1:3-22 1H NMR Spectrum (CDC13) of Compound X. 20 MHz p i in i ii i [ti um >i| 11 iM MH[n n m ii|iiiiiMii|iin rinijTrrin in jn tn im |iiiiiiiii|im im i|iii» iiii|iM iiiin n iiiiriii|n iirin i]in iiu ii|TH )iiiii|iiiPiiiii|iiiiiii 150 100 50 PPM Figure 1:3-23 i3C NMR Spectrum (CDCl^) of Compound X. II * U 19 13 M W * • 1400 M tA OVAVtNUMltt m o MOO M O SOO * 0 0 Figure 1:3-24 IR Spectrum (CHClj) of Compound XI. of Compound 1:3-24 Figure (CHClj) Spectrum IR HOC HOC MO ICO UOC MOO MOO NOlSllllSNVM 90MHz JL JA_ Ll. i L-! 1 J I -i- i A 1 * l I I > > « k 1 1 J .1 1 * I . < I It i_L i i k 1 1 1 i i i I i i . J J J 1 1 1 1 J ‘ 1 t J J « * ‘ lililljA lttl. I ■ I -1 1 > ■ * I ^ ^ 3 2 1 P P M Figure 1:3-25 NMR Spectrum (CgD^; 74°)) of Compound XI. 20 MHz , - j J H J U U v NfwuyAu/1 p n iiin i|iiiiiiiii|iiiiinii|iMiiiiM|iiiiniii|iiiiiiiii|im inrrjtfrm Tit[ruiiiui]i>Tr»iT»jnTmntpTTTiiii)|iirm rrr|im »niijiiiiiiiii|iiiirni> |iiiniiii|m iiiiii 150 1 00 50 PPM Figure 1:3-26 13C NMR Spectrum (C6Dg; 70°) of Compound )a. n*»KM - ij i] « u *- u k t t c «oo wcftOt a JLi-L I___ J j L i A n IVA 90 MHz r I I T ~ l1 | I I I I I I ; T f | I I I I T T T T I | I I I I I I I I 6 5 4 Figure 1:3-28 NMR Spectrum (dg-acetone) of Compound XII. 20 MHz )MnMin|niiiim|ii'riniiipii'mimiiiiirnniiiiimniiimnnimmii|)iiiiirrniiiiiim|iiiiiiin|iiiimi>|nmnii|miimim>iiiur|ii'iimr|iiinim|nnnin 150 100 50 PPM 13 Figure 1:3-29 C NMR Spectrum (CDCl^) of Compound XII. ttAHSMIIHOM iu e :-0 IR Spectrum (CHCl^) 1:3-30Figure Compound of XIII i CAT* K I M M V A W t MO e tii 90M H z i J i l l ■ L * | | j | J 1 j | i ! t_i I 1 1 1 t L i l i . I > . I r I I . . . I . . 1 ! j , . iff * * ■ > I » > -1 1 f 1 * t 1 ^ 1 1 ‘ ‘ 1 1 1 d 1 ^ ‘ 1 1 * * 1 1 1 LJ 7 6 5 4 3 2 1 PPM Figure 1:3-31 H NMR Spectrum {dg-acetone) of Compound XIII. 144 +4 ■o 250 3 0 0 nm -1200 -12 255 -16- 2300 -22 210 Figure 1:3-32 CD Spectrum (MeOH) of Compound XIV. TtANtwrtUON iu e :-3 IR Spectrum (CHC13) 1:3-33Figure Compound of XIV. W A «W »I« C M 4 Stit 90 MHz PPM i Figure 1:3-34 H NMR Spectrum (dg-acetone) of Compound XIV. WCtNt TftANSMrWO* i ue :-5 IR Spectrum (CHCl^)Figure 1:3-35 Compound of XV. «t«M wia wia «t«M at* *AAA v A 90 MHz ill I l.i 11 m il m i 111 i.i Lli i.i Liu i I in 1111 n 1111111 in 1111111111 In n 1111111 PPM 7 Figure 1:3-36 NMR Spectrum (CDC13) of Compound XV. 20 MHz T T T r r T TT PPM Figure 1:3-37 13C NMR Spectrum (dg-acetone) of Compound XV. j=- 4000 woe Meo mo hoc w aooo mo mo poo we i**> * I WAVfNUMtll c * r* Figure 1:3-38 XR Spectrum (CHClj) of Compound XVI. 90 MHz i i r r |—1~r' i'"> t ' r i r t i r i i r r r r r f t r r r r i i i f j r i r i i i i r i 151 Figure 1:3-39 NMR Spectrum (CDCl^) of Compound XVI. 0 4 9 7 TMN10AHSIO* •0 i ue :-0 IR Spectrum (CHCl^)Figure 1:3-40 Compound of XVII. 1 ftj MX) HOC0 *AVfHUMMI CM* DOO < 152 JJL .m JIa MA v 90 MHz JLa Jl M IllllIilli Li u 111 n 1111 n Lit! L1 Li 11111 n 111 LI 11 M 11 n 11 Llli 11 LLII i 11 n I n LI 111 l PPM 7 Figure 1:3-41 H NMR Spectrum (CDCl^) of Compound XVII 153 20 MHz r~r | I—l~T [-1 I I I I—T"'I | T " 1 1 |- I I " r | T"j i I I ■i" I l~T ' I"' I ' 'I i I ' I PPM 150 100 50 13 ui Figure 1:3-42 C NMR Spectrum (CDCl^) of Compound XVII *= LIST OF REFERENCES 1) G.A. Petrides, "A Field Guide To Trees and Shrubs", 2nd edition, The Peterson Field Guide Series, Houghton Mifflon Company, Boston, 1972, p. 203. 2) M.N. Fernald, "Gray's Manual of Botany", 8th edition, American Book Company, New York, 1950, p. 676. 3) S.J. Angyal and V. Bender, _j. Chem. Soc., 1961, 4718. 4) E.E. Dickey, _J. Org. Chem. , 1958, 23, 179- 5) F. Hideto and H. Takayoshi, Mokuzai Gakkaishi, 1977, 23, 405; CA 87:197244h . 6) C.L. Chen, and H.M. Chang, Phytochemistry, 1978, 17, 779- 7) F. Van Valen, Proc. K. Ned. Akad. Wet., Ser. C., 1978, 81, 355, CA 89:21T95T s . 8) R.W. Wes, K, Krevitz, and S. Behzadan, Lipids, 1976, 11, 18. 9) P. Demuth, and F.S. Santamour, Jr., Bull. Torrey Bot. Club, 1978, 105, 65; CA 89:20325a. 10) C.D. Hufford, M.J. Funderburk, J.M. Morgan, and L.W. Robertson, Pharm. Sci., 1975, 64, 789. 11)M.A. Buchanan and E.E. Dickey, J. Org. Chem., 1960, 25, 1389. 155 156 12) R. Ziyaev, A. Abdusamatov, S.Y. Yunusov, Khim. Prir. Soedin, 1973, 67: CA 78:159939v. 13) R. Ziyaev, A. Abdusamatov, S.Y. Yunusov, Khim. Prir. Soedin, 1977, 715: CA 88:121478n. 14) P.D. Senter and C.L. Chen, Phytochemistry, 1977, 16, 2015. 15) R. Ziyaev, O.N. Arslanova, A. Abdusamatov, S.Y. Yunusov, Khim. Prir. Soedin, 1980, 428: CA 93:164329k. 16) A. Abdusamatov, R. Ziyaev, S.Y. Yunusov, Khim. Prir. Soedin, 1975, 813: CA 84:150806r. 17) C.L. Chen, H.M. Chang, E.B. Cowling, C.H. Hsu and R.P. Gates, Phytochemistry, 1976, 15, 1161. 18) Other references can be found under the following Chemical Abstracts listings: 80:60055h; 8l:177117g; 81:60821a; 81:60824d; 82:86460c; 85:l60380g; 85:17144c; 93:P31772. 19) R.W. Doskotch and F.S. El-Feraly, J. Pharm. Sci., 1969, 58, 877. “ 20) R.W.Doskotch and F.S. El-Feraly, J.Org.Chem., 1970, 35, 1928. ~ 21) R.W. Doskotch, C.D. Hufford and F.S. El-Feraly, J. Org, Chem., 1972, 37, 2740. “ 22) R.W. Doskotch, S.L. Keely, Jr., and C.D. Hufford, _J Chem. Soc. Chem. Commun., 1972, 1137- 23) R.W.Doskotch, S. L,Keely, Jr., C.D.Hufford and F.S.El- Feraly, Phytochem. , 1975, 14, 769. 24) R.W.Doskotch, F. S.El-Feraly, E. H.Fairchild, and 157 C.T. Huang, ib id ., 1976, 402. 25) R.W. Doskotch, F.S. El-Feraly, E.H. Fairchild, and C.T. Huang, Org. Chem., 1977. 42, 3614. 26) R.W.Doskotch, E.H.Fairchild, C.T.Huang, J.H.Wilton, M.A.Beno, and G.C.Christoph, J. Org.Chem. , 1980, 45, 1441. 27) N.H. Fischer, E.J. Olivier and H.D. Fischer, Forschritte d_. Chem. Org. Naturst. , 1978, 38, 47. 28) E.H. Fairchild, 1976, Ph.D. Dissertation, The Ohio State University, Columbus, Ohio. 29) F.S. El-Feraly, 1969, Ph.D. Dissertation, The Ohio State University, Columbus, Ohio . 30) R.W. Doskotch, and F.S. El-Feraly, Cam. J. Chem., 1969, 47, 1139. 31) T.A. Geissman and T. Saitoh, Phytochemistry, 1972, 11, 1160. 32) S. K.Talapatra, A.Patra, B.Talapatra, Phytochem., 1973, 12, 1827. 33) W. Stocklin, T.G. Waddell, and T.A. Geissman, Tetrahedron, 1970, 26, 2397. 34) K.Tori, I.Horibe, K.Kurizaraa, H.Tada and K.Takeda, ibid., 1971, 1393- H.Tada and K.Takeda, J.C.S.Chem.Commun., 1971, 1391. 35) R.W.Doskotch, F.S.El-Feraly and C.D.Hufford, Can.J.Chem., 1971, 49, 2103 . 36) A. Horeau in "Stereochemistry, Fundamentals and Methods", Vol. 3, H.B. Kagan, Ed., Thieme, Stuttgart, 1977, p. 51. W. Herz and H, Kagan, _J. Org. Chem. ,1967, 32, 216. 158 37) Private communication with F.S. El-Feraly; University of Mississippi; to be published. 38) R.W. Denny, and A. NIckon, "Sensitized Photooxygenation of Olefins", in "Organic Reactions", Volume 20, John Wiley and Sons, New York, 1973, p. 133* 39) K. Tori, I. Horibe, Y. Tamura, K. Kuriyama, H. Tada, and K. Takeda, Tetrahedron Letters, 1976, 387. 40) E.D. Brown and J.K. Sutherland, Chem. Commun., 1968, 1060. 41) T.A, Dullforce, G.A. Sim, D.N. White, J.E. Kelsey, and S.M. Kupchan, Tet. Lett., 1968, 973- 42) H. Furukawa, K.H. Lee, T. Shinger, R. Meek, and C. Piantadosi , J^. Org. Chem. , 1973, 38, 1724. 43) F. Bohlmann and C. Zdero, Phytochem. , 1977, 16, 778. 44) W. Herz, J. Poplawski, and R.P. Sharma, J. Org. Chem., 1975, 45, 199. 45) A,A. Devreese, P.J. DeClercq, and M. Vandewalle, Tet. L ett., 1980, 4767. 46) W.C.Still, M.Kahn, and A.Mitra, J.Org.Chem., 1978, 43, 2923. 47) Private communication with F. Harraz, Ohio State University; to be published. 48) L.M. Jackman and S. Sternhall, "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry", 2nd edition, Pergamon Press Ltd., Oxford, 19 6 9 - 49) K. Tori, I. Horibe, K. Kuriyama, Chem. Commun., 1971, 1393. 159 50) K. Tori, I. Horibe, Y. Tamura, K. Kuriyama, H . Tada and K. Takeda, Tet. Lett., 1976, 387. 51) K.Tori, I. Horibe, K. Kuriyama and K. Takeda, Chem. Commun., 1970, 957- 52) K. tori, I. Horibe, H. Yoshioka and T. Mabry, J. Chem. Soc. (B), 1971, 1084. PART II: ISOLATION, IDENTIFICATION AND ABSOLUTE STRUCTURE ELUCIDATION OF COMPOUNDS FROM SIMMONDSIA CALIFORNICA. 160 Chapter 1: The Isolation and Structural Determination of Three New Cyanoglycosides and Other Constituents from Simmondsia californica. INTRODUCTION Simmondsia californica (chinesis), (family Buxaceae) , commonly known as jojoba is a dioecious, evergreen desert shrub which grows naturally in southern United States and northwestern Mexico. The seeds, which are brown and vary in size from a coffee bean to a large peanut, are produced at a rate of 5-10 pounds p^r year by a single mature shrub 1 2 . The seeds contain 50% of a liquid wax whose major constituents are C^q, C^t and C^ monoesters3* This pale yellow, odorless oil has recently received widespread attention as a replacement for spermaceti (sperm whale o il)^ ’3. However, the economic value of an oilseed crop usually depends upon utilization of the resulting meal as a livestock feed. In the case of jojoba, the meal is very high in protein, but contains a bitter principle, and causes extreme weight loss in laboratory animals2’3 rendering it unsuitable for such use. Investigations were prompted to establish the identity 161 162 of the bitter, toxic substanceCs). As a result, Elliger et ji al. isolated two cyanomethylenecyclohexyl glycosides, simmondsin and simmondsin-2'-ferulate. The structures and relative stereochemistry had been established^*^ ^ , but the absolute stereochemistry was unknown. In addition, the presence of two monomethoxyglycosides was indicated but were never isolated in pure form, nor fully characterized1*. Related compounds include dasycarponin from Thalictrum £ dasycarpum(Ranunculaceae), and lithospermaside from Thaiictrum rugosum^, Thalic trum revolutum^, Lithospermum purpureo-caeruleum, Lithospermum officinale (Boraginaceae)®, and Grjffonia simplicifolia Baill. (Caesalpinaceae)^. Together, these compounds make up a growing class of novel cyanoglycosides atypical of the classical cyanogenic glycosides (e.g. amygdalin, prunasin, dhurrin). CN OH HO HQ 7, MeO I I OMe Simmondsin CN 163 OH RO HO CH,0 i OCH, Simmond sin-2'-Ferulate 0CHo NC HOCH, HO HO OH HO i i HO Lithospermas ide NC HOCH HO HO HO Dasycarpon in Our investigation includes the isolation and structure determination of two new demethylsimmondsin glycosides, a didemethylglycoside, and other polar constituents from Simmondsia californica. DISCUSSION AND RESULTS Isolation of the glycosides and other compounds from Simmondsia californica (chinensis) was accomplished by use of a droplet counter-current chromatograph in which fractionation occurs by a form of liquid - liquid partition chromatography. Separation of 6.1gms of MeOH/EtOAc (1:1) soluble material (described in EXPERIMENTAL) using the lower layer of CHCl^:CH^CN:MeOH: (4: 1:4:2) as the mobile phase, resulted in the isolation of nine fractions as illustrated in Figure 11:1-1. The individual compounds isolated from these fractions include simmond sin(LA), 5-0- demethylsimmondsin(TB), 6-0-demethylsimmondsin(1^), 5,6-di- O-demethylsimmondsin (_ID) , sucrose, ( + )-pinitol (VIA) , and a Dragendorf-positive aromatic compound whose structure remains unknown. The NMR of the four glycosides (IA-ID) immediately indicated that they contained the same basic skeleton, but differed in the number of methoxy groups which were present. Simmond sin (I_A) was obtained as large granular crystals, mp 158-9° (95-100° reported; hydrated form^) from fraction C, and was identified by comparison of its physical constants with literature values (mp, IR, UV, and NMR)^. 164 165 216 mg 1 9 0 Instrument R1 kaklkai Co. (Model DCC-A) Sample Wt. G.lgms. Pressure 6 — 8.5kg/cm 1 8 0 Plow Rate 19ml/hr (53 drops/ml) Fraction Vol. 40ml Solvent CHC13 s CH-CN:MeOH;H-0(4;1:4:2) 1 7 0 mobile - lower layer 1 6 0 1 5 0 HO, ,0-Glu 1 4 0 ISO- OMe 1 2 0 - 110- 100 9 0 Weiohtfma.) A 162 8 0 8 72 C 1121 7 0 - 0 468 338 5,6-Demethyl E 6 0 Simmondsin and F 1598 (+)-Pin!tol G 132 sa H 1416 | 217 _ 4 0 - 3 0 20 - 10 - a t------1------V ‘ T 100 110 120 130 Fraction No. ure ll:l-l Droplet Counter-Current Chromatographic Separation of Simmondsia Glycosides 166 CN OMe HQ OGIu MO, OH R1 R2 IA Me Me HO IB H Me RIO OH IC Me H OR2 OH ID H H VIA Its IR spectrum (Figure 11:1-10) contains a 2220cm”1 peak indicative of a conjugated n itrile, a 1645 and 800cm_1 absorption corresponding to a conjugated, trisubstituted double bond, and multiple peaks in the 1050 to 1130cm”1 region for C-0 stretching and 0-H bending frequencies. The optical rotation was found to be -73°t and the UV spectrum contained a maxima at 217nm (log € =3-95) corresponding to an a, 0-unsaturated nitrile. Since the NMR analysis was found to be useful in the study of these glycosides, a short review of the simmondsin spectrum (Figure 11:1-11 and Table 11:1-1) is given. Downfield at 5.79ppm, a doublet is observed for the olefinic proton (H-2) which is coupled to the axial C-4 proton. This is typical of allylic coupling1® which generally ranges from 0-3Hz, being at a maximum when the dihedral angle is 90°, but zero at 0 or 180°* The C-4 proton appears as a dd at 4.70ppm (J2 ^ 2 .0 ,^ 5“9-8Hz) . Proton H-8 is deshielded relative to H-4 and appears 167 Table 11:1-1 1H and 1 3C NMR Data for Simmondsia Glycosides 1A-ID. 1H NMR DATA3 H IA IB IC ID 2 5 .79fd 5 .77,d 5.78,d 5.78 ,d (2.0) (1.6) (2. 1) (2.1) 4 4.70,dd 4,61,dd 4.75,ddb 4.70,dd (2.0,9.8) (1.6,9.6) (2. 1, 10.0) (2.1,9.8) 5 3-28,dd 3.59,dd 3.19,dd 3.50,dd (4.0,9.8) (3.3,9.6) (3.2,10.0) (3.3,9.8) 6 4.05,ddd 3.82,ddd 4.42,ddd 4. 16,ddd (3.2,3.6 (3.3,3.5) (3.0,3-2, (3.0,3.2, 4.0) 3.7) 3.3) 7ax 1.74,ddd 1.78,ddd 1.88,ddd 1.92,ddd (3.4,3.6, (3*5,3.7, (3.7,3.7, (3.0,3.8, 15.9) 15.7) 15.9) 15.8) 7eq 2.60,ddd 2.56,ddd 2.44,ddd 2.41,ddd (3.2,3.3, (3.3,3.3, (3.0,3.0, (3-2,3.2, 15.9) 15.7) 15.9) 15.8) 8 4.96,dd 4.97 ,dd 5.01,dd 5.01 ,dd (3-3,3-4) (3.3,3.7) (3-0,3.7) (3.2,3.8) 5-0Mec 3.45,s - 3 .45,s - 6-0Mec 3.42,s 3•42, s - - 1 ' 4.55, d 4.56,d 4.57 , d 4.57,d (7.9) (7.9) (7.9) (7.9) 2 ' , 3 ,d 4 ’ ,5 ’ 3.25-3.55 3.25-3.55 3.25-3.55 3.25-3.55 6a 1 3.86, d 3-86,d 3-85,d 3.86,dd (12.4) (12. 1) (11.7) (1.12.4) 6b' 3.74,dd 3.74,dd 3-74,dd 3.75,dd (4.0,12.4) (3.4,12.1) (3.3.11.7) (1.3.5.12.4) 168 Table 11:1-1 (continued) 13C NMR DATAe C IA IB IC ID 1 164.7(s) 164.8(s) 164.8(s) 165.0(s) 2 95.7(d) 95.6(d) 95.4(d) 95.4(d) 3 117.6(s) 117.6(s) 117.5(s) 117.7(s) 4 69.6(d) 70.9(d) 65.5(d) 70.4(d) 5 84.6(d) 78.5(d) 85.4(d) 76.4(d) 6 75.0(d) 75.3(d) 69.2(d) 69.5(d) 7 30.7(t) 30.7(t) 34.0(t) 34.3(t) 8 76.8(d) 76.9(d) 77.6(d) 77.9(d) 5-0Me 57.7(d)* - 57.7(q) - 6-OMe 5 8. 0 ( q )* 57.9(q) - - 1 ' 103.4(d) 103.3(d) 103.6(d) 103-6(d) 2* 73.5(d) 73.5(d) 73.5(d) 73.6(d) 3’ 76.6(d) 76.5(d) 76.3(d) 76.3(d) 4 * 69.9(d) 69.9(d) 69.9(d) 7 0 .1(d) 5 ' 76.5(d) 76.5(d) 76.6(d) 76.7(d) 6 ' 61.0(t) 61.0(t) 60.9(t ) 61.0(t) a300MHz/D2o with DSS standard. bFrom 315°K spectrum. cBased on Ref. 11, page 240, and demethyl glycosides. ^Non-lst order. e2CMHz/D20; spectrometer reference of dioxane (D 2O) sample as standard ( <5C 67.4ppm). May be interchanged. 169 as a double doublet at 4.96ppm indicating that it must lie within the plane of the double bond and can only be coupled to H-7ax and H-7eq, but not H-2. This was substantiated 'through decoupling experiments in which H-2 was irradiated without any apparent collapse of H-8. Since H-8 has two small coupling constants (3*8Hz; observed in CD^CN at 90MHz), it must be equatorial and adjacent to a methylene group forming a dihedral angle of approximately 60° with both C-7 protons. The C-7 protons are observed as doublet of double doublets with coupling constants in the range of 3, 3» and 1 6Hz, and form the AB portion of an ABXX * system. The larger coupling constant results from geminal coupling, and the smaller from the adjacent, equatorially-oriented C-6 and C-8 protons. The C-5 proton, located at 3*28ppm (CD^CN, 90MHz), appears as a double doublet with coupling constants of 4 and 9.6Hz suggesting an axial orientation in which it forms a 60° dihedral angle with H-6, but is trans-diax ial to H-4. The hydroxyl group must be located at C-4 since the corresponding proton collapses to a double doublet from a doublet of double doublets on addition of D 2O as observed in CD^CN at 90MHz. Also, H-4 is deshielded to 5-92ppm on formation of the pentaacetate derivative confirming the assignment. The glucose moiety can only be located at carbons 5, 6 170 or 8- Positions 5 and 6 are immediately ruled out since on treatment with HC1, an a,0-unsaturated lactone (modified .CN HO, OGIu HO MeO‘ IA MeO' IVA OMe OMe aglycone) is formed bearing two methoxy substituents at these positions^. This lactone can only be formed by generation of a new hydroxyl group on the same side as the n itr ile, formation of an imino-ester, and subsequent hydrolysis. As previously discussed, the H-8 proton must be equatorial which places the glucose moiety in an axial position, a phenomenon known as a^ ’ ^) 5tr ain 11 ' 1 ^ r 1 ^. This type of steric interference operates in Y-substituted allylic systems, such that even when moderately sized substituents are involved, the interference is greater than that found in 1,3-diaxial interactions. The 13C NMR spectrum (Figure 11:1-12) contains a singlet at l64.7ppm for the carbon of the n itrile, and a singlet and doublet at 117.6 and 95.7ppm respectively, for the olefinic carbons. Two triplets are observed at 61.0 and 30.7ppm for C-6* and C-5, respectively, and a doublet at 103.^ppm for the anomeric carbon of glucose. The latter confirms the stereochemistry of C-l1 as 0 since a- and /3-D-glucopyranose 171 have reported values of 100.6 and 104.6ppm, respectively1^. The two methoxyl groups were observed as quartets at 57.7 and 58.0ppm. In addition, eight doublets were observed in the 65 to 85ppm region corresponding to oxygenated carbons. The assignment of all carbons (Table 11:1-1) was made by comparison of the resonances of the four glycosides and 1-0- methyl-/3-D glucopyranoside (Table 11:1-2). A discussion of these is given later in the chapter. Table 11:1-2 13C NMR Data for 1 -0-Methyl-/5-D-Glucopyranoside( VA) a . c SC *cb 1' 103.7(d) 104. 2 2 ' 73.7(d) 74. 1 3 ' 76.3(d) 76. 9 4 ' 70.2(d) 70.7 5 * 76.3(d) - 76.8 6 ' 61.3 C t ) 61. 9 MeO 57.7(q) 57. 9 a0btained at 20.123MHz in D2O using the spectrometer reference of a dioxane/D 20 sample as standard. bValues measured in D2O relative to external CS2 and converted to ppm from TMS using the factor 193. 5ppm, and taken from reference 15, page 461. 172 Of the three new glycosides, only 6-0-demethylsimmondsin (IC) crystallized (mp 211-12°): 5-0-demethylsimmondsin( IB) and 5, 6-di-0-demethylsimmond sin (I_D) were both obtained as a non-cry3talline hard glass. All compounds (Iji, H>) have negative optical rotations (-85.7°, -109°, and -90°, respectively) similar in magnitude to simmondsinC-73°), as well as UV maxima (217, 218, and 220nm, respectively) indicating the presence of the , -unsaturated nitrile. The IR spectra of each contain absorptions for the n itrile, double bond, hydroxy and/or methoxy functions. The NMR spectra (300MHz) revealed that all compounds contained the doublet at 5 .8ppm for the olefinic H-2 proton, as well as characteristic absorptions for H-4, H-7ax, H-7eq and H-8 protons. Compounds _IB and both contained a single methoxy as indicated by the 3H singlets at 3.42 and 3.45ppm, respectively. Compound _ID was totally devoid of such absorptions indicating that it was the 5,6-di-O- demethyl derivative of simmondsin. Compound Hi was formulated as 5-0-demethylsimmondsin and JX as the 6-0- demethyl derivative on the basis that H-5 is shifted downfield to 3*59ppm in simmondsin TB from 3*28ppm whereas H-6 undergoes the same downfield shift to 4.42ppm in _I£ from 4.05ppm. This suggests that these protons are located on carbons bearing newly formed hydroxyl groups. This is in agreement with ID, the didemethyl compound, in which both H-5 and H-6 are located downfield. These assignments in 173 H-5 and H-6 are located downfield. These assignments in chemical shifts and conformation comply with the following rules18: (1) axial -OMe groups (3-42 and 3-42ppm in IA and IB) appear further upfield relative to those in the equatorial position (3*45 and 3-45ppm in ^A and I£ ), (2) equatorial protons attached to carbons bearing -OMe or -OH groups are deshielded by 0.6ppm relative to the corresponding axial protons (4.06 and 4.l6ppm for H-6 in 1A and ID vs 3-25 and 3*50ppm for H-5 in IA and ID). NC R1 R2 IA Me Me IB H Me IC Me H HO H ID HH OGIu OR1 OR 2 Since the initial evidence suggested that the three new glycosides were demethyl derivatives of simmondsin, it was necessary to establish whether they were of the same abso lute and relative stereochemistry. Treatment of each of the new glycosides with Mel/Ag^O/DMF at room tem p era tu re' 18 produced permethylsimmondsin (I_I) , identical by tic, [«], IR, MS, and H NMR to that obtained from simmondsin under the same conditions. Permethyl simmondsin (I_I) was obtained as a colorless oil having an optical rotation of -46°, and exhibited end absorption in the UV at 210nm (log e = 3*88). The material analyzed correctly for C21H 25 NOg, and was devoid of hydroxyl absorptions in the IR spectrum (Figure 174 11:1-22). Absorptions for the n itrile (2222cm-1 ), and double bond (3008 and 1642cm-1 ), as well as an intense, broad absorption at 1100cm-1 corresponding to the C-O-C stretching frequency were readily observed. The 1H NMR spectrum (Figure 11:1-23) contained seven methoxy singlets in the 3*3 to 3»6ppm region (Table 11:1-3). The spectrum integrated properly, and protons at newly methoxylated centers are shifted upfield (e.g. H-4; 4.72 to 4.24ppm) as was observed in the glycosides themselves. Examination of the H-4 coupling constants shows c —'41 b decreasing from 9 .5(simmondsin) to 7.8Hz indicating that the C-4 methoxy group may also be experiencing A^1,3) strain. By shifting below the plane of the double bond, the dihedral angle of H-4 and H-5 would decrease accounting for the smaller coupling. Further evidence for this i3 that a second coupling constant for H-2 (1.3 and 0.6 Hz) is observed indicating that both of the allylic protons have diverged from the plane of the double bond. This conformational shift would also decrease the dihedral angle of H-7ax with the neighboring protons accounting for the increase in its coupling constant. The carbon spectrum (Figure 11:1-24; Table 11:1-3) contained the necessary 21 carbons, seven of which were quartets in the 57 to 61ppm region. In order to verify the location of the OH groups as assigned from the observed chemical shifts in the 1H NMR 175 Table 11:1-3 1H NMR (300MHz, CDC1 IMS) and 13C NMR (20MHz, CDC1^, TMS) Data For Permethylsimmondsin( I I ). H Multiplicity (J_ in Hz) 2 5. 49, dd(1.3, 0.6)a 4 4. 24, dd (7. 8, 1. 3)a 6 3. 75, dd(3•2, 3.8, 5.9) 7ax 1. 70, ddd(3•8, 3.8, 14.2) 7eq 2.36, ddd(5.9, 5.9, 14.2) 8 4.72, brdd(3.8, 5.9) 1 ' 4.35, complex OMe' s 3.61, 3.60, 3.51, 3.49, 3.42, 3.41, 3.34, s ’s ,3 ',V 3-1-3 .6, complex and buried under 5 ' ,6A ’ , 6B* methoxy singlets C C 5C Cb 1 161.9(s) 1 ' 103.3(d) 4 79.4(d) 2 95.4(d) 6 * 71.5(t) 6 74.9(d) 3 116. 1 (s) OMe 60.5(q) 2 ' 80.6(q) 60.5(q) 5 86.6(d) 60.2(q) 3’ 83.3(d) 59. 1(q) 7 31.5(t) 58.7(q) 4 ' 83.4(d) 58.2(q) 8 76.0(d) 57.0(q) 5 ' 74.8(d) a0btained from 90HHz spectrum in CDCl^ with TMS. bAll assignments in this column were made by reference to other derivatives and may be interchangeable. 176 spectrum of the glycosides, each compound was acetylated with Ac2Q/Pyr at RT. Simmondin (I_A) gave the corresponding pentaacetate IIIA whose physical constants agreed with literature values^. Glycosides H* an(* .IC both produced hexaacetates (IIIB and II1C, respectively) whereas _ID gave a heptaacetate, H ID. This was confirmed by the number of 3H singlets in the 1H NMR spectra at 2.0ppm (Table 11:1-4) and mass spectral analysis. Since the glucosyl moiety itself must form a tetraacetate, there remains 1, 2, 2 and 3 acetates respectively on the aglycone portion of IIIA-IIID. In forming the pentaacetate of simmondsin, H-4 is deshielded from 4.72ppm to 5-92ppm as expected. Likewise, the H-5 and H-6 protons in IIIB and IIIC, respectively, are also deshielded. In compound IIID, H-4, H-5 and H-6 are all deshielded. Confirmation of the positions of the various protons was gained from double irradiation experiments (Table 11:1-5) of 6-0-demethylsimmondsin hexaacetate(IIIC), and from the individual proton multiplicities and coupling constants. In order to perform a more thorough analysis of the proton and carbon absorptions, as well as to provide additional evidence for the structure proof, it was deemed necessary to obtain the corresponding aglycones. Acid hydrolysis (2N HC1) of each glycoside gave the corresponding modified aglycone (IVA-D) which is formed via an imino-ester 177 Table 11:1-4 1H NMR Data For The Peracetylated Glycosides IIIA-Da. H IIIA IIIB IIIC HID 2 5.42 ,d 5.45,-d 5.45 ,d 5.46,d (1.8) (1. 9 )b (2. 0) (2,0) 4 6.03,dd 6.16 ,dd 6. 14,dd 6.21,dd (1.8 ,8.9) (1 .9b,9.8) (2.0,9.6) (2.0,10.5) 5 3.19,dd 4.79 »dd 3-15,dd 4.82,dd (2.8,8.9) (2.9,9.8) (3.2,9-6) (3-4,10.5) 6 3-81,brm 3.80,ddd 5.34,ddd 5.27,ddd (2.8,3.2, (2.9,3*4, (2.9,3.2, (2.8,2.8, 4.5) 3.6) 3.3) 3-4) 7ax 1.62,ddd 1.67,ddd 1.76,ddd 1.86,ddd (3.2,3.2, (3.6,3.6, (3.3, 3.8, (2.8,3.7, 15.0) 15.3) 15. 6) 15.8) 7eq 2.44,ddd 2.48,ddd 2.46,ddd 2.46,ddd (4.5,4.5, (3.4,3.4, (2.6,2.9, (2.8,2.8, 15.0) 15.3) 15.6) 15.9) 8 4.79,dd 4.85,dd 4.83,dd 4.87,dd (3.2,4.5) (3.4,3.6) (2.6,3.8) (2. 8,3-7) 5-OHe 3.42,s - 3.36,s - 6-OMe 3•35,s 3.31, s - - Ac 2.00,s 2.00, s 2.00, s 2.00, s 2.02,s 2.02,s 2.02,s 2.02,s 2.04,s 2.04,s 2.03,s 2.02, s 2.08,s 2 .07,s 2.08,s 2 .02,s 2.14,s 2.08,s 2.14,3 2.07,s 2. 11 ,s 2 .17,s 2 .12,s 2.14,s 178 Table 11:1-4 (continued) H IIIA 11 IB m e HID 1 ' 4.70,d 4.69,d 4.66,d 4.67,d (7.8) (7.8) (8.1) (7.9) 2' , 3',4* 5. 12 ,mc 5 . 1 3, mc 5. 13,mc 5 . 1 3, mc (5.0-5.2) (5.0-5.25) (5.0-5.25) (5.0-5.25) 5 ’ 3.60,ddd 3.66,ddd 3.66,ddd 3.67,ddd (2.4,4.2, (2.0,4.0, (2.2,4.2) (2.2,4.2, 9.6) 9.6) 9.6) 6 'a 4.24,dd 4.27,dd 4.28,dd 4.31,dd (2.4,12.2) (2.0,12.3) (2.2,12.5) (2.2,12.3) 6 *b 4.03,dd 3.99»dd 3.97,dd 3.95,dd (4.2,12.2) (4.0,12.3) (4.2,12.5) (4.2,12.3) aDetermined at 300MHz in CDCl^ with TMS as standard. ^From 90MHz spectrum (CDCl^). °Non 1st order. 179 Table 11:1-5 Double Irradiation Experiments for 6-0-Demethylsimmondsin Hexaacetate(IIIC)a . Proton *H Results After Irradiation H-4 6.14 H-5, 3.18, dd collapsed to a d(3.2) H-2, 5.49, d collapsed to a s H-5 3. 18 H-6, 5*38, ddd collapsed to a dd H-4, 6.14, dd collapsed to a d(2.0) H-6 5. 38 H-5, 3-18, dd collapsed to a d(9.6) H-7ax, 1.79, ddd collapsed to a dd(3*8,15.6) H-7eq, 2.50, ddd collapsed to a dd(2.6,15.6) H-8 4. 87 H-7ax, 1.79, ddd collapsed to a dd(3-3, 15.6) H-7eq, 2.50, ddd collapsed to a dd(2.9, 15.6) aObtained at 80MHz in CDCl^ with TMS as standard. The observed residual coupling constants are listed in parentheses in H2 . 180 Table 11:1-6 1 Data for Acetylated Glycosides IIIA-D3 . c IIIA 111B m e HID 1 159.3(a) 159.0(s) 159.8(s) 158.9(3) 2 95.7(d) 96.4(d) 95.6(d) 96.4(d) 3 115. Ms) 115.4(s) 115.6(s) 115.3(s) 4 70.8(d) 69.2(d) 70.3(d) 68. 6(d)e 5 82.5(d) 75. 1(d)d 82.3(d) 73.2(d) 6 74.3(d) 74.6(d)d 66.7(d) 68. 0(d)c 7 30.6(t) 31.2(t) 33. 1 (t) 3 3 - 3 C t) 8 75.7(d) 76.2(d) 77.0(d) 77.0(d) 5-OMe 5 8 . 2 ( q ) 58.3(q) — 6-OMe 56.6(q) 57.0(q) - — 1 * 100.7(d) 101.5(d) 102.2(d) 102.5(d) 2 ,b 72.8(d) 73. Kd) 73.0(d) 73.0(d) 3,b 68.2(d) 68.5(d) 68.3(d) 68.4(d) 4 ,b 70.8(d) 71.3(d) 71.3(d) 71.3(d) 5' 72.0(d) 72.4(d) 72.4(d) 72.6(d) 6 ’ 61.4 (t) 61.6(t) 61 - 4 C t ) 61.4(t) Ac C=0 168. 3 168.2 168.5 168. 3 (s's) 168.7 168. 8 1 69. 1 169. 1 169. 1 169. 3 169.4 169. 3 169.9 1 70. 2 170. 2 169.7 170. 2 170. 2 170. 6 1 70. 0 - 170. 4 171.2 170. 4 - -- 170.9 Ac CH3 20. 3 20. 5 20. 6 20.5 (q' s) — 20. 8 20. 7 20. 9 — 21 . 0 aCDC1^ and Me^Si at 20MHz. b»c»d Interchangeable. 181 Table 1 1 :1 -7 SFORD Results for Simmondsin Pentaacetate (IIIA)a. Irradiated Corresponding Carbons Region Protons Collapsed 1. 64 7ax 30.6 2. 43 7eq 30.6 3.21 5 82. 5 3.70 6 74. 2 5 ' 72. 0 4. 03 6a' 61 . 4 4.26 6b* 61 . 4 4. 78 8 75.7 1 • 100. 7 5. 10 2 ' 72. 9^ 3 ’ 68. 2b 4 ’ 70. 8b 5.44 2 95. 8 6. 01 4 70. 8 aSpectra were determined at 2 0 .123MHz using Me^Si as standard in CDCl^. bThese assignments are not absolute and may be interchanged. intermediate. In addition permethylsimmondsin was also hydrolyzed to yield the permethyl aglycone IVE. Compounds IVA and IVE are chloroform soluble and could be extracted directly from the hydrolysis mixture to give crystalline material. Compounds IVB-IVD, however, were unextractable even with n-BuOH and required chromatography over silica gel. All were obtained in crystalline form having negative optical rotations of similar magnitude and exhibiting end absorption in the UV spectra. Aglycone R 1 R2 R3 mp t>3D UV £ ) IVA H Me Me 139° -191° 208nm (4.12) IVB HH Me 151° -175° 207nm (4. 16) IVC H Me H 115° -189° 206nm (4.11) IVD H H H 184° -199° 206nm (4. 19) IVE Me Me Me 74° -144° 206nm (4.13) OR1 RIO, .O---- OR3 O R20' OR2 i1 O R 3 183 A side product from the hydrolysis of 6-0- demethylsimmondsin was also isolated and identified from its 1 mp, IR, and H NMR as 2-hydroxy-5-methoxyphenylacetic acid (VII). The substitution pattern of this compound provides conclusive evidence for the assignment of the methoxy substituent at C-5. Compound VII was previously isolated as a degradation product from the acid hydrolysis of simmondsin( LA), and was identified by conversion to the known 2, 5-dimethoxyphenylacetic acid and its corresponding lactone^. The IR spectra of the aglycones exhibit multiple bands associated with the carbonyl stretching frequency. This phenomena may be observed in both the infrared and Raman spectra, and is common to unsaturated 5 and 6-membered lactones in which the double bond is conjugated to the mmm ir IVEIVA IVB rvc IVD 184 carbonyl and bears an a-proton. The relative band intensities are dependent upon solvent polarity and temperature, but independent of concentration^. During the analysis of the NMR spectra, it was observed that several of the coupling constants had changed dramatically relative to those in the glycosides. This was expected since the iJ values of the glycosides are a direct consequence of *3) strain. On removal of the bulky glucose substituent the ring is allowed to flip back and forth between somewhat equally-balanced conformers. ,CN HO, M eO ' MeO OMe OMe I"' NC OH HO H OMe HO NC OH 3 OMe The conformer on the right allows for a nucleophilic attack of the hydroxyl group on the n itrile to form the imino-ester intermediate followed by hydrolysis to the unsaturated -lactone. Such a ring flip is easily verified by comparison of the H-7ax and H-7eq m ultiplicities and coupling constants. As observed in Figure 11:1-2, there is a dramatic change in the patterns of these two protons between the glycoside (right trace) and the aglycone (le ft). As mentioned earlier, both of the glycoside C-7 protons analyzed for a doublet of double doublets (ddd) which had the appearance of a doublet of triplets. This analysis was verified through double irradiation techniques, and the same pattern was observed in the permethylated and peracetylated compounds. However, in examining the conformation of the aglycone, we find that H-7ax is now trans-diaxial to two protons resulting in three large coupling constants of similar magnitude. This results in a double doubletd, which has the appearance of a quartet. The pattern for the equatorial C-7 proton of the aglycone is very complex and required double irradiation experiments for its full elucidation. It can be rationalized by noting that the proton is in a planar zig zag configuration with H-5 and experiences long range "M, W or tail-to-tail" coupling, generally 1-2Hz in magnitude 20 . This was confirmed by irradiating H-5 in 1VB (300MHz) and observing collapse of H-7eq (Figure 11:1-3) to a slightly NC OH OCH HO -H OGIu hP OH OMe H-7eq 186 Figure 11:1-2 Analysis of H-7ax Multiplicity in the NMR Spectrum. OH HQ OCH OH HO' OM« H-2 H-5 Figure 11:1-3 Analysis of H-7eq Multiplicity in the NMH Spectrum. 188 Table 11:1-8 1H NMR Data for Simmond3ia Aglycones IVA-Ea. H IVE IVA IVB IVC IVD 2 6 . 15 ,dd 5 . 98dd 5 .9 7 ,dd 5.99, d 5 .96 ,d ( 1. 9, 0. 6 ) (1. 9 ,0 . 6 ) (1. 6 ,0.7) (1 .9 ,0 .7 ) ( 1. 6 ) 4 4.52,d 4 .8 9 ,brd 4 .7 6 ,dd 4 .9 4 ,d 4 .7 5 ,db (3.5) (3.5) (3 .5 ,3 .5 ) (3.5) (3.5) 5 3 -87,m 3 .8 1 ,m 4 .2 3 ,m 3. 61 ,dddb 4 . 03»brdd buried (2 . 2, 3-5) ( 1. 3, 2. 6 , (2 .7,3-5) (2.7,3-5) 3 .2 ,3 .5 ) 6 3.7 4 ,ddd 3.8 7 ,ddd 3 .8 2 ,ddd 4 . 5 ,dddb 4 .2 1 ,dddb (2 .4 ,3 .9 , (2 . 2, 3 . 8, (2 . 6 ,4. 1, (2 . 7 , 3 . 8, (2 .7 ,4 . 1, 11. 8) 11.4) 12. 1) 12. 2) 12. 0) 7 ax 1. 57 ,ddd 1.5 7 ,ddd 1 .6 1 ,ddd 1.60,ddd 1.64,ddd ( 11. 8) (11.4) ( 12. 1) ( 12. 2) ( 12. 0) (11.4) (11.4) (11.4) (11.4) (11.4) (10.5) ( 10. 8) ( 10. 8) ( 10. 8) ( 10. 8) 7 eq 2. 51 ,ddd 2 . 53 ,ddd 2 . 5 ,ddq 2 .4 2 ,ddd 2 .4 0 ,ddd (3 .9 ,6 .7 , ( 3- 8, 6 .4, ( 1.3 ,4 .1 , (3 .8 ,6 .4 , (4 .1 ,6 .4 , 10.5) 10. 8) 6.4 ,1 0 .8 ) 10. 8) 10. 8) 8 5 . 02,ddd 5.11 ,ddd 5. 13,ddd 5 . 12,ddd 5. 13,ddd ( 1 .9 , 6 .7, (1 .9 ,6 .4 , ( 1 . 6 ,6.4, (1 . 9, 6 . 4, (1 . 6 , 6 . 4, 11.4) 11.4) 11.4) 11.4) 11.4) 4-OMe 3-30,s - - - - 5-OMe 3.44,3 3.43,3 - 3.45, s - 6 -OHe 3•37,s 3 .3 8 ,s 3.37,s - - OHC 5 .0 6 ,brs 3•93, d 3 .0 5 ,br 4 .9 9 ,brd (3 .2 ,0 -5 ) (2.9) - 4.97 , d 3 •9 ,br 3. 12, brs (3. 5, C-4 ) unknown aDetermined at 90MHz in acetone-d^ with TMS as standard. ^Pattern observed after D^O added. c Exchanges with D^O, 189 Table 11:1-9 Double Irradiation Experiments for Simmondsin Aglycone( IVA) . Proton Result After Irradiation H-2 5 . it H-8, ddd collapsed to dd(6.4,11.4) H-4 4. 89 H-5, dd collapsed to d(2.5) H-5b'‘ 3.81 H-4, d collapsed to s H-7ax, ddd collapsed to dd(11.4, 11.4) H-6 b,c 3.83 H-7eq, ddd collapsed to dd(6.4,11.4) H-7ax 1. 72 H-6 , ddd collapsed to buried m H-7eq, ddd collapsed to m H-8, ddd collapsed to dd(1.5,6.4) H-7eq 2.59 H-6 , ddd collapsed to dd(2.5)k H-7ax, ddd collapsed to dd( 11. 4 , 11.4) H-8, ddd collapsed to dd(1.5»11.4) H-8 5.11 H-2, ddd collapsed to s H-7ax, ddd collapsed to dd(11,4,11.4) H-7eq, ddd collapsed to dd(3.7,1l.4) OH 3. 21d 5-OMe 3. 48* 6 -OMe 3.44® aObtained at 300 MHz in CDCl^ with TMS after D^O exchange. bP artially buried. c Irradiated simultaneously. ^Exchanges with D2O. eEstablished by NOE experiments. 190 broadened doublet of quartets. This residual broadening was of interest since H-2 possessed a 0.6Hz coupling in addition to its normal coupling with the allylic C -8 proton. Irradiation of H-2 (IVB: 300MHz) produced not only the expected collapse of H -8 to a double doublet of double doublets), but also collapse of H-7eq to a very neat doublet of octets confirming the long-range, 5 -bond coupling of H-7eq with H-2. Thus, H-7eq couples with a ll protons in the molecule except H-4. Addition information from the ^H NMR spectra includes: (1) a 0.4 to 0.8ppm downfield shift of the proton on the carbon which is demethylated, and ( 2) non-application of the substituent chemical shift rule since the presence of the a,p-unsaturated lactone deshields the axial 5-OMe relative to the 6 -OMe group. This assignment has been v erified by Nuclear Overhauser Experiments (Figure 11:1-4 and 11:1-5) in which the methoxy at C-5 was found to interact with H-4 and H-5 (4.7 and 6.9% enhancement, respectively), whereas the methoxyl at C -6 interacts with H-5, H-6 and H-7eq ( 5 . 0, 5.8, and 2. 0% enhancement, respectively). The broad band and off-resonance decoupled NMR spectra were acquired for each compound, and together with information from single frequency ORD experiments, it was possible to make exact assignments for each carbon of the five aglycones (Tables 11:1-10 to 12). The two downfield singlets (173 and 170ppm) were assigned to H - 4 H-S 300 MHz 5 1 2 K 4.7% 3 2 K 6.9% 3 2 K J 1------1------1------1------1 1 I 1 I I i i I t t t i t i i i i I 5 4 3 Figure 11:1-4 Nuclear Overhauser Experiment of Aglycone IVA: C-5 Methoxy. 192 H -5 H -i 5 .0 % 8 K 5.8% 2j0\ 8 K 2 5 6 K 4 K J I 1------1______I______I______1 , 1 J______* *______I______ll-l______I__ 5 4 PPM Figure 11:1-5 Nuclear Overhauser Experiment of Aglycone IVA: C -6 Methoxy. 193 Table II—1;10 1^C NMR Data for Aglycones IVA-E3 . C IVE IVA IVB IVC IVDb 1 172.5(s) 173. 0(s) 173.0(s) 173.0(s) 173. 1 2 119.9(d) 116 . 9(d) 117. 0(d) 116.9(d) 117.0 3 167.8(a) 170.4(s) 170.5(s) 170.4(s) 170. 6 4 75.5(d) 66 . 3(d) 68 . 8(d) 6 5 .8(d) 69. 0 5 79.2(d) 80.9(d) 71. 6 (d) 84.4(d) 74. 9 6 76.5(d) 76.2(d) 7 5 .9(d) 66 . 3(d) 66.2 7 3 3 .8(t ) 3 4 .0( t ) 33-8(t) 37.7(t) 37.0 8 7 8 .3(d) 78.4(d) 78.4(d) 78.4(d) 78.5 4-OMe 5 7 .3(q) - - -- 5-OMe 59•3(q) 59•2(q) - 5 9 -2(q) - 6 -OMe 57.0(q) 57.0(q) 5 6 .7(q) — — aSpectra were determined at 20MHz using Me^^Si as internal standard in acetone-dg. ^Sample was saturated, but concentration was too low to allow collection of ORD or SFORD experiments. For multiplicities refer to Table I I : 1-12. 194 Table II: 1-11 NMR Data for Simmond sin AglyconetIVA)a . c IVAb IVA° 1 173*5(s) 173-0(s) 2 116.7(d) 116.9(d) 3 169.3(s) 170.4(a) 4 65.4(d) 66.3(d) 5 79.7(d) 80.9(d) 6 75.2(d) 76 . 2(d) 7 32. 6 (t) 3 4 .0(t) 8 7 8 .3(d) 78.4(d) 5-OMe 59. Kq) 5 9 .2(q) 6 -OMe 57.2(q) 57.0(q) aSpec tra were determined at 20MHz using Me^si as internal bCDCl^. °Acetone standard ”d6 * 195 Table 11:1-12 1 3C NMR Data for Aglycone IVDa . c I VDb IVDC 1 176.5Cs) 173. 1 2 117.9(d) 117.0 3 1 70. U s) 170.6 4 68 . 2(d) 69.0 5 73.4 d) 74.9 6 65.7(d) 66.2 7 35 .2 (t ) 37.0 8 79.8(d) 78.5 aSpectra were determined at 20MHz. b D^O using the spectrometer reference of a dioxane/D 20 standard (8C 67.4). cSaturated in acetone-d^ with TMS standard. 196 carbons 1 and 3, resp ectiv ely . These assignments are based on both the longer relaxation time of the 173ppm peak, and the fact that the same peak appeared as a doublet in a gated undecoupled spectra, whereas a broad multiplet was observed in the 170ppm region for C-3* On proceeding through the series from the permethylated to the totally demethylated aglycone, it can be observed that carbons 1, 2 , 3 and 8, and the methoxy carbons experience very little shielding from the various substituent changes. The relative shift changes for carbons 4, 5, 6 and 7 can be seen in Figure 11:1-6 which demonstrates two phenomena in agreement with the literature^: first, a consistent 9- 12ppm upfield shift is observed when the substituent is changed from MeO to HO; and, secondly, there is an average 2ppm downfield shift (range 0-4ppm) in the resonances of the carbons adjacent to the site of substitution. An exception to this is C -6 which remains constant between some of the substitutions. This may be explained by the fact that it lies directly across the ring from the extended 7r-system and experiences varying degrees of shielding which are stronger than or equal to to the small a-effects of the neighboring substituents depending on the conformation. Comparison of this data to the series of acetates shows dramatic similarities (Figure 11:1-7). Further comparison with 2,3,4,6-0-tetraacetyl-1-0-methyl- 0 -D-glucopyranoside 6 5 P P M ■ 197 Figure 11:1-6 Graphical Representation of Aglycone Carbon Absorptions. 198 (Vji) i and the carbon assignments obtained from a Birdsall plot of the residual coupling versus the proton chemical shifts, allowed assignments of all carbons in the peracetate series (Table 11:1-6 and 11:1-7). The same type of assessment and comparison with 1-O-methyl-/ 9-D- glucopyranoside allowed assignments for all carbons in the glycoside series (Figure 11:1-8 and Tables 11:1-1 and 1 1 :1-2 ). The only difference in the direction of shielding of any carbon within the 3 series pertains to carbon 4. This may be explained by the fact that the acetate has no / 9- e ff e c t . This resu lts in a loss of the methoxy / 9-effect and a net movement upfield. In the case of the glycosides and aglycones, the /9-effect of the -OH group is stronger than that of the methoxy and the net resu lt is a slig h t downfield shift on demethylation. The remaining compounds which were isolated include sucrose and (+)-pinitol(VIA). Sucrose was shown to be identical with an authentic sample by mp, c o -tlc , [° ]Df IR, and H NMR, and conversion to the octaacetate whose physical p i data were consistent with the literature . The last compound is the cyclitol methyl ether, (+)-pinitol(IVA) . Fraction G crystallized from MeOH:EtOAc (4:1) to give 90mg of crystals which melted at 184.5-185°, and on continued heating (195-215°), slowly became blood-red in color. C-4 C-5 C-3 C-2 72.9 71.8 71.3 VB 6 8 . 5 C-5 C-8 C-6 \C -4 8 2 . 5 IIIA 7 5 .7 7 4 .3 7 2 .8 7 2 . 0 7 0 .8 68.2 7 T- T^T / i * \ 1 / / 75.1 74.6 73.1 72 A 71,3 69.2 68.5 IIIB 7,7 _ J . - \ s- - \ i 1 ! 8 2 . 3 MIC 7 7 . 0 7 3 .0 7 2 ,4 7 1 .3 7 0 .3 6 8 . 3 6 6 . 7 I / \ MID 7 7 . 0 7 3 .2 7 2 . 6 7 1 .3 68.6 68.0 7 3 . 0 H 68.4 I T T 80 75 7 0 PPM Figure 11:1-7 Graphical Representation of Carbon Absorptions in the VO Peracetylated Derivatives. VO 0 C - 3 C - 5 ' c-a C - 4 76.3 73.7 70.2 VA 1 i C - 5 c-%>t C -6 \C -4 64.6 76.6 n.e 69.21 175.0 73 * IA 69.6 1 * I A 1 1 / I t / / i * I * 76.5 76.9 76.6 75.3 70.9 69.9 73.5 IB 7 T 6 5 .4 77,6 76.6 76.3 69.9 69.2 73.5 1C 66.5 i ! I'' ,1 77.9 76.7 76.3 73.6 70^ 70.1 69.5 ID ft—76.4 “I” r~ T 85 "E" 75 70 65 PPM Figure 11:1-8 Graphical Representation of the Glycoside Carbon Absorptions. ro o o 20.1 OMe OR VIA R - H VIB B - Ac OR Acetylation of VIA with acetic anhydride and pyridine at room temperature gave a pentaacetate(VIB) which showed five sharp singlets at approximately 2.Oppm in the NMR spectrum (Figure 11:1-53), each integrating for three hydrogens. A sixth three hydrogen singlet corresponding to the single methoxy was observed at 3*^7ppm. A double doublet at 3* 62ppm integrated for "* H and was assigned to the proton on the carbon bearing the methoxyl. The two coupling constants are large (approximately 9Hz each) indicating a trans-diax ial relationship with the neighboring protons giving rise to the following partial structure. The remaining three carbons allow for eight possible isomeric compounds. Four of these are immediately eliminated since an optical rotation of +64°(c=0.47, H 2O) was observed for compound VIB, indicating that a plane of symmetry was lac king. Ac OAc H The four remaining structures consist of two enantiomeric pairs, ie , (+/-)-pinitol and (+/-)-ononitol. Comparison of data from the isolated material to that r e p o r t e d ^ ' for the dextrarotary isomers showed agreement with that of (+ ) - p in it o l(VIA). The IR and 1H NMR were consistent with the structure of VIA. As a conclusion, Figure 11:1-9 presents a summary of the compounds discussed, and the chemical transformations used to establish their structures. 203 Table 11:1-13 Physical Data for the 0-Methyl Ethers of Inositol and Myoinositol. Methylcyclitol MP [«]D Pen taacetate MP [a]D (+)-Pinitol 186-7° + 65° 98-9° + 11° (5-0-Methylinositol) (H20 ) (EtOH) (+ )-0nonitol 172° +7° 122/ - 12° C 6 -O-Methylmyoinositol) (h2o) 131° (CHCI3) Isolated Material 185° + 65° + 13° (H20 ) (MeOH) HO OH HO OCH, H OH OCH HO (+)-Pinitol ( + )-Ononitol 204 CN CN HO, IIA {>60%} OR2 IIB (80%} IIC (78%) tU H2 IA IB IC M e XD 2N HCI CN Ac O A c Ac RIO i R1 R2 R RI R2 IIIA He He IVA H he he 11 IB Ac He IVB H H Me m e He Ac IW H He H IIID Ac Ac IVD H H H IVB Me he Me Figure 11:1-9 Summary of Simmondsia Glycosides and their Derivatives. EXPERIMENTAL* Plant Material. Simmondsia californ ica (jojoba) seeds were obtained from Anthony C. Waiss, Jr. at the Western Regional Research Laboratory of the U. S. Department of Agriculture in Berkeley, C alifornia. A voucher specimen is on f ile . Visualization Reagents. All simmondsin-related compounds (glycosides, peracetates, permethylates and modified aglycones) and carbohydrates were visualized on silica gel tic plates using p-anisaldehyde spray reagents or 1^ vapors. All compounds appeared as various shades of dark blue or v io le t after heating 5-10 minutes at 110°. The methyl cyclitol, (+)-pinitol, could only be visualized with 10% H^SO^ in Et20 and 30-60 minutes of heating at 110°, or spraying with alkaline AgNO^. Refer to Appendix for formulations. ft See Appendix on p .364 205 206 Isolation of Simmondsia G lycosides( IA-ID) . The 3eeds (855 gms) of Simmondsia californ ica were ground to 20 mesh size, packed into a glass percolator, and extracted in sequence with hexane, ethyl acetate and 95$ ethanol. This yielded 455, 10.5, and 106gm residues, respectively, for a total of 67% extractable material. The 95% ethanol fraction ( 105 gm) was dissolved in 500ml of distilled H20 and extracted consecutively with chloroform, ethyl acetate, and n-butanol (with backwashing) to afford 5.3, 2.5, and 8. 8gm fractions, respectively. The dried aqueous residue (90gm) was dissolved in 150ml MeOH at 40°C and an equal volume of ethyl acetate added. The mixture was stirred for one hour and filtered. After repeating the MeOH/EtOAc extraction a second time, there remained 25.1 gms of insoluble material. The soluble material ( 65 . 0gms) was composed of 5 major components by tic with 20% MeOH in CHCl^ on silica gel. The individual glycosides were separated by chromatography of 6 . 1gms of the MeOH/EtOAc solubles on a droplet counter-current chromatograph [Rikakikai Co.; Model DCC-A: pressure = 6-8.5kg/cm^; flow rate = 19ml/hr (53 drops/ml); 40ml/fraction; descending mode]. The following solvent system was U3ed for the separation: CHCl^:CH^CN:Me0H:H20 (4:1:4:2); the lower organic layer being used as the mobile phase. This afforded a total of nine fractions, as illustrated in Figure 11:1-1. 207 Isolation of Simmondsln( IA) . Fraction C, obtained from the DCCC (droplet counter- current chromatograph) separation, crystallized from MeOH/EtOAc (1/6) to give 338mg of granular crystals. This material was identical with that previously isolated from Simmondsia californ ica (chinensis): mp 96-8° (80% aqueous MeOH:acetone (1:2), 158-9° [MeOH:EtOAc:acetone (1:1:1)]; t i c , Rf=0.24, 20% MeOH in CHCl^; [a ]D _73°(c=0.19, MeOH); IR (KBr) vMAX 3380 (O-H, H-bonded), 2220 (C=N), 1645 (C=C), 1130-1050 (C-O-C), and 800 (trisubstituted olefinic C-H); UV (MeOH) XMA)£ 21 7nm (log « =3.95); CD (c = 4 . 1 9x 1 0"3M, MeOH) l^ ]250 °» ^ ^ 226 “30,000, t0]2oo - I 1*, 000; 1H NMR (DgO) , refer to Table 11:1-1; 13C NMR (D20), refer to Table 11:1-1. Isolation of 5-0-Demethylsimmondsin(IB). A 626mg fraction was further purified by reverse phase chromatography on the DCCC using the lower phase of CHCl^:CH^CN:MeOH:H2O(4 : 1:4 :2) as the stationary phase: pressure = 9kg/cm ; rate = 10drops/min; tube volume = 200drops (4.1ml). After 140ml of solvent was eluted, a 535mg fraction was obtained. Final purification was accomplished by chromatography of 175mg of residue over 19gms of silica gel 60 (230-400 mesh; E.M. Co.) using the lower phase of the DCCC solvent system plus 5ml of MeOH per 100ml of solvent. After 95ml eluted, 1G5mg of pure material 208 was obtained. 5-0-Demethylsimmondsin(Hi) is characterized as a n o n -cry sta llin e, hard glass which becomes a viscous syrup above 93°» t i c , Rf = 0.26, lower phase of CHCl3;CH3CN:Me0H:H20 <4:1:^:2); [a]D -8 5 .7°(c=0.205, MeOH); IR (KBr) vMAX 3^100 (0-H, H-bonded), 2220 (C=N), 1640 (C=C) , 1075-1040 (C-0 str , and O-H bend), 801cm"1 (C=C-H); UV (MeOH) XMAX 217nm (log t = 4.02); 1H NMR (D2o), refer to Table 11:1-1; 1^C NMR (D^o), refer tc Table 11:1-1; mass spectrum CEI), m/£ 361 CO.2%, M+) , 330.1193 (0.7%, M-OMe; Cl4H20NOg requires 330.1189), 182 (6%, M-Glu), 149 (52%), 122 (88%), 94 (37%), 74 (35%), 61 (37%), 43 (100%); (Cl, isobutane), m/e 362 (1.1%). Isolation of 6-0-Demethyl 3immondsin( IC). Fraction D crystallized from MeOH/EtOAc (1:1) to yield 152mg of small rosette-like crystals. This material was identified as 6-0-demethylsimmondsin(I£) having the following physical constants: mp 211- 212°; tic, Rf=0 . 25 , 28% MeOH in CHC13; [a]D -109°< c = 0.226; H2o); IR (KBr) v MA)( 31*110 (0-H, H-bonded), 2235 (C=N), 1647 (C=C), 1110-1020 (C-0 and 0-H), and 795cm_1 (C=C-H); UV (H20) Amax 2l8nm (log £ = 4.03); "'h NMR (CD3CN), refer to Table 11:1-1; 13C NMR (D2o), refer to Table 11:1-1; mass spectrum (El), m/je 330.1193 (0.4%, M-OMe; C1J4H20NOg requires 330. 1189), 182 (4%, M-Glu), 163 (13%), 133 (14.5%), 109 (13%), 78 (26%), 74 (100%), 60 209 (39%), 43 (53%); (Cl, isobutane), m/e 362 (0.6%). Isolation of 5, 6 -D l-0-Demethyl 3lmmondsin(ID) . A 900mg fraction, in which H) was the major component, was further purified by reverse phase chromatography on the DCCC using the lower layer of CHCl^:CH^CN:MeOH:H20 (4:1:4:2) as the stationary phase: pressure = 5 kg/cm^ (71psi); tube volume = 3 . 2ml (200 drops); two individual passes under the same conditions. After the second separation, a 655mg peak emerged after 85ml. This material (I44mg) was further purified by passage over 25gms of Sephadex LH-20 (1.7 O.D. x 44.5cm) using 90% EtOH as solvent: flow rate = 12drops/min; tube volume = 3ml (168 drops). Further chromatography was required on a cellulose column. The c ellu lo se column was prepared by making a slurry of 44gms of Avicel (superfines, FMC Corp.) in 110ml of the upper layer of EtOAc:Pyr:H20(2:1.06:2), and packing the column under pressure (15psi) in approximately 1. 5 cm layers. Each layer was allowed to fully settle and then firmed down with a tamping rod fitte d with a cork. Total bed volume was 1.9 O.D. x 25.5cm. A final pressure head of 15 psi was applied and the column allowed to elute and eq u ilib riate for 24 hours. A 38mg sample of the above mixture was applied with minimum solvent. A 12.5psi pressure head was applied for elution: flow rate = 210 4drops/min; tube volume = 3ml (125 drops); void volume = 36ml (determined with Aviation Oil Blue). After 72ml of solvent eluted, 32.6mg of a pale yellow viscous oil was obtained. The yellow impurity was removed by chromatography of 32mg of material over 2gms of silica gel 60 (230-400 mesh) using the lower layer of CHCl^CH^CN-.MeOH-.^O (4: 1:4:2) as the eluting solvent to give 28.3mg of 5,6-di-O- demethylsimmondsin (IJD) as a non-crystalline, hard glass: melting range of 90-107°; tic, Rf=0.42, EtOAc:Pyr:H20 (2:1.06:2; upper layer), or Rf=0.18, CHCl^:CH^CN:MeOH:H20 (4:1:4:2; lower layer); [a]D -90 o( c=0.16, MeOH); IR (KBr) VMAX 3380 C0-H» H-bonded), 2220 (C=N>, 1640 (C=C), 1100-1035 (C-0 str , and 0-H bend), 800 cm”1 (C=C-H); UV (MeOH) *HAX 220nm (log * = 3-98); 1H NMR 13C NMR (D20), refer to Table 11:1-1; mass spectrum (El), m/e 347 (1.2%), 185 (4%), 168 (11%), 162 (7%), 149 (38%), 133 (40%), 122 (51%), 116 (21%), 94 (18%), 85 (28%), 73 (91%), 60 (100%); (Cl, isobutane), m/e 348 (14%). Isolation of (+)-Pinitol(IVA). Fraction G crystallized from Me0H:Et0Ac (4:1) to give 98mg of crystals. This material was identified as (+)-pinitol(IVA) having the following physical constants in agreement with literature sources18 (Table 11:1-13): mp s 184.5-5°; WD +64.5°Cc = 0.47, H20); IR (KBr) vMftX 3400 (0- 211 H), 1000-1500 (C-0); 1H NMR (D^O, DSS), «H 3-58 (3H, S, -OMe), 3*2-4.0 ( 6 H, complex). Isolation of Sucrose. Fraction H (1.416gms) crystallized from MeOH:EtOAc (2:1) to give 412mg of granular crystals, identical with an 1 authentic sample of sucrose by mp, co-tlc, CJDr IR, and H NMR, and consistent with literature values^. Permethylation of glycosides IA-ID 2fS. A 421mg sample of simmondsin(IA) was dissolved in 8.5ml of dry DMF (10-25x the weight of the glycoside) followed by 0.3ml of Mel (3eq per OH group) and 521mg of Ag 20 (2eq per OH group). The suspension was stirred at room temperature under a dry argon atmosphere for 91 hours during which further additions of Mel and Ag 20(as above) were made approximately every twenty-four hours. When the reaction was complete, as shown by t ic , 20ml of chloroform was added and stirring continued for 20 minutes. The suspension was then suction filtered, and the yellow precipitate washed three times with 10ml portions of chloroform. The filtrate was extracted with H 2O and evaporated to give 560mg of a viscous golden-red o il. Chromatography over 30gms of s ilic a gel (EM; 230-400 mesh; 6 % H20) with 1.5J MeOH in CHCl^ gave a 170mg of pure oil identified as permethylsimmondsin(II) . 212 After rechromatography of other column fractions, a total of 251mg of material was obtained corresponding to an overall yield of 50 . 2%. Permethylsimmondsin (I_I_) was obtained as a colorless oil; t i c , Rf=0.27i 2% MeOH/CHCl^; [a ]p -63° C-0-C); UV (MeOH) * MAX 210nm (lo g « = 3.88); 1H NMR (CDClg), refer to Table 11:1-3; 1 NMR (CDCl^), refer to Table 11:1-3; mass spectrum (El): m/e 210 (27.4%, M-C1QH1gO^), 178 (26%, M-C1qh 1906 , -MeOH), 101 (93*), 88 (100%), 75 (64%); MS (Cl, isobutane), m/e 446 (19*9%, M-H+). Analysis. Calcd for C21H35 NOg: C, 56.6; H, 7.92; N, 3.14%. Found: C, 56.17; H, 3.08; N, 3-05%. Higher yields of permethylated products could be obtained if additions of Mel and Ag 20 were made every 8 hours and the reaction stopped after a total of 48 hours. 5 - 0-demethylsimmondsin(Ij3, 30mg), 6 - 0-demethylsim- mondsin(I£, 26.3mg) and 5 ,6-di-0-demethylsimmondsin(ID, 21mg) were each permethylated by the above procedure (8 hour routine) to give permethyl simmondsin (I_I) in yields of 80.2, 78.1, and 85.6%, respectively. In each case, the isolated product was identical to that obtained from simmondsin by c o -tlc , [®3p IR, NMR, and MS. 213 Simmondsin Pentaacetatet IIIA). A 500mg fraction of simmondsin(LA) was dissolved in 5 ml of pyridine and 4ml of acetic anhydride and stirred at room temperature overnight. The reaction was complete after 16 hours as shown by tic (Rf =0.63, 10% MeOH/CHCl^), and 10ml of H2O added and stirred for 30 minutes. The reaction mixture was extracted with CHCl^ (3x10ml) and the combined CHCl^ backwashed with (10ml). On evaporation, the CHCl^ layer gave 732mg of oil which crystallized from EtOAc:Hex to give 677mg of simmondsin pentaacetate(IIIA) as long white needles whose physical data was identical with literature values: mp. 167-7-5° (EtOAc : Hex ); [o]D _in°(c = 0.45, MeOH); IR (CHC13) vMAJ( 3020 (C=C-H), 2225 ( C=N), 1760 (C=0), 1642 CC=C), 1140-1050 CC-0); UV (MeOH) *MAX 217nm (log c = 4.07); NMR, refer to Table 11:1-4; 1 NMR, refer to Table 11:1-6; CD (c=3 - 21x10-3M, MeOH) 1012H5 °* Ctf]222 - 44>000» [tf]200 -12,500; mass spectrum (El), m/e 331 (23%, M- C12H16n05) » 238 (11% M-GLU), 178 (9%), 169 (100%), 163 (14%), 146 (17%), 137 (44%), 115 (24%), 109 (66%), and 60 (40%); (Cl, isobutane), m/e 586 (37.4%). 5-0-Demethylsimmondsin Hexacetate(IIIB) . A 117mg sample of IIIB was dissolved in 2ml of pyridine and 1ml of acetic anhydride and 3tirred at room temperature overnight. The reaction was quenched after 21.5 hours by 214 addition of 2ml of MeOH and stirring for one hour. The solution was evaporated to dryness, and partitioned between CHCl^ and H20 . Evaporation of the CHCl^ layer gave 213mg of a very viscous pale yellow o il which refused c r y sta lliz a tio n . Chromatography over 35gms of s i l ic a gel 60 (230-400 mesh) using 1.5% MeOH in CHCl^ as the eluting solvent produced 128mg of pure 5 - 0 -demethylsimmonsdin hexaacetate(IIIB) as a colorless, non-crystalline, hard glass having the following physical data: melting range 62-79°; tic, Rf= 0.42, 2% MeOH in CHCI3 ; [a]D -39-5° (c= 0.385, MeOH); IR (CHC13 ) vMAX 3028 (CH3-C=0), 2225 (C=N), 1765 (0=0), 1643 (CbC), 1205-1240 (br, C-0); UV (MeOH) * MAX 213nm (lo g * = 4.09); NMR (CDCl^), refer to Table 11:1-4; 13C NMR (CDCl^), refer to Table 1 1 : 1- 6 ; mass spectrum (E l), m/e 347 (3%, M-c 3H1605N), 331 (22%, M- C13H16°6N )* 266 (41*’ M-C 6Hn 05 ), 207 (13%), 174 (16%), 169 (100%), 164 (31%), 157 (20%), 146 (32%), 139 (23%), 109 (53%), 98 (26%), 83 (100%); (Cl, isobutane), m/e 614 (3%). 6-0-Demethylsimmondsin Hexaacetate(IIIC) . To 2ml of pyridine and 1.5ml of acetic anhydride was added 78mg of 6- 0 -demethylsimmondsin (IC^) with stirring at room temperature. After 18.5 hours, 5ml of MeOH was added, stirring maintained for 30 min, and the solution evaporated to dryness. Chromatography over silica gel 60 (230-400 215 mesh) using 1% MeOH in CHCl^ as solvent (Rf=0.2) gave pure 6-0-demethylsimmondsin hexaacetate(IIIC) . This material precipitated as an amorphous solid by dissolving in a minimum amount of MeOH/i-P^O and cooling at 0° for 5-6 hours, further cooling in a dry ice-acetone bath for 30 min, and suction filtering. The following physical data was obtained: melts slowly above 82° (softens above 64°); - 3 3 .9°(c= 0.3375, MeOH); IR (CHC13) vMAX 3020 (CH3-C=0), 2222 ( C=N ); 1765 (C =0); 1642 (C=C); 1210-1240 (C-0); UV (MeOH) ^MAX 219nm (log « = 4.15); ^ NMR (CDCl^), refer to Table 11:1-4; 13C NMR (CDCl^), refer to Table 11:1-6; mass spectrum (E l), m/e 613 (0.1%, M+), 347 (1.5%, M-C13H1gO^N), 331 (32%, M-C13H1606 N), 266 £18%, M-C6 Hn 05), 169 ( 100%), 146 (28%), 109 (62%), 98 (24%), and 60 (31%); (Cl, isobutane), m/e 614 (9-4%). 5 , 6 -Di-0-Demethylsimmondsin Heptaacetate(IIID) . With stirring at room temperature, 68 mg of HID was dissolved in 1ml of pyridine and 0. 5ml of acetic anhydride. After 21 hours, 3ml of MeOH was added and stirred for 1 hour. The solution was evaporated to dryness, and then applied to a 32gm s ilic a gel 60 PF-254 bed using 1% MeOH in CHCl^ as solvent. This gave 114mg of HID as a clear, colorless, hard glass having the following physical data: melting range 62-74°; tic, Rf=0.44, 2% MeOH in CHC13; [a]D 216 - 3 3 .5°(c= 0.325, MeOH); IR (CHC13) vMAX 3020 (CH3-C=0), 2225 ( C=N), 1760 (CrO), 1650 (C=C), 1205-1250 (br, C-0); UV (MeOH) Xjy|AX 214nm Clog * = 4. 07); ^ NMR (CDCl^), refer to Table 11:1-4; 1 NMR (CDCl^), refer to Table 11:1-6: mass spectrum CEI), m/e 642.2053 (1.1%, M+H, ^28^36^16 reQuires 642.2034), 641 (0.1%, M+), 581 (3%, M-AcOH), 568 (1.1%, M- CH3C02CH2), 331 (52%, M-C14H16 N0?) , 294 (40%, M-Glu), 182 (52%), 169 ( 100%), 140 (84%), 115 (52%), 87 (61%), and 43 ( 100%). Sucrose Octaacetate. A 55mg sample of sucrose (isolated from jojoba) was dissolved.in 2ml of pyridine, 1ml of acetic anhydride and stirred overnight. The reaction was quenched by addition of 5ml of MeOH and stir rin g for 30 min. Evaporation in vacuo gave 103mg of colorless oil. A total of 73mg crystallized from 95% EtOH at 4° as long needles whose mp, [ Q]D, IR, and NMR were consistant with literature values'^. Acetylation of (+)-Pinitol(VIA) . A 68 . 6mg sample of (+ )- p in it o l(VIA) was dissolved in 1. 0ml of pyridine and 0. 5 ml of acetic anhydride at room temperature with stirring. After 26 hours, 5ml of MeOH was added, stirred , and the solution evaporated to dryness. The oil refused crystallization and was chromatographed over 217 12gms of s ilic a gel 60 (EM, 230-400 mesh, 6 % f^O) using CHCl^ and 0.5% MeOH in CHCl^ as the eluting solvents. A colorless, viscous oil (106mg) was obtained and characterized as ( + )-pinitol pentaacetate(VIB)^: +12.6°(c= 0.47, MeOH); IR (CHC13 ) 302O (CH3-CO), 1760 (0=0), 1380 (CH3-C0), 1210-1240 (C-0), 1060 (C-0); 1H NMR (CDC13), 8h 5.5-5.1 (5H, m), 3.62 ( 1H, dd, J=9.2, 9.2Hz), 3.47 (3H, s, -OMe), 2.17 (6 H, s, Ac), 2.09 (3H, s, Ac), 2.05 (3H, s, Ac), 1.99(3H, s, Ac). Preparation of 1-0-Methyl-2,3.4, 6-Tetraacetyl-fl-D- Glucopyranoside (VB). A 54 0mg sample of 1-0-methyl-£-D-glucopyranoside(VA , Sigma) was dissolved in 10ml of anhydrous Pyr (5°C) and 5ml of acetic anhydride ( 5 °C) and stirred for 19 hours maintaining the temperature at 5°. The reaction was quenched by addition of H£Q ( 10ml), stirring for three hours (5°), and evaporated under vacuum to 5ml. The solution was diluted with 50 ml of H20 and extracted with CHC 13 (5 x 10ml). The combined CHC13 was washed with 50ml of H2O, and evaporated to give 1095mg of a viscous oil which crystallized from MeOH to give 926mg of long needles (92% yield) identified as 1 -0-methyl-2 , 3, 4, 6 -te tr aacetyl-/J-D- glucopyranoside(VB). The [Q3D and mp were identical with that previously rep orted ^ . The IR, UV, MS, "'h HMR and CMR 218 were consistent with the structure of VB. Hydrolysis of Simmondsin(IA). Simmondsin ( 56 lmg) was dissolved in 25ml of Ac0H;HC1:H20 (4:5:1) and heated on a steam bath for 8 hours at which time i t was shown by tic (Rf = 0.39, 10* MeOH in CHC13) to be completely hydrolyzed. The solution was evaporated onto 2gms of s ilic a gel and packed onto a column containing 60gms of s i l ic a gel 60 (70-230 mesh, EM). Chloroform and increasing amounts of MeOH in CHCl^ were used for elu tion . A 61.5mg fraction was obtained which cry stallized from EtOAc/Hex as long needles (31mg). The material was identical by mp, IR, NMR (Table 11-1:8), and MS to that previously reported for simmondsin aglycone(IVA)^. Unreported data includes: -1 91 . 4°(c=0.21, MeOH); UV (MeOH) *MAX 208nm (log c = 4.12); CD (3-83 x 10~3M, MeOH) ^ ^270 ^ ^240 -7600(shl), £^207 “ 56, 000 and t 1 -16,000; 13C NMR (dg_acetone), refer to Table 11:1-10. Ac id Hydrolysis of 5-0-Demethylsimmondsin(IB) . A 305mg sample of _I8 was dissolved in 30ml of 2N HCL and heated for three hours on a steam bath. The 2-neck flask contained a condenser and N 2 inlet. When complete (tic, 165 MeOH/CHCl^ on silica gel), the reaction mixture was cooled, and solid NaHCO^ slowly added to the rapidly stirred 219 solution until pH 7 was obtained. The solution was transferred to a flask containing lOgms of silica gel 60 (230-400 mesh) and evaporated to dryness. The silica was packed into a 1.75cm I.D. column which contained 5gms of s i l ic a gel 60 packed in 16% MeOH in CHCl^. The column was then washed with 150ml of solvent which gave 6 lmg of yellow oil on evaporation. This was chromatographed on 12gms of silica gel 60 (230-400 mesh) using 8% MeOH in CHCl^ as elu en t. After 54ml emerged, a 51 mg peak was eluted which c ry sta llized from 10% MeOH/EtOAc to give 45mg of cry sta ls. R ecrystallization from 10% MeOH/EtOAc gave 35mg of fine needle rosettes: mp 150-1°; tic, Rf=0.47, 16% MeOH in CHC1^; [a]D = - 175° (0-H), 1770 (C=0), 1775 (C=0), 1725 (C=0), 1655 (C=C), 1115 (C-0); UV (MeOH) XMAX 207nm (log « = 4.16); CD (c = 3.2 x 10-3m, MeOH) [03265 0, [0 ]245 -3900(shl), E* ] 212 -52,000 and [0 ] 195 -7800; 1H NMR (acetone-dg), refer to Table 11:1-8; ^ 3c NMR (acetone-dg), refer to Table 11:1-10; mass spectrum (El): m/e 200.0689 (40%; CgH 120 5 requires 200.0685), 182 (26%, M-H20 ), 150 (30%, M-H20, -MeOH), 140 (13%), 124 (26%), 122 (27%), 112 (25%), 109 (23%), 96 (23%), 85 (25%), 74 ( 97%), and 43 ( 100%). 220 Acid Hydrolysis of 6-0-Demethylslmmondsin(IC). A 239mg sample of was dissolved in 24ml of 2N HCl and heated on a steam bath for 3 hours. The 2-neck flask was fitted with a cold water condenser and Ng inlet. When the reaction was complete, as shown by t ic (16% MeOH in CHCl^ on silica gel G), the rapidly stirred solution was neutralized to pH 7 by slow addition of solid NaHCO^. The pale yellow solution was transferred to a 125ml erlenmeyer flask containing 10gms of silica gel 60 (230-400 mesh) and evaporated to dryness under vacuum at 40°C. The dried s ilic a was then packed into a 1.75cm I.D. column which contained 5 gms of s ilic a gel (230-400 mesh) packed in 16% MeOH/CHCl^. After the s ilic a gel had se ttle d , the column was washed with 150 ml of solvent which gave 111 mg of yellow oil on evaporation. This material was further chromatagraphed over 10gms of s i l ic a gel 60 (230-400 mesh, 1 O.D. x 33cm) using 8% MeOH/CHCl^ as solvent. After 77ml of solvent eluted, 41mg of material was obtained. Further chromatography over 12gms of silica gel 60 using 9% acetone in EtOAc (R^sO.21) as solvent gave a co lorless oil. This crystallized from 5% acetone in EtOAc as needles and was characterized as 6 - 0-demethylsimmondsin aglycone (IVC) having the following physical constants: mp 115-5.5°; [oJD -189°(c=0.245, MeOH); IR (KBr) vMAX 1750 (c=0), 1659 (CsC), 1070 (C-0-C), 800 (C=C-H); UV (MeOH) *MAX 206nm (log = 4.11); CD (c = 3.0 x 10" 3M, MeOH) [e j^ o , [G]245-9700, 221 [ e 3237_9300 (7.6%, M+, C9h120 5 requires 200.0685), 182 (15%, M-H 20 ), 168 (15%, MeOH), 155 (20%), 150 (12%), 141 (63%), 122 (28%), 112 (45%), 109 (63%), 74 ( 100%). Also, a 53mg peak eluted ahead of 6-0-Demethylsimmondsin aglycone (elution volume-38ml). The heavy oil crystallized from EtOAc/Hex to give 34mg of rosettes and was id en tified as 2-hydroxy-5-methoxy phenyl acetic acid(VII) by comparison of its mp, IR and NMR to literature values^. Hydrolysis of 5 ,6-Di-0-Demethylsimmondsin(ID) . The 5 ,6-di-0-demethyl glycoside, (_ID), was hydrolyzed by addition of 1.Ogm of material to 78ml of 2N HC1 in a 2-neck flask fitted with a cold-water condenser and argon inlet, and heating on a steam bath for 3-25 hours. This was then neutralized by addition of solid NaHCO^ to the rapidly stirred solution. The material was adsorbed onto 15gms of silica and packed on top of 5gms of silica, previously packed in a glass column using 20% MeOH/CHCl^ as solvent. After 225ml of solvent eluted, and the solvent evaporated, 233mg of an orange oil was obtained. This material was further chromatographed over 32gms of s ilic a gel 60 (230-400 mesh) using 16% MeOH in CHCl^ (Rf =0.21) as solvent. After 222 139ml had eluted, a 103mg peak emerged containing a c o lo r le ss, viscous o il which crystallized from 5% MeOH/acetone as colorless prisms: mp 183-4°; [a]^_i9g°(c = 0.185, MeOH); IR (KBr) vMAX 348O (0-H), 3380 (0-H), 2940 (C- H), 1750 (C=0), 1665 (C=C), 1010 (C-0), 800 (C=C-H); UV (MeOH) *MAX 215nm (log e = 4.19); CD (c=5.48 x 10"3M, MeOH) [tf]275 °t [*]245 “6000(shld), [tf]210 “59,000, [*3195 -10,000; 1H NMR (acetone-d6), refer to Table 11:1-8; 13C NMR, refer to Table 11:1-10 (d^-acetone), and Table 11:1-12 (D20); mass spectrum (El), m/e 186.0533 (3.9*, M+, cgH100^ requires 186.0528), 168 (9%, M-H20), 150 (14%, M-2H20), 127 (100%), 122 (54%), 109 (97%), 84 (18%), 60 (35%), and 44 (69%). Methyl Simmondsin Aglycone(IVE) . Simmondsin (1.0gm) was dissolved in 20ml of dry DMF, and 1 .24gms of Ag20 added followed by 0.72ml of Mel. The solution was stirred in a room temperature water bath under a dry Argon atmosphere. Similar additions of Mel and Ag2o were made every 10 hours. After 48 hours the reaction was complete by tic, and 20ml of CHCl^ added, stirred for 10 minutes and filtered. The precipitate was washed well with CHCl^ (5x15ml), and the filtrate reduced to 15ml. This was dissolved in 40ml of H20 and extracted with CHCl^ (4x15ml), and the combined CHCl^ washed with 20ml of H20 . On evaporation, the CHCl^ gave 1. 21gms of pale yellow oil whose major component was permethylsimmondsin( I_^) by t ic and NMR. This was transferred to a 200ml RB flask, 96ml of 2N HC1 added, and heated on a steam bath for 2.25 hours. After removing and cooling to room temperature, solid NaHCO^ was added to the rapidly stirred solution until pH7 was obtained. Extraction with CHCl^ and evaporation, gave 746mg of red-orange o il which was chromatographed over 70gms of silica gel 60 (230-400 mesh; 6 % H20 ) using CHCl^ as solvent. After 225ml eluted, a 104mg peak emerged which cry sta llized from 25% EtOAc/Hex as long rods (54mg). This material was identified as methylsimmondsin aglycone (IVE): mp 74-4.5°; tic Rf=0.42, 2% MeOH in CHC13; [a]D- 144°(c=0.165, MeOH); IR (CHC13) vmax 3010 (C=C-H), 2937 (C-H), 1760 (C=0), 1660 (C=C) , 1460 (CH2 scissor), and 1050-1150 (C-0); UV (MeOH) *MAX 215nm ( log « = 4.06); CD (c=3* 29x 10~3M, MeOH), [0327qO, ^ ^240 " 5800, [^3233 -5100(min), £^207 “52,000, -15,000; "*H NMR (d^-acetone) , refer to Table 11:1-8; "*3C NMR (d^-acetone), refer to Table 11:1-10; mass spectrum (El), m/je 229 (0.7%* M + 1 ), 228. 1005 (6.0%, M+, ^^H^gOg requires 228.0998), 155 ( 100%), 126 (77%), 88 (20%), 75 (30%), 69 (18%), and 44 (13%). Continued elution of the silica gel column gave a 182mg component after a total of 365ml of solvent eluted. This material was identified as permethylsimmondsin by tic and NMR. Changing the solvent to 1% Me0H/CHCl 3 and elution of 22H 220ml gave a 154mg of oil which readily crystallized from 10% EtOAc in hexane as long needles and was identified as 2, 3* 4, 6 -tetramethyl-a-£-glucose: mp 101- 2°; tic RfsO.^e, 4% MeOH in CHC13; [a]D+i20(c=0.228, MeOH); IR (CHCI 3) vMAX 3595 (0-H), 3400 (0-H), 1200-1050 (C-0), 855 (anomeric C-H bend of ct-sugars); ^ NMR (CDC13), 5H 5.33 (1H, dd, H -1) , 3.85 (m), 3-68 (m), 3.2 (m), 3-63 (3H, -OMe), 3*54 (3H, -OMe), 3-52 (3H, -OMe), 3.40 (3H, -OMe); 13C NMR (CDC13), $c 91.0, 83.4, 82.4, 79.9, 71.7, 70.3, 60.8, 60.4, 59.3, and 58-9- Tt iu e 111 IR Spectrum Figure (KBr) 11:1-10 Simmondsin(IA) of 225 90 MHz U u _ j ^ A j M 11 I II I t 11 L I I [ I 1 I I 1 I 1 I I t I 1 1 I I n It I I M I I II I I I I II I i II II I I I I I I I I I 6 ^ 4 3 2 1 PPM rv> rv> Figure 11:1-11 H NMR Spectrum (Dj®) Simmondsin(IA) o> 20 MHz Wl wv 4w |1 u j i F H r | n n r i T i T | T T r r > r i T | T T r r T n n r [ i T m TTTT|TTiTTTii I | | n r T M M n p i m t i i I | f! I i i i f r r ^ i r i n i I I J I i n i i i T i j i i T i T i n r p i r T i U M f i m ii I I | i n i r p i u | >i | n m i i i F | m n i i i i | n i n r r f T p n M H i T j T i n u n i 200 150 100 50 13 227 Figure 11:1-12 C NMR Spectrum (D20) of Simmondsin(IA) Figure 11:1-13 IR Spectrum (KBr) of 5-0-Demethylsimmondsin(IB) 228 9 0 M H z _L aA______AA______... ■J i * L„. 4 I 1 1 * > » I | ■ < _J 4 1 i ■■ X J »______^ 3 2 ^ P PM 229 Figure 11:1-14 NMR Spectrum (D2 O) of 5-0-Demethylsimmondsin(IB). 20MHz |iiiiMiii[iiiiiiiiiiiiiM>ii<[iiii)iiii|ii)iiiiii|irm inr[nn iiiii|iiiniin]innnii|iiinn»i|iniiiiii[iiiiiiiiiiiim nm iir» w niiinnn|H)iiim |iiiiuiii|iinriiii|)iiim ir|wnw iiipnirin|iiiniiii 200 150 100 50 PPM 230 Figure 11:1-15 13C NMR Spectrum (D20) o£ 5-0-Demethylsimmcmdsin(IB). •CaCCHT itAMlwtMiON i ue 111 IR Spectrum (KBr) 6-0-Demethylsimmondsin(IC). 11:1-16Figure of UJ ru 9 0 MHz 1 a j \A lv _ A V J W . -* * ■- » * - L . j — J i i 1 — j- i * -L_I_ LJ-J PPM 232 Figure 11:1-17 NMR Spectrum (D20) of 6-0-Demethylsimmondsin(IC) 90 MHz |iiiiiiiii|iiiiiriii|i>iinm jirwmn| iinn|iiniiiii|rm irTii| iiiijirm rn[nwiiHr|iriiiiiii|>iiitiiii|iiiiiini|nn(tni|)ii)iiiii|iiiiiimjii i[iiimm|i iiiiriiHmniiim iliiinitir 200 1 5 0 1 0 0 5 0 PPM 233 Figure 11:1-18 *3C NMR Spectrum ID2°) 6-0-Demethylsimmondsin(XC) WAVMMMI CM-’ i figure 11:1-19 IR Spectrum (KBr) of 5 , 6 -Di-O-demethylsimmondsin(ID) nee S 90MHz a A_A a . [ A I I l_ L I m I L l l l I 1 I I I 1 I I I I 1 I I I L 1 III 1 It I I I 1 I i_l I 1 I I L L L I 235 l’ i ]iiiiTiiHfim iiitr|iiiiiiiii|iiim iii|Hiim ii|iiim iii|iiiiiiiii|riiiniii|iiim iii|iiHTtrn|nn i iim iiiiirin |Triu m )jiriiiiin |iMiiTm ]iiiiirMi|iiiiin n |in iim i 150 100 50 PPM 13 Figure 11:1-21 C NMR Spectrum (D2 O) of 5,6-Di-0-demethylsimmondsin(ID). TftAtllMrssiOfi iue 112 I Setu (H1 o Permethy1simmondsin(II) (CHC1 of Spectrum IR ) 11:1-22 Figure 237 90M H z 238 Figure 11:1-23 H NMR Spectrum (CDCl^) of Permethylsimmondsin (ij) 20MHz jTTTTmi)|nn niii|iiiiinii|)niii)H|iiiirTrFi|nmirirjimnirTpniirro|'i iimiiii| [HMTniijMniiiii| |iii i|iitPiin ifm n M H [ 1 5 0 1 0 0 5 0 figure 11:1-24 13C NMR Spectrum (CDCl^) of Permethylsimmondsin(II). 239 TtKNIMIlSlDN iue 112 IR Spectrum (CHCl^) 11:1-25Figure Simmondsin of Pentaacetate(11IA) 240 300M Hz _JL ILL J . I J . » —■— t A. A -A. > I ... A M. 1 t - J- — -A. PPM Figure 11:1-26 1H NMR Spectrum (CDC13) of Simmondsin Pentaacetate(I1IA). 90 MHz i»i *>*Yr LI jm i iTttrjTi imiji 11» n ii ii i rjimnTif[i mi ii11 [ i iniiiTi iTrnnTi]iirMriiFniTfiim j»iiiim|nnmir|inrimipininininiiiTii|iin»n i|iiiiim rjiiiimii|m iiirrijrinTf»r|iHTm »|iinrfT 200 100 50 PPM 2U2 Figure 11:1-27 13c NMR Spectrum (CDClj) of Simmondsin Pentaacetate(IIIA). iu e 112 IR Spectrum 11:1-28 (CHCl^)Figure 5-O-Demethylsimmondsin of Hexaacetate(IB). FftCENf T«4MtMlSStON 1000 MOO m |M 0 UV oe to 243 90M H z J l/l ^ A J LI I I I LJ Ll I I L I 1 I I 1 I 1 I 1 L I 1 I I 1 I 1 I I I I I I I I I I I I I I l l I I I 1 I I I I I I I I 1 I I I I I 1 PPM Figure 11:1-29 ^ HHR Spectrum (CDC13) 0f 5_0-Demethylsimmond 5 in Hexaacetate(IB). trtr? 20 MHz | '■■Jl|mnri|irii'iiiiT|iriiiiiri|nm m i|in iin ii[n iiririi|iri» mi[iim iiin iii)im i|iiiiii«ii|iiiiM m |n n Tim|iMniiir|nTi'iiiri|)iiiiiiii[riin irii|iiii m n 150 100 50 PPM 13, Figure 11:1-30 C NMR Spectrum (CDC13) of 5-0-Demethylsimmondsin Hexaacetate(IB) . noc noc TtANlMllUQN moo Mttj me . o *>-“- no: 6-O-Demethylsimmondsin Hexaacetate(IIIC) . iu e 113 IR Spectrum (CHCl-j)Figure 11:1-31 of » w not mo o m o m iowMoouecno c wc o o no no no c cw ocm ouoouoeuconeoH M oinoew m WV*UE f I WAVt*UUIEI I Of* 9tr? 90MHz 1- L 1 I 1_1 t I I L t i 1 I 1 I i 1 I I I I I I I I I 1 I I I I I I L 1 I ■ 1 ■ I . I I i 1 I I I i I 1 t I I 1 1 1 I I I I 1 1 -L L i 1 l_L 1 1 1 7 6 5 4 3 2 1 PPM Figure 11:1-32 NMR Spectrum (CDCl^) of 6-0-Demethylsimmondsin Hexaacetate(IIIC). I tr£ 9 0 M Hz pm m i ri |i 1111111 iji i rii i ii i[ ii m n n |ii 11 m i > [i r it n n i j 1111111 n| n m r m |i 11 in nrji>m 1111 |n ii rrrr r^ 111111 ithtti i r|ii i i i i i iiju rn ri n | ’JTTfflM'l 1 1 5 0 1 0 0 5 0 Figure 11:1-33 13C NMR Spectrum (CDC13) of 6-0-Demethylsimmondsin Hexaacetate(111C). 8trS m Kuctm imm^mission m < K 5/6-Di-O-demethylsimmondsin Heptaacetate(HID) . u iue 113 IR Spectrum tCHCl3) 11:1-34Figure of WAVaCNCTH 0* CM4 I U W M V A W taoo 6tr2 9 0 MHz _AJL I 1 I I I II I L 1.1 1L I I L_L LL I Ll Ll [ I I I I I I I I I I I I I I 1 I I I I I I I 1 I I I I M I I I I I I I I I 5 4 3 2 PPM Figure 11:1-35 1H NMK Spectrum (CDCl^) of ro LTl 5 ,6-Di-O-demethylsimmondsin Heptaacetate(IIID) . o 20MHz J *>*A* **«■»«■ <*J [>piiiiiir|iiim iiiji Figure 11:1-36 13C NMR Spectrum (CDC13) of 5, 6-Di-0-demethylsimmondsin Heptaacetate (II.ID) . tt*NSMf$SlON i ue 113 IR Spectrum (CHCl^)Figure 11:1-37 Simmondsin of Aqlycone(IVA) M o u n iu fi fi iu n u o M WAYEHLMlli WAYEHLMlli CM~* 252 9 0 MHz A JllljL L 1 L iJ I 1 L LI H I i 1 I L I_1 L I I U [ L 1 i I I 1 L I 1 1 1 1 M i I I 1 I 1 L 1 I L 1 I I II I L 1 II 1 II PPM 253 Figure 11:1-38 H NMR Spectrum (d^-acetone) of Simmondsin Aglycone(IVA) 20 MHz 4 \ pir iiiir|iirriini|Miiiiiii|niniiiMnmini|irrrm iijiriiiriii|iiniiiii[iiniirii|iiniini[riiriiHf|fMMHii|iiiiiiinjitiiiMtijiim iin|iiiniiii|i'rntnrr|HHiim 150 100 50 PPM Figure 11:1-39 13C NMR Spectrum (CDC13) of Simmondsln Aglycone(IVA). n s s t4»4CM iu e 114 IR Spectrum (KBr) 5-0-Demethylsimmondsin 11:1-40Figure of Aglycone(IVB) . 255 9 0 MHz — u * - I I L I LI I L lI L L I L L I I 1 1 I I L L_L I L I I L 1 1 l j LL_t 1 1 I LI I 1 LI 1 I II I I I I I I 1 I 1 1 6 5 4 3 2 1 PPM Figure 11:1-41 1H NMR Spectrum (d^-acetone) of ru 5, 6-Dernethylsimmondsin Aglycone( IVti) . ^ 20MHz ii njiyi rv> ( ...... J|*^ ...... ” | ...... tl,[",,','|')||||ii*iiJiin'n'rr]'W'iiimi|iiiin rri|iiiiiiMi|itm in i|m in in |>iiiri'rm m iiin'r|iiiim n |n m iiir|iiim iin iMin ii>jm im ii 1 5 0 1 0 0 5 0 Figure 11:1-42 ,3C NMR Spectrum (Dg-acetone) of 257 5-0-Demethylsimmondsin Aglycone(IVB). 258 Figure 11:1-43 IR (KBr) Spectrum of Figure 11:1-43 Spectrum IR (KBr) Aglycone(IVC). 6-0-Demethylsimmondsin N O M im M V U 9 0 MHz 1L . A j I I it II I L i n I 1 I I I I I M I I I 1 I I I 1 I It I I I 1 I I I II I I 1 I I I I I I L_L.l1.LJ-II I LLI 6 5 4 3 2 1 PPM Figure 11:1-44 NMR Spectrum (d^-acetone) of 259 6-0-Demethylsimmondsin AglyconeCIVC). 20MHz i n' i nijiM *n^nrn »i i>> r» « i y n ^ P11'1" 1’! T|ii.TTT.ri|..rr i"l II11II111111111 III III 11 |l 111II1111 IT|I 11 . 11 K < I 11II [ II I I M 111 Jl H I U l l f J I I I H H I I j IT'II I TTVI ...... I J 111TTH 111 III 1111 H I III H I 111 1 5 0 1 0 0 5 0 1 1 Figure 11:1-45 C NMR Spectrum (dg-acetone) of 260 6-0-Demethylsimmondsin Aglycone(IVC). 90M H z 2_ A . | l i l l j L VULA. L I L L.L I i 1 l . i I I I I J L I t I I I I I I I I I I I ■ 1 1 1 1 111111 111 1111111111 PPM 6 5 4 261 Figure 11:1-46 IR Spectrum (KBr) of ,5,6-Di demethylsimmondsin Aglycone(IVD) 9 0 M H z i _U 1 I 1 LL.L I l_Li L l ! 1 I 1 L J I L I I 1 I M I I I I I I I I I M I I I 1 I 1 I I I I L I I I I I 1 I I I 1 I I 5 4 3 2 PPM Figure 11:1-47 NMR Spectrum (d^-acetone) of 262 5,6-Di-O-demethylsimmondsin Aglycone(IVD). 2 0 M H z nrrTrmrnr 'i ii|irmmi|iiMtMM|i | ii i i i i t i ? | i i i I n ' i t i | i i n n i i i | i i 11 ii i r t | n ii i r r n 11 ii i M iri|i>iM M ri| i riTMi 150 100 50 PPM Figure 11:1-48 13C NMR Spectrum (D20) of 263 5,6-Di-O-demethylsimmondsin Aglycone (IVD). Figure 11:1-49 IR Spectrum (CHC13) of Methylsimmondsin Aglycone(IVE ) t? 9 S 90MHz I _ A _ J U j l 1 _ N V iA _ _ jL . 1 I I L I L-LI I j J I L I 1 1.1 1 I i J I I L l L L i t I l J _ U _ L n I L i I L t I 1 1 1 I I I I I I I 1 I I I L I t I PPM 6 5 4 3 2 1 265 Figure 11:1-50 NMR Spectrum (d,-acetone) of Methylsimmondsin Aglycone(IVE). 20MHz |iiiiiiiii[iniiMiniiiiim i|Mnniinimi>Mi|ninirrij T™ i[iiintirniiiiiiiiMiim iiii[iiii)iiii|iimnM|»im »ini«iiiim i|m iiiiu|iiiiiiin 1 5 0 100 5 0 P P M 1 ^ 6 ure 11:1-51 C NMR Spectrum (d -acetone) of Methylsimmondsin Aglycone{IVE) . T T T*AN*MH*TOH 3 iu e 115 I Setu (KBr) (+)-Pinitol(VIA) IR Spectrum of Figure 11:1-52 f W A V B M C 1 H* « M W A V K M M H I I H M M K V A W -1 K C rrr w ^ T U U 14 * 267 90MHz l l u L l i i i Lj_i l i 1.L1 m 1iili.L lliIiill1ll.i i Im I I I I I I I I L I I I PPM 5 4 3 2 1 268 Figure 11:1-53 NMR Spectrum (D-0) of (+)-Pinitol(IVA). Chapter 2: Determination of the Absolute Stereochemistry of Simmondsin and Related Cyanoglycosides. INTRODUCTION In 1973 E lliger and coworkers** isolated simmondsin, and postulated the existence of two monomethoxyglycosides from Simmondsia californica (chinesis). Although the structure was correctly determined, the absolute stereochemistry of the compound was never approached. Our investigation includes the determination of the absolute configuration of simmondsin(_V) and its related glycosides. This involved the conversion of simmondsin(V) to dimethyl dasycarponilideCTV) using the steps outlined in Figure 11:2-1. DimethyldasycarponilideCIV) is a derivative of dasycarponinC I_I) , a cyanoglycoside of known absolute stereochemistry^, previously isolated from Thaiictrum dasycarpum^. The physical properties of dimethyldasy- carponilide from both sources were identical, including the optical rotation and CD Cotton effects indicating they are of the same mirror image. Further evidence is obtained from the CD spectrum of simmondsin aglycone benzoate and application of the exciton chirality rule. 269 270 CN NC HO, OGIu AcO, OGIu ta tra Ac OCH OCH NC Glu< OH HO OCH M al/Ag^o 180 DMF NC C H ,0 OCH OCH Figure 11:2-1 Conversion of Simmondsin and Dasycarponin to Dimethyldasycarponi1ide. DISCUSSION AND RESULTS Since an attempted permethylation of simmondsin aglycone(I^) was low yielding and consisted of many side products, it was decided that permethylation and subsequent hydrolysis would be a more successful route to dimethyl dasycarponilide( HO . Treatment of dasycar pon in (I_I) with 17 18 Ag20 and Mel in DMF at room temperature ’ gave permethyldasycarponlnCIII). NC GluO H HO II HO H 1 I OCH OH Ag_0/ Mel NC tetraMe H GluOw 2N HCI 'H IV och 3 OCH 271 272 This material was isolated as a light oil having an optical rotation of +7°, and a UV maxima at 251nm (loge=4.19). The IR spectrum lacked absorptions between 3^00 and 3600cm-1 indicating the absence of hydroxyl groups. However, in addition to the nitrile stretch at 2220cm-1, an intense absorption was seen at 1100cm-1 characteristic of the C-0-C stretch of acyclic ethers. The results of the proton and carbon NMR spectra are given in Table 11:2-1. Worth noting are the presence of six methoxy groups, the uncoupled C-2 proton (5.34,s) indicating that H-4 lies close to the plane of the double bond, and the AB q at 6.24 and 6.01 for H-7 and H-8, respectively. Satisfactory chemical analysis of III could not be obtained since the compound is unstable at room temperature and continually decomposes on silica gel. However, mass spectral analysis (El) showed a molecular ion at m/e 413, and the expected m/«s 414 peak in the chemical ionization mass spect-um. Hydrolysis of III was obtained by heating on a steam bath with 2NHC1 for two and one half hours. Extraction with CHCl^ followed by chromatography over silica gel gave dimethyl dasycarponi 1 ide( I_V) in 50% yield. As expected, its spectral characteristics closely parallel those of dasycarponilide, the modified aglycone from dasycarponin. All protons and carbons were identified through decoupling experiments and comparison of the data with that from known compounds (Table 11:2-2). 273 Table 11:2-1 1H and 13c NMR Resonances for Permethyl Dasycarponin(III). a b Proton multiplicity8 Carbon oc , m u ltip licity H-2 5 . 34,s C-1 153.0,s H-4 4 .8 8 ,d C-2 99.8,d (4.5) C-3 116.7,s H-5 3-95,dt C-5 86 .9 ,d <1.5,3.2) H-6 4 .2 6 ,m C-7 135.2,d H-7 6.24.AB q C-8 126.4,d (10.8) H-8 6 .0 1 ,AB q C-1 ' ■ 104.6,d (10.8) C-6 * 71. 0t H-1 ’ 4 . 54fd OMe 6 0 .8 ,q (7.3) 60 .8 ,q 60 .3 ,q H-2'-6 » 2 .8 -3 .7 59 .4 ,q 5 9 .2 ,q 57 .2 ,q -OMe 3.63ts unass igned 8 4 .0,d 3 . 58,s 79 .3 ,d 3.52,s 77.5,d 3.51,s 76.0,d 3-47,s 7 5 .6,d 3.38,s 75 .2 ,d aDetermined at 90MHz in CDCl^ with TMS. Coupling constants are given in parenthesis in Hz. ^Determined at 20MHz in CDCl^ with TMS. 274 Table 11:2-2 and * NMR Resonances for Dimethyl Dasycarponilide(IV). Proton 5h, multiplicity® Carbon 8 , multiplicity** c H-2 5 .93»d C-1 173.2,s (1.9) C-2 113.7,d H-4 5 .2 3 ,dd (1.9,10.5) C-3 162.8,s H-5 3 .4 3 ,dd C-4 81 .7 ,d (3 .8 ,1 0 .5 ) C-5 8 4 .0,d H-6 4 .2 9 ,dd (3 .8 ,5 .4 ) C-6 7 5 .3 ,d H-7 6 .4 9 ,dd C-7 137. 1 ,d (5 .4 ,9 .5 ) C-8 123.6,d H-8 6 .7 5 ,d (9.5) 5-OMe 5 9 .6 ,q OMe 3-52,s , 6 H 6-0Me 5 8 .4 ,q aDetermined at 90MHz in dg-acetone with TMS. Coupling constants given in parenthesis. ^Determined at 20MHz in dg-acetone with TMS. 275 The isomerization of 3immondsin(\0 was expected to proceed to VI without great difficulty since the product would be of greater conformational stability than the starting m aterial. However, when a deoxygenated sample of V was irradiated with a high pressure mercury lamp using m- methoxyacetophenone as sensitizer, an equilibrium mixture was obtained after three hours. This conversion could not be driven further even with th ir ty -fiv e hours of irradiation. In addition, the desired product C isosimmondsint VI_)) could not be isolated since it was unstable in the presence of hydroxylated solvents, temperatures above 25°C» and AgNO^ adsorbed silica gel. This decomposition is most likely attributed to an interaction between the n itr ile and syn hydroxy group. NC OH GluO H 0M« HO OGIu M* OM* 0M» V VI When simmondsin pentaacetate(VII) was irradiated in the presence of m-methoxyacetophenone, an equilibrium mixture (48:52; VII:VIII) of stable products was obtained (Figure 11:2-2), allowing isolation of isosimmondsin 276 NC CN HO OGIu HO OGIu UNSTABLE M eO MeO OMe OMe (37:63) ,CN NC AcO AcO O G Iu- tetraAc (48*52) tetraAc h u MeO MeO s e n s i t i z e r MeO M eO VII VIII Figure 11:2-2 Isomerizations of Simmondsin(V) and Simmondsin Pentaacetate(V II). 277 pentaacetate(VIII) as a colorless hard glass. Although most of the data collected for VIII wa 3 sim ilar to that of simmondsin pentaacetate (Table 11:2-3 to 11:2-5), the CD and NMR spectra have very noticeable differences. As a result of the conformational ring flip, H-2 is now allylically coupled with the axial C -8 proton rather than H-4 which must lie in the plane of the double bond. Also, the patterns for the C-7 protons resulting from vicinal and geminal coupling have changed from two doublet of tr ip le ts (ddd's) to a multiplet and a quartet-like doublet of double doublets. The latter results from three large coupling constants (a ll approximately 12.1Hz) indicating that the C-7 proton must be trans diaxial to both H -6 and H-8. The third large coupling is to the geminal H-7. The change in the magnitude of the coupling constants as observed between the glycoside and aglycone requires a ring flip from one chai1' form to the other and strongly confirms the axial orientation of the glucose moiety resulting from A^1*^ strain 1 ^' 1 ^ As mentioned, the isomerization reaction only resulted in a 48:52 ratio between simmondsin pentaacetate(VII) and its isomer. However, many reactions require long (20-40 hrs.) irradiation times to attain completion. Thus, it was desirable to demonstrate reverse isomerization to confirm that the product ratio was the result of a true equilibrium and not short irradiation times. Indeed, irradiation of 278 Table 1:2-3 A comparison of the Physical Properties of Simmondsin Fentaacetate(VII) and Isosimmondsin Pentaacetate(VIII). Isosimmondsin Simmondsin Pentaacetate Pentaacetate mp non-crystalline 167-7.5° softens above 63 ( -46° -41° uv 211nm 21 7nm (MeOH) (log c = 4 . 10) (log £ =4.07) IR 3020, 2222, 3020,2225, (CHC13) 1760,1645, 1760,1642, 1370,1215, 1375,1230, 1020- 1100, 1140-1050, 893 912 CD [0]231+655O (MeOH) [0 ]225 0 [0] 222 -44200 [0]2O6-5°300 ms 585(M+) (El) 347,331, 347,331, 271,238, 271,238, 206 ,178, 206,178, 169( 100%), 169(100%), 163,157, 146,109, 146,109, 88 88 (Cl) 586(37%) 279 Table 11:2-4 1H NMR (300MHz, CDCl.. ) Data for Simmondsin Pentaaoetate(VII) and Isosimmondsin Pentaacetate(VIII). Proton VIII VII H-2 5.74,d 5.4 2 ,d J(2,8)=1.9 J(2 , 4) = 2. 0 H-4 5.95,d 6 . 03,ddd J (4 , 5) = 3. 7 J (4 ,5)=8.9 H-5 3 .68 ,m 3 - 19,dd J (5 ,6 ) = 2. 8 H-6 3 .5 4 ,ddd 3 .8 1 ,brm J(6 , 7ax) = 1 2. 1 J(6,7eq)=4.5 H-7ax 1. 55,ddd 1.62,ddd J(6 , 7ax) = 12. 1 J(6 , 7ax)= 3 * 2 J(7ax, 7eq)= 12.1 J(7ax,7eq)=15.0 J(8,7ax)=12 .1 J(8, 7ax)= 3-2 H-7eq 2. 20,m 2.4 4 ,ddd H-8 4.3 7 ,ddd 4 .7 9 ,dd J(7ax,8)=12.1 J(7eq,8)=4.5 H-1 ' 4 .61,d 4 .7 0 ,d J(1',2')=7.8 J (1 ',2 * )«7. 8 H-2’-4• 4.9 -5.2 5.0-5.2 non- 1st order non- 1st order H-5* 3 .6 4 ,ddd 3 .6 0 ,ddd H-6 'A 4. 17,dd 4 .2 4 ,dd J (51 , 6 'A)=4 . 4 J(5 ',6*A)=2.4 H-6 'B 4.01,brd 4 .0 3 ,dd J(6 *A , 6 ’B) = 12.2 J(6 fAf 6'B)s12.2 J (5 ' ,6 B) s4 . 2 5-OMe 3-32,s,3H 3.42,s,3H 6-0 He 3 .4 5 ,s ,3H 3* 35,s , 3H OAc 2 .0 1 , s , 3H 2 .1 4 ,s , 3H 1. 97, s , 3H 2. 08,s , 3H 1. 9 3 ,s , 6H 2. 04,s, 3H 1. 91,s,3H 2. 02,s , 3H 2. 00, s , 3H 280 Table 11:2-5 NMR (20MHz, CDC1_) Data for Simmondsin Pentaacetate(VII) and Isosimmondsin Pentaacetate (VIII). Carbon Isosimmondsin Simmondsin Pentaacetate Pentaacetate C-1 155.8,s 159.3,s C-2 100. 2,d 9 5 .7 ,d C-3 115.3,3 115.M,s C-4 69. 2,d 7 0 .8 ,d C-5 76 . d 8 2 .5 ,d C-6 75. 1 ,d 71*. 3,d C-7 32. 0(t 30. 6 , t C-8 72. il.d 7 5 .7 ,d C-1 ' 9 8 .M,d 100. 7,d C-2' 7 2 .4,d 72. 8,d C-3 ' 68 .4 ,d 68 . 2,d C-4 ' 71. 0,d 7 0 .8,d C-5' 72. 0,d 7 2 .0,d C-6 ' 6 1 .7,d 6 1 .4 ,t 5-OMe 59. l,q 5 8 .2 ,q 6 -OMe 5 6 .7 ,q 56 . 6 , q ii o o 170. 4 ,s 170. 2,s 170. 0,s 169.9,s 169.3,s 169. 1, s 169. 1,s 168.7,s 168 . 6 , s 168.3,s ch3co 20. 8, q 20. 3, q 20.5 , q 281 isosimmondsin pentaacetate(VIII) in a similar manner produced a 46:54 mixture of simmondsin pentaacetate(VII) with its isomer (Figure 11:2-2). This equilibration was also found to be independent of solvent polarity since sim ilar ratios were obtained in hexane/benzene( 1: 1), MeOH/benzene (1:1 and 1:4), and MeOH. I n itia l attempts to isomerize simmondsin(V) fa iled . However, VII and VIII were isolated in a 37:63 ratio when siramondsin(V) was irradiated for three hours in the presence of m-methoxyacetophenone, the solvent evaporated at 25° in vacuo, and the remaining residue immediately acetylated with AC2O and pyridine. This supplied indirect evidence that simmondsin does isomerize, but to an unstable product. The hydrolysis of isosimmondsin pentaacetate(VIII) with 2N HC1 went smoothly to produce isosimmondsin aglycone( IX) in 21% yield. Its physical constants are contrasted with those of 3immondsin aglycone in Tables 11:2-6 and 11:2-7. Formation of the y-lactone requires a second ring flip to bring the hydroxyl into an equatorial position. This is verified by the patterns of the C-7 protons in the NMR spectrum being present as doublets of tr ip le ts . Also, H-2 resumes its allylic coupling with H-4 as was the case in simmondsin pentaacetate(VII) . The dehydration of isosimmondsin aglycone(JHC) was more difficult than anticipated. It was expected that the axial leaving group would eliminate ea sily via a 282 Table 11:2-6 A Comparison of the Physical Properties of Simmondsin Aglycone(I_) and Isosimmondsin Aglycone( IX). Isosimmondsin Simmondsin Aglycone Aglycone mp 129- 30° 138.5-9° 0 t- on [a]D -31° 1 UV 214nm 208nm (MeOH) (log c =3. 96) (log € =4.12) IR 3740,3010, 3610,3418, ( CHC1t) 1791,1760, 3015,1785, J 1665,1150- 1758 , 1662 , 1050,915, 1125-1130, 871,690, 867,691 CD [0]245-141OO ----- (MeOH) [0 ]235-125OO(min) ^^240-7600(3hl) [^ 3 213-32300 [0]2O7-559OO 13C NMR (d^-acetone) (d 6 -acetone) C-1 173.0,s 173.0,s C-2 115.1,d 116.9,d c-3 169.6,3 170.4 ,s C-4 8 2 .4 ,d 66 . 3,d C-5 88. 2,d 80 .9 ,d C-6 7 9 .0,d 76 . 2,d C-7 3 * . 8,d 3 4 .0 ,t C-8 6 5 .8 ,d 7 8 .4 ,d 5-0me 5 8 .3 ,q 57.0,q 6 -OMe 59 -9 ,q 5 9 .2 ,q ms 214(M+) 214(M+) (El) 182(100%) 182,153, 150,122, 141,112, 109, 88, 109,88, 71,45 71(100%) 283 Table 11:2-7 1H NMR (300MHz,d6 -acetone) Data for Simmondsin Aglycone(I^) and Isosimmondsin Aglycone( IX). Proton Isosimmondsin Aglycone Simmondsin Aglycone H-2 5.97,d 5 . 98,dd J(2,4)= 1. 8 J(2, 7eq)=0.6 J (2 ,8 )= 1 .9 H-4 5. 32,ddd 4 .89,brd J (4 ,6 ) = 0. 8 J ( 4 ,5)= 3.5 J(2,4)= 1. 8 JCU,5) = 9. 6 H-5 3. 22,d 3.81 ,m J (5 ,6 ) = 2. 7 J ( 5 ,6 )=2 .2 J (4 ,5)=9•3 JC4,5)=3.5 H-6 4 . 10,ddd 3 .8 7 ,ddd J( 4 ,6 ) = 0 . 8 J ( 5 ,6 )=2 .2 J(6 , 8) = 0 . 8 J(6 , 7eq)=3-8 J(6 , 7eq)=2.4 J(6 , 7ax) = 1 1. 4 J (5 ,6 ) = 2. 7 J(6 , 7ax)=3 .8 H-7ax 1.69,ddd 1. 57 ,ddd J(7ax, 8)=2. 4 J(6 , 7ax ) = 1 1. 4 J(6 ,7ax)=3 .8 J(7ax, 8)=11.4 J(7ax,7eq)=15.5 J(7ax, 7eq) = 10 H-7 eq 2.4 9 ,ddd 2. 53,ddd J(6 , 7eq)=2.4 J(6,7eq)=3.8 J C7eq,8) = 3*4 J(7eq,8)=6.4 J(7ax,7eq)=15.5 J(7ax, 7eq>= 1 0 H-8 4 .76,m 5 . 11,ddd J(6 , 8)=0. 8 J(2 , 8)=1 .9 J C7ax,8) = 2 . 4 J(7eq,8)=6.4 J C7eq,8) = 3*4 J(7ax,8)=11.4 J(8, OH)=9. 0 OH 4 .14,d 5 . 06 ,brs J(8,0H)=9*0 5-OMe 3.55,s 3-43,2 6 -OMe 3•49,s 3 .38, s 284 trans-jB-eliroination and extend conjugation serving as a driving source for the reaction. The elimination was finally achieved by preparing the mesylate(X) from mesyl chloride and triethylamine2\ and heating the latter at 150 ° in the presence of DMSO for eighty minutes. This gave dimethyl dasycarponilide(rv) in 25.4% yield which was found to be identical in all respects (mp, co-tlc, IR, UV, CD, NMR, 13C NMR, and ms) with that obtained from dasycarponin(II). / H O: H c h 3 | OM? CH3 0 C H o c h 3 OCH X Since the measurement of the CD gave identical results (sign and magnitude) for dimethyl dasycarponilide from both sources, carbons 4, 5 and 6 of simmondsin(V^) must have the same absolute stereochemical arrangement as in rv. The relative stereochemistry of H -8 is known from 1H NMR, NOE and decoupling studies of simmondsin(V) and its derivatives giving rise to the above final structure. Since the three new glycosides (th is volume, Chapter 1 of Part II) and simmondsin were converted to a sin gle permethylated product all having the same sign of optical rotation, then all four glycosides must be of the same absolute stereochemistry. The absolute stereochemical assignment is further substantiated from the CD measurements of simmondson aglycone benzoate(JCI) . This compound was prepared by reaction of 1^ with pyridine and benzoyl chloride at room temperature (Table 11:2-8). An examination of the group orientation along the C^c^ bond shows that the allylic benzoate is left-handed and must give rise to a negative Cotton effect at 230nm as observed. The CD (Figure 11:2-3) exhibited a negative Cotton effect at 230nm ([0] -57600), a second Cotton effect at 205nm ([0] -44000), and a third at 195nm ([0] -106000). This is in agreement with those 286 Table 11:2-8 1H NMR Data (90MHz, CDC1 ) for Simmondsin Aglycone Benzoate(XI) Proton multiplicity J (Hz) H-2 6 . 20,d J(2 , 8)= 1. 6 H-4 6 .09id J ( 4 ,5)-3. 5 H-5 3 .9 6 ,ddd J(5,7eq)=1. 8 J(5,6)=2.5 «J (4 , 5) = 3 • 5 H-6 3 -78dq J(5,6)=2.5 J(6 ,7eq)=3. 8 J(6 , 7ax)= 12. 1 H-7eq 2. 68 ,brdq J( 5 ,7eq) = 1. 8 J(6 , 7eq) = 3 ■ 8 J(8. 7eq)=6 .4 J(7ax,7eq) = 11. 8 H-7ax 1.38,ddd J(6 , 7ax) = 12. 1 J(7ax, 7eq) = 1 1. 8 J(8 , 7ax ) = 11. 4 H-8 5 . 02,ddd J(2 , 8)=1 .6 J(8, 7eq)=6 .4 J(7ax,8)=11. 4 -OMe 3.57,s 3 .4 6 ,s aromatics 7 .3 -8 .1 ,m (5H) .+106000 +100 + 8 0 - + 6 0 - OCCzH + 4 0 - +20 'o 2 5 0 ^00 2 2 5 2 7 5 - 2 0 - CD - 4 0 - -4 4 0 0 0 -5 7600 Figure 11:2-3 UV and CD Spectra for Simmondsin Aglycone Benzoate {XI). 288 predicted from the application of exciton coupling to non degenerate systems composed of two d ifferen t chromophores^^. Allylic alcoholic benzoate systems are known to exhibit an a allowed n-+n transition at 230nm from the benzoate, and a a second allowed transition at 195nm from the C=C. According to the exciton chirality rule, the second Cotton effect at shorter wavelength should be opposite in sign to the f i r s t . The CD Cotton e ffe c ts of simmondsin aglycone(_I) and isosimmondsin aglycone(JUC) are also in agreement with the above results. The observed Cotton effects in these compounds are the result of two effects: ( 1)the "lactone rule" which is useful in determining the configuration at the y-carbon of an a,j 3-unsaturated lactone , and ( 2 ) the " a lly lic oxygen" e f f e c t ^ which w ill dominate over the former. Both rules predict a negative cotton effect for the £ n+n transition in _I and JX, The CD spectra (Figure 11:2-4) show this to be the case although significant differences are seen in the magnitude of the transitions. The reason for this must result from 1, 3-diaxial interactions since the relative stereochemistry at carbons 3> 4 and 8 are identical in terms of the y-carbone and allylic hydroxyl. Isosimmondsin aglycone(DC) possesses a 1,3-diaxial interaction between bulky groups (hydroxy and methoxy) which is absent in simmondsin aglycone( 1^). This w ill result in a conformational twist, changing the dihedral angle between 289 + 10- isosimmondsin aglycone 200 25 0 [ e ] -7600 *■ 2 4 0 - 10- M -14100 245 - 2 0 - -30- -3 2 3 0 0 213 - 4 0 - simmondsin aglycone -50— - 6 0 - Figure 11:2-4 CD Spectra of Simmondsin Aglycone (I_) and Isosimmondsin Aglycone (IX.) . 290 the hydroxyl and the double bond on which the magnitude of the transition is dependent. HO, .OH MeO I I I I OMe OMe I IX ISO OMe H .OMe H OH OH Me OMe X C = 0 C H X>Y HO (-) In summary, the absolute configuration of simmondsin and related glycosides from Simmondsia californica has been determined by conversion of simmondsin to the dimethyl derivative of dasycarponilide. The absolute stereochemistry 291 of the latter is known from the application of the non empirical '’dibenzoate chirality rule." EXPERIMENTAL* Methylation of Pa3ycarponin(II). To 10ml of DMF, 5l0mg of dasycarponin(ri) was added followed by 0.36ml of Mel and 631mg of Ag£0 and stirred in a room temperature bath for eighty-three hours. During this time, similar additions of Mel and Ag£0 were made approximately every twelve hours until the reaction was complete by tic. Afterwards, 20ml of CHCl^ was added, stirred for ten minutes, and filtered through a sintered glass funnel. The yellow precipitate was washed with CHCl^ (5x10ml) and evaporated to give 684mg of yellow oil. This was chromatographed over 60gms of silica gel 60 (230-400 mesh) using 1.5% MeOH in CHCl^ as solvent. Fractions totaling 418mg (65%) were obtained and identified as permethyldasycarponin(I I I ): ligh t o il; t ic , R^=0.22, 75% EtOAc in hexane; visualization, purple with p-anisaldehyde; [a ]D+7°(c= 0.6025, MeOH); IR (CHC13>, wMAX 3010 (C=C-H), 2220 (C=N), 1100 (C-0-C) and 850cm“ 1 (CsC-H); UV (MeOH), XMAX 251nm (log < =4.19); 1H NMR (90MHz, CDClj), refer to Table 11:2-1; 13C NMR (20MHZ, CDCl^), refer to Table 11:2-1: mass £ See Appendix on p .364 292 293 spectrum (E l), m/e 413(2.3%, M+) , 238 (23%), 187 (45%), 178 (22%), 111 (30%), 101 (84%), 88 (100%), and 75 (53%) (Cl), m/e 414 (34%). The low yield of this product was found to be related to the compound’s instability at room temperature and on silica gel. Prep-layer chromatography was found to be the best cleanup method since the compound continues to decompose on silica columns. As a resultof its instability, chemical analysis (%C, 5.45; %H, 7.06; %N, 3-02: calc'd for C20h31no8: *c* 58*1! 7.56; %N, 3.39) did not give satisfactory results. Hydrolysis to Dimethyl Dasycarponilide(IV). Permethyl dasycarponin ( I I I : 265mg, 0.641meq) was added to 25ml of 2NHC1 in a 100ml RB flask fitte d with a H£0 condenser and heated on a steam bath for two and one half hours. The orange-red solution was rapidly stirred and solid NaHCO^ added until pH 7.0 was obtained. Extraction with CHCl^ (4x20ml), and washing the combined organic layers with H2O (15ml), followed by evaporation of the CHCl^ gave 161mg of yellow residue. Chromatography over 32gms of s i l ic a gel 60 (230-400mesh) using 1% MeOH in CHCl^ gave four fractions. Fractions 2 and 3 contained 17mg of permethyl dasycarponin, and fraction 4 ( 66 mg) was id en tified as 2, 3, 4 ,6 -tetramethyl glucose by mp, co-tlc, [«]q, and H IJMR. 294 Fraction 1 contained 59mg (50.2% yield) of dimethyl dasycarponilide(IV) which crystallized as prisms from EtOAc/hexane: mp 82-82.5°; t ic , Rf=0.l6, 0.5% MeOH/CHCl3 ; visualization, yellow with p-anisaldehyde; [c]^+ 379(e=0. 05 , MeOH); UV (MeOH), Amax 252nra (log £ =4.25); IR (CHC13), v MAX 3010 (C=C-H), 1797 and 1757 (C=0), 1650 and 1595 (d ien e), 1155-1055 (C-0-C), 903, 860 and 857cm“1 (C=C-H); CD (c=2.55x10“3M, MeOH), [0 ]3O5 +550, [0 3275 +45600, 3^63 +70600, [0 ]24o 0, Ce]230 -12200, [0 ]2l8 °» ^ ] 205 +12500, and 0; NMR (90MHz, d^-acetone) , refer to Table 11:2-2; ^3C NMR (20MHz, d^-acetone), refer to Table 11:2-2; mass spectrum (E l), m/js 196.0740 (30%, M + , ^ 0^ 12^4 requires 196.0736), 164 (39%, M-MeOH), 151 (39%), 137 (100%), 108 (30%), 95 (52%) and 78 (30%). Isomerization of Sitnmondsin Pentaacetate(VII) . Simmondsin pentaacetate (VII: 60Gmg, 1.023meq) and 645mg (4.3meq) of m-methoxyacetophenone were dissolved in 200ml of spectrograde Me0H:benzene (1:4) in a 250ml standard taper pyrex erlenmeyer. The solution was cooled in an ice bath and purged with dry N2 for one and one half hours while stirring. The vessel was fitted with a cold finger and irradiated for three and one half hours with a high pressure mercury lamp. Evaporation of solvent gave 1 .32gms of pale yellow oil which was chromatographed over 60gms of silica 295 gel 60 PF-254 using CHCl^ and 1% HeOH in CHC13 as solvent. The sensitizer eluted in the first 100ml. Fraction 6 (263 mg) was found to contain starting material and crystallized from EtOAc:hexane to give 24lmg of cry sta llin e simmondsin pentaacetate(VII). Fractions 3 and 4 (299mg) were further purified to give 285mg of isosimmondsin pentaacetate(VIII) as a colorless hard glass: softens to a viscous oil above 63°; tic, Rf=o.36, 2% MeOH in CHCl^; [a ]D.46°(c = 0 . 658, MeOH); IR (CHC^), wMAX 3020 (C=C-H and CH3-C0), 2937 and 2835 (C-H), 2222 (C=N), 1760 (CsO), 1645 (C=C), 1370 (CH3C00), 1215 CC-0), 1100-1020 (C-O-C), and 893 (C=C-H); UV (MeOH), XMAX 211nm (log e =4.10); CD (c=3.42x10"3M, MeOH), C0]25O 0, [032uo +1170, C03231 +6550, [ *]225 °» [ * ]215 -31600, [0 ]2o6 “50300, [0] 195 -24900; 1H NMR (300MHz, CDC13), refer to Table 11:2-4; 13C NMR (20MHz, CDC13), refer to Table 11:2-5; mass spectrum (El), m/e 585 (0.1%, H+) 1 331 (60%, Cu h 130 9 + >, 238 (47%, M-C 1 4h ^ 30 , Q) , 169 (100%), 109 (91%), and 88 (43%); (Cl), m/e 586 (1.8%, M+H). Analysis. Found: %C, 52.89; %H, 6.08; %N , 2.34. Calc'd for C26 H35 N014: %C, 53.33; %H, 6.02, and %N, 2.40. Isomerization of Isosimmondsin Fentaacetate(VIII). Isosimmondsin pentaacetate (V III: 70mg, 0.119meq) and m- methoxyacetophenone (I80mg, 1. 20meq) were dissolved in 13ml 296 of spectrograde MeOH in a pyrex irradiation vessel and purged with dry N2 for two hours while being rapidly stirred. The solution was irradiated for three hours using a high pressure Hg vapor lamp with continuous water-cooling, rapid stirring and slow N2 purge. When complete (tlcj 3% MeOH in CHCl^), the methanol was evaporated in vacuo at 35°» and the oily residue applied to a column containing 5 gms of s ilic a gel (70-230mesh; E.M.) packed in 25% CHCl^ in benzene. Elution with the column solvent and 0.5% MeOH in CHCl^ gave four fractions. Fraction 1 contained the sensitizer and fraction 4 consisted of minor side products (6.4mg) which were not further investigated. Fraction 2 (35 . 7mg) was further purified over 5 gms of silica gel to give 22mg of pure isosimmondsin pentaacetate(VIII) identical with the starting material (mp, tic, and 1H HMR). Fraction 3 (30.0mg) cry sta llized from EtOAc/hexane to give 17.5mg of long needles. This was identical with simmondsin pentaacetate(VII) by mp, c o -tlc , IR and "*H NMR. The two major products were obtained in a 9**% yield in a 5^:46 (VIII:VII) ratio. Isomerization and Acetylation of Simmondsin(V). Simmondsin (V: 368mg, 0.98meq) was added to a pyrex irradiation vessel which contained 17ml of spectrograde MeOH and 450mg of m-methoxyacetophenone ( 3meq), and purged with 297 dry N2 for three hours. The solution was then irrad iated with a high pressure Hg lamp for three hours with continuous water-cooling, rapid stirring and N2 purging. On completion, the solvent was evaporated at 20- 25 ° in vacuo. The pale yellow, oily residue was dissolved in 4ml of pyridine, 2ml of Ac20 added and the solution stirred for twelve hours. Evaporation to dryness gave a yellow oil which was chromatographed over 45gms of s ilic a gel 60 PF-254 using CHCl^ as solvent. Elution with 0.5% MeOH in CHCl^ gave 333mg of isosimmondsin pentaacetate (VIII: identical by c o -tlc , IR and 1H NMR), and 198.5mg of simmondsin pentaacetate( VII) which cry sta llized from EtOAc/hexane as long needles (identical by mp, co-tlc, IR and 1H NMR). These two major products were obtained in 92.6% yield in a 63:37 (VIII:VII) ratio. Ac id Hydrolysis of Isosimmondsin Pentaacetate(VIII) . Isosimmondsin pentaacetate (VIII: 1. 5 gms, 2. 56 meq) was heated on a steam bath for two hours with 150ml of 2NHC1 in a 250ml round bottom flask fitted with a condensor. The solution was allowed to cool to room temperature, and the pH raised to 4 with the addition of solid NaHCO^. Extraction with CHC1^ (5x20ml), washing of the organic layer with distilled H2o ( 35 ml), and evaporation gave 164mg of crystalline residue. Recrystallization from EtOH:hexane 298 ( 1: 1) gave prisms which were id en tified as isosimmondsin aglycone (IX: yield 21%): mp 129-30°; tic, Rf=0.23, silica gel, 2.5% MeOH in CHCl^; visualization, orange with p- anisaldehyde; [“ Jp —31°tc=0-183* MeOH); IR (CHCl^), 3470 (0-H), 3010 (C=C-H), 1791 and 1760 (C=0), 1665 (C=C), 1150-1050 (C-0-C), and 690cm ' 1 (C=C-H); UV (MeOH), AMAX 2l4nm (log£=3*96); NMR (300MHz, dg_acetone), refer to Table 11:2-7; NMR (20MHz, d^-acetone), refer to T a b lell: 2-6; CD (c= 3 • 35x 1 0** 3M, MeOH), Ctf]2g0 0 ; l0l2H 5 -14100, 191235 -12600, [tf]213 -32300, [^]2oo 0; ma3S spectrum (E l), m/e 214 (1%, M+), 182.0584 (100%, M-MeOH, C9h 10°4 requires 182. 0579), 154 (14%), 150 (30%, M-2MeOH), 109 (59%), 88 (84%), 71 ( 86 %), and 45 (94%). Dehydration of Isosimmondsin Aglycone(IX). Isosimmondsin aglycone (.IX.: 28mg, 0.131meq) was dissolved in 1ml of dry CH 2C12 at 0° in a 5ml RB fitte d with a drying tube and septum. To this 361 (0.26meq) of triethylamine was added followed by the slow addition (15 min) of 17.51 (0.22meq) of mesyl chloride (X). After stirring at 0° for 45 minutes, a tic (2% MeOH in CHCl^: the mesylate is yellow and aglycone orange with p-anisaldehyde) indicated completion: total time 65 minutes. The pale yellow solution was added to ice (5ml) in a 30ml separatory funnel using 10ml of cold CH2C12. The aqueous layer was 299 extracted with two additional 5ml lota of CH 2C12 , and the combined CH2C12 successively washed with 1% HC1 (cold) , saturated NaHCO^, and H2O (5ml each). On evaporation of the CHC1^, 36-5mg of c r y sta llin e mesylate was obtained: [mp 73°» 1H NMR (CDC13), Ah 3.03 Heating of th is material in 1.5ml of d is tille d DMSO to 150 ° in an o il bath for 80 minutes produced a dark reddish gold solution. This was transferred to a seperatory funnel using 10ml of CHCl^ and 20ml of 5% NaHC03. The CHCI 3 layer was further washed with 20ml of 2% HC1 and 15ml portions of H20 (7 tim es). The aqueous layers were combined and extracted with CHCl^ (3x20ml), and the extraction process repeated on the combined CHCl^ (60ml). Evaporation of the to ta l CHC1^ gave 28mg of an orange red o i l. Prep tic purification (20x20cm; 0.5mm) using 1% MeOH in CHCl^ gave 6.5mg of material (25.4% yield from aglycone). Final purification on a second preparative layer ( 20x 20cm; 0. 5mm; 84% EtOAc/hexane) after combination with an additional 4.5mg of material gave 9. 0mg of crystalline dimethyl dasycarpon- ilide(^V) that crystallized from 25% EtOAc in hexane as prisms: mp 82-2.5°; tic, Rf=0.43, 2% MeOH in CHCl^ and Rf=0.28, 76 % EtOAc in hexane; visualization, yellow with p- anisaldehyde; [a]D +386(c=0.033, MeOH); IR, UV, CD, 1H NMR and MS were id en tical with dimethyl da 3ycarponilide obtained from dasycarponin( 11). 300 Benzoylatlon of Simmondsin Aglycone(I ). Simmondsin aglycone (_I: 66mg, O.308meq) was dissolved in 2.5ml of Pyr, 0.04ml of benzoyl chloride added, and stirred at room temperature for twenty hours. The dark orange solution was evaporated to dryness and partitioned between 10ml each of NaHCO^ and CHCl^. The aqueous layer was extracted three times more with 5ml portions of CHCl^, and the combined CHCl^ washed once with 10ml of H20. Evaporation gave 233mg of orange oil which was chromatographed over 31gms of silica gel 60 (230-400 mesh; 6% H20) usin8 0.6% MeOH in CHCl^ as solvent. After 160ml eluted, 95mg (96.8% yield) of crystalline material was obtained on evaporation of the solvent. This was recrystallized from 30% hexane in EtOH to give 72mg of long needles id en tified as simmondsin aglycone benzoate(X_I): mp 148.5-149° tic, Rf=0.42; 1.6% MeOH/CHCl^; visualization, pale yellow with p-anisaldehyde ; [q]q -1 03°Cc = 0.17, MeOH); IR (CHC13), vmax 3020 (C =C-H), 1790 and 1770 (C=0), 1735 ( C =0, benzoate), 1668 (C=C), 1606, 1590, 1496, 1456 (arom CsC), 1256 (C-0), 1150-1025 (C-0, ether), 710cm-1 (C-H, arom); UV (MeOH), AMAX 228nm (log£=4.33; K band), 274nm (log£=3.10; B band), and 282nm (log£=3.01; shl); CD (c = 3.52x10-3 M, MeOH), C^]26Q -11OO,[0]25O -4800, t^3230 -57600, [«]215 -12800, C0]2O5 -44000, [«]201 °» ^ 1 9 5 +106000; 1H NMR (90MHz, CDCl^jj refer to Table 11:2-8; mass spectrum (El), m/e 318.1111 (6%, M+, c17Hlg06), requires 318.1103, 245 (15%), 196 (6%, M -C^COOH), 139(9%), 105(100%, C6H5CO+>, and 77 (18%, C6 h5+). MC INC koo KC m MM MM MM UK MO MM IW MM M MM HH UM OM OK HOC HIM »M W Ht to. 1 I VMVfNUUMI CM-1 I -__ _ Figure 11:2-5 IR Spectrum (CHCl^) of Permethyl Dasycarponin. 302 9 0 M H z L _ <1 — i I I I I I I I L I I 1 I I I I 1 I I 1 I I I I I I 1 I I I I I I I I I 1 L I I I I I I I I I PPM 6 5 4 3 2 303 Figure 11:2-6 H NMR Spectrum (CDCl^) of Permethyl Dasycarponin. 20 MHz nT|Tnimn|iim iiii|m iiiinjiin in ii[i' jim i m i|i ii ntin |ii ii m ii|iin ruii| mm in |i 11 mu i|iim im iirn im i| ii inTTiiiim i PPM 150 100 50 ut 13 o Figure 11:2-7 C NMR Spectrum {CDCl^) of Permethyl Dasycarponin . t W A V M S O T H > * 3R- m 31 n X » m I WAVfMUMIII CAT* I Figure 11:2-8 IR Spectrum (CHCl^) of Dimethyl Dasycarponilide. 305 90 MHz WJL _ * J j U L 1 LLLi Ll I - u L l L L I_LlLLI I LI LlI I I I i l l II I I I I I I I 1 I I I I I I I I I I L I I I I I I I [ I I I L PPM o Figure 11:2-9 H NMR Spectrum (dg-acetone) of Dimethyl Dasycarponilide, o\ ii i|i in |iiniiiii|ii(^iiT|TTnnn(p(iTnnTT| PPM 150 100 50 0 307 13 Figure 11:2-10 C NMR Spectrum (dg-acetone) of Dimethyl Dasycarponilide. 308 +70 0 ] m J 7°6OO' +60 +50- +40 +30- +20 { e ] +12500 •o 200 250 300 nm -1 0 ['$] -12200 2 3 0 Figure 11:2-11 CD Spectrum (MeOH) of Dimethyl Dasycarponilide. fftANtMtU'OM -10 Mi MOO i ue 121 IR Spectrum (CHCl^)Figure 11:2-12 Isosimmondsin of Pentaacetate. 11 3700 O M MOO» UOO * T A C H l M U N f V A W MOO 309 9 0 M Hz J U J U u J U J ! J lM t Jl m aA IIh K r * s J —i—r—|— i—i— i—i—i—I—i—i—i— i—i—i—i—i—i—i—r—r "( > I I ' I I I I • | I I I—1—I—I—I—I—I- f—I—I—| I | | I I—|—|—| ( T ' 6 5 4 3 2 1 P P M Figure 11:2-13 NMR Spectrum (CDC13) of Isosimmondsin Pentaacetate. 310 20 MHz ***>wyvL^' piiniiir|rmiim |iiiiiimMn m iiniiiMim[irHiiiii|iiitmii|nTT>TFii[iiiiiirii| niTjm m nr|m (iMM|iimmr mrrmTimim m im um iiiim im iim m ? 1 5 0 1 0 0 S [> P P M 311 Figure 11:2-14 13C NMR Spectrum (CDCl^) of Isosimmondsin Pentaacetate. i 3 1 2 + 2 0 + 10 [ 0] +6550 1---- 1 231 O 200 2 5 0 -1 0 - 20- -3 t - -4 0 -5 0 3 0 0 206 Figure 11:2-15 CD Spectrum (MeOH) of Isosimmondsin Pentaacetate. ■MOO hoq a g o u t m o » no mo uoo • WtVEMMIII Of1 Figure 11:2-16 IR Spectrum (CHClj) of Isosimmondsin Aglycone LO 9 0 MHz AA. 6 5PPM 4 3 2 1 Figure 11:2-17 NMR Spectrum (dg-acetone) of Isosimmondsin Aglycone. trl£ 20 MHz rHt»M>Hni»iiiiiii[iiiniiii|iinMiii|nnmrqiiiiinrT|miirnn^iiniinn l|linTTlTq n u ilMHIIHT>IIHIIIIIIIlT]IIIIIIIIIHHIH III|lllim iH I PPM 150 100 50 315 13 Figure 11:2-18 C NMR Spectrum (dg-acetone) of Isosimmondsin Aglycone, MtcRVT iue 121 I Setu (Hl) f imnsn gyoe Benzoate. Simmondsin Aglycone (CHCl^) of Spectrum IR 11:2-19 Figure U) Ov 9 0 M Hz j V k i J»«L L— J u i ulll LI I I LLI I I LI I I I H I I I I I I I I I I I 1 I I I I I I I I I I I I I II I I I I I I II I I I I I I I I I I I I I I I I I I I II I I I I 8 7 6 5 4 3 2 1 317 Figure 11:2-20 1H NMR Spectrum (CDC13) of Simmondsin Aglycone Benzoate LIST OF REFERENCES 1) A.J. Verbiscar and T.F. Banigan, J. Agricul. Food Chem., 1978, 26, 1456. “ 2) A.N. Booth, C.A. Elliger and A.C. Waiss, Jr., Life Sciences. 1974 15, 1115. 3) C.A. Elliger, A.C. Waiss, Jr., and R.E.Lundin, J.C.S. Perkin I, 1973, 2209- " 4) C.A. Elliger, A.C. Waiss,Jr. and R.E. Lundin, Phytochem., 1974, 13, 2319. 5) C.A. E lliger, A.C. Waiss, Jr., and R.E. Lundin, J. Org. Chem., 1974, 39, 2930. 6) D. Lee, 1968, Ph.D. dissertation, University of West Indies, Part II, P87-145. 7) J. Wu, E.H. Fairchild, J.L. Beal, T. Tomimatsu, R.W. Doskotch, J. Nat. Prod., 1979, 42, 500. 8) A. Sosa, F. Winterraitz, R. Wylde, and A. Pavia, Phytochemistry, 1977, 16, 707. 9) D. Dwuma-Badu, W.H. Watsum, E.M. Gapalakrishna, T.U. Okarter, J.E. Knapp, P.L. Schiff, Jr., and D.J. Slatkin, LLoydia, 1976, 39, 385. 10) L.M. Jackman and S. Sternhall, "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry", 2nd ed ition , Pergamon Press Ltd., Oxford, 1969, pg. 239. 318 319 11) F. Johnson and D.T. Dix, J. Am. Chem. Soc., 1971» 93, 5931. ~ 12) F. Johnson and S.K. Malhotra, ibid.., 1965, 87, 5492. 13) S.K. Malhotra and F. Johnson, ib id ., 1965, 87, 5493* 14) T.E. Walker, R.E. London, T.W. Whaley, R. Barker and N.A. Matwiyoff J^ Amer. Chem. Soc., 1976, 98, 5807* 15) J.B. Strothers, "Carbon-13 NMR Soectroscopy", Academic Press, New York, 1972, pg. 461. 16) L.M. Jackman and S. Sternhall, "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry", 2nd edition , Pergamon Press Ltd., Oxford, 1969, pg. 240. 17) K.G. W allenfels, G. Bechtler, R. Kuhn, H. Trischmann, and H. Egge , Angew. Chem. Internat. E d it., 1963, 2, 515. 18) H.G. Walker, Jr., M. Gee and R.M. McCready, jJ. Org. Chem., 1962, 27, 2100. 19) R.N. Jones, C.L. Angell, T. Ito and R.J. Smith, Can. J.* Chem., 1959, 37, 2007. 20) L.M. Jackman and S. Sternhall, "Applications of Nuclear Magnetic Resonance Spectroscopy in Organic Chemistry", 2nd edition , Pergamon Press Ltd., Oxford, 1969, Pg. 334. 21) Sir Ian Heilbron and H.M. Bunbury, "Dictionary of Organic Compounds", Oxford University Press, 1953, Volume 4, pg. 385. 22) "The C yclitols", by Theodore Posternak, Holden-Day, Inc., Publishers, 1965, pg. 127-143. 320 23) " B eilstein ’s Handbuck der Organishen Chemie", 3rd and 4th Supplement, edited by Hans G. Boit, Springer - Verlag Berlin, 1975, 17(4), 3132. 24) R.K. Crossland and K.L. Servis, J. Org. Chem., 1970, 35, 3195. “ 25) N. Harada, J. Iwabuchi, Y. Yokota, and H. Uda, J. Am. Chem. Soc., 1981, 103, 5590. 26) I. Uchida and K. Kuriyama, T et. L ett., 1974, 3761. 27) A.F. Beecham, Tetrahedron, 1971, 27, 5207. PART III: ISOLATION AND STRUCTURE ELUCIDATION OF ANTIFUNGAL COMPOUNDS PRESENT IN HARDWOOD BARK COMPOST. 321 Part III: Isolation and Structure Elucidation of Antifungal Compounds Present in Hardwood Bark Compost. INTRODUCTION Use of composted hardwood bark for the container production of woody ornamentals and other nursery stock has steadily replaced the use of peat over the last three decades. This is the result of a much lower incidence of Pythium and Phytopthora root rots in compost-potted i/s peat- potted plants. A review of the biochemistry and methodology of composting has been published1. More recently, Hoitink, et al.^ has discussed the effect of various critical factors (N£, aeration, microflora, and pH) on the success of the composting process, whereas L. Basham^ has reviewed the use of hardwood bark compost in disease control. An investigation into the antifungal compounds present in hardwood bark compost is described herein. Three compounds have been isolated from the most active partition fraction and partially identified. The isolation and characterization of three related compounds is still in progress. In addition, elemental sulfur has been isolated from the hexane partition fraction accounting for the majority of the activity there. 322 DISCUSSION AND RESULTS Replacement of peat by composted hardwood bark as a container media for production of nursery stock has led to an investigation of the antifungal components found to be inhibitory to disease-producing Phytophthora species. Screening techniques which included bioautography, sporangial growth assays, and zoospore germination assays indicated a variety of compounds to be present which varied in a c tiv ity against d ifferen t organisms as well as mechanism of action (growth inhibition v_s germination inhibition). The Cladosporium cucumerinum thin layer bioassay follows the basic procedure of K een1* and involves the application of a known quantity of sample ( 0 .5 - 0 .05mg) on a 2 5 0 ju. tic plate (20x20cm) and developing the plate in the appropriate solvent. After drying for thirty minutes, the plate is sprayed with a zoospore suspension. The suspension is prepared by flooding the sporulating organism (grown on PDA for one week at 22°) and filtering through several layers of cheesecloth. The plates were then sprayed with half strength melted PDA until the plates glistened. 323 324 Incubation with continuous light in a moist chamber resulted in white zones of inhibition against a black mycelial background. The sporangial assay, originally developed by Rao^, uses iso la te 544 of Phytophthora cinnamomi grown on Difco lima bean agar for forty-eigh t hours at 2S°C, and is performed by placing leached mycelial disks (two per plate) cut from hyphal mats in 1ml sa lt solutions containing the te st sample (dissolved in 3/xl of MeOH or DMF). Three sporangial counts are made per disk with two replicates per sample. The zoospore germination assay (PSGA), using isolate 599 of Phytophthora citrophthora, is performed by addition of 0.1ml of zoospore suspension (1-3x10^ spores/ml) onto plates of M1 medium containing the sample to be tested (previously dissolved in HeOH or DM SO). The percent germination is counted after two to three hours. The composted hardwood bark was prepared from a mixture of trees in which 70-80% was derived from oak (Quercus: 60% overall), ash (Fraximus) , and hickory (Carya), and the remainder from maple ( Acer), elm ( Ulmus), and tulip poplar (Liriodendron) , with walnut (Juglans) and black cherry (Prunus) being very minor components. The bark was composted in windrows for six months and stored at 4° until needed. The moisture content was determined to be 39% after heating for twelve hours at 106°. 325 Bark Compost <3916 water) (60.33kg wet weight; 36.8kg dry weight) 956 EtOH extraction (total 855 liters) orange-brown tar (62Ugms = 1.7% dry weight basis) chloroform/water extraction I1 chloroform interface 1 water solubles (24gms) solubles (4 0 7 gm s) L | EtOAC water I extraction hexane/90% MeOH extraction I EtOAc water solubles solubles (5gms) (159gms) hex ane 90% MeOH solubles solubles (187gms) (162gms) interface 2 (33•^gms) 1 J I 1) evaporate I 2) dissolve in hexane ! MeOH (1:1) t 3) cool to H and filter I I\ ~ 1 If solubles white crystals (27.9gms) (5.56gms) Figure III-1 Extraction and Partitioning Scheme for Composted Hardwood Bark. Table III-1 Bioassay Results of Crude Bark Compost Partition Fractions. Fraction Phytophthora Spore Germination Assay 3 t i c *1 P.o itrophthora P,c actorum P.c innamomi Assay Interface 1 inactive inactive inactive 1 c EtOAc 55% ------2c h2o inactive inactive inactive in actived Hexane 60% inactive inactive 1 e f 90% MeOH 5% 13% 37% several Interface 2 inac tive inactive inactive 1 f Crystals inactive ------1 g aAssay conducted at 500ppm. Activity reported as % germination and is the average of two determinations. Inactive designates greater than 90% germination. bCladosporium bioautographic assay conducted at 0.2, 0.4, 0.6 and 0.8mg. The value designates the number of sites of inhibition observed. Superscripts designate the tic solvent used on silica gel plates. c30% MeOH in CHC13 (1/2x) and 4% MeOH in CHC13 (1x). d10% H2o in MeOH (1/2x) and MeOH (1x). e 1% MeOH in CHCl^ (1/2x) and benzene (1x). **15% MeOH in CHC13 (1/2x) and CHCI3 (1x). gCHCl3 (lx). hAssay conducted at 125ppm. 327 Extraction of 60.33kg of compost with a to tal of 855 liters of 95% EtOH gave 624gms of an orange-brown tar on evaporation. Using solvent-solvent partitioning, the initial extract was subdivided into seven fractions as illustrated in Figure 1II-1. Bioassaying of this material in a spore germination assay^ showed that four of the seven extracts were inactive (Table III-1). The hexane solubles were inactive against Phytophthora cactoriumi and £. cinnamomi, but demonstrated some activity against citrophthora at 500ppm. Moderate activity was also seen in th EtOAc solubles. However, the 90% MeOH solubles were by far the most effective against all three species. Testing against Cladosporium using the bioautographic technique showed all but the H20 solubles to have at least one active component(s ), and the 90% MeOH solubles to be again the most active. It was interesting to observe the inactivity of the H20 soluble fraction in all assays. The initial hypothesis was that the activity would be concentrated in this fraction since water leachates from pots of compost have been shown to be quite effective in lysing zoospores and cysts as well p as inhibiting sporangial production. In addition, the crude ethanolic extract and seven partition fractions were tested against the following gram-(+) and gram-(-) bacteria, yeast and fungi using the standard agar-disk assay to determine their effectiveness 328 against human pathogens. 1) Bac illu s su b til is ATCC 6633 2) Staphylococcus aureus ATCC 6538 3) Escherichia coli ATCC 9637 4) Proteus vulgari3 ATCC 9484 5) Mycobacterium smegmatis ATCC 14468 6) Candida albicans ATCC 10231 7) Saccharomyces cerevisiae ATCC 2366 8) Aspergillus niger ATCC 16888 9) Trichophyton mentagrophytes ATCC 9972 10) Botrytis a lii ATCC 9435 11) Microsporium gypseum ATCC 11395 12) Mucor rouxii ATCC 24905 All fractions were found to be ineffective at 2 and 4mg per disc; the largest zone of inh ib ition was 2.0mm. The hexane solubles (164gms) were chromatographed over 1.7kg of silicic acid using a stepwise gradient from hexane to 12} MeOH in CHCl^ (Figure III-2) to give 12 fractions. Only the first three showed significant activity in spore germination and bioautographic assays. Minor activity was also observed for fractions six and seven in the Cladosporium tic assay (0.5mg) but was not further investigated. When fraction 2 was further chromatographed over activity grade II silicic acid using pet ether (37-45°) as solvent, yellow crystals were isolated from fractions 5, 329 6, and 7 and identified as sulfur^ (overall yield 0.00216% or 22ppm in dry compost) . This material produced an intense zone of inhibition in the bioautographic assay when tested at O.OSmg. The activity of the adjacent fractions (#1 and 3) from the f ir s t column were shown to result from the presence of sulfur in these fractions as well. Sulfur was also found to be active against _P. cactorum and _P. cinnamomi although the initial hexane fraction was not. This results from a significant increase in the concentration of the material during the purification process. In addition, when comparative t i c ' s and bioautographic assays were run, it was found that the active component of the f ir s t and second interfaces was also due to the presence of sulfur [zone of inhibition at Rj. = 0.55 on silica gel G plates using pet ether (37-M50 ) as solven t]. Thus, the bulk of the a c tiv ity in the hexane solubles was due to elemental sulfur although some very minor inhibitory agents were also present. However, this activity was only observed in the Cladosporium tic assay; not the spore germination assay. The most active fraction presented a challenging isolation problem. Initial work showed that the active materials would tend to "bleed off" a silica gel column rather than elute as bands of localized a c tiv ity . Also, further chromatographic fractionations seem to decrease activity suggesting a synergistic effect between the active components. 330 Hexane Solubles (60% germination - 500ppm) 164gms t ! 1.7kg silicic acid ! hexane— >benzene— > i chloroform-->12%MeOH I in chloroform ! (step-gradient) 12 fractions #1 - 3.96gms - 37% (500ppm) #2 - 3 -44gms - 0% (500ppm) #3 - Q.45gms - 59% (500pptn) #4-012 - 95% (500ppm> (spore germination assay using £. citrophthora) Fraction #2 (3* 4gms) 350 gms s i l i c i c acid (act. II) pet ether (37-45°) “1— -i— I i I I i 1 2 3 4 5 6 7 8 9 10 995 104 24 28 31 1 600 265 155 471 94 mg dissolve in hot chloroform and f ilt e r yellow bispyramidal crystals 795mg (overall yield = 0.00216%) -form of sulfur mp 117-8 23% germ. (25ppm)-P. cactorum 29% germ. (50ppra)-F. cinnamomi Figure III-2 Chromatographic Fractionation of Hexane Soluble Materials 331 Isolation of the active components of the 90S MeOH solubles required a second partitioning scheme as outlined in Figure III-3. This gave two dissimilar, but active fractions (A and B; Table III-2). The most active, fraction A, was consecutively treated with a series of solvents at room temperature or in the cold to produce several p recip itates which were inactive when tested at 500 and 250ppm (Table III-3). Also, when the precipitates were tested in all possible recombinations of three and four samples, no increase in biological activity (synergistic e ffe c t) could be observed at 125, 250 and 500ppm. Chromatography over Sephadex LH-20 using MeOH, and a linear gradient from CHCl^ to 10% MeOH in CHCl^ on a second column, gave nine fractions of variable activity (Group C) . Compound A was isolated after further workup of fractions C2 acd C3 (222gms). A second pass over Sephadex LH-20 with only CHCl^ as solvent, gave a fraction which was further chromatographed over 175gms Of silica gel using hexanerether (1:1) as solvent with increasing concentrations of MeOH to give a 254mg fraction (<2% germination at 25ppm; PSGA). Final purification was obtained over 37gms of silica using increasing concentrations of MeOH in CHCl^ for elution. This resulted in 115mg which crystallized from EtOAcrhexane to give long needles with mp 60.5-61° (50% germination at 50ppm; PSGA). The physical constants for compound A are listed in Table III-4. The compound 332 90% MeOH solubles-138gms (5% germination at 500ppm) I J 600m l 70% aq MeOH, ! extract with chloroform (600 ! and 2x 300ml) and backwash i i 70% aq MeOH hexane:chloroform ! solubles ! extract with (104gms-77%) ! chloroform ( 3x 300ml) ! backwash I — 1 - ir ~ ’ ’ Ii 70% aq MeOH solubles CHOI., solubles ( 4 .86gms-3.6%) (25gms-l8%) Fraction A - 15.3gms I I a) dissolve in solvent (80- I 100ml) at RT or with cooling ! b) filter i i Soluble Material 5 precipitates all (11.2gms) inactive (408gms-27%) I i 250gms Sephadex LH-20 i solvent: MeOH ! two passes 1I Fraction M2 - 591gms (<2% germination at 250ppm; P. citrophthora) » I 138gms Sephadex LH-20 i 650ml chloroform and a linear gradient from chi orofo rm to 10% MeOH in chlorofo rm ( 1. 25L each) » ■" ■“ 1 r J 1 1 ' 1 1 t i 1 1 1 1 Cl C2 C3 C4 C5 C6 C7 is C 9 1. 22 1.33 0. 89 0. 65 0.25 0• 33 0. 57 0.55 0.07 gms 71 11 <2 <2 <2 <2 <2 5. 1 92 200ppm 92 53 11 31 4 <2 19 49 96 50ppm Fig ure I I 1-3 Parti tio ning and Chromatographic Fractio nation of 90% MeOH Solubles 333 Table III-2 Spore Germination Assay Results3 for the 90S MeOH Partition Fractions. Test Test Fraction Fraction Fraction Organism Level A B C P. citrophthora 250ppm <2% 22% 52% 125ppm 93% 92% 92% P. cactorum 250ppm <2% 6% - 125ppm 69% 58% — P. cinnamomi 250ppm 39% 3% - 125ppm 63% 37% aReported as % germination; average of two readings. Table III-3 Weights and Spore Germination Results3 for the Precipitates from the Hexane:Chloroform Solubles (Fraction A). Solvent Temp. Wt. 500ppm 250ppm acetone RT 606mg 95% 98% acetone 0° 787mg 94% 97% ethyl ether RT 2 .42gm 93% 96% o I methanol ro o 2G4mg 94% 97% o i ethyl acetate ro o 62mg 90% 95% solubles — 11.21gm <2% <2% aReported as % germination of £. citrophthora: average of two readings. Table III-4 Physical constants for Compound A. Activity P. c itrophthora spore germination assay 1 1 germination at 100ppm 50% germination at 50ppm 96% germination at 25ppm (Delayed germination and stunting observed) mp 60.5-61° (needles from EtOAc/hexane) tic RfsO.M; acetone:MeOH:water (1:1:1) (silanized silica gel) IR 3585 (0-H s t r .) (CHC1-) 2940 (C-H asym. s t r ., CH2 ) 2860 (C-H sym. s t r ., CH2) 1733 (C=0 s t r . , ester) 1469 (C-H, CH2 scissor) 1378 (C-H, sym bend, CH^) 1250-1180 (C-0 s t r ., asym. and sym.) NMR (300MHz, CDC13, TMS) 4.12 (q , 2H, J = 7. 1, -0-CH2-CH3) 3.39 ( brm, 1H, HO-C-H) 2.29 (t, 2H, J = 7.5, -CH2-C-0) 2.11 (br3, 1H, HO; lost with D2o) 1.62 (brt, 2H) 1.32 (brs, methylene envelope) 335 Table III-4 (continued) 13C NMR (20MHz, CDCl^ IMS) 173-9 (s, C=0) 74.5 (d, HO-C-H) 60.2 (t, CH3-CH2-0-) 34.4 (t) 33-7 (t) 29.5 (t) 29.2 (t) 29.1 (t) 25.6 (t) 24.9 (t) 14.3 (q, CH3-CH2-0-) ms electron impact m/£ 368 (4%, M+-HpOp) 339 (7%) 323 (14%) 376 (20%) 201 (36%) 155 (65%) 125 (23%) 111 (22%) 97 (32%) 83 (42%) 69 (46%) 55 ( 100%) chemical ionization 403 (1.4%) 336 exhibited a peak at m/e 368 (4%) in the electron impact mass spectrum corresponding to the loss of and a peak at m/e 403 (1.4%, M+H) in the chemical ionization mass spectrum. The lack of the optical rotation could be a characteristic of the compound even if chiral with a weak specific rotation, but could also result from a racemic mixture, or plane of symmetry within the molecule. The IR 1 spectrum showed a strong absorption at 1733cm characteristic of a saturated ester, and a band at 3585cm* ^ indicating the presence of a hydroxyl. Otherwise, only C-H bending and stretching absorptions are observed. The NMR spectrum was highly informative. The most unusual feature was a tr ip le t at 4.12ppm and a quartet at 3.39ppm. Together with the presence of the carbonyl in the IR, this is suggestive of an ethyl ester. In addition, a triplet is observed at 2.29ppm characteristic of a methylene deshielded by a carbonyl. The fact that it is a triplet requires an adjacent methylene group. The broad absorption at 2.11ppm is the hydroxyl proton. On addition of D20 this pattern disappears and the broad 1H multiplet at 3.39ppm sharpens suggesting a degree of coupling between the two. The la st absorption is at 1.32ppm corresponding to a methylene envelope. From this information the following partial structures can be drawn. 337 0 OH CH3-CH2-0-C-CH2-CH2-11 and -C-1 H This is suggestive of hydroxylated fatty acid ester. However, long chain hydrocarbons generally exhibit a triplet at 0.85ppm for the terminal methyl group as seen in myristic acid (Figure III-4). This methyl absorption is lacking in Compound A suggesting either a terminal ring substituent, a dimeric or symmetrical molecule, or possibly an w-trisub- stituted species (e.g. trichloro- or tribromo) whose identity is not evident from the IR or NMR. The last piece of data is the NMR taken in CDCl^ at 20MHz. This displayed a total of 11 peaks: a singlet at 173-9ppm for a carbonyl, no olefinics, two oxygenated carbons at 74.5(d) and 60.2(t)ppm for the hydroxyl carbon and ethoxy methylene respectively, and a quartet at 14.3ppm for the ethoxy methyl. All other peaks appear to be triplets in the 23.0-35.0ppm region. The fact that a second quartet for the terminal methyl was not seen provides additional evidence for a symmetrical, or terminal ring- substituted compound. Thus far, the evidence suggested the presence of 11 carbons and 3 oxygens (hydroxyl and ester'). Without hydrogens this only contributes 180 to the molecular weight suggesting a symmetrical molecule since the molecular ion is m/e 402. In addition, a peak at m/e 201 is also 338 9 0 MHz 1 I -1 i i *—l L i 1 I I L- I 1 l t I I I I 1 L I 1 1 I I I I 1 i I 3 2 1 PPM Figure III-4 NMR Spectrum (CDClj) of Myristic Acid. 339 present from a cleavage product, or a doubly-charged species. The following proposed structure is in agreement with the above data. However, more information is required for its absolute verification. 0 OH OH 0 ii i i i CH3-CH2-0-C-CH2-(CH2)6-CH-CH-(CH2>6 -ch2-c-o-ch 2-ch3 Compound B was isolated from fraction C4. This involved a second pass over Sephadex LH-20 with pure chloroform, and further chromatography over s ilic a gel 60 using a step gradient of MeOH:hexane:EtgO (from 4:67:129 to 16:28:56). This resulted in a 192mg fraction found to be active against citophthora (37% germination at 25ppm PSGA). Repeated recrystallization gave 109mg of spherical ( amorphous) crystals from EtOH/hexane. This material consistently lost activity in the final stages of purification which suggests a syn ergistic e ffe ct from the presence of other more active constituents. This compound, however, seems to be more active as a growth inhibitor (severe stunting in hyphal structure) rather than a germination inhibitor although a delay in germination has been observed. An investigation into the nature of the other component(s) is still in progress . The physical constants for Compound B are listed in Table III-5. 340 Table III-5 Physical Constants for Compound B. A ctivity P. cinnamomi spore germination assay 7*3% germination at 200ppm 77% germination at 150ppra 83% germination at 100ppm (delayed germination and extreme stunting observed) mp 67.5-68.5° (EtOAc/hexane-amorphous) (softens above 66°) tic RfsO.54; acetone:MeOH:water (8:7:7) (silanized silica gel) IR 3620 (0 -H s t r . , fre e) (CHC13) 3575 (0 -H s t r . , 1,2 -d iol) 3445 (0-H str. , intermol. H-bonded) 2935 (C-H asym . str ., ch2 > 2860 (C-H sym. s t r . , ch2 > 1727 (C=0 s t r . , ester) 1465 (C-H, ch2 sc is sor) 1375 (C-H sym. bend , ch3) 1250- 1180 (C-0 s t r . , asym. and sym. 1 H NMR (90MHz, cdci3 , TMS) 4.12 (q, 2H, J =7.0, ch3—ch2- 0-) 3.64 (t, 2H * 1-6 .2 , H0-CH2- ch2-) 3-39 (m, 2H, two H-C -0H) 2.30 (t, 2H, J =7.3, -ch2-c= 0 ) 1.96 (m, H0-; lo st with D^O) 1.33 (brs, methylene envelope) 1.25 (t, 3H, J=7.0, -0-CH2-CH3 ) 341 Table III-5 (continued). 13c NMR (20MHz, CDC13, IMS) 174.0 (s, 0=0) 74.5 (d, HO-C-H) 62.8 (t , H0-CH2-CH2) 60.2 (t , CH3-CH2- 0 -) 33.6 (t) 32.7 (t) 29*6 29.5 29-4 29*2 29- 1 25.8 25.6 24.9 14.2 (q, CH3-CH2-0-) 342 Except for two major differences, the NMR spectrum is virtually identical with that observed for compound A except for two major points. An additional two hydrogen tr ip le t appears at 3-64ppm, and the ratio between the intensity of the methylene envelope (1. 33ppm) and the ethoxy methyl group has increased from almost 1:1 to 2:1. The NMR spectrum contains 16 carbon absorptions. Interestingly, 11 of them are identical with those found in compound A. There is one additional oxygenated carbon (62.8ppm) and four possible triplets in the 24-35ppm region. These differences may be accounted for if one terminal ethoxy ester was now present as an alcohol. The ttf-hydroxylated carbon would account for the tr ip le t at 3.64ppm in the NMR spectrum and the 62.8ppm absorption in the 13C NMR. Also, since the molecule is no longer symmetrical, an increase in the number of carbon absorptions and the number of protons is observed. The 1H NMR spectrum of B integrates for 40 protons, and many of the absorptions in the proton and carbon spectra are identical with those seen for Compound A. Although NMR spectra have shown at least four more compounds to be related fatty acids, only one more will be presented since the others are in various stages of isolation and characterization. Fractions C5 and C6 were combined with sim ilar fractions of the previous columns and chromatographed twice over Sephadex LH-20 using a CHCl^ - 343 Table III-6 Physical Constants for Compound C. Activity _P. oitrophthora spore germination assay <2% germination at 25ppm <2% germination at 10ppm 12% germination at 5ppm (extreme stun ting) mp 39_no° (fin e needles from EtOAc/hexane) tic Rf=0.41; acetone:MeOH:water (4:3:3) (silanized silica gel) IR 3620 (0-H s t r . ( free) (CHC1i ) 2930 (C-H asym. s t r ., CH2) 2860 (C-H sym. s t r ., CH2) 1720 (C=0 str., carboxylic acid) 1460 (C-H, CH2 scissor) 1300-1200 (C-0 s tr .) 1H NMR (90MHz. CDC13, IMS) 5.67 (br, 2H, H0-CH2-and H00C-) 5.34 (t, 2H, _J=4. 6 , -CH2-CH=CH-CH2-) 3-65 (t, 2H, J=6.4, HO-CH2_CH2-) 2.34 (t, 2H, J=7.3, -CH2-C=0) 2.01 (ra, 4H) 1.31 (brs, methylene envelope) 344 10% MeOH in CHCl^ linear gradient. A 6l3mg fraction was eluted with 4% MeOH in CHCl^. Further chromatography over silica gel 60 (230-400 mesh) using a step gradient of MeOH in CHC1^ gave a 230mg fraction which was finally purified over 34gms of s ilic a to give 17mg of Compound C 098% germination of _P. c itophthora at 25ppm). The physical constants for Compound C are listed in Table III-6. The NMR of compound C again is related to those of compounds A and B. Differences include loss of the ethoxy group resulting in a free acid whose a-methylene is present at 2.34ppm. A two hydrogen tr ip le t appears in the o lefin ic region (5.34ppm) indicating the presence of a cis or trans double bond. The hydroxy and carboxylic acid protons are present at 5.67, and the tu-methylene appears at 3*65ppm sim ilar to that found in compound B. The three compounds which are discussed represent only a few of the many compounds which are present within the 90% MeOH soluble extract. These compounds appear to be synergistically active as shown by preliminary test results of a mixture of Compound A (0.45mg), Compound B (0.45mg), Compound C (0.85mg), and an unreported fatty acid (0.35mg) [14% germination at 5ppm, and 31% germination at 2 ppin; PSGA]. It must be mentioned, however, that the ethyl esters isolated (Compound A and Compound B) most lik e ly are not naturally-occurring and are formed during extraction of the compost with 95% ethanol. Future studies to verify this 345 would require GC analysis of the TMS derivatized extract fresh compost (extracted with a d ifferen t solven t), and comparison to the TMS derivatives of the known compounds isolated . It has long been known that the long chain saturated (and unsaturated) fatty acids are active against many fungi^. Undecanoic acid appears to be the most active within the series, with the activity dropping off by increasing or decreasing the carbon chain length. It has been reported recently by Seidel® that fermentation of a 240-380° petroleum fraction by Candida guilliermondi gives 50-70J unsaturated acid by-products which is effective in controlling Phytophthora late blight on greenhouse tomatoes. Many naturally-occurring hydroxy acids, polyhydroxy acids, keto acids, epoxy acids and dicarboxylic acids derive from the substitution of hydrogen by oxygen in the fatty acid chain. Depolymerized cutin has been thoroughly investigated over the past few years, and several oxygenated fatty acids have been isolated from a variety of sources^'These monomers consist of palmitic and stearic acid derivatives which possess various combinations of hydroxy, epoxy and keto groups at the 9, 10 and w-positions. CH3- (CH2) 7-CH=CH- (CH2) 7-COOH oleic acid w-hydroxylation ch2- (ch2) ?-ch= ch- (ch2) 7-cooh OH t epoxidatlon hydration CH.- (CH, )-,-CH—CH- (CH,) ,-COOH hydrationreduction ch2-(ch2)?-ch-ch2-(ch2)?-COOH CH.-(CH.),-CH~CH-(CH-),-COOH | 2 2 7 | | 2'7 OH OH OH Figure III-5 Proposed Biosynthetic Pathway for the C10 Cutinic Acids 1 8 9 tr £ 347 The proposed biosynthetic pathway^ for the family of acids is shown in Figure III-5. These acids con stitu te the monomeric units of cutin and suberin, the structural components of the outer layer of plants. It is postulated that these layers act to protect the plant from dehydration, microorganisms, and other harmful agents. The monomeric units of these polymeric coats have been shown to be active against various organisms 1 1 and may constitute an effective means of controlling microorganisms by the plant kingdom. EXPERIMENTAL* Many of the preliminary investigations of this project have been previously discussed^ by Miss L.M. Basham of the Department of Plant Pathology at the Ohio State University with whom this has been a cooperative effort. Miss Basham maintained cultures and performed all bioassays during the course of this investigation as well as aiding in the chromatography of these materials. All of the bioassay techniques used (sporangial assay, zoospore assay, bioautographic assay) have been adequately discussed by Miss Basham and, thus, need not be reviewed here. Plant Material. The composted hardwood bark was obtained by Dr. Harry A. Hoitink, Ohio Agricultural Research and Developement Center, Wooster, Ohio, from Paygro, Inc. in South Charlestn, Ohio. The material was prepared from a variety of tree species and composted in windrows for six months. The moisture content was determined to be 39% by heating 453gms of compost at 106° for twelve hours. All material was maintained at 4° n See Appendix on p . 3 6 4 3^8 349 until needed for the individual studies. Extraction and Partitioning. The compost (60.33kg, wet weight; 36.8kg dry weight) was percolated seventeen times with a total of 855 liters of 95% EtOH to yield 624gms of orange-brown tar (1.69% on a dry weight basis). The solvent was removed using a Precision Scientific steam stripper at 45-47°- The biological activity of the various extracts was followed by the testing of all partition fractions and chromatographic fractions with the spore germination assay and/or the Cladosporium tic assay. Partitioning of the crude ethanol extract is portrayed in Figure III-1. The assay results from these fractions are listed in Table III-1. Chromatography of Hexane S olu b les. The hexane soluble fraction (187gms) was dissolved in 300ml of hexane:MeOH (3:1), cooled to 4° and filte r e d . After repeating four times, 14.29 gms of light brown crystals were obtained which were e sse n tia lly id en tical in composition (by tic) to those obtained from the second interface by a similar procedure. The soluble material (164gms) was chromatographed over 1. 7kg of s i l i c i c acid (grade II) using hexane as the initial solvent (8500ml). The solvent polarity was then increased stepwise from hexane to benzene (2, 5, 12, 30, 60, 80 and 100%) and from benzene to chloroform (3, 6, 15, 35, 70 and 100%) using .5-2. 0L of each solvent. The active material was eluted with hexane, and 2 and 5% benzene in hexane. The first three of twelve fractions were active, with fraction 2 being the most active. Further chromatography of fraction 2 over 350gms of silicic acid (70-230 mesh; heated at 120° for 12 hours) using only petroleum ether (37-45°) as solvent gave a total of ten fraction s. Fractions 5, 6 and 7 (1176mg) were combined and crystallized from hot chloroform and recrystallized from benzene:CHCl^ (1:4) to give 394mg of bright yellow bispyramidal crystals of sulfur: mp 117-8°; tic, Rf=0.55, pet ether (37-45°)» silica gel; visualizes only with vapors; burns and emits an odor of SO2; no IR (CHCl^) or NMR (CDCl^) spectrum; forms a black precipitate of lead sulphide in sodium fusion test 1 2 . Fractionation of the 90% MeOH Solubles. The 90% MeOH fraction (138gms) was partitioned as illustrated in Figure III-3 giving rise to three fractions of variable a c tiv ity . The most active (A: 15-3gms) was consecutively dissolved and filtered in 80-100ml each of acetone (room temperature and 0°), ethyl ether (room temperature), MeOH (-20°), and EtOAc (-20°); refer to Table III-3. The resulting final filtrate (11.2gms) was chromatographed over 250gms of Sephadex LH-20 (1700ml bed; 6.1x58.2cm column) using MeOH as solvent to give five fractions. Fraction 2 (6.07gms) eluted after 500ml and was- 351 rechromatographed a second time on the same column giving rise to four fractions. A 5.91gm fraction (M-2; <2% germination at 250ppm; PSGA) was further chromatographed over 138gms of Sephadex LH-20 (3*2x79*3cm column; bed volume=635ml; void volume 200ml) using 650ml of chloroform followed by a linear gradient from chloroform to 10% MeOH in chloroform (1.25L each). This gave nine fractions (C series, Figure III-3). Isolation of Compound A. Fractions C2 and C3 (2.22gms total; 11 and <2% germ, at 200ppm resp ectively; PSGA) were combined and chromatographed over 138gms of Sephadex LH-20 a second time using only CHCl^ as solvent. After 255ml eluted, 1.98gm of material was obtained and chromatographed over 175gms of silica gel 60 (230-400 mesh; 2.9x48.9cm; 1.15ml/min) using hexane:Et20 (1: 1; H20 sat'd) as solvent with increasing concentrations of MeOH (1, 2, 4, 8, 16 and 32%; 500-700ml each). A to ta l of twenty-one fractions were obtained. Fractions 7 and 8 (253mg; eluted with 1 and 2% MeOH; <2% germ, at 25ppm; PSGA) were combined and chromatographed over 37gms of s ilic a gel 60 (230-400 mesh; 1.45x42cm column; 0.5ml/min) using H2q- sat'd CHCl^ with increasing amounts of MeOH. Compound A (115mg) eluted with 1 and 2% MeOH in CHCl^. This cry sta llized from EtOAc/hexane to give long needles: a ll physical constants are listed in Table III-4. 352 Isolation of Compound B. Fraction C4 (645mg; <2% germ, at 100ppm; PSGA) was chromatographed over 138gms of Sephadex LH-20 a second time using CHCl^ and 5% MeOH in CHCl^ as solvent. Fraction 5 was eluted with 5% MeOH in CHCl^ and combined with similar fractions of previous columns to give 626mg. Chromatography over 65gms of silica gel 60 (230-400 mesh; 5% H20) using a step gradient of MeOH:hexane:Et20 (from 2:33:65 to 16:28:56; 150ml each) resulted in the elution of 192mg of Compound B (37%germ, at 25ppm; PSGA) after 424ml. This crystallized from abs EtOH/hexane to give spherical (amorphous) crystals: all physical constants are listed in Table III-4. Isolation of Compound C. Fractions C5 and C6 (581mg) were combined with sim ilar fractions from previous columns and chromatographed over 138gms of Sephadex LH-20 a second time using a linear gradient from CHCl^ to 10% MeOH in CHCl^ as solvent. This gave 6l 3mg (eluted with 4% MeOH in CHCl^; <2% germ. at 100ppm; PSGA) which was further chromatographed over 48gms of silica gel 60 (230-400 mesh; 10% H2o ; 1.3 x 47.3cm bed) eluting with 2, 4, 10 and 20% MeOH in CHCl^ (150ml each; CHCl^ was H20 -s a t'd ). Fraction 3 (230mg; eluted with 4% MeOH) was finally purified over 34gms of silica gel 60 (230-400 mesh) eluting with 2, 4, 6 , 8 and 16% MeOH in H20- s a t ’d hexane:Et20 (1:2). Compound C (17mg; 798% germ at 353 25ppm; PSGA) eluted with 2% MeOH: all physical constants are listed in Table III-4. 354 «i»4:i;;::ii m;i;;»u Li««;i?wiiiiwii Figure XII—6 Figure XII—6 IR Spectrum for (CHCl^) Compound 90 MHz D^O I I I I I L. L 1 1 I I—L 1 L -1 LllllllLllLllllllllllllll.il PPM 4 3 2 1 355 Figure III-7 NMR (CDCl^) Spectrum for Compound A. 2 0 M Hz j^im jiiiriiiiip m iinn in iiiiinin ijm ti'n iiiiiiipTiiim n iiiiiiin iiiin iMniiii^ in iiin iiKlim iiiniTH'J’fTT ’niiiiiiiiii^miniijinniiiiimniiiii....!— j— '"' q Figure III-8 13C NMR Spectrum (CDCl^) for Compound A. ItAHUMIUQM iue I- IR Spectrum (CHC13) III-9 Figure Compound for B. 357 90 MHz ■ill, 1 PPM Figure 111-10 XH NMR (CDC13) Spectrum for Compound B. 8 5 3 20 MHz f r f 1111 f 11 1 1 1 1 f 111 r i j ) i M 111 f 111111 i > r PPM 150 100 50 Figure III-ll 13C NMR Spectrum (CDCl^) for Compound B. 359 nm ifr f ^EH » U M M U CM-1 Figure 111-12 IR Spectrum (CHCl^) for Compound C. 360 90 MHz 7 6 5 4 3 2 1 PPM Figure 111-13 *H NMR (CDCl^) Spectrum for Compound C. List of References 1) R.P. Poincelot, The Conn. Agr. Exp. Sta., New Have, Conn., Bull. 754, 1975. 2) H.A. Hoitink, J.H. Wilson, H.A. Poole, Proceedings of the 2nd Woody Ornamental Disease Workshop, March 21-22, 197"5, University ot M issouri. 3) L.M. Basham, M.S. Thesis, Department of Plant Pathology, The Ohio State University, 1980. 4) N.T. Keen, J.J.Sim s, D.C. Erwin, E. Rice, and J.E. Partridge, Phytopathology, 1971, 61, 1084. 5) B. Rao, A.F. Schmitthenner, and H.A. Hoitink, Proc. Am. Phytopathol. Soc., 1977, 4, 174. 6 ) Chemical Rubber Co., "Handbook of Chemistry and Physics” , 52nd Edition, 1972. 7) N.E. Rigler and G.A. Greathouse, Amer. Jour. Bot., 1940, 27, 701. 8) H. Seidel, B. Voigt, U. Thust, H. P feiffer and H. Sandvoss, Ger. (East) 144706, CA 94 : 169419 v . 9) "Recent Advances in the Chemistry and Biochemistry of Plant Lipids", edited by T. Gaillard and E.I. Mercer, Academic Press Inc., London, 1975. 362 10) "Recent Advances in Phytochemistry", Volume 11, "The Structure, Biosynthesis and Degradation of Wood", edited F.A. Loewus and V.C. Runeckles, Plenum Press, New York, 1977. 11) P.E. Kolattukudy, J.J. Walton, and R.P. Kushawaha, Biochemistry, 1973. 12, 4488. 12) Vogel, "Textbook of Practical Organic Chemistry", 3rd Edition, John Wiley and Sons, N.Y. 1966, page 1039-41. Append ix Experimental Procedures Solvents and Reagents - all solvents for chromatography and chemical transformations were redistilled and reagents were of analytical purity. Thin-Layer Chromatography Plates - layers of 0.25 or 0.50mm were made using a Shandon Southern Unoplan Spreader. After spreading, the plates were air-dried for 15-20 minutes, oven-dried for 20-25 minutes at 110-120°C, and then allowed to stand at room temperature overnight before using. Adsorbents: 1) Silica Gel 60 G for tic (according to Stahl), EM Reagents, catalog no. 7731 - analytical tic plates. 2) S ilic a Gel 60 PF-254 containing gypsum, EM Reagents, catalog no. 7749 - preparative tic plates. 3) Silica Gel 60, particle size 0.040-0.063Mm* ^ Reagents, catalog no. 9385 - flash and gravity column chromatography. 4) Silica Gel 60, particle size 0.063-0.200/im, EM Reagents, catalog no. 7734 - column chromatography of crude e x tr a c ts. 364 365 5) S ilic a Gel 60 PF-254 for preparative layer chromatography, EM Reagents, catalog no. 7747 - for final column chromatographic pur ification . 6 ) S ilic a Gel 60 HF-254, sila n ized , for thin layer chromatography, EM Reagents, catalog no. 7750 - for tic and column chromatography. 7) A vicel, superfines, FMC Corp., Lot E-425 - for tic and column chromatography. 8 ) Sephadex LH-20, Pharmacia Fine Chemicals, p a rticle size 25-100 microns. Spray Reagents: 1) p-Anisaldehyde - made by the slow addition of 5ml of concentrated H2SO4 to a well-stirred solution of 5ml of p- anisaldehyde in 90ml of 95% EtOH; spray and heat to 110° for 5-10 minutes. 2) Sul furic Ac id a) Slowly add 10ml of concentrated 8280^ to 90ml of ice- chilled ether with stirring; spray and heat to 110° for 5-10 m in u tes. b) Slowly add 50ml of concentrated H2 S 0 || to 50ml of ice- chilled absolute ethanol with stirring; spray and heat to 110° for 5 minutes or as required for visualization. 3) Alkaline AgNO^ - reagent I is made by diluting 1ml of saturated silver nitrate with 200ml of acetone and addition of 2- 10ml of H2O to dissolve the white precipitate; reagent II is made by dissolving 2gms of NaOH in minimum H2O and 366 diluting with 100ml of MeOH: spray I, then II and heat 110° for 1 to 2 minutes. Chromatographs: 1) Droplet Counter-current Chromatograph, Model DCC-A, Tokyo Rikakikai Co., Ltd. 2) Rotation Locular Counter-current Chromatograph, Tokyo Rikakikai Co., Ltd. 3) Preparitive High Pressure Liquid Chromatograph, System 500 A, Waters Associates. Melting Points - uncorrected and determined on a Fisher- Johns hot stage or Thomas-Hoover Unimelt apparatus. Optical Rotations - obtained on a Perkin Elmer 241 Polarimeter at the Na-D line (589nm) in a 1dm cell using the stated solvent, CD Spectra - determined on a Durrum-Jasco ORD/UV-5 with the Sproul Scientific SS-20 modification, or CARY Model 60 Spectropolarimeter and determined in methanol. Infrared Spectra - determined on a Beckman 4230 Infrared Spectrometer in NaCl cells of 0 .1mm path length using CHCl^ or CH2CI2 as solvent, or as KBr p e lle ts. Ultraviolet Spectra - recorded on a Beckman 5260 Ultraviolet and Visible Spectrometer using MeOH as solvent. - l ie and —H NMR Spectra - obtained in the stated deuterated solvent using TMS or DSS as internal standard on a 1) Bruker WP-80 2) Bruker HX-90E, or 367 3) Bruker WM 300. Chemical shifts are listed in ppm on the delta scale, coupling constants reported in Hz, and multiplicities given as: s, singlet; d, doublet; m, multiplet; t, triplet; q, quartet; br, broadened signal. Mass Spectra - low and high resolution were measured on a AEI MS-9 Mass Spectrometer and provided by Mr. C. Weisenberger, Chemistry Department, The Ohio State University; low resolution were also obtained on a DuPont Model 21-491 by Mr. John Fowble, College of Pharmacy, The Ohio State University.