STUDIES TOWARD the TOTAL SYNTHESIS of MANZAMINE a DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degr

Home , Total synthesis, Tryptophol

STUDIES TOWARD THE TOTAL SYNTHESIS

OF MANZAMINE A

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of the Ohio State University

Brian B. Filippini

The Ohio State University

1995

Dissertation Committee: Approved by

Dr. David J. Hart

Dr. Viresh H. Rawal

Dr. Matthew Platz Adviser

Department of Chemistry UMI Number: 9533971

This microform edition is protected against unauthorized copying under Title 17, United States Code.

UMI 300 North Zeeb Road Ann Arbor, Ml 48103 To My Family

ii ACKNOWLEDGEMENTS

i thank Professor David Hart, my adviser, for his guidance and encouragement throughout my stay at OSU. I can’t imagine working for anyone else. He lelt me alone when I needed my space, but was nearby to hear any cries for help. I also thank Professor Paul Sampson at Kent State University for sparking my interest in organic chemistry.

Next, I am grateful to all the past and present Hart group members who made the graduate school experience a bit easier. Special thanks go to my longtime labmate Anne-laure, my buddy Alyx, Dave Y., Vicky, Tony, Ram, Dan, Vincent, Ying, David E., Stephane, and David C.

These people are more than co-workers, they are my friends.

Furthermore, I thank the technical support staff at Ohio St.: Carl Engelman, Drs. Dirk

Friedrich and Charles Cottrell for NMR services, Dr. Judith Gallucci for X-ray crystallographic analyses, Mr. David Chang for mass spectral analyses, and Dr. Kurt Loening for assistance in nomenclature.

Finally, I thank my family for their emotional and financial support throughout my undergraduate and graduate studies. Their love and presence has enabled me to endure the tough times, enjoy the good times, and put life into perspective.

I could ask for no more. VITA

September 12, 1968 Bom, Salem, Ohio

1985-1989 B.S., Chemistry, Kent State University, Kent, Ohio

1989-1991 .Teaching Assistant, The Ohio State University, Columbus

1991-1992, 1994-1995 Research Assistant, The Ohio State University, Columbus

1991-1994 National Needs Graduate Fellow,The Ohio State University

PRESENTATIONS

Filippini, B. B.; Campbell, J. A.; Hart, D. J. "Studies Toward the Total Synthesis of Manzamine A",

25th Central Regional Meeting of the American Chemical Society, Pittsburgh, PA, October

1 9 9 3 .

Filippini, B. B.; Campbell. J. A.; Hart, D. J. "Studies Directed Toward the Total Synthesis of

Manzamine A" Joint Great Lakes and Midwest Regional ACS Meeting, Ann Arbor, Ml, June

1 9 9 4 .

FIELD OF STUDY

Major Field: Chemistry

Studies in Organic Chemistry IV TABLE OF CONTENTS

DEDICATION...... ii

ACKNOWLEDGEMENTS...... iii

VITA...... iv

TABLE OF CONTENTS...... v

LIST OF TABLES...... vii

LIST OF FIGURES...... viii

LIST OF SCHEMES...... ix

CHAPTER PAGE

I. BACKGROUND AND SYNTHETIC STUDIES

A. Introduction...... 1

B. Background...... 2

C. Synthetic Studies Toward the Manzamine Alkaloids ...... 12

v II. PREVIOUS STUDIES AND SYNTHESIS OF ABCD TETRACYCLIC ENONES

A. Retrosynthetic Analysis ...... 31

B. Results and Discussion...... 33

III. STUDIES FOR INCORPORATION OF p-CARBOLINE UNIT OF MANZAMINE A

A. Background...... 55

B. Results and Discussion...... 62

IV. STUDIES FOR INCORPORATION OF THIRTEEN-MEMBERED RING...... MANZAMINE A

A. Introduction...... 8 6

B. Background...... 87

C. Results and Discussion...... 8 8

V. MOST ADVANCED SYNTHETIC INTERMEDIATES AND CONCLUSIONS

A. Results and Discussion...... 107

B. Conclusions...... 110

EXPERIMENTAL SECTION...... 113

LIST OF REFERENCES...... 217

APPENDICES

A. 1H and 13C NMR Spectra of Selected Compounds 223

B. X-Ray Crystallographic Data...... 412 LIST OF TABLES

TABLE PAGE

1 . Product Ratio from Reduction of Ketone 2 1 0 ...... 47

2 . Selected nOe Experiments with 396 ...... 93

3. Bond Lengths for 151 ...... 413

4. Bond Angles for 151 ...... 414

vii LIST OF FIGURES

FIGURE PAGE

1 . Manzamine A and Manzamine A Hydrochloride ...... 1

2. Other Members of the Manzamine Family of Alkaloids ...... 4

3. Ircinol Alkaloids...... 9

4. Alkaloids Supporting Baldwin's Proposed Cycloaddition...... 1 0

5. X-Ray Structure of Enone 151 ...... 52

6 . Selected p-Carboline Natural Products ...... 56

7. Alcohol 393 ...... 93

8 . X-Ray Structure of Alcohol 415 ...... 99

viii LIST OF SCHEMES

SCHEME PAGE

I. Baldwin's Proposed Biosynthesis of Manzamine C...... 5

II. Baldwin’s Proposed Biosynthesis of Manzamine B Tetracycle ...... 6

III. Baldwin's Proposed Biosynthesis of Manzamine A...... 7

IV. Interconversions from Ircinal A ...... 8

V. Interconversions from Ircinal B ...... 8

VI. Conversion of Manzamine B into Manzamine ..J...... 9

VII. Biosynthetic Precursors of Manzamine C ...... 10

VIII. Proposed Biosynthesis of Keramaphidin B ...... 11

IX. Proposed Biosynthesis of Madangamine A ...... 12

X. Nakagawa's Synthesis of Manzamine A ...... 13

XI. Gerlach’s Synthesis of Manzamine C ...... 14

XII. Simpkin's Approach to Manzamine A...... 15

XIII. Leonard's Approach to Manzamine A ...... 15

XIV. Baldwin's Biomimetic Synthetic Studies ...... 16

XV. Yamamura's Approach to Manzamine A ...... 17

XVI. Marko's Approach to Manzamine A...... 18

XVII. Hart’s Free Radical Approach to Manzamine A ...... 19

XVIII. Overman's Approach to Ircinal A and Manzamine A ...... 20

XIX. Winkler’s Preparation of Photocyclization Precursor ...... 21 ix XX. Winkler's Preparation of Manzamine A Tetracycle ...... 22

XXI. Condensation Substrates for Martin's Cyclization Precursor...... 23

XXII. Martin's Preparation of Manzamine A Tetracycle ...... 24

XXIII. Nakagawa’s Preparation of Manzamine A Tetracycle ...... 25

XXIV. Pandit’s Preparation of a Racemic Manzamine A Tricycle ...... 26

XXV. Pandit's Preparation of a Homochiral Manzamine A Tricycle ...... 27

XXVI. Pandit's Preparation of ABCE Tetracycle ...... 28

XXVII. Pandit's Preparation of ABCD Tetracycle with p-Carboline...... 29

XXVIII. Retrosynthetic Analysis for Manzamine A ...... 32

XXIX. Preparation of C/s-Octahydroisoquinoline 156 ...... 34

XXX. Two Step Conversion of 156 -> 170 ...... 35

XXXI. Addition of a Lithium Acetylide to Aldehyde 170 ...... 35

XXXII. Kishi's Use of CrCl2 -NiCl2 for the Halichondrins...... 36

XXXIII. Kishi's Use of CrCl2-NiCl2 for Palytoxin ...... 37

XXXIV. Preparation of lodoalkyne 181 and lodoalkene 182 ...... 37

XXXV. CrCl2 -NiCl2 Mediated Couplings with 170 ...... 38

XXXVI. Preparation of Acylsulfonamide 189 ...... 39

XXXVII. Explanation for Retention of Configuration ...... 40

XXXVIII. Strategy to Invert Allylic Alcohol 174 ...... 41

XXXIX. Inversion of Allylic Alcohol 188 ...... 42

XL. Thompson's Inversion of a Benzylic Alcohol with Azide ...... 42

XLI. Proof of Stereochemistry for Attempted Inversion ...... 43

XLII. Wipf's Inversion of Hydroxyl via Oxazoline Intermediate ...... 44

XLIII. Manipulations of Secondary Amide 2 0 4 ...... 45

XLIV. Plan for Inversion of Imino Ether ...... 46

XLV. Reduction of Propargyl Ketone 210 ...... 47 x XLVI. Recycling of Alcohol 188 ...... 48

XLVII. Preparation of Homoallylic Alcohol 215 ...... 49

XLVIII. Epoxidation of Alcohol 2 1 5 ...... 50

XLIX. Completion of Enone 151 ...... 51

L. Preparation of Enone 152...... 53

LI. Labile Acetate of ABCD Tetracycle ...... 54

Lll. Bischler-Napieralski Reactions...... 57

LMI. Kende's Bischler-Napieralski for p-Carboline Preparation ...... 57

LIV. Protic Pictet-Spengler Reactions ...... 58

LV. Rinehart's Use of Pictet-Spengler for Eudistomin Alkaloids...... 58

LVI. Cook’s Use of Pictet-Spengler for Pyridindolol ...... 59

LVII, Queguiner's p-Carboline Synthesis ...... 59

LVIII. Bogeys p-Carboline Synthesis ...... 60

LIX. Molina's p-Carboline Synthesis...... 60

LX. Ciufolini’s p-Carboline Synthesis...... 61

LXI. Plan for Addition of p-Carboline ...... 62

LXII. 1.3-Oxygen Rearrangement with Cr (VI) Reagent ...... 63

LXIII. Palladium-Mediated Allylic Acetate Rearrangements ...... 64

LXIV. Model System for Acetate Rearrangement ...... 64

LXV. Preparation of 1-Amino-p-Carboline ...... 65

LXVI. Preparation of 1 -Chloro-p-Carboline ...... 6 6

LXVII. Synthesis of 1-Bromo-p-Carboline ...... 6 6

LXVIII. Bracher’s Results with Phosphorous Oxybromide ...... 67

LXIX. Corcoran's Preparation of 2-lodopyridine ...... 67

LXX. Preparation of 1-lodo-p-Cait>oline ...... 6 8

LXXI. Additions of 1-Lithio-p-Carboline Anions...... 6 8 xi LXXII. An Unexpected Acelate Rearrangement ...... 69

LXXIII. Preparation ot Model Enone 300 ...... 70

LXXV. Additions to Enone 3 0 0 ...... 72

LXXVI. Addition of Acyl Anion Equivalent to Enone 151 ...... 72

LXXVII. Dithiane Model ...... 73

LXXVIII. Model Aldehyde for Pictet-Spengler Reaction ...... 74

LXXIX. Cook's Aprotic Pictet-Spengler Reaction ...... 75

LXXX. Preparation of /V-Allyltryptamine ...... 76

LXXXI. p-Carboline 332 via Pictet-Spengler ...... 77

LXXXII. Alternative Preparation of p-Carboline 332 ...... 78

LXXXIII. Palladium-Mediated Addition of p-Carboline Unit ...... 79

LXXXIV. Examples of Palladium-Mediated Couplings ...... 80

LXXXV. Preparation of Enone 3 3 6 ...... 80

LXXXVI. Preparation and Plan for Diol 349 ...... 81

LXXXVII. Unexpected Oxidation of Diol 3 4 9 ...... 82

LXXXVIII. Attempted Addition to a P-Alkoxyketone ...... 83

LXXXIX. Alternative Strategy Adding a Acyl Anion Equivalent ...... 84

XC. Oxidation of Epoxy Alcohol 217 ...... 85

XCI. New Strategy to Introduce Acyl Anion Equivalent ...... 85

XCII. Strategy for Introducing Thirteen-Membered Ring ...... 8 6

XCIII. Corey’s Macrolactamization ...... 87

XCIV. Bai’s Macrolactamizations ...... 87

XCV. Preparation of Orthoester 377 ...... 89

XCVI. Preparation of Orthoester 376 ...... 90

XCVII. Preparation and Use ol Bromide 385 ...... 91

XCVIII. Organometallic Additions to Enone 152 ...... 92 xii XCIX. Pandit's Attempted Organometallic Additions ...... 94

C. Olefin Metathesis Approach for Thirteen-Membered Ring ...... 95

Cl. Ramberg-Backlund Approach for Thirteen-Membered Ring ...... 96

Cll. Martin’s Metatheses ...... 96

CHI. Pandit's Organometallic Additions...... 97

CIV. Addition of Grignard Reagents to Enone 1 5 2 ...... 97

CV. Functionalization of Ally! Group ...... 99

CVI. Deprotection of Lactam Nitrogen ...... 100

CVII. Literature Reductions of Amides via Iminoethers ...... 101

CVIII. Attempted Reduction of Amide 410 ...... 102

CIX. Literature Reductions of Lactams via Thiolactam ...... 102

CX. Preparation of Amine 420 ...... 103

CXI. Exocyclic Amide Approach ...... 104

CXII. Alkylation of Amide 418 ...... 105

CXIII. Plan for Prepration fo Alcohol 447 ...... 107

CXIV. Synthesis of Ketone 4 5 4 ...... 108

CXV. Attempted Addition of p-Carboline to Ketone 4 5 7 ...... 109

CXVI. Attempted Elimination and Use of Enone 4 5 5 ...... 110

CXVII. Synthesis of Ketone 454 from Benzoic Acid ...... 111

xiii CHAPTER I.

MANZAMINE A: BACKGROUND AND SYNTHETIC STUDIES.

A. Introduction

The objective of this research was to continue studies toward the total synthesis of manzamine A ( 1 ) (Figure 1 ) . 1 The results of these studies will be reported in several chapters.

The first chapter will present background information and other groups' studies toward the synthesis of the manzamine family of alkaloids. The second chapter will describe a retrosynthetic analysis, present previous progress on this approach to the synthesis of manzamine A and detail the preparation of several tetracyclic intermediates. The third chapter will address studies towards the incorporation of the p-carboline unit. Efforts toward construction ol the thirteen membered ring will be presented in chapter four, and chapter five will present the most advanced intermediates en route to manzamine A and summarize these studies.

34 33 Manzamine A (1)

Figure 1. Manzamine A and Manzamine A Hydrochloride.

1 B. Background

Manzamine A (1) is a polycyclic marine alkaloid isolated in 1986 by Higa and co-workers

from the sponge Haticlona sp. collected off Manzamo, Okinawa, Japan. 1 This compound has

been reported to inhibit the growth of mouse leukemia cells having an IC 5 0 of 0.07 pg/mL.

Manzamine A hydrochloride (2) (Figure 1) was isolated as colorless crystals from methanol,

decomposing when heated above 240°C and having a specific rotation at the sodium D-line of

+50° (c 0.28, CHCI 3 ) at 20°C. The molecular formula of the free amine was determined by

HREIMS to be C 3 6 H4 4 N4 O (nVz 548.3510) and by LRFABMS (M++1 at m/z 549). The 13C NMR

data indicated that all 36 carbons were unique, 17 sp2- and 19 sp3-hybridized atoms, and the UV

spectral data in methanol suggested the presence of a p-carboline chromophore. The 1H NMR

spectrum suggested the presence of two di- and one trisubstituted olefin, and the IR spectrum

revealed a tertiary alcohol by the stretch at 1065 cm '1. Although the structure could not be

elucidated from these standard spectral methods, the structure and absolute configuration were

secured by x-ray crystallography, and reported as shown in Figure 1. The standard ring system

lettering and atom numbering of manzamine A is also shown in this illustration.

In 1987, Nakamura and co-workers independently isolated the same compound from the

Okinawan marine sponge Pellina sp., and named the molecule keramamine A . 2 Nakamura

reported keramamine A to have antimicrobial activity, showing a minimum inhibitory concentration

versus Staphylococcus aureus of 6.3 pg/mL. The specific rotation of the hydrochloride salt of

keramamine A was reported as +44.3° (c 1.09, CHCI 3 ), the material decomposed above 200°C,

and UV data indicated the presence ot a p-carboline chromophore. The molecular formula was

determined to be C 3 6 H4 4 N4 O for the free base by HREIMS ( m/e 548.3495), and single x-ray crystal analysis data of the HCI salt secured the structure as identical to that reported by Higa for

manzamine A hydrochloride.

Manzamine A is only one member of a growing family of p-carboline alkaloids. For example, in 1987 Higa reported the isolation of additional polycyclic alkaloids from the sponge Haliclona sp. Manzamine B (3) (Figure 2} was isolated as colorless crystals from ethyl acetate, and

its structure was secured by x-ray crystallography .3 The tetracyclic nucleus of manzamine B

closely resembles the polycyclic nucleus of manzamine A , but differs by the absence of the C r

C2 double bond and the C 2 1 -N4 bond, and the presence of a C 2 -C3 epoxide. The structure of

manzamine C (4) (Figure 2) was also unambiguously confirmed by x-ray crystallography .3 It is the

structurally least complex member of the manzamine family. Manzamines B and C showed IC 5 0 values of 6 and 3 pg/mL, respectively, against P388 mouse leukemia. These numbers have led to the belief that the p-carboline moiety is important for the observed activity, although the mode of action is not known. When Nakamura isolated keramamine A from the sponge Pellina sp., he also discovered another bioactive molecule given the name keramamine B (5), and the structure was determined by standard spectral methods and reported as shown in Figure 2 .2 Keramamine

B showed antimicrobial activity, as demonstrated by its minimum inhibitory concentration of 25 pg/mL against Staphyloccus aureus.

A third species of sponge, Xestospongia sp., was found by Higa in 1988 to be another natural source of manzamine alkaloids. These new compounds were named manzamines E ( 6 ) and F (7) (Figure 2 ) 4 Both compounds showed IC5 0 values of 5 pg/mL against P388 murine leukemia cells. Manzamines E and F were reported to have the identical pentacyclic nucleus and p-carboline framework as manzamine A, but to differ in oxidation state at Cts and C 3 4 . Manzamine

F was found to have the same spectral data as keramamine B, thus leading to a revision of the structure previously assigned to this natural product. Since the manzamines have been isolated from various sponge species, it has been suggested that a symbiotic microorganism present in all these sponges is the actual source of the manzamine alkaloids.

Manzamines H (8 ) and J (9) (Figure 2) were isolated from a fourth sponge genre, Ircinia sp., in 1992 by Kobayashi ,5 Manzamine H differs from manzamine A by the absence of the C 21 -

N4 bond, and the presence of the tetrahydro-p-carboline moiety, whereas manzamine J is dissimilar only by lacking the C2 1 -N4 bond. These two compounds exhibited cytotoxicities against L1210 murine leukemia with IC 5 0 values of 1.3 and 2.6 (ig/mL and against KB human

epiderimoid carcinoma cells with IC 5 0 values of 4.6 and greater than 10 pg/mL, respectively.

OH OH

Manzamine B (3) Manzamine C (4) 5 Manzamine E ( 6 )

OH OH OH OH

Manzamine F (7) Manzamine H ( 8 )

Figure 2. Other Members of the Manzamine Family of Alkaloids.

Although a number of members of the manzamine family of alkaloids have been isolated, their discoveries had been accompanied by statements that their biosyntheses remained a

mystery. This void was filled by Baldwin in 1992 when he proposed that the biosyntheses of the

manzamines orginates from three simple building blocks .6 Baldwin first presented a straightforward sequence for manzamine C, starting from tryptophan (10), acrolein ( 1 1 ), and symmetrical dialdehyde 12 via sequential reductive couplings with ammonia (Scheme I).

Scheme I. Baldwin's Proposed Biosynthesis of Manzamine C.

1 2 4

The proposed biosynthesis for manzamine A is far more elegant (Scheme II). Beginning with the same three-carbon and ten-carbon units, it was suggested that condensation with ammonia could afford bisdihydropyridine 13. Tautomerization provides iminium ion 14, which could undergo an endo Diels-Alder reaction to yield iminium ion 15. This step would set the relative stereochemistry and regiochemistry required by manzamine A. Redox exchange between the iminium ion and tertiary amine could give cation 16, and opening of the iminium ion would afford the aldehyde and secondary amine of 17. The resulting tetracyclic nucleus closely resembles manzamine B. Condensation of tryptophan onto the aldehyde would give 18, and epoxidation of the double bond at C 2 -C3 would afford manzamine B (3). Deprotonation of Ci and epoxide opening would provide the tertiary allylic alcohol of manzamine A as in 19, and allylic oxidation in 6 the 1 1 -membered ring could provide 20. Displacement of the hydroxyl by the secondary amine completes Baldwin's proposed biosynthesis of manzamine A (Scheme III).

Scheme II. Baldwin's Proposed Biosynthesis of Manzamine B Tetracycle.

CHO

OHC CHO NH,

CHO 1 3 OHC II

1 4

CHO

1 5 1 6 1 7 7

Scheme III. Baldwin's Proposed Biosynthesis of Manzamine A.

1 0

20 1 9

When published, this sequence had no supporting evidence, such as the isolation of any structurally similar intermediates from the sponges. Since his initial report, however, a number of alkaloids consistent with Baldwin's pathway have been isolated. Furthermore, simple interconversions of several manzamine alkaloids and short syntheses of these polycycles from complex natural precursors have been accomplished by simple chemical transformations. In the paper describing manzamines H and J, Kobayashi also reported two other novel alkaloids, ircinals

A (21) (Scheme IV) and B (22) (Scheme V ) . 5 These compounds are closely related to manzamines A and J, having identical pentacyclic and tetracyclic nuclei, respectively. Whereas the manzamines have a p-carboline unit at C i, the ircinals have a formyl group in this position. This occurrence is consistent with Baldwin's theory. The conversion of ircinal A to manzamine A was accomplished by Kobayashi (Scheme IV). Thus, treatment of ircinal A with tryptamine and trifluoroacetic acid afforded the Pictet-Spengler product 23. Oxidation with dichlorodicyanobenzoquinone (DDQ) fully aromatized the p-carboline unit to yield manzamine A (1) (Scheme IV).

Ircinal B, upon Pictet-Spengler condensation, provided manzamine H ( 8 ), and treatment with

DDQ gave manzamine J (9) (Scheme V).

Scheme IV. Interconversions from Ircinal A.

CHO

OH 1 0 OH DDQ

50% 50%

2 1 2 3

Scheme V. Interconversions from Ircinal B. Manzamine B (3) was also converted into 9 by treatment with sodium hydride to provide the Ci -Cz

double bond and the tertiary hydroxyl at C 3 (Scheme VI).

Scheme VI. Conversion of Manzamine B Into Manzamine J.

NaH OH

54%

3 9

Kobayashi and co-workers further isolated ircinots A (24) and B (25) from the sponge Ircirtia sp.

(Figure 3)7 These alkaloids differ from ircinals A and B by the presence of a hydroxymethyl group

at C1 instead of a formyl group, and they also have the opposite absolute configuration.

CHjOH

2 4 2 5

Figure 3. Irclnol Alkaloids.

Baldwin's scheme focused on a C3 unit (acrolein) and a C10 unit (dialdehyde 1 2 ), and their reductive couplings with ammonia. Kobayashi's recent isolation of keramaphidin C (26) and keramamine C (27) from the sponge Amphtmeden sp. supports this aspect of the theory

(Scheme VII ) .8 Keramaphidine C may arise from condensation of ammonia onto dialdehyde 12 10

and subsequent reduction to yield the natural product 26. Michael addition of amine 26 to

acrolein and Pictet-Spengler condensation of tryptamine provides tetrahydro-p-carboline 27,

identilied as keramamine C. Oxidation to the fully aromatic p-carboline unit leads to manzamine C.

Scheme VII. Biosynthetic Precursors of Manzamine C.

NH,

CHO CHO

: > N CHO

1 2 2 6

2 7

Figure 4. Alkaloids Supporting Baldwin's Proposed Cycloaddition.

Isolation of the ircinals and ircinols supported Baldwin's theory for the origin of the p- carboline unit, and the two manzamine C precursors supported the "C3-C10" theory, but these alkaloids gave no support for the proposed cycloaddition. Several polycyclic marine alkaloids 11

isolated recently, however, appear to be either manzamine precursors or arise from a similar Diels-

Alder pathway. Xestocyclamine A (28) and ingenamine (29) appear to be formed via

intramolecular cycloadditions of a bis(hydropyridine ) . 9 - 1 0 These compounds differ only in the

position of the double bond in the 13-membered ring and the stereochemistry of the hydroxyl

group (Figure 4). A similar process could provide two-carbon homolog xestocyclamine B (30)

(Figure 4). Ingenamine was isolated from the marine sponge Xestospongia ingens, and the

xestocyclamines were isolated from the sponge Xestospongia sp., the same sponge genus that

Higa reported to be sources of manzamines E and F.

The natural alkaloid most similar to one of Baldwin's proposed biosynthetic precursors is

keramaphidin B (31),11 a pentacyclic compound isolated by Kobayashi from the sponge

Amphimedon (Scheme VIII). This molecule closely resembles intermediate 15 presented by

Baldwin (Scheme II), and may be formed by cycloaddition of 32. Keramaphidin B requires

reduction of the iminium ion (33) to amine 31 instead of the redox exchange proposed by

Baldwin.

Scheme VIII. Proposed Biosynthesis of Keramaphidin B.

3 2 3 3 3 1

Madangamine A (34) has also been isolated from the sponge Xestospongis ingens by

Andersen, and although it does not appear to be a manzamine precursor, a similar cycloaddition pathway of bis(hydropyridine) 35 seems likely for its formation (Scheme IX ) . 1 2 The discovery of

such alkaloids supports Baldwin's theory and has led to the general acceptance of this pathway for

the biogenesis of the manzamine family of alkaloids.

Scheme IX. Proposed Biosynthesis of Madangamlne A.

3 5

C. Synthetic Studies Toward the Manzamine Alkaloids.

Since the first member of the manzamine family of alkaloids was discovered in 1986,

research groups around the world have initiated programs focusing on the total synthesis of these

compounds. Two syntheses of the simplest member, manzamine C (4), have been reported. In

1989, Nakagawa and co-workers published the first total synthesis of manzamine C (Scheme X ) . 13

Ditosylate 36 was prepared in three steps from silyloxyacetylene 37. A double displacement with

p-toluenesulfonamide and deprotection gave azacycloundecene 38 in good yield. The aromatic

segment of 4 was prepared by condensation of tryptamine with ethyl chlorooxalate, followed by

Bischlar-Napieralski cyclization with POCI3 , and aromatization to provide p-carboline ester 39.

Heating a mixture of ester 39 and amine 38 provided amide 40. Reduction of 40 with UAIH 4 afforded manzamine C (4). 13

Scheme X. Nakagawa’s Synthesis of Manzamine C.

Tsi OTs OSiMejtBu 3 Steps a,b

3 7 3 6

N, N, COaEt 3 9 4 0

(a) TsNH2 , NaOH, n-Bu4 t; (b) Red-AI; (c) 38, A; (d) L 1AIH4

Gerlach and co-workers reported a total synthesis of manzamine C in 1993 using a similar approach (Scheme XI ) . 14 The p-carboline portion was introduced using an aprotic Pictet Spengler condensation of W-benzyltryptamine (41) with aldehyde 42 to yield tetrahydro-p-carboline 43.

Sequential dehydrogenation with Pd/C and conversion of the ortho ester into a methyl ester provided 44. Carbamate 45 was prepared in eight steps from 5-hexynoic acid, and was cyclized and reduced to provide amine 38. Condensation of amine 38 with ester 44 afforded 40, the final intermediate in Nakagawa's synthesis.

Manzamine A has been the most sought-after member of the family, but its total synthesis has yet to be accomplished. Its complex cyclic array coupled with its biological activity, highest among the family, has made it an attractive synthetic target. Most of the approaches towards manzamine A have focused on initial preparation of the AB or ABC ring system, and although a 14

variety of approaches have been reported, the most common strategy involves a cycloaddition to

prepare the c/s-fused perhydroisoquinoline framework.

Scheme XI. Gerlach's Synthesis of Manzamine C.

CHO

Bn 4 2 Bn a,b I H

4 3 O 4 4

NHtBoc

3 8 4 0

(a) Pd-C, 90%; (b) H 2 SO4 ; KOH; MeOH, 90%; (c) TFA, 93%; (d) LiA!H 4 , 92%; (e) 4-DMAP, 44, 73%.

Simpkins and Imbroisi have reported the preparation of a bicyclic intermediate with an eight-carbon side chain present for incorporation of the thirteen membered ring (Scheme XII ) . 15

Dienophile 46 was prepared in four steps from y-valerolactam, and diene 47 was obtained in nine steps from 5-hexyn-l-ol. Although simpler model systems gave better results, cycloaddition between 46 and 47 using zinc bromide in dichloromethane gave the cycloadduct 48 in low yield

(27-46%). Closure of the thirteen membered ring was investigated despite the problematic cycloaddition. A variety of conditions provided low yields of two cyclized products, tertiary amide

49 and imino ether 50, neither of which were well-characterized. No improvements in this route have been reported by Simpkins. 15

Scheme XII. Simpkins Approach to Manzamine A.

MsO SPh SPh

MsO 4 8 4 7

SPh SPh

5 0 4 9

(a) ZnBr2 , 27-46%; (b) various bases and conditions: NaH, K2 C0 3 , CsF.

Scheme XIII. Leonard’s Approach to Manzamine A.

u w CON Me, Me02C MeaN— ^ Me2N— ^

bSO, ^ r - b — r-b • Bn' V* 5 1 5 2 5 3 5 4

(a) BnNH2, NaBH3CN, 63%; (b) CH2 =CHCOCI, Et3 N, 81%; (c) A, 100%

Leonard and co-workers have taken an intramolecular Diels-Atder approach to the bicyclic nucleus

(Scheme XIII ) . 1 6 Sulfolene 51 served as a precursor to aldehyde 52, which was converted to unsaturated amide 53 by reductive amination with benzylamine and sodium cyanoborohydride followed by /V-acylation using acryloyl chloride. Heating 53 at reflux for 24 h in toluene lead to sulfur dioxide extrusion and cycloaddition to provide 54 in a quantitative yield. A trans.cis ratio of

7:1 at the ring juncture was observed. This ratio increased to 9:1 when the benzyl group was 16 replaced by an allyl group. The major ( trans) stereochemical relationship is the opposite to that required by the manzamine A skeleton.

Scheme XIV. Baldwin's Biomlmetfc Synthetic Studies. n. k > 5 6 5 7 5 5 5 9 5 8 k

6 OHC

CH. NH 1

6 2 6 1 6 0

(a) CH3 CH2 Br, 8 6 %; (b) NaBH4l 55%; (C) m-CPBA, 87%; (d) (CF 3 CO)2 0: (e) pH 8.3 TRIS/HCI buffer, then NaBH4, 10% from 57; (f) m-CPBA, 75%; (g) (CH 3 CO)2 0 , 50%.

Baldwin has initiated a research program to model his proposed biosynthesis of manzamine A (Scheme XIV ) . 17 These studies focus on a Diels-Alder reaction between a 1,6- dihydropyridine and conjugated iminium ion 55, prepared in four steps from 3-methylpyridine

(56) via amine oxide 57. Treatment of 55 with a pH 8.3 buffer and subsequent reduction with sodium borohydride provided the endo cycloadduct 58 in 10% yield from 57. The major product of the reaction was tetrahydropyridine 59. Selective oxidation gave 60, and treatment with acetic anhydride provided amide 61. Completion of the isoquinoline bicycle of the manzamines will require cleavage of the N 1-C6 bond (61 -> 62). 17

Yamamura and co-workers have prepared a potential manzamine A precursor containing the AD ring system (Scheme XV ) . 18 Their strategy focused on an intramolecular Diels-Alder reaction of 63 that would form the BCE rings in one step, affording an ABCDE pentacycle. The

AD bicycle preparation began with y-valerolactam, which was converted into aldehyde 64 in seven steps. Addition of lithium acetylide 65 to the aldehyde followed by Mitsonobu reaction with amine equivalent 6 6 afforded acylsulfonamide 67. Cyclization precursor 6 8 was prepared in another seven steps, and the azocine ring of 69 was formed in high yield by treatment of 6 8 with potassium f-butoxide.

Scheme XV. Yamamura's Approach to Manzamine A.

SES H SPh SPh

,OCH.

3 steps OR

(a) LiC=C(CH2 )4 0 THP (65), 55%; (b) Ph3 P, SESNHt-Boc (6 6 ), 8 8 %; (C) KO/-Bu, 85%.

An intramolecular conjugate addition was used by Marko and co-workers in an expedient preparation of the tricyclic ABC system of manzamine A (Scheme XVI ) . 1 9 Indole 3-carbox- aldehyde (70) was converted to amine 71 by reductive amination with butylamine and sodium 18

borohydride. Michael addition of 71 to acrolein followed by a Wadsworth-Emmons reaction of

aldehyde 72 with phosphonate 73 gave a good yield of a 1:1 mixture of E,E- and Z,E-cyclization

precursors 74. Treatment of 74 with 1 equivalent of lithium hexamethyldisilazide provided a 1:1

mixture of tetracyclic adducts 75 and 76. The aromatic ring is to serve as a handle for

incorporation of the azocine ring. A different intramolecular Diels-Alder approach was also

reported by Marko, but these studies were abandoned due to mediocre yields and low selectivity

for the c/s-perhydroisoquinoline ring fusion required by manzamine A .2 0

Scheme XVI. Marko's Approach to Manzamine A.

CHO a,b

H H H 7 0 7 1 7 2 d

Bus

e +

H COgCHg H COjCHj H COaCH3

7 6 7 5 7 4

(a) BuNH2 ; (b) NaBH4i 75-80%; (c) CH 2 =CHCHO, DBU (d) (Me0) 2 P(0)C H 2 C H =C H -C 0 2 CH3 (73). KOf-Bu, 75% ( 2 steps); (e) LiHMDS, 35-50%.

Hart and McKinney reported a synthesis of the ABC portion of manzamine A in an approach featuring free radical cyclization to form the AB ring system and subsequent C ring formation using an electophile-initiated closure (Scheme XVII ) . 21 Dihydrobenzoic acid 77 was converted into amide 78 in three steps. Free radical cyclization of 78 gave a 4:1 mixture of c/s and frans-perhydroisoquinolines 79, respectively, in good yield. Cyclization precursor 80 was prepared from 79 using a five-reaction sequence, and cyclization to afford 81 was accomplished

using iodine. Dehydrohalogenation using DBU afforded pyrroloisoquinoline 82 to complete the

preparation of the ABC ring system. Although this paper marked the first published route to the

manzamine tricycle, the route was abandoned as the sequence was not amenable to large scale

synthesis.

Scheme XVII. Hart's Free Radical Approach to Manzamine A.

SePh

N— OCH3 7 7 7 8 5 steps 7 9 CD NHCOaEt

8 0

(a) (PhO)2 PON3, PhSeCH 2 CH2 NH2, 87%; (b) UAIH 4 , 49%; (c) CH 3 COCI, 84%; (d) n-Bu 3 SnH, AIBN, 67% (4:1 ds'.trans) (e) l2, K 2 C 03, 76%; (f) DBU, 62%.

The preparation of a highly functionalized and enantipure tricycle has been reported by Overman

(Scheme XVIII ) . 2 2 This approach focused on sequential fusing of the A and C rings to a

substituted cyclohexenone, followed by incorporation of the formyl unit found in the ircinal family

of natural products. Enantiomerically pure enone 83 was prepared from D-(-)-quinic acid (84),

and converted into tertiary amide 85 in four steps. Oxidative cleavage of the terminal vinyl group

and reductive amination of the resulting aldehyde with benzylamine and sodium triacetoxyborohydride provided compound 8 6 in good yield. Treatment of 8 6 with formaldehyde

and formic acid gave 87. Conversion of the secondary alcohol into a carbamate, protecting group exchange of the A-ring nitrogen, and sequential treatment with eerie ammonium nitrate and 20 camphorsulfonic acid affected nitrogen deprotection and ring closure to tricyclic enamide 8 8 in high yield.

Scheme XVIII. Overman's Approach to Irclnal A and Manzamine A.

(AnsJBnNOC^^I

O 8 4 8 3

r f S (AnsJBnNOC^^LJ

(AnsJBnNOC^" O O

O 88 8 7 86

oo o

8 9 9 0 9 1

(a) Bu3 S n C H 2 CH=CH2. TBSOTf; p-TsOH, 8 8 %; (b) DBU, TBSCI, 84%; (c) LiHMDS, ICH2 CONBn(Ans); (d) Na2 S2 0 4 , 75% <2 steps); (e) 0 s 0 4, N al04; (f) BnNH2 , NaB(OAc)3 H, (Bo c )20, 75% (2 steps); (g) HCHO, HC02 H; K2 C 0 3, 75%; (h) CIC02 Me; (i) CAN; (j) CSA, 82% (3 steps); (k) MMPP; CSA, 59%; (I) (BnOCH2 )2 CuLi, TMSCI; (m) Pd(OAc)2, 55% (2 steps); (n) BCI3, (o) Dess-Martin periodinane, 65%, (2 steps); Ans = p-methoxybenzyl.

The B-ring enone moiety was incorporated by oxidation of 8 8 with magnesium monoperoxyperphthalic acid (MMPP) and treatment of the ensuing epoxide with camphorsulfonic 21 acid, resulting in rearrangement and ^-elimination to afford 89 in good yield. The formyl unit at the

3 -carbon was then introduced by addition of lithium dibenzyloxymethyf cuprate, trapping the intermediate enolate as the silyl enol ether, and oxidation with palladium diacetate to provide the

3-substituted enone 90. Deprotection of the benzyl group was accomplished using boron trichloride, and Dess-Martin periodinane oxidation of the primary alcohol completed the preparation of p-formy( enone 91.

Scheme XIX. Winkler's Preparation of Photocyclization Precursor.

NCOzMe 9 6 Bn c,d

Bn 9 7 "O a,b 9 2

9 6 Bn c,d HO-i

9 5

(a) LiHMDS, 93, 53%; (b) NaBH4, 92% (overall); (c) H2, Pd; (d) 96, 54% (2 steps).

The ABCD ring sytem has been prepared by a number of groups. An approach by the

Winkler group brings the eight and six membered rings together using an alkylation (Scheme XIX) 22

(92 + 93 -» 94 + 95), followed by tetracycle formation using 2 + 2 cycloaddition and Mannich chemistry (Scheme XX ) . 2 3 Thus, alkylation of ketone 92 with iodide 93, and reduction of the resulting ketone provided a 4:1 cis:trans mixture of diastereomeric alcohols 94 and 95, respectively. Debenzylation of the separated alcohols and condensation with p-dicarbonyl 96 provided vinylogous amides 97 and 98.

Scheme XX. Winkler's Preparation of Manzamine A Tetracycle.

H H

MeOjC b-d

HO" * * 1

d NCOjMe 9 9

9 7 MeO,C" N>— 7 \- " ° (I b-d H-«^— N

1 00 1 0 2

NCOaMe MeOjC Me0 2C"

HO- HO--

9 8 1 0 3 1 0 2

(a) hv, (b) Et3 N-HCI; (c) 4-DMAP, 50%, 2 :1 ratio of 99:100 (3 steps): (d) Swern (50-60%); (e) NaOMe; (f) hv, -78°C; (g) EtsN-HCI; (h) 4-DMAP, 36% (3 steps). Irradiation of the major (c/'s) alcohol 97 afforded pyrans 99 and 100. Sequential treatment of these enol ethers with triethylamine hydrochloride, 4-{/V,N-dimethylamino)pyridine, and Swern oxidation provided a 2:1 ratio of ketones 101 and 102. The minor stereoisomer 102 contained the relative stereochemistry occurring in manzamine A. Epimerization with sodium methoxide gave the wrong ketone, epimeric at C 3 4 . Subjecting the minor (trans) alcohol 98 to a similar reaction sequence gave tetracycle 103, which upon Swern oxidation also provided dione 102.

Although this approach allows rapid formation of the tetracyclic nucleus, the level of functionality and tow diastereoselectivity could hamper completion of this approach to manzamine A.

Scheme XXI. Condensation Substrates for Martin's Cyclization Precursor.

COCI

TBDPSO

103 104 106 105

Martin has reported the asymmetric synthesis of a highly functionalized tetracyclic intermediate .2 4 This approach constructs the ABC ring system via an intramolecular Diels-Atder reaction and closure of the D ring by olefin metathesis. Enantiopure methyl pyro-D-glutamate

(103) was converted in seven steps to pyrrole derivative 104 (Scheme XXI). Condensation of acid chloride 104 with amine 105 [prepared in four steps from carbamate 106 (Scheme XXI)] provided Diels-Alder precursor 107 in excellent yield (Scheme XXII). Heating compound 107 in toluene provided tricycle 108 stereoselectively in good yield. Incorporation of the azocine ring began with removal of the Boc group and condensation of the liberated amine with 5-hexenoyl chloride to afford amide 109. Deprotection of the TBDPS group was followed by Swern oxidation, and a Wittig reaction provided the second terminal alkene of 110. Treatment of 110 24 with Grubbs' molybdenum catalyst 111 accomplished the metathesis to yield tetracycle 112 in good yield.

Scheme XXII. Martin's Preparation of Manzamine A Tetracycle.

1 0 4

1 0 5 OTBDPS 1 0 7 1 08 OTBDPS

c,d

e-<

OTBDPS

1 1 2 1 1 0 1 0 9

(a) Et3 N, 90%; (b) 160°C, 74%; (c) TMS-I, 65%; (d) CH 2 =CH(CH2 )3 COCI, 70%; (e) HF/Pyridine, 90%;

Nakagawa and co-workers have also reported a synthesis of a manzamine A tetracycle

(Scheme XXIII ) .25 Their approach features an intermolecular cycloaddition to prepare the AB ring system, a conjugate addition of an amide onto an enone for C ring construction, and an amine- activated ester condensation for completion of the D ring. Enone 113 was prepared by thermal cycloaddition of Danishefsky diene 114 with acyl sulfonamide 115. The SEM group of 113 was removed using TFA, and treatment of the resulting trifluoroacetamide with DABCO provided the tricyclic nucleus of 116 in good yield as a 1:1 mixture of diastereomers. Ketalization using 25

ethylene glycol and reductive cleavage of the phenylsulfonamide afforded tricycle 116.

Reduction of the ester moiety and cleavage of the trifluoroacetamide was accomplished by using

a mixture of lithium borohydride and trimethyl borate, resulting in a primary alcohol that was

oxidized to aldehyde 117 with PCC. Wittig reaction of 117 with ylide 118 gave the desired

unsaturated carboxylic acid in an E/Z ratio of 2/5. Esterification of the acid using

pentafluorophenol and DCC gave cyclization precursor 119. The lactam 120 was prepared by

treating 119 sequentially with TFA and 4-DMAP. The stereochemistry of this compound was

proven by x-ray crystallography.

Scheme XXlll. Nakagawa’s Preparation of Manzamine A Tetracycle.

mO M At?ocf’_ctri PhS02'"

1 1 5 PhSO, OTMS

MeOaC SCOCF. OCH3 1 1 3 1 1 6 1 1 4 g-i

120 119 (E/Z =2/5) 1 1 7

(a) p-cymene; (b) CSA, 6 6 % (2 steps); (c) TFA, 77% ; (d) DABCO, 75-85%; (e) (CH 2 0 H)2 , PPTS, 96% (49:47 p-C 0 2 Me:a-CC> 2 Me); (f) Na, anthracene, 83%; (g) UBH 4 , B(OMe) 3 , (h) NaOH, (f- Bo c )20, 87% (2 steps); (i) PCC, 67%; (j) Ph3 P=CH(CH2 )3 COOK (118); (k) C6 F5 OH, DCC, 91% (2 steps); (I) 4-DMAP; (m) DPPA, 58% (2 steps). Pandit and co-workers have made the greatest strides toward the total synthesis of

manzamine A .2 6 As reported in a series of articles, Pandit has incorporated the azocine (D) ring,

the thirteen-membered (E) ring, and the p-carboline unit. However, all these subunits have yet to

be installed in the same molecule. His approach involves an intramolecular Diels-Alder reaction for

preparation of the ABC ring system, olefin metathesis for closing the thirteen membered ring,

classical condensation chemistry for formation of the p-carboline, and azocine lactam

incorporation via an amine-activated ester condensation. Pandit has also reported both racemic

and enantioselective syntheses of the ABC ring system.

Scheme XXIV. Pandit's Preparation of a Racemic Manzamine A Tricycle.

f-BuS

Ph Sf-Bu

f-BuS

1 2 1 1 2 9 1 2 6

(a) NaH, ICH2 CH2 N H C 02Et (123); (b) p-TsOH, A, 58%; (c) LDA, CH2 =NMe 2 l (125), 49%; (d) CH3 I; (e) DBU, 76% (2 steps); (f) AgOTf, DiPEA, 77%; (g) A, PhCH3, 96%.

Scheme XXIV outlines the preparation of racemic tricycle 121. Alkylation and condensation of thioester 122 with iodide 123 afforded dihydropyrrole 124. Condensation of

124 with Eschenmoser's salt (125) followed by A/-methylation and elimination afforded diene 27

126. Condensation of thioester 126 with amine 127 [from benzylamine (128)] gave cyclization

precursor 129 in good yield. Heating 129 in toluene provided tricycle 121 stereoselectively in

excellent yield.

Scheme XXV. Pandit's Preparation of a Homochiral Manzamine A Tricycle.

f-BuS C02Me S' O O a-b X 4 steps 1 I X r '^ N H C B Z ______^ f ^ N H C B Z Sf-Bu OH TBDPSO TBDPSO CBZ 1 3 2 131 1 22 1 33

c-f

COgMfi H | rW*° h-k r Ph Ph f CBZ CBZ TBDPSO CBZ TBDPSO

1 3 0 13 5 134

(a) NaH, 131; (b) p-TsOH, A, 49% (2 steps); (c) TMSOTf, Et3 N, CH2 =NMe 2 l, 85%; (d) Mel; (e) DBU; (f) AgOTf, DIPEA, 126, 65-70%; (g) A, PhCH3, 69% (135), 21% (diastereomer); (h) 0 s 0 4; (i) NaHS03, (j) PTSA, 78% (3 steps); (k) TBAF, 90%.

The preparation of homochiral tricycle 130 is outlined in Scheme XXV and follows the same strategy used to prepare the racemic analog. Optically pure iodide 131 was prepared in four steps from L-(+)-serine derivative 132. Treatment of 131 with the sodium enolate of thioester 122, followed by dehydration, provided enantiopure 133 in modest yield. Cyclization precursor 134 was then prepared and the Diels-Alder reaction provided a 69% yield of tricycle

135 and 2 1 % of a diastereomer (opposite stereochemistry at C 5 , C6 , and Cio). The major isomer (135) was converted to ketone 130 by oxidation with osmium tetroxide, reductive workup with sodium bisulfite, acid catalyzed dehydration with PTSA, and alcohol deprotection with TBAF. A derivative of 130 was prepared, and x-ray crystallography confirmed its structure. Overall, tricycle

130 was prepared in 13 steps and 12% yield from L-(+)-serine.

Scheme XXVI. Pandit's Preparation of ABCE Tetracycle.

CO^Mo c h 2o t b d p s

COjEt COaEt O 121 1 3 7 1 3 8

TBDPSOH-jC CHpOTBDPS CHjOTBDPS

O 1 3 6 141 1 39

(a) LiBH4; (b) TBDPSCI, 70% (2 steps); (c) O s04, then H+, 6 6 %; (d) CH2 =CHCH2 CI, 60%; (e) NaH, 93%; (f) 9-BBN, H 2 O2 ; (g) Dess-Martin periodinane; (h) Ph 3 P=CH2, 48% (3 steps); (i) Li in NH3, then Bn 20; (j) CH2 =CH(CH2)4I (140), KOH, 77% ( 2 steps); (k) (Cy3 P)2 CI2 Ru=CHCH=CPh2 (142).

The Pandit group has also converted racemic tricycle 1 2 1 into pentacycle 136, the first advanced manzamine A intermediate containing the thirteen membered (E) ring (Scheme XXVI).

The carbomethoxy group of 121 was converted to a primary silyl ether by reduction to the alcohol and protection with TBDPSCI. Sequential treatment with osmium tetroxide and acid gave ketone

137. Allylmagnesium chloride added stereoselectivitely to the a-face of 137, and the resulting alcohol was protected as cyclic carbamate 138. The allyl group was homologated to a 3-butenyi unit using a hydroboration/oxidation/Wittig sequence to afford tetracycle 139. The A-ring nitrogen was deprotected using lithium in ammonia and a six-carbon unit was introduced by N- alkylation using potassium hydroxide and iodide 140. Treatment of 141 with metathesis catalyst

142 afforded the macrocycle of 136 in modest yield.

Scheme XXVII. Pandit's Preparation of ABCD Tetracycle with p-carbollne.

COpMe COgM© C02Me H H H

1 30 1 4 3 1 44

1 47 1 46 1 4 5

(a) NaBH3 CN, H+; (b) Dess-Martin periodinane; (c) Ph 3 P=CH(CH2)3COOK (118); (d) HBr, HOAc; (e) Py-BOP; (f) LiBH4; (g) tryptamine. HCI; (h) Pd/C.

Finally, a brief communication by Pandit reported incorporation of the azocine ring and p- carboline unit onto tricycle 130 (Scheme XXVII). Treatment of 130 with sodium cyanoborohydride and acid reduced the enamine while also deprotecting the primary alcohol. Dess-Martin periodinane oxidation afforded aldehyde 143. Wittig reaction of 143 with ylid 118 installed the olefin and acid moities of 144. Lactam cyclization was accomplished by removal of the benzyl carbamate with HBr/HOAc followed by pyridine-BOP to afford tetracycle 145. The p-carboline was constructed by reducing the carbomethoxy group to the corresponding primary alcohol, oxidation to aldehyde 146, and condensation with tryptamine and dry HCI. This sequence yielded the tetrahydro-p-carboline, and dehydrogenation with Pd/C provided heptacycle 147.

Much of the chemistry presented in this chapter was carried out concurrently with the studies to be described in this thesis. The following chapter will address our approach to manzamine A, present previous results from this research group, and describe the improved synthesis of tetracyclic intermediates. CHAPTER II.

PREVIOUS STUDIES AND SYNTHESIS OF ABCD TETRACYCLIC ENONES.

A. Retrosynthetic Analysis.

The retrosynthetic analysis followed in our approach to manzamine A (1) is outlined in

Scheme XXVIII. It was projected that the thirteen membered ring of 1 could be formed via lactam-

ization of a substrate of type 148. Addition of an organometallic reagent of type 149 to an enone

such as 150 was to provide the macrolactamization precursor. It was imagined that enone 150 could be prepared from either enone 151 or 152. For example, a 1,2-addition of 153 to 151 followed by oxidative transposition would give 150. On the other hand, 1.4-addition of 153 to

152 followed by oxidative reintroduction of the double bond would also provide 150. It was felt that tricycle 154 would sen/e as a precursor to both enones (151 and 152) via closure of N 4 (as a nucleophile) onto C 4 (as an electrophile) followed by adjustment of oxidation states. Azocine

154 was to be prepared using an intramolecular displacement (155 -> 154) and 155 was to be prepared from analogously substituted perhydroisoquinoline 156. Finally, 156 was to be prepared from benzoic acid using a new procedure for the preparation of perhydroisoquinolines.

At the time when the author became involved with this project, past studies evolving from this retrosynthetic analysis had resulted in syntheses of enones 151 and 152, but only on very small scales.2 7 As a great deal of chemistry had to be investigated beyond this point, this sequence had to be repeated and carried out on larger amounts of material. Some of the transformations in this series of reactions were problematic, so further studies on those steps were also necessary. As one of the objectives of this research was to incorporate improvements

31 32

Scheme XXVlll. Retrosynthetic Analysis for Manzamine A.

R10 ,C ' Mel 1 4 9

R10,C

1 4 8

A r' or

Met Rz

1 5 2 151 II

OAc OAc OAc A r' * rtS ° >— OCH,

1 54 155 1 56

Ar = p-methoxyphenyl; R1, R2, R 3 = protecting groups; X = leaving group 33

into the established syntheses of these two compounds, the chemistry described in this chapter

is a combination of earlier work developed by Campbell, and improvements on this route. Thus,

improved syntheses of enones 151 and 152 will be described in the remainder of this chapter.

II. Results and Discussion.

The scale-up of the synthesis of perhydroisoquinoline 156 proceeded smoothly as only

operational changes were made and chromatography was kept to a minimum (Scheme XXIX).

Thus, Birch reduction of benzoic acid (157) and alkylation of the intermediate dianion with 2-

bromoethyl methyl ether (158) provided dihydrobenzoic acid 159 in high yield .2 0 Treatment of crude 159 with diphenylphosphoryl azide and pyrrolidine followed by iodolactonization of crude

160 afforded lactone 161 in good yield after crystallization .2 7 Keck allylation of 161 using allyltri-

n-butyltin (163) gave lactone 162 in good yield .2 9 Chromatography was required at this point to remove unreacted stannane 163, but this separation was trivial as 163 could be quickly eluted from the column. Lactone 162 was opened with p-anisidinylmagnesium bromide (164) to cleanly provide amide 165. The liberated alcohol (165) was then protected as the corresponding acetate to afford crude 166 in excellent yield .2 7 Johnson-Lemieux oxidation of the terminal vinyl group of amide 166 gave a mixture of three compounds, aldehyde 167 and isomeric carbinol lactams 168.30 As all components of this mixture could be converted to perhydroisoquinoline

156, no separation was performed at this stage. Instead, conversion of 167 and 168 to 169 was accomplished by heating the mixture with methanol and acidic Dowex resin. Reduction of

169 was carried out using sodium cyanoborohydride and trifluoroacetic acid to afford c/'s-fused perhydroisoquinoline 156 in good yield .2 7 Chromatography was required at this stage since

156 was a low melting solid that could not be crystallized from the reaction mixture. Through these first nine steps of the synthesis, 51 g of benzoic acid could be converted into 28 g of perhydroisoquinoline 156 in 24% overall yield. In addition, chromatography was needed for only two separations and crude reaction mixtures could be carried on in several instances. 34

Scheme XXIX. Preparation of C/s-Octahydroisoquinollne 156.

j O H O , C - P HOaC ' i o c h 3 1 5 7 1 59

II \ \ o N | 1 o c h 3 OCH3 OCHa

^ 1 6 5 R = H 1 6 2 1 6 1

' C „ . . c o * v

CHO OAc OAc OAc RO.

Ar*

OCH3 OCH3 OCH3

1 6 7 / 168 R = H 1 5 6

hCx 169 R = Me

(a) Li. NH3; then CH 3 OCH2 CH2Br (158). 92%; (b) pyrrolidine.DPPA, Et 3 N, DMF, 95%; (c) l2, THF-H2 0 , 69%; (d) CH 2 =CHCH2 SnBu3 (163), AIBN, PhH, A, 63%; (e) p-MeO-CeH^NHMgBr (164), 94%; (f) Ac 2 0 , Et3 N, 4-DMAP, 99%; (g) OSO 4 . Nal04, f-BuOH-THF-H2 0 ; (h) Dowex H+, CH3 OH-THF; (i) NaBH3 CN, TFA, CH2 CI2. 67% (three steps).

The next task was conversion of 156 into azocine 154. This began with the two step conversion of methyl ether 156 into aldehyde 170, which was accomplished using the same 35

reagents as previously reported, but the purifications were simplified (Scheme XXX). Treatment of

156 with boron tribromide provided a thick oil as the crude reaction mixture, from which the majority of alcohol 171 could be crystallized. Chromatography of the mother liquor provided additional 171. Swern oxidation converted alcohol 171 to aldehyde 170. Most of the product could be obtained cleanly by crystallization of the crude mixture and additional 170 was obtained by chromatography of the residue.

Scheme XXX. Two Step Conversion of 156 -» 170.

OAc OAc OAc

CHO OCH, OH

156 171 170

(a) BBr3, CH2 Cl2 , -78°C -> -15°C, 69%; (b) (COCI) 2 , DMSO, CH2 CI2, -78°C, then EI 3 N, 8 6 %.

Scheme XXXI. Addition of a Lithium Acetylide to Aldehyde 170.

OAc OAc OAc Lj ^ — (CH2)4OTHP

-78 C

0 Y ifT I— OTHP Ilf r— OTHP O

170 173 174

The next task involved attaching the six-carbon unit necessary for incorporation of the azocine ring. This transformation presented two major challenges. First, we wished to form this carbon-carbon bond in high yield, as this step is a key bond construction in the synthesis. 36

Second, we wanted to achieve high diastereoselectivity in this bond construction. Previous studies had accomplished this step by adding acetylide 172 to aldehyde 170, giving a 60% yield of a mixture of two diastereomeric propargyl alcohols 173 and 174 in a 2.5:1 ratio, respectively

(Scheme XXXI). However, problems arose in reproducing the chemical yield and in separating the alcohol products Irom unconverted starting aldehyde 170. For example, the separation of 170 from 173 and 174 could only be accomplished using MPLC on scales of one gram or less. Thus, the incomplete conversion of 170 to 173 and 174, presumably due to enolization of 170, severely limited scale-up at this stage of the synthesis. Furthermore, the enolization problem was not alleviated by use of organocerium reagents or appropriate vinyllithium reagents.

Scheme XXXII. Kishl’s Use of CrCl 2 -NlCl2 for the Halichondrins.

(CH2)4 CH3

(CHz)4Me

176 R = p-anisylidinyl 1 7 7

Use of an admixture of chromium (II) chloride and nickel (II) chloride to mediate the coupling of alkenyl or alkynyl halides to aldehydes appeared to offer an attractive solution to this problem. The reaction conditions are reported to be mild and very selective for additions to aldehydes over ketones, esters, or amides. Such selectivity has been chronicled in methodology studies by Hiyama ,31 Takai,3 2 and Oshima 3 3 and the mechanism has been detailed by Takai .34

The work described in these papers have been applied to complex synthetic problems, as demonstrated by the Kishi group in the synthesis of palytoxin ,3 5 (+)-ophiobolin C ,3 6 the halichondrins,3 7 and taxane natural products .38 For example in halichondrin studies, iodoalkyne 37

175 was coupled with aldehyde 176 to give propargyl alcohol 177 (Scheme XXXII) and

iodoalkene 178 was coupled with aldehyde 179 to provide allylic alcohol 180 in a good yield

during studies directed toward palytoxin (Scheme XXXIII).

Scheme XXXIII. Kishi’s Use of CrCI 2 -N IC l2 for Palytoxin.

I OBn

OBn OBn OBn BnO ,OBn / BnO I 1 7 8 OBn OH OBn ^ ' 6 :'

OBn 1 7 9

Scheme XXXIV. Preparation of lodoalkyne 181 and Iodoalkene 182.

ex. ■OH

1 8 4 1 85 1 8 3

-OTHP e oTHp d -OTHP

1 8 2 181 1 8 6

(a) SOCI2> pyridine, 60%; (b) UNH 2 NH3, Fe(N0 3 )3 «9H20 , 84%; (c) DHP, p-TsOH, 95%; (d) n- BuLi, THF, -78°C, then l2, 8 6 %; (e) K0 2 C-N=N-C02 K, AcOH, 44%. 38

The conversion of 170 to 173 and 174, or the alkenes derived therefrom, required the

preparation of iodoalkyne 181 and iodoalkene 182 (Scheme XXXIV). These iodides were

prepared from 5-hexyn-1-ol (183).39 Thus, 2-tetrahydropyranyl methanol (184) was converted to chloride 185 using thionyl chloride, and a double elimination afforded alcohol 183 in high yield. The hydroxyl group was protected as a tetrahydropyranyl ether (186). Deprotonation of

186 with n-butyllithium and treatment of the resulting acetylide with iodine gave 181. Reduction of alkyne 181 diimide prepared in situ from dipotassium azodicarboxylate and acetic acid gave cis- vinyl iodide 182.40

Scheme XXXV. CrCl 2 -N IC l2 Mediated Couplings with 170.

OAc

-(CH2)4OTHP

173 174

173 :174 = 5 : 4 (90%) OAc

H2> Pd on BaS0 4 , Pyr., 1 atm, rt, 98%

1 7 0 OAc OAc

/ \ c h 2j4oth p

182, CrCI2, NiCI2, O ' S — OH O ' N> — OH THF, rt, 48 h OTHP ( j -

1 8 7 1 8 8

187 :188 = 60 : 40 (80%) Although DUF and DMSO have been more widely reported as the solvent in these

coupling reactions, THF was used in the experiments described herein. Treatment of a slurry of

chromium (II) chloride and catalytic nickel (II) chloride in THF with a solution of iodoalkyne 181 and

aldehyde 170 yielded a 5:4 mixture of inseparable diastereomeric alcohols 173 and 174,

respectively, in 90% yield (Scheme XXXV). Unfortunately, the diastereoselectivity was low, so

addition of vinyl iodide 182 was attempted in hopes of improving this ratio. Although the

selectivity was slightly higher ( 1 .6 :1 ), the chemical yield was lower (80%) and reaction time

considerably longer. Thus, for practical purposes, it was decided to proceed via the iodoalkyne

coupling. The mixture of propargylic alcohols 173 and 174 were converted to alcohols 187 and

188 by catalytic hydrogenation at one atmosphere in near quantitative yield, and the two

diastereomers were separated at this stage. Although the separation required a chromatography,

70% of 10 grams of the mixture could be separated over 800 grams of silica gel in one working

day.

Scheme XXXVI. Preparation of Acylsulfonamide 189.

OAc OAc

SESNHf-Boc (6 6 ), Ph3P, DEAD

THF, rt, 93% OTHP OTHP

1 8 7 1 89

The next stage of the synthesis called for conversion of the allylic alcohol to a suitably protected amine. This had been accomplished prior to the outset of this research as outlined in

Scheme XXXVI. Thus treatment of 187 with amine equivalent 6 6 in THF with triphenylphosphine and diethyl azodicarboxylate provided acylsulfonamide 189 in high yield. The reader 40

should note that this reaction occurred with retention of configuration at the allylic center. As now

will be seen, this created problems with attempts to incorporate 188 in the synthetic scheme.

In summary, although we had achieved a high chemical yield in adding the six-carbon unit,

the problem of low diastereoslectivity had not been solved. This resulted in attempts to invert the

stereochemistry at the carbinol center in alcohol 188. We envisioned two potential strategies for

solving this problem: (1 ) inversion of configuration at the carbinol center by substitution

chemistry, or (2) oxidation of alcohols 173 and 174 followed by reduction with a hydride source,

hoping that C-H bond formation would proceed with higher diastereoselectivity than had C-C

bond formation.

Scheme XXXVII. Explanation for Retention of Configuration.

OAc OAc OAc

Ar' O OR OTHP OTHP (_ /° THP

190 R = CHO 1 8 7 191 c187 R = H

(a) HCOOH, Ph3 P, DEAD; (b) LiOH*H2 0.

The inversion chemistry was first investigated. Previous studies from our group had shown that treatment of allylic alcohol 187 with formic acid under Mitsunobu conditions yielded ester 190 that, upon hydrolysis, gave the starting alcohol 187 with retention of configuration at the carbinol center .2 7 The process outlined in Scheme XXXVII accounts lor this observation.

Once the hydroxyl is converted into a good leaving group, an intramolecular displacement occurs to form cyclic iminium ether intermediate 191. This step accomplishes the first inversion. The external nucleophile then attacks the original carbinol center in a second inversion step, 41

regenerating the starting alcohol. This reasoning also explains the retention of configuration

seen in Scheme XXXVI for preparation of acylsulfonamide 189.

Our strategy was devised to take advantage of the formation of intermediate 191

(Scheme XXXVIII). This imidate has two potential hydrolysis sites. Hydrolysis at the original

carbinol center would result in recovered starting material (pathway a). However, hydrolysis at the

original carbonyl carbon (pathway b) would require the carbonyl oxygen to become the alcohol

oxygen, and only one inversion would take place. Thus, our plan was to convert the alcohol to a

good leaving group, followed by intramolecular displacement and hydrolysis at the imino ether

carbon to provide the inverted alcohol.

Scheme XXXVlii. Strategy to Invert Allylic Alcohol 174.

OAc OAc

OTHPOTHP

1 88 3 X == leavingleavingX group

OAc

OTHP OTHP

1 91 1 8 7

This premise was proven reasonable, as treament of alcohol 188 with mesyl chloride followed by hydrolysis with saturated aqueous sodium bicarbonate provided a 25% yield of alcohol 187 (Scheme XXXIX). Unfortunately, the reaction was not clean, the mass balance was 42

low, and no starting alcohol was recovered. Potential loss of mass could result from Sn 2' reactions

of 191 or decomposition of the intermediate mesylate. However, no by-products from either of

these alternative pathways were observed. Although the results from these studies seemed

somewhat encouraging, the complex reaction mixtures and low chemicals yield caused this

approach to be abandoned.

Scheme XXXIX. Inversion of Allylic Alcohol 188.

OAc OAc

XP MsCI. ElgN ^ A(. Q P

Vv_r'!u 2. NaHCOs, q OTHP H2Q (7 |— OTHP & 25%

1 8 8 1 8 7

Scheme XL. Thompson's Inversion of a Benzyllc Alcohol with Azlde.

91% 1 8 9

1 92 1 9 3 1 9 4

(a) DPPA, DBU, PhMe

Ultimately the hydroxyl group had to be replaced by a nitrogen. Our next set of inversion studies used a nitrogen nucleophile to potentially accomplish the inversion. Thompson and co workers reported inversion of optically active benzylic alcohols could be accomplished without racemization by treating those alcohols with diphenylphosphoryl azide and 1 ,8 - diazabicyclo[5.4.0]undec-7-ene (DBU) (Scheme XL ) . 41 For example, this process converted alcohol 192 into the corresponding phosphate 193, which was in turn displaced by azide to yield the inverted product 194. These conditions appeared attractive, as azide would hopefully be highly competitive for displacement of the leaving group before neighboring group participation by the amide.

Scheme XLI. Proof of Stereochemistry for Attempted Inversion.

OAc OAc OAc

OTHP OTHP OTHP

1 8 8 1 9 5 1 9 8

OAc OAc OAc

OTHP OTHP OTHP

1 89 1 9 6 1 9 7

(a) DPPA, DBU, PhCH3 , 69%; (b) Ph 3 P, THF-H2 0, 65%; (c) N3 CC>2 f-Bu, DMSO, 33%; (d) TBAF, THF, 83%.

Treatment of alcohol 188 under Thompson's conditions did provide an azide in 69% yield. However, the structure was eventually shown to be 195, the product of a double inversion.

The structure of 195 was established by the chemical correlation outlined in Scheme XLI. Since the stereochemistry of the acylsulfonamide 189 was already clearly established, correlation of

195 with a derivative of this material seemed logical. It seemed that f-butyl carbamate 196 would be easily accessible from acyl sulfonamide 189 by deprotection, and that azide 195 could be converted into its corresponding carbamate 197 via a two step sequence. In the event, azide

195 was reduced to primary amine 198 upon treament with triphenylphosphine in wet THF .4 2

Acylation of 198 with f-butyl azidoformate 4 3 in DMSO gave f-butylcarbamate 197.44 Acyl sulfonamide 189 was converted to f-butyl carbamate 196 by removal of the [J-trimethylsilyl- ethanesulfonyl group with tetrabutylammonium fluoride ,4 5 and the 1H NMR spectra were compared. This analysis quickly revealed that we had two different stereoisomers, and that retention of configuration had again occurred at the carbinol center. The azide had not been competitive with the intramolecular reaction, and again the time had come to change approaches.

Scheme XLII. Wlpf's Inversion of Hydroxyl via Oxazoline Intermediate.

2 0 3 2 0 2

(a) CH302CNS02(C2H5)3 (199), THF, 81%; (b) 0.3 N HCI-THF; (c) K2C 03, pH 9.5, 77%.

Our results indicated that the lactam had to be modified if we wanted to accomplished the inversion. Thus, our nexl strategy involved removal of the p-methoxyphenyl group from the lactam nitrogen followed by attempts to conduct the inversion on the resulting secondary lactam.

One report that gave us hope that the secondary lactam might be useful is shown in Scheme XLII.

Wipf and Miller used Burgess' reagent (199) and subsequent hydrolysis of the resulting oxazoline to successfully invert hydroxyl stereochemistry of peptides .4 6 For example, secondary amide 200 was treated with Burgess reagent, (methoxycarbonyl)sulfamyl triethylammonium hydroxide, inner salt (199) to provide cyclic iminoether 201. Sequential treatment with hydrochloric acid and potassium carbonate resulted in clean hydrolysis at the original carbonyl carbon to afford inversion product 202. Alcohol 202 could then be converted back into peptide

200 via oxazoline 203 using the same reaction sequence.

Scheme XLIfl. Manipulations of Secondary Amide 204.

OAc OAc OAc

Ar' .Qi> o ^ o ^ 0 ^ OCHj OH OCH, 1 5 6 2 0 4 2 0 8

c '^ d ______|

OAc

f-Boc<

o c h 3 2 0 9

(a) CAN, H2 O-CH3 CN, 0°C, 54%; (b) BBr3, CH2 CI2, 0°C, 208 in 48% + 204 in 35%; (c) f-Boc) 2 0, Et3 N, 67%; (d) BBr3, CH2 CI2, 0°C, 204 in 70%, 208 in 20%.

Our studies began with removal of the p-methoxyphenyl group. Ceric ammonium nitrate

(CAN) is reputed to efficiently remove p-methoxyphenyl groups from p-lactams .4 7 Unfortunately, 46 treatment of allylic alcohol 188 with CAN resulted in a complex mixture from which the desired secondary amide appeared to be a minor component. The presence of the allylic alcohol and THP group in the side chain appeared to be complicating matters, so deprotection of perhydroisoquinoline 156 was investigated, the hope being that it could be elaborated to a useful substrate. Treatment of 156 with CAN at 0°C in aqueous acetonitrile provided secondary amide 204 in 54% yield (Scheme XLIII). We had planned convert 204 to allylic alcohol 205, use

Wipf’s procedure to provide imino ether 206, and then hydrolyze 206 to the desired allylic alcohol 207 (Scheme XLIV). Unfortunately, problems were encountered. Deprotection of methyl ether 204 with boron tribromide afforded a 48% yield of the desired alcohol 208 and 35% of recovered 204 (Scheme XLIII). Swern oxidation of alcohol 208, however, failed to provide the desired aldehyde. We surmised that the secondary amide may have adversely effected the oxidation and next decided to protect the amide with an electron-withdrawing group to decrease the chance of neighboring group participation. Thus, treatment of lactam 204 with di-terf-butyl dicarbonate provided a 67% yield of W-acyl carbamate 209.48 Unfortunately, the f-Boc group was removed faster than the methyl ether upon treatment of 209 with boron tribromide. Ether 204 was the major product and alcohol 208 was a minor product. As this approach seemed to have problems on a number of fronts, it was also abandoned.

Scheme XLIV. Plan for Inversion via Imino Ether.

OAc OAc OAc

OTHP OTHP OTHP

205 206 207 47

Scheme XLV. Reduction of Propargyl Ketone 210.

OAc OAc

Mn0 2 Ar' CH2CI2, v P O " S — OH 97% OTHP OTHP t J 173 + 174 2 1 0

Reducing Agents i OAc OAc

O OH O OH O' OTHP O" OTHP 1 7 4 1 7 3

Table 1. Product Ratio from Reduction of Ketone 210.

Reducing Agent Ratio 173:174 NaBH4> CeCt3, CH 3 OH, 0°C .4 9 2 :1 NaBH4l CeCIa, CH 3 OH, -60°C. 3 :1 DIBAL-H, PhMe, -78°C complex mixture LiBH(CH{CH3 )CH2 CH3) 2:1 Zn(BH4 )2 , various temperatures 5 0 No reaction NaBH4, CeCI3l EtOH, -78°C. 1.5:1 LiAIH(01Bu)3l THF, 0°C to rt complex mixture

Our attention was then turned towards an oxidation/reduction strategy. Alcohols 173 and 174 were converted to ynone 210 with activiated manganese dioxide in high yield ,51 and a variety of reducing agents were surveyed. These results are summarized in Scheme XLV and

Table 1. Although none ot the conditions gave excellent diastereoselection, the reduction with sodium borohydride and cerium trichloride provided a 3:1 ratio of 173:174. If propargyl alcohols 48

173 and 174 had been separable, this sequence would have allowed for greater than 90%

conversion of alcohols 173 and 174 to alcohol 173 via reduction, separation, reoxidation, and

reduction. Unfortunately, 173 and 174 were not separable, and allylic alcohols 187 and 188

were not amenable to this reaction sequence. Nevertheless, the oxidation/reduction sequence

described here allows the highest throughput of material as the chemical yields are excellent

(95% overall) and the diastereoselectivity is the highest yet seen for this system.

Despite the increased conversion of alcohol 174 to alcohol 173 and attempted

inversions of 188, we still wished to devise a sequence to better make use of alcohol 188. This

goal was achieved by degradation of 188 to aldehyde 170 (Scheme XLVI). Thus, epoxidation of

188 with vanadyl acetoacetonate and f-butyl hydroperoxide 5 2 afforded epoxyalcohol 2 1 1 as

one diastereomer, and oxidative cleavage with periodic acid 52 gave perhydroisoquinoline 170 in

a 50% yield for the two-step sequence.

Scheme XLVI. Recycling of Alcohol 188.

OAc OAc OAc

> — OH OTHP

188 211 170

(a) VO(acac)2, f-BuOOH, PhH, A, 70%; (b) H5 I06, THF, 71%.

Returning to the mainline synthesis, acylsulfonamide 189 had been prepared in high yield from alcohol 187. As the nitrogen for the azocine ring was now in place, the next task involved construction of the eight-membered ring as shown in Scheme XLVII. This required removal of the deprotecting groups and cyclization. 49

Scheme XLVII. Preparation ot Homoallyllc Alcohol 215.

OAc OAc

a-b

f-Boc

OTHP SES SES OH

1 8 9 2 1 2

OAc

OTs SES

214 R = Ac 2 1 3

215 R = H

(a) p-TsOH, MeOH, THF, rt. 48 h; (b) DMSO, A, 25 min, 70% (two steps); (c) p-TsCI, 4-DMAP, Et3 N, CH2 CI2, 95%; (d) KH, 18-C-6, Bu 4 N+|-, PhCH3, A; (e) LiOH*H 2 0, THF-MeOH-H 2 0 , 80% (two steps).

Deprotection of the THP and f-Boc groups had been previously achieved using

trimethyisilyl iodide, prepared in situ from trimethylsifyl chloride and sodium iodide ,5 4 but this

procedure gave irreproducible results. Instead, a two-step sequence was used, involving initial

removal of the THP group with catalytic p-toluenesulfonic acid in methanol and subsequent thermal cleavage of the f-Boc group in DMSO. This sequence provided hydroxysulfonamide 212 in 70% yield. The azocine ring was then closed without complication by the established

sequence. Sulfonamide 212 was converted to 213 using tosyl chloride, 4-DMAP, and triethylamine. Subjecting tosylate 213 to potassium hydride in the presence of l 8 -crown-6 and tetra-n-butylammonium iodide afforded azocine 214 in high yield. Hydrolysis of the acetate using 50 lithium hydroxide monohydrate provided homoallylic alcohol 215 in 80% overall yield from 212.

The relative stereochemistry of 215 was proven previously by x-ray crystallography .2 7

Recalling the retrosynthetic analysis outlined in Scheme XXVIII, homoallylic alcohol 215 is an intermediate with all the features of 154, a projected precursor of either tetracyclic enone

151 or 152. Therefore, the five-membered ring of the ABCD tetracyclic core was next to be completed. This was to be accomplished by attack of nucleophilic N 4 onto an electrophilic C 4 .

Earlier work by Campbell and Hart investigated electrophilic activation of the C 3 -C4 double bond of an amine derived from 215. These studies were abandoned as only complex mixtures were obtained when selenium, mercury, iodine, and bromine electrophiles were used, and attention was turned to the preparation of a C 3 -C4 epoxide. This required selective epoxidization of the C 3 -

C4 double bond in the presence of the azocine double bond. Furthermore, as the C ring would be closed by a nucleophilic displacement, the a-face nitrogen would need to open a p-face epoxide. These two requirements made an epoxidation directed by the hydroxyl group at C 1 ideal (Scheme XLVIII). Previous studies had used non-anhydrous f-butylhydroperoxide and catalytic vanadyl acetoacetonate for the directed epoxidation. These conditions gave marginal results, as only a 60% conversion to the desired epoxide 216 was obtained after initial reaction, separation, and recycling of unreacted starling material.

Scheme XLVlll. Epoxidation of Alcohol 215.

rsH1 OH OH

2 1 5 2 1 6 2 1 7

(a) VO(acac)2 , f-BuOOH, CICH2 CH2 CI-CH2 Cl2 -PhH; 216, 60%; (b) Mo(CO)6 , r-BuOOH, PhH, A; 216, 85%, 217, 10%. 51

Scheme XLIX. Completion of Enone 151.

OH OMOM OMOM

SES SES

2 1 6 2 1 8 2 1 9

OH OMOM

OAc OAc

151 221 220

(a) MOMCI, /-Pr2 NEtp 8 8 %; (b) CsF, DMF, A, 8 6 %; (c) Ac2 0 , /-Pr2 NEt, 90%; (d) LiBF4, CH 3 CN, H ^ , 71%; (e) (COCI)2l DMSO, -78°C, then Et 3 N, then basic alumina, 83%.

A more effiecient conversion of 215 -> 216 was desired, so other peroxide sources and catalysts were sought. A procedure by Sharpless was followed for reliable preparation of stable solutions of f-butylhydroperoxide in toluene 5 5 instead of using commercial 90% t- butylhydroperoxide dried over magnesium sulfate. The use ol this anhydrous peroxide and vanadyl acetoacetonate did not improve results, so attention was turned to other catalysts.

Examples throughout the literature have detailed the use of molydenum hexacarbonyl for catalysis of directed epoxidations with allylic alcohols and homoallylic esters or alcohols .5 2 This catalyst gave markedly improved results for epoxidation of alcohol 215. Heating 215 to reflux in benzene with anhydrous f-butylhydroperoxide and catalytic molydenum hexacarbonyl afforded 52 epoxyalcohol 216 in 85% yield. Overoxidation was a minor problem, as a 10% yield ot bisepoxide

217 was also observed.

The stage was set for closing the C ring (Scheme XLIX), but protection of the hydroxyl group was first necessary as the cyclization would liberate a secondary alcohol in the product. As described by Hart and Campbell, alcohol 216 was treated with chloromethyl methyl ether and diisopropylethylamine to provide alcohol 218. Heating 218 with cesium fluoride in DMF deprotected the D-ring nitrogen, and epoxide opening occurred to afford tetracycle 214 in high yield. Enone 151 was then prepared by a protection/deprotection sequence followed by oxidation/elimination chemistry. Thus, alcohol 219 was converted into tetracycle 220 in 85% yield upon treatment with acetic anhydride and diisopropylethylamine. Deprotection of the methoxymethyl acetal was capricious, as treatment with trimethylsilyl iodide afforded alcohol 2 2 1 in variable yields. Reproducibly good yields were obtained when lithium tetrafluoroborate was used for the deprotection .5 6 This reagent provided alcohol 221 in 71% yield along with 9% of recovered 220. Swern oxidation of 221 followed by chromatography over basic alumina afforded enone 151 in 83% yield. The structure of 151 was proven by x-ray crystallography 02A (Figure 5).

C6A C4A C3A< C5A C7A

N IA l CIA C8A C9 •— N2A C17A t O C 2 3 A C 2 2 A 01A CIO'C16J C 2 4 A C21A Cll A C15 0 3 A Cf ' t ! 3 A

Figure 5. X-Ray Structure of Enone 151. Scheme L. Preparations of Enone 152.

OH OMOM OMOM

SES

2 1 6 2 1 9 222

OPiv OPiv

Ar' OH

SES

2 2 7 2 2 8 1 5 2

(a) MOMCI, r-Pr2 NEt, 8 6 %, (b) CsF, DMF, A, 81% for 219, 8 8 % for 228; (c) DMSO, then Et 3 N, then basic alumina, 65%.

Transposed enone 152 had also been prepared previously, but in low yield .2 7 Swern oxidation of alcohol 219 gave (3-alkoxyketone 222 in high yield, but elimination of the 13- substituent provided only 34% of enone 152 on a small scale (Scheme L). As a carboxylate was

eliminated cleanly to provide enone 151 in high yield, such a strategy seemed reasonable for enone 152. Although protecting the C-| hydroxyl as an acetate may have seemed an obvious choice, earlier results from Campbell and Hart had shown that the acetyl group of 223 was prone to migration or cleavage upon treatment wtih cesium fluoride to give a mixture of alcohols 224-

226 (Scheme LI). 2 7 We surmised that a bulkier ester would be less susceptible to such side reactions. Conversion of alcohol 216 into the corresponding pivaloate 227 was achieved in high yield with pivaloyl chloride in pyridine .5 7 Healing 227 with cesium fluoride in DMF gave ABCD 54 tetracycle 228 in 8 8 % yield and Swern oxidation followed by chromatography over basic alumina provided enone 152 in 65% yield.

Scheme LI. Labile Acetate of ABCD Tetracycle.

OAc OR1 OH

° A— N o \_ kj o \ _ N

2 2 3 224 R1 = Ac, R2 = H 2 2 6 225 R1 =H, R 2 = Ac

To summarize, tetracyclic enone 151 was prepared in overall 1.4% yield for 26 steps, enone 152 was prepared in 2.0% yield for 24 steps, and we were pleased to have reasonable quantities and of these enones in hand. Although the syntheses were lengthy, 151 and 152 had levels of functionality that allowed us to begin exploring introduction of the p-carboline unit and thirteen membered ring. CHAPTER III.

STUDIES FOR INCORPORATION OF p-CARBOLINE

UNIT OF MANZAMINE A.

A. Background

p-Carbolines are found in many natural products apart from the manzamine family of

alkaloids. Other selected p-carboline natural products are presented in Figure 6 and include

harman derivatives 226 and 227,58 the eudistomins (228-231 ) , 5 9 lavendamycin (232)60 and

its synthetic esters (233-234),'61 -6 2 pyridindolol (235),63 and alkaloids found in the leaves and

bark of Picrasma javanica B! 6 4 These compounds exhibit a wide range of biological activities,

including antitumor and antiviral properties, making methodologies for their preparation the

objective of many synthetic studies. As a prelude to studies directed toward the synthesis of the p-carboline substructure of manzamine A, general strategies for the syntheses of p-carbolines will be described in this section.

The most classical methods of p-carboline construction are the Bischler-Napieralski reaction 6 5 and the Pictet-Spengler reaction .6 6 The former method is represented in Scheme Lll.

For example, heating N-tormyltryptophan (236) with phosphorous oxychloride and polyphosphoric acid provides norharman (237) via cyclodehydration, decarboxylation, and dehydrogenation. Dihydro-p-carboline 238 can be prepared using tryptamine derivative 239 instead of a tryptophan amide.

55 56

X p O d c cN o I CH3 H

226 6 -Hydroxyharman 227 6 -Hydroxynorharman 228 Eudistomin J

h 2n N— H

229 Eudistomin H 230 Eudistomin K 231 Eudistomin M

CH,OH

H— N' OH OH

R = H, Lavendamycin 233 R = CH3> Lavendamycin methyl ester 234 R = CH2CH3l Lavendamycin ethyl ester 235 Pyridindolol

Figure 6. Selected p-Carbollne Natural Products.

The Kende group has used this popular method of p-carboline formation in their preparation of lavendamycin methyl ester (Scheme LIU ) .61 Amide 240 was cyclodehydrated and subsequently dehydrogenated by heating with polyphosphate esters to afford substituted heterocycle 241. 57

Scheme Lll. BIschler-Napleralski Reactions.

poci3 111 PPA, A 'N' 36% I H 236 2 3 7

POCI3, A

90%

Ph Ph 239 238

Scheme Llll. Kende's BIschler-Napleralski for |3-Carbollne Preparation.

c o 2c h3 COzCH3 PPE, A. 40%

H—N H—N

2 4 0 2 4 1

The Pictet-Spengler cyclization involves reaction of a tryptamine derivative with an aldehyde in the presence of acid to provide a tetrahydro-p-carboline. Scheme LIV shows the reaction of tryptamine with acetaldehyde and sulfuric acid to provide tricycle 242. Tryptophan and formaldehyde react under physiological conditions (pH 6.5) to afford amino acid 243 in high yield. Rinehart and co-workers have used standard protic Pictet-Spengler chemistry to condense tryptamine with aldehyde 244 en route to tetrahydro-p-carboline 245, an intermediate in syntheses of various members of the eudistomin family (Scheme LV)59. The Cook group has developed aprotic Pictet-Spengler methodology, to be discussed in more detail later within this chapter, in their synthesis of pyridindotol (235)63. Heating tryptophan methyl ester 246 with aldehyde 247 in benzene provided a high yield of tetrahydro-p-carboline 248 (Scheme LVI).

Scheme Liv. Protic Pictet-Spengler Reactions.

CH3CHO

H H CH,

2 4 2

HCHO, pH 6.5

H H 2 4 3

Scheme LV. Rinehart's Use of Pictet-Spengler for Eudistomin Alkaloids.

CHO COOH

H Eudistomins H H COaH

2 4 5

Queguiner and co-workers have prepared 6 -hydroxyharman (226) and 6 -hydroxynorharman (227) by bringing substituted pyridine and phenyl rings together by a Suzuki coupling, followed by a cyclization to complete the tricycle (Scheme LVIf ) . 5 0 -6 7 Boronic acid 249 and iodopyridine 250 were coupled with tetrakis(triphenylphosphine)palladium( 0 ) to afford biaryl 59

251. Heating with pyridinium chloride affected five-membered ring closure to afford 227 in high

yield.

Scheme LVI. Cook’s Use of Pictet-Spengler for Pyridindolol-

CHO

CO,CH. COjCHa tx

i n NH, 2 4 7 ‘N ■ *- 2 3 5 I H PhH, A, 90% 2 4 6 2 4 8

Scheme LVII. Quegulner's p-Carboline Synthesis.

CHoO. 250, a F NHCOtBu NHCOtBu

249 251 227

(a) Pd(PPh3 )4 , K2 CO3 , A, 73%; (b) pyridinium chloride, A, then NH 4 OH, 89%

Boger 6 8 Molina,62 and Ciufolini6 9 have used less traditional methods for building the

substituted p-carboline units appearing in lavendamycin esters (233 -> 234). Boger applied an

inverse electron-demand Diels-Alder followed by a palladium-mediated coupling to prepare a p- carboline unit (Scheme LVIII). Thus, a cycloaddition between enamine 252 and triazine 253 provided biaryl 254. Selective conversion of one carbethoxy group to an amine afforded 255, and intramolecular coupling of the amine and the aryl bromide with tetrakis(lriphenyl)phosphine- pafladium(O) gave p-carboline 256 in high yield. 60

Scheme LVlll. Boger's (3-Carboline Synthesis.

ElO,C C02Et CH E t O ^ ^ N ^ C O aEt A. 50% ✓ N EtOjC Brvs[ 1 1 + EtOaC Jl ^ N*

2 5 2 2 5 3 2 5 4 several steps

c o 2c h . HZN c o 2ei

h— n; h— n ; CH.

2 3 3 2 5 6 2 5 5

Scheme LIX. Molina's p-Carbollne Synthesis.

CH. COjEl CO,Et Bu3 P, 100% N=PBu.

2 5 7 2 5 8

1 0% Pd/C, 45% CHO OBn 2 5 9

OBn H-N'

2 3 4 260 61

Molina used a tandem aza Wittig/electrocyclic ring closure for incorporating the p-carboline

unit in lavendamycin ethyl ester (Scheme LIX). Treatment of indolyl azide 257 with tri-n-

butylphosphine provided intermediate iminophosphorane 258 in quantitative yield, which upon treament with aldehyde 259 gave the targeted intermediate 260.

Ciufolini constructed the p-carboline moiety using a modified Knoevenagel-Stobbe reaction to construct the pyridine and subsequent thermolytic nitrene insertion to complete p- carboline formation (Scheme LX). Ketal 261 was prepared via a hetero Diels Alder reaction, and treatment of 261 with hydroxylamine hydrochloride provided pyridine 262. After oxidation of the benzylic methyl into an aldehyde, heating resulted in nitrene formation and insertion into the pyridine ring to provide the p-carboline 263 in high yield.

Scheme LX. Clufollnl's p-Carbollne Synthesis.

Me OEt CH

OBn OBn

89%

261 2 6 2

(a) Se02, 87% (b) A, 83%.

CHO

OBn

h— n;

2 3 3 2 6 3 62

B. Results and Discussion

Our plan was to incorporate the p-carboline unit in one piece as outlined in Scheme LXI. It was hoped that 1 -lithio-p-carboline 264 would undergo a 1.2-addition to enone 151 to provide tertiary ally lie alcohol 265. An oxidative rearrangement was to provide enone 150 which would then serve as a substrate for incorporation of the thirteen membered ring. Before attempting any studies on the tetracyclic system, we felt that three basic questions needed to be addressed: ( 1 )

What was the best method by which to accomplish the 1,3 oxygen transposition? (2) could a 1- lithio-p-carboline species be prepared? and (3) how efficiently would this carbanion add to enones?

Scheme LXI. Plan for Addition of p-carbollne.

Li R 2 6 4

151 2 6 5 1 50

The first goal was to determine conditions for the oxidative rearrangement. Pyridinyl alcohol 266 was attractive as a simple model system since it contained a tertiary allylic alcohol and a pyridine ring (Scheme LXII). 2-Lithiopyridine (267), prepared via metal-halogen exchange from

2-bromopyridine (268),70 added to 2-cyclohexen-1-one (269) in good yield to afford model substrate 266. Chromium (VI) species are well known to accomplish the desired type of oxidative rearrangement and thus, they were surveyed for one-pot rearrangement and oxidation of 266.71

Unfortunately, these reagents gave poor results. The best results were obtained using pyridinium dichromate, yet even this attempt gave only an 18% yield of enone 270. All trials were plagued by low mass balance as the chromium oxidants seemed to deoompose 266. Focus was then shifted to stepwise oxygen transpostion-oxidation sequences. Palladium-catalyzed allylic acetate rearrangements have been reported, and this method seemed to be a plausible solution to the problem (Scheme LXIIf). For example, Overman has isomerized acetate 271 into acetate

272 using catalytic bis(acetonitrile)palladium(ll) chloride, and Oehlschlager and co-workers converted tertiary allylic acetate 273 into the more highly substituted alkene 274 in high yield using the same catalyst.7 2 -73

Scheme LXII. 1,3-Oxygen Rearrangement with Cr (VI) Reagent.

n-BuLi

268 267 266 270

Thus, alcohol 266 was converted into acetate 275, and this rearrangement was investigated. Tertiary acetate 275 was transposed using bis(benzonitrile)palladium(ll) chloride into secondary acetate 276 in 42% yield along with 18% of recovered 275 (Scheme LXIV).

Despite the low yield, we were pleased that the oxygen functionality rearranged. Our next task was to work on a (Tcarboline system. Scheme LXIIl. Palladium-Mediated Allylic Acetate Rearrangements.

OAc

(CH3CN)2 PdCI2, (cat,)

(CH3CN)2 PdCI2, (cat.)

2 7 3 2 7 4

Our strategy for addition of the (1-carboline unit focused on preparation of 1-lithio-p- carbolines. We envisioned this carbanion to arise from metal-halogen exchange of a 1-halo-fi- carboline. It was imagined that halide introduction might be achieved via diazonium chemistry of known amine 277, Thus, norharman (237)74 was prepared in 33% yield from tryptophan and formaldehyde, and a subsequent Chichibabin reaction 7 5 provided 1-amino-p-carboline (277)74 in 70% yield (Scheme LXV). Although the literature reports conditions of diazotization for pyridine and indole derivatives, mimicking these conditions with amine 277 resulted in decomposition and diazonium chemistry was abandoned .76

Scheme LXIV. Model System for Acetate Rearrangement.

266 275 276

(a) Ac20 , 4-DMAP, Et3 N, CH2 CI2, 87%; (b) (PhCN) 2 PdCI2 (10 mot%), THF, 65°C, 42%. 65

Scheme LXV. Preparation of 1*Amino*p*carbollne.

H H H NHj

2 3 7 2 7 7

(a) HCHO, then K 2 Cr2 0 7 in HOAc, 33%; (b) NaNH2> 70%.

A literature search for 1 -halo-p-carbolines revealed that chloride 278 had been prepared

(Scheme LXVI ) . 7 7 Thus, 3-carboethoxy-2-piperidone (279) was converted into phenylhydrazone 280 in 76% yield by decarboxylation and addition to benzenediazonium chloride .78 Tetrahydro-p-carboline 281 was then prepared using a Fischer indole synthesis and treatment of 281 with dichlorodicyanobenzoquinone provided dihydro-p-carboline 282 in 8 6 % yield .7 7 Literature protocol was followed lor smooth conversion of oxo-p-carboline 282 into chloride 278. Since metal-halogen exchange is generally not efficient for aryl chlorides, this methodology was extended to the preparation of 1-bromo-p-carboline 283. This proved not to be as straightforward as the synthesis of 278. Heating oxo-p-carboline 282 with neat phosphorous oxybromide in the presence of phosphorous pentabromide did not provide any 1 - bromo-p-carboline 2B3. Instead di- and tri-bromo-p-carboline species 284 and 285 were obtained (Scheme LXVII). These products seemed to arise from initial substitution of bromide for oxygen and subsequent bromination of the aromatic ring(s) by the molecular bromine liberated during the reaction. Thus, bromination of the substitution product needed to be suppressed.

This goal was achieved by using toluene as the reaction solvent instead of neat phosphorus oxybromide, and 1 -bromo-p-cartooline (283) was obtained in 31% yield. 66

Scheme LXVI. Preparation of 1-Chloro-p-Carbollne.

nz+ cr o j f * ) OEt

CC*N O N O a* H

2 7 9 2 8 0

2 7 8

(a) KOH, then HCI (to pH 3), 76%; (b) 70% aq. HC 0 2 H, 87%;

Scheme LXVII. Synthesis of 1-Bromo-p-Carboline.

POBr3l PBr5,

A 31%

2 8 2 2 8 3

POBr3, PBrs, A

Br ■'OgC> H Br

285 (24%) 284 (19%) 67

Scheme LXVIII. Bracher's Results with Phosphorous Oxybromide.

Br Br a b

H Br H O H Br

2 8 5 2 8 2 2 8 3

(a) POBr3l A, 51%; (b) POBr3, PhOCH3l 8 8 %.

In 1994, after our studies were complete, Bracher also reported that tri-bromo-p-carbotine

285 was obtained by heating oxo-p-carboline 282 with fused phosphorous oxybromide

(Scheme LXVIII ) .7 9 He also reported that polybromination could be overcome by using anisole as the reaction solvent, giving 1-bromo-p-carboline 283 in high yield. Once again, anisole acted as a bromine scavenger in this reaction, preventing further bromination of the heterocycle. This report is consistent with our observation that toluene gave only "marginal" results when used as the solvent for this reaction.

We also wished to prepare 1-iodo-p-carboline (286) for our lithiation studies. Corcoran had reported the conversion of 2-chloropyridine (287) to 2-iodopyridine (288) upon treatment with sodium iodide and acetyl chloride (Scheme LXIX ) . 0 0 This process proceeded through N- acetyl complex 289. The acyl group activates the aromatic ring, facilitating nucleophilic substitution. Subjecting 1-chloro-p-carboline 278 to similar conditions gave 42% of the desired iodide 286 and 21% of /V-acetyl-p-carboline 290 (Scheme LXX).

Scheme LXIX. Corcoran's Preparation of 2-lodopyrldlne.

c h 3cn, a 54%

287 289 288 68

Scheme LXX. Preparation of 1-lodo-p-carboline.

Nal, AcCI ii n N' CH3CN, H I A CH,

2 7 8 286 (42%) 290 (21%)

Scheme LXXI. Additions of 1-Lithlo-p-carboiine Anions.

c-d OH

2 9 2

e-f g-h I I OH Me

2 8 6 2 9 3 2 9 4

(a) NaH. DMF; (b) MOMCI, 6 6 %; (c) f-BuLi (2.2 eq.), -78°C, E1 20; (d) 2-cyclohexen-1-one, 6 6 %; (e) NaH, DMF; (1) Mel, 85%; (g) n-BuLi, -78°C, THF; (h) 2 -cyclohexen -1 -one. 33%.

Although the yields were not optimal, methods for preparing p-carbolines 283 and 286 were established, and the next task was to convert these compounds into appropriate lithium reagents (Scheme LXXI). Treatment of 1-bromo-p-carboline 283 with sodium hydride and chloromethyl methyl ether gave protected p-carboline 291 in 6 6 % yield. Metal/halogen exchange was accomplished using 2 .2 equivalents of f-butyllithium and addition of the resulting lithium anion to 2-cyclohexen-l-one afforded tertiary allylic alcohol 292 in 6 6% yield. Iodide 287 69

was also used as a precursor to a potentially less hindered organolithium. Thus, N-methyl-p-

carboline 293 was prepared from sodium hydride and methyl iodide, and n-butyllithium affected conversion to the corresponding carbanion. Again, 2-cyclohexen-1-one was used to trap the

anion, affording alcohol 294 in modest yield.

Next, oxygen transposition needed to be accomplished on the ji-carboline system. To our pleasant surprise, treatment of alcohol 292 with acetic anhydride and 4-DMAP did not yield

acetate 295, but instead secondary acetate 296. Scheme LXXII outlines this rearrangement.

Alcohol 292 may first be converted into tertiary acetate 295. Unlike the analogous pyridine system, the indole-like nitrogen of the 0 -carboline could donate electrons into the pyridine ring, enabling the acetate group to dissociate and give iminium ion 297. Attack of acetate as indicated would provide 296. This potential site of alkylation, also present in manzamine A, could provide a rationale for the unknown origin of biological activity in the natural product.

Scheme LXXII. An Unexpected Acetate Rearrangement.

Ac20 , 4-DMAP

OAc 2 9 2 2 9 6

OAc

295 297 70

Scheme LXXIII. Attempted Addition of (3-Carbolfne to Enone 151.

2 9 8

THF, -78 C + No Addition Product 151

Scheme LXXIV. Preparation of Model Enone 300.

OAc OR

Ar'

° /—OH O "S—OMOM n y OTHP | OTHP

^ ' s 301301a a R = Ac >— / b ^ V 301b R = H

Ar' Ar' O ) — OMOM O J — OMOM OTHP OTHP

3 0 0 3 0 2

(a) MOMCI, /-Pr2 NEt, 77%; (b) LiOH-H2 0. MeOH-THF-H 2 0, 94%; (c) (COCI)2> DMSO, -78°C, then Et3 N. 77%; (d) DBU, CH2 CI2, 77%. 71

As addition of the p-carboline unit to 2-cyclohexen-1-one and subsequent oxygen

transpostion had been successful, extending this process to tetracycle 151 became the task at

hand. Unfortunately, treatment of enone 151 with 1-lithio-p-carboline 298 afforded no addition

product (Scheme LXXIII). The only product isolated from the reaction mixture was N -

methyloxymethyl-p-carboline (299), presumably from protonation of the lithium species. An

explanation for this result could be that carbanion 298 acts as a base instead of a nucleophile,

deprotonating enone 151 to give p-carboline 299. The enolate of 151, presumed to be

generated, then seems to decompose as no starting ketone was recovered.

The assumption that 2-cyclohexen-l-one would be an appropriate model for anion

additions seemed naive at this point. Instead, a complex model more closely related to tetracycle

151 was desired. A supply of alcohol 188 was available and a short reaction sequence was used

to prepare bicyclic enone 300, a more suitable model system (Scheme LXXIV). Alcohol 188 was converted into homoallylic alcohol 301 via treatment with chloromethyl methyl ether followed by

lithium hydroxide. Swern oxidation of 301 provided p.y-unsaturated enone 302, and the double bond was moved into conjugation by treatment with DBU to afford enone 300 in good yield.

Although the p-carboline unit was not successfully added to this system, some conclusions were reached from two important experiments (Scheme LXXV). Treatment of enone

300 with N-methyl-1 -lithio-p-carboline 303 again afforded no addition product, only protonated p- carboline 299. It seemed as though the p-carboline anion was acting as a base and not a

nucleophile. In contrast to this result, 2-lithiopyridine added to enone 300 to provide alcohol

304, albeit in a low but unoptimized yield. These results could be explained through a si eric argument. The p-carboline species is a bulkier anion than the pyridyl anion, thus less likely to add to sterically hindered systems. Consequently, it may be more prone to act as a base. Although the enone system of perhydroisoquinoline 300 lacks the y-substitution present in enone 151, these results resemble those obtained in the attempted addition of 298 to tetracycle 151. 72

Scheme LXXV. Additions to Enone 300.

No Addition OMOM UWIUMOMOM OTHP OTHP

As the p-carboline anion would not add to complex enones 151 and 300, an alternative strategy for introduction of this substructure was needed. The addition of an acyl anion equivalent followed by condensation of a tryptamine derivative seemed like a reliable method of p- carboline formation. Several acyl anion equivalents are well-established ,81 as are alkoxymethyl lithium species,8 2 but we chose to use 1,3-dithiane because of its availability. Again, it seemed appropriate to conduct model studies for dilhiane addition, oxygen transposition, and the Pictet-

Spengler reaction.

Scheme LXXVI. Addition of Acyl Anion Equivalent to Enone 151.

151 306 (40%) 307 (15%) 73

Before spending time and energy on the model system, we first examined the addition of

2-!ithio-1,3-dithiane (305) to enone 151 (Scheme LXXVt). Treatment of 151 with anion 305 afforded 1,2-addition product 306 in 40% yield and conjugate adduct 307 as a minor product

(15%). Each product was formed as a single diastereomer of undetermined stereochemistry. We suspected that the lower basicity dithiane anion was responsible for the observed results. After our disappointment with the p-carboline anion, we were pleased with this result and began our model studies.

Addition of 2-lithio-1,3-dithiane to 2-cyclohexe-i-one gave alcohol 308 (Scheme

LXXVII). Conversion to the tertiary acetate 309 provided material to test the palladium-catalyzed oxygen transposition in the presence of the dithiane group. Treatment of acetate 309 with bis(benzonitrile)palladium(ll) chloride affected oxygen rearrangement to give secondary acetate

310 in 57% yield. This established that the thioacetal could tolerate the reaction conditions.

Scheme LXXVII. Dithiane Model.

308 309 310

(a) 2-lithio-1.3-dithiane, THF, -78°C; (b) AC 2 O, 4-DMAP, Et3 N; 76%; (PhCN)2 PdCl2 , THF, A, 57%.

Aldehyde 311 was the model system chosen for Pictet-Spengler condensation chemistry. This material was prepared via the three step sequence outlined in Scheme LXXVIII.

Addition of 2-lithio-1,3-dithiane to 3-ethoxy-2-cyclohexen-l-one (312) followed by an acidic work-up gave enone 313 in 64% yield. Methylmagnesium bromide added cleanly to enone 314, and deprotection of the thioacetal was achieved using thallium trinitrate trihydrate ,6 3 providing aldehyde 311 in 46% yield (two steps). 74

Scheme LXXVIII, Model Aldehyde for Pictet-Spengler Reaction.

p e Sw JS o CHO

OH Me M e

3 1 2 3 1 3 3 1 4 3 1 1

(a) 2-lithio-1,3-dithiane, -78°C, then 10% HCI, 64%; (b) MeMgBr, THF, -78°C, 90%; (c) TI(N03)3*3H20 , 49%.

Initial trials for Pictet-Spengler reactions between tryptamine and aldehyde 311 were

carried out using trifluoroacetic acid at -15°C as described by Waldmann.84 Unfortunately, these

conditions gave a complex reaction mixture and thus, attention was turned to the use of an aprotic

Pictet-Spengler. Cook has published extensively in this area.63'85 Aprotic condensations can be

accomplished by reaction of A/-alkyl substituted tryptamines in the presence of an aldehyde to

yield cyclization products (Scheme LXXIX). For example, heating /V-benzyltryptamine (315) with benzaldehyde (316) in benzene afforded a near quantitative yield of tetrahydro-p-carboline 317.

Conversely, the same reaction conditions applied to benzaldehyde and tryptamine provided only imine 318. These observations have been explained by the reactivity of imine 318 versus iminium ion 319. The imine lacks the necessary electrophilicity for cyclization onto the indole ring, but the more reactive iminium ion can undergo this process. Interestingly, imine 320 derived from benzaldehyde and tryptophan methyl ester 321, has the requisite reactivity for the cyclization to proceed to tetracycle 322. 75

Scheme LXXIX. Cook's Aprotic Pictel-Spengler Reaction. G a ' \ H

3 1 8

. Ph Ph N I r h H Ph

3 1 5 3 1 9 3 1 7

CO^Mo cxr* a „

3 2 1 3 2 0 3 2 2

(a) PhCHO (316), A, PhH.

We needed to choose an appropriate group for the A/-substitution of tryptamine in our system. Any method established for the model system needed to be tolerant of the level of functionality present in tetracyclic substrates. After consideration, an allyl group seemed attractive as the reaction conditions for its removal have been reported to be very selective.06 Therefore,

/V-allyl-tryptamine (323) was prepared in five steps from indole as outlined in Scheme LXXX.

Electrophilic substitution of indole (324) with oxalyl chloride gave acid chloride 3 2 5 .8 7

Conversion into ethyl ester 326 and reduction to the primary alcohol 327 with lithium aluminum 76 hydride proceeded smoothly (75%).08 Substitution of the hydroxyl group gave bromide 328 in good yield,87 and reaction with allylamine provided allyltryptamine 323 in excellent yield.

Scheme LXXX. Preparation of N-Allyltryptamlne (323).

C oN I H

3 2 4

O X 1-,— CX»— O X- ■ I ■ ■ H H H 323 328 327

(a) (COCI)2 , Et20 , 92%; (b) EtOH, Et 3 N, 84%; (c) UAIH 4 , THF, A, 89%; (d) PBr3, Et20 , 68%; (e) CH2=CHCH2Br, H2O-Me0H, 94%.

The aprotic Pictet-Spengler reaction between aldehyde 311 and tryptamine 323 provided tetrahydro-p-carboline 329a and 329b as a mixture of diastereomers (Scheme LXXXI), the stereochemistry at the epimeric center to be lost when fully aromatized. As expected, the allyl group was removed in high yield by heating with tris(triphenyl)phosphinerhodium(l) chloride to provide secondary amines 330a and 330b.86 Aromatization was accomplished by dehydrogenation with dichforcdtcyanobenzoquinone to give a mixture of two compounds. The major product was identified as dihydro-p-carboline 331 and the minor product was the fully aromatized p-carboline 332. 77 Scheme LXXXl. p-Carbollne 332 via Pictet-Spengler.

CHO CO?... PhH, A 6 OH Dean-Stark Me OH ■ K 75% Me 329a and 329b 3 2 3 31 1

(Ph3P)3RhCl. DABCO, EtOH 90%

DDQ

OH OH OH Me Me Me 332 (19%) 331 (44%) 330a and 330b

tautomerize

OH OH 3 3 3 Me 3 3 4 Me

[O]

OH Me

330a and 330b

OH Me OH 331 332 Me 78

Unfortunately, dihydro compound 331 was not converted into 332 upon further exposure to

DDQ or heating with palladium on carbon in ethanol. Scheme LXXXI outlines a possible explanation for this result. Oxidation of amine 330 could provide imine 333 or imine 331. While imine 333 can tautomerize to enamine 334 that can undergo further oxidation, such a scenario is not possible for imine 331. Thus, any amine 330 that is oxidized to imine 331 cannot proceed on to the fully aromatic product 332.

Two other model studies for direct introduction of the p-carboline onto a cyclohexa-1,3- dione derivative were examined. For example, allylic alcohol 332 was prepared through another short sequence (Scheme LXXXII). Bromide 283 was converted to the 1,9-dianion as described by Bracher through sequential treatment with potassium hydride and f-butyllithium.79 Addition to

3-ethoxy-2-cyclohexen-1-one and acid hydrolysis provided enone 336. Methylmagnesium bromide then added to 336 smoothly to afford alcohol 332. This compound will be screened for biological activity as it contains the p-carboline unit and a potential alkylation site similar to manzamine A.

Scheme LXXXII. Alternative Preparation of p-Carboline 332.

H Br O OH Me 2 8 3 3 3 6 3 3 2

(a) KH, THF, 0°C; (b) f-BuLi, -78°C, THF; (c) 3-ethoxy-2-cyclohexen-1-one, -78°C->rt, (d) 10% aq. HCI, 28%; (e) MeMgBr, THF, -78°C, 88%. 79

One final method for introduction of the p-carboline to an A8CD tetracycle involved palladium-mediated coupling of p-carbolines of type 337 with an enone of type 338 (Scheme

LXXXIII). Various palladium catalysts have been used to couple 1 -halo-p-carbolines with other N- containing aromatic heterocycles. For example, Bracher has reported the reaction of 1-chloro-p- carboline 278 with simple aryl boronic acids 340 and 341 using catalytic tetrakis(triphenylphosphine)palladium(O) to alford products and 342 and 343.89 Furthermore, Laborde and co- workers have coupled aryl chloride 344 with cyclohexene 345 to afford 346 in high yield. These examples show the tolerance of a wide variety of functionality for such a transformation (Scheme

LX X XIV ).90 Again, model systems to establish the feasibility of this transformation were considered. A quick model to test the coupling was between p-tributylstannyl-2-cyclohexen-1- one (347) and p-carboline 283, which afforded enone 336 in low yield (Scheme LXXXV). Side products of this reaction appeared to be homocoupled bis(p-carboline) and bis(cyclohexenone), but full characterization these two compounds was not aggressively pursued.

Scheme LXXXIII. Palladium-Mediated Addition of p-Carbollne Unit.

Palladium catalyst

3 3 8 X = B(OH)2,1, SnR3 3 3 9 80

Scheme LXXXIV. Examples of Palladium-Mediated Couplings.

B(OH);

CHO 341

Pd(Ph3P)4, 87%

CHO 3 4 2 2 7 8 3 4 3

H B oc-N SnBu-i o o x y OEt 3 4 5 OEt B oc-N T r r Pd(Ph3P)4, 76%

3 4 4

Scheme LXXXV. Preparation of Enone 336.

SnBu-i

3 4 7 (Ph3P)4Pd, A, 25%

283 336 While determining the optimal conditions for the palladium mediated coupling,

incorporating a p-iodo or p-stannyl group into a tetracyclic enone was necessary. A conversion of

216 into diketone 348 was examined.

Scheme LXXXVI. Preparation and Plan for Dlol 349.

OH OH

CsF, A [O] Ar*-& ~ 7" OH Ar" o \ DMF u SES 90% 2 1 6 3 4 9 3 4 8

Thus, treatment of 216 with cesium fluoride gave diol 349 in 90% yield (Scheme

LXXXVI). As oxidations using Dess-Martin periodinane have become increasingly popular,91 it was decided to use such a hypervalent iodoso compound for this step. Frigerio has reported that iodoxybenzoic acid (350), prepared from 2-iodobenzoic acid and potassium bromate, could serve as a mild but efficient oxidant of alcohols to ketones.92 Diol 349 was oxidized with acid

350 (Scheme LXXXVII), and the crude product treated with diazomethane to trap any p-diketone as an enol ether. It was noticed, however, that no reaction occurred during the attempted methylation. Further interpretation of spectral data strongly suggested that only one hydroxyl group in 349 had been oxidized and the other alcohol remained untouched. As we surmised that the oxidation product was a p-hydroxyketone, treatment with acetic anhydride and chromatography over basic alumina should provided either enone 151 or 152. In the event, this sequence cleanly afforded enone 151, confirming that p-hydroxyketone 351 was the oxidation product. 1H NMR decoupling experiments further supported the structure 351. Although this result was unexpected, a plausible explanation is presented in Scheme LXXXVII. Upon addition of acid 350 to amine 349. an acid-base reaction may occur to initially provide ammonium salt 82

352. Hydrogen bonding between the hydroxyl oxygen ot C 3 and the proton of the ammonium ion could decrease the nucleophilicity of this hydroxyl group, rendering it unreactive toward the oxidant. Regardless of the origin of this selectivity, we were very pleased by this result as it distinguished between the two secondary alcohols of diol 352.

Scheme LXXXVII. Unexpected Oxidation of Diol 349.

OH OH OH

3 5 0 Ar" OH Ar" OH DMSO, N-H 53%

3 4 9 3 5 2

AC2O, EtsN OH

quantitative

151 3 5 1

Two other alternative strategies were aiso investigated lor incorporating the one-carbon unit required by ircinal A. As p-alkoxyketone 222 was not converted to enone 151 when passed over basic alumina, it seemed possible that related compounds might undergo 1,2-addition rather than p-elimination upon treatment with an appropriate carbanion. As this approach was pursued before the oxidation described in Scheme LXXXVII was discovered, that chemistry was not used. 83

Instead, alcohol 228 was converted into tetracycle 353 in high yield by treatment with

chloromethyl methyl ether, and the pivaloate was removed by treatment with tetrabutylammonium

hydroxide to yield alcohol 354 (Scheme LXXXVIII).93 Swern oxidation gave p-alkoxyketone 355.

Unfortunately, 2-lithio-1,3-dithiane did not add to 355 to afford 356, but instead caused p-

elimination to afford enone 151.

Scheme LXXXVIII. Attempted Addition to a p-Alkoxyketone.

OPiv OPiv OH

OH Ar" OMOM OMOM

2 2 8 3 5 3 3 5 4

r ' " i

OH OMOM Ar" OMOM

3 5 6 151 3 5 5

(a) MOMCI, /-Pr2NEt, CH2CI2, 79%; (b) 40% aq. Bu4N+OH-, MeOH, 66%; (c) (COCI)2, DMSO, then Et3N, 50%; (d) 2-lithio-1,3-dithiane, THF, -78°C, 50%.

A plan to add a stablized ylid to epoxyketone 357 is shown in Scheme LXXXIX. It was hoped that oxidation of epoxyalcohol 216 would provide ketone 357. Treatment of 357 with ylid 358 or 359 would then provide bisepoxide 360.94 Tetracycle formation with cesium fluoride 84 would yield 361, followed by oxidation to ketone 362 and elimination to give enone 363. The hydroxymethyl group at Ci would serve as a handle for p-carboline incorporation. As bisepoxide

217 was an epoxidiation by-product and could not be reintroduced into the synthesis, it was used to scope out this chemistry. Alcohol 217 was oxidized using iodoxybenzoic acid (350), and although ketone 364 seemed to be the initial product in the crude reaction mixture, chromatography over silica gel afforded the enone 365 resulting from epoxide opening (Scheme

XC). One attempt was made to treat crude 364 with ylid 358, but a complex mixture was obtained from this reaction. As epoxyketone 364 appeared to be quite sensitive to acid or base, this route was also abandoned.

Scheme LXXXIX. Alternative Strategy Adding an Acyl Anion Equivalent.

N—SES Me2S(0)=CH2

2 1 6 3 5 7 3 6 0

363 362 361 65

Scheme XC. Oxidation of Epoxy Alcohol 217.

DMSO, 60%

2 1 7 3 6 4 3 6 5

Our plan lor incorporation of the p-carboline unit was revised at this point. Addition of a formyl anion equivalent to a ketone similar to 351 would provide diol 366 (Scheme XCI). It was imagined that oxidation of the secondary hydroxyl group and subsequent elimination would provide enone 367. This intermediate is analogous to enone 150 in the retrosynthetic analysis in Scheme XXVIII. The results of these investigations will be described in Chapter V.

Scheme XCI. New Strategy to Introduce an Acyl Anion Equivalent.

3 51 3 6 6 3 6 7 CHAPTER IV.

STUDIES FOR INCORPORATION OF THIRTEEN-MEMBERED

RING OF MANZAMINE A.

A. Introduction.

A major goal of the work described in this chapter was incorporation of the thirteen membered ring of manzamine A. As outlined in the retrosynthetic analysis (Scheme XXVIII) and also in Scheme XCII, the original plan called for macrolactamization to accomplish this step. The addition of an organometallic substrate of type 149 to enones 150,152, or 367 was to afford the cyclization precursor. Thus, this sequence required the synthesis of an appropriately substituted eight carbon unit, development of an efficient organometallic addition, and proper choice of a reagent to promote macrolactamization.

Scheme XCII. Strategy for Introducing Thlrteen-Membered Ring.

R'OzC Mel 1 4 9

OH □ O

1 1 48 150 R2 = p-carboline 152 R2 = H 367 R2 = 1,3-dithianyl

86 87

B. Background.

Although macrolactamizations have been relatively scarce due to recurring problems of polymerization, protocols have been developed for such cyclizations. For example, Corey generated a mixed anhydride to activate a carboxyl group in a synthesis of (±)-A/-methyl- maysenine.95 Treatment of 368 with mesitylenesulfonyl chloride facilitated lactamization to give the nineteen-membered ring of 369 (Scheme XCIII). Maytansine was also prepared by Corey using this protocol for amide bond formation.96

Scheme XCIII. Corey's Macrolactamizatlon.

Me Me MeO, NHMe Me MeO, MeOTBS Me OTBS

2. 2,4,6-Me3C6H2S 0 2CI, Me MeO f-Pr2NEt, 65% Me MeO

3 6 8 3 6 9

Scheme XCIV. Bars Macrolactamizations.

ch3

ct c h 3 o t b d m s Br OTBDMS

MeO^ .. NH JL yJV .Me MeO. Me Me Me 3 7 0 OTBS

Me MeO

3 7 2

370 or (PyrS)2-PPh3 (375)

n = 6-10, 70-90% with 370 n = 5-7, 35-70% with 375

373 374 88

Bai and co-workers have used pyridinium iodide 370 to close the macrocycle of 372 from precursor 371 in good yield (Scheme XCIV).97 Furthermore, the Bai group has extended this methodology to cyclize a variety of a,io-amino acids 373 to afford macrolactams 374 in variable yields using either 370 or dipyridinedisulfide triphenylphosphine (375) as the coupling reagent.98 General methods for macrocyclizations have been reviewed by Paterson99 and

Hesse100 and will not be discussed herein.

C. Results and Discussion.

The first objective for this portion of the research was to prepare an appropriate precursor to an organometallic of type 149. As a macrolactamization was planned, one terminus of the eight-carbon chain was to be at the acid oxidation level while the other end was to be a surrogate for a carbanion. Two carbon homologation of 5-hexyn-1-ol, whose preparation was described in

Chapter II, was expected to provide an expedient entry into such compounds. Thus. 5-octynoic acid derivative 376 became the target molecule to serve as the organometallic precursor

(Scheme XCVI).

The preparation of orthoester 377 was carried out by the procedure of Corey as outlined in Scheme XCV.101 Oxidation of 5-hexyn-1-ol (183) with chromic acid gave 5-hexynoic acid

(378) in 50% yield.102 Treatment of 378 with thionyl chloride gave acid chloride 379 in good yield. Condensation of 379 with alcohol 380 (prepared from 2,2-bishydroxymethyM -propanol

(381)) provided ester 382 in 91% yield. Isomerization of 382 using boron trifluoride etherate gave the desired orthoester 377 in 56% yield. 89

Scheme XCV. Preparation of Orthoeater 377.

°O H ------/ r\ s “ = ----- V /~°H ~ " — = ----- V C02H ■ w OH OH

381 380 183 378

d V COCI VC o \ /

377 382 379

(a) CO(OEt)2, EtOH, KOH, 43%; (b) Cr0 3 , AcOH, 50%; (c) SOCi2> A; 72%; (d) 380. pyridine. CH2CI2l 91%; (e) BF3*Et20 . -20°C, 56%.

The two-carbon homologation was the next task (Scheme XCVI). We planned to

accomplish this transformation by addition of the lithium acetylide of 377 to ethylene oxide. Initial

trials involving treatment of the acetylide of 377 (prepared using n-butyllithium) with ethylene

oxide gave poor yields (less than 10%) of the desired alcohol 383. Nicolaou has reported

excellent yields for reactions of acetylides with ethylene oxide in the presence of N,N,N,N-

tetramethylethylenediamine (TMEDA).103 Using this additive in our system afforded 383 in 84%

yield. As the orthoester moiety of 383 was very sensitive to acid hydrolysis, conversion of the

hydroxyl group to a halide had to be accomplished using neutral to basic conditions. Thus,

treatment of alcohol 383 with p-toluenesulfonyl chloride provided tosylate 384 in 72% yield, and

384 was converted into bromide 376 in variable yields (30-75%) using lithium bromide in acetone. The orthoester was problematic for this step as hydrolysis seemed to occur to yield polar side-products. In general, the orthoesters in this sequence were very sensitive to hydrolysis. 90

Only basic aqueous work-ups were performed and all chromatography was performed over basic

alumina or silica gel pretreated with triethylamine.

Scheme XCVI. Preparation of Orthoester 376.

1. n-BuLi, then TMEDA, -78°C, THF HO 2. Ethylene oxide, 84% 3 7 7 3 8 3

LiBr, acetone, A Br Ts< 30-75% 3 7 6 3 8 4

Since the orthoester caused problems, octynol derivatives 385 and 386 were prepared via a similar sequence (Scheme XCVII). It was imagined that oxidation of the oxygenated terminus of 385 or 386 could be accomplished once the chain was attached to the tetracyclic nucleus of mazamine A. Protected alcohol 186 was prepared as described earlier, and sequential treatment with n-butyllithium, TMEDA, and ethylene oxide afforded alcohol 387 in 85% yield. Conversion of 387 to tosylate 388 was achieved by treatment with p-toluenesulfonyl chloride. Tosytate 388 was then converted to either bromide 385 or iodide 386 in high yield by heating with the appropriate lithium halide in acetone. Bromide 385 served as the precursor to an eight-carbon organometallic. Thus, ireatment of 385 with 2.2 equivalents of f-butyllithium and quenching with

2-cyclohexen-1 -one afforded alcohol 389 in 50% yield. Enolization of the enone seemed to be a competitive process as octynol derivative 390 was observed as a side product. 91

Scheme XCVII. Preparation and Use of Bromide 385.

HO Ts c o t h p a-b

THPO THPO

1 86 3 8 7 3 8 8

d o re

/ HO.

+ THPO THPO THPO

3 9 0 3 8 9 385 X = Br 386 X = I

(a) n-BuLi, THF, -78°C, then TMEDA; (b) ethylene oxide, THF, 85%; (c) p-TsCI, Et 3 N, 4-DMAP, CH2 CI2 , 93%; (d) Lil, acetone, A, 90%; (e) LiBr, acetone, A, 91%;

The next task was the addition of carbanion 391 to enone 152 (Scheme XCVIII). The

carbanion was prepared with f-butyllithium, and enone 152 was added to the lithium reagent.

The reaction was not clean and the mass balance was not good, but a variable yield (20-40%) o1 a

compound tentatively assigned as the desired 1,2-adduct (392) was isolated. Although the

signals in the 1H NMR spectrum were consistent for the target, the integration was inconsistent,

leading to some uncertainly in the assignment. As addition of the lithium reagent was not

effective, small organometallic reagents were surveyed for addition to enone 152. No addition was observed when enone 152 was treated with methyllithium or methyl magnesium bromide, but treatment of 152 with methyl cerium chloride gave a very clean 1,2-addition (Scheme XCVIII).

This reaction afforded alcohol 393 in 64% yield along with 20% of recovered starting material. It seemed likely that methyllithium and methyl magnesium bromide had enolized enone 152 while the cerate, reputed for its low basicity and high nucleophilicity, gave the 1 ,2 -adduct.104 92

Scheme XCVIII. Organometallic Additions to Enone 152.

■OTHP

OH 391 THPO. THF, -78°C 20-40%

1 5 2 3 9 2

OH CH3CeCI 2 Ar" CH.

THF, -78°C 64% (80% based on recovered 152)

15 2 3 9 3

We were very pleased to see that organometallic additions to this enone could be accomplished, but the stereochemistry of this addition was also of great interest. A series of 1H

NMR nOe experiments designed 1o determine stereochemistry at the tertiary alcohol center of

393 are summarized in Table 2 and Figure 6 . A tentative stereochemical assignment was based on these experiments. Irradiation of H 5 (4.1 ppm) showed and enhancement of the CH 3 group, but the enhancement was not as large as one might expect for a cis relationship. As expected, irradiation of the methyl group gave a modest enhancement of Hs. However, no enhancement was observed at H 6 or H7 (CH2 N group), and at least a small enhancement would be expected if the addition had occurred to give the (J-alcohol. Thus, the cis relationship of the hydroxyl group and the amine nitrogen, resulting Irom cerate addition from the (Mace, was tentatively assigned. 93

Table 2. Selected nOe Experiments with 393.

Irradiated H ppm Enhanced H ppm (%)

H3 6.01 H2 5.59 (6.5%) H4 1.45(4.1%) H2 5.59 H3 6.01 (9.0%) H11orH12 3.31(5.8%) H6 orH7 2.71 (2.3%) H10 4.25 H8orH9 5.77(1.0%) H6 or H7 2.79 (2.3%) H5 4.10 H®orH9 5.77 (2.6%) H6 or H7 2.79 (2.8%) H4 1.45(3.8%) H4 1.45 H3 6.01 (9.5%) H5 4.10 (6.5%) H11orH12 3.35 H2 5.59 (7.4%) H5 4.10 (0.8%) H11 or H12 2.97(19%) H6orH7 2.75 H2 5.59(2.1%) H5 4.10 (2.2%)

H" OH Ar CH.

H10

3 9 3

Figure 7. Alcohol 393

It was hoped that transmetallation of organolithium reagent 391 to a cerate would provide a nucleophilic source of the eight carbon unit. Unfortunately, all attempts to converting 391 into the cerate met with failure as no 1,2-addition product was observed upon quenching with cyclohexenone. Pandit's group has observed similar results when attempting organometallic additions (Scheme XCIX).105 For example, upon treatment of ketone 394 with Grignard reagent

395, no addition product 396 was observed and starting ketone was partially recovered. As enolization was suspected to be a competitive process, the Grignard was converted into the corresponding cerate. Again, this reagent gave no observed addition. However, it was 94

commented that upon “inspection of Dreiding models” that enone 397 should undergo stereo

selective nucleophilic addition more easily than ketone 394.

Scheme XCIX. Pandit's Atlempted Organometallic Additions.

CO,CH, CQ2Me H i

Met OTBS TBSO PH o \ _ ’ 395 - \ * O ' — N Cbz Cbz

394 396

397

A number of problems had been encountered by this point. The eight carbon chain had

not added in good yield, and its oxidation level needed to be adjusted. Furthermore, the p-

methoxyphenyl group of the lactam nitrogen still needed to be removed. Although eerie

ammonium nitrate (CAN) had been used to obtain a moderate yield of deprotected

perhydroisoquinoline 204 (Scheme XLIII), alcohol 392 had a basic nitrogen and a tertiary allylic

alcohol, functionality potentially incompatible with CAN. These factors led to the exploration of other approaches.

Alternative plans for incorporation of the thirteen-membered ring involved adding two

smaller carbon units to N4 and C4 followed by closing the macrocycle via formation of the carbon- 95

carbon double bond. One strategy is presented in Scheme C. This metathesis approach would

require addition of a 3-butenyl unit to the B*ring carbonyl group of enone 367. Furthermore, the

lactam nitrogen would need to carry a 5-hexenyl unit to afford tetracycle 39B. An intramolecular

metathesis reaction between the terminal vinyl groups would provide pentacycle 399. The

second approach is outlined in Scheme Cl and uses Ramberg-Backlund chemistry for carbon-

carbon double bond formation.106 Analogous to 398, incorporating an allyl group at the B-ring

carbonyl group and a 4-pentenyl group on the lactam nitrogen would provide 400.

Hydroboration of the terminal alkenes would give diol 401 and double displacement would

provide sulfide 402. Oxidation to sulfone 403, a-halogenation, and Ramberg-Backlund

extrusion of sulfur dioxide would provided 399.

Scheme C. Olefin Metathesis Approach for Thlrteen-Membered Ring.

OH □ OH

150 R = p-carboline 3 9 8 3 9 9 152 R = H 367 R = 1,3-dithianyl

Pandit and Marlin have both used olefin metatheses in their studies toward the total synthesis of manzamine A.24*26 Martin has also asserted the generality of this method by cycling a,a>-dienes 404-407 to cycloalkenes 408-411 (Scheme Cll).107 Good to excellent yields were observed when closing five- to eight-membered rings 408-411, but no macrocycle formation was observed for the n = 7 substrate 412. Even under very dilute conditions, only dimerization was observed for 412. Martin has applied this methodology in an approach to FR-900482.108 96

One important aspect of Martin's and Pandit's metatheses is that in each case, the nitrogen is protected with an acyl group. Our system would require this reaction to be carried out in the presence of a tertiary amine.

Scheme Cl. Ramberg-Backlund Approach to Thirleen-Membered Ring.

..c±x a " .□ OH

HO, OH

150 R = 0-carboline 4 0 0 401 152 R = H 367 R =1,3-dithianyl □

OH OH □ OH

3 9 9 4 0 3 4 0 2

Scheme Cll. Martin's Metatheses.

|) PhMe2CCH=Mo=NI216-{/-Pr)2C6H3][OCMe(CF3)2]2

404-407, n = 0-3 408-411, n = 0-3, 65-90% 412, n = 7 97

The first goal of these investigations was to add a three or four carbon organometallic to enone 151 and secure the stereochemistry of this addition. Pandit had reported that 3-butenyl magnesium bromide did not add to ketone 137 to yield 413, but allyl magnesium bromide gave clean 1,2-addition to afford cyclic carbamate 138 (Scheme Clll).26

Scheme Clll. Pandit's Organometallic Additions.

COjCH; c o 2c h 3 c o 2c h 3

o COjEt O

1 3 8 1 3 7 41 3

(a) CH2=CHCH2MgCI, then NaH, 55%; (b) CH2=CHCH2CH2MgBr.

Scheme CIV. Addition of Grignard Reagents to Enone 152.

4 1 5 152 4 1 4

(a) CH2 =CHCH2MgCI, THF, -78°C, 90%; (b) CH2=CHCH2CH2MgBr, THF, -78°C, 60% (based on recovered 152).

We wished to see if these observations held true for our system. 3-Butenylmagnesium bromide was added to enone 152, but surprisingly only 1,4-addition, not 1,2-addition, was observed (Scheme CIV). The reaction proceeded stereoselectively as only one diastereomer of 98 ketone 414, tentatively assigned as the p-isomer, was observed along with unreacted starting material. Although this result was intriguing and seemed to indicate that the enone carbonyl of

152 was hindered towards nucleophilic addition, it did not construct the proper carbon-carbon bond necessary lor the metathesis approach. Addition of 3-butenyllithium was also attempted but no addition product was observed.109 Again, as with other lithium reagents, the mass balance was low for recovered 152.

Attention was then turned to the addition of allylmagnesium chloride to enone 152 as allyl Grignards are well-known to undergo efficient 1,2-additions. Treatment of enone 152 with an excess of allyl magnesium chloride gave adduct 415 in 90% yield as a single diastereomer.

Coordination of the Grignard to the carbonyl oxygen seemed to be effective in controlling the site of addition for this system. Amine 415 was converted to its hydrochloride salt, and a crystal was grown from 1,2-dichloroethane and hexanes. Solvent trapped in the crystal lattice evaporated quickly such that high quality crystallographic data could not be collected. Nevertheless, these data conclusively indicated that p-face addition of the organometallic had occurred. The cis relationship established between the tertiary amine and the hydroxyl group match the stereochemistry found in manzamine A. The crystal structure for this compound is presented in

Figure 8. The allyl group could be used for the Ramberg-Backlund approach presented in

Scheme Cl or be homologated into a 3-butenyl unit via a sequence similar to that such as reported by Pandit. For use in the Ramberg-Backlund approach, hydroboration of the terminal vinyl group was needed. It was envisioned that alcohol 416 could be prepared by using 9-BBN and subsequent oxidation with trimethylamine N-oxide.110 Unfortunately, this reaction was unsuccessful as no hydroboration occurred and only starting material was observed in the 1H

NMR of the reaction mixture (Scheme CV). Establishing the feasiblity of this hydroboration could be studied using alcohol 417 as a simple model. This diene could also be used in preliminary studies for selective addition of thiols to terminal alkenes.111 These studies have yet to be completed. 0 3 99 C23

C27

1C 19 C25 C21 C12 C26 Cl 3, CiB

cie CU N2 101 C17 02 IC10 CIS C14

C9 C2

C4 C6 C5

Figure 8. X-Ray Structure of Alcohol 415.

Scheme CV. Functlonallzation of Allyl Group

OH 1. 9-BBN

2, Me3N+-0'

4 1 5 4 1 6

sn □ HO.

418 100

The next goals were deprotection of the lactam nitrogen, reduction of the amide, and incorporation of a five or six carbon unit. Removal of the p-methoxyphenyl group from 214 was achieved using eerie ammonium nitrate to provide secondary amide 416 in modest yield and nitro compound 419 as a minor side product (Scheme CVI). Unfortunately, nitration of the aromatic ring most likely rendered the phenyl group unreactive toward oxidation, thus material converted into 419 was no longer of use in the synthetic scheme.

Scheme CVI. Deprotecflon of Lactam Nitrogen.

OAc OAc OAc

CAN, A r' ~ SES CHaCN, SES SES H2O H,CO' NO.

214 418 (40%) 419 (7%)

As the secondary amide 418 was now in hand, reduction of the amide to the amine was investigated. Strong reductants like lithium aluminum hydride were ruled out, as reduction of an imino ether intermediate was more attractive as a mild method for converting the lactam to the amine. Kuehne has reported the reduction of simple amides and lactams, such as 420, to the corresponding amines 421 using phosphorous oxychloride and sodium borohydride via iminium ion 422 (Scheme CVII).112 A two-step procedure has been reported by Borch. For example, conversion of caprolactam (423) into iminoether 424 and subsequent reduction by sodium borohydride afforded amine 425.113 This procedure worked nicely for a variety of secondary and tertiary amides. Ohno and co-workers reported the conversion of caprolactam into iminoether

426 in excellent yield upon treatment with diazomethane and silica gel.114 With these options, the protocol by Borch seemed appropriate for our system. Thus, treatment of 418 with triethyloxonium tetrafluoroborate afforded iminoether 427 in near quantitative yield. Attempted 101

reduction of this intermediate to amine 428 with sodium cyanoborohydride and trifluoracetic acid

or sodium borohydride met with failure (Scheme CVIII).

Scheme CVII. Literature Reductions of Amides via Imlnoethers.

NaBH4 •Ph POCI3 Ph Ph 70% O'

4 2 0 422, X, Y = Cl, OPOCI2 4 2 1

Et30 +BF4" NaBH4 O'N OCHjCHj O o 92% 4 2 4 4 2 3 4 2 5

CH2N2, silica gel OCH, O N 95% O

4 2 3 4 2 6

Focus was next turned to reduction of an iminothioether instead of an iminoether, as the former would undergo reduction more readily. Amides have been converted into amines via the three-step process outlined in Scheme CIX by Raucher and co-workers.115 Lawesson's reagent

(429) converted amide 430 into thiolactam 4 3 1 .116 Treatment with Meerwein's reagent afforded thioimidate 432, and reduction to the amine 433 was then accomplished with sodium borohydride. Sundberg and co-workers have also developed methodology by which secondary and tertiary amides, such as 434, have been converted into thioimides or thioimidates (435) and reduction with sodium borohydride or sodium cyanoborohydride provided the corresponding amine (436).117 102

Scheme CVIII. Attempted Reduction of Amide 418.

OAc OAc OAc

El30 + NaBH4 or „SES EtO SES SES CH2Cl2, NaBHgCN 98%

41B 4 2 7

Scheme CIX. Literature Reductions of Lactams via Thlolactam.

o Sv S Ar Ar. , V S * . S <429>

Ar= p-methoxyphenyl

70%

4 3 0 431

Et30 +BF4‘

BFi SEt

NaBH4

94%

Ts COaMe Ts

4 3 3 4 3 2

SCH, Mel NaBH4 (96%) or

Ph Phr^*0 Dk NaBH3CN (89%) Ph

434 435 436 103

As Lawesson's reagent (429) has become popular for conversion of amides into thioamides, this method was used to convert lactam 418 into the thiolactam 437 (Scheme CX).

The reaction was very sluggish, even upon heating at reflux in toluene and using an excess of

Lawesson's reagent. Furthermore, the excess Lawesson's reagent and associated by-products were difficult to remove, requiring chromatography three times to afford a 33% yield of isolated thiolactam 437. This compound was converted smoothly into iminothiol 438 upon treatment with methyl iodide and sodium carbonate in 71% yield. Alternatively, the impure thiolactam 437 from one chromatography could be converted into thiol 438 in 48% yield for the two-step sequence. Treatment of thiol 438 with sodium cyanoborohydride and trifluoroacetic acid afforded secondary amine 428 in 83% yield. Although a three-step protocol was required for the reduction, secondary amide 418 was converted into amine 428 in an overall 40% yield.

Scheme CX. Preparation of Amine 428.

OAc Ar OAc

PhCH3, A

SES Ar= p-methoxyphenyt

4 1 8 4 3 7

CH3l, Na^Os, 48% (2 steps) OAc OAc

NaBH3CN, TFA

428 438 104

Scheme CXI. Exocyclic Amide Approach.

OAc OAc

CH2=CHCH2CH2COCI -Oi> SES V n ' s e s Et3N, CH2CI2, 94% o

4 3 9 4 4 0

LiOH*H2Op THF-MeOH-H 20 , 90%

Mo(CO)6, t-BuOOH SES ,SES PhH, A, 38%

4 4 2 4 4 1

The plan next called for attachment of a five-carbon chain to the amine nitrogen. This

chain was smoothly introduced to 428 upon treatment using 4-pentenoyl chloride to provide

exocyclic amide 440 in 94% yield (Scheme CXI). Deprotection of of the acetate was achieved with lithium hydroxide, affording a 90% yield of homoallylic alcohol 441. Both exocyclic amides

440 and 441 showed both amide geometrical isomers by *H NMR at room temperature, and some signals were still broad upon heating to 338 K. Homoallylic alcohol 441 was convened into epoxyalcohol 442 using molybdenum hexacarbonyl and f-butylhydroperoxide in modest yield, and again the 1H NMR and 13C NMR spectra were broad at 343 K. To our dismay, heating to 373

K in hopes of fully resolving all signals resulted in decomposition of 442. Because of this 105

characterization problem, a better strategy seemed to be leaving the endocyclic amide intact and

introducing the five carbon chain as an alkyl group rather than as an acyl group.

A variety of reaction conditions were surveyed for the alkylation of secondary amide 416

(Scheme CXII). Treatment of 418 with potassium hydride in DMSO at room temperature followed

by addition of an excess of 4-bromopentene (443) afforded tertiary amide 444 in 42% yield. No

improvement was observed when using 4-iodopentene as the alkylating agent. The acetate was

removed cleanly from 444 with lithium hydroxide to provide a 90% yield of homoallylic alcohol

445. We were, however, not satisfied with the results of the alkylation. Extending the strategy

used for cyclization of the azocine ring (Scheme XLVII in Chapter II), the alkylation was carried out

in toluene with 18-crown-6 to aid in solvation. Performing this reaction at room temperature gave

the best results. Acetate 444 and alcohol 445 were both present in the product mixture, so the

crude material was treated with lithium hydroxide to provide a 54% yield of atcohol 445 for the

two-step process.

Scheme CXII. Alkylation of Amide 418.

OAc OAc OH

SES SES SES

4 1 8 4 4 4 4 4 5

(a) KH. 18-Crown-6, PhMe, rt, then CH 2=CHCH2CH2CH2Br (441), (b) LiOH-H20, MeOH-THF- H20 , 54% (two steps).

In summary, this chapter has presented methodology for the introduction of an ally! group

and 4-pentenyl chain to advanced manzamine A intermediates, opening potential routes for incorporating the thirteen-membered ring. The stereochemistry of allyl group addition was proven 106 by x-ray crystallography. The next chapter will describe our efforts at the forefront of the synthesis, using the methods presented in the Chapters ll-IV. CHAPTER V.

MOST ADVANCED SYNTHETIC INTERMEDIATES

AND CONCLUSIONS.

A. Results and Discussion.

Using the methodology presented in the previous three chapters, this chapter will describe results at the forefront of our effort towards a total synthesis of manzamine A. The strategy is outlined in Scheme CXIII. Tertiary amide 448 should be a useful substrate for the

reaction sequence described in Chapter II for elaboration of the tetracyclic nucleus. Diol 446 is analogous to diol 349 (Scheme LXXXVII), and it was imagined that selective oxidation of the Ci hydroxyl would provide a p-hydroxyketone. Adjustment of oxidation states and addition of a dithiane unit and allyl group would provide alcohol 447. Finally, Ramberg-Backlund chemistry would be investigated for closing the thirteen-membered ring. This late intermediate (447) would also contain all the necessary carbons for completing irdnal A.

Scheme CXIII. Plan for Preparation of Alcohol 447.

445 4 4 6 447

107 108

Preparation of tetracycle 446 proceeded as planned (Scheme CXIV). The directed epoxidation was accomplished as described for alcohol 214, but with poorer selectivity. Treatment of alcohol

445 with catalytic molybdenum hexacarbonyl and Fbutyl hydroperoxide afforded epoxyalcohol

448 in 50% yield and bisepoxide 449 in 20% yield. The terminal vinyl group, however, remained untouched. The tetracyclic nucleus was closed by removal of the SES group from 448 with cesium fluoride and subsequent ring closure to provide 446 in high yield.

Scheme CXIV. Synthesis of Ketone 454.

OH OH OH

SES SES SES

4 4 5 4 4 9 4 4 8

i^ i OH

R1 d-e OH OH

4 5 0 4 4 6 451 R1 = OH, R2 454 R1, R2 = O

(a) Mo(CO)e (cat), f-BuOOH, PhH, A, 448, 50%; 449, 20%; (b) CsF, DMF, A, 85%; (C) iodoxybenzoic acid, DMSO. 53%; (d) 2-lithio-1,3-dithiane, THF, -78°C, 75%; (e) DMSO, (COCI)2, Et3N, 55%.

Diol 446 was converted into p-hydroxyketone 450 through a selective oxidation with iodoxybenzoic acid in DMSO. Thus, the secondary hydroxyls at Ci and C 3 of diol 446 were differentiated and the introduction of a handle for construction of the P-carboline unit could be studied. It was envisioned that an organometallic might add to Ci of 453 via the following 109

process. The first equivalent of an organometallic should deprotonate the alcohol to provide an

alkoxide, rendering the p-oxygen immune to elimination. The second equivalent should then

undergo 1,2-addition to the ketone. This scenario was observed with 2-lithio-l ,3-dithiane as

clean addition to ketone 450 afforded diol 451 in 63% yield as one diastereomer, along with

16% of recovered 450. This result marked our most successful addition of an acyl anion

equivalent to a tetracyclic nucleus. The presence of the alkoxide may have decreased the

propensity for enolization, thus decreasing side reactions resulting from enolate formation.

Encouraged by this result, addition of p-carboline anion 452 to ketone 450 was attempted.

Unfortunately, no addition product 453 was observed (Scheme CXV). Either carbanion 452 was too large and steric hinderance prevented the addition, or enolization remained the preferred

reaction pathway.

Scheme CXV. Attempied Addition of p-Carboline to Ketone 457.

4 5 0 4 5 3

Our next goals were the oxidation of the C 3 alcohol of 451, f)-elimination of the hydroxyl group, and addition ol allylmagnesium chloride to the ketone. Swern oxidation did convert diol

451 into p-hydroxyketone 454 in 53% yield, but the dehydration was surprisingly difficult and has yet to be accomplished (Scheme CXVI). Mimicking the protocol used previously for p- elimination with of p-hydroxyketones, treatment of 454 with acetic anhydride gave no enone

455. Presumably the tertiary hydroxyl of 454 is very hindered and thus, unreactive towards 110 acetylation. The use of trifluoroacetic anhydride gave no better results. Attention was next

turned to Martin's sulfurane (456) as it has been reported to be effective for dehydration . 1 1 8

Again, no elimination was observed and further studies are needed. Once enone 455 is

prepared, completion of alcohol 457 will be attempted by addition of allylmagnesium chloride to

4 55 .

Scheme CXVI. Attempted Elimination and Use of Enone 455.

AC2O or Tf20, Et3N, 4-DMAP

Ph2S[OC(CF3)2Ph]2, ^ [| (456) V

4 5 4 4 5 5 4 5 7

B. Conclusions.

This thesis has detailed continuing studies toward a total synthesis of manzamine A (1).

Improved routes to two previously reported tetracyclic enones and the synthesis of other

advanced intermediates have been described. The p-carboline unit has been added to simple

model systems but not to a tetracyclic nucleus. However, a potential precursor of the p-carboline unit, 2-lithio-1,3-dithiane, has been added to two tetracyclic ABCD systems, thus also providing a potential route to ircinal A (21). Although an approach to adding an eight-carbon organometallic to tetracyclic enone was not fruitful, smaller carbon units have been added in good yields to the A and B rings of advanced intermediates. These chains hopefully will be brought together to form the thirteen membered ring. A summary of the synthesis of ketone 452, the most advanced intermediate in these studies, is presented in Scheme CXVII. 111 A number of questions need to be investigated in future studies . The synthesis of 452 requires twenty-eight linear steps. A more convergent approach would be welcome as lack of large quantities of advanced intermediates often has been a problem throughout these studies.

Scheme cxvu. Synthesis of Ketone 454 from Benzoic Acid.

OTHP CHO

m-o | 6 8 %

OAc

OTHP OTHP

454 446 445 112

(a) Li. NH3, -33°C, then BrCH 2 CH2 OCH3; (b) DPPA, pyrrolidine, Et 3 N; (c) l2, H2 0-THF; (d) CH2»CHCH2 SnBu3, AIBN, PhH, A; (e) BrMgC 6 H4 OCH3> THF, 0°C -► rt; (f) A c ^ , Et3N, CH2 CI2; (g) OSO4 (cat), Nal04; (h) MeOH, Dowex-H+; (i) NaBH 3CN, TFA, CH2 CI2; (j) BBr3, -78°C -> -15°C, CH2 CI2, (k) DMSO, (COCI)2. then Et 3 N; (I) ICCCH2 CH2CH2 CH2 OTHP, CrCI2, NiCI2 (cat.), THF; (m) Mn0 2 . CH2 CI2; (n) CeCI3, NaBH4l MeOH, -78°C; (o) H2, Pd on BaS04, pyridine, 1 atm; (p) SESNHf-Boc, DEAD, Ph3 P, THF; (q) THF, MeOH, p-TsOH; (r) DMSO, A; (s) p-TsCI, Et3 N, 4- DMAP, CH2 CI2; (t) KH, 18-C-6, PhCH3, A; (u) CAN, CH3 CN-H2 0; (v) KH, 18-C-6, PhCH3, rt, then CH2 =CHCH2 CH2 CH2 Br; (w) U0H-H 2 0, MeOH-THF-H 2 0; (x) Mo(CO) 6 , f-BuOOH, PhH, A; (y) CsF, DMF, A; (z) lodoxybenzoic add, DMSO, rt; (aa) 2-lithio-1,3-dithiane, THF. -78°C; (bb) DMSO, (COCI)2, then Et 3 N.

Furthermore, the route presented in this thesis is racemic, so an enantioselective synthesis would be of great interest. The addition of other acyl anion equivalents or alkoxymethyl organometallics, rather than a 1,3-dithiane unit, may also be necessary as the deprotection of thioketals can be capricious. Furthermore, closure of the thirteen membered ring remains unchartered.

Addressing these challenges will complement the studies described in this thesis and hopefully lead to the synthesis of manzamine A. Experimental Section

All melting points were taken with a Thomas-Hoover capillary melting point apparatus and are uncorrected as are all boiling points. Proton nuclear magnetic resonance spectra (1H NMR) were recorded on a Bruker AC-200, Bruker AM-250, Bruker AM-300, or Bruker AC-300 spectrometers and are recorded in parts per million from internal chloroform, benzene, or dimethylsulfoxide on the 5 scale. The 1H NMR spectra are reported as follows: chemical shift

[multiplicity (s = singlet, d - doublet, t = triplet, q = quartet, m = multiplet, br = broad), coupling constants in hertz, integration, interpretation]. Assignments for spectra denoted by a (♦ ) have been determined by extensive decoupling experiments. Carbon-13 nuclear magnetic spectra

(13C NMR) were obtained with a Bruker AM-250, Bruker AM-300 or Bruker AC-300 spectrometers and are recorded in parts per million from internal chloroform, benzene, or dimethylsulfoxide on the S scale. Sets of peaks enclosed within brackets, for example [33.4, 33.6], denotes signals from a diastereomeric mixture. The 13C NMR spectra are reported as follows: chemical shift

(multiplicity determined from off-resonance decoupled or DEPT spectra). Infrared spectra were taken with a Perkin-Elmer 1600 (FT-IR) instrument. Mass spectra were obtained on a Kratos MS-

30 or Kratos VG70-250S instrument at an ionization energy of 70 eV. Compounds of which an exact mass is reported exhibited no significant peaks at m/z greater than that of the parent.

Combustion analyses were performed by Atlantic Microlab. Norcross, GA.

Solvents and reagents were dried and purified prior to use when deemed necessary: tetrahydrofuran, diethyl ether and benzene were distilled from sodium metal: dichloromethane, triethylamine, toluene, dimethyiformamide, and dimethyl sulfoxide were distilled from calcium hy

113 dride; W.N.A/.W-tetramethylethylenediamine was distilled from o-butyllithium. Reactions requiring

an inert atmosphere were run under argon. Analytical thin-layer chromatography was conducted

using EM Laboratories 0.25 mm thick precoated silica gel 60F-254 plates. Column

chromatography was performed over EM Laboratory silica gel (70-230 mesh). All organolithium

reagents were titrated prior to use with (±)-menthol using 1 ,1 0 phenanthroline as the indicator , 1 19

The abbreviations used in this thesis follow the guidelines suggested by the Journal of Organic

Chemistry, 120 The order of experimental procedures follow their order of appearance in the text.

o c h 3

1 5 9

i-(2-Methoxyethyl)-2,5-cyclohexadiene*i-carboxylic acid (159). To a

solution of 51.9 g (0.425 mol) of benzoic acid in 390 mL of THF in a 3-L round bottom flask

equipped with a dry ice-acetone condenser and cooled in a dry ice-acetone bath was condensed

1.5 L of ammonia. To the resulting mixture was added 6.70 g (0.967 mol) of lithium metal in 0.2 g

portions over 30 min. The reaction held a dark blue color for 25 min, and 75 mL (0.799 mol) of 2-

bromoethyl methyl ether was added via syringe. The pale yellow mixture was stirred in the cooling

bath for 2 h, and 55 g (1.0 mol) of ammonium chloride was carefully added. The ammonia was

allowed to evaporate overnight, and the residue was dissolved in 800 mL of water and extracted

with three 150-mL portions of ether. The aqueous layer was cooled in an ice-water bath and the

pH was adjusted to 1 by addition of 350 mL of concentrated aqueous hydrochloric acid. The

resulting pink mixture was extracted with five 250-mL portions of CH 2CI2 , the combined organic

layers were dried (MgS 0 4 ), and concentrated in vacuo to yield carboxylic acid 159 as an orange oil, 71.3 g (92%). The spectral data collected for this material were identical to that previously reported. This material was suitable for use in the next reaction .27 115

° ^ o c h 3

1 6 0

N-(i-[2‘meihoxyethyl]-2,5-cyclohexadlen-1-ylcarbonyl)pyrrolidine (160).

To a solution of 60.1 g (0.330 mol) of carboxylic acid 159 in 550 mL of N.N-dimethylformamide cooled in an ice-water bath was added sequentially 32 mL (0.383 mol) of pyrrolidine, 100 g (0.364 mol) of diphenylphosphoryl azide, and 90 mL (0.649 mol) of triethylamine. The cooling bath was removed after 1 h and the reaction mixture was stirred for 24 h followed by dilution with 1 L of ether. The mixture was washed with four 200-mL portions of brine, and the combined aqueous layers were extracted with 500 mL of ether. The combined organic layers were dried (MgSO^, concentrated in vacuo, and further concentrated at 45-50°C for 6 h at 1.2 mm Hg to yield 73.9 g

(95%) of amide 160 as a thick orange oil. The spectral data collected for this material were identical to that previously reported .2 7 This material was suitable for use in the next reaction.

OCHj

161

(±J-(1flal5R"l8ft*)-8-lodo-1-(2-methoxyethyl)-6-oxablcyclo[3.2.1 Joct-2- en-7-one (161). To a mixture of 77.0 g (0.328 mol) of amide 160 in 1000 mL of THF and 1000 mL of water was added 240 g (0.945 mol) of iodine in several portions. The dark mixture was stirred at room temperature for 24 h and then diluted with 2200 mL of ether. The mixture was transferred to a 6 -L Erlenmeyer flask and saturated aqueous sodium bisulfite was added until the iodine color was discharged. The mixture was split into four portions, and each aliquot was 116 washed with 200 mL of saturated aqueous sodium bisulfite, 200 mL of water, and 150 mL of brine. The combined aqueous layers were extracted with two 500-mL portions of ether. The combined organic layers were dried (MgS 0 4 ), concentrated in vacuo, and the solid residue was recrystallized from ether (8 mL ether per 1 g iodolactone) to yield 69.4 g (69%) of iodolactone

161 as a white solid collected in four crops: m.p. 70-73°C (lit2 7 75.5-77°C). The spectral data collected for this material were identical to that previously reported.

Sn(CHj,CH2CH2CH3)3

1 6 3

Allyl tri-n-butylstannane (163). Procedure A : 2 9 To a suspension of 15.1 g

(0.63 mol) of magnesium turnings in 190 mL of THF cooled in an ice-water bath was added a mixture of 35 mL (0.43 mol) of allyl chloride, 75 mL (0.28 mol) of tri-n-butylstannyl chloride, and

150 mL of THF dropwise over a 1 h 15 min period. The resulting gray slurry was heated to a gentle reflux for 3 h, cooled to rt, and then quenched by the addition of 250 mL of saturated aqueous ammonium chloride. The solid was removed by suction filtration and washed with ether. The resulting two layers were separated and the aqueous layer was extracted with two 200-mL portions of ether. The combined organic layers were washed with 150 mL of brine, dried

(MgSC>4 ), and concentrated in vacuo. The residue was distilled through a Vigreaux column to yield 90 g (98%) of allyl tri-n-butylstannane as a clear: bp 98-103°C (0.3 mm Hg). Spectral data were consistent with those reported in the literature .2 9

Procedure B : 121 To a suspension of 36.3 g (1.51 mol) of magnesium turnings in 500 mL of ether was added 80 mL (0.93 mol) of allyl bromide in 80 mL of ether over 2.5 h at a rate such that the exothermic reaction maintained a gentle reflux. The reaction was then heated at 35°C for

2 h, cooled to room temperature, and 159 mL (0.312 mol) of bis(tributylstannyl)oxide in 80 mL of ether was added dropwise over 2 h, again at a rate such that the exothermic reaction maintained a gentle reflux. The resulting mixture was heated to reflux for 1 .5 h and stirred at room temperature 117

(or 11 h. The reaction was quenched by the addition of 150 mL of saturated aqueous ammonium chloride dropwise, and the supernatant solution was decanted through glass wool. The residual solid was washed with 800 mL of hexanes, and the combined organic layers were washed with

100 mL brine, dried (MgSC> 4 ), and concentrated in vacuo. The residue was distilled through a

Vigreaux column to yield 174 g (85%) of allyl tri-n-butylstannane as a clear, colorless liquid: bp 96-

102°C (0.3 mm Hg).

o c h 3

1 6 2

(±)-(iff*,5S*,8/?*)-8-Allyl-l-(2-methoxyethyl)-6-oxablcyclo[3.2.l]oct-2- en-7-one (162). A mixture of 34.1 g (111 mmol) of iodolactone 161, 72 mL (233 mmol) of allyltri-n-butylstannane (163), and 200 mL of benzene was purged with Ar gas for 1 h, and 2.19 g

(13 mmol) of AIBN was added quickly in one portion. The reaction was heated at 80-82°C for 9 h and cooled to room temperature. To the mixture was added 250 mL of aqueous saturated potassium fluoride, the mixture was stirred overnight, and the resulting solid was removed by suction filtration. The tiltercake was washed with ether. The filtrate was washed with three 125-mL portions of brine, dried (MgS 0 4 ), and concentrated in vacuo to yield a pale yellow oil. This material was chromatographed over 150 g of flash silica gel topped with 1 0 g of activity grade I basic alumina (eluted with ethyl acetate-hexanes, 1:12) to yield 15.7 g (63%) of lactone 162 as a clear, colorless oil. The spectral data collected lor this material were identical to that previously reported .2 7 118 OH

OCH3

1 6 5

(±J-(lS*,5S*,6/?*)*6-Allyl-5-hydroxy-1-[(2-methoxyethyl]-2-cyclohexene-

1-carbox-p-anisidlde (165). To a solution of 75.1 g (610 mmol) of p-anisidine in 650 mL of

THF cooled in an ice-water bath was added 167 mL (434 mmol) of a 2.6 M solution of methyl

magnesium bromide in ether over a 45 min period. The mixture stirred in the cooling bath for 15

min, and 32.3 g (145 mmol) of lactone 162 in 150 mL of THF was added dropwise to the cooled

anion. The bath was removed and the reaction was stirred at room temperature for 6 h. The dark

mixture was poured over 200 g of ice, 600 mL of 10% aqueous hydrochloric acid was added, and

the mixture was extracted with 1 000 mL and 600 mL portions of dichloromethane. The combined

organic layers were washed with 300 mL of 10% aqueous hydrochloric acid, 300 mL of saturated

aqueous NaHCOs, and dried (MgSCXi). Concentration of the organic layers in vacuo and drying at

40°C and 0.15 mm Hg for 15 h yielded 46.8 g (94%) of alcohol 165 as a deep red oil. This

material gave spectral data identical to those of an authentic sample 2 7 and was used directly in the next reaction.

o c h 3

1 6 6

(±l'(1S*,5S*,6/?*)-6-Allyl-5-hydroxy-1*[(p-me!hoxyphenyl)carbamoyl]-2- cyclohexene-1-carbox-p-anlsidtde acetate (ester) (166). To a solution of 46.8 g (136 119

mmol) of crude alcohol 165 in 650 mL of dichloromethane cooled in an ice water bath was added

50 mL (360 mmol) of triethylamine, 59 mL (620 mmol) of acetic anhydride, and 1.94 g (15.9 mmol) of 4-/V,/V-dimethylaminopyridine. The cooling bath was removed after 30 min and the reaction

stirred at room temperature for 5 h. The dark mixture was poured into 300 mL of water and the

mixture was extracted with two 500-mL portions of dichloromethane. The combined organic layers were washed with two 300-mL portions of 10% aqueous hydrochloric acid, 400 mL of saturated aqueous sodium bicarbonate, dried (MgSCU), concentrated in vacuo, and dried at 0.2 mm Hg for 12 h to yield 52.4 g (99%) of acetate 166 as a red solid (mp 73-77°C). This material gave spectral data identical to those of an authentic sample 2 7 and was used directly in the next reaction.

OAc

1 5 6

(±M4a/?*,5S*,8a/?*)-3,4,4a,5,6,8a-Hexahydro-5-hydroxy-8a-(2-methoxy- eth yl)-2-(p-m etho xyphenyl)-l(2/V )-isoqulnollnone 5-acetate (156). To a solution of

43.0 g (111 mmol) of acetate 166 in 1000 mL of /-butyl alcohol, 130 mL of water, and 180 mL of

THF cooled in an ice/water bath was added 1.0 g (3.9 mmol) of osmium tetroxide in 50 mL of water. To the resulting solution was added 49.2 g (230 mmol) of sodium periodate in 2 g portions over a 4.5 h period. The reaction was stirred in the cooling bath for 6 h, and was split into four portions, each of which was extracted with 600 and 300 mL portions of dichloromethane. The combined organic layers were dried (MgSO^, concentrated in vacuo, and dried at 0.2 mm Hg and

35°C for 4 h to yield a mixture of aldehyde 167 and hemiamidal 168 as a thick brown oil. 120

A heterogeneous mixture of the aldehyde 167 and hemiamidal 168, 7.7 g of Dowex 50 x 8 resin (rinsed with methanol), 800 mL of methanol, and 400 mL of tetrahydrofuran was stirred with heating at 45°C for 20 h. The dark mixture was cooled to room temperature, the resin was removed by filtration, rinsed with methanol, and the filtrate was concentrated in vacuo and dried at

0.1 mm Hg for 3 h to yield hemiamidal 169 as a brown foam.

To the crude hemiamidal 169 and 24.5 g (408 mmol) of sodium cyanoborohydride in 200 mL of dichloromethane cooled in an ice water bath was added 80 mL (1.0 mol) of trifluoroacetic acid over a 1 h period. The cooling bath was removed and the reaction stirred for 1 h. The mixture was cooled again in an ice water bath, and 150 mL of 10% aqueous hydrochloric acid was added dropwise followed by the dropwise addition of 3 N aqueous NaOH (approximately 450 mL) until the aqueous layer was basic. The mixture was extracted with two 700-mL portions of dichloromethane, the combined organic layers were washed with 300 mL of 10% aqueous hydrochloric acid, 300 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo. The residue was chromatographed over 350 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:3, to ethyl acetate-hexanes, 1:1) to yield 27.5 g (67%) of bicyclic lactam 156, (mp 69-74°C (lit 2 7 82-83°C)J. This material gave spectral data identical to those of an authentic sample.

OAc

1 7 1

(±><4a/?*,5S’\8afl*)-3,4,4a,5,6,8a-Hexahydro-5-hydroxy-8a-(2-hydroxyethyl)-2-(p-methoxyphenyl)-l(2H)-isoqulnollnone 5-acetate (171). To a solution of

17.1 g (46.0 mmol) of lactam 156 in 1 L of dichloromethane cooled in a dry ice-acetone bath was 121 added 100 mL (100 mmol) o( a 1.0 M solution of boron tribromide in dichloromethane over a 20 min period. The resulting mixture stirred in the cooling bath for 30 min, was stirred in a bath at -18 to -15°C for 1 h 5 min, and then 24 mL of ether was added. The mixture was cannulated, with swirling, into a mixture of 400 g ice and 500 mL of saturated sodium bicarbonate over 10 min. The mixture was transferred to a separatory funnel, shaken, and the layers were separated. The aqueous layer was extracted with three 500-mL portions of dichloromethane. The combined organic layers were dried (MgSCU) and concentrated in vacuo to yield and orange residue. This material was crystallized from 100 mL ethyl acetate and 30 mL hexanes to provided 7.97 g of alcohol 171 as an off-white solid. The filtrate was concentrated in vacuo and chromatographed over 70 g silica gel (eluted with ethyl acetate-hexanes, 1:1, to ethyl acetate-hexanes, 3:1) to yield an additional 3.34 g of alcohol 171 as an off-white solid (11.31 g total, 69%): mp = 115-125°C (lit

135.5-137°C) The spectral data for this material was identical to that previously reported.27

OAc

1 7 0

(±)-(4a/7*,5S*,8afl*)-3,4,4a,5,6,-Hexahydro-5-hydroxy-2-(p-meihoxy- phenyl)-i-oxo-8a(1H)-isoqulnolineacetaldehyde (170). To a solution of 4.0 mL (46 mmol) of oxalyl chloride in 200 mL of dichloromethane cooled in a dry ice-acetone bath was added

7.1 mL (100 mmol) of dimethyl sulfoxide. The resulting mixture stirred in the cooling bath for 35 min, and to the reaction was added 8.31 g (23.1 mmol) of alcohol 171 in 150 mL of dichloromethane dropwise via an addition funnel. The reaction was stirred in the cooling bath for

1 h 30 min, followed by addition of 16 mL (120 mmol) of triethylamine. The reaction was stirred in the cooling bath for 1 h, warmed to room temperature, was diluted with 1 L of dichloromethane. 122

The organic layer was washed with 100 mL of 10% aqueous hydrochloric acid, 100 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo to yield a thick yellow oil. This material was chromatographed over 100 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:2 (2 L) to ethyl acetate-hexanes, 1:1) to yield 7.06 g ( 8 6 %) the desired aldehyde 170 as a white foam. Upon sitting, the aldehyde slowly crystallized to a pale yellow solid. A portion of this material was recrystallized from ethyl acetate-hexanes (1:1) to yield the aldehyde as white needles (mp 104-105.5°C). This material exhibited spectral data identical to that previously reported for aldehyde 170.27

Anal, calcd. for C2 0 H2 3 NO4 : C, 67.21; H, 6.49. Found C, 67.22; H, 6.47.

1 8 5

2-(C hlorom ethyl)tetrahydropyran (185).39 To a mixture of 90 mL (0.80 mol) of alcohol 184 in 160 mL of pyridine was added 71 mL (0.97 mol) of thionyl chloride dropwise over 8 h such that the reaction temperature did not exceed 45°C. The resulting brown mixture was stirred at room temperature for 48 h and was extracted with six 250-mL portions of ether. The combined organic layers were washed with two 250-mL portions of water, 250 mL of saturated aqueous sodium bicarbonate, 250 mL of brine, and was dried (MgS 0 4 ). Concentration in vacuo afforded a yellow oil that was distilled through a Vigreaux column to yield 64.6 g (60%) of the desired chloride as a clear colorless oil: bp 85-86°C at 70 mm Hg (lit 3 9 55.0-55.5°C at 6 mm Hg).

1 8 4

5-H exyn -l-o l (184).39 To an orange mixture of 400 mg (1.0 mmol) of ferric nitrate nonahydrate in 2000 mL of liquid ammonia removed from a dry ice/acetone bath was added 41.0 123

g (1.78 mol) of sodium metal in small pieces over 2 h. Cooling was applied to the mixture as

necessary using the dry ice/acetone bath during the addition. Upon completion of addition, the

mixture was stirred without cooling for 45 min, and to the resulting gray slurry was added 64.5 g

(450 mmol) of chloride 185 via syringe over 10 min. The resulting mixture was stirred at -33°C for

4.5 h. The reaction was quenched by the slow addition of 105 g (2.1 mol) of solid ammonium chloride, the ammonia was evaporated overnight, and the residue was stirred with 2 L of ether for

24 h. The resulting gray solid was removed by filtration through a Buchner funnel, the filtrate was concentrated in vacuo, and the residue was distilled through a Vigreaux column to provide 36.22 g (82%) of 5-hexyn-1-ol (184) as a clear colorless oil: bp 98-101°C at 70 mm Hg (lit 3 9 53°C at 1.2 mmHg).

OTHP

1 8 6

(±)-(Tetrahydro-2H -pyran-2-yl)oxy 5-hexynyl acetal (186). To a mixture of

20.06 g (0.204 mol) of 5-hexyn-1-ol and 17.79 g (0.211 mol) of 3,4-dihydro-2H-pyran was added

55 mg (0.29 mmol) of p-toluenesulfonic acid. The mixture became warm to the touch, was cooled in an ice/water bath for 10 min, the bath was removed, and the reaction was stirred for 3 h. The mixture was chromatographed over 300 g of flash silica gel (eluted with ethyl acetate-hexanes,

1:50, to ethyl acetate-hexanes, 1:33) to yield 35.57 g (96%) of alkyne 186 as a clear, colorless oil. The spectral data for this material were identical to that previously reported .27

OTHP

181

(±.)-6-lodo-(tetrahydro-2W-pyran-2-yl)oxy 5-hexynyl acetal (181). To a solution of 17.2 g (94.5 mmol) of alkyne 186 in 400 mL of THF cooled in a dry ice-acetone bath was added 110 mL (165 mmol) of 1.5 M solution of n-butyllithium in hexanes dropwise over 30 124

min. The mixture was stirred in the cooling bath for 45 min and 51.2 g (2 0 2 mmol) of iodine in 150

mL of THF was added dropwise over 30-35 min. The resulting red mixture was stirred in the

cooling bath for 5 h, and was then poured into 500 mL of saturated sodium bisulfite. The mixture

was extracted with two 600-mL portions of ether, and the combined organic layers were washed

with 200 mL of saturated aqueous sodium bisulfite, 400 mL of water, two 300-mL portions of

saturated sodium bicarbonate, and 400 mL of brine. The organic layer was dried (MgS 0 4 ) and

concentrated in vacuo. The residue was chromatographed over 200 g of flash silica gel (eluted

with ethyl acetate-hexanes, 1:50) to yield 25.5 g ( 8 8 %) of alkyne 181 as a yellow oil: IR (neat)

2940, 1453 c m 1: 1H NMR (CDCI 3 , 300 MHz) 8 1.41-1.89 (m, 10H, CH 2 manifold), 2.39 (t, J= 6.9

Hz, 2 H, CCCH2), 3 37 (dt, J= 9.8, 6.1 Hz, 1H, OCH2), 3.48 (m, 1 H, OCH2), 3.72 (dt, J = 9.8, 6.2

Hz, 1H, OCH2), 3 83 (m, 1H, OCH2). 4.55 (m, 1 H, OCHO): 13C NMR (CDCI3 , 75.5 MHz) 8 -7.2 (s),

19.5 (t), 20.5 (t), 25.3 (t), 25.4 (t), 28.7 (t), 30.6 (t), 62.2 (t), 66.7 (t), 94,3 (s). 98.7 (d); exact mass

cacld for C1 1H1 7IO2 m/2 308.0274, found m/2 308.0262.

OTHP

1 82

(±J-Z-6-lodo-(tetrahydro-2H-pyran-2-yl)oxy 5-hexenyl acetal (182). To a

mixture of 1.09 g (3.55 mmol) of iodoalkyne 181 in 7 mL of methanol was added 0.76 g (3.90

mmol) of dipotassium azodicarboxylate (DAPA), and to the resulting yellow mixture was added

240 pL of glacial acetic acid dropwise over 1 h 30 min. The mixture was stirred at room temperature for 5 h, and to the mixture was added 215 mg (1.11 mmol) of DAPA and 50 pL of glacial acetic acid over 40 min. The reaction was stirred for 1 2 h, and to the mixture was added 505 mg (2.6 mmol) of DAPA and 150 pi of glacial acetic acid over 1.5 h. The reaction was stirred for 5 h, and to the mixture added 550 mg (2.84 mmol) of DAPA and 150 pi of glacial acetic acid over 45 min. The reaction was stirred for 19 h, and was poured into 100 mL of saturated aqueous sodium bicarbonate. The mixture was extracted with two 100-mL portions of dichloromethane, the 125

combined organic layers were dried (MgS 0 4 ), and concentrated in vacuo. The residue was

chromatographed over 100 g of silica gel (eluted with ethyl acetate-hexanes, 1:99} to yield 483

mg (44%) of the desired olefin as a pale yellow oil: IR (neat) 2938, 2866, 1453,1439 cm'1, 1H-

NMR (CDCI3 , 300 MHz) 6 1.46-1.83 (m, 10H, CH 2 manifold), 2.16 (m, 2H, =CHCd2 ), 3.39 (td, J =

1 0 , 6 Hz, 1 H, CHOCtb). 3-48 (m, 1 H, =C(CH2 )3 Ctl2 ). 3.74 (td, J = 1 0 , 6 Hz, 1 H, CHOChte), 3.85

(m, 1 H, -C(CH 2 )3 Ctl2 ). 4.57 (m, 1H, OCHO), 6.15 (m, 2 H. CH=CH); 13C NMR (CDCI3 , 75.5 MHz)

6 19.5 (t), 24.6 (t), 25.4 (t), 29.1 (t), 30.6 (t), 34.4 (t), 62.2 (t), 67.1 (t), 82.3

exact mass calcd. for m/z C 1 1 H 1 9 IO2 310.0430. found m/z 310.0408. Continued elution

provided 402 mg (37%) of recovered starting material.

OAc OAc

OTHP OTHP

173 174

(±H4a/?*,55*,8aff*)-3,4,4a,5,6,8a-Hexahydro-5-hydroxy-8a-{(2$*)-2-hy-

droxy-8-[tetrahydro-2H-pyran-2-yl)oxy]-3-octynol)-2-(p-methoxyphenyl)-l(2H)-

Isoqulnolinone 5-acetate (173) and (±_)-(4a/?*,5S*,8af?*)-3,4,4a,5,6,8a-

Hexahydro-5-hydroxy-8a-{(2f7*)-2-hydroxy-8-[tetrahydro-2H-pyran-2-yl)oxy]-3-

octynol)-2-(p-methoxyphenyl)-1(2H)-isoqulnollnone 5-acetate (174). In a flask

equipped with an Ar inlet and addition funnel, a mixture of 11.4 g (92.7 mmol) of chromium (II) chloride and 140 mg ( 1.1 mmol) of nickel (II) chloride was flamed dried under vacuum. To the solid

mixture was added 200 mL of THF. The resulting gray-green slurry was cooled to -25°C and a mixture of 6.85 g (19.2 mmol) of aldehyde 170 and 16.3 g (52.9 mmol) of iodide 181 (dried by azeotropic removal of water with toluene) in 125 mL of THF was added dropwise over a 25 min period. The resulting mixture was stirred in the bath as it slowly warmed to 10°C over 2.5 h, and 126

the reaction color changed from green-gray to a rusty orange-brown. After 1 h at 10°C, the

mixture was poured into 300 mL ol water and extracted with three 500-mL portions of ether. The

combined organic layers were washed with two 100-mL portions of brine, dried (MgSO^, and

concentrated in vacuo. The residual orange oil was chromatographed over 300 g of flash silica gel

(eluted sequentially with ethyl acetate-hexanes, 1:10 (1 L) to ethyl acetate-hexanes, 1:3 (2 L), to

ethyl acetate-hexanes, 1:1 (until completion)) to yield unreacted iodoalkyne and 8.99 g (67%) of a

1.3:1 mixture of the diastereomeric alcohols 173:174 (ratio determined by integration of OCH 3

peak in *H-NMR) a very viscous orange oil. This mixture exhibited spectral data identical to those

previously reported .2 7

OAc OAc

OTHP OTHP

187 188

(±J-(4a/7*,5S*t8aR*)-3,4,4at5)6,8a-Hexahydro-5-hydroxy-8a-{(2S*)-2-hy-

droxy-8-[tetrahydro-2H-pyran-2-yl)oxy]-3-octenol)-2-(p-methoxyphenyl)-l(2M)-

Isoquinolinone 5-acetate (187) and (±J-(4a/?*,5S*,8afl*)-3,4,4a,5,6,8a-

Hexahydro-5-hydroxy-8a-{(2/7*)-2-hydroxy-8-[tetrahydro-2/y-pyran-2-yl)oxy]-3- octenol)-2-(p-methoxyphenyl)-1(2H)-isoqulnollnone 5-acetate (188). From

Alcohols 173 and 174: To a mixture 9.25 g (17.2 mmol) of alcohols 173 and 174 in 95 mL of pyridine was added 1.00 g of 5% palladium on barium sulfate. The heterogeneous mixture was evacuated, purged with hydrogen gas three times, and hydrogenated at room temperature and one atmosphere of hydrogen until 425 mL (17.3 mmol) of hydrogen gas was consumed. The black mixture was evacuated, opened to the atmosphere, passed through a pad of Celite and 127

rinsed with 700 mL of dichloromethane. The filtrate was further diluted to 1 L with dichloromethane, washed with three 125-mL portions of 10% aqueous hydrochloric acid, 150 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo to yield a thick orange oil. This material was chromatographed twice over 800 g and 400 g of flash silica gel, respectively (eluted with ethyl acetate-hexanes, 1:3, until elution of first isomer) to yield 2.24 g

(24%) of a yellow foam identified as alcohol 188. This material exhibited spectra data identical to that previously reported. Further elution of the column yielded 426 mg (5%) of a mixture of alcohols 188 and 187, and elution with ethyl acetate-hexanes, 1:1, yielded 6.35 g (69%) of alcohol 187 as a yellow foam, with identical spectral data to that previously reported .2 7

From Aldehyde 170: In a flask equipped with an Ar inlet and addition funnel, a mixture of 495 mg (4.03 mmol) of chromium (II) chloride and 4 mg (0.031 mmol) of nickel (II) chloride was flame-dried under vacuum. To the solid mixture was added 2 mL of THF at room temperature. To the resulting green-gray slurry was added 92 mg (0.26 mmol) of aldehyde 170 and 428 mg (1.38 mmol) of vinyl iodide 182 (dried by azeotropic removal of water with toluene) in 5 mL of THF dropwise at room temperature. The resulting slurry was stirred at room temperature for 48 h, over which time the reaction became an orange-brown color. The reaction was poured into 10 mL of water and was extracted with four 50-mL portions of ether. The combined organic layers were washed with 30 mL of water, 40 mL of brine, dried (MgSC> 4 ), and concentrated in vacuo to yield a yellow oil. This material was chromatographed over 60 g of silica gel (eluted with ethyl acetate- hexanes, 1:2) to yield 40 mg (29%) of alcohol 188 as a clear foam. Continued elution provided

71 mg (51%) of alcohol 187 as a clear foam. These two isomeric alcohols exhibited spectral data identical to those previously reported for the respective compounds .2 7 128 OAc

OTHP

1 8 9

ferf-Butyl (±J-[(lR* 12Z)-l-{[(4af7*,SS*,8a/7*)-2,3,4,4a,5,6-hexahydro-5-

hydroxy-2-(p-methoxyphenyl)-l-oxo-8a(1H)-isoqulnolyl]methyl)-7-[(tetrahydro-

2 H -pyran- 2 -yl)oxy}- 2 -heptenyl]{( 2 -(trlmethylsllyl)ethyl]sulfonyl)carbamate, acetate (ester) (189). To a solution of 6.32 g (22.5 mmol) of sulfonamide 6 6 . 6.27 g (23.8 mmol) of triphenylphosphine, and 350 mL of THF cooled in an ice-water bath was added 6.28 g

(11.6 mmol) of alcohol 187 in 50 mL of THF over a 10 min period. The resulting yellow solution was stirred for 5 min, and 4.0 mL (25.3 mmol) of diethyl azodicarboxylate was added dropwise via syringe. The reaction was stirred in the cooling bath for 10 min, at room temperature for 8 h, and was concentrated in vacuo. Solid impurities were removed by crystallization from ethyl acetate- hexanes and filtration. The filtrate was concentrated in vacuo and the residue was chromatographed over 400 g of activity grade II basic alumina (eluted with ethyl acetate-hexanes,

1:6 ) to yield 8.69 g (93%) of acyl sulfonamide 189 as a white foam. Spectral data for this material were identical to that reported previously .27

OAc

OTHP

195 129

(±M4a/?*,5S*,8a#?*)-8a-[(2S*,3Z)-2-Azldo-8-[(tetrahydro-2H'pyran-2- yl)oxy]*3-octenyl]-3,4,4a,5,6,8a-hexahydro-5-hydroxy-2-(p-methoxyphenyl)- l(2H)-lsoqulnolone acetate (ester) (195). To a solution of 223 mg (0.41 mmol) of alcohol

188 in 4 mL of toluene was added 256 mg (200 pL, 0.93 mmol) of diphenylphosphorylazide and

204 mg (220 pL.1.34 mmol) of 1,8-diazabicyclo[5.4.0]undec-7-ene via syringe. The reaction mixture was stirred at room temperature for 24 h and then diluted with 80 mL of dichloromethane.

The mixture was washed with 25 mL of 10% aqueous HCI, 25 mL of saturated aqueous NaHCOa, dried (MgSC> 4 ), and concentrated in vacuo to yield an orange oil. This material was chromatographed over 15 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:5) to yield 165 mg (71%) of azide 195 as an oil: IR (neat) 2095,1730, 1649, cm'1; 1H NMR (CDCI 3 , 300 MHz)* 5

1.35-2.00 (m, 14H, CH 2 manifold), 2.09 (s, 3H, COCH3), 2 .2 1 (m, 2 H, CHOAcCfcb and

CH2 CHN3 ), 2.37 (dt, J= 12.0, 3.0 Hz, 1H, CHOAcCtfc), 2.57 (dt, J= 12.4, 0.6 Hz, 1H. CH), 3.01

(dm, J = 13.0 Hz, 1 H, CU2 CHN 3 ), 3.37 (dt, J = 9.8, 6.5 Hz, 1 H, NCH2), 3.48 (m, 2H,

CtfcOCHOCtte). 3.73 (m, 3H, CtfcOCHOCtb ( 1 H), CHN3 , and NCH2), 3.80 (s, 3H, OCH3 ), 3.84

(m, 1 H, ChbOCHOCfcb), 4.55 (m, 1 H, OCHO), 5.19 (m, 1 H, CHOAc), 5.63 (m, 2 H, NCHCH=Ctl),

5.68 (m, 1 H, CHOCH2 CH=), 5.81 (dm, J = 9.0 Hz, 1 H CHOCH2 CH=CH), 6.90 (d, J = 9.0 Hz, 2 H,

ArH), 7.11 (d, J = 9.0 Hz, 2H, ArH); 13C NMR (CDCI3 , 75.5 MHz) 6 19.5 (t), 19.7 (t), 21.2 (q), 22.5

(t), 25.3 (t). 26.6 (t), 29.2 (t), 30.6 (t). 34.2 (t), 36.7 (d), 41.4 (t), 50.5 (S), 51.0 (t), 55.3 (q), 62.1 (t).

65.1 (d), 67.1 (t), 68.1 (d), 98.7 (d), 114.3 (d), 123.3 (d), 127.1 (d), 130.4 (d), 131.6 (d), 132.6 (d),

136.2 (s), 158.0 (s), 170.1 (s), 171.6 (s); exact mass calcd for C3 iH 4 2 N4 C>6 m/z 566.3107; calcd for M+-N2 m/z 538.3045, found m/z 538.3050. 130 OAc

H,CO'

1 9 6

(±J-(4afl*,5S*,8af?*)-8a-[(2S*,3Z)-2-AmIno-8-l(tetrahydro-2H-pyran-2- yl)oxy]-3-octenyl]-3,4,4a,5,6,8a-hexahydro-5-hydroxy-2*(p*methoxyphenyl)-

1(2W)-isoqulnolone acetate (ester) (196). To a solution of 78 mg (0.14 mmol) of azide

195 in 3 mL of THF was added 80 pL (4.4 mmol) of water and 61 mg (0.23 mmol) of triphenylphosphine. The resulting mixture was stirred at room temperature for 36 h and was concentrated in vacuo. The residue was chromatographed over 5 g of silica gel (eluted with ethyl acetate (50 mL) to ethyl acetate-methanol, 10:1 (970 mL) to ethyl acetate-methanol-triethylamine,

15:5:1) to yield 53 mg (70%) of amine 196 as a clear colorless oil: IR (neat) 3443, 1729, 1644 cm’ 1; 1H NMR (CDCI3 , 250 MHz)* 5 1.32-2.00 (m, 14H, CH2 manifold), 2.07 (s. 3H, COCH3 ),

2.15 (dd, J= 13.7, 8.0 Hz, 1 H. NCHCtfc), 2.15 (m, 1 H, CHOCtb). 2.36 (dt, J= 17.2, 5.5 Hz, 1 H,

CHOCH2 ). 2.52 (dm, J= 12.0 Hz, 1H, CfciCHOAc), 2.84 (br s, 2 H, NH2), 2.93 (dd, J= 13.8, 4.5 Hz,

1 H, NCHCtd2 ). 3.33 (m, 2 H, ChbOCHOCtb), 3.48 (m, 2 H, CHN and CONCH2 (1 H)), 3.71 (m, 2 H,

ChfcOCHOCtte), 3.77 (s, 3H, OCH3 ), 3.83 (m. 1H, CONCH2), 4.53 (br s, 1H, OCHO), 5.15 (m, 1 H,

CHOAc), 5.54 (m, 2H, NCHChhCtl), 5.63 (ddd, J= 10.0, 5.2, 2.1 Hz, 1 H, CCH=CJd). 5.78 (d, J =

10.0 Hz, 1H, CCH=), 6.87 (d, J= 8.9 Hz, 2H, ArH), 7.11 (d. J= 8.9 Hz, 2H, ArH); 13C NMR (CDCI 3 )

6 19.6 (t), 19.8 (t), 21.2 (q), 22.7 (t), 25.3 (t), 26.6 (t), 29.6 (t), 30.6 (t), 37.0 (d), 37.3 (t), 41.4 (t),

50.6 (t). 51.4 (s), 53.7 (d), 55.3 (q), 62.2 (t), 67.3 (t), 68.5 (d). 98.7 (d), 114.3 (d), 122.9 (d), 124.8

(d), 127.1 (d), 131.8 (d). 136.3 (s), 139.4 (d), 158.0 (s), 170.2 (s), 171.9 (s); exact mass calcd for

C3 1 H4 4 N2 O 6 m/z 540.3201, found m/z 540.3021. 131 OAc

tBoc

H,CO' OTHP

1 9 7

fert-Butyl (±J-[(lR*J2Z)-1-[t4aS‘,5R*,8aS*)-2,3,4,4a,5,6-hexahydro-5- hydroxy-2-(p-methoxyphenyl)-i-oxo-8a(1H)-lsoqulnolyl]methyl]-7-[(tetrahydro-

2H-pyran-2-yl)oxy]-2-heptenyl]carbamate, acetate (ester) (197). To a solution of 18 mg (0.033 mmol) of amine 198 in 3 mL of dimethyl sulfoxide was added 30 pL (0.21 mmol) of t- butylazidoformate via syringe. The reaction was stirred at room temperature for 10 h, and was concentrated under high vacuum. The residue was chromatographed over 4 g of silica gel (eluted with ethyl acetate-hexanes, 1:1) to yield 7 mg (33%) of carbamate 197 as a clear, colorless oil: IR

(neat) 3320, 1730, 1704, 1650 cm'1; 1H NMR (CDCI 3 , 250 MHz) 4 6 1.43 (s, 9H, C(CH3)3), 1.46-

1.91 (m, 13H, CH2 manifold), 2.09 (s, 3H, COCH3), 2.14 (m, 2H, NCHCU 2 and CHOAcCtte). 2.35

(td, J = 11.2, 5.5 Hz, 1H, CHOAcCfcte). 2.51 (m, 1H, CfcLCHOAc). 2.93 (dd, J= 15.1, 4.2 Hz, 1 H,

NCHCtfc), 3-35 (dt, J = 9.7, 6.6 Hz, 1H. NCH2), 3.42-3.55 (m, 2H, CtbOCHOCfcfe), 3.70 (dt, J =

9.7, 6.6 Hz, 1H, NCH2), 3.70-3.92 (m, 2H, CtfcOCHOCtk), 3.80 (s, 3H, OCH3), 4.01 (br s, 1H,

NH), 4.55 (m, 2H, OCHO and CHN), 4.58 (m, 1H, NCH), 5.18 (m, 1H, CHOAc), 5.49 (m, 2H,

CHCU=Cbl). 5.64 (dm. J= 10.1 Hz, 1H, CCH=Ctl). 5.81 (d, J= 10.3 Hz, 1H, CCH=CH), 6.90 (d, J=

8.7 Hz, 2H, ArH), 7.17 (d, J = 8.6 Hz, 2H, ArH); 13C NMR (CDCI3, 75.5 MHz) 8 [19.54 (t), 19.58(t)],

19.9 (t). 21.2 (q). [22.44(t), 22.47(1)], 25.4 (t), 26.6 (t), 28.3 (q), [29.4 (t), 29.6 (t)]. 30.6 (t), 34.9 (t).

36.8 (d), 41.3 (t). [50.5 (t), 50.7 (t)], 51.6 (S), 52.9 (d), 55.3 (q), [62.19 (t). 62.25 (t)J, 67.2 (t), 68.6

(d), 77.6 (S), [98.74 (d), 98.79 (d)], 114.4 (d), 122.8 (d), 126.2 (d). 127,4 (d), 131.9 (d), 135.7 (d),

136.4 (s), 155.1 (s), 158.1 (s), 170.1 (s), 171.8 (s); exact mass calcd for C3 6H5 2 N 2 0 a m/z

640.3727, found m/z 640.3716. 132 OAc

x tBoc

OTHP

1 9 6

rerf-Butyl (±.H(l#?\2Z)-l-[l4a/?*,5S*,8aH*)-2,3,4,4a,5,6-hexahydro-5- hydroxy-2-(p-melhoxyphenyl)-i-oxo-8a(1H)-lsoqulnolyl]methyl]-7-[(tetrahydro-

2H -pyran-2-yl)oxy]-2-heptenyl]carbam ate, acetate (ester) (196). To a solution of 26 mg (0.032 mmol) of acylsulfonamide 189 in 2 mL of THF at room temperature was added 53 pL

(0.053 mmol) of 1.0 M tetra-n-butylammonium fluoride in THF. The resulting mixture stirred at room temperature for 3.5 h, was diluted with 50 mL of ether, and washed with 20 mL of water and

20 mL of brine. The organic layer was dried (MgS 0 4 ) and concentrated in vacuo to yield a pale yellow oil. This material was chromatographed over 4 g of silica gel (eluted with ethyl acetate- hexanes, 1:1) to yield 17 mg (83%) of carbamate 196 as a clear, colorless oil: IR (neat) 3341,

1730, 1711, 1648 cm'1: 1H NMR (CDCI3, 250 MHz)* 81.42 (s, 9H, C(CH3)3), 1.46-1.93 (m, 14H,

CH2 manifold), 2.09 (s, 3H, COCH3)), 2.10-2.27 (m, 2H, CHOAcCtb and NCHCJdfc). 2.28 (dd, J =

17.4, 5.5 Hz, 1H, NCHCH2 ), 2.40 (dt, J= 17.4, 5.5 Hz, 1H. CHOAcChte), 2.65 (m, 1H, CH), 3.35

(dt, J= 9.6, 6.5 Hz, 1H, NCH2), 3.47 (m, 2H, C tbO C H O C tb), 3 60-3.89 (m, 3H, NCH2 (1H), and

CtfcOCHOCtb) (2H)), 3.80 (s, 3H, OCH3), 4.55 (m, 1H, OCHO), 4.61 (m, 1H, CHN), 4.85 (brs,

1H, NH), 5.22 (m, 2H, CHOAc and NCCH=), 5.41 (dt, J= 10.7, 7.3 Hz, 1H. NCCH=CH), 5.60 (ddd,

J= 10.1, 4.9, 1.8 Hz. 1H, CCH=CM), 5.70 (d, J= 10.1 Hz, 1H, CCH=), 6.90 (d, J= 8.9 Hz, 2H,

ArH), 7.19 (d, J= 8.7 Hz, 2H, ArH); 13C NMR (CDCI3, 75.5 MHz) 8 19.5 (t). 20.0 (t), 21.1 (q), 25.4

(t), 26.3 (t), 26.5 (t), 27.4 (t), 28.4 (q), 29.3 (t), 30.6 (t), 37.5 (d), 42.8 (t), 45.7 (d). 50.6 (s), 50.8 (t),

55.3 (q), 62.1 (t), 67.4 (t), 68.7 (d), 77.3 (s), 98.6 (d), 114.3 (d), 122.8 (d), 127.4 (d), 130.9 (d),

131.5 (d), 1 3 2 .2 (d), 136.7 (s), 154.6 (s), 158.1 (s), 170.1 (s), 172.1 (s): exact mass calcd for

C36 H5 2N2Og m/z 640.3727, found m/z 640.3729. 1 3 3

OAc

° ' I OCHa

2 0 4

(±>(4a#?*,5S*,8afl*)-3,4,4a.5.6.8a-Hexahydro*5-hydroxy-8a-(2-methoxy-

ethyl)-1(2H)-lsoquinollne acetate (ester) (204). To a solution of 289 mg (0.77 mmol) of

amide 156 in 8 mL of acetonitrile cooled in an ice/water bath was added 1.27 g (2.3 mmol) of eerie

ammonium nitrate in 10 mL of water over 20 min. The resulting orange solution was stirred at 0°C

for 25 min and then poured into 80 mL of water. The mixture was extracted with four 75 mL

portions of ethyl acetate, and the combined organic layers were washed with two 50-mL portions

of saturated sodium bicarbonate. The combined aqueous layers were extracted with 40 mL of

ethyl acetate, and the combined organic portions were washed with two 50-mL portions of

saturated aqueous sodium bisulfite. The aqueous layers were combined and extracted with 50

mL of ethyl acetate. The combined organic layers were washed with 50 mL of saturated aqueous

sodium bicarbonate, 50 mL of brine, dried (MgSCXO, and concentrated in vacuo. The residue was

chromatographed over 15 g of silica gel (eluted with acetone-hexanes, 1:3) to yield 121 mg (59%)

of amide 204 as an off-white solid: mp 124-125°C; IR (neat) 3513, 1732, 1643 cm-1; 1H NMR

(CDCI3 , 250 MHz) 5 1.64-1.92 (m, 3H, CCH 2 and NCH2 Chl2 (1 H)). 2.06 (S, 3H, COCH3 ), 2.07 (m,

1 H, NCH2 Ctl2 ). 2.31-2.50 (m, 3H, =CHCU2 and CH), 3.28 (S, 3H, OCH3 ), 3.29 (m, 2 H, NCH2),

3.48 (t. J= 6.7 Hz, 2H, OCH2), 5.15 (ddd, J= 10.2, 6.1, 4.1 Hz, 1H, CHOAc), 5.59 (ddd, */= 10.0,

5.2, 2.0 Hz, 1H, CCH=), 5.68 (dt, J = 10.1, 1.3 Hz, 1H, CCH=Ct|), 6.19 (br s, 1H, NH); 13C NMR

(COCI3 , 75.5 MHz) 5 19.8 (t), 21.2 (q), 26.7 (t), 36.9 (t), 37.7 (d), 41.2 (t), 49.2 (s), 58.5 (q), 6 8 .8 134

(d), 69.3 (t), 123.0 (d), 131.6 (d), 170.3 (s). 174.4 (s); exact mass calcd. for C1 4 H2 1 NO 4 m/z

267.1471, found m/z 267.1477.

OAc

2 0 8

(i)-(4aW*,5S*,8a/7*)-3I4,4a,5,6 18a-Hexahydro-5*hydroxy-8a-(2-hydroxyethyl)-1(2H)-isoqulnoline acetate (ester) (208). To a solution of 77 mg (0.29 mmol) of amide 204 in 6 mL of dichloromethane cooled in a dry ice-acetone bath was added 1.2 mL (1.2 mmol) of 1.0 M boron tribromide in dichloromethane. The reaction was stirred at -78°C for 15 min, at 0°C for 1 h 15 min, and was poured into 40 mL of saturated aqueous sodium bicarbonate. The mixture was extracted with two 75-mL portions of dichloromethane, dried (MgS 0 4 ), and chromatographed over 8 g of silica gel (eluted with ethyl acetate) to yield 27 mg (35%) of unreacted starting material. Continued elution with ethyl acetate-methanol, 1 0 :1 , yielded 35 mg

(48%) of the alcohol 204 as an off-white solid: IR (CH 2 CI2 ) 3395, 1718,1654,1648, 1636 cm-1:

1H-NMR (CDCI3 , 300 MHz) 5 1.64-1.71, 1.82-1.91, 2.08-2.28 (m, 7H, NCH 2 CH2 C]± OCHCfcfc,

CCH2), 2.05 (S, 3H, CH3 ), 3.28 (m, 2H, CH 2 0), 3.70 (m, 2H, NCH2), 3.87 (br.s, 1H, OH), 5.10

(ddd, J= 10, 6 , 4 Hz, 1 H, CHO), 5.64 (ddd, J = 10, 4.5, 1.5 Hz, 1 H, CH=CtlCH2 ). 5.72 (dd, J= 10,

1 Hz, 1H, CCJd=CH), 6.81 (S, 1 H, NH); 1 3C-NMR (CDCI3 , 75.5 MHz) 6 19.3 (t), 21.1 (q), 26.7 (t),

38.9 (d), 40.5 (t), 41.0 (t), 49.2 (s). 58.9 (1), 68.5 (d), 124.1 (d), 129.9 (d), 170.4 (s), 175.8 (s); exact mass calcd. for C1 3 H1 9NO4 m/z 253.1314, found m/z 253.1337. 135 OAc

n o

o c h 3

2 0 9

ferf-Butyl (±J-(4aR*,5S*,8aff‘)-3,4,4a,5,6,8a-hexahy

335 mg (67%) of the desired amide 209 as a thick pale yellow oil: IR (neat) 2978,1725 (broad) cm*1; 1H-NMR (CDCI3 , 250 MHz)* 5 1.50 (s, 9H, C(CH3)3), 1.60 (m, 1H, Ctl2 CH2 N), 1.78 (td, J =

15, 6 Hz, 1 H, OCH2 CH2 ), 193-2.10 (m, 2 H, CJd2 CH2N and OCH2 CH2 ), 2 06 (s, 3H, COCH3),

2.31-2.47 (m, 3H, OCHCfcland OCH2 Ctl2 ), 3.24 (s, 3H, OCH3 ), 3.36 (td, 1 2 , 4 Hz, 1H, CH2 N),

3.46 (m, 2H, OCH2), 3.85 (dt, J= 12, 4.5 Hz, 1 H, CH2 N), 5.06 (m, 1H, CHOAc), 5.63 (ddd, J= 10,

5, 2 Hz. 1H, CH2 CU=), 5.72 (dd, J= 10,1.3 Hz, 1H, CH2CH=Ctl); 13C-NMR (CDCI3 . 75.5 MHz) 8

19.7 (t), 21.0 (q), 26.5 (t), 27.8 (q), 37.7 (d), 38.0 (t), 45.5 (t), 52.0 (s), 58.3 (q), 68.5 (d), 68.7 (t),

82.4 (s). 123.3 (d), 131.3 (d), 152.6 (s), 170.1 (s), 173.9 (s); exact mass calcd. for C1 9H3 0 NO6 m/z

368.2074, found m/z 368.1980. 136 OAc OAc

° ""1 0 " I OCHj OH

204 208

(±H4a/?t,5S*,8a/?*)-3,4,4at5,6,8a-Hexahydro-5-hydroxy-8a-(2-methoxy- e t h y l) - l ( 2 /y)-isoqulnollne acetate (ester) (204) and (±>(4a/?*,5S*,8a/?*)-

3,4,4a,5l6,8a-Hexahydro-5-hydroxy-8a-(2-hydroxyethyl)-1(2H)-lsoquinollne acetate (ester) (208). To a solution of 51 mg (0.139 mmol) of carbamate 209 in 4 mL of dichloromethane cooled in a dry ice-acetone bath was added 250 pL (0.250 mmol) of 1.0 M boron tribromide in dichloromethane. The resulting mixture was stirred in the cooling bath for 1 h, at 0°C for 1 h, and then poured into 100 mL of dichloromethane. The organic layer was washed with 25 mL of saturated aqueous sodium bicarbonate, dried (MgSC> 4 ), and concentrated in vacuo. The residue was chromatographed over 8 g of silica gel (eluted with ethyl acetate) to yield 26 mg (70%) of methoxylactam 204 with spectral data identical to that previously reported (vide supra). Further elution with ethyl acetate-methanol, 10:1 afforded 7 mg (20%) hydroxylactam 208 with spectral data identical to that previously reported (vide supra).

OAc

OTHP

2 0 6

(+)-(4a/?*l5S*,8a/?*)-3t4,4a,5,6,8a-Hexahydro-5-hydroxy-2-(p-methoxyphenyl)-8a-[2-oxo-8-[(tetrahydro-2H-pyran-2-yl)oxy]-3'OCtynyl]-1(2H)-lsoqulno- llne acetate (ester) (206). To a solution of 9.70 g (17.9 mmol) of propargyl alcohols 173 and 174 in 200 mL of dichloromethane was added 14.5 g <167 mmol) of manganese dioxide.

The resulting dark slurry was stirred at room temperature for 18 h. The resulting dark mixture was

filtered through 40 g of silica gel, eluting with 900 mL of ethyl acetate. The filtrate was

concentrated in vacuo to yield 9.50 g (98%) of propargyl ketone 206 was as a thick orange oil-

foam: IR (neat) 2940, 2210, 1731, 1650 cm'1; 1H NMR (CDCI 3 , 300 MHz) 4 5 1.46-2.00 (m, 13H,

CH2 manifold), 2.04 (s, 3H, O2 CCH3 ), 2.10 (m, 1H, =CHCJti2 ). 2.34-2.45 (m, 3H, =CHCH2 and

=CCH2), 2.53 (dt, J = 12.4, 3.5 Hz, 1H, CH), 2.85 (d, J = 18.9 Hz, 1H, COCH2), 3.33-3.55 and

3.68-3.88 (m, 6 H, CtbOCHOCtk and NCH2), 3.75 (s, 3H, OCH3 ), 4.54 (m, 1H, OCHO), 5.04

(ddd, J= 10.3, 5.9, 4.3 Hz, 1H, CHOAc), 5.64 (m, 2H, CH=CH), 6 .8 6 (d, J = 8.9 Hz, 2 H, ArH), 7.14

(d, 8 .8 Hz, 2H, ArH); 13C NMR (CDCI3 , 75.5 MHz) S 18.7 (t), 19.4 (t), 20.3 (t), 21.1 (q), 24.5 (t),

25.3 (t), 26.4 (t), 28.7 (t), 30.6 (t), 38.2 (d), 48.3 (s), 50.5 (t), 53.0 (t), 55.3 (q), 62.1 (t), 66.5 (t),

68.7(d), 80.8 (s), 94.3 (s), 98.7 (d), 114.3 (d), 124.2 (d), 127.4 (d), 130.2 (d). 136.6 (S), 158.0 (S),

170.2 (s), 172.1 (s), 185.1 (s); exact mass calcd for C3 1 H3 9 NO7 m/z 537.2728, found m/z

537.2726.

OAc OAc

OTHP OTHP

173 174

(±M4a/?*,5S*,8af?*)-3,4,4a,5,6,8a-Hexahydro-5-hydroxy-8a-{(2$*)*2*hydroxy-8-[tetrahydro-2H-pyran-2-yl)oxy]-3-octynol}-2-(p-methoxyphenyl)-l(2H)-

Isoqulnollnone 5-acetate (173) and (±J-(4aft*,5S*,8a/?*)-3,4,4a,5,6,8a-

Hexahydro-5-hydroxy-8a-{(2/7*)-2-hydroxy-8-[tetrahydro-2H-pyran-2-yl)oxy]-3- octynol}-2-(p-methoxyphenyl)-1(2H)-isoquinollnone 5-acetate (174). To a mixture of 9.50 (17.3 mmol) of ynone 206 in 350 mL of methanol was added 7.71 g (20.7 mmol) of cerium 138

trichloride heptahydrate. The mixture was stirred at rt for 15 min, cooled in a dry ice-acetone bath,

and 707 mg (20.8 mmol) ot sodium borohydride was added in one portion. The reaction mixture

was stirred in the cooling bath for 1 h 30 min and 10 mL of water was added. The reaction was

concentrated to approximately one-fourth of the original volume, and was partitioned between

800 mL of dichloromethane and 150 mL of water. The organic layer was dried (MgS 0 4 ) and

concentrated in vacuo to yield 9.25 g (97%) of a 3:1 mixture of diastereomeric alcohols 173 and

174 (determined by integration of the OCH 3 peak in the 1H NMR spectrum), as a thick orange

foam.

OAc

OTHP

21 1

(±M4a/?*,5S*,8a#?*)-8a-[(2S*)-3,4-Epoxy-2-hydroxy-8[[tetrahydro-2W-

pyran-2-yl)oxy]octyl]-3,4,4a,5,6,8a-hexahydro-5-hydroxy-2-(p-methoxyphenyl)-

l(2H)-lsoqulnolone 5-acetate (211). To a solution of 3.40 g (6.28 mmol) of alcohol 188 in

100 mL of benzene was added 25 mg (0.094 mmol) of vanadyl acetylacetonate and 3.5 mL (7.2

mmol) of 2.07 M f-butyl hydrogen peroxide in toluene. The resulting deep red mixture was

heated at 40°C for 2 h, cooled to room temperature, and diluted with 100 mL of ether, The

mixture was washed with 40 mL of saturated aqueous sodium bisulfite, 40 mL of saturated

aqueous sodium bicarbonate, 40 mL of brine, dried (MgS 0 4 ), and concentrated in vacuo to yield

a thick brown oil. This material was chromatographed over 80 g of flash silica gel (eluted with ethyl

acetate-hexanes, 1:1 (1.5 L) to ethyl acetate-hexanes, 3:1) to yield 710 mg ( 2 1 %) of impure

recovered starting material. Continued elution provided 1 .99 g (57%) of epoxide 211 as a thick orange oil: IR (neat) 3426, 1729, 1633, 1605 cm '1; 1H NMR (CDCI 3 , 300 MHz)* 5 1.49-2.20 (m, 139

16H, CH2 manifold), 2.05 (s, 3H, COCH3 ), 2.42 (m, 2H, OH and -CHCtte). 2.64 (dd, J = 14.8,

10.3 Hz, 1 H, CHOHCfcfc), 2.66 (m, 1H, CH), 2.80 (dd, J= 8.0, 4.0 Hz, 1H, CHCJJO), 2.97 (m, 1 H,

CHCHOCH), 3.38-3.53 and 3.66-3.88 (m, 7H, CtiOH, NCH2, and CtteOCHOCfcfc), 3.77 (s, 3H,

OCH 3 ), 4.56 (m, 1H, OCHO), 5.10 (m, 1H, CHOAc), 5.62 (ddd, J = 10.0, 5.0, 2.3 Hz, 1H,

-CUCH2), 5.72 (d. J= 10.0 Hz, 1 H, CCH=), 6 .8 6 (d, J = 8.9 Hz. 2H, ArH), 7.08 (d, J= 8.9 Hz, 2 H,

ArH); 13C NMR (CDCI3 , 75.5 MHz) 5 19.5 (t), 19.7 (t), 21.1 (q), 23.3 (I), 25.4 (t), 26.7 (t), 27.4 (t),

29.4 (1), 30.6 (1), 36.6 (d), 42.6 (I), 49.3 (I), 50.8 (s), 55.3 (q). 57.8 (d), 59.3 (d), 62.2 (t), 65.9 (d),

67.2 (t), 68.5 (d), 98.7 (d), 114.4 (d), 123.1 (d), 127.2 (d), 131.9 (d), 136.4 (s), 158.2 (s), 170.0

(s), 173.8 (s); exact mass calcd for C3 1 H4 3 NO0 m/z 557.2990, found m/z 557.2985.

OAc

1 7 0

{±)-4afl*,5S*,8a/?*)-3,4,4a,5,6*Hexahydro-5-hydroxy-2-(p-methoxyphenyl)-l-oxo- 8 a(lH)-lsoquinollneacetaldehyde (170). To a solution of 1.95 g (3.50 mmol) of epoxy alcohol 211 in 30 mL of dry THF was added 1.88 g (8.25 mmol) of periodic acid in one portion. The resulting mixture was stirred at room temperature for 1 h 45 min, and was then diluted with 80 mL of water. The mixture was extracted with 400 mL of ether, and the organic layer was washed with 50 mL of saturated aqueous sodium bicarbonate, 70 mL of brine, dried

(MgS0 4 ), and concentrated in vacuo to yield a thick pink oil. This material was chromatographed over 30 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:2) to yield 0.93 g (74%) of aldehyde 170 as a light pink foam. The spectral data for this material was identical to that previously reported .2 7 140 OAc

212

ferf-Butyl (±)*[( 1 /7*,2Z)-l-[[(4afl*,5S*,8aR*)-2,3,4,4a,5,6-hexahydro-5- hydroxy-2-(p-methoxyphenyl)-l-oxo-8a(1H)-lsoqulnolyl]methyl]-7-[(tetrahydro-

2 /y-pyran- 2 -yl)oxy}- 2 -heptenyl](( 2 -(trlmethylsilyl)ethyl]sulfonyl]carbamate, acetate (ester) (212) A mixture of 8.65 g (10.8 mmol) of acylsulfonamide 189, 450 mg (2.37 mmol) of p-toluenesulfonic add, 400 mL of THF and 50 mL of methanol was stirred for 40 h and then concentrated in vacuo. The residue was dissolved in 200 mL of dichloromethane, washed with 30 mL of saturated aqueous sodium bicarbonate, dried (MgSCXO, and concentrated in vacuo to yield a clear white foam. This material was dissolved in 150 mL of dimethyl sulfoxide, heated at

170°C for 25 min, and cooled to room temperature. The mixture was diluted with 1 L of ether, washed with four 80-mL portions of brine, and the combined aqueous layers were backwashed with 400 mL of ether. The combined organic layers were dried (MgS 0 4 ) and concentrated in vacuo. The residue was chromatographed over 80 g of flash silica gel (eluted with ethyl acetate- hexanes, 1:3 (2 L) to ethyl acetate-hexanes, 1:1) to yield 4.48 g (67%) of hydroxy sulfonamide

212 as a white foam. This material exhibited spectral data identical to that reported previously .27

OAc

213 141

(±)-A/-[i/l*I2Z)-l-[[(4aff*,5S*,8af7*)-2,3,4,4at5,6-Hexahydro-5-hydroxy-2-

(p-methoxyphenyl)-1-oxo-8a(1//)-isoqulnolyl]methyl]-7-hydroxy-2-hepienyl]-2- trimethylsllyl)elhanesulfonamide] 5'-acetate 7-p-toluenesulronate (213). To a solution of 2.16 g (3.50 mmol) of alcohol 212 in 40 mL of dichloromethane was added 1.01 g

(5.30 mmol) of p-toluenesulfonyl chloride, 34 mg (0.28 mmol) of 4-N,N-dimethylaminopyridine, and 2.0 mL (14 mmol) of triefhylamine. The resulting mixture was stirred at room temperature for

10 h, was diluted to 300 mL with dichloromethane, and washed sequentially with 40 mL of 10% aqueous hydrochloric acid and 40 mL of saturated aqueous sodium bicarbonate. The organic layer was dried (MgSCXO, concentrated in vacuo, and the residue was chromatographed over 50 g of silica gel (eluted with ethyl acetate-hexanes, 1:4 (1.5 L) to ethyl acetate-hexanes, 3:7) to yield

2.58 g (95%) of tosylate 213 as a white foam. This material exhibited spectral data identical to that previously reported .2 7

OAc

2 1 4

(±)-W -[i/?\2Z)-l-[[(4a/?*l5S*,8aR*)-2,3,414a,5,6-Hexahydro-5-hydroxy*2*

(p-methoxyphenyl)-1-oxo-8a(1H)-lsoquinolyl]methyl]-1,2,5,6l7,8-hexahydro-1-

[[2-(trlmethylsilyl)ethyl]sulfonyl]azoclne acetate (ester) (214). To a mixture of 4.45 g (5.74 mmol) of tosylate 213 in 600 mL of toluene was added 2.40 g (9.09 mmol) of 18-crown-6,

2.27 g (6.14 mmol) of tetrabutylammonium iodide, and 310 mg (7.75 mmol) of potassium hydride.

The resulting mixture was heated at 120°C for 2 h, and then cooled to rt. The mixture was poured into 300 mL of ethyl acetate, washed with 300 mL of water, and the organic layer was dried

(MgSC>4 ) and concentrated in vacuo. This material was chromatographed over 120 g of flash silica 142 gel (eluted with dichloromethane-ethyl acetate, 9:1) to yield 3.06 g (89%) ol azocine 214 as a white solid: mp 157-160.5°C (lit2 7 162-163°C). The spectral data for this material were identical to that previously reported. OH

2 1 5

(±.)-AHlfl*)-l-t[(4aH*,5S*,8afl*)-2,3J4,4a,5,6-Hexahydro-5-hydroxy-2-(p- methoxyphenyl)-l-oxo-8a(1H)-isoqulnolyl]meihyl]-l,2,5,6,7,8-hexahydro-1-[[2-

(trlmethylsllyl)ethyl]sulfonyl]azocine (215). To a mixture of 2.48 g (3.20 mmol) of tosylate 213, 1.08 g (4.09 mmol) of l 8 -crown-6 , 1.38 g (3.73 mmol) of tetrabutylammonium iodide, and 350 mL of toluene was added 360 mg (9.0 mmol) of potassium hydride (originally 35% weight in mineral oil, washed with two 5-mL portions of hexanes and evacuated). The resulting mixture was heated at 105-115°C for 1.5 h, was cooled to room temperature, cooled in an ice- water bath, and quenched by the addition of 0.5 mL of glacial acetic acid. The reaction mixture was concentrated in vacuo to yield a yellow solid, which was dissolved in 50 mL of THF, 50 mL of methanol, and 25 mL of water. To the mixture was added 380 mg (9.05 mmol) of lithium hydroxide monohydrate, and the mixture was heated at 40-45°C for 45 min, cooled to room temperature, and quenched by the addition of 1 mL of glacial acetic acid. After concentrating to approximately one-fifth of the original volume, the residue was diluted with 700 mL of dichloromethane, was washed with 200 mL of saturated aqueous sodium bicarbonate and 200 mL of water. The organic layer was dried (MgS 0 4 ) and concentrated in vacuo to yield a yellow solid. This material was chromatographed over 130 g of silica gel (eluted with dichloromethane-ethyl acetate, 9:1) to yield

1.44 g (80%) of tricycle 215 as a white solid: mp 196.5-200°C (lit 2 7 216.7-217.5°C). This material exhibited spectral data identical to that previously reported. SES LJ 216 217

(±)-(2/?*)-l,2,5,6,7,8-Hexahydro-2-[[1a#?#,3S*,3a#?*,7aS*,7bS*)-octahydro-3-hydroxy-6-(p-methoxyphenyl)-7-oxoxolreno[/?]lsoqulnolln-7a(2/y)- yl]methyl]-1-[[2-(trlmethylsllyl)ethyl]sulfonyl]azoclne (216) and (±>(2#?*)-2- l[(la/?*,3S\3a/?*,7aS*,7bS*)-Oclahydro-3-hydroxy-6-(p-methoxyphenyl)-7- oxooxlreno[ft]isoquinolln-7a<2H)-yl]methyl]-3[[2-(trlmethylsilyl)ethyl]sul1onyl]-9- oxa-3-azablcyclo[6.1.0]nonane (217). To a mixture of 1.41 g (2.52 mmol) of homoallylic alcohol 215, 25 mg (0.095 mmol) of molybdenum hexacarbonyl, and 300 mL of benzene was added 900 pL (3.60 mmol) of 4.0 M f-butylhydroperoxide in toluene. The resulting pink mixture was heated to 92°C over 1 h followed by 1.5 h at 92°C. To the mixture was added an additional

100 pL (0.40 mmol) of 4.0 M f-butylhydroperoxide in toluene, and after 20 min the reaction was cooled to room temperature. The mixture was poured into 1 L of ether, washed with 100 mL of saturated aqueous sodium bisulfite, 100 mL of saturated aqueous sodium bicarbonate, 100 mL of brine, dried (MgSC> 4 ), and concentrated in vacuo. The residual white solid was chromatographed over silica gel (eluted with ethyl acetate-hexanes, 1:1) to yield 1.23 g (85%) of epoxy alcohol 216 as a white solid (mp 189-190.5°C), exhibiting spectral data identical to that previously reported .2 7

Further elution afforded 230 mg (11%) of bisepoxide 217 as a white solid: mp 108-110°C; IR

(CH2 CI2 ) 3478, 1636, 1606, cm -1; 1H NMR (CDCI 3 , 300 MHz)* 6 0.02 (s. 9H, Si(CH3)3), 1.09 (m,

2 H, CH2 Si), 1 .6 8 and 1.89 (m, 5H, CH 2 manifold). 2.05-2.31 (m, 6 H, CH and CH2 manifold), 2.54

(dd, J = 14.9, 6.3 Hz, 1 H, NCHCtte). 2.61 (dd, J = 14.9, 4.5 Hz, 1 H NCHCM2 ). 2 . 8 8 (d, J= 8.3 Hz,

1H, OH). 2.99 (t, J= 8.3 Hz, 2H, SCH2), 3.08 (m, 1H, NCCOCH), 3.17 (dt, J= 15.1, 5.2 Hz, 1H, 144

SO2 NCH2 ), 3.30 (dd, J= 6.2, 4.2 Hz, 1 H, NCCHO), 3.44 (m, 2 H, CCHOCJdand SO2 NCH2), 3.48

(d, J = 3.8 Hz, 1H, CCHO), 3.55 (m, 1 H, NCH2 ), 3.79 (s. 3H, OCH3 ), 3.99-4.14 (m, 3H, NCH,

CHOH, NCH2), 6 .8 8 (d, */ = 8.9 Hz, 2 H, ArH), 7.12 (d, J= 8 .8 Hz, 2H, ArH). 13C NMR (CDCI3 , 62.9

MHz) 6 -2.0 (q), 10.3 (t), 22.2 (t), 22.7 (t), 25.1 (t), 27.6 (t), 29.9 (t), 39.4 (t), 39.9 (d), 44.3 (S), 48.6

(t), 50.4 (t), 51.5 (t). 53.9 (d), 55.4 (q), 55.7 (d), 57.3 (d), 58.4 (d), 68.5 (d), 114.4 (d), 127.0 (d),

136.4 (s), 158.1 (s), 171.7 (s), one sp3 carbon missing; exact mass calcd. tor C2 gH4 4 N2 SSi0 7 m/z 592.2639, found m/z 592.2641.

OMOM-

2 1 8

(±)"(2H*)*1,2,5,6,7,8-Hexahydro-2-[[la/?*,3S*,3a/?*,7aS*l7bS*)-octahydro-3-methoxymethoxy-6-(p-methoxyphenyl)-7-oxoxolreno[/i]lsoqulnolln-

7a(2W)-yl]methyl]-1-[[2-(trlmethylsllyl)ethyl]sulfonyl]azoclne (218). To a solution of

420 mg (0.73 mmol) of alcohol 216 in 8 mL of dichloromethane was added 600 pL (3.4 mmol) of diisopropylethylamine and 310 pL (4.1 mmol) of chloromethyl methyl ether. The resulting yellow mixture was stirred at room temperature for 24 h, diluted with 150 mL of dichloromethane, and then washed with 25 mL of 10% aqueous hydrochloric acid and 25 mL of saturated aqueous sodium bicarbonate. The organic phase was dried (MgSC> 4 ), concentrated in vacuo, and chromatographed over 15 g of silica gel (eluted with ethyl acetate-hexanes, 1:2 (400 mL) to ethyl acetate-hexanes, 1:1) to yield 360 mg ( 8 6 %) of azocine 218 as a white solid (mp 176-177°C).

This material exhibited spectra data identical to that previously reported .27 145 OMOM

Tetradecahydro-7-hydroxy-5(methoxymethoxy)-2-(p-methoxyphenyl)-lH-azoclno-[1',2,:1,5]pyrroloE2l3-i]lsoqulnolin-l-one (219). A mixture of 232 mg (0.374 mmol) of azocine 218, 280 mg (1.84 mmol) of cesium fluoride, and 6 mL of dry DMF was stirred and heated at 80°C for 19 h. The mixture was cooled to rt, and the solvent was removed under a vacuum of 0.5-1.5 mm Hg over 3 h. The resulting orange residue was chromatographed over 8 g of activity II basic alumina (eluted with ethyl acetate (150 mL) to ethyl acetate-methanol 15:1) to yield 139 mg (81%) of alcohol 219 as an orange foam. This material exhibited identical spectral data to that previously reported .2 7

OMOM

220

Tetradecahydro-7-hydroxy-5(methoxymethoxy)-2-(p-methoxyphenyl)-1H-azoclno-[i,,2':i,5]pyrrolo[2l3-l]isoqulnolln-l-one acetate (ester) (220). To a solution of 135 mg (0.296 mmol) of alcohol 219 in 4 mL of dry dichloromethane was added 200 pL (2.1 mmol) of acetic anhydride and 300 pL (1.7 mmol) of diisopropylethylamine. The resulting orange 146 mixture was stirred at room temperature for 2 2 h and then concentrated in vacuo to yield and orange oily residue. This material was chromatographed over 7 g of activity II basic alumina (eluted with ethyl acetate) to provide 134 mg (91%) of the desired acetate 220 as an orange foam. The spectral data for this material were identical to that previously reported .2 7

221

Tetradecahydro-5,7-dihydroxy-2-(p-methoxyphenyl)-l/y-azoclno-[1,,2,:1,5] pyrrolo[2,3-i]isoquinolln-1-one 7-acetate (221). To a mixture of 175 mg (0.35 mmol) of tetracycle 220 in 7 mL of acetonitrile was added 320 mg (0.340 mmol) of lithium tetrafluoroborate and 5 drops of water. The resulting mixture was stirred and heated at 75°C for 6 h, cooled to rt. and concentrated in vacuo. The resulting oily orange residue was chromatographed over 20 g of activity II basic alumina (eluted with ethyl acetate-hexanes, 1:1 (300 mL) to ethyl acetate (200 mL) to ethyl acetate-methanol, 25:1 (200 mL) to ethyl acetate-methanol, 10:1) to providel 6 mg (9%) of unreacted starting material and 113 mg (71%) of alcohol 221 as an off-white foam. The spectral data for this material was identical to that previously reported .2 7 151

/’0/-(4aft*,7aS*,14aSM5a5*)-2,3,4,4a,9,1O,11,12,14a,l5‘Dectihydro-2-

(p-methoxyphenyl)-l H-azoclno[1 ',2*:1,5]pyrrolo[2,3/]lsoqulnoline-1,5{7aM)- dlone (151). To a solution of 80 pL (0.92 mmoi) of oxalyl chloride in 3 mL of dichloromethane cooled in a dry ice-acetone bath was added 120 pL ( 1 .7 mmol) of dimethylsulfoxide. The resulting mixture stirred in the cooling bath for 45 min and 105 mg (0.23 mmol) of alcohol 221 in 3 mL of dichloromethane was added dropwise. The resulting mixture was stirred at -78°C for 1 h 30 min,

250 pL (1.8 mmol) of triethylamine was added, and the reaction was stirred for 1 h in the cooling bath and 1 h at room temperature. The reaction mixture was concentrated in vacuo and chromatographed over 15 g of activity II basic alumina (eluted with ethyl acetate-hexanes, 1.1) to provide 75 mg (83%) of enone 151 as a yellow foam. This material was crystallized from dichloromethane-hexanes to afford white crystals (mp 173-174.5°C). Spectral data were identical to those previously reported .2 7

Anal, calcd for C2 4 H2 8 N2 O3 : C, 73.44; H, 7.19. Found; C, 73.20; H, 7.25.

( H 3 0 3 C Xo 148

(±)-(2R*)-1,2,5,6,7,8-Hexahydro-2-[[(laf?*,3S*,3a/?\7aS*,7bS*)- octahydro-3-hydroxy-6-(p-methoxyphenyl)-7-oxooxrleno[/r]lsoqulnolln-7a(2H)- yl]methyl]-i-[[2-(trimethylsllyl)ethyl]sulfonyl]azoclne plvalate ester (227). To a solution of 420 mg (0.73 mmol) of azodne 216 in 10 mL of pyridine was added 1.2 mL (9.8 mmol) of trimethylacetyl chloride. The resulting mixture was stirred at room temperature for 24 h and concentrated in vacuo. The residue was dissolved in 200 mL of dichloromethane, washed with

25 mL of 10% aqueous hydrochloric acid, 20 mL ol saturated aqueous sodium bicarbonate, dried

(MgS0 4 ), and concentrated in vacuo. The residual white solid was chromatographed over 20 g of silica gel (eluted with ethyl acetate-hexanes, 1:2) to yield 473 mg (98%) of the desired azocine

227 as a white solid (mp 238.5-240°C): IR (CDCI3 ): 1725, 1623, 1602 cm’1; 1H NMR (CDCI3 ,

300 MHz) 4 8 0.03 (S, 9H, Si(CH3)3), 1 .1 3 (m, 2H, CH2 Si), 1.21 (s, 9H, C{CH3)3), 1.55-2.00 (m, 6 H,

CH2 manifold), 2 .1 0 (m, 1 H, =CHCtb). 2 42-2.65 (m, 4H, NCHCtkOH), CtlCHOAcCte), 2.85 (m,

1H, sCHChte), 2.87 (dd, J = 14.4, 4.4 Hz, 1H, NCHCtfc), 3.04 (ddd, J= 10.8, 5.7, 1.8 Hz, 2H,

S 0 2 CH2), 3.46 (dm, 16.1 Hz, 1H, NCH2), 3.58-3.78 (m, 3H, ArNCH 2 and S 0 2 NCH2), 3.80 (S,

3H, OCH3 ), 3.82 (m, 1H, CCHO), 4.23 (q, 4.3 Hz, 1H, CHOCHCH2), 4.73 (m, 1H, NCH), 5.35

(m, 1H, CHOAc), 5.63 (m, 2H, CH=CH), 6.93 (d, J= 8.9 Hz, 2H, ArH), 7.17 (d, J= 8.9 Hz, 2H, ArH);

13C NMR (CDCI3 , 62.9 MHz) 5 -1.9 (q), 10.3 (t), 19.7 (t), 24.0 (t), 24.9 (t), 25.7 (t), 27.2 (q), 30.9

(t), 37.2 (d), 37.4 (t), 38.9 (S), 46.4 (t), 49.7 (t), 50.9 (t), 54.0 (d), 55.4 (q), 59.2 (d), 67.8 (d), 75.6

(d), 114.7 (d), 127.6 (d), 128.5 (d), 131.1 (d), 136.0 (s), 158.4 (s), 173.6 (s), 177.4 (s), the 13C

NMR is missing one aliphatic singlet; exact mass calcd. for C3 4 H5 2 SSiN2 0 7 0*^660.3267, lound m/z 660.3250. 149 o

(CH3)3C

2 2 8

(±>(4afl*,5S\7/?*,7a/7*,i4afl*,l5aS*)-2,3,4,4a,5,6,7,7a,9,10,11,14a,

l5-Tetrahydro-5,7-dlhydroxy-2-(p-methoxyphenyl)-i W-azoclnop’^ ’M.Sl-

pyrroloia.S-lllsoqulnolin-l-one 5-pivalate (228). A mixture of 560 mg (3.68 mmol) of

cesium fluoride, 345 mg (0.523 mmol) of azocine 227, and 10 mL of A/.AAdimethylformamjde was

heated at 89°C for 17 h. The solvent was removed in vacuo, and the residue was

chromatographed over 20 g of activity grade II basic alumina (eluted with ethyl acetate-hexanes,

1:1 (150 mL) to ethyl acetate-methanol, 15:1 (until completion)) to yield 229 mg ( 8 8 %) of alcohol

228 as a yellow foam: IR (CHCI3) 3426, 1717, 1639, 1219 cm'1; 1H-NMR (CDCI3 , 300 MHz*) 8

1.16 (s, 9H, C(CH3 )3 )1.3-1.7, 1.7-1.8 , 1.85-2.10 (m, 1 0 H, CH2 manifold), 2.10-2.30 (m, 2H, CH2

manifold), 2.51 (dd, J = 13,8 Hz, 1H, NCHCtte), 2.58 (m, 1H, CH), 2.88 (t, J= 6 Hz, 1H, NCHChfe),

3.24 (d, J= 5 Hz, 1H, OH), 3.62 (d, J= 3 Hz, 1H, NCH), 3.72 (ddd, J= 13.7, 4.0, 1.0 Hz, 1 H,

CONCH2)), 3.80 (s, 3H, OCH3 ), 3.87 (ddd, J= 13, 10, 5.5 Hz, 1 H, CONCH2), 4.02 (s, 1 H, CHOH),

4.23 (m, 1H, NCH), 5.09 (q. J= 4 Hz, 1 H, CHOPiv), 5.36 (t, J= 10.5 Hz, 1H, NCHCH=), 5.94 (dt. J

- 11, 7 Hz, 1H, CHCH=Chl), 6.89 (d. J = 9 Hz, 2H, ArH), 7.21 (d, J= 9 Hz, 2H ArH); 1 3C-NMR

{CDCI3 , 75.5 MHz) 8 23.2 (t), 27.2 (q), 27.3 (t), 27.6 (t), 29.2 (t), 30.7 (t). 39.0 (S), 39.7 (d), 43.9 (t),

47.5 (t), 48.4 (t), 50.1 (s), 54.0 (d), 55.4 (q), 65.2 (d), 65.5 (d), 71.7 (d), 114.0 (d), 126.2 (d). 129.3

(d). 133.9 (d), 136-0 (s), 157.6 (s), 175.7 (s), 178.3 (s); exact mass calcd. for C2 9H4 0 N2 O5 m/z

496.2939, found mz 496.2903. 150

1 5 2

(±M 4afl*,7aS\l4aSM 5aS*)-2,3,4,4a,9,i0,i i,i2,i4a,i5-Decahydro-2-

(p-methoxyphenyl)-1H-azoclno[ri2,:1,5]pyrrolo[2,3-/]isoqulnollne-1,7(7aH)- dlone (152). To a solution of 105 |j.L ( 1 .2 1 mmol) of oxalyl chloride in 2.0 mL of dichloromethane cooled in a dry ice-acetone bath was added 180 jiL (2.5 mmol) of dimethylsulfoxide. The resulting mixture was stirred in the cooling bath for 55 min, and a solution of 203 mg (0.41 mmol) of alcohol 228 in 1 mL of dichloromethane was added dropwise via syringe. The resulting mixture was stirred in the cooling bath for 1 h 30 min, and 180 pL (1.3 mmol) of triethylamine was added. The reaction mixture continued to stir in the cooling bath for 30 min and was then warmed to room temperature. The mixture was poured in 15 mL of saturated aqueous sodium bicarbonate, and extracted with two 70-mL portions of dichloromethane. The combined organic layers were dried (MgS 0 4 ) and concentrated in vacuo. The residue was chromatographed over 8 g of activity II basic alumina (eluted with ethyl acetate-hexanes, 1:4 (200 mL) to ethyl acetate-hexanes, 1:1) to yield 105 mg (65%) of enone 152 as a pale yellow solid [mp

180-185°C (dec)). The spectral data were identical to that previously reported :2 7 13C NMR

(CDCI3 , 75.5 MHz) (previously unreported) 5 24.6 (t), 27.2 (t), 27.6 (t), 29.2 (t), 37.7 (d), 44.7 (t),

46.4 (t), 48.3 (t), 54,7 (d), 55.2 (s), 55.4 (q), 70.0 (d), 114.3 (d), 127,2 (d), 129.3 (d),130.6 (d).

134.8 (d). 135.7 (s), 149.8 (d), 158.1 (s), 171.1 (s), 197.0 (s). 1-(2-Pyridlnyl)-2-cyclohexen-1-ol (266). To a solution of 40 mL (54.4 mmol) of a

1.36 M solution of n-butyllithium in hexanes at '7 6 °C was added 8.59 g (54.4 mmol) of 2 - bromopyridine in 16 mL of ether. The resulting red mixture was stirred at -75°C for 2.5 h, and the resulting red solution was added via cannula to a solution of 5.23 g (54.5 mmol) of 2-cyck>hexen-

1-one in 15 mL of ether at -76°C. Upon completion of addition, the olive green reaction mixture was warmed slowly to room temperature and the color changed to yellow. The solution was stirred for 2 h, 100 mL of water was added, and after 15 min the reaction mixture was extracted with five

75-mL portions of ether. The combined organic layers were washed with 75 mL of brine, dried

(M g S 0 4 ), and concentrated in vacuo to yield 10.23 g of an orange oil. This oil was chromatographed over 125 g of silica gel (eluted with ethyl acetate-hexanes, 1:8) to yield 6.65 g

(70%) of alcohol 266 as a yellow oil: IR (neat) 3396 cm'1; 1H NMR (CDCI 3 , 300 MHz) 5 1.73-2.25

(m, 6 H, (CH2 )3 ), 5.14 (brs, 1H. OH), 5.67 (d, J= 10 Hz, 1H, =CH), 6.07 (ddd, J = 10, 4.5, 3 Hz, 1H,

=CH), 7.16 (ddd, J= 7.4, 5, 1 Hz, 1H, H5-), 7.31 (dt, J= 8 , 1 Hz, 1 H, H3 -), 7.65 (td, J= 7.5, 1.7 Hz,

1H, H4’), 8.50

71.1 (s), 120.1 (d), 121.8 (d), 130.6 (d). 131.3 (d), 136.5 (d), 147.4 (d), 164.7 (s); exact mass calcd for Ci 1H1 3 NO m/z 175.0997, found m/z 175.1004. 3-(2-P yrldlnyl)-2-cyclohexen-l-on e (270). To a solution of 3.12 g (8.31 mmol) of pyridinium dichromate in 10 mL of dimethylformamide under an inert atmosphere was added

7 4 4 mg (4.25 mmol) of alcohol 266 in 1 mL of dimethyHormamide. The black reaction mixture was stirred at 65°C for 17 h and then poured into 100 mL of 3 N aqueous NaOH. The aqueous solution was extracted with eight 100-mL portions of ether. The combined organic layers were washed with 100 mL of brine, dried (MgS04), and concentrated to yield 198 mg of an orange oil.

This material was chromatographed over 10 g of silica gel (eluted with ethyl acetate-hexanes, 1:5) to yield 131 mg (18 %) of ketone 270 as a yellow oil: IR (neat) 1670 cm*1; 1H NMR (CDCI 3 , 250

MHz) 6 2.14 (qu, J = 6 Hz, 2 H, CH2 Ctl2 CH2 ), 2.49 (t, J = 7 Hz, 2 H, (C=0 )CH2 ), 2.90 (Id, J= 6 , 1.5

Hz, 2H, =CCH2), 6.70 (t, 1.5 Hz, 1 H, =CH(C=0)), 7.27 (ddd, J= 6 , 3, 1Hz, 1 H, H5 ), 7.60 (dt, J

= 7, 1Hz, 1 H, H3’), 7.72 (td, J= 7.5, 1.7 Hz, 1H, H4 ), 8.65 (dt, J= 4, 2 Hz, 1H, HeO: 13C NMR

(CDCI3 , 75.5 MHz) 5 22.5 (t), 26.3 (t), 37.5 (t). 120.9 (d), 124.0 (d). 126.6 (d), 136.5 (d), 149.4 (d),

155.8 (s), 158.6 (s), 200.1 (s); exact mass calcd for C1 1 H 1 1 NO m/z 173.0841, found m/z

173.0841.

2 7 5

1-(2>Pyrldlnyl-2-cyclohex-i-enol acetate (275). A mixture of 1.17 g (6.71 mmol) of alcohol 266, 72 mg (0.59 mmol) of 4-A/,/V-dimethylaminopyridine, 1.06 g (10.0 mmol) of acetic anhydride, and 1.4 mL (10.0 mmol) of triethylamine was stirred under an inert atmosphere at room temperature for 13 h. The mixture was concentrated in vacuo and the resulting orange- brown oil was chromatographed over 25 g of silica gel (eluted with ethyl acetate-hexanes, 1:5) to yield 1.35 g (87%) of ester 275 as a pale yellow oil: IR (neat) 1732 c m 1; *H NMR (CDCI 3 , 300

MHz) 5 1.60-2.25 (m with s at 2.0, 9H, (CH2 ) 3 and (C=0)CH3), 6.16 (dt, 10, 3.8 Hz, 1 H, =CH), 153

6.50 (dqu, J= 10, 1 Hz, 1 H, =CH), 7.14 (ddd, J = 7.5, 4.8, 1.1 Hz, 1H, Hs-), 7.31 (dt, J= 8 ,1 Hz,

1H, H3 -). 7.64 (td, J= 7.5,1.8 Hz, 1 H, H4-), 8.56 (dq, J= 4.8, 0.9 Hz. 1H, H6 ); 13C NMR (CDCI3 ,

62.9 MHz) 8 18.3 (I), 21.7 (q), 24.7 (t), 36.6 (t), 80.2

136.3 (d), 148.8 (d), 162.9 (s), 169.7 (s): exact mass calcd for C 1 3 H15 NO2 m/z 217.1104, found

m/z 217.1104.

2 7 6

3-(2-Pyridlnyl)<2*cyclohexen-1-ol acetate (276). To a solution of 334 mg (1.54

mmol) of ester 34 in 6 mL of THF under an inert atmosphere was added 59 mg (0.155 mmol) of

b/s(benzonitrile)palladium (II) chloride. The yellow solution was heated at 65°C for 46 h during which the reaction became heterogenous. The solvent was removed in vacuo to yield a yellow- brown oil which was chromatographed over 1 2 g of silica gel (eluted with ethyl acetate-hexanes;

1:10) to yield 61 mg (18%) of unreacted starting material, and 141 mg (42%) of ester 35 as a pale yellow oil: IR (neat): 1732 cnr1; 1H NMR (CDCI3,200 MHz) 51.72-1.99 (m, 4H, CH 2CH2), 2.05(s,

3H, (C=0)CH3), 2.47-2.69 (m, 2H, =CCH2), 5.48 (m, 1H, OCH), 6.54 (dt, J= 3.8, 1.9 Hz, 1H,

=CH), 7.14 (ddd, J= 7.4, 4.8, 1.2 Hz, 1H, H5-). 7.42 (dt, 8.0, 1.0 Hz. 1H, H3-), 7.63 (td, J= 8 , 2

Hz, 1H, H4>), 8.54 (dq, J= 4.8, 0.9 Hz, 1H, H6>); 13C NMR (CDCI3, 62.9 MHz) 8 19.1 (t), 21.1 (q),

25.6 (t), 27.8 (t), 6 8 .6 (d), 119.5 (d). 122.2 (d), 125.2 (d), 136.2 (d), 141.1 (s), 148.7 (d), 157.4 (s),

170.5 (s); exact mass calcd for C-|3 H i5 N0 2 m/z 217.1104, found ntfz217.1095. 154

( H

2 8 0

2,3-D loxopiperidlne 3-phenylhydrazone (280).78 To a mixture of 7.08 g (41 mmol) of 3-carboethoxy-2-piperidone (279) in 80 mL of water was added 2.54 g (45 mmol) of potassium hydroxide. The resulting mixture was stirred at room temperature for 14 h, and was brought to pH 3 by the addition of 8 mL of 10% aqueous hydrochloric add while cooling to 0°C.

In a second flask, benzenediazonium chloride was prepared by the simultaneous addition of 4.21 g (45.2 mmol) of aniline and 80 mL (45.8 mmol) of a 0.57 M solution of aqueous sodium nitrite to a solution of 90 mL of 10% aqueous hydrochloric acid cooled in an ice/water bath. After

10 min, the mixture was partially neutralized by the addition of 80 mL of saturated aqueous sodium bicarbonate. The resulting mixture was added to the amido acid over a 10 min period, and after an additional 5 min, the yellow solution was brought to pH 5 by the addition of solid sodium acetate.

The reaction stirred at 0°C for 6.5 h, the yellow solid was collected and recrystallized from aqueous methanol to afford 6.36 g (76%) of hydrazone 280 as yellow needles: mp 235-240°C

(dec) (lit 7 8 244-245°C (dec)).

H o

2 8 1

2,3,4,9*T etrah yd ro >lH

(oil bath temperature = 130°C) tor 30 min. The hot solution was diluted with water until an oil separated from solution, and methanol was added until the oil returned to solution. The reaction 155

was cooled to yield 4.30 g (87%) of the desired p-carboline was an off-white solid: mp 182-185°C

(lit7 8 184°C).

H H O

2 8 2

2 l9 'D lh y d rO '1H -p yrld o -[3 l4-b]-lndol-one (282).77 To a mixture of 4.29 g (23.1

mmol) of p-carboline 281 in 80 mL of dioxane was added 6.45 g (28.4 mmol) of-2,6-dichloro-3,4- dicyanobenzoquinone, and the mixture was stirred overnight at room temperature. The reaction was concentrated in vacuo, the residue was washed with 80 mL of saturated aqueous sodium

bicarbonate, and the solid was collected by Buchner filtration. The solid was washed with water and recrystallized from aqueous methanol to yield 3.69 g (87%) of 1-oxo-p-carboline as a tan solid:

mp 251 -254°C (lit7 7 255-257°C).

H Bf

2 8 3

l-B ro m o -9 H -p y rld o [3 ,4 -b ]ln d o le (283). A mixture of 1.01 g (5.49 mmol) of p- carboline 282 1.59 g (34.5 mmol) of phosphorous oxybromide, 1.46 (3.39 mmol) of phosphorous pentabromide, and 160 mL of toluene was heated at 120°C for 15 h. The reaction was cooled to room temperature and quenched by the addition of 20 mL of water and saturated aqueous sodium bicarbonate until basic. To the biphasic mixture was added 30 mL of methanol, the mixture was extracted with three 200-mL portions of ether, the organic layers were combined, dried (MgSCXt), and concentrated in vacuo. The residual yellow solid was dissolved in methanol, solid-bound to 3 g of silica gel, and chromatographed over 35 g of silica gel (eluted with ethyl acetate-hexanes, 1:8) to yield 425 mg (31%) of 1-bromo-p-carboline 283 as a yellow solid: 1H NMR (DMSO-de, 300 MHz) 8 7.28 {td, J= 7, 1.2 Hz, 1H, H6), 7.58 (td, J= 8.3, 1.2 Hz, 1H, H 7 ),

7.65 (d. 8.1 Hz, 1H, Hfl), 8.14 (d, J= 5.1 Hz, 1H, H4), 8.16 (d, J= 5.1 Hz, 1H, H3), 8.23 (d, J= 8

Hz, 1H, H5 ), 11.8 (s, 1 H, NH); 13C NMR (DMSO-d6, 62.9 MHz) 8 112.5 (d), 114.9 (d), 120.1 (d),

121.0 (s), 122.1 (d), 123.9 (s), 128.7 (d), 129.3 (s), 134.7 (s), 138.1 (d), 140.6 (S); exact mass calcd. for C n H 7 N2 79Br m/z 245.9793, found m/z 245.9787. A portion of this material was recrystallized from ethyl acetate-hexanes to provide an analytically pure sample (mp 154-155°C).

Anal. Calcd. for C n H 7 N2 Br: C, 53.48; H 2.86. Found: C, 53.39; H, 2.91.

2 7 8

1-Chloro-9H-pyrido[3,4-b]indole (278).77 A mixture of 990 mg (5.38 mmol) of p- carboline 282, 541 mg (2.59 mmol) of phosphorous pentachloride, and 12 mL of phosphorous oxychloride was stirred and heated (oil bath temperature = 1 0 0 °C) for 15 h and cooled to room temperature. The mixture was poured over ice and neutralized with solid sodium bicarbonate and saturated aqueous sodium bicarbonate until basic by pH paper. The resulting yellow-brown solid was collected by suction filtration, dissolved in methanol, and solid bound to 7 g of silica gel. This material was chromatographed over 35 g of silica gel (eluted with ethyl acetate-hexanes, 1:9) to yield 813 mg (75%) of p-carboline 278 as a pale yellow solid: mp 174.5-176°C (lit 7 7 178-181°C).

286 290

1-lodo-9H-pyrldo[3,4-b]fndole (286) and 9-Acetyl*9H-pyrldo[3,4-b]indole

(290). To a mixture of 3.83 g (25.6 mmol) of flame-dried sodium iodide, 548 mg (2.71 mmol) of 157

1 -chloro-p-carboline (278), and 25 mL of dry acetonitrile was added 1.2 mL (16.8 mmol) of freshly distilled acetyl chloride. The resulting yellow slurry was heated and stirred at 74°C for 20 h, was cooled to room temperature, and was diluted with 25 mL of water. The iodine color was quenched by adding 2 mL of saturated aqueous sodium bisulfite and the mixture was neutralized by adding solid sodium bicarbonate. The organic components were extracted with two 75-mL portions of dichloromethane, and the combined organic layers were dried (MgS 0 4 ), concentrated in vacuo, and chromatographed over 15g of silica gel (eluted with ethyl acetate-hexanes, 1:8 (600 mL) to yield 322 mg (42%) of 1 -iodo-p-carboline (286) as a pale yellow solid: IR (CH 2 CI2 ), 1626, 1539,

1497, 1454 cm'1; 1H NMR (CDCI 3 , 300 MHz) 57.30 (ddd, J= 8.0, 6.3, 1.8 Hz, 1 H, H6), 7.55 (m,

2 H, H7 and H0), 7.84 (dd, J= 5.2, 0.5 Hz, 1 H, H4), 8.05 (dt, J= 8.0, 0.8 Hz, 1H, Ha), 8.23 (d, J =

5.2 Hz, 1 H, H3), 8.56 (s. 1H, NH); 13C NMR (CDCI3 . 75.5 MHz) 6 100.8 (s), 111.8 (d), 114.5 (d).

120.7 (d). 122.1 (d), 122.2 (s), 128.1 (S), 128.9 (d), 138.6 (s). 139.4 (s), 140.1 (d); exact mass calcd. for C1 1 H 7 IN2 m/z 293.9655, found m/z 293.9646. A portion of this material was recrystallized from hexanes-dichloromethane to yield pale yellow crystals, mp 131-133°C.

Continued elution with ethyl acetate yield 117 mg ( 2 1 %) of a tan solid identified as N-acetyl-p- carboline (290): IR (CH 2 CI2 ) 1694, 1619,1423, 1367 cm'1; 1H NMR (CDCI 3 , 300 MHz) 5 2.69 (s,

3H, CH3), 7.27 (t, J= 7.5 Hz, 1H, H6 or H7 ), 7.44 (td, J= 7.3,1.2 Hz, 1H, H6 or H7 ), 7.60 (dd, J =

5.1, 1 H, 1.0 Hz, H4 ), 7.78 (dt, J = 7.7, 0.6 Hz, 1 H, H5 or H3), 7.96 (d, J = 7.5 Hz, 1H, H5 or Hs),

8.44 (d, J= 5.1 Hz, 1H, H3), 9.38 (s, 1H, H-|); 13C NMR {CDCI 3 , 75.5 MHz) 527.3 (q), 113.7(d),

116.0 (d), 121.0 (d), 123.8 (d), 124.0 (S), 129.7 (d), 132.0 (S), 134.9 (s), 137.9 (s), 138.5 (S),

143.2 (d), 169.1 (s); exact mass calcd. fo rC i3 H io N 2 0 m/z 210.0794, found m/z210.0789.

291 l-B ro m o -9 -(m e th o x y m e th y l)-9 H -p y rld o [3 ,4 -b ]ln d o le (291). To a mixture ot

104 mg (2.6 mmol) of 60% sodium hydride in mineral oil (washed with two 3-mL portions of hexanes) in 4 mL of A/,/V-dimethyfformamide cooled in an ice/water bath was added 414 mg ( 1 .6 8 mmol) of 1-bromo-p-carboline in 4 mL of DMF. The resulting yellow-brown mixture was stirred in the cooling bath for 1 h, and 240 pL (3.2 mmol) of chloromethyl methyl ether was added. The reaction was stirred at room temperature for 3 h and was quenched by the addition of 5 mL of 1 N aqueous sodium hydroxide and 20 mL of water. The mixture was extracted with three 30-mL portions of dichloromethane, the combined organic layers were washed with 20 mL of brine, dried

(MgS0 4 ), and concentrated in vacuo. The residue was chromatographed over 25 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:12) to yield 324 mg ( 6 6 %) of the protected p-carboline

291 as a thick yellow oil: IR (neat) 3057, 1650, 1622, 1543 cm 1; 1H NMR (CDCI 3 , 300 MHz) 5

3.30 (S, 3H, CH3), 5.98 (s, 2H, CH2). 7.30 (m, 1H, H 6 ), 7.57 (m, 2H, H7 and Hg), 7.79 (d, J= 5.1

Hz, 1H, H4 ), 8.00 (td, J= 7.9, 0.8 Hz, 1H, H5), 8.19 (d, J= 5.1 Hz, 1H, H3); 13C NMR (CDCI3, 75.5

MHz) 8 55.6 (q), 74.1 (t), 110.7 (d), 114.0 (d), 121.1 (S), 121.26 (d), 121.34 (d), 129.2 (d), 132.2

(S), 132.6 (s), 133.1 (s), 138.8 (d), 142.0 (s); exact mass calcd for C1 3H 1 iN 2 0 79Br m/z290.0055, found m/z 290.0062.

2 9 2

(±)-1-[9-(Methoxymethyl)-9H*pyrldo[3,4-b]lndoM-ylI-2-cyclohexen-l-ol

(292). To a solution of 2 mL of diethyl ether and 1.6 mL (1.88 mmol) of f-butyllithium in pentane cooled in a dry ice/acetone bath was added 250 mg (0.86 mmol) of p-carboline 291 in 2 mL of ether dropwise. The resulting deep yellow mixture was stirred for 50 min, and 200 pL (2.1 mmol) of 2-cyclohexen-l-one was added. After 10 min, the cooling bath was removed and the reaction 159

stirred at room temperature for 2.5 h. To the resulting slurry was added 10 mL of water, the

mixture was extracted with two 50-mL portions of ether, dried (MgS 0 4 ), and concentrated in

vacuo. The residue was chromatographed over 35 g of silica gel (eluted with ethyl acetate-

hexanes, 1:10 (225 mL) to ethyl acetate-hexanes, 1:5) to afford 175 mg ( 6 6 %) of alcohol 292 as

a pale yellow oil: IR (neat) 3226, 1620 cm'1, 1H NMR (CDCI 3 , 250 MHz) 5 1.87 (m, 2H,

CH 2 CU2 CH 2 ). 2.21-2.42 (m, 4H, C tb C ^ C h b ). 3.23 (s, 3H, CH3), 5.85 (d, J= 10.5 Hz, 1 H,

NCH2), 6.01 (d, J = 10 Hz, 1H, CCH=), 6.12 (d, J= 10 Hz, 1 H, NCH2), 6.17 (dd, 10, 4 Hz, 1H,

CH2 CH=), 7.35 (ddd, J = 8 , 5.5, 2.5 Hz, 1 H, H6), 7.59 (m, 2 H, H7 and H0), 7.96 (d, J = 5.1 Hz, 1 H,

H4), 8.11 (d. J= 8 Hz, 1H, H5), 8.43 (d, J = 5.1 Hz, 1H, H3), OH missing in 1H NMR; 13C NMR

(CDCI3> 62.9 MHz) S 18.8 (t), 24.7 (t), 34.9 (t), 55.2 (q), 70.5 (s), 76.4 (t), 111.4 (d). 114.2 (d),

121.0 (d), 121.1 (d), 122.3 (s). 128.8 (d), 131.1 (d), 132.2 (d), 132.6 (S), 133.6 (s), 136.6 (d),

142.4 (s), 147.9 (s); exact mass calcd for C1 9H2 0 N2 O2 m/z 308.1525, found m/z 308.1524.

2 9 3

1-lodo-9-methyl-9/f-pyrldo[3,4‘d]fndole (293). To a solution of 126 mg (0.43 mmol) of 1-iodo-p-carboline (286) in 5 mL of THF at rl was added 37 mg (1.54 mmol) of sodium hydride in one portion. The resulting yellow-orange mixture was stirred at rt for 45 min, and 35 pL

(0.55 mmol) of methyl iodide was added. The reaction was stirred at rt for 2.5 h, quenched by the addition of 0.5 mL of methanol, and was poured into 25 mL of water. The mixture was extracted with 120 mL of dichloromethane, the organic layer was dried (M gS04), and concentrated in vacuo. The residue was chromtographed over 12 g of flash silica gel (eluted with ethyl acetate- hexanes. 1:20 ) to yield 1 1 2 mg (85%) of iodide 293: IR (CH 2 CI2 ) 1622 cm'1; 1H NMR (CDCI3,

300 MHz) 8 4.26 (S, 3H, CH3), 7.31 (td, J= 7.1, 0.8 Hz, 1H, H 6 ), 7.47 (d, J = 8 .4 Hz. 1H, H7 ), 7.63

(ddd, J - 8.2, 7.0, 1.2 Hz, 1H, H6), 7.88 (d, J= 5.0 Hz, 1 H, H4), 8.09 (dt, J= 7.9, 0.8 Hz, 1H, H5), 8.18 (d, J - 5.0 Hz, 1H, H3); 13C NMR (CDCI 3 , 75.5 MHz) 5 31.7 (q), 96.6 (s), 109.8 (d). 114.1 (d),

120.0 (s). 120.4 (d), 121.2 (d), 128.8 (d), 130.0 (s), 137.2 (s). 139.3 (d), 142.5 (s); exact mass calcd for C12H9 IN2 01*307.9812, found 01*307.9804.

,OH

2 9 4

(±)-1-[9-Methyl-9H-pyrldo[3,4-b]lndol-1-yl]-2-cyclohexen-1-oL (294). To a solution of 27 mg (0.088 mmol) of p-carboline 293 in 2 mL of dry THF cooled in a dry ice acetone bath was added 70 pL (0.11 mmol) of o-butyllithium via syringe. The resulting bright yellow mixture was stirred in the cooling bath for 35 min, and 22 pL (0.23 mmol) of 2-cyck>hexen-1 -one was added to the mixture via syringe. The reaction was stirred in the cooling bath for 1 h 30 min, warmed to room temperature over 20 min, and was poured in to 25 mL of water. The mixture was extracted with 50 mL ol dichloromethane, the organic layer was dried (MgS 0 4 ), and concentrated in vacuo to yield a brown oil. This material was chromatographed over 6 g of silica gel (eluted with ethyl acetate-hexanes, 1:15 (160 mL) to ethyl acetate-hexanes, 1:7) to yield 8 mg (33%) alcohol

294 as a pale yellow film; IR (neat) 3208 cm 1; 1H NMR (COCI 3 ), 300 MHz) 5 1.84-1.94 (m, 2H,

CH2 CH2 CH2), 2.14-2.44 (m, 4H, CtteC H ^tfc). 4.20 (S, 3H, CH3), 6.04 (d, J - 10.0 Hz, 1H,

CCH-), 6.16 (ddd, J - 10.0, 4.4, 3.7 Hz, 1 H, CHCH-CfcD, 7.32 (t, J - 7.8 Hz, 1 H, H6 ). 7.45 (d, J -

8.4 Hz, 1H, Hg), 7.63 (t, J = 8.2 Hz, 1H, H 7), 7.98 (d, 5.1 Hz. 1H, H4), 8.15 (d, J = 7.9 Hz, 1 H,

H5), 8.37 (d, 5.1 Hz, 1H, H3), OH is missing in 1H NMR;13 C-NMR (CDCI3 , 75.5 MHz) 5 18.9 (t),

24.7 (t), 34.1 (q), 37.2 (t), 70.0 (s), 109.9 (d), 114.2 (d), 120.0 (d), 120.9 (d), 121.1 (S), 128.5 (d),

131.0 (d), 131.3 (s), 133.0 (d), 134.3 (s), 135.0 (d), 142.7 (s). 147.3 (s); exact mass calcd for

C i8H18N20 m/z 278.1420, found m/z 278.1425. 161

OAc

2 9 6

(+)-3-[9-(Methoxymethyl)-9M-pyrldo[3,4-£>]indoM-yl]-2-eyelohexan-1-ol acetate (ester) (296). A mixture of 125 mg (0.41 mmol) of alcohol 292, 0.70 mL (5.0 mmol) of triethylamine, 0.45 mL (4.8 mmol) of acetic anhydride, and 7 mg (0.057 mmol) of 4 -N,N- dimethylaminopyridine was stirred under Ar gas for 12 h. An additional 7 mg (0.057 mmol) of 4-

A/,Af-dimethylaminopyridine was added, and the reaction was stirred for 24 h. The mixture was concentrated in vacuo, and the residue was chromatographed over 1 0 g of silica gel (eluted with ethyl acetate-hexanes, 1:10 (100 mL) to ethyl acetate-hexanes, 1:5) to yield 90 mg (63%) of acetate 296 as a yellow oil: IR (neat) 1731,1620cm'1, 1H NMR (CDCI 3 , 250 MHz) 81.91-2.10 (m,

4H, CHOCtbCtb), 2.08 (s, 3H, COCH3 ), 2.63 (m, 2 H, =CCH2), 3.16 (S, 3H. OCH3 ), 5.52 (m, 1 H,

CHO), 5.76 (d, 10.5 Hz, 1H, OCH2 0), 5.83 (d, J= 10.5 Hz, 1 H, OCH2 0), 5.95 (m, 1H, =CH),

7.33 (ddd, 8 , 5. 3 Hz, 1H, H6), 7.59 (m, 2H, H 7 and H8), 7.89 (d, J = 5 Hz, 1H, H4), 8.12 (d, J =

8 Hz, 1H, Hs), 8.48 (d, J = 5 Hz, 1H, H3); 13C NMR (CDCI3 , 62.9 MHz) 8 19.3 (t), 21.3 (q), 27.9 (t),

28.7 (t). 55.6 (q), 68.2 (d), 74.9 (t), 110.5 (d), 113.5 (d), 120.8 (d), 121.5 (d), 121.8 (s), 126.9 (d),

128.8 (d), 131.3 (s), 133.0 (s), 139.3 (d), 142.4 (s), 143.0 (S), 145.4 (s), 170.7

OAc

OMOM OTHP

301a 162

(±)-(4a/?*,5S*,8a»*)-3,4,5,6,8a-Hexahydro-5-hydroxy-8a-l(2S*,3Z)-2-

(methoxymethoxy)-8-((t®trahydro-2H-pyran-2-yl)oxy]-3-octenyl]-2-(p-molhoxy- phenyl)-1(2H)-lsoqulnolone acetate (eater) (301a). To a solution of 510 mg (0.946 mmol) of alcohol 188 in 20 mL of dichloromethane was added 636 mg (600 pL, 7.90 mmol) of chloromethyl methyl ether and 1.2 mL ( 6 . 8 mmol) of diisopropylamine. The resulting yellow- orange solution was stirred at room temperature for 22 h, and the reaction was diluted to 300 mL with dichloromethane. The mixture was washed with 50 mL of 10% aqueous hydrochloric acid and 50 mL of saturated aqueous sodium bicarbonate. The organic layer was dried (MgS 0 4 ), concentrated in vacuo, and was chromatographed over 2 0 g of silica gel (eluted with ethyl acetate- hexanes, 1:3) to yield 424 mg (77%) of acetate 301a as a thick yellow oil: IR (neat) 1729,1644,

1511 cm"1; 1H NMR (CDCI3, 300 MHz) 8 1.49-1.95 (m, 13H, CH2 manifold), 2.07 (s, 3H, COCH3),

2.09-2.24 (m. 4H, CH 2 manifold), 2.33 (dt, J = 17.8, 6 .2 Hz, 1 H, AcOCHChl2). 2.76 (m, 1H.

AcOCHCJd2).3-32 (s, 3H. CH2OCH3), 3.33-3.73 (m. 5H, CtfcOCHOCtte and NCtte), 3.75 (S. 3H.

ArOCH3 ), 3.81 (m, 1H, NCH2 ), 4.55 (m, 1 H, OCHO), 4.55 (d, J- 6.0 Hz, 1H, OCH2 O), 4.59 (d, J =

6.0 Hz, 1 H, OCH2O), 4.77 (td, J = 9.6, 4.3 Hz, 1H, MOMOCH), 5.14 (ddd, J - 10.4, 6.4, 3.8 Hz,

1 H, CHOAc), 5.30 (dd, J= 10.3, 9.3 Hz, 1 H, MOMOCJdCH=), 5.54 (m, 2H, MOMOCHCH=CH and

AcOCHCH2 C £H . 5.73 (dt, 10.0, 1.2 Hz, 1H, COCCH=), 6.84 (d, J= 9.0 Hz, 2H, ArH), 7.13 (d,

J - 9.0 Hz, 2H, ArH); 13C NMR (CDCI3 , 75.5 MHz) 5 19.5 (t), 20.1 (t), 21.1 (q), 25.4 (t), 26.3 (t),

26.7 (t), 27.1 (t), 29.2 (t), 30.7 (t), 36.6 (d), 42.8 (t), 49.5 (s), 50.4 (t), 55.3 (q), 55.9 (q), 62.1 (t),

67.3 (t), 68.2 (d), 68.7 (d), 94.1 (t), 98.6 (d), 114.1 (d), 122.0

133.5 (d), 136.9 (s), 157.7 (s), 169 9 (s), 172.5 (s); exact mass calcd. for C3 3 H 4 7 NO 8 m/z

585.3303, found m/z 585.3303. 163 OH

H,C< OTHP

3 0 1 b

(±)-(4aJ?\ 5S*,8a/?*)-3,4,48,5,6,8a-Hexahydro-5-hydroxy-8a-[(2S*,3Z)-2-

(methoxymethoxy)-8-[(tetrahydro-2W-pyran-2-yl)oxy]-3-octenyl]-2*(p-methoxy-

p h e n y l)-i( 2 H)-isoqulnolone (301b). To a solution of 307 mg (0.527 mmol) of acetate

301a in 10 mL of methanol, 10 mL of tetrahydrofuran and 5 mL of water was added 41 mg (0.976

mmol) of lithium hydroxide monohydrate, and the reaction was heated at 35°C tor 1.5 h. The

mixture was concentrated to approximately one-fifth of the original volume, diluted to 100 mL with

dichloromethane, washed with 25 mL of water, dried (MgS 0 4 ), and concentrated in vacuo. The

residue was chromatographed over 15 g of silica gel (eluted with ethyl acetate-hexanes, 1:1) to

yield 268 mg (94%) of alcohol 301b as a very thick yellow oil: IR (neat) 3418,1633,1606 cm'1;

1H NMR (CDCI3 , 300 MHz), S 1.42-1.87 (m, 14H, CH 2 manifold and OH), 2.08-2.18 (m, 3H, CH 2

manifold), 2.33 (m, 1H, CtteCHOH), 2.60 (m, 1H, ACOCHCH 2 ), 2.77 (dd, 14.7. 9.9 Hz, 1H,

MOMOCHCU2 ), 3-36 (s, 3H, CH2 OCH3 ), 3.38-3.77 (m, 5H, CtfcOCHOCtb and NCH2), 3.79 (S,

3H, A1OCH3 ), 3.88 (m, 1H. NCH 2 ). 4.13 (m, 1H, CtlOH), 4.55 (d, J- 6.2 Hz, 1H, OCH2 O), 4.59

(m, 1H, OCHO), 4.65 (d, J * 6.2 Hz. 1H, OCH2 O), 4.72 (m, 1H, MOMOCH). 5.34 (dd, 10.7,9.2

Hz, 1H, MOMOCHCtH, 5.58 (m, 2 H, MOMOCHCH-Ctl and HOCHCH2 CB-), 5.75 (d, J - 1 0 .1

Hz, 1H.COCCH-), 6 .8 6 (d, J - 9.0 Hz, 2H, ArH), 7.16 (d. J- 9.0 Hz, 2H, ArH); 13C NMR (75.5

MHz, CDCI3 ) 5 (19.41 (t), 19.47 (t)J, 19.6 (t), [25.30 (t), 25.32 (t)J, 25.8 (t), (26.97(t), 27.03 (t)].

[28.78 (t), 28.91 (t)], 30.4 (t), 30.6 (t), [39.65 (d), 39.84(d)], 43.1 (t), 49.7 (s), 50.8 (t), 55.3 (q),

55.7 (q), [62.3 (t), 62.6 (t)], [65.2

(d)], 114.1 (d), 122.9 (d). [130.19 (d), 130.24(d)], 132.8 (d), 133.1 (d). 133.3 (d), 137.0 (s), 157.7

(s), 172.9 (s): exact mass calcd. for C3 1 H4 5 NO7 m/z 543.3198, found m/z 543.3198. 164

OMOM LJ" 3 0 2

(±)-(4a/?*,8a/7*)-2t3,4>4a,6,Ba-H«xahydro-8a-[(2S*l3Z)-2-(m«thoxy-

methoxy)-8-[(tetrahydro-2H-pyran-2-yl)oxy]-3-octenyl]-2-(p-methoxyphenyl)-

1,5-isoqulnolinedione (302). To a solution of 0.45 mL (5.14 mmol) of oxalyl chloride in 50 mL of dichloromethane cooled in a dry ice-acetone bath was added 0.80 mL (11.3 mmol) of dimethylsulfoxide via syringe. The resulting mixture was stirred in the cooling bath for 1 h 15 min, followed by addition of 1.36 g (2.51 mmol) of alcohol 301b in 50 mL of dichloromethane. The resulting mixture was stirred in the cooling bath for 1 h 30 min, and 2.3 mL (16 mmol) of triethylamine was added. The reaction was stirred in the cooling bath for 30 min and at room temperature for 1 h. The mixture was diluted to 300 mL with dichloromethane, washed with 50 mL of 10% aqueous hydrochloric acid, 50 mL of saturated aqueous sodium bicarbonate, dried

(MgS0 4 ), and concentrated in vacuo. The residue was chromatographed over 40 g of silica gel

(eluted with ethyl acetate-hexanes, 1:1) to yield 1.04 g (77%) of a thick yellow oil identified as p,y- unsaturated ketone 302: IR (neat) 1715, 1650 cm' 1 ; 1 H-NMR (CDCI3 , 300 MHz)* S 1.40-1.82

(m, 1 1 H, CH2 manifold), 1.96-2.14 (m, 3H, OCHCH=CHCJd2 and COCH), 2.36 (m, 1 H, CH2 manifold), 2.55 (dd, J= 14.7, 9.7 Hz, 1 H, MOMOCHCH2 ). 2.85-3.05 (m, 2 H, COCH2 ), 3.32 (s, 3H,

CH2 OCH3 ). 3.35 (m, 2 H, OCH2), 3.45 (m, 1 H, NCH2 ), 3.68 (m, 2 H, OCH2), 3.75 (s, 3H, ArOCH3 ),

3.81 (m. 1H, NCH 2 ), 4.47 (d, J = 6.2 Hz, 1H, OCH2 O), 4.54 (m, 1H, OCHO), 4.58 (d, J= 6.2 Hz,

1H, OCH2 0 ). 4.68 (Id. J = 9.6, 3.5 Hz, 1 H, MOMOCH), 5.25 (dd, J = 10.9, 9.2 Hz, 1H,

MOMOCHCJJ=). 5.55 (dt,J= 10.9, 7.4 Hz. 1 H. MOMOCHCH=CH), 5.79 (dt. J= 1 0 .1 , 3.5 Hz, 1 H,

COCH2 CH), 5.97 (d, J= 1 0 .1 Hz. 1 H, COCH2 CH=Cfcl), 6.85 (d, J= 8.9 Hz, 2 H. ArH), 7.10 (d, J = 165

8.9 Hz, 2H, ArH), 7.10 (d, J = 8.9 Hz. 2H, ArH); 1 3 C-NMR (CDCI3, 75.5 MHz) 5 19.5 (t), 22.6 (t),

25.4 (t), 26.2 (t), 27.2 (t), 29.3 (t), 30.6 (t), 37.6 (t), 43.7 (t). 49.8 (d), 50.0 (t), 50.9 (s), 55.3

55.8 (q), 62.1 (I), 67.2 (t), 68.2 (d), 93.8 (t). 98.7 (d), 114.2 (d), 122.5 (d), 127.1 (d), 129.5 (d),

132.7 (d), 133.6 (d), 136.1 (s), 157.9 (s), 171.7 (s), 208.8 (s); exact mass calcd. for C3 1 H4 3 N0 7 m/z 541.3041, found m/z 541.3042.

OMOM OTHP

3 0 0

(±)-{4a/?*t8a/T*)-2,3f4,4a,8,8a-Hexahydro-8a-[(2S#t3Z)-2-(methoxymethoxy)-8-[(tetrahydro-2W-pyran-2-yl)oxy]-3-oct«nyl]-2-(p-methoxyphenyl)-

1,6-iaoqulnolinediona (300). To a solution of 1.03 g (1.90 mmol) of ketone 302 in 1 0 0 mL of dichloromethane was added 430 pL (2.89 mmol) of 1 ,8-diazabicyclo[5.4.0]undec-7-ene. The resulting mixture was stirred at room temperature for 2.5 h, diluted with 300 mL of dichloromethane, and washed with 50 mL of 1 0 % aqueous hydrochloric acid and 50 mL of saturated aqueous sodium bicarbonate. The organic layer was dried (MgS 0 4 ), concentrated in vacuo, and chromatographed over 30 g of silica gel (eluted with ethyl acetate-hexanes, 1:3 (400 mL) to ethyl acetate-hexanes, 3:2) to yield 794 mg (77%) of a,p-unsaturated ketone 300 as a very thick yellow oil: IR (neat) 1667, 1651 cm'1; 1 H-NMR (CDCI3 , 250 MHz)* 5 1.42-1.82 (m, 11H,

CH2 manifold), 2.02-2.13 (m. 2 H, =CHChl2 CH2 ). 2.14-2.22 (m, 1 H. CH2 manifold), 2.42 (m, 1 H,

CH2 manifold), 2.45 (dd, J = 14.8, 9.6 Hz, 1 H. MOMOCHCH2 ). 2.55 (dt, J = 19.3, 1 .8 Hz, 1 H.

COCH=CHCH2 ), 2.99 (dt, J= 19.3, 1 .8 Hz, 1 H, COCH=CHCH2 ), 3.17 (dd, J = 8.5, 3.6 Hz, 1 H,

COCH), 3.32-3.40 (m, 1 H, OCH2 ), 3.36 (s, 3H, CH20Ctl3). 3 46-3.50 (m, 1 H, NCH2 ), 3.62-3.88

(m, 4H, NCH2 and CifcOCHOCtfc). 3-76 (s, 3H, A1OCH3 ), 4.49 (d, J= 6.4 Hz. 1 H, OCH2 O), 4.54 166 (m, 1H, OCHO), 4.62 (d, J = 6.4 Hz, 1 H, OCHgO), 4.68 (m, 1H, MOMOCH), 5.26 (t, J = 10.7 Hz,

1 H, OCHCH=), 5.56 (dt, 10.7, 7.4 Hz, 1 H, OCHCH=Ctl), 6 09 (d, J = 1 0 .2 Hz, 1 H, COCH=),

6.85 (d, J = 8 . 8 Hz, 2H, ArH), 6.95 (dt, J = 10.1, 4.0 Hz, 1H. COCH=CtD, 7.04 (d, J= 8.9 Hz, 2H,

ArH); 1 3 C-NMR (CDCI3 , 75.5 MHz) S 19.5 (I), 21.4 (t), 25.4 (t), 26.2 (t), 27.2 (t), 29.2 (t). 30.6 (t),

33.4 (t), 41.0 (t), 45.5 (s), 47.4 (d), 50.0 (t), 55.3 (q), 56.0 (q), 62.1 («), 67.2 (t), 68.0 (d). 93.9 (t),

98.7 (d), 114.3 (d), 127.2 (d), 128.4 (d), 129.5 (d), 133.7 (d), 136.1 (s), 148.6 (d), 158.1 (s).

172.8 (s), 198.3 (s); exact mass calcd. for C 3 1 H4 3 NO 7 m/z 541.3041, found m/z 541.3056.

LJ-'5 — OMOM 3 0 4

(±3>(4aff*,8a5*)-3,4,4a,5,8l8a-Hexahydro>5>hydroxy-8a-[(25*,32r)-2-

(methoxymethoxy)-8-[(tetrahydro-2tf-pyran-2-yl)oxy]-3-octenyl]-2-(p-methoxyphenyl)-5-(2-pyridyi)-1(2H)-isoquinolone (304). To a solution of 100 pL of 2- bromopyridine in 5 mL of dry THF cooled in a dry ice-acetone bath was added 630 pL (1.0 mmol) of o-butyllithium in hexanes. The resulting red mixture was stirred in the cooling bath for 2.5 h and ketone 300 in 2 mL of dry THF was added via syringe. The reaction mixture was stirred in the cooling bath for 30 min, opened to the atmosphere, quenched by adding 1 mL of methanol, and poured in 30 mL of water. The mixture was extracted with 120 mL of dichloromethane, the organic layer was dried (MgS 0 4 ), and concentrated in vacuo to yield a yellow oil. This residue was chromatographed over 15 g of silica gel (eluted with ethyl acetate-hexanes, 1:1) to yield 32 mg

(33%) of allylic alcohol 304 as a pale yellow oil: IR (neat) 3417, 1644 cm '1; 1H NMR (CDCI 3 , 250

M Hz)* 6 1.42-1.84 (m, 1 2 H. CH2 manifold), 1.93 (dd. J = 14.8, 4.0 Hz, 1 H, MOMOCHCtb), 2.04 167 (m, 2 H, MOMOCHCH=CHCH2 ). 2.18 (dd, J= 14.8, 8 .0 Hz, 1 H, MOMOCHCtfc), 2.27 (dm, J =

18.2, 1 H, CCH=CHCtl2 ). 2.73 (dd, J = 6 .1 , 4.4 Hz, 1 H, COCH), 3.21 (dd, J= 18.2, 4.4 Hz, 1 H.

C C H =C H C Ji2 ). 3.31 (s, 3H, CH2 OCtb), 3 33-3.52 and 3.65-3.90 (m, 6 H, NCH2 and

CtbOCHOCtfc), 3.79 (s. 3H, ArOCH3), 4.43 (d, J =6 .6 Hz, 1H, 0CH 2 0), 4.47 (m, 1H, MOMOCH).

4.57 (m, 1 H.OCHO), 4.61 (d, J=6.5Hz, 1 H, O C H ^ ), 5.05 (brs, 1 H, OH), 5.19 (dd, J= 1 0 .8 , 9.5

Hz, 1 H, MOMOCHCH=). 5.50 (dt, J= 1 1 .2 , 1 .2 Hz, 1 H, MOMOCHCH=CtD 5.63 (d, J = 9.9 Hz. 1 H,

COCH=), 6.09 (ddd, J= 9.9, 5.1, 2.5 Hz, 1 H, COCH=Ctf) 6.87 (d, J = 8.9 Hz, 2 H, ArH), 7.19-7.26

(m, 3H, ArH, NCHCM). 7.45 (d, J= 8.0 Hz. 1H. NCCH), 7.72 (td, J = 7.8, 1.6 Hz, 1H, NCCHCfci),

8.56 (d, J = 4.7 Hz, 1H, NCH); 13C NMR(CDCI3, 75.5 MHz) 8 18.5 (t), 19.6 (t). 25.4 (t), 26.2 (t),

27.5 (t), 29.3 (t), 30.7 (t), 31.2 (t). 43.2 (s), 43.9 (t), 45.4 (d), 50.2 (t). 55.3 (q), 56.0 (q), 62.3 (t),

67.3 (t), 68.0 (d), 73.9 (s), 93.7 (t), 98.8 (d), 114.1 (d), 120.8 (d), 122.3 (d), 127.6 (d), 129.3 (d),

129.8 (d), 133.2 (d), 137.2 (d), 137.3 (s), 147.2 (d), 157.8 (s), 164.4 (s), 173.1 (s), missing one sp2 doublet; exact mass calcd for C 3 eH 4 gN2 0 7 m/z 620.3464, found m/z 620.3462. n

3 0 7

(±)-{4a/?*,7aSM4a»M SaS^-S-m-Dfthtan^-yl^.S^^a.SJa.O.IO.11,12,

14a,15-dodecahydro-5-hydroxy-2-(p-methoxyphenyl)-1 M-azocino[1 \2'-1,5J- pyrrolo[2,3-/]isoquinolin-1-ona. (307). To a solution of 63 mg (0.52 mmol) of 1,3- dithiane in 3 mL of THF cooled in a dry ice-acetone bath was added 0.30 mL (0.48 mmol) of 1 .6 M n-butyllithium in hexanes via syringe. The resulting mixture was stirred in the cooling bath for 45 min, and 6 8 mg (0.173 mmol) of enone 151 in 3 mL of THF was added dropwise. The resulting 168 yellow mixture was stirred in the cooling bath for 30 min, and was then quenched at low temperature with 1 mL of methanol. The resulting mixture was concentrated in vacuo to yield a

brown oily residue. This material was chromatographed over 8 g of activity II basic alumina (eluted with ethyl acetate-hexanes, 1:5 (100 mL), to ethyl acetate-hexanes, 1:1) to yield 38 mg (40%) of alcohol 307 as an orange foam: IR (neat) 3426, 1633 cm'1; 1H NMR (CDCI 3 , 300 MHz)* 5 1.25-

2.59 (m, 15H, CH 2 manifold and OH), 2.82-3.03 (m, 4H, CH 2 manifold), 3.11 (m, 1H, C(OH)CH),

3.59 (dt, 1 2 .1 , 5.5 Hz, 1 H, NCH2), 3.77 (s. 3H. OCH3 ), 3.90 (dt, J= 11.7, 8.2 Hz. 1H, NCH2),

4.06 (br s, 1H, NCHC), 4.18 (br s, 1 H, NCHCH=), 4.55 (s, 1 H, SCHS), 5.58 (t, J = 10.3 Hz. 1H.

NCHCH=). 5.91 (q, J= 1 0 .2 Hz, 1 H, NCHCH=Cfcl), 6.05 (dd, J= 10.3, 3.3 Hz, 1 H. C(OH)CH=CJd),

6.27 (d, J= 10.3 Hz. 1H, C(OH)CU=), 6 . 8 6 (d, J= 9.0 Hz, 2 H, ArH), 7.11 (d, 9.0 Hz, 2H. ArH);

13C NMR (CDCI3 , 75.5 MHz) 8 19.6 (t), 26.9 (t), 28.0 (t), 28.1 (t), 30.5 (t), 30.7 (t). 42.3 (d), 45.3 (t),

48.3 (t), 49.0 (s), 49.4 (t), 55.3 (q), 55.6 (d), 56.5 (d), 59.8 (d), 74.3 (s), 114.2 (d), 127.3 (d),

128.0, 129.0, 131.5, 133.0, 137.1 (s), 157.9 (s), 175.1 (s); exact mass calcd for C2 sH3 6 N2 S2 0 3 m/z 512.2170, found m/z 512.2179.

3 0 9

(±)-1-m-Dlthian-2-yl-2-eyclohaxen-1-ol acetate (309). To a solution of 850 mg

(3.94 mmol) of alcohol 306 in 1 0 mL of dichloromethane was added 1 .2 mL (12.7 mmol) of acetic anhydride, 32 mg (0.26 mmol) of 4-N,N -dimethylaminopyridine, and 1.6 mL (11.6 mmol) of triethylamine. The resulting yellow mixture was stirred at room temperature for 20 h, diluted to 100 mL with dichloromethane, washed with 2 0 mL of 1 0 % aqueous hydrochloric acid, 2 0 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo. The residue 169 was chromatographed over 35 g of silica gel (eluted with ethyl acetate-hexanes, 1:12) to yield 770

mg (76%) of acetate 309 as a clear, colorless oil: IR (neat) 1726 cm *1; 1H NMR (CDCI 3 , 300 MHz)

8 1.64-2.19 (m, 8 H, CH2 manifold), 1.99 (s. 3H, CH3), 2.77-2.93 (m, 4H, (SCH2)2), 5.20 (s, 1H,

CHS2), 5.82 (ddd, 10, 4.5, 2.7 Hz, 1H. CH 2 CH), 6.11 (dm, J = 10 Hz, 1H, CCH); 13C NMR

(CDCI3, 75.5 MHz) 6 17.8 (t), 21.8 (q), 24.7 (t), 26.0 (t), 29.8 (t), 30.8 (1), 31.1 (t), 55.3 (d), 80.9 (s),

127.0 (d), 133.8 (d), 170.2 (s); exact mass calcd. for C i2 H ie 0 2 S2 m/z 258.0748, found m/z

258.0751.

3 1 0

(±)-3-m-Dlthlan-2-yl-2-cyclohexen-1-ol acetate (310). A mixture of 135 mg

(0.523 mmol) of acetate 309, 41 mg (0.11 mmol) of b/s(benzonitrile)palladium (II) chloride, and 7 mL of THF was stirred and heated at 75°C for 72 h. The reaction was cooled to room temperature, concentrated in vacuo, and chromatographed over 1 0 g of silica gel (eluted with ethyl acetate- hexanes, 1 :1 0 ) to afford 77 mg (57 %) of acetate 310 as a pale yellow oil: IR (neat) 1730 cm*1; 1H

NMR (CDCI3 , 300 MHz) S 1.62-1.89 (m, 4H, SCH 2CH2 . CHCH2 CJd2 ), 2 .0 0 (s, 3H. CH3), 2 .0 2 -2 .2 2

(m, 4H, CHCH2 CH2 Chj2 ). 2.81-2.96 (m, 4H, (SCH 2 )2 ), 4.51 (s, 1 H, S2 CH), 5.27 (m, 1 H, CHO),

5.88 (m, 1H, CH=C); 13C NMR (CDCI 3 , 75.5 MHz) 8 19.1 (t), 21.2 (q), 25.3 (t), 26.8 (t). 27.8 (t),

31.2 (t), 52.8 (d), 67.9 (d), 124.5 (d), 141.3 (s), 170.5 (s); exact mass calcd. for C i2 H i8 0 2 S2 m/z

258.0748, found m/z 258.0740. 170 3 1 3

(±)-3-m -Dithian-2-yl-2-cyclohexen~1-one (313). To a solution of 1.15 g (9.58 mmol) of 1 ,3-dithiane in 30 mL of dichloromethane cooled in a dry ice-acetone bath was added

6 .1 mL (9.76 mmol) of 1 .6 M n-butyllithium in hexanes. The resulting mixture was stirred in the cooling bath for 1 .5 h, and 1.38 g (9.72 mmol) of 3-ethoxy-2-cyck>hexen-1 -one in 1 0 mL of THF was added over 3 min. The resulting mixture stirred in the cooling bath for 5 min. the bath was removed, and the reaction mixture was stirred at room temperature for 3 h. To the mixture was added 30 mL of 10% aqueous hydrochloric acid, the resulting mixture was stirred for 30 min, and the mixture was extracted with 400 mL of dichloromethane. The organic layer was washed with

100 mL of water, 1 0 0 mL of saturated aqueous sodium bicarbonate, dried (MgSC^), and concentrated in vacuo. The residue was chromatographed over 60 g of silica gel (eluted with ethyl acetate-hexanes, 1:10) to yield 1.37 g (67%) of enone 313 as a yellow oil: IR (neat) 1681 cm'1; 1H NMR (CDCI3, 300 MHz) 5 1.81-2.15 (m, 4H, SCH 2 CH2 and C O C H ^tla), 2.37 (t, J = 6 . 6

Hz. 2H. CH2 C=), 2.49 (t. J = 6 . 6 Hz, 2H, COCH2), 2.89 (m, 4H, CJ^SCHSCHa), 4.63 (s, 1H,

SCHS). 6.18 (s, 1H, =CH); 13C NMR (CDCI3 , 75.5 MHz) 8 2 2 .6 (t), 25.1 (t), 281. (t), 30.4 (t), 37.3

(t). 51.8 (d), 127.5 (d), 160.4 (s), 199.2 (s).

3 1 4

(±)-3-m-Dithian-2-yl-1-methyl-2*cyclohex«n-1-ol (314). To a solution 5.75 g

(26.9 mmol) of enone 313 in 300 mL of THF cooled in a dry ice/acetone bath was added 30 mL

(75 mmol) of a 2.5 M solution of methyl magnesium bromide in ether. The resulting mixture was stirred in the cooling bath for 1 h and at room temperature for 7 h. The reaction was quenched with 200 mL of saturated aqueous ammonium chloride, extracted with 900 mL of 171 dichloromethane, dried (MgS 0 4 ), and concentrated in vacuo. The residue was chromatographed over 200 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:6) to yield 5.55 g (90%) of alcohol 314 as a clear yellow oil: IR (neat) 3420 cm'1; 1H NMR (CDCI 3 , 300 MHz) 5 1.25 (s, 3H,

CH3), 1.54-1.73 (m, 4H, S C H ^ tte, CCH 2 CH2 ). 1-79 (m, 2 H, C(CH3 )CJd2 ). 2.04-2.15 (m, 3 H,

=CCH2 and OH), 2.79-2.95 (m, 4H, (SCH2)2), 4.47 (s, 1H, S 2 CH), 5.79 (s. 1 H, =CH); 13C NMR

(CDCI3, 75.5 MHz) 5 19.7 (t), 25.4 (t), 27.1 (t), 29.1 (q), 31.3 (t), 37.2 (t), 52.9 (d), 68.1 (s), 132.3

(d), 137.8 (s); exact mass calcd. for C1 1 H i3 OS2 m/z 230.0800, found m/z 230.0801.

CHO

31 1

( ± ) - 3 -H y d ro x y - 3 -methyM-cyclohexene- 1 -carboxaldehyd 0 . (311). To a solution of 348 mg (1.51 mmol) of dithiane 311 in 9 mL of methanol and 6 mL of THF cooled in an ice/water bath was added 2.0 g (4.5 mmol) of thallium trinitrate trihydrate. The resulting white slurry was stirred in the cooling bath for 10 min, passed quickly through a column of 30 g of activity

II basic alumina (eluted with ethyl acetate), and the eluant was concentrated in vacuo. The residue was chromatographed over 20 g of activity II basic alumina (eluted with ethyl acetate) to yield 103 mg (49%) of aldehyde 311 as a pale yellow oil: IR (neat) 3359, 1682, 1643, 1626 cm'1; 1H NMR

(CDCI3, 300 MHz) 8 1.40 (s, 3H, CH3), 1.65-1.81 (m, 4H, HOCtbCfcb), 1.91 (s. 1H, OH), 2.03-

2.28 (m. 2 H. =CCH2), 6.52 (s, 1 H, C=CH), 9.45 (s, 1 H, CHO); 13C NMR (CDCI3, 75.5 MHz) 8 18.8

(t), 21.2 (t), 28.3 (q), 37.7 (1). 68.5 (s), 140.7 (s), 152.9 (d), 194.5 (d); exact mass calcd. for

C0 H i2 O 2 m/z 140.0838, found m/z 140.0873.

o 172 325

3-lndolylglyoxylyl chloride (325).07 To a solution of 1.63 g (13.9 mmol) of indole

in 30 mL of ether cooled in an ica/water bath was added 2.5 mL (28.5 mmol} of oxalyl chloride.

Within 1-2 min, a yellow precipitate formed, and the reaction mixture stirred in the cooling bath for

25 min. Filtration and washing the product with ether provided 2.66 g (92%) of add chloride 325 as a yellow solid: mp 130°C (dec) (lit 8 7 128°C).

i H

3 2 6

Ethyl 3-lndoylglyoxylate (326).88 A mixture of 2.63 g (12.7 mmol) of acid chloride

325, 1 .6 mL (1 1.5 mmol) of triethylamine, and 25 mL of ethanol was stirred and heated in an oil bath at 95-100°C for 45 min. The deep magenta mixture was cooled to room temperature, sat for

7 h, and the preciptate was collected and washed with 15 mL of ethanol to provide 2.31 g (84%) of ester 326 as a light tan solid: mp 180-183°C (lit 8 8 186°C).

i H

3 2 7

Tryptophol (327). 8 8 To a solution of 2.30 g (10.6 mmol) of ester 326 in 50 mL of THF cooled in an ice/water bath was added 1 .95 mL (51.4 mmol) of lithium aluminum hydride . The resulting gray slurry was heated to reflux (oil bath temperature = 85°C) for 2.5 h, was cooled in an ice water bath and 8 mL of water was added dropwise. The slurry sat overnight, the precipitate was collected and washed with THF. The filtrate was dried (MgS 0 4 ), and concentrated in vacuo to yield 1.52 g (89%) of alcohol 327 as a pale yellow solid: mp 50-54°C (lit 8 8 54-55°C) 173

I H

3 2 8

Tryptophyl Bromide (320).87 To a solution of 1.51 g (9.41 mmol) of tryptophol

(327) in 40 mL of ether cooled in an ice/water bath was added 0.34 mL (3.58 mmol) of

phosphorous tribromide in 20 mL of ether dropwise over a 10 min period. After 15 min, the ice

bath was removed and the reaction stirred at room temperature for 24 h. The supernatant was

decanted from the red-purple residue, the residue was washed with two 50-mL portions of ether,

and the combined organic layers were washed with 50 mL of saturated sodium bicarbonate, 50

mL brine, dried (MgSOj*) and concentrated in vacuo to yield 1.44 g ( 6 8 %) of the bromide 328 as a

light pink solid: mp 94-96°C (lit 100-102°C ) .8 7

i H

3 2 3

Np-Allyltryptamine (323). A mixture of 775 mg (3.37 mmol) of tryptophyl bromide

(328), 15 mL of water, 15 mL of methanol, and 15 mL of allylamine was stirred and heated at 85-

90°C (oil bath temperature) for 1 2 h. The mixture was cooled to room temperature and

concentrated in vacuo. The residue was dissolved in 200 mL of dichloromethane, washed with

40 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo.

The residue was chromatographed over 2 0 g of silica gel (eluted with ethyl acetate-methanol,

20:1 (1L) to ethyl acetate-methanol-triethylamine, 90:9:1) to yield 631 mg (94%) of allylamine 323

as a thick yellow-orange oil: IR (neat) 3413,1643 cm'1; 1H NMR (CDCI 3 , 200 MHz) 5 1.36 (br s, 1 H.

CH2NH), 3.02 (m, 4H, NCH2 CH2), 3.30 (dt, J = 6.0, 1.4 Hz, 2H, NCtteCH=), 5.10 (m, 2H, =CH2),

5.90 (ddt, J= 17, 10, 6 Hz, 1H, CH=CH2). 7.01 (d, J = 2 .2 Hz, 1 H, H i). 7.14 (m, 2H, H4 and H5), 174

7.36 (dt, J ® 7, 1 Hz, 1H, He), 7.66 (dd, J = 8 , 1 Hz. 1H. H3). 8.62 (s. 1 H, =CHNfcD; 13C NMR

(CDCI3 , 75.5 MHz) 5 25.5 (t), 49.1 (t). 52.0 (t), 11.3 (d), 113.0 (s), 116.6 (t), 118.7 (d), 119.1 (d),

121.8 (d). 122.4 (d), 127.3 (s). 135.9 (d). 136.5 (s); exact mass calcd. for C 1 3 H 1 6 N 2 m/z

200.1315, found m/z 200.1318.

329a and 329b

(±)-(1/?*)-3-(2-Allyl-2l3,4,9-tatrahydro-1 W-pyrldo[3,4-b]indol-1-yl)-1- methyl-2-cyclohexen-1-ol (329a and 329b). To a mixture of 273 mg (1.95 mmol) of aldehyde 311 in 20 mL of benzene in an apparatus equipped with a Dean-Stark trap was added

452 mg (2.26 mmol) of tryptamine 323. The resulting mixture was heated to reflux for 18 h, cooled to room temperature, and concentrated in vacuo to yield a dark orange oil. This residue was chromatographed over 60 g of flash silica gel (eluted with ethyl acetate-hexanes, 1:5) to yield

188 mg (30%) of an amorphous yellow solid identified as one diastereomer of alcohol 329: IR

(CH 2 CI2) 3310 c m 1; 1H NMR (CDCI3 . 250 M H z)* 5 1.45 (s, 3H, CH3). 1.61-1.99 (m, 6 H,

CH2 CH2 CH2), 2.35 (br s, 1H, OH), 2.50 (td. J= 11.0, 4.0 Hz. 1H. NCH2 CH2), 2.76 (t, J= 15.1 Hz,

IH , NCH2 CH2 ), 2.85 (m, 1H, NCH 2 CH2), 2.88 (dd, J= 14.0, 8.0 Hz, 1H, N C tf^ H ® ), 3.27 (dm, J =

I I . 2 Hz, 1 H, NCH2 CH2), 3.42 (dd, 14.3, 4.2 Hz, 1 H, NCtbCH®), 4.01 (s, 1 H, NCH), 5.23 (m,

2H, =CH2 ), 5.84 (s, 1H, =CfciCOH). 5.91 (m, 1 H, NCH2 Cfcl=), 7.10 (m. 2H, H6 and H7). 7.29 (d, J =

7.3 Hz, 1 H, H5 or H8), 7.50 (d, J= 7.0 Hz, 1H, H5 or H8), 8.65 (s. 1 H, NH); 13C NMR (CDCI3 , 75.5

MHz) 8 19.7 (t), 21.5 (t), 24.1 (t), 30.1 (q), 38.2 (t), 48.6 (t), 56.8 (t), 66.7 (d). 69.1 (s), 109.2 (s),

110.0 (d). 117.4 (t), 117.9 (d), 118.9 (d), 121.0 (d), 127.2 (s), 133.2 (d), 133.5 (s), 135.6 (d),

136.1 (s), 140.7 (s); exact mass calcd. for C2 ^H2 8 N20 m/z 322.2047, found m/z 322.2046.

Continued elution of the column (ethyl acetate-hexanes, 1:5) yielded 212 mg (34%) of a yellow 175 foam, identified as a mixture of diasteromeric alcohols 329a and 329b. Further elution of the column (ethyl acetate-hexanes, 1 :1 ) yielded 60 mg ( 1 0 %) of a yellow foam identifed as the diastereomer 329b: IR (neat) 3404 cm'1; 1H NMR (CDCI 3 , 300 MHz)* 6 1.39 (s, 3H, CH3), 1.57-

1.83 (m, 6 H, CH2 CH2 CH 2 ). 2 .1 2 (s. 1 H. OH), 2.60 (ddd, J = 11.7, 8.5, 4.6 Hz, 1 H. NCH2 CH2),

2.79 (m. 2 H, NCH2 CH2 ), 2.98 (dd, J= 14.1, 7.4 Hz, 1 H, NCtfcCHs), 3.25 (dt. J= 11.7, 4.4 Hz. 1 H,

NCH2 CH 2 ), 3 43 (ddt, J = 15.7. 5.3, 1.6 Hz, 1 H, NCtbCH=), 4.01 (s, 1 H, NCH), 5.24 (m, 2 H,

=CH2), 5.68 (s, 1H, =CHCOH), 5.92 (m, 1 H, NCH2 Chi=), 7.11 (td, 7.1, 1.2 Hz, 1 H, H6 or H7).

7.17 (td, J= 7.1, 1.4 Hz, 1H, H6 or H7), 7.31 (dt, J = 8.0, 1.0 Hz, 1H, Hs or H0), 7.51 (dd, J - 7.1.

1 .2 Hz, 1 H, H5 or H8). 7.58 (s, 1 H, NH); 13C NMR (CDCI3 , 75.5 MHz) 5 2 0 . 0 (t), 20.7 (t), 24.5 (t),

29.7 (d), 38.2 (t), 47.8 (t), 56.7 (t), 65.4 (d), 6 8 . 8 (s), 109.6 (s), 110.7 (d), 117.3 (t). 118.1 (d),

119.2 (d), 121.3 (d), 127.3 (s), 132.7 (s). 133.3 (d), 135.8 (d), 140.7 (s); exact mass calcd. for

C 2 1 H2 6 N2 O m/z 322.2047, found m/z 322.2044.

3 3 0 a

(±)-(1 ft*)-1-Methyl-3-(2,3,4t9-tetrahydro-1 W-pyrido(3,4-b]indol-1-yl)-2- cyclohexen- 1-ol. To a degassed mixture of 98 mg (0.30 mmol) of allylamine diastereomer

329a in 9 mL of ethanol and 1 mL of water were added 39 mg (0.042 mmol) of tris(triphenylphosphine)rhodium (I) chloride and 29 mg (0.26 mmol) of 1,4-diazabicyclo[2.2.2]octane. The apparatus was evacuated and purged with Ar gas three times, and the mixture was heated (oil bath

T = 90-95 °C) for 2.5 h. The reaction was cooled to room temperature, concentrated in vacuo, and the residue was chromatographed over 15 g of silica gel (eluted with ethyl acetate (50 mL) to ethyl acetate-methanol-triethylamine, 89:10:1) to yield 77 mg (90%) of amino alcohol 330a as a light brown foam: IR (neat) 3286 cm'1; 1H NMR (CDCI 3 , 300 MHz)* 81.39 (s, 3H, CH3), 1.60-2.01 176 (m. 6 H, CH2 CH2 CH2), 2.32 (br s, 2 H, CHNU and OH), 2.72 (dm, J = 15.3 Hz, 1 H, NCH^Hz), 2.83

(m, 1H, NCH2 CH2), 3.05 (ddd. J= 12.2, 9.5, 4.7 Hz, 1H, NCH 2 CH2), 3.37 (ddd, 12.2, 5.1. 3.0 Hz,

1 H, NCH^H^, 4.57 (s, 1 H, NCH), 5.77 (s, 1 H, =CfclCOH), 7.0B (td, J = 7.1, 1.3 Hz, 1 H, H6 o rH 7),

7.13 (td, J = 7.1, 1.5 Hz, 1H, H6 o rH 7), 7.29 (dt, J = 7 .1 , 1 .6 Hz, 1H. H5 or H0), 7.50 (dd, J = 6.9,

1.5 Hz, 1H, Hs or Ha), 8.56 (s, 1 H, =CNH); 13C NMR (CDCI3 , 75.5 MHz) 5 19.6 (t), 22.3 (t), 24.3

(t), 29.9 (q), 37.9 (t), 43.2 (t). 60.4 (d), 6 8 . 8 (s), 109.4 (s), 110.8 (d). 118.0 (d), 119.0 (d), 121.3

(d), 127.4 (s). 132.4 (d), 133.5 (s), 135.8 (s). 140.4 (s); exact mass calcd. for C1 8 H2 2 N 2 O m/z

282.1734, found m/z 282.1709.

3 3 6

3-[9H-pyrido[3,4-J>]indol-1-yl]-2-cyclohexen*1*one (336). From 3-Ethoxy-

2-cyclohexen-1-one: To a solution of 359 mg (1.45 mmol) of p-carboline 283 in 14 mL of dry

THF was added 6 6 mg (1.65 mmol) of potassium hydride in one portion. The resulting yellow mixture was stirred at rt for 20 min, and was cooled in a dry ice/ acetone bath. To the cooled reaction mixture was added 1 . 8 mL (3.06 mmol) of a 1 .6 M solution of f-butyllithium in pentane.

The resulting deep red mixture was stirred in the cooling bath for 45 min, and ethoxy enone 312 in 6 mL of THF was added. The reaction was stirred in the bath for 1.5 h and at rt for 5 h. To the mixture was added 60 mL of 1 0 % hydrochloric acid and followed by stirring at rt for 45 min. The reaction was made basic by addition of solid sodium carbonate, and the mixture was partitioned between 1 0 0 mL of water and 300 mL of dichloromethane. The organic layer was dried (MgS 0 4 ) and concentrated in vacuo to yield a yellow solid. This material was solid bound to 5 g silica gel and chromatographed over 40 g of silica gel (eluted with ethyl acetate-hexanes, 1:1) to yield 106 mg (28%) of enone 336 as a yellow solid, mp 232-233.5°C, IR (CH 2 CI2) 3400, 1656 cm*1; 1H 177

NMR (CDCI3 , 300 MHz) 6 2.27 (qu, J = 6 .2 Hz, 2 H, CH2 CH2 CH2 ), 2.65 (t. 7.2 Hz. 2 H. COCH2),

3.16 (Id, J = 6.0 Hz, 1.4 Hz, 2H, CH2 C=), 6.80 (t, J= 1.4 Hz, 1H, =CH). 7.30 (ddd, J= 8.0, 5.7, 2.3

Hz, 1H, H6), 7.55 (m, 2H, H7 and He), 7.97 (d, J= 5.2 Hz, 1 H, H4), 8.12 (d, J = 7.8 Hz, 1H. H8).

8.54 (d, J= 5.1 Hz, 1 H, H3), 9.66 (s. 1 H. NH); 13C NMR (CDCI3 , 75.5 MHz) 5 2 2 . 6 (I), 27.9 (t), 37.7

(t), 111.8 (d), 115.4 (d), 120.5 (d), 121.2 (s), 121.5 (d). 127.6 (d), 129.0 (d), 130.7 (s). 133.6 (s).

138.4 (d), 140.1 (s), 140.7 (s), 160.0 (s), 200.9 (s); exact mass cackf for C7 i H i 4 N20 m/z

262.1107, found m/z 262.1108.

Anal. Calcd. for C1 7H i4 N2 0: C, 77.84; H. 5.38. Found: C, 77.59; H, 5.44.

From Palladium Coupling: A mixture of 32 mg (0.083 mmol) of stannyl enone 347,

25 mg (0.085 mmol) of iodo-p-carboline 283 and 1 mL of dry DMF was degassed by bubbling Ar gas for 2 min. To the mixture was added 2 mg of (bis(triphenylphosphine)palladium (II) chloride, and the mixture was heated to 120-125°C for 18 h. The reaction was cooled to room temperature, diluted with 1 00 mL of ether, and washed with 40 mL of brine. The organic layer was dried (M gS04), and conentrated in vacuo to yield an yellow oily solid. This material was chromtographed over 5 g silica gel (eluted with ethyl acetate-hexanes, 1:3) to yield 9 mg of a yellow solid, which was recystallized from dichloromethane-hexanes to provide 6 mg (25%) of enone 336.

OH c h 3

3 3 2

(±)-1-Methyl*3-[9H-pyrido[3,4-b]indol-1-yl]-2-cyclohexen-1-ol (332).

From Enone 336: To a solution of 152 mg (0.58 mmol) of enone 336 in 2 0 mL of dry THF cooled in a dry ice/acetone bath was added 1.5 mL (2.1 mmol) of 1.4 M solution of methyllithium in ether. The reaction was stirred in the cooling bath for 2 h and was quenched with 1 mL of 178 methanol. The reaction was diluted to 1 0 0 mL with dichtoromethane, washed with 25 mL of water, dried (NasCOa), and concentrated in vacuo. The residue was chromatographed over 8 g of silica

gel (eluted with ethyl acetate-hexanes, 1:1) to yield 140 mg ( 8 8 %) of alcohol 332 as a yellow solid: IR (CH2 CI2) 3287, 1626 cm’1; 1H NMR (CDCI 3 , 300 MHz) 5 1.45 (s, 3H, CH3), 1.81-2.04 (m,

4H, C(0)CH 2 CH2). 2.53 (dm. J = 18.0 Hz, 1 H, =CHCfcl2 ), 2.84 (dm, J= 18.0 Hz, 1 H, =CCH2 ), 3.21

(br s, 1H, OH). 6.35 (s, 1H, =CH), 7.26 (m, 1H. H6), 7.48 (m, 2H. H7 and H8), 7.84 (d, J= 5.2 Hz,

1H, H4), 8.10 (d, J = 8.0 Hz, 1H, H5), 8.42 (d, J = 5.3 Hz, 1 H, H3), 9.42 (s, 1 H, NH); 13C NMR

(CDCI3 , 75.5 MHz) 6 19.9 (t), 27.6 (t), 29.6 (q), 37.5 (t), 69.1 (s), 111.5 (d), 113.5 (d), 119.9 (d),

121.5 (d). 121.6 (s). 128.3 (d), 129.6 (s), 132.9 (s), 134.0 (d), 138.1 (d), 138.5 (s). 140.4 (s),

144.0 (s); exact mass calcd for m fcC ieH -ig^O 278.1420, found m/z 278.1419. A portion of this material was recrystallized from ethyl acetate-hexanes to provide off-white crystals (mp 180-

182°C).

Anal, calcd. for C1 8 H t8 N2 0 : C, 77.67; H, 6.52. Found: C, 77.57; H. 6.52.

From Dehydrogenation of Alcohol 334: To a mixture of 92 mg (0.326 mmol) of amino alcohol 334 in 4 mL of ethanol and 2 mL of dichloromethane was added 384 mg (1.69 mmol) of 2,3-dichloro-5,6-dicyanobenzoquinone. The resulting dark mixture stirred at rt for 40 h, diluted in 70 mL of dichloromethane, and was washed with 20 mL of saturated aqueous sodium bicarbonate. The organic layer was dried (MgS04), and concentrated in vacuo. The residue was chromatographed over 2 0 g of silica gel (eluted with ethyl acetate to ethyl acetate-methanol, 1 0 :1 ) to yield 17 mg (19%) of alcohol 332 as a pale yellow oil/film. Continued elution provided 40 mg

(44%) of the dihydro-p-carboline 331. u 3 4 9

(±)*(4aW#I5S*,7/T*,7a/f*,14a/T*,1 SaS*)-2,3,4,4a,5,6,7,7a,9,10,11,12,

14a,15-Tatradacahydro-5,7-dlhydroxy-2-(p-mathoxyphanyl)>1 W-azoelno-

[1‘,2':1,5]pyrrolo[2,3i]iaoquinolin-1-ona (349). To a mixture of 50 mg (0.087 mmol) of

epoxy alcohol 216 in 6 mL of dry DMF was added 35 mg (0.23 mmol) of cesium fluoride. The

resulting mixture was stirred at 90°C for 24 h and then cooled to rt. The mixture was diluted with

2 0 mL of dichloromethane. filtered to remove solids, concentrated in vacuo, and further

concentrated under high vacuum. The residue was chromatographed over 500 mg of silica gel

[eluted with ethyl acetate-hexanes, 1 :1 (5 mL) to ethyl acetate (5 mL), to ethyl acetate-methanol,

10:1 (0.5% triethylamine)] to yield 33 mg (90%) of diol 349 as an oily orange solid. The spectral data were identical to that previously reported .2 7

3 5 1

15-Dodecahydro-7-hydroxy-2-(p-methoxyphenyl)-1 f/-azocim-[1 \2':1,5]-

pyrrolo(2,3-i]isoquinolin-1,5-dione (351). A mixture of 21 mg (0.051 mmol) of diol 349,

52 mg (0.19 mmol) of iodoxybenzoic acid (350), and 2 mL of dimethyl sulfoxide was stirred at rt for 160 7 h. The mixture was poured into 20 mL of saturated aqueous sodium bicarbonate, extracted with

50 mL of dichloromethane, and dried (Na 2 C0 3 ). The organic layer was concentrated in vacuo and

chromatographed over 1 g of silica gel (eluted with ethyl acetate-hexanes, 3:1 (15 mL) to ethyl

acetate) to yield 11 mg (53%) of 0-hydroxyketone 351 as a pale yellow film: IR (neat) 3387,1714,

1644 cm'1; 1H NMR (C6D 6,200 MHz)* 8 1.15-2.15 (m, 14H, CH2 manifold) 2.17 (dd, J = 13.8, 4.0

Hz, 1H, NCHCHa), 2.30-2.52 (m. 4H, CH2 manifold), 2.61 (dd, J = 13.8, 2.5 Hz. 1H, NCHCtfc),

3.02 (dd, J= 10.8,4.9 Hz, 1H, CHC=0), 3.27 (m, 1H, CONCH2(1H) and s, 3H, OCH3), 3.75 (d, J =

3.5 Hz, 1H, NCHCH), 3.94 (m, 1H, CONCH2), 4.03-4.23 (m, 2H, NCtJCH= and CJdOH), 5.54 (t, J =

10.2 Hz, 1H, NCHCU=), 5.91 (m, 1H, NCHCH=CU), 6.74 (d, J=9.0 Hz, 2H, ArH), 7.06 (d, J= 9.0

Hz, 2H, ArH); 13C NMR (C6D6, 75.5 MHz) 8 19.5 (t), 28.0 (t), 29.8 (t), 42.5 (t), 45.9 (t), 46.4 (t),

48.6 (t), 49.6 (d), 54.7 (q), 54.9 (d), 67.2 (d), 69.4 (d), 114.4 (d), 127.3 (d), 128.5 (d), 130.7 (d),

134.1 (d), 137.4 (s), 158.4 (s), 173.0 (s), 207.4 (s), The 13C NMR was missing one aliphatic

singlet; exact mass calcd for C24 H3 6 N2C>4 m/z 410.2207, found m/z 410.2211.

(CH-jUC ' O

OMOM

3 5 3

14a,15-Tetradecahydro-5-hydroxy*7-(niethoxymethoxy)-2-(p-methoxyphenyl)-

1 H-azoclno[1,,2':1,5]pyrrolo[2,3-/]isoqulnolion-1-one pivalata (eater) (353). To a solution of 114 mg (0.230 mmol) of alcohol 228 in 3 mL of dichloromethane was added 240 pL

(3.16 mmol) of chloromethyl methyl ether and 600 pL (3.44 mmol) of diisopropylethylamine. The resulting orange mixture was stirred at room temperature for 3 days, quenched by adding 0.5 mL 181 of methanol, and then concentrated in vacuo. The residue was chromatographed over 10 g of silica gel (eluted with ethyl acetate-methanol, 25:1) to yield 98 mg (79%) of tetracycle 353 as an orange foam: IR (CH 2 CI2) 1718, 1638 cm'1; 1H NMR (CDCI 3 , 300 MHz) 8 1.18 (s, 9H, (CH 3 )3),

1.40-2.30 (m, 1 2 H, CH2 manifold), 2.38-2.50 (m, 2 H, CH and CH2 manifold), 2 .8 8 (t, J= 12 Hz,

1H, NCHCH2 ). 3.34 (s, 3H, CH2 OCJtb), 3.48 (br s, 1H, NCUCHO), 3.64 (m. 2H. NCH2), 3.78 (s,

3H, ArOCHg), 4.04 (m, 1H, CHOMOM), 4.19 (br s, 1 H, NCHCH=), 4.55 (d, J= 6 .8 Hz, 1 H,

OCH2 0), 4.64 (d, J = 6 .8 Hz, 1 H, OCH2 0), 5.54 (m, 1H, CHOPiv), 5.51-5.91 (m, 2H, CH=CH),

6.87 (d, J= 9.0 Hz, 2H, ArH), 7.13 (d, J= 9.0 Hz, 2H, ArH); 13C NMR (CDCI 3 , 62.9 MHz) 8 21.8 (t),

27.1 (t), 27.2 (q), 28.1 (t), 28.7 (t), 29.5 (t), 38.9 (s), 43.9 (d), 46.8 (t), 48.4 (t), 50.2 (t), 51.3 (s),

55.3 (d) 55.5 (q), 66.2 (d), 69.7 (d). 70.4 (d), 95.0 (I), 114.2 (d), 126.9 (d), 129.4 (d), 132.6 (d),

137.2 (s), 157.7 (s), 174.7 (s), 178.3 (s), aliphatic quartet not resolved; exact mass calcd for

C3 1 H4 4 N2 0 6 m/z 540.3199, found m/z 540.3195.

OMOM

3 5 4

14a,15-Tetradecahydro-5-hydroxy-7-(methoxymethoxy)-2-(/>-methoxyphenyl)-

1H-azoclno[1',2>:1,5]pyrrolo[2,3-/]isoquinolion-1-one (354). A mixture of 8 8 mg

(0.163 mmol) of pivalate 353 in 4 mL of methanol and 4 mL (6.00 mmol) of 40% aqueous tetrabutylammonium hydroxide was stirred at room temperature for 48 h. The reaction was concentrated to approximately one-half of the original volume, was diluted with 15 mL of water, and was extracted with two 100-mL portions of dichloromethane. The combined organic layers were dried (MgS 0 4 ), concentrated in vacuo, and the residue was chromatographed over 1 0 g of 182 activity II basic alumina (eluted with ethyl acetate-hexanes, 1:1) to yield 49 mg ( 6 6 %) of alcohol

354 as a pale yellow foam: IR (neat) 3450,1644 cm'1; 1H NMR (CDCI 3 , 250 MHz)* 5 1.2-2.4 (m,

1 1 H, CH2 manifold), 2.44 (m, 3H, =CHCh2 <1 H), NCHCfcfc (1H), CH), 2.81 (br s, 1 H, NCHCH2 ).

3.35 (s, 3H, CH2 OCH3 ). 3.43 (ddd, 11,7,1 Hz, 1 H, CONCH2), 3.77 (s, 3H, A1OCH3 ), 3.87 (br

s, 1H, NCHCHO), 4.05 and 4.20 (m, 4H, CfciOH, CHOMOM, NCUCH=, OH), 4.34 (Id, J= 11. 7 Hz.

1H, CONCH2), 4.48 (d, J= 7.1 Hz, 1H, 0CH2 0), 4.69 (d, 7.1 Hz, 1H, 0CH 2 0), 5.46 (t, 10

Hz, 1 H, NCHCH=), 5.95 (dt, 1 0 . 8 Hz, 1 H, NCHCH=CH), 6.85 (d, J = 9.0 Hz, 2 H, ArH), 7.08 (d,

J = 9.0 Hz, 2 H, ArH); 13C NMR (CDCI3 , 62.9 MHz) 823.4 (t), 27.3 (t), 27.8 (t), 29.0 (t), 31.2 (t), 41.6

(d), 45.9 (t), 47.0 (t), 47.6 (s), 49.9 (I), 54.5 (d), 55.4 (q), 55.7 (q). 61.6 (d), 71.7 (d), 72.6 (d), 94.3

(t), 114.3 (d), 127.3 (d), 129.8 (d), 133.5 (d), 137.2 (s). 157.8 (s), 173.6 (s); exact mass calcd for

C2 6 H3 6 N2 0 5 m/z 456.2624, found m/z 456.2586.

OMOM

3 5 5

<±>(4a/?*,5S*,7ff*,7aRM4aflM5aS*)-2,3,4,4a,6,7,7a,9,10,11,12,14a,

15-Dodecahydro-7-(methoxymethoxy)-2-(p-methoxyphenyl)-1 W-azocine-

[1',2‘:1,5]-pyrrolo(2,3-i]isoquinolin-1,5-dione (355). To a solution of 16 mg (11 pL,

0.126 mmol) of oxalyl chloride in 1 mL of dichloromethane cooled in a dry ice-acetone bath was added 55 mg (50 pL, 0.465 mmol) of dimethyl sulfoxide via syringe. The mixture was stirred in the cooling bath for 45 min and 24 mg (0.0053 mmol) of alcohol 354 in 1 mL of dichloromethane.

The mixture was stirred in the cooling bath for 1.5 h, and 73 mg (100 pL, 0.72 mmol) of 183 triethylamine was added dropwise. The resulting mixture stirred in the cooling bath for 1 h, warmed to room temperature, stirred for 3 h, and was concentrated in vacuo. The residue was chromatographed over 8 g of activity II basic alumina (eluted with ethyl acetate-hexanes, 1 :1 , to ethyl acetate) to yield 11 mg (46%) of ketone 355 as a pale orange oil: IR (neat) 1716,1647 cm'1;

1H NMR (CDCI3, 250 MHz) 8 1 .5-3.0 (m, 15H), 3.31 (s, 3H, CH2 OCH3 ). 3.34-3.51 (m, 1 H, NCH2),

3.65-3.83 (m, 2H, NCHCO and NCH2), 3.78 (s, 3H. ArOCH3), 4.30 (br s. 1H. CHO). 4.36 (br s, 1H,

NCHCH=), 4.46 (d, J = 7.1 Hz, 1H, 0C H 2 0 ), 4.60 (d, J = 7.1 Hz, 1H. 0C H 2 0 ), 5.48 (m, 1H,

NCHCH=). 5.99 (m. 1 H, NCHCH=Cfcl), 6 . 8 6 (d, J= 8.9 Hz, 2 H, ArH), 7.04 (d. J= 8.9 Hz, 2 H, ArH); exact mass calcd for C26 H 3 4 N2Ob m/z 454.2469, found m/z 454.2471. Continued elution provide 10 mg (42%) of unreacted 354.

OH H,CI SES

3 6 5

(±)-(2fl*)-2-[[(4afl*,8/?*,8aS*)-2,3,4,4a,5t8-Hexahydro-8-hydroxy-2-{p- methoxyphenyl)-1,5-dloxo-8a(1 H)-isoquinolinoyl]methyl]-3-[[2-(trimethylsilyl)- ethyl]sulfonyl]-9-oxa-3-azabicyclo[6.1.0]nonane (365). To a mixture of 10 mg (0.018 mmol) of alcohol 216 in 2 m l of dimethyl sulfoxide at rt was added 15 mg (0.054 mmol) of iodoxybenzoic acid. The reaction mixture was stirred at rt for 3 h, was poured into 50 mL of dichloromethane, and was washed with 15 mL of water. The organic layer was dried (MgSC> 4 ) and concentrated in vacuo to yield a white solid residue. This material was chromatographed over 500 mg of silica gel (eluted with ethyl acetate-hexanes, 1:2 (10 mL) to ethyl acetate-hexanes, 1:1) to provide 6 mg (60%) of enone 365 as a white solid: mp 211-212°C, IR (neat) 3323, 1672, 1622, cm'1: 1H NMR (CDCI3 , 300 MHz)* 8 0.04 (s, 9H, Si(CH3)3), 1.08 (m, 2 H, C H ^ i), 1.50-2.12, 2.28 184 (m, 6 H, CH2 manifold), 2.41 (m, 1 H. C -0-C C H 2), 2 . 6 8 (m, 1 H, C-0-CCH2), 2.80-2.92 (m, 3H.

NCHCtfc and C O NCtb (1 H)), 3.06-3.16 (m, 4H, CH2 S02, COCH, NHCHCH-O-Cb), 3.50 (m, 3H,

S 0 2 CH2 and NCH), 3.55 (dd, J = 9.3, 4.4 Hz, 1H, NCHCtl), 3.75 (m, 1H, CONCH2), 3.78 (s, 3H,

OCH3), 4.64 (d, J= 10.9 Hz, 1 H, CfclOH), 5.92 (d, J= 11.1 Hz, 1H, OH), 6.06 (dd. J= 10.2, 2.4 Hz,

1H, COCH=) 6.90 (AB quartet, J= 9.0 Hz, 4H, ArH), 7.12 (dd, J= 1 0 .2 , 1.5 Hz, 1 H, COCH=CjJ);

13C NMR (CDCI3 . 75.5 MHz) 5-2.1 (q), 10.3 (t), 18.3 (t), 21.9 (t), 25.1 (t), 29.9 (t), 35.7 (t), 45.6 (d),

48.1 (t), 50.7 (s), 51.3 (t), 52.2 (t), 55.4 (q), 56.9 (d), 57.3 (d), 58.2 (d). 70.1 (d). 114.6 (d), 127.2

(d), 127.3 (s), 134.4 (s), 156.3 (d), 158.7 (s), 175.2 (s), 195.9 (s); exact mass calcd for

C2 2 H4 2 N2 0 7 SSi m/z 590.2484, found m/z 590.2482. Continued elution of the column afforded

2 mg (2 0 %) of recovered starting material.

3 7 8

5-Hexynoic Acid (378).102 To a mixture of 99.3 g (0.993 mol) of chromium trioxide,

900 mL of glacial acetic acid, and 1 0 0 mL of water cooled in an ice/water bath was added 35.2 g

(0.359 mol) of 5-hexyn-1-ol in 800 mL of glacial acetic acid dropwise over 3 h. The ice bath was removed, the reaction mixture was stirred at rt for 46 h, and the excess chromium reagent was destroyed by adding 50 mL of isopropanol. The reaction mixture was concentrated in vacuo to yield a black-green tar which was dissolved in 1 L of 1 0 % aqueous hydrochloric acid. The resulting mixture was extracted with four 400-mL portions of ether, the combined ether layers were washed with three 250-mL portions of cold 3 N aqueous sodium hydroxide, and the combined basic aqueous layers were brought to pH 1 by the addition of concentrated aqueous hydrochloric acid. The acidic aqueous mixture was extracted with four 400-mL portions of ether, the combined ether layers were washed with 2 0 0 mL of brine, dried (MgS 0 4 ), and concentrated in vacuo. The residue was distilled through a Vigreaux column to yield 2 0 .1 g (50%) of acid 378 as a dear, colorless oil: bp 92-94°C at 0 .6 mm Hg (lit102 95-96°C at 0 .6 mm Hg). 185

Cl o

379

5-Hexynoyl Chloride (379). To a flask containing 20.0 g (179 mmol) of 5-hexynoic acid cooled in an ice/water bath was added 2 1 . 2 g (13.0 mL, 180 mmol) of thionyl chloride dropwise over a 15 min period. After 30 min, the reaction was warmed to rt. stirred for 2.5 h, and was heated at 75-80°C for 1 h. The reaction was cooled to room temperature, and the mixture was distilled through a Vigreaux column to yield 16.76 g (72%) of acid chloride 379 as a clear, colorless oil: bp 77-80°C at 16 mm Hg .

3 8 0

3-Hydroxymethyl-3-methyloxetane (380). 1 01 A mixture of 36.68 g (305 mmol) of

2,2-bishydroxymethyM-propanol (381), 35.1 g (36 mL, 297 mmol) of diethyl carbonate, 300 mg

(5.36 mmol) of potassium hydroxide, and 2.1 mL (35.8 mmol) of ethanol was stirred and heated to

120°C over 50 min. The bath temperature was held at 120-140°C as 30 mL of ethanol was removed by distillation through a Vigreaux column. The pressure was then reduced from 1 atm to

50 mm Hg, and the temperature was elevated to 210°C with all the distillate being collected (bp

105-145°C). This crude product was redistilled to yield 13.12 g (43%) of oxetane 380 as a clear, colorless oil: bp 95-99°C at 2 mm Hg (lit 101 80°C at 40 mm Hg).

382 166 (3*Methyl-3-oxatanyl)methyl 5-hexynoate (382). 101 To a solution of 12.2 g

(119 mmol) oxetane 380 and 9.8 g (10.0 mL, 124 mmol) of pyridine in 60 mL of dry dichloromethane cooled in an ice/water bath was added 16.5 g (126 mmol) of 5-hexynoyl chloride. The resulting white slurry was stirred in the cooling bath for 7 h. The mixture was diluted with 200 mL of dichloFomethane, washed with 50 mL of water, and dried (MgS 0 4 ). Concentration in vacuo afforded a pale yellow oil that was chromatographed over 400 g of silica gel (pretreated with 1 % triethylamine in hexanes; eluted with ethyl acetate-hexanes, 1:7) to yield 22.57 g (91 %) of ester 382 as a clear, colorless oil.

3 7 7

1-(4-pentynyl)-4-methyl-2,6,7-trioxabicyclo[2.2.2]octane (377). 1 0 1 To a solution of 6.54 g (33.4 mmol) of ester 382 in 36 mL of dry dichloromethane cooled in a bath at

-20°C was added 1 .02 mL (8.3 mmol) of boron trifluoride etherate via syringe. The resulting mixture was stirred in the cooling bath for 19 h and 2.4 mL (17.6 mmol) of triethylamine was added. The reaction was warmed to rt, diluted with 300 mL of ether, sat for 2 h, and the white precipitate was removed by filtration. The filtrate was concentrated in vacuo, and the residue was chromatographed over 300 g of silica gel (pretreated with 1% triethylamine in hexanes, eluted with ethyl acetate-hexanes, 1:10) to provide 3.65 g (56%) of orthoester 377 as a white solid, mp

47-49.5°C. Continued elution afforded 997 mg (15%) of unreacted 382.

3 8 3

7-(4'Methyl-2(6l7'trloxabicyclo[2,2,2]oct-1-yl)-3-heptyn-1ol (363). To a solution of 1.45 g (7.40 mmol) of orthoester 377 in 20 mL of dry THF cooled in a dry ice/ acetone bath was added 6 .1 mL (8.5 mmol) of a 1 .4 M solution of n-butyllithium in hexanes. The resulting 187 mixture was stirred in the cooling bath for 30 min, 1.4 mL (9.3 mmol) of N,N,N,N-

tetramethylethylenediamine was added, and the reaction was stirred for 10 min. To that mixture

was added 6.3 g (143 mmol) of freshly distilled ethylene oxide in 18 mL of THF, the reaction was

stirred in the cooling bath for 1 h, the bath was removed, and stirring was continued for 6 h. The

mixture was concentrated in vacuo, the residue was partitioned between 30 mL of water and 100

mL of ether, and the aqueous layer was extracted wtih 100 mL of ether. The combined organic

layers were washed with 30 mL of brine, dried (MgS 0 4 ), and concentrated in vacuo. The residue

was chromatographed over 70 g of silica gel (pretreated with triethylamine and eluted with ethyl

acetate-hexanes, 1 :1 ) to yield 1.49 g (84%) of alcohol 383 as a clear colorless oil: IR (neat) 3426

cm'1; 1H-NMR (CDCI3, 250 MHz) 8 0.78 (s. 3H, CH3 ), 1.64 (m, 2H, CH2 CH2 CH2 ), 1.75 (m, 2H,

CH2 CO3 ), 2 .0 2 (br s, 1 H, OH), 2.17 (tt, J= 7,2.4 Hz, 2 H, CH2 CCCH2 ). 2.38 (It, J = 6 .2 , 2.4 Hz, 2 H,

HOCH2 Cfci2 ). 3.64 (br s, 2 H, HOCtfc), 3.87 (s, 6 H, C{CH2)3); 13 C-NMR (CDCI3 , 62.9 MHz) 814.5

(q), 18.5 (t). 22.8 (t), 23.2 (t), 30.2 (s), 35.5 (t), 61.3 (t), 72.5 (t), 76.8 (s), 82.2 (s), 108.9 (s); exact

mass calcd. for C1 3 H2 1 O4 (M+1) m/z 241.1440, found rrVz 241.1444.

3 8 4

7-(4-Methyl-2t6,7-trioxabicyclo[2,2,2]oct-1-yl)-3-heptyn-1ol p-toluenesulfonate (384). To a solution of 1.39 g (5.79 mmol) of alcohol 3 83 in 70 mL of dichloromethane cooled in an ice/water bath was added 1.37 g (7.19 mmol) of p-toluenesulfonyl chloride, 48 mg (0.39 mmol) of 4-A/,W-dimethylaminopyridine, and 1.5 g ( 2 . 0 mL, 14.3 mmol) of triethylamine. The resulting mixture was stirred for 20 min in the cooling bath and at room temperature for 13 h. The reaction was diluted with 70 mL of dichloromethane, washed with 30 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo. The residue was chromatographed over 70 g of activity grade II basic alumina (eluted with ethyl acetate-hexanes, 1 :8 (800 mL) to ethyl acetate-hexanes, 1:5) to yield 1.77 g (72%) of tosylate 188

384 as a pale yellow oil: IR (neat) 2932, 2877, 1399 cm'1; 1H NMR (CDCI 3 , 300 MHz) 5 0.76 (s,

3H, (CH2 )3 CCfcb), 1 -56 (m, 2 H, CH2 CU2 CH2 ), 1.67 (m, 2 H, CH2 C 03), 2.06 (tt, J= 6.9, 2.4 Hz, 2 H,

CCCH2), 2.42 (s, 3H, ArCH3), 2.45 (tt, 7.3, 2.4 Hz, 2H, T s O C H ^ tb ), 3.85 (s,6 H, (OCH2)3),

4.00 (t, J= 7.3 Hz, 2H, TsOCH^, 7.31 (dd, J= 8.5, 0.6 Hz, 2H. ArH), 7.76 (d, J= 8.5 Hz, 2H, ArH);

13C NMR (CDCI3, 62.9 MHz) 8 14.4 (q), 18.3 (t), 19.5 (t), 21.5 (q), 22.5 (t), 30.1 (s), 35,5 (t), 68.2

(t), 72.4 (t), 74.0 (s), 82.3 (s), 108.7 (s), 127.8 (d), 129.8 (d). 132.9 (s), 144.7 (s); exact mass calod. for C2 oH2 e06S m/z 394.1451, found m/z 394.1444.

3 7 6

1-(7-bromo-4-heptynyl)-4-methyl-2,6,7-trioxabicyclo[2.2.2]octane (376).

To a flame-dried flask containing 302 mg (3.48 mmol) of lithium bromide was added 343 mg (0.871 mmol) of tosylate 384 in 2 0 mL of acetone. The resulting mixture was heated at 65°C for 7 h, cooled to room temperature, concentrated in vacuo, and the residue was partitioned between

1 0 0 mL of dichloromethane and 2 0 mL of saturated aqueous sodium bicarbonate. The organic layer was dried (MgS 0 4 ), concentrated in vacuo, and chromatographed over 25 g of activity grade

II basic alumina (eluted with ethyl acetate-hexanes, 1:15) to yield 194 mg (74%) of bromide 376 as a clear colorless oil: IR (neat) 2936,1438,1351 cm'1; 1H NMR (CDCI3, 250 MHz) 5 0.78 (s, 3H,

CH3), 1.56-1.78 (m, 4H, CH^ifcCfckCOs), 2.15 (tt. J = 6.9, 2.4 Hz. 2H, CCCH2), 2.67 (tt, J = 7.4,

2.3 Hz, 2H, BrCH2 CU2 ), 3.38 (t, J = 7.5 Hz, 2H. BrCH2), 3.87 (s, 6 H, (OCH2)3); 13C NMR (CDCI3,

62.9 MHz) 8 14.5 (q), 18.5 (t), 2 2 .6 (1), 23.3 (t), 30.16 (s), 30.19 (t), 35.6 (t), 72.5 (t), 77.1 (s), 82.1

(s), 108.7 (s); exact mass calcd. fo rC i3 H i9 0 379Br nVz 302.0518, found m/z 302.0508.

H OTHP

387 189 (±)-8-[(Tatrahydro-2H-pyran-2-yl)oxy]-3-octyn-1-ol (387). To a solution of

5.10 g (28.0 mmol) of alkyne 186 in 180 mL of THF cooled in a dry ice-acetone bath was added

2 0 mL (30 mmol) of a 1.5 M solution of n-butyllithium in hexanes via syringe. The resulting pale yellow mixture was stirred in the cooling bath for 50 min, 6.0 mL (40 mmol) of N,N,N,N- tetramethylethylenediamine was added, the mixture was stirred for 10 min, and 15 g (340 mmol) of ethylene oxide was distilled into the reaction. The reaction mixture was stirred in the cooling bath for 1 h and at rt for 24 h. The reaction mixture was concentrated in vacuo, the residue was dissolved in 750 mL of dichloromethane, was washed with 1 0 0 mL of 1 0 % aqueous hydrochloric acid and 100 mL of saturated aqueous sodium bicarbonate, dried (MgSC> 4 ), and concentrated in vacuo. The residue was chromatographed over 1 0 0 g of flash silica gel (eluted with ethyl acetate- hexanes. 1:4) to yield 5.38 g (85%) of alcohol 387 as a clear colorless oil: IR (neat) 3441 cm'1; 1H

NMR (CDCI3 , 250 MHz) 5 1.45-1.90 (m, 1 0 H, CH2 manifold), 1.98 (t, J= 6 Hz, 1 H, OH), 2.19 ( tt, J

= 7, 2.4 Hz, 2 H, HOCH2 CH2 CCCtd2 ). 2 41 (tt, J= 6 , 2.4 Hz, 2 H, HOCH^tfc), 3.40 (dt, J= 1 0 , 6

Hz, 1 H, C H^TH P), 3.48 (m, 1 H. OfcOCHOCtk), 3.65 (q. J = 6 Hz, 2H. CtfcOH), 3.72 (dt. J= 10,

6 Hz, 1 H, CH2 OTHP), 3.84 (m, 1 H, CH2 OCHOCH2 ). 4.57 (m, 1 H, OCHO); 13C NMR (CDCI3 , 62.9

MHz) 5 18.5 (t), 19.5 (t), 23.1 (t), 25.4 (t), 25.7 (t), 28.9 (t), 30.7 (t), 61.3 (t), 62.3 (t), 67.0 (t), 76.7

(s), 62.2 (s), 98.9 (d); exact mass calcd. for C-| 3 H2 2 0 3 m/z226.1569, m/z found 226.1463.

3 8 8

(±}-8-[(Tetrahydro-2M-pyran-2-yl)oxy]-3-octyn-1-ol p-toluenesulfonate

(388). To a solution of 1.47 g (6.52 mmol) of alcohol 387, 39 mg (0.32 mmol) of A-N,N- dimethylaminopyridine, and 20 mL of dry dichloFomethane cooled to 0°C was added 1.52 g (7.94 mmol) of p-toluenesulfonyl chloride and 1 .8 mL (13.0 mmol) of triethylamine. The resulting mixture was stirred in the cooling bath for 30 min, at room temperature for 2 h, and was diluted with

80 mL of dichloFomethane. The organic layer was washed wtih 20 mL of water, 20 mL of 10% 190 aqueous hydrochloric acid, 20 mL of saturated aqueous sodium bicarbonate, and dried (MgS 0 4 ).

Concentration in vacuo yielded a yellow oil which was chromatographed over 30 g of silica gel

[(eluted with ethyl acetate-hexanes, 1:10 (220 mL) to ethyl acetate-hexanes, 1:6 (until

completion)] to yield 2.32 g (93%) of tosylate 386 as a clear colorless oil: IR (neat) 2942, 2B68,

1598,1453 cm*1; 1H NMR (CDCI3, 250 MHz) 5 1.45-1.90 (m, 10H, CH 2 manifold), 2.12 ( tt, J = 7,

2.5 Hz, 2 H. CCCba, 2.45 (s, 3H, CH3), 2.50 ( tt, J= 7.5, 2.5 Hz. 2 H. TsOCH2 Ctl2 ), 3.38 (dt, J = 1 0 ,

6 Hz, 1H, CH2 OTHP), 3.49 (m, 1H, OCHOCtb), 3.73 (dt, J = 1 0 , 6 Hz, 1H, CH2 OTHP), 3.83 (m,

1H, OCHOCtte), 4.05 (t, J = 7 Hz. 2 H. TsOCH2), 4.56 (m, 1 H, OCHO), 7.33 (dd, J = 8.5, 0.5 Hz,

2H, ArH), 7.82 (dt. J = 8.5, 2 Hz, 2H, ArH); 13C NMR (CDCI3 , 62.9 MHz) 5 18.4 (t), 19.6 (t), 19.7 (t),

21.6 (t), 25.4 (t), 25.5 (t), 28.8 (t), 30.7 (t). 62.3 (t), 66.9 (t). 68.2 (1). 74.1 (s). 82.5 (s), 98.8 (d),

127.9 (d), 129.8 (d), 132.9 (s), 144.8 (s); exact mass calcd. for C 2 oH2 80sS m/z 380.1658, found

m/z 380.1651.

3 8 5

(±)-8-Bromo-(tetrahydro-2M-pyran-2-yl)oxy 5-octynyl acetal (365). To a mixture of 2.27 g (5.94 mmol) of tosylate 388 in 40 mL of acetone was added 2.32 g (26.7 mmol) of lithium bromide. The resulting mixture was heated at 55-60°C for 8 h, cooled to room temperature and concentrated in vacuo. The residue was partitioned between 250 mL of dichloromethane and 75 mL of water. The organic layer was washed with 50 mL of saturated aqueous sodium bicarbonate, dried (MgS 0 4 ), and concentrated in vacuo to yield a pale yellow oil.

This material was chromatographed over 30 g of silica gel (eluted with ethyl acetate-hexanes,

1:10) to yield 1.56 g (91%) of bromide 385 as a clear, colorless oil: IR (neat) 2940, 2867, 1136 cm*1; *H NMR (CDCI3 . 250 MHz) 5 1.40-1.90 (m, 10H, CH 2 manifold), 2.20 (tt, J = 7.0, 2.3 Hz, 2H,

CCChb, 2.70 ( tt. 7.5, 2.3 Hz. 2 H, B rC H ^ tk), 3.41 (t, J = 7.0 Hz, 2 H, CH2 Br), 3.44 (dt, J = 1 0 ,

6 Hz, 1H, CH2 OTHP) 3.50 (m, 1H, OCHOCfcb), 3.76 (dt, J = 10, 6 Hz, 1H, CH2 OTHP), 3.86 (m, 191

1H, OCHOCtb), 4.58

25.4 (t), 25.5 (t). 28.7 (t). 30.1 (I), 30.6 (t), 62.1 (t), 6 6 . 8 (t), 77.0 (s), 82.2 (s). 98.6 (d); exact mass

calcd. for C i3 H2 i 7 9 Br0 2 m/z 288.0725, found m/z 288.0721.

,OTHP

3 8 6

(±)-8-lodo-(tetrahydro-2W-pyran-2-yl)oxy 5-octynyl acatal (386). To a

solution of 2.10 g (5.50 mmol) of tosylate 388 in 20 mL of acetone was added 3.40 g (22.7 mmol)

of lithium iodide. The resulting mixture was stirred at room temperature for 15 h, was heated at 55-

60°C for 1 h, and was cooled to room temperature. The mixture was concentrated in vacuo,

diluted in 200 mL of ether, washed with 40 mL of water. 40 mL of saturated sodium thiosulfate, 40

mL of saturated aqueous sodium bicarbonate, and 40 mL of brine. The organic layer was dried

(MgS0 4 ), concentrated in vacuo, and was chromatographed over 60 g of silica gel (eluted with

ethyl acetate-hexanes, 1:15) to yield 1.66 g (90%) of iodide 386 as a pale yellow oil: IR (neat)

2939, 2867, 1249, 1 2 0 0 cm'l; NMR (CDCI3, 300 MHz) 6 1.45-1.85 (m, 1 0 H, CH2 manifold),

2.16 (tt, J = 7.5, 2.5 Hz, 2 H, CCCH2 ), 2.72 (tt, J = 7.5, 2.5 Hz, 2H, ICH2 Chl2 ), 3.18 (t, J = 7.4 Hz,

2 H, CH2 I), 3.38 (dt, J = 10, 6 Hz, 1 H, CH2OTHP), 3.48 (m, 1 H, OCHOCtb), 3.73 (dt. 1 0 , 6 Hz,

1 H, CH2 OTHP). 3.83 (m, 1H, OCHOCtb). 4.56 (m, 1 H, OCHO); 13C NMR (CDCI3 , 62.9 MHz) 5

2.5 (t), 18.5 (t), 19.5 (t), 24.0 (t), 25.4 (t), 25.5 (t), 28.8 (t). 30.6 (t), 62.2 (t), 66.9 (t), 78.9 (s), 82.1

(s), 98.7 (d); exact mass calcd. for C 1 3 H2 1 IO2 m/z 336.0586, found m/z 336.0584.

,OTHP

3 8 9

(±)-1-[8-[(Tetrahydro-2tf-pyran-2-yl]-3-octynyl]-2-cyclohexen-1-ol (389).

To a solution fo 1 .6 mL (1 .8 mmol) of 1 .1 M f-butyllithium in pentane in 4 mL of ether cooled in a 192 dry ice-acetone bath was added 235 mg (0.813 mmol) of bromide 385 in 2 mL of ether. The

resulting clear mixture was stirred in the cooling bath for 25 min and 108mg (110 pL, 1.15 mmol) of

neat 2 -cyck)hexen -1 -one was added via syringe. The pale yellow mixture was stirred in the

cooling bath for 1 h and at room temperature for 30 min. The reaction was quenched with 15 mL

of water, extracted with 75 mL of dichloromethane, dried (MgS 0 4 ). and concentrated in vacuo.

The residue was chromatographed over 14 g of silica gel (eluted with ethyl acetate-hexanes,

1:10) to yield 125 mg (50%) of allyiic alcohol 389 as a clear, colorless oil: IR (neat) 3444,1719 cm'

1; 1H NMR (CDCI3 , 250 MHz) S 1.45-1.90 (m, 17H, OH and CH 2 manifold), 2 .0 0 (m, 2 H, CH2 C=C),

2.18 (tt, J = 7, 2.5 Hz, 2 H, CCCH2), 2.27 (tt, J = 7.5, 2.5 Hz, 2H, CH2 CC). 3.40 (dt, J = 1 0 , 6 Hz,

1 H, CH2 OTHP), 3.50 (m, 1H, OCHOCH2 ). 3.75 (dt, J = 1 0 , 6 Hz, 1H, CH2 OTHP), 3.85 (m. 1 H,

OCHOCH2 ). 4.58 (m, 1 H, OCHO), 5.63 (d, 1 0 Hz, 1 H, =CHC), 5.79 ( dt, J = 1 0 , 3 Hz. 1 H,

CH2 CH=): 13C NMR (CDCI3 , 75.5 MHz) 5 13.2 (t), 18.6 (t), 19.0 (t), 19.6 (t), 25.2 (t), 25.5 (t), 25.8

(t), 28.9 (t), 30.7 (t), 35.4 (t), 41.1 (t), 62.3 (t), 67.0 (t), 69.4 (s), 77.2 (s), 80.5 (s), 98.8 (d), 130.0

(d), 132.0 (d); exact mass calcd. for C1 9H3 0 O3 m/z 306.2195, found m/z 306.2182.

OTHP

3 9 2

dodecahydro-7-hydroxy-2-(p-methoxyphenyl)-7-[8-[(tetrahydro-2H-pyran-2-yl]-

3-octynyl]-1 W-azocino[1,,2':1,5]-pyrrolo[2,3-/]lsoquinolin-1-one (392). To a mixture of 180 pL (0.198 mmol) of f-butyllithium in 2 mL of ether cooled in a dry ice-acetone bath was added 29 mg ( 0 .1 0 0 mmol) of bromide 385 in 1.5 mL of ether. The resulting mixture was stirred in the cooling bath for 15 min, and 16 mg (0.041 mmol) of enone 152 in 2 mL of ether and

1 .5 mL of THF was added dropwise over 10 min. The resulting mixture was stirred in the cooling 193 bath for 45 min and quenched with 2 mL of methanol. The solvent was removed in vacuo, the

residue was filtered through a Kimwipe plug, and the filtrate was concentrated in vacuo. The

residue was chromatographed over 9 g of activity II basic alumina (eluted with ethyl acetate-

hexanes, 1 :2 ( 2 0 0 mL) to ethyl acetate-hexanes, 2 :1) to yield 1 0 mg (41%) of alcohol 392 as a

pale yellow oil/film: IR (neat) 3441,1643 cm*1; 1H NMR (CDCI 3 , 300 MHz) 5 1.5-2.4 (m, 27H), 2.42

(tt, 6 .2 , 2.5 Hz, 1 H, CH2 C»C), 2.79 (br s, 1H), 3.00 (br s, 1H), 3.38 (dt, J = 9.7, 6.2 Hz, 1 H,

OCH2), 3.45-3.95 (m, 6 H, CJd^CHOChfc (3H), NCHC, NCH2), 3.78 (s, 3H, OCH3 ), 4.28 (m, brs,

1H, NCtiCH=), 4.57 (m, 1H. OCHO). 5.56 (m, 1H, NCHCH=), 5.90 (m, 1H, NCHCH=Ctl), 5.95 (dd,

J = 9.6, 2.5 Hz, 1H, CCH=CH), 6.09 (dd, J = 9.6, 2.9 Hz, 1H, CCH=Ctt), 6.87 (d, J = 8.9 Hz, 2H,

ArH), 7.04 (d. 8.9 Hz, 2 H, ArH).

3 9 3

dodecahydro-7-hydroxy-2-(p-methoxyph«nyl)-7-methyt-1/y-azoclno[1>,2,:1,5]-

pyrro lo [2,3-/)i8o q u in o lin -1-o n e (393). To a flask containing freshly-dried cerium

trichloride under an Ar gas balloon was added 2 mL of THF. The resulting milky slurry was stirred at

room temperature for 2h, was cooled in a dry ice-acetone bath for 30 min, and 195 pL (0.25 mmol)

of 1 .3 M solution of methyl lithium in ether was added. The resulting near-black mixture was stirred

in the cooling bath for 20 min, and to the dark mixture was added 24 mg (0.061 mmol) of enone

152 in 5 mL of dry THF drop wise over 10 min. The reaction mixture was stirred in the cooling bath for 40 min, the bath was removed, and the flask was opened to atmospheric moisture. The

heterogenous mixture was wanned to rt, passed quickly through a column of 3 g of activity II basic 194 alumina (eluted with 30:1 ethyl acetate:methanol), and the filtrate was concentrated in vacuo. The residue was chromatographed over 7 g of activity II basic alumina (eluted with 1:1 hexanes:ethyl acetate (120 mL) to 1:2 hexanes:ethyl acetate (60 mL) to ethyl acetate) to yield 6 mg (25%) of unreacted enone 152 and 16 mg (64%) of alcohol 393: IR (free base, neat) 3384, 1644 cm*1;

1H NMR (300 MHz, CDCI3) 5 1.40 (s 3H. CCH3 ), 1.55-2.00 (m, 5H, CONCH2 CH2 ) and CH2 manifold), 2.07-2.35 (m, 6 H, CH=CHCfcl2 , CHNCfcfc (1 H), NCHCtfc. and 1 H of CH2 manifold). 2 .8 6

(m, 2 H, CChCH= and CHNChfc). 3 12 (br s, 1 H, OH), 3.51 (dd, 12.7, 4.8 Hz, 1H, CONCH2),

3.73-3.83 (m, 2 H, CONCH2 and NCfciCO, 3.78 (s. 3H, OCH3 ), 4.33 (br s, 1 H, NCHCH=), 5.62

(ddd, v/= 1 0 .6 , 8 .2 , 1.5 Hz, 1 H, N C H C tH , 5.90 (m, 1 H, NCHCH=CH), 5.92 (dd, J = 9.5, 2 . 0 Hz.

1 H, COCH=), 5.97 (dd, J = 9.5, 2.4 Hz, 1 H, COCH=Chl)„ 6.87 (d, J = 9.0 Hz, 2H, ArH), 7.06 (d, J =

9.0 Hz. 2 H, ArH); 1H NMR (300 MHz, C6 De) 5 1.20-1.37 (m, 5H. CH 2 manifold), 1.43 (s, 3H, CH3),

1.45-2.17 (m, 6 H, CH2 manifold), 2.71 (br s, 1H, OH), 2.77 (m, 1H, CHNCfck), 2.97 (dd, J = 11.7,

4.5 Hz, 1H, CONCH2 ). 3.24 (s. 4H, OCH3 and CCHCH=), 3.31 (td, J = 1 2 .1 , 4.8 Hz, 1 H, CONCH2),

4.09 (s, 1H, NCHCO), 4.25 (m, 1 H, NCHCH=), 5.59 (dd. J = 9.5, 2.7, 1 H, COCH=Cb), 5.77 (m, 2 H,

NCHCM=CU), 6 .0 1 (dd, J = 9.5, 2.7 Hz, 1 H, COCH=CJd), 6.70 (d. J = 9.0 Hz, 2 H, ArH), 6.96 (d, J =

9.0 Hz, 2H, ArH); 13C NMR (HCI salt, 62.9 MHz, CDCI 3 ) 523.6 (t), 25.7 (t), 27.1 (t), 29.7 (t), 31.5

(q). 34.1 (d), 44.2 (t). 48.4 (t), 51.7 (t), 53.3 (s), 55.5 (q), 57.9 (d), 66.7 (s), 73.8 (d), 114.6 (d),

124.1 (d), 127.1 (d), 129.8 (d), 135.8 (s), 137.1 (d), 141.5 (d), 158.5 (s), 171.2 (s); exact mass calcd for C2 sH3 2 N2 0 3 m/z 408.2413, found m/z 408.2378.

414 195 (±H 4aft*t7a/?M4a/?M5aS*)-5-(3-biJUnyl)-2,3,4,4at5,6,7,7a,9,10,

11,12,14a,15-dodecahydro-2-(p-methoxypheny l)-1 H-azocino[1 \2'-1 ,5]-

pyrrolo[2,3-J]isoquinolin-1,7(7aH)-dione (414). To a solution of 20 mg (0.051 mmol) of

enone 152 in 4 mL of THF cooled in a dry ice-acetone bath was added 165 pL (0.125 mmol) of a

0.76 M solution of 3-butenylmagnesium bromide via syringe. The resulting pale yellow solution

was stirred in the cooling bath for 2 0 min and was quenched at low temperature by the addition of

1 mL of methanol. The reaction mixture was concentrated in vacuo to yield a yellow oily residue.

This material was chromatographed over 5 g of activity II basic alumina (eluted with ethyl acetate-

hexanes, 1:2 (120 mL) to ethyl acetate (50 mL)) to provide impure ketone 414 and 10 mg (50%)

of recovered 152. Impure ketone 414 was chromatographed over 4 g of silica gel (eluted with

ethyl acetate-hexanes, 1:2) to afford 7 mg (31%) of ketone 414 as a pale yellow oil: IR (neat)

1714,1644 cm'1; 1H NMR (CDCI3, 300 MHz) 8 1.2-2.7 (m, 20H), 3.65 (m, 2H, NCH2), 3.79 (s, 3H,

OCH3 ), 3.91 (s, 1H, NCHCO), 4.29 (br s, 1H, NCHCH=), 5.01 (dd, J= 9.9, 1 .6 Hz, 1H, =CH2), 5.07

(dq, J= 15.4, 1.6 Hz, 1 H, =CH2), 5.40 (I. J= 1 0 .0 Hz, 1 H, NCHCH=). 5.77 (ddt, J= 17.0,10.0, 6.4

Hz, 1H, Cti=CH2), 5.96 (m, 1H, NCHCH=CJd), 6 .8 8 (d, J= 9.0 Hz, 2H, ArH), 7.08 (d, J = 9.0 Hz,

2H, ArH); exact mass calcd. for C2 eH 3 6 N2 0 3 m/z 448.2728, found m i 448.2713.

4 1 5

15-dodecahydro-7-hydroxy-2-(p-methoxyphenyl)-1 H-azocinoH'^'iI.SJpyrrolo-

[2,3-/]isoqulnolln-1-one (415). To a solution of 20 mg (0.051 mmol) of enone 152 in 6 mL of dry THF cooled in a dry ice acetone bath was added 60 pL (0.120 mmol) of 2.0 M allyl magnesium chloride in THF. The resulting mixture stirred in the cooling bath for 25 min, and the reaction was quenched at low temperature by adding 100 pL of methanol. The reaction was concentrated in vacuo and the residue was chromatographed over 1 g of silica gel (eluted with ethyl acetate (10 mL) to 7:1 ethyl acetate: methanol) to yield 20 mg (90%) of alcohol 415 as an off- white solid: mp (free base) 83-8B°C; mp (hydrochloride salt) 198-201 °C; data for hydrochloride salt: IR (neat, free base) 3418,1644 c m 1; 1H NMR (HCI salt, 300 MHz, C0CI3) 8 1.35 (m. 1H, CH2 manifold), 1.85-2.15 (m, 3H, CH 2 manifold), 2.17-2.37 (m, 6 H, CH2 manifold), 2.45 (m, 1 H,

NCHCfcb), 2.72 ( brs, 1 H, NCH2), 3.12 (dd, J = 1 2 .2 ,10.4 Hz. 1 H, NCHCfcfc), 3.47 (dd, J = 1 2 .6 ,

5.6 Hz, 1 H, CONCH2 ). 3 54 (br s, 1 H, NCH2), 3.71 (td, J = 1 2 .6 , 4.7 Hz, 1 H, CONCH2), 3.78 (s,

3H, OCH3 ). 3.97 (br s. 1 H, OH). 4.40 (br s, 1 H, NCHCO), 4.81 (br s. 1 H. NCHCH=), 5.13 (m. 2H,

=CH2), 5.67 (t, J = 9.8 Hz, 1 H, NCHCU=). 5.78 (m. 1 H, CH=CH2 ), 5.95 (m, 2 H, CCH=CH), 6.16 (q,

J = 8.1 Hz, 1H, CHCH=CH), 6 . 8 8 (d, J = 9.0 Hz, 2H. ArH), 7.01 (d, J = 9.0 Hz, 2H, ArH), OM is missing from 1H NMR; ’ H NMR (free base, 300 MHz, CDCI 3 ) 8 1.32 (m, 1H, CH2 manifold), 1.57-

1.98 (m. 4H, CONCH 2 CJd2 ) and CH2 manifold), 2.03-2.20 (m, 2 H, =CHCH2), 2.20-2.41 (m, 5H,

NCHCH2 . CH2 =CHCtl2 (1H), CHNCHo (1 H), one of CH 2 manifold), 2.48 (dd, J = 14.0, 6.4 Hz, 1 H,

CH=CHCh2 ), 2.85 (br s, 1 H, =CHCtiCH2), 3.02 (m. 1 H, CHNCH2 ), 3-10 (br s, 1 H, OH), 3.49 (dd, J

= 11.3, 5.7 Hz, 1 H, CONCH2 ). 3.76 (m. 1 H, CONCH2 ), 3.77 (s, 3H, CH3), 3.93 (s, 1 H, NCHCO),

4.31 (m, 1H, NCtlCH=), 5.10 (m, 2H, =CH2), 5.57 (ddd, J = 10.7, 7.4, 1.2 Hz, 1H, NCHCd=), 5.86

(m, 2H, CHCH=Cfcl and CH=CH2), 5.94 (dd, J = 9.6, 2.4 Hz. 1H, COCH=CH), 6.07 (dd, J = 9.6, 2.9

Hz, 1H, COCH=CfcD, 6 . 8 6 (d, J = 9.0 Hz, 2 H, ArH), 7.03 (d, J = 9.0 Hz, 2H, ArH); 13C NMR

(hydrochloride salt, 75.5 MHz, CDCI 3 ) 8 23.8 (t), 25.6 (t), 26.3 (t), 27.1 (t), 34.6 (d), 42.7 (t), 47.1

(t), 48.3 (t), 51.5 (t), 53.3 (s), 55.4 (t), 57.9 (d), 6 8 . 6 (s), 72.1 (d), 114.5 (d), 119.9 (t), 124.5 (d),

127.1 (d), 130.4 (d), 132.0 (d). 135.6 (d), 135.9 (s), 140.5, 159.4 (s), 170.7 (s); 13C NMR (free base, 75.5 MHz. CDCI3 ) 8 24.3 (t), 26.4 (t), 27.1 (t), 30.0 (t), 36.3 (d), 44.9 (t), 47.0 (t), 48.1 (t),

51.1 (t), 55.0 (s), 55.4 (q), 58.6 (d), 68.4 (s), 73.9 (d), 114.3 (d), 118.4 (t), 127.1 (d), 128.0 (d),

131.4(d), 133.5 (d), 136.6 (s), 136.8 (d), 158.0 (s). 174.0 (s), 13C NMR is missing one sp2 doublet; exact mass calcd for C 2 7 H 3 4 N2 O3 (free base) nVz434.2571, found m/z 434.2566. 197

OAc

418 419

(±}-(2fl*)-2-[[4afl*,SSe,8*fl*)-2,3,4,4aJ5,6-Hexahydro-5-hydroxy-1-oxo-

8a(1 W)-isoquinolyl]methyl]-1,2,5,6,7,8-hexahydro-1[[2-(triinethylsllyl)ethyl]8ul-

fonyljazocine acatata (astar) (418) and (±j>(2R*)-2-[[4a/7*,5S*,8a/7*)-

2,3,4,4a,5,6-Hexahydro-5-hydroxy-2-(p-methoxy-fn-nltrophenyl)-1-oxo-8a(1 H)~

iaoquinolyl]mathyl]-1,2,5,6,7,8-haxahydro-1[[2-(trimathylsilyl)athyl]sul1onyl]-

azocina acatata (aatar) (419). To a solution of 499 mg (0.84 mmol) of /V-aryl lactam 214 in 6

mL of acetonitrile cooled in an ice/water bath was added 1.30 g (2.37 mmol) of eerie ammonium

nitrate in 6 mL of water over a 1 0 min period. The reaction was stirred in the cooling bath for 1 h,

and was then diluted with 2 0 0 mL of dichloromethane. The mixture was washed with 50 mL of

saturated aqueous sodium bisulfite, the layers were separated, and the aqueous layer was diluted

with 50 mL of water. The aqueous layer was extracted with 250 mL of chloroform, and the

combined organic layers were washed with 75 mL of saturated sodium bicarbonate, and dried

(MgS0 4 ), Concentration in vacuo yielded an orange-brown residue that was chromatographed over 15 g of silica gel (eluted with ethyl acetate-hexanes, 1 :3) to yield 34 mg (7%) of /V-aryl lactam

419 as a yellow oil: IR (neat) 1731, 1651 cm'1; 1H NMR (CDCI3, 300 MHz) 5 0.03 (s, 9H, SiMe3),

1.04 (m, 2H, CH2 Si). 1.5-2.08 (m, 8 H, CH2 manifold). 2.08 (s, 3H, COCH3), 2.10-2.35 (m, 2H,

CtLCHOAcCfcia), 2.42 (m, 1 H, CUCHOAcCfcb), 2.70 (m. 1 H), 2.72 (dd, J = 14.4, 8.4 Hz, 1 H,

NCHCtb). 2.97 (td, J = 1 2 .8 , 5.4 Hz, 1 H, S 0 2 CH2), 3.03 (td, J = 12.7, 5.4 Hz, 1 H, S 0 2 CH2), 3.39

(m, 1 H, SC^NCH^, 3.53 (m. 1 H. SC^NCH^, 3.71 (m, 1 H, CONCH2), 3.84 (m, 1H. CONCH2), 3.96

(s, 3H, OCH3), 4.79 (1H, br s, NCH), 5.24 (m, 1H, CHOAc), 5.41 (ddd, J = 11.9, 3.7, 1.0 Hz, 1H, 198 CCH=CH), 5.57-5.77 (m, 3H, CCJ*=CH, NCHCtf=Ctl). 7.08 (d, J= 9.0 Hz. 1 H. ArH). 7.47 (dd. J=

8.9. 2.6 Hz. 1H, ArH). 7.82 (d. J= 2.6 Hz, 1H, ArH); 13C NMR (CDCI3 , 300 MHz) 5 -2.1 (q). 10.2 (1).

19.9 (t). 21.1 (q). 23.8 (t), 24.3 (1), 25.4 (t). 26.6 (t), 36.8 (d), 39.8 (t). 45.4 (t), 49.7 (t). 50.1 (t).

50.4 (s). 53.9 (d). 56.7 (q). 68.1 (d). 113.9 (d), 123.4 (d). 123.6 (d), 128.1 (d). 130.6 (d). 132.1

(d). 132.3 (d). 135.8 (s). 139.3 (s). 151.6

C 3 lH 4 5 N3 0 gSSi m/z 647.2699, calcd for M+-CH 3 m/z 632.2465, found m/z 632.2485.

Continued elution afforded 159 mg (38%) of secondary amide 416 as an off-white solid: IR

(CH2CI2) 3346.1729.1660 cm’1; 1H-NMR (CDCI3 . 300 MHz)* 5 -0.06 (s. 9H. Si(CH3)3l 0.95 (td,

J = 13.7, 4.4 Hz, 1H, CH2 Si), 1.09 (td, J = 13.8, 4.2 Hz, 1 H, CH2 Si), 1.35-1.75 ( m, 5H, CH2

manifold), 1 .8 6 (dd, J= 1 0 .6 , 2.5 Hz, 1 H, NCHC^), 1.90-2.05 (m. 3H. CH 2 manifold), 2.06 (s, 3H,

COCH 3 ), 2.15 (dt, J = 10.4, 2.3 Hz, 1 H, CfclCHOAcCila), 2.36 (dt, J = 17.7, 5.5 Hz, 1H,

CHCHOAcCtU), 2.50 (dm, J = 1 0 .8 Hz, 1 H, CfclCHOAcCfcb). 2.84 (dd, 13.9, 10.4 Hz, 1 H,

NCHCtk), 2.93 (td, 13.7, 4.4 Hz, 1 H, S 0 2 CH2), 3.09 (td, J= 13.8, 4.4 Hz, 1 H. SQ2CH2), 3.35

(m, 3H, S0 2 NCH2 and CONHCU2 (1 H)), 3.74 (ddd, J= 13.8, 10.9, 2 .8 Hz, 1 H, CONHCtk). 4.81

(m, 1H, NCH), 5.18 (ddd, J= 10.2, 6.3 , 3.8 Hz, 1H, CHOAc), 5.41 (ddd, J - 11.9, 3.1,1.0 Hz, 1H,

NCHCH=), 5.58 (m, 2 H, NCHCH=CM and CHOAcCH2 CtL=), 5.68 (dd, J = 1 0 .1 , 1 .0 Hz, 1 H,

CHOAcCH2 CH=CU). 6.07 (s, 1H, NH); 13C NMR (CDCI3 , 75.5 MHz) 8 -2.1. (q), 10.2 (t), 19.4 (t),

21.0 (q), 23.9 (t), 24.5 (t), 25.3 (t), 26.7 (t), 36.6 (d), 40.1 (1), 41.0 (t), 45.5 (t), 49.0 (t). 49.4 (s),

53.4 (d), 68.2 (d), 123.2 (d), 128.4 (d), 129.4 (d), 131.8 (d), 170.1 (s), 174.0 (s); exact mass calcd for C24H4oN20sSSi m/z 496.2429, found m/z 496.2413. A portion of this material was recrystallized from dichloromethane-hexanes to provide white crystals (mp 204.5-205.0°C).

Anal. Calcd. for C2 4 H4 oN2 0 5 SSi: C, 58.03; H, 8.12. Found: C, 57.93; H, 8.15. 199

SES

4 2 7

(±)-(2/?*)-2-[[{4afl*,5S*p8a/?*)-1-Ethoxy-4,4«t5,6-tetrahydro-5-hyclroxy-

8a(3W)-i*oquinolyl]methyl]-1,2,5,6,7,8-hexahydro-1-[(2-(trlmethylsilyl)ethyl]- sulfonyl] azocina acatata (astar). (427). To a solution of 29 mg (0.060 mmol) of amide

418 in 3 mL of dry dichloromethane at rt was added 30 mg (0.16 mmol) of triethyloxonium tetrafluoroborate. The resulting mixture was stirred at rt for 2.5 h, was diluted with 40 mL of dichloromethane, and was washed with 1 0 mL of ice-cold aqueous saturated sodium bicarbonate.

The organic layer was dried (K 2 CO3 ) and concentrated in vacuo to yield 30 mg (98%) of imino ether 427 as a brown oil: IR (neat) 1720, 1630 cm'1; 1H NMR (CDCI3, 300 MHz) 5 0.02 (s, 9H,

Si(CH3)3), 0.85 (m, 1H, CH2 manifold), 0.94 (td, J = 13.8, 4.2 Hz, 1H, CH2Si), 1.07 (td, J= 13.8,

4.2 Hz, 1H, CH2 Si), 1.24 (t, J = 7.0 Hz, 3H, CH2C tb ). 1.51 (dd, J = 12.5, 4.8 Hz, 1H, CH2 manifold), 1.50-1.90 (m, 3H, CH2 manifold), 1.73 (dd, J= 13.8, 4.2 Hz, 1H, NCHCtfc), 1.95-2.15

(m, 2 H, CH2 manifold), 2.05 (s. 3H, COCH3), 2.18 (dm, J= 1 2 .6 Hz, 1 H, CHOAcCtb). 2.26 (m, 1 H,

CHOAcCtD. 2.42 (dm, J= 1 2 .8 Hz, 1 H, CHOAcCfcy, 2.79 (dd, J = 13.8, 1 1.3 Hz, 1 H. NCHCtk),

2.90 (m, 1H, =CHCH 2 CH2), 2.93 (Id, J= 13.8, 4.4 Hz, 1H, S 0 2 CH2), 3.07 (td, J = 13.6, 4.4 Hz,

1H, S 0 2 CH2), 3.36 (m, 2 H, =NCH and S 0 2 NCH2), 3.62-3.79 (m, 2H, =NCH 2 and S 0 2 NCH2),

3.98 ( two q, J = 7.1 Hz, 2H, OCH2), 4.67 (m, 1H, NCH), 5.15 (ddd, J = 11.8 , 6.4 , 3.8 Hz, 1H,

CHOAc), 5.39 (dq, J = 1 1 .8 ,1.3 Hz, 1 H, CCH=CH), 5.56 (m, 2 H, NCHCfcL=CH), 5.67 (d, J = 11.3

Hz, 1H, CCH=); 13CNMR (CDCI3 , 75.5 MHz) 8 -2.1 (q), 10.3 (t), 14.1 (q), 19.3 (t), 21.0 (q), 23.8 (t),

24.3 (t), 25.4 (t), 26.7 (t), 36.5 (d), 39.4 (t), 45.4 (s), 45.7 (t). 46.2 (t), 49.3 (1), 53.7 (d), 60.6 (t), 200 6 8 .1 (d), 122.9 (d), 127.7 (d). 129.3 d), 131.3 (cJ), 162.8 (s), 170.2 (s); exact mass calcd for

C2 6 H4 4 N2 0 5SSi m/z 524.2742, found m/z 524.2750.

OAc

4 3 7

(±)-(2fl*)-2-[[(4a/?*,5S*I8aft*)-2,3,414a15t6-hexahydro-5-hydroxy-1- thioxo-8a(1H)-isoquinolyl]methyl]-1,2,5,6,7,8-hexahydro-1-[[2-(trfmethyl- sllyl)ethyl]-sulfonyl] azocine acetate (ester) (437). To a mixture of 57 mg (0.12 mmol) of amide 418 in 1 mL of dry toluene was added 109 mg (0.27 mmol) of Lawesson's reagent. The resulting mixture was heated at 120°C for 8 h, cooled to room temperature, and concentrated in vacuo. The residue was chromatographed three times, first over 1 g of silica gel (eluted with ethyl acetate-hexanes, 1:9 (15 mL) to ethyl acetate-hexanes, 1:3), then over 1 g of silica gel (eluted with ethyl acetate-hexanes, 1:9(10 mL) to ethyl acetate-hexanes, 1 :5), and finally over 1 g of silica gel (eluted with dichloromethane-ethyl acetate, 49:1) to yield 20 mg (33%) of the thiolactam 437 as a clear film: IR (neat): 1729 cm'1; 1H NMR (CDCI 3 ,300MHz)* 60.03 (s, 9H, Si(CH3)3), 0.87 (m,

1 H, CH2 manifold). 0.98 (td, J = 14.1, 4.6 Hz, 1H, CH2 Si), 1.11 (td, J= 13.4, 4.8 Hz, 1H, CH 2 Si).

1.45-1.90 (m, 5H, CH 2 manifold), 1.95-2.10 (m, 2H, CH 2 manifold), 2.07 (s, 3H, COCH3 ), 2.15 (dd,

J= 14.6, 6.0 Hz. 1H, NCHCtfe). 2.35 (dt, J= 17.6, 5.3 Hz, 1H, CJdCHOAc), 2.58 (dm, J= 11.9 Hz,

1 H, CHOAcCtk), 2 .6 6 (m. 1 H, CHOAcCtb), 2.92 (td, J = 13.6, 4.5 Hz, 1 H, S 0 2 CH2), 3.05 (td, J =

13.7 , 4.5 Hz, 1H, S 0 2 CH2), 3.21 (dd. J= 14.5, 8 .1 Hz, 1 H, NCHCJd^, 3.40 (m, 3H, S 02 CH2 and

CONCH2 (1H)), 3.71 (tm, J = 11.6 Hz, 1H. CONCH2), 4.78 (m, 1H, NCH), 5.29 (m, 1H, CHOAc),

5.29-5.61 (m, 3H, NCHCH^CRand CCU=CH), 5.99 (dd, J= 1 0 .2 , 1.1 Hz, 1H, CCH=CtJ). 8.50 (s,

1H, NH); 13C NMR (CDCI3 , 75.5 MHz) 8 -2.1 (q), 10.2 (t), 18.9 (t), 21.1 (q), 23.8 ( 1), 24.3 (t), 25.6 201 (I). 26.5 (I), 35.8 (d), 43.0 (t). 44.3 (t). 45.44 (t), 49.6 (t). 53.2 (s). 53.7 (d), 68.5 (d), 122.2 (d),

127.7 (d). 130.7 (d), 135.0 (d). 170.2 (s), 208.0 (s); exact mass calcd for C2 4 H4 oN2 0 4 SSi m/z

512.2201, found m/z 512.2220.

SES

4 3 8

(±)-(2/?*)-1,2,5,6,7,8-Hexahydro-2-[[(4afl*,5S*,8afl*)-4,4a,5,6- tatrahydro-5-hydroxy-1-(methylthio)-8a(3M)-lsoquinolyl]methyl]-1,2,5,6,7,8-

hexahydro-1-[[2-(trimethylsllyl)ethyl]-sulfonyl] azoclne acatata (ester).(438). To a solution of 15 mg (0.029 mmol) of thiolactam 437 in 400 pL of dichloromethane was added 60 mg (0.43 mmol) of potassium carbonate. To the heterogenous mixture was added 50 pL of methyl iodide. At 1 .5 h intervals, 250 pL of didchloromethane and 50 pL of methyl iodide were added. After a total of 6 h, the reaction was concentrated in vacuo and chromatographed over

500 mg of silica gel ( eluted with ethyl acetate-hexanes, 1:9) to yield 11 mg (71 %) of thioimidate

438 as a clear film: IR (neat) 1732. 1620 c m 1; 1H NMR (CDCI3, 300 MHz)* 5 0.03 (s, 9H,

Si(CH3)3), 0.85 (m, 1H, CH 2 manilfold), 0.95 (td. 13.8, 4.2 Hz. 1 H, CH2 SO, 1.10 (Id, J = 13.8,

4.1 Hz, 1 H, CH2 Si), 1.45-1.80 (m, 5H, CH 2 manifold). 1.96 (dd, J= 14.8 , 4.9 Hz, 1 H, NCHCfcte).

2 .0 0 (m. 1 H, CH2 manifold), 2.05 (s, 3H, COCH3), 2.17 (td, J = 1 0 .1 , 2 . 6 Hz, 1 H. CHOAcC^)-

2.22 (s, 3H, SCH3), 2.30 (td, J = 16.6, 5.1 Hz, 1H, CHCHOAc). 2.48 (dm, J, = 12.1 Hz, 1H,

CHOAcCH2), 2.70 (dd. J= 14.8, 10.4 Hz, 1 H, NCHCH2 ). 2.77 (m, 1 H, =CHCH2 CH2), 2.92 (td, J =

13.8, 4.2 Hz, 1 H, SO2CH2 ), 3.12 (td. J = 13.8, 4.2 Hz, 1 H, SO2 CH2 ), 3.37 (ddd, J = 14.1, 5.5, 2.3 202 Hz, 1H, S 0 2 NCH2), 3.48 (ddd. J = 16.6, 11.5, 4.5 Hz, 1H, C=NCH2), 3.73 (ddd, J = 13.3, 10.5,

2.5 Hz, 1 H, S p 2 NCH2), 3.96 (ddd. J = 16.8, 4.4, 1.7 Hz, 1 H, C=NCH2), 4.73 (m, 1 H, NCH), 5.21

(ddd, J = 1 0 .0 , 6.3, 3.7 Hz, 1 H. CHOAc), 5.42-5.61 (m, 3H, CHCH=dd and CH2 CJd=), 5.86 (d, J =

1 0 .0 Hz, 1H, C H ^ H -C H ); 13C NMR (CDCI3, 75.5 MHz) S -2.1 (q), 10.2 (t), 12.7 (q), 19.1 (t). 21.1

(q), 24.1(t), 24.8 (1), 25.5 (t), 26.9 (t). 36.3 (d). 40.8 (1), 45.9 (t), 49.0 (s), 49.6 (t), 50.2 (t), 53.7 (d),

68.4 (d), 122.8 (d), 128.5 (d), 128.8 (d), 132.0 (d), 170.0 (s), 170.3 (s); exact mass calcd for

C2 5 H4 2 N2 0 4 S2Si m/z 526.23358, found m/z 526.2387.

Alternatively, a mixture of 130 mg (0.262 mmol) of amide 418, 270 mg ( 0 . 6 6 8 mmol) of

Lawesson's reagent and 2 mL of toluene was stirred and heated at 130°C for 8 h, at rt for 10 h, and at 125°C for 1 h. The mixture was concentrated in vacuo and chromatographed over 3 g of

silica gel (eluted with dichloromethane (50 mL) to dichloromethane-ethyl acetate, 30:1) to yield 75

mg of a mixture of the desired thiolactam, contaminated with Lawesson's reagent, as a thick orange oil. To a mixture of this material in 4 mL of acetone was added 1 mL of methyl iodide and

150 mg of sodium carbonate. The resulting slurry was stirred at rt for 1 2 h, was filtered through a

Kimwipe plug, and was concentrated in vacuo. The residue was chromatographed over 1 g of silica gel (eluted with ethyl acetate-hexanes, 1:9) to yield 6 6 mg (48% from the secondary amide

418) of the iminothioether as a clear, colorless film.

OAc

4 2 8

(±)-(2A?*)-2-[[{4a/?*, 5S*,8a/7*)-2,3,4,4a, 5,6-hexahyd ro-5-hydroxy-8a(3 «)- isoqulnolyl]methyl]-1,2,5,6,7,8-hexahydro-1-[[2-(trlmethylsilyl)ethyl]-sulfonyl] azocine acatata (astar). (428). To a solution of 64 mg (0.122 mmol) of thiolactam 438 in 203 3.5 mL of absolute ethanol was added 0.5 mg of bromocresol green. To the blue-green mixture was added 51 mg (0.797 mmol) of sodium cyanoborohydride, resulting in a sky-blue solution. To the mixture was added 25 pL of trifluoroacetic acid over the next 30 min in four portions, each aliquot being added after the mixture changed from yellow back to blue. The reaction was stirred with 3 mL of 10% aqueous hydrochloric acid for 30 min, saturated aqueous sodium bicarbonate was added until the yellow mixture turned blue, and 5 mL of 0 .1 N aqueous sodium hydroxide was added. The aqueous layer was extracted with three 40-mL portions of dichloromethane, dried

(NagCOa), and concentrated in vacuo to yield 49 mg (83%) of the desired amine was a thick yellow oil: IR (neat) 3443,1728, cm'1; 1H NMR (300 MHz, CDCI3)* 8 0.05 (s, 9H, Si(CH3)3). 0.99 (td, J=

13.8, 4.5 Hz, 1H, CH 2 Si), 1.09 (td, J-= 13.8 ,4.5 Hz, 1 H, CH2 Si), 1.37-1.82 (m, 9H, CH 2 manifold and NH), 1.98-2.10 (m, 2 H, NHCfcb). 2.05 (s, 3H, COCH3), 2.16 (dd, J = 13.5, 10.5 Hz.1 H,

NCHCtla), 2.44 (m, 2 H, CHCHOAcCtte), 2.51 (d, J = 1 2 .2 Hz, 1 H, NCH2 C). 2.82 (m, 1H,

=CHCM2 CH2), 2 . 8 6 (td, J = 13.2, 5.0 Hz, 1H, S 0 2 CH2), 2.96 (td. J = 13.7, 4.8 Hz, 1H, CH 2 S 0 2),

3.07 (dm, J = 11.0 Hz, 1 H, CHOAcCHa), 3.23 (d, J = 12.0 Hz, 1H, NCH2 C), 3.28 (m, 1H,

S 0 2 NCH2), 3.60 (m, 1 H, S 0 2 NCH2), 4.56 (d. 9.8 Hz, 1H, NCH), 5.31 (m, 1H, CHOAc), 5.50 (d,

.7=10.0 Hz, 1 H, CCH=), 5.56 (m, 2 H, CHCH=Cti), 5.74 (ddd, J = 1 0 .0 , 4.1, 2.9 Hz, 1 H, CCH=CH).;

13C NMR (75.5 MHz, CDCI3 ), 6 -2.0 (q), 1 0 .2 (t), 2 1 .1 (q), 21.9 (t), 24.0 (t), 25.4 (t), 25.6 (t), 27.3

(t), 39.5 (s), 41.0 (d), 45.7 (t), 45.8 (t), 46.4 (t), 49.9 (t), 52.7 (t), 56.7 (t), 69.2 (d), 125.9 (d), 128.1

(d), 132.3 (d), 132.6 (d), 170.6 (s); exact mass calcd for C2 4 H4 2 N2 C>4 SSi m/z 482.2637, found m/z 482.2624. PLEASE NOTE

Page(s) missing in number only; text foNows. rmmSQ as reoeivwj*

UMI 205

4 4 0

(±H4af?*,5S*,8a/?*)-8a-[[(2/?*)-l,2,5,6l7l8-Hexahydro*l~[[2-trlmethyl-

sllyl)ethyl]sulfonyl]-2

pentenoyl)-5-lsoqulnollnol.acetate (ester) (440). To a mixture of 48 mg (0.10 mmol) of

amine 426 in 2 mL of dichloromethane at room temperature was added 30 pL (0.25 mmol) of 4-

pentenyl chloride and 60 pL (0.43 mmol) of triethylamine. The reaction mixture was stirred at rt for

1 h 40 min, and to the mixture was added 2 mL of 1 N NaOH. The mixture was stirred at rt for 10

min, and was then diluted with 50 mL of dichloromethane. The mixture was washed with two 10- mL portions of saturated aqueous sodium bicarbonate, dried (MgSC> 4 ), and concentrated in vacuo. The residue was chromatographed over 1 g of silica gel (eluted with ethyl acetate-

hexanes, 1:9 (8 mL) to ethyl acetate-hexanes, 1:3) to yield 49 mg (94%) of exocyclic amide 440

as a yellow glass: IR (neat): 1729, 1643 c m 1; 1H NMR (DMSO-d6. 300 MHz. 336K)* 8 0.03 (S,

9H, SiCH3)3), 0.93 (m. 2H, CH2Si), 1.30-2.00 (m, 11H, CH2 manifold), 2.01 (S, 3H, COCH3), 2.08

(dd, J - 14.3, 9.7 Hz, 1H, NCHCH2), 2.19, 2.28, 2.31-2.41 (m, 5H, CHCHOAcCik, COCU2),

2.70-3.10, 3.32-3.52 (m, 8 H, CH2NCH2, CH2S02CH2), 4.51 (d, J - 8.2 Hz, 1H, NCH), 4.93 (dq, J

- 10.1, 0.7 Hz. 1H. =CH2), 5.02 (dd, J - 17.1, 1.6 Hz, 1H, -C H 2), 5.21 (m, 1H, CHOAc), 5.40-5.57

(m. 3H, CCH=CH and NCHCH-). 5.63 (m, 1H. NCHCH-Ctl), 5.80 (m, 1H, CM-CH2); 13C NMR

(DMSO-dg. 75.5 MHz, 336K), 8 -1.8 (q), 10.1 (t), 21.0 (q), 24.0 (t), 25.3 (t), 25.7 (t), 27.3 (t), 29.0

(t), 31.6 (t). 40.1 (d). 44.8 (t), 45.6 (t). 49.5

133.2 (d), 138.1 (d), 169.83 (s), 169.89 (s), the 13C NMR spectrum is missing two vinylic 206 doublets, one aliphatic singlet, and two aliphatic triplets; exact mass calcd for C2gH 4 eN2OsSSi m/z 564.3055. found m/z 564.3053.

4 4 1

(±)-(4a/?*,5S*, 8aff*)-8a-[[{2/?#)-1,2,5,6,7,8-Hexahydro-1-[[2-trlmethylsllyl)ethy l]sulfonyl]-2-azoclnyl]methyl]-1,2,3,4,4a,5,6,8a-octahydro-2-(4- pentenoyl)-5-lsoqulnollnol. (441). To a mixture of 48 mg (0.085 mmol) of amide 440 in 1 mL of methanol, 1 mL of THF, and 0.5 mL of water at room temperature was added 7 mg (0.18 mmol) of lithium hydroxide monohydrate. The resulting mixture was stirred at rt for 30 min, was poured into 20 mL of water, and was extracted with 85 mL of dichloromethane. The organic layer was dried (MgSC>4 ), and concentrated in vacuo to yield 40 mg (90%) of alcohol 441 as a thick yellow oil: IR (neat) 3414, 1633 c m 1: *H NMR (300 MHz, DMSO-d6, 338K) 6 0.04 (s, 9H,

Si(CH3 >3 ), 0.93 (m, 2H, CH2Si), 1.25-2.38 (m, 16H, CH2 manifold), 2.02 (dd, 14.0, 9.3 Hz, 1H,

NCHCtfc), 2 55-3.10, 3.28, 3.52 (m, 8H, CH2NCH2 and C H jjN S O ^ H ^ , 4.08 (m, 1H, CHO), 4.40

(d, J - 3.6 Hz, 1H, OH), 4.52 (d, J - 8.3 Hz, 1H, NCH), 4.91 (d. J- 10.9 Hz, 1H, -C H 2), 5.01 (d, J -

17.2 Hz. 1H, =CH2), 5.38 (br s. 1H. NCHCH.=), 5.53 (m, 2H, CCH-CH), 5.64 (br s, 1H,

NCHCH-ChD. 5.78 (ddt, J- 17.0. 10.7, 6.5 Hz, 1H, CH-CH2); 13C NMR (DMSO-d6, 75.5 MHz,

338K) 6 -1.8 (q), 10.1 (t), 23.7 (t), 24.8 (t), 25.6 (t), 29.1 (t), 31.0 (t). 31.6 (t), 43.4 (d), 45.0 (t). 45.3

(t), 49.2 (t), 52.7 (d). 64.2 (d), 114.9 (t), 127.2 (d). 131.3 (d), 133.4 (d). 138.1 (d). 169.7 (s), the

13C NMR spedrum is missing three aliphatic triplets, one aliphatic singlet, one vinyl doublet; and shows an extra alkyl quartet and doublet; exact mass calcd for C27H46N20 2SSi m/z 522.2950, found m/z 522.2948. 207 OH

4 4 2

(±)-(la/?*,3S*13a»*,7aS*,7bS*)-7a-[[(2ff*)-l>2,5,6,7,8*Hexahydro-l[[2-

(trlmethylsilyl)ethyl]-2-azoclnyl]methyl]decahydro-6-(4-pentenoyl)oxlreno[/r}lsoqulnolin-3-ol . (442). To a solution of 8 mg (0.015 mmol) of alcohol 441 in 1 mL of benzene was added 1 mg (0.004 mmol) of molybdenum hexacarbonyl and 10 jiL (0.020 mmol) of 2.0 M t- butylhydroperoxide in toluene. The reaction was healed at 85°C for 6 h, and cooled to room temperature. The mixture was diluted with 40 mL of dichloromethane, washed with 15 mL of water, the organic layer was dried (MgS 0 4 ) and concentrated in vacuo. The residue was chromatographed over 1 g of silica gel (eluted with ethyl acetate-hexanes, 1:2, to ethyl acetate- hexanes, 1:1) to yield 3 mg (37%) of epoxy alcohol 442 as a white film: IR (neat) 3418, 1633 cm *

1; 1H NMR (DMSO-de, 300 MHz, 343 K) 5 0.04 (s, 9H, Si(CH3)3), 0.94 (m, 2H, CH2Si), 1.5-1.9,

2.05-2.4 (m, 20H), 2.7-3.0, 3.32-3.55 (m, 6H), 3.91 (br s, 1H, CHO), 4.31 (m, 2H), 4.52 (m, 1H,

NCH), 4.92 (dd, J = 10.3, 1.1 Hz, 1H, =CH2), 5.02 (dd, J = 17.3, 1.5 Hz, 1H, =CH2), 5.61 (m, 2H,

NCHChL=CH), 5.85 (m, 1H, CH=CH2); exact mass calcd. for C2 7 H4 6 N2OsSSi m/z 538.2899, found m/z 538.2909.

OAc OH

444 445 208

(±)-(2/7*)>2[[4afl*,55*l8aR*)-2,3,4,4a,5l6-Hexahydro-5-hydroxy-2‘(4' pentenyl)-i-oxo-8a(lH)-lsoqulnolyl]methyl]-1,2p5,6,7,8-hexahydro-1-[[2-(trimethylsllyl)sulfonyl]azoclne acetate (ester) (444) and (±_)-(2fl*)-

2[[4a#?*P5S*,8aR*)-2,3p4p4ap5,6-Hexahydro-5-hydroxy-2-(4-pentenyl)-i-oxo-

8a(lH)-lsoqulno!yl]methyl]-1,2,5,6,7t8-hexahydro-1-[[2-(trimethylsilyl)sulfonyl]azoclne (445). From Amide 418: To a heterogeneous mixture ot 273 mg (0.55 mmol) ol amide 418 in 20 mL of dry toluene was added 160 mg (0.61 mmol) of 18-crown-6 and

45 mg (1.1 mmol) of potassium hydride. The dark brown mixture was stirred at rl for 5 min, and 1 .6 mL (13.5 mmol) of 5-bromo-1-pentene was added in one portion. The reaction mixture was stirred at rt for 2.5 h, and was quenched by the addition of 3 mL of saturated aqueous ammonium chloride. The mixture was poured into 50 mL of water and extracted with two 150-mL portions of dichloromethane. The combined organic layers were dried (MgSCU) and concentrated in vacuo.

The residue was chromatographed over 20 g of flash silica gel (eluted with ethyl acetate-hexanes,

1:9 (400 mL) to ethyl acetate-hexanes, 1:3 (200 mL) to ethyl acetate-hexanes, 1:1) to yield 134 mg (43%) of acetate 444 as an off-white solid: IR (neat) 1731, 1635 cm'1; 1H NMR (CDCI 3 , 300

MHz) 5 0.02 (s, 9H, Si(CH3 )3 ), 0.93 (td, J = 13.8, 4.3 Hz, 1 H, CH2 Si), 1.08 (td, J= 13.7, 4.3 Hz,

1H, CH2 Si), 1.40-2.00 (m, 12H, CH2 manifold), 2.05 (s, 3H, COCH3 ), 2.12 (dt, J = 10.4, 2.4 Hz.

1 H, CHCHOAcCtb). 2 .3 4 (dt, J= 17.6, 5.3 Hz, 1 H, CHCHOAcCtb). 2.48 (d, J = 12.4 Hz, 1H,

CtiCHOAcCtb). 2 .7 9-2 .9 9 (m, 3 H, SO2CH2 (1H), NCHCH=CHCb2 (1H), NCHCtb (1 H)), 3.07 (td,

J= 13,7, 4.3 Hz, 1 H, S 0 2CH2), 3.18-3.50 (m, 5H, CH 2 NCH2 and S 0 2 NCH2), 3.72 (ddd, J= 13.7,

10.8, 2.5 Hz, 1 H, CH2 NCH2 or S 0 2 NCH2), 4.72 (m, 1 H, NCH), 4.94-5.04 (m, 2 H, =CH2), 5.14

(ddd, J = 10.3, 6.3, 3.9 Hz, 1 H, CHOAc), 5.27 (ddd, J = 11.9, 3.3, 1.4 Hz, 1H, NCHCfcW, 5.39-

5.58 (m, 2H, NCHCH=CH and CCH=Ctl), 5.67 (d, J = 10.1 Hz. 1H, CCH=), 5.79 (ddt, J = 17.0,

10.3, 6.5 Hz, Hz, 1H, CiJ=CH2); 13C NMR (CDCI3 , 75.5 MHz) 5-2.1 (q), 10.2 (t), 19.4 (t), 21.0 (q),

23.8 (t), 24,3 (t), 25.2 (t), 25.9 (1), 26.7 (t), 31.0 (t), 36.8 (d), 40.2 (t), 45.5 ( 1), 46.9

(t), 49.4 (S), 53.4 (d), 68.3 (d), 114.9 (t), 122.5 (d), 127.9 (d), 129.7 (d), 132..6 (d), 137.7 (d), 209

170.1 (s), 171.2 (s); exact mass calcd. for C a g ^ g ^ O s S S im/z 564.3055, found m/z 564.3054.

A portion of this material was recrystallized from dichloromethane-hexanes to yield white crystals

(mp 113.5-115°C).

Anal. Calcd. for C2 9H4 sN2 C)5 SSi : C, 61.67; H, 8.56. Found: C. 61.70; H, 8.49,

Continued elution provided 43 mg (15%) of homoallylic alcohol 445 as an off-white solid: IR (neat)

3426, 2932, 1622, 1400 cm'1; 1H NMR (CDCI 3 , 300 MHz) 5 0.03 (s, 9H, Si(CH3)3), 1.03 (m, 2 H

CH2 Si), 1.43-1.75 (m, 6 H, CH2 manifold) 1.79 (dd, J = 14.0, 4.0 Hz, 1 H, NCHCfci2 ),1 82-2.09 (m,

7H, CH2 manifold), 2.29 (dt, J = 17.6, 5.4 Hz, 1 H, CMCHO), 2.44 (d, 12.2 Hz, 1H, CHOCtte,

2.67 (dd, J = 14.1, 9.0 Hz, 1 H, NCHCtI2 ). 2.70 (m, 1 H, NCH=CHCtd2), 2.92 (m, 2H, S 0 2 CH2),

3.17-3.61 (m, 6 H, CH2 NCH2 and S 0 2 NCH2), 4.15 (ddd, J = 10.1, 6.1, 3.9 Hz, 1 H, CHO) 4.64 (m,

1H, NCH), 4.97-5.05 (m, 2 H, =CH2), 5.37 (dm, J = 11.9 Hz, 1 H, NCHCJd=), 5.46-5.59 (m, 2H,

NCHCH=Chl and CCH=CH), 5.63 (d, J = 1 0 .2 Hz, 1 H, CCH=), 5.79 (ddt, J = 17.0, 10.3, 6 . 6 Hz,

1H, CU=CH2); 13C NMR (CDCI3, 75.5 MHz), 8 -2.1 (q). 10.1 (t), 18.6 (t), 24.1 (t), 24.4 (t), 25.5 (t),

26.0 (t), 29.6 (t), 31.0 (t), 39.9 (d), 41.1 («), 45,6 (t), 47.4 (t), 47.9 (t), 49.8 (t), 50.1 (s), 54.0 (d),

65.3 (d). 114.9 (t), 123.6 (d), 127.1 (d), 130.7 (d). 132.0 (d), 137.7 (d), 171.6 (s); exact mass calcd for C 2 7 H 4 6 N 2 0 4 SSi m /z 522.2950, found m /z 522.2928. A portion of this material was recrystallized from dichloromethane-hexanes to provide white crystals (mp 123.5-124.5°C).

Anal. Calcd. for C27H46N20 4SSi : C, 62.03; H, 8.87. Found: C, 62.00; H, 8.82.

Continued elution provided 53 mg ( 1 9 %) of recovered 418.

Alcohol 445 from Acetate 444: To a mixture of 117 mg ( 0 .2 1 mmol) of acetate 444 in 2 mL of methanol, 2 mL of THF, and 1 rnL of water was added 20 mg (0.50 mmol) of lithium hydroxide monohydrate. The mixture was stirred at rt for 1 h, concentrated in vacuo, and chromatographed over 5 g of silica gel (eluted with ethyl acetate-hexanes, 1:9) to provided 100 mg (91%) of the desired alcohol as a pale yellow solid.

Alcohol 445 from Amide 418: To a mixture of 55 mg (0.11 mmol) of amide 418 in

15 mL of toluene was added 8 mg (0.20 mmol) of potassium hydride and 39 mg (0.15 mmol) of 18- 210

crown-6 . The mixture became a brown color and 0.50 mL (4.2 mmol) of 5-bromo-1-penlene was

added via syringe. The reaction stirred at rt for 2.5 h, and 3 mL of saturated aqueous ammonium

chloride was added. The mixture was poured into 30 mL of water and extracted with 100 mL of

dichloromethane. The organic layer was dried (MgS 0 4 ) and concentrated in vacuo. The residue

was dissolved in 1 mL of methanol, 1 mL of THF, and 0.5 mL of water, and to the mixture was

added 11 mg (0.28 mmol) of lithium hydroxide monohydrate. The mixture was stirred at room

temperature for 1 h, and concentrated to approximately one-fifth of the original volume. The

residue was partitioned between 10 mL of water and 30 mL of dichloromethane. The organic

layer was dried (MgSC> 4 ) and concentrated in vacuo. The residue was chromatographed over 2 g

of silica gel (eluted with ethyl acetate-hexanes, 1:9 (8 mL) to ethyl acetate-hexanes, 1:3) to

provide 45 mg ( 6 8 %) of alcohol 418.

OH OH

SES SES

448 449

(±)-(2fl*)-l, 2,5,6,7,8-Hexahydro-2-[{(l afl*,3S*, 3a/7\7aS*,7bS*)-

octahydro-3-hydroxy-6-(4-pentenyl)-7-oxooxlreno[/r]isoqulnolln-7a(2H)-

yl]methyl]-1-[[2-(trlmethylsllyl)ethyl]sulfonyl]azocine (448) and (±>(2ff*)-2-

[[(1a/?*,3S\3a/?*,7aS*,7bS*)-Octahydro-3-hydroxy-6-(4-pentenyl)-7- oxooxlreno[/?]lsoqulnolln-7a(2H)-yl]methyl]-3-[[2(trimethylsilyl)ethyf]sulfonyl-9- oxa-3-azablcyclo[6.l.0]nonane (449). To a solution of 48 mg (0.092 mmol) of alcohol

445 in 2 mL of benzene at rt was added 2 mg (0.008 mmol) of molybdenum hexacarbonyl and 80 pL (0.17 mmol) of 2.1 M f-butyl hydrogen peroxide in toluene. The mixture was heated in an oil bath at 105°C for 3.5 h, cooled to rt, and concentrated in vacuo. The residue was chrom

atographed over 2 g of silica gel (eluted with ethyl acetate-hexanes, 1:3 ) to yield 7 mg (15%) of

recovered starting material. Continued elution provided 25 mg (50%) of epoxy alcohol 448 as a

clear, thick oil: IR (neat) 3418, 1622 c m 1; 1H NMR (CDCI 3 , 300 MHz) 5 0.04 (s, 9H, Si(CH3)3),

1.04 (m, 2H, CH2 Si) 1.45-2.15 (m, 15H, CH2 manifold), 2.42 (m, 1H, CtiCHOAc), 2.60 (dd, J =

13.8, 9.5 Hz, 2 H, NCHCld2 ).2 .6 2 (m, 1 H, NCHCH=CHCb2 ), 2.92 (t, J = 9.0 Hz, 2 H, S0 2 CH2),

3.24-3.62 (m, 8 H, CH2 NCH2, CH2 NS02, CH-O-CH), 4.00 (br.s, 1H, CHO), 4.60 (m, 1 H, NCH),

4.94-5.06 (m, 2H, =CH2), 5.37 (dm, J= 12.5 Hz, 1H, NCHCtf=), 5.57 (m, 1H, NCHCH=CH). 5.78

(ddt, J - 16.9, 10.2, 6.7 Hz, 1H, CH=CH2); 13C NMR (CDCI3, 75.5 MHz) S -2.1 (q), 10.1 (t), 20.8

(t), 24.5 (t), 25.2 (t), 25.5 (t), 26.0 (t), 28.8 (t), 31.0 (t), 38.5 (d), 39.0 (t), 44.6 (s). 45.6 (t), 47.0 (t),

48.0 (t), 50.5( t), 52.8 (d), 54.0 (d), 59.2 (d), 66.5 (d), 114.9 (t), 129.0 (d), 129.9 (d), 137.8 (d),

170.8 (s); exact mass calcd for C2 7 H4 6 N2 0 5 SSi m/z 538.2899, found m/z 538.2905. Continued

elution provided 10 mg (20%) of bisepoxide 449 as a pale yellow film: IR (neat) 3418, 2928,

1622, 1494 cm'1; 1H NMR (CDCI3, 300 MHz) 5 0.05 (s, 9H, Si(CH3)3), 1.06 (m, 2H, CH 2 Si), 1.45-

2.15 (m, 13H, CH2 manifold and OH), 2.24 (m, 2H, CtlCHOHCtte (2H)), 2.42 (dd, 15.0, 6.0 Hz,

1 H, NCHCH2 ), 2.53 (dd, J = 15.0, 4.6 Hz, 1 H, NCHCB2 ), 2.80 (m, 1 H, CHCHOHCfck (1 H)), 2.92

(m, 2H, S 0 2 CH2), 3.03-3.35, 3.38-3.50 (m, 9H, CH-O-CH, CH-O-CH, S 0 2 NCH2), CH2 NCH2 (3H)),

3.77 (m, 1H, CH2 NCH2), 3.93 (br s, 1H, CHO), 4.02 (m, 1 H, NCH), 4.98 (m, 2H, =CH2), 5.79 (ddt,

J=» 17.0. 10.2, 6.7 Hz, 1H, CH=); 13C NMR (CDCI3, 75.5 MHz) 5 -2.0 (q), 10.2 (t), 22.2 (t), 22.6 (t),

25.0 (t), 25.9 (t), 27.7 (t), 30.0 (t), 31.0 (t), 39.3 (1), 39.6 (t), 43.4 (s), 46.5 (t), 47.9 (t), 48.8 (t), 51.3

(t), 54.2 (d), 55.8 (d). 57.2 (d), 57.4 (d), 58.2 (d), 69.0 (d). 115.0 (t), 137.6 (d), 170.8 (s); exact mass calcd.for C27H46N20$SSim/z 554.2848; found m/z 554.2853. 212

4 4 6

(±J-(4a#f,5S*I7/?*l7a«M 4a/?*,15aS,)-2,3t4I4a,5,6,7,7a,9,10,11p12p

14a,15-Tetradecahydro-5I7-dlhydroxy*2-(4-penienyl)-lH-azoclno-[1'I2*:1,5] pyrrolo[ 2 ,3 l]lsoquinolin-1 -one (446). To a mixture of 115 mg (0.214 mmol) of epoxy alcohol 448 in 3 mL of DMF was added 140 mg (0.927 mmol) of cesium fluoride. The reaction was stirred and heated at 90-95°C for 20 h. The mixture was cooled to rt, diluted with 20 mL of dichloromethane, and the solid residues were removed by filtration. The filtrate was concentrated in vacuo, and the residual DMF was removed under vacuum (1.0 mm Hg). The residue was chromatographed over 2 g of silica gel (eluted with ethyl acetate-hexanes, 1:1 (10 mL) to ethyl acetate-methanol, 7:1 (10 mL) to ethyl acetate-methanol, 3:1) to yield 6 8 mg (85%) of diot 446 as a yellow glass: IR (neat) 3351, 1610 c m 1; 1H NMR (C 6 D6, 300 MHz) 4 5 1.15-1.50 (m, 3H, CH2 manilold), 1.50-1.63 (m, 3H, CH2 manifold), 1.65-1.90 (m, 4H, CH 2 manifold), 1.90-2.14 (m, 7H,

CH2 manifold and CCH), 2.37 (dd, J = 11.7, 7.7 Hz, 1H, CHNCfck), 2.58 (t, J = 11.5 Hz, 1H,

CHNCHa), 2.74 (m, 1 H, CONCH2), 3.08 (dt, J = 13.6, 7.4 Hz. 1 H, CONCH2), 3.43

7.0 Hz, 1H, CONCH2), 3.88 (m, 2H, NCHC and CONCH2), 3.97 (s, 1H. CHOH), 4.06 (m, 1H,

NCiJCH=), 4.27 (s, 1H, CHOH (1H)), 4.61 (brs, 1H, OH), 4.76 (d, J= 9.2 Hz, 1H, CHOH), 4.98 (dt,

J = 10.1, 1.0 Hz, 1H, =CH2), 5.05 (dq, J = 17.2, 1.6 Hz, 1H, =CH2), 5.50 (t, J = 10.5 Hz, 1H,

NCHCtL=), 5.77 (ddt, J = 17.9, 10.1, 6.7 Hz, 1 H, CH=CH2), 5.91 (dt, J = 10.4, 7.8 Hz, 1 H,

NCHCH=CJd): 13C NMR (C6 D6, 75.5 MHz) 5 23.3 (t), 26.3 (t), 27.7 (1), 28.2 (t). 29.5 (t), 30.2 (t),

31.5 (t), 41.2 (d), 44.8 (t), 46.3 (t), 46.9 (t), 47.9 (t), 54.3 (d), 65.1 (d), 68.5 (d), 72.7 (d), 115.0 (1), 213

130.9 (d), 133.6 (d), 138.5 (d), 174.8 (s), the 13C NMR spectrum is missing one aliphatic singlet; exact mass calcd for C2 2 H3 4 N2 O3 m/z 374.2571, found m/z 374.2534.

O H

4 5 0

15-Dodecahydro-7-hydroxy>2-{4-pentenyl)'1H-azoclne-[rl2,:1,5]-pyrrolo(2l3-

/Jisoqulnolln-1,5-dfone (450). To a mixture of 6 8 mg (0.18 mmol) of diol 446 in 5 mL of

DMSO at rt was added 118 mg (0.42 mmol) of iodoxybenzoic acid. The resulting mixture was stirred at rt for 4 h, and was partitioned between 60 mL dichloromethane and 10 mL of saturated aqueous sodium bicarbonate. The organic layer was dried (Na 2 C0 3 ), concentrated in vacuo, and

DMSO was removed in vacuo (1.0 mm Hg). The residue was chromatographed over 2 g of silica gel (eluted wilh ethyl acetate-hexanes, 1:1 (10 mL) to ethyl acetate) to provide 36 mg (53%) of the desired ketone: IR (neat) 3389,1715,1634 cm'1; 1H NMR (CeD 6 , 300 MHz) 81.12-1.60 (m, 7H,

CH2 manifold), 1.75-2.12 (m, 7H, CH2 manifold), 2.25 (d, J= 13.5 Hz, 1H, COCH), 2.39-2.54 (m,

3H, CH2 NCH2 (1H), COCH2 (1H), CHNCtfc (1H)), 2.62 (dd, J= 14.4, 2.8 Hz, 1H, COCH2), 2.81 (t,

10.1 Hz, 1 H, CHNCd2 ). 2.98 (m, 1 H, CH2 NCH2). 3.18 (m, 1 H, CH2 NCH2), 3.41 (td, J= 1 2 .1 ,

4.5 Hz, 1H, CH2 NCH2), 3.93 (d. 4.2 Hz, 1H, NCHCHO), 4.20 (m, 1H, NCtiCH=), 4.29 (m, 1H,

CHO), 4.96 (m, 2H, =CH2), 5.59 (t, J= 10.0 Hz, 1 H, N C H C tH , 5.70 (ddt, J= 16.8, 10.1, 6.7 Hz,

1 H, CH=CH2), 5.87 (m, 1 H, NCHCH=CH), 1H NMR spectrum is missing the OH; 13C NMR (C6 De,

75.5 MHz) 5 18.8 (t), 26.2 (t), 27.7 (t), 28.0 (t), 29.1 (t), 31.3 (t), 43.1 (t), 44.4 (t), 45.1 (t), 46.4 (t), 214

47.3 (t). 47.9 (t), 49.5 (d), 53.6 (s), 67.9 (d), 68.7 (d), 115.0

172.1 (s), 207.0 (s); exact mass calcd tor C2 2 H3 2 N2 O3 m/z 372.2415, found m/z 372.2406.

OH OH

4 5 1

(±J-{4a/7*,7n*t7aR*l14aR*,l5aS*)-5-m>Dithlan-2-yl-2,3,4,4a,5,6,7l7a,9,

10,11,12,14a,15-tetradecahydro-5,7-dlhydroxy*2-(4-pentenyl)-1H-azocino-

[1al2,-1,5]pyrrolo[2,3-/]lsoquinolln-1‘One (451). To a mixture of 67 mg (0.56 mmol) of

1,3-dithiane in 3 mL of THF cooled in a dry ice-acetone bath was added 250 pL (0.38 mmol) of n- butyl lithium in hexanes. The mixture was stirred in the cooling bath for 1 h and ketone 450 was added in 1 mL of THF dropwise. The reaction mixture was stirred in the bath for 1 h 15 min, and was quenched by the addition of 200 pL of methanol. The reaction was poured into 45 mL of dichloromethane, washed with 10 mL of saturated sodium bicarbonate, dried (I^ C O a ), and concentrated in vacuo to yield 21 mg (63%) of diol 451 as a yellow film: IR (neat) 3359,1614 cm*

1; 1H NMR (CeDe, 300 MHz)* 8 1.05-1.65 (m, 9H, CH 2 manifold), 1.67 (dd, J = 12.8, 3.8 Hz, 1 H,

NCHCJJ2 ). 1 85-2.05 (m, 5H, CH 2 manifold), 2.15-2.40 (m, 5H, CH2 manifold), 2.45 (m, 4H,

(CH2 )2 S), 2.64 (dd, J = 6.5, 3.2 Hz, 1H, CCH), 2.82 (m, 1 H, CH2 NCH2), 3.08 (ddd, J = 13.4, 9.6,

6 .1 Hz, 1 H, CH2 NCH2), 3.34

4.03 (m, 1H, NCiJCH=), 4.23 (br s, 1H, CHOH), 4.32 (br s, 1H, OH), 4.43 (s,1H, S2CH), 4.93 (dt, J

- 1 0 .2 , 1.0 Hz, 1 H, =CH2 ),5.00 (d, J = 15.6 Hz, 1 H, =CH2), 5.27 (t. J = 10.0 Hz, 1 H. NCHCti=).

5.53 (s, 1 H, OH), 5.67-5.79 (m, 2 H. NCHCH=ChLand CH=CH2); 13C NMR (C6 D6. 75.5 MHz) 8 215

19.3 (t), 25.9 (t) (two C), 27.3 (t). 27.5 (t). 29.1 (t), 30.4 (t), 30.8 (t), 31.1 (t), 31.8 (t). 41.2 (d). 44.3

(t). 45.4 (I), 46.0 (I), 47.3 (t), 49.5 (s), 53.5 (d). 57.8 (d), 63.6 (d), 68.0

130.0 (d), 133.6 (d), 138.0 (d), 174.9 (s); exact mass calcd for C 2 6 H4 0 N2 O3 S2 m/z 492.2483, found m/z 492.2439. Continued elution provided 4 mg (16%) of recovered starting material.

r ' l

4 5 4

(±)*{4a#?#,7a #?*,14afl*,i5aS*)-5-/n-Dlthian-2-y 1-2,3,4,48,5,6,7,7a,9,10,

11,12,14a,15-dodecahydro-5-hydroxy-2-(4*pentenyl)-1H-azocino[l',2’-i,5]- pyrrolo[2,3-/]lsoquinolin-1,7(7aH)-dlone (454). To a solution of 15 pL (0.17 mmol) of oxalyl chloride in 1 mL of dichloromelhane cooled in a dry ice-acetone bath was added 30 pL

(0.42 mmol) of dimethyl sulfoxide. The mixture was stirred in the cooling bath for 35 min and 21 mg (0.043 mmol) of diol 451 in 1 mL of dichloromethane was added dropwise. The reaction mixture stirred in the bath for 1.5 h and 30 pL (0.22 mmol) of triethylamine was added. The mixture stirred in the cooling bath for 45 min, at rt for 30 min, and was poured into 30 mL of dichloromethane. The mixture was washed with 5 mL of saturated aqueous sodium bicarbonate, dried (Na 2 C0 3 ), and concentrated in vacuo. The residue was chromatographed over 1 g of silica gel (eluted with ethyl acetate-hexanes, 1:1) to yield 13 mg (62%) of ketone 454 as an oily yellow solid: IR (neat) 3334, 1715, 1632 cm'1; 1H NMR (CeD6, 300 MHz) 5 0.90-2.40 (m, 22H), 2.45

(dd. J = 12.5, 7.2 Hz, 1H), 2.62 (d, J = 13.4 Hz, 1H, COCH2). 2.63-2.77 (m, 2H), 3.08 (m. 1H,

CH2NCH2), 3.22 (m. 1H, CH2NCH2), 3.38 (m, 1H, CH2NCH2), 3.42 (d. J= 13.4 Hz, 1H, COCH2), 3.51 (td, J= 11.5, 5.0 Hz, 1 H, CH2 NCH2). 4.09 (m , 1 H, NCHCH2=). 4.20 (S. 1H, NCHC), 4.50 (S,

1H, SCHS), 4.94 (dt, J = 10.1, 1.0 Hz, 1H, =CH2), 5.01 (ddd, 17.2, 3.5, 1.5 Hz. 1H, =CH2),

5.57 (t, J= 9.8 Hz, 1H, NCHCth), 5.76 (m, 2H, NCHCH=CH and CH=CH2); 13C NMR (C6 D6, 75.5

MHz) 8 19.5 (t), 25.6 (t), 26.3 (t), 27.5 (t), 28.0

(t). 47.2 (t), 47.8 (t), 48.1 (t), 48.2 (t), 54.8 (s), 55.2 (d), 57.2 (d), 72.0 (d), 80.9 (S), 114.8 (t), 129.7

(d), 133.7 (d), 138.7 (d), 172.8 (S), 208.5 (s); exact mass calcd for C2eH38N20 2S2 m/z 474.2377, found m/z 474.2395. List of References

1 . Sakai, R.; Higa, T.; Jetford, C. W.; Bernardinelli, G. J. Am. Chem. Soc. 1986, 108, 6404.

2. Nakamura, H.; Deng, S.; Kobayashi, J.; Ohizumi, Y.; Tomotake, Y.; Matsuzaki, T.; Hirata, Y. Tetrahedron Lett. 1987,28,621.

3. Sakai, R.; Kohmoto, S.; Higa, T.; Jefford, C. W.; Bernardinelli, G. Tetrahedron Lett. 1987, 28, 5493.

4. Ichiba, T.; Sakai, R.; Kohmolo, S.; Saucy, G.; Higa, T. Tetrahedron Lett. 1988, 29, 3083.

5. Kondo, K.; Shigemori, H.; Kikuchi, Y.; Ishibashi, M.; Sasaki, T.; Kobayashi, J.J. Org. Chem. 1992, 57, 2480.

6 . Baldwin, J. E.; Whitehead, R. C. Tetrahedron Lett. 1992, 33, 2059.

7. Tsuda, M.; Kawasaki, N.; Kobayashi, J. Tetrahedron 1994, 5 0 ,7957.

8 . Tsuda, M.; Kawasaki, N.; Kobayashi, J.Tetrahedron Lett. 1994, 3 5 ,4387.

9. Rodriguez, J.; Crews, P. Tetrahedron Lett. 1994, 35, 4719.

10. Kong, F.; Andersen, R. J.; Allen, T. M. Tetrahedron Lett. 1994, 35, 1643.

11. Kobayashi, J.; Tsuda, M.; Kawasaki, N.; Matsumoto, K.; Adachi, T.Tetrahedron Lett. 1994, 35, 4383.

12. Kong, F.; Andersen, R. J.; Allen, T. M. J. Am. Chem. Soc. 1994, 116, 6007.

13. Torisawa, Y.; Hashimoto, A.; Nakagawa, M.; Hino, T. Tetrahedron Lett. 1989,38,6549.

14. Nowak, W.; Gerlach, H. Liebigs Ann. Chem. 1993,153.

15. Oliveira Imbrosi, D.; Simpkins, N. S. Tetrahedron Lett. 1989, 30, 4309. Oliveira Imbrosi, D.; Simpkins, N. S. J. Chem. Soc. Perkins Trans. 1,1991,1815.

16. Leonard, J.; Fearnley, S. P.; Hickey, D. M. B. Syntett. 1992, 272.

17. Baldwin, J. E.; Claridge, T. D. W.; Heupel, F, A.; Whitehead, R. C. Tetrahedron Lett. 1994, 35, 7829.

18. Li, S.; Kosemura, S.; Yamamura, S. Tetrahedron Lett. 1994, 35, 8217.

19. Marko, I. E.; Southern, J. M.; Adams, H. Tetrahedron Lett. 1992, 33, 4657.

20. Marko, I. E.; Chesney, A, Synlett. 1992, 275.

217 218 21. Hart, D. J.; McKinney, J. A. Tetrahedron Lett. 1989, 30, 2611.

22. Kamenecka, T. M.; Overman, L. E. Tetrahedron Lett. 1994, 35, 4279.

23. Winkler, J. D.; Siegel, M. G.; Stelmach, J. E. Tetrahedron Lett. 1994, 35, 6509.

24. Martin, S. F.; Rein, T.; Liao, Y. Tetrahedron Lett. 1991,32,6481. Martin, S. F.; Liao, Y.; Wong, Y.; Rein, T. Tetrahedron Lett. 1994, 35, 691.

25. Torisawa, Y.; Nakagawa, M.; Arai, H.; Lai, Z.; Hino, T.; Nakata, T.; Oishi, T. Tetrahedron Lett. 1990, 31,3195. Torisawa, Y.; Nakagawa, M.; Hosaka, T.; Tanabe, K.; Lai, Z.; Ogata, K.; Nakata, T.; Oishi, T.; Hino, T. J. Org. Chem. 1992, 57,5741. Nakagawa, M.; Torisawa, Y.; Hosaka, T.; Tanabe, K.; Da-te, T.; Okamura, K.; Hino, T. Tetrahedron Lett. 1993,34, 4543. Ma, J.; Nakagawa, M.; Torisawa, Y.; Hino, T. Heterocycles, 1994, 38, 609. Nakagawa, M.; Torisawa, Y.; Hosaka, T.; Tanabe, K.; Da-te, T.; Okamura, K.; Hino, T. Tetrahedron Lett. 1993, 34, 4543.

26. Brands, K. M. J.; Meekel, A. A, P.; Pandit, U. K. Tetrahedron 1991, 47, 2005. Borer, B. C.; Deerenberg, S.; Bieraugel, H.; Pandit, U. K.; Tetrahedron Lett. 1994, 35, 3191. Pandit, U. K.; Borer, B. C.; Bieraugel, H.; Deerenberg, S. Pure andAppl. Chem. 1994, 66, 2131.

27. Campbell, J. A.; Ph.D. Thesis, The Ohio State University, 1993. Campbell, J. A.; Hart, D. J. Tetrahedron Lett. 1992, 33, 6247.

28. Roblin, R. O. Jr; Lampen, J. O.; English, J. P.; Cole. Q. P.; Vaughan, J. R. Jr.J. Am. Chem. Soc. 1945, 67, 290.

29. Keck, G. E.; Enholm. E. J.; Yates, J. B.; Wiley. M. R. Tetrahedron, 1985, 41, 4079.

30. Pappo, R.; Allen, D. S. Jr.; Lemieux, R. U.; Johnson, W. S. J. Org. Chem. 1956, 21, 478.

31. Hiyama, T.; Okude, Y.; Kimura, K.; Nozaki, H. Bull. Chem. Soc. Jpn. 1982, 55, 561.

32. Takai, K,; Kimura, K.; Kuroda, T.; Hiyama, T.; Nozaki, H.; Tetrahedron Lett. 1983, 24, 5281.

33. Takai, K.; Kuroda, T.; Nakatsukasa, S.; Oshima, K.; Nozaki, H.Tetrahedron Lett. 1985, 26, 5585.

34. Takai, K.; Tagashira, M.; Kuroda, T.; Oshima, K.; Utimoto, K; Nozaki, H. J. Am. Chem. Soc. 1986, 108, 6048.

35. Jin, H.; Uenishi, J.; Christ, W.J.; Kishi, Y.J. Am. Chem. Soc. 1986, 108, 5644.

36. Rowley, M.; Tsukamoto, M.; Kishi, Y. J. Am. Chem. Soc. 1989, 111, 2735,

37. Aicher, T. D.; Kishi, Y. Tetrahedron Lett. 1987, 28. 3463.

38. Kress, M. H.; Ruel, R.; Miller, W. H.; Kishi, Y. Tetrahedron Lett. 1993, 34, 5999, 6003.

39. Tutariello, J. J.; Trybulski, E. J. J. Org. Chem. 1974, 39, 3378.

40. Hamersma, J. W.; Snyder, E. I. J. Org. Chem. 1965, 30, 3985. 219 41. Thompson, A. S.; Humphrey, G. R.; DeMarco, A. M.; Mathre, D. J.; Grabowski, E. J. J. J. Org. Chem. 1993, 58, 5886.

42. Nagarajan, S.; Ganem, B. J. Org. Chem. 1987, 52, 5044.

43. Carpino, L. A.; Carpino, B. A.; Crowley, P. J.; Giza, C. A.; Terry, P. H. Org. Syn. 1964, 44,15.

44. Hansen, J. B.; Nielson, M. C.; Ehrbar, U.; Buchardt, O. Synthesis 1982, 404.

45. Campbell, J. A.; Hart, D. J.; J. Org. Chem. 1993, 58, 2900.

46. Wipf, P.; Miller, C. P. J. Org. Chem. 1993, 58, 1575.

47. Kronenthal, D. R.; Han, C. Y.; Taylor, M. K.; J. Org. Chem. 1982, 47, 2765.

48. Flynn, D. L.; Zelle, R. E.; Grieco, P. A. J. Org. Chem. 1983, 48, 2424.

49. Wadsworth, D. H.; Geer, S. M; Detty, M. R. J. Org. Chem. 1987, 52, 3662.

50. Boutin, R. H,; Rapoport, H. J. Org. Chem. 1986,51,5320. Gemal, A. L.; Luche, J.-L. J. Am. Chem. Soc. 1981, 103, 5454.

51. Gensler, W. J.; Johnson, F.; Sloan, D. B. J. Am. Chem. Soc. 1960, 82, 6074.

52. Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993,95,1307.

53. Boehm, M. F.; Prestwich, G. D. J. Org. Chem. 1986, 51, 5447.

54. Olah, G. A.; Narang, S. C.; Gupta, B. G. B.; Malhotra, R. J. Org. Chem. 1979, 44,1247. Olah, G. A.; Husain, A.; Singh, B. P.; Mehrotra, A. K. J. Org. Chem. 1983, 48, 3667.

55. Hill, J. G.; Rossiter, B.E.; Sharpless, K. B.; J. Org. Chem. 1983, 48, 3607.

56. Ireland, R. E.; Varney, M. D. J. Org. Chem. 1986, 51, 635.

57. Nicolaou, K. C.; Webber, S. E. Synthesis 1986, 453.

58. Rocca, P.; Marsais, F.; Godard, A.; Queguiner, G. Tetrahedron Lett. 1994, 35, 2003 and references within.

59. Rinehart, K. L. Jr.; Kobayashi, J.; Harbour, G. C.; Gilmore, J.; Mascal, M.; Holt, T. G.; Shield. L. S.; Lafargue, F. J. Am. Chem. Soc. 1987, 109, 3378.

60. Doyle, T. W.; Balitz, D. M ; Grulich, R. E; Nettleton, D. E. Tetrahedron. Lett. 1981 ,22, 4595.

61. Kende, A. S.; Ebetino, F. H.; Tetrahedron Lett. 1984, 25, 923.

62. Molina, P.; Murda, F.; Fresneda, P.M. Tetrahedron Lett. 1994, 35, 1453.

63. Soerens, D.; Sandrin, J.; Ungemach, F.; Mokry, P.; Wu, G. S.; Yamanaka, E.; Hutchins, L.; DiPierro, M.; Cook, J. M. J. Org. Chem. 1979, 44, 535, and references within.

64. Arbain, D ; Sargent, M. V. Aust. J. Chem. 1987,40,1527.

65. Whaley, W. M.; Govindachari, T. R. Org. Reactions, 1951, 6, 74. 220

6 6 . Whaley, W. M.; Govindachari, T. R. Org, Reactions, 1951, 6,152.

67. Rocca, P.; Marsais, F.; Godard, A.; Queguiner, G. Tetrahedron, 1993, 49, 49.

6 8 . Boger, D. L.; Duff, S. R.; Panek, J. S.; Yasuda, M. J. Org. Chem. 1985, 50, 5782, 5790.

69. Ciufolini, M. A.; Bishop, M. J.; J. Chem. Soc., Chem. Commun., 1993, 1463.

70. Davies, S. G.; Shipton, M. R. J. Chem. Soc., Perkins Trans. 1. 1991, 501.

71. Examples of various chromium (VI) oxidants: Chromium trioxide dipyridine: Collins, J. C.; Hess, W. W.; Frank, F. J. Tetrahedron Lett, 1968, 3363. Ratcliffe, R.; Rodehorst, R. J. Org. Chem. 1970, 35, 4000. Pyrdinium chlorochromate: Dauben, W. G.; Michno, D. M.; J. Org. Chem. 1977,42, 682. Corey, E. J.; Suggs, J. W.; Tetrahedron Lett. 1975, 2647. Babler, J. H.; Coghlan, M. J. Syn. Commun. 1976, 6 ,469. Pyridinium Dichromate: Corey, E. J.; Schmidt, G. Tetrahedron Lett. 1979,399. Majetich, G.; Condon, S.; Hull, K.; Amad, S. Tetrahedron Lett. 1989, 30, 1033.

72. Overman, L. E.; Knoll, F. M.; Tetrahedron Lett. 1979, 321. Review: Overman, L. E. Angew. Chem. Intl. Ed., Engl. 1984, 23, 579.

73. Oehlschlager, A. C.; Mishra, P.; Dhami, S. Can. J. Chem. 1984,62,791.

74. Snyder, H. R.; Walker, H. G.; Werber, F. X. J. Am. Chem. Soc. 1949, 71, 527.

75. Leffler, M.T.; Org. Reactions 1942, 1, 91.

76. Melhado, L. L.; Leonard, N. J. J. Org. Chem. 1983, 48,5130. Moriya, T.; Hagio, K.; Yoneda, N. Bull. Chem. Soc. Jpn. 1975, 48, 2217. Craig, L. C. J. Am. Chem. Soc. 1934, 56, 231.

77. Barker, M. D.; Woodward, P. R.; Lewis, J. R. UK Patent Application GB 2 155 462 A, 1 985.

78. Abramovitch, R. A.; Shapiro, D. J. Am. Chem. Soc. 1956, 77, 4589.

79. Bracher, F.; Hildebrand, D. Tetrahedron 1994, 5 0 ,12329.

80. Corcoran, R. C.; Bang, S. H. Tetrahedron Lett. 1990, 31, 6757.

81. Chemla, F.; Julia. M.; Uguen, D.; Zhang, D. Synlett. 1991, 501. Satoh, T.; Motohashi, S.; Yamakawa, K. Tetrahedron Lett. 1986,27,2889. Ogura, K.; Yamashita, M.; Tsuchihashi, G. Tetrahedron Lett. 1978,19,1303. Shiner, C.S.; Tsunoda, T.; Goodman, B. A.; Ingham, S.; Lee, S.; Vorndam, P.E.; J. Am. Chem. Soc. 1989, 111, 1381.

82. Still, W. C. J. Am. Chem.Soc. 1978, 100, 1481.

83. Jones, P. S.; Ley, S. V.; Simpkins, N. S.; Whittle, A. J. Tetrahedron, 1986, 42, 6519.

84. Waldmann. H.; Schmidt, G.; Jansen, M.; Geb, J. Tetrahedron Lett. 1993,34,5867.

85. Sandrin, J.; Hollinshead, S. P.; Cook, J. M. J. Org. Chem. 1989, 54. 5636.

8 6 . Laguzza, B. C.; Ganem, B. Tetrahedron Lett. 1981, 2 2 , 1483. 221 87. Neumeyer, J. L.; Moyer, U. V.; Leonard, J. E. J. Med. Chem. 1969, 12, 450.

8 8 . Nogrady, T.; Doyle, T. Can. J. Chem. 1964, 42, 485.

89. Bracher, F.; Hildebrand, D. Liebigs. Ann. Chem. 1992, 1315.

90. Laborde, E.; Lesheski, L. E.; Kiely, J. S. Tetrahedron Lett. 1990,31,1837.

91. Dess, D. B.; Martin, J. C. J. Org. Chem. 1983,48,4155. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277. Ireland, R. E.; Liu, L. J. Org. Chem. 1993, 58, 2899.

92. Frigerio, M.; Santagostino, M. Tetrahedron Lett. 1994, 35, 8019.

93. van Boeckel, C. A. A.; van Boom, J. H.; Tetrahedron. Lett. 1979, 3561.

94. Corey, E. J.; Chaykovsky, M. J. Am. Chem. Soc. 1965,87,1353.

95. Corey, E. J.; Weigel, L. O.; Floyd, D.; Bock, M. G. J. Am. Chem. Soc. 1978, 100, 2916.

96. Corey, E. J.; Weigel, L. O.; Chamberlin, A. R.; Cho, H.; Hua, D. H. J. Am. Chem. Soc. 1980, 102, 6613.

97. Bai, D.; Bo, Y.; Zhou, Q. Tetrahedron Lett. 1990, 31, 2161.

98. Bai, D.; Shi, Y. Tetrahedron Lett. 1992, 33, 943.

99. Paterson, I.; Mansuri, M. M. Tetrahedron 1985, 41, 3569.

100. Stach, H.; Hesse, M. Tetrahedron 1988, 44,1573.

101. Corey, E. J,; Raju, N. Tetrahedron Lett. 1983, 24, 5571.

102. Just, G.; Luthe, C.; Oh, H.; Montgomery, J. Syn. Commun. 1979,9,613.

103. Nicolaou, K. C.; Petasis, N. A.; Li, W. S.; Ladduwahetty, T.; Randall, J. L.; Webber, S. E.; Hernandez, P. E. J. Org. Chem. 1983, 48, 5400.

104. Imamoto, T.; Kusumoto, T.; Tawarayama, Y.; Sugiura, Y.; Mita, T.; Hatanaka, Y.; Yokoyama, M.; J. Org. Chem. 1984, 49, 3904.

105. Brands, K. M. J. Ph.D. Thesis, Universiteit van Amsterdam, 1990.

106. Paquette, L. A. Org. Reactions 1977, 25,1. Clough, J. A. in Comprehensive Organic Synthesis 1991, 3, 861, Pergammon Press.

107. Martin, S. F.; Liao, Y.; Chen, H.; Patzel, M.; Ramser, M. N. Tetrahedron Lett. 1994, 35. 6005.

108. Martin, S. F.; Wagman, A. S. Tetrahedron Lett. 1995, 3 6 ,1169.

109. Negishi, E.; Swanson, D. R.; Rousset, C. J. J. Org. Chem. 1990, 55, 5406.

110 Kabalka, G. W.; Hedgecock, H. C. J. Org. Chem. 1975, 40, 1776.

111. Rabilloud, G. Bull. Soc. Chim. Fr. 1967, 384. 222 112. Kuehne, M. E.; Shannon, P. J. J. Org. Chem. 1977, 42, 2082.

113. Borch, R. F. Tetrahedron Lett. 1968,61.

114. Nishiyama, H.; Nagase, H.; Ohno, K. Tetrahedron Lett. 1979,4671.

115. Raucher, S.; Klein, P. Tetrahedron Lett. 1980,21,4061.

116: Scheibye, S.; Pedersen, B. S.; Lawesson, S. O. Bull. Soc. Chim. Betg. 1978, 87, 229.

117. Sundberg, R. J.; Walters, C. P.; Bloom, J. D. J. Org. Chem. 1981, 46,3730.

118. Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327. Arhart, R. J.; Martin, J, C.; J. Am. Chem. Soc. 1972, 94,5003. Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1974, 96, 4604.

119. Watson, S. C.; Eastham, J. F.;J. Organomet. Chem. 1967, 9 ,165.

120. Abbreviations are given under "Guidelines for Authors" in J. Org. Chem. 1995, 60, 12A.

121. Halligan, N. G.; Blaszczak, L. C. Org. Syn. Coll. Vol. VIII, 1993,23. APPENDIX A: 1H AND 13C NMR SPECTRA OF SELECTED COMPOUNDS

223 I

OCH,

157 BF-V-209, CDCI 3 , 200 MHz

j jLJI I

OCH3 160 BF-V-214, CDCI3i 200 MHz * impurity

T -1 i 1 j ?■■■> ~1 r 1 I r ro 2.5 2.0 1.5 1.0 fo Ul o ch3

161 BF-V-170, CDCI3, 200 MHz f

a ^ A a lull JaL 226

\ \

I I \< v

- t o

o ch3

162 BF-V-183, CDCI3, 200 MHz t~r 53i 227 - r r -?!*' 3.5 3.0 2,5 1.5 1.0 b H,CO'

165 BF-VII-212, COCI3

4 166 BF-V-239, CDCI3, 250 MHz 229 i

OAc

OCH3 156 BF-V-253-D, CDCI3, 200 MHz /

* - r - i - r ■, , , , r *■ ■ r | I , f ■ 1 ; , , , . 1 7—J-T ' I ' 230 7.0 6 5 6.0 5.5 5.0 4.5 4.0 3.5 1.5 10 .5 PPH OAc

' H,CO OH 171 BF-V-303-B/C, CDCI3, 300 MHz

1—I—r 1 s r ' I ' ' ' ! I ' 231 4 0 3.0 2 5 2.0 1.5 1.0 PPM OAc

170 BF-VI-10, CDCI3, 300 MHz

r r

DLL

, , r ’ 1 ■> ! ...... i (232 S.S 8 0 8-5 8.0 7.5 7.D 6.5 6.0 5.5 233

in (M

sea.1

N X

o o CM

o a

CO O r" r^-

> u. CO

nuoaiNl i i n t £gra : 7.0 6.5 BF-V-114, CDCI3,MHzBF-V-114,200 6.0 183 5.5 5.0 OH . 4.0 4.5 PPM 3.5 3.0 2.5 2.0 . 5 1.5

1.0 234 i N lCBRiL BF-IV-271, CDCIBF-IV-271, *impurity 186 3 . 200 MHz.200 OTHP

PPH

3.5 235 236 3.5

co o

LL OTHP

181 BF-VII-271, CDCI3 , 75.5 MHz 237 I

JNTEGRAL BF-IV-198-B, CDCI3,BF-IV-198-B,MHz 250 5.0 182 OTHP PPH 03 co ro I 1----- ' OTHP

182 BF-IV-198-B, CDCI 3, 62.9 MHz

<£> OAc

OH OTHP

173 and 174 BF-V-69-B, CDCI3, 300 MHz 240

7.5 1.0 6.5 6.0 5.5 5.0 4.0 3.5 3.0 2.0 1.5 1.0 5 PPM OAc

O ^ -O H

| ^ J - 0THP

187 BF-VII-72-B, CDCI3, 300 MHz

J f

6.0 5.5 OAc

OH H,CO' OTHP

188 BF-VII-72-D, CDCI3, 300 MHz

Jf 242

PPM OAc

tBoc j a OTHP SES

189 BF-VII-211, C0CI3l 300 MHz H,CO OTHP

195 BF-VII-43-C, CDCI3, 300 MHz 244 OAc

H3CO OTHP

195 BF-VII-43-C, CDCI3, 75.5 MHz 555 HaCO BF-VII-48-B,CDCI 198 NH2 “ > 3 OAc , 250 MHz,250

PPM 246 PPk > ° ^ O X H BF-VIII'51-B,CDCI3, MHz75.5 rr 198 OAc OTHP

1 H .25 B f

OAc

r-Boc

H,CO OTHP

197 BF-VII-55-B, CDCI3, 250 MHz

ih s J V 1 A _____ VJi \ W \ 1 V U W U UA UAI

5.0 3.0 2.5 2.0 1.5 PPH I

OAc

l-Boc H,CO OTHP

197 BF-VII-55-B, CDCI3l 75.5 MHz rm s BF-VII-56-B,COCI * impurity* 196 J X K W i 3 , 250 MHz 250 , OTHP

250 OTHP 251 0 ^ och3 204 r BF-IV-78-B, CDCI3, 250 MHz t 12.496 252 T'-'f “ T“ n 1 ' ■ ...... 111 > 11111111 6.5 5.0 <1.57.0 4.0 3.0 2.5 2.0 1.5 1.0 .53.5 0 ^ OCHj 204 BF-IV-76-B, CDCI3l 75.5 MHz * impurity

¥*W J^^rNNi253 OAc

OH 208 BF-V-269-C, CDCI 3, 300 MHz • CH2CI2

a AI w _ ^ 254

7.5 7.0 5.5 5.0 VI c V 3 * S -V - r* fl g 0 0 •£ - O 1 ' O C ' r r CJ ! - ■

OAc

208 BF-V-269-C, CDCI 3, 75.5 MHz 255 r

o c h3 209 BF-V-290-B, CDCI 3, 300 MHz

S i I f J f . Ja _ m .4hA». vM JW. JL 256 TCrnr

OAc (CH3)3COY o

OCH3 209 BF-V-295-B, CDCi3, 75.5 MHz 257 OAc

JO" OTHP

210 BF-VII-137-B, CDCI3, 300 MHz

JUL I I 258

>— 1— 1— 1— r ■ 1— 1— 1— 1— 1— I— r*—T ’ —1— 1— t — p— 1 ‘ -1 l » r » -< | » ' ' ' I 1 7.5 7.0 6,5 6.0 S.s 5 0 OAc

JCX OTHP O" 210 BF-VII-137-B, CDCI3, 75.5 MHz

iJujL j jjAi W MyuAjujiA L I 259 T ^ T n W r ^ T ir fr 7 fF *rTryT^^^*lWp OAc

O 3— OH HjCO O 0THP 211 BF-VIM15-C, CDCI3, 300 MHz

7.5 5.5 PPM 260 OAc

OTHP

211 BF-V1I-115-C, CDCI3, 75.5 MHz

10 o> OAc

H,CO OH SES

212 BF 1V-297-B, CDCI3, 300 MHz ♦ ethyl acetate

J J

W K K 262 INTEGRAL 8.5 BF-V-152, CDCIBF-V-152, . 6.0 6.5 213 3 , 200 MHz,200 U . M i l J . . — k\ \k) \ki\l y s' 3.5 / 3.0 ' T ’ l 2,5

S 263 i i

OAc

H3CO'

214 BF-VIII-12-B, CDCIg, 300 MHz

j U \ k TMK 264

\ 215 BF-VI-47-A, CDCI3i 300 MHz

PPM 265 OH

SES

216 BF-VI-47-A, CDCI3, 300 MHz

AW 266 5.5 PPH OH

SES H,CO

217 BF-V-272-B, CDCI3l 300 MHz * impurity ♦CH 2CI2

1 1 1 1 1 1 1 v ■ 1 ■ 1 1 1 ’ 1 1 1 1 1 1 1 1 1 ' ■ i 1 ■ 1 1 1 1 7-5 7-0 6.5 6.0 S.5 5.0 4.5 267 OH

SES H,CO'

217 BF-V-136'D, CDCI3i 62.9 MHz * impurity

i 268 OMOM

,SES

218 BF-V-280, CDCI 3, 250 MHz

kLojului kk u u 269

PPM OMOM

OAc

CO rr ° \-H

220 BF-VI-9-B, CDCI3) 250 MHz

ijLLl i u 221 BF-V-288-B, CDCI3, 300 MHz

- L . flL. .A. -j A uu u 271 i_t— i 1 3.5 3.0 2.5 2.0 1.5 1.0 .5 H,CO'

151 BF-V-289-B, CDCI 3, 300 MHz

0 La ilc U I

1 1 I 1 ' ' ' | ’ 1 ■ 'r | 1 1 1 i 1 1 1 1 1 j 1 f 'T ’ 7.0 6,5 6.0 5.5 5.0 4.5 4.0 PPM SES HaC0‘

227 BF-V-12-B, CDCI3, 300 MHz

* c h 2ci2

6 0 3 .5 2 .5 PPM 273

i I 2 / 5 MT 'M

OPiv

H,CO- SES

227 BF-V-12-B, CDCI3l 62.9 MHz 274 ■■ ■■ i .i— ...------— ir~ij-~i_r -#sWr SiWiQyi'ni i.n 'in f/y>^ OPiv

228 BF-V-22-B, CDCI3, 300 MHz

/ 1 1 m ^ jA l J

UUJIJ \)o j fc a t s. *i «1 r* 3 a ■» ZH a WUJ k w\ * b *i r“ n i

3 .0 2 .5 1 .5 1.0 PPM OPiv

'OH C r " y h 0 X ~ N

228 BF-V-15-C, CDCI3, 62.9 MHz 276 IfWn' BF-VII'51, CDCI3, 300 MHz * CH2CI2 277 2 7 8

*dd 266 BF-IM21-C,CDCI3, 300 MHz

1 11 11 1 11 1 1 1 n - t - 279 I I ' Y ' ' I ■ | I ■ T 1- 1 M 1 ' 1 I 1 1 1 1 | 1 1 1 1 | 1 1 » 1 | ' r » t * | 1 1 1 i | j i f * \ i 1 i 9-5 9 .0 a . 5 8.0 7 .5 |—r7- 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 .5 PPM 266 BF-II-121-C, CDCI3, 75.5 MHz 280 in t e g r a l BF-II-142-C, CDCI3, 300 MHzBF-II-142-C,CDCI3, 300 H20 *H 270 PPM

281 270 BF-II-142-C, CDCI3, 75.5 MHz 262 275 BF-II-159-B, CDCI3, 300 MHz

4 4,0 3.535 5 5 283 03 * o~ V Iif 275 BF-II-159-B, CDCI3, 62.9 CDCI3, BF-II-159-B, MHz

TrTTTTT Hdd IMTEGflAL . . 0.5 9.0 9.5 .6 0706.5 7.0 7.56.0 FI-5-, CDCI3lMHzBF-II-154-B,200 276 6.0 OAc

5.5 PPM T' ’T 3.0 2.5 2.0

1.5 285 276 BF-II-154-B, CDCI3, 62,9 MHz 286 H Cl

278 BF-1II-116-B, DMSO-d6, 300 MHz 287 I-- 1 - 0.0 6.5 278 BF-lll-116-B, DMSO-dg, 62.9 MHz H B r

283 BF-III-136-B, DMSO-d6l 300 MHz

I 289 < —r"“'—i 1 —i—’—1—' ^*” 1—■— —<—r S.O 8 .5 S.O 7 .5 7.0 283 BF-III-136-B, DMSO-d6, 62.9 MHz 290 286 BF-VI-193-B, CDCI3, 300 MHz ♦ machine artifact * impurity

I ■ I ■ ' I 1 1 ' 1 I 1 1 i "l I 1 I 1 ■ ■ ■ r - r - i ■ i | - 4-Q 3.5 3.0 2.5 2.0 1.5 J.O .5 292 1 '(JVVAW'MW’ 286 BF-VI-193-B, CDCI3l 75.5 BF-VI-193-B, CDCI3l MHz

Hr1 d 290 BF-VI-198-C, CDCI3l 300 MHz

L. w JOl — 293

' T '1 ’~T T I T" ' J 1 n" ’ ! 1 rrr 5.3 4.5 4.0 3.5 3.0 1.5 1.0 .5 PPM c

CH.

290 BF-VM98-C, CDCI3, 75.5 MHz 294 291 BF-V-191-B, CDCI3, 300 MHz

■ i 1 ■ I *" T " 1- t-T-7- 1 1 I" 1 “r" r r ' l 1 "r-r r 4.5 4.0 3.5 3.0 2.5 2.0 1.5 i.o .5 295 I

291 BF-V-191-B, CDCl3t 75.5 MHz

JL to 0> N V 0 ' (£j)

292 BF-lll-153-B, CDCI3, 250 MHz 297 I755F • ! 1 “■—I—•” 1 1 y ■ ■ ■ 1 ■ M -f-r-r- rTrr 4.5 4.Q 3.5 3.0 2.5 2.0 1.5 1.0 .5 120.764 BF-III-153-B,CDCl3t MHz 62.9 292

4T1 rv fikc £

” TIL

293 BF-VI-252, CDCI3l 300 MHz

llill 299 1

293 BF-VI-252, CDCI3, 75.5 MHz 300 294 BF-VI-269-B, CDCI3 i 300 MHz 301 - i 1 ,T' n 1 '^"i 1 • 1 ■ i ' i 1 - i ' • . j T . . . j | . 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 .5 OH

294 BF-VI-269-B, CDCl3, 75.5 MHz 302 I fW OAc 296 BF-UM59-B, CDCI3, 250 MHz ♦ ethyl acetate

♦ Jutli HjL JL SLL 303 I ■ ’ ' ■ I ■ ’ ■ I ■ ' - « | ■ -T , % T . j . T1 . , 1—. I . f'1 1 .* ,■ ' I -1 -TT -r' 9.5 9.0 0.5 0.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 PPM 296 BF-III-159-B, CDCI3, 62.9 MHz 304 iktJtlidiiiuiMilihJ linllLiUwiyU t il lltllUilJlllaAll^Ljilllldiil jilllLli U lU l& ltW lliml rrp i^ ip OAc

OMOM H,CO'

301a BF-VI-196, CDC!3, 300 MH z

II // / JUL

4.5 305

i OAc

[TV ■OMOM ^f0THP

301a BF-VI-196, CDCI3l 75.5 MHz

w Wv'dUlwJkA^ 306 OH

OMOM j c x OTHP

301b BF-VI-199-B, CDCI3, 300 MHz

3.5 3.0 2.0 1.5 1.0 307 OH

H3CO OMOM OTHP

301b 8F-VI-199-B, CDCI3, 75.5 MHz 308 0

OMOM jCr OTKP

302 BF-VI-208, CDCI3, 300 MHz 309

/ 5 7 0 6.5 6.0 5.5 5.0 o

>~OMOM HaCO OTHP U" 302 BF-VI-209, CDCI3, 75.5 MHz 310 o

OMOM jCr OTHP

300 BF-VI-206, CDCI3, 250 MHz 311 ^ 0 5 — O M O M h 3c o OTHP

300 BF-VI-206, CDCI3, 75.5 MHz HXO' OTHP

304 BF-VI-264-C, CDCl3) 250 MHz 313 j j 1 1 ■ ■ i ' ■ ■ f 'j ■ ■ • ' i »* r T i ' ’ I 1 9.5 9.0 6.5 0,0 7.5 7.0 5.5 5.0 4.5 4.0 3.5 3.0 PPM -U4.107

> — OMOM H3co U" OTHP 304 BF-VI-264-C, CDCI3, 75.5 MHz 306 BF-VI-187-B, CDCI3l 300 MHz

JU l-J V 315 BF-VI-187-B,CDCI3> MHz 75.5 1 7 5 .IQS 306

OH 316 308 BF-VI-36-B, CDCI3l 300 MHz 308 BF-VI-39-B, CDCI3( 75.5 MHz 318

w 310 BF-VI-44-C, CDCI3l 300 MHz

k . >I.M u ukA; i l .930 T 7 2 5 I 319 OAc

BF-VI-44-C, CDCI3, 75.5 MHz

CJ w o 313 BF-V1-75-C, CDCb, 300 MHz

* CH2CI2

i , ■ i • ■ i ■ ''I'' ' ’ ! 1—1 r"‘ I ' ’ ' 1 I 1 3.0 2.5 2.0 1.5 1.0 .5 321 ri

313 BF-VI-75-C, CDCI3, 75.5 MHz 322 4/Mm "I WfllMtVVVfVV'1 314 BF-VI-76-B, COCI3 , 300 MHz

u u 323

' i ■ r‘ ■ 1 | * » 1 1 3 1 > 1 1 I ' f ' i '— I 3.5 3.0 2.5 2.0 1.5 1 0 .5 OH

BF-VI-76-B, CDCI3, 75.5 MHz 324 CHO

311 BF-VI-94, CDCI3> 250 MHz * impurity

4.5 4.0 3.5 3,0 2.5 2.0 1.5 1.0 .5 325 CHO

Me 311 BF-VI-94, CDCI3 i 75.5 MHz

JJ 326

i 1

1 INTEGRAL * ■ 1 . 9.0 9.5

1 ■ 1

1 8.5 I 1 I T e.o BF-VI-137, CDCIBF-VI-137, • ♦ CH♦ 7.5

1 ■

2 It] CI 7.0 H 2 i ...... * impurity * 323

6.5 1 3 ,MHz200 1111

Jl 6.0 1 ''i ) K \)

5.5 1

I PPM 5.0 4.5 4.0 3.5 7 u 3.0 f . .5 2.5 1 2.0 I 1 '

01 -S| PO 323 BF-VI-137, CDCI3, 75.5 MHz * CH3

329a

BF-VI-151-C, CDCI3, 250 MHz * impurity

/ J E H

B.O 7.0 6.5 6,0 5.5 5.0 PDU ’CH. OH 329a BF-VM38-B, CDCI 3, 75.5 MHz 330 I

OH Me

329b BF-VM51-E, CDCI3, 300 MHz

6.5 4.5 4.0 3.5 3.0 2.5 2.0 1-5 1.0 5 331 PPM Jgj

^ OH Me 329b BF-VM51-E. CDCI3, 75.5 MHz

03 03 ro OH Me 330a ( BF-VM55-B, CDCI3, 75.5 MHz

. r - , - [-T-T , t | t r I I"*"' 1 ' I ' ’ ' T I"' ' I ' 11 1 I ' ’ ,1 I 1 1 1 ' I 1 9.5 9.0 .5 0.0 7.5 7.0 6.5 6.0 5.5 5.0 333 PPM - CH,

330a BF-VI-155-B, CDCI3l 75.5 MHz

W W f ^ v^ lywwwtf 1 334 OH Me 332 BF-VI-166-B, CDCI3, 300 MHz 335 I" ' ' ■ 1 i 1 r "l 1 1 1 1 1 ' ^ 1 i t ■ | ‘ ^ i r 1 j 1-1 1 . , 1 - , , 1 j 1 ’r T “ r 9.5 9.0 B.5 0.0 7.5 H 7,0 6.5 6.0 5.5 5 0 <.5 *.0 PPH OH Me 332 BF-VI-166-B, C0CI3, 75.5 MHz

Oi f

336 BF-VII-284-C, CDCI3 i 300 MHz

JLA Jl 337

10.5 10 0 5.5 PPM o

336 BF-VII-284-C, CDCI3, 75.5 MHz

W M | ruiVhfnrfpurtnil^M'^JLj 03 \

HiCO'

349 BF-VII-253-B, CDCI3, 300 MHz

ill A__ J*L w u u u

1 i 1 1 1 ■ r 1 ' ' t 1 I 1 ■*—1 ■ 1 1 ’ ' 7.5 7.0 6.5 6.0 5.5 5.0 3.5 3.0 339 PPM o

351 BF-VII-270-B, C6 D6, 200 MHz 340 r W r ’ OH H X O ' ^ °U

351 BF-VII-270-B, C6D6, 75.5 MHz BF-V1-82-B, CDCI3, 300 MHz

{/ II / JUL ^ r\ _ J U. J lJ

-t—r “f 342 4.5 4.0 3.5 PPM OPiv

OMOM

353 BF-V-189-B, CDCI3 , 62.9 MHz

JjL. 9J l m Ufrw JhM1^1 .343 OH

OMOM

354 BF-V-171-B, CDCI3, 300 MHz * impurity ♦ CH 2CI2

I \/y r\ / 11 _JK_ aAa aL . Jl

r 7.5 7.0 6.5 6.0 5.5 3.5 rd O O l GIBn 0 10 IT ^ J l '1 tc ic S N lO u0 tv -

94.330 h. rs r . s

tn in in 1n q a Hf; \l ((

OMOM

354 BF-V-194-C, CDCl3, 62.9 MHz 345 OMOM

355 BF-VI-89-A, CDCI 3, 250 MHz * impurity 346 6.5 3.5 2.5 2.0 PPM o

365 BF-VII-241-B, CDCI3 , 300 MHz S. 057 S. n im

| i 99.587

OH H,CO SES

365 BF-VII-241-B, CDCI3 , 75.5 MHz 348 } 0 OH

380 BF-V-130, CDCI3, 250 MHz

r r 349 I

“j INTEGRAL o n 10.0 BF-V-44, CDCI3, 200 MHzCDCI3,BF-V-44,200 378 COOH

PPM e.o 350 BF-V-52, CDCI3, 250 MHz 5J557 351 0 PPM 382 BF-V-51-B, CDCI3, 200 MHz

U k

r - p - r 1 ■ I ■ 1 inrnr ' I 1 T^- I ' 1-1 1 I T-1 " ’ ’-T" 7.5 7.0 5.5 5 .0 4 .5 4.0 3.5 3.0 2.5 2.0 i.5 1,0 .5 PPM 352 377 BF-V-53-0, CDCI3, 300 MHz

1 ■1 3.5

3.0 2.5 353 I

> J W © -

383 BF-V-87-B, CDCI3, 250 MHz

> ■ ■ ■ ■ i ■ ■ ■ ■ i ■ ■ ■ ■ v 1 ■ ■ i 1 1 1 t~ ' 1 - t 1 -r~ ' ‘ I ' 354 7 5 7.0 6.5 6.0 5.5 5.0 4.5 3.0 2.5 HO

383 BF-V-87-B, CDCI 3. 62.9 MHz 355 *1^ V»Ki» fW iCWwrfVWlfiW*' € 3 6 TsO BF-V-107-C,CDCI V 384 3 , MHz 300

-’-‘I'-''

4.0 jsl 3.5 1 I'r' 3.0

2.5

2.0 356 TsO

384 BF-V-107-C, CDCI3, 62.9 MHz 357 I

B,-T o ? 376 BF-V-225-B, CDCI3, 250 MHz

u 358 '■ i ■ ...... i 1 '""i * 1 1 1 T ' ' I 1 I * ■ ' t I I I ■ . ■ 7 5 7.0 6 5 6.0 3.5 3.0 2.5 2 .0 i.5 1.0 .5

i « * “1 a - Sa 2 q .10EMU BF-V-225-B,CDCI3lMHz 62.9 376 *

n ■* 0 * ^ a *

b r « a r a b r. B O - " IW IW ' r " m : 9 3 n 4 ►!- [

h c • c *

a p« 359 -OTHP HO-T

387 BF-V-241-B. CDCI3l 250 MHz 360 I

PPM HO BF-V-241-B,CDCI3 T - 387 r _ \ i 62.9 MHz62.9 OTHP A

m . ■* M 361 OTHP TsO

388 BF-V-243 B, CDCI3, 250 MHz

♦ CH2CI2 362

PPM CJ o> \( OTHP 388 BF-V-243-B, CDCU, 62.9 CDCU, BF-V-243-B, MHz TsO

SUB'FFT -OTHP

B r -T

385 BF-V-254-B, CDCI3, 250 MHz

—CO 4.0 3.5 3.0 2.5 2 . 0 1.5 1.0 .5 at PPM u -OTHP — r

385 BF-V-254-B, CDCI3, 75.5 MHz

JL mMi 365 i

OTHP

386 BF-VI-2-B, CDCI3, 300 MHz 366 6.5 5.5 4.5 PPM ■OTHP , _ y

386 BF-VI-2-B, CDCIg, 75.5 MHz 367 JLUVfft HO,

389 BF-V-279-B. CDCI3, 250 MHz * impurity 368

3.5 3.0 1.5 5 I

OTHP

HO.

389 BF-V-279-B, CDCl3, 75.5 MHz 369 1 BF-V-257-C, CDCI3, 300 MHz * impurity

/ J JJ AJWjjv. 370 ^y Q yLMe

393 BF-V-149-B, CDCI 3, 300 MHz f * impurity

Hi / J UU 'UU U KA)

—1 j . . ■ ■ r U 1 1 1 1 ' I r ’ •' 1 T -J 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 5 PPH BF-VII-149-B, CgD6l 75.5 MHz

/ uif 372 ■'“ I 6. 3 . 5 3 .0 393-H C I BF-VI1-159, CDCI 3, 75.5 MHz * impurity 373 I

H,CO

414 BF-VII-54-A , CDCI3, 300 MHz 374

5.5 5.0 2.5 2.0 PPM BF-VII-83-B, CDCI3l 300 MHz

^ J L ^ . / _^U

) UU ) U 375 i ■ 1 * * r * ...... 3.5 3.0 2.5 2.0 1.5 1.0 .5 r r Q ^ h3c o ^ 0

415*HCI BF-VII-218, CDCI 3, 300 MHz

* CH2Ct2 (

a A A - J \ 376 377 OH H,CO'

415»HCI BF-VII-218, CDCI3l 75.5 MHz 378 OAc

.SES

418 BF-VIM31-C, CDC)3> 300 MHz 379

2.5 ppm 1

OAc

^.SES

418 BF-VII-131-C, CDCI3l 75.5 MHz 380 OAc

H,CO NO- SES

419 BF-VHI-52-E, CDCI3, 300 MHz * impurity 381

s.5 5.0 4.5 4.0 3.0 2.0

i I .? 178.3M ^ 170 . lf lf i BF-VIII-52-E,CDCI3l MHz 75.5 no 2 * impurity* 419 SES OAc

Y vm

• • I- 382 427 BF-VII-176-A, CDCI3i 300 MHz * ch2ci2 383 OAc

EtO SES

427 BF-VII-176-A, CDCI3, 75.5 MHz 384 OAc

SES

437 BF-VII-154-C, COCI3, 300 MHz

J.O 6.56.0 5.5 2 .5 PPM CDcn 208.029 BF-VII-154-C,CDCI 437 OAc SES 3 ,75.5 MHz

to.isa h3c s \ ,s e s

438 BF-VII-183-B, CDC!3, 300 MHz

I) I UU1UUJUUI j I

,-r"r f '■ ' 1 ' S' 1 I r ■(■ * ■ ■ r - i ’-v ' ^ * T1 6 .5 5 .5 5 0 4.5 4.0 3.5 3.0 S.5 2.0 1.5 PPM PPM BF-VII-183-B,CDCI 438 3

, 75.5MHz ,

' T l'T T B ESI 61 61 ESI OAc

H' ,SES

439 BF-VII-159, CDCI 3, 300 MHz

/ M

I ...... I ’ • • i 1 1 1 1 i ir^- 389 7.S 7.0 6.5 6.0 5,5 5.0 4.5

I OAc

.SES

y *

439 BF-VII-159, CDCI3 i 75.5 MHz 440 BF-VII-166,336 K, DMSO-d6, 300 MHz

// / / / i f II i JUc

AH J I 391 ' i ' 1 1 ' r 1 1 1"" r , « ' j i F ■i i L. i . . . . r . r t 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 PPH OAc

SES

440 BF-VII-166, 338 K, DMSO-d6, 75.5 MHz 441 BF-VII-242, 338 K, DMSO-dg, 300 MHz 393 OH

SES

441 BF-VII-242, 338 K, DMSO-d6, 75.5 MHz 394 I f v s L w ^ J i' LwW OH

SES

442 BF-VII-252-B, DMSOd6, 300 MHz, 343 K *H 20

T 395

! ■- ■ . r | . f " . 1 T 1 1 T 1 ■ 1 1 1 ■ ■ | ■ » i * t * i 1 ■ | ■ i ■ i | > i f- t-j- 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 PPH OAc

444 BF-VII-282'B, CDCI3, 300 MHz

• » ■ ■ ■ T~ r ’ 1 * i~' i 7.5 7.0 6.5 6.0 5.5 5.0 4.5 I

PPM BF-VII-282-B, CDCI3,MHzBF-VII-282-B, 75.5 444

53.446 I

SES 398

5.5 PPM 6.5 445 BF-VII-300, CDCI3, 75.5 MHz 399 Mi 448 BF-VI1-304-B, CDCI3, 300 MHz

* c h 2ci2 400 OH

SES

448 BF-VII-304-B, CDCI3, 75.5 MHz 401 o o CO

2 - (/) *7 uJ I 449 BF-VII-304-C, CDCI3l 75.5 MHz 403 m a m w i OH

446 BF-Vlll-3, C6D6, 300 MHz 404 *****

446 BF-VIII-3. C6D6, 75.5 MHz

m MWir a w * 405 450 BF-Vlll-20, C6D6, 300 MHz i

j k j K 406 f

S ZQ?,P57

T n .e o 408

5 m PPM

m i

CO 10 CO

LL n

451 BF-VIII-22-B, C6D6l 75.5 MHz 409 OH

454 BF-Vltl-45-B, C6D6. 300 MHz

r ' V a . U) u i

...... I I " 1 1 p 1' ” ~r 7.5 7.0 6'5 6.0 5.5 5.0 4.5 4.0 PPM r n-m r

454 BF-VIII-45-B, C6D6l 75.5 MHz

‘ * * < m |< tyU. APPENDIX B: X-RAY CRYSTALLOGRAPHIC DATA

412 413

Table 3. Bond Lengths for 151.

Atom Atom Distance Atom Atom D istance 0(1 A) C(2A) 1.227(4) C(16A) C(17A) 1.541(5) 0(2A) C(6A) 1.220(5) C(18A) C(19A) 1.374(5) 0(3A) C(21 A) 1.371(4) C(18A) C(23A) 1.378(5) 0(3A) C(24A) 1.418(5) C(19A) C(20A) 1.395(5) N(1A) C(2A) 1.355(5) C(20A) C(21A) 1.373(5) N(1 A) C(3A) 1.479(5) C(21A) C(22A) 1.387(5) N(1A) C(18A) 1.443(4) C(22A) C(23A) 1.380(5) N(2A) C(9A) 1.482(4) 0(1 B) C(2B) 1.225(4) N(2A) C(10A) 1.473(4) 0(2B) C(6B) 1.216(5) N(2A) C(16A) 1.482(4) Q(3B) C(21B) 1.386(5) C(1A) C(2A) 1.533(5) 0(3B) C(24B) 1.413(5) 0(1 A) C(5A) 1.540(5) N(1B) C(2B) 1.359(5) C(1A) C(9A) 1.538(5) N(1B) C(3B) 1.474(5) C(1A) C(17A) 1.565(5) N(1B) C(18B) 1.445(5) C(3A) C(4A) 1.519(5) N(2B) C(9B) 1.486(4) C(4A) C(5A) 1.521(5) N(2B) C(1 OB) 1.462(4) C(5A) C(6A) 1.515(5) N(2B) C(16B) 1.492(4) C(6A) C(7A) 1.462(5) 0(1 B) C(2B) 1.532(5) C(7A) C(8A) 1.327(5) C(1 B) C(5B) 1.541(5) C(8A) C(9A) 1.496(5) C(1B) C(9B) 1.544(5) C(10A) C(11 A) 1.541(5) C(1B) C(17B) 1.564(5) C(11A) C(12A) 1.526(6) C(3B) C(4B) 1.515(5) C(12A) C(13A) 1.531(6) C(4B) C(5B) 1.525(5) C(13A) C(14A) 1.505(6) C(5B) C(6B) 1.516(5) C(14A) C(15A) 1.314(6) C(6B) C(7B) 1.460(5) C(15A) C(16A) 1.505(5) C(7B) C(8B) 1.328(5) C(8B) C(9B) 1.494(5) C(10B) C(1 IB) 1.544(5) C(11B) C(12B) 1.535(6) C(12B) C(13B) 1.516(6) C(13B) C(14B) 1.513(6) C(14B) C(15B) 1.322(6) C(15B) C(16B) 1.507(5) C(16B) C(17B) 1.536(5) C(18B) C(19B) 1.378(5) C(18B) C(23B) 1.363(5) C(19B) C(20B) 1.382(5) C(20B) C(21B) 1.376(5) C(21B) C(22B) 1.372(6) C(22B) C(23B) 1.394(5)

Distances are in angstroms. Estimated standard deviations in the least significant figure are given in parentheses. 414

Table 4. Bond Angles for 151.

Atom Atom Atom A n g le Atom Atom Atom A n g le C(21 A) 0(3A) C(24A) 117.7(3) N(2A) C(9A) 0(1 A) 102.8(3) C(2A) N(1A) C(3A) 126.4(3) N(2A) C(9A) C(8A) 114.2(3) C(2A) N(1A) C(18A) 118.5(3) 0(1 A) C(9A) C(8A) 114.5(3) C(3A) N(1A) C(18A) 114.9(3) N(2A) C(10A) C(11A) 112.7(3) C{9A) N(2A) C(10A) 112.4(3) C(10A) C(11A) C(12A) 116.0(4) C(9A) N(2A) C(16A) 103.3(3) C(11A) C(12A) C(13A) 116.3(4) C(10A) N(2A) C(16A) 115.7(3) C(12A) C(13A) C(14A) 114.9(4) C(2A) C(1A) C(5A) 112.2(3) C(13A) C(14A) C(15A) 125.1(5) C(2A) C(1A) C(9A) 110.4(3) C(14A) C(15A) C(16A) 124.4(4 C(2A) 0(1 A) C(17A) 107.3(3) N(2A) C(16A) C(15A) 113.8(3) C(5A) 0(1 A) C(9A) 112.2(3) N(2A) C(16A) C(17A) 101.6(3) C(5A) 0(1 A) C(17A) 111.8(3) C(15A) 0(16) 0(17A) 113.8(4) C(9A) C(1A) C(17A) 102.3(3) 0(1 A) C(17A) C(16A) 106.1(3) 0(1 A) C(2A) N(1 A) 120.9(4) N(1A) C(18A) C(19A) 119.8(4) 0(1 A) C(2A) C91 A) 120.5(4) N(1 A) C(18A) C(23A) 120.6(4) N(1A) C(2A) 0(1 A) 118.4(4) C(19A) C(18A) C(23A) 119.6(4) N(1A) C(3A) C(4A) 112.7(3) C(18A) C(19A) C(20A) 120.5(4) C(3A) C(4A) C(5A) 110.4(3) C(19A) C(20A) C(21A) 119.7(4) 0(1 A) C(5A) C(4A) 110.1(3) 0(3A) C(21 A) C(20A) 125.1(4) C(1A) C(5A) C(6A) 112.4(3) 0(3A) C(21A) C(22A) 115.1(4) C(4A) C(5A) C(6A) 112.8(3) C(20A) C(21 A) C(22A) 119.8(4) 0(2A) C(6A) C(5A) 122.0(4) C(21A) C(22A) C(23A) 120.2(4) 0(2A) C(6A) C(7A) 120.4(4) C(18A) C(23A) C(22A) 120.2(4) C(5A) C(6A) C(7A0 117.4(4) C(21B) 0(3B) C(24B) 117.3(4) C(6A) C(7A) C(8A) 121.7(4) C(2B) N(1B) C(3B) 126.8(3) C(7A) C(8A) C(9A) 124.2(4) C(2B) N(1B) C(18B) 118.1(3) C(3B) N(1B) C(18B) 114.8(3) N(2B) C(10B) C(11B) 112.7(3) C(9B) N(2B) C(1 OB) 111.6(3) C(10B) C(11B) C(12B) 116.1(4) C(9B) N(2B) C(16B) 103.2(3) C(11B) C(12B) C(13B) 116.3(4) C(10B) N(2B) C(16B) 115.7(3) C(12B) C(13B) C(14B) 114.7(4) C(2B) 0(1 B) C(5B) 111.6(3) C(13B) C(14B) C(15B) 125.7(4) C(2B) 0(1 B) C(9B) 110.3(3) C(14B) C(15B) C(16B) 123.9(4) C(2B) 0(1 B) C(17B) 108.3(3) N(2B) C(16B) C(15B) 113.8(3) C(5B) C(1B) C(9B) 111.2(3) N(2B) C(16B) C(17B) 101.3(3) C(5B) 0(1 B) C(17B) 112.3(3) C(15B) C(16B) C(17B) 114.0(3) C(9B) C(1B) C(17B) 102.8(3) 0(1 B) C(17B) C(16B) 106.2(3) 0(1B) C(2B) N(1B) 121.4(4) N(1B) C(18B) C(19B) 120.4(4) 0(1B) C(2B) 0(1 B) 120.7(4) N(1B) C(18B) C(23B) 120.5(4) N(1B) C(2B) 0(1 B) 117.8(4) C(19B) C(18B) C(23B) 119.1(4) N(1B) C(3B) C(4B) 113.1(4) C(18B) 0(19B) C(20B) 120.4(4) C(3B) C(4B) C(5B) 110.7(3) C(19B) C(20B) C(21B) 119.6(4) 0(1 B) C(5B) C(4B) 109.4(3) 0(3B) C(21B) C(20B) 115.0(4) C(1B) C(5B) C(6B) 113.1(4) C(20B) C(21B) C(22B) 120.9(4) 0(2B) C(6B) C(5B) 122.8(4) C(21B) C(22B) C(23B) 118.4(4) 0(2B) C(6B) C(7B) 120.8(4) C(18B) C(23B) C(22B) 121.6(4) C(5B) C(6B) C(7B) 116.3(4) C(6B) C(7B) C(8B) 121.8(4) C(7B) C(8B) C(9B) 124.0(4) N(2B) C(9B) C(1B) 102.8(3) N(2B) C(9B) C(8B) 114.2(3) C(1B) C(8B) C(8B) 115.0(3)

Angles are in degrees. Estimated standard deviations in the least significant figure are given in parentheses.