Studies Toward the Asymmetric Total Synthesis of Mitomycin C

STUDIES TOWARD THE ASYMMETRIC TOTAL SYNTHESIS OF MITOMYCIN C

DISSERTATION

Presented in Partial Fulfillment of the Requirement for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Wei Chen, B.S.

* * * * *

The Ohio State University 2003

Dissertation Committee: Approved by Professor Robert S. Coleman, Adviser

Professor David J. Hart  Adviser Professor Leo A. Paquette Department of Chemistry

ABSTRACT

The mitomycins are a structurally unique class of naturally occurring compounds. Nearly all members of mitomycins family have shown broad spectrum antibiotic and potent antitumor activity against tumors resistant to other antineoplastic agents. One particular member of the family, mitomycin C, has been used clinically for cancer chemotherapy because of its broad spectrum activity against solid tumors.

In this dissertation, a novel stereoselective approach to the ring system of the mitomycins is described. The synthesis was based on a convergent strategy involving a stereocontrolled addition of a phenyl silyl enol ether to a pyrroline N-acyliminium ion followed by an intramolecular palladium-catalyzed aryl triflate amination to afford the (9R*,9aR)-tetrahydropyrrolo[1,2-a]indole ring system.

Based on similar strategy, an enantioselective synthesis of the fully functionalized mitosanes and mitosenes were achieved with high efficiency. An approach toward the isomitomycin system and efforts to introduce the key C9a methoxy group are also described.

Dedicated to my family

iii

ACKNOWLEDGMENTS

I wish to thank my adviser, Professor Robert S. Coleman, for his patience,

guidance, encouragement, intellectual support and enthusiasm that made this thesis

possible, and for the enormous amount of time he spent on correcting both my stylistic and scientific errors.

I thank Professor David J. Hart and Professor Leo A. Paquette for spending time to serve on my committee in their busy schedule.

I would also like to thank my friends, most of which have left OSU to move on to their new careers, for their helps in both my life and my scientific career. Their friendship highlighted my six years at The Ohio State University.

I am grateful to my groupmates, in particular Dr. Ronnie Perez, Dr. Thomas

Richardson, Dr. Jason McCary, Dr. Srinivas Gurrala, Jason Guernon, Jossian

Oppenheimer, Ruhul Garg and many names too long to be listed, who have contributed time, experience and advice in my research.

Finally I would like to thank my family for their endless love and support.

Without them, none of my achievements would have been possible.

VITA

December 13, 1975………………………… Born – Jingdezhen, China

1997 ……………………………………… B. S. Chemistry, Peking University

1997- present Graduate Teaching and Research Associate, The Ohio State University

PUBLICATIONS

Research Publication

1. Coleman, R. S.; Chen, W. “A Convergent Approach to the Mitomycin Ring System,” Org. Lett. 2001, 3, 1141-1144.

FIELDS OF STUDY

Major Field: Chemistry

TABLE OF CONTENTS

Page

Abstract...... ii

Dedication...... iii

Acknowledgment...... iv

Vita...... v

List of Tables...... vi

List of Figures...... ix

Chapters:

1. The Chemistry and Biology of Mitomycins...... 1

1.1 Introduction...... 1 1.2 Isolation and Structure Determination of the Mitomycins and Structurally Related Natural Products...... 3 1.3 The Chemistry of Mitomycins...... 7 1.4 The Biology of Mitomycins...... 11

2. Previous Synthetic Studies...... 20

2.1 Kishi’s Total Syntheses of Mitomycins...... 24 2.2 Fukuyama’s Total Syntheses of Mitomycins...... 26 2.3 Danishefsky’s Total Syntheses of Mitomycin K and FR-900482. . . 29 2.4 Jimenez’s Total Synthesis of mitomycin K...... 35 2.5 Fukuyama’s Total Syntheses of FR-900482...... 37 2.6 Terashima’s Total Synthesis of (+)-FR-900482...... 44 2.7 Martin’s Formal Total Synthesis of FR-900482...... 47 2.8 Williams’ Total Synthesis of (+)-FR-900482 and (+)-FR-66979. . . 50 2.9 Ciufolini’s Total Synthesis of FR-66979...... 53 2.10 Cha’s Approach to Mitomycin System...... 56

2.11 Sulikowski’s Approach to Mitosenes...... 60 2.12 Tandem Radical Cyclization Approaches...... 61 2.13 Vedejs’s Approaches to Aziridinomitosene...... 64 2.14 Michael’s Enantioselective Synthesis of Aziridinomitosene. . . . . 69 2.15 Jones’ Stereoselective Synthesis of Mitosane...... 71 2.16 Miller’s Enantioselective Synthesis of Mitomycin Core Structure. 73

3. An Overview of the Synthetic Strategy...... 76

4. A Convergent Synthesis of Mitomycin Ring System...... 79

4.1 Retrosynthetic Analysis...... 79 4.2 Synthesis of N-Acyl-2-hydroxy-pyrrolidine and Allylstannane. . . 80 4.3 Lewis Acid Catalyzed Addition of Allylstannane to Iminium Ion. 84 4.4 Synthesis of Silyl Enol Ether and Its Addition to Iminium Ion. . . . 86 4.5 Mitosane Ring Cyclization...... 89 4.6 The Determination of The Relative Stereochemistry...... 94 4.7 Rationale for the Observed Diastereoselection...... 98 4.8 Experimental...... 100

5. Stereoselective Synthesis of Mitosanes and Mitosenes...... 109

5.1 Retrosynthetic Analysis...... 109 5.2 Synthesis of the Iminium Ion Precursor...... 114 5.3 Synthesis of Silyl Enol Ether Precursor and Attempted Coupling Reaction...... 122 5.4 Synthesis of Cinnamyl Stananne and Its Coupling Reaction. . . . . 126 5.5 Mitosane B Ring Cyclozation...... 131 5.6 Unexpected Mitosane Conversion to Mitosene...... 139 5.7 Conclusion...... 142 5.8 Experimental...... 143

6. Synthetic Studies Toward Isomitomycins...... 162

6.1 Introduction...... 162 6.2 Efforts of Aminohydroxy Functionalities Installation From Mitosane Intermediates...... 167 6.3 Installation of trans Aminohydroxy Functionalities by Asymmetric Sharpless Aminohydroxylation...... 173 6.4 Lewis Acid Catalyzed Addition of Stannane to Iminium Ion. . . . . 178 6.5 Studies Toward the Installation of Isomitomycin Ring...... 180 6.6 Conclusion...... 186 6.7 Experimental...... 187

vii

7. Studies Toward Oxidative Installation of Methoxy Moiety 195

7.1 Introduction...... 195 7.2 Intramolecular Oxidative Cyclization...... 198 7.3 Conclusion...... 203 7.4 Experimental ...... 204

References...... 208

Appendix: Selected NMR Spectra...... 223

viii

LIST OF FIGURES

Figure Page

1.1 The Mitomycin Family...... 5

1.2 FR-900482 and FR-66979...... 6

1.3 Interconversion of Mitomycins...... 8

1.4 Mitomycin Solvolysis in Dilute HCl...... 9

1.5 Reductive Activation of Mitomycin/DNA Cross-Linking...... 15

1.6 Disulfide Substituted Analogs of Mitomycin C...... 19

2.1 Strategies toward Construction of Mitomycin Ring System...... 23

2.2 Kishi’s Total Synthesis of Mitomycin A...... 25

2.3 Interconversion between Mitomycin A and Isomitomycin A...... 26

2.4 Fukuyama’s Total Synthesis of Mitomycin C...... 28

2.5 Danishefsky’s Plan to Construct FR-900482 Ring System ...... 30

2.6 Further Studies Toward FR-900482 by Danishefsky...... 31

2.7 Danishefsky’s Total Synthesis of Mitomycin K...... 32

2.8 Danishefsky’s Total Synthesis of FR-900482...... 34

2.9 Jimenez’s Total Synthesis of Mitomycin K...... 36

2.10 Retrosynthetic Analysis of Fukuyama’s FR-900482 Syntheses. . . . 38

2.11 Fukuyama’s 1992 Total Synthesis of FR-900482...... 40

2.12 Fukuyama’s 2002 Total Synthesis of (+)-FR-900482...... 43

2.13 Retrosynthetic Analysis of Terashima’s Total Synthesis of (+)-FR- 900482...... 46

2.14 Martin’s Formal Total Synthesis of FR-900482...... 49

2.15 Williams’ Total Synthesis of (+)-FR-900482...... 52

2.16 Ciufonili’s Approach to Benzazocene...... 53

2.17 Ciufolini’s Total Synthesis of FR-99674...... 55

2.18 Reactions of Dialkoxytitanacyclopropane...... 57

2.19 Cha’s Approach to the Mitomycin Ring System...... 59

2.20 Sulikowski’s Approach to Mitosenes...... 60

2.21 Ziegler’s Approach to 9a-Desmethoxymitomycins...... 62

2.22 Jones’ and Parsons’ Approaches...... 63

2.23 Vedejs’ First Approach to Aziridinomitosene...... 65

2.24 Vedejs’ Latter Approach to Aziridinomitosene...... 68

2.25 Michael’s Enantioselective Synthesis of Aziridinomitosene...... 70

2.26 Jones’ Stereocontrolled Synthesis of Mitosane...... 72

2.27 Miller’s Enantioselective Synthesis of Mitomycin Core Structure. . 75

3.1 Retosynthetic Analysis...... 77

4.1 Retrosynthetic Plan...... 80

4.2 Syntheses of N-Acyl-2-hydroxypyrrolidines...... 81

4.3 Synthesis of Allylstannane...... 83

4.4 Lewis Acid Catalyzed Addition Reaction of Allylsilane...... 84

4.5 Allylstannane Addition and Attempted Cyclization...... 86

4.6 Synthesis of Silyl Enol Ether and Its Addition Reaction...... 88

4.7 Manipulations of Protecting Groups...... 91

4.8 Selective O- vs. N-Triflation...... 92

4.9 Palladium-Catalyzed Cyclization...... 93

4.10 Completion of the Convergent Approach to Mitomycin Ring System...... 94

4.11 Illustrations of Important Observed NOE Effect...... 96

4.12 Correlations between Observed NOE Effect and Calculated Proton-Distances...... 97

4.13 Origin of Diastereoselectivity...... 99

5.1 Retrosynthetic Analysis of Mitosanes...... 110

5.2 Modifications of Synthetic Strategy...... 113

5.3 Synthesis of Iminium Ion Precursor from D-Ribose...... 115

5.4 Transformations from Primary Alcohol to Carbamate Protected Amine...... 116

5.5 Attempted Debenzylation...... 118

5.6 Use of The Allyl Protecting Group...... 120

5.7 Pyrrolidine Ring Formation...... 121

5.8 Synthesis of Iminium Ion Precursor...... 122

5.9 Functional Group Installations on the Aromatic Ring...... 124

5.10 Synthesis of Silyl Enol Ether...... 125

5.11 Failed Coupling Reaction...... 126

5.12 Synthesis of Cinnamyl Stannane...... 127

5.13 Stannane Coupling with Pyrrolidine Iminium Ion...... 127

5.14 Synthesis of Bis-dinitrobenzoic Ester...... 129

5.15 X-Ray Crystal Structure of 171...... 130

5.16 Cleavage of the Terminal Double Bond...... 133

5.17 Ozone Cleavage of the Terminal Double Bond...... 134

5.18 Synthesis of Aminophenol...... 135

5.19 Proposed B Ring Cyclization...... 136

5.20 Oxidative Cyclization...... 138

5.21 Unexpected Mitosane Conversion to Mitosene...... 141

6.1 Attempted C9a Methoxy Installation in Literature...... 163

6.2 Interconversion between Mitomycin A and Isomitomycin A...... 164

6.3 Retrosynthetic Analysis...... 166

6.4 Attempted Aminoalcohol Installation on Alkene 169...... 168

6.5 Attempted Aminoalcohol Installation on Silyl Ether 176...... 170

6.6 Retrosynthetic Analysis...... 173

6.7 Synthesis of the Sharpless Aminohydroxylation Precursor...... 174

6.8 Initial Effort for Sharpless Asymmetric Aminohydroxylation. . . . . 175

6.9 Sharpless Asymmetric Aminohydroxylation...... 176

6.10 Use of p-Methoxybenzyl Protecting Group...... 178

6.11 Reductive Cylization...... 179

6.12 BF3⋅Et2O Catalyzed Coupling Reaction...... 180

xii

6.13 Ozone Cleavage of the Terminal Double Bond...... 181

6.14 Aziridine Formation...... 182

6.15 Functional Group Manipulations...... 183

6.16 Possible Formation of Isomitomycin Ring...... 185

7.1 Nitrogen Assisted Methanol Elimination...... 196

7.2 Acid Stable 2-Alkoxypyrrolidine...... 197

7.3 Intramolecular Oxidative Cyclization...... 197

7.4 Synthesis of Model Study Precursor...... 197

7.5 Intramolecular Oxidative Cyclization on the Model System...... 201

7.6 Other Model Systems Examined...... 202

xiii

CHAPTER 1

THE CHEMISTRY AND BIOLOGY OF MITOMYCINS

1.1 Introduction

The mitomycins are a structurally unique class of naturally occurring

compounds first isolated in the late 1950’s by Japanese microbiologists from

fermentation cultures of the microorganism Streptomyces caespitosus.1 Nearly all members of this family have shown broad spectrum antibiotic and potent antitumor activity against tumors resistant to other antineoplastic agents. One particular member of the family, mitomycin C, has been used clinically for cancer chemotherapy since the 1960’s because of its broad spectrum activity against solid tumors. Mitomycin C remains an important component of combination chemotherapy of breast, lung, and

prostate cancer; it is among the few drugs effective against colorectal cancer and it is

the drug of choice for intravesical administration in superficial bladder

1. Hata, T.; Sano, Y.; Sugawara, R.; Matsumae, A.; Kanamorei, K.; Shima, T.; Hoshi, T. “Mitomycin, A New Antibiotic from Streptomyces,” J. Antibiot. Ser. A 1956, 9, 141-146; Wakaki, S.; Marumo, H.; Tomioka, K.; Shimizu, G.; Kato, E.; Kamda, H.; Kudo, S.; Fujimoto, Y. “Isolation of New Fractions of Antitumor Mitomycins,” Antibiot. Chemother. 1958, 8, 228-240. 1

cancer.2 Mitomycin C is the single most active agent for the treatment of non-small cell lung cancer.3

In addition to its antitumor activity, mitomycin C has a variety of specific biological effects in mammalian cells or microorganisms, including selective inhibition of DNA synthesis, recombination, chromosome breakage, sister chromatid exchange, and induction of DNA repair (SOS response) in bacteria.4 More recently,

several interesting effects of mitomycin C on mammalian gene expression have been

reported: induction of transcription of DT-diaphorase in human colon tumors,5 induction or suppression of glutethimide inducible genes in chick embryos,6 suppression of P-glycoprotein expression and multidrug resistance in human KB cancer cells,7 and induction of the human Y-box-binding protein and consequent

2. Carter, S. K.; Crooke, S. T. “Mitomycin C: Current Status and New Developments,” Academic Press, 1979.

3. Spain, R. C. “The Case for Mitomycin in Non-Small Cell Lung Cancer,” Oncology, 1993, 50 (Suppl. 1), 35-52.

4. Szybalski, W.; Iyer, V. N. “Cross-Linking of DNA by Enzymatically or Chemically Activated Mitomycins and Porfiromycins, Bifunctionally ‘Alkylating’ Antibiotics,” Fed. Proc. 1964, 23, 946-957; Carrano, A. V.; Thompson, L. H.; Stretka, D.G.; Minkler, J. L.; Mazrimas, J. A.; Fong, S. “DNA Crosslinking, Sister Chromatic Exchange and Specific Locus Mutations,” Mutat. Res. 1979, 63, 175-188.

5. Yao, K.-S.; Godwin, A. K.; Johnson, C.; and O’Dwyer, P. J. “Alternative Splicing and Differential Expression of DT-Diaphorase Transcipts in Human Colon Tumors and in Peripheral Mononuclear Cells in Response to Mitomycin C Treatment,” Cancer Res. 1996, 56, 1731-1736.

6. Caron, R. M.; Hamilton, J. W. “Preferential Effects of the Chemotherapeutic DNA Cross-Linking Agent Mitomycin C on Inducible Gene Expression in vitro Environ,” Mol. Mutagen. 1995, 25, 4- 11.

7. Ihnat, M.; Lavrivriere, J. P.; Warren, A. J.; La Ronde, N.; Blaxall, J. R.; Pierrre K. M.; Turpie, B. W.; Hamilton, J. W. “Suppression of P-Glycoprotein Expression and Mutidrug Resistance by DNA Cross-Linking Agents,” Clin. Cancer Res. 1997, 3, 1339-1346. 2

regulation of the multidrug gene resistance gene 1.8 These observations highlight the

possible new applications of mitomycin C in cancer chemotherapy.

1.2 Isolation and Structure Determination of the Mitomycins and

Structurally Related Natural Products

In 1956, mitomycin A (1) and B (2) were isolated from Streptomyces caespitosus by Hata and co-workers at Kitasato Institute in Japan and were found to possess potent antitumor and antibiotic activities.9 Two years later, Wakagi and co-

workers in the same institute reported the isolation of mitomycin C (3) from the same

fermentation broth at a higher pH, a related compound of superior antitumor activity.

Immediately following this discovery, DeBoer of Upjohn Company isolated the N-

methylmitomycin C,10 also called mitomycin D or porfiromycin (4) from a different species, Streptomyces ardus. In 1981, researchers in Japan discovered a new type of mitomycin, 10-decarbamoyloxy-9-dehydromitomycin B (5),11 which was later named

8. Ohga, T.; Koike, K.; Ono, M.; Makino, Y.; Itagaki, Y.; Tanimoto, M.; Kuwano, M.; Kohno, K.; “Role of the Human Y-Box-Binding Protein YB-1 in Cellular Sensitivity to the DNA Damaging Agents Cisplatin, Mitomycin C, and Ultraviolet Light,” Cancer Res. 1996, 56, 4224-4228.

9. Hata, T.; Sano, Y.; Sugawara, R.; Matsumae, A.; Kanamorei, K.; Shima, T.; Hoshi, T. “Mitomycin, A New Antibiotic from Streptomyces,” J. Antibiot. Ser. A 1956, 9, 141-146.

10. DeBoer, C.; Dietz, A.; Lummus, N. E.; Savage, G. M. “Porfiromycin, a New Antibiotic. I. Discovery and Biological Activities,” Antimicrob. Agents Ann. 1961, Vol. Date 1960, 17-22.

11. Urakawa, C.; Tsuchiya, H.; Nakano, K.-I. “New Mitomycin, 10-Decarbamoyloxy-9-dehydromitomycin B from Streptomyces Caespitosus,” J. Antibiot. 1981, 34, 243-244. 3

as mitomycin H. There are currently 17 mitomycins known (Figure 1.1). Only mitroromycin (6) is reported to possess no biological activity.12

12. Lefemine, D. V.; Dann, M.; Barbatschi, F.;Hausmann, W. K.; Zbinovsky, V.; Monnikendum, J.; Bohonos, N. J. “Isolation and Characterization Mitiromycin and other Antibiotics,” J. Am. Chem. Soc. 1962, 84, 3184-3185. 4

10 O OCONH2 O OCONH2 X 8a 9 OMe MeO OH 7 9a 6 1 4a NNY NNMe 4 O 3 O

X Y mitomycin A (1) MeO H mitomycin C (3) NH2 H mitomycin B (2) porfiromycin (4) NH2 Me

O O O O MeO OMe MeO NH NNMe NNMe O O

mitomycin K (5) mitiromycin (6)

Figure 1.1 The Mitomycin Family

Structurally related compounds FR-900482 (7) and FR-66979 (8) were recently

isolated from Streptomyces sandaensis No. 6897 by the Fujisawa Pharmaceutical

Co.13 Both compounds have been shown to exibit potent antitumor activities. FR-

900482 exists as a 2:1 mixture of tautomers (7a and 7b) that interconvert presumably

via intermediate keto form 9 (Figure 1.2).

OCONH OH 2 OH FR-900482 R=CHO (7) FR-66979 R=CH OH (8) O NH 2 RN

OCONH2 OCONH O OCONH OH 2 OH OH 2 OH OH NH O NH O NH OHC N OHC N OHC N HO 7a 9 7b

Figure 1.2 FR-900482 and FR-66979

The assignment of the absolute configuration of mitomycins has undergone

several reversals. The correct assignment was finally established in 1983, twenty-

seven years after the discovery of mitomycins, by Hirayama and co-workers based on

13. Iwami, M.; Kiyoto, S.; Terano, H.; Kohsaka, M.; Aoki, H.; Imanaka, H. “A New Antitumor Antibiotic, FR-900482. I. Taxonomic Studies on the Producing Strain: A New Species of the Genus Streptomyces,” J. Antibiot. 1987, 40, 589-593.

the analysis of X-ray crystallographic data of 2-N-(p-bromobenzoyl)mitomycin A.14

This assignment was coincident with the discovery of the absolute configuration by

Fukuyama and Yang in their total synthesis of mitomycin C.15

1.3 The Chemistry of Mitomycins

Mitomycins can undergo interconvertion,16,17 and this property greatly facilitates

the syntheses of mitomycins. For example, mitomycin A (1) can be converted to

mitomycin C (3) by treatment with methanolic ammonia. Hydrolysis of mitomycin C

followed by methylation produces mitomycin A. Mitomycin G can be formed by treatment of mitomycin A with DBU. (Figure 1.3)

14. Shirahata, K.; Hirayama, N. “Revised Absolute Configuration of Mitomycin C. X-Ray Analysis of 1-N-(p-bromobenzoyl)mitomycin C,” J. Am. Chem. Soc. 1983, 105, 7199-7200.

15. Fukuyama, T.; Yang, L. “Practical Total Synthesis of (±)-Mitomycin C,” J. Am. Chem. Soc. 1989, 111, 8303-8304.

16. Remers, W. A.; Dorr, R. T. “Alkaloids: Chemical and Biological Perspectives,” Relletier, S. W. Ed. Wiley, New York, 1988, 1-67.

17. Remers, W. A. “The Chemistry of Antitumor Antibiotics,” Vol. 1, Wiley, New York, 1979. 7

O OCONH2 O MeO MeO OMe DBU OMe NNH NNH O O mitomycin A (1) mitomycin G

1) OH- NH3/MeOH 2) CH2N2

O OCONH2 O OCONH2 H2N H2N OMe OMe MeI/K2CO3 NNH NNMe O O mitomycin C (3) porfiromycin (4)

Figure 1.3 Interconversion of Mitomycins

Mitomycins are relatively stable to basic conditions in contrast to their reactivity under acidic conditions. Mitomycins undergo facile elimination of methanol under mildly acidic conditions to give an unstable aziridinomitosene, which reacts further to yield solvolyzed products.18,19 The major products of hydrolysis in

dilute aqueous acid (0.05−0.1 N HCl) were cis- and trans-1-hydroxy-2-amino

mitosenes. The regio- and stereochemistry of these 1,2-disubstituted mitosenes were

determined. Without exception the aziridine opens with retention of configuration

18. Remers, W. A.; Dorr, R. T. “Alkaloids: Chemical and Biological Perspectives,” Relletier, S. W. Ed. Wiley, New York, 1988, 1-67.

19. Remers, W. A. “The Chemistry of Antitumor Antibiotics,” Vol. 1, Wiley, New York, 1979. 8

incorporating oxygen at C1 and leaving nitrogen at C2. In all aqueous acid hydrolyses cases the cis-aminohydrin predominates.20 Exceptions were found in the methanol/acetic acid or Dowex/methanol solvolysis of mitomycin C, the major product being the trans-methoxy amine.21 (Figure 1.4)

O OCONH2 MeO OMe

NNMe O

0.05 N HCl

O OCONH2 O OCONH2 MeO MeO OH + OH N N O NHMe O NHMe

Figure 1.4 Mitomycin Solvolysis in Dilute HCl22

20. Taylor, W. G.; Remers, W. A. “Structure and Stereochemistry of Some 1,2-Disubstituted Mitosenes from Solvolysis of Mitomycin C and Mitomycin A,” J. Med. Chem. 1975, 18, 307-311.

21. Cheng, L.; Remers, W. A. “Comparative Stereochemistry in the Aziridine Ring Openings of N- methylmitomycin A and 7-Methoxy-1,2-(N-methylaziridino)mitosene,” J. Med. Chem. 1977, 20, 138-141.

22. Because of the wrong assignment of the absolute configuration of mitomycins at the time the research was conducted, the stereochemistry in reference 21 is opposite to that shown in this figure. 9

On prolonged exposure to dilute acid, the C7 substituent also suffers hydrolysis to give the 7-hydroxyquinone. Harshly acidic conditions are required for carbamate hydrolysis.

Historically the chemistry associated with the reduction of mitomycins is a result of the mechanistic investigations of their antitumor activity. These investigations provided crucial information for constructing the delicate functionalities that helped to develop synthetic strategy. The deactivation of the nucleophilic character of the B ring nitrogen provided by the vinylogous amide-like resonance with the quinone and is critical for the maintainance of the angular methoxy or hydroxy group. Once the quinone is reduced to dihydroquinone, the sensitive angular methoxy oxygen is rapidly expelled with the assistance from the now nucleophilic nitrogen. Although mitomycin derivatives bearing C9a methoxy groups in dihydroquinone form, leucomitomycins (products of reduction of quinone of mitomycins B and C), were synthesized and characterized, little chemistry could be done on those extremely sensitive compounds.23

23. Danishefsky, S.; Ciufolini, M. “Leucomitomycins,” J. Am. Chem. Soc. 1984, 106, 6424-6425. 10

1.4 The Biology Of Mitomycins24,25

Soon after mitomycin C was discovered, studies of the molecular pharmacology revealed an extraordinary property of this class of antitumor antibiotics: mitomycin C and other members of this class were found to cross-link the complementary strands of DNA in an interstrand manner26 accompanied by observations of monoalkylation of single stranded of DNA.27 Mitomycins still remain the only known natural antibiotics that possess this ability. Azinomycins28 and

24. Tomasz, M.; Palom, Y. “The Mitomycin Bioreductive Antitumor Agents: Cross-Linking and Alkylation of DNA as the Molecular Basis of Their Activity,” Pharmacol. Ther. 1997, 76, 73-87; and references therein.

25. Tomasz, M. “Molecular Aspects of Anticancer Drug-DNA Interactions,” Neidel, S. and Waring, M. eds., 1979, Vol. 2, 312-349; and references therein.

26. Iyer, V. N.; Szybalski, W. “A Molecular Mechanism of Mitomycin Action: Linking of Complementary DNA Strands,” Proc. Natl. Acad. Sci. 1963, 50, 355-362.

27. Weissback, A.; Lisio, A. “Alkylation of Nucleic Acids by Mitomycin C and Porfiromycin,” Biochemistry, 1965, 4, 196-200; Tomasz, M.; Mercado, C. M.; Olson, J.; Chatterjie, N. “The Mode of Interaction of Mitomycin C with DNA and Polynucleotides in vivo,” Biochemistry 1974, 13, 4878-4887; Lown, J. W.; Bergleiter, A.; Jobson, D.; Morgan, A. R. “Studies Related to Antitumor Antibiotics: V. Reaction of Mitomycin C with DNA Examined by Ethidium Fluorescence Assay,” Can. J. Biochem. 1976, 54, 110-119.

28. Coleman, R. S.; Perez, R. J.; Burk, C. H.; Navarro, A. “Studies on the Mechanism of Action of Azinomycin B: Definition of Regioselectivity and Sequence Selectivity of DNA Cross-Link Formation and Clarification of the Role of the Naphthoate,” J. Am. Chem. Soc. 2002, 124, 13008- 13017; Alcaro, S.; Ortuso, F.; Coleman, R. S. “DNA Cross-Linking by Azinomycin B: Monte Carlo Simulations in the Evaluation of Sequence Selectivity,” J. Med. Chem. 2002, 45, 861-870; Hartley, J. A.; Hazrati, A.; Kelland, L. R.; Khanim, R.; Shipman, M.; Suzenet, F.; Walker, L. F. “A Synthetic Azinomycin Analog with Demonstrated DNA Cross-Linking Activity: Insights into the Mechanism of Action of this Class of Antitumor Agent,” Angew. Chem. Int. Ed. 2000, 39, 3467-3470; Zang, H.; Gates, K. S. “DNA Binding and Alkylation by the "Left Half" of Azinomycin B,” Biochemistry 2000, 39, 14968-14975; Armstrong, R. W.; Salvati, M. E.; Nguyen, M. “Novel Interstrand Cross-Links Induced by the Antitumor Antibiotic Carzinophillin/Azinomy- cin B,” J. Am. Chem. Soc. 1992, 114, 3144-3145.

bioxalomycin29 could be two possible exceptions, although in both cases the evidence for cross-linking with DNA are still preliminary and the experiments were conducted in vitro with purified or synthetic DNA. It has already been demonstrated that

mitomycins exert their antibiotic activities through the inhibition of DNA replication

resulting from mitomycin-DNA cross-linking.30

Structurally, mitomycin C contains three groups that can damage cells: the quinone that can participate in free radical reactions generating superoxide, and aziridinyl and urethane (carbamate) functions that can take part in DNA alkylation.

However, neither mitomycin C nor its derivatives are reactive toward DNA at pH 7-

8.31 Unlike many other antibiotics, mitomycin C requires in vivo activation to bind covalently to DNA.32,33 Upon enzymatic,34 metabolic,35 electrochemical,36 or

29. Williams, R. M.; Herberich, B. “DNA Interstrand Cross-Link Formation Induced by

Bioxalomycin α2,” J. Am. Chem. Soc. 1998, 120, 10272-10273.

30. Szybalski, W.; Iyer, V. N. “Cross-Linking of DNA by Enzymatically or Chemically Activated Mitomycins and Porfiromycins, Bifunctionally ‘Alkylating’ Antibiotics,” Fed. Proc. 1964, 23, 946-957.

31. Iyer, V. N.; Szybalski, W. “A Molecular Mechanism of Mitomycin Action: Linking of Complementary DNA Strands,” Proc. Natl. Acad. Sci. 1963, 50, 355-362; Kumar, G. S.; He, Q.; Behr-Ventura, D.; Tomasz, M. “Binding of 2,7-Diaminomitosene to DNA: Model for the Recognition of DNA by Activated Mitomycin C,” Biochemistry 1995, 34, 2662-2671; Rodighiero, G.; Marciani Magno, S.; Dell Acqua, F.; Vedaldi, D.; “Studies on the Mechanism of Action of Mitomycin C,” Farmaco. Ed. Sci. 1978, 3, 651-666.

32. Tomasz M.; Lipman R.; Chowdary D.; Pawlak J.; Verdine G. L.; Nakanishi K. “Isolation and Structure of A Covalent Cross-Link Adduct between Mitomycin C and DNA,” Science 1987, 235, 1204-1208.

33. For a comprehensive review of in situ activation for DNA targeting agents, see: Wolkenberg, S. E.; Boger, D. L. “Mechanisms of in situ Activation for DNA-Targeting Antitumor Agents,” Chem. Rev. 2002, 102, 2475-2495.

34. Peterson, D. M.; Fisher, J. “Autocatalytic Quinone Methide Formation from Mitomycin C,” Biochemistry, 1986, 25, 4077-4084; Iyer, V. N.; Szybalski, W. “Mitomycins and Porfiromycin: Chemical Mechanism of Activation and Cross-Linking of DNA,” Science 1964, 145, 55-58.

chemical37 reduction, a cascade of spontaneous transformations occurs and efficient

alkylation of DNA or trapping of nucleophiles is observed with DNA cross-linking

proceeding at remarkable rates (<1 min).38 Although several enzymes have been

found that reduce mitomycins, it still remains unknown with certainty which

biologically related agent is responsible for the process in vivo.

In 1964, after observing the metabolically activated mitomycin cross-linking

with DNA, Iyer and Szybalski proposed a reductive activation mechanism,39 which after decades of experimental investigation, has been verified with little modification.

This hypothesis has served as the organizing framework for the experimental inquiry into the precise molecular mode of action of the mitomycins in the following decades.

The original Iyer-Szybalski hypothesis identified the C1 aziridine and C10 carbamate as two important sites for mitomycin DNA cross-linking. Reduction of mitomycin releases the N4 lone pair of electrons from vinylogous amide resonance with the quinone, enabling the spontaneous expulsion of methoxide from C9a under

35. Tomasz, M.; Lipman, R. “Reductive Metabolism and Alkylating Activity of Mitomycin C Induced by Rat Liver Microsomes,” Biochemistry 1981, 20, 5056-5061.

36. Andrews, P. A.; Pan, S.-S.; Bachur, N. R. “Electrochemical Reductive Activation of Mitomycin C,” J. Am. Chem. Soc. 1986, 108, 4158-4166; Kohn, H.; Zein, N.; Lin, X. Q.; Ding, J. Q.; Kadish, K. M. “Mechanistic Studies on the Mode of Reaction of Mitomycin C under Catalytic and Electrochemical Reductive Conditions,” J. Am. Chem. Soc. 1987, 109, 1833-1840.

37. Hoey, B. M.; Butler, J.; Swallow, A. J. “Reductive Activation of Mitomycin C,” Biochemistry 1988, 27, 2608-2614.

38. Cera, C.; Egbertson, M.; Teng, S. P.; Crothers, D. M.; Danishefsky, S. J. “DNA Cross-Linking by Intermediates in the Mitomycin Activation Cascade,” Biochemistry 1989, 28, 5665-5669.

39. Iyer, V. N.; Szybalski, W. “Mitomycins and Porfiromycin: Chemical Mechanism of Activation and Cross-Linking of DNA,” Science 1964, 145, 55-58.

physiological conditions40 to form iminium 10 (Figure 1.5). The iminium 10 will then spontaneously rearrange to indole 11. The driving force for the whole process is the stability of the aromatic indole system. The ring strain of the “benzylic-like” aziridine is released by the subsequent conversion to the extended quinone methide

12.41 Intermediate 12 sets up the stage for mitomycin DNA cross-linking.

40. Under anhydrous condition, the elimination process is not spontaneous: Danishefsky, S.; Ciufolini, M. “Leucomitomycins,” J. Am. Chem. Soc. 1984, 106, 6424-6425.

41. Support for this intermediate is indirect. See: Tomasz, M.; Lipman, R. “Reductive Metabolism and Alkylating Activity of Mitomycin C Induced by Rat Liver Microsomes,” Biochemistry 1981, 20, 5056-5061; Bean, M.; Kohn, H. “Studies on the Reaction of Mitomycin C with Potassium Thiobenzoate under Reductive Conditions,” J. Org. Chem. 1985, 50, 293-298; Gargiulo, D.; Musser, S. S.; Yang, L.; Fukuyama, T.; Tomasz, M. “Alkylation and Crosslinking of DNA by the Unnatural Enantiomer of Mitomycin C: Mechanism of the DNA-Sequence Specificity of Mitomycins,” J. Am. Chem. Soc. 1995, 117, 9388-9398; and references cited therein. 14

O OCONH2 OH OCONH2 H2N OMe reduction H2N OMe

NNH NNH O OH

OH OCONH2 OH OCONH2 - H2N H B H2N

NNH NNH OH OH 10 11

H O OCONH2 OH OCONH2 H2N H2N DNA1 DNA1 N N

OH NH2 OH NH2 12 13

OH DNA2 OH DNA2 H2N H2N DNA1 DNA1 N N

OH NH2 OH NH2 14

Figure 1.5 Reductive Activation of Mitomycin/DNA Cross-Linking

Quinone methide 12 is highly unstable. It alkylates guanine residues in DNA at the exocyclic 2-amino positions, forming a monoalkylation adduct with DNA. The oxidized quinone form of the indole compound 13 was isolated and characterized.42

The second electrophilic center of the mitomycin is the iminium ion 14 formed by the elimination of the C10 carbamate. Carbon 10 is the other site of alkylation in mitomycin/DNA cross-link formation, forming a second bond with the

2-amino group of guanine on the complementary strand.43 The C1 alkylation always precedes reaction at C10.44 Both inter- and intrastrand cross-links are observed when mitomycin reacts with DNA, although the former predominates.45

Mitomycin alkylation is sequence specific. Guanines in CpG sequence are strongly preferred sites of the first monoalkylation step compared with other NpG dinucleotide sequences.46 Investigation of the mechanism of this 5′-d(CpG) selectivity of the monoalkylation step revealed that it is not the 5′-dC residue that enhances the reactivity of the adjacent guanine with mitomycin C, but rather the dG

42. Tomasz, M.; Lipman, R.; Verdine, G. L.; Nakanishi, K. “Reassignment of the Guanine-Binding Mode of Reduced Mitomycin C,” Biochemistry 1986, 25, 4337-4344; Tomasz, M.; Lipman, R.; McGuinness, B. F.; Nakanishi, K. “Isolation and Characterization of A Major Adduct between Mitomycin C and DNA,” J. Am. Chem. Soc. 1988, 110, 5892-5896.

43. Tomasz, M.; Lipman, R.; Chowdary, D.; Pawlak, J.; Verdine, G.L.; Nakanishi, K. “Isolation and Structure of A Covalent Cross-Link Adduct between Mitomycin C and DNA,” Science 1987, 235, 1204-1208.

44. Tomasz, M.; Chawla, A. K.; Lipman, R. “Mechanism of Monofunctional and Bifunctional Alkylation of DNA by Mitomycin C,” Biochemistry 1988, 27, 3182-3187.

45. Bizanek, R.; McGuinness, B. F.; Nakanishi, K.; Tomasz, M. “Isolation and Structure of An Intrastrand Cross-Link Adduct of Mitomycin C and DNA,” Biochemistry 1992, 31, 3084-3091.

46. Kohn, H.; Li, V.-S.; Schiltz, P.; Tang, M. S. “On the Origins of the DNA Sequence Selectivity of Mitomycin Monoalkylation Transformations,” J. Am. Chem. Soc. 1992, 114, 9218-9220; Kumar, S.; Lipman, R.; Tomasz, M. “Recognition of Specific DNA Sequences by Mitomycin C for Alkylation,” Biochemistry 1992, 31, 1399-1407. 16

residue on the opposite strand that is responsible for this effect. It is proposed that the

2-amino group of the opposite-strand guanine forms a specific H-bond with the C10 oxygen atom of the activated mitomycin C, and that this noncovalent interaction facilitates the convalent bonding by the 2-amino group of the target guanine.47

The mitomycin C interstrand cross-linking is “absolutely” specific to the 5′- d(CG) duplex sequence48 with the agent spanning the distance between the two strands across a one base-pair step within the minor groove.49 Most of the cross- linking selectivity is a result of the first alkylation step. A second contribution to the selectivity of the cross-linking stems from the preferred orientation of the covalent monoadduct in the DNA minor groove where it is poised to form the cross-link without significant structural reorganization of the DNA duplex.50

An interesting DNA sequence specificity was observed for the DNA alkylation of antitumor antibiotic FR-66979. It was demonstrated it forms DNA

47. Sastry M.; Fiala R.; Lipman R.; Tomasz M.; Patel D. J. “Solution Structure of the Monoalkylated Mitomycin C-DNA Complex,” J. Mol. Biol. 1995, 247, 338-359; Li, V.-S.; Choi, D.; Wang, Z.; Jimenez, L. S.; Tang, M.-S.; Kohn, H. “Role of the C-10 Substituent in Mitomycin C-1-DNA Bonding,” J. Am. Chem. Soc. 1996, 118, 2326-2331.

48. Borowy-Borowski, H.; Lipman, R.; Tomasz, M. “Recognition between Mitomycin C and Specific DNA Sequences for Cross-Link Formation,” Biochemistry 1990, 29, 2999-3006; Millard, J. T.; Weidner, M. F.; Raucher, S.; Hopkins, P. B. “Determination of the DNA Crosslinking Sequence Specificity of Reductively Activated Mitomycin C at Single-Nucleotide Resolution: Deoxyguanosine Residues at CpG are Crosslinked Preferentially,” J. Am. Chem. Soc. 1990, 112, 3637-3641; Li, V. -S.; Kohn, H. “Studies on the Bonding Specificity for Mitomycin C-DNA Monoalkylation Processes,” J. Am. Chem. Soc. 1991, 113, 275-283.

49. Norman, D.; Live, D.; Sastry, M.; Lipman, R.; Hingerty, B. E.; Tomasz, M.; Broyde, S.; Patel, D. J. “NMR and Computational Characterization of Mitomycin Cross-Linked to Adjacent Deoxyguanosines in the Minor Groove of the d(T-A-C-G-T-A).d(T-A-C-G-T-A) Duplex,” Biochemistry 1990, 29, 2861-2875.

50. Kumar, S.; Lipman, R.; Tomasz, M. “Recognition of Specific DNA Sequences by Mitomycin C for Alkylation,” Biochemistry 1992, 31, 1399-1407; Teng, S. P.; Woodson, S. A.; Crothers, D. M. “DNA Sequence Specificity of Mitomycin Cross-Linking,” Biochemistry 1989, 28, 3901-3907. 17

cross-links that were analogous51 to the mitomycin C-induced cross-links. FR-66979 cross-links are also specific to 5′-d(CG) DNA sequences. Monoalkylation of this compound with DNA showed no selectivity for 5′-d(CG), however, in contrast to

alkylation to mitomycin C.52

The DNA cross-linking property of mitomycin C in cells relies on the

enzymatic reduction. Recently, two semisynthetic mitomycin analogs BMY-25067

(BMS-181174) and KW-2149 were obtained by simple modification of the 7-

substituent of mitomycin C. They were designed to undergo a nonenzymatic

intramolcular reduction of the quinone ring by the 7-position sulfhydryl group that

was generated by the disulfide exchange with the glutathione or another external

thiol.53 (Figure 1.6) They exhibit antitumor activity superior to mitomycin C, are

active against mitomycin C-resistant tumors, and are both under clinical trials.

51. Fukuyama, T.; Goto, S. “Synthetic Approaches toward FR-900482. I. Stereoselective Synthesis of A Pentacyclic Model Compound,” Tetrahedron Lett. 1989, 30, 6491-6494.

52. Williams, R. M.; Rajski, S. R.; Rollins, S. B. “FR 900482, A Close Cousin of Mitomycin C that Exploits Mitosene-Based DNA Crosslinking,” Chem. Biol. 1997, 4, 127-137.

53. Doyle, T. W.; Vyas, D. M. “Second Generation Analogs of Etoposide and Mitomycin C,” Cancer Treat. Rev. 1990, 17, 127-131; Kono, M.; Saitoh, Y.; Kasai, M.; Sato, A.; Shirahata, K.; Morimoto, M.; Ashizawa, T. “Synthesis and Antitumor Activity of A Novel Water Soluble Mitomycin Analog; 7-N-[2-[[2-(g-L-Glutamylamino)ethyl]dithio]ethyl]mitomycin C,” Chem. Pharm. Bull. 1989, 37, 1128-1130. 18

O OCONH2 RSS(CH2)2HN OMe

NNH O

R = p-nitrophenyl BMY-25067

R = HO2C(NH2)CH(CH2)2CONH(CH2)2 KW-2149

O OCONH2 O OCONH2 - RSS(CH2)2HN OMe S(CH2)2HN OMe G-SH NNH NNH O O

- OH OCONH2 H O OCONH2 GSS(CH ) HN N 2 2 OMe G-SH OMe

NNH S NNH OH Me O

Figure 1.6 Disulfide Substituted Analogs of Mitomycin C

CHAPTER 2

PREVIOUS SYNTHETIC STUDIES

Immediately after the structural elucidation was accomplished in 1962, this group of complex and delicate natural products attracted the attention of synthetic organic chemists. Numerous efforts at the total synthesis of mitomycins and related compounds have been reported. Investigations utilizing modern synthetic methods are still being conducted even after four decades extensive research. The passion and interests on mitomycins does not persist without reason.1 The fascinating and novel array of structures of the mitomycin family provoked a long-term concentration for synthetic chemists. The density of functionalities housed on this small molecular framework leads to complicated interactions that are not easily unearthed and sometimes beyond the current knowledge of organic chemistry. The total synthesis of the mitomycins, especially mitomycin C, has continuously proven to be a tremendous challenge. The accomplishment of the synthesis has been seen as a great testimony to the reasoning of organic chemistry. The new methodologies developed

1. It is interesting to find out that Fukuyama and Ciufolini, who both successfully synthesized mitomycin C/FR-900482/FR-66979 as professors, have worked on mitomycins during their graduate/postdoctoral studies. 20

along with the total synthesis have greatly strengthened the power of organic chemists

to synthesize difficult and complex molecules.

Although those numerous efforts in a number of laboratories have generated

more than twenty approaches over the last four decades, to date there are only two

successful total syntheses of mitomycin C and two of mitomycin K. On one hand, it

suggests the mitomycins are very challenging molecules in terms of synthetic targets;

on the other hand, this suggests there remains a lot of work to be done in this area,

which might explain the fact that a considerable number of new synthetic approaches

have been recently reported. A number of reviews have been published in this area.2

From the strategic viewpoint, most of the previous approaches toward the mitomycin system could be classified systematically into eight categories. (Figure 2.1) Approach b is the only successful approach3 toward the synthesis of mitomycin C; approaches d

and e are successful in the total synthesis of mitomycin K.4 Most of the investigations

were aimed at the synthesis of more stable mitosanes or mitosenes. This chapter will

only briefly review the successful total syntheses of mitomycins in the 1970’s and

1980’s. Instead, the focus and emphasis are given to some of the most recent progress

and the approaches similar to our strategy.

2. Coleman, R. S. “Total Synthesis of Natural Product DNA Cross-Linking Agents,” Curr. Opinion Drug Disc. Dev. 2001, 4, 435-449; Danishefsky, S. J.; Schkeryantz, J. M. “Chemical Explorations Driven by An Enchantment with Mitomycinoids – A Twenty Year Account,” Synlett 1995, spec. issue, 475-490; Fukuyama, T.; Yang, L. “Total Synthesis of Mitomycins,” In: Studies in Natural Products Chemistry Atta-ur-Rahman Ed., Elsevier Science, 1993, 13, 433-471; Remers, W. A.; Dorr, R. T. “Alkaloids: Chemical and Biological Perspectives,” Relletier, S. W. Ed. Wiley, New York, 1988, 1-67; Remers, W. A. “The Chemistry of Antitumor Antibiotics,” Vol. 1, Wiley, New York, 1979.

3. Fukuyama used an indirect way to synthesize mitomycins.

4. Approach e was realized via mitosene intermediate. Thus it could be seen as approaches d+g. 21

Most of the synthetic approaches are effective at constructing the mitomycin

hetereocyclic ring system, but might lack of generality, efficiency and convergency.

While the first total synthesis of mitomycins by Kishi is a true landmark of organic synthesis, the synthesis itself suffers from the length and poor stereocontrol. Total

syntheses of mitomycin K by Danishefsky and Jimenez rely heavily on the

transformation from C9 ketone to C9−C10 double bonds, and are thus unlikely to be a

feasible route to mitomycin A/B type of molecules. Moreover, all four total syntheses

are racemic. An enantioselective, or even better, an asymmetric total synthesis of

mitomycins has not yet accomplished. These deficiencies have encouraged us to

investigate a potentially general and conceptually simple synthesis of mitomycins.

We are especially interested in the synthesis of mitomycin C, not only because of its potent antitumor activity and important clinical usage but also because of its general conversion to other types of mitomycins including mitomycin K. Our efforts toward the total synthesis of mitomycin C will be described in detail in the following several chapters.

9 7 b 9a c AB 1 NN 6 4a NNC D N 2 mitosane e

f g N N N

mitosene

Figure 2.1 Strategies toward Construction of Mitomycin Ring System

2.1 Kishi’s Total Syntheses of Mitomycins

The first total synthesis of mitomycins was achieved by Kishi in 19775 more than two decades after mitomycins were discovered. Though very lengthy in practice, it is masterful at the strategic and planning level and is a true milestone in organic synthesis.

Kishi’s strategy was to construct an eight-membered ring through an intramolecular Michael addition of a free primary amine to the quinone, followed by a transannular closure to form the B and C rings. (Figure 2.2) The troublesome elimination of methanol was avoided since the bridgehead nitrogen was connected to the quinone, which stabilized the system, and the reaction conditions were mild. The use of the unusual 3-acetoxypropyl protecting group on the azirindine nitrogen was a key to the success of the synthesis. The mild deprotection conditions made the survival of this delicate molecule possible.

Although the yields of most of the steps of this lengthy synthesis are moderate to good, the overall yield (0.06%) is unsurprisingly low for this impractical 45-step operation. The dihydroxylation reaction of olefin using OsO4 was not stereoselective;

the whole synthesis is racemic. However this first total synthesis demonstrated the

enormous difficulties in the synthesis of such a compact and highly functionalized molecule. This eight member ring approach later becomes a standard method for the

total synthesis of structurally related FR-900482.

5. Fukuyama, T.; Nakatsubo F.; Cocuzza A. J.; Kishi, Y. “ Synthetic Studies toward Mitomycins. III. Total Syntheses of Mitomycin A and C,” Tetrahedron Lett. 1977, 49, 4295-4298. 24

OBn 1. H2O2, K2CO3 MeO MeO 2. i-Pr2NLi, CH3CN 3. Jones oxidation 8 steps 61% OMe OBn 54%

OBn OBn OBn MeO O MeO OMe OMe 7 steps 5 steps OBn CN 46% OBn CN 66%

OBn OBn OBn OBn OMe OMe OsO MeO OMe 4 MeO OMe OH pyridine 10 steps OAc OH 87% 22% OBn 1:1 d.s. OBn OAc

OBn OBn OBn OMe OBn OMe MeO OMe MeO OMe

NH 3 steps N(CH2)3OAc 78% OBn OBn NBn2 NBn2

OH O OMe 1. HBF4 1. H2, Pd/C MeO OMe 2. COCl2, HNMe2 N(CH2)3OAc 2. O2 3. NH3 48% HN O 65%

O O OCONH2 1. NaOMe OCONH2 MeO OMe 2. DMSO-DCC MeO OMe

3. HClO4 NN(CH2)3OAc NNH 35% O O Mitomycin A (1)

Figure 2.2 Kishi’s Total Synthesis of Mitomycin A

2.2 Fukuyama’s Total Syntheses of Mitomycins

Ten years after Kishi’s remarkable total synthesis of mitomycin, Fukuyama reported the second and a more practical synthesis of mitomycin A and C.6 Fukuyama utilized a known interconversion between mitomycin A and isomitomycin A. (Figure

2.3)7

OCONH O OCONH O 2 2 OMe MeO H OMe MeO N NNH N O O

mitomycin A isomitomycin A

O OCONH2 MeO OMe

N N O albomitomycin A

Figure 2.3 Interconversion between Mitomycin A and Isomitomycin A

6. Fukuyama, T.; Yang, L. “Total Synthesis of (±)-Mitomycins via Isomitomycin A,” J. Am. Chem. Soc. 1987, 109, 7881-7882; Fukuyama, T.; Yang, L. “Practical Total Synthesis of (±)-Mitomycins C,” J. Am. Chem. Soc. 1989, 111, 8303-8304. This was the second time Fukuyama finished the synthesis of mitomycins: also see referece 5 in this chapter.

7. Kono, M.; Saitoh, Y.; Shirahata, K.; Arai, Y.; Ishii, S. “Albomitomycin A and Isomitomycin A. Products of Novel Intramolecular Rearrangement of Mitomycin A,” J. Am. Chem. Soc. 1987, 109, 7224-7225. 26

There are obvious advantages to synthesize mitomycins through isomitomycin

intermediates. The introduction of aziridine, which was very troublesome and lengthy in many other approaches, was accomplished in a much simpler manner by a single intramolecular azidoolefin cyclization. (Figure 2.4) Moreover, elimination of methanol, the most serious and challenging problem in mitomycin synthesis would no

longer be a threat since the methoxy group was located at a rigid bridge-head position

where elimination would result in an anti-Bredt8 compound.

Again, the 3-acetoxypropyl protecting group was used to protect the

secondary pyrrolidine amine as Kishi did for the aziridine nitrogen in the original

approach. Further efforts to achieve asymmetric synthesis proved unsuccessful by this

route. A chemical resolution of a late intermediate was conducted in order to provide

optically pure (−)-isomitomycin A.

8. Shea, K. J. “Recent Developments in the Synthesis, Structure, and Chemistry of Bridgehead Alkenes,” Tetrahedron 1980, 36, 1683-1715. 27

COPh OH OBn MeO MeO MeO

6 steps 7 steps NH N3 2 76% OMe 85% OMe OMe

OSiMe3 OBn EtS O OSiMe3 SEt ∆ MeO O O SnCl4, pyridine 86% 95% N3 OMe

OSiMe3 OCONHCOCCl OBn OBn 3 SEt SO2Et MeO O MeO O O 5 steps OAc N 81% N OMe OMe

OCONH2 O OMe O OCONH2 MeO NH NH3 H2N OMe

4 steps N 85% NNH 47% O O

Figure 2.4 Fukuyama’s Total Synthesis of Mitomycin C

2.3 Danishefsky’s Total Syntheses of Mitomycin K and FR-900482

Another member of mitomycin family, mitomycin K was synthesized by

Danishefsky and co-workers in 1993.9 Mitomycin K has a characteristic methylene group at C9 and C10, not present in the more traditional congeners. The presence of

exo olefin poses another locus of potential structural vulnerability and may confer

electrophilicity to the compound even in the absence of reductive activation by virtue

of its conjugation with the quinone.

Historically this remarkable report of the concise total synthesis of mitomycin

K is actually a “by-product” of their exploration towards the total synthesis of the

related natural product FR-900482. While their initial twenty years efforts toward

construction of mitomycins A, B and C had generated several approaches toward

mitosanes and mitosenes and greatly enriched the understanding of the chemistry and

biology of mitomycins, unfortunately their journey did not provide an approachable

route to the total synthesis of mitomycins.

Based on their long term involvement in hetero Diels-Alder reaction,10

Danishefsky proposed an idea with admirable simplicity of construction FR-900482 ring system employing a cycloaddition of a suitable diene with an appropriate aryl nitroso dienophile. Given the apparent difficulties in the operation of a C7−C8

9. Benbow, J. W.; Schulte, G. K.; Danishefsky, S. J. “The Total Synthesis of (±)-Mitomycin K,” Angew. Chem. Int. Ed. 1992, 31, 915-917; Benbow, J. W.; McClure, K. F.; Danishefsky, S. J. “Intramolecular Cycloaddition Reactions of Dienyl Nitroso Compounds: Application to the synthesis of Mitomycin K,” J. Am. Chem. Soc. 1993, 115, 12305-12314.

10. Boger, D. L. “Heterodiene Additions,” Comp. Org. Syn. Trost, B. M.; Fleming, I. Eds. Pergamon Press, Oxford, 1991, 5, 451-550. 29

closure via nucleophilic attack from C7 to an electrophile at C8, it would be desirable if the hetereo Diels-Alder reaction could be conducted on an intromolecular basis by having the diene connected with the aromatic ring through a C7 tether. (Figure 2.5)

OCONH OH 2 OR OH 7 OR 8 O O NH O RN 1 N N 15 16

O OR OR OR N N N NH O O HO 18 17

Figure 2.5 Danishefsky’s Plan to Construct FR-900482 Ring System

To construct FR-900482 ring system, the desired hetero Diels-Alder reaction

of compound 16 has to occur regioselectively in the bridged mode to generate intermediate compound 15. However there is also possibility that cycloaddition would occur in fused mode to produce intermediate compound 17. Danishefsky and

co-workers carefully evaluated this and quickly realized the opportunity that

compound 17 could be a very good intermediate leading to mitomycins through

rearrangement to compound 18, though there were serious concerns about the viability of the existence of compound 17. Further investigation revealed that this C7 tether intramolecular Diels-Alder method happened in the fused mode and did not constitute a viable route to reach FR-900482. (Scheme 2.6)

OMe OMe OH OMe OMe CHO + Li MeO2CNO2 MeO2CNO2

OMe O OMe O OMe OMe

N MeO2C MeO2CNO O

Figure 2.6 Further Studies Toward FR-900482 by Danishefsky

At this point, it was not difficult to recognize that with suitable substitution on

the A ring, the fused intramolecular Diels-Alder reaction might serve as a precursor to mitomycins. Thus, further modifications of the starting material paved the way to the total synthesis of mitomycin K. (Figure 2.7)

OMe OMe OH OMe OMe MeO CHO MeO hν + Li NO2 80% NO2 45% OMe OMe 19

OMe O OMe OMe O MeO [4+2] MeO OMe

NO N OMe OMe O

OMe O OMe O MeO OMe MeO OMe SPh N N N N N 4 steps R= S N N OMe 29% OMe OH R S 21 22

OMe O 1. n-Bu3SnH MeO OMe 2. hν Me3SiCH2Li 3. Raney nickel N NMe 21% OMe 90% 23

SiMe3 O MeO OH MeO OMe MeO OMe 1. silver(II) picolinate NNMe N NMe 2. PPTS O OMe low yields 24 mitomycin K

Figure 2.7 Danishefsky’s Total Synthesis of Mitomycin K

It is interesting to note that the photolysis of compound 19 did not yield the

expected compound 20, an analog of the product obtained in the FR-900482 model

studies, but indeed a much more advanced intermediate 21. It was presumed that

compound 21 was actually the product of further photolysis of 20. The C9 carbonyl group plays a key role to tie up the pyrroloindole nitrogen as a vinylogous amide to prevent the elimination of the bridgehead methoxy group.

The C3 aminal removal turned out to be problematic. It was removed by a low yielding five-step process to accommodate the aziridine. The final installation of the

C9 methylene group was realized using a Peterson olefination11 by reaction of compound 23 with trimethylsilylmethyllithium followed by acid promoted elimination, a process interrupted by an extremely low yielding (8-16%) oxidation of the dimethyl dihydroquinone to the corresponding quinone using silver(II) picolinate.

Nevertheless, a rather concise total synthesis of racemic mitomycin K was accomplished.

Danishefsky and co-workers then re-focused on the goal of a total synthesis of

FR-900482. A different strategy was used that featured an intermolecular hetero

Diels-Alder reaction and an intramolecular Heck reaction. As a consequence, a strategically creative total synthesis of FR-900482 was achieved.12 (Figure 2.8) It is

still the only FR-900482 synthesis that does not go through an eight-membered ring

intermediate.

11. Hudrlik, P. F.; Peterson, D. “Stereospecific Olefin-forming Elimination Reactions of β- Hydroxyalkylsilanes,” J. Am. Chem. Soc. 1975, 97, 1464 – 1468.

12. Schkeryantz, J. M.; Danishefsky, S. J. “Total Synthesis of (±)-FR-900482,” J. Am. Chem. Soc. 1995, 117, 4722-4723. 33

OBn OBn OCH2OMe I 1. SmI2 I HO + 2. Oxone MeO CNO MeO CNO 2 2 85% 2 26 27

OBn OH OBn OAc OCH2OMe OCH2OMe I I hν 80% O O NCO Me MeO CN 7 steps MeO CN 2 2 35% 2 28

OBn 1. K2CO3, MeOH OCH2OMe 2. Swern oxidation I (Ph3P)4Pd, Et3N O NCO Me 3. Ph3PCH2 2 93% MeO2CN 75%

OBn OBn O OCH OMe OCH OMe 2 1. OsO4, NMO 2

O NCO2Me 2. DIAD, Ph3P O NCO2Me MeO2CN 77% MeO2CN

29 30

OH OCONH OBn OH 2 OCH2OMe OH SmI2 92% 10 steps O NCO2Me O NH MeO2CN 31% OHC N 31 FR-900482 (7)

Figure 2.8 Danishefsky’s Total Synthesis of FR-900482

The synthesis started with the o-nitrosoiodo aromatic system 26, prepared from methyl vanillate. Cycloaddition of 26 and 27 provided an 80% yield of compound 28. After seven steps of trivial synthetic transformations, the double bond functionality was transformed into aziridino methylcarbamate. Further standard operations converted primary acetate into C7−C12 double bond. Intramolecular Heck arylation using (PPh3)4Pd proceeded smoothly to give exo-olefin 29. Upon

osmylation and oxirane formation, compound 30 and its diasteromer were obtained in

10:1 ratio. Reductive epoxide ring opening afforded primary alcohol 31 in an

excellent yield. Standard protecting group and oxidation state manipulation furnished

racemic FR-900482 in 10 steps.

2.4 Jimenez’s Total Synthesis of mitomycin K

It might be fair to say Jimenez’s report in 199613 of a concise total synthesis

of mitomycin K was based on several key transformations of Danishefsky’s work.

However the significance of this report was not diminished. Jimenez’s strategy

involved a key transformation of mitosene to mitomycin by utilizing a MoO5⋅HMPA

oxidation to install the C9a methoxy group. (Figure 2.9) The synthesis is very

efficient in terms of the amazing simplicity and the relatively high yielding; it is also

unexceptional in respect to its oxidation strategy.

13. Wang, Z.; Jimenez, L. S. “A Total Synthesis of (±)-Mitomycin K. Oxidation of the Mitosene C9-

C9a Double Bond by (Hexamethylphosphoramido)oxodiperoxomobdenum (VI) (MoO5⋅HMPA),” Tetrahedron Lett. 1996, 37, 6049-6052; Wang, Z.; Jimenez, L. S. “Synthesis of An Aziridinomitosene Analog,” J. Org. Chem. 1996, 61, 816-818. 35

OTBS MeO Me MeO NaH CHO + 8 steps SMe2 Me Me N 19% H I OTBS

OTBS OTBS MeO MeO 1. NaN3 MoO5⋅HMPA N3 Me N O 2. CH3SO2Cl Me N MeOH OTBS 3 steps OTBS OMs 29%+46%β 59%

OTBS O OTBS O MeO MeO OMe 1. Ph P OMe N3 3 Me3SiCH2Li N 2. MeOTf N NMe Me Me 76% OTBS OMs 55% OTBS

SiMe TBSO 3 O OH MeO MeO OMe PCC OMe

Me N NMe 63% Me N NMe TBSO O mitomycin K

Figure 2.9 Jimenez’s Total Synthesis of Mitomycin K

2.5 Fukuyama’s Total Syntheses of FR-900482

Fukuyama described the first synthesis of FR-900482 in 199214 in a linear and

lengthy fashion that relied on a transannular hemiketalization similar to that used in

Kishi’s mitomycin synthesis. Very recently, a more fascinating and efficient

enantioselective total synthesis of (+)-FR-90048215 based on the earlier synthesis was

reported by the same author using a relative convergent approach through a facile

construction of the N-hydroxybenzazocine intermediate.

Retrosynthetically, both syntheses share similar plans. (Figure 2.10) Although

the 1992 synthesis is linear and the 2002 one is relatively convergent, the two share

the same intermediate pentacyclic intermediate 32 and similar N-hydroxybenzazocine

33; both syntheses construct the C11−N bond as the last step to cyclize the eight

membered ring. In the second synthesis, the chirality was introduced by the terminal

acetylene 38 that also served as a construction unit for the convergent synthesis of the

eight membered ring.

14. Fukuyama, T.; Xu, L.; Goto, S. “Total Synthesis of (±)-FR-900482,” J. Am. Chem. Soc. 1992, 114, 383-385.

15. Suzuki, M.; Kambe, M.; Tokuyama, H.; Fukuyama, T. “Facile Construction of N- Hydroxybenzazocin: Enantioselective Total Synthesis of (+)-FR-900482,” Angew. Chem. Int. Ed. 2002, 41, 4686-4688. 37

OCONH O OH 2 OBn OH O

O NH PMPO OO OHC N N

PMP = p-methoxyphenyl FR-900482 (7) 32

OR O

O R N R 33 1992 2002

OR OBn OH OBn SPh O

PMPO N MeO2C NO2 Ac O 34 35

OBn OBn O O CHO OSiMe3 X + +

PMPO N3 MeO2C NO2 RO 36 37 38

Figure 2.10 Retrosynthetic Analysis of Fukuyama’s FR-900482 Syntheses

The 1992 synthesis is outlined in Figure 2.11. The synthesis started from N- benzylamine 39, which was prepared in 41% from ethyl acetoacetate and benzylamine. It was converted to azide 36 by a nine-step sequence of operations in

44% overall yield. Lewis acid catalyzed addition of 2-(trimethylsiloxy)furan and

Michael addition of thiophenol to the resulting reactive butenolides followed by acetylation and reductive removal of benzylic acetate with triethylsilane provided an azidolactone, which was further reduced to give eight-member benzazocine 34. A five-step sequence of operations included amine protection, oxidative olefin formation, acetate hydrolysis, epoxidation and Swern oxidation, which converted aminol alcohol 34 into epoxide 40. Hydromethylation of 40 and reduction was followed by silylation of the primary hydroxy group, reductive removal of the amide acetate, oxidation of the resuting secondary amine to the hydroxylamine, acetate ester formation and Swern oxidation to yield ketone 41. Hydrazinolysis of the acetate, deprotection of the t-butyldimethylsilyl (TBS) ether and protection of the diol as the corresponding acetonide afforded pentacyclic compound 32. Eight-step transformations including conversion of the acetonide into carbonate, oxidative deprotection of the p-methoxyphenyl (PMP) ether, pyridinium chlorochromate (PCC) oxidation of the resulting alcohol to aldehyde and standard aziridine ring formation afforded 42 in 38% yield. The final ammonolysis of 42 gave (±)-FR-900482 in 95% yield. This marked the accomplishment of the first total synthesis FR-900482 that took more than 40 steps with a 0.4% overall yield.

OH OBn

CO2Me CHO 9 steps MeO O 7 steps NHBn 44% N3 37% 39 36

OBn OH OBn O SPh O PMPO 5 steps PMPO N 65% N H Ac 34 40

OSit-BuMe2 O OBn O 1. NH2NH2 OBn 2. n-Bu4NF O 7 steps O 3. Me C(OMe) O O 29% PMPO 2 2 N N 96% OPMP OAc 41 32

O O OCONH OH OH 2

O NH3 OH 10 steps O NH 95% O NH 34% OHC N OHC N

42 FR-900482 (7)

Figure 2.11 Fukuyama’s 1992 Total Synthesis of FR-900482

The 2002 approach by Fukuyama and co-workers is more efficient in terms of

the length and the overall yield of the synthesis. Moreover, because it used a chiral

starting material to construct the eight membered ring, the whole synthesis is

enantioselective. (Figure 2.12)

The journey commenced with Sonogashira coupling16 of acetylene 44 and aryl

triflate 43, an intermediate prepared in Danishefsky’s synthesis of FR-900482, to

provide aryl acetylene 45. Regioselective transformation of the alkyne to the required ketone was realized by a novel conjugate addition of a secondary amine to this o-

nitroaryl acetylene followed by acidic hydrolysis of the intermediate enamine.

Oxidation state modification followed by standard protecting group manipulation,

routine epoxidation and Dess-Martin oxidation gave epoxyaldehyde 46 in 44% yield

in a straitforward manner. The eight membered ring was beautifully closed by hydrogenation using Pt/C as a catalyst to form the hydroxylamine without the need to adjust the oxidation state of nitrogen. Protection of hydroxylamine as the acetonide ether followed by deprotection of the silyl group and Swern oxidation furnished

ketone 47. Hydroxylmethylation and deprotection of the hydroxylamine acetonide

were accomplished with formalin in the presence of catalytic amount LiOH followed

by acidic workup. The resulting hydroxylamine attacked the ketone in situ to form the

hemiacetal, which was treated with 2-methoxypropene in acid to afford pentacyclic

acetonide. The methyl ester was reduced by i-Bu2AlH and the resultant benzyl

alcohol was protected as the p-methoxyphenyl (PMP) ether to furnish the pentacyclic

16. Sonogashira, K. “Development of Pd-Cu Catalyzed Cross-Coupling of Terminal Acetylenes with sp2-Carbon Halides,” J. Organomet. Chem. 2002, 653, 46-49. 41

compound 32, a key intermediate in their 1992 synthesis. The enantioselective synthesis of (+)-FR-900482 was thereby accomplished in 11 extra steps in 31% yield by modifying the protocol established in the 1992 synthesis.

OBn O O O OTf Pd(OAc)2 + O OBn 75% MeO C NO 2 2 OSit-BuMe2 OSit-BuMe2 MeO2C NO2 43 44 45

OSi(i-Pr) OBn 3 1. H2, Pt/C O 2. 2-methoxypropene 9 steps 3. n-Bu4NF 44% MeO2C NO 2 O 4. Swern oxidation

OBn O O OBn 1. HCHO, LiOH O 2. 2-methoxypropene O MeO2C N 3. i-Bu AlH O O OOMe 2 N 4. p-methoxyphenol, Ph P, DEAD 3 OPMP 47 32

OCONH OH 2 OH

O NH 11 steps OHC N 31% (+)-FR-900482 (7)

Figure 2.12 Fukuyama’s 2002 Total Synthesis of (+)-FR-900482

2.6 Terashima’s Total Synthesis of (+)-FR-900482

In 1997, Terashima and co-workers reported the first enantioselective

synthesis of (+)-FR-90048217 by a convergent but lengthy route that suffered from

protecting group problems and from poor stereochemical control.

The retrosynthetic analysis is outlined in Figure 2.13. It is clear that FR-

900482 can be derived without significant difficulty from intermediate 50 by protecting group manipulations and oxidation state adjustment. The most crucial step is the intramolecular aldol reaction of the highly functionalized dialdehyde 51 to construct the eight membered azirinobenzazocin 50. The cyclization precursor 50 in turn could be elaborated by coupling the aromatic segment 52 and the enantiomerically pure aliphatic segment 53.

In general, the formation of eight membered ring systems by intramolecular ring closure reactions is often difficult due to the steric strain and unfavorable entropy factors. However, in the case of dialdehyde 51, with the help of the strain of the azirindine and aromatic systems, the intramolecular cyclization was accomplished by using lithium hexamethyldisilazide. Unfortunately, stereochemistry at C8 could not be controlled. The stereochemistry of C8 hydroxymethyl group turned out to be opposite from the desired sense. Four extra steps were necessary for epimerization

17. Yoshino, T.; Nagata, Y.; Itoh, E.; Hashimoto, M.; Katoh, T.; Terashima, S. “Total Synthesis of An Enantiomeric Pair of FR900482. 2. Syntheses of the Aromatic and the Optically Active Aliphatic Segments,” Tetrahedron 1997, 53, 10239-10252; Katoh, T.; Nagata, Y.; Yoshino, T.; Nakatani, S.; Terashima, S. “Total Synthesis of An Enantiomeric Pair of FR900482. 3. Completion of the Synthesis by Assembling the Two Segments,” Tetrahedron 1997, 53, 10253-10270. 44

and this first enantioselective total synthesis was then accomplished in a 57-step sequence of reactions.

OCONH OCONH OH 2 OBn 2 OH OAc

O NH BnOCH O O NTs OHC N 2 N

(+)-FR-900482 (7) 48

OH OH OBn O OBn OH

89 NTs NTs N N BnOCH2O OH BnOCH2O OH

49 50

OBn CHO CHO NTs N BnOCH O 2 CO2Allyl 51

OBn t-BuMe SiO 2 CO2CH2CCl3 OSit-BuMe2 N +

NHCO2Allyl O BnOCH2O OTf 52 53

Figure 2.13 Retrosynthetic Analysis of Terashima’s Total Synthesis of (+)-FR-

900482

2.7 Martin’s Formal Total Synthesis of FR-900482

Recently, Martin and co-workers reported an elegant formal total synthesis of

FR-90048218 by constructing a late intermediate of Fukuyama’s 1992 synthesis. A ring-closing metathesis was used to close the eight membered ring. Although the whole synthesis is racemic, one of the intermediate was successfully synthesized in enantiomeric enriched fashion by enzymatic resolution, which suggested this route could also be used for an enantioselective synthesis. (Figure 2.13)

The formal total synthesis started with aryl diol compound 54. Differentiation of the two hydroxyl groups was successful by mono-protection of one as a p- methoxybenzyl (PMB) ether. Subsequent silylation of the remaining alcohol, Raney nickel reduction and trichloroethyl carbamate (Troc) protection of the resulting aryl amine followed by allylation on the aryl nitrogen affords the Troc protected allylamine 55 in a straitforward manner. Conversion of the primary silyl ether to allylic alcohol proceeded smoothly by standard procedures to give ring-closing metathesis precursor 56. The diene 56 underwent facile ring-closing metathesis with the first generation Grubbs’ ruthenium catalyst to yield desired eight membered benzazocine 57. Routine reductive deprotection of Troc protecting group, m- chloroperbenzoic acid (m-CPBA) oxidation and acetyl ester protection of the resulting hydroxylamine was followed by another m-CPBA oxidation to afford the epoxide. The formal total synthesis was accomplished after t-butyldimethylsilyl

18. Fellows, I. M.; Kaelin, D. E.; Martin, S. F. “Application of Ring-Closing Metathesis to the Formal Total Synthesis of (+)-FR-900482,” J. Am. Chem. Soc. 2000, 122, 10781-10787. 47

(TBS) ether was used to replace the p-methoxybenzyl ether and the benzyl ether

protected benzyl alcohol was converted into a p-methoxyphenyl benzyl ether to provide the key Fukuyama intermediate 41.

OH OPMB OBn OBn 1. HF⋅pyr OH OSi(i-Pr)3 2. Swern oxidation 5 steps 3. MgBr NO2 68% N OBn OBn CO2CH2CCl3 65% 54 55

OPMB OPMB OBn PCy3 OBn OH OH Cl Ru 1. Zn, HOAc Cl Ph 2. m-CPBA PCy3 3. Ac O N 78% N 2 4. m-CPBA OBn CO2CH2CCl3 OBn CO CH CCl 2 2 3 43%

56 57

OPMB OTBS OBn OH OBn OH 1. CF3COOH 2. t-BuMe SiCl O 2 O 3. H , Pd/C N 2 N 4. p-methoxyphenol, PPh3, DEAD OBn OAc OPMP OAc 20% 58 41

Figure 2.14 Martin’s Formal Total Synthesis of FR-900482

2.8 Williams’ Total Synthesis of (+)-FR-900482 and (+)-FR-66979

A very recent report by Williams and co-workers19 described a convergent,

concise, and enantioselective synthesis of (+)-FR-66979 and (+)-FR-900482 by an

unusual strategy to deprotect the N-p-methoxybenzyl group and simultaneously form

the hydroxylamine hemiketal in a single operation. The labile aziridine functionality

was installed in the very beginning and carried intact through the synthesis.

The synthesis (Figure 2.15) started from aziridine 60, prepared in 10 steps

from 2-butene-1,4-diol. Deprotonation of the methyl group of aryl precursor 61

followed by nucleophilic attack onto the aldehyde 60 constructed C7−C8 bond in the

FR-900482 system. The resulting secondary hydroxyl group was protected as a silyl ether and the C11−N bond was formed to afford eight membered ring compound 62

by the nucleophilic attack of the aryl amine to the C11 aldehyde functionality

followed by in situ reduction of the resulting imine. Reaction of 62 with p-

methoxybenzyl bromide, followed by removal of the diethylisopropylsilyl protecting group with tris(dimethylamino)sulfonium difluorotrimethylsilicate (TSAF) in

DMF/H2O and subsequent oxidation with Dess-Martin periodinane afforded ketone

63. Hydromethylation with i-Pr2NLi (LDA) and CH2O proceeded smoothly to furnish

to the aldol adduct 64 as a 1:1 mixture of diastereomers in 50% yield. An extra

epimerization step was required to convert the undesired diastereomer to the desired

19. Judd, T. C.; Williams, R. M. “Concise Enantioselective Synthesis of (+)-FR66979 and (+)- FR900482: Dimethyldioxirane-Mediated Construction of the Hydroxylamine Hemiketal,” Angew. Chem. Int. Ed. 2002, 41, 4683-4685. 50

one. Compound 64 was then protected as a t-butyldimethylsilyl (TBS) ether. The removal of p-methoxybenzyl (PMB) protecting group, oxidative formation of the hydroxylamine and in situ formation of hemiketal was accomplished in one step using dimethyloxirane as an oxidant to furnish tetracyclic compound 65. Removal of the t- butyldimethylsilyl (TBS) protecting group followed by reaction of the resulting primary alcohol with trichloroacetyl isocyanate installed the urethane/carbamate moiety at C13. Removal of methoxymethyl (MOM) protecting group on the phenol was successful with Me3SiBr in the presence of acid sensitive aziridine functionality.

Final reduction of both carbomethoxy groups with LiBH4 amazingly afforded the natural product (+)-FR-66979 (8). Further oxidation of the benzylic alcohol to aldehyde by Swern oxidation furnished another natural product (+)-FR-900482.

OCH2OMe OHC HO Me NCO2Me + HO 10 steps PMBO 7 steps MeO CNO 21% 2 2 32% 60 61

OSiEt (i-Pr) MeOCH O O MeOCH2O 2 1. PMBBr 2 2. TASF NCO2Me 3. Dess-Martin NCO2Me MeO C MeO C 2 HN 65% 2 N PMB TASF = tris(dimethylamino)sulfonium 62 63 difluorotrimethylsilicate

OH MeOCH2O O 1. t-BuMe2SiOTf (i-Pr)2NLi/CH2O NCO Me 2. dimethyldioxirane 50% 2 d.s. 1:1 MeO2C N 29% PMB

1. n-Bu4NF OSit-BuMe2 OCONH2 MeOCH2O 2. Cl3CCONCO HO OH OH 3. Me3SiBr

4. LiBH4 O NCO2Me O NH MeO2CN 5. Swern oxidation OHC N

13% 65 (+)-FR-900482 (7)

Figure 2.15 Williams’ Total Synthesis of (+)-FR-900482

2.9 Ciufolini’s Total Synthesis of FR-66979

Very recently, Ciufonili and Ducray reported a total synthesis of FR-66979 using a very innovative approach.20 This synthesis involved a new methodology21 to assemble benzazocene efficiently through an unusual fragmentation of a silylated aziridine orchestrated by a preliminary homo-Brook22 transposition. (Figure 2.16)

OH O O SiMe SiMe3 SiMe3 3 base

N N N

OH OSiMe3

N N

Figure 2.16 Ciufonili’s Approach to Benzazocene

20. Ducray, R.; Ciufolini, M. A. “Total Synthesis of (±)-FR66979,” Angew. Chem. Int. Ed. 2002, 41, 4688-4691.

21. Ciufolini, M. A.; Chen, M.; Lovett, D. P.; Deaton, M. V. “Application of Ene-Like Reactions of Aldehydes with Vinyl Ethers: Facile Assembly of Benzazocenone Intermediates for Mitomycinoids,” Tetrahedron Lett. 1997, 38, 4355-4358.

22. Hudrlik, P. F.; Hudrlik, A. M.; Kulkarni, A. K. “Protodesilylation Reactions of Simple β- Hydroxysilanes (and α-Hydroxysilanes). Homo-Brook rearrangements,” J. Am. Chem. Soc. 1982, 104, 6809-6811. 53

The synthesis (Figure 2.17) started from the readily available aldehyde 66.

Nucleophilic addition of lithiated allytrimethylsilane to compound 66 in the presence of (i-PrO)4Ti followed by thermo- and photo-chemical formation of aziridine to furnished aziridine compound 67, which underwent a homo-Brook triggered fragmentation to benzazocenol 68. This is probably the most efficient route so far to construct the eight membered nitrogen containing heterocycles in mitomycinoids syntheses.

Straightforward operations including the reported procedures and perruthenate oxidation converted benzazocenol 68 into ketone 69. Hydrazine mediated acetate removal initiated the in situ hemiketal formation. Carefully monitored debenzylation using hydrogenolyis catalyzed by Pd/C afforded tetracyclic compound 70, which was elaborated to (±)-FR-66979 by a modification of known procedures.

OBn OBn OBn 1. allyltrimethylsilane OBn OH CHO BuLi, Ti(Oi-Pr)4 n-Bu4NOH SiMe 2. ∆ 3 N 49% 3 3. hν N H OBn 62% OBn 66 67

OBn OBn OH O OBn 1. m-CPBA OBn 2. Ac2O 1. N2H4 O 3. m-CPBA 2. H2, Pd/C HN 4. TPAP, NMO N OBn OBn 51% OAc 68 TPAP = tetrapropylammonium 69 perruthenate NMO = 4-methylmorpholine N-oxide

OH O OH OAc OH 1. 2-methoxypropene O OMs 2. LiN3 O O O N 3. Ac2O N N3 OH d. CH3SO2Cl OAc 70 32% 71

O O OCONH OAc OH 2 O 1. Ph P OH 1. CF3COOH OMs 3 2. COCl2 O 2. NH3 O NH N N 28% N 31% OAc 3 OH 72 FR-66974 (8)

Figure 2. 17 Ciufolini’s Total Synthesis of FR-99674

2.10 Cha’s Approach to Mitomycin System

The main difficulty in the synthesis of the mitomycins stems from the introduction and preservation of the extremely labile C9a-hydroxy or -methoxy group. In 1997, Cha and co-workers reported an interesting method for the stereocontrolled construction of mitomycin C9−C9a bond and formation of C9a hydroxy functionality.23

Cha’s approach was based on the coupling reactions of in situ generated dialkoxytitanacyclopropane 73 and various carbonyl compounds24 (Figure 2.18), which in turn were built upon the cyclopropanation of carboxylic esters originally developed by Kulinkovich.25

23 Lee, J.; Ha, J. D.; Cha, J. K. “New Synthetic Method for Functionalized Pyrrolizidine, Indolizidine, and Mitomycin Alkaloids,” J. Am. Chem. Soc. 1997, 119, 8127-8128.

24 Lee, J.; Kang, C. H.; Kim, H.; Cha, J. K. “Intramolecular Hydroxycyclopropanation of Vinyl Carboxylic Esters,” J. Am. Chem. Soc. 1996, 118, 291-292.

25 Kulinkovich, O. G.; Sviridov, S. V.; Vasilevskii, D. A. “Titanium(IV) Isopropoxide-Catalyzed Formation of 1-Substituted Cyclopropanols in the Reaction of Ethylmagnesium Bromide with Methyl Alkanecarboxylates,” Synthesis 1991, 234-235. 56

2 equiv i-Pr-O O-i-Pr c-C5H9MgCl Ti R ClTi(O-i-Pr)3 R 73

1 O HO R 73 R1 OR2 R

i-Pr-O O-i-Pr Ti O R R1 2 R O

Figure 2.18 Reactions of Dialkoxytitanacyclopropane

In contrast, it was suspected that imide 74 would lead to a different product:

the initial adduct 75 might be resistant to subsequent cyclopropanation and thus

sufficiently stable to be isolated upon hydrolysis. Indeed, the reaction went on

smoothly to afford compound 76 as the major product.26 The intramolecular reactions of imide 77 also underwent similar mechanism to produce correspondent acylaminal 78. (Figure 2. 19)

Extension to o-imidostyrene derivatives to construct mitomycin ring system was an obvious application to this method. When the unsubstituted o-imidostyrene 79

26. As stated in the paper, “none of the coupling reactions has been optimized as yet.” 57

was subject to the typical reaction condition at 0°C, the desired mitomycin B type

compound 80 was obtained in fair yield.

Cha’s approach provides a very efficient method for constructing C9−C9a

bond, and more importantly a facile way to introduce C9a hydroxy group. However,

there are still a lot more challenges remain unsolved. The C3 carbonyl is crucial for

the survival of C9a hydroxy group. While it was problematic to remove the C3

carbonyl in Danishefsky’s total synthesis of mitomycin K, in this case it will be even

more difficult to selectively remove this functionality in the presence of quinone which appears to be the only possible way to stabilize the C9a hydroxy group.

Moreover, the effect to the imide nitrogen of the electron rich aromatic ring, which is necessary to introduce the quinone functionality, needs to be investigated.

O-i-Pr HO O i-Pr-O Ti R CH2CH2R R O H2O N n-Bu N n-Bu c-C H MgCl 5 9 N n-Bu O ClTi(O-i-Pr)3 O 74 O 76 75 R=CH2CH2OTIPS 51%

O-i-Pr i-Pr-O Ti HO CH2OH O c-C5H9MgCl O ClTi(O-i-Pr)3 O2 N N 45% N O O O 77 78

i-Pr-O O-i-Pr CH2OH Ti c-C5H9MgCl OH O ClTi(O-i-Pr)3 O O2 N N 40% N

O O O 79 80

Figure 2.19 Cha’s Approach to the Mitomycin Ring System

2.11 Sulikowski’s Approach to Mitosenes

Sulikowski and co-workers reported27 an asymmetric approach to the construction of mitosene ring system by forming C9−C9a bond through an

enantioselective intramolecular C−H insertion reaction of meso diazoester with poor

enantioselectivity. (Figure 2.20)

O O N2 NN CO2Me CO2Me HH CO2Me O O N N N O Cu(I)OTf O O

O α-CO2Me 43% 51% ee β-CO2Me 13% 26% ee β-CO2Me 25% 16% ee α-CO2Me 10% 4% ee

CO2Me dichlorodicyanoquinone O N O

Figure 2.20 Sulikowski’s Approach to Mitosenes

27. Lim, H.-J.; Sulikowski, G. A. “Enantioselective Synthesis of a 1,2-Disubstituted Mitosene by a Copper-Catalyzed Intramolecular Carbon-Hydrogen Insertion Reaction of a Diazo Ester,” J. Org. Chem. 1995, 60, 2326-2327; Lee, S.; Lim, H.-J.; Cha, K. L.; Sulikowski, G. A. “Asymmetric Approaches to 1,2-Disubstituted Mitosenes Based on the Intramolecular Cyclization of Diazoesters,” Tetrahedron 1997, 53, 16521-16532. 60

2.12 Tandem Radical Cyclization Approaches

Ziegler’s approach28 to 9a-desmethoxymitomycins involved a tandem radical

cyclization of chiral aziridinyl bromide 81 to afford tetracyclic alcohol 82, which was

followed by regular functional group and protecting group manipulations to give 9a- desmethoxymitomycin A (83). (Figure 2.21)

A similar strategy applied to the synthesis of 9a-desmethoxymitomycin K

(84). A new methodology was developed to install the required C9−C10 double bond

by using facile fragmentation of β-alkyl-β-aryl-α-oxo-γ-butyrolactone.

28. Ziegler, F. E.; Belema, M. “Cyclization of Chiral Carbon-Centered Aziridinyl Radicals: A New Route to Aziridino[2′,3′:3,4]pyrrolo[1,2-a]indoles,” J. Org. Chem. 1994, 59, 7962-7967; Ziegler, F. E.; Berlin, M. Y. “A Synthesis of (+)-9a-Desmethoxymitomycin A via Aziridinyl Radical Cyclization,” Tetrahedron Lett. 1998, 39, 2455-2458; Ziegler, F. E.; Berlin, M. Y.; Lee, K.; Looker, A. R. “Formation of 9,10-Unsaturation in the Mitomycins: Facile Fragmentation of β- Alkyl-β-aryl-α-oxo-γ-butyrolactones,” Org. Lett. 2000, 2, 3619-3621.

OBn OBn CH2OH MeO MeO n-Bu SnH Br 3 N 6 steps N 55% H 48% OMe OMe N CO2t-Bu 81

OBn O CH2OH CH2OCONH2 MeO H MeO H

7 steps NNCO2t-Bu NNH 20% OMe O 82 83

OBn O CH2OH BnO MeO 1. propargyl alcohol, TsOH MeO H Br N 2. n-Bu3SnH N NCO2t-Bu OMe N 39% CO2t-Bu OMe 1. m-CPBA 2. O3 3. Me2S 93%

BnO BnO O O MeO H MeO KN(SiMe3)2, O2 H 85% N NCO2t-Bu N NCO2t-Bu OMe OMe 84

Fire 2.21 Ziegler’s Approach to 9a-Desmethoxymitomycins

The approaches of Jones29 and Parsons30 are similarly based on tandom

radical cyclization, although in both of cases much more simplified model system were investigated. Jones’ approach utilized a radical cylization to construct the

C1−C9a bond in mitosene 85; Parsons’ approach aimed at constructing the C9−C9a

bond in mitosane 86. (Figure 2.22)

Br n-Bu3SnH 9a 1 N 79% N

Br CO2Et OMe MeO CO2Et 9 n-Bu3SnH 9a 50% N N 7.3:1 isomer ration OMe MeO O O 86

Figure 2.22 Jones’ and Parsons’ Approaches31

29. Dobbs, A. P.; Jones, K.; Veal, K. T. “Indole Radical Cyclisations: A Rapid Route to Mitosenes,” Tetrahedron Lett. 1995, 36, 4857-4860.

30. Allan, G. M.; Parsons, A. F.; Pons, J.-F. “Tandem Radical Cyclisation and Translocation Approaches to Biologically Important Mitomycin Ring Systems,” Synlett 2002, 1431-1434.

31. In Parsons’ approach, the relative stereochemistry of 86 along the C9-C9a bond was not determined. 63

2.13 Vedejs’s Approaches to Aziridinomitosene

Recently, Vedejs and co-workers reported two different approaches to

aziridinomitosenes. Although there was tremendous effort put into the syntheses of mitomycins, mitosane and mitosene, most of those approaches were aimed at

constructing B, C or D ring with quinone A ring introduced as a six-membered

aromatic system and the A ring skeleton kept intact during the synthesis. Few reports

have been addressed at constructing the A ring. However the first approach by

Vedejs32 is an exception. This non-traditional approach relied on a key intramolecular

[3+2] cycloaddition reaction between an azomethine ylide and an alkyne for

simultaneous construction of six-membered A ring and five-membered B ring of the

fully elaborated aziridinomitosene system. (Figure 2.23)

32. Vedejs, E.; Klapars, A.; Naidu, B. N.; Piotrowski, D. W.; Tucci, F. C. “ Enantiocontrolled Synthesis of (1S,2S)-6-Desmethyl-(methylaziridino)mitosene,” J. Am. Chem. Soc. 2000, 122, 5401-5402. 64

OTBS OH OAc OSit-BuMe 1. Dess-Martin 2 2. LiCCCH2OTBS O O O NMe 3. Ac2O NMe N N N I I

OAc OSit-BuMe2 OAc OSit-BuMe2

1. AgOSO2CF3 O O CN 2. BnMe3NCN NMe NMe N N

OAc OSit-BuMe2 OAc OSit-BuMe2 OAc OSit-BuMe2 CN O CN N NMe N NMe NMe N O O

O OCONH2

N NMe O

Figure 2.23 Vedejs’ First Approach to Aziridinomitosene

The second approach of Vedejs33 to aziridinomitosene is retrosynthetically similar to Ziegler’s approach. While both of approaches aimed at constructing the

C3−N4 and C1−C9a bonds and shared the same method for forming the C3−N4 bond, the strategies for building the C1−C9a bond are different. Ziegler used a tandem radical cyclization; Vedejs employed a tandem anionic Michael addition-cyclization.

(Figure 2.24)

It is interesting to note the initial attempt to effect metal exchange and internal

Michael addition revealed a complex situation. (Figure 2.24) Treatment of stannane

87 with excess methyllithium followed by work-up with deuterated ethanol afforded cyclized mitosane 88 as a minor product and a high yield of its monodeuterio derivative 89a along with the corresponding de-stannylated diderterio compound 89b

(88:89a:89b = 9:55:36). This suggested that the indole C−H lithiation is faster than tin-lithium exchange and prevents Michael addition. It was suspected that the monodeuterated 89a, easily available from treatment with PhLi and EtOD, should be a better substrate for the cyclization process because of the slower undesired indole lithiation caused by a primary kinetic isotope effect. Indeed, treatment of monodeuterated stannane 89a with four equivalents of methyllithium followed by quenching with EtOH gave a dramatically altered ratio of products with 78% of desired cyclized mitosane 91. Quenching intermediate 90 with phenylselenyl chloride furnished mitosene 92, presumed to be the result of a spontaneous elimination of the resulting selenide intermediate. Classical conditions to remove triphenylmethyl

33. Vedejs, E.; Little, J. “Aziridinomitosenes by Anionic Cyclization: Deuterium as a Removable Blocking Group,” J. Am. Chem. Soc. 2001, 124, 748-749, 66

(trityl) protecting group afforded aziridinomitosene 93 in 65% yield. Earlier studies revealed that this kind of mitosene azirindine compound with functionalized aromatic system is not stable. In this case, the stability of aziridine 93 might benefit from the

C10 carbonyl stabilization and also the absence of the hydroxy or methoxy group on the aromatic ring

CO2Et D CO2Et CO2Et H H 1.MeLi, -65°C D SnBu + R N 3 2. EtOD N NCPh3 N 89a:89b:88 NCPh NCPh 3 55: 36 :9 3 87 88 89a R = SnBu3 89b R = D 1. PhLi, -78°C 2. EtOD

CO Et CO Et 2 2 H CO2Et D D MeLi D EtOH N SnBu3 N Li -65°C 78% N NCPh3

NCPh3 NCPh3 89a 90 91

PhSeCl 80%

CO2Et CO2Et

1. Et3SiH, MsOH

2. i-Pr2NEt N NCPh3 NNH 65% 92 93

Figure 2.24 Vedejs’ Latter Approach to Aziridinomitosene

2.14 Michael’s Enantioselective Synthesis of Aziridinomitosene

Very recently, Michael reported an enantioselective synthesis of aziridino-

mitosene34 using an intramolecular Heck coupling with an unusual strategy aiming at

constructing the C3−N4, C9a−N4 and aryl−C9 bonds.

The synthesis started from the “chiral pool” compound (−)-2,3-O-

isopropyllidene-D-erythronolactone (94), readily available from D-isoascorbic acid.35

The lactone ring was opened by treatment with the anion of 2-bromoaniline followed by mesylation of the alcohol and sodium hydride promoted cylization to give lactam

95. Thionation with Lawesson’s reagent and transformation of thiolactam into a vinylogous urethane afforded Heck coupling precursor 96. Palladium acetate promoted intramolecular Heck coupling proceeded well to give mitosene compound

97, which was transformed into azidoalcohol 98. The final aziridine installation was

36 accomplished by Staudinger reaction using the phosphorus (III) reagent P(OMe)3.

34. Michael, J. P.; de Koning, C. B.; Petersen, R. L.; Stanbury, T. V. “Asymmetric Synthesis of A Tetracyclic Model for the Aziridinomitosenes,” Tetrahedron Lett. 2001, 42, 7513-7516.

35. Thus, the synthesis is enantioselective, not asymmetric as the authors claimed in title.

36. Nakatsubo, F.; Fukuyama, T.; Cocuzza, A. J.; Kishi, Y. “Synthetic Studies toward Mitomycins. 2. Total Synthesis of dl-Porfiromycin,” J. Am. Chem. Soc. 1977, 99, 8115-8116. 69

O 1. 2-bromoaniline Br O O 2. MeSO2Cl 1. Lawesson's Reagent O 3. NaH O O N 2. Zn, BrCH2CO2Et 80% O 81% 94 95

Br CO Et CO2Et 2 1. CF3COOH Pd(OAc) 2. SOCl O 2 O 2 N 99% N 3. NaN3 O O 96 97

CO2Et CO2Et

1. MeSO2Cl N3 2. P(OMe) N 3 N NH OH 98 99

Figure 2.25 Michael’s Enantioselective Synthesis of Aziridinomitosene

2.15 Jones’ Stereoselective Synthesis of Mitosane

A novel and short synthesis of the mitosane skeleton using a stereoselective reduction of η6-arene chromium carbonyl complex was introduced by Jones and co- workers.37 η6-Arylchromium complex 101, prepared from indole 100, upon reduction by sodium cyanoborohydride in the presence of trifluoacetic acid afforded mitosane

102. The hydride added to the C9a carbon exclusively syn to the chromium complex with 95:5 diastereoselectivity. Further lithiation at the C9 benzylic carbon followed by addition of formaldehyde afforded complex 103 with the incoming electrophile trans to the C9a hydrogen. Photochemical decomplexation and transformation of primary alcohol to carbamate afforded mitosane 104.38

37. Jones, G. B.; Guzel, M.; Mathews, J. E. “Stereoselective Route to Mitosanes via Tricarbonyl η6 Arene Chromium Complexes,” Tetrahedron Lett. 2000, 41, 1123-1126.

38. We appreciate Professor Jones for the helpful discussions and the kind provision of NMR spectra of some of their key compounds. However, their 1H NMR spectra didn’t match the spectrum of any mitosane 104 isomer synthesized in our lab. Details of our synthesis of mitosane 104 will be discussed in detail in Chapter 4. 71

Cr(CO)6 NaCNBH3

N 96% N CF3COOH (CO)3Cr 92% 100 101 95:5 d.s.

OH OCONH2 1. hν H H H s-BuLi 2. PhOCOCl N N (CHO)n 3. NH3 N (CO) Cr (CO) Cr 3 90% 3 77% 102 103 104

Figure 2. 26 Jones’ Stereocontrolled Synthesis of Mitosane

2.16 Miller’s Enantioselective Synthesis of Mitomycin Core Structure

Recently, a concise and enantioselective synthesis of the tetracyclic

mitomycin core skeleton was reported by Miller and co-workers.39 The synthesis proceeded through an eight membered ring intermediate constructed by ring closing metathesis and relied on a transannular aminal formation similar to that used in

Kishi’s total synthesis of mitomycin C. The enatioselectivity was achieved in a rapid

fashion by screening a moderate sized peptide library of acylation atalysts for kinetic

resolution of a secondary alcohol intermediate. (Figure 2.27)

The synthesis started from commercially available isatin, which was

converted to ring closing metathesis precursor 105 through nomal protecting group

and functional groups manipulations. Ring closing metathesis followed by lithium

aluminum hydride reduction afforded racemic secondary alcohol 106. Kinetic

resolution by peptide-catalyzed acylation furnished chiral alcohol 107 with very high

enatiomeric excess. Epoxidation and nitric acid catalyzed ketal and t-butyl carbamate

protecting group removal followed by spontaneous transannular C9a aminal

formation afforded epoxide 109. However, the stereochemistry of the C9a methoxy

group is opposite to the desired conformation. Epimerization of the C9a methoxy

group and regular transformation of epoxide to aziridine afforded mitomycin core

structure 110.

39. Papaioannou, N.; Evans, C. A.; Blank, J. T.; Miller, S. J. “Enantioselective Synthesis of a Mitosane Core Assisted by Diversity-Based Catalyst Discovery,” Org. Lett. 2001, 3, 2879-2882. 73

Miller’s synthesis provides a new and relative short approach for

enantioselective synthesis of mitomycins, which has not yet been accomplished.

While it adopts Kishi’s eight-membered ring strategy, it was apparently aimed at the

total synthesis of mitomycin K type of molecules through a known Danishefsky intermediate. However, there are severe challenges that remain unsolved. In this simplified model synthesis, though the stereochemistry of C9a methoxy group initially emerged in the undesired configuration, it was still possible to correct it by taking an extra step to epimerize this stereocenter. However, as in the case of

Danishefsky’s early efforts40 toward the total synthesis of mitomycin K, they were

unable to epimerize the incorrect C9a methoxy stereochemistry in the late

intermediates with the fictionalized aromatic or quinone systems, though enormous

effort was invested. This could be the biggest challenge for applying this method to

the total synthesis of mitomycins.

40. Danishefsky, S. J.; Schkeryantz, J. M. “Chemical Explorations Driven by An Enchantment with Mitomycinoids – A Twenty Year Account,” Synlett 1995, spec. issue, 475-490. 74

O O O O 1. ring closing metathesis O 7 steps 2. LiAlH4 N N H 46% 83% CO2t-Bu Isatin 105

O OH O OH O O 1. kinetic resolution 1. Oxone 2. recrystallization 2. Swern oxidation N 40%, 99% ee N 93% CO2t-Bu CO2t-Bu 106 107

O O O O 1. Me SiN O OMe 3 3 OMe HNO3 2. HCl, MeOH O MeOH N O 3. MeSO2Cl N NH N 4. PPh3 CO t-Bu 81% 2 36% 108 109 110

Figure 2.27 Miller’s Enantioselective Synthesis of Mitomycin Core Structure

CHAPTER 3

AN OVERVIEW OF SYNTHETIC STRATEGY

As described in Chapter 2, during the past forty years numerous synthetic approaches had been reported towards the mitomycin family of natural products.

However, even though we felt that many synthetic approaches to the mitomycins and related systems were effective at construction of the target heterocyclic ring system, certain deficiencies remained with respect to poor efficiency and with issues of stereocontrol in the construction of the parent hetereocyclic ring system, and in particular, the C9 and C9a stereogenic centers. In the course of planning our synthetic approach toward the mitomycins, we realized that a convergent and flexible route utilizing more modern organometallic methods would allow a general and efficient synthesis of mitomycins as well as some analogs of biological interest.

Examination of mitomycin structure revealed that retrosynthetic disconnection at the C9−C9a and N4−C4a bonds would divide the molecule into two major fragments: the aromatic fragment and the pyrrolidine fragment. (Figure 3.1)

O OCONH2 X 8a 9 OMe

9a 1 4a NNY 4 O

R OR X X MeO + RO N X CO R OR 2

Figure 3.1 Retosynthetic Analysis

While there was almost no literature precedent of previous approaches to mitomycin system by constructing the bond between nitrogen and aromatic ring in the late stage using relatively mild reaction conditions, we believed with the help of new organometalllic methods this could be achieved efficiently.

Besides of the obvious efficiency in the synthesis provided by the convergent design, our strategy provides more advantages over the previous approaches. First, the troublesome stereocontrol in the C9 and C9a stereo centers would no longer be a problem by the facile addition reaction of allylstanane or similar compounds to the pyrrolidine iminium ion formed by treating the pyrrolidine fragment with Lewis acids

with literature precedent that generally good stereocontrol is achieved; second, with the independent constructions of the two fragments, many functional group incompatibility problems in some of the previous approaches could be avoided; third, the aziridine functionality or its chiral precursor could be introduced earlier so as to avoid the late modification of the C1-C2 bond which was a typical problem of previous approaches; fourth, the early introduction of aziridine functionality could provide an enantioselective total synthesis of mitomycins, which has not been achieved yet.

However, our strategy will not install the methoxy group, a required functionality for mitomycins, at the C9a center. A late stage oxidation of C−H bond

to C−O bond needs to be conducted. Our efforts toward this transformation will be

discussed in Chapter 7.

CHAPTER 4

A CONVERGENT SYNTHESIS OF MITOMYCIN RING SYSTEM

4.1 Retrosynthetic Plan

Our first synthetic objective was the simplified mitomycin model compound

111 with the 9R*,9aR relationship1 in the tetrahydropyrrolo[1,2-a]indole ring system.

The B ring of 111 was initially envisioned to rise by a two-reaction process: an

palladium-catalyzed intramolecular Buchwald-Hartwig aryl amination2 of 112 between pyrrolidine amine and aryl triflate/halides proceeded by a stereoselective addition of allylstannane 113 to pyrrolidinone-derived N-acyliminium ion 114

1. The stereochemical series with a cis relationship between C9−H and C9a−H is denoted trans by Chemical Abstract Service. Before 1983, the 9R*,9aR series with a cis relationship between C9−H and C9a−H was denoted as cis. Because of this confusing reversal of notation, we have avoided using cis and trans altogether in naming our compounds. Thus, the 9R*,9aR stereoisomer has the C9-H and C9a-H on the same face of the molecule, whereas the 9R*,9aS stereoisomer has them on opposite sides.

2. For review, see: Hartwig, J. F. “Transition Metal Catalyzed Synthesis of Arylamines and Aryl Ethers from Aryl Halides and Triflates: Scope and Mechanism,” Angew. Chem. Int. Ed. 1998, 37, 2046-2067. 79

generated in situ by treatment of N-acyl-2-hydroxypyrrolidine with Lewis acid.3

(Figure 4.1)

OCONH2 9 H H AB9a N C N X H 111 112

Sn(n-Bu)3

+ N CO R X 2

113 114

Figure 4.1 Retrosynthetic Plan

4.2 Synthesis of N-Acyl-2-hydroxy-pyrrolidine and Allylstannane

The synthesis of N-acyl-2-hydroxypyrrolidine was quite straightforward and

well stated in the literature,4 though the initial attempts were made more difficult by

3. Example of Lewis acid catalyzed allylstannane addition to iminium ion: Klitzke, C. F.; Pilli, R. A. “Enhanced trans Diastereoselection in the Allylation of Cyclic Chiral N-Acyliminium Ions. Synthesis of Hydroxylated Indolizidines,” Tetrahedron Lett. 2001, 42, 5605-5608.

the broadened 1H NMR signals of compounds 115 and 116 caused by line broadening due to the slowly interconverting rotamers about the carbamate N-acyl bond.

LiEt BH (t-BuOCO)2O, Et3N 3 O N O N THF, −78 °C HO N CH2Cl2, 25 °C H CO2t-Bu CO2t-Bu 95% 91% 115

ClCO2Bn, n-BuLi LiEt3BH O O HO N THF, −78 °C N THF, −78 °C N H CO2Bn CO2Bn 86% 92% 116

Figure 4.2 Syntheses of N-Acyl-2-hydroxypyrrolidines

Many issues had been taken into consideration in choosing a suitable functional group on the aromatic ring to facilitate the aryl amination. In practice, both aryl halides5 and aryl triflates6 would react with a secondary amine, but we felt the

4. For example: Oba, M.; Terauchi, T.; Hashimoto, J.; Tanaka, T.; Nishiyama, K. “Stereoselective Synthesis of (2S,3S,4R,5S)-Proline-3,4,5-d3,”Tetrahedron Lett. 1997, 38,5515-5518.

5. Wolfe, J. P.; Buchwald, S. L. “An Improved Catalyst System for Aromatic Carbon-Nitrogen Bond Formation: The Possible Involvement of Bis(Phosphine) Palladium Complexes as Key Intermediates,” J. Am. Chem. Soc. 1996, 118, 7215-7216; Driver, M. S.; Hartwig, J. F. “A Second- Generation Catalyst for Aryl Halide Amination: Mixed Secondary Amines from Aryl Halides and Primary Amines Catalyzed by (DPPF)PdCl2,” J. Am. Chem. Soc. 1996, 118, 7217-7218.

6. Wolfe, J. P.; Buchwald, S. L. “Palladium-Catalyzed Amination of Aryl Triflates,” J. Org. Chem. 1997, 62, 1264-1267; Louie, J.; Driver, M. S.; Hamann, B. C.; Hartwig, J. F. “Palladium- Catalyzed Amination of Aryl Triflates and Importance of Triflate Addition Rate,” J. Org. Chem. 1997, 62, 1268-1273. 81

easy installation of triflate from the corresponding phenol would allow the synthesis

to have more flexibility to accommodate other functional groups.

Allylstannane 120 was synthesized starting from an E/Z mixture of 2-

propenylphenol 117. Triflation using freshly distilled triflic anhydride in the presence

of Et3N afforded phenol triflate 118 in almost quantitative yield. Thermodynamically

controlled photochemical bromination on the allylic position furnished trans-allyl

bromide 119 as a single regio- and stereoisomer. (Figure 4.3)

Existing methods of synthesizing allystannanes from allyl halides can be

summarized in three categories: radical stannylation,7 nucleophilic attack of an

8 9 allyllic metal complex to n-Bu3SnCl and reductive allylic stannylation. However

because of the aryl triflate functionality, none of those methods could be applied.

Palladium(0) catalyzed transformation to allylstannane, modified from reaction

condition for allylsilane formation,10 only provided allylstannane 120 in very low

yield. Mass spectrometry analysis revealed that the major product of this reaction was

a double stannylation compound resulting from the second stannylation at the aryl

triflate. Variation of reaction conditions to increase the yield was not successful.

7. Mitchell, T. N.; Reimann, W.; Nettelbeck, C. “Hydrostannation of 1,1-Bis(trimethylstannyl)- ethylene and Some Related Olefins: First Steps Toward Constructing A Karplus-Type Curve For 3J(Sn-Sn),” Organometallics 1985, 4, 1044-1048.

8. Lee, A. S.-Y.; Dai, W.-C. “A Facile and Highly Efficient Sonochemical Synthesis of Organo- stannane via Barbier Reaction,” Tetrahedron 1997, 53, 859-868.

9. Tabuchi, T.; Inanaga, J.; Yamaguchi, M. “A Mild and Efficient Method for the Formation of Allylstannanes Utilizing Samarium Iodide-Induced Polarity Inversion of π-Allyl Palladium Complexes,” Tetrahedron Lett. 1987, 28, 215-216.

10. Matsumoto, H.; Kasahara, M.; Matsubara, I.; Takahashi, M.; Arai, T.; Hasegawa, M.; Nakano, T.; Nagai, Y. “Conversion of Disilanes to Functional Monosilanes. XII. The Palladium(0)-Catalyzed Dechlorinative Silylation of Benzylic Chlorides with Methylchlorodisilanes. A Facile Route to Benzylmethylchlorosilanes,” J. Organomet. Chem. 1983, 250, 99-107. 82

(CF3SO2)2O, Et3N N-bromosuccinimide, hν CH2Cl2, −15−25 °C benzene, 80 °C OH OTf 99% 92% 117 118

Br Sn(n-Bu)3

(n-Bu3Sn)2, Pd(PPh3)4 neat, 120 °C OTf OTf 12% 119 120

Figure 4.3 Synthesis of Allylstannane

Similar reaction condition was used to make allylsilane 121. In contrast,

allylsilane was isolated in 56% yield. Lewis acid catalyzed addition of allysilane 121

with N-t-butylcarbamate (Boc) protected 2-hydroxypyrrolidine 115 proceeded

smoothly, however, to provide an undesired α-cross-coupling product 122. (Figure

4.4)

Br SiMe3

Me3SiSiMe3, Pd(PPh3)4 neat, 120 °C OTf OTf 56% 119 121

HO N 115 CO t-Bu 2 N BF ·Et O, CH Cl OTf 3 2 2 2 CO2t-Bu −78−25 °C 73% 122

Figure 4.4 Lewis Acid Catalyzed Addition Reaction of Allylsilane

4.3 Lewis Acid Catalyzed Addition of Allylstannane to Iminium Ion

Addition of allylstannane to 2-hydroxypyrrolidine 115 catalyzed by BF3⋅Et2O

proceeded smoothly to afford the allylically transposed compound 123 in 38% yield.

This moderate yield was partly due to the heavy contamination of stannane by- product inseparable from the desired allylstannane starting material. Again, because of the slowly interconverting rotamers about the carbamate N-acyl bond, the 1H NMR

signal of compound 123 were obscured by line broadening. This made it very difficult

to identify the diastereoselectivity and relative stereochemistry about the newly

formed C9-C9a bond. Actually, this was a global problem through the whole

synthesis. t-Butylcarbamate protecting group removal was achieved by treating

compound 123 with trifluoroacetic acid in CH2Cl2 to afford pyrrolidine 124 in almost

quantitative yield. No double bond migration product was detected.

Attempts to close the mitosane B ring by intramolecular Buchwald-Hartwig

reaction on compound 124 revealed very complex results. While the mass

spectrometric analysis of the crude of reaction mixture had continuously suggested

that the cyclized compound was formed, variation of reaction conditions did not

provide the desired mitosane compound 125 at all, but only an unidentified mixture of

compounds. (Figure 4.5)

Sn(n-Bu)3

HO N 115 CO2t-Bu N

OTf BF3·Et2O, CH2Cl2 OTf CO2t-Bu −78−25 °C 120 38% 123

CF3COOH N x CH Cl , 25 °C 2 2 OTf HN 99% 124 125

Figure 4.5 Allylstannane Addition and Attempted Cyclization

A rationale for the failure of cyclization might be that the reaction conditions

(relatively strong base and high temperature) were too harsh for a homobenzylic terminal double bond. Thermodynamically controlled double bond migration might have happened and thus complicated the reaction results. Transformation of the terminal double bond to the required alcohol functionality prior the cyclization seemed to be a key to the synthesis.

4.4 Synthesis of Silyl Enol Ether and Its Addition to Iminium Ion

It became necessary to make some modifications to our initial synthetic plan

with the poor yield of the transallylation addition of allylsilane to iminium ion and the

low efficiency in the allylstannane synthesis. The timing to convert double bond to

alcohol also need be taken into consideration. We felt a synthesis starting from the

silyl enol ether would possibly be more efficient than from the allylstannane

precursor.

Trimethylsilyl enol ether 128 was synthesized starting from 2-allylphenol.

Protection as a benzyl ether 126 was achieved most efficiently using benzyl bromide

and potassium carbonate in refluxing acetone. The primary reason for choosing benzyl protecting group here is that it could be efficiently removed later in one step together with benzylcarbamate protecting group. Standard ozonolysis cleaved the terminal double bond to furnish aldehyde 127, which was transformed to the unstable trimethylsilyl enol ether 128 by following Duhamel’s protocol.11 (Figure 4.6) The

overall efficiency for this 3-step operation was 54% based on the 2-allylphenol.

11. Duhamel, P.; Hennequin, L.; Poirier, J. M.; Tavel, G.; Vottero, C. “1,5-Dicarbonyl Compounds. A General Preparation Method,” Tetrahedron 1986, 42, 4777-4786.

BnBr, K2CO3 1. O3, CH2Cl2, −78 °C

acetone, 58 °C 2. Me2S, CH2Cl2, 25 °C OH OBn OBn

126 127

OSiMe3 116 O HO N

Me3SiCl, Et3N CO2Bn H MeCN, 0 °C Me3SiOTf, CH2Cl2 N OBn OBn 54% from −78-25 °C CO2Bn 2-allylphenol 128 96% 129

Figure 4.6 Synthesis of Silyl Enol Ether and Its Addition Reaction

Initial attempts to perform a Mukaiyama type coupling12 between silyl enol ether 128 and 2-hydroxypyrrolidine 116 using BF3⋅Et2O as the Lewis acid were

discouraging since there was no indication the desired corresponding coupling

product aldehyde 129 was formed (via mass spectrometric analysis). However the

detection of a Mukaiyama aldol reaction product between silyl enol ether 128 and

aldehyde 127 provided some insights into this reaction process. The silyl enol ether

128 was unstable to even the mild Lewis acid BF3⋅Et2O, which effected a quick

12. Mukaiyama, T.; Banno, K.; Narasaka, K. “New Cross-Aldol Reactions. Reactions of Silyl Enol Ethers with Carbonyl Compounds Activated by Titanium Tetrachloride,” J. Am. Chem. Soc. 1974, 96, 7503-7509.

conversion back to aldehyde 127.13 The activation of 2-hydroxypyrrolidine by

BF3⋅Et2O was apparently too mild to permit effective formation of the iminium ion

needed for a cross-coupling reaction. This hypothesis suggested a stronger Lewis acid

might be helpful for this reaction. Indeed, carefully tuning the catalyst Lewis acidity

revealed that an excellent 96% yield of the coupling reaction was obtained when a

strong Lewis acid trimethylsilyl triflate was used as the catalyst.

Establishment of the relative stereochemistry and the reaction

diastereoselectivity was not possible at this juncture due to the broadened room

temperature 1H NMR signal as a typical result of N-carbamate protection, although

high temperature NMR signal suggested the reaction was diastereoselective. The

stereochemistry and diastereoselectivity were established later by following the

cyclization to the tetrahydropyrroloindole (vide infra).

4.5 Mitosane Ring Cyclization

The reduction of aldehyde 129 turned out to be more difficult than expected.

Typical aldehyde reduction reagents such as NaBH4 and i-Bu2AlH failed to reduce

the aldehyde, even under prolonged reaction time and ambient reaction temperature.

A powerful reducing agent Super Hydride (LiEt3BH, THF, −78 °C) was used and the

corresponding alcohol 130 was isolated in almost quantitative yield. (Figure 4.7)

13. Because of the difficulty associated with the purification of silyl enol ether, there might be some amount of water remained in the silyl enol ether compound. 89

Carrying the primary alcohol through the cyclization process as the

unprotected form was also planned, however this tactic was abandoned because of literature precedent that under the Buchwald-Hartwig aryl amination condition, alcohols can also participate aryl ether formation.14 Although no previous evidence

indicating O-arylation is favored over N-arylation or vice versa, at that time we felt in

our system O-arylation might be favored because of steric factors (i.e., primary

alcohol vs. secondary pyrrolidine amine).

A triisopropylsilyl ether was chosen to protect the primary alcohol, primarily

because of its stability and well established diverse removal conditions.15 Standard

16 silyl ether formation condition (i-Pr3SiOTf, 2,6-lutidine, CH2Cl2, 0 °C) afforded

silyl ether 131 in 92% yield. Removal of the O-benzyl ether and N-benzylcarbamate

(N-Cbz) protecting groups by hydrogenolysis (1 atm H2, Pd/C, MeOH) afforded the corresponding aminophenol 132 that was used as it was obtained from the reaction

mixture. Proton NMR spectroscopy revealed that compound 132 existed as a >6:1

ratio of spectroscopically distinguishable mixture of diastereomers. This was the most

direct evidence for the level of diastereoselectivity in the addition process of silyl

14. Hartwig, J. F. “Transition Metal Catalyzed Synthesis of Arylamines and Aryl Ethers from Aryl Halides and Triflates: Scope and Mechanism,” Angew. Chem. Int. Ed. 1998, 37, 2046-2067; Mann, G.; Incarvito, C.; Rheingold, A. L.; Hartwig, J. F. “Palladium-Catalyzed C-O Coupling Involving Unactivated Aryl Halides. Sterically Induced Reductive Elimination to Form the C-O Bond in Diaryl Ethers,” J. Am. Chem. Soc. 1999, 121, 3224-3225; Palucki, M.; Wolfe, J. P.; Buchwald, S. L. “Synthesis of Oxygen Heterocycles Via A Palladium-Catalyzed C-O Bond- Forming Reaction,” J. Am. Chem. Soc. 1996, 118, 10333-10334; Palucki, M.; Wolfe, J. P.; Buchwald, S. L. “Palladium-Catalyzed Intermolecular Carbon-Oxygen Bond Formation: A New Synthesis of Aryl Ethers,” J. Am. Chem. Soc. 1997, 119, 3395-3396.

15. Kocienski, P. J. “Protecting Groups,” G. Thieme, New York, 1994; Greene, T. W.; Wuts, P. G. M. “Protective Groups in Organic Synthesis,” 3rd ed., Wiley, New York, 1999.

16. Corey, E. J.; Cho, H.; Rüecker, C.; Hua, D. H. “Studies with Trialkylsilyltrifluoromethyl- sulfonates: New Syntheses and Applications,” Tetrahedron Lett. 1981, 22, 3455-3458. 90

enol ether to the N-acyliminium ion. Unfortunately this mixture of diastereomers were chromatographically inseparable. (Figure 4.7)

O OH

LiEt3BH i-Pr3SiOTf, 2,6-lutidine H H N THF, −78 °C N CH2Cl2, 25 °C OBn OBn CO2Bn 99% CO2Bn 92% 129 130

OSi(i-Pr)3 OSi(i-Pr)3

H2, Pd/C H N MeOH, 25 °C H N H OBn CO2Bn OH 131 132

Figure 4.7 Manipulations of Protecting Groups

With the phenol and free secondary pyrrolidine amine in place, we next investigated the installation of the phenol triflate functionality, required for the key

Buchwald-Hartwig intramolecular aryl amination. Generally a secondary amine is more nucleophilic than a regular phenol to favor N- over O-triflation. This was proved true when relatively weak bases such as Et3N, pyridine or 2,6-lutidine were

used together with triflic anhydride ((CF3SO2)2O). Stronger bases like NaH and

LiN(SiMe3)2 revealed a complex situation with almost no selectivity. We realized this

selectivity change, from N-triflation to no selectivity, was caused by a simple fact that 91

strong bases could completely deprotonate the phenol and form a more reactive phenolate anion. We suspected the “hard” electrophile N-phenyltriflimide (PhNTf2)

would favor reaction at the presumed “hard” phenolate nucleophile. Indeed, treatment of aminophenol 132 with NaN(SiMe3)2 and PhNTf2 in THF furnished phenyl triflate

133 cleanly (83% from carbamate 131). (Figure 4.8) Phenyl triflate 133 set the stage

for the final operation required for building the targeted tetrahydropyrroloindole

system.

OSi(i-Pr)3 OSi(i-Pr)3

NaN(SiMe3)2, PhN(CF3SO2)2 H H N THF, 0 °C N OH H OTf H 83% from carbamate 131 132 133

Figure 4.8 Selective O- vs. N-Triflation

Attempts for palladium-catalyzed intramolecular aryl amination17 were

successful to furnish mitosane 134, but only in moderate yields. (Figure 4.9) This is

partly due to the thermodynamically favored air oxidation of the unstable

dihydroindole compound to stable aromatic indole compound. Color changes from

17. Wolfe, J. P.; Buchwald, S. L. “Palladium-Catalyzed Amination of Aryl Triflates,” J. Org. Chem. 1997, 62, 1264-1267; Louie, J.; Driver, M. S.; Hamann, B. C.; Hartwig, J. F. “Palladium- Catalyzed Amination of Aryl Triflates and Importance of Triflate Addition Rate,” J. Org. Chem. 1997, 62, 1268-1273.

yellow to green, blue and darker were observed during chromatography, which

indicated the progressing oxidation that occurred on silica gel.

OSi(i-Pr)3 OSi(i-Pr)3 Pd(OAc)2, BINAP H H N Cs2CO3, Toluene, 100 °C OTf H N 44% 133 134

Figure 4.9 Palladium-Catalyzed Cyclization

Surprisingly, mitosane compound 134 was sensitive to both acids and bases. It decomposed immediatedly in untreated CDCl3; treatment with n-Bu4NF in THF in

order to remove the silyl protecting group also lead to decomposed products. Mild

desilylation condition (HF⋅pyridine, THF, 25 °C)18 furnished primary alcohol 135 in

83% yield. Final installation of the carbamate functionality by following literature

protocol19 afforded mitosane 111, which marked the completion of the synthesis.20

(Figure 4.10)

18. Wipf, P.; Kim, H. “Total Synthesis of Cyclotheonamide A,” J. Org. Chem. 1993, 58, 5592-5594.

19. Dijksman, W. C.; Verboom, W.; Egberink, R. J. M.; Reinhoudt, D. N. “Synthesis of Mitomycin C Analogs. 1. Introduction of the Urethane Function at C-10 of the Pyrrolo[1,2-a]indole Skeleton,” J. Org. Chem. 1985, 50, 3791-3797.

20. Coleman, R. S.; Chen, W. “A Convergent Approach to the Mitomycin Ring System,” Org. Lett. 2001, 3, 1141-1144.

OSi(i-Pr)3 OH H HF⋅pyridine H N THF, 25 °C N 83% 134 135

OCO2Ph OCONH2 H H PhOCOCl NH3

pyridine, 25 °C N CH2Cl2, −40 °C N 59% 136 2 steps 111

Figure 4.10 Completion of the Convergent Approach to Mitomycin Ring System

4.6 The Determination of The Relative Stereochemistry

It was possible to separate the two diastereomers that were formed in the addition step of silyl enol ether to N-acyliminium ion at the stage of mitosane 111.

The major isomer was recrystallized from hot ethyl acetate; the minor isomer was separated by preparative thin-layer chromatography.

The relative stereochemistry of the C9 and C9a stereogenic centers of the major diastereomer of mitosane 111 was determined to be 9R*,9aR by correlation

with published 1H and 13C NMR spectral data.21 In addition, nuclear Overhauser enhancement studies on both diastereomers were consistent with and confirmed unequivocally the proposed stereochemical assignments. Figure 4.11 shows energy- minimized structures (MMX force field) of the major and minor diastereomers of

111. A good correlation was observed between the calculated proton-distances and experimental observation of the strength of the nOe effect. (Figure 4.12) In the major stereoisomer [(9R*,9aR)-111], the most definitive enhancements were between

C10−H and C1−H (4.4%) and between C9−H and C9a−H (5.7%). Notably absent were enhancements between C9−H and either C1−H and between the C10−H's and

C9a−H. In the minor diastereomer [(9R*,9aS)-111], the definitive enhancements were between C9−H and C1−Hα (3.4%) and between C9a−H and C10−H (2.7%).

Completely absent were enhancements between C10−H's and either C1−Hα or

C1−Hβ. These nuclear Overhauser enhancement studies served to definitively

establish the relative stereochemistry at the C9 and C9a stereogenic centers.

21. Dijksman, W. C.; Verboom, W.; Egberink, R. J. M.; Reinhoudt, D. N. “Synthesis of Mitomycin C Analogs. 1. Introduction of the Urethane Function at C-10 of the Pyrrolo[1,2-a]indole Skeleton,” J. Org. Chem. 1985, 50, 3791-3797. 95

4.4%

10 1

9a 9

5.7% 10 OCONH2 9 H (9R*,9aR)-111 [major] 9a 1 N

2.7%

9a 9

3.4% 10 OCONH2 9 H (9R*,9aS)-111 [minor] 9a 1 N

Figure 4.11 Illustrations of Important Observed NOE Effect

H10 10 H OCONH2 9 H H9a H1β 1α N H

(9R*, 9aR)-111

irradation observed nOe effect (calculated proton-distance) C1-Hα C1-Hβ C9-H C9a-H C10-H C10-H C1-Hα   0.0% (3.7Å) 6.7% (2.5Å) 1.1% (2.7Å) 0.0% (3.5Å) C1-Hβ   0.0% (3.8Å) 0.0% (3.1Å) 4.4% (2.3Å) 1.5% (2.7Å) C9-H 0.0% (3.7Å) 0.0% (3.8Å)  4.9% (2.3Å) 1.3% (3.1Å) 3.3% (2.6Å) C9a-H 4.0% (2.5Å) 0.8% (3.1Å) 5.7% (2.3Å)  0.0% (3.9Å) 0.0% (3.5Å) C10-H 0.5% (2.7Å) 1.8% (2.7Å) 5.0% (2.6Å) 0.0% (3.9Å)   C10-H 0.0% (3.5Å) 2.2% (3.1Å) 2.2% (3.1Å) 0.0% (3.5Å)  

H10 10 H OCONH2 9 H H9a H1β 1α N H

(9R*, 9aS)-111

irradation observed nOe effect (calculated proton-distance) C1-Hα C1-Hβ C9-H C9a-H C10-H C10-H C1-Hα   5.0% (2.4Å) 0.0% (3.1Å) 0.0% (4.1Å) 0.0% (4.8Å) C1-Hβ   0.0% (3.1Å) 6.5% (2.5Å) 0.0% (3.5Å) 0.0% (4.5Å) C9-H 3.4% (2.4Å) 0.0% (3.8Å)  1.1% (3.0Å) 7.4% (2.6Å) 0.0% (3.1Å) C9a-H 0.0% (3.1Å) 1.7% (2.5Å) 1.4% (3.0Å)  2.7% (2.6Å) 1.5% (2.6Å) C10-H 0.0% (4.1Å) 0.0% (4.8Å) 0.0% (3.1Å) 4.4% (2.6Å)   C10-H 0.0% (3.5Å) 0.0% (4.5Å) 4.5% (2.6Å) 1.7% (2.6Å)  

Figure 4.12 Correlations between Observed NOE Effect and Calculated Proton-

Distances

4.7 Rationale for the Observed Diastereoselection

A rationale for the observed diastereoselection in the addition of enol ether

128 with the iminium ion derived from 116 is shown in Figure 4.13.22 Considering

the possible orientations of the two reactive π-systems, the least sterically crowded arrangement is the synclinal orientation sc1 that leads to the desired stereochemical

array in the addition product. The alternative synclinal orientations sc2, sc3, and sc4 place the pyrrolinium ring over the aromatic ring or silyl enol ether, making all of these orientations nonviable energetically. Both antiperiplanar orientations suffer steric crowding, severely in ap1 between pyrrolinium and aromatic rings, or to a

lesser degree in ap2 between the carbamoyl group and the O-benzyl phenol on the

aromatic ring. Thus, from this analysis, it would appear that the two least crowded

orientations are sc1 and ap2, where the sc1 orientation predominates to afford the

observed diastereoselection. The minor diastereomer presumably arises from

orientation ap2. A bulkier carbamate such as t-butylcarbamate might help to reduce

the formation of this minor diastereomer.

22. Russowsky, D.; Petersen, R. Z.; Godoi, M. N.; Pilli, R. A. “Addition of Silylated Carbon Nucleophiles to Iminium and Cyclic N-Acyliminium Ions Promoted by InCl3,” Tetrahedron Lett. 2000, 41, 9939-9942; Maldaner, A. O.; Pilli, R. A. “Addition of Carbon Nucleophiles to Substituted N-Acyliminium Ions. A Stereoselective Route to Trans-Fused Decahydroquinoline Systems,” Tetrahedron Lett. 2000, 41, 7843-7847; Pilli, R. A.; Russowsky, D. “The Stereochemistry of the Addition of Carbon Nucleophiles to Imines and Iminium Ions,” Trends Org. Chem. 1997, 6, 101-123; Yamamoto, Y.; Komatsu, T.; Maruyama, K. “Diastereofacial Selectivity in the Reaction of Allylic Organometallic Compounds with Imines. Stereoelectronic Effect of Imine Group,” J. Org. Chem. 1985, 50, 3115-3121. 98

OR OHC CO R CO2R H 2 H N Ar N Ar H favored sc1

OR OR OR RO C 2 CO R RO2C N N H N 2 Ar H Ar H Ar H H H sc2 sc3 sc4

OR OR H H Ar H Ar H N N CO2R RO2C

1 2 ap ap

Figure 4.13 Origin of Diastereoselectivity

4.8 Experimental23

H N OBn CO2Bn 2-[1-(2-Benzyloxyphenyl)-2-oxoethyl]pyrrolidine-1-

carboxylic acid benzyl ester (129).24 Trimethylsilyl triflate (0.49 mL, 2.7 mmol) was

slowly added to a solution of 128 (587 mg, 2.66 mmol) in CH2Cl2 (10 mL) at −78 °C.

The reaction mixture was stirred for 1 h at −78 °C and a solution of 116 (804 mg,

2.70 mmol) in CH2Cl2 (2.7 mL) was added. The reaction mixture was stirred for 15 min at −78 °C and for 2 h at 25 °C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH

7, 10 mL) was added. The aqueous layer was extracted with EtOAc (3 × 10 mL) and

the combined organic extracts were dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (5 × 20 cm

silica, 10% EtOAc/hexane) to afford pure 129 (1.095 g, 96%) as a pale yellow oil: 1H

NMR (CDCl3, 500 MHz) δ 9.90-9.67 (br, 1H), 7.45-7.09 (br, 11H), 7.09-6.82 (br,

3H), 5.17-4.85 (br, 4H), 4.73-4.67 (br, 1H), 4.25-4.01 (br, 1H), 3.53-3.02 (br, 2H),

13 2.11-1.34 (br, 4H); C NMR (CDCl3, 125 MHz) δ 199.69, 170.72, 155.45, 141.11,

23. All 1H and 13C NMR spectra were recorded on a Brüker DRX-500. Chemical Shifts are reported 1 13 in ppm relative to CDCl3 peak at 7.24 ppm ( H) or 77.0 ppm ( C). All mass spectroscopy was performed by the Campus Chemical Instruments Center at The Ohio State University on a Micromass Q-TQF2 instrument. Infrared (IR) spectra were recorded on a Perkin-Elmer 1600 FT- IR spectrometer. All reactions unless otherwise specified were run under an inert atmosphere of nitrogen or argon using clean, dried glassware. All solvents were freshly distilled before use. Yields reported refer to isolated material determined to be pure by NMR spectroscopy and thin layer chromatography (TLC) unless specified in the text.

24. Naming of all compounds was provided by AutoNom 2.1, a software contained in ChemOffice 6.0.1. 100

136.88, 130.86, 128.33, 128.18, 128.00, 127.96, 127.82, 127.64, 127.53, 127.13,

127.00, 126.93, 126.56, 123,53, 121.08, 112.26, 70.15, 66.44, 57.13, 56.18, 46.13,

28.00, 23.24; HRMS (ESI) m/z 452.1846 (calcd. for C27H27NO4 + Na: 452.1838).

H N OBn CO2Bn 2-[1-(2-Benzyloxyphenyl)-2-hydroxyethyl]pyrrolidine-1-

carboxylic acid benzyl ester (130). Super Hydride (1.0 M solution in THF, 0.038

mL, 0.038 mmol) was slowly added to a solution of 129 (14.8 mg, 0.034 mmol) in

THF (1.5 mL) at −78 °C. The reaction mixture was stirred for 30 min at −78 °C and

aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 2 mL) was added. The aqueous layer

was extracted with EtOAc (3 × 3 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (0.5 × 10 cm silica, 10% EtOAc/hexane) to afford pure 130

1 (14.8 mg, 99%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 7.64-7.56 (br, 1H),

7.43-7.27 (br, 9H), 7.20-7.16 (br, 1H), 6.98-6.93 (br, 1H), 6.93-6.89 (br, 1H), 5.22-

5.12 (br, 2H), 5.06-4.97 (br, 2H), 4.50-4.43 (br, 1H), 3.89-3.81 (br, 1H), 3.62-3.54

(br, 1H), 3.52-3.32 (br, 2H), 3.28-3.21 (br, 1H), 1.94-1.64 (br, 3H), 1.55-1.46 (br,

13 1H); C NMR (CDCl3, 125 MHz) δ 157.15, 156.42, 141.04, 136.66, 128.53, 128.23,

128.19, 128.14, 127.99, 127.75, 127.71, 127.56, 127.45, 126.93, 126.87, 126.81,

101

126.50, 120.57, 111.85, 70.02, 66.98, 63.00, 57.49, 47.47, 43.00, 28.82, 23.28;

HRMS (ESI) m/z 454.1991 (calcd. for C27H29NO4 + Na: 454.1995).

OSi(i-Pr)3

H N OBn CO2Bn 2-[1-(2-Benzyloxyphenyl)-2-triisopropylsilanyloxy-

ethyl]pyrrolidine-1-carboxylic acid benzyl ester (131). Triisopropyl

trifluoromethanesulfonate (0.54 mL, 2.0 mmol) was slowly added to a solution of 130

(403.4 mg, 0.94 mmol) and 2,6-lutidine (0.35 mL, 3.0 mmol) in CH2Cl2 (5 mL) at 0

°C. The reaction mixture was stirred for 6 h at 0 °C and saturated aqueous NaHCO3

(5 mL) was added. The aqueous layer was extracted with EtOAc (3 × 10 mL) and the combined organic extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash chromatography (5 × 20 cm

silica, 10% EtOAc/hexane) to afford pure 131 (515.2 mg, 92%) as a colorless oil: 1H

NMR (CDCl3, 500 MHz) δ 7.42-7.25 (br, 10H), 7.16-7.08 (br, 2H), 6.91-6.79 (br,

2H), 5.15-4.82 (br, 4H), 4.43-4.23 (br, 1H), 4.11-3.83 (br, 2H), 3.73-3.02 (br, 3H),

13 1.88-1.46 (br, 4H), 1.02-0.68 (br, 21H); C NMR (CDCl3, 125 MHz) δ 157.30,

155.49, 137.39, 137.02, 129.71, 128.43, 128.36, 128.29, 128.20, 127.97, 127.72,

127.64, 127.36, 127.28, 127.22, 127.17, 127.09, 120.83, 112.06, 70.35, 66.56, 65.51,

59.62, 46.43, 28.60, 23.40, 22.57, 17.87, 11.91; IR (neat) νmax 3449.3, 2941.7,

102

2890.1, 1681.7, 1452.1, 1415.0, 1237.5, 1110.6, 882.5, 751.1, 695.6 cm-1; HRMS

(ESI) m/z 610.3316 (calcd. for C36H49O4NSi + Na: 610.3329).

OSi(i-Pr)3

H N OH H 2-(1-Pyrrolidin-2-yl-2-triisopropylsilanyloxyethyl)phenol

(132). A slurry of 10% Pd on activated carbon (10 mg, catalytic) was added in one

portion to a solution of 131 (515.2 mg, 0.87 mmol) in methanol (10 mL) at 25 °C.

The reaction mixture was stirred for 12 h at 25 °C under 1 atm H2. The mixture was

filtered and the filtrate was evaporated in vacuo to give 132 as a pale yellow heavy oil

1 (363 mg, 99%), that was used without further purification: H NMR (CDCl3, 500

MHz) δ 7.12-7.06 (app t, 1H, J = 7.9 Hz), 6.93-6.86 (m, 2H), 6.78-6.73 (app t, 3H, J

= 12.2, 6.7 Hz), 4.10-3.96 (m, 2H), 3.48-3.44 (app q, 1H, J = 10.3, 3.7 Hz), 3.02-2.90

(m, 1H), 2.80-2.71 (m, 1H), 1.86-1.68 (m, 2H), 1.67-1.58 (m, 1H). 1.56-1.45 (m, 1H),

1.48-0.90 (m, 21H).

OSi(i-Pr)3

H N OTf H 2-(1-Pyrrolidin-2-yl-2-triisopropylsilanyloxyethyl)phenyl

trifluoromethanesulfonic acid ester (133). A solution of NaN(SiMe3)2 (0.6 M in

toluene, 1.40 mL, 0.84 mmol) was slowly added to a solution of 132 (253.7 mg, 0.70

103

mmol) in THF (2 mL) at 0 °C. The reaction mixture was stirred for 30 min at 0 °C

and PhNTf2 (275 mg, 0.77 mmol) was added. The mixture was stirred for 1 h at 0 °C

and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 3 mL) was added. The aqueous layer was extracted by EtOAc (3 × 5 mL) and the combined organic extracts were

dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was

purified by flash chromatography (2 × 20 cm silica, EtOAc) to afford pure 133 (288

1 mg, 83%) as a pale yellow oil: H NMR (CDCl3, 500 MHz) δ 7.49 (d, J = 7.9 Hz,

1H), 7.32-7.22 (m, 3H), 4.04 (dd, J = 10.4, 5.5 Hz, 1H), 3.94 (dd, J = 9.2, 5.5Hz, 1H),

3.62-3.52 (m, 1H), 3.08-2.98 (m, 2H), 2.98-2.91 (m, 1H), 1.78-1.70 (m, 1H), 1.68-

13 1.58 (m, 2H), 1.28-1.18 (m, 1H), 1.05-0.88 (m, 21H); C NMR (CDCl3, 125 MHz)

δ 147.90, 135.25, 130.48, 128.01, 127.89, 121.05, 119.10 (q, J = 321 Hz, CF3), 65.68,

19 59.90, 46.55, 30.46, 25.61, 25.06, 17.87, 11.84; F NMR (CDCl3, 235 MHz, CFCl3 =

0.00 ppm) δ 74.39; IR (film) νmax 2944.6, 2987.1, 2363.4, 1593.5, 1489.9, 1463.1,

1419.5, 1287.5, 1248.6, 1214.0, 1140.6, 1005.9, 889.7, 768.2, 682.7 cm-1; HRMS

(ESI) m/z 496.2161 (calcd. for C22H36SO4F3SiN + H: 496.2164).

OSi(i-Pr)3 H

N 8-Triisopropylsilanyloxymethyl-2,3,8,8a-tetrahydro-1H-

3a-azacyclopenta[a]indene (134). A mixture of 133 (20 mg, 0.04 mmol), Pd(OAc)2

(9.0 mg, 0.04 mmol), BINAP (racemic, 31.1 mg, 0.05 mmol), and Cs2CO3 (19.5 mg,

104

0.06 mmol) in toluene (0.2 mL) in a vial equipped with a Teflon-lined lid was

warmed to 100 °C for 18 h and was allowed to cool to 25 °C. The reaction mixture was filtered and the filtrate was evaporated in vacuo. The residue was purified by flash chromatography (0.5 × 7 cm silica, hexane) to afford 14 (6.1 mg, 44%) as a pale

1 yellow oil: H NMR (CDCl3, 500 MHz) δ 7.16-7.08 (m, 2H), 6.75 (app t, J = 7.9 Hz,

1H), 6.58 (d, J = 7.9 Hz, 1H), 4.20 (dd, J = 10.4, 6.1 Hz, 1H), 4.06-4.00 (m, 1H), 3.95

(app t, J = 9.2 Hz, 1H), 3.65 (app dd, J = 14.6, 9.2 Hz, 1H), 3.45 (m, 1H), 3.14 (m,

1H), 1.97-1.82 (m, 2H), 1.82-1.76 (m, 1H), 1.48-1.40 (m, 1H), 1.22-1.05 (m, 21H);

13 C NMR (CDCl3, 125 MHz) δ 155.18, 130.33, 128.01, 124.03, 118.93, 110.48,

69.04, 63.63, 51.75, 45.42, 26.11, 24.96, 18.10, 12.05; IR (film) νmax 2940.9, 2864.7,

2360.1, 1603.7, 1458.1, 1092.2, 881.7, 781.3, 751.7 cm-1; HRMS (ESI) m/z 346.2582

(calcd. for C21H35NOSi + H: 345.2566).

OH H

N (2,3,8,8a-Tetrahydro-1H-3a-azacyclopenta[a]inden-8-yl)-

methanol 135. A dilute HF⋅pyridine solution (0.18 mL, consisting of HF⋅pyridine

(Aldrich, 0.5 mL) diluted with 2 ml dry pyridine at −20 °C in a plastic bottle under

N2) was added dropwise to a solution of 134 (20 mg, 0.058 mmol) in THF (0.58 mL) in a plastic reaction vessel at 25 °C. The reaction was stirred for 48 h at 25 °C and cold saturated aqueous NaHCO3 (3 mL) was added. The aqueous layer was extracted

105

with EtOAc (3 × 3 mL) and the combined organic extracts were dried (Na2SO4),

filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash

chromatography (0.5 × 10 cm silica, 33% EtOAc/hexane) to afford pure 135 (9.1 mg,

1 83%) as a pale yellow liquid: H NMR (CDCl3, 500 MHz) δ 7.17 (app t, J = 7.9 Hz,

1H), 7. 13 (d, J = 7.9 Hz, 1H), 6.80 (app t, J = 7.9 Hz, 1H), 6.63 (d, J = 7.9 Hz, 1H),

4.19 (dd, J = 11.0, 5.5 Hz, 1H), 4.12-4.06 (m, 1H), 3.98 (dd, J = 11.0, 8.5Hz, 1H),

3.67 (app dd, J = 14.7, 8.5 Hz, 1H), 3.50 (m, 1H), 3.17 (m, 1H), 2.02-1.88 (m, 2H),

13 1.85-1.79 (m, 1H), 1.79-1.68 (br s, 1H, OH), 1.52 (m, 1H); C NMR (CDCl3, 125

MHz) δ 155.05, 129.62, 128.23, 123.85, 119.14, 110.69, 68.76, 62.81, 51.74, 45.01,

26.16, 24.98; HRMS (ESI) m/z 190.1232 (calcd. for C12H15NO + H: 190.1231).

OCONH2 9 H

9a N Pyrrolo[1,2-a]indole 111 Phenyl chloroformate (44 mg, 0.28

mmol) was added to a solution of 135 (44 mg, 0.23 mmol) in dry pyridine (2 mL) at 0

°C. The reaction mixture was stirred for 18 h at 25 °C and water (2 mL) was added.

The aqueous layer was extracted with EtOAc (3 × 5 mL) and the combined organic

extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The

residue was dissolved in CH2Cl2 (2 mL) and was cooled to −40 °C whereupon liquid

NH3 (5 mL) was added. After refluxing for 6 hours at −40 °C, the reaction mixture

was warmed to 25 °C and NH3 was evaporated. The reaction mixture was diluted

106

with CHCl3 (15 mL) and EtOAc (3 mL) and was washed with aqueous NaOH (4%,

10 mL) and saturated aqueous NaCl (10 mL). The organic layer was dried (Na2SO4),

filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash

chromatography (0.5 × 10 cm silica, EtOAc) to afford 111 as a mixture of

diastereomers as a colorless solid (32 mg, 59%). The major isomer (9R*,9aR)-111

was isolated by recrystallization (EtOAc). The minor isomer (9R*,9aS)-111 was

isolated by preparative TLC (10 × 20 cm silica PF254, 50 µm, EtOAc).

1 (9R*,9aR)-111: H NMR (CDCl3, 500 MHz) δ 7.11 (app t, 1H, J = 8.5, 7.9 Hz), 7.07

(d, 1H, J = 7.9 Hz), 6.74 (dd, 1H, J = 7.9, 7.3 Hz), 6.56 (d, 1H, J = 7.9 Hz), 4.64 (br s,

2H, NH2), 4.55 (dd, 1H, J = 6.1, 5.5 Hz, C10-H), 4.29 (dd, 1H, J = 8.5, 5.5 Hz, C10-

H), 3.96 (ddd, 1H, J = 9.8, 8.5, 5.5 Hz, C9a-H), 3.71 (ddd, 1H, J = 9.6, 8.5, 6.1 Hz,

C9-H), 3.44 (ddd, 1H, J = 10.7, 8.8, 3.7 Hz, C3-H), 3.12 (ddd, 1H, J = 10.7, 8.5, 7.7

Hz, C3-H), 1.78-1.95 (m, 2H, C2-H), 1.68 (m, 1H, C1-H), 1.36 (m, 1H, C1-H); 13C

NMR (CDCl3, 125 MHz) δ 156.58, 154.81, 128.98, 128.41, 124.03, 119.35, 110.76,

68.64, 64.94, 51.80, 41.78, 26.00, 25.07; IR (film) νmax 2984.5, 1741.4, 1447.8,

-1 1373.7, 1239.8, 1047.1 cm ; HRMS (ESI) m/z 233.1294 (calcd. for C13H16N2O2 + H:

233.1290).

1 (9R*,9aS)-111: H NMR (CDCl3, 500 MHz) δ 7.12 (m, 1H, 2H), 6.75 (app t, 1H, J =

7.3 Hz), 6.59 (d, 1H, J = 7.9 Hz), 4.60 (br s, 2H, NH2), 4.23 (dd, 1H, J = 10.4, 6.4

Hz, C10-H), 4.13 (dd, 1H, J = 11.9, 6.4 Hz, C10-H), 3.71 (ddd, 1H, J = 9.5, 6.4, 3.1

107

Hz, C9a-H), 3.50 (m, 1H, C9-H), 3.40 (m, 1H, C3-H), 3.13 (m, 1H, C3-H), 1.91 (m,

13 1H, C1-H), 1.84 (m, 2H, C2-H), 1.35 (m, 1H, C1-H); C NMR (CDCl3, 125 MHz) δ

156.65, 154.91, 129.60, 128.77, 125.25, 119.40, 111.35, 68.82, 67.82, 51.98, 46.55,

30.83, 25.68.

108

CHAPTER 5

STEREOSELECTIVE SYNTHESES OF MITOSANES AND

MITOSENES

5.1 Retrosynthetic Analysis

After the stereoselective and convergent synthesis of the parent mitosane ring

was successfully accomplished, we moved on to our next target: the stereoselective

synthesis of fully functionalized mitosanes. The success of Lewis acid catalyzed silyl

enol addition to a pyrrolidine derived iminium ion and the efficiency of the

intramolecular Buchwald-Hartwig aryl amination in the earlier synthesis provided an

exciting chance to probe the synthesis of the more complicated system. The initial

restrosynthetic plan is detailed in Figure 5.1. This plan envisioned that mitosane 137

could be derived from electron rich aromatic compound 138 by oxidation.

Construction of 138 would be realized in a two-reaction process analogous to the previous synthesis: a stereoselective addition of silyl enol ether 140 to the iminium ion derived from 141 followed by an intramolecular Buchwald-Hartwig aryl amination of 139. 109

O OR1 OR5 OR1 MeO MeO 8 H 7 H 2 AB OR 1 N C NH 4a N Me Me 5 2 3 O OR4 OR 137 138

OR1 OR5 2 H OR MeO OR3 N Me X H OR4 139

5 OSiMe3 OR R2O OR3 MeO + HO N Me X CO t-Bu OR4 2 140 141

Figure 5.1 Retrosynthetic Analysis of Mitosanes

110

Although this strategy proved successful in the previous synthesis, and good

stereocontrol was achieved in the addition process, a careful examination of the

synthetic plan revealed some key challenges associated with this synthetic route.

First, the regioselectivity in the oxidative quinone formation process would be

hard to achieve. The C4a-C7 oxidation might compete with the desired C5-C8

oxidation when R4 and R5 are alkyl groups. One way to overcome this is to convert

C5 and C8 substituents from alkoxy group to hydroxy group to direct the

regioselective oxidation. However, this requires extra steps1 and sometimes harsh

reaction conditions.

Second, another challenge is associated with the intramolecular Buchwald-

Hartwig aryl amination. For this process, palladium insertion into the aryl halide or

aryl triflate bond is vital for the subsequent aryl-nitrogen bond formation. However

there is still no satisfactory method to effectively induce palladium insertion reaction

on an electron rich system. Moreover, the normal aryl amination reaction requires

relative high temperature (80−100 °C) and strong base (e.g., t-BuOK, NaH, Cs2CO3).

This will limit the choice of protecting groups and sometimes require unnecessary or extraneous protecting groups on some of the functional groups.

With these considerations in mind, we realized a few modifications would be necessary for the above synthetic plan. A new strategy was adopted for the construction of the B-ring utilizing an intramolecular Michael addition of secondary

1. Kishi took 4 extra steps to do this: Fukuyama, T.; Nakatsubo F.; Cocuzza A. J.; Kishi, Y. “ Synthetic Studies toward Mitomycins. III. Total Syntheses of Mitomycin A and C,” Tetrahedron Lett. 1977, 49, 4295-4298. 111

pyrrolidine amine to the quinone system followed by a one-pot aerobic oxidation of the resulting dihydroquinone to return the system to quinone oxidation state. (Figure

5.2)

As outlined in Chapter 4.7, based on the rationale we proposed for the trans selectivity in the addition of silyl enol ethers to pyrrolidine derived iminium ions, we suspected a bulky carbamate protecting group would reduce the formation of the undesired minor diastereomer. Thus, a t-butyl carbamate was chosen to protect the pyrrolidine nitrogen.

112

O OR1 OR1 O OR2 MeO H H MeO AB OR3 Me N C NH HN Me O O 137 142

OR1 OR5 2 H OR MeO OR3 HN Me OR4 143

5 OSiMe3 OR R2O OR3 MeO + HO N Me CO t-Bu OR4 2 144 141

Figure 5.2 Modifications of Synthetic Strategy

113

5.2 Synthesis of the Iminium Ion Precursor

We realized that there were a large number of protecting group choices for R2 and R3 in pyrrolidine 141 type of compounds and that the absolute stereochemistry on

C9a would be dependent on this choice. We suspected a bulky protecting group of R2 would be helpful to control the face selectivity on C9a in the addition process. Thus, an acetonide protecting group on both the 2- and 3-hydroxy fuctionalities was chosen.

The acetonide would also help to stablize the molecule that is normally sensitive to acid. The Lewis acid used to catalyze the coupling would serve to deprotect the acetonide on C3 oxygen.

With these considerations in mind, we started our synthesis from D-ribose

(145). D-Ribose reacted with benzyl alcohol in refluxing acetone catalyzed by sulfuric acid to provide acetonide 146. Standard two-step process involving mesylation and SN2 azide displacement converted the primary alcohol to azide 148.

(Figure 5.3)

114

HO OBn HO OH O O acetone, BnOH MeSO2Cl, Et3N

H2SO4, 58 °C O O CH2Cl2, −20 °C OH OH 40% 93% 145 146

MsOOBn N OBn O 3 O NaN3 DMF, 50 °C O O O O 88% 147 148

Figure 5.3 Synthesis of Iminium Ion Precursor from D-Ribose

115

For the optimization purposes, primary alcohol 146 was effeciently converted to azide 148 in almost quantitative yield by a modified Mitsunobu reaction2 using zinc azide bis-pyridine complex3 as the azide source.4 One pot transformation using

Pd(OH)2/C, Et3SiH and (t-BuOCO)2O afforded the t-butyl carbamate (Boc) protected amine 149 in 96% yield.5 (Figure 5.4)

N3 OBn HO OH O O diethyl azodicarboxylate, PPh3

Zn(N3)2⋅(pyr)2, toluene, 25 °C O O OH OH 99% 145 148

t-BuO CHN OBn 2 O Pd(OH)2/C, Et3SiH (t-BuOCO) O, MeOH 2 O O 96% 149

Figure 5.4 Transformations from Primary Alcohol to Carbamate Protected Amine

2. Mitsunobu, O. “The Use of Diethyl Azodicarboxylate and Triphenylphosphine in Synthesis and Transformation of Natural Products,” Synthesis 1981, 1-28.

3. Agrell, I. “The Crystal Structure of [Zn(N3)2(C5H5N)2],” Acta Chem. Scand. 1970, 24, 1247-1261.

4. Viaud, M. C.; Rollin, P. “Zinc Azide Mediated Mitsunobu Substitution. An Expedient Method for the One-Pot Azidation of Alcohols,” Synthesis, 1990, 130-132.

5. Kotsuki, H.; Ohishi, T.; Araki, T. “A New Facile Method for the Chemoselective Reductive Transformation of Azides to N-(tert-Butoxycarbonyl)amines,” Tetrahedron Lett. 1997, 38, 2129- 2132. 116

With the primary amine protected as a t-butyl carbamate, what remained was

to remove the benzyl protecting group on the acetal to prepare for the oxidative ring

fragmentation to form the pyrrolidine system.

However this debenzylation process was more complicated than expected. On

a small scale (~25 mg), forcing reaction conditions (85 psi H2, Pd(OH)2/C, MeOH, 25

°C, 48 h) were required to achieve reasonably good yield (85%) of hemiacetal 150.

On a larger scale (~1.5 g), however, no reaction conditions gave a satisfactory yield of the hydrogenolysis product. (Figure 5.5)

117

t-BuO CHN OBn t-BuO CHN OH 2 O 2 O

O O O O

149 150

scale catalyst reaction condition yield

25 mg Pd(OH)2/C 85 psi H2, MeOH, 25 °C, 48 h 85%

1.5 g Pd(OH)2/C 85 psi H2, MeOH, 25 °C, 48 h 10%

1.5 g Pd(OH)2/C 85 psi H2, MeOH, 25 °C, 120 h 8%

1.5 g PdCl2 85 psi H2, MeOH, 25 °C, 48 h 4%

1.5 g Pd(OAc)2 85 psi H2, MeOH, 25 °C, 48 h 4%

1.5 g Pd/C 85 psi H2, MeOH, 25 °C, 48 h 5%

1.5 g Pd black 85 psi H2, MeOH, 25 °C, 48 h 24%

1.5 g Pd black 85 psi H2, HOAc, MeOH, 25 °C, 48 h 20%

1.5 g Pd/C NH4OCHO, MeOH, 65 °C, 48 h no reaction

1.5 g Me3SiI, CH2Cl2, 0 °C, 5 min s.m. decomp.

1.5 g BF3⋅Et2O, CH2Cl2, 0 °C, 5 min s.m. decomp.

Figure 5.5 Attempted Debenzylation

118

Those efforts suggested that benzyl protection group on the acetal was not the optimal protecting group for this system. An alternative protecting group that would

be removed more easily without changing the established reaction route needed to be

investigated.

An allyl protecting group on the acetal was used. Using reaction conditions

developed in the previous synthesis including acetonide formation and zinc azide- mediated Mitsunobu displacement converted D-ribose to azide 152. Unfortunately the one pot reaction conditions used to convert the azide to the corresponding t-butyl

carbamate protected amine would also reduce the double bond on the allyl protecting

group. Instead, a standard two-step process was used involving primary amine

formation followed by carbamate protection. Allyl protecting group on the acetal was

removed in one step using NiCl2(dppp) and Et3Al to afford hemiacetal 150 in 86%

yield.6

This modified process was very reliable. Up to 5 g of compound 150 was

successfully synthesized in one sequence of operations.

6. Taniguchi, T.; Ogasawara, K. “Extremely Facile and Selective Nickel-Catalyzed Allyl Ether Cleavage,” Angew. Chem. Int. Ed. 1998, 37, 1136-1137. 119

HO O HO OH O O acetone, allyl alcohol

H2SO4, 58 °C O O OH OH 39% 145 151

N O 3 O

diethyl azodicarboxylate, PPh3 PPh3, H2O

Zn(N3)2⋅(pyr)2, toluene, 25 °C O O toluene, 25 °C 94% 152

H N O t-BuO CHN O 2 O 2 O (t-BuOCO)2O

O O Et3N, CH2Cl2, 25 °C O O 88%, 2 steps 153 154

t-BuO CHN OH 2 O NiCl2(dppp), Et3Al

toluene, 0−25 °C O O 86% 150 dppp = 1,3-bis(diphenylphosphino)propane

Figure 5.6 Allyl Protecting Group

120

The hemiacetal 150 was subjected to iodosobenzene/iodine mediated ring fragmentation to afford the unstable key five-membered pyrrolidine 155 in moderate yield.7,8 It is presumed that iodosobenzene cleaved the C-C bond adjacent to the hemiacetal followed by the intramolecular nucleophilic attack of the carbamate

nitrogen to form the desired five-membed pyrrolidine ring. (Figure 5.7)

t-BuO CHN OH 2 O O OCHO PhIO, I2 O O O CH2Cl2, 25 °C N 69% CO2t-Bu 150 155

t-BuO CHN O t-BuO CHN O t-BuO CHN O 2 O 2 O 2 O

O O O O O O

156

Figure 5.7 Pyrrolidine Ring Formation

7. Saltzman, H.; Sharefkin, J. G. “Iodosobenzene,” Org. Synth., 1973, Coll. Vol. V, 658-659.

8. Armas, P.; Francisco, C. G.; Suarez, E. “Hypervalent Iodine Reagents: Preparation of Chiral Synthesis Building Blocks by Fragmentation of Anomeric Carbohydrate Alkoxy Radicals,” Angew. Chem. Int. Ed. 1992, 31, 772-774; Armas, P.; Francisco, C. G.; Suarez, E. “Fragmentation of Alkoxy Radicals: Tandem β-Fragmentation-Cycloperoxyiodination Reaction,” Tetrahedron Lett. 1992, 33, 6687-6690; Armas, P.; Francisco, C. G.; Suarez, E. “Fragmentation of Carbohydrate Anomeric Alkoxy Radicals. Tandem β-Fragmentation-Cyclization of Alcohols,” J. Am. Chem. Soc. 1993, 115, 8865-8866; Francisco, C. G.; Freire, R.; Gonzalez, C. C.; Suarez, E. “Fragmentation of Carbohydrate Anomeric Alkoxy Radicals. Synthesis of Azasugars,” Tetrahedron: Asym. 1997, 8, 1971-1974. 121

The unstable formic ester on the pyrrolidine compound 155 was not a suitable

substituent for further investigation. Reductive removal of the formic ester by Super

Hydride (LiEt3BH) was successful in almost quantitative yield. The resulting

secondary alcohol 156 was protected as a benzyl ether. (Figure 5.8) Benzyl ether 157

served as the pyrrolidine iminium ion precursor in the following studies.

O OCHO O OH O OBn LiEt3BH BnBr, NaH, n-Bu4NI O O O N THF, −78 °C N DMF, THF, 25 °C N CO2t-Bu 99% CO2t-Bu 89% CO2t-Bu 155 156 157

Figure 5.8 Synthesis of Iminium Ion Precursor

5.3 Synthesis of Silyl Enol Ether Precursor and Attempted Coupling

Reaction

Similar to many reported approaches to the mitomycins, we started our

aromatic precursor synthesis from 2,6-dimethoxytoluene (158), an inexpensive

9 commercially available compound. Friedel-Crafts alkylation by Cl2CHOMe and

9. Fleming, I. “Improving the Friedel-Crafts Reaction,” Chemtracts 2001, 14, 405-406; and references cited therein.

122

10 SnCl4 in CH2Cl2 afforded benzaldehyde 159 in 93% yield. Reaction using TiCl4 as the Lewis acid was not as efficient. Standard Baeyer-Villiger reaction11 conditions

(m-chloroperoxybenzoic acid) followed by KOH catalyzed hydrolysis in methanol12 of the resulting formic ester furnished the desired phenol 160 in 79% yield. To facilitate the further installation of substituent ortho to the phenol, phenol 160 was converted to methoxymethyl (MOM) ether 161. This reaction sequence is very convenient at large scale: benzaldehyde 159 was purified by recrystalization in hot hexane; phenol 160 and methoxymethyl ether 161 were purified by distillation under high vacuum.13

10. Meier, H.; Kretzschmann, H.; Kolshorn, H. “[abc]-Annealated [18]Annulenes,” J. Org. Chem. 1992, 57, 6847-6852.

11. Renz, M.; Meunier, B. “100 Years of Baeyer-Villiger Oxidations,” Eur. J. Org. Chem. 1999, 737- 750.

12. Maruyama, K.; Nagai, N.; Naruta, Y. “Lewis Acid-Mediated Claisen-Type Rearrangement of Aryl Dienyl Ethers,” J. Org. Chem. 1986, 51, 5083-5092.

13. A major by-product for synthesizing benzaldehyde 159 is HCl gas. It is very important to keep reaction flask open to release gas in large reaction scale. 123

CHO MeO Cl2CHOMe, SnCl4 MeO 1. m-CPBA

CH2Cl2, 0 °C 2. KOH, MeOH, 0 °C 93% OMe OMe 79% 158 159

OH OCH2OMe MeO MeO MeOCH2Cl, NaH THF, DMF OMe 89% OMe 160 161

Figure 5.9 Functional Group Installations on the Aromatic Ring

Methoxymethyl ether directed ortho lithiation of compound 161 with n-BuLi was efficient. Subsequent treatment of the phenyl anion with allylbromide afforded allylbenzene 162 in 88% yield. No meta lithiation product was observed, although in previous reports a methyl ether has sometimes served as a lithiation directing group.

The cleavage of the terminal double bond to form aldehyde was surprisingly problematic. Standard ozonolysis condition (O3, CH2Cl2, −78 °C; Me2S, 25 °C) gave

only 6% desired aldehyde 163. We assumed that the low yielding was a result of

competing oxidation of the electron rich aromatic ring. Johnson-Lemieux oxidation14

(NaIO4, OsO4, dioxane, H2O, 0 °C) initially gave even lower yield, but switching

14. Pappo, R.; Allen Jr., D. S.; Lemieux, R. U.; Johnson, W. S. “Osmium Tetroxide-Catalyzed Periodate Oxidation of Olefinic Bonds,” J. Org. Chem. 1956, 21, 478-479. 124

solvent systems to Et2O/H2O dramatically increased the yield to 65%. Aldehyde 163

was then converted to silyl enol ether 164 in 89% crude yield, however with a

cis/trans ratio 2.8:1 in favor of cis double bond form. (Figure 5.10)

MeOCH2O OCH2OMe MeO MeO 1. n-BuLi, THF, −78 °C NaIO4, OsO4

2. allylbromide, THF, 0 °C Et2O, H2O, 0 °C OMe 88% OMe 65% 161 162

OSiMe3 OCH2OMe MeOCH2O

MeO O NaI, Me3SiCl, Et3N MeO MeCN, 0−25°C cis: trans 2.8:1 89% OMe OMe 163 164

Figure 5.10 Synthesis of Silyl Enol Ether

Although the cis/trans selectivity on the double bond was not satisfactory in the silyl enol ether synthesis, we continued our investigation with this mixture.

Unfortunately coupling reaction of silyl enol ether 164 with pyrrolidine 157 catalyzed by a variety of Lewis acid failed to provide corresponding aldehyde 165.

125

OSiMe O MeOCH O 3 MeOCH O 2 O OBn 2 OH MeO Lewis acid MeO + O x OBn N N CO t-Bu CO t-Bu OMe 2 OMe 2 164 157 165

Figure 5.11 Failed Coupling Reaction

We realized that silyl enol ether 164 was too sensitive toward Lewis acid but

pyrrolidine 157 was overly stable. The reactivity of the two species did not match

each other. These prompted us to move to synthesizing a more stable aryl precursor.

Because of the dominant undesired α-cross-coupling of allylsilanes (Chapter 4.2,

Figure 4.4), we decided to investigate the reaction of the corresponding allylstannane.

5.4 Synthesis of the Cinnamyl Stannane and Its Coupling Reaction

We commenced the synthesis of allylstannane from methoxymethyl ether 161.

Methoxymethyl ether directed ortho lithiation with n-BuLi followed by subsequent

treatment of freshly distilled acrolein afforded allyl alcohol 166 in 79% yield. This secondary alcohol was converted to acetate 167 following standard procedures.

Reductive stannylation of acetate 167 furnished trans-cinnamyl stannane 168 in 77%

126

yield with excellent regio- and stereocontrol.15 No cis or terminal double bond by- products were detected. This methodology was originally developed for allyl halides and occurred without thermodynamically driven double bond migration. Our synthesis expanded the scope of the application. (Figure 5.12)

MeOCH2O MeOCH2O OH MeO MeO 1. n-BuLi, THF, −78 °C AcCl, Et3N

2. acrolein, THF, 0 °C CH2Cl2, 25 °C OMe 79% OMe 89% 161 166

MeOCH2O OAc MeOCH2O MeO MeO SmI2, Pd(PPh3)4, n-Bu3SnCl Sn(n-Bu)3 THF, 0-25 °C OMe 77% OMe 167 168

Figure 5.12 Synthesis of Cinnamyl Stannane

15. Tabuchi, T.; Inanaga, J.; Yamaguchi, M. “A Mild and Efficient Method for the Formation of Allylstannanes Utilizing Samarium Iodide-Induced Polarity Inversion of π-Allyl Palladium Complexes,” Tetrahedron Lett. 1987, 28, 215-216. 127

The key coupling reaction of allystannane 168 with pyrrolidine 157 catalyzed by

BF3⋅Et2O proceeded smoothly to afford the allylically transposed compound 169 in

75% yield. (Figure 5.13)

Sn(n-Bu)3 O OBn MeOCH O MeOCH O 10 2 O 157 2 OH N H MeO MeO 9 9a CO2t-Bu OBn N BF3⋅Et2O, CH2Cl2, −78 °C CO2t-Bu OMe 75% OMe 168 169

Figure 5.13 Stannane Coupling with Pyrrolidine Iminium Ion

The amount of BF3⋅Et2O must be stoichiometric. Excess reagent caused the further deprotection of methoxymethyl group on the aromatic ring, which was not desirable at that stage. Only one stereoisomer was isolated from the reaction mixture.

However, because of line broadening in 1H NMR signal, we were unable to determine

the absolute and relative stereochemistry about C9−C9a bond at that stage.

Hydrogenation of 169 with 1 atm H2 and Pd/C in MeOH hydrogenated the

double bond and at the same time removed the benzyl protecting group. The resulting

diol 170 was converted to bis-dinitrobenzoic ester 171, which was a nicely crystalline

compound. Single crystal X-ray structure determination established the absolute

stereochemistry at C9 as R and C9a as S, as we expected. (Figure 5.14, 5.15) 128

MeOCH O MeOCH O 2 H OH 2 H OH MeO H2, Pd/C MeO OBn OH N MeOH, 25 °C N CO t-Bu CO t-Bu OMe 2 OMe 2 169 170

MeOCH O 2 H ODNB MeO 9 3,5-dinitrobenzoyl chloride, DMAP 9a ODNB N CH2Cl2, 25 °C CO t-Bu OMe 2 171 DMAP = 4-(dimethylamino)pyridine DNB = 3,5-dinitrobenzoyl

Figure 5.14 Synthesis of bis-Dinitrobenzoic Ester

129

Figure 5.15 X-Ray Crystal Structure of 17116

16. Irrelevant hydrogen atoms and both of the dinitrobenzoyl groups were deleted for clarity of illustration. 130

We assumed the allylstannane addition to iminium ion also went through the

same transition state of the silyl enol ether addition to iminium ion which favored a

trans diastereoselectivity between C9-C10 and C9a−H bonds by the less steric

demanding synclinal transition state. (Figure 4.13) We had expected a bulky

carbamate might reduce the undesired cis diastereoselectivity and we did not observe

other diastereomers in this experiment, presumably in part due to the bulky t-butyl

carbamate. The absolute stereochemistry is obviously a result of the chirality of the

iminium ion precursor. When an excess of this precursor was used, we observed

significant unreacted material which suggested the collapse of the actonide actually

happened after the stannane addition, rather when Lewis acid was added. Thus, the

acetonide controlled the face selectivity of the stannane addition, which determined

the absolute stereochemistry at C9a.

5.5 Mitosane B Ring Cyclization

By switching from silyl enol ether to allylstannane, we had efficiently coupled

the aromatic and pyrrolidine precursors. As a result, we had to deal with an old

problem that troubled us when we prepared the silyl enol ether precursor: cleavage of the terminal double bond. Presumably it was to be even more complicated than before. The aromatic ring remained electron rich and still sensitive to oxidation; the compound we were dealing was more complicated containing more functional groups and a sensitive benzylic stereogenic center.

131

Initial efforts employing ozonolysis and reduction protocol (O3, MeOH, −78

°C; NaBH4, 0 °C) lead to very low yield (< 5%) of alcohol 174. Johnson-Lemieux

oxidation gave aldehyde 173 in 40% yield plus 29% of diol 172 as a by-product. This

prompted us to investigate the efficiency of the two-step process. Unfortunately,

stoichiometric OsO4 dihydroxylation followed by NaIO4 cleavage gave a lower yield.

Although the yield was not satisfactory, we made aldehyde 173 by Johnson-Lemieux

procedure and reduced the aldehyde to primary alcohol 174 in 78% yield. (Figure

5.16)

Because of the carbamate protecting group on the pyrrolidine nitrogen, 1H

NMR spectra of compounds 169, 172 and 173 were broadened and provided a very limited amount of information. However the 1H NMR signals of primary alcohol 174

were clear, though still far from the sharpness of typical NMR signal. We suspected

this as a result of hydrogen bonding between the primary alcohol and the carbonyl on the carbamate that stopped the slow rotation along the N−CO bond, a primary cause

for the NMR line broadening. We were surprised to find that almost all proton signals

were doubled in the 1H NMR spectrum of 174. Although ozonolysis and reduction only provided very low yield of 174, there was only one set of proton signals on 1H

NMR. Clearly epimerization at the C9 stereogenic center happened, most likely at the

stage of aldehyde 173.

132

OH HO MeOCH O MeOCH O 2 HHOH 2 OH MeO MeO OsO4 OBn OBn N Et2O, H2O, 0 °C N CO t-Bu CO t-Bu OMe 2 59% OMe 2 169 172

OsO , NaIO 4 4 NaIO4 Et O, H O, 0 °C 2 2 Et2O, H2O, 0 °C 40% 34% + 29% 172

OH O MeOCH O MeOCH O 2 H OH 2 H OH MeO LiEt3BH MeO OBn OBn N THF, −78 °C N CO t-Bu 78% CO t-Bu OMe 2 OMe 2 174 173

Figure 5.16 Cleavage of the Terminal Double Bond

Obviously aldehyde 173 was not a suitable intermediate to move the synthesis forward. There was little literature precedent of this kind of transformation without going through aldehyde intermediate besides ozonolysis/reduction and

17 Me3SiO3H/LiAlH4. We focused on optimizing ozonolysis reaction and found out under a slow ozone flow and carefully controlled reaction conditions, the reaction

17. Reference for Me3SiO3H/LiAlH4: Posner, G. H.; Oh, C. H.; Milhous, W. K. “Olefin Oxidative Cleavage and Dioxetane Formation using Triethylsilyl Hydrotrioxide: Applications to Preparation of Potent Antimalarial 1,2,4-Trioxanes,” Tetrahedron Lett. 1991, 32, 4235-4238.

133

furnished alcohol 175 in 62% yield with no observation of epimerization at C9.

(Figure 5.17)

OH MeOCH O MeOCH O 2 HHOH 2 OH MeO MeO 1. O3, MeOH, -78 °C OBn OBn N 2. NaBH4, MeOH, 0 °C N CO t-Bu CO t-Bu OMe 2 62% OMe 2 169 175

Figure 5.17 Ozone Cleavage of the Terminal Double Bond

From the previous experience, we knew that if we removed t-butyl carbamate protecting group in alcohol 175 under acidic conditions, we would eliminate a molecule of water to form styrene-like compound. But it would be desirable if protection on the primary alcohol could be avoided. Thus reductive removal

18 19 conditions such as Red-Al (NaAlH2(OCH2CH2OCH3)2) and LiBH4/Me3SiCl were investigated but failed to remove the targeted t-butyl carbamate. The primary alcohol was protected as a t-butyldiphenylsilyl ether. Treatment of 176 with trifluoacetic acid

18. This condition was developed to remove methyl carbamate. Lenz, G. R. “Synthesis of 7- Oxygenated Aporphine Alkaloids from A 1-Benzylideneisoquinoline Enamide,” J. Org. Chem. 1988, 53, 4447-4452.

19. This condition was developed to remove benzyl carbamate. Giannis, A.; Sandhoff, K. “Lithium Borohydride (Sodium Borohydride)-Chlorotrimethylsilane, An Unusually Strong and Versatile Reducing Agent,” Angew. Chem. Int. Ed. 1989, 28, 218-220. 134

removed the t-butyl carbamate protecting group and also the methoxymethyl

protecting group to afford aminophenol 177 in 79% yield. (Figure 5.18)

OH MeOCH O 2 H OH MeO t-BuPh2SiCl, Et3N OBn N DMAP, CH2Cl2, 25 °C CO t-Bu 99% OMe 2 175

OSit-BuPh OSit-BuPh MeOCH O 2 OH 2 2 HHOH OH MeO MeO OBn CF3COOH OBn N HN CH2Cl2, 25 °C CO2t-Bu OMe 79% OMe 176 177

Figure 5.18 Synthesis of Aminophenol

As described in the retrosynthetic analysis, our plan was to close the B ring in mitosane system by an intramolecular Michael addition of pyrrolidine amine to the quinone α,β-unsaturated ketone system. Aminophenol 177 provided a good chance to execute this plan. We envisioned that a mild oxidant that selectively oxidize the electron rich aromatic ring without oxidizing the secondary pyrrolidine amine.

Regiocontrol should not be a problem in this case with the phenol and

135

methoxylphenol ether in para relationship on the aromatic ring as the directing groups. Once the quinone was formed it would be a very good Michael acceptor for the intramolecular attack by the pyrrolidine amine. This intramolecular Michael addition would set up the mitosane B-ring and return the quinone ring to aromatic dihydroquinone oxidation state, which would be re-oxidized to the quinone by oxygen in the air during workup. (Figure 5.19)

OSit-BuPh OSit-BuPh OH 2 O 2 HHOH OH MeO [O] MeO OBn OBn HN N H OMe O 177 178

OH OSit-BuPh2 O OSit-BuPh2 MeO H MeO H OH O2 OH N N OH OBn O OBn 179 180

Figure 5.19 Proposed B Ring Cyclization

From the synthetic design viewpoint, this one-pot oxidation-cyclization- oxidation operation avoided the possibly troublesome palladium catalyzed N-aryl

136

bond formation and the difficult regiocontrol in oxidative quinone formation. It

incorporated three steps in a one-pot procedure, which would require several

operations otherwise. However, we realized one key step could be problematic. The

Michael addition of the pyrrolidine amine to quinone system was a 5-endo-trig cyclization, which is disfavored according to Baldwin’s rules.20 Reviewing literature

precedent revealed that although this empirical rule did predict and explain many

experimental results, it does not “represent a solitary island in the ocean of

stereoelectronic phenomena” and there were some reported examples in violation of

the rules.21,22 Thus we decided to continue investigating this key cyclization.

Initially we thought that oxygen would be a mild and effective oxidant for the

electron rich aromatic system. Unfortunately no reaction occurred upon extended exposure of 177 to O2. A similar lack of reactivity with CuCl2/O2 and Ag2O were

observed. Silver (II) oxide oxidation was too harsh and formed a mixture of

unidentified polar products. Ceric ammonium nitrate oxidation was successful, but

only in moderate yield. Efforts were invested to optimize the reaction yield with no

success. (Figure 5.20)

The decomposition of the mitosane compound 186 during the concentration

process was observed and probably due to the instability of this compound. In the

ceric ammonium nitrate oxidation, we observed significant new polar impurities on

20. Baldwin, J. E. “Approach Vector Analysis: A Stereochemical Approach to Reactivity,” J. Chem. Soc., Chem. Comm. 1976, 738-741.

21. Johnson, C. D. “Stereoelectronic Effects in the Formation of 5- and 6-Membered Rings: the Role of Baldwin's rules,” Acc. Chem. Res. 1993, 26, 476-482.

22. Actually the mitomycin/isomytomycin conversion is an example of violations of Baldwin’s rules. 137

TLC and significant new mass peaks showing up on mass spectrometric analysis after the concentration process. However, in most cases there was always some mitosane compound remaining.

OSiPh2(t-Bu) O OSiPh2(t-Bu) OH OH H MeO H MeO [O] OH OBn HN N O OBn OMe 177 180

entry oxidant reaction condition yield

1 O2 MeOH, H2O, 25 °C, 24 h no reaction

2 O2 MeOH, H2O, 65 °C, 24 h no reaction

3 CuCl2, O2 MeOH, H2O, 25 °C, 24 h no reaction

4 Ag2O MeOH, H2O, dark, 25 °C, 24 h no reaction

5 AgO MeOH, H2O, dark, 0 °C, 30 min 0%

6 (NH4)3Ce(NO3)6 MeCN, H2O, 25 °C, 5 min 15 ~ 25%

7 DDQ THF, H2O, 0 °C, 10 min 0%

(DDQ = 2,3-dichloro-5,6-dicyanoquinone)

Figure 5.20 Oxidative Cyclization

138

5.6 Unexpected Mitosane Conversion to Mitosene

We had planned for the cis diol relationship in mitosane 180 to provide us a

good opportunity to install the aziridine functionality. Although we were not satisfied

with the low yield in the mitosane formation process, we continued our investigation.

An interesting result was obtained when we tried to convert the secondary

alcohol on mitosane 180 to a mesylate using Et3N and MeSO2Cl. Instead of the

corresponding mesylate, mitosene 181 was isolated from the reaction mixture. We

assumed this reaction process occurred through a base-catalyzed re-aromatization and

air oxidation process. The driving force for this conversion is the thermodynamic

stability of the aromatic indole structure in 184. This process was supported by the conversion of mitosane 180 to mitosene 181 mediated by Et3N and CH2Cl2. (Figure

5.21)

This unexpected conversion of mitosane 180 to mitosene 181 reflects the

difficulty of synthesizing the targeted quinone-indoline system. The base sensitive

nature of mitosane 180 also explained the low efficiency in the cyclization process.

Proton NMR monitored cyclization reaction revealed that ceric ammonium nitrate

oxidation of the aromatic system to quinone was a fast reaction process. However,

without the basic workup procedure, in NMR tube the intramolecular Michael

addition would not occur. Unfortunately the presence of base would also further

destroy the resulting mitosane 180, thus cause significant loss of the desired product.

139

This also explained the observation of formation of mitosene 181 in the cyclization process, which was a result of treatment with base rather the over-oxidation.

140

O OSit-BuPh2 O OSit-BuPh2 MeO H Et N, MeSO Cl MeO OH 3 2 OH

N CH2Cl2, −20 °C N O OBn 56% O OBn

180 181

OSit-BuPh2 OSit-BuPh2 O OH B MeO H MeO H OH N OH N O OBn OH OBn

182 184

OSit-BuPh2 O B MeO H

N OH OH OBn 183

Figure 5.21 Unexpected Mitosane Conversion to Mitosene

141

5.7 Conclusion

We have successfully synthesized mitosane 180 and mitosene 181 in an

enantioselective fashion. Because of the unexpected base sensitivity of mitosane 180,

it does not apprear to be a suitable precursor for installation of the aziridine

functionality unless the quinone system is reduced to dihydroquinone and

immediately protected. A good way to approach the mitosane-aziridine system might

be installation of the aziridine prior to the formation of the quinone-indoline system.

Some of our efforts of introduce aziridine precursors prior to the ring cyclization will be discussed in the following chapter.

142

5.8 Experimental

HO OBn O

O O (6-Benzyloxy-2,2-dimethyltetrahydrofuro[3,4-d][1,3]dioxol-

4-yl)methanol (146). Concentrated H2SO4 (0.12 mL) was slowly added to a solution

of D-ribose (1.50 g, 10.0 mmol) and benzyl alcohol (10 mL) in acetone (6.0 mL) at

25 °C. The reaction mixture was warmed at reflux for 3 h. The reaction mixture was

allowed to cool to 25 °C and Na2CO3 (1.50 g) was added. The mixture was filtered

and the filtrate was evaporated in vacuo. The residue was heated at 150 °C under high

vacuum (0.2 mm Hg) to remove benzyl alcohol. The residue was purified by flash

chromatography (5 × 10 cm silica, 20% EtOAc/hexane) to afford pure 146 (1.12 g,

1 40%) as a colorless solid: H NMR (CDCl3, 500 MHz) δ 7.27 (m, 5H), 5.16 (s, 1H),

4.80 (d, 1H, J = 6.1 Hz), 4.72 (d, 1H, J = 11.6 Hz), 4.64 (d, 1H, J = 6.1 Hz), 4.51 (d,

1H, J = 11.6 Hz), 4.39 (t, 1H, J = 3.7 Hz), 3.61 (m, 2H), 3.22 (dd, 1H, J = 9.2, 3.7

13 Hz), 1.45 (s, 3H), 1.28 (s, 3H); C NMR (CDCl3, 125 MHz) δ 136.31, 128.40,

127.94, 111.88, 107.69, 88.07, 85.64, 81.33, 69.74, 63.67, 26.13, 24.48; HRMS (ESI)

m/z 303.2217 (calcd. for C15H20O5 + Na: 303.2209).

143

N OBn 3 O

O O 4-Azidomethyl-6-benzyloxy-2,2-dimethyltetrahydro-

furo[3,4-d][1,3]dioxole (148). Zinc azide bis-pyridine complex (58.0 mg, 0.19 mmol)

was slowly added to a solution of 146 (70.0 mg, 0.25 mmol), Ph3P (131 mg, 0.50

mmol), and diisopropyl azodicarboxylate (0.098 mL, 0.50 mmol) in toluene (1.0 mL)

at 25 °C. The reaction mixture was stirred for 16 h at 25 °C and was evaporated in

vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 10%

EtOAc/hexane) to afford pure 148 (77.0 mg, 99%) as a colorless oil: 1H NMR

(CDCl3, 500 MHz) 7.30 (m, 5H), 5.18 (s, 1H), 4.73 (d, 1H, J = 12.6 Hz), 4.68 (d, 1H,

J = 6.1 Hz), 4.61 (d, 1H, J = 6.1 Hz), 4.49 (d, 1H, J = 12.20 Hz), 4.31 (t, 1H, J = 6.7

Hz), 3.48 (dd, 1H, J = 12.2, 7.9 Hz), 3.29 (dd, 1H, J = 12.2, 6.7 Hz), 1.46 (s, 3H),

13 1.30 (s, 3H); C NMR (CDCl3, 125 MHz) δ 136.89, 128.55, 128.07, 112.77, 107.73,

85.62, 85.30, 82.16, 69.58, 53.74, 26.42, 24.92; HRMS (ESI) m/z 328.1288 (calcd.

for C15H19N3O4 + Na: 328.1274).

t-BuO CHN OBn 2 O

O O (6-Benzyloxy-2,2-dimethyltetrahydro-furo[3,4-

d][1,3]dioxol-4-ylmethyl)carbamic Acid tert-Butyl Ester (149). Triethylsilane (0.32

mL, 2.0 mmol) was added dropwise over 30 min to a mixture of 148 (101.7 mg, 0.33

mmol), Pd(OH)2/C (9 mg) and di-tert-butyl dicarbonate (0.11 mL, 0.50 mmol) in

144

MeOH (1.5 mL) at 25 °C. The reaction mixture was stirred for 30 min at 25 °C. The

mixture was filtered and the filtrate was evaporated in vacuo. The residue was

purified by flash chromatography (1 × 10 cm silica, 10% EtOAc/hexane) to afford

1 pure 149 (120.1 mg, 96%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 7.33-7.25

(m, 5H), 5.13 (s, 1H), 5.09-4.99 (br, 1H), 4.68 (br, 1H), 4.65 (br, 1H), 4.61 (br, 1H),

4.47 (br, 1H), 4.29 (m, 1H), 3.27 (br, 2H), 1.49 (s, 9H), 1.43 (s, 3H), 1.27 (s, 3H);

HRMS (ESI) m/z 402.1878 (calcd. for C20H29NO6 + Na: 402.1893).

HO O O

O O (6-Allyloxy-2,2-dimethyltetrahydrofuro[3,4-

d][1,3]dioxol-4-yl)methanol (151). Concentrated H2SO4 (1.2 mL) was slowly added

to a solution of D-ribose (15.0 g, 100.0 mmol) and allyl alcohol (68 mL) in acetone

(60 mL) at 25 °C. The reaction mixture was warmed at reflux for 3 h. The reaction

mixture was allowed to cool to 25 °C and Na2CO3 (15.0 g) was added. The mixture

was filtered and the filtrate was evaporated in vacuo. The residue was purified by

flash chromatography (10 × 10 cm silica, 20% EtOAc/hexane) to afford pure 151

1 (10.0 g, 44%) as a colorless oil: H NMR (CDCl3, 400 MHz) δ 5.82 (m, 1H), 5.22

(dd, 1H, J = 17.2, 2.6 Hz), 5.18 (dd, 1H, J = 10.4, 2.4 Hz), 5.06 (s, 1H), 4.79 (d, 1H, J

= 6.0 Hz), 4.57 (d, 1H, 5.9), 4.36 (s, 1H), 4.17 (dd, 1H, J = 5.4, 1.3 Hz), 4.02 (dd, 1H,

J = 6.4, 1.2 Hz), 3.61 (m, 2H), 3.17 (dd, 1H, J = 10.1, 3.2 Hz), 1.43 (s, 3H), 1.26 (s,

13 3H); C NMR (CDCl3, 100 MHz) δ 133.01, 118.18, 112.02, 107.86, 88.30, 85.86,

145

81.44, 68.82, 63.86, 26.26, 24.61; HRMS (ESI) m/z 253.1031 (calcd. for C11H18O5 +

Na: 253.1052).

N O 3 O

O O 4-Allyloxy-6-azidomethyl-2,2-dimethyltetrahydrofuro[3,4-

d][1,3]dioxole (152). Zinc azide bis-pyridine (842.0 mg, 2.8 mmol) was slowly added

to a solution of 151 (842.8 mg, 3.66 mmol), Ph3P (1.84 g, 7.0 mmol), and diisopropyl azodicarboxylate (1.10 mL, 7.0 mmol) in toluene (18 mL) at 25 °C. The reaction

mixture was stirred for 16 h at 25 °C and was evaporated in vacuo. The residue was purified by flash chromatography (5 × 10 cm silica, 10% EtOAc/hexane) to afford

1 pure 152 (877.1 mg, 94%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 5.93 (m,

1H, C2′-H), 5.33 (dd, 1H, C3′-H, J = 17.1, 1.2 Hz), 5.25 (dd, 1H, C3′-H, J = 10.4, 1.2

Hz), 5.19 (s, 1H, C1-H), 4.71 (d, 1H, C2-H, J = 6.1 Hz), 4.66 (d, 1H, C3-H, 6.1), 4.34

(t, 1H, C4-H, J = 7.3 Hz), 4.26 (m, 1H, C1′-H), 4.04 (m, 1H, C1′-H), 3.45 (dd, 1H,

C5-H, J = 12.8, 6.1 Hz), 3.26 (dd, 1H, C5-H, J = 12.8, 7.9 Hz), 1.53 (s, 3H, CH3),

13 1.37 (s, 3H, CH3); C NMR (CDCl3, 125 MHz) δ 133.46, 117.65, 112.71, 107.66,

85.48, 85.21, 82.10, 68.38, 53.72, 26.39, 24.90; HRMS (ESI) m/z 278.1120 (calcd.

for C11H17N3O4 + Na: 278.1117).

146

t-BuO CHN O 2 O

O O (6-Allyloxy-2,2-dimethyltetrahydrofuro[3,4-

d][1,3]dioxol-4-ylmethyl) carbamic Acid tert-Butyl Ester (154). A slurry of Ph3P

(144 mg, 0.55 mmol) was added to a solution of 152 (127.5 mg, 0.50 mmol) in

toluene (2.5 mL) at 25 °C. The reaction mixture was stirred for 1 h at 25 °C and water

(0.5 mL) was added. The reaction mixture was stirred for 12 h at 25 °C and was

evaporated in vacuo. The residue was dried under high vacuum for 12 h and dissolved

in CH2Cl2 (2.5 mL) at 25 °C. Triethylamine (0.56 mL, 4.0 mmol) was added followed

by the addition of di-tert-butyl dicarbonate (0.13 mL, 0.55 mmol). The reaction

mixture was stirred for 3 h at 25 °C and saturated aqueous NaHCO3 (2 mL) was

added. The aqueous layer was extracted with EtOAc (3 × 2 mL) and the combined

organic extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in

vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 10%

EtOAc/hexane) to afford 154 (144.9 mg, 88%) as a colorless oil, which solidified on

1 standing: H NMR (CDCl3, 500 MHz) δ 5.85 (m, 1H, C2’-H), 5.27 (app d, 1H, C3′-

H, J = 17.1 Hz), 5.16 (br d, 1H, C3′-H, J = 10.4 Hz), 5.07 (br s, 2H, C1-H collapse

with NH), 4.61 (s, 2H, C2-H collapse with C3-H), 4.26 (t, 1H, C4-H, J = 5.5 Hz),

4.15 (m, 1H, C1′-H), 3.96 (m, 1H, C1′-H), 3.27 (br, 2H, C5-H), 1.44 (s, 3H, CH3),

13 1.41 (s, 9H, C(CH3)3), 1.27 (s, 3H, CH3); C NMR (CDCl3, 125 MHz) δ 155.91,

133.44, 117.40, 112.33, 107.78, 86.20, 85.58, 82.02, 79.33, 68.38, 43.56, 28.32,

26.38, 24.83; HRMS (ESI) m/z 352.1749 (calcd. for C16H27NO6 + Na: 352.1736).

147

t-BuO CHN OH 2 O

O O (6-Hydroxy-2,2-dimethyltetrahydrofuro[3,4-

d][1,3]dioxol-4-ylmethyl)carbamic Acid tert-Butyl Ester (150). A solution of Et3Al

(1.9 M in toluene, 0.10 mL) was slowly added to a solution of 154 (27.3 mg, 0.083 mmol) and NiCl2(dppp) (3 mg) in toluene (1 mL) at 0 °C. The reaction mixture was

stirred for 5 min at 0 °C and 4 h at 25 °C and water (0.1 mL) was added. The reaction

mixture was stirred for 1 h at 25 °C and was evaporated in vacuo. The residue was

purified by flash chromatography (0.5 × 10 cm silica, 15% EtOAc/hexane) to afford

1 150 (20.6 mg, 86%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 5.43 (s, 1H),

5.17-5.07 (br, 1H), 4.60 (m, 2H), 4.40 (br, 1H), 4.19 (br, 1H), 3.43 (br, 1H), 3.20 (br,

13 1H), 1.44 (br, 3H), 1.41 (br, 9H), 1.24 (br, 3H); C NMR (CDCl3, 125 MHz) δ

156.81, 112.44, 103.05, 86.59, 86.23, 82.06, 79.89, 44.18, 28.35, 26.45, 24.91;

HRMS (ESI) m/z 312.1432 (calcd. for C13H23NO6 + Na: 312.1423).

O OCHO

O N CO2t-Bu 6-Formyloxy-2,2-dimethyltetrahydro-[1,3]dioxolo[4,5-

b]pyrrole-4-carboxylic Acid tert-Butyl Ester (155). Freshly prepared PhIO (70.4

mg, 0.32 mmol) was slowly added to a solution of 150 (37.1 mg, 0.128 mmol) and I2

(33.0 mg, 0.13 mmol) in CH2Cl2 (100 mL) at 25 °C. The reaction mixture was stirred

for 24 h at 25 °C and a saturated aqueous Na2S2O3 solution (20 mL) was added. The

148

organic layer was washed with saturated aqueous NaCl (20 mL), dried (Na2SO4),

filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash

chromatography (1 × 10 cm silica, 10% EtOAc/hexane) to afford pure 155 (25.3 mg,

1 69%) as a colorless solid: H NMR (CDCl3, 500 MHz, 27 °C) δ 8.06 (s, 1H), 5.94-

5.74 (br, 1H), 5.01-4.91 (br, 1H), 4.79-4.73 (br, 1H), 4.05-3.90 (br, 1H), 3.36 (t, 1H, J

1 = 10.4 Hz), 1.49 (br, 3H), 1.44 (br, 9H), 1.34 (br, 3H); H NMR (CDCl3, 500 MHz,

60 °C) δ 8.04 (s, 1H), 5.85 (br, 1H), 4.95 (m, 1H), 4.76 (t, 1H, J = 9.2 Hz), 3.97 (br t,

1H), 3.38 (t, 1H, J = 10.4 Hz), 1.49 (s, 3H), 1.44 (s, 9H), 1.34 (s, 3H); 13C NMR

(CDCl3, 125 MHz) δ 159.71, 153.49, 112.98, 81.01, 69.83, 60.22, 45.95, 28.36,

26.82, 26.28; HRMS (ESI) m/z 310.1264 (calcd. for C13H21NO6 + Na: 310.1267).

O OH

O N CO2t-Bu 6-Hydroxy-2,2-dimethyltetrahydro-[1,3]dioxolo[4,5-b]pyrrole-

4-carboxylic Acid tert-Butyl Ester (156). A solution of LiEt3BH (1.0 M in THF, 1.6

mL) was slowly added over 30 min via syringe to a solution 155 (212.8 mg, 0.74

mmol) in THF (4 mL) at −78 °C. The reaction mixture was stirred for 30 min at −78

°C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 2 mL) was added. The

aqueous layer was extracted with EtOAc (3 × 2 mL) and the combined organic

extracts were washed with saturated aqueous NaCl (2 mL), dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 15% EtOAc/hexane) to afford pure 156 (203.7 mg,

149

1 99%) as a colorless oil: H NMR (CDCl3, 500 MHz, 60 °C) δ 5.88-5.70 (br, 1H), 4.50

(br t, 1H), 4.02 (br, 1H), 3.87-3.73 (br, 1H), 3.07 (br t, 1H), 2.45-2.37 (br, 1H), 1.46

13 (s, 3H), 1.43 (s, 9H), 1.33 (s, 3H); C NMR (CDCl3, 125 MHz) δ 153.60, 112.39,

88.60, 80.60, 78.47, 69.55, 49.36, 28.35, 26.69, 26.20; HRMS (ESI) m/z 282.1324

(calcd. for C12H21NO5 + Na: 282.1317).

O OBn

O N CO2t-Bu 6-Benzyloxy-2,2-dimethyltetrahydro-[1,3]dioxolo[4,5- b]pyrrole-4-carboxylic Acid tert-Butyl Ester (157). A slurry of NaH (19.2 mg, 0.80 mmol) was added in one portion to a solution of 156 (102 mg, 0.40 mmol) and n-

Bu4NI (15.0 mg, 0.04 mmol) in THF (4 mL) and DMF (0.4 mL) at 0 °C. The reaction

mixture was stirred for 30 min at 0 °C and benzyl bromide (0.070 mL, 0.60 mmol)

was added. The mixture was stirred for 1 h at 0 °C and saturated aqueous NaHCO3 (4 mL) was added. The aqueous layer was extracted with EtOAc (3 × 4 mL) and the combined organic extracts were washed with saturated aqueous NaCl (4 mL), dried

(Na2SO4), filtered, and the filtrated was evaporated in vacuo. The residue was

purified by flash chromatography (1 × 10 cm silica, 15% EtOAc/hexane) to afford

1 pure 157 (84.3 mg, 89%) as a white solid: H NMR (CDCl3, 500 MHz) δ 7.39-7.25

(br, 5H), 5.80-5.75, 5.67-5.61(2 parts, br, 1H), 4.71-4.54 (br m, 3H), 3.85-3.71 (br m,

1H), 3.70-3.63 (br t, 1H), 3.23 (br t, 1H), 1.51 (br d, 3H), 1.43 (s, 9H), 1.34 (br s, 3H).

13C NMR (CDCl3, 125 MHz) δ 153.91 (153.68, rotamer), 137.52 (137.47), 128.53,

150

128.07 (128.00), 112.71 (112.58), 88.20 (88.05), 80.56, 78.29, 75.38 (74.85), 72.10,

46.75 (45.94), 28.33, 27.05, 26.37 (26.10); HRMS (ESI) m/z 372.1798 (calcd. for

C19H27NO5 + Na: 372.1787).

CHO MeO

OMe 2,4-Dimethoxy-3-methylbenzaldehyde (159). A solution of SnCl4

(1.0 M in CH2Cl2, 26 mL) was added dropwise to a solution of 2,6-dimethoxytoluene

(3.04 g, 20.0 mmol) in CH2Cl2 (60 mL) at 0 °C. The reaction mixture was stirred for

1 h at 0 °C and Cl2CHOCH3 (2.4 mL, 26.0 mmol) was slowly added. The reaction

mixture was stirred for 1 h at 0 °C and was allowed to warm to 25 °C. The reaction

mixture was poured on crushed ice (40 mL) and was stirred for 1 h at 25 °C. The

aqueous layer was extracted with EtOAc (3 × 200 mL) and the combined organic

extracts were washed with saturated aqueous NaCl (50 mL), saturated aqueous

NaHCO3 (50 mL), a saturated NaCl solution (50 mL), dried (Na2SO4), filtered, and

the filtrate was evaporated in vacuo. Recrystallization from hexane (100 mL) afforded

1 159 as colorless, needle-like crystals (3.33 g, 93%): H NMR (CDCl3, 500 MHz) δ

10.21 (s, 1H), 7.74 (d, 1H, J = 9.2 Hz), 6.74 (d, 1H, J = 9.2 Hz), 3.89 (s, 3H), 3.84 (s,

3H), 2.15 (3H).

151

OH MeO

OMe 2,4-Dimethoxy-3-methylphenol (160). m-Chloroperbenzoic acid

(10.6 g, 37 mmol) was added in one portion to a solution of 159 (3.33 g, 18.5 mmol)

in CH2Cl2 (50 mL) at 0 °C. The reaction mixture was stirred for 3 h at 0 °C. The reaction mixture was washed with saturated aqueous K2CO3 (3 × 10 mL) and was

evaporated in vacuo. The residue was dissolved in MeOH (40 mL) at 0 °C and KOH

(1.57 g, 28 mmol) was added. The reaction mixture was stirred for 1 h at 0 °C and

was evaporated in vacuo. The residue was dissolved in ether (300 mL) and washed

with dilute aqueous HCl (0.1 M, 100 mL). The organic layer was washed with water

(2 × 100 mL), saturated aqueous NaCl (100 mL), dried (Na2SO4), filtered, and the

filtrate was evaporated in vacuo. The residue was distilled under vacuum to afford

1 160 as a colorless oil (2.46g, 79%, 98 °C/0.2 mm Hg): H NMR (CDCl3, 500 MHz) δ

6.73 (d, 1H, J = 8.6 Hz), 6.53 (d, 1H, J = 8.6 Hz), 5.21 (br, 1H), 3.76 (s, 3H), 3.75(s,

3H), 2.16 (s, 3H).

OCH2OMe MeO

OMe 1,3-Dimethoxy-4-methoxymethoxy-2-methylbenzene (161).

A slurry of NaH (3.26 g, 136 mmol) was added in one portion to a solution of 160

(11.4 g, 68.0 mmol) in THF (340 mL) and DMF (34 mL) at 0 °C. The reaction

mixture was stirred 1 h at 25 °C and chloromethyl methyl ether (5.68 mL, 74.8 mmol)

152

was slowly added. The reaction mixture was stirred 1 h at 25 °C and ether (500 mL)

was added. The reaction mixture was washed with saturated aqueous NaHCO3 (100

mL) and saturated aqueous NaCl (100 mL). The organic layer was dried (Na2SO4),

filtered, and the filtrate was evaporated in vacuo. The residue was distilled under

vacuum to afford 160 as a colorless oil (11.72 g, 89%, 64 °C/0.2 mm Hg): 1H NMR

(CDCl3, 500 MHz) δ 6.90 (d, 1H, J = 9.2 Hz), 6.51 (d, 1H, J = 9.2 Hz), 5.12 (s, 2H),

13 3.79 (s, 3H), 3.76 (s, 3H), 3.50(s, 3H), 2.13 (s, 3H); C NMR (CDCl3, 125 MHz) δ

153.64, 149.28, 144.38, 129.85, 114.49, 105.48, 96.08, 60.49, 56.11, 55.85, 8.95;

HRMS (ESI) m/z 235.0933 (calcd. for C11H16O4 + Na: 235.0947).

MeOCH2OOH MeO

OMe 1-(3,5-Dimethoxy-2-methoxymethoxy-4-methylphenyl)-

prop-2-en-1-ol (166). A solution of n-BuLi (1.91 M in hexane, 0.63 mL, 1.20 mmol)

was slowly added to a solution of 166 (158 mg, 0.75 mmol) in THF (3 mL) at 0 °C.

The reaction mixture was stirred for 1 h at 0 °C and freshly distilled acrolein (0.067

mL, 1.0 mmol) was added. The reaction mixture was allowed to warm to 25 °C and

saturated aqueous NaHCO3 (2 mL) was added. The aqueous layer was extracted with

EtOAc (3 × 2 mL) and the combined organic extracts were dried (Na2SO4), filtered,

and the filtrate was evaporated in vacuo. The residue was purified by flash

chromatography (1 × 10 cm silica, 10% EtOAc/hexane containing 5% Et3N) to afford

1 pure 166 (146 mg, 79 %) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 6.57 (s,

153

1H), 6.09 (m, 1H), 5.52 (s, 1H), 5.42 (d, 1H, J = 17.1 Hz), 5.23 (d, 1H, J = 11.0 Hz),

5.07 (d, 1H, J = 6.1 Hz), 5.03 (d, 1H, J = 6.1 Hz), 3.76 (s, 3H), 3.74 (s, 3H), 3.54 (s,

13 3H), 3.22 (br, 1H), 2.09 (s, 3H); C NMR (CDCl3, 125 MHz) δ 154.75, 151.31,

141.98, 139.04, 134.03, 120.66, 114.62, 104.20, 99.71, 69.09, 60.22, 57.52, 55.68,

8.89; HRMS (ESI) m/z 291.1180 (calcd. for C14H20O5 + Na: 291.1203).

MeOCH2OOAc MeO

OMe Acetic acid 1-(3,5-dimethoxy-2-methoxymethoxy-4-

methylphenyl)allyl ester (167). Acetyl choloride (0.071 mL, 1.0 mmol) was slowly

added to a solution of 166 (70.7 mg, 0.26 mmol) and Et3N (0.21 mL, 1.5 mmol) in

CH2Cl2 (1 mL) at 25 °C. The reaction mixture was stirred for 3 h at 25 °C and

saturated aqueous NaHCO3 (1 mL) was added. The aqueous layer was extracted with

EtOAc (3 × 1 mL) and the combined organic extracts were dried (Na2SO4), filtered,

and the filtrate was evaporated in vacuo. The residue was purified by flash

chromatography (1 × 10 cm silica, 10% EtOAc/hexane + 5% Et3N) to afford pure 167

1 (76.6 mg, 89%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 6.69 (d, 1H, C1’-H,

J = 5.5 Hz), 6.57 (s, 1H, Ar-H), 6.02 (m, 1H, C2’-H), 5.24 (d, 1H, C3’-H, J = 17.7

Hz), 5.20 (d, 1H, C3’-H, J = 10.4 Hz), 5.06 (s, 2H, COCH2OAr), 3.78 (s, 3H, Me-H),

3.76 (s, 3H, Me-H), 3.59 (s, 3H, Me-H), 2.10 (s, 3H, Me-H), 2.09 (s, 3H, MeCO-H);

13 C NMR (CDCl3, 125 MHz) δ 169.68, 154.62, 151.51, 141.50, 136.11, 129.84,

154

121.26, 115.94, 104.29, 99.50, 70.66, 60.26, 57.67, 55.73, 21.18, 8.95; HRMS (ESI)

m/z 333.1309 (calcd. for C16H22O6 + Na: 333.1314).

MeOCH2O MeO Sn(n-Bu)3

OMe Tri-n-butyl-[3-(3,5-dimethoxy-2-methoxy-

methoxy-4-methylphenyl)allyl]stannane (168). A solution of SmI2 (0.1 M in THF,

1.78 mL) was added over 30 min to a solution of 167 (22.0 mg, 0.071 mmol),

Pd2(PPh3)4 (4.1 mg, 3.55 µmol), and n-Bu3SnCl (0.028 mL, 0.11 mmol) in THF (0.3 mL) at 25 °C. The reaction mixture was stirred for 12 h at 25 °C and was evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 5%

1 Et3N/hexane) to afford pure 168 (26.1 mg, 77 %) as a pale yellow oil: H NMR

(CDCl3, 500 MHz) δ 6.61 (s, 1H), 6.49 (d, 1H, J = 15.9 Hz), 6.33 (m, 1H), 4.99 (s,

2H), 3.78 (s, 3H), 3.76 (s, 3H), 3.38 (s, 3H), 2.09 (s, 3H), 1.97 (d, 2H, J = 8.5 Hz),

13 1.49 (m, 6H), 1.29 (m, 6H), 0.87 (m, 15H); C NMR (CDCl3, 125 MHz) δ 154.43,

151.89, 140.36, 131.47, 130.21, 119.39, 118.55, 101.85, 99.37, 60.32, 57.57, 55.60,

29.03, 27.29, 16.41, 13.65, 9.50, 8.89; HRMS (ESI) m/z 565.2299 (calcd. for

C26H46O4Sn + Na: 565.2316).

155

MeOCH O 2 H OH MeO OBn N CO t-Bu OMe 2 4-Benzyloxy-2-[1-(3,5-dimethoxy-2-

methoxymethoxy-4-methylphenyl)allyl]-3-hydroxypyrrolidine-1-carboxylic Acid tert-Butyl Ester (169). Boron trifluoride etherate (0.38 mL, 3.0 mmol) was added

over 30 min to a solution of 157 (797.2 mg, 3.0 mmol) in CH2Cl2 (15 mL) at −78 °C.

The reaction mixture was stirred for 30 min at −78 °C and stannane 168 (2.15 g, 3.98

mmol) was added. The reaction mixture was stirred for 30 min at −78 °C and aqueous

K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 15 mL) was added. The aqueous layer was

extracted with EtOAc (3 × 15 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (5 × 10 cm silica, 15% EtOAc/hexane) to afford pure 169

1 (1.22 g, 75%) as a pale yellow oil: H NMR (CDCl3, 500 MHz) δ 6.56-6.45 (br, 1H),

6.19-6.03 (m, 1H), 5.05-4.87 (m, 4H), 4.54-4.438 (m, 2H), 4.27-3.36 (br, 16H), 2.59-

13 2.37 (br, 1H), 2.08 (br, 3H), 1.49-1.40 (br, 9H); C NMR (CDCl3, 125 MHz) δ

155.17, 154.76, 151.61, 141.99, 138.78, 137.15, 132.06, 128.54, 128.10, 127.80,

119.58, 115.76, 104.74, 99.63, 79.27, 71.95, 70.71, 67.80, 60.16, 57.60, 55.81, 48.48,

44.82, 29.66, 28.44, 8.84; HRMS (ESI) m/z 566.2680 (calcd. for C30H41NO8 + Na:

566.2724).

156

OH MeOCH O 2 H OH MeO OBn N CO t-Bu OMe 2 4-Benzyloxy-2-[1-(3,5-dimethoxy-2- methoxymethoxy-4-methylphenyl)-2-hydroxyethyl]-3-hydroxypyrrolidine-1-carboxylic Acid tert-Butyl Ester (175). A stream of O3 in O2 was slowly bubbled into a solution of 169 (164 mg, 0.30 mmol) in MeOH (7 mL) at −78 °C. The progress of the reaction was carefully monitored by TLC and N2 was bubbled into the reaction mixture for 30 min immediately after the reaction was judged complete. The reaction mixture was warmed to 0 °C and NaBH4 (330 mg, 8.0 mmol) was added. The reaction mixture was stirred for 1 h at 0 °C and was evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 20% EtOAc/hexane) to

1 afford pure 175 (92.5 mg, 62%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ

7.33-7.25 (m, 5H), 6.98 (s, 1H), 4.95-4.90 (app q, 2H), 4.58-4.52 (br, 1H), 4.52-4.43

(app q, 2H), 4.36-4.31 (m, 1H), 4.31-4.26 (br, 1H), 3.87-3.80 (br, 1H), 3.79 (br, 1H),

3.78 (s, 3H), 3.72 (s, 3H), 3.65-3.58 (m, 1H), 3.58-3.51 (br, 1H), 3.48-3.42 (br, 1H),

3.36 (s, 3H), 2.97-2.91 (br d, 1H), 2.58-2.40 (br, 1H), 2.12 (s, 1H), 2.09 (s, 3H), 1.54-

13 1.44 (br, 9H); C NMR (CDCl3, 125 MHz) δ 157.12, 154.66, 151.03, 141.75, 137.08,

132.02, 128.53, 128.12, 127.76, 119.27, 105.71, 99.74, 80.73, 71.95, 71.49, 64.61,

62.81, 60.13, 57.35, 55.68, 48.55, 41.20, 28.33, 8.80; HRMS (ESI) m/z 570.2668

(calcd. for C29H41NO9 + Na: 570.2674).

157

OSit-BuPh MeOCH O 2 2 H OH MeO OBn N CO t-Bu OMe 2 4-Benzyloxy-2-[2-(tert-butyldiphenylsilanyloxy)-1-

(3,5-dimethoxy-2-methoxymethoxy-4-methylphenyl)ethyl]-3-hydroxy-pyrroli-

dine-1-carboxylic Acid tert-Butyl Ester (176). tert-Butyldiphenylsilyl chloride

(0.063 mL, 0.24 mmol) was added to a solution of 175 (102 mg, 0.19 mmol), Et3N

(0.034 mL, 0.24 mmol), and 4-(dimethylamino)pyridine (10 mg) in CH2Cl2 (1 mL) at

25 °C. The reaction mixture was stirred for 3 h at 25 °C and saturated aqueous

NaHCO3 (1 mL) was added. The aqueous layer was extracted with EtOAc (3 × 1 mL)

and the combined organic extracts were dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm

silica, 10% EtOAc/hexane) to afford pure 176 (149 mg, 99%) as a colorless oil: 1H

NMR (CDCl3, 500 MHz) δ 7.60-7.50 (m, 2H), 7.41-7.20 (m, 13H), 6.58-6.38 (br,

1H), 5.10-4.90 (m, 2H), 4.54-4.44 (m, 2H), 4.27-4.18 (br, 1H), 4.14-3.98 (br, 2H),

3.97-3.82 (br, 2H), 3.74 (s, 3H), 3.69-3.64 (br, 3H), 3.57-3.48 (br, 4H), 3.45-3.32 (br,

1H), 2.52-3.38 (br, 1H), 2.20-2.12 (br, 3H), 1.44-1.18 (br, 9H), 1.00-0.86 (br, 9H);

13 C NMR (CDCl3, 125 MHz) δ 155.38, 155.19, 154.68, 151.72, 151.63, 143.13,

142.83, 137.28, 135.60, 135.52, 133.75, 133.62, 133.46, 133.36, 131.87, 129.56,

129.46, 129.30, 128.53, 128.06, 127.72, 127.63, 127.40, 127.27, 119.36, 119.10,

104.22, 99.70, 99.64, 79.83, 79.35, 72.10, 71.12, 66.20, 65.80, 60.23, 57.32, 55.65,

55.50, 48.38, 48.09, 28.34, 28.08, 26.83, 26.72, 19.11, 8.89 (many are rotamers);

HRMS (ESI) m/z 808.3880 (calcd. for C45H59O9NSi + Na: 808.3851). 158

OSit-BuPh OH 2 H OH MeO OBn HN

OMe 4-Benzyloxy-2-[2-(tert-butyldiphenylsilanyloxy)-1-

(2-hydroxy-3,5-dimethoxy-4-methylphenyl)ethyl]pyrrolidin-3-ol (177). Trifluo- acetic acid (1 mL) was added to a solution of 176 (149 mg, 0.19 mmol) in CH2Cl2 (1

mL) at 25 °C. The reaction mixture was stirred for 10 min at 25 °C, cooled to −40 °C

and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 5 mL) was added followed by the

addition of NaOH (520 mg). The aqueous layer was extracted with EtOAc (3 × 1 mL)

and the combined organic extracts were dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm

silica, 50% EtOAc/hexane) to afford pure 177 (97.1 mg, 79%) as a pale yellow oil: 1H

NMR (CDCl3, 500 MHz) δ 7.60 (m, 3H), 7.44-7.23 (m, 12H), 6.19 (s, 1H), 4.53 (s,

2H), 4.15 (m, 1H), 3.94 (m, 2H), 3.74 (s, 3H), 3.69 (s, 3H), 3.49 (m, 1H), 3.25 (m,

1H), 2.98 (dd, 1H, J = 9.8, 4.9 Hz), 2.70 (d, 1H, J = 4.9 Hz), 2.49 (dd, 1H, J = 9.8, 4.9

13 Hz), 2.09 (s, 3H), 0.96 (s, 9H); C NMR (CDCl3, 125 MHz) δ 153.44, 150.91,

143.74, 137.38, 135.55, 134.76, 133.62, 133.16, 129.73, 129.45, 128.50, 128.05,

127.72, 127.69, 127.54, 124.42, 119.54, 105.8179.11, 72.82, 72.07, 65.29, 64.10,

60.85, 56.25, 53.78, 47.66, 30.29, 26.78, 19.11, 8.85; HRMS (ESI) m/z 642.3195

(calcd. for C38H47NO6Si + H: 642.3245).

159

O OSit-BuPh2 MeO H OH N O OBn 2-Benzyloxy-8-(tert-butyldiphenylsilanyloxymethyl)-

1-hydroxy-6-methoxy-5-methyl-2,3,8,8a-tetrahydro-1H-3a-azacyclopenta[a]- indene-4,7-dione (180). A solution of (NH4)3Ce(NO3)6 (11.0 mg, 8.7 µmol) in H2O

(0.08 mL) was added to a solution of 177 (5.6 mg, 8.7 µmol) in MeCN (0.2 mL) at 25

°C. The reaction mixture was stirred for 5 min at 25 °C, cooled to 0 °C and saturated aqueous NaHCO3 (0.2 mL) and EtOAc (0.5 mL) were added. The aqueous layer was

extracted with EtOAc (3 × 0.5 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (0.5 × 10 cm silica, 5% EtOAc/hexane) to afford pure 180

1 (1.1 mg, 21%) as a pink oil: H NMR (CDCl3, 500 MHz) δ 7.74 (m, 3H), 7.49-7.34

(m, 12H), 5.14 (m, 1H), 5.04 (d, 2H, J = 3.7 Hz), 4.84 (d, 1H, J = 11.6 Hz), 4.70 (d,

1H, J = 12.2 Hz), 4.47 (dd, 1H, J = 12.2, 7.3 Hz), 4.31 (dd, 1H, J = 12.2, 7.3 Hz),

4.04 (dd, 1H, J = 12.2, 7.3 Hz), 4.01 (s, 3H), 3.42 (dd, 1H, J = 14.6, 7.3Hz), 3.35 (dd,

1H, J = 14.0, 6.7 Hz), 2.77 (d, 1H, J = 1.8 Hz), 2.13 (s, 1H), 2.00 (s, 3H), 1.15 (s,

9H); HRMS (ESI) m/z 646.2655 (calcd. for C37H41NO6Si + Na: 646.2601).

160

O OSit-BuPh2 MeO OH N O OBn 2-Benzyloxy-8-(tert-butyldiphenylsilanyloxy-

methyl)-1-hydroxy-6-methoxy-5-methyl-2,3-dihydro-1H-3a-azacyclopenta[a]in-

dene-4,7-dione (181). A solution of MeSO2Cl (2.8 µl, 0.036 mmol) in CH2Cl2 (0.1 mL) was slowly added to a solution of 180 (7.4 mg, 0.012 mmol) and Et3N (8.4 µl,

0.060 mmol) in CH2Cl2 (0.1 mL) at −20 °C. The reaction mixture was stirred for 3 h

at 25 °C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 0.2 mL) was added. The

aqueous layer was extracted with EtOAc (3 × 0.2 mL) and the combined organic

extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash chromatography (0.5 × 10 cm silica, 5% EtOAc/hexane) to afford 181 with some contamination of 180 (3.9 mg, 56%) as a pink oil: 1H NMR

(CDCl3, 500 MHz) δ 7.65 (m, 3H), 7.43-7.26 (m, 12H), 5.21 (s, 1H), 5.07 (d, 1H, J =

15.3 Hz), 4.95 (d, 1H, J = 15.0 Hz), 4.75 (d, 1H, J = 11.9 Hz), 4.61 (d, 1H, J = 11.9

Hz), 4.54 (m, 1H), 4.44 (dd, 1H, J = 13.1, 6.7 Hz), 4.42 (dd, 1H, J = 13.1, 4.0 Hz),

3.97 (s, 3H), 3.35 (d, 1H, J = 1.2 Hz), 1.94 (s, 3H), 1.11 (s, 9H); HRMS (ESI) m/z

644.2418 (calcd. for C37H39NO6Si + Na: 644.2445).

161

CHAPTER 6

SYNTHETIC STUDIES TOWARD ISOMITOMYCINS

6.1 Introduction

As discussed previously, although our strategy was quite successful in

building the mitomycin ring system, it will not allow the installation of the methoxy

group at the C9a center that is a required functionality for mitomycins. Examination of the literature precedent1 revealed that direct oxidation after the mitomycin ring

system was built, either at the quinone or aromatic stage, would not lead to a

successful installation of methoxy/hydroxy functionality at C9a center necessary for

synthesizing mitomycin A, B or C. (Figure 6.1) Thus, an oxidative introduction of

methoxy/hydroxy functionality must occur before the mitomycin ring system is

installed.

1. Danishefsky, S. J.; Schkeryantz, J. M. “Chemical Explorations Driven by An Enchantment with Mitomycinoids – A Twenty Year Account,” Synlett 1995, spec. issue, 475-490; and references cited therein. 162

9a N X N

O O OR

9a N X N O O

Figure 6.1 Attempted C9a Methoxy Installation in Literature

In 1987, Kono et al discovered an interconversion between mitomycin A and isomitomycin A.2 (Figure 6.2) We felt this interesting interconversion could be useful for our approach to the mitomycins.

2. Kono, M.i; Saitoh, Y.; Shirahata, K.; Arai, Y.; Ishii, S. “Albomitomycin A and Isomitomycin A. Products of Novel Intramolecular Rearrangement of Mitomycin A,” J. Am. Chem. Soc. 1987, 109, 7224-7225.

163

OCONH O OCONH O 2 2 OMe MeO H OMe MeO N NNH N O O

mitomycin A isomitomycin A

O OCONH2 MeO OMe

N N O albomitomycin A

Figure 6.2 Interconversion between Mitomycin A and Isomitomycin A

There were obvious advantages to synthesize the mitomycins through

isomitomycin intermediates. First, this approach would give us flexibility in terms of

timing and methods to introduce the C9a methoxy/hydroxy group; second, the

approach through isomitomycins would not require significant modification of our

established strategy. We hoped an intermediate synthesized in the mitosane synthesis could be used directly in this approach.

Apparently there are two timings for introducing the methoxy/hydroxy group functionality on C9a position: prior to or after the isomitomycin ring formation. To

avoid the potential problem of elimination of methanol, we felt a late stage

164

installation of this sensitive functionality would be desirable. The original retrosynthetic analysis of constructing isomitomycin A is detailed in Figure 6.3.

165

1 OCONH2 OR O OMe O H H R2 MeO N MeO N

N N O O isomitomycin A 185

OR1 OR1 O MeOCH2O NHR3 H R2 H MeO N MeO OH N N H R2 O H OH OMe 186 187

MeOCH O 2 H OH MeO OBn N CO t-Bu OMe 2 169

Figure 6.3 Retrosynthetic Analysis

166

We envisioned that isomitomycin A would be derived from isomitomycin-

type compound 185, which in turn could be constructed through either an

intramolecular Mitsunobu reaction or an SN2 displacement of a mesylate. The six membered piperidine ring could be formed by intramolecular Michael addition of compound 187 in a manner analogous to our mitosane synthesis. We supposed this addition process would be easier since this 6-endo-trig addition is favored according

to Baldwin’s rules. We also recognized that the trans aminoalcohol relationship in

compound 187 could be potentially derived from the cis diol in compound 169.

Compound 169 was synthesized in the previous mitosane synthesis with good

stereocontrol and synthetic efficiency. We hoped this could simplify our efforts

toward constructing isomitomycin system.

6.2 Efforts of Aminoalcohol Functionalities Installation From

Mitosane Intermediates

We started our efforts of installing the trans-aminoalcohol functionalities

from alkene 169. However both Mitsunobu reaction and the conversion of the

secondary alcohol to mesylate failed. We suspected the sensitive homobenzylic

double bond caused the reaction to fail as it could easily migrate to form a conjugated

system with the aromatic ring. (Figure 6.4)

167

MeOCH O MeOCH O 2 H OH 2 H N3 MeO Zn(N3)2⋅(pyr)2, DIAD MeO OBn X OBn N PPh3, THF, 40 °C N CO t-Bu Boc OMe 2 OMe 169 188

MeOCH O MeOCH O 2 H OH 2 H OMs MeO MeSO2Cl, pyridine MeO OBn X OBn N CH2Cl2, 0 °C N CO t-Bu CO t-Bu OMe 2 OMe 2 169 189

Figure 6.4 Attempted Aminoalcohol Functionalities Installation on Alkene 169

168

Another obvious alternative candidate for this cis-diol to trans-aminoalcohol

transformation is the terminal double bond cleavage product. We continued our

investigation with silyl ether 176. Direct azide displacement under Mitsunobu

conditions at room temperature afforded only unreacted starting material. At higher temperatures (e.g., 60 °C) the reaction gave a less polar compound in 68% yield that

was later confirmed as the dehydration product by mass spectrometric analysis.3

Mesylation on the secondary alcohol was successful, however azide displacement at room temperature did not proceed at room temperature and at higher temperatures

(e.g., 40 °C) elimination occurred. (Figure 6.5)

3. We don’t have affirmative evidence for the double bond position. We would guess the migration product was likely. 169

OSit-BuPh OSit-BuPh MeOCH O 2 MeOCH O 2 2 H OH 2 MeO Zn(N3)2⋅(pyr)2, DIAD MeO OBn OBn N PPh3, THF, 60 °C N CO t-Bu 68% CO t-Bu OMe 2 OMe 2 176 190

NaN3, DMF 40 °C 87%

OSit-BuPh OSit-BuPh MeOCH O 2 MeOCH O 2 2 HHOH 2 OMs MeO MeSO2Cl, pyridine MeO OBn OBn N N CH2Cl2, 0 °C CO2t-Bu CO2t-Bu OMe 76% OMe 176 191

Figure 6.5 Attempted Aminoalcohol Functionalities Installation on Silyl Ether 176

These unsuccessful attempts illustrate the difficulties associated with working with this highly functionalized pyrrolidine ring. They also suggested the trans- aminoalcohol functionalities must be installed prior to the formation of the pyrrolidine ring.

170

6.3 Installation of trans-Aminoalcohol Functionalities by Asymmetric

Sharpless Aminohydroxylation

With the failure to convert the secondary alcohol to the corresponding amine, we realized that an earlier installation of the aminoalcohol groups would be necessary.

Consequently, we revised our plan for the construction of 187. In our new plan, compound 187 would be derived by Lewis acid catalyzed addition of cinnamyl stannane 168 to a 2-hydroxypyrrolidine 192 in which the trans-aminoalcohol

functionalities had already been installed. 2-Hydroxypyrrolidine 192 would be most

conveniently synthesized from aminoalcohol 193. (Figure 6.6) There are a variety methods available for synthesizing aminoalcohols in literature: nucleophilic azide opening of epoxides;4 enzymatic aminohydroxylation;5 transformation from amino acids6 and Sharpless asymmetric aminohydroxylation.7 Of those, we were particularly

4. For example: Acena, J. L.; Arjona, O.; Leon, M. L.; Plumet, J. “Total Synthesis of (+)-7- Deoxypancratistatin from Furan,” Org. Lett. 2000, 2, 3683-3686.

5. Kimura, T.; Vassilev, V. P.; Shen, G. –J.; Wong, C. –H. “Enzymatic Synthesis of β-Hydroxy-α- amino Acids Based on Recombinant D- and L- Threonine Aldolases,” J. Am. Chem. Soc., 1997, 119, 11734-11742.

6. Chen, S. -T.; Wang, K. -T. “A New Synthesis of O-Benzyl-L-serine,” Synthesis, 1989, 36-37.

7. Schlingloff, G.; Sharpless, K. B. “Asymmetric Aminohydroxylation,” Asym. Oxidation React. 2001, 104-114; Andersson, M. A.; Epple, R.; Fokin, V. V.; Sharpless, K. B. “A New Approach to Osmium-Catalyzed Asymmetric Dihydroxylation and Aminohydroxylation of Olefins,” Angew. Chem. Int. Ed. 2002, 41, 472-475; Fokin, V. V.; Sharpless, K. B. “A Practical and Highly Efficient Aminohydroxylation of Unsaturated Carboxylic Acids,” Angew. Chem. Int. Ed. 2001, 40, 3455- 3457; Rubin, A. E.; Sharpless, K. B. “A Highly Efficient Aminohydroxylation Process,” Angew. Chem. Int. Ed. 1997, 36, 2637-2640; Li, G.; Angert, H. H.; Sharpless, K. B. “N-Halocarbamate Salts Lead to More Efficient Catalytic Asymmetric Aminohydroxylation,” Angew. Chem. Int. Ed. 1996, 35, 2813-2817.

171

interested in the Sharpless asymmetric aminohydroxylation primarily because of its convenience, simplicity and efficiency.

172

OR1 MeOCH2O NHR3 MeO 9 9a OH N R2 OMe 187

Sn(n-Bu)3

3 MeOCH2O R HN OH MeO + HO N R2 OMe 168 192

O OH NHR2 MeO NHR3 193

Figure 6.6 Retrosynthetic Analysis

173

We realized that it could be difficult to achieve effective stereocontrol in the

addition of stannane 168 to 2-hydroxypyrrolidine 192. Because of the planned oxidation at C9a position, the absolute stereochemistry at C9a would become relatively less important. The C9 absolute stereochemistry would be important, however, because the R configuration would lead to mitomycin A and C whereas the

S configuration product would lead to mitomycin B.

We started our synthesis from commercially available methyl 4- bromocrotonate. Direct azide displacement by NaN3 in dioxane/water proceeded

smoothly and the corresponding azide 195 was converted to t-butyl carbamate

protected aminocrotonate 196 in one step with some contamination of olefin over-

reduction by-product.

O O NaN3 Br N MeO MeO 3 H2O, dioxane, 25 °C 194 87% 195

O (t-BuOCO)2O, Et3SiH, Pd(OH)2/C NHCO t-Bu MeO 2 MeOH, 25 °C 63% 196 + 30% olefin reduction product

Figure 6.7 Synthesis of the Sharpless Aminohydroxylation Precursor

174

Aminocrotonate 196 was subjected to the Sharpless aminohydroxylation

condition. After extensive optimization, this reaction produced only about 15% of the

desired aminohydroxylation product with about 30% dihydroxylation by-product.

(Figure 6.8) Crotonates 194 and 195 were also tested, but with no success in producing the desired aminohydroxylation product in appreciable yield.

O O OH NaOH, t-BuOCl, K2OsO4⋅2H2O NHCO t-Bu NHCO t-Bu MeO 2 MeO 2 H2NCO2Bn, (DHQD)2AQN n-PrOH, H2O, 25 °C NHCO2Bn 196 15% 197 + 30% dihydroxylation product

Figure 6.8 Initial Effort for Sharpless Asymmetric Aminohydroxylation

We suspected the failure of Sharpless aminohydroxylation was related to the

acidic proton of the terminal NHCO2t-Bu group. We assume the NHCO2t-Bu group

could also react with t-BuOCl to form a product that was no longer a substrate for

Sharpless aminohydroxylation. This assumption prompted us to double protect the

terminal amine.

The choice of the second protecting group was difficult. We felt another

carbamate or an amide would make the nitrogen electron deficient and thereby

175

potentially change the regioselectivity of the aminohydroxylation.8 Rather, we would prefer an alkyl protecting group that would survive the Sharpless aminohydroxylation conditions and could be easily removed later. Thus we decided to investigate benzyl protecting groups. We had hoped the benzyl group could be removed by hydrogenolysis.

Sharpless aminohydroxylation of crotonate ester 198 with benzyl and t-butyl carbamate double protecting groups on the terminal amine went on smoothly to afford the desired aminoalcohol 199 in 55% yield. The secondary alcohol was converted to acetate. By comparing the 1H NMR spectra of both compounds we determined the

regioselectivity was in desired sense by the observation of a 0.9 ppm downfield shift

of the CH-OAc proton in the 1H NMR spectrum of 200. (Figure 6.9)

O Bn N NaOH, t-BuOCl, K2OsO4⋅2H2O, H2NCO2Bn MeO CO2t-Bu (DHQD)2AQN, n-PrOH, H2O, 25 °C 198 55%

O OH Bn AcCl, pyridine O OAc Bn N N MeO CO t-Bu 2 CH2Cl2, 25 °C MeO CO2t-Bu NHCO2Bn NHCO2Bn 199 200

Figure 6.9 Sharpless Asymmetric Aminohydroxylation

8. Sometimes the regioselectivity of Sharpless asymmetric dihydroxylation is hard to predict. Morgan, A. J.; Masse, C. E.; Panek, J. S. “Reversal of Regioselection in the Sharpless Asymmtric Aminohydroxylation of Aryl Ester Substrates,” Org. Lett. 1999, 1, 1949-1952. 176

Unfortunately we were unable to remove the benzyl protecting group under a

wide variety of reaction conditions. Consequently, we decided to investigate the use

of a p-methoxybenzyl protecting group. We were concerned that the oxidative

Sharpless condition might remove the oxidation-sensitive p-methoxybenzyl group,

however investigation revealed that at 0 °C the Sharpless asymmetric

aminohydroxylation proceeded smoothly to give desired aminoalcohol 203 in 67%

yield. Because of the slight contamination of benzyl carbamate, it was difficult to

determine the enantioselectivity of the reaction. Instead it was determined after the p-

methoxybenzyl group was removed. Chiral HPLC (15% isopropanol/water, 0.46 cm

× 25 cm, OJ, Chiralcel, Daicel Chemical Ind., LTD.) determined the enantiomeric excess of 204 was 92% ee.

177

O O (t-BuOCO) O, Et N p-methoxybenzyl amine 2 3 Br NHPMB MeO MeO CH2Cl2, 25 °C EtN(i-Pr)2, THF, 25 °C 95% 194 67% 201

O PMB NaOH, t-BuOCl, K2OsO4⋅2H2O, H2NCO2Bn N MeO CO t-Bu 2 (DHQD)2AQN, n-PrOH, H2O, 0 °C 67% 202

O OH PMB O OH (NH ) Ce(NO ) N 4 3 3 6 NHCO2t-Bu MeO CO2t-Bu MeO MeCN, H2O, 25 °C NHCO2Bn NHCO2Bn 83% 203 204

Figure 6.10 Use of p-Methoxybenzyl Protecting Group

6.4 Lewis Acid Catalyzed Addition of Stannane to Iminium Ion

With the aminoalcohol 204 in hand, we continued our investigation of the preparation of the pyrrolidine iminium ion precursor. Treatment of aminoalcohol 204 with large excess of i-Bu2AlH in THF at −78 °C smoothly effected the cyclization to afford the unstable 2-hydroxypyrrolidine 205 in 79% yield. (Figure 6.11)

178

O OH BnO2CHN OH i-Bu AlH NHCO t-Bu 2 MeO 2 THF, −78 °C HO N NHCO2Bn 79% CO2t-Bu 204 205

Figure 6.11 Reductive Cylization

The Lewis acid catalyzed reaction of 2-hydroxypyrroline 205 and stannane

168 was less efficient than in our earlier investigations. Because of the instability of

the 2-hydroxypyrrolidine 205, a different procedure was adopted for effective

coupling: stannane 168 and 2-hyrodroxypyrrolidine 205 were pre-mixed in CH2Cl2 at

–78 °C prior to the addition of Lewis acid. Among the Lewis acids tested, only

BF3⋅Et2O was effective at catalyzing the coupling reaction. Weaker Lewis acid such

as MgBr2⋅Et2O were not strong enough to promote the reaction and stronger Lewis

acid such as SnCl4 reacted unproductively with the unstable 2-hydroxypyrrolidine

prior to coupling. The yield for BF3⋅Et2O catalyzed reaction was moderate and a

mixture of two chromatographically inseparable disastereomers was produced.

(Figure 6.12) Because of the line broadening in the 1H NMR signal, we had no

information about the diastereoselectivity and the absolute stereochemistry was not

obtainable by NMR. The two diastereomers were carried through the subsequent

investigations.

179

Sn(n-Bu)3 BnO2CHN OH MeOCH O MeOCH O 2 2 NHCO2Bn MeO BF ⋅Et O MeO HO N 3 2 + OH CO2t-Bu CH2Cl2, −78 °C N 56% CO t-Bu OMe OMe 2 205 168 206

Figure 6.12 BF3⋅Et2O Catalyzed Coupling Reaction

6.5 Studies Toward the Installation of Isomitomycin Ring

It proved difficult to accomplish synthetic transformations with aminoalcohol

206 because of the sensitive homobenzylic double bond. Cleavage of the double bond

and transformation to alcohol functionality became our priority task. However

ozonolytic cleavage of the terminal double bond followed by NaBH4 reduction based

on our established protocol gave a very low yield of primary alcohol 207. (Figure

6.13)

180

OH MeOCH2O MeOCH O NHCO2Bn 2 NHCO2Bn MeO 1. O3, MeOH, −78 °C MeO OH OH N 2. NaBH4, MeOH, 0 °C N CO2t-Bu 13% CO t-Bu OMe OMe 2 206 207

Figure 6.13 Ozone Cleavage of the Terminal Double Bond

It would be difficult to continue our investigation with such low yields (13%

and 56%) in two key steps. We compared the ozonolysis reaction with the reaction of

169 (Figure 5.17) and suspected that a double protecting group on the NHCO2Bn

amine might be necessary. Besides complicating the situation, a selective double

protection at this stage would be difficult to achieve and the choices of protecting

groups were more limited. Instead we adjusted our synthetic plan and decided to form

the aziridine ring at this stage of the synthesis. We hoped that the relatively neutral

Mitsunobu reaction conditions would allow a clean transformation without affecting

the terminal double bond. Indeed, the aminoalcohol cyclized smoothly to afford

aziridine 208 in 79% yield. (Figure 6.14)

181

MeOCH O MeOCH O 2 NHCO2Bn 2 MeO MeO NCO Bn PPh3, DIAD 2 OH N THF, 25 °C N CO2t-Bu 79% CO2t-Bu OMe OMe DIAD = diisopropyl azodicarboxylate 206 208

Figure 6.14 Aziridine Formation

Ozonolytic cleavage of the double bond of aziridine 208 followed by reduction with NaBH4 went smoothly − as we predicted − to give alcohol 209 in 55% yield. Dichloromethane was added as a co-solvent to improve the reaction yield.

Alcohol 209 was further protected as the corresponding t-butyldiphenylsilyl ether

210. Ceric ammonium nitrate oxidized the electron rich aromatic system to quinone.

(Figure 6.15)

182

MeOCH2O MeO NCO Bn 2 1. O3, MeOH, CH2Cl2, −78 °C

N 2. NaBH4, MeOH, CH2Cl2, 0 °C CO t-Bu OMe 2 55% 208

OH MeOCH2O

MeO NCO2Bn t-BuPh2SiCl, Et3N, DMAP N CH2Cl2, 25 °C CO t-Bu OMe 2 96% 209

OSit-BuPh2 OSit-BuPh2 MeOCH2O O

MeO NCO2Bn MeO NCO2Bn (NH4)3Ce(NO3)6 N N MeCN, H2O, 25 °C CO t-Bu CO2t-Bu OMe 2 87% O 210 211

Figure 6.15 Functional Group Manipulations

183

We continued our investigation to explore the key intramolecular aziridine

Michael addition. Cleavage of the benzyl carbamate protecting group on the aziridine by hydrogenolysis proceeded smoothly. We expected that the quinone moiety would also be reduced, but should be oxidized back to quinone when exposed to air. We had hoped the resulting free aziridine would attack the quinone in situ to cyclize to the six membered isomitomycin ring, and that the newly formed dihydroquinone would be oxidized back to quinone in the presence of oxygen. Unfortunately the reaction didn’t proceed the way we planned. Mass spectrometric analysis suggested the product was the aminoquinone 213, although we were unable to isolate this compound. The reaction residue quickly decomposed on both silica gel (with or without treatment with 5% Et3N/hexane) and on neutral alumina gel. (Figure 6.16)

We were surprised by this result and suspected that the secondary aziridine

nitrogen might not be nucleophilic enough to participate in the desired cyclization.

Consequently we treated the reaction mixture with bases to promote the cylization,

but treatment with bases either destroyed the molecule (e.g., KOt-Bu) or had no effect

at promoting the cyclization (e.g., Et3N, pyridine).

As we examined models of the molecule, we realized that the free rotation of the molecule about the C9-C9a bond was important for the desired Michael addition to occur and that it was possible for the bulky t-butyldiphenylsilyl group to hinder this rotation. Treatment of the reaction mixture with (n-Bu)4NF in THF removed the silyl

protecting group and mass spectrometric analysis of the reaction mixture confirmed

184

the existence of isomitomycin compound 214, which unfortunately decomposed during the purification process.

OSit-BuPh OSit-BuPh O 2 O 2 CO2t-Bu MeO NCO2Bn MeO N X N N Boc O O 212 211

1 atm H2, Pd/C MeOH, 25 °C X

OSit-BuPh O 2 OH O CO2t-Bu MeO 9 NH (n-Bu) NF 9a 4 MeO N N THF, 25 °C CO t-Bu N O 2 O 213 214

Figure 6.16 Possible Formation of Isomitomycin Ring

185

6.7 Conclusion

We have successfully developed an efficient route for the installation of trans aminoalcohol functionalities on the pyrrolidine precursor and investigated the synthesis of isomitomycin ring system. The poor diastereoselectivity and the inseparable feature of the two diastereomers that were formed in the addition of allyl stannane to trans-aminoalcohol iminium ion made investigations very difficult. The lack of C9a methoxy/hydroxy group is a possible cause for the instability of some of the intermediate. The earlier introduction of this important group might be helpful.

186

6.8 Experimental

O PMB N MeO CO2t-Bu 4-[tert-Butoxycarbonyl-(4-methoxybenzyl)amino]-

but-2-enoic acid methyl ester (202). p-Methoxybenzylamine (1.43 mL, 11.0 mmol)

was slowly added to a solution of methyl 4-bromocrotonate (1.40 mL, 10.0 mmol)

and EtN(i-Pr)2 (6.64 mL, 33.0 mmol) in THF (50 mL) at 25 °C. The reaction mixture

was stirred for 12 h at 25 °C and saturated aqueous NaHCO3 (50 mL) was added. The

aqueous layer was extracted with EtOAc (3 × 50 mL) and the combined organic

extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The

residue was dissolved in CH2Cl2 (50 mL) at 25 °C and Et3N (3.34 mL, 24.0 mmol)

and (t-BuOCO)2O (2.76 mL, 12.0 mmol) were added. The reaction mixture was

stirred for 3 h at 25 °C and saturated aqueous NaHCO3 (50 mL) was added. The

aqueous layer was extracted with EtOAc (3 × 50 mL) and the combined organic

extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The

residue was purified by flash chromatography (5 × 10 cm silica, 10% EtOAc/hexane)

1 to afford pure 202 (2.35 g, 71%) as a colorless oil: H NMR (CDCl3, 500 MHz, 60°C)

δ 7.12 (d, 2H, J = 8.5 Hz), 6.84 (d, 2H, J = 8.5 Hz), 6.79 (m, 1H), 5.82 (d, 1H, J =

15.9 Hz), 4.35 (br, 2H), 3.88 (br, 2H), 3.77 (s, 3H), 3.72 (s, 3H), 1.47 (s, 9H); HRMS

(ESI) m/z 358.1617 (calcd. for C18H25NO5 + Na: 358.1631).

187

O OH NHCO t-Bu MeO 2 NHCO2Bn 2-Benzyloxycarbonylamino-4-tert-butoxycarbonyl- amino-3-hydroxybutyric acid methyl ester (204). A slurry of benzyl carbamate

(234 mg, 1.55 mmol) was added in one portion to a solution of NaOH (60.0 mg, 1.50

mmol) in H2O (3.8 mL) and n-PrOH (2.3 mL) at 25 °C. After the solution was

completely clear, t-BuOCl (0.17 mL, 1.50 mmol) was added to the reaction mixture

and the solution was stirred in the dark for 5 min at 25 °C. The reaction mixture was

cooled to 0 °C in the dark and a solution of (DHQD)2AQN (21 mg, 0.025 mmol) in n-

PrOH (2.0 mL) was added followed by the addition of 202 (167 mg, 0.50 mmol) and

K2OsO4⋅2H2O (7.4 mg, 0.020 mmol). The reaction mixture was stirred for 6 h at 0 °C

in the dark and saturated aqueous Na2S2O3 (5 mL) was added. The aqueous layer was extracted with EtOAc (3 × 5 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (1 × 10 cm silica, 5% acetone/CH2Cl2) to afford 204 (168.2

mg, 67%) as a colorless oil, with slight contamination of benzyl carbamate that was

used without further purification. A solution of (NH4)3Ce(NO3)6 (605 mg, 1.1 mmol)

in H2O (2 mL) was added to a solution of 204 (168.2 mg, 0.34 mmol) in MeCN (6

mL) at 25 °C. The reaction mixture was stirred for 5 min at 25 °C and EtOAc (5 mL)

was added. The aqueous layer was extracted with EtOAc (3 × 5 mL) and the

combined organic extracts were dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm

188

silica, 20% EtOAc/hexane) to afford pure 204 (112.0 mg, 56% for 2 steps) as a white

1 solid: H NMR (CDCl3, 500 MHz) δ 7.32 (m, 5H), 5.80 (br, 1H), 5.28 (br, 1H,

BnO2CN-H), 5.10 (br s, 2H, C6H5C-H), 4.44 (br, 1H, BnO2CHNC-H), 4.20 (br, 1H,

HOC-H), 3.76 (br, 1H), 3.73 (s, 3H), 3.35 (br, 1H, t-BuO2CNC-H), 3.00 (br, 1H, t-

13 BuO2CNC-H), 1.40 (s, 9H); C NMR (CDCl3, 125 MHz) δ 170.79, 159.28, 156.08,

136.19, 128.41, 128.06, 127.87, 80.04, 71.63, 67.23, 56.51, 44.00, 28.25; HRMS

(ESI) m/z 405.1609 (calcd. for C18H26N2O7 + Na: 405.1638).

BnO2CHN OH

HO N CO2t-Bu 3-Benzyloxycarbonylamino-2,4-dihydroxypyrrolidine-1-

carboxylic Acid tert-Butyl Ester (205). A solution of i-Bu2AlH (1.0 M in THF, 1.21

mL) was slowly added to a solution of 204 (84.8 mg, 0.22 mmol) in THF (2 mL) at –

78 °C. The reaction mixture was stirred for 14 h at –78 °C and aqueous

K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 5 mL) was added. The aqueous layer was

extracted with EtOAc (3 × 5 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (1 × 10 cm silica, 25% EtOAc/hexane) to afford pure 205

1 (61.5 mg, 67%) as an unstable pale yellow oil: H NMR (CDCl3, 500 MHz, 60 °C) δ

7.33 (m, 5H), 5.60-5.20 (br, 2H), 5.12 (br, 2H), 4.40-4.15 (br, 1H), 3.97 (br, 1H),

13 3.83-3.43 (br, 2H), 3.15 (m, 1H), 1.47 (s, 9H); C NMR (CDCl3, 125 MHz) δ

189

160.77, 154.75, 135.87, 128.58, 128.33, 128.17, 81.05, 79.06, 74.81, 67.40, 59.48,

50.25, 28.27; HRMS (ESI) m/z 375.1558 (calcd. for C17H24N2O6+ Na: 375.1532).

MeOCH O 2 NHCO2Bn MeO OH N CO t-Bu OMe 2 3-Benzyloxycarbonylamino-2-[1-(3,5-dimethoxy-2- methoxymethoxy-4-methylphenyl)allyl]-4-hydroxy-pyrrolidine-1-carboxylic

Acid tert-Butyl Ester (206). Boron trifluoride etherate (0.42 mL, 3.00 mmol) was slowly added to a solution of 205 (898.6 mg, 2.55 mmol) and 168 (1.79 g, 3.32 mmol) in CH2Cl2 (12 mL) at –78 °C. The reaction mixture was stirred for 30 min at –

78 °C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 15 mL) was added. The aqueous layer was extracted with EtOAc (3 × 15 mL) and the combined organic extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 25% EtOAc/hexane)

1 to afford pure 206 (838.0 mg, 56%) as a pale yellow oil: H NMR (CDCl3, 500 MHz)

δ 7.30 (m, 5H), 6.54-6.50 (br, 1H), 6.17 (m, 1H), 5.24 (m, 2H), 5.29-4.96 (br, 6 H),

13 4.69-3.16 (br, 15 H), 2.10-2.03 (br, 3H), 1.44 (br, 9H); C NMR (CDCl3, 125 MHz)

δ 153.31, 151.80, 151.56, 141.69, 135.92, 135.84, 128.53, 128.49, 128.22, 127.99,

127.80, 119.78, 119.44, 105.72, 99.63, 99.57, 80.25, 79.83, 71.80, 67.29, 67.07,

66.90, 65.92, 60.35, 60.16, 57.62, 55.72, 52.72, 52.08, 51.70, 28.29, 8.85, 8.73

(mixture of diastereomers; some peaks belong to rotamers); HRMS (ESI) m/z

609.2766 (calcd. for C31H42N2O9 + Na: 609.2783).

190

MeOCH2O

MeO NCO2Bn N CO t-Bu OMe 2 2-[1-(3,5-Dimethoxy-2-methoxymethoxy-4-

methylphenyl)allyl]-3,6-diazabicyclo[3.1.0]hexane-3,6-dicarboxylic Acid 6-

Benzyl Ester 3-tert-Butyl Ester (208). Diisopropyl azodicarboxylate (0.039 mL,

0.20 mmol) was slowly added to a solution of 206 (118.2 mg, 0.20 mmol) and PPh3

(104.8 mg, 0.40 mmol) in THF (10 mL) at 25 °C. The reaction mixture was stirred for

1 h at 25 °C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 5 mL) was added.

The aqueous layer was extracted with EtOAc (3 × 5 mL) and the combined organic

extracts were dried (Na2SO4), filtered, and the filtrate was evaporated in vacuo. The

residue was purified by flash chromatography (1 × 10 cm silica, 10% EtOAc/hexane)

1 to afford pure 208 (89.7 mg, 79%) as a pale yellow oil: H NMR (CDCl3, 500 MHz) δ

7.33 (m, 5H), 6.57-6.46 (br, 1H), 6.23 (br, 1H), 5.28-4.94 (br, 8 H), 4.42-3.31 (br, 13

13 H), 3.15-2.63 (br, 2H), 2.08-2.00 (br, 3H), 1.47 (br, 9H); C NMR (CDCl3, 125

MHz) δ 158.74, 152.33, 152.04, 141.85, 136.24, 136.01, 129.59, 129.17, 128.45,

128.04, 127.65, 119.41, 119.04, 106.36, 98.74, 98.61, 80.90, 80.65, 67.00, 66.90,

66.45, 66.37, 65.89, 61.39, 61.28, 58.32, 54.60, 51.99, 51.87, 51.71, 45.03, 44. 58,

39.44, 39.35, 38.76, 28.75, 8.67, 8.54 (mixture of diastereomers; some peaks belong to rotamers); HRMS (ESI) m/z 591.2697 (calcd. for C31H40N2O8 + Na: 591.2683).

191

OH MeOCH2O

MeO NCO2Bn N CO t-Bu OMe 2 2-[1-(3,5-Dimethoxy-2-methoxymethoxy-4-

methylphenyl)-2-hydroxyethyl]-3,6-diazabicyclo[3.1.0]hexane-3,6-dicarboxylic

Acid 6-Benzyl Ester 3-tert-Butyl Ester (209). A stream of O3 in O2 was slowly

bubbled into a solution of 208 (90 mg, 0.158 mmol) in MeOH (14 mL) and CH2Cl2 (4

mL) at –78 °C. The progress of the reaction was carefully monitored by TLC and N2 gas was bubbled for 30 min at –78 °C immediately after the reaction was complete.

The reaction mixture was warmed to 0 °C and NaBH4 (18.2 mg, 0.48 mmol) was

added. The reaction mixture was stirred for 3 h at 0 °C and the solvents were

evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm

silica, 15% EtOAc/hexane) to afford pure 209 (49.7 mg, 55%) as a colorless oil: 1H

NMR (CDCl3, 500 MHz, 60 °C) δ 7.34 (m, 5H), 6.56-6.42 (br, 1H), 5.27-4.96 (br,

4H), 4.74-4.52 (br, 1H), 4.23-3.19 (br, 15 H), 2.84-2.65 (br, 1H), 2.60-2.38 (br, 1H),

2.12-2.07 (br, 3H), 1.42-1.30 (br, 9H); HRMS (ESI) m/z 595.2610 (calcd. for

C30H40N2O9 + Na: 595.2632).

OSit-BuPh2 MeOCH2O

MeO NCO2Bn N CO t-Bu OMe 2 2-[2-(tert-Butyldiphenylsilanyloxy)-1-(3,5-

dimethoxy-2-methoxymethoxy-4-methylphenyl)ethyl]-3,6-diazabicyclo[3.1.0]-

192

hexane-3,6-dicarboxylic Acid 6-Benzyl Ester 3-tert-Butyl Ester (210). tert-

Butyldiphenylsilyl chloride (0.052 mL, 0.2 mmol) was slowly added to a solution of

209 (33.2 mg, 0.058 mmol), Et3N (0.069 mL, 0.50 mmol), and 4-

(dimethylamino)pyridine (5 mg) in THF (2 mL) at 25 °C. The reaction mixture was

stirred for 12 h at 25 °C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 2 mL)

was added. The aqueous layer was extracted with EtOAc (3 × 2 mL) and the

combined organic extracts were dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm

silica, 5% EtOAc/hexane) to afford pure 210 (45.9 mg, 96%) as a colorless oil: 1H

NMR (CDCl3, 500 MHz, 60 °C) δ 7.62-7.26 (m, 15H), 6.54-6.35 (br, 1H), 5.12-4.86

(br, 4H), 4.80-4.39 (br, 1H), 4.23-3.27 (br, 15 H), 2.96-2.84 (br, 1H), 2.73-2.42 (br,

13 1H), 2.18-2.07 (br, 3H), 1.42-1.20 (br, 9H), 1.06-0.92 (br, 9H); C NMR (CDCl3,

125 MHz) δ 158.13, 157.49, 152.59, 150.76, 147.82, 145.11, 135.58, 132.19, 128.84,

128.77, 128.69, 128.56, 128.44, 128.34, 128.28, 127.98, 127.53, 127.43, 127.49,

127.30, 123.47, 120.48, 120.37, 110.32, 99.70, 85.82, 85.72, 69.08, 68.90, 67.54,

67.47, 64.99, 62.05, 62.00, 59.79, 56.72, 51.60, 51.45, 51.30, 47.89, 46. 83, 40.82,

39.66, 39.06, 28.99, 28.25, 28.11, 27.32, 26.85, 26.77, 19.11, 8.44, 8.32 (mixture of diastereomers; some peaks belong to rotamers); HRMS (ESI) m/z 833.3859 (calcd. for C46H58N2O9Si + Na: 833.3804).

193

OSit-BuPh O 2

MeO NCO2Bn N CO t-Bu O 2 2-[2-(tert-Butyl-diphenylsilanyloxy)-1-(5- methoxy-4-methyl-3,6-dioxocyclohexa-1,4-dienyl)ethyl]-3,6-diazabicyclo[3.1.0]- hexane-3,6-dicarboxylic Acid 6-Benzyl Ester 3-tert-Butyl Ester (211). A solution

of (NH4)3Ce(NO3)6 in H2O (0.1 mL) was added to a solution of 209 (8.1 mg, 0.010

mmol) in MeCN (0.3 mL) at 25 °C. The reaction mixture was stirred for 5 min at 25

°C and saturated aqueous NaCl (0.5 mL) was added. The aqueous layer was extracted with EtOAc (3 × 0.5 mL) and the combined organic extracts were washed with

saturated aqueous NaHCO3 (0.5 mL), dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (0.5 × 10 cm

silica, 15% EtOAc/hexane) to afford pure 211 (6.7 mg, 87%) as a bright yellow oil:

1 H NMR (CDCl3, 500 MHz, 60 °C) δ 7.69-7.53 (m, 4H), 7.42-7.26 (m, 11H), 6.67-

6.28 (br, 1H), 5.16-4.89 (br, 2H), 4.56-4.38 (br, 1H), 4.17-3.84 (br, 6H), 3.45-3.24

(br, 1H), 3.12-2.72 (br, 3H), 1.96-1.77 (br, 3H), 1.38-1.17 (br, 9H), 1.08-0.96 (br,

9H); HRMS (ESI) m/z 773.3171 (calcd. for C43H50N2O8Si + Na: 773.3221).

194

CHAPTER 7

STUDIES TOWARD OXIDATIVE INSTALLATION OF

METHOXY MOIETY

7.1 Introduction

One of the most challenging problems in synthetic approaches to the mitomycins is the installation and preservation of the C9a methoxy/hydroxy group. A

C9a oxidative transformation of C-H bond to C-O bond is an interesting but difficult approach to this problem.

In our initial design we planned to perform an oxidative C-O bond formation at the C9a center after the isomitomycin ring system was built. However, our work in that area suggested that was not a viable route. Our mitosane synthesis and a few previous literature reports1 proved that the C9a methoxy/hydroxy group is important

to the stability of the molecule. Without the C9a methoxy/hydroxy group (i.e., where

1. Ziegler, F. E.; Berlin, M. Y. “A Synthesis of (+)-9a-Desmethoxymitomycin A via Aziridinyl Radical Cyclization,” Tetrahedron Lett. 1998, 39, 2455-2458; Danishefsky, S.; Berman, E. M.; Ciufolini, M.; Etheredge, S. J.; Segmuller, B. E. “A Stereospecific Route to Aziridinomitosanes: the Synthesis of Novel Mitomycin Congeners,” J. Am. Chem. Soc. 1985, 107, 3891-3898. 195

there is C9a−H bond), the molecule is less stable than its mitomycin form. Thus, we

realized that an early stage oxidative installation of C-O bond might be necessary if

we were to approach this system based on our established strategy.

The early stage installation of C-O moiety at C9a center would bring another

problem to the synthesis: the preservation of this sensitive functional group through

the remaining series of transformations. We expected that the elimination process

would be thermodynamically favored particularly in the presence of acid. (Figure 7.1)

R1 H R1 R1 OMe

N N N R2 R2 R2 215 216 217

Figure 7.1 Nitrogen Assisted Methanol Elimination

A carbamate protecting group on nitrogen would be beneficial for stabilizing molecules such as 215, however the methoxy group would still be unstable in the presence of acid. Actually, this is the basis for our Lewis acid catalyzed coupling strategy.

However, we suspected that for the 218 type of molecules, even if in some cases the nitrogen atom would assist the cleavage of the C9a-O bond, it would be

196

likely that the nearby hydroxy group would be kinetically favored to add back to the iminium ion double bond to bring system back to 218 stage, thereby providing the molecule with reasonably good stability toward acids. (Figure 7.2)

R1 H R1 O HO ( )n ( )n N N

218 219

Figure 7.2 Acid Stable 2-Alkoxypyrrolidine

Based on this premise, we proposed an intramolecular oxidative cyclization strategy to approach the C9a oxidation problem. We planned to use a mild oxidant to oxidize the C9a carbon followed by an in situ intramolecular trapping by the primary alcohol on the resulting iminium ion to form the heterocylic ring system. (Figure 7.3)

R1 OH R1 R1 H HO O ( )n [O] ( )n ( )n N N N

220 219 218

Figure 7.3 Intramolecular Oxidative Cyclization

197

We understood that in the real system we would have an electron rich

aromatic system, and would thereby limit the range of useful oxidant. Moreover, we

were concerned about the regioselectivity of the oxidation. We assumed oxidation on the tertiary carbon would be favored if this reaction were a thermodynamic controlled process.

We believed that the choice of tether length also played a key role.

Supposedly five membered ring tether would cyclize faster2 and the resulting hetero-

cycle would be more stable. However we felt it would be difficult to remove the 2- hydroxyethyl group on nitrogen later. The six membered ring tether should cyclize

equally well and the literature precedent had proved that it could be removed under very mild conditions.3

7.2 Intramolecular Oxidative Cyclization

The synthesis of our model system for the intramolecular oxidative cyclization

was based on our established synthetic protocol and utilized some intermediates we synthesized in our previous investigations. Because we expected the use of some

2. For five membered ring cyclization, this is a 5-endo-trig cyclization which is disfavored according to Baldwin’s rules.

3. Fukuyama, T.; Nakatsubo F.; Cocuzza A. J.; Kishi, Y. “ Synthetic Studies toward Mitomycins. III. Total Syntheses of Mitomycin A and C,” Tetrahedron Lett. 1977, 49, 4295-4298; Fukuyama, T.; Yang, L. “Total Synthesis of (±)-Mitomycins via Isomitomycin A,” J. Am. Chem. Soc. 1987, 109, 7881-7882. 198

acidic oxidants, the methoxymethyl phenyl ether was transformed to methyl phenyl ether. (Figure 7.4)

199

Sn(n-Bu)3

MeOCH2O MeO BF3⋅Et2O + HO N CH2Cl2, −78 °C CO2Bn OMe 94% 168 116

MeOCH2O OH MeO MeO BF3⋅Et2O

N CH2Cl2, 0 °C N CO2Bn 86% CO2Bn OMe OMe 221 222

OMe MeO Me2SO4, K2CO3 1 atm H2, Pd/C N acetone, 58 °C MeOH, 25 °C CO Bn OMe 2 55% 2 steps 223

OMe OMe MeO MeO BrCH CH CH OH, NaI 2 2 2 N HN K CO , n-BuOH, 100 °C 2 3 OMe OMe 86% HO 224 225

Figure 7.4 Synthesis of the Model Study Precursor

200

4 In Corey’s synthesis of aspidophytine, K3Fe(CN)6 was used to affect an

similar oxidative cyclization with high efficiency and beautiful simplicity. The

electron-rich aromatic system in that case was not affected. Because of this, we

examined the intramolecular cyclization using K3Fe(CN)6 as an oxidant.

Unfortunately, even under forcing conditions (K3Fe(CN)6, NaHCO3, n-BuOH, H2O,

60 °C) there was no indication of any oxidation. Starting material was recovered in all cases. We investigated other oxidants such as Pb(OAc)4 and Mn(OAc)3, and only

Pb(OAc)4 promoted the intramolecular oxidative cyclization, albeit in low yield and

wrong regioselectivity. Most of the starting materials were recovered. Changing

reaction parameters such as temperature (25 °C, 60 °C), solvent (MeCN, acetone) and

base (with/without NaHCO3) were ineffective at increasing the reaction yield or

altering regioselectivity. (Figure 7.5)

OMe MeO OMe Pb(OAc) MeO N 4 MeCN, 25 °C NO OMe 15% + 80% s. m. OMe HO 225 226

Figure 7.5 Intramolecular Oxidative Cyclization on the Model System

4. He, F.; Bo, Y.; Altom, J. D.; Corey, E. J. “Enantioselective Total Synthesis of Aspidophytine,” J. Am. Chem. Soc. 1999, 121, 6771-6772. 201

Clearly the kinetic hydrogen atom abstraction played a predominant role in this disappointing regioselectivity. Other forms of tethers on pyrrolidine nitrogen were synthesized, but proved unreactive towards any of the three oxidants. (Figure

7.6)

OMe OMe OMe MeO MeO MeO

N N N

OMe OMe OMe COOH OH COOH

227 228 229

Figure 7.6 Other Model Systems Examined

202

7.3 Conclusion

We have developed a novel and interesting strategy to introduce the C9a-O moiety in the mitomycins. Unfortunately the Pb(OAc)4 promoted intramolecular

oxidative cyclization occured with the wrong regioselectivity. However, the stability

of the resulting 2-alkoxypyrrolidine 226 in the relatively acid reaction condition suggested that this strategy might still be useful in preserving the C9a-O moiety in the mitomycin system. A bulky substituent adjacent to the competing carbon might be helpful in control the regioselectivity.

203

7.4 Experimental

MeOCH2O MeO

N CO Bn OMe 2 2-[1-(3,5-Dimethoxy-2-methoxymethoxy-4-

methylphenyl)allyl]pyrrolidine-1-carboxylic acid benzyl ester (221). Boron

trifluoride etherate (0.22 mL, 1.35 mmol) was slowly added to a solution of 116

(294.4 mg, 1.33 mmol) in CH2Cl2 (7 mL) at –78 °C. The reaction mixture was stirred

for 1 h at –78 °C and 168 (719.3 mg, 1.33 mmol) was added. The reaction mixture

was stirred for 30 min at –78 °C and aqueous K2HPO4/KH2PO4 buffer (1.0 M, pH 7,

8 mL) was added. The aqueous layer was extracted with EtOAc (3 × 5 mL) and the

combined organic extracts were dried (Na2SO4), filtered, and the filtrate was

evaporated in vacuo. The residue was purified by flash chromatography (5 × 10 cm

silica, 10% EtOAc/hexane) to afford pure 221 (569.3 mg, 94%) as a pale yellow oil:

1 H NMR (CDCl3, 500 MHz) δ 7.32 (m, 5H), 6.49 (br, 1H), 6.12 (br, 1H), 5.17 (d, 1H,

J = 12.8 Hz), 5.14 (d, 1H, J = 15.3 Hz), 5.01-4.90 (br, 4H), 4.29 (br, 1H), 3.95 (br,

1H), 3.82-3.60 (br, 6H), 3.59-3.43 (br, 4H), 3.35 (br, 1H), 2.09 (s, 3H), 1.87 (m, 2H),

1.67 (m, 2H); HRMS (ESI) m/z 478.2234 (calcd. for C26H33NO6 + Na: 478.2206).

204

OH MeO

N CO Bn OMe 2 2-[1-(2-Hydroxy-3,5-dimethoxy-4-methylphenyl)-

allyl]pyrrolidine-1-carboxylic acid benzyl ester (222). Boron trifluoride etherate

(0.12 mL, 1.0 mmol) was slowly added to a solution of 116 (455 mg, 1.0 mmol) in

CH2Cl2 (5 mL) at 0 °C. The reaction mixture was stirred for 2 h at 0 °C and aqueous

K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 5 mL) was added. The aqueous layer was

extracted with EtOAc (3 × 5 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (2 × 10 cm silica, 10% EtOAc/hexane) to afford pure 222

1 (352.8 mg, 86%) as a pale yellow oil: H NMR (CDCl3, 500 MHz) δ 6.43 (br, 1H),

6.22 (m, 1H), 5.20-4.92 (br, 4H), 4.24 (br, 1H), 4.15 (br, 1H), 3.79-3.66 (br, 6H), 3.52

(br, 1H), 3.36 (m, 1H), 2.12 (s, 3H), 1.93-1.71 (m, 4H); HRMS (ESI) m/z 434.1907

(calcd. for C24H29NO5 + Na: 434.1944).

OMe MeO

OMe 2-[1-(2,3,5-Trimethoxy-4-methylphenyl)propyl]pyrrolidine

(224). A slurry of K2CO3 (2.07 g, 15.0 mmol) was added in one portion to a solution

of 116 (226 mg, 0.55 mmol) in acetone (10 mL) at 25 °C followed by the addition of

Me2SO4 (0.28 mL, 3.0 mmol). The mixture was warmed to reflux for 16 h and was

205

allowed to cool to 25 °C. The mixture was filtered and the filtrate was evaporated in

vacuo. The residue was dissolved in methanol (10 mL) and 10% palladium over activated carbon (10 mg) was added. The reaction mixture was stirred for 3 h at 25 °C under 1 atm H2. The reaction mixture was filtered and the filtrate was evaporated in

vacuo. The residue was purified by flash chromatography (1 × 10 cm silica, 5%

1 Et3N/EtOAc) to afford pure 224 (80.2 mg, 55% for 2 steps) as a colorless oil: H

NMR (CDCl3, 400 MHz) δ 6.36 (s, 1H), 3.73 (s, 3H), 3.70 (s, 6H), 3.01 (dd, 1H, J =

15.7, 8.6 Hz), 2.87 (m, 1H), 2.79 (br, 1H), 2.62 (dd, 1H, J = 16.7, 8.1 Hz), 2.03 (s,

3H), 1.89 (m, 1H), 1.67 (m, 3H), 1.50 (m, 1H), 1.33 (m, 1H), 0.68 (t, 3H, J = 7.6 Hz);

HRMS (ESI) m/z 316.1899 (calcd. for C17H27NO3 + Na: 316.1889).

OMe MeO

OMe

HO 3-{2-[1-(2,3,5-Trimethoxy-4-methylphenyl)propyl]-

pyrrolidin-1-yl}propan-1-ol (225). 3-Bromo-1-propanol (0.024 mL, 0.26 mmol) was added to a mixture of 224 (76.6 mg, 0.26 mmol), K2CO3 (138 mg, 1.0 mmol) and NaI

(5 mg) in n-BuOH (2 mL) at 25 °C. The reaction mixture was stirred for 12 h at 100

°C and was allowed to cool to 25 °C. The mixture was filtered and the filtrate was evaporated in vacuo. The residue was purified by flash chromatography (1 × 10 cm

1 silica, 5% Et3N/EtOAc) to afford pure 225 (78.3 mg, 86%) as a colorless oil: H

NMR (CDCl3, 500 MHz) δ 6.42 (s, 1H), 3.78 (s, 3H), 3.77 (s, 3H), 3.75 (s, 3H), 3.73

206

(m, 2H), 3.70 (m, 1H), 3.30 (m, 1H), 3.20 (m, 1H), 3.07 (dt, 1H, J = 12.2, 3.7 Hz),

2.44 (m, 2H), 2.08 (s, 3H), 2.07 (m, 1H), 1.89 (m, 1H), 1.75 (m, 2H), 1.61 (m, 3H),

13 1.47 (m, 1H), 0.78 (t, 3H, J = 7.3 Hz); C NMR (CDCl3, 125 MHz) δ 153.92,

151.81, 145.86, 133.45, 118.48, 104.67, 69.97, 64.78, 60.69, 60.03, 55.92, 55.82,

54.16, 41.99, 29.36, 26.15, 22.83, 20.46, 12.64, 8.69; HRMS (ESI) m/z 374.2319

(calcd. for C20H33NO4 + Na: 374.2308).

OMe MeO NO

OMe 6-[1-(2,3,5-Trimethoxy-4-methylphenyl)propyl]-

hexahydropyrrolo[2,1-b][1,3]oxazine 226. A slurry of Pb(OAc)4 (11.1 mg, 0.025

mmol) was added in one portion to a solution of 225 (7.8 mg, 0.022 mmol) in MeCN

(0.1 mL) at 25 °C. The reaction mixture was stirred for 12 h at 25 °C and aqueous

K2HPO4/KH2PO4 buffer (1.0 M, pH 7, 0.5 mL) was added. The aqueous layer was

extracted with EtOAc (3 × 0.5 mL) and the combined organic extracts were dried

(Na2SO4), filtered, and the filtrate was evaporated in vacuo. The residue was purified

by flash chromatography (0.5 × 10 cm silica, 10% EtOAc/hexane) to afford pure 222

1 (1.2 mg, 15%) as a colorless oil: H NMR (CDCl3, 500 MHz) δ 6.47 (s, 1H), 4.63 (d,

1H, J = 4.3 Hz), 3.96 (m, 1H), 3.79 (s, 3H), 3.78 (s, 3H), 3.76 (s, 3H), 3.64 (dt, 1H),

3.34 (m, 1H), 3.05 (m, 1H), 3.01 (m, 1H), 2.93 (m, 1H), 2.10 (s, 3H), 1.90 (m, 2H),

1.76 (m, 3H), 1.56 (m, 3H), 0.76 (t, 3H, J = 7.9 Hz); HRMS (ESI) m/z 372.2142

(calcd. for C20H31NO4 + Na: 372.2151).

207

REFERENCES

2. Carter, S. K.; Crooke, S. T. “Mitomycin C: Current Status and New Developments,” Academic Press, 1979.

3. Spain, R. C. “The Case for Mitomycin in Non-Small Cell Lung Cancer,” Oncology, 1993, 50 (Suppl. 1), 35-52.

6. Caron, R. M.; Hamilton, J. W. “Preferential Effects of the Chemotherapeutic DNA Cross-Linking Agent Mitomycin C on Inducible Gene Expression in vitro Environ,” Mol. Mutagen. 1995, 25, 4-11.

208

9. Hata, T.; Sano, Y.; Sugawara, R.; Matsumae, A.; Kanamorei, K.; Shima, T.; Hoshi, T. “Mitomycin, A New Antibiotic from Streptomyces,” J. Antibiot. Ser. A 1956, 9, 141-146.

10. DeBoer, C.; Dietz, A.; Lummus, N. E.; Savage, G. M. “Porfiromycin, a New Antibiotic. I. Discovery and Biological Activities,” Antimicrob. Agents Ann. 1961, Vol. Date 1960, 17-22.

11. Urakawa, C.; Tsuchiya, H.; Nakano, K.-I. “New Mitomycin, 10- Decarbamoyloxy-9-dehydromitomycin B from Streptomyces Caespitosus,” J. Antibiot. 1981, 34, 243-244.

14. Shirahata, K.; Hirayama, N. “Revised Absolute Configuration of Mitomycin C. X-Ray Analysis of 1-N-(p-bromobenzoyl)mitomycin C,” J. Am. Chem. Soc. 1983, 105, 7199-7200.

15. Fukuyama, T.; Yang, L. “Practical Total Synthesis of (±)-Mitomycin C,” J. Am. Chem. Soc. 1989, 111, 8303-8304.

16. Remers, W. A.; Dorr, R. T. “Alkaloids: Chemical and Biological Perspectives,” Relletier, S. W. Ed. Wiley, New York, 1988, 1-67.

17. Remers, W. A. “The Chemistry of Antitumor Antibiotics,” Vol. 1, Wiley, New York, 1979.

18. Remers, W. A.; Dorr, R. T. “Alkaloids: Chemical and Biological Perspectives,” Relletier, S. W. Ed. Wiley, New York, 1988, 1-67.

19. Remers, W. A. “The Chemistry of Antitumor Antibiotics,” Vol. 1, Wiley, New York, 1979.

209

20. Taylor, W. G.; Remers, W. A. “Structure and Stereochemistry of Some 1,2- Disubstituted Mitosenes from Solvolysis of Mitomycin C and Mitomycin A,” J. Med. Chem. 1975, 18, 307-311.

21. Cheng, L.; Remers, W. A. “Comparative Stereochemistry in the Aziridine Ring Openings of N-methylmitomycin A and 7-Methoxy-1,2-(N-methylaziridino)- mitosene,” J. Med. Chem. 1977, 20, 138-141.

22. Danishefsky, S.; Ciufolini, M. “Leucomitomycins,” J. Am. Chem. Soc. 1984, 106, 6424-6425.

23. Tomasz, M.; Palom, Y. “The Mitomycin Bioreductive Antitumor Agents: Cross-Linking and Alkylation of DNA as the Molecular Basis of Their Activity,” Pharmacol. Ther. 1997, 76, 73-87; and references therein.

24. Tomasz, M. “Molecular Aspects of Anticancer Drug-DNA interactions,” Neidel, S. and Waring, M. eds., 1979, Vol. 2, 312-349; and references therein.

25. Iyer, V. N.; Szybalski, W. “A Molecular Mechanism of Mitomycin Action: Linking of Complementary DNA Strands,” Proc. Natl. Acad. Sci. 1963, 50, 355-362.

26. Weissback, A.; Lisio, A. “Alkylation of Nucleic Acids by Mitomycin C and Porfiromycin,” Biochemistry, 1965, 4, 196-200; Tomasz, M.; Mercado, C. M.; Olson, J.; Chatterjie, N. “The Mode of Interaction of Mitomycin C with DNA and Polynucleotides in vivo,” Biochemistry 1974, 13, 4878-4887; Lown, J. W.; Bergleiter, A.; Jobson, D.; Morgan, A. R. “Studies Related to Antitumor Antibiotics: V. Reaction of Mitomycin C with DNA Examined by Ethidium Fluorescence Assay,” Can. J. Biochem. 1976, 54, 110-119.

27. Coleman, R. S.; Perez, R. J.; Burk, C. H.; Navarro, A. “Studies on the Mechanism of Action of Azinomycin B: Definition of Regioselectivity and Sequence Selectivity of DNA Cross-Link Formation and Clarification of the Role of the Naphthoate,” J. Am. Chem. Soc. 2002, 124, 13008-13017; Alcaro, S.; Ortuso, F.; Coleman, R. S. “DNA Cross-Linking by Azinomycin B: Monte Carlo Simulations in the Evaluation of Sequence Selectivity,” J. Med. Chem. 2002, 45, 861-870; Hartley, J. A.; Hazrati, A.; Kelland, L. R.; Khanim, R.; Shipman, M.; Suzenet, F.; Walker, L. F. “A Synthetic Azinomycin Analog with Demonstrated DNA Cross-Linking Activity: Insights into the Mechanism of Action of this Class of Antitumor Agent,” Angew. Chem. Int. Ed. 2000, 39, 3467-3470; Zang, H.; Gates, K. S. “DNA Binding and Alkylation by the "Left Half" of Azinomycin B,” Biochemistry 2000, 39, 14968-14975; Armstrong, R. W.; Salvati, M. E.; Nguyen, M. “Novel Interstrand Cross-Links Induced by the

210

Antitumor Antibiotic Carzinophillin/Azinomy-cin B,” J. Am. Chem. Soc. 1992, 114, 3144-3145.

28. Williams, R. M.; Herberich, B. “DNA Interstrand Cross-Link Formation Induced by Bioxalomycin α2,” J. Am. Chem. Soc. 1998, 120, 10272-10273.

29. Szybalski, W.; Iyer, V. N. “Cross-Linking of DNA by Enzymatically or Chemically Activated Mitomycins and Porfiromycins, Bifunctionally ‘Alkylating’ Antibiotics,” Fed. Proc. 1964, 23, 946-957.

30. Iyer, V. N.; Szybalski, W. “A Molecular Mechanism of Mitomycin Action: Linking of Complementary DNA Strands,” Proc. Natl. Acad. Sci. 1963, 50, 355-362; Kumar, G. S.; He, Q. –Y.; Behr-Ventura, D.; Tomasz, M. “Binding of 2,7-Diaminomitosene to DNA: Model for the Recognition of DNA by Activated Mitomycin C,” Biochemistry 1995, 34, 2662-2671; Rodighiero, G.; Marciani Magno, S.; Dell Acqua, F.; Vedaldi, D.; “Studies on the Mechanism of Action of Mitomycin C,” Farmaco. Ed. Sci. 1978, 3, 651-666.

31. Tomasz M.; Lipman R.; Chowdary D.; Pawlak J.; Verdine G. L.; Nakanishi K. “Isolation and Structure of A Covalent Cross-Link Adduct between Mitomycin C and DNA,” Science 1987, 235, 1204-1208.

32. For a comprehensive review of in situ activation for DNA targeting agents, see: Wolkenberg, S. E.; Boger, D. L. “Mechanisms of in situ Activation for DNA- Targeting Antitumor Agents,” Chem. Rev. 2002, 102, 2475-2495.

33. Peterson, D. M.; Fisher, J. “Autocatalytic Quinone Methide Formation from Mitomycin C,” Biochemistry, 1986, 25, 4077-4084; Iyer, V. N.; Szybalski, W. “Mitomycins and Porfiromycin: Chemical Mechanism of Activation and Cross- Linking of DNA,” Science 1964, 145, 55-58.

34. Tomasz, M.; Lipman, R. “Reductive Metabolism and Alkylating Activity of Mitomycin C Induced by Rat Liver Microsomes,” Biochemistry 1981, 20, 5056- 5061.

35. Andrews, P. A.; Pan, S.-S.; Bachur, N. R. “Electrochemical Reductive Activation of Mitomycin C,” J. Am. Chem. Soc. 1986, 108, 4158-4166; Kohn, H.; Zein, N.; Lin, X. Q.; Ding, J. Q.; Kadish, K. M. “Mechanistic Studies on the Mode of Reaction of Mitomycin C under Catalytic and Electrochemical Reductive Conditions,” J. Am. Chem. Soc. 1987, 109, 1833-1840.

36. Hoey, B. M.; Butler, J.; Swallow, A. J. “Reductive Activation of Mitomycin C,” Biochemistry 1988, 27, 2608-2614.

211

37. Cera, C.; Egbertson, M.; Teng, S. P.; Crothers, D. M.; Danishefsky, S. J. “DNA Cross-Linking by Intermediates in the Mitomycin Activation Cascade,” Biochemistry 1989, 28, 5665-5669.

38. Iyer, V. N.; Szybalski, W. “Mitomycins and Porfiromycin: Chemical Mechanism of Activation and Cross-Linking of DNA,” Science 1964, 145, 55- 58.

39. Under anhydrous condition, the elimination process is not spontaneous: Danishefsky, S.; Ciufolini, M. “Leucomitomycins,” J. Am. Chem. Soc. 1984, 106, 6424-6425.

40. Support for this intermediate is indirect. See: Tomasz, M.; Lipman, R. “Reductive Metabolism and Alkylating Activity of Mitomycin C Induced by Rat Liver Microsomes,” Biochemistry 1981, 20, 5056-5061; Bean, M.; Kohn, H. “Studies on the Reaction of Mitomycin C with Potassium Thiobenzoate under Reductive Conditions,” J. Org. Chem. 1985, 50, 293-298; Gargiulo, D.; Musser, S. S.; Yang, L.; Fukuyama, T.; Tomasz, M. “Alkylation and Crosslinking of DNA by the Unnatural Enantiomer of Mitomycin C: Mechanism of the DNA- Sequence Specificity of Mitomycins,” J. Am. Chem. Soc. 1995, 117, 9388-9398; and references cited therein.

41. Tomasz, M.; Lipman, R.; Verdine, G. L.; Nakanishi, K. “Reassignment of the Guanine-Binding Mode of Reduced Mitomycin C,” Biochemistry 1986, 25, 4337-4344; Tomasz, M.; Lipman, R.; McGuinness, B. F.; Nakanishi, K. “Isolation and Characterization of A Major Adduct between Mitomycin C and DNA,” J. Am. Chem. Soc. 1988, 110, 5892-5896.

42. Tomasz, M.; Lipman, R.; Chowdary, D.; Pawlak, J.; Verdine, G.L.; Nakanishi, K. “Isolation and Structure of A Covalent Cross-Link Adduct between Mitomycin C and DNA,” Science 1987, 235, 1204-1208.

43. Tomasz, M.; Chawla, A. K.; Lipman, R. “Mechanism of Monofunctional and Bifunctional Alkylation of DNA by Mitomycin C,” Biochemistry 1988, 27, 3182-3187.

44. Bizanek, R.; McGuinness, B. F.; Nakanishi, K.; Tomasz, M. “Isolation and Structure of An Intrastrand Cross-Link Adduct of Mitomycin C and DNA,” Biochemistry 1992, 31, 3084-3091.

45. Kohn, H.; Li, V.–S.; Schiltz, P.; Tang, M. S. “On the Origins of the DNA Sequence Selectivity of Mitomycin Monoalkylation Transformations,” J. Am. Chem. Soc. 1992, 114, 9218-9220; Kumar, S.; Lipman, R.; Tomasz, M.

212

“Recognition of Specific DNA Sequences by Mitomycin C for Alkylation,” Biochemistry 1992, 31, 1399-1407.

46. Sastry M.; Fiala R.; Lipman R.; Tomasz M.; Patel D. J. “Solution Structure of the Monoalkylated Mitomycin C-DNA Complex,” J. Mol. Biol. 1995, 247, 338- 359; Li, V.-S.; Choi, D.; Wang, Z.; Jimenez, L. S.; Tang, M.-S.; Kohn, H. “Role of the C-10 Substituent in Mitomycin C-1-DNA Bonding,” J. Am. Chem. Soc. 1996, 118, 2326-2331.

47. Borowy-Borowski, H.; Lipman, R.; Tomasz, M. “Recognition between Mitomycin C and Specific DNA Sequences for Cross-Link Formation,” Biochemistry 1990, 29, 2999-3006; Millard, J. T.; Weidner, M. F.; Raucher, S.; Hopkins, P. B. “Determination of the DNA Crosslinking Sequence Specificity of Reductively Activated Mitomycin C at Single-Nucleotide Resolution: Deoxyguanosine Residues at CpG are Crosslinked Preferentially,” J. Am. Chem. Soc. 1990, 112, 3637-3641; Li, V.-S.; Kohn, H. “Studies on the Bonding Specificity for Mitomycin C-DNA Monoalkylation Processes,” J. Am. Chem. Soc. 1991, 113, 275-283.

48. Norman, D.; Live, D.; Sastry, M.; Lipman, R.; Hingerty, B. E.; Tomasz, M.; Broyde, S.; Patel, D. J. “NMR and Computational Characterization of Mitomycin Cross-Linked to Adjacent Deoxyguanosines in the Minor Groove of the d(T-A-C-G-T-A).d(T-A-C-G-T-A) Duplex,” Biochemistry 1990, 29, 2861- 2875.

49. Kumar, S.; Lipman, R.; Tomasz, M. “Recognition of Specific DNA Sequences by Mitomycin C for Alkylation,” Biochemistry 1992, 31, 1399-1407; Teng, S. P.; Woodson, S. A.; Crothers, D. M. “DNA Sequence Specificity of Mitomycin Cross-Linking,” Biochemistry 1989, 28, 3901-3907.

50. Fukuyama, T.; Goto, S. “Synthetic Approaches toward FR-900482. I. Stereoselective Synthesis of A Pentacyclic Model Compound,” Tetrahedron Lett. 1989, 30, 6491-6494.

51. Williams, R. M.; Rajski, S. R.; Rollins, S. B. “FR 900482, A Close Cousin of Mitomycin C that Exploits Mitosene-Based DNA Crosslinking,” Chem. Biol. 1997, 4, 127-137.

52. Doyle, T. W.; Vyas, D. M. “Second Generation Analogs of Etoposide and Mitomycin C,” Cancer Treat. Rev. 1990, 17, 127-131; Kono, M.; Saitoh, Y.; Kasai, M.; Sato, A.; Shirahata, K.; Morimoto, M.; Ashizawa, T. “Synthesis and Antitumor Activity of A Novel Water Soluble Mitomycin Analog; 7-N-[2-[[2- (g-L-Glutamylamino)ethyl]dithio]ethyl]mitomycin C,” Chem. Pharm. Bull. 1989, 37, 1128-1130.

213

53. Coleman, R. S. “Total Synthesis of Natural Product DNA Cross-Linking Agents,” Curr. Opinion Drug Disc. Dev. 2001, 4, 435-449; Danishefsky, S. J.; Schkeryantz, J. M. “Chemical Explorations Driven by An Enchantment with Mitomycinoids – A Twenty Year Account,” Synlett 1995, spec. issue, 475-490; Fukuyama, T.; Yang, L. “Total Synthesis of Mitomycins,” In: Studies in Natural Products Chemistry Atta-ur-Rahman Ed., Elsevier Science, 1993, 13, 433-471; Remers, W. A.; Dorr, R. T. “Alkaloids: Chemical and Biological Perspectives,” Relletier, S. W. Ed. Wiley, New York, 1988, 1-67; Remers, W. A. “The Chemistry of Antitumor Antibiotics,” Vol. 1, Wiley, New York, 1979.

54. Fukuyama, T.; Nakatsubo F.; Cocuzza A. J.; Kishi, Y. “ Synthetic Studies toward Mitomycins. III. Total Syntheses of Mitomycin A and C,” Tetrahedron Lett. 1977, 49, 4295-4298.

55. Fukuyama, T.; Yang, L. “Total Synthesis of (±)-Mitomycins via Isomitomycin A,” J. Am. Chem. Soc. 1987, 109, 7881-7882; Fukuyama, T.; Yang, L. “Practical Total Synthesis of (±)-Mitomycins C,” J. Am. Chem. Soc. 1989, 111, 8303-8304. This was the second time Fukuyama finished the synthesis of mitomycins: also see referece 5 in this chapter.

56. Kono, M.i; Saitoh, Y.; Shirahata, K.; Arai, Y.; Ishii, S. “Albomitomycin A and Isomitomycin A. Products of Novel Intramolecular Rearrangement of Mitomycin A,” J. Am. Chem. Soc. 1987, 109, 7224-7225.

57. Shea, K. J. “Recent Developments in the Synthesis, Structure, and Chemistry of Bridgehead Alkenes,” Tetrahedron 1980, 36, 1683-1715.

58. Benbow, J. W.; Schulte, G. K.; Danishefsky, S. J. “The Total Synthesis of (±)- Mitomycin K,” Angew. Chem. Int. Ed. 1992, 31, 915-917; Benbow, J. W.; McClure, K. F.; Danishefsky, S. J. “Intramolecular Cycloaddition Reactions of Dienyl Nitroso Compounds: Application to the synthesis of Mitomycin K,” J. Am. Chem. Soc. 1993, 115, 12305-12314.

59. Boger, D. L. “Heterodiene Additions,” Comp. Org. Syn. Trost, B. M.; Fleming, I. Eds. Pergamon Press, Oxford, 1991, 5, 451-550.

60. Hudrlik, P. F.; Peterson D. “Stereospecific Olefin-forming Elimination Reactions of β-Hydroxyalkylsilanes,” J. Am. Chem. Soc. 1975, 97, 1464 – 1468.

61. Schkeryantz, J. M.; Danishefsky, S. J. “Total Synthesis of (±)-FR-900482,” J. Am. Chem. Soc. 1995, 117, 4722-4723.

214

62. Wang, Z.; Jimenez, L. S. “A Total Synthesis of (±)-Mitomycin K. Oxidation of the Mitosene C9-9a Double Bond by (Hexamethylphosphoramido)oxodiperoxomobdenum (VI) (MoO5⋅HMPA),” Tetrahedron Lett. 1996, 37, 6049- 6052; Wang, Z.; Jimenez, L. S. “Synthesis of An Aziridinomitosene Analog,” J. Org. Chem. 1996, 61, 816-818.

63. Fukuyama, T.; Xu, L.; Goto, S. “Total Synthesis of (±)-FR-900482,” J. Am. Chem. Soc. 1992, 114, 383-385.

64. Suzuki, M.; Kambe, M.; Tokuyama, H.; Fukuyama, T. “Facile Construction of N-Hydroxybenzazocin: Enantioselective Total Synthesis of (+)-FR-900482,” Angew. Chem. Int. Ed. 2002, 41, 4686-4688.

65. Sonogashira, K. “Development of Pd-Cu Catalyzed Cross-Coupling of Terminal Acetylenes with sp2-Carbon Halides,” J. Organomet. Chem. 2002, 653, 46-49.

66. Yoshino, T.; Nagata, Y.; Itoh, E.; Hashimoto, M.; Katoh, T.; Terashima, S. “Total Synthesis of An Enantiomeric Pair of FR900482. 2. Syntheses of the Aromatic and the Optically Active Aliphatic Segments,” Tetrahedron 1997, 53, 10239-10252; Katoh, T.; Nagata, Y.; Yoshino, T.; Nakatani, S.; Terashima, S. “Total Synthesis of An Enantiomeric Pair of FR900482. 3. Completion of the Synthesis by Assembling the Two Segments,” Tetrahedron 1997, 53, 10253- 10270.

66. Fellows, I. M.; Kaelin, D. E.; Martin, S. F. “Application of Ring-Closing Metathesis to the Formal Total Synthesis of (+)-FR-900482,” J. Am. Chem. Soc. 2000, 122, 10781-10787.

67. Judd, T. C.; Williams, R. M. “Concise Enantioselective Synthesis of (+)- FR66979 and (+)-FR900482: Dimethyldioxirane-Mediated Construction of the Hydroxylamine Hemiketal,” Angew. Chem. Int. Ed. 2002, 41, 4683-4685.

68. Ducray, R.; Ciufolini, M. A. “Total Synthesis of (±)-FR66979,” Angew. Chem. Int. Ed. 2002, 41, 4688-4691.

69. Ciufolini, M. A.; Chen, M.; Lovett, D. P.; Deaton, M. V. “Application of Ene- Like Reactions of Aldehydes with Vinyl Ethers: Facile Assembly of Benzazocenone Intermediates for Mitomycinoids,” Tetrahedron Lett. 1997, 38, 4355-4358.

70. Hudrlik, P. F.; Hudrlik, A. M.; Kulkarni, A. K. “Protodesilylation Reactions of Simple β-Hydroxysilanes (and α-Hydroxysilanes). Homo-Brook rearrangements,” J. Am. Chem. Soc. 1982, 104, 6809-6811.

215

71. Lee, J.; Ha, J. D.; Cha, J. K. “New Synthetic Method for Functionalized Pyrrolizidine, Indolizidine, and Mitomycin Alkaloids,” J. Am. Chem. Soc. 1997, 119, 8127-8128.

72. Lee, J.; Kang, C. H.; Kim, H.; Cha, J. K. “Intramolecular Hydroxycyclopropanation of Vinyl Carboxylic Esters,” J. Am. Chem. Soc. 1996, 118, 291-292.

73. Kulinkovich, O. G.; Sviridov, S. V.; Vasilevskii, D. A. “Titanium(IV) Isopropoxide-Catalyzed Formation of 1-Substituted Cyclopropanols in the Reaction of Ethylmagnesium Bromide with Methyl Alkanecarboxylates,” Synthesis 1991, 234-235.

74. Lim, H.–J.; Sulihowski, G. A. “Enantioselective Synthesis of a 1,2- Disubstituted Mitosene by a Copper-Catalyzed Intramolecular Carbon- Hydrogen Insertion Reaction of a Diazo Ester,” J. Org. Chem. 1995, 60, 2326- 2327; Lee, S.; Lim, H.–J.; Cha, K. L.; Sulikowski, G. A. “Asymmetric Approaches to 1,2-Disubstituted Mitosenes Based on the Intramolecular Cyclization of Diazoesters,” Tetrahedron 1997, 53, 16521-16532.

75. Ziegler, F. E.; Belema, M. “Cyclization of Chiral Carbon-Centered Aziridinyl Radicals: A New Route to Aziridino[2′,3′:3,4]pyrrolo[1,2-a]indoles,” J. Org. Chem. 1994, 59, 7962-7967; Ziegler, F. E.; Berlin, M. Y. “A Synthesis of (+)- 9a-Desmethoxymitomycin A via Aziridinyl Radical Cyclization,” Tetrahedron Lett. 1998, 39, 2455-2458; Ziegler, F. E.; Berlin, M. Y.; Lee, K.; Looker, A. R. “Formation of 9,10-Unsaturation in the Mitomycins: Facile Fragmentation of β- Alkyl-β-aryl-α-oxo-γ-butyrolactones,” Org. Lett. 2000, 2, 3619-3621.

76. Dobbs, A. P.; Jones, K.; Veal, K. T. “Indole Radical Cyclisations: A Rapid Route to Mitosenes,” Tetrahedron Lett. 1995, 36, 4857-4860.

77. Allan, G. M.; Parsons, A. F.; Pons, J.–F. “Tandem Radical Cyclisation and Translocation Approaches to Biologically Important Mitomycin Ring Systems,” Synlett 2002, 1431-1434.

78. Vedejs, E.; Klapars, A.; Naidu, B. N.; Piotrowski, D. W.; Tucci, F. C. “ Enantiocontrolled Synthesis of (1S,2S)-6-Desmethyl-(methylaziridino)- mitosene,” J. Am. Chem. Soc. 2000, 122, 5401-5402.

79. Vedejs, E.; Little, J. “Aziridinomitosenes by Anionic Cyclization: Deuterium as a Removable Blocking Group,” J. Am. Chem. Soc. 2001, 124, 748-749,

216

80. Michael, J. P.; de Koning, C. B.; Petersen, R. L.; Stanbury, T. V. “Asymmetric Synthesis of A Tetracyclic Model for the Aziridinomitosenes,” Tetrahedron Lett. 2001, 42, 7513-7516.

81. Nakatsubo, F.; Fukuyama, T.; Cocuzza, A. J.; Kishi, Y. “Synthetic Studies toward Mitomycins. 2. Total Synthesis of dl-Porfiromycin,” J. Am. Chem. Soc. 1977, 99, 8115-8116.

82. Jones, G. B.; Guzel, M.; Mathews, J. E. “Stereoselective Route to Mitosanes via Tricarbonyl η6 Arene Chromium Complexes,” Tetrahedron Lett. 2000, 41, 1123-1126.

83. Papaioannou, N.; Evans, C. A.; Blank, J. T.; Miller, S. J. “Enantioselective Synthesis of a Mitosane Core Assisted by Diversity-Based Catalyst Discovery,” Org. Lett. 2001, 3, 2879-2882.

84. Danishefsky, S. J.; Schkeryantz, J. M. “Chemical Explorations Driven by An Enchantment with Mitomycinoids – A Twenty Year Account,” Synlett 1995, spec. issue, 475-490.

85. For review, see: Hartwig, J. F. “Transition Metal Catalyzed Synthesis of Arylamines and Aryl Ethers from Aryl Halides and Triflates: Scope and Mechanism,” Angew. Chem. Int. Ed. 1998, 37, 2046-2067.

86. Example of Lewis acid catalyzed allylstanane addition to iminium ion: Klitzke, C. F.; Pilli, R. A. “Enhanced trans Diastereoselection in the Allylation of Cyclic Chiral N-Acyliminium Ions. Synthesis of Hydroxylated Indolizidines,” Tetrahedron Lett. 2001, 42, 5605-5608.

87. For example: Oba, M.; Terauchi, T.; Hashimoto, J.; Tanaka, T.; Nishiyama, K. “Stereoselective Synthesis of (2S,3S,4R,5S)-Proline-3,4,5-d3,”Tetrahedron Lett. 1997, 38,5515-5518.

88. Wolfe, J. P.; Buchwald, S. L. “An Improved Catalyst System for Aromatic Carbon-Nitrogen Bond Formation: The Possible Involvement of Bis(Phosphine) Palladium Complexes as Key Intermediates,” J. Am. Chem. Soc. 1996, 118, 7215-7216; Driver, M. S.; Hartwig, J. F. “A Second-Generation Catalyst for Aryl Halide Amination: Mixed Secondary Amines from Aryl Halides and Primary Amines Catalyzed by (DPPF)PdCl2,” J. Am. Chem. Soc. 1996, 118, 7217-7218.

89. Wolfe, J. P.; Buchwald, S. L. “Palladium-Catalyzed Amination of Aryl Triflates,” J. Org. Chem. 1997, 62, 1264-1267; Louie, J.; Driver, M. S.; Hamann, B. C.; Hartwig, J. F. “Palladium-Catalyzed Amination of Aryl

217

Triflates and Importance of Triflate Addition Rate,” J. Org. Chem. 1997, 62, 1268-1273.

90. Mitchell, T. N.; Reimann, W.; Nettelbeck, C. “Hydrostannation of 1,1- Bis(trimethylstannyl)-ethylene and Some Related Olefins: First Steps Toward Constructing A Karplus-Type Curve For 3J(Sn-Sn),” Organometallics 1985, 4, 1044-1048.

91. Lee, A. S.-Y.; Dai, W.-C. “A Facile and Highly Efficient Sonochemical Synthesis of Organo-stannane via Barbier Reaction,” Tetrahedron 1997, 53, 859-868.

92. Tabuchi, T.; Inanaga, J.; Yamaguchi, M. “A Mild and Efficient Method for the Formation of Allylstannanes Utilizing Samarium Iodide-Induced Polarity Inversion of π-Allyl Palladium Complexes,” Tetrahedron Lett. 1987, 28, 215- 216.

93. Matsumoto, H.; Kasahara, M.; Matsubara, I.; Takahashi, M.; Arai, T.; Hasegawa, M.; Nakano, T.; Nagai, Y. “Conversion of Disilanes to Functional Monosilanes. XII. The Palladium(0)-Catalyzed Dechlorinative Silylation of Benzylic Chlorides with Methylchlorodisilanes. A Facile Route to Benzylmethylchlorosilanes,” J. Organomet. Chem. 1983, 250, 99-107.

94. Duhamel, P.; Hennequin, L.; Poirier, J. M.; Tavel, G.; Vottero, C. “1,5- Dicarbonyl Compounds. A General Preparation Method,” Tetrahedron 1986, 42, 4777-4786.

95. Mukaiyama, T.; Banno, K.; Narasaka, K. “New Cross-Aldol Reactions. Reactions of Silyl Enol Ethers with Carbonyl Compounds Activated by Titanium Tetrachloride,” J. Am. Chem. Soc. 1974, 96, 7503-7509.

96. Hartwig, J. F. “Transition Metal Catalyzed Synthesis of Arylamines and Aryl Ethers from Aryl Halides and Triflates: Scope and Mechanism,” Angew. Chem. Int. Ed. 1998, 37, 2046-2067; Mann, G.; Incarvito, C.; Rheingold, A. L.; Hartwig, J. F. “Palladium-Catalyzed C-O Coupling Involving Unactivated Aryl Halides. Sterically Induced Reductive Elimination to Form the C-O Bond in Diaryl Ethers,” J. Am. Chem. Soc. 1999, 121, 3224-3225; Palucki, M.; Wolfe, J. P.; Buchwald, S. L. “Synthesis of Oxygen Heterocycles via A Palladium- Catalyzed C-O Bond-Forming Reaction,” J. Am. Chem. Soc. 1996, 118, 10333- 10334; Palucki, M.; Wolfe, J. P.; Buchwald, S. L. “Palladium-Catalyzed Intermolecular Carbon-Oxygen Bond Formation: A New Synthesis of Aryl Ethers,” J. Am. Chem. Soc. 1997, 119, 3395-3396.

218

97. Kocienski, P. J. “Protecting Groups,” G. Thieme, New York, 1994; Greene, T. W.; Wuts, P. G. M. “Protective Groups in Organic Synthesis,” 3rd ed., Wiley, New York, 1999.

98. Corey, E. J.; Cho, H.; Rüecker, C.; Hua, D. H. “Studies with Trialkylsilyltrifluoromethyl-sulfonates: New Syntheses and Applications,” Tetrahedron Lett. 1981, 22, 3455-3458.

99. Wolfe, J. P.; Buchwald, S. L. “Palladium-Catalyzed Amination of Aryl Triflates,” J. Org. Chem. 1997, 62, 1264-1267; Louie, J.; Driver, M. S.; Hamann, B. C.; Hartwig, J. F. “Palladium-Catalyzed Amination of Aryl Triflates and Importance of Triflate Addition Rate,” J. Org. Chem. 1997, 62, 1268-1273.

100. Wipf, P.; Kim, H. “Total Synthesis of Cyclotheonamide A,” J. Org. Chem. 1993, 58, 5592-5594.

101. Dijksman, W. C.; Verboom, W.; Egberink, R. J. M.; Reinhoudt, D. N. “Synthesis of Mitomycin C Analogs. 1. Introduction of the Urethane Function at C-10 of the Pyrrolo[1,2-a]indole Skeleton,” J. Org. Chem. 1985, 50, 3791-3797.

102. Coleman, R. S.; Chen, W. “A Convergent Approach to the Mitomycin Ring System,” Org. Lett. 2001, 3, 1141-1144.

103. Dijksman, W. C.; Verboom, W.; Egberink, R. J. M.; Reinhoudt, D. N. “Synthesis of Mitomycin C Analogs. 1. Introduction of the Urethane Function at C-10 of the Pyrrolo[1,2-a]indole Skeleton,” J. Org. Chem. 1985, 50, 3791-3797.

104. Russowsky, D.; Petersen, R. Z.; Godoi, M. N.; Pilli, R. A. “Addition of Silylated Carbon Nucleophiles to Iminium and Cyclic N-Acyliminium Ions Promoted by InCl3,” Tetrahedron Lett. 2000, 41, 9939-9942; Maldaner, A. O.; Pilli, R. A. “Addition of Carbon Nucleophiles to Substituted N-Acyliminium Iions. A Stereoselective Route to Rrans-Fused Decahydroquinoline Systems,” Tetrahedron Lett. 2000, 41, 7843-7847; Pilli, R. A.; Russowsky, D. “The Stereochemistry of the Addition of Carbon Nucleophiles to Imines and Iminium Ions,” Trends Org. Chem. 1997, 6, 101-123; Yamamoto, Y.; Komatsu, T.; Maruyama, K. “Diastereofacial Selectivity in the Reaction of Allylic Organometallic Compounds with Imines. Stereoelectronic Effect of Imine Group,” J. Org. Chem. 1985, 50, 3115-3121.

105. Mitsunobu, O. “The Use of Diethyl Azodicarboxylate and Triphenylphosphine in Synthesis and Transformation of Natural Products,” Synthesis 1981, 1-28.

219

106. Agrell, I. “The Crystal Structure of [Zn(N3)2(C5H5N)2],” Acta Chem. Scand. 1970, 24, 1247-1261.

107. Viaud, M. C.; Rollin, P. “Zinc Azide Mediated Mitsunobu Substitution. An Expedient Method for the One-Pot Azidation of Alcohols,” Synthesis, 1990, 130-132.

108. Kotsuki, H.; Ohishi, T.; Araki, T. “A New Facile Method for the Chemoselective Reductive Transformation of Azides to N-(tert- Butoxycarbonyl)amines,” Tetrahedron Lett. 1997, 38, 2129-2132.

109. Taniguchi, T.; Ogasawara, K. “Extremely Facile and Selective Nickel- Catalyzed Allyl Ether Cleavage,” Angew. Chem. Int. Ed. 1998, 37, 1136-1137.

110. Saltzman, H.; Sharefkin, J. G. “Iodosobenzene,” Org. Synth., 1973, Coll. Vol. V, 658-659.

111. Armas, P.; Francisco, C. G.; Suarez, E. “Hypervalent Iodine Reagents: Preparation of Chiral Synthesis Building Blocks by Fragmentation of Anomeric Carbohydrate Alkoxy Radicals,” Angew. Chem. Int. Ed. 1992, 31, 772-774; Armas, P.; Francisco, C. G.; Suarez, E. “Fragmentation of Alkoxy Radicals: Tandem β-Fragmentation-Cycloperoxyiodination Reaction,” Tetrahedron Lett. 1992, 33, 6687-6690; Armas, P.; Francisco, C. G.; Suarez, E. “Fragmentation of Carbohydrate Anomeric Alkoxy Radicals. Tandem β-Fragmentation- Cyclization of Alcohols,” J. Am. Chem. Soc. 1993, 115, 8865-8866; Francisco, C. G.; Freire, R.; Gonzalez, C. C.; Suarez, E. “Fragmentation of Carbohydrate Anomeric Alkoxy Radicals. Synthesis of Azasugars,” Tetrahedron: Asym. 1997, 8, 1971-1974.

112. Fleming, I. “Improving the Friedel-Crafts Reaction,” Chemtracts 2001, 14, 405- 406; and references cited therein.

113. Meier, H.; Kretzschmann, H.; Kolshorn, H. “[abc]-Annealated [18]Annulenes,” J. Org. Chem. 1992, 57, 6847-6852.

114. Renz, M.; Meunier, B. “100 Years of Baeyer-Villiger Oxidations,” Eur. J. Org. Chem. 1999, 737-750.

115. Maruyama, K.; Nagai, N.; Naruta, Y. “Lewis Acid-Mediated Claisen-Type Rearrangement of Aryl Dienyl Ethers,” J. Org. Chem. 1986, 51, 5083-5092.

116. Pappo, R.; Allen Jr., D. S.; Lemieux, R. U.; Johnson, W. S. “Osmium Tetroxide-Catalyzed Periodate Oxidation of Olefinic Bonds,” J. Org. Chem. 1956, 21, 478-479.

220

117. Tabuchi, T.; Inanaga, J.; Yamaguchi, M. “A Mild and Efficient Method for the Formation of Allylstannanes Utilizing Samarium Iodide-Induced Polarity Inversion of π-Allyl Palladium Complexes,” Tetrahedron Lett. 1987, 28, 215- 216.

118. Reference for Me3SiO3H/LiAlH4: Posner, G. H.; Oh, C. H.; Milhous, W. K. “Olefin Oxidative Cleavage and Dioxetane Formation using Triethylsilyl Hydrotrioxide: Applications to Preparation of Potent Antimalarial 1,2,4- Trioxanes,” Tetrahedron Lett. 1991, 32, 4235-4238.

119. This condition was developed to remove methyl carbamate. Lenz, G. R. “Synthesis of 7-Oxygenated Aporphine Alkaloids from A 1-Benzylideneiso- quinoline Enamide,” J. Org. Chem. 1988, 53, 4447-4452.

120. This condition was developed to remove benzyl carbamate. Giannis, A.; Sandhoff, K. “Lithium Borohydride (Sodium Borohydride)-Chlorotrimethyl- silane, An Unusually Strong and Versatile Reducing Agent,” Angew. Chem. Int. Ed. 1989, 28, 218-220.

121. Baldwin, J. E. “Approach Vector Analysis: A Stereochemical Approach to Reactivity,” J. Chem. Soc., Chem. Comm. 1976, 738-741.

122. Johnson, C. D. “Stereoelectronic Effects in the Formation of 5- and 6- Membered Rings: the Role of Baldwin's rules,” Acc. Chem. Res. 1993, 26, 476- 482.

123. Danishefsky, S. J.; Schkeryantz, J. M. “Chemical Explorations Driven by An Enchantment with Mitomycinoids – A Twenty Year Account,” Synlett 1995, spec. issue, 475-490; and references cited therein.

124. For example: Acena, J. L.; Arjona, O.; Leon, M. L.; Plumet, J. “Total Synthesis of (+)-7-Deoxypancratistatin from Furan,” Org. Lett. 2000, 2, 3683-3686.

125. Schlingloff, G.; Sharpless, K. B. “Asymmetric Aminohydroxylation,” Asym. Oxidation React. 2001, 104-114; Andersson, M. A.; Epple, R.; Fokin, V. V.; Sharpless, K. B. “A New Approach to Osmium-Catalyzed Asymmetric Dihydroxylation and Aminohydroxylation of Olefins,” Angew. Chem. Int. Ed. 2002, 41, 472-475; Fokin, V. V.; Sharpless, K. B. “A Practical and Highly Efficient Aminohydroxylation of Unsaturated Carboxylic Acids,” Angew. Chem. Int. Ed. 2001, 40, 3455-3457; Rubin, A. E.; Sharpless, K. B. “A Highly Efficient Aminohydroxylation Process,” Angew. Chem. Int. Ed. 1997, 36, 2637- 2640; Li, G.; Angert, H. H.; Sharpless, K. B. “N-Halocarbamate Salts Lead to More Efficient Catalytic Asymmetric Aminohydroxylation,” Angew. Chem. Int. Ed. 1996, 35, 2813-2817.

221

126. Chen, S. -T.; Wang, K. -T. “A New Synthesis of O-Benzyl-L-serine,” Synthesis 1989, 36-37.

127. Kimura, T.; Vassilev, V. P.; Shen, G. -J. Wong, C. -H. “Enzymatic Synthesis of β-Hydroxy-α-amino Acids Based on Recombinant D- and L- Threonine Aldolases,” J. Am. Chem. Soc. 1997, 119, 11734-11742.

128. Sometimes the regioselectivity of Sharpless asymmetric dihydroxylation is hard to predict. Morgan, A. J.; Masse, C. E.; Panek, J. S. “Reversal of Regioselection in the Sharpless Asymmtric Aminohydroxylation of Aryl Ester Substrates,” Org. Lett. 1999, 1, 1949-1952.

129. Ciufolini, M.; Etheredge, S. J.; Segmuller, B. E. “A Stereospecific Route to Aziridinomitosanes: the Synthesis of Novel Mitomycin Congeners,” J. Am. Chem. Soc. 1985, 107, 3891-3898.

130. He, F.; Bo, Y.; Altom, J. D.; Corey, E. J. “Enantioselective Total Synthesis of Aspidophytine,” J. Am. Chem. Soc. 1999, 121, 6771-6772.

222

APPENDIX

SELECTED NMR SPECTRA

223

-0.5 0.0

0.5 24.2 24.2

1.0 1.3 1.3

1.5 2.0 2.0 1.2 1.2

2.0 1.3 1.3

2.5

1.0 1.0

2.0 2.0 3.0 1.1 1.1 3.5

) 1.2 1.2

Pr - 1.0 4.0 i ( N H OSi 133 H OTf 4.5 5.0 5.5 6.0 6.5

7.0

3.9 3.9 1.0 1.0 7.5 8.0 8.5 . 9

224

0 0 25 25 50 50 75 75 3 ) Pr - i ( N H OSi 133 100 100 H OTf 125 125 150 150 175 175 200 200

225

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 21.5 21.5 1.1 1.1

1.5 1.5

0.9 0.9 2.1 2.1 2.0 2.0 2.5 2.5

3.0 3.0

1.0 1.0 1.1 1.1

3.5 3.5

0.8 0.8

3 0.1

) 0.4 0.4

Pr 0.8

- i 0.8

( 4.0 4.0 0.8 0.8 OSi H N 4.5 4.5 134 5.0 5.0 5.5 5.5

6.0 6.0

1.0 1.0 6.5 6.5 1.0 1.0

7.0 7.0 2.0 2.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

226

0 0 25 25 50 50 75 75 3 ) Pr - i ( 100 100 OSi H N 134 125 125 150 150 175 175 200 200 225 225

227

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 0.9 0.9

1.5 1.5 1.0 1.0

1.1 1.1 2.2 2.2 2.0 2.0 2.5 2.5

3.0 1.0 1.0 3.0

1.0 1.0 0.8 0.8 3.5 3.5

OH H

N 0.9

135 0.9 4.0 4.0 0.8 0.8 4.5 4.5 5.0 5.0 5.5 5.5

6.0 6.0

1.0 1.0 6.5 6.5 1.0 1.0

7.0 2.0 2.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . .

228

0 0 25 25 50 50 75 75 OH H N 135 100 100 125 125 150 150 175 175 200 200

229

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 1.1 1.1

1.5 1.5

1.2 1.2 2.2 2.2 2.0 2.0 2.5 2.5

3.0 3.0 1.0 1.0 1.0 1.0

3.5 3.5 1.0 1.0

2 1.1 1.1

4.0 4.0 1.0 1.0

111

OCONH H - 1.1 1.1

4.5 4.5

N 9aR ) 1.8

9R *, ( 5.0 5.0 5.5 5.5

6.0 6.0

0.9 0.9 6.5 6.5 1.0 1.0

7.0 2.0 2.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

230

0 0 25 25 50 50 75 75 2 111 OCONH H - 100 100 N 9aR )

9R *, ( 125 125 150 150 175 175 200 200

231

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 0.0 0.0

1.5 1.5 3.6 3.6 2.0 2.0 2.5 2.5

3.0 3.0 1.1 1.1

1.1 1.1 1.1 1.1

3.5 3.5 2 1.2

4.0 4.0 1.1 1.1

111 OCONH H 1.2 - N

9aS )

4.5 2.1 2.1 4.5 9R *, ( 5.0 5.0 5.5 5.5

6.0 6.0

1.1 1.1 6.5 6.5 1.2 1.2

7.0 7.0 2.0 2.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

232

0 0 25 25 50 50 75 75 100 100 2 111 OCONH H - N 9aS )

125 125 9R *, ( 150 150 175 175 200 200

233

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.1 3.1 3.1 3.1 1.5 1.5 2.0 2.0 2.5 2.5

3.0 3.0

1.0 1.0 1.0 1.0 3.5 3.5

O 1.0 1.0

O 4.0 4.0

O 1.0

152 1.0

1.0 1.0 4.5 4.5 1.0 1.0 3 N

5.0 5.0

1.0 1.0

1.0 1.0 1.0 1.0

5.5 5.5 1.0 1.0 6.0 6.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

234

0 0 25 25 50 50 75 75 100 100 O O O 152 O 3 N 125 125 150 150 175 175 200 200

235

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.2 3.2 12.3 12.3 1.5 1.5 2.0 2.0 2.5 2.5

3.0 3.0 2.0 2.0

3.5 3.5 1.0 1.0

4.0 4.0

1.0 1.0

O 1.0 1.0 O O

154

4.5 4.5

O 2.0

1.8 1.8 5.0 5.0

CHN 1.0

2 1.1 1.1 BuO

- t 5.5 5.5 0.9 0.9 6.0 6.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

236

0 0 25 25 50 50 75 75 O O 100 100 O 154 O CHN 2 125 125 BuO - t 150 150 175 175 200 200

237

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.0 3.0 12.5 12.5 1.5 1.5 2.0 2.0 2.5 2.5

3.0 3.0

1.1 1.1 0.8 0.8 3.5 3.5 OH O

150 4.0 4.0

O 0.9 0.8 0.8

4.5 4.5 2.2 2.2 CHN 2 BuO

5.0 5.0 t 0.7 0.7 1.0 1.0 5.5 5.5 6.0 6.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

238

0 0 25 25 50 50 75 75 OH O O 150 100 100 O CHN 2 BuO - t 125 125 150 150 175 175 200 200

239

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.2 3.2 13.0 13.0 1.5 1.5 2.0 2.0 2.5 2.5

3.0 3.0 1.0 1.0

3.5 3.5 0.9 0.9 4.0 4.0 °C ) 27

Bu -

t 4.5 4.5

2 OCHO 1.0 1.0 MHz,

N CO 155 1.0 1.0 500 500

5.0 5.0 O , 3 O CDCl

(

5.5 5.5 1.0 1.0 6.0 6.0 6.5 6.5 7.0 7.0

7.5 7.5 1.0 1.0 8.0 8.0 8.5 8.5 . . 9

240

0 0 25 25 50 50 75 75 °C ) 27

Bu - t 2 OCHO 100 100 MHz, N CO 155 500 500

O , 3 O CDCl ( 125 125 150 150 175 175 200 200

241

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.3 3.3 13.2 13.2 1.5 1.5 2.0 2.0

2.5 2.5 -0.2 -0.2

3.0 3.0 1.0 1.0 3.5 3.5

°C ) 1.0 1.0 4.0 4.0 60

Bu - t 2 OCHO MHz, N CO 155 4.5 4.5

500 500

O 1.0 ,

3 O 1.0 1.0 5.0 5.0 CDCl (

5.5 5.5 0.9 0.9 6.0 6.0 6.5 6.5 7.0 7.0

7.5 7.5 1.0 1.0 8.0 8.0 8.5 8.5 . . 9

242

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.4 3.4 13.3 13.3 1.5 1.5

2.0 2.0 0.6 0.6

2.5 2.5

0.7 0.7 3.0 3.0 0.4 0.4

3.5 3.5 1.0 1.0

°C ) 0.7 0.7 Bu 4.0 4.0 60 -

2 OH

N 0.4

156 MHz,

0.7 0.7 0.4 0.4 4.5 4.5 O 500 500

O , 3 5.0 5.0 CDCl (

5.5 5.5 1.0 1.0 6.0 6.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9 9

243

0 0 25 25 50 50 75 75 °C ) Bu 60 -

t 100 100 2 OH N CO 156 MHz, O 500 500

O , 3 CDCl ( 125 125 150 150 175 175 200 200

244

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

3.0 3.0

8.9 8.9 2.9 2.9 1.5 1.5 2.0 2.0 2.5 2.5

3.0 3.0 1.0 1.0

3.5 3.5 2.0 2.0 4.0 4.0 °C ) 60

Bu - t

2 OBn 4.5 4.5 3.1 3.1 MHz, N CO 157 500 500

, O 3 5.0 5.0 O CDCl (

5.5 5.5 0.9 0.9 6.0 6.0 6.5 6.5

7.0 7.0 5.9 5.9 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

245

0 0 25 25 50 50 75 75 °C ) 60

Bu - t 2 OBn MHz, 100 100 N CO 157 500 500

, O 3 O CDCl ( 125 125 150 150 175 175 200 200

246

-0.5 -0.5 0.0 0.0 0.5 0.5 1.0 1.0 1.5 1.5

2.0 2.0

3.0 3.0 0.0 0.0 2.5 2.5

3.0 3.0 3.1 3.1

3.5 3.5 6.6 6.6 4.0 4.0 OMe 2 4.5 4.5 OCH OMe 161

5.0 5.0 2.2 2.2 MeO 5.5 5.5

6.0 6.0 1.1 1.1

6.5 6.5 1.0 1.0 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . .

247

0 0 25 25 50 50 75 75 OMe 2 100 100 OCH OMe 161 MeO 125 125 150 150 175 175 200 200

248

-0.5 -0.5 0.0 0.0 0.5 0.5 1.0 1.0

1.5 1.5 3.1 3.1 2.0 2.0 2.5 2.5

3.0 3.0

1.0 1.0 3.0 3.0

3.5 3.5 6.3 6.3 4.0 4.0 4.5 4.5 OH 166 OMe O

2.1 2.1 5.0 5.0

1.0 1.0 1.0 1.0

MeO MeOCH 1.0 1.0

5.5 5.5

1.0 1.0 6.0 6.0 1.0 1.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . .

249

0 0 25 25 50 50 75 75 100 100 OH 166 OMe O 2 MeO MeOCH 125 125 150 150 175 175 200 200

250

-0.5 -0.5 0.0 0.0 0.5 0.5 1.0 1.0 1.5 1.5

2.0 6.3 6.3 2.0 2.5 2.5

3.0 3.0

3.0 3.0 3.5 3.5 6.2 6.2 4.0 4.0 c A 4.5 4.5 O OMe O

167

2.1 2.1 5.0 5.0 2.1 2.1 MeO MeOCH

5.5 5.5 1.0 1.0

6.0 6.0

1.0 1.0 6.5 6.5 1.0 1.0 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . .

251

0 0 25 25 50 50 75 75 100 100 c A O OMe O 167 2 MeO MeOCH 125 125 150 150 175 175 200 200

252

-0.5 -0.5 0.0 0.0

0.5 0.5 15.4 15.4

1.0 1.0

7.1 7.1 6.8 6.8

1.5 1.5

1.7 1.7 3.2 3.2 2.0 2.0 2.5 2.5 3 3.0 3.0 Bu) n-

Sn(

3.4 3.4 3.5 3.5 7.1 7.1 168 4.0 4.0 OMe O 2 4.5 4.5

MeO MeOCH 2.1 2.1 5.0 5.0 5.5 5.5

6.0 6.0

1.0 1.0 1.0 1.0

6.5 6.5 1.0 1.0 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

253

0 0 25 25 50 50 75 75 3 Bu) n- Sn( 100 100 168 OMe O 2 125 125 MeO MeOCH 150 150 175 175 200 200

254

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0

8.3 8.3 0.9 0.9

1.5 1.5

3.0 3.0 2.0 2.0 1.1 1.1 0.8 0.8

2.5 2.5 0.9 0.9

3.0 3.0

0.9 0.9

1.1 1.1 3.5

3.5 1.0 1.0

OBn 3.1 4.0 4.0 1.1 1.1 Bu - t OH 2 4.0 4.0

H N CO

1.8 1.8

175 2.0 0.8 0.8 4.5 4.5

OMe O

2 2.0 2.0 5.0 5.0 MeO MeOCH 5.5 5.5 6.0 6.0

6.5 6.5 0.9 0.9

7.0 7.0 5.2 5.2 7.5 7.5 8.0 8.0 8.5 8.5 . . 9 9

255

0 0 25 25 50 50 75 75 OBn Bu - t OH 2 H N CO OH 100 100 175 OMe O 2 MeO MeOCH 125 125 150 150 175 175 200 200

256

-0.5 -0.5 0.0 0.0 0.5 0.5 1.0 1.0 1.5 1.5 2.0 2.0 2.5 2.5 3.0 3.0 3.5 3.5 4.0 4.0 OBn °C ) Bu - t OH 60 4.5 4.5 2

H N CO OH MHz, 175 5.0 5.0 500 500

, OMe O 3 2 5.5 5.5 CDCl ( MeO MeOCH 6.0 6.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

257

-0.5 -0.5 0.0 0.0

0.5 0.5 9.5 9.5

1.0 1.0

5.1 5.1 4.7 4.7 1.5 1.5

2.0 2.0

3.1 3.1 0.9 0.9 2.5 2.5

3.0 3.0

1.5 1.5

4.7 4.7

3.5 3.5

3.1 3.1

2.0 2.0 2 2.2 4.0 4.0

OBn 1.0 1.0 °C ) Bu BuPh - OH

- t t 2 60 60 2.0

4.5 4.5 H N CO OSi MHz,

176 2.0 2.0 5.0 5.0 500 500

OMe , O 3 2 CDCl 5.5 ( 5.5 MeO MeOCH

6.0 6.0 0.9 0.9 6.5 6.5

7.0 7.0

11.4 11.4 2.1 2.1 7.5 7.5 8.0 8.0 8.5 8.5 . . 9 9

258

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 12.5 12.5

1.5 1.5 3.0 3.0

2.0 2.0 1.1 1.1

2.5 2.5 0.7 0.7

3.0 3.0

0.6 0.6

2 0.9 0.9 3.5 3.5 BuPh

- t OH OBn

OSi 3.1 H 1.3 1.3 4.0 4.0

N 1.2 1.2

180 1.2 1.2

4.5 4.5 O O

1.1 1.1

1.4 1.4 1.8 1.8

5.0 5.0 MeO 1.0 1.0 5.5 5.5 6.0 6.0 6.5 6.5 7.0 7.0

7.5 7.5 4.4 4.4 8.0 8.0 8.5 8.5 . . 9

259

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 8.8 8.8

1.5 1.5 3.1 3.1 2.0 2.0 2.5 2.5 3.0 3.0

2 0.9 0.9 BuPh - t 3.5 3.5 OH OBn

OSi

3.0 3.0 N 1.2 4.0 4.0 181

O O

1.1 1.1

1.1 1.1 4.5 4.5 1.1 1.1 1.1 1.1

MeO

1.2 1.2

1.1 1.1 5.0 5.0 1.0 1.0 5.5 5.5 6.0 6.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

260

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 9.2 9.2 1.5 1.5 2.0 2.0

2.5 2.5 0.7 0.7

3.0 3.0 0.9 0.9 3.5 3.5 Bu

t 2 3.8 4.0 4.0

NHCO

Bn 0.9

2 0.8 0.8 OH 4.5 4.5 NHCO 204 O

5.0 1.9 1.9 5.0

MeO 0.8 0.8

5.5 5.5 0.8 0.8 6.0 6.0 6.5 6.5

7.0 7.0 5.0 5.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

261

-0.5 -0.5 0.0 0.0 0.5 0.5

1.0 1.0 9.2 9.2 1.5 1.5 2.0 2.0 2.5 2.5

3.0

3.0 0.8 0.8 0.7 0.7

3.5 3.5

°C ) 0.6 0.7 0.7

60 Bu

- 0.9 t

OH 4.0 4.0

MHz, 0.5 N CO

205 0.6 0.6 500

, 4.5 4.5 3 CHN HO 2 CDCl ( BnO

5.0 5.0 2.1 2.1

0.3 0.3

0.8 0.8 0.6 0.6 5.5 5.5 6.0 6.0 6.5 6.5

7.0 7.0 5.0 5.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9

262

-0.5 -0.5 0.0 0.0

0.5 0.5 3.0 3.0

1.0 1.0

1.6 1.6 1.3 1.3

1.5 1.5

3.1 3.1

1.0 1.0 3.0 3.0 2.0 2.0

2.5 2.5

1.0 1.0

0.8 0.8

1.1 1.1 1.0 1.0 3.0 3.0

3.5 3.5

6.0 6.0 3.1 3.1 4.0 4.0 HN 4.5 4.5 224 OMe OMe 5.0 5.0 MeO 5.5 5.5

6.0 6.0 1.0 1.0 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9 9

263

-0.5 -0.5 0.0 0.0

0.5 0.5 3.0 3.0

1.0 1.0 1.1 1.1

1.5 1.5 4.0 4.0

2.1 2.1

1.0 1.0

4.1 4.1 2.0 2.0 1.9 1.9

2.5 2.5

1.0 1.0 3.0 3.0

1.0 1.0 0.9 0.9

3.5 3.5

0.8 0.8

3.0 3.0 3.5 3.5 5.9 5.9 4.0 4.0 N HO 225 4.5 4.5 OMe OMe 5.0 5.0 MeO 5.5 5.5

6.0 6.0 0.9 0.9 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9 9

264

0 0 25 25 50 50 75 75 N HO 225 100 100 OMe OMe MeO 125 125 150 150 175 175 200 200

265

-0.5 -0.5 0.0 0.0

0.5 0.5 3.0 3.0

1.0 1.0

1.6 1.6 1.5 1.5

2.7 2.7 1.9 1.9

2.0 2.0 2.8 2.8

2.5 2.5

0.8 0.8

0.9 0.9

0.8 0.8 3.0 3.0 0.7 0.7

3.5 3.5

0.9 0.9

8.2 8.2 0.6 0.6 4.0 4.0

4.5 4.5 NO 0.7 0.7 226 5.0 5.0 OMe OMe 5.5 5.5 MeO

6.0 6.0 0.7 0.7 6.5 6.5 7.0 7.0 7.5 7.5 8.0 8.0 8.5 8.5 . . 9 9

266