<<

The Pennsylvania State University

The Graduate School

Eberly College of Science

STEREOSELECTIVE SYNTHESIS OF (Z)-2-ACYL-2-ENALS VIA

RETROCYCLOADDITIONS OF 5-ACYL-4-ALKYL-4H-1,3-DIOXINS:

APPLICATIONS IN NATURAL PRODUCT SYNTHESIS

A Thesis in

Chemistry

By

Ronald A. Aungst, Jr.

© 2007 Ronald A. Aungst, Jr.

Submitted in Partial Fulfillment

of the Requirements

for the Degree of

Doctor of Philosophy

August 2007 The thesis of Ronald A. Aungst, Jr., was reviewed and approved* by the following:

Raymond L. Funk Professor of Chemistry Thesis Advisor Chair of Committee

Steven M. Weinreb Russell and Mildred Marker Professor Of Natural Products Chemistry

Kenneth S. Feldman Professor of Chemistry

J. Martin Bollinger, Jr. Associate Professor of Biochemistry and Molecular Biology Associate Professor of Chemistry

Ayusman Sen Professor of Chemistry Head of Chemistry

*Signatures are on file in the Graduate School ABSTRACT

Reported herein are several approaches for the synthesis of substituted acrolein derivatives via the retrocycloaddition of substituted 4H-1,3-dioxins.

Upon thermal or Lewis acid mediated retro hetero Diels-Alder reactions of these substrates, an appropriately substituted acrolein derivative is generated

(stereoselectively in most cases) which can be utilized as a substrate for a variety of cycloaddition reactions. These cycloadditions, including Diels-Alder, hetero Diels-Alder, and [4 + 3], have provided access to several ring systems with excellent control over regio- and stereoselectivity.

A strategy for the stereocontrolled synthesis of (Z)-2-acylenals has been developed through utilization of a retrocycloaddition reaction of 5-acyl-4-alkyl-4H-

1,3-dioxins. This approach has provided concise synthetic routes to several natural products containing the 5-acyl-2H-3,4-dihydropyran substructure.

Application of this methodology has been showcased in the natural product syntheses of both the immunosuppressant loganetin and the cytotoxin euplotin A.

This methodology has also been investigated in a synthetic approach towards the synthesis of the xenicin core.

Similarly, retrocycloadditions of 4H-4-alkyl-5-(trialkylsilyloxy)-1,3-dioxins proceed smoothly in refluxing to afford (Z)-2-(trialkylsilyloxy)-2-alkenals with complete stereoselectivity. These enals undergo Sasaki-type [4 + 3] cyclization with in the presence of Lewis acids, in many instances with excellent regio- and/or stereoselectivity.

iii Also, throughout our investigations of forming 1,3-dioxins it was found that these retrocyloaddition precursors could be formed by a variety of reaction sequences. An example of a deconjugation approach for the formation of dioxins is exemplified in the development of a thermally labile solid-phase linker.

A similar approach was also utilized in the synthesis of the 1,3-dioxin precursors for the preparation of the labdane/clerodane ring system.

Subsequent thermolysis of this substrate provided clean conversion, via a one- pot retro- hetero Diels-Alder reaction and successive intramolecular Diels-Alder reaction, to provide the requisite trans-decalin ring system with an axial bridgehead aldehyde.

Finally, in our studies toward the synthesis of 1,3-dioxins, a Diels-Alder approach to the sesquiterpene illudin C ring system was investigated.

Unfortunately, this approach proved unproductive; however, it did lead to an alternative approach to illudin C. A convergent total synthesis of illudin C is described. The tricyclic ring system of the natural product was quickly assembled from cyclopropane and precursors via a novel oxime dianion coupling reaction and a subsequent intramolecular nitrile oxide-olefin cycloaddition.

iv

TABLE OF CONTENTS

LIST OF SCHEMES ix

LIST OF FIGURES xii

LIST OF TABLES xiii

ACKNOWLEDGEMENTS xiv

CHAPTER 1. Stereoselective Synthesis of (Z)-2-Acyl-2-enals 1

1.1 Previous synthesis and utility of 2-acyl-2-enals 1

1.2 Novel preparation of substituted acroleins via 3

retro Diels-Alder reactions of 1,3-dioxins

1.2.1 Stereoselective generation of 3-alkylacroliens 4

1.2.2 Synthesis of 2-acyloxy and 2-amido acroleins 5

1.2.3 Synthesis of 2-alkyl and 2-acylacroleins 8

1.3 Development of methodology for the stereoselective 10

generation of (Z)-2-acyl-2-enals

1.3.1 Synthesis of 4-alkyl-1,3-dioxan-5-ones 11

1.3.2 Regioselective generation of the vinyl triflates 13

1.3.3 Palladium catalyzed coupling reactions to yield 14

4,5-disubstituted-4H-1,3-dioxins

1.3.4 Investigation of the stereoselectivity of the 16

retrocycloaddition reactions of 5-acyl-4-alkyl-4H-1,3-dioxins

1.3.5 Investigation of the stereoselectivity of the 19

v hetero Diels-Alder reaction of (Z)-2-acyl-2-enals

CHAPTER 2. Application of (Z)-2-Acyl-2-enals in 22

Natural Product Synthesis

2.1 Background on the iridoid natural products 22

2.2 An approach to the iridoid natural products 26

2.2.1 An approach to the synthesis of the 26

immunosuppressant (±)-loganetin

2.2.2 Completion of the synthesis of (±)-loganetin 31

2.3 Total synthesis of the cytotoxin (±)-euplotin A 35

2.3.1 Biological importance of euplotin A 35

2.3.2 Structural determination of the euplotins 37

2.2.3 Biosynthetic route to the euplotins 39

2.3.4 Retrosynthetic analysis for the synthesis of euplotin A 41

2.3.5 Synthesis and alkylation of the dioxin side chain 44

2.3.6 Key retrocycloaddition/cycloaddition reaction 48

2.3.7 Completion of the (±)-euplotin A synthesis 55

2.4 Studies toward the total synthesis of the 60

xenicin class of compounds

2.4.1 Biological importance of the xenicins 60

2.4.2 Structural elucidation of xenicin 61

2.4.3 Synthetic approach to the xeniolides 62

2.4.2 A concise retrosynthetic analysis into the xenicanes 65

vi 2.4.3 Studies towards synthesis of the xenicane 66

ring system via an intramolecular cycloaddition

2.4.4 A concise retrosynthetic analysis into the xenicins 73

utilizing an intermolecular hetero Diels-Alder reaction

2.4.5 Studies towards synthesis of the xenicane 75

ring system via an intermolecular cycloaddition

CHAPTER 3. Alternative Approaches to 1,3-Dioxins 89

3.1 Stereoselective generation of (Z)-2-(trialkylsilyloxy)-2-enals 89

3.1.1 Application of (Z)-2-(trialkylsilyloxy)-2-enals 91

in [4 + 3] cycloaddition reactions

3.2 Deconjugation route to form 1,3-dioxins 101

3.2.1 Approaches to the first thermally labile solid-phase linker 101

3.2.2 An approach to the labdane and clerodane natural 110

product ring systems

3.3 Diels-Alder approach to synthesize 1,3-dioxins: 115

Application in the Illudin C synthesis

3.3.1 Biological importance of the illudin compounds 115

3.3.2 Retrosynthetic analysis for a Diels-Alder 118

approach to the illudin C ring system

3.3.3 Synthesis and application of the Diels-Alder 119

substrate for the illudin C model system

3.3.4 Retrosynthetic analysis for an intramolecular 121

vii nitrile-oxide cycloaddition approach to the illudin C ring system

3.3.5 Synthesis and application of the nitrile-oxide cycloaddition 122

substrate for the total synthesis of illudin C

3.4 Conclusions 127

EXPERIMENTAL SECTION 129

REFERENCES 202

viii LIST OF SCHEMES

Scheme 1. Previous utilization of 2-acyl-2-enals in organic synthesis 2 Scheme 2. Previous approaches to 2-acyl-2-enals 3 Scheme 3. Initial discovery of 1,3-dioxin retrocycloadditions 4 Scheme 4. Proposed route to taxane-A ring synthons 5 Scheme 5. Preparation and Diels-Alder reactions of 2-acyloxyacroliens 6 Scheme 6. Preparation of 2-triflamidoacroleins 7 Scheme 7. Uses of 2-amidoacroleins in natural product synthesis 8 Scheme 8. Preparation and utility 2-alkyl and 2-arylacroliens 9 Scheme 9. Preparation and utility of 2-acylacroliens 10 Scheme 10. Hydrazone approach to 4-alkyl-1,3-dioxan-5-ones 11 Scheme 11. Cyclohexylimine approach to 4-alkyl-1,3-dioxan-5-ones 12 Scheme 12. Preparation of 4-alkylidiene-1,3-dioxan-5-ones 12 Scheme 13. Regioselective synthesis of the vinyl triflate 36d 13 Scheme 14. Conformational analysis of the retrocycloaddition reaction 16 Scheme 15. Thermal isomerization of methyl 2-formyl-2-pentenoate 18 Scheme 16. Investigation of the enolization pathway to 19 olefin isomerization Scheme 17. Conformation analysis of the hetero Diels-Alder 20 reaction of enal 50 Scheme 18. Partridge’s synthesis of loganin 23 Scheme 19. Büchi’s synthesis of loganin 24 Scheme 20. Trost’s synthesis of loganin 25 Scheme 21. Claisen condensation approach to loganin 25 Scheme 22. Hiroi’s approach to loganin 26 Scheme 23. Retrosynthetic analysis of (±)-loganetin 27 Scheme 24. Loganetin retrocycloaddition/cycloaddition substrate 28 Scheme 25. Conformational analysis of the loganetin 29 cycloaddition reaction Scheme 26. Lewis acid promoted formation of (E)-retrocycloadduct 32 Scheme 27. Completion of total synthesis of (±)-loganetin 33 Scheme 28. Isolation of hydrolysis product of euplotin C 37 Scheme 29. Biosynthetic pathway to euplotin C 40 Scheme 30. Proposed biosynthetic pathway to euplotins A and B 41 Scheme 31. Retrosynthetic analysis of (±)-euplotin A 42 Scheme 32. Rate of 2-acyl-2-enal isomerization vs. 44 cycloaddition reaction Scheme 33. Initial approach to euplotin A alkylation substrate 45 Scheme 34. Ring closure of a xanthate radical 46 Scheme 35. Synthesis of alkyl iodide 113 47 Scheme 36. Preparation of the retrocycloaddition/cycloaddition 48 substrate 99 Scheme 37. Key retrocycloaddition/cycloaddition reaction of dioxin 99 49 Scheme 38. Installation of the ketone sidechain of euplotin A 50 Scheme 39. Retro-ene reaction of the Weinreb amide 51 ix Scheme 40. Attempted conversion to the acetoxy acetal 51 Scheme 41. Oxetane ring opening with ethanethiol 52 Scheme 42. Synthesis of alkyl iodide 126 53 Scheme 43. Preparation of the retrocycloaddition/cycloaddition 54 substrate 129 Scheme 44. Retrocycloaddition/cycloaddition reaction of dioxin 129 54 Scheme 45. Completion of the total synthesis of (±)-euplotin A 55 Scheme 46. Leumann’s total synthesis of coraxeniolide A (part 1) 63 Scheme 47. Leumann’s total synthesis of coraxeniolide A (part 2) 64 Scheme 48. An intramolecular hetero Diels-Alder approach 66 to the xenicins Scheme 49. Synthesis of allyl iodide 151 67 Scheme 50. Synthesis of homoallyl iodide 155 67 Scheme 51. Preparation of the retrocycloaddition/cycloaddition 68 substrate 147 Scheme 52. Retrocycloaddition and attempted cycloaddition of 146 69 Scheme 53. Use of strong Lewis acids for cycloaddition of enal 146 70 Scheme 54. Use of weak Lewis acids for cycloaddition of enal 146 70 Scheme 55. Synthesis of homoallyl iodide 160 71 Scheme 56. Second generation retrocycloaddition/cycloaddition 71 substrate Scheme 57. Retrocycloaddition/cycloaddition reaction of dioxin 163 72 Scheme 58. Intermolecular cycloaddition approach to the xenicin core 73 Scheme 59. Retrosynthetic analysis of RCM approach 74 Scheme 60. Cycloaddition reaction of 48 with trans-enol ethers 75 Scheme 61. Cycloaddition reaction of 48 with trans-enamine 178 76 Scheme 62. Cycloaddition reactions of 48 with tertiary enamides 77 Scheme 63. Cycloaddition reaction of 48 with secondary enamide 184 78 Scheme 64. Synthesis of the alkenyl substituted acylenal 172 79 Scheme 65. Kuwajima approach to aldehyde 194 80 Scheme 66. Synthesis of secondary enamide 196 80 Scheme 67. Key intermolecular hetero cycloaddition of 172 and 196 81 Scheme 68. Investigation for synthesis of enecarbamate 199 81 Scheme 69. Preparation of aldehyde 204 82 Scheme 70. Synthesis of the Curtius rearrangement substrate 206 83 Scheme 71. Alternative approach to the Curtius rearrangement 83 substrate Scheme 72. Enecarbamate synthesis via Curtius rearrangement 83 Scheme 73. Cycloaddition reaction of enal 172 and enecarbamate 210 84 Scheme 74. Attempts at N-alkylation or hydrolysis of cycloadduct 211 85 Scheme 75. Attempted cycloaddition of tertiary enecarbamate 214 85 Scheme 76. Preparation of enecarbamates and eneureas 86 Scheme 77. Potential synthesis of 2-acyloxy and 2-amido-2-enals 89 Scheme 78. Stereoselective synthesis of 2-acyloxy and 90 2-amido-2-enals Scheme 79. Literature examples of [4 + 3] cycloaddition reactions 91 x Scheme 80. Sasaki [4 + 3] cycloaddition reaction 92 Scheme 81. Proposed use of dioxin retrocycloaddition for the 93 Sasaki reaction Scheme 82. Harmata [4 + 3] example 93 Scheme 83. Synthesis of (Z)-2-(trialkylsilyloxy)-2-enals 240 94 Scheme 84. Intramolecular [4 + 3] cycloaddition reaction 97 Scheme 85. Investigation into silyl transfer step 98 Scheme 86. Mechanistic studies of the Sasaki [4 + 3] 99 cycloaddition reaction Scheme 87. Proposed thermally labile linker 103 Scheme 88. Synthesis of dioxin 263 103 Scheme 89. Resin attachment of the first generation linker 104 Scheme 90. Attachment and release of stearic acid 105 Scheme 91. Utility of allylic sulfone 269 106 Scheme 92. Preparation and attachment of the second generation 106 linker Scheme 93. Preparation and attachment of the third generation linker 108 Scheme 94. Solid and solution phase synthesis of a diester 108 Scheme 95. Thermal release of acid 285 109 Scheme 97. Retrosynthetic analysis of potamogetonyde 111 Scheme 98. Synthesis of b-ketophosphonate 295 112 Scheme 99. Synthesis of the labdane trans-decalin ring system 298 113 Scheme 100. Synthesis of the 9-alkyl trans-decalin ring system 301 114 Scheme 101. Dioxin synthesis via Diels-Alder reaction 115 Scheme 102. Mechanism of action of illudin S 117 Scheme 103. Retrosynthetic analysis of illudin C 118 Scheme 104. Preparation of cyclopropylidiene 314 119 Scheme 105. Attempted Diels-Alder reaction to provide dioxin 316 120 Scheme 106. Second generation retrosynthetic analysis of illudin C 122 Scheme 107. Vilsmeier-Haack formylations 123 Scheme 108. Optimized Vilsmeier-Haack formylation 123 Scheme 109. Synthesis of ketone 321 124 Scheme 110. Aldol by-product 331 formation 125 Scheme 111. Synthesis of nitrile-oxide cycloaddition substrate 125 Scheme 112. [3 + 2] cycloaddition of nitrile-oxide 318 126 Scheme 113. Completion of the total synthesis of illudin C 126

xi LIST OF FIGURES

Figure 1. Natural products with the 5-acyl-3,4-dihydro-2H-pyran 22 core substructure Figure 2. Stereochemical analysis of cycloadduct 83 30 Figure 3. Stereochemical analysis of cycloadduct 88 34 Figure 4. 1H NMR comparison of literature and experimental values 34 Figure 5. Cytotoxic products isolated from Euplotus crassus 36 Figure 6. nOe analysis of euplotin C 38 Figure 7. Natural products with the all cis-ring fusions 38 Figure 8. Biosynthetic precursors of the euplotins 39 Figure 9. Tietze’s molecular mechanics calculations 43 Figure 10. An approach to the lignans 45 Figure 11. nOe analysis of cycloadduct 96 49 Figure 12. Single crystal X-ray analysis of cycloadduct 130 55 Figure 13. NMR comparison of authentic and synthetic euplotin A 57 Figure 14. Subclasses of the xenicanes 60 Figure 15. Reproduction of the single crystal x-ray structure of xenicin 62 Figure 16. Additional examples of (Z)-2-(trialkylsilyloxy)-2-enals 95 Figure 17. Known and proposed linker release mechanisms 102 Figure 18. Examples of labdane/clerodane natural products 110 Figure 19. The illudin class of natural products 116

xii LIST OF TABLES

Table 1. Coupling conditions for 5-aryl and 15 5-acyl-4-alkyl-4H-1,3-dioxins syntheses Table 2. Retrocycloaddition reactions of 5-aryl and 17 5-acyl-4-alkyl-4H-1,3-dioxins Table 3. Hetero Diels-Alder reaction of 2-acyl-2-enals 21 Table 4. Cytotoxicity testing on euplotin A and its analogues 59 Table 5. Cycloaddition reactions of enecarbamates and 87 urea enamides Table 6. [4 + 3] Cycloaddition reactions of 96 (Z)-2-(trialkylsilyloxy)-2-enals

xiii ACKNOWLEDGEMENTS

Firstly, I would like to take this opportunity to thank my advisor Professor

Raymond Funk for his years of mentoring and teaching throughout my time at

Penn State. Without a doubt Ray’s wealth of knowledge, ability to teach, and driving work ethic has prepared me well for my entry into the industry. Though we may have “butted” heads a few times, I appreciate everything he has done to prepare me for the multiple challenges associated with organic chemistry.

Additionally, I would like to thank my committee members, Prof. Steven Weinreb,

Prof. Kenneth Feldman, and Prof. Martin Bollinger for their direction, encouragement and for serving on my thesis committee.

I would also like to take this chance to thank my colleagues with whom I worked with throughout my years in the Funk group. The lab would not have been quite the same without Tom Greshock, Seth Crawley, and Jim Fuchs, to keep it enjoyable even when we were pounding away at the bench all week long.

Additionally, I would like to express my gratitude to my current employer,

Albany Molecular Research, Inc., for their patience and understanding while I have been finalizing my Ph.D. The management team has been very understanding and supportive throughout this time and I owe them thanks for that.

The support and encouragement that I have received from my family during and after my graduate studies has been nothing short of exceptional. It was their

xiv understanding and love they have shown me throughout my graduate career at

Penn State has helped me stay focused on obtaining my goals.

Finally, and without a doubt, most importantly, I would like to thank my loving wife Dawn. She has been with me every step of the way and supported me no matter what was happening, whether it was working seven days a week and late evenings or sacrificing time to allow me to complete this thesis while I was working as well. She became somewhat of a graduate school widow at times, but stuck with me throughout it all. I am forever indebted to her for everything she has done for me and without her support I may never have made it.

xv Chapter 1

Stereoselective Synthesis of (Z)-2-Acyl-2-enals

1.1 Previous synthesis and utility of 2-acyl-2-enals

2-Acyl-2-enals are useful reactants for several types of organic transformations. One application was realized in Woodward’s synthesis of cephalosporin C which features a Michael addition of b-lactam 1 to 2-formylenal

2 (Scheme 1).1 Acylenals have also been utilized as dienophiles in Diels-Alder reactions with several dienes (Scheme 1).2 The secologanin (4) synthesis reported by Tietze exemplifies their use as heterodienes in inverse-electron demand hetero Diels-Alder reactions to form 5-acyl-3,4-2H-dihydropyrans 3

(Scheme 1).3

However, the exploitation of 2-acyl-2-enals in organic synthesis is limited by the paucity of methods for their construction and, in particular, stereocontrolled preparations. Woodward formed enal 2 via an aldol reaction of the sodium salt of malonaldehyde (5) and b,b,b-trichloroethyl glyoxylate (6) followed by subsequent dehydration of the resulting aldol adduct (Scheme 2).1

This compound was not isolated due to its high reactivity, and was used directly

in the aforementioned Michael reaction. Another method for the preparation of 2-

acyl-2-enals involves the conjugate addition of Grignard reagents to dialdehyde

Scheme 1. Previous utilization of 2-acyl-2-enals in organic synthesis

Michael Addition Reaction OCH CCl O O O 2 3 NH H OCH CCl O O H H 2 3 H H 80ºC N H N S H H O Boc n-octane H O O N S 1 2 Boc Diels-Alder Reaction - Dienophile O O H CHO

O 80ºC C6D6 100%

Diels-Alder Reaction - Heterodiene

O O COCCl3 O COCCl O 3 CO CH H KF 2 3 toluene OHC H H O 120ºC O H O O OEt 1.75 h OEt S O-ß-Glu O Ph 3 4 S Ph 14% correct diastereomer

7, the dimethyl enamine derivative of triformylmethane.4 Acid catalyzed

hydrolysis of the Michael adducts afforded the corresponding 2-formyl-2-enal

(Scheme 2). Similarly, these compounds could not be isolated, but were directly reacted with ethyl vinyl ether to afford hetero Diels-Alder adducts 8. A limitation of the previous two methods is the inability to vary the 2-acyl substituent.

Consequently, methods for the preparation of unsymmetrical 2-acyl-2-enals have been developed. Thus, sodium periodate oxidation of selenide 9 and concomitant

[2,3]-sigmatropic rearrangement gave rise to the carbomethoxy substituted enal

10 as a 70 : 30 ratio of Z:E isomers (Scheme 2).5 A final way of preparing symmetrical and unsymmetrical 2-acyl-b-substituted enals is Tietze’s tandem

Knovenagel hetero Diels-Alder reaction sequence.6 In this process an aldehyde

2

Scheme 2. Previous approaches to 2-acyl-2-enals

Aldol/ Dehydration Method O O ONa H O O CCl 1.) H O OCH2CCl3 H 3 2 H H H H O 2.) n-octane 5 6 80 °C O O

Grignard Addition O H O O THF 1. HClO4, H2O O OEt -40-25 °C H H BuMgBr H O H 2. OEt MgBr Me N O Bu 2 7 Bu N 63% 8 Oxidative Rearrangement O Cl O O

NaIO4, MeOH H3CO H3CO H H2O, NaHCO3 Z : E SePh 20 °C,63% 70 : 30 10 9

Knovenagel Reaction O O OEt O OEt O O O KF, toluene ; 2.2 : 1 Cl3C H Cl C H+ 3 H CCl3 H 110 °C Et 11 12 O undetermined Z : E mixure

is condensed with a 1,3-dicarbonyl compound, such as 4,4,4-trichloro-3- oxobutanal (11), to afford the desired acylenal 12, which is trapped in a hetero

Diels-Alder reaction with an enol ether (Scheme 2). Again, the acylenal was not

isolated and was formed as an undetermined mixture of Z : E stereoisomers.

1.2 Novel preparation of substituted acroleins via

retro Diels-Alder reactions of 1,3-dioxins

The previous methods for preparation of 2-acyl-2-enals illustrate the need

for a stereocontrolled method for their preparation under much less harsh

conditions. Extension of 1,3-dioxin-based methodology that has been previously 3

discovered in the Funk labs was considered to be a viable solution to these

shortcomings. This novel methodology is briefly summarized below.

1.2.1 Stereoselective generation of 3-alkylacroleins

Formation of (E)-a,b-unsaturated aldehydes under mild conditions via

retrocycloaddition reactions of 4-substituted-4H-1,3-dioxins (Scheme 3) was reported in 1988 by Bolton and Funk.7 In particular, it was discovered that 4H-

1,3-dioxin (13) could be metalated with sec-butyllithium and alkylated at the allylic position to give the 4-substituted-4H-1,3-dioxin 14. Heating dioxin 14 in toluene at reflux afforded the retrocycloadduct 17 in good overall yield (Scheme

Scheme 3. Initial discovery of 1,3-dioxin retrocycloadditions

hex sec-BuLi, -78ºC; O O HexylBr, -78-25ºC O O 14 13 79%

toluene, 110ºC

H O O O O hex H hex 15 16

74%

H O H O

hex hex 17

4

3). The stereoselective formation of the (E)-isomer was ascribed to a preferential

retrocycloaddition through the boat-like conformer 15 in which the C(4)

substituent is in the pseudoequatorial position rather than the boat-like conformation 16, which is destabilized by a flagpole, flagpole interaction between the hexyl substituent and the axial lone pair on O(1).

1.2.2 Synthesis of 2-acyloxy and 2-amido acroleins

The preparation of 1,3-dioxins possessing C(5) substituents was next examined. The compounds represent precursors to 2-substituted acroleins. The initial efforts in this area were stimulated by the need to prepare taxol A-ring synthons.8 Thus, it was believed that a highly regioselective and stereoselective

Diels-Alder reaction of 2-(acyloxy)acroleins 19 with substituted dienes would lead to the desired taxane A-ring cycloadducts 20 (Scheme 4). Prior to the report by

Scheme 4. Proposed route to taxane-A ring synthons

5

Funk, no methodology had been previously reported for synthesis of 2-

(acyloxy)acroleins. Based upon the initial discovery of the thermal lability of 1,3-

dioxins, efforts were put forth to examine the analogous retrocycloadditions of an

appropriately C(5) substituted dioxin 18, in turn available by acylation of 2,2-

dimethyl-1,3-dioxan-5-one (21). This proved to be straightforward. For example,

subjection of ketone 21 to triethylamine and catalytic DMAP in the presence of

isopropyl chloroformate afforded the desired 1,3-dioxin 18 (Scheme 5).

Scheme 5. Preparation and Diels-Alder reactions of 2-acyloxyacroliens

It was found that dioxins of this type also underwent smooth thermal

retrocycloaddition to yield 2-(acyloxy)acroleins such as 19, which were found to be highly reactive in the Diels-Alder reaction with highly substituted dienes yielding the desired taxane A-ring synthons 20.8

6

The preparation of dioxins possessing nitrogen substituents at C(5) was

also investigated. For example, conversion of the parent dioxinone 21 to the N-

benzylimine derivative, followed by sulfonylation provided 5-triflamido-4H-1,3- dioxin 22 (Scheme 6).9 As expected, this dioxin underwent a smooth retrocycloaddition upon thermolysis to afford 2-triflamidoacroleins 23 (Scheme 6).

Scheme 6. Preparation of 2-triflamidoacroleins

A variety of 2-amidoacroleins have been prepared according to this

protocol and subsequently employed in several total syntheses.10 For example,

an intermolecular 2-amidoacrolein cycloaddition has facilitated the rapid

construction of the novel tricyclic ring system of the immunosuppressant

FR901483 (Scheme 7).10a Moreover, the assembly of the ring systems of both fasicularin and lepadiformine have benefited from this methodology.10b,c It should

also be noted that it was discovered that 2-amidoacroliens can also function as

heterodienes, as well as dienophiles, in this case providing 4-aminoglycosides

(Scheme 7).11 Finally, these reactants have been utilized as excellent Michael

acceptors for electrophilic aromatic substitution reactions and facilitated the total

syntheses of aphanorphine and lennoxamine (Scheme 7).12

7

Scheme 7. Uses of 2-amidoacroleins in natural product synthesis

As Dienophile

TIPS O O TIPS HO P HO O O O H O O H O H HO N N N MeO N N H MeO O OMe O O OMe FR901483 O

As Heterodiene OEt O O O 100 °C OEt O H toluene O O N Ph 98% N Ph S O Ph S F C N O O CF3 3 S CF 3 22 23 O 4-aminoglycosides

As a Michael Acceptor

H O O O O N MeAlCl2, H N C H , O 6 6 N N rt, 4 h O O 62% O O O OH aphanorphine

1.2.3 Synthesis of 2-alkyl and 2-acylacroleins

As discussed in previous sections, the dioxinone 21 had served as a useful compound for the preparation of dioxins substituted with heteroatoms at

C(5). It seemed likely that dioxins substituted with carbon substituents, in particular, acyl substituents at C(5) could also be prepared from dioxinone 21 via the corresponding triflate 24.13 Indeed, this triflate participated in a number of coupling reactions en route to 5-alkyl and 5-aryl dioxins (Scheme 8). Moreover,

8

thermal retrocycloaddition of the resultant coupled products produced 2-alkyl and

2-aryl acroleins such as 25 in excellent yields (Scheme 8).

Scheme 8. Preparation and utility 2-alkyl and 2-arylacroliens

OMe

MeO O O 2 equiv. O H TIPS TMSO TMSO O a-Acrolein toluene TIPS Cation 4:1 110 °C, 5 h 100 %

H NEt O O 3 Hex2CuLi O O OTf O (CH2)5CH3 (CH ) CH DMAP O THF, –20 °C 2 5 3 O Tf O O 90 % 21 2 24 76 % 25 63% 1. (HO)2B 60 % Pd O

2. NH2OMe 89 % H H O O O 130 °C, 1 h O o-dichlorobenzene 96 % N N N MeO MeO

Most importantly, it was found that dioxins bearing 5-acyl substituents

could be prepared from triflate 24. In this case the coupling reactions involved palladium mediated carbon monoxide insertion reactions or Heck reaction of enol ethers and subsequent regioselective hydrolysis of the vinyl ether (Scheme 9).13

Upon thermolysis, the resultant 2-acyldioxins 26 and 28 also afforded the desired

retrocycloaddition products, in this case 2-acylacroleins such as 29 (Scheme 9).

However, these products were relatively unstable and were generally trapped in situ in subsequent Diels-Alder reactions (Scheme 9). Thus, thermolysis of 2- acylacroleins in the presence of dienes such as isoprene afforded the expected

Diels-Alder adducts 27. Alternatively, if an enol ether was employed as a

9

Scheme 9. Preparation and utility of 2-acylacroleins

trapping agent then an inverse electron demand hetero Diels-Alder reaction ensued to afford a 5-acyl-3,4-dihydro-2H-pyran such as 30 (Scheme 9).

1.3 Development of methodology for the stereoselective

generation of (Z)-2-acyl-2-enals

Based upon this substantial precedent, it was believed that a straightforward extension of the aforementioned methodology13 to the preparation of (Z)-2-acyl-2-enals would be possible through a stereoselective retrocycloaddition reaction of 4-substituted-5-acyl-4H-1,3-dioxins.14 However, this would require the development of a procedure for introducing substituents at the C(4) of the parent dioxanone 21 as well as regioselective formation of the vinyl triflate.

10

1.3.1 Synthesis of 4-alkyl-1,3-dioxan-5-ones

Initial studies for direct alkylation of the enolate of dioxanone 21 had met with little success. As reported by Majewski,15 generation of the enolate of the dioxanone, even at low temperatures, resulted in formation of polymeric products as a result of self-condensation reactions (indeed, dioxinone 21 polymerizes upon standing at ambient temperatures overnight). In order to avoid the self- condensation reaction, ketone 21 was initially converted into the N,N- dimethylhydrazone by refluxing with 1,1-dimethylhydrazine overnight (Scheme

10).16 Initial metalation of the hydrazone with tert-butyllithium followed by alkylation with alkyl halides, gave the alkylated hydrazones 31 as mixtures of geometric isomers (3 : 1, Z : E). The resultant alkylation products were then converted back to ketones 32 via ozonolysis or copper mediated hydrolysis17 of the hydrazone (Scheme 10).

Scheme 10. Hydrazone approach to 4-alkyl-1,3-dioxan-5-ones

The alkylation of the cyclohexyl imine 33 derivative of ketone 21 was also

explored. Imine formation proceeded uneventfully and upon concentration was

sufficiently pure for direct alkylation (Scheme 11). The imine 33 was metalated

11

with lithium diethylamide followed by the addition of an alkyl iodide to provide the

alkylated imine, which upon aqueous workup was hydrolyzed to the desired

alkylated ketones 32 in excellent yields (Scheme 11).18

Scheme 11. Cyclohexylimine approach to 4-alkyl-1,3-dioxan-5-ones

O NCy O LiNEt2, THF R 2.0 equiv. CyNH2 –78 - –35 °C; O O CHCl , Na SO O O 3 2 4 O O RX; NH4Cl (aq) 50 °C, 2 h 21 33 32 RX = a. BuI 82% d. I(CH2)2OTES 87%

Introduction of a substituent at the C(4) carbon was not limited, however,

to these alkylation protocols. The aldol reaction was also investigated and

initially was accomplished using the hydrazone chemistry. However poor yields

led us to turn to the Mukiyama aldol reaction using recently reported aqueous

conditions.19 The trimethylsilylenol ether derivative 34 of ketone 21 was easily prepared and was relatively stable if stored below 0 °C (Scheme 12). Reaction of the dioxin 34 with butyraldehyde, in the presence of a catalytic amount of ytterbium(III) triflate, led to formation of the desired b-hydroxy ketone as a 1 : 1

Scheme 12. Preparation of 4-alkylidene-1,3-dioxan-5-ones

1. butyraldehyde O OTMS 0.1 equiv. Yb(OTf)3 O (4 : 1) THF : H2O TMSCl, Et3N 64% O O O O CH2Cl2, 0 °C O O 2. MsCl, Et3N Z : E 5 : 1 91% CH2Cl2, –78 °C-rt 21 34 99% 35

12

mixture of stereoisomers (Scheme 12). These products could then be easily

dehydrated via formation and elimination of the resultant mesylate to provide a 5

: 1 (Z : E) mixture of enones 35 (Scheme 12).

1.3.2 Regioselective generation of the vinyl triflates

With the alkylated dioxanones 32 in hand, investigation into the

regioselective formation of the desired vinyl triflate was investigated. The

standard conditions employed for conversion of the parent ketone 21 to the

triflate (Tf2O, DMAP, Et3N) only provided recovered starting ketone 32d. It was

hoped that the alkylated dioxinone would be less prone to base-promoted polymerization and permit enolate formation at low temperatures. This was indeed the case. Simultaneous addition of the ketone and N- phenyltrifluoromethanesulfonimide to NaHMDS at –78 °C generated the kinetic

enolate, which was trapped in situ to provide the vinyl triflate 36d (Scheme 13).20

However, poor yields were obtained if the ketone/triflamide mixture was added at

a rate such that the internal temperature rose above –70 °C, so precautions were

taken to minimize any exotherm (i.e. syringe pump addition).

Scheme 13. Regioselective synthesis of the vinyl triflate 36d

O OTf OTES NaHMDS OTES

O O PhNTf2, THF O O –78 °C, 93% 32d 36d

13

1.3.3 Palladium catalyzed coupling reactions to yield

4,5-disubstituted-4H-1,3-dioxins

Studies on the palladium mediated couplings of the vinyl triflates 36 were extensive. Use of the previous conditions utilized for the Stille reaction of phenyl tributylstannane with the unsubstituted vinyl triflate 24 worked very well (Table 1, entry a).13 However, in order to introduce an acyl substituent at the C(5) position a variety of conditions were explored. Although the Heck reaction of ethyl vinyl ether13 and vinyl triflate 24, followed by hydrolysis was successful in installing the methyl ketone (entry b), utilization of the carbon monoxide insertion reaction conditions failed. A variety of palladium catalysts, as well as ligands, bases, solvents and CO pressures were investigated, until optimal conditions were finally uncovered.

Based upon the observations of Fu, it was hoped that an electron donating ligand might be optimal for the oxidative addition step of the electron rich triflate

36.21 Indeed this was the case, as use of the electron rich bidentate ligand 1,3- bis(diphenylphosphino)propane (dppp) and Pd(II) acetate were found to be the optimal pairing for this system (Table 1). The yields could also be improved by optimization of the solvent and base and depended upon the type of acyl substituent that was being introduced (Table 1).

14

Table 1. Coupling conditions for 5-aryl and 5-acyl-4-alkyl-4H-1,3-dioxins syntheses

Entry Starting Material Coupling Conditions Product Yield

OTf .02 equiv. Pd2(dba)3 55% .08 equiv. Ph3As a 1.2 equiv.PhSnBu3 O O DMF, 25 °C, 20 h 36a O O 37

OTf O .06 equiv. Pd(OAc)2 93% b 40 equiv. ethyl vinyl ether O O 2 equiv. Et N, DMSO, O O 36a 3 38 + 48 h; H3O

OTf O N(OMe)Me

.07 equiv. Pd(PPh3)4 c 1 atm CO 17% O O 15 equiv. HN(OMe)Me O O 36a 39 DMF, 40 °C, 20 h

OTf .08 equiv. Pd(OAc)2 O OMe OTES .08 equiv. dppp d 2.2 equiv. DIPEA OTES 86% O O 40 equiv. MeOH O O 36d 40 5 equiv. K2CO3, THF 25 °C, 24 h

O N(OMe)Me OTf .08 equiv. Pd(OAc)2 .08 equiv. dppp OTES OTES e 0.5 equiv. Et3N 81% O O 5 equiv. HN(OMe)Me, DMF O O 36d 41 25 °C, 24 h

OTf .08 equiv. Pd(OAc)2 O NMe2 .08 equiv. dppp OTES OTES f 2.2 equiv. DIPEA 90% O O 10 equiv. HNMe2, THF O O 36d 25 °C, 24 h 42

15

1.3.4 Investigation of the stereoselectivity of the

retrocycloaddition reactions of 5-acyl-4-alkyl-4H-1,3-dioxins

On the basis of the previously discussed stereocontrolled

retrocycloadditions of dioxin 14 (Scheme 3),7 we anticipated that the

retrocycloaddition of dioxins 37-42 would proceed preferentially through the boat

conformer eq-43 rather than the boat conformer ax-43 which suffers from the

flagpole-flagpole interaction with the O(3) axial lone pair (Scheme 14). However,

in this case, the preference for eq-43 might be attenuated by an A1,2 interaction between the C(5) acyl substituent and the C(4) equatorial substituent. This concern did not seem to be warranted since we found that the retrocycloaddition of amides 41 and 42 (Table 2, entries f and g) proceeded with high stereoselectivity and afforded stable compounds. Interestingly, the more facile

(50 °C) retrocycloaddition of ester 40 (Table 2, entry e) was less stereoselective

(Table 2). Moreover, the product enal 49 could not be isolated and decomposed upon concentration of the reaction mixture.

Scheme 14. Conformational analysis of the retrocycloaddition reaction

16

Table 2. Retrocycloaddition reactions of 5-aryl and 5-acyl-4-alkyl-4H-1,3-dioxins

17

Further experimentation revealed that the Z/E stereoisomeric ratios for the

2-acyl-2-enals are likely the result of thermodynamically controlled isomerizations. Thus, the 97 : 3 stereoisomeric mixture of amides 50 was photoisomerized (Hanovia 500 W, toluene, 3 h) to a different mixture of isomers

(Z : E, 80 : 20) that upon heating (toluene, 1 h) afforded a ratio of isomers (Z : E,

96 : 4) nearly identical to that initially obtained. Similarly, when we performed

Dess-Martin oxidation of the known methyl E-2-(hydroxymethyl)-2-pentenoate the reaction gave rise to a mixture of stereoisomers (Z : E, 70 : 30) of methyl 2- formyl-2-pentenoate (Scheme 15).22 It would appear that 2-acylenals are particularly prone to thermal isomerization, since dioxin 37 afforded a single stereoisomer of enal 44 and photoisomerized stereoisomeric mixtures of enal 44 did not equilibrate upon thermolysis. In an attempt to understand the pathway for this isomerization, dioxin 41 was thermolyzed in the presence of an equivalent amount of an analogous dioxin possessing a 1,1-dideuteriohexyl substituent instead of the 2-(triethylsilyloxy)ethyl substituent (Scheme 16). This reaction provided enal 50 without incorporation of deuterium and an enal derived from the deuterated dioxin, which was fully deuterated in the allylic position, which suggests that an enolization pathway is not responsible for the Z/E isomerization.

Scheme 15. Thermal isomerization of methyl 2-formyl-2-pentenoate

18

This experiment led us to conclude that the isomerization is most likely due to a

Michael addition/retro Michael addition pathway initiated by an unidentified

nucleophile in the solvent.

Scheme 16. Investigation of the enolization pathway to olefin isomerization

1.3.5 Investigation of the stereoselectivity of the

hetero Diels-Alder reaction of (Z)-2-acyl-2-enals

Each of the acylenals could then be converted by subsequent treatment

with an alkyl vinyl ether to an endo/exo mixture of the desired 5-acyl-3,4-dihydro-

2H-pyrans (Table 3). The cis- cycloadduct was preferentially formed through a presumed endo- transition state (Scheme 17). The stereochemistry of the cycloadduct 52 was rigorously determined by formation of lactone 53 from the major product of 52 (desilylation with TBAF and lactonization with sodium hydride) (Scheme 17). This conformationally locked lactone 53 exhibited the diagnostic axial anomeric proton resonance as a doublet of doublets with J = 2.3,

9.8 Hz. From this information we were then able to determine through nOe analysis that the bridgehead proton was also in the axial position. Unfortunately,

19

Scheme 17. Conformation analysis of the hetero Diels-Alder reaction of enal 50 H H O O Oi-Bu O O OMe N 8 N H OMe O OTES i-BuO H TESO O 5 equiv. 52a CH Cl exo 2 2 : O 12 kbar 24 h, 71% N OMe OTES O 50 H O Oi-Bu H OMe O O 92 N N H OMe O OTES 52b TESO endo 1. 1.5 equiv.TBAF THF, 0 °C, .25 h 98% 2. 1.1 equiv. NaH O Hb DMF, 25 °C, 24 h O O O 95%

53 H H a Ja,b = 9.8 Hz

nOe the minor cycloadduct 52a could never be isolated in high enough purity or quantity so the resultant lactone of this product was not prepared. The stereochemical assignments of the remaining cycloadducts 54, 55, and 56 were based solely upon the assignment for 52b.

20

Table 3. Hetero Diels-Alder reaction of 2-acyl-2-enals

entry Starting Material Conditions Product Yield (cis/trans)

O O Oi-Bu O Oi-Bu 5 equiv. OMe a 80% O N C7H8, 90 °C N 20 h (100/0) Bu OMe O Bu 39 54 H O O O Oi-Bu Oi-Bu 2 equiv. 81% b OMe MeO (74/26) CDCl3, 25 °C 20 h O OTES 72:28 OTES Z:E 55 49 H O O O Oi-Bu Oi-Bu OMe 5 equiv. 71% c N (92:8) N CH2Cl2, 12 kbar 24 h OMe O OTES 97:3 OTES Z:E 52 50 H O Oi-Bu O O Oi-Bu 85% d 5 equiv. Me N (93:7) CH2Cl2, 12 kbar 2 NMe2 24 h O OTES

56 OTES 51

21

Chapter 2

Application of (Z)-2-Acyl-2-enals in Natural Product Synthesis

2.1 Background on the iridoid natural products

The initial investigation of this new methodology for the stereoselective generation of (Z)-2-acyl-2-enals indicated significant potential in natural product synthesis, in particular, those compounds that contain the 5-acyl-3,4-dihydro-2H- pyran core substructure (Figure 1). This substructure is present in numerous natural products such as pumiloside, plumericin and euplotin A. In addition,

Figure 1. Natural products with the 5-acyl-3,4-dihydro-2H-pyran core substructure

22

reduced congeners 5-(1-hydroxyalkyl)-3,4-dihydro-2H-pyrans can be found in a

number of natural products including the antiangiogenesis compound

epoxyquinol A and several members of the antimitotic xenicin class of

compounds. Accordingly, the development of methodology for the construction of

these ubiquitous substructures has received a considerable amount of attention

by the synthetic community over the years.

Scheme 18. Partridge’s synthesis of loganin

The most studied member of the 5-acyl-3,4-dihydro-2H-pyran class of

natural products is the iridoid loganin,23 a prominent biogenetic precursor to a variety of other natural products containing the 5-acyl-3,4-dihydro-2H-pyran core substructure. Two syntheses were reported each of which featured installation of the dihydropyran ring via a photoannelation/retroaldol reaction (deMayo reaction) upon a substituted cyclopentene ring using the tricarbonyl compound 58.23a,b

Partridge utilized the fully functionalized cyclopentene 57 which was prepared by an asymmetric hydroboration of 5-methylcyclopentadiene to afford the desired

23

Scheme 19. Büchi’s synthesis of loganin

enantiomer for the [2 + 2] cycloaddition (Scheme 18).23b Büchi, alternatively utilized the symmetrical cyclopentene 63, and thus had to subsequently install the C(7) methyl (Scheme 19).23a

Both of these photochemical transformations resulted in a mixture of products resulting from the photocycloaddition and concomitant retroaldol reaction to directly afford the key 5-acyl-3,4-dihydro-2H-pyran core ring system.

The Partridge cycloaddition/retroaldol sequence resulted in a mixture of regioisomers as well as a mixture of diastereomers, of which the desired product

60 was obtained in 22 % yield (67 : 33 ratio of 63 : 61/62). Both of the syntheses were carried onto the final target.

More recent approaches to the dihydropyran core of loganin have typically involved oxidative procedures. For example, ozonolysis of a b,g-unsaturated cyclopentane carboxylate, followed by acidic workup of the dialdehyde has been the most common of this type of approach (Scheme 20). Thus, Trost employed a palladium catalyzed coupling to form the bicyclic ring system 66.23d Subjection of

24

Scheme 20. Trost’s synthesis of loganin

ketone 66 to a Shapiro reaction led to the desired a,b-unsaturated ester 67 which was deconjugated to provide the key cyclopentene 68 for oxidative cleavage.

Upon exposure to ozone, both double bonds were cleaved to form the desired carbonyl compounds. Upon workup with acetic acid, the dialdehyde cyclized to provide the dihydropyran core 70. Both Isoe23e and Demuth23f have employed similar sequences, which utilize an osmium catalyzed cleavage of the b,g- unsaturated ester. Moreover, the dialdehyde intermediate 69 has been generated by a number of approaches; for example, through acylation of esters such as ester 71 (Scheme 21).

Scheme 21. Claisen condensation approach to loganin

25

Scheme 22. Hiroi’s approach to loganin

A final oxidative approach to the 5-acyl-3,4-dihydropyran ring system of

loganin was utilized by Hiroi (Scheme 22).23f This synthesis employed lead

tetraacetate in a ring cleavage reaction of the vicinal hydroxy thioether 74 to provide a cyclic hemithioacetal 75. Deacetoxylation with catalytic p- toluenesulfonic acid then provided the desired dihydropyran 76.

Although no loganin synthesis to date has utilized a hetero Diels-Alder reaction to incorporate the 5-acyl-3,4-dihydro-2H-pyran ring system, this is by far the most expedient way to construct this substructure. The Tietze example shown in Scheme 1 demonstrated the utility of this reaction in natural product synthesis namely of secologanin (4) and sweroside.3a

2.2 An approach to the iridoid natural products

2.2.1 An approach to the synthesis of the immunosuppressant (±)-loganetin

It was believed that the loganin aglycon, loganetin, could be quickly accessed by use of the newly developed methodology for generation of (Z)-2- acyl-2-enals through an intramolecular cycloaddition reaction (Scheme 23). The

26

Scheme 23. Retrosynthetic analysis of (±)-loganetin

C(1) anomeric hemiacetal could be prepared by hydrolysis of carbamate 77. In turn, the ring system of loganetin would arise via the intramolecular hetero Diels-

Alder reaction through the (Z)-endo-a transition state of the (Z)-2-acyl-2-enal wherein the methyl and hydroxyl substituents emerge on the convex face of the newly formed cis-fused bicyclic ring system (Scheme 23). Finally, the retrocycloaddition/cycloaddition substrate 78 could be synthesized by utilizing chemistry developed by Hoppe. It was expected that the titanium reagent 79 would add to aldehyde 80 through the transition state shown to afford the desired anti- homoaldol adduct 78, by analogy to the many examples documented in the

Hoppe laboratories.24

27

Scheme 24. Loganetin retrocycloaddition/cycloaddition substrate

O i-Pr N O O O O O i-Pr O O i-Pr (i-PrO)3Ti OTES 1. HF, 99% 79 O N H i- Pr O O 2. Swern O O O 79% O O OH O 95% 40 80 78

The synthesis was initiated with the preparation of aldehyde 80.

Desilylation of dioxin 40 (page 15, Table 1) was accomplished by utilizing HF at

low temperature in order to avoid acid catalyzed retrocycloaddition. Oxidation of

the resultant alcohol occurred in a straightforward fashion using Swern conditions

to provide the key aldehyde substrate 80 for use in the Hoppe homoaldol

reaction (Scheme 24).31 Anti-addition of the titanium anion 79 proceeded smoothly to afford the desired dioxin 78 (as a single stereoisomer), providing the key substrate for the retrocycloaddition/cycloaddition reaction sequence

(Scheme 24).

Based upon the assumption that the kinetic product of the thermal retrocycloaddition of 78 would possess the (Z) configuration, four transition states are possible for the subsequent hetero Diels-Alder reaction (Scheme 25). It was expected that the endo- transition states would be favored over the exo-

transition states since the former lead to a less strained cis- rather that trans- fused bicyclic product. This of course depends upon the asynchronicity of the transition state. A transition state wherein the C-C bond formation is much more advanced than the C-O bond formation would be expected to lead to the trans- ring fusion, although it could not benefit from the putative secondary molecular orbital overlap. The (Z)-endo-a transition state where the substituents emerge

28

on the convex face of the molecule leading to 82 is highly preferred over the (Z)- endo-b transition state, which leads to 81 bearing concave substituents. Thus, it seemed possible that a 5-acyldihydropyran possessing substituents for a loganetin total synthesis could be rapidly assembled using this methodology.

Scheme 25. Conformational analysis of the loganetin cycloaddition reaction

29

Figure 2. Stereochemical analysis of cycloadduct 83

Thermolysis of dioxin 78 in refluxing toluene for 1h afforded a single

product. Utilizing 2D NMR spectroscopic studies, the fully elucidated

stereochemical structure was determined to actually be the trans-fused bicyclic ring system 83 derived from the (Z)-exo-b transition state (Figure 2). COSY NMR was used to assign the respective proton resonances and NOESY NMR allowed stereochemical assignments. With Ha showing nOe’s to Hd, He, and Hf, it was deduced that the methyl and hydroxyl substituents were on the opposite face of the molecule from Ha. The lack of nOe to Hb was an indicator that the ring fusion was likely not the expected cis-ring fused bicyclic. Irradiation of Hb in turn also lacked the nOe to back to Ha but did show one to Hc, which confirmed the trans- ring fused system and also showed that the carbamate was in the axial position.

Finally, the stereochemistry of the ring fusion was further indicated by the large coupling constant between Ha and Hb (Ja,b = 11.8 Hz).

The formation of a single product from this transformation suggests that the kinetic stereoselectivities of both the retrocycloaddition and cycloaddition are excellent. If the product from the retrocycloaddition undergoes rapid thermal isomerization prior to cycloaddition to provide the 73 : 28 (Z : E) mixture of enals previously observed (Table 2, entry e), then a mixture of products should have

30

been obtained since the E-enal can only cyclize through the exo transition state to afford a product with cis-ring fusion. Thus, we conclude that the cycloaddition proceeds in a very early transition state involving the Z-enal. This observation could be useful for preparing rare examples of natural products which possess trans-fused ring systems such as those previously mentioned in Figure 1: euplotin A, the xenicins, and epoxyquinol A.

2.2.2 Completion of the synthesis of (±)-loganetin

One solution for correcting the stereochemical problem for the loganetin synthesis would rely upon access to the alternate heterodiene stereochemistry

(i.e. the (E)-2-acyl-2-enal) that, as previously mentioned, could only undergo cycloaddition via an exo- transition state (vida infra). We had previously determined that the stereochemistry of the retrocycloaddition products could be altered utilizing a Lewis acid (Scheme 26). Thus, we found that the retrocycloaddition of the Weinreb amide dioxin 41 with BF3 in chloroform at room temperature gave a ratio (40 : 60, Z : E) that was different from those obtained from heating in toluene (97 : 3, Z : E). Moreover, the ratio did not change upon extended exposure to BF3 at room temperature (Scheme 26). One speculative explanation could be that equatorial complexation of the Lewis acid with the incipient acetone carbonyl is preferred and transition state 85 is sterically less encumbered than the alternative transition state 86. Thus, retrocycloaddition through transition state 85 provides the (E)-2-acyl-2-enal 50 preferentially over

31

Scheme 26. Lewis acid promoted formation of (E)-retrocycloadduct

the alternative (Z)-50 product. It was hoped that the analogous ester 40 might also provide the E isomer as the major product. However, we had previously shown that ester 49 (Table 2) isomerizes at ambient temperature and, consequently, it is not possible to determine the kinetic ratio of retrocycloaddition products under these conditions. Indeed, ester 40 provided the same ratio of products that had previously been observed (73 : 28, Z : E) under purely thermal conditions. However, it was hoped that the (E)-enal would be kinetically trapped in the intramolecular cycloaddition before isomerization could occur and thus provide the cis-ring fusion of the bicyclic system.

Protection of the hydroxyl group of 78 was required in order to accommodate the stoichiometric Lewis acid. To that end, the free alcohol was converted to the TBDMS ether and the resultant dioxin was subjected to the BF3 catalyzed retrocycloaddition/cycloaddition reaction (Scheme 27). We were gratified to observe that two cycloadducts (3 : 1 mixture) were formed in this reaction. The minor product 89 (upon desilylation) was identical to the previously

32

obtained product from the thermal retrocycloaddition/cycloaddition reaction of 78.

This product must arise from an analogues exo transition state of the intermediate (Z)-enal. The stereochemistry of the major isomer, 88, was determined via NOESY and COSY experiments (Figure 3). Irradiation of Ha showed significant nOe’s to Hb, Hc, and Hg. The nOe to Hb provided evidence that the ring fusion may in fact be the desired cis- stereochemistry.

Scheme 27. Completion of total synthesis of (±)-loganetin

O O i-Pr TBSOTf O O i-Pr (i-Pr)2NEt O N O N i-Pr i-Pr CH2Cl2, rt O O OH O 65% O O OTBS O

78 87 BF3•OEt2 toluene –20 °C - rt 78%

TBSO i-Pr i-Pr H i-Pr H i-Pr H N H N O O TBSO H O O H O O H H BF H 3 H BF3 O O (Z)-exo-b O (E)-exo-a O

OH O O H i-Pr i-Pr O N O N O H H HO 6 M HCl i-Pr H i-Pr (aq) O O H TBSO TBSO 1,4-dioxane O O 73% H H loganetin 3 : 1 O O O O

88 89

Also, it was clear that the carbamate moiety was in the pseudoequatorial position of the dihydropyran with the nOe to Hc. Irradiation of Hb in turn also had a strong nOe back to Ha as well as to Hc, which confirmed the cis-ring fused

33

Figure 3. Stereochemical analysis of cycloadduct 88

system (the coupling constant between Ha and Hb also supports the

stereochemical assignments, Figure 3) and definitively established that the

carbamate was in the pseudoequatorial position. Another nOe was observed

between Hb and the methyl substituent of the cyclopentene ring, providing

evidence that the substituents were on the convex face of the compound.

Irradiation of Hd in turn showed an nOe to He (and vice-versa) but not to any other protons on the ring system. Thus, the major isomer possessed the desired cis- ring fusion stereochemistry. This product arose from the (E)-enal cycloaddition through the (E)-exo-a transition state (the endo- transition state is

Figure 4. 1H NMR comparison of literature and experimental values

34

conformationally not possible for this type of ). Final conversion to the desired natural product, loganetin, was accomplished by simultaneous hydrolysis of the silyl protecting group and the carbamate functionality to furnish the racemic loganetin (Scheme 27). The 1H NMR resonances of the synthetic material were very similar to the those of the natural product which have been reported in the literature (Figure 4).23i,j

2.3 Total synthesis of the cytotoxin (±)-euplotin A

2.3.1 Biological importance of euplotin A

Euplotins A, B, and C are potent cytotoxins formed by the marine ciliated protist Euplotus crassus (Figure 5).25 These compounds were harvested in very low quantities from nearly 90 million cells, yielding only 2.5 mg of euplotin A, 1 mg of euplotin B, and 15 mg of euplotin C.25 A number of biological assays were performed on each of these compounds that showed cytotoxic properties towards related species of the Euplotus genus (minuta and vannus) resulting in the inhibition of cellular fission and/or death yet the E. crassus species was unaffected. Perhaps the most interesting of these assays showed that direct cellular to cellular contact was required for transmittance of the cytotoxins to other species. Thus, when E. crassus was separated from either E. minuta or E. vannus by a 20 mm filter no death of either of the two species was observed.

35

Figure 5. Cytotoxic products isolated from Euplotus crassus

However, if the mesh was removed both the E. minuta and E. vannus species were eliminated.

The mechanism of action of these compounds was not known at the time of this research, but recent investigations into the apoptotic activity of euplotin C have provided a better understanding.25d It has been reported recently that exposure of both mouse and rat tumor cell lines to euplotin C leads to activation of ryanodine receptors, release of Ca2+ stores in the endoplasmic reticulum and cytochrome C from the mitochondria, as well as activation of capsase-12. These factors, whether they are initiated independently or concomitantly, ultimately lead to activation of capsase-3 and ultimately apotosis.25d There are also currently efforts being afforded to investigating the antimicrobial activity of this compound as well.25e

Initial observations showed that upon exposure to sea water (pH = 8-9) all three cytotoxins are readily decomposed by hydrolysis of the acetate moiety and concomitant ring opening to the corresponding dialdehyde 90a. It is possible that these potentially toxic dialdehydes are also generated by cellular esterases. In

36

Scheme 28. Isolation of hydrolysis product of euplotin C

fact, when euplotin C was exposed to mildly basic conditions (K2CO3 in

MeOH/H2O) a mixture of 4 regio- and diastereoisomeric acetals 90b-e were isolated and characterized (Scheme 28).25c

2.3.2 Structural determination of the euplotins

The structures of each of the euplotin analogues were elucidated by

spectroscopic means, including high-resolution mass spectra, IR spectra as well

as a variety of NMR experiments. 2D COSY experiments as well as HMBC and

HMQC experiments afforded the 1H-1H and 1H-13C required for assignment of the individual 1H and 13C resonances.25b The stereochemistry of both euplotin A and

B were based upon experimental and molecular mechanics (MM) calculated coupling constants of the 1H NMR. The experimental J values matched closely with the structure assigned as the highly strained 2,6-trans-ring fused

37

Figure 6. nOe analysis of euplotin C

compounds as opposed to its all cis- diastereomer (MM calculations for euplotin

C showed the 2,6-trans ring system to be 8 kcal/mol more strained than the all cis- stereoisomer). It is important to point out that there have been natural products with this core ring system that do indeed possess the much less strained all cis- configuration such as the antibiotic udoteatriol (recently synthesized)26a and the alleopathic agent duroin (Figure 7).26b

Figure 7. Natural products with the all cis-ring fusions

Due to the limited availability of both euplotin A and B, detailed NMR

experiments were carried out on the more abundant euplotin C to elucidate the

stereochemical assignments. Detailed 1D nOe and 2D NOESY experiments showed the stereochemistry of the 2,6-ring fusion to be unequivocally the trans-

isomer as was initially proposed for euplotins A and B (Figure 6).25b

38

Determination of the absolute configuration of euplotin C was also

accomplished. Upon hydrolysis, as depicted in Scheme 28, the inseparable

mixture of acetals 90c-d were converted to their Mosher esters and the absolute

configuration was determined utilizing the standard in depth NMR analysis of

chemical shift changes between the diastereomers.25c The absolute configuration

is correctly depicted in Figure 5.

2.3.3 Biosynthetic route to the euplotins

In addition to the three euplotin tricyclic acetals an additional compound

was subsequently isolated from E. crassus. This compound, known as

preeuplotin, is believed to be the biosynthetic precursor for each of the

euplotins.25b,c Preeuplotin has been identified as the C(12) deprenyl analogue of udoteal a biogenetic precursor of udoteatrial (Figure 8).25b

Figure 8. Biosynthetic precursors of the euplotins

A biosynthetic route to the euplotins from preeuplotin has been proposed

by Pietra and coworkers (Scheme 29).25b The hypothesis involves initial

hydrolysis of the C(1) acetate followed by conjugate addition of the resultant on

39

the enal functionality. An additional hydrolysis of the C(15) acetate would provide a trialdehyde 96, which can be hydrated in a cascade cyclization reaction to provide the complete carbon skeleton of the euplotin compounds. This step must clearly benefit from enzymatic catalysis to offset the significant amount of strain in the tricyclic acetal 93. Further dehydration at C(14) and reacylation of the C(15) hydroxyl would provide euplotin C.

Scheme 29. Biosynthetic pathway to euplotin C

It is uncertain though whether oxidation of C(8) occurs before or after the cyclization to provide euplotin B (or in the case of euplotin A the oxidation of C(8) as well as reduction of the C(10) – C(11) double bond). However, if oxidation does indeed occur prior to this proposed cyclization pathway, then from our perspective there exists a particularly attractive alternative cyclization pathway.

With the C(8) carbon fully oxidized to the ketone, the resultant Z- acylenal could

40

Scheme 30. Proposed biosynthetic pathway to euplotins A and B

potentially undergo an inverse-electron demand hetero Diels-Alder reaction.

Cycloaddition with the C(1) vinyl acetate moiety through an exo- transition state would yield the 2,6-trans-ring fusion embodied in the euplotins (Scheme 30). The

anomeric acetal of the resultant 5-acyl-3,4-dihydropyran 94 could then be

hydrolyzed and epimerized to provide the cyclic hemiacetal 95. An acid promoted

cyclization of the hemiacetal 95 might then afford euplotins A and B.

2.3.4 Retrosynthetic analysis for the synthesis of euplotin A

It was postulated that euplotin A could be formed utilizing the dioxin-based

methodology previously discussed for formation of (Z)-2-acyl-2-enals. Thus,

euplotin A should be available from the hetero Diels-Alder adduct 96 by

introduction of the ketone side chain from the Weinreb amide followed by the

challenging, stereoselective exchange of the methoxy substituent for an acetate

41

Scheme 31. Retrosynthetic analysis of (±)-euplotin A

moiety (Scheme 31).14 Although the retrocycloaddition of dioxin 99 should

proceed stereoselectively to afford (Z)-2-acyl-2-enal 97, the stereochemical outcome of the subsequent heterocycloaddition reaction was not certain. Thus, an endo- transition state would lead to the significantly less strained all-cis isomer. The stereochemistry required for a euplotin A synthesis would be derived from an exo- transition state. Indeed, our previous studies during the loganetin total synthesis suggested that an exo- transition state would be preferred. In this case the trans- cycloadduct 83 was kinetically favored over the cis- diastereomer

82 even though it was calculated to be thermodynamically disfavored by 1.7 kcal/mol. However, molecular mechanics calculations predict a much more substantial energy difference for the two possible euplotin cycloaddition products 42

Figure 9. Tietze’s molecular mechanics calculations

(96 and 98) of 6.4 kcal/mol. Nonetheless, it was hoped that an early, highly

asynchronous transition state exo-97 as well as the cis-enol ether moiety would combine to favor the exo transition state. This prediction was tenuous, at best, based upon calculations using PCMODEL (v 6.0). We employed bond order parameters of 0.1 and 0.9 for the forming C-O and C-C bonds, respectively, and calculated an energy difference of 0.9 kcal/mol favoring the exo transition state.

However, if the bond orders used by Tietze (C-O 0.1, C-C 0.7) for a more synchronous heterocycloaddition were used, then the exo transition state preference was reduced to 0.5 kcal/mol (Figure 9).27 Even more disconcerting was the calculation of the exo- transition structure for the stereoisomeric E-2- acyl-2-enal heterodiene 101 (2.8 kcal/mol lower than exo-Z, C-O 0.1, C-C 0.9).

This alternative pathway to the undesired all-cis cycloadduct 98 was an especially serious concern if the rate of Z/E isomerization is faster than that of the cycloaddition (Scheme 32).

43

Scheme 32. Rate of 2-acyl-2-enal isomerization vs. cycloaddition reaction

2.3.5 Synthesis and alkylation of the dioxin side chain

The synthesis of the precursor to acylenal 97, dioxin 99, was initiated with the preparation of the known Paterno-Buchi photocycloadduct 102 of ethyl glyoxylate and furan (Scheme 33).28 Subsequent reduction of the ester moiety and protection as the silyl ether provided oxetane 103 (Scheme 33).

Stereoselective acid promoted ring opening of the acetal moiety of 103 afforded the required dihydrofuran 104 (Scheme 33).29

Initial attempts to deoxygenate the hydroxyl group were attempted using the Barton-McCombie protocol.30 The xanthate 105 was synthesized using standard conditions and subjected to radical deoxygenation conditions with tributyltin hydride and AIBN (Scheme 33).

44

Scheme 33. Initial approach to euplotin A alkylation substrate

O 1. LAH, Et2O H HO OTBS H 0 °C, 0.5 h, 76% OTBS MeOH, AcOH OEt O H O 2. TBSCl, imid., reflux, 20 h, OMe O H 60% O H DMF, rt, 2 h, O 84% 102 103 104

CS2, NaH MeI, 5 h 78%

SSnBu3 S S O Bu3SnH, AIBN OTBS S O OTBS toulene, reflux S O OTBS H H H 2 min.,110 °C, H OMe O OMe 97% OMe O O 107 106 105

However, instead of the desired reduction, the radical intermediate 106 was trapped by the enol ether moiety providing the bicyclic furanyl compound 107

(Scheme 33). This is a common ring system seen in the lignan natural products such as acuminatolide (Figure 10) and could be of synthetic value.31

Figure 10. An approach to the lignans

Radical induced thiolactonizations have been previously observed.32 For example, Iwasa reported that xanthate 109 gave the thiolactone 110 when subjected to tributyltin hydride and AIBN (Scheme 34). This cyclization was driven by the ring-opening of the allylic cyclopropane moiety in the ring system.

45

Scheme 34. Ring closure of a xanthate radical

All other attempts at radical initiated removal of the xanthate either resulted in cyclization as well or in no reaction.33 An alternative approach for deoxygenation was required in order to continue with the synthesis of the required iodide.

Initially the hydroxyl was converted to a number of good leaving groups

(iodide, mesylate, and tosylate) in hopes that an Sn2 displacement with a hydride source would allow this transformation.34 Yet, all attempts at reduction of this carbon resulted in rearomatization of the furan by elimination of methanol or no reaction at all.

These experiments led to the realization that the deoxygenation reaction might have to be accomplished prior to the reduction of the ester substituent of the

Paterno-Buchi product. To that end the oxetane ring of 102 was directly ring opened in methanol (Scheme 35).34 The resultant secondary alcohol 111 was converted to the a-acetoxy ester and subjected to a samarium diiodide reduction35 providing the desired deoxygenated dihydrofuran 112 (Scheme 35).

46

Scheme 35. Synthesis of alkyl iodide 113

O O MeOH, TFA, 0 °C, 1 h; OEt H OEt HO O H NaHCO3 (s) O H 99% OMe O 102 111

1. Ac2O, pyridine CH2Cl2, DMAP (cat.) 0 °C, 2 h

2. SmI2, HMPA EtOH, rt, 5 min, 99% (2 steps) I O 1. LAH, Et2O, 0 ºC, 5 min, 91% OEt H H 2. MsCl, Et3N OMe CH2Cl2, –78 °C OMe O 0.75 h, 100% O 3. NaI, acetone 113 112 reflux, 2 h, 95%

The synthesis of the key synthon was completed by reduction of the ester

followed by a two-step conversion to iodide 113 (Scheme 35).

The synthesis of dioxin 99 was accomplished by our standard protocol

(Scheme 36).14 Metalation of the cyclohexyl imine 33 with lithium diethylamide

followed by addition of iodide 113 afforded the alkylated dioxanone 114 in 85%

yield after hydrolytic workup with aqueous ammonium chloride (Scheme 36).

Conversion of 114 to triflate 115 was achieved in the usual fashion by kinetic

enolate formation and in situ trapping of the enolate with N-

phenyltrifluoromethanesulfonimide followed by purification on Florisil (Scheme

36). Finally, the Weinreb amide 99 was formed by a palladium catalyzed carbon

monoxide insertion reaction with 115 in the presence of N,O-dimethyl-

hydroxylamine (87%, Scheme 36).

47

Scheme 36. Preparation of the retrocycloaddition/cycloaddition substrate 99

Cy I N O OMe O O H H 33 O O O OMe LiNEt2, THF, -78 °C; O 113, –78 ºC - rt; 113 NH4Cl (aq) 114 85% NaHMDS,

PhNTf2, THF –78 °C, 96%

N O OTf OMe O OMe H H Pd(OAc)2, dppp, O O CO, MeNHOMe, O O O O DIPEA, DMF, rt 18 h, 87% 99 115

2.3.6 Key retrocycloaddition/cycloaddition reaction

The key tandem retrocycloaddition/cycloaddition reaction was then

investigated (Scheme 37). We were pleased to discover that heating 99 in

toluene at reflux for 1 hour afforded the desired 2,6-trans-fused cycloadduct 96 as the major isomer accompanied by the all cis- minor isomer (4.5 : 1), in 72% yield (Scheme 37). It is presumed that the retrocycloaddition of dioxin 99

proceeded stereoselectively to afford almost exclusively the Z-isomer,14 which

cyclized via the transition states exo-97 and endo-97 (Scheme 37). The isomers could be separated and the stereochemical assignment for the major isomer was confirmed by 2D NMR spectroscopy.

48

Scheme 37. Key retrocycloaddition/cycloaddition reaction of dioxin 99

O N H H H O O OMe O N O O O OMe H H N 4.5 toluene, 110 °C H 96 O H O H 1h, 72% O : O O H OMe O H H O 1 N 99 exo-97 H H H O OMe O O H 98

The stereochemistry of the major isomer, 96, was determined via NOESY

and COSY experiments (Figure 11). In particular, irradiation of Ha showed a

significant nOe to He but no nOe with Hb, thus providing evidence that the ring

fusion may in fact possess the desired trans- stereochemistry. Irradiation of Hb,

in turn, confirmed this stereochemical assignment by providing nOe’s to Hc and

Hd but none to Ha. Further evidence of this trans ring fusion was seen in the substantially large coupling constant between Ha and Hb (J = 10.4 Hz).

Irradiation of the acetal methane proton resonance for Hc further confirmed this cis-ring fusion assignment for the bicyclic acetal substituent since nOe’s with Hb and Hd were observed. Further support for this assignment was indicated by the small coupling constants of Hb with Hc and Hd (Jb,c = 3.2 Hz and Jb,d = 6.8 Hz).

Figure 11. nOe analysis of cycloadduct 96

49

Finally, irradiation of the remaining acetal proton resonance assigned for He,

showed the return nOe back to Ha confirming the assignment of the methoxy

substituent being on the same face as the cyclic acetal proton Hc.

Scheme 38. Installation of the ketone sidechain of euplotin A

O Li N H H H H H H O O OMe Et2O/pentane OMe O O (1 : 1), –50 °C, 1 h O O H 63% H 96 116 Li

THF, –78 °C, 0.5 h

- O N H H H H H H H O N - OMe O O OMe H H O O i-Pn O O H i-Pn H 117a 117b

H O N N H H H H H H O O OMe OMe O i-Pn O O i-Pn O H H 117c 118

Final conversion of cycloadduct 96 to the natural product required the functional

group manipulation of the Weinreb amide to the ketone sidechain followed by

conversion of the cyclic methyl acetal to the glycosidic acetate. Addition of

isopentyl lithium to amide 96 in THF resulted in both 1,2 and 1,4-addition products 116 and 117c, respectively (Scheme 38). Another byproduct was also isolated, namely, amide 118. Apparently, the enolate 117a derived from conjugate addition underwent a retro-ene reaction to provide the secondary methylamide 118 and formaldehyde (Scheme 38).36 This type of transformation

50

has been previously observed. Graham reported that the Weinreb amide 119 undergoes retro-ene reactions upon treatment with LDA to provide the secondary methyl amide 120 and formaldehyde (Scheme 39).37 Fortunately, we found that a less polar solvent system (Et2O, pentane) avoided this side reaction and resulted in exclusive formation of the desired 1,2-addition product 116.

Scheme 39. Retro-ene reaction of the Weinreb amide

O O MeO O MeO LDA N O N MeO H O N O –78 °C C H H H H C 3 H2 119 120

All that remained for the completion of the total synthesis of euplotin A was the transformation of the methyl acetal moiety of the tricycle 116 into the glycoside acetate. Standard conditions for this transformation involve the generation of an oxonium ion by acid promoted loss of methanol, followed by in situ capture by acetic acid.38 Unfortunately, and as anticipated, all attempts to

Scheme 40. Attempted conversion to the acetoxy acetal

51

effect this transformation resulted in ring opening products 121, no doubt a

consequence of the release of the significant ring strain (Scheme 40).

Clearly, in order to effect this final transformation a method for the

generation of the desired cyclic oxonium ion was required. Accordingly, we

turned to a cyclic monothioacetal (ie. replacement of the methoxy substituent

with a thioethyl substituent). It was envisaged that mercury Lewis acids could

chemoselectively trigger the desired oxonium ion formation. This would mandate

incorporation of the thioethyl substituent in the synthesis at the initial oxetane

ring-opening step.

Scheme 41. Oxetane ring opening with ethanethiol

O O H HO OEt OEt EtSH, BF3, O H CH3CN, –40 °C 2 min, 70% SEt O O 102 122

H+, EtSH

O O O

HO OEt HO OEt HO OEt H H SEt SEt O O EtS O 122 123 124

Accordingly, oxetane 102 was subjected to acid promoted stereoselective ring opening using ethanethiol instead of methanol (Scheme 41). Some of the

desired product 122 was obtained using the previously employed conditions.

However, the majority of the product was furan 123, the result of facile

aromatization. In addition, the product resulting from double addition of the thiol

52

to provide 124 was obtained (Scheme 41). After examining a number of protic acids and solvents it was eventually discovered that a Lewis acid catalyzed ring opening of 102 using BF3•OEt2 in acetonitrile in the presence of ethanethiol gave the desired trans-1,2-disubstituted dihydrofuran 122 in 70% yield (Scheme 41).

Scheme 42. Synthesis of alkyl iodide 126

O 1. Ac2O, pyridine O I DMAP (cat.) 1. LAH, Et2O HO OEt CH2Cl2, 0 °C, 2 h, 89% OEt 0 °C, 0.75 h, 97% H H H 2. SmI2, HMPA 2. I2, PPh3, imidazole SEt EtOH, THF, rt SEt , rt, 10 min SEt O 10 min, 84% O 82% O 122 125 126

The dihydrofuran 122 was converted to the desired iodide 126 using the same reaction sequence employed for the methyl acetal derivative. Thus, acylation of the secondary alcohol 122 followed by reductive removal of the acetate furnished ester 125 (Scheme 42). As before, the ester was reduced to the primary alcohol. However, instead of using the two-step procedure to form iodide 126, direct synthesis of the iodide was accomplished with iodine and triphenyphosphine. Aromatization of the more acid labile compound to the furan, a by-product seen whenever the thioacetal was placed in even slightly acidic conditions, was minimized using this protocol (Scheme 42).

The retrocycloaddition/cycloaddition substrate 129 was formed similarly to dioxin 99. The aza-enolate derivative of imine 33 was alkylated with iodide 126 affording dioxinone 127, which was subsequently converted to the pivotal acyldioxin 129 using the previously established protocol (Scheme 43). Again, all

53

Scheme 43. Preparation of the retrocycloaddition/cycloaddition substrate 129

Cy I N O SEt O O H H 33 O O O SEt LiNEt2, THF, –78 °C; O 126, –78 °C - rt; 126 NH4Cl (aq) 127 62% NaHMDS,

PhNTf2, THF –78 °C, 85%

N O OTf SEt O SEt H H Pd(OAc)2, dppp, O O CO, MeNHOMe, O O O O Et3N, DMF, rt 18 h, 91% 129 128 reactions were carried out under the least acidic conditions possible to minimize the ethanethiol elimination side reaction. This included avoiding subjection of these products to silica-gel chromatography and careful hydrolysis of the alkylated imine back to ketone 127 (Scheme 43).

Scheme 44. Retrocycloaddition/cycloaddition reaction of dioxin 129

O O 4.5 : 1 N O N O SEt N H H H toluene, 110 °C H H O H O H O 1h, 72% O SEt O O H SEt O O O H H H 130 129 exo

We were pleased to discover that the acyl dioxin 129 underwent a smooth retrocycloaddition/cycloaddition reaction to again deliver a 4.5 : 1 mixture of 5- acyldihyropyrans 130 (Scheme 44), the major isomer of which possessed the desired relative stereochemistry as indicated by the diagnostic 1H NMR proton coupling constants and NOESY experiments. Moreover, this assignment was

54

confirmed upon single-crystal X-ray analysis of the major isomer. It is of interest to note that the angles for C(11-6-7) and C(2-1-8), 126.4° and 117.6°, respectively, are similar to those calculated (Figure 12) and reflect the significant strain associated with this ring system.

Figure 12. Single crystal X-ray analysis of cycloadduct 130

2.3.7 Completion of the (±)-euplotin A synthesis

It was with some trepidation that we examined the remaining two functional group transformations upon the potentially labile, strained thioacetal

130, namely, introduction of the isopentyl ketone and acetal functionalities. While the amide 130 could be directly converted to the isopentyl ketone upon treatment with isopentyl-lithium, the yield (39%) was significantly reduced due to the

55

Scheme 45. Completion of the total synthesis of (±)-euplotin A

O O H H H H Et2O, H2O H H N KOt-Bu, rt, 24 h HO SEt SEt MeO 98 % O O O O H H 130 131 iso-pentylLi THF/Et2O/pentane 0 °C, 24 h, 74% O H H H O H Hg(OAc)2 H H OAc CH3CN, rt O O SEt H + b 1.5 h, 67 % O O 5.5 : 1 H Euplotin A 132 competing conjugate addition reaction. Consequently, the amide was hydrolyzed to the corresponding carboxylic acid 131, via conditions developed by

Gassman39 (typical hydrolysis conditions KOH/MeOH resulted in ring opening products), which could be cleanly converted to the desired ketone 132 (Scheme

45). Fortunately, the thioacetal moiety of this product could be selectively activated for exchange with an acetoxy substituent by treatment with mercuric acetate to afford racemic euplotin A as well as the b-anomer (5.5 : 1) (Scheme

45).40 The spectral properties of the synthetic material were nearly identical to those found for authentic euplotin A (Figure 13).

56

Figure 13. NMR comparison of authentic and synthetic euplotin A

Authentic Euplotin A 1H NMR

Synthetic Euplotin A 1H NMR

57

Authentic Euplotin A 13C NMR

Synthetic Euplotin A 13C NMR

58

Table 4. Cytotoxicity testing on euplotin A and its analogues

The examination of the biological properties of euplotin A has not been extensively investigated due to its relatively low abundance in comparison to euplotin C. Accordingly, Dr. Michael Smith of the Hershey Medical School performed cytotoxicity assays of acetals 116 and 132 as well as synthetic racemic euplotin A against two cancer cell lines. We were pleased to learn that

132 as well as the natural product show promising activity (Table 4). In addition, a DNA damage assay was performed via a PCR amplification process and showed a negative test for euplotin A, in keeping with a subsequent report that euplotin C may operate via ultimate activation of capsase-3.25d

59

2.4 Studies toward the total synthesis

of the xenicin class of compounds

2.4.1 Biological importance of the xenicins

The xenicanes are a growing class of over 100 diterpenes and norditerpenes isolated from coelenterates (soft corals and gorgonians) and brown algae.41 The vast majority of the xenicanes exhibit potent biological properties that include cytotoxicity, anti-inflammatory and antifungal activities. A notable structural feature of the majority of the xenicanes is the strained trans- cyclononene ring. In many cases the C(7),C(8) trans-double bond is further

Figure 14. Subclasses of the xenicanes

60

oxidized to the corresponding epoxide. The xenicanes can be subdivided into three main groups according to the type of ring that is present connecting C(4a) –

C(11a) carbons (Figure 14).41l The xenicin group contains a 3,4-dihydropyran ring at this junction site and is represented by the first xenicane that was isolated, xenicin,41m as well as 9-deacetoxy-14,15-deepoxyxeniculin.41a,j Xenicanes containing a tetrahydropyranone ring in place of the 3,4-dihydropyran are referred to as xeniolides. Examples of this class of compounds include coraxeniolide-A and xeniaol.41b Finally, members of the xenaphyllane subgroup, such as xeniaphyllenol A, possess a cyclobutane ring fused at C(4a) and

C(11a).41l

2.4.2 Structural elucidation of xenicin

The structure of the diterpenoid xenicin, isolated in 1977 from the soft coral Xenia elongata, was determined by a number of spectroscopic methods.41m

The high resolution mass spectrum as well as elemental analysis showed the compound to have the molecular formula C28H38O9. The infrared spectrum showed two carbonyl absorptions corresponding to acetate stretches. Finally,

NMR experiments helped to elucidate a partial structure but the full structure could not be determined directly via spectroscopic methods. Instead, single crystal x-ray diffraction led to the final determination of not only the carbon structure and its relative stereochemistry, but the absolute stereochemistry

61

Figure 15. Reproduction of the single crystal x-ray structure of xenicin

as well (Figure 15).

Structures that have been isolated more recently have been determined in the same manner. However, due to the significant expansion in the area of 2D

NMR including COSY, HMBC, HMQC, DEPT, and NOESY experiments, the connectivity as well as the stereochemistry of these structures has also been proven via spectroscopic information.

2.4.3 Synthetic approach to the xeniolides

In view of the interesting structures and biological properties of the xenicanes, it is surprising that only one synthetic effort toward their synthesis has been reported to date. This work resulted in a rather lengthy (28 steps) synthesis of coraxeniolide A (Scheme 46) by Leumann,42a which drew heavily upon the

62

Scheme 46. Leumann’s total synthesis of coraxeniolide A (part 1)

classic Grob-fragmentation-based caryophyllene total synthesis reported by

Corey.42b

Leumann’s synthesis started from the Hajos-Parrish diketone 133 derived

from Scheme-(+)-proline. Initial chemo- and stereoselective reduction and protection of the saturated ketone followed by another stereoselective reduction of the a,b-unsaturated ketone provided the desired monoprotected diol 134

(Scheme 46). A two carbon homologation of the allylic alcohol was

accomplished by initial enol formation followed by a highly stereoselective [1,3]

sigmatropic rearrangement to yield the desired aldehyde 135. Protection of the

of the aldehyde as the dimethyl acetal was uneventful and allowed for a

diastereoselective epoxidation to provide a excellent ratio (11:1, a:b) of the

desired epoxide 136. Ring opening with cyanide anion followed by base- promoted equilibration of the resulting nitrile and subsequent protection of the tertiary alcohol gave the nitrile 137. A two-step reduction sequence of nitrile 137 via an intermediate aldehyde provided the alcohol 138 required for cyclization to the Grob-fragmentation substrate. This cyclization was accomplished by

63

deprotection of the acetal and both the tertiary and secondary alcohol moieties

with aqueous HCl to produce the lactol 139. Finally, oxidation to the lactone and

selective formation of the tosylate from the secondary alcohol gave the desired

fragmentation substrate 140 (Scheme 47).

Scheme 47. Leumann’s total synthesis of coraxeniolide-A (part 2)

Although the Grob-fragmentation was possible via the lactone substrate

140 in excellent yields, it was later found that the lactone functionality was not compatible with the introduction of the exocyclic olefin (Scheme 47). Thus, the lactone was reduced back to the lactol and then exposed to NaH to affect the stereoselective fragmentation. Subsequent protection of lactol 141 was

necessary to perform the Tebbe olefination, which was followed by deprotection

and reoxidation to the desired lactone to furnish the complete ring system 142 of

the xeniolide natural products. Finally, conversion to the final natural product

was realized by alkylation of the lactone to provide a 1 : 5.7 mixture of

diastereomers 143 and 144 that were epimeric at the C(4) position. The

64

undesired a-epimer 143 was the major product. Fortunately, equilibration of the

desired isomer 144 could be accomplished with catalytic TBD (1,5,7-

triazabicyclo[4.4.0]dec-5-ene) to give the desired product 144 as well as 143

(separable, 3 : 1 mixture). This ratio corresponded to the calculated

thermodynamic ratio one would expect, which showed that the a-diastereomer

was preferred by 0.6 kcal/mol over its C(4) epimer.

2.4.2 A concise retrosynthetic analysis into the xenicanes

The synthetic potential of (Z)-2-acylenals as heterodienes was previously

documented in the context of the loganetin and euplotin A total syntheses. We

decided to apply this methodology to the stereoselective synthesis of the strained

bicyclic trans-cyclononene ring system of the xenicane compounds via an intramolecular hetero Diels-Alder reaction.

Similar to the approach taken in the euplotin A synthesis, the prenyl side chain would be introduced via addition of an alkyl lithium to Weinreb amide 145

(Scheme 48). The resultant ketone would be stereoselectively reduced and acylated to complete the side chain (Scheme 48). Final conversion of the methyl acetal to the natural product could then be accomplished by acid catalyzed transformation to the glycosidic acetate (Scheme 48). The bicyclic ring system

145 could in turn be constructed via an intramolecular heterocycloaddition of a

(Z)-2-acyl-2-enal 146. The inverse electron demand hetero Diels-Alder

65

Scheme 48. An intramolecular hetero Diels-Alder approach to the xenicins

cycloaddition of the trans- enol ether was expected to occur via transition state endo-146. However, molecular mechanics calculations suggested that the endo-146 transition state was only slightly preferred over exo-146 (0.24 kcal/mol, PC Model), although secondary molecular orbital interactions would further favor the endo- transition state. The cycloaddition substrate 146 would be available via stereoselective retrocycloaddition of dioxin 147 providing the (Z)- acylenal 146.

2.4.3 Studies towards synthesis of the xenicane

ring system via an intramolecular cycloaddition

The first step in the investigation for the key cycloaddition step was preparation of the retrocycloaddition/cycloaddition substrate. Preparation of the alkyl iodide side chain for attachment to dioxin 147 was undertaken utilizing a

66

Scheme 49. Synthesis of allyl iodide 151

known synthesis (Scheme 49).43 3-Methyl-3-buten-1-ol (148) was easily converted into the desired protected b-hydroxy ketone 149 followed by Horner

Wadsworth-Emmons reaction to give an inseparable mixture of geometric isomers. Dibal-H reduction to the allylic alcohol 150 followed by formation of iodide 151 provided the substrate for alkylation of trans-4-methoxy-3-buten-2-one

(152).

Scheme 50. Synthesis of homoallyl iodide 155

ODMTS MeO NaHMDS, HMPA MeO O THF, –78 °C; 111, 83% O 152 153

CpTiMe2, THF, 65 °C 45%

I 1. TBAF, THF ODMTS 99% MeO MeO 2. I2, PPh3 imid., 93% 155 154

Standard procedures for alkylation of iodide 151 onto ketone 152 typically resulted in 1,4-addition of the base into the vinylogous ester. However, slow 67

addition of 152 to NaHMDS in THF at –78 °C via syringe pump, followed by

addition of HMPA at such a rate as to maintain the internal temperature below –

70 °C provided the highly reactive enolate. Dropwise addition of iodide 151 at –

78 °C then afforded the product 153 in excellent yield (Scheme 50). The

resultant vinylogous ester 153 was converted to the desired dienol ether 154

through a Petasis-Tebbe reaction (Scheme 50).44 Desilylation and iodination of

triene 154 provided the iodide side-chain 155 (Scheme 50).

Scheme 51. Preparation of the retrocycloaddition/cycloaddition substrate 147

I O 1. LiNEt2, –78 °C OMe MeO NCy O O ; NH Cl O O 4 155 33 156 85% NaHMDS PhNTf2, –78 °C 92%

O N OTf OMe O OMe Pd(OAc)2 dppp, Et3N O O O O MeNH(OMe) DMF, rt 147 68% 157

Dioxin 147 was synthesized by the standard procedures used in the

previous syntheses.14 As before, alkylation was accomplished by reaction of

iodide 155 with the aza-enolate of imine 33 (Scheme 51). Kinetic deprotonation

of the resultant ketone 156 with NaHMDS followed by trapping with triflamide provided triflate 157 (Scheme 51). Finally, palladium mediated carbonylation of triflate 157 under the typical conditions supplied the desired dioxin 147 (Scheme

51).

68

Scheme 52. Retrocycloaddition and attempted cycloaddition of 146

Initially, it was hoped that subjection of dioxin 147 to our standard conditions for retrocycloaddition would produce an acylenal that would subsequently undergo the desired heterocycloaddition. Indeed, exposure of dioxin 147 to refluxing toluene did provide the (Z)-acylenal 146 (Scheme 52). However, further heating only resulted in decomposition of the acylenal. The transformation was also attempted at lower temperatures (i.e. refluxing benzene or room temperature) yet the it still only produced decomposition products or no reaction at all, respectively

(Scheme 52). High-pressure (12 kbar) conditions were also examined, but did not afford any of the desired cycloadduct (Scheme 52). In addition, strong Lewis acids such as TiCl4, SnCl4, BF3 and EtAlCl2 were utilized at low temperatures (–

78 °C) but provided no identifiable products (Scheme 53). Weaker Lewis acids including Yb(fod)3, Eu(fod)3, and Me2AlCl showed no reaction with 146 even at elevated temperatures before the starting material began to decompose

69

Scheme 53. Use of strong Lewis acids for cycloaddition of enal 146

(Scheme 54). The Lewis acids were also utilized under high-pressure (12 kbar) again affording complete decomposition of 146 (Scheme 54).

Scheme 54. Use of weak Lewis acids for cycloaddition of enal 146

We speculated that the dienol ether functionality was most likely responsible for the instability of 146. Accordingly, we decided to investigate the preparation and cycloaddition of a mono-enol ether such as 163 that also embodies functionality that could later be converted to the exocyclic olefin. To that end, the vinylogous ester 153 was reduced under Lüche45 conditions followed by protection of the resultant alcohol 158 as the MOM ether (Scheme

55). Standard deprotection of silyloxy primary alcohol 159 and conversion to 70

Scheme 55. Synthesis of homoallyl iodide 160

iodide 160 provided the substrate for alkylation onto dioxanone 21 (Scheme 55).

Typical protocols for the alkylation and conversion to vinyl triflate 162 (Scheme

56)14 proceeded in excellent yields. Moreover, the carbonylation reaction was much more successful, providing the retrocycloaddition/cycloaddition substrate

163 in exceptional yields (Scheme 56), presumably due to the absence of the more labile dienol ether functionality.

Scheme 56. Second generation retrocycloaddition/cycloaddition substrate

OMOM I O 1. LiNEt2, –78 °C OMe MeO NCy O O OMOM ; NH Cl O O 4 160 33 161 87% NaHMDS PhNTf2, –78 °C 85%

OMOM OMOM O N OTf OMe O OMe Pd(OAc)2 dppp, Et3N O O O O MeNH(OMe) DMF, rt 163 82% 162

71

Subjection of amide 163 to refluxing toluene again provided the retrocycloadduct 164 (Scheme 57). Unfortunately, further subjection of enal 164 to these conditions resulted in decomposition (Scheme 57). A variety of high- pressure cycloadditions as well as Lewis acid catalyzed reactions (TiCl4, SnCl4,

BF3, EtAlCl2, Yb(fod)3, Eu(fod)3, and Me2AlCl) also resulted in decomposition of enal 164 (Scheme 57). Thus, efforts to form the xenicin ring system via an intramolecular hetero Diels-Alder reaction were terminated and we directed our attention to an alternative strategy that would employ an intermolecular heterocycloaddition.

Scheme 57. Retrocycloaddition/cycloaddition reaction of dioxin 163

72

2.4.4 A concise retrosynthetic analysis into the xenicins utilizing an

intermolecular hetero Diels-Alder reaction

A less risky approach to the synthesis of the xenicins might be

accomplished by an initial stereoselective intermolecular acylenal hetero Diels-

Alder cycloaddition reaction, providing a dihydropyran that might be suitably functionalized for subsequent transition metal catalyzed cyclization to afford the strained nine membered ring. Specifically, trans-3,4-dihydro-2H-pyrans 167 could be easily constructed and then closure of the alkenyl substituents could be

Scheme 58. Intermolecular cycloaddition approach to the xenicin core

effected using either Suzuki46 or olefin metathesis reactions47 to afford the trans- cyclononene ring 166 (Scheme 58). Both of these reactions have been used extensively in the preparation of medium and large rings.46, 47

73

Scheme 59. Retrosynthetic analysis of RCM approach

We decided to initially pursue the olefin metathesis based strategy for the

construction of the trans-cyclononene ring (Scheme 59). Based on literature

precedent, it seems reasonable that a cyclononene ring could be constructed from the olefinic side-chains of dihydropyran 170.47,48 However, it seemed likely

that the thermodynamically preferred cis-cyclononene would be obtained from this reaction. The cyclononene 169 could be inverted via the corresponding epoxide 168 by treatment with Ph2PLi according to the Gassman protocol that was successful for producing trans-cyclooctene.49 The stereoselectivity of the epoxidation reaction is of no consequence since each betaine intermediate can eliminate on the periphery of the cyclononane to produce cyclononene 166 atropisomers that will interconvert.50 Prior to initiating this olefin inversion sequence, the silyl ether will be first converted to the corresponding selenide and epoxidation with m-CPBA will proceed with concomitant oxidation of the selenide followed by thermally promoted syn-elimination of the selenoxide. Finally, the

74

trans-enol ether 171 and heterodiene 172 should provide the endo- cycloadduct

170 to correctly set the relative stereochemistry of the three stereogenic centers, although the precedent for this cycloaddition was negligible.

2.4.5 Studies towards synthesis of the xenicanes

ring system via an intermolecular cycloaddition

An investigation of the cycloaddition reactions of 2-acylenals with trans- enol ethers was initially investigated to determine their endo/exo selectivity. For example, the cycloaddition of acylenal 48 with enol ether 173 under high pressure conditions gave, as expected, the endo- cycloadduct 174 as the major

Scheme 60. Cycloaddition reaction of 48 with trans-enol ethers

isomer (Scheme 60). It is of interest to note that the diastereoselectivity at the hydroxyl bearing carbon was good, although the relative configuration is unknown. Unfortunately, we were unable to repeat this reaction so other trans- enol ethers 175-177 were also investigated. Again, these reactions also

75

Scheme 61. Cycloaddition reaction of 48 with trans-enamine 178

provided no discernable products under a variety of conditions including: thermal,

Lewis acid catalyzed, high pressure, or combinations thereof.

The need to increase the reactivity of the dienophile was evident, so trans-

enamine dienophiles were also investigated. The cycloaddition of pyrrolidine

enamine 178 with enal 48 proceeded smoothly under relatively mild conditions

(Scheme 61). The stereochemistry of the cycloadduct was determined by

NOESY experiments, but all attempts to convert the highly labile aminal 179 to

the desired acetal 180 failed (Scheme 61). This made it necessary to identify a dienophile that was intermediate in reactivity to the enol ether and enamine that would afford a cycloadduct that could be converted to the acetal. To this end, our attention turned towards an enamide substrate. To the best of our knowledge

76

Scheme 62. Cycloaddition reactions of 48 with tertiary enamides

these compounds had not previously been utilized in hetero cycloadditions of this type.

The trans-enamide dienophiles 182 and 183 were initially prepared since they could also be converted to secondary enamides via retro conjugate addition reactions vida infra. Thus, condensation of 3-aminopropionitrile with butyraldehyde provided (E)-imine 181 quantitatively (Scheme 62). Subjection of this imine to acetyl chloride/triethylamine as well as benzyl chloroformate/2,6- lutidine provided the trans-enamides 182 and 183 (Scheme 62). Both of these tertiary enamides were employed as dienophiles in high-pressure hetero Diels-

Alder reactions with hetero diene 48, neither of which provided cycloaddition products (Scheme 62).

77

Scheme 63. Cycloaddition reaction of 48 with secondary enamide 184

It was hoped that the sterically less encumbered and more planar

secondary enamides would undergo the desired cycloaddition. The secondary enamides 184 and 185 were readily obtained by subjecting enamides 182 and

183 to potassium tert-butoxide in THF (Scheme 63). Subjection of these enamides to high-pressure cycloaddition reactions with enal 48 satisfyingly afforded cycloaddition products 186 and 187 (Scheme 63). The stereochemistry of the resultant cycloadducts was determined through NOESY studies and found to be exclusively the endo-cycloadducts.

With reassuring precedent for the endo-selectivity of the trans-secondary enamides now in hand, we directed our attention to the preparation of alkenyl substituted acylenal 172 and enamide 196 required for the eventual olefin metathesis reaction. Our standard protocol for the formation of enal 172 was initially investigated.14 Alkylation of dioxin 21 with 1-iodo-3-methyl-3-butene followed by triflate formation was straightforward (Scheme 64). However, the

Heck product 190 was formed exclusively when we used the typical 78

carbonylation conditions at atmospheric pressure of CO. However, if the solvent

was saturated with CO before addition of the palladium source, and the reaction

was run under 500 psi of CO pressure, then a 3 : 1 mixture of the desired dioxin

was formed relative to the Heck product (Scheme 64). The dioxin 191 was

subjected to standard thermolysis conditions to provide enal 172 (Scheme 64).

Scheme 64. Synthesis of the alkenyl substituted acylenal 172

NCy O OTf

LiNEt2 I NaHMDS

O O NH4Cl 83% O O PhNTf2 O O 86% 33 188 189

Pd(OAc)2, dppp HN(OMe)Me CO (500 psi) 72%

O O N O O N N H O 110 °C, toluene O O H 1 h, 87% O O 3 : 1 O O (Z/E) 97 : 3 172 191 190

The retro Michael procedure for formation of the secondary enamide

analogous to 169 was next investigated. Monoprotection and oxidation of 1,4- butenediol (192) provided the a,b-unsaturated aldehyde 193 necessary for conjugate addition of a 3-butenyl side chain (Scheme 65). We employed the

Kuwajima protocol for cuprate 1,4-additions and subsequent hydrolysis of the resultant silylenol ether provided the desired aldehyde 194 (Scheme 65).51

79

Scheme 65. Kuwajima approach to aldehyde 194

As per the standard protocol, the imine was formed from aldehyde 194 and amine 139 and then acetylated to afford enamide 195 (Scheme 66). Base promoted retro Michael addition of enamide 195 provided the desired secondary enamide 196 for use in the cycloaddition reaction (Scheme 66).

Scheme 66. Synthesis of secondary enamide 196

The key intermolecular hetero cycloaddition was now investigated. Initial subjection of 172 and 196 to high-pressure (12 kbar) provided the endo- cycloaddition product 197 as a 2 : 1 mixture of diastereomers (Scheme 67).

Unfortunately, the maximum yield that could be obtained was 50%, due, in part, to the slow decomposition of enal 172.

80

Scheme 67. Key intermolecular hetero cycloaddition of 172 and 196

It was believed that enecarbamate such as 199 might be a more electron rich heterodienophile and participate in a more productive cycloaddition.

Conversion of 194 to the benzyloxy carbamate 199 was accomplished in a similar fashion as previously reported, but, unfortunately, this reaction could not be optimized and only provided poor yields (Scheme 68).

Scheme 68. Investigation for synthesis of enecarbamate 199

Use of a Curtius rearrangement was an alternative approach that was examined to form dienophiles such as 199. However, an aldehyde that was one carbon shorter than 194 would be needed in order to make the correct substrate for the Curtius rearrangement. Conversion of the aldehyde to the a,b-

81

unsaturated acyl azide, via formation of the a,b-unsaturated carboxylic acid,

would provide the correct substrate.

Scheme 69. Preparation of aldehyde 204

The appropriate aldehyde was synthesized in a straightforward fashion

from ethyl cyanoacetate (200). Alkylation of 200 with 3-butenyl iodide (201) and

NaH provided a separable 5 : 1 mixture of mono- 202 and bis-alkylated products respectively (Scheme 69). Chemoselective reduction of the ester functionality and protection of the resultant alcohol afforded nitrile 203, which was further reduced to the desired aldehyde 204 with Dibal-H (Scheme 69). Standard

Horner-Wadsworth-Emmons reaction of aldehyde 204 with trimethyl phosphonoacetate and NaH provided the desired a,b-unsaturated ester 205.

However, all attempts to saponify the ester resulted in concomitant deprotection of the silyl ether to afford hydroxy acid 206 (Scheme 70). Thus, another approach for formation of the a,b-unsaturated carboxylic acid was investigated.

82

Scheme 70. Synthesis of the Curtius rearrangement substrate 206

It occurred to us that the methodology previously used for preparation of

secondary enamides via the retro Michael addition reaction could be extended to

formation of carboxylic acids. Thus, the appropriate phosphonate was formed

Scheme 71. Alternative approach to the Curtius rearrangement substrate

via an Arbuzov reaction and was reacted with aldehyde 204 as before (Scheme

71). The resultant ester 207 was subjected to retro Michael reaction with

potassium tert-butoxide to give the a,b-unsaturated carboxylic acid 208 in

excellent yields (Scheme 71). Acid 208 was then converted to the acyl azide 209

followed by Curtius rearrangement in the presence of tert-butanol to provide the

BOC enamide 201 for use as the heterodienophile in the hetero Diels-Alder reaction (Scheme 72).52

Scheme 72. Enecarbamate synthesis via Curtius rearrangement

83

With a viable route to the enecarbamate in hand we turned to the investigation of the cycloaddition reaction. Initial subjection of 172 and 201 to high-pressure conditions (12 kbar) provided the endo-cycloaddition product 211 as a 2 : 1 mixture of diastereomers (76%, Scheme 73). As hoped, the yield was improved over the enamide counterpart. Initially, transfer of the silyl ether to the exocyclic olefin was examined. However, all the attempts at desilylating the primary alcohol resulted in epimerization at the anomeric carbon to give 212 and/or ring opened products (Scheme 73).

Scheme 73. Cycloaddition reaction of enal 172 and enecarbamate 210

84

Scheme 74. Attempts at N-alkylation or hydrolysis of cycloadduct 211

Efforts were then focused on placing a substituent on the carbamate nitrogen in order to prevent these epimerizations and ring openings. Again, all of the reactions that were undertaken, including: alkylation, acylation, and silylation, provided similar results as in the desilylation reactions discussed before (Scheme

74). Also, hydrolysis of the BOC carbamate via standard TFA/CH2Cl2 resulted in complete decomposition. Moreover, no reaction was observed if the Ohfune procedure was used (Scheme 74).53

Scheme 75. Attempted cycloaddition of tertiary enecarbamate 214

85

An obvious solution to introduce a substituent on the enecarbamate

nitrogen of cycloadduct 211 was to incorporate the substituent before the

cycloaddition reaction. Thus, N-alkylation of the enecarbamate 210 was

accomplished by deprotonation with NaHMDS followed by addition of methyl

iodide (Scheme 75). Subjection of the resultant tertiary enamide 214 to the

standard cycloaddition reaction with 172 provided only decomposition of enal 172

(Scheme 75). Again, Lewis acids were also utilized in the high-pressure

cycloadditions [Yb(fod)3 and Eu(fod)3] as well as a number of solvents, all

provided similar results of hetero diene 172 decomposition (Scheme 75).

Scheme 76. Preparation of enecarbamates and eneureas

The enecarbamate 215 and urea enamides 216 and 217 were also

prepared (Scheme 76) and used in successful cycloaddition reactions (Table 5).

However, the N-methylated analogue 218 produced only a trace amount of

product in the cycloaddition reaction as seen via 1H NMR.

In conclusion, work on this target has been abandoned, but it is interesting to note the difference in reactivity of the various dienophiles that were investigated in this study. Since the standard enol ethers did not participate in this cycloaddition due to their tendency for decomposition and lower reactivity, it

86

Table 5. Cycloaddition reactions of enecarbamates and urea enamides

was necessary to utilize the highly reactive enamine substrates. These compounds did in fact participate in the cycloaddition at a much increased rate, at room temperature in fact; however, the resultant cycloadducts were unable to be converted back to the desired acetals without full decomposition of the products. In an attempt to find a substrate that was moderate in reactivity to the two substrates above, investigation into the use of enamides as substrates was investigated. To our knowledge this was the first time that these substrates were utilized in the role of a dienophile. The reactivity of these substrates could be moderated by increasing the electron donating capabilities of the “amide”

87

nitrogen with a switch in the carbonyl group on this moiety. Switching to a more nucleophilic carbamate or urea allowed small moderations in the reactivity of these dienophiles. Unfortunately, these cycloadducts were not able to be further converted to the tertiary enamide/enecarbamate/eneurea for investigation into the key ring closing metathesis reaction.

88

Chapter 3

Alternative Approaches to 1,3-Dioxins

3.1 Stereoselective generation of (Z)-2-(trialkylsilyloxy)-2-enals

The previous methodology discussed in Chapter One for the stereoselective synthesis of (Z)-2-acyl-2-enals is not limited to acyl substitution at

C(2). As shown in earlier work, 2-acyloxyacroleins and 2-amidoacroleins can be generated by thermal retrocycloaddition of the appropriately substituted 4H-1,3- dioxin.8,9 Thus, it seemed quite likely that (Z)-2-acyloxy-2-enal 224 and (Z)-2- amido-2-enal 225 could be prepared by retrocycloadditions of the corresponding

4-alkyl substituted dioxins (Scheme 77).

Scheme 77. Potential synthesis of 2-acyloxy and 2-amido-2-enals

The substituted dioxins could be synthesized from the appropriate 4-alkyl-

1,3-dioxanones 32iv and 228. The acyloxy substituent was introduced using the conditions reported by Funk and Yost.8

89

Scheme 78. Stereoselective synthesis of 2-acyloxy and 2-amido-2-enals

Ketone 32 iv was subjected to acetic anhydride in the presence of

triethylamine and afforded both regioisomers of the 1,3-dioxin, the major product being the desired less substituted dioxin 226 (Scheme 78). The analogous amido dioxin 229 was formed by initial formation of the methyl imine, followed by standard N-acylation to provide exclusively the least substituted olefin of dioxin

229 (Scheme 78). Both substrates were then heated to provide the Z- retrocycloaddition products 227 and 230 (Scheme 78). These additional examples of stereoselective retrocycloadditions encouraged us to investigate the preparation and retrocycloadditions of 4-alkyl-5-triaklysilyloxy-4H-1,3-dioxins that would lead to 2-trialkysilyloxy-2-enals, reactants that might participate in stereoselective [4 + 3] cycloaddition reactions.54

90

3.1.1 Application of (Z)-2-(trialkylsilyloxy)-2-enals

in [4 + 3] cycloaddition reactions

The direct construction of seven-membered rings via [4 + 3] cyclization reactions is an attractive strategy for preparing this frequently observed natural product substructure.55 Accordingly, a considerable amount of effort has

Scheme 79. Literature examples of [4 + 3] cycloaddition reactions

91

been directed toward the discovery of three-atom component reactants that will

participate in this type of reaction. Several examples of this reaction are shown

Scheme 79.

56 We were particularly attracted by the method reported by Sasaki that

involves the treatment of 2-(trimethylsilyloxy)acrolein (231) with a Lewis acid in

the presence of a diene to afford, after acidic workup, a 2-hydroxycyclohept-4- enone (232, Scheme 80). Specifically, attempts to expand the scope of this reaction by employing b-substituted enal analogues related to propenal 231 in both inter- and intramolecular [4 + 3] cycloaddition reactions were not reported,57 perhaps due to the inaccessibility of the requisite (Z)-2-(trialkylsilyloxy)enals.

Scheme 80. Sasaki [4 + 3] cycloaddition reaction

A convenient, stereoselective preparation of (Z)-2-(trialkylsilyloxy)-2- alkenals 234 by application of the 1,3-dioxin-based methodology was envisaged.

Thus, based on this previous work it was anticipated that 4-alkyl-5-

(trialkylsilyloxy)-4H-1,3-dioxins 233 would be available by regioselective silylation of the corresponding 4-alkyl-1,3-dioxin-5-ones and would undergo facile retrocycloadditions in refluxing toluene to provide the (Z)-enals 234 (Scheme

81).58 A stereoselective preparation of these enals was considered to be critical for investigating the regio- and stereoselectivity of the subsequent [4 + 3]

92

cycloaddition reactions, thereby providing insight into the mechanism of this

intriguing transformation.

Scheme 81. Proposed use of dioxin retrocycloaddition for the Sasaki reaction

OSiR OSiR3 3 O R H R1 H D 1 R1 OSiR3 R2 O O O O R2 233 234 235

While our work was underway, Harmata reported that 2-(triethylsilyloxy)-

and 2-(triisopropylsilyloxy)acroleins, 237a and 237b respectively, also participate

in [4 +3] cycloaddition reactions and could be prepared from the 2- methoxydioxinone 236 via the more labile 2-methoxy-1,3-dioxin intermediates

(Scheme 82),58 that were generated in situ and underwent concomitant retrocycloadditions.

Scheme 82. Harmata [4 + 3] example

O O OSiR3 OSiR R3SiOTf Sc(OTf)3 3 H O O O TEA, C6H6 furan O OMe 236 237a R = Et, 50% 237b R = i-Pr, 72%

The synthesis of the 5-(silyloxy)dioxins 238 was easily accomplished. As

before, the aza-enolate derivative of the imine 33 underwent efficient alkylation

and upon hydrolytic workup afforded the desired butylated dioxinone 32i. Kinetic

deprotonation of ketone 32i with NaHMDS58 afforded the less-substituted enolate

93

Scheme 83. Synthesis of (Z)-2-(trialkylsilyloxy)-2-enals 240

that could be O-silylated with several trialkylsilyl halides (Scheme 83). As

expected, each of the resulting (trialkylsilyloxy)dioxins 238 was smoothly

converted to only the Z-stereoisomer of the (silyloxy)enals 240 in nearly

quantitative yields in refluxing toluene. The stereoselectivity is presumably a

consequence of preferential retrocycloaddition through the boat like conformer

eq-239 rather than boat like conformer ax-239 that is destabilized by a flagpole-

flagpole interaction between the butyl group and axial lone pair.

Thermodynamically controlled isomerization to (Z)-enals 240 was ruled out by

heating a mixture of 240iii and its E-isomer (83:17), obtained by

photoisomerization of 240iii (Hanovia 500 W, toluene), and observing no change

in the ratio of isomers. Finally, the (Z)-2-(tert-butyldimethylsilyloxy)-2-enals 241 and 242 were also prepared by straightforward application of this protocol (Figure

16).

94

Figure 16. Additional examples of (Z)-2-(trialkylsilyloxy)-2-enals

We were pleased to discover that (Z)-2-(trialkylsilyloxy)alkenals smoothly participated in Lewis acid catalyzed [4 + 3] cyclization reactions with a variety of dienes (Table 6).58 Several observations merit comment. In all cases the silyl group is cleanly transferred to what was the aldehyde oxygen. Moreover, all of

the cycloadducts possess a cis-stereochemical relationship (as determined by

NOESY studies) between the b-alkyl substituent of the enal and the newly

formed silyloxy substituent. In addition, endo- adducts are uniformly preferred over the exo- counterparts with the cyclic dienes furan and .

Moreover, the stereoselectivity is better for smaller silyl substituents (compare entries a vs. b and c vs. d). The endo/exo ratio can also be significantly improved by the choice of the Lewis acid catalyst (entry c). Not surprisingly, acyclic dienes are not as reactive as the cyclic dienes, but acceptable yields were obtained with , isoprene, and trans-. It is of interest to note that the regioselectivity of the cyclization with isoprene is also sensitive to the substituents on the silyl group (entries g and h) and that the acyclic diene

piperylene favors the exo-adduct in a highly regioselective and stereoselective

cyclization (entry i).

95

Table 6. [4 + 3] Cycloaddition reactions of (Z)-2-(trialkylsilyloxy)-2-enals

96

Scheme 84. Intramolecular [4 + 3] cycloaddition reaction

This methodology was not limited, however, to the intermolecular version

of the [4 + 3] cycloaddition. The retrocycloaddition of the appropriately

substituted 5-(trialkylsilyloxy)-1,3-dioxins 243 can be catalyzed by a Lewis acid

with concomitant intramolecular [4 + 3] cyclization of the resulting 2-

(trialkylsilyloxy)-2-enal 244 to afford fused bicyclic adducts 246 of potential value

in natural product synthesis (Scheme 84). The Lewis acid catalyzed

retrocycloaddition was necessary in this case since thermolysis of the dioxin 243i

in refluxing toluene gave rise to product 245 (relative stereochemistry unknown)

derived from an intramolecular Diels-Alder reaction of the intermediate 2-(tert- butyldimethylsilyloxy)-2-enal 244i.

In an attempt to better understand the mechanism and reactive species of this reaction, a crossover experiment was investigated that suggests that the silyl group is transferred both intra- and intermolecularly. Thus, when the cyclization shown in entry b was performed in the presence of equivalent amounts of enals

240iii and 242, the endo- adduct 247a (as in entry b) was obtained accompanied by the endo- adduct 247b (as in entry a) in a ratio of 86 : 14, respectively. 97

Similarly, the endo- cyclization product 248b derived from enal 242 (TBS ether) was accompanied by the corresponding endo-adduct 248a in a similar ratio of

86:14, respectively.

Scheme 85. Investigation into silyl transfer step

The mechanistic details of this type of [4 + 3] cycloaddition are still of some debate. Recent literature precedent from Davies showed that the mechanism of a similar [4 + 3] reaction with cyclic and acyclic enals, proceeds via an initial Diels-Alder cycloaddition reaction at low temperature (-78 °C) followed by subsequent ring expansion in the presence of the stoichiometric

Lewis acid at 0 °C. (Scheme 86).59a Davies and coworkers were able to isolate the [4 + 2] cycloadducts by quenching the reaction at -78 °C, but upon warming to 0 °C in the presence of stoichiometric aluminum chloride a ring expansion occurred to provide the pentultament seven-membered ring. However, when trying to repeat the Harmata example and quenching at low temperature to isolate the [4 + 2] adduct 249, the Davies group was only able to isolate the seven-membered ring products 250. They did discover that if the reaction was run under microwave conditions (instead of Lewis acid catalysis) that they were in fact able to obtain the Diels-Alder cycloadduct 249. This product was then subjected to Harmata’s conditions [Sc(OTf)3, 0 °C-rt] and a facile ring expansion

98

to the desired [4 + 3] product 250 was observed. Thus, this opens up the

possibility that this reaction may in fact occur via a tandem Diels-Alder/ring-

expansion mechanism.

Scheme 86. Mechanistic studies of the Sasaki [4 + 3] cycloaddition reaction

An additional study by Domingo and coworkers, also recently published,

investigated the mechanism of the “Sasaki” type [4 + 3] cycloaddition via a computational means.59b This work provides evidence that this reaction actually proceeds in a stepwise fashion via the intermediate zwitterion 251 as a result of

1,4-nucleophilic attack of furan onto the acrolein moiety. This process was calculated to have a relatively low barrier of activation (DE = 3.1 kcal/mol) in the

first transition state (TS 1) and formation of the zwitterion 251 is a slightly

99

exothermic reaction. Polarization of the HOMO for the 2-(silyloxy)acrolein favors

nucleophilic attack from the carbonyl carbon onto the C(5) position of the furan

subunit, thus providing the formal [4 + 3] cycloaddition products preferentially

over the [4 + 2] products. This process is again calculated to have a low barrier

of activation in the transition state (TS 2, DE = 1.6 kcal/mol) and the resultant

intermediate 252 is the consequence of an exothermic process (DE = -2.2

kcal/mol). The final step in this process, transfer of the silyl group from the

oxonium ion of intermediate 252 to the oxygen on the a-carbon, is actually

calculated to be the rate determining step with a transition state energy barrier

(TS 3) of 9.2 kcal/mol. This computational analysis seems to fit the observations

that were realized in our studies as well, since larger silyl groups and Lewis acids

tended to slow the reaction sequence down significantly.

In conclusion, we have demonstrated that 2-(trialkylsilyloxy)-2-enals constitute another class of useful reactants that can be generated with high stereoselectivity by retrocycloaddition reactions of substituted 4H-1,3-dioxins.

The [4 + 3] cyclizations of these enals documented herein substantially expand the scope and limitations of the original Sasaki account. Although the precise structure of the allyl cation intermediate generated in these reactions requires further clarification, the ability to control the regio- and stereoselectivity of these cyclization reactions by choice of the silyl substituents as well as the Lewis acid makes the 2-(trialkylsilyloxy)-2-enals particularly attractive among the various three-atom components for [4 + 3] cycloadditions.

100

3.2 Deconjugation route to form 1,3-dioxins

It has been found that 5-alkyl-1,3-dioxins can be readily prepared by deconjugation of conjugated 5-alkylidiene-1,3-dioxanes. Utilization of this methodology has been exemplified in the preparation of the first thermally labile solid-phase linker. Similarly, the deconjugation route has also been exploited for preparation of the labdane/clerodane natural products ring systems.

3.2.1 Approaches to the first thermally labile

solid-phase linker

Solid-phase chemistry has become a useful tool in the development of libraries of structurally diverse small molecules. These libraries can be assembled quickly to generate products for biological testing, providing rapid access for lead discovery and development in the pharmaceutical industry.60

Liberation of resin bound compounds following a solid-phase synthesis can be achieved through activation of the linker. The linker is the means by which the starting materials are attached to the resin and the products are subsequently released (Figure 17). Several strategies have been developed for cleaving the linker and range from acid labile linkers (i.e. Rink resin),61 to nucleophilically cleaved linkers (i.e. Kaiser oxime resin),62 or even photochemically labile linkers (i.e. Nb resin),63 each having its own pros and cons

(Figure 17).

101

Figure 17. Known and proposed linker release mechanisms

Herein, we report the development of a thermally labile linker allowing for release

of acid and base sensitive compounds from a resin under purely thermal

conditions (Figure 17).

It was envisaged that a thermally labile linker could be designed based upon

previous studies in this laboratory concerned with the thermally promoted retro

hetero Diels-Alder reactions of 1,3-dioxins.7-14,58 Attachment of a 1,3-dioxin to a polymeric resin at C(5) would provide a linker to which a number of substrates could be attached to a C(2) hydroxyethyl substituent for use in solid-phase synthesis. Thermolysis of resin 254 would effect a retro cycloaddition to afford the substituted enal 255, still attached to the resin, and the b-alkoxy aldehyde

256 that would undergo concomitant retro Michael addition to liberate 258 and acrolein (257) as a by-product (Scheme 87).

102

Scheme 87. Proposed thermally labile linker

The preparation of this thermally labile linker took advantage of

methodology developed for the synthesis of 1,3-dioxins via deconjugation

reactions of 5-alkylidene-1,3-dioxanes. Accordingly, synthesis of the desired

conjugated system was accomplished by initial formation of the cyclic acetal 260

through condensation of acrolein, benzyl alcohol and tris-

(hydroxymethyl)nitromethane (259) (Scheme 88).15 Upon reduction of the nitro moiety to the vicinal amino-alcohol 260 and oxidative cleavage with sodium

Scheme 88. Synthesis of dioxin 263

periodate, the desired 1,3-dioxanone 261 was obtained (Scheme 88).15 Horner-

Wadsworth-Emmons reaction provided the desired a,b-unsaturated ester 262,

which was deconjugated with catalytic DBU to afford dioxin 263 (Scheme 88). 103

The remaining two obstacles, deprotection of the primary alcohol and attachment of the substrate to the bead, were next investigated. All attempts at reductive removal of the benzyl group via hydrogenation resulted in reduction of the olefin as well. However, saponification to the acid followed by Birch reduction64 did in fact provide the deprotected hydroxy acid 264 (Scheme 89).

Scheme 89. Resin attachment of the first generation linker

Activation of the acid as the trichlorophenolic ester 265 followed by acylation of the amino functionality on the ArgoGel resin (Aldrich) afforded the resin bound linker 266 (Scheme 89).65

To test the utility of the new linker, stearic acid was coupled with the alcohol portion of 266 (Scheme 90). Subjection of the resin to refluxing toluene did indeed provide the retrocycloaddition products: resin bound enal 267 and aldehyde 268 (Scheme 90). Further heating of the retrocycloadduct 268 allowed

104

liberation of stearic acid and acrolein through retro Michael reaction (65% yield

for attachment of linker and stearic acid followed by release from the beads,

Scheme 90). Unfortunately, other substrates (ie. silyl protected lactic acid and 4-

(silyloxymethyl)benzoic acid) that were investigated for attachment to the linker

were not as successful, resulting in no discernable products upon thermolysis.

Also, attachment of linker 265 onto the resin was not as successful as desired,

leading to the need for an alternative approach for attachment of the linker to the

polystyrene beads.

Scheme 90. Attachment and release of stearic acid

The second-generation approach to this linker involved attachment of the

linker onto the bead via a direct alkylation reaction, hopefully a more efficient

approach than the alternative amide bond formation. We had previously found

that the allylic sulfone 269 could be cleanly metalated with n-butyllithium and

then alkylated with several alkyl iodides (Scheme 91). Thus, it was believed that

this methodology could be applied to incorporation of the dioxin into the appropriately functionalized resin.

105

Scheme 91. Utility of allylic sulfone 269

A synthetic sequence analogous to the one used to construct dioxin 263 was employed for the preparation of our new target dioxin 274. The b- hydroxypropanal 271 was converted to an acetal that was reduced to amino alcohol 272 (Scheme 92). Oxidative cleavage of the amino alcohol functionality

Scheme 92. Preparation and attachment of the second generation linker

with sodium periodate then provided ketone 273 (Scheme 92). Horner-

Wadsworth-Emmons reaction under Masamune-Roush conditions provided

106

dioxin 274 in a single step (Scheme 92).66 Alkylation of the metalated sulfone

274 with iodomethylated polystyrene afforded the desired resin. Finally, the sulfone functionality of resin 275 was reductively removed67 and the alcohol deprotected to provide the resin bound linker 276 (Scheme 92). As before, stearic acid was attached to the resin through DCC coupling, followed by thermal decomposition of the linker to release stearic acid (99% yield for the five step process of attachment, reduction, deprotection, esterification, and release).

Although resin attachment was successful, there were still problems of attaching other substrates to resin bound linker 276. Although it was possible to attach benzoic acid and release it from the resin (79% overall yield), attachment of other acids (such as those mentioned previously) were not as fruitful. This may have been due to the close proximity of the polystyrene bead to the primary alcohol moiety of 276. A final generation of this linker should allow for further spacing between the two moieties, as well as reduce the number of steps necessary to access the linker upon its attachment to the resin. Thus, our final approach focused on an ether linkage, which was hoped would provide the best solution to both of these concerns.

The previously outlined strategies were combined to afford the desired linker. To that end, ketone 273 was subjected to a Horner-Wadsworth-Emmons reaction to provide the a,b-unsaturated ester 277 (Scheme 93). Deconjugation followed by reduction to the resultant homoallylic alcohol 278 provided the desired linker for attachment to the resin (Scheme 93). The linker was attached to the resin in a straightforward fashion via a Williamson ether synthesis68 using

107

Scheme 93. Preparation and attachment of the third generation linker

chloromethyl polystyrene. Subsequent deprotection of the silyl ether provided the resin bound alcohol 279 (Scheme 93).

Attachment and release of stearic acid was again easily accomplished

(72% overall yield). Moreover, it was now possible to effect a DCC coupling of the substituted benzoic acid 280 to the resin followed by deprotection of the silyl ether and another DCC mediated coupling to provide the desired substrate 281

Scheme 94. Solid and solution phase synthesis of a diester

(Scheme 94). A solution phase synthesis was run in parallel (utilizing benzyl chloride as the resin surrogate) with the solid-phase synthesis to ensure

108

Scheme 95. Thermal release of acid 285

optimal conditions for linker attachment and diester formation were achieved

before proceeding with the solid-phase approach (Scheme 94).

With the polymer bound substrate 281 in hand, optimal conditions for the thermal release of the substrate through tandem retro hetero Diels-Alder/ retro

Michael addition reactions were studied. Initial attempts to induce the tandem reaction at 110 °C, the temperature at which most of the 1,3-dioxin retrocycloadditions have been previously performed, did release the diester 284 from the resin (Scheme 95). However, the retro Michael addition was very slow at this temperature and took several days to complete. It was found that thermolysis of the neat beads at 135 °C for four hours afforded the pure acid 285 in 48% yield, based upon initial loading of the Merrifield resin at 1.83 mmol/g

(Scheme 95).

In conclusion, a thermally labile linker has been developed allowing for substrate release from resins under purely thermal conditions. Use of an ether linkage for attachment of this linker to the resin provides a stable functional group that does not decompose under the conditions required for its release. This linkage also allows for efficient attachment of carboxylic acids. Finally, the thermal liberation of the ester product is noteworthy since it would not be

109

possible to release these compounds using the standard acidic or nucleophilic conditions required for cleavage from other linkers.

3.2.2 An approach to the labdane and

clerodane natural product ring systems

The labdane and clerodane diterpenes are two structurally similar classes of natural products that exhibit a wide variety of biological properties.69 The more interesting members of these classes of compounds (and challenging from a total synthesis perspective) invariably possess an angular carbon substituent in a higher oxidation state [C(10) for labdanes and C(5) for clerodanes].70 Thus, two recently isolated labdanes potamogetonyde71a and cacofuran A,71b show antiviral

(Herpes simplex virus type 1, IC50 = 8 ?g/mL; acyclovir = 2-5 ?g/ mL)

Figure 18. Examples of labdane/clerodane natural products

and cytotoxic (P388/K562, IC50 = 1 ?g/mL) activities, respectively, and tanabalin72 is an insect antifeedant clerodane (Figure 18). While the major source for these natural products has been from terrestrial plants, the recent isolation of cacofuran B from a sponge71b signifies that marine organisms have recognized the pharmacological value of this ring system as well.

110

Scheme 97. Retrosynthetic analysis of potamogetonyde

Herein a strategy is presented for the construction of the trans-decalin ring system 286 of the labdane molecules, while simultaneously introducing a formyl substituent at one of the bridgehead carbons (Scheme 97). This strategy involves an endo-intramolecular Diels-Alder reaction of 2-substituted enal 287, derived through retrocycloaddition of 1,3-dioxin 288 (Scheme 97). In turn, dioxin

288 could be formed utilizing the deconjugation method developed in the solid phase linker project to isomerize the exocyclic olefin of dioxane 289 into the dioxane ring (Scheme 97).

Thus, the initial focus was the preparation of the b-keto phosphonate 295 necessary for the synthesis of dioxane 289. 4-Pentyn-1-ol (290) was protected and converted to propargyl alcohol 291 under standard conditions (Scheme

98).73 Iodide 292 was prepared via a regioselective, directed hydrostannylation.74

Iodination of the resulting vinyl stannane and protection of the primary alcohol

(Scheme 98) as a triisopropylsilyl ether. The diene 293 was prepared by Stille 111

coupling with tributyl(vinyl)tin,13 followed by selective deprotection of the

triethylsilyl ether and oxidation of the resultant primary alcohol provided aldehyde

294 (Scheme 98). Finally, addition of a phosphonate anion and subsequent

oxidation afforded the desired b-keto phosphonate 278 (Scheme 98).

Scheme 98. Synthesis of b-ketophosphonate 295

The synthesis of the retrocycloaddition/cycloaddition substrate could now

be addressed. Condensation of b-keto phosphonate 295 with 2,2-dimethyl-1,3- dioxan-5-one (21) followed by treatment with DBU provided the deconjugated ketone 296 (Scheme 99). Initial cycloaddition studies were carried out on this simplified dioxin 296 (Scheme 99) that lacks the furanyl substituent present in dioxin 288. Thus, thermolysis of 296 provided the retrocycloaddition product within three hours, which upon further heating overnight cyclized through the endo-transition state 297 to provide only the trans-decalin ring system 298.

112

Scheme 99. Synthesis of the labdane trans-decalin ring system 298

The stereochemistry was established via nOe experiments, the most

diagnostic of which are indicated in structure 298 (Scheme 99). To the best of

our knowledge, this represents the first example of an intramolecular

cycloaddition reaction of a 2-methylene-7,9-decadienal. Molecular mechanics

(PC Model) calculations suggested that the endo-297 transition state would be highly favored over the exo-297 counterpart, due in large part to a destabilizing interaction between the silyloxymethyl carbon and the C(5) axial hydrogen in the exo-297 transition state.

With this reassuring result in hand, we turned our attention to the cycloaddition of a trienal that possessed a substituent at the allylic a-keto carbon in order to further establish the preferred endo- transition state. We were pleased to find that deprotonation of the ketone 296 was completely regioselective and afforded a dienolate, which was methylated to provide dioxin 299 (Scheme 100).

Thermolysis of dioxin 299 effected a remarkably clean transformation to a single product, aldehyde 301. Again, nOe experiments were used to establish the

113

stereoselectivity of the cycloaddition, in this case proceeding via the endo- transition state 300 possessing the methyl substituent in the equatorial position. It seems likely that the retrocycloaddition/cycloaddition of dioxin 288 will proceed with similar stereoselectivity and that this methodology is useful or the synthesis of labdanes/clerodanes possessing oxygenated C(10)/C(5) substituents.

Scheme 100. Synthesis of the 9-alkyl trans-decalin ring system 301

114

3.3 Diels-Alder approach to synthesize 1,3-dioxins:

Application in the Illudin C synthesis

A final strategy that was envisioned for the preparation of 1,3-dioxins was through Diels-Alder reactions (Scheme 101). Cycloaddition of a 4,5-dialkylidene-

1,3-dioxane 302 with a dienophile should provide the desired dioxin 303. Further heating of the cycloadduct 303 should provide the retrocycloaddition product enone 304. This type of transformation was projected to be a key step in the synthesis of the biologically active illudin compounds.

Scheme 101. Dioxin synthesis via Diels-Alder reaction

3.3.1 Biological importance of the illudin compounds

The illudins are an intriguing class of sesquiterpenes isolated from several

fungi.75 Their varied substitution patterns are suggestive of nature’s attempts to

optimize their cytotoxic properties. These natural products can be structurally

categorized according to the degree and position of unsaturation in the unusual

tricyclic ring system and the site of tertiary hydroxyl substituents. Three different

types of illudins result from this analysis (Figure 19). Thus, illudin M/S embody a

115

conjugated C(4)-C(5)-C(9)-C(8) diene and C(2) tertiary hydroxyl, whereas an unconjugated C(5)-C(9), C(2)-C(10) diene and C(4) tertiary hydroxyl are found in the illudin C/C2/C3 and illudinic acid group. Illudin A/B share features common to both of the aforementioned groups, namely, C(2) and C(4) tertiary hydroxyl substituents and C(5), C(9) sp2 –hybridized carbon atoms.

Figure 19. The illudin class of natural products

The cytotoxicity and anticancer activity of illudin S has been most extensively investigated.76 The target of the compound is believed to be DNA, although an illudin-DNA adduct has yet to be isolated since illudin S does not spontaneously bind to DNA. It has been proposed that illudin S is bioactivated, perhaps by conjugate addition of an endogenous nucleophile

(glutathione/NADH)79b to the enone moiety (Scheme 102). The resultant labile cyclohexadienol 305 then undergoes facile nucleophilic attack by DNA (or other

116

Scheme 102. Mechanism of action of illudin S

biological targets) on the cyclopropane ring to produce a stable aromatic

compound. The low therapeutic index of illudin S has precluded its development

as a chemotherapeutic agent. However, the semisynthetic illudin analogue, 6-

(hydroxymethyl)acylfulvene (HMAF),77 shows outstanding activity against breast, colon, lung, pancreas, prostate, and skin cancers and is now in various Phase I,

II, and III clinical trials.78

Illudin C,78c C2,78d and C378d and illudinic acid78c have only recently been isolated, and the latter three have been shown to possess antimicrobial activity against methicillin-resistant Staphylococcus aureus (MRSA).78d,e The anticancer

activity of these compounds has not been reported, although cytotoxicity in a

mammalian cell culture system has been demonstrated.78e Although the total

synthesis of the illudin S/M type of natural products has been extensively 117

investigated,79 no synthetic entry to the illudin C variant is presently available. In view of the successful analogue breakthrough in the illudin S/M series, an efficient synthetic entry to the illudin C type of natural products is highly desirable in order to more fully investigate the cytotoxic properties of these compounds and congeners.

3.3.2 Retrosynthetic analysis for a Diels-Alder

approach to the illudin C ring system

The initial approach to illudin C involved the use of 1,3-dioxin chemistry to install the a,b-unsaturated carbonyl functionality. It was envisaged that the tertiary hydroxyl group could be installed via oxidation of the tetra-substituted

Scheme 103. Retrosynthetic analysis of illudin C

olefin in enone 308, followed by ring opening of epoxide 307 and elimination to provide illudin C (Scheme 103). In turn, the enone functionality would be

118

established through the aforementioned retrocycloaddition of dioxin 309 (Scheme

103). It was believed that dioxin 309 could be formed through an intramolecular cycloaddition reaction of diene 310 (Scheme 103).

3.3.3 Synthesis and application of the Diels-Alder

substrate for the illudin C model system

The initial studies were directed toward a model system that lacked the geminal dimethyl group found in the natural product. Synthesis of the key cycloaddition substrate for this model study was accomplished by utilizing a

Mukiyama aldol reaction as shown in Scheme 14.18 Thus, reaction of silylenol ether 34 with aldehyde 312 (available by PCC oxidation of 5-heptyn-1-ol) provided a 1 : 1 mixture of diastereomers (Scheme 104). Dehydration of the b- hydroxy ketones 313 via formation and elimination of a mesylate intermediate provided the desired enone (Scheme 104). A variety of procedures were studied

Scheme 104. Preparation of cyclopropylidiene 314

for formation of cyclopropylidiene 314, including a number of Wittig reactions,

Julia reactions as well as Peterson olefinations. However, the only procedure that

119

provided any of the desired diene was through in situ formation of the cyclopropyl

ylide with potassium tert-butoxide, albeit in low yields (Scheme 104).80

A study of the cycloaddition/retrocycloaddition reaction was then initiated

on dioxane 314. Subjection of the substrate to refluxing toluene only resulted in

recovery of starting material (Scheme 105). Unfortunately all other attempts at

higher temperatures resulted in decomposition of the starting material (Scheme

105). Similarly, use of Lewis acids also caused decomposition.

Scheme 105. Attempted Diels-Alder reaction to provide dioxin 316

The failure of this particular Diels-Alder reaction sequence was not unexpected as there is limited precedent for cycloadditions of cyclopropylidienes.81 At the time of this research, de Meijere had reported in two papers the limitations of these dienes for use in cycloaddition reations.81a,b His research showed that activation of the cyclopropylidiene as a silylenol ether as well as utilization of highly activated dienophiles did provide access to the desired cycloadducts.81a Later reports from this group also explored the utility of using palladium as a source of activation to provide the desired cycloadducts using an unactivated cyclopropylidiene and activated dienophile.81b,c However, there was only one example of utilizing an as the dienophile (dimethyl

120

acetylene dicarboxylate)81c and no reports of an intramolecular cycloaddition with

these substrates.

3.3.4 Retrosynthetic analysis for an intramolecular nitrile-oxide

cycloaddition approach to the illudin C ring system

During investigation of the previously discussed dioxin-based approach to

the illudins we envisaged an alternative strategy for the synthesis of these natural

products that features a nitrile-oxide cycloaddition. The new retrosynthetic analysis for the synthesis of illudin C is outlined in scheme 106.82 On the basis of the pioneering investigations of Curran and Kozikowski,83 it was envisaged that the a-methylene ketone functionality of illudin C could be introduced via dehydration of the corresponding b-hydroxy ketone, in turn available from the hydrogenolysis-hydrolysis reaction of isoxazoline (317, Scheme 106). The conformationally restricted alkenyl nitrile oxide 318 was anticipated to be an excellent substrate for an intramolecular nitrile-oxide cycloaddition to furnish isoxazoline 317 and could be generated by the standard protocol, namely, oxidation of oxime 319 (Scheme 106). While the obvious route to oxime 319 was from the corresponding aldehyde, this would have necessitated prior protection of the aldehyde moiety in order to affect a planned vinyl anion coupling with cyclopropyl ketone 321 (Scheme 106). Instead, we were intrigued by the possibility of employing dianion 320, prepared by metalation of the corresponding

121

Scheme 106. Second generation retrosynthetic analysis of illudin C

b-halo unsaturated oxime, in a more direct synthesis of oxime 319 (Scheme

106).82 Little in the way of precedent for this transformation could be found in the literature, although the preparation and alkylation of the dianion derivatives of saturated ketoximes (a-Li, O-Li)84 offered some encouragement to pursue this especially convergent approach.

3.3.5 Synthesis and application of the nitrile-oxide cycloaddition

substrate for the total synthesis of illudin C

To that end, we considered the preparation of cyclopentenecarboxaldehyde 323 using a Vilsmeier-Haack procedure (Scheme

107). Several examples in the literature suggested that this compound could be obtained from 3,3-dimethylcyclopentan-1-one (322) by regioselective attack of the bromo(dimethylamino)methyl cation on the less encumbered enol (Scheme

107).85 However, a nearly equal mixture of aldehyde 323 and the regioisomeric product 2-bromo-5,5-dimethylcyclopent-1-enecarboxaldehyde (324), (1.1 : 1)

122

Scheme 107. Vilsmeier-Haack formylations

86 were obtained using PBr3/DMF in methylene chloride (Scheme 107).

Consequently, we turned to the known silylenol ethers 325,87 but unfortunately treatment with PBr3/DMF afforded the same mixture of regioisomers, presumably through initial hydrolysis of the silylenol ether (Scheme 107). However, it was found that when the triethylsilylenol ether 325 was subjected to the Vilsmeier

Scheme 108. Optimized Vilsmeier-Haack formylation

reagent prepared from POBr3 and DMF in the less polar solvent methylene chloride, aldehyde 323 was cleanly obtained in satisfactory yields and little contamination of the regioisomeric product (Scheme 108).88 To the best of our knowledge, this is the first example of a regiospecific Vilsmeier-Haack haloformylation of a silylenol ether.82 Aldehyde 323 was then converted to the E- oxime 326 using the standard protocol in 82% yield (Scheme 108).

123

Scheme 109. Synthesis of ketone 321

The preparation of the electrophilic component for the pending coupling reaction, ketone 321 (Scheme 109), was more straightforward and was initiated by bis-alkylation of 2,4-pentanedione with 1,2-dibromoethane to provide the known cyclopropane 327 (Scheme 109).89 Monoreduction of diketone 327 to ketoalcohol 328 could be accomplished with Li(O-t-Bu)3AlH (58%) and was accompanied by only minor amounts of the corresponding diol (Scheme 109).

Although dehydration of the alcohol 328 to olefin 321 took place upon treatment with the Burgess reagent, the transformation was best performed by initial conversion of alcohol 328 to the unstable iodide (82%) followed by distillation from neat DBU to afford the volatile ketone 321 in 58% yield (Scheme 109).

We were now in the position to examine the two-step merger of the cyclopropane and cyclopentene sectors of illudin C (Scheme 110).82 Indeed, the dianion 320 could be prepared by addition of t-BuLi (3 equiv) at –78 °C to bromo oxime 326

(Scheme 110). If the internal temperature was allowed to rise above –78 °C during the subsequent addition of cyclopropyl ketone 321, only deprotonation of the ketone occurred, forming enolate 330 and a,b-unsaturated oxime 329

(Scheme 110). Circumstantial evidence for the generation of the enolate 330 was based on the isolation of the aldol product 331 (Scheme 110). To minimize the unwanted proton transfer, cyclopropyl ketone 321 was diluted with THF and

124

added slowly via syringe pump (1 h) to dianion 320 affording the desired oxime

319 in 68% yield (Scheme 111).

Scheme 110. Aldol by-product 331 formation

We were pleased to discover that the final bond of the tricyclic system

formed readily upon treatment of oxime 319 with chloramine-T (EtOH, 40 °C, 6 h)90 to deliver the isoxazoline 317 as a single diastereomer (99%) (Scheme 112).

The cis-stereochemical relationship of the C(4) methyl substituent and C(2) hydrogen atom in the resultant tetracycle was assigned using nOe experiments

The diastereoselectivity is a consequence of preferential cycloaddition of the

nitrile oxide through transition state conformer 318 wherein the smaller C(4)

Scheme 111. Synthesis of nitrile-oxide cycloaddition substrate

hydroxyl substituent, vis-à-vis the C(4) methyl substituent, occupies an equatorial

position bisecting the cyclopropane ring (Scheme 112).

125

Scheme 112. [3 + 2] cycloaddition of nitrile-oxide 318

Transformation of the nitrile oxide-olefin cycloadduct 317 to illudin C was

uneventful (Scheme 113).82 Thus, Raney-Ni catalyzed hydrogenation of the N-O

bond followed by in situ hydrolysis of the resulting b-hydroxy imine according to

91 the Curran protocol (H2O, B(OH)3) furnished the formal aldol adduct 332 in 72%

yield (Scheme 113). The enone of illudin C was initially furnished by mesylate

formation (MsCl, Et3N) and subsequent elimination with DBU (Scheme 113). A

one-pot dehydration was possible by standard conversion of the diol 332 to the

mesylate (MsCl, Et3N) followed by addition of DBU to provide racemic illudin C

(73%) identical in all respects with published spectra (Scheme 113).

Scheme 113. Completion of the total synthesis of illudin C

O N HO O O Ra-Ni, MsCl, NEt3 H , B(OH) ° 2 3 CH2Cl2, –78 C;

MeOH/H2O DBU, rt 72% 73% OH OH OH 317 332 Illudin C

In summary, we have completed the first total synthesis of illudin C in 10

steps and 8.2% overall yield starting from allyl alcohol and isobutyraldehyde.87 In

the course of the synthesis we have demonstrated that silylenol ethers can be

used in regiospecific Vilsmeier-Haack-type reactions, that b-halo unsaturated

126

oximes permit halogen metal exchange, and that an intramolecular nitrile oxide-

olefin cycloaddition facilitates an exceptionally concise route to the illudin C

tricyclic ring system.81

3.4 Conclusions

In conclusion, methodology for the stereocontrolled synthesis of (Z)-2- acylenals has been developed via retrocycloaddition reactions of 5-acyl-4-alkyl-

4H-1,3-dioxins.14 This methodology has provided concise synthetic routes to several natural products containing the 5-acyl-2H-3,4-dihydropyran substructure.

The first application of the methodology was utilized in a short stereoselective synthesis of the mild immunosuppressant loganetin. Application of this methodology has also been showcased in the first total synthesis of the cytotoxin euplotin A, which was achieved in 12 steps from furan and in 3% overall yield.14

Finally, the limitations of this methodology were illuminated in both the intramolecular and intermolecular cycloadditions directed toward the synthesis of the xenicin class of natural products.

The methodology previously described was not limited to the preparation of enals possessing 2-acyl substitution, but also could be easily adapted to the preparation of 2-(trialkylsilyloxy)-2-enals.58 These reactive intermediates were useful as allyl cation precursors for a number of [4 + 3] cycloaddition reactions, which were highly diastereoselective.

127

New approaches for the construction of 1,3-dioxins were also discovered.

Thus, it was found that these retrocyloaddition precursors could be formed by

double bond isomerization of the appropriately substituted 4-alkylidene-1,3- dioxanes. This methodology has been utilized for the synthesis of a number of newly developed thermally labile linkers for solid phase organic synthesis. Our approach to the labdane/clerodane ring system has also exploited this strategy.

Finally, an alternative approach to 1,3-dioxins that featured a Diels-Alder reaction of bisalkylidene dioxanes was unsuccessful. However, this investigation did stimulate the discovery of an alternative approach to illudin C. Thus, the first total synthesis of illudin C, which exploited a nitrile-oxide cycloaddition as the key step,81 was accomplished in 10 steps with an overall yield of 8% starting from allyl alcohol and isobutyraldehyde.

128

Experimental Section

General Methods: 1H NMR spectra were recorded on Bruker ACE-200 MHz,

Bruker DPX-300 MHz, Bruker AMX-360 MHz, or Bruker DRX-400 MHz

spectrometers. All NOESY and COSY experiments were performed on a Bruker

DRX-400 MHz spectrometer. 1H NMR spectra were recorded as d values relative

13 to the CDCl3 peak at 7.26 ppm. C NMR were recorded on Bruker ACE-200

MHz, Bruker DPX-300 MHz, Bruker AMX-360 MHz, or Bruker DRX-400 MHz

13 spectrometers. C NMR shifts are referenced to the CDCl3 peak at 77.0 ppm.

High and low resolution mass spectra were provided by the Mass Spectrometry

Facility at the Pennsylvania State University, on a Kratos MS 9/50 mass spectrometer. IR spectra were recorded on a Perkin-Elmer 1600 FT-IR spectrometer. All melting points were determined in open Pyrex capillary tubes on a Thomas Hoover melting point apparatus and are uncorrected. Thin layer chromatography was done on Sigma-Aldrich aluminum backed silica-gel plates.

Flash chromatography was performed using IBN 60 silica-gel (230-400 mesh) unless otherwise indicated. All solvents were dried by distillation over the appropriate drying agent and under an Ar atmosphere. Also, all reactions requiring oxygen free and dry conditions were performed under an Ar atmosphere with flamed dried glassware.

129

4-butyl-2,2-dimethyl-1,3-dioxan-5-one (32i). To a solution of 2,2-dimethyl-1,3- dioxan-5-one (21)92 (5.00 g, 38.4 mmol) in benzene (125 mL) was added cyclohexylamine (8.80 mL, 76.8 mmol) and molecular sieves (4Å, 5.00 g). The mixture was stirred overnight at rt, filtered, and concentrated to yield the crude imine 33 (8.20 g). The imine was dissolved in THF (38 mL) and added dropwise at –78 °C to a solution of lithium diethylamide [formed by addition of n- butyllithium (2.5 M in hexane, 14.9 mL, 37.3 mmol) to a solution of diethylamine

(4.00 mL, 38.8 mmol) in THF (38 mL) at –35 °C]. The mixture was warmed to –

35 °C over 2 h and recooled to –78 °C. Iodobutane (3.44 mL, 29.9 mmol) was

added and the solution was warmed to rt over 2 h. A solution of saturated

aqueous NH4Cl (60 mL) was added and the resultant mixture was stirred at rt

overnight. The aqueous layer was extracted with Et2O and the combined

extracts were washed with brine, dried (Na2SO4) and concentrated. Purification

of the crude residue by silica-gel chromatography (ethyl acetate-hexane, 5 : 95)

1 gave a colorless oil (4.63 g, 83%): H NMR (200 MHz, CDCl3) d 0.83 (t, J = 7.0

Hz, 3 H), 1.15-1.50 (m, 5 H), 1.34 (s, 3 H), 1.37 (s, 3 H), 1.76 (dtd, J = 5.2, 8.8,

10.2 Hz, 1 H), 3.88 (d, J = 17.0 Hz, 1 H), 4.08-4.15 (m, 1 H), 4.16 (dd, J = 1.4,

13 17.0 Hz, 1 H); C NMR (50 MHz, CDCl3) d 13.8, 22.3, 23.7, 23.8, 27.1, 28.0,

65.4, 74.5, 100.5, 209.6; IR (neat) 2957, 1748 cm-1; HRMS (M-H+) calcd for

C10H17O3 185.1178, found 185.1174.

130

2,2-dimethyl-4-[2-(triethylsilyloxy)ethyl]-1,3-dioxan-5-one (32iv). To a solution of 2,2-dimethyl-1,3-dioxan-5-one (21)15 (5.70 g, 43.6 mmol) in benzene

(215 mL) were added molecular sieves (4Å, 6.00 g) and cyclohexylamine (10 mL,

87.0 mmol). The solution was stirred at rt overnight, filtered, and concentrated to

yield the crude cyclohexylimine 33 (8.23 g). The imine was dissolved in THF (39

mL) and added dropwise to a solution of lithium diethylamide [formed by addition

of n-BuLi (2.5 M in hexane, 14.6 mL, 36.5 mmol) to a solution of diethylamine

(4.00 mL, 39.0 mmol) in THF (39 mL) at –35 °C] at –78 °C. The solution was

warmed to –35 °C over 2 h, recooled to –78 °C and a solution of 1-iodo-2-

(triethylsilyloxy)ethane (6.97 g, 24.4 mmol) was added. The resultant mixture

was warmed to rt over 2 h and quenched with saturated aqueous NH4Cl. This

mixture was stirred at rt overnight and extracted with Et2O. The combined

extracts were washed with brine, dried (Na2SO4) and concentrated. Purification

of the crude residue by silica-gel chromatography (ethyl acetate-hexane, 1 : 32)

1 gave a colorless oil (6.12 g, 87%); H NMR (200 MHz, CDCl3) d 0.58 (qd, J = 2.3,

7.4 Hz, 6 H), 0.95 (t, J = 7.4 Hz, 9 H), 1.42 (s, 3 H) 1.45 (s, 3 H), 1.68 (ddt, J =

4.8, 8.8, 9.5 Hz, 1 H), 2.10 (dddd, J = 1.3, 3.9, 4.8, 9.5 Hz, 1 H), 3.70 (dd, J =

1.3, 4.8 Hz, 1 H), 3.74 (d, J = 4.8 Hz, 1 H), 3.98 (d, J = 16.9 Hz, 1 H), 4.28 (dd, J

= 1.4, 16.9 Hz, 1 H), 4.49 (ddd, J = 1.4, 3.9, 8.8 Hz, 1 H); 13C NMR (50 MHz,

CDCl3) d 4.4, 6.8, 23.5, 24.1, 31.8, 57.9, 66.6, 71.3, 100.7, 209.9; IR (neat) 2935,

-1 + 1749 cm ; HRMS (M+Na ) calcd for C14H28O4SNa 311.1655, found 311.1651. 131

2,2-dimethyl-4-[2-(triethylsilyloxy)ethyl]-4H-[1,3]dioxin-5-

trifluoromethanesulfonate (36iv). A solution of ketone 32iv (1.00 g, 3.47

mmol) and N-phenyltrifluoromethanesulfonimide (1.61 g, 4.51 mmol) in THF (17

mL) was added dropwise over 1 h to a solution of NaHMDS (1.0 M in THF, 4.51

mL, 4.51 mmol) in THF (15 mL) at –78 °C. The mixture was warmed slowly over

2 h to rt and poured onto saturated aqueous NaHCO3. The aqueous layer was

extracted with hexane and the combined extracts were washed with saturated

aqueous Na2CO3, dried (Na2SO4), and concentrated. The crude material was

purified by Florisil chromatography (hexane, Florisil deactivated with 10%

1 triethylamine) to give a colorless oil (1.24 g, 93%); H NMR (200 MHz, CDCl3) d

0.58 (qd, J = 1.3, 7.6 Hz, 6 H), 0.95 (t, J = 7.6 Hz, 9 H), 1.49 (s, 3 H), 1.52 (s, 3

H), 1.73 (ddt, J = 4.8, 9.5, 14.0 Hz, 1 H), 2.08 (dtd, J = 2.8, 9.5, 14.0 Hz, 1 H),

3.76 (m, 2 H), 4.62 (dt, J = 1.6, 9.5 Hz, 1 H), 6.71 (d, J = 1.6 Hz, 1 H); 13C NMR

(50 MHz, C6D6) d 4.7, 6.9, 20.5, 27.1, 34.3, 57.8, 65.5, 101.4, 120.0 (q, J = 240

Hz), 132.2, 137.6; IR (neat) 1680, 2957 cm-1; HRMS (M+Na+) calcd for

C15H27O6SSiF3Na 443.1147, found 443.1168.

132

methyl 2,2-dimethyl-4-[2-(triethylsilyloxy)ethyl]-4H-[1,3]dioxin-5-carboxylate

(40). To a solution of triflate 36iv (2.00 g, 5.2 mmol) in methanol (8.5 mL) and

THF (173 mL) were added dppp (161 mg, 0.39 mmol), N,N-diisopropylethylamine

(2.0 mL, 11.4 mmol), and K2CO3 (3.60 g, 26.0 mmol) followed by palladium(II) acetate (88 mg, 0.39 mmol). The solution was placed under a CO atmosphere

(balloon pressure) and stirred at rt overnight. The mixture was poured onto

saturated aqueous NaHCO3 and extracted with Et2O. The combined extracts

were washed with brine, dried (Na2SO4) and concentrated. Purification of the

crude residue by silica-gel chromatography (ethyl acetate-hexanes, 1 : 32, silica- gel deactivated with 10% triethylamine) gave a colorless oil (1.48 g, 86%); 1H

NMR (200 MHz, CDCl3) d 0.55 (qd, J = 1.3, 7.9 Hz, 6 H), 0.92 (t, J = 7.9 Hz, 9 H),

1.40 (s, 3 H), 1.44 (s, 3 H), 1.78 (tdd, J = 5.0, 6.4, 8.1 Hz, 1 H), 2.28 (dtd, J = 2.7,

6.4, 7.6 Hz, 1 H), 3.53-3.80 (m, 2 H), 3.65 (s, 3 H), 4.58 (ddd, J = 1.0, 2.7, 8.1

13 Hz, 1 H), 7.47 (d, J = 1.0 Hz, 1 H); C NMR (50 MHz, CDCl3) d 4.3, 6.6, 21.5,

27.4, 35.6, 50.8, 58.7, 65.3, 100.9, 107.8, 152.8, 165.8; IR (neat) 1631, 1713 cm-

1 + ; HRMS (M+Na ) calcd for C16H30O5SiNa 353.1760, found 353.1768.

133

N-methoxy-N-methyl-2,2-dimethyl-4-[2-(triethylsilyloxy)ethyl]-4H-[1,3]-

dioxin-5-carboxamide (41). To a solution of triflate 36iv (370 mg, 0.096 mmol)

in DMF (19 mL) were added dppp (79 mg, 0.19 mmol), N,O-

93 dimethylhydroxylamine (712 mg, 19.2 mmol) and Et3N (54 mL, 0.384 mmol) followed by palladium(II) acetate (43 mg, 0.19 mmol). The solution was placed under a CO atmosphere (balloon pressure) and stirred at rt overnight. The mixture was poured onto a saturated aqueous NaHCO3 and extracted with Et2O.

The combined extracts were washed with brine, dried (Na2SO4) and concentrated. Purification of the crude residue by Florisil chromatography (ethyl acetate-hexanes, 1 : 3, Florisil deactivated with 10% triethylamine) gave a

1 colorless oil (280 mg, 81%): H NMR (200 MHz, CDCl3) d 0.58 (qd, J = 1.3, 7.9

Hz, 6 H), 0.94 (t, J = 7.9 Hz, 9 H), 1.47 (s, 3 H), 1.48 (s, 3 H), 1.65 (tdd, J = 4.3,

8.5, 11.5 Hz, 1 H), 2.00 (dtd, J = 2.6, 7.9, 11.5 Hz, 1 H), 3.22 (s, 3 H), 3.65 (s, 3

H) 3.67-3.81 (m, 2 H), 4.74 (ddd, J = 1.4, 2.6, 8.5 Hz, 1 H), 7.22 (d, J = 1.4 Hz, 1

13 H); C NMR (50 MHz, CDCl3) d 4.3, 6.7, 21.5, 27.6, 32.6, 35.9, 58.8, 60.9, 65.4,

100.2, 110.9, 148.2, 167.0; IR (neat) 1651, 1615 cm-1; HRMS (M+Na+) calcd for

C17H33O5NSiNa 382.2026, found 382.2016.

134

N,N-dimethyl-2,2-dimethyl-4-[2-(triethylsilyloxy)ethyl]-4H-[1,3]dioxin-5- carboxamide (42). To a solution of triflate 36iv (852 mg, 2.22 mmol) in THF (74 mL) were added dppp (68 mg, 0.170 mmol), dimethylamine (2.0 M in THF, 11 mL,), and diisopropylethylamine (850 mL, 4.88 mmol) followed by palladium(II) acetate (37 mg, 0.170 mmol). The solution was placed under a CO atmosphere

(balloon pressure) and stirred at rt overnight. The mixture was poured onto saturated aqueous NaHCO3 and extracted with Et2O. The combined extracts were washed with brine, dried (Na2SO4) and concentrated. The crude material was purified by silica-gel chromatography (ethyl acetate-hexanes, 1 : 3, silica-gel deactivated with 10% triethylamine) to give a colorless oil (763 mg, 90%): 1H

NMR (200 MHz, CDCl3) d 0.53 (q, J = 7.6 Hz, 6 H), 0.89 (t, J = 7.6 Hz, 9 H), 1.40

(s, 3 H), 1.43 (s, 3 H), 1.63 (tdd, J = 4.5, 6.5, 9.2 Hz, 1 H), 1.85 (dtd, J = 2.6, 7.9,

9.2 Hz, 1 H), 2.97 (s, 6 H), 3.57-3.77 (m, 2 H), 4.66 (ddd, J = 1.3, 2.6, 9.2 Hz, 1

13 H), 6.62 (d, J = 1.3 Hz, 1 H); C NMR (50 MHz, CDCl3) d 4.3, 6.6, 21.8, 27.6,

35.6, 36.9, 58.7, 65.5, 100.1, 111.6, 143.6, 168.1; IR (neat) 1649, 1622 cm-1.

135

(E)-2-phenylhept-2-enal (44). A solution of dioxin 37 (20 mg, 0.08 mmol) in d8- toluene (1 mL) was heated at 110 °C for 8 h and then concentrated to give a

1 colorless oil (15 mg, >99%); H NMR (300 MHz, CDCl3) d 0.88 (t, J = 7.3 Hz, 3

H), 1.40 (h, J = 7.3 Hz, 2 H), 1.48 (p, J = 7.3 Hz, 2 H), 2.37 (q, J = 7.2 Hz, 2 H),

6.72 (t, J = 7.2 Hz, 1 H), 9.61 (s, 1 H).

methyl 2-formyl-5-(triethylsilyloxy)pent-2-enoate (49). A solution of ester 40

(85 mg, 0.26 mmol) in CDCl3 (1 mL) was heated at 50 °C for 36 h and concentrated to give a colorless oil (70 mg, 99%) as an inseparable mixture of

1 stereoisomers (Z : E, 72 : 28); H NMR (200 MHz, CDCl3) d 0.60 (q, J = 8.1 Hz, 6

H), 0.92 (t, J = 8.1 Hz, 9 H), major 2.81 (q, J = 7.5 Hz, 2 H), minor 2.94 (q, J = 7.1

Hz, 2 H), 3.65-3.88 (m, 2 H), minor 3.82 (s, 3 H), major 3.83 (s, 3 H), major 7.22

(t, J = 7.5 Hz, 1 H), minor 7.54 (td, J = 2.7, 7.1 Hz, 1 H), major 9.65 (s, 1 H),

13 minor 10.08 (d, J = 2.7 Hz, 1 H); C NMR (50 MHz, CDCl3) major isomer: d 4.3,

6.6, 33.7, 52.0, 60.8, 134.6, 157.0, 189.1; minor isomer: 4.3, 6.6, 33.0, 52.1,

61.0, 130.0, 157.3, 190.8; IR (neat) 1703, 1720 cm-1; HRMS (M+H+) calcd for

C13H25O4Si; 273.1522, found 273.1533.

136

N-methoxy-N-methyl-2-formyl-5-(triethylsilyloxy)pent-2-enamide (50). A

solution of Weinreb amide 41 (690 mg, 1.92 mmol) in toluene (19 mL) was

heated at 100 °C for 2 h. The solution was concentrated and the crude material

was purified by silica-gel chromatography (ethyl acetate-hexane, 1 : 3) to give a

colorless oil (533 mg, 93%) as an inseparable mixture of stereoisomers (Z : E, 97

1 : 3); H NMR (200 MHz, CDCl3) d 0.61 (qd, J = 1.3, 7.4 Hz, 6 H), 0.95 (t, J = 7.4

Hz, 9 H), major 2.61 (q, J = 6.2 Hz, 2 H), minor 2.88 (q, J = 6.3 Hz, 2 H), 3.24 (d,

J = 15.8 Hz, 3 H), 3.60 (d, J = 15.8 Hz, 3 H), 3.79 (t, J = 6.2 Hz, 2 H), 6.88 (t, J =

6.2 Hz, 1 H), major 9.44 (s, 1 H), minor 9.91 (d, J = 0.5 Hz, 1 H); 13C NMR (90

MHz, CDCl3) d 4.1, 6.5, 31.6, 33.6, 60.6, 61.5, 141.3, 152.6, 189.5; IR (neat)

-1 + 1694, 1659 cm ; HRMS (M+H ) calcd for C14H28O4NSi 302.1788, found

302.1779.

137

(Z)-N,N-dimethyl-2-formyl-5-(triethylsilyloxy)pent-2-enamide (51). A solution of amide 42 (200 mg, 0.58 mmol) in toluene (4 mL) was heated at 100 °C for 1 h and then concentrated to give a colorless oil (158 mg, 85%); 1H NMR (360 MHz,

CDCl3) d 0.54 (q, J = 8.0 Hz, 6 H), 0.88 (t, J = 8.0 Hz, 9 H), 2.51 (q, J = 6.1 Hz, 2

H), 2.82 (s, 3 H), 2.99 (s, 3 H), 3.74 (t, J = 6.1 Hz, 2 H), 6.80 (t, J = 6.1 Hz, 1 H),

13 9.39 (s, 1 H); C NMR (90 MHz, CDCl3) ? 4.6, 7.0, 34.2, 34.7, 38.1, 61.1, 143.0,

153.9, 165.8, 190.6; IR (neat) 1692, 1643 cm-1; HRMS (M+H+) calcd for

C14H28O3NSi 286.1838, found 286.1857.

methyl 6-isobutoxy-4-[2-(triethylsilyloxy)ethyl]-5,6-dihydro-4H-pyran-3- carboxylate (55). To a solution of esters 50 (101 mg, 0.37 mmol) in CH2Cl2 (2 mL) was added isobutyl vinyl ether (242 mL, 1.86 mmol) and the mixture was

stirred at rt for 24 h. The solution was concentrated and the crude material was

purified by silica-gel chromatography (ethyl acetate-hexanes, 1 : 32, silica-gel

deactivated with 10% triethylamine) to afford a colorless oil (112 mg, 81%) as an

inseparable mixture of diastereomers (cis : trans, 74 : 26): 1H NMR (300 MHz,

CDCl3) d 0.58 (q, J = 8.0 Hz, 6 H), major 0.85 (dd, J = 1.4, 6.7 Hz, 6 H), minor

138

0.89 (d, J = 2.3 Hz, 6 H), 0.94 (t, J = 8.0 Hz, 9 H), minor 1.18 (dt, J = 7.9, 11.8

Hz, 1 H), minor (ddd, J = 6.4, 9.7, 11.8 Hz, 1 H), 1.74-1.94 (m, 4 H), minor 2.06

(dt, J = 1.9, 12.0 Hz, 1 H), major (dt, J = 2.7, 14.3 Hz, 1 H), 2.61-2.73 (m, 1 H),

major (dd , J = 6.4, 9.1 Hz, 1 H), minor 3.27 (dd , J = 6.9, 9.3 Hz, 1 H), major

3.55 (dd , J = 6.6, 9.1 Hz, 1 H), 3.62-3.75 (m, 1 H), minor 3.68 (s, 3 H), major

3.69 (s, 3 H), minor 4.97 (dd, J = 2.3, 8.9 Hz, 1 H), major 5.08 (t, J = 2.7 Hz, 1 H),

13 7.45 (s, 1 H); C NMR (75 MHz, CDCl3) major isomer: ? 4.4, 6.7, 19.2, 24.9,

28.4, 28.8, 35.7, 51.0, 61.1, 75.9, 98.6, 110.8, 151.8, 167.9; minor isomer: d 4.3,

6.7, 19.1, 27.3, 28.4, 31.2, 37.7, 51.0, 61.2, 76.2, 98.7, 109.9, 153.1, 167.7; IR

-1 + (neat) 1712, 1635 cm ; HRMS (M+Na ) calcd for C19H36O5SiNa 395.2229, found

395.2286.

N-methoxy-N-methyl-6-isobutoxy-4-[2-(triethylsilyloxy)ethyl]-5,6-dihydro-

4H-pyran-3-carboxamide (52). To a solution of amides 50 (394 mg, 1.31 mmol) in CH2Cl2 (3 mL) was added isobutyl vinyl ether (852 mL, 6.54 mmol) and the

solution subjected to high pressure (12 kbar) overnight. The solution was then

concentrated and the crude material was purified by silica-gel chromatography

(ethyl acetate-hexanes, 1 : 9) to give a colorless oil (369 mg, 71%) as an

inseparable mixture of diastereomers (cis : trans, 92 : 8); 1H NMR (300 MHz,

CDCl3) d 0.62 (q, J = 7.8 Hz, 6 H), 0.81 (d, J = 6.7 Hz, 3 H), 0.85 (d, J = 6.7 Hz, 3

139

H), 1.75 (h, J = 6.7 Hz, 1 H), 1.82-1.93 (m, 3 H), 2.08-2.19 (m, 1 H), minor 3.97

(s, 3 H), major 3.98 (s, 3 H), 3.03 (dd , J = 6.2, 9.1 Hz, 1 H), 3.17 (s, 3 H), 3.53

(dd, J = 6.7, 9.1 Hz, 1 H), 3.75 (dd, J = 6.1, 6.7 Hz, 2 H), major 4.74 (dd, J = 3.1,

4.4 Hz, 1 H), minor 4.89 (dd, J = 2.5, 5.2 Hz, 1 H), 7.29 (d, J = 1.7 Hz, 1 H); 13C

NMR (90 MHz, CDCl3) d 4.8, 7.2, 19.7, 27.6, 28.9, 31.8, 34.3, 36.9, 61.1, 61.3,

76.3, 113.6, 147.2, 169.6; IR (neat) 2956, 1645 cm-1; HRMS (M+H+) calcd for

C20H40O5NSi 402.2676, found 402.2709.

N,N-dimethyl-6-isobutoxy-4-[2-(triethylsilyloxy)ethyl]-5,6-dihydro-4H-pyran-

3-carboxamide (56). To a solution of amide 51 (157 mg, 0.55 mmol) in CH2Cl2

(3 mL) was added isobutyl vinyl ether (360 mL, 2.75 mmol) and the solution was subjected to high pressure (12 kbar) overnight. The mixture was then concentrated to give a colorless oil (180 mg, 85%) as a mixture of diastereomers

1 (cis : trans, 93 : 7); H NMR (360 MHz, CDCl3) d 0.50 (q, J = 7.9 Hz, 6 H), 0.84

(d, J = 6.6 Hz, 6 H), 0.87 (t, J = 7.9 Hz, 9 H), 1.52-1.74 (m, 2 H), 1.75-1.86 (m, 2

H), 1.98 (ddd, J = 2.5, 6.8, 13.6 Hz, 1 H), 2.75 (dt, J = 5.1, 14.5 Hz, 1 H), minor

2.92 (s, 6 H), major 2.95 (s, 6 H), 3.18 (dd, J = 6.4, 9.1 Hz, 1 H), 3.46-3.61 (m, 3

H), 4.91 (dd, J = 2.6, 5.6 Hz, 1 H), major 6.39 (d, J = 0.6 Hz, 1 H) minor 6.43 (d, J

13 = 0.8 Hz, 1 H); C NMR (75 MHz, CDCl3) major isomer: d 4.2, 6.6, 19.1, 27.6,

28.33, 31.5, 36.2, 60.7, 75.6, 98.7, 113.9, 142.4, 170.3 minor isomer: d 3.9, 4.6,

140

19.1, 25.8, 28.3, 32.1, 37.1, 60.3, 75.1, 96.8, 114.9, 142.5, 169.9; IR (neat) 2955,

-1 + 1650cm ; HRMS (M+H ) calcd for C20H40O4NSi 386.2727, found 386.2723.

N-methoxy-N-methyl-6-isobutoxy-4-(2-hydroxyethyl)-5,6-dihydro-4H-pyran-

3-carboxamide. To the mixture of dihydropyans 52 (266 mg, 0.660 mmol) in

THF (3 mL) was added TBAF (1.0 M in THF, 795 mL) at 0 °C. The solution was

stirred for 2 h and poured onto saturated aqueous NH4Cl and extracted with

Et2O. The combined extracts were dried (Na2SO4) and concentrated to give a

1 yellow oil (190 mg, 99%); H NMR (360 MHz, CDCl3) d 0.87 (d, J = 6.7 Hz, 6 H),

1.71 (h, J = 6.7 Hz, 1 H), 1.80-1.97 (m, 4 H), 2.79-2.91 (m, 1 H), 2.92-3.03 (m, 1

H), minor 3.22 (s, 3 H), major 3.23 (s, 3 H), 3.27 (dd , J = 6.2, 9.1 Hz, 1 H), 3.49-

3.61 (m, 2 H), 3.53 (dd , J = 6.8, 9.1 Hz, 1 H), 3.6 (s, 3 H), minor 4.99 (dd, J =

2.5, 6.8 Hz, 1 H), major 5.06 (t, J = 3.2 Hz, 1 H), minor 7.21 (d, J = 0.8 Hz, 1 H),

13 major 7.22 (d, J = 0.6 Hz, 1 H); C NMR (90 MHz, CDCl3) d 19.2, 24.9, 28.4,

31.3, 34.1, 37.8, 60.1, 60.4, 75.8, 98.2, 111.4, 148.9, 169.4; IR (neat) 3427,

-1 + 1643cm ; HRMS (M+H ) calcd for C14H26O5N 288.1811, found 288.1787.

141

6-isobutoxy-4,4a,5,6-tetrahydro-3H-pyrano[3,4-c]pyran-1-one (53). To a

solution of the above alcohols (41 mg, 0.14 mmol) in DMF (5 mL) was added

NaH (60% dispersion in mineral oil, 6.0 mg, 0.16 mmol). The solution was stirred at rt for 48 h and poured onto brine, extracted with Et2O, washed with brine, dried

1 (Na2SO4) and concentrated to give a colorless oil (27 mg, 83%); H NMR (300

MHz, CDCl3) d minor 0.81 (d, J = 6.7 Hz, 3 H), minor 0.83 (d, J = 6.7 Hz, 3 H),

major 0.86 (d, J = 6.7 Hz, 3 H), major 0.87 (d, J = 6.7 Hz, 3 H), 1.42-1.64 (m, 2

H), 1.83 (h, J = 6.7 Hz, 1 H), 1.90 (ddt, J = 2.3, 4.4, 13.6 Hz, 1 H), minor 2.01

(ddd, J = 1.5, 5.0, 13.0 Hz, 1 H), major 2.13 (ddd, J = 2.3, 5.0, 13.1 Hz, 1 H),

2.69 (ttd, J = 2.2, 4.9, 12.1 Hz, 1 H), major 3.31 (dd, J = 6.8, 9.2 Hz, 1 H), minor

3.32 (dd, J = 6.7, 9.2 Hz, 1 H), minor 3.56 (dd, J = 6.7, 9.2 Hz, 1 H), major 3.73

(dd, J = 6.5, 9.2 Hz, 1 H), 4.28 (ddd, J = 2.5, 11.4, 12.7 Hz, 1 H), 4.45 (ddd, J =

2.0, 4.4, 11.4 Hz, 1 H), major 5.09 (dd, J = 2.3, 9.8 Hz, 1 H), minor 5.22 (t, J =

1.9 Hz, 1 H), minor 7.62 (d, J = 2.2 Hz, 1 H), major 7.65 (d, J =2 .0 Hz, 1 H); 13C

NMR (75 MHz, CDCl3) d 19.1, 19.2, , 21.4, 29.2, 29.7, 33.5, 68.2, 76.4, 101.6,

-1 + 104.1, 155.4; IR (neat) 1705, 1616 cm ; HRMS (M+Na ) calcd for C12H18O4Na

249.1103, found 249.1098.

142

methyl 2,2-dimethyl-4-(2-hydroxyethyl)-4H-[1,3]dioxin-5-carboxylate. To a

solution of ester 40 (376 mg, 1.14 mmol) in CH3CN (11 mL) at –30 °C was added

HF (5.0% in CH3CN, 7.0 mL) dropwise. The solution was stirred for 0.5 h at –30

°C and poured onto saturated aqueous NaHCO3 and extracted with EtOAc. The

combined extracts were washed with brine, dried (Na2SO4) and concentrated to

1 provide the alcohol as a colorless oil (247 mg, >99%); H NMR (200 MHz, CDCl3)

d 1.41 (s, 3 H), 1.45 (s, 3 H), 1.40 (s, 3 H), 1.91 (tdd, J = 5.0, 6.4, 7.5 Hz, 1 H),

2.28 (dtd, J = 3.0, 6.4, 7.6 Hz, 1 H), 2.52 (br s, 1 H), 3.64 (s, 3 H), 3.60-3.80 (m, 2

H), 4.65 (ddd, J = 1.1, 3.0, 7.5 Hz, 1 H), 7.49 (d, J = 1.1 Hz, 1 H); 13C NMR (50

MHz, CDCl3) d 21.3, 27.4, 34.9, 51.0, 59.9, 67.8, 100.9, 107.1, 153.2, 165.6.

methyl 2,2-dimethyl-4-(2-oxoethyl)-4H-[1,3]dioxin-5-carboxylate (80). To a solution of oxalyl chloride (109 mL, 1.25 mmol) in CH2Cl2 (3.20 mL) at –60 °C was added DMSO (194 mL, 2.73 mmol) dropwise. The mixture was stirred at –60 °C

for 0.25 h and then a solution of the above alcohol (246 mg, 1.14 mmol) in

CH2Cl2 (1.0 mL) was added. This mixture was stirred for another 0.25 h and

Et3N (793 mL, 5.69 mmol) was added. The solution was allowed to warm to rt,

143

poured onto saturated aqueous NaHCO3 and extracted with CH2Cl2. The

combined extracts were washed with brine, dried (Na2SO4) and concentrated to

1 provide aldehyde 80 as a colorless oil (231 mg, 95%); H NMR (360 MHz, CDCl3) d 1.49 (s, 6 H), 2.73 (ddd, J = 2.5, 7.3, 16.8 Hz, 1 H), 3.09 (ddd, J = 1.7, 3.7, 16.8

Hz, 1 H), 3.71 (s, 3 H), 4.99 (ddd, J = 2.5, 3.7, 7.3 Hz, 1 H), 7.60 (s, 1 H), 9.76

(dd, J = 1.7, 2.5 Hz, 1 H).

dioxin 78. To a solution of TMEDA (211 mL, 1.40 mmol) in Et2O (1.40 mL) at –

78 °C was added n-BuLi (2.5 M in hexane, 560 mL, 1.40 mmol) dropwise. The

mixture was stirred at –78 °C for 0.5 h when a solution of carbamte 7994 (140 mg, 0.70 mmol) in Et2O (0.70 mL) was added. This mixture was stirred for another 0.5 h at –78 °C and Ti(Oi-Pr)4 (620 mL, 2.10 mmol) was added and the mixture stirred for 0.5 h at –78 °C. A solution of aldehyde 80 (50 mg, 0.23 mmol) in Et2O (250 mL) was added dropwise and the reaction mixture was warmed to –

30 °C over 1 h. The resultant solution was poured onto saturated aqueous

NaHCO3 and extracted with CH2Cl2. The combined extracts were washed with brine, dried (Na2SO4), concentrated and the crude material was purified by silica- gel chromatography (ethyl acetate-hexane, 1 : 3) to give a colorless oil as an inseparable mixture (3 : 1) of diastereomers 90 (77 mg, 79%).

144

cycloadduct 83. A solution of dioxin 78 (62 mg, 0.15 mmol) in toluene (3 mL) was heated at reflux for 1 h and concentrated. The crude material was purified by silica-gel chromatography (ethyl acetate-hexane, 1 : 3) to give a colorless oil (48

1 mg, 90%) as a single diastereomer: H NMR (400 MHz, C6D6) d 0.92 (d, J = 6.9

Hz, 3H), 0.95 (d, J = 6.9 Hz, 6 H), 0.97 (d, J = 6.9 Hz, 6 H), 1.17 (br s, 1 H), 1.42

(ddd, J = 4.4, 12.0, 13.7 Hz, 1 H), 1.61 (td, J = 2.3, 12.0 Hz, 1 H), 1.67 (dq, J =

6.7, 18.0 Hz, 1 H), 2.54 (tdd, J = 2.3, 7.5, 12.0 Hz, 1 H), 3.11 (dt, J = 7.5, 13.7

Hz, 1 H), 3.37 (s, 3 H), 4.02 (ddd, J = 4.4, 6.7, 6.7 Hz, 1 H), 6.82 (d, J = 2.2 Hz, 1

H), 7.52 (d, J = 2.5 Hz, 1 H).

dioxin 87. To a solution of alcohols 78 (85 mg, 0.21 mmol) in CH2Cl2 (1.0 mL) at

–78 °C was added TBSOTf (71 mL, 0.31 mmol), followed by DIPEA (108 mL, 0.62

mmol). The mixture was warmed to rt over 12 h, poured onto saturated aqueous

NaHCO3 and extracted with CH2Cl2. The combined extracts were washed with brine, dried (Na2SO4), concentrated and the crude material was purified by silica- gel chromatography (ethyl acetate-hexane, 1 : 9) to give a colorless oil (63 mg,

145

58%) as a 3 : 1 mixture of diastereomers: major isomer: 1H NMR (360 MHz,

CDCl3) d 0.05 (s, 3 H), 0.07 (s, 3H), 0.90 (s, 9H), 0.99 (d, J = 7.0 Hz, 3 H), 1.23

(d, J = 6.8 Hz, 12 H), 1.37 (ddd, J = 2.3, 10.4, 13.8 Hz, 1 H), 1.43 (s, 3 H), 1.46

(s, 3H), 2.24 (ddd, J = 2.3, 9.7, 13.9 Hz, 1 H), 2.77 – 2.85 (m, 1 H), 3.68 (s, 3 H),

3.84-4.08 (m, 2 H), 3.95 (ddd, J = 2.1, 3.3, 9.7 Hz, 1 H), 4.56 (ddd, J = 0.9, 2.1,

10.4 Hz, 1 H), 4.63 (dd, J = 6.6, 9.7 Hz, 1 H), 6.96 (dd, J = 0.9, 6.8 Hz, 1 H), 7.49

13 (d, J = 0.9 Hz, 1 H); C NMR (90 MHz, CDCl3) d -4.7, -4.0, 15.4, 18.2, 21.8,

25.9, 27.5, 36.4, 36.3, 50.9, 65.3, 70.8, 100.9, 108.8, 112.5, 152.7, 165.8; IR

-1 1 (neat) 1714, 1633 cm . minor isomer: H NMR (200 MHz, CDCl3) d 0.06 (s, 6 H),

0.83 (s, 9H), 1.03 (d, J = 7.0 Hz, 3 H), 1.23 (d, J = 6.8 Hz, 12 H), 1.37 (m, 1 H),

1.39 (s, 3 H), 1.48 (s, 3H), 2.24 (ddd, J = 2.3, 9.3, 13.9 Hz, 1 H), 2.77 – 2.85 (m,

1 H), 3.65 (s, 3 H), 3.64-4.12 (m, 2 H), 3.85 (ddd, J = 3.1, 5.3, 12.3 Hz, 1 H), 4.44

(br d, J = 8.8 Hz, 1 H), 4.86 (dd, J = 6.5, 9.7 Hz, 1 H), 7.06 (d, J = 6.5 Hz, 1 H),

13 7.49 (s, 1 H); C NMR (50 MHz, CDCl3) d -4.7, -4.0, 15.4, 18.1, 21.6, 25.9, 27.5,

35.3, 39.3, 50.9, 66.0, 72.0, 100.9, 108.3, 111.7, 134.5, 152.8; IR (neat) 1714,

1633 cm-1.

146

cycloadduct 88. To a solution of the tert-butyldimethylsilyl ether 87 (114 mg,

0.22 mmol) in toluene (20 mL) at –20 °C was added BF3•OEt2 (33 mL, 0.26 mmol). The mixture was warmed to rt over 12 h, poured onto saturated aqueous

NaHCO3 and extracted with Et2O. The combined extracts were washed with brine, dried (Na2SO4), concentrated and the 3 : 1 mixture of diastereomers were

purified by silica-gel chromatography (Et2O-hexane-benzene, 1 : 9 : 10) to give a

1 colorless oil (80 mg, 78%): major isomer: H NMR (400 MHz, CDCl3) d 0.04 (s, 3

H), 0.07 (s, 3H), 0.87 (s, 9 H), 1.07 (d, J = 7.0 Hz, 3 H), 1.21(d, J = 6.8 Hz, 3 H),

1.11 (m, 14 H), 1.64 (ddd, J = 3.9, 9.9, 13.0 Hz, 1 H), 2.10 (ddd, J = 5.3, 7.5, 14.3

Hz, 1 H), 2.14 – 2.24 (m, 2 H), 3.14 (q, J = 8.5 Hz, 1 H), 3.71 (s, 3 H), 3.73-3.98

(m, 2 H), 4.08 (ddd, J = 2.1, 3.3, 5.7 Hz, 1 H), 6.30 (d, J = 3.7 Hz, 1 H), 7.41 (d, J

13 = 1.6 Hz, 1 H); C NMR (100 MHz, CDCl3) d -5.0, -4.7, 15.1, 18.1, 18.2, 25.7,

25.8, 31.8, 39.4, 41.4, 44.6, 51.1, 74.7, 112.0, 153.2, 167.8; IR (neat) 1713, 1637

-1 1 cm : minor isomer: H NMR (400 MHz, CDCl3) d 0.04 (s, 3 H), 0.07 (s, 3H), 0.84

(s, 9 H), 1.36 (d, J = 6.9 Hz, 3H), 0.95-1.17 (br m, 22 H), 1.42 (ddd, J = 4.4,

12.0, 13.7 Hz, 1 H), 1.61 (td, J = 2.3, 12.0 Hz, 1 H), 1.67 (dq, J = 6.7, 18.0 Hz, 1

H), 2.40 (tdd, J = 2.3, 7.5, 12.0 Hz, 1 H), 3.11 (dt, J = 7.5, 13.7 Hz, 1 H), 3.37 (s,

3 H), 4.32 (ddd, J = 4.4, 6.7, 6.7 Hz, 1 H), 6.52 (d, J = 2.2 Hz, 1 H), 7.38 (d, J =

2.5 Hz, 1 H).

147

loganetin. To a solution of the tert-butyldimethylsilyl ether of cis-cycloadduct 88

(17 mg, 0.034 mmol) in 1,4-dioxane (0.5 mL) was added HCl (6.0 M in H2O, 333 mL). The solution was stirred at rt for 48 h, poured onto saturated aqueous

NaHCO3 and extracted with EtOAc. The combined extracts dried (Na2SO4), concentrated and the crude material was purified by silica-gel chromatography

(MeOH-EtOAc-hexane, 0.3 : 5 : 4.7) to give a colorless oil (6.0 mg, 73%): 1H

NMR (400 MHz, CDCl3) d 1.14 (d, J = 6.9 Hz, 3 H), 1.23 (br s, 1H), 1.58 (ddd, J =

4.7, 8.6, 13.7 Hz, 1 H), 1.52–1.74 (br m, 1 H), 1.89 (m, 1 H), 1.98 (ddd, J = 6.9,

8.1, 12.7 Hz, 1 H), 2.32 (dd, J = 7.4, 14.1 Hz, 1 H), 3.19 (q, J = 8.4, 1 H), 3.72 (s,

3 H), 4.18 (br m, 1 H), 4.99 (d, J = 5.1 Hz, 1 H), 7.41 (s, 1 H); 13C NMR (100

MHz, CDCl3) d 13.2, 31.4, 41.3, 42.2, 46.4, 51.2, 74.3, 95.5, 111.9, 167.7; IR

(neat) 3448, 1686 cm-1.

148

6-carboethoxy-2,7-dioxabicyclo[3.2.0]hept-3-ene (102).34 A solution of ethyl

glyoxylate (33.9 g, 0.332 mol) in furan (330 mL) at 0 °C was irradiated with a 550

W high-pressure mercury lamp under argon in a Pyrex filtered photochemical

reactor. After 28 h the furan was distilled off and the residue purified by silica-gel

flash chromatography (ethyl acetate-hexane, 1 : 4) to give a white solid (36.5 g,

1 65%): mp 35-36 °C; H NMR (200 MHz, CDCl3) d 1.32 (t, J = 7.2Hz, 3 H), 3.76

(dddd, J = 1.0, 2.9, 3.0, 4.1 Hz, 1 H) 4.27 (q, J = 7.2 Hz, 2 H), 4.88 (d, J = 3.0 Hz,

1 H), 5.43 (t, J = 2.9 Hz, 1 H), 6.47 (d, J = 4.1 Hz, 1 H), 6.66 (dd, J = 1.0, 2.9 Hz,

13 1 H); C NMR (50 MHz, CDCl3) d 14.0, 49.1, 61.5, 86.1, 103.4, 108.7, 148.9,

-1 + 170.5; IR (neat) 1748, 1606 cm ; HRMS (M+H ) calcd for C8H11O4 171.0657, found 171.0642.

ethyl trans-(2-ethylsulfanyl-2,3-dihydrofuran-3-yl)hydroxyacetate (120). To a solution of oxetane 102 (3.00 g, 17.6 mmol) and ethanethiol (6.50 mL, 88.2 mmol) in CH3CN (18 mL) was added BF3•OEt2 (223 mL, 1.76 mmol) dropwise at

–40 °C. The reaction mixture was stirred at –40 °C for 5 min, quenched with saturated aqueous Na2CO3, extracted with Et2O, dried (Na2SO4) and

149

concentrated. Purification of the crude residue by silica-gel chromatography

1 (Et2O-CH2Cl2-hexane, 0.7 : 4 : 5) gave a colorless oil (2.86 g, 70%): H NMR

(200 MHz, CDCl3) d 1.24 (t, J = 7.1 Hz, 3 H), 1.25 (t, J = 7.4 Hz, 3 H), 2.61 (dq, J

= 7.4, 12.9 Hz, 1 H), 2.72 (dq , J = 7.4, 12.9 Hz, 1 H), 3.09 (d, J = 5.6 Hz, 1 H),

3.08-3.14 (m, 1 H), 4.09 (t, J = 5.6 Hz, 1 H), 4.18 (qd, J = 2.8, 7.1 Hz, 1 H), 4.25

(qd, J = 2.8, 7.1 Hz, 1 H) 4.95 (t, J = 2.7 Hz, 1 H), 5.61 (d, J = 5.2 Hz, 1 H), 6.32

13 (dd, J = 2.0, 2.7 Hz, 1 H); C NMR (50 MHz, CDCl3) d 14.3, 14.7, 24.9, 50.5,

54.0, 61.9, 71.0, 85.7, 89.5, 100.3, 101.3, 145.9, 172.9; IR (neat) 3477, 1731,

1620 cm-1.

ethyl trans-acetoxy(2-ethylsulfanyl-2,3-dihydrofuran-3-yl) acetate. To a

solution of alcohol 122 (3.62 g, 15.6 mmol) in CH2Cl2 (63 mL) were added

pyridine (3.78 mL, 46.8 mmol), acetic anhydride (2.21 mL, 23.4 mmol), and 4-

(dimethylamino)pyridine (190 mg, 1.56 mmol) at 0 °C. The mixture was stirred 2

h at 0 °C, quenched with saturated aqueous NaHCO3, extracted with CH2Cl2.

The combined organic layers were washed with brine, dried (Na2SO4), and

concentrated. Purification of the crude residue by silica-gel flash

chromatography (ethyl acetate-hexanes, 1 : 19) gave a colorless oil (3.81 g,

1 89%): H NMR (200 MHz, CDCl3) d 1.25 (t, J = 7.2 Hz, 3 H), 1.27 (t, J = 7.4 Hz, 3

H), 2.10 (s, 3 H), 2.63 (dq, J = 7.4, 12.8 Hz, 1 H), 2.74 (dq, J = 7.4, 12.8 Hz, 1 H),

150

3.26 (dddd, J = 1.9, 2.8, 4.4, 5.1 Hz, 1 H), 4.19 (qd, J = 3.0, 7.2 Hz, 2 H), 4.90 (t,

J = 2.8 Hz, 1 H), 4.95 (d, J = 4.4 Hz, 1 H) 5.63 (d, J = 5.1 Hz, 1 H), 6.31 (dd, J =

13 1.9, 2.8 Hz, 1 H); C NMR (50 MHz, CDCl3) d?14.0, 14.8, 20.4, 25.0, 51.5, 61.5,

72.1, 85.9, 99.7, 146.3, 168.0, 170.1; IR (neat) 1748, 1621 cm-1; HRMS (M+Na+) calcd for C12H18O5SNa 297.0773, found 297.0788.

ethyl trans-(2-ethylsulfany1-2,3-dihydrofuran-3-yl)acetate (125). To a solution of the above acetate (3.81 g, 13.9 mmol) in HMPA (23 mL) was added ethanol (956 mL, 16.6 mmol) followed by SmI2 (0.1 M in THF, 347 mL). The purple solution was stirred 10 min and quenched with 0.1 N HCl. The aqueous layer was extracted with Et2O and the combined organic layers were washed with aqueous saturated NaHCO3 and brine, dried (Na2SO4) and concentrated.

Purification of the crude residue by silica-gel flash chromatography (Et2O-

1 hexane, 1 : 12) gave a colorless oil (2.53 g, 84%): H NMR (200 MHz, CDCl3) d

1.17 (t, J = 7.2 Hz, 3 H), 1.22 (t, J = 7.4 Hz, 3 H), 2.30 (dd, J = 7.4, 15.4 Hz, 1 H),

2.40 (dd, J = 7.4, 15.4 Hz, 1 H), 2.58 (dq, J = 7.4, 12.8 Hz, 1 H), 2.68 (dq, J =

7.4, 12.8 Hz, 1 H), 3.02 (tddd, J = 1.7, 2.7, 4.6, 7.4 Hz, 1 H), 4.06 (q, J = 7.2 Hz,

2 H), 4.99 (t, J = 2.7 Hz, 1 H), 5.33 (d, J = 4.6 Hz, 1 H), 6.19 (dd, J = 1.7, 2.7 Hz,

13 1 H); C NMR (50 MHz, CDCl3) d 14.0, 14.7, 24.8, 39.0, 45.8, 60.3, 89.7, 104.0,

151

-1 + 144.5, 170.9; IR (neat) 1732, 1617 cm ; HRMS (M+H ) calcd for C10H17O3S

217.0898, found 217.0896.

trans-2-(2-ethylsulfanyl-2,3-dihydrofuran-3-yl)ethanol. To a solution of ester

125 (2.53 g, 11.7 mmol) in Et2O (46 mL) was added LiAlH4 (444 mg, 11.7 mmol) at 0 °C. The mixture was stirred for 40 min at 0 °C then EtOAc (2 mL) was added dropwise at 0 °C. Saturated aqueous Rochelle’s salt was added dropwise until H2 evolution ceased. The aqueous layer was extracted with EtOAc and the combined organic layers were washed with saturated aqueous Rochelle’s salt and brine, dried (Na2SO4), and concentrated. Purification of the crude residue by silica-gel flash chromatography (Et2O-CH2Cl2, 1 : 9) gave a colorless oil (1.98 g,

1 97%): H NMR (200 MHz, CDCl3) d?1.27 (t, J = 7.4 Hz, 3 H), 1.66 (q, J = 6.6 Hz,

2 H), 2.41 (s, 1H), 2.63 (dq, J = 7.4, 12.8 Hz, 1 H), 2.71 (dq, J = 7.4, 12.8 Hz, 1

H), 2.78-2.89 (m, 1 H), 3.62 (t, J = 6.6 Hz, 2 H), 5.00 (t, J = 2.8 Hz, 1 H), 5.35 (d,

13 J = 5.2 Hz, 1 H), 6.22 (dd, J = 1.9, 2.8 Hz, 1 H); C NMR (50 MHZ, CDCl3) d

14.8, 25.0, 37.4, 46.6, 60.2, 90.5, 104.8, 143.8; IR (neat) 3356, 1619 cm-1;

+ HRMS (M-OH ) calcd for C8H13OS 157.0687, found 157.0688.

152

trans-2-ethylsulfanyl-3-(2-iodoethyl)-2,3-dihydrofuran (126). To a solution of the above alcohol (1.40 g, 8.04 mmol) in benzene (133 mL) was added imidazole

(657 mg, 9.65 mmol), triphenylphosphine (2.53 g, 9.65 mmol), and iodine (2.24 g,

8.84 mmol). The mixture was stirred 10 min at rt and poured onto a 1 : 1 mixture of saturated aqueous NaHCO3 and 10% aqueous Na2S2O3. The aqueous layer was extracted with Et2O and the combined organic layers were dried (Na2SO4) and concentrated. The residue was dissolved in THF (25 mL), iodomethane was added (2.50 mL, 40.2 mmol) and the mixture was stirred 2 h, filtered and concentrated. Purification of the crude residue by silica-gel flash chromatography (hexane, silica-gel deactivated with 10% triethylamine) gave a

1 colorless oil (1.78 g, 78%): H NMR (200 MHz, CDCl3) d 1.32 (t, J = 7.4 Hz, 3 H),

2.00 (dtd, J = 7.1, 7.4, 10.9 Hz, 2 H), 2.67 (dq, J = 7.4, 12.8 Hz, 1 H), 2.74 (dq, J

= 7.4, 12.8 Hz, 1 H), 2.79-2.91 (m, 1 H), 3.15 (td, J = 1.4, 7.4 Hz, 2 H), 5.04 (t, J

= 2.7 Hz, 1 H), 5.32 (d, J = 4.9 Hz, 1 H), 6.29 (dd, J = 1.8, 2.7 Hz, 1 H); 13C NMR

(50 MHz, CDCl3) d 1.7, 14.9, 25.2, 38.8, 50.3, 89.9, 103.5, 144.7; IR (neat) 1618,

-1 + 1446 cm ; HRMS (M-I ) calcd for C8H13OS 157.0687, found 157.0676.

153

4-[2-(2-ethylsulfanyl-2,3-dihydrofuran-3-yl)ethyl]-2,2-dimethyl-1,3-dioxan-5- one (127). To a solution of 2,2-dimethyl-1,3-dioxan-5-one (21)15 (419 mg, 3.22 mmol) in benzene (20 mL) was added molecular sieves (4Å, 420 mg) and cyclohexylamine (737 mL, 6.44 mmol). The mixture was stirred at rt overnight, filtered and concentrated. The crude imine 33 was dissolved in anhydrous THF

(6.5 mL) and added dropwise to a solution of lithium diethylamide [formed by addition of n-BuLi (2.5 M in hexane, 1.21 mL) to a solution of diethylamine (333 mL, 3.22 mmol) in THF (3.5 mL) at –30 °C] at –78 °C. The solution was warmed over 2 h to –25 °C and recooled to –78 °C. Iodide 126 (571 mg, 2.01 mmol) in

THF (4 mL) was added dropwise and the solution was warmed to rt overnight.

The mixture was hydrolyzed by addition of saturated aqueous NH4Cl (10 mL) followed by stirring at rt for 6 h. The aqueous layer was extracted with Et2O and the combined organic layers were washed with the saturated aqueous NaHCO3 and brine, dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel flash chromatography (ethyl acetate-hexane, 1 : 32, silica-gel deactivated with 10% triethylamine) gave a colorless oil (360 mg, 62%): 1H NMR

(360 MHz, CDCl3) d 1.28 (t, J = 7.4 Hz, 3 H), 1.39 (s, 3 H), 1.42 (s, 3 H), 1.50-

1.59 (m, 3 H), 1.83-1.93 (m, 1 H), 2.66 (dq, J = 7.4, 12.8 Hz, 1 H), 2.68-2.73 (m,

1 H), 2.73 (dq, J = 7.4, 12.8 Hz, 1 H), 3.95 (d, J = 16.9 Hz, 1 H), 4.11-4.18 (m, 1

H), 4.21 (dd, J = 1.4, 16.9 Hz, 1 H), 5.00 (t, J = 2.7 Hz, 1 H), 5.30 (d, J = 4.9 Hz, 154

13 1 H), 6.23 (dd, J = 1.9, 2.7 Hz, 1 H); C NMR (90 MHz, CDCl3) d?14.8, 23.5, 23.8,

25.0, 25.5, 30.0, 49.1, 66.5, 74.1, 90.3, 100.7, 104.5, 143.9; IR (neat) 1746, 1618

-1 + cm ; HRMS (M+H ) calcd for C14H23O4S 287.1317, found 287.1314.

4-[2-(2-ethylsulfanyl-2,3-dihydroforan-3-yl-ethyl]-2,2-dimethyl-4H-

[1,3]dioxin-5-trifluoromethanesulfonate (128). A solution of ketone 127 (358

mg, 1.25 mmol) and N-phenyltrifluoromethanesulfonimide (625 mg, 1.75 mmol)

in THF (12 mL) was added to NaHMDS (1.0 M in THF, 1.75 mL) in THF (9 mL) at

–78 °C over 2 h. The solution was warmed to rt over 2 h and quenched with

saturated aqueous NaHCO3. The aqueous layer was extracted with Et2O and

the combine organic layers were washed with saturated aqueous Na2CO3 and

brine, dried (Na2SO4) and concentrated. Purification of the crude residue by

Florisil flash chromatography (hexane, Florisil deactivated with 10%

1 triethylamine) gave a colorless oil (445 mg, 85%): H NMR (360 MHz, CDCl3) d

1.32 (t, J = 7.5 Hz, 3 H), 1.48 (s, 3 H), 1.51 (s, 3 H), 1.51-1.69 (m, 3 H), 1.78-1.88

(m, 1 H), 2.67 (dq, J = 7.5, 12.8 Hz, 1 H), 2.73-2.78 (m, 1 H), 2.78 (dq, J = 7.5,

12.8 Hz, 1 H), 4.46-4.48 (m, 1 H), 5.04 (t, J = 2.7 Hz, 1 H), 5.31 (d, J = 4.8 Hz, 1

H), 6.26 (dd, J = 1.8, 2.7 Hz, 1 H), 6.72 (d, J = 1.1 Hz, 1 H); 13C NMR (90 MHz,

CDCl3) d 15.3, 21.1, 25.5, 27.6, 27.7, 29.4, 49.5, 68.5, 90.8, 101.8, 104.9, 118.9

155

(q, J = 319 Hz), 131.4, 138.1, 144.5; IR (neat) 1679, 1619 cm-1; HRMS (M+H+) calcd for C15H22O6S2F5 419.0810, found 419.0800.

N-methoxy-N-methyl-4-[2-[2-ethylsulfanyl-2,3-dihydrofuran-3-yl)-ethyl]-2,2-

dimethyl-4H-[1,3]dioxine-5-carboxyamide (129). A solution of the triflate 128

(29 mg, 0.069 mmol), N,O-dimethyl hydroxylamine (103 mg, 2.77 mmol), dppp (7

mg, 0.017 mmol), triethylamine (5 mL, 0.035 mmol), and palladium(II) acetate (4

mg, 0.017 mmol) in DMF (2 mL) was placed under an atmosphere of CO (balloon

pressure) and stirred 10 h at rt. The solution was poured onto saturated aqueous

NaHCO3 and extracted with CH2Cl2, dried (Na2SO4) and concentrated.

Purification of the crude residue by silica-gel flash chromatography (ethyl acetate-hexane, 1 : 19, silica-gel deactivated with 10% triethylamine) gave a

1 colorless oil (22 mg, 91%): H NMR (400 MHz, CDCl3) d 1.24 (t, J = 7.5 Hz, 3 H),

1.41 (s, 6 H), 1.44-1.58 (m, 3 H), 1.62-1.74 (m, 1 H), 2.59 (dq, J = 7.5, 13.0 Hz, 1

H), 2.61-2.68 (m, 1 H), 2.69 (dq, J = 7.5, 13.0 Hz, 1 H), 3.16 (s, 3 H), 3.59 (s, 3

H), 4.53-4.55 (m, 1 H), 4.97 (t, J = 2.7 Hz, 1 H), 5.23 (d, J = 4.8 Hz, 1 H), 6.17

13 (dd, J = 1.7, 2.7 Hz, 1 H), 7.21 (d, J = 0.8 Hz, 1 H); C NMR (100 MHz, CDCl3) d

15.3, 22.1, 25.4, 28.1, 30.1, 30.4, 33.5, 49.6, 61.3, 68.5, 91.0, 100.7, 105.4,

110.8, 144.1, 149.2, 167.3; IR (neat) 1650, 1614 cm-1.

156

cycloadduct (130). A solution of amide 129 (237 mg, 0.68 mmol) in toluene (13

mL) was heated at 110 °C for 1h. The solution was cooled and concentrated.

Recrystalization from hexanes gave the pure 2,6-trans cycloadduct 130.

Concentration and further purification of the crude residue by silica-gel flash

chromatography (Et2O-benzene, 3 : 22) of the filtrate gave 130 (2,6-trans : 2,6- cis, 4.5 : 1) as a white solid (151 mg, 72%): major isomer 1H NMR (400 MHz,

CDCl3) d 1.25 (t, J = 7.5 Hz, 3 H), 1.52 (tdd, J = 9.0, 11.6, 13.1 Hz, 1 H), 1.77

(tdd, J = 3.0, 7.3, 13.1 Hz, 1 H), 1.96 (ddd, J = 3.6, 7.1, 11.3 Hz, 1 H), 2.18 (ddd,

J = 2.6, 5.0, 11.3 Hz, 1 H), 2.19-2.32 (m, 2 H), 2.48 (dddd, J = 3.0, 5.7, 7.1, 9.3

Hz, 1 H), 2.62 (dq, J = 7.5, 12.8 Hz, 1 H), 2.70 (dq, J = 7.5, 12.8 Hz, 1 H), 3.16

(s, 3 H), 3.54 (s, 3 H), 5.12 (d, J = 5.7 Hz, 1 H), 5.73 (d, J = 3.6 Hz, 1 H), 7.12 (d,

13 J = 2.6 Hz, 1 H); C NMR (100 MHz, CDCl3) d 14.9, 26.0, 28.6, 31.8, 33.1, 36.7,

46.3, 51.1, 60.6, 91.2, 103.2, 115.9, 149.9, 166.8; IR (neat) 1633, 1599 cm-1;

+ HRMS (M+H ) calcd for C14H22O4NS 300.1270, found 300.1288. minor isomer

1 H NMR (400 MHz, CDCl3) d 1.24 (t, J = 7.4 Hz, 3 H), 1.69 (dtd, J = 6.3, 12.8,

19.1 Hz, 1 H), 1.70-1.75 (m, 1 H), 2.07 (dtd, J = 1.7, 5.8, 12.1 Hz, 1 H), 2.58 (dq,

J = 7.4, 12.8 Hz, 1 H), 2.57-2.62 (m, 1H), 2.68 (dq, J = 7.4, 12.8 Hz, 1 H), 2.69-

2.75 (m, 1 H), 2.88 (td, J = 7.1, 10.5 Hz, 1 H), 3.20 (s, 3 H), 3.56 (s, 3 H), 4.89 (d,

J = 5.1 Hz, 1 H), 5.68 (d, J = 3.7 Hz, 1 H), 7.25 (s, 1 H); 13C NMR (75 MHz,

CDCl3) d 14.9, 25.5, 31.3, 33.0, 33.3, 34.1, 40.6, 51.3, 60.8, 87.9, 99.2, 112.1,

157

-1 + 146.5, 168.5; IR (neat) 1648, 1449 cm , HRMS (M+H ) calcd for C14H22O4NS

300.1270, found 300.1293.

ketone (132). To a solution of amide 130 (41 mg, 0.13 mmol) in a 1 : 1 pentane /

ether solution (2 mL) was added isopentyllithium (0.08 M in 3 : 2 pentane / ether,

3.50 mL) at –40 °C. The solution was stirred 0.5 h and quenched with saturated

aqueous NH4Cl. The aqueous layer was extracted with Et2O and the combined

organic layers were dried (Na2SO4) and concentrated. Purification of the crude

residue by Florisil flash chromatography (ethyl acetate-hexane, 1 : 19, Florisil

deactivated with 10% triethylamine) gave ketone 132 (17 mg, 39%) as colorless

1 oil: H NMR (300 MHz, CDCl3) d 0.82 (d, J = 6.2 Hz, 6 H) 1.24 (t, J = 7.4 Hz, 3

H), 1.39-1.55 (m, 3 H), 1.57-1.69 (m, 1 H) 1.74 (tdd, J = 3.6, 6.9, 10.3 Hz, 1H),

1.87 (ddd, J = 3.5, 6.9, 11.7 Hz, 1 H), 2.01 (dddd, J = 2.5, 4.7, 10.3, 11.7 Hz, 1

H), 2.24 (dt, J = 8.5, 14.1 Hz, 1 H), 2.39-2.55 (m, 4 H), 2.61 (dq, J = 7.4, 12.8 Hz,

1 H), 2.68 (dq, J = 7.4, 12.8 Hz, 1 H), 5.09 (d, J = 5.7 Hz, 1 H), 5.76 (d, J = 3.5

13 Hz, 1 H), 7.39 (d, J = 2.5 Hz, 1 H); C NMR (75 MHz, CDCl3) d 14.9, 22.3, 22.4,

26.1, 27.8, 29.1, 32.2, 33.7, 35.6, 35.7, 46.1, 50.9, 91.5, 103.6, 122.9, 155.3,

-1 + 198.0; IR (neat) 1658, 1591 cm ; HRMS (M+H ) calcd for C11H27O3S 311.1681, found 311.1667.

158

carboxylic acid (131). To a solution of amide 130 (155 mg, 0.52 mmol) in Et2O

(5.2 mL) was added potassium tert-butoxide (384 mg, 3.42 mmol) and H2O (31 mL, 1.71 mmol). The solution was stirred at rt overnight then poured onto saturated aqueous KH2PO4. The aqueous layer was extracted with CH2Cl2 and the combined extracts were dried (Na2SO4) and concentrated to give a white

1 solid (130 mg, 98%): mp 118-120 °C (dec); H NMR (300 MHz, CDCl3) d 1.31 (t,

J = 7.4 Hz, 3 H), 1.63 (tdd, J = 8.5, 11.7, 13.2 Hz, 1 H), 1.83 (tdd, J = 2.7, 7.0, 9.8

Hz, 1 H), 1.98 (ddd, J = 3.6, 7.0, 11.7 Hz, 1 H), 2.15 (tdd, J = 2.7, 5.0, 13.2 Hz, 1

H), 2.31 (dt, J = 8.5, 17.2 Hz, 1 H), 2.37-2.56 (m, 2 H), 2.68 (dq, J = 7.4, 12.8 Hz,

1 H), 2.76 (dq, J = 7.4, 12.8 Hz, 1 H), 5.17 (d, J = 5.6 Hz, 1 H), 5.82 (d, J = 3.6

13 Hz, 1 H) 7.58 (d , J = 2.7 Hz, 1 H); C NMR (75 MHz, CDCl3) d 14.9, 26.1, 29.0,

29.7, 32.0, 33.3, 35.0, 46.2, 50.7, 51.1, 91.4, 103.6, 112.2, 156.9, 171.3; IR

-1 + (neat) 2929, 1662 cm ; HRMS (M-H ) calcd for C12H15O4S 255.0691, found

255.0688.

159

ketone (132). To a solution of carboxylic acid 131 (10 mg, 0.039 mmol) in THF

(580 mL) was added isopentyllithium (0.34 M in 3 : 2 pentane / ether, 580 mL)

dropwise at 0 °C. The solution was stirred overnight at 0 °C and acetone (29 mL,

0.39 mmol) was added at 0 °C. The solution was stirred 5 min at 0 °C and

poured onto saturated aqueous NaHCO3. The aqueous phase was extracted

with CH2Cl2 and the combined organic layers were dried (Na2SO4) and

concentrated. The crude material was purified by silica-gel chromatography

(ethyl acetate-hexanes, 1 : 9, silica-gel deactivated with 10% triethylamine) to

1 give a colorless oil (9.0 mg, 74%); H NMR (300 MHz, CDCl3) d 0.82 (d, J = 6.2

Hz, 6 H) 1.24 (t, J = 7.4 Hz, 3 H), 1.39-1.55 (m, 3 H), 1.57-1.69 (m, 1 H) 1.74

(tdd, J = 3.6, 6.9, 10.3 Hz, 1H), 1.87 (ddd, J = 3.5, 6.9, 11.7 Hz, 1 H), 2.01 (dddd,

J = 2.5, 4.7, 10.3, 11.7 Hz, 1 H) 2.24 (dt, J = 8.5, 14.1 Hz, 1 H), 2.39-2.55 (m, 4

H), 2.61 (dq, J = 7.4, 12.8 Hz, 1 H), 2.68 (dq, J = 7.4, 12.8 Hz, 1 H) 5.09 (d, J =

5.7 Hz, 1 H), 5.76 (d, J = 3.5 Hz, 1 H), 7.39 (d, J = 2.5 Hz, 1 H); 13C NMR (75

MHz, CDCl3) d 14.9, 22.3, 22.4, 26.1, 27.8, 29.1, 32.2, 33.7, 35.6, 35.7, 46.1,

50.9, 91.5, 103.6, 122.9, 155.3, 198.0; IR (neat) 1658, 1591 cm-1; HRMS (M+H+) calcd for C11H27O3S 311.1681, found 311.1667.

160

euplotin A. To a solution of ketone 132 (23 mg, 0.074 mmol) in CH3CN (2 mL)

was added mercuric acetate (47 mg, 0.15 mmol). The solution was stirred for 1.5

h at rt and poured onto CH2Cl2 and H2O. The aqueous layer was extracted with

CH2Cl2 and the combined organic layers were dried (Na2SO4) and concentrated

to give a mixture of diastereomers (a : b, 5.5 : 1). Purification of the crude

residue by silica-gel flash chromatography (ethyl acetate-hexanes, 1 : 9) gave euplotin A as a colorless oil (15 mg, 67%): major isomer: 1H NMR (610 MHz,

CDCl3) d 0.88 (d, J = 6.4 Hz, 6 H), 1.49-1.61 (m, 3 H), 1.71 (dddd, J = 8.5, 10.6,

11.8, 12.2 Hz, 1 H), 1.93 (dddd, J = 3.1, 7.2, 10.6, 13.9 Hz, 1 H), 1.98 (dddd, J =

2.4, 4.8, 11.8, 11.8 Hz, 1 H), 2.03 (ddd, J = 3.3, 6.9, 11.8 Hz, 1 H), 2.11 (s, 3 H),

2.35 (dt, J = 8.5, 13.9 Hz, 1 H), 2.54 (td, J = 3.8, 7.2 Hz, 2 H), 2.61 (ddd, J = 4.8,

6.9, 12.2 Hz, 1 H), 2.70 (ddd, J = 3.1, 6.9, 9.5 Hz, 1 H), 5.92 (d, J = 3.3 Hz, 1 H),

13 6.06 (d, J = 3.1 Hz, 1 H), 7.47 (d, J = 2.4 Hz, 1 H); C NMR (75 MHz, CDCl3) d

21.2, 22.3, 22.4, 27.8, 29.3, 30.9, 33.6, 35.7, 36.2, 45.9, 50.2, 103.8, 104.6,

122.9, 155.1, 169.9, 197.8; IR (neat) 1749, 1658, 1592 cm-1; HRMS (M+H+) calcd

1 for C17H25O5 309.1706, found 309.1702. minor isomer: H NMR (300 MHz,

CDCl3) d 0.89 (d, J = 6.2 Hz, 6 H), 1.44 - 1.61 (m, 3 H), 1.76 (ddd, J = 9.5, 12.1,

13.2 Hz, 1 H), 1.89 - 2.06 (m, 3 H), 2.10 (s, 3 H), 2.26 (ddd, J = 2.6, 5.3, 10.6 Hz,

1 H), 2.51 (td, J = 2.1, 7.2 Hz, 2 H), 2.57 (p, J = 6.5 Hz, 1 H), 2.94 (ddd, J = 3.1,

7.2, 15.4 Hz, 1 H), 5.66 (d, J = 4.4 Hz, 1 H), 6.25 (d, J = 7.2 Hz, 1 H), 7.46 (d, J =

161

13 2.6 Hz, 1 H); C NMR (75 MHz, CDCl3) d 21.2, 22.3, 22.4, 25.8, 27.8, 28.8, 33.7,

35.3, 35.7, 42.8, 50.3, 96.9, 103.4, 123.6, 155.5, 169.5, 198.0; IR (neat) 1751,

-1 + 1643 cm ; HRMS (M+Na ) calcd for C17H24O5Na 331.1521, found 331.1494.

dihydropyran 186. A solution of enal 48 (40 mg, 0.20 mmol) and enamide 184

(68 mg, 0.60 mmol) in CH2Cl2 (400 mL) were pressurized at 12 kbar for 48 h then concentrated. The crude material was purified by silica-gel chromatography

1 (EtOAc- CH2Cl2, 3 : 1) to provide a colorless oil (52 mg, 83%): H NMR (400

MHz, CDCl3) d 0.87 (t, J = 7.1 Hz, 3 H), 0.93 (t, J = 7.5 Hz, 3 H), 1.18-1.31 (m, 5

H), 1.41-1.61 (m, 5 H), 1.73 (ddd, J = 5.3, 6.8, 12.0 Hz, 1 H), 2.04 (s, 3 H), 2.76

(q, J = 5.0 Hz, 1 H), 3.23 (s, 3 H), 3.64 (s, 3 H), 5.58 (dd, J = 6.9, 9.6 Hz, 1 H),

6.27 (d, J = 9.5, 1 H), 6.98 (d, J = 1.6 Hz, 1 H).

162

2,2-dimethyl-4-(3-methylbut-3-enyl)-1,3-dioxan-5-one (188). To a solution of

15 2,2-dimethyl-1,3-dioxan-5-one (21) (5.00 g, 38.4 mmol) in CHCl3 (40 mL) were

added Na2SO4 (20 g) and cyclohexylamine (8.80 mL, 76.8 mmol). The solution

was stirred at rt overnight, filtered, and concentrated to yield the crude

cyclohexylimine 33. The imine was dissolved in THF (40 mL) and added

dropwise to a solution of lithium diethylamide [formed by addition of n-BuLi (2.5

M in hexane, 15.2 mL, 38 mmol) to a solution of diethylamine (4.37 mL, 42.2.0

mmol) in THF (43 mL) at –35 °C] at –78 °C. The solution was warmed to –35 °C

over 2 h, recooled to –78 °C and a solution of 1-iodo-3-methyl-3-butene (7.46 g,

38.1 mmol) was added. The mixture was warmed to rt over 2 h and quenched

with saturated aqueous NH4Cl. This mixture was stirred at rt overnight and

extracted with Et2O. The combined extracts were washed with brine, dried

(Na2SO4) and concentrated. Purification of the crude residue by silica-gel

chromatography (ethyl acetate-hexane, 1 : 32) gave a colorless oil (6.73 g, 83%);

1 H NMR (360 MHz, CDCl3) d 1.41 (s, 3 H) 1.44 (s, 3 H), 1.58-1.69 (m, 2 H), 1.72

(s, 3 H), 2.04 (ddd, J = 3.5, 7.1, 9.2 Hz, 1 H), 2.10-2.19 (m, 2 H), 3.95 (d, J = 16.9

Hz, 1 H), 4.20 (ddd, J = 1.5, 3.3, 8.8 Hz, 1 H), 4.25 (dd, J = 1.5, 16.9 Hz, 1 H),

13 4.71 (dq, J = 0.9, 13.2 Hz, 2 H); C NMR (90 MHz, CDCl3) d 22.2, 23.6, 24.0,

26.2, 32.9, 66.6, 73.8, 100.8, 110.9, 144.6, 145.7, 209.8; IR (neat) 2935, 1748

cm-1.

163

2,2-dimethyl-4-(3-methylbut-3-enyl)-4H-1,3-dioxin-5- trifluoromethanesulfonate (189). A solution of ketone 188 (6.73 g, 31.78 mmol) and N-phenyltrifluoromethanesulfonimide (12.5 g, 34.9 mmol) in THF (100 mL) was added dropwise over 2 h to a solution of NaHMDS (1.0 M in THF, 34.9 mL, 34.9 mmol) in THF (82 mL) at –78 °C. The mixture was warmed slowly over

2 h to rt and poured onto saturated aqueous NaHCO3. The aqueous layer was extracted with hexane and the combined extracts were washed with saturated aqueous Na2CO3, dried (Na2SO4), and concentrated. The crude material was purified by Florisil chromatography (hexane, Florisil deactivated with 10%

1 triethylamine) to give a colorless oil (9.03 g, 93%); H NMR (360 MHz, CDCl3) d

1.51 (s, 3 H), 1.53 (s, 3 H), 1.72 (ddt, J = 6.5, 8.2, 15.9 Hz, 1 H), 1.75 (s, 3 H),

1.97 (dtd, J = 2.1, 8.3, 14.0 Hz, 1 H), 2.12-2.18 (m, 2 H), 4.47 (ddd, J = 1.4, 3.0,

7.6 Hz, 1 H), 4.73 (dq, J = 0.9, 2.1, 14.1 Hz, 2 H), 6.73 (d, J = 1.4 Hz, 1 H); 13C

NMR (90 MHz, CDCl3) d 20.7, 22.3, 27.2, 28.4, 31.6, 32.0, 67.9, 101.4, 110.5,

118.0 (q, J = 321 Hz), 131.5, 137.8, 144.6; IR (neat) 1681, 1652 cm-1.

164

N-methoxy-N-methyl-2,2-dimethyl-4-(3-methylbut-3-enyl)-4H-1,3-dioxin-5- carboxamide (191). To a solution of triflate 189 (3.00 g, 9.09 mmol) in THF (35 mL) were added dppp (375 mg, 0.91 mmol), N,O-dimethylhydroxylamine (2.8 g,

45.5 mmol) and Et3N (634 mL, 4.55 mmol). CO was bubbled through this solution for 0.5 h at 0 °C then palladium(II) acetate (43 mg, 0.19 mmol) was added. The solution was placed under a CO atmosphere (500 psi) and stirred at rt overnight.

The mixture was poured onto a saturated aqueous NaHCO3 and extracted with

Et2O. The combined extracts were washed with brine, dried (Na2SO4) and concentrated to give a 3 : 1 mixture of the desired amide 191 and Heck product

190. Purification of the crude residue by silica-gel chromatography (ethyl acetate-hexanes, 1 : 4, silica-gel deactivated with 10% triethylamine) gave the major product as a colorless oil (6.55 g, 72%): major product: 1H NMR (360

MHz, CDCl3) ? 1.44 (s, 1 H), 1.45 (s, 3 H), 1.56 (tdd, J = 6.8, 8.1, 10.9 Hz, 1 H),

1.67 (s, 3 H), 1.82 (tdd, J = 2.6, 8.5, 13.7 Hz, 1 H), 2.09 (t, J = 7.6 Hz, 2 H), 3.19

(s, 3 H), 3.60 (s, 3 H), 4.57 (ddd, J = 1.5, 2.6, 7.0 Hz, 1 H), 4.63 (dq, J = 1.1, 4.2

13 Hz, 2 H), 7.21 (d, J = 1.4 Hz, 1 H); C NMR (90 MHz, CDCl3) ? 21.6, 22.4, 27.7,

30.7, 32.8, 33.0, 60.7, 67.8, 100.2, 109.8, 110.7, 145.4, 167.1; IR (neat) 1651,

-1 1 1614 cm : minor Heck product: H NMR (360 MHz, CDCl3) d 1.23 (s, 3 H), 1.40

(s, 3 H), 1.45 (s, 3 H), 1.53 (tdd, J = 8.3, 9.5, 11.0 Hz, 1 H), 1.91 (dt, J = 6.6, 11.0

Hz, 1 H), 2.03 (ddd, J = 1.5, 6.6, 11.0 Hz, 1 H), 2.42 (d, J = 14.8 Hz, 1 H), 2.51 165

(d, J = 14.8 Hz, 1 H), 3.11 (s, 3 H), 3.62 (s, 3 H), 4.55 (ddd, J = 1.8, 6.9, 9.0 Hz,

13 1 H), 6.17 (d, J = 1.8 Hz, 1 H); C NMR (90 MHz, CDCl3) d 22.0, 27.8, 27.9,

29.7, 36.9, 40.0, 43.0, 60.9, 71.9, 99.1, 122.9, 134.6; IR (neat) 1681, 1650 cm-1.

N-methoxy-N-methyl-2-formyl-6-methylhepta-2,6-dienamide (172). A solution of amide 191 (1.40 g, 5.21 mmol) in toluene (52 mL) was heated at 100

°C for 1 h. The solution was concentrated and the crude material was purified by silica-gel chromatography (ethyl acetate-hexane, 1 : 3) to give a colorless oil (958 mg, 87%) as an inseparable mixture of stereoisomers (Z : E, 97 : 3); 1H NMR

(360 MHz, CDCl3) d 1.71 (s, 3 H), 2.21 (t, J = 7.4 Hz, 2 H), major 2.50 (q, J = 7.5

Hz, 2 H), minor 2.88 (qd, J = 5.2, 6.3 Hz, 2 H), 3.27 (s, 3 H), 3.54 (s, 3 H), 4.73

(br d, J = 19.3 Hz, 2 H), major 6.71 (t, J = 7.4 Hz, 1 H), minor 6.75 (td, J = 0.5,

13.2 Hz, 1 H), major 9.40 (s, 1 H), minor 9.69 (d, J = 0.5 Hz, 1 H); 13C NMR (90

MHz, CDCl3) d 22.0, 27.9, 35.7, 61.4, 111.1, 140.4, 143.4, 154.7, 165.5,

189.8;IR(neat) 1694, 1644 cm-1.

166

ethyl 2-cyanohex-5-enoate (202). To a solution of ethyl acetonitrile (200) (5.8

mL, 54.6 mmol) in THF (55 mL) was added NaH (60% in mineral oil, 1.46 g, 36.4

mmol). The mixture was stirred at 0 °C for 1 h, then 1-iodo-3-butene (201) was

added. The solution was stirred at reflux overnight, poured onto a saturated

aqueous NaHCO3, and extracted with Et2O. The combined extracts were

washed with brine, dried (Na2SO4) and concentrated to give a 3 : 1 mixture of the

desired nitrile 202 and bisalkylated nitrile. Purification of the crude residue by

silica-gel chromatography (ethyl acetate-hexanes, 1 : 12) gave the major product

1 as a colorless oil (4.35 g, 72%): major product: H NMR (360 MHz, CDCl3) d

1.29 (t, J = 6.6 Hz, 3 H), 1.97-2.13 (m, 2 H), 2.26 (m, 2 H), 3.50 (dd, J = 6.4, 7.8

Hz, 3 H), 4.23 (q, J = 6.6 Hz, 2 H), 5.07 (ddd, J = 1.1, 2.2, 10.2 Hz, 1 H), 5.11

(dq, J = 1.1, 17.2 Hz, 1 H), 5.72 (ddt, J = 6.8, 10.2, 17.2 Hz, 1 H); 13C NMR (90

MHz, CDCl3) d 13.9, 28.8, 30.5, 36.6, 62.7, 116.2, 117.1, 135.3, 165.9; IR (neat)

2983, 1747 cm-1.

2-(hydroxymethyl)hex-5-enenitrile. To a solution of nitrile 202 (3.80 g, 22.7

mmol) in THF (44 mL) at 0 °C was added NaBH4 (2.58 g, 68.2 mmol) and LiCl

(2.89 g, 68.2 mmol) followed by slow addition of MeOH (44 mL). The mixture

was stirred at 0 °C for 1 h then warmed to rt overnight. The mixture was poured

onto a saturated aqueous NH4Cl and extracted with Et2O. The combined

167

extracts were washed with brine, dried (Na2SO4) and concentrated. Purification

of the crude residue by filtration through a plug of silica-gel with Et2O gave a

1 colorless oil (1.86 g, 66%): H NMR (360 MHz, CDCl3) d 1.61-1.75 (m, 2 H), 2.13

(dq, J = 7.3, 14.6 Hz, 1 H), 2.24 (dq, J = 6.4, 14.6 Hz, 1 H), 2.71 (ddt, J = 5.9,

9.2, 11.8 Hz, 1 H), 3.55 (br s, 3 H), 3.67 (br d, J = 6.0 Hz, 2 H), 4.98 (ddd, J =

0.6, 2.2, 11.2 Hz, 1 H), 5.04 (dq, J = 0.6, 17.2 Hz, 1 H), 5.72 (ddt, J = 6.4, 11.2,

13 17.2 Hz, 1 H); C NMR (90 MHz, CDCl3) d 27.3, 28.9, 30.7, 34.0, 61.9, 116.2,

120.7, 136.1; IR (neat) 3441, 2244 cm-1.

2-[(tert-butyldimethylsilyloxy)methyl]hex-5-enenitrile (203). To a solution of

the above alcohol (1.86 g, 14.9 mmol) in DMF (28 mL) was added imidazole

(2.43 g, 35.7 mmol) and TBSCl (2.70 g, 17.9 mmol). The mixture was stirred at rt

for 1 h, poured onto a saturated aqueous NaHCO3 and extracted with hexanes.

The combined extracts were washed with brine, dried (Na2SO4) and concentrated. Purification of the crude residue by filtration through a plug of silica-gel with hexane gave a colorless oil (2.99 g, 84%): 1H NMR (360 MHz,

CDCl3) d 0.09 (s, 6 H), 0.91 (s, 9 H), 1.72 (ddd, J = 5.8, 7.7, 17.6 Hz, 1 H), 1.75

(ddd, J = 5.8, 7.7, 17.6 Hz, 1 H), 2.21 (dqt, J = 1.3, 7.8 16.8 Hz, 1 H), 2.31 (ddt, J

= 1.3, 6.2, 16.8 Hz, 1 H), 2.71 (dq, J = 6.0, 8.8 Hz, 1 H), 3.71 (dd, J = 5.9, 9.9 Hz,

1 H), 3.74 (dd, J = 6.2, 9.9 Hz, 1 H), 5.05 (dq, J = 1.2, 10.2 Hz, 1 H), 5.11 (dq, J

= 1.7, 15.4 Hz, 1 H), 5.72 (ddt, J = 6.2, 10.2, 15.4 Hz, 1 H); 13C NMR (90 MHz,

168

CDCl3) d -5.5, -5.4, -3.0, 18.2, 25.7, 27.7, 30.9, 34.3, 62.9, 116.3, 120.6; IR (neat)

2955, 2360 cm-1.

2-[(tert-butyldimethylsilyloxy)methyl]hex-5-enal (204). To a solution of nitrile

203 (3.56 g, 14.9 mmol) in CH2Cl2 (30 mL) at –78 °C was added Dibal-H (1.5 M in toluene, 14.9 mL, 22.3 mmol). The mixture was stirred to rt to over 1 h, poured onto a saturated aqueous NaHCO3 and saturated aqueous Rochelle’s Salt, extracted with CH2Cl2. The combined extracts were washed with saturated aqueous Rochelle’s Salt, dried (Na2SO4) and concentrated. Purification of the crude residue by filtration through a plug of silica-gel with CH2Cl2 gave a colorless

1 oil (2.74 g, 76%): H NMR (360 MHz, CDCl3) d 0.01 (s, 6 H), 0.84 (s, 9 H), 1.53

(ddd, J = 6.5, 7.8, 15.6 Hz, 1 H), 1.76 (dq, J = 7.5, 15.6 Hz, 1 H), 2.21 (dqt, J =

1.3, 7.8 16.8 Hz, 1 H), 2.31 (ddt, J = 1.3, 6.2, 16.8 Hz, 1 H), 2.06 (qt, J = 1.3, 7.6

Hz, 1 H), 2.41 (dddd, J = 2.1, 4.8, 8.4, 13.4 Hz, 1 H), 3.80 (dd, J = 6.3, 10.3 Hz, 1

H), 3.85 (dd, J = 4.8, 10.3 Hz, 1 H), 4.94 (dq, J = 1.2, 10.2 Hz, 1 H), 5.01 (dq, J =

1.2, 15.4 Hz, 1 H), 5.73 (ddt, J = 6.2, 10.2, 15.4 Hz, 1 H); 13C NMR (90 MHz,

CDCl3) d -5.7, -5.6, -5.3, 18.3, 24.5, 25.7, 30.9, 53.3, 61.7, 115.3, 137.7, 204.4;

IR (neat) 1728, 1642 cm-1.

169

(E)-2-cyanoethyl 4-[(tert-butyldimethylsilyloxy)methyl]octa-2,7-dienenoate

(207). To a solution of the propionitrile Horner-Emmons reagent (2.23 g, 10.1 mmol) in THF (20 mL) at 0 °C was added NaH (60% in mineral oil, 370 mg, 9.30 mmol). The ylide was stirred at rt for 1 h, then aldehyde 204 (2.04 g, 8.42 mmol) in THF (9 mL) was. The mixture was stirred at rt for 1 h, poured onto a saturated aqueous NH4Cl and extracted with Et2O. The combined extracts were dried

(Na2SO4) and concentrated. Purification of the crude residue by filtration through

1 a plug of silica-gel with Et2O gave a colorless oil (2.67 g, 94%): H NMR (360

MHz, CDCl3) d 0.02 (s, 6 H), 0.85 (s, 9 H), 1.45 (tdd, J = 5.8, 8.9, 14.7 Hz, 1 H),

1.63 (dddd, J = 5.0, 6.7, 9.2, 15.9 Hz, 1 H), 1.92-2.12 (m, 2 H), 2.40 (ddt, J = 5.4,

11.1, 14.1 Hz, 1 H), 2.72 (t, J = 6.3 Hz, 2 H), 3.56 (dd, J = 6.2, 10.0 Hz, 1 H),

3.61 (dd, J = 5.7, 10.0 Hz, 1 H), 4.33 (t, J = 6.4 Hz, 2 H), 4.92-5.05 (m, 2 H), 5.75

(ddt, J = 6.3, 10.2, 15.4 Hz, 1 H), 5.86 (dd, J = 0.9, 15.8 Hz, 1 H), 6.89 (dd, J =

13 8.9, 15.8 Hz, 1 H); C NMR (90 MHz, CDCl3) d -5.6, -5.5, -5.4, 18.0, 18.1, 25.8,

29.3, 31.0, 44.5, 58.4, 65.2, 114.9, 116.7, 137.9, 165.6; IR (neat) 2254, 1732,

1651 cm-1.

170

(E)-4-[(tert-butyldimethylsilyloxy)methyl]octa-2,7-dienenoic acid (208). To a

solution of ester 207 (2.67 g, 7.92 mmol) in THF (30 mL) at 0 °C was added KOt-

Bu (1.0 M in THF, 8.70 mL, 8.70 mmol) all at once. The mixture was stirred at 0

°C for 0.25 h, poured onto a saturated aqueous KH2PO4 filtered through celite

and the aqueous phase extracted with EtOAc. The combined extracts were dried

(Na2SO4) and concentrated. Purification of the crude residue by filtration through

1 a plug of silica-gel with Et2O gave a colorless oil (2.23 g, 99%): H NMR (360

MHz, CDCl3) d 0.03 (s, 6 H), 0.87 (s, 9 H), 1.45 (tdd, J = 5.8, 9.0, 14.7 Hz, 1 H),

1.63 (m, 1 H), 1.96-2.14 (m, 2 H), 2.41 (ddt, J = 5.4, 11.1, 14.7 Hz, 1 H), 3.57

(dd, J = 6.1, 9.8 Hz, 1 H), 3.61 (dd, J = 5.7, 9.8 Hz, 1 H), 4.92-5.05 (m, 2 H), 5.76

(ddt, J = 6.6, 10.2, 16.9 Hz, 1 H), 5.85 (d, J = 15.8 Hz, 1 H), 6.93 (dd, J = 8.9,

13 15.8 Hz, 1 H); C NMR (90 MHz, CDCl3) d -5.5, -5.4, -5.3, 18.2, 25.8, 29.4, 30.5,

31.1, 31.2, 44.5, 65.3, 115.0, 138.1, 153.6; IR (neat) 3077, 1698, 1652 cm-1.

(E)-ethyl 4-[(tert-butyldimethylsilyloxy)methyl]hepta-1,6-dienylcarbamate

(215). To a solution of acid 208 (408 mg, 1.43 mmol) in acetone (3.0 mL) at 0 °C

was added DIPEA (312 mL, 1.79 mmol). The mixture was stirred at 0 °C for 0.25

171

h and methyl chloroformate (122 mL, 1.57 mmol) was added. The mixture was

stirred at 0 °C for 0.25 h and NaN3 (186 mg, 2.86 mmol) in H2O (475 mL) was

added. The solution was poured onto H2O, extracted with CH2Cl2. The combined

extracts were dried (Na2SO4) and toluene (1.5 mL) was added. The solvent was

reduced on a rotovap only, until only toluene was left. This crude acyl azide was

added dropwise to a refluxing solution of EtOH (417 mL, 7.15 mmol), BHT (cat) in

toluene (7 mL). The mixture was refluxed for 3 h and concentrated. The crude

material was purified by silica-gel chromatography (EtOAc-hexanes, 1 : 9) to give

1 a colorless oil (320 mg, 69%): H NMR (360 MHz, CDCl3) d 0.02 (s, 6 H), 0.89 (s,

9 H), 1.23-1.35 (m, 2 H), 1.61 (ddd, J = 6.1, 7.1, 13.2 Hz, 1 H), 1.98 (dq, J = 7.0,

14.7 Hz, 1 H), 2.04-2.19 (m, 2 H), 3.48 (dd, J = 6.3, 9.8 Hz, 1 H), 3.61 (dd, J =

5.8, 9.8 Hz, 1 H), 4.15 (q, J = 7.1 Hz, 2 H), 4.79 (dd, J = 9.3, 13.8 Hz, 1 H), 4.93

(dt, J = 0.9, 10.2 Hz, 1 H) 5.00 (dq, J = 1.6, 17.1 Hz, 1 H), 5.78 (ddt, J = 6.6,

10.2, 13.2 Hz, 1 H), 6.32 (d, J = 10.1 Hz, 1 H), 6.49 (dd, J = 11.4, 13.5 Hz, 1 H);

13 C NMR (90 MHz, CDCl3) d -5.4, -5.3, 14.5, 18.3, 25.9, 30.7, 31.2, 42.4, 61.2,

67.1, 111.5, 114.4, 124.1, 138.9, 153.7; IR (neat) 3310, 1704, 1678 cm-1.

(E)-3-{3-[(tert-butyldimethylsilyloxy)methyl]hepta-1,6-dienyl}-1,1- dimethylurera (216). To a solution of acid 208 (202 mg, 0.71 mmol) in acetone

(1.0 mL) at 0 °C was added DIPEA (155 mL, 0.89 mmol). The mixture was stirred

172

at 0 °C for 0.25 h and methyl chloroformate (61 mL, 0.78 mmol) was added. The

mixture was stirred at 0 °C for 0.25 h and NaN3 (92 mg, 1.42 mmol) in H2O (250

mL) was added. The solution was poured onto H2O, extracted with CH2Cl2. The

combined extracts were dried (Na2SO4) and toluene (1 mL) was added. The solvent was reduced on a rotovap only, until only toluene was left. This crude acyl azide was added dropwise to a refluxing solution of BHT (cat) in toluene (7 mL). The mixture was refluxed for 3 h, cooled and dimethylamine (40% in H2O,

446 mL, 3.55 mmol) was added. The solution was stirred for 0.5 h and concentrated. The crude material was purified by silica-gel chromatography

(EtOAc-hexanes, 1 : 9) to give a colorless oil (195 mg, 84%).

(E)-1-{3-[(tert-butyldimethylsilyloxy)methyl]hepta-1,6-dienyl}-1,3,3- trimethylurera (218). Urea 216 (47 mg, 0.15 mmol) was dissolved in THF (500 mL) and NaH (60 % in mineral oil, 7.0 mg, 0.17 mmol) was added at 0 °C. The solution was stirred 0.25 h and iodomethane (27 mL, 0.43 mmol) was added. The

mixture was stirred at rt for 2 h, poured onto a saturated aqueous NaHCO3 and

extracted with Et2O. The combined extracts were dried (Na2SO4) and

concentrated. Purification of the crude residue by silica-gel chromatography

(EtOAc-hexanes, 1 : 9) gave a colorless oil (35 mg, 72%)1H NMR (360 MHz,

CDCl3) d 0.02 (s, 6 H), 0.87 (s, 9 H), 1.27 (tdd, J = 5.3, 9.4, 13.5 Hz, 1 H), 1.65

(tdd, J = 4.4, 6.8, 13.5 Hz, 1 H), 1.98 (dq, J = 6.6, 14.4 Hz, 1 H), 2.05-2.19 (m, 2 173

H), 2.84 (s, 6 H), 2.96 (s, 3 H), 3.47 (dd, J = 6.5, 9.8 Hz, 1 H), 3.51 (dd, J = 5.9,

9.8 Hz, 1 H), 4.50 (dd, J = 9.2, 14.2 Hz, 1 H), 4.92 (ddd, J = 0.9, 2.0, 10.2 Hz, 1

H), 4.93 (dq, J = 1.6, 17.1 Hz, 1 H), 5.78 (ddt, J = 6.6, 10.2, 13.1 Hz, 1 H), 6.48

13 (d, J = 14.2 Hz, 1 H); C NMR (90 MHz, CDCl3) d -5.4, -5.3, 18.3, 25.9, 31.1,

31.3, 33.4, 38.8, 43.1, 67.6, 108.2 114.2, 162.4; IR (neat) 3310, 1704, 1678 cm-1.

cycloadduct 220. A solution of enal 172 (250 mg, 1.18 mmol) and enamide 215

(387 mg, 1.18 mmol) in CH2Cl2 (1 mL) were pressurized at 12 kbar for 48 h then

concentrated. The crude material was purified by silica-gel chromatography

(EtOAc-hexanes, 1 : 3) to provide a colorless oil as a 2.3 : 1 mixture of

1 diastereomers (462 mg, 73%): H NMR (400 MHz, CDCl3) d minor 0.04 (s, 3 H), minor 0.06 (s, 3 H), major 0.09 (s, 3 H), major 0.10 (s, 3 H), minor 0.88 (s, 9 H), major 0.92 (s, 9 H), 1.23 (t, J = 7.1 Hz, 3 H), minor 1.32-1.41 (m, 1 H), major

1.43-1.52 (m, 1 H), 1.54-1.65 (m, 1 H), major 1.66 (s, 3 H), minor 1.69 (s, 3 H),

1.73-1.82 (m, 1 H), 1.89-2.14 (m, 3 H), 2.18 (td, J = 2.4, 8.8 Hz, 3 H), 2.83 (br dt, J = 3.6 Hz, 1 H), 3.23 (s, 3 H), minor 3.52 (dd, J = 6.8, 12.8 Hz, 1 H), major

3.59 (dd, J = 7.1, 11.9 Hz, 1 H), minor 3.64 (s, 3 H), major 3.66 (s, 3 H), 3.69 (dd,

J = 5.3, 10.3 Hz, 1 H), 4.15 (q, J = 6.9 Hz, 2 H), major 4.64 (d, J = 15.0 Hz, 2 H), minor 4.67 (d, J = 14.2 Hz, 2 H), 4.92 (d, J = 9.6 Hz, 1 H), 4.99 (dq, J = 1.4, 17.0

174

Hz, 1 H), 5.15 (t, J = 8.9 Hz, 1 H), minor 5.48 (dd, J = 8.1, 9.7 Hz, 1 H), minor

5.57 (br s, 1 H), 5.74 (ddt, J = 6.6, 10.2, 13.1 Hz, 1 H), major 6.39 (d, J = 7.6, 1

H), minor 7.06 (d, J = 1.0 Hz, 1 H), major 7.12 (d, J = 0.8 Hz, 1 H). 13C NMR (90

MHz, CDCl3) d -5.6, -5.58, -5.53, -5.46, 14.4, 18.2, 18.3, 22.5, 22.6, 25.3, 25.8,

25.9, 26.0, 28.0, 29.4, 32.0, 32.1, 32.2, 32.7, 33.7, 33.8, 35.2, 38.6, 39.3, 39.9,

41.7, 60.82, 60.85, 61.2, 64.3, 79.6, 80.3, 109.8, 110.3, 113.2, 114.9, 115.3,

137.9, 138.2, 145.2, 145.5, 147.4, 148.4, 155.4, 155.6, 168.7, 168.9; IR (neat)

3306, 1734, 1648 cm-1.

tert-butyl-(4-butyl-2,2-dimethyl-4H-[1,3]dioxin-5-yloxy)dimethylsilane

(238iii). A solution of ketone 32i (500 mg, 2.69 mmol) in THF (27 mL) was

added dropwise over 1.5 h to a solution of NaHMDS (1.0 M in THF, 3.50 mL,

3.50 mmol) in THF (18 mL) at –78 °C. TBSCl (526 mg, 3.50 mmol) was added

and the solution was warmed to rt over 2 h. Saturated aqueous NaHCO3 was

added and the aqueous layer extracted with hexane. The combined organic

layers were washed with saturated aqueous Na2CO3, dried (Na2SO4) and

concentrated. Purification of the crude residue by silica-gel chromatography

(hexane, silica-gel deactivated with 10% triethylamine) yielded a colorless oil

(743 mg, 92%): 1H NMR (200 MHz, CDCl3) d 0.12 (s, 3 H), 0.14 (s, 3 H), 0.90 (t,

J = 8.0 Hz, 3 H), 0.91 (s, 9 H), 1.23-1.50 (m, 4 H), 1.42 (s, 3 H), 1.45 (s, 3 H),

175

1.57 (ddd, J = 2.2, 7.0, 14.4 Hz, 1 H), 1.70 (dtd, J = 3.2, 7.2, 14.4 Hz, 1 H), 4.10

(ddd, J = 1.1, 3.2, 7.0 Hz, 1 H), 6.10 (d, J = 1.1 Hz, 1 H); 13C NMR (50 MHz,

CDCl3) d -4.7, -4.4, 14.0, 17.9, 20.9, 22.7, 25.6, 26.5, 27.8, 30.7, 69.9, 98.3,

-1 + 126.2, 134.8; IR (neat) 2997, 1471 cm ; HRMS (M+H ) calcd for C16H32O3Si

301.2199, found 301.2188.

2-(tert-butyldimethylsilyloxy)hept-2-enal (240iii). A solution of dioxin 238iii

(743 mg, 2.47 mmol) in toluene (20 mL) was heated at 110 °C for 2 h. The

solution was cooled and concentrated to give a colorless oil (615 mg, 99%); 1H

NMR (400 MHz, CDCl3) d 0.16 (s, 6H), 0.92 (t, J = 7.2 Hz, 3 H), 0.95 (s, 9 H),

1.37 (h, J = 7.2 Hz, 2 H), 1.45 (p, J = 7.2 Hz, 2 H), 2.34 (q, J = 7.2 Hz, 2 H), 5.75

13 (t, J = 7.2 Hz, 1 H), 9.12 (s, 1 H); C NMR (75 MHz, CDCl3) d -4.1, 13.8, 18.6,

22.5, 25.8, 30.6, 136.6, 151.4, 189.0; IR (neat) 1693, 1635 cm-1; HRMS (M+H+) calcd for C13H26O2Si 243.1780, found 243.1804.

2-butyl-4-(tert-butyldimethylsilyloxy)-8-oxabicyclo[3.2.1]oct-6-en-3-one

(entry a). Dimethyl aluminum chloride (1.0 M in hexane, 144 mL, 0.144 mmol)

176

was added dropwise to a solution of enal 240iii (35 mg, 0.144 mmol) and furan

(53 mL, 0.722 mmol) in CH2Cl2 (500 mL) at –78 °C. After stirring overnight at –78

°C, saturated aqueous NaHCO3 was added and the mixture was warmed to rt.

The aqueous layer was extracted with CH2Cl2 and the combined organic extracts

were dried (Na2SO4) and concentrated. Purification of the crude residue by

silica-gel chromatography (benzene-hexane, 2 : 3, silica-gel deactivated with

10% triethylamine) yielded a mixture (endo : exo, 77 : 23) as colorless oils (34

1 mg overall, 76%): endo isomer: H NMR (400 MHz, CDCl3) d 0.05 (s, 3 H), 0.14

(s, 3 H), 0.88 (t, J = 7.1 Hz, 3 H), 0.90 (s, 9 H), 1.02 (ddd, J = 5.2, 8.3, 9.0 Hz, 1

H), 1.29–1.40 (m, 2 H), 1.33 (q, J = 7.1 Hz, 2 H), 1.78 (ddd, J = 5.5, 11.0, 13.9

Hz, 1 H), 2.65 (dt, J = 5.2, 8.1 Hz, 1 H), 4.29 (dd, J = 0.5, 5.1 Hz, 1 H) 4.83 (dd, J

= 1.7, 5.1 Hz, 1 H), 4.90 (dd, J = 1.5, 4.6 Hz, 1 H), 6.27 (dd, J = 1.5, 6.1 Hz, 1 H),

13 6.33 (dd, J = 1.7, 6.1 Hz, 1 H); C NMR (75 MHz, CDCl3) d -5.5, -4.6, 13.9, 18.4,

22.8, 24.7, 25.7, 29.6, 55.3, 78.4, 81.4, 81.8, 133.2, 133.3, 206.3; IR (neat) 2955,

-1 + 1728 cm ; HRMS (M+Na ) calcd for C17H30O3SiNa 333.1862, found 333.1884;

1 exo isomer: H NMR (300 MHz, CDCl3) d 0.00 (s, 3 H), 0.08 (s, 3 H), 0.83 (t, J =

7.5 Hz, 3 H), 0.84 (s, 9 H), 1.17 – 1.40 (m, 2 H), 1.26 (q, J = 7.5 Hz, 2 H), 1.65 (q,

J = 6.8 Hz, 1 H), 2.24 (t, J = 7.7 Hz, 1 H), 4.31 (dd, J = 0.4, 5.1 Hz, 1 H), 4.70

(bs, 1 H), 4.75 (dd, J = 1.6, 5.1 Hz, 1 H), 6.21 (dd, J = 1.6, 6.1 Hz, 1 H), 6.25 (dd,

13 J = 1.6, 6.1 Hz, 1 H); C NMR (75 MHz, CDCl3) d -5.6, -4.6, 13.8, 18.4, 22.4,

25.7, 29.6, 30.0, 55.8, 77.9, 81.5, 81.6, 132.3, 134.7, 208.1; IR (neat) 2955, 1726

-1 + cm ; HRMS (M+Na ) calcd for C17H30O3SiNa 333.1862, found 333.1869.

177

2-butyl-4-(tert-butyldimethylsilyloxy)bicyclo[3.2.1]oct-6-en-3-one (entry c).

Tin(IV) chloride (1.0 M in CH2Cl2, 145 mL, 0.145 mmol), was added dropwise to a solution of enal 240iii (35 mg, 0.145 mmol) and cyclopentadiene (48 mg, 0.722 mmol) in CH2Cl2 (500 mL) at –78 °C. After stirring overnight at –78 °C, saturated aqueous NaHCO3 was added and the mixture was warmed to rt. The aqueous layer was extracted with CH2Cl2 and the combined organic extracts were dried

(Na2SO4) and concentrated to give a mixture of diastereomers (endo : exo, 80 :

20). Purification of the crude residue by silica-gel chromatography (benzene– hexane, 1 : 4, silica-gel deactivated with 10% triethylamine) gave a yellow oil (37

1 mg, 73%): endo isomer: H NMR (400 MHz, CDCl3) d 0.03 (s, 3 H), 0.13 (s, 3 H),

0.88 (t, J = 6.7 Hz, 3 H), 0.90 (s, 9 H), 1.09 (ddd, J = 5.6, 9.7, 18.5 Hz, 1 H),

1.25–1.39 (m, 4 H), 1.84 (d, J = 11.3 Hz, 1 H), 1.88 (dtd, J = 2.0, 5.2, 15.4 Hz, 1

H), 2.13 (p, J = 5.5 Hz, 1 H), 2.29 (ddd, J = 3.0, 4.8, 8.2 Hz, 1 H), 2.84 (ddd, J =

2.7, 3.0, 5.2 Hz, 1 H), 2.91 (p, J = 2.6 Hz, 1 H), 4.17 (dd, J = 0.4, 3.6 Hz, 1 H),

6.03 (dd, J = 2.6, 5.7 Hz, 1 H), 6.10 (dd, J = 2.7, 5.7 Hz, 1 H); 13C NMR (90 MHz,

CDCl3) d -5.5, -4.5, 13.9, 18.5, 22.8, 25.8 (3 C), 27.4, 29.8, 41.4, 43.2, 48.2, 54.9,

80.7, 135.1, 135.2, 209.1; IR (neat) 2952, 1725 cm-1; HRMS (M+H+) calcd for

1 C18H33O2Si 309.2249, found 309.2224. exo isomer: H NMR (400 MHz, CDCl3) d

0.03 (s, 3 H), 0.09 (s, 3 H), 0.87 (t, J = 6.2 Hz, 3 H), 0.88 (s, 9 H), 1.25-1.35 (m, 1

H), 1.60-1.65 (m, 1 H), 1.66 (ddt, J = 1.5, 5.1, 10.1 Hz, 1 H), 1.81 (dtd, J = 2.0,

178

5.1, 15.4 Hz, 1 H), 1.84 (d, J = 11.4 Hz, 1 H), 2.20 - 2.25 (m, 1 H), 2.53 (d, J =

11.4 Hz, 1 H), 2.63 (dt, J = 2.7, 5.1 Hz, 1 H), 2.72 (dd, J = 3.6, 7.8 Hz, 1 H), 3.69

(dt, J = 1.3, 3.5 Hz, 1 H), 5.87 (dd, J = 2.8, 5.8 Hz, 1 H), 6.16 (dd, J = 2.6, 5.8 Hz,

13 1 H); C NMR (90 MHz, CDCl3) d -5.2, -4.9, 13.9, 18.2, 22.6, 25.6 (3 C), 25.8,

29.8, 30.0, 30.2, 32.5, 42.1, 46.3, 54.7, 76.2, 135.1, 135.2, 210.5; IR (neat) 2954,

-1 + 1719 cm ; HRMS (M+H ) calcd for C18H33O2Si 309.2249, found 309.2224.

2-benzyl-4-(tert-butyldimethylsilyloxy)bicyclo[3.2.1]oct-6-en-3-one (entry e).

Dimethyl aluminum chloride (1.0 M in hexane, 200 mL, 0.199 mmol), was added dropwise to a solution of enal 242 (50 mg, 0.181 mmol) and cyclopentadiene (60 mg, 0.905 mmol) in CH2Cl2 (500 mL) at –78 °C. After stirring overnight at –78 °C, saturated aqueous NaHCO3 was added and the mixture was warmed to rt. The aqueous layer was extracted with CH2Cl2 and the combined organic extracts were dried (Na2SO4) and concentrated to give a single isomer. Purification of the crude residue by silica-gel chromatography (benzene–hexane, 1 : 4, silica-gel deactivated with 10% triethylamine) gave a colorless oil (37 mg, 63%): 1H NMR

(400 MHz, CDCl3) d 0.03 (s, 3 H), 0.13 (s, 3 H), 0.90 (s, 9 H), 1.78 (d, J = 11.5

Hz, 1 H), 2.07 (dt, J = 5.5, 11.5 Hz, 1 H), 2.28 (dd, J = 10.6, 14.2 Hz, 1 H), 2.62

(dt, J = 3.3, 10.7 Hz, 1 H), 2.67 (dt, J = 2.6, 5.3 Hz, 1 H), 2.93 (dt, J = 3.4, 5.3 Hz,

179

1 H), 3.35 (dd, J = 3.8, 14.2 Hz, 1 H), 4.23 (d, J = 3.4 Hz, 1 H), 6.15 (dd, J = 2.5,

5.8 Hz, 1 H) 6.18 (dd, J = 2.5, 5.7 Hz, 1 H), 7.14-7.31 (m, 5 H); 13C NMR (100

MHz, CDCl3) d -5.4, -4.5, 18.5, 25.8, 33.9, 41.2, 41.9, 48.3, 56.6, 80.8, 109.6,

126.0, 128.5, 129.1, 134.8, 135.8, 140.4, 208.4; IR (neat) 2929, 1724 cm-1;

+ HRMS (M+H ) calcd for C21H31O2Si 343.2093, found 343.2091.

2-benzyl-7-(tert-butyldimethylsilyloxy)cyclohept-4-enone (entry f). Ethyl aluminum dichloride (1.0 M in hexane, 140 mL, 0.140 mmol), was added dropwise to a solution of enal 242 (50 mg, 0.181 mmol) and 1,3-butadiene (3.60

M in CH2Cl2, 200 mL, 0.722 mmol) in CH2Cl2 (500 mL) at –78 °C. After stirring overnight at –78 °C, saturated aqueous NaHCO3 was added and the mixture was warmed to rt. The aqueous layer was extracted with CH2Cl2 and the combined organic extracts were dried (Na2SO4) and concentrated to give a single isomer.

Purification of the crude residue by silica-gel chromatography (benzene–hexane,

1 : 4, silica-gel deactivated with 10% triethylamine) gave a colorless oil (27 mg,

1 65%): H NMR (400 MHz, CDCl3) d -0.03 (s, 3 H), 0.09 (s, 3 H), 0.85 (s, 9 H),

1.98 (ddd, J = 3.5, 9.8, 13.1 Hz, 1 H), 4.40 (dd, J = 4.7, 7.8 Hz, 1 H), 5.76 (m, 2

13 H), 7.11-7.30 (m, 5 H); C NMR (100 MHz, CDCl3) d -5.5, -4.7, 18.4, 25.8, 29.2,

180

34.4, 36.6, 51.9, 78.1, 126.2, 126.9, 128.4, 129.1, 129.5, 139.9, 211.2; IR (neat)

-1 + 2928, 1724 cm ; HRMS (M+H ) calcd for C20H31O2Si 331.2093, found 331.2107.

2-butyl-7-(tert-butyldimethylsilyloxy)-4-methylcyclohept-4-enone (entry g).

Titanium(IV) chloride (1.0 M in hexane, 413 mL, 0.413 mmol) was added

dropwise to a solution of enal 240iii (100 mg, 0.413 mmol) and isoprene (206 mL,

2.06 mmol) in CH2Cl2 (1.50 mL) at –78 °C. After stirring overnight at –78 °C,

saturated aqueous NaHCO3 was added and the mixture was warmed to rt. The

aqueous layer was extracted with CH2Cl2 and the combined organic layers were

dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel

chromatography (benzene-hexane, 7 : 13, silica-gel deactivated with 10%

triethylamine) yielded an inseparable mixture of regioisomers (C(4)-methyl : C(5)- methyl, 89 : 11) as a colorless oil (69 mg, 54%): 1H NMR (400 MHz, d6-benzene) d minor 0.87 (s, 3 H), major 0.89 (s, 3 H), minor 0.23 (s, 3 H) major 0.25 (s, 3 H),

0.87 (t, J = 7.2 Hz, 3 H), minor 0.98 (s, 9 H), major 1.00 (s, 9 H) 1.08–1.27 (m, 6

H), major 1.49 (d, J = 0.7 Hz, 3 H), minor 1.62 (d, J = 1.2 Hz, 3 H), 1.78 (dd, J =

10.3, 15.9 Hz, 1 H), 1.83 (dd, J = 10.6, 15.9 Hz, 1 H), 1.85–1.92 (m, 1 H), 2.17

(dt, J = 6.4, 14.1 Hz, 1 H), 2.35 (dtd, J = 4.6, 8.1, 9.9 Hz, 1 H), 2.49 (dtd, J = 1.2,

5.0, 15.6 Hz, 1 H), major 4.18 (dd, J = 4.5, 6.6 Hz, 1 H), minor 4.24 (dd, J = 4.2,

8.0 Hz, 1 H), minor 5.29 (t, J = 6.8, 1 H), major 5.32 (t, J = 6.4 Hz, 1 H); 13C NMR

(75 MHz, d6-benzene) d -5.1, -3.9, 14.2, 18.8, 23.1, major 26.1, minor 26.7, 30.2, 181

minor 31.0, major. 31.3, 36.8, 39.4, major 48.9, minor 49.9, minor 78.7, major

79.6, major 120.8, minor 123.6, minor 135.0, major 138.1, 210.0; IR (neat) 2955,

-1 + 1723 cm ; HRMS (M+Na ) calcd for C18H34O2SiNa 333.2226, found 333.2223.

7-butyl-2-(tert-butyldimethylsilyloxy)-3-methylcyclohept-4-enone (entry i).

Ethyl aluminum dichloride (1.0 M in hexane, 158 mL, 0.158 mmol), was added dropwise to a solution of enal 240iii (35 mg, 0.144 mmol) and trans-piperylene

(72 mL, 0.722 mmol) in CH2Cl2 (500 mL) at –78 °C. After stirring overnight at –78

°C, saturated aqueous NaHCO3was added and the mixture was warmed to rt.

The aqueous layer was extracted with CH2Cl2 and the combined organic extracts were dried (Na2SO4) and concentrated to give a two diastereomers (endo : exo, 5

: 95). Purification of the crude residue by silica-gel chromatography (benzene– hexane, 1 : 9, silica-gel deactivated with 10% triethylamine) gave a colorless oil

1 (22 mg, 50%): exo isomer: H NMR (400 MHz, CDCl3) d -0.03 (s, 3 H), 0.09 (s, 3

H), 0.87 (t, J = 7.2 Hz, 3 H), 0.91 (s, 9H), 1.13 (d, J = 7.1 Hz, 3 H), 1.18-1.40 (m,

5 H), 1.61 (br d, J = 10.6 Hz, 1 H), 1.82 (ddd, J = 7.1, 13.4, 20.0 Hz, 1 H), 2.09

(m, 1 H), 2.22-2.41 (m, 3 H), 4.13 (d, J = 8.7 Hz, 1 H), 5.48 (ddd, J = 1.1, 3.4,

13 12.5 Hz, 1 H), 5.77 (m, 1H); C NMR (100 MHz, CDCl3) d -5.3, -4.6, 14.0, 18.4,

19.4, 22.7, 25.8, 28.8, 29.7, 30.4, 39.1, 52.1, 81.7, 128.0, 134.7, 212.3; IR (neat)

-1 + 1721, 1463 cm ; HRMS (M+H ) calcd for C18H35O2Si 311.2406, found 311.2414.

182

1 endo isomer: H NMR (400 MHz, CDCl3) d -0.01 (s, 3 H), 0.11 (s, 3 H), 0.87 (t, J

= 7.2 Hz, 3 H), 0.91 (s, 9H), 1.08 (d, J = 7.2 Hz, 3 H), 1.12-1.40 (m, 5 H), 1.63 (br d, J = 5.6 Hz, 1 H), 1.69 (d, J = 4.8 Hz, 1 H), 1.71-1.82 (m, 2 H), 2.04-2.12 (m, 2

H), 2.37 (dtt, J = 1.7, 5.0, 15.0 Hz, 1 H), 2.53 (ddd, J = 5.1, 7.1, 12.2 Hz, 1H),

2.82-2.91 (m, 1 H), 4.42 (d, J = 4.2 Hz, 1 H), 5.52 (dd, J = 5.6, 11.3 Hz, 1 H),

5.60 (dtd, J = 1.7, 5.6, 11.3 Hz, 1 H); IR (neat) 1725, 1462 cm-1; HRMS (M+H+) calcd for C18H35O2Si 311.2406, found 311.2428.

4-hepta-4,6-dienyl-2,2-dimethyl-1,3-dioxan-5-one. To a solution of 2,2- dimethyl-1,3-dioxan-5-one (21)15 (3.30 g, 25.4 mmol) in benzene (100 mL) was added cyclohexylamine (5.80 mL, 50.7 mmol) and molecular sieves (4Å, 3.50 g).

The mixture was stirred overnight at rt, filtered, and concentrated to yield the crude imine. The imine (3.53 g, 16.7 mmol) was dissolved in THF (32 mL) and added dropwise at –78 °C to a solution of lithium diethylamide [formed by addition of n-butyllithium (2.5 M in hexane, 6.27 mL, 15.7 mmol) to a solution of diethylamine (1.73 mL, 16.7 mmol) in THF (17 mL) at –35 °C]. The mixture was warmed to –35 °C over 2 h and recooled to –78 °C. 1-Iodohepta-4,6-diene (2.32 g, 10.4 mmol) in THF (20 mL) was added and the solution was warmed to rt over

2 h. A solution of saturated aqueous NH4Cl (60 mL) was added and the resultant mixture was stirred at rt overnight. The aqueous layer was extracted with Et2O and the combined extracts were washed with brine, dried (Na2SO4) and

183

concentrated. Purification of the crude residue by silica-gel chromatography

(ethyl acetate-hexane, 3 : 97) gave a colorless oil (1.93 g, 82%): 1H NMR (360

MHz, CDCl3) d 1.43 (s, 3 H), 1.46 (s, 3 H), 1.48-1.62 (m, 2 H), 1.85-1.93 (m, 1 H),

2.12 (br q, j = 7.2 Hz, 1 H), 3.97 (d, J = 16.8 Hz, 1 H), 4.20-4.23 (m, 1 H), 4.25

(dd, J = 1.5, 16.8 Hz, 1 H), 4.96 (br d, J = 10.1 Hz, 1 H), 5.09 (br d, J = 16.9 Hz, 1

H), 5.69 (dt, J = 7.0, 15.2 Hz, 1 H), 6.05 (br dd, J = 10.1, 15.2 Hz, 1 H), 6.26 (dt,

13 J = 10.1, 16.9 Hz, 1 H); C NMR (50 MHz, CDCl3) d 23.4, 23.8, 24.5, 27.8, 32.1,

66.4, 74.3, 100.5, 114.8, 131.3, 134.3, 137.0, 209.3; IR (neat) 1748, 1652, 1603 cm-1.

4-(hepta-4,6-dienyl)-5-(tert-butyldimethylsilyloxy)-2,2-dimethyl-4H-

[1,3]dioxin (243i). A solution of the above ketone (500 mg, 2.23 mmol) in THF

(22 mL) was added dropwise over 1.5 h to a solution of NaHMDS (1.0 M in THF,

2.90 mL, 2.90 mmol) in THF (15 mL) at –78 °C. TBSCl (436 mg, 2.90 mmol) was

added and the solution was warmed to rt over 2 h. Saturated aqueous NaHCO3

was added and the aqueous layer extracted with hexane. The combined organic

layers were washed with saturated aqueous Na2CO3, dried (Na2SO4) and

1 concentrated to yield a colorless oil (743 mg, 92%): H NMR (360 MHz, CDCl3) d

0.12 (s, 3 H), 0.13 (s, 3 H), 1.42 (s, 3 H), 1.45 (s, 3 H), 1.47-1.63 (m, 2 H), 1.72-

1.79 (m, 1 H), 2.10 (br qd, J = 2.5, 6.3 Hz, 1 H), 4.10 (ddd, J = 1.3, 2.9, 6.3 Hz, 1

184

H), 4.92 (dd, J = 1.7, 10.2 Hz, 1 H), 5.06 (dd, J = 1.3, 16.9 Hz, 1 H), 5.69 (dt, J =

6.9, 15.1 Hz, 1 H), 6.04 (dd, J = 10.2, 15.1 Hz, 1 H), 6.10 (d, J = 1.3 Hz, 1 H),

13 6.23 (dt, J = 10.2, 16.9 Hz, 1 H); C NMR (90 MHz, CDCl3) d -4.7, -4.4, 17.9,

20.9, 23.8, 25.6 (3 C), 27.8, 30.5, 32.4, 69.8, 98.2, 114.5, 126.3, 131.1, 134.5,

-1 + 135.2, 137.3; IR (neat) 2955, 1653 cm ; HRMS (M+H ) calcd for C19H35O3Si

339.2355, found 339.2333.

5-(tert-butyldimethylsilyloxy)-2,3,3a,5,6,8a-hexahydro-1H-azulen-4-one

(246). Dimethyl aluminum chloride (1.0 M in hexane, 113 mL, 0.113 mmol) was added dropwise to dioxin (243i) (38 mg, 0.113 mmol) in CH2Cl2 (500 mL) at –78

°C. After stirring overnight at –78 °C, saturated aqueous NaHCO3 was added and the solution was warmed to rt. The aqueous layer was extracted with CH2Cl2 and the combined organic layers were dried (Na2SO4) and concentrated.

Purification of the crude residue by silica-gel chromatography (benzene-hexane,

3 : 17, silica-gel deactivated with 10% triethylamine) gave a mixture of diastereomers (endo : exo, 79 : 21) as colorless oils (23 mg overall, 74%) : endo isomer 1H NMR (400 MHz, d6-benzene) d 0.09 (s, 3 H), 0.25 (s, 3 H), 1.04 (s, 9

H), 1.15 (dtd, J = 2.0, 9.1, 18.4 Hz, 1 H), 1.32–1.44 (m, 2 H), 1.49 (dt, J = 6.7,

14.3 Hz, 1 H), 1.70 (dtd, J = 4.0, 7.0, 9.1 Hz, 1 H), 1.95 (tdd, J = 3.5, 7.4, 10.8

Hz, 1 H), 2.02 (ddd, J = 8.3, 11.0, 11.0 Hz, 1 H), 2.21 (ddddd, J = 1.4, 2.4, 4.9,

185

10.8, 14.7 Hz, 1 H), 2.32 (ddd, J = 4.0, 8.3, 14.7 Hz, 1 H), 2.40 (tdd, J = 6.3, 9.1,

12.8 Hz, 1 H), 4.00 (dd, J = 4.0, 10.6 Hz, 1 H), 5.58 (tdd, J = 2.4, 4.9, 8.3 Hz, 1

13 H), 5.69 (ddd, J = 2.5, 3.6, 10.6 Hz, 1 H); C NMR (90 MHz, CDCl3) d -5.3, -4.6,

18.5, 23.6, 25.2, 25.8 (3 C), 34.1, 34.4, 43.4, 57.7, 77.8, 208.9; IR (neat) 1721,

-1 + 1472 cm ; HRMS (M+H ) calcd for C16H29O2Si 281.1937, found 281.1952. exo

isomer: 1H NMR (400 MHz, d6-benzene) d 0.09 (s, 3 H), 0.27 (s, 3 H), 1.03 (s, 9

H), 1.17 – 1.31 (m, 2 H), 1.45 (dq, J = 4.6, 117.5 Hz, 1 H), 1.52 (dt, J = 7.0, 17.5

Hz, 1 H), 1.63 (tdd, J = 3.4, 4.7, 8.0 Hz, 1 H), 2.16 (dddd, J = 0.8, 4.8, 6.8, 15.1

Hz, 1 H), 2.17 (dddd, J = 5.8, 8.1, 10.5, 15.1 Hz, 1 H), 2.43 – 2.52 (m, 2 H), 2.65

(ddd, J = 6.1, 7.4, 8.1 Hz, 1 H), 4.28 (t, J = 4.8 Hz, 1 H), 5.35 (dt, J = 2.4, 10.7

13 Hz, 1 H), 5.62 (tdd, J = 2.9, 6.0, 10.7 Hz, 1 H); C NMR (75 MHz, CDCl3) d -5.3,

-4.5, 18.5, 23.6, 25.8 (3 C), 26.0, 33.7, 33.8, 41.7, 51.7, 79.2, 126.5, 134.6,

-1 + 208.9; IR (neat) 1726, 1472 cm ; HRMS (M+H ) calcd for C16H29O2Si 281.1937,

found 281.1952.

{2-[2-(benzyloxy)ethyl]-5-nitro-1,3-dioxan-5-yl}methanol. To a solution of tris(hydroxymethyl)nitromethane (259) (15.0 g, 99.3 mmol) in CHCl3 (500 mL)

was added acrolein (4.42 mL, 66.2 mmol), benzyl alcohol (22.6 mL, 218.4 mmol),

and p-toluenesulfonic acid monohydrate (252 mg, 1.32 mmol). The mixture was

refluxed with a Soxhlet extractor containing 4Å molecular sieves for 72 h. The

186

solution was poured onto saturated aqueous NaHCO3 extracted with CH2Cl2 and

the combined organic layers were dried (Na2SO4) and concentrated. Purification

of the crude residue by silica-gel chromatography (EtOAc-hexane, 1 : 3) gave a 3

: 1 mixture of diastereomers as colorless oils (12.3 g overall, 63%) : major

1 isomer H NMR (200 MHz, CDCl3) d 1.90 (q, J = 5.9 Hz, 2 H), 2.97 (br s, 1 H),

3.56 (t, J = 5.9 Hz, 2 H), 3.65 (s, 2 H), 3.79 (d, J = 9.9 Hz, 2 H), 4.47 (s, 2 H),

4.71 (t, J = 5.9 Hz, 1 H), 4.79 (d, J = 9.9 Hz, 2 H), 7.26-7.40 (m, 5 H); 13C NMR

(50 MHz, CDCl3) d 34.5, 63.2, 65.1, 68.2, 72.9, 87.1, 100.2, 127.7, 127.8, 128.4,

-1 1 137.9; IR (neat) 3416, 1550 cm ; minor isomer: H NMR (200 MHz, CDCl3) d

1.94 (q, J = 6.1 Hz, 2 H), 2.93 (br s, 1 H), 3.57 (t, J = 6.1 Hz, 2 H), 3.96 (dd, J =

1.3, 10.4 Hz, 2 H), 4.20 (d, J = 3.5 Hz, 2 H), 4.39 (dd, J = 1.0, 10.4 Hz, 2 H), 4.50

13 (s, 2 H), 4.67 (t, J = 5.3 Hz, 1 H), 7.26-7.41 (m, 5 H); C NMR (50 MHz, CDCl3) d

34.3, 62.3, 65.0, 67.2, 72.9, 82.5, 100.5, 127.6, 128.3, 137.9; IR (neat) 3417,

1543cm-1.

2-[2-(benzyloxy)ethyl]-1,3-dioxan-5-one (261). To a solution of nitro-alcohols

(12.1 g, 40.6 mmol) in EtOH (40 mL) was added Raney-Ni (50% in H2O, 4.0 g).

The mixture was placed under H2 (1400 psi) at 80 °C for 3 d, filtered, and

concentrated. The crude amino-alcohols 260 were pure enough to be taken onto the oxidation. The amino-alcohols were dissolved in H2O/MeOH (1 : 1, 80 mL)

187

and KH2PO4 (5.53 g, 40.6 mmol) was added and the solution cooled to 0 °C. A

solution of NaIO4 (8.69 g, 40.6 mmol) in H2O (80 mL) was added dropwise over 1

h and the mixture was warmed overnight to rt. The solution was extracted with

CH2Cl2 and the organic layers were combined, dried (Na2SO4) and concentrated

1 to yield a pure colorless oil (7.80 g, 82%) : H NMR (200 MHz, CDCl3) d 2.03 (q,

J = 6.0 Hz, 2 H), 3.62 (t, J = 6.0 Hz, 2 H), 4.25 (dd, J = 1.5, 17.8 Hz, 2 H), 4.39

(dd, J = 1.5, 17.8 Hz, 2 H), 4.52 (s, 2 H), 5.06 (t, J = 6.0 Hz, 1 H), 7.31-7.37 (m, 5

13 H); C NMR (50 MHz, CDCl3) d 34.4, 65.2, 72.8, 72.9, 97.8, 127.6, 128.3, 138.2,

-1 + 204.2; IR (neat) 2867, 1738 cm ; HRMS (M+H ) calcd for C13H17O4 237.1127,

found 237.1139.

methyl 2-{2-[2-(benzyloxy)ethyl]-1,3-dioxan-5-ylidene}acetate (262). To a

solution of trimethoxy phosphonoacetate (1.31 mL, 9.1 mmol) in THF (91 mL) at

0 °C was added NaH (60% in mineral oil, 370 mg, 9.30 mmol). The ylide was

stirred at rt for 1 h, then a solution of ketone 261 (2.15 g, 9.1 mmol) in THF (40

mL) was added. The mixture was stirred at rt for 12 h, poured onto a saturated

aqueous NH4Cl and extracted with Et2O. The combined extracts were dried

(Na2SO4) and concentrated. Purification of the crude residue by filtration through

1 a plug of silica-gel with Et2O gave a colorless oil (2.66 g, >99%): H NMR (200

MHz, CDCl3) d 1.95 (q, J = 6.3 Hz, 2 H), 3.60 (t, J = 6.3 Hz, 2 H), 3.70 (s, 3 H),

188

4.29-4.43 (m, 3 H), 4.51 (s, 2 H), 4.90 (t, J = 6.3 Hz, 1 H), 5.57 (d, J = 15.4 Hz, 1

13 H), 5.68 (br s, 1 H), 7.26-7.37 (m, 5 H); C NMR (50 MHz, CDCl3) d 35.0, 51.4,

65.3, 66.0, 69.9, 72.9, 99.2, 114.5, 127.5, 127.6, 128.3, 138.3, 149.9, 165.6; IR

-1 + (neat) 1716, 1667 cm ; HRMS (M+Na ) calcd for C16H20O5Na 315.1208, found

315.1222.

methyl 2-{2-[2-(benzyloxy)ethyl]-4H-1,3-dioxin-5-yl}acetate (263). To a

solution of the above a,b-unsaturated ester (1.79 g, 6.13 mmol) in benzene (25

mL) was added DBU (184 mL) and the mixture was heated at 50 °C for 72 h. The

solution was cooled and concentrated. Purification of the crude residue by

filtration through a plug of silica-gel with Et2O gave a colorless oil (1.69 g, >99%):

1 H NMR (200 MHz, CDCl3) d 2.02 (q, J = 6.2 Hz, 2 H), 2.89 (s, 2 H), 3.62 (t, J =

6.2 Hz, 2 H), 3.68 (s, 3 H), 4.15 (dd, J = 0.9, 15.0 Hz, 1 H), 4.38 (dd, J = 1.1, 15.0

Hz, 1 H), 4.51 (s, 2 H), 5.00 (t, J = 6.2 Hz, 1 H), 6.47 (br s, 1 H), 7.30-7.35 (m, 5

13 H); C NMR (50 MHz, CDCl3) d 34.3, 34.7, 51.8, 65.1, 66.2, 72.9, 96.6, 107.3,

127.4, 127.5, 128.3, 138.2, 141.2, 171.3; IR (neat) 1738, 1675 cm-1; HRMS

+ (M+Na ) calcd for C16H20O5Na 315.1208, found 315.1232.

189

2-{2-[2-(benzyloxy)ethyl]-4H-1,3-dioxin-5-yl}acetic acid. To a solution of dioxin 248 (1.52 g, 5.19 mmol) in MeOH (16 mL) at 0 °C was added KOH (1.0 M in H2O, 15.6 mL). The mixture was warmed to rt and stirred for 2 h at this temperature. The solution was placed on the rotovap and MeOH was removed under vacuum. The aqueous layer was washed with Et2O then acidified with saturated aqueous KH2PO4 (pH = 5) and saturated with NaCl. The aqueous layer was extracted with CH2Cl2 and the combined extracts were dried (Na2SO4)

and concentrated. The crude acid was used directly in the birch reduction: 1H

NMR (200 MHz, CDCl3) d 2.01 (q, J = 6.1 Hz, 2 H), 2.91 (s, 2 H), 3.62 (t, J = 6.1

Hz, 2 H), 4.15 (dd, J = 0.9, 15.1 Hz, 1 H), 4.40 (dd, J = 1.1, 15.1 Hz, 1 H), 4.53

(s, 2 H), 5.00 (t, J = 6.1 Hz, 1 H), 6.40 (br s, 1 H), 7.31-7.41 (m, 5 H).

2-[2-(2-hydroxyethyl)-4H-1,3-dioxin-5-yl}acetic acid (264). Liquid NH3 (300

mL) was condensed in a dry reaction vessel at –78 °C and lithium wire (1.02 g,

146 mmol) was added in small pieces. A solution of the above acid (1.70 g, 6.09

mmol) in t-BuOH (2.92 mL, 30.5 mmol) and THF (60 mL) was added dropwise to

the blue solution and stirred for 0.5 h. Solid NH4Cl was added until the lithium

190

was quenched, and the resultant mixture was allowed to boil off the solvents

overnight. The solids were dissolved with saturated aqueous KH2PO4 (pH = 5)

and then saturated with NaCl. The aqueous layer was extracted with CH2Cl2 and

EtOAc, the combined extracts were dried (Na2SO4) and concentrated. The white

1 solid (1.04 g, 90 %) was pure by NMR analysis: H NMR (200 MHz, CDCl3) d

2.00 (q, J = 6.3 Hz, 2 H), 2.94 (s, 2 H), 3.62 (br s, 2 H), 3.81 (td, 1.8, 6.3 Hz, 2

H), 4.19 (dd, J = 0.9, 15.1 Hz, 1 H), 4.45 (dd, J = 1.1, 15.1 Hz, 1 H), 5.10 (t, J =

6.3 Hz, 1 H), 6.51 (br s, 1 H).

{2-[2-(triisopropylsilyloxy)ethyl]-5-nitro-1,3-dioxan-5-yl}methanol. To a solution of aldehyde 271 (14.6 g, 63.4 mmol) in CHCl3 (250 mL) was added was added tris(hydroxymethyl)-nitromethane (259) (14.4 g, 95.1 mmol) and p- toluenesulfonic acid monohydrate (1.2 g, 6.34 mmol). The mixture was refluxed with a Soxhlet extractor containing 4Å molecular sieves for 72 h. The solution was poured onto saturated aqueous NaHCO3 extracted with CH2Cl2 and the combined organic layers were dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel chromatography (EtOAc-hexane, 1 : 3) gave a single diastereomers as a colorless oil (9.6 g overall, 41%) : 1H NMR (360 MHz,

CDCl3) d 0.94-1.08 (m, 21 H), 1.77 (q, J = 5.8 Hz, 2 H), 3.09 (br s, 1 H), 3.73 (t, J

= 5.8 Hz, 2 H), 3.75 (br s, 2 H), 3.81 (d, J = 12.9 Hz, 2 H), 4.74 (t, J = 5.9 Hz, 1

191

13 H), 4.79 (d, J = 12.9 Hz, 2 H); C NMR (90 MHz, CDCl3) d 11.8, 17.8, 37.4, 58.2,

63.3, 68.3, 87.3, 100.4; IR (neat) 3466, 1553 cm-1.

{2-[2-(triisopropylsilyloxy)ethyl]-5-amino-1,3-dioxan-5-yl}methanol (272). To a solution of the above nitro-alcohol (3.56 g, 9.80 mmol) in EtOH (20 mL) was added Raney-Ni (50% in H2O, 400 mg). The mixture was placed under H2 (1400 psi) at 80 °C overnight, filtered, and concentrated. The pure amino-alcohol 257 was isolated as a clear tan oil (3.23 g, 99%): 1H NMR (200 MHz, d6-acetone) d

0.94-1.08 (m, 21 H), 1.78 (q, J = 5.9 Hz, 2 H), 2.50 (br s, 1 H), 3.23 (s, 2 H), 3.62

(d, J = 12.9 Hz, 2 H), 3.71 (d, J = 12.9 Hz, 2 H), 3.73 (t, J = 5.9 Hz, 2 H), 4.64 (t,

13 J = 5.9 Hz, 1 H); C NMR (50 MHz, CDCl3) d 12.6, 18.3, 39.0, 51.2, 59.6, 65.5,

74.9, 100.5, 206.0; IR (neat) 3355, 3288, 1586 cm-1.

2-[2-(triisopropylsilyloxy)ethyl]-1,3-dioxan-5-one 258. The amino-alcohol 272

(1.17 g, 3.51 mmol) was dissolved in MeOH (8 mL) and KH2PO4 (525 mg, 3.86 mmol) was added and the solution cooled to 0 °C. A solution of NaIO4 (826 mg,

3.86 mmol) in H2O (8 mL) was added dropwise and the mixture was warmed

192

overnight to rt. The solution was extracted with CH2Cl2 and the organic layers

were combined, dried (Na2SO4) and concentrated to yield a pure colorless oil

1 (870 mg, 82%) : H NMR (200 MHz, CDCl3) d 0.92-1.15 (m, 21 H), 1.91 (q, J =

6.1 Hz, 2 H), 3.81 (t, J = 6.1 Hz, 2 H), 4.25 (dd, J = 0.9, 17.7 Hz, 2 H), 4.38 (dd, J

13 = 1.5, 17.7 Hz, 2 H), 5.06 (t, J = 6.0 Hz, 1 H); C NMR (50 MHz, CDCl3) d 11.4,

17.8, 36.9, 58.1, 72.5, 98.0, 205.2; IR (neat) 2943, 1745 cm-1.

methyl 2-{2-[2-(triisopropylsilyloxy)ethyl]-1,3-dioxan-5-ylidene}acetate

(274). To a solution of trimethoxy phosphonoacetate (1.48 mL, 9.16 mmol) in

THF (91 mL) at 0 °C was added NaH (60% in mineral oil, 366 mg, 9.16 mmol).

The ylide was stirred at rt for 1 h, then a solution of ketone 273 (2.52 g, 8.33

mmol) in THF (40 mL) was added. The mixture was stirred at rt for 12 h, poured

onto a saturated aqueous NH4Cl and extracted with Et2O. The combined

extracts were dried (Na2SO4) and concentrated. Purification of the crude residue

by filtration through a plug of silica-gel with Et2O gave a colorless oil (2.99 g,

1 >99% crude yield): H NMR (300 MHz, CDCl3) d 0.92-1.15 (m, 21 H), 1.87 (q, J =

6.0 Hz, 2 H), 3.70 (s, 3 H), 3.81 (t, J = 6.0 Hz, 2 H), 4.29-4.45 (m, 4 H), 4.91 (t, J

= 6.0 Hz, 1 H), 5.59 (d, J = 15.4 Hz, 1 H), 5.68 (br s, 1 H).

193

methyl 2-{2-[2-(triisopropylsilyloxy)ethyl]-4H-1,3-dioxin-5-yl}acetate. To a

solution of a,b-unsaturated ester 274 (2.99 g, 8.33 mmol) in benzene (33 mL)

was added DBU (250 mL) and the mixture was heated at 50 °C for 18 h. The

solution was cooled and concentrated. Purification of the crude residue by

filtration through a plug of silica-gel with Et2O gave a colorless oil (2.75 g, 92%):

1 H NMR (360 MHz, CDCl3) d 0.99-1.11 (m, 21 H), 2.02 (dt, J = 5.4, 6.4 Hz, 2 H),

2.89 (s, 2 H), 3.67 (s, 3 H), 3.84 (t, J = 6.4 Hz, 2 H), 4.16 (d, J = 15.0 Hz, 1 H),

4.37 (d, J = 15.0 Hz, 1 H), 5.00 (t, J = 5.4 Hz, 1 H), 6.47 (br s, 1 H); 13C NMR (90

MHz, CDCl3) d 11.9, 17.9, 29.7, 34.9, 37.4, 51.9, 58.5, 66.3, 96.7, 107.4, 171.4;

IR (neat) 2942, 1744, 1674 cm-1.

2-{2-[2-(triisopropylsilyloxy)ethyl]-4H-1,3-dioxin-5-yl}ethanol (275). To a solution of the above ester (1.51 g, 4.22 mmol) in Et2O (14 mL) was added LiAlH4

(160 mg, 11.7 mmol) at 0 °C. The mixture was stirred for 40 min at 0 °C, then

H2O (160 mL) was added dropwise, followed by NaOH (3.0 M in H2O, 160 mL),

and lastly more H2O (480 mL). The mixture was filtered and concentrated to give

194

1 a colorless oil (1.30 g, 99%): H NMR (360 MHz, CDCl3) d?0.92-1.05 (m, 21 H),

1.78-1.91 (m, 2 H), 2.03 (q, J = 5.4 Hz, 3 H), 2.80 (br s, 1 H), 3.53 (t, J = 6.4 Hz,

2 H), 3.76 (t, J = 5.8 Hz, 2 H), 4.05 (d, J = 15.5 Hz, 1 H), 4.23 (d, J = 15.5 Hz, 1

13 H), 4.89 (t, J = 5.4 Hz, 1 H), 6.35 (br s, 1 H); C NMR (50 MHZ, CDCl3) d 11.8,

17.8, 32.2, 37.3, 45.6, 58.4, 60.6, 66.2, 96.3, 110.7, 139.6; IR (neat) 3406, 1673

cm-1.

2-bromo-4,4-dimethylcyclopent-1-enecarboxaldehyde (323). To a solution of

DMF (513 mL, 6.63 mmol) in CH2Cl2 (10 mL) was added POBr3 (1.58 g, 5.52 mmol) at rt. The solution was stirred at rt for 1 h as a white precipitate formed. A solution of silyl enol ether 325 (500 mg, 2.21 mmol) in CH2Cl2 (2 mL) was added to the mixture and the resultant slurry was stirred 72 h at rt and poured onto ice

(5 g). The solution was neutralized with NaHCO3 and extracted with hexane/Et2O (9 : 1). The combined organic layers were washed with saturated aqueous NaHCO3, dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel chromatography (ethyl acetate-hexane, 3 : 97, silica-gel deactivated with 10% triethylamine) gave exclusively one regioisomer as a

1 colorless oil (289 mg, 64%): H NMR (200 MHz, CDCl3) d 1.11 (s, 6 H), 2.29 (t, J

13 = 2.2 Hz, 2 H), 2.67 (t, J = 2.2 Hz, 2 H), 9.83 (s, 1 H); C NMR (50 MHz, CDCl3) d 29.3 (2 C), 37.5, 43.8, 56.8, 139.0, 139.2, 189.2; IR (neat) 1675, 1608 cm-1.

195

2-bromo-4,4-dimethylcyclopent-1-enecarboxaldehyde oxime (326). To a

solution of aldehyde 323 (325 mg, 1.60 mmol) in EtOH (4 mL) was added a

solution of H2NOH•HCl (167 mg, 2.40 mmol) and sodium acetate (197 mg, 2.40

mmol) in EtOH/H2O (2.70 mL, 1 : 1) dropwise at 0 °C. The mixture was stirred

1.5 h at rt and the EtOH was concentrated off. Brine was added and the

aqueous layer was extracted with CH2Cl2. The combined organic layers were dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel chromatography (Et2O- CH2Cl2-hexane, 4 : 45 : 51) provided a white solid (285 mg, 82%): 1H NMR (200 MHz, d6-acetone) d 1.12 (S, 6 H), 2.32 (td, J = 0.7, 2.0

Hz, 2 H), 2.56 (t, J = 2.0 Hz, 2 H), 7.93 (s, 1 H), 10.39 (s, 1 H); 13C NMR (50

MHz, d6-acetone) d 29.6 (2C), 38.0, 46.4, 55.8, 123.0, 134.4, 145.7; IR (neat)

-1 + 3286, 1626 cm ; HRMS (M+H ) calcd for C8H13ONBr 218.0181, found 218.0198.

1-[1-(1-hydroxyethyl)-cyclopropyl]ethanone (328). Lithium tri(tert- butoxy)alumninohydride (15.9 mL, 1.0 M in THF, 15.9 mmol) was added

89 dropwise to a solution of dione 327 (2.00 g, 15.9 mmol) in Et2O (80 mL) at –78

°C. The solution was slowly warmed to rt and stirred overnight. A saturated

aqueous solution of sodium potassium tartrate was added and the aqueous layer

196

was extracted with CH2Cl2. The combined organic layers were washed with

saturated aqueous sodium potassium tartrate, dried (Na2SO4) and concentrated.

Purification of the crude residue by silica-gel chromatography (ethyl acetate-

1 benzene, 1 : 4) yielded a colorless oil (1.18 g, 58%): H NMR (200 MHz, CDCl3) d 0.86 (dd, J = 3.1, 5.2 Hz, 2 H), 0.96–1.06 (m, 2 H), 1.04 (d, J = 6.6 Hz, 3 H),

2.02 (s, 3 H), 3.36 (br d, J = 6.6 Hz, 1 H), 3.67 (br p, J = 6.6 Hz, 1 H); 13C NMR

(50 MHz, CDCl3) d 11.8, 13.8, 19.8, 24.5, 36.9, 68.8, 209.8; IR (neat) 3440, 1682 cm-1.

1-(1-vinylcyclopropyl)ethanone (321). To a solution of ketoalcohol 328 (1.20

g, 9.35 mmol) in benzene (50 mL) was added imidazole (700 mg, 10.28 mmol),

triphenylphosphine (2.70 g, 10.28 mmol) and lastly iodine (2.66 g, 10.47 mmol).

The solution was stirred 1 h at rt and poured onto a 1 : 1 solution of 10%

Na2S2O3 and saturated aqueous NaHCO3. The aqueous layer was extracted with hexane and the combined organic layers were dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel chromatography

1 (Et2O–pentane, 2 : 23) afforded a yellow oil (1.81 g, 82%): H NMR (200 MHz,

CDCl3) d 1.07 (ddd, J = 4.1, 6.6, 9.4 Hz, 1 H), 1.22 (ddd, J = 4.3, 6.2, 9.4 Hz, 1

H), 1.44 (ddd, J = 4.1, 6.2, 9.5 Hz, 1 H), 1.60 (ddd, J = 4.3, 6.6, 9.5 Hz, 1 H), 1.81

(d, J = 7.2 Hz, 3 H), 2.08 (s, 3 H), 4.64 (q, J = 7.2 Hz, 1 H). The above iodide

(1.63 g, 6.85 mmol) was immediately subjected to elimination by addition of DBU 197

(2.05 mL, 13.7 mmol). The neat mixture was heated to 85 °C at reduced

pressure (20 mmHg) as the product was distilled over and trapped at –78 °C to

1 yield a volatile colorless oil (440 mg, 58%): H NMR (360 MHz, CDCl3) d 1.01 (q,

J = 3.5 Hz, 2 H), 1.35 (q, J = 3.5 Hz, 2 H), 2.15 (s, 3 H), 4.99 (dd, J = 1.1, 17.1

Hz, 1 H), 5.07 (dd, J = 1.1, 10.5 Hz, 1 H), 6.46 (dd, J = 10.5, 17.1 Hz, 1 H); 13C

NMR (90 MHz, CDCl3) d 18.8 (2 C), 27.7, 34.0, 114.6, 136.7, 207.6; IR (neat)

1697, 1639 cm-1.

2-[1-hydroxy-1-(1-vinylcyclopropyl)ethyl]-4,4-dimethylcyclopent-1-

enecarboxaldehyde oxime (319). To a solution of oxime 326 (100 mg, 0.46

mmol) in THF (2.5 mL) at –78 °C was added tert-butyllithium (837 mL, 1.7 M in

pentane, 1.42 mmol) dropwise. The resultant yellow solution was stirred at –78

°C for 1.5 h. A solution of ketone 321 (66 mg, 0.60 mmol) in THF (2 mL) was

added dropwise over 1 h and the resultant solution was slowly warmed to rt over

1 h. The mixture was poured onto saturated aqueous NaHCO3 and the aqueous

layer was extracted with Et2O. The combined organic extracts were dried

(Na2SO4) and concentrated. Purification of the crude residue by silica-gel

chromatography (EtOAc-isopropanol-hexane, 1 : 0.01 : 4) yielded a white solid

1 (77 mg, 68%): H NMR (360 MHz, CDCl3) d 0.56 (ddd, J = 3.3, 5.0, 8.6 Hz, 1 H),

0.77 (ddd, J = 4.4, 5.8, 8.6 Hz, 1 H), 0.84 (ddd, J = 4.4, 5.0, 9.7 Hz, 1 H), 0.89

198

(ddd, J = 3.3, 5.8, 9.7 Hz, 1 H), 1.05 (s, 3 H), 1.07 (s, 3 H), 1.15 (s, 3 H), 2.24 (dt,

J = 1.8, 17.0 Hz, 1 H), 2.36 (dt, J = 1.6, 17.0 Hz, 1 H), 2.41 (br s, 2 H), 4.94 (dd, J

= 1.3, 17.5 Hz, 1 H), 4.96 (dd, J = 1.3, 10.3 Hz, 1 H), 6.14 (dd, J = 10.3, 17.5 Hz,

1 H), 8.2 (br s, 1 H), 8.83 (s, 1 H); 13C NMR (50 MHz, d6-acetone) d 11.1, 12.3,

25.3, 29.6, 32.1, 36.7, 48.4, 51.6, 76.2, 111.5, 130.6, 141.9, 149.7, 150.1; HRMS

+ (M+H ) calcd for C15H24O2N 250.1807, found 250.1786.

cycloadduct 317. To a solution of hydroxyoxime 319 (97 mg, 0.39 mmol) in

EtOH (4 mL) was added chloramine-T (134 mg, 0.59 mmol) at rt. The mixture was heated to 40 °C for 6 h then concentrated. The residue was dissolved in

EtOAc and the organic phase was washed with 1.0 M NaOH (3x) and brine (2x).

The organic layer was dried (Na2SO4) and concentrated to provide a white solid

1 as a single diastereomer (95 mg, 99%): H NMR (400 MHz, CDCl3) d 0.18 (ddd, J

= 4.6, 5.3, 10.1 Hz, 1 H), 0.55 (ddd, J = 3.5, 4.9, 10.1 Hz, 1 H), 0.72 (ddd, J =

4.6, 4.9, 10.0 Hz, 1 H), 0.77 (ddd, J = 3.5, 5.3, 10.0 Hz, 1 H), 1.12 (s, 3 H), 1.15

(s, 3 H), 1.29 (s, 1 H), 1.43 (s, 3 H), 2.31 (dt, J = 1.8, 17.8 Hz, 1 H), 2.39 (dt, J =

1.8, 12.7 Hz, 1 H), 2.44 (dt, J = 1.8, 12.7 Hz, 1 H), 2.54 (dt, J = 1.8, 17.8 Hz, 1

H), 3.43 (dd, J = 7.9, 13.1 Hz, 1 H), 3.88 (dd, J = 10.3, 13.1 Hz, 1 H), 4.29 (dd, J

13 = 7.9, 10.3 Hz, 1 H); C NMR (90 MHz, CDCl3) d 3.0, 5.4, 24.5, 28.2, 29.3, 29.6,

199

38.4, 45.4, 46.5, 49.8, 68.8, 71.2, 129.1, 155.5, 155.8; IR (neat) 3438, 1630,

-1 + 1587 cm ; HRMS (M+H ) calcd for C15H22O2N 248.1651, found 248.1652.

ketodiol 332. Cycloadduct 317 (92 mg, 0.37 mmol) was dissolved in a

MeOH/H2O (3 mL, 5 : 1) solution. Boric acid (49 mg, 0.79 mmol) and Raney-Ni

(20 mg, 50% in H2O) were added and the mixture was stirred under balloon pressure of hydrogen for 2 h. The solution was poured onto saturated aqueous

NaHCO3 and the aqueous layer was extracted with CH2Cl2. The combined organic layers were washed with saturated aqueous NaHCO3, dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel chromatography

(EtOAc-isopropanol-hexane, 1 : 0.05 : 1) to afford a colorless oil (68 mg, 73%) :

1 H NMR (360 MHz, CDCl3) d 0.35 (dt, J = 4.5, 9.5 Hz, 1 H), 0.42 (dt, J = 4.5, 9.5

Hz, 1 H), 0.74 (dt, J = 4.5, 9.7 Hz, 1 H), 0.86 (dt, J = 4.5, 9.7 Hz, 1 H), 0.99 (s, 3

H), 1.04 (s, 3 H), 1.11 (s, 3 H), 1.70 (t, J = 3.2 Hz, 1 H), 2.34 (t, J = 1.6 Hz, 2 H),

2.40 (dt, J = 1.6, 18.2 Hz, 1 H), 2.48 (dt, J = 1.6, 18.2 Hz, 1H), 3.72 (dd, J = 3.2,

10.4 Hz, 1 H), 3.84 (dd, J = 3.2, 10.4 Hz, 1 H), 4.53 (br s, 1 H), 5.31 (br s, 1 H);

13 C NMR (90 MHz, CDCl3) d 5.7, 13.9, 21.1, 28.1, 29.1, 29.2, 37.7, 43.9, 47.6,

57.3, 63.1, 38.2, 137.4, 167.1, 199.1.

200

illudin C. To a solution of ketodiol 332 (43 mg, 0.17 mmol) in CH2Cl2 (1.7 mL) at

–78 °C was added Et3N (72 mL, 0.52 mmol) and MsCl (16 mL, 0.21 mmol) dropwise. The solution was warmed to 0 °C over 2 h then DBU (51 mL, 0.34 mmol) was added and the mixture was stirred at rt overnight. Saturated aqueous

NaHCO3 was added and the aqueous layer was extracted with CH2Cl2. The combined organic layers were dried (Na2SO4) and concentrated. Purification of the crude residue by silica-gel chromatography (MeOH-CHCl3, 1 : 99) to yield a white solid (29 mg, 73%): 1H NMR (400 MHz, d6-benzene) d 0.19 (ddd, J = 3.5,

6.6, 9.8 Hz, 1 H), 0.75 (ddd, J = 4.9, 6.6, 9.6 Hz, 1 H), 0.79 (dt, J = 4.9, 5.1, 9.6

Hz, 1 H), 0.90 (s, 3 H), 0.95 (s, 3 H), 1.17 (s, 3 H), 1.19 (ddd, J = 3.5, 5.1, 9.8 Hz,

1 H), 1.75 (s, 1 H), 2.18 (dt, J = 2.0, 20.0 Hz, 1 H), 2.47 (dt, J = 1.8, 20.0 Hz, 1

H), 2.48 (dt, J = 1.8, 16.7 Hz, 1 H), 2.51 (dt, J = 2.0, 16.7 Hz, 1 H), 4.86 (d, J =

1.6 Hz, 1 H), 6.15 (d, J = 1.6 Hz, 1 H); 13C NMR (90 MHz, d6-benzene) d 5.2,

13.2, 26.1, 29.3, 29.4, 33.4, 37.5, 44.8, 47.9, 70.1, 135.9, 148.4, 169.2, 169.3,

185.7; IR (neat) 3443, 1657, 1605 cm-1.

201

References

1 Woodward, R. B.; Heusler, K.; Gosteli, J.; Naegeli, P.; Oppolzer, W.; Ramage, R.; Ranganathan, S.; Vorbruggen, H. J. Am. Chem. Soc. 1966, 88, 852.

2 Liotta, D.; Saindane, M.; Barnum, C. J. Am. Chem. Soc. 1981, 103, 3224.

3 Tietze, L. F.; Meier, H.; Nutt, H. Liebigs Ann. Chem. 1990, 253. For reviews, see: (a) Desimoni, G.; Tacconi, G. Chem. Rev. 1975, 75, 651. (b) Tietze, L. F.; Beifuss, U. Angew. Chem., Int. Ed. Engl. 1993, 32, 131. (c) Waldmann, H. Synthesis 1994, 535. (d) Tietze, L. F. Chem. Rev. 1996, 96, 115. (e) Tietze, L. F.; Kettschau, G. Top. Curr. Chem. 1997, 189, 1. (f) Tietze, L. F.; Modi, A. Med. Res. Rev. 2000, 20, 304.

4 Arnold, Z.; Kryshtal, G. V.; Kral, V.; Dvorak, D.; Yanovskaya, L. A. Tetrahedron Lett. 1988, 29, 2861.

5 Leouge, P.; Paulmier, C. J. Am. Chem. Soc. 1985, 6, 1225.

6 For the preparation of alkylidenemalonaldehydes, see: (a) Tietze, L.-F.; Glusenkamp, K.-H.; Holla, W. Angew. Chem., Int. Ed. Engl. 1982, 21, 793. (b) Arnold, Z.; Kryshtal, G. V.; Kral, V.; Dvorak, D.; Yanovskaya, L. A. Tetrahedron Lett. 1988, 29, 2861. For the in situ generation 2-(trichloroacetyl)-2-enals, presumably as a mixture of stereoisomers, see: (c) Tietze, L. F.; Meier, H.; Nutt, H. Chem. Ber. 1989, 122, 643. (d) Tietze, L. F.; Meier, H.; Nutt, H. Liebigs Ann. Chem. 1990, 253.

7 Funk, R. L.; Bolton, G. L. J. Am. Chem. Soc. 1988, 110, 1290.

8 Funk, R. L.; Yost, K. J. J. Org. Chem. 1996, 61, 2598.

9 Funk, R. L.; Maeng, J.-H. Org. Lett. 2001, 3, 1125.

10 (a) Funk, R. L.; Maeng, J.-H. Org. Lett. 2001, 3, 1125. (b) Funk, R. L.; Greshock, T. J. Org. Lett. 2001, 3, 3511. (c) Funk, R. L.; Maeng, J.-H. Org. Lett. 2002, 4, 331.

11 Funk, R. L.; Maeng, J.-H.; Zhang, X.-C. unpublished results.

12 Funk, R. L.; Fuchs, J. R. Org. Lett. 2001, 3, 3923.

13 Funk, R. L.; Fearnley, S. P.; Gregg, R. J. Tetrahedron 2000, 56, 10275.

14 Funk, R. L.; Aungst, R. A. J. Am. Chem. Soc. 2001, 123, 9455.

202

15 Majewski, M.; Gleave, D. M.; Nowak, P. Can. J. Chem. 1995, 73, 1616.

16 (a) Enders, D.; Bockstiegel, B. Synthesis 1989, 493. (b) Molander, G. A.; Cameron, K. O. J. Org. Chem. 1993, 58, 5931.

17 Yamashita, M.; Mino, T.; Fukui, S. J. Org. Chem. 1997, 62, 734.

18 Cuvigny, T.; Larchevêque, M.; Valette, G. Tetrahedron 1979, 35, 1745.

19 Kobayashi, S.; Hachiya, I. Tetrahedron Lett. 1992, 33, 1625.

20 (a) McMurry, J. E.; Scott, W. J. Tetrahedron Lett. 1983, 24, 979. (b) Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1992, 33, 6299.

21 (a) Fu, G. C.; Littke, A. F. J. Org. Chem. 1999, 64, 10. For more recent references, see: (b) Martinelli, J. R.; Freckmann, D. M. M.; Buchwald, S. L. Org. Lett. 2006, 8, 4843 and references therein.

22 Charette, A. B.; Côté, B.; Monroc, S.; Prescott, S. J. Org. Chem. 1995, 60, 6888.

23 (a) Büchi, G.; Carlson, J. A.; Powel, J. E.; Tietze, L.-F. J. Am. Chem. Soc. 1970, 92, 2165. (b) Partridge, J. J.; Chadha, N. K.; Uskokovié, M. R J. Am. Chem. Soc.1973, 95, 532. (c) Tietze, L.-F. Chem. Ber. 1974,107, 2499. (d) Trost, B. M.; Nanninga, T. N. J. Am. Chem. Soc. 1985, 107, 1293. (e) Isoe, S.; Kon, K. Helv. Chim. Acta 1983, 66, 755. (f) Demuth, M.; Chanrasekhar, S.; Schaffner, K. J. Am. Chem. Soc. 1984, 106, 1092. (g) Hiroi, K.; Miura, H.; Kotsuji, K.; Sato, S. Chem. Lett. 1981, 559. (h) Chang, N.-C.; Tai, H.-M.; Chang, M.-Y.; Lee, A.-Y. J. Org. Chem. 1999, 64, 659, and references therein. (i) Jensen, S. R.; Lyse- Petersen, S. E.; Nielsen, B. J. Phytochem. 1979, 18, 273. (j) Young, M. C.; Braga, M. R.; Dietrich, S. M.; Gottlieb, H. E.; Trevisan, L. M.; Bolzani, V. S. Phytochem. 1992, 31, 3433.

24 (a) Hoppe, D.; Hanko, R. Angew. Chem. Int. Ed. Eng. 1982, 21, 372. (b) Hoppe, D.; Brönneke, A. Tetrahedron Lett. 1983, 24, 1687. (c) Hoppe, D. Angew. Chem. Int. Ed. Eng. 1984, 23, 932. (d) Hoppe, D.; Zschage, O. Angew. Chem. Int. Ed. Eng. 1989, 28, 69.

25 (a) Dini, F.; Guella, G.; Giubbilini, P.; Mancini, I.; Pietra, F. Naturwissenschaften 1993, 80, 84. (b) Guella, G.; Dini, F.; Tomei, A.; Pietra, F. J. Chem. Soc., Perkin Trans. 1 1994, 17, 161. (c) Guella, G.; Dini, F.; Pietra, F. Helv. Chim. Acta 1996, 79, 710. (d) Cervia, D.; Martini, D.; Garcia-Gil, M.; di Giuseppe, G.; Guella, G.; Dini, F.; Bagnoli, P. Apoptosis 2006, 11, 829. (e) Savoia, D.; Avanzini, C.; Allice, T.; Callone, E.; Guella, G.; Dini, F. Antimicrobial Agents and Chemotherapy 2004, 48, 3828.

203

26 (a) Chang, M.-Y.; Lin, C.-H.; Chang, N.-C. Tetrahedron 2004, 80, 1581. (b) Yus, M.; Lorenzo, E.; Alonso, F. Tetrahedron Lett. 2000, 41, 1661.

27 Tietze, L. F.; Geissler, H.; Fennen, J.; Brumby, T.; Brand, S.; Schulz, G. J. Org. Chem. 1994, 59, 182.

28 Zamojski, A.; Kozluk, T. J. Org. Chem. 1977, 42, 1089.

29 Just, G.; Hambalek, R. Tetrahedron Lett. 1990, 31, 4693.

30 Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans. 1 1975, 1574.

31 Takano, S.; Ohkawa, T.; Tamori, S.; Satoh, S.; Ogasawara, K. J. Chem. Soc., Chem. Comm. 1988, 189.

32 (a) Iwasa. S.; Yamamoto, M.; Kohmoto, S.; Yamada, K. J. Chem. Soc., Perkin Trans. 1 1991, 1173. (b) Bachi, M. D.; Bosch, E.; Denenmark, D.; Girsh, D. J. Org. Chem. 1992, 57, 6803.

33 (a) Hartwig, W. Tetrahedron 1983, 39, 2609. (b) Barton, D. H. R.; Jaszberenyi, J. C. Tetrahedron Lett. 1989, 30, 2619.

34 (a) Brown, H. C.; Krishnamurthy, S. J. Org. Chem. 1976, 41, 3065. (b) Holder, R. W.; Matturro, M. G. J. Org. Chem. 1977, 42, 2166.

35 Kusuda, K.; Inanaga, J.; Yamaguchi, M. Tetrahedron Lett. 1989, 30, 2945.

36 Graham, S. L.; Scholz, T. H. Tetrahedron Lett. 1990, 31, 6269.

37 (a) Graham, S. L.; Scholz, T. H. Tetrahedron Lett. 1990, 31, 6269. (b) Keck, G. E.; McHarhy, S. F.; Murry, J. A. Tetrahedron Lett. 1993, 43, 6215.

38 (a) Whitesell, J. K.; Fisher, M.; Jardin, P. D. S. J. Org. Chem. 1983, 48, 1556. (b) Sammes, P. G.; Whitby, R. J. J. Chem. Soc., Perkin Trans. 1 1987, 195. (c) Vaghefi, M. M.; Bernacki, R. J.; Dalley, N. K.; Wilson, B. E.; Robins, R. K. J. Med. Chem. 1987, 30, 1383. (d) Knapp, S.; Kukkola, P. J. J. Org. Chem. 1990, 55, 1632. (e) Usman, N.; Matulic-Adamic, J.; Haeberli, P. J. Org. Chem. 1995, 60, 2563. (f) Matsuda, A.; Hattori, H.; Nozawa, E.; Iino, T.; Yoshimura, Y.; Shuto, S.; Shimamoto, Y.; Nomura, M.; Fukushima, M.; Tanaka, M.; Sasaki, T. J. Med. Chem. 1998, 41, 2892 and references therein.

39 Gassman, P. G.; Hodgson, P. K. G.; Balchunis, R. J. Am. Chem. Soc. 1976, 98, 1275.

204

40 Sarma, J. C.; Borbaruah, M.; Sarma, D. N.; Barua, N. C.; Sharma, R. P. Tetrahedron 1986, 42, 3999.

41 (a) Kashman, Y.; Groweiss, A. J. Org. Chem. 1980, 45, 3815. (b) Scheuer, P. J.; Schwartz, R.; Zabel, V.; Watson, W. H. Tetrahedron 1982, 37, 2725. (c) Davies-Coleman, M. T.; Hooper, G. J. Tetrahedron 1995, 36, 9973. (d) Fusetani, N.; Asano, M.; Matsunaga, S.; Hashimoto, K. Tetrahedron 1989, 45, 1647. (e) Fusetani, N.; Asano, M; Matsunaga, S.; Hashimoto, K. Tetrahedron 1987, 47, 5837. (f) Bouaicha, N.; Pesando, D; Puel, D. J. Nat. Prod. 1993, 56, 1747. (g) Ochi, M.; Kataoka, K.; Tatsukawa, K.; Kotsuki, H.; Shibata, K. Heterocycles, 1993, 36, 41. (h) Ochi, M.; Kataoka, K.; Tatsukawa, A.; Kotsuki, H; Shibata, K. Ibid. 1994, 38, 151. (i) Miyamoto, T.; Takenaka, Y.; Yamada, K; Higuchi, T. J. Nat. Prod. 1995, 58, 924. (j) Hooper,G. J.; Davies-Coleman, M. T.; Schleyer, M. J. Nat. Prod. 1997, 60, 889. (k) Rho, J.-R.; Oh, M-S.; Jang, K.; Cho, K. W.; Shin, J. J. Nat. Prod. 2001, 64, 540. (l) Pfander, H.; Liu, G.; Smith, T. C. Tetrahedron Lett. 1995, 36, 4979. (m) van der Helm, D.; Schmitz, F. J.; Vanderah, D. J.; Steudler, P. A.; Ciereszko, L. S.; Elkstrand, J. D. J. Am. Chem. Soc. 1977, 99, 5780.

42 (a) Renneberg, D.; Pfander, H.; Leumann, C. J. J. Org. Chem. 2000, 65, 9069. (b) Corey, E. J.; Mitra, R. B.; Uda, H. J. Am. Chem. Soc. 1964, 86, 485.

43 Greene, A. E.; Kanazawa, A.; Delair, P.; Pourashraf, M. J. Chem. Soc., Perkin Trans. 1 1997, 1911.

44 Petasis, N. A.; Bzowej, E. I. J. Am. Chem. Soc. 1990, 112, 6392.

45 Lüche, J.-L.; Gemal, A. L. J. Am. Chem. Soc. 1979, 101, 5848.

46 . (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457. (b) Chemler S. R.; Trauner, D; Danishefsky, S. J. Angew. Chem. Int. Ed. 2001, 40, 4544.

47 (a) Fürstner, A. Angew. Chem. Int. Ed. 2000, 17, 3010. (b) Grubbs, R. H.; Chang, S. Tetrahedron, 1998, 54, 4413.

48 For the preparation of oxacyclononenes, see: (a) Delgado, M.; Martín, J. D. Tetrahedron Lett. 1997, 38, 6299. (b) Crimmins, M. T.; Choy, A. L. J. Org. Chem. 1997, 62, 7548.

49 (a) Vedejs, E.; Fuchs, P. L. J. Am. Chem. Soc. 1973, 95, 822. (b) Vedejs, E.; Snoble, K. A. J.; Fuchs, P. L. J. Org. Chem. 1973, 38, 1178. (c) Mori, M.; Uozumi, Y.; Kimura, M.; Ban, Y. Tetrahedron 1986, 42, 3793.

205

50 Pietra, F; Guella, G.; Chiasera, G.; N'Diaye, I. Helv. Chim. Acta 1994, 77, 1203.

51 Nakamura, E.; Matsuzawa, S.; Horiguchi, Y.; Kuwajima, I. Tetrahedron Lett. 1986, 27, 4029.

52 Pais, G. C. G.; Maier, M. E. J. Org. Chem. 1999, 64, 4551.

53 Sakaitani, M.; Ohfune, Y. Tetrahedron Lett. 1985, 26, 5543.

54 Funk, R. L.; Aungst, R. A. Org. Lett. 2001, 3, 3553.

55 (a) Noyori, R.; Hayakawa, Y. Org. React. 1983, 29, 163. (b) Hoffmann, H. M. R. Angew. Chem., Int. Ed. Engl. 1984, 23, 1. (b) Mann, J. Tetrahedron 1986, 42, 4611. (c) Hosomi, A.; Tominaga. Y. Comprehensive Organic Synthesis; Trost, B. M, Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 5, pp 593-615. (d) Rigby, J. Org. React. 1997, 51, 351. (e) Harmata, M. Tetrahedron 1997, 53, 6235.

56 Sasaki, T.; Ishibashi, Y.; Ohno, M. Tetrahedron Lett. 1982, 23, 1693.

57 Blackburn, C.; Childs, R. F.; Kennedy, R. A. Can. J. Chem. 1983, 61, 1981.

58 Harmata, M.; Sharma, U. Org. Lett. 2000, 2, 2703.

59 (a) Davies, H. M.; Dai, X. J. Am. Chem. Soc. 2004, 126, 2692. (b) Saez, J. A.; Arno, M.; Domingo, L. R. Org. Lett. 2003, 5, 4117.

60 Green, J. J. Org. Chem. 1995, 60, 4287.

61 Rink, H. Tetrahedron Lett. 1987, 28, 3787.

62 Kaiser, E. T.; DeGrado, W. F. J. Org. Chem. 1980, 45, 1295.

63 Holmes, C. P.; Jones, D. G. J. Org. Chem. 1995, 60, 2318.

64 (a) Rao, A. V. R.; Gurjar, M. K.; Nallanganchu, B. R.; Bhandari, A. Tetrahedron Lett. 1993, 34, 7081. (b) Williams, R. M.; Aoyagi, Y. Tetrahedron 1998, 54, 13045.

65 Lloyd-Williams, P.; Albericio, F.; Giralt, E. Tetrahedron 1993, 49, 11065.

66 Masamune, S.; Roush, W. R.; Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfield, A. P.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183.

67 Orita, A.; Watanabe, A.; Tsuchiya, H.; Otera, J. Tetrahedron 1999, 55, 2889.

206

68 Waldmann, H.; Brohm, D.; Metzger, S.; Bhargava, A.; Müller, O.; Lieb, F. Angew. Chem. Int. Ed. 2002, 41, 307.

69 Hanson, J. R.; Nat. Prod. Rep. 2001, 18, 88 and reference 1 therein.

70 For synthetic approaches to labdanes possessing alkoxymethyl or acyloxymethyl angular carbon substituents, see; (a) Arnó, M.; González, M. A.; Marín, M. L.; Zaragozá, R. J. Tetrahedron Lett. 2001, 42, 1669. (b) Tokoroyama, T.; Fujimori, K.; Shimizu, T.; Yamigawa, Y.; Monden, M.; Iio, H. Tetrahedron 1988, 44, 6607. (c) Hagiwara, H.; Uda, H. J. Chem. Soc., Chem. Commun. 1987, 1351. (d) Yoshikoshi, A.; Kato, M.; Matsumura, Y.; Heima, K.; Fukamiya, N.; Kabuto, C. Tetrahedron 1987, 711. (e) Matsumoto, T.; Endo, Y.; Okimoto, M. Bull. Chem. Soc. Jpn. 1983, 56, 2018. (f) Matsumoto, T.; Usui, S. Bull. Chem. Soc. Jpn. 1982, 55, 1599. (g) Yoshikoshi, A.; Kato, M.; Heima, K.; Matsumura, Y. J. Am. Chem. Soc. 1981, 103, 2434.

71 (a) Kittakoop, P.; Wanasith, S.; Watts P.; Kramyu, J.; Tanticharoen, M.; Thebtaranonth, Y. J. Nat. Prod. 2001, 64, 385. (b) Tanaka, J.; Marriot, G.; Higa, T.; Higa, T. J. Nat. Prod. 2001, 64, 1468.

72 Kubo, I.; Jamalamadaka, V.; Kamikawa, T.; Takahashi, K.; Tabata, K.; Kusumi, T. Chem. Lett. 1996, 441.

73 (a) Smith, A. B.; Levenberg, P. A.; Suits, J. Z. Synthesis 1986, 184. (b) Takano, S.; Iwabuchi, Y.; Ogasawara, K. J. Chem. Soc., Chem. Comm. 1989, 1371.

74 Zhang, H. X.; Guibé, F.; Balavoine, G. J. Org. Chem. 1990, 55, 1857.

75 (a) Matsumoto, T.; Shirahama, H.; Ichihara, A.; Fukuoka, Y.; Takahashi, Y.; Mori, Y.; Watanabe, M. Tetrahedron 1965, 21, 2671. (b) Arnone, A.; Cardillo, R.; Nasini, G.; de Pava, O. V. J. Chem. Soc., Perkin Trans. 1 1991, 733. (c) Arnone, A.; Cardillo, R.; Modugno, V.; Nasin, G. Gazz. Chim. Ital. 1991, 121, 345. (d) Lee, I.-K.; Jeong, C. Y.; Cho, S. M.; Yun, B. S.; Kim, Y. S.; Yu, S. H.; Koshino, H.; Yoo, I. D. J. Antibiot. 1996, 49, 821. (e) Dufresne, C.; Young, K.; Pelaez, F.; Gonzalez, A.; Valentino, D.; Graham, A.; Platas, F.; Bernard, A.; Zink, D. J. Nat. Prod. 1997, 60, 188.

76 (a) Kelner, M. J.; McMorris, T. C.; Estes, L.; Rutherford, M.; Montoya, M.; Goldstein, J.; Samson, K.; Starr, R.; Taetle, R. Biochem. Pharmacol. 1994, 48, 403. (b) Tanaka, K.; Inoue, T.; Tezuka, Y.; Kikuchi, T. Chem. Pharm. Bull. 1996, 44, 273. (c) McMorris, T. C.; Kelner, M. J.; Wang, W.; Moon, S.; Taetle, R. Chem. Res. Toxicol. 1990, 3, 574.

207

77 (a) Weinreb, S. M.; McMorris, T. C.; Anchel, M. Tetrahedron Lett. 1971, 12, 3489. (b) McMorris, T. C.; Kelner, M. J.; Wang, W.; Diaz, M. A.; Estes, L. A.; Taetle, R. Experientia 1996, 52, 75. (c) McMorris, T. C.; Kelner, M. J.; Wang, W.; Yu, J.; Estes, L. A.; Taetle, R. J. Nat. Prod. 1996, 59, 896. (d) McMorris, T. C.; Yu, J.; Gantzel, P. K.; Istes, L. A.; Kelner, M. J. Tetrahedron Lett. 1997, 38, 1687. (e) MacDonald, J. R.; Muscoplat, C. C.; Dexter, D. L.; Mangold, G. L.; Chen, S. F.; Kelner, M. J.; McMorris, T. C.; Von Hoff, D. D. Cancer Res. 1997, 57, 279. (f) McMorris, T. C.; Yu, J.; Ngo, H.-T.; Wang, H.; Kelner, M. J. J. Med. Chem. 2000, 43, 3577.

78 These studies are being undertaken by MGI Pharma. For an extensive compilation of relevant literature references, see: http://www.mgipharma.com/Products_in_Development/MGI_114/body_mgi_114. html.

79 (a) Matsumoto, T.; Shirahama, H.; Ichihara, A.; Shin, H.; Kawawa, S.; Sakan, F.; Matsumoto, S.; Nishida, S. J. Am. Chem. Soc. 1968, 90, 3280. (b) Matsumoto, T.; Shirahama, H.; Ichihara, A.; Shin, H.; Kagawa, S.; Sakan, F.; Miyano, K. Tetrahedron Lett. 1971, 12, 2049. (c) Kigoshi, H.; Imamura, Y.; Mizuta, K.; Niwa, H.; Yamada, K. J. Am. Chem. Soc. 1993, 115, 3056. (d) Kinder, F. R., Jr.; Bair, D. W. J. Org. Chem. 1994, 59, 6965. (e) Padwa, A.; Curtis, E. A.; Sandanayaka, V. P. J. Org. Chem. 1997, 62, 1317. (f) McMorris, T.; Hu, Y.; Yu, J.; Kelner, M. J. J. Chem. Soc., Chem. Commun. 1997, 315. (g) Pirrung, M. C.; Kaliappan, K. P. Org. Lett. 2000, 2, 353. (h) Brummond, K. M.; Lu, J. L.; Petersen, J. J. Am. Chem. Soc. 2000, 122, 4915. (i) Armone, A.; Merlini, L.; Nasini, G.; de Pava, O. V.; Zunino, F. J. Chem. Soc. Perkin Trans. 1 2001, 610.

80 (a) Utimoto, K.; Tamura, M.; Sisido, K. Tetrahedron 1973, 29, 1169. (b) Maercker, A.; Daub, V. E. E. Tetrahedron 1994, 50, 2439.

81 (a) Thiemann, T.; Kohlstruck, S.; Schwaer, G.; de Meijere, A. Tetrahedron Lett. 1991, 32, 3483.(b) Braese, S.; de Meijere, A. Angew. Chem. 1995, 107, 2741. (c)Yuecel, B.; Arve, L.; de Meijere, A. Tetrahedron 2005, 61, 11355.

82 Funk, R. L.; Aungst, R. A.; Chan, C. Org. Lett. 2001, 3, 2611.

83 (a) Curran, D. P. J. Am. Chem. Soc. 1982, 104, 4024. (b) Kozikowski, A. P.; Stein, P. D. J. Am. Chem. Soc. 1982, 104, 4023. (c) Kozikowski, A. P. Acc. Chem. Res. 1984, 17, 410. (d) Curran, D. P. Adv. Cycloadd. 1988, 1, 129.

84 (a) Henoch, F. E.; Hampton, K. G.; Hauser, C. R. J. Am. Chem. Soc. 1969, 91, 676. (b) Jung, M. E.; Blair, P. A.; Lowe, J. A. Tetrahedron Lett. 1976, 17, 1439. (c) Kofron, W. G.; Yeh, M.-K. J. Org. Chem. 1976, 41, 439.

85 (a) Iwata, C.; Ida, Y.; Miyashita, K.; Nakanishi, T.; Yamada, M. Chem. Pharm.

208

Bull. 1982, 30, 2738. (b) Karlsson, J. O.; Frejd, T. J. Org. Chem. 1983, 48, 1921. (c) Coates, R. M.; Muskopf, J. W.; Senter, P. A. J. Org. Chem. 1985, 50, 3541.

86 Frejd, T.; Karlsson, J. O. J. Org. Chem. 1983, 48, 1921.

87 (a) Magnus, P. D.; Nobbs, M. S. Synth. Commun. 1980, 10, 273. (b) Exon, C.; Nobbs, M. S.; Magnus, P. D. Tetrahedron 1981, 37, 4515.

88 Tanimoto, S.; Reddy, C. P. Synthesis 1987, 575.

89 Podder, R. K.; Sarkar, R. K.; Ray, S. C. Indian J. Chem. 1988, 27B, 530.

90 Hassner, A.; Rai, K. M. L. Synthesis 1989, 57.

91 Curran, D. P. J. Am. Chem. Soc. 1983, 105, 5826.

92 Hoppe, D.; Schmincke, H.; Kleemann, H.-W. Tetrahedron 1989, 45, 687.

93 (a) Sibi, M. P. Org. Prep. Proceed. Int. 1993, 25, 15. (b) Overhand, M.; Hecht, S. M. J. Org. Chem. 1994, 59, 4721.

94 Hoppe, D.; Hanko, R.; Brönneke, A.; Litchtenberg, F.; van Hülsen, E. Chem. Ber. 1985, 118, 2822.

209 VITA

Ronald A. Aungst, Jr.

Ronald, son of Ronald and Norma Aungst, was born on June 10, 1974 in

Williamsport, Pennsylvania. He attended Lycoming College in Williamsport,

Pennsylvania, where he graduated magne cum laude with Bachelor of Arts degrees in Chemistry and Biology in May of 1996. Ronald then went onto pursue a Ph.D. in synthetic organic chemistry at the Pennsylvania State University under the guidance of Professor Raymond L. Funk. His research has been directed towards the development of methodology for the stereoselective synthesis of (Z)-

2-acyl-2-enals and their application in natural product synthesis. In 2002, Ronald began his career in the industry working as a Senior Research Scientist at

Albany Molecular Research, Inc. (AMRI). In January of 2005, Ronald went to

Singapore for a two year term in order to help set up AMRI’s Singapore subsidiary.