Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
Copyright
By
Regan Andrew Jones
2009
The Dissertation Committee for Regan Andrew Jones Certifies that this is the
approved version of the following dissertation:
The Asymmetric Total Synthesis of (+)-Geniposide via Phosphine-
Catalyzed [3+2] Cycloaddition
Committee:
Michael J. Krische, Supervisor
Stephen F. Martin
Philip D. Magnus
Hung-wen Liu
Sean Kerwin
The Asymmetric Total Synthesis of (+)-Geniposide via Phosphine-
Catalyzed [3+2] Cycloaddition
by
Regan Andrew Jones, B. A.
Dissertation
Presented to the Faculty of the Graduate School of
The University of Texas at Austin
in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy
The University of Texas at Austin
May, 2009
Dedication
To my wife Laura for all her continual love and support.
v
Acknowledgements I want to first thank the God and Father of my Lord Jesus Christ, who brought me
here to Austin for graduate school, loved me, saved me, and has carried me throughout
this program. I also want to thank Him for all the wonderful people he has brought into
my life.
I want to thank my wife, parents, and mother in law for their constant love and
encouragement, as well as their emotional and financial support. I also want to thank
TiAnna and Charles for buying me a new copy of Microsoft Word. I am also grateful to
Ronnie Smith, Peter Webber, Soo-Bong Han, and Ming-Yu Ngai for all their help and
friendship.
I want to thank my research advisor Dr. Michael Krische for allowing me to be in
his group and for financially supporting me to do this research. I also want to thank him
for his kind encouragement and guidance.
Finally I would like to express my gratitude to Vanessa Williams and Soo-Bong
Han for proofreading my dissertation and to Eduardas Skucas for allowing me to run
some final experiments in his lab.
vi
The Asymmetric Total Synthesis of (+)-Geniposide via Phosphine-
Catalyzed [3+2] Cycloaddition
Publication No._____________
Regan Andrew Jones, Ph.D.
The University of Texas at Austin, 2009
Supervisor: Michael J. Krische
The iridoids are a large family of monoterpenoid natural products that possess a
wide range of biological activities. A great deal of research has already been done in the
field of iridoid total synthesis, but limitations still remain. Specifically, few syntheses of
iridoid -glycosides have been reported. This work describes the 14 step asymmetric
total synthesis of the iridoid -glycoside (+)-geniposide utilizing a phosphine-catalyzed
[3+2] cycloaddition as the key step. Other noteworthy steps in the synthesis include a
palladium-catalyzed kinetic resolution and a previously unutilized method for iridoid
glycosidation. In addition to describing the synthesis of (+)-geniposide, this dissertation
will also review 1) phosphine-catalyzed cycloaddition reactions and 2) previous
enantioselective total syntheses of iridoid glycosides.
vii
Table of Contents
Chapter 1 Review of Phosphine-Catalyzed Cycloadditions.........................................1
1.1 Introduction ......................................................................................................1
1.2 [3+2] Cycloaddition of Allenes and Alkynoates with Electron Deficient
Alkenes ..................................................................................................................1
1.2.1 Initial Report of Phosphine-Catalyzed [3+2] Cycloadditions of Allenes
with Electron Deficient Alkenes.....................................................................1
1.2.2 Mechanism of Phosphine-Catalyzed [3+2] Cycloaddition......................2
1.2.3 Regioselectivity of Phosphine-Catalyzed [3+2] Cycloaddition...............4
1.2.4 [3+2] Cycloaddition of Allenoates with Diethyl Maleate and Diethyl
Fumurate........................................................................................................5
1.2.5 Electron Deficient Alkynes as 1,3-Dipole Precursors in [3+2]
Cycloaddition.................................................................................................6
1.2.6 Substituted Allenoates and Alkynoates in [3+2] Cycloaddition..............8
1.2.7 [3+2] Cycloaddition Reactions with C60 Fullerene ............................... 10
1.2.8 [3+2] Cycloaddition for Spirocycle Formation -L-Glutamate Analogues
.................................................................................................................... 11
1.2.9 [3+2] Cycloaddition for Spirocycle Formation -Synthesis of (-)-Hinesol
.................................................................................................................... 12
1.2.10 [3+2] Cycloaddition of 1,1-Dicyanoalkenes....................................... 14
1.2.11 [3+2] Cycloaddition of Furanones ..................................................... 15
1.2.12 [3+2] Cycloaddition of 3-Substituted-Chromones and Quinoline-1,3-
Dicarboxylate............................................................................................... 16
1.2.13 [3+2] Cycloaddition Reactions Using Activated Allylic Bromides,
Acetates, and Carbonates ............................................................................. 17
viii
1.2.14 Chiral Auxilaries in [3+2] Cycloaddition ........................................... 20
1.2.15 Enantioselective Phosphine-Catalyzed [3+2] Cycloadditions ............. 21
1.2.16 Intramolecular Phosphine-Catalyzed [3+2] Cycloadditions................ 26
1.3 Phosphine-Catalyzed [3+2] Cycloaddition of Imines with Allenoates and
Alkynoates ........................................................................................................... 33
1.3.1 [3+2] Cycloaddition of N-Tosyl Imines and Allenoates ....................... 33
1.3.2 Mechanism of [3+2] Cycloaddition between N-Tosyl Imines and
Allenoates.................................................................................................... 34
1.3.3 Regioselectivity of [3+2] Cycloaddition between N-Tosyl Imines and
Allenoates.................................................................................................... 35
1.3.4 [3+2] Cycloaddition Between N-Tosyl Imines and Alkynoates ............ 36
1.3.5 [3+2] Cycloaddition Between Imines and Substituted Allenoates ........ 37
1.3.6 [3+2] Cycloaddition of Imines and Alkynyl Ketones ........................... 40
1.3.7 [3+2] Cycloaddition of Allenes and N-(thio)-phosphoryl Protected
Imines .......................................................................................................... 41
1.3.8 Enantioselective [3+2] Cycloaddition of Imines .................................. 42
1.4 [3+2] Cycloaddition for Pyrrole Synthesis ...................................................... 43
1.5 Phosphine-Catalyzed [4+2] Cycloadditions of Allenes.................................... 45
1.5.1 Initial Phosphine-Catalyzed [4+2] Cycloaddition................................. 45
1.5.2 Mechanism of Phosphine-Catalyzed [4+2] Cycloaddition.................... 45
1.5.3 Scope of Phosphine-Catalyzed [4+2] Cycloaddition ............................ 46
1.5.4 Enantioselective Phosphine-Catalyzed [4+2] Cycloaddition ................ 48
1.5.5 Phosphine-Catalyzed [4+2] Cycloaddition with 1,1-dicyanoalkenes .... 49
1.5.6 Phosphine-Catalyzed [4+2] Cycloaddition of Allenyl Ketones............. 51
1.6 Phosphine-Catalyzed [4+3] Cycloaddition ...................................................... 52
ix
1.7 Phosphine-Catalyzed [6+3] Cycloaddition ...................................................... 52
1.8 Phosphine-Catalyzed [8+2] Cycloaddition ...................................................... 53
1.9 Miscellaneous Cycloadditions......................................................................... 54
1.9.1 Phosphine-catalyzed Synthesis of 1,3-Dioxin-4-ylidenes ..................... 54
1.9.2 Phosphine-catalyzed [4+2] Cycloaddition of 3-Formylchromones ....... 56
1.9.3 Phosphine-Catalyzed Cycloaddition of Trienoates............................... 57
1.10 Conclusion.................................................................................................... 58
1.11 References .................................................................................................... 59
Chapter 2 Review of Enantioselective Total Syntheses of Iridoid Glycosides........... 63
2.1 Introduction .................................................................................................... 63
2.2 General Discussion of Iridoid Glycoside Formation ........................................ 64
2.2.1 Koenigs-Knorr Type Glycosidation of Iridoids .................................... 64
2.2.2 2nd Strategy for Iridoid Glycoside Synthesis ........................................ 67
2.3 Review of Enantioselective Iridoid Glycoside Syntheses................................. 70
2.3.1 Introduction......................................................................................... 70
2.3.2 Enantioselective Total Synthesis of (-)-Loganin .................................. 71
2.3.3 Enantioselective Total Synthesis of (+)-Semperoside A....................... 73
2.3.4 Enantioselective Total Synthesis of (-)-Brasoside and (-)-Littoralisone 74
2.4 Conclusions .................................................................................................... 77
2.5 References ...................................................................................................... 77
Chapter 3 Intramolecular Approach to (+)-Geniposide ............................................ 80
3.1 Intramolecular Cycloaddition Retrosynthetic Analysis .................................... 80
3.2 Coumalate Intramolecular Cycloaddition Substrate......................................... 81
x
3.2.1 Coumalate Intramolecular Cycloaddition Synthesis ............................. 82
3.2.2 Attempted Coumalate Intramolecular Cycloaddition............................ 82
3.3 Pyranone Intramolecular Cycloaddition Substrate ........................................... 84
3.3.1 Design of 1st Generation Pyranone Intramolecular Cycloaddition
Substrate ...................................................................................................... 84
3.3.2 Synthesis of 1st Generation Pyranone Intramolecular Cycloaddition
Substrate ...................................................................................................... 84
3.3.3 1st Generation Pyranone Intramolecular Cycloaddition ........................ 85
3.3.4 Design of 2nd Generation Pyranone Intramolecular Cycloaddition
Substrate ...................................................................................................... 86
3.3.5 Synthesis of 2nd Generation Pyranone Intramolecular Cycloaddition
Substrate ...................................................................................................... 86
3.3.6 Cycloaddition of 2nd Generation Pyranone Intramolecular Substrate.... 87
3.3.7 Stereochemical Determination of Cycloaddition Product ..................... 88
3.3.8 Transition State Model for Intramolecular [3+2] Cycloaddition........... 89
3.4 Elaboration Of Cycloaddition Product to (+)-Geniposide ................................ 90
3.4.1 Retrosynthetic Analysis for Intramolecular [3+2] Cycloadduct ............ 90
3.4.1 Acetal Opening ................................................................................... 91
3.4.2 Alkene Isomerization to Direct Regiochemistry in Acetal Opening...... 92
3.4.2 Alkene Isomerization via Diol Elimination.......................................... 93
3.4.3 Alkene Isomerization via Base Mediated Epoxide Opening ................. 94
3.4.4 Proposal of New Synthetic Route to (+)-Geniposide............................ 95
3.5 Experimental Procedures ................................................................................ 96
3.6 1H and 13C NMR Spectra .............................................................................. 112
3.7 References .................................................................................................... 122
xi
Chapter 4 Intermolecular Approach to (+)-Geniposide .......................................... 124
4.1 Intermolecular Cycloaddition Retrosynthetic Analysis .................................. 124
4.2 Intermolecular [3+2] Cycloaddition Reaction................................................ 125
4.2.1 Intermolecular [3+2] Cycloaddition with Butynoate .......................... 125
4.2.2 Intermolecular [3+2] Cycloaddition with Allenoate ........................... 125
4.2.3 Stereochemical Determination of [3+2] Cycloadduct......................... 127
4.2.4 Regiochemical Analysis of Intermolecular [3+2] Cycloaddition ........ 128
4.2.5 Diastereoselectivity of Intermolecular [3+2] Cycloaddition ............... 130
4.3 Palladium-Catalyzed Kinetic Resolution ....................................................... 130
4.3.1 General Scheme for Enantioselective Synthesis ................................. 130
4.3.2 Mechanistic Outline of Palladium-Catalyzed Kinetic Resolution ....... 130
4.3.3 Precedent for Palladium-Catalyzed Kinetic Resolution ...................... 131
4.3.4 Palladium-Catalyzed Kinetic Resolution Optimization ...................... 132
4.3.5 Determination of Absolute Stereochemistry....................................... 133
4.3.6 Transition State Model for Kinetic Resolution................................... 134
4.4 Retrosynthetic Analysis of (+)-Geniposide from Cycloadduct ....................... 136
4.5 One Carbon Homologation of [3+2] Cycloadduct ......................................... 137
4.6 Reduction of Ethyl Ester ............................................................................... 138
4.6.1: Selectivity of Ester Reduction .......................................................... 138
4.6.2: Optimization of DIBAL-H Reduction............................................... 139
4.7 Esterification of the Nitrile............................................................................ 140
4.7.1 Discussion of Classical Esterification Methods.................................. 140
4.7.2 Platinum-Catalyzed Nitrile Hydration................................................ 141
4.7.3 Esterification of Amide ..................................................................... 141
xii
4.8 Introduction of the β-Glycoside..................................................................... 143
4.8.1 Discussion of Iridoid Glycoside Formation........................................ 143
4.8.2 Glycosidation using trichloracetimidate............................................. 143
4.8.3 Formation of glycosidation substrate ................................................. 144
4.8.4 Organotin-Catalyzed Deprotection .................................................... 145
4.8.5 Successful Glycosidation of Lactol.................................................... 146
4.9 Global Deprotection...................................................................................... 146
4.10 Conclusion.................................................................................................. 147
4.11 Experimental Procedures............................................................................. 149
4.12 1H and 13C NMR Spectra and HPLC Traces................................................ 165
4.13 References .................................................................................................. 180
Vita............................................................................................................................. 182
xiii
List of Tables
Table 1.1: First Intermolecular phosphine-catalyzed [3+2] cycloaddition........................2
Table 1.2: Butynoates as 1,3-dipole precursor in phosphine-catalyzed [3+2]
cycloaddition...................................................................................................................7
Table 1.3: Lu�s synthesis of spirocyclic compounds .....................................................13
Table 1.4: One pot [3+2] cycloaddition/aldehyde malonitrile condensation ..................15
Table 1.5: Scope of [3+2] Cycloaddition reactions with allylic bromides, acetates, and
carbonates .....................................................................................................................19
Table 1.6: [3+2] Cycloaddition of 1,1-dicyanoalkenes with activated allylic tert-butyl
carbonates .....................................................................................................................20
Table 1.7: [3+2] cycloaddition using chiral auxiliaries..................................................21
Table 1.8: First Enantioselective [3+2] Cycloaddition ..................................................22
Table 1.9: Enantioselective [3+2] cycloaddition with β-substituted enones ...................23
Table 1.10: Enantioselective [3+2] cycloaddition with phosphine-containing α-amino
acids..............................................................................................................................24
Table 1.11: Enantioselective [3+2] cycloadditions with allenyl ketones ........................26
Table 1.12: First intramolecular [3+2] cycloaddition ....................................................28
Table 1.13: [3+2] cycloaddition for synthesis of coumarins ..........................................30
Table 1.14: Intramolecular [3+2] cycloaddition of aromatic allylic bromides................32
Table 1.15: Intramolecular [3+2] cycloaddition of aliphatic allylic bromides ................33
Table 1.16: Initial studies of [3+2] cycloadditions with N-tosyl imines .........................34
Table 1.17: [3+2] Cycloaddition of Tosyl Imines with butynoates ................................37
xiv
Table 1.18: [3+2] cycloaddition of substituted allenoates with N-tosyl imines ..............38
Table 1.19: [3+2] cycloaddition of substituted allenoates with various imines ..............39
Table 1.20: PPh2Me-Catalyzed [3+2] cycloaddition of substituted allenoates with various
imines ...........................................................................................................................39
Table 1.21: [3+2] cycloaddition of alkynyl ketones with N-tosyl imines .......................40
Table 1.22: [3+2] cycloaddition of various alkynyl ketones with N-tosyl imines...........41
Table 1.23: Enantioselective [3+2] cycloaddition of DPP imines ..................................43
Table 1.24: Scope of pyrrole formation.........................................................................45
Table 1.25: Scope of imine in [4+2] cycloaddition........................................................47
Table 1.26: Benzyl substituted allenes in [4+2] cycloaddition.......................................47
Table 1.27: Enantioselective phosphine-catalyzed [4+2] cycloaddition.........................48
Table 1.28: Phosphine-catalyzed [4+2] cycloaddition for cyclohexene synthesis ..........49
Table 1.29: Substitution affects in [4+2] cycloaddition for cyclohexene synthesis ........50
Table 1.30: Phosphine-catalyzed synthesis of 1,3-dioxan-4-ylidenes ............................55
Table 3.1: Coumalate intramolecular cycloaddition reaction.........................................83
Table 3.2: 1st generation pyranone intramolecular cycloaddition...................................86
Table 3.3: 1st generation pyranone intramolecular cycloaddition...................................88
Table 4.1: Intermolecular [3+2] cycloaddition with butynoate ....................................125
Table 4.2: Kinetic resolution of allylic pivalate...........................................................133
Table 4.3: Reduction of ethyl ester .............................................................................140
xv
List of Figures
Figure 1.1: Regiochemical analysis of intermolecular [3+2] cycloaddition......................5
Figure 1.2: Regioselectivity in [3+2] cycloadditions with N-tosyl imines......................36
Figure 2.1: Iridoid natural products...............................................................................63
Figure 3.1: Single crystal X-ray diffraction analysis of cycloadduct 3.20......................88
Figure 3.2: Transition state model for intramolecular [3+2] cycloaddition ....................90
Figure 4.1: Single crystal X-ray diffraction analysis of [3+2] cycloadduct ..................127
Figure 4.2: Orbital analysis of 1,3-dipole....................................................................129
Figure 4.3: Orbital analysis of allylic pivalate dipolarophile .......................................129
Figure 4.4: Diastereochemical model for intermolecular [3+2] cycloaddition .............130
Figure 4.5: Determination of absolute stereochemistry ...............................................134
Figure 4.6: Model for predicting stereochemistry in asymmetric allylic alkylation......135
Figure 4.7: Transition state for kinetic resolution........................................................136
xvi
List of Schemes
Scheme 1.1: Failed [3+2] cycloaddition with unactivated alkenes...................................2
Scheme 1.2: Proposed mechanism for [3+2] cycloaddition reaction................................3
Scheme 1.3: Stepwise [3+2] Cycloaddition.....................................................................4
Scheme 1.4: Water-catalyzed intermolecular 1,2-proton transfer .....................................4
Scheme 1.5: [3+2] cycloaddition with dimethyl maleate and dimethyl fumurate .............6
Scheme 1.6: Mechanism of 1,3-dipole formation from electron deficient alkynes ...........6
Scheme 1.7: Tributylphosphine-catalyzed cycloaddition with diethyl maleate and
fumurate..........................................................................................................................8
Scheme 1.8: [3+2] cycloaddition of substituted allenoates with electron deficient alkenes
........................................................................................................................................9
Scheme 1.9: [3+2] Cycloaddition of functionalized allenoate with fumurate...................9
Scheme 1.10: [3+2] Cycloaddition of substituted alkynoates with diethyl fumurate ......10
Scheme 1.11: [3+2] Cycloaddition of functionalized alkynoates ...................................10
Scheme 1.12: [3+2] cycloaddition of [60] fullerene ......................................................11
Scheme 1.13: Synthesis of podophyllotoxin derivative .................................................11
Scheme 1.14: Synthesis of conformationally restricted L-glutamate analogues .............12
Scheme 1.15: [3+2] cycloaddition of dehydroaminoacids .............................................12
Scheme 1.16: Total synthesis of (-)-hinesol ..................................................................14
Scheme 1.17: [3+2] Cycloaddition of 1,1-dicyanoalkenes.............................................14
Scheme 1.18: Phosphine-catalyzed condensation of benzaldehyde with malonitrile ......14
Scheme 1.19: [3+2] Cycloaddition with furanones........................................................16
xvii
Scheme 1.20: [3+2] Cycloaddition with sulfinylfuranone .............................................16
Scheme 1.21: [3+2] cycloaddition of 3-substituted chromones......................................17
Scheme 1.22: [3+2] cycloaddition of quinoline-1,3-dicarboxylate ................................17
Scheme 1.23: [3+2] Cycloaddition reaction with allylic bromides.................................17
Scheme 1.24: Proposed Mechanism of [3+2] cycloaddition with allylic bromides ........18
Scheme 1.25: Scope of dipolarophile in [3+2] cycloaddition of allylic bromides ..........19
Scheme 1.26: Dynamic kinetic asymmetric transformation...........................................25
Scheme 1.27: Total synthesis of (±)-hirsutene ...............................................................29
Scheme 1.28: Investigation of activating group in coumarin synthesis ..........................31
Scheme 1.29: Mechanism of [3+2] cycloadditions with N-tosyl imines ........................35
Scheme 1.30: Synthesis of pentabromopseudilin via [3+2] cycloaddition .....................37
Scheme 1.31: [3+2] Cycloaddition with N-(thio)-phosphoryl Protected Imine ..............42
Scheme 1.32: Pyrrole Synthesis via [3+2] cycloaddition...............................................43
Scheme 1.33: Mechanism of pyrrole formation.............................................................44
Scheme 1.34: Representative phosphine-catalyzed [4+2] cycloaddition........................45
Scheme 1.35: Mechanism of [4+2] cycloaddition .........................................................46
Scheme 1.36: Formal synthesis of (±)-alstonerine and (±)-macroline ............................48
Scheme 1.37: Mechanistic explanation of [4+2] phosphine effects ...............................50
Scheme 1.38: Cycloaddition of allenyl ketones.............................................................51
Scheme 1.39: Mechanism of [4+2] cycloaddition/dimerization of ketones ....................52
Scheme 1.40: Phosphine-catalyzed [4+3] cycloadditions ..............................................52
Scheme 1.41: Phosphine-catalyzed [6+3] cycloaddition................................................53
xviii
Scheme 1.42: Phosphine-Catalyze [8+2] cycloaddition.................................................53
Scheme 1.43: Mechanism of phosphine-catalyzed [8+2] cycloaddition.........................54
Scheme 1.44: Mechanism of 1,3-dioxan-4-ylidenes formation......................................55
Scheme 1.45: Phosphine-catalyzed [4+2] cycloadditions with acetylene carboxylates ..56
Scheme 1.46: Mechanism of phosphine-catalyzed [4+2] cycloadditions with acetylene
carboxylates ..................................................................................................................57
Scheme 1.47: Phosphine-catalyzed [3+2] cycloaddition with trienoates........................58
Scheme 2.1: Koenigs-Knorr strategy for glycosidation .................................................65
Scheme 2.2: Glycosidation with α-acetobromoglucose .................................................65
Scheme 2.3: Glycosidation with 1,2-anhydro-α-D-glucose triacetate ............................66
Scheme 2.4: Problems with 1st glycosidation strategy...................................................66
Scheme 2.5: Formation of iridoid dimers ......................................................................67
Scheme 2.6: 2nd general glycosdiation strategy .............................................................68
Scheme 2.7: Glycosidation of loganin...........................................................................68
Scheme 2.8: Formation of iridoid dimers ......................................................................69
Scheme 2.9: Tietze glycosidation method .....................................................................69
Scheme 2.10: Mechanism of Glycosidation ..................................................................70
Scheme 2.11: Total synthesis of (-)-loganin..................................................................71
Scheme 2.12: Asymmetric hydroboration-oxidation .....................................................71
Scheme 2.13: Photoannulation of acetate......................................................................72
Scheme 2.14: Glycosidation in total synthesis of loganin..............................................73
Scheme 2.15: Synthesis of (+)-semperoside A ..............................................................73
xix
Scheme 2.16: Glycosidation reaction in (+)-semperoside A synthesis ...........................74
Scheme 2.17: Mercury-mediated cyclization in (+)-semperoside A synthesis ...............74
Scheme 2.18: Synthesis of (-)-Brasoside and (-)-littoralisone........................................75
Scheme 2.19: Proline-catalyzed Michael addition.........................................................76
Scheme 2.20: Final stages of (-)-brasoside Synthesis ....................................................76
Scheme 2.21: Final stages of (-)-littoralisone Synthesis ................................................77
Scheme 3.1: First generation retrosynthetic analysis of (+)-geniposide .........................81
Scheme 3.2: Coumalate intramolecular cycloaddition substrate ....................................81
Scheme 3.3: Synthesis of coumalate cycloaddition substrate alcohol ............................82
Scheme 3.4: Synthesis of coumalate cycloaddition substrate.........................................82
Scheme 3.5: Design of 1st generation pyranone intramolecular cycloaddition substrate.84
Scheme 3.6: Synthesis of 1st generation pyranone intramolecular cycloaddition substrate
......................................................................................................................................85
Scheme 3.7: Design of 2nd generation pyranone intramolecular cycloaddition substrate 86
Scheme 3.8: Synthesis of 2nd generation pyranone intramolecular cycloaddition alcohol
......................................................................................................................................87
Scheme 3.9: Synthesis of 2nd generation pyranone intramolecular cycloaddition substrate
......................................................................................................................................87
Scheme 3.10: Retrosynthetic analysis for intramolecular [3+2] cycloaddition product ..91
Scheme 3.11: Acetal opening to incorrect regioisomer..................................................92
Scheme 3.12: Alkene isomerization to prepare for acetal opening.................................93
Scheme 3.13: Alkene isomerization via dihydroxylation/elimination sequence .............94
xx
Scheme 3.14: Alkene isomerization via base-mediated epoxide elimination .................94
Scheme 3.15: Alkene isomerization via halohydrin.......................................................95
Scheme 4.1: Second generation retrosynthetic analysis of (+)-geniposide ...................125
Scheme 4.2: Intermolecular [3+2] cycloaddition with allenoate ..................................127
Scheme 4.3: Regioselectivity of intermolecular [3+2] cycloaddition...........................128
Scheme 4.4: Regiochemical transition state of intermolecular [3+2] cycloaddition .....129
Scheme 4.5: Enantioselective total synthesis of (+)-geniposide...................................130
Scheme 4.6: Mechanism of catalyzed kinetic resolution .............................................131
Scheme 4.7: Related palladium-catalyzed kinetic resolution .......................................132
Scheme 4.8: Related palladium-catalyzed reaction with alcohols ................................132
Scheme 4.9: Derivative of pivalate .............................................................................134
Scheme 4.10: Retrosynthetic analysis of (+)-geniposide to [3+2] cycloadduct ............137
Scheme 4.11: Formation of α,β-unsaturated nitrile......................................................137
Scheme 4.12: Attempted conversion of ketone to unsaturated ester.............................138
Scheme 4.13: Selectivity issues in reduction of α,β-unsaturated ester..........................139
Scheme 4.14: Esterification of nitrile ..........................................................................141
Scheme 4.15: Hydration of nitrile ...............................................................................141
Scheme 4.16: Conversion of amide to carboxylic acid ................................................142
Scheme 4.17: Allylic acetate protection ......................................................................143
Scheme 4.18: Iridoid glycoside formation...................................................................143
Scheme 4.19: Proposed glycosidatin of lactol .............................................................144
Scheme 4.20: Synthesis of lactol ................................................................................144
xxi
Scheme 4.21: Intramolecular tranesterification ...........................................................145
Scheme 4.22: Glycosidation of lactol..........................................................................146
Scheme 4.23: Attempted global deprotection..............................................................147
Scheme 4.24: Final hydrolysis and esterification ........................................................147
1
Chapter 1 Review of Phosphine-Catalyzed Cycloadditions 1.1 Introduction
Cycloaddition reactions are inherently powerful. They form multiple bonds in a
single manipulation, and allow for efficient entry into cyclic frameworks. Cycloadditions
become even more powerful when they can be conducted in a stereocontrolled fashion.
In light of the utility of cycloadditions, it is not surprising that as the field of
organocatalysis has advanced in recent years, a variety of new organocatalyzed
cycloadditions have been developed.1 One subset of these organocatalyzed
cycloadditions, are nucleophilic phosphine-catalyzed cycloadditions.2 The purpose of
this chapter is to give a general review of these phosphine-catalyzed cycloadditions.
1.2 [3+2] Cycloaddition of Allenes and Alkynoates with Electron Deficient Alkenes 1.2.1 Initial Report of Phosphine-Catalyzed [3+2] Cycloadditions of Allenes with
Electron Deficient Alkenes
The first phosphine-catalyzed [3+2] cycloaddition of electron deficient alkenes
with allenes was reported by Lu in 1995.3 Lu found that a [3+2] cycloaddition occurred
between latent 1,3-dipole ethyl-2,3-butadienoate (1.1) and dipolarophile ethyl-acrylate
(1.2) upon treatment with a catalytic amount of triphenylphosphine (Table 1.1, entry 1).
Two regioisomeric [3+2] cyclodoadducts 1.3 and 1.4 were formed in a combined 76%
yield and in a 75:25 regioisomeric ratio respectively. The reaction could also be
catalyzed by tributylphosphine to give products 1.3 and 1.4 in similar yield and
regioselectivity (entry 2). Furthermore, the [3+2] cycloaddition proceeded with several
2
other electron deficient alkenes to give cycloadduct products in good yields with modest
regioselectivities (entries 3-6).
EtO2C
PR3 (10 mol%)Benzene
25 °C
EWG
EtO2CEtO2C
1.1 1.2 1.3 1.4
EWG
EWG
Entry EWG (1.2) PR3 Ratio (1.3:1.4) % Yield 1 CO2Et PPh3 75:25 76 2 CO2Et PBu3 75:25 66 3 CO2Me PPh3 80:20 81 4 CO2Me PBu3 85:15 66 5 COMe PPh3 63:37 55 6 CN PPh3 83:17 79
Table 1.1: First Intermolecular phosphine-catalyzed [3+2] cycloaddition
Unfortunately, when the [3+2] cycloaddition was attempted with inactivated
olefins, such as 1-hexene 1.5, none of the desired [3+2] cycloaddition product was
observed (Scheme 1.1). However, the dimerization product of ethyl-2,3-butadienoate
(1.1), product 1.6, was isolated in 20% yield.
EtO2C
PPh3 (10 mol%)Benzene
25 °CEtO2C
1.1 1.5 1.6
CO2Et
Scheme 1.1: Failed [3+2] cycloaddition with unactivated alkenes 1.2.2 Mechanism of Phosphine-Catalyzed [3+2] Cycloaddition
Lu proposed a general mechanism for the [3+2] cycloadition (Scheme 1.2).3 This
mechanism begins with addition of phosphine to ethyl-2,3-butadienoate (1.1) to produce
1,3-dipole 1.7. The 1,3-dipole 1.7 is composed of two resonance structures, α-1.7 and γ-
1.7. The 1,3-dipole 1.7 then reacts with alkene dipolarophile 1.2 in a formal [3+2]
3
cycloadditon to produce zwitterionic intermediate 1.8. Next, zwitterionic intermediate
1.8 undergoes a 1,2-proton transfer to produce 1.9. Subsequent elimination of phosphine
affords the major product cycloadduct 1.3 and regenerates the phosphine catalyst
(Scheme 1.2).
R3P
R3P
EtO2C
R3P R3P
α−1.7 γ−1.7
EtO2C EtO2C
CO2Et
12
EtO2C
1.8
R3P12
EtO2C
1.9
R3P
EtO2C1.3
CO2Et
1.1 1.2
R3P
R3P
EtO2C1.7
R3P
EtO2C
1.7
HH
CO2Et CO2Et
CO2Et
1.2
Scheme 1.2: Proposed mechanism for [3+2] cycloaddition reaction Recently, several in depth computer modeling studies have been conducted
investigating the mechanism of the [3+2] cycloaddition.4,5,6 These studies confirm the
general mechanism originally proposed by Lu and provide some additional insight into
the reaction mechanism. Most notably, these studies revealed that the [3+2]
cycloaddition event between 1,3-dipole 1.7 and acrylate 1.2 to form cycloadduct
intermediate 1.9 is not a concerted reaction but rather a stepwise process (Scheme 1.3).
4
Ph3P
EtO2C
1.1
R3P
EtO2C
1.7
CO2Et
EtO2C
R3P CO2Et
EtO2C1.9
PR3
H
CO2Et
1.2 1.10
Scheme 1.3: Stepwise [3+2] Cycloaddition These studies also demonstrated that the 1,2 proton transfer of ylide 1.8 to
intermediate 1.9 is not an intramolecular reaction. Rather it is an intermolecular reaction
catalyzed by trace amounts of water in the reaction media (Scheme 1.4).
EtO2C
1.9
R3P CO2Et
EtO2C
1.8
PR3
H
CO2Et1,2 ProtonTransfer
EtO2C
1.11
R3P CO2EtH
H
EtO2C
1.8
PR3
H
CO2Et
OH H
HOH
EtO2C
R3P CO2EtH
1.12
OH H
Scheme 1.4: Water-catalyzed intermolecular 1,2-proton transfer 1.2.3 Regioselectivity of Phosphine-Catalyzed [3+2] Cycloaddition
Although the regioselectivity of the [3+2] cycloaddition reaction was modest, the
head to tail cycloadduct 1.3 was formed in approximately a 3:1 preference over
regioisomeric head to head cycloadduct 1.4 (Figure 1.1). This regiochemical preference
can be explained qualitatively using frontier molecular orbital (FMO) theory. The
HOMO coefficient of the 1,3-dipole 1.7 should be higher at the α-C since when the anion
is placed at the α-C it is stabilized by the adjacent electron withdrawing ester moiety, (α-
1.7). This stabilization is not present when the anion is placed on the γ-C, (γ-1.7).
Furthermore, the LUMO coefficient of the electron deficient alkene 1.2 should be highest
5
at the carbon β to the electron withdrawing group. To maximize HOMO-LUMO orbital
overlap, the reaction should proceed through transition state 1.13, which gives rise to the
major regioisomeric product 1.3. In addition to this general FMO analysis, computer
modeling studies of the regiochemical outcome of the cycloaddition have also been
conducted.5,7
EtO2C
PR3 (10 mol%)Benzene
25 °C
CO2Et
EtO2CEtO2C1.1 1.2 1.3
3
1.4
1
CO2Et
CO2Et
γ
α
Ph3P
γ
α
Ph3P
EtO2C
EtO2C
CO2Et
1.2
CO2Et
α−1.7
γ−1.7 1.2
EtO2C
1.3
CO2Et
EtO2C1.4
CO2Et
:
1.13
1.14
R3P
EtO2C
1.7
Figure 1.1: Regiochemical analysis of intermolecular [3+2] cycloaddition 1.2.4 [3+2] Cycloaddition of Allenoates with Diethyl Maleate and Diethyl Fumurate
In his initial study,3 Lu also showed that doubly-activated β-dipolarophiles
undergo facile [3+2] cycloaddition. Diethyl maleate (1.15) and diethyl fumurate (1.16)
underwent cycloaddition with ethyl-2,3-butadienoate (1.1) to afford cis-1.17 and trans-
1.18 in 46% yield and 67% yield respectively (Scheme 1.5). These results are notable, as
they show that the geometry of the electron deficient alkene dictates the stereochemical
outcome of the reaction.
6
EtO2C
PR3 (10mol%)Benzene
25 °C
46%
CO2Et CO2EtEtO2C1.1 1.15 cis-1.17
EtO2C
PR3 (10 mol%)Benzene
25 °C
67%
CO2Et CO2EtEtO2C
1.1 1.16 trans-1.18
CO2Et
CO2EtCO2Et
EtO2C
Scheme 1.5: [3+2] cycloaddition with dimethyl maleate and dimethyl fumurate 1.2.5 Electron Deficient Alkynes as 1,3-Dipole Precursors in [3+2] Cycloaddition
Lu also showed that electron deficient alkynes could serve as precursors to the
1,3-dipole intermediate 1.7.3 This realization was significant since electron deficient
alkynes are more readily available then electron deficient allenes. Specifically, Lu found
that ethyl-2-butynoate (1.19) could be converted into 1,3-dipole 1.7 upon treatment with
phosphine (Scheme 1.6). Mechanistically, this is assumed to occur through addition of
phosphine to the β-position of alkyne 1.19 to produce vinyl anion 1.20. Vinyl anion 1.20
then undergoes a 1,3-proton transfer to form the 1,3-dipole 1.7. It has recently been
proposed that this 1,3-proton transfer is mediated by catalytic amount of water in the
reaction media.6
R3P R3P
1.20 1.7
EtO2C EtO2C
R3P
EtO2C1.7
EtO2C
Me H
H
1.19
PR3
Scheme 1.6: Mechanism of 1,3-dipole formation from electron deficient alkynes
Accordingly, the cycloaddition of ethyl-2-butynoate (1.19) and ethyl acrylate
(1.2) catalyzed by tributylphosphine afforded products 1.3 and 1.4 in a combined 85%
7
yield in an 89:11 rr (Table 1.2, entry 1). Ethyl-2-butynoate (1.19) also underwent
smooth [3+2] cycloaddition with methyl acrylate (1.2a) and acrylonitrile 1.2b (entries 2-
3). Unfortunately, when the reaction was conducted using methyl-vinyl-ketone (1.2c),
none of the desired cycloaddition products were isolated (entry 4). This failure resulted
from polymerization of methyl vinyl ketone (1.2b) upon treatment with
tributylphosphine.
PBu3 (10 mol%)Benzene
25 °C
EWG
EtO2CEtO2C1.19 1.2 1.3 1.4
EtO2C
EWG
EWG
Entry Substrate EWG (1.2) Products Ratio (1.3:1.4) % Yield 1 1.2 CO2Et (1.3:1.4) 89:11 85 2 1.2a CO2Me (1.3a:1.4a) 84:16 78 3 1.2b CN (1.3b:1.4b) 93:7 80 4 1.2c COMe (1.3c:1.4c) - -
Table 1.2: Butynoates as 1,3-dipole precursor in phosphine-catalyzed [3+2] cycloaddition
These butynoate cycloadditions were also attempted using triphenylphosphine as
catalyst. Unfortunately, only a trace amount of the cycloaddition products could be
observed at highly elevated reaction temperatures. It is believed that the more
nucleophilic trialkylphosphines are required in cycloadditions using butynoates because
triarylphosphines are not sufficiently nucleophilic to convert the butynoate into the
required 1,3-dipole intermediate (Scheme 1.6).8 Additionally, it was found that
triethylamine was unable to catalyze the reaction.
The phosphine-catalyzed [3+2] cycloaddition of ethyl-2-butynoate (1.19) with
diethyl maleate (1.15) and diethyl fumurate (1.16) was also attempted (Scheme 1.7).3 In
8
contrast to the cycloaddition with ethyl-2,3-butadienoate (1.1), both diethyl maleate
(1.15) and diethyl fumurate (1.16) produced the same product trans-1.18 in 91% yield
and 88% yield respectively. This surprising stereochemical result stems from rapid
isomerizaition of dimethyl maleate (1.15) to dimethyl fumurate (1.16) through the action
of tributylphosphine.9
MePBu3 (10mol%)
Benzene
25 °C
91%
CO2Et CO2EtEtO2C
1.19 1.15 trans-1.18
PBu3 (10 mol%)Benzene
25 °C
88%
CO2Et CO2EtEtO2C1.19 1.16 trans-1.18
CO2Et
CO2EtCO2Et
EtO2C
CO2Et
Me
CO2Et
Scheme 1.7: Tributylphosphine-catalyzed cycloaddition with diethyl maleate and fumurate After Lu described these initial results on the phosphine-catalyzed [3+2]
cycloaddition, numerous other studies of the cycloaddition by various groups were
reported. These reports will be reviewed in the following sections.
1.2.6 Substituted Allenoates and Alkynoates in [3+2] Cycloaddition
After his initial studies, Lu reported results on the phosphine-catalyzed [3+2]
cycloadditions using substituted allenoates and butynoates.10 For instance, methyl
substituted allenoate 1.21 was treated with triphenylphosphine in the presence of ethyl
acrylate (1.2) to afford [3+2] cycloaddition product 1.22 along with phosphine coupling
product 1.23 in a combined 59% yield in a 64:36 ratio, respectively (Scheme 1.8). The
9
cycloadduct 1.22 was formed as a 67:33 trans:cis mixture of diastereomers. Similar
results were obtained when vinyl sulfone 1.24 was used as the dipolarophile.
EtO2C
PPh3 (10 mol%)Benzene
25 °C
59%
CO2Et
EtO2C
1.21 1.2
Me
EtO2C
Me
CO2Et
1.22
64(trans:cis = 67:33)
Me
CO2Et
1.23
36
EtO2C
PPh3 (10 mol%)Benzene
25 °C
67%
SO2Ph
EtO2C1.21 1.24
Me
EtO2C
Me
SO2Ph
1.25
59(trans:cis = 65:35)
Me
SO2Ph
1.26
41
Scheme 1.8: [3+2] cycloaddition of substituted allenoates with electron deficient alkenes
Additionally, methyl substituted allenoate 1.21 was reacted with diethylfumurate
(1.16) to afford triester 1.27 in 69% yield as a single diastereomer (Scheme 1.9).
EtO2C
PPh3 (10 mol%)Benzene
25 °C
69%
CO2Et CO2EtEtO2C1.19 1.16 trans-1.27
CO2EtEtO2CMeMe
Scheme 1.9: [3+2] Cycloaddition of functionalized allenoate with fumurate Examples of substituted alkynoates in the [3+2] cycloaddition include the
phosphine-catalyzed [3+2] cycloaddition of ethyl-2-heptynoate (1.28) with
diethylfumurate (1.16) using tributylphosphine as catalyst to produce product 1.29 in
73% yield (Scheme 1.10). The product of phosphine-catalyzed isomerization of 1.28,
ethyl-2,4-heptadienoate (1.30), was also isolated in 11% yield.11
10
PBu3 (15 mol%)Benzene
25 °CCO2Et
CO2EtEtO2C
1.28 1.16 1.29
CO2EtEtO2C
H9C4
EtO2C
H9C4
Me CO2Et
1.3073% 11%
Scheme 1.10: [3+2] Cycloaddition of substituted alkynoates with diethyl fumurate Additionally, the cycloaddition of substituted alkynoate 1.31 with diethylfumurate
(1.16) gave cycloadduct 1.32 in 47% yield as a 5:1 mixture of trans:cis isomers (Scheme
1.11).
PBu3 (15 mol%)Benzene
25 °C
47%CO2Et
CO2EtEtO2C1.31 1.16 1.32
CO2EtEtO2C
EtO2C
OO OO
trans:cis = 5:1
Scheme 1.11: [3+2] Cycloaddition of functionalized alkynoates
1.2.7 [3+2] Cycloaddition Reactions with C60 Fullerene
Another interesting application of the phosphine-catalyzed [3+2] cycloaddition
reported simultaneously by Wu12 and Walton13 is the reaction of ethyl-2,3-butadienoate
(1.1) with [60] fullerene 1.33 to give cycloadduct 1.34 in 42-43% yield (Scheme 1.12).
Walton also reported that the reaction proceeded in 23% yield when ethyl-2-butynoate
(1.19) was used as the latent 1,3-dipole precursor.
PBu3 (10 mol%)Toluene
25 °C
42-43%CO2Et CO2Et
1.33 1.1 1.34
11
Scheme 1.12: [3+2] cycloaddition of [60] fullerene
Wu later elaborated on this reaction to synthesize a novel [60] fullerene derivative
of the antineoplastic natural product podophyllotoxin through the [3+2] cycloaddition of
allene 1.35 with [60] fullerene 1.33 to give derivative 1.36 in 70% yield (Scheme 1.13).14
PBu3 (10 mol%)Toluene
25 °C
70%
1.36
1.33
O
O
O
O
OCH3OCH3
H3CO
O
O
O
O
O
OCH3OCH3
H3CO
O
O
O
1.35
Scheme 1.13: Synthesis of podophyllotoxin derivative 1.2.8 [3+2] Cycloaddition for Spirocycle Formation - L-Glutamate Analogues
Pyne has used the phosphine-catalyzed [3+2] cycloaddition reaction to synthesize
a variety of analogs of therapeutically useful molecules. For example, conformationally
restricted L-glutamate analogs were synthesized through the phosphine-catalyzed [3+2]
cycloaddition of ethyl-2,3-butadienoate (1.1) and oxazolidinone 1.37 (Scheme 1.14). 15
Cycloaddition products 1.38 and 1.39 were isolated in 49% yield and 17% yield
respectively along with a 27% yield of butadienoate self cycloaddition product 1.6
(Scheme 1.1). Both 1.38 and 1.39 were formed in 77:23 dr. Cycloadducts 1.38 and 1.39
were subsequently converted into L-glutamate analogues.
12
EtO2C
PBu3 (10 mol%)Benzene
25 °C
1.1
BzN O
O
H Ph
BzN O
O
H Ph
EtO2CBzN O
O
H Ph
CO2Et
1.37 1.38
49%
1.39
17%
Scheme 1.14: Synthesis of conformationally restricted L-glutamate analogues Pyne was also able to synthesize L-glutamate analogues through a
tributylphosphine-catalyzed [3+2] cycloaddition of ethyl-2-butynoate (1.19) with
dehydroaminoacid 1.40 (Scheme 1.15).16 Notably, only one regioisomeric product 1.41
was formed in 98% yield. The high regioselectivity of this transformation can be
explained by FMO analysis. Specifically, the LUMO coefficient of the alkene of
dehydroamino acid 1.40 should be increased at the terminal position of the alkene due to
the electron withdrawing imino and ester functionalities. Consequently, the alkene is
highly polarized and reacts with high regioselectivity.
Me PBu3 (10 mol%)Benzene
25 °C
98%1.19 1.40
EtO2C
CO2MeNPh2C
N
CO2MeCPh2
1.41
CO2Et
Scheme 1.15: [3+2] cycloaddition of dehydroaminoacids
1.2.9 [3+2] Cycloaddition for Spirocycle Formation - Synthesis of (-)-Hinesol
In 2002 Lu reported the phosphine catalyzed [3+2]-cycloaddition of electron-
deficient exocyclic alkenes 1.42 with tert-butyl allenoate 1.41 to form spirocyclic
compounds 1.43 and 1.44 (Table 1.4).17 The reaction proceeded across a broad range of
substrates in high yield and high dr favoring cycloadduct 1.43 (Table 1.3, entries 1-10).
13
The tert-butyl allenoate 1.41 was used as the 1,3-dipole precursor instead of the more
commonly used ethyl-2,3-butadienoate (1.1) because the bulky tert-butyl group
significantly enhanced the diastereoselectivity of the transformation. Lu also showed that
the reaction would proceed when the corresponding tert-butyl butynoate was used as the
latent 1,3-dipole. However, these reactions proceeded in significantly lower yields.
PPh3 (10 mol%)Toluene
reflux
1.41 1.42 1.43 1.44
CO2tBu
O O O
CO2tBu
CO2tBu
Entry 1.42 R (Ratio: 1.43:1.44) % Yield
1 2
O
R
R=H R = OMe
(91:9) (92:8)
98 95
3 4 5
O
R
R = H R = Br
R = OMe
(95:5) (93:7) (92:8)
99 98 96
6
O
(80:20) 78
7 O
(78:22) 63
8 9 RN
O
R = Boc R = Ts
(92:8) (91:9)
92 93
10 BnN
NBn
O
(74:26) 90
Table 1.3: Lu�s synthesis of spirocyclic compounds Lu later showcased this method of spirocycle formation in the first total synthesis
of the natural product (-)-hinesol (1.47) (Scheme 1.16).18
14
MePPh3 (10 mol%)
Toluene
25 °C
60%1.421.45 1.46
O O
CO2tBu
CO2tBuMe Me
Me
Me
Me
MeOH
1.47
Scheme 1.16: Total synthesis of (-)-hinesol 1.2.10 [3+2] Cycloaddition of 1,1-Dicyanoalkenes
A phosphine-catalyzed [3+2] cycloaddition between ethyl-2,3-butadienoate (1.1)
and 2-benzylidenemalonitrile (1.48) was reported by Lu in 2006 (Scheme 1.17).19 The
reaction afforded cycloadduct 1.49 as a single regioisomer in 89% yield. Notably, at the
time of this report, this reaction represented the first example of a phosphine-catalyzed
[3+2] cycloaddition with a β-substituted dipolarophile that was not activated by two
electron withdrawing groups.
EtO2C
PPh3Toluene
25 °C
89%1.1 1.48 1.49
CNNC
Ph EtO2C Ph
CN
CN
Scheme 1.17: [3+2] Cycloaddition of 1,1-dicyanoalkenes Lu was able to develop a one pot three component coupling using this reaction.
This discovery was driven by an earlier report20 which showed that a condensation
reaction occurred between benzaldehyde (1.50) and malonitrile (1.51) upon treatment
with triphenylphosphine to produce dipolarophile (1.48) (Scheme 1.18).
Ph
OPPh3Toluene
reflux1.50 1.51 1.48
CNNC
Ph
CNNC
Scheme 1.18: Phosphine-catalyzed condensation of benzaldehyde with malonitrile
15
Thus when various aldehydes 1.52, malonitrile (1.51), and ethyl-2,3-butadienoate
(1.1) were treated with triphenylphosphine, modest to good yields of cycloadducts 1.53
could be isolated (Table 1.4). In this reaction, ethyl-2,3-butadienoate (1.1) had to be
added slowly by syringe pump addition, and molecular sieves were necessary to obtain
high yields. Although a variety of aryl aldehydes participated in the transformation
(entries 1-8), aliphatic aldehydes were not viable substrates (entry 9).
R
O
PPh3 (10 mol%)TolueneMol Sieves
reflux
1.52 1.51
CNNC
EtO2C
1.1 1.53
EtO2C Ph
CN
CN
Entry R (1.52) % Yield (1.53) 1 Ph 86 2 p-MeOC6H4 56 3 p-FC6H4 76 4 p-ClC6H4 78 5 α-Napthyl 69 6 2-Pyridyl 53 7 2-Furyl 74 8 Cinnamyl 26 9 n-propyl -
Table 1.4: One pot [3+2] cycloaddition/aldehyde malonitrile condensation 1.2.11 [3+2] Cycloaddition of Furanones
The phosphine-catalyzed [3+2] cycloaddition of ethyl-2-butadienoate 1.1 with
methyl furanone 1.54 was reported in 2008 by Ruano and Martín (Scheme 1.19).21 A
75% yield of the cis-fused bicycle 1.55 was isolated in 69% yield as a single diastereomer
when two equivalents of methyl furanone 1.54 were used. The diastereoselectivity of the
reaction is believed to be controlled by the acetal methoxy group. The success of this
cycloaddition was surprising since β-substituted unsaturated esters do not typically
16
participate in the [3+2] cycloaddition.3 This reaction probably proceeds because the
alkene of furanone 1.54 is within an electron poor ring system.
1.1
O
O
OMe
1.54
O
H
H
O
OMe
1.55
PPh3 (30 mol%)Benzene
25 °C
75%EtO2C EtO2C
Scheme 1.19: [3+2] Cycloaddition with furanones The authors also found that optically pure sulfinylfuranone 1.56 participated in
the cycloaddition to give adduct 1.57 in 96% yield as a single diastereomer (Scheme
1.20). The authors postulate that the diastereoselectivity of this reaction was controlled
completely by the ethoxy group of 1.56.
1.1
O
O
OEt
1.56
O
TolOS
H
O
OEt
1.57
PPh3 (30 mol%)Benzene
25 °C
96%EtO2C EtO2C
SO
Tol
Scheme 1.20: [3+2] Cycloaddition with sulfinylfuranone 1.2.12 [3+2] Cycloaddition of 3-Substituted-Chromones and Quinoline-1,3-
Dicarboxylate
In 2000, Ishar reported the [3+2] cycloaddition of ethyl-2-butadienoate (1.1) with
3-substitued-chromones 1.58 to give products 1.59 in 72-74% yield (Scheme 1.21).22 In
this reaction the [3+2] cycloaddition was followed by spontaneous deformylation of the
3-formyl residue.
17
PPh3Benzene
80 °CO
O H
O
O
O
R
CO2Et
H
H
3 examples74-72% yield
R = H, Cl, Me1.58 1.591.1
CO2Et
Scheme 1.21: [3+2] cycloaddition of 3-substituted chromones In 2008 a related study Beifuβ and Al-Masoudi showed that structurally related 4-
quinoline-1,3-dicarboxylate 1.60 also participated in effective [3+2] cycloaddition to
provide cycloadduct 1.61 in 60% yield (Scheme 1.22).23
PPh3Benzene
23 °C
60%N
OCO2Et
N
O
R
CO2Et
CO2Et
H
1.60 1.61
CO2Et EtO2C
Scheme 1.22: [3+2] cycloaddition of quinoline-1,3-dicarboxylate 1.2.13 [3+2] Cycloaddition Reactions Using Activated Allylic Bromides, Acetates,
and Carbonates
In 2003, Lu described a procedure for phosphine-catalyzed [3+2] cycloaddition of
activated allylic bromide 1.62 with succinimide 1.63 to produce cycloadduct 1.64 in 88%
yield (Scheme 1.23).24
Br
EtO2C
1.62
EtO2CNPh
O
O
1.63
NPh
H
H
O
O
1.64
PPh3 (10 mol%)K2CO3 (1.5 Equiv)
Toluene
90 °C
88%
Scheme 1.23: [3+2] Cycloaddition reaction with allylic bromides
18
Lu proposed a general mechanism for this reaction that begins with SN2
displacement of the bromide in compound 1.62 with phosphine to produce phosphonium
salt 1.65 (Scheme 1.24). Deprotonation of the phosphonium salt 1.65 with potassium
carbonate provides 1,3-dipole 1.66 which participates in a [3+2] cycloaddition with
succinimide 1.63 to provide the cycloadduct 1.64.
Br
EtO2CPPh3
1.62
EtO2C
1.65
Ph3P BrH
EtO2C
1.66
Ph3P
K2CO3EtO2C
NPh
O
O1.63 NPh
H
H
O
OPPh31.64
Scheme 1.24: Proposed Mechanism of [3+2] cycloaddition with allylic bromides The scope of the [3+2] cycloaddition with phthalimide 1.63 was also investigated
with regard to the allylic compound 1.67 (Table 1.5). The reaction proceeded well with
phenyl (entry 2) and alkyl (entry 3) substituted allylic bromides to produce adducts 1.68a
and 1.68b in decent yield and in greater than 93:3 dr. Furthermore, it was discovered that
allylic acetate 1.67c (entry 4) and allylic tert-butyl carbonate 1.67d (entry 5) participated
in the cycloaddition effectively. Notably, in the cycloaddition of tert-butyl carbonate
1.67d, no potassium carbonate was required since the tert-butyl carbonate degrades to
form tert-butoxide in situ.
19
R
EtO2C
1.67
EtO2CNPh
O
O
1.63
NPh
H
H
O
O
1.68
PPh3 (10 mol%)K2CO3 (1.5 Equiv)
TolueneR'R'
Entry Allylic Substrate R R� T °C Product % Yield (1.68) 1 1.67 Br H 90 °C 1.68 88 2 1.67a Br Ph 110 °C 1.68a 68 3 1.67b Br nPr 110 °C 1.67b 60 4 1.67c OAc H 70 °C 1.67c 76 5 1.67d OBoc H 110 °C 1.67d 74
Table 1.5: Scope of [3+2] Cycloaddition reactions with allylic bromides, acetates, and carbonates
Lu also found that the [3+2] cycloaddition proceeded with other dipolarophiles
such as chalcone 1.69 and diester 1.70 to give the corresponding cycloadducts 1.71 and
1.72 in good yield (Scheme 1.25).
Br
EtO2C
1.62
PPh3 (10 mol%)K2CO3 (1.5 Equiv)
Toluene
CO2Ph
CO2Et
1.70
O
1.69
O
1.71
CO2Et
CO2Et
CO2Ph
EtO2C
1.72
Br
EtO2C
1.62
PPh3 (10 mol%)K2CO3 (1.5 Equiv)
Toluene
110 °C
70%
110 °C
72%
Scheme 1.25: Scope of dipolarophile in [3+2] cycloaddition of allylic bromides In a subsequent study, Lu reported that activated allylic tert-butyl carbonate 1.73
underwent effective cycloaddition with β-substituted 1,1-dicyanoalkenes 1.74 to give
cycloadducts 1.75 (Table 1.6).25 Electron neutral, electron rich, and electron deficient
aryl substituted 1,1-dicyanoalkenes participated in the cycloaddition in excellent yield
20
(entries 1-3). n-Propyl susbstituted 1,1-dicyanoalkene also underwent [3+2]
cycloaddition in high yield (entry 4). However the bulkier isopropyl substrate completely
suppressed the cycloaddition reaction (entry 5).
BocO
EtO2C
1.73 1.74 1.75
EtPh2P (10 mol%)TolueneNC CN
R
EtO2C
CNCN
R25 °C
Entry (1.74) R % Yield (1.75) 1 Ph 90 2 4-MeO-C6H4- 96 3 4-NO2-C6H4- 87 4 n-Pr 89 5 i-Pr -
Table 1.6: [3+2] Cycloaddition of 1,1-dicyanoalkenes with activated allylic tert-butyl carbonates 1.2.14 Chiral Auxilaries in [3+2] Cycloaddition
Another interesting addition to this methodology reported by Pyne was the use of
chiral auxiliaries in the [3+2] cycloaddition reaction.26 Remarkably, the auxiliaries
affected both the regioselectivity and the diastereoselectivity of the reaction. For
instance, the cycloaddition between ethyl-2-butynoate (1.19) and hydantoin 1.76
produced cycloaddition products 1.77 and 1.78 in 81% overall yield, in a regioisomeric
ratio of 98:2 respectively (Table 1.7, entry 1). However, when chiral auxiliary 1.79 was
used the regioselectivity of the reaction was reversed (entry 2). The reaction produced
both products 1.77 and 1.78 in a combined 61% yield in an approximate 11:89
regioisomeric ratio. Product 1.77 was formed as a 1:1 mixture of diastereomers and
product 1.78 was formed in >98% de. Pyne proposed that the regiochemical reversal in
the cycloaddition is caused by the electronic effects of the chiral auxiliary rather than
21
sterically driven. This proposal is supported by the fact that when camphor sultam
auxiliary 1.80 was used in the cycloaddition, the opposite regioisomer, product 1.77, was
formed exclusively in 74% yield albeit as a 1:1 mixture of diastereomers (entry 3).27
Notably, Pyne investigated these reactions in his work involving the synthesis of
carbocyclic hydantoins27 and his work in the synthesis of 2-azaspiro[4.4]nonan-1-ones.28
Me PBu3 (10 mol%)Benzene
25 °C
1.19 X = OEt
BnN NBn
O
BnN NBn
O
BnN NBn
O
1.76 1.77 1.78
COX O O O
ON
O
Bn
MeMe
NS OO
1.80, X =1.79, X =
CO2X
XO2C
Entry Alkyne Ratio (1.77:1.78) % Yield % de (1.77:1.78)
1 1.19 98:2 81% - 2 1.79 11:89 61% 0:98 3 1.80 100:0 74% 0:0
Table 1.7: [3+2] cycloaddition using chiral auxiliaries 1.2.15 Enantioselective Phosphine-Catalyzed [3+2] Cycloadditions
The first example of an enantioselective phosphine-catalyzed [3+2] cycloaddition
using chiral tertiary phosphines was reported by Zhang in 1997.29 Zhang found that
novel tertiary chiral monophosphine 1.82 gave superior results compared to other known
chiral phosphines in the enantioselective phosphine-catalyzed [3+2] cycloaddition of
ethyl-2,3-butadienoate (1.1) with various electron deficient alkenes 1.81 to form
cycloadducts 1.83 and 1.84 (Table 1.8). The best enantioselectivities and
22
regioselectivities were obtained when the ester group (E) of 1.81 was iso-butyl rather
than methyl, ethyl, or tert-butyl (entries 1-4). The best results were obtained when the
reaction temperature was decreased to 0 °C. Under these conditions the cycloadduct 1.83
could be obtained in 88% yield and 93% ee as a single regioisomer (entry 5). The
cycloaddition was also attempted using diethyl maleate and diethyl fumurate as the
dipolarophile though modest yields and enantioselectivities were obtained (entries 6 and
7). While Zhang had developed the first enantioselective phosphine-catalyzed [3+2]
cycloaddition, the scope of the method was limited.
EtO2CE EEtO2C
1.1 1.81 1.83EtO2C
1.84
E
PPh iPriPr
1.82
R1 R2R2
R2
R1R1
Entry E R1 R2 Solvent T (°C) (1.83:1.84) % Yield % ee 1 COOEt H H Benzene rt 97:3 76 81 2 COOMe H H Benzene rt 96:4 87 79 3 COOiBu H H Benzene rt 100:0 92 88 4 COOtBu H H Benzene rt 95:5 75 88 5 COOiBu H H Toluene 0 100:0 88 93 6 COOEt H COOEt Toluene 0 - 49 79 7 COOMe COOMe H Benzene rt - 84 36
Table 1.8: First Enantioselective [3+2] Cycloaddition
After Zhang�s initial study, Fu reported an enantioselective phosphine-catalyzed
[3+2] cycloaddition using chiral tertiary phosphine 1.85 (Table 1.9).30 Phosphine 1.85
successfully catalyzed the [3+2] cycloaddition of ethyl-2,3-butadienoate (1.1) with a
broad range of β-substituted enones 1.86 to give the major the regioisomeric product 1.87
in good yield and high ee. Notably, this work represents the first example of a β-
substituted singly activated dipolarophile undergoing successful [3+2] cycloaddition.
23
Furthermore, these β-substituted enones 1.86 prefer to form the regioisomeric
cycloadduct 1.87 over cycloadduct 1.88. This regiochemical preference is opposite to
that observed in [3+2] cycloaddition involving non-β-substituted dipolarophiles.
Presumable the steric hindrance of the β-substituent causes this regiochemical shift.
EtO2C EtO2C
1.1 1.86 1.87
EtO2C
1.88
1.85
P tBu
R
O
R1
O
R1
R
R
O
R1Toluene
rt
Entry R (1.86) R1 (1.86) (1.87:1.88) % Yield % ee 1 Ph Ph 13:1 64 88 2 Ph 4-Cl-C6H4 7:1 76 82 3 Ph 4-Me-C6H4 20:1 61 87 4 Ph 4-MeO-C6H4 >20:1 54 88 5 4-Cl-C6H4 Ph 9:1 74 87 6 4-MeO- C6H4 Ph 10:1 67 87 7 2-furyl Ph 3:1 69 88 8 2-quinolyl Ph 20:1 52 88 9 4-Cl-C6H4 2-(5-Me-furyl) >20:1 54 89
10 Ph 2-thienyl 6:1 74 90 11 C≡C-C5H11 Ph 6:1 65 85 12 C≡C-TES Ph >20:1 70 87 13 C5H11 Ph >20:1 39 75
Table 1.9: Enantioselective [3+2] cycloaddition with β-substituted enones Another example of an enantioselective [3+2] cycloaddition using a phosphine-
containing α-amino acid catalyst 1.89 was published by Miller in 2007.31 Benzyl
allenoate 1.90 underwent cycloaddition with a variety of electron deficient exocyclic
alkenes 1.91 to provide cycloadducts 1.92 and 1.93 in modest enantioselectivity and
excellent regioselectivity (Table 1.10). Both aromatic and heteroaromatic substituents
were tolerated in the reaction (Entries 1-4). However, a slight decrease in regioselectivity
was observed when an acyclic substrate was used (entry 5)
24
1.90 1.91 1.92 1.93
BnO2CO O O
CO2Bn
BocHNOMe
O
PPh2
1.89 (10 mol %)Toluene
-25 °C
CO2Bn
Entry (1.91) ( 1.92:1.93) % Yield % ee (1.92)
1
O
MeO
(99:1) 95 84
2
O
O
(94:6) 68 65
3
O
O
(>99:1) 75 76
4
O
NAc
(>99:1) 53 71
5
O
Me
(85:15) 75 70
Table 1.10: Enantioselective [3+2] cycloaddition with phosphine-containing α-amino acids Miller also established that a dynamic kinetic asymmetric transformation could be
achieved using catalyst 1.89 (Scheme 1.26). When racemic chiral allene 1.94 was
reacted with dipolarophile 1.95 using catalyst 1.89, (100 mol %), the cycloaddition
product 1.96 was isolated in 94% yield and 91% ee (Scheme 1.26). This result is possible
because the chirality of the allene 1.94 is destroyed in the formation of 1,3-dipole
intermediate 1.97. The reaction also proceeded at lower catalyst loading but in
significantly diminished yield.
25
1.94 1.95
Ph
O
BocHNOMe
O
PPh21.89 (100 mol %)
Toluene
-25 °C
94% Yield91% ee
PhMeH
R3P
BnO2C
1.97
CO2Bn
Me
CO2Bn
Ph
OPh
Me
1.96
Scheme 1.26: Dynamic kinetic asymmetric transformation Finally, researchers at Merck found that allenyl ketones 1.98 could be used as 1,3-
dipole precursors in asymmetric phosphine-catalyzed [3+2] cycloaddition with exocyclic
alkenes 1.99 to form spirocycles 1.100 and 1.101 (Table 1.11).32 This report was
significant since it was the first time an allenyl ketone had been used as a 1,3-dipole
precursor. After initial screening, it was found that the commercial chiral phosphine-
catalyst DIOP 1.102 provided excellent regioselectivity and moderate yield and
enantioselectivity in the [3+2] cycloadditions of allenyl ketones 1.98 with several
exocyclic alkenes 1.99 (entries 1-5).
26
1.98 1.99 1.100 1.101
O O ODIOP 1.102 (20 mol %)
Toluene
25 °CMe
O OMe
MeO
O
O
PPh2PPh2
Entry (1.99) (Ratio: 1.100:1.102) % Yield % ee (1.100)
1
O
(95:5) 58 61
2
O
MeO
(91:9) 64 77
3
MeO
O
(80:20) 73 53
4
MeO
O
(90:10) 82 46
5
O
O
BnO
(95:5) 84 71
Table 1.11: Enantioselective [3+2] cycloadditions with allenyl ketones 1.2.16 Intramolecular Phosphine-Catalyzed [3+2] Cycloadditions
In 2003 Krische reported the first example of an intramolecular phosphine-
catalyzed [3+2] cycloaddition.33 Specifically, 1,7-enynes 1.103 underwent
tributylphosphine-catalyzed intramolecular [3+2] cycloaddition to produce diquinane
products 1.104 in 71-86% yields (Table 1.12). All the products were formed in >95:5
d.e. except for in the case of the oxygen tethered substrate where an epimeric product was
also isolated in 10% yield (entry 6). The reaction was compatible with cyclopropyl
(entries 5-7), aryl (entry 1), heteroaryl (entry 2), and thioester functionalities (entries 3,8).
27
However, 1,7-enynes that employed enoates as the dipolarophile did not participate in the
reaction to a significant extent.
O
R1
O
R1
H
H
OR2
O R2 PBu3 (10 mol%)EtOAc110 °C
Sealed Tube
1.103 1.104
Entry (1.103) Product (1.104) % Yield
1
O PhO
MeO
MeO
O
H
H
OPh
76
2 O
O
MeO
O
MeO
O
H
H
O O
74
3
O SEtO
MeO
MeO
O
H
H
OSEt
78
4
O MeO
MeO
MeO
O
H
H
OMe
77
5
OO
MeO
MeO
O
H
H
O
86
28
6
O
OO
MeO
MeO
O
O
H
H
O
75
7 O
O
Me
Me
O
H
H
O
71
8 O SEt
O
Me
Me
O
H
H
OSEt
75
Table 1.12: First intramolecular [3+2] cycloaddition This intramolecular phosphine-catalyzed [3+2] cycloaddition reaction was
subsequently utilized by Krische in the total synthesis of the linear triquinane natural
product (±)-hirsutene 1.105 (Scheme 1.27).34 [3+2] cycloaddition of substrate (E)-1.106
gave product 1.107 as a single diastereomeric product in 88% yield. Product 1.107 was
further elaborated to the natural product (±)-hirsutene 1.105. A noteworthy aspect of this
work was the discovery that when diastereomeric cycloaddition substrate (Z)-1.106 was
subjected to the [3+2] cycloaddition conditions, an epimeric cycloadduct epi-1.107 was
isolated in 73% yield. This result revealed that the intramolecular [3+2] cycloaddition is
a stereospecific transformation.
29
H
HMe
Me
MeH
Hirsutene
MeMe
Me OMeO2C
Me
88%
H
HMe
Me
MeMe
O
110 °CSealed Tube
MeO2CPBu3 (10 mol%)
EtOAc
MeMe
MeMeO2C
73%
H
HMe
Me
MeMe O
110 °CSealed Tube
MeO2CPBu3 (10 mol%)
EtOAc
(E)-1.106
O
Me
(Z)-1.106
1.107
epi-1.107
1.105
Scheme 1.27: Total synthesis of (±)-hirsutene Kwon has also reported an intramolecular phosphine catalyzed [3+2]
cycloaddition to produce an assortment of highly functionalized coumarins (Table
1.13).35 Substrates 1.108 could be converted to cyclopentene-fused dihydrocoumarins
1.109 when treated with a catalytic amount of tributylphosphine. Both electron neutral
(entries 1-2), electron donating (entries 3-6) and electron withdrawing (entries 6-7) aryl
substituents were tolerated in the reaction. However, the 5-nitro substrate produced the
corresponding cycloadduct in only 9% yield (entry 8).
30
rt
PBu3 (20 mol%)THF
O
CO2Et
O O O
EtO2C
R R1.108 1.109
Entry (1.108) % Yield (1.109) 1 H 96 2 3-methyl 98 3 3-methoxy 74 4 4-methoxy 94 5 5-methoxy 70 6 5-fluoro 91 7 5-bromo 93 8 5-nitro 9
Table 1.13: [3+2] cycloaddition for synthesis of coumarins Kwon investigated the effects of changing the activating group of the alkene
dipolarophile in the cycloaddition and found that when sulfone 1.110 was used, a 63%
yield of the corresponding cycloadduct 1.111 could be isolated (Scheme 1.28).
Furthermore, when nitro-alkene 1.112 was used as substrate, only a 48% yield of product
1.113 could be isolated along with a 12% yield of tricyclic nitronate 1.114 when
triphenylphosphine was used as catalyst. This nitronate side product 1.114 could be
isolated in 62% yield when the reaction was conducted in benzene using tris(p-
fluorophenyl)-phosphine as catalyst. Kwon postulates two possible mechanisms for the
formation of this unexpected product 1.114 in her paper.35
31
rt
63%
PBu3 (20 mol%)THF
O
SO2Ph
O O O
PhO2S
rt
48%
PPh3 (20 mol%)THF
O
NO2
O O O
O2N
rt
48%
P(p-FC6H5)3(20 mol%)Benzene
O O
1.110 1.111
1.112 1.113
ONO
1.114
Scheme 1.28: Investigation of activating group in coumarin synthesis Finally, Tang and coworkers have published studies an intramolecular
phosphine-catalyzed [3+2] cycloaddition of allylic bromides.36,37 For example, allylic
bromide 1.115a participates in a intramolecular [3+2] cycloaddition to form
benzobicyclo[4.3.0] compound 1.116a and 1.117a in 95% yield in a 91:9 ratio,
respectively (Table 1.14, entry 1). Similarly, the electron poor p-bromo substrate 1.115b
(entry 2), and electron rich p-methoxy substrate 1.115c (entry 3) also underwent the
[3+2] cycloaddition in similar yield. Tang also found that when cesium carbonate was
used as base, the opposite alkene isomers 1.17 could be formed as the major product
(entries 3-6).37
32
Base (150 mol%)PPh3 (20 mol%)
Toluene
1.115
OEtO2C
Br
CO2Et
1.116
H
HCO2Et
EtO2C H
HCO2Et
EtO2C
1.117R R R
Entry R (1.115) Substrate Base Temp. (°C) Ratio (1.116:1.117) % Yield 1 H 1.115a Na2CO3 80 (91:9) 95 2 4-Br 1.115b Na2CO3 80 (90:10) 96 3 4-MeO 1.115c Na2CO3 80 (91:9) 96 4 H 1.115a Cs2CO3 80-90 (18:82) 96 5 4-Br 1.115b Cs2CO3 80-90 (18:82) 89 6 4-MeO 1.115c Cs2CO3 80-90 (19:81) 83
Table 1.14: Intramolecular [3+2] cycloaddition of aromatic allylic bromides Tang additionally showed that the aliphatic substrates could also be used in the
intramolecular [3+2] cycloaddition to produce bicyclo-[3.3.0] ring systems.36 To this
end substrate 1.118a could be converted to bicyclo-[3.3.0] ring system 1.119a in 73%
yield in >20:1 dr (
Table 1.15, entry 1). In the same way, an oxygen tethered substrate 1.118b proceeded to
give the cycloadduct 1.119b in 88% yield and in excellent diastereoselectivity (entry 2),
and N-tosyl tethered substrate 1.118c gave [3+2] adduct 1.119c in a diminished 56%
yield and >20:1 dr (entry 3).
33
Cs2CO3 (150 mol%)PPh3 (20 mol%)
Toluene
1.118 1.119
X CO2R
CO2R
Br
X
CO2R
CO2RH
H90 °C
Entry Substrate R X Product % Yield d.r.
1 1.118a Me CH2 1.119a 73 >20:1 2 1.118b Et O 1.119b 88 >19:1 3 1.118c Me NTs 1.119c 56 >20:1
Table 1.15: Intramolecular [3+2] cycloaddition of aliphatic allylic bromides
1.3 Phosphine-Catalyzed [3+2] Cycloaddition of Imines with Allenoates
and Alkynoates
1.3.1 [3+2] Cycloaddition of N-Tosyl Imines and Allenoates
In 1998 Lu reported the phosphine catalyzed [3+2] cycloaddition of allenoates
and N-tosylimines to form pyrrollidines.38 Reaction of methyl-2,3-butadienoate 1.120
with benzaldimine 1.121 in the presence of 10 mol% triphenylphosphine yielded
heterocycle 1.122 in 98% yield (Table 1.16, entry 1). The reaction also proceeded in
high yield with an electron rich aryl imine 1.121a (entry 2), and an electron poor imine
1.121b (entry 3). Unfortunately, when aliphatic 2-methyl-4-pentenyl imine 1.121c was
used in the cycloaddition only trace amounts of the cycloaddition product could be
isolated from the reaction (entry 4). Furthermore, when 2-furyl imine 1.121d was used as
substrate regioisomeric adducts 1.122d and 1.123d were isolated in 83% yield in an
85:15 regioisomeric ratio (entry 6). Attempts to catalyze the reaction with nitrogen based
catalysts were unsuccessful. In addition to this report by Lu, Shi has reported one
example of a related phosphine-catalyzed [3+2] cycloaddition.39
34
MeO2C
PPh3 (10 mol%)Benzene
25 °C
NTs
NTsMeO2C
1.120 1.121 1.123
R
RNTs
MeO2C
1.122
R
Entry Imine R Product % Yield 1 1.121 Ph 1.122 98 2 1.121a 4-MeO-C6H4 1.122a 98 3 1.121b 4-NO2-C6H4 1.122b 88 5 1.121c 2-methyl-4-pentenyl 1.122c trace 6 1.121d 2-furyl 1.122d:1.123d 85:15
Table 1.16: Initial studies of [3+2] cycloadditions with N-tosyl imines 1.3.2 Mechanism of [3+2] Cycloaddition between N-Tosyl Imines and Allenoates
Lu proposed a general mechanism for this reaction that is analogous to the [3+2]
cycloaddition of allenoates with acrylates (Scheme 1.29).3 First, addition of
triphenylphosphine to methyl-2,3-butadienoate 1.120 forms 1,3-dipole intermediate
1.124. 1,3-Dipole 1.124 then adds to imine 1.121 to give N-tosyl anion 1.125. Next,
intramolecular attack of the nitrogen anion of 1.125 onto the vinyl residue provides five
membered heterocycle 1.126. 1,2-proton transfer of heterocycle 1.126 to intermediate
1.127 is followed by elimination of phosphine to give the observed product 1.122.
35
Ph3P
Ph3P
MeO2C
NTs
MeO2C1.125
Ph3P NTs
MeO2C1.126
Ph3P
NTs
MeO2C
1.122
NTs
1.120 1.121
Ph3P
Ph3P
MeO2C
1.124
Ph
NTs
1.121Ph
NTs
MeO2C
Ph3P
PhH
Ph
Ph Ph
H
1.127 Scheme 1.29: Mechanism of [3+2] cycloadditions with N-tosyl imines
1.3.3 Regioselectivity of [3+2] Cycloaddition between N-Tosyl Imines and Allenoates
The regiochemical preference for major adduct 1.122 is analogous to the [3+2]
cycloaddition of allenoates with acrylates, and can be predicted by frontier molecular
orbital theory (Figure 1.2). The HOMO coefficient of the dipole 1.124 is highest at the
α-position (See Section 1.2.3), and the LUMO coefficient of the imine 1.121 is highest at
the carbon of the C=N imine double bond. For maximum orbital overlap the reaction
would proceed through transition state 1.128 to give major product 1.122. Minor
regioisomer 1.123, which is observed in some cases, could form through transition state
1.129. In addition to this general FMO analysis, computer modeling studies of the
regiochemical outcome of the cycloaddition have also been conducted.7
36
γ
α
Ph3P
γ
αPh3P
MeO2C
MeO2C
1.121α−1.124
γ−1.124 1.121
1.128
1.129
MeO2C
PPh3 (10 mol%)Benzene
25 °C
NTs
NTsMeO2C
1.120 1.121 1.123Minor
Ph
PhNTs
MeO2C1.122Major
Ph
NTs
Ph
NTs
Ph
NTs
MeO2C
1.122
Ph
NTsMeO2C
1.123
Ph
Figure 1.2: Regioselectivity in [3+2] cycloadditions with N-tosyl imines 1.3.4 [3+2] Cycloaddition Between N-Tosyl Imines and Alkynoates
In later studies, Lu reported that the [3+2] imine cycloaddition also proceeded
when alkynoates were used as the dipolarophile in the presence of tributylphosphine.8,10
Accordingly phenyl imine 1.121 and butynoate 1.19 participated in effective [3+2]
cycloaddition to afford products 1.122 along with side product 1.130 in 98% yield in a
87:13 ratio (Table 1.17, entry 1). A mechanism for the formation of sideproduct 1.130 is
described in detail by Lu. Electron deficient and electron rich aryl imines also
participated in the cycloaddition in good yields with similar regioselectivities (entry 2-4).
Notably, an alkyl imine provided the corresponding adduct 1.122 in 57% yield without
formation of side product 1.130 (entry 5). This result was surprising because reaction of
methyl-2,3-butadienoate 1.120 with alkyl imines provided only trace amount of the
cycloadduct.
37
PBu3 (10 mol%)Benzene
25 °C
NTs NTs
RO2C
R = Et (1.19) 1.121 1.130
R'
NTs
RO2C
1.122
R' R'
TsHN R'Me
CO2R
Entry R (1.19) R� (1.121) (1.122:1.130) % Yield 1 Et C6H5 87:13 98 2 Me 2-MeO-C6H4 81:19 95 3 Et 4-MeO-C6H4 90:10 86 4 Me 4-Cl-C6H4 85:15 87 5 Me t-Bu 100:00 57
Table 1.17: [3+2] Cycloaddition of Tosyl Imines with butynoates Lu also showed that this methodology could be used in the synthesis of the marine
antibiotic pentabromopseudilin (Scheme 1.30).8 Phosphine catalyzed cycloaddition of
methyl-2,3-butadienoate 1.120 with N-tosyl-imine 1.131 provided adduct 1.132 in 96%
yield. Product 1.132 could be elaborated into pentabromopseudilin 1.133 in 5 additional
steps.
MeO2C
PPh3Benzene
25 °C
96%
NTs
1.120 1.131 1.1331.132
OMe NTs OMe
MeO2C
NTs OH
Br
Br
Br
Br
Br
Scheme 1.30: Synthesis of pentabromopseudilin via [3+2] cycloaddition 1.3.5 [3+2] Cycloaddition Between Imines and Substituted Allenoates
Kwon has reported the diastereoselective phosphine catalyzed [3+2] cycloaddition
of imines and substituted allenoates.40 A variety of substituted allenoates 1.134 were
reacted with phenyl N-Tosyl imine 1.121 using tributylphosphine as catalyst to give
excellent yields of diastereomeric cycloadducts 1.135 and 1.136 (Table 1.18). In all
38
cases the cis-adduct was formed as the major product. Increasing the bulkiness of the
allenoate substituent resulted in a direct increase in the cis-selectivity of the reaction
(entries 1-6).
MeO2C
PPh3 (20 mol%)Benzene
25 °C
NTs NTs
EtO2C
1.134 1.121 1.136
Ph
NTs
EtO2C1.135
Ph
R R
Ph
R
Entry R (1.134) (1.135:1.136) % Yield 1 Me 91:9 89 2 Et 95:5 99 3 n-Pr 96:4 98 4 i-Pr 100:0 99 5 t-Bu 100:0 99 6 Ph 100:0 99
Table 1.18: [3+2] cycloaddition of substituted allenoates with N-tosyl imines
The affects of the N-Tosyl imine 1.121 aryl substituents were also explored in the
reaction with various allenoates 1.134 (Table 1.19). In all examples the reaction
proceeded in quantitative or nearly quantitative yield to give cis-cycloadduct 1.137 as the
sole product (entries 1-9).
39
MeO2C
PPh3 (20 mol%)Benzene
25 °C
NTs
1.134 1.121
R'
NTs
EtO2C
1.137
R'
R R
Entry R (1.134) R� (1.121) % Yield (1.137) 1 R = iPr 2-F-C6H4 97 2 R = iPr 4- i-Pr-C6H4 96 3 R = iPr 4-CF3-C6H4 96 4 R = C6H5 2-Cl-C6H4 99 5 R = C6H5 3-Cl-C6H4 99 6 R = C6H5 4-MeO-C6H4 99 7 R = tBu 4-CN-C6H4 >99 8 R = tBu 4-MeO-C6H4 >99 9 R = tBu 1-Napthyl >99
Table 1.19: [3+2] cycloaddition of substituted allenoates with various imines
In addition to Kwon�s work, Shi has reported the successful
dimethylphenylphosphine-catalyzed cycloaddition of methyl substituted butadienoate
1.134 with several N-tosyl imines 1.121 (Table 1.20).41 All of the reactions gave high
selectivity for the cis cycloadduct 1.138. However electron neutral (entry 1) and electron
poor aryl imines (entries 4-5) gave higher yields than the corresponding electron rich
imines (entries 2-3).
MeO2C
PPh2Me (20 mol%)DCM
25 °C
NTs NTs
EtO2C
1.134 1.121 1.139
R
NTs
EtO2C1.138
R
Me Me
R
Me
Entry R (1.121) (1.138:1.139) % Yield 1 C6H5 13:1 95 2 4-MeO-C6H4 15:1 31 3 4-Me2N-C6H4 >30:1 28 4 4-Cl-C6H4 16:1 72 5 4-NO2-C6H4 16:1 84
Table 1.20: PPh2Me-Catalyzed [3+2] cycloaddition of substituted allenoates with various imines
40
1.3.6 [3+2] Cycloaddition of Imines and Alkynyl Ketones
The phosphine-catalyze [3+2] cycloaddition of aryl N-tosyl imines 1.121 with
alkynyl ketones 1.140 was reported by Xue in 2008.42 Cycloaddition of phenyl ketone
1.140 with phenyl tosyl imine 1.121 in the presence of tributylphosphine gave cis-
cycloadduct 1.141 in 90% yield (Table 1.21, entry 1). The reaction proceeded in similar
yield with electron deficient imines (entries 2-3) but the yield was slightly diminished
when the electron rich p-MeO-phenyl aryl imine was used (entry 4). Similarly, the
reaction yield was lower when an ortho-halo-substituted aryl imine was used (entry 5).
PBu3 (20 mol%)PhMe
25 °C
NTs
1.140 1.121
R
O
Me
O
NTs
R
Me
1.141
Entry R (1.121) % Yield (1.141) 1 C6H5 90 2 4-Br-C6H4 90 3 4-NO2-C6H4 92 4 4-MeO-C6H4 64 5 2-Br-C6H4 79
Table 1.21: [3+2] cycloaddition of alkynyl ketones with N-tosyl imines
Xue also tested the scope of the alkyne 1.141 in the reaction and found that
electron rich aryl groups promoted the reaction (Table 1.22, entry 1-2), while an electron
deficient group resulted in a decrease in yield (entry 3). Surprisingly, when a p-nitro-
phenyl ketone was used, only a trace amount of the product 1.142 was isolated (entry 4).
An aliphatic ketone was also tested but gave a complex mixture of products (entry 5).
41
PBu3 (20 mol%)PhMe
25 °C
NTs
1.141 1.121
PhR
O
Me
R
O
NTs
Ph
Me
1.142
Entry R (1.141) % Yield (1.142) 1 4-MeO-C6H4 99 2 1-furanyl 84 3 4-Br-C6H4 80 4 4-NO2-C6H4 - 5 n-C3H7 -
Table 1.22: [3+2] cycloaddition of various alkynyl ketones with N-tosyl imines 1.3.7 [3+2] Cycloaddition of Allenes and N-(thio)-phosphoryl Protected Imines
The cycloaddition of allenes with N-(thio)-phosphoryl protected imines was
reported by He in 2008 (Scheme 1.31).43 Although Lu and Kwon had reported a few
examples without tosyl protected imines, most of the early work on the [3+2]
cycloaddition reaction of imines involved protection of the imine nitrogen with a tosyl
group,.8,40 The purpose of this study was to develop cycloaddition reactions on imines
with an easily cleaved protecting group, since removal of the tosyl group can be difficult.
To this end the authors found that the N-thiophosphoryl imines 1.143 served as suitable
substrates in the [3+2] cycloaddition with butadienoates 1.1 and 1.134. Eleven examples
were reported ranging from 41-99% yield. When methyl substituted allene 1.134 was
used as the latent 1,3-dipole the cis-product 1.145 was favored.
42
EtO2C
PR3 (20 mol%)DCM
25 °CN N
EtO2C
1.1 R = H1.134 R =Me
1.143 1.145
Ar
N
EtO2C
1.144
Ar
R R
Ar
R
PS
(EtO)2P PS
(OEt)2
S
(OEt)2
11 examples41-99%
Scheme 1.31: [3+2] Cycloaddition with N-(thio)-phosphoryl Protected Imine 1.3.8 Enantioselective [3+2] Cycloaddition of Imines
The first example of an enantioselective phosphine-catalyzed [3+2] cycloaddition
between allenes and imines was reported by Marinetti in 2006.44 Unfortunately, the
enantioselectivities obtained in this paper were low, with the highest reported ee being
61%. Shortly after this report, Gladysz described another enantioselective variant of the
reaction using a chiral rhenium phosphine complex. However, the enantioselectivities in
this paper were also low ranging from 51-60% ee.45
High enantioselectivities could not be obtained in the reaction until Jacobsen
reported an enantioselective allene imine [3+2] cycloaddition catalyzed by
phosphinothiourea catalyst 1.146 (Table 1.23).46 Through substrate screening it was
found that the aryl diphenylphosphinoyl (DPP) protected imines 1.147 provided the
highest enantioselectivities. A variety of aryl DPP-imines participated in the
cycloaddition with ethyl-2,3-butadienoate 1.1 to provide the cycloaddition products 1.148
in good yields (68-90%) and excellent enantioselectivity (94-98% ee).
43
EtO2C
1.146 (cat.)H2O (20 mol%)Et3N (5 mol%)Toluene
-30 °CN
1.1 1.147
Ar
N
EtO2C
1.148
Ar
PO
(Ph)2PO
(Ph)2
PPh2
NH
NH
Bn2N
O
SMe
1.146
Entry Ar (1.147) 1.146 (mol%) % Yield (1.148) ee (%) 1 C6H4 10 84 99 2 4-F-C6H4 10 72 95 3 4-MeO-C6H4 20 80 97 4 3-NO2-C6H4 10 70 95 5 2-Br-C6H4 10 90 95 6 3-pyridyl 10 85 95 7 2-furyl 20 79 94
Table 1.23: Enantioselective [3+2] cycloaddition of DPP imines 1.4 [3+2] Cycloaddition for Pyrrole Synthesis
In 2005, Yamamoto reported a phosphine-catalyzed [3+2] cycloaddition between
electron deficient alkynes and isocyanides to form pyrroles.47 Reaction of ethyl
isocyanate 1.150 and ethyl-2-butynoate 1.19 with catalytic amounts of 1,3-
bis(diphenylphosphino)propane (dppp) provided pyrrole 1.151 in 60% yield (Scheme
1.32).
dppp ( 15 mol%)Dioxane
100 °C
1.19 1.150 1.151
Me CO2Et NC CO2EtNH
CO2Et
CO2EtMe
Scheme 1.32: Pyrrole Synthesis via [3+2] cycloaddition Mechanistically, this reaction is thought to proceed by initial addition of
phosphine to ethyl-2-butynoate 1.19 to produce vinyl anion 1.152 (Scheme 1.33). The
44
vinyl anion 1.152 then deprotonates ethyl isocyanate 1.150 to give vinyl phosphonium
intermediate 1.153 and anion 1.154. Next a [3+2] cycloaddition between intermediates
1.153 and 1.154 occurs to form adduct 1.155. Proton migration of adduct 1.155 is
followed by elimination of the phosphine to give compound 1.156. Finally, 1,5 hydrogen
shift of compound 1.156 produces the pyrrole product 1.151.
1.156
1.150
Me CO2Et
NC CO2Et
R3P
CO2Et
R3P
Me
1.152
1.19
CO2Et
R3P
Me1.153
H
NC CO2Et
CN
R3PMe CO2Et
H
CO2Et
CN CO2Et
CO2EtMe
H
1.151
NH
CO2Et
CO2EtMe
1.154
1.155
Scheme 1.33: Mechanism of pyrrole formation The scope of the reaction was explored and several aliphatic substituted alkynes
1.157 (Table 1.24, entries 1-3) provided the corresponding pyrroles 1.158 in moderate
yield. Notably, even an unprotected alcohol provided the corresponding product in 59%
yield (entry 2). However, a tert-butyl substituted alkyne did not produce any of the
desired product (entry 4). Aryl substituted alkynes also participated in the reaction in
good yield (entries 5-7).
45
dppp ( 15mol%)Dioxane
100 °C
1.157 1.150 1.158
R CO2Et NC CO2EtNH
CO2Et
CO2EtR
Entry R (1.157) % Yield (1.158) 1 CH3CH2 72 2 HO(CH2)4 59 3 cyclo-C6H11 66 4 t-Bu - 5 C6H5 79 6 4-MeO-C6H5 79 7 4-CF3-C6H5 48
Table 1.24: Scope of pyrrole formation
1.5 Phosphine-Catalyzed [4+2] Cycloadditions of Allenes
1.5.1 Initial Phosphine-Catalyzed [4+2] Cycloaddition
A phosphine-catalyzed [4+2] annulation for the synthesis of tetrahydropyridines
was reported by Kwon in 2003.48 As a representative example, reaction of ethyl-2-
methyl-2,3-butadienoate 1.159 with imine 1.121 led to the formation of [4+2]
cycloaddition product 1.160 in nearly quantitative yield (Scheme 1.34).
PBu3 (20 mol%)CH2Cl225 °C
98%
Me
CO2Et
1.121 1.159
NPhTs
CO2Et
NTsPh
1.160
Scheme 1.34: Representative phosphine-catalyzed [4+2] cycloaddition
1.5.2 Mechanism of Phosphine-Catalyzed [4+2] Cycloaddition
The mechanism of this transformation begins with addition of phosphine to the
allene 1.159 to produce intermediate 1,3-dipole α-1.161 which is in resonance with
structure γ-1.161 (Scheme 1.35). Addition of γ-1.161 to the imine 1.121 produces
46
intermediate 1.162 which goes through a series of proton transfers to 1.164. Finally,
elimination of the PBu3 provides product 1.160.
NPhTs
CO2Et
1.160
Me
CO2Et
αMe
CO2Etγ
PBu3
αMe
CO2Etγ
PBu3
NTsPh
NPhTs
CO2Et
H
PBu3
NHPhTs
CO2EtPBu3
NHPhTs
CO2EtPBu3
NPhTs
CO2EtPBu3
PBu3
1.159 α-1.161 γ-1.161
1.121
1.162 1.163a 1.163b
1.164
PBu3
Scheme 1.35: Mechanism of [4+2] cycloaddition 1.5.3 Scope of Phosphine-Catalyzed [4+2] Cycloaddition
Kwon investigated the scope of this reaction and found that a variety of aryl and
heteroaryl imines 1.121 provided the cycloadducts 1.160 in excellent yield (Table 1.25,
entries 1-4). However, substrates with acidic protons failed completely (entries 5 and 6).
Finally, the aliphatic tert-butyl imine provided the corresponding adduct in 86% yield
(entry 7), but the n-propyl imine did not participate in the cycloaddition (entry 8).
47
PBu3 (20 mol%)CH2Cl225 °C
Me
CO2Et
1.121 1.159
NPhTs
CO2Et
NTsR
1.160
Entry R (1.121) % Yield (1.160) 1 Ph 98 2 4-MeO-C6H4 99 3 4-NO2-C6H4 86 4 2-furyl 97 5 2-pyrrol 0 6 2-HO-C6H4 0 7 t-Bu (1.xx) 86a
8 n-Pr 0 a 3 equiv of Na2CO3 Added
Table 1.25: Scope of imine in [4+2] cycloaddition The scope of the substitution of the allene was also investigated (Table 1.26).
Several benzyl substituted allenes 1.165 participated in the [4+2] cycloaddition with
phenyl substituted imines 1.121 to provide the cis-disubstituted tetrahydropyridines 1.166
in excellent yield and dr.
PBu3 (20 mol%)CH2Cl225 °C
CO2Et
1.121 1.165
NPhTs
CO2Et
NTsPh
1.166
RR
Entry R (1.165) % Yield (1.166) dr 1 C6H5 99 98:2 2 2-F-C6H4 99 97:3 3 3-MeO-C6H4 99 98:2 4 2-Me-C6H4 82 88:12
Table 1.26: Benzyl substituted allenes in [4+2] cycloaddition After her initial study Kwon used this phosphine-catalyzed [4+2] to access known
intermediate 1.167 in the formal synthesis of the indole alkaloids (±)-alstonerine and (±)-
macroline (Scheme 1.36).49
48
[4+2]
CO2Et1.168
1.169
CO2EtNMe
NNs N
MeNNs
CO2Et
CO2Et
NMe
NMe
H
H
OHPBu3
1.1671.170Known intermediate
Scheme 1.36: Formal synthesis of (±)-alstonerine and (±)-macroline 1.5.4 Enantioselective Phosphine-Catalyzed [4+2] Cycloaddition
In 2007, Fu reported an asymmetric variant of Kwon�s [4+2] cycloaddition.50
Using chiral phosphine catalyst 1.85, allene 1.171 could be reacted with a variety of aryl
and heteroaryl imines 1.121 to afford [4+2] cycloaddition products 1.172 in excellent
yield and enantioselectivity with high cis diastereoselectivity (Table 1.27).
25 °CCO2Et
1.121 1.171
NRTs
CO2Et
NTsR
1.172
CO2EtCO2Et 1.85 (5 mol%)
P tBu
CH2Cl2
Entry R (1.121) % Yield (1.172) % ee dr 1 C6H5 93 98 91:9 2 3-Me-C6H4 98 98 93:7 3 3,4,5-(MeO)3-C6H2 86 96 96:4 4 4-MeO-C6H4 42 98 93:7 5 2-Cl-C6H4 75 60 79:21 5 2-NO2-C6H4 98 68 96:4 6 2-furyl 98 97 87:13 7 3-pyridyl 0 97 91:9
Table 1.27: Enantioselective phosphine-catalyzed [4+2] cycloaddition
49
1.5.5 Phosphine-Catalyzed [4+2] Cycloaddition with 1,1-dicyanoalkenes
Finally in 2007, Kwon described another variant of the [4+2] cycloaddition for
cyclohexene synthesis (Table 1.28).51 Kwon observed that in the [4+2] cycloaddition of
allene 1.159 with alkene dinitrile 1.48 one of two products, either cyclohexene 1.173 or
1.174, could be isolated from the reaction. Notably, when the electron rich phosphine
hexamethyl phosphorus triamide (HMPT) was used as catalyst, isomer 1.173 could be
formed exclusively in good yield (entries 1-3). However, when the electron poor
phosphine tris(p-fluorophenyl)phosphine was used as catalyst, the opposite regioisomer
1.174 could be accessed in excellent yield (entries 4-6).
PR3 (20 mol%)Benzene
∆x
Me
CO2Et
1.48 1.159
R
CO2Et
R
1.173
CN
CN
NC CN
CO2Et1.174
NCNC
R
Entry R (1.48) PR3 Product % Yield 1 Ph P(NMe)2 1.173 98 2 4-MeO-C6H4 P(NMe)2 1.173 94 3 4-Br-C6H4 P(NMe)2 1.173 86 4 Ph P(4-FC6H4)3 1.174 93 5 4-MeO-C6H4 P(4-FC6H4)3 1.174 90 6 4-Br-C6H4 P(4-FC6H4)3 1.174 85
Table 1.28: Phosphine-catalyzed [4+2] cycloaddition for cyclohexene synthesis Kwon postulates that the bifurcation of the reaction products is based on the
electronic properties of the phosphines and can be explained through analysis of the 1,3-
proton transfer of initial phosphine addition adduct 1.175 to intermediate 1.176 (Scheme
1.37). When electron rich phosphines are used this proton transfer is slow and dipole
1.175 reacts with dipolarophile 1.148 to give regioisomer 1.173 as the sole product.
When an electron withdrawing phosphine is used the proton transfer is fast and
50
intermediate 1.176 reacts with the dipolarophile 1.148 to produce the other regioisomer
1.174.
CO2EtPR3
CO2EtPR3
1.175 1.176
1.148Ph
CO2Et
Ph
1.173
CN
CNNC CN
CO2Et1.174
PhCN
CN1.148
NCNC
Ph
Me
CO2Et
PR3
1,3 ProtonTransfer
H
1.159
Scheme 1.37: Mechanistic explanation of [4+2] phosphine effects Kwon also studied the effects of substitution on the allene in this reaction. Thus
the [4+2] cycloaddition of 1.148 with substituted allenes 1.177 was attempted using
HMPT as catalyst, and the corresponding substituted cyclohexenes 1.178 could be
isolated in high yields with good selectivity for the cis isomer in most cases (Table 1.29).
PR3 (20 mol%)Benzene
CO2Et
1.148 1.177
Ph
CO2Et
Ph
1.178
CN
CN
NC CNRR
Entry R (1.177) PR3 % Yield (1.178) cis:trans 1 Ph P(NMe)2 93 82:18 2 4-MeO-C6H4 P(Nme)2 92 78:28 3 4-Br-C6H4 P(Nme)2 96 84:16 4 CO2Et P(NMe)2 96 66:33 5 Et P(NMe)2 98 92:8 6 iPr P(NMe)2 7 34:66
Table 1.29: Substitution affects in [4+2] cycloaddition for cyclohexene synthesis
51
1.5.6 Phosphine-Catalyzed [4+2] Cycloaddition of Allenyl Ketones
A mechanistically unrelated phosphine-catalyzed [4+2] cycloaddition was
observed by researchers at Merck in the dimerization of allenyl ketones.32 Treatment of
allenyl ketones 1.179 with triphenylphosphine produced the corresponding pyrans 1.180
in modest yield in approximately an 85:15 E/Z ratio (Scheme 1.38).
PPh3 (20 mol%)Tolune
25 °C
5 Examples26-54% Yield
1.179
O
R O R
Me
O
R
1.180R = Me, Ar
Scheme 1.38: Cycloaddition of allenyl ketones The mechanism of this cycloaddition begins with addition of phosphine to the
allene 1.179 to produce 1,3-dipole 1.181 (Scheme 1.39). Addition of the dipole 1.181b to
another molecule of allenyl ketone 1.179 produces intermediate 1.182. The alkoxide of
1.182 then adds intramolecularly to the electrophilic alkene to form enolate pyran 1.183.
Elimination of triphenylphosphine from 1.183 is followed by isomerization to provide the
observed product 1.180.
52
PPh3
PPh3
1.181a1.179
R
O PPh31.181b
R
O
PPh3
1.182
R
O
OO
R R
PPh3
R
OR
O
O
R O
R
O
R
Me
1.179
1.183 1.184 1.180R O R O
Scheme 1.39: Mechanism of [4+2] cycloaddition/dimerization of ketones
1.6 Phosphine-Catalyzed [4+3] Cycloaddition
A phosphine-catalyzed [4+3] cycloaddition between ethyl-2-butadienoate (1.1)
and 3-(N-aryliminomethyl)chromenes 1.185 has been reported by Ishar (Scheme 1.40).22
Initial [4+3] cycloaddition between 1.1 and 1.185 affords cycloadduct intermediate 1.186.
Intermediate 1.186 undergoes a subsequent rearrangement under the reaction conditions
to afford product 1.187. Six examples are reported ranging from 55-64% yield.
PPh3 (cat.)Benzene
O
NR
OAr
∆xO
RO
CO2Et
N
CO2Et
Ar
O
RO
N
CO2Et
Ar6 examples55-64% yield
1.1
1.185 1.186 1.187 Scheme 1.40: Phosphine-catalyzed [4+3] cycloadditions
1.7 Phosphine-Catalyzed [6+3] Cycloaddition
Lu has reported a phosphine catalyzed [3+6] cycloaddition between tropone 1.188
and various allylic acetate esters 1.189 to give adducts of type 1.190 (Scheme 1.41).52
Six examples were reported with various alkyl and aryl esters 1.189 to provide the
53
corresponding [6+3] adducts 1.190 in 85-95% yields. The reaction also proceeded with
allylic bromides, chlorides and carbonates. The mechanism of this reaction is analogous
to the related [3+2] cycloaddition (see section 1.2.13).
CO2ROAc
O PPh3 (5 mol%)K2CO3Toluene
∆x
6 examples85-95% yield
O
COR
1.189 1.188 1.190
Scheme 1.41: Phosphine-catalyzed [6+3] cycloaddition
1.8 Phosphine-Catalyzed [8+2] Cycloaddition
Ishar reported a phosphine-catalyzed [8+2] cycloaddition between troponone
1.188 and various electron deficient allenes 1.191 to give adducts 1.192 (Scheme 1.42).53
The reaction proceeds in high yields with several different allenyl ketones 1.191. The
reaction also proceeds with ethyl-2-butadienoate (1.1) in good yield, but a small amount
of a [6+4] cycloadduct is formed in addition to the [8+2] adduct.
OPPh3 (cat.)Benzene
∆x
5 examples82-95% yield1.191 R= Alkyl, Aryl
1.1 R = OEt1.188 1.192
O
HCOR
COR
Scheme 1.42: Phosphine-Catalyze [8+2] cycloaddition Mechanistically, this reaction is proposed to proceed by addition of 1,3-dipole
1.193 to the tropone 1.188 to give alkoxide 1.194 (Scheme 1.43). Alkoxide 1.194 then
adds into the electron deficient alkene to give enolate 1.195. To finish, enolate 1.195
54
eliminates triphenylphosphine to provide the observed product 1.192 and regenerate the
catalyst.
O
1.191 1.188 1.194
OCOR COR
PPh3
1.193
PPh3
1.195
O
HCOR
PPh3
COR
PPh3
1.192
O
HCOR
PPh3
Scheme 1.43: Mechanism of phosphine-catalyzed [8+2] cycloaddition
1.9 Miscellaneous Cycloadditions
1.9.1 Phosphine-catalyzed Synthesis of 1,3-Dioxin-4-ylidenes
Kwon has reported a phosphine-catalyzed synthesis of 1,3-dioxan-4-ylidenes
from aldehydes and allenes (
Table 1.30).54 Isopropyl allene 1.196 undergoes annulation with two equivalents of
aldehyde 1.197 to 1,3-dioxan-4-ylidenes 1.198. The reaction gives exclusive cis-
selectivity for the substituents of 1.198 and is selective for the E-alkene. Electron
deficient aryl and heteroaryl aldehydes give the corresponding products in high yields
(entries 1-3), but electron rich aldehydes give diminished yields (entries 4-5).
Unfortunately, butyraldehyde fails to produce any of corresponding product (entry 6).
55
PMe3 (20 mol%)CHCl3CO2iPr
1.197 1.196
O O
1.198
R
R
CO2iPrR
O
25 °C
Entry R (1.197) % Yield (1.198) E:Z 1 4-pyridyl 99 8:1 2 4-CF3-C6H4 99 7:1 3 4-NO2-C6H4 84 8:1 4 Ph 54 100:0 5 3-MeO-C6H4 47 100:0 6 n-Pr - -
Table 1.30: Phosphine-catalyzed synthesis of 1,3-dioxan-4-ylidenes The mechanism of this reaction involves initial formation of 1,3-dipole 1.199a-b
from the allene 1.196 (Scheme 1.44). This is followed by addition of the 1,3-dipole
1.199b to aldehyde 1.197 to form alkoxide intermediate 1.200. Next, alkoxide 1.200
reacts with a second equivalent of aldehyde 1.197 to give intermediate 1.201.
Intramolecular addition of the alkoxide of 1.201 to the electron deficient alkene produces
enolate 1.202 which eliminates the phosphine to produce product 1.198.
CO2iPrCO2iPr
PMe3
CO2iPrPMe3
OR
OR
CO2iPrPMe3
OR
CO2iPrPMe3
O
O
PMe3
1.196 1.99a 1.99b
1.197
1.200 1.201 1.202
OR
1.197 O
R
R
Me3P
R
CO2iPr
OO
1.198
R
R
CO2iPr
Scheme 1.44: Mechanism of 1,3-dioxan-4-ylidenes formation
56
1.9.2 Phosphine-catalyzed [4+2] Cycloaddition of 3-Formylchromones
A phosphine-catalyzed [4+2] cycloaddition of 3-formylchromones 1.203 and
acetylene carboxylates 1.204 to produce cycloadducts 1.205 was reported by Waldmann
and Kumar in 2008 (Scheme 1.45).55 Eleven examples were reported ranging from 60-
99% yield.
PR3 (30 mol%)Toluene
25 °C
11 Examples60-99% yield
O
O H
O
O
O
R
1.205
R
O
CO2R
CO2R1.204
1.203
R
R
Scheme 1.45: Phosphine-catalyzed [4+2] cycloadditions with acetylene carboxylates A general mechanism for this transformation was proposed by Waldmann and
Kumar that begins with addition of phosphine to the acetylene carboxylate 1.204 to form
ylide 1.206 (Scheme 1.46). The ylide 1.206 then adds to 1.203 to give enolate
intermediate 1.207. The enolate 1.207 then eliminates triphenylphosphine and produces
the product 1.205.
57
O
O
O
RO2C PR3
RO2C
PR3
RO2C
PR3 O
O H
O
CO2RPR3 O
O
1.205
O
CO2R
1.207
1.203
1.206
PR3
1.204 1.206
Scheme 1.46: Mechanism of phosphine-catalyzed [4+2] cycloadditions with acetylene
carboxylates
1.9.3 Phosphine-Catalyzed Cycloaddition of Trienoates
Recently, the phosphine-catalyzed [3+2] cycloaddition of trienoate 1.208 with
1,1-dicyanoalkenes 1.209 to give cycloadducts 1.210 was reported by Shi (Scheme
1.47).56 Fifteen examples of the trienoate [3+2] cycloaddition with malonitriles
proceeded in yields ranging from 29-99%. Additionally twelve examples of the
cycloaddition of trienoate 1.208 with N-tosyl imines 1.211 were reported to give pyrroles
1.212 in yields ranging from 48-77%.
58
EtO2C
PBu3 (50mol%)THF
60 °C
15 Examples29-99%1.208 1.209 1.210
CNNC
R EtO2C R
CN
CN
ArAr Ar Ar
EtO2C
PBu3 (50 mol%)THF
80 °C
12 Examples48-77%
NTs
1.208 1.211 1.212
R
NTs
EtO2C R
ArAr Ar Ar
Scheme 1.47: Phosphine-catalyzed [3+2] cycloaddition with trienoates
1.10 Conclusion
Phosphine-catalyzed cycloaddition reactions have progressed a great deal in the
past 14 years since Lu�s initial report of a phosphine-catalyzed [3+2] cycloaddition
reaction. Several different types of interesting cycloadditions have been developed.
Additionally, through the use of chiral phosphines a few of the cycloadditions have been
rendered asymmetric. However, application of the phosphine-catalyzed cycloadditions
seems to be limited to a handful of organic chemists. Possibly through further refinement
of the reactions, phosphine-catalyzed cycloadditions will become more broadly used in
the synthetic community.
59
1.11 References
1 For books and reviews on organocatalysis: (a) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2001, 40, 3726. (b) Jarvo, E. R.; Miller, S. C. Tetrahedron 2002, 58, 2481. (c) List, B. Tetrahedron 2002, 58, 5573. (d) Dalko, P. I.; Moisan, L. Angew. Chem. Int. Ed. 2004, 43, 5138. (e) Ballini, R.; Bosica, G.; Palmieri, A.; Petrina, M. Chem. Rev. 2005, 105, 933. (f) Berkessel, A.; Gröger, H. Asymmetric Organocatalysis, Wiley-VCH, Weinheim, Germany, 2005. (g) Lelais, G.; MacMillan, D. W. C. Aldrichimica Acta, 2006, 39, 79. (h) Enders, D.; Grondal, C.; Huttl, M. R. M. Angew. Chem. Int. Ed. 2007, 46, 1570. 2 For reviews on phosphine organocatalysis: (a) Lu, X.; Zhang, C.; Xu, Z. Acc. Chem. Res. 2001, 34, 535. (a) Roush, W. R.; Methot, J. L. Adv. Synth. Catal. 2004, 346, 1035. (b) Long-Wu, Y.; Zhou, J; Tang, Y. Chem. Soc. Rev. 2008, 37, 1140. 3 Zhang, C.; Lu, X. J. Org. Chem. 1995, 60, 2906. 4 Mercier, E.; Fonovic, B.; Henry, C.; Kwon, O.; Dudding, T. Tetrahedron Lett. 2007, 48, 3617. 5 Xia, Y.; Liang, Y.; Chen. Y.; Wang, M.; Jiao, L.; Huang, F.; Liu, S.; Li, Y.; Yu, Z.-X. J. Am. Chem. Soc. 2007, 129, 3470. 6 Liang, Y.; Liu, S.; Xia, Y.; Li, Y.; Yu, Z.-X. Chem. Eur. J. 2008, 14, 4361. 7 Dudding, T.; Kwon, O.; Mercier, E. Org. Lett. 2006, 8, 3643. 8 Xu, X.; Lu, X. J. Org. Chem. 1998, 63, 5031. 9 (a) Ganguly, S.; Roundhill, D. M. J. Chem. Soc. Chem. Commun. 1991, 639. (b) Larpent, C.; Meignan, G. Tetrahedron Lett. 1993, 34, 4331. 10 Xu, Z.; Lu, X. Tetrahedron Lett. 1999, 40, 549. 11 (a) Trost, B. M.; Li, C. J.; J. Am. Chem. Soc. 1994, 116, 10819. (b) Guo, C.; Lu, X. J. Chem. Soc., Perkin Trans. 1 1993, 1921. 12 Shu, L.-H.; Sun, W.-Q.; Zhang, D.-W.; Wu, S.-H.; Wu, H.-M.; Xu, J.-F.; Lao, X.-F. Chem. Comm. 1997, 79.
60
13 O�Donovan, B. F.; Hitchcock, P. B.; Meidine, M. F.; Kroto, H. W.; Taylor, R.; Walton, D. R. M. Chem. Comm. 1997, 81. 14 Guo, L.-W.; Gao, X.; Zhang, D.-W.; Wu, S.-H.; Wu, H.-M. Chin. J. Chem. 2002, 20, 1430. 15 Pyne, S. G.; Schafer, K.; Skelton, B. W.; White, A. W. Chem. Comm. 1997, 2267. 16 Ung, A. T.; Schafer, K.; Lindsay, K. B.; Pyne, S. G.; Amornraksa, K.; Wouters, R.; Van der Linden, I.; Biesmans, I.; Lesage, A. S. J.; Skelton, B. W.; White, A. H. J. Org. Chem. 2002, 67, 227. 17 Du, Y.; Lu, X.; Yu, Y. J. Org. Chem. 2002, 67, 8901. 18 Du, Y.; Lu, X. J. Org. Chem. 2003, 68, 6463. 19 Lu, X.; Lu, Z.; Zhang, X. Tetrahedron 2006, 62, 457. 20 Yadav, J. S.; Reddy, B.; Narsaiah, A. V.; Nagaiah, K. Eur. J. Org. Chem. 2004, 546. 21 García Ruano, J.-L.; Núñez, A.; Martín, M. R.; Fraile, A. J. Org. Chem. 2008, 73, 9366. 22 Kumar, K.; Kapoor, R.; Kapur, A.; Ishar, M. P. S. Org. Lett. 2000, 2, 2023. 23 Al-Soud, Y. A.; Al-Masoudi, N. A.; Hass, T.; Beifuß, U. Lett. Org. Chem. 2008, 5, 55. 24 Du, L.; Lu, X.; Zhang, C. Angew. Chem. Int. Ed. 2003, 42, 1035. 25 Feng, J.; Lu, X.; Kong, A.; Hun, X. Tetrahedron 2007, 63, 6035. 26 Pham, T. Q.; Pyne, S. G.; Skelton, B. W.; White, A. H. Tetrahedron Lett. 2002, 43, 5953. 27 Pham, T. Q.; Pyne, S. G.; Skelton, B. W.; White, A. H. J. Org. Chem. 2005, 70, 6369. 28 Yong, S. R.; Williams, M. C.; Pyne, S. G.; Ung, A. T.; Skelton, B. W.; Turner, P. 2005, 61, 8120.
61
29 Zhu, G.; Chen, Z. Jiang, Q.; Xiao, D.; Cao, P. Zhang, X. J. Am. Chem. Soc. 1997, 119, 3836. 30 Wilson, J. E.; Fu, G. C. Angew. Chem. Int. Ed. 2006, 45, 1426. 31 Cowen, J. C.; Miller, S. J. J. Am. Chem. Soc. 2007, 129, 10988. 32 Wallace, D. J.; Sidda, R. L.; Reamer, R. A. J. Org. Chem. 2007, 72, 1051. 33 Wang, J.-C.; Ng, S.-S.; Krische, M. J. J. Am. Chem. Soc. 2003, 125, 3682. 34 Wang, J.-C.; Krische, M.-J. Angew. Chem. Int. Ed. 2003, 42, 5855. 35 Henry, C. E.; Kwon, O. Org. Lett. 2007, 9, 3069.
36 Ye, L.-W.; Sun, X.-L.; Wang, Q.-G.; Tang, Y. Angew. Chem. Int. Ed. 2007, 46, 5951. 37 Ye, L.-W.; Han, X.; Sun, X-L.; Tang, Y. Tetrahedron 2008, 64, 1487. 38 Xu, Z.; Lu, X. Tetrahedron Lett. 1997, 38, 3461. 39 Shi, Y.-L. Shi, M. Org. Lett. 2005, 7, 3057. 40 Zhu, X.-F.; Henry, C. E.; Kwon, O. Tetrahedron 2005, 61, 6276. 41 Zhao, G.-L.; Shi, M. J. Org. Chem. 2005, 70, 9975. 42 Meng, L.-G.; Cai, P.; Guo, Q.; Xue, S. J. Org. Chem. 2008, 73, 8491. 43 Zhang, B.; He, Z.; Xu, S.; Wu, G. He, Z. Tetrahedron 2008, 64, 9471. 44 Jean, L.; Marinetti, A. Tetrahedron Lett. 2006, 47, 2141. 45 Scherer, A.; Gladysz, J. A. Tetrahedron Lett. 2006, 47, 6335. 46 Fang, Y.-Q.; Jacobsen, E. N. J. Am. Chem. Soc. 2008, 130, 5660. 47 (a) Kamijo, S.; Kanazawa, C.; Yamamoto, Y. Tetrahedron Lett. 2005, 46, 2563. (b) Kamijo, S.; Kanazawa, C.; Yamamoto, Y. J. Am. Chem. Soc. 2005, 127, 9260.
62
48 Zhu, X.-F.; Lan, J.; Kwon, O. J. Am Chem. Soc. 2003, 125, 4716. 49 Tran, Y. S.; Kwon, O. Org. Lett. 2005, 7, 4289. 50 Wurz, R. P.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 12234. 51 Tran, Y. S.; Kwon, O. J. Am. Chem. Soc. 2007, 129, 12632. 52 Du, Y.; Feng, J.; Lu, X. Org. Lett. 2005, 7, 1987. 53 Kumar, K.; Kapur, A.; Ishar, M. P. S. Org. Lett. 2000, 2, 787. 54 Zhu, X.-F.; Henry, C. E.; Wang, J.; Dudding, T.; Kwon, O. Org. Lett. 2005, 7, 1387. 55 Waldmann, H.; Khedkar, V.; Dückert, H.; Shürmann, M.; Oppel, I. M.; Kumar, K. Angew. Chem. Int. Ed. 2008, 47, 6869. 56 Guan, X.-Y.; Shi, M. J. Org. Chem. ASAP
63
Chapter 2 Review of Enantioselective Total Syntheses of Iridoid Glycosides
2.1 Introduction
Chapter 1 The iridoids are a large family (>1000 members) of monoterpenoid natural
products characterized by a dihydropyran ring system (Figure 2.1). 1 This family of
natural products can be generally divided into two classes, carbocyclic iridoids and
secoiridoids.2 In carbocylic iridoids, the dihydropyran ring system is cis-fused to a
cyclopentane ring (2.1, 2.5-2.8), whereas in secoiridoids (2.2, 2.4), this cyclopentane ring
is cleaved between C-7 and C-8.
O
OGlu
CO2Me
HO
H
H
2.5
Geniposide
O
OGlu
CO2MeH
H
OHC
2.4
Secologanin
O
OGluHO
H
H
2.6
Aucubin
O
OGluHO
H
H
2.7
Catalpol
O
Me
H
H
2.8
Nepatalactone
Me
O
1 O 2
34
OR
6
7
8
10
59
H
H
2.1
Carbocyclic Iridoid
1 O
34
OR
6
8
10
59
H
H
2.2
Secoiridoid
711 11
1 O
O OOH
OHOH
HO
H
H
2.3
Iridoid Glycoside
O
HO
Figure 2.1: Iridoid natural products
Several iridoids are attached to a sugar unit and are therefore referred to as iridoid
glycosides. Typically, the sugar unit of iridoid glycosides is attached at C-1 of the iridoid
carbon skeleton through a linkage (2.3, 2.4-2.7), however other glycoside linkages do
exist (Figure 2.1).2
64
Numerous iridoids possess useful biological activities3 and as a result, the iridoids
have been popular synthetic targets.2,4,5 However, there are relatively few total syntheses
of iridoid glycosides.5 This is most likely due to the difficulties inherent to installation of
the complex -glycoside linkages in these compounds. Furthermore, due to the inherent
chirality of the naturally derived sugar subunits, these syntheses are most efficient when
the iridoid carbon skeleton is formed in an enantioselective fashion so that diastereomeric
mixtures are not formed in the final glycosidation step. This chapter will review the
development of methods for the enantioselective total synthesis of iridoid glycosides.
2.2 General Discussion of Iridoid Glycoside Formation
2.2.1 Koenigs-Knorr Type Glycosidation of Iridoids
In 1983 Tietze reviewed the existing methods for iridoid glycosidation and
outlined two basic strategies.5 The first strategy is illustrated using generic iridoid
structures in Scheme 2.1. This approach utilizes the hemiacetal of iridoid aglucon 2.9 as
a nucleophile and an activated α-glucose derivative 2.10 as an electrophile in a SN2 type
reaction. Successful application of this strategy should produce the β-glycosidated
product 2.11. Unfortunately, this classical Koenigs-Knorr glycosidation strategy gives
poor results in the synthesis of iridoid glycosides. Typically the reaction either fails
completely, or produces the desired glycosidation product in very low yield.
65
O
OH
H
H
OOR
OROR
RO
LGO
OβO
OROROR
RO
H
H
2.9 2.10 2.11
CO2Me CO2Me
Scheme 2.1: Koenigs-Knorr strategy for glycosidation
For example, Tietze attempted the glycosidation of 6-O-acetylloganin-aglycon
2.12 with acetate protected α-bromoglucopyranose 2.13 in the presence of silver
perchlorate (Scheme 2.2).6 A disappointing 13% yield of the desired β-glycoside loganin
(2.14) was isolated after solvolysis of the acetate protecting groups. Additionally, the
epimeric α-glycoside 2.15 was produced in 5.4% yield.
O
OH
H
H
OOAc
OAcOAc
AcO
Br
O
O OOH
OHOH
HO
H
H
2.12
2.13
13%
2.14
CO2Me CO2Me
Me Me
AcO HO O
O
H
H
5.4%
2.15
CO2Me
Me
HO1. AgClO42. Deprotection
OHOOH
OHOH
Scheme 2.2: Glycosidation with α-acetobromoglucose
In another representative study, Tietze attempted the glycosidation of 6-O-
acetylloganin-aglycon 2.12 with 1,2-anhydro-α-D-glucose-triacetate (2.16) in the
presence of a lewis acid (Scheme 2.3). After hydrolysis of the initial products, only the
unnatural α-glycoside 2.15 could be isolated in 24% yield.7
66
O
OH
H
H
OAcOAcO
2.12
2.16
CO2Me
Me
AcO O
O
H
H
2.15
CO2Me
Me
HO
1. BF3.OEt22. Deprotection
OHOOH
OHOH
O
AcO
24%
Scheme 2.3: Glycosidation with 1,2-anhydro-α-D-glucose triacetate
In addition to these studies, there have been other reports of the failure of this 1st
glycosidation strategy in iridoid β-glycoside synthesis.8,9,10 The general failure of this
method to provide satisfactory results in the synthesis of iridoid β-glycosides stems from
several factors. First the hemiacetal oxygen of the iridoid aglucon 2.9 is a weak
nucleophile because of adjacent electron withdrawing groups (Scheme 2.4).5 As a result,
effective addition to the glycosidation reagent 2.10 is difficult.
O
OH
H
H
OOR
OROR
RO
LGO
OβO
OROROR
RO
H
H
2.9 2.10 2.11
CO2Me CO2Me
Scheme 2.4: Problems with 1st glycosidation strategy
Second, the yields of these reactions are attenuated by the formation of unwanted
iridoid dimers of type 2.18 (Scheme 2.5).6,7 These dimers are thought to form through
the intermediacy of oxocarbenium intermediate 2.17. The oxocarbenium intermediate
2.17 can be formed via ionization of the hemiacetal oxygen of compound 2.9 in the
67
presence of a lewis acid. Addition of a second molecule of 2.9 to the oxocarbenium ion
2.17 produces iridoid dimer 2.18.
O
OH
H
H
2.9
O
O
H
H
O
H
HO
H
H
2.18
2.17
Lewis Acid
CO2Me CO2Me
CO2Me
CO2Me
O
OH
H
H
CO2Me
2.9
Scheme 2.5: Formation of iridoid dimers In light of the deficiencies of this first strategy of iridoid glycoside synthesis,
novel glycosidation methods were developed. These methods will be discussed in the
subsequent section.
2.2.2 2nd Strategy for Iridoid Glycoside Synthesis
The 2nd general iridoid glycosidation strategy for iridoid glycoside formation
involves converting the iridoid carbon skeleton into an electrophile, and using the
hemiacetal of the glycosidation reagent as the nucleophile (Scheme 2.6). This general
strategy can be accomplished through ionization of the hemiacetal oxygen of
representative iridoid aglucon 2.9 to produce oxocarbenium intermediate 2.17. The
electrophilic intermediate 2.17 can then react with the hemiacetal oxygen of β-
glucopyranose derivative 2.19 to give the desired β-glycosidation product 2.11.
O
OH
H
H
2.9
O
H
H
2.17
Lewis AcidO
OROROR
RO
2.19
HO O
O OOR
OROR
RO
H
H
2.11
CO2Me CO2Me CO2Me
68
Scheme 2.6: 2nd general glycosdiation strategy The first example of this strategy for iridoid glycosidation was reported by Büchi
in the total synthesis of loganin (Scheme 2.7).11 Reaction of acetate protected racemic
loganin hemiacetal 2.12 with β-�-tetracetylglucose 2.20 in the presence of a boron
trifluoride diethyl etherate provides a low 1.4% yield of loganin pentaacetate 2.21. In
subsequent studies, Partridge9 was able to increase the yield of this same transformation
to 17% employing optically active 2.12.
O
OH
H
HO
OAcOAcOAc
AcO O
O OOAc
OAcOAc
AcO
H
H
2.12 2.20 2.21
CO2Me CO2Me
Me Me
AcO AcOBF3.OEt2HO
Scheme 2.7: Glycosidation of loganin Tietze has also studied this identical transformation in depth, and reports that 6-O-
acetylloganin-aglycon 2.12 upon reaction with β-�-tetracetylglucose 2.20 affords a
combined 32.3 % yield of epimeric pentaacetates 2.21 and 2.22, in addition to a 43.2%
yield of the iridoid dimer 2.23 (Scheme 2.8).6 Hydrolyis of the acetate groups of the
combined mixture of epimeric pentaacetates 2.21 and 2.22 allowed for isolation of
loganin (2.14) in 9.1 % yield and its epimeric α-isomer 2.15 in 11% yield. These results
were noteworthy, as they revealed that dimerization of iridoid aglucon 2.12 to compound
2.23 was favored over glycosidation. Furthermore, they showed that epimerization12 of
β-�-tetracetylglucose 2.20 was occurring prior to glycosidation, resulting in the
formation of the undesired iridoid α-glycoside 2.22.
69
O
OH
H
H
OOAc
OAcOAc
AcO
2.12
2.20CO2Me
Me
AcOBF3.OEt2
HO
O
O
H
H
O
H
H
CO2Me
CO2Me
HO
Me
HO
Me
2.23
O
O
H
H
CO2Me
Me
AcO
OAcOOAc
OAcOAc
2.22
O
O OOAc
OAcOAc
AcO
H
H
2.21
CO2Me
Me
AcO
Scheme 2.8: Formation of iridoid dimers Notably, Tietze was able to develop a novel method13 of glycosidation to
overcome the problems in this second glycosidation stategy. Specifically, it was found
that when the acetate protected hemiacetal 2.24 was reacted with TMS-β-
glucopyranoside 2.25 in the presence of catalytic amounts of TMSOTf, loganin
pentaacetate 2.21 could be isolated in 75% yield as a 12:1 β:α mixture of epimers
(Scheme 2.9).14
O
OAc
H
HO
OAcOAcOAc
AcO
2.24 2.25
CO2Me
Me
AcOTMSO O
O OOAc
OAcOAc
AcO
H
H
2.21
CO2Me
Me
AcO-40 °C
cat. TMSOTfSO2(liq.)
Scheme 2.9: Tietze glycosidation method It is postulated that the reaction proceeds by ionization of the acetate of
compound 2.24 to form oxocarbenium intermediate 2.26 (Scheme 2.10). The
trimethylsiloxy group of the glycosidation reagent 2.25 then reacts with oxocarbenium
ion 2.26 to give loganin pentaacetate 2.21 and regenerate the catalyst. It is presumed that
acetate protection of the hemiacetal of 2.24 serves to prohibit the dimerization process
observed in the previous glycosidation reaction. Additionally, TMS protection of the
70
glycosidation reagent 2.25 prevents epimerization thereby ensuring high levels of
stereoselectivity. These combined effects account for the excellent results obtained when
using this methodology.
O
OAc
H
H
OOAc
OAcOAc
AcO
2.24 2.25
CO2Me
Me
AcO TMSOO
H
H
2.26
CO2Me
Me
AcOTMSOTf
OTf
TMSOAc TMSOTf
2.21
Scheme 2.10: Mechanism of Glycosidation MacMillan has successfully applied this same glycosidation strategy to the total
synthesis the iridoid natural products brasoside and littoralisone.15 And currently, this is
the only method for iridoid glycoside formation at the C-1 position that provides the
desired iridoid glycosides in high yields with excellent β-selectivity.
2.3 Review of Enantioselective Iridoid Glycoside Syntheses
2.3.1 Introduction
To date, there have only been three enantioselective total syntheses of iridoid
glycosides that access the iridoid carbon skeleton in enantiopure form prior to the
glycosidation step. Tietze has reported the total syntheses of the iridoid glycosides
hydroxyloganin, 7-epihydroxyloganin, and hydroxyloganic acid. 16 However, this work
will be omitted from this review since it relies on the resolution of a late stage racemic
iridoid aglucon intermediates, and because it is analogous to the asymmetric total
synthesis of loganin by Partridge9 that will be discussed in detail.
71
2.3.2 Enantioselective Total Synthesis of (-)-Loganin
The first reported enantioselective synthesis of an iridoid β-glycoside was the
asymmetric synthesis of (-)-loganin (2.14) reported by Partridge in 1973.9 Partridge was
able to access loganin (2.14) in a concise 5 steps from 5-methyl-cyclopentadiene (2.27)
and diformyl ester 2.28 (Scheme 2.11). Unfortunately, several of these steps proceeded
in low yield. The key steps in the synthesis were an asymmetric hydroboration-
oxidation, and a [2+2] cycloaddition.
2.27
Me OH
MeO2C CHO 5 steps
2.28
O
O OOH
OHOH
HO
H
H
2.14
CO2Me
Me
HO
Scheme 2.11: Total synthesis of (-)-loganin
The synthesis started with an asymmetric hydroboration-oxidation of 5-methyl-
cyclopentadiene 2.27 using (+)-di-3-pinanylborane to produce trans-alcohol 2.29 in 33%
yield and ≥ 95% ee (Scheme 2.12). Next, the trans-alcohol 2.29 was converted to cis-
acetate 2.30 in two subsequent steps.
2.27 2.302.29
1. (+)-R2BHR=pinanyl
2. H2O2, NaOH
33% Yield95% ee
Me Me
HO 2 steps
Me
AcO
Scheme 2.12: Asymmetric hydroboration-oxidation
With acetate 2.30 in hand, the key photoannulation reaction with diformyl ester
2.28 was explored (Scheme 2.13). Treatment of racemic acetate 2.30 and diformyl ester
2.28 with UV light afforded a combined 33% yield of isomeric products 2.12, 2.31, and
72
2.32. The desired adduct 2.12 was the major product being formed in 22% yield.
Mechanistically, this reaction occurs via an initial [2+2] cycloaddition between
compounds 2.30 and 2.28 to produce cyclobutane intermediate 2.33. Cyclobutane 2.33
then undergoes a retroaldol reaction to generate dialdehyde intermediate 2.34. Finally,
the dialdehyde 2.34 cyclizes to produce the major product 2.12. Partridge also
conducted this reaction using optically enriched acetate 2.30 to afford a 20% yield of
adduct 2.12.
2.30
Me
AcO hυ
OH
MeO2C CHO
Me
AcO
H
H
CO2MeCHO
OH
retro-aldol
Me
AcO
H
HO
CO2Me
O
O
OH
H
H
CO2Me
Me
AcO O
OH
H
H
CO2Me
2.28
[2+2]
AcO
Me
O
OH
H
H
CO2Me
Me
AcO
2.33 2.34
22% Yield
2.12
2.31 2.3233%
Scheme 2.13: Photoannulation of acetate The final step of the synthesis was the aforementioned glycosidation of
intermediate 2.12 with β-�-tetracetylglucose 2.20 using boron trifluoride to give loganin
pentaacetate 2.21 in 17% yield (Scheme 2.14). Conversion of pentaacetate 2.21 to
loganin (2.14) had previously been reported.17
73
O
OH
H
HO
OAcOAcOAc
AcO O
O OOAc
OAcOAc
AcO
H
H
2.12 2.20 2.21
CO2Me CO2Me
Me Me
AcO AcOBF3.OEt2HO
17%
Scheme 2.14: Glycosidation in total synthesis of loganin 2.3.3 Enantioselective Total Synthesis of (+)-Semperoside A
The total synthesis of the iridoid glycoside (+)-semperoside A (2.35) was
described in 2004 by Vidari.18 Vidari was able to complete the total synthesis of (+)-
semperoside A (2.35) in 10 steps and 17% overall yield from the known19
enantiomerically pure lactone 2.36 (Scheme 2.15). The synthesis is unique since it
involves the formation of a β-glycoside at the C-3 position of the iridoid carbon skeleton
rather than the usual C-1. Key steps in the synthesis include the formation of the β-
glycoside, and a mercury catalyzed cyclization.
1
O3
OOH
OHOH
HOH
H
2.35
Semperoside A
Me
OO
OO
O
OH
10 steps
2.36
17%
Scheme 2.15: Synthesis of (+)-semperoside A The synthesis starts with conversion of known lactone 2.36 to enol ether 2.37 in 6
steps (Scheme 2.16). Enol ether 2.37 is then glycosidated with benzyl protected α-
bromoglucopyranose 2.38 in the presence of potassium carbonate to give the E-β-
glycoside 2.39 in 95%.
74
O
O
OH
2.36
6 Step O
O
OSEM
Me2.37
OH
OOBn
OBnOBn
BnO
Br2.38
OSEMO
OBnOBnOBn
BnO
2.39Me
O
O
O
K2CO3
95%
Scheme 2.16: Glycosidation reaction in (+)-semperoside A synthesis The glycosidation reaction was followed by deprotection of the SEM protecting
group to afford alcohol 2.40 (Scheme 2.17). After that, intramolecular addition of the
primary alcohol of 2.40 to the enol ether was promoted using mercury trifluoroacetate to
generate organomercurial intermediate 2.41. Reduction of intermediate 2.41 with sodium
borohydride gives product 2.42 in 50% yield over two steps. Final deprotection of the
benzyl groups of 2.42 yields (+)-semperoside A (2.35) in 95% yield.
OHO
OBnOBnOBn
BnO
H
2.40Me
O
O
O
Hg(OCOCF3)2O
OOBn
OBnOBn
BnOH
H
2.41Me
OO
OHg
O
OOBn
OBnOBn
BnOH
HMe
OO
OH
2.42
Pd/C H2
95%O
OOH
OHOH
HOH
HMe
OO
OH
2.35
NaBH4
50%
Scheme 2.17: Mercury-mediated cyclization in (+)-semperoside A synthesis 2.3.4 Enantioselective Total Synthesis of (-)-Brasoside and (-)-Littoralisone
In 2005 MacMillan reported the enantioselective total synthesis of the iridoid
natural products (-)-brasoside 2.43 and (-)-littoralisone 2.44 from a common intermediate
75
(Scheme 2.18).15 In this elegant work (-)-brasoside (2.43) is accessed in 13 steps from (-
)-citronellol (2.45), and (-)-littoralisone (2.44) is accessed in 13 steps and 13% overall
yield. The key step in this synthetic approach was an organocatalytic Michael reaction.
O
O OOH
OHOH
HO
H
H
2.43
(-)-Brasoside
Me
OO
O
O OOH
OHOH
O
H
H
2.44
(-)-Littoralisone
Me
OO
O
HH
OH
OH
2.45
Me
Scheme 2.18: Synthesis of (-)-Brasoside and (-)-littoralisone The synthesis starts with conversion of (-)-citronellol 2.45 to the key Michael
addition substrate enal-aldehyde 2.46 in 6 synthetic operations (Scheme 2.19). Enal-
aldehyde 2.46 was then treated with a catalytic amount of L-proline, and under optimized
conditions, a 91% yield of the desired Michael addition products 2.47 and 2.48 were
formed in 91% yield in a 10:1 ratio. Product 2.47 could also be acylated in situ to form
acetate 2.49 in an overall 83% yield from enal-aldehyde 2.46.
OH
2.45
Me
6 StepsO
2.46
Me
TBDPSOO L-Proline (cat.)
Me
TBDPSO
O
H
HOH
O
Me
TBDPSOO
91 %
10:12.47:2.48
H
H
2.47 2.48
O
2.46
Me
TBDPSOO L-Proline (cat.)
then Acylation
Me
TBDPSO
O
H
HOAc83 %
2.49
76
Scheme 2.19: Proline-catalyzed Michael addition After synthesis of the core iridoid skeleton, acetate 2.49 was converted to lactone
2.50 in 4 steps (Scheme 2.20). Compound 2.50 was then glycosidated according to the
method of Tietze,14 using β-TMS-glucopyranoside 2.25. Subsequent hydrolysis of the
acetate protecting groups yielded (-)-brasoside (2.43) in 82% yield.
Me
TBDPSO
O
H
HOAc
2.49
O
OAc
H
HMe
OO
4 steps
O
O OOH
OHOH
HO
H
HMe
OO
1. TMSOTf (cat.)2. MeOH, Et3N
82%
OOAc
OAcOAc
AcOTMSO
2.50 2.25 2.43
Scheme 2.20: Final stages of (-)-brasoside Synthesis Lactone 2.50 could also be converted to (-)-litoralisone (2.44) in two steps
(Scheme 2.21). Glycosidation of lactone 2.50 using differentially protected β-TMS-
glucopyranose 2.51 afforded glycoside 2.52 in 74% yield. Finally, treatment of
compound 2.52 with UV light affected a [2+2] cycloaddition that created the cyclobutane
functionality of (-)-litoralisone (2.44). This was followed by in situ deprotection to
produce the natural product 2.44 in 84% yield.
77
O
O OOH
OHOH
O
H
H
2.44
Me
OO
O
HH
OH
O
O OOBn
OBnOBn
O
H
HMe
OO
O
BnO
OOBn
OBnOBn
ROTMSO
R = p-benzyloxy-cinnamyl
TMSOTf (cat.)O
OAc
H
HMe
OO
74%
hυH2, Pd/C
2.50 2.52
84%
2.51
Scheme 2.21: Final stages of (-)-littoralisone Synthesis
2.4 Conclusions
Although there has been extensive work in the total synthesis of iridoid natural
products, limitations still remain. Several methods have been developed for the synthesis
of the iridoid carbon skeletons. However, methods for the glycosidation of iridoids are
still limited. Currently there is only one method that gives the desired iridoid β-
glycosides in acceptable yields. Future work in this area will hopefully address this
problem, and provide more concise enantioselective routes towards the iridoids
glycosides.
2.5 References
78
1 For structural reviews on iridoids see: (a) El-Naggar, L. J.; Beal, J. L. J. Nat. Prod. 1980, 43, 649. (b) Boros, C. A.; Stermitz, F. R. J. Nat. Prod. 1990, 53, 1055. (c) Dinda, B.; Debnath, S.; Harigaya, Y. Chem. Pharm. Bull. 2007, 55, 159. (d) Dinda, B.; Debnath, S.; Harigaya, Y. Chem. Pharm. Bull. 2007, 55, 689. 2 Franzyk, H. Prog. Chem. Org. Nat. Prod. 2000, 79, 1.
3 For reviews on iridoid biological activity see: (a) Buzogany, K.; Cucu, V. Farmacia 1983, 31, 129. (b) Tietze, L.-F. Angew. Chem. Int. Ed. Engl. 1983, 22, 829. (c) Ghisalberti, E. L. Phytomedicine 1998, 5, 147. (d) Tundis, R.; Loizzo, M. R.; Menichini, F.; Statti, G. A.; Menichini, F. Mini-Rev. Med. Chem. 2008, 8, 399. 4 For reviews on iridoid synthesis see: (a) Bianco, A. Stud. Nat. Prod. Chem. 1990, 7, 439. (b) Bianco, A. Pure Appl. Chem. 1994, 66, 2335. (c) Isoe, S. Stud. Nat. Prod. Chem. 1995, 16, 289. (d) Nangia, A.; Prasuna, G.; Rao, P. B. Tetrahedron, 1997, 53, 14507. 5 Tietze, L.-F. Angew. Chem. Int. Ed. Engl. 1983, 22, 828.
6 (a) Tietze, L.-F.; Niemeyer, U.; Marx, P. Tetrahedron Lett. 1977, 39, 3441. (b) Tietze, L.-F.; Niemeyer, U. Chem. Ber. 1978, 111, 2423. 7 Tietze, L.-F.; Marx, P. Chem. Ber. 1978, 111, 2441.
8 Merz, K. W.; Lehmann, H. Arch. Pharm. 1957, 290, 543.
9 Partridge, J. J.; Chadha, N. K.; Uskoković, M. R. J. Am. Chem. Soc. 1973, 95, 532.
10 Halpern, O.; Schmid, H. Helv. Chem. Acta. 1958, 41, 1109.
11 (a) Büchi, G.; Carlson, J. A.; Powell, J. E.; Tietze, L.-F. J. Am. Chem. Soc. 1970, 92, 2165. (b) Büchi, G.; Carlson, J. A.; Powell, J. E.; Tietze, L.-F. J. Am. Chem. Soc. 1973, 95, 540. 12 Georg, A. Helv. Chim. Acta, 1932, 15, 924. 13 (a) Tietze, L.-F.; Fischer, R. Angew. Chem. Int. Ed. Engl. 1981, 20, 969. (b) Tietze, L.-F.; Fischer, R. Angew. Chem. 1981, 93, 1002. 14 Tietze, L.-F., Fischer, R. F.; Remberg, G. Liebigs Ann. Chem. 1987, 971.
79
15 Mangion, I. K.; MacMillan, D. W. C.; J. Am. Chem. Soc. 2005, 127, 3639.
16 (a) L.-F. Tietze, Angew. Chem. 1973, 85, 763. (b) L.-F. Tietze, Angew. Chem. Int. Ed. Engl. 1973, 12, 757. (c) Tietze, L.-F. Chem. Ber. 1974, 107, 2499. 17 Battersby, A. R.; Hall, E. S.; Southgate, R. J. Chem. Soc. C, 1969, 721.
18 Piccinini, P.; Vidari, G.; Zanoni, G. J. Am. Chem. Soc. 2004, 126, 5088.
19 Zanoni, G.; Agnelli, F.; Meriggi, A.; Vidari, G. Tetrahedron Asymmetry: 2001, 12,
1779.
80
Chapter 3 Intramolecular Approach to (+)-Geniposide
3.1 Intramolecular Cycloaddition Retrosynthetic Analysis
In the course of a program dedicated to the development of new methods for
phosphine catalyzed C-C bond formations, we had previously investigated1 new
applications of Lu�s phosphine catalyzed [3+2] dipolar cycloaddition reaction of electron
deficient allenoates and alkynoates with alkenes.2 During the course of these studies, we
recognized that the general carbon skeleton of the iridoid natural products could be
accessed efficiently using this methodology. In order to showcase the utility of this
strategy for iridoid synthesis, we decided to apply it to the synthesis of the iridoid
glycoside (+)-geniposide (3.1) (Scheme 1).3 (+)-Geniposide (3.1) embodies the common
structural features of several iridoid glycosides making the synthesis relevant to a broad
range of iridoid natural products. In addition, the molecule displays antitumor4 and anti-
inflammatory5 activity and its aglycone, genipin, has recently garnered attention as an
effective treatment for type II diabetes.6
Retrosynthetic analysis of (+)-geniposide (3.1) started with disconnection of the
β-glycoside to leave behind the core 6-5 cis-fused cyclopentapyran skeleton 3.2 (Scheme
3.1). It was envisioned that the core skeleton of 3.2 could be accessed efficiently via an
intramolecular phosphine-catalyzed [3+2] cycloaddition. This cycloaddition could be
conducted on a substrate of type 3.3 in which a latent 1,3 dipole, or electron deficient
alkyne, is tethered to an appropriately oxygenated dipolarophile. We anticipated that
activation of alkyne 3.3 by a catalytic amount of a trialkylphosphine would give rise to
81
dipole 3.4, which could react with the appended alkene dipolarophile to provide
cycloadduct 3.5. Initial studies towards the realization of this synthetic route focused on
the design and synthesis of [3+2] cycloaddition substrates of type 3.3.
O
RO2C OR
[3+2]
O
O
CO2Me
HO OOH
OHOH
HO
H
H
H
H
PR3
3.1
3.5 3.4 3.3
O
RO2C OR
H
H
3.2
CO2Me
O
ORO2CRO2C
OR3P
O
Scheme 3.1: First generation retrosynthetic analysis of (+)-geniposide
3.2 Coumalate Intramolecular Cycloaddition Substrate
The first goal in our synthesis was to design a substrate for our key phosphine-
catalyzed intramolecular [3+2] cycloaddition. The first substrate proposed was
coumalate derived substrate 3.6 (Scheme 3.2). This substrate was chosen because
successful [3+2] cycloaddition of 3.6 gives rapid access to the carbon skeleton of (+)-
geniposide in formation of cycloadduct 3.7.
O
O
OO
MeO2CO
O
OO
MeO2C
[3+2]
O
O
CO2Me
HO OOH
OHOH
HO
H
H
3.13.6 3.7
Scheme 3.2: Coumalate intramolecular cycloaddition substrate
82
3.2.1 Coumalate Intramolecular Cycloaddition Synthesis
Synthesis of cycloaddition substrate 3.6 began with THP-protection of propargyl
alcohol (3.8) using 3,4-dihydro-2H-pyran and TsOH to afford alkyne 3.9 in 86% yield
(Scheme 3.3).7 Alkyne 3.9 was then deprotonated with MeLi, and the resulting acetylide
anion was trapped with methyl chloroformate to produce ynoate 3.10 in 41% yield.8,9
Deprotection of ynoate 3.10 to desired alcohol 3.11 was realized in 82% yield using
TsOH in methanol.10
OH OTHPOTHP
MeO2C
DHP, TsOHDCM
1. MeLi, THF2. ClCO2Me
-78 °C
41%
25 °C
86%
OH
MeO2C
TsOH, MeOH
25 °C
82%
3.8 3.9 3.10 3.11
Scheme 3.3: Synthesis of coumalate cycloaddition substrate alcohol
After having synthesized alcohol 3.11, commercially available coumalic acid
(3.12) was converted into acid chloride 3.13 in 57% yield using PCl5 (Scheme 3.4).11
The acid chloride was then acylated with alcohol 3.11 to give substrate 3.6 in 52% yield.
O
O
OO
MeO2CO
O
OHO
O
O
OCl
PCl5, Et2O
50 °C
57%
Et3N, DCM
0 °C
52%
HOCO2Me
3.12 3.13
3.11
3.6
Scheme 3.4: Synthesis of coumalate cycloaddition substrate 3.2.2 Attempted Coumalate Intramolecular Cycloaddition
With the desired intramolecular substrate 3.6 in hand, the key phosphine
catalyzed [3+2] cycloaddition reaction was attempted. Unfortunately, when the standard
83
conditions12 for this reaction were used, none of the desired cycloadduct 3.7 could be
isolated from the reaction (
Table 3.1, entry 1). Further attempts at changing various reaction parameters to favor
formation of the desired product 3.7 were unsuccessful (entries 2-6). One possible reason
that this substrate failed to undergo cycloaddition because the alkene of 3.6 was not
sufficiently electrophilic. This proposal is substantiated by the fact that in previous
intramolecular phosphine-catalyzed [3+2] cycloaddition reactions, enones participate
readily in the reaction, whereas less reactive enoates do not.1a This realization, coupled
with the failure of substrate 3.6 to undergo cycloaddition, compelled us to design a new,
more reactive cycloaddition substrate for our synthesis.
O
O
OO
MeO2CO
O
O
O
MeO2C
PR3Sealed Tube
3.6 3.7
Entry Phosphine Solvent Temp (°C) Conc. (M) Time (h) Yield 1 PBu3 EtOAc 110 0.1 24 - 2 PBu3 Toluene 110 0.1 48 - 3 PBu3 EtOAc 110 0.01 48 - 4 PBu3 EtOAc 50 0.1 48 - 5 PBu3 EtOAc 25 0.1 48 - 6 PPh3 EtOAc 110 0.1 48 - 7 PCy3 EtOAc 110 0.1 48 -
Table 3.1: Coumalate intramolecular cycloaddition reaction
84
3.3 Pyranone Intramolecular Cycloaddition Substrate
3.3.1 Design of 1st Generation Pyranone Intramolecular Cycloaddition Substrate
Since coumalate derived substrate 3.6 did not undergo cycloaddition due to a lack
of reactivity, a new more reactive substrate was sought. This search led to the proposal
of substrate 3.14, in which a pyranone based ring system is tethered through an acetal
linkage to an electron deficient alkyne (Scheme 3.5). Substrate 3.14 should be more
reactive than the previous substrate 3.6 towards cycloaddition since its dipolarophile is
doubly activated by both the adjacent electron withdrawing ketone at C-1, and the
electron withdrawing acetal at C-4. Furthermore, it was thought that substrate 3.14
would be superior to the coumalate derived substrate 3.6 since its cycloaddition product
3.15 maps on to the carbon skeleton of (+)-geniposide (3.1) more directly.
4 O
1
O
O
O
Me O
O
OO
[3+2]
O
O
CO2Me
HO OOH
OHOH
HO
H
H
3.13.14 3.15
Scheme 3.5: Design of 1st generation pyranone intramolecular cycloaddition substrate 3.3.2 Synthesis of 1st Generation Pyranone Intramolecular Cycloaddition Substrate
The synthesis of substrate 3.14 was accomplished in 2 steps from commercially
available furfuryl alcohol (3.16) (Scheme 3.6). The first step was oxidation of furfuryl
alcohol (3.16) to lactol 3.17 in 78% yield using m-CPMA.13 Lactol 3.17 was then
85
coupled with commercially available 2-butynoic acid (3.18) to give the cycloadduct
substrate 3.14 in 52% yield.
OOH
m-CPBA, DCM
25 °C
78%
O
O
OH
DCC, DMAP, DCM0 °C
52%
HO2C Me
O
O
O
O
Me
3.18
3.16 3.17 3.14
Scheme 3.6: Synthesis of 1st generation pyranone intramolecular cycloaddition substrate 3.3.3 1st Generation Pyranone Intramolecular Cycloaddition
With cycloaddition substrate 3.14 in hand, the phosphine-catalyzed [3+2]-
cycloaddition was attempted using the previously developed reaction conditions1a (Table
3.2, entry 1). Unfortunately, none of the desired cycloadduct 3.15 could be isolated from
the reaction. Variation of the reaction time and phosphine catalyst was also unsuccessful
(entries 2-4). The failure of this reaction was surprising since a highly activated
dipolarophile was being used. However, one potential problem with substrate 3.14 is that
the sp2-hybridized ester tether is conformationaly restricted, and therefore may not be
flexible enough to allow the latent 1,3-dipole to interact effectively with the alkene of
3.14. This potential problem led us to propose a second generation pyranone
cycloaddition substrate with a more flexible tether.
86
O
O
O
O
MeO
O
OO
Sealed Tube
110 ºC
EtOAc
3.14 3.15
Entry Phosphine Time (h) Yield 1 PBu3 24 - 2 PBu3 48 - 3 PPh3 48 - 4 PCy3 48 -
Table 3.2: 1st generation pyranone intramolecular cycloaddition 3.3.4 Design of 2nd Generation Pyranone Intramolecular Cycloaddition Substrate
The second generation pyranone intramolecular cycloaddition substrate 3.19,
incorporated a less conformationally restricted ether tether (Scheme 3.7). It was hoped
that this more flexible tether would allow the latent 1,3 to dipole interact effectively with
the alkene of 3.19. Successful cycloaddition of 3.19 would give rise to adduct 3.20,
which corresponds well to the carbon skeleton of (+)-geniposide (3.1).
[3+2]
O
O
CO2Me
HO OOH
OHOH
HO
H
H
3.13.19 3.20
O
O
O
MeO2C
O
O
O
MeO2C
Scheme 3.7: Design of 2nd generation pyranone intramolecular cycloaddition substrate 3.3.5 Synthesis of 2nd Generation Pyranone Intramolecular Cycloaddition Substrate
Synthesis of substrate 3.19 began with THP protection of 3-butyn-1-ol (3.21) in
75% yield to give alkyne 3.22 (Scheme 3.8).14 Alkyne 3.22 was then deprotonated with
nBuLi at -78 °C, and the resulting acetylide anion was trapped with methyl chloroformate
87
to give ynoate 3.23 in 93% yield.15 The THP group of 3.23 was then removed using
TsOH in methanol to furnish alcohol 3.24 in 86% yield.16
CO2Me
OTHPOH
DHP, TsOHDCM
25° C
75%OTHP
1. n-BuLi, THF2. ClCO2Me
-78° C
93%
TsOH, MeOH
25° C
86%
CO2Me
OH
(3.21) 3.22 3.23 3.24
Scheme 3.8: Synthesis of 2nd generation pyranone intramolecular cycloaddition alcohol
The pyranone substructure of cycloaddition substrate 3.19 was accessed from
furfuryl alcohol (3.16) (Scheme 3.9). Oxidation of furfuryl alcohol (3.16) using NBS and
water was followed by in situ acylation of the oxidized product with acetic anhydride to
furnish allylic acetate 3.25 in 53% yield.17 Allyic acetate 3.25 was then transformed to
the desired substrate 3.19 in 88% yield through a palladium-catalyzed allylic alkylation
reaction with alcohol 3.24.
OOH NBS, THF/H2O 4:1
Ac2O, NaHCO3
0 to 25 °C
53%
O
O
OAc
Pd2dba3.CHCl3PPh3, DCM
0 °C
88%
O
O
O
MeO2C
MeO2C
HO
3.16 3.25 3.19
3.24
Scheme 3.9: Synthesis of 2nd generation pyranone intramolecular cycloaddition substrate 3.3.6 Cycloaddition of 2nd Generation Pyranone Intramolecular Substrate
After having synthesized 3.19 the key [3+2] cycloaddition was attempted using
10 mol% of PBu3 in toluene at reflux (Table 3.3, entry 1). Gratifyingly, we were able to
isolate the desired cycloadduct 3.20 in 54% yield with ≥95:5 dr. TLC analysis during the
course of the reaction indicated that some type of polymerization was occuring. In an
88
attempt to inhibit this unfavorable side reaction, the phosphine catalyst loading was
lowered to 2.5 mol% (entry 2). Unfortunately, a virtually identical yield was obtained.
However, when the solvent concentration was lowered to 0.05 M (entry 3) an excellent
82% yield of 3.20 was obtained.
O
O
O
MeO2C
O
O
O
MeO2C
PBu3, PhMe
110 °C
3.19 3.20
Entry Mole % Conc. (M) Time (h) % Yield 1 10 0.1 1 54 2 2.5 0.1 1 55 3 2.5 0.05 1 82
Table 3.3: 1st generation pyranone intramolecular cycloaddition 3.3.7 Stereochemical Determination of Cycloaddition Product
In order to determine the stereochemistry at the bridgehead positions, (C-2, C-6,
C-9, C-11), a crystal structure of 3.20 was obtained (Figure 3.1). The crystal structure
revealed that the hydrogens at these bridgehead positions were all syn to one another.
O
O
O
MeO2C
2
116
9
H
H
3.20
Figure 3.1: Single crystal X-ray diffraction analysis of cycloadduct 3.20
89
3.3.8 Transition State Model for Intramolecular [3+2] Cycloaddition
A transition state model was proposed in order to account for the high levels of
diastereoselectivity in the [3+2] cycloaddition reaction (Figure 3.2). This model was
based on two premises. First, the acetal tether of 3.19 should direct the 1,3-dipole to a
single diastereotopic face of the alkene. Consequently, if the acetal tether is pointed
downward, as is depicted in cis-3.26 and trans-3.26, the dipole will be forced to react
with the bottom diastereotopic face of the alkene. The second premise was that the
reaction should proceed through an exo transition state, (see both cis-3.26 and trans-
3.26), where the 1,3-dipole is pointing out of the plane of the page, and away from the
pyranone ring system. An endo transition state should be disfavored since it would place
the bulky PBu3 directly underneath the pyranone ring system, causing an unfavorable
steric interaction. These two premises leave two possible transition state models, (cis-
3.26 and trans-3.26), that differ only in the geometry about the 1,3-dipole. The geometry
at the 1,3-dipole is important because it dictates the stereochemistry at position C-9 of the
product 3.20. Stereochemical model cis-3.26 has a cis-1,3-dipole, where the carbon
tether is on the same face of the 1,3-dipole as PBu3. In stereochemical model trans-3.26,
the carbon tether and PBu3 groups are on opposite faces of the 1,3-dipole. Transition
state model trans-3.26 gives rise to the observed product β-3.20. It is assumed that
transition state trans-3.26 is favored because it does not suffer from the steric interaction
between the carbon tether and the PBu3 that is present in cis-3.26. Furthermore, product
α-3.20 may be disfavored due to the strain present in the trans-fused five-membered ring.
90
O9
MeO2C O
O
O
O
O
MeO2C
O9
MeO2C O
OHH
[3+2]
OO
MeO2CPBu3
O
H
OO
MeO2CPBu3
HO
H H
cis- Dipole trans-Dipole
Observed Product
Favored Transition State
H
H
H
H
3.19 α−3.20 β−3.20
cis-3.26 trans-3.26
Figure 3.2: Transition state model for intramolecular [3+2] cycloaddition
3.4 Elaboration Of Cycloaddition Product to (+)-Geniposide
3.4.1 Retrosynthetic Analysis for Intramolecular [3+2] Cycloadduct
After completion of the key [3+2] cycloaddition, retrosynthetic analysis was done
to determine what steps were required to convert cycloaddition product 3.20 into (+)-
geniposide (3.1). It was concluded that four main synthetic challenges needed to be
addressed (Scheme 3.10). First, the 5-membered ring of the acetal of cycloadduct 3.20
had to be opened with some type of oxygenated nucleophile to produce alcohol 3.27.
Second, the alkene of 3.27 must be isomerized out of conjugation with the methyl ester to
produce allylic alcohol 3.28. Third, the C-C bond of methyl ester 3.28 needed to be
cleaved to produce 3.29. Finally, the α,β unsaturated methyl ester of (+)-geniposide (3.1)
needed to be installed upon ketone 3.29.
91
O
O
C-C Bond Formation
ORHO
H
HO
O
CO2Me
HO OOH
OHOH
HO
H
H
3.1 3.29
O
MeO2C O
O
Alkene Isomerization
Acetal Opening
O
MeO2C O
ORHO
H
H
H
H
3.27 3.20
O
MeO2C O
C-C Bond Cleavage
ORHO
H
H
3.28
Scheme 3.10: Retrosynthetic analysis for intramolecular [3+2] cycloaddition product 3.4.1 Acetal Opening
The first of these four main tasks that was undertaken was the acetal opening. To
setup this transformation, ketone 3.20 was reduced using sodium borohydride to form
alcohol 3.30 in 80% yield and in ≥10:1 dr (Scheme 3.11). Alcohol 3.30 was then treated
with TsOH and methanol in an attempt to open the five membered ring of the acetal and
gain access to methyl acetal 3.32. Surprisingly, when this reaction was conducted, an
undesired product, lactone 3.33, was isolated in 73% yield in what appeared to be a single
diastereomers by HNMR. Lactone 3.33 is thought to arise from opening of the 6-
membered ring of the acetal of 3.30 to give intermediate 3.34, which then undergoes
spontaneous lactonization to give lactone 3.33.
92
O
MeO2C O
O
NaBH4, MeOH
O
MeO2C
O
OH
0 °C
85%
TsOH, MeOH
O
MeO2C OH
OMeHO
0 °C
73%TsOH, MeOH0 °C
H
H
H
H
H
H
OHOH
O
H
H
MeO2C
OMe
3.20 3.30 3.32
3.34
OHO
O
H
H
O
OMe
3.33
Scheme 3.11: Acetal opening to incorrect regioisomer 3.4.2 Alkene Isomerization to Direct Regiochemistry in Acetal Opening
It was anticipated that the regiochemical problem in the acetal opening of 3.30
could be overcome by isomerizing the alkene of 3.30 (Scheme 3.12) prior to conducting
the acetal opening. Isomerization of alkene 3.30 to the more strained alkene 3.35 should
bring strain into the 5-membered acetal ring of 3.35. Hopefully when the acetal opening
of 3.35 was attempted, the 5-membered ring would open preferentially to relieve this
strain, and produce the desire product 3.36. Precedent for this transformation, is found in
the known conversion of acetal 3.37 to allylic alcohol 3.38 upon treatment with m-CPBA
in methanol.18
93
O
O
OHMeO2C
Strained Alkene
AlkeneIsomerization
O
MeO2C OH
O
O
O
OH OHHO
O
OAc OAcAcO
AcO OMe
1. m-CPBA, MeOH2. Ac2O, Pyridine
25 °C
34% 2 steps
H
H
H
H
H
H
H
H
O
MeO2C OH
OMeHO
H
H
H+MeOH
3.30 3.35 3.36
3.37 3.38
Scheme 3.12: Alkene isomerization to prepare for acetal opening 3.4.2 Alkene Isomerization via Diol Elimination
It was proposed that the alkene isomerization could be accomplished in a two step
procedure involving 1) oxidation of alkene 3.30 to diol 3.39, and 2) elimination of the
secondary alcohol with the β-H of 3.39 to produce alkene 3.40 (Scheme 3.13). To this
end, alkene 3.30 was dihydroxylated using a catalytic amount of osmium tetroxide.
Dihydroxylation was followed by spontaneous lactonization to furnish lactone-diol 3.39
in 58% yield. Next, elimination of the secondary alcohol of 3.39 to produce allylic
alcohol 3.40 was attempted using Tf2O and N-methyl-imidazole at 100 °C.19
Unfortunately, this method did not give any of the desired product. Additional attempts
were made to convert the secondary alcohol of diol 3.39 into a leaving group, which upon
a base-mediated elimination would produce 3.40, but these efforts were also fruitless.
These E2-type elimination reactions usually proceed through an anti-periplanar transition
state. However, the syn relationship between the secondary alcohol of and β-H of diol
3.39 requires a less favorable syn-periplanar transistion state for the elimination to occur.
94
It was reasoned that the elimination failed due to these stereochemical issues, and a
different method for alkene isomerization was sought.
O
H
H
MeO2C
O
OH
58%
OsO4, NMMOAcetone, H2O
O
H
HO
HO
HO
OO
Tf2O, NMI
100° C O
H
HO
HO OO
Hβ
3.30 3.39 3.40
Scheme 3.13: Alkene isomerization via dihydroxylation/elimination sequence 3.4.3 Alkene Isomerization via Base Mediated Epoxide Opening
The next strategy for accomplishing the isomerization of alkene 3.30 to the
desired substrate 3.40 involved epoxidation of alkene 3.30 with methy(trifluoromethyl)-
dioxirane (Scheme 3.14).20 The epoxidation was accompanied with spontaneous
lactonization to afford epoxy-lactone 3.41 in 53% yield. A base mediated epoxide
opening was then attempted on 3.41 in hopes of accessing allylic alcohol 3.40. It was
believed that this elimination would be more facile than the previously discussed
elimination of 3.39, since base mediated epoxide eliminations are known to proceed
through a transition state where the β-hydrogen and the epoxide are syn to one another.21
However, when the elimination was attempted using base, none of the desired allylic
alcohol 3.40 could be isolated.
O
MeO2C
O
OH Trifluoroacetone,Oxone, EDTACH3CN, H2O
O
O
OO
O
0 °C
53%
Base
O
O
OO
HO
Hβ
H
H
H
H
H
H
3.30 3.41 3.40
Scheme 3.14: Alkene isomerization via base-mediated epoxide elimination
95
One last effort was made to access the desired isomerized product 3.40 from
epoxide 3.41. The epoxide 3.41 was opened, using an equimolar mixture of titanium
tetrachloride and titanium isopropoxide, to produce chlorohydrin 3.42 in 84% yield
(Scheme 3.15). It was then proposed that a simple E2 elimination of the chloride and the
β-H of 3.42 would give the desired allylic alcohol 3.40. Unfortunately the product 3.40
could not be isolate from these reaction.
O
O
OO
OTiCl4, Ti(OiPr)4
DCM
0 °C
84%
O
O
OO
HO
Cl O
O
OO
HOBase
Hβ
H
H
H
H
H
H
3.41 3.42 3.40
Scheme 3.15: Alkene isomerization via halohydrin 3.4.4 Proposal of New Synthetic Route to (+)-Geniposide
After encountering serious difficulties in the acetal opening and alkene
isomerization of substrate 3.30, it was concluded that this synthetic route needed to be
abandoned. Although interesting chemistry had been developed, this route did not prove
to be effective for accessing the iridoid natural product (+)-geniposide (3.1). Future
synthetic studies focused on the development of a route to (+)-geniposide (3.1) through
an inter-molecular phophine-catalyzed [3+2] cycloaddition.
96
3.5 Experimental Procedures
General Procedures
All reactions were run under an atmosphere of argon under anhydrous conditions unless
otherwise indicated. Dichloromethane (DCM) was distilled from calcium hydride.
Tetrahydrofuran (THF) and ethyl ether (Et2O) were both distilled from sodium and
benzoquinone. Triethylamine (Et3N) was distilled from calcium hydride. All other
commercial reagents were used directly without further purification. Analytical thin-
layer chromatography (TLC) was carried out using 0.2-mm commercial silica gel plates
(DC-Fertigplatten Kieselgel 60 F254). Visualization of the chromatograms was
accomplished using UV light and vanillin, anisaldehyde, or permanganate stain with
heating. Solvents for chromatography are listed as volume:volume ratios. Preparative
column chromatography using silica gel was performed according to the method of
Still.22 Infrared spectra were recorded on a Nicolet 380 FTIR. High-resolution mass
spectra (HRMS) were obtained on a Waters Micromass Autospec or a Varian FTICR as
m/z (relative intensity). Accurate masses are reported for the molecular ion (M+1, M or
M-1) or a suitable fragment ion. Melting points were obtained on a Thomas-Hoover
Unimelt apparatus. Nuclear magnetic resonance spectra (1H NMR and 13C NMR) were
recorded with a Varian (400 MHz or 300 MHz) spectrometer as indicated and reported in
delta (δ) units, parts per million (ppm) referenced to the residual solvent signal as an
internal standards. Coupling constants are reported in hertz (Hz).
97
O
O
OO
MeO2C
6-Oxo-6H-pyran-3-carboxylic acid 3-methoxycarbonyl-prop-2-ynyl ester (3.6)
A flame-dried argon flushed flask was charged with γ-hydroxy-butynoate 3.11 (200 mg,
1.75 mmol), Et2O (5.8 mL, 0.3 M), and triethylamine (0.367 mL, 2.63 mmol, 150 mol%).
The solution was cooled to 0 ºC and a solution of acid chloride 3.13 (416 mg, 2.63 mmol,
150 mol%) in Et2O (2.9 mL, 0.6M) was added. The reaction was stirred and allowed to
warm to ambient temperature over 1 h. The reaction was then diluted with Et2O and
quenched with a saturated aqueous solution of NH4Cl. The organic layer was separated
and washed with saturated solutions of NaHCO3, water, and brine. The combined
organic layers were then dried over Na2SO4, concentrated in vacuo and purified by flash
column chromatography, (SiO2, 5:3 hexanes:EtOAc), to furnish the title compound as a
white solid (216 mg, 52%).
1H NMR: (400 MHz, CDCl3): δ 8.36 (dd, J = 2.4, 1.0 Hz, 1H), 7.78 (dd, J = 9.9, 2.7 Hz,
1H), 6.37 (dd, J = 9.9, 2.7 Hz, 1H), 5.00 (s, 2H), 3.80 (s, 3H). 13C NMR: (75 MHz, CDCl3): δ 161.9, 159.2, 158.9, 152.9, 141.1, 115.3, 110.9, 80.1,
78.2, 52.9, 52.0.
HRMS: Calcd. for C11H9O6 (M+1) 237.0399, Found: 237.0397.
FTIR: (neat): 3091, 2250, 1765, 1713, 1431, 1255, 1242, 1165, 1076, 962, 939, 843,
780, 770, 748 cm-1.
MP: 65-66 ºC
98
OTHP
2-Pro-2-ynyloxy-tetrahydro-pyran (3.9)
A flame-dried argon flushed flask was charged with propargyl alcohol (3.8) (10.6 mL,
178.4 mmol), DCM (60 mL, 3M), and p-toluenesulfonic acid (613 mg, 3.6 mmol, 2
mol%). The solution was cooled to 0 ºC and 3,4-dihydro-2H-pyran (17.9 mL, 196.2
mmol, 110 mol%) was added dropwise to the solution and the reaction was allowed to
warm to ambient temperature over 3 h. The reaction was diluted with DCM and washed
with a saturated solution of NaHCO3, water, and brine. The combined organic layers
were dried over Na2SO4, concentrated in vacuo, and purified by flash column
chromatography, (SiO2, 30:1 hexanes:EtOAc), to furnish the title compound as a clear oil
(21.45 g, 86%). The spectral data for this compound has been previously reported.7
CO2Me
OTHP
4-(Tetrahydro-pyran-2-yloxy)-but-2-ynoic acid methyl ester (3.10)
A flame-dried argon flushed flask was charged with alkyne 3.9 (10g, 71.3 mmol) and
THF (240 mL, 0.3 M). The solution was cooled to -78 ºC and 1.6 M MeLi in Et2O (49.1
mL, 78.5 mmol, 110 mol%) was added dropwise as the reaction turned black. The
reaction was stirred at -78 ºC for 1h and then warmed to -20 ºC for 1h. Methyl
chloroformate (6.6 mL, 85.6 mmol, 120 mol%) was added dropwise and the reaction was
stirred at -20 ºC for an additional 1 h and then warmed to ambient temperature for 2 h.
The reaction mixture was poured into a saturated solution of aqueous NaHCO3. This
solution was extracted with Et2O and the combined organic layers were then washed with
water. The organic solution was dried over Na2SO4, concentrated in vacuo, and purified
by flash column chromatography (SiO2, 9:1 hexanes:EtOAc) to yield the product as clear
99
light yellow oil (5.84 g, 41%). The spectral data for this compound has been previously
reported.9
CO2Me
OH
4-Hydroxy-but-2-ynoic acid methyl ester (3.11)
A flask was charged with alkyne 3.10 (4.43 g, 22.4 mmol) and MeOH (112 mL, 0.2M),
and then cooled to 0 ºC. p-Toluenesulfonic acid (425 mg, 2.24 mmol, 10 mol%) was
added and the reaction was allowed to slowly warm to ambient temperature over 19 h.
The reaction was quenched with triethylamine (0.623 mL, 4.47 mmol, 20 mol%),
concentrated in vacuo, and purified by flash column chromatography, (SiO2, 1:1
hexanes:EtOAc), to furnish the title compound as a clear light yellow oil (2.10g, 82%).
The spectral data for this compound has been previously reported.23
O
O
OCl
6-Oxo-6H-pyran-3-carbonyl chloride (3.13)
A flame-dried argon flushed flask was charged with coumalic acid (3.12), (500 mg, 3.57
mmol), Et2O (7.1 mL, 0.5M), and PCl5 (1.115 g, 5.36 mmol, 150 mol%). The resulting
suspension was heated to 60 ºC under a reflux condenser until the coumalic acid
dissolved (~ 1 h). The remaining PCl5 was filtered, and the filtrate was diluted with
petroleum ether and cooled to induce crystallization. The resulting precipitate was
isolated by filtration and dried under vacuum to provide light yellow crystals (322 mg,
57%). The spectral data for this compound has been previously reported.24
100
O
O
O
O
Me
But-2-ynoic acid 5-oxo-5,6-dihydro-2H-pyran-2-yl ester (3.14)
A flame-dried argon flushed flask was charged with 6-hydroxy-6H-pyran-3-one 3.17
(580 mg, 5.08 mmol), tetrolic acid (3.18) (640 mg, 7.62 mmol, 150 mol%), and DCM (51
mL, 0.1 M). The solution was cooled to 0 ºC and 4-(dimethylamino)pyridine (62 mg,
0.51 mmol, 10 mol%), and N,N'-Dicyclohexylcarbodiimide (2.096 g, 10.16 mmol, 200
mol%) were added to the reaction. The reaction was stirred for 20 minutes at 0 ºC and
then ran through a column of silica gel (Et2O, 0.1% Et3N). The filtrate was concentrated
in vacuo and purified two times by flash column chromatograpy (SiO2 5:1
hexanes/EtOAc 0.1% Et3N) to furnish the title compound as a white solid (476 mg, 52%).
1H NMR: (400 MHz, CDCl3): δ 6.78 (dd, J = 10.3, 3.8 Hz, 1H), 6.25 (d, J = 3.42 Hz,
1H), 6.06 (d, J = 10.3 Hz, 1H), 4.30 (d, J = 17.1 Hz, 1H), 4.03 (d, J = 17.1 Hz 1H), 1.82
(s, 3H). 13C NMR: (100 MHz, CDCl3): δ 192.4, 151.2, 140.9, 128.3, 87.6, 86.9, 70.9, 66.6, 3.0.
HRMS: Calcd. for C9H9O4 (M+1) 181.0501, Found: 181.494.
MP: 62-63 ºC
101
O
OH
O
6-Hydroxy-6H-pyran-3-one (3.17)
A flame-dried argon flask was charged with furfuryl alcohol (3.16) (10g, 8.810 mL,
101.937 mmol) and DCM (510 mL, 0.2 M). The solution was cooled to 0 ºC and 70-75%
m-chloroperoxybenzoic acid (37.693 g, 152.906 mmol, 150 mol%) was added portion-
wise to the solution. The reaction was allowed to slowly warm to ambient temperature
for 6 h during which solid m-chlorobenzoic acid precipitated from the solution. The
solution was cooled to -78 ºC for 15 minutes and the solid m-chlorobenzoic acid was
filtered. The filtrate was concentrated in vacuo and purified by flash column
chromatography, (SiO2, 2:1 hexanes:EtOAc 1% acetic acid to 1:1 hexanes:EtOAc 1%
acetic acid), to furnish the title compound as a light yellow solid (9.06 g, 78%). The
spectral data corresponded to that of the previously reported.25
102
MeO2C
O
O
O
5-(5-Oxo-5,6-dihydro-2H-pyran-2-yloxy)-pent-2-ynoic acid methyl ester (3.19)
A flame-dried argon flushed flask was charged allylic acetate 3.25 (100 mg, 0.640 mmol,
120 mol%), and with δ-hydroxy-butynoate 3.24 (98 mg, 0.768 mmol, 120 mol%), and
DCM (1.3 mL, 0.5M). The solution was cooled to 0 ºC and triphenylphosphine (17 mg,
0.064 mmol, 10 mol%) and Pd2(dba)3.CHCl3 (17 mg, 0.016 mmol, 2.5 mol%) were
added. The reaction was stirred for 0.5 h at 0 ºC. The reaction was diluted with ether,
and quenched with a saturated solution of aqueous sodium bicarbonate. The aqueous
layer was extracted with Et2O and then the combined organic layers were dried over
Na2SO4, concentrated in vacuo, and purified by flash column chromatography (SiO2, 2:1
hexanes:EtOAc) to furnish the title compound as a clear light yellow oil (125 mg, 87%).
1H NMR: (400 MHz, CDCl3): δ 6.83 (dd, J = 10.4, 3.3 Hz, 1H), 6.06 (d, J = 10.4 Hz,
1H), 5.19 (dd, J = 3.3, 0.6 Hz, 1H), 4.41 (d, J = 16.8 Hz, 1H), 4.02 (dd, J = 16.8, 0.4 Hz,
1H), 3.89 (dt, J = 9.7, 6.5 Hz, 1H) 3.70 (dt, J = 9.7, 6.5 Hz, 1H), 3.67 (s, 3H), 2.61 (t, J =
6.5 Hz, 2H). 13C NMR: (100 MHz, CDCl3): δ 194.1, 153.6, 143.6, 127.7, 93.0, 85.6, 73.6, 66.1, 65.8,
52.5, 20.0.
HRMS: Calcd. for C11H13O5 (M+1) 225.0763, Found: 225.0763.
FTIR: (neat): 2954, 2241, 1705, 1254, 1104, 1078, 1048, 1001, 858, 751 cm-1.
103
O
OMeO2C
O
H
H
5-Oxo-2a,4a,5,6,7a,7b-hexahydro-2H-1,7-dioxa-cyclopenta[cd]indene-4-carboxylic
acid methyl ester (3.20)
A flame-dried argon flushed flask was charged with ynoate 3.19 (9g, 40.14 mmol) and
undistilled PhMe (803 mL, 0.05 M). The reaction was heated to 110 ºC under a reflux
condenser and freshly distilled tri-n-butylphosphine (0.248 mL, 1.00 mmol, 2.5 mol%)
was added and the reaction was stirred for 2 h. Air was bubbled through the solution to
oxidize the tri-n-butylphosphine and the solution was cooled in an ice bath. The solution
was concentrated in vacuo and purified by flash column chromatography (2:1:3
hexanes:EtOAc:DCM) to furnish the title compound as a white solid (7.36 g, 82%).
1H NMR: (400 MHz, CDCl3): δ 6.86 (t, J = 2.2 Hz, 1H), 5.23 (d, J = 4.7 Hz, 1H), 4.27
(d, J = 18.4 Hz, 1H), 4.24 (t, J = 8.8 Hz, 1H), 4.04 (dd, J = 8.6, 4.7 Hz, 1H), 3.98 (d, J =
18.4 Hz, 1H), 3.76-3.31 (m, 2H), 3.67 (s, 3H), 3.29 (dt, J = 8.9, 4.6 Hz, 1H). 13C NMR: (100 MHz, CDCl3): δ 206.9, 163.7, 146.8, 135.0, 101.5, 71.6, 69.9, 54.5, 51.8,
49.6, 48.9.
HRMS: Calcd. for C11H13O5 (M+1) 225.0763, Found: 225.0766.
FTIR: (neat): 2913, 1732, 1699, 1319, 1250, 1197, 1108, 1078, 1018, 945, 931, 755 cm-
1.
MP: 86-87 ºC.
104
OTHP 2-But-3-ynyloxy-tetrahydro-pyran (3.22)
A flame-dried argon flushed flask was charged with 3-butyn-1-ol (3.21) (15g, 214
mmol), and DCM (1070 mL, 0.2 M). The solution was cooled to 0 ºC and p-
toluenesulfonic acid (369 mg, 2.14 mmol, 1 mol%) was added followed by 3,4-dihydro-
2H-pyran (25.2 mL, 278 mmol, 130 mol%). The reaction was slowly warmed to ambient
temperature for 15 h. The solution was neutralized with solid NaHCO3 (539 mg, 6.42
mmol, 3 mol%) and a saturated aqueous solution of NaHCO3 (375 mL) and stirred for 15
minutes. The organic layer was separated, washed with brine, and then dried over
sodium sulfate. The organic solution was then concentrated in vacuo, and purified by
flash column chromatography (SiO2, 19:1 hexanes:EtOAc) to furnish the title compound
as a clear liquid (24.84 g, 75%). The spectral data corresponded to that of the previously
reported material.14
105
CO2Me
OTHP 5-(Tetrahydro-pyran-2-yloxy)-pent-2-ynoic acid methyl ester (3.23)
A flame-dried argon flushed flask was charged with alkyne 3.22 (1.98 g, 12.87 mmol),
and THF (26 mL, 0.5 M). The solution was cooled to -78 ºC and 2.5 M nBuLi in hexanes
(5.66 mL, 14.15 mmol, 110 mol%) was added drop-wise. The reaction was stirred at -78
ºC for 0.5 h and then methyl chloroformate (1.19 mL, 15.41 mmol, 120 mol%) was
added. The reaction was stirred at -78 º for an additional 1h and then warmed to ambient
temperature over 1 h. The reaction was quenched with a saturated solution of aqueous
ammonium chloride and extracted with Et2O. The combined organic layers were washed
with brine, dried over Na2SO4, concentrated in vacuo, and purified by flash column
chromatography (SiO2, 9:1 hexanes:EtOAc) to furnish the title compound as a yellow oil
(2.55 g, 93%). The spectral data corresponded to that of the previously reported
material.15
CO2Me
OH 5-Hydroxy-pent-2-ynoic acid methyl ester (3.24)
A flame-dried argon flushed flask was charged with 3.23 (2.55 g, 12.01 mmol), and
MeOH (60 mL, 0.2 M). The solution was cooled to 0 ºC and p-toluenesulfonic acid (21
mg, 0.12 mmol, 1 mol%) was added. The reaction was warmed to ambient temperature
over 7 h. The reaction was diluted with Et2O and quenched with a saturated solution of
NaHCO3. The aqueous layer was washed with Et2O 5x and the combined organic layers
were then washed with a small amount of brine and dried over sodium sulfate. The
solution was concentrated in vacuo, and purified by flash column chromatography (SiO2,
2:1 hexanes:EtOAc) to furnish the title compound as a clear oil (1.32 g, 86%). The
spectral data corresponded to that of the previously reported material.16
106
O
OAc
O
Acetic acid 5-oxo-5,6-dihydro-2H-pyran-2-yl ester (3.25)
A flame-dried argon flushed flask was charged with furfuryl alcohol (3.16) (4.41 mL,
50.97 mmol), and 4:1 solution of THF:H2O (25.45 mL, 2 M). The reaction was cooled to
0 ºC and a finely ground mixture of N-bromosuccinimide (9.98 g, 56.07 mmol, 110
mol%) and NaHCO3 (8.56g, 101.94 mmol, 200 mol%) was added portionwise over 15
minutes. Acetic anhydride (9.62 mL, 101.93 mmol, 200 mol %) was added and the
reaction was slowly warmed to ambient temperature overnight. The reaction was
neutralized with solid NaHCO3 and a saturated solution of aqueous NaHCO3. The
aqueous layer was washed with EtOAc 5x. The combined organic layers were then
washed with brine, dried over Na2SO4, concentrated in vacuo, and purified by flash
column chromatography (SiO2, 5:1 hexanes:EtOAc) to furnish the title compound as an
orange oil (4.244 g, 53%). The spectral data corresponded to that of the previously
reported material.17
107
O
OHMeO2C
O
H
H
5-Hydroxy-2a,4a,5,6,7a,7b-hexahydro-2H-1,7-dioxa-cyclopenta[cd]indene-4-
carboxylic acid methyl ester (3.30)
A flask was charged with cycloadduct 3.19 (100 mg, 0.892 mmol) and MeOH (9.92 mL,
0.1 M). The reaction was cooled to 0 ºC and sodium borohydride (34 mg, 0.892 mmol,
100 mol%) was added to the reaction. The reaction was stirred at 0 ºC for 40 minutes.
The reaction was quenched with a saturated aqueous solution of ammonium chloride and
extracted with DCM 5x. The combined organic layers were washed with brine and dried
over Na2SO4. The solution was concentrated in vacuo and purified by flash column
chromatography (5:4 hexanes:EtOAc) to furnish the title compound as a white solid in ≥
10:1 dr (162 mg, 80%).
1H NMR: (400 MHz, CDCl3): δ 6.82 (t, J = 2.2 Hz, 1H), 5.23 (d, J = 5.7 Hz, 1H), 4.57
(d, J = 9.6 Hz, 1H), 4.19-4.13 (m, 1H), 3.85 (dd, J = 9.4, 2.7 Hz, 1H), 3.79-3.68 (m, 3H),
3.77 (s, 3H), 3.48-3.41 (m, 2H), 2.99 (dt, J = 9.2, 5.5 Hz, 1H). 13C NMR: (100 MHz, CDCl3): δ 166.2, 146.6, 136.8, 100.8, 67.4, 65.2, 65.1, 52.1, 49.6,
46.5, 42.2.
HRMS: Calcd. for C11H15O5 (M+1) 227.0919, Found: 227.0923.
FTIR: (neat): 3519.0, 2945.8, 1711.0, 1235.1, 1203.1, 1064.0, 1026.1, 1013.7, 952.8,
914.5, 858.5, 760.0.
MP: 73-76 ºC
108
OHO
O
H
H
O
OMe
3-Hydroxymethyl-4-methoxy-3,3a,3b,4,6,6a-hexahydro-cyclopenta[1,2-c;3,4-
c']difuran-1-one (3.33)
A flame-dried argon flushed flask was charged with 3.30 (100 mg, 0.442 mmol), MeOH
(2.2 mL, 0.2 M), p-toluenesulfonic acid (4 mg, 0.022 mmol, 5 mol%), and stirred
overnight. The reaction was quenched with solid K2CO3 and stirred for 15 minutes. The
solid K2CO3 was filtered. The filtrate was concentrated in vacuo on to silica gel and
purified by flash column chromatography (SiO2, 2:1 EtOAc:hexanes) to furnish the title
compound as a clear film (74 mg, 73%).
1H NMR: (400 MHz, CDCl3): δ 6.63 (q, J = 1.2 Hz, 1H), 5.70 (d, J = 5.1 Hz, 1H), 4.26
(dd, J = 8.8, 4.3 Hz, 1H), 3.90-3.84 (m, 2H), 3.82 (dd, J = 11.5, 3.7 Hz, 1H), 3.75-3.71
(m, 1H), 3.74 (s, 3H), 3.57 (ddd, J = 8.9, 5.2, 0.8 Hz, 1H), 3.44-3.38 (m, 2H), 2.35 (br s
1H). 13C NMR: (100 MHz, CDCl3): δ 164.8, 144.3, 137.8, 109.6, 87.3, 70.0, 64.6, 54.2, 51.8,
50.9, 49.6.
HRMS: Calcd. for C11H15O5 (M+1): 227.0919, 227.0921.
FTIR: (CHCl3): 3468, 2950, 2161, 2030, 1979, 1712, 1278, 1195, 1100, 1077, 1052,
1004, 751 cm-1.
109
O
OH
OH
O
HO
HO
2a,3-Dihydroxy-octahydro-1,5,6-trioxa-cyclopenta[jkl]-as-indacen-2-one (3.39)
A flask was charged with alcohol 3.30 (2.34 g, 10.34 mmol), N-methyl morpholine oxide
(1.82 g, 15.51 mmol, 150 mol%), and 9:1 acetone:H2O (25.9 mL, 0.5 M). Osmium
tetroxide (79 mg, 0.310 mmol, 3 mol%) was added and the reaction was stirred for 3 h at
ambient temperature. The reaction was diluted with EtOAc, and then sodium
metabisulfite (2.95 g, 15.51 mmol, 150 mol%) was added and the reaction was stirred for
15 minutes. The solution was filtered through a pad of silica with isopropanol. The
resulting filtrate was concentrated in vacuo and purified by flash column chromatography
(SiO2, EtOAc). The material was purified by flash column chromatography a second
time (SiO2, 15:1 DCM:IPA) to furnish the title compound as a white solid (1.38 g, 58%).
1H NMR: (400 MHz, d6-DMSO): δ 5.53 (s, 1H), 5.51 (s, 1H), 4.90 (d, J = 5.1 Hz, 1H),
4.45 (d, J = 9.6, 1H), 3.85-3.79 (m, 4H), 3.50 (dd, J = 13.8, 2.2 Hz, 1H), 3.16 (t, J = 10.4
Hz, 1H), 2.85-2.81 (m, 1H), 2.67-2.61 (m, 1H). 13C NMR: (100 MHz, d6-DMSO): δ 176.4, 101.2, 83.8, 78.1, 70.1, 69.8, 64.1, 51.6, 42.9,
39.2.
HRMS: Calcd. for C10H13O6 (M+1): 229.0712, Found: 229.0712.
FTIR: (neat): 3422, 3380, 1770, 1208, 1194, 1116, 1096, 1078, 1054, 1032, 1017, 984,
954, 933, 653 cm-1.
MP: 189-192 ºC (decomp).
110
O
OH
OH
O
O
Epoxide (3.41)
A flask was charged with alcohol 3.30 (250 mg, 1.105 mmol), acetonitrile (8.5 mL, 0.13
M), and a 1x10-4 M aqueous solution of ethylenediaminetetraacetic acid disodium salt
(6.5 mL, 0.17 M). The solution was cooled to 0 ºC and 1,1,1-trifluoroacetone (3.3 mL,
36.47 mmol, 3300 mol%) was added via a pre-cooled syringe. A mixture of Oxone
(1.358 g, 2.21 mmol, 200 mol%) and NaHCO3 (557 mg, 6.63 mmol, 600 mol%) was
added in one portion and the reaction was stirred for 1 h at 0 ºC. Two additional
equivalent portions of Oxone and NaHCO3 were added at hourly intervals for a total of 3
additions after which the reaction was deemed complete by TLC. The resulting
suspension was filtered and washed with DCM. The aqueous filtrate was extracted with
DCM and the combined organic layers were washed with brine and dried over Na2SO4.
The organic solutions were concentrated in vacuo and purified by flash column
chromatography (SiO2, 3:2 EtOAc:hexanes) to furnish the title compound as clear
crystals (124 mg, 53%).
1H NMR: (400 MHz, d6-DMSO): δ 4.98 (d, J = 5.1 Hz, 1H), 4.78 (ddd, J = 9.9, 3.5, 1.3
Hz, 1H), 4.05 (t, J = 9.3 Hz, 1H), 3.95-3.91 (m, 3H), 3.67 (dd, J = 13.9, 3.5, 1H), 3.30-
3.27 (m, 1H), 3.11 (dt, J = 9.2, 5.7, Hz, 1H), 2.90 (dt, J = 8.9, 5.1 Hz, 1H). 13C NMR: (100 MHz, d6-DMSO): δ 169.9, 102.1, 72.6, 72.2, 70.1, 68.4, 65.4, 51.2, 47.0,
36.4.
HRMS: Calcd. for C10H11O5 (M+1): 211.0606, Found: 211.0607
FTIR: (neat): 2910, 1771, 1249, 1129, 1107, 1049, 1026, 984, 970, 939, 923, 909, 901,
722 cm-1.
MP: 141-142 °C
111
O
OH
OH
O
HO
Cl
3-Chloro-2a-hydroxy-octahydro-1,5,6-trioxa-cyclopenta[jkl]-as-indacen-2-one (3.42)
A flame-dried argon flushed flask was charged with epoxide 3.41 (100 mg, 0.476 mmol),
and DCM (4.8 mL, 0.1 M). The reaction was cooled to 0 ºC and titanium tetrachloride
(0.026 mL, 0.238 mmol, 50 mol%) and titanium isopropoxide (0.070 mL, 0.238 mmol,
50 mol%) were added. The reaction was stirred at 0 ºC for 1 h and then concentrated in
vacuo onto silica gel and purified by flash column chromatography (SiO2, 4:1
EtOAc:Hexanes) to furnish the title compound as a white solid (99 mg, 84%).
1H NMR: (400 MHz, d6-DMSO): δ 6.87 (s, 1H), 5.22 (d, J = 5.8 Hz, 1H), 4.64 (dt, 1H, J
= 8.2, 2.1 Hz, 1H), 4.54 (d, J = 8.2 Hz, 1H), 3.88 (dd, J = 13.3, 2.4, 1H), 3.82 (dd, J =
9.1, 8.0 Hz, 1H), 3.55 (t, J = 9.1 Hz, 1H), 3.51 (dd, J= 13.2, 1.5 Hz, 1H), 3.25-3.15 (m,
1H), 3.00 (dd, J = 12.0, 8.2 Hz, 1H), 2.87 (ddd, J = 11.9, 10.0, 5.9 Hz, 1H).
HRMS: Calcd. for C10H12O5Cl (M+1): 247.0373, Found: 247.0373.
112
3.6 1H and 13C NMR Spectra
113
O
O
OO
MeO2C
3.6
114
O
O
O
O
Me
3.14
115
MeO2C
O
O
O
3.19
116
O
OMeO2C
O
H
H
3.20
117
O
OHMeO2C
O
H
H
3.30
118
OHO
O
H
H
O
OMe
3.33
119
O
OH
OH
O
HO
HO
3.39
120
O
OH
OH
O
O
3.41
121
O
OH
OH
O
HO
Cl
3.42
122
3.7 References
1 a) Wang, J.-C.; Ng, S.-S.; Krische, M. J. J. Am. Chem. Soc. 2003, 125, 3682. (b) Wang, J.-C.; Krische, M. J. Angew. Chem. Int. Ed. 2003, 42, 5855. 2 Zhang, C.; Lu, X. J. Org. Chem. 1995, 60, 2906. 3 Other structurally related iridoid natural products were initially targeted, however geniposide proved to be the final target. For the sake of clarity, this review will describe our synthetic progress in iridoid synthesis as if geniposide were the original target. 4 (a) Ueda, S.; Iwahashi, Y.; Tokuda, H. J. Nat. Prod. 1991, 54, 1677. (b) Lee. M.-J.; Hsu, J.-D.; Wang, C.-J.. Anticancer Res. 1995, 15, 411. 5 Koo, H.-J.; Lim, K.-H.; Jung, H.-J.; Park, E.-H. J. Ethnopharmacol. 2006, 103, 496. 6 Zhang, C.-Y; Parton, L. E.; Ye, C. P.; Krauss, S.; Shen, R.; Lin, C.-T; Porco, J. A.; Lowell, B. B. Cell Metab. 2006, 3, 417. 7 Blond, G.; Bour, C.; Salem, B.; Suffert, J. Org. Lett. 2008, 10, 1075. 8 Rossi, R.; Carpita, A.; Cossi, P. Tetrahedron 1992, 48, 8801. 9 Leonard, M. S.;Carrol, P. J.; Joullié M. M. J. Org. Chem. 2004, 69, 2526. 10 Tamaru, Y.; Kimura, M.; Tanaka, S.; Kure, S.; Yoshida, Z-. I. Bull. Chem. Soc. Jpn. 1994, 67, 2838. 11 See, Fried, J.; Elderfield, R. C. J. Org. Chem. 1941, 6, 577 and references cited therein. 12 Wang, J.-C.; Ng, S.-S.; Krische, M. J. J. Am. Chem. Soc. 2003, 125, 3682. 13 Lefebvre, Y. Tetrahedron Lett. 1972, 2, 133. 14 Caussanel, F.; Deslongschamps, P.; Dory, Y. L. Org. Lett. 2003, 5, 4799. 15 Blazykowski, C.; Harrak, Y.; Gonçlaves, M.-H.; Cloarec, J. -M.; Dhimane, A.-L.; Fensterbank, L.; Malacria, M. Org. Lett. 2004, 6, 3771. 16 Maguire, R. J.; Munt, S. P.; Thomas, E. J. J. Chem. Soc. Perkin Trans. 1 1998, 1, 2853.
123
17 Caddick, S.; Khan, S. Frost, L. M.; Smith, N. J.; Cheung, S.; Pairaudeau, G. Tetrahedron 2000, 56, 8953. 18 Marco-Contelles, J.; Juliana Ruiz-Caro, J. J. Org. Chem. 1999, 64, 8302. 19 Trost, B.; Krische, M. J. J. Am. Chem. Soc. 1999, 121, 6131. 20 Yang, D.; Wong, M.-K.; Yip, Y.-C. J. Org. Chem. 1995, 60, 3887. 21 Hodgson, D. M.; Gibbs, A. R.; Lee, G. P. Tetrahedron, 1996, 14361. 22 Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923. 23 Harland, P. A.; Hodge, P. Synthesis 1982, 3, 223. 24 Csihony, S.; Mika, L. T.; Vlád, G.; Barta, K.; Mehnert, C. P.; Horváth, I. T. Collect. Czech. Chem. Commun. 2007, 72, 1094. 25 Hoffmann, H. M. R.; Krumwiede, D.; Mucha, B. Oehlerking, H. H. Prahst, G. W. Tetrahedron, 1993, 49, 8999.
124
Chapter 4 Intermolecular Approach to (+)-Geniposide
4.1 Intermolecular Cycloaddition Retrosynthetic Analysis
The second generation retrosynthesic analysis of (+)-geniposide (4.1) begins with
disconnection of the β-glycoside to structure 4.2 (Scheme 4.1). It was proposed that the
α,β-unsaturated methyl ester of 4.2 could be accessed through a one carbon homologation
of ketone 4.3. Ketone 4.3 was envisioned to arise from the key intermolecular
phosphine-catalyzed [3+2] dipolar cycloaddition between 1,3-dipole intermediate 4.4 and
dipolarophile (S)-4.6. The 1,3-dipole intermediate could be accessed from ethyl-2-
butynoate (4.5) upon treatment with a trialkylphosphine. Finally, it was hoped that
compound (S)-4.6 could be synthesized in enantiopure form, through a palladium-
catalyzed kinetic resolution of racemic substrate rac-4.6.
O
EtO2C OR
[3+2]
O
O
CO2Me
HO OOH
OHOH
HO
H
H
H
H
4.1
4.3 4.5
O
OR
H
H
4.2
CO2Me
HO
O
∗ O
OAc
O
(S)-4.6
KineticResolution
O
OAc
O
rac-4.6
EtO2C
∗ O
EtO2C OAc
R3P
O
PR3
4.4 (S)-4.6
125
Scheme 4.1: Second generation retrosynthetic analysis of (+)-geniposide
4.2 Intermolecular [3+2] Cycloaddition Reaction
4.2.1 Intermolecular [3+2] Cycloaddition with Butynoate
The first attempt at the key phosphine-catalyzed [3+2] cycloaddition incorporated
allylic acetate rac-4.61 as the dipolarophile, and commercially available ethyl-2-
butynoate (4.5) as the latent 1,3-dipole (Table 4.1). The reaction was attempted using
PBu3 as catalyst and toluene as solvent. Unfortunately none of the desired cycloadduct
4.8 could be isolated from the reaction mixture. The temperature, time, and
concentration of the reaction were altered in an attempt to gain access to desired product
4.8, but these efforts were also unsuccessful (entries 1-6).
O
OAc
OPBu3 (10 mol%)
Toluene
O
OAc
O
EtO2C
Me
EtO2C4.5 rac-4.6 4.8
Entry Conc. (M) Temp (°C) Time (h) % Yield 1 0.1 25 24 - 2 0.1 50 5 - 3 0.1 80 2 - 4 0.5 25 24 - 5 0.5 80 3 - 6 0.5 110 1 -
Table 4.1: Intermolecular [3+2] cycloaddition with butynoate 4.2.2 Intermolecular [3+2] Cycloaddition with Allenoate
Since the key [3+2] cycloaddition did not proceed when ethyl-2-butynoate (4.5)
was used as a precursor to 1,3-dipole intermediate 4.4, we switched to using
126
commercially available ethyl-2-butadienoate (4.7) as the 1,3-dipole intermediate
precursor (Scheme 4.2). We believed that ethyl-2-butadienoate (4.7) might promote the
desired [3+2] cycloaddition more readily than ethyl-2-butynoate (4.5) because it gives
direct formation of the 1,3-dipole intermediate upon addition of phosphine.2
Furthermore we found experimentally that allylic pivalate 4.9 served as a suitable
dipolarophile for the [3+2] cycloaddition with ethyl-2-butadienoate (4.7) (Scheme 4.2).
The allylic pivalate 4.9 could be accessed in two steps through 1) oxidation of furfuryl
alcohol (4.10) to lactol 4.11 in 78% yield, and 2) pivalate protection of the lactol 4.11 in
80% yield (Scheme 4.2). Under optimized reaction conditions, the desired phosphine-
catalyzed [3+2] cycloaddition between ethyl-2-butadienoate (4.7) and pivalate 4.9
proceeded in 63% yield to afford cycloadduct 4.12. The reaction required two
equivalents of pivalate 4.9, but the excess substrate can be recovered from the reaction in
96% recovery. In addition to product 4.12, other isomeric cycloaddition products were
also formed in the reaction. However these products were formed as complex mixtures
and could not be thoroughly characterized. It is postulated that the bulky pivalate
protecting group of 4.9 is beneficial for the [3+2] cycloaddition because it suppresses the
formation of other isomeric cycloadducts. Additionally, the pivalate protecting group of
4.9 provides an opportunity to render the synthesis asymmetric through a palladium-
catalyzed kinetic resolution which is discussed in section 4.3.
127
O
OPiv
O
O
OPiv
O
EtO2CEtO2C
PPh3 (10 mol %),Toluene
200 mol%4.7 4.124.9
H
H110 °C
63%
OOH
0 to 25 °C
78%
O
OH
O PivCl, DMAP,lutidine, DCM
0 to 25 °C
80%
O
OPiv
O
m-CPBA, DCM
4.94.114.10
Scheme 4.2: Intermolecular [3+2] cycloaddition with allenoate 4.2.3 Stereochemical Determination of [3+2] Cycloadduct
In order to establish the stereochemistry at the three stereogenic, (C-4, C-8, C-9),
centers of cycloadduct 4.12, a crystal structure was obtained (Figure 4.1). The crystal
structure revealed that the two bridgehead hydrogens, at C-4 and C-8, were syn to one
another, and that the pivalate group at C-9 was also in a syn relationship to these two
bridgehead hydrogens at C-4 and C-8. This stereochemical result was desirable as the
three contiguous stereocenters are set in an analogous fashion to the corresponding
stereocenters found in the target molecule (+)-geniposide (4.1)
O
O
984
H
H
4.12
OPivEtO2C
Figure 4.1: Single crystal X-ray diffraction analysis of [3+2] cycloadduct
128
4.2.4 Regiochemical Analysis of Intermolecular [3+2] Cycloaddition
Since the intermolecular [3+2] cycloaddition was regio- and diastereoselective, it
was essential to explain the stereochemical outcome of the [3+2] cycloaddition. The first
stereochemical aspect of the reaction that needed to be explained was the regiochemistry.
Specifically why was cycloadduct 4.12 formed preferentially over the regioisomeric
product 4.13 (Scheme 4.3)?
O
OPiv
O
O
O
EtO2CEtO2C OPiv
O
O
OPiv
EtO2C[3+2]
Observed4.7 4.9 4.12 4.13
Scheme 4.3: Regioselectivity of intermolecular [3+2] cycloaddition
This observed regiochemical outcome can be explained through FMO analysis of
the 1,3-dipole intermediate 4.14 (Figure 4.2), and the allylic pivalate dipolarophile 4.9
(Figure 4.3). First, in the resonance structure of 1,3-dipole intermediate 4.14, the
carbanion of the 1,3-dipole can reside either at the γ-carbon of the 1-3-dipole, as shown
in structure γ-4.14, or at the α-carbon, as shown in α-4.14. However, when the carbanion
is placed at the α-carbon, it is stabilized by the adjacent electron withdrawing ester
functionality. This carbanion stabilizing effect is not present in γ-4.14. As a result, the
HOMO coefficient at the α-C of 4.14 should be larger than the HOMO coefficient at the
γ-C. This qualitative analysis is supported by FMO analysis of related 1,3-dipoles, which
reveals that the HOMO coefficient at the α-C is 0.69 and the HOMO coefficient at the γ-
C is 0.55.3
129
γ
αPh3P
OOEt
γ
α
Ph3P
OOEt
α−C Larger HOMO
γ−4.14 α−4.14
Ph3P
OOEt
4.14
Figure 4.2: Orbital analysis of 1,3-dipole
Second, simple analysis of the resonance structure of allylic pivalate dipolarophile
4.9 reveals that the LUMO coefficient at the β-C of should be larger than the α-C because
resonance structure β-4.9 has a positive charge at that carbon (Figure 4.3).
α
β O
O
OPiv
α
β O
O
OPiv
β-Position Larger LUMO
4.9 β−4.9
Figure 4.3: Orbital analysis of allylic pivalate dipolarophile In order to maximize the HOMO-LUMO orbital overlap, the reaction should
proceed through transition state 4.15, which would give rise to the observed product 4.12
(Scheme 4.4).
γ
α
Ph3P
OOEt
α
βO
O
OPiv
O
OPiv
O
O
O
EtO2CEtO2C OPiv
PPh3
Observed
4.7 4.9 4.12
[3+2]
4.15
Scheme 4.4: Regiochemical transition state of intermolecular [3+2] cycloaddition
130
4.2.5 Diastereoselectivity of Intermolecular [3+2] Cycloaddition
The diastereochemical outcome of the intermolecular [3+2] cyloaddition reaction
can be explained through the use of simple steric arguments. The large pivalate
protecting group of 4.9 blocks the top face of the alkene (Figure 4.4). This forces the 1,3-
dipole to attack the bottom diastereotopic face of the alkene. This mode of attack gives
rise to the observed stereochemistry of the reaction.
OO
PPh3CO2Et
O
O
Me
Me
Me
H
4.14
4.9
Figure 4.4: Diastereochemical model for intermolecular [3+2] cycloaddition
4.3 Palladium-Catalyzed Kinetic Resolution
4.3.1 General Scheme for Enantioselective Synthesis
After having successfully completed our key [3+2] cycloaddition, we wanted to
access allylic pivalate 4.9 in enantiopure form and thereby render our synthesis
asymmetric (Scheme 4.5).
O
OPiv
O
O
O
EtO2CEtO2C OPiv
PPh3 (10 mol%)Toluene
110 °C
63%
H
HO
O
CO2Me
HO OOH
OHOH
HO
H
H
4.14.7 4.9 4.12
*
Scheme 4.5: Enantioselective total synthesis of (+)-geniposide 4.3.2 Mechanistic Outline of Palladium-Catalyzed Kinetic Resolution
131
Our strategy for reaching this goal was to do a palladium-catalyzed kinetic
resolution of allylic pivalate rac-4.9 to produce enantiomerically pure (S)-4.9 (Scheme
4.6). Mechanistically, we envisioned this to occur through selective ionization of the
undesired enantiomer of allylic pivalate rac-4.9 with a chiral palladium-(0) source to
produce palladium π-allyl intermediate 4.16. The electrophilic palladium-(II) π-allyl
intermediate 4.16, could then react with a generic nucleophile to give substitution product
4.17, and leave behind the desired allylic pivalate (S)-4.9 in enantioenriched form.
PdO
OPiv
O
Pd(0)L*, Nu
O
OPiv
O
O
Nu
O
O
O
*L*L
II
Nurac-4.9 (S)-4.94.16 4.17
Scheme 4.6: Mechanism of catalyzed kinetic resolution 4.3.3 Precedent for Palladium-Catalyzed Kinetic Resolution
Precedent for the proposed kinectic resolution is found in Trost�s synthesis of (+)-
aflatoxin B1 and B2a, where an asymmetric palladium-catalyzed kinetic resolution of a
related γ-acyloxybutenolides 4.18 is accomplished using phenol 4.19 as nucleophile
(Scheme 4.7). 4
OPd2dba.CHCl3Na2CO3O
OBoc
OHCHOH3CO
O
O
OCHO
H3CO
NH
NH
O O
PPh2 Ph2P
1.0 equiv. 0.45 equiv.
91% Theoretical
89% ee
4.18 4.19 4.20
132
Scheme 4.7: Related palladium-catalyzed kinetic resolution
Additionally, our proposal was guided by the elegant work of Feringa,5 wherein
he shows that the palladium-catalyzed allylic substitution of pyranone allylic acetate 4.6
with simple alcohols proceeds in high yields with retention of configuration (Scheme
4.8).
O
OAc
OPd(OAc)2, POPh3
DCM
O
OR
O
4.6
ROH
R= Me, Et, i-PrOH
95-96% Yield
4.21
Scheme 4.8: Related palladium-catalyzed reaction with alcohols 4.3.4 Palladium-Catalyzed Kinetic Resolution Optimization
We chose as standard reagents for our kinetic resolution, p-nitrobenzyl alcohol
4.22 as nucleophile, palladium allyl-chloride dimer as a catalyst, Trost ligand (R,R)-4.23
as the chiral ligand, 2,6-lutidine as base, and DCM as solvent (Table 4.2). The reaction
was first attempted at 25 °C and after 4 h a 54% theoretical yield of allylic pivalate (S)-
4.9 was isolated in 75% ee (entry 1). In this reaction, it appeared by TLC that water was
entering into the reaction and reacting with allylic pivalate 4.9 despite the fact that dry
solvents were used. To address this problem, sodium sulfate was added as a dessicant
and a substantial increase in ee to 83% was observed (entry 2). Further optimization
revealed that a decrease in reaction temperature to 4 °C gave a significant increase in
yield to 62% and a modest increase in ee to 84% (entry 3). Running the reaction with
sodium sulfate at 4 °C gave a substantial increase in ee to 95%, however the yield
decreased to 36% (entry 4). Fortunately, a decrease in solvent concentration from 0.1 M
133
to 0.05 M resulted in a substantial increase in yield to 54% with a similar 93% ee (entry
5). Other dessicants were tested in the reaction, and it was found that magnesium sulfate
was superior to sodium sulfate in terms of yield (entry 6). Next, the reaction was
attempted without base and the magnesium sulfate loading was increase to 200 mol%.
This resulted in a significant increase in yield to 90%, albeit with a decrease in ee to 78%
(entry 7). Finally it was found that the loading of p-nitrobenzyl alcohol 4.19 was
increased to 55 mol%, the allylic pivalate (S)-4.9 could be isolated in 70% yield and 92%
ee (entry 8). Furthermore, under these optimized conditions byproduct 4.24 could be
isolated in 96% yield and 60% ee.
O
OPiv
O
O
OPiv
ONO2
HO
O
O
O[η3−C3Η5PdCl]2 (1.0 mol %),
DCM
NH
NH
O O
PPh2 Ph2P
(R,R)-4.23 (3.0 mol%)
NO2
rac-4.9 (S)-4.9 4.244.22
Entry Nu: (mol%) Base (mol%) Additive (mol%) Conc. (M) Temp. (°C) Time (h) Yield (%) ee (%)1 50 Lutidine (100) - 0.1 25 4 54 75 2 50 Lutidine (100) Na2SO4 (100) 0.1 25 4 54 83 3 50 Lutidine (100) - 0.1 4 26 62 84 4 50 Lutidine (100) Na2SO4 (100) 0.1 4 21 36 95 5 50 Lutidine (100) Na2SO4 (100) 0.05 4 48 54 93 6 50 Lutidine (100) MgSO4 (100) 0.05 4 43 56 93 7 50 - MgSO4 (200) 0.05 4 46 90 78 8 55 - MgSO4 (200) 0.05 4 48 70 92
Table 4.2: Kinetic resolution of allylic pivalate 4.3.5 Determination of Absolute Stereochemistry
The absolute stereochemistry of (S)-4.9 was determined by converting it into 4,5-
dichlorophthalimide derivative 4.26 in 65% yield through another palladium catalyzed
134
substitution reaction with 4,5-dichlorophthalimie (4.25) (Scheme 4.9). An x-ray crystal
structure of compound 4.26 was obtained and the absolute stereochemistry was shown to
be the (S)-enantiomer (Figure 1.1).
(S)-4.9
(S) O
OPiv
O O
N
O
(3.0 mol %)PPh3, (9.0 mol %)Et3N, MgSO4, THF
PdCl
ClPd
HNO O
Cl Cl
OO
Cl Cl
25 ºC
65%
4.25
4.26 Scheme 4.9: Derivative of pivalate
O
N
O
OO
Cl Cl4.26
Figure 4.5: Determination of absolute stereochemistry 4.3.6 Transition State Model for Kinetic Resolution
A stereochemical model was needed to explain the results of the kinetic
resolution. Trost has reported a model for predicting the stereochemical outcome of
palladium-catalyzed asymmetric allylic alkylations using diphenylphosphino benzoic
ligands of type (R,R)-4.23 (Figure 4.6).6 This model assumes that ligand (R,R)-4.23
complexes palladium in a C2 symmetric fashion in a 1:1 palladium:ligand ratio to form a
complex of type 4.27. The phenyl rings of complex 4.27 orient themselves in a propeller
135
like fashion to avoid interaction with one another. This propeller conformation of the
phenyl rings results in two vertical phenyl �walls�, and two horizontal phenyl �flaps�.
The resulting C-2 symmetric chiral pocket that is created can be illustrated effectively by
cartoon structure 4.28 where the front left and the back right quadrants are blocked by
phenyl walls, and the front right and back left quadrants are open. In the case of
ionization of a leaving group from a palladium π-allyl complex in this chiral
environment, as depicted in structure 4.29, ionization will prefer to occur at the front
right quadrant, rather than the front left quadrant, to avoid steric interaction with the
phenyl wall. Furthermore, ionization from the back two quadrants of 4.29 is disfavored
because the palladium ligand complex tilts away from the π-allyl, and ionization occurs
exo to the palladium-ligand complex to maximize orbital overlap.6
4.28
NH HNOO
P PPd
Flap
Wall
Pd
4.29
4.27
Figure 4.6: Model for predicting stereochemistry in asymmetric allylic alkylation If this model is applied to the kinetic resolution of allylic pivalate 4.9 using
palladium ligand complex 4.29, the pivalate leaving group of the (R)-enantiomer of the
136
allylic pivalate (R)-4.9, would be placed in a favorable position for ionization under the
right quadrant flap (Figure 4.7, left structure). However when the (S)-enantiomer, (S)-
4.9, interacts with the palladium complex 4.29, the pivalate group is forced to reside
underneath the sterically encumbered �wall� of the complex (right structure in Figure
4.7). This model suggests that ionization of the (R)-4.9 pivalate enantiomer is
significantly faster than ionization for the (S)-4.9 pivalate enantiomer thereby allowing
for successful kinetic resolution.
OO
Pd
OPivO
O
Pd
PivO
4.294.29
(R)-4.9 (S)-4.9 Figure 4.7: Transition state for kinetic resolution
4.4 Retrosynthetic Analysis of (+)-Geniposide from Cycloadduct
After having successfully completed the key intermolecular [3+2] cycloaddition
and palladium-catalyzed kinetic resolution, further retrosynthetic analysis was done to
see what synthetic tasks needed to be completed to convert [3+2] cycloadduct 4.12 into
(+)-geniposide (4.1) (Scheme 4.10). It was proposed that the β-glycoside of (+)-
geniposide (4.1) could be accessed from compound 4.30. Compound 4.30 could be
acquired via esterification of nitrile 4.31. The allylic alcohol of nitrile 4.31 could be
formed through reduction of the ethyl ester of structure 4.32. And finally, structure 4.32
could be prepared from cycloadduct 4.12 through a one carbon homologation of the
ketone.
137
O
OPIv
O
O
CO2Me
HO OOH
OHOH
HO
H
H
H
H
4.1 4.31
O
OPiv
H
H
4.30
CO2Me
HO
CN
O
EtO2C OPiv
H
H
4.12
O
O
OPiv
O
EtO2C
4.7 (S)-4.9
*
HO
O
EtO2C OPIv
H
H
CN
4.32 Scheme 4.10: Retrosynthetic analysis of (+)-geniposide to [3+2] cycloadduct
4.5 One Carbon Homologation of [3+2] Cycloadduct
In order to convert cycloadduct 4.12 into α,β-unsaturated nitrile 4.32 a one carbon
homologation of the ketone of 4.12 needed to be accomplished (Scheme 4.11). A
classical method ketone homologation is to add cyanide to the ketone to produce the
cyanohydrin. This was accomplished by treating ketone 4.12 with an excess of
potassium cyanide and AcOH in ethanol to provide cyanohydrin 4.33. The crude
cyanohydrin 4.33 was then dehydrated with thionyl chloride and pyridine to from the α,β-
unsaturated nitrile 4.32 in 60% yield over 2 steps.
O
OPiv
O
EtO2C
O
OPivEtO2C
KCN, AcOH,EtOH
25 °C
SOCl2, Pyridine,DCE
O
OPivEtO2C
CNHO CN
80 °C
60%2 Steps
H
H H
H
H
4.12 4.33 4.32
Scheme 4.11: Formation of α,β-unsaturated nitrile
It should be noted at this point, that the two step conversion of ketone 4.12 to
methyl ester 4.50 was attempted (Scheme 4.12). This could be theoretically
138
accomplished by formation of enol trifalte 4.51 which could then be converted to ester
4.50 through a palladium-catalyzed coupling reaction of the triflate 4.51 with carbon
monoxide and methanol. This route would have provided access to the ester functionality
that is in the target (+)-geniposide 4.1 without the intermediacy of the nitrile 4.32.
However, we were unable to enact this process, due to the the instability of the enol
triflate 4.51. The corresponding enol phosphonates were also explored in this
transformation but were also unsuccessful.
O
OPiv
O
EtO2C
O
OPivEtO2C
Pd, CO,MeOH
O
OPivEtO2C
CO2MeH
H H
H
H
4.12 4.51 4.50
OTfN-PhenylTriflimide,LHMDS
Scheme 4.12: Attempted conversion of ketone to unsaturated ester.
4.6 Reduction of Ethyl Ester
4.6.1: Selectivity of Ester Reduction
With α,β-unsaturated nitrile 4.32 in hand, efforts were directed toward the
selective reduction of the ethyl ester to allylic alcohol 4.31 (Scheme 4.13). Although
reduction of the ethyl ester appears simple at first glance, several selectivity issues arise
when this transformation is analyzed in detail. First, the reduction must be
chemoselective, since two other reducible groups, the nitrile and the pivalate, are also
present in 4.32. Additionally, the ethyl ester is α,β-unsaturated and the reduction must
therefore be regioselective for 1,2 reduction.
139
O
OPivEtO2C
O
OPiv
CN CN
HO
H
H
H
H
4.314.32 Scheme 4.13: Selectivity issues in reduction of α,β-unsaturated ester 4.6.2: Optimization of DIBAL-H Reduction
With these selectivity issues in mind, the reduction of 4.32 was attempted using
several different reducing reagents. However, DIBAL-H was the only reagent that gave
significant quantities of the desired allylic alcohol 4.31. Our initial conditions for the
reduction used three equivalents of the DIBAL-H reagent, THF as solvent, and the
reaction was conducted at -78 °C (Table 4.3, entry 1). After 2 h the reaction was
quenched, and a 31% yield of the desired allylic alcohol 4.31 was isolated. The low yield
of 4.31 resulted from the reaction failing to go to completion. This result was surprising
since DIBAL-H reductions of esters are typically facile processes. In an attempt to
increase the reactivity of our ester 4.32 we switched to the less sterically encumbered
methyl ester 4.34 and attempted the reaction under the same conditions (entry 2).
Unfortunately a virtually identical yield was obtained. We then tried to increase the
reactivity of the DIBAL-H reagent by running the reaction in a less polar solvent (entries
3-4). Unfortunately, when the reaction was conducted in ether or toluene, reduction of
the nitrile of 4.34 began to compete with reduction of the ester and the yield of the allylic
alcohol 4.31 was diminished. Fortunately, when the reaction time was increased to 20 h
a 47% yield of the product could be isolated (entry 5). This result led us to decrease the
reaction temperature even further to -80 °C using a cryocool that could hold that
140
temperature accurately for 24 h. Gratifyingly, when the reaction was conducted under
these conditions on substrate 4.32 an acceptable 62% yield of allylic alcohol 4.31 was
obtained (entry 6).
O
OPivRO2C
O
OPiv
DIBAL-H (3 Equiv.)Solvent (0.1M)
CN CN
HO
H
H
H
H
R = Et 4.32R = Me 4.34
4.31
Entry Ester Solvent Temp (°C) Time (h) Yield %
1 Et THF -78 2 31 2 Me THF -78 2 30 3 Me Ether -78 2 23 4 Me Toluene -78 2 6 5 Me THF -78 20 47 6 Et THF -80 24 62
Table 4.3: Reduction of ethyl ester
4.7 Esterification of the Nitrile
4.7.1 Discussion of Classical Esterification Methods
The next step in the synthesis was the conversion of nitrile 4.31 to methyl ester
4.30. Classical methods for nitrile esterification involve the use of strongly acidic or
strongly basic conditions (Scheme 4.14). Unfortuately, these classical methods were not
suitable for our synthesis. The enol-ether moiety of 4.31 is acid sensitive, and the
pivalate group is base sensitive. Consequently, the esterification of nitrile 4.31 needed to
be conducted under virtually neutral conditions.
141
O O
OPiv
CO2MeCN
OPiv
Strong Acid/Base
HO HO
H
H H
H
4.31 4.30 Scheme 4.14: Esterification of nitrile 4.7.2 Platinum-Catalyzed Nitrile Hydration
It was proposed that the nitrile esterification of 4.31 could be conducted under
very mild reaction conditions by doing the conversion in a stepwise procedure. The first
step in this procedure was hydration of nitrile 4.31 to amide 4.35 in 87% yield using the
platinum catalyst 4.36 developed by Ghaffar and Parkins (Scheme 4.15).7 Notably, this
method for nitrile hydration allowed us to gain access to amide 4.35 in high yield using
essentially neutral reaction conditions.
PMe
O O
OPiv
CONH2CN
80 °C
87%
Pt PMe
MeOH
HOH
MeMe
Me
PO
OPiv
EtOH, H2O
HO
H
H HO
H
H
4.31 4.35
4.36 (20 mol%)
Scheme 4.15: Hydration of nitrile 4.7.3 Esterification of Amide
Next, amide 4.35 needed to be converted into the desired methyl ester 4.30.
Again, classical conditions for this transformation involve strongly acidic or basic
reaction conditions and were therefore unsuitable. However, we found that the amide
4.35 could be converted into the corresponding carboxylic acid 4.37 using sodium nitrite.
Thus when amide 4.35 was dissolved in a 2:1 mixture of acetic acid:acetic anhydride and
142
treated with sodium nitrite, carboxylic acid 4.37 could be isolated from the reaction
mixture (Scheme 4.16). However, this reaction also produced allylic acetate 4.38.
Allylic acetate 4.38 is thought to arise from protection of the allylic alcohol of 4.35 by
acetic anhydride.
O
OPiv
CO2H
HO
H
HO
OPiv
CONH2 Ac2O, AcOH,NaNO2
0 to 25 °C
HO
H
HO
OPiv
CO2H
AcO
H
H
4.35 4.37 4.38
Scheme 4.16: Conversion of amide to carboxylic acid Fortunately, this acetate protection could be used to our advantage. By changing
the reaction conditions we were able to get allylic acetate 4.38 as the sole product
(Scheme 4.17). Specifically this was accomplished by first dissolving the amide 4.35 in
acetic anhydride and adding two equivalents of Et3N. This solution was stirred for 1 h
until the allylic alcohol was fully protected. Next, AcOH and sodium nitrite were added
to conduct the diazotination reaction. This one-pot procedure gave allylic acetate
protected carboxylic acid 4.38. The crude acid 4.38 was then converted into the desired
methyl ester 4.39 using TMS-diazomethane. The overall yield for this two step
transformation was 74%.
143
O
OPiv
CO2Me
AcO
H
H
O
OPiv
CO2HTMSCH2N2,CHCl3, MeOH
0 °C
74%2 steps
AcO
H
HO
OPiv
CONH2Ac2O, Et3N,
then AcOH, NaNO2
25 °C then 0-5 °CHO
H
H
4.35 4.38 4.39
Scheme 4.17: Allylic acetate protection
4.8 Introduction of the β-Glycoside
4.8.1 Discussion of Iridoid Glycoside Formation
The last major synthetic transformation that needed to be accomplished was
installation of the β-glycoside onto 4.39 to give (+)-geniposide (4.1) (Scheme 4.18).
Installation of β-glycosides onto iridoid natural products is an extremely challenging task.
In fact syntheses of iridoid glycosides are rare due to this challenge. The difficulty in this
transformation results from the complex 1,1-diacetal linkage that must be made between
the β-glycoside and the carbon skeleton of the iridoid. Classical methods of
glycosidation typically give very low yields and poor selectivities.8
O
O
CO2Me
HO OOH
OHOH
HO
H
H
4.1
O
OPiv
H
H
4.39
CO2Me
AcO
OOH
OHOH
HOHO
Scheme 4.18: Iridoid glycoside formation 4.8.2 Glycosidation using trichloracetimidate
It was proposed that the β-glycoside formation could be accomplished on
proposed lactol substrate 4.40 using α-O-glycosyl-trichloroacetimidate9 4.41 as the
glycosidation reagent in the presence of a lewis acid to give a protected from of (+)-
144
geniposide 4.42 (Scheme 4.19). This trichloroacetimidate glycosidation method
developed by Schmidt, was chosen because it is known to proceed with weak
nucleophiles, and gives high β-selectivity of the glycoside when the proper conditions are
used.10
O
O
CO2Me
OOAc
OAcOAc
AcO
H
H
4.42
O
OH
H
H
4.40
CO2Me
OOAc
OAcOAc
AcO
4.41
L.A
RO ROOCl3C
NH
Scheme 4.19: Proposed glycosidatin of lactol 4.8.3 Formation of glycosidation substrate
In order to attempt the proposed glycosidation, lactol 4.40 had to be accessed
from pivalate 4.39. The most obvious way to accomplish this would be to first deprotect
the acetate and the pivalate of 4.39 under basic conditions to give the free lactol 4.43,
which is also known as genipin (Scheme 4.20). Subsequently, the primary alcohol of
genipin (4.43) could be selectively protected to give the desired glycosidation substrate
4.40.
O
OH
CO2Me
HO H
H
O
OPiv
CO2Me
AcO H
HBase
O
OHRO
CO2MeH
H
Protect
4.39 4.404.43
Scheme 4.20: Synthesis of lactol Unfortunately, this deprotection/protection strategy was unfeasible because the
lactol of genipin 4.43 undergoes rapid decomposition under mildly basic conditions. In
145
fact, it has been reported that genipin 4.43 is so sensitive to base that it decomposes when
boiled in neutral water.11 It was therefore necessary to find a method for deprotection of
4.39 under neutral reaction conditions.
4.8.4 Organotin-Catalyzed Deprotection
It has been reported by Otera, that the transesterification of esters can be
accomplished under virtually neutral conditions using organotin catalyst 4.44 (Scheme
4.21).12 In light of this, it was proposed that selective deprotection of the acetate group of
compound 4.39 could be accomplished through the action of Otera�s organotin catalyst
4.44 and MeOH. Surprisingly, when this reaction was conducted it was found that, under
optimized reaction conditions, lactol 4.46 was produced in 73% yield (Scheme 4.21).13
Lactol 4.46 is thought to arise from initial allylic acetate deprotection to furnish allylic
alcohol intermediate 4.45. Next, intermediate 4.45 undergoes an intramolecular
transesterification of the pivalate ester to provide lactol 4.46. It is postulated that the
driving force for the intramolecular transesterfication is derived from moving the bulky
pivalate group away from the congested cis-fused 6-5-ring system to the less substituted
primary alcohol.
O
OPiv
CO2Me
AcO H
H
O
CO2MeH
O
OH
CO2Me
PivO H
H
O
C(Me)3
HO
Bu2Sn O
Cl SnBu Bu
O
SnBu Bu
Cl
SnBu2Cl
Cl
PhMe 100 °C
O
4.44 (10 mol%)
4.39 4.464.45
MeOH
60 to 85 °C 73% yield
Scheme 4.21: Intramolecular tranesterification
146
It is noteworthy that solvent and temperature effects play a key role in this
transformation. The initial acetate removal from compound 4.39 to intermediate 4.45
was found to proceed most effectively in MeOH at 60 °C. However, the intramolecular
pivalate transfer of intermediate 4.45 to lactol 4.46 proceeded most effectively in toluene
at 100 °C. Consequently, the optimized reaction conditions involved initially conducting
the reaction in a sealed tube at 60 °C using methanol as solvent. After the initial acetate
deprotection to intermediate 4.45 was deemed complete by TLC, the sealed tube was
opened, toluene was added, and the reaction was heated to 85 °C until the methanol had
fully evaporated. The reaction vessel was then sealed and heated to 100 °C to promote
the final intramolecular transesterification to lactol 4.46.
4.8.5 Successful Glycosidation of Lactol
The formation of lactol 4.46 provided a perfect opportunity to attempt the
glycosidation reaction. Gratifyingly, it was found when lactol 4.46 was treated with α-O-
glycosyl-trichloroacetimidate 4.41 in the presence of boron trifluoride diethyl-etherate the
desired β-glycoside 4.47 was formed in 70% yield (Scheme 4.22).
O
O
CO2Me
OOAc
OAcOAc
AcO
H
H
4.47
O
OH
H
H
4.46
CO2Me
OOAc
OAcOAc
AcO
4.41
PivO PivOOCl3C
NH
BF3.OEt2, DCE
-20 °C
70%
Scheme 4.22: Glycosidation of lactol
4.9 Global Deprotection
Global deprotection of the pivalate and acetate protecting groups of 4.47 was
attempted using lithium hydroxide in methanol to give the target compound (+)-
147
geniposide (4.1) (Scheme 4.23). However under these conditions, methoxide added to
the β-position of the enoate of 4.47 producing compound 4.48 along with (+)-geniposide
(4.1) in approximately a 1:2 ratio as determined by H1NMR, respectively. This problem
was further exacerbated by the fact that these compounds 4.1 and 4.48 were inseparable
by column chromatography.
O
OGlu
CO2MeH
H
4.1
HO
LiOH, MeOH
25 °CO
O
CO2Me
OOAc
OAcOAc
AcO
H
H
4.47
PivO
O
OGlu
CO2MeH
H
4.48
HO
OMe
Scheme 4.23: Attempted global deprotection Fortunately, the global deprotection could be accomplished in two steps via initial
hydrolysis of 4.47 using LiOH in aqueous acetonitrile to carboxylic acid 4.49.14 Then the
intermediate acid 4.49 was esterified with TMS-diazomethane to yield the target
compound (+)-geniposide 4.1 in 61% yield over two steps (Scheme 4.24).
O
O
CO2Me
OOAc
OAcOAc
AcO
H
H
4.47
PivO
LiOH, CH3CN, H2O40 °C
61%2 Steps
O
O
CO2H
OOH
OHOH
HO
H
H
4.49
HO
TMSCH2N2, CHCl3,MeOH, 0 °C
4.1
Scheme 4.24: Final hydrolysis and esterification
4.10 Conclusion
In sum, the enantioselective total synthesis of (+)-geniposide 4.1 has been
completed in 14 steps in 1.8% overall yield.15 Key transformations in this synthesis
include an intermolecular phospine-catalyzed [3+2] cycloaddition reaction, and a
148
palladium-catalyzed kinetic resolution. Other interesting facets of the synthesis are 1) the
use of mild reaction conditions, and 2) the first use of an α-O-glycosyl-
trichloroacetimidate in the formation of an iridoid glycoside. This synthesis also
represents a formal synthesis of (+)-genipin, since the conversion of (+)-geniposide to
(+)-genipin has been previously reported.16
149
4.11 Experimental Procedures
General Procedures
All reactions were run under an atmosphere of argon under anhydrous conditions unless
otherwise indicated. Dichloromethane (DCM), dichloroethane (DCE), tetrahydrofuran
(THF), and toluene (PhMe) were obtained from a Pure-Solv MD-5 Solvent Purification
System (Innovative Technology, inc). Methanol was distilled from magnesium turnings
and iodine. Pyridine was dried with and stored over sodium hydroxide pellets.
Anhydrous solvents were transferred using oven-dried syringes. Thionyl Chloride
(SOCl2) was distilled from quinoline prior to use. Boron trifluoride diethyl etherate was
distilled prior to use. p-nitrobenzyl alcohol was recrystallized from DCM prior to use.
Magnesium Sulfate (MgSO4) was dried in an oven prior to use in the asymmetric kinetic
resolution. All other commercial reagents were used directly without further purification.
Analytical thin-layer chromatography (TLC) was carried out using 0.2-mm commercial
silica gel plates (DC-Fertigplatten Kieselgel 60 F254). Visualization of the
chromatograms was accomplished using UV light and vanillin, anisaldehyde,
permanganate, cerium molybdate stain with heating. Preparative column
chromatography using silica gel was performed according to the method of Still.17
Infrared spectra were recorded on a Nicolet 380 FTIR. High-resolution mass spectra
(HRMS) were obtained on a Waters Micromass Autospec or a Varian FTICR as m/z
(relative intensity). Accurate masses are reported for the molecular ion (M+1, M or M-1)
or a suitable fragment ion. Melting points were obtained on a Thomas-Hoover Unimelt
apparatus. Nuclear magnetic resonance spectra (1H NMR and 13C NMR) spectra were
recorded with a Varian (400 MHz) spectrometer and reported in parts per million (ppm)
referenced to the residual protio solvent signal as an internal standards. Coupling
constants are reported in hertz (Hz). Optical rotations were measured on a ATAGO AP-
300 automatic polarimeter at a path length of 1 dm.
150
O
O
CO2Me
HO OOH
OHOH
HOH
H
7-Hydroxymethyl-1-(3,4,5-trihydroxy-6-hydroxymethyl-tetrahydro-pyran-2-yloxy)-
1,4a,5,7a-tetrahydro-cyclopenta[c]pyran-4-carboxylic acid methyl ester (4.1)
A flask was charged with 4.47 (50 mg, 0.078 mmol), LiOH.H2O (49 mg, 1.171 mmol,
1500 mol%), and a 7:3 mixture of CH3CN:H2O (1.56 mL, 0.05 M). The reaction was
heated to 40 °C under a reflux condenser for 16.5 h and then quenched with AcOH
(0.090 mL, 1.56 mmol, 2000 mol%). The resulting solution was passed through a
column of Dowex®50WX8-200 ion-exchange resin with water and concentrated in
vacuo. The crude material was dissolved in 1:1 MeOH:CHCl3 (4 mL, 0.02 M) and a 2.0
M solution of TMS-diazomethane in hexane was added (0.078 mL 200 mol%) and the
reaction was stirred at ambient temperature for 1h. After 1 h an additional portion of
TMS-diazomethane in hexane was added (0.078 mL 200 mol%) and the reaction was
deemed complete after another 30 minutes. The excess TMS-diazomethane was
quenched with acetic acid and the resulting solution was concentrated in vacuo on to
silica gel. The material was purified by flash column chromatography (SiO2, 9:1
CHCl3:MeOH to 4:1 CHCl3:MeOH) and then ran through a column of Dowex®50WX8-
200 ion-exchange resin with MeOH. The filtrate was concentrated to furnish (+)-
geniposide as a white solid (18.4 mg 61%). The spectral data corresponded to that of the
previously reported material.18,19 Additionally, the spectral data for the material
corresponded to the spectral data of an authentic sample of (+)-geniposide purchased
from AvaChem Scientific LLC in a side by side comparison. 1H NMR: (400 MHz, CD3OD): δ 7.51 (s, 1H), 5.79 (s, 1H), 5.16 (d, J = 7.5 Hz, 1H), 4.70 (dd, J = 7.9, 2.1 Hz, 1H) 4.31 (d, J = 14.7, 1H), 4.18 (d, J = 14.4, 1H), 3.84 (d, J = 11.6, 1H), 3.70 (s, 3H), 3.63 (dd, J = 12.1, 5.2 Hz, 1H), 3.40-3.34 (m, 2H), 3.28-3.26 (m, 1H), 3.24-3.15 (m, 2H), 2.81 (dd, J = 16.2, 8.4 Hz, 1H), 2.72 (t, J = 7.7 Hz, 1H), 2.12-2.06 (m, 1H).
151
13C NMR: (100 MHz, CD3OD): δ 169.5, 153.3, 144.8, 128.3, 112.5, 100.3, 98.2, 78.4, 77.8, 74.9, 71.5, 62.7, 61.4, 51.7, 47.0, 39.7, 36.6. HRMS: Calcd. For C17H23O10 (M-1): 387.1293, Found: 387.1297. FTIR: (neat): υ 3362, 2920, 2487, 1697, 1630, 1282, 1160, 1074, 1037, 942, 893, 822, 795, 767 cm-1. M.P.: 123-124 °C
[α]24 D = +24.25 (C = 0.660, EtOH)
152
O
OPiv
O
2,2-Dimethyl-propionic acid 5-oxo-5,6-dihydro-2H-pyran-2-yl ester (rac-4.9)
A flame-dried argon flushed flask was charged with lactol 4.10 (2.0g, 17.53 mmol),
DCM (88 mL, 0.2M), 2,6-lutidine (4.08 mL, 35.06 mmol, 200 mol%) and DMAP (107
mg, 0.88 mmol, 5 mol%). The solution was cooled to 0 ºC and trimethylacetyl chloride
(3.24 mL, 26.29 mmol, 150 mol%) was added. The reaction was slowly warmed to room
temperature and stirred for 72 h. The reaction was diluted with ether and washed with
water, a saturated solution of sodium bicarbonate, a 5% solution of aqueous CuSO4 and
brine. The combined organic layers were dried over Na2SO4, concentrated in vacuo,
(Caution: compound rac-4.9 is volative under high vacuum), and purified by flash
column chromatography, (SiO2, 8:1 pentane:ether), to furnish the title compound as a
white solid (2.78 g, 80%).
1H NMR: (400 MHz, CDCl3): δ 6.90 (dd, J = 10.4, 3.6 Hz, 1H), 6.43 (d, J = 3.4 Hz,
1H), 6.22 (d, J = 10.6, 1H), 4.43 (d, J = 16.8, 1H) 4.17 (d, J = 17.1, 1H), 1.21, (s, 9H). 13C NMR: (100 MHz, CDCl3): δ 193.3, 176.7, 142.4, 128.5, 86.4, 67.1, 39.1, 26.8.
HRMS: Calcd. For C10H15O4 (M+1): 199.0970, Found: 199.0974.
FTIR: (neat): υ 2956, 1731, 1699, 1686, 1281, 1264, 1132, 1102, 1026, 1006, 989, 912,
879, 865, 778 cm-1.
MP: 45-46 ºC
O
O
OPiv
153
2,2-Dimethyl-propionic acid 5-oxo-5,6-dihydro-2H-pyran-2-yl ester ((S)-4.9))
A flame dried argon flushed flask was charged with DCM (20 mL, 0.05 M), magnesium
sulfate (400 mg, 200 wt %), p-nitrobenzyl alcohol (4.22) (85 mg, 0.554 mmol, 55 mol%),
pivalate rac-4.9 (200 mg, 1.01 mmol, 100 mol%) and Trost ligand (R,R)-4.23 (21 mg,
0.030 mmol, 3 mol%), respectively. The solution was cooled to 4 ºC and allyl palladium
chloride dimer (3.7 mg %, 0.010 mmol, 1.0 mol%) was added. The flask was sealed with
a cap and parafilm and stirred for 48 hours at 4 ºC. The reaction was diluted with ether,
and washed with a saturated solution of aqueous sodium bicarbonate, and brine. The
combined organic layers were dried over sodium sulfate, concentrated in vacuo (Caution:
compound (S)-4.9 is volative under high vacuum), and purified by flash column
chromatography (SiO2, 5:1 pentane:ether to 2:1 hexanes:EtOAc) to furnish the title
compound as a white solid (70.5 mg, 70%, 92% ee) and the p-nitrobenzyl derivative 4.24
as a light yellow solid (96 mg, 96%, 60% ee). Sixty-five milligrams of 92% ee
compound (S)-4.9 was recrystallized from pentanes twice to furnish 43 mg of the title
compound in >99% ee (66% recovery).
[α]24 D = +151.00 (c= 1.00, CHCl3).
HPLC: (Chiralpak AD-H column, 2% i-PrOH/hexanes, 0.5 mL/min, 230 nm), tmajor =
15.1 min, tminor = 21.0 min; ee = 92%.
(Chiralpak AD-H column, 2% i-PrOH/hexanes, 0.5 mL/min, 230 nm), tmajor = 15.7 m; ee
= >99%.
154
O
O
EtO2C OPivH
H
1-(2,2-Dimethyl-propionyloxy)-4-oxo-1,3,4,4a,5,7a-hexahydro-cyclopenta[c]pyran-7-
carboxylic acid ethyl ester (4.12)
A flame-dried argon flushed flask was charged with PhMe (45mL, 0.2M with respect to
ethyl butadienoate), and racemic pivalate 4.9 (3.535g, 17.836 mmol, 200 mol %). The
reaction vessel was heated to 110 °C and a catalytic amount of triphenylphosphine (234
mg, 0.892 mmol, 10 mol%) was added. Ethyl-2-butadienoate 4.7 (1.035 mL, 9.918
mmol, 100 mol%) was added dropwise. The reaction was stirred for 0.5 h. The reaction
mixture was cooled to ambient temperature and directly purified by flash column
chromatography (SiO2, 9:1 petroleum ether:ether to 5:1 hexanes:ethyl acetate) to
furnished the title compound as a white solid (1.732 g, 63%) and the unreacted pivalate
4.9 as a white solid (1.695 g, 96% theoretical recovery). The reaction was also conducted
using >99%ee (S)-4.9. 1H NMR: (400 MHz, CDCl3): δ 6.90 (d, J = 2.1 Hz, 1H), 6.45 (d, J = 0.7 Hz, 1H), 4.27-
4.14 (m, 2H), 4.18 (d, J = 18.1 Hz, 1H), 4.09 (d, J = 18.1 Hz, 1H), 3.47 (d, J = 8.9 Hz,
1H), 3.27 (dd, J = 15.6, 8.7 Hz, 1H), 2.90-2.78 (m, 2H), 1.28, (t, J = 7.2 Hz, 3H), 1.20, (s,
9H). 13C NMR: (100 MHz, CDCl3): δ 208.4, 175.9, 163.4, 144.7, 134.8, 90.5, 66.6, 60.7,
49.8, 46.0, 38.9, 36.3, 29.9, 14.1.
HRMS: Calcd. For C16H23O6 (M+1): 311.1500, Found: 311.1495
FTIR: (neat): 2976, 1735, 1717, 1702, 1257, 1161, 1149, 1130, 1104, 1077, 1025, 993,
944, 929, 848, 840, 761 cm-1.
MP: 75-76 ºC
[α]24 D = +72.00 (c= 1.00, CHCl3).
155
O
O
O
NO2
6-(4-Nitro-benzyloxy)-6H-pyran-3-one (4.24) 1H NMR: (400 MHz, CDCl3): δ 8.24 (d, J = 6.8 Hz, 2H), 7.54 (d, J = 8.9, 2H), 6.93,
(dd, J = 10.4, 3.2 Hz, 1H), 6.21, (d, J = 10.3 Hz, 1H), 5.32 (dd, J = 3.4, 0.6, 1H), 4.97 (d,
J = 13.0, 1H) 4.77 (d, J = 13.0, 1H), 4.45 (d, J = 16.8, 1H), 4.14 (d, J = 16.8, 1H). 13C NMR: (100 MHz, CDCl3): δ 194.0, 147.5, 144.4, 143.5, 128.1, 128.0, 123.7, 92.7,
69.3, 66.3 ppm.
HRMS: Calcd. For C12H12NO5 (M+1): 250.0719, Found: 250.0715
FTIR: (neat): 2916, 2853, 1483, 1516, 1346, 1334, 1106, 1054, 1034, 1005, 987, 976,
858, 832, 771, 735 cm-1.
MP: 108-110 ºC
HPLC: (Chiralpak AD-H column, 2% i-PrOH/hexanes, 1.0 mL/min, 254 nm), tminor =
47.1 min, tmajor = 51.8 min; ee = 68%.
156
O
O
N
Cl Cl
O O
(R)-5,6-dichloro-2-(5-oxo-5,6-dihydro-2H-pyran-2-yl)isoindoline-1,3-dione (4.26)
A flame-dried argon flushed flask was charged with THF, (2.5 mL, 0.1 M), >99% ee
pivalate (S)-4.9, (50 mg, 0.252 mmol), 4,5-dichlorophthalimide 4.25 (49 mg, 0.227
mmol, 90 mol%), MgSO4 (50 mg, 100 wt %), triphenylphosphine (6 mg, 0.023 mmol, 9
mol%), palladium allyl chloride dimer (3 mg, 0.008 mmol, 3 mol%), and triethylamine
(0.035 mL, 0.252 mmol, 100 mol%). The reaction was then stirred at room temperature
for 5 h. Afterwards, the reaction was diluted with EtOAc, and washed with a 5% solution
of CuSO4 and then brine. The combined organic layers were dried over sodium sulfate,
concentrated in vacuo, and purified by flash column chromatography (SiO2, 3:1 EtOAc:
hexanes) to furnish the title compound as a white solid (46 mg, 65% based on 4,5-
dichlorophthalimide). 1H NMR: (400 MHz, CDCl3): δ 8.00 (s, 2H), 7.02 (dd, J = 10.5, 2.6 Hz, 1H), 6.42 (dd, J
= 10.5, 2.3 Hz, 1H), 6.25 (m, 1H), 4.48 (d, J = 16.6 Hz, 1H), 4.31 (dd, J = 16.6, 1.2 Hz,
1H). 13C NMR: (400 MHz, CDCl3): δ 192.8, 165.1, 143.6, 140.0, 130.5, 129.5, 126.0, 72.6,
70.6.
HRMS: Calcd. For C13H7NO4Cl2Na+1: 333.9650, Found 333.9644.
FTIR: (neat): υ 1725, 1703, 1380, 1361, 1341, 1124, 1082, 895, 754, 748, 741 cm-1.
MP: 180 °C
[α]24 D = +124.28 (c= 0.89, CHCl3).
157
O
OPiv
CN
HO
H
H 2,2-Dimethyl-propionic acid 4-cyano-7-hydroxymethyl-1,4a,5,7a-tetrahydro-
cyclopenta[c]pyran-1-yl ester (4.31)
A flame-dried argon flushed flask was charged with racemic ester 4.32 (100 mg, 0.313
mmol) and THF (3.1 mL, 0.1M) and then cooled to -80 ºC. Reagent grade DIBAL-H
(167 µL, 0.939 mmol, 300 mol%) was added and the reaction was stirred for 24h at -80
ºC. The reaction was quenched with acetic acid (0.090 µL, 1.565 mmol, 500 mol%) and
partitioned between ethyl acetate and an aqueous solution of saturated Rochelle�s salt.
The biphasic solution was stirred vigorously for 1h until the aluminum salts precipitated
into the aqueous layer. The salts were filtered, and the organic layer was washed with
water and brine. The combined organic layers were dried over Na2SO4, concentrated in
vacuo, and purified by flash column chromatographed (SiO2, 3:1 petroleum ether:ether to
15:1 DCM:EtOAc) to furnish the title compound as a yellow oil (54mg, 62%). The
reaction was also conducted using optically active 4.32. 1H NMR: (400 MHz, CDCl3): δ 6.99 (s, 1H), 6.34 (d, J = 2.1, 1H), 5.81 (s, 1H), 4.27 (d,
J = 13.7 Hz, 1H), 4.22 (d, J = 13.7, 1H), 3.12 (s, 2H), 2.75 (d, J = 16.8, 1H), 2.48 (d, J =
16.1, 1H), 1.22 (s, 9H). 13C NMR: (100 MHz, CDCl3): δ 176.4, 153.2, 140.2, 128.7, 117.9, 94.4, 89.1, 60.1,
46.0, 38.9, 36.9, 33.4, 26.8.
HRMS: Calcd. For C15H20NO4 (M+1): 278.1392, Found: 278.1398.
FTIR: (neat): 3452, 2975, 2214, 1748, 1634, 1196, 1109, 1053, 1029, 982, 915, 824,
754 cm-1.
[α]24 D = -52.22 (c= 0.536, CHCl3).
158
O
CN
EtO2C OPiv
H
H
4-Cyano-1-(2,2-dimethyl-propionyloxy)-1,4a,5,7a-tetrahydro-cyclopenta[c]pyran-7-
carboxylic acid ethyl ester (4.32)
A flame dried argon flushed flask was charged with enantiopure ketone 4.12 (446 mg,
1.437 mmol), ethanol (7.2 mL, 0.2 M), and potassium cyanide (468 mg, 7.185 mmol, 500
mol%). Acetic acid (0.411 mL, 7.185 mmol 500 mol%) was added dropwise to the
solution and the reaction was stirred at ambient temperature for 5 hours. The solution
was diluted with ether, and then washed with water, and brine. The combined organic
layers were dried over Na2SO4, concentrated in vacuo, and taken on to the next step
without further purification. To a flask charged with the crude cyanohydrin, 4.33 was
added DCE (14.4 mL, 0.1 M), pyridine (455 mg, 5.748 mmol, 400 mol%), and thionyl
chloride (0.209 mL, 2.874 mmol, 200 mol%). The reaction was immediately immersed
in an 80 ºC oil bath and stirred under a reflux condenser for 2.5 hours. The reaction was
diluted with ether and then washed with water, a 5% aqueous solution of CuSO4, and
brine. The combined organic layers were dried over Na2SO4, concentrated in vacuo, and
purified twice by flash column chromatographed (SiO2, 3:1 hexanes: Et2O) to furnish the
title compound as an orange oil (277 mg, 60%). 1H NMR: (400 MHz, CDCl3): δ 6.99 (d, J = 0.7 Hz, 1H), 6.92 (d, J = 5.0, 2.6 Hz, 1H),
6.85 (d, J = 3.1 Hz, 1H), 4.29-4.17 (m, 2H), 3.37-3.35, (m, 1H), 3.22 (t, J = 7.2 Hz, 1H),
2.89-2.81 (m, 1H), 2.71 (dd, J = 18.5, 2.4 Hz, 1H), 1.31 (t, J = 7.2, 3H), 1.21 (s, 9H). 13C NMR: (100 MHz, CDCl3): δ 176.0, 163.5, 153.6, 145.5, 132.9, 117.6, 93.8, 88.3,
60.8, 45.6, 38.8, 37.2, 32.7, 26.7, 14.1.
HRMS: Calcd. For C17H22NO5 (M+1): 320.1503, Found: 320.1498
FTIR: (neat): 2974, 2217, 1737, 1701, 1642, 1269, 1103, 1021, 998, 922, 836, 771 cm-1
[α]24 D = +14.00 (c= 1.00, CHCl3).
159
O
OPiv
CONH2
HO
H
H
2,2-Dimethyl-propionic acid 4-carbamoyl-7-hydroxymethyl-1,4a,5,7a-tetrahydro-
cyclopenta[c]pyran-1-yl ester (4.35)
A flask was charged with racemic nitrile 4.31 (130 mg, 0.469 mmol), 2:1 ethanol:water
(2.4 mL, 0.2M) and platinum catalyst 4.36 (40 mg, 0.094 mmol, 20 mol%). The reaction
was heated to 80 ºC under a reflux condenser and stirred for 3h. The solution was
concentrated in vacuo and the crude material was chromatographed with (SiO2, EtOAc)
to furnish the title compound as a white solid (120 mg, 87%). The reaction was also
conducted using optically active 4.31.
1H NMR: (400 MHz, d6-DMSO): δ 7.17 (s, 1H), 7.16 (brs, 1H), 6.86 (brs, 1H), 5.80 (d,
J = 6.5 Hz, 1H), 5.69 (d, J = 1.0 Hz, 1H), 4.84 (t, J = 5.3 Hz, 1H), 4.06-3.92 (m, 2H),
3.21 (dd, J = 14.7, 7.2 Hz, 1H), 2.82 (t, J = 7.2 Hz, 1H), 2.69 (dd, J =16.2, 8.4 Hz, 1H),
2.04-2.00 (m, 1H), 1.16 (s, 9H). 13C NMR: (400 MHz, d6-DMSO): δ 176.0, 167.7, 145.7, 142.9, 126.8, 115.2, 90.8,
59.0, 44.8, 38.3, 37.8, 33.8, 26.5.
HRMS: Calcd. For C15H22NO5 (M+1): 296.1498, Found: 296.1498
FTIR: (neat): 3377, 3197, 1750, 1667, 1631, 1592, 1195, 1091, 1051, 1017, 985, 953,
829, 736.
MP: 154-155 ºC (decomp.)
[α]24 D = +32.57 (c = 0.645, EtOH).
160
O
OPiv
CO2Me
AcO
H
H
7-Acetoxymethyl-1-(2,2-dimethyl-propionyloxy)-1,4a,5,7a-tetrahydro-
cyclopenta[c]pyran-4-carboxylic acid methyl ester (4.39)
A flame-dried argon flushed flask was charged with amide 4.35 (50 mg, 0.169 mmol),
acetic anhydride (1.11 mL, 0.15 M), and triethylamine (0.047 mL, 0.338 mmol, 200
mol%) and the reaction was stirred for 1 h at 25 °C. The solution was cooled to 0-5 ºC
and acetic acid (0.56 mL, 0.3 M) was added. Sodium nitrite (117 mg, 1.69 mmol, 1000
mol%) was added and the reaction was stirred at 0-5 ºC for 16.5 h. The reaction was
quenched with water and ran through a column of Dowex®50WX8-200 ion-exchange
resin with MeOH. The filtrate was concentrated in vacuo to provide the crude acid 4.38.
Crude acid 4.38 was dissolved in 1:1 MeOH:CHCl3 (3.38 mL, 0.05 M) and cooled to 0
ºC. A 2.0 M solution of TMS-diazomethane in Et2O was added in portions, (0.169 mL,
0.338 mmol, 200 mol%), at 15 minute time intervals until the reaction was deemed
complete by TLC. The reaction was quenched with acetic acid, concentrated in vacuo,
and purified by flash column chromatography (SiO2, 5:1 hexane:Et2O) to furnish the title
compound as a white solid (44 mg, 74%). The reaction was also conducted using
optically active 4.35. 1H NMR: (400 MHz, CDCl3): δ 7.44 (d, J = 1.0 Hz, 1H), 5.92, (s, 1H), 5.88 (d, J = 7.2
Hz, 1H), 4.63 (s, 2H), 3.73 (s, 3H), 3.31-3.25 (m, 1H), 2.93-2.86 (m, 2H), 2.26-2.17 (m,
1H), 2.07 (s, 3H), 1.24 (s 9H). 13C NMR: (400 MHz, CDCl3): δ 176.7, 170.6, 167.3, 151.6, 136.5, 132.6, 111.1, 91.7,
61.9, 51.3, 45.1, 38.8, 38.6, 34.7, 26.8, 20.8.
HRMS: Calcd. For C18H23O7 (M-1): 351.1444, Found:351.1439.
FTIR: (neat): 2939, 1751, 1736, 1712, 1634, 1228, 1202, 1124, 1082, 1055, 1030, 970,
958, 765 cm-1.
MP: 71-72 °C
[α]25 D = +146.49 (c= 0.164, CHCl3).
161
OOAc
OAcOAc
AcO
OCl3C
NH (2R,3R,4S,5R,6R)-2-(acetoxymethyl)-6-(2,2,2-trichloro-1-iminoethoxy)tetrahydro-
2H-pyran-3,4,5-triyl triacetate (4.41)
A flame-dried argon flushed flask was charged with tetraacetyl-glucose (200 mg,
0.574 mmol), DCM (5.7 mL, 0.1M), and cesium carbonate (37 mg, 0.115 mmol, 20
mol%) respectively. The reaction was stirred at ambient temperature for 3 h. The
reaction was filtered through celite, concentrated in vacuo, and purified by flash column
chromatography (SiO2, 3:1 hexane:EtOAc 1% Et3N) to furnish the title compound as a
thick oil (237 mg, 84%). The spectral data for this compound has been previously
reported.9
162
O
OH
CO2Me
PivO
H
H
7-(2,2-Dimethyl-propionyloxymethyl)-1-hydroxy-1,4a,5,7a-tetrahydro-
cyclopenta[c]pyran-4-carboxylic acid methyl ester (4.46)
An oven-dried sealed tube was charged with racemic acetate 4.39 (40 mg, 0.114 mmol),
MeOH (1.14 mL, 0.1 M), and Otera�s catalyst 4.44 (10 mg, 0.011 mmol, 10 mol%)
respectively. The reaction vessel was sealed and heated to 70 ºC for 20 h. The reaction
vessel was then opened and PhMe (1.14 mL, 0.1 M) was added. The open reaction vessel
was then heated to 85 °C for 0.5 h until the methanol fully evaporated from the solution.
The reaction was then sealed and heated to 100 °C for an additional 1.5 h. The reaction
was directly purified by flash column chromatographed (SiO2, 5:1 hexane:ethyl acetate)
to furnish the title compound as a white solid (25.6 mg, 73%) in approximately a 5:1
epimeric ratio at the lactol stereocenter. A similar reaction was also conducted with
optically active 4.39. 1H NMR: (400 MHz, CDCl3): 7.52 (d, J = 1.0 Hz, 1H), 5.92 (d, J = 1.0 Hz, 1H), 4.99-
4.95 (m, 1H) 4.83-4.76 (m, 2H), 4.67 (d, J =13.3 Hz, 1H), 3.72 (s, 3H), 3.22-3.12 (m,
1H), 2.93-2.86 (m, 1H), 2.44 (dt, J = 8.0, 2.0 Hz, 1H) 2.09-2.01 (m, 1H), 2.21 (s, 9H). 13C NMR: (100 MHz, CDCl3): δ 179.6, 167.8, 152.7, 138.4, 132.2, 110.5, 96.5, 63.2,
51.3, 46.9, 38.9, 38.8, 36.4, 27.1.
HRMS: Calcd. For C16H23O6 (M+1): 311.1495, Found: 311.1493.
FTIR: (neat): 3551, 2966, 1708, 1629, 1284, 1191, 1163, 1142, 1131, 1100, 1085, 947,
928, 888, 832, 764 cm-1.
MP: 88-89 ºC
[α]25 D = +80.84 (c= 1.064, CHCl3).
163
O
OH
CO2Me
PivO
H
H
7-(2,2-Dimethyl-propionyloxymethyl)-1-hydroxy-1,4a,5,7a-tetrahydro-
cyclopenta[c]pyran-4-carboxylic acid methyl ester (4.46)
A flask was charged with commercially available (+)-genipin 4.43 (500 mg, 2.210
mmol), DCM (22 mL, 0.1M), and pyridine (0.267 mL, 3.315 mmol, 150 mol%). The
reaction was cooled to 0 °C and trimethylacetylchloride (0.299 mL, 2.431 mmol, 110
mol%) was added and the reaction was allowed to warm to room temperature overnight
under a balloon of argon. The reaction was diluted with Et2O washed and then diluted
with a saturated aqueous solution of NH4Cl, a 5% aqueous solution of CuSO4, and brine.
The combined organic layers were concentrated in vacuo, and purified by flash column
chromatographed (SiO2, 4:1 hexane:ethyl acetate) to furnish the title compound as a
white solid (555 mg, 81%) in approximately a 5:1 epimeric ratio at the lactol
stereocenter.
164
O
O
CO2Me
PivO OOAc
OAcOAc
AcOH
H
7-(2,2-Dimethyl-propionyloxymethyl)-1-(3,4,5-triacetoxy-6-acetoxymethyl-
tetrahydro-pyran-2-yloxy)-1,4a,5,7a-tetrahydro-cyclopenta[c]pyran-4-carboxylic
acid methyl ester (4.47)
A flame-dried argon flushed flask was charged with α-O-glycosyl-trichloroacetimidate
4.41 (159 mg, 0.322 mmol 200 mol%), lactol 4.46 (50 mg, 0.161 mmol 100 mol %), and
DCM (0.805 mL, 0.2 M). The solution was cooled to -20 ºC, and freshly distilled boron
trifluoride diethyl etherate (0.010 mL, 0.081 mmol, 50 mol %) was added. The reaction
was stirred at -20 ºC for 20 hours. The reaction was diluted with EtOAc and washed with
saturate of NaHCO3 and then brine. The organic solution was dried over magnesium
sulfate, concentrated in vacuo, and purified 3 times by flash column chromatography (1st
columin: SiO2, DCM to hexanes to 4:1 hexanes:EtOAc to 2:1 hexanes:EtOAc), (2nd
column: SiO2 2:1 hexanes:EtOAc), (3rd column, 15:1 DCM Et2O to 9:1 DCM:Et2O) to
furnish the title compound as light yellow film (64.4 mg, 62% yield). 1H NMR: (400 MHz, CDCl3): δ 7.41 (d, J = 1.2 Hz, 1H), 5.79 (d, J = 1.6 Hz, 1H), 5.23
(t, J = 9.5 Hz, 1H), 5.19 (d, J = 5.3 Hz, 1H), 5.12 (t, J = 9.6 Hz, 1H), 5.01 (dd, J = 9.7,
8.1 Hz, 1H), 4.86 (d, J = 8.0 Hz, 1H), 4.68 (s, 2H), 4.26 (dd, J = 12.5, 8.0 Hz, 1H), 4.13,
(dd, J = 12.4, 2.4 Hz, 1H), 3.74-3.69 (m, 1H), 3.712 (s, 3H), 3.23-3.18 (m, 1H), 2.94-2.91
(m, 1H), 2.84 (dd, J = 16.9, 7.9 Hz, 1H), 2.22-2.17 (m, 1H), 2.09 (s, 3H), 2.03 (s, 3H),
2.01 (s, 3H), 1.96 (s, 3H), 1.21 (s, 9H). 13C NMR: (100 MHz, CDCl3): δ 177.9, 170.6, 170.2, 169.3, 169.0, 167.2, 150.8, 136.8,
130.4, 112.1, 96.5, 95.1, 72.4, 72.0, 70.6, 68.0, 61.7, 61.5, 51.2, 46.7, 38.8, 38.3, 33.1,
27.1, 20.6, 20.5, 20.5, 20.3.
HRMS: Calcd. ForC30H41O15 (M+1): 641.2445, Found: 641.2446.
FTIR: (neat): 2958, 1746, 1367, 1278, 1214, 1152, 1077, 1036, 962, 903, 824 cm-1.
[α]25 D = +25.84 (c= 1.006), CHCl3.
165
4.12 1H and 13C NMR Spectra and HPLC Traces
166
O
O
CO2Me
HO OOH
OHOH
HOH
H
4.1
167
O
OPiv
O
4.9
168
O
O
OPiv
4.9 Racemic
92% ee
169
>99%ee
170
O
O
EtO2C OPivH
H
4.12
171
O
O
O
NO2
4.24
172
Racemic
67% ee
173
OO N
Cl
Cl
O
O
4.26
174
O
OPiv
CN
HO
H
H
4.31
175
O
CN
EtO2C OPiv
H
H
4.32
176
O
OPiv
CONH2
HO
H
H
4.35
177
O
OPiv
CO2Me
AcO
H
H
4.39
178
O
OH
CO2Me
PivO
H
H
4.46
179
O
O
CO2Me
PivO OOAc
OAcOAc
AcOH
H
180
4.13 References
1 The synthesis of allylic acetate 4.6 is described in chapter 3.
2 Lu, X.; Zhang, C.; Xu, Z. Acc. Chem. Res. 2001, 34, 535.
3Yong, L.; Song, L.; Xia, Yuanzhi, X.; Yahong, L.; Zhi-Xiang, X. Chem. Eur. J. 2008,
14, 4361.
4 Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 2003, 125, 3090.
5 Deen, H. V. D.; Oeveren, A. V.; Kellogg, R. M.; Feringa, B. L. Tetrahedron Lett. 1999,
40, 1775.
6 (a) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 4545. (b) Trost, B. M.;
Machacek, M. R.; Aponick, A. Acc. Chem. Res. 2006, 39, 747. 7 (a) Ghaffer, T.; Parkins, A. W. Tetrahedron Lett. 1995, 36, 8657. (b) Ghaffer, T.;
Parkins, A. W. J. Mol. Catal. A 2000, 160, 249. 8 Tietze, L. �F. Angew. Chem. Int. Ed. Engl. 1983, 22, 828. 9 Kværnø, L.; Ritter, T.; Werder, M.; Hauser, H.; Carreira, E, M. Angew. Chem. Int. Ed. 2004, 43, 4653. 10 Schmidt, R. D. Angew. Chem. Int. Ed. Engl. 1986, 25, 212.
11 Djeerassi, C. Gray, J. D. Kingel, F. A. J. Org. Chem. 1960, 25, 2174.
12 Otera, J.; Dan-oh, N., Nozaki, H. 1991, 56, 5307. 13 Lactol 4.46 can also be accessed in one step from commercially available (+)-genipin. See supporting information for details. 14 Mouriès, C.; Deguin, B.; Koch, M.; Tillequin, F. Helv. Chim. Acta. 2003, 86, 147.
181
15 This yield is based on a 70% theoretical yield for the kinetic resolution. 16 (a) Endo, T.; Taguchi, H. Chem. Pharm. Bull. 1973, 21, 2684. (b) Tanaka, M.; Kigawa, M.; Mitsuhashi, H.; Wakamatsu, T. Heterocycles, 1991, 32, 1451. (c) Isoe, S. Stud. Nat. Prod. Chem. 1995, 16, 289. 17 Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923. 18 Güvenalp, Z.; Kili�, N.; Kazaz, C.; Kaya, Y.; Demirezer, Ö. Turk. J. Chem. 2006, 30, 515. 19 Morota, T.; Sasaki, H.; Nishimura, H.; Sugama, K.; Chin, M; Mitsuhashi, H. Phytochemistry, 1989, 28, 2149.
182
Vita
Regan Andrew Jones was born on May 8th 1981 in Vancouver, Washington. He
graduated from Woodland High School, Woodland WA in 1999. From 1999 to
2003 he attended Occidental College in Los Angeles, CA, and in 2003 he
received his B.A. in chemistry from Occidental College. In 2003 he enrolled in
the graduate program in chemistry at the the University of Texas at Austin.
Permanent Address: 2610 Lewis River Road, P.O. Box 575, Woodland, WA,
98674.
This dissertation was typed by the author.